• Author / Uploaded
• Glasberg M

Citation preview

ENCYCLOPEDIA OF FOOD AND HEALTH

This page intentionally left blank

ENCYCLOPEDIA OF FOOD AND HEALTH EDITORS-IN-CHIEF

BENJAMIN CABALLERO PAUL M. FINGLAS FIDEL TOLDRA´

AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB 225 Wyman Street, Waltham MA 02451 Copyright © 2016 Elsevier Ltd. All rights reserved No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notice Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers may always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN 978-0-12-384947-2 For information on all publications visit our website at http://store.elsevier.com Printed and bound in the United Kingdom.

Acquisition Editor: Rachel Gerlis Content Project Manager: Justin Taylor Cover Designer: Maria Ineˆs Cruz

EDITORS-IN-CHIEF Benjamin Caballero is professor of International Health and of Maternal and Child Health (Bloomberg School of Public Health), and professor of pediatrics (School of Medicine) at Johns Hopkins University. He obtained his MD from the University of Buenos Aires, his MSc in biochemistry from the University of San Carlos, and his PhD in neuroendocrine regulation from MIT, in Cambridge, MA. He started his academic career as assistant professor of pediatrics at Harvard Medical School and director of the Nutrition Unit of Boston Children’s Hospital, and subsequently became the founding director of the Center for Human Nutrition at Johns Hopkins University, in Baltimore. Prof. Caballero has focused his research on child nutrition and health in developing countries. In particular, he has explored the combination of undernutrition and overweight that has become increasingly prevalent in low- and middle-income countries. He was a member of the Food and Nutrition Board of the Institute of Medicine, National Academy of Sciences, USA, and of a number of expert panels created by the Institute, including the Dietary Reference Intakes (DRI) Committee, the Expert Panel on Macronutrient Requirements, and the Childhood Obesity Task Force. He was also a member of the Dietary Guidelines for Americans Advisory Committee, of the Scientific Advisory Board of the Food and Drug Administration (FDA), and of a number of advisory committees of the National Institutes of Health (USA). He is the editor-in-chief of the Encyclopedia of Food Sciences and Nutrition, a 10-volume work on food production, consumption and biological effects. He is also editor-in-chief of the Encyclopedia of Human Nutrition, which received the Book of the Year Award from the British Medical Association. His Guide to Dietary Supplements summarizes the current scientific basis for the use of mineral and vitamin supplements. His book The Nutrition Transition: Diet and Disease in the Developing World explored the impact of demographic and economic development on diet- and lifestyle-related diseases in developing countries. His book Obesity in China summarizes research conducted in rural and urban China to track the impact of socioeconomic development on health outcomes. He is also coeditor of a widely used textbook on human nutrition, Modern Nutrition in Health and Disease. He is a member of the Spanish Academy of Nutritional Sciences, and a Fellow of the American Society for Nutrition and of the Royal Society of Medicine (UK). Recent awards include the Donald Medearis Lectureship from the Massachusetts General Hospital/Harvard Medical School, the Mataix Prize for lifetime achievements in nutrition science from the Spanish Academy of Nutritional Sciences, the Ancel Keys Prize for achievements in international public health, and the Thompson–Beaudette Lectureship from Rutgers University.

Paul Finglas joined the Institute of Food Research in 1981 and is currently head of the Food Databanks National Platform and Research Leader in Food and Health at the Institute (http://www. ifr.ac.uk/science/platform/FD/default.html). He has, for most of his science career, been involved in a wide range of research in food composition and analysis, and the nutritional effects of micronutrients in food and health research. Paul has considerable experience of co-coordinating both national and international projects (e.g., EuroFIR, TDS-EXPOSURE, Bacchus and QualiFY (all EU FP7), and is currently of the spin-out EuroFIR AISBL, a non-profit international association based in Belgium, from one of these projects. Paul has a broad range of experience in science publishing and is currently editor of the journals Food Chemistry and Trends in Food Science and Technology, and was one of the coeditors for the Encyclopedia of Food Science and Nutrition (2nd Ed.). Paul has a degree in chemistry from Aston University in Birmingham and has published over 150 publications on a wide range of topics in food science and nutrition.

v

vi

Editors-in-Chief

Fidel Toldra´ holds a BSc in chemistry (1980), high degree on food technology (1981) and PhD in chemistry from the University of Valencia (1984). Professor Toldra´ was a Fulbright postdoctoral scholar at Purdue University in West Lafayette (US, 1985–86) and visiting scientist at the University of Wisconsin-Madison (1991 and 1995), and the Institute of Food Research-Bristol (UK, 1987). Currently, he is research professor at the Instituto de Agroquı´mica y Tecnologı´a de Alimentos (CSIC), in Paterna, Valencia (Spain). He is also associate professor of food technology at the Polytechnical University of Valencia. Prof. Toldra´ has focused his research on food biochemistry and its relationship with nutrition, quality and safety. He has filed 12 patents, directed 22 PhD thesis and published over 245 manuscripts in recognized scientific journals and more than 115 chapters of books. His h-index is 41. Prof. Toldra´ has authored two books and edited/co-edited more than 30 books for major publishers like CRC Press, Wiley-Blackwell, Elsevier and Springer. Prof. Toldra´ is the European editor of Trends in Food Science and Technology (2005–) and associate editor of Meat Science (2014–); he was the editor-in-chief of Current Nutrition & Food Science (2005–2012), section editor of the Journal of Muscle Foods (2009–2010) and guest editor of 12 special journals issues. He is a member of the editorial boards of Food Chemistry, Food Analytical Methods, Journal of Food Engineering, Journal of Food and Nutrition Research, The Open Nutrition Journal, The Open Enzyme Inhibition Journal, Recent Patents in Agriculture, Food and Nutrition, Food Science & Nutrition and Current Opinion in Food Science. He has been a member of the Scientific Panel on food additives, flavorings, processing aids and materials in contact with foods (periods 2003–2008) and the Scientific Panel on flavorings, enzymes, processing aids and materials in contact with foods (2008–2015) of the European Food Safety Authority (EFSA) acting as Chairman of the Working groups on Irradiation (2009–2010), Processing Aids (2011–2014) and Enzymes (2010–2015). He was a member of FAO/WHO group of experts to evaluate chlorine-based disinfectants in the processing of foods (2008–2009). He was a member of the Executive Committee of the European Federation of Food Science and Technology (EFFOST, 2002–2009). He is a Fellow of the International Academy of Food Science and Technology (IAFOST, 2008) and of the Institute of Food Technologists (IFT, 2009–). He received the Iber Award on Food and Cardiovascular Diseases (1992), the Institute Danone award in Food, Nutrition and Health (2001), the International Prize for Meat Science and Technology from the International Meat Secretariat (2002), GEA award on RþD activity from the Valencian Community (2002), and the Distinguished Research Award (2010) and Meat Processing Award (2014), both from the American Meat Science Association.

EDITORIAL ADVISORY BOARD Siaˆn Astley has worked extensively with individuals and organizations throughout Europe from a variety of disciplines including research, food and biotech industries, and the media. She is the author of more than 300 popular science articles for magazines and trade publications as well as 25 peer-reviewed papers, and she was awarded her Diploma in Science Communication in 2009 (Birkbeck University of London). After 14 years as a bench-scientist, Siaˆn became communications manager for NuGO, one of the first FP6 networks of excellence, and was the European communications manager for the Institute of Food Research in Norwich (UK) until April 2012. Currently, she is the training and communications manager for the European Food Information Resource (EuroFIR AISBL) supporting training within EU-funded research projects and networks, and communication of research activities.

David J. Baer is a supervisory research physiologist with the US Department of Agriculture’s Beltsville Human Nutrition Research Center located in Beltsville, Maryland. He serves as the research leader for the Center’s Food Components and Health Laboratory and also serves as the director of the Center’s Human Studies Facility. Dr. Baer conducts controlled dietary intervention studies to investigate the relationship between diet and the risk for chronic degenerative diseases, especially cardiovascular disease, cancer, and diabetes in people. His research also includes studies on the health impacts of weight gain and determining the calorie content of foods. Some of the dietary interventions he has investigated include the effects of different types of protein, fats and fatty acids, fiber, margarine, butter, plant sterols, salad dressings, nuts, whole grains, berries, alcohol, and tea on overall nutrition and health. In addition to dietary intervention studies, Dr. Baer is involved in research studies to validate food survey methodologies and in developing new methods for dietary assessment. Dr. Baer earned his bachelor’s degree from the University of Illinois and his master’s and doctorate in nutrition from Michigan State University.

vii

viii

Editorial Advisory Board

Marina Carcea was awarded a master degree in agricultural sciences at the University of Pisa, Italy “cum laude” in 1980, and a PhD in food science also “cum laude.” She is currently a senior researcher in the Research Center on Food and Nutrition of the Council for research in agriculture and analysis of agricultural economy (CRA-NUT formerly INRAN National Research Institute on Food and Nutrition) and she was the director of the Cereals Research Programme in INRAN. CRA-NUT is a primary research institute in Italy under the egis of the Ministry of Agriculture. Dr. Carcea joined INRAN in 1989 after having worked in Italian and English universities (Queen Elizabeth College, King’s College, and University of London) and after a two-year contract with the Food and Agriculture Organization (FAO) of the United Nations (UN), Rome. She has a vast experience in the field of research on foods, cereals in particular. In recent years, her main research interests have been: chemical characterization and study of the functional properties of cereal components; study of the interactions between components and of the interrelationships between the biochemical properties of components and the technological properties of the raw material and derived products; development of new, cereal-based products; development of methods to assess technological parameters of the raw material; nutritional value of cereals; and developments of protocols for quality assurance of cereals, food authenticity. She has taken part and/or co-ordinated several research projects within national or international programs (European Commission, FAO) involving several institutions. She is the author of more than 160 scientific publications, mostly in international journals, eight book chapters, and two scientific books. She delivered lectures on her research activity at about 150 national and international congresses and she seats in several national and international committees (Italian Ministry of Agriculture, Codex Alimentarius, and European Commission) regarding food and nutrition topics. She is also a member of the editorial board of scientific journals. From 1994 to 2006, she has also been a lecturer of food science and technology at the University of Tor Vergata, Rome, Italy. She is a founding member of AISTEC, the Italian Association of Cereal Science and Technology. Since 1996, she is an elected member in the Executive Committee of the same association and since 2009, president of the association. Since 2000, she is the Italian National Delegate of the International Association for Cereal Science and Technology (ICC) and she was also the president of the same association for 2011–2012. In 2004, she was the first woman to be awarded the International Harald Perten Prize for her excellent research achievements in the field of cereal science and technology. She is also a member of the Georgofili Academy in Florence, Italy.

Lawrence J. Cheskin graduated from Dartmouth Medical School and completed a fellowship in gastroenterology at Yale–New Haven Hospital. He is an associate professor of health, behavior, and society at the Johns Hopkins Bloomberg School of Public Health, with a joint appointment in International Health–Human Nutrition, and in medicine (GI) at the Johns Hopkins University School of Medicine. Dr. Cheskin is also a founder and director of the Johns Hopkins Weight Management Center, a comprehensive treatment program for obesity. In his research, Dr. Cheskin has studied the effects of medications on body weight, the gastrointestinal effects of olestra, how cigarette smoking relates to dieting and body weight, and the effectiveness of lifestyle and dietary changes in weight loss and weight maintenance. He is also the author of four books: Losing Weight for Good, New Hope for People with Weight Problems, Better Homes and Gardens’ 3 Steps to Weight Loss, and Healing Heartburn. Dr. Cheskin has appeared on television news programs and lectured to both professional and lay audiences on the topics of obesity and weight control.

Editorial Advisory Board

ix

Nigel Cook is a graduate of the University of Dundee. After postdoctoral research in the Universities of Aberdeen and Leicester, he moved to the Central Science Laboratory (now the Food and Environment Research Agency (FERA)) at the Food Science Laboratory, Torry, Aberdeen in September 1994, before relocating to new facilities in York. At FERA, he studies the transmission of pathogens, particularly enteric viruses, through foods and the environment. He has a visiting professorship at the Katholieke Universiteit Leuven in Belgium. He is a councilor of the International Association for Food and Environmental Virology. He is a project leader within the standardization working group ISO TC34 SC9 WG6, currently developing a standard for detection of Cryptosporidium and Giardia on berry fruits and leafy green vegetables. He was a coordinator of the European Framework 7 project “Integrated monitoring and control of foodborne viruses in European food supply chains (VITAL),” and a chair of COST Action 929 “A European Network for Environmental and Food Virology” from 2006 to 2010. Between 2009 and 2014, he was a member of various European Food Safety Authority’s Working Groups preparing opinions on the risk of foodborne viruses, and represented the European Communities on the Codex Committee on Food Hygiene Working Group developing Guidelines on the Application of General Principles of Food Hygiene to the Control of Viruses in Food. He was a member of the UK Advisory Committee on the Microbiological Safety of Food’s Viral Infections Subgroup. He was the founding editor of the journal Food and Environmental Virology, published by Springer. Luca Simone Cocolin graduated in 1994 in food science with a grade of 110/110 and remark followed by food biotechnology PhD studies from 1995 to 1998. In February 1999, he defended his thesis acquiring the title of PhD in food biotechnology. From 1998 to 2001, he received a scholarship from the Friuli Venezia Giulia region (Italy). From November 1, 2001, he was an assistant professor at the University of Udine, Faculty of Agriculture, Food Science Department, Italy, and in October 1, 2006, he became an associate professor at the University of Torino, Italy. In January 2014, he had the habilitation for full professor and from June 2015, he is the full professor in food microbiology at the University of Torino. From September 2008, he is an executive board member of the International Committee on Food Microbiology and Hygiene (ICFM) part of the International Union of Microbiological Societies (IUMS) (http://www.icfmh.org/). From January 2008, he is the editor-in-chief of the International Journal of Food Microbiology and he is a member of the editorial board of Applied and Environmental Microbiology, Food Analytical Methods, Frontiers in Food Microbiology, and Frontiers in Nutrition and Food Science Technology. He regularly reviews paper for Food Microbiology, Meat Science, Journal of Applied Microbiology, and Letters in Applied Microbiology. He is a co-author of more than 180 papers on national and international journals and he attended national and international congresses with oral and poster presentations. On Scopus (www.scopus.com, consulted on March 2015) he has 172 documents reviewed, which were cited 3520 times, resulting in an h index of 33.

His scientific interests comprise: development, optimization, and application of molecular methods for the detection, quantification, and characterization of foodborne pathogens; study of the microbial ecology of fermented foods (mainly sausage, cheese, and wine) by using culture independent and dependent methods; bioprotection: molecular characterization of bacteriocin production and its study in vitro and in situ; selection of new putative probiotics from artisanal fermented foods; and study of the human microbiome and its influence on human health. Christopher Duggan, for the past 25 years, has been performing clinical trials in the fields of pediatric nutrition, gastroenterology, and global health. His early work centered on the management of diarrheal diseases in children, including trials that demonstrated the feasibility and efficacy of oral rehydration solutions (ORS) for diarrhea management in the United States and globally. In collaboration with colleagues at Harvard TS Chan School of Public Health and Muhimbili University of Health and Allied Sciences in Dar es Salaam, Tanzania, Dr. Duggan and colleagues are evaluating the efficacy of micronutrient supplementation in infants and young children born to women with or at risk of HIV infection. Recent studies include the development of new biomarkers of environmental enteric dysfunction as well as the evaluation of nutritional status on neurodevelopment. With colleagues at St John’s Research Institute in Bangalore, India, he is evaluating the efficacy of maternal vitamin B12 supplementation on biochemical and clinical parameters during pregnancy and infancy. He is a course co-director of the Bangalore, Boston Nutrition Collaborative (http://bbnc.globalhealth.harvard.edu). Past and present research support has come from the National Institutes of Health, the Gates Foundation, and the World Health Organization. In addition to his global health research interests, he is a pediatric gastroenterologist and a nutrition physician at Boston Children’s Hospital where he directs the Center for Nutrition (http://www.childrenshospital.org/nutrition). He is a medical director of the Center for Advanced Intestinal Rehabilitation, one of the largest centers in the United States for the care of children with intestinal failure/chronic diarrhea syndromes (http://www.childrenshospital.org/cair). He is also the course co-director of an inaugural Harvard College course “Nutrition and Global Health” and mentors undergraduate, graduate, and postdoctoral students at HMS and HSPH (http://www.hsph.harvard.edu/nutrition-and-global-health/). He is a professor of pediatrics at Harvard Medical School and a professor in the Department of Nutrition at the Harvard TS Chan School of Public Health (http://www.hsph.harvard.edu/faculty/christopher-duggan/).

x

Editorial Advisory Board Jed William Fahey’s current research concerns elucidating the mechanisms of how plants protect themselves against unfavorable and stressful conditions, and how this understanding can be translated to chemoprotection of eukaryotic mammalian systems. This work draws on elements of natural product chemistry, enzymology, nutritional epidemiology, and clinical research in order with isothiocyanates (e.g., sulforaphane) and glucosinolates. His work led to the discovery that broccoli sprouts are an exceptionally rich and consistent source of phytochemicals that induce the detoxification of carcinogens, and to the development of methods for their detection and for assessing their metabolism in humans. He discovered that one of the inducers, sulforaphane, has potent antibiotic activity against Helicobacter pylori, a causative agent of peptic ulcer and stomach cancer, and followed up with trials in animals and in H. pylori-infected humans. Ongoing collaborations examine the effects of broccoli, Moringa, and the other plants and their phytochemicals against a range of chronic diseases. Dr. Fahey has for years taught courses in chronic disease prevention and nutrition at both medical and public health schools.

Manuel Franco is an associate professor at University of Alcala´ in Madrid (Spain) where he leads the social and cardiovascular epidemiology research group (http://www3.uah.es/cardiosocialepi/). He is also an adjunct professor at Johns Hopkins University (Baltimore). Prof. Franco’s work focuses on the social determinants of cardiovascular diseases and its major risk factors as diet. His methodological interests include the measurement of the urban environment and large social and economic changes in relation to cardiovascular health. He is the lead investigator of the Heart Healthy Hoods, study funded by the European Research Council, that will study the urban environment in relation to cardiovascular health in Madrid (http://hhhproject. eu/). This longitudinal study will be collecting neighborhood level data (via audits, Google Street view, photovoice, and qualitative methods) and linking them to clinical outcomes collected from patients enrolled at the City of Madrid primary healthcare clinics. Prof. Franco trained in Spain and Germany to obtain his MD and obtained his PhD from Johns Hopkins Bloomberg School of Public Health working with Dr. Ana Diez-Roux in the MESA study on food environment and dietary patterns. He has published over 30 international high impact articles and collaborates with universities in the United States, Europe, and Latin America.

Maria Glibetic is a research director of Centre of Research Excellence in nutrition research, Institute for Medical Research in Belgrade, University of Belgrade, Serbia, and member of executive board of directors for food data association EuroFIR AISBL. She is an experienced basic and nutritional scientist with over 250 scientific publications and presentations. Maria has considerable experience in leading national and international projects and since 2006, she participated in nine EU funded projects including EuroFIR, EURRECA, BaseFOOD, CHANCE, BACCHUS, and ODIN. Maria and her team are responsible for the creation of the first online national food database, for designing food data management system, and for the development of different nutritional tools for intake analysis. She was a principal leader of many nutrition intervention studies evaluating the plant bioactive component effects on human cardiovascular health. She leads postgraduate department for integrated nutritional sciences at University of Belgrade, where she teaches two courses.

Linda Harvey obtained her PhD from the University of East Anglia, UK. She is currently the head of the Human Nutrition Unit at the Institute of Food Research, Norwich, UK. Her research interests include micronutrient requirements, bioavailability, and metabolism. Ronald Jackson received his bachelor’s and master’s degrees from Queen’s University and doctorate from the University of Toronto. His time in Vineland, Ontario, and subsequent sabbatical at Cornell University, redirected his interest in Botrytis toward viticulture and enology. As part of his teaching duties at Brandon University, he developed the first wine technology course in Canada. For many years he was a technical advisor to the Manitoba Liquor Control Commission, developing sensory tests to assess candidates of its sensory panel, and was a member of its external tasting panel. He is the author of Wine Science: Principles and Applications, 4th edition (2014), Wine Tasting: A Professional Handbook, 2nd edition (2009), Conserve Water, Drink Wine, and chapters and technical reviews in other multiple books and encyclopedia. He is retired in Bronte, Ontario, but remains active writing, cycling, doing yoga, and traveling, as well as being a fellow in the Cool Climate Viticulture and Oenology Institute, Brock University, St. Catharines, Ontario, Canada.

Editorial Advisory Board

xi

Joe P. Kerry is a senior college lecturer and the head of the food packaging research group in the School of Food and Nutritional Sciences at University College Cork (UCC). He received his doctorate in microbiology at University College Galway in 1995. Prof. Kerry is also a qualified member of the Institute of Packaging. He is very involved in national and international research projects both at fundamental and applied levels. Primary research interests address various aspects of food packaging, shelf-life stability, food composition, and numerous aspects of food quality, particularly in relation to muscle foods. He also has very strong links with industry and his research team assists companies in relation to many aspects of new food product development. He has over 220 publications in peer-reviewed international journals, over 300 presentations at major international conferences, along with several other significant publications. His expertise includes use and manipulation of modified atmosphere packaging systems for use with foods, use of extrusion technology for the manufacture of food products/packaging materials, and applications and sensor/new technology developments within the area of food packaging, especially in the area of smart packaging. Fre´de´ric Leroy, after studying bio-engineering at Ghent University, obtained a PhD in applied biological sciences at the Vrije Universiteit Brussel in 2002, where he continued his academic career at the research group of Industrial Microbiology and Food Biotechnology (faculty of sciences and bio-engineering sciences). As associate professor, his lecturing activities include courses in food science and technology (i.e., “Nutrition,” “Technology of animal products,” “Food microbiology and ecology,” and “Quantitative and predictive microbiology”). Dr. Leroy’s research primarily deals with the ecology and functional roles of bacterial communities in (fermented) foods, in particular with respect to the generation of quality, safety, and/or nutritional and health advantages. Focus is mostly on meat products, but other food systems are also being studied, including fermented milks and sourdough breads. In addition, his research interests relate to elements of tradition and innovation in foods, both from a technological and societal point of view.

Catherine M. Logue completed her undergraduate and postgraduate degrees in Ireland and earned a PhD in meat microbiology from the University of Ulster, UK in 1996. Dr. Logue was a faculty member at North Dakota State University from 1999 to 2011 rising through the ranks of assistant to associate and full professor. In 2011, she re-located to Iowa State University’s College of Veterinary Medicine and is a professor of veterinary microbiology and preventive medicine. Dr. Logue is also the director of faculty and staff advancement and equity for the college. Her research interests focus on foodborne pathogens of food animals and the contamination of meat and meat products destined for human consumption. Her research studies the detection, isolation, and characterization of a range of foodborne pathogens such as Salmonella, Campylobacter, Listeria, Escherichia coli, and methicillin-resistant Staphylococcus aureus (MRSA) in poultry, bovine, and swine. She also focuses her research on antimicrobial resistance in commensals and pathogens of production animals. She has been an author and a co-author on more than 90 peer-reviewed papers and book chapters as well as more than 150 abstracts and presentations at national and international meetings.

F. Xavier Malcata graduated in chemical engineering in 1986 from the University of Porto (Portugal), received a PhD in chemical engineering/food science from University of Wisconsin, Madison (USA) in 1991, and his habilitation in food science and engineering by Portuguese Catholic University in 2002. He was the dean of College of Biotechnology of Portuguese Catholic University, the chairman of Portuguese Society of Biotechnology, Portuguese representative at VI and VII European Union Framework Programs of research and development, expert for European Food Safety Agency, and a co-ordinator of Portuguese Engineering Accreditation Board in chemical engineering for Northern Region. He is currently a full professor at University of Porto. His major research interests have focused on technological improvement of traditional Portuguese foods and upgrade of byproducts thereof, development of nutraceutical ingredients and functional foods, design and optimization of enzymatic reactors for edible oil processing, characterization of plant proteases toward cheese and whey cheese manufacture, production of starter and nonstarter cultures from adventitious microflora, and optimized application of unit operations to food processing. With an academic career of independent research and teaching for more than two decades, Prof. Malcata published more than 450 papers in refereed journals worldwide, wrote 11 books, and prepared more than 45 chapters for edited books. Among many international distinctions, he was

xii

Editorial Advisory Board

recipient of Ralph H. Potts Memorial Award in 1991 by American Oil Chemists’ Society (AOCS, USA), Foundation Scholar Award – Dairy Foods in 1998 by American Dairy Science Association (ADSA, USA), Young Scientist Research Award in 2001 by AOCS, Canadian/ International Constituency Investigator Award in physical sciences and engineering in 2002 and 2004 by Sigma Xi (USA), Danisco International Dairy Science Award in 2007 by ADSA, Scientist of the Year Award in 2007 by European Federation of Food Science and Technology (the Netherlands), Samuel C. Prescott Award in 2008 by Institute of Food Technologists (IFT, USA), International Leadership Award in 2008 by International Association for Food Protection (IAFP, USA), Elmer Marth Educator Award in 2011 by IAFP, Distinguished Service Award in 2012 by ADSA, and William V. Cruess Award in 2014 by IFT. He has been elected for the honor societies of food science (Phi Tau Sigma, USA), scientific research (Sigma Xi, USA), and engineering (Tau Beta Pi, USA). He was also elected for fellow of IFT, ADSA, AOCS, and International Academy of Food Science and Technology. Gopinadhan Paliyath is a professor at the Department of Plant Agriculture, University of Guelph, and the research program director for “Food for Health,” under the UG/OMAFRA partnership. Dr. Paliyath is a biochemist and has an interest in various aspects of fruits and vegetables, specifically the nutraceutical components and their mechanism of action. He obtained his BSc Ed degree (botany and chemistry) from the University of Mysore, MSc degree (botany) from the University of Calicut, and PhD degree (biochemistry) from the Indian Institute of Science, Bangalore. Subsequently, he did postdoctoral work at Washington State University, University of Waterloo, and University of Guelph. Dr. Paliyath’s research is focused on the biochemistry of plant senescence, specifically pertaining to postharvest biology and technology of fruits and vegetables. Investigations on the role of phospholipase D (PLD) in membrane homeostasis and signal transduction have led to advances in the understanding of the mechanism of membrane deterioration that occur during stress and senescence. Another aspect of his research is focused on understanding the mechanism of action of food components in disease prevention. The efficacy, bio-accessibility, bioavailability, and molecular mechanisms of action of nutraceuticals in fruits and processed products in relation to their cancer-preventive and anti-inflammatory actions are being investigated using mammalian cell lines, and animal model systems. Dr. Paliyath has developed technologies and products for enhancing the shelf life and quality of fruits and vegetables based on PLD inhibition. R&D activities relevant to the industry sector include: (1) optimization of an enhanced freshness formulation for application to various fruits, vegetables, and flowers; (2) developing methods for nutraceutical carriers that would enhance the functional food quality and delivery (e.g., stabilizing lycopene in tomato juice, sauce, etc., for health beneficial effects); and (3) developing novel technologies to enhance the cancer-preventive ingredients in fruit products, etc. Patents awarded include: (1) # 6,514,914 (US) and 2,298,249 (Canada); (2) #7,198,811 (USA), 4141387-1 (Japan), 260738 (Mexico), 1469736 (Turkey), 028 284763 (China), and 223077 (India). The patents describe the use of nanoformulations based on hexanal and other generally regarded as safe (GRAS) ingredients for enhancing the shelf life and quality of fruits, vegetables, and flowers by pre or postharvest treatments. These technologies are currently being evaluated for extending shelf life and quality of mango in India and Sri Lanka with the assistance of the Canadian International Food Security Research fund. The collaboration involves researchers from Canada, India (Tamil Nadu Agricultural University), and Sri Lanka (Industrial Technology Institute). Dr. Paliyath is also the research program director for the food and health theme-related activities under the OMAF/MRA/University of Guelph research partnership. He serves on the editorial board of several journals. He is a member of American Chemical Society and Canadian Society of Plant Biologists. (Total-refereed publications in journals – 92; patents and intellectual properties – 2; disclosures – 4; chapters in books – 27; nonrefereed contributions – 10; research reports – 28; conference proceedings – 88; edited books – 9; book reviews – 6 (Google Scholar: h index – 31, i10 index – 68, citations – 4332; RG score – 35.63)). Yolanda Pico´ is a full professor of nutrition and food science at the University of Valencia since 1998. She is currently the head of the research group on food and environmental safety of the University of Valencia. Her research interests are the development of new analytical methods to determine organic contaminants in food and the environment, identification of unknown compounds by liquid chromatography–mass spectrometry, micro-extraction separations, and environmental and food safety. To the date, she is the author of nearly 200 peer-reviewed papers, 170 scientific papers in journals of SCI, 25 book chapters, and editor of four books on food and environmental safety.

Editorial Advisory Board

xiii

Vieno Piironen is a professor of food chemistry at the Department of Food and Environmental Sciences, University of Helsinki, Finland. She received her PhD in food chemistry at the University of Helsinki in 1987 and has approximately 35 years of experience in food research and education on bachelor, master, and PhD levels. She has participated actively in international research and education projects and networks. Her research has focused especially on chemical and nutritional properties, reactions and analysis of lipids, vitamins, and other bioactive compounds. Research on vitamins has been active from the beginning of 1980s. She has studied both lipid- and watersoluble vitamins; their chemical and nutritional properties and importance in foods and diets as well as factors influencing vitamin levels. In addition, development of analytical methods for different vitamers as well as validation and harmonization of the methods through international collaboration have been among the priorities. Recent collaboration projects have focused on enhancing vitamin contents in cereal-based foods by plant breeding, utilization of vitamin-rich grain fractions, and bioprocessing. Currently, the research focus lies on investigating microbial in situ synthesis of folate, vitamin B12, and other B vitamins in cereal and legume matrices as a means to improve nutritional quality of foods and to develop new food applications. In lipid research, different lipid classes and their chemical and enzymatic reactions in food matrices are studied. Diverse methods are used to study proceeding of oxidation from primary products to monomeric oxides, volatiles, and polymerization products and to study possibilities to control oxidation. Controlling enzymatic reactions leading to off-flavors in cereal and legume matrices is also among the interests. Phytosterols and their conjugates have been studied as natural food components belonging to the dietary fiber complex. On the other hand, questions related to sterol enrichment such as oxidation susceptibility and mechanisms as well as factors affecting oxidation reactions have been of interest. She has also studied nutrients and anti-nutrients in legumes and more recently started research on utilization of high value components in microalgae. She has approximately 160 papers in international journals and a number of other publications. David Rodrı´guez-La´zaro is a doctor in veterinary medicine (DVM), specialized in food science (BSc and MSc) and molecular microbiology (PhD). He is a senior scientist at ITACyL and an assistant professor of microbiology at the University of Burgos. He has performed research stays in the Danish Institute for Food and Veterinary Research (Denmark), the University of Prague (Czech Republic), the Food and Environmental Research Agency (UK), and the University of Bristol (UK). He was a Leverhulme visiting professor in the Institute of Advanced Sciences in the University of Bristol during the years 2004 and 2005 and Marie Curie research fellow in the faculty of medical and veterinary sciences in the University of Bristol (UK) until 2007. His research interest is focused on the establishment of reliable, quantitative molecular strategies for detection of important food-borne pathogens from environmental sources and various types of foodstuffs, the characterization of the prevalence of the main foodborne pathogens in food and food-related environments, and the development of emergent food preservation processes and their effects in the microbial virulence. He has participated in a number of coordinated EU-funded projects such as PROMISE, BASELINE, VITAL, FOOD-PCR, SACROHN, and MONI-QA, having established active links with the leading European research groups working in food safety. He has published more than 100 international scientific papers and book or book chapters regarding food safety. He is currently a member of the editorial board of Applied and Environmental Microbiology, International Journal of Food Microbiology, Food and Environmental Virology, and International Journal of Food Contamination and the editor-in-chief of the journal Food Analytical Methods. He was awarded with the XV Jaime Ferra´n Award in 2013 by the Spanish Society for Microbiology for his promising scientific career in microbiology. Turid Rustad is a professor and the head of the food science group at the Department of Biotechnology, Norwegian University of Science and Technology. The main research focuses on the biochemistry of marine raw materials, the relationship between biochemistry and quality, and changes in raw material properties during processing. Studies of enzymatic activities in different raw materials have been linked to studies of changes in the biochemistry of these raw materials. She has worked with characterization of composition and enzymatic processes in a wide range of different raw materials, such as fish, fish by-products, and zooplankton in relation to different storage and processing methods such as chilling, heating, superchilling, and frozen storage.

xiv

Editorial Advisory Board

Noel W. Solomons has lived and worked in Guatemala for 40 years. He was born and educated in Massachusetts in the United States. As a young child, he became an amateur naturalist and was a nature counselor at various summer camps; this would guide him to a career in science. In his young adulthood, he would participate in the civil rights and anti-war movements, only to become disillusioned by the intractable nature of the injustice elements in the fabric of American society. As a physician by graduate training, he performed his university studies at Harvard College and Harvard Medical School; it was during overseas electives in his medical training that he visited Peru and Colombia and committed to an expatriate life trajectory outside of his homeland. Clinical training included a residency in internal medicine and infectious diseases at the Hospital of the University of Pennsylvania and specialization in gastroenterology and clinical nutrition at the University of Chicago. He became a resident of Guatemala in 1974 as an affiliated investigator at the Institute of Nutrition of Central America and Panama. He would later commute for eight years to a faculty position in the Department of Nutrition and Food Science of the Massachusetts Institute of Technology. Assuming a full-time Guatemala commitment in 1985, he co-founded the Center for Studies of Sensory Impairment, Aging and Metabolism (CeSSIAM) where he remains its scientific director. Over 40 local university theses have been completed by Central American students in that institution as well as an equal number of master’s degree research projects from international students from Europe, and North and South America. He has supervised doctoral dissertations for 12 PhD candidates from the United States, Canada, Germany, Spain, and the Netherlands through CeSSIAM. Dr. Solomons has 332 publications indexed on Medline. In addition, he has edited two books and contributed over 100 articles, reviews, editorials, and commentaries in nonindexed venues and over 50 book chapters. These are dedicated to the scientific and academic interests of his career including: clinical nutrition; human growth and body composition; lactose maldigestion; dietary intake, nutritional status, intestinal absorption, and food fortification related to various micronutrients (vitamins, trace elements, and essential fatty acids); complementary feeding; nutrition in aging and chronic disease; and the interaction of malnutrition and infection. Among the honors bestowed upon Dr. Solomons are the International Nutrition Prize of the International Union of Nutritional Sciences and the Kellogg Prize of the Society for International Nutrition Research. He is a fellow of the American Society of Nutrition. He is an academic member of the Guatemalan Academy of Medical, Physical and Natural Sciences and the Spanish Academy of Nutrition and Food Science. He was the awardee of the 2010 National Medal for Science and Technology for Guatemala. He has been a visiting professor in university courses in Mexico, Peru, Brazil, Indonesia, and Spain. He currently holds adjunct professorial appointments at the Boston University School of Public Health, and the Friedman School of Nutrition Science and Policy and the Department of Community Medicine and Public Health, both at Tufts University. He is a founding board of directors, member of the Hildegard Grunow Foundation in Munich and the Essential Nutrient Foundation of Singapore. Finally, Dr. Solomons is a coordinator for Central America of the Nevin Scrimshaw International Nutrition Foundation in Boston, and an associate editor for the Foundation’s Food and Nutrition Bulletin. He serves on editorial boards for ten scientific journals. Maria Tsimidou is a professor of food chemistry and the head of the Laboratory of Chemistry and Technology in the School of Chemistry at the Aristotle University of Thessaloniki (AUTh), Greece. Her teaching is food chemistry, analysis, quality control, and food legislation. Research interests are related to virgin olive oil chemistry, quality and authenticity, saffron chemistry, authenticity and quality, antioxidant activity of plant extracts and constituents, new sources of targeted bioactive compounds (squalene, carotenoids, and phenols), and analytical procedures for their determination. She has published many research papers, review articles, and contributions to scientific books and encyclopedias on the above-mentioned topics. Currently, she is an associate editor in the European Journal of Lipid Science and Technology and chairs the COST ACTION FA1101 “Saffronomics.”

Editorial Advisory Board

xv

Jorge Welti-Chanes earned his degree in biochemical engineering (1976) and master of science in food engineering (1978) at Tecnolo´gico de Monterrey (ITESM, Mexico), later he moved to Spain to perform his doctoral studies in chemistry, in the area of food technology, obtaining his degree at the University of Valencia. He is currently the national director of graduate studies at School of Engineering and Sciences at Tecnolo´gico de Monterrey also is professor and researcher in the areas of biotechnology and food at the same institution. He started his academic activity in 1976 as a university professor of ITESM, has additionally been a full professor at the National Polytechnic Institute (IPN, Mexico) and the University of the Americas, Puebla, Mexico (UDLA). He has an experience of 37 years as a teacher and university researcher, 20 of which were spent in combination with the development of administration work in education, science and technology. In the UDLA, he was teaching in the Departments of Chemistry and Biology and Chemical Engineering and Food, in the latter was responsible for the leadership for a period of a year and subsequently became dean of the School of Engineering (1986–1988). From January 1989 to June 2002, he was an academic vice chancellor at UDLA. He has published 14 books and has over 200 scientific publications in refereed journals and books, has given more than 250 presentations at international conferences. He is an associate editor of the journals Food Engineering Reviews and Journal of Food Science and participates as a member of the editorial boards of Journal of Food Engineering and Current Opinion in Food Science. In May 2011, he received the Life Achievement Award by the International Association for Engineering and Food (IAEF), for his career as a researcher and academic worldwide, and in January 2014, the Romulo Garza Award from the Tecnolo´gico de Monterrey for the impact of their research work and as recognition for being one of the most productive researchers in the life of Tecnolo´gico de Monterrey. He has been the president of ISOPOW and IAEF and is the currently president of the International Society of Food Engineering (ISFE). Peter J. Wilde graduated in biophysics at the University of East Anglia in 1985 and has been researching the colloidal and interfacial properties of food systems at the Institute of Food Research (IFR) for over 25 years. IFR is the only publicly funded UK research institute that focuses on the underlying science of food and health to address the global challenges of food security, diet, and health, healthy aging, and food waste. IFR is the one of eight institutes that receives strategic funding from the Biotechnology and Biological Sciences Research Council (BBSRC). It also receives funding from government agencies and departments, the EU, charities, and industry, from the UK and overseas. Pete’s research expertise is the interfacial behavior of proteins and other surface active components in food relevant systems. The aim is to determine how the molecular and interfacial processes control the functionality of foams and emulsions. Currently, the functional aspects of his research have focused on improving the dietary impact of emulsified foods. These include fundamental studies on how interfacial layers control emulsion rheology to develop novel fat reduction strategies; the design of interfacial structures to control lipid digestion to promote satiety or the delivery of fat-soluble nutrients and drugs; and to determine the physico-chemical role played by the salivary film in perceiving fat content in emulsions. The impact of this research will be to aid the rational design of foods with enhanced nutritional benefits to address the global challenges of obesity, type 2 diabetes, and other major diet-related conditions.

This page intentionally left blank

HOW TO USE THE ENCYCLOPEDIA All articles in the encyclopedia are arranged alphabetically as a series of entries.

See also: Anemia: Causes and Prevalence; Anemia: Prevention and Dietary Strategies; Iron: Biosynthesis and Significance of Heme; Iron: Physiology of Iron.

1. Contents Your first point of reference will likely be the contents. The complete contents list appears at the front of each volume providing volume and page numbers of the entry. We also display the article title in the running headers on each page so you are able to identify your location and browse the work in this manner. 2. Cross-references The majority of articles within the encyclopedia have an extensive list of cross-references that appear at the end of each article, for example:

3. Index The index provides the volume and page number for where the material is located, and the index entries differentiate between material that is a whole article; is part of an article, part of a table, or in a figure. 4. Contributors A full list of contributors appears at the end of volume 5.

xvii

This page intentionally left blank

INTRODUCTION Until a few decades ago, virtually all known health effects of foods were related to their content of essential nutrients. The clinical description of most diet-related illnesses mirrored the signs of essential nutrient deficiencies, such as pellagra, beriberi, and others. Consequently, the key public health concern regarding diet was ensuring that everyone consumed enough food. It was only in the past 50 years that large-scale epidemiological observations began to associate chronic diseases like diabetes and cardiovascular disease with nonessential diet constituents such as saturated fat, fiber, and cholesterol. Taking advantage of the emergence of digital informatics, these studies were able to manipulate increasingly large sets of data and provide, for the first time, a picture of the secular changes in the health of large populations and its association with what they ate regularly. These findings progressively shifted the concern from eating enough to avoiding excessive consumption of certain foods. Eating enough was replaced by eating well. But it turned out that defining how to eat well is far more complex than defining minimum needs of essential nutrients. First, there is no single paradigm to study those relationships, given the wide variety of biological mechanisms and the long exposures involved. Second, many of the experimental models used to define essential nutrient needs are not applicable to the study of long-term effects of diets in free-living populations. And it is now clear that experiments with isolated dietary compounds do not reflect the actual effects of the complex food matrix we consume daily. Finally, while the discovery of essential nutrients and their role in health was the domain of a few specialties speaking a common language (primarily biochemists and physiologists), the study of the long-term effects of whole diets in humans must of necessity involve epidemiologists, social and behavioral scientists, food scientists, clinicians, policy experts, etc., making far more difficult the development of consensus and foundational concepts. It is thus not surprising that today we have still not achieved a stable consensus on how to eat ‘well.’ Furthermore, while few nonscientists would care about the minimum requirement of a vitamin to sustain life, there are plenty of opinions among nonscientists on how to eat ‘well.’ Our goal in preparing this encyclopedia has been to contribute to the understanding of that complex diet–health relationship by providing a multidisciplinary, integrative and accurate source of information. We aim to serve the needs not only of established and in-training scientists, but also of the increasingly important group of professionals who are key to disseminate and sustain the practice of science: journalists, science writers, science administrators, fund raisers, donors, and policymakers. In preparing this work, we had the enormous advantage of working with one of the publishers with the most extensive expertise in major reference works, Elsevier. This first edition builds on the impressive breadth of knowledge of over 922 authors and on the tireless work of our editorial advisory board. We are very grateful to all of them. Benjamin Caballero Paul Finglas Fidel Toldra´

xix

This page intentionally left blank

VOLUME 2 TABLE OF CONTENTS Editors-in-Chief Editorial Advisory Board

v vii

How to use the Encyclopedia

xvii

Introduction

xix

Chemometrics

1

F Marini

Cherries (Prunus Spp.): The Fruit and Its Importance

10

W Loescher

Chilled Foods: Effects on Shelf-life and Sensory Quality

14

D Bermu´dez-Aguirre and J Welti-Chanes

Chilled Foods: Modified Atmosphere Packaging

19

LM Cunha and SC Fonseca

Chilled Foods: Packaging Under Vacuum

23

M Rossi

Chilled Foods: Principles

28

GG Amador-Espejo and ME Ba´rcenas Pozos

Chlorophyll

37

C Yilmaz and V Go¨kmen

Cholecalciferol: Properties and Determination

42

AK Hewavitharana and FP Gomes

Cholesterol: Absorption, Function and Metabolism

47

V Vucˇic´ and Z Cvetkovic´

Cholesterol: Factors Determining Blood Cholesterol Levels

53

Z Rasic-Milutinovic, G Perunicic-Pekovic, D Jovanovic, N Simovic, Z Gluvic, D Ristic-Medic, and M Glibetic

Cholesterol: Properties, Processing Effects, and Determination

60

T Dinh and L Thompson

Choline: Physiology

70

SH Zeisel

Choline: Properties and Determination

73

MM Phillips

Chromatography: Combined Chromatography and Mass Spectrometry

79

Z Zhang, X Hu, and P Li

Chromatography: Focus on Multidimensional GC

85

C Cordero, C Cagliero, E Liberto, B Sgorbini, P Rubiolo, and C Bicchi

Chromatography: High-Performance Liquid Chromatography

93

H Gika, G Kaklamanos, P Manesiotis, and G Theodoridis

xxi

xxii

Volume 2 Table of Contents

Chromatography: Supercritical Fluid Chromatography

100

C Galea, D Mangelings, and YV Heyden

Chromium: Physiology

108

JB Vincent

Chromium: Properties and Determination

114

JB Vincent

Cider (Cyder; Hard Cider): The Product and Its Manufacture

119

E Coton, M Coton, and H Guichard

Cirrhosis

129

S Honigbaum, J Lucas, and KB Schwarz

Citrus Fruits

136

AC Matheyambath, P Padmanabhan, and G Paliyath

Clostridium botulinum

141

A Harris

Clostridium: Occurrence and Detection of Clostridium perfringens

146

R Labbe´ and V Juneja

Clostridium: Food Poisoning by Clostridium perfringens

149

K Miyamoto and M Nagahama

Clostridium: Occurrence and Detection of Clostridium botulinum and Botulinum Neurotoxin

155

JW Austin

Cobalamin (Vitamin B12): Metabolism and Disorders

160

E Andre`s and N Dali-Youcef

Cobalt: Properties and Determination

166

F Ca´mara-Martos and R Moreno-Rojas

Cobalt: Toxicology

172

F Ca´mara-Martos and R Moreno-Rojas

Cocoa: Composition and Health Effects

179

DD Mellor

Cocoa: Production, Chemistry, and Use

185

A Caligiani, A Marseglia, and G Palla

Codex Alimentarius

191

I Stankovic

Codex Alimentarius Commission: Role in International Food Standards Setting

197

V Kotwal

Coenzymes and Cofactors

206

RB Rucker and W Chowanadisai

Coffee: Analysis and Composition

225

MC Cid and M-P de Pen˜a

Coffee: Decaffeination

232

AS Franca

Coffee: Health Effects

237

R Tofalo, G Renda, R De Caterina, and G Suzzi

Coffee: Types and Production

244

LR Batista, SM Chalfoun de Souza, CF Silva e Batista, and RF Schwan

Colon: Diseases and Disorders

252

R Arbizu and S Nurko

Colon: Structure and Function R Arbizu and S Nurko

259

Volume 2 Table of Contents

Colors: Health Effects

xxiii

265

D Villan˜o, C Garcı´a-Viguera, and P Mena

Colors: Properties and Determination of Natural Pigments

273

A Giuliani, L Cerretani, and A Cichelli

Colors: Properties and Determination of Synthetic Pigments

284

E Diacu

Condensed Milk

291

SD Kalyankar, MA Deshmukh, CD Khedkar, SS Deosarkar, and AR Sarode

Consumer Protection Legislation

296

K Purnhagen and B van der Meulen

Controlled Atmosphere Storage: Applications for Bulk Storage of Foodstuffs

301

Z Escobedo-Avellaneda and J Welti-Chanes

Controlled Atmosphere Storage: Effect on Fruit and Vegetables

308

A Valdez Fragoso and H Mu´jica-Paz

Convenience Food

312

TA Brunner

Cooking: Domestic Techniques

316

AJ Rosenthal

Copper: Physiology

321

J Bertinato

Cream: Clotted Cream

327

RS Chavan, A Kumar, and S Bhatt

Cream: Types of Cream

331

SS Deosarkar, CD Khedkar, SD Kalyankar, and AR Sarode

Cured Foods: Health Effects

338

J Ruiz-Carrascal

Cystic Fibrosis, Nutrition in

343

S Sabharwal

D

345

Dahi

345

CD Khedkar, SD Kalyankar, SS Deosarkar, and AM Patil

Dairy Products: Dietary and Medical Importance

352

F Visioli

Date Palm: A Wealth of Healthy Food

356

KM Farag

Diarrheal Diseases

361

Z Bhutta and S Syed

Dietary Exposure Assessment

373

D Arcella, F He´raud, and M Gilsenan

Dietary Fiber: Bran

378

A Kamal-Eldin

Dietary Fiber: Determination

383

R Mongeau and SPJ Brooks

Dietary Fiber: Energy Value

392

SPJ Brooks and R Mongeau

Dietary Fiber: Physiological Effects IT Johnson

400

xxiv

Volume 2 Table of Contents

Dietary Fiber: Properties and Sources

404

R Mongeau and SPJ Brooks

Dietary Practices

413

AFG Cicero and T Stallone

Dietary References: US

418

J Dwyer and NJ Armstrong

Dietary Surveys: National Food Intake

432

MC Ocke´, CTM van Rossum, and EJ de Boer

Drying: Effect on Nutrients, Composition and Health

439

SV Crowley and JA O’Mahony

Drying: Physical and Structural Changes

446

SV Jangam, AS Mujumdar, and B Adhikari

Drying: Principles and Types

456

JM Barat and R Grau

E

463

Eating Disorders

463

CC Schreyer, S Makhzoumi, JW Coughlin, and AS Guarda

Eggs: Composition and Health Effects

470

ML Fernandez and CJ Andersen

Eggs: Use in the Food Industry

476

CG Belyavin

Elderly: Nutrition Requirements

480

R Chernoff

Emerging Foodborne Enteric Bacterial Pathogens

487

SJ Forsythe

Emulsifiers: Types and Uses

498

R Miller

Energy Metabolism

503

JA Coss-Bu and NM Mehta

Energy: Intake and Energy Requirements

511

DJ Millward

Enteral Feeding

519

DL Waitzberg and RS Torrinhas

Enzymes: Analysis and Food Processing

524

T Haertle´

Enzymes: Functions and Characteristics

532

D Talens-Perales, J Marı´n-Navarro, and J Polaina

Escherichia coli and Other Enterobacteriaceae: Food Poisoning and Health Effects

539

JL Smith and PM Fratamico

Escherichia coli and Other Enterobacteriaceae: Occurrence and Detection

545

Essential Oils: Isolation, Production and Uses

552

S Fanning, L Rogers, K Power, and PO´ Gaora

CM Cook and T Lanaras

Essential Oils: Properties, Composition and Health Effects

558

G Buchbauer and IM Wallner

Ethnic Foods OI Bermudez

563

Volume 2 Table of Contents

Extrusion Cooking: Chemical and Nutritional Changes

xxv

569

JAG Areˆas, CM Rocha-Olivieri, and MR Marques

Extrusion Cooking: Principles and Practice

576

L Moscicki

F

581

Famine, Hunger, and Undernourishment

581

R Mila`-Villarroel, C Homs, J Ngo, J Martı´n, and M Vidal

Fat Replacer

589

RS Chavan, CD Khedkar, and S Bhatt

Fats: Classification and Analysis

596

M Narva´ez-Rivas and M Leo´n-Camacho

Fats: Production and Uses of Animal Fats

604

SB Smith and DR Smith

Fatty Acids: Determination and Requirements

609

M Narva´ez-Rivas and M Leo´n-Camacho

Fatty Acids: Essential Fatty Acids

615

B Lands

Fatty Acids: Fatty Acids

623

S Petrovic and A Arsic

Fatty Acids: Metabolism

632

PC Calder

Fatty Acids: Trans Fatty Acids

645

AH Lichtenstein

Fermented Foods: Composition and Health effects

649

D Ansorena and I Astiasara´n

Fermented Foods: Fermented Meat Products

656

F Leroy and L De Vuyst

Fermented Foods: Fermented Milks

661

CD Khedkar, SD Kalyankar, and SS Deosarkar

Fermented Foods: Fermented Vegetables and Other Products

668

R Di Cagno, P Filannino, and M Gobbetti

Fermented Foods: Origins and Applications

675

A Bevilacqua, M Sinigaglia, and MR Corbo

Fermented Foods: Use of Starter Cultures

681

PM Malo and EA Urquhart

Fish Oils: Composition and Health Effects

686

C Jacobsen

Fish Oils: Production and Properties

693

AK Carvajal and R Mozuraityte

Fish: Dietary Importance and Health Effects

699

HK Mæhre, I-J Jensen, and K-E Eilertsen

Fish: Fish in the Human Diet

706

B Blakistone, R Kleiner, and J McGuire

Fish: Processing

710

SP Aubourg

Flavor Enhancers: Characteristics and Uses D Baines and M Brown

716

xxvi

Volume 2 Table of Contents

Folic acid and Folates: Physiology and Health Effects

724

C Wittho¨ft and M Hefni

Food Additives: Classification, Uses and Regulation

731

GA Blekas

Food Allergies

737

SL Taylor and JL Baumert

Food Allergies: Occurrence and Analysis

743

S Sforza and B Prandi

Food and Agriculture Organization of the United Nations E Casadei and J Albert

749

Chemometrics F Marini, University of Rome “La Sapienza”, Rome, Italy ã 2016 Elsevier Ltd. All rights reserved.

Introduction Chemometrics can be defined as the chemical discipline, which makes use of mathematical, statistical, and logical methods to extract meaningful information from experimental data and to optimize processes and/or products. It is evident from this definition that it accompanies all stages of the chemical measurement process, from sampling to interpretation passing through the definition of the optimal experimental conditions and data collection and processing. Even if it finds its roots in analytical chemistry, chemometrics is highly interdisciplinary and its domain of application is becoming wider and wider. However, since the very beginning, food-related issues have constituted an important field of application of chemometric techniques. Indeed, foodstuffs are rather complex matrices and their authentication often relies on a holistic characterization through the measurement of different chemical indexes or through the recording of whole instrumental fingerprints. The construction of traceability models for the verification of labeling compliance of products with a designated origin, the monitoring and control of food production processes, the correlation of volatile compound profiles with the sensory evaluation of a trained panel, and the indirect quantification of one or multiple analytes based on cheap and rapid spectroscopic techniques are only a few emblematic examples of application where the use of chemometrics is not only advisable but also necessary.

wavelengths or detector intensities at different retention times, in the case of spectroscopic or chromatographic profiles, respectively). The resulting data are then multivariate and each sample may be described by a numerical row vector xi, collecting the results of all the measurements performed:
 xi ¼ xi1 xi2 xi3 . . . xip [1] p being the total number of variables. Accordingly, when more than a single sample is analyzed, the data can be arranged into a matrix X, having as many rows as the number of samples (n) and, as already stated in eqn [1], as many columns as the number of variables (p): 0 1 x11 . . . x1p B C C [2] X¼B @ ⋮ O ⋮A xn1

...

xnp

Each row of the matrix X, then, contains the results of the experimental measurement performed on a particular sample, while each column is made of the values of a single variable along all the samples. One of the advantages of the matrix representation in eqn [2] is that it has a geometric counterpart: Each variable can be thought as an axis in a (hyper-)space having p dimensions so that the value of the matrix element xij represents the coordinate of the ith object along the jth direction. Accordingly, each sample can be described as a point in the p-dimensional space (see Figure 1).

Chemical Data: Types and Representation

Experimental Design

As anticipated, the main goal of chemometrics is to extract useful and meaningful information from chemical data. Therefore, prior to illustrating the chemometric toolbox with all its plethora of methods, it is essential to understand the data, which constitute the basis for modeling. A chemical system is usually characterized and described by a set of measured or calculated variables, which can be either quantitative or qualitative in nature. In the former case, variables are defined on a numerical (interval or ratio) scale, can undergo any kind of algebraic operation, and can assume an infinite (continuous) set of values; examples of these types of indices are pH, concentration, and temperature. On the other hand, qualitative variables may assume only a discrete set of values, which are often categorical or expressed in the form of attributes: the presence or absence of a constituent, genuine/adulterated, and sensory panel evaluation on a 1–5 scale are all examples of such kind. Usually, one variable only is not enough to characterize a chemical system, so that multiple descriptors are recorded on the same sample: these may come from different instruments or analytic procedures (e.g., total acidity, peroxide number, and concentration of iron or calcium) or be acquired with the same platform (e.g., absorbances at different

Although this aspect is often still neglected by many chemists, chemometrics enters the analytical process long before the data processing and interpretation stage, on one hand as it is necessary to sketch the most appropriate sampling strategies in order to have representativeness and to meet specific requirements (e.g., a target accuracy or precision) and, on the other, as the quality of the data obtained and the possibility of retrieving the information sought strongly rely on a careful experimental planning. Indeed, as already pointed out by Fisher, “Statistical procedure and experimental design are only two different aspects of the same whole, and that whole comprises all the logical requirements of the complete process of adding to natural knowledge by experimentation.” By the term experimental design, one indicates the family of techniques whose aim is to identify a parsimonious set of experimental conditions. The aim of experimental design techniques is (a) to understand the effect of controlled variables (factors) on one or more responses (often with the final intent of optimization) and (b) to define an empirical model (response surface) for the relation between the responses (dependent variables) and the factors (independent variables), which may be used either to support the optimization process

Encyclopedia of Food and Health

http://dx.doi.org/10.1016/B978-0-12-384947-2.00779-0

1

2

Chemometrics

8 x2 = [1 3 7]

Variable 3

6

x1 = [4 5 2] 4 2 0 6 4

4 Variable 2

3 2

2 1

0 0

Variable 1

Figure 1 Geometric representation of data vectors as points in the multidimensional space of the variables.

Selection of factors and responses

Definition of experimental domain Screening designs Discard unimportant factors Selection of the strategy

Simultaneous designs

Sequential designs

Two level factorial designs

Selection of optimum

Interactions

No interactions

Multilevel factorial designs

Multilevel OVAT

Selection of optimum

Selection of optimum

Figure 2 Flowchart of experimental design.

defined in point (a) or for prediction purposes. In both cases, a systematic and efficient mapping of the experimental domain, at the same time looking for the minimum number of experiments to be performed, is sought. Given the scopes summarized earlier, the family of experimental designs encompasses a wide range of techniques of increasing complexity, as evidenced by the pipeline shown in Figure 2. Screening designs are the most parsimonious ones, as they include only the experiments needed to estimate in a semiquantitative fashion the main effect of the investigated factors, not taking into account the possible interactions among variables. A linear relation is assumed between the response(s) y and each of the f controlled factors: y ¼ b0 þ b1 x1 þ b2 x2 þ . . . þ bf xf

[3]

where b0 is an offset term, corresponding to the predicted response in the center of the experimental domain, while

b1. . .bf are estimates of the effect of the factors 1. . .f on y. A direct consequence of the assumption of linearity is that only two levels are investigated for each factor and the resulting strategies are on one hand the so-called saturated or supersaturated factorial and on the other the Plackett–Burman designs. Screening designs are normally used to have information on which of possible factors may be discarded as very likely unaffecting the response. After screening, it is necessary to choose the strategy that may be more appropriate for the specific problem: Indeed, there are two options available, the factorial approach and the sequential approach. The sequential strategy is most suited when there are only a few factors, the goal is optimization of (possibly) a single response, and there is no interest in calculating the response surface. It operates by selecting a small number of initial experiments and then iteratively choosing the coordinates for the successive experiment based on the coordinates and the responses of the previous ones, up to when a stopping/optimality criterion is met. On the other hand, in the factorial approach, the whole set of

Chemometrics experiments to be conducted is defined a priori, before any of those is actually carried out: This approach is in general to be preferred, but it is for sure the one to be adopted when the effect of controlled factors on multiple responses has to be investigated and also when modeling is the purpose. In this context, two level factorial designs, that is, factorial designs in which each factor is controlled only at two values, allow to assess the main effect of the factors and at the same time to verify the presence of interactions. Indeed, they assume a linear model with mixed terms, which, in the case of two factors, takes the form y ¼ b0 þ b1 x1 þ b2 x2 þ b12 x1 x2

[4]

where b12 is the term that accounts for the interaction between factor 1 and factor 2, while all other terms have the same meaning as in eqn [3]. Although such mathematical assumption is often too approximate to lead to reliable estimation of the main effects and interaction terms or to accurate predictions of the response values, still, two level factorial designs allow to have a semiquantitative understanding of the relation between dependent and independent variables and to assess the presence or absence of interactions. When the linearity assumption is released, so that nonlinear (almost always second- or third-order polynomial) relationships are modeled, factors cannot be controlled at two levels only anymore, so that multilevel designs are needed. The use of multilevel designs, the most famous of which are the central composite, the Box-Behnken, and the Doehlert ones, together with the family of optimal designs (D-optimal, A-optimal, and so on), allows to better approximate the response surface, which, in the case of three factors, usually takes the form y ¼ b0 þ b1 x1 þ b2 x2 þ b3 x3 þ b12 x1 x2 þ b23 x2 x3 þ b13 x1 x3 þ b11 x21 þ b22 x22 þ b33 x23

[5]

where b11, b22, and b33 account for the effect of the squared factors and all the other terms have the same meaning as in eqns [2] and [3]. Specific designs may also be used when the factors to be investigated are the constituents of a mixture or when they are qualitative variables.

Exploratory Analysis Exploratory analysis is a strategy/philosophy that makes use of a variety of techniques to summarize the main characteristics of a data matrix in an easy-to-understand form, often with visual graphs, without using a statistical model or having formulated a hypothesis. Its main purposes are to maximize insight into a data set, to uncover underlying structure, to extract important variables, to detect outliers and anomalies, to test underlying assumptions, and to develop parsimonious models. All these characteristics suggest that exploration is and should always be the first step in data analysis, even when the final goal is some kind of predictive modeling. According to the definitions reported earlier, exploratory data analysis is normally carried out by means of two families of techniques: projection methods and clustering algorithms. The need for the projection method is evident when thinking that, in the situations when many variables are measured on

3

the samples, which is what normally occurs with chemical systems, graphical inspection of the data in the original p-dimensional space described in the section ‘Chemical Data: Types and Representation’ may result difficult or even unfeasible. Indeed, as the name suggests, projection methods operate by projecting the data on a relevant subspace, which can be used to summarize the data conveniently and explore the data set using plots and figures, in order, for instance, to find patterns, relation, differences, and similarities between objects and variables. The most famous and used multivariate projection method is for sure principal component analysis (PCA), whose aim is to find the best low-rank approximation of the original data in a least squares sense. This is equivalent to state that PCA identifies those directions in the multivariate space, which in turn account for the maximum explained variance and which are mutually orthogonal, and projects the original data onto this reduced subspace. Mathematically, the PCA model can be expressed by the following equation: T ¼ X P

nF

np pF

[6]

where T is the matrix collecting the coordinates (scores) of the sample onto the new reduced (F usually  p) set of variables (called principal components), while the matrix P contains the coefficients of the projection (loadings). The same relation can be rearranged as in eqn [7], to show the so-called bilinear nature of the PCA decomposition: ^ þ E ¼ TPT þ E X¼X

[7]

where the matrix E (residuals) accounts for the portion of the variability originally present in the data, which is not explained by the PCA model and the hat (^) indicates that X is approximated by the model. Indeed, bilinear modeling, that is, the approach that assumes that a data matrix X may be approximated by the product of two low-rank matrices, one representing the coordinates of the samples onto a reduced subspace and the other expressing the variable contribution to the definition of the subspace, is one of the core elements of chemometrics. Due to its characteristics, PCA represents the optimal tool to perform exploratory data analysis, as it can provide information about the relationship among samples and among variables, together with the possibility of noise filtering and outlier detection. Graphical representation of the samples in the principal component space (scores plot) allows to identify the presence of clusters or groups in the data or to evidence some trends; on the other hand, correlation between the experimental variables or irrelevant descriptors may be assessed by visually inspecting the elements of the loadings matrix (loadings plot). An example of the application of PCA in the context of food analysis may be observed in Figure 3, where the scores and the loadings plots for a problem involving the authentication of honey samples are shown. In particular, the data set was made on the results of 15 chemical analyses on 73 honey samples coming from different botanical origin (sulla, heather, eucalyptus, chestnut, honeydew, and wildflower). Inspection of the scores plot in Figure 3(a) evidences the presence of different clusters in the data: If no information about the origin of samples were available, one would have stated that there were seven different groups of honey; however, integrating the scores plot with the information available,

4

Chemometrics

4 3

PC 2 (26.97%)

2 1 0 honeydew eucalyptus chestnut sulla heather wildflower

−1 −2 −3 −4 −4

−3

−2

−1

0

1

2

3

4

PC 1 (39.08%)

(a)

0.5 Conductivity 0.4

Color Specific rotation

pH

0.3

PC 2 (26.97%)

0.2

Free acidity Total acidity

0.1 % DP2 0

HMF

13C/12C(proteins) Moisture

−0.1

%dextrose Lactones

Diastase −0.2 13C/12C(whole)

−0.3

%fructose −0.4 (b)

−0.6

−0.4

−0.2

0

0.2

0.4

0.6

PC 1 (39.08%)

Figure 3 Principal component analysis of honey data set. (a) Scores and (b) loadings plots.

it is possible to affirm that there are six clusters, one for each of the different floral origin of the samples and that the group corresponding to honeydew is further split in two subclusters. Moreover, in general, it is also possible to see how the clusters corresponding to multifloral origin (honeydew and wildflower), which are separated from the unifloral ones along PC2, are less homogeneous, as it could be expected. Information about the variables can, instead, be obtained by looking at the loadings plot reported in Figure 3(b): At first, one can observe that dextrose and lactones are correlated, as well as color and specific rotation or free and total acidity, as they fall very close to one another in the PC space; on the other hand, the carbon isotope ratio in the proteic fraction does not seem to contribute much to the model as its loadings are almost zero. Lastly, comparison of the scores and the loadings plot allows to interpret the differences observed among samples in terms of the original variables. As stated, unifloral honeys are separated from multifloral ones along the second principal component: looking at the loadings plot, this means that the

former are characterized by a higher fructose content and carbon isotope ratio, at the same time showing a lower pH, conductivity, color, and specific rotation. On the other hand, wildflower is separated from honeydew along PC1, and the same component differentiates also among the various unifloral honeys: this means that in going from honeydew to wildflower or from chestnut to heather, there are a systematic increase of dextrose, lactones, free and total acidity, and hydroxymethylfurfural (HMF) and a corresponding decrease of diastase and DP2. The second family of techniques normally used for exploratory data analysis are clustering algorithms. As the name suggests, the aim of clustering techniques is to look for the presence of groups (usually of samples, but it is possible to apply the same concepts to variables or to both) in the data, based on the calculation of some kind of similarity or dissimilarity index. Indeed, the underlying idea is that object belonging to the same group will be similar to one another and dissimilar to the object belonging to other clusters, so that

Chemometrics the definition of a proper measure to quantify (dis-)similarity plays a key role for this family of methods. In many applications, dissimilarity is expressed in terms of a distance measure in the multivariate space, but other indexes may also be used. Once the proper way of quantifying (dis-)similarity is chosen, clustering may be carried out according to two different approaches: hierarchical and partitional. In hierarchical clustering, as the name suggests, objects are ranked from the most similar to the most dissimilar in a hierarchical fashion, and the main result is a graphical representation called a dendrogram. In particular, at the beginning of the hierarchical clustering procedure, each sample is considered a cluster per se; at each step, the two most similar clusters are merged together and the procedure is continued until all objects are joined into a single group. This approach is called agglomerative (bottom up) and it is the most widely used: hierarchical clustering may also be carried out in a divisive way (top down), by starting from a single cluster and iteratively arriving to as many groups as the number of objects, but this approach is more computationally intensive and, therefore, it is rarely adopted. Here, it must be stressed that, whatever the approach, hierarchical clustering does not provide a unique grouping: partitioning of object into clusters may be achieved only by cutting the dendrogram at a specific level, which usually corresponds to a large distance among the merged groups. As an example, the dendrogram corresponding to agglomerative hierarchical clustering for the honey data set already presented in the case of PCA is reported in Figure 4. The dendrogram in Figure 4 shows very clearly the presence of clusters among the data, and also, as already evidenced by PCA, the groups of honeydew and wildflower samples are less homogenous than the other ones (their final merging distance is higher). Moreover, it can be also seen how the identified number of clusters would vary, depending on the level at

5

which the dendrogram be cut: If one chooses a threshold distance of 8 (dashed line), then honeydew samples would be seen as two different groups and a total number of seven clusters would be identified; instead, if a threshold distance of 10 (continuous line) be adopted, then the number of detected groups would be six. On the other hand, partitional clustering attempts to directly decompose the data into a predefined number of groups by minimizing some criterion, which should reflect the local structure. The most famous member of this family of algorithm is k-means where the criterion function is the sum of the squared distances of each sample to its nearest cluster centroid: E¼

XN i¼1

jjx i  mCðxi Þ jj

2

[8]

where mCðxi Þ is the vector collecting the coordinates of the centroid of the cluster closer to xi.

Calibration When dealing with food analysis or characterization, exploratory data analysis, although necessary, may not be sufficient, as many problems may require the prediction of one or more quantitative or qualitative properties of the samples. Calibration is the use of empirical data and prior knowledge for determining how to predict quantitative information Y from available measurement X via some mathematical transfer function f: ^ þ EY ¼ f ðX Þ þ EY Y¼Y

[9]

where the matrix of residuals EY collects the variance in Y not accounted for by the regression model, that is, the difference between the actual values of Y and those predicted by the

25

Distance

20

15

10

5

0

DDDDDDDDDDDWWWWWWWWWWWWWHHHHHHHHHHHHE EEEE EEEEE EECCCCCCCCCCCCCSSSSSSSSSSSS Sample Index

Figure 4 Dendrogram resulting from hierarchical clustering of honey data. Dashed and dotted lines (corresponding to distance values of 8 and 10, respectively) indicate two possible threshold values, which may be used to cut the dendrogram in order to define sample grouping.

6

Chemometrics

model (Yˆ). In food analysis, calibration plays a key role as quite often, the direct measurements of the properties of interest may not be feasible or it may be too expensive, time-consuming, or inaccurate, so that quantification is usually based on secondary measures (e.g., chromatographic peak area, absorption or emission intensity, current, and voltage). In this context, the simplest relation between the experimental measurements collected in X and the properties to be predicted is a linear one: ^ ¼ XB Y

[10]

where B is a matrix of the so-called regression coefficients, which constitute the model parameters and uniquely define the mathematical relation. In particular, each column of the regression coefficient matrix collects the weights associated with the independent variables for the prediction of the corresponding column of the Y matrix, and this information can be used also for the sake of interpretation. Indeed, in principle, the larger the absolute value of the coefficient bij (the ith row and jth column element of B), the higher the contribution of the ith X-variable to the prediction of the jth response. However, particular care should be taken when inspecting the values of B, as their magnitude directly reflects the scales of both X and Y. Moreover, when dealing with instrumental fingerprints, interpretation can be further hindered by the presence of overlapping signals: As an example, one may consider the case when a spectroscopic measurement is used to predict the concentration of an analyte in a solution. If no interferent is present, the regression coefficients would resemble the spectrum of the pure analyte; however, if the solution contains also an interferent, whose spectrum partially overlaps with that of the analyte, the regression vector no longer looks like the pure spectrum because negative parts and shifts in position of peak maximum are introduced. Accordingly, there may be cases when a negative regression coefficient is correctly obtained for a variable that is positively correlated with the response. Operationally, the most straightforward way of estimating the regression coefficients in eqn [10] is provided by multiple linear regression (MLR). MLR is the multivariate generalization of the univariate least squares method; the values of the coefficient are calculated as the ones, which minimize the sum of squared residuals:     ^ 2 min jjEY jj2 ¼ min jjY  Yjj B

B

  ^ ¼ TV ) V ¼ TT T 1 TT Y Y

[13]

V being the matrix of regression coefficients relating Y to T. By combining eqns [13] and [7], it is possible to obtain the model coefficients directly in terms of the original independent matrix X: ^ ¼ TV ¼ XPV ¼ XBPCR ) BPCR ¼ PV Y

[14]

Due to the nature of PCA decomposition, PCR provides a reliable answer to the drawback discussed earlier for MLR, even if it may not provide the best low-rank representation of the data to be used for calibration purposes. Indeed, principal components are calculated as those directions in space that account for the maximum variance in the data set; however, when multiple sources of spurious (irrelevant) variation are present in the data set, the directions of maximum variance may not account for the correlation with the responses to be predicted. On the other hand, partial least squares regression (PLS) makes active use of the information in Y to define the low-rank subspace onto which the data should be projected. In PLS, both the X- and the Y-blocks are decomposed in a bilinear fashion, and the axes of the low-dimensional subspace (called latent vectors) are chosen so that the scores of Y (U) have maximum covariance with those of X (T) and are linearly dependent, through what is called the inner relation (last one of the following equations): T ¼ XR Y ¼ UQT þ EY U ¼ TC

[15]

[11]

where R is a matrix of weights, governing the projection of the X-block, Q are the loadings for the Y-block, and C is a diagonal matrix of coefficients. Also, in the case of PLS, it is possible to combine the equations in eqn [15] to obtain a matrix of regression coefficients directly relating Y to X:

[12]

Y ¼ UQT þ EY ¼ TCQT þ EY ¼ XRCQT þ EY ¼ XBPLS þ EY BPLS ¼ RCQT [16]

Accordingly, the matrix B is estimated as  1 BMLR ¼ X T X X T Y

orthogonal variables, represents a way of overcoming the previously mentioned limitations. In particular, principal component regression (PCR) is a calibration method that originated by directly combining a preliminary dimensionality reduction step by means of PCA with the calculation of a MLR model on the resulting scores. In mathematical terms, at first, the matrix X is decomposed into the product of scores and loadings according to eqn [7], and successively, the scores are used as predictors to build the MLR model:

Unfortunately, even if MLR is the simplest method for linear regression, it is quite often inapplicable to the matrices resulting from food analysis and characterization. Indeed, in order for the term (XTX)1 in eqn [12] to be estimated accurately, the matrix X should meet some mathematical requirements, which are rarely fulfilled in problems involving instrumental fingerprinting of real-world samples: the number of samples should be lower than the number of predictors and the variables should be as uncorrelated as possible from one another. In this context, the bilinear approach introduced in the section ‘Exploratory Analysis’ for PCA, by involving the projection of the samples onto a low-dimensional space of

However, inspection of regression coefficients is not the only tool for interpretation, when dealing with PLS: the bilinear nature of the relations reported in eqn [15] suggests that scores and loadings plots also constitute a valid support to model understanding and diagnostics. Although, for all the methods described so far, a linear relationship between the property (or properties) to be predicted and the secondary measurements is assumed, this need not always be the case: the important is to have a defined calibration equation in order to be able to make predictions on future samples, so that several nonlinear algorithms have

Chemometrics been proposed in the literature to tackle with more complex functional dependencies. In this framework, since a detailed discussion would be far beyond the scope of the present article, it is just worth mentioning, among the various possibilities, kernel- or dissimilarity-based approaches, neural networks, or locally weighted regression.

Classification Food-related issues may not always call for a quantitative prediction, and rather, problems such as the authentication of a good, its quality control, and traceability (just to cite a few) involve the assessment of one or more qualitative properties. For instance, one may be interested in assessing whether a product is organically grown or not, or if a wine was produced in Italy, Spain, South Africa, or Chile, or, again, if a food will be good, acceptable, or bad according to consumer preferences. From a chemometric standpoint, all those methods, which deal with the possibility of predicting one or more qualitative responses on a set of samples, belong to the family of classification tools. Indeed, classification techniques aim at building models, which, based on the values of the measured variables, assign a sample to a category or class, the latter being a group of objects sharing similar characteristics. Accordingly, in the language of classification techniques, the wine authentication problem, cited earlier as an example, would involve four categories (‘Spain,’ ‘Italy,’ ‘South Africa,’ and ‘Chile’), while the three classes ‘good,’ ‘bad,’ and ‘acceptable’ would be considered for the food preference one. Since samples can be represented as points in the multivariate space of the variables, classification may be seen under a geometric perspective as the search for surfaces identifying regions of space where it is more likely to find objects belonging to a particular category. In this context, it is particularly useful to operate a distinction between two possible approaches: discrimination and class modeling. Discriminant techniques partition the space in as many regions as the number of categories in the data set, so that if an object falls in the region corresponding to a particular class, it is univocally assigned to it; as a consequence, each sample is predicted to belong to one and only one of the categories postulated by the problem. On the other hand, modeling techniques, as the name suggest, try to model each category independently on the others and operate by identifying a region of the multivariate space where it is likely to find samples from that particular class (the model space): If a sample falls within that region, it is accepted by the class model; otherwise, it is rejected. Accordingly, when more than one category is modeled, a sample can be accepted by only one class (and then be univocally assigned to it), by more than one (i.e., confused), or by none (and be considered an outlier). In the remainder of the section, discriminant and modeling approaches will be further discussed through the illustration of two widely used methods, respectively, partial least squares discriminant analysis (PLS-DA) and soft independent modeling of class analogies (SIMCA). PLS-DA is a discriminant classification technique based on the PLS algorithm already described in the section ‘Calibration,’ and it was introduced to overcome the limitations suffered by traditional methods such as linear (LDA) and quadratic (QDA)

7

discriminant analysis, in the presence of ill-conditioned X matrices. Indeed, the same kind of problems (high number of highly correlated variables), which make MLR unsuitable to build regression models, hinders the applicability of LDA and QDA for classification. Accordingly, after finding a suitable coding that allows to transform a classification problem into a regression one, the use of the PLS algorithm represents a solution to these drawbacks. In particular, PLS-DA is based on coding the information about class belonging into a binary dummy matrix Y having as many rows as the number of samples and as many columns as the number of classes: the matrix element yij will be equal to 1 if the ith sample belongs to class j or to zero if it does not. A PLS model is then built between the experimental matrix X and the dummy matrix Y, and classification is achieved on the basis of the predicted values Yˆ, usually via the introduction of a suitable threshold. An example of application is reported in Figure 5, where the use of PLS-DA on chromatographic data to discriminate olive oils from the PDO Sabina from other extra virgin oils is shown. In this case, since there are only two categories, due to the symmetry of the problem, the dummy matrix Y boils down to a vector y in which 1 indicates ‘Sabina’ and 0 ‘other oils.’ Classification is then achieved by setting a threshold of 0.5 to the predictions: all the samples for which the predicted response is higher than the threshold are classified as Sabina oils, while all the others are recognized as from other origins. Differently than what happens for discriminant methods, modeling techniques focus on capturing the similarity among samples belonging to the same category, rather than the differences between individuals from competing classes. Under many respects, they can be considered as outlier detection techniques, as their aim is to verify whether a sample fits the model of a particular category or not. In particular, SIMCA operates by describing the class-related variability in the experimental fingerprint using a PCA model of appropriate dimensionality: XG ¼ TG PTG þ EG

[17]

where the subscript G indicates that only the samples from category G are used to define the projection. Then, for each sample, an overall distance to the class model, measuring the extent of outlyingness, is defined as the combination of the distance to the model space (which is a function of the residuals) and the distance within the model space (which accounts for the distance of the sample scores to the origin of the PC space). Accordingly, if the distance to the model is below a prespecified threshold, the sample is accepted by the category; otherwise, it is rejected. When more than a single category is modeled, a straightforward way of representing the results of SIMCA is the so-called Coomans plot, which is shown in Figure 6 for the same data set used to exemplify PLS-DA. The axes of the Coomans plot represent the sample distances to the two investigated categories, and the thresholds used to define acceptance/rejection by the class models divide the plot in four different regions: Objects falling in the uppermost left region of the plot are univocally accepted by the class ‘Sabina,’ while those mapped onto the rightmost lower part are uniquely accepted by ‘other oils’; the samples falling in the lowermost left part of the plot are confused between the two categories, while those in the uppermost right regions are considered as outliers by both classes.

8

Chemometrics

1.2 Other origins Sabina

1

Y predicted

0.8 0.6 0.4 0.2 0 −0.2 −0.4

10

0

20

30 Sample Index

40

50

Figure 5 Illustration of how classification is accomplished in PLS-DA: samples are assigned to one or the other class based on the predicted y values and the green dashed line indicates the classification threshold. Accordingly, all samples except the two Sabina oils that fall below the threshold are correctly classified.

Distance to the model “other origins”

8 Other origins Sabina

7 6 5 4 3 2 1 0

0

1

2

3

4

5

6

7

8

Distance to the model “Sabina” Figure 6 SIMCA modeling of the olive oil data set: Coomans plot. The dashed lines indicate the acceptance thresholds used by the two class models.

A Continually Increasing Toolbox Although the topics presented in the previous sections cover most of the fields of application of chemometrics to food analysis and characterization, the chemometric toolbox is continuously evolving to match the increase in the complexity of the problems to be tackled and, at the same time, in the availability of high-throughput instrumentation. For instance, multiway and multiset resolution techniques may be used to extract chemically relevant profiles from data collected by means of hyphenated techniques or, in general, resulting from experiments where signals are recorded as a function of different sources of variability. On the other hand, hyperspectral image analysis techniques allow to extract both spatial

information (e.g., texture and homogeneity) and spectral information from the samples, while data fusion approaches combine information from multiple sources into a holistic characterization of the sample for both exploratory and predictive purposes. In general, one may affirm that chemometric techniques constitute an essential and valid tool for all of those who are involved at different levels in the characterization and analysis of foodstuff.

See also: Authenticity of Food; Food Fraud; Infrared Spectroscopy: Applications.

Chemometrics

Further Reading Bevilacqua M, Marini F, Biasioli F, and Gasperi F (2013) Advances in analysis of instrumental food sensory quality data. In: Kilcast D (ed.) Instrumental assessment of food sensory quality, pp. 313–352. London: Woodhead Publishing. Bevilacqua M, Nescatelli R, Bucci R, Magrı` AD, Magrı` AL, and Marini F (2014) Chemometric classification techniques as a tool for solving problems in analytical chemistry. Journal of AOAC International 97: 19–28. Brereton R (2009) Chemometrics for pattern recognition. New York, NY: Wiley. Bro R, van den Berg F, Thybo A, Andersen CM, Jørgensen BM, and Andersen H (2002) Multivariate data analysis as a tool in advanced quality monitoring in the food production chain. Trends in Food Science and Technology 13: 235–244. Brown SD, Tauler R, and Walczak B (eds.) (2009) Comprehensive chemometrics. Chemical and biochemical data analysis. Oxford: Elsevier. Forina M, Lanteri S, and Armanino C (1987) Chemometrics in food chemistry. Topics in Current Chemistry 141: 91–143.

9

Leardi R (2003) Chemometrics in data analysis. In: Lees M (ed.) Food authenticity and traceability, pp. 299–320. London: Woodhead Publishing. Leardi R (2008) Chemometric methods in food authentication. In: Sun D-W (ed.) Modern techniques for food authentication, pp. 585–616. New York, NY: Academic Press. Marini F (2013) Chemometrics in food chemistry. Oxford: Elsevier. Martens H and Martens M (2001) Multivariate analysis of quality: an introduction. New York, NY: Wiley. Martens H and Næs T (1991) Multivariate calibration, 2nd ed. New York, NY: Wiley. Munck L, Nørgaard L, Engelsen SB, Bro R, and Andersson CA (1998) Chemometrics in food science – a demonstration of the feasibility of a highly exploratory, inductive evaluation strategy of fundamental scientific significance. Chemometrics and Intelligent Laboratory Systems 44: 31–60. Oliveri P and Downey G (2012) Multivariate class modeling for the verification of foodauthenticity claims. Trends in Analytical Chemistry 35: 74–86.

Cherries (Prunus spp.): The Fruit and Its Importance W Loescher, Michigan State University, East Lansing, MI, USA ã 2016 Elsevier Ltd. All rights reserved.

Introduction: Cherry Taxonomy and Types Commercially, the two most important cherry species are sweet cherry (Prunus avium, L.) and tart cherry (Prunus cerasus L.), both tree fruits native to Southeastern Europe and Western Asia. They are closely related and graft-compatible and will hybridize to form interspecific (Duke) cultivars. Sweet cherry (diploid, with a base chromosome number of 8 and a somatic number of 16) probably originated between the Black Sea and the Caspian Sea, but it spread into Europe in ancient times. Tart cherry (tetraploid, with a base chromosome number of 16 and somatic chromosome number of 32) is native to the same areas as sweet cherry, and there is good evidence that crosses between Prunus avium and the ground cherry (Prunus fruticosa Pall) gave rise to tart cherry. There are other cherry species, but most, for example, Nanking cherry (Prunus tomentosa), have limited commercial value as fruits. Sweet cherries can be divided into two major types based on fruit characteristics. Heart-type cherries are ovoid or heartshaped with relatively soft flesh, often ripening early. Most of the commercially important cultivars, however, are of the Bigarreau type with firmer, crisp-fleshed fruit, ripening mid to late season. Fruit flesh may be red or yellow, and the skin may be dark (red to nearly black) or light (yellow-red to yellowwhite). Many sweet cherry cultivars grown throughout the world originated in Europe, but a number of important ones were selected or bred in local cherry districts. European cultivars grown in the United States are Napoleon (Royal Ann), Black Tartarian, Eagle, Early Purple, Early Rivers, Elkhorn, Hedelfingen, Knight’s Early Black, Lyon, and Schmidt. The cultivars Windsor, Van, Sam, Vista, Victor, Sue, Vega, Summit, and Stella were developed in Canada. Chinook and Rainier were developed in Washington. Bing, Lambert, Black Republican, Corum, and Hoskins were selected and developed in Oregon. Chapman, Burbank, Bush Tartarian, and the new cultivars Mona, Larian, Jubilee, Berryessa, and Bada originated in California. Recent introductions include Ulster and Hudson from New York and Angela from Utah. The most important sweet cherry cultivars in the Western United States, where over 80% of the US crop is produced, have been dark-fruited, crisp-fleshed cultivars: Bing (the leading cultivar in North America), Van, and Lambert. But others may be available because of their use as pollenizers or as the result of recent fresh market demand for large, light-colored, and crisp-fleshed fruits from cultivars like Rainier. Firmness, size, color, and soluble solids are all important market considerations, and growers in regions where summer rains are prevalent, for example, the eastern United States and Eastern Europe, are at a disadvantage because the main cultivars are often the softer-fleshed, rain cracking-resistant types, for example, Emperor Francis, Hedelfingen, and Schmidt. In these regions, light-fleshed cultivars, Rainier, Napoleon (Royal

10

Ann), Corum, and Emperor Francis, are best for making into maraschino cherries (because pigment is undesirable), but a few are nonetheless grown for the fresh market. Napoleon is also used for canning. Bing is mainly a fresh market cultivar, and Lambert is used both for canning and fresh market. Black Republican and other very firm, dark cherries are good for freezing. Tart cherry fruit are generally soft, juicy, and depressedglobose in shape, but colors may range from the Morello types with red to dark red flesh and juice to the Amarelle types with nearly colorless juice and flesh. Although new cultivars are being tested, there are only a few tart cherry cultivars commonly grown in North America, ranging from the light red Early Richmond, to the medium redskinned Montmorency, to the late dark red English Morello, but Montmorency is still the standard. In Western Europe, Schattenmorelle and Sternsbaer are common, but many others are grown in Russia, Slovenia, Romania, and Hungary. Most, unlike sweet cherry, are more or less self-fertile and generally do not require pollenizers. Almost all of those grown in the United States and Western Europe are harvested mechanically and sold for processing, primarily as a frozen or canned ingredient for use in manufactured food products such as pies, but more recently as a dried fruit product, and in Europe and other areas, there have been uses developing for juice, liqueur, and marmalade production and combinations with yogurt.

Production Areas Turkey, the United States, Iran, and Russia are large producers of both sweet and tart cherries (FAO Statistics, Table 1). In some areas of northern Europe, tart cherries are, after apples, the second most important fruit grown. Otherwise, within Europe, tart cherry production is concentrated in Eastern Europe, Slovenia, Hungary, and Romania, while sweet cherry production is more common in Western Europe, Italy, Switzerland, France, and Spain. Sweet cherry production is increasing in the Southern Hemisphere, New Zealand, Australia, and Chile, for fresh shipments to northern markets in their winter season. In the United States, sweet cherry cultivation has also been increasing, with production mostly in the west, not only in Washington but also in Oregon and California. Most US tart cherry production occurs near the Great Lakes, primarily in Michigan with 70–75% of the US crop, which with New York, Wisconsin, and Pennsylvania totals 90–95% of the US crop. Average world production values for sweet and tart cherries are now approximately 1 200 000 and 2 200 000 metric tons, respectively. Wide annual supply fluctuations, especially regionally, in both sweet and tart cherries characterize production and create high risks in product availability and price change for producers, processors, and marketers. Annual US

Encyclopedia of Food and Health

http://dx.doi.org/10.1016/B978-0-12-384947-2.00138-0

Cherries (Prunus spp.): The Fruit and Its Importance Table 1 Country Turkey The United States Iran Italy Spain Chile Uzbekistan Syria Ukraine Russia Romania Greece Poland Austria China France Germany Lebanon Serbia Bulgaria

11

Average values and yields of tart (sour) and sweet cherries (top 20 producers) over the years 2010–12 Sour cherries average annual yield (MT)

Sour cherries value ($1000)

Sweet cherries average annual yield (MT)

Sweet cherries value ($1000)

Total average yield (MT)

Total average value ($1000)

188 388 179 667

115 033 109 708

445 734 324 057

566 649 411 964

634 122 503 723

681 682 521 672

166 733 165 893 102 744 77 159 76 687 55 677 32 267 29 575 17 833 16 333 13 821 12 238 9096 7308 6778 6616 6494 5338

101 811 101 297 62 737 47 114 46 826 33 997 19 702 18 059 10 889 9973 8439 7472 5554 4462 4138 4039 3965 3259

200 864 111 006 95 179 78 716 80 333 67 540 72 800 71 567 74 225 47 567 39 727 53 954 32 000 41 135 30 290 22 833 24 322 24 842

255 353 141 118 120 998 100 069 102 125 85 861 92 548 90 980 94 359 60 470 50 504 68 590 40 680 52 293 38 507 29 027 30 919 31 581

367 598 276 898 197 923 155 875 157 021 123 217 105 067 101 142 92 058 63 900 53 548 66 192 41 096 48 443 37 069 29 449 30 816 30 180

357 163 242 415 183 735 147 183 148 951 119 858 112 250 109 039 105 249 70 443 58 943 76 063 46 234 56 755 42 645 33 066 34 884 34 840

Source: United Nations FAO statistics.

tart cherry production ranged, for example, from 38 000 to 133 000 metric tons in the last several years, but there has been a gradual downward trend in average production since the mid-1960s to about 120 000 tons (in 2014), with a farm value of about $50 million, but the processed value is at least thrice that. Sweet cherry production, however, has been increasing, especially recently as markets develop in Japan and the Far East of the Pacific Rim for fresh cherries grown in the Western United States and elsewhere. Between 2010 and 2012, the world’s cherry acreage increased by 4.2%, reaching over 400 000 ha, according to the United Nations Food and Agriculture Organization (FAO). Turkey has the largest cherry acreage worldwide (increasing its share from 11% to 12% in 2012), with the United States being the second with 8.7%. Italy is third with Syria (at 7.4%) and then Iran and Spain, with shares of 7.2% and 6%, respectively. Chile’s share increased (from 3.4% to 3.8%). Turkey also increased its share in the world’s production, with 21.3%. The United States was second with 17%, while Iran, Syria, and Italy reduced their shares to 8.9%, 4.6%, and 3.6%, respectively. Chile ranked sixth in world production, with a 4% share of the total in 2012, a sharp increase compared with the 2.8% in 2010.

Growth and Management Flowering and fruit set – Sweet cherry flowers are in clusters of two to four usually borne laterally on short spurs on 2-year-old twigs or near the base of longer 1-year-old shoots. Floral initiation takes place in July, after the crop is harvested, and only on buds where the subtending leaves opened relatively early in summer. Flower buds are of the unmixed type and do not give

rise to a lateral or bourse shoot. As a result, flowering spurs, unlike those on apples and pears, do not remain productive. Flowers generally have a single pistil but in very hot summers may form two pistils that result in undesirable double fruits. With few exceptions, for example, ‘Stella’ (and its progeny) and the new ‘Lapins’ and ‘Sweetheart’ cultivars, commercial sweet cherries are self-sterile (self-incompatible) and therefore require another cultivar for pollination. There are, however, intrasterile groups, where none of the group will crosspollinate any other member of the group. Bing, Lambert, and Napoleon are one such group. Tart cherry flowers develop much like sweet cherry, with buds of two to four flowers on either spurs or lateral buds. Tart cherry cultivars range from compatible to self-incompatible. Montmorency, for example, is only partially compatible but is always grown without a pollinator. Fruit set, however, clearly limits yield on the fully incompatible cultivars, but overcropping may occur in other cultivars with excessive flowering or fruit set resulting in too few leaves or leaf buds to develop fruit of adequate size and quality. In both sweet and tart cherries, flower development and fruit set may frequently be harmed by late frosts, although tart cherry is hardier and generally blooms later than sweet cherry. Wide annual supply fluctuations are consequently common in major growing areas for both species due to spring frosts or to low midwinter temperatures where lack of wood hardiness is a contributing factor. Sweet cherries are less hardy than apples, but some tart cherry cultivars may be as hardy as the apple cultivars McIntosh or Northern Spy. In addition, the bestquality sweet cherry cultivars tend to be more susceptible to rain cracking and do best in regions with dry summer growing conditions. Fruit also develops good quality in regions often too cool for peaches or apricots. Climate also dictates that

12

Cherries (Prunus spp.): The Fruit and Its Importance

cherries be grown where winter chilling temperatures (about 1000 h for most sweet cherries, longer for tart cherry) are adequate to break rest; thus, cherry culture is generally limited to cooler temperate regions. Tree size and rootstocks – Tree size plays a central role in production of quality fruit. Dwarf trees have many advantages: Light penetrates better, favoring photosynthesis; the tree produces more and better fruit; spraying can be done more efficiently, usually with reduced use of chemicals; and dwarf trees are easier to harvest. Cherries are no exception, but dwarfing rootstocks have not until recently been available for either sweet or tart cherry. The common rootstocks, ‘Mazzard’ and ‘Mahaleb,’ only slightly affect tree size, if at all. ‘Colt’ is similar and may be somewhat drought- and cold-susceptible. Recently, however, dwarfing rootstocks have been developed in several breeding programs, and is being tested. For example, of 17 cherry rootstocks developed in Giessen, Germany, most produce relatively large trees, but two of these rootstocks give trees about 25% of the standard, and several rootstocks developed in Belgium may also be promising. Harvesting and handling – Sweet cherries are almost all hand-harvested, particularly those intended for the fresh market. Avoiding pitting and bruising throughout harvest, sorting, and packing is a major problem in delivering high-quality fruit to the fresh market. Bruise susceptibility of some white- or yellow-fleshed cultivars may even require field packing to minimize loss. Mechanical harvesting of tart cherries for processing, however, has been a major technological development, which substantially reduces grower costs. A grower and his family plus some high school students (a crew of 6–8) can mechanically harvest as much as 200–300 hand pickers formerly did. The results include huge savings in direct labor costs, large reductions in housing costs, and substantial savings in labor fringe costs. The US tart cherry industry used essentially completely mechanical harvesting during the 1970s, although there have since been additional improvements in equipment and techniques. Some aspects of cherry processing have also substantially changed, which has led to greater efficiencies and product quality in the cherry industry. Almost all tart cherry processors have adopted electric-eye sorting equipment, which substantially reduces in-plant sorting labor, and destemming equipment efficiently removes the stems from mechanically harvested cherries. Although picking of sweet cherries for the fresh market is still by hand, subsequent handling has been improved dramatically very recently with the substitution of hydraulic flumes for conveyor belts to reduce bruising and pitting throughout sorting and packing.

Quality Factors Soluble solids (primarily hexose sugars and sorbitol) and fruit color (depending on the type) are the best indicators of quality for both sweet and tart cherries, although fruit acid level may be important in tart cherry. Except for soluble carbohydrates, vitamins A and C, and certain flavonoids that may be important as antioxidants in some cultivars, cherries are relatively low in nutrients, but calcium, iron, magnesium, phosphorus, and copper contents are high compared to apple, peach, grape,

and strawberry. High-quality tart cherry fruit typically has at least 15% soluble solids, while sweet cherries should have nearly 20% (or higher). Standards for harvesting and marketing may, however, often be lower. Optimum conditions also often vary with use. To facilitate brining (bleaching in sulfur dioxide solutions for maraschino cherries), fruit may be picked prematurely before color and soluble solids are adequate for the fresh market. Stem fruit removal force is carefully monitored for tart cherries that will be mechanically harvested, and abscission may be brought on by treatment with ethephon, which releases ethylene, expediting abscission and fruit drop in response to mechanical shaking.

Disorders, Diseases, and Pests Disorders – In processing brined sweet and tart cherries, the solution pocket problem involves subepidermal splits in the flesh, which fill with brine solution and with ruptured cell contents. Time of harvest, degree of turgidity at brining, temperature, or any procedures that reduce either the sugar or water content of the fruit will tend to decrease the problem. Rain cracking (swelling followed by rupture of the epidermis) of sweet cherries occurs mostly during the harvest period when the fruit is mature or nearly so and has been wet with rain for some time. Primary cause is absorption of water directly through the skin of the fruit and not through the root system. Cultivar cracking susceptibility has been tested extensively. In testing, Bing, one of the best-quality cultivars, was worst, followed by Napoleon, Lambert, Emperor Francis, Giant, Schmidt, Yellow Spanish, and Montmorency, which did not crack. In another ranking, Bing invariably cracked very badly and was followed by Lambert, Giant, Gil Peck, and Hedelfingen. In still another test, Van, Merton Glory, Vega, and Vista were very susceptible, while Emperor Francis, Schmidt, and Sam were less, and Sue, Kristin, Ulster, and Early Rivers were the least susceptible. From the long-range viewpoint, breeding programs under way ultimately may produce desirable crack-resistant cherries. Cracking may be reduced by some chemical treatments, for example, rain-activated calcium sprayers, but results with hormone (auxin (NAA) and gibberellin (GA3)) applications are equivocal. In some parts of the world, for example, Europe, covering trees with plastic film has been widely used to avoid cracking. Elsewhere, this approach or a high tunnel method has also been used to promote earlier flowering and fruiting, thus capturing high early market returns that justify the extra costs. Pitting of sweet cherry is a condition in which areas near the surface of the fruit become sunken, forming dimples or pits, and may occur before or after harvest, and there are at least three different sources: usually from bruising during handling, from feeding by sucking insects such as the soldier bug, and perhaps from physiological injuries, for example, adverse lowtemperature stress during postharvest cooling or growing conditions. Diseases – Bacterial canker, one of the most important sweet and tart cherry pathogens, is caused by two different pathogens, Pseudomonas syringae and Pseudomonas morsprunorum, and is characterized by oozing of gum (gummosis) at infection sites. Disease development is most prevalent during the

Cherries (Prunus spp.): The Fruit and Its Importance cool, wet periods of early spring. Crown gall, caused by Agrobacterium tumefaciens, can affect sweet and tart cherry rootstocks and is characterized by galls forming usually near infection sites caused by wounds, sometimes man-made, for example, cultivation injuries, or due to damage from subterranean chewing insects or rodents. Some rootstocks, however, are only moderately susceptible, and some hybrids may be tolerant. Brown rot, caused by the fungi Monilinia fructicola or M. laxa, affects both sweet and tart cherries and reduces yield in infected and decaying blossoms, twigs, and fruit. Fruit decay after harvest is also a problem. The fungi persist in mummified fruit on the tree and the ground and infection continues from these inocula the following spring. Growing regions with cooler and drier summers provide some relief. Cherry leaf spot, Blumeriella jaapii (Coccomyces hiemalis), is the most serious disease affecting tart cherry and most ground cherries. Infection occurs in the spring on expanding leaves and continues throughout the season under favorable conditions, for example, high humidity. Severely infected leaves become chlorotic and abscise, and if defoliation is severe, fruit may not ripen properly and tree vigor and hardiness are reduced. Powdery mildew (Podosphaera oxyacanthae) is similar. Other fungal pathogens may sometimes be important, causing blights, crown or root rots, and replant (orchard reestablishment) problems. Cherry dieback is thought to be a complex of several disorders, one of which may be mycoplasma disease. X-disease, leafhopper-transmitted and often devastating, is also due to mycoplasma. Several viruses cause poor vegetative growth, reduce yields, and may even result in tree death, but others are symptomless. Among the most severe is prunus necrotic ringspot virus, which is pollen-transmitted and present in all cherry-growing areas of the world. Little cherry (prune dwarf) disease is another pollentransmitted virus and very destructive. Prunus stem pitting disease is caused by the tomato ringspot virus and is spread by nematodes. Pests – Bird damage can be very serious, and some areas may require protective netting to reduce predation. The cherry fruit fly passes the winter in the soil as a pupa, adult flies emerge in late spring, and females feed on surfaces of leaves and fruit and lay eggs in the nearly ripe fruit. On hatching, the larvae (maggots) feed on the fruit flesh. The larvae are easily killed by holding fruit near 0  C, but fumigation, until recently most often with methyl bromide, may be required to meet quarantine restrictions for shipping overseas. Other pests include black cherry aphid, plum curculio, European red mite, peach tree borer, and two-spotted mite.

Economic Problems and Future Developments Although the tart cherry industry is facing a serious problem of excessive productive capacity in some years and persistent

13

oversupplies, this industry has adopted new technologies and practices in the last 10 years, which substantially improve its cost efficiency and productivity. Much of the newly planted acreage uses efficient trickle irrigation and closely planted orchard systems that also involve hedging techniques. These recent new techniques, especially in combination, provide large increases in yields per hectare and hence substantial reductions in costs. Considerable genetic diversity still exists in Eastern Europe and Russia, the center of origin for both tart and sweet cherries. Although breeding programs have been limited, exploiting that diversity should do much to overcome the growing, handling, and processing problems that face growers using the industry standards, the sweet ‘Bing’ and the tart ‘Montmorency’ in the United States. Sweet cherry growers especially need dwarfing rootstocks and spur types for growth control, and all growers need cultivars with disease and pest resistance, less self-sterility, and a range of maturities so that there are longer seasons for fresh markets. Sweet cherry growers also need cultivars with rain-cracking resistance and, for postharvest fresh market quality, resistance to bruising. Tart cherry growers need new cultivars for diversifying and strengthening marketing options, for example, fresh and frozen juice products, dyes for cosmetics and the food processing industry, and dry stem scars and small freestone pits to facilitate handling and processing. A combination of new marketing strategies and products for tart cherries and advances in breeding of both sweet and tart cherries would clearly benefit the economic potential of the entire cherry industry.

See also: Apples; Berries and Related Fruits; Drying: Effect on Nutrients, Composition and Health; Fruit Juices; Peaches and Nectarines; Plums and Related Fruits; Strawberries.

Further Reading Ayala M, Zoffoli JP, and Lang G (2014) Proceedings of the sixth international cherry symposium (2009). Acta Horticulturae 1020: 1–536. Brettin TS, Karle R, Crowe EL, and Iezzoni AF (2000) Chloroplast inheritance and DNA variation in sweet, tart, and ground cherry. Journal of Heredity 91: 75–79. Brown SK, Iezzoni AF, and Fogle HW (1996) Cherries. In: Janick J and Moore JN (eds.) Fruit breeding, vol. I: tree and tropical fruits, pp. 213–255. New York: Wiley. Iezzoni A, Schmidt H, and Albertini A (1990) Cherries (Prunus spp.). In: Moore JN and Ballington Jr. JR Jr. (eds.) Genetic resources of temperate fruit and nut crops, pp. 110–173. Wageningen: International Society for Horticultural Science. Kappel F, Lang G, Azarenko A, et al. (2013) Performance of sweet cherry rootstocks in the 1998 NC-140 regional trial in western North America. Journal of the American Pomological Society 67: 186–195. Lang GA (2000) Precocious, dwarfing, and productive – how will new cherry rootstocks impact the sweet cherry industry? HortTechnology 10: 719–725. Lang GA (2013) Tree fruit production in high tunnels: current status and case study of sweet cherries. Acta Horticulturae 987: 73–81. Webster AD and Looney NE (eds.) (1996) Cherries: crop physiology, production and uses. New York: Oxford University Press 464 pp.

Chilled Foods: Effects on Shelf-life and Sensory Quality D Bermu´dez-Aguirre and J Welti-Chanes, Tecnolo´gico de Monterrey, Monterrey, Nuevo Leon, Mexico ã 2016 Elsevier Ltd. All rights reserved.

Introduction The commercial cold chain has been evolved together with the progress on Food Science and Technology. In the past, chilled food was a term used for those items that need to be kept under refrigeration because of quick microbial spoilage. Sometimes, these products were seafood and fish caught in remote areas of the world and needed to be transported to specific markets. However, nowadays, the chilled food chain also includes some novel products that must be kept under refrigerated conditions; temperature abuse is not an option because of the microbial risk inherent in the product. These novel products are called cooked-chilled foods or ready-to-eat meals. During the storage of chilled foods, temperature must be kept at specific values and recorded to ensure the microbial safety of the product. Consumers must be aware of the importance of controlling the temperature of these products and make sure their responsibility to handle the product from the store to the final consumption. Even if the product is kept with ice on a boat or kept on a supermarket on the fridge, temperature must be carefully monitored to ensure that spoilage microorganisms are growing slowly or the microbial growth is delayed. Several microbial species have been identified in specific products and these are the ones that should be monitored during the storage; microorganisms such as Bacillus spp. are frequently found in chilled goods and it is recognized as one of the foodborne organisms, and high counts of this sporeformer microorganism can promote gastrointestinal diseases. Pathogens such as Listeria monocytogenes, or even spores of Clostridium botulinum, can be identified in some chilled foods, because of a poor pasteurization process, lack of hygienic measures, cross contamination, or underprocessing of the product. Furthermore, sensory quality of chilled foods can be drastically compromised because of some chemical reactions taking place during storage, such as rancidity, changes in color and flavor, or changes in texture. Refrigerated temperatures delay some of these chemical reactions but do not eliminate them completely. Even if the reaction rate is slow, the chemical processes are taking place, and if the product is stored for several weeks, noticeable changes on sensory properties could be observed. This article presents a brief discussion about some microbial and sensory changes in chilled foods. The specific microbial groups of representative food products are included as well as some of the novel approaches to minimize microbial risks and to improve the current technologies used to preserve chilled foods. Sensory changes in some refrigerated products and some novel strategies used by food technologists are also included.

Chilled Foods The term chilled foods includes a number of products: some of them are ready-to-eat, others require some quick preparation

14

by the consumer, and another category includes chilled products that will be used as ingredients for other products. In general, chilled foods must be kept at a temperature 5  C to ensure the microbial quality of the product through the chilled chain until they reach the consumer. Chilled foods include entre´es, pasta, vegetables, fresh soups, salad dressing, desserts, deli products, dips, ready-to-eat meals, different kinds of meat, fish and seafood, and poultry products. All of them, regardless of the product, must follow the food safety regulations in each country for processing, handling, transportation, storage, and final consumption. Often, a hazard analysis and critical control point (HACCP) program is followed to process, preserve, handle, prepare, transport, and package chilled foods. Some of the main concerns of chilled foods are related with chemical, sensory, and microbial changes of the product by the end of the shelf life when the consumer is eating the product. Chemical changes on the product can seriously compromise the nutritional quality of the product leading to a food item without the original nutrient content. Sensory changes are related more with the acceptability of the product by the consumer; even though these changes do not put at risk the health of the user, they might affect the perception and consumption of the product, making it unsuitable for eating. Finally, one of the most important characteristics of the chilled foods is the microbial quality; even though the growth of the microorganisms is delayed during storage of these products because of low temperature, there are some species such as psychrophilic microorganisms that can grow and represent a risk for the consumption. Besides, some of the emerging pathogens might stay alive on the product and grow if the temperature is not controlled, producing a high risk of foodborne illnesses.

Shelf Life The shelf life of the product depends on several factors such as initial microbial counts, the quality of the raw ingredients, the processing technology, the addition of antimicrobials, and the use of preservation factors such as decrease of pH or aw and storage temperature. The chemical composition of the product will also be an important factor to consider during the shelf life studies, as the richer in nutrients the product is, the faster its microbial growth. Furthermore, aw is an important fact to consider during the shelf life, especially for chilled foods that have high moisture content. Water is important not only for microbial growth but also for the promotion of chemical reactions on the product. Regardless of the chemical composition of the product and the initial microbial population, an additional aspect to consider during the shelf life is the packaging material and the packaging conditions of the product. Some packaging materials represent a strong barrier against moisture, oxygen, and temperature, which together will extend considerably the shelf

Encyclopedia of Food and Health

http://dx.doi.org/10.1016/B978-0-12-384947-2.00144-6

Chilled Foods: Effects on Shelf-life and Sensory Quality life of the product. However, on the chilled food chain, some products are not packaged because they are sold ‘fresh and raw’ such as fish or produce or some products have a very weak barrier with very basic packaging materials. Some products might be packaged using some specific conditions such as vacuum and modified or controlled atmospheres, which also contribute to the extension of the shelf life. Also, several preservation factors have been tested together with low temperatures during food processing to extend the shelf life of different products; physical and chemical hurdles have been used as shown in Table 1.

Microbial Shelf Life Fish and seafood Several marine products such as seafood and fish have a short shelf life because of the chemical reactions taking place together with the postmortem changes. All of these chemical changes can promote faster microbial growth of the product. Spoilage microorganisms on fish include aerobic and proteolytic bacteria, coliforms, and lactic acid species. Pathogens on fish include Vibrio cholera, L. monocytogenes, Escherichia coli, and Salmonella spp., among others (Table 2). Numerous studies have been conducted to incorporate some compounds on the ice that keeps the temperature of chilled foods low. Some of the compounds have antimicrobial properties to delay the microbial growth and extend the shelf life of the product. Studies have been done with plants and herbal extracts, for example, thyme, oregano, clove, basil, and rosemary, among others. Other compounds that have been tested together with ice include some organic acids, alone and in combination. Some examples are citric, ascorbic, and lactic acids. The food products tested include marine species such as seafood (anchovies) and different kinds of fish (sardines and mackerel, among others). In most of the cases, the shelf life of the

product has been considerably extended when comparing with control samples kept only on ice; microbial loads reported for muscles are considerably lower for those treated products compared with control samples. Counts of aerobes, anaerobes, psychrotrophs, Enterobacteriaceae, and lipolytic and proteolytic microorganisms are affected because of the presence of acids and free radicals. This traditional method to preserve fish, using ice, is commonly used for transportation from the origin to the final market of the fish. Boxes with  30% of ice are used to keep the fish with low microbial growth and minimize the chemical reactions during transportation. In the last few years, a new approach has been studied to reduce the amount of ice required for transportation of fish. Superchilled fish has 10–15% of ice since the temperature of the product is reduced to about 1–2  C below its freezing point. The ice is surrounding the product, protecting it from microbial spoilage and enzymatic reactions, creating an ice shell within the product. Other studies on keeping chilling temperatures on fish include the use of ice with different shapes such as the traditional flake ice or the use of small spherical ice crystals known as slurry, flow, or fluid ice. The latest provides an extended shelf life of the fish rather than the flake ice (up to two times longer) because of the direct contact of the crystals within a bigger superficial area of the fish delaying microbial growth and preserving the texture of the product. Other approaches that have been researched to extend the shelf life of fish include the use of chilling slurry, edible films, Table 2 List of microorganisms found in chilled food according to the product Product

Microorganisms

Product

Microorganisms

Fish and seafood

Lactic acid bacteria (LAB) Vibrio spp. Listeria monocytogenes Escherichia coli Pseudomonas spp. H2S-producing bacteria Pseudomonas spp. Lactic acid bacteria (LAB) Brochothrix thermosphacta Enterobacteria

Fresh produce

Coliforms

Table 1 Examples of preservation factors used together with refrigerated conditions to extend the shelf life of chilled foods Food item Fish and seafood Ready-to-eat meals (meats and vegetables) Ready-to-eat meals (beef) Fish Shrimp Seafood Fish Fish Ready-to-cook meats and fish Fish Fish Sausages Fish

Physical factors Irradiation High hydrostatic pressure Vacuum packing Vacuum packing

Different shapes of ice crystals Super chilling (partly freezing)

Chemical factors Herbal extracts Rabbit meat

Essential oils Gas flushing Organic acids Modified atmosphere packaging (MAP) Chitosan

15

Cookedchilled foodsa

Bacteriocins (nisin) Olive oil a

L. monocytogenes

Lamb

Yeast and molds Lactic acid bacteria (LAB) Salmonellaa E. coli O157:H7a L. monocytogenesa E. coli Lactic acid bacteria (LAB) Pseudomonas spp. Yersinia enterocolitica

Clostridium botulinum Bacillus cereus Clostridium perfringens

No common microorganisms but they might be present if the food has not been properly handled.

16

Chilled Foods: Effects on Shelf-life and Sensory Quality

and modified and/or controlled atmosphere. The use of chilling slurry has been tested in cod, basically using seawater slurry ( 2  C) that can reduce the temperature of the product faster than regular ice. During the shelf life, chilling slurry has shown better results on the fish because of the low microbial growth and the freshness of the product. However, some of the problems associated with the use of chilling slurry are associated with the weight gain and salt intake of the fish, both considered drawbacks for the product.

Meat products One of the meat that are highly valued is lamb; some of the most important markets are away from the highest consumers, for example, the lamb from New Zealand needs to travel to foreign markets that sometimes takes several weeks to reach its final destination. The vacuum-packed lamb needs to keep a temperature below 1.5  C to ensure a shelf life between 60 and 70 days. Some of the bacteria that can be found in lamb meat include E. coli, Pseudomonas spp., lactic acid bacteria, and even Yersinia enterocolitica (Table 2). It is essential to keep the product under strict temperature conditions to delay the microbial growth; besides, such as in other meat products, cross contamination can occur during the handling of the animal.

Miscellaneous foods As part of this category, there is a group of very complex foods that include all the cooked-chilled foods that have meat, poultry, fish, and vegetables as part of the list of ingredients. Sales on these products have been drastically increased in the last few years because of the variety of products, innovative concepts, and developed products that fit several lifestyles. The average shelf life of these products is about 5 days when the temperature has been 3  C. However, if the temperature is not well controlled, the products might represent a risk for consumption because of the kind of bacteria that can grow on them. These products are generally pasteurized and packaged such as mashed potatoes, pork chops, ratatouille, minced fish, spaghetti, and mac and cheese. A comprehensive list of these products is presented in Table 3. However, one of the main risks of these products is associated with the presence of some pathogenic microorganisms such as L. monocytogenes and spores of C. botulinum that can survive the pasteurization and be latent on the product. These cooked-chilled foods are also known as refrigerated processed foods of extended durability (REPFED) ready-to-eat meals. Depending on the thermal treatment applied to this kind of products, there are three categories of REPFEDs: (a) Mild thermal treatment at 70  C for 2 min. Basically, this treatment is applied to inactivate L. monocytogenes in at Table 3

least 6 log reduction. These products are pasteurized in the package used to sell them. However, this mild thermal treatment cannot inactivate spores in the product and represents a risk for the consumer if the food is not adequately handled by the consumer until the final consumption. (b) Medium thermal treatment at 90  C for 10 min. This treatment has the goal to inactivate the strains of nonproteolytic vegetative cells of C. botulinum, Bacillus cereus, and L. monocytogenes. The product is pasteurized in the package used to sell it, and because the thermal treatment is not really strong, some spores can resist the process and survive. However, the temperature used for this process can produce some thermal damage on the spores reducing the possibility of producing a foodborne problem. (c) Repackaged chilled foods. This kind of products is pasteurized in some packaging material and after is repackaged in the intended final package. Sometimes, these products are pasteurized in opened containers and after packaged to be sold. The big risks in these products are related with cross contamination after the pasteurization with pathogenic strains. Cold storage is one of the hurdles used on these products to extend the shelf life and delay the microbial growth; however, most of the time, a previous thermal treatment and some other preservation factors (such as reduction of pH, aw, and antimicrobials) are used together. One of the microorganisms that have been also associated with chilled foods is B. cereus. The spores of this microorganism can survive high temperature during conventional pasteurization but it can also survive refrigeration temperatures and grow during the storage of food, representing a microbiological concern. B. cereus is frequently associated with foodborne gastroenteritis. Several foodborne outbreaks have been reported in the food industry in the last decades because of the presence of spore-forming bacteria, mainly from the genera Bacillus. The vehicle of these microorganisms has been mainly associated with the vegetables that are part of most of these cookedchilled foods.

Chemical Shelf Life One of the main chemical reactions taking place on fish is oxidation of the lipids; these reactions known as rancidity are responsible of the generation of off-flavors and odors on the product. Marine products have a characteristic odor that can be easily noticeable. However, when there is lipid oxidation

Examples of cooked-chilled foods (REPFEDs) available on the market; classification is made based on the main ingredient

Meat

Poultry

Fish

Vegetable

Pasta

Pork chops Meatballs with tomato paste Veal stew Beef burgers Sliced ham Roast beef

Chicken and rice Curry chicken Chicken with vegetables Grilled chicken Turkey meat Poultry sausages

Salmon and rice Fish in sauce Minced fish Crab cakes Lobster cakes Surimi

Mashed potatoes Spinach mash Ratatouille Carrots Leak and potato mash Rice with vegetables

Spaghetti Cannelloni Penne with vegetables Mac and cheese Lasagna Fresh pasta salad

Chilled Foods: Effects on Shelf-life and Sensory Quality because of the changes in pH during the shelf life, this odor changes to a very unpleasant and putrid smell because of the several biochemical reactions taking place, in addition to microbial growth. Also, there are chemical changes associated with proteins and microbial growth together with enzymatic activity, releasing some nitrogen that is associated with fish deterioration. Some metabolites coming from the microbial activity such as amino acids are related with the protein hydrolysis taking place on fish and are also responsible for changes on the texture of the product. On the other side, in some meat products, the chemical reactions taking place during the chilled storage are part of the natural process of tenderization of the product. Changes in pH, water-soluble compounds, and lipids on meat muscle during the storage will promote some desirable chemical reactions that provide the product with the characteristic flavor during the cooking process. For example, on beef, the biochemical changes during the postmortem period involving the adenosine triphosphate (ATP) degradation will produce specific flavor precursors on the product once it is cooked. These changes involve the reaction between sugars and amino acids producing specific volatile compounds. Nevertheless, the degree of production of these compounds is also influenced by other factors such as the diet and race of the animal, the season, the animal age, and the conditions of slaughtering, among others. Because the quality of the product depends on the control on the chemical reactions taking place during the storage, several efforts focus on how to optimize these chemical changes. Some examples are the reduction of temperature keeping the product in special chambers, the use of special packaging materials and conditions, the incorporation of some chemicals on the ice, and the use of antioxidants and radical scavengers, among others. The use of antioxidants has been extensively documented; several compounds have been tested in different food products to delay microbial growth, to inhibit enzymes or reduce the chemical reactions catalyzed by them, and to scavenge the free radicals in those products having rancidity problems during storage.

Sensory Quality The basic sensory evaluation of food products includes the assessment of odor, flavor, taste, texture, and appearance. During chilling of foods, some physicochemical changes take place; in some cases, these changes are part of the conditioning process of the product, for example, the postmortem changes mentioned earlier (such as ATP degradation) on fish, beef, pork, and poultry products. The flavor and aroma compounds in different kinds of meat are present on the water-soluble and lipid fractions of the product. Furthermore, these biochemical changes will impact the texture, general acceptability, and overall sensory quality of the product. The chemical reactions associated with the sensory characteristics of the food involve sugars, lipids, and amino acids. Several studies have shown that the conditioning process of beef and pork meat provides better products in terms of sensory characteristics when the product is kept at chilled conditions as long as 21 days. Important and significant changes are observed in aroma, flavor, taste, and texture characteristics

17

such as tenderness, juiciness, and chewiness when the meat is allowed for a long conditioning process. Another example is the use of CO2 snow and brine chilled ( 2  C) in ground beef; both techniques are able to delay the microbial growth on the meat up to 21 days. The use of CO2 snow produced also a better texture on the product having a more tender meat; meanwhile, the chilled option showed cooking losses on the product. Some specific products from beef, such as the heart and liver, which are used as by-products for other industries, are preserved as chilled foods. However, when the product is removed from the animal and immediately packaged under vacuum conditions, the meat is better preserved in terms of weight loss, off-flavor generation, and minimum microbial growth. Then, it is highly recommended for this kind of products to package the product under vacuum just as the postmortem period starts. The use of herbs has also been used in the meat industry to improve the sensory quality of some products such as chilled lamb. Some extracts from rosemary have antioxidant effects, and once applied to chilled lamb, the product can have a longer shelf life not only with minimum lipid oxidation but also with excellent sensory properties in terms of flavor, texture, color, and aroma. The main sensory issues observed on fish during the shelf life are related to the loss of freshness observed on the flesh, the changes on pigmentation, the presence of off-flavors, and the appearance of the skin. Regarding some of the novel applications in chilled foods, some of the previously mentioned studies using herbal extracts to extend the shelf life of fish have shown positive results regarding sensory characteristics. When the ice contains some herbs such as oregano, thyme, and rosemary, the product acquires a similar taste and flavor, making it more appealing to the consumer. Furthermore, when fish is stored under specific preservation factors, the shelf life can be extended and the quality improved. For example, studying the use of superchilled ( 2  C) and chilled (4  C) salmon and having the same fish under modified atmosphere, it is possible to extend the shelf life considerably. Products under superchilled conditions are able to show good quality and lower microbial counts (98% of Cr3þ passes through without being absorbed. This lack of knowledge is in stark contrast to what has been established for ferric iron (Fe3þ), with a similar charge to size ratio as Cr3þ. Dietary iron is probably primarily in the ferric form. Absorption of iron takes place in the proximal portion of the duodenum. Unlike Cr3þ, whose reduction potential is such that it should not readily be reduced under the conditions in the gastrointestinal tract, ferric ions can be reduced chemically or enzymatically by a brush border ferrireductase of the enterocytes. Subsequently, iron enters the enterocytes as ferrous iron via the transmembrane protein DMT1 (divalent metal transporter 1), which is probably responsible for the entry of a variety of divalent metal ions. The transported iron is then stored or is released from the enterocyte by the basolateral transmembrane protein ferroportin. Iron probably passes through the enterocyte in the ferrous state (Fe2þ) and is returned to the ferric state (Fe3þ) by ferroxidases for transport in the bloodstream by the protein transferrin (vide infra). The failure of Cr3þ ions to be reduced requires a unique absorption system to be present for chromium, compared with other proposed essential metals ions, if chromium is actively absorbed and essential. However, the preponderance of evidence suggests that chromic ions are passively absorbed, that is, absorbed via simple diffusion. In rats, chromium has been established to be absorbed via passive diffusion. Only a small percentage (10%. It is homogenized and UHT-processed, filled aseptically, or sterilized in the container. It is a popular product that is mainly used for whitening coffee, as well as for

335

imparting a pleasant flavor to coffee. It is also used in the preparation of food and drinks, and for direct consumption. Coffee cream has a minimum shelf-life of 4 months at room temperature, and it normally contains 10–12 g fat/100 g, and, less often, 15–20 g fat/100 g. Its shelf life is similar to the shelf life of UHT milk. The important quality attributes of coffee cream are taste, whitening power, and stability in hot coffee. Following separation and standardization, the cream is heat-treated at 90–95
C and cooled to 6
C. Stabilizers such as phosphates or citrates may be added to increase the pH and to reduce the concentration of ionic calcium, thereby reducing the susceptibility of casein micelles to aggregation during the sterilization process or on addition to hot coffee. Cream is subjected to two-stage homogenization and sterilization. Traditionally, coffee cream was sterilized in cans or bottles, but since 1996, continuous-flow sterilization in a UHT plant, followed by aseptic packaging, has largely replaced the former process. Both homogenization steps operate at a total pressure of 20 MPa and a temperature of 70
C. Flow-sterilized coffee cream is subsequently cooled to 25
C and aseptically filled, usually into plastic cups (7.5–15 g capacity) or preformed cups or cans (100–200 g capacity).

Importance of homogenization Coffee cream must be homogenized. This prevents a fat layer or fat plug from forming in the container, thus improving taste, whipping power, and stability. It has a direct influence on the flocculation stability of coffee cream. When sterilizing cream in the pack, homogenization has to take place before the sterilization, which again is a double stage process (20 MPa) using the same pressure. Flocculation of cream in hot coffee is due to casein precipitation.

Desirable attributes of coffee cream In addition to good sensory properties, the two main criteria for assessing the quality of coffee creams are as follows: Stability of the emulsion: This is the ability of the particulate constituents (primarily lipid and protein particles) to remain evenly dispersed throughout the coffee cream during its long shelf-life. Whitening effect: This is the ability to form a milky, homogenous mixture on addition to coffee, irrespective of the composition, brewing conditions, and temperature of the coffee. Like UHT milk, homogenized UHT creams exhibit a gelation phenomenon, generally attributed to the activity of heatresistant proteolytic enzymes, after prolonged storage. One important defect with respect to coffee creams is the phenomenon known as ‘feathering.’ Feathering is the formation of macroscopic aggregates containing proteins and lipids through the acid-induced flocculation of partially proteincovered lipid globules in hot coffee drinks. Although aggregates always form in hot coffee, the visible aggregates (i.e., 100 mm) render the coffee cream defective, because the consumer commonly perceives such aggregates as an indication of the sourness of the product. The susceptibility of coffee cream to feathering is often referred to as its coffee stability.

336

Cream: Types of Cream

Cream Liqueur Cream liqueurs are a group of liqueurs that contain milk cream as an ingredient. The roots of cream liqueur lie in home-made products that can be traced back for centuries, such as the Scottish product Atholl Brose. Atholl Brose consists of milk cream and honey, to which whisky, steeped with oatmeal to impart a pleasant nutty flavor, is added. In the second half of the twentieth century, commercial-scale attempts to produce Atholl Brose and other home-made cream liqueurs largely failed, until 1974, when Baileys Irish Cream Liqueur became the first commercialized cream liqueur. The most famous cream liqueurs are the Irish product Baileys Irish Cream and the South African product Amarula.

Composition of cream liqueur In terms of commercial practice, a standard cream liqueur does not exist. Each manufacturer decides the preferred components and their concentrations. This flexibility enables a product of desired organoleptic properties and a sufficiently long shelflife. Table 3 shows the composition of a standard cream liqueur, which appears to represent most commercially available products. The 14 g/100 g concentration of ethanol in the product equates to 17 g/100 ml ethanol. Of the ingredients, sodium caseinate is used as an emulsifier, and citrate is added to prevent serum separation during storage.

Technology of cream liqueur Although there is no universal manufacturing process for cream liqueur, the principle steps of mixing the ingredients and homogenizing the cream base are universal. The sodium caseinate can be dissolved in hot water (85
C), after which time the other ingredients are added to produce the cream base. Ethanol may be added to the cream base either before or after homogenization. The function of homogenization is to reduce the lipid globule size, thereby prolonging the physical stability of the product by preventing creaming and/or the formation of the cream plug on storage. A general rule-ofthumb for homogenization efficiency is that >98% of the milk lipid globules should have a diameter 3.00) Maximum residual amounts Maximum residual amounts Maximum residual amounts Only hard, semihard, and semisoft cheese Only cheese milk of hard, semihard, and semisoft cheese Only hard, semihard, and semisoft ripened products Only dairy-based cheese analogue

Cured Foods: Health Effects which the initial content of free amines is enormous, curing implies a very high risk for the formation of unacceptable amounts of NAs. In fact, addition of nitrites to fish and seafood is banned or severely restricted by most food legislation. Some unusual NAs, such as N-nitrosodi-n-butylamine and N-nitrosodibenzylamine, have also been detected in cured meat products packaged in elastic rubber nettings, formed as a result of the interaction of the residual nitrite in the cured meats with amine additives in the rubber nettings. NPIP in cured products due to the simultaneous presence of nitrite and pepper can occur either by the oxidative cleavage of the amide bond of piperine and a subsequent nitrosation of piperidine or by a direct nitrosation of the already available piperidine. In addition to the formation of NA in the food, nitrates and nitrites are also involved in their formation in vivo in the human body, mainly due to the acidic conditions of the stomach. Gastric nitrosation proceeds through reaction between amines from the food and nitrosating agents derived from the diet, both nitrites present directly in the food and nitrates reduced to nitrites in the mouth. In fact, nitrate is resorbed in the intestine and excreted through the saliva; thereafter, it is partially reduced to nitrite by nitrate reductase-containing bacteria in the oral cavity. Some studies have suggested that around 90% of ingested nitrite comes from the saliva. Nitrite intake in most countries mainly comes from cured meats, while the main source of nitrates is green leafy vegetables. In fact, most global food legislation has limited the use of nitrates for cooked cured meat products, so that the actual amount of nitrite can be more precisely controlled (Table 1). Moreover, there has been a trend toward a reduction in allowed levels of nitrate and nitrite in cured meats, so that nowadays, the residual levels are much lower than they used to be 30 years ago. Nevertheless, EU legislation still allows the use of nitrate in several traditional cured meat products and in cured cheese (Table 1). In the case of traditional cured meat products, processors claim that without the addition of nitrates, it would be difficult to achieve the features that consumers expect in this type of products, while in cheeses, the allowance is based in the prevention of early blowing during the processing of less acidic varieties. The use of such high levels of nitrates allowed in pickled herring by the EU normative is remarkable, and as such, usage entails a risk for NA formation. Nitrosation in the intestinal track seems to be markedly diminished by the presence of ascorbate, so that its presence in the food would limit the in vivo formation of NAs. However, some findings suggest that in the presence of fat, ascorbate could actually enhance the formation of NA in the stomach. The parallel intake of polyphenols or other plant compounds with antioxidant activity could also help in the reduction of in vivo nitrosation, while other studies suggest that heme from meat could play a role in boosting nitrosation in the large intestine. Without question, more research is needed in this area, which will potentially allow designing cured food product formulas targeted at reducing nitrosation of amines, both in the food and in vivo. As discussed previously, the potential involvement of cured products in the development of cancer caused by N-nitroso compounds could be due to both the formation of NAs (and other N-nitroso compounds) during product processing and storage and their contribution to the total nitrate and nitrite

341

intake, which would be in turn related to in vivo nitrosation. Indeed, the role of NAs in cancer was already described in the 1950s. Later, in the 1960s and 1970s, several papers linked the consumption of cured meats to the development of different types of cancer. Although subsequent studies failed to confirm that nitrite intake caused cancer, the Assembly of Life Sciences in the United States was prompted to commission a report on the toxicity of nitrate and nitrite, which was only partially reassuring. In 2003, the Joint FAO/WHO Expert Committee on Food Additives stated that the epidemiological studies showed no consistently increased risk for cancer with increasing consumption of nitrate. Subsequent epidemiological studies have similarly failed to show a convincing link between nitrate or nitrite intake and cancer. Nevertheless, the International Agency for Research on Cancer (IARC) published a monograph in 2010 in which it was concluded that ingested nitrate or nitrite under conditions that result in endogenous nitrosation is probably carcinogenic to humans. However, this statement does not specifically refer to cured products. In fact, and as explained earlier, the major sources for nitrate and nitrite intake in most countries are saliva, vegetables, and water. More specifically related to cured products, different epidemiological studies relating the consumption of cured meats or fish with a higher risk for the development of different types of cancer have been published and have been widely communicated in the media. Moreover, the World Cancer Research Fund published guidelines for cancer prevention, one recommendation being to limit the consumption of red meat and eliminate processed meat consumption entirely. Although not specifically mentioned, it is assumed that such a risk is attributed to curing, although other factors specifically linked to the processing of meats could also be involved. Such claims have been very controversial, and a number of papers and articles have been published highlighting the fact that such published claims have lacked epidemiological evidence. In fact, it has been pointed out that studies linking the consumption of processed meat to cancer seriously failed to completely consider several additional factors that can contribute to chronic disease, including participants’ behavior as to alcohol and tobacco use, exercise, weight, and access to health care. At any rate, if processed meats are truly involved in a higher risk of colorectal cancer, it is unlikely that this is due to residual nitrate or nitrite, because green leafy vegetables show much higher levels and seem to show no epidemiological relationship to cancer. Moreover, both nitrate and nitrite are formed endogenously and recycled with the saliva. Nevertheless, some studies suggest that the concomitant ingestion of heme and nitrosation agents could boost the endogenous formation on N-nitroso compounds in the colon. Others have hypothesized that N-nitrosamides, rather than NAs, could be related to the development of some types of cancers, but little is known about their actual content or their specific involvement in this type of disease.

Other Potential Harmful Effects of Cured Foods Cases of infant methemoglobinemia as a consequence of wellwater with a high nitrate content occurred in the past, and that is the reason why the maximum allowed amount of nitrates in

342

Cured Foods: Health Effects

water is regulated in most countries. In fact, it is nitrite that causes methemoglobinemia, and in this sense, it could be thought that the consumption of food containing residual nitrite, like cured foods, could somehow be related to this type of disease. Nevertheless, the levels of dietary nitrites causing this condition are much higher than those that can be found in cured foods. Yet, there are reports linking cases of methemoglobinemia to the consumption of cured products with high levels of nitrites. For example, in one such incidence dating from the 1950s, an outbreak of methemoglobinemia occurred after consuming wieners and bologna containing more than 5000 ppm of nitrite. At any rate, episodes of cured food involved in cases of acute nitrite poisoning are most likely due to processing mistake.

blood flow to type II muscles. Given the importance of type II fiber recruitment and metabolism to performance in highintensity intermittent exercise, and the likelihood that tissue hypoxia develops to a greater extent during such exercise compared to continuous low-intensity exercise, these recent studies therefore provide a clear rationale for nitrate supplementation to enhance performance during multiple sprint sport.

See also: Cancer: Diet in Cancer Prevention; Food Additives: Classification, Uses and Regulation; Nitrites and Nitrates.

Further Reading Potential Beneficial Effects of Nitrate and Nitrite Nitrate conversion into nitrite, and subsequently into NO, is behind the potential positive health effects of intake of these inorganic salts. Most of these possible effects have been shown for vegetable extracts, very rich in nitrates, and it is unlikely that any significant effect could be due to the residual levels of nitrate or nitrite found in cured products. Nevertheless, as a part of the diet, cured products no doubt contribute to the total intake of nitrate and specially nitrite, since they are estimated to be the third source of these salts, after vegetables and water. NO is an endogenously produced modulator of vascular tone. Indeed, as a consequence of eating a meal rich in nitrates, a subsequent fall in both systolic pressure and diastolic pressure has been described. It seems clear that the reduction of ingested nitrate into nitrite is behind this effect, since in an experiment in which subjects had to eliminate their saliva after nitrate ingestion, the nitrite peak in blood disappeared and so too did the effect on blood pressure. NO also plays a key role in vascular homeostasis by maintaining vessels in their relaxed state; in experiments carried out in humans, dietary nitrate supplementation attenuated the impairment in endothelial function seen in ischemia reperfusion injury. NO also plays a role in the inhibition of platelet adhesion and aggregation. In this sense, dietary supplementation with either nitrate or beetroot juice has resulted in significant inhibition of platelet aggregation. Widespread media attention has focussed on the positive effect of nitrate on muscle efficiency and fatigue resistance. In different experiments, a short-term dietary supplementation with sodium nitrate resulted in improved muscular efficiency and a reduction in oxygen consumption during submaximal exercise in healthy subjects. NO has numerous physiological signaling functions that may impact on muscle fatigue and performance: regulation of blood flow, muscle contractility, glucose and calcium homeostasis, and mitochondrial oxygen consumption. The reduction of nitrite to NO is facilitated when oxygen availability is limited, which agrees with the fact that nitrate supplementation appears to be particularly effective in enhancing performance in hypoxia and ischemia. Recent studies suggest that nitrate supplementation may specifically improve calcium handling and force production in type II muscle fibers and result in preferential distribution of

Cross AJ, Pollock JRA, and Bingham SA (2003) Haem, not protein or inorganic iron, is responsible for endogenous intestinal N-nitrosation arising from red meat. Cancer Research 63: 2358–2360. Dellisanti A, Cerutti G, and Airoldi L (1996) Volatile N-nitrosamines in selected Italian cheeses. Bulletin of Environmental Contamination and Toxicology 57: 16–21. Drabik-Markiewicz G, Dejaegher B, De Mey E, Kowalska T, Paelinck H, and Vander Heyden Y (2011) Influence of putrescine, cadaverine, spermidine or spermine on the formation of N-nitrosamine in heated cured pork meat. Food Chemistry 126: 1539–1545. Gangolli SD, van den Brandt P, Feron V, et al. (1994) Nitrate, nitrite, and N-nitroso compounds. European Journal of Pharmacology 1: 1–38. Herrmann SS, Duedahl-Olesen L, and Granby K (2015a) Occurrence of volatile and non-volatile N-nitrosamines in processed meat products and the role of heat treatment. Food Control 48: 163–169. Herrmann SS, Duedahl-Olesen L, and Granby K (2015b) Formation and mitigation of N-nitrosamines in nitrite preserved cooked sausages. Food Chemistry 174: 516–526. Honikel KO (2008) The use and control of nitrate and nitrite for the processing of meat products. Meat Science 78: 68–76. Hord NG, Tang Y, and Bryan NS (2009) Food sources of nitrates and nitrites: the physiologic context for potential health benefits. American Journal of Clinical Nutrition 90: 1–10. IARC (1978) IARC monographs on the evaluation of carcinogenic risks to humans, vol. 17: some N-nitroso compounds. Lyon: IARC. Lijinsky W and Epstein SS (1970) Nitrosamines as environmental carcinogens. Nature 225: 21–23. Pegg RB and Shahidi F (2008) Nitrite curing of meat: the N-nitrosamine problem and nitrite alternatives. Trumbull: Wiley-Blackwell. Pierre FHF, Martin OCB, Santarelli RL, et al. (2013) Calcium and a-tocopherol suppress cured-meat promotion of chemically induced colon carcinogenesis in rats and reduce associated biomarkers in human volunteers. American Journal of Clinical Nutrition 98: 1255–1262. Skibsted LH (2011) Nitric oxide and quality and safety of muscle based foods. Nitric Oxide 24: 176–183. Stuff J, Goh ET, Barrera SL, Bondy ML, and Forman MR (2009) Construction of an Nnitroso database for assessing dietary intake. Journal of Food Composition and Analysis 22(S): S42–S47. Tricker AR and Preussmann R (1991) Carcinogenic N-nitrosamines in the diet: occurrence, formation, mechanisms and carcinogenic potential. Mutation Research 259: 277–289. Yurchenko S and Molder U (2006) Volatile N-nitrosamines in various fish products. Food Chemistry 96: 325–333.

Relevant Websites http://www.cmc-cvc.com/en/nutrition-health/nitrite-cured-meat-products – Canadian Meat Council. http://www.efsa.europa.eu/ – EFSA. http://epic.iarc.fr/ – EPIC study. http://www.meatami.com/ – North American Meat Institute.

Cystic Fibrosis, Nutrition in S Sabharwal, Boston Children’s Hospital, Boston, MA, USA ã 2016 Elsevier Ltd. All rights reserved.

Definition Cystic fibrosis (CF) is a multiple organ system disease, affecting the lungs and pancreas. Nutritional status is impaired by inadequate intake, increased energy needs, and pancreatic insufficiency. Pancreatic insufficiency results in maldigestion and malabsorption of nutrients and fat-soluble vitamins. Pancreatic enzyme supplementation and optimizing nutritional deficiencies prevent growth failure and improve other outcomes in patients with CF, including quality of life, resistance to infection, and chronic lung disease prevention, leading to longer life expectancy in patients with this disease. CF is a result of the defects in the CF transmembrane conductance regulator (CFTR), which is responsible for the excretion of salt, in turn creating viscous secretions in multiple organ systems. For decades, CF was thought to be a disease only of childhood given the lower life expectancy. Largely because of improvements in nutrition, the average life expectancy of patients with CF has advanced well into adulthood.

Clinical Manifestations Defects in the CFTR result in multisystem disease including lung, liver, and gastrointestinal (GI) diseases, notably pancreatic insufficiency. While the majority of CF mutations cause lung disease, growth and nutritional status are closely tied to lung disease and affected by the type of CF mutation. Body mass index (BMI) being maintained at  50% for age has been shown to correlate with the percent predicted forced expiratory volume in 1 s of greater than or equal to 90%, which is an excellent marker of lung function. For CF patients ages greater than or equal to 20 years, it is recommended that women maintain a BMI at or above 22 and that men maintain a BMI at or above 23. Nutritional failure in CF is multifactorial. Insufficient production of pancreatic enzymes leads to fat malabsorption, which can be exacerbated by bile salt abnormalities if there is concurrent liver disease. Progressive pulmonary infection leads to increased work of breathing and therefore increased caloric needs. Chronic lung infection can reduce appetite and lead to an inflammatory catabolic process in the body. Other factors affecting nutrition include CF-related diabetes mellitus, altered motility of the GI tract, and small bowel bacterial overgrowth.

Management CF derived its name from the cysts and fibrosis noted in the pancreas of patients with CF. The pancreas is responsible for secreting enzymes that aid in the absorption of nutrients including fat-soluble vitamins A, D, E, and K. In order to counteract this malabsorption, pancreatic enzyme replacement

Encyclopedia of Food and Health

therapy is used. Lipase is the enzyme responsible for fat absorption. Pancreatic enzyme replacement is based on the dosage of lipase in the supplement (500–2500 lipase units per kilogram of body weight per meal or –1 WAZ for mortality) –2 to 70 years Pregnancy 14–18 years 19–30 years 31–50 years Lactation 14–18 years 19–30 years 31–50 years

Total watera (l day
1)

Carbohydrate (g day
1)

Total fiber (g day
1)

Fat (g day
1)

Linoleic acid (g day
1)

a-Linolenic acid (g day
1)

Proteinb (g day
1)

0.7* 0.8*

60* 95*

ND ND

31* 30*

4.4* 4.6*

0.5* 0.5*

9.1* 11.0

1.3* 1.7*

130 130

19* 25*

NDc ND

7* 10*

0.7* 0.9*

13 19

2.4* 3.3* 3.7* 3.7* 3.7* 3.7*

130 130 130 130 130 130

31* 38* 38* 38* 30* 30*

ND ND ND ND ND ND

12* 16* 17* 17* 14* 14*

1.2* 1.6* 1.6* 1.6* 1.6* 1.6*

34 52 56 56 56 56

2.1* 2.3* 2.7* 2.7* 2.7* 2.7*

130 130 130 130 130 130

26* 26* 25* 25* 21* 21*

ND ND ND ND ND ND

10* 11* 12* 12* 11* 11*

1.0* 1.1* 1.1* 1.1* 1.1* 1.1*

34 46 46 46 46 46

3.0* 3.0* 3.0*

175 175 175

28* 28* 28*

ND ND ND

13* 13* 13*

1.4* 1.4* 1.4*

71 71 71

3.8* 3.8* 3.8*

210 210 210

29* 29* 29*

ND ND ND

13* 13* 13*

1.3* 1.3* 1.3*

71 71 71

Note: This table (taken from the DRI reports; see www.nap.edu) presents the recommended dietary allowances (RDAs) in bold type and adequate intakes (AIs) in ordinary type followed by an asterisk (*). An RDA is the average daily dietary intake level, sufficient to meet the nutrient requirements of nearly all (97–98%) healthy individuals in a group. It is calculated from an estimated average requirement (EAR). If sufficient scientific evidence is not available to establish an EAR, and thus calculate an RDA, an AI is usually developed. For healthy breastfed infants, an AI is the mean intake. The AI for other life stages and gender groups is believed to cover the needs of all healthy individuals in the groups, but the lack of data or uncertainty in the data prevents one from specifying with confidence the percentage of individuals covered by this intake. a Total water includes all water contained in food, beverages, and drinking water. b Based on g protein per kg of body weight for the reference body weight, for example, for adults 0.8 g kg
1 day
1 body weight for the reference body weight. c Not determined. Source: Reprinted with permission from Dietary Reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids (2002/2005) and Dietary reference intakes for water, potassium, sodium, chloride, and sulfate (2005) by the National Academy of Sciences, courtesy of the National Academies Press, Washington, DC. The report may be accessed via www.nap.edu.

Table 5 Dietary reference intakes (DRIs): acceptable macronutrient distribution ranges (Food and Nutrition Board, Institute of Medicine, National Academies) Range (percent of energy) Macronutrient

Children, 1–3 years

Children, 4–18 years

Adults

Fat n
6 polyunsaturated fatty acidsa (linoleic acid) n
3 polyunsaturated fatty acidsa (a-linolenic acid) Carbohydrate Protein

30–40 5–10 0.6–1.2 45–65 5–20

25–35 5–10 0.6–1.2 45–65 10–30

20–35 5–10 0.6–1.2 45–65 10–35

Macronutrient

Recommendation

Dietary cholesterol Trans-fatty acids Saturated fatty acids Added sugarsb

As low as possible while consuming a nutritionally adequate diet As low as possible while consuming a nutritionally adequate diet As low as possible while consuming a nutritionally adequate diet Limit to no more than 25% of total energy

a

Approximately 10% of the total can come from longer-chain n
3 or n
6 fatty acids. Not a recommended intake. A daily intake of added sugars that individuals should aim for to achieve a healthful diet was not set. Source: Reprinted with permission from Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids (2002/2005) by the National Academy of Sciences, courtesy of the National Academies Press, Washington, DC. The report may be accessed via www.nap.edu. b

Dietary References: US AMDR for an individual can only be calculated once the total energy expenditure (TEE) is estimated and that it is not based on experimental data. The UL is the highest level of daily nutrient intake that is likely to pose no risk of adverse health effect to almost all individuals. These are limits set for safety reasons and are not derived from a continuous distribution like the EAR that permits statistical manipulation of the probabilities of exceeding the value. UL is determined using a risk assessment model adapted for nutrients that is set based on expert judgment for the adverse effect of interest for the nutrient. Table 6 presents the UL for vitamins, and Table 7 presents the UL for mineral elements (Figure 2). The other DRIs describe energy intakes. They include the resting energy requirement, or REE, and the TEE, an estimate of the energy expenditure needed to maintain current body weight and activity levels. There is no RDA for energy since an excess above one’s energy requirement would result in weight gain. An individual’s resting energy expenditure only describes needs at rest. The estimated energy requirement (EER) is the estimated average dietary intake associated with an individual of a specific sex, age, height, weight, and activity level. The EER values are determined from equations that reflect the energy expenditure of individuals of normal body composition using data from doubly labeled water experiments. The EER and TEE are the same under conditions other than growth, pregnancy, or lactation; the EER prediction equations also account for the deposition of new tissue during pregnancy or growth and production of milk in the case of lactation. Many different equations are used to estimate the TEE or REE, depending on the individual’s height, weight, age, and current activity level. Some use a multiplier reflecting physical activity level, ranging from 1 for sedentary, to 1.48 for moderately active, to higher for the very active. For hospitalized patients, different multipliers are sometimes used to account for metabolic stresses due to illness. All of these equations are approximations but useful to obtain a rough rule of thumb on energy needs. The DRIs are based on a ‘reference person’ of a given weight and body composition. Those references are based on typical characteristics of American and Canadian citizens in these regards. They present reference values for each nutrient by age, sex, and physiological condition (pregnancy or lactation) for healthy people.

Applications The DRIs are used to assess nutritional adequacy and make recommendations for dietary planning of individuals and groups. They are also used as standards for establishing nutrient recommendation for those who are ill. The DRIs for individuals suffering from disease only change when a specific nutrient requirement is affected by a disease or condition.

Assessment The DRIs are used to assess the adequacy of nutrient intakes of individuals and populations from recalls or records of usual intake for micronutrients, protein, essential fatty acids, and carbohydrates. They are also used as a rule of thumb for

425

assessing the intakes of individuals, but caution is needed. Usual intake and not a single day’s intake must be considered. Moreover, since an individual’s requirement is never known, only qualitative statements about adequacy are possible. The EAR is the average requirement for a group, but by definition, it does not meet the requirements of half of the population. In contrast, the RDA meets the requirements of virtually all individuals. Usual intakes at or above the RDA mean a low probability of inadequacy since the RDA is calculated to meet 97–98% of the needs of the population. Note that intakes below the RDA may be sufficient for some individuals whose requirements are low. However, we never know if a given individual’s requirement is high or low. The likelihood of nutrient inadequacy increases as usual intakes fall further and further below the RDA, but intakes below the RDA cannot be assumed to be inadequate. Although definitive statements cannot be made, mean intakes of an individual below the EAR very likely need to be improved because the probability of inadequacy is by definition at least 50%. An individual’s observed mean intake between the EAR and the RDA probably needs to be improved, because the probability of inadequacy is more than 50 but less than 97%. Only if an individual’s intake has been reported for a large number of days can one have a considerable degree of confidence that the diet is inadequate. For nutrients without an EAR or RDA, those who have intakes above the AI can be considered to have reasonably good intakes. The AI is an intake at or above which there is low probability of inadequacy. Therefore, many intakes are already likely to be in excess of requirements; an intake that is low by the AI standard may still be perfectly adequate. The UL is the usual intake above which there is an increased risk of adverse effects for almost all individuals in the general population. When assessing nutrient intake, the UL can only show the likelihood of adverse effects from usual (chronic) intakes above the UL, not intake on a single day. It is important to note that with most nutrients, the UL applies to all sources of dietary intake, but for one nutrient, magnesium, only pharmacological sources are included. Dietary assessment of an individual is only one component of a complete nutritional status evaluation. It should be combined with clinical, biochemical, and anthropometric data to provide a valid picture of nutritional status, especially in those who are significantly different than the ‘reference person.’ The goal of assessing nutrient intakes of a population using the DRI is to determine the prevalence of inadequate or excessive nutrient intakes within it. There are several methods available to estimate usual intake distributions from dietary intake data on two or more days’ worth of intakes. More information and a link to software are available through Iowa State University at http://www.cssm.iastate. edu/software/side/. The statistical techniques are well worked out, and today, monitoring of dietary adequacy and excess is done on an ongoing basis in the US National Health and Nutrition Examination Survey.

Planning The primary goal of planning an individual’s diet is to devise an intake pattern that ensures nutrient adequacy while also minimizing risk of excess and maximizing individual

Table 6 Life stage/ group Infants 0–6 months 6–12 months Children 1–3 years 4–8 years Males 9–13 years 14–18 years 19–30 years 31–50 years 51–70 years >70 years Females 9–13 years 14–18 years 19–30 years 31–50 years 51–70 years >70 years

Dietary reference intakes (DRIs): tolerable upper intake levels, vitamins (Food and Nutrition Board, Institute of Medicine, National Academies) Vitamin A (mg day
1)a

Vitamin C (mg day
1)

Vitamin D (mg day
1)

Vitamin E (mg day
1)b,c

Vitamin K (mg day
1)

Thiamin (mg day
1)

Riboflavin (mg day
1)

Niacin (mg day
1)c

Vitamin B6 (mg day
1)

Folate (mg day
1)c

Vitamin B12 (mg day
1)

Pantothenic acid (mg day
1)

Biotin (mg day
1)

Choline (g day
1)

Carotenoidsd

600 600

NDe ND

25 38

ND ND

ND ND

ND ND

ND ND

ND ND

ND ND

ND ND

ND ND

ND ND

ND ND

ND ND

ND ND

600 900

400 650

63 75

200 300

ND ND

ND ND

ND ND

10 15

30 40

300 400

ND ND

ND ND

ND ND

1.0 1.0

ND ND

1700 2800

1200 1800

100 100

600 800

ND ND

ND ND

ND ND

20 30

60 80

600 800

ND ND

ND ND

ND ND

2.0 3.0

ND ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

1700 2800

1200 1800

100 100

600 800

ND ND

ND ND

ND ND

20 30

60 80

600 800

ND ND

ND ND

ND ND

2.0 3.0

ND ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

Pregnancy 14–18 years 19–30 years 31–50 years Lactation 14–18 years 19–30 years 31–50 years

2800

1800

100

800

ND

ND

ND

30

80

800

ND

ND

ND

3.0

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

2800

1800

100

800

ND

ND

ND

30

80

800

ND

ND

ND

3.0

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

3000

2000

100

1000

ND

ND

ND

35

100

1000

ND

ND

ND

3.5

ND

Note: A tolerable upper intake level (UL) is the highest level of daily nutrient intake that is likely to pose no risk of adverse health effects to almost all individuals in the general population. Unless otherwise specified, the UL represents total intake from food, water, and supplements. Due to a lack of suitable data, ULs could not be established for vitamin K, thiamin, riboflavin, vitamin B12, pantothenic acid, biotin, and carotenoids. In the absence of a UL, extra caution may be warranted in consuming levels above recommended intakes. Members of the general population should be advised not to routinely exceed the UL. The UL is not meant to apply to individuals who are treated with the nutrient under medical supervision or to individuals with predisposing conditions that modify their sensitivity to the nutrient. a As preformed vitamin A only. b As a-tocopherol; applies to any form of supplemental a-tocopherol. c The ULs for vitamin E, niacin, and folate apply to synthetic forms obtained from supplements, fortified foods, or a combination of the two. d b-Carotene supplements are advised only to serve as a provitamin A source for individuals at risk of vitamin A deficiency. e ND ¼ not determinable due to the lack of data of adverse effects in this age group and concern with regard to the lack of ability to handle excess amounts. Source of intake should be from food only to prevent high levels of intake. Source: Reprinted with permission from Dietary reference intakes for calcium, phosphorous, magnesium, vitamin D, and fluoride (1997); Dietary reference intakes for thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and choline (1998); Dietary reference intakes for vitamin C, vitamin E, selenium, and carotenoids (2000); Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc (2001); and Dietary reference intakes for calcium and vitamin D (2011) by the National Academy of Sciences, courtesy of the National Academies Press, Washington, DC. These reports may be accessed via www.nap.edu.

Table 7 Life stage/ group Infants 0–6 months 6–12 months Children 1–3 years 4–8 years Males 9–13 years 14–18 years 19–30 years 31–50 years 51–70 years >70 years Females 9–13 years 14–18 years 19–30 years 31–50 years 51–70 years >70 years

Dietary reference intakes (DRIs): tolerable upper intake levels, elements (Food and Nutrition Board, Institute of Medicine, National Academies)

Arsenica

Boron (mg day
1)

Calcium (mg day
1)

Copper Chromium (mg (mg day
1) day
1)

Fluoride (mg day
1)

Iodine (mg day
1)

Iron (mg Magnesium Manganese Molybdenum day
1) (mg day
1)b (mg day
1) (mg day
1)

Nickel (mg day
1)

Phosphorus (g day
1)

Selenium (mg day
1)

Siliconc

Vanadium (mg day
1)d

Sodium Zinc (g (mg day
1) day
1)

Chloride (g day
1)

NDe ND

ND ND

1000 1500

ND ND

ND ND

0.7 0.9

ND ND

40 40

ND ND

ND ND

ND ND

ND ND

ND ND

45 60

ND ND

ND ND

4 5

ND ND

ND ND

ND ND

3 6

2500 2500

ND ND

1000 3000

1.3 2.2

200 300

40 40

65 110

2 3

300 600

0.2 0.3

3 3

90 150

ND ND

ND ND

7 12

1.5 1.9

2.3 2.9

ND ND

11 17

3000 3000

ND ND

5000 8000

10 10

600 900

40 45

350 350

6 9

1100 1700

0.6 1.0

4 4

280 400

ND ND

ND ND

23 34

2.2 2.3

3.4 3.6

ND

20

2500

ND

10 000

10

1100

45

350

11

2000

1.0

4

400

ND

1.8

40

2.3

3.6

ND

20

2500

ND

10 000

10

1100

45

350

11

2000

1.0

4

400

ND

1.8

40

2.3

3.6

ND

20

2000

ND

10 000

10

1100

45

350

11

2000

1.0

4

400

ND

1.8

40

2.3

3.6

ND

20

2000

ND

10 000

10

1100

45

350

11

2000

1.0

3

400

ND

1.8

40

2.3

3.6

ND ND

11 17

3000 3000

ND ND

5000 8000

10 10

600 900

40 45

350 350

6 9

1100 1700

0.6 1.0

4 4

280 400

ND ND

ND ND

23 34

2.2 2.3

3.4 3.6

ND

20

2500

ND

10 000

10

1100

45

350

11

2000

1.0

4

400

ND

1.8

40

2.3

3.6

ND

20

2500

ND

10 000

10

1100

45

350

11

2000

1.0

4

400

ND

1.8

40

2.3

3.6

ND

20

2000

ND

10 000

10

1100

45

350

11

2000

1.0

4

400

ND

1.8

40

2.3

3.6

ND

20

2000

ND

10 000

10

1100

45

350

11

2000

1.0

3

400

ND

1.8

40

2.3

3.6

Pregnancy 14–18 years 19–30 years 31–50 years Lactation 14–18 years 19–30 years 31–50 years

ND

17

3000

ND

8000

10

900

45

350

9

1700

1.0

3.5

400

ND

ND

34

2.3

3.6

ND

20

2500

ND

10 000

10

1100

45

350

11

2000

1.0

3.5

400

ND

ND

40

2.3

3.6

ND

20

2500

ND

10 000

10

1100

45

350

11

2000

1.0

3.5

400

ND

ND

40

2.3

3.6

ND

17

3000

ND

8000

10

900

45

350

9

1700

1.0

4

400

ND

ND

34

2.3

3.6

ND

20

2500

ND

10 000

10

1100

45

350

11

2000

1.0

4

400

ND

ND

40

2.3

3.6

ND

20

2500

ND

10 000

10

1100

45

350

11

2000

1.0

4

400

ND

ND

40

2.3

3.6

Note: A tolerable upper intake level (UL) is the highest level of daily nutrient intake that is likely to pose no risk of adverse health effects to almost all individuals in the general population. Unless otherwise specified, the UL represents total intake from food, water, and supplements. Due to a lack of suitable data, ULs could not be established for vitamin K, thiamin, riboflavin, vitamin B12, pantothenic acid, biotin, and carotenoids. In the absence of a UL, extra caution may be warranted in consuming levels above recommended intakes. Members of the general population should be advised not to routinely exceed the UL. The UL is not meant to apply to individuals who are treated with the nutrient under medical supervision or to individuals with predisposing conditions that modify their sensitivity to the nutrient. a Although the UL was not determined for arsenic, there is no justification for adding arsenic to food or supplements. b The ULs for magnesium represent intake from a pharmacological agent only and do not include intake from food and water. c Although silicon has not been shown to cause adverse effects in humans, there is no justification for adding silicon to supplements. d Although vanadium in food has not been shown to cause adverse effects in humans, there is no justification for adding vanadium to food and vanadium supplements should be used with caution. The UL is based on adverse effects in laboratory animals and these data could be used to set a UL for adults but not children and adolescents. e ND ¼ not determinable due to the lack of data of adverse effects in this age group and concern with regard to the lack of ability to handle excess amounts. Source of intake should be from food only to prevent high levels of intake. Source: Reprinted with permission from Dietary reference intakes for calcium, phosphorous, magnesium, vitamin D, and fluoride (1997); Dietary reference intakes for thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and choline (1998); Dietary reference intakes for vitamin C, vitamin E, selenium, and carotenoids (2000); Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc (2001); Dietary reference intakes for water, potassium, sodium, chloride, and sulfate (2005); and Dietary reference intakes for calcium and vitamin D (2011) by the National Academy of Sciences, courtesy of the National Academies Press, Washington, DC. These reports may be accessed via www.nap.edu.

Dietary References: US

Risk of inadequacy

1.0

EAR RDA

UL

0.5

1.0

0.5

0.0

Risk of adverse effects

430

0.0 Observed level of intake

Figure 2 The relationship between different descriptors of dietary reference intakes. Reprinted with permission from Dietary reference intakes: the essential guide to nutrient requirements, 2006 by the National Academy of Sciences, courtesy of the National Academies Press, Washington, DC.

preferences, including meal patterns, culture, religious beliefs, and taste. Perfectly adequate diets from the nutrient standpoint that are not eaten are nutritionally useless. When planning for individuals, the nutrient goal should be at least the RDA or AI to minimize risk of inadequacies. There is no increased risk of adverse effects when an individual’s intake is greater than the RDA as long as it remains below the UL. Similarly, when planning for populations or particular groups, it is important to ensure that virtually all individuals exceed the EAR without exceeding the UL. Many tools are available for diet planning. The USDA provides daily food plan guides (http://www.choosemyplate.gov/ myplate/index.aspx) by food groups for those with varied daily calorie needs. Additionally, the USDA periodically issues the Thrifty Food Plan (http://www.cnpp.usda.gov/USDAFoodPlansCostofFood.htm), which is a guide to selecting foods and menus for individuals using Supplemental Nutrition Assistance Program benefits to maximize the amount and nutritional value of food while receiving benefits.

Challenges, Additional Research, and Future Steps The development of the DRIs has been important in improving national assessment and monitoring of dietary intakes and planning of federal and other programs, but research challenges remain.

Tolerable Upper Intake Level The UL is the level of nutrient intake below which adverse health effects are unlikely. Adverse events have not been collected routinely on a systematic basis in studies of high levels of nutrient intakes requirements, and therefore, there are often very little data upon which to base the UL. Moreover, dose–response studies are rarely done at high levels, and in some cases, they would be unethical, yet in order to accurately determine a health risk associated with an intake above the UL, it is necessary to know the dose–response relationship between chronic intake of the nutrient and the observed adverse effect chosen. More data are needed in order to estimate dose–

response curves for most nutrients. A second problem with the UL is that the level of severity of the adverse effects upon which the UL is based varies greatly from one nutrient to another. For example, exceeding the UL on a chronic basis for pyridoxine is a major concern since it may produce irreversible nerve damage. In contrast, exceeding the UL for choline is likely of lesser concern since the UL is based on a fishy body odor. A final problem with the UL is its characteristics; it is based on a toxicological model with many safety factors that are based on expert judgments but not necessarily on empirically determined data. The UL is a defined as a sharp cutoff and is difficult to manipulate from the statistical standpoint.

Protein Requirement The current DRI for protein is based on the analysis of nitrogen balance studies, which have many limitations. Nitrogen balance may give falsely high estimates because it is easy to overestimate nitrogen intake and underestimate excretion due to incomplete collections and difficult to measure dermal and miscellaneous nitrogen losses that vary greatly under some environmental conditions. Protein requirements can also be determined using other techniques, such as stable isotopes, carbon balance, and amino acid flux. However, before new methods are adopted, there is a need to conduct comparative studies using the proposed techniques to harmonize them with the nitrogen balance studies of requirements. One alternative technique that has been proposed to estimate requirements of essential amino acids is the indicator amino acid oxidation method to determine indispensable amino acid (IDAA) requirements in humans. It is based on the idea that when one IDAA is deficient for protein synthesis, then the excess of all other IDAAs will be oxidized, including the amino acid marked with a stable isotope. Proponents claim that this method is robust, minimally invasive, more rapid, and sensitive to changes in amino acid intake than nitrogen-based studies. Studies using it have identified amino acid requirements that are much higher than those based on nitrogen balance.

Dietary References: US Better Experimental Data on Subgroups Many of the DRIs for specific life stage groups have been extrapolated from adults, due to the lack of data on the specific nutrient and life stage group. More studies of these groups are needed in order to set more accurate DRIs.

DRIs for Nonnutrient Bioactives? Nonnutrient bioactive constituents may have significant health benefits. They include food components such as fiber, carotenoids, glucosinolates, and flavonoids, among others. With the exception of fiber, DRIs are not established for these components. There are many challenges in determining recommendations for consumption of bioactives including their variable bioavailability and bioactivity, errors in intake measurement, and health benefits based on surrogate biomarkers of effect rather than health outcomes. Historically, much of the discussion around setting recommendations for bioactives was about the paucity and quality of scientific data available. Phenols, polyphenols, and flavonoids were excluded from consideration by the DRI panel at the time the DRIs were last revised because of the lack of sufficient data. Since then, there has been a substantial increase in research and available data, although randomized clinical trials of intakes of these constituents with health outcomes are still few in number. Some other types of guidance may be more appropriate than to establish DRI for them.

Chronic Disease Endpoints More attention needs to be paid to methods for linking nutrient intakes to chronic disease endpoints/outcomes and setting recommendations based on these.

Updates and Harmonization It may be time for reevaluation of the DRI development process. Perhaps, it should establish criteria for which nutrients are most in need of revision and then concentrate on revising one nutrient at a time rather than many nutrients. At present, the Food and Nutrition Board and Health Canada have received nominations for many nutrients thought to be in need of revision, of which four were selected for further consideration (vitamin E, sodium, magnesium, and omega 3 fatty acids). It remains to be seen what nutrients will be chosen. Finally, there is a need to harmonize efforts on setting requirements with other authoritative bodies in Europe, Australia, New Zealand, and elsewhere, to avoid needless overduplication of work.

See also: Elderly: Nutrition Requirements; Energy: Intake and Energy Requirements; Fatty Acids: Determination and Requirements; Infants:

431

Nutritional Requirements; Pregnancy: Metabolic Adaptations and Nutritional Requirements; Protein: Requirements.

Further Reading Institute of Medicine (2011) Dietary reference intakes for calcium and vitamin D. Washington, DC: The National Academies Press. Institute of Medicine (2006) Dietary reference intakes: the essential guide to nutrient requirements. Washington, DC: The National Academies Press. Institute of Medicine (2005a) Dietary reference intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein, and amino acids (macronutrients). Washington, DC: The National Academies Press. Institute of Medicine (2005b) Dietary reference intakes for water, potassium, sodium, chloride, and sulfate. Washington, DC: The National Academies Press. Institute of Medicine (2003a) Dietary reference intakes: applications in dietary planning. Washington, DC: The National Academies Press. Institute of Medicine (2003b) Dietary reference intakes: guiding principles for nutrition labeling and fortification. Washington, DC: The National Academies Press. Institute of Medicine (2001a) Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. Washington, DC: The National Academies Press. Institute of Medicine (2001b) Dietary reference intakes: proposed definition of dietary fiber. Washington, DC: The National Academies Press. Institute of Medicine (2000a) Dietary reference intakes for vitamin C, vitamin E, selenium, and carotenoids. Washington, DC: The National Academies Press. Institute of Medicine (2000b) Dietary reference intakes: applications in dietary assessment. Washington, DC: The National Academies Press. Institute of Medicine (1998a) Dietary reference intakes for thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and choline. Washington, DC: The National Academies Press. Institute of Medicine (1998b) Dietary reference intakes: a risk assessment model for establishing upper intake levels for nutrients. Washington, DC: The National Academies Press. Institute of Medicine (1998c) Dietary reference intakes: proposed definition and plan for review of dietary antioxidants and related compounds. Washington, DC: The National Academies Press. Institute of Medicine (1997) Dietary reference intakes for calcium, phosphorus, magnesium, vitamin D, and fluoride. Washington, DC: The National Academies Press. King JC and Garza C (2007) Harmonization of nutrient intake values. Food and Nutrition Bulletin 28(S1): S3–S12. Munro IC (2006) Setting tolerable upper intake levels for nutrients. The Journal of Nutrition 136: 490S–492S.

Relevant Websites http://choosemyplate.gov – USDA MyPlate. http://www.dietitians.ca/Knowledge-Center/Live-Events/Online-Courses/DietaryReference-Intakes.aspx – Dietitians of Canada: Dietary Reference Intakes Online Tutorials. http://fnic.nal.usda.gov/dietary-guidance/dietary-reference-intakes – USDA Natural Agriculture Library: Dietary Reference Intakes. http://www.iom.edu/About-IOM/Leadership-Staff/Boards/Food-and-Nutrition-Board. aspx – Institute of Medicine Food and Nutrition Board. http://ods.od.nih.gov/Health_Information/Dietary_Reference_Intakes.aspx – National Institutes of Health Office of Dietary Supplements: Dietary Reference Intakes.

Dietary Surveys: National Food Intake MC Ocke´, CTM van Rossum, and EJ de Boer, National Institute for Public Health and the Environment, Bilthoven, The Netherlands ã 2016 Elsevier Ltd. All rights reserved.

Introduction National food intake surveys are surveys that monitor the food consumption of groups of individuals that are representative for a national population or important population groups within a country. National food intake surveys provide insight into the consumption of foods, the intake of energy and nutrients, and the exposure to potentially harmful chemical substances. Regular repetitions of national food intake surveys show dietary trends in a population. Apart from national food intake surveys, other sources of national food data are available in countries. These data from food balance sheets and household budget and expenditure surveys are sometimes used as a surrogate for food intake data. However, each of them provides information about a different level in the flow of food production to consumption and on different aggregation levels. Food balance sheets give information on the national per capita food supply, and household budget surveys focus on food acquisition at the level of households, sometimes supplemented with foods acquired out of home. The data of food balance sheets and household budget surveys are thus not directly comparable with data from food consumption surveys (see Figure 1). Only national food intake surveys allow to estimate the population distributions of consumption of foods, intake of energy and nutrients, and exposure to chemicals and to do this for subgroups within a population such as by gender and age groups. Many countries conduct national food intake surveys. For the European Union, food intake data of 22 member states are available in the comprehensive food consumption database of the European Food Safety Authority (EFSA). The number of countries that carry out food intake surveys on a regular basis rather than incidentally is however limited. The most well-know

long-standing national food intake surveys are probably those in the United States, starting from the National Food Consumption Survey in 1935 to the ongoing dietary surveys as part of the National Health and Nutrition Examination Survey (NHANES). This article highlights major uses of national food intake data, important aspects about the design and methodology of food intake surveys and their European harmonization, limitations associated with food intake surveys, and future directions.

Major Uses and Stakeholders of National Food Intake Data Data from national food intake surveys together with data on the composition of foods and information on nutritional status are needed to develop and evaluate health, nutrition, and food policy. Examples are the evaluation of diet from a health perspective, simulating the impact of dietary policy measures, describing trends in food and nutrient intakes, and deriving optimal levels of food fortification. Moreover, the food intake data are essential to develop food frequency questionnaires, dietary indexes, and food-based dietary guidelines. Similarly, data from national food intake surveys in combination with data on contaminants, additives, or chemical constituents in foods are essential for the assessment of exposure to hazardous substances, as part of the risk assessment process. This information is also useful for the evaluation of new food legislation. Examples are dietary exposure assessment, derivation of safe maximum fortification levels, communication of risks associated with the food chain, estimations of the effects of particular diseases caused by infectious agents that can be transmitted between animals and humans, and identification of core foods for total diet studies.

Food available but not acquired

FOOD BALANS SHEETS National food supply

FOOD CONSUMPTION SURVEYS food consumed by individuals

HOUSEHOLD BUDGET SURVEYS Household food acquisition

Out of home food acquisition

Foods not consumed (by humans)

Figure 1 Relationships between food supply data, food acquisition data, and food consumption data.

432

Encyclopedia of Food and Health

http://dx.doi.org/10.1016/B978-0-12-384947-2.00237-3

Dietary Surveys: National Food Intake

Dietary exposure to food contaminants, additives

433

Evaluate nutritional quality, e.g. % adequate

Input for risk communication

Describe dietary trends

Derive safe maximum levels for foods

Simulate impact of food policies National food intake data

Develop food frequency questionnaires

Derive optimal levels of food fortification

Develop dietary indexes

Describe diet costs

Derive food-based dietary guidelines

Describe environmental impact of diets

Figure 2 Examples of uses of food intake surveys.

Additionally, food intake data can be used for the assessment of the environmental impact of diet or to describe dietary costs (see Figure 2). Stakeholders of national food intake data are ministries of health, ministries of agriculture, food industry, dieticians, patient organizations, branch organizations, journalists, universities, and national and international health and food safety organizations like the World Health Organization (WHO), Food and Agriculture Organization (FAO), and EFSA. Scientists in public health or nutrition institutes, food safety agencies, universities, and research departments of the larger food industries often have an interest to conduct specific statistical analyses in national food intake databases themselves.

Study Design, Data Collection, and Handling Two features are critical for the success of national food intake surveys. The first critical aspect is to obtain a study population that is representative of the target population. The second challenge is to collect food intake data that are valid and precise and contain sufficient detail for their purposes.

Study Population When planning a national food intake survey, the first step to be taken is defining the target population. The target population is the entire set of individuals to which findings of the survey are to be extrapolated. Typically, a broad age group within a national population is defined as the target population for a food intake survey. The target population may not be the same as the study population, that is, the population that can be defined accurately and reached in a study. In order to precisely define the study population, it is necessary to agree on

exclusion criteria. Institutionalized subjects are often not included in food intake surveys as their inclusion makes the sampling procedures and/or the field surveys too complicated. Also, language barriers are commonly used as exclusion criteria. In order to obtain a representative study population for the target population, excluding subjects due to language barriers or other practical reasons should be kept to a minimum. The sampling frame is the list of units from which the sample is drawn. The availability of sampling frames varies from country to country. For national food intake surveys, the most complete lists can be chosen among population registers, electoral lists, and census lists. Alternative sampling frames are telephone lists, lists of schools or day care centers, market research panels, and lists of medical centers or general practitioners. Great care must be taken when making decisions on the sampling frame, particularly with regard to bias due to undercoverage. Overcoverage of the sampling frame normally can be detected and dealt with during the fieldwork. Specific subgroups within a population might need a different sampling frame, since electoral list is only suitable for adults and list of day care centers or schools is especially suitable for young or older children, respectively. Sampling from the sampling frame can be done in various ways. It is considered a good survey practice to use a sampling design that yields a probability sample. Under such a design, each sample unit of the sampling frame has a nonzero probability of being selected. It does not necessarily require that every person has the same probability of being selected. Stratified sampling, for example, sampling by subgroups of age and gender, is often conducted in food intake surveys to ensure that there are a sufficient number of people in each subgroup to allow meaningful analysis. For practical reasons, cluster sampling or multistage sampling can be applied. In cluster

434

Dietary Surveys: National Food Intake

sampling, first, a random sample of naturally occurring units or groupings within the population is taken (e.g., areas, cities, or schools). Then, either all members of the chosen clusters or a random selection from among them is included in the sample. Multistage sampling is an extension of cluster sampling where a hierarchy of clusters are chosen going from larger to smaller. The required size of a food intake survey population depends on the variability in intake, the statistics of interest (i.e., mean, percentiles, or variance), and the level of precision needed. Information on intake variability (from previous studies) is therefore needed in order to determine the number of subjects to include in a survey. In the European Food Consumption and Validation (EFCOVAL) project, the sample size per country was estimated to be, on average, equal to 2000 subjects for a desired precision of 5% in nutrient and food group intake. In simple random samples, high percentiles can be assessed with sufficient accuracy if the minimum sample sizes for the

95th, 97.5th, and 99th percentiles can be estimated to be equal to 160, 320, and 800, respectively. When results are presented for different subgroups within a study population, the minimum sample size has to be available for each subgroup. The expected response rate should be taken into account in determining the sample size for achieving a certain number of actual participants. In the case of recruitment in waves through a year, adjustments in the number of invited participant can be made according to the actual response rates for specific strata in previous waves. Moreover, it is important to collect information on nonrespondents in order to eventually correct for lack of representativeness of the study population for the target population. For active nonparticipants, it is possible to administer a concise nonrespondent questionnaire (among those willing); for passive nonparticipant with which no contact was established, only information from the sampling frame can be used (see Figure 3).

Sample from sampling frame, e.g. population register

Periodic subsample for recruitment in waves (every month, season etc,)

Recruitment, e.g. by invitation letter and brochure

Passive nonparticipant

Participant

First 24-h dietary recall, other measurements e.g. general questionnaire, height/weight 1-6 weeks in between

Second 24-h dietary recall

Figure 3 Example of the logistic flow of a food intake survey.

Active nonParticipant

Concise non-response questionnaire

Dietary Surveys: National Food Intake Dietary Assessment Traditionally, four main methods of dietary assessment are distinguished. These are food records, dietary recalls, food frequency questionnaires, and dietary history questionnaires. Food frequency questionnaires capture usual intake, but they are not quantitatively precise and thus cannot estimate the percentage of the population with inadequate intakes or too high intakes. This was clearly demonstrated in a biomarkerbased validation study in the United States, the Observing Protein and Energy Nutrition (OPEN) Study. The traditional dietary history also captures usual intake but is extremely labor-intensive and involves a high subject burden with an interview of several hours, whereas the modern version of a dietary history method has similar problems to food frequency questionnaire with estimating intake quantitatively. Therefore, food frequency methods and dietary history methods are not suitable for national food intake surveys. Well-conducted food records and dietary recalls are better able to quantify food intake for the specific days they cover. Since usual intake is of interest for food intake surveys, at least two repeated nonconsecutive food records or dietary recalls are required together with statistical modeling of usual intake. With the statistical modeling, the within-person day-to-day variation in the intake data can be accounted for. For foods that are important contributors of nutrients and that are consumed by less than 50% of the population, additional information on the frequency of consumption has added value. Examples of these products are types of dietary supplements and liver. Whether a food record or dietary recall is the preferred dietary assessment method for food intake surveys depends on the survey population. In populations with notably low response rates, the preferred method is the method with less respondent burden; this is the dietary recall. The dietary recall method is also preferred in populations with high illiteracy levels. In other populations, food records might be preferred. The food record is a prospective method and therefore does not have the disadvantage that it relies on memory such as the retrospective 24 h dietary recall method. However, it is wellknown that people change their diets because of keeping a record of everything that is consumed. For harmonized panEuropean food intake surveys, EFSA advises to apply dietary recalls as dietary assessment method, particularly because of the lower burden to the participants. In order to obtain intake data that are usable for many purposes, it is important to collect information on all foods, beverages, and dietary supplements consumed. The information should contain a sufficient level of detail for the different purposes, for example, information on fat content, sweetening (sugar and/or artificially sweetened), fortification, preparation method, conservation method, and package material. Brand name information can help to ascertain aspects like sweetening and fortification that are often not known by consumers. It is also necessary to estimate the consumed amount for each eaten food. There are different options, which can be used in combination: estimation in household units like glasses or spoons, estimation in natural units, for example, for eggs or apples, estimation in commercial units for products like chocolate bars and sweets, estimation using food models or food photographs, and estimation in gram or milliliter.

435

To conduct a high-quality standardized 24 h dietary recall, it is advised to use a computer-administered software for 24 h recalls. With the software, the interviewer is guided by automatic response routing to ask tailored questions and prompts for specific foods. Using multiple steps, the system helps to ensure completeness and consistency across interviewers. There are automated links to food lookup tables and data quality checks. It is of major importance to test any software before applying it in a survey; preferably, a validation study should be conducted. Examples of carefully developed and extensively tested and validated software are the five-step multiple-pass method used in the NHANES in the United States and GloboDiet, previously known as EPIC-Soft, used in food intake surveys of Austria, Belgium, France, Germany, Malta, the Netherlands, and Switzerland. Specific subgroups within a population might need a different dietary assessment method. For example, a 24 h dietary recall is less suitable for young children in case the child is often not together with the parent or caretaker that is interviewed. Also for older adults with a higher likelihood of impaired short-term memory, a 24 h dietary recall is less suitable. In both age groups, a dietary record or a food diary and 24 h dietary recall could be an option. For less developed countries, more emphasis is needed for feasible and robust methods.

Supplemental Information to be Collected from Participants Depending on the specific aims of a food intake survey, other information is collected besides the dietary data. All food intake surveys collect information on background characteristics of the participants, usually including sociodemographic information and information on dietary practices. This enables to stratify the results by sociodemographic subgroups and to judge the representativeness of the study population. Because of the high prevalence of overweight, additional information on physical activity and anthropometric information like height, weight, and waist circumference is also collected in many food intake surveys. Information on body weight is also of prime importance if chemical exposure from the diet is to be compared with health limits. These health limits are usually expressed per kilogram body weight. Moreover, some surveys collect information on health and others on food choice determinants, biological marker for nutritional status, or aspects important for microbiological risk associated with food consumption. It is outside the scope of this article to detail methods for collecting these data.

Administration Protocol The study design must consider that in order to capture the interseasonal variability in consumption patterns, subjects must be uniformly distributed over the four different seasons and each seasonal sample must represent each survey stratum (e.g., on age and sex). This can be done by recruitment of participant in waves throughout the year (see Figure 2). It is of prime importance to monitor response and the completed number of participants and to adjust procedures if response or the completed number of participants falls behind.

436

Dietary Surveys: National Food Intake

The survey calendar must be organized in such a way that all days of the week are represented proportionally in the food consumption data. Specific emphasis on an adequate proportion of weekdays and weekend days at population group level is needed, since this is the main source of variation in intake over the days. For food intake surveys with nonconsecutive dietary records or recall, it is advised to have an interval of 1–6 weeks in between both administrations. More than 6 weeks in between, which might be an option theoretically, might result in losing study participants and is therefore not advised. The administration of the interviews is usually done face to face, although in some surveys, telephone interviews are conducted. Face-to-face interviews can be conducted at a study center or at the home of the participant. The latter has the advantage that consumed foods and used household measures are available to show to the interviewer. Interviewers should preferably be trained dietitians or nutritionist; although, with more extensive training on consumed foods and recipes, also interviewers with other backgrounds are used in practice. During the survey period, refresher trainings are recommended. Moreover, it is of prime importance to monitor the quality of the interviewers and provide them feedback to improve their work.

Data Cleaning and Analyses After data collection, data cleaning is an important phase. It is recommended to start data cleaning immediately after the first data are received from the interviewers, rather than waiting until the survey period is finalized. Data cleaning should include checking for missing, inconsistent, and extreme values and considering data-related remarks made by interviewers. When the food intake data are checked, further handling can include linkage to nutrient databases for foods and dietary supplements, databases with concentrations of chemicals in foods, or other food-related databases. More information on nutrient databases can be found in this encyclopedia. Subsequently, statistical analyses can be performed. In these analyses, survey weights can be used to make the data representative for the target population. A commonly applied correction technique for lack of representativeness is weighting adjustment. It assigns an adjustment weight to each survey respondent. Persons with characteristics that are underrepresented get a weight larger than one, and those in overrepresented groups get a weight smaller than one. In the computation of survey results, not just the values of the variables are used, but the weighted values. When usual intake is of interest, intake or exposure per person per day should be calculated and then usual intake modeling should be applied. In these models, the withinperson variation in intake is removed from the data. Various tools to model usual intake exist. In the Statistical Program to Assess Dietary Exposure (SPADE), usual intake can be modeled for the combination of intake from foods and drinks as well as dietary supplements measured on two specific days. In the Monte Carlo Risk Assessment (MCRA) tool, usual exposure to chemical substances can be modeled, dealing, for example, with occurrence data below the detection limit and having the option to check the input data quality.

An important step in data analyses is evaluation of dietary intake against dietary reference values for nutrients and against health limits for food chemicals. Procedures for proper evaluation are described elsewhere in this encyclopedia.

Limitations Two major challenges of food intake surveys are the internal validity and the external validity. Internal validity of national food intake surveys is never perfect, since diet cannot be measured without error; both random error (including day-to-day variation) and systematic bias (social desirability) are present in the data. Important aspects within dietary recalls are the correctness of the reported foods (both omissions and intrusions occur), the completeness of food description, and the correctness of portion size estimation. In practice, dietary assessment based on self-reporting usually leads to underestimation of energy intake. Moreover, dietary assessment in combination with using food composition databases has additional limitations for specific components in the diet. This is, for example, the case for dietary exposure assessment of mycotoxins. Since the concentration of mycotoxins in foods is very variable due to, for instance, weather and storage conditions, duplicate diet studies or total diet studies are better equipped to estimate mycotoxin exposure. Another example is estimation of sodium intake. This is difficult because of the large variation in sodium concentrations in processed foods and recipes, as well as the difficulty of estimating discretionary salt intake. Dietary assessment of sodium (rather than urinary excretion) does however enable identification of important dietary sources of sodium, which can inform public health interventions to lower sodium intake. What is important is whether the dietary assessment method is suitable for providing useful analytic measurement for a given purpose and context. For many purposes and in many contexts, 24 h dietary recall data from national food intake surveys proved to be useful in helping to address important research and policy questions, despite their known errors. Strategies to optimize the impact of food surveys are to have more transparency of raw research data, consistent data-staging techniques, improved data analysis, and reporting indicators for dietary intake quality such as indicators for underreporting. External validity is impacted by low response rates that often occur. In practice, response rates between 26 and 97 are observed in national food intake surveys, with the majority between 42% and 71%. Low participation rates and selective nonparticipation may cause bias to the survey results based on participants alone. It is known that respondents and nonrespondents might differ in their socioeconomic and demographic status as well as in their health and lifestyle behaviors. Therefore, for achieving a representative study population, it is of utmost importance to use all means to obtain the highest response rate possible. Strategies to optimize response rates are choosing data collection methods with a lower burden for the respondents, providing a suitable incentive, flexibility in recruiting and participation (times available for the interview, second call if a no-show, etc.), personal contact in recruitment, and awareness of the study by media coverage.

Dietary Surveys: National Food Intake

Harmonization of National Food Intake Data Across Countries in the European Union National food intake surveys in Europe are heterogeneous with respect to dietary assessment methodology and number of days for which dietary data are collected. This hampers the comparison of results across countries, for example, in dietary exposure assessment using the data of the comprehensive food consumption database such as conducted by EFSA. Therefore, EFSA stimulates European Union member states to collect national food intake data in a harmonized way and prepared guidance for this in 2009 with an update in 2014. Moreover, EFSA provides seed money to countries that conduct food intake surveys in accordance with EFSA guidance. This is called the EU Menu process. The guidance of EFSA was partly based on the results of previous European research projects, such as summarized in the succeeding text. In the European Food Consumption Survey Method (EFCOSUM) project, recommendations for reliable and comparable pan-European collection of food intake data were formulated. As the most suitable method to collect internationally comparable data on population means and distributions of actual intake, the 24 h recall was recommended, to be conducted at least twice. This would also allow for the estimation of usual intake by a modeling technique that separates intraand interindividual intake. GloboDiet, at that time called EPIC-SOFT, was considered the most suitable software for a standardized data collection in a pan-European survey. For a number of micronutrients, that is, vitamin D, folate, sodium, iron, and iodine, the use of biomarkers was recommended. Subsequently, the EFCOVAL project further developed and validated the pan-European food consumption method recommended by the EFCOSUM project. The validation study confirmed that the repeated 24 h dietary recall using GloboDiet in combination with a food propensity questionnaire and modeling of usual intake is a suitable method for a pan-European food intake survey. This conclusion applied to healthy adults and possibly to children aged 7 years and older. As next step, it was recommended to provide and standardize an implementation plan to facilitate this methodology in European countries. The plan should account for maintenance and updates, sampling designs, national surveillance programs, tailored capacity building and training, and linkage to food composition and occurrence databases. The objective of the PILOT-PANEU project was to develop, test, and evaluate the applicability of tools and procedures for conducting a pan-European food intake survey including adolescents, adults, and elderly people. The dietary assessment methodology consisted of two nonconsecutive 24 h recalls performed with GloboDiet methodology. The consortium made a number of recommendations for the improvement and further development of the tools and procedures. Assuming that the recommended modifications are made, the tools and procedures developed and tested were considered suitable for the EFSA EU Menu process in adolescents, adults, and the elderly. Similarly, in the ‘Pilot study for the Assessment of Nutrient intake and food Consumption Among Kids in Europe’ (PANCAKE), the feasibility of tools and procedures for a pan-European food intake survey among children 0–10 years was tested. This included two alternative dietary assessment

437

methods. One method consisted of a 3-day food diary, which was checked with a parent, and data were entered afterward using GloboDiet. The alternative method consisted of two nonconsecutive 1-day food diaries followed by GloboDiet completion interviews. Both dietary assessment methods with related tools and administration protocols were evaluated as feasible. The administration protocol with two 1-day food diaries with completion interviews was judged to offer more advantages for the EFSA EU Menu survey in children 0–10 years. It offered more complete description of the consumed foods, was more suitable to estimate usual intake, and resembled more the method in adults.

Future Directions In the future, it is important to improve data collection methods and to collect food intake data in connection to other information. First, improving data collection methods should focus on methods that have a small participant burden or that are attractive to participants, so that response rates increase. The use of modern technologies might help for this purpose, although for some population groups, it might not be appropriate. Additionally, future data collection methods should be less labor-intensive than current data collection methods, since the costs of the current methods are high and increase because of the dynamic and increasingly complex food market. Examples of promising data collection methods are online 24 h dietary recalls, devices that automatically take pictures of all food consumed, and apps for food records using bar codes and food pictures for portion sizes, or automatic recognition of photographed consumed food and volumes. These promising technologies need to be extensively tested in the general population and important subgroups of the population and validated positively before they can be incorporated in food intake surveys. At the moment, various interesting developments and validation studies with modern technologies are ongoing. Second, it has added value to conduct national food intake surveys in connection to biomarker and other data collection. Combining with nutritional status monitoring and collecting information of biomarker of food intake allow for better interpretation of food intake results and provide insight in the validity of the results. Combining with data collection on other lifestyle information has the advantage that risk groups with multiple unhealthy lifestyles can be identified. Moreover, in order to get insight in ways to improve dietary intake, it is an asset to collect food intake data in combination with information on determinants of food choice, the food environment, and economical aspects of diet and health. Such integrated food intake surveys ask for ingenious study designs so that the participant burden can be kept acceptable, and not all information has to be collected from each participant.

Conclusions Conducting regular national food intake surveys is important for the development and evaluation of nutritional and food safety policies. The current best practice in food intake surveys

438

Dietary Surveys: National Food Intake

is to obtain a representative sample from a national population register. For each participant, at least two nonconsecutive food records or 24 h dietary recalls combined with a short food frequency questionnaire need to be collected supplemented with at least sociodemographic background information. Statistical modeling should be used to estimate usual intake. For pan-European harmonization of food intake surveys, dietary recalls rather than food records are advised. Standardized collection of 24 h dietary recalls with much detail can best be done using validated software such as GloboDiet. Overall, study preparation and data collection and data handling of food intake surveys with much detail in the food description are labor-intensive and therefore costly. Using the collected dietary data to their full potential will however lead to a return on investment. Potential uses are manifold. Moreover, investments can become smaller if the same methods are used repeatedly and shared with other groups. Output and lessons can increase if the collected data are easily discovered, user-friendly to access and understand, and freely available. The trend toward open data helps here. This article described that two features are critical for the success of national food intake surveys. The first is the internal validity or validity of the collected food intake data. The second is the external validity or the representativeness of the survey population for the target population. In the practice of food intake surveys, we have to deal with suboptimal situation that diet cannot be measured without error and decreasing response rates. Future directions should address both these challenges and time and cost-efficiency.

See also: Dietary Exposure Assessment; Dietary Practices; Dietary References: US; Food Composition Databases.

Conway JM, Ingwersen LA, and Moshfegh AJ (2004) Accuracy of dietary recall using the USDA five-step multiple-pass method in men: an observational validation study. Journal of the American Dietetic Association 104(4): 595–603. de Boer EJ, Slimani N, van’t Veer P, et al. (2011) The European Food Consumption Validation Project: conclusions and recommendations. European Journal of Clinical Nutrition 65(Suppl. 1): S102–S107. Dekkers AL, Verkaik-Kloosterman J, van Rossum CT, and Ocke MC (2014) SPADE, a new statistical program to estimate habitual dietary intake from multiple food sources and dietary supplements. Journal of Nutrition 144(12): 2083–2091. Edwards PJ, Roberts I, Clarke MJ, et al. (2009) Methods to increase response to postal and electronic questionnaires. Cochrane Database of Systematic Reviews (3): MR000008. European Food Safety Authority (2014) Guidance on the EU Menu methodology. EFSA Journal 12(12): 77. Hebert JR, Hurley TG, Steck SE, et al. (2014) Considering the value of dietary assessment data in informing nutrition-related health policy. Advances in Nutrition 5(4): 447–455. Kipnis V, Subar AF, Midthune D, et al. (2003) Structure of dietary measurement error: results of the OPEN biomarker study. American Journal of Epidemiology 158(1): 14–21 discussion, 22–26. Kirkpatrick SI, Subar AF, Douglass D, et al. (2014) Performance of the automated selfadministered 24-hour recall relative to a measure of true intakes and to an interviewer-administered 24-h recall. American Journal of Clinical Nutrition 100(1): 233–240. Merten C, Ferrari P, Bakker M, et al. (2011) Methodological characteristics of the national dietary surveys carried out in the European Union as included in the European Food Safety Authority (EFSA) Comprehensive European Food Consumption Database. Food Additives and Contaminants. Part A: Chemistry, Analysis, Control, Exposure & Risk Assessment 28(8): 975–995. Ocke M, Brants H, Dofkova M, et al. (2014) Feasibility of dietary assessment methods, other tools and procedures for a pan-European food consumption survey among infants, toddlers and children. European Journal of Nutrition . Slimani N, Casagrande C, Nicolas G, et al. (2011) The standardized computerized 24-h dietary recall method EPIC-Soft adapted for pan-European dietary monitoring. European Journal of Clinical Nutrition 65(Suppl. 1): S5–S15. Thompson FE and Subar AF (2013) Dietary assessment methodology. In: Coulston AM, Boushey CJ, and Ferruzzi MG (eds.) Nutrition in the prevention and treatment of disease, 3rd ed., pp. 5–46. Amsterdam: Elsevier. Vandevijvere S, Monteiro C, Krebs-Smith SM, et al. (2013) Monitoring and benchmarking population diet quality globally: a step-wise approach. Obesity Reviews 14(Suppl. 1): 135–149.

Further Reading Ahuja JK, Moshfegh AJ, Holden JM, and Harris E (2013) USDA food and nutrient databases provide the infrastructure for food and nutrition research, policy, and practice. Journal of Nutrition 143(2): 241s–249s. Andersen LF, Lioret S, Brants H, et al. (2011) Recommendations for a trans-European dietary assessment method in children between 4 and 14 years. European Journal of Clinical Nutrition 65(Suppl. 1): S58–S64. Brussaard JH, Lowik MR, Steingrimsdottir L, et al. (2002) A European food consumption survey method – conclusions and recommendations. European Journal of Clinical Nutrition 56(Suppl. 2): S89–S94.

Relevant Websites http://www.cdc.gov/nchs/nhanes.htm – Centers for Disease Control and Prevention. National Health and Nutrition Examination Survey. http://www.efsa.europa.eu/en/datexfoodcdb/datexeumenu.htm – European Food Safety Authority. EU Menu. http://www.rivm.nl/en/Topics/D/Dutch_National_Food_Consumption_Survey – National Institute for Public Health and the Environment. Dutch National Food Consumption Survey.

Drying: Effect on Nutrients, Composition and Health SV Crowley and JA O’Mahony, University College Cork, Cork, Ireland ã 2016 Elsevier Ltd. All rights reserved.

Introduction In their natural state, most foods contain a significant quantity of water. In foods with a high water activity (aw), spoilage and/or pathogenic microorganisms can grow; in addition, degradative physicochemical changes occur more rapidly in high aw foods. One of the most effective methods of increasing both the shelf life and safety of foods is through dehydration. Traditional drying methods, such as sun-drying, are slow and manual in nature, and can incur losses in yield due to difficult-to-predict factors relating to the local climate and indigenous wildlife. These traditional methods of dehydration are still successfully used in developing countries and also in the production of artisan or minimally processed food. However, the modern food supply chain is a global one, and the increased affluence of consumers has resulted in a strong demand for the year-round availability of a diverse range of foods. In industrial countries, this demand has contributed to the general adoption of mechanized drying technologies (e.g., sprayand freeze-drying) and a proliferation of food powders. Food powders are nutrient-dense, low-volume products that can be easily transported nationally and internationally. The drying process itself can also be harnessed to encapsulate nutrients deemed to have specific health benefits, such as probiotics, fatty acids, and phytochemicals. These encapsulated nutrients can be stored for use as versatile ingredients in the fortification of a wide range of foods, which could allow their targeted delivery to a larger proportion of the world population. In this article, the influence of the drying process and powder storage on the stability of key nutrients in different food systems is discussed. Certain dried foods, such as herbs and spices, are omitted, as the low levels at which they are consumed were considered to render their nutritional contribution to the average diet negligible. A comprehensive overview of the effects of drying on a range of foods and nutrients is provided, including fruits, fruit juice, vegetables, egg, milk, wine, algae, infant formulas, enteral formulas, bacteria, and plant-derived oils. This overview is based on recent research, which has primarily focused on the effect of drying conditions (e.g., temperature and flow rate) and postdrying storage conditions (e.g., relative humidity and light exposure) on the stability of key nutrients. The effect of bulk density on nutrient dosage in nutritional powders and the role of accelerated stability tests and fortification overages in formulation design, two increasingly important topics in modern manufacturing practices for dried foods, are also highlighted before a general closing discussion on the link between dried foods and health.

Effect of Drying on Nutrients Changes in Macronutrients During Drying and Powder Storage Much of the focus of this article is on the changes that occur in the micronutrients present in foods during drying and

Encyclopedia of Food and Health

subsequent storage. Although reactions occurring between macronutrients, such as those involving proteins and carbohydrates during nonenzymatic browning, are of nutritional consequence, individual macronutrients in food are generally not impacted to a significant degree by the drying process alone. In this section, the changes that occur in the macronutrient fraction are discussed briefly. The factors influencing the degradation of macronutrients during the drying of foods are also summarized in Table 1.

Protein Protein is naturally present in many common dried products, including milk and egg powders, and can play a significant role in the encapsulation of lipids and probiotics. In addition, advances in processing technology have resulted in an increased availability of dairy-derived (whey protein and casein) and plant-derived (soy and rice) protein powders, bars, and beverages, which are popular product categories among young adults engaged in athletic activities; on the other hand, an aging or ‘graying’ global population is likely to result in increased instances of diseases such as pathological sarcopenia, which is accompanied by muscle wastage. In both cases, consumption of adequate quantities of protein is essential to aid in the process of muscle synthesis. Modification of the structure of proteins (i.e., denaturation and aggregation) has been associated with altered biofunctional properties such as allergenicity and digestibility, but these effects are generally not attributed to the drying process itself. Additionally, these structural modifications can occur in predrying processing, such as the pasteurization and evaporation treatments applied to some dairy systems. In the presence of carbohydrates, however, proteins may undergo structural modifications due to the Maillard reaction, a nonenzymatic browning process. This aspect is covered in ‘Carbohydrate.’

Carbohydrate Carbohydrates are important structural components in dried food matrices. In powders, they are frequently the primary encapsulating material for other nutrients (e.g., fatty acids and probiotics). Degradative changes associated with carbohydrates of nutritional consequence are mostly limited to the generation of Maillard reaction products (MRPs) through nonenzymatic browning reactions between reducing sugars and amino acids. The generation of MRPs has been linked with increases in cytotoxicity, mutagenicity, and carcinogenicity; however, MRPs have also been associated with positive effects such as increased antioxidant activity. In addition to health effects, MRPs are associated with color and flavor development in foods. However, while the sensory attributes that Maillard browning imbues on certain foods, such as bakery and confectionary products, are often desirable, browning in food powders is typically considered a defect. Investigations of MRP generation in infant formula powders have been frequent, as

http://dx.doi.org/10.1016/B978-0-12-384947-2.00242-7

439

440 Table 1

Drying: Effect on Nutrients, Composition and Health General properties of the main macronutrients in foods related to the influence of drying on their nutritional quality

Macronutrient name(s)

Main degradative reaction

Dryinginduced changes

Storageinduced changes

Proteins

Denaturation

Negligible

Negligible

Predrying thermal processes (e.g., pasteurization and evaporation)

Carbohydrates/ proteinsa

Browning

Minor

Significant

Drying technology, drying temperature (spray-drying), feed composition (type of sugar), storage conditions (relative humidity, temperature, light exposure), crystallization of sugars

Lipids

Oxidation

Significant

Significant

Emulsion state (e.g., fat globule size distribution, emulsifier type), encapsulation efficiency, protein/fat ratio, crystallization of sugars

Key influencing factors

Some potential adverse nutritional consequences Protein digestibility+ Protein allergenicity* Cytotoxicity* Mutagenicity* Carcinogenicity* Lysine levels in food+ Cytotoxicity* Genotoxicity* PUFAb biofunctionality+

a

Nonenzymatic browning requires amino acids in addition to carbohydrates. PUFA, polyunsaturated fatty acid.

b

they are nutritional products high in both lactose and protein. MRPs can be generated during spray-drying, with reports highlighting the impact that spray-drying has had on infant formulas, resulting in greater loss of lysine than observed under processes of sterilization; however, it is also worth pointing out that higher concentrations of MRPs have also been measured in sterilized infant formulas compared with spraydried formulas. Different drying techniques and feed compositions can influence the physical properties (e.g., porosity and morphology) and chemical behavior (e.g., crystallization and water sorption) of the dried material during storage, which in turn can influence the generation of MRPs. MRP generation also occurs in powders during storage. Environmental conditions during the storage of infant formula powders (e.g., relative humidity, temperature, and light and oxygen exposure) have a strong impact on the rate of generation of MRPs.

Lipids Lipids are a primary contributor to the energy density and texture of foods. Lipid-containing foods are typically emulsions, consisting of lipids dispersed in a continuous water phase (O/W emulsion). Emulsified systems are intrinsically unstable, and to improve their stability, multiple strategies are employed, including the reduction in fat globule size through physical disruption (e.g., indent homogenization and microfluidization), increasing the viscosity of the continuous phase (e.g., addition of hydrocolloids), and the addition of amphiphilic molecules that adsorb at the oil–water interface and reduce interfacial tension (e.g., lecithin). When emulsions are dried, the continuous phase is removed and a solid carbohydrate/protein matrix forms around the fat. Effective encapsulation reduces the levels of free fat at particle surfaces and is dependent on the properties of the emulsion to be dried, the nature and level of the encapsulant, and the drying conditions. Oxidation of specific components in the lipid fraction, namely, fatty acids and cholesterol, is associated with losses in nutritional quality and is discussed in more detail in ‘Fatty acids’ and ‘Cholesterol,’ respectively.

Changes in Micronutrients During Drying and Powder Storage Fatty acids Consumption of foods that are rich in polyunsaturated fatty acids (PUFAs) has been associated with numerous health benefits. Soybeans are an example of a food that is naturally rich in PUFAs. Microwave-assisted drying has been reported to result in less PUFA degradation in soybeans compared to forcedconvection drying, and PUFA degradation was independent of drying temperature if soybeans were dried to the same final moisture content. It has been demonstrated that sundrying or freeze-drying resulted in higher retention of seaweed PUFAs compared to oven-drying. Although drying can result in PUFA degradation, it may be preferable to canning as a method of preserving seaweed, with one study reporting higher levels of total PUFAs in five dried seaweeds compared to two canned varieties. Infant milk formulas are frequently fortified with PUFAs, as long-chain PUFAs are abundant in human milk and are believed to contribute to the development of cognition and visual perception. Infant formulas contain metals that are known to be pro-oxidizers, such as iron. Iron-binding proteins, such as lactoferrin, have been shown to protect against oxidation when added to infant formulas. Methods for the encapsulation of PUFA-rich oils (e.g., fish oil) have been developed, but the protective effect of the encapsulant can diminish during processing, and oxidation of PUFAs remains a challenge. It has been reported that fortification of infant formula with microencapsulated fish oil resulted in higher peroxide values after 18 months. Reduced oxidation has been shown in infant formulas where casein was the dominant protein, compared to whey protein-dominant formulas, while higher levels of carrageenan or lecithin have also been demonstrated to reduce oxidation in infant formulas. Higher levels of PUFAs, elevated storage temperatures, and long storage duration generally result in increased oxidation. There is ongoing research into the potential for the encapsulation of plant-derived oils that are high in PUFAs, such as walnut and chia oils, using spray-drying, with some recent

Drying: Effect on Nutrients, Composition and Health research showing that homogenization of a chia oil emulsion had a greater impact on the effectiveness of sodium caseinate–lactose encapsulation than the temperature at which the emulsion was spray-dried. Encapsulation of oils by spraydrying in the presence of hydroxypropyl methylcellulose, with or without rosemary oil (antioxidant), improved the oxidative stability of walnut oil but resulted in greater oxidation in chia oil; it was suggested that the poor stability of encapsulated chia oil was linked to chemical damage caused during spray-drying.

441

lower phenol losses during drying of extracts of phenols from black currant pomace. Encapsulation of a phenol extract derived from olive pomace by spray-drying at different inlet temperatures (130 or 160  C), feed flow rates (5 or 10 ml min1), and maltodextrin levels (100 or 500 g l1) has also been studied, with the highest polyphenol levels at an inlet temperature of 130  C, a feed flow rate of 10 ml min1, and a maltodextrin level of 100 g l1; it was also found that the powder underwent the least degradation of phenols when stored in dark conditions at low temperatures.

Cholesterol The oxidation of cholesterol with a concomitant increase in the levels of cholesterol oxidation products (COPs) can occur during the manufacture of egg powders, which is a concern as COPs have been linked with disease in humans. In other dried products, such as spray-dried milk, cream, and infant formula powders, the levels of COPs are less of a concern. High outlet air temperatures and the generation of nitrogen oxides in a gasfired air heater during spray-drying resulted in increased generation of COPs. In addition, COPS can form during storage of egg powders, but this can be controlled by the presence of antioxidants in the powder, such as ascorbyl palmitate and tocopherols. Spray-dried egg powders contained more COPs than freeze-dried varieties, with the generation of COPS being higher in those powders produced from liquid whole egg than liquid egg yolk alone.

Phytochemicals Phytochemicals are plant-derived substances, such as polyphenols and flavonols, which are purported to have various bioactive properties. Studies have found that the levels of phenolic compounds in dried kale were highest when freeze-drying was used compared to air-drying, while blanching before drying led to reduced levels in kale dried by either method. Recent research has shown that Irish brown seaweed dried in a hot air oven at different temperatures (25, 35, or 40  C) increased the levels of phytochemicals as drying temperature increased and that increasing temperature of rehydration of dried Irish brown seaweed in water results in faster rehydration times but lower levels of phenols. It has been shown that drying conditions have no impact on the phytochemical contents of apricots, currants, or prunes, but do have an effect on grapes, while sun-drying has little effect on the levels of phenolic compounds in Italian bell peppers. A comparison of air- and freeze-drying for their ability to retain phytochemicals in physiologically dropped, unmatured citrus fruits reported that while air-drying was better at retaining flavonols, freeze-drying exhibited superior retention of phenols. A phenol-rich ingredient, named piraltin, has been manufactured by freeze-drying red wine; piraltin contained 70% of the polyphenols in the wine and was alcohol-free. Phenol levels in spray-dried bayberry juice, with maltodextrin as encapsulant, were nearly 100% of those in the original juice; when stored at 4  C, the levels of polyphenols in the bayberry juice powder only decreased by 83% of viable bacteria remained; the survival of the freeze-dried probiotic during storage was best at the intermediate levels of moisture, which were studied (2.8–5.6%). Researchers have spray-dried L. paracasei inoculated in a skim milk/yeast medium using an inlet temperature of 175  C and an outlet temperature of 68  C and reported that 15% of the probiotic was lost during drying; when cheddar cheese was fortified with the powder, the researchers found that the probiotic population grew during ripening. The survival of Saccharomyces boulardii, a probiotic yeast, was greatest at acid pH and high inlet temperature, deemed by the authors to be due to optimal protein crust formation around the yeast in these conditions. Encapsulation of B. animalis subsp. lactis BB-12 in sweet whey resulted in reduced viability at acid pH and no difference in salt tolerance (bile or NaCl); when the probiotic was added to a refrigerated dairy dessert, it showed a consistently higher viability during 6 weeks of storage. Freeze-drying of B. longum 1941 in different protein–sugar encapsulants revealed that milk proteins (skim milk, sodium caseinate, and whey protein concentrate) outperformed soy proteins and that sugar alcohols (glycerol and mannitol) outperformed maltodextrins, in terms of postdrying viability and acid and bile tolerance. Experiments where single droplets containing L. plantarum A17 in addition to skim milk, whey protein isolate, lactose, or trehalose were dried revealed that the dairy ingredients were capable of protecting the bacteria to a greater degree than the sugars during drying, due to the formation of a crust that limited the thermal load applied to the encapsulated bacteria. An interesting approach to the drying of probiotics is their codrying in the presence of prebiotics. Prebiotics are carbohydrates that cannot be digested by humans, but that can be used as a nutrient source by probiotics, thereby stimulating the

442

Drying: Effect on Nutrients, Composition and Health

growth of the latter. For this reason, they are sometimes collectively referred to as ‘synbiotics.’ Recent studies have shown that Bifidobacterium species can be effectively encapsulated using inulin, oligofructose, oligofructose-enriched inulin, or polydextrose, while Lactobacillus species have been encapsulated using fructooligosaccharide. Other combinatory approaches at improving the manufacture of probiotics include the use of spray–freeze-drying, which combines spray-drying and freeze-drying technologies and has primarily been used in the production of biopharmaceuticals. Spray–freeze-drying was successfully used to dry L. casei, with 98% of the probiotic surviving the drying process.

Nucleotides Nucleotides in the diet are thought to contribute positively to immune function, gastrointestinal health, and absorption of other micronutrients. Nutritional sources of nucleotides mainly include mammalian milks. It has long been known that human milk has innately high levels of nucleotides, and this has resulted in efforts to humanize infant formulas through fortification with nucleotides. Significantly lower levels of adenosine 50 -monophosphate and cytidine 50 monophosphate have been measured in reconstituted skim milk powder compared to raw milk, indicating that the spraydrying process may cause degradation of nucleotides. It was also reported by the same group that pasteurized milk had lower levels of these nucleotides, and as the skim milk powders were pasteurized and evaporated before drying, it is difficult to ascertain the influence of the drying step in isolation; it is also worth noting that in-container sterilized and powdered milks were the only samples to contain significantly lower levels of orotate. A recent study found that pasteurization of human milk actually increased the levels of free nucleotide monophosphates; on the other hand, high-pressure processing caused a decrease. Casein-dominant infant formulas have been shown to have naturally higher concentrations of nucleotides compared with whey protein-dominant formulas, whether the latter had been fortified or not. Additionally, the same author indicated that minimal degradation of nucleotides occurred in the powders studied during 1 year of storage. Recently, nucleotides in 11 infant formulas were analyzed and it was found that the majority had levels that were in general agreement with their label claim; however, one formula was reported to have nucleotide levels that were 90% less than what the label claims, indicating major losses during processing. More research is required to identify the potential role of spray-drying and other processing steps on the degradation of nucleotides in infant formulas.

Vitamins Foods such as fruits and vegetables are widely known to be important natural sources of vitamins, while formulated products such as infant formulas are increasingly fortified with vitamins. Vitamin D3, which is now routinely added to milk and infant milk products, was shown to undergo negligible degradation in the production of spray-dried milk. Vitamin C was shown to degrade during the air-drying of tomatoes, with higher drying temperatures resulting in greater losses; immersion in salt or sugar solutions prior to drying was found to inhibit this degradation. An inverse relationship between

air-drying temperature and vitamin C levels in both tomatoes and papaya has also been reported. Decreases in vitamins A, E, and C were measured during storage of infant formula powders for up to 18 months; the greatest losses were reported for vitamin C, and levels of all vitamins decreased more rapidly when powders were stored at elevated temperatures (40  C vs. 25  C). Tracking the degradation of vitamins A and E in infant formula powders after the packaging had been opened revealed that, while the levels of these vitamins did decrease over 70 d storage, the levels remained within regulatory limits. Enteral formula powders underwent reductions in vitamins A, B1, and E during storage for up to 6 months at 30  C that were sufficient to compromise the ability of those formulas to satisfy their label claims for RDAs, particularly in adverse conditions (high aw and lengthy storage). Efforts to increase the retention of vitamins in powders by advanced drying techniques (nanospraying or electrospraying with encapsulants) have recently been investigated.

Minerals The greatest potential for loss of minerals in products intended for drying is when the feed is in its liquid state. Mineral fallout in neutral-pH liquid products (e.g., cow milk, soy milk, and infant formula) can be caused by the use of insoluble calcium salts such as calcium carbonate, phosphate, and citrate, but this is less of a concern in acid products such as fruit juices. The stability of insoluble salts in liquids can be improved through the use of thickeners to increase the viscosity of the solvent or micronized forms of calcium that sediment less rapidly. Mineral loss from food during drying or during postdrying storage has not been demonstrated to be an issue affecting the nutritional quality of foods.

Other Considerations When Drying Nutritional Formulations Influence of Bulk Density on Nutrient Dosage in Nutritional Powders Powdered protein supplements are currently very popular among the physically active. In addition, powders are available that, once reconstituted, are intended to function as complete meal replacers, where they are the sole source of sustenance for the consumer. Critically, these products require the consumer to add a defined quantity of powder to a defined quantity of water to create a beverage with defined levels of key nutrients. In products such as these, infant formula powders being another example, the manufacturer must ensure that the powder has a consistent bulk density so that the consumer can scoop the correct quantity of powder into the liquid. In spray-drying, bulk density is influenced primarily by the viscosity of the feed, degree of aeration of the feed, type of atomizer, and agglomeration and/or instantization. If pneumatic conveying is used to transport powders within a manufacturing facility, additional care must be taken to prevent agglomerate breakage, which can result in deviation of bulk density from specified values. The manner in which bulk density can affect nutrient dosage in these products is illustrated in Figure 1.

Drying: Effect on Nutrients, Composition and Health

Multiple factors affect the bulk density of spray-dried powders

443

Scoops are precision engineered for powders with a specific bulk density

-Feed viscosity - Feed aeration - Atomizer type - Agglomeration

Equivalent volume of scooped powder

- Instantisation

=

- Pneumatic conveying

equivalent mass of nutrients per scoop if bulk density is not equivalent

Out-of-spec bulk density

Figure 1 Influence of bulk density on nutrient dosage in spray-dried nutritional formulations, such as infant formula and meal replacers.

Role of Accelerated Stability Tests and Fortification Overages In the development of new nutritional products, the final formulation and shelf life are determined by the integration of nutrient and physical stability testing data. For extended shelf life powders, the long shelf life necessitates the implementation of accelerated stability testing programs that complement traditional ambient stability tests. The principal environmental factors that are modulated during accelerated stability tests are temperature, light exposure/intensity, and relative humidity. Typical ambient and accelerated stability testing storage conditions for powders are 25  C/60% relative humidity and 40  C/75% relative humidity, respectively. For sensitive nutrients (e.g., B vitamins and PUFAs), the formulation overages required are normally determined based on measured losses during processing and stability testing. The principal factors affecting the overages required for individual nutrients vary but are generally determined by inherent stability of the nutrient, regulatory/quality limits, choice of drying technology and operating parameters, packaging material, transport and storage conditions, and the required shelf life. Some general considerations for nutrient stability testing are that ambient and accelerated programs should run in parallel and stability data should be generated using the actual commercial packaging material in all types and sizes that are manufactured. These concepts are illustrated in Figure 2.

Drying and Health: Overview and Future Perspectives Most foods, in their natural, high-moisture state, are prone to rapid deterioration due to microbial spoilage and/or chemical

degradation. Spoiled food is unpalatable and potentially toxic and generally has poor nutritional value. Dried foods and food powders have shelf lives that far exceed those of undehydrated varieties, and it may take many years before any spoilage eventually occurs. It would be a mistake to simply compare the nutritional value of a freshly harvested food with the same food recently dried, because the relative dynamics of degradation of each over time are also an important factor determining their potential health-giving properties. For certain nutrient-dense foods, it may not be possible to produce them in certain regions of the world, or where they can be produced, the supply may be contaminated or otherwise untrustworthy; in cases such as this, dried foods are a lowvolume product that can be transported from a reliable source at low cost and low environmental impact. Some of the research that has been highlighted in this article has demonstrated that the nutritional value of foods can indeed depreciate when being transformed into a dried product; however, many of these studies have also shown that this effect can be minimized or even eliminated if the drying process is carefully controlled. The consumption of bioactive substances (e.g., phytochemicals, probiotics, and PUFAs) in sufficient quantities to extract their purported benefits is difficult to achieve in many regions, particularly for the socioeconomically disadvantaged. Encapsulation by drying affords the possibility of a range of bioactive-fortified foods becoming commercially available. The availability of such products may alleviate the burden on individuals to consistently source and acquire the requisite foods, which may be prohibitively expensive or otherwise inaccessible. Moreover, it is thought that targeted food fortification strategies may be needed to alleviate region-specific

444

Drying: Effect on Nutrients, Composition and Health

Regulatory limits Target = innate + fortification Max. regulatory limit

Fortification = Target – innate Target Label claim Min. regulatory limit

Overage = Target – label claim

Innate Ingredient A

+

+

Fortification Overage

Ingredient B

Ambient stability tests

integrated with

General considerations: Ambient and accelerated tests should be run in parallel Stability data should be generated using actual commercial packaging material in all types/sizes to be manufactured.

Accelerated stability tests Variables: temperature, light exposure and intensity, relative humidity Sensitive nutrients (e.g., B vitamins, PUFAs), require an overage Overage is based on losses during processing and stability testing.

Figure 2 Principles of stability testing and fortification overages for sensitive nutrients during formulation of nutritional product.

deficiencies in certain vitamins, for example, vitamin D, the status of which is thought to be lower in northern latitudes due to a lack of sunshine; the effective incorporation of such fatsoluble vitamins into stable food systems is likely to require effective encapsulation technology. Dried food is a valuable alternative to fresh food, where the latter is unsafe or unavailable. The nature of a given food needs to be considered during the design of appropriate processes for its dehydration. In general, dried foods should be stored in cool ( 17 kg m 2), moderate (16–16.99 kg m 2), severe (15–15.99 kg m 2), and extreme (BMI < 15 kg m 2). Fear of fatness need not be verbally acknowledged if the patient is of significantly low weight and consistently engages in behavior that prevents weight gain. Patients may recognize that they are too thin but remain overly concerned with the size of specific body parts, such as the stomach, hips, or thighs. Frequent or obsessional body checking and weighing are often used to evaluate weight and shape. Individuals with AN are divided into two subtypes based on an assessment of disordered eating behaviors in the past 3 months. Patients who engage solely in restricting behaviors and excessive exercise, without regular (weekly) binge eating or purging behaviors, are diagnosed with AN, restricting subtype, whereas those whose restricting behaviors are accompanied by at least weekly binge eating and/or purging behaviors are diagnosed with the binge eating/purging subtype of AN. The 12-month prevalence rate among females is low (0.4%) and approximately 10% of cases are male. Of note,

http://dx.doi.org/10.1016/B978-0-12-384947-2.00243-9

463

464

Eating Disorders

AN has the highest standardized mortality rate of any psychiatric condition with fivefold risk of any cause of death and 18fold risk of death by suicide compared to an age-matched female population.

Bulimia Nervosa BN is characterized by recurrent episodes of binge eating and inappropriate compensatory behaviors. A binge is defined as eating a larger amount of food than typically normal in a discrete period of time, associated with a sense of loss of control over eating. Lack of control is defined as the inability to refrain from eating or to stop eating once an individual has begun. Compensatory behaviors are employed to prevent weight gain and include self-induced vomiting; misuse of laxatives, diet pills, or diuretics, restricting, and excessive exercise. Both the binge eating episodes and compensatory behaviors must occur on average at least once a week for 3 months. As in AN, excessive emphasis on body shape and weight as a measure of self-evaluation and self-esteem is a core criterion of BN. Individuals with BN are typically of normal weight or overweight; if significantly underweight, the diagnosis of AN-purging subtype trumps the diagnosis of BN. Within BN, the level of severity depends on the frequency with which individuals engage in compensatory behaviors and ranges between mild (1–3 episodes per week), moderate (4–7 episodes per week), severe (8–13 episodes per week), and extreme (14 or more episodes per week). BN is slightly more common than AN, with 12-month prevalence rates of 1–5% in females. Similar to AN, male patients represent approximately 10% of cases.

Binge Eating Disorder Binge Eating Disorder (BED), defined as recurrent consumption of large amounts of food in a fairly short period of time accompanied by feelings of loss of control (without the compensatory behaviors seen in BN), has recently been changed from a provisional diagnosis in need of further research in the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV), to a unique psychiatric condition in the DSM-V. Binge episodes are associated with eating until uncomfortably full, eating rapidly, eating alone, eating in the absence of hunger, and/or feeling disgusted with oneself after binge eating. The DSM-V requires that at least three of these five features accompany binge episodes on average once a week or more for 3 months to make a diagnosis. Binge episodes differ from overeating in that they are often secretive, associated with shame, and occur in the context of negative affect and cognitions. BED is associated with decreased quality of life and increased lifetime co-occurrence of medical (e.g., irritable bowel syndrome, fibromyalgia, type 2 diabetes, and obesity) and psychiatric disorders, including major depressive disorder, bipolar disorder, anxiety disorders, substance use disorders, and body dysmorphic disorder. Although not all individuals with BED are overweight or obese, approximately 30% of those seeking weight loss treatment meet criteria for BED in comparison to 2–5% prevalence rates in community samples. BED is particularly prevalent among those pursuing bariatric surgery, ranging from 10% to 50%, and preoperative BED may

indirectly predict postsurgical weight loss mediated by postoperative eating behavior.

Other Eating Disorders Avoidant/Restrictive Food Intake Disorder (ARFID) is an eating or feeding disturbance in which individuals avoid or restrict their oral intake of food in the absence of an underlying medical condition. This disturbance results in one or more of the following: significant weight loss, or in children, a failure to meet expected weight gain, nutritional deficiency, dependence on oral nutritional supplements or enteral feeding, or a significant impairment in psychosocial functioning. The avoidance or restriction of food in ARFID is not in response to scarcity of food or cultural and developmental norms such as picky eating in preschool-aged children. Typically, the avoidance/restriction results from a lack of interest in eating or food, or an avoidance of eating due to sensory characteristics of food, such as texture or smell. Unlike other eating disorders, individuals with ARFID do not experience a disturbance in the perception of body shape or weight. In addition to these disorders, DSM-V distinguishes a fifth broadly defined category of Other Specified Feeding or Eating Disorders. The clinical presentations of these syndromes may be behaviorally similar to AN, BN, or BED and cause significant impairment and distress in the individual’s life. However, the symptoms do not meet full diagnostic criteria for one of the earlier mentioned conditions, or there is insufficient information to make a diagnosis. Examples of these disorders include atypical AN, subclinical BN or BED, purging disorder, and night eating syndrome. In atypical AN, all the criteria for AN are met; however, despite significant weight loss, the individual’s weight is within or above the normal range. Subclinical BN or BED may be diagnosed when the individual engages in binge eating or inappropriate compensatory behaviors less than once a week and/or for less than 3 months. Purging disorder is characterized by recurrent purging behaviors with the goal of preventing weight gain in the absence of binge eating. Night eating syndrome is defined as recurrent episodes of nocturnal eating in which an individual consumes an excessive amount of food after an evening meal or will eat after awakening from sleep. These episodes of eating are not in response to external influences or related to a medical or psychiatric disorder. Finally, unspecified Feeding or Eating Disorder may be diagnosed if a patient presents with symptoms of disordered eating that cause significant distress or impairment in functioning and is most commonly used when there is insufficient information to determine a diagnosis.

Disordered Eating Behaviors Certain disordered eating behaviors are observed cross diagnostically. Further, these behaviors may occur in nonclinical samples and potentially signify the need for early intervention. Though restricting behaviors are thought to be primarily associated with AN, they are seen in other eating disorders as well, especially BN. Individuals who engage in restricting behaviors attempt to limit their dietary intake with the goal of weight loss or to prevent weight gain. Common restricting behaviors

Eating Disorders include reducing the amount and frequency of meals and food portions, avoiding calorie-dense foods, and narrowing one’s food repertoire such that consumption is limited to specific low-calorie ‘safe’ food items. In addition, individuals will often engage in prolonged periods of fasting. For example, after a binge episode, individuals may forgo food for an extended period of time to prevent weight gain. Binge eating and purging behaviors also occur trans-diagnostically. Binge eating episodes are defined as consuming an unusually large amount of food (about two meals worth) in a discrete period of time coupled with a sense of loss of control over eating. When a sense of loss of control is associated with eating small amounts of food, this is termed a subjective binge episode. Subjective binge episodes are more frequently reported by patients with binge eating/purging type of AN than by those with BN. Common purging behaviors include self-induced vomiting and laxative, enema, or diuretic abuse. Uncommon but important additional purging behaviors employed in the service of weight loss by individuals with eating disorders can include insulin omission in type I diabetics and wasting expressed breast milk in postpartum mothers. The frequency of purging behavior use varies, with some individuals purging exclusively after binge eating episodes, whereas others may purge after consuming small amounts of food or after every meal. Additional behaviors are observed across all eating disorder diagnoses. These include chewing and spitting and grazing. Chewing and spitting out food before swallowing is common, often associated with a sense of lack of control over eating, and can involve large amounts of food and feelings of shame/guilt afterward. This behavior may be utilized as a way to taste ‘forbidden foods’ without ingesting calories. In other cases, chewing and spitting is best conceptualized as an alternative to bingeing and purging and is perceived as less medically risky, with lower likelihood of electrolyte abnormalities or other medical complications than vomiting. Grazing is the unplanned, repetitive ingestion of small amounts of food that may occur in a compulsive manner, such as repeatedly eating from a container of food when walking past it. Although grazing is frequently observed in nonclinical and eating disorder samples, when accompanied by a sense of loss of control over eating, it is associated with increased psychopathology. Grazing behavior has received significant attention in the bariatric surgery literature, particularly since grazing postsurgically is associated with suboptimal weight loss and/or weight regain, and efforts are currently under way to determine the best way to define grazing both clinically and for research purposes. Excessive exercise is a frequently observed behavior employed in the service of weight loss among individuals diagnosed with AN or BN and is commonly defined by dysfunction in the quality and quantity of exercise. Unlike normal exercise, excessive exercise manifests as a strict and compulsive adherence to exercise regimes, preoccupation with exercise, feelings of guilt if exercise is not completed, and exercising despite illness, injury, or physical complications. These behaviors are utilized to prevent weight gain, influence one’s shape and/or weight, or reduce feelings of distress related to food consumption. In addition, individuals may engage in increased frequency, duration, and intensity of exercise. Excessive exercise is associated with increased psychological distress, psychopathology, and risk of relapse.

465

Laboratory Studies of Food Consumption/ Macronutrient Selection Laboratory studies of eating behavior have consistently shown that individuals with eating disorders will engage in disordered eating under controlled laboratory conditions. Compared to healthy controls, patients with AN select fewer food items, eat fewer calorie-dense foods, and have a lower total caloric intake in laboratory test meals in which they are presented with a large variety of foods of different calorie densities. Macronutrient selection primarily consists of carbohydrates ( 55%) and protein ( 30%), with limited caloric intake from fat (1000 ng, whereas the LD50 of Stx2a is 6.5 ng. Stx1 and Stx2 are AB5-type toxins; the B-pentamer of the holotoxin binds to globotriaosylceramide (Gb3) present on host microvascular endothelial cell surfaces (the kidney, intestine, and brain) followed by endocytosis of the holotoxin. A subunit of the A portion of the toxin is an N-glycosidase that acts on the 28S RNA of the 60S ribosomal subunit leading to the inhibition of protein synthesis and apoptosis of endothelial cells, particularly those of the kidneys with resultant damage of the renal endothelial cells. There are swelling of the renal cells and detachment of the cells from the basement membrane and formation of fibrin thrombi in the kidney capillaries. Narrowing of the capillary lumens leads to a reduced blood supply to glomeruli with a loss of kidney function resulting in HUS. HUS is characterized by hemolytic anemia (destruction of red blood cells), acute kidney failure (uremia), and low platelet count (thrombocytopenia). Early symptoms of STEC infection include watery diarrhea and abdominal cramps. Approximately 90% of confirmed STEC infections result in HC, and 5–15% of the HC cases progress to HUS. The mortality rate for HUS ranges from 5% to 10%. A few patients suffer long-term consequences affecting the gastrointestinal tract, liver, cardiovascular system, or central nervous system (CNS); however, most patients recover without major consequences. Very young and elderly patients show the most susceptibility to complications and death due to HUS. Antibiotics are not recommended in the treatment of STEC-infected patients and can increase the risk for development of HUS due to toxin release from lysed STEC cells. Fluid hydration may be necessary in diarrheic patients. Patients with severe bloody diarrhea may need blood transfusions, and in cases of HUS, dialysis may be necessary. The main reservoir for STEC is the intestinal tract of cattle. However, other ruminants, including sheep, goat, buffalo, guanaco, deer, and elk, and nonruminants, including cats, dogs, pigs, horses, rabbits, and chickens, also can carry STEC. Contact with cattle, the cattle environment, and food products originating from cattle provides major risk factors for STEC

infection to humans. Recreational, drinking, and irrigation waters contaminated with cattle feces can be sources of STEC. Infected individuals can transmit STEC person-to-person, and food may be contaminated by cross contamination or by an infected food preparer. In the United States, STEC O157:H7 was declared an adulterant in beef in 1994 after a large outbreak linked to contaminated ground beef occurred, and recently, STEC serogroups O26, O45, O103, O111, O121, and O145 were declared adulterants, since these serogroups cause the majority of the non-O157 STEC infections and illness similar to that caused by STEC O157:H7. For the period 2009– 13 in the United States, O157:H7 STEC outbreaks were associated with foods, including spinach (1 outbreak), Romaine lettuce (1), Lebanon bologna (1), beef (3), hazel nuts (1), cheese (1), and prepackaged cookie dough (1). STEC serogroup O26 was associated with outbreaks linked to ground beef and raw clover sprouts. An outbreak linked to STEC O121 involved frozen food products, and O145 was associated with two outbreaks, one linked to Romaine lettuce and the other due to an unknown source. Outbreaks are often associated with STEC O157:H7, whereas non-O157 STEC are more often associated with sporadic cases and less often with outbreaks. However, it is now estimated that non-O157 strains cause a greater number of illnesses than E. coli O157:H7. Based on the Centers for Disease Control and Prevention’s FoodNet program in the United States for the years 2000–10, STEC O157:H7 was the cause of 5688 infections (74.0%), whereas non-O157 STEC caused 2006 infections (26.1%). The top six non-O157 STEC caused 1416 cases (18.4%). Serogroup O26 was responsible for 5.8% of STEC cases, followed by serogroups O103 (5.0%), O111 (4.2%), O121 (1.4%), O45 (1.2%), and O145 (0.8%). However, estimates of the total STEC foodborne infections indicate that the non-O157 serogroups cause 64% of the STEC disease cases in the United States. In the European Union, in 2011, nine STEC serogroups were responsible for 4022 cases of disease. STEC O157:H7/H was responsible for 54% of the STEC cases, and STEAEC (Shiga toxin-producing enteroaggregative E. coli) O104 caused 26% of the disease cases. Serogroup O26 accounted for 7% of the cases. Serogroups O103, O91, O145, O128, O111, and O146 were responsible for fewer cases in the descending order. In 2010, nine serogroups (in the descending order, O157, O26, O103, O145, O91, O63, O111, O128, and O146) accounted for 2110 cases; STEC O157:H7 caused 71% of the STEC illnesses followed by 12% for serogroup O26. There was no report of STEAEC cases in the European Union in 2010. Of 504 isolates of STEC isolated for the period 2001–09 in Australia, the following serogroups were the most common: O157 (58%), O111 (13.7%), O26 (11.1%), O113 (3.6%), O55 (1.3%), and O86 (1.0%). The information on STEC isolates from patients in the United States, the European Union, and Australia indicate that serogroup O157 was the most common cause of STEC illness. Strict hygiene during animal slaughtering, vegetable and fruit harvesting, product handling, shipping, and food preparation is critical in controlling and preventing STEC-induced disease.

Enteropathogenic E. coli Enteropathogenic E. coli (EPEC) are an important cause of persistent and potentially fatal diarrhea in infants and young

Escherichia coli and Other Enterobacteriaceae: Food Poisoning and Health Effects children in developing countries. EPEC possess the LEE pathogenicity island similar to the Shiga toxin-producing E. coli but do not produce Stx. The formation of A/E lesions is characteristic of EPEC and is dependent on LEE island genes encoding T3SS, which acts in a similar way to the secretory system in STEC and allows intimate attachment of the bacteria to the host cell. LEE genes encode for effector proteins that subvert host cellular processes and induce gastrointestinal symptoms. Non-LEE-encoded genes produce effector proteins that inhibit phagocytosis, activate innate immune responses, and have a role in colonization and pathogenic responses. The E. coli adherence factor plasmid (pEAF) has the bfp operon, which encodes the bundle-forming pilus. The EPEC strains that carry the eae and bfpA genes are termed typical EPEC, whereas the atypical EPEC strains have the eae gene but lack bfpA. The typical EPEC strains are a major cause of infantile diarrhea in developing countries but are rare in the developed world. Diarrhea manifests as a severe watery stool with a large amount of mucus but is rarely bloody. The atypical EPEC induce a less severe illness. Treatment consists of prevention of dehydration induced by diarrhea. Humans appear to be the only reservoir for typical EPEC, whereas both animals and humans are reservoirs for atypical EPEC. Contaminated food and water have been associated with EPEC diarrhea. EPEC infections are generally sporadic, and food vehicles are often unknown. Raw chicken and beef, cold pork, meat pie, and other foods, as well as water, have been implicated in a few outbreaks. However, any foods exposed to the feces of an infected human may induce diarrhea.

Enteroinvasive E. coli Enteroinvasive E. coli (EIEC) induce an invasive inflammatory colitis with copious watery diarrhea. EIEC are closely related to Shigella species but do not produce Shiga toxin. A number of EIEC strains contain the sen gene, which encodes for Shigella enterotoxin-2 (ShET-2). ShET-2 is translocated to the host cell via T3SS and may participate in the inflammation of the intestinal epithelial cells. The virulence effector genes, the sen gene, and T3SS genes are located on a large plasmid, pInv. EIEC does not carry the LEE pathogenicity island. When ingested, EIEC reach the colon, gain access to the colonic submucosa through the M cells, and are then taken up by macrophages. The bacterial cells escape from the macrophages and invade and replicate in the colonocytes. Bacterial virulence effectors are secreted via T3SS into the host cells, which enable the bacteria to evade the immune system and to promote cell-to-cell dissemination. The role of T3SS in EIEC is internalization of the bacteria into the host cell; A/E lesions are not seen in EIEC infection. EIEC-contaminated food and water are associated with the illness. EIEC infections caused by EIEC are rarely detected in the United States; however, a large outbreak occurred in Texas in 1992 in which 370 people became ill after consuming guacamole. Foods are most likely contaminated by infected food handlers, and EIEC infections may also occur via person-to-person transmission. Outbreaks associated with EIEC in several countries have been reported with water, imported cheese, vegetables, and potato salad as the vehicles of infection. Rehydration therapy and antibiotics are recommended as treatments for EIEC infection. Since humans are

541

the only known reservoir of EIEC, it is probable that most EIEC infections are due to food prepared by an ill individual.

Enterotoxigenic E. coli Enterotoxigenic E. coli (ETEC) infection induces a watery diarrhea, which may be a mild, self-limiting disease or a severe purging cholera-like disease. The infection may attack children in developing countries or other areas with poor hygiene and unclean water. In addition, ETEC is a major cause of traveler’s diarrhea. The organism colonizes the surface of the small bowel and produces enterotoxins, which induce the diarrhea. ETEC elaborate two types of plasmid-encoded enterotoxins: heat-labile enterotoxins (LT-1 and LT-2) and heat-stable enterotoxins (STa and STb). LT-1 is an AB5 toxin closely related to cholera toxin. LT intoxication of host intestinal cells leads to permanent activation of adenylate cyclase with increased levels of cyclic AMP (adenosine monophosphate). Cyclic AMP activates protein kinase A, leading to phosphorylation of ion channels and secretion of electrolytes into the intestinal lumen. The increased gut osmolarity leads to a watery diarrhea. STa is a small, single-peptide toxin, which binds to the host cell guanylate cyclase, leading to increased cyclic GMP (guanosine monophosphate) levels. Chloride secretion is increased, and NaCl absorption is inhibited. The increased osmolarity in the intestinal lumen results in diarrhea. ETEC strains may produce LT, ST, or both. Rehydration therapy and antibiotics are recommended as treatments for ETEC infection. Humans are the reservoir of ETEC, and ingestion of food or water contaminated by an infected person is often the likely cause of illness. Foods implicated in infections include Brie cheese, curried turkey, mayonnaise, crab meat, deli foods, and salads.

Enteroaggregative E. coli Enteroaggregative E. coli (EAEC) include a heterogeneous population of E. coli strains that do not produce LT or ST and that adhere in a stacked brick pattern to HEp-2 cells and to each other. These strains cause a watery diarrhea in both industrialized and third-world countries. Diarrhea can be of relatively long duration ( 14 days) in some cases. EAEC infects children and immunocompromised individuals, and these pathogens are an important cause of traveler’s diarrhea. The major features of EAEC pathogenesis are colonization of the intestinal mucosa with mucoid biofilm formation, elaboration of enterotoxins, and mucosal inflammation. The plasmid (pAA)-encoded AAF (aggregative adherence fimbriae) and AggR (regulator) are required for colonization; however, AAF is not present in all EAEC strains, and therefore, other adhesins are likely involved in EAEC colonization. Adhesion is followed by secretion of Pet (plasmid-encoded toxin) and EAST1 toxins (plasmid-encoded EAEC heat-stable toxin 1). EAST1 is a small polypeptide enterotoxin that induces diarrhea by secreting cyclic GMP. The pet gene is located on pAA and encodes Pet, a heat-labile protein toxin belonging to the serine protease autotransporter family. Pet is an enterotoxin inducing cytoskeletal changes and epithelial cell rounding. In addition, the enterotoxin ShET-1 may be found in some EAEC. However, the role of enterotoxins in EAEC-induced diarrhea is not clear. EAEC strains that show the stacked brick phenotype and that

542

Escherichia coli and Other Enterobacteriaceae: Food Poisoning and Health Effects

contain pAA are termed typical EAEC, whereas those strains that have the stacked brick phenotype but lack pAA are termed atypical EAEC. Both atypical and typical strains induce diarrhea. Rehydration therapy and antibiotics are recommended as treatments for EAEC infection. Ingestion of food has been associated with EAEC-induced diarrhea; however, the reservoir of the organism is unknown. Salsa in Mexico, unpasteurized cheeses in Italy, and school lunches in Japan have caused outbreaks of EAEC-induced illness. It is probable that infected food handlers are the source of the organism.

Diffusely Adherent E. coli Diffuse adherence is characterized by uniform attachment of bacteria to HeLa or HEp-2 cells, whereas localized adherence is when bacteria attach in groups to only a few sites on the cultured cell surface. Diffusely adherent E. coli (DAEC) are a poorly characterized heterogeneous population of E. coli showing diffuse attachment to cell surfaces and are believed to be a cause of diarrhea. The DAEC pathotype colonizes the small bowel and may be responsible for watery diarrhea in children. Diffuse adherence is mediated by both fimbrial (Dr and F1845) and afimbrial (Afa) adhesions. A secreted autotransporter toxin (SAT) has been demonstrated in DAEC strains associated with diarrhea. SAT induces lesions on tight junctions of epithelial cells causing increased permeability and may be a virulence mechanism in DAEC. However, the mechanism of diarrhea induction in DAEC strains is not certain.

Shiga Toxin-Producing Enteroaggregative E. coli The Shiga toxin-producing enteroaggregative E. coli (STEAEC) may be a newly emerging pathotype of E. coli. The serotype O104:H4 caused a large outbreak associated with close to 4000 cases of illness, over 800 cases of HUS, and over 50 deaths in Europe in 2011. This pathotype is unusual in that the organism carried the pAA-virulence plasmid (and its virulence genes) of EAEC strains and a stx2-harboring prophage. Thus, this STEAEC strain presents the stacked brick phenotype of EAEC and production of Shiga toxin. The STEAEC strain is not a pathogenic E. coli with new virulence factors but is rather a pathogen that has combined the virulence properties of the EAEC with the production of Shiga toxin of the STEC pathotype. The strong morbidity and mortality induced by this strain may be due to its strong adherence to the host cell leading to higher levels of Stx transfer. The serotype O111:H2 with both Stx2 and EAgg (AggR) was reported in 1997 as a cause of a case of HUS, and in 2012, an outbreak induced by serotype O111: H21 positive for Stx2c and aggR occurred in Ireland. Therefore, it is probable that the transfer of the Shiga toxin gene into other EAEC serotypes will produce STEAEC serotypes that may cause serious outbreaks of Shiga toxin-induced HC and HUS in the future.

Recent Outbreaks in the United States due to E. coli Pathotypes For the years 2009–10, the Centers for Disease Control and Prevention reported that there were 58 outbreaks due to STEC serotypes (53 due to STEC O157:H7 and 5 due to non-O157

STEC) with 651 total cases. In 2009, there were three outbreaks due to ETEC with 85 cases, and in 2010, EPEC caused one outbreak with 7 cases. The vehicles of infection were not provided. There were no reported outbreaks caused by EIEC, EAEC, or DAEC for 2009–10.

Nonintestinal Pathogenic E. coli The nonintestinal pathogenic E. coli, which are a cause of illness outside of the gut, are termed extraintestinal pathogenic E. coli (ExPEC). The intestinal pathogenic E. coli (IPEC) are obligate pathogens, whereas the ExPEC are facultative pathogens that reside in the gut of a fraction of individuals and behave as harmless commensal organisms. In general, the IPEC have well-defined virulence traits, whereas ExPEC strains are very diverse and have few common virulence factors among them. Thus, it is difficult to define ExPEC by virulence properties. At the present time, ExPEC and nonpathogenic E. coli cannot be differentiated by molecular epidemiological techniques, and therefore, there is a limited understanding of the biology of ExPEC. Urinary tract infections (UTIs) are common bacterial infections, particularly in women, and extract extremely high health costs in terms of morbidity, treatment, and loss of productivity. Uropathogenic E. coli (UPEC) are ExPEC that cause  80% of community-acquired UTIs and  30% of nosocomial UTIs. The UPEC reside in the gastrointestinal tract and can be transferred to the urinary tract. Infection of the urethra is termed urethritis, that of the bladder is termed cystitis, and that of the kidneys is termed pyelonephritis. Individuals at risk for acquiring UTIs include the very young, elderly, females, pregnant women, and individuals who are immunocompromised or who are undergoing urinary catheterization. The UPEC possess a number of virulence traits that make them efficient pathogens of the urinary tract such as specific adhesins; several toxins including hemolysin, cytotoxic necrotizing factor type 1 (CNF1), and a protease toxin (SAT); and iron acquisition systems. The primary reservoir for UPEC is the human intestinal tract; however, the original source (e.g., possibly avian, food, or environmental source) of UPEC is not clear. The pathology leading to UTIs by UPEC includes adherence and colonization, evasion of host defenses, and damage to host cells and tissues. The UPEC are transferred from the rectum to the periurethral area and ascend the urethra into the bladder. Type 1 fimbriated UPEC attach to the bladder epithelial cells and trigger apoptosis and exfoliation, as well as invasion and multiplication of UPEC in the bladder cells. If the UTI is not treated, the organisms may ascend the ureters and induce pyelonephritis. Pyelonephritis can cause irreversible kidney damage, kidney failure, and death. Some individuals may have high bacterial counts in the urine but are asymptomatic. Treatment of UTIs is becoming more complicated due to increased resistance of UPEC to antimicrobial compounds. Neonatal meningitis E. coli (NMEC) strains are ExPEC that represent the most frequent cause of gram-negative meningitis in the newborn. Bacterial meningitis is an inflammation of the meninges with high morbidity and mortality. Severe neurological sequelae occur in 30–50% of infants who survive

Escherichia coli and Other Enterobacteriaceae: Food Poisoning and Health Effects meningitis. The neonate (neonatal period is the first 28 days after birth) acquires NMEC during birth or from the environment, and the organisms invade and multiply in the bloodstream. The NMEC pathogen then crosses the blood–brain barrier (BBB). The BBB is a structural and functional barrier made up of the brain microvascular endothelial cells (BMECs), which separates the CNS from the vascular component of the body. The BBB regulates the transition of macromolecules into the CNS and protects the CNS against entry of microbes and toxins that may be present in the blood. There are several stages of NMEC pathogenesis: bacteremia, binding of bacteria to the surface of the BMEC, and invasion of the BMEC followed by invasion of the meninges and the CNS. Meningitis is associated with bacteremia counts of >103 CFU ml 1 blood. Outer membrane protein A (OmpA), K1 capsule (carried by the predominant strains of E. coli that cause neonatal meningitis), and O-lipopolysaccharide are necessary for survival and multiplication of NMEC in the circulatory system. OmpA and FimB (fimbriae) of NMEC bind to BMEC, and the organism invades the BMEC via a transcellular traversal mechanism in which the bacteria penetrate the cells without disrupting them. OmpA, Ibe proteins, and CNF1 induce actin cytoskeletal rearrangement and formation of microvillus-like protrusions, which facilitate entry of the organisms into the BMEC. NMEC leave the BMEC and invade the meninges and CNS, and then, they multiply and induce the release of proinflammatory compounds. In addition, there are brain edema and increased intracranial pressure with meningitis and neuronal injury. Neonatal meningitis is treated with appropriate antibiotics. Avian pathogenic E. coli (APEC) are responsible for a number of economically important poultry diseases. Interestingly, some human ExPEC and some APEC strains have similar phylogenetic backgrounds and share some of the same virulence genes. This relationship between ExPEC and APEC suggests that APEC may be zoonotic pathogens that can lead to human diseases such as UTIs or neonatal meningitis through the ingestion of poultry meat. Recent studies support the role of chicken meat as a potential source of organisms inducing UTIs to humans; however, food has not been shown to be related to neonatal meningitis.

Other Enterobacteriaceae Cronobacter sakazakii The genus Cronobacter consists of seven species, namely, C. sakazakii, C. malonaticus, C. turicensis, C. dublinensis, C. muytjensii, C. condimenti, and C. universalis. Cronobacter species are gram-negative, motile by peritrichous flagella, non-sporeforming, facultatively anaerobic, rod-shaped bacteria. Cronobacter species are more thermotolerant when compared with other genera in the Enterobacteriaceae family, and they can grow in temperatures over a range of 6–47  C. The reservoir of Cronobacter is unknown; however, plants are a possible source, and the organisms have been isolated from a wide range of food products. Most Cronobacter species have been associated with human illness; however, the majority of the infections in infants and neonates have been associated with C. sakazakii, and this species has been the most studied. Infants that ingest

543

rehydrated powdered milk formula contaminated with C. sakazakii suffer bacteremia, meningitis, and necrotizing enterocolitis; mortality is greater than 50%. Infections among immunocompromised adults, particularly the elderly, have been reported. The outer membrane protein, OmpA, is an important virulence factor in the development of meningitis in newborn rats. OmpA-positive Cronobacter cross the intestinal barrier, multiply in the blood, and invade the brain endothelial cells. The pathogen crosses the BBB and eventually causes meningitis. Cronobacter lacking the ompA gene do not bind to intestinal epithelial cells, do not cross the intestinal barrier, and do not lead to pathology. Other potential virulence factors include biofilm formation and membraneassociated efflux pumps that expel a broad range of inhibitory compounds. However, it is clear that knowledge of the virulence mechanisms of the Cronobacter species is limited.

Shigella The Shigella species are small, gram-negative, nonmotile, facultatively anaerobic rods, which do not produce spores or capsules. The optimum growth temperature is 37  C. The organisms are the causative agents of bacillary dysentery (shigellosis). All of the species except Shigella sonnei have several serotypes. Humans are the natural host of the Shigella species. The organisms are transmitted via the fecal–oral route by infected individuals and contaminated food and water. Shigellosis is an invasive infection of the colon that ranges from short-term watery diarrhea to inflammatory bowel disease. In inflammatory bowel disease, watery diarrhea proceeds to bloody mucoid stools, abdominal cramps, and tenesmus. In healthy individuals, shigellosis is self-limited and resolves without sequelae. In malnourished infants and young children in developing countries, shigellosis can result in acute, life-threatening complications. HIV-infected and immunocompromised individuals may suffer persistent diarrhea and malnutrition if infected by Shigella. Cellular invasion and spread of infection involve invasion of the enterocytes, intracellular multiplication, intra- and intercellular spread, killing of the host cells, and tissue damage. Shigella dysenteriae, but not other Shigella species, produce Shiga toxin. Shiga toxin is encoded chromosomally in Shigella dysenteriae, whereas it is phage-encoded in STEC. Similar to STEC, infection by Shigella dysenteriae can induce HUS. The virulence of Shigella is dependent on a virulence plasmid encoding T3SS components. T3SS is necessary for invasion and secretion of virulence factors into intestinal epithelial cells leading to inflammatory destruction of the epithelial lining of the intestine. Rehydration therapy and antibiotics are used to treat shigellosis.

Salmonella enterica Salmonella spp. are rod-shaped, gram-negative, non-sporeforming, predominantly motile with peritrichous flagella, chemoorganotrophic, and facultatively anaerobic. The range of temperature for growth is 2–54  C, and the optimum is 35– 37  C. There are more than 2000 different serovars based on flagellar, carbohydrate, and lipopolysaccharide characteristics. Salmonella causes disease in both animals and humans. Serovars Typhi, Paratyphi, and Sendai cause enteric fever, whereas most

544

Escherichia coli and Other Enterobacteriaceae: Food Poisoning and Health Effects

serovars cause enterocolitis and diarrhea. The elderly, infants, and immunocompromised individuals are at the highest risk for Salmonella infections. Ingestion of food or water contaminated with nontyphoidal Salmonella is followed by crampy, abdominal pain, diarrhea, nausea, and vomiting. Enterocolitis in children is marked by increased severity of inflammation, bloody diarrhea, long duration, and increased risk of complications. Salmonellosis is self-limited in immunocompetent hosts. A small number of infected individuals develop a sequela known as reactive arthritis. Important virulence genes are carried on the Salmonella pathogenicity islands (SPI-1 through SPI-5). Other virulence traits include the virulence plasmid (pSLT), adhesins, flagella, and proteins involved in biofilm formation. Antibiotics are recommended in severely ill patients; however, it is important to realize that antibiotic resistance is common in the salmonellae. Hydration therapy is necessary in cases of severe diarrhea.

Yersinia Yersiniosis is a self-limiting gastrointestinal disease in immunocompetent individuals; however, in young children and immunocompromised individuals, the disease is more severe and may result in death. Sequelae such as reactive arthritis may occur after infection. The organism is rod-shaped, gramnegative, psychrotrophic (can grow at 4  C), and facultatively anaerobic. Yersiniosis is a foodborne disease caused mainly by Y. enterocolitica and Y. pseudotuberculosis, and the fleatransmitted Y. pestis rarely causes foodborne illness. Pigs are the major reservoir of Y. enterocolitica; pig tonsils are a common site of Yersinia colonization. The absence of the virulence plasmid (pYV) is associated with lack of pathogenicity in Y. enterocolitica. pYV encodes the Yersinia adhesin YadA, which mediates mucus and epithelial cell attachment and host cell invasion. A T3SS is also encoded by pYV and is involved in the injection of virulence-related effector proteins, the Yersinia outer membrane proteins (Yops), into the host cell. The Yop proteins interfere with host signal induction (evasion of immune responses), disrupt host actin cytoskeleton, and induce apoptosis in host cells. Chromosomal genes encoding the invasion protein (Inv) and the attachment-invasion locus (Ail) have a role in cell binding and invasion by Yersinia. The most common vehicle for yersiniosis is contaminated food, including meat, oysters, fish, raw milk, and cheese. Antibiotics are used to treat yersiniosis.

See also: Diarrheal Diseases; Emerging Foodborne Enteric Bacterial Pathogens; Escherichia coli and Other Enterobacteriaceae: Occurrence and Detection; Food Poisoning: Classification; Food Poisoning: Epidemiology; Food Poisoning: Tracing Origins and Testing;

Salmonella: Detection; Salmonella: Properties and Occurrence; Salmonella: Salmonellosis; Shigella; Yersinia enterocolitica: Properties and Occurrence; Yersinia enterocolitica: Detection and Treatment.

Further Reading CDC (2013) Surveillance for foodborne disease outbreaks – United States, 2009–2010. Morbidity and Mortality Weekly Report (MMWR) 62: 41–47. Clements A, Young JC, Constantinou N, and Frankel G (2012) Infection strategies of enteric pathogenic Escherichia coli. Gut Microbes 3: 71–87. Coburn B, Grassl GA, and Finley BB (2007) Salmonella, the host and disease: a brief review. Immunology and Cell Biology 85: 112–118. Croxen MA and Finlay BB (2010) Molecular mechanisms of Escherichia coli pathogenicity. Nature Reviews Microbiology 8: 26–38. Drummond N, Murphy BP, Ringwood T, Prentice MB, Buckley JF, and Fanning S (2012) Yersinia enterocolitica: a brief review of the issues relating to the zoonotic pathogen, public health challenges, and the pork production chain. Foodborne Pathogens and Disease 9: 179–189. European Food Safety Authority (2013) EU summary report on trends and sources of zoonoses, zoonotic agents, and food-borne outbreaks in 2011. EFSA Journal 11(4): 3129. Fa`brega A and Vila J (2013) Salmonella enterica serovar Typhimurium skills to succeed in the host: virulence and regulation. Clinical Microbiology Reviews 26: 308–341. Farfan MJ and Torres AG (2012) Molecular mechanisms that mediate colonization of Shiga toxin-producing Escherichia coli strains. Infection and Immunity 80: 902–913. Healy B, Cooney S, O’Brien S, et al. (2010) Cronobacter (Enterobacter sakazakii): an opportunistic foodborne pathogen. Foodborne Pathogens and Disease 7: 339–350. Kaper JB, Nataro JP, and Mobley HL (2004) Pathogenic Escherichia coli. Nature Reviews Microbiology 2: 123–140. Mellata M (2013) Human and avian extraintestinal pathogenic Escherichia coli: infections, zoonotic risks, and antibiotic resistance trends. Foodborne Pathogens and Disease 10: 916–932. Melton-Celsa A, Mohawk K, Teel L, and O’Brien A (2012) Pathogenesis of Shiga toxinproducing Escherichia coli. Current Topics in Microbiology and Immunology 357: 67–103. Scallan E, Hoekstra RM, Angulo FJ, et al. (2011) Foodborne illness acquired in the United Statesdmajor pathogens. Emerging Infectious Diseases 17: 7–15. Smith JL, Fratamico PM, and Gunther NW (2007) Extraintestinal pathogenic Escherichia coli. Foodborne Pathogens and Disease 4: 134–163. Wang F, Yang Q, Kase JA, Meng J, Clotilde LM, Lin A, and Ge B (2013) Current trends in detecting non-O157 Shiga toxin-producing Escherichia coli in food. Foodborne Pathogens and Disease 10: 665–677.

Relevant Websites http://www.cdc.gov/cronobacter/technical.html – Centers for Disease Control and Prevention. http://www.cdc.gov/nczved/divisions/dfbmd/diseases/shigellosis/ – Centers for Disease Control and Prevention. http://www.cdc.gov/ncidod/dbmd/diseaseinfo/yersinia_g.htm – Centers for Disease Control and Prevention. http://www.cdc.gov/ecoli/ – Centers for Disease Control and Prevention. http://www.cdc.gov/salmonella/general/ – Centers for Disease Control and Prevention. http://www.fda.gov/Food/FoodborneIllnessContaminants/ CausesOfIllnessBadBugBook/ – U.S. Food and Drug Administration.

Escherichia coli and Other Enterobacteriaceae: Occurrence and Detection S Fanning, L Rogers, K Power, and P O´ Gaora, University College Dublin, Dublin, Ireland ã 2016 Elsevier Ltd. All rights reserved.

Enterobacteriaceae All members of the Enterobacteriaceae possess some common bacteriological features. They are gram-negative, rod-shaped facultative anaerobes. Bacterial isolates are classified as being either motile by means of peritrichous flagella or nonmotile, being devoid of any flagella. Members of this family are generally 0.3–1.0  0.6–6.0 mm in size and are non-spore-forming. They can ferment glucose and are catalase-positive (except for Shigella dysenteriae) and oxidase-negative, and they reduce nitrate to nitrite. Table 1 provides a summary of the general characteristics of some of the members of this family. Enterobacteriaceae family members are associated with a wide range of disease (Figure 1). Illnesses associated with members of the Enterobacteriaceae include wound infections, urinary tract infections (UTI), gastroenteritis, meningitis, pneumonia, septicemia, and hemolytic uremic syndrome (HUS). While some of these species are true pathogens, others are regarded as opportunistic. Members of this bacterial family can grow readily on a variety of microbiological culture media. They are also widely dispersed throughout the environment being found in soil, water, plants, and the gastrointestinal tract of animals and humans.

pathogenic bacteria from thriving in the environment of the intestine.

Relevance of E. coli to public health

Pathogenic strains of E. coli cause a variety of illnesses such as food poisoning with severe stomach cramps, watery or bloody diarrhea, vomiting, and nausea. Severe complications include kidney failure, seizures, and stroke. The main route of transmission is through fecal–oral contamination, due to poor hygienic conditions. E. coli has a very fluid genome that enables the uptake of many different mobile genetic elements. This genomic fluidity has had a large impact on the evolution of pathogenic serovars with phages, plasmids, and pathogenicity islands playing a role in their diversity and evolution. This has resulted in the emergence of pathogenic strains that are associated with their own virulence factors and disease symptoms. Pathogenic strains are classified into a number of pathotypes as follows:

• • • • • •

Verotoxigenic E. coli (VTEC) Enterotoxigenic E. coli (ETEC) Enteropathogenic E. coli (EPEC) Enteroaggregative E. coli (EAEC) Enteroinvasive E. coli (EIEC) Diffusely adherent E. coli (DAEC)

The Genus Escherichia The genus Escherichia was described originally by Theodore von Escherich who first discovered Escherichia coli in 1885. This bacterial genus contains five species with E. coli being the best known. Other species include E. albertii, E. fergusonii, E. hermannii, and E. vulneris. A number of species are pathogenic; however, the majority are commensals or opportunistic pathogens.

Features associated with isolates of the E. coli genus

Escherichia coli cells are 0.5 mm in length and 2.0 mm in diameter. Optimal growth occurs at a temperature of 37  C, though some strains can grow at temperatures up to 49  C. The optimal range of pH is between 6.0 and 7.0, with few being able to grow at a pH < 4 or > 9. E. coli grown on MacConkey agar forms small, red/pink, smooth, and circular colonies. When further assessed, these are motile by means of peritrichous flagella (Table 1).

Environments known to contain E. coli

These bacteria can be found in the gastrointestinal tract of warm-blooded animals in which it colonizes within hours of birth. Strains can also be found throughout the environment, in water, food, and soil. Most E. coli isolates are harmless and play an important role as constituents of the normal biota or microbiome found in a healthy human intestinal tract. They aid digestion in the gut, produce vitamin K2, and prevent

Encyclopedia of Food and Health

Pathotypes of Escherichia coli Important to Human Health Verotoxigenic E. coli

VTEC strains are defined by the production of cytotoxins that disrupt protein synthesis within host cells. These toxins are referred to as either verotoxins (vt) due to their activity against Vero cells in vitro or Shiga-like toxin 1 (stx1) and Shiga-like toxin 2 (stx2) due to their similarity to the Shiga toxin produced by Shigella dysenteriae. Dissemination of these strains occurs via the fecal–oral route arising from substandard hygienic conditions. An infectious dose can be as low as 10–100 cells. Ruminant animals are an important reservoir for VTEC. Infections can occur through contact with an infected individual, directly from an animal shedding the organism or its fecal matter, via the consumption of a contaminated food containing the bacterium or through the consumption of contaminated water. Symptoms of a VTEC infection can range from mild diarrhea to severe bloody diarrhea. Potentially fatal complications arising from VTEC infections include HUS and thrombotic thrombocytopenic purpura. Enterohemorrhagic E. coli (EHEC) is a subset of VTEC that is classified as pathogenic to humans. Usually, strains are classified as EHEC when they contain a verotoxin-encoding gene in combination with an eae gene that encodes intimin, an outer membrane protein that is involved in attachment to epithelial cells. Some eae-negative strains may also be EHEC.

http://dx.doi.org/10.1016/B978-0-12-384947-2.00259-2

545

546 Table 1

Escherichia coli and Other Enterobacteriaceae: Occurrence and Detection Typical characteristics of Enterobacteriaceae

Features Size (mm) Motility Optimal Temperature or range ( C) pH range Flagella Encapsulated Pathogen category

Escherichia coli

Salmonella

Yersinia

Cronobacter

Enterobacter

Citrobacter

0.5  2.0 Motile

3.4  0.7 Motile

0.8  1–2 Nonmotile (between 25 and 35  C)

31 Motile

0.6  2–3 Motile

1  2–6 Motile

37 6–7 Peritrichous Yes Major

30–37 6.5–9 Peritrichous Yes Major

28 4–10 Peritrichous Yes Major

6–45 5–10 Peritrichous Yes Opportunistic

30–37 4.4–9 Peritrichous Yes Opportunistic

37 5–8 Peritrichous No Opportunistic

Enterobacteriaceae

Major food-borne pathogens

Escherichia Salmonella coli species

VTEC S. Typhimurium ETEC S. Enteritidis EHEC EPEC EIEC DAEC

Yersinia species

Opportunistic food-borne pathogens

Cronobacter species

Y. pestis C. sakazakii Y. enterocolitica Y. pseudotuberculosis

Enterobacter species

E. cloacae E. aerogenes

Citrobacter species

C. freundii C. koseri

Klebsiella species

K. pneumoniae

Figure 1 Enterobacteriaceae of importance to public health.

They produce highly potent toxins and can form attaching and effacing lesions on epithelial cells, a characteristic typically associated with EPEC strains. In terms of public health, E. coli O157 is one of the most commonly detected EHEC serotypes, although other serotypes are also being recognized.

Outbreaks of EPEC infection have been linked to the consumption of contaminated drinking water as well as some meat products. In the case of tEPEC, humans are thought to be the only reservoir while both animals and humans are reservoirs for aEPEC.

Enteropathogenic E. coli

Enterotoxigenic E. coli

EPEC strains are noted for their ability to produce attaching and effacing (A/E) lesions. In this case, bacteria attach to the host epithelial cell membrane disrupting the cell surface resulting in the effacement of microvilli at the site of adherence. EPEC contains a 35 kbp pathogenicity island called the locus of enterocyte effacement, which encodes genes responsible for the attaching and effacing lesions. These strains are further classified into typical enteropathogenic E. coli (tEPEC) and atypical enteropathogenic E. coli (aEPEC) based on the presence and absence of the E. coli adherence factor plasmid that is only found in tEPEC strains. tEPEC is a well-documented cause of gastroenteritis in infants. aEPEC is more prevalent than tEPEC in both developed and developing countries and is an important risk factor in juvenile endemic diarrhea and diarrheal outbreaks.

ETEC strains are characterized by their expression of enterotoxins and the presence of host-specific fimbriae required for attachment to intestinal cells. Two types of toxins, heatstable (ST) and heat-labile (LT) toxins, are produced by ETEC. Different strains can harbor either one or both of these toxin-encoding loci. The toxins act to stimulate the intestinal lining leading to an increase in fluid secretion resulting in diarrhea. These pathogenic E. coli are the leading cause of traveler’s diarrhea. Symptoms range from mild to severe cholera-like illness characterized by watery stool(s), vomiting, and stomach pains lasting up to 3 days. The infectious dose is estimated at 108 cells but is lower in immunocompromised patients. ETEC is mainly transmitted through food and water contaminated with human and animal feces. In humans, ETEC

Escherichia coli and Other Enterobacteriaceae: Occurrence and Detection infections are widespread though rarely life-threatening and usually self-limiting. It is a major pathogen in animals and can even cause death.

Enteroinvasive E. coli

These bacteria are characterized by their ability to invade epithelial cells and disseminate from cell to cell while replicating within. EIEC differ from typical E. coli in that they are nonmotile and cannot decarboxylate lysine nor ferment lactose. EIEC strains possess an invasion plasmid that encodes the virulence genes conferring EIEC invasive capacity. EIEC strains are highly related to Shigella species at the genetic, biochemical, and pathogenic levels. They cause a syndrome identical to bacillary dysentery more often linked to Shigella species, with profuse diarrhea and high fever. The infectious dose of EIEC is estimated at 106 cells whereas that of Shigella is similar to EHEC, in the range of ten to several hundred cells. There are no known animal reservoirs of EIEC. Transmission usually occurs through contact with infected humans through the fecal–oral route. EIEC infections can also occur through ingestion of contaminated food or water.

Enteroaggregative E. coli

EAEC strains are characterized by their ability to adhere to epithelial cells in an aggregative manner resulting in a stackedbrick pattern. They do not secrete LT or ST enterotoxins. EAEC pathogenesis involves adherence to the intestinal mucosa resulting in increased production and deposition of a mucus biofilm leading to mucosal toxicity due to inflammation and cytokine release. EAEC strains can cause persistent infection by evading the immune system of the host. These organisms are associated with acute or persistent diarrhea with little or no fever and no vomiting. They are especially prevalent among children in developing countries. EAEC strains are considered emerging pathogens, as VTEC O111:H2 strains isolated from a HUS outbreak in France also possessed genetic markers of EAEC E. coli. EAEC outbreaks have been linked to infant food including formula, milk, and water. In the summer of 2011, a novel strain of E. coli O104:H4 emerged and caused a serious food-borne outbreak in Lower Saxony, Germany. This outbreak was associated with serious illness such as bloody diarrhea with a high incidence of HUS. Epidemiological investigations reported 3816 cases with 845 instances of HUS and 54 deaths. At first, E. coli O104:H4 was thought to be an EHEC strain, but it was later identified as an EAEC that had acquired the ability to produce Shiga toxin.

Diffusely adherent E. coli

DAEC strains are characterized by their adherence to Hep-2 cells in a diffuse pattern. Approximately 75% of these bacteria express the Afa/Dr family of adhesins, but not the typical virulence genes of other E. coli pathotypes. DAEC are a major cause of UTI and can cause watery diarrhea typically without blood but with vomiting. Infection is characterized by the growth of fingerlike cellular projections on the surface of infected epithelial cells that envelop the bacteria. Sources associated with DAEC infections include

547

contaminated food such as undercooked beef, contaminated water, and contact with infected animals.

Enterobacteriaceae of Importance to Human Health The genus Salmonella

This bacterial genus was named after Daniel Elmer Salmon following the discovery of the type strain Salmonella enterica in 1885. The Salmonella genus contains three species, S. bongori, S. enterica (including six subspecies), and S. subterranea. The genus is subdivided into more than 2500 serovars based on antigenic features. Salmonella enterica and its serovars are most commonly associated with human disease.

Features associated with isolates of the Salmonella genus

Salmonella cells are 2–3.4 mm long and 0.7–9.0 mm in diameter, with rounded ends. Most have peritrichous flagella, though S. subterranea has one lateral flagellum. Optimal growth occurs within a pH range from 6.5 to 9.0 and across a temperature range of 30–37  C. Salmonella species do not grow at pHs 9.5 nor at temperatures 42  C. Salmonellae are acid-tolerant and can grow at lower pH limits of 3.7–4.4. Salmonella species can produce hydrogen sulfide from thiosulfate. The motility of Salmonella isolates is strain-dependent and specific to the host environment.

Environments known to contain Salmonella

Salmonella species are isolated from aquatic and sedimentary environments where they can multiply though these are regarded as transitional environments between host infections. Some strains such as S. subterranea are specifically adapted for living in soil. Salmonellae possess the ability to survive in lowmoisture conditions and are associated with many food-borne outbreaks especially in the ready to eat food market.

Relevance of Salmonella to public health

Salmonella species cause illnesses including typhoid fever, paratyphoid fever, and food-borne illnesses. These bacteria are zoonotic pathogens, being present in (food-producing) animals from which they can transfer to humans. S. enterica subspecies enterica serovars Typhimurium and Enteritidis are commonly reported members of this genus. These bacteria can exhibit resistance to one or more antibiotics, compromising treatment in some infected individuals. Nontyphoidal salmonellosis is food-borne, and consequently, the food chain is an important route of infection for humans. Salmonella typhimurium is one of the serovars most often associated with cases of food-borne infections. Clinical signs associated with salmonellosis include diarrhea, dehydration, fever, and abdominal pain that can last for up to 7 days. There is a greater risk of infection in the young, the elderly, and immunocompromised individuals.

The genus Yersinia

Yersinia pestis (the ‘plague bacillus’) was first identified by Alexandre Yersin in 1894 and classified as Pasteurella pestis soon thereafter. Von Logham reclassified Pasteurella pestis and Pasteurella rodentium into the new genus of Yersinia in 1944. The name was accepted as an official genus in 1980. The genus Yersinia contains 17 species, of which only 3 are known to be

548

Escherichia coli and Other Enterobacteriaceae: Occurrence and Detection

pathogenic to humans. These include Y. pestis, Y. enterocolitica, and Y. pseudotuberculosis.

Features associated with isolates of the Yersinia genus

Yersinia species form gray-white translucent colonies after 24 h of growth on blood agar plates. After 48 h on MacConkey agar, they form clear colonies. Yersinia strains are nonmotile at temperatures between 25 and 35  C (Table 1). Y. enterocolitica and Y. pseudotuberculosis are psychotropic, being capable of growth at temperatures of 4  C or even lower.

Environments known to contain Yersinia

Rodents are the natural reservoirs of Yersinia species, although other mammals can occasionally serve as hosts. Yersinia pestis can be transmitted to humans subcutaneously through flea bites and can become airborne during pandemics. Due to the ability of enteropathogenic Yersinia species to grow at low temperatures, refrigerated foodstuffs can become contaminated with these organisms. Swine and wild animals are common reservoirs.

Relevance of Yersinia to public health

The pathogenic species of Yersinia have been linked with a wide range of illnesses in humans including Crohn’s disease, yersiniosis, mesenteric lymphadenitis pseudoappendicitis, and systemic infectious disease, commonly known as plague. Y. pestis is the causative agent of the plague, which has been responsible for three human pandemics throughout history. These are the Justinian plagues, recorded from the sixth to eighth centuries, the Black Death from the fourteenth to nineteenth centuries, and the modern plague from the nineteenth century to the present time. Y. enterocolitica is primarily a food-borne pathogen found in some food-producing animals such as pigs and other mammals. After ingestion of contaminated water or food, Y. enterocolitica colonizes the intestine causing yersiniosis, an acute gastrointestinal condition. Symptoms include fever, abdominal pain, vomiting, and diarrhea. Treatment usually consists of an aggressive course of antimicrobial chemotherapy. Y. pseudotuberculosis is the least common of the three Yersinia pathogenic strains. It causes an illness characterized by fever and acute abdominal pain arising from mesenteric lymphadenitis, an inflammation of the lymph nodes.

The genus Cronobacter

Originally, members of this bacterial genus were described as yellow-pigmented Enterobacter cloacae being subsequently reclassified as Enterobacter sakazakii in 1980. A further and more recent reclassification was reported in 2007 with the acceptance of the new genus name, Cronobacter, based on a polyphasic study of a large collection of these isolates. Currently, the genus Cronobacter contains seven species, namely, C. sakazakii, C. malonaticus, C. turicensis, C. muytjensii, C. dublinensis, C. condimenti, and C. universalis.

Features associated with isolates of the Cronobacter genus

Cronobacter are 3 mm long and 1 mm in diameter. Cronobacter species produce two colony morphotypes, one being described as glossy and the other having a matte texture. Some 80% of strains identified produce a yellow pigment on tryptone soya

agar at 25  C. These strains can grow over a wide range of temperatures from 5  C to 44–47  C depending on the isolate. Optimal growth occurs in pH ranging from five to ten with some isolates being more susceptible to acid exposure. Cronobacter have a notable tolerance to desiccation, being able to withstand dry conditions for up to 2 years in infant formula. Cronobacter strains are motile by means of peritrichous flagella.

Environments known to contain Cronobacter

Cronobacter has been isolated from a variety of food and environmental sources such as dairy products, dried meats, powdered infant formula (PIF), households, livestock facilities, food manufacturers, and PIF production facilities.

Relevance of Cronobacter to public health

Cronobacter species are regarded as opportunistic pathogens linked with life-threatening infections in neonates. Clinical presentation can include necrotizing enterocolitis, bacteremia, and meningitis. These organisms can also infect older adults, although these cases tend not to be life-threatening.

The genus Enterobacter

The genus Enterobacter was accepted in 1960 with Enterobacter cloacae being designated as the type species. With the application of modern typing techniques, nine species are currently recognized in this genus. These are E. cancerogenus, E. aerogenes, E. mori, E. ludwigii, E. kobei, E. soli, E. asburiae, E. cloacae, and E. hormaechei.

Features associated with isolates of the Enterobacter genus

Enterobacter species are motile by four to six peritrichous flagella. Some are encapsulated (Table 1). Generally, these bacteria are 0.6–1.0  2–3 mm in size. Enterobacter species produce round, iridescent, flat, nonpigmented, irregular-edged colonies, when grown on nutritive agar. The optimum growth temperatures are between 30 and 37  C, though they can grow in temperatures up to 44  C.

Environments known to contain Enterobacter

Enterobacter can be found on human skin, plants, soil, water, sewage, intestinal tracts of animals, including humans, dairy products; and clinical specimens such as feces, urine, blood, sputum, and wound exudates.

Relevance of Enterobacter to public health

Enterobacter species are considered opportunistic pathogens, rarely causing disease in healthy individuals. The species E. cloacae and E. aerogenes are the main pathogens in the genus. These organisms are associated with bacteremia, lower respiratory tract infections, skin and soft tissue infections, UTIs, endocarditis, osteomyelitis, arthritis, and ophthalmic infections. Outbreaks involving E. cloacae have been reported in the ICUs of hospitals. E. cancerous is rarely associated with human infections and is considered a plant pathogen.

The genus Citrobacter

The genus Citrobacter was first proposed in 1932. Citrobacter contains 11 species: C. amalonaticus, C. farmeri, C. braakii,

Escherichia coli and Other Enterobacteriaceae: Occurrence and Detection C. freundii, C. gillenii, C. murliniae, C. sedlakii, C. werkmanii, C. youngae, C. koseri, and C. rodentium.

Features associated with isolates of the Citrobacter genus

Citrobacter species are 1.0  2.0–6.0 mm in size. They are found either singly or in pairs, are devoid of a capsule, and are motile. C. freundii form small, circular, convex dark pink colonies on MacConkey agar. Rough or mucoid forms have also been reported. Citrobacter species grow optimally at a temperature of 37  C.

Environments known to contain Citrobacter

Citrobacter are found in a variety of environmental sources, including soil and water, and in the human intestines. They are rarely the primary source of illness, though some strains can cause infections of the urinary tract, sepsis, and infant meningitis.

Relevance of Citrobacter to public health

Citrobacter species are not regarded as significant etiological agents in human disease. C. freundii and C. koseri have mainly been isolated from urinary and respiratory tract infections. Citrobacter can cause septicemia in patients that display a number of predisposing factors. Citrobacter have also been found to cause meningitis, septicemia, and pulmonary infections in neonates and young children, and some of these cases have been linked to contaminated batches of PIF. Citrobacter is considered an opportunistic pathogen.

Detection of Members of the Enterobacteriaceae Family Using Conventional and Molecular Methods Conventional microbiology Conventional microbiological-based techniques make use of morphological and biochemical differences between genera and bacterial isolates recovered from different sources. Selective and differential bacterial culture media for enrichment and isolation of strains have been developed.

Examples of detection using bacterial morphology and reactions on specific culture media

Macroscopic morphology can aid with the identification of features visible to the naked eye. It describes colony appearance in terms of its shape, texture, pigment, speed of growth when incubated, and the growth pattern. Microscopic morphology allows for an examination of individual bacterial cells for their shape, size, Gram stain, and acid-fast reactions to aid identification. The goal is to achieve an initial presumptive identification of the bacterium on the culture plate. The application of specially designed culture media is a method commonly used to aid the identification of key pathogens of significance to public health. As an example, E. coli O157 must be identified quickly due to its association with low infectious doses. Its impact on human health makes it the focus of numerous commercial diagnostic companies. Many of these exploit the inability of VTEC O157:H7 strains to ferment the carbon source contained in sorbitol MacConkey agar. When grown on this media, VTEC O157:H7 produces a colorless colony distinguishing it from other organisms on the same plate. However, more recently, some E. coli O157 have

549

been identified with the ability to ferment sorbitol so other means for confirmatory identification must be developed and applied. Non-O157 strains may sometimes be differentiated through the use of Chromocult and Rainbow agar on the basis of their colony color. However, currently, there are no suitable commercial agars available for this purpose. Similarly, Salmonella species are cultured from food (after a period of resuscitation) and from clinical samples directly onto selective agar media, usually xylose–lysine–deoxycholate agar, and incubated at 37  C for 24 h. Colonies can be preliminarily classified based on their morphology.

Examples of diagnostic biochemical tests

Commercially available biochemical galleries exploit the expression or otherwise of different phenotypic characteristics to identify bacteria. Some of these tests exploit the biochemical capacity of an isolate in terms of its ability to ferment a range of sugars, susceptibility to various antibacterial agents including drugs, and the capacity to metabolize complex polymers. In addition, some strategies will allow for the assessment of the motile nature of an isolate (see Table 1 in the preceding text).

Molecular-based diagnostic microbiology Polymerase chain reaction

The polymerase chain reaction (PCR) is an in vitro version of the natural DNA replication process that occurs within all living cells. It can be applied to rapidly amplify a discrete segment of genomic DNA. Characteristically, double-stranded DNA is denatured, separating it into two discrete strands. Primers, short oligonucleotide sequences of synthetic DNA, hybridize to opposite DNA strands, and, following the addition of a thermostable DNA polymerase and the four dNTPs, a round of synthesis is initiated, extending the DNA from the synthetic primer on each strand. Up to 30 reiterated cycles of denaturation, annealing, and extension are completed within a typical PCR reaction, amplifying the original DNA target exponentially. In a more recent development of the conventional PCR approach, quantitative real-time PCR (qPCR) emerged. This strategy enables quantification of target-specific amplicons. In this approach, a housekeeping gene is chosen to act as a normalized comparator against which a variable DNA target is measured, resulting in quantification of the latter. qPCR can be applied for the quantification of individual genes and bacteria of importance to human health. As an example, VTEC organisms can be detected by targeting their unique virulence genes, including stx and eae (as described earlier). Real-time PCR assays have been developed that allow the detection of the major VTEC-associated genes enabling the differentiation of strains in real time. These targets can be combined into multiplex assay platforms that can facilitate the detection of several different VTEC serogroups simultaneously. Various PCR-based methods for the detection of Cronobacter strains exist. These include a conventional 1,6-aglucosidase-based PCR and a dnaG-based qPCR that can detect all seven of the Cronobacter species. When used in combination with a species level PCR assay, individual species of the genus can be identified. This PCR method used the rpoB gene for which specific primers have been designed for each species.

550

Escherichia coli and Other Enterobacteriaceae: Occurrence and Detection

Pulsed-field gel electrophoresis

Pulsed-field gel electrophoresis (PFGE) makes use of suitable restriction enzymes to digest bacterial DNA at a limited number of sites within the genome, producing large or macro-DNA fragments that can be separated based on their size. The careful separation of these DNA fragments is achieved through the application of a periodical alternating electric field to a gel matrix. Comparison of these restriction profiles can indicate whether or not isolates are epidemiologically linked. PFGE has been used to aid with the tracking of many food-borne outbreaks and is regarded as the gold standard in molecular subtyping. PFGE has been standardized between different laboratories allowing for the comparison of PFGE profiles. PulseNet is an initiative that was originally developed at the CDC that facilitates the global tracking of food-borne pathogens.

Multilocus sequence typing

Multilocus sequence typing is a genome-based protocol used for the characterization of bacterial species by sequencing internal regions of typically seven housekeeping genes. To enable accurate sequencing, fragments of between 450 and 500 base pairs are used. Bacterial species display sufficient polymorphism within the chosen genes to assign an identifier to the alleles detected. By combining the sequences of (usually up to) seven loci, a species-specific allelic profile can be generated for each bacterial isolate.

Optical mapping

Optical mapping is a recently developed technique that generates high-resolution, ordered, whole-genome restriction maps from single DNA molecules independently of available sequence information. Purified DNA is isolated from a bacterium of interest and digested with a suitable restriction endonuclease. The DNA is then stained, scanned, accurately measured, and subsequently assembled into a complete genome restriction map. These maps can be compared to discover genetic differences, such as insertions or deletions between strains. Optical mapping can be used to distinguish between different species and isolates. Optical mapping was applied to distinct E. coli O104:H4 German outbreak isolates in order to characterize differences. In this case, regions of differences contributing to the pathogenicity of this unusual strain were identified.

A number of NGS technologies now exist for this purpose, such as

• • • • •

Illumina (Solexa) sequencing, Roche 454 sequencing, Pacific Biosciences single-molecule real-time sequencing, SOLiD sequencing, Life Technologies Ion Torrent sequencing.

Despite the differences in the chemistries used, all the methods in the preceding text share some common features such as the random fragmentation of the template DNA, amplification of the target to be sequenced on a solid surface such as a bead or glass slide, and direct step-by-step detection of each nucleotide, following incorporation. Sequencing technologies provide for the determination of the base order from thousands of small fragments that are subsequently assembled to generate a consensus genome sequence. Once assembled, this draft sequence must then be checked to confirm that correct assembly has taken place, followed by careful gene annotation and importantly the discovery of features of interest, including the identification of resistance and virulence genes along with other phenotypic data on a molecular basis. NGS is a high-resolution and therefore accurate and reliable means of bacterial characterization. In the 2011 German outbreak mentioned earlier, NGS technology facilitated the whole-genome characterization at an early stage of this incident. An E. coli O104:H4 outbreak isolate and a historic E. coli O104:H4 HUS isolate from 2001 were characterized using Ion Torrent’s NGS in combination with optical mapping. Within 62 h, draft genomes were available enabling the genes to be examined. It was found that the strains carried genes associated with both EAEC and EHEC. Hence, it was proposed that the outbreak strain was a highly pathogenic hybrid of EAEC and EHEC strains.

See also: Emerging Foodborne Enteric Bacterial Pathogens; Escherichia coli and Other Enterobacteriaceae: Food Poisoning and Health Effects; Salmonella: Detection; Salmonella: Properties and Occurrence; Salmonella: Salmonellosis; Shigella.

Further Reading Whole-genome sequencing

Genome sequencing involves the decoding of a bacterium’s complete genetic code, more specifically the determination of nucleotide order, within the chromosome. The technology underpinning this approach has evolved following a number of advancements over the years, and it is the most powerful analytic approach to molecular determination.

Next-generation sequencing

Next-generation sequencing (NGS) or high-throughput sequencing describes a number of sequencing technologies in common use. These approaches generate large volumes of sequence data at relatively low cost, thereby facilitating the rapid sequencing of DNA and RNA. This technology is significantly quicker and cheaper than the traditional Sanger-based technology.

Bolton DJ, O’Sullivan J, and Duffy G, et al. (eds.) (2007) Methods for detection and molecular characterisation of pathogenic Escherichia coli. Ireland: Pathogenic Escherichia coli Network. http://www.antimicrobialresistance.dk/data/images/ protocols/e%20coli%20methods.pdf. EFSA Panel and on Biological Hazards (BIOHAZ) (2013) Scientific opinion on VTECseropathotype and scientific criteria regarding pathogenicity assessment. EFSA Journal 11(4). EFSA Panel and on Biological Hazards (BIOHAZ) (2014) Scientific opinion on the evaluation of molecular typing methods for major food-borne microbiological hazards and their use for attribution modelling, outbreak investigation and scanning surveillance: part 2 (surveillance and data management activities). EFSA Journal 12(7). Ekiri AB, Landblom D, Doetkott D, Olet S, Shelver WL, and Khaitsa ML (2014) Isolation and characterization of shiga toxin-producing escherichia coli serogroups O26, O45, O103, O111, O113, O121, O145, and O157 shed from range and feedlot cattle from postweaning to slaughter. Journal of Food Protection 77(7): 1052–1061. Fricke WF and Rasko D (2014) Bacterial genome sequencing in the clinic: bioinformatic challenges and solutions. Nature Reviews Genetics 15(1): 49–55.

Escherichia coli and Other Enterobacteriaceae: Occurrence and Detection Kaper JB, Nataro JP, and Mobley HLT (2004) Pathogenic Escherichia coli. Nature Reviews Microbiology 2: 123–140. Latreille P, Norton S, Goldman BS, Henkhaus J, Miller N, Barbazuk B, Bode HB, Darby C, Du Z, Forst S, Gaudriault S, Goodner B, Goodrich-Blair H, and Slater S (2007) Optical mapping as a routine tool for bacterial genome sequence finishing. BMC Genomics 8: 321. Leekitcharoenphon P, Nielsen EM, Kaas RS, Lund O, and Aarestrup FM (2014) Evaluation of whole genome sequencing for outbreak detection of Salmonella enterica. PloS One 9(2), p.e87991. Mellmann A, Harmsen D, Cummings C, et al. (2011) Prospective genomic characterization of the German enterohemorrhagic Escherichia coli O104:H4 outbreak by rapid next generation sequencing technology. PloS One 6(7) p.e22751. Metzker ML (2010) Sequencing technologies – the next generation. Nature Reviews Genetics 11(1): 31–46.

551

Pennington TH (2014) E. coli O157 outbreaks in the United Kingdom: past, present, and future. Infection and Drug Resistance 7: 211–222. Quinn PJ and Markey BK (2003a) Enterobacteriaceae 1. In: Concise review of veterinary microbiology, pp. 38–42. Oxford: Blackwell Publishing. Quinn PJ and Markey BK (2003b) Concise review of veterinary microbiology. Oxford: Blackwell Publishing. Robinson ER, Walker TM, and Pallen MJ (2013) Genomics and outbreak investigation: from sequence to consequence. Genome Medicine 5(4): 36.

Relevant Websites http://www.cdc.gov/ecoli/ – CDC Escherichia coli. www.cdc.gov/pulsenet – CDC PulseNet.

Essential Oils: Isolation, Production and Uses CM Cook, Hellenic Agricultural Organization-Demeter, Thermi, Greece T Lanaras, Aristotle University of Thessaloniki, Thessaloniki, Greece ã 2016 Elsevier Ltd. All rights reserved.

Background Essential oils are volatile, aromatic components of herbs and spices and have been used since ancient times in flavoring, food preservation, medicine, and perfumery. Some wellknown aromatic and spice plants from which essential oils are produced commercially are shown in Table 1. The plant families Lamiaceae and Apiaceae are particularly rich in aromatic plants. The essential oil can be found in the flowers, leaves, roots, rhizomes, fruits, seeds, wood, and resins of these plants. On the surface of leaves and calyces, the essential oil is located in specialized cells called glandular trichomes. Essential oils are synthesized by plants, as secondary metabolites, and are bioactive compounds with biological properties. They are considered to play a role in plant protection against herbivores making them unpalatable; in the attraction of pollinators; in plant protection against fungal, viral, and bacterial infections; in communication and signaling between other plants; and in allopathy by reducing the competition of neighboring plant species, among others. The essential oil content of plant material can vary with season, bioclimatic zone, growing conditions, and stress. Essential oils can easily be isolated from plant material by hydrodistillation and are liquids at room temperature. They are a complex mixture of terpenes and terpenoids, which make up the characteristic aroma of each plant, and it is the composite effect of all the constituents present in it. They can be present from a mere trace up to around 95% of the essential oil. Most often, main components are precursors, intermediates, and end products of a particular biosynthetic pathway, and their relative proportions may depend on the time of year or the developmental stage of the plant. Terpenes are chemical compounds of low molecular weight, which are made up of isoprenoid units consisting of 5 carbon atoms, and these can join together to form units of C10 mono-, C15 sesqui-, C20 diterpenes. Oxygenated terpenes or terpenoids can be aldehydes, alcohols, ketones, acids, esters, ethers, and others. Natural biodiversity is reflected by plants and their essential oils. It is possible to have several different species producing an essential oil with the same composition and major component (same biosynthetic pathway), for example, carvacrol in ‘oregano’ plants: Origanum vulgare subsp. hirtum (‘Greek’), O. onites (‘Turkish’), and Thymbra capitata (‘Spanish’ oregano). On the other hand plants of the same species, for example, Mentha spicata can have completely different aromas: ‘peppermint’ (menthol), ‘spearmint’ (carvone), ‘wild mint’ (piperitone and piperitenone oxide), and ‘lavender’ (linalool and linalyl acetate). These are known as chemotypes, plants of the same species having terpenoids arising from different biosynthetic pathways. A return to the use of natural products in everyday life has stimulated interest in different aspects of essential oils and prospective uses in medicine, cosmetics, and the food industry. This interest is reflected in the marked increase in research

552

publications on essential oils over the last twenty years (Figure 1) focused primarily on not only essential oil composition and biological properties but also isolation methodologies, species variation, chemotaxonomy, and others.

Isolation The volatile nature, general insolubility in water, and solubility in organic solvents are properties of essential oils that can be exploited when designing isolation methods. The nature of the plant material and the lability of essential oil components should also be considered. Traditional and commercially most commonly used methods are distillation using water and/or steam, cohobation, maceration, enfleurage, and cold pressing. However, new technologies are continually being developed in order to decrease extraction times and energy costs.

Traditional Technology Hydrosteam distillation Distillation using water and/or steam is the most widely used and cost-effective method for the production of the majority of essential oils worldwide. It involves the application of heat and moisture to vaporize and release the essential oils from the plant material followed by the cooling of the vapor mixture and the separation of the oil from the water. Although the boiling point of most essential oil components generally ranges from 150 to 300
C, they can be evaporated with steam/boiling water at 100
C, as the combined vapor pressures of two immiscible liquids will be equal to atmospheric pressure at a lower temperature than that at which the essential oil will vaporize. Hydrodistillation is the most simple of processes and has been used by man for centuries for the extraction of essential oils and is used mostly by small-scale producers. The plant material is almost completely covered by water in the chamber. The water is boiled using an external heat source, and the essential oils together with the steam are condensed and separated. Essential oil content of aromatic plants is determined using a Clevenger-type distillation apparatus, according to the European Pharmacopoeia. Disadvantages of hydrodistillation are that the process is slow and requires energy for heating, distillation rates may vary if the heat source is not controlled, and the direct heat source may cause charring of plant material in the base of the chamber. Prolonged contact of essential oils with the hot water can cause the hydrolysis of some essential oil constituents, for example, esters can hydrolyze to acids and alcohols. However, advantages of hydrodistillation are high yields of essential oil and good recovery of constituents. It is simpler and faster than steam distillation. In steam and water distillation, direct contact with the boiling water is avoided by supporting the plant material on a grid. Advantages are that the process is faster and requires less fuel

Encyclopedia of Food and Health

http://dx.doi.org/10.1016/B978-0-12-384947-2.00261-0

Essential Oils: Isolation, Production and Uses Table 1 Some well-known aromatic plants and spices used commercially for their essential oil Plant family

Common (and scientific name)

Anacardiaceae Annonaceae Apiaceae

Mastic (Pistacia lentiscus var. chia) Ylang-ylang (Cananga odorata) Dill (Anethum graveolens), parsley (Petroselinum crispum), anise (Pimpinella anisum), coriander (Coriandrum sativum), cumin (Cuminum cyminum), caraway (Carum carvi) Chamomile (Chamaemelum nobile and Matricaria chamomilla), tarragon (Artemisia dracunculus), yarrow (Achillea millefolium) Frankincense (Boswellia sp.) Valerian (Valeriana officinalis) ‘Greek’ oregano (Origanum vulgare subsp. hirtum), peppermint (Mentha x piperita), spearmint (Mentha spicata), sage (Salvia officinalis), lavender (Lavandula sp.), rosemary (Rosmarinus officinalis), basil (Ocimum basilicum), marjoram (Origanum majorana), lemon balm (Melissa officinalis), pennyroyal (Mentha pulegium), thyme (Thymus sp.), patchouli (Pogostemon cablin) Bay (Laurus nobilis), cinnamon (Cinnamomum sp.) Nutmeg (Myristica fragrans) Clove (Syzygium aromaticum), eucalyptus (Eucalyptus sp.), wintergreen (Gaultheria sp.) Jasmine (Jasminum sp.) Black pepper (Piper nigrum) Citronella grass (Cymbopogon nardus, C. winterianus) Damask rose (Rosa x damescena and R. centifolia) Lemon (Citrus limon), bergamot orange (C. bergamia) Sandalwood (Santalum sp.) Star anise (Illicium verum) Cardamom (Amomum sp. and Elettaria sp.), ginger (Zingiber officinale), turmeric (Curcuma longa)

Asteraceae Burseraceae Caprifoliaceae Lamiaceae

Lauraceae Myristicaceae Myrtaceae Oleaceae Piperaceae Poaceae Rosaceae Rutaceae Santalaceae Schisandraceae Zingiberaceae

553

than hydrodistillation while giving high reproducible yields. The method is used in developing countries as portable field distillation units, combining simple construction, ease of operation, and low cost with capacities of 100–2000 kg plant material. In direct steam distillation, steam is generated in a boiler (external source) that is then passed through the plant material, which is supported on a grid in the chamber. Steam at atmospheric pressure has a temperature of 100
C; however, the use of high-pressure steam correspondingly increases the temperature enabling a more rapid and complete distillation of essential oil. The advantages are reduced fuel costs due to higher thermal efficiency and suitability for large-scale production (1–3 ton plant material). However, setup costs are quite high and stills should be of stainless steel. Distillation with cohobation: most essential oils have very low solubility in water; however, the solubility of rose oil, for example, is quite high. To avoid the loss in the condensing water, this water is returned to the still when using water and steam distillation for redistillation.

Solvent extraction In the perfume industry, essential oil production is carried out using volatile solvents such as petroleum ether and hexane at 50
C. This temperature is lower than that of distillation reducing the possibility of degradation of thermally labile compounds, and thus, essential oils are considered to have a more natural odor. After filtration, the essential oil is recovered by reducing the volume of solvent under vacuum evaporation; however, this can lead to severe losses of essential oil volatile components. Solvent extraction is advantageous when plant material is very limited (e.g., single leaf analysis) or where the major essential oil constituents are of interest rather than the essential oil content. Disadvantages are that nonvolatile compounds may be coextracted and that toxic solvent residues remain in the extract and can then contaminate food and fragrances. The process is also more expensive than distillation and creates waste disposal problems for the used solvents.

Enfleurage Enfleurage is the extraction of essential oils using cold fat and has been used traditionally in the production of perfumes from fresh flower petals with low essential oil content, for example, jasmine and violets. The process has limited application on a large scale as it is very labor-intensive and has been superseded by modern processes such as extraction with volatile solvents. Dry flower petals are placed on glass plates coated with a thin layer of a high-quality odorless fat. Emitted essential oils are absorbed by the fat, and after extraction (in the order of 24 h or more hours), the petals are removed (defleurage) and replaced by fresh petals. This is an accumulative process that is repeated until the end of the flowering season. The essential oil that is absorbed in the fat is subsequently extracted with alcohol.

Number of publications

3000

2000

1000

Cold pressing (expression) 2000

2005 Year

2010

Figure 1 Trends in essential oil publications: 1995–2014. Web of Science (all databases), Thomson Reuters (2015) Thomson Reuters Web of Science. Available online: http://apps.webofknowledge.com (2015) (Accessed 30 January 2015).

This mechanical method is almost exclusively used in the production of citrus fruit (e.g., lemon, orange, and bergamot) essential oils. Pressure is used to crush the peel (pericarp) by rolling the fruit against sharp objects. Essential oil glands located in pits on the pericarp are punctured, releasing the essential oil, which is washed away with water. The essential

554

Essential Oils: Isolation, Production and Uses

oil and water are separated by centrifugation. The process is carried out at low temperature so there is no thermal degradation of the essential oil components. Essential oil can be extracted before or after the extraction of the juice. Traditionally, expression was carried out by hand.

New ‘Green’ Technology In recent years, research has centered on designing more efficient extraction procedures, which give higher yields (process intensification) with lower energy consumption and lower ambient operational temperatures. Conventional steam distillation and solvent extraction procedures although being generally simple have disadvantages of requiring long extraction times and large volumes of solvents. Industries are moving towards the application of greener nonconventional technologies with an emphasis on safer, sustainable, environmentally friendly, and more economical protocols. These nonconventional technologies include supercritical fluid extraction (SFE), solid-phase microextraction (SPME), ultrasoundassisted extraction (UAE), and microwave-assisted extraction (MAE), which are increasingly finding applications at a pilot and industrial level. It is interesting to see whether the trends in essential oil publications with respect to methods of isolation (MAE, solvent extraction, SPME, SFE, and hydrodistillation) for the period 1995–2014 (Figure 2) follow the trends in the development of new nonconventional technologies. Hydrodistillation is by far the primary isolation method reported in publications that would appear to be a result of the simplicity and low cost of the method. However, rising trends are also observed for SFE and SPME as might be expected.

Supercritical fluid extraction (SFE)

Number of publications

Carbon dioxide is a supercritical fluid when it is subjected to temperatures and pressures above its critical temperature (304.25 K) and pressure (7.39 MPa). It then expands like a gas but has the density of a liquid. Supercritical CO2 is an

1500

important industrial solvent, which is odorless, colorless, nonflammable, and of low toxicity and environmental impact, compared to other organic solvents used in essential oil extraction, such as hexane and acetone, which are toxic and flammable. Other advantages of supercritical CO2 extraction are relatively low temperature, shorter extraction times, selectivity, solvating power that varies with temperature and pressure enabling fractional extraction, absence of toxic residues that remain in extracts, absence of oxidation of essential oil compounds, ease of isolation of analytes from the supercritical CO2 by pressure reduction (evaporation), and CO2 recycling by condensation. For essential oil extraction by SFE fractionation, supercritical CO2 at temperatures of 40–50
C and a pressure of about 11 MPa is passed into a heated extraction column packed with plant material. Usually, the plant material is incubated for a period of time (static mode) followed by a flow of new fluid (dynamic mode). The supercritical CO2 with dissolved compounds is subjected to pressure reduction to remove the CO2. Isolated essential oils may contain small amounts of other coextracted compounds such as cuticular waxes.

Solid-phase microextraction (SPME) SPME is a rapid, solvent-free sample extraction technique. The SPME apparatus consists of a fiber holder and assembly containing a 1–2 cm thin, fused-silica retractable fiber coated with a thin polymeric coating such as polydimethylsiloxane (stationary phase). The sample is placed in a vial sealed with a septum. The fiber is extended through the needle into the headspace above the sample. The volatile analytes establish an equilibrium between the sample matrix, the gas phase above the sample, and the polymeric coating on the fusedsilica fiber to which the analytes adsorb. After a given extraction time, the fiber is withdrawn and can be inserted directly into the injection port of a capillary GC column where the volatile analytes are rapidly, thermally desorbed. As no solvent is injected, there is no solvent front to mask component peaks

microwave extraction solvent extraction solid phase microextraction supercritical fluid extraction hydrodistillation

1000

500

0

1995–1999

2000–2004 2005–2009 Year

2010–2014

Figure 2 Trends in essential oil publications with respect to methods of isolation (microwave-assisted extraction, solvent extraction, solid-phase microextraction, supercritical fluid extraction, and hydrodistillation) for 5 year periods from 1995 to 2014. Web of Science (all databases), Thomson Reuters (2015) Thomson Reuters Web of Science. Available online: http://apps.webofknowledge.com (2015) (Accessed 30 January 2015).

Essential Oils: Isolation, Production and Uses eluting early from the column improving detection limits and resolution. SPME is used in the analysis of flavor and fragrance components in foods and beverages and is easily automated. It can be applied to very small samples and also to living systems.

Ultrasound-assisted extraction (UAE) Ultrasound (US) has been used traditionally in biochemistry as a procedure for the production of plant extracts, as it causes the rupture of cell membranes and the release cell contents into the medium. The main factors that cause cell breakage are the water jets and shock waves caused by the collapse of cavitation bubbles generated by US. Part of the US pressure wave is dissipated as heat. The application of US to essential oil isolation enhances the homogenization of the plant material and thereby decreases the extraction time and the total energy consumed in the process. Analytes are more concentrated in the extract. The process is clean with no residues left in the extract. Combined with other extraction procedures, it can improve yield, selectivity, and product quality. Recently, it has been shown that UAE using large-scale multiple transducer flow reactors operating at high power density (700 W l1) promotes process intensification in the isolation of essential oils as compared to classical maceration in hydroalcoholic extracts. US has also been used to improve hydrodistillation extraction (sono-Clevenger) by greatly reducing extraction time without reducing yield or essential oil quality. Combined US and hydrodistillation extractions have been implemented on an industrial scale.

Microwave-assisted solvent extraction (MAE) MAE is an extraction technique that combines microwave radiation technology with traditional solvent extraction. Microwave radiation is used to heat the solvent and plant material. Advantages of MAE are shorter extraction times (minutes rather than many hours), the use of less solvent, higher extraction yield, higher-quality products with lower cost, and lower energy consumption and CO2 emissions when compared to Soxhlet or solvent extraction with stirring. Apparatus is simpler and less expensive than other methods such as SFE. Under optimized conditions (solvent volume, power, and heating time), the quality of essential oil and yield obtained are similar to those obtained by hydrodistillation.

Solvent-free microwave extraction (SFME) Solvent-free microwave extraction (SFME) combines microwave heating and ‘dry’ distillation (no added solvent or water) under conditions of atmospheric pressure and 100
C. Isolation and concentration of volatile compounds are performed in a single step. The internal heating of water within the plant material swells the plant cells and leads to rupture of the essential oil glands. Essential oil is evaporated by the in situ water of the plant material, and the distillate is passed through a cooling system outside the microwave. Excess water is returned to the extraction vessel to restore the in situ water of the plant tissue. SFME significantly reduces the time, energy, and plant material required for extraction. The reduction in extraction time and small amounts of water utilized decrease the likelihood of the hydrolysis, esterification, and/or oxidation of essential oil compounds.

555

Production Production technology should be optimized and standardized in order to improve the overall yield and quality of the essential oil product. Raw plant materials, extraction yield and extraction time, and quality (composition, color, and texture) are important factors for industrial production, which looks for efficient, fast, and economic processes. Subsequently, storage conditions and containers should be optimized for maximum shelf life. The quality of the raw plant material and an adequate supply for industry can be guaranteed by sustainable agricultural practices. Indiscriminate collection from the wild can lead to the reduction of wild resources, loss of biodiversity, and plant material of inconsistent quality. Cultivation of selected plants with high essential oil and biomass yields and of desired chemical composition (quality) is preferable. Harvesting times and postharvest drying and processing conditions must also be optimized and standardized so as to minimize any deterioration in quality and yield (appearance and loss of essential oil). Extraction procedures should be carried out in stills and vats made of noncorrosive materials (e.g., stainless steel and glass) of optimal design and under optimal standardized conditions (temperature and time of extraction) for reproducible results. Plant materials such as seeds, roots, and wood may need to be crushed, ground, or soaked prior to extraction. Distilled essential oils with a high moisture content must be clarified using desiccants or on an industrial scale by high speed centrifugation. Moisture, oxygen, high temperature, and light adversely affect the storage life of the essential oil. Essential oil composition (quality) can be analyzed using gas chromatography (separation of compounds) coupled with detection methods such as mass spectrometry (MS). For example, the gas chromatographic analysis of oregano (Origanum vulgare subsp. hirtum) essential oil (Figure 3) indicates many component peaks and the presence of four main components p-cymene (5), g-terpinene (6), thymol (9), and the major component carvacrol (10). Some characteristics of these components are given in Table 2. The standardization and quality control of essential oils ensure product quality and safety in worldwide markets. Standards are important and aid in the detection of adulterated and synthetic essential oils (addition of synthetic compounds and mixing with cheaper lower-quality oils, or solvents, to increase bulk). Standardization is an important reference for industry providing the properties of the natural components of products. The International Organization for Standardization technical committee ISO/TC 54, Essential oils, has developed 1. specific monographs for quality standardization of all essential oils; 2. standardized analytic methods to control the quality of international standards of analytical methods and specifications; 3. requirements for transport, labeling, and marking; and 4. nomenclature and botanical names, with the aim of expanding global trade in essential oils and ensuring their quality and safety. ISO/TC 54 aims to 1. facilitate the global trade in essential oils, 2. enhance the quality of essential oils on the market, 3. protect the health of essential oil consumers, and 4. enhance the safety of essential oils. Many essential oils (or components) demonstrate toxicity and/or are hazardous in pregnancy and for persons with asthma or epilepsy and thus should be treated as medicines. The

Essential Oils: Isolation, Production and Uses

Pk #

TRACE GC-Channel 1

10

556

3000

3000

OH

2000

2000

1000 12

11

8

7

3 4 5

1 2

6

9

1000

mAU

mAU

OH

0

0 0

10

20

40

30

50

60

70

Time (min)

Figure 3 Gas chromatograph of Origanum vulgare subsp. hirtum (oregano) essential oil. Experimental details: the oil was analyzed by GC-FID using a ThermoQuest Trace GC with an HP-5MS capillary column (5% diphenyl-, 95% dimethyl- polysiloxane, non-polar, 30m  0.25 mm ID  0.25 mm film thickness) coupled with a flame ionization detector. Carrier gas: helium. Flow rate: 1.0 ml min1. Column temperature was 60
C for 5 min, increased 2
C min1 until 150
C and then increased 6
C min1 until 280
C. The injector and detector temperatures were 240
C. Kovats indexes were calculated by comparison of the relative retention times of the separated essential oil components with those of an homologous series of C8-C26 n-alkanes standards.

Table 2 Some characteristics of the four main compounds of oregano essential oil analyzed by gas chromatography (see Figure 3) Peak Compound number name

Kovats index

Molecular formula

Formula weight

Chemical abstract service number

5 6 9 10

1024 1059 1290 1299

C10H14 C10H16 C10H14O C10H14O

134 136 150 150

99-87-6 99-85-4 89-83-8 499-75-2

p-Cymene g-Terpinene Thymol Carvacrol

Compiled from data given in Adams, R. P. (2007). Identification of essential oil components by gas chromatography/mass spectrometry (4th ed). Allured Publishing Corporation: Carol Stream, IL

committee has also participated in the revision of and new European Pharmacopoeia monographs for essential oils.

Food Industry About 60 % of essential oil production is used in the food industry primarily for flavoring and seasoning of food products such as cured meats, alcoholic and soft drinks, ice cream, and confectionary, among others. Essential oils also represent a source of natural antimicrobials for food preservation. Carvacrol, a major component of oregano essential oil, has many beneficial bioactivities in the preservation of food, which include antioxidative properties in food (fats and oils); the inhibition of foodborne human pathogenic bacteria, fungi, and parasites in human foods (eggs, meat and poultry products, milk, and leafy vegetables) and animal feed; and also the inhibition of microbial and fungal toxin production. The many health promoting effects of carvacrol create a potential for its use as a multifunctional food in pure and encapsulated form and in edible films.

Pharmaceutical Industry

Uses Essential oils contain many bioactive compounds (e.g., terpenoids), which, apart from their fragrance and flavor, also show a wide range of biological activities and properties. For example, some of these properties are antibacterial, antifungal, antioxidant, insecticidal, insect repellent, acaricidal, larvicidal, anthelmintic, anti-inflammatory, cytotoxic, antibiotic, anticarcinogenic, analgesic, and local anesthetic. As such, these natural products from more than 300 essential oils are widely used in the food and beverage industry; in the pharmaceutical industry and medicine; in perfumery, cosmetics, and healthcare products (toiletries); and in veterinary science and alternative medicine (aromatherapy). As the uses are many, it is useful to examine them on a general level and then focus on some recent advances relevant to food and health.

Essential oils are generally not recommended for internal use undiluted due to their toxicity, nor for direct application to the skin, with the risk of dermatitis, allergic reaction, or skin photosensitization. However, topical application of some essential oils such as lavender and oregano when diluted in plant oils aids in the healing of wounds and burns. Essential oils are used as analgesics, expectorants, decongestants (menthol and eucalyptus), antiseptics, flavors to neutralize unpleasant tastes, and others. Recently, some essential oil constituents such as geraniol have shown potential antitumor effects, inducing growth inhibition and apoptosis in cancer cells. Mice fed with diets supplemented with geraniol (25–75 mmol kg1) showed a dose- and timedependent growth inhibition of tumor growth in vivo together with an induction of apoptosis. It also reduced cholesterogenesis. The doses of geraniol were nontoxic to the mice; the mevalonate biosynthetic pathway was inhibited. As such, geraniol could be a potential candidate for cancer chemotherapy.

Essential Oils: Isolation, Production and Uses Animal Feed Supplements and Animal Health Essential oils have been evaluated for their effects on growth performance and meat quality of animals raised for meat production (e.g., poultry, pigs, and rabbits) as a safe and natural feed supplement alternative to the overuse of nontherapeutic antibiotics (growth promoters) and the subsequent development of antibiotic resistance in bacterial pathogens of humans and animals. For example, the supplementation of quail (Coturnix coturnix) diets with juniper oil (100 and 150 mg kg1) caused a significant increase in live weight, live weight gain, and carcass yields. Further, the dietary juniper oil reduced thiobarbituric acid levels in raw thigh meat samples in storage and exhibited antioxidant activity, preventing lipid oxidation in stored meat. Generally, essential oils as animal feed supplements can protect against the colonization of pathogenic microorganisms in the gut, reduce the fermentation process and the production of toxic metabolites, and have a positive effect on gut microbiota and animal welfare. Many essential oil constituents (e.g., trans-cinnamaldehyde, thymol, carvacrol, citronellal, neral, linalool, and limonene) are important insect repellents and are effective against mosquitoes, fleas, ticks, flies, and lice, which can transmit infective diseases such as malaria.

Acknowledgment The authors thank Evaggelia Samara for the technical assistance and acknowledge the financial support of the National Strategic Reference Framework (NSRF) Greece, Research Funding Program, Action ARISTEIA II (NATURAL AROMA-4204), for the essential oil analysis.

557

See also: Essential Oils: Properties, Composition and Health Effects; Flavor Enhancers: Characteristics and Uses; Spices and Flavoring Crops: Fruits and Seeds.

Further Reading Alexandru L, Cravotto G, Giordana L, Binello A, and Chemat F (2013) Ultrasoundassisted extraction of clove buds using batch- and flow-reactors: a comparative study on a pilot scale. Innovative Food Science and Emerging Technologies 20: 167–172. Bordas A and Bermejo E (2012) Essential oils. A fresh look at the oldest known remedy and beauty booster. ISO Focusþ July-August: 12–13. Capuzzo A, Massimo ME, and Occhipinti A (2013) Supercritical fluid extraction of plant flavors and fragrances. Molecules 18: 7194–7238. Chen H-C, Chi H-S, and Lin L-Y (2013) Headspace solid-phase microextraction analysis of volatile components in Narcissus tazetta var. chinensis Roem. Molecules 18: 13723–13734. Delazar A, Nahar L, Hamedeyazdan S, and Sarker SD (2012) Microwave-assisted extraction in natural products isolation. Methods in Molecular Biology 864: 89–115. Engelberth J (2010) Secondary metabolites and plant defense. In: Taiz L and Zeiger E (eds.) Plant physiology, 5th ed., pp. 369–400. Massachusetts, USA: Sinauer Associates Inc. Filly A, Fernandez X, Minuti M, et al. (2014) Solvent-free microwave extraction of essential oil from aromatic herbs: From laboratory to pilot and industrial scale. Food Chemistry 150: 193–198. Friedman M (2014) Chemistry and multibeneficial bioactivities of carvacrol (4-isopropyl-2-methylphenol), a component of essential oils produced by aromatic plants and spices. Journal of Agricultural and Food Chemistry 62: 7652–7670. Galle M, Crespo R, Kladniew BR, et al. (2014) Suppression by geraniol of the growth of A549 human lung adenocarcinoma cells and inhibition of the mevalonate pathway in culture and in vivo: Potential use in cancer therapy. Nutrition and Cancer-An International Journal 66: 888–895. Hyldgaard M, Mygind T, and Meyer RK (2012) Essential oils in food preservation: Mode of action, synergies, and interactions with food matrix components. Frontiers in Microbiology 3: 1–24. Yesilbag D, Cengiz SS, Cetin I, Meral Y, and Biricik H (2014) Influence of Juniper (Juniperus communis) oil on growth performance and meat quality as a natural antioxidant in quail diets. British Poultry Science 55: 495–500.

Essential Oils: Properties, Composition and Health Effects G Buchbauer and IM Wallner, University of Vienna, Vienna, Austria ã 2016 Elsevier Ltd. All rights reserved.

Antioxidative Properties of Essential Oils

Sources and Production Essential oils are a group of plant secondary metabolites and are obtained only by steam distillation (according to ISO rule 9235) from different plant organs especially from aromatic plants. Those volatile compounds vary in odor and flavor. The manner of this distillation method is important because it determines the quality of the essential oil. Inappropriate production can result in the loss of bioactivity. Essential oils are concentrated liquids of complex mixtures and contain a lot of bioactive compounds.

Patterns of Consumption Essential oils are lipophilic compounds, which are able to cross membranes very easily. Hence, essential oils are well absorbed not only from the intestine but also through the skin and the lungs. Because of their transdermal and pulmonary absorption, they are used for many medical applications in ointments, balms, and bath additives.

Availability, Absorption, and Metabolism A high amount of essential oils is rapidly absorbed after dermal, oral, or pulmonary administration and crosses the blood–brain barrier. They are able to interact with the receptors in the central nervous system (CNS) and to affect biological functions such as relaxation, sleep, and digestion. Essential oil constituents can be metabolized by the liver in the form of polar compounds following limited phase I enzyme metabolism by conjunction with glucuronate or sulfate. Furthermore, they can be exhaled via the lungs as volatiles or as CO2. Sulfate and glucuronide forms have been detected in urine and in plasma. Because of the fast metabolism and short half-life of active compounds, a minimum risk of accumulation in the body is suggested. The respiratory tract exerts the most rapid way of entry, followed by the dermal pathway. Essential oils and their metabolites can also be absorbed and delivered to the body through oral ingestion.

Nutritional properties of foodstuffs are affected by oxidation of oils. Today, the tendency goes towards using natural compounds such as essential oils for producing functional food. A present study has tested Cinnamon zeylanicum and Zataria multiflora Boiss as two natural preservatives. The use of this essential oil prevents oxidation rate and reduces preliminary and secondary oxidation products compared with butylated hydroxyanisole (BHA). Foods, which contain fat and oils, can be oxidized slowly during storage. As a result, different oxidation products occur, which cause rancidity and reduction of the sensory properties of foodstuffs. Lipid oxidation and fungal growth reduce the shelf life of food products. To solve these problems, manufacturers use antioxidants and preservatives. In the majority of cases, the producers of foodstuff use synthetic additives such as BHA and butylated hydroxytoluene as antioxidants, but the even low toxicity of those products has restricted their use. Recently, aromatic plants and spices serve as a source of biologically active substances such as antioxidants in foodstuffs. Today, the focus is on natural antioxidants that can replace synthetic additives that might be carcinogenic. Cinnamon, which belongs to the Lauraceae family, provides a variety of oils with different aroma characteristics. This aromatic plant contains the highest phenolic contents and strongest antioxidant activity. The main components of cinnamon are cinnamaldehyde and methyl eugenol. Studies have shown good antioxidative and antifungal activities of Cinnamon zeylanicum compared with control samples. The components that are responsible for antiradical activity of cinnamon are eugenol, cinnamaldehyde, cinnamic acid, and 1,8-cineole. The antifungal nature of cinnamon belongs to the high phenolic contents especially carvacrol, thymol, cinnamaldehyde, and eugenol. Food manufacturers need antioxidants that resist at high temperature during baking. Studies have been shown that concentrations of 500 ppm of Cinnamon zeylanicum essential oil can be used instead of BHA. Foodstuff, which contains this oil, might have nutritive and functional advantages compared to BHA. Consumption of foodstuff, which includes natural additives, can help us to prevent health disorders caused by oxidation, for example, aging, atherosclerosis, and cancerogenesis.

Inhibition of Foodborne Pathogens by Essential Oils

Health Effects Essential oils and especially their constituents have a lot of health effects such as antibacterial, antiviral, and antifungal activities. Further fields of applications are anticancer therapy and therapies for cardiovascular and nervous system disorders. Additionally, they are used to reduce the level of cholesterol and regulate the glucose level. Besides, they are useful in the treatment of gynecological diseases. Essential oils are commonly used in the food and cosmetic industries.

558

Foodborne illness that is caused by microorganisms is a growing public health concern. The percentage of people that are suffering from foodborne illness are up to 30%. Unsafe food is responsible for illness and results in several deaths every year. Nowadays, there are an increased concern of food safety and a demand from customers of natural products, which are free from synthetic additives. As a result, the use of natural antimicrobial compounds, such as essential oils, herbs, and spices for food, is gaining interest. Essential oils offer a relatively safe status and a potential for multipurpose functional use.

Encyclopedia of Food and Health

http://dx.doi.org/10.1016/B978-0-12-384947-2.00262-2

Essential Oils: Properties, Composition and Health Effects Especially essential oils from the Lamiaceae and Apiaceae families exert bactericidal properties against microorganisms that contaminate food products; therefore, they are considered as natural food preservatives. Additionally, the widest spectrum of action as well as the greatest antibacterial activity is shown by phenolic compounds, for example, thymol and carvacrol. Above all, Thymus vulgaris L., Origanum vulgare L., and Satureja hortensis L. play an important role. The phenylpropanoid eugenol, which is a phenolic compound, is also known for its antimicrobial properties, and eugenol occurs especially in the essential oils of clove and of cinnamon leaves. All these essential oil compounds, especially as carvacrol, thymol, eugenol, cinnamaldehyde, and cinnamic acid, are used in food to control natural spoilage process and to prevent growth of microorganisms. Moreover, essential oils could be used for the reduction of biogenic amines and as a flavoring agent in Gouda cheese, which has a positive health effect. Biogenic amines affect physiological functions such as immune response, brain activity, and gastric acid secretion. However, if biogenic amines occur in high amounts in food, they can cause health damage. Different concentrations of the essential oil from Z. multiflora Boiss (Lamiaceae) were added to milk, to test the effect of this oil on biogenic amines production such as tyramine and histamine and microbial counts in Gouda cheese. Thus, the concentration of tyramine and histamine was significantly reduced at the end of the maturation period in comparison with the control group. The higher the concentration of the essential oil, the higher was the decrease in biogenic amines and microbial counts. Z. multiflora essential oil was most effective against yeasts, but there was only a low reduction in Enterobacteriaceae counts.

Essential oils and spasmolytic activities Present studies have shown promising spasmolytic activities from Cuminum cyminum L. Generally, cumin is used as an antioxidant and flavor compound. Besides these properties, it also acts as an antiseptic, analgesic, anti-inflammatory, and sedative and is used against stomach disorders, diarrhea, and spasms. The antispasmodic activity of the hydrodistilled fruit extract of Cuminum cyminum was tested on isolated of guinea pig ileum. At the beginning, concentration-dependent responses of acetylcholine – which induces such spasms – on pig ileum were recorded and then a distinctive antispasmodic activity was shown. The anticholinergic drug atropine was used as the standard antispasmodic agent. These results proved that acetylcholine alone causes contraction of ileum, but when acetylcholine was given in combination with the hydrodistilled fruit extract, a significant decrease of contraction was noted. This high degree of spasmolytic activity was caused by blocking the cholinergic receptors.

Essential Oils and Anticancer Activities Cancer is a worldwide disease that can occur at any age and it is the second cause of mortality particularly in low-income countries. Until 2030, the cancer mortality could increase by 50% to reach 15 million worldwide. There is a relationship between the production of reactive oxygen species and the origin of oxidation and inflammatory, facts which can lead to cancer. Oxidative stress acts as a DNA-damaging agent, which

559

furnishes an increase of the mutation rate within cells and promotes oncogenetic transformation. ROS can activate signaling pathways that contribute tumor development through the regulation of cellular proliferation, angiogenesis, and metastasis. It is worth to note that many cytotoxic molecules of plant origin that possess a chemopreventive potential are used in chemotherapy, for example, thymoquinone as an active ingredient of Nigella sativa L. (Ranunculaceae). It should be considered that 30–40% of all kinds of cancer can be prevented with a healthy lifestyle and dietary measures, and it is obvious that nutrition has an impact on the cancer process. There are dietary components, which are able to promote cancer progression, and there are some that act as chemopreventive agents. N. sativa exerts a positive effect as mediating inflammation and cancerous cell growth especially to treat colon cancer. Nuclear factor-kappa B activation was decreased by thymoquinone in a dose-dependent manner. The chemotherapeutic effect of this substance is comparable with the synthetic 5-fluorouracil. The inhibiting effect of the essential oil from N. sativa on colon carcinogenesis may be linked with the suppression of cell proliferation in the colon mucosa. The essential oil of black cumin seeds is known for its cytotoxic and apoptotic/necrotic properties in a cancer cell line. An animal study has shown that the addition of N. sativa seeds to the diet is able to protect against oxidative stress, inflammatory response, and carcinogenesis. Furthermore, this study reported that the daily intake of these seeds (and thus of the volatile oil) results in the reduction of chromosomal aberrations and damaged cells. Additionally, animal models showed that thymoquinone blocked angiogenesis in vitro and in vivo, thus preventing the growth of cancerous cells. In conclusion, the anticancer and anti-inflammatory properties of N. sativa and its essential oil containing thymoquinone may be attributed partly to the suppression of NF-kappa B activation pathway. This essential oil can also be considered as a potential immunosuppressive agent. Because of its nutritional quality, N. sativa is a potential source of diet-based strategies, which may play a role in improving human health. Besides, monoterpenes that are components of nearly all essential oils and especially of citrus fruits can be used to prevent cancer. They show ideal chemopreventive properties, such as antitumor activity, commercial availability, low cost, oral bioavailability, and low toxicity. Because of this reason, monoterpene studies are proceeding in human clinical trials for chemotherapeutic activity. Several dietary monoterpenes are known for their antitumor activity. It has been shown that (þ)-limonene exerts chemopreventive activity against rodent mammary, skin, liver, lung, and fore-stomach cancers. (þ)-Limonene, the main constituent of orange oil, for example, is widely used as a flavoring agent for fruit juices, soft drinks, ice cream, and so on. The monoterpenes (þ)-limonene and perillyl alcohol and their metabolism products showed a high degree of oral bioavailability. Essential oil constituents such as citral are also known for their anticancer potential. Citral can be obtained from herbs such as lemongrass, melissa, and verbena. Citral consists of two isomeric acyclic monoterpene aldehydes, namely, geranial (trans-citral, or citral A) and neral (cis-citral, or citral B). This key component is commonly used as a food additive. Studies showed that concentrations of 44.5 mM citral induced

560

Essential Oils: Properties, Composition and Health Effects

apoptosis in cancer cell lines, a concentration that is comparable to the one of citral in a cup of tea prepared from 1 g of lemongrass. Apoptosis is linked with DNA fragmentation and induction of caspase-3 catalytic activity, whereby the abunsaturated aldehyde group is responsible for this apoptotic effect and might be a core structure for the design of proapoptotic drugs. Essential oils are also able to inhibit the formation of N-nitrosodimethylamine in vegetables. Vegetables are rich in nitrate that is a precursor of nitrite. Crops contain different concentrations of nitrate, dependent on the species, genetic, and environmental factors and cultivation conditions, whereas the concentration of nitrite in plants is commonly very low. NNitroso compounds such as nitrosamines that are formed by N-nitrosation of secondary amines with nitrite in an acidic environment act as strong carcinogens. Several foods contain the precursors of this reaction. Meats, cured meat products, and smoked fish are sources of secondary amines. Nitrite often occurs in hams and sausages as an antimicrobial or coloring substance. Essential oil constituents, such as carvone and limonene, are able to activate the glutathione S-transferase that reduces the risk of cancer. The terpene hydrocarbon (þ)limonene, which represents one of the major compounds of citrus essential oils, exhibits anticarcinogenic properties against mammary, lung, stomach, and skin cancer. It has been reported that the essential oils of citrus show an inhibitory effect on the formation of N-nitrosodimethylamine. Especially Citrus junos Siebold ex Tanaka (Rutaceae), also known as yuzu, which is a popular sour citrus fruit in Japan, shows an inhibitory effect on the formation of N-nitrosodimethylamine. Because the Japanese diet is richer in vegetables than in other countries, their intake of nitrate is very high, and thus, they have a higher risk to suffer from stomach cancer. Bacteria in saliva have an influence on the N-nitrosodimethylamine formation, because they induce the reduction of nitrate. Although the rate of formation of nitrosamines is proportional to the square of nitrite concentration, it has been considered that nitrite in vegetables especially cabbage, celery, spinach, and mint is not mainly responsible for the formation of N-nitrosodimethylamine. In most cases, oral bacteria are responsible for the reduction of nitrate into nitrite. Terpene hydrocarbons such as myrcene, a-terpinene, and terpinolene show an inhibitory effect and cause the decrease of N-nitrosodimethylamine formation. In conclusion, yuzu oil showed an inhibitory activity on the formation of N-nitrosodimethylamine in the presence of vegetables and saliva.

Essential Oils and Diabetes Essential oils from different plants can also have a positive effect on diabetes and hypertension. Nowadays, there is a tendency to manage type-2 diabetes and hypertension with natural sources. There are two major ways to handle this: the scavenging of free radicals and the inhibition of key enzymes, which are involved in starch digestion such as a-amylase and aglucosidase. By those enzymes, starch is converted into glucose, thus increasing its concentration. Inhibition of these enzymes leads to a delay of the absorption of glucose followed by a moderate postprandial blood glucose elevation. There is also a relationship between diabetes and an increased

generation of free radicals and a defective antioxidant defense system. Additionally, oxidative stress is involved in the diabetogenic process. For that reason, antioxidant-rich foods have a good dietary intervention in the management of this disease. The use of the essential oil of black pepper seeds in relation with diabetes and hypertension is known in folk medicine. Aromatic spice plants such as black pepper show good antioxidant properties, due to phenolic contents of the essential oil, which possesses antimicrobial, antihypertensive, anticonvulsive, and sedative activities. The essential oil is extracted from the seeds and leaves from black pepper and contains mono- and sesquiterpenes, besides phenolic compounds that are effective in preventing diabetes in animal models. The essential oil of black pepper is established for its radical scavenger abilities and its ferric-reducing antioxidant activity attributed to the presence of pinene and 1,8-cineole. Another therapeutic approach is the inhibition of the starch-metabolizing enzymes. In this case, the activity of a-glucosidase is more strongly inhibited by the essential oil than the one of a-amylase. This circumstance has a therapeutical purpose towards synthetic a-amylase and a-glucosidase inhibitors. The use of those natural products is important in preventing some side effects of the synthetic inhibitors. The essential oil of black pepper can inhibit also the angiotensin-1-converting enzyme (ACE) activity. This fact is a therapeutic approach in the treatment of hypertension, which is one of many complications associated with type-2 diabetes. In vitro, the ACE activity was inhibited in a concentrationdependent manner by the essential oil, because it is able to block the conversion of angiotensin-1 into the vasoconstrictor angiotensin-2 that causes the hypertension.

Essential Oils and Mood Disorders Furthermore, some essential oil constituents, such as (E)methyl isoeugenol (MIE), is often used as food flavor. This essential oil from Pimenta pseudocaryophyllus (Gomes) L.R. Landrum (Myrtaceae) leaf has calming properties. This ubiquitous food additive appears attractive for the treatment of mood disorders. Studies demonstrated the anxiolytic- and antidepressant-like activities of MIE, suggesting the participation of serotonergic pathways. Among psychiatric diseases, mood disorders are most common, that is, widespread with a prevalence of up to 20% worldwide. Because of the low remission rate and the high rate of nonresponse to current treatment, it is necessary to develop new therapeutic agents, such as consumption of functional food. With this basic food nourishment as well as health benefits are covered. Today, aromatic plants have been largely explored as functional ingredients in the pharmaceutical and food industries. The treatment of neural disorders with naturally occurring food flavors, such as MIE, seems to be more acceptable than the use of pharmacotherapies. Oral administration of MIE showed an increase in sleep duration and a soothing activity of the CNS. An antiseizure property of MIE was hypothesized too, because it acts similar to diazepam, a CNS-calming compound. This hypothesis is further supported by the antiseizure properties of aromatic compounds with a similar chemical structure as MIE, such as methyleugenol, eugenol, and 1-nitro-2-phenylethane. Besides MIE, in diets, the widely consumed essential oil of lemon is also known for its antidepressant-like effect and

Essential Oils: Properties, Composition and Health Effects therefore acts as functional food ingredient. Because of many side effects, such as nausea and anorexia, by using conventional medicaments against depression, there is a growing interest in the search of new effective antidepressants, such as functional foods that worldwide are used in the kitchen. Animal studies have been shown that the volatile oil from lemon has increased the metabolic turnover of dopamine in the hippocampus and of serotonin in the prefrontal cortex and striatum. The oral administration of lemon essential oil does not show any toxic effects or an influence on weight, blood, and organs of rodents. Upon oral administration of 400 mg kg1 lemon essential oil, the dopaminergic activity in the striatum and the hippocampus was significantly enhanced on account of an increased concentration of dopamine and a decreased turnover. It is known that depressed patients suffer from a dysfunctional 5-hydroxytryptamine (5-HT) system; thus, selective 5-HT reuptake inhibitors enhance the activation of various 5-HT receptor subtypes. In conclusion, the essential oil of lemon has a great potential to be used as an antidepressant functional food ingredient.

Essential Oils and Osteoporosis Furthermore, essential oils can also be useful in the treatment of osteoporosis. This disease is a major health problem in aging humans – especially women – when low bone mass leads to osteoporotic fractures. The bone turnover is increasing and it comes to a marked decrease in trabecular bone mineral density (BMD) and bone mineral content. Osteoporotic fractures reduce the quality of life of the patient and they are a burden to health care. Hence, it is desirable from medical and economical points of view to prevent the loss of bone mass. A nutritional approach would be an inexpensive opportunity to prevent low bone mass. Studies have shown that essential oils from sage, rosemary, and thyme and essential oils from pine, juniper, and eucalyptus inhibit or at least retard the activity of osteoclasts. The essential oil of pine is the most potent one and also the bitter orange-peel oil is very potent, for which monoterpenes, such as thujone, eucalyptol, camphor, borneol, menthol, and thymol, are responsible. Studies have shown that pine oil reduces the loss of trabecular BMD in animal models, and in vitro studies report that these monoterpenes directly inhibit the osteoclast activity. (1R,3R,4S)-()Menthol that is the most widely used monoterpenic alcohol in human nutrition inhibits bone resorption in vivo and in vitro, but the latter effect is weak. Also, the consumption of monoterpene-rich fruits and vegetables is linked with a greater BMD in humans. Animal studies have shown that vegetables, salads, and herbs, as a part of human nutrition, are able to inhibit bone resorption. Essential oils exert a positive effect on the bone metabolism when they are added to food. Therefore, essential oil-containing herbs are possible candidates for a dietary approach to osteoporosis.

Essential Oils and Their Effect on the Gastrointestinal Tract Essential oils are also famous for their positive effect on the gastrointestinal tract. Mentha x piperita L., which belongs to the Lamiaceae family, contains a volatile oil with the major compound ()-menthol. This essential oil is used against irritable

561

bowel syndrome, because of its influence on gastric motility. Peppermint tea is commonly used against acute and chronic gastritis, enteritis, and disturbances of the GI tract by the antispasmodic effect of menthol. Furthermore, peppermint tea is commonly used as an antiemetic in pregnant women, besides ginger and cannabis. This volatile oil shows the same effect as 5-HT3 receptor antagonists, 5-HT4 receptor agonists, and anticholinergics. Those receptors are also involved in the pathogenesis of irritable bowel syndrome. Peppermint oil and ()menthol are able to completely inhibit 5-HT3 receptor in a concentration-dependent manner by relaxing the smooth muscle and antagonizing the serotonin-induced stimulation. This antagonistic effect on the 5-HT3 receptor channel influences in a positive manner the disturbed motility during symptoms of irritable bowel syndrome and emesis. The cationic influx through 5-HT3 receptor channels was inhibited by peppermint oil and its main constituent ()-menthol. The influence of peppermint essential oil and ()-menthol leads to the reduction of serotonin-induced contraction in animal models. In addition, they possess a direct relaxant effect. Essential oils often are used as food additives to complement present therapies of gastritis and peptic ulcers. This disease is caused by an increased density of Helicobacter pylori in the gastric mucosa. A nutritional approach can help people with asymptomatic gastritis to manage the infection and to decrease the development of the disease. Those essential oils that exhibit the strongest bactericidal potential against H. pylori P1 were also active against other Helicobacter strains. The most effective single essential oils constituent against H. pylori are carvacrol, isoeugenol, nerol, citral, and sabinene. The widespread infection with H. pylori is well known as the major etiologic factor in chronic active type B gastritis, gastric ulcers, and gastric cancer. Nowadays, a triple therapy such as the combination of proton pump inhibitor and two antibiotics is common, but increasing antibiotic resistance is also responsible for an eradication treatment failure. Therefore, there is a growing interest to find alternative treatments, for example, the development of new nutritional approaches. Especially patients with an asymptomatic gastritis profit from this nutritional approach.

Essential Oils and Alzheimer’s Disease Essential oils can also be used in the treatment of Alzheimer’s disease, the most common cause of dementia in the aged population. It results in cognitive decline and mental deterioration that is the result of massive and progressive loss of neurons from different regions of the brain. Disorders of the CNS are linked with neurotransmitter disturbances and insufficiencies in cholinergic functions. Cholinesterase is responsible for the hydrolyzation of choline, and if there is a high concentration of this enzyme, then a lower concentration of choline in the synaptic gap results. Investigations have shown that there is a relation between increased levels of cholinesterase enzymes and Alzheimer’s disease. This fact leads to the hypothesis that this cognitive decline in patients is linked to the progressive cholinergic degeneration. Hence, promising approaches for the treatment of Alzheimer’s disease are cholinesterase inhibitors that are able to enhance the level of cholinergic neurotransmitters in the brain. Acetylcholinesterase (AChE) and

562

Essential Oils: Properties, Composition and Health Effects

butyrylcholinesterase (BChE) are the major cholinesterase enzymes that are responsible for the decreasing choline levels. Today, Alzheimer’s disease patients are treated with AChE inhibitors, synthetic compounds that reveal some toxicity during prolonged use. Hence, there is a great demand for alternative drugs to treat Alzheimer’s disease. It has been suggested that dietary supplements with antioxidants and free radical scavengers may decrease the mild cognitive impairment of Alzheimer’s disease. A high amount of plants has been traditionally used for the enhancement of cognitive function and alleviation of several symptoms linked with Alzheimer’s disease. Nowadays, essential oils of edible plants are considered to inhibit AChE and BChE. Those volatile oils are of great interest, because of their availability, few side effects, low toxicity, and their biodegradability. Studies have shown that the essential oils of Citrus aurantifolia (Christm.) Swingle (Rutaceae), Alpinia galanga (L.) Willd. (Zingiberaceae), and Melissa officinalis L. (Lamiaceae) possess the highest inhibitory activity against AChE, whereas Melissa officinalis, Citrus aurantifolia, and Ocimum gratissimum L. (Lamiaceae) exhibited the strongest inhibitory properties against BChE. Therefore, the leaf essential oil of Citrus aurantifolia with the constituents (þ)-limonene, ()camphor, citronellol, cis-ocimene, and 1,8-cineole is a promise for the prevention and treatment of Alzheimer’s disease, because it inhibits both enzymes. 1,8-Cineole that is one of the most potent AChE inhibitors also occurs in other plants such as Rosmarinus officinalis L. (Lamiaceae).

Essential oils and atherosclerosis The essential oil of sage could be used as food ingredient with anti-inflammatory and antiatherogenic properties. Cholesterol deposition in the intima of large- and medium-size arteries together with a chronic inflammatory process leads to atherosclerosis. Oxidized low-density lipoproteins play a key role in early inflammation. This type of LDL that is not recognized by the LDL receptor is taken up by the scavenger receptors in monocytes–macrophages and endothelial cells. Hence, it comes to the accumulation of cholesterol in macrophages and to the formation of foaming cells. Moreover, oxidized LDL is able to induce the expression of proinflammatory cytokines such as TNF-a, Il-1b, and IL-6 in macrophages and endothelial cells. Essential oils from spices, which are added to food, present various biological activities, such as antioxidant and anti-inflammatory properties. As mentioned earlier, the oxidation of LDL results in this atherosclerotic lesion and the initiation of the inflammatory cascade. Therefore, oxidized LDL is able to induce the expression of proinflammatory cytokines, such as IL-1b, TNF-a, and IL-6 from monocytes and macrophages. IL-1b and TNF-a play an important role in the initial amplification of the inflammatory response. Moreover, they are able to induce the expression of adhesion molecules by endothelial cells and promote secretion of different cytokines and chemokines by monocytes. Furthermore, they are involved in the process of foam cell formation, especially by means of induction of growth factors. Activated macrophages, lymphocytes, fibroblasts, and vascular smooth muscle cells produce

during stimulation IL-1 and TNF-a. The inflammatory cytokine IL-6 plays an important role in the development and progression of atherosclerosis. It has been reported that the essential oil of Salvia officinalis L. (Lamiaceae) is able to inhibit COX-2, thus exerting anti-inflammatory properties. Also, supercritical extracts of Salvia officinalis are effective inhibitors of oxidized LDL and induced proinflammatory cytokines. This sage extract mainly consists of camphor, borneol, and 1,8-cineole that is a strong inhibitor of TNF-a, IL-1b, and IL-6 production. Borneol and camphor also exhibit excellent anti-inflammatory properties. Therefore, sage extract has a great potential to be used as an anti-inflammatory agent to prevent atherosclerosis.

See also: Essential Oils: Isolation, Production and Uses; Herbs: Composition and Dietary Importance.

Further Reading Alimohammadi M, Es’haghi Gorji M, Jahed Khaniki G, Nabizadeh Nodehi R, Noori N, and Rastkari N (2014) The evaluation of Zataria multiflora Boiss. Essential oil effect on biogenic amines formation and microbiological profile in Gouda cheese. Letters in Applied Microbiology 59(6): 621–630. Arranza E, Jaimea L, Lopez de la Hazasa MC, Regleroa G, Santoyoa S, and Vicentea G (2014) Supercritical sage extracts as anti-inflammatory food ingredients. Industrial Crops and Products 54: 159–166. Barzegar M, Kordsardouei H, and Sahari MA (2013) Application of Zataria multiflora Boiss. and Cinnamon zeylanicum essential oils as two natural preservatives in cake. Avicenna Journal of Phytomedicine 3(3): 238–247. Benjakul S and Tongnuanchan P (2014) Essential oils: extraction, bioactivities, and their uses for food preservation. Journal of Food Science 79(7): 1231–1249. Catherine AA, Deepika H, and Negi PS (2012) Antibacterial activity of eugenol and peppermint oil in model food systems. Journal of Essential Oil Research 24(5): 481–486. Crowell PL (1999) Prevention and therapy of cancer by dietary monoterpenes. The Journal of Nutrition 129(3): 775–778. Edward K, Monika S, and Magorzata W (2012) Recent patents regarding essential oils and the significance of their constituents in human health and treatment. Recent Patents on Anti-Infective Drug Discovery 7(2): 133–140. Fajemiroye JO, Galdino PM, Paula JAMD, Rocha FF, and Akanmu MA (2014) Anxiolytic and antidepressant like effects of natural food flavour (E)-methyl isoeugenol. Food & Function 5(8): 1819–1828. Felix R, Lozano A, Mu¨hlbauer RC, Palacio S, and Reinli A (2003) Common herbs, essential oils, and monoterpenes potently modulate bone metabolism. Bone 32(4): 372–380. Ferreira AR, Mata AT, Nogueira JMF, et al. (2007) Antioxidant and antiacetylcholinesterase activities of five plants used as Portuguese food spices. Food Chemistry 103(3): 778–786. Hauk F, Heimes K, and Verspohl EJ (2011) Mode of action of peppermint oil and ()menthol with respect to 5-HT3 receptor subtypes: binding studies, cation uptake by receptor channels and contraction of isolated rat ileum. Phytotherapy Research 25(5): 702–708. Muhlbauer RC (2006) Are vegetables, salads, herbs, mushrooms, fruits and red wine residue that inhibit bone resorption in the rat a promise of osteoporosis prevention? Current Nutrition and Food Science 2(1): 69–78. Nagori BP, Saini N, and Singh GK (2014) Physicochemical characterization and spasmolytic activity of essential oil of cumin. International Journal of Biology, Pharmacy and Allied Sciences 3(1): 78–87. Sadiq BM and Tauseef SM (2010) Nigella sativa: reduces the risk of various maladies. Critical Reviews in Food Science and Nutrition 50(7): 654–665. Wallner, I. M. (2015). Essential oils: properties, composition and health effects. Master Thesis, University of Vienna.

Ethnic Foods OI Bermudez, Tufts University School of Medicine, Boston, MA, USA ã 2016 Elsevier Ltd. All rights reserved.

Introduction Consumers of ethnic foods often have distinctive cultural characteristics, including language, social norms, and behaviors that provide them with a sense of connectedness. From a cultural standpoint, ethnic foods serve as the familiar link with the past and help those who are accustomed to them to maintain and promote among peers their ethnic identity. In countries like the United States, with a very diverse ethnic population, the production, distribution, and consumption of ethnic foods are an integral feature of the overall American culture. Native Americans are direct descendants from the original population groups distributed throughout the country before the European colonization. Since then, they have contributed to the cultural dimensions of ‘ethnic foods’ with their food production developed around the ‘three sisters’ (corn, beans, and squash) system. Moreover, they developed a rich and highly nutritive traditional cuisine with many similitudes to those found in Central America. For example, in Guatemala, the ‘three sisters’ system was termed as the ‘milpa.’ During and after the colonization, several waves of immigrants from Europe, Africa, and Asia brought from their countries of origin their ethnic foods and dishes. Then, their cuisine evolved from the homeland traditional dishes to the modified versions of ethnic cuisines that currently exist in these modern times. Similarly, more recent waves of immigrants from Central and South America, the Caribbean, and the four corners of the globe contributed to the rich diversity of the foods present today in US households. Several cooking techniques and eating styles from the original traditions brought by original, earlier, and new American settlers were exchanged and adopted by them. Traditional dishes like tacos, tamales, pizza, and noodles and Chinese dishes (e.g., lo mein) gave origin to new dishes, different from their native, original versions. Those ethnic food transformations are likely to continue as the American population becomes more diverse and interrelated, with evolving sense of taste, flavor, and smell for what they term as ‘ethnic foods.’

Ethnic Diversity of the American Population The United States has a very diverse population. This diversity was long in existence in the precolonial times with more than 300 tribal groups of Native Americans. Since then, the United States had transformed into a ‘plurality nation.’ This pluralism is reflected in the ancestry of the early settlers in the country, which are those that contributed to the ethnic diversity of the American cuisine. In the current times, those from German (16%), Irish (11%), English (8%), and Italian (6%) descents are the largest ethnic groups that contributed to the diversity of the white population (see Figure 1). The Census Bureau

Encyclopedia of Food and Health

defines as white those having origins in any of the original peoples of Europe, the Middle East, or North Africa. Furthermore, it is the plurality of the more recent immigrants, along with that from the African-American population that has given this country the ethnic diversity observed in the early periods of the twenty-first century. According to the 2010 Census Bureau, the total population reached about 309 million people. The racial and ethnic heterogeneity of this population is represented in Table 1. As seen in this Table 1, whites were the largest group (75%). Other ethnic groups include black or African-Americans (14%), Asians (6%), AmericanIndian and Alaska Natives (2%), and Native Hawaiian and other Pacific Islanders (0.4%). Based on ethnicity, which is reported by the Census Bureau as having or not Latino or Hispanic background, 16% of the total US population is of Latino or Hispanic origin. This diversity of the major racial and ethnic groups in the United States is reflected in the large number of ethnic groups that, because of their geographic, socioeconomic, and cultural origins, have contributed to the rich, mixed, and festive ethnic foods that every day are present in the American households.

Ethnic Foods Present in the American Homes Based on the geographic, climate, and, particularly, the plurality of the American population, one could find an extraordinarily rich ethnic cuisine in most households across the country. Most ethnic cuisines have been influenced by centuries of food exchanges between indigenous groups and immigrants from all over the world. As result, the American cuisine is a unique multiethnic blend. No single eating pattern has prevailed over all the others, and no ethnic or cultural group has emerged with an unchanged food pattern. However, some ethnic cuisines are more popular than others, mainly due to the high number of immigrants from the latitudes where those ethnic foods and traditions originated. In general, foods that are regarded as ethnic foods by a cultural group vary over time and with the degree of acculturation of the group consuming them. Even within major race and ethnic subgroups, many variations in ethnic foods exist. Some foods only bear a passing resemblance to the original cuisine from which they were derived. However, all of them fulfill important cultural roles for the diverse American population. For example, in the United States, pizza, tacos, chili with meat, and chicken chow mien are very different from the Italian, Mexican, and Chinese dishes that inspired them. Hispanic-Americans with a Mexican-American heritage differ in their food preferences from Puerto Ricans, Colombians, and Chileans, whose eating patterns involve a blend of Spanish, African, Native Central/South American, and continental North American influences. Similarly, Asian-Americans of Filipino, Japanese, or Chinese ancestry have very different food choices and traditions. Among the over 300 different Native

http://dx.doi.org/10.1016/B978-0-12-384947-2.00263-4

563

564

Ethnic Foods

German, 15.5

Irish, 11.2

Other Groups, 38.9

English, 8.4

American, 6.5

Russian, 1.0 Scotch-Irish, 1.1 Swedish, 1.3 Norwegian, 1.4

Italian, 5.6 Dutch, 1.5

Scottish, 1.8

French, 2.8

Polish, 3.1

Figure 1 Ancestry of the American population, based on the 2010 US Census data. Reproduced from US Census Bureau. (2010). 2010 American Community Survey. American Fact Finder: People Reporting Ancestry.

Table 1

Racial and ethnic population groups in the United States

Race and Latino or Hispanic origin

Number

Percenta

Total population Race categories White Black or African-American Asian American Indian and Alaska native Native Hawaiian and other Pacific Islander Some other race Latino or Hispanic ethnicity (from any race categories)

308 745 538

100.0

231 040 398 42 020 743 17 320 856 5 220 579 1 225 195 21 748 084 50 477 594

74.8 13.6 5.6 1.7 0.4 7.0 16.3

Table 2 Examples of ethnic cuisines and representatives dishes commonly eaten by the American population Ethnic cuisine Puerto Rican Mexican Italian Japanese

Examples of popular dishes Rice with pigeon peas, tostones (fried green plantains), habichuelas (stewed pink beans), pasteles (similar to tamales), and rice pudding Tamales, tortillas (corn- or wheat-based), tacos, enchiladas, guacamole, and mole poblano Pizza, risotto, spaghetti, lasagna, ciabatta bread, cheeses, and wines Sashimi, sushi, soba, sukiyaki, yakitori, and miso soup

a

Percentages estimated from the total population. Source: US Census Bureau. (2010). 2010 Census Briefs. Overview of Race and Hispanic Origin: 2010.

American groups, the staples and other foods that they regard as unique and distinctive to their cultures also vary. The traditional ‘Mediterranean’ diet varies in terms of foods and seasonings included in their cuisine. These differences are reflecting the European and North African Mediterranean locations and cultural norms of the Mediterranean countries where the Mediterranean diets originated. These traditional diets include liberal amounts of fruits, vegetables, legumes, grains, and fish. High amounts of monounsaturated compared with saturated fats, low amounts of meats, and moderate consumption of alcohol are also observed. Although there is a debate about the extent to which such a regimen is exportable, some ‘Mediterranean’ ethnic foods have become very popular in America. Examples of those foods include olives and olive oil, breads, couscous, pastas, artichokes, arugula, grapes, dates, figs, almonds, chickpeas, basil, bay leaf, and parsley. The ethnically diverse cuisines found in the United States include the contributions of most ethnic groups. Those groups

brought with them their native recipes and some of their traditional food ingredients. And then, throughout the times, they modified and adjusted those recipes to the availability and cost of foods and seasonings in order to make them acceptable by the general population. Table 2 contains a short list of examples of ethnic cuisines present in the US households. Germans, for example, brought and modified hot dogs and hamburgers. Italians, with their basic ingredients wheat, tomato sauce, and olive oil, also influenced the American foodways. They introduced classic dishes such as pizza and spaghetti, to the point that the Italian cuisine is the third most popular in the country. Ethnic cuisines of Asian origins, particularly represented by Chinese and Japanese ethnic dishes, adapted to the American palate. Chinese-American foods represent the second most popular ethnic cuisine in the United States. Those ethnic cuisines contributed with several unique ethnic dishes. Noodles-based dishes and a large variety of vegetable dishes (e.g., broccoli and tomatoes) are part of the incorporation of non-Chinese foods added to satisfy the American preferences.

Ethnic Foods Table 3

565

Useful sources to consult on ethnic cuisines, ethnic foods, and food behaviors

Description of sources Scientific journals Journal of the Academy of Nutrition and Dietetics Nutrition Today American Journal of Clinical Nutrition Journal of the American College of Nutrition Journal of Nutrition Appetite Public Health Nutrition Journal of Food Composition and Analysis Ethnicity and Disease British Food Journal Proceedings of the Nutrition Society Organizations, programs, and other resources Dietary Guidelines for Americans US Food Classification: MyPlate USDA Cultural and ethnic food and nutrition resources Society for the Study of Ingestive Behavior American Institute of Wine & Food Center for Food Safety Center for Science in the Public Interest Oldways Preservation Trust Sally’s Place Slow Food USA UN Food and Agriculture Organization (FAO) FAO food-based dietary guidelines

From Latin America, Mexicans contributed a large variety of ethnic foods. An extensive list of Mexican foods, including chili peppers, chocolate, black beans, tacos, ‘burritos,’ and tamales, is part of the rich offerings found across the country. Currently, the Mexican cuisine is the most popular ethnic food category in the country. Because the large Mexican population in the United States, along with their rich and varied traditional cuisine, the general American population has incorporated many of those Mexican ethnic foods into their eating habits, diversifying the culinary landscape of ‘American’ ethnic foods. Ethnic food preferences and choices are important aspects of the cultural heritage they represent. Some of the many sources available on food-related aspects of culture are summarized in Table 3. It includes the cultural anthropology literature on foodways, reports, and nutrition information materials that describe foods, their uses, and related traditions.

Ethnic Differences in Food Classification Systems Ethnic groups vary in their foodways, social and physical contexts in which food preparation, serving, and eating occur. Table 4 describes the factors that are thought to be involved in the development of ethnically distinctive cuisines and ethnic foods. Nutritional anthropologists subscribe to various theories on how ethnic food patterns have developed (see Table 5). No one theory appears to predominate over the other. Individuals in different ethnic groups have different ways of categorizing foods, which are sometimes at variance with

Electronic sites http://www.andjrnl.org/ http://journals.lww.com/nutritiontodayonline/pages/default.aspx http://ajcn.nutrition.org/ http://americancollegeofnutrition.org/content/the-journal http://jn.nutrition.org/ http://www.journals.elsevier.com/appetite/ http://journals.cambridge.org/action/displayJournal?jid¼PHN http://www.journals.elsevier.com/journal-of-food-composition-and-analysis/ http://www.ishib.org/wordpress/?page_id¼1388 http://www.emeraldinsight.com/journal/bfj http://journals.cambridge.org/action/displayJournal?jid¼PNS http://www.health.gov/dietaryguidelines/ http://www.choosemyplate.gov/food-groups/ http://www.nal.usda.gov/fnic/pubs/ethnic.pdf http://www.ssib.org/web/ http://www.aiwf.org/ http://www.centerforfoodsafety.org/about-us http://www.cspinet.org/ http://oldwayspt.org/ http://www.sallybernstein.com/ http://www.slowfoodusa.org/ http://www.fao.org/home/en/ http://www.fao.org/nutrition/nutrition-education/food-dietary-guidelines/en/

Table 4 cuisines

Factors involved in the development of ethnically distinctive

Factors

Examples

Ecological factors Individual factors Social factors

Geography, climate, range of native plants and unusual species in area Philosophical, religious, and moral attitudes

Economic factors

Table 5

Social structures, social relationships, social status, migration patterns (both voluntary and involuntary) Trade, food prices, food distribution, imports, and exports

Theories on how ethnic food patterns developed

Theory

Comments

Environmentalism

Ethnic food traditions are determined primarily by environmental and external influences Humans mold their environments and diets quite independently of their physical surroundings Both culture and environment are important Historical influences on how foods have evolved and developed over time Use of foods to identify or please people socially and how foods are used in society

Cultural determinism Cultural ecology Cultural history Functionalism

medical views of foods as sources of nutrients that are based on physiological and biochemical characteristics. They may also differ in their attitudes toward medical care and their health behaviors and in their beliefs about the therapeutic

566 Table 6

Ethnic Foods Some ways in which foods are classified in different cultures

Food classification system Food habits Gastronomic Religious/ philosophical Opposing categories or properties Medicinal Social Emotional significance Nutritional Hygienic

Descriptions ‘Core,’ secondary core, and peripheral foods Foods versus nonfoods, edible versus nonedible Sacred versus profane, allowed versus forbidden Hot versus cold, yin versus yang, or other ways of ‘balancing food intakes to prevent or cure disease’ Foods for the prevention of diseases or as treatment for certain diseases or psychological states Symbolic of a culture or belief system Foods for the sick, party foods Macro- and micronutrients Wholesomeness, cleanliness, freedom from environmental contaminants, microorganisms, or other undesirable elements

Source: Kaufman-Kurzrock, D. L. (1989). Cultural aspects of nutrition. Topics in Clinical Nutrition 4(2), 1–6.

and preventive properties of foods and the appropriate balance between different foods. For example, there are systems in various cultures that categorize foods into medicinal versus nonmedicinal, hot versus cold, or yin versus yang. These are classification schemes based on properties other than nutrient content (see Table 6). Food classification systems may affect nutritional status because they cause the exclusion of healthful foods or food groups from the diet, encourage consumption of harmful items, or produce a false sense of security among eaters that all of their nutritional needs are met when in fact they are not.

Dietary Assessment of Ethnic Food Consumption Patterns Familiarity with ethnic foods is important for dietary assessment, particularly if they are imports from other countries. The foods themselves, their names, composition, portion sizes, and frequency of consumption vary. Accurate assessment of nutrient intakes depends on a good knowledge of these factors. Some ethnic foods may be especially rich sources of nutrients or other food constituents, such as phytochemicals. During the last 40 years, Americans experienced increases in a series of nutritional and health threats associated with unhealthy eating practices, including obesity, type 2 diabetes, hypertension, cardiovascular diseases, and certain types of cancers. To halt those detrimental effects, people, in general, are responding favorably to the new policies and recommendations to improve eating and health-related practices. For some groups, those recommendations are being implemented with the use of ethnic foods considered healthy alternatives. To confirm those beliefs, food and nutrition experts need to analyze the growing number of ethnic foods that are being

incorporated into the already diversified, ethnic cuisine practiced by Americans. The application of tools and techniques for dietary assessment may be needed in order to provide appropriate nutritional guidance for those groups adapting new ethnic foods. One challenge in evaluating diets that include many ethnic foods is the appropriate choice of methods for dietary assessment. Methods such as semiquantitative food frequency questionnaires are often designed to capture intakes of important nutrients for those who conform to the predominant food culture. These are often inappropriate for subgroups with very different dietary patterns. Subgroup members may call the foods by different names, consume items not included on the food lists, or vary in amount and frequency of consumption. The use of such instruments to assess intakes of individuals with unconventional dietary preferences can lead to highly erroneous conclusions. Subgroup-specific food frequency questionnaires that take account of these factors are sometimes available, and when they are, these should be used. For example, well-validated ethnic-specific food frequency questionnaires have been developed for Japanese and Chinese-Americans living in Hawaii and for Puerto Ricans living in the Northeast. Another potential problem in the dietary assessment of ethnic cuisines is that nutrient databases for converting dietary intakes into nutrients for ethnic foods may not be available. In research studies on individuals who use many ethnic foods, nutrient databases that include entries for ethnic favorites are needed to obtain complete information about intakes. Whenever possible, food tables that are specific to the country or ethnic group should be used. Country-specific tables of food composition and nutrient databases are becoming increasingly available, and these are the best option. For individuals who eat largely ethnic cuisines in countries where the parent food culture is different, special food composition tables for the ethnic foods that are most popular in mainstream diets are usually unavailable. Even for ethnic foods that are included in nutrient databases, food composition information may be incomplete, inappropriate, or lacking. Moreover, nutrient values for foods that are rarely consumed outside of the ethnic group may be absent, especially if the ethnic group is small. This is a challenge with many local foods consumed by Alaskan natives for which there is a gap in nutrient composition of those foods. The dietary assessment task is more difficult when no country or ethnic-specific food tables are available. Recipes made from locally available ingredients or imported foods may differ in their food composition from recipe values in food tables used in Western countries. For example, a typical curry dinner as served in India, Ghana, and England varies greatly in its nutrient composition. For this reason, country-specific food consumption tables are needed.

Ethnically Appropriate Dietary Planning and Interventions The consumption of ethnic foods is usually interpreted as a cultural phenomenon. However, it has physiological implications. Sometimes, alterations in eating habits are necessary to achieve nutritional needs, but this is a difficult goal to achieve. Food habits are among the oldest and most deeply entrenched aspects of many cultures, and they cannot be easily changed.

Ethnic Foods Table 7

567

Examples of the importance of ethnically appropriate nutrition interventions

Resources for nutrition intervention

Electronic sites

National Heart, Lung, and Blood Institute (NHLBI): Materials for American Indian/ Alaska Native Audiences NHLBI: American Indian Health NHLBI: Materials for Asian American/Native Hawaiian/Other Pacific Islander NHLBI: Materials for Latino/Hispanic Americans NHLBI: On the Move to Better Heart Health for African Americans

http://www.nhlbi.nih.gov/health/healthdisp/an.htm

USDA: Cultural and ethnic food and nutrition resources US Department of Health and Human Services. Office of Disease Prevention and Health Promotion: Dietary Guidelines for Americans US Department of Agriculture. US Food Classification: MyPlate Oldways Preservation Trust: Inspiring good health through cultural food traditions and lifestyles Illinois State University: Healthy habits at the holidays Mayo Clinic. Mediterranean diet: a heart-healthy eating plan Winham, D.M., 2009. Culturally tailored foods and CVD prevention. American Journal of Lifestyle Medicine 3, 64S Lemacks, J. et al., 2013. Interventions for improving nutrition and physical activity behaviors in adult African American populations. Preventing Chronic Disease 10, 120256

When they are changed, they often produce unexpected and unwelcome reactions. There is no simple formula for success, but ethnic specificity and cultural sensitivity are important ingredients in crafting successful nutrition planning and interventions. Table 7 provides some examples from the recent literature of ethnically appropriate nutrition interventions. The expertise of nutritional anthropologists who specialize in cultural anthropology as it relates to food and foodways is helpful. Some general principles that must be kept in mind are discussed in the succeeding text. Some foods are more important than others to eaters and are the base of their food systems. Those foods constitute a continuum or gradient that consists of the ‘core diet,’ which includes universal, regular staple foods, the ‘secondary core’ of foods that are used widely but not universally, and ‘peripheral’ foods that are used infrequently or during specific holidays or celebrations. Such groupings are useful in planning nutritional interventions since they often suggest opportunities for making acceptable changes and possible pitfalls that need to be avoided. Food habits are always difficult to change. The most resistant habits of all involve foods in the core diet, especially those that are also viewed by the eaters as ethnically appropriate and vital to group cohesion. The language used about foods and the names of favorite foods themselves must be used in recruitment. Ethnic-specific nutrition information, recommendations, dietary guidance, and education are also vital if interventions are to be appropriate to the dietary patterns and culture of intended recipients. In many countries, ethnicity, socioeconomic status, educational level, geographic location, and religious and ethical views are often confounded. As these other factors vary, so will the appropriateness of the intervention.

http://americanindianhealth.nlm.nih.gov/index.html http://www.nhlbi.nih.gov/health/healthdisp/aapi.htm http://www.nhlbi.nih.gov/health/healthdisp/lat.htm http://www.nhlbi.nih.gov/health/resources/heart/africanamerican-index http://www.nal.usda.gov/fnic/pubs/ethnic.pdf http://www.health.gov/dietaryguidelines/ http://www.choosemyplate.gov/food-groups/ http://oldwayspt.org/ http://stories.illinoisstate.edu/student-affairs/healthpromotion-and-wellness/healthy-habits-holidays-2/ http://www.mayoclinic.org/healthy-living/nutrition-andhealthy-eating/in-depth/mediterranean-diet/art-20047801 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2782861/ http://www.cdc.gov/pcd/issues/2013/12_0256.htm

The importance of including ethnic foods in determining the ultimate acceptance of nutrition interventions varies within and between ethnic groups. In general, those who are least acculturated to the larger mainstream society, those with language barriers, the ill, the geographically isolated, the economically disadvantaged, traditionalists, and the religiously conservative may regard ethnic foods as particularly important. Those who are innovators and desire acceptance in the larger culture may regard ethnic foods as less important. The degree to which new foods are incorporated into eating styles varies depending on contact between cultures, food availability, trade, and social status of the group introducing the food and philosophical, religious, and individual factors. Immigrants differ in the extent to which they wish to remain unassimilated in their foodways and true to their native cuisines. Ethnic borrowing and mingling rather than rigid preservation of unique eating traditions often occur. Ethnic-specific food-based dietary guidelines and graphics to guide food choices present nutritionally appropriate food patterns in a culturally acceptable manner for many ethnic groups. Many guidelines are currently available. They have the advantage of including ethnic foods as a part of overall diets. These guidelines are helpful as starting points for developing population-based nutrition education materials for healthy persons. Graphics that depict appropriate food choices have the additional advantage of being understandable even to those who may be illiterate. Special attention must be paid to the ethnic appropriateness of nutritional counseling for those who are ill or in especially vulnerable groups from the physiological standpoint. Therapeutic diet manuals are sometimes available or obtainable that can be helpful in designing diets that provide medical nutrition therapy that is culturally appropriate.

568

Ethnic Foods

Additional useful techniques for crafting nutrition information, education, and counseling interventions include the use of ethnic and regional terms for foods and recipes. Interventions should be adapted to be culturally appropriate. Food preparation and cooking demonstrations can be used to show how to incorporate recommendations into existing food patterns. Nutrition education and counseling should include attention to ethnic issues. Community social structures in the ethnic group are helpful to make contacts with the target ethnic groups. Those who develop nutritional interventions that involve food enrichment, fortification, or special supplements must also pay attention to the role of ethnic foods and food patterns in the diet. In many cultures, the target groups for such efforts are those who are nutritionally vulnerable because of their age, sex, and socioeconomic status. And they are likely to have eating practices that are deeply embedded in their culture. Nutrition counselors must remember that care of the sick is heavily intertwined with food and health beliefs, social relationships, and acculturation. In summary, nutrition workers must avoid cultural biases and ethnocentricity in dealing with their clients in both dietary assessment and planning.

See also: Amaranth; Authenticity of Food; Bananas and Plantains; Cashew Nuts; Cassava: The Nature and Uses; Date Palm: A Wealth of Healthy Food; Food–Herbal Medicine Interface; Legumes in the Diet; Quinoa.

Further Reading Administration on Aging and US Department of Health and Human Services, 2005. Guidelines for Culturally and/or Linguistically Competent Agencies. Backstrom A, Pirttila-Backman A-M, and Tuorila H (2003) Dimensions of novelty: a social representation approach to new foods. Appetite 40(3): 299–307. Bermudez OI and Tucker KL (2003) Trends in dietary patterns of Latin American populations. Cadernos de Sau´de Pu´blica 19(Suppl. 1): S87–S99. Bermudez OI and Tucker KL (2004) Cultural aspects of food choices in various communities of elders. Generations (Journal of the American Society on Aging) 28(3): 22–27.

Broussard-Marin L and Hynak-Hankinson MT (1989) Ethnic food: the use of Cajun cuisine as a model. Journal of the American Dietetic Association 89(8): 1117–1121. Kaufman-Kurzrock DL (1989) Cultural aspects of nutrition. Topics in Clinical Nutrition 4(2): 1–6. Khan LK and Martorell R (1997) Diet diversity in Mexican Americans, Cuban Americans and Puerto Ricans. Ecology of Food and Nutrition 36: 401–415. Koehler KM (1989) Core, secondary, and peripheral foods in the diets of Hispanic, Navajo, and Jemez Indian children. Journal of the American Dietetic Association 89(4): 538–540. Lee JH, Hwang J, and Mustapha A (2014) Popular ethnic foods in the United States: a historical and safety perspective. Comprehensive Reviews in Food Science and Food Safety 13(1): 2–17. Lemacks J, Wells BA, Ilich JZ, and Ralston PA (2013) Interventions for improving nutrition and physical activity behaviors in adult African American populations. Preventing Chronic Disease 10: 120256. Lin H, Bermudez OI, and Tucker KL (2003) Dietary patterns of Hispanic elders are associated with acculturation and obesity. Journal of Nutrition 133(11): 3651–3657. Nestle M (1995) Mediterranean diets: historical and research overview. American Society for Nutrition 61: 135–205. Sanjur D (1995) Hispanic foodways, nutrition, and health. Boston: Allyn and Bacon.

Relevant Websites http://www.nal.usda.gov/fnic/pubs/ethnic.pdf – Food and Nutrition Information Center, & USDA. http://stories.illinoisstate.edu/student-affairs/health-promotion-and-wellness/healthyhabits-holidays-2/ – Illinois State University. http://www.mayoclinic.org/healthy-living/nutrition-and-healthy-eating/in-depth/ mediterranean-diet/art-20047801 – Mayo Clinic. http://americanindianhealth.nlm.nih.gov/index.html – National Heart Lung and Blood Institute. http://www.nhlbi.nih.gov/health/healthdisp/an.htm – National Heart Lung and Blood Institute. http://www.nhlbi.nih.gov/health/healthdisp/aapi.htm – National Heart Lung and Blood Institute. http://www.nhlbi.nih.gov/health/healthdisp/lat.htm – National Heart Lung and Blood Institute. http://www.nhlbi.nih.gov/health/resources/heart/african-american-index – National Heart Lung and Blood Institute. http://oldwayspt.org/ – Oldways Preservation Trust. http://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml? pid¼ACS_10_1YR_B04006&prodType¼table – U.S. Census Bureau. http://www.census.gov/prod/cen2010/briefs/c2010br-02.pdf – U.S. Census Bureau.

Extrusion Cooking: Chemical and Nutritional Changes JAG Areˆas, CM Rocha-Olivieri, and MR Marques, Universidade de Sa˜o Paulo, Sa˜o Paulo, Brazil ã 2016 Elsevier Ltd. All rights reserved.

Introduction The Extrusion Cooking of Foods Extrusion cooking has become one of the most prominent technologies for thermal processing of food for a number of reasons. It was adapted from the plastic industry and first used for food processing in the 1920s to cook and model cereals in several forms. It evolved in the 1970s to the high-shear, highpressure, high-temperature machines, which has become ever since the major process used throughout the food industry to generate expanded ready-to-eat products and textured vegetable protein. In the 1990s, modulated double-screw extruders were developed, and more feed materials with distinct water content could be processed. However, the most successful configuration for extrusion cooking is still the one that through a high shear at high temperature and pressure in a single screw creates the final texture by orienting and cross-linking the food macromolecules in a tridimensional stable structure, generating a wide variety of food products. In this extruder configuration, the feed material, rich in carbohydrates and proteins and with a low moisture content, is conveyed through an Archimedes screw within a barrel forming a molten mass that is expelled through a die. During this process, the molten mass experiences elevated shear, temperature, and pressure, with resident times at high temperatures of 180
C) and rotation (>100 rpm) in combination with low moisture (20% w/w) they are unstable and tend to gels, which are plastic, and spreadable, which provides shortening-like texture. Potato maltodextrin is produced by solubilizing the slurry by jet cooking and treating it with enzymes until the required degree of hydrolysis is achieved (Paselli SA2, C* Pur 01906). Potato starch contains amylose molecules with longer chains as compared to corn and wheat, which retrograde less readily, thereby reducing the tendency to cause turbidity and an undesirable texture in foods. Paselli SA2 is applied in products such as low-fat mayonnaise, sauces, imitation cheese, and frozen desserts, whereas C* Pur 01906, is used in baked goods. Preprepared gels consisting of one part maltodextrin and three parts water provide 1 kcal g
1. Corn maltodextrin is obtained from limited hydrolysis of cornstarch. Maltrin M040, a product of Grain Processing Corp., Muscatine, Iowa, is a spray-dried corn maltodextrin (DE ¼ 5) and is able to partially or totally replace fat in

margarine, frozen desserts, salad dressings, and extruded cereals and snacks. Tapioca maltodextrin is obtained by heating tapioca starch in the presence of hydrochloric acid, which yields a gel and is characterized by a bland flavor, smooth mouthfeel, and a texture similar to that of hydrogenated fat, for example, N-oil, N-oil II. N-oil is used as a 20–35% aqueous solution, contributes only 1 kcal g
1, and is used in products like salad dressings, puddings, and margarine. Oat maltodextrin is obtained by partial enzymatic hydrolysis of oat flour, oat fiber, or oat-bran. Oat maltodextrin (Otrim1, Otrim-5, Otrim-10) gels are useful for fat sparing in food and food products like yogurt, baked products, and meat products. Oatrim-5, invented and patented by the USDA at the National Center for Agricultural Research, is the only carbohydrate-based FS and contains the unique combination of b-glucan and low-DE maltodextrin. Wheat-based maltodextrin is produced by microfibrillation of wheat fibers and maltodextrin, for example, Vitacel, produced by Rettenmaier, is composed of 70% wheat fibers and 30% maltodextrin. Vitacel can be used instead of hydrocolloids as a thickening and binding agent for stabilization of emulsions, foams, and liquid media. Disadvantages associated with the usage of maltodextrins is that the amylopectin present in it has a tendency to retrograde slowly, giving rise to a phenomena of setback in low-fat spoonable salad dressings. Low-DE maltodextrins also suffer from freeze–thaw stability and unreliable heat and acid stability. Energy output of all the maltodextrin-based FMs is about 4 kcal g
1.

Polydextrose

Polydextrose was invented by Hans Rennhard at Pfizer Central Research Laboratories, United States, in the mid-1970s.

592

Fat Replacer

Table 4

Microcrystalline cellulose and polydextrose-based fat replacers

Trade name

Type

Usage level (%)

Avicel RC/CL

Cellulose gel

2–5.0

>3.5

Tabulose

Powdered MCC

0.5–20

–

Novagel RCN10; RCN-15 Polydextrose, Litesse, Litesse II Sta-Lite

MCC þ Guar gum

0.5–5.0

–

Polymer of glucose with sorbitol and citric acid (89:10:1) Polydextrose (fast dissolving)

5%

5–6

–

–

pH stability

Applications in food products

Manufacturer

Salad dressings, dips, spreads, bakery, products, dairy products, meat products Mayonnaise, salad dressings, sauces, dietary products, bakery products, imitation cheese products Bakery and snack foods, beverages, confectionary, dairy products, salad dressings Pastry, confectionary products, dressings, spreads, bakery fillings, toppings, chilled desserts Confectionary and bakery products

FMC Co., PA, USA Blanver Farmoquimica Ltd., USA FMC Corporation, PA, USA Pfizer Inc., New York, USA A. E. Staley Manufacturing Company, IL, USA

Compiled from various sources: Roller (1996), Sharma and Ganeshkumar (1996), Akoh (1998), Sharma et al. (1998), Sandrou and Arvanitoyannis (2000), Chavan and Prajapati (2009), Tiwari (2005).

Polydextrose is composed of randomly cross-linked glucose polymers with all types of glucosidic bonds, although 1–6 bonds are predominant (min. 90%), sorbitol end-groups (max. 2%), and monoester bonds with citric acid. Polydextrose exists in five forms viz., coarse powder, fine powder, type N, type K, and type F. Polydextrose is used as a low-calorie bulking agent that can replace all or part of the sugars and some of the fats in foods while maintaining a pleasant texture and mouthfeel. Polydextrose can also be used as a humectant, texturizer, thickener, stabilizer, and cryoprotectant (Table 4). Litesse®, a white- to cream-colored powder that has a pH (10% solution) of 2.5–3.5 and melts above 130  C, is clear, bright, and nonsticky; exhibits no crystallization; and supplies only 1 kcal g
1. According to the FDA, Litesse® may be used in frozen dairy desserts, baked goods, confections, frostings, salad dressings, gelatins, puddings, fillings, hard and soft candy, and chewing gum. Litesse® II, specifically developed for light foods, provides a higher level of sugar and fat replacement. The FDA has recognized polydextrose as a ‘carbohydrate.’

Cellulose-based FRs Cellulose is a polymer based on b-(1 ! 4)-linked D-glucose. The most widely available cellulose-based FR are microcrystalline cellulose (MCC) and methylcellulose (MC) gums.

Microcrystalline cellulose

Developed in 1964, is a nonfibrous form of cellulose, ranging in size from 25 to less than 1 mm. MCC is composed of anhydroglucose units linked through a b (1–4) glycosidic bond, which is hydrophilic, linear, and high molecular weight, and it accounts 75–95% by weight. The remaining 5–20% is soluble in nature (Table 4). MCC is produced using a-cellulose, which is composed of paracrystalline and crystalline regions of microfibrils by employing acid depolymerization and agglomeration, which is followed by drying to give porous particles of MCC. In case of colloidal grade, soluble hydrocolloids such as carboxymethyl cellulose, xanthan gum, or alginate are added to

prevent aggregation during drying. On proper dispersion, the cellulose crystallites set up a three-dimensional network that is highly thixotropic, is temperature stable, and adds body, while imparting a clean mouthfeel. Functionality of MCC is affected by different factors like adequate shear, order of addition, hard water/electrolytes presence, and pH of the food. MCC, when used in cakes at levels from 2% to 4%, improves the sponginess, increases cake volume, provides better handling properties, and creates improved product appearance. In doughnuts, powdered grade MCC (1–3%) can reduce the fat absorption. MCC is indigestible, and FDA regulations permit a maximum addition of 1.5% (w/w) of finished ice cream, with an exception in the case of fruit sherbets, wherein a maximum of 0.5% (w/w) can be added.

MC gums

Hydroxypropylemethyl-cellulose (HPMC) and MC are gums derived from chemical modification of cellulose derived from wheat, maize, potato, and/or rice. The cellulose is made soluble by adding sodium hydroxide and then treating with methyl chloride to form MC and along with propylene oxide to give HPMC, which is further dried using hot air, followed by grinding and packaging. MC and HPMC are described in terms of degree of substitution (DS) and molar substitutions (MSs). DS is the amount of substituent groups on the anhydroglucose units of cellulose, and MS is the average number of molecules of substituent that have been substituted per anhydroglucose unit. DS and MS affect physicochemical properties of MC gums like water retention, sensitivity to electrolytes, dissolution temperatures, and gelation characteristics. Methoxy substitution in MC and HPMC corresponds to a DS range of 1.49–2.00 and 3.0–12.0%, respectively. In fried foods MC and HPMC cause reduction in fat uptake, lower caloric value, improve cooking economy, provide more moisture retention, and provide improved yield. In cake and yeast-leavened doughnuts, oil reduction of about 26–28% can be achieved along with enhancement of air entrainment and promotion of uniform and fine cell size in crumb structure. In liquid foods like sauces

Fat Replacer and dressings, MC gums act as a stabilizer and provide pourability, texture, and viscosity. In frozen dairy products, their film-forming property, thickening capability, and lubricity mimics the feel of fat. MC is labeled either as ‘methylcellulose’ or E-461 and HPMC as ‘hydroxpropyl methylcellulose’ or E-464.

Dietary fiber-based FRs Dietary fibers can be used to replace fat due to their wide range of technological functionalities, and soluble fibers, such as pectin and gums, possess a higher WHC of almost 20 times their weight of water as compared to cellulosic fibers. WHC is affected by concentration, temperature, presence of certain ions, and pH of the food system. Fibers, such as pectin, gums, b-glucans with mixed bonds, and polysaccharides extracted from algae, form highly viscous solutions, whereas that of inulin is minimal. The capacity of a fiber to bind fat depends on the porosity, which is useful for avoiding excessive absorption of frying fat in batters. Dietary fiber also acts as a protective agent against cardiovascular diseases, diverticulosis, constipation, irritable colon, colon cancer, and diabetes.

Pectin

Pectin is a complex mixture of polysaccharide composed of a galacturonan backbone of which variable proportions can be methyl-esterified and is usually obtained from citrus fruit and apple. D-galacturonic acid units are linked by a-(1, 4)glucosidic bonds. Commercial pectins are divided according to the degree of esterification into low methocxyl (LM) pectins and high methoxyl (HM) pectins. HM pectins can also be chemically amidated to obtain low methyl-esterified and amidated (LMA) pectins. HM pectins are used mainly in sugar-acid gels, whereas LM and LMA pectins are used in pectate gels. Functionality of pectin is influenced mostly by the molecular weight and the level and distribution of methyl-esters. Slendid®, manufactured by Copenhagen Pectin, Denmark, is produced by extraction from plant, purification, and isolation of the pectin followed by a controlled desertification with the help of either acid or base. Currently, it is used for fat sparing in confectionary products, dairy products, and bakery fillings. To mimic the physical and sensory characteristics of emulsified fat, the particle size of 25–50 mm and usage level of 1.5–4.0% is recommended. It is possible to reduce the fat content in a frankfurter from 25% to 35% to 3–5% by using a wet preparation of 2% Slendid® 110. Pectin gives a trophic effect, increases villus height and crypt depth in the small intestine, stimulates intestinal cell proliferation and activity of brush border membrane enzymes, and stimulates short-chain fatty acid production in the cecum. Pectin gives a net energy of 9 kJ g
1 on digestion. In the United States, pectin is given a Generally Recognized As Safe (GRAS) status, whereas in the European Union it is given a designation of E440 as a food additive.

Inulin

Inulin is an oligomer found in plants such as chicory and Jerusalem artichoke and has a sweetening power of 30–65% that of sucrose and a high degree of polymerization (DP) of 2–60. Inulin is produced by extracting inulin from chicory roots by diffusing the roots in hot water followed by refining and last spray-drying the concentrate. At a concentration of

593

40–45%, it forms a gel or cre`me having a fatty creamy feel with characteristics like high water binding, stability against freeze thaw, and inhibition of syneresis in mayonnaise and salad dressings. Inulin with lower a DP of 25 is generally available for high-performance fat replacement. Inulin cre`me has been successfully applied in fat-reduced table spreads, frozen desserts, cheese products, meat products, fillings, sauces, and meat replacers (Table 4).

b-glucan

b-glucan is a cell wall polysaccharide present in oat, barley, and other grains, and recently yeast b-glucan has been used as a FR in mayonnaise. b-glucan preparation is used to partially substitute for vegetable oil in low-fat products such as salad dressings, ice creams, yogurts, and cheese. b-glucans can also be used for their therapeutic values, as they are found to be immunomodulators, antitumorogenic, and antiviral agents for treatment of hypercholesterolemia and stabilization of glycemia.

Bacterial cellulose

It is produced by some Acetobacter, which is a high-value raw material for producing dietetic and dessert foods due to its high WHC. Instead of xanthan gum, bacterial cellulose at a concentration of 3% can be used in ice creams. Bacterial cellulose improves the quality of pastry products by reducing their stickiness.

Z-trim

Z-trim refers to zero calories and can be used for replacing fat and some of the glycemic materials (starches, sugars, and syrups). It provides an aqueous gel fiber structure without taste and imparts a smooth texture. Z-Trim gels also provide insoluble fibers to be used separately or with other fibers such as Oatrim. Z-Trim gel can be used for considerable reduction in calories depending on the amount of fat and carbohydrates replaced in food formulations.

Gum-based FRs Gums are hydrocolloids, which are long-chain, high-molecular-weight polymers that dissolve or disperse in water. These gums are used along with a combination of other ingredients viz. starches, bulking agents, emulsifiers, and flavor to yield a finished product similar to a full-fat system. Viscosifying and gelling property are affected by temperature, pH, solvent quality, ionic strength, and presence of specific ions. Hydrocolloids usually come from: (a) plant materials such as seaweed, seeds, roots, and tree exudates; (b) microbial biosynthesis; and (c) chemical modification of natural polysaccharides.

Locust bean gum

It is derived from carob seed, Ceratonia Siliqua and consists of D-Mannopyronosyl backbone with attached Dgalactopyranosyl units existing in a ratio of 4:1. It is having a limited solubility in cold water but on heating to 80  C for 10 min it hydrates fully, resulting in a highly viscous pseudoplastic solution. Prolonged heating, high rate of shear, and the time of heating can irreversibly degrade the locust bean gum solution.

594

Fat Replacer

Guar gum

‘Dairy Lo,’ a thermally denatured WPC, interacts with water and prevents iciness, provides opacity, controls viscosity, and stabilizes air cells along with a fat-sparing effect in dairy product when used at a level of 2–5%. The formulations supplemented with WPC showed greater elasticity, firmness, chewiness, and gumminess compared to the control sample, thus contributing to the texture of the fat-free dairy desserts. Some of the other protein-based FRs available in the market are K-Blazer®, Ultra-BakeTM, Ultra-FreezeTM, and Lita®.

Protein-Based FRs

Fat-Based FRs

Protein-based FR is derived from proteins found in eggs, milk, and other foods (Table 5). The concept of using proteins as FRs is related to proteins with a particular spheroidal structure, ranging from 0.1 to 2.0 mm in size and is usually produced by microparticulation process. Simplesse® (NutraSweet, Deerfield, IL) was the first FR developed with protein and has received a GRAS status to be used in frozen dessert. It contains 53% protein and is a microparticulated spray-dried powder that mimics emulsified fat by forming a dispersed phase of particles that are free to move independently. It is a multifunctional dairy ingredient made from whey protein concentrate (WPC) that undergoes a unique microparticulation process that results in uniform, spherical, and deformable protein particles, much like fat globules, which range in size from 0.1 to 3 mm. This process helps to reduce the tendency to aggregate the spherical droplets and form gel on heating. These droplets roll around on the tongue, providing a smooth, creamy texture without chalkiness or graininess. Simplesse increases the opacity of the product by forming small spherical droplets of protein that allow more scattering of light. Simplesse retains the biological property of the protein used, so individuals who are allergic to egg or milk proteins can experience allergic reactions to it.

Fat-like substances, which are resistant to hydrolysis by digestive enzymes, comprise another major category of materials being promoted as partial or full replacements for oils and fats in bakery and other food products.

Guar gum or guran is the endosperm of the seed Cyampsis tetragonolobus containing a backbone of (1 ! 4)-b-Dmannopyranosyl units, with every second unit bearing a (1 ! 6)-a-D-galactopyranosyl unit. Lower mannose:galactose ratio (1.5–2) in guar gum indicates that there are few intermolecular interactions, and consequently, less heat is required for solubilization (i.e., at 25–40  C). Guar gum is usually utilized at a concentration < 1% and is stable between pH 3.5 and 9.0.

Table 5

Structured lipids A structured lipid is a triglyceride obtained by the hydrolysis and random transesterification of medium-chain triglyceride (MCTs) and long-chain triglyceride. Structured lipids provide approximately half of the calories of the normal edible oil. Caprenin, an MCT derived from coconut, palm kernel, and canola oil, mimics the physical properties of cocoa butter or confectionary fat. Another example of MCT is Neobee M-5, derived from coconut oil, which prevents bloom formation in chocolate and sticking of uncoated nutritional bars to packaging material by migrating to the surface of these products. Salatrim (short- and long-chain acyltriglyceride molecules) is a ‘family’ of triacylglycerols produced by the interesterification of highly hydrogenated vegetable oils with triacylglycerols of acetic and/or propionic and/or butyric acids. Salatrim preparations are available to emulate cocoa butter, as well as for use in baked products and filled dairy products.

Protein-based fat replacers

Product name

Type

Applications in food products

Manufacturer

SimplesseW

Microparticulate whey protein concentrate Modified whey protein concentrate

Dairy products, spreads, bakery product, salad dressing, dips, mayonnaise, frostings, sauces, soups Milk/dairy products, baked goods (cheesecake), frostings, salad dressing, mayonnaise-type products Salad dressing, mayonnaise, frozen dessert, sour cream, baked goods, and dairy products Baked goods

CP Kelco

Dairy-LoW K-BlazerW LitaW Leancreme™ Optipep™ N-Flate ULTRA-BAKETM ULTRAFREEZETM

A blend of protein, food starch, gums, and emulsifiers Corn zein Microparticulated whey protein Hydrolyzed whey protein concentrates and isolates Nonfat milk, gums, emulsifiers, and modified starch Starch, modified vegetable protein, and xanthan gum Egg-white, milk protein, corn syrup solids, and modified starch

Baked goods Yogurts, ice cream, cheese, and drinking milk

Cultor, New York, NY Kraft Food Ingredients Corp, Memphis, TN Opta Food Ingredients, Inc., Cambridge, MA SPX-APV brand Carbery

Sports nutrition products and nutrition bars Cake mixes, salad dressing, icings, glazes, desserts, ice cream, and ground beef Baked goods

Compiled from various sources: Roller (1996), Sharma and Ganeshkumar (1996), Akoh (1998), Sharma et al. (1998), Sandrou and Arvanitoyannis (2000), Chavan and Prajapati (2009), Tiwari (2005).

Fat Replacer Sugar polyesters Olestra is a mixture of hexa-, hepta-, and octaesters of sucrose prepared by esterifying sucrose with long-chain fatty acids isolated from edible fats and oils. Olestra can be liquid or solid at room temperature based on the fat source in the sucrose polyester. Olestra has the organoleptic and thermal properties of fat and thus it can serve as a zero-calorie replacement for fat in a variety of foods like savory snacks such as potato and corn chips.

Esterified propoxylated glycerol Esterified propoxylated glycerol (EPG) is a family of propylene oxide derivatives with a structure similar to that of natural fat. It is been claimed that EPG is resistant to enzymatic hydrolysis and that it can be substituted for fats and oils in products such as table spreads, frozen desserts, salad dressings, and bakery products.

595

effect on overall health. Some research suggests that individuals who consume a diet that is reduced in fat and calories and includes use of fat-modified products have a better overall nutrient profile than do individuals who do not use any fatmodified products. The recent increase in the availability of FRs in the market raises questions about the cumulative impact of using them in multiple food products and the potential interaction with medications and food ingredients. Within the context of a healthy dietary pattern, FRs, when used judiciously, may provide some flexibility in dietary planning, although additional research with clinical trials is needed to fully determine the longer-term health effects.

See also: Fats: Classification and Analysis; Fatty Acids: Determination and Requirements; Fatty Acids: Fatty Acids; Food and Agriculture Organization of the United Nations.

Dailkyl dihexadecymalonate A fatty alcohol ester of malonie and alkylmalonic acids, dailkyl dihexadecymalonate (DDM) in combination with soybean oil has been used to produce potato and tortilla chips. DDM has also been tried in products like mayonnaise and margarine.

Trialkoxytricarballate A tricarballylie acid, esterified with fatty alcohol, is being evaluated as an oil substitute to produce acceptable margarine and mayonnaise-based products.

Safety of FRs The use of FRs as a food additives to reduce the fat content in food products also raises the concern of consumers’ safety. The safety of the currently used FRs is ensured by the GRAS status by the FDA. FRs could be used and consumed by the consumers in large quantities in fat-less or reduced-fat products. Many of the carbohydrate-based, protein-based, and fat-based FRs have not shown major health concerns except for polydextrose and Olestra. Polydextrose can have a laxative effect, which requires a labeling disclaimer when present at specified levels. Olestra may cause leaky and fatty stools and loss of fatsoluble vitamins. In general, there is limited evidence at the present time on the long-term adverse consequences associated with the consumption of these or any other reduced-fat foods by adults. However, further research is required to study the safe use of these products by children and adults and to evaluate fully their long-term health effects.

Conclusion Due to the increasing demand for low-fat food products, reduced-calorie and low-fat food markets are showing a dynamic growth. As pointed out in the beginning of this article, the FR can mimic one or more roles of fat in food products. Classification of an FR usually is based on the nutrient source. They provide the functional and sensory qualities of fats in foods and are absorbed and metabolized normally. The introduction of olestra, which is a nonabsorbable FS that can affect nutrient absorption, raises questions about its potential

Further Reading Akoh CC (1998) Fat replacers. Food Technology 52: 47–53. Chavan RS and Prajapati PS (2009) Carbohydrate-based fat replacers – a review. Indian Journal of Dairy Science 62: 1–14. Cheskin LJ, Zorich N, Miday R, and Filloon T (1998) Gastrointestinal symptoms following consumption of olestra or regular triglyceride potato chips: a controlled comparison. JAMA 279(2): 150–152. Clegg SM (1996) The use of hydrocolloid gums as fat mimetics. In: Roller S and Jones SA (eds.) Handbook of fat replacers, pp. 191–212. Boca Raton, FL: CRC Press, Chapter 9. Huyghebaert A, Dewenttinck K, and deGrey W (1996) Fat replacers. IDF Bulletin 317: 10–15. Judith WR (2002) Fat substitutes and health: an advisory from the Nutrition Committee of the American Heart Association. Circulation 105: 2800–2804. http://dx.doi.org/ 10.1161/01.CIR.0000019402.35632.EB. Kaur L, Singh N, and Singh J (2004) Factors influencing the properties of hydroxypropylated potato starches. Carbohydrate Polymers 55: 211–223. Lucca PA and Tepper BJ (1994) Fat replacers and functionality of fats in foods. Trends in Food Science and Technology 5: 12–19. Mitchell HL (2004) Capture the opportunity: sustained fermentation with Litesse, polydextrose and lactitol – the health benefits. Innovations in Food Technology 13: 24–26. Ognean CF, Neli D, and Ognean M (2006) Fat replacers – review. Journal of Agroalimentary Processes and Technologies XII: 433–442. Osborn HT and Akoh CC (2002) Structured lipids: novel fats with medical, neutraceutical, and food applications. Comprehensive Reviews in Food Science and Food Safety 3: 110–120. Roller S (1996) Starch-derived fat mimetics: maltodextrins. In: Roller S and Jones SA (eds.) Handbook of fat replacers, pp. 99–118. Boca Raton, FL: CRC Press, Chapter 6A. Sandrou DK and Arvanitoyannis IS (2000) Low-fat/calorie foods. Current status and perspectives. Critical Reviews in Food Science and Nutrition 40: 427–447. Sharma A and Ganeshkumar C (1996) Fat replacers as future dietary regimes: an overview. Indian Dairyman 48: 27–35. Stanton JM (1990) Fat substitute update. Food Australia 42: 472–475.

Relevant Websites http://www2.ca.uky.edu/agc/pubs/fcs3/fcs3208/fcs3208.pdf. http://www.circulationaha.org. http://www.cfs.purdue.edu/fn/fn453/pdf_full/Fat_Replacers.pdf. http://www.fda.gov/bbs/topics/ANSWERS/2003/ANS01245.html. http://www.ift.org//media/Knowledge%20Center/Science%20Reports/Scientific% 20Statu%20Summaries/fatreplacers_0398.pdf. http://journal-of-agroalimentary.ro/admin/articole/74033L66_FAT_REPLACERS_final. pdf.

Fats: Classification and Analysis M Narva´ez-Rivas and M Leo´n-Camacho, Instituto de la Grasa (C.S.I.C.), Seville, Spain ã 2016 Elsevier Ltd. All rights reserved.

Introduction Fats are essential nutrients and concentrated sources of energy of the diet, supplying about 9 kcal g
1. They are soluble in most organic solvents, such as hexane and ethyl ether, but basically insoluble in water. By classical definition, fats are referred to total lipids, which can be classified into two different fractions depending on if they form or fail to form soaps when blended with sodium hydroxide, considered as saponifiable and unsaponifiable fractions, respectively. The different compounds that can be part of the saponifiable fraction are shown in the next scheme. The unsaponifiable matter contains terpenic (sterols, 4-methylsterols, 4,40 -dimethylsterols, carotenoids, tocopherols, and tocotrienols) and aliphatic compounds (saturated and unsaturated hydrocarbons and fatty alcohols). The different ways of analysis of the fat compounds are going to be described in the succeeding text.

Saponifiable Fraction Triacylglycerols Triacylglycerols consist of the trihydric alcohol glycerol esterified with fatty acids. They represent the main lipid fraction of foods. For the analysis of molecular species of triacylglycerols, chromatographic and spectrometric methods are available. However, stereospecific determination of positional distributions of fatty acids in them remains a substantial technical challenge. Nowadays, the most used methods for their analysis are HPLC and GC.

Reversed-phase HPLC Good resolution of triacylglycerols by reversed-phase highperformance liquid chromatography is possible. The retention times are dependent on the total number of carbon atoms of the three fatty acids of the molecule, with each double bond reducing the effective chain length of the species by the equivalent of about two carbon atoms. The most frequently used columns have been those with phases of the octadecylsilyl (C18, ODS, or ODS-2) type, since stationary phases similar in chain length to the fatty acyl chains maximize the interactions and give a higher efficiency. Silver ion HPLC has been also used. The 250  4.6 mm column with 5 mm packing, the 100  3 mm column with a 3 mm packing, and the 100  2.1 mm column with 1.7 mm packing have been used, the last one being used in terms of saving time and solvent and with no loss of resolution. Referring to the mobile phases, the most used are – acetonitrile–acetone, – acetonitrile–isopropanol, – acetonitrile–dichloromethane,

596

– acetonitrile–ethanol–hexane, – acetonitrile–tetrahydrofuran. They improve the solubility of triacylglycerols and have an effect on the selectivity and efficiency of the separation. The detector used and the nature of triacylglycerols may determine the choice of mobile phase and the choice of an isocratic or a gradient system. For example, acetonitrile–acetone mixtures are not well suited to saturated components of higher molecular weight, since they tend to precipitate out of solution. For such purpose, the acetonitrile–dichloromethane combination may be the best available. Isocratic eluants are utilized with detectors in which the change in the composition of the mobile phase affects the obtained signal, as in the case of the refractive index (RI) detector. The gradient of elution can be used with detectors such as UV and evaporative light scattering detectors (ELSDs). The column temperature should be fixed and controlled to achieve reproducible retention times. With low temperatures, solubility effects may come into play, and the more saturated molecular species may precipitate out of solution. The detection systems for quantitative analysis of triacylglycerol species separated by HPLC have been RI, UV, evaporative light scattering, and mass spectrometry detectors. One of the most used for the mentioned purpose is the RI detector, in which the RI changes with the chain length and with the unsaturation number. The response factor is considered the same (1) for all triacylglycerols in the practice. This detector is affected by the temperature and composition of the mobile phase, and due to this, the analysis needs to be carried out with thermostated cells and isocratic elution. In some cases, flow gradients can be used. Another limitation of this detector is its low sensitivity. However, RI detection gives good results and is one of the most adequate for the quantitative analysis of these compounds. It has been reported that UV detection can be used to get good quantitative data at wavelength between 200 and 220 nm, where the ester bond absorbs. This detector shows the highest sensitivity. Nevertheless, the absorption of several solvents used in the mobile phase, like acetone, the unsaturations, and the conjugation of these unsaturations are a problem for the use of this detector in triacylglycerol analysis. An alternative to the detectors mentioned is the ELSD, which is universal and simple. This shows the disadvantages that a calibration method must be taken into account and that the linear range is between 50 and 250 mg. Below 50 mg, the sensitivity decreases, which is a problem for the analysis of the minor compounds. But this detector shows the advantages that no solvent peak appears and the resolution of the peaks is good and it is not affected by the changes in the solvents of the mobile phase. For both detecting and identifying molecular species of triacylglycerols, mass spectrometry is used, showing a good sensitivity, although a careful calibration is also needed.

Encyclopedia of Food and Health

http://dx.doi.org/10.1016/B978-0-12-384947-2.00274-9

Fats: Classification and Analysis The choice of injection solvent is another important practical point that should be mentioned. Samples should be injected in a small volume of the mobile phase (5–10 ml). Hexane is not employed as injection solvent because a single component can emerge as double peaks due to its properties, which are so similar to those of the stationary phase that it competes with this for the solute molecules.

597

resolving triglyceride mixtures even better than nonpolar phases, but they cannot reach temperatures over 300  C. The polarizable phenyl methyl silicone phases are capable of enduring temperatures of about 360–370  C for a long time, separating triacylglycerols by both chain length and the number of unsaturation. The fused silica capillary DB-17ht column (30 m  0.32 mm i.d. and 0.15 mm film thickness) shows a really good resolution for the different molecular species of triacylglycerols. Detection is carried out by using a flame ionization detector (FID), and the response factors need to be considered, which depend on a certain extent on the ratio of the carrier gas. The purity and nature of the carrier gas, flow rate, rate of temperature change, and whether the carrier gas is maintained at constant flow or constant pressure rate are other practical factors that have to be considered. An example of triacylglycerol analysis by GC-FID is shown in Figure 1.

High-temperature GC Nowadays, the use of high-temperature GC with flame ionization detection for the separation and quantification of triacylglycerols is a good alternative to HPLC, solving its problems. Triacylglycerols are not at all easy to volatilize for gas chromatography purposes since their molecular weight is quite high. The injection procedure is of great importance to avoid any sample discrimination and achieve good quantification. Modern gas chromatographs usually come equipped to achieve such purposes with cold on-column injection, split injection, and/or programmed temperature vaporizer injection. The use of cold on-column injection is one of the most desirable to reduce or eliminate thermal decomposition (e.g., the pyrolysis of the most unsaturated triacylglycerols), mass discrimination, and other undesirable effects associated with classical split/ splitless injection. Using this injection technique, the injector temperature should be lower than the solvent boiling point. The new split/splitless injectors minimize mass discrimination loss and make possible to apply this technique to triacylglycerol analysis with similar results to those obtained by using cold on-column injection. Another aspect that needs to be addressed is the nature of the stationary phase. The temperature stability of this is highly important. The most stable to high temperatures is the nonpolarity phases of the dimethylpolysiloxane type, which permit separation by chain length. However, high-polarity or polarizable phases allow remarkable separations by both carbon atom number and degree of unsaturation, but they are less stable to high temperatures. Columns coated with polar stationary phases, such as cyanopropyl silicone, are capable of

Diacylglycerols and Monoacylglycerols Diacylglycerols are esters of the trihydric alcohol glycerol in which two of the hydroxyl groups are esterified with fatty acids, and they are found as 1,2- and 1,3-isomers. They are generated by acidic and enzymatic hydrolyses of triacylglycerols during the transformation and storage of fats, and 1,3-isomers can be generated from 1,2-diacylglycerols due to an isomerization process, since the 1,3 species is more thermodynamically stable. In tissues, 1,2-diacylglycerols can also be generated from phosphatidylinositol (PI) by the action of phospholipase C. The determination of molecular species of diacylglycerols in fats can be accomplished by a previous extraction and purification by column chromatography or solid phase extraction (SPE) and a subsequent chromatographic method, GC being the most appropriate for that purpose. An aliquot of the fat in hexane and a suitable internal standard (e.g., 1,3dipalmitoyl-glycerol and dinonadecanoine) are added prior to

mVolts 50

9

40 30 14

20

8 10

15 13

10

16

4 5 12

0 5

3

6 7

10

11 12

17

15

20 Time (Min)

Figure 1 Chromatogram of the triacylglycerol profile of an Iberian pig subcutaneous fat sample, where 1, PPP; 2, MOP; 3, PPS; 4, POP; 5, POPo þ PLP; 6, PLPo þ MLO; 7, PSS; 8, PSO; 9, POO; 10, PLO; 11, PLL þ PoLO; 12, SOS; 13, SOO; 14, OOO; 15, SOL; 16, OOL; 17, OLL. Reproduced from Journal of Agricultural and Food Chemistry, 2012, 60, 1645–1651.

598

Fats: Classification and Analysis

mVolts 2 Diglycerides

75

5

50 3 1

25

4 0 −9 5

10

15

20

25

30 Minutes

Figure 2 Gas chromatogram of the polar fraction from 0 to 35 min corresponding to the subcutaneous fat of Iberian dry-cured ham. Labeled peaks are identified as follows: 1, palmitic acid; 2, oleic acid; 3, 1-monoolein; 4, cholesterol; 5, 1,3-dimyristin (IS). Reproduced from Journal of Agricultural and Food Chemistry, 2007, 55, 10953–10961.

the purification step, in which silica or diol SPE columns are utilized. Purification comprises several steps using different solvent systems. The first is carried out passing a solution to eliminate the major compounds (e.g., hexane/methylene chloride/ethyl ether (89:10:1)) through the column, which is discarded. In the second step, the diacylglycerol fraction is obtained by passing a solution, generally chloroform/methanol (2:1), that is evaporated to dryness in a rotary evaporator under reduced pressure, and the residue is treated with 200 ml of the silylating reagent and left at room temperature for a few minutes, obtaining the corresponding trimethylsilyl derivatives of 1,2- and 1,3diacylglycerols. Monoacylglycerols are also extracted with the diacylglycerol fraction, and they can be analyzed together by GC using a proper temperature program and a proper column. An example of this is shown in Figure 2. The analysis of the silylated fraction is performed by GCFID using thermostable midpolarity GC columns (e.g., fused silica capillary DB-17ht polyamide-coated column, 50% phenyl methylpolysiloxane, 65% phenyl- and 35% dimethylpolysiloxane, and 65% methyl phenyl silicone). The peak of 1,2-diacylglycerols appears before their corresponding 1,3isomers. With respect to monoacylglycerols, they are at trace levels in fats, and their analytic determination is carried out with a procedure similar to diacylglycerols, trimethylsilyl derivatization, and GC-FID analysis but with nonpolar capillary columns, or their combination as it has been mentioned in the preceding text.

Phospholipids Phospholipids are a class of lipids that have two fatty acyl molecules esterified at the sn-1 and sn-2 positions of glycerol and a head group linked by a phosphate residue at the sn-3 position. They have amphipathic property since the head group forms a hydrophilic region that determines the type of phospholipid, and the fatty acyl side chains are hydrophobic.

Two different categories are distinguished: glycerophospholipids (phosphatidylethanolamine (PE), phosphatidylcholine (PC), phosphatidylinositol (PI), phosphatidylserine (PS), and cardiolipin (CL)), the derivatives of glycerol, and sphingophospholipids (sphingomyelin: SPH), those that do not contain glycerol. Phospholipids have been determined traditionally by gravimetric methods and spectroscopic methods, like molecular or atomic absorption spectroscopy. However, these are not usually used nowadays. The thin-layer chromatography (TLC) technique using adsorbent (silica) coated on glass strips is used for the separation of phospholipids as a separation technique previous to the quantification of these by another procedure or even for their direct quantification by densitometry or thincography. The type of mobile phase used and whether the TLC is in one or two dimensions are critical in the separation of the different groups. This technique has some disadvantages, such as a limited capacity to resolve the phosphatides in their different molecular species, techniques of detection more or less complex with significant errors in the quantification, and a relatively long time period required for analysis. Nowadays, the method used for the separation and determination of phospholipid classes is the HPLC. Silica columns are the most used for such purposes since this can give excellent results and is relatively inexpensive and robust. Nevertheless, amino columns have been also used, but they need to be periodically regenerated with ammonia solutions. So, they are not widely used. The most used detectors in the study of phospholipid classes are the UV and the ELSD. The response of UV detector is dependent on the degree of unsaturation in the molecule, and different wavelengths have been used, but the most frequently used are 205 and 208 nm (absorption of the ester bond). Besides, in these detectors, another problem is the absorption of some mobile phases at these wavelengths. That is why there are solvents that cannot be used with these detectors, like chloroform and acetic acid. So, the quantification of

Fats: Classification and Analysis phospholipids using this detector is difficult. On the other hand, ELSD gives a stable flat baseline and is very sensitive, and its response is not dependent on the number of double bonds in the molecule, showing the advantage of simplicity and universality. To use this detector in quantitative analysis, the response factors of the phospholipids must be calculated, due to the minor components. Referring to the mobile phases used, an isocratic eluant has rarely been used. With a UV detector, an isocratic mode or a linear gradient can be applied. Binary gradients can be employed using ELSD, and the following have been used in several works: (a) chloroform–methanol–ammonium hydroxide and (b) chloroform–methanol–water–ammonium hydroxide. More complicated gradients of elution with three or four eluants have been used, but to a lesser degree than the binary ones. A wide range of different mobile phases have been used to separate the different phospholipid classes, since unfortunately not all phases have the same effectiveness for the different types of samples and the composition of the mobile phase can become critical for each type of matrix. Binary gradients that have been successfully used with different types of matrix are (a) chloroform–methanol–ammonia solution and (b) chloroform–methanol–triethylamine–water. An example is shown in Figure 3. The use of reversed-phase HPLC allows the separation of the different molecular species of each phospholipid class. Normally, the different phospholipid classes are separated and collected by HPLC or TLC. One or two columns in series can be used depending on the resolution desired and the mobile phase used. Acetonitrile–methanol–water is the most commonly used mobile phase in reversed-phase HPLC using an RP18 column. The last thing to mention about phospholipids is that they can be purified previously to their chromatographic analysis using SPE to isolate them, and the best results are obtained when silica columns are employed, allowing a high recovery.

599

1 6

2

4 5 3

7

0

2.5

5

7.5

10

12.5

15

17.5

20 min

17.5

20 min

(a)

2

5 6.8

7

7.2 7.4 7.6 7.8

1

8

8.2 min

2

6

3 4

Waxes Waxes are organic compounds that consist of long alkyl chains. They are insoluble in water but soluble in organic nonpolar solvents. The waxy material of the fats can include primary and secondary alcohols, diols, ketones, b-diketones, triacylglycerols, hydrocarbons, sterol esters, and aliphatic aldehydes. Silica gel column chromatography, TLC, SPE, and HPLC (using silica) have been used to separate them from other lipid constituents and to isolate individual classes of waxes for more detailed analysis. On the other hand, high-temperature gas chromatography following trimethylsilylation has been used in recent works, often in combination with mass spectrometry, to achieve simultaneous identification and quantification of the various molecular species. Semipolar columns can be used for their analysis, although a better resolution is obtained with phenyl methyl silicone columns, which can endure higher temperatures. All the methods for the determination of waxes consist of a separation from other lipid constituents using silica gel column chromatography, SPE, or HPLC and subsequent quantification using GC. In the official method, the waxes are determined by separating them according to the number of carbon atoms. The fractionation step is carried out by chromatography on a

0

2.5

5

7.5

10

12.5

15

(b)

Figure 3 HPLC chromatogram of phospholipid fractions (a) from the standard solution and (b) from the subcutaneous fat of Iberian pig. 1, Cardiolipin; 2, phosphatidylethanolamine; 3, phosphatidylinositol; 4, phosphatidylserine; 5, phosphatidylcholine; 6, sphingomyelin; 7, lysophosphatidylcholine. Reproduced from Journal of Chromatography A, 2011, 1218, 3453–3458.

hydrated silica gel column after the addition of the internal standard (e.g., lauryl arachidate). The eluted fraction is recovered and subsequently injected on-column in a GC capillary column. SE-54 or SE-52 semipolar columns can be used, although a better resolution can be achieved with phenyl methyl silicone columns, which can tolerate higher temperatures. Waxes can be determined simultaneously with fatty acid methyl esters and fatty acid ethyl esters by capillary gas chromatography, which is a method widely applied in laboratories. A method based on through oven transfer adsorption–desorption (TOTAD) interface is another analytic

600

Fats: Classification and Analysis

alternative. The lipid extract, with an internal standard diluted in n-heptane, is injected directly with no sample pretreatment step other than filtration. The wax ester fraction is separated from the triglycerides by normal-phase liquid chromatography, and the TOTAD interface transfers it to the GC to be analyzed. This method allows an automatic analysis of different wax esters, being simpler and faster than other methods, although its resolution is poorer. A method in which samples are prepared by normal-phase liquid chromatography has also been proposed for the analysis of wax esters, but the precolumn is attached to the inlet of the column in the GC  GC instrument by means of a press-fit connector, the first dimension being performed with a PS-255 column (20 m  0.25 mm i.d.) and the second dimension with a SOP-50 (50% phenyl polysiloxane) (1.5 m  0.15 mm i.d.).

Unsaponifiable Fraction The main classes of compounds that constitute the unsaponifiable fraction of food are hydrocarbons, carotenes, tocopherols, tocotrienols, linear fatty alcohols, triterpenic alcohols, methyl sterols, and sterols. Other components that are part of the unsaponifiable fraction are carotenoid pigments, triterpenic dialcohols, diterpenic alcohols, and phytol, but they are not always present in food matrices. The unsaponifiable fraction is got doing a saponification with ethanolic potassium hydroxide at high temperature. After this, distilled water is added to the solution, and the mixture is extracted several times with portions of a nonpolar solvent such as hexane, heptanes, diethyl ether, chloroform, or their mixtures thereof. The unsaponifiables will remain in the nonpolar phase, while the fatty acid salts will partition to the polar phase. The nonpolar extracts are combined and washed several times with portions of a mixture of ethanol–water (1:1), until the wash reaches neutral pH. Finally, the hexane solution is dried under a stream of nitrogen or by rotoevaporation and redissolved in an appropriate amount of solvent for cleanup by either TLC or SPE or directly for derivatization. The fractionation of the different classes present in the extract can be

carried out by HPLC. In Figure 4, there is an example of the fractionation using the HPLC technique. The same preparative procedure is utilized for the analysis of all unsaponifiable components. The trimethylsilyl derivatives (except for the hydrocarbons) of the purified fractions are analyzed using nonpolar capillary gas chromatography columns. On the other hand, a direct analysis of the silylated unsaponifiable components obtained from different food products can be carried out using a thermostable polar gas chromatography column (65% phenyl–35% dimethylpolysiloxane).

Hydrocarbons In the natural lipid systems of food matrices, hydrocarbons are present in quite small amounts, and most of them have an odd number of carbon atoms. Cyclic, polycyclic aromatic, and ramified hydrocarbons are also present in small amounts. A good separation of hydrocarbons is provided by silica TLC of natural lipids using a mobile phase of n-hexane-diethyl ether (7:3, v/v), eluting just after the solvent front. To improve the isolation of the unsaponifiable fraction of lipids, TLC and liquid chromatography are being replaced by HPLC methods, and HPLC-GC off-line methods that allow isolation and quantification of the hydrocarbon fraction have been reported. The column used for such purposes is a silica column and the mobile phase is hexane/ ethyl acetate. SPE is also used, like in the case of polycyclic aromatic hydrocarbons, which are considered as contaminants. After their purification, the hydrocarbons are analyzed by GC-FID or GC-MS, the last technique allowing their simultaneous identification. For polycyclic aromatic hydrocarbon analysis, low-bleed and low-polarity columns have been used, like DB-XLB columns. Linear, branched, and terpenic hydrocarbon analyses are performed using nonpolar columns of (95%) dimethyl–(5%) diphenylpolysiloxane type. An example of hydrocarbon analysis by GC-FID is shown in Figure 5.

Alcohols and Sterols Linear alcohols are constituted by a homologue series of primary fatty alcohols that generally contain 20–30 carbon

Linear Hydrocarbons nRIU Sterols 300000 Terpenic Hydrocarbons

250000 200000 150000

Linear and Terpenic alcohols

100000 50000 0 –50000 –100000 0

5

10

15

20

Figure 4 HPLC chromatogram of unsaponifiable fraction from the subcutaneous fat of Iberian pig.

25

min

Fats: Classification and Analysis

601

I.S.

(a) I.S. 5

7

3

9

5´

3´

KAURENE 11 13 15

7´ 13´

11´

1

0

SQUALENE

C20H38

10

20

(b)

30

17

19

21

15´

40

50

60

Minutes

Figure 5 Chromatograms of hydrocarbons obtained from the subcutaneous fat of Iberian pig by LC (a) and HPLC (b). Peaks: 1, 1-dodecene; 3, 1-tetradecene; 30 , tetradecane; 5, 1-hexadecene; 50 , hexadecane; 7, 1-octadecene; 70 , octadecane; 9, 1-eicosene; 11, 1-docosene; 110 , docosane; 13, 1-tetracosene; 130 , tetracosane; 15, 1-hexacosene; 150 , hexacosane; 17, 1-octacosene; 19, 1-triacontene; 21, 1-dotriacontene.

atoms. The triterpenic alcohols have a steroid structure and are present at different levels in vegetal lipids. These are also known as 4,40 -dimethylsterols. Both linear and triterpenic alcohols are often badly separated in silica TLC. To achieve a better TLC separation, a multiple development technique with a slightly different mobile phase for the second development can be carried out. Then, they are determined using the GC technique. Sterols are steroid alcohols and they occur naturally in plants (phytosterols), animals, and fungi. Cholesterol (cholest-5-en-3b-ol) is by far the most abundant sterol in animal tissue. This can be esterified to long-chain fatty acids (cholesterol esters). These cholesterol esters are much less polar than free cholesterol. Cholesterol oxidizes slowly in tissues or foods forming a range of different products with additional hydroperoxy, epoxy, hydroxy, or keto groups. These compounds are called oxysterols, cholesterol oxides, or cholesterol oxidation products (COPs). Enzymatic methods from commercially available kits are used to determine the cholesterol content. It is necessary to hydrolyze the cholesterol ester fraction to determine the total cholesterol content. In the analysis of sterols, these are first isolated from lipid extracts by TLC or column chromatography, following hydrolysis if necessary. Individual components can then be determined by GC, often after conversion to trimethylsilyl ether derivatives to give sharper peaks. The addition of an internal standard allows obtaining accurate qualitative and quantitative results by GC. The most common internal standards for sterol quantification are 5a-cholestane, dihydrocholesterol (5acholestan-3b-ol), epicoprostanol (5b-cholestan-3a-ol), and

betulin. The internal standard is typically added to the sample prior to alkaline or acidic hydrolysis, undergoing the same extraction conditions as the sterol fraction, or it can be added after extraction but prior to derivatization. Acidic hydrolysis is necessary to free the sterols when the sample has a complex protein or carbohydrate matrix and/or steryl glycosides. If this is needed, the internal standard is added to the weighed sample, and then, it is hydrolyzed with hydrochloric acid at 100  C. After the acidic hydrolysis, lipids are extracted using a nonpolar solvent. After solvent removal, saponification of the lipid fraction is carried out as described in the preceding text. Once saponification has been done, the unsaponifiable fraction may contain other lipids besides sterols including hydrocarbons, carotenoids, tocopherols, free fatty acids, and other triterpenes. For the unsaponifiable fraction of a refined fat or oil sample, it is usually acceptable to proceed with derivatization and GC analysis without further sample cleanup, and no problems with interference by these compounds are presented. However, in other types of samples, the sterol fraction needs to be purified. For such purpose, silica, C18, and aminopropyl SPE cartridges have all been used. Nowadays, sterols are usually analyzed as trimethylsilyl ethers, since their volatility, peak shape, and response factors improve. Direct injection of nonderivatized sterols results in broader peaks and a lower FID response. The silylating reagents used are N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) and N-methyltrimethylsilyltrifluoroacetamide (MSTFA), as well as mixtures of pyridine, hexamethyldisilazane, and trimethylchlorosilane (9:3:1, v/v/v). Derivatization reactions may proceed

602

Fats: Classification and Analysis

completely in several minutes (about 15 or 20 min) at room temperature or heated to complete the reaction when the hydroxyl group on the sterols is hindered in any way. After derivatization, sterols can immediately be analyzed by GC or after carefully evaporating the derivatization reagents with nitrogen at low temperatures to prevent any sterol ethers and redissolving them in an organic solvent, like chloroform, to an appropriate concentration before injecting. Split mode injection is usually used and the injector temperature usually set from 250 to 300  C. Fused silica capillary columns with polysiloxane phases of low to midpolarity are used for their analysis. The most common column phase used for sterol analysis is composed of (95%) dimethyl–(5%) diphenylpolysiloxane. A nonpolar phase of (100%) dimethylpolysiloxane can be used, but a good separation of sterol peaks and their respective stanol peaks is not achieved with this column phase. Midpolarity columns (14% cyanopropyl-phenyl-methylpolysiloxane) have been suggested for a better resolution of D5-saturated and D5-unsaturated sterols. For the identification of individual components of sterol fraction, mass spectrometry may be required. HPLC can be employed for the analysis of sterols, using UV and mass spectrometric detection for their quantification and identification, respectively. An HPLC-/atmospheric pressure chemical ionization-MS method has been developed for the analysis of these compounds, in which no previous derivatization is needed. Several stationary reverse phases (C8, C18, and zirconia-based adsorbent) have been studied for this purpose, and the best separation of sterols has been obtained with the C8 phase. Increasing the column temperature accelerates their elution markedly. Nevertheless, different mobile phases have been used. Rigorous attention in the analysis of the COPs should be attempted because they tend to be present at rather low levels, and there is a danger of further oxidation or side reactions during the analytic process. Both hot and cold saponifications, using either ethanolic or methanolic potassium hydroxide (KOH), have been used in the analysis of COPs. Their enrichment can be carried out using either silica or aminopropyl SPE. Besides, the sample should be protected against light, oxygen, and high-temperature exposures, since they promote lipid oxidation. HPLC and GC techniques have been used for the analysis of COPs, but GC has given better results than HPLC. Samples are derivatized to trimethylsilyl ethers using BSTFA, Tri-Sil reagent (Pierce Chemical Co., Rockford, IL), or the mix of hexamethyldisilazane–trimethylchlorosilane–anhydrous pyridine (3:1:9, v/v/v) prior to GC and GC-MS analyses. Fused silica open tubular capillary columns coated with (5%) phenyl methyl silicone and columns coated with (100%) dimethylpolysiloxane, such as DB-1, have been used for their analysis. Similar results are obtained using DB5-MS and DB1MS columns. Although GC has been the most used technique for the analysis of COPs, some HPLC methods have been reported. A method using reversed-phase LC/atmospheric pressure chemical ionization mass spectrometry for the successful determination of cholesterol and its oxidized compounds in processed foods such as pork, beef, chicken, and eggs has been described. Diode array ultraviolet detection and laser light scattering detection (LLSD) have been compared, concluding that the use of LLSD in the HPLC of cholesterol and its oxidized

products could provide a reliable tool for the determination of these compounds, since the use of this allowed the detection of cholestanetriol and a- and b-epoxides. Between the different internal standards used for their quantification are 5a-colestane, betulin, 19-hydroxycholesterol, and betamethasone-17, 19-dipropionate. 4-Methylsterol and 4,4-dimethylsterol fractions are analyzed in a similar way to other sterols.

Tocopherols and Tocotrienols These compounds consist of a polar chromanol ring and a hydrophobic 16-carbon side chain attached to the ring via the C2 atom. The difference is that tocopherols contain saturated phytyl side chains, while tocotrienols have isoprenyl side chains with three double bonds. They display antioxidant properties and are active as vitamins (vitamin E), making them particularly important for human health. They are amphipathic and lipid-soluble and are easily oxidized when subjected to heat, light, and alkaline conditions. Tocotrienols are present in small amounts in food lipids. However, palm oil, grape-seed oil, and the annatto lipid fraction contain a relatively high amount of these components. The determination of the tocopherols and tocotrienols as trimethylsilyl derivatives can be carried out on a short GC column of medium polarity (OV-17), with both flame ionization and mass spectrometric detection, within a short time of analysis, achieving a very good separation of the two classes of components. However, the methods that are usually recommended involve HPLC with fluorescence detection. These compounds can be extracted using solvent or alkaline hydrolysis extractions. Supercritical fluid or pressurized liquid extractions are alternative techniques to traditional solvent extraction methods that may be used to extract tocopherols and tocotrienols. HPLC is the most widely used technique for the analysis of these compounds since they are stable under HPLC conditions and easy to dissolve in appropriate solvents, and there are several detectors that can be combined with this technique to detect them, fluorescence detection and UV (l ¼ 290–300 nm) detection being the most commonly used. Aside from this, both normal-phase chromatography and reversed-phase chromatography can be applied. Normal-phase using silica columns is commonly applied when all eight tocopherols and tocotrienols need to be separated, achieving better selectivity when hexane is used with a relatively strong polar modifier 1,4-dioxane in the mobile phase than with a weak modifier such as tert-butyl methyl ether. However, reversed-phase might also be utilized when only some vitamins are of interest or when mixtures of fat-soluble vitamins and free and esterified tocopherols and/or tocotrienols have to be separated. The external standard method is usually used to quantify tocopherols and tocotrienols because it is difficult to find a compound that would not interfere with their analysis.

See also: Fatty Acids: Determination and Requirements; Fatty Acids: Fatty Acids; Triacylglycerols: Characterization and Determination;

Fats: Classification and Analysis Triacylglycerols: Structures and Properties; Vegetable Oils: Composition and Analysis.

Further Reading Abidi SL (2001) Chromatographic analysis of plant sterols in foods and vegetable oils. Journal of Chromatography A 935: 173–201. Hamilton RJ and Rossell JB (eds.) (1986) The analysis of oils and fats. London: Elsevier Applied Science. Harwood J and Aparicio R (2013) Handbook of olive oil: analysis and properties. New York: Springer. International Organization for Standardization, ISO 12228:1999 (1999) Animal and vegetable fats and oils-determination of individual and total sterols content-gas chromatographic method. Geneva, Switzerland. Jurado JM, Jime´nez-Lirola A, Narva´ez-Rivas M, Gallardo E, Pablos F, and Leo´nCamacho M (2013) Characterization and quantification of 4-methylsterols and 4,4dimethylsterols from Iberian pig subcutaneous fat by gas chromatography–mass spectrometry and gas chromatography–flame ionization detector and their use to authenticate the fattening systems. Talanta 106: 14–19.

603

Kamal-Eldin A, Gorgen S, Pettersson J, and Lampi AM (2000) Normal-phase highperformance liquid chromatography of tocopherols and tocotrienols – comparison of different chromatographic columns. Journal of Chromatography A 881: 217–227. Kuksis A (ed.) (1978) Handbook of lipid research: fatty acids and glycerides, vol. 1. New York: Plenum Press. Laakso P (2005) Analysis of sterols from various food matrices. European Journal of Lipid Science and Technology 107: 402–410. Lercker G and Rodriguez-Estrada MT (2000) Chromatographic analysis of unsaponifiable compounds of olive oils and fat-containing foods. Journal of Chromatography A 881: 105–129. Narva´ez-Rivas M, Gallardo E, Rı´os JJ, and Leo´n-Camacho M (2011) A new highperformance liquid chromatographic method with evaporative light scattering detector for the analysis of phospholipids. Application to Iberian pig subcutaneous fat. Journal of Chromatography A 1218: 3453–3458. Nielsen MM and Hansen A (2008) Rapid high-performance liquid chromatography determination of tocopherols and tocotrienols in cereals. Cereal Chemistry 85: 248–251.

Relevant Website http://lipidlibrary.aocs.org/ – AOCS Lipid Library.

Fats: Production and Uses of Animal Fats SB Smith and DR Smith, Texas A&M University, College Station, TX, USA ã 2016 Elsevier Ltd. All rights reserved.

Introduction The composition of lipids in fatty tissues of meat varies in response to diet, the time at which the lipids are deposited, and adipose tissue depot. The fatty acid composition of pork is especially sensitive to dietary manipulation, whereas that of beef and lamb is affected primarily by the age of the animal and fat depot, although fatty acid composition of ruminant tissues can be modified by dietary means. For example, pasture or grass feeding causes fat depots to be high in saturated fatty acids, whereas grain feeding increases the concentration of monounsaturated fatty acids. Between 45% and 50% of the entire live weight of livestock species does not enter directly into the human food chain. The by-products of animal slaughter and fabrication include muscle trim and a large quantity of the carcass fat (especially subcutaneous and internal fat depots) that are not used in processed meats. In addition, many internal organs are included in total by-products. Altogether, these by-products contain approximately 20% extractable fat. With rendering, some of this extractable fat can be recovered and converted into more useful and profitable materials such as lard or tallow, allowing part of carcass fat to reenter the human food chain. The quality, and especially the fatty acid composition of the rendered materials, determines the fate of the finished product.

Sources of Lipid in Meat There are four sources of lipid in meat that are used for meat processing or rendering: the muscle fibers; subcutaneous fat; intermuscular (seam) fat between muscle groups; and intramuscular (interfascicular and marbling) fat. Very lean beef, pork, or lamb, in which all subcutaneous and intermuscular fat has been removed, contains approximately 1% extractable lipid. At the other extreme, even lean trim from Japanese Black (or wagyu) cattle can contain over 40% extractable lipid. This extraordinary concentration of lipid can be attributed largely to lipid within the muscle.

Fatty Acid Composition of Subcutaneous, Intramuscular, and Muscle Lipids Lipids from fat, lean trim, and organs destined for rendering contain the saturated fatty acids palmitic acid (16:0) and stearic acid (18:0) and their delta-9-desaturase products, the monounsaturated fatty acids palmitoleic acid (16:1n 7) and oleic acid (18:1n 9). Palmitic acid is produced by the combined activities of acetyl-coenzyme A carboxylase and fatty acid synthase, whereas stearic acid is produced by the addition of two carbons to palmitic acid via a fatty acid elongase. Palmitoleic acid and oleic acid are the products of the delta 9

604

desaturation of palmitic acid and stearic acid, respectively, catalyzed by the enzyme stearoyl-CoA desaturase (SCD). Palmitic, palmitoleic, stearic, and oleic acids are always present in all tissues of animals (Figures 1 and 2), but the relative concentration of each fatty acid is dictated primarily by the activity of the enzyme SCD. The polyunsaturated fatty acids typically consumed by livestock species consist primarily of linoleic acid (18:2n 6) and a-linolenic acid (18:2n 3). In the rumen of cattle, sheep, and goats, polyunsaturated fatty acids are extensively isomerized to cis-9, trans-11 conjugated linoleic acid (CLA) and other CLA isomers (Figure 3). However, the concentration of CLA isomers in ruminant tissues does not exceed 2% of total fatty acids because CLA is almost quantitatively hydrogenated to stearic acid in the rumen, although small portions of trans-vaccenic acid and other fatty acid trans-isomers escape the rumen and are absorbed in the gastrointestinal tract. For this reason, ruminant tissues are relatively high in stearic acid and naturally occurring trans-fatty acids. The contribution of both de novo fatty acid synthesis and fatty acid desaturation to the fatty acid composition of fat depots is especially apparent in bovine fat (Figure 1). Stearic acid constitutes more than 80% of the fatty acids available for absorption from the duodenum in beef cattle, yet the most abundant fatty acids in bovine fat are palmitic acid (the product of de novo synthesis) and oleic acid (from desaturation of stearic acid). Typically, abdominal fat is more highly saturated than subcutaneous or intramuscular fat. In beef cattle, intramuscular fat is also higher in stearic acid than is subcutaneous fat (Figure 1). Unlike the situation for beef, intramuscular fat from pork contains less stearic acid than does subcutaneous fat (Figure 2). As monogastrics, swine have adipose tissues with fatty acid composition that more closely resembles that of dietary fatty acids. Thus, diets enriched with oleic or linoleic acid will cause the lipids associated with pork to become similarly enriched with these fatty acids. Polyunsaturated fatty acids are especially susceptible to oxidation during long-term storage or rendering. This results in the formation of undesirable aldehydes and ketones, causing the rendered fat to be less acceptable. Feeding pigs diets enriched with polyunsaturated fatty acids also causes the carcasses to be oily and may lead to production problems such as soft pork bellies. Enrichment of carcass lean trim and fat with oleic acid, which is a solid at typical freezer temperatures, does not appear to accelerate lipid oxidation. The majority of the fatty acids are stored as highly nonpolar triacylglycerols (Figure 4). Triacylglycerols coalesce into the large lipid vacuoles that are the central features of adipocytes. The triacylglycerol structure consists of a glycerol (i.e., threecarbon alcohol) backbone containing three fatty acids in ester linkages. The glycerol primarily is derived from glycerol-3phosphate, which in turn is derived from the metabolism of glucose or, in the liver, from the phosphorylation of free

Encyclopedia of Food and Health

http://dx.doi.org/10.1016/B978-0-12-384947-2.00273-7

Fats: Production and Uses of Animal Fats Desaturation of Fatty Acids

SUBCUTANEOUS PERIRENAL

WEIGHT PERCENTAGE

50

Whereas ruminal hydrogenation of unsaturated fatty acids makes it difficult to enrich beef with unsaturated fatty acids, desaturation of fatty acids available for digestion in the duodenum ensures that the plasma, muscle, and fat depots are not overly enriched with saturated fatty acids, especially stearic acid. The activity of SCD has been demonstrated in the liver, muscle, subcutaneous fat, and intestinal mucosa of pigs and cattle. The greatest SCD activity is located within the subcutaneous fat. However, substantial activity also is observed in intestinal mucosal cells, where it may function to reduce the concentration of stearic acid in chylomicrons and very-low-density lipoproteins produced within mucosal cells. In this manner, SCD activity within mucosal cells largely influences the composition of fatty acids available for deposition in other tissues.

INTRAMUSCULAR

40

30

20

10

0 16:0

18:0

18:1

18:2

BOVINE ADIPOSE TISSUE FATTY ACIDS Figure 1 Fatty acid composition of bovine adipose tissues. Reprinted with permission from Rule, D. C., Smith, S. B. and Romans, J. R. (1995). Fatty acid composition of muscle and adipose tissue of meat animals. In: Smith, S. B. and Smith, D. R. (eds.) Biology of fat in meat animals: current advances, pp. 144–165. Savoy, IL: American Society of Animal Science. SUBCUTANEOUS PERIRENAL

60

INTRAMUSCULAR

WEIGHT PERCENTAGE

INTERMUSCULAR

50

40

30

20

10

0

16:0

18:0

605

18:1

18:2

SWINE ADIPOSE TISSUE FATTY ACIDS Figure 2 Fatty acid composition of porcine adipose tissues. Reprinted with permission from Rule, D. C., Smith, S. B. and Romans, J. R. (1995). Fatty acid composition of muscle and adipose tissue of meat animals. In: Smith, S. B. and Smith, D. R. (eds.) Biology of fat in meat animals: current advances, pp. 144–165. Savoy, IL: American Society of Animal Science.

glycerol by glycerol kinase. In adipose tissue and intestinal mucosal cells, 2-monoacylglycerol (derived from partial hydrolysis of triacylglycerols) provides a portion of the carbon backbone for triacylglycerol synthesis.

Fatty Acid Composition and Melting Points of Lipids Longer-chain fatty acids have higher melting points than shorter-chain fatty acids, and more saturated fatty acids have higher melting points than unsaturated fatty acids. Because of the abundance of stearic and palmitic acids, animal fats are typically solids at room temperature. At 90–100  F, rendered pork fat will separate with a liquid upper layer and solid material in the lower layer. The solid portion is known as stearin (from the Greek for animal fat) and is composed primarily of triacylglycerols enriched with stearic acid. The liquid portion is composed of the glycerol backbone in the form of esters with the less saturated fatty acids, primarily oleic acid, and is known as lard oil or olein (from the Latin for oil). Due to differences in fatty acid composition, the melting points of fat from different species vary. Lamb fat typically has the highest melting point, followed by beef fat and then pork fat. With a high polyunsaturated fat content, poultry fat has the lowest melting point of the four species (80–110  F). Related to the higher internal temperature, the internal fats have higher saturated fatty acid content and correspondingly higher melting points. The fat in the outer layers of the animal body is less saturated with lower melting points. This corresponds to the lower temperature near the body surface and the need to maintain a more liquid state. For example, beef kidney fat has a melting point of 104–122  F, whereas the external subcutaneous fat has a melting point of only 89–110  F. Beef brisket subcutaneous fat has an unusually low melting point (77  F), due to its high concentration of oleic acid. Pork leaf fat has a melting point of 110–118  F; pork back fat has a melting point of 86–104  F. Tallow and lard melting points are standardized as titers, which are the minimum temperature at which fat congeals. For edible tallow, the minimum titer is 105  F, whereas for edible lard, the minimum titer is 100  F. Major differences in fatty acid composition are observed across fat depots of beef carcasses (Figure 5). Brisket fat contains over 6% palmitoleic acid and less than 10% stearic acid, whereas subcutaneous fat overlying the flank contains less than 3% palmitoleic acid and more than 15% stearic acid. The highly consistent, negative relationship between palmitoleic acid and stearic acid demonstrated in several studies indicates that the concentrations of these fatty acids are coordinately

606

Fats: Production and Uses of Animal Fats

Oleic acid 18:1c9

Conjugated linoleic acid 18:2c9, t11

Linoleic acid 18:2c9, c12

Figure 3 Structures of linoleic acid, oleic acid, and the cis-9,trans-11 isomer of conjugated linoleic acid. Large filled circles represent carbon; large shaded circles represent oxygen; small shaded circles represent hydrogen.

controlled by SCD. The concentration of stearic acid in adipose tissue lipids is the primary determinant of lipid melting points (Figure 6), and even lipids extracted from fat depots of the same carcasses can exhibit a remarkable range of melting points. These differences in lipid melting point can have major practical importance.

Procedures for Rendering Fats Early rendering involved the addition of water to the byproducts in an open kettle or the injection of steam into a sealed autoclave. This type of process is referred to as ‘tanking’ and could be used for both edible and inedible products. The fat produced by this method was relatively light in color, but the increased presence of added water resulted in an elevation in the free fatty acid content, reducing the quality and making the product more susceptible to oxidation. Therefore, for economic reasons, dry rendering is most commonly used today.

In the past, each slaughter facility had its own steamjacketed kettle and rendered its own fat. Although the kettle was jacketed in steam, the by-products were cooked in their own juices with no added water, and therefore, this was considered a dry rendering method. Currently, in the United States, there has been a general shift to centralized rendering operations that collect animal by-products and ship them to a central rendering facility. Rendering fat involves applying heat, extracting moisture, and separating the fat. To accomplish this, by-products are first ground into a consistent particle size and then cooked at 115–145  C for up to 90 min. With the heating process, microorganisms such as bacteria and viruses are inactivated (an advantage over other waste product disposal methods). While the by-products are cooked, the jacket steam pressure must be controlled, the mixture agitated, and the temperature monitored for the desired end point. Over the years, there have been advances in the uses of the rendered products and in the rendering methods. Today, more energy-efficient, continuous

Fats: Production and Uses of Animal Fats

607

10

Palmitoleic acid (16:1n-7) g/100 g total fatty acids

9 8 7 6 5 4 3 2 1 0 0

5

10

15

20

25

30

Stearic acid (18:0), g/100 g total fatty acids Figure 5 Palmitoleic acid (16:1n 7) plotted as a function of stearic acid (18:0) in subcutaneous fat. Lipids were extracted from eight subcutaneous fat depots, taken from 50 carcasses of unknown background. Lipids from brisket fat contained the greatest concentration of palmitoleic acid and least stearic acid, whereas lipids from flank fat contained the least palmitoleic acid and the most stearic acid. Reproduced with permission from Turk, S. N. and Smith, S. B. (2009). Carcass fatty acid mapping. Meat Science 81: 658–663.

Figure 4 Typical structure of a triacylglycerol: sn 1 fatty acid (on left), oleic acid; sn 2 fatty acid, palmitic acid; sn 3 fatty acid, linoleic acid. This triacylglycerol would be common in porcine adipose tissue. In bovine and ovine adipose tissues, palmitic acid would occupy the sn 1 position and stearic or oleic acid would occupy the sn 2 position. Large filled circles represent carbon; large shaded circles represent oxygen; small shaded circles represent hydrogen.

processes are used to allow the reuse of process vapors to preheat or dry the materials. Filtering and bleaching systems, as well as refining equipment to remove free fatty acids, may be part of the rendering process. During rendering, the melted fat floats to the top of the unit by virtue of its lesser density, whereas protein and bone solids settle to the bottom of the rendering kettle. Traditionally, the rendered fat then was skimmed from the top of the rendering kettle. This has been replaced with a screw press that draws off the melted fat, and the melted fat is stored and/or transported in tanks.

Chemistry of Rendered Fats Rendered fats consist primarily of triacylglycerols (Figure 4). The characteristics of the rendered fat are determined by the composition of fatty acids attached to the glycerol backbone,

which in turn is determined by the source of the raw materials used in rendering. The most abundant fatty acids of animal fat triacylglycerols are palmitic, stearic, and oleic acids, typically comprising 20–25%, 10–30%, and 30–55%, respectively, of the total fatty acids in lean and fat trim. Significant quantities of linoleic and a-linoleic acids are contained in triacylglycerols, especially in rendered poultry fat. Lipids from ruminant tissues also contain measurable amounts of odd-chained fatty acids, branched-chain fatty acids, trans-fatty acids, and conjugated fatty acids, all of which are products of ruminal fermentation. Lard and tallow also contain free fatty acids (primarily oleic acid), but these must be no more than 0.5% in edible lard and 0.75% in edible tallow. When the rendered fat contains more saturated fatty acids, the fat is referred to as hard fat. It is more solid at room temperature and has a higher melting point due to its high concentration of stearic acid. When the rendered fat contains more unsaturated fatty acids, the fat is referred to as a soft fat, which is not as solid at room temperature due to a greater abundance of oleic acid.

Oxidative Rancidity Rendered fat high in polyunsaturated fatty acids will react chemically with oxygen, resulting in off-flavors and odors

608

Fats: Production and Uses of Animal Fats fat. Tallow is primarily derived from rendered beef fat. Choice white cooking grease is derived from pork fat. Yellow grease (not to be confused with off-colored white cooking grease) is restaurant-quality grease or cooking oil and may be from blended sources. Rendered fat provides concentrated sources of energy for use in feed for poultry, aquaculture, and pets. Other uses of rendered fat include soap, candles, and biodiesel.

50 Brisket Chuck Flank Loin Plate Rib Round Sirloin

45

40

Slip points, ⬚C

Conclusions 35

Fatty acid composition of animal fats varies with animal species, diet, and fat depot. Stearic acid is the primary determinant of lipid melting points, although high concentrations of linoleic acid can effectively reduce melting points. Rendered fats have a great deal of functionality, depending on the source of the fat and the separation techniques. Rendering provides an outlet for animal by-products that would otherwise be unusable. Differential rendering of fat trims that vary in melting points improves the functionality and usefulness of the rendered fat.

30

25

20

15 4

6

8

10

12

14

16

18

20

22

24

26

18:0, g/100 g total fatty acids Figure 6 Lipid melting point (slip point) plotted as a function of stearic acid (18:0) in lipids from eight subcutaneous adipose tissue depots. Lipids from brisket adipose tissue had melting points less than 17  C, whereas lipids from the flank had melting points greater than 45  C. Reproduced with permission from Turk, S. N. and Smith, S. B. (2009). Carcass fatty acid mapping. Meat Science 81: 658–663.

known as rancidity. Oxidative rancidity is catalyzed by the presence of heat, light, salt, and iron, as well as other elements. Because rancidity is not chemically defined or quantifiable, measurements of oxidation products are often used to indicate quality of the product. One such test is the peroxide value test that measures the milliequivalents (me) of peroxide per kilogram fat. A low peroxide value (