240 x 159 /Pantone Rhodamine Red C & 137C
28mm
WOODHEAD PUBLISHING IN FOOD SCIENCE, TECHNOLOGY AND NUTRITION
The first part of the book reviews factors affecting sweet taste perception. It includes chapters on how taste cells respond to sweet taste compounds, genetic differences in sweet taste perception, the influence of tasteodour and taste-ingredient interactions and ways of measuring consumer perceptions of sweet taste. Part II discusses the main types of sweet-tasting compounds: sucrose, polyols, low-calorie and reduced-calorie sweeteners. The final part of the book looks at ways of improving the use of sweettasting compounds, including the range of strategies for developing new natural sweeteners, improving sweetener taste, optimising synergies in sweetener blends and improving the use of bulk sweeteners. With its distinguished editor and international team of contributors, Optimising sweet taste in foods will be a standard reference for the food industry in improving low-fat and other foods.
Optimising sweet taste in foods
A sweet taste is often a critical component in a consumer’s sensory evaluation of a food product. This book summarises key research on what determines consumer perceptions of sweet taste, the range of sweet-tasting compounds and the ways in which their use in foods can be optimised.
WOODHEAD PUBLISHING IN FOOD SCIENCE, TECHNOLOGY AND NUTRITION
Professor William J. Spillane works within the Chemistry Department of the National University of Ireland, Galway. He has published widely on non-nutritive sweeteners such as sulfamates as well as other aspects of sweet taste quality.
9 780849 376030
Woodhead Publishing
ISBN-13:978-948-1-84569-0 ISBN-10:1-84569-008-7
9 789481 845690
CRC Press LLC 6000 Broken Sound Parkway, NW Suite 300 Boca Raton FL 33487 USA CRC order number WP7603 CRC Press ISBN-10: 0-8493-7603-3
9 780849 376030 ISBN-13:978-948-1-84569-0 ISBN-10:1-84569-008-7
9 789481 845690
Spillane
Woodhead Publishing Ltd Abington Hall Abington ISBN-10: 0-8493-7603-3 Cambridge CB1 6AH England www.woodheadpublishing.com
Optimising sweet taste in foods Edited by W. J. Spillane
Optimising sweet taste in foods
Related titles: Flavour in food (ISBN-13: 978-1-85573-960-4; ISBN-10: 1-85573-960-7) The first part of the book reviews ways of measuring flavour. Part II looks at the ways flavour is retained and released in food. It considers the way flavour is retained in particular food matrices, how flavour is released during the process of eating, and the range of influences governing how flavour is perceived by the consumer. Flavour in food guides the reader through a complex subject and provides the essential foundation in both understanding and controlling food flavour. Improving the fat content of foods (ISBN-13: 978-1-85573-965-9; ISBN-10: 1-85573-965-8) Dietary fats have long been recognised as having a major impact on health: negative in the case of consumers' excessive intake of saturated fatty acids, positive in the case of increasing consumers' intake of long-chain n-3 polyunsaturated fatty acids (PUFAs). However, progress in ensuring that consumers achieve a nutritionally-optimal fat intake has been slow. This important collection reviews the range of steps needed to improve the fat content of foods whilst maintaining sensory quality. Texture in food, Volume 1: Semi-solid foods (ISBN-13: 978-1-85573-673-3; ISBN-10: 1-85573-673-X) Understanding and controlling the texture of semi-solid foods such as yoghurt and ice cream is a complex process. With a distinguished international team of contributors, this important collection summarises some of the most significant research in this area. The first part of the book looks at the behaviour of gels and emulsions, how they can be measured and their textural properties improved. The second part of the collection discusses the control of texture in particular foods such as yoghurt, ice cream, spreads and sauces. Texture in food, Volume 2: Solid foods (ISBN-13: 978-1-85573-724-2; ISBN-10: 1-85573-724-8) With its distinguished editor and international team of contributors, this authoritative book summarises the wealth of recent research on what influences texture in solid foods and how it can be controlled to maximise product quality. The first part of the book reviews research on understanding how consumers experience texture, and how they perceive key textural qualities. Part II considers the instrumental techniques used for analysing texture. The final part examines how the texture of particular foods may be better understood and improved. Details of these books and a complete list of Woodhead titles can be obtained by: · visiting our website at www.woodheadpublishing.com · contacting Customer Services (email:
[email protected]; fax: +44 (0) 1223 893694; tel.: +44 (0) 1223 891358 ext. 30; address: Woodhead Publishing Limited, Abington Hall, Abington, Cambridge CB1 6AH, England)
Optimising sweet taste in foods Edited by William J. Spillane
Published by Woodhead Publishing Limited Abington Hall, Abington Cambridge CB1 6AH England www.woodheadpublishing.com Published in North America by CRC Press LLC 6000 Broken Sound Parkway, NW Suite 300 Boca Raton, FL 33487 USA First published 2006, Woodhead Publishing Limited and CRC Press LLC ß 2006, Woodhead Publishing Limited The authors have asserted their moral rights. This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. Reasonable efforts have been made to publish reliable data and information, but the authors and the publishers cannot assume responsibility for the validity of all materials. Neither the authors nor the publishers, nor anyone else associated with this publication, shall be liable for any loss, damage or liability directly or indirectly caused or alleged to be caused by this book.
Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming and recording, or by any information storage or retrieval system, without permission in writing from Woodhead Publishing Limited. The consent of Woodhead Publishing Limited does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from Woodhead Publishing Limited for such copying.
Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress. Woodhead Publishing Limited ISBN-13: Woodhead Publishing Limited ISBN-10: Woodhead Publishing Limited ISBN-13: Woodhead Publishing Limited ISBN-10: CRC Press ISBN-10: 0-8493-7603-3 CRC Press order number: WP7603
978-1-84569-008-3 (book) 1-84569-008-7 (book) 978-1-84569-164-6 (e-book) 1-84569-164-4 (e-book)
The publishers' policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp which is processed using acid-free and elementary chlorine-free practices. Furthermore, the publishers ensure that the text paper and cover board used have met acceptable environmental accreditation standards. Project managed by Macfarlane Production Services, Dunstable, Bedfordshire (e-mail:
[email protected]) Typeset by Godiva Publishing Services Ltd, Coventry, West Midlands Printed by TJ International Limited, Padstow, Cornwall, England
Contents
Contributor contact details . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xi
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . W. J. Spillane, National University of Ireland ± Galway, Ireland
xv
Part I 1
Factors affecting sweet taste perception
Stimulation of taste cells by sweet taste compounds . . . . . . . . . . . . M. Naim, The Hebrew University of Jerusalem, Israel, L. Huang, Monell Chemical Senses Center, USA, A. I. Spielman, College of Dentistry, New York University, USA and M. E. Shaul and A. Aliluiko, The Hebrew University of Jerusalem, Israel 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Peripheral organization of taste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 The sweet taste receptor(s) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 Downstream signaling components . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Putative cellular mechanism(s) for the signal termination of sweet taste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3
3 6 7 14 17 23 23 24
vi
Contents
2 Genetic differences in sweet taste perception . . . . . . . . . . . . . . . . . . . V. B. Duffy, J. E. Hayes and M. E. Dinehart, University of Connecticut, USA 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 PTC/PROP tasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Sweetness: relation to PROP, fungiform papillae density, T2R38 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Relating genetic taste markers with dietary sweet behaviors . 2.5 Sweet liking: relation to PROP, fungiform papillae density, T2R38 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 Variation in other bitter markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7 SAC Gene: human homolog and hedonic response to sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.8 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Children's liking of sweet tastes and its biological basis . . . . . . . . M. Y. Pepino and J. A. Mennella, Monell Chemical Senses Center, USA 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Ontogeny of sweet taste preferences: from fetal life to adolescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Role of experiential factors on sweet taste preferences . . . . . . 3.4 Physiological properties of sweet tastes . . . . . . . . . . . . . . . . . . . . . . 3.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Taste±odour interactions in sweet taste perception . . . . . . . . . . . . . D. Valentin, C. Chrea and D. H. Nguyen, CSEG, France 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Overview of the literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Factors affecting taste±odour interactions . . . . . . . . . . . . . . . . . . . . 4.4 Mechanisms of taste±odour interactions . . . . . . . . . . . . . . . . . . . . . . 4.5 Implications for food product development . . . . . . . . . . . . . . . . . . 4.6 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.7 Sources of further information and advice. . . . . . . . . . . . . . . . . . . . 4.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Taste±ingredient interactions modulating sweetness . . . . . . . . . . . . M. Lindley, Lindley Consulting, UK 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Interactions between sweeteners and bitter compounds . . . . . . 5.3 Interactions between sweeteners and acids . . . . . . . . . . . . . . . . . . . 5.4 Interactions of sweet and salty compounds . . . . . . . . . . . . . . . . . . .
30 30 36 37 41 42 45 46 49 49 54 54 55 58 59 60 61 61 66 66 67 69 75 79 80 81 81 85 85 86 88 92
Contents 5.5 5.6 5.7
Interactions of sweet compounds with other sweet compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6 Measuring consumers' perceptions of sweet taste . . . . . . . . . . . . . . S. Issanchou and S. Nicklaus, INRA, France 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods to determine consumers' perceptions of sweet taste . 6.2 6.3 Overview of consumers' perception of sweetness . . . . . . . . . . . . 6.4 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5 Sources of further information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Part II
vii 92 94 94 97 97 97 112 121 122 122
Types of sweet tasting compounds
7 Sucrose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J. M. Cooper, British Sugar plc, UK 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Sugar manufacture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3 Sugar products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4 Properties of sugar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5 Sugar functionality in food products . . . . . . . . . . . . . . . . . . . . . . . . . 7.6 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.7 Sources of further information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.8 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.9 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Polyols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . M. E. Embuscado, McCormick & Company, Inc, USA 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 Types of polyols, chemical structures and their manufacture 8.3 Physicochemical and functional properties . . . . . . . . . . . . . . . . . . . 8.4 Physiological properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.5 Advantages and disadvantages of using polyols . . . . . . . . . . . . . . 8.6 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7 Mixed sweetener potential of polyols . . . . . . . . . . . . . . . . . . . . . . . . 8.8 Regulatory status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.9 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.10 Sources of further information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.11 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Low-calorie sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . S. E. Kemp, Kemps Research Solutions Ltd, UK 9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
135 135 137 140 142 145 149 150 151 151 153 153 154 156 160 162 163 166 168 170 171 171 175 175
viii
Contents 9.2 9.3 9.4 9.5 9.6 9.7 9.8 9.9
Low-calorie sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Health-related developments in low-calorie sweeteners . . . . . . Market-related developments in low-calorie sweeteners . . . . . . Implications for food product design . . . . . . . . . . . . . . . . . . . . . . . . . Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sources of further information and advice . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
176 216 222 230 235 239 240 240
10 Reduced-calorie sweeteners and caloric alternatives . . . . . . . . . . . . G-W. von Rymon Lipinski, Consultant, Germany 10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2 Reduced-calorie sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3 Alternative caloric sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.4 Established caloric alternatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.5 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.6 Sources of further information and advice . . . . . . . . . . . . . . . . . . . 10.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
252
Part III 11
252 254 259 266 275 276 276
Improving sweet tasting compounds and optimising their use in foods
Analysing and predicting properties of sweet-tasting compounds D. E. Walters, Rosalind Franklin University of Medicine and Science, USA 11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.2 QSAR models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.3 Pharmacophore models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.4 Receptor models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.5 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.6 Sources of further information and advice . . . . . . . . . . . . . . . . . . . 11.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12 Discovering new natural sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. D. Kinghorn, The Ohio State University, USA and N.-C. Kim, Lovelace Respiratory Research Institute, USA 12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.2 Commercially used natural sweeteners . . . . . . . . . . . . . . . . . . . . . . . 12.3 Approaches to natural sweetener discovery . . . . . . . . . . . . . . . . . . 12.4 Improvement of sweet taste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.5 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.6 Sources of further information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
283 283 284 285 287 288 288 288 292 292 296 298 300 301 302 303
Contents 13 Molecular design and the development of new sweeteners . . . . . J. Polanski, University of Silesia, Poland 13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2 The historical development of sweetness consumption . . . . . . . 13.3 Commercial alternative sweetener discoveries . . . . . . . . . . . . . . . 13.4 Molecular design ± novel approaches novel chances . . . . . . . . . 13.5 Screening and visualising new sweeteners candidates . . . . . . . . 13.6 Molecular design in commercial sweetener development . . . . 13.7 From discovery to commercial products . . . . . . . . . . . . . . . . . . . . . 13.8 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.9 Sources of further information and advice . . . . . . . . . . . . . . . . . . . 13.10 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Developing new sweeteners from natural compounds . . . . . . . . . . A. Bassoli, DISMA, University of Milan, Italy 14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.2 Importance of developing new sweeteners from natural compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3 Methods of designing new sweeteners from natural compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.4 Synthesising new sweeteners from natural compounds . . . . . . . 14.5 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Improving the taste of sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. E. Walters, Rosalind Franklin University of Medicine and Science, USA 15.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15.2 Blending sweeteners to provide synergy . . . . . . . . . . . . . . . . . . . . . 15.3 Blending sweeteners to improve taste profile . . . . . . . . . . . . . . . . 15.4 Blending sweeteners to improve temporal profile . . . . . . . . . . . . 15.5 Using additives to improve taste quality of sweeteners . . . . . . 15.6 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15.7 Sources of further information and advice . . . . . . . . . . . . . . . . . . . 15.8 Reference list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Analysing and predicting synergy in sweetener blends . . . . . . . . . P. Laffort, Centre des Sciences du GouÃt, France 16.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16.2 Sweet taste measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16.3 Response of a mixture vs. responses to its components: the perceptive models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16.4 Response of a mixture vs. concentrations of its components: the ÿ (Gamma) family models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16.5 Other interaction models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ix 307 307 308 309 310 313 317 318 320 321 322 327 327 327 329 336 340 341 344 344 344 345 346 346 347 347 347 349 349 350 356 361 366
x
Contents 16.6 16.7 16.8 16.9
17
18
Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix: precise technical details . . . . . . . . . . . . . . . . . . . . . . . . . .
369 371 371 373
Bulk sweet tasting compounds in food product development . . . M. Lindley, Lindley Consulting, UK 17.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.2 Characteristics of bulk sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.3 Sensory properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.4 Physical functionalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.5 Bulk sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17.7 Sources of further information and advice . . . . . . . . . . . . . . . . . . . 17.8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
375
Hydrocolloid±sweetener interactions in food products . . . . . . . . . . D. Cook, The University of Nottingham, UK 18.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.2 Hydrocolloid±sweetener interactions . . . . . . . . . . . . . . . . . . . . . . . . . 18.3 Mechanisms of hydrocolloid±sweetener interactions . . . . . . . . . 18.4 Implications for food product development . . . . . . . . . . . . . . . . . . 18.5 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18.6 Sources of further information and advice . . . . . . . . . . . . . . . . . . . 18.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
19
375 376 377 380 383 385 386 386 388 388 390 394 396 399 400 401
Future directions: using biotechnology to discover new sweeteners, bitter blockers and sweetness potentiators . . . . . . . . . R. McGregor Linguagen Corp., USA 19.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19.2 Understanding taste at the molecular level enables discovery of novel sweeteners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19.3 The use of bitter blockers to improve sweet taste . . . . . . . . . . . . 19.4 Future trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19.5 Sources of further information and advice . . . . . . . . . . . . . . . . . . . 19.6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
406 410 410 411 412
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
415
404 404
Contributor contact details
(* = main contact)
Editor Professor William J. Spillane Chemistry Department National University of Ireland Galway Ireland Tel: 353 91 492475 Fax: 353 91 525700 Email:
[email protected]
PO Box 12, Rehovot 76-100 Israel Tel: 972-8-9489276 Fax: 972-8-9476189 E-mail:
[email protected] E-mail:
[email protected] E-mail:
[email protected] Dr Liquan Huang Monell Chemical Senses Center 3500 Market Str. Philadelphia, PA 19104 USA E-mail:
[email protected]
Chapter 1 Professor Michael Naim,* Mrs Merav E. Shaul and Mr Alexander Aliluiko Institute of Biochemistry, Food Science and Nutrition Faculty of Agricultural, Food and Environmental Quality Sciences The Hebrew University of Jerusalem
Professor Andrew I. Spielman New York University College of Dentistry 345E 24th Str. New York, NY 10012 USA E-mail:
[email protected]
xii
Contributors
Chapter 2
Chapters 5 and 17
Valerie B. Duffy,* J. E. Hayes and M. E. Dinehart School of Allied Health, University of Connecticut 358 Mansfield Road, Box U-101 Storrs, CT 06269-2101 USA
Dr Michael G. Lindley Lindley Consulting 17 Highway Crowthorne Berkshire RG45 6HE UK
E-mail:
[email protected]
Tel: 01344 771015 Email:
[email protected]
Chapter 3
Chapter 6
Professor Julie Mennella* Member and Director of Education Outreach Monell Chemical Senses Center 350 Market Street Philadelphia, PA 19104-3308 USA
Sylvie Issanchou and Sophie Nicklaus* INRA FLAVIC (UMR INRA-ENESAD) 17 rue Sully BP 86510 21 065 Dijon Cedex France
Tel: +01215 898 9230 E-mail:
[email protected] Yanina Pepino Instituto de InvestigacioÂn MeÂdica Mercedes y MartõÂn Ferreyra CoÂrdoba Argentina
Chapter 4 Dominique Valentin,* C. Chrea and D. H. Nguyen CSEG Campus Universitaire 15 rue Hugues Picardet 21000 Dijon France Tel/Fax: 03 80 68 16 57 E-mail:
[email protected]
Email:
[email protected];
[email protected]
Chapter 7 Dr Julian M. Cooper Head of Food Science British Sugar plc Oundle Road Peterborough PE2 9QU UK Tel: +44 (0)1733 422478 E-mail:
[email protected]
Contributors
Chapter 8 Dr Milda E. Embuscado McCormick & Company, Inc. 204 Wight Avenue Hunt Valley, MD 21031-1501 USA Tel: 410-527-6009 Fax: 410-527-6527 E-mail:
[email protected]
Chapter 9 Dr Sarah E. Kemp Kemps Research Solutions Ltd 10 Overend Close Bradwell Village Milton Keynes MK13 9EJ UK Tel: +44 (0)1908 227575 E-mail:
[email protected]
Chapter 10 Professor Gert von Rymon Lipinski Schlesienstraûe 62 65824 Schwalbach Germany Tel: +49 6196 9023651 Fax: +49 6196 9023652 E-mail:
[email protected]
Chapters 11 and 15 D. Eric Walters Department of Biochemistry and Molecular Biology The Chicago Medical School Rosalind Franklin University of Medicine and Science
xiii
3333 Green Bay Road North Chicago, IL 60064 USA Tel: 847-578-8613 Fax: 847-578-3240 E-mail:
[email protected]
Chapter 12 Professor A. Douglas Kinghorn Jack L. Beal Professor and Chair College of Pharmacy The Ohio State University 500 West 12th Avenue Columbus, OH 43210-1291 USA E-mail:
[email protected]
Chapter 13 Professor Jaroslaw Polanski University of Silesia Institute of Chemistry ul. Szkolna 9 PL-40-006 Katowice Poland E-mail:
[email protected]
Chapter 14 Dr Angela Bassoli DISMA Dipartimento di Scienze Molecolari Agroalimentari UniversitaÁ di Milano Via Celoria 2 I-20133 Milano Italy Tel: +39 02 5031 6815 Fax: +39 02 5031 6801 E-mail:
[email protected]
xiv
Contributors
Chapter 16
Chapter 19
Dr Paul Laffort CESG-CNRS 15 Rue Hugues Picardet 21000 Dijon France
Richard McGregor Director, Technology Assessment Linguagen Corp. 2005 Eastpark Boulevard Cranbury, NJ 08512-3515 USA
E-mail:
[email protected]
Chapter 18 Dr David Cook Division of Food Sciences University of Nottingham Sutton Bonington Campus Loughborough Leics LE12 5RD UK Tel: 0115 9516245 Fax: 0115 9516162 E-mail:
[email protected]
Tel: 609 860 1500 Fax: 609 860 5900 E-mail:
[email protected]
Introduction W. J. Spillane, National University of Ireland ± Galway, Ireland
The `sweetness' industry is a multimillion pound growth industry worldwide especially in the developed world. In 2004 over 2,000 reduced sugar and sugar free foods were introduced. In academe and in industrial laboratories there are numerous groups active in taste studies of many types and this is reflected in the volume of diverse scientific publications, patents etc. that have appeared. There have been quite a number of books and conferences on sweetness and taste and some of these meetings take place at regular intervals. It is a few years since a book has appeared that addresses the topic of sweetness and in this present volume, which we hope will be a timely contribution to the field, we have brought together an array of experts in the sweetness area. Each is an authority in his/her own area and in inviting scientists to contribute we have tried to cover areas that have not been touched on in recent times. We hope that this will make this volume of particular interest to a variety of researchers in the taste domain and to a wider circle of scientists who may wish to have an update in various sweetness fields. The book is in three sections: Part I deals with factors affecting sweet taste perception, Part II examines some types of sweet tasting compounds and Part III looks at improving sweet tasting compounds and optimising their use in foods. In Part I Chapter 1 has an important and timely discussion on the stimulation of taste cells by sweet tasting compounds including a useful resume on the stimulation of the TIR1, TIR2 and TIR3 sweet taste receptors. The intriguing subject of genetic differences in sweet taste perception is addressed in Chapter 2. Chapter 3 highlights the liking of sweet taste by children from infancy to adolescence. The surprising role of taste±odour interactions in sweet taste perception is assessed in Chapter 4. Taste±ingredient interactions that are responsible for modulating sweetness are considered in Chapter 5. Finally in this
xvi
Introduction
section of the book the sensory evaluation and factors affecting it of consumers' perceptions of sweet taste are focused on in Chapter 6. Some classes of sweet tasting compounds are described in Part II. Starting in Chapter 7 with the key sweetener sucrose emphasis had been put on its manufacture, processing, products, properties and myriad uses. The three general types of polyols (hydrogenated mono- and di-saccharides and mixtures of hydrogenated saccharides/polysaccharides) are dealt with in Chapter 8. The advantages and disadvantages of using polyols are weighed up and the applications and potential of these important sweeteners are looked at. One of the longer chapters in the book (Chapter 9) covers developments in the substantial area of low calorie sweeteners looking particularly at topics such as regulatory position and safety, health trends, their role in weight control, diabetes, dental caries, market economics and other important aspects of their wide usage. This part of the book concludes with Chapter 10 on reduced calorie sweeteners and caloric alternatives. This chapter includes a lengthy review of D-tagatose, the L-sugars and discussion of alternative caloric sweeteners, such as isomaltulose, trehalose, hydrolysed lactose and galactose, leucrose and starch hydrolysates among other types. In Part III the focus has been on the improvement of sweet tasting compounds and optimising their use. In Chapter 11 attention is given to analysing and predicting sweetness in which structure-taste relationships both qualitative and quantitative are reviewed and pharmacophore and receptor models are discussed. The discovery of new natural sweeteners is the thrust of Chapter 12. In it the authors give information on commercially used natural sweeteners, they detail the approaches to the discovery of such compounds and they show how improvements in sweet taste can be achieved using natural sweeteners. In Chapter 13 the molecular design and development of new sweeteners is discussed. The author looks at commercial alternative sweetener discoveries, molecular design, screening and visualising new sweetener candidates and the movement to commercial products. Chapter 14 outlines the development of new sweeteners from natural compounds. Methods of designing new sweeteners from natural compounds using QSARs and modelling and the synthesis of new sweeteners from natural compounds are addressed. Improving the taste of sweeteners through blending to provide better synergy, better taste profiles and better quality is examined in Chapter 15. In the next chapter synergy is the key topic and the author analyses and predicts synergy in various sweetener blends by examining a number of models available for synergistic analyses. Chapter 17 provides an interesting overview of bulk sweet tasting compounds in food product development. In this chapter many aspects of bulk sweeteners are delved into including flavour, sweetness, texture, appearance and physical properties. Hydrocolloid sweetener interactions are examined in Chapter 18 including discussions of the mechanisms of sweetener-hydrocolloid interactions and criteria for hydrocolloid selection in food product development. The use of biotechnology to discover new sweeteners, bitter blockers and sweetness potentiators is discussed as a future trend in Chapter 19.
Part I Factors affecting sweet taste perception
1 Stimulation of taste cells by sweet taste compounds M. Naim, The Hebrew University of Jerusalem, Israel, Liquan Huang, Monell Chemical Senses Center, USA, A. I. Spielman, College of Dentistry, New York University, USA and M. E. Shaul and A. Aliluiko, The Hebrew University of Jerusalem, Israel
1.1
Introduction
Perception of sweet sensation is known to be associated in humans and many mammals with a unique pleasure. Studies examining facial expressions of neonates following taste stimulation indicate appealing responses to sweet taste and aversion to bitter taste (Steiner, 1973), and newborns prefer sugar solutions to water and their preference increases with sugar concentration (Desor et al., 1973). Furthermore, their preference for different sugars corresponds to adult perception (Maller and Desor, 1974). These studies suggest that the response to sweet taste is innate. The high threshold for the appealing taste of sugar and actual preference (Cagan and Maller, 1974) lead mammals to select high-caloric foods. In contrast, as a counter defense and screening mechanism, a strong rejection of bitter stimuli is due to the fact that traditionally most bitter compounds were toxic and therefore are perceived at much lower concentrations than sweet compounds. These innate behavioral mechanisms have apparently had a significant effect on survival during phylogenetic development. Moreover, it has long been recognized that the preference for sugars may change with physiological state. Human subjects prefer a 5% sucrose solution to 30% sucrose solution, but can be induced to prefer the more concentrated solution once blood glucose level is decreased to 50 mg% following insulin injection (Mayer-Gross and Walker, 1946). Similar results are found in rats (Jacobs, 1958). A strong preference for sweet high-fat foods and fat is associated with opioids (MarksKaufman and Kanarek, 1981; Marks-Kaufman and Lipeles, 1982; Drewnowski
4
Optimising sweet taste in foods
et al., 1992; Drewnowski, 1991) which appear to be released under mild stress. Administration of naloxone, an opioid antagonist, reduces the hedonic preference for sugar (Drewnowski et al., 1995). Other studies have shown that oral stimulation by sweet taste substances may, via the cephalic phase of endocrine secretion, release insulin, which in turn may participate in stimulating eating (Brand et al., 1982). The intake of food items containing refined sugars and high caloric density constituents has increased significantly in the last century (Grand, 1974), and has been linked to metabolic disorders such as diabetes and obesity, termed `diabesity' (Shafrir, 1997), considered to be the main threats to human health in the 21st century (Zimmet et al., 2001). In response to such health hazards, during the last four decades many chemical studies were initiated to explore low-calorie alternative sweeteners with high sweet potency (DuBois et al., 1981; DuBois et al., 1993; Nofre and Tinti, 1987; Nofre et al., 1988; Tinti and Nofre, 1996). The resulting synthetic and some natural non-sugar sweeteners include a large collection of diverse compounds such as sulfamates, flavonoids, oximes, amino acids, peptides, proteins, guanidines and terpenoids. These discoveries were based on structure±function relationship studies which have essentially proposed a single sweet taste receptor for all sweeteners (Shallenberger and Acree, 1967; Kier, 1972; van der Heijden et al., 1985a, 1985b). Tinti and Nofre have prepared extremely potent sweeteners and proposed the multipoint attachment (MPA) model of sweetness (Tinti and Nofre, 1996; Nofre and Tinti, 1996). This model was in line with the notion of a single sweet taste receptor but with multiple binding sites between the high potency sweeteners and the receptor molecule. The nutritional justification of alternative sweeteners to sugars has led to increased consumption of low-calorie soft-drinks, creating a multi-billion dollar economic target for the food industry. However, the sweet taste of sugars, especially that of sucrose, is regarded as pure with optimal sensation in humans, whereas many non-sugar sweeteners possess inferior sweet quality. Psychophysical sensory studies in humans using the multi-dimensional similarity (MDS) analysis (Schiffman et al., 1979) clearly showed that the sweetness of a variety of non-sugar sweeteners may be located at a different site in the MDS sweet similarity map from that of sugar sweeteners. Cross-adaptation studies of sweeteners (Schiffman et al., 1981; Lawless and Stevens, 1983) have suggested for a long time that either different sweet taste receptors or different mechanisms are involved for sweet sensation. A further difference between non-sugar sweeteners and sugars is a phenomenon known as synergism. Sweet taste synergism occurs mainly between pairs of non-sugar sweeteners but not between pairs of sugars (Ayya and Lawless, 1992; DuBois, 2004; Schiffman et al., 1995). Moreover, in most cases, stimulation by non-sugar sweeteners does not lead to the maximal sweet taste intensity level achieved with sugars during psychophysical experiments, even though these sweeteners are more potent when intensity is evaluated on a weight basis (Moskowitz, 1977; DuBois et al., 1991). Sugars and sugar alcohols show linear concentration±response relationships while high-potency sweeteners yield hyperbolic dependence. According to
Stimulation of taste cells by sweet taste compounds
5
Moskowitz (1977), it may be related to the fact that as one increases the concentration of the non-sugar sweeteners, the additional taste qualities that such sweeteners possess become predominant and do not permit a further increase in sweetness intensity as occurs for sugars. In fact, it was suggested that the concentration-intensity slope be used as an additional parameter for evaluation of the sweet taste quality of a sweetener (Moskowitz, 1977). It is apparent that there are multiple reasons that may be related for the above dissimilarity in sweet taste between sugars and non-sugar sweeteners. It is well known that many (though not all) non-sugar sweeteners possess additional taste sensations such as bitter, metallic and licorice. Although the molecular basis for such sensations is not clear, the sweeteners saccharin and acesulfam K were recently found to stimulate both bitter and sweet taste receptors (Kuhn et al., 2004) (see more discussion on sweet taste receptors below). A further distinguishing factor that makes clearly non-sugar sweeteners inferior to sugars is their temporal properties. Time±intensity relationship studies indicated that, compared to sucrose, it takes a longer time for the sensation of a non-sugar sweetener to reach a maximal sweet taste intensity (slow taste onset), and more time (sometimes minutes) for the sweetness to be extinguished (lingering aftertaste) (Birch et al., 1980; Larson-Powers and Pangborn, 1978; Naim et al., 1986; DuBois et al., 1977). This lingering phenomenon is also known to occur for a variety of bitter stimuli. Interestingly, sweeteners which produce the sweet aftertaste may also induce `sweet water aftertaste', a phenomenon where water tastes sweet for several minutes after tasting a sweetener (van der Wel, 1972; Bartoshuk, 1987; Naim et al., 1986). This is also true for stimulation by some sweetener receptor antagonists (D'Angelo and DuBois, 1999). The delay in sweet taste termination following the tasting of nonsugar sweeteners is not unique to humans. It was also shown to occur in old world monkeys during behavioral studies (e.g., baboons, (Rogatka et al., 1985)). The mechanism of lingering taste is unknown. One question is whether such a delay is due to peripheral or central nervous system events. For instance, in olfaction, the memory of odors can persist, e.g., in the neuronal tissue of insects, even as cells go through metamorphosis (Ray, 1999). In the brain, taste sensation which is associated with malaise may form long-term conditioned taste aversion through memory which depends on identified biochemical mechanisms (Berman and Dudai, 2001). However, electrophysiological recordings of taste nerves at the periphery (Hellekant, 1994) have indicated `lingering aftertaste' in response to stimulation by non-sugar sweeteners. Furthermore, time-course measurements have shown that the transient onset and termination of IP3 release in taste cells following stimulation by some bitter tastants known to possess lingering aftertaste is delayed beyond 500 ms (Spielman et al., 2002). These results suggest that the delay in taste-signal termination induced by some non-sugar sweeteners and bitter tastants is at the periphery, at the taste-cell level. To date, the molecular basis for the `slow taste onset' and the `lingering aftertaste' phenomena is not known, even though it has significant implications regarding the acceptance of a variety of food products.
6
Optimising sweet taste in foods
In the current chapter we will discuss the complex signaling which leads to sweet taste sensation events occurring at the level of the taste bud cells. Furthermore, we will analyze current experimental data assumed to provide better understanding of biochemical events related to sweet taste quality.
1.2
Peripheral organization of taste
The peripheral gustatory system is exposed to a variety of physical, chemical and biological insults. Extreme hot, cold, irritating, acidic and non-sterile stimuli may have damaging effects on peripheral taste receptors. Therefore, unlike all other sensory systems that have peripheral neuronal receptors, the peripheral gustatory system has evolved to have receptor cells that are rapidly-renewing, specialized epithelial cells. This adaptation ensures rapid regeneration of taste receptor cells that become damaged. The peripheral organ in the gustatory system is comprised, in decreasing hierarchical order of: taste papilla, taste bud and taste cell. The largest is the taste papilla (Spielman and Brand, 1997). Visible to the naked eye, taste papillae in humans come in at least four different shapes and are located on the tongue, soft and hard palate, pharynx, epiglottis and larynx. The largest of the four types, the circumvallate (CV) papillae, are located on the dorsal surface of the tongue between the anterior 2/3 and posterior 1/3 along a V shaped line. The number of CV papillae varies from species to species. There are between three and 13 papillae in humans. In other animals, their number varies from just one in rodents to 25 in cows. On the postero-lateral border of the tongue are the foliate papillae, each of which is a pocket-shaped invagination lined with taste buds. The invaginations protect the taste buds from direct physical damage. The third type of papillae, the mushroom shaped fungiform papillae, cover the anterior dorsal surface of the tongue. Compared with other gustatory papillae, they are the largest in number with 50±200 in humans. Finally, the extralingual papillae are located on the soft and hard palate, the pharynx, epiglottis and larynx, with the most prominent being the taste stripes (or Geschmackstreifen from the original German term), located on both sides of the palatal midline at the transition of the soft and hard palate. Interestingly, the most abundant and recognizable structures, the filiform papillae are nongustatory. When overgrown and stained by beverages and food dye these threadlike papillae tend to provide a characteristic colour to the tongue. Similarly, excessive shedding of the papillae provides a `white coat' on the tongue. On the other hand, loss of the filiform papillae creates a dry and smooth appearance which is associated with a variety of oral and systemic conditions such as dry mouth, anaemia and Scarlet fever. The filiform papillae act like `VelcroÕ' to help secure food to the tongue so that the bolus can be moved around the oral cavity (Spielman, 1998). Within each taste papilla there are varying numbers of taste buds visible under light and electron microscopes. The number of taste buds varies from one
Stimulation of taste cells by sweet taste compounds
7
papilla to the next. In humans, fungiform papillae contain 1±10, CV papillae contain 100±200, while foliate papillae may have from a few hundred to a few thousand buds. The taste bud is the functional unit of the peripheral taste organ. It is onionshaped and contains 50±100 continuously-maturing taste receptor and supporting cells. Most of the taste cells in a bud are shielded from the oral cavity. Only the apical tip of a few taste cells contains taste receptor proteins that is exposed to the oral cavity through a 3±5 micron wide opening, referred to as the taste pore. Unlike any other sensory system, taste cells have a rapid turnover rate of 10.5 days, which is about twice as fast as the surrounding epithelial cells. The progenitor cells, also known as basal cells, are located in an epithelial layer at the base of the taste bud that corresponds to the germinal layer of the epithelium. As taste receptor cells continuously grow and mature, they migrate from the basal area of the bud toward the taste pore over a period of about 10 days, its turnover time. Exposure of an individual receptor cell to taste stimulants is limited usually to a single meal. Within a few hours of a chemosensory experience, the exposed taste receptor cells are shed into the oral cavity and the dead cells are washed away by saliva. The fungiform- and taste-stripe-containing taste buds are bathed in saliva from the bilateral major salivary glands (parotid, submandibular, sublingual). In contrast, taste buds in the circumvallate and foliate taste papillae are washed primarily by the saliva of the von Ebner's salivary glands located in the body of the tongue with openings into the trenches that surround these papillae. Within the taste bud, several taste receptor cells, certainly not all, will synapse with first order neurons belonging to either the facial (chorda tympani), glossopharyngeal or vagus nerves. In a recent study synapses were observed in a few taste receptor cells (Yang et al., 2000). In another study, the specific receptor cells, called Type II, associated with bitter, sweet and umami taste transduction was devoid of conventional synapse (Clapp et al., 2004). These cells either communicate first with adjacent taste cells in the bud before synapsing, or they have subsurface cysternae of smooth endoplasmic reticulum instead of traditional synapses (Clapp et al., 2004). The sensory pathway for taste involves a chain of neurons in series. First order neurons connect the periphery to the nucleus of the solitary tract in the central nervous system where they synapse second order neurons, cross the midline and connect with the thalamus on the opposite side, while the third order neurons connect the thalamus to the cortex. The sensory pathway for taste involves neural connections with the salivary nuclei to activate a reflex pathway and the perception of taste depends on the signals reaching the sensory cortex.
1.3
The sweet taste receptor(s)
1.3.1 Sweet taste is mediated by G protein coupled receptors Earlier biochemical studies indicated that sweet taste is detected by proteinaceous membrane receptors that are located on the tip of taste receptor
8
Optimising sweet taste in foods
cells. Proteolytic (pronase) treatment of the surface of human tongue could temporarily remove sweet perception, indicating that the sweet receptors are the membrane-bound surface proteins (Hiji, 1975). Radiolabeled sugars and sweeteners such as sucrose, fructose, glucose, synthetic sweeteners (cyclamate and saccharin) and sweet-tasting proteins monellin and thaumatin could bind to the purified homogenate fractions from human and bovine papillae. This binding activity was taste papillae-dependent, saturable and reversible, could be competitively inhibited by unlabeled sugars and sweeteners, indicating that there were specific receptors on taste bud cells that could reversibly bind to sweet compounds (Cagan, 1971; Lum and Henkin, 1976; Cagan and Morris, 1979). Additionally, photoaffinity labeling demonstrated that the intensely sweettasting protein thaumatin I could bind to a protein of approximately Mr 50,000 from monkey circumvallate papillae (Shimazaki et al., 1986). This finding was further corroborated in other studies. Colloidal gold-labeled thaumatin was shown to bind to the microvilli in the taste pore of monkey taste buds, as expected, where the sweet receptors are located (Farbman et al., 1987; Menco and Hellekant, 1993). In an attempt to purify the sweet receptor proteins, affinity columns of covalently attached thaumatin and monellin were used to isolate thaumatin/monellin-bound proteins from cow, pig and rat gustatory membranes. These bindings could be displaced by the sweet peptide aspartame or the sweet taste modifier gymnemic acid, indicating that there is a common receptor for some sweet proteins and the peptide aspartame or gymnemic acid (Persaud et al., 1988). However, the isolation of sweet receptor proteins has been hampered largely due to three problems: the scarcity of taste buds, the lack of a tight binding between taste membranes and their presumed receptors (Bruch et al., 1988), and lack of cultured taste cell lines. Further biochemical studies indicated that sweet receptors are G-protein coupled receptors (GPCRs). Activation of these receptors in the membranes derived from taste bud-containing lingual epithelium by sucrose and other sugars significantly enhanced the activity of adenylyl cyclase (AC), one of the effector enzymes in G-protein-mediated signaling cascades. This response was concentration-dependent, and was mediated by guanine-nucleotide-binding proteins (G-proteins) found in the taste cell membranes (Striem et al., 1989; Bruch and Kalinoski, 1987). 1.3.2 Discovery of the sweet taste receptors Over the last three decades, animal behavioral assays demonstrated that inbred mouse strains such as C57BL/6 and DBA/2 displayed a marked difference in detection thresholds for synthetic sweet compounds such as saccharin and other sweeteners (Fuller, 1974). This trait exhibited a simple mendelian inheritance (Lush, 1989; Phillips et al., 1994; Lush et al., 1995). Genetic mapping of recombinant strains identified a Sac locus, named after saccharin, on the distal end of mouse chromosome 4 as the genetic determinant for this trait (Phillips et al., 1994; Lush et al., 1995; Blizard et al., 1999). In addition, the peripheral
Stimulation of taste cells by sweet taste compounds
9
nerve responses to sucrose were affected by the Sac locus as well (Bachmanov et al., 1997), suggesting that this locus encodes a sweet taste receptor or some other key sweet taste transduction element. In 1999, by random sequencing of subtracted rat taste papilla-enriched cDNA libraries, Hoon and colleagues isolated cDNAs that encoded two putative GPCRs: TR1 and TR2 (now referred to as T1R1 and T1R2) (Hoon et al., 1999). These two genes are expressed in a subset of taste bud cells and expression levels of each gene showed distinct topographical distributions across the tongue and palate that corresponds to the taste sensitivity map of the tongue. Based on its topographical expression pattern, T1R1 was suggested to be putative sweet receptor. However, high resolution genetic mapping eliminated T1R1 as a Sac-encoded sweet receptor (Li et al., 2001). With the availability of mouse and human genome sequences and bioinformatics tools, data mining of DNA sequences of the Sac locus on mouse chromosome 4 and a syntenous region on human chromosome 1 identified a novel putative 7-transmembrane domain receptor, T1R3 (Bachmanov et al., 2001; Kitagawa et al., 2001; Max et al., 2001; Montmayeur et al., 2001; Sainz et al., 2001). Subsequent analyses revealed that the allelic variations between sweet-sensitive and -insensitive mouse strains resulted in amino acid substitutions in T1R3 receptor that corresponded to the Sac phenotypes, suggesting that it is the sweet receptor that is encoded by this Sac locus. Introgression by serial backcrossing of a small T1R3 gene-containing chromosomal fragment from a sweet taster mouse strain (C57BL/6ByJ) to a sweet non-taster strain (129P3/J) conferred the latter high sensitivity to sweeteners (Bachmanov et al., 2001). Further confirmation of T1R3 gene as Sac is from a transgenic experiment that demonstrated the rescue of the sweet taste deficit by introduction of the T1R3 gene from a sweet taster strain into a sweet non-taster strain (Nelson et al., 2001). Based on the sequence similarity of the heptahelical domain, T1R3, as well as T1R1 and T1R2, is classified as a member of Class C GPCR family (Pin et al., 2003). Many members of this class including calcium sensing receptor (CaSRs), putative pheromone receptor V2Rs, neurotransmitter receptors mGluRs and GABABRs, as well as T1R1, T1R2 and T1R3 have a large N-terminal extracellular domain. Some of the receptors in this class form homo- or heterodimers. Double labeling in situ hybridization indicated some taste bud cells coexpressed T1R1 and T1R3, or T1R2 and T1R3 while a few cells express T1R3 alone. Thus T1R-expressing cells can be categorized into three types: cells coexpressing T1R1+T1R3; cells co-expressing T1R2+T1R3; cells expressing T1R3 alone (Montmayeur et al., 2001; Nelson et al., 2001). These expression patterns suggested that T1R3, like other members of Class C GPCRs, may form heterodimers with T1R1 and T1R2. Co-expression of T1Rs in a heterologous system in fact confirmed that the dimeric receptor of T1R2 and T1R3 functions as a sweet receptor and human T1R2/T1R3 can be activated by all sweet compounds tested, including sugars: fructose, galactose, glucose, lactose, sucrose and maltose; amino acids: glycine and D-tryptophan; sweet proteins: monellin and thaumatin; synthetic sweeteners: acesulfame K, aspartame,
10
Optimising sweet taste in foods
cyclamate, dulcin, neotame and saccharin. The order of the sensitivity of the receptors to these substances approximates that from animal behavioral tests. Interestingly, T1R3 can also form a heterodimeric receptor with T1R1 that can be activated by umami stimuli (Nelson et al., 2001, 2002; Li et al., 2002). Therefore, T1R3 appears to be a common monomer of sweet and umami receptors. Phylogenetic analysis showed that T1R3 is most closely similar to T1R1 and T1R2 with ~30% amino acid identity, while T1R1 and T1R2 are ~40% identical to each other (Hoon et al., 1999; Bachmanov et al., 2001; Kitagawa et al., 2001; Max et al., 2001; Montmayeur et al., 2001; Sainz et al., 2001; Nelson et al., 2001). Unlike most other GPCRs, which usually share high similarity across closely related species (e.g. another receptor of the Class C, calcium-sensing receptor has >90% amino acid identity among human, rodent and bovine sequences) (Chattopadhyay et al., 1996), taste receptors including T1R sweet, umami receptors and T2R bitter receptors display only about 70% sequence identity between human and rodents. This suggests that species-specific taste receptors may have different ligand interaction properties, and each may have evolutionarily tuned to their distinct ecological niches. Psychophysical, behavioral and electrophysiological studies showed that humans and old world primates can detect the sweetness of a subset of substances including monellin, brazzein, thaumatin, neotame, aspartame, cyclamate and neohesperidin dihydrochalcone that new world monkeys and rodents cannot. Furthermore, human sweet taste is subject to the suppression by lactisole while rodents are insensitive to this compound (Danilova et al., 1998, 2002; Hellekant et al., 1997; Sclafani and Perez, 1997; Brouwer et al., 1973; Naim et al., 1982; Schiffman et al., 1999; Johnson et al., 1994). In line with these observations, heterologously expressed human T1R2 and T1R3 dimeric receptors can respond to the stimulation of these sweeteners, and the responses are susceptible to lactisole inhibition while rodent receptors cannot be activated by these compounds and the response to other sweeteners are not suppressed by lactisole (Li et al., 2002). Unlike bitter T2R receptors, which have a short extracellular amino-terminus, T1R receptor proteins consist of several functional domains (Fig. 1.1): a large extracellular amino-terminal domain (ATD), followed by a cysteine rich domain (CRD), a heptahelical transmembrane domain (HD) and an intracellular carboxyl-terminal tail (Hoon et al., 1999; Bachmanov et al., 2001; Kitagawa et al., 2001; Max et al., 2001; Montmayeur et al., 2001; Sainz et al., 2001; Nelson et al., 2001; Adler et al., 2000). Based on the crystal structure of another member of Class C receptors, mGluR1, T1Rs are likely to have a clam shell-like `Venus flytrap module' (VFTM) in the ATD domain that is responsible for ligand binding. Mutagenesis and interspecies domain swapping between human and rodent T1R2 and T1R3 receptors have determined particular domains and specific amino acid residues to which the species-specific sensitivity to ligands and inhibitors are attributable. For example, human T1R2 VFTM but not rat T1R2 VFTM can be activated by aspartame and neotame; human T1R3 CRD domain and extracellular loops 2 and 3 of HD domain can interact with brazzein
Stimulation of taste cells by sweet taste compounds
11
Fig. 1.1 Schematic representation of a dimeric sweet receptor structure. Each monomer consists of a Venus flytrap module (VFTM), a cysteine rich domain (CRD), a heptahelical transmembrane domain (HD) and an intracellular carboxyl-terminal tail (C-tail) (Pin et al., 2003). Sweeteners including aspartame and neotame bind to the VFTM module of T1R2 while brazzein and cyclamate bind to the CRD domain and extracellular loops on T1R3, respectively (Jiang et al., 2004; Xu et al., 2004). Sweet inhibitor lactisole binds to the pocket comprised of T1R3 transmembrane helices (Jiang et al., 2005; Winnig et al., 2005; Xu et al., 2004).
and cyclamate, respectively; lactisole docks to the human T1R3 binding pocket comprised of the seven transmembrane helices, while rodent counterparts do not interact with these compounds (Xu et al., 2004; Jiang et al., 2004, 2005; Winnig et al., 2005). These data also demonstrate that multiple ligand binding sites exist in the two monomers of the dimeric sweet receptor T1R2/T1R3, which may explain the earlier observations indicative of multiple sweet receptors (Schiffman et al., 1981). Surprisingly but logically, lactisole and cyclamate that suppress/enhance sweet sensation, respectively, also exert the same effect on umami taste by interacting on the T1R3 moiety of the T1R1/T1R3 umami receptor (Xu et al., 2004). 1.3.3 Discrimination of taste modalities at taste bud cells In situ hybridization and immunohistochemistry showed that T1R3 is expressed in about 30% of taste bud cells from all types of taste papillae and palate while T1R1 in turn, is mostly expressed in fungiform and palate, less in foliate, and rarely in circumvallate (Fig. 1.2). In contrast, T1R2 is expressed mostly in circumvallate, less in palate and foliate, and rarely in fungiform (Montmayeur et al., 2001; Max et al., 2001; Nelson et al., 2001). Each of T1R1 and T1R2 is coexpressed with T1R3 in a taste bud cell. And some taste bud cells may express T1R3 alone. The bitter T2R receptors are expressed in a different subset of taste
12
Optimising sweet taste in foods
Fig. 1.2 Expression of taste GPCRs in taste bud cells. Sweet receptor (T1R2/T1R3 or T1R3 alone), umami receptor (T1R1/T1R3) and bitter receptors (T2Rs) are expressed in discrete subsets of taste bud cells (Adler et al., 2000; Hoon et al., 1999; Nelson et al., 2001) . Sweet receptor cells are most abundant in circumvallate papilla, less in foliate and very few in fungiform. In contrast, umami receptor cells are rare in circumvallate, more in foliate and abundant in fungiform (Nelson et al., 2001). Bitter receptor cells, which can express multiple different bitter receptors, are more abundant in circumvallate and foliate than in fungiform (Adler et al., 2000).
bud cells (Adler et al., 2000; Nelson et al., 2001). This non-overlapping cellular expression profile suggests that sweet, umami and bitter taste modalities are discriminated at the taste bud cell level. Knockout of T1R3 receptors markedly reduced the sensitivity to sweet stimuli in behavioral tests but had no effect on bitter, sour and salty tastes (Damak et al., 2003; Zhao et al., 2003). The residual response to sweeteners in T1R3-null mice was abolished by double knockout of both T1R2 and T1R3 receptors. Transgenic introduction of human T1R2 into mice generated animals with humanized sweet taste preferences. These mice
Stimulation of taste cells by sweet taste compounds
13
could detect substances that taste sweet to humans, but normally not to rodents. In addition, transgenic expression of a modified opioid receptor in T1R2expressing cells resulted in the attraction of animals to a synthetic opiate that is normally tasteless to wild type mice. And expression of a human receptor T2R16 for a bitter compound phenyl-B-D-glucopyranoside in mouse T1R2expressing sweet cells generated animals that exhibit strong attraction to this bitter compound (Mueller et al., 2005). Theses studies demonstrated that taste modalities are detected and segregated by the taste bud cells, rather than receptors themselves. 1.3.4 Could sweet tastants stimulate non-taste receptors in taste cells? In a recent study (Zubare-Samuelov et al., 2003) we found that some sweet and bitter tastants, e.g. the non-sugar sweeteners, saccharin, D-tryptophan and neohesperidin dihydrochalcone (NHD), and the bitter tastants cyclo(Leu-Trp) and limonin, stimulated pigment aggregation in a Xenopus laevis melanophore cell line as does the native hormone melatonin. Thus, these tastants were able to generate physiological responses independent of taste sensation. Furthermore, as the native melatonin, these tastants stimulated the pigment aggregation by stimulation of the inhibitory pathway of AC as pre-treatment of melanophores with pertussis toxin almost abolished the tastant-induced cAMP reduction. Importantly, the presence of luzindole (melatonin receptor antagonist) almost abolished the inhibition of cAMP formation induced by saccharin, D-tryptophan, and cyclo(Leu-Trp) and the presence of 2-adrenergic receptor antagonist, yohimbine, almost abolished the inhibition of cAMP formation induced by NHD. These data demonstrated that saccharin and D-tryptophan are agonists of the melatonin receptors and NHD is an agonist of the 2-adrenergic receptors under the experimental conditions. It becomes evident that these tastants may stimulate receptors that are not identified taste receptors. Interestingly, both the 2-adrenergic and the melatonin receptors were reported to be present in taste bud cells (Zubare-Samuelov et al., 2003; Herness et al., 2002). If such sweeteners, in addition to their ability to stimulate the T1R2/T1R3 sweet receptors, can also stimulate melatonin, 2-adrenergic receptors, and perhaps other non-taste receptors in taste cells, they may affect the signaling output in taste cells. Should these events occur within the subsecond time course of sweet taste sensation, they may affect taste sensation which may be related to the lack of pure sweetness when tasting non-sugar sweeteners, or even modify time-taste intensity relationships. Although the PLC 2 pathway appears to be essential for the cellular transduction of sweet taste (Zhang et al., 2003), to the best of our knowledge no biochemical data are yet available to show the direct activation of specific signaling pathways in taste cells by either the T1R2/R3 or the T1R3 homodimeric sweet taste receptors. In a recent study (Ozeck et al., 2004), sucrose and non-sugar sweeteners such as aspartame, cyclamate and saccharin stimulated the human T1R2/T1R3 sweet receptor expressed in HEK293 cells to
14
Optimising sweet taste in foods
activate the inhibitory G-protein signaling pathways (in particular AC) in a pertussis toxin sensitive manner. These data suggest that either the T1R2/T1R3 GPCRs are coupled to the Gi subunits in taste cells to induce sweet taste transduction or these receptors can be coupled to more than one transduction pathway, further demonstrating a lack of specificity of the transduction output.
1.4
Downstream signaling components
In general, signaling pathways for GPCRs are characterized by the presence of a GTP binding protein (G-protein) which is a coupling component between GPCRs, effector enzymes and ionic channels (Simon et al., 1991; Kristiansen, 2004). There are two main classes of G-proteins. The first are heterotrimeric Gproteins that associate with GPCRs and are involved in signal transduction (Wong, 2003; Kristiansen, 2004). The second are small cytoplasmic G-proteins. The heterotrimeric G-proteins are composed of , and subunits, with molecular masses of about 39±45, 35±39 and 6±8 kDa, respectively. About 28 distinct G-protein subunits, which are products of 16 different genes and splice variants have been identified. and subunits are tightly associated and can be regarded as one functional unit. Five different and 12 different subunits have been identified (Kristiansen, 2004). Although not all interactions are possible, many combinations occur which increased the diversity and specificity of signaling. The heterotrimeric G-proteins are divided into four families based on the degree of primary sequence similarities of their subunits: Gs (Gs and Golf), Gi/o subfamily which inhibits AC and regulates ion channels (Gtr, Gtc, Ggust, Gi1-3, Go, and Gz), Gq/11 which activates PLC (Gq, G11, G14, and G15/16), and G12(G12 and G13) that activates Na+/H+ exchanger pathway (Wong, 2003; Kristiansen, 2004; Cabrera-Vera et al., 2003). The G-protein gustducin (Ggust) was first identified in taste cells along with some other Gproteins (McLaughlin et al., 1992). Upon activation by the GPCR, the -subunit binds GTP in exchange for GDP but both - and subunits interact specifically with various effectors in cells including ion channels in cell membranes and enzymes both in the cytosol and in cell membranes. In several signaling systems, specific `regulators of G-protein signaling' (RGS) proteins accelerate GTPase activity of the -subunits. Interestingly, RGS21, a novel RGS was recently found to be co-expressed with sweet T1R2/T1R3 and bitter T2R taste receptors in subpopulations of taste cells (von Buchholtz et al., 2004). Even though considerable research efforts have been made during the last 15 years trying to elucidate possible cellular mechanisms, the nature of the cellular signals responsible for sweet taste transduction is complex and to date is far from being fully clarified. Early studies indicated high AC (Kurihara and Koyama, 1972) and PDE (Kurihara, 1972) activities in gustatory epithelium, and the presence of these enzymes in the microvilli of cells in rabbit taste buds was also shown (Asanuma and Nomura, 1982). Furthermore, cAMP content could increase in intact bovine taste papillae in response to sucrose stimulation
Stimulation of taste cells by sweet taste compounds
15
(Cagan, 1974). When the patch-clamp technique became available to record electrophysiological responses in single taste cells, intracellular administration of the cyclic nucleotides cAMP and cGMP into taste cells of frogs and mice (Avenet et al., 1988; Tonosaki and Funakoshi, 1988) decreased potassium conductance, leading to depolarization. In hamsters, fungiform taste buds responded to stimulation by sucrose and some non-sugar sweeteners by inducing action potentials. These responses were mimicked by stimulation with membranepermeant analogs of cAMP and cGMP (Cummings et al., 1993). Additional biochemical studies employing crude membrane preparations of taste tissue and isolated circumvallate (CV) taste bud sheets have suggested the involvement of the AC cascade in sugar taste transduction (Naim et al., 1991; Striem et al., 1989, 1991). However, additional biochemical experiments with isolated CV taste buds indicated that the non-sugar sweeteners SC45647 and saccharin did not induce cellular increase in cAMP, but rather an increase in IP3 and parallel elevation of intracellular Ca2+, as observed by confocal microscopy (Bernhardt et al., 1996). This indicates two different signaling pathways for sweet taste in a single cell responsive to sweet sucrose and SC45647 but not to the bitter quinine. In summary, the above data suggest that sweet taste transduction utilizes more than one pathway (Kinnamon, 1996; Lindemann, 1996, 2001). In the early studies, some of the controversial results may stem from a lack of appropriate methodology for monitoring post-receptor intracellular chemical signals in subsecond time course that parallels taste sensation. For example, our previous biochemical studies (Striem et al., 1989, 1991) (also supported by cellular electrophysiological studies (e.g., ref. (Avenet et al., 1988)) proposed that the AC is involved in sugar taste transduction. However, when we examined the same question on a subsecond time frame (Spielman et al., 1996; Breer et al., 1990), we found, following sucrose stimulation in intact taste bud cells that cGMP rather than cAMP increases in the first 100 ms. (Krizhanovsky et al., 2000). It was then proposed that cAMP does not appear to be involved in the early detection (ms time range) of sweet taste of sugar, but it is elevated later in the sequence of events (seconds time range or longer) and may involve adaptation and signal termination (Krizhanovsky et al., 2000; Kinnamon and Varkevisser, 1998; Varkevisser and Kinnamon, 2000). Controversial data also appeared in relation to how cAMP may be involved in Ca2+ inflow following stimulation by sugar. For example, an inhibitor of protein kinase A (PKA) was found to block the cAMP-induced decrease in potassium conductance in frog taste cells (Avenet et al., 1988) but not in hamster taste cells (Varkevisser and Kinnamon, 2000). Furthermore, a cyclic nucleotide-gated channel, the CNGgust was found in taste cells (Misaka et al., 1998) which may contribute to depolarization and Ca2+ inflow (Lindemann, 2001). It is still unclear how data from the above studies fit into the overall mechanism(s) of sweet taste transduction. The findings that -gustducin knockout mice exhibited reduced (though not completely eliminated) behavioral and electrophysiological responses to bitter-
16
Optimising sweet taste in foods
and sweet, but not salty- or sour-taste stimulations (Wong et al., 1996) has led to the following hypothesis (Kinnamon, 1996). Gustducin may mediate bitter-taste transduction by its -subunit via activation of PDE and subsequent reduction of cAMP, while the associated -subunits may activate PLC 2 and induce an increase in IP3 following non-sugar stimulation. This can explain the involvement of both AC and PLC pathways in mediating sweet taste transduction induced by sugar and non-sugar sweeteners, respectively. This finding leads to the assumption that cells which respond to sweet are not the ones which respond to bitter stimulation (Lindemann, 1996, 2001). More recent studies have indicated that TRPM5 (Transient Receptor Potential ion channel), and PLC 2 are present in taste cells (Perez et al., 2002). TRPM5 appears to be a non-specific monovalent cation channel which is important to cell depolarization. It is activated by M concentrations of calcium ions (Hofmann et al., 2003; Liu and Liman, 2003; Prawitt et al., 2003). Most importantly, TRPM5- or PLC 2-knockout mice almost completely lose their sensitivity to sugar, non-sugar sweeteners, bitter and umami taste stimulation, as evidenced by behavioral and electrophysiological experiments (Zhang et al., 2003). TRPM5's function was evident following sweet and bitter stimulation in cells expressing a G-protein that couples a wide range of GPCRs to PLC; every T1R- and T2R-positive cell in the circumvallate (CV), foliate fungiform and palate taste buds co-expressed TRPM5, and the need for the co-existence of TRPM5 and PLC 2 in the taste cells was strongly suggested (Zhang et al., 2003). These new data led to the hypothesis that a common signaling pathway for the generation of these tastes exists, with PLC 2 potentially being one and TRPM5 being the last post-receptor component in the cellular-transduction chain needed to elicit these tastes. These data are in contrast, at least in part, with the aforementioned data suggesting the involvement of various post-receptor transduction pathways for sweet, bitter and umami tastes. The demonstration that the PLC 2±TRPM5 pathway is an essential part for taste signaling has led to the suggestion that other pathways may only modulate taste sensation but are not needed for the basic detection of sweet, bitter and umami tastes. 1.4.1 Could receptor-independent mechanisms be involved in sweet taste transduction? In contrast to sugars, the chemical structures of non-sugar sweeteners (as well as bitter tastants) are very diverse, all of these tastants are amphipathic, i.e., contain both hydrophobic and hydrophilic domains, and therefore they are putative membrane-permeant compounds. A variety of amphipathic neuropeptides, venom peptides and non-peptide substances have been found to be direct activators of G-proteins and are likely to activate transduction pathways due to their ability to permeate the plasma membrane, in addition to their action on specific receptors (Higashijima et al., 1988; Mousli et al., 1990). Furthermore, some of these taste compounds elicit taste and taste nerve responses (Bradley,
Stimulation of taste cells by sweet taste compounds
17
1973; Hellekant et al., 1987; Fishberg et al., 1933) following intravenous or intralingual administration, independent of stimulation of putative receptors at the apical surface of the tongue. For example, already in the 1930s, Fishberg and colleagues (Fishberg et al., 1933) used intravenous injection of sodium saccharin to measure the blood circulation time in humans from the time of saccharin injection into the peripheral vein until subjects indicated that they could taste the compound. For the amphipathic tastants to interact directly with membrane transduction components located downstream to the GPCRs, they need to be able to permeate deeply into the membrane and/or translocate to the cytosolic side of the plasma membrane. Some amphipathic tastants were found to be able to permeate liposomes, isolated taste bud cells (Peri et al., 2000), and taste bud cells in vivo (Zubare-Samuelov et al., 2005) (see discussion below). In line with these results, some amphipathic non-sugar sweeteners and the bitter tastant quinine are direct activators of G-proteins in vitro (Spielman et al., 1992; Naim et al., 1994). However, the evidence that such a phenomenon exists in vivo remains to be proven.
1.5 Putative cellular mechanism(s) for the signal termination of sweet taste The efficiency of receptors signaling may be regulated by the receptor number and by their presence on the cell surface (Gainetdinov et al., 2004). It is known that often GPCR signaling desensitizes rapidly as a consequence of receptor phosphorylation. This phosphorylation is mediated by two families of protein kinases (Bohm et al., 1997; Hamm and Gilchrist, 1996; Pitcher et al., 1998). One family of kinases is second-messenger-dependent (heterologous desensitization). For example, cAMP-dependent protein kinase, PKA, phosphorylates the -adrenergic receptors. Similar second messenger-dependent regulation of receptor desensitization may occur via protein kinase C (PKC), mitogenactivated protein (MAP) and additional kinases. The same kinases can also phosphorylate effectors such as AC and PLC (Gainetdinov et al., 2004; Hamm and Gilchrist, 1996). The second family of kinases that desensitize GPCRs (homologous desensitization) are the GPCR kinases (GRKs). In contrast to the second messengers-dependent kinases, GRKs discriminate between the inactive and agonist-activated states of the receptor (catalytically activated by stimulated receptors) (Gainetdinov et al., 2004; Pitcher et al., 1998), and specifically phosphorylate the agonist-activated form of GPCRs. Thus, when ligands are involved, the phosphorylation of GPCRs by GRKs appears essential: deletion of GRK phosphorylation sites in the receptor results in stronger signaling responses to agonist stimulation in transfection studies (Mueller et al., 1997). GRKs promote the binding of the phosphorylated receptors to arrestin proteins which further uncouple the receptors from G-proteins and prevent further stimulation of G-proteins and downstream signaling pathways (Pitcher et al., 1998; Gainetdinov et al., 2004). Subsequently, desensitization of the receptors leads to
18
Optimising sweet taste in foods
their removal from the cell surface (internalization). They may then be degraded in lysosomes (downregulation) or recycling to the cell surface (resensitization). There are seven known types of GRK subtypes which are classified based on their sequence and function. GRK1 and GRK7 are primarily expressed in the visual system, GRK4 is primarily expressed in testes. Therefore, it is believed that GRK2, GRK3, GRK5 and GRK6 subtypes account for the regulation of most of the GPCRs found throughout the body (Benovic et al., 1987; Gainetdinov et al., 2004). What GRK subtypes are present in the gustatory system? GRK5 and potentially GRK2 and GRK3 have been previously identified in bovine CV papilla (Premont et al., 1994), though there did not seem to be any evidence of their presence in taste cells. Recently, two studies were published on the presence of GRKs in taste bud cells (Masuho et al., 2005; Naim et al., 2004; Zubare-Samuelov et al., 2005). The presence of the bitter T2Rs (Adler et al., 2000) and the sweet (e.g., T1R3 and T1R2) (Max et al., 2001, Nelson et al., 2001) taste receptors in subpopulations of cells present in rodent CV and other taste papillae has been demonstrated. Evidently, for the GRKs to function in the signal termination of taste receptors they must be present in the same cell subpopulations that contain the above receptors. Our recent study (Naim et al., 2004; Zubare-Samuelov et al., 2005) using RT-PCR suggested the presence of T2R4 and T1R3 along with
Fig. 1.3 RT-PCR analysis of GRK1, GRK2, GRK3, GRK5, GRK6, T2R4 and T1R3 mRNA in circumvallate (CV) taste bud sheets and non-sensory epithelium (EP). cDNA was synthesized from CV and EP RNA and then amplified by PCR using specific primers for GRK1, GRK2, GRK3, GRK5, GRK6, T2R4 or T1R3; GAPDH was used as an internal reference gene. CON designates parallel PCR, omitting the RT step and using GRK2 primers. From Zubare-Samuelov et al. (2005). Used with permission by The American Physiological Society.
Stimulation of taste cells by sweet taste compounds
19
GRK2, GRK3, GRK5 and GRK6 in isolated taste bud cell sheets of rat CV papillae (Fig. 1.3). The same GRKs were found to be expressed in non-sensory epithelium but as expected, no PCR products for the T2R4 and T1R3 receptors were observed in non-sensory epithelial cells. Furthermore, antibodies against GRK2, GRK5 and GRK6 (with weak staining of GRK3) yielded clear, positive immunostaining in the 10 M frozen sections of the CV papilla. However, only GRK5 was clearly stained in the CV taste bud cells; GRK2 and GRK6 were notably stained in the surrounding epithelium but their presence in the taste cells remains questionable. These immunostaining experiments cannot provide evidence for T2R4 and T1R3 receptor expression in the same subpopulation of taste cells that contain the GRKs. However, since GRK5 appeared to be expressed in almost all taste bud cells, it is likely that it co-expresses in cells expressing T2R4 and T1R3. The study in the mouse (Masuho et al., 2005), using immunohistochemistry of the CV papilla, suggested the differential distribution of GRK2, GRK3 and GRK5 in the CV papilla while only GRK2 was present in taste bud cells. It remains to be determined whether the difference between the results in these two studies are related to species differences (rats vs. mice) or the different methodology for GRKs identification led to different observations. Nevertheless, these results call for further investigation to verify that GRK5 and possible other GRKs are indeed expressed in the same taste cells as the T1R2/T1R3 receptors, and to investigate whether they functionally coupled to these receptors. 1.5.1 Amphipathic tastants permeate taste cells in vivo and inhibit receptor-related kinases in vitro A few hypotheses may be proposed to explain the delay in taste-signal termination produced by many non-sugar sweeteners and bitter tastants. According to some studies (see Introduction), this phenomenon appears to be related to events occurring in taste cells and not in the central nervous system (Hellekant, 1994; Spielman et al., 2002). Interestingly, although sugar and nonsugar sweeteners appear to stimulate similar or identical sweet GPCRs, the `lingering' taste sensations are uniquely produced by non-sugar sweeteners (as well by bitter tastants). Because such amphipathic tastants, unlike sucrose, rapidly permeate isolated taste bud cells in situ (Peri et al., 2000), a recent study (Zubare-Samuelov et al., 2005) was initiated to investigate whether they may, after their permeation through the taste cell plasma membrane, interact with downstream transduction components which affect GPCR signal termination (e.g., GRKs, PKA), in addition to their extracellular interaction with GPCRs. For example, inhibition of receptor phosphorylation by GRKs can delay GPCRsignal termination, as shown in vision (Sitaramayya and Liebman, 1983; Gan et al., 2000, 2004) and other systems (Stadel et al., 1983; Pao and Benovic, 2005; Teli et al., 2005). The first step was undertaken to further investigate whether amphipathic tastants can indeed permeate taste cells via the apical oral route under
20
Optimising sweet taste in foods
physiological conditions and not only in isolated taste bud preparations (Peri et al., 2000). The auto-fluorescence that some sweet and bitter amphipathic tastants possess was used as before to monitor tastant permeation into taste cells using two different procedures (Zubare-Samuelov et al., 2005). First, in-situ confocal imaging of intact CV papilla surgically removed without collagenase treatment indicated evident dynamics of tastants permeation (the sweeteners saccharin and D-tryptophan and the bitter tastants caffeine, cyclo(Leu-Trp), naringin and quinine) into the CV cells located around the circular inner trench where most of the CV taste buds are located. The elimination of collagenase treatment in this procedure prevented the exposure of the basolateral side of taste bud cells to tastants and thus allowed permeation through the apical side of the membrane only, the route by which tastants interact with sensory taste cells under physiological conditions. In these experiments, the addition of the membrane-impermeant quencher, KI, had no effect on tastants fluorescence, indicating the presence of such tastants inside the cytosol or on the cytosolic side of the membrane rather than being adsorbed on the extracellular surface of the cells. In the second procedure, the oral cavity of anesthetized rats was stimulated on-and-off with tastant solutions for 90 s, followed by immediate animal sacrifice and an HPLC determination of the tastants' intracellular concentrations (Zubare-Samuelov et al., 2005): millimolar levels of these tastants were found inside the taste cells (Table 1.1). Together, the results provided evidence that these and perhaps additional (though not necessarily all) amphipathic tastants rapidly permeate taste cells under physiological conditions, either through the taste bud pore or through the tight junction, and that such permeation also occurs in non-sensory lingual epithelial cells. Subsequently, in the same study, the effect of tastants on GRK2and GRK5-phosphorylated rhodopsin, a well-studied model for GPCR phosphorylation in vitro (Pitcher et al., 1998), was investigated. The amphipathic non-sugar sweeteners cyclamate, saccharin, NHD and D-tryptophan, and the bitter tastants caffeine, quinine, limonin and naringin, with diverse chemical structures, were found to inhibit GRK2 and GRK5 phosphorylation of rhodopsin Table 1.1 Estimated permeation of amphipathic tastants into CV taste bud cells via the apical side Tastant
Extracellular Efflux (%) during conc. (mM) during 25 min 90 s of oral stimulation collagenase trt.
D-Tryptophan Quinine Cyclo(Leu-Trp) Caffeine
30 2 2 10
86* 75* 64* 70*
9.5 5.6 3.0 2.5
Intracellular conc. (mM) after 90 s of oral stimulation 6.4** 5.5** 0.96** 13.7*
1.0 0.8 0.1 3.6
Data are the means SEM of four to nine replicates, each derived from one or two rats. * indicates significant values (P < 0.03) and ** (P < 0.001). From Zubare-Samuelov et al. (2005). Used with permission of The American Physiological Society.
Stimulation of taste cells by sweet taste compounds
21
Fig. 1.4 Tastants inhibit GRK2- (a) and (b) and GRK5- (c) and (d) induced phosphorylation of rhodopsin in a concentration-dependent manner. GRK2 and GRK5 were incubated with rhodopsin and one of the following compounds: NHD, saccharin, caffeine or naringin at the stated concentrations. Note that for saccharin phosphorylation, each concentration shows two bands, one for NaCl (control for sodium) and one for sodium saccharin. CON phosphorylation level when tastants were not present. Results are the means SEM of three independent experiments. From Zubare-Samuelov et al. (2005). Used with permission of The American Physiological Society.
(a GPCR) and PKA phosphorylation of casein in vitro. Their effects depended on the kinase being tested. A concentration-dependence, tested for NHD, saccharin, caffeine and naringin, was evident (Fig. 1.4) and depending on solubility, 90 to 100% inhibition could be shown. Although inhibition could already be observed at relatively low concentrations, e.g., 2.5 mM caffeine, 600 M NHD, 125 M naringin and 100 M limonin, the tastant concentrations needed for kinase inhibition under the experimental conditions were at the millimolar level, higher than the micromolar levels usually used with other kinase inhibitors in vitro and in vivo medically (Sasaki et al., 2002) or experimentally in non-taste systems (Vera et al., 2001). Moreover, these tastants do not appear to be very specific since they inhibited PKA as well as the GRKs, and may very well inhibit additional kinases. Nevertheless, the range of tastant concentrations applied in this study (Zubare-Samuelov et al., 2005) matches that used in taste stimulation during sensory and biochemical studies (Hellekant, 1994; Nelson et al., 2001; Rosenzweig et al., 1999; Schiffman et al., 1979; Wong et al., 1996; Zhang et al., 2003) and the tastant levels were found inside
22
Optimising sweet taste in foods
taste bud cells within seconds after their extracellular application (Table 1.1). The same tastants have also been recently demonstrated to permeate other epithelial cells in situ, e.g., Xenopus laevis melanophores (Zubare-Samuelov et al., 2003). Therefore, although these relatively high concentrations of amphipathic tastants suggest low potency, their physiological significance as kinase inhibitors in taste cells may result from their access to the cytosolic side of the taste cells, and thus to GRKs, which are present in taste tissue (Fig. 1.3), and to PKA or PKC which have been indirectly shown to be active in taste cells (Varkevisser and Kinnamon, 2000). Naringin (as well as some other flavonoids) has been found to inhibit phosphorylation, e.g., naringin inhibits AMPKactivating kinase (Larsen et al., 2002). Caffeine and theophylline have also been recently found to be phosphoinositide 3-kinase inhibitors (Foukas et al., 2002). However, nothing has been reported in relation to GRKs or PKA. The above results have led to a new hypothesis which may partly explain the lingering aftertaste produced by non-sugar sweetener and bitter tastants (Fig. 1.5). Amphipathic tastants stimulate sweet taste GPCRs located on the
Fig. 1.5 Hypothetical pathways for desensitization of sweet taste receptors (R). Nonsugar sweeteners (T) stimulate R from the extracellular side to activate G-proteins (, ,
). -Subunits, in turn, activate PLC 2. Receptor-mediated signaling may be desensitized by G-protein-coupled receptor kinases (GRKs) which phosphorylate R and promote the binding of arrestin proteins to the receptor, uncoupling the receptor from Gproteins. Alternatively, the receptor may be phosphorylated and desensitized by second messenger-dependent kinases, e.g. protein kinase A (PKA) or protein kinase C (PKC). Intracellular inhibition of these receptor phosphorylations, e.g. directly by the membranepermeant T, would expect to inhibit desensitization and therefore delay taste signaling termination. Modified from Pitcher et al. (1998) and based on Zubare-Samuelov et al. (2005).
Stimulation of taste cells by sweet taste compounds
23
extracellular surface of taste cells, and concomitantly also permeate to the cytosolic side of the cell membrane or to the cytosol under physiological conditions. Thus, such tastants have access to interact directly with GRKs or with other receptor-related kinases (e.g., PKA, PKC). Inhibition of GRK- and/or PKA/PKC-induced phosphorylation of GPCRs should lead to a delay in signal termination, and therefore may extend the taste response (e.g., lingering). In conclusion, results discussed in this chapter show the complexity of the cellular events occurring in taste cells upon stimulation by sweeteners. It is likely that such diverse cellular responses following the stimulation by different sweeteners, can explain the tremendous differences in sweet sensation of these sweeteners as implied to sweet taste quality.
1.6
Future trends
To further explore ways to optimize sweet taste quality, more studies at the cellular and molecular level are required. At the level of the sweet receptor(s), a better understanding of the interaction(s) with different sweet ligands is to be achieved. Each of these interactions may have a different affinity characteristic and, in view of the data on other GPCRs indicating their ability to activate more than a single transduction pathway, different sweet tastants may stimulate different signaling pathways which are related to different sweet taste qualities. More research is needed to investigate if some sweeteners can stimulate nontaste receptors that are present in taste cells. One may hypothesize that such effects, if occurring within the time course of taste sensation, may also be related to inferior sweet taste quality. Finally, investigations of peripheral taste signal termination mechanisms, and the possible interference with such mechanisms are imperative since they may explain the undesirable temporal properties produced by a variety of non-sugar sweeteners. To validate such mechanisms, it is necessary to further investigate the newly discovered taste GPCRs to verify their possible interactions with GRKs and other kinases, and to verify the inhibitory effect of taste stimulants in vivo. This hypothesis of signal termination is bold precisely because it involves a dual mechanism: first the ligand interacts extracellularly with GPCRs to initiate the transduction chain, second, the same ligand may interact intracellularly with downstream shut-off components to affect signal termination.
1.7
Acknowledgements
This research was supported by Grant 2003015 from The US-Israel Binational Science Foundation (BSF) and by Grant IS-3366 from The US-Israel Binational Agricultural Research and Development Fund (BARD).
24
1.8
Optimising sweet taste in foods
References
and (2000) Cell, 100, 693±702. ASANUMA, N. and NOMURA, H. (1982) Chem. Senses, 7, 1±9. AVENET, P., HOFMANN, F. and LINDEMANN, B. (1988) Nature, 331, 351±354. AYYA, N. and LAWLESS, H. T. (1992) Chem. Senses, 17, 245±259. ADLER, E., HOON, M. A., MUELLER, K. L., CHANDRASHEKAR, J., RYBA, N. J. P.
ZUKER, C. S.
BACHMANOV, A. A., REED, D. R., NINOMIYA, Y., INOUE, M., TORDOFF, M. G., PRICE, R. A. BEAUCHAMP, G. K.
(1997) Mamm. Genome, 8, 545±548.
and
BACHMANOV, A. A., LI, X., REED, D. R., OHMEN, J. D., LI, S., CHEN, Z., TORDOFF, M. G., DE JONG, P.
and BEAUCHAMP, G. K. (2001) Chem. Senses, 26, 925±933. BARTOSHUK, L. M. (1987) In Sweetness (Ed, Dobbing, J.) Springer-Verlag, London, pp. 33±46. BENOVIC, J. L., MAYOR, F. Jr, STANISZEWSKI, C., LEFKOWITZ, R. J. and CARON, M. G. (1987) J. Biol. Chem., 262, 9026±9032. BERMAN, D. E. and DUDAI, Y. (2001) Science, 291, 2417±2419. BERNHARDT, S. J., NAIM, M., ZEHAVI, U. and LINDEMANN, B. (1996) J. Physiol. (Lond.), 490, 325±336. BIRCH, G. G., LATYMER, Z. and HOLLAWAY, M. (1980) Chem. Senses, 5, 63±78. BLIZARD, D. A., KOTLUS, B. and FRANK, M. E. (1999) Chem. Senses, 24, 373±385. BOHM, S. K., GRADY, E. F. and BUNNETT, N. W. (1997) Biochem. J., 322, 1±18. BRADLEY, R. M. (1973) Am. J. Physiol., 224, 300±304. BRAND, J. G., CAGAN, R. H. and NAIM, M. (1982) Ann. Rev. Nutr., 2, 249±276. BREER, H., BOEKHOFF, I. and TAREILUS, E. (1990) Nature, 345, 65±68. BROUWER, J. N., HELLEKANT, G., KASAHARA, Y., VAN DER WEL, H. and ZOTTERMAN, Y. (1973) Acta. Physiol. Scand., 89, 550±557. BRUCH, R. C. and KALINOSKI, D. L. (1987) J. Biol. Chem., 262, 2401±2404. BRUCH, R. C., KALINOSKI, D. L. and KARE, M. R. (1988) Ann. Rev. Nutr., 8, 21±42. J., WU, C., WEST, D. B., CHATTERJEE, A., ROSS, D. A.
CABRERA-VERA, T. M., VANHAUWE, J., THOMAS, T. O., MEDKOVA, M., PREININGER, A., MAZZONI,
and HAMM, H. E. (2003) Endocr. Rev., 24, 765±781. (1971) Biochim. Biophys. Acta, 252, 199±06. CAGAN, R. H. (1974) In Sugars in Nutrition (Eds, Sipple, H. L. and McNutt, K. W.) Academic Press, New York, pp. 19±36. CAGAN, R. H. and MALLER, O. (1974) J. Comp. Physiol. Psychol., 87, 47±5. CAGAN, R. H. and MORRIS, R. W. (1979) Proc. Natl. Acad. Sci. USA, 76, 1692±1696. CHATTOPADHYAY, N., MITHAL, A. and BROWN, E. M. (1996) Endocr. Rev., 17, 289±307. CLAPP, T. R., YANG, R., STOICK, C. L., KINNAMON, S. C. and KINNAMON, J. C. (2004) J. Comp. Neurol., 468, 311±321. CUMMINGS, T. A., POWELL, J. and KINNAMON, S. C. (1993) J. Neurophysiol., 70, 2326±2336. M. R.
CAGAN, R. H.
DAMAK, S., RONG, M., YASUMATSU, K., KOKRASHVILI, Z., VARADARAJAN, V., ZOU, S., JIANG, P.,
and MARGOLSKEE, R. F. (2003) Science, 301, 850±853. and DUBOIS, G. E. (1999) In XXI Association of Chemoreception Sciences Meeting Sarasota, FL, April 16., pp. Poster Presentation. DANILOVA, V., HELLEKANT, G., TINTI, J. M. and NOFRE, C. (1998) J. Neurophysiol., 80, 2102± 2112. DANILOVA, V., DANILOV, Y., ROBERTS, T., TINTI, J. M., NOFRE, C. and HELLEKANT, G. (2002) J. Neurophysiol., 88, 579±594. DESOR, J. A., MALLER, O. and TURNER, R. E. (1973) J. Comp. Physiol. Psychol., 84, 496±501. NINOMIYA, Y.
D'ANGELO, L.
Stimulation of taste cells by sweet taste compounds
25
(1991) In Chemical Senses: Appetite and Nutrition, Vol. 4 (Eds, Friedman, M. I., Tordoff, M. G. and Kare, M. R.) Marcel Dekker, Inc., New York, pp. 69±83. DREWNOWSKI, A., KRAHN, D. D., DEMITRACK, M. A., NAIRN, K. and GOSNELL, B. A. (1992) Physiol. Behav., 51, 371±379. DREWNOWSKI, A., KRAHN, D. D., DEMITRACK, M. A., NAIRN, K. and GOSNELL, B. A. (1995) Am. J. Clin. Nutr., 61, 1206±1212. DUBOIS, G. E. (2004) Proc. Natl. Acad. Sci. (USA), 101, 13972±13973. DUBOIS, G. E., CROSBY, G. A., STEPHENSON, R. A. and WINGARD, R. E. Jr (1977) J. Agric. Food Chem., 25, 763±772. DUBOIS, G. E., CROSBY, G. A., LEE, J. F., STEPHENSON, R. A. and WANG, P. C. (1981) J. Agric. Food Chem., 29, 1269±1276. DREWNOWSKI, A.
DUBOIS, G. E., WALTERS, D. E., SCHIFFMAN, S. S., WARWICK, Z. S., BOOTH, B. J., PECORE, S. D.,
GIBES, K., CARR, B. T. and BRANDS, L. M. (1991) In Sweeteners: Discovery, Molecular Design and Chemoreception, Vol. ACS Symposium Series 450 (Eds, Walters, D. E., Orthoefer, F. T. and DuBois, G. E.) Am. Chem. Soc., Washington, DC, pp. 261±276. DUBOIS, G. E., WALTERS, D. E. and KELLOG, M. S. (1993) In Sweet-Taste Chemoreception (Eds, Mathlouthi, M., Kanters, J. A. and Birch, G. G.) Parthenon Press, London, pp. 237±267. FARBMAN, A. I., OGDEN-OGLE, C. K., HELLEKANT, G., SIMMONS, S. R., ALBRECHT, R. M. and VAN DER WEL, H. (1987) Scanning Microsc., 1, 351±357. FISHBERG, A. M., HITZIG, W. M. and KING, F. H. (1933) Proc. Soc. Exptl. Biol. Med., 30, 651±652. FOUKAS, L. C., DANIELE, N., KTORI, C., ANDERSON, K. E., JENSEN, J. and SHEPHERD, P. R. (2002) J. Biol. Chem., 277, 37124±37130. FULLER, J. L. (1974) J. Hered., 65, 33±36. GAINETDINOV, R. R., PREMONT, R. T., BOHN, L. M., LEFKOWITZ, R. J. and CARON, M. G. (2004) Ann. Rev. Neurosci., 27, 107±144. GAN, X. Q., WANG, J. Y., YANG, Q. H., LI, Z., LIU, F., PEI, G. and LI, L. (2000) J. Biol. Chem., 275, 8469±8474. GAN, X., MA, Z., DENG, N., WANG, J., DING, J. and LI, L. (2004) J. Biol. Chem., 279, 49741± 49746. GRAND, F. (1974) In Sugar in Nutrition (Eds, Sipple, H. L. and McNutt, K. W.) Academic Press, New York, pp. 401±437. HAMM, H. E. and GILCHRIST, A. (1996) Curr. Opin. Cell. Biol., 8, 189±196. HELLEKANT, G. (1994) In Thaumatin (Eds, Witty, M. and Higginbotham, J. D.) CRC Press, London, pp. 99±113. HELLEKANT, G., ROBERTS, T., AF SEGERSTAD, C. H. and VAN DER WEL, H. (1987) Ann. New York Acad. Sci., 510, 356±358. HELLEKANT, G., NINOMIYA, Y. and DANILOVA, V. (1997) Physiol. Behav., 61, 829±841. HERNESS, S., ZHAO, F. L., KAYA, N., LU, S. G., SHEN, T. and SUN, X. D. (2002) J. Physiol. (Lond.), 543, 601±614. HIGASHIJIMA, T., UZU, S., NAKAJIMA, T. and ROSS, E. M. (1988) J. Biol. Chem., 263, 6491± 6494. HIJI, T. (1975) Nature, 256, 427±429. HOFMANN, T., CHUBANOV, V., GUDERMANN, T. and MONTELL, C. (2003) Curr. Biol., 13, 1153±1158. HOON, M. A., ADLER, E., LINDEMEIER, J., BATTEY, J. F., RYBA, N. J. P. and ZUKER, C. S. (1999) Cell, 96, 541±551.
26
Optimising sweet taste in foods
JACOBS, H. L.
(1958) J. Comp. Physiol. Psychol., 51, 304±310.
JIANG, P., JI, Q., LIU, Z., SNYDER, L. A., BENARD, L. M., MARGOLSKEE, R. F.
Biol. Chem., 279, 45068±45075.
and MAX, M. (2004) J.
JIANG, P., CUI, M., ZHAO, B., LIU, Z., SNYDER, L. A., BENARD, L. M., OSMAN, R., MARGOLSKEE, R. F.
and
(2005) J Biol Chem, 280, 15238±15246. and MACDOUGALL, D. B. (1994) Chem. Senses, 19, 349-58. KIER, L. M. (1972) J. Pharm. Sci., 61, 1394±1397. KINNAMON, S. C. (1996) Nature, 381, 737±738. KINNAMON, S. C. and VARKEVISSER, B. (1998) ECRO XIII Congress, Abstract No. 18, Siena, Italy. KITAGAWA, M., KUSAKABE, Y., MIURA, H., NINOMIYA, Y. and HINO, A. (2001) Biochem. Biophys. Res. Commun., 283, 236±242. KRISTIANSEN, K. (2004) Pharamacol. Therap., 103, 21±80. KRIZHANOVSKY, V., AGAMY, O. and NAIM, M. (2000) Am. J. Physiol. Cell Physiol., 279, C120±C125. MAX, M.
JOHNSON, C., BIRCH, G. G.
KUHN, C., BUFE, B., WINNIG, M., HOFMANN, T., FRANK, O., BEHRENS, M., LEWTSCHENKO, T.,
and MEYERHOF, W. (2004) J. Neurosci., 24, 10260±10265. (1972) FEBS Lett., 27, 279±281. KURIHARA, K. and KOYAMA, N. (1972) Biochem. Biophys. Res. Comm., 48, 30±34. LARSEN, A.-K. R., MOLLER, M. T. N., BLANKSON, H., SAMARI, H. R., HOLDEN, L. and SEGLEN, P. O. (2002) J. Biol. Chem., 277, 34826±34835. LARSON-POWERS, N. and PANGBORN, R. M. (1978) J. Food Sci., 43, 41±46. LAWLESS, H. T. and STEVENS, D. A. (1983) Chem. Senses, 7, 309±315. SLACK, J. P., WARD, C. D.
KURIHARA, K.
LI, X., INOUE, M., REED, D. R., HUQUE, T., PUCHALSKI, R. B., TORDOFF, M. G., NINOMIYA, Y., BEAUCHAMP, G. K.
and
BACHMANOV, A. A.
(2001) Mamm. Genome, 12, 13±16. and ADLER, E. (2002) Proc. Natl. Acad.
LI, X., STASZEWSKI, L., XU, H., DURICK, K., ZOLLER, M.
Sci. USA, 99, 4692±4696. (1996) Physiol. Rev., 76, 719±766. LINDEMANN, B. (2001) Nature, 413, 219±225. LIU, D. and LIMAN, E. R. (2003) Proc. Natl. Acad. Sci. USA, 100, 15160±15165. LUM, C. K. and HENKIN, R. I. (1976) Biochim. Biophys. Acta, 421, 380±394. LUSH, I. E. (1989) Genet. Res., 53, 95±99. LUSH, I. E., HORNIGOLD, N., KING, P. and STOYE, J. P. (1995) Genet. Res., 66, 167±174. MALLER, O. and DESOR, J. A. (1974) In Oral Sensation and Perception: Development in the Fetus and Infant (Ed, Bosma, J.) US Govt Printing Office, Washington, DC, pp. 279±291. MARKS-KAUFMAN, R. and KANAREK, R. B. (1981) Psychopharmacology (Berl), 74, 321±324. MARKS-KAUFMAN, R. and LIPELES, B. J. (1982) Nutr. Behav., 1, 33±46. MASUHO, I., TATEYAMA, M. and SAITOH, O. (2005) Chem. Senses, 30, 281±290. LINDEMANN, B.
MAX, M., SHANKER, Y. G., HUANG, L., RONG, M., LIU, Z., CAMPAGNE, F., WEINSTEIN, H., DAMAK, S.
and
(2001) Nat. Genet., 28, 58±63. (1946) Br. J. Exp. Pathol., 27, 297±298. MCLAUGHLIN, S. K., MCKINNON, P. J. and MARGOLSKEE, R. F. (1992) Nature, 357, 563±569. MENCO, B. P. and HELLEKANT, G. (1993) Microsc. Res. Tech., 26, 133±141. MISAKA, T., KUSAKABE, Y., EMORI, Y., ARAI, S. and ABE, K. (1998) Ann. N.Y. Acad. Sci, 855, 150±159. MONTMAYEUR, J. P., LIBERLES, S. D., MATSUNAMI, H. and BUCK, L. B. (2001) Nat. Neurosci., 4, 492±498. MOSKOWITZ, H. R. (1977) In The Chemical Senses and Nutrition (Eds, Kare, M. R. and MARGOLSKEE, R. F.
MAYER-GROSS, W.
and
WALKER, J. W.
Stimulation of taste cells by sweet taste compounds
27
Maller, O.) Academic Press, New York, pp. 71±101. and LANDRY, Y. (1990) Trends Pharmacol. Sci., 11, 358±362. MUELLER, K. L., HOON, M. A., ERLENBACH, I., CHANDRASHEKAR, J., ZUKER, C. S. and RYBA, N. J. P. (2005) Nature, 434, 225±229. MUELLER, S. G., WHITE, J. R., SCHRAW, W. P., LAM, V. and RICHMOND, A. (1997) J. Biol. Chem., 272, 8207±8214. NAIM, M., ROGATKA, H., YAMAMOTO, T. and ZEHAVI, U. (1982) Physiol. Behav., 28, 979±986. NAIM, M., DUKAN, E., ZEHAVI, U. and YARON, L. (1986) Chem. Senses, 11, 361±370. NAIM, M., RONEN, T., STRIEM, B. J., LEVINSON, M. and ZEHAVI, U. (1991) Comp. Biochem. Physiol. B., 100, 455±458. NAIM, M., SEIFERT, R., NURNBERG, B., GRUNBAUM, L. and SCHULTZ, G. (1994) Biochem. J., 297, 451±454. NAIM, M., ZUBARE-SAMUELOV, M., PERI, I. and SHAUL, M. (2004) FASEB J, 18, A313 (abst 213.22). NELSON, G., HOON, M. A., CHANDRASHEKAR, J., ZHANG, Y., RYBA, N. J. P. and ZUKER, C. S. (2001) Cell, 106, 381±390. NELSON, G., CHANDRASHEKAR, J., HOON, M. A., FENG, L., ZHAO, G., RYBA, N. J. P. and ZUKER, C. S. (2002) Nature, 416, 199±202. NOFRE, C. and TINTI, J.-M. (1987) United States Patent 4,645,678. NOFRE, C. and TINTI, J.-M. (1996) Food Chem., 56, 263±274. NOFRE, C., TINTI, J. M. and CHATZOPOULOS-OUAR, F. (1988) Chemical Abstracts, 109, 190047k. OZECK, M., BRUST, P., XU, H. and SERVANT, G. (2004) Eur. J. Pharmacol., 489, 139±149. PAO, C. S. and BENOVIC, J. L. (2005) J. Biol. Chem., 280, 11052±11058. PEREZ, C. A., HUANG, L., RONG, M., KOZAK, J. A., PREUSS, A. K., ZHANG, H., MAX, M. and MARGOLSKEE, R. F. (2002) Nat. Neurosci., 5, 1169±1176. PERI, I., MAMRUD-BRAINS, H., RODIN, S., KRIZHANOVSKY, V., SHAI, Y., NIR, S. and NAIM, M. (2000) Am. J. Physiol. Cell Physiol., 278, C17±C25. PERSAUD, K. C., CHIAVACCI, L. and PELOSI, P. (1988) Biochim. Biophys. Acta, 967, 65±75. PHILLIPS, T. J., CRABBE, J. C., METTEN, P. and BELKNAP, J. K. (1994) Alcohol Clin. Exp. Res., 18, 931±941. PIN, J. P., GALVEZ, T. and PREZEAU, L. (2003) Pharmacol. Ther., 98, 325±354. PITCHER, J. A., FREEDMAN, N. J. and LEFKOWITZ, R. J. (1998) Ann. Rev. Biochem., 67, 653±692. PRAWITT, D., MONTEILH-ZOLLER, M. K., BRIXEL, L., SPANGENBERG, C., ZABEL, B., FLEIG, A. and PENNER, R. (2003) Proc. Natl. Acad. Sci. USA, 100, 15166±15171. PREMONT, R. T., KOCH, W. J., INGLESE, J. and LEFKOWITZ, R. J. (1994) J. Biol. Chem., 269, 6832±6841. RAY, S. (1999) Neurosci. Lett., 259, 37±40. ROGATKA, H., NAIM, M. and ZEHAVI, U. (1985) Chem. Senses, 10, 269±277. ROSENZWEIG, S., YAN, W., DASSO, M. and SPIELMAN, A. I. (1999) J. Neurophysiol., 81, 1661± 1665. SAINZ, E., KORLEY, J. N., BATTEY, J. F. and SULLIVAN, S. L. (2001) J. Neurochem., 77, 896±903. SASAKI, Y., SUZUKI, M. and HIDAKA, H. (2002) Pharamacol. Therap., 93, 225±232. SCHIFFMAN, S. S., REILLY, D. A. and CLARK, T. B. 3rd (1979) Physiol. Behav., 23, 1±9. SCHIFFMAN, S. S., CAHN, H. and LINDLEY, M. G. (1981) Pharmacol. Biochem. Behav., 15, 377±388. SCHIFFMAN, S. S., BOOTH, B. J., CARR, B. T., LOSEE, M. L., SATTELY-MILLER, E. A. and GRAHAM, B. G. (1995) Brain Res. Bull., 38, 105±120. MOUSLI, M., BUEB, J.-L., BRONNER, C., ROUOT, B.
28
Optimising sweet taste in foods
and GIBES, K. M. (1999) Chem. Senses, 24, 439±447. SCLAFANI, A. and PEREZ, C. (1997) Physiol. Behav., 61, 25±29. SHAFRIR, E. (1997) Diabetes Metab., 22, 131±148. SHALLENBERGER, R. S. and ACREE, T. E. (1967) Nature, 216, 480±482. SHIMAZAKI, K., SATO, M. and NAKAO, M. (1986) Biochim. Biophys. Acta, 884, 291±298. SIMON, M. I., STRATHMANN, M. P. and GAUTAM, N. (1991) Science, 252, 802±808. SITARAMAYYA, A. and LIEBMAN, P. A. (1983) J. Biol. Chem., 258, 12106±12109. SPIELMAN, A. I. (1998) Crit.Rev. Oral Biol. Med., 9, 267±291. SPIELMAN, A. I. and BRAND, J. G. (1997) In Encyclopedia of Human Biology, Vol. 8 (Ed, Dulbecco, R.) Academic Press, San Diego, pp. 455±466. SPIELMAN, A. I., HUQUE, T., WHITNEY, G. and BRAND, J. G. (1992) In Sensory Transduction (Eds, Corey, D. P. and Roper, S. D.) The Rockefeller University Press, New York, pp. 307±324. SCHIFFMAN, S. S., BOOTH, B. J., SATTELY-MILLER, E. A., GRAHAM, B. G.
SPIELMAN, A. I., NAGAI, H., SUNAVALA, G., DASSO, M., BREER, H., BOEKHOFF, I., HUQUE, T.,
and BRAND, J. G. (1996) Am. J. Physiol. Cell Physiol., 270, C926±C931. and NAIM, M. (2002) In Chemistry of Taste, Mechanisms, Behaviors, and Mimics, Vol. ACS Symposium Series 825 (Eds, Given, P. and Paredes, D.) American Chemical Society, Washington, DC, pp. 18±31. STADEL, J. M., NAMBI, P., SHORR, R. G. L., SAWYER, D. F., CARON, M. G. and LEFKOWITZ, R. J. (1983) Proc. Natl. Acad. Sci. USA, 80, 3173±3177. STEINER, J. (1973) In Fourth Symposium on Oral Sensation and Perception (Ed, Bosma, J. F.) US Department of Health, Education and Welfare, National Institutes of Health, Bethesda, MD, pp. 254±278. STRIEM, B. J., PACE, U., ZEHAVI, U., NAIM, M. and LANCET, D. (1989) Biochem. J., 260, 121±126. STRIEM, B. J., NAIM, M. and LINDEMANN, B. (1991) Cell. Physiol. Biochem., 1, 46±54. TELI, T., MARKOVIC, D., LEVINE, M. A., HILLHOUSE, E. W. and GRAMMATOPOULOS, D. K. (2005) Mol. Endocrinol., 19, 474±490. TINTI, J.-M. and NOFRE, C. (1996) In The Contribution of Low- and Nonvolatile Materials to the Flavor of Foods (Eds, Pickenhagen, W., Ho, C.-T. and Spanier, A. M.) Allured Publishing Corp., Carol Stream, IL, pp. 77±93. TONOSAKI, K. and FUNAKOSHI, M. (1988) Nature, 331, 354±356. VAN DER HEIJDEN, A., VAN DER WEL, H. and PEER, G. H. (1985a) Chem. Senses, 10, 57±72. VAN DER HEIJDEN, A., VAN DER WEL, H. and PEER, G. H. (1985b) Chem. Senses, 10, 73±88. VAN DER WEL, H. (1972) In Olfaction and Taste IV (Ed, Schneider, D.) Wissenschaftliche Verlagsgesellschaft GmbH, Stuttgart, pp. 226±233. VARKEVISSER, B. and KINNAMON, S. C. (2000) J. Neurophysiol., 83, 2526±2532. WHITNEY, G.
SPIELMAN, A. I., YAN, W., KRIZHANOVSKY, V., ROSENZWEIG, S., YAMAMOTO, T.
VERA, J. C., REYES, A. M., VELASQUEZ, F. V., RIVAS, C. I., ZHANG, R. H., STROBEL, P., SLEBE, J. C., NUNEZ-ALARCON, J.
and Golde,
D. W.
(2001) Biochemistry, 40, 777±790.
and (2004) Eur. J. Neurosci., 19, 1535±1544. WINNIG, M., BUFE, B. and MEYERHOF, W. (2005) BMC Neurosci, 6, 22. WONG, G. T., GANNON, K. S. and MARGOLSKEE, R. F. (1996) Nature, 381, 796±800. WONG, S. K. (2003) Neurosignals, 12, 1±12. XU, H., STASZEWSKI, L., TANG, H., ADLER, E., ZOLLER, M. and LI, X. (2004) Proc. Natl. Acad. Sci. USA, 101, 14258±14263. YANG, R., CROWLEY, H. H., ROCK, M. E. and KINNAMON, J. C. (2000) J. Comp. Neurol., 424, 205±215. VON BUCHHOLTZ, L., ELISCHER, A., TAREILUS, E., GOUKA, R., KAISER, C., BREER, H. CONZELMANN, S.
Stimulation of taste cells by sweet taste compounds ZHANG, Y., HOON, M. A., CHANDRASHEKAR, J., MUELLER, K. L., COOK, B., WU, D., ZUKER, C. S. RYBA, N. J. P.
(2003) Cell, 112, 293±301.
29 and
and (2003) Cell, 115, 255±266. ZIMMET, P., ALBERTI, K. G. M. M. and SHAW, J. (2001) Nature, 414, 782±787. ZUBARE-SAMUELOV, M., PERI, I., TAL, M., TARSHISH, M., SPIELMAN, A. I. and NAIM, M. (2003) Am. J. Physiol. Cell Physiol., 285, C1255±C1262. ZUBARE-SAMUELOV, M., SHAUL, M. E., PERI, I., ALILUIKO, A., TIROSH, O. and NAIM, M. (2005) Am. J. Physiol. Cell Physiol., 289, C483±C492. ZHAO, G. Q., ZHANG, Y., HOON, M. A., CHANDRASHEKAR, J., ERLENBACH, I., RYBA, N. J. ZUKER, C. S.
2 Genetic differences in sweet taste perception V. B. Duffy, J. E. Hayes, M. E. Dinehart, University of Connecticut, USA
2.1
Introduction
Sweet taste serves as a sensory cue for energy as well as a source of pleasure. The ability to respond to and prefer sweet taste is observed in newborn infants (Mennella and Beauchamp, 1998; Steiner et al., 2001) and probably present before birth (El-Haddad et al., 2002). However, neither the stimulus±intensity relationship nor the intensity±hedonic relationship are uniform across individuals. The impact of these sensory relationships on human preference for and intake of sweet foods is of interest to supporting food enjoyment while promoting health and preventing obesity and chronic disease. High intakes of sugars have historically created concern for human nutrition, particularly those sugars added to foods in relationship with the quality of the diet (e.g., vitamins and minerals provided) and risk of overweight or obesity (Murphy and Johnson, 2003; WHO/FAO, 2003). Nutritionists make a distinction between those sugars naturally occurring in milk and fruits and those added to foods or beverages during processing or preparation. Added sugars are typically consumed in soft drinks, artificial fruit beverages, candy, cookies, cakes, and pastries (DHHS/USDA, 2005). Organizations such as the Institute of Medicine (IOM), the United States Department of Agriculture (USDA), the Department of Health and Human Services (DHHS) and the World Health Organization (WHO) have issued recommendations on intakes of added sugar. The Dietary Reference Intakes, established by the IOM from a critical review of the literature, recommend that intakes of added sugars not exceed 25 percent of total energy, beyond which, dietary quality suffers (IOM, 2002). Although no specific intake value is set, the 2005 Dietary
Genetic differences in sweet taste perception
31
Guidelines for Americans recommends that foods and beverages be selected with little added sugars or caloric sweeteners (DHHS/USDA, 2005). Added sugars, fats, and alcohol are combined and referred to as discretionary calories (i.e., energy). Individuals are advised to consume discretionary energy only after including the recommended servings of nutrient-dense foods from each of the food groups (DHHS/USDA, 2005). Based on the consensus of an expert panel, the WHO recommends intake not exceed 10 percent of total energy (WHO/ FAO, 2003). Most empirical evidence does not support that a sweet tooth causes obesity (Rodin, 1975; Thompson et al., 1977; Grinker, 1978; Drewnowski et al., 1985; Macdiarmid et al., 1998) although some studies disagree (Cabanac and Duclaux, 1970; Conner et al., 1988). Whether this myth persists because of an attribution error (Drewnowski et al., 1985, 1992) or vilification (Fischler, 1987; Rozin, 1987) is unclear. Nonetheless, high consumers of added sugar may overconsume total energy or, conversely, compensate for surplus energy by intakes of nutrient-dense foods (Kant, 2000). The public health recommendations about added sugars must be balanced with the reality that sweetness is an important component in the pleasure of eating. The American Dietetic Association (ADA), an organization of food and nutrition professionals, believes that all foods can be part of a healthy diet when eaten in the appropriate portion size and when combined with regular physical activity (Freeland-Graves and Nitzke, 2002). The ADA strives to avoid `good' and `bad' labels for food choices and instead looks at all foods in the context of the total diet (Freeland-Graves and Nitzke, 2002). Nutrition and health professionals can assist consumers in incorporating sweet tasting foods into diets that are enjoyable, healthful and meet individual health and nutrition needs (Duffy and Sigmand-Grant, 2004). Adding sweetness to foods with high nutrient quality may increase the palatability, the chance that they are consumed. To alter the traditional aphorism, `a spoonful of sugar helps the fruits, vegetables and fiber go down.' Although myriad psychosocial, physiological and cognitive factors influence sweet preference and intake in adults, this chapter will address oral sensory factors that affect preference for and intake of sweet foods and beverages. First is a review of the literature on human variation in sweet intensity and sweet preference, and how variability in these measures associates with differences in sweet intake. Following this is a review of genetic variation in taste, described by phenotypic markers, anatomical markers, and emerging gene markers, and how these markers explain some of the variability in sweet sensations, preference and intake. 2.1.1 Individual differences in sweet intensity Although genetic diversity in sweet response is well documented in animal models (Bachmanov et al., 2001) and reviewed elsewhere (Beauchamp et al., 2002), less is known about genetic diversity in humans. Under a basic taste
32
Optimising sweet taste in foods
paradigm (see Delwiche, 1996), sweetness is typically considered a unitary percept (Halpern, 2002), that is, a singular perceptual experience wherein all sweet sensations are qualitatively similar. Although the idea of a common structural motif (i.e., universal glycophore) (Shallenberger and Acree, 1967; Eggers et al., 2000) is parsimoniously elegant, a single binding site cannot entirely explain sweet taste (DuBois, 2004). Moreover, experiencing sweetness as a singular percept does not preclude multiple transduction mechanisms. Individual sweet receptors may contain distinct binding pockets that bind diverse classes of ligands (Xu et al., 2004) and individuals taste receptor cells (TRCs) may express diverse receptors, as occurs in some bitter TRCs (Glendinning and Hills, 1997; Adler et al., 2000). Broad tuning could also result if signals from narrowly tuned TRCs converge at some point prior to reaching the anatomical structures responsible for perception. While issues of TRC tuning (Amrein and Bray, 2003) and the implications for sensory coding (Erickson, 2000) are beyond the scope of this chapter, evidence supportive of multiple sweet receptor mechanisms in humans and variation between individuals and across sweeteners is discussed below. Evidence for multiple sweet mechanisms Three distinct lines of psychophysical evidence support the idea of multiple sweet mechanisms in humans. One measure of mechanism independence is the failure to cross-adapt; in a cross-adaptation paradigm, pretreatment with one tastant reduces the response to the test stimuli if the two tastants share a similar mechanism (McBurney et al., 1972; Keast and Breslin, 2002). McBurney (1972) found that sucrose cross-adapted all 13 sweet tastants tested whereas saccharin did not. Subsequently, Lawless and Stevens (1983) tested sucrose and three high intensity sweeteners for self- and cross-adaptation. All sweeteners self-adapted and sucrose cross-adapted aspartame, saccharin, and dihydrochalcone. Conversely, when one of the high intensity sweeteners was the adapting stimulus, cross-adaptation was incomplete. These findings were later extended by the demonstration that the ability of sugar sweeteners (sucrose, fructose) and high-intensity sweeteners (acesulfame-K, dulcin) to cross adapt is related to the number of shared binding motifs (Froloff et al., 1998). The second piece of evidence for multiple mechanisms is the existence of synergy; if two sweet compounds act via the same mechanism, mixtures of the two should behave in an additive manner. Instead, mixtures of acesulfame-K and aspartame exhibit synergy (Ayya and Lawless, 1992; Lawless, 1998; Keast et al., 2004), wherein the mixture elicits greater sweetness than that predicted via addition. Finally, in other studies, sweet blockers ± like sodium-PMP (Schiffman et al., 1999) and zinc sulfate (Keast et al., 2004) ± inhibit some but not all sweeteners. Collectively, these findings strongly support the hypothesis of multiple sweet receptors in humans. Variation across individuals In contrast to ample behavioral variation in animals (Beauchamp et al., 2002), reports of phenotypic variation in humans are scattered at best. In an early report
Genetic differences in sweet taste perception
33
by Richter and Campbell (1939), 13 of 45 young adults exhibited elevated sucrose recognition thresholds (>17.5 mM), well in excess of the population mode (5.8 mM) and the modern accepted value (3 mM). When matching sucrose solutions to the intensity of a 29 mM sodium chloride reference, Faurion and colleagues (1993) found the subjects (n 55) required solutions that ranged in concentration from 30 to 300 mM, an entire order of magnitude. From our laboratory, 87 adults (mean age = 26 4 yrs, 42 females, 45 males) reveal variability in intensity ratings of a single sucrose (0.32 M) and aspartame (1 mM) concentration. Both sweeteners produce a normal distribution with ratings ranging from near weak to near very strong (Fig. 2.1) on the general Labeled Magnitude Scale (gLMS) (Bartoshuk et al., 2004b). Ratings normalized
Fig. 2.1 Histograms of intensity ratings of sucrose and aspartame on the general Labeled Magnitude Scale (Bartoshuk et al., 2004b), where 6 = weak, 17 = moderate, 35 = strong, 53 = v. strong.
34
Optimising sweet taste in foods
to the intensity of tones have similar shaped distribution and degree of variability. The variability of intensity of these sweeteners does not appear to be due to idiosyncratic intensity ratings. Comparing the variance for each sweetener to a tone of equal mean intensity shows significantly greater variance for sucrose (F(86) 1.85, p < 0:01) and aspartame (F(81) 2.40, p < 0:01). Variation across sweeteners Even within individuals, not all sweeteners elicit equivalent responses. Among simple carbohydrates, glucose exhibits a markedly different recognition function from sucrose and fructose (Kennedy et al., 1997). Eylam and Kennedy (1998) subsequently identified some individuals (16 of 92 screened) who were hypogeusic to glucose but not fructose (n 12) or vice versa (n 4). In another report (Faurion, 1993), some subjects exhibited diverse responses to fructose and sucrose sweetness. Likewise, the relative sweetness of fructose to sucrose, expressed as a ratio, shows a large amount of inter-subject variability (Fontvieille et al., 1989). High-intensity sweeteners evoke sweet sensations at low concentrations and are diverse in chemical structure. Variability in the sweetness from high-intensity sweeteners has been attributed in part to the presence of bitter or metallic side tastes and to duration of sweetness (lingering). From our laboratory, we find a correlation between intensity ratings of sucrose and aspartame, a high-intensity sweetener that is purported to limit less desirable sensory characteristics (Fig. 2.2). Nonetheless, high residuals (circled) may represent distinct responses to these sweeteners, rather than measurement error.
Fig. 2.2 Joint distribution of intensity ratings from sucrose and aspartame on the general Labeled Magnitude Scale (Bartoshuk et al., 2004b). Circled are individuals who exceed the residual seen across the rest of the sample.
Genetic differences in sweet taste perception
35
2.1.2 Individual differences in sweet hedonics Numerous researchers have noted great diversity in sweet liking and acceptance (Cameron, 1947; Pangborn, 1970; Thompson et al., 1977; Witherly et al., 1980; Meiselman, 1987; Booth et al., 1987). Histograms from our laboratory demonstrate this diversity in the same sample of adults shown in previous figures (Fig. 2.3). With increasing intensity, liking functions usually take the form of an inverted U-shape (Moskowitz, 1971, 2002). Merely averaging across a sample may obfuscate individual differences in liking (Lundgren et al., 1978; Moskowitz and Moskowitz, 2000). Thompson and colleagues (Thompson et al., 1977) identified Type I individuals, who showed an inverted U liking function with increasing sucrose concentration, in contrast to Type II individuals, who showed a monotonic increase with concentration. Pangborn and her students recapitulated these hedonic functions and noted two additional functions; individuals who showed a monotonic decrease in liking and those who showed a flat relationship between concentration and liking (Lundgren et al., 1978; Witherly et al., 1980), patterns which were later described (Drewnowski, 1987) as Type III and Type IV responses, respectively.
Fig. 2.3 Histograms of liking/disliking ratings of sucrose and aspartame on the general Labeled Magnitude Scale (Bartoshuk et al., 2004b), where +/-6 = weakly, 17 = moderately, 35 = strongly, 53 = v. strongly.
36
Optimising sweet taste in foods
Fig. 2.4 Individuals categorized as dislikers and likers based on hedonic ratings on the general Labeled Magnitude Scale (Bartoshuk et al., 2004b) to increasing concentration of sucrose.
Alternatively, individuals have been classified as sweet likers or dislikers based on a positive or negative slope for that individual's hedonic-intensity function (Looy and Weingarten, 1992); a classification scheme which would collapse Type I (inverted U) and Type III (monotonic decrease) into dislikers, with Type II representing likers. Figure 2.4 shows adults classified by this method: 21 individuals could be classified as dislikers (type I and type III responders), and 42 as likers.
2.2
PTC/PROP tasting
The ability to taste the bitterness of the thiourea compounds phenylthiocarbamide (PTC) and PROP is a well documented phenotypic polymorphism (Richter and Clisby, 1941; Hall et al., 1975; Lawless, 1980; Bartoshuk et al., 1994). The existence of this taste blindness was discovered serendipitously in 1931 when, during the synthesis of PTC, Fox (1931) accidentally released some into the air. A colleague commented on the intense bitter taste of the dust while Fox tasted nothing. The distribution of thresholds to PTC or PROP is bimodal; nontasters exhibit elevated thresholds (low sensitivity) while tasters show lower thresholds (higher sensitivity) (see Fox, 1931; Harris and Kalmus, 1949). Early family studies suggested that PTC nontasters carried two recessive alleles and tasters carried either one or both dominant alleles (Snyder, 1931; Blakeslee, 1932). Following the lead of Fischer and Griffen (1961), PROP eventually replaced PTC as the marker of choice because of concerns with the safety of PTC and because the sulfurous odor of PTC could provide a nontaste cue in a detection task (Lawless, 1980).
Genetic differences in sweet taste perception
37
2.2.1 Supertasting and PROP Using scaling, Bartoshuk and colleagues (1994) found that tasters could be further separated into medium tasters, for whom concentrated PROP solutions are mildly bitter, and supertasters, for whom it is intensely bitter. Supertasters cannot be identified via thresholds as the distributions between those who are sensitive and extremely sensitive to PROP overlap (Reed et al., 1995). The underlying reason behind supertasting is unclear, although a `general taste function' (Lawless, 1979a), differences in fungiform papillae density (Bartoshuk et al., 1994; Delwiche et al., 2001) or some sort of central gain (Green and George, 2004; Green and Hayes, 2004) remain as possible explanations. The fungiform papillae hypothesis is supported by observations that supertasters have, on average, the greatest number of fungiform papillae and taste buds as assessed with videomicroscopy (Bartoshuk et al., 1994) and minimal magnification (Tepper and Nurse, 1997; Delwiche et al., 2001). Early work on the genetics of PTC tasting suggested a two-locus model where one locus controls PTC tasting and the other, a more general taste ability (Olson et al., 1989). The genetic control of fungiform papillae density is unknown at present. More recently, the response to PTC (Kim et al., 2003) and PROP (Duffy et al., 2004b; Bufe et al., 2005) associates with the T2R38 receptor gene, which is located on chromosome 7q36. Three SNPs (single nucleotide polymorphisms) result in five distinct haplotypes for this gene, two of which are common (Kim et al., 2003). The two common molecular haplotypes (PAV and AVI) are named for their respective three amino acid substitutions: Pro49Ala, Ala262Val, and Val296Ile. From non-human primates (Kim et al., 2003; Wooding et al., 2004), the ancestral haplotype is PAV, which associates with tasting, while the other common form is AVI, which associates with nontasting. The polymorphism of the T2R38 gene across the globe could result from balancing selection (Wooding et al., 2004) but also purifying selection and random genetic drift (Kidd et al., 2004). Although the T2R38 gene explains a large proportion of the variability in threshold response to PTC (Wooding et al., 2004; Prodi et al., 2004) and PROP (Duffy et al., 2004c; Mennella et al., 2005), this gene cannot explain supertasting (Bartoshuk et al., 2004a; Duffy et al., 2004c; Hayes et al., 2005).
2.3 Sweetness: relation to PROP, fungiform papillae density, T2R38 Bartoshuk (1979) noted that magnitude estimates of the sweetness of both saccharin and sucrose were lower in subjects who exhibited elevated PROP thresholds (i.e., nontasters). Subsequently, a magnitude matching protocol confirmed that PROP nontasters tasted sucrose, saccharin, and dihydrochalcone as less sweet than did tasters (Gent and Bartoshuk, 1983). Similarly, subjects with lower PROP thresholds exhibit lower recognition thresholds for both sucrose and fructose (Pasquet et al., 2002). Since threshold is often a poor
38
Optimising sweet taste in foods
Fig. 2.5 Associations between PROP threshold and intensity of sucrose (top) and aspartame (bottom).
predictor of suprathreshold response (Bartoshuk, 1978; Pangborn, 1980) and does not identify supertasters (Bartoshuk et al., 1994), PROP effects on sweetness are attenuated when classifying PROP tasting with thresholds. Figure 2.5 shows data from adults in our laboratory to demonstrate this statement. Most studies report an association between PROP bitterness and the sweetness of sucrose solutions (Duffy et al., 2003; Bartoshuk, 2000; Bartoshuk et al., 2003; Marks et al., 1992), although the effect size may be underestimated in some because of context and ceiling effects (Bartoshuk, 2000; Bartoshuk et al., 2003; Lawless, 1983; Lawless et al., 2000). Conversely, studies that fail to find an association may have context (Lawless, 1979b) or ceiling effects (Drewnowski et al., 1997b, 1998; Smagghe and Louis-Sylvestre, 1998; Ly and Drewnowski, 2001), which emphasizes the importance of psychophysical
Genetic differences in sweet taste perception
Fig. 2.6
39
Associations between the bitterness of concentrated PROP and intensities of sucrose (top) and aspartame (bottom).
methods (Bartoshuk et al., 2003). Data from our laboratory (Fig. 2.6) show PROP effects on sucrose and aspartame intensity. Partial coefficients are reported in a model that included the intensity of 1000 Hz tones at 74 dB as a non-oral sensory standard. PROP effects on sweetness are blunted or nonexistent when using category scales that do not permit valid comparisons of intensity ratings across subjects (Prutkin et al., 2000). The relationship between PROP bitterness and sweetness is mediated in part by the number of fungiform papillae. This point is demonstrated in multiple regression analyses to explain variability in sweetness of 1 M sucrose across 85 adults (20 to 59 years of age) from our laboratory. Those with more fungiform papillae (sr = 0.32, p 0:001) and who were younger (sr ÿ0.38, p 0:001)
40
Optimising sweet taste in foods
tasted greater sweetness. Sweetness also varied by T2R38 gene (sr 0.23, p 0:02). In analysis of variance with planned comparisons and using t-tests with the error term generated by the ANOVA, the AVI/AVI nontasters reported significantly greater intensity than did those who were PAV/PAV tasters (p < 0:05). A similar finding was seen for aspartame intensity across a cohort of 55 adults (20 to 40 years) from our laboratory. Greater sweetness of 3.2 mM aspartame was tasted by those with greater numbers of fungiform papillae (sr 0.32, p < 0:05) and influenced by the T2R38 genotype (sr = 0.26, p 0:05). In pairwise comparisons, AVI/AVI nontasters reported significantly greater intensity than did the PAV/PAV tasters (p 0:05). Mixtures can blunt or even reverse PROP effects on sweetness. In mixtures of simple tastants, sucrose sweetness is attenuated in medium and supertasters in two-component mixtures by high concentrations of quinine (240 to 360 mM) (Prescott et al., 2001; Yiee et al., 2002, 2003) and citric acid (14 mM) (Yiee et al., 2002, 2003); the suppression is seen also in three- and four-component mixtures (Yiee et al., 2002, 2003). Sucrose added to strong coffee mostly blunts the bitterness for tasters (Ly and Drewnowski, 2001) and supertasters (Yiee et
Fig. 2.7
Associations between PROP bitterness and sweetness (top) and bitterness (bottom) from DewarsÕ Scotch (Lanier et al., 2005).
Genetic differences in sweet taste perception
41
al., 2002, 2003). Within other non-alcoholic beverages, a positive relationship is seen between PROP and sweetness in a commercial lemonade, which was tasted less sour than commercial grapefruit juice (Yiee et al., 2002, 2003). In contrast, sour/bitter or irritating qualities in beverages may conceal PROP effects on sweetness, particularly for supertasters, as seen with grapefruit juice and tonic water (Yiee et al., 2002, 2003), orange juice or citric acid/carbonated mixtures (Prescott et al., 2004). Components of a mixture can also completely reverse PROP effects on sweetness as seen in perceptually complex beverages such as scotch (Fig. 2.7); greater bitterness and possibly irritation, suppresses the sweetness in supertasters (Lanier et al., 2005). High-intensity sweeteners with bitter tastes act perceptually as mixtures as some are known for having bitter or metallic side tastes (Ayya and Lawless, 1992; Schiffman et al., 1995). One study, which assessed PROP tasting by threshold, found that nontasters tasted less suprathreshold bitterness from saccharin (Bartoshuk, 1979). Subsequent work suggests that these PROP effects were due to differences in general taste sensitivity (Lawless, 1979a; Olson et al., 1989) rather than a shared receptor mechanism. Modern scaling methods found that while acesulfame K and saccharin bitterness covary within subjects, they were independent of PROP bitterness (Horne et al., 2002). The independence of PROP tasting from acesulfame K and saccharin bitterness was subsequently confirmed by evidence in humans that PROP tasting associates with the T2R38 genotype (Duffy et al., 2004c) while saccharin and acesulfame K bitterness occur via T2R43/44 (Kuhn et al., 2004). Thus, product developers and sensory specialists are cautioned that screening panelists for PROP tasting may not identify individuals sensitive to the side tastes of some high intensity sweeteners.
2.4 Relating genetic taste markers with dietary sweet behaviors The idea that individual differences in taste perception may influence preference and intake dates back over 40 years (Fischer et al., 1963; Kang et al., 1967), though work in this area has typically focused on associations between PROP/ PTC tasting and foods/beverages thought to be bitter (see Mattes and Labov, 1989; Mattes, 1994; Drewnowski et al., 1999, 2000; Kaminski et al., 2000; Duffy et al., 2004c). Recent advances in molecular genetics have allowed the direct associations between T2R38 haplotype and dietary behaviors (Duffy et al., 2004a, 2004c; Mennella et al., 2005). The general hypothesis for these studies has been that genetic variation in taste and oral sensation influences food and beverage sensations and thus preference. Since we consume what we prefer and avoid what we do not, genetic variation in taste and oral sensation ultimately affects dietary selection. Below we explore this general hypothesis applied to sweet preference and intake by utilizing a number of markers of genetic variation in taste and oral sensation.
42
Optimising sweet taste in foods
2.5 Sweet liking: relation to PROP, fungiform papillae density, T2R38 Classification of sucrose likers and dislikers by PROP threshold reveals that dislikers are almost exclusively tasters while likers are evenly split between tasters and nontasters (Looy and Weingarten, 1992). In agreement, data from our laboratory shows that dislikers and likers (as shown in Fig. 2.4) differ in PROP bitterness. Dislikers (n 21) were significantly more likely to be PROP tasters, both medium and supertasters, whereas likers (n 42) were significantly more likely to be PROP nontasters (Fig. 2.8). A number of studies do not find differences in PROP tasting among sweet liker and disliker groups (Drewnowski et al., 1997a, 1997b, 1998). The PROP relationship with sweet liker/disliker classification is probably not unique to PROP per se, but rather due to an overall difference in oral sensation. Genotyping for the T2R38 receptor was available in a sub-sample of those classified as dislikers (10 of 21) and likers (29 of 42). A chi-square statistic close to zero indicates that the observed distribution equaled expected for T2R38 genotype across dislikers (4 of 10 were AVI/AVI nontasters; 6 of 10 were AVI/PAV or PAV/PAV tasters) and likers (13 of 29 were AVI/AVI nontasters; 16 of 29 were AVI/PAV or PAV/PAV tasters). Thus, the liker/disliker distribution difference in PROP tasting probably is less likely to result from the presence or absence of a functional PTC/PROP receptor and more likely to result from factors such as density of fungiform papillae or central response to sweetness. The inability to uncover phenotypic PROP effects in some studies likely results from misclassification bias that attenuates measurement of the underlying relationship. Use of psychophysical methods that do not accurately characterize individual differences in oral sensations will result in the failure to uncover sensory differences that associate with liking and disliking.
Fig. 2.8 The joint distribution of PROP nontasters (PROP=moderate) and tasters by sucrose liking/disliking categories. Two of 21 dislikers were nontasters, 19 were tasters; 30 of the likers were nontasters, 12 were tasters.
Genetic differences in sweet taste perception
43
Studies from our laboratory have seen a relationship between PROP bitterness and preference for sweet foods in adults of normal adiposity (body mass index 25) assessed by questionnaire and by ratings for sampled sweet foods. Since high cognitive restraint may influence preference ratings for sweet and high-energy foods (e.g., Lahteenmaki and Tuorila, 1993; Kanarek et al., 1995), we used validated instruments to exclude individuals with high levels of cognitive restraint from participation (e.g., Herman and Polivy, 1980; Polivy et al., 1994; Stunkard and Messick, 1985). Liking of sweet foods (ice cream, doughnuts, cookies, cake, pie, milk chocolate) assessed via questionnaire decreased with increasing PROP bitterness in women (n 24), but showed a flat association in men (n 22) (Duffy and Bartoshuk, 2000). Subsequent analyses reveals that individuals with low fungiform papillae and low PROP tasting reported greater liking for sweet foods than those with high papillae and high PROP tasting (t(34) 2.05, p < 0:05). Thus, the sex effect on sweet preference could have resulted from differences in density fungiform papillae. In a separate study of 38 females and 44 males (mean age 26 4 years) (Duffy et al., 2003), we examined associations between sweet preference (questionnaire and sampled sweet foods) and PROP as well as quinine, another bitter marker of variability in taste (see below for more discussion about quinine and sweet behaviors). PROP only tended to explain variability in preference across sweet foods from questionnaire, yet was a significant predictor of preference ratings for sampled sweet foods (Fig. 2.9). Those who tasted PROP as least bitter reported significantly greater preference for sampled foods (white icing, milk chocolate, marshmallow, grape jelly), independent of sex, adiposity, level of hunger and sweetness of these foods. Additionally, those who tasted PROP as least bitter consumed significantly more energy from added sugars (assessed via nonconsecutive food records) and tended to consume more sweet foods (assessed by food frequency survey). These findings show that PROP can explain variability in sweet liking and intake among younger adults (mean age below 30), who are primarily of European ancestry, of normal adiposity and have low levels of dietary restraint. Other studies have not seen PROP effects on sweet hedonics. Conceivably, the failure of some (Drewnowski et al., 1997b, 1998; Ly and Drewnowski, 2001) to find an effect could result from either scaling issues (Duffy and Bartoshuk, 2000; Bartoshuk et al., 2003) or the inclusion of samples with diversity in ancestry, body weight, and control over eating. Recent work from the Mennella laboratory shows that the T2R38 genotype explains variability in sweet preference in 143 children (ages 5 to 10 years) and their mothers (mean age 35 years), who were African American or non-Hispanic white (Mennella et al., 2005). Children and mothers provided preference ratings for sampled sucrose solutions, recalled their favorite cereals, and, for mothers, how much sugar they added to coffee. Children who were tasters by genotype preferred higher sucrose concentration than children who had nontaster genotypes, findings that disagree PROP phenotype-sweet relationships found in our laboratory and others (Looy and Weingarten, 1992; Lin, 2003). Genotype did not predict sweet preference in mothers because of the influences of ancestry
44
Optimising sweet taste in foods
Fig. 2.9 Preference for sampled sweet foods (white icing, milk chocolate, marshmallow, grape jelly) by PROP bitterness (top) and T2R38 genotype (bottom).
on sweet preferences ± individuals of African ancestry report greater preferences for sweets than those of European ancestry (Mennella et al., 2005; Greene et al., 1975; Schiffman et al., 2000). We were able to analyze the relationship between T2R38 genotype and dietary sweet behaviors in a subset of individuals (n 55) from our laboratory for whom this genotyping was available. In multiple regression equations controlling for sex, age, and adiposity as described (Duffy et al., 2003), the T2R38 genotype did not explain significant variance in preference for sweet foods from questionnaire (p 0:66), intake of energy from added sugars
Genetic differences in sweet taste perception
45
(p 0:72), or frequency of consuming sweet foods (p 0:48). The T2R38 genotype did predict significant variance in preference for sampled sweet foods such that those who had the nontaster genotype (AVI/AVI) reported significantly greater preference for these foods than did the homozygote taster (PAV/PAV) as shown in Fig. 2.9. Interestingly, this agrees with findings of Mennella et al. (2005), who found an interaction between genotype and ancestry for sucrose preference in the mothers. For African American mothers, tasters by genotype showed a higher sucrose preference than did nontasters by genotype, although not significant because the means plus variances overlapped. However, for mothers of European ancestry, nontasters by genotype reported a significantly higher sucrose preference than did tasters by genotype. Thus, when comparing groups with similar ancestry and age, our findings and those from Mennella's laboratory (2005) are consistent ± nontasters have greater sweet preference than do tasters by genotype or supertasters by phenotype. However, the finding is not robust across all measures of sweet preference and intake and across groups that vary in age and ancestry, suggesting that other measures of variance in taste must be considered when trying to explore sensory factors driving sweet behaviors. In summary, the work described in the previous section raises more questions than it answers. It appears that individuals who vary in sweet preference do have differences in oral sensation that may or may not be dependent on the presence or absence of a single bitter receptor. Individuals who taste PROP as more bitter show less liking for foods or beverages that are very sweet ± possibly because of greater innervation density (e.g., numbers of fungiform papillae) or differences in the manner in which tastes, nontaste oral sensations (creaminess, viscosity) and flavor (including retronasal olfaction) are perceived. This section also suggests the need to find additional markers of oral sensory variation that can explain differences in sweet preference.
2.6
Variation in other bitter markers
Fischer and colleagues (1961) were the first to examine variability in PROP and quinine tasting pertaining to food preferences. They compared the characteristic bimodal threshold distribution for PROP with the normal distributions produced by stimuli representing all four taste qualities, with particular emphasis on quinine thresholds (Fischer and Griffin, 1961). Fischer believed that his quinine measure reflected a general ability to taste; combining this general measure with the specific ability to taste PTC/PROP compounds produced a group of `acute tasters' who had more food dislikes, tended to avoid smoking (Fischer et al., 1963) and were thinner (Fischer et al., 1966). Using PROP and quinine as markers of variation in taste has been shown to enhance explanation of alcohol (Lanier et al., 2005) and sweet (Duffy et al., 2003) preference and intake. A subset of individuals exists who are discordant in PROP and quinine bitterness. These individuals taste greater bitterness from one marker relative to
46
Optimising sweet taste in foods
the other and, more interestingly, differ in dietary behaviors. Those tasting PROP as less than `strongly bitter' but quinine as above `strongly bitter' have greater preference for and intake of sweet foods than those with the opposite bitter taste profile (Duffy et al., 2003). Recent work in our laboratory has produced similar results; those tasting quinine as relatively more bitter than PROP also consume alcoholic beverages most frequently (Dinehart et al., 2005). Quinine was found to explain variability in a number of measures of sweet preference and intake while PROP only explained variability in some of these behaviors, particularly in younger participants (Dinehart et al., 2005; Duffy et al., 2003). Discordance between PROP and quinine bitterness may result from damage to the chorda tympani nerve (Bartoshuk et al., 1995), a branch of cranial nerve VII that supplies taste to the anterior two-thirds of the tongue. Taste damage from this nerve releases inhibition on cranial nerves IX and X and can result in increases in quinine bitterness as perceived with the whole mouth (Kveton and Bartoshuk, 1994; Halpern and Nelson, 1965; Yanagisawa et al., 1998). Another explanation for discordance between PROP and quinine bitterness may be that bitterness of these compounds results from differential expression of multiple bitter receptors (Margolskee, 2002). Since PROP and quinine do not totally cross-adapt they may stimulate different taste receptors, which would explain differences in bitter perception between the two (McBurney, 1972). Supportive of the independence of quinine from PROP and PTC is the observation that, at least among Japanese dental students, PTC nontasters are more quinine sensitive than are PTC tasters (Sato et al., 1997). It is uncertain whether quinine represents another genetic taste variant, a measure of overall taste sensation, or an environmentally mediated pattern of sensation. Nonetheless, we believe that using multiple markers of variation in taste and oral sensation improves the ability to explain differences in preference for and intake of sweets.
2.7 SAC Gene: human homolog and hedonic response to sweeteners In mice, varied strains differ as to whether they are `taster' mice that prefer sweet solutions or `nontaster' strains that failed to prefer sweet solutions to water. These phenotypic differences were linked to allelic variation in the saccharin preference (Sac) locus, the T1R3 taste receptor, which forms a heterodimer with T1R1 (Max et al., 2001; Montmayeur et al., 2001; Sainz et al., 2001; Nelson et al., 2001; Bachmanov et al., 2001). The T1R3 gene from a region on chromosome 1 (1p36) appears to be the human homolog of the mouse SAC, a chromosomal region that also contains T1R1, T1R2 genes (Liao and Schultz, 2003). T1R2 and T1R3 have complementary functional roles to form a heteromeric taste receptor complex that responds to structurally diverse sweeteners (Xu et al., 2004).
Genetic differences in sweet taste perception
47
Table 2.1 Description of SAC gene (Liao and Schultz, 2003) groups (mean SD) Heterozygote Sex Age* BMIy FP
2 F; 4 29 26.3 25
M 4 4.9 9
Homozygote 15 F; 19 M 25 4 23.4 2.8 23 7
* p < 0:1 y p < 0:05
Reed and colleagues (2002) presented preliminary data on behaviors toward sucrose and aspartame in humans who varied on the human homolog of the mouse SAC, T1R3. The psychophysical data were from our laboratory, the Reed laboratory sequenced DNA from whole bloods. The sample was from adults previous described (Duffy et al., 2003); only those who reported being of European ancestry are shown (Table 2.1). Six heterozygous and 34 homozygous subjects were identified. The heterozygotes tended to be older and averaged significantly heavier, although the BMI distribution between the two groups was not significantly different (Fisher's Exact Test, p 0:16). The subjects rated the intensity and liking/disliking for a sucrose concentration series (Fig. 2.10), and intensity and liking/disliking for a single concentration of aspartame (3.2 mM). A group by concentration ANCOVA that controlled for age and BMI showed significant main effects of group (F(1, 35) 6.09, p 0:02) and concentration (F(2,76) 6.58, p 0:002): the heterozygotes showed a steeper rise in hedonic response to increasing concentrations of sucrose. The sucrose intensities between the two groups were overlapping. A similar trend was seen for aspartame. In ANCOVA controlling for covariance from age and BMI, the heterozygotes tended to rate the 3.2 mM aspartame as more pleasant (p < 0:1) but not vary in its intensity (p 0:65). These data show that another gene marker may explain some variability in sweet liking in humans and, that according to preliminary analyses, was not associated with perceived sweetness. 2.7.1 Implication for designing foods that meet consumers' sweet taste Sweetness is a pleasurable sensation yet the consumption of sweet foods and beverages has been under increasing scrutiny by nutrition professionals as well as the lay public because of its potential connection with weight management and dietary quality. Our obesiogenic environment provides ready access to good tasting, high-energy foods and surrounds us with technology that minimizes energy expenditure. The food supply needs to offer foods and beverages that taste sweet but, when ingested, do not displace healthful, nutrient-rich foods. Additionally, adding sweeteners to the food supply should ultimately aim to improve dietary quality while promoting food enjoyment (Duffy and Sigmand-Grant, 2004).
48
Optimising sweet taste in foods
Fig. 2.10 Intensity (top) and hedonic (bottom) functions of sucrose in individuals who vary in the human SAC gene (Liao and Schultz, 2003).
The present chapter has reviewed genetic variation in taste and oral sensation and how markers of this variation have been shown to associate with dietary sweet behaviors, primarily in studies that employ convenience samples of adults recruited for academic research in contrast to efforts by industry that use consumer or trained panels for product acceptability testing or market research. The studies support that markers of variability in taste do explain some of the variability in sweet behaviors. Although the ability of laboratory tests to predict consumption has been questioned (Lucas and Bellisle, 1987), and the failure of taste tests to predict market performance has been discussed (Garber et al.,
Genetic differences in sweet taste perception
49
2003), we contend that laboratory testing can, and does predict real-world consumption. However, the observation that `for hedonic measurements, group averages may be misleading or even completely artifactual (Lundgren et al., 1978)' emphasizes the criticality in identifying different market segments (Garber et al., 2003; Wansink, 2003) for whom the product is intended. As noted by Buck (2003), consumers vary widely in their acceptance of commercially available products. Some suggest that identifying potential segmentation variables (usage frequency, personality and lifestyle factors) can enhance the value of sensory testing (Wansink, 2003). At a minimum, the decision as to whether a trained panel should be according to specific markers of taste, or show a representational cross-section is entirely dependent on the goals of the research, whether market or sensory research. While we do not anticipate the appearance of products tailored to specific genotypes, both sensory practitioners and basic researchers (Cardello, 2003) can benefit from an increased understanding of how genetic variation in taste and oral sensation can influence liking.
2.8
Acknowledgements
The project was supported by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, grant number 2003-35200-12943 and NIH Institutes of Deafness and Communication Disorders grant number DC00283.
2.9
References
and ZUKER, C. S. (2000) Cell, 100, 693±702. AMREIN, H. and BRAY, S. (2003) Cell, 112, 283±284. AYYA, N. and LAWLESS, H. T. (1992) Chem Senses, 17, 245±259. BACHMANOV, A. A., TORDOFF, M. G. and BEAUCHAMP, G. K. (2001) Chem Senses, 26, 905± 913. BARTOSHUK, L. (1978) Am J Clin Nutr, 31, 1068±1077. BARTOSHUK, L. M. (1979) Science, 205, 934±935. BARTOSHUK, L. M. (2000) Chem Senses, 25, 447±460. BARTOSHUK, L. M., DUFFY, V. B. and MILLER, I. J. (1994) Physiol Behav, 56, 1165±1171. BARTOSHUK, L., DUFFY, V., REED, D. and WILLIAMS, A. (1995) Neurosci Biobehav Rev, 20, 79±87. BARTOSHUK, L. M., DUFFY, V. B., FAST, K., GREEN, B. G., PRUTKIN, J. and SNYDER, D. J. (2003) Food Qual Pref, 14, 125±138. ADLER, E., HOON, M. A., MUELLER, K. L., CHANDRASHEKAR, J., RYBA, N. J.
BARTOSHUK, L. M., DAVIDSON, A., KIDD, J., KIDD, K., SPEED, W., PAKSTIS, A., REED, D. R., SNYDER, D. J.
and
DUFFY, V. B.
(2004a) Chem Senses, 30, A78±A88.
BARTOSHUK, L. M., DUFFY, V. B., GREEN, B. G., HOFFMAN, H. J., KO, C. W., LUCCHINA, L. A., MARKS, L. E., SNYDER, D. J.
and WEIFFENBACH, J. M. (2004b) Physiol Behav, 82, 109±114. and BACHMANOV, A. A. (2002) In Chemistry of
BEAUCHAMP, G. K., REED, D. R., TORDOFF, M. G.
50
Optimising sweet taste in foods
taste: mechanisms, behavior, and mimics (Eds, Given, P. and Paredes, D.) American Chemical Society, Washington, DC, pp. 40±51. BLAKESLEE, A. (1932) Proc Nat Acad Sci, 18, 120±130. BOOTH, D. A., CONNER, M. T. and MARIE, S. (1987) In Sweetness (Ed, Dobbing, J.) SpringerVerlag, London, New York, pp. 143±160. BUCK, D. (2003) Food Quality and Preference, 14, 37±38. BUFE, B., BRESLIN, P. A., KUHN, C., REED, D. R., THARP, C. D., SLACK, J. P., KIM, U. K., DRAYNA, D.
and
(2005) Curr Biol, 15, 322±237. and DUCLAUX, R. (1970) Science, 168, 496±497. CAMERON, A. T. (1947) The taste sense and the relative sweetness of sugars and other sweet substances, Sugar Research Foundation, New York, 72 pp. CARDELLO, A. (2003) Food Quality and Preference, 14, 27±30. CONNER, M. T., HADDON, A. V., PICKERING, E. S. and BOOTH, D. A. (1988) J Appl Psychol, 73, 275±280. DELWICHE, J. (1996) Trends in Food Science & Technology, 7, 411±415. DELWICHE, J. F., BULETIC, Z. and BRESLIN, P. A. (2001) Physiol Behav, 74, 329±337. DHHS/USDA (2005) Washington, DC: US Government Printing Office. DINEHART, M., BARTOSHUK, L. and DUFFY, V. (2005) In 26th Annual Meeting of AChemS, Vol. 30 Chem Senses, Sarasota FL, pp. A62. DREWNOWSKI, A. (1987) In Sweetness (Ed, Dobbing, J.) Springer-Verlag, London, New York. DREWNOWSKI, A., BRUNZELL, J. D., SANDE, K., IVERIUS, P. H. and GREENWOOD, M. R. (1985) Physiol Behav, 35, 617±622. DREWNOWSKI, A., KURTH, C., HOLDEN-WILTSE, J. and SAARI, J. (1992) Appetite, 18, 207±221. DREWNOWSKI, A., HENDERSON, S. A. and SHORE, A. B. (1997a) Chem Senses, 22, 27±37. DREWNOWSKI, A., HENDERSON, S. A., SHORE, A. B. and BARRATT-FORNELL, A. (1997b) Physiol Behav, 62, 649±655. DREWNOWSKI, A., HENDERSON, S. A. and BARRATT-FORNELL, A. (1998) Physiol Behav, 63, 771±777. DREWNOWSKI, A., HENDERSON, S. A., LEVINE, A. and HANN, C. (1999) Public Health Nutr, 2, 513±519. DREWNOWSKI, A., HENDERSON, S. A., HANN, C. S., BERG, W. A. and RUFFIN, M. T. (2000) J Am Diet Assoc, 100, 191±197. DUBOIS, G. E. (2004) Proc Natl Acad Sci USA, 101, 13972±13973. DUFFY, V. B. and BARTOSHUK, L. M. (2000) J Am Diet Assoc, 100, 647±655. DUFFY, V. and SIGMAND-GRANT, M. (2004) J Am Diet Assoc, 104, 255±275. DUFFY, V. B., PETERSON, J., DINEHART, M. and BARTOSHUK, L. M. (2003) Top Clin Nutr, 18, 209±220. DUFFY, V., DAVIDSON, A., BARTOSHUK, L., KIDD, J. and KIDD, K. (2004a) Experimental Biology meeting abstracts, The FASEB Journal, 19, abstract no. 6780. Washington, DC. MEYERHOF, W.
CABANAC, M.
DUFFY, V. B., DAVIDSON, A., KIDD, J., KIDD, K., SPEED, W., PATKSTIS, A., REED, D. R., SNYDER, D. J.
and BARTOSHUK, L. M. (2004b) In AChemS XVI Association for Chemoreception Sciences, Sarasota, FL.
DUFFY, V. B., DAVIDSON, A., KIDD, J., KIDD, K., SPEED, W., PATKSTIS, A., REED, D. R., SNYDER, D. J.
and
(2004c) Alcohol Clin Exp Res, 28, 1629±1637. and SHALLENBERGER, R. S. (2000) Food Chemistry, 68, 45±49. EL-HADDAD, M. A., ISMAIL, Y., GUERRA, C., DAY, L. and ROSS, M. G. (2002) Am J Obstet Gynecol, 187, 898±901. BARTOSHUK, L. M.
EGGERS, S. C., ACREE, T. E.
Genetic differences in sweet taste perception
51
(2000) Physiol Behav, 69, 3±13. and KENNEDY, L. M. (1998) Ann N Y Acad Sci, 855, 170±174. FAURION, A. (1993) In Sweet-taste chemoreception (Eds, Mathlouthi, M., Kanters, J. A. and Birch, G. G.) Elsevier Applied Science, London, New York. FISCHER, R. and GRIFFIN, F. (1961) Experientia, XVII, 36±38. FISCHER, R., GRIFFIN , F., ENGLAND, S. and GARN, S. (1961) Science, 191, 1328. FISCHER, R., GRIFFIN, F. and KAPLAN, A. R. (1963) Med Exp Int J Exp Med, 210, 151±167. FISCHER, R., GRIFFIN, F. and ROCKEY, M. A. (1966) Perspect Biol Med, 9, 549±577. FISCHLER, C. (1987) In Sweetness (Ed, Dobbing, J.) Springer-Verlag, London, New York. ERICKSON, R. P. EYLAM, S.
FONTVIEILLE, A. M., FAURION, A., HELAL, I., RIZKALLA, S. W., FALGON, S., LETANOUX, M.,
and SLAMA, G. (1989) Diabetes Care, 12, 481±486. (1931) Science News Letter, 9, 249. FREELAND-GRAVES, J. and NITZKE, S. (2002) J Am Diet Assoc, 102, 100±108. FROLOFF, N., LLORET, E., MARTINEZ, J. M. and FAURION, A. (1998) Chem Senses, 23, 197±206. GARBER, J., LAWRENCE L., HYATT, E. M. and STARR, J., RICHARD, G. (2003) Food Qual Pref, 14, 3±15. GENT, J. F. and BARTOSHUK, L. M. (1983) Chem Senses, 7, 265±272. GLENDINNING, J. I. and HILLS, T. T. (1997) J Neurophysiol, 78, 734±745. GREEN, B. G. and GEORGE, P. (2004) Chem Senses, 29, 617±628. GREEN, B. G. and HAYES, J. E. (2004) Chem Senses, 29, 53±60. GREENE, L., DESOR, J. and MALLER, O. (1975) J Comp Physiol Psychol, 89, 279±284. GRINKER, J. (1978) Am J Clin Nutr, 31, 1078±1087. HALL, M. J., BARTOSHUK, L. M., CAIN, W. S. and STEVENS, J. C. (1975) Nature, 253, 442±443. HALPERN, B. P. (2002) In Steven's handbook of experimental psychology: Vol. 1. Sensation and perception (Eds, Stevens, S. S. and Pashler, H. E.) John Wiley & Sons, New York, pp. 653±690. HALPERN, B. P. and NELSON, L. M. (1965) Am J Physiology, 209, 105±110. HARRIS, H. and KALMUS, H. (1949) Ann Eugen, 15, 24±31. HAYES, J. E., KIDD, J., KIDD, K., BARTOSHUK, L. M. and DUFFY, V. B. (2005) Chem Senses, 30, A34. HERMAN, C. and POLIVY, J. (1980) In Obesity (Ed. Stunkard, A.) W. B. Saunders, Philadelphia, PA, pp. 208±225. HORNE, J., LAWLESS, H. T., SPEIRS, W. and SPOSATO, D. (2002) Chem Senses, 27, 31±38. INSTITUTE OF MEDICINE OF THE NATIONAL ACADEMY OF SCIENCES (2002) National Academy of Sciences, Washington DC. KAMINSKI, L. C., HENDERSON, S. A. and DREWNOWSKI, A. (2000) Physiol Behav, 68, 691±697. KANAREK, R. B., RYU, M. and PRZYPEK, J. (1995) Physiology and Behavior, 57, 821±826. KANG, Y. S., CHO, W. K. and URN, K. S. (1967) Eugenics Quarterly, 14, 1±6. KANT, A. K. (2000) Am J Clin Nutr, 72, 929±936. KEAST, R. S. J. and BRESLIN, P. A. S. (2002) Chem Senses, 27, 123±131. KEAST, R. S., CANTY, T. M. and BRESLIN, P. A. (2004) Chem Senses, 29, 513±521. KENNEDY, L. M., EYLAM, S., POSKANZER, J. E. and SAIKKU, A. R. (1997) Food Chemistry, 60, 311±321. KIDD, K. K., DAVIDSON, A. C., PAKSTIS, A. J., SPEED, W. C., BARTOSHUK, L. M., DUFFY, V. B. and KIDD, J. R. (2004) PTC-tasting and negative evidence for balancing selection. American Society of Human Genetics meeting. (October) Toronto. Available at www.ashg.org/genetics/ashg04s/index.html KIM, U. K., JORGENSON, E., COON, H., LEPPERT, M., RISCH, N. and DRAYNA, D. (2003) Science, 299, 1221±1225. TCHOBROUTSKY, G.
FOX, A. L.
52
Optimising sweet taste in foods
KUHN, C., BUFE, B., WINNIG, M., HOFMANN, T., FRANK, O., BEHRENS, M., LEWTSCHENKO, T.,
(2004) J Neurosci, 24, 10260±10265. (1994) Laryngoscope, 104, 25±29. LAHTEENMAKI, L. and TUORILA, H. (1993) Physiol Behav, 57, 81±88. LANIER, S. A., HAYES, J. E. and DUFFY, V. B. (2005) Physiol Behav, 83, 821±831. LAWLESS, H. T. (1979a) Chem Senses Flavour, 4, 249±258. LAWLESS, H. T. (1979b) J Comp Physiol Psychol, 93, 538±547. LAWLESS, H. T. (1980) Chem Senses, 5, 247±256. LAWLESS, H. T. (1983) J Test Eval, 11, 346±349. LAWLESS, H. T. (1998) Chem Senses, 23, 447±451. LAWLESS, H. T. and STEVENS, D. A. (1983) Chem Senses, 7, 309±315. LAWLESS, H. T., HORNE, J. and SPEIRS, W. (2000) Chem Senses, 25, 85±92. LIAO, J. and SCHULTZ, P. G. (2003) Mamm Genome, 14, 291±301. LIN, B. (2003) Pediatric Dentistry, 25, 37±42. LOOY, H. and WEINGARTEN, H. P. (1992) Physiol Behav, 52, 75±82. LUCAS, F. and BELLISLE, F. (1987) Physiol Behav, 39, 739±743. SLACK, J. P., WARD, C. D.
KVETON, J. F.
and
and
MEYERHOF, W.
BARTOSHUK, L. M.
LUNDGREN, B., JONSSON, B., PANGBORN, R. M., SONTAG, A. M., BARYLKO-PIKIELNA, N.,
and YOSHIDA, M. (1978) Chem Senses Flavour, 3, 249±265. LY, A. and DREWNOWSKI, A. (2001) Chem Senses, 26, 41±47. MACDIARMID, J. I., VAIL, A., CADE, J. E. and BLUNDELL, J. E. (1998) Int J Obes Relat Metab Disord, 22, 1053±1061. MARGOLSKEE, R. F. (2002) J Biol Chem, 277, 1±4. MARKS, L. E., BORG, G. and WESTERLUND, J. (1992) Chem Senses, 17, 493±506. MATTES, R. and LABOV, J. (1989) J Am Diet Assoc, 89, 692±694. MATTES, R. D. (1994) Physiol Behav, 56, 1229±1236. PIETRZAK, E., GARRUTI, R. S., CHAIB, M. A.
MAX, M., SHANKER, Y. G., HUANG, L., RONG, M., LIU, Z., CAMPAGNE, F., WEINSTEIN, H., DAMAK, S.
and
(2001) Nat Genet, 28, 58±63. (1972) Percept Psychophy, 11, 225±227. MCBURNEY, D. H. S., DAVID V. and SHICK, T. R. (1972) Percept Psychophys, 11, 228±232. MEISELMAN, H. L. (1987) In Sweetness (Ed, Dobbing, J.) Springer-Verlag, London, New York, pp. 261±276. MENNELLA, J. A. and BEAUCHAMP, G. K. (1998) Nutr Rev, 56, 205±211. MENNELLA, J. A., PEPINO, M. Y. and REED, D. R. (2005) Pediatrics, 115, e216±222. MONTMAYEUR, J. P., LIBERLES, S. D., MATSUNAMI, H. and BUCK, L. B. (2001) Nat Neurosci, 4, 492±498. MOSKOWITZ, H. R. (1971) Am J Psychol, 84, 387±405. MOSKOWITZ, H. R. (2002) In Chemistry of taste: mechanisms, behavior, and mimics (Eds, Given, P. and Paredes, D.) American Chemical Society, Washington, DC. MOSKOWITZ, H. and MOSKOWITZ, R. B. (2000) J Sens Stud, 15, 263±284. MURPHY, S. P. and JOHNSON, R. K. (2003) Am J Clin Nutr, 78, 827S±833S. NELSON, G., HOON, M. A., CHANDRASHEKAR, J., ZHANG, Y., RYBA, N. J. and ZUKER, C. S. (2001) Cell, 106, 381±390. OLSON, J. M., BOEHNKE, M., NEISWANGER, K., ROCHE, A. F. and SIERVOGEL, R. M. (1989) Genet Epidemiol, 6, 423±434. PANGBORN, R. M. (1970) Psychonom Sci, 21, 125±126. PANGBORN, R. M. (1980) In Carbohydrate sweeteners in foods and nutrition (Eds, Koivistoinen, P., HyveÁonen, L. and Scandinavian Agricultural Research Workers' Association.) Academic Press, London, New York. MARGOLSKEE, R. F.
MCBURNEY, D. H.
Genetic differences in sweet taste perception
53
and HLADIK, C. M. (2002) Appetite, 39, 167±173. and MCFARLANE, T. (1994) J Abnorm Psychol, 103, 505±510. PRESCOTT, J., RIPANDELLI, N. and WAKELING, I. (2001) Chem Senses, 26, 993±1003. PRESCOTT, J., SOO, J., CAMPBELL, H. and ROBERTS, C. (2004) Physiol Behav, 82, 459±469. PASQUET, P., OBERTI, B., EL ATI, J. POLIVY, J., HERMAN, C. P.
PRODI, D. A., DRAYNA, D., FORABOSCO, P., PALMAS, M. A., MAESTRALE, G. B., PIRAS, D., PIRASTU, M.
and
ANGIUS, A.
(2004) Chem Senses, 29, 697±702.
PRUTKIN, J., FISHER, E. M., ETTER, L., FAST, K., GARDNER, E., LUCCHINA, L. A., SNYDER, D. J., TIE, K., WEIFFENBACH, J.
and
BARTOSHUK, L. M.
(2000) Physiol Behav, 69, 161±173. and PRICE, R. A. (1995) Chem Senses,
REED, D. R., BARTOSHUK, L. M., DUFFY, V., MARINO, S.
20, 529±533.
REED, D. R., LI, X., CHEN, Z., BACHMANOV, A. A., BEAUCHAMP, G. K., TORDOFF, M. G., MAX, M.,
and DUFFY, V. B. (2002) Obes Res, 10, 67 (abstract). RICHTER, C. P. and CAMPBELL, K. H. (1939) Am J Physiol ± Legacy Content, 128, 291±297. RICHTER, C. P. and CLISBY, K. H. (1941) Am J Physiol, 134, 157±164. RODIN, J. (1975) J Comp Physiol Psychol, 89, 1003±1009. ROZIN, P. (1987) In Sweetness (Eds, Dobbing, J. and International Life Sciences Institute) Springer-Verlag, London, New York. SAINZ, E., KORLEY, J. N., BATTEY, J. F. and SULLIVAN, S. L. (2001) J Neurochem, 77, 896±903. SATO, T., OKADA, Y., MIYAMOTO, T. and FUJIYAMA, R. (1997) Chem Senses, 22, 547±551. SCHIFFMAN, S. S., BOOTH, B. J., LOSEE, M. L., PECORE, S. D. and WARWICK, Z. S. (1995) Brain Res Bull, 36, 505±513. SCHIFFMAN, S. S., BOOTH, B. J., SATTELY-MILLER, E. A., GRAHAM, B. G. and GIBES, K. M. (1999) Chem Senses, 24, 439±447. SCHIFFMAN, S. S., GRAHAM, B. G., SATTELY-MILLER, E. A. and PETERSON-DANCY, M. (2000) Nutrition, 16, 886±893. SHALLENBERGER, R. S. and ACREE, T. E. (1967) Nature, 216, 480±482. SMAGGHE, K. and LOUIS-SYLVESTRE, J. (1998) Appetite, 30, 325±339. SNYDER, L. H. (1931) Science, 74, 151±152. STEINER, J. E., GLASER, D., HAWILO, M. E. and BERRIDGE, K. C. (2001) Neurosci Biobehav Rev, 25, 53±74. STUNKARD, A. and MESSICK, S. (1985) Journal of Psychosomatic Research, 29, 71±83. TEPPER, B. J. and NURSE, R. J. (1997) Physiol Behav, 61, 949±954. THOMPSON, D. A., MOSKOWITZ, H. R. and CAMPBELL, R. G. (1977) Physiol Behav, 19, 335± 337. WANSINK, B. (2003) Food Qual Pref, 14, 23±26. WHO/FAO (2003) Food and Agriculture Organization of the United Nations, Geneva, Switzerland, pp. 54±71. WITHERLY, S. A., PANGBORN, R. M. and STERN, J. S. (1980) Appetite, 1, 53±63. WOODING, S., KIM, U. K., BAMSHAD, M. J., LARSEN, J., JORDE, L. B. and DRAYNA, D. (2004) Am J Hum Genet, 74, 637±646. XU, H., STASZEWSKI, L., TANG, H., ADLER, E., ZOLLER, M. and LI, X. (2004) Proc Natl Acad Sci USA, 101, 14258±14263. YANAGISAWA, K., BARTOSHUK, L. M., CATALANOTTO, F. A., KARRER, T. A. and KVETON, J. F. (1998) Physiol Behav, 63, 329±335. YIEE, J., DUFFY, V. and BARTOSHUK, L. (2002) Appetite, 39, 108 (abstract). YIEE, J. H., DUFFY, V. B. and BARTOSHUK, L. M. (2003) Chem Senses, 28, 134 (abstract). MARGOLSKEE, R., BARTOSHUK, L. M.
3 Children's liking of sweet tastes and its biological basis M. Y. Pepino and J. A. Mennella, Monell Chemical Senses Center, USA
3.1
Introduction
The liking for things that taste sweet is universal and a hallmark of development. At birth, sweet liking ensures the acceptance of that which infants need to survive ± the sweet taste of their mothers' milk. Heightened preferences for sweet tastes, which persist throughout childhood and adolescence, may have an ecological basis since, in nature, sweet tasting foods, such as fruits, are associated with energy-producing sugars, minerals, and vitamins (Beauchamp, 1999). Although strong evidence is lacking, it has been suggested that such preferences evolved to solve a basic nutritional problem of attracting children to sources of high energy during periods of maximal growth (Simmen and Hladik, 1998; Drewnowski, 2000; Coldwell and Oswald, 2004). In more modern times, children express their liking for sweets by consuming foods other than nature's fruits. Since the 19th century, sweet candies have been the first material goods that children spend their own money on (Woloson, 2002). In this chapter, we review the scientific literature that suggests that children's liking for all that is sweet is not solely a product of modern-day technology and advertising, but reflects their basic biology. To this end, we summarize the insights gleaned from scientific research on the ontogeny of human behavioral responses to sweet tastes. We review the scientific literature that documents the innate heightened preferences for sweet tastes and how this innate preference can be modified by experiences early in life. This review serves as a foundation for a discussion on the physiological and motivational properties of sugars.
Children's liking of sweet tastes and its biological basis
55
3.2 Ontogeny of sweet taste preferences: from fetal life to adolescence 3.2.1 Fetus Although the response of human fetuses to sweet taste has never been directly investigated, indirect evidence suggests that changes in the chemosensory properties of amniotic fluid may modulate intrauterine behaviors. Athough there were a number of methodological flaws (see Mistretta and Bradley, 1977), two studies reported increased swallowing following the injection of sweet stimuli such as saccharin (Liley, 1972; De Snoo, 1937) in the amniotic fluid of pregnant women and decreased swallowing following the injection of a bitter-tasting substance (Liley, 1972) thus suggesting that prenatal monitoring of taste stimuli dissolved in amniotic fluid is possible during late gestation. Perhaps stronger evidence that the fetus can respond to taste stimuli comes from research on premature infants who were six to nine months' gestational age. To overcome methodological limitations and because premature infants often have immature suck-swallow coordination, innovative methods were developed that avoided the risk of fluid aspiration by embedding the taste substance in a nippleshaped gelatin medium that released small amounts of the sweet substance when mouthed or sucked. Infants born pre-term and tested between 33 and 40 weeks postconception produced stronger and more frequent sucking responses when offered the sucrose-sweetened nipple compared with a latex nipple (Maone et al., 1990). Pre-term infants, who had been fed exclusively via gastric tubes, exhibited more non-nutritive sucking in response to glucose than to water (Tatzer et al., 1985). Taken together, these data suggest that the ability to detect sweet tastes and, in turn, modulate a variety of behaviors is evidenced during fetal life. The preference for sweet tastes is not only present early in ontogeny, but research on other infant primates revealed that it is remarkably well conserved phylogenetically (Steiner et al., 2001). 3.2.2 Infants Within hours after birth, the liking for sweet tastes is unequivocal. In fact, newborns are sweet connoisseurs. They respond to even dilute sweet tastes, differentiate varying degrees of sweetness and will consume more of a sugar solution when compared to water (Nowlis and Kessen, 1976; Desor et al., 1973). When a sweet solution is placed in the oral cavity, the baby's face relaxes, resembling an expression of satisfaction which is often followed by a smile (Steiner, 1977; Rosenstein and Oster, 1988). That the positive facial expressions elicited by sweet tasting sugars are reflex-like is supported by several findings. First, single response components, such as tongue movements, can be reliably elicited in newborns by sweet tastes in a concentration-dependent manner (e.g., Nowlis and Kessen, 1976; Weiffenbach, 1977). Second, neonates born with severe developmental malformations of the central nervous system (e.g., anencephalic infants) react to the sweet taste (and other tastes) like a normal term born neonate (Steiner, 1977).
56
Optimising sweet taste in foods
Consistent with that observed for intake and facial expression, infants can express that they perceive and respond positively to sweet stimuli via a variety of other behaviors. Sweet taste solutions of sucrose and aspartame in the oral cavity reliably elicits mouthing and sucking movements and hand-to-mouth contacts (e.g., Miller et al., 1994; Barr et al., 1999; Blass et al., 1989), both of which are feeding-related behaviors. Newborns also respond to sweet stimulation with changes in autonomic response (e.g., Blass and Watt, 1999; Crook and Lipsitt, 1976; Haouari et al., 1995; Lipsitt, 1977; Lipsitt et al., 1976; Rao et al., 1997). However, the behavioral state of the baby modified the direction of the change. For example, heart rates increased proportionally when increasing concentration of sucrose were placed in the mouths of non-agitated infants (Ashmead et al., 1980). Conversely, heart rate decreased when sweet tastes were introduced to infants who were agitated, resulting in overall calmness (e.g., Blass and Watt, 1999). Furthermore, electroencephalographic recordings of 2- to 3-day-old newborns revealed that infants respond to sucrose administration with an asymmetry of brain electrical activity, a response which is usually associated with hedonically positive emotional reactions or approach behavior (Fox and Davidson, 1986). Although each measure has its limitations, the convergence of findings obtained from different methodologies suggested that from birth infants are quite sensitive to and prefer sweet tastes. 3.2.3 Children and adolescents The heightened preference for sweet taste is universal and evident among children around the world (e.g., Brazil: Tomita et al., 1999; France: Bellisle et al., 1990; Iraq: Jamel et al., 1997; Israel: Steiner et al., 1984; Mexico: Vazquez et al., 1982; Netherlands: De Graaf and Zandstra, 1999; and North America: Beauchamp and Cowart, 1987; Desor et al., 1975). Both cross-sectional and longitudinal studies revealed that the preference for sweets remains heightened throughout childhood (Pepino and Mennella, 2005; Mennella et al., 2005a) and early adolescence (Desor et al., 1975) and declines to adult levels during late adolescence (Desor and Beauchamp, 1987). In a cross-sectional study that measured sweet preference in more than 750 participants, 50% of the children and adolescents, but only 25% of the adults, selected the 0.60 M sucrose concentration as their favorite solution (Desor et al., 1975). To put this in perspective, a 0.60 M sucrose concentration is equivalent to approximately 12 spoonfuls of sugar in 230 ml of water (an 8 oz glass), whereas a typical cola has a 0.34 M sugar concentration. The level of sucrose that children prefer, as measured in the laboratory, is significantly related to children's preferences for sweet tasting foods such as cereals (Liem and Mennella, 2002; Mennella et al., 2005a) and beverages (Olson and Gemmill, 1981; Mennella et al., 2005a). Sweet taste can clearly drive consumption in the young. Adding sugar to beverages such as Kool Aid (Beauchamp and Moran, 1984) and solid foods such as spaghetti (Filer, 1978),
Children's liking of sweet tastes and its biological basis
57
ricotta and jicama (Sullivan and Birch, 1990) increased both liking and acceptance by children. Findings from the USDA Department of Agriculture 1994±1996 Continuing Survey of Food Intakes by Individuals of more than 15,000 Americans who were two years of age or older (Guthrie and Morton, 2000) paralleled the findings from basic research of an age-related decline in sweet preferences. That is, the proportion of energy obtained from added sweeteners peaks in adolescents, with approximately 20% of energy derived from added sugars, and declines to approximately 12.4% of energy for those 65 years and older (Guthrie and Morton, 2000). What causes the developmental shift in sweet preferences and consumption remains a mystery, but this age-related decline in sweet preference has been observed in other mammals (Wurtman and Wurtman, 1979; Marlin, 1983). One explanation may be that children are less sensitive to sweet tastes, thereby requiring larger amounts of sugars. However, there is no strong evidence that the heightened sweet preference evidenced during infancy and childhood is due to decreased sensitivity to sweet tastes (see Cowart, 1981 for review). Whereas some investigators may report that children are more sensitive than adults when small regions of the anterior tongue are stimulated with sucrose (Stein et al., 1994), others found that taste sensitivity during childhood does not differ from that during adulthood (Cowart, 1981). Another explanation is that the heightened sweet preferences early in life are linked to the growing child's need for calories (Drewnowski, 2000). Recent findings by Coldwell and Oswald (2004) lend support to this hypothesis. Elevento 15-year-old children were divided into likers and dislikers based on their sucrose preferences. Although there were no differences between these two groups in sucrose detection thresholds, age, body mass index, percent of body fat, plasma leptin, dietary restraint, pubertal development or gonadal hormone levels, growth rate was significantly lower in dislikers when compared to likers. These data suggest that the age-related decline in sucrose preferences may be related to the cessation of growth (Coldwell and Oswald, 2004). 3.2.4 Individual differences Despite such age-related changes in sweet preferences, individual variations exist at both ends of the age spectrum (Desor et al., 1977; Beauchamp and Moran, 1982; Enns et al., 1979; Drewnowski et al., 1985; Pliner and Fleming, 1983; Mennella et al., 2005a). Genetic variation may contribute to some differences in sweet taste preferences (see McDaniel and Reed, 2004). We recently demonstrated that alleles of the bitter taste receptor TAS2R38 gene could explain much of the phenotypic variation in the sensitivity to the bitter compound, propylthiouracil and partially explained individual differences in children's sweet preferences (Mennella et al., 2005a; see also Keller and Tepper, 2004). Here individuals homozygous for the bitter-insensitive allele are referred to as AA, heterozygotes as AP, and those homozygous for the bitter-sensitive allele as PP. The data
58
Optimising sweet taste in foods
revealed that AP and PP children preferred significantly higher concentrations of sugars in liquids when compared with children homozygous for the bitterinsensitive allele (AA). Similar findings were observed for foods such as cereals. Future research is needed to determine how variation in other taste receptor genes, particularly those involved in sweet taste perception, contribute to individual differences in preferences and food habits. Unlike children, there was no correspondence between TAS2R38 genotypes and sweet preference in adults. Here the effects of race/ethnicity were the strongest determinants, thus suggesting that cultural forces and experience may override this genotype effect on sweet preferences. To the point, several studies demonstrated that African-American, infants, children, teenagers as well as adults preferred higher concentrations of sucrose when compare to similarly aged Caucasians (Beauchamp and Moran, 1982; Desor et al., 1975; Bacon et al., 1994; Pepino and Mennella, 2005). Differential exposure to sweets foods, as will be discussed in the following section, may contribute to these race/ethnicity differences since the tradition of feeding sweetened water or adding sugar to the children's food is practiced significantly more among African American (Pepino and Mennella, 2005; Liem and Mennella, 2002) and Hispanic (Mennella et al., 2005b) mothers than Caucasians. Nevertheless, although there are individual differences in sweet preferences that are determined in part by our genetics and are shaped by the culture in which we live, there is no doubt that children, as a group, prefer significantly higher concentrations of sugars in foods and liquids than do adults.
3.3
Role of experiential factors on sweet taste preferences
Considering the strong innate preference that humans have for sweet tastes, it is not surprising that a review of ancient and current-day customs of the first liquids or foods given to infants prior to the first breast feed, hence the term prelacteal feed, (Morse et al., 1990) revealed a striking commonality in their sweet taste (Jerome, 1977; Mennella, 1997). Sugared waters or teas are perhaps the most popular pre-lacteals given to infants throughout the world (Morse et al., 1990; Mennella, 1997; Mennella et al., 2005b). To what extent do these early experiences alter or modulate the development of sweet preferences later in life? Longitudinal studies (Beauchamp and Moran, 1982, 1984) revealed that babies who were routinely fed sweetened water (e.g., water sweetened with table sugar, Karo syrup or honey) during the first months of life exhibited a greater preference for sweetened water when tested at six months (Beauchamp and Moran, 1982) and then again at two years of age (Beauchamp and Moran, 1984) when compared to those who had little or no experience with sweetened water. A more recent cross-sectional study on 6- to 10-year-old children supported these findings and revealed that such feeding practices may have longer term effects on the preference for sweetened water than previously realized (Pepino and Mennella, 2005).
Children's liking of sweet tastes and its biological basis
59
Whether early experience with sweetened water modifies overall liking for sweetened foods is less uncertain. In fact, there is no compelling data suggesting that repeated exposure to sugar water results in a generalized heightened hedonic response to sweetness. Being fed sugar water during infancy did not heighten 2year-old children's preference for sweetened versus the unsweetened Kool-Aid. All children, regardless of whether they were fed sugar water, ingested significantly more of the sweetened when compared to the plain Kool-Aid (Beauchamp and Moran, 1984). Moreover, 4- to 7-year-old children whose mothers reported adding sugar to their foods on a routine basis were significantly more likely to prefer apple juices with added sugar and cereals with higher sugar contents than similar age children whose mothers reported never adding sugar to the foods at home (Liem and Mennella, 2002). These data support the hypothesis that the context in which the taste experience occurred is an important factor and through familiarization, children develop a sense of what should, or should not, taste sweet (Beauchamp and Cowart, 1985).
3.4
Physiological properties of sweet tastes
Tasting a sweet substance not only elicits feeding-related behaviors but also modifies crying behaviors and physiological responses. A small amount of a sweet solution placed on the tongue of a crying newborn exerts a rapid, calming effect which persists for several minutes (Blass and Hoffmeyer, 1991; Barr et al., 1999). Sweet tastes can also attenuate heart rate increases, blunt expressions of pain, and calm both preterm and full-term infants who have been subjected to painful events such as heel stick or circumcision (Stevens et al., 2004 for review). That sucrose attenuated a negative electroencephalographic response to a painful procedure (Fernandez et al., 2003) suggests that sucrose blocks pain afferents which, in turn, diminishes stress and cardiac changes. Because noncaloric sweet substances such as aspartame mimic the calming effects of sucrose (Barr et al., 1999; Bucher et al., 2000) and because the administration of sucrose by direct stomach loading is not effective (Ramenghi et al., 1999), afferent signals from the mouth, rather than gastric or metabolic changes, appear to be responsible for the analgesic properties of sweet tastes. Other flavors and orosensory components also reduce pain in infants and may act synergistically with sweet tastes. These include the flavor of mothers' milk (Upadhyay et al., 2004) and some of its constituents (i.e., fat, protein) (Barr et al., 1996; Blass, 1997). Orotactile stimulation from a pacifier (Blass and Hoffmeyer, 1991; Blass and Watt, 1999; Carbajal et al., 1999), maternal skin-toskin contact (Gray et al., 2000) or being held before and during the painful procedure (Gormally et al., 2001) also produce the analgesic effect, but the effects are transient when compared to when sweet taste is in the oral cavity (Smith et al., 1990). Using sterile sugar water to reduce pain in infants is now routine practice in several hospital nurseries because of its natural simplicity, viability and efficacy (Anand, 2001). However, more research is needed to
60
Optimising sweet taste in foods
determine sweet taste's efficacy in reducing pain after repeated administrations (Stevens et al., 2004; Mennella et al., 2003; Vazquez et al., 1982). Sweet tastes continue to have analgesic properties during childhood. The presence of sucrose, but not water, in the oral cavity delayed 8- to 11-year-old children's reporting of pain onset when undergoing a cold-induced pain stimulus test, the Cold Pressor Test (Miller et al., 1994). Research from our laboratory recently confirmed these findings and revealed that sucrose's efficacy in reducing pain is related to the hedonic value of sweet taste for the child (Pepino et al., 2004). The more children like sucrose, the better it works in increasing pain tolerance.
3.5
Conclusions
Liking sweet tasting foods and liquids reflects the child's basic biology. When given the choice, it is not surprising that children will eat foods and beverages that are high in sugar content. The heightened preference for sweet tastes during development may have ensured the acceptance of nature's first food ± mothers' milk, as well as nature's sweet tasting foods such as fruits which are high in energy, minerals and vitamins. However, we are now living in an age of abundance with a diet that contains highly concentrated sugars in foods and beverages, in part, because of modern technological means of refining sugars. In other words, there is great mismatch between the environment in which humans evolved and the one we currently reside in. Organizations worldwide (e.g., Joint World Health Organization/Food and Agriculture Organization Expert Consultation on Diet, Nutrition and Prevention of Chronic Disease and United States Department of Agriculture) recommend limiting the intake of free sugars to less than 10% of total energy (WHO/Food and Agriculture Organization of the United Nations Expert Consultation, 2003; Welsh et al., 1993). However, at least in the United States, estimates suggest that the levels of consumption far exceed recommended levels with adolescents consuming around 20% of total energies from added sweeteners (Guthrie and Morton, 2000). Whether the food habits of children are determined by genetics or shaped by the culture in which we live or both, there is no doubt that children, as a group, prefer significantly higher concentrations of sugars in foods and liquids than do adults. However, there are striking individual and group differences in the levels of sweetness preferred. For example, across all age groups, African-American non-Hispanics prefer significantly higher levels of sweetness than Caucasians (Beauchamp and Moran, 1982; Desor et al., 1975; Bacon et al., 1994; Pepino and Mennella, 2005; Krebs-Smith, 2001). Therefore, it is important to realize that trying to limit sweet food /beverages consumption may be more difficult for some children or certain ethnic groups because of the inherent hedonic value of sweet tastes and how sweets make them feel. More knowledge about the factors that contribute to preferences for sweet-tasting foods and beverages in children,
Children's liking of sweet tastes and its biological basis
61
a generation who will struggle with obesity and diabetes, may suggest population-based strategies to overcome diet-induced disease and promote healthy eating habits.
3.6
Acknowledgements
Preparation of this manuscript was supported in part by NIH Grants AA09523 and HD37119.
3.7
References
(2001), `Consensus statement for the prevention and management of pain in the newborn', Arch Pediatr Adolesc Med, 155, 173±80. ASHMEAD D H, REILLY B M and LIPSITT L P (1980), `Neonates' heart rate, sucking rhythm, and sucking amplitude as a function of the sweet taste' J Exp Child Psychol, 29 (2), 264±81. BACON A W, MILES J S and SCHIFFMAN S S (1994), `Effect of race on perception of fat alone and in combination with sugar' Physiol Behav, 55 (3), 603±6. BARR R G, YOUNG S N, ALKAWAF R and WETHEIM L (1996), `Does mature hind milk taste calm crying infants?' Ped Res, 39 (4 Part 2), 16A. BARR R G, PANTEL M S, YOUNG S N, WRIGHT J H, HENDRICKS L A and GRAVEL R (1999), `The response of crying newborns to sucrose: is it a ``sweetness'' effect?' Physiol Behav, 66 (3), 409±17. BEAUCHAMP G K (1999), `Factors affecting sweetness', in Corti A, Low-calorie sweeteners: Present and future, Basel, Karger, World Rev Nutr Diet, 10±17. BEAUCHAMP G K and COWART B J (1985), `Congenital and experiential factors in the development of human flavor preferences', Appetite, 6 (4), 357±72. BEAUCHAMP G K and COWART B J (1987), `Development of sweet taste', in Dobbing J, Sweetness, Berlin, Springer-Verlag, 127±38. BEAUCHAMP G K and MORAN M (1982), `Dietary experience and sweet taste preference in human infants', Appetite, 3 (2), 139±52. BEAUCHAMP G K and MORAN M (1984), `Acceptance of sweet and salty tastes in 2-year-old children', Appetite, 5 (4), 291±305. BELLISLE F, DARTOIS A M, KLEINKNECHT C and BROYER M (1990), `Perceptions of and preferences for sweet taste in uremic children', J Am Diet Assoc, 90 (7), 951±4. BLASS E M (1997), `Milk-induced hypoalgesia in human newborns', Pediatrics, 99 (6), 825±9. BLASS E M and HOFFMEYER L B (1991), `Sucrose as an analgesic for newborn infants', Pediatrics, 87 (2), 215±18. BLASS E M and WATT L B (1999), `Suckling- and sucrose-induced analgesia in human newborns', Pain, 83 (3), 611±23. BLASS E M, FILLION T J, ROCHAT P, HOFFMEYER L B, ROCHAT P and METZGER M A (1989), `Sensorimotor and motivational determinants of hand-mouth coordination in 1±3day-old human infants', Dev Psychol, 25 (6), 963±75. BUCHER H U, BAUMGARTNER R, BUCHER N, SEILER M and FAUCHERE J C (2000), `Artificial ANAND K J S
62
Optimising sweet taste in foods
sweetener reduces nociceptive reaction in term newborn infants', Early Hum Dev, 59 (1), 51±60. CARBAJAL R, CHAUVET X, COUDERC S and OLIVIER-MARTIN M (1999), `Randomised trial of analgesic effects of sucrose, glucose, and pacifiers in term neonates', BMJ, 319 (7222), 1393±7. COLDWELL S E and OSWALD T K (2004), `Growth-rate differs between children with adultlike versus child-like sugar preference', J Dent Res, 83 (Spec Iss A), Abst 0740. COWART B J (1981), `Development of taste perception in humans: sensitivity and preference throughout the life span', Psychol Bull, 90 (1), 43±73. CROOK C K and LIPSITT L P (1976), `Neonatal nutritive sucking: effects of taste stimulation upon sucking rhythm and heart rate', Child Dev, 47 (2), 518±22. DE GRAAF C and ZANDSTRA E H (1999), `Sweetness intensity and pleasantness in children, adolescents, and adults', Physiol Behav, 67 (4), 513±20. DE SNOO K (1937), `Das trinkende Kind im Uterus' [The drinking child in-utero], Monatsschr Geburtsh Gynaekol [International Monthly Review of Obstetrics and Gynaecology], 105, 88±97. DESOR J A and BEAUCHAMP G K (1987), `Longitudinal changes in sweet preferences in humans', Physiol Behav, 39 (5), 639±41. DESOR J A, MALLER O and TURNER R E (1973), `Taste in acceptance of sugars by human infants', J Comp Phys Psychol, 84, 496±501. DESOR J A, GREENE L S and MALLER O (1975), `Preferences for sweet and salty in 9- to 15year-old and adult humans', Science, 190 (4215), 686±7. DESOR J A, MALLER O and GREENE L S (1977), `Preference for sweet in humans: Infants, children, and adults', in Weiffenbach J M, Taste and Development: the genesis of sweet preference, Wahington, D.C, U.S. Government Printing Office, 161±72. DREWNOWSKI A (2000), `Sensory control of energy density at different life stages', Proc Nutr Soc, 59 (2), 239±44. DREWNOWSKI A, BRUNZELL J D, SANDE K, IVERIUS P H and GREENWOOD M R (1985), `Sweet tooth reconsidered: taste responsiveness in human obesity', Physiol Behav, 35 (4), 617±22. ENNS M P, VAN ITALLIE T B and GRINKER J A (1979), `Contributions of age, sex and degree of fatness on preferences and magnitude estimations for sucrose in humans', Physiol Behav, 22 (5), 999±1003. FERNANDEZ M, BLASS E M, HERNANDEZ-REIF M, FIELD T, DIEGO M and SANDERS C (2003), `Sucrose attenuates a negative electroencephalographic response to an aversive stimulus for newborns', J Dev Behav Pediatr, 24 (4), 261±6. FILER L J (1978), `Studies of taste perception in infancy and childhood', Ped Bas, 5±9. FOX N A and DAVIDSON R J (1986), `Taste-elicited changes in facial signs of emotion and the asymmetry of brain electrical activity in human newborns', Neuropsychologia, 24 (3), 417±22. GORMALLY S, BARR R G, WERTHEIM L, ALKAWAF R, CALINOIU N and YOUNG S N (2001), `Contact and nutrient care giving effects on newborn infant pain responses', Dev Med Child Neurol, 43 (1), 28±38. GRAY L, WATT L and BLASS E M (2000), `Skin-to-skin contact is analgesic in healthy newborns', Pediatrics, 105 (1), e14. GUTHRIE J F and MORTON J F (2000), `Food sources of added sweeteners in the diets of Americans', J Am Diet Assoc, 100 (1), 43±51. HAOUARI N, WOOD C, GRIFFITHS G and LEVENE M (1995). `The analgesic effect of sucrose in full term infants: a randomised controlled trial', BMJ, 310 (6993), 1498±500.
Children's liking of sweet tastes and its biological basis
63
and COWELL C R (1997), `Sweet preference, consumption of sweet tea and dental caries; studies in urban and rural Iraqi populations', Int Dent J, 47 (4), 213±17. JEROME N W (1977), `Taste experience and the development of a dietary preference for sweet in humans: ethnic and cultural variations in early taste experience', in Weiffenbach J M, Taste and Development: the genesis of sweet preference, Washington, DC, US Government Printing Office, pp. 235±48. KELLER K L and TEPPER B J (2004), `Inherited taste sensitivity to 6-n-propylthiouracil in diet and body weight in children', Obes Res, 12 (6), 904±12. KREBS-SMITH S (2001), `Choose beverages and foods to moderate your intake of sugars: Measurement requires quantification', Am Soc Nutr Sci, 131 (2S-1), 527S±535S. LIEM D G and MENNELLA J A (2002), `Sweet and sour preferences during childhood: role of early experiences', Dev Psychobiol, 41 (4), 388±95. LILEY A W (1972), `Disorders of amniotic fluid', in Assali, N S, Patho-physiology of gestation, Vol II. Fetal-placental disorders, New York, Academic, 157±206. LIPSITT L P (1977), `Taste in human neonates: Its effects on sucking and heart rate', in Weiffenbach J M, Taste and Development: The genesis of sweet preference, Washington, DC, US Government Printing Office, pp. 125±42. LIPSITT L P, REILLY B M, BUTCHER M J and GREENWOOD M M (1976), `The stability and interrelationships of newborn sucking and heart rate', Dev Psychobiol, 9 (4), 305± 10. MAONE T R, MATTES R D, BERNBAUM J C and BEAUCHAMP G K (1990), `A new method for delivering a taste without fluids to preterm and term infants', Dev Psychobiol, 23 (2), 179±91. MARLIN N A (1983), `Early exposure to sugars influences the sugar preference of the adult rat', Physiol Behav, 31 (5), 619±23. MCDANIEL A H and REED D R (2004), `The human sweet tooth and its relationship to obesity', in Moustaid-Moussa N and Berdanier C, Genomics and Proteomics in Nutrition, New York, Marcel Dekker, Inc, 49±67. MENNELLA J A (1997), `The flavor world of infants: A cross cultural perspective', Nutrition Today, 32, 142±51. MENNELLA J A, FORESTELL C A and PEPINO M Y (2003). `The flavor world of infants', Perspectives on Swallowing and Swallowing Disorders, 12, 10±19. MENNELLA J A, PEPINO M Y and REED D R (2005a), `Genetic and environmental determinants of bitter perception and sweet preferences', Pediatrics, 115 (2), e216±22. MENNELLA J A, TURNBULL B, ZIEGLER P J and MARTINEZ H (2005b), `Infant feeding practices and early flavor experiences in Mexican infants: An intra-cultural study', J Am Diet Assoc, 105 (6), 908±15. MILLER A, BARR R G and YOUNG S N (1994), `The cold pressor test in children: methodological aspects and the analgesic effect of intraoral sucrose', Pain, 56 (2), 175±83. MISTRETTA C M and BRADLEY R M (1977), `Taste in Utero: Theoretical Considerations', in Weiffenbach J M, Taste and Development: The genesis of sweet preference, Washington, DC, US Government Printing Office, pp. 51±69. MORSE J M, JEHLE C and GAMBLE D (1990), `Initiating breastfeeding: a world survey of the timing of postpartum breastfeeding', Int J Nurs Stud, 27 (3), 303±13. NOWLIS G H and KESSEN W (1976), `Human newborns differentiate differing concentrations of sucrose and glucose', Science, 191 (4229), 865±6. OLSON C M and GEMMILL K P (1981), `Association of sweet preference and food selection among four to five year old children', Ecol Food Nutr, 11, 145±50. JAMEL H A, SHEIHAM A, WATT R G
64
Optimising sweet taste in foods
and MENNELLA J A (2004), `Analgesic effects of intraoral sucrose: The more they like the sweet taste, the better it works?' Abstract presented at the XXVI Annual Meeting for the Association for Chemoreception Sciences, Sarasota, FL. PEPINO M Y and MENNELLA J A (2005), `Factors contributing to individual differences in sucrose preference', Chem Senses, 30 (Suppl 1), i319±20. PLINER P and FLEMING A S (1983), `Food intake, body weight, and sweetness preferences over the menstrual cycle in humans', Physiol Behav, 30 (4), 663±6. RAMENGHI L A, EVANS D J and LEVENE M I (1999), ```Sucrose analgesia'': absorptive mechanism or taste perception?', Arch Dis Child Fetal Neonatal Ed, 80 (2), F146±7. RAO M, BLASS E M, BRIGNOL M M, MARINO L and GLASS L (1997), `Reduced heat loss following sucrose ingestion in premature and normal human newborns', Early Hum Dev, 48 (1±2), 109±16. ROSENSTEIN D and OSTER H (1988), `Differential facial responses to four basic tastes in newborns', Child Dev, 59 (6), 1555±68. SIMMEN B and HLADIK C M (1998), `Sweet and bitter taste discrimination in primates: scaling effects across species', Folia Primatol (Basel), 69 (3), 129±38. SMITH B A, FILLION T J and BLASS E M (1990), `Orally mediated sources of calming in 1-to 3-day-old human infants', Dev Psychol, 26 (5), 731±7. STEIN N, LAING D G and HUTCHINSON I (1994), `Topographical differences in sweetness sensitivity in the peripheral gustatory system of adults and children', Brain Res Dev Brain Res, 82 (1±2), 286±92. STEINER J E (1977), `Facial expressions of the neonate infant indicating the hedonics of food-related chemical stimuli', in Weiffenbach J M, Taste and development: the genesis of sweet preference, Washington, DC, US Government Printing Office, pp. 173±88. STEINER J E, SGAN-COHEN H D and NAHAS J (1984), `Sweet preference and dental caries among Bedouin youth in Israel', Community Dent Oral Epidemiol, 12 (6), 386±9. STEINER J E, GLASER D, HAWILO M E and BERRIDGE K C (2001), `Comparative expression of hedonic impact: affective reactions to taste by human infants and other primates', Neurosci Biobehav Rev, 25 (1), 53±74. STEVENS B, YAMADA J and OHLSSON A (2004), `Sucrose for analgesia in newborn infants undergoing painful procedures', Cochrane Database Syst Rev, (3), CD001069. SULLIVAN S A and BIRCH L L (1990), `Pass the sugar, pass, the salt: experience dictates preference', Dev Psychol, 26 (4), 546±51. TATZER E, SCHUBERT M T, TIMISCHL W and SIMBRUNER G (1985), `Discrimination of taste and preference for sweet in premature babies', Early Hum Dev, 12 (1), 23±30. TOMITA N E, NADANOVSKY P, VIEIRA A L and LOPES E S (1999), `Taste preference for sweets and caries prevalence in preschool children', Rev Saude Publica, 33 (6), 542±6. UPADHYAY A, AGGARWAL R, NARAYAN S, JOSHI M, PAUL V K and DEORARI A K (2004), `Analgesic effect of expressed breast milk in procedural pain in term neonates: a randomized, placebo-controlled, double-blind trial', Acta Paediatr, 93 (4), 518±22. VAZQUEZ M, PEARSON P B and BEAUCHAMP G K (1982), `Flavor preferences in malnourished Mexican infants', Physiol Behav, 28 (3), 513±19. WEIFFENBACH J M (1977), `Sensory mechanisms of the newborn's tongue', in Weiffenbach J M, Taste and Development: the Genesis of Sweet Preference, Washington, DC, US Government Printing Office, pp. 205±34. WELSH S, DAVIS C and SHAW A (1993), USDA Food Guide: Background and Development, Hyattsville, MD: United States Department of Agriculture, Report No. 1514. PEPINO M Y, KENNEDY J
Children's liking of sweet tastes and its biological basis
65
(2002), Refined Tastes: Sugar, Confectionery, and Consumers in Nineteenth-Century America. Baltimore, Maryland, The Johns Hopkins University Press.
WOLOSON W A
WORLD HEALTH ORGANIZATION/FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED
NATIONS EXPERT CONSULTATION (2003), Diet, Nutrition and the Prevention of Chronic Diseases. Technical Support Series 916, Geneva, Switzerland, World Health Organization. WURTMAN J J and WURTMAN R J (1979), `Sucrose consumption early in life fails to modify the appetite of adult rats for sweet foods', Science, 205 (4403), 321±2.
4 Taste±odour interactions in sweet taste perception D. Valentin, C. Chrea and D. H. Nguyen, CSEG, France
4.1
Introduction
The sense of smell plays a major role in food flavour perception: individuals who lose their sense of smell often report that food has no taste anymore. Yet, from a neuroanatomical point of view taste and smell are very different senses. Taste is perceived primarily on the tongue whereas odours are perceived in the upper part of the nasal cavity either directly or via the back of the mouth. But does this neuroanatomical dissociation imply that taste and odour perception are independent? Probably not! Indeed, information coming from the gustatory and the olfactory systems are likely to be combined at a higher level of processing in the brain to give rise to a unique perception referred to as `flavour' (Abdi, 2002; Prescott, 1999). The question thus is: What is the nature of this combination? Is it an additive combination whereby taste and odour perception are simply added to form an overall perception or can we observe interactions between taste and odours? In this context, taste odour interaction refers to a modification in perceived taste intensity in the presence of an odour. For example, a sweet solution will taste sweeter in the presence of a vanilla aroma even though the vanilla aroma possesses no taste properties. The aim of this chapter is to examine whether such interactions exist for sweet taste perception. We begin with an overview of the literature on the effect of odour on sweet taste perception. Then we present the factors that might affect taste±odour interactions and the different underlying mechanisms proposed in the literature. Finally we highlight some implications of taste±odour interactions for food product development.
Taste±odour interactions in sweet taste perception
4.2
67
Overview of the literature
Early work suggested that odour and taste were independent. Murphy et al. (1977), for example, examined the perceived odour intensity, taste intensity and overall intensity of mixtures of sodium saccharin and ethyl butyrate. They found that the overall intensity of the mixtures approximate the simple sum of the intensities of the unmixed components (90%). The same pattern of results was also obtained by Murphy and Cain (1980) for sucrose-citral mixtures. However, in both studies, the authors indicate that, despite this apparent independency between taste and aroma, participants tend to perceive olfactory stimulation as taste. For example, Murphy and Cain (1980) report that whereas citral had no effect on sweet perception, it was judged to `have taste magnitude when presented alone'. This effect, referred to as `olfactory referral' by the authors, does not result from the stimulation of receptors in the oral cavity since it can be abolished by pinching the nostrils (Murphy and Cain, 1980; Schifferstein and Verlegh, 1996). It was first attributed to trigeminal stimulations (Murphy and Cain, 1980). According to these authors `the trigeminal system may serve to bind the anatomically and physiologically distinct olfactory and taste systems into a single perceptual system during eating' (see Murphy and Cain, 1980, p. 605). As we shall see in Section 4.3, other explanations are proposed to account for this effect. Since these first demonstrations of taste±odour confusion many studies have investigated the effect of odours on sweet taste perception over the past 20 years. Although those studies use different experimental paradigms, the general principle remains the same. Participants are presented with a series of solutions made of a sweetener alone and a sweetener plus an odorant. Their task is to estimate the sweet (and aroma) intensity of each solution. An odour-induced taste enhancement is said to occur if the perceived sweet intensity of the mixture is greater than the perceived sweet intensity of the sweetener alone. And inversely an odour-induced taste suppression is said to occur if the perceived sweet intensity of the mixture is smaller than the perceived sweet intensity of the sweetener alone. The picture emerging from these studies is somewhat confusing (cf. Table 4.1). However, despite some divergence in the results, most studies showed that sweet intensity can be enhanced by odour. For example, Frank and Byram (1988) reported that the perceived sweetness of whip cream samples increased when a strawberry aroma was added. The same type of effect was obtained by other authors in presence of several aromas including almond, caramel, coffee, lemon, peach and vanilla and for several sweeteners such as sucrose, fructose, aspartame and saccharine. This effect occurs both when subjects swallow the test solutions as well as when they spit it (Frank et al., 1989). In addition, Cliff and Noble (1990) using a peach aroma showed that increasing the concentration of aroma increased not only the intensity of perceived sweetness of a glucose solution but also its duration. More striking, Algom et al. (1993) indicated that both perceived and mentally imagined mixtures of different concentrations of glucose and orange aroma showed the
68
Optimising sweet taste in foods
Table 4.1 Overview of the impact of odours on sweet taste perception. SS Single scale ratings, MS multiple scale ratings D detection, M matching. indicates an enhancement effect, ÿ a suppression effect, no effect, not applicable Authors
Odorant
Tastant
SS
M
D
De Araujo et al. (2003) Djordjevic et al. (2004) Frank and Byram (1988) Frank et al. (1989) Frank et al. (1990) Frank et al. (1993) Nguyen (2000) Schifferstein and Verlegh (1996) Stevenson (2001) Stevenson et al. (1999) van der Klaauw and Frank (1994) van der Klaauw and Frank (1996) van der Klaauw and Frank (1996) van der Klaauw and Frank (1993) van der Klaauw and Frank (1993) Clark and Lawless (1994) Frank et al. (1991) van der Klaauw and Frank (1993) Frank et al. (1993) van der Klaauw and Frank (1993) van der Klaauw and Frank (1993) Frank et al. (1991) van der Klaauw and Frank (1993) Lawless and Schlegel (1984) Nguyen (2000) Schifferstein and Verlegh (1996) Frank et al. (1991) Clark and Lawless (1994) Nguyen (2000) Small et al. (2004) Lavin and Lawless (1998) Prescott (1999) Prescott et al. (2004) Stevenson et al. (1995) Stevenson et al. (1999) Djordjevic et al. (2004) Schifferstein and Verlegh (1996) Stevenson et al. (1995) Stevenson et al. (1999) Bingham et al. (1990) Stevenson et al. (1999) Frank and Byram (1988) Prescott (1999) Frank et al. (1991) Stevenson et al. (1999)
Strawberry
Sucrose
ÿ ÿ
Frank et al. (1993)
MS
Lemon
Sucrose Fructose Saccharine Aspartame Aspartame Sucrose
Vanilla
Aspartame Sucrose
Waterchestnut
Milk Sucrose
Ham
Sucrose
Lychee
Sucrose
Maltol
Sucrose
Peanut butter
Sucrose
Peanut Acetyl methyl carbamol Almond
Aspartame Sucrose
,ÿ ÿ ÿ ÿ ÿ ÿ ÿ
Sucrose
Fructose Saccharine Aspartame Wintergreen
,ÿ
Taste±odour interactions in sweet taste perception Table 4.1
69
Continued
Authors
Odorant
Tastant
SS
MS
M
D
Stevenson et al. (1999) Stevenson et al. (1999) Frank et al. (1991) van der Klaauw and Frank (1994) Stevenson et al. (1999) Stevenson et al. (1999) Stevenson et al. (1999) Stevenson et al. (1999) Prescott (1999) Algom et al. (1993) Cliff and Noble (1990) Prescott et al. (2004) Prescott (1999)
Angelica oil Caramel Chocolate Coffee Damascene Eucalyptol Mango Maracuja Oolong tea Orange Peach Prune Raspberry
Sucrose Sucrose Aspartame Sucrose Sucrose Sucrose Sucrose Sucrose Sucrose Sucrose Sucrose Sucrose Sucrose
ÿ
ÿ ÿ
same patterns of interaction. In other words imagined odours interact with perceived sweetness the same way as perceived odours (Djordjevic et al., 2004). Schifferstein (1997), however, indicates that in the case of mentally imagined mixtures participants might not base their response on mental images but rather on their knowledge of sensory interaction. Finally, Sakai et al. (2001) found that sweetness enhancement of aspartame by vanilla persists when vanilla is presented simultaneously with, but not dissolved in, the taste stimuli. So, to summarise it seems that sweet taste perception might be modified in the presence of an odour and that this effect is not due to changes in the physicochemical properties of tastants in the presence of odorants since it can be observed when odorants and tastants are presented separately. Yet, despite the many studies demonstrating an enhancement of sweet taste in the presence of an odour, the conditions of appearance of such an effect are far from being clear. Indeed, as can be noted on Table 4.1, all odours do not have the same effect on sweetness perception. Some odours such as strawberry, vanilla, lemon, almond, caramel, maracuja and lychee tend to increase sweetness intensity of sucrose or aspartame. Other odours tend to have no effect (peanut butter, ham, chocolate, mango and wintergreen) or tend to decrease sweetness intensity of sucrose or aspartame (liquorice, almond, damascene, and angelica oil). Moreover, the same odour does not seem to always induce a similar effect. For example, lemon seems sometimes to enhance sweet taste intensity, sometimes to suppress it and other times to have no effect.
4.3
Factors affecting taste±odour interactions
Three major classes of factors that we refer to as `task-driven', `stimulus-driven' and `subject-driven' factors have been proposed to explain the variability of the effect of odour on sweet taste.
70
Optimising sweet taste in foods
4.3.1 Task-driven factors A first explanation for the variability of the effects observed in Table 4.1 might be found in differences in the methodology used by the authors. For example, Lawless and Schlegel (1984) found a taste odour interaction in mixtures with variable sucrose and citral concentrations using Thurstonian scale values derived from triangle tests but not using attribute ratings. Inversely, Djordjevic et al. (2004) failed to find an effect of strawberry odour on sweet taste perception with a forced choice detection task whereas other authors found an effect with intensity scaling. These results indicate that the instructions given to the participants as well as the strategy implied by the task performed play an influential role in taste±odour interactions. Instructions given to the subjects Most studies of taste±odour interaction use intensity ratings as measurements. Several studies showed that, in this type of experimental set up, the instructions given to the subjects are crucial. In particular, Frank and his collaborators (Frank et al., 1990; van der Klaauw and Frank, 1996) and Clark and Lawless (1994) showed that sweetness enhancement depends upon the response alternatives given to the subjects. For example, in Frank et al. (1993), participants were asked to evaluate either the sweetness of sucrose and sucrose-strawberry mixtures or the sweetness, sourness and fruitiness of the same mixtures. When sweetness was the only alternative, an enhancement was observed: the sucrosestrawberry mixture was rated as sweeter than the sucrose solution. No such enhancement was observed when participants had to rate simultaneously the sweetness, the fruitiness, sourness and fruitiness of the mixtures. In another condition, participants were asked to first rate the global intensity of the mixture and then separate this global intensity in six ratings (sweet, salty, sour, bitter, fruity and other). A suppression effect was observed: The sucrose-strawberry mixture was rated as less sweet than the sucrose solution. According to Frank et al. (1993), providing participants with alternative scales helps them separate their perception in taste and odour perception and thus reduces olfactory referral. Further work by van der Klaauw and Frank (1996) examined if the effect of the instructions given to the subjects was due to the number of alternatives proposed to the participants or to the appropriateness of these alternatives. Six rating conditions were used: 1) sweet only, 2) sweet and bitter, 3) sweet, bitter, and floral, 4) sweet and floral, 5) sweet and fruity, 6) sweet, or fruity or bitter (participants in this condition had to make only one rating per stimulus but were not told before the presentation of the stimulus if this rating was sweet, fruity or bitter). As shown in Fig. 4.1, the number of alternatives did not play an important role in determining sweet taste enhancement: no decrease in odour-induced enhancement was observed when the number of alternatives increased. In contrast, the appropriateness of alternatives seems to be crucial: inappropriate alternatives such as bitter did not reduce taste enhancement. Finally, no enhancement was observed in the last condition thus indicating that attending to different
Taste±odour interactions in sweet taste perception
71
* p>0.05 (adapted from van der Klaauw and Frank (1996).
Fig. 4.1 Changes in sweetness perception (sweetness rating for the mixture minus sweetness rating for sucrose alone) for strawberry sucrose mixtures under six ratings condition.
components of the overall sensation is more important than actually making multiple ratings. Strategy implied by the task As we just saw the number of rating scales available to describe a mixture might affect the appearance of taste±odour interactions. According to Frank and his collaborators this effect can be explained in terms of attentional strategies. Providing participants with multiple scales encourages them to adopt an analytical approach whereas providing them with a single scale encourages them to adopt an integrative or synthetic approach. More specifically Frank (2002) indicates that `When subjects are asked to attend to multiple stimulus attributes, the instructions encourage them to disentangle the concepts for each of the attributes and this constricts the conceptual boundaries of the attributes and minimizes dimensional interactions' (see Frank, 2002, p. 142). For example, providing only a sweetness scale to evaluate a sucrose-strawberry mixture might lead to a wider interpretation of the sweetness concept including sensation such as fruitiness. By contrast providing both a sweetness and a fruitiness scale might lead to a more narrow sweetness concept excluding fruitiness sensation. An implication of this interpretation is that flavour perception (i.e., the combination of taste and odour) should be analysable into its taste and odour components by individuals trained to focus their attention on these components but not by untrained consumers. In agreement with this hypothesis, Bingham et al. (1990) found that untrained participants rated a mixture of maltol and sucrose as sweeter than an equivalent concentration of sucrose alone but not participants who have been trained to use an analytical tasting strategy. Such effects of
72
Optimising sweet taste in foods
training, however, are not always found. Indeed, Stevenson (2001) and Stevenson and Case (2003) found no difference in the ability of trained and untrained participants to separate an odour-sucrose mixture into its components. In other words, as Stevenson (2001) put it, `enhancement effects, and thus confusion between odours and tastes, may be more cognitively impenetrable than has sometimes been assumed' (see Stevenson, 2001, p. 242). 4.3.2 Stimuli-driven factors A second explanation for the variability of the effects observed in Table 4.1 might be linked to the nature of the odorant and tastant mixtures. Indeed Frank and Byram (1988), for example, found sweetness to be enhanced by strawberry odour, but not by peanut butter odour. According to Stevenson et al. (1995; see also Stevenson et al., 1998) the degree of association between tastant and odorant is crucial in taste±odour interactions. More specifically, these authors suggest that an interaction will occur when the odour and the taste have previously been experienced conjointly. Other authors mention as a determining factor the nature of the relationship between tastant and odorant. The harmony between tastant and odorant was first studied by Murphy and Cain (1980) who failed to show an effect of this factor on taste±odour interaction. Their results reveal the same pattern of additivity for both a harmonious (citral and sucrose) and an inharmonious (citral and sodium chloride) mixture. However, this result has to be interpreted with some caution because the degree of `harmony' between tastant and odorant was not estimated by the participants but was based on assumptions from the authors. Later, Frank et al. (1991) asked subjects to evaluate the similarity of pairs of odorants and tastants. They found that the more similar tastant and odorant pairs yielded the stronger effect of odour on taste. Using a similar method, Schifferstein and Verlegh (1996) examined the effect of the congruency between taste and odour. They found a taste enhancement for congruent mixtures (sucrose-strawberry, sucrose-lemon) but not for incongruent mixtures (sucroseham). In agreement with this result, a single cell recording study (Rolls and Bayliss, 1994) indicates that some cells of monkey orbitofrontal cortex respond to both olfactory and gustatory stimuli but mostly for congruent stimuli such as glucose and fruity odours. However, Schifferstein and Verlegh reported that the degree of congruence was not linearly correlated with the amount of odourinduced taste enhancement. They suggested that `the congruency judgements are not effective in explaining the enhancement because they do not reflect the degree of association between two elements but they reflect how much somebody likes the combination of two elements' (see Schifferstein and Verlegh, 1996, p. 102). A better predictor might be the smelled taste of the odorants (e.g., the sweetness rating of a strawberry aroma) as proposed by Stevenson et al. (1999). Indeed these authors reported that the degree to which an odour smelled sweet predicted about 60% of odour-induced sweetness enhancement or suppression. In this study caramel, maracuja, strawberry and lychee were found to enhance sweet perception whereas angelica oil and damascene suppressed it.
Taste±odour interactions in sweet taste perception
73
To evaluate if the different concepts used in the literature tapped the same aspect of the relationship between odorant and tastant, Nguyen (2000) asked four groups of participants to rate the harmony, the congruency, the similarity and the smelled taste of four mixtures. The four mixtures were chosen based on a free association task whose goal was to evaluate the cognitive association between odours and tastes (Sauvageot et al., 2000). Two of them were associated: sucrose-vanilla and acid-lemon. That is some subjects spontaneously answered acid when prompted with the word `lemon' and some sweet when prompted with the word `vanilla'. The two others were not associated. That is no subjects answered acid to vanilla or sweet to lemon. A fifth group of participants had to rate the sweetness or the sourness of the mixtures and two control mixtures made of pure sucrose or acid. Contrast analyses showed that, although the four rating scales were positively correlated, only smelled taste (40%), and similarity (29%), predicted the observed odourinduced taste enhancement. Moreover, a very high correlation was observed between similarity and smelled taste indicating that these two scales measure the same aspect of the odorant±tastant relationship, probably what Stevenson (2001) calls the confusion between odour and taste. The fact that harmony and congruence ratings did not predict taste enhancement is in agreement with previous data by Murphy and Cain (1980) and Schifferstein and Verlegh (1996). These aspects of the relationship between tastants and odorants might be more difficult to estimate and more prone to individual variations than smelled taste and similarity. Yet, the most surprising aspect of this study is that an odour-induced taste enhancement effect was observed for the acid-vanilla mixture. This mixture was considered as being incongruent, not harmonious and not similar. Furthermore, Sauvageot et al.'s (2000) study showed that no subjects associated the term `sour' with the term `vanilla' even in a forced choice association task in which the possible choices corresponded to the four fundamental tastes (sweet, salty, sour, bitter). Finally, the vanilla aroma was not perceived to smell `sour'. Thus, the positive effect of vanilla on sourness cannot be explained in terms of relationship, cognitive association, or perceptual similarity between odorant and tastant. This last result indicates rather that the similarity or confusability between odorant and tastant might not be a necessary condition for enhancement to occur. This last point will be discussed later in the chapter. 4.3.3 Subject-driven factors Additional constraints for the observation of an effect of odour on sweet taste seem to be subject-driven. We already saw that participants' degree of expertise might play an important role in the ability to separate or integrate olfactory and gustatory components of a mixture. Other factors affecting taste odour interactions include participants' culture and individual differences.
74
Optimising sweet taste in foods
Culture effect A first demonstration of the effect of culture on odour-induced sweetness enhancement resulted from an attempt at replicating Frank and Byram (1988) with French participants (Nguyen, 2000). In both Frank and Byram and Nguyen studies, participants had to rate the sweetness intensity of four whip cream samples containing respectively 0, 0.25, 0.60 and 1.20 M sucrose with and without the presence of a strawberry aroma. Whereas Frank and Byram observed a positive effect of strawberry on sweetness for all sucrose concentrations, Nguyen observed a positive effect only for the second concentration. Moreover, this effect was much smaller than that observed by Frank and Byram (2 vs. 6 points on a 20 points scale). This result can be explained in terms of food consumption habits and cognitive association between odorants and tastants. Whereas the association between strawberry and sweet is very frequent in American food (milk shake, ice cream, yogurts), it is less frequent in French food. This was confirmed in the free association task carried out by Sauvageot et al. (2000). For French subjects, strawberry is more often associated with a sour taste or a red colour than with a sweet taste as it is for American subjects. A link between food consumption habits and taste odour interaction was also reported in Nguyen et al. (2001). French and Vietnamese participants were presented with two sets of samples: a sweet set made of a sucrose solution, a sucrose-vanilla mixture and a sucrose-lemon mixture and a sour set made of an acid solution, a acid-vanilla mixture and a acid-lemon mixture. They were instructed to swallow the samples and to rate their sweet or sour intensity on a 10 cm unstructured line scale. An analysis of variance revealed a one-way interaction between type of mixtures and cultures. Pairwise comparisons showed that this interaction was mainly due to the sweet mixtures. Whereas for French subjects sweetness enhancement was higher in the presence of vanilla than of lemon, the opposite was observed for Vietnamese subjects. As for strawberry aroma, this cultural effect can be put in perspective with culinary association. Vanilla is very often used in France to flavour sweet dishes but not in Vietnam. In contrast, the association between lemon and sweet is less frequent in France than in Vietnam where lemon soft drinks are very popular. A study by van der Klaauw and Frank (1994) showed that in addition to cultural differences, individual differences might play a role in odour±taste interactions. Participants were given three solutions of sucrose (0.18, 0.30, 0.435 M), a mixture of 0.30 M sucrose and strawberry aroma, and a mixture of 0.30 M sucrose and coffee aroma. Their task was either to rate the sweet intensity of the five samples or to match each sample with a sucrose solution from a range of nine concentrations going from 0.13 to 0.55 M. While on the whole strawberry aroma and to a lesser degree coffee aroma enhanced the sweetness of sucrose solutions, participants clearly differed in their probability to show an enhancement. Interestingly, these individual differences were consistent across the two tasks. An individual showing an enhancement effect with the scaling method was likely to show the same effect with the matching task. The authors interpreted these individual differences as an indication that
Taste±odour interactions in sweet taste perception
75
some participants are better at separating their perception into olfactory and gustatory components.
4.4
Mechanisms of taste±odour interactions
Despite the different factors that might intervene the odour-induced taste enhancement or suppression effect seems to be a robust effect. The question that naturally arises, however, is at which level does this effect occur? It is unlikely to occur at the receptor level. Since as already mentioned sweet taste enhancements have been observed even in situations in which the odorant and the tastant were presented separately (Sakai et al., 2001; Djordjevic et al., 2004), it is more probable that it occurs at the central processing level. However, the central mechanisms involved in this effect are still controversial and three main views seem to emerge from the literature. 4.4.1 Response bias For some authors such as Clark and Lawless (1994), odour-induced taste enhancement could be explained in terms of response biases. These authors base their argumentation on the fact that a taste enhancement is observed only when subjects are provided with a single scale (cf. sections 4.3.1 and 4.3.2). Clark and Lawless (1994) interpreted this phenomenon as a `dumping' effect. According to their interpretation, when a participant is not given an appropriate rating scale to express a sensation, he/she `dumps' this sensation onto the only available scale(s). This behaviour would `correspond to a general tendency to integrate all information to increase subjective confidence in the judgments' (see Clark and Lawless, 1994, p. 591). For example, when subjects are asked to evaluate the sweetness intensity of an aspartame solution to which a strawberry aroma was added, they `dump' the strawberry sensation onto the sweetness scale and thus a sweetness overestimation is observed. An opposite effect can also be observed when too many scales are used. For example Clark and Lawless (1994) observed that an aspartame solution without any added aroma was rated as less sweet when participants were asked to rate both the sweetness and the flavour intensity than when they were asked to rate only the sweetness intensity. This suppression effect could be due `to a tendency to use response category with equal frequency' (p. 592). Within this framework, it is unclear whether taste±odour interactions are a reliable perceptual effect or simply a measurement artefact. Indeed, intensity rating methods indicate that odours might increase sweetness ratings but does that imply that it increases sweetness perception? To answer this question a few authors examined taste±odour interactions to see if they can be shown with paradigms different from intensity ratings. Djordjevic et al. (2004) for example used a sweet taste detection task. Participants were given pairs of samples made of a weak sucrose solution and a blank solution. They were asked to sip each solution immediately after having
76
Optimising sweet taste in foods
sniffed or imagined a strawberry or a ham odour and to indicate which of the two solutions tasted stronger. In both perceived and imagined conditions, detection of sucrose was poorer when combined with a ham odour than with no odour. No difference was observed with strawberry odour. This result clearly shows that taste±odour interaction can be obtained with a paradigm free of the response biases described by Clark and Lawless (1994). Nguyen et al. (2000) proposed another bias-free method to measure the effect of vanillin on sucrose. This method, based on Garner (1974) filtering paradigm, evaluates if a categorization task performed on sweetness perception is affected by changes in vanilla aroma. Participants had to make categorization judgements according to the level of the perceived sweetness intensity of samples made of two concentrations of sucrose and two concentrations of vanillin. Results show that categorization performance was systematically lower when the concentration in vanillin changed in addition to the concentration of sucrose. This indicates that participants were unable to ignore the olfactory component when categorizing the solutions according to the gustatory component. Taken together the Djordjevic et al. (2004) and Nguyen et al. (2000) results suggest that taste±odour interactions cannot be explained solely in terms of response biases. 4.4.2 Learned synesthesia For other authors such as Stevenson et al. (1998), the effect of odour on sweet taste perception could be explained in terms of learned synesthesia. Synesthesia is defined as a systematic association between two sensations corresponding to two different sensory modalities. A classical example of synesthesia can be found in Cytowic (1993). For the patient described in this book, each different taste had a different feel. Mint for instance felt like `smooth, cool columns of glass'. Stevenson et al. (1998, see also Stevenson et al., 1995) showed that synesthesia can be learned through a combined exposure to two sensations. Especially, they showed that repeated pairing of a previously neutral odorant with sucrose results in an increased perceived sweetness of the odorant. Participants attended five sessions, one session of pre-test, three sessions of conditioning and a session of post-test. During pre-test, participants had to rate among other ratings (liking, overall intensity, sourness, familiarity) the sweetness of lychee and water chestnut odours. During each conditioning session participants performed six triangle tests among which one consisted of water chestnut (or lychee) paired with sucrose (conditioning trial) and one of lychee (or water chestnut) and water (control condition). The other triangle tests served as fillers. During the post-test, participants had to rate again the sweetness of lychee and water chestnut odours and to answer an awareness test on the same odours. Results indicated that sweetness perceived intensity increased between pre-and post-tests for the odour paired with sucrose (water chestnut or lychee) but not for that paired with water. This increase in perceived sweetness occurred independently of participants' awareness that the
Taste±odour interactions in sweet taste perception
77
odour was paired with sucrose during the conditioning test and did not disappear after 24 h of delay. More recently, Prescott et al. (2004), using a similar paradigm, found that even a single co-exposure with sucrose was sufficient to enhance the perceived sweetness of a prune and a water chestnut odour. However they also found that a synthetic perceptual strategy during the co-exposure was necessary to observe a taste enhancement in the presence of the same odours. Indeed, participants instructed to evaluate the overall flavour of sweet-prune (or water chestnut) flavoured solutions during the conditioning phase showed an enhancement whereas participants instructed to evaluate separately the sweetness and the aroma of the same solutions did not. Prescott et al. (2004) concluded that `exposure alone is a necessary, but not sufficient, cause of sweetness enhancement effects' (see Prescott et al., 2004, p. 335). Both Stevenson et al. (1998) and Prescott et al. (2004) results emphasize the role of associative learning in odour-induced sweetness enhancement. However, can we really interpret this effect in terms of learned synesthesia as proposed by Stevenson et al. (1998)? Synesthesia is the involuntary physical experience of a cross-modal association. It is characterized by idiosyncratic, vivid and irrepressible perceptions and does not seem to rely on any form of learning (Cytowic and Wood, 1982). In contrast, taste±odour interactions seem to result from perceptual learning and to be culture-dependent rather than idiosyncratic. An interpretation in terms of central integration as proposed by Prescott et al. (2004) seems more appropriate. Consistent with this interpretation, White and Prescott (2002) showed that an odour can prime the cognitive system to expect a particular type of taste based on past flavour experiences. More specifically, reaction time to identify a taste in the presence of an orthonasally delivered odour decreases if the taste and the odour have been encountered conjointly in a previous conditioning phase. Likewise, White and Prescott (2001) found that participants identified faster a sweet taste in the presence of a congruent odour such as strawberry than in the presence of an incongruent odour such as grapefruit. 4.4.3 Central integration Integration between odour and taste could occur in a brain region called the orbitofrontal cortex (Rolls, 2004). This brain structure receives projections from the olfactory cortex and the gustatory cortex. As already mentioned, Rolls and Bayliss (1994) found in the orbitofrontal cortex of monkeys some neurons that respond to both olfactory and gustatory inputs. Among these bimodal neurons, some responded to odours and tastes that occur together in food. For example, a given neuron will respond preferentially to a sweet taste in a taste discrimination task and to a fruit odour in an odour discrimination task. Moreover, Rolls et al. (1996) showed that the response of some olfactory neurons in the monkey orbitofrontal cortex depends upon the taste with which the odour is associated. Rolls (1997) suggested that bimodal neurons would develop from unimodal
78
Optimising sweet taste in foods
neurons through associative learning of tastes and odours occurring naturally together in food. Imaging techniques have been used recently to assess the role of central brain structures in flavour perception by humans. Small et al. (1997) used positron emission tomography (PET) to compare cerebral blood flow (CBF) changes due to the processing of olfactory (e.g. strawberry), gustatory (e.g sucrose), and combined olfactory gustatory stimuli (e.g. strawberry, sucrose). They found that bimodal processing of taste and odours produces significant CBF decreases relative to unimodal processing of identical tastes and odours. This decrease in CBF can be related to suppression effects observed in psychophysical experiments. More recently de Araujo et al. (2003), using an event-related functional magnetic resonance imaging (fMRI) study, reported an activity in the orbitofrontal cortex when a tastant (sucrose or monosodium glutamate) and a retronasally delivered odorant (strawberry or chicken-like) were presented simultaneously. This activity was greater than the sum of the activities obtained when tastant or odorant were presented separately, thus indicating that an interaction occurred. A similar finding was also reported by Small et al. (2004) thus indicating that orthonasal and. retronasal olfaction might interact differently with taste. Indeed, Small et al. (1997) found a taste suppression effect with orthonasally delivered odours. This differential effect of orthonasal and retronasal olfaction on taste was not, however, always found in psychophysical studies (Sakai et al., 2001). Small et al. (2004) also showed that superadditive responses were observed only when the taste±odour pair was congruent (sucrose-vanillin as opposed to NaCl-vanillin). Such patterns of superadditivity seem to be the physiological correlate of psychophysical data by Dalton et al. (2000). These authors found that participants could detect the presence of benzaldhyde at 30% below its detection threshold when the odorant was combined with sodium saccharin at 30% below its detection threshold. In a second study, Breslin et al. (2001) failed to find a superadditivity effect when benzaldehyde was presented in combination with monosodium glutamate. 4.4.4 Towards a dual model The literature reviewed so far indicates that two different cognitive mechanisms might be involved in taste±odour interactions. The first one would not require any previous association between odorants and tastants and would give rise to small amplitude effects. It would correspond to a general tendency to integrate all information into a total intensity impression (dumping effect) and thus would be sensitive to the number of rating scales. In contrast, because it acts more like a response bias than a perceptual enhancement, it would not be sensitive to subjects' history and thus would not be mixture dependent. The second one would correspond to a real perceptual enhancement and give rise to high amplitude effects. It would result from the co-encoding of an odorant and a tastant and thus would be mixture and culture dependent (Nguyen, 2000).
Taste±odour interactions in sweet taste perception
79
Fig. 4.2 Changes in sour perception (sourness rating for the mixture minus sourness rating for citric acid alone) for vanilla citric acid and lemon citric acid mixtures under two rating conditions (adapted from Valentin and Nguyen, 2001).
To test this dual mechanism hypothesis, Valentin and Nguyen (2001) evaluated the effect of lemon and vanilla aroma on sourness ratings. The experiment consisted of two intensity tasks. In the first intensity task (1 question condition) subjects had to indicate the sourness of an acid solution, a vanillaacid solution and a lemon-acid solution. In the second intensity task (3 questions condition) subjects were given the same solutions but were instructed to rate the sour, vanilla and lemon intensity of the mixtures on three different scales. As illustrated in Fig. 4.2, student t-tests revealed a significant odour-induced taste enhancement for both vanilla and lemon in the 1 question condition with a higher amplitude effect for lemon than for vanilla. In the 3 questions condition, the effect of vanilla on sourness was not significant whereas the effect of lemon is significant. This result indicates that providing several scales to subjects does not have the same effect on odour-induced taste enhancement depending on the association between tastants and odorants. Subjects seem to be able to separate taste and odour sensations only for taste±odour mixtures in which the components are not cognitively associated, similar, harmonious, congruent, and familiar (e.g. acid-vanilla). For other mixtures (e.g. acid-lemon), subjects seem to be able to separate only partially their sensations: providing several scales decreases the odour-induced taste enhancement effect but does not suppress it.
4.5
Implications for food product development
Odour-induced sweet taste enhancement or suppression appear to be a robust effect albeit not a strong one. Indeed, while odour and taste sensation in the mouth can, to some extent, be analysed, sensory integration might be more frequent in natural eating conditions. This is reflected by the spontaneous use of the term `sweet' to describe food related odours (Dravnieks, 1985). The question thus is can we use this natural tendency to integrate sensations to enhance the sweetness of a food product via its aromas? The answer is yes, but the result will depend on several factors. First, all odours will not be equally useful to enhance
80
Optimising sweet taste in foods
sweet taste. Odours confusable with sweet taste because they have been experienced often in conjunction with a sweet taste are the best candidates to enhance sweet taste. Second, the ability of an odour to enhance sweet taste does not seem to be universal: an efficient sweet taste enhancer in a given culture might not be as useful in another one. Finally interactions between sweet taste and odours might vary as a function of individual sensitivity and strategy in food tasting. Individuals trained to analyse their sensations might be less prone to taste odour interactions than other individuals. Taste odour interactions also have broad implications for descriptive analyses in which multiple attributes of complex foods are rated. For some time scientists in sensory evaluation thought that panellists could be trained to separate their sensations. The data presented here show that this is not always the case. For example, it is quite possible that even for a trained panellist the sweetness produced by a fruity aroma cannot be separated from the sweetness produced by sucrose. This raises the question of what do we want to measure when we ask panellists to rate the sweetness of a food product? The overall sweetness of the product or the sweetness yielded by the taste component? According to Prescott (1999), `this distinction might be artificial [. . .] since the sweetness of the flavour components may be represented cognitively, and even perhaps neurally, as equivalent. They may be functionally equivalent as well' (Prescott, 1999, p. 355). In other words, in some cases, such as understanding consumer preferences, it might be preferable to measure the overall perceived sweetness of a food product. In other cases, where it is necessary to estimate only the sweetness produced by the taste component of a food, the rating scales provided to the panellists should be chosen with care to help them separate their sensations. Indeed instructing a panellist to focus only on taste perception will not be enough to avoid odour-induced taste enhancement.
4.6
Future trends
One potentially interesting line of research will be to further examine the effect of the degree of cultural experience with particular combinations of tastants and odorants in odour-induced sweet taste enhancement. Cultural factors are one of the most powerful determinants of which food we consume. Nevertheless, there has been surprisingly little research on how perception of, and preference for food might vary across cultures. To understand odour-induced sweet enhancement as a function of cultural experience may provide some new insights in the understanding of cultural differences in preferences for sweet foods. A second line of research will be to extend the study of odour±taste interactions to multiple interactions in sweet taste perception. It is well known that when eating food, sensations such as somatosensory, visual or auditory sensations influence food perception and preference (Delwiche, 2004). This may be particularly relevant also from a neuroanatomical point of view to evaluate how the different sensory inputs interact with each other and how they are
Taste±odour interactions in sweet taste perception
81
integrated in the brain. In this perspective, imaging techniques will be very useful. Finally, focus should be given to a more general consequence of sweet taste enhancement: the improvement of palatability and the compensation for chemosensory losses in populations such as the elderly. Recently, it has been demonstrated that the addition of a flavour enhancer can improve a number of nutritional and immune parameters such as increasing the number of lymphocyte cells or increasing secretion rate of specific antibodies (Schiffman, 2002). As flavour enhancement can be achieved in several ways, in addition to further investigations into the contribution of neural integration of sensory inputs, it is crucial to examine also other sources of sweet taste enhancement such as modification of receptor mechanisms.
4.7
Sources of further information and advice
4.7.1 Reviews A few reviews are of interest to acquire further information and advice on taste±odour interactions. Prescott (1999) provides a broad body of literature on taste±odour interactions, and also raises the question of the different origins of odour±taste interactions and provides evidence that taste enhancement is a psychological construct. Frank (2002), based on a large number of studies from his laboratory, proposes a model that attempts to characterize taste±odour interactions when making judgements about the sensory magnitude of complex chemosensory stimuli. Recommendations are furthermore made based on studies of response context and flavour research. Delwiche (2004) provides a large scope of investigations on flavour perception. This paper reviews studies that look at the impact of different sensory cues on the perception of taste, odour and taste±odour mixtures, as well as the impact of certain physical interactions on these perceptions such as influence of irritation, temperature, colour, texture and sound. Readers interested in neuroanatomical issues of odour±taste interactions will find some thorough information in Rolls (1999). 4.7.2 Current research and interest groups The Monell Chemical Senses Center in Philadelphia (USA) and the Sensory Science Research Centre at the University of Otago (New Zealand) are the main references in terms of research groups investigating taste±odour interactions.
4.8
References
(2002), `What can cognitive psychology and sensory evaluation learn from each other?', Food Qual Pref, 13, 445±451. ALGOM D, MARKS L E and CAIN W (1993), `Memory psychophysics for chemosensation: perceptual and mental mixtures of odor and taste', Chem Senses, 18, 151±160. ABDI H
82
Optimising sweet taste in foods
and PERRING K D (1990), `Sensory studies with sucrose-maltol mixtures', Chem Senses, 15, 447±456. BRESLIN, P A, DOOLITTLE H and DALTON P (2001), `Subthreshold integration of taste and smell: The role of experience in flavour integration', Chem Senses, 26, 1035. CLARK C C and LAWLESS H T (1994), `Limiting response alternative in time-intensity scaling: an examination of the halo-dumping effect', Chem Senses, 19, 538±594. CLIFF M and NOBLE A C (1990), `Time-intensity evaluation of sweetness and fruitiness and their interaction in a model solution', J Food Sci, 55, 450±454. CYTOWIC R E (1993) The Man Who Tasted Shapes: A Bizarre Medical Mystery Offers Revolutionary Insights into Reasoning, Emotion, and Consciousness. New York: Putnam. CYTOWIC R E and WOOD F B (1982) `Synesthesia I: A review of major theories and their brain basis'. Brain and Cognition, 1, 23±35. DALTON P, DOOLITTLE N, NAGATA H and BRESLIN P A (2000), `The merging of the senses: integration of subthreshold taste and smell', Nature Neurosci, 3, 431±432. DE ARAUJO I E, ROLLS E T, KRINGELBACH M L, MCGLONE F and PHILLIPS N (2003), `Tasteolfactory convergence, and the representation of the pleasantness of flavour, in the human brain', Eur J Neurosci, 18, 2059±2068. DELWICHE J (2004), `The impact of perceptual interactions on perceived flavour', Food Qual Pref, 15, 137±146. DJORDJEVIC J, ZATORRE R J and JONES-GOTMAN M (2004), `Effects of perceived and imagined odors on taste detection', Chem Senses, 29, 199±208. DRAVNIEKS A (1985), Atlas of Odor Character Profiles, Philadelphia: ASTM. FRANK R A (2002), `Response context affects judgments of flavour components in foods and beverages', Food Qual Pref, 14, 139±145. FRANK R A and BYRAM J (1988), `Taste-smell interactions are tastant and odorant dependent', Chem Senses, 13, 445±455. FRANK R A, DUCHENY K and MIZE S J S (1989), `Strawberry odor, but not red color, enhances the sweetness of sucrose solutions', Chem Senses, 14, 371±377. FRANK R A, WESSEL N and SHAFFER G (1990), `The enhancement of sweetness by strawberry odor is instruction-dependent', Chem Senses, 15, 576±577. FRANK R A, SHAFFER G and SMITH D V (1991), `Taste±odor similarities predict taste enhancement and suppression in taste±odor mixture', Chem Senses, 16, 523. FRANK R A, VAN DER KLAAUW N J and SCHIFFERSTEIN H N J (1993), `Both perceptual and conceptual factors influence taste±odor and taste±taste interactions', Percept Psychophys, 54, 343±354. GARNER W R (1974) The Processing of Information and Structure. Potomac, MD: Lawrence Erlbaum Associates. LAVIN J G and LAWLESS H T (1998) `Effect of color and odor on judgments of sweetness among children and adults', Food Qual Pref, 9, 283±289. LAWLESS H and SCHLEGEL M P (1984), `Direct and indirect scaling of sensory differences in simple taste and odor mixtures', J Food Sci, 49, 44±51. MURPHY C and CAIN W C (1980), `Taste and olfaction: independence vs. interaction', Physiol Behav, 24, 601±605. MURPHY C, CAIN W S and BARTOSHUK L M (1977), `Mutual action of taste and olfaction', Sensory Processes, 1, 204±211. Á l'eÂtude des interactions entre les entreÂes sensorielles: NGUYEN H D (2000), Contribution a Effets d'une odeur sur la perception d'une saveur, TheÁse de Doctorat de l'Universite de Bourgogne, Ensbana, Dijon. BINGHAM A F, BIRCH G G, DE GRAAF C, BEHAN J M
Taste±odour interactions in sweet taste perception
83
and VALENTIN D (2000), `Perceptual separability and taste± odour interactions', Poster presented at the European Chemoreception Research Organisation conference, Brighton, England, 20±24 July. NGUYEN H D, VALENTIN D, LY M H, CHREA C and SAUVAGEOT F (2001), `When does smell enhance taste? The effect of culture and odorant/tastant relationship', Poster presented at the 4th Pangborn Sensory Science Symposium Dijon, France. PRESCOTT J (1999), `Flavour as a psychological construct: implications for perceiving and measuring the sensory qualities of foods', Food Qual Pref, 10, 349±356. PRESCOTT J, JOHNSTONE V and FRANCIS J (2004), `Odor±taste interactions: effects of attentional strategies during exposure', Chem Senses 29, 331±340. ROLLS E T (1997), `Taste and olfactory processing in the brain and its relation to the control of eating', Crit Rev Neurobiol, 11, 263±287. ROLLS E T (1999), The Brain and Emotion, New York: Oxford University Press. ROLLS E T (2004), `The functions of the orbitofrontal cortex', Brain Cogn, 55, 11±29. ROLLS E T and BAYLISS L L (1994), `Gustatory, olfactory, and visual convergence within the primate orbitofrontal cortex', J Neurosci, 14, 5437±5452. ROLLS E T , CRITCHLEY H, MASON R and WAKEMAN E A (1996), `Responses of neurons in the primate taste cortex to the glutamate ion and to inosine 50 -monophosphate', Physiol Behav, 59, 991±1000. SAKAI N, KOBAYAKAWA T, GOTOW N, SAITO S and IMADA S (2001), `Enhancement of sweetness ratings of aspartame by a vanilla odor presented either by orthonasal or retronasal routes', Percept Motor Skills, 92, 1002±1008. SAUVAGEOT F, NGUYEN H. D. and VALENTIN D (2000), `Les mots eÂvoquent-ils des saveur? Une comparaison entre eÂtudiants de France, du Vietnam et des USA', Sciences des Aliments, 20, 491±522. SCHIFFERSTEIN H N J (1997), `Perceptual and imaginary mixtures in chemosensation', J Exp Psychol: Hum Percept and Perf, 23, 278±288. SCHIFFERSTEIN H N J and VERLEGH P W J (1996), `The role of congruency and pleasantness in odor-induced taste enhancement', Acta Psychol, 94, 87±105. SCHIFFMAN S S (2002), `Flavor enhancement and its positive health benefits', Aroma-chol rev, X, 1±5. SMALL D M, JONES-GOTMAN M, ZATORRE R J, PETRIDES M and EVANS A (1997), `Flavor processing: more than the sum of its parts', Neuroreport, 8, 3913±3917. SMALL D M, VOSS J, MAK Y E, SIMMONS K B, PARRISH T and GITELMAN D (2004), `Experiencedependent neural integration of taste and smell in the human brain', J Neurophysiol, 92, 1892±1903. STEVENSON R J (2001), `Is sweetness taste enhancement cognitively impenetrable? Effect of exposure, training and knowledge', Appetite, 36, 241±242. STEVENSON R J and CASE T J (2003), `Preexposure to the stimulus elements, but not training to detect them, retards human odour-taste', Learn Behav Proc, 61, 13±25. STEVENSON R J, PRESCOTT J and BOAKES R A (1995), `The acquisition of taste properties by odors', Learn Motiv, 26, 1±23. STEVENSON R J, BOAKES R A and PRESCOTT J (1998), `Changes in odor sweetness resulting from implicit learning of a simultaneous odor-sweetness association: an example of learned synesthesia', Learn Motiv, 29, 113±132. STEVENSON R J, PRESCOTT J and BOAKES R A (1999), `Confusing tastes and smell: How odours can influence the perception of sweet and sour tastes', Chem Senses, 24, 627±635. VALENTIN D and NGUYEN H D (2001), `Peut-on parler d'aro à mes sucreÂs? Le point de vue de NGUYEN H D, DACREMONT C
84
Optimising sweet taste in foods
la psychologie cognitive', Proc Colloque Alliance 7, Cedus, Paris 2002. and FRANK R A (1993), `Taste±smell interactions with multiple sweeteners', Chem Senses, 18, 580. VAN DER KLAAUW N J and FRANK R A (1994), `Matching and scaling of taste±smell mixtures: Individual differences in sweetness enhancement by strawberry odor', Chem Senses, 19, 567±568. VAN DER KLAAUW N J and FRANK R A (1996), `Scaling component intensity of complex stimuli: The influence of response alternatives', Environment International, 22, 21±31. WHITE T and PRESCOTT J (2001), `Odors influence speed of taste naming', Poster presented at the XXIII AChems meeting, Sarasota FL, 23±25 April. WHITE T and PRESCOTT J (2002), `Learned associations with smells influence taste naming speed', Poster presented at the European Chemoreception Research Organisation conference. Erlangen, Germany, 23±27 July. VAN DER KLAAUW N J
5 Taste±ingredient interactions modulating sweetness M. Lindley, Lindley Consulting, UK
5.1
Introduction
In mixtures of substances with different taste qualities, the individual components can still be recognised but are usually perceived as less potent than when unmixed. This phenomenon is known as `mixture suppression' (Bartoshuk, 1975) and is normally observed when compounds of differing taste qualities are mixed together. Sensory interactions between sweet tasting compounds, and between sweeteners and compounds of different taste qualities, influence the overall taste quality delivered in many food and beverage products, as well as potentially having important commercial implications. From practical standpoints, interactions between sweet tasting compounds are particularly relevant due to the sometime occurrence of sweetness synergy. Understanding the interactions between sweet tasting compounds and acidulants is also important because of the use of sweeteners and acids in combination in many food and beverage products. Sweet and bitter compounds are found together in some foods and beverages, so it is also appropriate to develop an understanding of these sensory interactions. Additionally, many high potency sweeteners deliver bitterness in addition to sweetness, thus confirming the need to understand how these sensations interact. In contrast, the use of sweet and salty tasting compounds together is not a widespread practice in foods and beverages, although inorganic salts can exert significant influences over the sensory delivery of many sweet compounds. Interactions between ingredients that elicit non-sweet taste sensations with sweet tasting compounds are important influences on overall taste and product
86
Optimising sweet taste in foods
quality. In addition, however, as has been mentioned, the taste profiles delivered by sweeteners, particularly high potency sweeteners, are often complex mixtures of tastes. High potency sweeteners frequently elicit taste sensations additional to sweetness, with associated bitter tastes being particularly prevalent. Additionally, cooling aftertaste sensations, coupled with liquorice notes, are frequent complications of high potency sweeteners. Ingredient interactions that result in these taste sensations being reduced in their intensity have been widely studied, not least because of the important commercial implications of success.
5.2
Interactions between sweeteners and bitter compounds
Many food and beverage products are both sweet and bitter. Important examples include chocolate confectionery and coffee products, both of which contain intrinsic bitterness that is balanced by added sweetness. The ability of sucrose to balance the bitterness of cocoa and of the bitterness of cocoa to reduce the perceived sweetness of sucrose, both reductions being a consequence of mixture suppression, is arguably the most important market example of this sensory phenomenon. 5.2.1 Mixture suppression of sweet and bitter compounds Mixture suppression effects between sweet and bitter tasting compounds are reliable; mutual suppression of sweetness and bitterness always occurs. There is clear evidence that these mixture suppression effects are not due to any chemical interactions in solution or competition of molecules for common receptor sites, but that the effects are peripherally mediated (Lawless, 1979, 1982). In the first of these studies (Lawless, 1979), removal of sweetness from bitter-sweet mixtures, either by adapting to sucrose or pre-treatment of the tongue with an extract solution of Gymnema sylvestra, caused the perceived intensity of bitterness to increase. Since neither adaptation nor sweetness inhibition by Gymnema affected the concentration of sucrose on the tongue, it was concluded that mixture suppression must be due to neural inhibition. As part of the same study, Lawless also showed that similar dependence of suppression was observed with mixtures of phenylthiocarbamide (PTC) and sucrose in which those assessors that perceived PTC as bitter exhibited stronger suppression of sweetness than assessors that were insensitive to the bitterness of PTC. It was concluded that these observations were also inconsistent with molecular interactions mediating suppression, further supporting the neural inhibition conclusion. The second study by Lawless (1982) used a dorsal flow technique to control the exposure of the tongue to quinine, sucrose and a quinine-sucrose mixture. Previously documented effects of adaptation, mixture suppression and release from suppression were observed, as anticipated. In addition, reliable residual bitter or sweet taste was measured following adaptation to quinine-sucrose and presentation of equi-molar unmixed quinine or equi-molar unmixed sucrose.
Taste±ingredient interactions modulating sweetness
87
This residual taste was significantly greater than following self-adaptation of the unmixed solutions. It was concluded that these observations suggest a peripheral mechanism is responsible for the suppression between sweet and bitter tastes in a mixture. In contrast, a subsequent study (Kroeze and Bartoshuk, 1985), that also confirmed bitterness suppression in mixtures of quinine hydrochloride and sucrose, concluded that bitterness suppression of quinine hydrochloride in mixtures with sucrose occurs centrally. However, irrespective of the mechanism of mutual mixture suppression of bitterness and sweetness, it is a reliable phenomenon. There are individual differences in sensitivity to these mixture suppression sensory effects that are largely related to the ability of each individual to taste 6-n-propylthiouracil (PROP) or PTC as bitter. It has been demonstrated unequivocally that individuals vary in the extent to which they taste both phenylthiocarbamide (PTC) and 6-n-propylthiouracil (PROP) as bitter (Fox, 1932; Bartoshuk, et al., 1994). For many years, it was assumed that there are two population sub-groups; those that taste PTC/PROP as intensely bitter at a given concentration and those that taste the same concentration as very slightly bitter or tasteless. Now, however, it is acknowledged that there are in fact three subgroups, non-tasters, medium tasters and super-tasters (Bartoshuk et al., 1992). Notwithstanding this evolution in the classification of PTC/PROP tasters, some studies have shown that there are clear differences in the suppressive effect of quinine hydrochloride on sucrose that appear to be dependent on an individual assessor's classification as super-taster, medium-taster or non-taster of PTC/ PROP. In one comprehensive study, Prescott et al. (2001) demonstrated that quinine suppresses the sweetness of sucrose, but primarily only for super-tasters and medium-tasters. These observations are consistent with the fact that supertasters and medium-tasters rated quinine hydrochloride as more intensely bitter than did non-tasters and so an important conclusion from these studies is that genetic differences in taste sensitivity may influence the perception of taste qualities differently in complex, formulated foods and beverages. If so, a PTC/ PROP super-taster might be more likely than a non-taster to perceive the bitterness of cocoa in chocolate or bitterness in fruit juices, for example, and these groups might therefore be likely to exhibit marked differences in their ratings of product acceptability. In food and beverage companies where product quality and acceptability are not judged in formal sensory assessment by a taste panel, but by key technical or marketing personnel, the potential dangers are obvious. However, other studies have failed to confirm these effects with quinine. For example, Mela (1989) and Schifferstein and Frijters (1991) both failed to demonstrate a relationship between sensitivity to PROP and the perception of quinine hydrochloride taste. Therefore, as has been noted by Mattes (1994), the PROP sensitivity/food dislikes relationships reported for particular foods have tended to be inconclusive, possibly symptomatic of the complexity of assessing food acceptability.
88
Optimising sweet taste in foods
5.2.2 Bitterness reduction in sweet/bitter high potency sweeteners It has been established that there are clear qualitative differences among sweeteners, principally among the high potency sweeteners, and that these are frequently expressed as being due to bitter or metallic tastes (Schiffman et al., 1979). The high potency sweeteners saccharin, acesulfame-K, stevioside and rebaudioside A are specific examples of sweeteners that also elicit slight bitterness. From a practical standpoint, routes to the amelioration of the bitter aftertaste of saccharin have been the focus of much attention, mainly because during the whole of the 1970s, saccharin was the only permitted high potency sweetener in the USA. A number of ingredient additives have been evaluated for their bitterness reducing characteristics when formulated with saccharin. Specifically, glucono-lactone, sodium gluconate and cream of tartar (potassium bitartrate) have been proposed by Eisenstadt (1972) as a route to bitterness elimination from saccharin. This formulation must be considered to have been a success since it has formed the basis of the successful `Sweet 'n' Low' table-top sweetener product that is based on the sweetener saccharin. Another interesting approach to the elimination of the bitter aftertaste of saccharin was described by Pampiano (1980). It was claimed that when the aqueous extract of Gentiana lutea is added at low levels to saccharin, the bitter aftertaste conventionally associated with saccharin is substantially reduced or eliminated and that higher levels of saccharin may be employed to produce higher sweetness levels than are normally used. This approach is particularly interesting because G. lutea has been utilised in extract form as a tonic and is itself bitter. Gentiopicrin, the active component responsible for the bitter taste of gentian, is present in G. lutea and yet the extract does act to prevent the bitter aftertaste conventionally associated with saccharin itself. In other words, it may be functioning through some form of competitive binding at the receptor for saccharin bitter taste on the tongue. Similar observations were made by Wong et al. (1992) who disclosed that the sweet taste quality of acesulfame-K in an acidic environment may be improved and enhanced by the addition of potassium chloride. They propose that potassium chloride should be added in an amount approximately equal to that of the acesulfame-K. Again, given the intrinsic bitter taste of potassium chloride, the mechanism whereby the bitterness of acesulfame-K is reduced may be due to competitive binding at a bitter receptor site on the tongue. Other ingredients that have been reported (Lindley, 1999) to improve that sensory delivery of saccharin through amelioration of its bitter aftertaste include calcium chloride, arabinogalactan, D-tryptophan and neodeosmin.
5.3
Interactions between sweeteners and acids
Foods and beverages are formulated across a broad range of pH levels and intensities of perceived acidity. The pH range extends from approximately pH
Taste±ingredient interactions modulating sweetness
89
2.5 to around neutrality and perceived acidities vary from the mild acid taste intensity of fermented milk products such as yoghurt through to the intense acidity of some fruit based hard-boiled confectionery products and some soft drinks. In many food products, sucrose and other carbohydrate sweeteners are used for a variety of reasons in addition to sweetness delivery, including the provision of bulk and the development of texture. Therefore, carbohydrate sweeteners tend to be used at concentrations sufficient to deliver the necessary bulk and texture and not for reasons of sweetness delivery having achieved a particular level. As a result, any sweetness modifying impact of acids on carbohydrate sweeteners is of practical relevance only in that the amount of acid added should be sufficient to achieve the desired sweetness-acidity balance. In contrast, high potency sweeteners are mono-functional ingredients, contributing sweetness alone to the foods and beverages in which they are formulated. In these circumstances, potentiation or inhibition of high potency sweetener derived sweetness by acids can have important commercial implications, particularly since most applications for high potency sweeteners are in media that are acidified and their main use is in beverages that have pronounced acid tastes. 5.3.1 Carbohydrate sweetener±acid interactions Conventional concepts of mixture suppression suggest that an acid taste should suppress the sweetness delivery of a sweet substance. This certainly seems to be the case when sucrose is mixed with acids, as was first described by Pangborn (1961). McBride and Finlay (1990) examined the effects of citric acid on sucrose and also found a suppression of sweetness. Similarly, Schifferstein and Frijters (1990), using the `equi-ratio' method in which the mixing ratio is held constant across a series of solutions, reported that citric acid has a slight suppressive effect on the sweetness of sucrose, and that the suppressive effect of sucrose on the acid taste intensity of citric acid is more marked. Prescott et al. (2001) also showed that the sweetness of sucrose was suppressed by increasing concentrations of citric acid and that increasing concentrations of sucrose suppressed the sourness of citric acid in a progressive fashion. Thus, these and other publications all agree that mixtures of sucrose and citric acid mutually suppress the taste intensities delivered by each component. Of relevance from a practical standpoint, it has been reported that the carbonation of sweetener solutions also affects the perceived intensity of sweetness. Carbonation, which of course releases carbonic acid into beverages, apparently suppresses sucrose sweetness at practical use levels, but, paradoxically, enhances sweetness at concentrations approaching threshold (Odake, 1996). The influence of acidity on the sweetness of fructose is a special case complicated by the fact that fructose in solution is present as a mixture of anomeric forms; -pyranose, -furanose and its keto-hexose form. Of these anomeric conformations, only the -pyranose anomer is believed to be sweet. As the pH of a solution of fructose is reduced, the equilibrium of fructose
90
Optimising sweet taste in foods
anomers is shifted in favour of the sweet -pyranose form, thus resulting in sweetness being enhanced (Osberger and Linn, 1978). It is for this reason that the sweetness of a fructose solution is at a maximum at cold temperatures and low pH. 5.3.2 High potency sweetener±acid interactions There is very little published literature on the influence of food acids on the sweet taste delivered by high potency sweeteners and much of what has been published is difficult to interpret due to the use of different sensory methods and the failure to quantify sweetness in terms of sucrose equivalence. This is illustrated by early work with aspartame that indicates equivalent potencies at pH 3.2 and in water (Cloninger and Baldwin, 1974). However, there is more reliable evidence that aspartame tastes sweeter relative to sucrose in an acid solution compared to a water solution (Fry, 1993) and there are other indications from the literature that acids may enhance sweetness (Lotz and Meyer, 1994). While it is not clear whether it is the sweetness of aspartame that is actually enhanced or the sweetness of the sucrose reference that is reduced, the commercial consequences are that less sweetener is required than would be expected by examination of sucrose equivalence values in water. Most of the few studies that have been conducted have examined sweetener performance in citric acid buffers. However, since the food acids exhibit different sensory profiles, and indeed are often marketed on that basis, it would be surprising if all food acids interacted in the same way with all high potency sweeteners. Unfortunately, there have been no systematic studies to examine possible differences in sensory effects, although Pettigrew and Silka (1988) did conclude that the sweetness of aspartame is enhanced more by phosphoric acid than it is by citric acid. What is perhaps surprising, given the commercial importance of this observation, is that it is not more widely appreciated within the industry and that the findings have not been judged sufficiently important to stimulate further study in to the phenomenon. As has been mentioned, different food acids exhibit different sensory properties, particularly with respect to their temporal properties. For example, malic acid is marketed as delivering a longer-lasting acid taste relative to the acid taste of citric acid. This difference is claimed to make malic acid more ideally suited for use with high potency sweeteners, particularly with those sweeteners that deliver longer-lasting sweet aftertastes. However, no systematic study appears to have been conducted to develop objective sensory data to underpin these claims and there are also no data examining the impact of acids on blended sweetener performance, but since blends tend to be used in the majority of cases today, this apparent oversight seems particularly surprising. The ability of different acids to modify the temporal properties of specific high potency sweeteners has been examined in individual cases with some potentially important conclusions. For example, Shamil (1998) identified tannic acid compounds as additives that have the capability to reduce the lingering
Taste±ingredient interactions modulating sweetness
91
sweet aftertaste of sucralose. Data were presented in which a cola beverage sweetened with high fructose corn syrup was compared with one sweetened with sucralose and another sweetened with sucralose plus a small concentration of tannic acid. Sensory assessors were required to rate the intensity of sweetness, sourness, bitterness and sweet aftertaste duration, concluding that only sweet aftertaste duration was affected by the tannic acid. A statistically significant reduction in sweet aftertaste duration was observed. However, although there were no statistically significant differences in the measured intensities of the other attributes, it is interesting that the scores of the tannic acid containing samples for both sourness and bitterness were higher than those for sucralose alone, thus indicating that although tannic acid may indeed reduce the lingering sweetness of sucralose, it may introduce further sensory complications. In fact, it seems likely that tannic acid is reducing the sweet aftertaste of sucralose by taking advantage of the `mixture suppression' sensory phenomenon already discussed. The other high potency sweetener whose temporal profile has been a subject of some scientific study is neotame. Prakash et al. (2001) reported a number of formulation approaches to reducing the lingering sweet aftertaste of neotame, including the use of hydrophobic organic acids (such as cinnamate) or hydroxyamino acids (such as serine and tyrosine). In this study, samples were evaluated at room temperature and panellists trained on sucrose solutions such that a rating of 10 for maximum sweetness intensity equates to a 10% sucrose solution. Starting as soon as each solution was taken into the mouth, perceived sweetness was recorded by moving a cursor along a 15 cm line scale. A number of timeintensity parameters were extracted from the recordings, including maximum sweetness intensity, area under the curve, rising area under the curve, falling area under the curve, time to maximum sweetness and duration of sweetness. Cinnamic acid was found to reduce the duration of sweetness of neotame, as measured by the `duration' and `falling area under the curve' parameters (Table 5.1) and was shown to produce a statistically significant 17% reduction in sweetness duration. The authors conclude it is unlikely that these effects were as a consequence of acid Table 5.1 Temporal profile data for colas sweetened with neotame and various levels of cinnamic acid (Prakash et al., 2001) Parameter
Maximum intensity Area under curve Rising area under curve Falling area under curve Time to maximum intensity (sec) Duration (sec)
Control (Neotame)
Neotame + 10 ppm cinnamic acid
8.56 236 56 180 10.8 59.3
* Statistically significant difference from control (p < 0:05)
8.55 209 55 154 10.8 49.2*
Neotame + 25 ppm cinnamic acid 8.78 221 55 165 10.6 50.3*
92
Optimising sweet taste in foods
taste delivery from cinnamic acid since further work demonstrated that it is more likely the effects are due to cinnamic acid acting competitively at the receptor than by direct interaction with the sweetener in solution. Since cinnamic acid has also been shown to exhibit other taste modifying characteristics as a bitter taste inhibitor, it seems likely that the effects are due to general taste modification rather than sweetener±acid interactions.
5.4
Interactions of sweet and salty compounds
As has been noted, there are few important examples of foods and beverages that deliver both sweet and salt taste sensations; some savoury soups and sauces, notably those based on tomato, constitute some of the products that are both sweet and salty. Other examples can be found in populations that occasionally sprinkle salt on fruit. Thus, the interactions between these two taste sensations have not been widely examined, although the aforementioned mixture suppression effects have also been reported in mixtures of sucrose and sodium chloride (Bartoshuk, 1975). Other reports suggest that some salts at specific uselevels can cause enhancement of sweet taste (van der Heijden et al., 1983), with potassium acetate, potassium chloride and sodium chloride all claimed to enhance sweetness at specific concentrations (Table 5.2; Birch, 1999). Table 5.2
Sweetness enhancing salts (Birch, 1999)
Salt Potassium acetate Potassium chloride Sodium chloride Potassium carbonate Magnesium acetate
Dosage required for significant enhancement (mg/l) 700 450 300 450 700
It has been reported that the flavour quality of tomato products can be influenced by sweetness. Rosett et al. (1997), working in formulated tomato soups, demonstrated that salt taste intensity correlated positively with tomato flavour and that salt taste was suppressed in versions formulated using milk, concluding that salt taste was probably masked by the lactose-derived sweetness from the milk.
5.5 Interactions of sweet compounds with other sweet compounds Sweetness synergy is said to occur when a mixture of two sweeteners is perceived as being sweeter than would be expected based on the sweetness of the
Taste±ingredient interactions modulating sweetness Table 5.3 1996)
93
Synergistic and non-synergistic pairs of high potency sweeteners (Birch,
Sweetener pairs exhibiting strong synergy Sweetener pairs exhibiting little/no synergy Aspartame ± Acesulfame-K Aspartame ± Saccharin Cyclamate ± Aspartame Saccharin ± Cyclamate Acesulfame-K ± Cyclamate
Saccharin ± Acesulfame-K Sucralose ± Saccharin Sucralose ± Acesulfame-K Sucralose ± Aspartame Sucralose ± Cyclamate
individual sweeteners. The sweetness synergy sensory phenomenon is discussed elsewhere in this volume. However, some sweetener±sweetener interactions are also reviewed in this chapter; partly for completeness, but also because of the taste quality benefits that can also be a consequence of blending sweeteners. This issue, in particular, has important implications for food product development. The mechanism whereby some pairs of sweeteners deliver higher sweetness intensities than predicted based on their individual perceived intensities is not known with certainty (Birch, 1996). However, the mechanism of synergy is not particularly important in practical terms; what is important are the commercial benefits of lower sweetener concentrations and hence lower costs. The most obvious practical benefit of sweetness synergy today is that between aspartame and acesulfame-K resulting in substantially lower costs of sweetness than would be the case if using aspartame alone. Some synergistic pairs of sweeteners and non-synergistic pairs of sweeteners are listed in Table 5.3. Examination of the sweetener pairs in this table suggests that sweetness synergy may be little more than a consequence of release from the effects of mixture suppression. A sweetener such as aspartame that is effective at reducing the perception of bitterness synergises with those sweeteners that also elicit a clear bitter taste, such as saccharin and acesulfame-K. Aspartame may simply suppress the bitterness of saccharin or acesulfame-K, thus allowing intrinsic saccharin/ acesulfame-K sweetness previously masked by their inherent bitterness to be released and perceived. While this is a plausible explanation for synergy, what is of practical consequence is the potential that must therefore exist for improvements in taste quality to be a result of such interactions between sweeteners. These improvements may be further enhanced because of the inevitable consequence of using lower concentrations of individual sweeteners in blends than when each might be used as the sole sweetener. Thus, there is a reduced potential for the side-tastes or after-tastes that are an intrinsic part of the taste delivered by high potency sweeteners to be perceived. There has been little academic study of the impact of sweetener blending on taste quality, but examination of market practices demonstrates clearly that industry has adopted the principle with enthusiasm. Products containing three
94
Optimising sweet taste in foods
and four high potency sweeteners blended together are commonly seen on supermarket shelves. As noted by Shamil (1997), `if you blend many different sweeteners (sometimes up to six or seven), you really do create a very sugar-like taste'. In effect, the greater the number of sweeteners used in a blend, the closer will be the approach to the taste of sucrose.
5.6
Future trends
There is a dearth of quantitative information on the impact of different acids on the performance, both quantitative and qualitative, of different high potency sweeteners and their blends. It would be very helpful to the users of high potency sweeteners to have such information, particularly if the information was relevant to their products. There may be potential for ingredient cost saving through the identification of specific enhancements, and also for improving overall taste quality, possibly by optimising the temporal delivery of sweetness and acidity. Thus, it would be a valuable addition to the literature to complete fully comprehensive quantitative and qualitative studies of sweetener±acid interactions. The experience that has been gained in recent years on blending high potency sweeteners in foods and beverages indicates that this trend will almost certainly continue. All current mainstream high potency sweeteners are synthetic chemicals and so do not offer the marketing benefit of `natural' label declarations. Thus, there is no significant marketing disadvantage of blending sweeteners in products. With the development of a quality natural sweetener, that situation might change, but the immediate approval of such a sweetener is not anticipated. The practical observations are that blending leads to taste quality benefits, and although it would be interesting to see published confirmation of this, even without that endorsement, the practice will continue.
5.7
References
(1975) `Taste mixtures: is mixture suppression related to compression?', Physiol. Behav., 14, 643±649. BARTOSHUK, L.M., FAST, K., KARRER, T.A., MARINO, S, PRICE, R.A. and REED, D.A. (1992) `PROP supertasters and the perception of sweetness and bitterness', Chem. Senses, 17, 594±601. BARTOSHUK, L.M., DUFFY, V.B. and MILLER, I.J. (1994) `PTC/PROP tasting: anatomy, psychophysics and sex effects', Physiol. Behav., 56, 1165±1171. BIRCH, G.G. (1996) `Towards an improved understanding of sweetener synergy', Trends Fd Sci. Technol, 7, 403±407. BIRCH, G.G. (1999) `Modulation of sweet taste', BioFactors, 9, 73±80. CLONINGER, M.R. and BALDWIN, R.E. (1974) `L-Aspartyl-L-phenylalanine methyl ester (aspartame) as a sweetener', J. Fd Sci., 39, 347±351. EISENSTADT, M.E. (1972) `Cyclamate-free calorie-free sweetener', US Patent 3,647,483. BARTOSHUK, L.M.
Taste±ingredient interactions modulating sweetness
95
(1932) `The relation between chemical constitution and taste', Proc. Nat. Acad. Sci. USA, 18, 115±120. FRY, J.C. (1993) `Relative sweetness of intense sweeteners ± guidelines for the technologist', Food Technol. Europe, Sept./Nov., 114±116. KROEZE, J.H.A. and BARTOSHUK, L.M. (1985) `Bitterness suppression as revealed by splittongue taste stimulation in humans', Physiol. Behav., 35, 779±783. LAWLESS, H.T. (1979) `Evidence for neural inhibition in bittersweet taste mixtures', J. Comp. Physiol. Psychol., 93, 538±547. LAWLESS, H.T. (1982) `Paradoxical adaptation to taste mixtures', Physiol. Behav., 29, 149± 152. LINDLEY, M.G. (1999) `New developments in low-calorie sweeteners', in Corti, A., Low calorie sweeteners: Present and future, Basel: Karger, 44±51. LOTZ, A. and MEYER, E. (1994) `Sweeteners in beverages ± new developments', Internat. Fd Marketing and Technol., 8, 4±6, 8±9. MATTES, R. (1994) `Influences on acceptance of bitter foods and beverages', Physiol. Behav., 56, 1229±1236. MCBRIDE, R.L. and FINLAY, D.C. (1990) `Perceptual integration of tertiary taste mixtures', Percept. Psychophys., 48, 326±330. MELA, D.J. (1989) `Bitter taste intensity: the effect of tastant and thiourea taster status', Chem. Senses, 14, 131±135. ODAKE, S. (1996) `Sweetness intensity in carbonated water', in Ikan, R. and Cramer, B (eds), Abstracts of the IUPAC symposium on sweeteners, Jerusalem, Abstract 51. OSBERGER, T.F. and LINN, H.R. (1978) `Pure fructose and its applications in reduced calorie foods', in Dwivedi, B.K., Low Calorie and Special Dietary Foods, West Palm Beach: CRC Press, 115±123. PAMPIANO, C. (1980) `Means and method of improving the taste of saccharin sweetened food products', US Patent 4,219,579. PANGBORN, R.M. (1961) `Taste interrelationships II. Suprathreshold solutions of sucrose and citric acid', J. Fd Sci., 26, 648±655. PETTIGREW, S.J. and SILKA, L.A. (1988) `Non-saccharide sweetened product', European Patent 0140517. PRAKASH, I., BISHAY, I.E., DESAI, N. and WALTERS, D.E. (2001) `Modifying the temporal profile of high potency sweetener neotame', J. Agric. Fd Chem., 49, 786±789. PRESCOTT, J., RIPANDELLI, N. and WAKELING, I. (2001) `Binary taste mixture interactions in PROP non-tasters, medium-tasters and super-tasters', Chem. Senses, 26, 993± 1003. ROSETT, T.R., KENDREGAN, S.L. and KLEIN, B.P. (1997) `Fat, protein and mineral components of added ingredients affect flavor qualities of tomato soups', J. Fd Sci., 62, 190± 193. SCHIFFERSTEIN, H.N.J. and FRIJTERS, J.E.R. (1990) `Sensory integration in citric acid/sucrose mixtures', Chem. Senses, 15, 87±109. SCHIFFERSTEIN, H.N.J. and FRIJTERS, J.E.R. (1991) `The perception of the taste of potassium chloride, sodium chloride and quinine hydrochloride is not related to PROP sensitivity', Chem. Senses, 16, 303±317. SCHIFFMAN, S.S., REILLY, D.A. and CLARK, T.B. (1979) `Qualitative differences among sweeteners', Physiol. Behav., 23, 1±9. SHAMIL, S. (1997) `Sweetener blends in beverages' in Lisansky, S.G. and Corti, A. Low calorie sweeteners, Newbury: CPL Press, pp. 129±136. FOX, A.L.
96
Optimising sweet taste in foods
(1998) `Reduction of lingering sweet aftertaste of sucralose', World Patent Application WO 98/20753. VAN DER HEIJDEN, A., BRUSSEL, L.B.P., KOSMEIJER, J.G. and PEER, H.G. (1983) `Effects of salts on perceived sweetness', Lebensm. Unters Forsch, 176, 371±375 WONG, L.L., FAUST, S.M. and CHERUKURI, S.R. (1992) `Enhanced sweetness of acesulfame-K in edible compositions', US Patent 5,106,632. SHAMIL, S.H.
6 Measuring consumers' perceptions of sweet taste S. Issanchou and S. Nicklaus, INRA France
6.1
Introduction
Sensory evaluation defined as the `systematic study of human response to physico-chemical properties' makes it possible to obtain information about the sensitivity of the human sense and about the four dimensions of the sensory perception, i.e. the quantitative, the qualitative, the temporal and the hedonic dimensions. Methods used to study responses to sweetness are similar to those used to study responses to other stimuli, in particular to other chemical stimuli. They will be described in this chapter. Used in different research contexts, applications of these methods showed that consumers' perception of sweet taste is very variable. Variations might be linked to factors related to the consumer, such as experience, age or physiological state. They might also depend on product-related factors such as variations in other taste, in texture, odour and colour. An overview of such variations will be drawn.
6.2 Methods to determine consumers' perceptions of sweet taste A key point about sensory measurement should be underlined: many factors can affect the answers given by participants in a sensory test. These factors come from three sources: the individuals, the products and the experimental conditions. Concerning the individuals, the most important factors, which can explain variations in answers on the short term for a given person, are the motivation, the attention and the physiological status (for instance, hungry vs. satiated). Inter-individual variations might be related to ageing, to genetic equipment and
98
Optimising sweet taste in foods
to experience. Concerning the products, the most important factors are the medium in which the sweet compounds are tested and the temperature. Concerning the experimental conditions, the most important factors are the method, the instructions and the context of sample presentation (e.g. sample space, order of sample presentation). 6.2.1 Measurement of sweetness sensitivity Sensitivity is defined by ISO (1992) as `Ability to perceive, identify and/or differentiate, qualitatively and/or quantitatively, one or more stimuli by means of the sense organs'. Ability to perceive refers to the measurement of detection or absolute threshold; ability to differentiate stimuli quantitatively refers to the measurement of difference threshold. However, as pointed out by Stevens et al. (1995), `a threshold is a more or less transient and imperfect reflection of a person's sensitivity'. Two identical stimuli may lead to slightly different sensations. In other words, sensory systems operate in a probabilistic mode and not in an all or nothing mode; thus the probability to detect a stimulus or to differentiate two stimuli at different concentrations respectively increases when the concentration of the stimulus increases or when the difference of concentration increases. For estimating a detection threshold the most current methods are based on the two-alternative forced choice (2-AFC) or on the three-alternative forced choice (3-AFC). The subjects receive pairs (in 2-AFC) or triads (in 3-AFC) of samples; within each pair or triad, one sample contains the tastant in the chosen medium and the other one or two sample(s) only contain the medium. Subjects must decide which sample contains the tastant. If the 2-AFC and 3-AFC methods are the most widely used methods, other experimental parameters (such as the volume of solution to be tasted, the rinsing procedure, the concentration range, the concentration step scale, the sequence of presentation of the pairs or triads, the number of replicates) vary considerably among experiments. One of the most important differences lies on the sequence of presentation of the different pairs or triads and in parallel on the way to define the threshold. Two procedures are based on 3-AFC but differ on the number of 3-AFC tests performed by each subject. The simplest and fastest procedure for the determination and calculation of a taste threshold is the method described by ASTM (1991b). However, as pointed out by GonzaÂlez-VinÄas et al. (1998), this very simple method gives no more than a rough estimation of a threshold. Costell et al. (1994) compared it to another ASTM method (1991a) and to the method proposed by Lundahl et al. (1986) and concluded that the E1432 ASTM standard (1991a) is the best option to compute individual threshold with a maximum of precision. Moreover, Bi and Ennis (1998) suggested a new statistical approach to compute a population threshold which is not based on a measure of central tendency of individual thresholds but which takes into account the two sources of variation in the raw data, i.e. the inter-person and the intra-person variabilities.
Measuring consumers' perceptions of sweet taste
99
Recognition threshold is defined as the lowest concentration at which the taste is correctly identified. The determination of the recognition threshold is sometimes performed at the same time as the detection threshold. 6.2.2 Measurement of sweetness intensity First, before asking an individual about the perceived intensity of a sensory attribute, one should check that he/she is able to recognise this attribute. In the case of sweetness, this preliminary check is not always performed as sweetness is easy to identify, even for consumers who are not trained to perform sensory evaluation. Second, it is necessary to distinguish the tools used for recording individuals' responses about perceived intensity from the measurement scale. The former ones are the `responses scales' while the latter is defined as `the formal relationship (for instance ordinal, interval or ratio) between the intensity of a sensory perception and the numbers used to represent values of that perception' (ISO, 2004). One tool is the ranking method. In the simplest case, two samples are presented and assessors are asked to indicate which sample is the sweetest. This paired comparison test is a forced choice procedure. For data analysis, two cases must be distinguished: the one-sided test if the experimenter has an a priori hypothesis concerning the direction of the difference (e.g. when the experimenter wants to know if two concentrations of the same compound lead to significantly distinguished perceived intensities) and the two-sided test if the experimenter has no a priori hypothesis (e.g. when the experimenter wants to know which one of two solutions or foods prepared with different sweeteners is sweeter). The paired comparison test and ranking procedures in general have the advantage of being easy to understand. Nevertheless, these methods only permit one to compare samples presented in the same session; they give information only about the significance of the difference, but not about the magnitude of the difference or about the level of sweetness intensity of each sample. If the experimenter wants to collect such information and in particular when the purpose is to establish a psychophysical function for a given sweet compound (i.e. to establish the relationship between the perceived intensity and the concentration of the sweet compound; see Chapter 16), other methods are to be used. The two main options are the methods using response scales and the magnitude estimation. Concerning the first group of methods, two types of scales are commonly used: the line or continuous scales and the category or discrete scales. Line scales are generally anchored at each end with a verbal label: for example CalvinÄo and GarcõÂa (1998) used a 120 mm line scale anchored at the left end with `none' and at the right end with `extremely intense'. For category scale, some or all categories are associated with a verbal label. Line scales permit unlimited fineness of differentiation (ISO, 2004) but need more training to achieve a good level of repeatability than category scales. Moreover, when using a discrete scale with a label for each category, there is a greater risk
100
Optimising sweet taste in foods
to get a measurement scale where equality of intervals is not achieved, i.e. which size of the differences in perceived intensity between two adjacent categories is not the same for all adjacent categories. When using a category scale it is preferable to choose a scale with seven to nine categories and with a label only at the two extreme categories. Magnitude estimation is a scaling method with which assessors are asked to assign values to the intensity of a product's attribute in such a way that the ratio between the assigned values and the attribute perceptions are the same (ISO, 2000). Thus if the rating of sweetness for the sample X is 10, and if the assessor perceives the sample Y as being twice as sweet as the sample X, the assessor would give a rating of 20 to Y. There are variants of the method: the first sample is an external reference and each subsequent sample is evaluated in comparison to this reference, or the assessors are asked to evaluate each sample by comparing it to the immediately preceding sample. When an external reference is used, the assessors are free to assign it a number or this number referred to as a modulus is fixed by the experimenter. Magnitude estimation is less susceptible to `end-effects' than category or line scaling. Moreover, a measurement scale obtained from magnitude estimation data is theoretically a ratio scale. However, there is considerable controversy concerning the superiority of magnitude estimation to other scaling procedures (see Stone and Sidel, 2004). Magnitude estimation seems particularly useful in psychophysics for studying the relationship between the concentration of stimuli and the perceived intensities. Nevertheless, Laffort et al. (2002) observed greater dispersion of the results with magnitude estimation than with category scaling (see Chapter 16). As noted in the standard, `the magnitude estimation method is not the most efficient technique for determining small differences between stimuli or for conducting assessments in the vicinity of a detection threshold' (ISO, 2000). The labelled magnitude scale (LMS) has emerged recently (Green et al., 1993) as a potential alternative to magnitude estimation and to linear and category response scales. This is a semantic scale of perceived intensity characterized by a quasi-logarithmic spacing of its verbal labels (Fig. 6.1). LMS gave psychophysical functions similar to those obtained with magnitude estimation when the upper bound of the scale was defined as the `strongest imaginable oral sensation' (Green et al., 1993). However, when the upper bound was defined as the `strongest imaginable sweetness', LMS produced steeper psychophysical functions than magnitude estimation (Green et al., 1996). Bartoshuk and coworkers (Bartoshuk et al., 2001) generalised the LMS by labelling the upper bound of the scale `strongest imaginable sensation of any kind'. According to them (Bartoshuk et al., 2003, 2004), the gLMS (general labelled magnitude scale) would permit unbiased group or individual comparison contrarily to line or category scales. Indeed, for these authors, low-sensitivity individuals stretch the line or category rating scales while very sensitive individuals will compress it to fit the context of their individual experiences. In other words, a similar rating given by two individuals with very different sensitivities will not correspond to the same perceived intensity.
Measuring consumers' perceptions of sweet taste
Fig. 6.1
101
The labeled magnitude scale for intensity measurement (after Green et al., 1993).
Composite methods also exist. Other authors, working on new sweeteners used a 15 cm line scale but presented six sucrose references (2, 5, 7.5, 10, 12 and 16% w/v) to normalize intensity ratings (DuBois et al., 1991). These sucrose standards were allotted the values 2, 5, 7.5, 10, 12 and 15, respectively on the 0± 15 point scale (Portmann and Kilcast, 1996). As children do not necessarily have the same perception as adults, it can be useful to conduct sensory evaluation of children-targeted products with children and not with adults. The question is then `Among these different methods, which ones can be used efficiently by children?'. James et al. (2004, 1999) found that 8to 9-year-old children, after being given an appropriate training, are able to use a magnitude estimation procedure but have a tendency to limit the range of numbers used in their estimates of high concentration of sucrose for an orange drink but not for sucrose solutions, nor for custard and shortbread. The same authors compared the use of magnitude estimation and of a six-category scale (labelled `not sweet', `a tiny sweet', `a bit sweet', `moderately sweet', `very sweet' and `super sweet') between children (mean age 9 years) and adults: they found some differences in their psychophysical functions but concluded that children were able to use both scales even if they failed to distinguish between two successive sucrose concentrations out of the seven that were tasted (James et al., 2004). Zandstra and de Graaf (1998) used a five-point category scale (labelled `not at all sweet' and `very sweet') and also observed a flatter psychophysical function for children compared to adults; however they did not find any difference on the responses collected for the colour evaluation of grey surfaces on a five-point scale which anchors were labelled `white' and `black'. Since the results obtained for children and adults differed according to the stimuli, one can conclude that the difference between children and adults is not due to a response bias. This difference can be due to the fact that children may be more influenced by other components in some foods products (James et al., 1999). It is also possible that children referred to a different sample space because they do not consume the same products, and as
102
Optimising sweet taste in foods
human judgements are always comparative, the usual range of sweetness they consume induces different answers. In conclusion, different types of scale can be used with children, trying to keep the procedure as simple as possible, but a key point is to train them sufficiently and to make sure that they understand the task before starting measurement. Despite the variety of methods available, none of them can be considered as perfect. A context effect linked to the range of stimuli have been observed for category scaling (Riskey et al., 1979), for the line scale method (Conner et al., 1987), for magnitude estimation (Diamond and Lawless, 2001), and for category scaling, magnitude estimation and labelled magnitude scale (Lawless et al., 2000), as illustrated in Fig. 6.2. Such a context effect is a problem when one wants to compare results obtained at different sessions where the ranges of stimuli were different. Concerning the discrimination ability of the different scales, the results are diverse, but one can conclude that no scale presents a strong advantage over the other ones (see for example Jeon et al. (2004) for a comparison of category and line scales). Whatever the response scale, it is recommended to use Williams Latin square designs for the presentation order of the different samples to balance for order and first-order carry-over effects (MacFie et al., 1989). In conclusion, it is difficult to recommend one particular method for evaluating perceived sweetness. A general rule is to take into account the specific objective of the study and the age and training level of the assessors. However, due to reasons explained previously, some authors recommend using a
Fig. 6.2 Context effect on sweetness intensity rating with different scales (after Lawless et al., 2000). Mean rating of a 10% sucrose solution in a low sweetness context (presented after a 5% sucrose solution) or in a high sweetness context (presented after a 20% sucrose solution).
Measuring consumers' perceptions of sweet taste
103
gLMS scale when the objective is to compare individuals or groups of individuals, for instance subjects varying in their PROP-tasting abilities (see Chapter 2). However, to our knowledge the superiority of the gLMS scale over a line or category scale to reveal differences between such groups of individuals has never been demonstrated by comparing data collected with a gLMS scale and a line or category scale with the same individuals. 6.2.3 Measurement of the temporal dimension Perception of sweetness as well as of other flavour characteristics is not a static phenomenon but a dynamic one. Indeed, once a food is put into the mouth, there is a delay before perception occurs, then the perceived intensity increases until a maximum; this maximal intensity can last a certain time before it decreases. So, when assessors are asked to give a single rating for perceived sweetness, they must average or integrate their perception over time (Lee III and Pangborn, 1986). Time-intensity methodology makes it possible to record such temporal information about perception. This method has been more widely used since computerized systems became available. The most current approach to analyse the intensity curves obtained by asking assessors to continuously record their perception is to extract parameters from the individual curves obtained for each sample and to perform analyses of variance on this data set. The most currently used parameters are the maximum intensity, the time-to-maximum intensity, the total duration, the area under the curve, the rate of increase and the rate of decrease (for an extensive list, refer to Cliff and Heymann, 1993). An average curve is obtained for each sample by averaging all individual values at given times and by connecting these averages. However, all experimenters observed that each individual produces a curve with a specific shape, so methods were developed to obtain more meaningful average curves (Overbosch et al., 1986, Liu and MacFie, 1990). Time-intensity is a method which requires a computerized system, more training than a unique rating task, more complicated analyses but has been proved useful to reveal differences between different sweet compounds and particularly intense sweeteners. 6.2.4 Measurement of sweetness quality Sweetness is sometimes assumed to vary qualitatively according to sweet compounds, in particular in the case of sweeteners. Descriptive sensory method has been used to describe such variations in sweetness qualities. In the standardized approach, assessors first generated individually their descriptors which are then discussed in group session or individually (ISO, 1995). Then descriptors are defined, some references can be provided for helping the assessors at this stage, redundant terms are eliminated. Once the list is established assessors are asked to quantify the perceived intensity for each descriptor. Using this approach, Ayya and Lawless (1992) reported finding six
104
Optimising sweet taste in foods
attributes to describe the taste of sweeteners but quoted and reported results only for sweetness and bitterness. This descriptive approach was also used to characterize mixtures of bulk and intense sweeteners: besides sweetness and sweetness persistence, six other flavour attributes (acid, bitter, burnt sugar, caramel, liquorice, menthol), four mouthfeel attributes (body, cooling effect, drying, smoothness) and five after-taste attributes (astringent, irritant, metallic, liquorice, sweet persistence) were rated (Portmann and Kilcast, 1998). However such an approach based on the use of words has some limits especially to describe sweetness quality. Indeed, in English, as well as in other European languages, there are four common words to describe tastes (to which umami or savoury might be added). It is thus difficult for an assessor to find other words to describe adequately and completely the taste of different sweeteners. It also requires selecting individuals with good abilities to describe their sensations and an important training phase to get a consensual list for the panel beyond the four common taste descriptors. Thus, to avoid problems linked to the use of words, one can use dissimilarity scaling. In this method, samples are presented in pairs; assessors are asked to rate the dissimilarity of the quality of the two samples on a line scale anchored at the left hand by `exactly the same' and at the right hand by an expression such as `completely different'. In order that differences reflect only differences in quality between the studied sweeteners, they are presented at concentrations leading to equi-intense perceptions. Multidimensional scaling is then applied to the matrix of mean distances or to the individual matrices to get a multidimensional space (Thomson and Tunaley, 1987). 6.2.5 Measurement of the hedonic dimension Any change in a product formulation, and especially any change in sweetener composition can modify sweetness perception and consequently liking. The most used approaches are based on measurement methods which are similar to the ones used for measuring sweetness intensity, presented earlier in this chapter. Instead of being asked to compare or rate the sweetness intensity, consumers are asked to compare or rate the affective response elicited by the products. Global liking can be assessed, but consumers can also be required to score how far each sample is from their ideal sweetness level. Besides these methods where overt questions are asked, behavioural measurements, such as intake measurement, are also used, assuming that the higher the liking, the higher the intake. Evaluation of global liking As for the evaluation of the quantitative dimension a paired comparison test can be used to evaluate preference, but in this case the consumer is asked `Which sample do you prefer?'. A two-tailed binomial test is used because the experimenter has no a priori concerning the direction of the preference. A paired preference questionnaire often includes a `No preference' option. Adding this
Measuring consumers' perceptions of sweet taste
105
response option was proposed because if there is no significant preference for one sample, the experimenter does not know if half of the panel prefers one sample and the other half prefers the other sample or if the whole panel has no preference. However, when using a questionnaire with a `No preference' option, it is not possible to handle the data using a binomial statistical analysis any more. Simply eliminating the `No preference' answers in order to use a binomial statistical analysis is not a proper way of doing! As recently demonstrated (Angulo and O'Mahony, 2005), this can lead to a distortion of the pattern of preferences. So, in conclusion, we recommend not to add a `No preference' option in the questionnaire or, if this option is used, not to eliminate the answers corresponding to this choice and to use multinomial statistics to analyse data. If several product alternatives must be compared, multiple paired tests can be used but this method is time- and sample-consuming and can become demanding for the consumers. The solution is then to use balanced incomplete block designs (e.g. AFNOR, 2000; Meilgaard et al., 1999). Different approaches have been proposed to analyse data collected with multiple paired comparisons. If the experimenter wants a global result over the whole panel, the method consists in converting raw data in a score or in a scale value and then in analysing the significance of the difference between these values (see for example, AFNOR (2000); Best and Rayner (2000) and Meilgaard et al. (1999)). However, such approaches assumed that preferences are transitive whereas it can happen that A is preferred to B, B is preferred to C, but C is preferred to A. An explanation is that when comparing samples A and B, the consumer focused on one sensory attribute and on another one when comparing samples A and C (KoÈster, 1998). Such non-transitivity raises problem for applying the statistical analysis described above (Gabrielsen, 2000). If the experimenter is interested in individual differences, methods have been developed permitting segmentation of the panel (Qannari et al., 2000). The paired preference test is very appealing since comparing two samples seems to be an easy task. Moreover, comparison methods are usually highly recommended since people have a better ability to compare stimuli than to give absolute judgements. Nevertheless, paired preference tests can lead to spurious results (KoÈster, 1981). Indeed, if for instance a consumer is asked to indicate his preference between two beers, one of which is sweeter than the other, he will most likely choose the sweetest. If one repeats this test using the highest sweetness of the previous trial as the lowest one in the next trial, the consumer might again choose the sweetest beer. After several steps, one can lead the consumer to a preference for an extremely sweet beer. But, if after the test one presents, under natural conditions (for instance, at a bar), the consumer with a beer of the sweetness level he has just preferred, he will very often reject it as being appallingly sweet. KoÈster (1981) argued that this comparison test is unnatural because the separate judgement in each paired comparison is made against a very limited background. The two pair members are only compared against each other. In such conditions, the normal everyday internal standards of the consumer are not taken into account when he/she makes each decision. So,
106
Optimising sweet taste in foods
presenting successive pairs in a paired preference test in a series of slowly ascending sweetness must be avoided. Moreover, a paired preference test permits comparison but if the experimenter wants information on the degree of liking, response scales must be used. Different response scales have been proposed. The most common one is the nine-point hedonic scale (see Fig. 6.3a) developed in the United States (Jones et al., 1955; Peryam and Pilgrim, 1957). The most important criticism concerning this response scale is the non-equality of the intervals. This nonequality was demonstrated (McEwan, 1988; Jones et al., 1955): for instance, the distance between categories 8 and 7, i.e. `like very much' and `like moderately' and between categories 3 and 2, i.e. `dislike moderately' and `dislike very much' are considerably larger than those between categories 3, 4, 5, 6 and 7. Moreover, when this scale is translated to other languages, the distances between intervals might have different values. To avoid this problem of interval inequality, it is recommended to use a continuous scale as the one presented in Fig. 6.3b. As computerized systems for data collection are more and more often used, it is no longer a time consuming task to convert marks on the scales into scores. With children, category scales might appear easier to use, in which case they are often accompanied by expressive faces to help children understanding the orientation of the scale (Meilgaard et al., 1999; LeÂon et al., 1999). As observed for ratings of sweetness intensity, the effect of the stimulus context has been demonstrated for ratings of pleasantness of soft drinks containing different sucrose concentrations (Riskey et al., 1979) as shown in Fig. 6.4: lower concentrations were judged more pleasant in a `positively skewed' context, i.e. when lower concentrations were presented more frequently.
Fig. 6.3
Examples of response scales for hedonic measurement.
Measuring consumers' perceptions of sweet taste
107
Fig. 6.4 Context effect on liking rating for sucrose solutions (after Riskey et al., 1979). In the `skew' context, very sweet solutions are more frequently presented than in the `normal' and in the `ÿskew' context.
Concerning the presentation order of the different samples, it is recommended, as for measuring intensity with response scales, to use Williams Latin square designs. Moreover, in the case of hedonic measurements it is also recommended to present a `dummy' sample in the first position to eliminate the first position bias, commonly observed in hedonic judgements (MacFie et al., 1996). For data analysis, if the samples only differ on sweetener concentration and if this difference in sample composition induces modification of the perceived sweetness intensity only, it is possible to look at the relationship between hedonic scores and sweetener concentration (for each consumer, for groups of consumers ± e.g. for comparing young and elderly subjects ± or over the whole panel). Moreover, if consumers rated both their liking and their perceived sweetness intensity (in which case it is recommended to ask these two ratings with different sample presentations, the first one being devoted to the hedonic ratings), it is possible to examine the concentration±intensity relationship (psychophysical function) and the intensity±liking relationship (psychohedonic function). It is thus possible to determine if inter-individual differences in the concentration± liking relationship are caused by differences in the psychophysical function and/ or by differences in the psychohedonic function (see Section 6.3). However, one must be conscious that even if formulations only differ on sweetener concentration, samples might be perceived as different not only on sweetness but also on other sensory characteristics due to interactions (for example a yoghurt with less sucrose can be perceived as less thick, more acid, less intense in aroma). In this case and if the experimenter supposes or wants to take into account the interindividual differences in patterns of hedonic responses and if a sufficient number of samples have been tested, multidimensional methods must be used. Internal preference mapping permits one to achieve a representation of the samples and of
108
Optimising sweet taste in foods
the consumers in the same multidimensional space (Greenhoff and MacFie, 1994). If such a graphic of the preference space can be useful to look at interindividual differences, it may be difficult to read and loses its interest when the number of consumers is large and, especially when the number of dimensions to be interpreted (i.e. representing more than half of the total information) is higher than 3 (Schlich, 1995; Vigneau et al., 2001). Thus, cluster analysis which permits to achieve a segmentation of the consumer panel into several homogeneous groups of consumers is a useful tool. Evaluation of ideal sweetness For sweetness, as for many attributes, the shape of the relationship between liking and concentration or perceived intensity is supposed to be an inverted-U or inverted-V shape if `the units of the acceptance-determining factor are chosen to be equally discriminable and the consumer test has been correctly designed to avoid biases' (Booth and Conner, 1990). However, in the case of sweetness, different authors observed different shapes which are presented in Fig. 6.5.
Fig. 6.5 Schematic representation of the four types of relationships observed between sucrose concentration and hedonic responses.
Measuring consumers' perceptions of sweet taste
109
Just-about right (JAR) response scales (see Fig. 6.6 for example) are commonly used to determine the optimum level for a given sensory attribute in a product. Since sweetness intensity is quite easy to evaluate by a consumer without any sensory training, and since sweetness may often be related to the concentration of one ingredient, JAR scales are particularly used for determining ideal levels of sweetness. Generally, the samples are the same for all consumers and the sample presentation order is pre-determined and is either randomized or follows a Williams Latin square design. The relative-to-ideal procedure developed by Booth et al. (1983) seems particularly relevant to determine unbiased individuals' most preferred level of an ingredient in a product. The stimulus range bias seems to be avoided by selecting stimuli so that the mean response score is close to the mid-point (the ideal) of the scale (Conner et al., 1987). In this procedure, the first sample corresponds to the middle of the tested range of the ingredient, the second sample is chosen depending on the consumer's response obtained for the first sample in order that it would be rated on the other side of the consumer's ideal point. The subsequent samples are selected in order
Fig. 6.5 (Continued)
110
Optimising sweet taste in foods
Fig. 6.6
Examples of just-about-right response scales for measurement of ideal sweetness level.
to get an alternation of responses on either side of ideal while extending the range towards each extreme. However, the experimenter must avoid samples that could elicit a rating at either end of the response scale in order to minimise end effects. A linear regression of relative-to-ideal ratings (distances of each response mark from the ideal) on the logarithm of the ingredient concentration is then calculated for each consumer (see Fig. 6.7) and the estimated slope can be related to the tolerance of deviations from ideal (the higher the slope, the lower the tolerance). Several authors wondered what was the best method to predict actual consumers' optimal sweetness and/or intake. Consequently, the ideal value obtained with one of the previously described methods was compared to the sweetener content either of the marketed products or of the sample eaten in the largest quantity in a consumption test where at each session, a sample at a given sweetener concentration was presented ad libitum. Thus, Bellisle and Lucas (1987) found for different sucrose levels in yoghurt a much higher mean ideal
Fig. 6.7 Linear regression of the ideal-relative sweetness on the logarithm of sucrose concentration in yoghurt (after Daillant and Issanchou, 1991). Concentration of sucrose in g added in 100 g yoghurt. The numbers close to the experimental points indicate the order of presentation. The estimated ideal level is 9.2 g in 100 g yoghurt.
Measuring consumers' perceptions of sweet taste
111
value (15.3%) with the hedonic scale (with a sip and spit procedure) than with the intake data (8.2%). Working also on plain yoghurt, Daillant and Issanchou (1991) found a lower discrepancy and in the opposite direction between relativeto-ideal ratings and a consumption test: 6.3 and 7.5%, respectively. Epler et al. (1998) noted that several authors (e.g. Vickers, 1988) found a lower mean optimal level of sweetness determined with JAR scales than in the marketed products. They proposed that JAR scales could give an underestimated value of the actual optimal sweetness. However, it is also possible that the sweetness is too high in many marketed products, at least for a large proportion of consumers. Comparing JAR scales (a continuous one and a category one) to hedonic ratings, Epler et al. (1998) found a lower sucrose content with the JAR scales than with the hedonic scale: 9.3 and 10.3% w/v of sucrose in lemonade respectively. These authors also found that in a paired preference test, 112 out of 195 consumers preferred the 10.3% sucrose lemonade and thus concluded that the hedonic scale would better predict the optimal sweetness than the JAR scale. However, this conclusion might be misleading considering the limitation of the paired preference test presented earlier. As pointed out by Shepherd et al. (1985) the discrepancy between the relative-to-ideal scale and the hedonic scale may be due to the fact that these scales do not exactly measure the same thing. Although consumers were asked in all studies but one (Bellisle and Lucas, 1987) to rate their liking for the sweetness, some consumers may have rated their overall liking while actually focused more on sweetness with the JAR scale. Since modifying the concentration of the sweet compound may affect sensory characteristics other than sweetness, it is possible that the sample at the optimal sweetness level could be too acid for instance and thus could not get the highest liking score. Moreover, such a discrepancy between JAR and hedonic scales may occur for sweetness since this sensory characteristic is related to an ingredient which is considered by many consumers to be `bad' for health (Epler et al., 1998). Bower and Boyd (2003) also observed a lower optimal sweetness in lemon juice with the JAR scale (8.8% w/v) than with the hedonic scale (10.3% w/v); however, they did not find any significance difference according to consumer's health concern and consumption patterns. Further comparisons of methods used to determine ideal sweetness, including characterization of consumer's psychological profile, would help to improve methods for sweet food development. Presently, it is clear that the JAR scale is leading to lower optimal sweetness levels compared to other methods. As modification of sweetener level can also affect other sensory characteristics we would recommend for sweet food development to use a hedonic scale and to ask consumers to rate their overall liking and not to focus their attention on sweetness. Thus, all sensory modifications induced by a modification in the sweetener level and important for a consumer hedonic appreciation would be taken into account. This recommendation is also based on previous results (Popper et al., 2004; Earthy et al., 1997) which demonstrated that asking questions on specific sensory attributes (with JAR or hedonic scales) had a significant effect on overall liking ratings.
112
Optimising sweet taste in foods
6.3
Overview of consumers' perception of sweetness
Our purpose is to highlight specific issues rather than to provide an exhaustive review of variations in consumers' perception of sweetness. Some aspects will be developed in other chapters of this volume. In this section, we will present factors of variations related to the consumers and factors related to the food. 6.3.1 Variations in perception of sweetness according to factors related to the consumers Inter- or intra-individual variations can be observed at different levels. Sweetness perception can be affected at the detection or recognition level or at a suprathreshold level. In this last case, the psychophysical function can be affected in different ways as illustrated in Fig. 6.8 (left section). In the case of intense sweeteners which can present sensory properties other than sweetness, the qualitative perception can be affected especially by a modification of the detection threshold. These differences can induce different hedonic answers: the relationship between liking scores and sweeteners concentrations will be affected as shown in Fig. 6.8 (right section). However, a modification of the relationship between liking scores and sweetener concentrations can also be due to a modification of the psychohedonic function. The mechanisms responsible for such changes are not clear. Only when both intensity and liking are measured, is it possible to determine if inter-individual differences in the concentration±liking relationship are caused by differences in the psychophysical function and/or by differences in the psychohedonic function (see for example de Graaf et al., 1996). Consumer's perception of sweetness is likely to vary according to a number of individual factors. Recent studies have shown in particular that genetic variability of the TAS2R38 receptor, linked to the perception of the bitterness of PTC, is also related to preference for sucrose and for sweet beverages and cereals in children (e.g. Mennella et al., 2005): this aspect is developed in Chapter 2. Though sweetness perception and preference might have a heritable component, they are also likely to vary according to experience, through culture, family habits or individual history: this will be briefly overviewed here. The development of individual sweet perception in infancy and childhood is developed in Chapter 3. We will describe evolution of sweetness perception with age; as well as potential variations according to physiological and psychological status. Our presentation is organized according to the different factors involved in variations of sweetness perception; however, it is not always easy to tear apart their respective effects. For instance, a higher desire for sweet taste (Schiffman et al., 2000a) and preference for sweeter solutions were shown in AfricanAmerican infants (Beauchamp and Moran, 1984), adolescents and adults (Bacon et al., 1994; Greene et al., 1975; Pepino and Mennella, 2005), compared to Caucasians. It is difficult to attribute this difference to genetic inheritance or to experience, since this cultural subgroup is known for consuming more sugar than Caucasians. In most studies quoted further, perception of sweetness is
Fig. 6.8
Schematic examples of different psychophysical functions and impact on the relationships between liking and concentration in case of identical psychohedonic function (after de Graaf et al., 1996).
114
Optimising sweet taste in foods
assessed by using a model food (or beverage) which variants are sweetened at different levels. The impact of experience Culture-related experience Cross-cultural studies show no (Druz and Baldwin, 1982; Lundgren et al., 1976; Laing et al., 1993) or few differences (Ishii et al., 1992) in thresholds for sucrose in different cultures; and similarly, no (Prescott et al., 1997; Laing et al., 1994; Bertino and Chan, 1986) or few differences (Bertino et al., 1983; Holt et al., 2000) in perceived sweetness of sweet solutions or foods in different cultures. Results concerning liking for sucrose solutions across cultures are contrasted. Liking for sucrose solutions does not differ between Australian and Japanese subjects (Prescott et al., 1992). In contrast, compared to American students, Asian students living in United States find sucrose solutions more pleasant: it was shown for Taiwanese (Bertino et al., 1983) and Chinese students (Bertino and Chan, 1986). However, in both cases, the breakpoint of the inverted-V shape was not different in both groups. In a comparison of Australian and Malaysian students, similar percentages of each group were either sweetness likers or dislikers; but Malaysians rated pleasantness of high sucrose solutions higher than Australians (Holt et al., 2000). Cultural differences in the preferred level of sweetness are more apparent in actual food products. Thus, Lundgren et al. (1978) observed that in coffee, the most frequent patterns of hedonic responses over different sucrose concentrations (as defined in Fig. 6.5) differ according to the country. Several studies compared populations of various ethnic origins living in the United States. Nigerians and Koreans preferred a sweetened version of a tomato juice contrary to Americans who preferred the unsweetened version; and Nigerians preferred a sweetened version of an apple sauce compared to Americans who preferred it plain, while Koreans liked all versions of this food (Druz and Baldwin, 1982). Taiwanese assigned higher pleasantness rating to cookies with low sucrose concentrations compared to Americans who preferred the more sweetened cookies (Bertino et al., 1983). Chinese living in the United States preferred highly sweetened cookies compared to Caucasian Americans but the difference in their breakpoint concentration was not significant (Bertino and Chan, 1986). Other studies compared Australian to Asian consumers. Prescott and co-workers (1997) observed a good agreement between Australian and Japanese regarding the optimum level of sucrose for each tested food. Laing and collaborators (1994) also observed similar responses for Australian and Japanese with jams and fruit juices but considerable disagreement with regard to sweetness liking in beverages, biscuits and chocolates. In these cases, liking was rated higher by Australian for products from the Australian market and conversely Japanese gave higher liking scores to products from the Japanese market. In another study, differences in sweetness liking were observed between Australian and Malaysian students but were food-specific, however no difference in breakpoint concentration appear for any of the tested foods (Holt et al., 2000). In the
Measuring consumers' perceptions of sweet taste
115
previous studies, as noted by the authors, there was evidence that familiarity with sweetness level in habitual foods influenced responses. In addition, one might underline that within each culture, the preferred sweetness is food-specific (Prescott et al., 1997; Bertino et al., 1983). This underlines the importance of the experience with the sweetness of a specific food in the development of sweet preference. The cultural differences might reflect difference in food availability across nations, or different habits of sugar consumption in specific foods. Individual experience That experience strongly determines preference for sweet taste was revealed with infants. An infant who receives sweetened water during the first six months of life will maintain the preference for sugar exhibited at birth, whereas this preference is not maintained in an infant not exposed to sweetened water. Such an effect is still apparent at the age of 24 months (Beauchamp and Moran, 1982, 1984). Four- to five-year-old children, exposed 15 times to a bland food (tofu), either sweetened or salted, prefer the version to which they were exposed even a few weeks after the exposure phase (Sullivan and Birch, 1990). The impact of taste experience in early childhood is developed in Chapter 3. The assumption that higher dietary sugar intake would reflect in higher preferred levels of sugar was the starting point for studies on the role of experience. The higher consumption of sweet foods and beverages by children and adolescents was indeed explained by their preference for sweeter solutions (Drewnowski, 1989). In healthy adults, some studies confirmed the link between sugar intake and preference for sweet foods: preferred levels of sucrose in lemonade (Pangborn and Giovanni, 1984), in oatmeal (Mattes and Mela, 1986), in tea (Jamel et al., 1996), in lime drink (Conner and Booth, 1988) and in various foods (Holt et al., 2000), were related to dietary carbohydrate or sugar intake. Habitual consumption of soft drinks was related to preferred sugar level in soft drinks (Tuorila-Ollikainen and MahlamaÈki-Kultanen, 1985). However, other studies did not confirm the relationship between usual intake of sugar and preference for sweet solutions (Bertino and Chan, 1986) or sweet foods (Stone and Pangborn, 1990; Mattes, 1985), or between discretionary usage of sugar and preference for sweet solutions or foods (Maller et al., 1982). Overall, it seems that experience does not generally influence sweetness perception but it might well influence preference for sweetness, in a foodspecific way. The impact of age Childhood and adolescence The sweet taste is universally liked and looked for, but perception of and preference for sweetness vary with age. It is not known to what extent young subjects' sensitivity to sugar differs from that of adults, because in general different methods have to be used and therefore direct comparisons between young subjects and adults are difficult to perform. Newborns exhibit positive
116
Optimising sweet taste in foods
facial reactions when presented with sucrose solutions (Steiner, 1979; Rosenstein and Oster, 1988), preferentially ingest sweet solutions compared to water (Beauchamp and Moran, 1982), and are able to distinguish different sucrose concentrations (Crook, 1978) as do 2 to 3-year-old children, who ingest more of a sweet solution when its concentration increases (Vasquez et al., 1982). This positive affective reaction to sweet taste might be related to its analgesic properties (Blass and Hoffmeyer, 1991). Preference for very sweet solutions continues during childhood and adolescence (Desor et al., 1975), however, by early adulthood, preferred sugar levels lower off as shown by cross-sectional (Grinker et al., 1976) or longitudinal studies (Desor and Beauchamp, 1987): the result of the former study is illustrated in Fig. 6.9. Preferred sucrose levels in foods are generally higher in children than in adolescents, and higher in adolescents than in adults (de Graaf and Zandstra, 1999; Monneuse et al., 1991; Zandstra and de Graaf, 1998). Whether this difference in preference is related to difference in perception is not clear. Perceived sweetness curves in function of sucrose concentration in water are similar between children and adults in most studies (James et al., 1999, 2004; de Graaf and Zandstra, 1999) but they are steeper in children than in adults in one study (Enns et al., 1979). In solid foods, psychophysical functions are similar between 8±9-year-old children and adults for custards and biscuits (James et al., 1999); however children's functions are flatter in orangeade (de Graaf and Zandstra, 1999; James et al., 1999; Zandstra and de Graaf, 1998) as shown in Fig. 6.10. Thus, compared to those of adults, children's perception of sweetness is more likely to be influenced by food context, in particular by taste context. Shortcomings in young children's ability to integrate sensory information might explain the difference in their perception compared to adults; rather than pure physiological differences at the peripheral level (Oram et al., 2001; Stein et al., 1994).
Fig. 6.9 Evolution of preference for sweetness with age: percentage of subjects preferring each sucrose concentration when they were 11±15 years old and when they were 19±25 years old (after Desor and Beauchamp, 1987).
Measuring consumers' perceptions of sweet taste
117
Fig. 6.10 Mean sweetness intensity and pleasantness of orange beverages varying in sucrose concentration by children (6±12), adolescents (13±18), young adults (19±34) and elderly subjects (65+) (adapted after Zandstra and de Graaf, 1998).
Ageing At the other extremity of life, changes concerning the quantitative and the qualitative dimensions of taste perception are documented, starting at the age of 40 or 50 years; they might result from ageing of the taste system, or from indirect impact of medicine use (Doty and Bromley, 2004). Sucrose and aspartame thresholds were found to be either higher in elderly subjects than in younger adults or not different in both populations (see Mojet et al., 2001 for a review). Overall, thresholds for sweet compounds slightly increase with age; this slight increase is not specific to this taste quality and is probably more pronounced in the case of salt or of bitter compounds (Mojet et al., 2001; Stevens, 1996). Recognition thresholds are also affected by ageing, in particular in the case of sucrose and saccharin, but not for aspartame (EasterbySmith et al., 1994). Not mutually exclusive hypotheses concerning the origin of the threshold increase have been proposed. The problem could take place at a peripheral level, at a neural level and at a psychological level (see Mojet et al., 2001 for discussion). Perceived sweetness of water solutions is generally similar in elderly and young adults (Enns et al., 1979; Bartoshuk, 1989; Murphy and Gilmore, 1989; Hyde and Feller, 1981; Weiffenbach et al., 1986, 1990; Cowart, 1989; Stevens and Lawless, 1981). However, some studies showed that although seniors are as able as younger adults to discriminate sucrose concentration levels in foods or in solutions (Zandstra and de Graaf, 1998; de Graaf et al., 1996; Mojet et al., 2003; Gilmore and Murphy, 1989; Warwick and Schiffman, 1990), they produce flatter psychophysical slopes than younger adults in water solutions (Mojet et al., 2003; Schiffman et al., 1981; Bartoshuk et al., 1986; Bartoshuk,
118
Optimising sweet taste in foods
1989) and in food products (Zandstra and de Graaf, 1998; de Graaf et al., 1994, 1996). Figure 6.10 illustrates this effect. However, the difference between the sweetness perception of seniors and of younger adults is smaller when it is tested in actual food products than when it is tested in water (Mojet et al., 2003), so the impact of ageing on taste evaluation might have been overestimated by the numerous studies conducted with aqueous solutions. In addition, the extent of qualitative interaction of sweetness with other tastes is either similar in young and elderly (Murphy and Gilmore, 1989; Pelletier et al., 2004) or more pronounced in elderly subjects (Stevens and Cain, 1993; Stevens, 1996; Mojet et al., 2004). Elderly subjects' preference for sweet solutions is either similar to that of younger adults (Enns et al., 1979) or oriented toward higher sugar concentration (Murphy and Withee, 1986). When actual foods were rated, most studies showed that elderly subjects prefer sweeter variants of a given food (Zandstra and de Graaf, 1998; de Graaf et al., 1994, 1996; Murphy and Withee, 1986; de Jong et al., 1996; Barylko-Pikielna et al., 2002); only one study showed that JAR assessment of yoghurts' sweetness does not differ in elderly and young subjects (Koskinen et al., 2003). In the absence of longitudinal data, it is difficult to conclude whether this higher preference for sweet foods in seniors is linked to physiological consequences of ageing, to psychological evolution with age or to a generational trend. The impact of gender Perceived sweetness is the same in both genders in general (James et al., 1999; Lundgren et al., 1978; Holt et al., 2000; Mojet et al., 2003); only two studies showed that females perceive higher sweetness in lime drinks (Laeng et al., 1993) and in custards and biscuits (Holt et al., 2000). Nevertheless, preference for sweet foods (sweeter variants) seems to be higher in males (Conner and Booth, 1988; Enns et al., 1979; Monneuse et al., 1991; Desor et al., 1975; Looy and Weingarten, 1991) or at least equal in both genders (Salbe et al., 2004). Alliesthesia decreases sweet liking more strongly in female than in male subjects (Laeng et al., 1993). Since perception of sweetness is equal in both genders, the higher liking for sweet foods in males is more likely to be related to a greater disinhibition (LaÈhteenmaÈki and Tuorila, 1994). Sweetness liking does not seem to depend on hormonal variations: menstrual cycle has no effect on preferences for sweet solutions (Pomerleau et al., 1991); and during pregnancy, no change in sweetness perception is observed (Duffy et al., 1998b), though increased preference for sweet foods is noticed (Duffy et al., 1998a). The impact of physiological factors Salivary flow Saliva may affect taste sensitivity in different ways such as through diffusion of taste compounds, chemical interaction with taste compounds, stimulation of taste receptors and protection of taste receptors (Matsuo, 2000). However, while
Measuring consumers' perceptions of sweet taste
119
saliva flow varies greatly between subjects, its impact on sweetness perception is either absent (Bonnans and Noble, 1995) or limited (Guinard et al., 1997). Drugs Many drugs are likely to modify taste, in particular sweet taste perception (Schiffman et al., 1998; Doty and Bromley, 2004): they generally diminish the intensity of perceptions. The most common abnormality in taste perception related to a drug treatment is a decreased ability to detect or recognize stimuli (Henkin, 1994). According to this author, sweetness perception is less commonly affected by drugs than other tastes. Nevertheless, many drugs can induce taste, as well as olfactory impairment and depending on the drug, the origin of the impairment might occur at different levels: at the central nervous system level, at the neural transmission level but mainly at the receptor level (receptor pathology per se or receptor-related mechanisms). Moreover, opioid antagonists have been shown to affect hedonic perception. For example, after naltrexone administration a significant, but slight, decrease in liking for sucrose solutions occurred while this drug did not alter sucrose detection and recognition thresholds (Arbisi et al., 1999). Weight status Though preference for sweet (and fat) foods was considered to potentially explain differences in corpulence, weight status does not seem to be related to sweetness perception (Frijters and Rasmussen-Conrad, 1982; Thompson et al., 1976; Drewnowski and Holden-Wiltse, 1992), neither to preference for sweet foods (Salbe et al., 2004; Conner and Booth, 1988; Frijters and Rasmussen-Conrad, 1982; Thompson et al., 1976; Drewnowski and Holden-Wiltse, 1992). Only a few studies showed a link between weight and sweet liking: BMI was inversely related to hedonic ratings for sweet solutions (Enns et al., 1979; Drewnowski et al., 1985), and was associated with consumption of high-sugar foods particularly in obese women (Macdiarmid et al., 1998; Drewnowski et al., 1992). Sweet preferences are higher in obese subjects having stronger fluctuations in body weight (Drewnowski and Holden-Wiltse, 1992). Interestingly, a longitudinal study revealed that subjects who prefer sugar and fat at the beginning of a follow-up were more likely to have gained weight ten years later (Salbe et al., 2004). Satiety Recognition thresholds for sucrose were significantly higher after a meal than during fasting state (Zverev, 2004). Detection thresholds for glucose were also found to decrease when subjects were satiated (Crovetti et al., 1997). These authors found that the higher the energy intake, the higher the threshold. However, they also observed that the increase in threshold is more important after eating meat balls than after eating baked macaroni, despite similar energy intake with both foods. This difference was attributed to the different satiating power of those foods: subjects indeed rated their fullness and satiety more intense after eating meat balls than after eating macaroni.
120
Optimising sweet taste in foods
Satiety was considered to have a potential impact on liking for sweet taste (Cabanac, 1971, 1979). However, results from experimental studies are not convincing (Moskowitz et al., 1976). Satiety leads to a decrease of sweet liking in female but not in male and in subjects with sweet tooth (see further) contrary to subjects less reactive to sweetness (Laeng et al., 1993). The opposite result was also shown: decrease in sweet dislike in sweet `dislikers' when fooddeprived compared to sated; and no change of liking in sweet likers (Looy and Weingarten, 1991). The impact of psychological factors `Sweet tooth' was defined as the liking for a higher level of sweetness in foods and a greater preference for sweet foods relative to non-sweet foods (Conner et al., 1988). According to this definition, and given results exposed above, sweet tooth is a feature of childhood and adolescence. Sweet tooth does not seem to be related to personality traits (Stone and Pangborn, 1990). However, among dieters, dimension of cognitive restraint, disinhibition (tendency to eat in response to an emotion or to an external signal such as the presence of food or other people who are eating), and hunger are related to the use and liking of sweet foods (LaÈhteenmaÈki and Tuorila, 1995). Positive attitudes toward sweet foods might explain sweet food usage and also liking for sweet foods in a lab setting (LaÈhteenmaÈki and Tuorila, 1994). 6.3.2 Variations in perception of sweetness according to factors related to the food Not only might perception of sweetness vary according to factors related to consumers, but sweetness is also likely to vary according to a number of parameters related to the food itself. Firstly, mixtures of sweet compounds (in particular with high potency sweeteners) often present enhanced sweetness compared to unmixed compounds, because of synergistic effects (e.g. Schiffman et al., 2000c). Such interactions are described in details in Chapters 11 and 16. Second, sweetness, like other tastes, can be modulated by the other characteristics from the same sensory modality, i.e. by other tastes (see Chapter 17), and from other sensory modalities such as odours (its influences are developed in Chapter 4), texture and colour. Not all interactions take place at the same level. Depending on the product composition, and in particular on the composition in proteins, polysaccharides and lipids, the quantity of released flavour compounds and the time course of this release may differ. The composition also affects the texture but in these cases, the interactions are clearly physical and chemical interactions (see for example Guichard et al., 2003). Nevertheless, texture-taste interactions are not always related to physical or chemical interactions but can occur at the perceptual level (Delwiche, 2004). In the case of interactions between other tastes or irritants and sweetness, the effect can occur at the receptor level and at the central level. In the case of interactions between colour and sweetness and between odour and sweetness, it is clear that the effect occurs at the central level.
Measuring consumers' perceptions of sweet taste
121
Influence of other tastes qualities on sweetness perception Tastes present in a food or a beverage can have different impacts on perception of sweetness depending on their own intensity and on sweetness intensity (see Keast and Breslin, 2002 for a review). When sweetness is at low intensity, saltiness has an enhancing effect whereas bitterness and sourness do not have a consistent influence. When sweetness is at medium intensity, the only consistent effect is a suppression of sweetness by bitterness. At high sweetness intensity, salty, bitter and sour tastes all have suppressing effects on sweetness. When more than two taste qualities are mixed, as often in real foods and beverages, interactions are more complex and rarely predictable (see Breslin and Beauchamp, 1997 for an example). Influence of other sensory modalities Chemical irritation induced by compounds such as capsaicin reduces the sweetness of sucrose or tomato soup (Prescott et al., 1993; Prescott and Stevenson, 1995), without having any effect on other taste qualities. Temperature might have an effect on perceived sweetness, with decreasing temperature decreasing the intensity of perceived sweetness, however this effect seems to be limited (Schiffman et al., 2000b). The reduction of the temperature of the tongue seems to be more important than the reduction of the temperature of the food or beverage. Interestingly, the decrease in sweetness with decrease in temperature is less marked with artificial sweeteners such as saccharin than with glucose or fructose (Green and Frankmann, 1988). It has also been demonstrated that colour may alter taste perception. Though a specific colour does not seem to have an influence on specific tastes (see Delwiche, 2004 for a review), it seems that increased colour level is associated with increased flavour intensity. Moreover, learned taste-colour associations are likely to influence perceived taste, in particular sweetness ratings. For instance, altering the relationship between green and yellow colours in lemon and lime flavoured sucrose solution changed the sweetness ratings (Roth et al., 1988). Texture has also an influence on perceived taste and the most studied case is this of viscous solutions. A vast array of studies showed that increasing the viscosity of a solution decreases the intensity of taste, and in particular of the sweet taste (Delwiche, 2004). Interactions of texture in solid foods with perceived sweetness have not been investigated in detail.
6.4
Future trends
Several methods are available to characterize the various aspects of consumers' perception of sweetness (sensitivity, intensity, time course and hedonic value). Research to improve methods and data analysis is however still on-going, in particular to take into account individual differences in perception. When developing researches on sensory or preference evaluation related to sweetness, a thorough analysis should be conducted beforehand in order to define as
122
Optimising sweet taste in foods
precisely as possible the research questions and, therefore, to choose the most adequate methods which are to be used. Different examples were given in this chapter to underline the biases inherent in some methods. For food and beverage product development, it is very important to keep in mind that sweetness perception and more particularly preference for sweetness is very likely to depend on the targeted population. Preference for sweetness evolves with age and according to different cultures; it also depends largely on experience with sweet food consumption. This makes it difficult to formulate a product which would be ideally suited for all countries or all age groups. Another difficulty in formulating lies in the fact that sweetness perception will not only depend on the amount of sugars or sweeteners, but also on the global context of the food: other taste-active compounds, temperature, colour and texture are likely to influence its perception. Clearly, more research is needed to better understand those interactions and maybe, in the long term, to establish reliable rules for improving and rationalizing sweet product development.
6.5
Sources of further information
Despite the variety of methods available to conduct sensory evaluation, practices in the field were harmonized by using the recommendations provided by standardization organizations such as ISO or ASTM. For the most up-to-date versions of the standards, visit www.iso.org or www.astm.org. Several books give insights on the methods and on data analysis, such as Sensory Evaluation Techniques, by Meilgaard et al. (1999, Boca Raton, CRC Press), Sensory Evaluation Practices, by Stone and Sidel (2004, London, Academic Press), Sensory Evaluation of Food, Principles and Practices, by Lawless and Heymann (1998, Fredericksburg, Aspen) and Measurement of Food Preference by MacFie and Thomson (1994, London, Blackie). Several publications are essentially devoted to presenting results of the study of perception. Journals such as Chemical Senses, Physiology & Behavior, Food Quality and Preference, Journal of Sensory Studies and Appetite are good sources of information. Conferences are also a good place to learn about the most recent developments in the field: more fundamental aspects would be typically presented at the conferences of the European Chemoreception Organisation (ECRO) or of the Association for Chemoreception Science (AChemS), whereas applied aspects would be showed during the Pangborn Sensory Science Symposium or the IFT meeting.
6.6 AFNOR
References (2000), MeÂthodologie. Directives geÂneÂrales pour la reÂalisation d'eÂpreuves heÂdoniques en laboratoire d'eÂvaluation sensorielle ou en salle en conditions controÃleÂes impliquant des consommateurs. XP V 09-500, Paris ± La DeÂfense, AFNOR.
Measuring consumers' perceptions of sweet taste
123
and O'MAHONY, M. (2005), The paired preference test and the 'No Preference' option: was Odesky correct?, Food Quality and Preference, 16, 425±34. ARBISI, P. A., BILLINGTON, C. J. and LEVINE, A. S. (1999), The effect of naltrexone on taste detection and recognition threshold, Appetite, 32, 241±9. ASTM (1991a), Standard practice for defining and calculating individual and group sensory thresholds from forced-choice data sets of intermediate size. E 1432-91, American Society for Testing Materials. ASTM (1991b), Standard practice for determination of odor and taste thresholds by a forced-choice ascending concentration series method of limits. E 679-91, American Society for Testing Materials. AYYA, N. and LAWLESS, H. T. (1992), Quantitative and qualitative evaluation of highintensity sweetener mixtures, Chemical Senses, 17, 245±59. BACON, A. W., MILES, J. S. and SCHIFFMAN, S. S. (1994), Effect of race on perception of fat alone and in combination with sugar, Physiology & Behavior, 55, 603±6. BARTOSHUK, L. M. (1989), Taste: robust across the age span?, Annals of the New-York Academy of Sciences, 561, 65±75. BARTOSHUK, L. M., RIFKIN, B., MARKS, L. E. and BARS, P. (1986), Taste and aging, Journal of Gerontology, 41, 51±7. BARTOSHUK, L. M., DUFFY, V. B., FAST, K., GREEN, B. G. and SNYDER, D. J. (2001), The General Labeled Magnitude Scale provides valid measures of genetic variation in taste and may be a universal psychophysical ruler, Appetite, 37, 126. BARTOSHUK, L. M., DUFFY, V. B., FAST, K., GREEN, B. G., PRUTKIN, J. and SNYDER, D. J. (2003), Labeled scales (e.g., category, Likert, VAS) and invalid across-group comparisons: what we have learned from genetic variation in taste, Food Quality and Preference, 14, 125±38. BARTOSHUK, L. M., DUFFY, V. B., CHAPO, A. K., FAST, K., YIEE, J. H., HOFFMAN, H. J., KO, C.-W. and SNYDER, D. J. (2004), From psychophysics to the clinic: missteps and advances, Food Quality and Preference, 15, 617±32. BARYLKO-PIKIELNA, N., MATUSZEWSKA, I., KOZLOWSKA, K., JERUSZKA, M. and ROSZKOWSKI, W. (2002), Perception of sweetness and sourness in apple juice varying in sucrose level in elderly and young adults and its relation to hedonic response, Polish Journal of Food and Nutritional Sciences, 11/52, 65±74. BEAUCHAMP, G. K. and MORAN, M. (1982), Dietary experience and sweet taste preferences in human infants, Appetite, 3, 139±52. BEAUCHAMP, G. K. and MORAN, M. (1984), Acceptance of sweet and salty tastes in 2-yearold children, Appetite, 5, 291±305. BELLISLE, F. and LUCAS, F. (1987), La mesure des preÂfeÂrences alimentaires: les tests brefs d'eÂvaluation sensorielle permettent-ils de preÂdire la consommation?, Sciences des Aliments, 7, 33±7. BERTINO, M. and CHAN, M. M. (1986), Taste perception and diet in individuals with Chinese and European ethnic backgrounds, Chemical Senses, 11, 229±41. BERTINO, M., BEAUCHAMP, G. K. and JEN, K. C. (1983), Rated taste perception in two cultural groups, Chemical Senses, 8, 3±15. BEST, D. J. and RAYNER, J. C. W. (2000), Extended statistical analysis of ranked data from balanced incomplete block and paired comparison designs, Journal of Sensory Studies, 15, 383±92. BI, J. and ENNIS, D. M. (1998), Sensory thresholds: concepts and methods, Journal of Sensory Studies, 13, 133±48. BLASS, E. M. and HOFFMEYER, L. B. (1991), Sucrose as an analgesic for newborn infants, ANGULO, O.
124
Optimising sweet taste in foods
Pediatrics, 87, 215±18. and NOBLE, A. C. (1995), Interaction of salivary flow with temporal perception of sweetness, sourness and fruitiness, Physiology & Behavior, 57, 569±74. BOOTH, D. A. and CONNER, M. T. (1990), Characterization and measurement of influences on food acceptability by analysis of choice differences: theory and practice, Food Quality and Preference, 2, 75±85. BOOTH, D. A., THOMPSON, A. and SHAHEDIAN, B. (1983), A robust, brief measure of an individual's most preferred level of salt in an ordinary foodstuff, Appetite, 4, 301±12. BOWER, J. A. and BOYD, R. (2003), Effect of health concern and consumption patterns on measures of sweetness by hedonic and just-about-right scales, Journal of Sensory Studies, 18, 235±48. BRESLIN, P. A. S. and BEAUCHAMP, G. K. (1997), Salt enhances flavor by suppressing bitterness, Nature, 387, 563. CABANAC, M. (1971), Physiological role of pleasure. A stimulus can feel pleasant or unpleasant depending upon its useful as determined by internal signals, Science, 173, 1103±7. CABANAC, M. (1979), Sensory pleasure, Quarterly Review of Biology, 54, 1±29. CALVINÄO, A. M. and GARCIÂA, M. (1998), Perceived intensity and persistence of heterogeneous taste mixtures, Food Science and Technology International, 4, 277±84. CLIFF, M. and HEYMANN, H. (1993), Development and use of time-intensity methodology for sensory evaluation ± A review, Food Research International, 26, 375±85. CONNER, M. T. and BOOTH, D. A. (1988), Preferred sweetness of a lime drink and preference for sweet over non-sweet foods, related to sex and reported age and body weight, Appetite, 10, 25±35. CONNER, M. T., LAND, D. G. and BOOTH, D. A. (1987), Effect of stimulus range on judgments of sweetness intensity in a lime drink, British Journal of Psychology, 78, 357±64. CONNER, M. T., HADDON, A. V., PICKERING, E. S. and BOOTH, D. A. (1988), Sweet tooth demonstrating: individual differences in preference for both sweet foods and foods highly sweetened, Journal of Applied Psychology, 73, 275±80. COSTELL, E., PASTOR, M. V., IZQUIERDO, L. and DURAN, L. (1994), Comparison of simplified methods for evaluation of sensory thresholds of added substances, Journal of Sensory Studies, 9, 365±82. COWART, B. J. (1989), Relationships between taste and smell across the adult life span, Annals of the New York Academy of Sciences, 561, 39±55. CROOK, C. K. (1978), Taste perception in the newborn infant, Infant Behavior and Development, 1, 52±69. CROVETTI, R., SANTANGELO, A. and RISO, P. (1997), Sweet taste reactivity and satiety, Nutrition Research, 17, 1417±25. DAILLANT, B. and ISSANCHOU, S. (1991), Most preferred level of sugar: rapid measure and consumption test, Journal of Sensory Studies, 6, 131±44. DE GRAAF, C. and ZANDSTRA, E. H. (1999), Sweetness intensity and pleasantness in children, adolescents, and adults, Physiology & Behavior, 67, 513±20. DE GRAAF, C., POLET, P. and VAN STAVEREN, W. (1994), Sensory perception and pleasantness of food flavors in elderly subjects, Journal of Gerontology. Series B: Psychological Sciences and Social Sciences, 49, 93±9. DE GRAAF, C., VAN STAVEREN, W. and BUREMA, J. (1996), Psychophysical and psychohedonic functions of four common food flavours in elderly subjects, Chemical Senses, 21, 293±302. BONNANS, S. R.
Measuring consumers' perceptions of sweet taste
125
and VAN STAVEREN, W. A. (1996), Effect of sucrose in breakfast items on pleasantness and food intake in the elderly, Physiology & Behavior, 60, 1453±62. DELWICHE, J. (2004), The impact of perceptual interactions on perceived flavor, Food Quality and Preference, 15, 137±46. DESOR, J. A. and BEAUCHAMP, G. K. (1987), Longitudinal changes in sweet preferences in humans, Physiology & Behavior, 39, 639±41. DESOR, J. A., GREENE, L. S. and MALLER, O. (1975), Preference for sweet and salty in 9- to 15year-old and adult humans, Science, 190, 686±7. DIAMOND, J. and LAWLESS, H. T. (2001), Context effects and reference standards with magnitude estimation and the labeled magnitude scale, Journal of Sensory Studies, 16, 1±10. DOTY, R. L. and BROMLEY, S. M. (2004), Effects of drugs on olfaction and taste, Otolaryngology Clinics of North America, 37, 1229±54. DREWNOWSKI, A. (1989), Sensory preferences for fat and sugar in adolescence and adult life, In Murphy, C., Cain, W. S. and Hegsted, D. M., Nutrition and the chemical senses in aging: Recent advances and current research needs, The New York Academy of Sciences, New York, pp. 243±50. DREWNOWSKI, A. and HOLDEN-WILTSE, J. (1992), Taste responses and food preferences in obese women: effects of weight cycling, International Journal of Obesity, 16, 639± 48. DREWNOWSKI, A., BRUNZELL, J. D., SANDE, K., IVERIUS, P. H. and GREENWOOD, M. R. C. (1985), Sweet tooth reconsidered: Taste responsiveness in human obesity, Physiology & Behavior, 35, 617±22. DREWNOWSKI, A., KURTH, C., HOLDENWILTSE, J. and SAARI, J. (1992), Food preferences in human obesity ± Carbohydrates versus fats, Appetite, 18, 207±21. DRUZ, L. L. and BALDWIN, R. E. (1982), Taste thresholds and hedonic response of panel representing three nationalities, Journal of Food Science, 47, 561±9. DE JONG, N., DE GRAAF, C.
DUBOIS, G. E., WAITERS, D. E., SCHIFFMAN, S. S., WARWICK, Z. S., BOOTH, B. J., PECORE, S. D.,
and BRANDS, L. M. (1991), Concentration-response relationships of sweeteners ± a systematic study, In Walters, D. E., Orthoefer, F. T. and DuBois, G. E., Sweeteners. Discovery, Molecular Design and Chemoreception, American Chemical Society, Washington DC, pp. 261±76. DUFFY, V. B., BARTOSHUK, L. M., STRIEGEL-MOORE, R. and RODIN, J. (1998a), Acceptance of salty, sweet and bitter food across pregnancy, Chemical Senses, 23, 560±1. DUFFY, V. B., BARTOSHUK, L. M., STRIGEL-MOORE, R. and RODIN, J. (1998b), Taste changes across pregnancy, Annals of the New-York Academy of Sciences, 855, 805±9. EARTHY, P. J., MACFIE, H. J. H. and HEDDERLEY, D. (1997), Effect of question order on sensory perception and preference in central location trials, Journal of Sensory Studies, 12, 215±37. EASTERBY-SMITH, V., BESFORD, J. and HEATH, M. R. (1994), The effect of age on the recognition thresholds of three sweeteners: sucrose, saccharin and aspartame, Gerodontology, 11, 39±45. ENNS, M. P., VAN ITALLIE, T. B. and GRINKER, J. A. (1979), Contributions of age, sex and degree of fatness on preferences and magnitude estimations for sucrose in humans, Physiology & Behavior, 22, 999±1003. EPLER, S., CHAMBERS IV, E. and KENNETH, K. E. (1998), Hedonic scales are a better predictor than just-about-right scale of optimal sweetness in lemonade, Journal of Sensory Studies, 13, 191±7. GIBES, K., CARR, B. T.
126
Optimising sweet taste in foods
and RASMUSSEN-CONRAD, E. L. (1982), Sensory discrimination intensity perception, and affective judgment of sucrose-sweetness in the overweight, The Journal of General Psychology, 107, 233±47. GABRIELSEN, G. (2000), Paired comparisons and designed experiments, Food Quality and Preference, 11, 55±61. GILMORE, M. M. and MURPHY, C. (1989), Aging is associated with increased Weber ratios for caffeine, but not for sucrose, Perception & Psychophysics, 46, 555±9. Â LEZ-VINÄAS, M. A. and MARTIN-ALVAREZ, P. J. (1998), Comparison of two simple GONZA methods for the measurement of detection thresholds from basic, umami and metallic tastes, Journal of Sensory Studies, 13, 299±314. GREEN, B. G. and FRANKMANN, S. P. (1988), The effect of cooling on the perception of carbohydrate and intensive sweeteners, Physiology & Behavior, 43, 515±19. GREEN, B. G., SHAFFER, G. S. and GILMORE, M. M. (1993), Derivation and evaluation of a semantic scale of oral sensation magnitude with apparent ratio properties, Chemical Senses, 18, 683±702. GREEN, B. J., DALTON, P., COWART, B., SHAFFER, G., RANKIN, K. and HIGGINS, J. (1996), Evaluating the `labeled magnitude scale' for measuring sensations of taste and smell, Chemical Senses, 21, 323±34. GREENE, L. J., DESOR, J. A. and MALLER, O. (1975), Heredity and experience: their relative importance in the development of taste preference in man, Journal of Comparative and Physiological Psychology, 89, 279±84. GREENHOFF, K. and MACFIE, H. J. H. (1994), Preference mapping in practice, In MacFie, H. J. H. and Thomson, D. M. H., Measurement of food preferences, Blackie, London, pp. 137±66. GRINKER, J., PRICE, J. and GREENWOOD, M. (1976), Studies of childhood obesity, In Novin, D., Wyrwicka, W. and Bray, D., Hunger: Basic mechanisms and clinical implications, Raven Press, New York, pp. 441±57. GUICHARD, E., TROMELIN, A., JUTEAU, A., REGA, B. and ROUDNITZKY, N. (2003), Physical and chemical interactions in several flavour compounds/food matrix systems, In Le QueÂreÂ, J. L. and EtieÂvant, P. X., 10th Weurman flavour research symposium, Lavoisier, Paris, Beaune, France, pp. 15±20. GUINARD, J.-X., ZOUMAS-MORSE, C., WALCHAK, C. and SIMPSON, H. (1997), Relation between saliva flow and flavor release from chewing gum, Physiology & Behaviour, 61, 591±6. HENKIN, R. I. (1994), Drug-induced taste and smell disorders, Drug Safety, 11, 318±77. HOLT, S. H. A., COBIAC, L., BEAUMONT-SMITH, N. E., EASTON, K. and BEST, D. J. (2000), Dietary habits and the perception and liking of sweetness among Australian and Malaysian students: a cross-cultural study, Food Quality and Preference, 11, 299±312. HYDE, R. J. and FELLER, R. P. (1981), Age and sex effects on taste of sucrose, NaCl, citric acid and caffeine, Neurobiology of Aging, 1, 315±18. ISHII, R., YAMAGUCHI, S. and O'MAHONY, M. O. (1992), Measure of taste discriminability for sweet, salty and umami stimuli: Japanese versus Americans, Chemical Senses, 17, 365±80. ISO (1992), Sensory Analysis. Vocabulary. NF ISO 5492, Paris ± La DeÂfense, AFNOR. Â lection de descripteurs pour l'eÂlaboration d'un profil ISO (1995), Recherche et se sensoriel, par approche multidimensionnelle. NF ISO 11035, Paris ± La DeÂfense, AFNOR. Â thode d'estimation de la grandeur. NF ISO 11056, ISO (2000), Analyse sensorielle. Me Paris ± La DeÂfense, AFNOR. FRIJTERS, J. E. R.
Measuring consumers' perceptions of sweet taste
127
(2004), Sensory analysis. Guidelines for the use of quantitative response scales. NF ISO 4121, Paris ± La DeÂfense. JAMEL, H. A., SHEIMAN, A., COWELL, C. R. and WATT, R. G. (1996), Taste preference for sweetness in urban and rural population of Iraq, Journal of Dentistry Research, 75, 1879±84. JAMES, C. E., LAING, D. G., ORAM, N. and HUTCHINSON, I. (1999), Perception of sweetness in simple and complex taste stimuli by adults and children, Chemical Senses, 24, 281±7. JAMES, C. E., LAING, D. G., JINKS, A. L., ORAM, N. and HUTCHINSON, I. (2004), Taste response functions of adults and children using different rating scales, Food Quality and Preference, 15, 77±82. JEON, S. Y., O'MAHONY, M. and KIM, K. O. (2004), A comparison of category and line scales under various experimental protocols, Journal of Sensory Studies, 19, 49±66. JONES, L. V., PERYAM, D. R. and THURSTONE, L. L. (1955), Development of a scale for measuring soldiers' food preferences, Food Research, 20, 512±20. KEAST, R. S. J. and BRESLIN, P. A. S. (2002), An overview of binary taste±taste interactions, Food Quality and Preference, 14, 111±24. KOSKINEN, S., KALVIAINEN, N. and TUORILA, H. (2003), Perception of chemosensory stimuli and related responses to flavored yogurts in the young and elderly, Food Quality and Preference, 14, 623±35. È STER, E. P. (1981), Sensory evaluation in a `natural environment', In Schreier, P., KO Flavour '81, Walter de Gruyter, Berlin, pp. 93±100. È STER, E. P. (1998), Les eÂpreuves heÂdoniques, In SSHA and Depledt, F., Evaluation KO sensorielle: Manuel meÂthodologique, Lavoisier, Paris, pp. 182±206. LAENG, B., BERRIDGE, K. C. and BUTTER, C. M. (1993), Pleasantness of a sweet taste during hunger and satiety: effects of gender and `sweet tooth', Appetite, 21, 247±54. LAFFORT, P., WALSH, R. M. and SPILLANE, W. J. (2002), Application of the U and gamma models in binary sweet taste mixtures, Chemical Senses, 27, 511±20. È HTEENMA È KI, L. and TUORILA, H. (1994), Attitudes towards sweetness as predictors of LA liking and use of various sweet foods, Ecology of Food and Nutrition, 31, 161±70. È HTEENMA È KI, L. and TUORILA, H. (1995), Three-factor eating questionnaire and the use LA and liking of sweet and fat among dieters, Physiology & Behavior, 57, 81±8. ISO
LAING, D. J., PRESCOTT, J., BELL, G. A., GILLMORE, R., JAMES, C., BEST, D. J., ALLEN, S., YOSHIDA, M. and YAMAZAKI, K. (1993), A cross-cultural study of taste discrimination with Australians and Japanese, Chemical Senses, 18, 161±8.
LAING, D. J., PRESCOTT, J., BELL, G. A., GILLMORE, R., ALLEN, S., BEST, D. J., YOSHIDA, M.,
and ISHII, R. (1994), Responses of Japanese and Australians to sweetness in the context of different foods, Journal of Sensory Studies, 9, 131±55. LAWLESS, H. T. and HEYMANN, H. (1998), Sensory evaluation of food, principles and practices, Aspen, Ferdericksburg, MD. LAWLESS, H. T., HORNE, J. and SPIERS, W. (2000), Contrast and range effects for category, magnitude and labeled magnitude scales in judgements of sweetness intensity, Chemical Senses, 25, 85±92. LEE III, W. E. and PANGBORN, R. M. (1986), Time-intensity: The temporal aspects of sensory perception, Food Technology, 40, 71±8 and 82. È STER, E. P. (1999), Measuring food liking in LEÂON, F., COURONNE, T., MARCUZ, M. C. and KO children: a comparison of non verbal methods, Food Quality and Preference, 10, 93±100. YAMAZAKI, K.
128
Optimising sweet taste in foods
and MACFIE, H. J. H. (1990), Methods for averaging time-intensity curves, Chemical Senses, 15, 471±84. LOOY, H. and WEINGARTEN, H. P. (1991), Effects of metabolic state on sweet taste reactivity in humans depend on underlying hedonic response profile, Chemical Senses, 16, 123±30. LUNDAHL, D. S., LUKES, B. K., MCDANIEL, M. R. and HENDERSON, L. A. (1986), A semiascending paired difference method for determining sensory thresholds of added substances to background media, Journal of Sensory Studies, 1, 291±306. LUNDGREN, B., PANGBORN, R. M., BARYLKO-PIKIELNA, N. and DAGET, N. (1976), Difference taste thresholds for sucrose in water and in orange juice: an interlaboratory study, Chemical Senses and Flavor, 2, 157±76. LIU, Y.-H.
LUNDGREN, B., JONSSON, B., PANGBORN, R. M., SONTAG, A., BARYLKO-PIKIELNA, N., PIETRZAK,
and YOSHIDA, M. (1978), Taste discrimination vs hedonic response to sucrose in coffee beverage. An interlaboratory study, Chemical Senses, 3, 249±65. MACDIARMID, J. I., VAIL, A., CADE, J. E. and BLUNDELL, J. E. (1998), The sugar-fat relationship revisited: differences in consumption between men and women of varying BMI, International Journal of Obesity, 22, 1053±61. MACFIE, H. J. and THOMSON, D. M. (1994), Measurement of food preferences, Blackie, London. MACFIE, H. J., BRATCHELL, N., GREENHOFF, K. and VALLIS, L. V. (1989), Designs to balance the effect of order of presentation and first-order carry-over effects in hall tests, Journal of Sensory Studies, 4, 129±48. MACFIE, H. J., EARTHY, P. and DELIZA, R. (1996), Measurement of consumer preference, in Consumer preference and sensory analysis (Ed, Poretta, S.) M. Freeman, Maarsen, The Netherlands, pp. 9±20. MALLER, O., CARDELLO, A., SWEENEY, J. and SHAPIRO, D. (1982), Psychological and cognitive correlates of discretionary usage of table salt and sugar by humans, In Steiner, J. E. and Ganchrow, J. R., Determination of Behavior by Chemical Studies, IRL Press, London, pp. 205±18. MATSUO, R. (2000), Role of saliva in the maintenance of taste sensitivity, Critical Reviews in Oral Biology and Medicine, 11, 216±29. MATTES, R. D. (1985), Gustation as a determinant of ingestion: methodological issues, The American Journal of Clinical Nutrition, 41, 672±83. MATTES, R. D. and MELA, D. J. (1986), Relationships between and among selected measures of sweet taste preferences and dietary intake, Chemical Senses, 11, 523±39. MCEWAN, J. A. (1988), Methodology and new applications in food acceptance research, PhD Thesis, University of Reading. MEILGAARD, M., CIVILLE, G. V. and CARR, B. T. (1999), Sensory evaluation techniques, Boca Raton, CRC Press. MENNELLA, J. A., PEPINO, M. Y. and REED, D. R. (2005), Genetic and environmental determinants of bitter perception and sweet preferences, Pediatrics, 115, e216±e22. MOJET, J., CHRIST-HAZELHOF, E. and HEIDEMA, J. (2001), Taste perception with age: generic or specific losses in threshold sensitivity to the five basic tastes?, Chemical Senses, 26, 845±60. MOJET, J., HEIDEMA, J. and CHRIST-HAZELHOF, E. (2003), Taste perception with age: generic or specific losses in supra-threshold intensities of five taste qualities?, Chemical Senses, 28, 397±413. MOJET, J., HEIDEMA, J. and CHRIST-HAZELHOF, E. (2004), Effect of concentration on taste± E., DOS SANTOS GARRUTI, R., MORAES, M. A. C.
Measuring consumers' perceptions of sweet taste
129
taste interactions in foods for elderly and young subjects, Chemical Senses, 29, 671±81. MONNEUSE, M.-O., BELLISLE, F. and LOUIS-SYLVESTRE, J. (1991), Impact of sex and age on sensory evaluation of sugar and fat in dairy products, Physiology & Behavior, 50, 1111±17. MOSKOWITZ, H. R., KUMRAIAH, V., HARMA, K. N., JACOBS, K. L. and SHARMA, S. D. (1976), Effects of hunger satiety and glucose load upon taste intensity and hedonics, Physiology & Behavior, 16, 471±5. MURPHY, C. and GILMORE, M. M. (1989), Quality-specific effects of aging on the human taste system, Perception & Psychophysics, 45, 121±8. MURPHY, C. and WITHEE, J. (1986), Age-related difference in the pleasantness of chemosensory stimuli, Psychology and Aging, 1, 312±18. ORAM, N., LAING, D. G., FREEMAN, M. H. and HUTCHINSON, I. (2001), Analysis of taste mixture by adults and children, Developmental Psychobiology, 38, 67±77. OVERBOSCH, P., ENDEN, J. C. V. D. and KEUR, B. M. (1986), An improved method for measuring perceived intensity/time relationships in human taste and smell, Chemical Senses, 11, 331±8. PANGBORN, R. M. and GIOVANNI, M. E. (1984), Dietary intake of sweet foods and of dairy fats and resultant gustatory responses to sugar in lemonade and to fat in milk, Appetite, 5, 317±27. PELLETIER, C. A., LAWLESS, H. T. and HORNE, J. (2004), Sweet-sour mixture suppression in older and young adults, Food Quality and Preference, 15, 105±16. PEPINO, M. Y. and MENNELLA, J. A. (2005), Factors contributing to individual differences in sucrose preference, Chemical Senses, 30, i319±i20. PERYAM, D. R. and PILGRIM, F. J. (1957), Hedonic scale method of measuring food preferences, Food Technology, 11, 9±14. POMERLEAU, C. S., GARCIA, A. W., DREWNOWSKI, A. and POMERLEAU, O. F. (1991), Sweet taste preference in women smokers ± Comparison with nonsmokers and effects of menstrual phase and nicotine, Pharmacology Biochemistry and Behavior, 40, 995±9. POPPER, R., ROSENSTOCK, W., SCHRAIDT, M. and KROLL, B. J. (2004), The effect of attribute questions on overall liking ratings, Food Quality and Preference, 15, 853±8. PORTMANN, M. O. and KILCAST, D. (1996), Psychophysical characterization of new sweeteners of commercial importance for the EC food industry, Food Chemistry, 56, 291±302. PORTMANN, M. O. and KILCAST, D. (1998), Descriptive profiles of synergistic mixtures of bulk and intense sweeteners, Food Quality and Preference, 9, 221±9. PRESCOTT, J. and STEVENSON, R. J. (1995), Effects of oral chemical irritation on tastes and flavors in frequent and infrequent users of chili, Physiology & Behavior, 58, 1117± 27. PRESCOTT, J., LAING, D., BELL, G., YOSHIDA, M., GILLMORE, R., ALLEN, S., YAMAZAKI, K. and ISHII, R. (1992), Hedonic responses to taste solutions ± a cross-cultural study of Japanese and Australians, Chemical Senses, 17, 801±9. PRESCOTT, J., ALLEN, S. and STEPHENS, L. (1993), Interactions between oral chemical irritation, taste and temperature, Chemical Senses, 18, 389±404. PRESCOTT, J., BELL, G. A., GILLMORE, R., YOSHIDA, M., O'SULLIVAN, M., KORAC, S., ALLEN, S. and YAMAZAKI, K. (1997), Cross-cultural comparisons of Japanese and Australian responses to manipulations of sweetness in foods, Food Quality and Preference, 8, 45±55. QANNARI, E. M., COURCOUX, P. and SEMEÂNOU, M. (2000), New models for the analysis of
130
Optimising sweet taste in foods
paired comparison data: segmentation of the panel, Food Quality and Preference, 11, 71±96. RISKEY, D. R., PARDUCCI, A. and BEAUCHAMP, G. K. (1979), Effects of context in judgments of sweetness and pleasantness, Perception & Psychophysics, 26, 171±6. ROSENSTEIN, D. and OSTER, H. (1988), Differential facial responses to four basic tastes in newborns, Child Development, 59, 1555±68. ROTH, H. A., RADLE, L. J., GIFFORD, S. R. and CLYDESDALE, F. M. (1988), Psychophysical relationships between perceived sweetness and color in lemon- and lime-flavored drinks, Journal of Food Science, 53, 1116±19+62. SALBE, A. D., DELPARIGI, A., PRATLEY, R. E., DREWNOWSKI, A. and TATARANNI, P. A. (2004), Taste preferences and body weight changes in an obesity-prone population, American Journal of Clinical Nutrition, 79, 372±8. SCHIFFMAN, S. S., LINDLEY, M. G., CLARK, T. B. and MAKINO, C. (1981), Molecular mechanism of sweet taste: relationship of hydrogen bonding to taste sensitivity for both young and elderly, Neurobiology of Aging, 2, 173±85. SCHIFFMAN, S. S., GRAHAM, B. G., SUGGS, M. S. and SATTELY-MILLER, E. A. (1998), Effect of psychotropic drugs on taste responses in young and elderly persons, Annals of the New-York Academy of Sciences, 855, 732±7. SCHIFFMAN, S. S., GRAHAM, B. G., SATTELY-MILLER, E. A. and PETERSON-DANCY, M. (2000a), Elevated and sustained desire for sweet taste in African-Americans: A potential factor in the development of obesity, Nutrition, 16, 886±93. SCHIFFMAN, S. S., SATTELY-MILLER, E. A., GRAHAM, B. G., BENNETT, J. L., BOOTH, B. J., DESAI, N.
and BISHAY, I. (2000b), Effects of temperature, pH and ions on sweet taste, Physiology & Behavior, 68, 469±81. SCHIFFMAN, S. S., SATTELY-MILLER, E. A., GRAHAM, B. G., BOOTH, B. J. and GIBES, K. M. (2000c), Synergism among ternary mixture of fourteen sweeteners, Chemical Senses, 25, 131±40. SCHLICH, P. (1995), Preference mapping: relating consumer preferences to sensory or instrumental measurements, In EtieÂvant, P. and Schreier, P., Bioflavour 95. Analysis/Precursor studies/Biotechnology, INRA Editions, Paris, pp. 135±50. SHEPHERD, R., FALEIGH, C. A., LAND, D. G. and FRANKLIN, J. G. (1985) In Progress in flavour research (Ed, Adda, J.) Elsevier, Amsterdam, pp. 103±10. STEIN, N., LAING, D. G. and HUTCHINSON, I. (1994), Topographical differences in sweetness sensitivity in the peripheral gustatory system of adults and children, Developmental Brain Research, 82, 286±92. STEINER, J. E. (1979), Human facial expressions in response to taste and smell stimulation, Advances in Child Development and Behavior, 13, 257±95. STEVENS, D. A. and LAWLESS, H. T. (1981), Age-related changes in flavor perception, Appetite, 2, 127±36. STEVENS, J. C. (1996), Detection of tastes in mixture with other tastes: issues of masking and aging, Chemical Senses, 21, 211±21. STEVENS, J. C. and CAIN, W. S. (1993), Changes in taste and flavor in aging, Critical Reviews in Food Science and Nutrition, 33, 27±38. STEVENS, J. C., CRUZ, L. A., HOFFMAN, J. M. and PATTERSON, M. Q. (1995), Taste sensitivity and aging: high incidence of decline revealed by repeated threshold measures, Chemical Senses, 20, 451±9. STONE, H. and SIDEL, J. L. (2004), Sensory evaluation practices, London, Academic Press. STONE, L. J. and PANGBORN, R. M. (1990), Preferences and intake measures of salt and sugar, and their relation to personality traits, Appetite, 15, 63±79.
Measuring consumers' perceptions of sweet taste
131
and BIRCH, L. L. (1990), Pass the sugar, pass the salt: Experience dictates preference, Developmental Psychology, 26, 546±51. THOMPSON, D. A., MOSKOVITZ, H. R. and CAMPBELL, R. G. (1976), Effects of body weight and food intake on pleasantness ratings for a sweet stimulus, Journal of Applied Physiology, 41, 77±83. THOMSON, D. M. H. and TUNALEY, A. (1987), Examination of the sensory characteristics of sweeteners using multidimensional scaling, In Martens, M., Dalen, G. A. and Russwurm Jr., H., Flavour Science and Technology, Wiley & Sons, Chichester, pp. 347±56. TUORILA-OLLIKAINEN, H. and MAHLAMaÈKI-KULTANEN, S. (1985), The relationship of attitudes and experiences of Finnish youths to their hedonic responses to sweetness in soft drinks, Appetite, 6, 115±24. VASQUEZ, M., PEARSON, P. B. and BEAUCHAMP, G. K. (1982), Flavor preferences in malnourished Mexican infants, Physiology & Behavior, 28, 513±19. VICKERS, Z. (1988), Sensory specific satiety in lemonade using a just right scale for sweetness, Journal of Sensory Studies, 3, 1±8. VIGNEAU, E., QANNARI, E. M., PUNTER, P. H. and KNOOPS, S. (2001), Segmentation of a panel of consumers using clustering of variables around latent directions of preference, Food Quality and Preference, 12, 359±63. WARWICK, Z. S. and SCHIFFMAN, S. S. (1990), Sensory evaluations of fat-sucrose and fat-salt mixtures: relationship to age and weight status, Physiology & Behavior, 48, 633±6. WEIFFENBACH, J. M., COWART, B. J. and BAUM, B. J. (1986), Taste intensity perception in aging, Journal of Gerontology, 41, 460±8. WEIFFENBACH, J. M., TYLENDA, C. A. and BAUM, B. J. (1990), Oral sensory changes in aging, Journal of Gerontology. Series A: Biological Sciences and Medical Sciences, 45, M121±M5. ZANDSTRA, E. H. and DE GRAAF, C. (1998), Sensory perception and pleasantness of orange beverages from childhood to old age, Food Quality and Preference, 9, 5±12. ZVEREV, Y. P. (2004), Effects of caloric deprivation and satiety on sensitivity of the gustatory system, BMC Neurosci, 5, 5. SULLIVAN, S. A.
Part II Types of sweet tasting compounds
7 Sucrose J. M. Cooper, British Sugar plc, UK
7.1
Introduction
Sucrose, commonly known as sugar, but more precisely -D-glucopyranosyl D-fructofuranoside has been reported to have the world's highest production of any single, pure, natural, organic chemical (Parker et al., 1977). Throughout the remainder of this chapter `sugar' will be used when referring to sucrose. This chapter will describe some of the history of sugar and detail many of the reactions and functional properties of sugar which result in a wide diversity of food products which have been developed using this versatile food ingredient. This book considers the sweet taste in foods and sweetness is one of few tastes which is innate and it has been argued that a preference for sweet taste evolved to ensure that that animals chose a diet adequate in certain vitamins and/ or minerals (Beauchamp and Cowart, 1987). Early man was exposed to sweet tastes from natural fruits, e.g. the date palm and honey that he collected by the smoking of bees from their nests. Today the two main crops that are cultivated specifically for their sugar are sugar cane and sugar beet. Sugar cane was first recorded in Asia around 8000 BC and it is postulated that it originated as a native plant from Melanesia (Artschwanger and Brandes, 1958). One of the earliest written reports of sugar was from Nearchus, one of Alexander the Great's commanders, he reports in 327 BC `In India exist reeds that produce honey, although there are no bees, from which an intoxicating drink can be produced, although the plant carries no fruit'. It is postulated that he was referring to sugar cane. The conquering armies spread sugar cane around the Far and Middle East and the first Europeans to taste sugar were likely to have been the soldiers of the Crusades in the 11th and 13th centuries. Columbus brought sugar cane to the New World on his second voyage in 1493. Sugar cane
136
Optimising sweet taste in foods
Fig. 7.1
Worldwide sugar production and consumption 1980±2004 (adapted from Licht 2005).
shaped much of the history of the New World and cultivation soon spread to many countries in South America. The `sugar islands' of the West Indies brought great wealth to England and France. Sugar cane now accounts for over 75% of the world's production of sugar (see Fig. 7.1). The major cane sugar producers are Brazil, India, China, Thailand, Mexico and Australia accounting for over 60% of the total cane sugar produced in 2003/4 and nearly 50% of the world's total production of sugar (Licht, 2005). Sugar from sugar beet is a much more recent development. In 1747 Andreas Magraff, a member of the Berlin Academy of Sciences discovered a `salt' in the juice of white beet (a wild form of Beta vulgaris L.) that was indistinguishable from `true perfect sugar'. Beet processing for sugar received a major boost in the
Sucrose 137 early 1800s when the European mainland was cut off from sugar originating in the Caribbean by the British naval blockades. Napoleon issued decrees promoting beet growing and the construction of sugar factories. Beet processing expanded during the 19th and 20th centuries up to the present day where over 20% of the world's sugar production is derived from sugar beet. Over 50% of the sugar produced from beet is grown in France, Germany, the USA, Poland, Russia and Turkey (Licht, 2005). Over the last 30 years worldwide consumption of sugar has more or less matched production (see Fig. 7.1). The production from beet has been constant and the increasing demand has been provided by increased production from cane. Today the trade in sugar is still a major political subject. In many countries worldwide the production of sugar is subject to quotas and tariffs that are the centre of much heated debate across the world.
7.2
Sugar manufacture
7.2.1 Sugar manufacture from sugar beet Sugar manufacture from sugar beet is a multi-stage process carried out on a large scale in most Western European countries, parts of the USA, Japan and other temperate climate countries. Sugar beet is a biennial crop that achieves maximum sugar content in the autumn of the first year of growth. At this point it is harvested and the sugar contained within the cells of the sugar beet is extracted and crystallised. Once the sugar beet has been harvested the level of sugar will reduce rapidly, consequently the extraction process must be carried out promptly to maximise extraction yield. Hence most beet processing is carried out between September and February each year resulting in a seasonal harvesting and processing regime sometimes referred to as the `campaign'. The main stages of the manufacturing process are as follows: · Beet processing ± removal of soil, extraction of sugar juice, separation of plant material · Purification/concentration ± removal of non-sugars, removal of water (storage of intermediate syrup) · Crystallisation and storage of sugar ± isolation of pure sugar, bulk storage. Most of these stages are continuous and are only carried out for part of the year. The aim of the process is to obtain a stable product (crystal sugar or intermediate syrup) that can be stored for year-round delivery (crystal sugar) or processing (intermediate syrup). Beet processing Sugar beet are delivered from the farm and stored on site before processing. The first stage is to remove the soil adhering to the beet; this is achieved by washing using re-circulated water in a series of rotary drums. The washed sugar beets are
138
Optimising sweet taste in foods
sliced into thin slices (cossettes) and extracted using hot water (ca. 72ëC) in continuous counter-current extractors. The vegetable matter is recovered using presses and then dried to produce animal feed. The juice, containing the sugar (approximately 16±18%) and nonsugar plant components, is then purified. Purification/concentration During the extraction process non-sugar components present in the beet are coextracted with the sugar. The next stages of the process are designed to remove as many of these interfering materials as possible. Milk of lime is added to the sugar containing juice and the action of the high pH and high level of calcium ions promote many degradation and precipitation reactions. Carbon dioxide is added to the limed juice and calcium carbonate is precipitated which removes more interfering materials e.g. colour. All the reactions take place at high pHs (>pH 9) and at elevated temperatures (>85ëC). The impurities are removed from the sugar juice by sedimentation and/or filtration. To promote the crystallisation of pure sugar, water within the dilute syrup is removed by evaporation. To achieve this in a cost effective manner it is necessary to use multiple effect evaporation which utilises the steam driven off dilute juices at high temperature to evaporate the water from more concentrated juices at lower temperatures. The use of pressure and vacuum is necessary to carry out this effectively. The condensed vapours from this stage are used for heating and washing duties within the process. Crystallisation and storage of sugar The final stages of evaporation result in a super-saturated solution from which pure sugar crystallises. The sugar crystals are grown to an optimum size ca 500± 700 m before they are removed from the surrounding syrup using centrifuges. The exhausted syrup remaining is known as molasses. The sugar crystals are washed with hot condensate prior to drying and cooling. The dried, cooled crystals are stored in controlled atmosphere (temperature and humidity) bulk silos prior to packing or bulk transport. 7.2.2 Sugar manufacture from sugar cane Sugar cane is grown from short lengths of cane, called `setts', which are ready to be harvested after 11 to 18 months. Each sett can produce six or seven crops before it needs to be replaced. Harvesting is carried out mechanically or by hand, with the dry leaves or `trash' often being burned off beforehand to assist the subsequent processing. Once harvested, the cut canes must be processed as quickly as possible to conserve the sugar and to prevent microbial degradation. This first stage of processing is carried out in factories close to the growing area. The canes are cleaned, crushed and shredded, then sprayed with hot water to extract the juice. The juice is filtered, concentrated by evaporation under vacuum, and
Sucrose 139 crystallised, before being removed from the remaining mother liquor by centrifuge. At this point, the sugar is partly purified and in a concentrated, crystallised, microbiologically stable form, suitable for bulk handling, storage and transport to refineries around the world. The main stages of the cane refining process are as follows: · · · · ·
Affination and melting Carbonatation and filtration Decolourisation Evaporation and crystallisation Separation and drying.
Affination and melting Raw sugar is made up of sugar crystals that have a thin film of impurities on the surface. It is mixed with hot (60ëC) impure syrup that softens the film. This mixture, known as magma, is a dark brown viscous mass. Magma is fed into centrifuges, which can spin at up to 1,050 rpm. This centrifugal force separates the raw crystals from the impure raw syrup. The sugar is sprayed with water for a few seconds during the spinning, which helps to remove the last traces of impure syrup. The clean raw sugar crystals from these centrifugal machines are then dissolved in water to produce raw melter liquor. Carbonation and filtration Milk of lime (slaked lime) is mixed with the raw melter liquor. Carbon dioxide gas is bubbled through the mixture. The carbon dioxide reacts with the lime to form chalk (calcium carbonate), which attracts the waxes, gums, resins and other impurities in the liquor at a temperature of 80±85ëC. This is then filtered which removes the chalk and about half of the colour, together with virtually all the fine debris and insoluble matter. The resulting pressed liquor is then ready for the next stage, Decolourisation. The chalk by-product can be used on agricultural land as a soil improver. Decolourisation The pressed liquor from carbonation is passed through tanks containing anionic resins. The first tanks contain acrylic resin, which remove larger colour molecules with subsequent tanks containing styrene resin, which removes the smaller colour molecules. The liquor can be further polished by passing it through a column of granular active carbon from a mineral source, after which the total colour removal exceeds 90%. Evaporation and crystallisation All the previous refining processes have been carried out on liquors that are 60± 67% concentration. An evaporator is used at this stage to remove some of the water prior to crystallisation. The resulting syrup is about 74% solids. This is fed
140
Optimising sweet taste in foods
into large vacuum pans where the sugar crystals are grown. The syrup is heated indirectly by steam to about 80ëC where it boils due to the vacuum applied to the vessel. The use of a vacuum and the resulting reduced temperature helps to minimise the creation of colour during the process. Separation, drying and storage. The mixture of sugar crystals and liquor (massecuite) is centrifuged to separate the white sugar crystals. The separated liquor, which still contains significant amounts of sugar, is sent to a second and then a third crystallisation stage, to ensure that the yield is maximised without compromising quality. A further fourth and fifth crystallisation can be applied for the production of sugars used mainly by industrial customers. The separated white sugar crystals discharged from the centrifuges still contain up to 1% moisture. This is removed by passing the sugar either into a rotating twostage dryer or a fluidised bed dryer, through which filtered, heated air is passed. The moisture level of the sugar at this stage is about 0.06%. Further drying occurs during conveying to the silos or packing areas and in conditioning silos.
7.3
Sugar products
The principal product of sugar manufacture from either sugar cane or sugar beet is white granulated sugar. White granulated sugar (white sugar or sugar) is the main commercial product of trade and is specified in several regulatory texts (EU Sugars for human consumption, USA Food Chemical Codex and Codex Alimentarius). The quality of white sugar is determined by its colour defined typically in ICUMSA units (International Commission for Uniform Methods of Sugar Analysis), sucrose content measured by polarisation, water, ash and reducing sugar content (see Table 7.1). Its mean particle size is usually between Table 7.1 Quality criteria for white sugar according to A. Codex Alimentarius, B. EU Council Directive 2001/111/EC and C. USA Food Chemical Codex Quality factor Sucrose content polarisation in ëZ Conductivity ash % by wt Invert sugar % by wt Loss on drying % by wt Colour (ICUMSA Units)
A
B
C
99.7 (min) 0.04 (max) 0.04 (max) 0.1 (max) 60 (max)
99.7 (min) Not specified 0.04 (max) 0.06 (max) 45 (max)
99.8±100.2 0.15 (max)* 0.1 (max) 0.1 (max) 75 (max)
* expressed as residue on ignition A = Codex Standard for Sugars. Codex Stan 212 ± 1999 (Amd 1-2001) B = Council Directive 2001/111/EC of 20 December 2001 relating to certain sugar intended for human consumption C = Food Chemical Codex 21 CFR part 184.1854
Sucrose 141 Table 7.2
Comparison of the particle sizes of screened and icing sugars UK sugars
Name
American sugars*
Particle size (m) Range Mean
Coarse Medium Fine Standard granulated Extra fine
800±2200 1100±1200 650±900 750 500±700 600 100±1000 500±700 200±600
360±440
Caster Powdered Icings
150±450 10±250 1±100
250±300 50±75 10±25
Name ConfectionersAA Coarse granulated Sanding Manufacturer's granulated Fine granulated Extra fine granulated Bakers special Powdered Icings
Particle size (m) Range Mean 400±1700 1100±1200 300±1200 650±750 200±900 600±700 100±800 450±550 100±600 100±500
320±420 300±350
100±400 10±150 1±80
200±300 50±75 10±20
* Adapted from data in Junk and Pancoast (1973)
400 and 700 microns and has a normal distribution about this mean, at this point the particle size range is quite wide and is a result of the natural crystallisation process. Granulated sugar can be delivered in a wide range of pack formats from small bags 1, 2 and 3 kg, 20 and 25 kg bags, 1 tonne Big Bags up to bulk tankers. Granulated sugar is typically the starting point for other sugar products: · Screened sugars are produced by passing granulated sugar over and through a series of defined aperture screens to provide a selected particle size range. Each of these sugars will have a different name and particle size depending on the country of origin (see Table 7.2). · Icing sugars are produced by milling granulated sugar to a fine powder. Typical mean particle sizes for these sugars can be from 8 microns to 50 microns depending on the potential application required. Due to their finely divided nature they are very prone to caking and lumping and to provide a usable shelf-life the sugar processor will usually add anti-caking or flow agents e.g. tri-calcium phosphate, sodium aluminium silicate, cornflour or other suitable additives. · Liquid sugars are produced by dissolving granulated sugar in potable water. They are usually delivered in bulk tankers or 1 tonne IBCs (Intermediate Bulk Containers) at 67ë Brix (Brix = g/100 g solution). Liquid product streams can also be produced direct from cane sugar refining after suitable decolourising and ion exchange treatments. · Invert syrups can be produced from liquid sugars by treatment with acids or enzymes. · Brown sugars are partially processed cane raw sugars or can also be prepared from white sugars (usually beet) by the admixing of cane molasses.
142
Optimising sweet taste in foods
7.4
Properties of sugar
Sugar is a multifunctional ingredient that contributes to the characteristics of a wide range of food products. This functionality in each product is derived from the physical properties of sugar and its many reactions and interactions with the other food ingredients present. The functionality is usually derived from sugar in solution. There are many excellent chapters in books on sugar solution properties (e.g. Reiser et al., 1995; Heitz, 1998), older data has been collected in one text by workers (Mageean et al., 1991) at Leatherhead Food Research Association (LFRA) (now Leatherhead Food International) this section highlights some key properties which influence finished product properties. The reader is referred to the above texts if more detailed information is required. 7.4.1 Solubility and crystallisation Solubility of sugar in water is of fundamental importance in defining the supersaturation, or driving force of sugar crystal growth. These properties influence the crystallisation of sugar on an industrial scale and indirectly can determine food product characteristics, e.g. fondants and fudge. A molecule of sucrose (sugar) has eight hydroxyl groups, three hydrophilic oxygen atoms and 14 hydrogen atoms. These contribute to the ready formation of hydrogen bonds with water molecules, thus sugar is easily and readily soluble in water. However, sugar does not dissolve in non-polar solvents. Solubility of sucrose is influenced by the presence, both amount and type, of impurities (soluble non-sugars) and temperature. As the temperature increases so does the solubility of sugar and similarly an increase of impurities will also result in increased sucrose solubility. Again both these factors are important in the manufacture of sugar (i.e. crystallisation from impure solutions) and the production of textural properties in food products (e.g. sugar `doctors' in confectionery manufacture, see Section 7.5.2). Several empirical equations have been derived to fit the experimental data for the solubility of pure sugar in water. In 1962 Vavrinecz (Vavrinecz, 1962) carried out a statistical evaluation of the data from 25 authors and derived equation 7.1 which was adopted by ICUMSA in 1978. DS 64:447 0:08222 t 1:6169 10ÿ3 t2 ÿ 1:558 10ÿ6 t3 ÿ 4:63 10ÿ8 t4
7:1
where DS = dry substance (% w/w) and t = temperature (ëC). This equation can be used to prepare a table providing the solubility of pure sugar at temperatures ÿ13ëC to 100ëC (see Table 7.3). The ready solubility of sugar in water and specifically the increased solubility in the presence of non-sugar solutes provide functional properties that can confer benefits for food products. Increased soluble solids can depress the freezing point, which is important in ice cream manufacture. Decreased water activity (aw) at higher solids ensures that sugar has an important role in the preservation of foods sensitive to bacterial spoilage, e.g. jams, preserves and conserves.
Sucrose 143 Table 7.3
Solubility of sugar in water calculated from Vavrinecz (1962)
Temperature (ëC)
g sugar per g water
Dry substance DS (%)
ÿ10 ÿ5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100
1.762 1.784 1.813 1.849 1.893 1.944 2.005 2.074 2.154 2.244 2.345 2.459 2.586 2.728 2.886 3.060 3.252 3.462 3.690 3.937 4.200 4.478 4.764
63.79 64.08 64.45 64.90 65.43 66.04 66.72 67.47 68.29 69.17 70.10 71.09 72.12 73.18 74.26 75.37 76.48 77.59 78.68 79.74 80.77 81.74 82.65
7.4.2 Inversion Most of the reactions of sugar occur in solution, even in relatively dry products the key reactions take place in the liquid phase. Sugar is classed as a nonreducing sugar and is relatively un-reactive in solution apart from the usual solution and colligative properties described previously. Sugar is relatively stable at high pHs and hence most of the sugar manufacturing processes are carried out at alkaline pHs. Buchholz (Buchholz et al., 1998) combined several sources of experimental data to provide a graph of the rate constant of sucrose hydrolysis at different temperatures and pHs. This plot demonstrated that the rate constant of sugar hydrolysis was at a minimum at about pH 8 for temperatures from 60ëC up to 140ëC. Most of the reactions associated with sugar are usually preceded by conversion to its component monosaccharides, glucose and fructose. Glucose and fructose are reducing sugars and are more reactive than sugar. Parker claimed that sucrose is one of the most acid-labile disaccharides known and its hydrolysis to invert is catalysed by heat and low pH (Parker, 1974). The process of conversion/hydrolysis of sugar to an equimolar mixture of glucose and fructose is known as inversion. The mixture is known as invert sugar or invert. The name invert is derived from the change of sign of the optical rotation of the sugars as the reaction proceeds. Sugar has a positive rotation
144
Optimising sweet taste in foods
whereas the mixture of glucose and fructose has a negative rotation of a plane of polarised light (optical rotation); thus old chemists observed this change when sugar was reacted with acid and referred to the newly formed sugars as invert. This reaction is important in the production of industrial syrups ± invert syrups and the reaction can be carried out using mineral acids followed by neutralisation, ion exchanges resins in the acid form or using the enzyme invertase. The formation of invert, i.e. glucose and fructose, is also important for the formation of colour and flavour compounds particularly in baked goods and confectionery. The inversion reaction is also important in beverage formulations as at the low pHs associated with carbonated soft drinks sugar is rapidly converted to invert and thus the sweetening system present in most carbonated beverages is predominantly invert and not sugar. 7.4.3 Colour and flavour formation The role of sugar in colour formation has been investigated for two main reasons ± in sugar manufacture the aim is to reduce colour formation and provide a white crystalline product meeting demanding quality standards (Table 7.1); in food manufacture the ability of sugar to hydrolyse and react with other food ingredients, e.g. proteins and amino acids to produce characteristic colours (and flavours) is actively encouraged and is fundamental for the organoleptic properties of most food products. The mechanisms by which sugar is converted to colour and flavour are similar in both instances. Thermal degradation of sugar at low pH results in caramel formation, in reaction with primary amines (amino acids) melanoidins are formed via the Maillard reaction. Both routes are highly complex reaction pathways resulting in a wide range of low molecular weight intermediates (usually associated with flavour) and high molecular weight polymeric materials (usually associated with colour). Sugar being a non-reducing material does not take part in the Maillard reaction so an important precursor is the hydrolysis to glucose and fructose, which are readily, reacted via the Maillard pathways (Shallenberger and Birch, 1975). The Maillard reaction is not a single reaction pathway but a multitude of reactions of specific types resulting in a complex mixture of many compounds. The complex mixture resulting from the reaction conditions, i.e. reactants, pH, temperature and time combine to deliver the characteristics of a product, notably the smell, taste and colour. The Maillard reaction is characterised by several reaction sequences notably Amadori and Heyns rearrangements to produce 1-amino-ketoses and 2-amino aldoses via nucleophilic substitution of reducing sugars by the amine (or amino acid), followed by a rearrangement. These intermediates react further to produce dicarbonyl compounds giving further fragmentation and condensation products. The reaction intermediates can then take part in Strecker degradation reactions resulting in pyrazines and pyrroles. In sugar manufacture sulphur dioxide can be added at the later stages of processing (evaporation) to inhibit the Maillard reaction and thus reduce colour formation.
Sucrose 145 Caramelisation reactions are degradation reactions of sugar, typically caused by heating solid sugars and solutions, which lead to the formation of brown colours and characteristic flavours (caramels). The reactions proceed via dehydration reactions producing precursors similar to the Maillard reaction (enols). Cyclisation and polymerisation reactions generate higher molecular weight compounds yielding the characteristic caramel properties. The main volatile component of sucrose caramel is 5-(hydroxymethyl)-2-furfuraldehyde. In products where there is limited opportunity for reaction with amino acids caramelisation is thought to be the predominant reaction producing flavour and colour and in situations where there are amino acids/proteins present the Maillard reaction is thought to be the dominant reactive pathway. However, investigations carried out by Imming and co-workers (Imming et al., 1994) concluded that they could find no conclusive proof to distinguish the caramels and melanoidins as different colour types and considered caramelisation products would be incorporated into the complex network of the Maillard reaction. The Maillard reaction and caramelisation are fundamental to the formation of colour and flavour in the majority of food products. The reactions have been the subject of many reviews and conferences. Buchholz et al. (1998) have reviewed the literature on both reactions and determined the relevance to sugar manufacture, more general reviews on the Maillard reaction have been undertaken by O'Brien et al. (1998), Finot et al. (1990) and Ledl and Schleicher (1990).
7.5
Sugar functionality in food products
The functional properties of sugar have been used over many centuries in the preparation of food products. Many of the characteristics of products ± bulk, texture, mouthfeel, flavour and colour are derived from the attributes which sugar delivers. The main sectors where sugar is employed and a breakdown of the percentages of the total volume used in the UK in 2003 are presented in Fig. 7.2. 7.5.1 Beverages Sugar plays an important role in beverages. The sweetness of sugar complements the traditional flavours employed in many soft drinks, for example, citrus and fruit flavours in juice beverages and caramel flavours in colas. The sweetness can also reduce the perceived acidity in carbonated beverages and sugar will contribute to the viscosity and mouthfeel of drinks. The clouds and pulps used to provide visual appeal to drinks are also stabilised in drinks by the viscosity imparted in solution by the sugar. As mentioned in a previous section inversion plays a major role in the sweetness profile of low acidity carbonated beverages. At the usual pH of carbonates, e.g. 2.8±4.0, sugar is rapidly hydrolysed to invert such that the sweetening system present in most commercial
146
Optimising sweet taste in foods
Fig. 7.2 UK sugar usage 2003 (adapted from LMC International data 2005).
carbonated beverages is predominantly invert and not sugar alone. The hydrolysis of sucrose to a mixture of glucose and fructose will also result in an increase in dry substance as a molecule of water will be added across the glycosidic linkage. In several alcoholic beverages, e.g. liqueurs, sucrose and more typically invert syrups are used to deliver the characteristic mouthfeel and moderate the alcohol burn in these products. One characteristic of sugar that is used to good effect in alcoholic beverages is as a fermentation substrate. Sugar is fermented by yeasts to provide alcohol and can be used to prime beer in bottles delivering the characteristic carbon dioxide fizz. 7.5.2 Confectionery In confectionery products sugar provides not only the desired sweetness to many products but also a wide range of functional properties that deliver the characteristic texture, flavour and colour associated with many products. In chocolate, sugar is generally assumed to be an inert ingredient with regard to the subtleties of flavour, contributing only sweetness. However, in dark chocolate, sugar is added for flavour purposes to offset the bitterness of cocoa solids and can also have an effect on processing techniques. The caramelisation and Maillard reactions (see Section 7.4.3) of sugar in milk chocolate, particularly crumb-based chocolates has a great influence on the finished flavour. Crumb based chocolate, most common in the UK, is based on adding sugar to milk followed by evaporation to give a condensed milk. The condensed milk is added to cocoa mass and then vacuum dried to give a powder referred to as crumb. Flavour can also be generated during the conching (mixing) process
Sucrose 147 where heat and vacuum are applied to the chocolate mass to remove undesirable flavours and develop more pleasant ones. The refining process in chocolate manufacture produces the very fine particles associated with chocolate texture. Sugar is co-milled (refined) with the other components of chocolate ± cocoa mass, fats, proteins, etc. Sugar usually exists in a crystalline form and when milled will typically behave in a brittle manner when subject to mechanical stress (milling). In roll refining, sugar crystals pass very quickly into high pressure areas produced by the fine tolerances on the rollers. Under these conditions sugar behaves a bit like sheet metal ± it forms flat sheets of amorphous sugar that has a high surface area and is able to absorb large quantities of the different flavour compounds. During roller refining of chocolate it has been estimated that 30±90% of the sugar becomes amorphous (Niediek, 1981). The subsequent re-crystallisation of the sugar crystals disperses the flavour throughout the chocolate mass. Chocolate made with pre-milled sugar to the correct particle size does not have the same flavour characteristics as chocolate made where the sugar is milled in situ. In other confectionery products sugar contributes to texture and structure of traditional products. In high boil sweets (candies) sugar contributes to the boiling point elevation allowing a mixture to be boiled at higher temperatures. Other sugars historically invert sugar derived from in situ inversion using organic acids (e.g. tartaric or citric acids) but more recently glucose syrups are mixed with sugar to give a mass that does not crystallise. These other sugars known as sugar `doctors' give rise to non-crystallising, super-cooled liquids or glasses typical of high boil sweets. Acids, colours and flavours are added to the hot pass prior to cooling to give the desired product attributes. In fudges, sugar crystallisation is positively encouraged to provide the characteristic short eating texture of these products. In gums and jellies, sugar contributes to the high viscosity required for moulding and setting of the gelling agents. The high solubility of sugar also contributes to the required shelf life of confectionery products and its ready solubility contributes to the flavour release and mouth feel of these products. 7.5.3 Bakery In baked goods, sugar can influence the structure and texture in several ways: sugar is soluble in most dough and batter systems. The amount of sugar present can influence the texture of the final products. This is particularly important in biscuits, for example crackers and water biscuits have a low level of sugar whereas semi-sweet and short dough biscuits can have 15Ð30% sugar. The hardness of biscuits can also be influenced by not only the amount of sugar but also the particle size of the sugar crystals. The larger crystals will take longer to dissolve in the dough during cooking and result in crisper biscuits (e.g. ginger nuts). In cakes and biscuits two of the major influences of sugar are on the gelatinisation of starch and the denaturation of proteins. Sugars increase the gelatinisation temperature of starch and thus delay the setting temperature. The
148
Optimising sweet taste in foods
same is also true for cooling, the gelatinised starch will gel at a higher temperature and thus set the structure of the cake. Protein denaturation has an impact on setting and structure forming and is most noticeable in meringues. The characteristic colours and flavours of baked goods are those associated with the Maillard reaction and caramelisation reactions and these are complemented by using brown sugars and treacles to promote these flavours and colours. Sugar and invert (produced in situ or added) contribute greatly to the humectancy of baked goods and crumb tenderness and soft eat are usually promoted by the use of invert. Interestingly if a standard cake is made using no sugar it typically will not rise and will be dry, bland and tasteless. The caloric value of the cake will also increase by some 25% on a weight basis as the fat present will be at a higher percentage of the total ingredients (Tsang and Clarke, 1988). 7.5.4 Pharmaceuticals The anecdotes `a spoonful of sugar helps the medicine go down' and `sugaring the pill' refer to the sweetening property of sugar and its ability to help the palatability of pharmaceutical preparations. In addition to sweetness, sugar also provides desirable functional properties. The low toxicity, high purity and diverse physicochemical properties of sugar account for its popularity in pharmaceutical applications. Sugar and other ingredients used in the food industry (e.g. starches, bulk sweeteners, thickeners, flavours, etc.) are used in the pharmaceutical and nutraceutical (functional foods) industries as excipients in dosage forms. Typical dosage forms include syrups, suspensions, capsules, tablets and creams. The dosage form acts as a vehicle by which the active ingredient is introduced to the body, and almost invariably contains excipients in addition to the active ingredient. A medicine should be safe and efficacious, and the primary function of formulation is to ensure that the active ingredient is delivered to its site of action at an appropriate concentration. An important aspect of this is to ensure patient compliance, i.e. the medicine is taken as prescribed. For example, many drugs have a bitter taste, so taste masking is an important aid to compliance. Sugar provides the sweetness required for the taste-masking role and thus ensures patient compliance. In addition to the sweetness, sucrose performs many functions in a range of products in the pharmaceutical sector. In the past, excipients were thought of as `inert', but this has now been shown not to be the case. Improper formulation using inappropriate excipients can seriously compromise or even inhibit drug activity. More recently excipients have been found to perform other functions in addition to the traditional ones described above (e.g. taste masking and/or diluent). The correct formulation of excipient and active can provide accurate delivery of the required dose with a reduction of side effects and targeted/controlled release to the site of therapeutic need. The development of particular sugar-based products meeting these specific
Sucrose 149 requirements (e.g. sugar spheres) has expanded the application of sugar products in the pharmaceutical sector. The volumes of sugar used in the pharmaceutical sector are not great when compared to the other routinely used binders and fillers like cellulosics, polyols, starches and lactose. Sugar does have some unique and desirable properties, e.g. sweetness, preservative properties, which are hard to replicate with other products (cf. the food sector) and thus there will always be opportunities for sugar-based products in this sector. The applications of sugar in the pharmaceutical sector have been reviewed by the author of this chapter (Cooper, 2002).
7.6
Future trends
This section is very much a personal view from the author and is written based on many years in the food industry and, hopefully, an informed view of its future. Sugar consumption and production have continued to grow over many centuries. Many of today's food products have been developed around the multifunctional properties of sugar. Over recent decades there has been a shift in consumption patterns from consumption in the home, i.e. table-top, home baking, etc., to greater usage in processed foods. Obesity is growing at an alarming rate both in the developed world and the developing nations. Sugar is being targeted by pressure groups and regulatory bodies in an attempt to reduce the calorie intake of the population. As has been highlighted elsewhere in this book, sugar alternatives (high intensity sweeteners, bulking agents and polyols) have been developed to attempt to replace the functionality of sugar in products without the resultant calories. Some have been more successful than others ± high intensity sweeteners have been successful in a range of diet beverages and over the past 20 years. LMC International data indicates that between 1995 and 2004 UK sugar usage in beverages has fallen by 105,000 tonnes sugar equivalent; over the same period aspartame and acesulfame K volumes have increased by 29,000 and 26,000 tonnes (sugar equivalent) respectively (LMC International, 2005). Polyols, in particular xylitol, have essentially replaced sugar in certain confectionery sectors, for example, in chewing gums it is very difficult to obtain sugar sweetened variants at many retail outlets. Attempts to replace the full functionality in baked goods and chocolate have been less successful. Most food manufacturers are aiming to produce a range of products that offer a choice of calorie intake without compromised taste and quality. It is very unlikely that there will be many new sugar replacers developed over the next decade. The time and cost of development and more particularly the clearance of new and novel food ingredients will inhibit all but the most foolhardy of ingredient companies. The experiences of sucralose will deter many companies from developing new ingredients in this area. The role of sugar and other carbohydrates in healthy nutrition is being highlighted and the recent trend of `low carbs' is likely to be superseded by `slow carbs'. Information on the nutritional impact of carbohydrates will provide
150
Optimising sweet taste in foods
food developers with the tools to design products with controlled responses and targeted/timed nutrition. The physical form of products will also have a role to play in product design and controlled delivery. Paramount in all new product development will be the need for optimal taste and flavour and the role of sweetness in satiety is only just beginning to be explored. In this new era of food development traditional ingredients like sugar will continue to play a role in delivering a balanced diet with un-compromised taste, quality and safety.
7.7
Sources of further information
History of sugar
(1949) The History of Sugar Vol 1, Chapman & Hall, London. (1950) The History of Sugar Vol 2. Chapman & Hall, London. MINTZ, S. W. (1985) Sweetness and Power, The place of sugar in modern history, Elisabeth Sifton Books, Viking, New York. DEERE, N. DEERE, N.
Sugar production/consumption data F. O. LICHTS,
Sugar Year Book 2005 and World Sugar Statistics 2005.
Sugar manufacture and technology
and SCHWARTZ, T. (1998) Sugar Technology, Beet and Cane Sugar Manufacture, Verlag Dr Albert Bartens KG, Berlin. MCGINNIS, R. A. (1982) Beet Sugar Technology (3rd edn), Beet Sugar Development Foundation, Fort Collins. CLARKE, M. A. and GODSHALL, M. A. (1988) Chemistry and Processing of Sugarbeet and Sugarcane. XII. Elsevier, Amsterdam. VAN DER POEL, P.W., SCHIWECK, H.
Sucrose properties
and PARKER, K. J. (1979) Sugar: Science and technology, Applied Science Publishers Ltd, Barking, Essex. MAGEEAN, M. P., KRISTOTT, J. U. and JONES, S. A. (1991) Physical properties of sugars and their solutions, Leatherhead Food RA, Scientific and Technical Surveys No 172 ISSN 0144-2074. MATHLOUTHI, M. and REISER, P. (1995) Sucrose ± Properties and Applications, Chapman & Hall, London. PENNINGTON N. L. and BAKER, C. W. (1990) Sugar. A user's guide to sucrose, Van Nostrand Reinhold, New York. BIRCH, G. G.
Sugar functionality in food products
(2000) The science of chocolate, The Royal Society of Chemistry, Cambridge. BECKETT, S. T. (1998) Industrial chocolate manufacture and use, Chapman & Hall, London. EDWARDS, W. P. (2000) The science of sugar confectionery, The Royal Society of Chemistry, Cambridge. BECKETT, S. T.
Sucrose 151 MANLEY, D. J. R.
Sussex.
(1991) Technology of biscuits, crackers and cookies, Ellis Horwood, West
Sugar associations/trade associations/research associations
The Sugar Bureau: www.sugar-bureau.co.uk The Sugar Association Inc.: www.sugar.org World Sugar Research Organisation (WSRO): www.wsro.org International Sugar Organisation (ISO): www.isosugar.org Sugar Processing Research Institute (SPRI): www.spriinc.org CEFS (Comite EuropeÂen des Fabricants de Sucre): www.cefs.org CEDUS (Centre d'Etudes et de Documentation du Sucre): www.lesucre.com
Sugar companies
British Sugar plc: www.britishsugar.co.uk Tate & Lyle: www.tate-lyle.co.uk Danisco: www.danisco.dk Sudzucker: www.suedzucker.de
Sugar journals
International Sugar Journal, International Media, Port Talbot. Journal of the American Society of Sugar Beet Technologists, ASSBT, Fort Collins. La Sucrerie Belge, SocieÂte GeÂneÂrale des Fabricants de Sucre de Belgique et de la Societe Technique et Chimique de Sucrerie de Belgique, Tienen. Sucrerie Francaise, SEPAIC, Paris. Sugar Industry Abstracts, CAB International, Wallingford. Zuckerindustrie, Verlag Dr Albert Bartens, Berlin.
7.8
Acknowledgements
The author would like to take this opportunity to thank several colleagues who have provided information in specific areas: Simon Houghton-Dodd of Tate & Lyle for the details on cane processing and colleagues at British Sugar for constructive discussions and suggestions on content.
7.9
References
and BRANDES, E. W. (1958) Sugarcane (Saccharum officinarum L.) Agriculture Handbook No. 122, US Department of Agriculture, Washington DC. BEAUCHAMP, G. K. and COWART, B. J. (1987) `Development of sweet taste' in Sweetness, London, Springer, 127±142. BUCHHOLZ, K., BLIESNER, K.M., BUCZYS, R., THIELECKE, K. and MIEHE, D. (1998) in Van der Poel, P.W., Schiweck, H., Schwartz, T., Sugar technology, beet and cane sugar manufacture, Verlag Dr Albert Bartens KG, Berlin, 163±200. COOPER, J. M. (2002) Pharmaceutical applications of sucrose or `does a spoonful of sugar really help the medicine go down?', International Sugar Journal, 104 (1243), 301±305. ARTSCHWAGER, E.
152
Optimising sweet taste in foods
and KATO, H. (1990) The Maillard reaction in food processing, human nutrition and physiology. Birkhauser, Basel/Boston/Berlin. HEITZ, F. (1998) `Sugar' in Van der Poel, P.W., Schiweck, H., Schwartz, T., Sugar technology, beet and cane sugar manufacture. Verlag Dr Albert Bartens KG, Berlin, 70±83. IMMING, R., BLIESNER, K. M. and BUCHHOLZ, K. (1994) The fundamental chemistry of colour formation in highly concentrated sucrose solutions. Zuckerindustrie, 119, 915±919. JUNK W. R. and PANCOAST H. M. (1973) Handbook of Sugar, AVI Publishing Co Inc., Westport, CT. LEDL, F. and SCHLEICHER, E. (1990) Die Maillard-Reaktion in Lebensmitteln und im menschlichen Korper ± neue Ergebnisse zu Chemie, Biochemie und Medizin, Angew Chem. 102, 597±626. LICHT, F. O. (2005) World Sugar Statistics 2005, 66th Edition. LMC INTERNATIONAL (2005) Sweetener Markets for 2015: The Outlook for the Structure of Demand for Caloric and Low Calorie Sweeteners, LMC International, Oxford, UK. MAGEEAN, M. P., KRISTOTT, J. U. and JONES, S. A. (1991) Physical properties of sugars and their solutions, Leatherhead Food RA, Scientific and Technical Surveys No 172 ISSN 0144-2074 NIEDIEK, E. A. (1981) Untersuchungen zum Einfluss der Aromasorption von Zucker auf die GeschmacksqualitaÈt von Schokoladen, Zucker und Susswarenwirtschaft, 34, 44±57. O'BRIEN, H. E., NURSTEN, H. E., CRABBE, M. J. C. and AMES, J. M. (1998) The Maillard reaction in foods and medicine, The Royal Society of Chemistry, Cambridge. PARKER, K. J. (1974) Sucrose as an industrial raw material, La Sucrerie Belge, 93, 15±27. PARKER, K. J., JAMES, K. and HURSFORD, J. (1977) Sucrose ester surfactant ± a solventless process and the products thereof, in Hickson, J. L. (ed). Sucrochemistry, American Chemical Society, Washington DC, 97±114. REISER, P., BIRCH, G. G. and MATHLOUTHI, M. (1995) Physical properties, in Mathlouthi, M., Reiser, P., Sucrose ± Properties and Applications, Chapman & Hall, London. SHALLENBERGER, R. S. and BIRCH, G. G. (1975) Sugar Chemistry, AVI, Westport, CT. TSANG, W. S. C. and CLARKE, M. A. (1988) Chemistry in sugar food processing, in Clarke, M. A. and Godshall, M. A., Chemistry and Processing of Sugarbeet and Sugarcane, Elsevier, Amsterdam. VAVRINECZ, G. (1962) Neue Tabelle u È ber die LoÈslichkeit reiner Saccharose in Wasser. Z. Zuckerind, 12, 481±487. FINOT, P. A., AESCHBACHER, H. U.
8 Polyols M. E. Embuscado, McCormick & Company, Inc., USA
8.1
Introduction
Polyols are reduced-calorie, sugar-free bulk sweeteners. They are sugar alcohols or polyhydric alcohols that are derived from saccharides by the reduction of the aldehyde or ketone group to an alcohol group through chemical or biochemical process. A majority of polyols are naturally occurring substances. Erythritol, sorbitol, xylitol and mannitol can be found in plants (Table 8.1) (Budavari, 1996; Oku and Noda, 1990; Roquette, 2001). These polyols are present in fruits such as plums, pears and grapes and are also found in vegetables. Lactitol, isomalt and hydrogenated starch hydrolyzate (HSH) are synthetic substances and have not been found in nature. Polyols have unique properties that are important in various food and pharmaceutical applications. Although each polyol has different chemical structure and physical properties, polyols as a group of carbohydrates have common characteristics. They are chemically and heat stable. They do not undergo Maillard browning reaction, which is desirable for specific food applications. They are reduced-calorie ingredients and their sweetness is close to sucrose but without the cariogenicity associated with sugars. Because the intensity of sweetness and sweetness profile of polyols are close to sugar, they can be used to replace sugar for bulk and sweetness. This is an advantage of polyols over intense sweeteners that require bulking agents or fillers when used as tabletop sugar replacements. The use of low- and reduced-calorie sweeteners has markedly increased during the last 25 years. Polyols are important reduced-calorie and sugar-free sweeteners. They combine synergistically with other low-calorie sweeteners to provide a more balanced sweetness that is comparable to sugars. Due to these
154
Optimising sweet taste in foods
Table 8.1 Polyol
Natural sources of polyols Fruits
Vegetables
Erythritol Grapes, melons, pears
Xylitol
Strawberries, raspberries, greengages Sorbitol Mountain ash ripe berries, plums, red thorn, pears, cherries, apples, hawthorn Mannitol Honeydew, figs, larches, pineapple
Mushrooms, cauliflower Cabbage
Other sources Wine, sake, soy sauce, miso paste, algae, lichen, grasses, human and animal tissue and body fluids Human metabolic intermediate, hardwood Seaweed, algae, blackstrap molasses, human metabolite
Mushrooms, Algae, manna ash, edible olives, asparagus, fungi, seaweed sweet potatoes, carrots
From various sources
unique properties, polyols are used in a number of food applications. More than 2,000 products containing polyols were launched globally between March and August, 2005 (Deis, 2005). This chapter will discuss the physicochemical, functional and physiological properties of polyols. It will also discuss the basic manufacturing process for each polyol. It will discuss various food and pharmaceutical applications of polyols including the advantages and disadvantages of using these reduced-calorie ingredients in food products. Discussion on future trends of polyols will also be included. This chapter will include the polyol information resources on the web such as the websites of government agencies, organizations and companies producing polyols.
8.2 Types of polyols, chemical structures and their manufacture There are three general types of polyols. They are hydrogenated monosaccharides, hydrogenated disaccharides and mixture of hydrogenated saccharides and polysaccharides. Table 8.2 summarizes the groupings of the polyols and their parent reducing sugars or carbohydrates. Thus sorbitol or glucitol, a six-carbon polyol, is the hydrogenation product of glucose, a six-carbon monosaccharide (Fig. 8.1). Erythritol, a four-carbon polyol, is produced through biochemical processes using starch as the initial raw material (Goossens and Gonze, 2000; Kasumi, 1995). It is a relatively new polyol having been commercially produced only a few years ago. The first step in the commercial production of erythritol is the liquefaction of the starch slurry through the use of enzymes. The liquefied slurry
Polyols 155 Table 8.2
Types of polyols
Type
Polyol
Parent reducing sugar
Hydrogenated monosaccharide
Erythritol Xylitol Sorbitol (glucitol) Mannitol Maltitol Lactitol Isomalt Hydrogenated glucose syrups (HGS) Hydrogenated starch hydrolyzate (HSH)
Erythrose Xylose Glucose Mannose Maltose Lactose Sucrose Glucose syrup
Hydrogenated disaccharide Mixture of hydrogenated saccharides and polysaccharides
Number of carbons 4 5 6 6 12 12 12
Starch hydrolyzate
is then fermented using a selected strain of osmophilic yeasts or fungi. The fermentation liquid is filtered and purified, concentrated and undergoes a crystallization process. The final product, erythritol, is available in powder or crystalline forms. Like erythritol, most polyols use starch as the initial material for their manufacture. Sorbitol, sorbitol syrup, maltitol, maltitol syrup, mannitol and hydrogenated starch hydrolyzate (HSH) are all derived from starches. HSH or polyglycitol was developed in the 1960s (Calorie Control Council, 2005). The broad term HSH is used to describe a product derived from the hydrogenation of oligosaccharides and polysaccharides produced through partial hydrolysis of starch. The starch can come from corn, wheat or potato. HSH containing one polyol as a major component is named after that polyol. For example, an HSH containing 50% or more of sorbitol is called sorbitol syrup and that with more than 50% maltitol is called maltitol syrup. HSH does not contain a specific polyol as a major component. Hydrogenated glucose syrup (HGS) contains sorbitol, maltitol, maltotriitol and hydrogenated polysaccharides. It is produced by the hydrogenation of high
Fig. 8.1
Hydrogenation of glucose to sorbitol.
156
Optimising sweet taste in foods
maltose-glucose syrup, which is prepared through a double enzymatic hydrolysis of starch. Xylitol, on the other hand, is a product made through the hydrolysis of wood, almond shells, straw, corncobs, sugar cane bagasse and other materials rich in xylan. The xylose produced from hydrolysis of xylan undergoes hydrogenation to produce xylitol. The liquor is then filtered, purified and concentrated. Xylitol is crystallized from the concentrated liquor. Maltitol is produced through the catalytic hydrogenation of maltose. Maltose is derived from the liquefaction of starch slurry into a low dextrose equivalent product. This is then enzymatically hydrolyzed into maltose syrup. The maltose syrup is purified and concentrated and then undergoes hydrogenation using nickel or other transition metal as catalyst. The syrup is purified, concentrated then maltitol is crystallized out from the concentrated liquid. Beet or cane sugar is the material used for the production of isomalt. The sucrose from these materials is first enzymatically rearranged to isomaltulose by an enzyme system present in Protaminobacter rubrum bacteria (Bornet, 1993). Isomaltulose then undergoes catalytic hydrogenation to produce isomalt. Lactitol is prepared through hydrogenation of lactose from whey. Lactose is known as the milk sugar. Lactitol was first discovered in 1920 and was first used in foods in the 1980s (Deis, 2000a). Different types of lactitol are produced depending on the conditions of crystallization. The anhydrate, monohydrate and dihydrate forms of lactitol are all available commercially. The chemical structures of selected polyols are shown in Fig. 8.2.
Fig. 8.2
8.3
Chemical structures of selected polyols.
Physicochemical and functional properties
Polyols have different physicochemical and functional properties (Tables 8.3 and 8.4) (Bornet, 1994; Calorie Control Council, 2003, 2005; Cerestar, 2005a,b; Danisco, 2005a,b; Griffin and Lynch, 1993; Olinger and Velasco, 1996; Palatinit, 2005a,b; Purac, 2005a,b; Roquette, 2001; SPI Polyols, 2001, 2005). In spite of these differences polyols have a number of common beneficial characteristics. They are reduced-calorie ingredients and have about 40% caloric values lower than sucrose (Table 8.3). Although their sweetness intensity is lower than sucrose, their sweetness profiles are similar to sucrose (Table 8.5).
Table 8.3
Properties of polyols as compared with sucrose
Carbohydrate
Molecular weight
Caloric value (kcal/g)
Melting point (ëC)
Glass transition (ëC)
Heat of solution (J/g)
Solubility (g/100g solution @ 20ëC)
Erythritol Xylitol Sorbitol Mannitol Maltitol Isomalt Lactitol Dihydrate Anhydrate HSH Sucrose
122 152 182 182 344 344 344
0.2±2.6 2.4±3.0 2.6±3.0 1.6±2.0 2.0±3.0 2.0 2.0
ÿ53.5 ÿ46.5 ÿ43.5 ÿ40.0 ÿ34.5 ÿ35.5
ÿ182 ÿ153 ÿ111 ÿ121 ÿ23 ÿ39 ÿ53
40 63 75 20 63 24.5 56
342
2.3±3.4 4.0
121.5 92±96 101 166±168 150 145 121±123 76±78 145±150 n/a Decomposes at 160±186
ÿ32.0
ÿ18
66.7
From various sources
158
Optimising sweet taste in foods
Table 8.4
Chemical properties
Carbohydrates
Heat stability
Acid/alkali stability
Chemical/enzymatic stability
Erythritol Xylitol Sorbitol Mannitol Maltitol Isomalt Lactitol Sucrose
>160ë >160ë >160ë >160ë >160ë >160ë >160ë Decomposes at 160±186ë
pH 2±12 pH 2±12 pH 2±12 pH 2±12 pH 2±12 pH 2±12 pH >3 Hydrolyzes at acidic or alkaline pH
Stable, no browning reaction Stable, no browning reaction Stable, no browning reaction Stable, no browning reaction Stable, no browning reaction Stable, no browning reaction Stable, no browning reaction Undergoes Maillard browning in the presence of acid, caramelization, hydrolyzed by enzymes
From various sources
Sugars and polyols are the only substances that elicit higher perception of sweetness and give purer and better sensation of sweetness than other sweeteners (Birch, 1999). Sweetness is one of the five basic tastes and is perceived as a pleasurable, pleasant taste. Table 8.5 shows the sweetening power, description of sweetness and other flavors associated with polyols as compared with sucrose. All polyols have been described to have a clean, natural and pleasant sweet taste similar to sucrose. All have lower sweetness levels than sucrose except xylitol and maltitol, which are the closest to sucrose in terms of sweetness intensity. Xylitol has a sweetness profile quite similar to sucrose and its sweetness level is from 87 to 100% that of sucrose. However, xylitol produces an intense cooling sensation making it undesirable as a tabletop sugar replacement. The cooling effect of maltitol has been described as similar to sucrose. It has a sweetness of about 85±95% of sucrose. Next to xylitol, mannitol and sorbitol have significant cooling effects. They are approximately 50±60% as sweet as sucrose. Erythritol is less sweet than sucrose and has the lowest heat of solution and yet it is not perceived as having as intense a cooling effect as xylitol. This is due to its low water solubility. The cooling effect of lactitol is slightly greater than sucrose while maltitol, isomalt, HSG and HSH have negligible cooling effects. The hygroscopicity of polyols is variable depending on form. For example, lactitol anhydride is hygroscopic while the monohydrate and dihydrate forms are nonhygroscopic like sucrose. Together with mannitol, lactitol monohydrate and erythritol are the least hygroscopic polyols. Maltitol in crystalline form is also nonhygroscopic. Isomalt has low hygroscopicity at 85% relative humidity and a temperature of 25 ëC. Sorbitol, on the other hand, is hygroscopic and while sorbitol and mannitol are isomers they have different hygroscopicity. The difference in molecular configuration affected their properties. While sorbitol is hygroscopic, mannitol is not. Sorbitol is used as a humectant while mannitol is used as an excipient. Figure 8.3 shows the relative hygroscopicity of sorbitol
Polyols 159 Table 8.5
Sweetness and taste of polyols
Carbohydrates*
Sweetening power (Sucrose = 1.0)
Erythritol
0.53±0.70
Xylitol
0.87±1.00
Sorbitol Mannitol Maltitol
0.50±0.60 0.50 0.85±0.95
Isomalt
0.45±0.60
Lactitol
0.30±0.40
HSH
0.30±0.40
Clean sweet taste, no aftertaste, less sweet than sucrose Pleasant sweet taste
1.00
Pleasant pure sweet taste
Sucrose
Description of sweetness
Other taste observed
Sweetness profile similar to sucrose with slight acidity and bitterness, no aftertaste, clean sweet taste Sweetest polyol, pleasant sweet taste Pleasant sweet taste Pleasant sweet taste Sweeter than other polyols, same sweetness as sucrose Sweet taste like sucrose, no aftertaste
Cooling sensation
Intense cooling sensation Cooling effect Cooling effect Low cooling effect similar to sucrose Mask bitter metallic aftertaste of other sweeteners, no cooling effect Cooling effect slightly stronger than sucrose Blend well with flavors, mask unpleasant offflavors
*10% in water at 20ëC From various sources
compared with maltitol, erythritol and mannitol. The moisture sorption isotherms were determined using an SGA-100 Symmetric Water Sorption Analyzer manufactured by VTI Corporation, Florida, USA. Based on this graph, sorbitol is most hygroscopic followed by maltitol. Erythritol and mannitol are the least hygroscopic. The most hygroscopic among all the polyols is xylitol. Solubility and hygroscopicity affect processability, shelf life, mouth feel and flavor release of food products. The selection of a polyol for a specific application will depend on these properties. Polyols are very stable to chemical and enzymatic attack. Unlike sucrose, these substances do not undergo molecular changes when heated above their melting points. Sucrose decomposes upon heating close to its melting point. Lactitol is the only polyol that is affected by heat. It is partly converted into lactitan, sorbitol and lower polyols when heated to high temperatures of 179± 240ëC (Mester et al., 2001). Polyols have high chemical stability and do not undergo Maillard browning like glucose and fructose (Table 8.4). Polyols do not react with amino acids to produce caramel-colored Maillard reaction products. They are stable under conditions used in food processing and thus
160
Optimising sweet taste in foods
Fig. 8.3
Water sorption isotherms of sorbitol, maltitol, erythritol and mannitol.
caramelization does not develop during baking and extrusion processes. Polyols are also not susceptible to enzymatic attack and cannot be metabolized by microorganisms such as S. mutans and other oral bacteria. This is the reason why polyols are considered as tooth friendly.
8.4
Physiological properties
Polyols have 25 to 95% less calories than sucrose (Table 8.3). Erythritol has the lowest caloric value while xylitol and HSH have the highest caloric values among the polyols. There is great demand for low-calorie and reduced-calorie ingredients such as polyols since the consumption of low-calorie, sugar-free food and beverages in the United States has more than doubled since 1984 (Fig. 8.4). This trend in consumption by consumers from other industrialized nations will likely be similar. The American Dietetic Association (ADA) has included the following conclusion in their 2004 position statement: Foods containing polyols can be labeled as sugar-free because they replace sugar sweeteners. They also contain less energy than sugars and have other potential health benefits (e.g., reduced glycemic response, decreased caries risk, prebiotic effects). The additional benefits of using of polyols as sweetener substitutes lie on their physiological properties. These are their noncariogenic and nonacidogenic
Polyols 161
Fig. 8.4
Consumer use of low-calorie, sugar-free food and beverages in the United States. Source: Calorie Control Council (2005).
properties. Polyols cannot be used as energy sources of oral microorganisms and thus these microorganisms cannot produce acid or glucan in the mouth. Acid and glucan have adverse effect on the tooth enamel. The noncariogenecity and nonacidogenicity of polyols make them important ingredients in tooth friendly products. The health claim `does not promote tooth decay' has been approved by the US Food and Drug Administration and can be used for sugar-free foods and beverages sweetened with polyols. The American Dental Association also released an official statement that the sugar-free foods do not promote dental caries. In Europe, the Scottish Intercollegiate Network (SIGN) was formed in 1993 to improve the quality of health care for patients in Scotland. SIGN has included xylitol recommendations to prevent dental caries. Lactitol has also been found to have prebiotic effects (Mester et al., 2001). Studies show that lactitol favors the growth of saccharolytic bacteria and inhibits the growth of proteolytic bacteria. Proteolytic bacteria produce ammonia, carcinogens and other substances harmful to the body. Polyols also produce low glycemic index and reduced insulin response making them suitable sweeteners for diabetics and for use in low carbohydrate diets. Glycemic index is the potential of a substance or food to increase blood glucose. For foods, those containing carbohydrates that breakdown quickly during digestion have high glycemic index while those that breakdown slowly and release glucose slowly into the bloodstream have low glycemic index. The Sidney University Glycemic Index Research Service has developed a website
162
Optimising sweet taste in foods
containing the glycemic indices of numerous substances and foods. The website address is www.glycemicindex.com. Sucrose (25 g test portion) has a glycemic index of 64 while glucose has a glycemic index of 103. Isomalt and lactitol have glycemic index of ÿ1 to 3 while xylitol and maltitol have a glycemic index of 7 and 30, respectively. The glycemic index of polyols is considered low. Although polyols are safe and non-toxic, there are limits to their daily intake. The maximum limit is determined by the laxation threshold in grams per day. Erythritol, maltitol, and HSH have a laxation threshold of 100 g/day or higher. This means that if an individual uses 100 g/day or less of these polyols, he will not experience abdominal discomfort such as bloating, abdominal cramps or diarrhea. Sorbitol, xylitol and isomalt have laxation threshold of 50 g/day while lactitol has laxation threshold of 20±50 g/day. Mannitol has the lowest laxation threshold (20 g/day). The higher the laxation threshold, the higher the amount of polyol the individual can consume without any abdominal discomfort. Thus erythritol, maltitol and HSH are well tolerated while mannitol is the least tolerated of the polyols due to its low laxation threshold. The abdominal symptoms when polyols are eaten will depend on the person's sensitivity and other foods he has eaten at the same time. If an individual has taken in some cough syrup containing sorbitol, he should consider eating less of a product with sorbitol because the effect will be additive.
8.5
Advantages and disadvantages of using polyols
There are a number of advantages in using polyols in food and pharmaceutical applications. Polyols are reduced-calorie, sugar-free ingredients and their usage promotes a positive role in diets (Calorie Control Council, 2005). They are excellent sugar substitutes because they possess a clean and pleasant sweetness profile similar to sucrose (Alonso and Setser, 1994; Birch, 1999; Calorie Control Council, 2005). They are extremely stable to heat, enzymes and chemical degradation. A number of polyols are naturally occurring substances. Several polyols have been in use by the food and the pharmaceutical industries for more than 50 years. Xylitol, for example, was first synthesized in 1891 by Emil Fischer and his associates and the preparation of xylitol crystals was first accomplished in 1942 (Budavari, 1996). Sorbitol, on the other hand, was first isolated by Joseph Boussingault from mountain ash berries juice in 1872 (Budavari, 1996). It was first isolated from berries by Embden and Griesbach in 1914 (Budavari, 1996). Sorbitol and mannitol became commercially available when it was produced in 1937 by the Atlas Powder Company, Wilmington, Delaware (now SPI Polyols, Inc.) (Le and Bowe Mulderrig, 2001). Senderens first described lactitol in the literature in 1920 (Budavari, 1996). Karrer and BuÈchi first prepared it in 1937 (Budavari, 1996). Maltitol was only available in syrup form but in 1940, maltitol in powder form was developed.
Polyols 163 Different forms of maltitol have been available since then and have been in use since 1981 (Hirao et al., 1983). In the 1960s, HSH were developed by a Swedish company. They have been in use by the food industry for many years. Erythritol is a relatively new commercial product in the US and Europe although it has been used in Japan since 1990. It was first isolated from algae, lichens and grasses in 1874 and was prepared using Aspergillus niger in 1948 (Budavari, 1996). Polyols have been well accepted as reduced-calorie sweeteners and this is supported by their long history of discovery and usage. They have been incorporated in a number of products. Due to their unique physicochemical and functional properties, polyols typically lengthen the shelf life of some products such as baked goods and candies. Polyols as safe ingredients are supported by numerous documents and studies reported in the scientific literature and international organizations like the World Health Organization. As with any ingredient, there are disadvantages of using polyols in food products. These are cost and digestive tolerance. Polyols are typically more expensive than the nutritive sweeteners such as sucrose, glucose, fructose and corn syrups. In certain cases, polyols are two to three times more expensive than the regular sweeteners. The digestive tolerance for these nutritive ingredients is also typically higher. Nevertheless, as previously discussed, polyols can offer unique properties and functionalities that the nutritive sweeteners cannot offer.
8.6
Applications
Table 8.6 summarizes the different food and pharmaceutical applications of polyols (Bakal et al., 1978; Calorie Control Council, 2003, 2005; Cerestar, 2005a,b; Danisco, 2005a,b; Deis, 2000a,b, 2005; Goossens and Gonze, 1997, 2000; Griffin and Lynch, 1993; Knehr, 2005; Mackay et al., 1978; Olinger and Velasco, 1996; Palatinit, 2005a,b; Purac, 2005a,b; Roquette, 2001; SPI Polyols, 2001, 2005; Turner, 2005; Van Hoef, 1999; Zouilias et al., 2002). The primary use of polyols is as a sugar replacement. Erythritol, isomalt, lactitol and HSH can be used as tabletop sweeteners. Polyols are also used extensively in the preparation of hard and soft candies, chewing gums, mint drops, toffees, fudge, chewy sweets, chocolate and other confectionery products. HSH are used to replace corn syrup in confectionery products where the preparation, heating and handling processes are not changed. An additional advantage of using HSH as corn syrup substitute instead of using other polyols is the lack of crystalline structure, which is favorable for the preparation of confectionery products. This property is also desirable for frozen desserts to inhibit sugar crystal formation. Polyols are not only used as sweeteners in food products. They are also used for their cooling effect and flavor enhancement properties. For example, xylitol and sorbitol are used to enhance the mint flavor perception in chewing gums.
Table 8.6
Applications of polyols
Polyol
Sugar replacement
Food uses
Pharmaceuticals
Erythritol Xylitol
Table top sweetener
Chocolate, bakery products, beverages Chewing gum, mints, gum drops, hard candy, dietary foods for diabetics
Syrups, excipients, oral care products Throat lozenges, cough syrups, children's chewable vitamins, diluent for sachets, toothpastes, mouthwashes Toothpaste, mouthwashes, cough syrups, creams, liquid medicines, parenteral applications, diluent for sachets, tablets Diluent for sachets, lyophilization carrier, tablets, parenteral applications Lozenges, cough syrup and liquid medicines
Sorbitol
Chewing gums, candies, frozen desserts, cookies, cakes, icings, fillings, dietary foods for diabetics
Mannitol
Dusting powder for chewing gums, coatings, dietary foods for diabetics Hard candies, chewing gums, chocolate flavored confectionery, chocolate coatings, pastilles, fondant, jellies, baked goods, ice cream, dietary foods for diabetics Hard candies, toffees, chewing gums, chewy candies, chewing gum coating, chocolates, baked goods, nutritional supplements, low glycemic products Ice cream, chocolate, hard and soft candies, baked goods, sugar-reduced preserves, chewing gums, sugar substitutes Confections, baked goods, hard candies, gum arabic pastilles, chewy sweets, fudges, toffees
Maltitol
Isomalt
Table top sweetener
Lactitol
Table top sweetener
HSH
Table top sweetener
From various sources
Cough drops, throat lozenges
Dentifrices, mouthwashes
Polyols 165 Some polyols can enhance or mask off-flavors. Lactitol and isomalt enhance flavor while HSH masks off-flavors. Sorbitol, on the other hand, enhances the flavor of cooked sausage and improves and stabilizes its color (Le and Bowe Mulderrig, 2001). Polyols are used in foods because of their humectant properties (Cerestar, 2005a,b; Danisco, 2005a,b; Roquette, 2001; SPI Polyols, 2001, 2005; Turner, 2005). Sorbitol and HSH are used as humectants for baked goods and shredded coconut. These polyols are also used as cryoprotectants of seafoods and Surimi products (Roquette, 2001; Cerestar, 2005a,b). They protect the products from freezer burn and at the same time they do not alter their flavor because the sweetness level of these polyols is lower. Nonhygroscopic polyols such as mannitol, erythritol and anhydrous isomalt are used as coating and dusting agents for candies and chewing gums. They are desirable because they prevent the product from becoming sticky and soggy. They are also used in the preparation of chocolate coatings and coatings for ice cream confections. Lactitol is also used because of the crispy texture it imparts to food products. Maltitol is used as a fat replacer because of its creamy mouth feel. It is used to partially replace fat in brownies, cakes and cookies. Maltitol is also used in dietetic foods such as granola bars, ice cream fillings, salad dressings and spreads due to its unique texture. One advantage of maltitol over other polyols is its low cooling effect. Maltitol can also be used as a tabletop sweetener but typically with the addition of aspartame to boost its sweetness. Xylitol has been approved as a direct food additive for special dietary uses and for diabetic diets (Cerestar, 2005a,b; Purac, 2005a,b; Danisco, 2005a,b). It is also used in parenteral nutrition and has been known to reduce dental caries, plaque formation and induce salivary flow which aids in the repair of damaged tooth enamel (Olinger and Pepper, 2001). Due to these benefits, xylitol are popularly used in chewing gums, gumdrops, hard candies and in pharmaceuticals such as lozenges, cough syrups, toothpastes and mouthwashes. Erythritol, the newest polyol in the market, has been approved for use in soft and hard candies, yogurt, chewing gum, low calorie beverages, chocolate, fermented milk, dehydrated foods, and in bakery products such as fat cream in cookies, cakes and pastries and in dietetic foods such as cookies and wafers (Cerestar, 2005a,b; Goossens and Gonze, 1997, 2000). It provides 0.2 calories per gram, the lowest caloric value among the polyols (Calorie Control Council, 2005). Figure 8.5 illustrates the most popular low-calorie, sugar-free products consumed by American adults. The most popular products are diet soft drinks (68%) followed by sugar-free light non-carbonated drinks (63%), sugar-free frozen desserts (48%), ice cream or frozen yogurt (48%), sugar substitutes (46%) and sugar-free gums (41%). Polyols play an important role in the successful formulation of these food products (Cerestar, 2005a,b; Calorie Control Council, 2005; Deis, 2000a,b, 2005; Roquette, 2001; SPI Polyols, 2005).
166
Optimising sweet taste in foods
Fig. 8.5
8.7
Most popular low-calorie, sugar-free products in the United States. Source: Calorie Control Council (2005).
Mixed sweetener potential of polyols
Except for xylitol and maltitol, all polyols have less sweetness level than sucrose. To compensate for the reduced sweetness, polyols have been blended with other polyols or other intense sweeteners to achieve the same pleasant and pleasurable sweetness obtained when sucrose is used. Table 8.7 summarizes the mixtures of polyols with other polyols or with intense sweeteners. The sweeteners in these blends were found to have sweetness synergy. Sweetness synergy is obtained when the sweetness intensity of a mixture of sweeteners is greater than the sum of the individual sweetness intensities. Generally, binary mixtures were found to have enhanced sweetness properties but there were reports that mixing more than two sweeteners can also produce a product with the desired sweetness intensity and profile. To prepare a tabletop sweetener, maltitol can be blended with sodium cyclamate or acesulfame K (Hutteau et al., 1998; Parke et al., 1999) or lactitol
Polyols 167 Table 8.7
Synergy for mixture of polyols and intense sweeteners
Polyol
Polyol
Intense sweetener
Erythritol Xylitol Maltitol Isomalt
Sorbitol, xylitol Sorbitol
Aspartame, acesulfame K
Lactitol
Xylitol, sorbitol, mannitol, maltitol syrup, HSH
HSH
Aspartame Acesulfame K, aspartame, sucralose, cyclamate, saccharin Aspartame, acesulfame K, saccharin, neotame, sucralose Acesulfame K, aspartame, neotame, saccharin, sucralose
From various sources
can be mixed with 0.3% aspartame or acesulfame K or 0.15% saccharine to obtain the same sweetness as sucrose (von Rymon Lipinski and Hanger, 2001). Xylitol was observed to have significant sweetness synergism with other polyols (Olinger and Pepper, 2001). Maltitol was found to exhibit positive sweetness synergy with sodium cyclamate and acesulfame K and negative sweetness synergy with aspartame (Hutteau et al., 1998). Erythritol has been blended with aspartame or acesulfame K at selected sweetness contribution or erythritol-intense sweetener ratios (Austin and Pierpoint, 1998). It was found that erythritol-acesulfame K at 5-95, 85-15, 95-5 and 99-1 ratios were significantly sweeter than the expected sweetness. With aspartame, combinations of erythritol-aspartame at 85-15, 95-5 and 99-1 gave significantly sweeter taste. Different results were obtained when sucrose was used instead of erythritol in the same mixtures. Polyols were found compatible with acesulfame K. These mixture of acesulfame K and a polyol increased the overall sweetness level of the mixture with the final blend having a full and well balanced sweetness (von Rymon Lipinski and Hanger, 2001). This mixture has been used in sugar-free confections and fruit preparations. The use of polyols as bulking agents for intense sweeteners is ideal because they are sugar-free, reduced-calorie ingredients. The typical water-soluble bulking agents such as maltodextrins and corn syrup solids are neither sugar-free nor reduced-calorie ingredients. A variety of fully functional and effective sugar-free, low-calorie sweeteners can only be achieved by blending polyols and intense sweeteners. Acesulfame K was tested in combination with maltitol using the ratio 1:150 (von Rymon Lipinski and Hanger, 2001). The mixture had the sweetness profile closer to sucrose and the aftertaste was significantly reduced. Other successful blends examined were acesulfame K-xylitol (1:100), acesulfame K-sorbitol (1:150-100) and acesulfame K-isomalt (1:250). HSH was also found to blend well with acesulfame K and aspartame to match the sweetness of sucrose in candies. Additional benefit of these blends was the masking of bitter notes coming from the intense sweetener.
168
Optimising sweet taste in foods
For sugarless chewing gums, polyols provide the desired cooling sensation. However, the sweetness intensity of polyols did not provide adequate sweetness level (Bakal, 2001). In these patents, addition of 0.1% to 0.2% saccharin in sugarless chewing gums containing a polyol improved the sweetness quality of the product and provided a long-lasting flavor (Bakal et al., 1978; Mackay et al., 1978). The polyols used were sorbitol, xylitol, maltitol or mannitol. The sweetness level of isomalt is 40 to 55% less than sucrose and blending isomalt with other sweeteners is necessary for some applications to increase its sweetness intensity. Sweetness synergy was acquired when isomalt was blended with other sugar alcohols and other intense sweeteners with the added benefit of masking the bitter metallic aftertaste of some of the intense sweeteners (Wijers and StraÈter, 2001). Kim et al. (2003) conducted a study on the sensory and physicochemical properties of selected sweetener blends. The ingredients used in their study were xylitol, sorbitol, isomalt, aspartame and stevioside. A blend of aspartamesorbitol, aspartame-xylitol and aspartame-isomalt (all at 0.01% to 8.0%, aspartame and polyol, respectively) gave lower bitterness and astringency and higher freshness than polyols mixed with stevioside. This supports previous findings on masking properties of isomalt as well its sweetness synergy with intense sweeteners.
8.8
Regulatory status
Table 8.8 summarizes the regulatory status of polyols as determined by three leading agencies: The Scientific Committee for Food of the European Union, the Joint Food and Agriculture/World Health Organization Expert Committee on Food Additives (JEFCA) and the US Food and Drug Administration (USFDA). The US generally recognized as safe (GRAS) affirmation petition states erythritol is intended for use as flavor enhancer, formulation aid, humectant, nutritive sweetener, stabilizer and thickener, sequestrant and texturizer (Calorie Control Council, 2005). Japan has been using erythritol since 1990. The approval of erythritol is still pending in the Europe Union (EU) (SPI Polyols, Inc., 2005). In the US, xylitol is approved as a food additive for use in foods for special dietary uses and is approved for use in the EU. Products in the US containing xylitol as the sweetener can claim the product to be `noncariogenic'. JECFA has given xylitol an acceptable daily intake (ADI) of `not specified', the safest category in which JECFA can place a food additive (Calorie Control Council, 2005). Sorbitol is GRAS for use as a direct additive to food according to the USFDA regulation 21 CFR 184.1835 (sorbitol) (Office of the Federal Register, 1999a) while mannitol is permitted in food on an interim basis according to CFR 180.25 (mannitol) (Office of Federal Register, 1999b). Foods containing sorbitol and mannitol may include a health claim on their label that these foods `do not promote tooth decay' (Calorie Control Council, 2005). The USFDA requires
Polyols 169 Table 8.8
Regulatory status of polyols
Polyol
Regulations
Erythritol
US FDA ± accepted a petition to affirm the Generally Recognized as Safe (GRAS) status for specific applications US FDA ± use in food is broadly permitted by food additive regulations (21CFR 180.25) Joint Food and Agriculture Organization/World Health Organization Expert Committee on Food Additives (JECFA) ± safe, established an acceptable daily intake (ADI): `not specified' The Scientific Committee for Food of the European Union (EU) acceptable for dietary use US FDA ± GRAS. Label statement required: `Excess consumption may have a laxative effect.' JECFA ± safe, ADI: `not specified' The Scientific Committee for Food of the EU: acceptable for use, no limit set on its use US FDA ± use in food is broadly permitted by food additive regulations (21CFR 180.25). Label statement required: `Excess consumption may have a laxative effect.' JECFA ± safe, temporary ADI of 0±50 mg/kg US FDA ± accepted a petition to affirm the GRAS status for specific applications JECFA ± safe, ADI: `not specified' The Scientific Committee for Food of the EU ± acceptable for use, no limit set on its use US FDA ± accepted a petition to affirm the GRAS status JECFA ± safe, ADI: `not specified' US FDA ± accepted a petition to affirm the GRAS satus for specific applications JECFA ± ADI: `not specified' The Scientific Committee for Food of the EU ± ADI: `not specified' US FDA ± accepted a petition to affirm the GRAS status JECFA ± ADI: `not specified' The Scientific Committee for Food of the European Union (EU) ± ADI: `not specified'
Xylitol
Sorbitol
Mannitol
Maltitol
Isomalt Lactitol
HSH
including the following label for foods whose reasonable foreseeable consumption may result in the daily ingestion of 50 grams of sorbitol: `Excess consumption may have a laxative effect' (Office of Federal Register, 1999a). The same label is required for mannitol (Office of Federal Register, 1999b). JECFA and the Scientific Committee for Food of the European Union have established that both sorbitol and mannitol are safe. The ADI for sorbitol is `not specified' while for mannitol the ADI is set to 50 mg/kg of body weight. Both
170
Optimising sweet taste in foods
sorbitol and mannitol are approved for use in the EU. Maltitol is classified as self-affirmed GRAS in the US and is approved for use in EU. The US petition describes the use of maltitol as a flavoring agent, formulation aid, humectant, nutritive sweetener, processing aid, sequestrant, stabilizer and thickener, surface-finishing agent and texturizer (Calorie Control Council, 2005). JECFA has established an ADI for maltitol of `not specified' (Calorie Control Council, 2005). The EU has found that maltitol is safe and can be used in food products. Isomalt is also classified as self-affirmed GRAS in the US while JECFA established an ADI of `not specified' for isomalt (Calorie Control Council, 2005). The EU has likewise approved the use of isomalt in food. Like erythritol, maltitol and isomalt, lactitol is also classified as self-affirmed GRAS in the US. In the EU, lactitol is allowed as a sweetener as regulated by the food directive No. 94/35/EC and can be used as a food additive as regulated in the food additives other than colors and sweeteners No. 95/2/EC (Mester et al., 2001). JECFA allocated lactitol an ADI `not specified', the safest food additive category (Calories Control Council, 2005). HSH is one of the five polyols which are self-affirmed GRAS in the US. The use of HSH has been approved by the JECFA and the EU with an ADI of `not specified'.
8.9
Future trends
Polyols are versatile reduced-calorie, sugar-free sweeteners. They are very stable to acidity and alkalinity, heat and chemical and enzymatic degradation. Polyols do not undergo Maillard reaction because they do not have a reducing group. Their sweetness profiles are similar to sucrose. The levels of sweetness of polyols range from 30% to 100% that of sucrose. Some polyols are hygroscopic while others are nonhygroscopic. Some members of the polyol family have intense cooling effect while others have negligible cooling effect. Polyols have been known to have sweetness synergy with other polyols or intense sweeteners. The sweetness level of a polyol can be adjusted by combining it with acesulfame K, for example. These mixtures typically have a well-balanced, cleaner and pleasant sweetness and flavor than the individual ingredient. Thus, the members of the polyol family have a wide spectrum of unique properties. This is indispensable to food scientists, product developers, nutritionists, dieticians and health care professionals who are involved in the formulation of food products, preparation of diets or formulation of dietary programs. Consumption of polyols has additional benefits. They have noncariogenic and nonacidogenic properties. In addition to their reduced caloric value, they have low glycemic index making them suitable ingredients for diabetics. With the obesity problem and aging population in the United States and other industrialized nations, increased concern of people about their general health and dental health problems in developing countries, polyols will continue to play a major role in food and pharmaceutical applications. More studies will be
Polyols 171 undertaken in developing synergistic mixtures of polyols and non-nutritive sweeteners. Manufacturing companies and researchers will focus on finding ways to improve the polyol manufacturing processes to economically produced these products. Other novel uses of polyols will be explored.
8.10
Sources of further information
Table 8.9 summarizes the information resources on the web. It includes the main polyol producing companies throughout the world. The suppliers' websites are unique and each contains a wealth of information about their polyol products. Palatinit, Purac, Roquette and SPI Polyols, Inc. websites contain valuable product information and application bulletins that can be downloaded into your computer. Current news about the company and their products, newsletters, trends, frequently asked questions (FAQ), contact information and their locations worldwide are also found in their websites including electronic requests for literature and samples. The websites of government agencies, universities and organizations also offer important information, which are specifically related to the charge or mission of that office or organization. Information related to regulations can be obtained or verified by visiting these sites. In addition to the websites listed in Table 8.9, other information resources on the web can be found by entering key words using search engines. Almost all agencies, universities and organizations like the World Health Organization, the American Dental Association, the American Dietetic Association and research centers have websites that will provide valuable information on polyols.
8.11
References
and SETSER C (1994), `Functional replacements for sugars in foods', Trends in Food Science and Technology, 5(5), 139±146. AUSTIN CL and PIERPOINT DJ (1998), `The role of starch-derived ingredients in beverage applications', Cereal Foods World, 43, 748±752. BAKAL AI (2001), `Mixed sweetener functionality', in O'Brien Nabors L, Alternative Sweeteners, New York, Marcel Dekker, 463±480. BAKAL AI, WITZEL F, MACKAY DAM and SCHOENHOLZ, D (1978), Long-lasting flavored chewing gum including sweetener dispersed in ester gums and method, US Patent 4,087,557. BIRCH GG (1999), `Modulation of sweet taste', BioFactors, 9, 73±80. BORNET FRJ (1993), `Low-calorie bulk sweeteners: nutrition and metabolism', in Khan R, Low-Calorie Foods and Food Ingredients, London, Blackie Academic & Professional, 36±51. BORNET FRJ (1994), `Undigestible sugars in food products', Am J Clin Nutr, 59(suppl), 763S±769S. BUDAVARI S (1996), The Merck Index, New Jersey, Merck Research Laboratories. ALONSO S
Table 8.9
Polyol information resources on the web
Source
Website
Name
Products
Information available
Suppliers
www.cerestar.com www.cerestarsugarfree.com
Cerestar
Product information, application overview, erythritol and isomalt section
www.danisco.com
Danisco Sweeteners
Sorbitol, maltitol, isomalt, mannitol ± all in powder and syrup, erythritol, xylitol Lactitol, xylitol
www.palatinit.com
Palatinit GmbH
Isomalt
www.purac.com www.roquette.com
Purac Roquette
www.spipolyols.com
SPI Polyols
www.towc.co.jp
Towa Chemical Industry Co., Ltd
www.caloriecontrol.org
Calorie Control Council
Lactitol, xylitol Maltitol and sorbitol ± powder and syrup, xylitol, hydrogenated glucose syrups Sorbitol and maltitol (powder and syrup), xylitol, mannitol, HSH (polyglycitol) Sorbitol, maltitol (both syrup and powder), mannitol, xylitol, lactitol, HSH, HGS
www.fda.gov
US Food and Drug Administration The Sidney University Glycemic Index Research Service
Government agencies, universities, organizations
www.glycemicindex.com
Product information, applications, literature Product properties, applications, manufacture Datasheets, applications, literature Product information, applications, www.maltitol-maltisorb.roquette.com Detailed product information, applications and applications bulletins Product list Information on cutting calorie and fats in diets, achieving and maintaining a healthy weight and low-calorie, reduced fat foods and beverages, low calorie sweeteners, polyols, list of ingredient companies and their products Approval, code of federal regulations Glycemic index of more than 1000 foods and ingredients
Polyols 173 (2003), Polyols in sugar-free and reduced-calorie foods and beverages, Atlanta, Calorie Control Council, www.caloriecontrol.org. CALORIE CONTROL COUNCIL (2005), Reduced calorie sweeteners: polyols, Atlanta, Calorie Control Council. CERESTAR (2005a), Product portfolio, Belgium, Cerestar, www.cerestarsugarfree.com CERESTAR (2005b), Specialty polyols, Belgium, Cerestar, www.cerestarsugarfree.com. DANISCO (2005a), Sweeteners, Copenhagen, Danisco A/S, www.danisco.com DANISCO (2005b), About sweeteners, Copenhagen, Danisco A/S. DEIS RC (2000a), `Polyol as functional ingredients with multiple uses', Food Product Design, 45(9), 418±421. DEIS RC (2000b), `Polyols in confectionery', The Manufacturing Confectioner, 80(10), 53±57. DEIS RC (2005), `How sweet it is ± using polyols and high-potency sweeteners', Food Product Design, 85(10), 57±71. GOOSSENS J and GONZE M (1997), `Nutritional and application properties of erythritol: a unique combination? Part I: nutritional and functional properties', Agro-Food Ind Hi Tech, 3±9. GOOSSENS J and GONZE M (2000), `Erythritol', The Manufacturing Confectioner, 80(1), 71±75. GRIFFIN WC and LYNCH MJ (1993), `Polyhydric alcohols', in Furia TE, Handbook of Food Additives, Ohio, CRC Press, 431±455. HIRAO M, HIJIYA H and MIYAKA T (1983), Anhydrous crystals of maltitol and the whole crystalline hydrogenated starch hydrolyzate mixture solid containing the crystals, and process for the production and use thereof, US Patent 4,408,041. HUTTEAU F, MATHLOUTHI M, PORTMANN MO and KILCAST D (1998), `Physicochemical and psychophysical characteristics of binary mixtures of bulk and intense sweeteners', Food Chemistry, 63(1), 9±16. KASUMI T (1995), `Fermentative production of polyols and utilization for food and other products in Japan', Jpn Agric Res Q, 29, 49±55. KIM Y, LEE J, KIM H, CHOI SJ, SHIN W-S and MOON TW (2003), `Sensory and physicochemical properties of selected sweetener blends containing polyols', Food Sci Biotechnol, 12(2), 151±156. KNEHR E (2005), `Carbohydrate sweeteners provide sweet results', Food Product Design, 85(5), 38±48. LE AS and BOWE MULDERRIG K (2001), `Sorbitol and Mannitol', in O'Brien Nabors L, Alternative Sweeteners, New York, Marcel Dekker, 317±334. MACKAY DAM, WITZEL F, DWIVEDI BK and BASANT K (1978), Long-lasting flavored chewing gum, US Patent 4,085,227. MESTER PH, VAN VELTHUIJSEN JA and BROKX S (2001), `Lactitol: a new reduced-calorie sweetener', in O'Brien Nabors L, Alternative Sweeteners, New York, Marcel Dekker, 297±316. OFFICE OF THE FEDERAL REGISTER (1999a), Code of Federal Regulations, Title 21, Section 184.1835. Washington, DC, US Government Printing Office. OFFICE OF THE FEDERAL REGISTER (1999b), Code of Federal Regulations, Title 21, Section 180.25, Washington, DC, US Government Printing Office. OKU T and NODA K (1990), `Influence of chronic ingestion of newly developed sweetener, erythritol on growth and gastrointestinal function of the rats', Nutr. Res., 10, 987± 996. OLINGER P and PEPPER T (2001), `Xylitol', in O'Brien Nabors L, Alternative Sweeteners, CALORIE CONTROL COUNCIL
174
Optimising sweet taste in foods
New York, Marcel Dekker, 463±480. and VELASCO VS (1996), `Opportunities and advantages of sugar replacements', Cereal Foods World, 41(3), 110±117. PALATINIT (2005a), Isomalt sugar-born sugar-like sugar-free, Germany, Palatinit Germany. PALATINIT (2005b), The isomalt family, Germany, Palatinit. PARKE SA, BIRCH GG, PORTMANN MO and KILCAST D (1999), `A study of the solution properties of selected binary mixtures of bulk and intense sweeteners in relation to their psychophysical characteristics', Food Chemistry, 67(3), 247±259. PURAC (2005a), `Lactitol ± a sweetener with health benefits', Confectioner Production, 14±15. PURAC (2005b), Lactitol and xylitol: functional sweeteners, Netherlands, Purac Biochem, www.purac.com. ROQUETTE (2001), Polyols, France, Roquette FreÁres. SPI POLYOLS (2001), Polyols comparison chart, Delaware, SPI Polyols, Inc., www.spipolyols.com. SPI POLYOLS (2005), SPI polyol product line, Delaware, SPI Polyols, Inc., www.spipolyols.com. TURNER J (2005), `Everybody loves Carbs', Food Product Design, 85(5), 12±23. VAN HOEF R (1999), `Lactitol for sugarfree compressed sweets', The Manufacturing, 79, 103±105. VON RYMON LIPINSKI G-W and HANGER LY (2001), `Acesulfame K', in O'Brien Nabors L, Alternative Sweeteners, New York, Marcel Dekker, 13±30. WIJERS M-C and STRaÈTER PJ (2001), `Isomalt', in O'Brien Nabors L, Alternative Sweeteners, New York, Marcel Dekker, 265±282. ZOUILIAS E, OREOPOULOU V and KOUNALAKI E (2002), `Effect of fat and sugar replacement on cookie properties', J Sci Food and Agri, 82, 1637±1644. OLINGER PM
9 Low-calorie sweeteners S. E. Kemp, Kemps Research Solutions Ltd., UK
9.1
Introduction
Low-calorie sweeteners are compounds that taste sweet and provide no calories, or compounds that have such an intensely sweet taste they can be used in food products at concentrations low enough not to contribute significantly to caloric content. They enable consumers to enjoy palatable, pleasant-tasting food whilst offering benefits over nutritive sweeteners, including assisting in weight control and reducing caries. They are also cheaper than nutritive sweeteners when compared on the basis of sweetening power. Earlier low-calorie sweeteners were discovered by serendipity and later lowcalorie sweeteners were specifically designed. This chapter gives an overview of the 11 low-calorie sweeteners approved for use around the world: AcesulfameK, alitame, aspartame, aspartame-acesulfame salt, cyclamate, neohesperidin dihydrochalcone (NHDC), neotame, saccharin, stevioside, sucralose and thaumatin. D-Tagatose is also included as a low-calorie bulk sweetener. Key developments related to low-calorie sweeteners are reviewed, including health developments, such as health trends, obesity, diabetes and dental caries, and market developments, such as economics, consumption, consumer perceptions, consumer attitudes and new sweetener design. The implications for food product design are discussed, including sweetener blends, flavour modification and functional properties. Future trends suggest that the market for low-calorie sweeteners will continue to grow, driven mainly by consumer health concerns. Recent advances in chemoreception sciences on the genetics, structure and functioning of the sweet and bitter taste receptors are opening new routes to design of low-calorie sweeteners.
176
Optimising sweet taste in foods
9.2
Low-calorie sweeteners
Many compounds have been found to be sweet and have no calories. Only a few are approved for use commercially. None yet meet the criteria for the perfect low-calorie sweetener (O'Brien Nabors, 2001b) (see Table 9.1). Low-calorie sweeteners have many uses (see Table 9.2). Some of these are reflected in the terms used to describe them, such as non-nutritive, intense, high intensity, high potency, alternative and artificial sweeteners, sugar replacers, sugar substitutes and macronutrient substitutes. Table 9.1
Properties of the perfect low-calorie sweetener Properties
· Have a similar taste profile to sucrose, possessing a clean sweetness quality, with no side tastes or odours, and a temporal taste profile with an immediate onset and no lingering aftertaste. · Non-calorific at normal usage levels. · Non-cariogenic. · Safe, with no short- or long-term effects on health either directly or through metabolites where they occur. · Natural in origin. · Commercially available at a competitive price. · Easy to use, e.g. soluble, colourless, odourless, easy to produce, store and process. · Stable under a range of processing and usage conditions, such as low pH and high temperature. · Inert and compatible with a wide range of food ingredients. · Stable on storage. · Provide some bulking effect and mouthfeel. · Biodegradable.
Table 9.2
Uses of low-calorie sweeteners Uses
· Provide a greater range of low-calorie food and beverage products to control calories, carbohydrate and specific sugar intake. · Assist in weight management. · Assist in control of dental caries. · Assist in the management of diabetes. · Enhance palatability of certain foods, such as nutrient dense foods or those for diabetics. · Enhance usability of pharmaceuticals and cosmetics, e.g. improve palatability of medicines. · Provide sweetness in times of sugar shortage, such as during First and Second World Wars. · Assist in cost effectiveness, e.g., by reducing costs over sugar or through sweet taste synergy.
Low-calorie sweeteners
177
9.2.1 Commercially available low-calorie sweeteners Currently, 11 low-calorie sweeteners are approved for use in various countries around the world. D-Tagatose is approved as a low-calorie bulk sweetener. This section gives information on history, overview and regulatory status (Table 9.3), sensory, chemical and physical properties (Table 9.4), safety, metabolic properties and physiological properties (Table 9.5), advantages, disadvantages and applications (Table 9.6) and structures (Fig. 9.1). 9.2.2 Other low-calorie sweeteners There are many intensely sweet compounds, but not all have been or can be widely commercialised. This section lists other naturally-occurring intense sweeteners (Table 9.7) and synthetically-produced intense sweeteners (Table 9.8) (O'Brien Nabors and Inglett, 1986; Kim and Dubois, 1991; Kinghorn and Compadre, 1991; Spillane, 1996; Kinghorn and Compadre, 2001). 9.2.3 Regulation The regulations and the regulatory process for low-calorie sweeteners are complex and vary by country. Gaining approval for a new sweetener can be long and expensive, taking 10±20 years in the US (Broulik, 1996). It is beyond the scope of this chapter to cover the regulatory process in depth. Brief descriptions of the main regulating bodies, legislation and terms are given in Table 9.9. Further details can be found on the websites listed in Section 9.7 and in the following references: Snodin (1991), Richards (1996), Broulik (1996), Tschanz et al. (1996), Shively-Knight (1997), Walker (1999), von Rymon Lipinski (1999), O'Brien Nabors (2001b) and Duffy and Sigman-Grant (2004). 9.2.4 Safety Low-calorie sweeteners undergo rigorous safety testing for approval for use in foods. Assessments are made of such aspects as acute toxicity, chronic toxicity, carcinogenicity, mutagenicity, reproductive toxicity, metabolism and breakdown products, taking into account how much is likely to be consumed, whether certain groups are more vulnerable and the products the sweetener will be used in. Testing establishes the Acceptable Daily Intake (ADI): The estimated amount per kilogramme of bodyweight that a person can safely consume on average every day over a lifetime without risk. The ADI is conservative and is usually 100 times less than the maximum level at which no observable effect occurs on animals, known as the No-Observed-Adverse-Effect Level (NOAEL). Usage levels are set and use is monitored so that consumption does not reach ADI levels. There is a tendency for the media to talk about a single ADI for sweeteners as a class but this is inappropriate. The ADIs of sweeteners are independent, as one sweetener does not alter the fate of another in the body. There is no metabolic or
Table 9.3
General and regulatory information
Low-calorie sweetener
History and overview
Acesulfame-K
Alitame
Regulation (General references and Fry, 2005)
Brand names (Õ or TM)
Permitted in US
Permitted in EU (E number)
Other information
In 1967, Clauss and Jensen accidentally discovered that the class of compounds dihydrooxathiazinone dioxides were sweet. They synthesised different structures in this class, including acesulfame-K in 1973, and it was selected for commercialisation due to ease of production.
Yes (Approved as generalpurpose sweetener)
Yes (E950)
Approval for use in over 100 countries.
Alitame was synthesised in the 1970s as part of a research programme to develop highpotency sweeteners through systematic synthesis of dipeptides with diverse structures carried out at Pfizer Central Research.
Food Additive Petition submitted to the FDA.
No
First approved for Aclame use in Australia in 1993. Approved for use in many other countries.
General references (Also used to provide information for Tables 9.4, 9.5, 9.6, 9.13 and Fig. 9.1) (Jackson et al., 1987; Kim and Dubois, 1991; Duffy and Sigman-Grant, 2004)
Sunett Mayer and Kemper Hermesetas (1991) Gold von Ryman Lipinski Sweet and Safe (1997) Sweetex Plus von Ryman Lipinski Sweet One and Hanger (2001)
Brennan and Hendrick (1983) Federal Register (1986) Glowaky et al. (1991) Hendrick et al. (1996) Auerbach et al. (2001)
Aspartame
The sweet taste of aspartame was accidentally discovered in 1969. It has similar sweetness to that of sucrose and its approval in the 1980s gave a step change in the sensory quality of diet products, particularly diet beverages. Since then, it has become extensively used in thousands of products throughout the world, although it has limited stability.
Yes (Approved as generalpurpose sweetener)
Yes (E951)
Widespread approval.
Canderel Equal NutraSweet Sanecta Sugar Twin Aspartil
Mazur et al. (1969) Beck (1974) Tschanz et al. (1996) Federal Register (1996) Stargel (1997) Butchko et al. (2001)
Aspartameacesulfame salt
First commercially available compound from the class of sweetener-sweetener salts. It has many advantages over using blends of the components, including enhanced sweet taste and stability. It is regarded in the EU and US as covered by existing safety data and regulations for the components, and must be used at levels and in applications specified for the components.
Yes (Approved as generalpurpose sweetener)
Yes (E962)
Clearance is being sought in many countries.
Twinsweet
Fry (1996) Hoek et al. (1999) Fry and Hoek (2001)
Table 9.3
Continued
Low-calorie sweetener
History and overview
Cyclamate (Cyclamic acid and its Na and Ca salts)
First synthesised and accidentally discovered to be sweet in 1937. Cyclamate was initially marketed as a table top sweetener in 1950 and used extensively in soft drinks in blends with saccharin in the 1960s. Cyclamate was classified as a GRAS sweetener by the FDA in 1958 and approved for use in many countries. However, it was subsequently banned in 1970 in the US and other countries as it was implicated as a bladder carcinogen in rats. The study was found to be wrongly interpreted and since then new studies have confirmed its safety. A Food Additive Petition for cyclamate was submitted in the US in 1973 and rejected in 1980. Another Food Additive Petition was submitted in 1982 and is currently being held in abeyance. Cyclamate continues to be widely used in Asia.
Regulation (General references and Fry, 2005) Permitted in US
Permitted in EU (E number)
Other information
No Food Additive Petition is pending.
Yes (E952) Limited levels.
Cyclamate is currently used in over 50 countries.
Brand names (Õ or TM)
General references
Audrieth and Sveda (1944) Price (1997) Bopp and Price (2001)
Neohesperidin DC
NHDC was synthesised and fortuitously discovered to be sweet in 1963. It exhibits double functionality, being both an intense sweetener and flavour modifier. Permitted in the EU as a sweetener in 1994 (EU, 1994) and a flavour modifier in 1995 (EU, 1995a). Recognised as GRAS as a flavour ingredient in 1996.
GRAS as a flavour ingredient at sweetness threshold.
Yes (E959) Permitted as sweetener and flavour modifier in specific applications at low levels.
Other countries have also permitted its use as a flavouring, some without limitation.
Neotame
Designed by Claude Nofre and Jean-Marie Tinti as part of a directed research programme on the discovery of low-calorie sweeteners utilising structureactivity models and is a derivative of aspartame. It is an intense sweetener and flavour enhancer.
Yes (Approved as generalpurpose sweetener)
No
Currently undergoing regulatory review for use as a general sweetener and flavour enhancer in various other countries.
Saccharin (Saccharin acid and its Na, K and Ca salts)
Saccharin and its sweetness was discovered in 1878 by Ira Remsen and Constantine Fahlberg and patented in 1885. It name derives from the latin for sugar: Saccharum. It is the cheapest and most widely used low-calorie sweetener. Saccharin was in commercial production by the 1890s, although it met opposition from sugar producers and received
Yes (Approved as generalpurpose sweetener)
Yes (E954)
Widespread approval.
Citrosa
Horowitz and Gentili (1963) Lindley (1996) Smith et al. (1996) Borrego et al. (1999) Borrego and Montijano (2001)
Nofre and Tinti (1991) Federal Register (1998; 1999) Witt (1999) Stargel et al. (2001)
Hermesetas Natrena Necta Sweet Shapers Saxin Spinkle Sweet Sugar Lite Sweetex Sweet 'n' Low Sweet Twin
Tisdel et al. (1974) Broulik (1996) Kathrani (1997) Pearson (2001)
Table 9.3
Continued
Low-calorie sweetener
History and overview
Saccharin (continued)
criticism for having no food value, but remained approved for use in the US at that time partly through the intervention of President Theodore Roosevelt. Saccharin use increased during World War I when sugar was rationed and was common in the US and Europe as a tabletop sweetener by 1917. During peacetime, it was mainly used by those who could not consume sugars for medical reasons. The use of saccharin became widespread with the introduction of cyclamate, when it was used in blends for improved taste, and with the increased concern over weight loss and dieting.
Regulation (General references and Fry, 2005) Permitted in US
Permitted in EU (E number)
Other information
Brand names (Õ or TM)
General references
Stevioside
Stevioside is the major sweet No constituent of the South American Sold only as plant Stevia rebaudiana Bertoni. herbal It is one of only two commercially preparations available, naturally occurring lowor dietary calorie sweeteners, the other being supplements. Thaumatin. It was first isolated in impure form in the 1900s and was developed as a commercial sweetener in the 1970s in Japan. Japan remains the heaviest user with China and South Korea accounting for most of the rest of consumption. JECFA (1998) and SCF (1999) have determined there is insufficient data to approve it for use as a sweetener, although it has been allocated a temporary ADI (JECFA, 2004).
No Sold only as herbal preparations or dietary supplements.
Japan, Argentina, Brazil, Paraguay and South Korea permit use of extracts from Stevia rebaudiana containing stevioside in foods and beverages, herbal preparations and dietary supplements. In some other countries, it is sold only as herbal preparations or dietary supplements.
Sato Stevia Steviron Marumilon 50 (50% stevioside and 50% rebaudiosides) Marumilon A (as above but also includes glycyrrhizin)
Bakal and Nabors (1986) Phillips (1987) Kinghorn and Soejarto (1991) Ishii (1999) Kinghorn (2001) Kinghorn et al. (2001)
Table 9.3
Continued
Low-calorie sweetener
History and overview
Sucralose
Sucralose resulted from a research programme carried out by Hough and his colleagues, with the support of Tate and Lyle Plc, at Queen Elizabeth College at the University of London in the 1970s, which showed that chlorination of sugar could result in intensely sweet compounds. Sucralose, possessing increased sweet intensity and stability, was chosen as the most promising candidate for commercialisation. It was patented in the UK in 1979, first approved by the FDA in April 1998 and was granted approval as a generalpurpose sweetener in August 1999.
Regulation (General references and Fry, 2005) Permitted in US
Permitted in EU (E number)
Other information
Yes (Approved as generalpurpose sweetener)
Yes (E955)
It is now approved for use in more than 40 countries.
Brand names (Õ or TM)
General references
Splenda
Jenner (1989) Jenner (1991) Jenner (1996) Roberts (1997) Goldsmith and Merkel (2001)
D-Tagatose
is a low-calorie bulk GRAS sweetener with 92% sweetness of status (selfsucrose. It occurs naturally in affirmed). Sterculia setigera gum and is also found in heated cows milk and occurs in other dairy products. The production and use of D-tagatose in food stuffs were patented in 1988 and 1991 respectively. GRAS status was approved in 2001 (FDA, 2001).
No Application for approval submitted in March 2005.
Approved for use GaioÕ tagatose in Korea, Australia and New Zealand
Bertelsen et al. (1999) Bertelsen et al. (2001) Calorie Control Council website (see Section 9.7)
Thaumatin
Thaumatin is a naturally-occurring sweet protein obtained from the arils of the fruits of the West African plant Thaumatococcus danielli (Bennett) Benth. (Marantaceae). It is one of the sweetest natural compounds known, but is more widely used for its flavour-modifying effects.
Yes (E957) Limited levels. Also approved as a flavour ingredient.
First permitted Talin as a food additive in Japan in 1979. Approved as a sweetener in Australia. Has extensive worldwide approval as a flavouring ingredient.
Van der Wel and Loeve (1972) Higginbotham (1986) Van der Wel (1986) Kinghorn and Compadre (1991) Green (1999) Kinghorn and Compadre (2001)
D-Tagatose
GRAS status as a flavour ingredient in a number of food categories.
Table 9.4
Sensory, chemical and physical properties
Low-calorie sweetener
Sensory propertiesab
Chemical and physical properties
Acesulfame-K
· 200 times as sweet as sucrose. · Half as sweet as saccharin, as sweet as aspartame and four to five times sweeter that sodium cyclamate. · An acidic environment slightly increases sweetness compared to a neutral environment. · Sweetness exhibits a rapid onset without lingering. · Reported to have a bitter taste at high concentrations. · Exhibits synergy when blended with aspartame and cyclamate, but not with saccharin.
· 6-methyl-1,2,3-oxathiazin-4(3H)-one-2,2-dioxide potassium salt (C4H4KNO4S) · White crystalline, monoclinic powder. · Dissolves readily in water at room temperature, maintaining neutral pH, with solubility increasing with temperature. Slightly soluble in ethanol. · Almost unlimited shelf-life at room temperature. · Heat stable at temperatures normally used for food additives. · Decomposes at temperatures in excess of 200ëC. · Breakdown products of hydrolytic decomposition under extreme conditions include acetone, CO2, ammonium salts, sulphate and amidosulfonate. · May be analysed using thin-layer chromatographic detection, liquid chromatography and isotachophoresis. · Continuous production is possible allowing large scale production.
Alitame
· Approx. 2,000 times sweeter than a 10% sucrose solution with concentrations in the range of 50 g/ml. · 2,900 times sweeter at threshold concentrations of sucrose. · Sweet taste potency is caused by the amide moiety. · Quality of sweetness is similar to that of sucrose with no off tastes. · Sweetness develops rapidly and lingers slightly. · Exhibits sweet taste synergy when used in combination with acesulfame-K and cyclamate.
· L-aspartyl-D-alanine amide · Crystalline, non-hygroscopic powder. · Very soluble in polar solvents, including water and alcohol, and insoluble in lipophilic solvents. Solubility increases with temperature and as pH deviates from the isoelectric point in aqueous solutions. · Relatively stable in aqueous solution, and is stable for more than a year at pH 5±8 at room temperature. Less stable a pH 2±4 due to hydrolysis. · Heat stable. · Can react with other food components, particularly high levels of reducing sugars such as glucose and lactose, in heated (semi-)liquid systems, e.g. baked goods, to form Maillard reaction products, which may results in a loss of sweetness.
· In acidic liquids, such as beverages, off-flavours may be produced from reaction with hydrogen peroxide, sodium bisulfite, ascorbic acid and some types of caramel colour below pH 4. Aspartame
· 200 (160±220) times as sweet as sucrose. · Clean sweet taste, with a slight delay in initial onset of sweetness and no aftertaste. · Exhibits sweetness synergy in combination with other sweeteners and can mask bitter flavour at levels below sweetness. · Acts as a flavour enhancer and extender.
· Dipeptide composed of 2 amino acids, L-aspartic acid and the methyl ester of L-phenylalanine; N-L--aspartyl-L-phenylalanine-1-methyl ester; (C14H18N2O5) · Slightly soluble in water, sparingly soluble in alcohol and insoluble in fats and oils. · In aqueous solution, stability depends on temperature and pH, being most stable between pH 3±5. · Very stable in dry conditions, except at extremely high temperatures. · Can withstand high-temperature short-time and UHT processing, such as pasteurisation and asceptic processing. · Unstable due to hydrolysis at high heat, so sweetness may be lost gradually at certain combinations of pH, moisture and temperature. Availability of the aspartyl moiety is important for the stability of aspartame when used alone in some low-moisture applications that contain flavours with high aldehyde content. In liquids and under certain conditions of moisture, temperature and pH, the ester bond is hydrolysed, forming the aspartylphenylalanine and methanol. The former can be hydrolysed to its amino acids components aspartate and phenylalanine. Methanol may also be hydrolised by the cyclisation of aspartame to form its diketopiperazine.
Aspartameacesulfame salt
· 350 times as sweet as sucrose in water · 400 times as sweet as 4% sucrose solution in citrate buffer. · Exhibits the same sweetness as an equimolar blend of the components, equivalent to 60:40 ratio by weight of aspartame to acesulfame-K, as it dissociates into components on dissolution.
· Salt of aspartame and acesulfame formed when two oppositely charged sweetener ions are combined. · Crystalline, non-hygroscopic.
Table 9.4
Continued
Low-calorie sweetener
Sensory propertiesab
Chemical and physical properties
Aspartameacesulfame salt (continued)
· 11% sweeter than equimolar blend on a weightto-weight basis, as it is free from potassium and has a lower moisture content than components. · Sensory properties are the same as blends of aspartame and acesulfame-K, including improved sweet quality, sweet taste synergy and greater sweetness stability.
· Dissolves readily in aqueous solution, dissociating into components immediately on dissolution. Slightly soluble in ethanol. · Stability on dissolution is the same as for individual components. · Stable as a dry solid on prolonged storage such as 60ëC over a year. · Produced by combining aspartame and acesulfame-K in aqueous acid solutions in a trans-salification reaction, with subsequent crystalisation, separation, washing and drying (Fry and Van Soolingen, 1997).
Cyclamate
· 30 times sweeter than sucrose. · Exhibits increasing levels of bitterness and aftertaste with increasing concentration, although this is not a problem at normal usage levels. · Calcium cyclamate is less sweet than sodium cyclamate and off-tastes start at lower concentrations. · Slow onset with some lingering. · Synergistic with acesulfame-K, alitame, aspartame, saccharin, stevioside and sucralose. · Enhances fruit flavours.
· Cyclohexylsulfamic acid (C6H13NO3S) and sodium and calcium cyclamate. · White, non-hygroscopic crystals or crystalline powder. · Cyclamic acid has good solubility in water but limited solubility in oils and non-polar solvents. · Cyclamic acid is a strong acid with a pH of 0.8±1.6 in 10% aqueous solution. · Sodium and calcium cyclamate are strong electrolytes, fairly neutral in character and little buffering capacity, and are sparingly soluble in ethanol. · Cyclamate solutions are stable to high and low temperatures through a wide pH range in the presence of oxygen and other food ingredients. · Produced using sulfonation of cyclohexylamine.
NHDC
· 400±600 times as sweet as sucrose at practical usage concentrations. · Approx. 1,800 times sweeter than sucrose on a weight basis at near threshold concentrations. · Long onset and persistence time, with a lingering menthol or liquorice-like cooling aftertaste.
· 2-O--L-rhamnopyranosyl-4'- -D-glucopyranosyl hesperetin dihydrochalcone (C28H36O15) · A glycosidic flavonoid and part of the flavonoid family of natural substances that occur in all higher order plants. · Has not been found in nature, although other structurally-related dihydrochalcones (DCs) have been found in 20 families of plants including a variety of sweet tea. DCs are open-ring derivatives of flavanones and consist of two C6 rings joined by a C3 bridge.
Neotame
· Reported to display synergism with most other intense and bulk sweeteners (Schiffman et al., 1995). · Reported to display synergy when used with maltol and/or ethyl maltol (Engels and Stagnitti, 1996). · Exhibits double functionality being both an intense sweetener and flavour modifier. · Acts as a flavour modifier and enhancer in sweet and savoury foods at concentrations below sweetness threshold by increasing mouthfeel perception and smoothing or blending the overall flavour profile and reducing bitterness perception (Lindley et al., 1993). · Modifies and suppresses side tastes in other sweeteners (von Ryman Lipinski, 1996). · Sensory properties reviewed in Lindley (1996).
· Off-white, monoclinic crystalline powder that is slightly hygroscopic. · Low solubility in water at room temperature. Solubility increases with temperature and it is highly soluble at high temperatures. · Stable at room temperature. · Hydrolyses in liquid media at high temperature and low pH. However, it can withstand pasteurisation in fruit-based soft drinks. · Stability is sufficient for most reasonable requirements of stability in food processing and storage. · May be analysed using HPLC. · Commercially produced from neohesperidin, which can be extracted from bitter orange and hydrogenated in the presence of a catalyst under alkaline conditions, or narangin, extracted from grapefruit.
· 10,000 (7,000±13,000) times sweeter than sucrose. · 30±60 times sweeter than aspartame. · Clean, sweet taste similar to sucrose with no significant off-taste. · Slower onset time and a longer linger time than sucrose, having a temporal profile close to aspartame. · Also functions as a flavour enhancer.
· N-[N-(3,3-dimethylbutyl)-L--aspartyl]-L-phenylalanine-1-methyl ester. · Derived from and is structurally similar to aspartame, being the Nalkyl derivative. · Better stability than aspartame due to N-substitution which prevents formation of the diketopiperazine derivative, making it more stable in baking and inert to some flavouring agents and reducing sugars. · Solubility is high in ethanol and sufficient in water and increases with increasing temperature. · Stable in dry form and in liquid form over a wide range of pHs and temperatures. · The major degredation product formed at very low levels is deesterified neotame formed by the hydrolysis of the methyl ester group. · Manufactured from aspartame and 3,3-dimethylbutyraldehyde via reductive alkylation followed by purification, drying and milling.
Table 9.4
Continued
Low-calorie sweetener
Sensory propertiesab
Chemical and physical properties
Saccharin
· 300 (200±700) times sweeter than 7% sucrose. · Slow onset of sweetness, with some lingering and bitter/metallic aftertaste. · Calcium saccharin has a shorter, cleaner aftertaste with less bitterness than sodium saccharin. · Synergistic with a number of intense sweeteners but shows little synergism with acesulfame-K or stevioside.
· Available in several forms: acid, ammonium, calcium, potassium and sodium saccharin. · Chemical name for saccharin is 3-oxo-2,3dihydrobenzo(d)isothiazol1,1-dioxide (C7H5NO3S) · Sodium saccharin is the most commonly used form because of its high solubility, high stability and superior economics. · Acid saccharin is a white odourless, crystalline powder. Slightly soluble in water, but sufficiently so to meet commercial needs. Sparingly soluble in ethanol. · Calcium saccharin has a solubility between the acid and sodium saccharin. · All forms are stable as powders and as aqueous solutions over a wide pH and temperature range. · Produced starting with either toluene or methyl anthranilate.
Stevioside
· 300 times sweeter than 0.4% w/v sucrose. · Slow onset of sweetness, with a bitter taste and a bitter/liquorice aftertaste. · Acts synergistically with aspartame, cyclamate and acesulfame-K. · Has flavour masking properties. · Some stevioside analogues, such as rebaudioside A, have better taste profiles than stevioside and a transglycosylated product of stevioside is sold commercially in Japan.
· Pure stevioside is a white crystalline material. · Highly soluble in ethanol but only slightly soluble in water. Reports on solubility in water are conflicting, possibly because it is often sold in extracts that vary in solubility. · Stable at 100ëC in solution at pH 3±9, but decomposes at alkaline pH. Is stable in acid conditions such as those found in acidulated beverages. · Relatively stable under normal elevated temperatures used in food processing. · Stevia rebaudiana, a member of the sunflower family, is native to Paraguay and is grown commercially in China, Brazil, Taiwan, Thailand and Vietnam. Stevioside is extracted from the leaves in aqueous solvent and refined and purified using a number of different methods including selective extraction into a polar organic solvent, decolourisation, precipitation, coagulation, adsorption, ion exchange and crystallisation.
Sucralose
· 600 times sweeter than sucrose. · Similar quality of sweetness to sucrose with no aftertaste and a slower rate of decay. · Synergistic with acesulfame-K and cyclamate.
· 1,6-dichloro-1,6-dideoxy- -D-fructofuranosyl-4-chloro-4-deoxy--Dgalactopyranoside (also known as trichlorogalactosucrose), (C12H19Cl3O8). · White, crystalline, non-hygroscopic powder. · Highly soluble in water and ethanol and has a viscosity similar to sucrose. · Low surface tension and does not cause foaming. · Good stability. Chlorination of sucrose at the 1 and 6 position of the fructose moiety and 4 on the glucose moiety makes it significantly more resistant to acids and enzymatic hydrolysis than sucrose. · Breaks down slowly into its component moieties in aqueous solution at a rate dependent on pH, temperature and time. It has been shown to be stable in carbonated soft drinks at room temperature for up to a year. · Dry sucralose will discolour slightly after storage, with discolouration occurring more quickly at higher temperatures. · Heat stable and can withstand heat-processing conditions typically used in food processing and manufacture. · Chemically inert and unlikely to react with other ingredients in formulations. · May be analysed using HPLC, GLC and refractive index. · Made from sugar using a process involving selective chlorination, in which the reactive 6-position of sucrose is selectively protected, followed by halogenation and then removal of protection.
D-Tagatose
· 92% the sweetness of a 10% sucrose solution. · Acts synergistically with aspartame and aspartame/aceculfame-K blends. · Acts as flavour enhancer and modifier so that sweetness onset times and bitter aftertastes are reduced and some flavours are improved. · Provides viscosity and mouthfeel, so that properties such as thickness and mouth drying, are improved.
· Naturally-occurring ketohexose sugar in which the fourth carbon is chiral and only differs from fructose at the fourth carbon atom. · White tetragonal bipyramidal crystalline non-hygroscopic powder. · Soluble in aqueous solution, with a viscosity lower than sucrose but higher than fructose and sorbitol. · Less stable than sucrose at extreme pHs. · Takes part in the Maillard reaction leading to browning. · Decomposes and caramelises at high temperatures but can be processed at high temperatures for a short time.
Table 9.4 Low-calorie sweetener
Continued Sensory propertiesab
Chemical and physical properties · Produced from lactose. Lactose is converted to galactose and glucose using the enzyme lactase. Galactose is fractionated by chromatography and converted to D-tagatose using isomerisation under alkaline conditions, which is then purified and crystallised.
Thaumatin
a
· 2000 (1600±3000) times sweeter than sucrose. · One of the sweetest natural compounds known. · Very lingering sweet taste with liquorice/ menthol aftertaste. · Acts synergistically with other intense sweeteners and polyols. · Often used at very low concentrations as part of a sweetener blend where it acts as a flavour modifier to improve sweetness profile, masking bitter, metallic and astringent notes, and enhancing mouthfeel. · Has sweet and flavour modifying properties at low concentrations that can be below sweetness levels.
· Naturally-occurring sweet protein with five molecules known: Thaumatin I, II, III, a and b. · Thaumatin I is a protein of 207 amino acid residues of molecular weight 22,209 daltons. · Cream-coloured powder. · Very soluble in water. · Stable to extremes of heat and temperature under typical processing conditions, although it is not stable when baked or boiled. · Reacts with tannins and loses its sweetness (Bakal, 2001). · Reacts with some synthetic colours and food gums (Wells, 1989).
Relative sweetness is the typical measure used to make comparisons of sweetness between compounds. This is the sweetness of a compound compared to sucrose on a weight basis in aqueous solution at standard temperature using a panel of assessors selected to have a good sense of taste. Relative sweetness depends on concentration and generally decreases with increasing sucrose concentration. b Perceived sweetness quality and intensity depends on a number of factors including concentration, temperature, composition, chemical and physical properties of the product, and individual sensitivity of the taster.
Table 9.5
Safety, metabolic properties and physiologic properties
Low-calorie sweetener
ADI mg/kg body weight/day
kcal/g
Metabolic and physiological properties
Safety
Acesulfame-K
0±15 (JECFA, 1990a; Federal register, 1988) 0±9 (SCF, 1985)
0
· Not metabolised by the human body and is excreted intact. · Non-cariogenic.
· Non-toxic and non-mutagenic. Found to be safe.
Alitame
0±1 (JECFA, 1996)
1.4
· 7±22% is excreted unchanged in faeces. · The remainder is hydrolysed to aspartic acid and alanine amide. · Aspartic acid is available for normal metabolism. Hence, alitame is partially calorific, although this is insignificant at typical usage levels (20±200 ppm). · Alanine amide is excreted in urine as glucuronide and sulfone. · Non-cariogenic.
· Alitame has undergone numerous safety tests and has been found to be safe as a dietary component.
Aspartame
0±40 (JECFA, 1980) 0±50 (Federal register, 1984)
4
· Metabolised into aspartate, phenylalanine and methanol, which are normal dietary components. · Although it is caloric, its contribution is negligible at concentrations used. · Individuals with the genetic disease, phenylketonuria (PKU), have decreased ability to metabolise phenylalanine and must restrict phenylalanine intake from all dietary sources, including aspartame. Products including aspartame must be labelled in many countries as containing a source of phenylalanine.
· The safety of aspartame has come under particular scrutiny and it has been associated with a range of conditions including allergic reactions, headaches, seizures, brain tumours, cognitive impairment, mood disorders, behavioural disorders and hyperactivity. Misinformation has been posted on the internet and circulated via email. Aspartame has undergone extensive safety testing in over 200 scientific studies, including post-marketing surveillance studies, and has been confirmed as safe in a variety of sub-groups in humans, including obese individuals, diabetics, lactating females and phenylketonuriacs. Recent critical
Table 9.5
Continued
Low-calorie sweetener
ADI mg/kg body weight/day
kcal/g
Aspartame (continued)
Metabolic and physiological properties
Safety
· Produces limited glycaemic response. · Non-cariogenic. · A detailed overview given in Stegink and Filer (1984) and Tschanz et al. (1996).
reviews of scientific publications and evidence refute associations with the conditions above and conclude that aspartame is safe for use in foods and beverages (Butchko et al., 2002; SCF, 2002; Duffy and Sigman-Grant, 2004). Many professional and regulatory bodies have published statements affirming the safety of aspartame (see www.aspartame.org). · It has been suggested that aspartame increases hunger and food intake but this has not been borne out by evidence, which shows that aspartame facilitates the control of body weight and long-term weight maintenance (see Blackburn (1999).
Aspartameacesulfame salt
Covered by ADIs set previously for component sweeteners. (JECFA, 2000)
4
As it dissociates into its component parts on dissolution without the presence of potassium, exposure is actually to aspartame or acesulfame and it presents no new toxicological issues.
· Considered as component sweeteners and is therefore deemed to be safe at usage levels set for those sweeteners by the SCF and FDA.
Cyclamate
0±11 expressed as cyclamic acid (JECFA, 1982)
0
· Cyclamate is incompletely absorbed and is excreted unchanged in urine. · It can cause diarrhoea at high doses. · In approximately 25% of the population, intestinal microflora convert non-metabolised cyclamate to cyclohexylamine although the rate of
· Safety issues are reviewed in Renwick (1997b). · Cyclohexamine is more toxic than cyclamate and its toxicity limits the use of and is used to set the ADI for cyclamate. · A study published in 1970 erroneously concluded that cyclamate caused tumours in the bladder of rats (Price et al., 1970). Cyclamate was
conversion varies considerably according to the individual, day and number of ingestion of cyclamate (Bopp et al., 1986). Cyclohexylamine is completely absorbed from the gastrointestinal tract and is primarily excreted unchanged in the urine (Bopp et al., 1986), with 1±2% being metabolised to cyclohexanol and trans-cyclohexane-1,2-diol (Renwick and Williams, 1972). · Non-cariogenic.
NHDC
0±5 (SCF, 1987)
Up to 2 · Metabolism is largely carried out by intestinal microflora and results in a range of metabolites, including aglycone, phoroglucinol and dihydroisoferulic acid. 90% of material is excreted in the first 24 hours, primarily in urine. · Caloric value based on the assumption that sugar residues are hydrolytically split and metabolised and aglycone is not extensively metabolised. Contribution is negligible due to low usage levels. · Non-cariogenic.
subsequently removed from GRAS status and banned for use in food and beverages in the US (Federal Register, 1969; 1970) and other countries. Since then at least 14 animal studies have been carried out on the safety of cyclamate and cyclohexylamine and none have found evidence that these compounds are carcinogenic (reviewed in Morgan and Wang (1985) and Bopp and Price (2001)). · Studies on cyclamate in Germany, Australia and Denmark showed a very small number of people who consumed more than the ADI averaged over a seven-day period. All but one case had less than a two-fold increase, which is not toxicologically significant. The remaining case had an intake three times the ADI but had a bizarre consumption profile (Renwick, 1999).
Table 9.5
Continued
Low-calorie sweetener
ADI mg/kg body weight/day
kcal/g
Metabolic and physiological properties
Safety
Neotame
0±18 (Federal Register, 2002) 0±2 (JECFA, 2003)
0
· Rapidly but incompletely absorbed. · Absorbed neotame is completely excreted in urine and faeces. · Major route of metabolism is to deesterified neotame, which is excreted in faeces. Methanol is also formed during this process at negligible amounts compared to safe levels. · A small amount of phenylalanine may be released in the plasma but this is not clinically significant for individuals with PKU. · Non-cariogenic.
· Numerous clinical studies have demonstrated that neotame is safe for human consumption.
Saccharin
0±5 (JECFA, 1993; EU, 1994)
0
· Not metabolised and is excreted unchanged, predominantly in urine. · Non-cariogenic. Inhibits bacterial growth and acidogenesis (Linke, 1987).
· The safety of saccharin has been extensively studied and determined to be safe for human consumption, although there has been controversy over its safety, predominantly due to findings of bladder tumours in some male rats fed high doses of sodium saccharide (Tisdel et al., 1974, reviewed in Cohen, 1997). This led to a requirement in some countries for products including saccharin to carry a warning label that saccharin may cause cancer in rats and it was banned in some countries, including Canada. However, subsequent human studies have found no increased risk of bladder cancer. In 1997, the International Agency for Research on Cancer determined saccharin results in rats are not relevant to man, as the mechanism of carcinogenicity in experimental animals does not operate in humans.
· Intake surveys revealed a problem of high saccharin intake in young children in the UK related to a particular type of soft drink concentrate and advice was issued to dilute concentrates to a greater extent for young children (Renwick, 1999). It has been suggested that caregivers in the US may want to limit intake of saccharin by young children because of the limited amount of data available on use in children (Duffy and SigmanGrant, 2004). Stevioside
0±2 Temporary ADI allocated by JECFA in June 2004 for steviol glucosides (stevioside). JECFA also requested further research data to be submitted by 2007 (JECFA, 2004).
· On oral administration, most stevioside is degraded by intestinal bacterial flora to steviol, steviolbioside and glucose, which are absorbed and further metabolised.
· A review of safety studies is given in Kinghorn et al. (2001). · Stevioside appears not to be toxic, but conflicting studies have been published on its biological effects, for example, effects on fertility and hypoglycaemic effects. The mixed findings in this area may be due to the use of extracts that may contain several active components, rather than the use of pure compounds, and the lack of studies in humans. · JECFA (1998) and SCF (1999) were unable to allocate an ADI as toxicological studies had been carried out using material that was poorly specified, of variable quality and with no information on constituents or contaminants (Walker, 1999). It was deemed that insufficient data was available. JECFA has now allocated a temporary ADI but has requested further information (JECFA, 2004). · Steviol is genotoxic. Many of the safety studies undertaken on stevioside used parenteral administration, which is not relevant to oral ingestion, as stevioside administered in this way is poorly absorbed and rapidly excreted in urine, producing no steviol.
Table 9.5
Continued
Low-calorie sweetener
ADI mg/kg body weight/day
kcal/g
Metabolic and physiological properties
Safety
Sucralose
0±15 (JECFA, 1990b)
0
· Mainly excreted unchanged in faeces. A small amount is absorbed and excreted unchanged in urine. · Non-cariogenic.
· Sucralose has undergone an extensive and thorough safety testing programme and review, with over a hundred studies conducted over a period of more than 20 years, and has been found to be safe for use. An overview of toxicity data is given in Grice and Goldsmith (2000).
D-Tagatose
Not specified, i.e., no need for a tolerance level to be set (JECFA, 2004).
1.5
· Only 20% of ingested D-Tagatose is absorbed in the small intestine and is then metabolised by the liver. The major portion is fermented in the colon by microflora to short-chain fatty acids. · Prebiotic. Promotes the growth of beneficial bacteria, decreases the growth of pathogenic bacteria and increases generation of short-chain fatty acids, particularly butyrate, which is hypothesised to be a key protective component of high fibre diets against colon cancer. · As is incompletely absorbed, it has a low glycaemic effect compared to other bulk sweeteners. · May produce mild flatulence and laxation, similar to other low digestible carbohydrates. · Non-cariogenic.
· Extensive safety and tolerance testing has been carried out and it has been found to be safe (as reviewed by Bertelsen et al., 2001).
Thaumatin
Not specified, i.e., no need for a tolerance level to be set. (JECFA, 1985)
4
· Although it is caloric, its contribution is negligible at concentrations used. · Non-cariogenic (Wells, 1989).
· JECFA have reviewed relevant studies and concluded that it is not toxic.
Table 9.6
Advantages, disadvantages and applications
Low-calorie sweetener
Advantages (General reference: Wells, 1989)
Disadvantages (General references: Wells, 1989; Lindley, 1999)
Applications
Acesulfame-K
· Very stable at a range of temperatures and pH. · Readily soluble. · Synergistic with a wide range of sweeteners.
· Bitter metallic aftertaste, thin mouthfeel. · Poor solubility in alcohol, however solubility adequate in aqueous blends. · Variable compatibility with flavourings.
· Used in a wide range of products, including foods, beverages, oral care products (toothpaste and mouthwash), pharmaceuticals (masks off-taste and flavours) and animal feed. Food products include carbonated, non-carbonated, diluted and powdered beverages, dairy products (yoghurts, ice cream, flavoured milk), baked good (cakes, cookies), jams, confectionery (candy, chocolate), chewing gum, pickled vegetables and table-top sweetener as a tablet, granules, powder and solutions. · Due to its good heat stability: ± Can be added to products before heating and cooking if required, e.g., in confectionery production. ± Stable in processing where heat is required, such as pasteurisation, sterilisation, baking, extrusion and spray drying. ± Suitable for use in warmer climates. · Useful in beverages, due to its prolonged stability at low pH in aqueous solution. Typically used blended with other intense and bulk sweeteners, e.g., aspartame. · Used in fermented milk products as not attacked by lactic acid bacteria and other bacteria. · May require encapsulation in gums due to rapid solubility in saliva.
Table 9.6
Continued
Low-calorie sweetener
Advantages
Disadvantages
Applications
Alitame
· Heat stable. · Readily soluble. · Good sweet taste quality with no off-tastes. · Synergistic when used with acesulfame-K and cyclamate.
· Some instability. · Small loss of sweetness in high-heat applications such as baking. · Compatibility depends on the ingredients present and the thermal and pH exposure during the manufacturing process. · Incompatible with certain beverage ingredients. · Unstable in most caramels.
· Specific applications include hard and soft candies, heat-pasteurised foods, neutral pH foods processed at high temperatures, e.g., sweet baked goods, dairy products, beverages, chewing gum, confections, jams, etc. · Often blended with other sweeteners, including saccharin.
Aspartame
· Good sweet taste quality. · Enhances fruit flavours. · Good compatibility.
· Unstable at neutral pH. · Breaks down under high-heat conditions. · Slow to dissolve, particularly in cold. Needs to be ground finely for quick dissolution but this can lead to clumping and poor flow. · May lead to foaming. · Stability of sweet taste can be a problem, particularly at neutral pH and during prolonged storage of low pH beverages. · Aspartame can react with the flavour causing a loss of both sweetness and flavours and shortening shelf-life, particularly low-moisture applications that contain flavours with high aldehyde content, such as sugar-free confectionery, particularly chewing gum.
· Used in a wide variety of food and beverage products, due to its good sensory properties. · Often used in blends, in which it can mask bitter flavour at levels below sweetness perception. · Acts as a flavour enhancer and extender, particularly to acid fruit flavours. Can be used to extend flavour perception in chewing gums. · Limited use in products involving excessive heat, such as baked goods, although it can survive pasteurisation.
Aspartameacesulfame salt
· Good sweet taste profile when compared to its component sweeteners. · 11% sweeter than blend as purity higher due to recrystallisation production process. · Offers the same advantages as blends of aspartame and acesulfame-K, such as greater sweetness stability, longer shelf-life, sweet taste synergy and improved sweet quality. · Less susceptible to attack by flavour aldehydes in lowmoisture applications as access to the aspartyl moiety is hindered. · Improved dissolution over aspartame. · Dissolution releases equimolar ratio continuously, where as individual sweeteners in blend dissolve at different rates giving a changed sweetness profile until full dissolution. · As it is crystalline, it overcomes the practical problems of creating a blend of powder mixes from the individual sweeteners such as dissolution time, hygroscopicity and homogeneity. · Savings on raw material handling.
· Suitable for a wide range of applications including beverages, dairy products, table-top sweeteners, confectionery including chewing gum and hard candy, and pharmaceutical products, including chewable tablets. · Particularly useful in powdered products. Aspartame and acesulfame-K crystallise in different forms, making it more difficult to create and maintain homogeneous mixtures, which can create a non-homogeneous sweet profile. Improved dissolution over aspartame is important in powdered products, particularly at cold temperatures such as for iced tea. Dissolves in half the time of aspartame at the same particle size. · In chewing gum encapsulation is not needed and it provides extended sweetness with a sweetness boost after 5±8 minutes chewing depending on product formulation and gum manufacturing process. Gives improved storage stability and shelflife as less susceptible to attack by aldehyde-rich flavours, such as cherry and cinnamon. · In hard candy disperses directly with better homogeneity in hot mass than aspartame blends, particularly in non-acid flavoured candy such as mint.
Table 9.6
Continued
Low-calorie sweetener
Advantages
Disadvantages
Applications
Cyclamate
· Good stability, solubility and compatibility. · Exhibits sweet taste synergy with acesulfame-K, aspartame, NHDC, saccharin and sucralose. · Enhances fruit flavours. · Economical.
· Chemical sweet taste. · Used in a wide range of product applications · Slow sweetness onset. including table-top sweeteners, beverages, · Lower relative sweetness, so need processed fruits, confections, chewing gums, salad relatively high use levels. dressings, jams, gelatine desserts, cured meats. · Poor public perception in some · Particularly useful in fruit products as it enhances countries due unfounded association with fruit flavours and can mask sourness of some citrus cancer. fruits. · Not available for use in the US. · Good in pharmaceutical and oral hygiene products as masks bitterness. · Frequently used in blends with saccharin, as it exhibits sweet taste synergy and improved sweet taste quality. · Not so good for baked products as it does not take part in the Maillard browning reaction and requires the addition of bulking agents to achieve the required texture. · The acid form can be used as an effervescent agent.
NHDC
· Good stability and compatibility. · Exhibits sweetness synergy in blends. · Masks bitterness. · Flavour enhancing and modifying properties.
· Slow sweetness onset. · Lingering sweet/liquorice/ menthol aftertaste. · Delayed sweetness onset. · Poor solubility and slow dissolution.
· Used in bakery products, beverages, chewing gum and animal feeds. · Unlikely to be used as sole sweetening agent due to its taste profile but plays an important role as a minor component in sweetener blends in which it contributes less than 10% sweetness. · Acts as a flavour modifier and enhancer in sweet and savoury foods at concentrations below sweetness threshold by increasing mouthfeel perception and smoothing or blending the overall flavour profile and reducing bitterness perception.
For example: ± Masks bitter taste in pharmaceuticals and beverages fortified with vitamins. ± Enhances perception of creaminess in low-fat margarines. Neotame
· Stable. · High intensity sweetness with no significant off-tastes. · Flavour enhancing and modifying properties.
Saccharin
· Good stability · Economical · Synergistic with aspartame, cyclamate, sucralose, alitame, sucrose and fructose.
· Uses include carbonated and powdered soft drinks, hot-packed still beverages, cakes, yoghurt and table-top sweeteners. · Stability has been tested using a designed programme based on a matrix of foods covering a range of the critical parameters of pH, temperature and water content from the food categories comprising the majority of commercial applications (carbonated and powdered soft drinks, cake, yoghurt and hot-packed still beverages). · Can be used alone or as blends. · Also functions as a flavour enhancer, particularly in fruit-based drinks, in which it maintains sourness, masks bitterness even at sub-threshold concentrations and eliminates beany flavours in soy products. · Slow sweetness onset. · Bitter/metallic aftertaste · Thin mouthfeel · Poor public perception due to association with poor taste quality and unfounded association with cancer.
· Used in a wide range of food applications including beverages, table-top sweeteners, processed fruit, chewing gum and confections, gelatine desserts, juices, jams, toppings, sauces and dressings. · Also used in a wide range of non-food applications, including nickel electroplating, pharmaceutical, personal care products, oral hygiene products, animal feeds and as a chemical intermediate.
Table 9.6 Low-calorie sweetener
Continued Advantages
Disadvantages
Applications · Typically used in blends with cyclamate, aspartame, acesulfame-K and sucralose and ternary blends with cyclamate and aspartame, and with sucrose and fructose, to improve its sweet profile, provide synergy and reduce costs. · Can be used as a standalone sweetener in a broad range of applications. Masking agents may be used to mask the bitterness.
Stevioside
· Good stability and compatibility. · Natural.
· Slow sweetness onset · Bitter/liquorice aftertaste · Not available for use in Europe and the US.
· Used in pickles, beverages, liquors, yoghurts, ice cream and sherbets, table-top sweeteners, soy sauces and pastes, dried seafood, seafood, fish and meat preparations, seasonings, confectionery, bread, chewing gum, medications, oral hygiene products and as a dietary supplements as a herbal tea, powdered leaf and liquid extract. · May be used with flavour masking and sweet enhancing agents. In Japan, it is used in combination with glycyrrhizin resulting in improvements in taste quality of both sweeteners and is said to mask the `pungency' of sodium chloride commonly used as a preservative in Japanese foods.
Sucralose
· Sugar-like taste profile. · Good stability, solubility and compatibility. · Exhibits synergy with other sweeteners. · Does not cause foaming.
· Some lingering sweet aftertaste
· Suitable for use in a range of applications including soft drinks, baked goods, desserts and toppings, canned fruits and vegetables, ice cream, dairy products, breakfast cereal, confectionery, table-top sweeteners and beverages. · Can be used in dry food applications with no expectation of discolouration during normal distribution. · Exhibits good stability in baked goods.
D-Tagatose
Compared with other bulk sweeteners: · Reduced-calorie, lowcarbohydrate bulk sweetener that provides texture and has similar sweetening power to sucrose. · Natural. · Prebiotic. · Low glycaemic response. · Soluble. · Good ability to crystallise. · Acts synergistically with aspartame and aspartame/ aceculfame-K blends. · Acts as a flavour enhancer. · Also functions as a stabiliser and humectant.
Compared with other bulk sweeteners: · Unstable at extremes of pH and temperature. · Caramelises and browns more rapidly than sucrose in baked products. · Cannot be labelled as sugar free.
· Can be used in beverages, breakfast cereals and products, soft drinks, frozen yoghurt, ice cream. · Suitable for use in confectionery, due to its ability to crystallise and caramelise, including chocolate, fondants, fudge, caramel, hard and soft candies, and chewing gum. · Enhances mint and lemon flavor in sugar free chewing gum and mints. · Boosts creaminess and toffee flavour.
Table 9.6
Continued
Low-calorie sweetener
Advantages
Disadvantages
Applications
Thaumatin
· Good stability. · Readily soluble. · Natural. · High sweetness potency, so can use at low levels. · Exhibits synergy with other sweeteners. · Also acts as a flavour enhancer.
· Slow sweetness onset. · Prolonged sweet taste with liquorice/ menthol aftertaste. · Incompatible with certain beverage ingredients. · Cannot be used in products to be baked or boiled.
· Often used a very low concentrations as part of a sweetener blend where it acts as a flavour modifier to improve sweetness profile, masking bitter, metallic and astringent notes, and enhancing mouthfeel. · Particularly useful in aiding flavour stability during storage of beverages by maintaining sweetness and masking unpleasant flavours that develop during degredation of lemon flavours. · Used in low- and full-fat yoghurts to improve taste and mouthfeel. · Improves and prolongs flavour and cooling effects in mint flavoured products, e.g., gums. · Used in oral hygiene products, such as toothpaste and mouthwash, to improve the flavour impact and cooling effect of mint products and mask metallic aftertastes caused by the use of saccharin. · Used in pharmaceutical, cosmetics and neutraceuticals for its ability to mask bitter and astringent tastes.
Low-calorie sweeteners
207
Fig. 9.1 Structures. (For information on the structure of thaumatin see Table 9.4). (afrom National Library of Medicine Specialized Information Services).
toxicological reason to consider intense sweeteners as a group as the only property they share is their ability to interact with sweet taste receptors. Therefore, it is safe to allow the use of blends in products (Renwick, 1997a; Walker, 1997). There is a steady increase in the number of low-calorie sweeteners approved for use, the number of countries in which they have approval and the number of products that they are used in. As consumption of low-calorie sweeteners has increased, both in variety and quantity of sweeteners and products containing
208
Optimising sweet taste in foods
Fig. 9.1 Continued
them, concerns have arisen that daily intakes may exceed the Acceptable Daily Intake (ADI). At least 15 intake studies have been performed since the 1970s in different countries across the world on various sweeteners approved at the time of study. These are reviewed by Renwick (1999). He concludes that the current intakes of intense sweeteners do not represent any concern for health. For all sweeteners, the mean and 90th percentile intakes were well below the ADI values.
Table 9.7
Natural intense and/or low-calorie sweeteners
Sweeteners
Chemical group
Origin
Relative sweetening power
Other sensory properties
Other information
Baiyunoside
Diterpene glycoside
Phlomis betonicoides Diels (Labiatae)
500
Very lingering aftertaste.
· The plant is used in Chinese medicine.
Brazzein
Protein
Fruits of the African plant Pentadiplandra brazzeana Baillon (Pentadiplandraceae)
500±2000
More sucrose-like sweet taste than other proteins.
Curculin
Protein
Fruits of Curculigo latifolia (Hypoxidaceae)
Has not been systematically studied
Has a sweet taste that disappears a few minutes after being held in the mouth. A sweet taste subsequently occurs on drinking water.
Glycyrrhizin
Triterpene glycoside
Rhizomes and roots of licorice and other species in the genus Glycyrrhiza.
50±100
Slow onset and lingering licorice aftertaste.
· Also known as glycyrrhizic acid. · Some of its derivatives are also sweet, notably its monoglucuronide. · Partially purified Glycyrrhiza extracts are widely used in Japan for sweetening and flavouring foods, beverages, medicines, cosmetics and tobacco. · In the US, ammoniated glycyrrhizin is included in the GRAS list of approved natural flavouring agents but is not approved as a sweetener. · References: Sela and Steinberg (1989), Spillane (1996).
Table 9.7
Continued
Sweeteners
Chemical group
Origin
Relative sweetening power
Other sensory properties
Hernandulcin
Bisabolane sesquiterpene
Aerial parts of the South American herb Lippia dulcis Trev. (Verbenaceae).
1000
Also bitter and has an aftertaste.
L-Sugars
Sugars
Various sources but occur rarely.
Same as D-sugars
Mabinlin I and II
Proteins
Seeds of the Chinese Similar in plant Capparis masakai quality to LeÂvl (Capparidaceae). thaumatin and monellin but less potent
Mogroside IV Cucurbitaneand type Mogroside V triterpene glycosides
Dried fruits of the Chinese vine lo han kuo (Siraitia grosvenorii (Swingle) C. Jeffrey)
233±392 250±425
Other information
· L-sugars are mirror images of D-sugars and behave in the same way, with similar sensory and functional properties, but are not metabolised and are therefore non-caloric (Levin, 1986). Has a sweet taste that disappears a few minutes after being held in the mouth. A sweet taste subsequently occurs on drinking water.
· It is used medically and the seeds are chewed for their sensory effects.
· The fruit, its juices and extracts are used as sweeteners · Plant used for mainly medicinal reasons in China and Japan. · Potential for further commercial development of mogroside V as is stable, soluble and data collected to date has shown no adverse safety issues.
Monatin
Root bark of Schlerochiton ilicifolius
800±1200
Monellin
Protein
Fruits of an African 3000 plant Dioscoreophyllum cumminsii (Stapf) Diels (Menispermaceae)
Osladin
Triterpene glycoside
Fern Polypodium vulgare L. (Polypodiaceae)
300±500
Pentadin
Protein
Fruits of the African plant Pentadiplandra brazzeana Baillon (Pentadiplandraceae)
500
Periandrins
Triterpenoid glycosides
Roots of Brazilian licorice, Periandra dulcis Mart. (Leguminosae)
Approx. 85±95
Phyllodulcin
Dihydroisocoumarin
Leaves of Hydrangea macrophylla Seringe var. thunbergii (Seibold) Makino (Saxifragaceae)
400±800
Polypodoside A Triterpene glycoside
Rhizomes of the licorice 600 fern from North America, Polypodium glycyrrhiza DC. Eaton
· References: Lindley (1999), Fry (2005). Slow onset and persistent aftertaste.
Slow onset and lingering licorice aftertaste.
Licorice taste and lingering aftertaste.
· The leaves of the plant are used in Japan to make a sweet tea.
Table 9.7
Continued
Sweeteners
Chemical group
Origin
Pterocaryoside Secodammarane Stem and leaves of A and B saponins Pterocarya paliurus Batal. (Juglandaceae) Rubusoside
Diterpene
Compounds Diterpene extracted from glycosides Stevia, other than stevioside: Rebaudioside A Rebaudioside B Rebaudioside C Rebaudioside D Rebaudioside E Steviolbioside Dulcoside A Dulcoside B
Relative sweetening power
Other sensory properties
50±100
Other information
· The leaves of the plant are used in China to sweeten foods.
Leaves of the Chinese 114 plant Rubus suavissimus S. Lee (Rosaceae)
Also has a bitter taste.
· The plant is used medicinally and as a health-promoting food ingredient and beverage in China and Japan.
South American plant Stevia rebaudiana
In general, have a lingering sweetness. Rebaudioside A has a bitter taste. The other compounds have not been systematically studied.
· Rebaudioside C and dulcoside B are the same compound. · Rebaudioside B and steviolbioside may not be compounds found naturally in the plant but may be formed by partial hydrolysis during the extraction process. · References: Crammer and Ikan (1987), Kinghorn and Soejarto (1991).
250±450 300±350 50±120 250±400 150±300 100±125 50±120 50±120
Low-calorie sweeteners Table 9.8
213
Synthetic intense sweeteners
Class
Compound
Relative sweetness
Other information
Guanidines
Bernadame Carrelame Lugduname Sucrononic acid
200 160 225 200
· Reference: Glaser (1999).
Oximes
SRI Oxime V
450
Perillartine
2000
Dulcin
200
Suosan Superaspartame
350 14 000
Thio derivative of superaspartame
50 000
Urea derivatives
000 000 000 000
Trihalogenated benzamides
3-(3-carbamoyl-2,4,6- 4000 tribromophehyl)propionic acid
Tryptophan derivatives
6-chloro-Dtryptophan
· No undesirable aftertaste. · Stable. · Bitter with methol/licorice off-taste. · Limited solubility in water. · Used commercially in Japan to sweeten tobacco. · Briefly approved for use in the US before it was found to cause liver damage in rats. Also bitter. · Combination of aspartame and cyanosuason. · Reference: Tinti and Nofre (1991).
· Slow onset and slightly lingering sweetness, with some bitterness.
1000
Intake by vulnerable groups has been investigated. Children may be expected to have higher intake relative to body weight. Estimates of intake are below acceptable levels. Low-calorie sweeteners are safe for use during pregnancy within ADI levels (Duffy and Sigman-Grant, 2004). The introduction of new sweeteners is not expected to raise concerns over consumption as they would reduce the intake of existing sweeteners and would not be able to capture the same proportion of the market as early sweeteners had at the time when intake studies were done. Only a major new use would increase consumption. Renwick (1999). Safety issues related to specific low-calorie sweeteners are given in Table 9.5.
Table 9.9
Regulatory definitions
Term
Description
The Joint Expert Committee on Food Additives (JECFA) of the United Nations Food and Agricultural Organization (FAO) and World Health Organisation (WHO)
JECFA is responsible for implementing the joint FAO/WHO programme on food additives to make evaluations of food additives and provide advice to member states on the control of additives and related health aspects. It carries out risk assessments of food additives by reviewing available safety and technical data, endorses substances for use in foods and allocates acceptable daily intake levels (ADI).
The Codex Alimentarius Commission (CAC)
This was established in 1962 and its purpose is to implement the joint FAO/WHO Food Standards Program to establish a framework for international food legislation. It is responsible for development of food standards to guide and promote harmonised definitions and requirements for foods. The Codex Committee for Food Additives and Contaminants (CCFAC) is the body responsible for food additives.
The Food and Drug Administration of the United States (FDA)
The FDA is responsible for regulating food products in the USA, including food additives as defined by the 1958 Food Additives Amendment to the Federal Food, Drug, and Cosmetic Act.
The Scientific Committee on Food of the Commission of the European Union (SCF)
The Scientific Committee for Food (SCF) was established in 1974 and reformed in 1997. It advises the Commission of the European Union (EU) on issues relating to the protection of health and safety of persons arising from the consumption of food. It was responsible for risk assessment of food additives until establishment of the EFSA.
European Food Safety Authority (EFSA)
Established in 2002 in response to a series of food safety scares with a main objective to `contribute to a high level of consumer health protection in the area of food safety, through which consumer confidence can be restored and maintained'. Responsibilities for safety of food and feed including providing scientific opinions and advice to the EU commission, risk assessment, monitoring of risk factors and diseases, approaches and methods for hazard and risk assessment and to prepare work for the future evaluation of health claims.
Panel on Food Additives, Flavourings, Processing Aids and Materials in Contact with Food (AFC)
AFC is an Expert Panel of the EFSA established in 2003 with responsibility for safety in the use of food additives, flavourings, processing aids and materials in contact with food.
The Sweeteners Directive
The EU adopted The Sweeteners Directive in 1994 (Directive 94/35/EC (EU, 1994)), as part of the initiative to harmonise member state laws on food additives. It harmonised member states' laws on sweeteners and provided for the use of acesulfame-K, aspartame, cyclamatic acid and its Na and Ca salts, NHDC, saccharin acid and its Na, K and Ca salts and thaumatin, as well as other nutritive sweeteners. Two amending directives have since been adopted: Directive 96/83/EC and Directive 2003/115/EC (EU 1996 and 2003). The directive specifies the foodstuffs approved sweeteners may be used in and the maximum usable dose for each foodstuff. Sweeteners may not be used in food for infants and young children. It also specifies labelling terms and requirements to indicate presence of sweeteners and to provide warnings, such as `contains a source of phenylalanine' in the case of aspartame and salt of aspartame and acesulfame. The most recent directive 2003/115/EC was adopted in 29 January 2003, came into force on 29 January 2004 and had to be implemented by member states by 29 January 2005. It introduces the use of two new sweeteners: Sucralose and salt of aspartame and acesulfame. Directive 95/31/EC as amended by Directives 98/66/EC, 2000/ 51/EC, 2001/52/EC and 2004/46/EC lay down specific criteria of purity concerning sweeteners for use in foodstuffs (EU, 1995b; 1998; 2000; 2001; 2004).
General purpose sweetener
May be used in accordance with good manufacturing practice to sweeten any food when a standard of identity does not preclude its use. (O'Brien Nabors, 2001b).
GRAS
Generally Recognised as Safe by the FDA. GRAS substances have scientific consensus on their safety based on a history of use prior to 1958 or on well-known scientific information.
Sweetener regulations typically specify:
· A `positive' list of approved sweeteners. · Foods in which the sweeteners can be used and/or foods in which they cannot be used. · Maximum use levels of sweeteners in product groups. · Specifications of identity and purity for each sweetener. · Labelling of sweeteners and labelling of products containing sweeteners. May also be covered by food labelling regulations.
A petition for a new sweetener may include:
· Information from the manufacturer and potential users demonstrating the need for the sweetener. · Physicochemical, analytical and manufacturing information on the sweetener, analysis in foodstuffs, proposed uses (e.g., stability and breakdown products) and estimates of consumer intake. · Information necessary to evaluate safety-in-use, e.g., toxicological data on the sweetener, significant impurities and/or breakdown products. · Data relevant to the environmental impact of the sweetener.
216
Optimising sweet taste in foods
9.3
Health-related developments in low-calorie sweeteners
9.3.1 Health trends Health was the major driver of the global food industry in 2004, followed by convenience (Sloan, 2005). Of the top ten global food trends, three were directly related to health: inherently healthy foods such as fruits, vegetables, salads, grains, nuts and yoghurts; foods with reduced levels of elements perceived to be `bad for you' such as fat, calories, salt; and treating medical conditions through diet (functional foods). Healthy elements were features of another four trends: take-out food; grazing including growth in the reduced-sugar sector; `farmfriendly' foods such as organic, natural, free-range and sustainably grown foods; and global brands. Health is likely to increase in importance in Western countries as the incidence of obesity and associated health problems continues to rise and the proportion of elderly increases, whilst in developing countries malnutrition continues to dominate. In terms of changes in health-related behaviour, trends are mixed. Healthier US trends include changes in the proportion of dietary saturated, polyunsaturated and monounsaturated fats, increased consumption of fruit and vegetables, decreases in smoking, and increased awareness, diagnosis and pharmacological treatment of risks factors such as high cholesterol and high blood pressure. Unhealthy US trends include increases in total calories, portion size, refined carbohydrates and fast food intake, all of which can increase obesity. Trends in physical activity remain unclear. There is evidence that US consumers are attempting to become healthier. Although incidences of obesity, diabetes, arthritis, disability and medication have increased, there has been a substantial decline in the prevalence of key cardiovascular risk factors such as high cholesterol, high blood pressure and smoking in the last 30±40 years, particularly amongst the obese (Gregg et al., 2005). Low-calorie sweeteners have an evolving role to play in improving consumer health. The emphasis on how low-calorie sweeteners can be of benefit has shifted from diabetes in the 1960s, to hyperactivity and behavioural issues in the 1990s, to obesity in the 2000s (Duffy and Sigman-Grant, 2004). The line between healthy and regular foods is becoming less obvious as health becomes a lifestyle choice. Health news shapes consumer preferences. Specific health concerns, such as obesity and type 2 diabetes, also act as motivators and the functional food market is growing. Low-calorie sweeteners are no longer viewed exclusively as medical or diet aids but as part of a generally healthy life-style (Bright, 1999). Recommendations have been made to cut sugar in the diet. Foods and beverages with added sugar have lower micronutrient contents than those containing naturally occurring sugars and several reports have linked high intakes of added sugar with low intakes of some micronutrients. The Dietary Reference Intakes set by the US Institute of Medicine recommends that added nutritive sweeteners should make up no more than 25% of total energy in diet, whilst WHO recommends 10% and the US Department of Agriculture (USDA)
Low-calorie sweeteners
217
recommends 6±10% (Duffy and Sigman-Grant (2004). The UK Nutritional Task Force have suggested that non-milk sugars should deliver no more than 10% of total dietary calories, a decrease of 25±30% on British sugar habits (Bright, 1999). Low-calorie sweeteners can play a part in a healthy life-style by replacing calories in sugar and increasing palatability and quality of healthy and lowcalorie diets, so acting as an aid to weight reduction, weight maintenance and oral health. Current dieting trends for low-carbohydrate diets are likely to increase consumption of low-calorie sweeteners. Non-nutritive sweeteners can improve dietary quality by lowering sugar intake and increasing the palatability of nutrient-rich foods, such as fruits and vegetables. The latter is of particular relevance in developing countries where low-calorie sweeteners can provide a cost-effective way to make diets rich in nutrients more palatable. Low-calorie sweeteners may also be important in reducing calories in the diet as life-styles become more sedentary due to increased technology (Bright, 1999), and as there is an increase in the elderly population, who need fewer calories yet must maintain fluid levels (Duffy and Sigman-Grant, 2004). The American Dietetic Association (ADA) has advocated the beneficial use of low-calorie sweeteners to improve health. It suggests that dietetic professionals provide consumers with guidance on how to safely formulate plans that meet individual health goals in the context of current dietary and physical activity recommendations, including low-calorie sweeteners as appropriate (Duffy and Sigman-Grant, 2004). 9.3.2 Obesity and weight control Obesity is a widespread problem with complex causes involving social, behavioural, cultural, physiological, metabolic and genetic factors. WHO reports that more than 1.3 billion people are overweight or obese world-wide (Sloan, 2005). The incidence of obesity in developed countries is increasing. In the US, the prevalence of obesity in adults aged 20±74 has increased from 13% to 31% in 25 years (Gregg et al., 2005). Being overweight or obese increases the risk of medical problems, such as hypertension, stroke, cardiovascular disease, arthritis, type 2 diabetes and certain forms of cancer. It reduces quality of life through decreased mobility, social and psychological effects. Obesity increases the risk of morbidity. A recent study has found that almost 112,000 more deaths occurred than expected in obese individuals in the US in 2000 (Flegal et al., 2005). The cause of weight gain is a higher caloric intake than expenditure. Weight loss can be achieved by decreasing energy intake by eating fewer calories and/or increasing energy expenditure through physical exercise and is associated with a reduction of health risks and medical problems. Low-calorie sweeteners can play a dual role in weight reduction by reducing calories from nutritive sweeteners and enhancing palatability of low-energy foods (Duffy and SigmanGrant, 2004). There has been much debate on the influence and mechanisms of sweetness and low-calorie sweeteners on energy intake and weight control. Nutritive
218
Optimising sweet taste in foods
sweeteners by themselves do not cause an increase in weight and, similarly, low-calorie sweeteners themselves neither promote weight gain nor weight loss. They achieve an uncoupling of sensory and caloric characteristics and can sweeten food without adding calories. Consumers can use this saving in calories to reduce or control weight or as an excuse to ingest calories in other forms. Overviews of the scientific literature in this field are given by Booth (1987), Rogers and Blundell (1989), Booth (1991), Renwick (1996), Renwick (1997a), Blackburn (1999) and Drewnowski (1999). Findings are often conflicting, understandably given the methodological difficulties and possibilities in conducting human studies on variables that are affected by a large number of psychological, behavioural and metabolic factors. The ADA recommends that further research is carried out on the influence of non-nutritive sweeteners on dietary quality and their impact on satiety, energy intake and weight management (Duffy and Sigman-Grant, 2004). A summary of key topics is given below. Sweetness, palatability and food intake Sweetness is a pleasurable sensation that makes food palatable. It is thought to be pleasurable because it signals the presence of calories and thereby acts as a psychological reinforcement in hungry organisms to eat. Palatability decreases satiety and increases appetite and food consumption. It has been suggested that the high palatability of sweet foods may override normal satiety signals, leading to over-consumption of energy dense foods and higher body weight. The obese and bulimics (Franko et al., 1994) may have greater preference for sweet taste than others. Using intense sweeteners separates palatability and pleasure of foods from energy density and it has been suggested that this may change intake, although little evidence has been found to support this view (see below). Effect of low-calorie sweeteners on perceived hunger In 1986, two studies using different methods (experimental and epidemiological) suggested that intense sweeteners enhanced appetite and caused increased calorie intake and weight gain. Although both studies had methodological flaws, this result was counter-intuitive to current thinking and attracted much interest and further study in this area. Since then, numerous studies have failed to support the finding of an increase in subjective hunger ratings after ingestion of low-calorie sweeteners, although one study using aspartame-sweetened chewing gum supported the concept of a `fragile effect'. Effect of low-calorie sweeteners on food intake It has been suggested that individuals may compensate for the lack of calories in foods sweetened with low-calorie sweeteners by eating more of other energy dense foods. Numerous studies have failed to support the finding of an increase in actual food intakes after ingestion of low-calorie sweeteners. In only one study did a high intensity sweetener, saccharin, increase food intake. The evidence suggests that low-calorie sweeteners do not increase food intake.
Low-calorie sweeteners
219
People tend to consume a fixed weight or volume of food. Varying the energy density of these foods, rather than the amounts eaten, is the prime mechanism of regulating energy intake. Therefore, including energy dilute foods in the diet should lead to reduced energy intake and weight loss. Energy-dilute foods are more satiating than energy-dense foods but tend to be bulky, unpalatable foods, such as whole grains, vegetables and fruit. Although they are effective in weight-control diets, over-weight individuals have difficulty adhering to them in the long-term. The use of low-calorie sweeteners increases palatability whilst decreasing energy density, although they seem to have no effect on satiety. Role of low-calorie sweeteners in weight control Most studies investigating the role of low-calorie sweeteners in weight control have shown that replacing foods in the diet with low-calorie versions containing low-calorie sweeteners reduces overall caloric intake. Calorie intake in studies using covert replacement of sugar with low-calorie sweeteners, have shown that compensation is incomplete, although sometimes an initial rapid fall that tended to revert towards normal later was found, indicating that people are unlikely to lose weight by using low-calorie sweeteners without intentional control of their total calorie intake. When low-calorie sweeteners replace nutritive sweeteners in products, there is a net reduction in calories of the product itself. Consumers can choose to use this reduction in calories to assist in weight loss or to eat another caloric food without an overall increase in calories. Both of these options benefit the consumer by allowing them wider choice without extra calorie intake. However, individuals may consciously choose to compensate for the lack of calories by eating more of other energy dense foods, which would lead to weight gain. Therefore, they will not help to reduce weight unless there is an overall intention to reduce calories. Studies on subjects in weight control programmes have shown that intense sweeteners can be helpful in making the regime more acceptable and successful. Several studies on aspartame have suggested that it may facilitate control of body weight and enhances weight maintenance over the long term. It is suggested that low-calorie sweeteners may work through sensory-specific satiety, a reduced palatability of just-consumed food relative to other foods, by satisfying the desire for the pleasurable sensation of sweetness without increased energy consumption. It has been suggested that in a weight-conscious society, the relationship between food preferences and energy density is most likely modulated not by energy needs, but by consumer diet-related attitudes and behaviour. Whilst diets high in caloric sweeteners themselves are not the cause of obesity, reducing caloric intake by substituting low-calorie sweeteners for caloric sweeteners as part of a balanced diet provides one route to control body weight. Historically, low-calorie sweeteners have been used as diet aids. Traditionally, they were used to replace sweetness in response to nutritional advice to reduce sugar in the diet. More recently, the trend in nutritional advice is to replace energy-dense
220
Optimising sweet taste in foods
foods with energy-dilute foods (Rolls et al., 2005), and so low-calorie sweeteners are used to reduce calories whilst maintaining sweetness as part of a calorie controlled diet. Current trends in dieting, such as the Atkins diet, advocate cutting carbohydrates including sugars. There are now many products formulated as reduced-calorie and low-calorie that include low-calorie sweeteners. Labelling of such products must adhere to regulations. 9.3.3 Diabetes Diabetes is a widespread, growing global health problem. A recent study found that the prevalence of diabetes in the US has increased by 55% over the last 40 years (Gregg et al., 2005). This was attributed to the increased proportion of the population who are classified as obese. It is estimated that 2.8% of the world's population (171 million people) had diabetes in 2000 and this is projected to rise to 4.4% (366 million people) by 2030 (Wild et al., 2004). There are three types of diabetes: type 1 diabetes in which insulin is not produced, accounting for 10±15% of diabetics; type 2 diabetes related to insulin resistance, accounting for 85±90% of diabetics, approximately 90% of whom are obese or overweight; and gestational diabetes which occurs during pregnancy. Diabetes affects taste perception and may alter preferred level of sweetness (Tepper et al., 1996). Diabetics need to monitor their carbohydrate intake in order to control blood glucose levels. Monitoring overall carbohydrate consumption is the most appropriate measure to maintain near-normal blood glucose levels, rather than the glycaemic response resulting from their consumption. It is widely recognised that sweeteners do not cause diabetes. Prior to the 1980s, it was recommended that sucrose be removed from the diet of diabetics. It has since been found that sucrose does not adversely affect glycaemic control. A limit of sucrose in the diet of 10% of energy is recommended, which is the same as for a normal healthy diet. This upper limit is to prevent elevation of triglycerides. Fructose produces a smaller rise in plasma glucose than sucrose and most starches and in this respect may offer an advantage as a sweetening agent in the diabetic diet. However, there are other disadvantages to having a high level of fructose in the diet (see below), including elevation of triglycerides, and it is not recommended as the sole sweetener for diabetics (Duffy and Sigman-Grant, 2004). Non-nutritive sweeteners do not affect glycaemic response and can help control overall carbohydrate intake by substituting for nutritive sweeteners (Duffy and Sigman-Grant, 2004). An overview of the role of intense sweeteners in diabetes management is given by Ha et al. (1999). Low-calorie sweeteners can aid in weight loss and weight control, which is an important factor in controlling type 2 diabetes. Low-calorie sweeteners have no advantage over sucrose in glycaemic control. One study reported by Ha et al. found three incidents of hypoglycaemia after consumption of a non-nutritive sweetener, which may be a disadvantage for type 1 diabetics, although the author did not attribute importance to this finding. An overview of foods and beverages
Low-calorie sweeteners
221
formulated for diabetics, including those containing low-calorie sweeteners, is given in Jackson et al. (1987). 9.3.4 Excess dietary fructose Current trends in health promotion emphasise the importance of reducing dietary fat intake. However, as dietary fat is reduced, the dietary carbohydrate content typically rises and the desired reduction in plasma cholesterol concentrations is frequently accompanied by an elevation of plasma triacylglycerol, which has been linked to coronary artery disease and heart disease. Fructose and sucrose have been shown to increase triglyceride levels in short-term studies, with fructose having a greater effect than sucrose. Indeed, it is now widely believed that in humans the metabolism of large amounts of fructose increases the synthesis of triacylglycerol and its release into the plasma in very low density lipoprotein. This is particularly true in low fat diets (Parks and Hellerstein, 2000; Duffy and Sigman-Grant, 2004). High levels of fructose intake can induce diarrhoea in children as they cannot absorb it. Some individuals suffer from inherited fructose intolerance (Duffy and Sigman-Grant, 2004). Low-calorie sweeteners can play a role by substituting for fructose and thereby reducing fructose intake. 9.3.5 Behavioural effects Sugars and sweeteners have been linked to behavioural effects, as summarised by Duffy and Sigman-Grant (2004). There have been claims that sugar is associated with hyperactivity, but these have not been supported. Claims that food additives in general lead to poor behaviour, including hyperactivity and difficulty in concentrating, have been made but more studies are needed in this area. Some studies on the addictive effects of sweeteners in animals have been undertaken. In the early 1990s, theories of the effect of sweeteners and sweetcontaining foods in relation to mood were proposed such that negative feelings and depression were seen at the same time that subjects noted increased intakes of sweeteners and carbohydrates in general. It was hypothesised that subjects were trying to alleviate symptoms of negativity using a pleasurable experience. Approved non-nutritive sweeteners have not been found to show significant behaviour effects when consumed within ADIs. 9.3.6 Dental health Development of dental caries is caused by many factors, including nutritive sweetener intake, frequency of eating, frequency of brushing teeth and fluoridation of water and tooth applications (Duffy and Sigman-Grant, 2004). Sweet taste exerts a stimulatory effect on parotid salivary secretion. This effect contributes to quick dissolution of sweet foods in the mouth. Oral bacteria interact with sugars released leading primarily to accumulation of the bacteria in
222
Optimising sweet taste in foods
large masses known as dental plaque. These bacteria ferment common dietary carbohydrates leading to the formation of acidic end-products. Acidic endproducts and other extracellular products, such as enzymes and toxins, cause trauma to hard and soft tissue resulting in oral diseases, dental caries and periodontal disease. Measures for prevention of dental caries include removal of dental plaque using good oral hygiene practices, reduction of acid dissolution of tooth enamel using fluoride and use of appropriate dietary habits, including reducing the available fermentable carbohydrates, notably sugar. Low-calorie sweeteners provide one means of reducing the detrimental effect of bacteria on teeth by allowing the removal of sugar from foods whilst maintaining food palatability. Linke (1987) gives a literature review of dental health as it relates to sugar substitutes. Low-calorie sweeteners are noncariogenic and do not enable growth of caries-forming bacteria. Aspartame, saccharin and acesulfame-K are all non-acidogenic and do not promote dental caries or plaque formation. Saccharin exhibits true inhibition of bacterial growth and acidogenesis, due to its influence on glycolytic enzymes. Intense sweeteners blended together, or with xylitol or fluoride, have been shown to exhibit synergism in their inhibition of cariogenic bacteria (Ziesenitz and Siebert, 1986; Brown and Best, 1988). Low-calorie sweeteners have an advantage over sugar replacement by sugar alcohols in that they are less likely to lead to adaptation of oral bacteria after long-term exposure. Low-calorie sweeteners are used in many oral hygiene products, such as toothpaste, mouth wash and dental chewing gums. Country-specific regulations govern the use of dental health claims on products. D-Tagatose is authorised and sucralose is currently under consideration by the FDA to use the health claim `does not promote tooth decay' (Federal Register, 2005).
9.4
Market-related developments in low-calorie sweeteners
9.4.1 Market economics Low-calorie sweeteners are more expensive than sucrose when considered on a weight basis (Pearson, 2001), but the most appropriate way to compare cost is on the basis of sweetening power relative to sucrose. This approach is not exact as relative sweetening power is a function of concentration and there is no universally agreed reference sweetness level at which sweet potencies are compared (Fry, 2005). Sweetening power relative to sucrose is often expressed as `tonnes sucrose equivalent' (TSE). When considered on sweetening power, low-calorie sweeteners are cheaper than sucrose (Fry, 1999). Key low-calorie sweeteners can be placed in order of increasing cost based on sweetening power as follows: saccharin, cyclamate, aspartame and acesulfame-K. Some of the cost saving over sucrose may be offset by the cost of additional ingredients needed to replace the functionality of sugars, e.g., bulking agents, whilst using synergistic blends of low-calorie sweeteners can significantly reduce costs further.
Low-calorie sweeteners
223
The global market for sweeteners is 180 million TSE. The low-calorie sweeteners market (excluding D-tagatose) is 15 million TSE (8.5% of total sweeteners market), of which saccharin accounts for two thirds of the volume. The size of the low-calorie sweetener market has been growing by almost 6% per year for more than 20 years (Fry, 2005). The world food use of low-calorie sweeteners (excluding D-tagatose) is just over 11 million TSE, of which saccharin accounts for 50% and aspartame accounts for 25%. In Western markets, 65% is used in soft drinks, 18±20% in table-top sweeteners, 6% in dairy products (predominantly low-calorie yoghurts) and 2% in sugar-free confectionary (Fry, 2005). Saccharin is the most widely used low-calorie sweetener, as it is the most economical, ranging from 20±65 times cheaper than sucrose based on sweetening power depending on the form of saccharin and concentration used (Pearson, 2001). It is the most widely used sweetener for food and beverage applications, but it is also used commercially in non-food applications, particularly nickel electroplating. Aspartame is the second highest selling sweetener by volume but generates the greatest value. Sales grew rapidly when it was launched, but have been static since the mid-1990s due to the increased use of blends. Aspartame when used alone cannot compete economically with blended sweeteners due to the increase in sweetness that can be achieved through synergy (Fry, 1999). 9.4.2 Consumption patterns Consumption patterns of sweeteners vary by country and depend on many factors including: · Regional/global factors, such as economics, price of sugars, price of sweeteners, patents and their expiry, regulations, etc. · Internal factors, such as imported level of sugars, imported level of sweeteners, existence of cartels, regulatory situation, political situation, economic situation, availability of food(s), food manufacturing industry, food processing methods, distribution conditions, etc. · Consumer factors, such as income, public perception, culture, life-style, nutritional education, cooking habits, eating habits, etc. Americas (North, Central and South) The Americas are the heaviest users of aspartame (Fry, 1999), accounting for 75% of world sales (Duffy and Sigman-Grant, 2004). In the US, light foods were traditionally considered as dietetic foods for people with diabetes or specific medical conditions, including clinical obesity. They could only be found in pharmacists and health food stores and were limited in variety, with poor taste and high price. Now products with good quality, variety and value are ubiquitous (O'Brien Nabors, 1999). Light and low-sugar products have become a permanent part of a healthy life-style for the majority of Americans, with 180 million people (approximately 90% of the population)
224
Optimising sweet taste in foods
using low-calorie and sugar-free foods and beverages (CCC, 2004). US consumers use them to maintain better general health and to reduce weight (Bright, 1999), although weight loss has not been the primary driver since 1986 (O'Brien Nabors, 1999). The top reasons for consumption in 1998 were `better overall health benefits' and `to eat or drink healthier foods and beverages'. The top reason for not using reduced-sugar products was disliking the taste (O'Brien Nabors, 1999). Latin American consumers historically perceived products sweetened with low-calorie sweeteners as having a bad taste and were concerned over the safety of sweetener compounds, although they were willing to accept beverages sweetened with intense sweeteners. The market is growing quickly as products containing sweeteners become more accepted, although individual countries vary in consumption patterns (Yokoyama, 1999). Asia Southern and Eastern Asia account for half of the global consumption of lowcalorie sweeteners on a sweetness basis (Fry, 2005). The Asian markets are the heaviest users of saccharin, cyclamates and stevioside and also rely heavily on glycyrrhizin (Fry, 1999). China's market has transformed since 1988, when sugar imports stopped increasing in line with gross domestic product and intense sweeteners grew to take up to one third of the market by 1999 at the expense of sugar (Fry, 1999). It is now the largest source and user of saccharin (Fry, 2005). In Japan, consumers have a preference for natural sweetness. They are not negative towards low-calorie sweeteners but feel they do not need them. Sweet taste is considered to be for children and even the word `sweet' is slightly negative, meaning simple or inexperienced. Usage of low-calorie sweeteners is low due to lower levels of obesity, use of traditional cooking methods and lack of sweet foods in the general diet, for example tea is typically drunk unsweetened and fruit is eaten at the end of a meal rather than a dessert dish (Ishii, 1999). Europe The European market has transformed as the EU has evolved into a single market without trade barriers. This process continues as the number of member states increases and the internal market grows. Regulation of the EU market, including sweeteners, has been harmonised, as laid out in The Sweeteners Directive (see Table 9.9). Overviews of issues related to harmonisation of sweetener regulations in the context of the expanding, single European market are given in Lisansky and Corti (1995, 1996). Europe is the heaviest user of acesulfame-K, sucralose and NHDC and relies heavily on aspartame (Fry, 1999). The market in Europe is divided along regional lines, as reported in a study by Bakker (1999). In Northern Europe (Holland, UK) there was a tendency towards more light, reduced-sugar and reduced-calorie products compared to
Low-calorie sweeteners
225
Southern Europe (Spain, France). Women had a higher consumption of lowsugar products than men, but there was no relationship between consumption and age. Low-sugar products had lower consumption than low fat products or products that combined low fat and low sugar. Avoidance of sugar seemed to be more strongly related to being over weight than avoidance of fat. The population in Europe is ageing and food consumption is expected to change according to the nutritional and energy needs of an ageing population. The elderly need fewer calories, which may lead to a decrease in overall food consumption (Bright, 1999). Developing countries In developing countries, the use of sweeteners is driven by price and economic factors. As these countries are the centres of growth for the world population, Fry (1999) suggests it is likely that the world-wide sweetener market will be driven by economics and cost, as well as concerns about calorie consumption. This may be the case until there is a change from food scarcity to food abundance. Africa, with the exception of South Africa, has a poorly developed food processing sector, with no efficient distribution system, no diet market, conservative consumer attitudes to new foods, little disposable income and a confused regulatory system. As sweetener intake is a function of processed food consumption, which in turn is a function of disposable income, sweetener consumption is very low. Most sugar is consumed as a bulk commodity and is seen as a luxury, due to its high cost often controlled by cartels. Consumer attitudes towards low-calorie sweeteners are generally negative due to the perception that they taste bad, based on the overuse of saccharin and prevalence of poorly formulated products, and negative reporting in the media. The sweeteners market is largely centred around sugar replacement in sectors in which sugar can easily be replaced, for example liquid and powdered beverages and sugar blended with low-calorie sweeteners to give increased sweetness at lower price (Lanton, 1999). There is a fear that permitting low-calorie sweetener use in countries where large proportions of the population are living at subsistence levels would have a negative nutritional impact by cutting energy intake, although a counter argument has been made that sugar in Africa is largely consumed by the affluent segment, who would not be prone to malnutrition (Lanton, 1999). However, low-calorie sweeteners have been used beneficially in poorer countries to make a limited and repetitive diet more palatable (Bright, 1999). They have also been used to provide low-cost taste improvement of highly nutritious, but relatively unpalatable products, such as mageu, a thin porridge made from maize meal (Lanton, 1999). Lanton has also suggested that the cost reduction made by using sweeteners could be used to fortify products with nutrients missing from diets, such as vitamins and minerals.
226
Optimising sweet taste in foods
9.4.3 Consumer preferences It has long been known that consumers prefer sweetened products over unsweetened products. It is also known that consumers prefer sweetener blends over a single sweetener (Drewnowski, 1999; von Rymon Lipinski and Hanger, 2001). Historically, food manufacturers producing low-calorie and diet food have aimed to produce products that have the same sensory properties as the regular versions in the belief that this is the preference of consumers. For sweet foods, this has meant attempting to mimic the taste and texture of sucrose-containing products. Early on, these attempts were not very successful, although reduced-calorie products currently produced are sensorically much closer to regular products. Expectation did not match experience, giving a more negative affect. However, it may be that the preferences of consumers are changing. There is anecdotal evidence that some consumers prefer diet products over regular products (Kuntz, 1995). Traditionally in the US, it was thought that the market consisted of 25±33% of diet, no-calorie beverage drinkers, whilst the remainder of the market preferred regular drinks and this group would not accept the different taste profile of diet drinks (Fry, 2005). Consumption of diet drinks has increased over time. More consumers are exposed to them and use them on a regular basis. The use of diet beverages is so widespread, it is possible that some consumers may never have experienced regular beverages. It may also be that those who consume diet products have learned that they have particular sensory properties that are different to regular products. They expect these products to have certain flavour characteristics and be less dense. Initial use may be for the benefit of reduced calories, but consumers have learned to like these products through repeated exposure. There are genetic variations in taste perception, so that individuals vary in their perception and preference for sweetness, and also for bitterness that may appear as a side taste in low-calorie sweeteners (Bartoshuk, 2000). Some individuals may perceive artificial sweeteners differently. For example, Bartoshuk (1979) reports that the bitter taste of saccharin is stronger to some individuals and Beauchamp (1999) reports anecdotal evidence that aspartame tastes like sucrose to some individuals whilst to others it has an off note. Other factors such as age, environment and culture also play a part in taste preference (Beauchamp, 1999), which means that consumers with different demographics and/or locations are likely to have different preferences for sweetness. 9.4.4 Consumer perceptions and attitudes Consumers recognise and accept the benefits of reduced-sugar and low-calorie products that include low-calorie sweeteners, but negative perceptions persist for a variety of reasons (see Table 9.10), despite the fact that many of the issues are no longer salient or are not supported by evidence. The taste quality of lowcalorie sweeteners is much improved and can come very close to sugar,
Low-calorie sweeteners Table 9.10
227
Public perceptions of low-calorie sweeteners
Reasons for poor perception · Poor sweet taste quality of early sweeteners. · Perception of saccharin as a poor, cheap substitute for sugar, associated with war-time deprivation. · Alleged carcinogenic effects of saccharin and cyclamate in the 1960s and 1970s. · Perception as `artificial' food additives rather than `natural' food components. As concerns over health and safety increase, there is an increase in concern over food ingredients and additives and a trend towards natural, organic, additive-free products. Sloan (2005) reports on a survey that paradoxically found mothers in the US are trying to restrict their children's intake of candy, sugary foods and high fructose syrups on the one hand, and artificial sweeteners on the other. The assignment of E-numbers to food additives was originally designed as a way to protect and re-assure the public of the safety of food additives, but these are now seen as a negative label. In the same way, low-calorie sweeteners have been viewed negatively by some consumers, as `artificial', `synthetic' and `chemical' with the negative associations these words bring. · Health-related and consumer groups, who encourage the reduction or elimination of sweetened foods and beverages from the diet. · Sensationalised, negative media and internet coverage, particularly in relation to aspartame.
especially with the use of blends. Alleged carcinogenic effects have proved unfounded and warning labels have been removed. The safety of low-calorie sweeteners has been re-affirmed by numerous studies and reviews undertaken by professional bodies (see Section 9.2.4 and Table 9.5). It seems that the issue of sweeteners can evoke strong feelings, irrespective of statistical argument and evidence (Duffy and Sigman-Grant, 2004). There is a need to educate the public in a sensible, reliable and easily understood informative manner on the benefits and perceived risks of low-calorie sweeteners (Grenby, 1996b). The ADA recognises that dietetic professionals play an important role in educating the public about the use, safety and health implications of sweeteners. They recommend that dieticians should provide consumers with science-based information on sweeteners and help them translate this information into a plan that safely meets individual health and dietary needs and goals in the context of current dietary and physical activity recommendations (Duffy and Sigman-Grant, 2004). A natural intense low-calorie sweetener may be more acceptable to consumers. Currently, there are only two: stevioside and thaumatin, but they are not in widespread use globally and do not have ideal sweet taste profiles. 9.4.5 Development of new low-calorie sweeteners There are several ways that can be used to identify and develop new intense sweeteners including:
228
Optimising sweet taste in foods
(a) Accidental discovery, e.g., saccharin, cyclamate and aspartame. This is less likely today due to safety and legal restrictions on tasting new compounds (Spillane, 1996). (b) Screening of literature for references on sweet taste, e.g., rediscovery of hernandulcin, a sweetener of natural origin known to the Aztecs (Spillane, 1996). (c) Development of structure-taste activity relationships (SARs) for different molecular classes, e.g., sucralose (Spillane, 1996). (d) `Purpose-built' design of sweeteners, e.g. alitame, which also involved method (a) and (b) (Spillane, 1996). (e) Development from knowledge of the structure of taste receptor proteins and mechanisms of taste perception. (f) Bio-divining or bio-prospecting. In the latter part of the 1990s, it was reported that there was little incentive to identify and manufacture new sweeteners as the range that existed was adequate for most purposes and the cost and time involved in developing and getting new sweeteners approved was prohibitive (Grenby, 1996b; Lindley, 1999). The priority shifted to developing technological advances to improve functionality of sweeteners, either by developing other ingredients to replace the functionality of sugar, such as bulking agents or taste improvers, or by processing sweeteners in new ways, such as agglomeration to improve handling and encapsulation to improve stability of aspartame. The development of the aspartame-acesulfame salt was a stepwise development that enabled new benefits to be gained from existing low-calorie sweeteners. Biotechnology is expected to bring major advances to the field of sweetness perception in three ways: · Through understanding the molecular basis of taste perception. · By understanding and delivering to individual variation in taste perception. · By production of natural sweeteners through genetic modification of plants. Understanding molecular basis of perception There have been significant advances in understanding the molecular receptor mechanisms involved in taste perception in the last five years (reviewed by Montmayeur and Matsunami, 2002; McGregor, 2004; Scott, 2004). Hoon et al. (1999) described genes that code for the sweet and bitter taste receptors, named T1R and T2R respectively. It has been shown recently that there are two sweet taste receptors, known as T1R2 and T1R3, that work together as a sweet detector (Nelson et al., 2001). Nutritive sweeteners may have different receptor mechanisms from non-nutritive sweeteners, as shown by studies in transgenic knockout mice (Damak et al., 2003), which may account for differences in taste between them. Bitter taste has a least 25 receptors known as TR2s (Adler et al., 2000). Once the structure of the sweet and bitter taste receptors that sense lowcalorie sweeteners are known, it will be possible to design sweet tasting
Low-calorie sweeteners
229
compounds with better or more intense sweet taste profiles using computer modelling techniques and rapid screening similar to methods employed in drug design by the pharmaceutical industry. It will also be possible to design taste modifiers that potentiate sweetness, perhaps enabling nutritive and/or natural low-intensity sweeteners to function as intense sweeteners, thereby reducing sugar levels and calories. The first taste modifier that blocks bitterness through interference with bitter taste transduction mechanisms is already being marketed commercially. In 2003, Linguagen was granted a patent (Margolskee and Ming, 2003) and FDA GRAS status for a bitter blocker, adenosine 50 -monophosphate (AMP). Bitter blockers have potential use in blocking unwanted bitter taste of low-calorie sweeteners. Individual variation in taste perception Recent advances in the field of genomics hold promise for the future. It has long been known that individuals vary in their taste abilities. Now, it is possible to combine precise molecular and perceptual information from individuals, so that all stages of the individual taste perception process from taste receptor genes to taste receptors and their mechanisms to taste response can be characterised, linked and understood (Kim et al., 2004). This approach, known as tastomics, has shown that individuals carry their own unique set of taste receptors that gives unique taste perception (Bufe et al., 2005). Similar investigations in the field of olfaction, when combined with taste research, will provide insights on individual variation in flavour perception. The growing field of nutrigenomics investigates the genetic variation that causes people to respond differently to food nutrients. In the future, these fields of genetic study may lead to personalised diets based on genetic make-up, that are tailored to meet individual nutritional requirements and flavour preferences, both of which are of relevance to the use of low-calorie sweeteners. Genetic modification Genetic modification may provide a route for production of natural low-calorie sweeteners. Easily-grown crop plants may be implanted with the genes necessary to synthesise naturally occurring low-calorie sweeteners (Fry, 2005). Attempts have been made to modify yeast cells to produce thaumatin (Weickmann et al., 1989) and more stable forms of monellin (Kim et al., 1991). New natural sweeteners Work to identify new natural intense sweeteners continues, as they can pass through the regulatory process more quickly and have a more acceptable image to consumers. There are many inaccessible areas, for example rainforests, containing undiscovered plant species that may yield new compounds with intense sweet taste. Some companies have undertaken commercial bio-divining agreements with local bodies that may yield new flavour components whilst providing a source of funding for ecological preservation. Potential new natural sweet tasting compounds can be rapidly screened using genomic techniques.
230
Optimising sweet taste in foods
9.5
Implications for food product design
Low-calorie sweeteners are added to food and beverage products to replace sugars in order to reduce calories and/or cost. These products need to be formulated to replace the properties lost when nutritive sweeteners are removed, so that they have an acceptable taste profile and desirable functional properties, including performance characteristics and sensory profile, such as flavour, texture, colour and appearance. Functionality depends on the food product, processing, application and conditions of use. Food product design needs to be undertaken in a systematic way, taking these factors into account. Food ingredients need to be compatible and stable. Low-calorie sweeteners are often used in combination with other low-calorie ingredients, such as fat replacers, which may make product design even more challenging. The ingredients and/or product must also be able to withstand different processing methods, such as pasteurisation, sterilisation, UHT treatment, extrusion, cooking, baking, drying, etc. Design needs to consider conditions of storage and use, with sufficient chemical and microbial shelflife taking into account factors such as temperature, pH and water activity. Formulation must meet consumer preferences and needs, taking cultural and market differences into account. Global products are a growing trend (Sloan, 2005) and careful consideration needs to be given so that their formulation meets varying local consumer preferences and local ingredient, processing and labelling regulations, whilst being similar worldwide. When used for sweetening properties, low-calorie sweeteners are competitively priced compared to the price of sucrose based on sweetness equivalence (see Section 9.4.1), but this cost saving may be off-set by the need to include additional ingredients to replace other functional properties of sucrose. Much effort, particularly in the last ten years, has focused on trying to find additives to improve the taste and functional properties of existing sweeteners. The following sections give an overview of key topics in food product design involving low-calorie sweeteners: sweetener blends, flavour design and modification and formulation to replace functionality. General references on application and use of low-calorie sweeteners in food product design include Lindley (1983), Jackson et al. (1987), MacKay (1987) and Nelson (2000). 9.5.1 Blends containing low-calorie sweeteners Sucrose continues to be the standard that the ideal low-calorie sweetener must attain, so that low-calorie sweeteners should have the same sensory and functional properties as sucrose at a competitive price when used in foods and beverages (see Table 9.11). None of the current low-calorie sweeteners match sucrose in sweet taste quality or temporal characteristics. They are commonly used in blends together and/or with nutritive sweeteners to give an improved sweet taste profile, such as masking off tastes. This often has the additional benefit of providing sweet taste synergy, so that the sweetener combination gives a higher perceived sweet intensity than would be expected from the sweet
Table 9.11 Functional properties of sugar Functional category
Functional property
Sensory properties
Sweet Odourless Contributes to perceived texture of foods
Physical texture properties
Contributes to perceived appearance of food Contributes to flavour Bulk, weight or size Thickness, body, viscosity and volume in liquids Gelling action Crystallisation process Fermentation Tenderness in baked goods
Solubility-related properties Chemical reactions Preservative action
Aids in ability to entrap gas Binding action Decreases water activity and increases water immobilisation Depresses freezing point Raises boiling point Caramelisation and Maillard reactions (with reducing sugars) Prevents microbial spoilage
Effects and examples
Influencing physical texture properties (see below). Examples of perceived texture influenced include fullness, mouthfeel, viscosity, thickness, crunchiness, crispness, etc. Examples of perceived appearance influenced include colour (see below) and surface appearance, e.g. frosting, shininess from glaze, etc. Counters bitterness and sourness. Also plays a role in flavour generation (see below). May need to be replaced with bulking ingredients. These improve mouthfeel and contribute to richness and fullness. A lack of thickness may cause problems during processing, for example with foaming on soft drinks lines. May need to be replaced with bulking agents. E.g. in jam. E.g. in production of hard candies, chocolate, ice cream, etc. E.g. in baked products. Influences temperature of starch gelatinisation and egg coagulation. Sugars compete with gluten for water in baked products, slowing development of gluten during mixing and making the final product more tender. Acts as a humectant, resisting changes in moisture and influencing shelf-life (staling and drying). Provides texture and volume, e.g. in cakes, ice cream, whipped desserts. Product remains homogeneous, e.g. in some sweets. Can improve stability and preservative action. E.g. in ice cream and frozen desserts. Important in the freeze-thaw cycle. Important in processing, e.g. hard candy production. Produces desirable changes in colour (browning) and flavour, e.g. in bakery products. Sugar reduces water activity at high concentrations and hence microbial spoilage. Additional preservatives may be needed if sucrose is not present, e.g. sulfides, sorbates and benzoates
Table 9.12 Benefits of using blends Benefit
Description
Quantitative sweetness synergy
When two sweeteners are used in combination, the sweetness of the mixture is greater than the sum of the sweetness of the individual components. In addition, maximum sweetness intensities from the psychophysical curve plateau of single sweeteners (8±15% sucrose equivalence) can be overcome with blends. For example, the common blend acesulfame-K/aspartame gives 30±40% sweetness enhancement. (For a psychophysical evaluation of synergy in binary and ternary sweetener mixtures see Schiffman et al. (1995) and Schiffman et al. (2000) respectively.)
Qualitative sweetness synergy
Sweet taste profile can be improved to more closely match the sensory properties of sucrose (Hanger et al., 1996). Bitterness and undesirable side tastes can be masked using the lingering sweetness of another sweetener. These off-tastes tend to occur at higher concentrations and are less evident at lower concentrations used in blends. Blends can be used to improve the temporal profile, such as overcoming delayed onset. Sweetness can be given a full mouth effect, character and body using blends of different sweeteners that stimulate a variety of sweet taste receptors throughout the mouth (Shamil, 1997). For example, in acesulfame-K/aspartame blends, aspartame broadens the taste profile and masks bitterness of acesulfame-K, giving a closer profile to sucrose.
Overcoming adaptation effects
Adaptation is the reduction in perception that occurs during prolonged exposure to a sensation. For sweetness, repeatedly taking sips of a beverage or mouthfuls of food containing a sweetener will cause adaptation, coupled with short periods of incomplete recovery of sweet taste perception when the mouth is cleared. Schiffman et al. (2003) found that the decline in sweetness intensity experienced over repeated exposure to a sweet stimulus (four sips of a beverage) could be reduced by using sweetener blends. Binary and ternary mixtures of 14 nutritive and non-nutritive sweeteners were investigated. Mixtures consisting of two or three sweeteners exhibited less reduction in sweetness intensity than a single sweetener at equivalent sweetness. Ternary mixtures of sweeteners were slightly more effective than binary mixtures.
Cost reduction through synergy
Using sweet taste synergy to reduce costs by producing the same sweetness at lower cost, or meeting cost constraints by producing the best quality sweetness possible at a fixed cost. For example, the blend of aspartame/saccharin is commonly used to overcome lower stability of aspartame and poorer taste of saccharin, whilst reducing cost over aspartame alone.
Improved sweetness stability during processing and storage
This can be achieved by blending sweeteners with a low stability, with those that are stable for longer, so that sweet taste is not lost over time during storage. The particular blend used can take into account product conditions, such as low pH in soft drinks, processing conditions, such as high temperature during sterilisation, usage conditions, such as cooking, and distribution conditions, such as excessive temperature in hot countries. For example, the greater stability of acesulfame-K and saccharin compensate for the lower stability of aspartame and consequent loss of sweetness over time in binary and ternary blends.
Improved versatility and range of applications
The range of sweeteners available for use in blends offers flexibility, so that precise sensory properties desired are achievable in a wider variety of products.
Comply with legal limits
Increase sweetness in applications where the use of one or more sweeteners is limited (Wells, 1989).
Providing acceptable taste across a broader range of consumers
Consumers have been found to prefer products sweetened with a blend over those sweetened with a single sweetener (Drewnowski, 1999). Carbonated beverages (colas) sweetened with a sweetener blend (acesulfame-K and aspartame) are preferred to those sweetened with a single sweetener (aspartame) von Rymon Lipinski and Hanger, 2001). There is genetic variation in individual sensitivities to sweet and bitter compounds. Certain groups of consumers may be adverse to particular sweeteners. For example, some people find the bitter taste of saccharin unacceptable. Using blends can overcome these unpleasant effects and offer variety to these consumer groups. Blends can also be used to better match specific cultural and market preferences.
Reducing exposure to a single sweetener
Thereby increasing the already large safety factor in intake levels (Renwick, 1997a).
Creating novel sweet sensations
Blends can be used to give new, unusual or unexpected sweet taste profiles, rather than attempting to mimic the taste profile of sucrose.
Synergism in cariogenic bacterial inhibition
Discussed in Section 9.3.6.
234
Optimising sweet taste in foods
tastes of the individual components. The amount of each sweetener that needs to be used in a blend is less than if used individually, lowering costs. The advantages of using blends are given in Table 9.12. The exact nature of the blend used, in terms of sweeteners and the ratios they are used in, depends on the product, the flavouring system, processing, distribution, shelf-life, usage conditions, target market and target consumer. References that give overviews and details of specific ratios and concentrations used in sweeteners blends include von Rymon Lipinski (1996), Knight (1997), Shamil (1997) and Bakal (2001). The first blend combination was saccharin and cyclamate. It has a less bitter aftertaste than the individual sweeteners and is used in a variety of applications, particularly beverages and table-top sweeteners. Blends commonly used include acesulfame-K/aspartame and aspartame/ saccharin. Three way blends (sucralose/acesulfame-K/cyclamate; saccharin/ acesulfame-K/aspartame) and four way blends (aspartame/acesulfame-K/ cyclamate/saccharin) are also used. Aspartame-acesulfame salt is a chemical blend that provides blended sensory properties in a single compound and offers advantages over physically blending the two components (see Table 9.6). Blending low-calorie sweeteners with nutritive sweeteners can give synergy whilst reducing calories and in some instances cost. For commercial reasons, e.g. where sucrose is expensive or difficult to get at consistent quality, intense sweeteners are used in combination with sucrose in a range of beverages. There has been legislation, for example in the US, to restrict the blending of lowcalorie sweeteners with nutritive sweeteners in products to prevent adulteration of sugar with cheaper alternative sweeteners. In 1993, a UK law requiring minimum sugar contents in soft drinks was abolished and this has led to a big shift in formulation of non-diet soft drinks for economic reasons, so that many regular drinks contain almost no calories, even though they are not labelled as lower-calorie or reduced sugar (Fry, 1999, 2005). In 1996, it was reported that the volume of the UK soft drinks sector that included intense sweeteners but were not marketed as diet or low-calorie was double that of the low-calorie sector (Gordon, 1996). Where taste is a key product characteristic, low-calorie sweeteners may be blended with nutritive sweeteners to improve the taste profile, for example, to enhance and extend sweetness in full sugared chewing gum. In table-top sweeteners, where taste profile is the key functionality, lowcalorie sweeteners may be blended with sucrose and fructose to produce a reduced-calorie, rather than low-calorie product with high quality sweet taste. 9.5.2 Flavour design and modification The flavour system of a product needs to be designed according to the sweetener(s) and flavours used. Sweeteners can act as flavour modifiers. What is a negative effect in one system may be a positive in another. For example, lingering sweetness is not appropriate in a beverage but is desirable in a chewing gum. In beverages, sweeteners with delayed onset can accentuate the flavour of a citrus drink but may bring out undesirable brown spice notes in a cola (Knight,
Low-calorie sweeteners
235
1997). If a sweetener with a bitter note is used in combination with bitter flavour notes, the final level of bitterness may be unacceptable. If a sweetener loses sweetness over time some of the perception of some of the lost sweetness can be replaced with other flavours. Using mixtures of sweeteners can minimise flavour problems that can occur when a single sweetener is substituted for sucrose (Nahon et al., 1996). Many factors can affect sweetness and flavour perception, such as temperature of consumption and viscosity of the system. A systematic approach needs to be undertaken, with the flavour system designed around the sweetener system being used, and taking into account the product system and product usage. Formulating flavour with low-calorie sweeteners is reviewed from a technical perspective in Kuntz (1995). Traditionally, formulations are made to match the taste of sugar. However, it may be more appropriate to make a new product with a new taste rather than try to simulate the taste of a full sugar product, particularly as consumer preferences are changing, especially in markets where low-calorie beverages have been available for a long time (see Section 9.4.3). A fuller mouthfeel and taste may no longer be seen as refreshing (Kuntz, 1995). Much effort has been put into modifying the taste of low-calorie sweeteners to reduce their undesirable side tastes and improve their sweet profile. Compounds used to improve the bitter/metallic after taste of saccharin include the use of calcium chloride, arabinogalactan, neosmin, d-tryptophan and cream of tartar. Recently, compounds called bitter blockers have been developed that could be used to block bitter off-tastes of low-calorie sweeteners (Margolskee and Ming, 2003). Sweet taste modifiers and enhancers include 2,4 and 3,5dihydroxybenzoic acid, tannic acid, salt at low levels, carbohydrates such as erythritol and xylitol said to modify sweet taste by reducing undesirable or lingering aftertastes and maltol and ethyl maltol said to enhance sweetness, round the flavour profile and mask bitterness and other off-flavours (Lindley, 1999). Natural flavour modifiers that work at very low levels include glycyrrhizin, which enhances sweetness whilst reducing bitter notes, and thaumatin. 9.5.3 Functionality Sugars and other nutritive sweeteners have a wide range of functional properties in foods and beverages that may need to be replaced when designing products without them (see Table 9.11) (Hegenbart, 1994; Arya, 1997). Some design considerations for specific food applications reformulated to include low-calorie sweeteners are given in Table 9.13.
9.6
Future trends
9.6.1 Future market trends It is forecasted that the market for low-calorie sweeteners will continue to expand and growth rates of 8±9% have been predicted. The main drivers of
Table 9.13 Design considerations for reformulation of food products and applications to include low-calorie sweeteners Application
Considerations
Baked goods
Use of high temperatures with carefully controlled water content to create and maintain the correct structure. Maillard reaction occurs. Low-calorie sweeteners used must be heat stable. May need to add bulking agents, such as polydextrose, to substitute for reduction of sugar and flour in reduced-calorie products. May need to modify flavour and colour ingredients.
Beverages
Require good dissolution, particularly for dry mix beverages, as well as good stability and long shelf-life at low pH. Low pH protects from microbiological storage. Heat may be used in processing, such as hot filling, pasteurisation, sterilisation and UHT treatments. Low foaming is important in carbonated beverages. Low-calorie sweeteners used must be stable with good sensory properties. Sweetener blends are typically used. May need to add bulking/thickening agents to increase viscosity lost when sugar is replaced. May need an increased level of preservatives, particularly in fruit beverages (Wells, 1989).
Breakfast cereal
Produced by extrusion or cooking then drying. Sugar is used to counter bitterness in the grain, act as a flavour enhancer for grain notes, produce colour, provide texture and mouthfeel, and influence processing behaviour, including extrusion. Flavour may be applied before or during cooking. Coatings are often applied prior to drying and may be designed as a glaze, frosty or dusting layer. They also serve as binders for adherence of dry ingredients, vehicles for vitamins, flavour carriers and free-flowing agents preventing clumping, e.g. of raisins (Hegenbart, 1996). Low calorie sweeteners used should be stable under processing conditions and, if used in a coating, dissolve easily in milk.
Chewing gum
Sugar-free chewing gums are bulked with sorbitol, mannitol and/or xylitol. Sugar-free gum may become brittle due to water loss and the crystalline nature of sorbitol can be important for extending chewy texture. However, the sweetness of these bulking agents is about half that of sucrose, so low-calorie sweeteners are used to increase sweetness, whilst providing the added benefit of being non-cariogenic. Encapsulation may be necessary if the sweeteners used dissolve quickly in saliva, e.g. acesulfame-K. Low-calorie sweeteners may also be used to enhance and extend sweetness in both sugar-free and full-sugared chewing gum. As the length of time flavour last depends on the extent of mixing of the flavour and sweetener with the gum base, it is better if sweeteners can be added when the product is hot during high torque mixing (Dubitsky, 1996).
Confectionery
Is commonly processed using heating or boiling. Texture properties are achieved through the equilibrium between sugar and glucose syrup and careful consideration of texture is needed when sugar is removed. Bulk sweetening agents are commonly used. For example, gum confections often contain sorbitol and xylitol and hardy candies often contain isomalt and lactitol. Low-calorie sweeteners need to be heat stable. They are useful for increasing sweetness where the amount of sugar that can be incorporated is limited, e.g. starch-based confectionery. They have not found widespread use in chocolate products due to problems in reproducing the textural characteristics produced by sugar crystals.
Dairy products (yoghurt and dairy desserts)
Need to replace texture and mouthfeel using bulking and thickening agents. Stability and preservation is important. Fruit flavoured dairy products often use blends of acesulfame-K and cyclamate. Acesulfame-K and aspartame is commonly used where temperature stability is not important.
Desserts, topping and fillings
May use pasteurisation, sterilisation or UHT. Starch-based puddings need heat treatment to gelatinise the starch. Sugars have a preservative action in pie fillings. Low-calorie sweeteners used need to withstand temperatures used during processing. Bulking agents to replace texture may be required, such as sorbitol and mannitol.
Dressings and condiments
Are mostly low pH products, but may be cooked during final use. Low-calorie sweeteners used need to be stable at low pH, often during prolonged storage, and may need to be heat stable.
Fermented products
The low-calorie sweeteners used need to be stable to bacteria used in fermentation, such as lactic acid bacteria, e.g. acesulfame-K.
Fruit and vegetables (preserved)
Fruit and vegetables preserved by canning or bottling typically undergo heat treatment and low-calorie sweeteners used need to withstand heat treatment. For fruits preserved using sugar, this can be replaced by glycerol, sorbitol and mannitol and low-calorie sweeteners can play a role in increasing sweetness. Whereas sugar in concentrated syrups can adversely affect fruit texture by osmotic action drawing out cellular fluids, use of low-calorie sweeteners can cause the converse with fruit cells swelling. Pectins can be used to increase syrup viscosity if necessary. Generally, low-calorie sweeteners used need to be stable during prolonged storage.
Ice cream and frozen confectionery
Sugar is important during freezing and crystalisation of water. It forms a large part of the total solids in ice cream and contributes to the texture and volume by enabling gas to be trapped. It is difficult to reproduce the texture properties of ice cream when sugar is removed. Low-calorie ice creams are traditionally made by removing fat and using emulsifiers and stabilisers to replicate texture characteristics.
Table 9.13 Continued Application
Considerations
Jams, marmalades and preserves
Often sterilised in jars. Removal of sucrose reduces shelf-life as water activity will be higher, so low-calorie products are more susceptible to microbial spoilage and preservatives may be needed to counteract this. Gelling agents, such as pectin, vegetable gums and modified starches, and bulking agents may also be necessary if sugar is removed. Flavour stability may be compromised and may need to be redesigned. Low-calorie sweeteners used must be stable during long-term storage.
Medicinal products
Medicinal products are traditionally formulated with sucrose (syrups, lozenges) and lactose (tablets, chewable tablets) to increase palatability, but often need to be taken on a regular basis, increasing the risk of dental caries. Low-calorie sweeteners are non-cariogenic and can overcome this problem. Low-calorie sweeteners can be used to replace the sweet taste of sugars in products formulated for diabetics. Products labelled as suitable for diabetics must conform to relevant regulations (Jackson et al., 1987).
Table-top sweeteners
Low-calorie sweeteners, used singly, in blends and blended with nutritive sweeteners, are available as tablets, powders and liquids for use in sweetening beverages, such as tea and coffee, sprinkling on food and in home baking. These products are often a mix of the carrier, the active sweetening ingredient(s) and an anti-caker or flow enhancer. Tablets must be readily soluble and a disintegrant, such as sodium bicarbonate, tartaric acid or sodium carboxymethylcellulose, may be necessary to aid dissolution of relatively insoluble low-calorie sweeteners, such as acesulfame-K and aspartame. Powders need to be homogeneous to prevent separating out during distribution and so may be co-spray dried. Agglomerated powders based on hydrolysed starch, lactose or dextrose may be used to provide bulk in powdered forms.
Low-calorie sweeteners
239
growth are health trends, economics, evolving consumer preferences and new market opportunities. The main constraints on growth are functional ingredient technology, supply, trend towards additive-free products and poor image (Fry, 2005). As most of these topics have been reviewed elsewhere in this chapter, only the remaining topics will be discussed here. · New market opportunities: A strategy for product formulation that provides a new market opportunity for low-calorie is `mid-cal'. This is a blend of nutritive and low-calorie sweeteners to give a product that is low in calories yet maintains taste and perceived texture qualities. This trend is already well established in the UK drinks market, where the change was made covertly, driven by lower costs. In the US, some mid-cal table-top sweeteners are available. The first drinks products are reaching the market and are being promoted as new brands with health benefits (Fry, 2005). · Functional ingredient technological constraints: Overcoming technological constraints in a cost effective manner and consumer-friendly format, such as low-calorie, non-laxative bulking agents (Fry, 2005). · Supply constraints: There may be problems in production capacity of aspartame and sucralose to meet demand in the short-term, although these are being addressed (Fry, 2005). 9.6.2 Technological advances The cost and time involved to develop and gain approval for new low-calorie sweeteners is increasingly prohibitive and, during the 1990s, the emphasis shifted to re-formulation of products for product improvement and better support of health claims (Grenby, 1996b). However, the search for improved low-calorie sweeteners has been revived by recent biotechnological developments in taste chemoreception sciences. These are providing a better understanding of the structure and functioning of the sweet and bitter taste receptors which are opening new routes for design and production of low-calorie sweeteners, including design based on receptor structure, modification of perception based on molecular mechanisms, individually-tailored perception based on genomics, individually-tailored diets combining tastomics and nutrigenomics and production of natural intense sweeteners using genetic modification. The ideal is to find a natural, stable, safe low-calorie sweetener with the taste properties of sucrose, as this would be more acceptable to consumers and potentially easier to gain regulatory approval (Lindley, 1999; Fry, 2005). Some efforts using new molecular and genetic technology, and bio-prospecting in remote areas, have been focused in this direction.
9.7
Sources of further information and advice
General information · European Food Information Council (EUFIC) (www.eufic.org)
240
Optimising sweet taste in foods
· International Food Information Council (IFIC) (www.ific.org) · International Sweeteners Association (ISA) (www.isabru.org) and ISA conference proceedings (Lisansky and Corti 1995; 1996; 1997; Corti, 1999) · Other conference proceedings (Inglett, 1974; Walters et al., 1991; Corti, 1999) · Book series Alternative sweeteners (O'Brien Nabors and Gelardi, 1986; O'Brien Nabors and Gelardi, 1991; O'Brien Nabors, 2001a) · Books series Developments in sweeteners (Hough et al., 1979; Grenby et al., 1983; Grenby, 1987) · Other books by Grenby including Progress in sweeteners (Grenby, 1989) and Advances in sweeteners (Grenby, 1996a) Regulatory and safety information · EFSA (www.efsa.eu.int) · FDA (www.fda.gov) · JEFCA (www.codexalimentarius.net/web/jecfa.jsp) · SCF (http://europa.eu.int/comm/food/fs/sc/scf/index_en.html) Health-related information · American Dental Association (www.ada.org) · American Diabetes Association (www.diabetes.org) · American Dietetic Association (ADA) (www.eatright.org) · CCC (www.caloriecontrol.org) · WHO (www.who.int/en/)
9.8
Acknowledgements
The author is grateful to ISA and John Fry for their assistance in obtaining information for this manuscript.
9.9
References
and ZUKER C S (2000), `A novel family of mammalian taste receptors', Cell, 100, 693±702. ARYA S S (1997), `The use of sweeteners in India', in Lisansky S G and Corti A, Lowcalorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 110±113. AUDRIETH L F and SVEDA M (1944), `Preparation and properties of some N-substituted sulfamic acids', J Org Chem, 9, 89±101. AUERBACH M H, LOCKE G and HENDRICK M E (2001), `Alitame', in O'Brien Nabors L, Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 31±40. BAKAL A I (2001), `Mixed sweetener functionality', in O'Brien Nabors L, Alternative ADLER E, HOON M A, MUELLER K L, CHANDRASHEKAR J, RYBA N J P
Low-calorie sweeteners
241
sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 463±480. BAKAL A I and O'BRIEN NABORS L (1986), `Stevioside', in O'Brien Nabors L and Gelardi R C, Alternative sweeteners, New York, Marcel Dekker, Inc., 295±307. BAKKER, D J G (1999), `Consumer behaviour and attitudes towards low-calorie products in Europe', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 146±158. BARTOSHUK L M (1979), `Bitter taste of saccharin: Related to the genetic ability to taste the bitter substance 6-n-propylthiouracil (PROP)', Science, 205, 934±935. BARTOSHUK L M (2000), `Comparing sensory experiences across individuals: Recent psychophysical advances illuminate genetic variation in taste perception', Chem Senses, 25, 447±460. BEAUCHAMP G K (1999), `Factors affecting sweetness', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 10±17. BECK C I (1974), `Sweetness, character, and applications of aspartic acid-based sweeteners', in Inglett G E, Symposium: sweeteners, Westport, CT, The AVI Publishing Company Inc., 164±181. BERTELSEN H, JENSEN B B and BUEMANN B (1999), `D-Tagatose ± A novel low-calorie bulk sweetener with prebiotic properties', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 98±109. BERTELSEN H, HANSEN S J, LAURSEN R S, SAUNDERS J and ERIKNAUER K (2001), `Tagatose', in O'Brien Nabors L, Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 105±127. BLACKBURN G L (1999), `Sweeteners and weight control', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 77±87. BOOTH D A (1987), `Evaluation of the usefulness of low-calorie sweeteners in weight control', in Grenby T H, Developments in sweeteners ± 3, London and New York, Elsevier Applied Science, 287±316. BOOTH D A (1991), `Sweeteners and body weight', in Marie S and Piggott J R, Handbook of sweeteners, Glasgow and London, Blackie and Son Ltd., 248±264. BOPP B A and PRICE P (2001), `Cyclamate', in O'Brien Nabors L, Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 63±86. BOPP B A, SONDERS R C and KESTERSON J W (1986), `Toxicological aspects of cyclamate and cyclohexylamine,' CRC Critical Reviews in Toxicology, 16, 213±306. BORREGO F and MONTIJANO H (2001), `Neohesperidin Dihydrochalcone', in O'Brien Nabors L, Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 87±104. BORREGO F, MONTIJANO H, LOPEZ-CREMADES F J, CANO J and PERIS A (1999), `Double Functionality of Sweeteners: A case to study', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 39±43.
242
Optimising sweet taste in foods
and HENDRICK M E (1983), US Patent No. 4,411,925 (25 October 1983). (1999), `Low-calorie sweeteners ± From molecules to mass markets', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 3±9. BROULIK F J (1996), `Regulatory processes for new sweeteners in the United States of America', in Grenby T H, Advances in sweeteners, Glasgow, Blackie Academic & Professional, 35±55. BROWN A T and BEST G M (1988), `Apparent synergism between the interaction of saccharin, acesulfame K and fluoride with hexitol metabolism by Streptococcus mutans', Caries Res, 22, 2±6. BUFE B, BRESLIN P A S, KUHN C, REED D R, THARP C D, SLACK J P, KIM U-K, DRAYNA D and MEYERHOF W (2005), `The molecular basis of individual differences in phenylthiocarbamide and propylthiouracil bitterness perception', Current Biology, 15, 322±327. BUTCHKO H H, STARGEL W W, COMER C P, MAYHEW D A and ANDRESS S E (2001), `Aspartame', in O'Brien Nabors L, Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 41±61. BRENNAN T M BRIGHT G
BUTCHKO H H, STARGEL W W, COMER C P, MAYHEW D A , BENNINGER C, BLACKBURN G L, DE SONNEVILLE L M, GEHA R S, HERTELENDY Z, KOESTNER A, LEON A S, LIEPA G U, MCMARTIN K E, MENDENHALL C L, MUNRO I C, NOVOTNY E J, RENWICK A G, SCHIFFMAN
S S, SCHOMER D L, SHAYWITZ B A, SPIERS P A, TEPHLY T R, THOMAS J A and TREFZ F K (2002), `Aspartame: Review of safety', Regul Toxicol Pharmacol, 35, S1±S93. CCC (2004), `Calorie Control Council National Consumer Surveys', available at www.caloriecontrol.org/lcchart.html (accessed 26 May 2005). COHEN S M (1997), `Saccharin safety update', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 92±100. CORTI A (1999), Low-calorie sweeteners: Present and future: World conference on LowCalorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG. CRAMMER B and IKAN R (1987), `Progress in the chemistry and properties of rebaudiosides', in Grenby T H, Developments in sweeteners ± 3, London and New York, Elsevier Applied Science, 45±64. DAMAK S, RONG M, YASUMATSU K, KOKRASHVILI Z, VARADARAJAN V, ZOU S, JIANG P,
NINOMIYA Y and MARGOLSKEE R F (2003), `Detection of sweet and umami taste in the absence of taste receptor T1r3', Science, 301, 850±853. DREWNOWSKI A (1999), `Sweetness, appetite, and energy intake: Physiological aspects', in Corti A, Low-calorie sweeteners: Present and future: World conference on lowcalorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 64±76. DUBITSKY M (1996), `Sugar-free chewing gums conquer markets', in Lisansky S G and Corti A, Low-calorie sweeteners: Openings in an expanding European market. Proceedings of a conference organised by the International Sweeteners Association, Brussels, Belgium, Newbury, CPL Press, 22±30. DUFFY V B and SIGMAN-GRANT M (2004), `Position of the American Dietetic Association: Use of nutritive and non-nutritive sweeteners', J Am Diet Assoc., 104, 255±275. ENGELS L and STAGNITTI G (1996), `Flavor modifying composition', WO Patent No. 96/ 17527.
Low-calorie sweeteners
243
(1994), `European Parliament and Council Directive 94/35/EC of 30 June 1994 on sweetness for use in foodstuffs', Official Journal of the European Communities, L237, 3, 10 September 1994. EU (1995a), `European Parliament and Council Directive 95/2/EC of 20 February 1995 on food additives other than colours and sweeteners', Official Journal of the European Communities, L61, 1, 18 March 1995. EU (1995b), `Commission Directive 95/31/EC of 5 July 1995 laying down specific criteria of purity concerning sweeteners for use in foodstuffs', Official Journal of the European Communities, L178, 1, 28 July 1995. EU (1996), `Directive 96/83/EC of the European Parliament and of the Council of 19 December 1996 amending Directive 94/35/EC on sweeteners for use in foodstuffs', Official Journal of the European Communities, L48, 4, 19 February 1997. EU (1998), `Commission Directive 98/66/EC of 4 September 1998 amending Directive 95/31/EC laying down specific criteria of purity concerning sweeteners for use in foodstuffs', Official Journal of the European Communities, L257, 35, 19 September 1998. EU (2000), `Commission Directive 2000/51/EC of 26 July 2000 amending Directive 95/ 31/EC laying down specific criteria of purity concerning sweeteners for use in foodstuffs', Official Journal of the European Communities, L198, 41, 4 August 2000. EU (2001), `Commission Directive 2001/52/EC of 3 July 2001 amending Directive 95/31/ EC laying down specific criteria of purity concerning sweeteners for use in foodstuffs', Official Journal of the European Communities, L190, 18, 12 July 2001. EU (2003), `Directive 2003/115/EC of the European Parliament and of the Council of 22 December 2003 amending Directive 94/35/EC on sweeteners for use in foodstuffs', Official Journal of the European Communities, L24, 65, 29 January 2004. EU (2004), `Commission Directive 2004/46/EC of 16 April 2004 amending Directive 95/ 31/EC as regards E955 sucralose and E962 salt of aspartame-acesulfame K', Official Journal of the European Communities, L114, 15, 21 April 2004. FDA (2001), Agency Response Letter, GRAS Notice No. GRN 000078, Center for Food Safety and Applied Nutrition Office of Food Additive Safety. FEDERAL REGISTER (1969), 34(202), 17063. FEDERAL REGISTER (1970), 35(167), 13644. FEDERAL REGISTER (1984), 49, 6672±6682. FEDERAL REGISTER (1986), 51(188), 34503. FEDERAL REGISTER (1988), 53(145), 28379. FEDERAL REGISTER (1996), 61, 33654±33656. FEDERAL REGISTER (1998), 63(27), 6762. FEDERAL REGISTER (1999), 64(25), 6100. FEDERAL REGISTER (2002), 67, 45310. FEDERAL REGISTER (2005), 70(92), 25496±25502. FLEGAL K M, GRAUBARD B I, WILLIAMSON D F and GAIL M H (2005), `Excess Deaths Associated With Underweight, Overweight, and Obesity', JAMA, 293, 1861±1867. FRANKO D L, WOLFE B E and JIMERSON D C (1994), `Elevated sweet taste pleasantness ratings in bulimia nervosa', Physiology & Behavior, 56(5), 969±973. FRY J (2005), `The future for low-calorie sweeteners', Food Science and Technology, 19(2), 26±30. FRY J C (1996), `Two-in-one ± an innovation in sweeteners', in Lisansky S G and Corti A, Low-calorie sweeteners: Openings in an expanding European market. Proceedings EU
244
Optimising sweet taste in foods
of a conference organised by the International Sweeteners Association, Brussels, Belgium, Newbury, CPL Press, 58±67. FRY J C (1999), `The world market for intense sweeteners', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 201±211. FRY J C and HOEK A C (2001), `Aspartame-acesulfame: Twinsweet', in O'Brien Nabors L, Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 481±498. FRY J C and VAN SOOLINGEN J (1997), `Sweetener salts', European Patent Application EP 0 768 041. GLASER D (1999), `The evolution of taste perception', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 18±38. GLOWAKY R C, HENDRICK M E, SMILES R E and TORRES A (1991), `Development and uses of alitame: A novel dipeptide amide sweetener', in Walters D E, Orthoefer F T and DuBois G E, Sweeteners: Discovery, molecular design and chemoreception, ACS symposium series 450, Washington DC, American Chemical Society, 57±67. GOLDSMITH L A and MERKEL C M (2001), `Sucralose' in O'Brien Nabors L, Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 185±207. GORDON I R (1996), `Cultural and legislative influences on the consumption of high intensity sweeteners in Europe', in Grenby T H, Advances in sweeteners, Glasgow, Blackie Academic & Professional, 273±284. GREEN C (1999), `Thaumatin: A natural flavour ingredient', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 129±132. GREGG E W, CHENG Y J, CADWELL B L, IMPERATORE G, WILLIAMS, D E, FLEGAL K M, NARAYAN,
K M V and WILLIAMSON D F (2005), `Secular trends in cardiovascular disease risk factors according to Body Mass Index in US adults', JAMA, 293, 1868±1874. GRENBY T H (1987), Developments in sweeteners ± 3, London, Elsevier Applied Science Publishers Ltd.. GRENBY T H (1989), Progress in sweeteners, London, New York, Elsevier Applied Science Ltd.. GRENBY T H (1996a), Advances in sweeteners, Glasgow, UK, Blackie Academic & Professional. GRENBY T H (1996b), `Preface', in Grenby T H, Advances in sweeteners, Glasgow, UK, Blackie Academic & Professional, v±vii. GRENBY T H, PARKER, K J and LINDLEY M G (1983), Developments in sweeteners ± 2, London, Elsevier Applied Science Publishers Ltd. GRICE H C and GOLDSMITH L A (2000), `Sucralose ± An overview of the toxicity data', Food Chem Toxicol, 38(2), S1±S6. HA M A, HA T K K and LEAN M E J (1999), `Role of intense sweeteners in diabetes management', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 88±97. HANGER L Y, LOTZ A and LEPENIOTIS, (1996), `Descriptive profiles of selected high
Low-calorie sweeteners
245
intensity sweeteners (HIS), HIS blends, and sucrose', J Food Sci, 61(2), 456±459. (1994), `Sweetener Shake-Out', Food product design, July (accessed at www.foodproductdesign.com on 15 June 2005). HEGENBART S (1996), `Sugar in the mornin'', Food product design, June (accessed at www.foodproductdesign.com on 15 June 2005). HENDRICK M E, MITCHELL H L and MURRAY P R (1996), `Features of alitame as a new high intensity sweetener', in Grenby T H, Advances in sweeteners, Glasgow, Blackie Academic & Professional, 226±239. HIGGINBOTHAM J D (1986), `Talin protein (thaumatin)', in O'Brien Nabors L and Gelardi R C, Alternative sweeteners. New York, Marcel Dekker, Inc., 103±134. HOEK A C, VLEUGELS L F W and GROENEVELD CH (1999), `Improved powder mix quality with Twinsweet', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 133± 139. HOON M A, ADLER E, LINDEMEIER J, BATTEY J F, RYBA N J P and ZUKER C S (1999), `Putative mammalian taste receptors: a class of taste-specific GPCRs with distinct topographic selectivity', Cell, 96, 541±551. HOROWITZ R M and GENTILI B (1963), `Dihydrochalcone derivatives and their use as sweetening agents', US Patent No. 3,087,821 (30 April 1963). HOUGH C A M, PARKER K J and VLITOS A J (1979), Developments in sweeteners ± 1, London, Elsevier Applied Science Publishers Ltd. INGLETT G E (1974), Symposium: Sweeteners, Westport, CT, The AVI Publishing Company, Inc. ISHII H (1999), `Consumer perceptions of products containing sweeteners in Asian countries', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 164±170. JACKSON K G, HOWELLS J and ARMSTRONG J (1987), `Sweeteners in special foods for diabetic, slimming and medical purposes', in Grenby T H, Developments in sweeteners ± 3, London, New York, Elsevier Applied Science Publishers Ltd., 213±262. JECFA (1980), `Evaluation of certain food additives and contaminants. 24th report of the Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 653, Geneva, WHO, 20. JECFA (1982), `Evaluation of certain food additives and contaminants. 26th report of the Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 683, Geneva, WHO, 27. JEFCA (1985), `Evaluation of certain food additives and contaminants. 29th meeting of the Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 29, Geneva, WHO, 35. JECFA (1990a), `Evaluation of certain food additives and contaminants. 37th report of the Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 806, Geneva, WHO, 20. JECFA (1990b), `Evaluation of certain food additives and contaminants. 37th report of the Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 806, Geneva, WHO, 21. JEFCA (1993), `Evaluation of certain food additives and contaminants. 41st Report of the HEGENBART S
246
Optimising sweet taste in foods
Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 837, Geneva, WHO, 17. JECFA (1996), `Evaluation of certain food additives and contaminants. 46th report of the Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 868, Geneva, WHO, 38. JECFA (1998), `Evaluation of certain food additives and contaminants. 51st report of the Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 891, Geneva, WHO, 35. JECFA (2000), `Evaluation of certain food additives and contaminants. 55th report of the Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 901, Geneva, WHO, 12. JECFA (2003), `Evaluation of certain food additives and contaminants. 61st report of the Joint FAO/WHO Expert Committee on Food Additives', WHO technical report series No. 922, Geneva, WHO, 30. JECFA (2004), `Evaluation of certain food additives and contaminants. 63rd report of the Joint FAO/WHO Expert Committee on Food Additives', Summary and Conclusion, Geneva, WHO, 2. JENNER M R (1989), `Sucralose: Unveiling its properties and applications', in Grenby T H, Progress in sweeteners, London and New York, Elsevier Applied Science Ltd., 121±141. JENNER M R (1991), `Sucralose: How to make sugar sweeter', in Walters D E, Orthoefer F T and DuBois G E, Sweeteners: Discovery, molecular design and chemoreception, ACS symposium series 450, Washington DC, American Chemical Society, 68±87. JENNER M R (1996), `The uses and commercial development of sucralose', in Grenby T H, Advances in sweeteners, Glasgow, Blackie Academic & Professional, 253±262. KATHRANI J C (1997), `Saccharin', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 88±91. KIM S-H and DUBOIS G E (1991), `Natural high potency sweeteners', in Marie S and Piggot J R, Handbook of sweeteners, Glasgow and London, Blackie, 116±185. KIM S-H, KANG C-H and CHO J-M (1991), `Sweet proteins. Biochemical studies and genetic engineering', in Walters D E, Orthoefer F T and DuBois G E, Sweeteners: Discovery, molecular design and chemoreception, ACS symposium series 450, Washington DC, American Chemical Society, 28±40. KIM U-K, BRESLIN P A S, REED D and DRAYNA D (2004), `Genetics of Human Taste Perception', Dent Res, 83(6), 448±453. KINGHORN A D (2001), Stevia: The genus stevia, Boca Raton, FL, CRC Press. KINGHORN A D and COMPADRE C M (1991), `Less common high-potency sweeteners', in O'Brien Nabors L and Gelardi R C, Alternative sweeteners. Second edition, revised and expanded, New York, Marcel Dekker, Inc., 197±218. KINGHORN A D and COMPADRE C M (2001), `Less common high-potency sweeteners', in O'Brien Nabors L, Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 209±233. KINGHORN A D and SOEJARTO D D (1991), `Stevioside', in O'Brien Nabors L and Gelardi R C, Alternative sweeteners. Second edition, revised and expanded, New York, Marcel Dekker, Inc., 157±171. KINGHORN A D, WU C D and SOEJARTO D D (2001), `Stevioside', in O'Brien Nabors L,
Low-calorie sweeteners
247
Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 167±183. KNIGHT I (1997), `Applications with sweetener blends', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 114±127. KUNTZ L A (1995), `Achieving flavour parity with alternative sweeteners', Food product design, August (accessed at www.foodproductdesign.com on 15 June 2005). LANTON B R (1999), `Use of intense sweeteners in Africa and nutrition', in Corti A, Lowcalorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 171±177. LEVIN G V (1986), `L-Sugars: lev-o-cal', in O'Brien Nabors L and Gelardi R C, Alternative sweeteners. New York, Marcel Dekker, Inc., 155±164. LINDLEY M (1999), `New developments in low-calorie sweeteners', in Corti A, Lowcalorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 44±51. LINDLEY M, BEYTS P K, CANALES I and BORREGO F (1993), `Flavor modifying characteristics of the intense sweetener neohesperidin dihydrochalcone', J Food Sci, 58, 592±594, 666. LINDLEY M G (1983), `Non-nutritive sweeteners in food systems', in Grenby T H, Parker, K J and Lindley M G, Developments in sweeteners ± 2, London, Elsevier Applied Science Publishers Ltd., 225±246. LINDLEY M G (1996), `Neohesperidine dihydrochalcone: Recent findings and technical advances', in Grenby T H, Advances in sweeteners, Glasgow, Blackie Academic & Professional, 240±252. LINKE H A B (1987), `Sweeteners and dental health: The influence of sugar substitutes on oral microorganisms', in Grenby T H, Developments in sweeteners ± 3, London and New York, Elsevier Applied Science Ltd., 151±188. LISANSKY S G and CORTI A J (1995), Low-calorie sweeteners: Harmonisation in Europe. Proceedings of a conference organised by the International Sweeteners Association, Brussels, Belgium, Newbury, CPL Press. LISANSKY S G and CORTI A (1996), Low-calorie sweeteners: Openings in an expanding European market. Proceedings of a conference organised by the International Sweeteners Association, Brussels, Belgium, Newbury, CPL Press. LISANSKY S G and CORTI A (1997), Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press. MACKAY D A M (1987), `The food manufacturer's view of sugar substitutes', in Grenby T H, Developments in sweeteners ± 3, London and New York, Elsevier Applied Science Ltd., 263±285. MARGOLSKEE R F and MING D (2003), `Inhibitors of the bitter taste response', US Patent 6,540,978. MAYER D and KEMPER F (1991), Acesulfame K, New York, Basel, Hong Kong, Marcel Dekker, Inc. MAZUR R H, SCHLATTER J M and GOLDCAMP A H (1969), `Structure-taste relationships of some dipeptides', J Am Chem Soc, 91, 2684±2691.
248
Optimising sweet taste in foods
(2004), `Taste modification in the biotech era', Food Tech, 58(5), 24±30. and MATSUNAMI H (2002), `Receptors for bitter and sweet taste', Current Opinion in Neurobiology, 12(4), 366±371. MORGAN R W and WANG O (1985), `A review of epidemiological studies of artificial sweeteners and bladder cancer', Food Chem Toxicol, 23, 529±533. NAHON D F, ROOZEN J P and DE GRAAF C (1996), `Sweetness flavour interaction in soft drinks', Food Chemistry, 56, 283±289. NELSON A L (2000), Sweeteners: Alternative, St Paul, AACC. NELSON G, HOON M A, CHANDRASHEKAR J, ZHANG Y, RYBA N J P and ZUKER C S (2001), `Mammalian sweet taste receptors', Cell, 106, 381±390. NOFRE C and TINTI J M (1991), `In quest of hyperpotent sweeteners', in Mathlouthi M, Kanters J A and Birch G G, Sweet-taste chemoreception, London, Elsevier Applied Science Ltd., 205±236. O'BRIEN NABORS L (1999), `Consumer perceptions of products containing sweeteners: USA', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 140±145. O'BRIEN NABORS L (2001a), Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc. O'BRIEN NABORS L (2001b), `Alternative sweeteners: An overview', in O'Brien Nabors L, Alternative sweeteners. Third edition, revised and expanded, New York, Marcel Dekker, Inc., 1±12. O'BRIEN NABORS L and GELARDI R C (1986), Alternative sweeteners. New York, Marcel Dekker, Inc. O'BRIEN NABORS L and GELARDI R C (1991), Alternative sweeteners, Second edition, revised and expanded, New York, Marcel Dekker, Inc. O'BRIEN NABORS L and INGLETT, G E (1986), `A review of various other alternative sweeteners', in O'Brien Nabors L and Gelardi R C, Alternative sweeteners, New York, Marcel Dekker, Inc., 309±323. PARKS E J and HELLERSTEIN M K (2000), `Carbohydrate-induced hypertriacylglycerolemia: Historical perspective and review of biological mechanisms', Am J Clin Nutr, 71, 412±433. PEARSON R L (2001), `Saccharin', in O'Brien Nabors L, Alternative sweeteners, Third edition, revised and expanded, New York, Marcel Dekker, Inc., 147±165. PHILLIPS K C (1987), `Stevia: Steps in developing a new sweetener', in Grenby T H, Developments in sweeteners ± 3, London and New York, Elsevier Applied Science Ltd., 1±43. PRICE J M, BIAVA C G, OSER B L, VOGIN E E, STEINFELD J and LEY H C (1970), `Bladder tumors in rats fed cyclohexylamine or high doses of a mixture of cyclamate and saccharin', Science, 167, 1131±1132. PRICE P (1997), `The advantages of cyclamate', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 82±87. RENWICK A (1997a), `The benefits and safety of intense sweeteners used alone and in combination', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 29±41. MCGREGOR R
MONTMAYEUR J-P
Low-calorie sweeteners
249
(1997b), `Update on cyclamate safety', in Lisansky S G and Corti A, Lowcalorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 102±108. RENWICK A G (1996), `Intense sweeteners and calorie control: The weight of a body of evidence', in Grenby T H, Advances in sweeteners, Glasgow, Blackie Academic & Professional, 208±225. RENWICK A G (1999), `Intake of intense sweeteners', in Corti A, Low-calorie sweeteners: Present and future: World Conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 178±200. RENWICK A G and WILLIAMS R T (1972), `The metabolites of cyclohexylamine in man and certain animals', Biochem J, 129, 857±867. RICHARDS B J (1996), `Essential information on regulatory and legislative matters pertaining to sweeteners in the United Kingdom and the European Union', in Grenby T H, Advances in sweeteners, Glasgow, Blackie Academic & Professional, 26±34. ROBERTS A (1997), `An overview of sucralose', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 76±80. ROGERS P J and BLUNDELL J E (1989), `Evaluation of the influence of intense sweeteners on the short-term control of appetite and caloric intake: A psychobiological approach', in Grenby T H, Progress in sweeteners, London and New York, Elsevier Applied Science Ltd., 267±289. ROLLS B J, DREWNOWSKI A and LEDIKWE J H (2005), `Changing the energy density of the diet as a strategy for weight management', J Am Diet Assoc., 105(5), 98±103. SCHIFFMAN S S, BOOTH B J, LOOSE M L and SATTELY-MILLER E A (1995), `Investigation of synergism in binary mixtures of sweeteners', Brain Res Bull, 38, 105±120. SCHIFFMAN S S, SATTELY-MILLER E A, GRAHAM B G, BOOTH B J and GIBES K M (2000), `Synergism among ternary mixtures of fourteen sweeteners', Chem Senses, 25, 131±140. SCHIFFMAN S S, SATTELY-MILLER E A, GRAHAM B G, ZERVAKIS J, BUTCHKO H H and STARGEL W W (2003), `Effect of repeated presentation on sweetness intensity of binary and ternary mixtures of sweeteners', Chem Senses, 28, 219±229. SCF (1985), Report of the Scientific Committee for Food. Sweeteners, Luxembourg, Commission of the European Communities, 16, 7. SCF (1987), Report of the Scientific Committee for Food on sweeteners, Luxembourg, Commission of the European Communities, 21 (CS/EDUL/-Final), 29, 11 December 1987. SCF (1999), Opinion on stevioside as a sweetener, Brussels, European Commission (17 June 1999). SCF (2002), Opinion of the Scientific Committee on Food: Update on the safety of aspartame, Brussels, European Commission, (SCF/CS/ADD/EDUL/222 Final), 4 December. SCOTT K (2004), `The sweet and the bitter of mammalian taste', Current Opinion in Neurobiology, 14(4), 423±427. SELA M N and STEINBERG D (1989), `Glycyrrhizin: The basic facts plus medical and dental benefits', in Grenby T H, Progress in sweeteners, London and New York, Elsevier RENWICK A
250
Optimising sweet taste in foods
Applied Science Ltd., 71±96. (1997), `Applications with sweetener blends in beverages', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 128± 133. SHIVELY-KNIGHT C (1997), `Intense sweeteners in the global marketplace ± an overview of international regulations', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 43±50. SLOAN A E (2005), `Top 10 global food trends', Food Techn, 59(4), 20±32. SMITH R L, NEWBERNE P, ADAMA T B, FORD R A and HALLAGAN J B (1996), `GRAS flavouring substances 17. 17th publication by the Flavors and Extracts Manufacturers' Association's Expert Panel on recent progress in the consideration of flavouring ingredients generally recognized as safe under the Food Additives Amendment', Food Techn, 50, 72±81. SNODIN D J (1991), `Sweeteners: statutory aspects', in Marie S and Piggott J R, Handbook of sweeteners, Glasgow and London, Blackie and Son Ltd., 265±294. SPILLANE W J (1996), `Molecular structure and sweet taste', in Grenby T H, Advances in sweeteners, Blackie Academic & Professional, Glasgow, 1±25. STARGEL W (1997), `The metabolism, safety and usefulness of aspartame', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 64±75. STARGEL W W, MAYHEW D A, COMER C P, ANDRESS S E and BUTCHKO H H (2001), `Neotame', in O'Brien Nabors L, Alternative sweeteners, Third edition, revised and expanded, New York, Marcel Dekker, Inc., 129±145. STEGINK L D and FILER JR. L J (1984), Aspartame: Physiology and biochemistry. New York, Marcel Dekker, Inc. TEPPER B J, HARTFIEL L M and SCHNEIDER S H (1996), `Sweet taste and diet in type II diabetes', Physiol Behav, 60, 13±18. TINTI J-M and NOFRE C (1991), `Design of sweeteners. A rational approach', in Walters D E, Orthoefer F T and DuBois G E, Sweeteners: Discovery, molecular design and chemoreception, ACS symposium series 450, Washington DC, American Chemical Society, 88±99. TISDEL M O, NEES, P O, HARRIS D L and DERSE H (1974), `Long term feeding of saccharin in rats', in Inglett G E, Symposium: Sweeteners, Westport, CT, The AVI Publishing Company Inc., 145±158. TSCHANZ C, BUTCHKO H H, STARGEL W W and KOTSONIS F N (1996), The clinical evaluation of a food additive: Assessment of aspartame, Boca Raton, FL, CPC Press. VAN DER WEL H (1986), `The sweet proteins', in Hudson B J F, Developments in food proteins ± 4, London, Elsevier Applied Science Ltd., 219±225. VAN DER WEL H and LOEVE K (1972), `Isolation and characterization of Thaumatin I and II, the sweet-tasting proteins from Thaumatococcus daniellii Benth.', J Biochem, 31, 221±225. VON RYMON LIPINSKI G-W (1996), `The blending of sweeteners ± applications and safety issues', in Grenby T H, Advances in sweeteners, Blackie Academic & Professional, SHAMIL S
Low-calorie sweeteners Glasgow, 263±272.
251
(1997), `Properties and applications of acesulfame-K', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 57±62. VON RYMON LIPINSKI G-W (1999), `The role of the Codex alimentarius', in Corti A, Lowcalorie sweeteners: Present and future: World Conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 218±229. VON RYMON LIPINSKI G-W and HANGER LY (2001), `Acesulfame K', in O'Brien Nabors L, Alternative sweeteners, Third edition, revised and expanded, New York, Marcel Dekker, Inc., 13±30. WALKER R (1997), `Risk criteria for risk evalution and tools for risk management', in Lisansky S G and Corti A, Low-calorie sweeteners: Proceedings of a seminar in India organised on February 5±6, 1997, in New Delhi by the International Sweeteners Association and the Confederation of Indian Food Trade & Industry, Newbury, CPL Press, 19±27. WALKER R (1999), `Natural versus `artificial' sweeteners: Regulatory aspects', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 117±124. WALTERS D E, ORTHOEFER F T and DUBOIS G E (1991), `Sweeteners: discovery, molecular design, and chemoreception', ACS symposium series, 450, Washington, DC, American Chemical Society. WEICKMANN J L, LEE J-H, BLAIR L C, GHOSH-DASTIDAR P and KODUIR R K (1989), `Exploitation of genetic engineering to produce novel protein sweeteners', in Grenby T H, Progress in sweeteners, London and New York, Elsevier Applied Science Ltd., 47±69. WELLS, A G (1989), `The use of intense sweeteners in soft drinks', in Grenby T H, Progress in sweeteners, London and New York, Elsevier Applied Science Ltd., 169±214. WILD S, ROGLIC G, GREEN A, SICREE R and KING H (2004), `Global prevalence of diabetes. Estimates for the year 2000 and projections for 2030', Diabetes Care, 27, 1047± 1053. WITT J (1999), `Discovery and development of neotame', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 52±57. YOKOYAMA S M (1999), `Consumer perceptions of products containing sweeteners: Latin America', in Corti A, Low-calorie sweeteners: Present and future: World conference on low-calorie sweeteners, April 25±28, 1999, Barcelona, Spain. World review of nutrition and dietetics, 85, New York and London, S. Karger AG, 159±163. ZIESENITZ S C and SIEBERT G (1986), `Nonutritive sweeteners as inhibitors of acid formation by oral microorganisms', Caries Res, 20, 498±502. VON RYMON LIPINSKI G-W
10 Reduced-calorie sweeteners and caloric alternatives G-W. von Rymon Lipinski, Consultant, Germany
10.1
Introduction
The human demand for sweet tasting foods and beverages has been met over the centuries to a greater and greater extent by the increasing availability of sucrose. Alternatives like honey and concentrated fruit juices have been gradually replaced by cane sugar imported from overseas. The demand for sugar could not be fulfilled when Napoleon blocked international trade, thereby bringing sugar supplies from the West Indies to a halt and at this time the development of alternatives to sugar began. Attention was drawn not only to sugar beet, but also to products of starch hydrolysis and these products became more and more important both as sweeteners and as raw materials for fermentations and functional food ingredients. A sweet tasting compound, glucose, obtained by hydrolysis of starch with mineral acids and identical to `grape sugar' was found in 1811 by Kirchberg (Lippmann 1929). Industrial production of glucose started in 1812 in Germany. Several attempts at commercial production failed as the products were not of sufficient purity and were less sweet than cane sugar. It is unclear whether the process of enzymatic hydrolysis was discovered around the same time, although the development of sweet tasting compounds from starch under the influence of malt had been discovered. An explanation for this process, however, was not possible, as enzymes were not known about at that time. With the availability of industrially produced enzymes cleaving starch without reversion activity, which limits the yield in acid hydrolysis, production of starch hydrolysates like glucose, maltose and syrups high in sweet oligosaccharides grew substantially. It was not much later that enzymes isomerising
Reduced-calorie sweeteners and caloric alternatives
253
glucose to fructose were considered for commercial application in order to produce sweeter syrups with a sweetness coming at least close to invert sugar. Isomerised starch syrups marketed as high fructose corn syrup have not only gained high commercial importance but have even been able to replace sucrose in a number of important applications like soft drinks outside Europe. Pure crystalline fructose has also become available in industrial quantities, but only since approximately the beginning of the last quarter of the 20th century. Lactose is produced from whey and has therefore been available for a long time. It has also been used in sizeable quantities for decades, although as a carrier for tablets rather than as a sweetener. As hydrolysis of lactose results in a sweeter product, lactose seems to be gaining in importance, but rather slowly. While the aforementioned products are fully metabolised in the human body, partially digestible, low-glycaemic and non-cariogenic carbohydrates have only aroused interest in the last few decades. Some of the partially digestible sweeteners developed, such as L-sugars or leucrose, have more or less been abandoned. Others, such as tagatose and isomaltulose, are still under development but have already found their way into commercial products. All products discussed here are bulk sweeteners and therefore have the potential to substitute sucrose. As a consequence, the functional properties of the reduced-calorie and caloric sweeteners are as important as their sweetness, as the functional properties determine their application. Particularly important properties in this context are solubility, equilibrium humidity and melting point in aqueous solution. A survey of some important properties is given in Table 10.1. Table 10.1 Some important properties of reduced-calorie sweeteners and caloric alternatives Sweetener
Sweetness
Solubility
(ca. values, (g/kg ca. values) 10% sol.) Fructose Galactose Glucose Isomaltulose Lactose Lactulose Leucrose Maltose Sorbose Tagatose Trehalose Sucrose * anhydrous D decomposition
1.2 0.65 0.6 0.5 0.25 0.5 0.5 0.35 0.6 0.9 0.4 1.0
800 680 500 290 150 950 600 500 450 550 430 660
Melting range (ëC)
Reduced calories
Reducing sugar
100±104 167 83 (146*) 123±124 202 (253*) 165±166 161±163 102±103 159±161 133±137 94±100 (210*) 185±186 D
no no no no no yes no no yes yes no no
yes yes yes yes yes yes no yes yes yes no no
254
Optimising sweet taste in foods
The technological properties of these alternatives are normally no obstacle for food use. Most of these alternatives are, however, much less sweet than sucrose and require combined use with intense sweeteners whenever products of the customary sweetness are required.
10.2
Reduced-calorie sweeteners
The conventional sweet carbohydrates sucrose, glucose and fructose are fully absorbed from the intestine when consumed in normal quantities. However, some other mono- and disaccharides are absorbed less readily which gives them a lower calorific value, as only part of the carbohydrate becomes available for direct energy gain. Microbial degradation of the unabsorbed part also makes a limited energy contribution. Several such products have been studied and developed over the course of the years. 10.2.1 D-tagatose Occurrence, chemical and physical properties Tagatose is a keto-sugar which is found in nature in small quantities. In dairy products, for example, it is formed from lactose during heating. It is a structural epimer to fructose inverted at C 4. Small amounts of tagatose are formed when dairy products are heated or stored for longer periods (Bertelsen et al. 2001a, Levin 2002). Tagatose forms white crystals similar to sucrose or other monosaccharides. Crystallisation results in anhydrous crystals of the -pyranose form. The solid form of tagatose melts in the range of 133±137ëC. The specific rotation is []20D: ÿ4 to ÿ5.6ë. Its equilibrium humidity is about 80% rh, in a similar range to that of sucrose and tagatose is therefore non-hygroscopic when stored under normal conditions. The solubility of tagatose at room temperature is around 550 g/kg of solution at 20ëC and increases to more than 850 g/kg of solution at 90ëC. The heat of solution is ÿ42.3 kJ/kg at 20ëC and ÿ84.1 kJ/kg at 37ëC which means that it is lower than the heat of solution of fructose but substantially higher than that of sugar alcohols. The negative heat of solution results in a cooling effect, but it is not as pronounced as for most sugar alcohols. In aqueous solution mutarotation results in an equilibrium of 71.3% -pyranose, 18.1% -pyranose, and smaller quantities of pyranoses and the open form. The viscosity of concentrated solutions of tagatose lies between the viscocity of sucrose and fructose. As a keto-sugar, tagatose is a reactive compound and more prone to caramelisation than sucrose. It also reacts readily in browning reactions. Stability seems good in the pH range of 3 to 8 and is therefore sufficient for common food applications. The sweetness intensity of tagatose is ca. 92% compared to a 10% solution of sucrose and therefore only slightly lower. In combination with aspartame or combinations of acesulfame K and aspartame tagatose shows synergy. It is claimed not only to reduce bitterness but also to increase sweetness
Reduced-calorie sweeteners and caloric alternatives
255
and to provide mouth-feel. These effects and also positive flavour modifications were observed at levels as low as 0.2% (Bertelsen et al. 2001a). Production Tagatose production starts from lactose, which is enzymatically hydrolysed to galactose and glucose. The mixture of monosaccharides is separated by chromatography, and galactose is converted to tagatose in the presence of calcium hydroxide. The calcium hydroxide is reacted with sulphuric acid to form calcium sulphate, which is subsequently removed. After the purification the solution is concentrated, and tagatose is crystallised (Levin 2002). Physiology Despite its similarity to fructose tagatose is only partly absorbed in the intestine. Radio-labelling showed that approximately 20% is passively absorbed by diffusion. This part is metabolised in the liver like fructose. While both fructose and tagatose are transformed to 1-phosphates, the rate of cleavage of the tagatose phosphate is only approximately 10% of the rate of cleavage of the fructose phosphate. This results in partial excretion of unmetabolised tagatose, which is assumed to be 20% of the absorbed amount or 5% of the ingested quantity (Normen et al. 2001). No interference by tagatose in fructose absorption and fructose metabolism was found after intake of tagatose-containing foods. Part of the unabsorbed tagatose can be fermented by the bacteria of the intestine, at least after adaptation. The caloric value of the absorbed part is assumed to be 3.75 kcal/g and studies aimed at determination of the metabolisable energy showed an energy value between 1.1 and 1.4 kcal/g. In the USA an energy value of 1.5 kcal/g is accepted for tagatose (Bertelsen et al. 2001a). Tagatose was found to have an influence on intestinal flora not only in animals but also in humans. Lactic acid bacteria and related organisms increased and pathogenic bacteria decreased. Therefore tagatose is claimed to be prebiotic (Bertelsen et al. 2001b). Owing to its limited absorption, tagatose has laxative properties when ingested in large amounts. Single doses of 20 g or divided doses of 30 g per day were tolerated without negative symptoms by the majority of test persons, and few experienced the symptoms known to be caused by osmotically active poorly absorbed substances (Buemann et al.1999a,b). The slow and incomplete absorption of D-tagatose results in a low glycemic impact. Consumption of 50 g portions of tagatose or glucose, dissolved in 200 mL water by 12 healthy persons resulted in relative glycaemic and insulinemic responses of only 3% (glucose gives a glycaemic response of 100%). After intake of up to 75 g of tagatose, an increase in plasma glucose or serum insulin was not found either in healthy people or in people suffering from diabetes Type II. If administered together with glucose, D-tagatose lowered the glycemic index of glucose by about 20%. When consumed before glucose or starch, tagatose has an attenuating effect on blood glucose levels (Donner et al.1999; Madenokoji et al. 2003; Szepesi 1996).
256
Optimising sweet taste in foods
The influence on blood uric acid levels observed after intake of high doses of tagatose could not be found when only 15 g were consumed. Fifteen grams is considered the normal daily intake of tagatose (Buemann et al. 2000). Studies on the potential cariogenicity using telemetric techniques for pH determination showed no critical decrease of plaque after consumption of a 10% solution of tagatose. Adaption of oral bacteria to tagatose did not change the results. It can therefore be considered non-cariogenic (Bertelsen et al. 2001a). Applications Tagatose is suitable to directly replace sucrose or monosaccharides in conventional products (Bertelsen et al. 2001a), including a variety of confectionery products such as chocolate, fondant, fudge, caramel and chewing gum. When replacing sucrose or monosaccharides in hard-boiled candy and gums, cooking under vacuum and combining tagatose with equal amounts of other sweeteners is recommended as tagatose promotes browning. Tagatose may be added to dairy and related products for its prebiotic properties. When added as a prebiotic, quantities should result in an intake of several grams per day. The flavour-modifying properties of tagatose can be exploited in beverages, especially low-calorie soft drinks to improve the sweetness profiles of the intense sweeteners used in these products. Only a few grams per litre in the beverage are required for this purpose (Bertelsen et al. 2001a). Applications in pharmaceuticals have also been proposed. These include lozenges, syrups and effervescent and chewable tablets (Levin 2002). Regulatory status While tagatose was originally evaluated as a food additive, it is now considered a novel food. Former evaluations by the Joint FAO/WHO Expert Committee on Food Additives resulted in numerical temporary acceptable intake (ADI) values. During the most recent evaluation a `not specified' ADI was allocated (Anonymous 2004). In the USA tagatose has GRAS status (Anonymous 2001a). Several countries like Australia and New Zealand or Korea permit tagatose as a food or novel food. In the European Union marketing as a novel food ingredient has become possible since 2005. 10.2.2 L-sugars Occurrence, chemical and physical properties With the exception of a few compounds, L-sugars are not very common in nature. The L-sugars of glucose, fructose and galactose in particular have been proposed for use as non-caloric sweetening agents, but several other L-sugar monosaccharides have also been described as potential sweeteners (Levin 1981). L-sugars are enantiomeric to the common D-sugars and are therefore identical in chemical and physical properties except for the rotation of polarised light passing through their solutions. They undergo equivalent reactions to their common counterparts.
Reduced-calorie sweeteners and caloric alternatives
257
The tastes of 10% solutions of various enantiomeric pairs of D- and L-sugars were compared by a taste panel. The solutions were all rated as rather sweet and the intensity of sweetness of a given D-sugar and its enantiomer were not distinguishable. The pairs tested included arabinose, xylose, glucose, rhamnose, galactose, glucoheptulose and fructose (Shallenberger et al. 1969). Production Several synthesis routes for L-sugars have been described. 4-Deoxypentenosides, which are readily derived from D-sugars, resemble glycals in structure and reactivity and can undergo stereoselective epoxidation and nucleophilic addition to produce L-sugars (Boulineau and Wei 2002). A novel route with L-ascorbic acid as a single common starting material for asymmetric synthesis of all eight diastereomers of L-hexoses has been described. Key steps involve stereoselective preparation of chiral hydroxy-, -unsaturated esters and their stereocontrolled dihydroxylation by OsO4 (Ermolenko et al. 2003). A specific multistep synthesis starting from L-ascorbic acid has been described for L-galactose (Kim et al. 2002). Enzymatic isomerisation of L-sugars is also possible (Horwath and Colonna 1984), and several laboratory studies on enzymatic and microbiological syntheses have also been published. Although some quantities of a few substances have been produced, it seems that no process suitable for large-scale production has been established, and supply seems limited to small quantities for research purposes. Physiology Although it was originally assumed that L-sugars would be non-caloric in energy balance, studies using 10% of L-sugars in the diet have shown that only L-glucose contributed virtually no energy (0:3 0:9 kJ/g), while L-fructose (6:9 0:9 kJ/g) and L-gulose (8:8 1:8 kJ/g) contributed 40% or more than 50%, respectively, of the energy of D-sugars. The assimilation of energy from L-fructose and L-gulose is assumed to be attributable to large bowel micro-organisms (Livesey and Brown 1995). L-monosaccharides have laxative properties and doses of 1±12 g per day were proposed as laxatives (Tarka et al. 1993). Applications L-sugars could be used in the same quantities as their D-counterparts as they have the same sweetness and technical properties. As they have not become commercially available, though, no applications can be described. 10.2.3 L-sorbose Occurrence, chemical and physical properties A particular L-sugar, L-sorbose, since it is a keto-sugar, was the subject of particular interest some time ago. L-sorbose is not only present in the fruits of sorbus species like the mountain ash (Sorbus aucuparia L.) but has become
258
Optimising sweet taste in foods
available also in large quantities as an intermediate of the commercial vitamin C synthesis. Sorbose forms colourless to white crystals of the -pyranoidic form with a melting range of 159±161ëC. The specific rotation is []20D: +43.4ë and crystalline sorbose is non-hygroscopic. The solubility of sorbose in water is ca. 450 g/ kg of solution at 20ëC and above 600 g/kg solution at 80ëC. At room temperature mutarotation results in only ca. 2% of crystals not of the -pyranoidic form. As a keto-sugar, sorbose behaves similarly to fructose but is more stable and reacts more slowly. Depending on the concentration in solution, sorbose has 60± 75% of the sweetness of sucrose. Synergistic sweetness effects are observed in blends with other sweetening agents, especially with sugar (Schiweck 1991b). Production Production of sorbose starts from D-glucose which is hydrogenated to Dsorbitol. Fermentative dehydration by microorganisms like Acetobacter yields L-sorbose (Reichstein and Gruessner 1934). Physiology Sorbose is only slowly absorbed and no increase in blood glucose levels was found after sorbose intake. It is therefore suitable for diabetics (Vidal-Sivilla 1958). Sorbose is not metabolised by the bacteria of the oral cavity and is therefore non-cariogenic (Havenaar et al. 1984). Haemolysis observed in animals after intake of large amounts of sorbose raised concerns about its safety for human consumption and these concerns have prevented sorbose from being used as a sweetener (Baer and Leeman 1999; Keller and Kistler 1977). Applications The properties of sorbose, especially its high melting point and lower reactivity compared to fructose would allow its use in foods and beverages. This has especially been shown in studies with confectionery products like hard and soft candy, chocolate and chewing gum or bakery products (Schiweck 1991b). It seems, however, that sorbose is not commercially used in foods. 10.2.4 Lactulose Occurrence, chemical and physical properties Lactulose, 4- -D-galactopyranosido-D-fructose, is formed during sterilisation of milk and dairy products and can serve as an indicator for sterilisation. Lactulose crystallises in three tautomeric forms, 74.5% -fructofuranosidic, 10% fructofuranosidic and 15.5% -fructopyranosidic, making its crystallisation very difficult to achieve (Schiweck 1991b). Its melting range is 165±166ëC. The specific rotation is []20D: ÿ50.4ë. The solubility in water is very high and reaches 95% (w/w) (Schiweck 1991b). At the low concentration of 5% (w/w) lactulose is almost half as sweet as sucrose, while at higher concentrations the sweetness increases to more than 60% of sucrose sweetness. Synergism up to 22% was found in blends of lactulose and sucrose (Parrish et al. 1979).
Reduced-calorie sweeteners and caloric alternatives
259
Production Production of lactulose starts from lactose or sweet whey and uses alkaline isomerisation with sodium hydroxide or organic amines, preferably in the presence of complexing agents like boric acid (Hicks and Parrish 1980; Hicks et al. 1984; Schiweck 1991b). After removal of salts the solution is concentrated and purified by chromatography. Continuous isomerisation is possible (Kozempel et al. 1995). As crystallisation is difficult, normally a lactulose syrup is produced. Physiology Only minute fractions below 3% of lactulose are absorbed in the intestine but intestinal bacteria are able to transform lactulose into lactic acid and short-chain fatty acids which can be partly utilised by the body. The energy value is therefore below 1 kcal/g. Lactulose is used in the treatment of constipation and hepatic encephalopathy as its fermentation in the colon by certain lactic acid bacteria has a local osmotic effect in the colon which results in increased faecal bulk. Fermentation of lactulose in the intestine counteracts detrimental species such as clostridia or salmonella and it is referred to as a bifidogenic factor (Schumann 2002). Applications Although many food applications have been described for lactulose, food uses outside the area of weaning and infant food seem very limited. An important field of use is in laxating pharmaceuticals. For these often a syrup containing 66.7% lactulose is produced (Kozempel et al. 1995). The recommended dose is in the range of 15±25 g/day.
10.3
Alternative caloric sweeteners
With increasing interest in the physiological properties of sweeteners, not only non-caloric and calorie-reduced products but also some of the carbohydrates which have been known for many years have been re-examined for their suitability for food use. Products with a low glycaemic index are presently being developed and have started to find their way into foods. Among these, trehalose and isomaltulose should especially be mentioned 10.3.1 Isomaltulose Occurrence, chemical and physical properties Isomaltulose, 6-O--D-glucopyranosyl-D-fructofuranose, is a reducing disaccharide consisting of a glucose molecule and a fructose molecule. It is naturally occurring in honey in levels up to 1% and can also be found in extracts of sugar cane (Irwin and Straeter 2001). Isomaltulose is produced in large quantities as an intermediate in the production of isomalt (Schiweck 1991c). The quantities used in foods are
260
Optimising sweet taste in foods
presently fairly small compared to those processed further. Isomaltulose has been used in Japan for some time and is also increasingly of interest as an alternative to sucrose in Europe and the USA. Isomaltulose is a colourless to white crystalline substance. It crystallises as a monohydrate of the -furanoid form, while in water at 34ëC, 20% and at 65ëC, 33% of the -furanoid form is observed. Its melting range is 123±124ëC. The specific rotation is []20D: +103± 104ë. Isomaltulose is virtually non-hygroscopic at 80% rh (Irwin and Straeter 2001; Schiweck 1991b,c). The solubility of isomaltulose at 20ëC is 290 g/kg of solution and increases to more than 700 g/kg of solution at 90ëC. The heat of solution is ÿ60.2 kJ/kg, which results in a cooling effect. Aqueous solutions of isomaltulose have a viscosity almost identical to sucrose solutions of the same concentration and temperature (Irwin and Straeter 2001; Schiweck 1991b,c). The acid catalysed hydrolysis of isomaltulose is slow compared to sucrose hydrolysis. A 20% (w/w) solution of isomaltulose adjusted to pH 2.0 was stable for 1 h at 100ëC, while a sucrose solution was completely inverted under the same conditions. As it is a reducing sugar, isomaltulose is not as stable as non-reducing sugars and undergoes Maillard reactions (Irwin and Straeter 2001). The sweetness intensity of isomaltulose is 48% compared to a 10% (w/w) solution of sucrose. It increases above 25% (w/w) and exceeds 50% of the sucrose sweetness at concentrations above 30% (w/w). The sweetness intensity is independent of the temperature. The taste profile of isomaltulose is similar to sucrose. It is also reported to mask off-flavours of other sweeteners (Irwin and Straeter 2001; Schiweck 1991b). Production Isomaltulose production starts from sucrose. It is formed in an enzymatic process using sucrose isomerase whereby the 1,2-glycosidic linkage between glucose and fructose is rearranged to form a 1,6-glycosidic linkage. Commercially the enzyme of Protaminobacter rubrum is used, although several other organisms also contain active enzymes (Irwin and Straeter 2001; Munir 1987; Sugitani et al. 1994; Wu and Birch 2004). The enzyme is either used in immobilised cells in reactor columns, in inactivated cells added to the sucrose solution and later removed by filtration or as an immobilised enzyme. The process allows use of fairly concentrated solutions of 50% (w/w) and more. Unreacted sucrose can be removed by fermentative degradation. Afterwards the solution is clarified and crystallised by evaporation and cooling. Re-crystallisation yields material of the required purity. Physiology In the human body isomaltulose is metabolised by the isomaltase-sucrase complex in the intestinal mucosa to glucose and fructose, which are identical to the glucose and fructose produced in the metabolism of sucrose. The small intestinal hydrolysis is, however, much slower than that of sucrose. but the
Reduced-calorie sweeteners and caloric alternatives
261
process continues to completion. The hydrolysis products glucose and fructose are absorbed and metabolized in the same way as those from other sources. Due to its full digestibility, isomaltulose provides the same physiological energy as sucrose (Lina et al. 2002). As isomaltulose is completely though slowly digested in the intestine, gastrointestinal intolerances even at doses up to 50 g/d have not been reported from human studies. The insulin response and plasma glucose levels following isomaltulose intake are significantly lower than those following glucose intake due to the slower rate of hydrolysis of isomaltulose in the small intestine. This has been studied in healthy as well as diabetic subjects. Isomaltulose has therefore a lower glycaemic index than sucrose (Kawai et al. 1985, 1989). Thus, isomaltulose is suitable for people with glucose intolerance diseases such as diabetics or preconditions. It has also been proposed for enteral nutrition (Kowalczyk et al. 2004). Isomaltulose does not promote growth of bacteria of the oral cavity. Studies on fermentation in vitro, in vivo and in situ pH telemetry studies showed that isomaltulose is not cariogenic. In addition, there are results which suggest that isomaltulose inhibits the formation of polymer glucans. Thus, it can be considered as tooth-friendly (Doerr et al. 2005; Irwin and Straeter 2001; Ooshima et al. 1990; Sasaki et al. 1985). Applications Substitution of sucrose with isomaltulose is possible without problems in many applications. These include bakery products, cereals, bars, confections, hard and soft candy, chocolate, chewing gum, sports, energy and dairy drinks and oral health products. The solubility of isomaltulose is sufficient for these applications. The lower solubility compared to sucrose may be a problem for products requiring a high solids level like jams, jellies and marmalades. Use of isomaltulose can be advantageous in cultured dairy products like yoghurts. The cultures used in these products like Lactobacillus acidophilus or Lactobacillus bifidus cannot metabolise isomaltulose. As a consequence, the stability of the products is improved and the sweetness level remains stable. The low hydrolysis in acidic beverages is an advantage over sucrose (Bernard and Kowalczyk 2005; Doerr et al. 2005; Irwin and Straeter 2001). The use of isomaltulose instead of sucrose in teas for young children allows production of palatable non-cariogenic products not causing the caries attributed to frequent consumption of conventional teas for children (Doerr et al. 2005). Isomaltulose has been used in a variety of products in Japan for several years. Regulatory status In the EU isomaltulose has been consumed in small quantities only as a constituent of different naturally occurring products like honey. It was therefore a novel food ingredient requiring pre-market approval. Isomaltulose has recently been granted this approval for general food use following Good Manufacturing Practice (Anonymous 2005). In the USA isomaltulose has self-affirmed GRAS status. It is also approved in Japan, Korea China and Taiwan.
262
Optimising sweet taste in foods
10.3.2 Trehalose Occurrence, chemical and physical properties Trehalose, -D-glucopyranosyl--D-glucopyranoside, consists of two glucose molecules linked in 1.1-position. It is a naturally occurring disaccharide, found in small quantities in several plants and animals. Reported levels in some foods are 0.9±1.9% in honey, 8±17% in commercially grown mushrooms and 15±20% in baker's yeast which results in the presence in bread up to ca. 1.5 g/kg of dry matter and especially in trehala manna, a secretion of beetles in the Iraqi desert which was collected by Bedouins and used as a sweetening agent. It has therefore been consumed in small quantities for a long time (Richards and Dexter 2001). Although trehalose has been known for many decades, interest in the compounds only increased recently when commercial quantities became available. It is used as a functional food ingredient rather than as a sweetener. Trehalose forms colourless to white crystals which melt in the range of 94±100ëC. During further heating the crystal water vaporises and the product solidifies again. The melting point of anhydrous trehalose is 210.5ëC. The specific rotation is []20D: +199ëC. Trehalose dihydrate is non-hygroscopic with an equilibrium humidity of 92% rh. The solubility of anhydrous trehalose is around 430 g/kg of solution at 20ëC and increases to ca. 800 g/kg of solution at 90ëC. The heat of solution of the dihydrate is +20.7 kJ/kg. It therefore does not have a cooling effect. Aqueous solutions of trehalose have a low viscosity. In 40% (w/w) solution viscosity is 5.7 cP at 20ëC and 4 cP at 37ëC and therefore similar to sucrose (Richards and Dexter 2001; Richards et al. 2002). As a non-reducing sugar trehalose is rather stable during storage and in solutions of pH 2 to 10. It does not participate in Maillard reactions (O'Brien 1996). The sweetness intensity of trehalose is 40% compared to a 10% (w/w) solution of sucrose. It increases slightly to 45% compared to sucrose for 22% (w/ w) solutions. The sweetness of trehalose is clean but persists longer than the sucrose taste. This effect becomes stronger with increasing concentration (Portmann and Birch 1995; Richards and Dexter 2001). Production Trehalose production starts from liquefied starch obtained by treatment with amylase. Isoamylase is used as a de-branching enzyme to cleave -1,6 linkages. The resulting maltooligosaccharides are treated with maltooligosyl trehalose synthase, which converts the -1.4 linkage at the reducing end of the oligosaccharide into a -1,1 linkage. Maltooligosyl trehalose trehalohydrolase hydrolyses the -1,4 linkage between the oligosaccharide moiety and the trehalose. The reaction mixture is decolorised with activated carbon, de-ionised with ion exchange resins, and concentrated by evaporation. Trehalose is obtained by crystallisation as the dihydrate (Nishimoto et al. 2002). Other potential production methods like fermentation, enzymatic conversion of maltose with phosphorylases or combined glucose acid reversion and enzymatic synthesis are of no commercial importance.
Reduced-calorie sweeteners and caloric alternatives
263
Physiology In the human intestine trehalose is rapidly cleaved to glucose by the enzyme trehalase. The resulting glucose is absorbed and metabolised just like glucose from other sources and provides the same energy (Ushijima et al. 1995). Single oral doses of up to 20±30 g as well as 40 g consumed within several hours were tolerated without the occurrence of gastrointestinal symptoms by healthy persons. Symptoms such as flatulence, watery stool and distension were reported to occur at higher doses (Elbein 1974; Richards et al. 2002). Trehalasedeficient persons may experience such symptoms after ingestion of smaller amounts of trehalose due to the osmotic activity of undigested trehalose in the gut. However, smaller amounts of trehalose are tolerated by such individuals without any such symptoms. Just as in the case of isomaltulose, insulin response and plasma glucose levels following trehalose intake are significantly lower than those following glucose intake, owing to its retarded digestion. The manufacture of trehalose-containing foods for diabetics has therefore also been proposed (Cooper et al. 2001). Diabetics will, however, have to add trehalose calories to their intake calculations. Trehalose has been proposed for parenteral nutrition, as it will be cleaved to glucose by trehalase which, depending upon the species, takes place in the serum, kidney, liver and bile. Any unmetabolised trehalose would be excreted with the urine (Sato et al. 1999). Trehalose can be fermented by Streptococcus mutans, but at a lower rate than sucrose. Telemetric record of plaque pH changes after consumption of trehalose-containing lozenge or chocolate showed that the pH did not reach levels below 5.7. Foods sweetened with trehalose can therefore be considered without promoting caries (Richards and Dexter 2001). Applications While trehalose can be used as a sweetener to replace conventional sweetening agents, like sucrose many of its applications are based on its technological functions. Trehalose stabilises proteins against denaturation such as the damage caused by freezing or drying and stabilises disulfide bonds, which prevents development of off-flavours. Coagulation of proteins during sterilisation is also reduced and levels of a few per cent are normally sufficient for such stabilising effects (Aguilera and Karel 1997). Starch retrogradation is retarded by trehalose in noodles, bread and cake. Again, levels well below 10% are required. In sliced fruit, fruit powder and dried herbs, addition of trehalose before drying resulted in dried products of good shelf stability which were easy to reconstitute and had better flavour than comparative products without trehalose (Roser 1991). In all these applications the sweetness provided by trehalose is limited if recognisable at all, as smaller quantities than those necessary for sweetening purposes are required. Regulatory status Evaluation by the Joint WHO/FAO Expert Panel for Food Additives resulted in the allocation of a `not specified' ADI (Anonymous 2001b). In the EU trehalose
264
Optimising sweet taste in foods
is basically a food ingredient not requiring approval. Trehalose produced by a novel process has, however, been approved as a novel food ingredient (Anonymous 2001c). In the USA trehalose is GRAS (Anonymous 2000). It is also approved in Japan, Korea and Taiwan. 10.3.3 Hydrolysed lactose and galactose Chemical and physical properties Owing to its limited sweetness and solubility lactose is used as a bulking agent rather than as a sweetener. Its two components, D-glucose and D-galactose are approximately twice as sweet as lactose. Therefore hydrolysis can increase the sweetness substantially. A variety of different forms of lactose have been discussed in the literature, and usually syrups rather than not dried products are preferred as the hydrolysed products are hygroscopic. The solubility of hydrolysed lactose in water at room temperature is ca. 37.5% (w/w) at room temperature. This is therefore higher than the solubility of lactose. D-galactose has a melting point of 167ëC. It is freely soluble in water, in which 68% (w/w) solutions can be prepared at 25ëC. It undergoes mutarotation resulting in a specific rotation of []20D: +80.2ë. Galactose has ca. 65% of the sweetness of sucrose, while syrups, based on dry matter, have an only a slightly lower sweetness at around 60%. The hydrolysed syrup contains two reducing sugars and is therefore much more prone to caramelisation and Maillard reactions than lactose (Boesig 1991). Production Lactose can be hydrolysed with -galactosidases or acids. Suitable acid catalysts for lactose hydrolysis are sulphuric acid or the acid form of cationic exchange resins. Enzymatic hydrolysis is the alternative production method and creates lower quantities of by-products. Enzymatic hydrolysis should be used for lactose hydrolysis in sweet whey syrups or foods. Suitable enzymes can be obtained from several organisms. They can be added to the substrate or used as carrierbound preparations. Thermo-stable enzymes with high conversion rates are available (Greenberg and Mahoney 1981; Mahoney 1985). The syrups can be used without further modification, in concentrated form or can be demineralised with ion exchangers. Galactose can be obtained by chromatographic separation of the hydrolysate. A feasible alternative is to add the enzyme to the food or to bring an immobilised enzyme in contact with the food, therefore producing the sweeter monosaccharides in situ. Physiology The glucose part of lactose hydrolysates is readily absorbed and metabolised in the same way as glucose from other sources. Galactose is actively absorbed from the intestine and metabolised in the body. For small quantities the metabolism is as fast as the metabolism of glucose while large doses are metabolised much more slowly. Part of the galactose may then be excreted in the urine. Large
Reduced-calorie sweeteners and caloric alternatives
265
doses lead to increased excretion of calcium and magnesium but also improve absorption of calcium. The human tolerance for oral ingestion is assumed to be 40±80 g/d (Flynn 1985; Lang 1979c). Hydrolysis improves tolerance for those with lactose intolerance caused by absence of galactosidase in the intestinal mucosa. Applications Syrups of hydrolysed lactose seem to have limited importance. In situ hydrolysis to sweeten dairy products like milk or yoghurt started around 1975. Speciality products with hydrolysed lactose are on the market in several countries (Holsinger and Kligerman 1991; Zadow 1986). 10.3.4 Leucrose Occurrence, chemical and physical properties Leucrose, 5- -D-glucopyranosido-D-fructopyranose, is produced by Leuconostoc mesenteroides and was discovered in fermentation broth containing this organism. Leucrose crystallises in the -fructopyranosidic form and does not mutarotate in aqueous solution. Its melting range is 161±163ëC. The specific rotation is []20D: ÿ6.8ë. The solubility in water is around 60% (w/w) at room temperature and increases to around 90% (w/w) at 90ëC (Schiweck 1991b). Leucrose is approximately half as sweet as sucrose. Production Leucrose can be produced by reaction of saccharose with -(1-6)-glucosyl transferase in the presence of fructose with at least a partial separation of the dextrans and iso-malto-oligosaccharides formed as by-products. Leucrose prepared by this method may have a purity of at least 98% (Schwengers and Benecke 1987). Physiology Leucrose was found to be easily digestible when given to humans as a single oral dose of 100 g, or when fed to rats at a level of 35 g/kg body wt daily. Weanling rats fed a 25% leucrose diet grew as well as rats fed a diet containing 25% sucrose or corn starch. The cleavage rate of leucrose in vitro by human digestive carbohydrases was 31% that of maltose and 63% that of sucrose. Blood glucose and fructose profiles in humans consuming leucrose tended to be lower than those in humans given sucrose, while insulin and C-peptide profiles were unaltered (Ziesenitz et al. 1989a). Despite its good digestibility leucrose was found to be non-cariogenic. No acid formation was observed after incubation of leucrose with suspensions of human dental plaque. Leucrose was a competitive inhibitor of acid formation from sucrose by Streptococcus mutans at neutral pH. Furthermore, it considerably inhibited the uptake of sucrose by S. mutans. In contrast to a group of rats fed with 30% sucrose, caries scores of a group fed with 30%
266
Optimising sweet taste in foods
leucrose were not significantly different from the starch group. In man, pH telemetry showed that plaque pH did not drop below pH 5.7 (Ziesenitz et al. 1989b) In a later study, however, it was found that some micro-organisms of the oral cavity could cleave leucrose to produce glucose and fructose which might be used as substrates for acid formation by bacteria contributing to caries formation (Peltroche-Llacsahuanga et al. 2001). Applications Although a number of potential applications to replace sucrose were described it seems that no commercial quantities have become available.
10.4
Established caloric alternatives
Several sweet carbohydrates which were discovered a long time ago have generated much more interest in recent decades as new technologies for their production have become available. Products of starch hydrolysis have not only gained substantial market shares in some countries, but have even replaced sucrose in a number of important applications like soft drinks. Among these sweet carbohydrates are oligosaccharide syrups as products of starch hydrolysis, high fructose corn syrups and, to a lesser extent, though they are still important, glucose and fructose. All these are used as sweetening and functional ingredients. Lactose is less sweet and therefore often used as a bulking ingredient and filler, rather than as a sweetener. Although these are not new products, their uses in foods have expanded markedly in the course of the last two to three decades. 10.4.1 Starch hydrolysates Properties Starch hydrolysis is an old technology, established almost 200 years ago. With the availability of industrially produced enzymes, a much wider variety of products has become available. Between maltodextrins and glucose, many products with different characteristics are now available. These products are normally characterised by their production route and the DE ( dextrose equivalent) value, a virtual glucose content in per cent determined by conventional chemical methods for reducing sugars (Table 10.2). Commercial availability of glucose isomerase led to the development of processes for the production of high-fructose corn syrups as a cheaper alternative to sucrose for industrial uses in a variety of countries. Starch hydrolysates can have very different properties, depending on the production process and the degree of starch degradation. They are normally characterised by their DE value, and virtual glucose level in drying matter determined by conventional analytical methods for the analysis of reducing sugars (Anonymous 2002; Tegge 1984). They are normally supplied as syrups concentrated to at least 70% dry matter, as they may be hygroscopic, but dried products are also available. Syrups can be
Table 10.2 Composition of starch hydrolysates DE value Carbohydrate*
28 34m 36 Level in dry matter in %
43
43m
53
63
66
95
DP 1 DP 2 DP 3
8 8 9
19 15 12
8 33 19
28 20 20
36 31 13
40 35 8
96 2.5 0.5
9 34 24
DE: dextrose equivalent m: maltose-rich syrups * DP 1: glucose, DP 2: maltose, DP 3: maltotriose
14 12 11
268
Optimising sweet taste in foods
freely mixed with water and dry products dissolve easily and quickly to yield clear solutions. While dry glucose monohydrate is of good shelf stability, syrups can undergo changes during prolonged storage like changing colour or glucose crystallisation. The sweetness of starch hydrolysates also depends on their composition and intensifies with increasing degradation as the level of oligosaccharides and glucose rises. The level of sweeteness (Table 10.3) comes close to the level of glucose for products with high DE values (Buck 2001). High-fructose corn syrups, in contrast, are characterised by their fructose content as they are always produced from high-glucose syrups. They often contain 72% dry matter with 42% fructose, 54% glucose and small amounts of oligosaccharides and are then as sweet as invert sugar. Syrups with fructose levels in the range of 80±95% are also available. Fructose levels between these two values can be obtained by blending (Anonymous 2002). Production Different processes are used for the production of starch hydrolysates and processes are either based on acids, enzymes, combined use of acids and enzymes or combinations of enzymes. Hydrolysis with acids uses hydrochloric acid or sulphuric acid and may be carried out in batches or continuously. Complete hydrolysis of starch with acids is not possible, as reversion limits the yield. Reversion is the re-arrangement and formation of new linkages under the catalytic influence of the acids and produces mostly 1,6-- and 1,6- -linkages. The upper limit for hydrolysis with acids is around DE 55, but the practical limit is DE 42. Starch suspensions of 35±40% are adjusted to pH 1.8±2 and heated to temperatures of up to 160ëC for 15±20 min. The reaction mixture is then neutralised and demineralised, treated with active carbon to remove coloured side-products and concentrated. For the first process step, which aims at limiting starch degradation and minimal reversion in a two-step process, hydrolysis temperatures in the range of 120±140ëC and shorter reaction periods are used (Tegge 1984). Enzymatic hydrolysis is possible with amylases of different types and with amyloglucosidase. To degrade 1,6-linkages, debranching enzymes like amyloglucosidase and preferably pullulanase are necessary. Enzymatic processes require less concentrated suspensions or solutions than acid hydrolysis. Selection of temperature-stable enzymes has resulted in process temperatures above those allowing microbial growth. Therefore such processes can be regarded as self-sterilising (Anonymous 2002; Tegge 1984). Combined processes either use acids for initial hydrolysis followed by enzymes or use two different types of enzymes. They allow determination of the reaction at any desired end-point and can proceed to glucose as the final product. Enzymes isomerising glucose to fructose can be isolated from many microorganisms like Bacillus, Streptomyces, Arthrobacter and Actinoplanes species. They need magnesium or cobalt ions for activation. Enzymes for commercial use have optimum activity at temperatures above 60ëC. They are normally used in immobilised form bound to a carrier (Buck 2001; Tegge 1984).
Reduced-calorie sweeteners and caloric alternatives
269
Production of high-fructose corn syrups starts from syrups with a high glucose content. Although the equilibrium of the reaction is at ca. 55% fructose, the reaction is normally terminated at lower fructose levels, as the rate of reaction decreases with increasing fructose concentration. Higher fructose levels can be obtained by chromatographic separation of the carbohydrates (Anonymous 2002; Buck 2001; Tegge 1984). After hydrolysis as well as isomerisation, syrups are demineralised with ion exchangers, decolorised with activated carbon and concentrated to the desired level of solids. Physiology The chain-length of hydrolysed starch products lies between that of starch and glucose. Therefore their digestibility is normally at least as good as that of starch. Products with a higher content of oligosaccharides or glucose may cause a fast increase in blood glucose levels similar to those seen after glucose consumption. Therefore these products have a high glycaemic index. High-fructose corn syrups are metabolised like invert sugar (Schwaiger 1991). Starch hydrolysates and high-fructose corn syrups can normally be metabolised to lactic acid by bacteria of the oral cavity and therefore contribute to caries formation. Applications Glucose syrups can vary widely in composition and are therefore very versatile. Their use is based on their functions not only as sweetening agents, nutrients or humectants, but also as exturising and crystallisation inhibiting agents. Important fields of application are confectionery products, in which they prevent crystallisation of glucose; ice cream, in which they prevent formation of large crystals of ice; and lactose crystallisation and bakery products, in which they prolong shelf-life (Buck 2001; Schwaiger 1991). High-fructose corn syrups are used in fine bakery wares, jams and jellies, canned fruit and sauces. Their main application is, however, in non-alcoholic beverages in which they have replaced sucrose in many countries outside the EU (Anonymous 2002; Buck 2001; Schwaiger 1991; Tegge 1984). Table 10.3 Sweetness and DE values of starch hydrolysates Product Syrup 27 Syrup 43 Syrup 52 Syrup 64 Syrup 78 Maltose Glucose Sucrose
Sweetness (on a dried basis) DE DE DE DE DE
DE: dextrose equivalent
0.27 0.38 0.43 0.58 0.63 0.43 0.61 1.0
270
Optimising sweet taste in foods
10.4.4 Glucose Occurrence, chemical and physical properties D-glucose is ubiquitous in nature. It is a common constituent of sweet fruits and honey and is also a building block of polysaccharides like starch and cellulose. Glucose is the most important substrate of energy metabolism of cells. The typical fasting level in human blood is 600±1000 mg/L. The nervous system, blood cells and some other parts of the human body are strictly dependent on glucose for energy gain. Glucose is a monosaccharide which is normally available as the -D-glucose monohydrate, a crystalline product forming white to colourless crystals. Anhydrous -D-glucose and also -D glucose are of very limited importance. The melting range of -D-glucose monohydrate is 83ëC while the anhydrous substance melts at 146ëC. The specific rotations are: -D-glucose is []20D: +112.2ë, -D-glucose []20D: +18.7ë. In aqueous solution mutarotation occurs which results in an equilibrium of []20D: +52.5ë. The solubility of glucose is lower than the solubility of sucrose. When dissolved, glucose has a cooling effect owing to its negative heat of dissolution of ÿ105.5 kJ/kg (Schwaiger 1991). Glucose is a reducing sugar and undergoes caramelisation and Maillard reactions. Production The production of glucose starts from starch degraded to syrups with a high glucose content. These syrups are concentrated to at least 75% dry matter and slowly cooled down over days. This results in a yield of around 60% of the glucose. Separation of the crystals, washing to remove the mother liquor and drying results in a product that has the required purity. Concentration of the mother liquor yields further product of lower purity (Anonymous 2002; Tegge 1984). Physiology Glucose is the most important substrate of cell metabolism. It is the carbohydrate fulfilling the task of being the energy transport system of the body. Several parts of the body like the nervous system or blood cells are completely dependent on glucose as the supplier of energy. Glucose is readily absorbed from the intestine. Any glucose not immediately needed for energy is stored as glycogen in the liver. The healthy organism tries to maintain a constant level of glucose in blood and serum by secretion of insulin to reduce too high a level and mobilises glucose from glycogen of the liver if levels fall below normal. As metabolism of glucose is insulin-dependent and type I diabetics lack insulin and type II diabetics have insufficient insulin release, they should avoid consumption of large amounts of glucose, at least without adequate medication (Lang 1979a; Szepesi 1996). Glucose is fully caloric. It can be metabolised to lactic acid by the bacteria of the oral cavity. Glucose is also an excellent substrate for almost all types of fermentation.
Reduced-calorie sweeteners and caloric alternatives
271
Applications Glucose has the disadvantage of a lower sweetness than sucrose. Its lower solubility can be an additional obstacle for applications in which readily and highly soluble sweeteners are required. Pure solid glucose monohydrate is used in tableted products intended to raise blood sugar levels quickly, for example, products consumed by athletes or diabetics in situations when their blood sugar levels are low. The large majority of glucose is used either in syrups in combination with sucrose for the manufacture of confectionery products or in high fructose corn syrups. This type of high fructose corn syrup has replaced sucrose in non-alcoholic beverages in countries having no specific taxes on their use (Anonymous 2002; Tegge 1984). 10.4.3 Maltose Occurrence, chemical and physical properties Maltose is a disaccharide normally formed as a product of starch degradation by -amylases of plants. It can be produced, therefore, in processes such as malting and is an important fermentable constituent of wort. Maltose, 4--D-glucopyranosido-D-glucopyranose, crystallises as the monohydrate and forms colourless to white crystals like the other sweet carbohydrates. Maltose hydrate melts in the range of 102±103ëC. It undergoes mutarotation in aqueous solution leading to a specific rotation of []20D: +130.4ë (Schwaiger 1991). The solubility of maltose hydrate is around 500 g/kg of solution at 20ëC. It does not have a cooling effect. Maltose is a reducing sugar and therefore participates in caramelisation and Maillard reaction. The sweetness intensity of maltose is 35% compared to a 10% (w/w) solution of sucrose. The sweetness is clean (Schwaiger 1991). Production Production of maltose is similar to glucose production. It starts from syrups rich in maltose and uses similar crystallisation conditions (Tegge 1984). Physiology Maltose and maltooligosaccharides are quickly cleaved to glucose in the intestine and absorbed. Therefore their glycaemic index is similar to glucose. Like glucose maltose and maltooligosaccharides are readily fermented by the bacteria of the oral cavity. They are also good substrates for fermentation, although the fermentation rate may be slightly delayed as adaptation of the organisms may be necessary. Applications In some countries syrups rich in maltose are used for brewing as their composition is similar to wort. They are also suitable for use in doughs with strong-fermenting yeasts as these are able to metabolise maltose quickly. As
272
Optimising sweet taste in foods
their sweetness is lower than that of syrups rich in glucose they are less suitable for applications in which a sweet syrup is advantageous (Schwaiger 1991). 10.4.4 Fructose Occurrence, chemical and physical properties Besides sucrose and glucose, fructose is one of the most common naturally occurring saccharides, especially in fruits, in which it can be the predominant monosaccharide. As fructose occurs in nectar, honey also contains substantial amounts of fructose. Fructose is a keto-hexose crystallising as the -pyranoid form in white to colourless crystals. Fructose melts in the range of 100±104ëC. The specific rotation is []20D: ÿ91 to ÿ93.5ë. It is slightly hygroscopic with an equilibrium humidity of 63% rh (Schiweck 1991a). The solubility of fructose is almost 800 g/kg of solution at 20ëC and increases to more than 950 g/kg of solution at 90ëC. The heat of solution is ÿ37.7 kJ/kg. In water, a temperature-dependent equilibrium between the - and -pyranoid, and -furanoid and linear form exists. Aqueous solutions of fructose have approximately half the viscosity of sucrose solutions of the same concentration (Schiweck 1991a; White and Osberger 2001). As a ketose, fructose is a reducing sugar and is therefore not fully stable, with optimum stability in the range of pH 3.5±5.5-hydroxymethyl-2-furfural is formed from fructose at low pH levels and higher temperatures, and difructose anhydrides are also formed from fructose in concentrated low pH solutions. Fructose readily reacts in Maillard reactions (Schiweck 1991a; White and Osberger 2001). Owing to the different fructose forms in aqueous solutions and the temperature-dependent changes, the sweetness intensity compared to sucrose is not only dependent on concentration, but also on temperature, and decreases with increasing temperature. Compared to a 100 g/l solution of sucrose 121% of the sucrose sweetness was determined, but at 190 g/l only 114%. The sweetness decreases by ca. 30% when the temperature increases from 20 to 60ëC. The sweetness of fructose is perceived quickly but also fades fast (Schiweck 1991a). Production Production of fructose can start from three different sources, high-fructose corn syrup, invert sugar and inulin, a fructose polycondensate occurring, among other places in the tubers of the Jerusalem artichoke (Helianthus tuberosus L.). Chemical or rather enzymatic hydrolysis of -2!1 linkages in inulin results in fructose. Despite improvement of enzyme performance, the common starting material for fructose production is a syrup rich in fructose obtained from invert sugar or high-fructose corn syrup by chromatographic separation of saccharides on cation exchangers in the Ca-form. Crystallisation is possible from solvents like ethanol. The common way, however, is crystallisation from a hot concentrated solution of at least 90% dry matter by cooling (Schiweck 1991a).
Reduced-calorie sweeteners and caloric alternatives
273
Physiology Fructose is absorbed from the intestine and partly metabolised already in the intestinal mucosa. The capacity for fructose absorption is, however, small compared with that for sucrose and glucose. It is readily converted in the liver with the result that levels of blood fructose are generally low. Intake of large amounts of fructose changes metabolic abilities as it by-passes the regulatory mechanism in glycolysis and can, if it replaces glucose for some time, temporarily result in poor glucose absorption. It can also shift fatty acid metabolism from oxidation to esterification of free fatty acids resulting in an increase in lowdensity blood lipids (Lang 1979b; Mayes 1993; Riby et al. 1993; Schiweck 1991a). Fructose is metabolised independently of insulin and is therefore suitable for foods specifically manufactured for diabetics. Its glycaemic index is ca. 20% of the glucose value (Crapo et al. 1993). A limited number of people suffer from hereditary fructose intolerance caused by inability to metabolise fructose (Odievre et al. 1978). They have to avoid fructose intake. Persons suffering from benign fructose intolerance are able to excrete fructose in urine. Applications Crystalline fructose is more expensive than sucrose, high-fructose corn syrup or glucose syrups. It has, however, a number of specific uses (Hanover and White 2002; Schiweck 1991a; White and Osberger 2001). Besides its sweetness contribution, fructose improves browning in bakery products owing to its reactivity in Maillard reactions and caramelisation properties and prolongs shelflife. It is used in low-solids fruit preparations for dairy products and jams and jellies in which it combines well with intense sweeteners. It does not readily recrystallise and is therefore suitable for confections with soft or liquid fillings and also reduces the water activity in these products. It is, however, less suitable for candy production. In some countries like Germany, chocolate and fine bakery wares for diabetics use fructose instead of sucrose. 10.4.5 Lactose Occurrence, chemical and physical properties Lactose is a constituent of cow's milk, found typically at concentrations of 4.5± 5.5%. It also occurs in milk of other animal species at concentrations of between 1.5 and 7.3% (w/w). Lactose, 4- -D-galactopyranosido-D-glucopyranose, is available as the lactose monohydrate and the anhydrous -lactose. -Lactose monohydrate releases water above ca. 120ëC to form the anhydrous substance. -Lactose melts at 202ëC, -lactose at 253ëC, and both melt under decomposition. Lactose shows mutarotation. The specific rotation is []20D: +52.7ë for -lactose monohydrate and []20D: +55.5ë for -lactose. Lactose dried in a special process has a low tendency to absorb water. Spray-dried lactose, however, is hygroscopic and can absorb water even at a relative humidity of 30% and readily
274
Optimising sweet taste in foods
absorbs water above 70% rh. The solubility of lactose is around 160 g/kg of solution at 20ëC and increases to ca. 600 g/kg of solution at 90ëC. Although it is a reducing sugar, lactose is rather stable during storage and food processing. The sweetness intensity of lactose is 35±40% compared to a 10% (w/w) solution of sucrose (Boesig 1991; Morrissey 1985). Production The source for commercial production of lactose is not milk itself but sweet whey which has either to be processed immediately or preserved to prevent growth of lactic acid bacteria and concomitant loss of lactose. As the first step, whey is adjusted to the isoelectric point of whey proteins, i.e. pH 4.8±5. Heating coagulates the proteins which are removed by filtration or centrifugation. During subsequent concentration, salts contained in the milk precipitate and are also removed. Lactose is crystallised by concentration to 60±65% dry matter and cooling. Normal yield is ca. 4% on the basis of the whey used. The remaining syrup is re-circulated back into the process. The crystallised lactose is not of sufficient purity for many applications and is therefore often purified by recrystallisation. Crystallisation is often carried out above 94ëC in order to obtain the more soluble and sweeter -lactose. Below this temperature -lactose monohydrate is obtained (Boesig 1991). Physiology Lactose has a special position in nutrition as it is more or less the only carbohydrate occurring in the diet of weanlings for months. In the intestine lactose is cleaved by lactase ( -glactosidase). The enzyme is located in the cells of the intestinal mucosa. Cleavage occurs therefore when lactose passes through the mucosa. In adults the activity of this enzyme is normally so low that cleavage of lactose is rather incomplete. After intake of large quantities, part of the ingested lactose may be absorbed without cleavage. It is then excreted with the urine as body cells are unable to metabolise lactose (Boesig 1991; Lang 1979d; Paige and Davis 1985). As absorption of lactose is slow, large doses can have a laxative effect, which occurs in healthy persons normally above doses of ca. 50 g only (Kleessen et al. 1997). Deficiency of lactase results in lactose intolerance which causes intestinal discomfort with symptoms like diarrhoea and flatulence. Inherited lactose intolerance is rather rare. Primary lactose intolerance caused by a loss of lactase in adults is fairly common among non-whites but a certain part of the population in Europe suffers from this deficiency, too. Secondary lactose intolerance is caused by damage to the intestinal mucosa. Approximately 5±10% of the population are supposed to suffer from this problem (Vesa et al. 2000). Lactose is of fairly low cariogenicity. The saliva lacks lactase, and only plaque bacteria can cleave lactose. The resulting decrease in pH, however, is normally limited. Some micro-organisms, but not all by far, are able to metabolise lactose. These include Lactobacilli and Streptococci in particular and also some other types of bacteria, some yeasts and a few moulds.
Reduced-calorie sweeteners and caloric alternatives
275
Applications Lactose is used as a bulking ingredient and filler for other special technological purposes, rather than as a sweetener, as its sweetness and solubility are too low for most hypothetical fields of application. It finds applications in foods in which dry solids are to be maintained at as high a level as in conventional products, but reduced sweetness is desired (Boesig 1991; Morrissey 1985). Lactose is important for the production of baby food, In baby food it does not only provide energy but also supports development of a healthy intestinal flora and aids absorption of calcium and trace minerals. In meat products undergoing lactic acid fermentation lactose is used as a nutrient for bacteria, especially starter cultures. In powdered products lactose may serve as the carrier. Lactose is widely used as a carrier for pharmaceutical tablets. Different granular sizes are available for granulation or direct tabletting. Although fermentability of lactose is limited it serves as a nutrient for speciality fermentations.
10.5
Future trends
Recent authorisations of tagatose and isomaltulose will result in increasing use of these sweetening agents, especially as they are produced by well-established producers of sweet carbohydrates and marketed by companies experienced in this area. Tagatose will benefit not only from its flavour modification properties and from the fact it is a reduced-calorie sweetening agent, but also from its sweetness intensity, which is similar to sucrose. It has, however, the disadvantage of being a laxative, similar to sugar alcohols. It is used as a flavour modifier in beverages at low levels. As a sweetener it will have to compete with sugar alcohols. Isomaltulose and also trehalose have a lower glycaemic index than sugar and starch hydrolysates. With increasing discussion of the glycaemic index or glycaemic load of carbohydrates and spreading awareness of the need to avoid high variations in the insulin and blood glucose levels, it seems probable that these sweetening agents, especially isomaltulose, will find markets in speciality products like those aimed at diabetics. This applies especially to isomaltulose, which has been produced in high quantities for at least 20 years as an intermediate of isomalt production. Lactose hydrolysis in dairy products has aroused more interest in recent years and may enjoy more widespread use replacing sucrose at least in part in sweetened products, where the replacement should be cost-competitive. It seems not very likely that any of the other reduced-calorie or new caloric alternatives will become commercially important in the coming years. Starch hydrolysates have replaced sucrose in several important applications in countries in which they can compete directly with sucrose. Growth in other countries and regions can be expected when the taxes imposed on these products to protect sugar are abandoned.
276
10.6
Optimising sweet taste in foods
Sources of further information and advice
Ample information on the established caloric sweeteners is available. The quantity of published information about reduced-calorie and new caloric sweeteners is not as large, as these sweeteners were either only developed or have come to be of more interest in recent years. Information on applications is often either published in patent applications or obtained from suppliers. The following websites can be consulted: Patents http://ep.espacenet.com/search97cgi/s97_cgi.exe?Action=FormGen&Template= ep/en/advanced.hts This site allows search for patents or published patent applications using different criteria including keywords, names of inventors or applicants. Tagatose http://www.gaio-tagatose.com/ http://www.naturlose.com/ Trehalose http://www.cargillhft.com/industry_products_trehalose.html Isomaltulose http://www.palatinit.com/en/Food_Ingredients/Palatinose/ http://www.cargillsweetness.com/1232.html
10.7
References
and KAREL M (1997) Preservation of biological materials under desiccation, Crit. Rev. Fd. Sci. Nutr. 37 (3), 287±309. ANONYMOUS (2000) US Food and Drug Administration. Response letter to GRAS notice 000045 dated 05 October 2000. ANONYMOUS (2001a) US Food and Drug Administration. Response letter to GRAS notice 000078 dated 25 October 2001. ANONYMOUS (2001b) Joint FAO/WHO Expert Committee on Food Additives Fifty-fifth meeting. WHO Technical Report Series No. 901 pp. 18±19. ANONYMOUS (2001c) Commission Decision authorising the placing on the market of trehalose as a novel food ingredient. Off. J. L 269, 17±19, 10 October. ANONYMOUS (2002) Nutritive Sweeteners from Corn. Corn Refiners Association, Washington, 7th Edition. ANONYMOUS (2004) Joint FAO/WHO Expert Committee on Food Additives. Sixty-first meeting, WHO Technical Report Series No. 922 pp. 23±26. ANONYMOUS (2005) Commission Decision authorising the placing on the market of isomaltulose as a novel food or novel food ingredient. Off. J. L 160, 28±30, 23 June and L 199, 90±91, 29 July. AGUILERA J M
Reduced-calorie sweeteners and caloric alternatives
277
and LEEMAN W R (1999) L-Sorbose but not D-tagatose induces hemolysis of dog erythrocytes in vitro, Regul. Toxicol. Pharmacol. 29, S43±S45. BERNARD J and KOWALCZYK J (2005) Gelatine-free soft caramel containing isomaltulose. WO2005.044.016, 19 May (to Suedzucker AG, Mannheim-Ochsenfurt). BERTELSEN H, HANSEN S J, LAURSEN R S, SAUNDERS J and ERIKNAUER K (2001a) Tagatose, in Alternative Sweeteners (Ed. L O'Brien Nabors) 3rd Edn, pp. 105±127 (Marcel Dekker, New York). BERTELSEN H, ANDERSEN H and TVEDE M (2001b) Fermentation of D-tagatose by human intestinal bacteria and dairy lactic acid bacteria, Microbial Ecology in Health and Disease 13, 87±95. BOESIG W (1991) Laktose und Laktosehydrolysate, in Handbuch Suessungsmittel (Eds. GW von Rymon Lipinski and H Schiweck) pp. 243±263 (Behr's, Hamburg). BOULINEAU F P and WEI A (2002) Synthesis of L-sugars from 4-deoxypentenosides, Org. Lett. 4, 2281±2283. BUCK A W (2001) High fructose corn syrup, in Alternative Sweeteners (Ed. L O'Brien Nabors) 3rd Edn, pp. 391±411 (Marcel Dekker, New York). BUEMANN B, TOUBRO S and ASTRUP A (1999a) Human gastrointestinal tolerance to Dtagatose, Regul. Toxicol. Pharmacol. 29, S71±S77. BUEMANN B, TOUBRO S, RABEN A and ASTRUP A (1999b) Human tolerance to single, high dose of D-tagatose, Regul. Toxicol. Pharmacol. 29, S66±S70. BUEMANN B, TOUBRO S, HOLST J J, REHFELD J F, BIBBY B M and ASTRUP A (2000) D-tagatose, a stereoisomer of D-fructose, increases blood uric acid concentration, Metabolism 49, 969±976. COOPER J M, ACASTER M A, GLEESON M, HEATH C and BOTHAM R L (2001) Nutritional compositions comprising trehalose for persons suffering from diabetes, GB 2,353,934, 14 March (to British Sugar plc, Peterborough). CRAPO P A, KOLTERMAN O G and OLEFSKY J M (1993) Effects of oral fructose in normal, diabetic, and impaired glucose tolerance subjects, Am J. Clin. Nutr. 58, 724S± 732S. DOERR T, GUDERJAHN L, KEME T, SENTKO A and KOWALCZYK J (2005) Isomaltulosecontaining instant beverage powder, EP 1,524,911, 27 April (to Suedzucker AG, Mannheim-Ochsenfurt). DONNER T W, WILBER J F and OSTROWSKI D (1999) D-Tagatose, a Novel Hexose: Acute Effects on Carbohydrate Tolerance in Subjects With and Without Type 2 Diabetes, Diabetes, Obesity and Metabolism 1, 285±291. ELBEIN A D (1974) The metabolism of ,-Trehalose, Adv. Carbohydr. Chem. Biochem. 30, 227±256. ERMOLENKO L, SASAKI N A and POTIER P (2003) Novel route to L-hexoses from L-ascorbic acid: Asymmetric synthesis of L-galactopyranose and L-talopyranose, Helv. Chim. Acta, 3578±3582. FLYNN A (1985) Nutritional significance of lactose: II. Metabolism and toxicity of galactose in Developments in Dairy Chemistry ± 3 (Ed. P O Fox) pp. 133±141. GREENBERG N A and MAHONEY R R (1981) Immobilisation of lactase (betagalactosidase) for use in dairy processing: a review, Process Biochem. 16(2), 2±8. HANOVER L M and WHITE S J (2002) Manufacturing, composition, and applications of fructose, Food Technol. Biotechnol. 40, 67±73. HAVENAAR R, DROST S J, DE STOPPELAAR J D, HUIS IN'T VELD J H J and BACKER DIRKS O (1984) Potential cariogenicity of Lycasin 80/55 in comparison to starch, sucrose, xylitol, sorbitol and L-sorbose in rats, Caries Research 18, 375±384. BAER A
278
Optimising sweet taste in foods
and PARRISH F W (1980) A new method for the preparation of lactulose from lactose, Carbohydrate Res. 82, 393±397. HICKS K B, RAUPP D L and SMITH P W (1984) Preparation and purification of lactulose from sweet cheese whey ultrafiltrate, J. of Agric. Food Chem. 32, 288±292. HOLSINGER V H and KLIGERMAN A E (1991) Applications of lactase in dairy foods and other foods containing lactose, Food Technol. 45(1), 93±95. HORWATH R O and COLONNA W J (1984). Process for preparing L-fructose from L-glucose by contacting L-glucose with xylose isomerase produced by a microorganism of the genus Candida, US 4,463,093, 31 July (to Nabisco Brands, Inc., Parsippany). Â Brien IRWIN W E and STRAETER P J (2001) Isomaltulose, in Alternative Sweeteners (Ed. L O Nabors) 3rd Edn, pp. 413±421 (Marcel Dekker, New York). KAWAI K, OKUDA Y and YAMASHITA K (1985) Changes in blood glucose and insulin after an oral palatinose administration in normal subjects, Endocrinol. Jpn. 32, 933±936. KAWAI K, YOSHIKAWA H, MURAYAMA Y, OKUDA Y and YAMASHITA K (1989) Usefulness of palatinose as a caloric sweetener for diabetic patients, Horm. Metab. Res. 21, 338± 340. KELLER P and KISTLER A (1977) The haemolytic effect of sorbose in dogs, Experientia 33, 1380±1382. KIM K S, CHO B H and SHIN I (2002) A new efficient method for the synthesis of Lgalactose, Bull. Korean Chem. Soc. 23, 1193±1194. KLEESSEN B, SYKURA B, ZUNFT H J and BLAUT M (1997) Effects of inulin and lactose on fecal microflora, microbial activity, and bowel habit in elderly constipated persons, Am. J. Clin. Nutr. 65, 1397±1402. KOWALCZYK J, KOZIANOWSKI G, KUNZ M and MOSER M (2004) Process for manufacturing a palatinose-containing food, EP 1,410,721, 21 April (to Suedzucker AG, Mannheim-Ochsenfurt). KOZEMPEL M F, KURANTZ M J, CRAIG J C JR and HICKS K B (1995) Development of a continuous lactulose process: Separation and purification, Biotechnol. Prog. 11, 592±595. LANG K (1979a) Biochemie der Ernaehrung, pp. 15±17 (Steinkopff, Darmstadt). LANG K (1979b) Biochemie der Ernaehrung, pp. 18±19 (Steinkopff, Darmstadt). LANG K (1979c) Biochemie der Ernaehrung, pp. 21±24 (Steinkopff, Darmstadt). LANG K (1979d) Biochemie der Ernaehrung, pp. 32±34 (Steinkopff, Darmstadt). LEVIN G V (1981) US 4,262,032, 14 April 1981 (to Biospherics Inc., Rockville). LEVIN G V (2002) Tagatose, the new GRAS sweetener and health product, J. Medical Food 5, 1±19. LINA B A R, JONKER D and KOZIANOWSKI G (2002) Isomaltulose (Palatinose): A review of biological and toxicological studies, Food Chem. Toxicol. 40, 1375±1381. LIPPMANN E O VON (1929) Geschichte des Zuckers (Springer, Berlin and Heidelberg). LIVESEY G and BROWN J C (1995) Whole body metabolism is not restricted to D-sugars because energy metabolism of L-sugars fits a computational model in rats, J. Nutr. 125, 3020±3029. MADENOKOJI N, IINO H, SHIMIZU T, HAYAKAWA J and SAKASHITA M (2003) Blunting effect of D-tagatose on blood glucose when administered orally with glucose in volunteer donors of boundary glycemic level, J. Jap. Soc. Clin. Nutr. 25, 21±28. MAHONEY R R (1985) Modification of lactose and lactose-containing dairy products with -galactosidase in Developments in Dairy Chemistry ± 3 (Ed. P O Fox) pp. 69± 109. MAYES P A (1993) Intermediary metabolism of fructose, Am. J. Clin. Nutr. 58, 754S±765S. HICKS K B
Reduced-calorie sweeteners and caloric alternatives
279
(1985) Lactose: Chemical and physicochemical properties, in Developments in Dairy Chemistry ± 3 (Ed. P O Fox) pp. 1±34. MUNIR M (1987) Production of isomaltulose using immobilised microorganisms, US 4,640,894, 3 February (to Suedzucker AG, Mannheim-Ochsenfurt). NISHIMOTO T, CHAEN H, SUGIMOTO T and MIYAKE T (2002) Preparation of high trehalose content product, EP 1,085,096, 2 January (to Hayashibara Co., Okayama). NORMEN L, LAERKE H N, JENSEN B B, LANGKILDE A M and ANDERSSON H (2001) Small-bowel absorption of D-tagatose and related effects on carbohydrate digestibility: an ileostomy study, Am. J. Clin. Nutr. 73, 105±110. O'BRIEN J (1996) Stability of trehalose, sucrose and glucose to nonenzymatic browning in model systems, J. Food Sci. 61, 679±682. ODIEVRE M, GENTIL C, GAUTIER M and ALAGILLE D (1978) Hereditary fructose intolerance in childhood. Diagnosis, management, and course in 55 patients, Am. J. Dis. Child 132, 605±608. OOSHIMA T, IZUMITANI A, TAKEI T, FUJIWARA T and SOBUE S (1990) Plaque formation of dietary isomaltulose in humans, Caries Res. 24, 48±51. PAIGE D M and DAVIS L R (1985) Nutritional significance of lactose: I. Nutritional aspects of lactose digestion, in Developments in Dairy Chemistry ± 3 (Ed. P O Fox) pp. 111±132. PARRISH F W, TALLEY F B, ROSS K D, CLARK J and PHILLIPS J G (1979) Sweetness of lactulose relative to sucrose, J. Food Sci. 44, 813±815. PELTROCHE-LLACSAHUANGA H, HAUK C J, KOCK R, LAMPERT F, LUTTICKEN R and HAAGSE G (2001) Assessment of acid production by various human oral micro-organisms when palatinose or leucrose is utilized, J Dental Res. 80, 378±384. PORTMANN M O and BIRCH G G (1995) Sweet taste and solution properties of ,-trehalose, J. Sci. Food Agric. 69, 275±281. REICHSTEIN T and GRUESSNER A (1934) Eine ergiebige Synthese der L-AscorbinsaÈure (CVitamin), Helv. Chim. Acta. 17, 311±315. RIBY J E, FUJISAWA T and KRETCHMER N (1993) Fructose absorption, Am. J. Clin. Nutr. 58, 754S±765S. RICHARDS A B and DEXTER L B (2001) Trehalose, in Alternative Sweeteners (Ed. L O'Brien Nabors) 3rd Edn, pp. 105±127 (Marcel Dekker, New York). MORRISSEY P A
RICHARDS A B, KRAKOWKA S, DEXTER L B, SCHMID H, WOLTERBEEK A P, WAALKENS-
and KURIMOTO M (2002) Trehalose: A review of properties, history of use and human tolerance, and results of multiple safety studies. Food Chem. Toxicol. 40, 871±898. ROSER B (1991) Trehalose, a new approach to premium dried foods. Trends Food Sci. Technol. 7, 166±169. SASAKI N, TOPITSOGLOU V, TAKAZOE I and FROSTELL G (1985) Cariogenicity of isomaltulose (palatinose), sucrose and mixture of these sugars in rats infected with streptococcus mutans E-49, Swed. Dent. J. 9(4), 149±155. SATO S, OKAMOTO K, MINAMI R, KOHRI H and YAMAMOTO S (1999) Trehalose can be used as a parenteral saccharide in rabbits, J. Nutrition 129, 158±164. SCHIWECK H (1991a) D-Fruktose, in Handbuch Suessungsmittel (Eds. G-W von Rymon Lipinski and H Schiweck) pp. 183±211 (Behr's, Hamburg). SCHIWECK H (1991b) Sonstige Zuckerstoffe und Polydextrose, in Handbuch Suessungsmittel (Eds. G-W von Rymon Lipinski and H Schiweck) pp. 243±263 (Behr's, Hamburg). SCHIWECK H (1991c) Isomalt (Palatinit), in Handbuch Suessungsmittel (Eds. G-W von BERENDSEN D H, SHIGOYUKI A
280
Optimising sweet taste in foods
Rymon Lipinski and H Schiweck) pp. 325±352 (Behr's, Hamburg). (2002) Medical, nutritional and technological properties of lactulose. An update, Eur J Nutr. 41 Suppl 1, 17±25. SCHWAIGER M (1991) Staerkeverzuckerungsprodukte, in Handbuch Suessungsmittel (Eds. G-W von Rymon Lipinski and H Schiweck) pp. 147±181 (Behr's, Hamburg). SCHWENGERS D and BENECKE H (1987) Processes for producing leucrose, US 4,693,974, 15 September (to Pfeifer & Langen KG, Cologne). SHALLENBERGER R S, ACREE T E and LEE C Y (1969) Sweet taste of D- and L-sugars and amino acids and the steric nature of their chemoreceptor site, Nature 221(5180), 555±556. SUGITANI T, TSUYUKI K, MIYATA Y, EBASHI T, OKUI H, NAKAJIMA Y and SAWADA K (1994) Process for preparing trehalulose and isomaltulose, US 5,336,617, 9 August (to Shin Mitsui Sugar Co., Tokyo and Suedzucker AG, Mannheim-Ochsenfurt). SZEPESI B (1996) Carbohydrates, in Present Knowledge in Nutrition (Eds. E E Ziegler and L J Filer Jr) pp. 33±43 (ILSI Press, Washington). TARKA S M JR, STANLEY M, SHIVELY C A, APGAR J L and KOCH K L (1993) L-sugar laxatives. US 5,219,573, 15 June (to Hershey Foods Corp., Hershey). TEGGE G (1984) Staerke, pp. 219±295 (Behr's, Hamburg). USHIJIMA K, FUJISAWA T and KRETCHMER N (1995) Study on digestion and absorption of trehalose in human intestine, Dig. Absorp. 18, 56±57. VESA T H, MARTEAU, P and KORPELA R (2000) Lactose intolerance, J. Am. Coll. Nutr. 19 (2 Suppl.), 165S±175S. VIDAL-SIVILLA S (1958) Die Wirkung hypertonischer Lo È sungen auf die intestinale Resorption verschiedener Zuckerarten, Pfluegers Archiv European Journal of Physiology 265, 389±404. WHITE J S and OSBERGER T F (2001) Crystalline fructose, in Alternative Sweeteners (Ed. L O'Brien Nabors) 3rd Edn, pp. 367±390 (Marcel Dekker, New York). WU L and BIRCH R G (2004) Characterization of Pantoea dispersa UQ68J: producer of a highly efficient sucrose isomerase for isomaltulose biosynthesis, J. Appl. Microbiol. 97, 93±103. ZADOW J G (1986) Lactose hydrolyzed dairy products, Food Technol. Aust. 38, 460±462, 471. ZIESENITZ S C, SIEBERT G, SCHWENGERS D and LEMMES R (1989a) Nutritional assessment in humans and rats of leucrose [D-glucopyranosyl-alpha(1!5)-D-fructopyranose] as a sugar substitute, J. Nutr. 119, 971±978. ZIESENITZ S C, SIEBERT G and IMFELD T (1989b) Cariological assessment of leucrose [Dglucopyranosyl-alpha(1!5)-D-fructopyranose] as a sugar substitute, Caries Res. 23, 351±357. SCHUMANN C
Part III Improving sweet tasting compounds and optimising their use in foods
11 Analysing and predicting properties of sweet-tasting compounds D. E. Walters, Rosalind Franklin University of Medicine and Science, USA
11.1
Introduction
A century ago, it was common practice for chemists to report the tastes of newly synthesized compounds. Cohn compiled the reported tastes of thousands of organic compounds, many of them sweet (Cohn, 1914). In 1919 Oertly and Myers wrote `The relation between the constitution of substances and their taste has always been of interest to the chemist and the physiologist' (Oertly and Myers, 1919). They attempted to systematize this relationship by identifying glucophores (functional groups which often taste sweet) and auxoglucs (functional groups which, when combined with glucophores, produce sweettasting compounds). The Shallenberger theory (Shallenberger and Acree, 1967) proposed that all sweet-tasting substances have a hydrogen bond donor and Ê . These models have proven to be acceptor, separated by a distance of about 3 A too simplistic (Crosby et al., 1979), but they have been useful starting points in understanding structure±taste relationships. In this chapter we consider more recent models used to rationalize and predict the sweetness of organic compounds. Several kinds of models have proven useful in rationalizing and predicting structure±taste relationships. These include quantitative structure±activity relationship (QSAR) models, in which sweetness potency is statistically correlated with structural properties; pharmacophore models, in which threedimensional patterns of functional groups in sweetener molecules are identified; and receptor models, in which the nature of the receptor active site is represented in three dimensions.
284
11.2
Optimising sweet taste in foods
QSAR models
In the QSAR approach, parameters based on chemical structure are tested for their ability to statistically correlate with a biological activity (in this case, with sweetness potency). The goals of a QSAR study are twofold: (1) to rationalize existing structure-activity data and identify parameters which contribute to potency, and (2) to permit the prediction of potency for new analogs, so that the most promising compounds can be synthesized and tested. Corwin Hansch, who pioneered the field of QSAR, published the first QSAR study of sweeteners (Deutsch and Hansch, 1966). In a series of substituted mnitroanilines, the sweetness potency was shown to correlate with the hydrophobicity and with the Hammett constant of the substituent. Subsequently the nitroanilines have been subjected to QSAR analysis using other substituent parameters (McFarland, 1971; Kier and Hall, 1976; Rao and Kumar, 1986). Among these, the best correlation (r 0:99) was obtained by Rao and Kumar, using van der Waals volume and resonance terms. Iwamura (1980) used Verloop's STERIMOL parameters (Verloop et al., 1976) to estimate the optimal size and shape of the binding site for a series of 38 oximes. Once again, hydrophobicity was positively correlated with potency. Three STERIMOL parameters (corresponding to the length of the substituent and the width in two directions) contributed to the final regression equation, suggesting an optimal size and shape for the hydrophobic binding site. Spillane and coworkers have reported several QSAR studies of cyclamate and related sulphamates (Spillane and McGlinchey, 1981; Spillane et al., 1983, 1996, 2000; Spillane, 1983; Drew et al., 1998). Their most recent study included 101 compounds (20 sweet and 81 non-sweet). Using this data set, a discriminant analysis method correctly classified about 80% of the compounds; a Tree-based analysis classified 86% correctly, but it did poorly with classification of sweet compounds. The discovery of aspartame (Mazur et al., 1969) led to extensive structureactivity studies; over 1000 analogs have been synthesized and tested. Van der Heijden performed a QSAR study of 33 sweet aspartame analogs (van der Heijden et al., 1979), employing STERIMOL. Their final regression equation utilizes a hydrophobic term and two STERIMOL parameters, and it produces a correlation coefficient of 0.81 between calculated and predicted potencies. As was the case for the nitroanilines, increased hydrophobicity led to increased potency. A more extensive QSAR study of aspartame analogs was reported by Iwamura (1981). Four classes of compounds were studied: aspartic amides, aspartylaminoethyl esters, aspartylaminopropionates, and aspartylaminoacetates. For each class, regression equations showed the importance of an electron-withdrawing effect on the amide bond, as well as the size and shape of the substituent as measured by STERIMOL parameters. A set of 22 aspartic peptides, arylureas, and guanidine-acetic acid derivatives was used to develop methodology which incorporates elements of QSAR, pharmacophore modeling and receptor modeling (Walters and Hinds, 1994). In
Analysing and predicting properties of sweet-tasting compounds 285 this approach, the structures are graphically superimposed in low-energy conformations in a way which presents a common pharmacophore pattern to the receptor. A series of model receptor sites are then generated around this structure set, using a genetic algorithm to produce models which give a high statistical correlation between calculated binding energy and experimentally measured sweetness potency. For the set of 22 diverse structures, models were derived with a mean correlation coefficient of 0.94. When a subset of 11 structures was used to generate models, the resulting models could predict potencies of the 11 omitted structures with a mean error of 0.4 log units. The results described in this section show that QSAR methods can, in fact, rationalize existing structure-activity data, identify parameters which contribute to potency, and predict sweetness potency for new analogs.
11.3
Pharmacophore models
A pharmacophore is a pattern of functional groups which can interact favourably with a receptor. The most thoroughly studied sweetener pharmacophore has been the AH-B pattern proposed by Shallenberger and Acree (1967). As has been pointed out by Crosby and others (Crosby et al., 1979; Rohse and Belitz, 1991), the AH-B pattern has serious limitations: numerous compounds which have the appropriate pattern of hydrogen bond donor (AH) and acceptor (B) do not taste sweet, and not all of the sweet compounds cited in the original paper have a hydrogen bond donor which is the proper distance from the hydrogen bond acceptor. Further, it is now known (Xu et al., 2004) that cyclamate binds to a different part of the receptor than do other sweeteners. Nevertheless, this model has stimulated extensive efforts to improve pharmacophore model representations of sweeteners. The first modification of the AH-B model was the addition of a steric barrier, to account for the observation that many D-aminoacids are sweet, while the corresponding L-isomers are not (Shallenberger et al., 1969). A few years later Kier proposed a third functional group, a dispersion or hydrophobic binding site Ê from the AH group and ~5.5 A Ê from the B group (Kier, 1972). located ~3.5 A Subsequent work extended this AH-B-X model to sugars (Shallenberger and Lindley, 1977) and dipeptides (van der Heijden et al., 1978). Van der Heijden later extended the three-site model by identifying optimal size and location for the hydrophobic group in nitroanilines, sulphamates, oximes, isocoumarins, saccharins, acesulphames, chlorosugars, tryptophans and ureas (van der Heijden et al., 1985a, 1985b). In these studies, it was assumed that each class of compounds might bind to a different receptor site, so placement of the hydrophobic site was calculated independently for each set. The results suggested that there may be three or four different receptor sites for these classes of compounds. Belitz and coworkers have described the AH and B sites as electrophile (e) and nucleophile (n) sites and have sorted sweeteners into two groups: one having
286
Optimising sweet taste in foods
Ê , and one having e/n separation of about 8 A Ê (Belitz e/n separation of about 2.5 A et al., 1979; Rohse and Belitz, 1991). For each group, they have superimposed a large number of sweet and non-sweet analogs in order to map the size and shape of the hydrophobic moiety. They therefore propose that there must be at least two receptor binding sites. Nofre and Tinti, in contrast, assumed that all sweeteners must bind to the same site. Their pharmacophore model (Tinti and Nofre, 1991b) incorporates the Shallenberger AH and B sites plus six others arranged at specific points in threedimensional space. While it is now clear that at least two different binding sites exist (Xu et al., 2004), this model facilitated the discovery of two classes of extremely potent sweeteners, the N-carbamoyl dipeptides and the guanidineacetic acids (Tinti and Nofre, 1991a). Aspartame and its analogs have inspired the development of several pharmacophore models. The primary challenge with these compounds is the selection of the proper conformation. Dipeptides are quite flexible and may have hundreds of energetically accessible conformations. Temussi and coworkers used NMR spectroscopy and conformational energy calculations to analyze the conformations of aspartame (Lelj et al., 1976; Temussi et al., 1978). They proposed that the receptor-active conformation is one in which the aspartate side chain has a torsion angle of ÿ60ë, and the phenylalanine has a torsion angle of 180ë. Later they showed that this conformation can be superimposed on a conformationally rigid naphthimidazolesulphonic acid which has high sweetness potency (Temussi et al., 1991), supporting their choice of aspartame conformation. Van der Heijden reinterpreted Temussi's NMR results and argued in favor of a conformation in which the aspartate side chain has a torsion angle of ÿ60ë, and the phenylalanine has a torsion angle of ÿ60ë (van der Heijden et al., 1978). Goodman used NMR spectroscopy of aspartame and of some retroinverso analogs to support a different conformation as the active one, one in which the aspartate side chain has a torsion angle of ÿ60ë, and the phenylalanine has a torsion angle of 60ë (Goodman et al., 1987, 1993; Douglas and Goodman, 1991). Walters and coworkers superimposed aspartame conformations onto the conformationally restricted high potency guanidine-acetic acid derivatives, leading to the proposal that the active conformation of aspartame is one in which the aspartate side chain has a torsion angle of 60ë, and the phenylalanine has a torsion angle of 180ë (Culberson and Walters, 1991; Walters et al., 2000). This was the conformation used by Walters and Hinds in the previously described QSAR work (Walters and Hinds, 1994). Venanzi and Venanzi took a completely different approach. Rather than defining the pharmacophore in terms of atoms and functional groups, they carried out ab initio calculations on a number of sweeteners and mapped patterns of molecular electrostatic potential around the molecules (Venanzi and Venanzi, 1988a, 1988b, 1991). Their studies focused on acesulphame, oximes, and nitroanilines, which have little or no conformational freedom. They found that the oximes and the nitroanilines show similar patterns of electrostatic potential, even though they have little chemical similarity.
Analysing and predicting properties of sweet-tasting compounds 287
11.4
Receptor models
In the previous section, we discussed three-dimensional models of the sweetener molecules in their supposed active conformations. An alternative approach is to model the receptor with which these sweeteners interact. There are two general approaches which can be taken in modeling the receptor binding site. First, in the absence of any information about the actual receptor, sweeteners can be superimposed in their active conformations, and a model binding site can be constructed around these templates. This model binding site can be represented as a surface, or as atoms or functional groups. If the amino acid sequence of the receptor is known, and it has sufficient homology to a protein for which an Xray crystal structure is known, a more detailed model may be constructed by homology modeling. In 1991 Culberson and Walters described a sweet receptor model based on their studies of dipeptides, ureas, and guanidine-acetic acid derivatives (Culberson and Walters, 1991). Conformational energy calculations were carried out for several of these compounds having high potency, and conformations were selected which permit superposition of several functional groups common to these molecules (carboxylate, NH groups and hydrophobic groups). A van der Waals surface was constructed around the superimposed sweeteners, and an electrostatic potential was mapped onto this surface. This model was used to identify promising targets for synthesis. It also was used in the design of a new series of arylureas, and it correctly predicted the active stereoisomer of compounds in this series (Madigan et al., 1989; Muller et al., 1991). Morini and coworkers used the Tinti and Nofre pharmacophore model (Tinti and Nofre, 1991b) as a guide in constructing a pseudoreceptor model (Bassoli et al., 2002). In the pseudoreceptor approach, active compounds are first superimposed in low energy conformations according to a pharmacophore pattern. Then amino acids are placed around the active compounds in such a way as to produce favorable binding interactions, simulating a proteinaceous receptor site. The pseudoreceptor model was based on a training set of 24 compounds. A correlation coefficient of 0.985 was found between calculated binding energies and experimentally measured sweetness potencies. Binding of an independent set of five test ligands was within 0.3±2.1 kcal molÿ1 of experimental values; for these five compounds, potencies were predicted with a mean error of 0.7 log units. In 2001 the sweet taste receptor was finally identified (Max et al., 2001; Nelson et al., 2001). Two homologous proteins, T1R2 and T1R3, combine to form a heterodimer which recognizes all sweeteners which have been tested. These proteins are members of the G protein coupled receptor family. Each has a large extracellular domain and seven hydrophobic transmembrane helices. The extracellular domain of T1R2 has been shown to mediate response to aspartame and neotame. The sweet taste receptor has significant homology with metabotropic glutamate receptors of the central nervous system; one of these has been shown
288
Optimising sweet taste in foods
to function as a homodimer, and its extracellular domain has been shown by Xray crystallography to bind glutamate (Kunishima et al., 2000). Thus, it is now possible to use homology modeling to construct models of the extracellular domains of the sweet taste receptor proteins. Max and Walters have each modeled the T1R3 extracellular domain (Max et al., 2001; Walters, 2002). More recently Temussi and coworkers used Swiss-Model (Peitsch, 1996) to build a homology-based model of the T1R2/T1R3 heterodimer (Spadaccini et al., 2003). They used this model to propose a binding mode for the sweet protein monellin.
11.5
Future trends
Now that the sweet taste receptor has been characterized, future research will be concentrated on receptor structure. There is room for improvement of the homology models. An important application of these models will be in the use of docking software to identify potential binding sites and ligands. The models may also be useful in designing single point mutation experiments which will identify the atomic interactions required for receptor binding. It is only a matter of time before the extracellular binding domains of the T1R2 and T1R3 are expressed, purified, and their structures determined by Xray crystallography. This should lead to an even more accurate picture of the receptor binding sites. It will be especially interesting if the receptors can be cocrystallized with different sweeteners; this will tell us how many different binding sites are utilized by the amazingly diverse array of known sweeteners.
11.6
Sources of further information and advice
An excellent review of the early history of structure±taste studies has been published (Crosby et al., 1979). The proceedings of an American Chemical Society symposium on the chemistry of sweeteners have been published (Walters et al., 1991). The proceedings of two International Union of Pure and Applied Chemistry symposia on sweeteners have been published (Kinghorn, 1997; Yamasaki and Tanaka, 2002).
11.7
References
and MORINI, G (2002) General pseudoreceptor model for sweet compounds: a semiquantitative prediction of binding affinity for sweettasting molecules. J. Med. Chem., 45, 4402±9. BELITZ, H-D, CHEN, W, JUGEL, H, TRELEANO, R, WIESER, H, GASTEIGER, J and MARSILI, M (1979) Sweet and Bitter Compounds: Structure and Taste Relationship. In Boudreau, JC (Ed.) Food Taste Chemistry. Washington, DC, American Chemical Society. BASSOLI, A, DREW, MG, MERLINI, L
Analysing and predicting properties of sweet-tasting compounds 289 (1914) Die Organischen Geschmacksstoffe [The Tastes of Organic Compounds], Berlin, Franz Siemenroth. CROSBY, GA, DUBOIS, GE and WINGARD, RE, JR. (1979) The Design of Synthetic Sweeteners. In ArieÈns, EJ (Ed.) Drug Design. New York, Academic Press. CULBERSON, JC and WALTERS, DE (1991) Development and Utilization of a ThreeDimensional Model for the Sweet Taste Receptor. In Walters, DE, Orthoefer, FT and Dubois, GE (Eds.) Sweeteners: Discovery, Molecular Design and Chemoreception. Washington, DC, American Chemical Society. DEUTSCH, EW and HANSCH, C (1966) Dependence of relative sweetness on hydrophobic bonding. Nature, 211, 75. DOUGLAS, AJ and GOODMAN, M (1991) Molecular basis of taste. A stereoisomeric approach. In Walters, DE, Orthoefer, FT and Dubois, GE (Eds.) Sweeteners: Discovery, Molecular Design, and Chemoreception. Washington, DC, American Chemical Society. DREW, MGB, WILDEN, GRH, SPILLANE, WJ, WALSH, RM, RYDER, CA and SIMMIE, JM (1998) Quantitative Structure-Activity Relationship Studies of Sulfamates RNHSO3ÿNa: Distinction between Sweet, Sweet-Bitter, and Bitter Molecules. J. Ag. Food Chem., 46, 3016±26. GOODMAN, M, CODDINGTON, J, MIERKE, DF and FULLER, WD (1987) A Model for the Sweet Taste of Stereoisomeric Retro-Inverso and Dipeptide Amides. J. Amer. Chem. Soc., 109, 4712±14. GOODMAN, M, YAMAZAKI, T, ZHU, YF, BENEDETTI, E and CHADHA, RK (1993) Structures of sweet and bitter peptide diastereomers by NMR, computer simulations, and X-ray crystallography. J. Amer. Chem. Soc., 115, 428±32. IWAMURA, H (1980) Structure±taste relationship of perillartine and nitro- and cyanoaniline derivatives. J. Med. Chem., 23, 308±12. IWAMURA, H (1981) Structure±sweetness relationship of L-aspartyl dipeptide analogues. A receptor site topology. J. Med. Chem., 24, 572±83. KIER, LB (1972) A molecular theory of sweet taste. J. Pharm. Sci., 61, 1394±7. KIER, LB and HALL, LH (1976) Molecular Connectivity in Chemistry and Drug Research. New York, Academic Press. KINGHORN, AD (1997) International Symposium on Sweeteners. Pure and Appl. Chem., 69, 655±727. COHN, G
KUNISHIMA, N, SHIMADA, Y, TSUJI, Y, SATO, T, YAMAMOTO, M, KUMASAKA, T, NAKANISHI, S, JINGAMI, H and MORIKAWA, K (2000) Structural basis of glutamate recognition by a dimeric metabotropic glutamate receptor. Nature, 407, 971±7. LELJ, F, TANCREDI, T, TEMUSSI, PA and TONIOLO, C (1976) Interaction of alpha-L-aspartyl-Lphenylalanine methyl ester with the receptor site of the sweet taste bud. J. Am. Chem. Soc., 98, 6669±75.
MADIGAN, DL, MULLER, GW, WALTERS, DE, CULBERSON, JC, DUBOIS, GE, CARTER, JS, NAGARAJAN, S, KLIX, RC, AGER, DJ and KLADE, CA (1989) Substituted Aryl Ureas as High Potency Sweeteners. Eur. Patent Appl. 89 115587.1, August 23, 1989; Eur. Patent Specification 0 355 819 A1, 1994. Europe. MAX, M, SHANKER, YG, HUANG, L, RONG, M, LIU, Z, CAMPAGNE, F, WEINSTEIN, H, DAMAK, S and MARGOLSKEE, RF (2001) Tas1r3, encoding a new candidate taste receptor, is allelic to the sweet responsiveness locus Sac. Nat. Genet., 28, 58±63. MAZUR, RH, SCHLATTER, JM and GOLDKAMP, AH (1969) Structure±taste relationships of some dipeptides. J. Am. Chem. Soc., 91, 2684±91. MCFARLAND, JW (1971) On the understanding of drug potency. In Jucker, E (Ed.) Progress in Drug Research. Basel, BirkhaÈuser Verlag.
290
Optimising sweet taste in foods
MULLER, GW, MADIGAN, DL, CULBERSON, JC, WALTERS, DE, CARTER, JS, KLADE, CA, DUBOIS, GE
and KELLOGG, MS (1991) High Potency Sweeteners Derived from beta-Amino Acids. In Walters, DE, Orthoefer, FT and Dubois, GE (Eds.) Sweeteners: Discovery, Molecular Design and Chemoreception. Washington, DC, American Chemical Society. NELSON, G, HOON, MA, CHANDRASHEKAR, J, ZHANG, Y, RYBA, NJ and ZUKER, CS (2001) Mammalian sweet taste receptors. Cell, 106, 381±90. OERTLY, E and MYERS, RG (1919) A new theory relating constitution to taste. Simple relations between the constitution of aliphatic compounds and their sweet taste. J. Amer. Chem. Soc., 41, 855±67. PEITSCH, MC (1996) ProMod and Swiss-Model: Internet-based tools for automated comparative protein modelling. Biochem. Soc. Trans., 24, 274±9. RAO, MNA and KUMAR, NA (1986) Quantitative Correlation Between Steric Factors and the Relative Sweetness of 2-Substituted 5-Nitroanilines. Indian Drugs, 23, 35±38. ROHSE, H and BELITZ, HD (1991) Shape of sweet receptors studied by computer modeling. In Walters, DE, Orthoefer, FT and Dubois, GE (Eds.) Sweeteners: Discovery, Molecular Design, and Chemoreception. Washington, DC, American Chemical Society. SHALLENBERGER, RS and ACREE, TE (1967) Molecular theory of sweet taste. Nature, 216, 480±2. SHALLENBERGER, RS and LINDLEY, MG (1977) A lipophilic-hydrophobic attribute and component in the stereochemistry of sweetness. Food Chem., 2, 145±53. SHALLENBERGER, RS, ACREE, TE and LEE, CY (1969) Sweet taste of D and L-sugars and amino-acids and the steric nature of their chemo-receptor site. Nature, 221, 555±6. SPADACCINI, R, TRABUCCO, F, SAVIANO, G, PICONE, D, CRESCENZI, O, TANCREDI, T and TEMUSSI, PA (2003) The mechanism of interaction of sweet proteins with the T1R2T1R3 receptor: evidence from the solution structure of G16A-MNEI. J. Mol. Biol., 328, 683±92. SPILLANE, WJ (1983) Quantitative structure-taste relationship studies of sulphamate sweeteners. Chem. and Ind. (London), 16±19. SPILLANE, WJ and MCGLINCHEY, G (1981) Structure-activity studies on sulfamate sweeteners II: semiquantitative structure±taste relationship for sulfamate (RNHSO3ÿ) sweeteners-the role of R. J. Pharm. Sci., 70, 933±5. SPILLANE, WJ, MCGLINCHEY, G, O MUIRCHEARTAIGH, I and BENSON, GA (1983) Structure± activity studies on sulfamate sweeteners III: structure±taste relationships for heterosulfamates. J. Pharm. Sci., 72, 852±6. SPILLANE, WJ, RYDER, CA, WALSH, MR, CURRAN, PJ, CONCAGH, DG and WALL, SN (1996) Sulfamate sweeteners. Food Chem., 56, 255±61. SPILLANE, WJ, RYDER, CA, CURRAN, PJ, WALL, SN, KELLY, LM, FEENEY, BG and NEWELL, J (2000) Development of structure±taste relationships for sweet and non-sweet heterosulfamates. J. Chem. Soc. Perkin Trans. 2, 1369±74. TEMUSSI, PA, LELJ, F and TANCREDI, T (1978) Three-dimensional mapping of the sweet taste receptor site. J. Med. Chem., 21, 1154±8. TEMUSSI, PA, LELJ, F and TANCREDI, T (1991) Structure±activity relationships of sweet molecules. In Walters, DE, Orthoefer, FT and DuBois, GE (Eds.) Sweeteners: Discovery, Molecular Design, and Chemoreception. Washington, DC, American Chemical Society. TINTI, J-M and NOFRE, C (1991a) Design of Sweeteners: A Rational Approach. In Walters, DE, Orthoefer, FT and DuBois, GE (Eds.) Sweeteners: Discovery, Molecular
Analysing and predicting properties of sweet-tasting compounds 291 Design, and Chemoreception. Washington, DC, American Chemical Society. and NOFRE, C (1991b) Why does a sweetener taste sweet? A new model. In Walters, DE, Orthoefer, FT and DuBois, GE (Eds.) Sweeteners: Discovery, Molecular Design, and Chemoreception. Washington, DC, American Chemical Society. VAN DER HEIJDEN, A, BRUSSEL, LBP and PEER, HG (1978) Chemoreception of Sweet-Tasting Dipeptide Esters: A Third Binding Site. Food Chem., 3, 207±11. VAN DER HEIJDEN, A, BRUSSEL, LBP and PEER, HG (1979) Quantitative structure-activity relationships (QSAR) in sweet aspartyl dipeptide methyl esters. Chem. Senses Flavour, 4, 141±52. VAN DER HEIJDEN, A, VAN DER WEL, H and PEER, HG (1985a) Structure±activity Relationships in Sweeteners. I. Nitroanilines, sulphamates, oximes, isocoumarins and dipeptides. Chem. Senses, 10, 57±72. VAN DER HEIJDEN, A, VAN DER WEL, H and PEER, HG (1985b) Structure±activity Relationships in Sweeteners. II. Saccharins, acesulfames, chlorosugars, tryptophans and ureas. Chem. Senses, 10, 73±88. VENANZI, TJ and VENANZI, CA (1988a) Ab initio molecular electrostatic potentials of perillartine analogues: implications for sweet-taste receptor recognition. J. Med. Chem., 31, 1879±85. VENANZI, TJ and VENANZI, CA (1988b) A Molecular Electrostatic-Potential Study of Acesulfame. Analyt. Chim. Acta, 210, 213±18. VENANZI, TJ and VENANZI, CA (1991) Electrostatic recognition patterns of sweet-tasting compounds. In Walters, DE, Orthoefer, FT and DuBois, GE (Eds.) Sweeteners: Discovery, Molecular Design, and Chemoreception. Washington, DC, American Chemical Society. VERLOOP, A, HOOGENSTRATEN, W and TIPKER, J (1976) New steric substituent parameters, New York, Academic Press. WALTERS, DE (2002) Homology-based model of the extracellular domain of the taste receptor T1R3. Pure and Appl. Chem., 74, 1117±23. WALTERS, DE and HINDS, RM (1994) Genetically evolved receptor models: a computational approach to construction of receptor models. J. Med. Chem., 37, 2527±36. WALTERS, DE, ORTHOEFER, FT and DUBOIS, GE (1991) Sweeteners: Discovery, Molecular Design and Chemoreception, Washington, DC, American Chemical Society. WALTERS, DE, PRAKASH, I and DESAI, N (2000) Active conformations of neotame and other high-potency sweeteners. J. Med. Chem., 43, 1242±5. XU, H, STASZEWSKI, L, TANG, H, ADLER, E, ZOLLER, M and LI, X (2004) Different functional roles of T1R subunits in the heteromeric taste receptors. Proc. Natl. Acad. Sci. USA, 101, 14258±63. YAMASAKI, K and TANAKA, O (2002) Special topic issue on the science of sweeteners. Pure and Appl. Chem., 74, 1101±316. TINTI, J-M
12 Discovering new natural sweeteners A. D. Kinghorn, The Ohio State University, USA and N.-C. Kim, Lovelace Respiratory Research Institute, USA
12.1
Introduction
There has been an increasing demand for novel alternative highly sweet, noncaloric and non-cariogenic sucrose substitutes for the diabetic and dietetic market, partly because sucrose consumption as a sweetener has been associated with nutritional and medical problems. Alternative sweeteners should have a sucrose-like taste quality with concomitant properties such as the lack of toxicity or cariogenicity, and should exhibit acceptable water solubility and heat stability. Sucrose substitutes that are at least 50±100 times sweeter than sucrose are termed `high-potency' or `high-intensity' sweeteners. At the moment, there are five FDA-approved high-intensity, non-nutritive sweeteners in the United States, namely, aspartame, acesulfame K, neotame, saccharin, and sucralose (Duffy et al., 2004). About 100 compounds, exclusive of monosaccharides, disaccharides, and polyols, have been discovered from green plants as natural sweeteners, and these belong to three major structural classes: isoprenoids, flavonoids, and proteins (Kim and Kinghorn, 2002a). Certain plant-derived natural compounds are currently used commercially in European, Asian, and Latin American countries as sweeteners, as will be discussed in more detail in the next section of this chapter. These compounds include the triterpene glycoside, glycyrrhizin (1), the semi-synthetic flavonoid glycoside, neohesperidin dihydrochalcone (2), the ent-kaurane diterpene glycosides, stevioside (3) and rebaudioside A (4), and the protein, thaumatin (see Fig. 12.1). However, none of these natural sweeteners is as yet approved by the FDA for use as a high-intensity sweetener in the United States. Thus far, all of the known natural product sweet-tasting compounds and sweetness modifiers have been isolated from green plants (ferns, monocotyledons,
Discovering new natural sweeteners
293
and dicotyledons), as opposed to lower plants, or microbial, or marine sources (Kurihara, 1992; Kinghorn et al., 1995; Suttisri et al., 1995). The highly sweet compounds from plants are representative of about 20 major structural types, with the principal groups being various terpenoids and steroids (isoprenoids), flavonoids, and proteins, in addition to compounds of other chemical classes such as an amino acid, a benzo[b]indeno[1,2-d]pyran, a dihydroisocoumarin, phenylpropanoids, and proanthocyanidins. In Table 12.1, the prototypes or most highly sweet example, for all of the chemical classes of the presently known highly sweet plant-derived compounds are summarized. A comprehensive list of natural sweeteners of plant origin was updated and published in a recent review (Kim and Kinghorn, 2002b). Table 12.1
Examples of highly sweet compounds from plants
Compound type/namea MONOTERPENE Perillartinec
SESQUITERPENES Bisabolane (+)-Hernandulcin
Plant name
Sweetness potencyb
Reference(s)
Perilla frutescens (L.) Britton (Labiatae)
370
Kinghorn and Soejarto (1986)
Lippa dulcis Trev. (Verbenaceae)
1500
Kinghorn and Soejarto (1986)
DITERPENES Diterpene acid 4 ,10-DimethylPine treee 1,2,3,4,5,10hexahydrofluorene-4,6dicarboxylic acidd
1300±1800f Kinghorn and Soejarto (1986)
ent-Kaurene glycosides Rebaudioside A Stevia rebaudiana (Bertoni) Bertoni (Compositae) Stevioside S. rebaudiana
242
Kinghorn and Soejarto (1986)
210
Kinghorn and Soejarto (1986)
Labdane glycoside Baiyunoside
Phlomis betonicoides Diels (Labiatae)
500
Kinghorn and Soejarto (1986)
Siraitia grosvenorii (Swingle) C. Jeffreyg (Cucurbitaceae)
250±425f
Kinghorn and Soejarto (1986)
TRITERPENES Cucurbitane glycoside Mogroside V
294
Optimising sweet taste in foods
Table 12.1 Continued Compound type/namea Cycloartane glycoside Abrusoside A
Dammarane glycoside Cyclocarioside A
Oleanane glycosides Glycyrrhizin Albiziasaponin B
Plant name
Sweetness potencyb
Reference(s)
Abrus precatorius L.; A. fruticulosus Wall et W.& A. (Leguminosae)
30
Choi et al. (1989a); Choi et al. (1989b)
Cyclocarya paliurus (Batal.) Iljinsk (Juglandaceae)
200
Yang et al. (1992)
Glycyrrhiza glabra L. (Leguminosae) Albizia myriophylla Benth. Leguminosae
Secodammarane glycoside Pterocaryoside A Pterocarya paliurus Batal. (Juglandaceae) STEROIDAL SAPONINS Osladin Polypodium vulgare L. (Polypodiaceae) Polypodoside A Polypodium glycyrrhiza DC. Eaton (Polypodiaceae) Telosma procumbens Telosmoside A15 (Hance) Merr. (Asclepiadaceae) PHENYLPROPANOIDS e.g., Foeniculum trans-Anetholeh vulgare Mill. (Umbelliferae) trans-Cinnamaldehyde Cinnamomum osmophleum Kanehira (Lauraceae) DIHYDROISOCOUMARIN Hydrangea macrophylla Phyllodulcini Seringe var. thunbergii (Siebold) Makino (Saxifragaceae) FLAVONOIDS Dihydrochalcone glycoside Neohesperidin Citrus aurantium L. (Rutaceae) dihydrochalconec
93±170f 600
Kinghorn and Soejarto (1986) Yoshikawa et al. (2002)
50
Kennelly et al. (1995)
500 600
Nishizawa and Yamada (1996) Kim et al. (1988)
1000
Huan et al. (2001)
13
Hussain et al. (1990b)
50
Kinghorn and Soejarto (1989)
400
Kinghorn and Soejarto (1986)
1000
Kinghorn and Soejarto (1986)
Discovering new natural sweeteners Table 12.1
295
Continued
Compound type/namea
Plant name
Dihydroflavonol Dihydroquercetin 3-O- Tessaria dodoneifolia acetate 40 -methyl etherc (Hook. & Arn.) Cabrera (Compositae)
Sweetness potencyb
Reference(s)
400
Kinghorn and Soejarto (1989)
PROANTHOCYANIDIN Selligueain A Selliguea feei Bory (Polypodiaceae)
35
Baek et al. (1993)
BENZO[b]INDENO[1,2d]PYRAN Hematoxylin Haematoxylon campechianum L. (Leguminosae)
120
Masuda et al. (1991)
AMINO ACID Monatin
PROTEINS Brazzein Monellin Thaumatin
Schlerochiton ilicifolius 1200±1400f Vleggaar et al. (1992) A. Meeuse (Acanthaceae) Pentadiplandra brazzeana Baillon (Pentadiplandraceae) Dioscoreophyllum cumminsii (Stapf) Diels (Menispermaceae) Thaumatococcus daniellii (Bennett) Benth. (Marantaceae)
2000
Ming and Hellekant (1994)
3000
Van der Wel (1972)
1600
Van der Wel (1972)
a The prototype or sweetest example of each known class of natural sweetener is listed, although for some categories more than one example is given. b Values of relative sweetness on a weight comparison basis to sucrose (= 1.0) are taken from the relevant literature source or from a review article/book chapter. c Semisynthetic derivative of natural product. d Semisynthetic sweetener. e Plant Latin binomial not given in the original reference. f Relative sweetness varied with the concentration of sucrose. g Formerly named Momordica grosvenorii Swingle and Thladiantha grosvenorii (Swingle) C. Jeffrey (Kinghorn and Kennelly, 1995). h Identified as a sweet-tasting constituent of six species in the paper cited. However, this compound has a much wider distribution in the plant kingdom. i The plant of origin may be crushed or fermented in order to generate phyllodulcin (6).
296
Optimising sweet taste in foods
Fig. 12.1
12.2
Structures of sweet natural products of plant origin and some derivatives.
Commercially used natural sweeteners
A reasonably high proportion of the compounds isolated from plant sources with a sweet taste have some commercial use, although several more have market potential. In the present section, only naturally occurring substances will be described. Of these, the currently used compounds include glycyrrhizin (1),
Discovering new natural sweeteners
297
neohesperidin dihydrochalcone (2), stevioside (3), rebaudioside A (4), mogroside V (5), phyllodulcin (6), and thaumatin. These are illustrated in Fig. 12.1 and will be briefly discussed, in turn. Glycyrrhizin (1), also known as glycyrrhizic acid, is an oleanane-type triterpenoid diglycoside isolated from the roots of Glycyrrhiza glabra L. (Leguminosae) and other species in the genus Glycyrrhiza (Kinghorn and Kennelly, 1995; Kitagawa, 2002). Glycyrrhizin (1) is 93±170 times sweeter than sucrose, depending on concentration. In Japan, root extracts of G. glabra (which contain >90% w/w pure glycyrrhizin) are used to sweeten foods and other products, such as cosmetics and medicines. Ammoniated glycyrrhizin has been on the Generally Recognized as Safe (GRAS) list of approved natural flavors in the United States, as a flavorant, flavor modifier, and foaming agent (Kinghorn and Compadre, 2001). Neohesperidin dihydrochalcone (2) is a semi-synthetic flavonoid glycoside produced from neohesperidin present in the peel of bitter orange (Citrus auranticum L.; Rutaceae), through catalytic hydrogenation under alkaline conditions. While various sweet-tasting dihydrochalcones have been isolated from natural sources (Kim and Kinghorn, 2002b), none of these is particularly highly sweet. Neohesperidin dihydrochalcone is not only highly sweet, but also has flavor-enhancing properties, and accordingly has a wide range of uses in foods and beverages in the countries of the European Union, where it is an authorized sweetener (Borrego and Montijano, 2001). Stevioside (3) and rebaudioside A (4) are ent-kaurene-type diterpene glycosides based on the aglycone steviol isolated from the leaves of the Paraguayan plant, Stevia rebaudiana (Bertoni) Bertoni (Compositae) (Kohda et al., 1976; Tanaka, 1997; Kinghorn et al., 2001), with stevioside (3) being the most abundant sweet compound in this plant part. The sweetness intensity of stevioside (3) has been rated as 210 times sweeter than sucrose, although this value varies with concentration. However, rebaudioside A (4) (the second most abundant S. rebaudiana entkaurene glycoside with a sweetness intensity rated as about 240 times sweeter than sucrose) is considerably more pleasant-tasting and more highly water-soluble than stevioside (3), and thus better suited for use in food and beverages. Extracts of S. rebaudiana containing stevioside and/or purified stevioside are permitted as food additives in Japan, South Korea, Brazil, Argentina, and Paraguay. In the United States, refined extracts of S. rebaudiana are currently widely available in health food stores as botanical `dietary supplements', and are sometimes packaged for use as table top sweeteners (Kinghorn et al., 2001). Mogroside V (5) is a cucurbitane-type triterpenoid glycoside isolated from the fruits of Siraitia grosvenorii (Swingle) C. Jeffrey (formerly Momordica grosvenorii Swingle) (Cucurbitaceae) (Takemoto et al., 1983). An extract of the dried fruits of S. grosvenorii, containing mogroside V (5) as the major sweet principle, is used in Japan as a sweetener for several foods and beverages. The sweetness intensity of mogroside V (5) has been rated as 250±425 times sweeter than sucrose, depending on concentration (Kinghorn and Compadre, 2001). A major corporation in the United States has expressed an interest in using extracts
298
Optimising sweet taste in foods
of S. grosvenorii and other Siraitia species in sweet juice compositions (Fischer et al., 1994). Phyllodulcin (6), a dihydroisocoumarin-type sweetener, occurs in glycosidic form in the leaves of Hydrangea macrophylla Seringe var. thunbergii (Siebold) Makino (Saxifragaceae) (`Amacha') and other species in this genus. After the fermentation of the leaves or by rubbing or crushing, the native glycosides are enzymatically hydrolyzed, and the sweet phyllodulcin (6, 400 times sweeter than 2% sucrose) is produced. The fermented leaves of H. macrophylla are used to prepare a sweet ceremonial tea in Japan, especially at `Hamatsuri', a Buddhist religious festival (Kinghorn and Compadre, 2001). Thaumatin (also known as TalinÕ protein) is constituted by two major sweet proteins, thaumatins I and II, obtained from the fruits of Thaumatococcus daniellii (Bennett) Benth. (Marantaceae), a plant of West African origin (Van der Wel, 1972; Kurihara, 1992). Thaumatins I and II have relative sweetness potencies of 1,600 and 3,000 when compared with sucrose on a weight basis. Thaumatin was first used as a sweetener in Japan, and is now approved in several other countries as a sucrose substitute. In the United States, however, it has GRAS status as a flavor adjunct, and is utilized in this manner in a wide range of foods and beverages (Kinghorn and Compadre, 2001).
12.3
Approaches to natural sweetener discovery
Three major approaches to access information on candidate plants with a potentially sweet taste may be proposed, prior to subsequent laboratory study. The first is a literature approach, through perusal of botanical and ethnobotanical scientific and popular texts, wherein accounts are searched on culinary uses of plants by indigenous or local communities. A specialized example of this approach is through the use of Index Kewensis, which serves as a taxonomic compendium of all published Latin binomial names of angiosperms and gynnosperms. A plant part that tastes or smells sweet may be assigned an epithet in Index Kewensis signifying such as `dulcificum', `dulcis', `glycyrrhiza', mellifera', `mellosa', and `saccharum' (Hussain et al., 1988). A second and more direct approach to the discovery of sweet-tasting plants is through field work, inclusive of interviews with members of indigenous communities and local healers or herbalists, followed by checking on the sweet taste of the plant or plants concerned. This approach can be accomplished most effectively in medicinal plant marketplaces in developing countries, and is followed up by searching for a specimen of the sweet plant concerned growing in the wild (Soejarto et al., 1983). However, as a consequence of the signing in 1992 of the United Nations Convention of Biological Diversity in Rio de Janeiro (the `Rio Summit') by the vast majority of countries of the world, it is now necessary to obtain prior informed consent before making inquiries on the sweet taste of plants. Scientific investigators must make provisions for compensation of persons in countries where plants are sourced, in lieu of the information
Discovering new natural sweeteners
299
obtained, such as the equitable sharing of benefits that may arise in the event of commercial development of indigenous traditional knowledge (Greaves, 1994). A third approach to finding previously undocumented sweet-tasting plants involves their organoleptic testing for sweetness, after appropriate safety precautions are made. This approach can be performed either in the field as a part of a plant collection expedition or in the laboratory (e.g., using herbarium specimens) (Soejarto et al., 1982; Kinghorn and Soejarto, 2002). At the beginning of their chemical examination in the laboratory, candidate sweet-tasting plants may be extracted initially with methanol-water (4:1), which serves as a general solvent for most plant secondary metabolites. After appropriate safety evaluation by toxicity testing, dried extracts are tested for the presence or absence of sweet taste. The type of toxicity testing carried out in our laboratory consists of acute toxicity testing and evaluation in a bacterial mutagenicity assay. For acute toxicity testing, the extract administered intraperitoneally (IP) at a dose of up to 2 g/kg body weight in suspension in sodium carboxymethylcellulose and animals are examined for up to 14 days after treatment for both mortality and loss of body weight. A forward mutation assay using Salmonella typhimurium strain TM677 has been employed both in the absence and presence of a metabolic activator. Extracts are tasted only if neither acutely toxic nor mutagenic. If found to be sweet-tasting, the initial dried extract is then dissolved in methanol-water (1:1), and partitioned sequentially with petroleum ether, ethyl acetate, and 1-butanol. When sweetness is detected in one or other of these fractions, the polarity of the solvent may serve as a useful predictor of the general chemical class exhibited by the pure sweet plant constituent(s) present. For example, sweet-tasting glycosides would be expected to occur in the 1-butanol-soluble extract (Kinghorn and Soejarto, 2002). In many cases, however, the presence of a sweet taste in a plant extract is most likely due to high concentration levels of sugars and polyols. Therefore, in the search for novel high-intensity sweeteners from plants, an initial `dereplication' stage is desirable. If the combined amounts of monosaccharides, disaccharides, and polyols exceeds 5% w/w in a given plant part, the resultant sweetness can be considered as being due to the presence of such `bulk' sweeteners. A suitable procedure using gas chromatography-mass spectrometry (GC-MS) has been developed for the purpose of ruling out the sweetness contribution from sugars and polyols in plants (Chung et al., 1997). Another group of common plant constituents that may taste sweet are phenylpropanoids such as trans-anethole and trans-cinnamaldehyde. If these compounds are in high concentration, they may contribute a sweet taste to the plant part under consideration and these phenylpropanoids usually partitioned into the plant petroleum ether fraction. Again, if present, they can be rapidly identified by GCMS (Hussain et al., 1990a). Once a compound is obtained in pure form after extraction, fractionation, and isolation steps, it is then subjected to a preliminary evaluation of safety via acute toxicity and mutagenicity testing as described above. If sweet-tasting, our laboratory has applied a threshold sensory test method using a small panel of
300
Optimising sweet taste in foods
human volunteers in good health to evaluate the sweetness potency of pure compounds relative to 2% w/v sucrose in water solution (Kim et al., 1988; Choi et al., 1989a; Baek et al., 1993; Kinghorn and Soejarto, 2002).
12.4
Improvement of sweet taste
There have been efforts to improve sweetness of naturally occurring sweeteners through their structural modification. For example, several attempts using various glycosylation methods have been made to increase the sweetness potency of glycyrrhizin (1). The group of the late Professor Osama Tanaka at Hiroshima University in Japan glycosylated glycyrrhetinic acid to afford various glycyrrhizin monoglycoside analogs using a chemical and enzymatic glycosylation procedure (Mizutani et al., 1994). A coupling reaction using mercury(II) cyanide [Hg(CN)2] for chemical glycosylation was effected, resulting in a significant enhancement of sweetness in the analogs obtained, especially the 3-O- -D-xylopyranoside (7) and 3-O- -D-glucuronide (MGGR, 8). The sweetness intensities of compounds 7 and 8 were rated as 544 and 941 times sweeter than sucrose, respectively. Such chemically modified products of glycyrrhizin also showed improved taste qualities (Tanaka, 1997). MGGR (8), in being more than five times sweeter than glycyrrhizin (1), as well as being readily soluble in water, is now used commercially as a sweetening agent in Japan (Mizutani et al., 1998). Over the years, there have been many attempts to improve the taste qualities of the major S. rebaudiana sweet steviol glycoside, stevioside (3), because of its sensory limitations (Kamiya et al., 1979; DuBois et al., 1984; Esaki et al., 1984; Mizutani et al., 1989; Ishikawa et al., 1990; Tanaka, 1997). Several synthetic studies on the structure-sweetness relationship of steviol glycosides have been conducted (Fukunaga et al., 1989; Mizutani et al., 1989; Ohtani and Yamasaki, 2002). For example, the sweetness-pleasantness of stevioside (3) may be increased by treating stevioside-galactosyl ester (Sgal), prepared by removal of the 19-O-glucosyl group of stevioside, and replacing it with a galactosyl group. Transglucosylation of Sgal with soluble starch using CGTase prepared from Bacillus macerans then affords a mixture of mono-, di-, tri-, and tetra--glycosylated compounds. The product with four glucosyl units attached at the C-13 position showed an enhanced sweetness (9, Sgal-2) (Mizutani et al., 1989). A rebaudioside A analog (10) was a (sodiosulfo)propyl group at C19 in place of a -glucosyl moiety showed improved sweetness qualities (DuBois et al., 1984). Stevioside (3) has been converted synthetically to rebaudioside A (4), by removing a glucose unit from stevioside at the C-13 position using amylase and then reintroducing synthetically two glucose units of different linkage to the remaining glucose unit at the C-13 position (Kaneda et al., 1977). In Japan, the largest market for the S. rebaudiana sweeteners to date, three different forms of stevia sweetener products are commercially available,
Discovering new natural sweeteners
301
namely, `stevia extract', `sugar-transferred stevia extract' (also known as `enzymatically modified stevia extract' and `glucosyl stevia'), and `rebaudioside A-enriched stevia extract' (Mizutani and Tanaka, 2002). `Stevia extract' is a powder or granule made by several industrial steps and standardized so as to contain more than 80% of steviol glycosides, inclusive of dulcoside A (3±5%), rebaudioside A (20±25%), rebaudioside C (5±10%), and stevioside (50±55%) (Shibasato, 1995; Mizutani and Tanaka, 2002). `Sugar-transferred stevia extract', a complex mixture of compounds, is made by transglycosylation of steviol glycosides present in commercially available `stevia extract' with a cyclomaltodextringlucanotransferase (CGTase)-starch system prepared from Bacillus macerans, followed by treatment with -amylase (Tanaka, 1997; Mizutani and Tanaka, 2002; Ohtani and Yamasaki, 2002). `Rebaudioside Aenriched extract' is made from improved varieties of S. rebaudiana, which produce more rebaudioside A (4) than the native Paraguayan species (Shibasato, 1995). Monellin is a sweet protein, about 3,000 times sweeter than sucrose, extracted from the African serendipity berry, Dioscoreophyllum cumminsii (Stapf) Diels (Menispermaceae) (Van der Wel, 1972; Kurihara, 1992). Monellin (or more accurately monellin 4) consists of two non-covalently associated polypeptide chains, A chain with 44 amino acid residues and B chain with 50 residues Ê . The (Kohmura et al., 2002). The crystal structure has been determined at 2.75 A individual A and B chains are not sweet, and the native conformation is essential for the sweet taste. Despite the fact that monellin is not used commercially, there remains a considerable interest in this compound in the scientific literature. Monellin was redesigned to increase thermal stability and renaturability using information from the crystal structure and genetic engineering techniques (Kim et al., 1991). Two chains of monellin were fused into a single chain using linkers and this newly fused protein showed the same sweetness with higher stability toward pH and temperature changes than monellin itself. This can be cloned and expressed in the bacterium Escherichia coli and the sweetness of the expressed protein was found to be the same as the natural sweetener. The relationship between the sweet taste and the structure of monellin was also examined by chemical modifications (Kohmura et al., 2002).
12.5
Future trends
There remains a strong demand for highly sweet non-caloric and non-cariogenic substances to substitute for sucrose in the diet. Despite the relatively small number of naturally occurring high potency sweeteners thus far discovered from higher plants, it is surprising that so many of these substances have some potentiality for commercialization, either in their naturally occurring or structurally modified form. Naturally occurring sweet-tasting compounds remain of considerable interest as sweeteners among consumers and manufacturers alike. Since 1995, preparations from S. rebaudiana have been used extensively in the
302
Optimising sweet taste in foods
United States as a `dietary supplement' and are widely available to the public. Stevioside and S. rebaudiana extracts thus far have shown no cases of clinical toxicity as documented in the biomedical literature, despite their ever-increasing use. In contrast, a large intake of glycyrrhizin as contained in licorice flavored confectionary may lead to pseudoaldosteronism (De Klerk et al., 1997; Van Rossum et al., 2001; Dalton, 2002). It can be expected that additional investigations on the safety of the S. rebaudiana sweeteners will continue to be conducted until a more complete understanding is obtained. It is of interest to note recent reports on the potential cancer chemopreventive activities of glycyrrhizin (1), stevioside (3), and mogroside V (5) (Konoshima and Takasaki, 2002). This suggests the potential use of natural sweeteners in `nutraceutical' or `functional food' compositions. Moreover, the increasing number of reports suggesting that certain diterpene and triterpene natural sweeteners have potential for use in the treatment of diabetes deserve closer scrutiny (Chen et al., 2005; Suzuki et al., 2005). It is possible that all of the more obvious candidate highly sweet plants have already been studied in the laboratory, and that in the future, novel natural sweetener chemotypes will be discovered either by following up on ethnobotanical leads of plants used for sweetening by indigenous peoples in remote areas, or by random organoleptic evaluation of plants collected in the field for other purposes (Kinghorn, 2002). In terms of the prospects of discovery of future highly sweet natural products from plants using ethnobotanical approaches, it will probably be necessary to access more remote geographical areas than previously in order to obtain candidate sweettasting plants. Thus, it is most advantageous in sweetener discovery projects from natural sources to work in a multidisciplinary team composed of botanists, natural products chemists, and biologists (Kinghorn and Soejarto, 2002). Also, since there has been considerable progress leading to the identification of the T1R family of receptors that respond to sweet stimuli (Nelson et al., 2001; Li et al., 2002), it is possible that new receptor binding assays can be developed to aid with the discovery of new natural sweeteners in the future (Marris, 2005), instead of relying on human panels to taste crude extracts, fractions, and pure isolates.
12.6
Sources of further information
(2002), `Safety evaluation of Stevia and stevioside', in Atta-ur-Rahman, Studies in Natural Products Chemistry, Amsterdam, Elsevier, 299±319. KINGHORN A D, ed. (2002), Stevia: the Genus Stevia, London, Taylor & Francis. O'BRIEN NABORS L, ed. (2001), Alternative Sweeteners: Third Edition, Revised and Expanded. Marcel Dekker, New York. Pure and Applied Chemistry, Special Issue on the Science of Sweeteners, volume 74, July 2002, iv and 1101±1316. GEUNS J M C
Discovering new natural sweeteners
12.7
303
References
BAEK N-I, CHUNG M-S, SHAMON L, KARDONO L B S, TSAURI S, PADMAWINATA K, PEZZUTO J M,
SOEJARTO D D and KINGHORN A D (1993), `Selligueain A, a novel highly sweet proanthocyanidin from the rhizomes of Selliguea feei', J Nat Prod, 56, 1532±1538. BORREGO F and MONTIJANO H (2001), `Neohesperidin dihydrochalcone', in O'Brien Nabors L, Alternative Sweeteners, Third Edition, Revised and Expanded, New York, Marcel Dekker, 87±104. CHEN T H, CHEN S C, CHAN P, CHU Y L, YANG H Y and CHENG J T (2005), `Mechanism of the hypoglycemic effect of stevioside, a glycoside of Stevia rebaudiana', Planta Med, 71, 108±113. CHOI Y-H, HUSSAIN R A, PEZZUTO J M, KINGHORN A D and MORTON J F (1989a), `Abrusosides A-D, four novel sweet-tasting triterpene glycosides from the leaves of Abrus precatorius', J Nat Prod, 52, 1118±1127. CHOI Y-H, KINGHORN A D, SHI Z, ZHANG H and TEO B K (1989b), `Abrusoside A: a new type of highly sweet triterpene glycoside', J Chem Soc, Chem Commun, 887±888. CHUNG M-S, KIM N-C, LONG L, SHAMON L, AHMAD W-Y, SAGRERO-NIEVES L, KARDONO L B S,
KENNELLY E J, PEZZUTO J M, SOEJARTO D D and KINGHORN A D (1997), `Dereplication of saccharide and polyol constituents of candidate sweet-tasting plants: isolation of the sesquiterpene glycoside mukurozioside IIb as a sweet principle of Sapindus rarak', Phytochem Anal, 8, 49±54. DALTON L (2002), `Licorice root is used worldwide as a flavor and a medicine', Chem Eng News (August 12), 37. DE KLERK G J, NIEUWENHUIS M G and BEUTLER J J (1997), `Lesson of the week: hypokalaemia and hypertension associated with use of liquorice flavoured chewing gum', BMJ, 314, 731±732. DUBOIS G E, BUNES L A, DIETRICH P S and STEPHENSON R A (1984), `Diterpenoid sweeteners. Synthesis and sensory evaluation of biologically stable analogues of stevioside', J Agric Food Chem, 32, 1321±1325. DUFFY V B, SIGMAN-GRANT M, POWERS M A, ELMORE D, MYERS E F, QUAGLIANI D, SPANO M,
and CONNOR S (2004), `Position of the American Dietetic Association: use of nutritive and nonnutritive sweeteners', J Am Diet Assoc, 104, 255±275. ESAKI S, TANAKA R and KAMIYA S (1984), `Synthesis and taste of certain steviol glycosides', Agric Biol Chem, 48, 1831±1834. FISCHER C M, HARPER H J, HENRY JR. W J, MOHLENKAMP JR. M K, ROMER K and SWAINE JR. R L (1994), Beverage Compositions and Sweetening Compositions which Contain Juice Derived from Botanical Subfamily Cucurbitaceae, The Procter & Gamble Company, United States. FUKUNAGA Y, MIYATA T, NAKAYASU N, MIZUTANI K, KASAI R and TANAKA O (1989), `Enzymatic transglucosylation products of stevioside: separation and sweetness evaluation', Agric Biol Chem, 53, 1603±1607. GREAVES T (1994), Intellectual Property Right for Indigenous Peoples: A Sourcebook. Oklahoma City, Oklahoma, Society for Applied Anthropology. HUAN V D, OHTANI K, KASAI R, YAMASAKI K and TUU N V (2001), `Sweet pregnane glycosides from Telosma procumbens', Chem Pharm Bull, 49, 453±460. HUSSAIN R A, KINGHORN A D and SOEJARTO D D (1988), `Sweetening agents of plant origin: literature search for candidate sweet plants', Econ Bot, 42, 267±283. HUSSAIN R A, LIN Y-M, POVEDA L J, BORDAS E, CHUNG B S, PEZZUTO J M, SOEJARTO D D and STITZEL K F, TAYLOR S, EARL R
304
Optimising sweet taste in foods
KINGHORN A D (1990a), `Plant-derived sweetening agents: saccharide and polyol constituents of some sweet-tasting plants', J Ethnopharmacol, 28, 103±115. HUSSAIN R A, POVEDA L J, PEZZUTO J M, SOEJARTO D D and KINGHORN A D (1990b), `Sweetening agents of plant origin: phenylpropanoid constituents of seven sweettasting plants', Econ Bot, 44, 174±182. ISHIKAWA H, KITAHATA S, OHTANI K, IKUHARA C and TANAKA O (1990), `Production of stevioside and rubusoside derivatives by transfructosylation of -fructofuranosidase', Agric Biol Chem, 54, 3137±3143. KAMIYA S, KONISHI F and ESAKI S (1979), `Synthesis and taste of some analogs of stevioside', Agric Biol Chem, 43, 3553±3557. KANEDA N, KASAI R, YAMASAKI K and TANAKA O (1977), `Chemical studies on sweet diterpene-glycosides of Stevia rebaudiana: conversion of stevioside into rebaudioside A', Chem Pharm Bull, 25, 2466±2467. KENNELLY E J, CAI L, LONG L, SHAMON L, ZAW K, ZHOU B-N, PEZZUTO J M and KINGHORN A D (1995), `Novel highly sweet secodammarane glycosides from Pterocarya paliurus', J Agric Food Chem, 43, 2602±2607. KIM J, PEZZUTO J M, SOEJARTO D D, LANG F A and KINGHORN A D (1988), `Polypodoside A, an intensely sweet constituent of the rhizomes of Polypodium glycyrrhiza', J Nat Prod, 51, 1166±1172. KIM N-C and KINGHORN A D (2002a), `Sweet-tasting and sweetness-modifying constituents of plants', in Atta-ur-Rahman, Studies in Natural Products Chemistry Vol 27 Bioactive Natural Products Part H, Amsterdam, Elsevier Scientific Publishers, 3±57. KIM N-C and KINGHORN A D (2002b), `Highly sweet compounds of plant origin', Arch Pharm Res, 25, 725±746. KIM S-H, KANG C-H and CHO J-M (1991), `Sweet Proteins', in Walters D E, Orthoefer F T and DuBois G E, Sweeteners, Discovery, Molecular Design, and Chemoreception, Washington DC, American Chemical Society, 28±40. KINGHORN A D (2002), `Overview', in Kinghorn A D, Stevia: the Genus Stevia, London, Taylor & Francis, 1±17. KINGHORN A D and COMPADRE C M (2001), `Less common high-potency sweeteners', in O'Brien Nabors L, Alternative Sweeteners, Third Edition, Revised and Expanded, New York, Marcel Dekker, 209±233. KINGHORN A D and KENNELLY E J (1995), `Discovery of highly sweet compounds from natural sources', J Chem Educ, 72, 676±680. KINGHORN A D and SOEJARTO D D (1986), `Sweetening agents of plant origin', CRC Crit Rev Plant Sci, 4, 79±120. KINGHORN A D and SOEJARTO D D (1989), `Intensely sweet compounds of natural origin', Med Res Rev, 9, 91±115. KINGHORN A D and SOEJARTO D D (2002), `Discovery of terpenoid and phenolic sweeteners from plants', Pure Appl Chem, 74, 1169±1179. KINGHORN A D, FULLAS F and HUSSAIN R A (1995), `Structure-activity relationships of highly sweet natural products', in Attar-ur-Rahman, Studies in Natural Product Chemistry Vol. 15, Structure and Chemistry (Part C), Amsterdam, Elsevier Science Publishers, 3±41. KINGHORN A D, WU C D and SOEJARTO D D (2001), `Stevioside', in O'Brien Nabors L, Alternative Sweeteners, Third Edition, Revised and Expanded, New York, Marcel Dekker, 167±183. KITAGAWA I (2002), `Licorice root. A natural sweetener and an important ingredient in Chinese medicine', Pure Appl Chem, 74, 1189±1198.
Discovering new natural sweeteners
305
and TANAKA O (1976), `New sweet diterpene glycosides from Stevia rebaudiana', Phytochemistry, 15, 981±983. KOHMURA M, MIZUKOSHI T, NIO N, SUZUKI E-I and ARIYOSHI Y (2002), `Structure±taste relationships of the sweet protein monellin', Pure Appl Chem, 74, 1235±1242. KONOSHIMA T and TAKASAKI M (2002), `Cancer-chemopreventive effects of natural sweeteners and related compounds', Pure Appl Chem, 74, 1309±1316. KURIHARA Y (1992), `Characterization of antisweet substance, sweet proteins, and sweetness-inducing proteins', Crit Rev Food Sci Nutr, 32, 231±252. LI X, STASZEWSKI L, XU H, DURICK K, ZOLLER M and ADLER E (2002), `Human receptors for sweet and umami taste', Proc Natl Acad Sci USA, 99, 4692±4696. MARRIS E (2005), `More flavour up front', Nature, 435, 559. MASUDA H, OHTANI K, MIZUTANI K, OGAWA S, KASAI R and TANAKA O (1991), `Chemical study on Haematoxylon campechianum: a sweet principle and new dibenz[b,d]oxocin derivatives', Chem Pharm Bull, 39, 1382±1384. MING D and HELLEKANT G (1994), `Brazzein, a new high-potency thermostable sweet protein from Pentadiplandra brazzeana B', FEBS Lett, 355, 106±108. MIZUTANI K and TANAKA O (2002), `Use of Stevia rebaudiana sweeteners in Japan', in Kinghorn A D, Stevia: the Genus Stevia, London, Taylor & Francis, 178±195. MIZUTANI K, MIYATA T, KASAI R, TANAKA O, OGAWA S and DOI S (1989), `Study on improvement of sweeteners of steviol bisglycosides: selective enzymatic transglucosylation of the 13-O-glycosyl moiety', Agric Biol Chem, 53, 395±398. MIZUTANI K, KURAMOTO T, TAMURA Y, OHTAKE N, DOI S, NAKAMURA M and TANAKA O (1994), `Sweetness of glycyrrhetic acid 3-O- -D-monoglucuronide and the related glycosides', Biosci Biotechnol Biochem, 58, 554±555. KOHDA H, KASAI R, YAMASAKI K, MURAKAMI K
MIZUTANI K, KAMBARA T, MASUDA H, TAMURA Y, IKEDA T, TANAKA O, TOKUDA H, NISHINO H,
KOZUKA M, KONOSHIMA T and TAKASAKI M (1998), `Glycyrrhetic acid monoglucuronide (MGGR): biological activities', in Ageta H, Aimi N, Ebizuka Y, Fujita T and Honda G, Toward Natural Medicine Research in the 21st Century, Amsterdam, Elsevier, 225±235. NELSON G, HOON M A, CHANDRASHEKAR J, ZHANG Y, RYBA N J P and ZUKER C S (2001), `Mammalian sweet taste receptors', Cell, 106, 381±390. NISHIZAWA M and YAMADA H (1996), `Intensely sweet saponin osladin: synthetic and structural study', Adv Exp Med Biol, 405, 25±36. OHTANI K and YAMASAKI K (2002), `Methods to improve the taste of the sweet principles of Stevia rebaudiana', in Kinghorn A D, Stevia: the Genus Stevia, London, Taylor & Francis, 138±159. SHIBASATO M (1995), `Current status of Stevia sweeteners and its application', Japan Food Sci, 51±58. SOEJARTO D D, KINGHORN A D and FARNSWORTH N R (1982), `Potential sweetening agents of plant origin. III. Organoleptic evaluation of Stevia leaf herbarium samples for sweeteners', J Nat Prod, 45, 590±599. SOEJARTO D D, COMPADRE C M, MEDON P J, KAMATH S K and KINGHORN A D (1983), `Potential sweetening agents of plant origin. II. Field search for sweet tasting Stevia species', Econ Bot, 37, 71±75. SUTTISRI R, LEE I-S and KINGHORN A D (1995), `Plant-derived triterpenoid sweetness inhibitors', J Ethnopharmacol, 47, 9±26. SUZUKI Y A, MURATA Y, INUI H, SUGIURA M and NAKANO Y (2005), `Triterpene glycosides of Siraitia grosvenori inhibit rat intestinal maltase and suppress the rise in blood glucose level after a single oral administration of maltose in rats', J Agric Food
306
Optimising sweet taste in foods Chem, 53, 2941±2946.
and OKUHIRA M (1983), `Studies on the constituents of Fructus Momordicae. I. On the sweet principle', Yakugaku Zasshi, 103, 1151± 1154. TANAKA O (1997), `Improvement of taste of natural sweeteners', Pure Appl Chem, 69, 675±683. VAN DER WEL H (1972), `Isolation and characterization of the sweet principles from Dioscoreophyllum cumminsii', FEBS Lett, 21, 88±90. VAN ROSSUM T G J, DE JONG F H, HOP W C J, BOOMSMA F and SCHALM S W (2001), ```Pseudoaldosteronism'' induced by intravenous glycyrrhizin treatment of chronic hepatitis C patients', Neth J Gastroenterol Hepatol, 16, 789±795. VLEGGAAR R, ACKERMAN L G J and STEYN P S (1992), `Structure elucidation of monatin, a high-intensity sweetener isolated from the plant Schlerochiton ilicifolius', J Chem Soc, Perkin Trans 1, 3095±3098. YANG D J, ZHONG Z C and XIE Z M (1992), `Studies on the sweet principles from the leaves of Cyclocarya paliurus (Batal.) Iljinskaya', Yao Hsueh Hsueh Pao, 27, 841±844. YOSHIKAWA M, MORIKAWA T, NAKANO K, PONGPIRIYADACHA Y, MURAKAMI T and MATSUDA H (2002), `Characterization of new sweet triterpene saponins from Albizia myriophylla', J Nat Prod, 65, 1638±1642. TAKEMOTO T, ARIHARA S, NAKAJIMA T
13 Molecular design and the development of new sweeteners J. Polanski, University of Silesia, Poland
13.1
Introduction
Although artificial sweeteners get better and better imitating sucrose their sweetness can still be distinguished from natural sugar. This makes a demand for a novel product which will be appreciated by the food industry. Similarly to pharmaceuticals, non-caloric sweeteners are mainly synthetic products developed by chemists. Basically, the most important objective of chemistry is to provide molecules triggering a desired answer of chemical or biological systems. This fact is well realised and can be commented after Hammond and Sharpless: `the most fundamental and lasting objective of synthesis is not a production of new compounds but the production of properties' (Kolb et al., 2001). Clearly more and more sophisticated tools are developed for the efficient translation of the molecular structure into the compound property space. However, nowadays the discovery in this field is still a complex issue that lacks a general approach. Pharmaceuticals (commonly: drugs) represent the most important market share among such products and consequently theoretical problems concerning the design and development of such substances are often depicted as drug design. Similarly, often all biologically active molecules targeted by chemists are referred to as drugs. On the other hand, the term molecular design has been coined to properly describe this research area. The design of a new pharmaceutical is extremely complicated. However, developing a new sweetener can be even more difficult, if we realised that the use of pharmaceuticals is usually restricted to a certain group of patients that apply them in small doses, generally, in a certain time period during illness. Sweeteners on the other hand are to be used without restrictions in much larger doses than pharmaceuticals. Frequently, even if it is in some sense risky, it is
308
Optimising sweet taste in foods
better to take a drug than to die from the disease. On the contrary, there is no similar pressure for the application of alternative sweeteners. Sucrose chemoreception is a subtle and complex effect that cannot be easily imitated by other substances, and even a minute differences in sweetness profile can cause human aversion.
13.2
The historical development of sweetness consumption
Nature decided sweet addiction of humans by favouring sweetness during natural selection. This was needed for the positive appraisal of food quality. Sweet is a synonym of benefit or favourable interaction. Sweet is a synonym for a person of a pleasant appearance, nice behaviour or just a success. On the contrary, a number of dangerous xenobiotics are marked negatively with bitterness which gives a warning to humans. This makes `adaptive sense, since sweetness is an indicator of calories in nature, and bitterness correlates with toxicity' (Rozin, 1989). The adaptive significance of taste chemoreception makes sense only with the provision that sweetness as such is a rare quality in nature. `These genetic predispositions evolved over thousands of years of human history, when foods ± especially foods high in energy density ± were relatively scarce' (Birch, 1999). Therefore, although we know a number of sweet tasting substances, sweetness is relatively exceptional. Serendipity clearly backed humans in their original search for sweetness. Natural fruits and honey have been sought after since prehistoric times. It was recognised from Stone Age paintings that honey was obtained as early as 20,000 years ago and Egyptians domesticated bees to raise it. Honey sweetness was surrogated by sugar cane which has been known as early as 1000 BC in India, where it originally came from Oceania (Van der Wel et al., 1987). A first authentic reference to honey and sugar is a birchbark scroll that dates back to the 4th century (Inglett, 1976). Europeans were trading with sugar from the 7th century, and in the 18th century Margraff discovered sucrose in red beets, which allow for the development of modern sucrose production in Europe to replace sugar cane delivery which had been blocked during a British blockade of Napoleonic France (Inglett, 1976). On the other hand, the uniqueness of sweetness decided its luxury. Sugar taxation clearly put the final seal on that fact. However, sugar prices (1400±1960) given in new English pence, steadily decreased from year to year from 875 in 1400; 210 (1600); 70 (1800); 17.5 (1900) to 3.7 in 1960, respectively (Ruprecht, 2001). High availability of sucrose and a significant increase in its consumption is a relatively novel effect that starts from 1850 to 1950, differing from country to country. Sugar availability `has contributed to destroying the old agricultural meal order which centered around the meal that was prepared by and consumed within the family. Sugar was associated with fast food, in the double sense of production and consumption' (Ruprecht, 2001). The availability or even an excess of sugar has deprived sweetness of its adaptive function; however; sweet desire and addiction
Molecular design and the development of new sweeteners
309
has survived. This explains the origin of the search and development of artificial non-caloric sweeteners.
13.3
Commercial alternative sweetener discoveries
From the molecular design point of view not a single sweetener of the currently approved and used commercial sweeteners (aspartame, acesulfame K, saccharin, sucralose or cyclamate) has been designed. All discoveries in this field were completely serendipitous (Sardesai and Waldshan, 1991). Constantine Fahlberg working with Ira Remsen at Johns Hopkins University in 1879 synthesising orthobenzoyl sulfimide and realised during eating dinner that his fingers tasted sweet and that they had discovered a new sweetener. A similar story describes the discovery of aspartame which was obtained by Jim Schlatter who worked in 1965 on new pharmaceuticals for gastric ulcers. He allegedly licked his finger to pick up a piece of paper and discovered the enormous sweetness of the compound. Similarly, paper picking in the lab provided another sweetener, acesulfame. Karl Clauss discovered this tastant in 1967. Cigarette smoking in the lab by a graduate student of the University of Illinois, Michael Sveda, who worked on new anti-pyretic drugs, led to the discovery of cyclamate sweetness in 1937. Another graduate student working for Tate & Lyle who was attempting
Fig. 13.1
Molecular formulae of the sweeteners having commercial importance.
310
Optimising sweet taste in foods
to find new sugar intermediates for organic synthesis discovered the sweetness of a halogentated sugar that is now known as sucralose. Shashikant Phadnis tasted the compound allegedly by language misunderstanding (Walters, 2001). This allowed Leslie Hough of the laboratory at King's College in London to develop a new sweetener (Hough, 1989). Neotame is a novel sweetener that differs from the others in a fact that it was obtained by the intentional modification of aspartame (Nofre and Tinti, 2000; Witt, 1999). As its application has been just approved by the American FDA (Neotame, 2002), it is the first alternative sweetener to be designed and approved for use.
13.4
Molecular design ± novel approaches novel chances
Although we still cannot do without a good measure of serendipity in this field, the costs of the technology involved in the development of a new drug means that the trial and error strategy is unacceptable. This justifies the conclusion that molecular design should improve the success ratio in drug discovery interpreted as screening novel drug candidates in a virtual chemical compound space. In fact, the expansion of rational techniques is still more and more evident. Basically, two strategies are possible for such research. In structure based design drug- or ligand-receptor interactions can be investigated in an effort to find novel effectors. Instead, in the cases when not enough data describes directly the target receptor the information on the apparent receptor is coded by the ligand structures. Thus, ligand based design investigates into the similarity between a series of ligands that are stimulating the receptor. Such indirect receptor investigations are frequently called receptor or pharmacophore mapping (Martin, 1997; van de Waterbeemd et al., 1997). Usually a chemical interpretation of pharmacophore emphasises a selection of a certain atom subset among all possible atoms within the series of ligands that forms a common or similar pattern for all active molecules. Thus, a pharmacophore in its most general meaning is just a receptor or receptor sector model deduced from a series of its ligands. In fact, no information is available on what the relation between a real receptor and this subset is. In taste chemistry glucophore is a sweet taste pharmacophore. Pharmacophore mapping is a broad strategy which includes a variety of experimental and computational approaches. In order to map any pharmacophore we need to know the structures of chemical compounds that bind this particular receptor. Thus, a synthesis of the compounds of diversified molecular structures is needed at first. This describes, usually qualitatively, structure±activity relationships (SAR). Chemical, physical and biological characteristics of the obtained compounds can now be measured and used in further computational analysis of these data. Generally, the active site cavity is defined by performing molecular superimposition of the active molecules and relating this to the inactive molecules. A comparison of the ligand series and their molecular superimposition is a key problem in receptor mapping. A variety of methods have been described that realise such a strategy.
Molecular design and the development of new sweeteners
311
Visualisation, screening, clustering or modelling SAR data are computational techniques that are used. Quantitative SAR or a QSAR approach is a method that attempts to model equations from the SAR data. Three-dimensional (3D) and four-dimensional (4D) QSAR have supplemented traditional QSAR that was originally developed by Hansch in the early 1960s (Hansch and Leo, 1995). The 3D QSAR (Kubinyi et al., 1998) technique that originated in the 1980s (Crammer et al., 1988) analyses three-dimensional molecular descriptors, while 4D QSAR (Hopfinger et al., 1997) derives the descriptors from the investigation of molecules' conformational space sampled by the enormous number of conformers exploring different spatial region. Actually, it is the likelihood of a formation of the common 3D patterns of a series of molecules that is sought after by the molecular dynamics simulations. This increases the chances for mapping a proper pharmacophore. More recently 5D QSAR method has also been published (Vedani and Dobler, 2002). Theoretically, a QSAR equation should allow us to predict the activity of a newly designed compound that is to be synthesised. However, during QSAR modelling we operate on a strictly finite set of molecules for which activity data are measured and described a priori. Eventually, before modelling we must have chosen the appropriate data for the compounds that are active. It means that QSAR is more an a posteriori analysis of the SAR data structure than a strict method for the activity prediction of a novel compound design. Traditional QSAR has been successfully applied for the optimisation of a variety of congeneric series (Boyd, 1990) and for the discussion of further problematic issues compare (Wermuth, 2001). Although 3D QSAR methodology is based on the statistical methods (Partial Least Squares Analysis) controlled by so-called predictive statistics, in fact also in this case modelling behaviour evidently predominates. The chances for the precise prediction of the activity for a single virtual molecule that does not exist, e.g., it is to be synthesised, would be much lower than that calculated for the molecules of the original series used either for modelling or testing the QSAR equation (Polanski et al., 2004). After all even a minute chemical structure modification can result in substantial activity changes. This similarity paradox (Bajorath, 2002) decides that a virtual molecule, in reality, cannot only be a more or less substantial outlier in a QSAR equation but can appear completely inactive. More and more this problem is realised and we are replacing traditional modelling techniques with more robust versions or instead of modelling we are using other data handling methods such as, for example, clustering or visualisation. Chemists depended upon molecular visualisation since they realised the fact that molecules are 3D objects. Today computer graphics facilities provide a sophisticated visualisation tool. Visualisation allows us to avoid the paradox of producing an excellent model that completely fails when prediction is attempted. Eventually, it is better to have a vague idea of the trends than an illusion of a proper prediction. The application of the self-organising neural networks (SOM) for visualisation of the molecular electrostatic potential can be an illustration of its potential in molecular design (Polanski et al., 2002; Zupan and Gasteiger,
312
Optimising sweet taste in foods
1999). In this aspect, a variety of different methods that oscillate between modelling and visualisation are used in current molecular design. These are not discussed further here and the reader is referred to a number of monographs available (Wermuth, 2003). Statistically, the probability of finding a drug in a random pool of molecules is usually characterised by a number of 1 to 10,000. In fact, this number refers to socalled lead structures explored in traditional drug design (Kubinyi, 2004). In such research a small set of virtual molecules is targeted in the search for better properties. A completely different strategy has been selected by Nature to control life processes. An extremely large pool of random compounds processed in an extremely large time provided peerless bioeffectors, the masterpieces of evolution. Millions of years were, however, needed for natural selection. Combinatorial chemistry, a method for drug discovery developed in the 1980s, attempts to imitate this strategy by the substantial increase of the population of molecular objects investigated. Contemporary combinatorial chemistry is a technology that includes a variety of different methods providing a large pool of so-called combinatorial libraries of potential drugs (Wermuth, 2003) for relevant references. Traditional combinatorial approaches are based on completely random synthesis. Paradoxically, this does not increase the chances for finding a new drug, because we change the space of the compounds tested. Now it is a pool level of hundreds of thousands of molecular objects which is to be screened to find hits that form a potential lead structure (Kubinyi, 2004). In fact, the significant expansion of the compound space investigated in recent years did not result in the increase of new drug approvals (Frantz, 2004; Schmid and Smith, 2004). Actually, ligand based discovery is often an oversimplification. Basically, we need complex information on drug-receptor interaction to search for its new ligands. Novel computational and experimental chemistry methods should still be developed and/or improved to deal efficiently with these types of problems. Molecular dynamics allowing for drug-receptor docking is an important computational option. Novel process-driven approaches to combinatorial molecular discovery, namely, dynamic combinatorial chemistry (Huc and Lehn, 1997; Otto et al., 2002) or click chemistry (Kolb et al., 2001) are new experimental methods that allow us to combine the advantages of the screening of large molecule populations and rational usage of target receptor structure. In these approaches that attempt in vitro selection an isolated receptor controls synthesis direction, selecting among many possible reagents. This `increases the concentration of the strong binders at the expense of the weak binding molecules' (Otto et al., 2002). Clearly, with the emergence of combinatorial chemistry the power of large numbers in drug design is better realised. The importance of probability issues is also well understood in contemporary QSAR. This decides that the 4D QSAR scheme is much more probabilistic in nature, if compared to the 3D QSAR. Binary QSAR is another probabilistic method developed for investigations into high throughput discovery data (Labute, 2004; Stahura et al., 2002). Is there any molecular peculiarity that makes, in all likelihood, a predisposition for a drug
Molecular design and the development of new sweeteners
313
function? The so-called drug-likeness concept (Lipinski et al., 1997) or ADMET profiling (Van der Waterbeemd and Gifford, 2003) attempt to answer that question allowing one for the early filtering of random molecular sets in order to increase the chances of finding drug candidates within virtual compound libraries. Eventually, present-day drug design will be based on genomics. Pharmacogenomics is expected to revolutionise molecular design by originating the completely new principle that all gene data will be available as a starting point for new drug developments. Gene cloning and target protein expression will provide structural data for the discovery of novel lead inhibitors in experimental (in vitro) and computational (in silico) screening (Dean et al., 2001; Roses, 2004). Basically, an efficient technology together with a precise knowledge of the real or possible ligand-receptor interactions should make drug discovery a dead certainty. However, this is still not the case. Predicting molecular properties and modelling chemical or biological effects are still a lottery and we still need to be fortunate in scientific research to succeed in this field. Let us only cite here one of the latest stories. A blue viagra pill, a billion-dollar pharmaceutical of Pfizer had been thoroughly designed as a cardiac drug. However, under clinical trials patients were cured of erectile dysfunction, a completely different ailment, which had been discovered quite quickly because the general health conditions of cardiac patients were poor. This also points to another advantage in pharmaceutical research in comparison to sweetener chemistry. A compound originally designed for cardiac treatment had succeeded in another field. Of course, this makes pharmaceutical R&D more profitable.
13.5
Screening and visualising new sweeteners candidates
Sweet taste is a sensory property that contributes to the characteristics of the chemical substance. In fact, in early organic chemistry when the sophisticated tools of contemporary chemistry were not available, testing compound taste has been widely practised (Van der Heijden, 1997). This random screening based strategy has broadened the chemists' knowledge of sweet tasting chemicals and stimulated the interest in the origins of sweet taste. Astonishingly, sweetness is a property of very different chemical compound classes. As early as 1898 Sternberg realised that hydroxyl and amino functions were important for sweetness. The requirements for sweet taste induction have been deduced by a comparison of different sweet tasting compounds. Glucophore mapping gave rise to a number of different theories that appeared during the years, see (Van der Heijden, 1997; Van der Wel et al., 1987) for details. However, several decades were needed to succeed with the relatively general structural sweetness model. An early Shallenberger theory based sweetness on two-point concerted hydrogen AH-B bonding (Shallenberger and Acree, 1967) has been supplemented by Kier's AH,B,X model (Kier, 1972) that involves a hydrophobic interaction site. The discovery of hypersweeteners brought the
314
Optimising sweet taste in foods
multipoint attachment theory that includes as many as eight interaction sites, described as B, AH, XH, G1, G2, G3, G4 and D (Nofre and Tinti, 1996). Just a few years after the discovery of the sweetness of m-nitroaniline (Blanksma and van der Weyden, 1940) the compound has been extensively derivatised and investigated to describe their structure-taste regularities (Blanksma and Hoegen, 1946; Verkade and Witjens, 1946). Saccharin is the first synthetic sweetener originating from the chemical lab. It is also the first commercial sweetener that was derivatised intentionally in similar investigations. This revealed that a `substitution in the benzene ring of saccharin with the electron-withdrawing nitro group gives a bitter tasting substance. Substitution with and electron-donating group results in a sweet taste' (Hamor, 1961). Similar investigations have been performed for the majority of alternative sweeteners, see (Van der Heijden, 1997) for an excellent review. Modelling taste QSAR is another direction in glucophore mapping. An example is an early QSAR model relating relative sweetness to octanol-water partition coefficients (Deutsch and Hansch, 1966) for a series of 2-amino-4nitrobenzenes that were obtained earlier (Blanksma and Hoegen, 1946). This pointed to the importance of the hydrophobic bonding for sweet taste chemoreception. On the other hand, this model can be improved by the inclusion of the Hammett constant (Hansch, 1970). Another two models for the same series reported the relationships between the sweetness and STERIMOL parameters (Iwamura, 1980) and neural network derived molecular descriptors (Polanski et al., 1998). For a number of similar investigations into other sweet tastants see (Van der Heijden, 1997). However, how can we obtain sweet tasting molecules that are needed for such a study, except through a serendipitous discovery? Let us explain it using an example from our laboratory. In 1967 during the investigations of the substantial effects of carboanion chemistry Aleksander Ratajczak synthesised 2,2-dimethyl1-phenylsulfonylalkanoic acid, ASA1 (Fig. 13.2), in Donald Cram's laboratory at UCLA. This compound provided an interesting model compound that could have been relatively easily obtained and the molecular symmetry implies a simplification of the compound stereochemistry (Ratajczak et al., 1967). During synthetic operations the sweet taste of this compound was discovered serendipitously, and this is claimed in the US patent (Cram and Ratajczak, 1969). Apparently, no further researches have been performed in this field for almost two decades. In the late 1980s the compound has been intentionally modified to obtain several dozen acyclic analogues, arylsulfonylalkanoic acids, ASA2 (Polanski and Ratajczak, 1993). More recently, a similar study provided a number of novel compounds, ASA3±ASA5 (Lysiak et al., 2005). What were the bases for the design of the above-mentioned analogues used in these two studies? The main hypothesis for the first study is that the cyclic structure is not essential for the compounds' activity. Thus, a characteristic 2,2-dimethyl substitution is replicated in the series of congeneric but acyclic structures. Actually, it appeared that this motif retains compound sweetness. This allowed us to model several QSARs for active compounds. Optimisation of the activity
Molecular design and the development of new sweeteners
Fig. 13.2
315
Drug design concept for ASA sweeteners derived from the parent ASA1. ASA2:. R= H or Me; X= alkyl, halogen, alkoxy.
according to the Topliss method (Topliss, 1977) gave the ASA compound of the maximal sweetness (Polanski and Ratajczak, 1993). We also attempted to model 3D QSAR and explain the molecular basis for the series using the compound Xray structures and the molecular descriptors derived from the processing of the molecular data by neural networks (Polanski et al., 1997). The importance for methyl substitution was tested in the second study. Actually, 2,4-dimethyl-1phenylsulfonylcyklohexanecarboxylic acid (ASA3) appeared to be the only active compound among the designed compounds that were obtained. This means that a double methyl substitution is needed in the cyclohexane ring to elicit sweet taste in these compounds. However, quite surprisingly, a majority of the synthetic precursors of ASA3, i.e., 2-methyl-1-cyklohexenecarboxylic acids ASA4 are potent sweet tastants, including the single -methyl group substituted analogue ASA5. Although such investigations might seem relatively outmoded, in the current practice the analysis of the SAR by screening potential receptor ligands designed on more or less intuitive similarity schemes, 2D or 3D substructure searches is still an important design strategy even in these cases when we have 3D receptor data. Paradoxically, the so-called fragment approach that insists on the importance of certain two-dimensional molecular frameworks for the drug-likeness and lead generation has just appeared. It has been found that a `group of 32 common shapes or frameworks accounted for 50% of the 5,120 molecules considered. Whether these fragments had intrinsic characteristics that gave them drug-like properties or their presence was a result of chemists' habits, familiarities or synthetic versatility was an issue that was recognized but not addressed' (Fattori, 2004). Probably all these reasons are important. Structural similarity increases the chances for the activity and chemical intuition replicates building blocks. This makes common archetypes, important building blocks deciding compound activity. Because organic synthesis is another bottleneck during drug development we cannot concentrate only on molecular design. This explains that we still often use the basically intuitive fragment approach in drug or lead generation because even if we know the receptor structure it is not easy to find potential synthetic targets and there is no guarantee that such compounds will appear active. This also explains the strategy for our investigations.
316
Optimising sweet taste in foods
We cannot cite here all the molecular design approaches in sweet taste chemistry. Only a few follow. Walters successfully applied a 3D technique of receptor surface modelling for glucophore mapping (Walters, 1995). In an interesting approach a series of isovanillyl and guanidine sweeteners were used to model the sweet taste pseudoreceptor. In such an approach a series of ligands is used for mimicking the unknown receptor structure by a series of model amino acids (Cohen, 1996). This could also be interpreted as a posteriori QSAR model. A publication title From molecules to receptor is the best description of this method (Bassoli et al., 2002). Finally, just recently the genes encoding sweet taste chemoreception have been located on the chromosome. The sweet receptor is a member of the GPCR family and is expressed on the surface of certain taste bud cells. Three genes code the T1R family of taste receptors (Hoon et al., 1999; Kitagawa et al., 2001; Max et al., 2001; Montmayeur et al., 2001; Sainz et al., 2001). The investigations of the preferences for the different sweeteners have been performed using the knocked animal models, e.g. genetic and physical mapping limited a critical genomic interval coding saccharine preference (Sac) to a 194 kb DNA fragment. This region was sequenced, which identified a gene (Tas1r3) encoding the third member of the T1R family of putative taste receptors, T1R3 (Bachmanov et al., 2001). Although the structure±function relationships of T1R receptors still remain largely unknown, it has been found that T1R2 is required for the interactions with aspartame and neotame (Xu et al., 2004). Lactisole inhibits the activity of the human T1R1/T1R3 receptor, and cyclamate does not activate the T1R1/T1R3 receptor (Xu et al., 2004). Thus, the investigations into sweet taste receptor allowed the identification of different functional roles of T1R3 and T1R2. This also proved the presence of multiple ligand binding sites on the sweet taste receptor. Recently, a 3D structure of the T1R3 receptor has been modelled by homology modelling (Walters, 2002). Definitely, structure-based design using sweet taste receptor data would significantly increase the potential for the development of novel sweeteners. However, so far these approaches concentrate on the explanation of the interactions of the different but known sweeteners. Antibody-based fluorescence polarisation assay to screen combinatorial libraries for sweet taste compounds has been developed recently for combinatorial and high throughput technology (Linthicum et al., 2001). Just recently the Senomyx reported SweetScreenHT, a sweet taste receptor-based assay system for highthroughput technology has been reported. `This assay system responded to many different sweet-tasting compounds including carbohydrate sweeteners and high potency artificial sweeteners and, therefore, could be used to identify novel sweet flavors and flavor enhancers' (Senomyx, 2004). Neural networks were used for screening virtual compound libraries of potential sweetener candidates (Polanski et al., 2000). So far there are no reports on combinatorial syntheses in the search for novel sweeteners; however; such approaches will definitely appear also in sweetener discovery in the future.
Molecular design and the development of new sweeteners
13.6
317
Molecular design in commercial sweetener development
What is a fundamental requirement that is needed for the development of a potential commercial sweetener? Paradoxically, the answer is in the intellectual property protection that is critical to the success in the present-day practice of medicinal chemistry. Contemporary pharmaceutical industry depends on patents that provide reward for the inventor who carried out the investigations and discloses the `invention in such a way that an expert could follow it without doing research' (Souleau, 2003). Of course, this is a formal prerequisite only, and we also need a sweet compound of the perfect taste quality, chemical stability and toxicological safety. Moreover, it should be possibly a top-earning drug, which means chemical synthesis should be relatively cheap. For a discussion of the synthetic problems in sweetener chemistry see Ager et al. (1998). Can we control of all these issues during sweetener development? The answer is no. Thus, we need a large pool of compounds to select a proper candidate. Let us illustrate this problem by the discovery of neotame. It was suggested that the sweet receptor may contain an additional hydrophobic binding site (Van der Heijden et al., 1978; Nofre and Tinti, 2000). Thus, it has been attempted to obtain sweeteners that could bind into such a site and several novel sweeteners have been discovered and claimed in several patents by (Nofre and Tinti, 2000). A modification of aspartame to include a hydrophobic moiety capable of binding this site (Fig. 13.3) provided a novel sweetener that has just been approved by the FDA for commercial applications. Neotame is a close aspartame congener. It has a clean sweet taste relatively close to sucrose. Higher stability in the neutral pH range is an important advantage in comparison to the parent aspartame. After consumption in humans only an insignificant release of methanol and phenylalanine occurs (Nofre and Tinti, 2000). This is important due to phenylketonuric human hazards (Gilman, 2004). Finally, high sweetness potency (neotame is 11,000 times sweeter than sucrose on a molar basis, and aspartame is only 170 times sweeter than sucrose) makes the production of this compound highly effective.
Fig. 13.3 The idea for the aspartame modification by the introduction of the second hydrophobic function, modified from (Nofre and Tinti, 2000).
318
13.7
Optimising sweet taste in foods
From discovery to commercial products
Saccharin is a first alternative sweetener developed into a commercial product. Its discovery and commercialisation gave rise to a controversy over fundamental issues in scientific research. Although the synthesis had been published jointly by Fahlberg and Remsen (1879) it is `Fahlberg who patented saccharin without including Remsen on the patent. Fahlberg went on to become wealthy. Remsen went on to become the president of Johns Hopkins University, and he spoke of Fahlberg as a scoundrel. It nauseates me to hear my name mentioned in the same breath with him' (Walters, 2001). Fahlberg who apparently consumed 10 g of the substance, noticing no adverse side effects, tested saccharin toxicology (Uher and WoÂjtowicz, 2003). Saccharin was developed as a cheap surrogate for the luxury substance sugar. It was much cheaper than sugar (about half the price in 1900) which decided its popularity in the German countryside. Saccharin consumption was 0.15 tonnes (or 83 tonnes as sugar equivalent unit) in 1888 and reached 118.9 tonnes (65,392 as sugar equivalent unit) as early as 1900 (Ruprecht, 2001). At that time sugar was still a political issue and due to sugar taxation saccharin development appeared to arouse public interest. Also, the sugar industry was disturbed. Apparently, this decided that saccharin should be blamed for digestive problems and it was banned in Germany, Spain, Portugal and the USA. Although the Comite de Hygiene in Paris, who had originally questioned saccharin consumption, revised their original reservations quite promptly, it took saccharin 15 years to reappear in the German market. Saccharin was used as sugar ersatz in times of sugar shortages, in particular during wartime. The adverse taste profile (bitter aftertaste) contributed to the low status of the compound (Ruprecht, 2001). Despite that, today a number of saccharin-sweetened products are available on the market. This includes tabletop sweeteners, baked goods, jams, chewing gum, canned fruit, candy, dessert toppings, salad dressings, cosmetic products, vitamins and pharmaceuticals. Synergistic mixtures with other sweeteners (cyclamate, aspartame, acesulfame) improved the saccharin taste profile. This got special interest and several blended tabletop sweeteners have been brought into the market (Saccharin, 2004). Originally, cyclamate was relatively expensive because it is only 30 times sweeter than sucrose. After discovery Dupont held the patent which was, however, later sold to Abbott. In 1950 Abbott applied for the approval of the compound which was planned to be used for masking bitter taste of some pharmaceuticals, especially antibiotics. In 1958 cyclamate was classified as a GRAS (Generally Recognised as Safe) substance and recommended as a tabletop sweetener for diabetics (Cyclamate, 1999). After controversial research in 1970 indicating cancer cases resulted in the lab animals treated with the sweetener it is still currently banned in the USA. However, it is approved in Europe and 50 other non-European countries (Spillane, 2004) and a re-approval application for the USA is still under the examination of FDA. Other important usage developed is low-calorie beverages, processed fruits, chewing gum, salad dressings, gelatin desserts, candies and bakery products (Cyclamate, 1999).
Molecular design and the development of new sweeteners
319
A chemical structure of aspartame makes this compound of particular interest for marketing because it is made up of two amino acids, natural food ingredients. Similar to a variety of common food that includes proteins (polypeptides), e.g. meat, it is metabolised in the body to the original amino acids, aspartic acid and phenylalanine. On the other hand, an aspartame molecule contains a methyl ester function hydrolysed to methanol that is further metabolised to formaldehyde. This led to toxicological concerns about the compound's safety, which are still under discussion. Aspartame toxicology has been thoroughly studied. In 1981 the FDA approved aspartame for use as a tabletop sweetener. The sweetening of various food and dry beverage mixes has also been approved. The 1983 FDA approval included carbonated beverages and a general use approval for food and beverages dates back to 1996 (Low calorie sweeteners, 2004). The pure sweet taste of aspartame, its chemical constitution related to natural products and a high sweetness value are factors that contributed much to the development of the modern market concept of alternative sweetener. It is also the recent trends of healthy food, low-sugar diet and healthy lifestyle that made this sweetener an attractive and widespread product. Aspartame marketing by Nutrasweet is based not on economical motivation (sugar ersatz) or even not a sugar substitution for disease conditions (diabetic patients). Aspartame was probably the first alternative sweetener attempting to provide a sugar replacement because of lifestyle advantages. It has gained better status than just a sugar ersatz and won relatively widespread consumer acceptance. In the contemporary `Coca-Cola society' it spreads throughout the whole world literally with Coca-Cola bottles that are sweetened with aspartame. On the other hand, aspartame-sweetened beverages can be easily distinguished from these blended with sugar with a clear sugar preference. Acesulfame discovered in Hoechst AG has been approved in 1983 in Great Britain. The US approval dated to 1988 and 1998 (in soft drinks). Nutrinova now sells acesulfame-K, a potassium sweetener salt, under the brand name Sunet (Nutrinova, 2004). Blends with other sweeteners are available on the market and are developed to imitate the sucrose-like sweetness profile, which is marketed as Sunnet Nutrinova Multi-Sweetener Concept, e.g. the Sucralose-Sunnet mixtures have been claimed in US 4 495 170, EP 64 361 patents. Various applications include tabletop tablets and the sweetening of beverages, confectionery, dietary products, baked foods and pharmaceuticals (Nutrinova, 2004). Neotame is a novel alternative sweetener composed of amino acids (Witt, 1999). Relative sweetness is approximately 30±40 times higher than aspartame; 7,000±13,000 times sweeter than sugar. Neotame's taste profile is comparable to sucrose. Neotame, a close aspartame congener is readily metabolised via natural biological processes. A clean, sweet taste is allegedly accompanied by unique flavour enhancement properties (Low calorie sweeteners, 2004). In 2002 the FDA approved neotame as a general-purpose sweetener. Neotame is also approved for use in Australia and New Zealand. Indicated applications are similar to those given for other alternative sweeteners and the usage in cooking and baking is especially underlined (Low calorie sweeteners, 2004).
320
Optimising sweet taste in foods
Alitame, discovered in Pfizer and given a brand name AclameTM, is another aspartame-like sweetener. It is constituted of the amino acids, L-aspartic acid and D-alanine. The compound awaits the FDA approval now and it is currently approved in Australia, New Zealand, Mexico and China (Low calorie sweeteners, 2004). Some regulatory issues concerning low-calorie sweeteners are discussed in (Lindley, 1999). The comparison of discovery and marketing timelines indicates a progressive delay between sweetener discovery and commercialisation (Bassoli, 2004). This illustrates a fact of increasing quality and safety requirements for food additives.
13.8
Future trends
Although the long-awaited break-through in molecular design is still in the future, the researches in this field are more and more important in contemporary pharmaceutical sciences. `The food industry could do the same [molecular design] type and quality of research as the pharmaceutical industry. They could synthesize new compounds to explore structure±activity relationships; they could isolate and characterize receptors and use them in sophisticated food additive screening assays; they could exploit the power of computational chemistry to postulate models for desired taste or functional properties. There is almost no activity at this level of basic research' (Mazur, 1990). At least in part, this last decade, Mazur's opinion on sweetener prospects is still relevant. Although Senomyx who invested in sweet receptor investigations provided an important breakthrough in sweet receptor chemistry (Senomyx, 2004), sweetener research has still to be developed to keep pace with pharmaceutical level projects. Present-day molecular design practised in pharmaceutical science shows the direction for further investigations in sweetener chemistry. Definitely, structure-based design using sweet taste receptor data will bring an important dimension to this. Pharmacogenomic, combinatorial chemistry and dynamic combinatorial approaches will probably appear also in sweetener design. However, the main issue for sweetener discovery and development prospects is whether the food industry will invest in the R&D sector to boost the researches and improve design performance. Current terms in the pharmaceutical industry decide that profits crucially depend on newly developed compounds. This makes companies anxious to invest in a single drug brought into the market for US$500±700 million (Kubinyi, 2004) or even as much as US$1.8 billion according to other data (Gilbert et al., 2003). Of course these numbers are of a statistical nature and also include the costs of a number of molecules designed and/or synthesised and then given up at the different stages of researches. Do we need a novel, highly expensive original sweetener molecule? Can the food industry pay comparable or even larger money for an original novel alternative sweetener? Without a doubt, economic factors will be a critical issue that appoints a time for a novel sweetener generation for the market. Eventually, in
Molecular design and the development of new sweeteners
321
pharma it is also industry rather than academia that decide market development. Senomyx's experience in taste investigations clearly illustrates that investments in sweet researches can be highly successful (FuÈlle, 2002). So far pharmaceutical research still seems to be more prosperous for academia than the sweet taste one. This can be easily illustrated by the impact factor (IF) of the respective journals, which illustrates an impact of the publications on the scientific audience. The IF of the Journal of Medicinal Chemistry that deals with pharmaceuticals is 4.820, while the IF of Food Chemistry and Journal of Agricultural and Food Chemistry are 1.204 and 2.102, respectively. Although the IF data are in some sense controversial the trend is obvious. Of course, this can be translated directly into the prospects for the financial support of research projects by science foundations. This also means that in the near future a large academic sweetener project is rather uncertain. On the contrary, smaller projects, especially devoted to synergistic taste improvements, seems to be attractive and to offer much cheaper alternative fields of research. Sweetness no longer has an adaptive function in modern society; however, there is no sign of people giving up their love of sweetness. This indicates a bright future for the alternative sweetener market. From the scientific point of view the sucrose-like non-nutritive sweetener has still to be developed, which in turn brightens the future for alternative sweetener researches.
13.9
Sources of further information and advice
The practice of medicinal chemistry is an excellent handbook providing fundamental knowledge on the current art of molecular design including economical and intellectual property issues (Wermuth, 2003). Guidebook on molecular modeling in drug design offers more detailed insight into theoretical issues (Cohen, 1996). On-line (ACS, 2000) reviews and lectures by Hugo Kubinyi give an excellent Web-based introduction to medicinal chemistry (Kubinyi, 2004). Interesting reviews on the investigations in sweetener development can be found in proceedings from several taste chemistry symposia (Walters et al., 1990; Matlouthi et al., 1993; Special Topic Issue on the Science of Sweeteners, 2002). Also, several books are available in this field, e.g. Shallenberger (1993). The updated critical reviews on the commercial sweeteners are available in the special issue of Agro Food Industry Hi-Tech (Bassoli, 2004; Spillane, 2004). In the same issue the reader can find an excellent review on the taste quality problems (Kilcast, 2004). Eric Walter's homepage offers a brief introduction to artificial sweeteners including discovery, chemistry and links to other pages (Walters, 2001). Commercial sweetener data can be found in the Web (Cyclamate, 1999; Neotame, 2002; Nutrinova, 2004; Saccharin, 2004). Finally, Senomyx's Web page offers the updated information on the current state of the sweet receptor investigations and pharmacogenomic approaches to sweetener development (Senomyx, 2004).
322
Optimising sweet taste in foods
13.10
References
(2000), `The Pharmaceutical Century, ten decades of drug discovery', http:// pubs.acs.org/journals/pharmcent, visited 3 January 2005. AGER D J, PANTALEONE D P, HENDERSON S A, KATRITZKY A R, PRAKASH I and WALTERS D E (1998), `Commercial, Synthetic Nonnutritive Sweeteners', Angew. Chem. Int. Ed., 37, 1802±17. ACS
BACHMANOV A A, LI X, REED D R, OHMEN J D, LI S., CHEN Z, TORDOFF M G, DE JONG P J, WU C,
and BEAUCHAMP G K (2001), `Positional Cloning of the Mouse Saccharin Preference (Sac) Locus', Chem. Senses, 26, 925±33. BAJORATH J (2002), `Virtual screening in drug discovery: Methods, expectations and reality', Current Drug Discovery, March, 24±28. BASSOLI A (2004), ```Chemistry-Nature'', still an open match for the discovery of new intensive sweeteners', Agro Food Ind. Hi Tec, 15, 27±29. BASSOLI A, DREW M G, MERLINI L and MORINI G (2002), `General pseudoreceptor model for sweet compounds: a semiquantitative prediction of binding affinity for sweettasting molecules', J. Med. Chem., 45, 4402±9. BIRCH L L (1999), `Development of Food Preferences', Annual Review of Nutrition, 19, 41±62. BLANKSMA J J and HOEGEN D (1946), `Sweet taste of 4-nitro-2-aminotoluene, 4-nitro-2aminobenzoic acid and 2-nitro-4-aminobenzoic acid', Recl. Trav. Chim. Pays-Bas, 65, 333±7. BLANKSMA J J and VAN DER WEYDEN P W M (1940), `Relationship between taste and structure in some derivatives of meta-nitroaniline', Recl. Trav. Chim. Pays-Bas, 59, 629±32. BOYD D B (1990), `Successes of Computer-Assisted Molecular Design', in Lipkowitz K B and Boyd D B, Reviews in Computational Chemistry, New York, VCH Publishers, 355±71. COHEN, N C (1996), Guidebook on molecular modeling in drug design, San Diego, Academic Press. CRAM D J and RATAJCZAK A (1969) US Patent 3 598 868. CRAMMER III R D, PATTERSON D E and BUNCE J D (1988), `Comparative Molecular Field Analysis (CoMFA). 1. Effect of shape on binding of steroids to carrier proteins', J. Am. Chem. Soc., 110, 5959±67. CYCLAMATE (1999), www.cyclamate.com, visited 3 January 2005. DEAN P M, ZANDERS E D and BAILEY D S (2001), `Industrial scale genomics-based drug discovery', Trends in Biotechn, 19, 288±92. DEUTSCH E W and HANSCH C (1966), `Dependence of relative sweetness on hydrophobic bonding', Nature, 211, 75. È ber die Oxydation des Orthotoluolsulfamids', Chem. FAHLBERG C and REMSEN I (1879), `U Ber., 12, 469±73. FATTORI, D D (2004), `Molecular recognition: the fragment approach in lead generation', Drug Discov. Today, 9, 229±38. FRANTZ S (2004), `2003 approvals: a year of innovation and upward trends', Nature Reviews Drug Discovery, 3, 103±5. È LLE H J (2002), `A Case Study on the Commercialization of Sensory Biotechnology by FU Senomyx., Inc.', Technology Commercialization Alliance, available from: http:// www.usc.edu/org/techalliance/Anthology.htm, visited 3 January 2005. GILBERT J, HENSKE P and SINGH A (2003), `Rebuilding big pharma's business model, IN WEST D B, CHATTERJEE A, ROSS D A
Molecular design and the development of new sweeteners
323
VIVO: The Business & Medicine Report, 11/01/2003, Windhover Information Inc., Vol 21(10), p. 73, available from: http://sis.windhover.com/windbuy/lpext.dll/ windbuy/iv/2003/2003800191.htm?f=templates&fn=document-frame.htm&q= %5Bfield%20author%3Agilbert%5D&x=Advanced&2.0#LPHit1, visited 3 January 2005. GILMAN V (2004), `What's that stuff? Artificial sweeteners', C&EN, 43, available from: http://pubs.acs.org/cen/whatstuff/stuff/8225sweeteners.html, visited 24 February 2005. HAMOR G H (1961), `Correlation of chemical structure and taste in saccharin series', Science, 134, 1416±17. HANSCH C (1970), `The use of in structure-activity correlations', J. Med. Chem., 13, 964±6. HANSCH C and LEO A (1995), Exploring QSAR Fundamentals and Applications in Chemistry and Biology, Washington DC, ACS. HOON M A, ADLER E, LINDEMEIER J, BATTEY J F, RYBA N J and ZUKER C S (1999), `Putative mammalian taste receptors: a class of taste-specific GPCRs with distinct topographic selectivity', Cell, 96, 541±51. HOPFINGER A J, WANG S, TOKARSKI J S, JIN B, ALBUQUERQUE M, MADHAV P J and DURAISWAMI C (1997), `Construction of 3D QSAR models using the 4D QSAR analysis formalism', J. Am. Chem. Soc., 119, 10509±24. HOUGH L (1989), `Sucrose, sweetness and sucralose', Int. Sugar JNL, 91, 23±29. HUC I and LEHN J-M (1997), `Virtual combinatorial libraries: Dynamic generation of molecular and supramolecular diversity by self-assembly', Proc. Natl. Acad. Sci. USA, 94, 2106±10. INGLETT G E (1976), `A History of sweeteners ± natural and synthetic', J. Toxicol. & Environ. Health, 2, 207±14. IWAMURA H (1980), `Structure-taste relationship of perillartine and nitro- and cyanoanilline derivatives', J. Med. Chem., 23, 306±12. KIER L B (1972), `A molecular theory of sweet taste', J .Pharm. Sci., 61, 1394±7. KILCAST D (2004), `Satisfying consumer demands for sweetness quality', Agro Food Ind. Hi Tec, 15, 36±38. KITAGAWA M, KUSAKABE Y, MIURA H, NINOMIYA Y and HINO A (2001), `Molecular genetic identification of a candidate receptor gene for sweet taste', Biochem. Biophys. Res. Commun., 283, 236±42. KOLB H C, FINN M G and SHARPLESS K B (2001), `Click chemistry: diverse chemical function from a few good reactions', Angew. Chem., Int. Ed. Engl., 40, 2004±21. KUBINYI H (2004), `Hugo Kubinyi, Lectures', http://www.kubinyi.de/lectures.html, visited 7 January 2005. KUBINYI H, FOLKERS G and MARTIN Y C (Eds.) (1998), 3D QSAR In Drug Design, Kluwer/ ESCOM, Dordrecht. LABUTE P (2004), `Binary QSAR: A New Technology for HTS and UHTS Data Analysis', available from: http://www.chemcomp.com/Journal_of_CCG/Features/ htsbqsar.htm, visited 3 January 2005. LINDLEY M G (1999), `New developments in low-calorie sweeteners', World Rev. Nutr. Diet., 85, 44±51. LINTHICUM D S, PATEL J and CAIRNS N (2001), `Antibody-based fluorescence polarization assay to screen combinatorial libraries for sweet taste compounds', Comb. Chem. High Throughput Screen., 4, 431±8. LIPINSKI C A, LOMBARDO F, DOMINY B W and FEENEY P J (1997), `Experimental and
324
Optimising sweet taste in foods
computational approaches to estimate solubility and permeability in drug discovery and development settings', Adv. Drug Deliv. Rev., 23, 3±25. LOW CALORIE SWEETENERS (2004), http://www.caloriecontrol.org, visited 3 January 2005. LYSIAK V, RATAJCZAK A, MENCEL A, JARZEMBEK K and POLANSKI J (2005), `A structure-taste study of arylsulfonyl(cyclo)alkanecarboxylic acids', Bioorg. Med. Chem., 13, 671±5. MARTIN Y C (1997), `Pharmacophore mapping', in Martin Y C, Willett P, Designing Bioactive Molecules Three-Dimensional Techniques and Applications, Washington, DC, ACS, 121±48. MATLOUTHI M, KANTERS J A and BIRCH G G (1993), Sweet Taste Chemoreception, London, Elsevier. MAX M, SHANKER Y G, HUANG L, RONG M, LIU Z, CAMPAGNE F, WEINSTEIN H, DAMAK S and MARGOLSKEE R F (2001), ' Tas1r3, encoding a new candidate taste receptor, is allelic to the sweet responsiveness locus Sac', Nat. Genet., 28, 58±63. MAZUR R H (1990), `The future of synthetic sweeteners', in Walters D E, Orthoefer F T, DuBois G E, Sweeteners Discovery, molecular design, and chemoreception, Washington, ACS, 313±19. MONTMAYEUR J P, LIBERLES S D, MATSUNAMI H and BUCK L B (2001), `A candidate taste receptor gene near a sweet taste locus', Nat. Neurosci., 4, 492±8. NEOTAME (2002), www.neotame.com, visited 3 January 2005. NOFRE C and TINTI J M (1996), `Sweetness reception in man: the multipoint attachment theory', Food Chem., 1996, 56, 263±74. NOFRE C and TINTI J-M (2000), `Neotame: discovery, property, utility', Food Chem., 69 245±57. NUTRINOVA (2004), www.nutrinowa.com, visited 3 January 2005. OTTO S, FURLAN R L and SANDERS J K (2002), `Dynamic combinatorial chemistry', Drug Discov. Today, 7, 117±25. POLANSKI J and RATAJCZAK A (1993), `A structure-taste study of a new class of artificial sweeteners arylsulphonylalkanoic acids', in Matlouthi M, Kanters J A and Birch G G, Sweet Taste Chemoreception, London, Elsevier, 185±203. POLANSKI J, RATAJCZAK A, GASTEIGER J, GALDECKI Z and GALDECKA E (1997), `Molecular modeling and X-ray analysis for the structure-taste study of arylsulfonylalkanoic acids', J. Mol. Struct., 407, 71±80. POLANSKI J, GASTEIGER J, WAGENER M and SADOWSKI J (1998), `The comparison of molecular surfaces by neural networks and its application to quantitative structure activity studies', Quant. Struct. Act. Relat., 17, 27±36. POLANSKI, J. GASTEIGER, J and JARZEMBEK, K (2000), `Self-organizing neural networks for screening and development of novel artificial sweetener candidates, Comb. Chem. & HTS, 3, 481±95. POLANSKI J, ZOUHIRI F, JEANSON L, DESMAEÈLE D, D'ANGELO J, MOUSCADET J-F, GIELECIAK R,
GASTEIGER J and LE BRET, M (2002), `Use of the Kohonen neural network for rapid screening of ex vivo anti-HIV activity of styrylquinolines, J. Med. Chem., 45 4647±54. POLANSKI J, GIELECIAK R and BAK A (2004), `Probability issues in molecular design: predictive and modeling ability in 3D-QSAR schemes', Comb Chem & HTS, 7, 793±807. RATAJCZAK A, ANET F A L and CRAM D J (1967), `Electrophilic substitution at saturated carbon. XXXII. Spectroscopic observation of enantiomer interconversion of 2,2dimethyl-1-phenylsulfonylcyclopropyl anions', J. Am. Chem. Soc., 89, 2072±7. ROSES A D (2004), `Pharmacogenomics and drug development: the path to safer and more
Molecular design and the development of new sweeteners
325
effective drugs', Nature Reviews Drug Discovery, 5, 645±56. (1989), `The role of learning in the acquisition of food preferences by humans', in Shepherd R, Handbook of the Psychophysiology of Human Eating, London, Wiley, 205±27. RUPRECHT W (2001), Consumption of sweeteners: an evolutionary analysis of historical development, available from http://www.business.auc.dk/druid/conferences/nw/ paper1/ruprecht.pdf, visited 3 January 2005. SACCHARIN (2004), www.saccharin.com, visited 3 January 2005. SAINZ E, KORLEY J N, BATTEY J F and SULLIVAN S L (2001), `Identification of a novel member of the T1R family of putative taste receptors', J. Neurochem., 77, 896± 903. SARDESAI V M and WALDSHAN T H (1991), `Natural and synthetic intense sweeteners', J. Nutr. Biochem., 2, 236±44. SCHMID E F and SMITH D (2004), `Is pharmaceutical R&D just a game of chance or can strategy make a difference?', Drug Discov. Today, 9, 18±26. SENOMYX (2004), www.senomyx.com, visited 3 January 2005. SHALLENBERGER R S (1993), Taste Chemistry, Glasgow, Blackie Academic and Professional. SHALLENBERGER R S and ACREE T E (1967), `Molecular theory of sweet taste', Nature, 216, 480±2. SOULEAU M (2003), `Legal aspects of product protection what a medicinal chemist should know about patent protection' in Wermuth C G The practice of medicinal chemistry, Amsterdam, Academic Press, 707±22. SPECIAL TOPIC ISSUE ON THE SCIENCE OF SWEETENERS (2002), Pure Appl. Chem., 74, 1101± 316. SPILLANE W J (2004), `Cyclamates Sweeteners that are here to stay', Agro Food Ind. Hi Tec., 15, 30±1. STAHURA F L, GODDEN J W and BAJORATH J (2002), `Differential Shannon Entropy Analysis Identifies Molecular Property Descriptors that Predict Aqueous Solubility of Synthetic Compounds with High Accuracy in Binary QSAR Calculations', J. Chem. Inf. Comput. Sci., 42, 550±8. TOPLISS J G (1977), `A manual method for applying the Hansch approach to drug design', J. Med. Chem. 20, 463±9. Â JTOWICZ H (2003), `Natural and synthetic non-saccharide sweeteners', UHER M and WO Wiad. Chem., 57, 505±46. VAN DER HEIJDEN A (1997), `Historical overview an structure-activity relationships among sweeteners', Pure & Appl. Chem., 69, 667±74. VAN DER HEIJDEN A, BRUSSEL L B P and PEER H G (1978), `Chemoreception of sweet tasting dipeptide esters: a third binding site', Food Chem. 3, 207±11. VAN DER WATERBEEMD H, CARTER R E, GRASSY G, KUBINYI H, MARTIN Y C, TUTE M S and WILLETT P (1997), `Glossary of Terms Used in Computational Drug Design (IUPAC Recommendations 1997)', Pure Appl. Chem. 69, 1137±52. VAN DER WATERBEEMD H and GIFFORD E (2003), `ADMET in silico modelling: towards prediction paradise?', Nat. Rev. Drug Disc., 2, 192±204. VAN DER WEL H, VAN DER HEIJDEN A and PEER H G (1987), `Sweeteners', Food Review International, 3, 193±268. VEDANI A and DOBLER M (2002), `5D-QSAR: the key for simulating induced fit', J Med Chem., 45, 2139±49. VERKADE P E and WITJENS P H (1946), `Researches on the alkoxy-aminonitrobenzenes. II. ROZIN P
326
Optimising sweet taste in foods
Synthesis and taste of the ten possible ethoxy and n-propoxy compounds, Recl. Trav. Chim. Pays-Bas, 65, 361±79. WALTERS D E (1995), `When There is No Receptor Crystal Structure: Building Useful Models of Receptor Sites', available from: NetSci's Science Center: Computational Chemistry: http://www.netsci.org/Science/Compchem/feature03.html, visited 3 January 2005. WALTERS D E (2001), `Everything you ever wanted to know about sweeteners', http:// www.rosalindfranklin.edu/cms/biochem/walters/sweet/sweethome.html, visited 3 January 2005. WALTERS D E (2002), `Homology-based model of the extracellular domain of the taste receptor T1R3', Pure Appl. Chem., 74, 1117±23. WALTERS D E, ORTHOEFER F T and DUBOIS G E (1990) Sweeteners Discovery, molecular design, and chemoreception, Washington, DC, ACS. WERMUTH C (2001), `The impact of QSAR and CADD methods in drug discovery, rational approaches to drug design', in Holtje H D and Sippl W, Rational approaches to drug design ± Proceedings of the 13th European Symposium on Quantitative Structure-Activity Relationships, Barcelona, Prous Science, 3±20. WERMUTH C G (2003), The practice of medicinal chemistry, Amsterdam, Academic Press. WITT J (1999), `Discovery and development of Neotame', World Rev. Nutr. Diet., 85, 52±7. XU H, STASZEWSKI L, TANG H, ADLER E, ZOLLER M and LI X (2004), `Different functional roles of T1R subunits in the heteromeric taste receptors', Proc. Natl. Acad. Sci. USA, 101, 14258±63. ZUPAN J and GASTEIGER J (1999), Neural Networks and Drug design for Chemists, Weinheim, VCH.
14 Developing new sweeteners from natural compounds A. Bassoli, DISMA, University of Milan, Italy
14.1
Introduction
To design a new sweetener a good lead compound has to be used. Both synthetic and natural active compounds have been used in the past and are still used as leads to generate a large class of active compounds. Section 14.2 describes the advantages (e.g., easy screening, available sources, safety prerequisites) and limitations (e.g., complicated structures, stereochemical requirements, instability) of natural compounds as leads for synthesis and structure±activity relationship studies. In Section 14.3 some methods are discussed for designing new products by modification of leads in order to obtain the maximum information from SAR and receptor modelling. The use of topological models, the identification of glucophores, methods to study the active conformation and to understand the role of stereochemistry in activity are discussed. Section 14.4 reports some practical examples of the development of analogues of new natural compounds, such as monatin. The choice of synthetic methods, the problems related to stereochemistry, the use of modelling to address the synthesis of new derivatives is discussed. Current advances and future trends in developing new analogues of natural compounds are discussed in Section 14.5.
14.2 Importance of developing new sweeteners from natural compounds The search for new sweet compounds in Nature has always been very active. Many natural sweet compounds are known, principally from the plant kingdom,
328
Optimising sweet taste in foods
belonging to many different chemical classes: amino acids, peptides and proteins; sugars and sugar derivatives; terpenes, steroids, polyketides and many other secondary metabolites have been reported to be sweet. Even some inorganic compounds such as some lead and beryllium salts are known to be sweet. Some natural sweet organic compounds are shown in Fig. 14.1. Most of them have been known for a long time since the habit to look for good tasting, therefore sweet plants, is innate in animals and humans. Also in recent years some new compounds have been isolated and reported to be sweet from ethnobotanical research. Eating is itself a `screening practice' that over the centuries has allowed mankind to taste millions of compounds and to select positively sweet components. In Nature the sweet taste is generally associated with edible substances; therefore for the chemist the choice of a natural compound as a lead is generally a good starting point with regard to safety prerequisites. In fact, in research for new taste active compounds it is common practice to submit new compounds to preliminary sensory evaluation using a restricted panel, with no or only a few toxicological tests to exclude acute toxicity risks. Only active compounds which could be interesting for further development are then studied in more detail since
Fig. 14.1 The structure of some natural sweet compounds.
Developing new sweeteners from natural compounds 329 an extensive toxicity study is usually very expensive and time consuming. To use a natural compound as a lead to develop new potentially active compounds also has some disadvantages. One of them is the fact that often natural compounds have complex structures: large molecular weight, many functional groups and several stereogenic centres are all factors that can pose severe limitations to the structural modifications to design new analogues to be synthesised. For instance molecules like stevioside, osladin or glycyrrhizin are not very useful as leads for a large number or analogues since the synthetic accessibility of the skeleton is limited. This inconvenience could be partly overcome if a common synthon could be found with a simple and accessible synthesis or by means of biotransformations. This synthon has then to be modified with simple and selective reactions able to remove, add or change a functional group into another. In structure±activity relationships (QSARs) the systematic variation of structural features is important in order to highlight some direct relationship between certain functional groups and biological activity. This approach is especially needed when the new set of analogues is not very large, therefore the relationships are developed using `classical' methods. Over the years, many natural compounds has been used as a lead to develop new analogues, but only a few of these were successful in giving a high number of new active compounds. In the next sections a description is given of the main problems encountered and which strategies can be used in order to obtain the maximum degree of information from natural compounds minimising the synthetic and analytical efforts in the synthesis of new analogues.
14.3 Methods of designing new sweeteners from natural compounds In order to design new derivatives, scientists have to take into account many factors. The first is the synthetic accessibility of derivatives. Very often a compound that seems to be a good candidate is practically un-accessible at least in a reasonable time and in amounts sufficient for analytical characterisation and preliminary sensory evaluation, i.e. usually at least 50±100 mg. In recent years the availability of the affinity in vitro tests with cloned receptors has made possible the screening of smaller amounts of compounds, usually in the milligram scale. Nevertheless, these tests are still not available commercially; their use in a pre-screening phase is extremely useful to detect the binding of a substrate to the receptor, but this condition does not necessarily imply automatically that a compound will be sweet in sensory evaluation. The second important feature to satisfy is the predictability of biological activity, that is usually indicated with the denomination of `rational' design. This means that the choice of the new derivatives to be synthesised is not random but is, at least in part, based on some knowledge of the biological mechanism underpinning the sweet taste. Taking this into account, the history of the discovery of sweet compounds can be roughly divided into three phases, illustrated in Fig. 14.2.
330
Optimising sweet taste in foods
Fig. 14.2 Time line of the discovery of sweet compounds.
Before 1967 the discovery of new sweet compounds was made exclusively by serendipity, since there was a complete lack of information on the sweet chemoreception mechanism. The second phase can be considered to start with the article of Shallenberger and Acree (1967): they recognised the existence in almost every sweet molecule of two functional groups corresponding to a hydrogen bond donor and a hydrogen bond acceptor, named AH and B respectively. Their role in recognition involved the creation of two parallel Ê apart, with two complementary sites on the receptor. hydrogen bonds at ca 3.5 A This second phase, which is partly still continuing, can be referred to as the era of the topological models. Many experimental indications suggests the existence of a protein (or a family of proteins) able to bind the sweet ligand; but, in the absence of information on the protein's structure, all the attention is focused on the molecular feature of the ligands, as the starting point to develop common structure±taste relationships. From the chemical point of view, this approach was already known in the study of biologically active compounds, where the concept of glucophore is equivalent to that of pharmacophore. This phase has been extremely fruitful in the design of new sweet compounds: starting from known leads, and following the indications of the topological models, many possibilities were opened to synthetic chemists to design and prepare large families of analogues. The results of activity gave a feedback to improve the initial model, by adding new glucophores, or giving more precise information on their relative position and function in the interaction with the putative receptor, so that a sort of iterative process involving design-synthesis-activity tests can continue. During this phase, the initial Shallenberger and Acree model was improved by many contributions among which those by Kier (1972) who added a third interaction site (first called X) corresponding to a hydrophobic region of the molecule, Temussi et al. (1990, 1992), DuBois et al. (1993), Goodman et al. (1994), Suami et al. (1994) were most important. The most complete version of these kind of models is the so called Multipoint Attachment Model (MPA) by Tinti and Nofre (1991, 1993), describing a hyperpotent sweetener as an ideal molecule containing eight glucophores of different importance, each of them able to interact with a corresponding site on the receptor protein. This model is an elaboration of the initial tripartite AH/B/X model and is able to explain the taste of hyperpotent complex molecules such as sucrononic acid and derivatives, discovered by the authors in the 1980s and 1990s (Tinti and Nofre, 1991). The topological models have been very useful to address the synthesis of new analogues of known sweet compounds.
Developing new sweeteners from natural compounds 331 An example of this process is represented by the family of the so-called isovanillic derivatives. The natural lead of isovanillic derivatives is phyllodulcin, a dihydroisocoumarin found in the leaves of Hydrangea Macrophylla (Asahina and Asano, 1929). Phyllodulcin has never been used as a sweetener: this is probably due to the fact that the natural source is quite expensive, and that it has a strong liquorice or anise aftertaste, as in the case of many other natural derivatives, that limits its possible applications. The total synthesis of optically active phyllodulcin has been described but it is probably not suitable for an industrial development. Another isovanillic derivative is neohesperidindihydrochalcone (NHDC) (Horovitz and Gentili, 1969) which is by contrast commercial; this is obtained by hydrogenation of bitter flavanones from low cost orange peels. During the years, many (Q)SARs and molecular modelling studies have been made on this class of compounds, which focused the attention on different aspects of their structure±taste relationships: identification and role of glucophores; active conformation; role of stereochemistry. 14.3.1 Identification and role of potential glucophores to design new analogues Since their discovery, the isovanillic derivatives were recognised to be sweet thanks to the presence of the isovanillic moiety on one of the aromatic rings, ring c. In fact the hydroxyl and o-methoxyl groups form a suitable couple of AH and B glucophores respectively, where AH is the hydroxyl and B is the methoxy group at a distance compatible with that postulated by Shallenberger and Acree (1967). To verify this starting hypothesis, several analogues were synthesised by modifying selectively some of these groups and/or the other possible AH/B system in the phyllodulcin molecule, made by the carbonyl group in the b ring and the adjacent hydroxyl in position 8 of the a ring (Dick, 1981). The elimination or substitution of one of these groups generally produced a decrease or the disappearance of the sweet taste. Some modified isovanillic derivatives are shown in Fig. 14.3. These studies established that the isovanillic substituents were essential interaction points for the biological activity. Nevertheless, the isovanillic ring alone does not give sweetness, therefore it is also important to understand the role of the remaining molecular fragments. Again, following the indications of the Shallenberger-Acree-Kier model, it was reasonable to hypothesise that the rings a and b in phyllodulcin and derivatives could correspond to the hydrophobic glucophore X (also called G in the MPA model). This glucophore is indeed quite undefined in its electronic and spatial definition since it corresponds to an area more than to a point. Consequently, some attempts to modify this region gave contrasting results on activity. To study the effects of ring a structure it is very useful to synthesise derivatives modified in this region, for instance having some hydrophobic substituents in various positions or with an aliphatic instead of the aromatic ring. By the way, this example illustrates very well the fact that the design of new derivatives is sometimes limited by synthetic procedures. For instance the
Fig. 14.3 The structure of some synthetic isovanillic derivatives. Modifications have been made to the isovanillic moiety as well as to the substituents and heteroatoms in a and b rings.
Developing new sweeteners from natural compounds 333 introduction of substituents on ring a is not obvious, due to the lack of commercially available appropriate synthons. Similarly, the simple substitution of ring a with a cyclohexane would introduce two new stereogenic centres on the molecule, making it much more difficult for the separation of single enantiomers and the attribution of the absolute configuration to each one. It was less easy, and still is, to establish the role of the other heteroatoms in the molecule. Actually it has been demonstrated that the sweet taste is conserved in the absence of the hydroxyl group in position 8. This is an advantage in the synthesis of derivatives since the contemporary presence of two hydroxyls makes critical the choice of appropriate and selective protective groups during the synthesis. In general, to reduce as much as possible the presence of functional groups on a new molecule while maintaining the activity is the general goal of the synthetic chemist. On the other side, the role of the heteroatom in the b ring is very important in the taste of derivatives (Arnoldi et al., 1991, 1993). Some heterocyclic analogues of phyllodulcin are shown in Fig. 14.3. They have been synthesised in order to study the effect of heteroatoms, their number and position in this ring. The carbocyclic analogue of phyllodulcin is not sweet, but the taste is very dependent on heteroatoms. In particular the best effect is obtained with two heteroatoms in the 1,3 positions, and improved by substitution of oxygen with sulfur. Again, the rationale which underlies the experimental design comes both from synthetic considerations and molecular modelling. In fact, the substitution of oxygen with sulfur, a so called isoster group, is a common practice in drug design that very often gave good results. Moreover, the introduction of two heteroatoms in the 1,3 positions instead of in the 1,4, is due to the fact that it is much more easy to access acetalic and thioacetalic derivatives using the appropriate diol or thiol and isovanillin as reagents. On the contrary, the 1,4 derivatives must be prepared using a more complicated procedure. At least in this case, the most simple derivatives to be prepared were also luckily the most active: in fact the derivatives with acetalic and thioacetalic structure are also the most active up to now in the isovanillic series and their synthesis ± as racemates is very easy also in gram scale. The use of some of these compounds as aroma components in the preparation of chewing gum has been recently patented (Patent 2003a). Nevertheless, it is not clear which potential glucophores the heteroatoms in b ring correspond to. The superimposition of such compounds with the MPA model did not give clear-cut answer to this question, therefore an indirect role, for instance in modifying the electronic density of aromatic ring a, cannot be excluded. This demonstrates that the topological models can be very useful guides for developing new analogues, but sometimes also chemical intuition and even serendipity can play a role in finding new and better derivatives. 14.3.2 The active conformation in designing new analogues The three dimensional structure of a biologically active compound is given not only from the primary structure, i.e. the sequence of functional groups, but also from the spatial arrangement that the molecule assumes in the active site, also
334
Optimising sweet taste in foods
referred to as the `active conformation'. The active conformation or a similar conformation could be possessed by the molecule itself when this is also the minimum energy conformation or when the molecule is relatively rigid. If the molecule is flexible the active conformation can be assumed when it reaches the active site, i.e. inside the receptor cavity, using a certain energy content. The information on active conformation is therefore very useful for the design of a new analogue and also to define the structure of the active site and therefore to improve the receptor model. In the family of isovanillic derivatives many analogues have been projected in order to study the active conformation (Arnoldi et al., 1991). Phyllodulcin and its analogues are relatively simple to study since the major degree of conformational freedom for the molecule consists in the rotation of ring c around the simple C±C bond and to a little extent, the flexibility of heterocyclic ring b. Some hints can be obtained with two strategies: the synthesis of rigid analogues and the relationships between conformation and activity (Arnoldi et al., 1998). Figure 14.4 shows the structure of some rigid isovanillic derivatives. In choosing their structures, the existence of two `limit conformations' have been taken into account: one having the rings a and c roughly co-planar (angle between the two planes about 0ë) and the other having the two planes approximately perpendicular (angle ca 90ë). The first type has been prepared using a steroid skeleton as a useful synthon to mimic the planar structure; the second class instead has a spiranic structure, which is easily obtained by spiroacetalisation of an appropriate substituted indanone. Other rigid compounds having a bowl shaped structure have been obtained mimicking the structure of the natural sweet compound haematoxylin (Arnoldi et al., 1995). The sensory evaluation showed that the rigid derivatives were all tasteless or less sweet than the parent compounds, therefore indicating an active conformation different from that of the limit conformations. Besides the use of rigid derivatives, information on the active conformation could be deduced by means of (Q)SARs using appropriate steric descriptors.
Fig. 14.4 The structure of rigid (bowl-shaped, planar or spiranic) derivatives in the isovanillic family.
Developing new sweeteners from natural compounds 335
Fig. 14.5
Geometric descriptors (which can then be confirmed by means of Principal Component Analysis (PCA)).
Some of these descriptors are shown in Fig. 14.5; they have been calculated or measured for a series of flexible isovanillic derivatives and a correlation sought using the Principal Component Analysis (PCA). The results of this simple approach are quite good: in fact the method permitted the identification of a probable active conformation and the ability to distinguish sweet from non-sweet compounds based on geometric characteristics. In particular, the enantiomers of the same compound were correctly classified according to their taste: one is sweet and the other tasteless. 14.3.3 Role of stereochemistry in designing new analogues The presence of stereogenic centres on a natural compound is generally an additional problem for the synthetic chemist. A good synthon is available only in a few cases and it must be cheap, stable, and easy to modify in the desired manner without racemisation or, alternatively, a good and economically convenient stereoselective synthetic procedure is needed to achieve the target compound with the correct stereochemistry and high optical purity. Actually the most successful commercial sweeteners are all achiral (e.g., saccharine, cyclamates and sucrononic acid) or, if they are chiral, they are derived by classical chiral synthons such as sugars (e.g., sucralose) or peptides (e.g., aspartame), for which the technology of classical transformation and biotransformation is accessible and convenient. In all other cases, even if the stereochemistry of the natural compound is relatively simple, it must be carefully controlled. In the isovanillic family the access to optically active
336
Optimising sweet taste in foods
derivatives with the thioacetalic structure was an important goal, since these compounds are the sweetest of the series. Unfortunately, even when the synthesis of racemates is very easy, the stereoselective introduction of a chiral centre on the thioacetalic carbon atom has not yet been reached. The two enantiomers of these derivatives have in fact been solved by means of chiral chromatography (Bassoli et al., 2000) and tasted separately, but this procedure is too expensive to be applied on a gram scale. A method based on chiral induction with organochromium derivatives has been developed to generate stereoselectively the thioacetalic derivatives (Bassoli et al., 1996), but it proved incompatible with the presence of the phenolic substituent on the isovanillic ring and therefore has not been applied, leaving open the question of an easy access to these very sweet derivatives.
14.4
Synthesising new sweeteners from natural compounds
The goal of using natural compounds as leads is general. The obtaining of high intensity thioacetalic sweeteners from phyllodulcin is an example, but several other natural compounds have been studied for this purpose. Sucrose is probably the best lead among natural compounds, since its taste profile is unique. Many sucrose derivatives have been synthesised and studied in recent years; nevertheless, sucrose is one of the most difficult compounds to be modelled, owing to the not straightforward identification of the glucophores and to the conformational freedom and flexibility. Actually, the only analogue of sucrose that had commercial success is the trichloroderivative, sucralose. Its development comes out initially from the serendipitous discovery of the sweet taste of some halogenated sugar derivatives, and then from the systematic design of analogues with one or more halogen atoms in various positions. In this example the structural modification to the original natural compound is quite big; this could also explain the fact that many years elapsed before the approval for introducing sucralose in the market was given. By the way, in structure±activity relationships it is common that a `strong diversity in structure' does not necessarily correspond to a `strong difference in activity' and vice-versa. This is well illustrated by the example of the sweet taste of monatin and derivatives. Monatin is a high-intensity sweet natural compound isolated from the roots of Schlerochiton ilicifolius, a spiny-leaved hardwood shrub growing in South Africa. Monatin is about 1,200±1,400 times more sweet than sucrose and has a very clean sweet taste without an aftertaste and the typical lingering and/or liquorice aroma that usually is present in many other sweeteners (Ackermann et al., 1991). Monatin is also very interesting from the chemical point of view. It has a modified amino acidic structure related to simple chiral synthons such as glutamic acid and tryptophan, which are themselves taste active compounds. It is intensely sweet and its taste profile is described as extremely pleasant and similar to that of sucrose; it has a relatively simple structure with many polar
Developing new sweeteners from natural compounds 337 groups thus it is very soluble in water and suitable for many possible applications. Therefore this molecule is a good candidate for practical applications as a sugar substitute. In the last few years monatin has been studied quite intensively as a natural compound, especially in the industrial area as demonstrated by the numerous patents filed on this subject. Monatin has two principal limitations: the first is that it is not very accessible since it is found in the roots of a plant (not in the leaves) that grows in a restricted area. The second limitation, from the chemical point of view, is the presence of two stereogenic centres. The one in C2 is an amino acidic residue and therefore is, at least in principle, easily introduced from the appropriate synthon. On the other hand, the C-4 chiral centre is a quaternary hydroxy acid and its stereoselective introduction is quite difficult. Owing to these two limitations, research on monatin has moved in the directions of: (a) finding an efficient stereoselective synthesis of natural 2S,4S-monatin; (b) understanding the relationship between taste and absolute configuration; in fact the synthesis of a racemate would have been much more easy to perform; (c) finding simple analogues with similar taste characteristics but more accessible. Some synthetic approaches to racemic monatin have been described in the open literature or in patents (Holzapfel et al., 1994; Patent, 1999). Moreover, some stereoselective syntheses of (-)-monatin have been reported (Nakamura et al., 2000, 2004; Tamura et al., 2003). Since 2002 many patents have been filed concerning monatin synthesis and its possible applications as a sweetener. In one of these patents (Patent, 2003b), monatin has been obtained from a cross aldol reaction of a specific pyruvic acid with oxalacetic or pyruvic acid, followed if necessary, by decarboxylation. Subsequently the carbonyl group of the ketoglutaric acid compound is then replaced with the amino group. In 2004 the same company obtained monatin by resolution of diastereomeric mixtures. Monatin has also been prepared by complex microbiological processes using mutated D-aminotransferases from glucose, tryptophane and indol-3-lactic acid (Patent, 2003c) or 4-(indol-3-ylmethyl)-4-hydroxy-2-oxoglutaric acid (Patent, 2004). Many of these methods are of little applicability on an industrial scale owing to the use of complicated chiral reagents or catalysts, multi-step procedures and low yields. In another patent the sensory properties of all the four monatin stereoisomers are described (Patent, 2003d). It was extremely interesting the finding that all the four stereoisomers of monatin were described to be sweet, with different intensities. In particular, the sweetest compound seems to be 2R,4R monatin, i.e. the enantiomer of the natural compound. A similar result was obtained by our research group (Bassoli et al., 2005). We have developed a chemoenzymatic approach, using a protease from Aspergillus oryzae, to the synthesis of all the four stereoisomers of monatin, that were obtained in high chemical and optical purity. In sensory evaluations, three proved sweet and one tasteless. This
338
Optimising sweet taste in foods
experimental result was initially surprising, and a lot of work was dedicated to the attribution of the absolute configuration of each enantiomer. In fact, obtaining single enantiopure compounds, or diastereomeric mixture or racemates has very different degrees of difficulty; it is therefore of the maximum importance to relate each enantiomer to its taste properties. The configuration of 2S,4S has been assigned to natural levorotatory monatin based on NOE NMR experiments on a cyclic derivative and the application of the empirical Clough-LutzJirgenson rule, but the crystal structure of the compound is not yet available. The stereogenic centre at C-2 was introduced non-specifically, and the resulting diastereomeric mixtures were separated by RP-HPLC. The absolute configuration of the intermediate products was assigned by X-ray diffraction of chiral derivatives. The absolute configurations of the final products were indeed established by comparing the retention times on a chiral HPLC column with those of known samples obtained in an independent way. Surprisingly, a sample of natural monatin analysed in the same conditions is shown to contain all the four stereoisomers. It is not clear if this reflects the actual content in the plant or is due to the possible isomerisation of the chiral centres during extraction and manipulation of monatin samples. Once again, it is very important to establish the exact structure of the natural compound, and which isomer/isomers is/are actually `natural' in order to address the synthesis of natural-identic compound and the synthesis of analogues. The sweet taste of monatin and its stereoisomers is dependent on stereochemistry. As usual, in trying to design some new analogues it is important to understand which are the relevant glucophores to interact with the receptor and which of them could possibly be eliminated in order to have a simpler structure. Some attempts have been made for this purpose. In such a molecule, it is not easy to identify the glucophores using topological models. For instance, both the terminal amino-acid and the central hydroxy-acid moiety could, in principle, correspond to the AH-B system. In fact the distances of these groups with the hydrophobic G group (identified in the indole ring) and the chemical characteristics are compatible with the MPA model in both cases. Therefore, initially it seemed more useful to prepare some monatin analogues by elimination of the alcoholic hydroxyl, which is more difficult to introduce by synthesis. In fact the deoxy monatin derivatives can be easily prepared by alkylation of pyroglutamic acid as a precursor. The derivatives obtained were tasteless in the tasting trials, indicating that the hydroxyl group plays an important role for the interaction of this molecule with the putative taste receptor. It is not easy to explain in detail what this role could be, but again some information could be obtained taking into account both the concepts of active conformation and the use of receptor models which can give some hints into the interactions between the receptor and the ligand, like the general pseudoreceptor model for sweet compounds (Bassoli et al., 2002). Pseudoreceptor modelling is a technique initially developed to study drugs and the interaction with their putative receptor, and has been successfully applied by our group to explain and predict semiquantitatively the sweet taste of many sweet compounds owing to different chemical classes.
Developing new sweeteners from natural compounds 339
Fig. 14.6 R,R monatin in the cavity of the pseudoreceptor model.
Figure 14.6 shows the R,R isomer of monatin inside the active site of the pseudoreceptor for sweet compounds. In the minimum energy conformation ± that in our modelling is also maintained into the active site ± R*,R* monatin has Ê ) that involves the terminal ±NH3+ group as the a strong hydrogen bond (1.99 A hydrogen donor and the oxygen of the C-4 hydroxyl group as the acceptor. The other diastereoisomer, R*, S* monatin, has a different minimum energy conformation with three intramolecular hydrogen bonds: one between H of hydroxyl in C-4 and carboxylate in C-2; one between H of COOH in C-4 and O of hydroxyl in C-4; another between H of NH3+ in C-2 and carboxylate in C-2. It is therefore possible that the hydroxyl group plays an important role in the taste of monatin, not as a glucophore itself, but indirectly, by introducing an important constraint which generates an active conformation. The sweet taste disappears also in all other deoxy monatin derivatives. On the other hand the other groups have an important role, since those lacking the carboxyl or the amino or even the indole ring are all tasteless: therefore, much information is to be gained in order to design new and potentially active derivatives of this natural compound.
340
14.5
Optimising sweet taste in foods
Future trends
What can we expect in the future in designing new active compounds from natural leads? First of all, the search for new active principles is and will be more and more active, due to the globalisation that allows researchers to extend their area of action all over the world. Moreover, the analytical instrumentations and techniques allow us to identify and to examine hundreds of new structures each year. At the same time, new theoretical and practical tools are going to become available to the researchers. One of these is undoubtedly the use of in vitro assays with cloned receptors. In the developing new active molecules probably one of the `rate determining steps' has been until now the time and costs of sensory analysis and of related toxicological protocols. Each new compound is needed in amounts of at least 100±200 mg to be tested even with a restricted panel and some more compound, often in gram scale, is need for preliminary toxicology studies. The use of in vitro tests is going to reduce strongly these problems: only minimal amounts of compound are needed (usually micromolar solutions are suitable) and no toxicology is needed. Of course the binding assays do not substitute for sensory evaluation: the perception of taste is a complex mechanism, that starts on the papillae with the receptor binding but needs then to be transmitted through the neurons and elaborated in the brain. Therefore there is no direct evidence that the affinity measured with the in vitro tests is linearly correlated with the relative sweetness assessed by sensory evaluations, and false positive as well as false negative results cannot be excluded. As the use of in vitro tests will become more diffuse, it seems likely that some systematic comparison between the two methods will be made in order to understand the potential and limitations of this methodology, but it is already clear that the possibility to assay many new derivatives, at least for a rough pre-screening, is a powerful tool in the hands of researchers. Another future scenario in designing new active compounds is opened up by the knowledge of the structure of the sweet taste receptor proteins that were recently disclosed in the literature (Max et al., 2001; Montmayeur et al., 2001; Kitagawa et al., 2001; Nelson et al., 2001; Li et al., 2001; Sainz et al., 2001). The sweet taste receptor is a G protein coupled receptor (GPCR) similar to the dimeric mGluR1 receptor. Both belong to class C of GPCRs, which includes also umami taste receptors. The main structural difference between the sweet receptor and mGluR1 receptor is that while the first is heterodimeric, mGluR1 is homodimeric, albeit with two slightly different conformations of its two chains (A and B). A unique feature is that, while the ligands of mGluR1 are either glutamate or closely related molecules, the ligands able to activate the sweet taste receptor vary widely in chemical constitution, ranging from sugars to amino acids, peptides, proteins and several other classes of organic compounds. In the last years many research groups contributed to modelling the structure of the sweet taste receptor, and particularly the active site. We have now some information coming from genetics, molecular biology and molecular modelling
Developing new sweeteners from natural compounds 341 which contributes to our understanding of how small sweeteners and large proteins (Temussi, 2002; Morini et al., 2005) can bind to the sweet taste receptor. The knowledge of the molecular details of this interaction will be in the future the key to designing new active compounds, mimicking the activity of natural agonists as sugar and amino acids, or synthetic sweeteners, as aspartame or saccharine, or macromolecules such as proteins. Moreover, this could allow us to understand complex mechanisms such as the phenomenon of the synergy among different sweet compounds, through the understanding of the mechanisms of allosteric modulation of the sweet taste receptor proteins. All this information will probably give in the future important results in the discovery of new sweet compounds.
14.6
References
and STEYN P S (1991), `Structure elucidation of monatin, a high-intensity sweetener isolated from the plant Scherochiton Ilicifolius', J. C. S. Perkin Trans. 1, 3095±3098. ARNOLDI A, BASSOLI A, MERLINI L and RAGG E (1991), `Isovanillyl sweeteners. Synthesis, conformational analysis and structure-activity relationships of some sweet oxygen heterocycles', J. Chem. Soc., Perkin Trans. 2, 1399±1406. ARNOLDI A, BASSOLI A, MERLINI L and RAGG E (1993), `Isovanillyl sweeteners. Synthesis and sweet taste of sulfur heterocycles', J. Chem. Soc., Perkin Trans. 1, 1359±1366. ARNOLDI A, BASSOLI A, BORGONOVO G and MERLINI L (1995), `Synthesis and sweet taste of optically active (-)-haematoxylin and of some (+)-haematoxylin derivatives', J. Chem. Soc., Perkin Trans. 1, 2447±2453. ARNOLDI A, BASSOLI A, BORGONOVO G, DREW G B, MERLINI L and MORINI G (1998), `Sweet isovanillyl derivatives: synthesis and structure±taste relationships of conformationally restricted analogues', J. Agric. Food Chem., 46, 4002±4010. ASAHINA Y and ASANO J (1929), `On the structure of hydrangenol and phyllodulcin', Chem. Ber., 62, 171±177. BASSOLI A, MERLINI L, BALDOLI C, MAIORANA S and DREW M G B (1996), `Asymmetric synthesis of (+) and (-)-2-(2-methoxyphenyl)-3,1-benzoxathiane', Tetrahedron Asymm., 7, 1903±1906. BASSOLI A, BORGONOVO G, DREW M G B and MERLINI L (2000), `Enantiodifferentiation in taste perception of isovanillic derivatives', Tetrahedron Asymm., 11, 3177±3186. BASSOLI A, DREW M G B, MERLINI L and MORINI G (2002), `A general pseudoreceptor model for sweet compounds: a semi-quantitative prediction of binding affinity for sweet tasting molecules', J. Med. Chem., 45, 4402±4409. BASSOLI A, BORGONOVO G, BUSNELLI G, MORINI G and DREW M G B (2005), `Monatin and its stereoisomers: chemoenzymatic synthesis and taste properties', Eur. J. Org. Chem., 1652±1658. DICK W E (1981), `Structure-taste correlations for flavans and flavanones conformationally equivalent to phyllodulcin', J. Agric. Food. Chem., 29, 305±312. DUBOIS G E, WALTERS D E and KELLOGG M S (1993), `The rational design of ultra-highpotency sweeteners', in Mathlouthi M, Kanters J A and Birch G G, Sweet taste chemoreception, London, Elsevier, 237±267. ACKERMAN L G J, VLEGGAAR R
342
Optimising sweet taste in foods
and KENT D (1994), `Conformational requirements for sweet-tasting peptides and peptidomimetics', Angew. Chem., Int. Ed. Engl., 33, 1437±1451. HOLZAPFEL C H, BISCHOFBERGER K and OLIVIER J (1994), `A simple cycloaddition approach to a racemate of the natural sweetener monatin' Synth. Comm., 24, 3197±3211. HOROVITZ R M and GENTILI B (1969), `Taste and structure in phenolic glycosides', J. Agric. Food Chem., 17, 696±700. KIER L B (1972), `A molecular theory of sweet taste', J. Pharm. Sci., 61, 1394±1397. KITAGAWA M, KUSAKABE Y, MIURA H, NINOMIYA Y and HINO A (2001), `Molecular genetic identification of a candidate receptor gene for sweet taste', Biochem. Biophys. Res. Comm., 283, 236±242. GOODMAN M, YAMAZAKI T, BENEDETTI E
LI X, INOUE M, REED D R, HUNQUE T, PUCHALSKI R B, TORDOFF M G, NINOMIIYA Y, BEAUCHAMP
G K and BACHMANOV A A (2001), `High-resolution genetic mapping of the saccharin preference locus (Sac) and the putative sweet taste receptor (T1R1) gene (Gpr70) to mouse distal Chromosome 4', Mamm. Genome, 12, 13±16. MAX M, SHANKER Y G, HUANG L, RONG M, LIU Z, CAMPAGNE F, WEINSTEIN H, DAMAK S and MARGOLSKEE R F (2001), `Tas1r3, encoding a new candidate taste receptor, is allelic to the sweet responsiveness locus Sac', Nature Genetics, 28, 58±63. MONTMAYEUR J P, LIBERLES S D, MATSUNAMI H and BUCK L B (2001), `A candidate taste receptor gene near a sweet taste locus', Nature Neuroscience, 4, 492±498. MORINI G, BASSOLI A and TEMUSSI P (2005), `From small sweeteners to sweet proteins: anatomy of the binding sites of the human T1R2-T1R3 receptor', J. Med. Chem., 48(17), 5520±5529. NAKAMURA K, BAKER T J and GOODMAN M (2000), `Total synthesis of monatin', Organic Letters, 2 (19), 2967±2970. NAKAMURA K, KOGISO K, NAKAJIMA T and KAYAHARA H (2004), `Total synthesis of monatin and the taste expression', Peptide Sci., 40, 61±64. NELSON G, HOON M A, CHANDRASHEKAR J, RYBA N J P and ZUKER C S (2001), `Mammalian sweet taste receptor', Cell, 106, 381±390. NOFRE C and TINTI J M (1991), `Why does a sweetener taste sweet? A new model', in Walters D E, Orthoefer F T and DuBois G E, Sweeteners discovery, molecular design and chemoreception, Washington, DC, American Chemical Society, 206± 213. PATENT (1999), US 5994559 A, Nov 30. Abushanab, E and Arumugam, S. `High-yield synthesis of the high-intensity natural sweetener Monatin'. PATENT (2003a), US 2003/0072842A1. Johnson, S and Greenberg, M J `Hydrophobic sweetener-containing chewing gum having prolonged sensory benefits'. PATENT (2003b), WO2003059865. Kawahara S, Amino Y; Mori K, Funakoshi N and Takemoto T `Processes for the preparation of glutamic acid compounds and intermediates thereof and novel intermediates used in the processes'. PATENT (2003c), WO2003091396. Abraham T W, Cameron D C, Dalluge J, Hicks P M, Hobson R J, McFarlan S C, Millis J and Rosazza J `Polypeptides and biosynthetic pathways for producing monatin'. PATENT (2003d), WO2003045914 (JP); EP 1449832 (ENG). Amino Y, Yuzawa K, Mori K and Takemoto T `Crystals of non-natural stereoisomer salts of monatin as sweeteners'. PATENT (2004), WO2004053125. Sugiyama M, Watanabe K, Kashiwagi T and Suzuki E, `D-aminotransferase mutants for stereoselective synthesis of glutamic acid derivatives: synthesis of (2R,4R)-monatin, a sweetener.
Developing new sweeteners from natural compounds 343 and SULLIVAN S L (2001), `Identification of a novel member of the T1R family of putative taste receptors', J. Neurochem., 77, 896± 903. SHALLENBERGER R S and ACREE T E (1967), `Molecular theory of sweet taste', Nature, 216, 480±482. SUAMI T, HOUGH L, TSUBOI M, MACHINAMI T and WATANABE N (1994), `Molecular mechanism of sweet taste. V. Sucralose and its derivatives', J. Carbohydr. Chem., 13, 1079±1092. TAMURA O, SHIRO T, TOYO H and ISHIBASHI H (2003), `Highly stereoselective synthesis of (-)monatin, a high-intensity sweetener, using chelation-controlled nitrone cycloaddition', Chem. Commun., 2678±2679. TEMUSSI P A (2002), `Why are sweet proteins sweet? Interaction of brazzein, monellin and thaumatin with the T1R2-T1R3 receptor', FEBS Lett., 526, 1±3. TEMUSSI P A, CASTIGLIONE MORELLI M A, LELJ F, NAIDER F, TALLON M and TANCREDI T (1990), `Conformation-activity relationship of sweet molecules. Comparison of aspartame and naphthimidazolesulfonic acids', J. Med. Chem., 33, 514±520. TEMUSSI P A, KAMPHUIS J, LELJ F, TANCERDI T and TONIOLO C (1992) `SAR of sweet molecules: conformational analysis of two hypersweet and two conformationally restricted aspartame analogues', Quant. Struc.-Act. Relat., 11, 486±491. TINTI J M and NOFRE C (1991), `Design of sweeteners', in Walters D E, Orthoefer F T and DuBois G E, Sweeteners discovery, molecular design and chemoreception, Washington, DC, American Chemical Society, 88±99. TINTI J M and NOFRE C (1993), `In quest of hyperpotent sweeteners', in Mathlouthi M, Kanters J A and Birch G G, Sweet taste chemoreception, London, Elsevier, 205± 236. SAINZ E, KORLEY J N, BATTEY, J F
15 Improving the taste of sweeteners D. E. Walters, Rosalind Franklin University of Medicine and Science, USA
15.1
Introduction
The blending of high potency sweeteners to improve sweet taste was first reported in 1921, when only two high potency sweeteners were in use (Paul, 1921). Since then, numerous sweeteners have been used in combinations of two or more at a time. There is almost always some advantage found: synergistic sweetness potency, improved taste profile, improved temporal profile (Walters, 1993; Verdi and Hood, 1993). Synergy is the observation of a higher level of sweetness from a blend than would be expected if the sweetness is additive. This permits the use of less sweetener, often resulting in cost savings. The existence of synergy suggests that different sweeteners may act at different sites on the receptor rather than competing for a single site. Improved taste profile is observed when different sweeteners have different off-tastes. The sweetness adds, while the off-tastes do not, and the overall taste quality is improved. Improved temporal profile can occur because some sweeteners have a fast onset of sweetness while others have a slower onset than sucrose; some sweeteners have a lingering sweetness while others clear quickly. By blending two or more sweeteners, a temporal profile can be achieved which is closer to that of sucrose than any of the individual components.
15.2
Blending sweeteners to provide synergy
There are several ways to define and measure synergy in sweetness blends. These are discussed in Chapter 16 of this volume. Here we review reports of synergy from the literature of the last 84 years. Paul reported that the degree of
Improving the taste of sweeteners
345
sweetness of saccharin is unproportionally increased by the addition of dulcin (Paul, 1921). Helgren and coworkers observed synergy when saccharin and cyclamate are combined (Vincent et al., 1955; Helgren, 1957). They speculated that at least part of this synergy may result from a decrease in the off-tastes, since less of each sweetener is used in the blend. Saccharin also shows synergy with neohesperidin dihydrochalcone (Ishii et al., 1972) and with glycine (Kravitz, 1960). A group in Japan reported synergy in two-way and three-way blends of saccharin, dulcin and cyclamate (Ueno, 1964). Yamane (1966) described this effect as a general phenomenon of sweeteners. Aspartame has been blended with many other sweeteners. Synergy has been reported for a blend of aspartame and saccharin (Hill and La Via, 1972; Yamaguchi et al., 1972). Frank et al. (1989) found 38% synergy for an aspartame/saccharin blend. Yamaguchi et al. (1972) also found synergy when aspartame and cyclamate are blended. Scott (1973) described synergy in blends of aspartame and saccharin, lactose, mannitol, sorbitol, or sucrose. Eisenstadt (1981) claimed synergy when aspartame is blended with monoammonium glycyrrhizinate, potassium bitartrate, and a sugar or sugar alcohol. Acesulfame-K has also been blended with many other sweeteners since its discovery in 1976. An acesulfame-K/aspartame blend was reported to show 36% synergy (Frank et al., 1989). Wells (1989) reported synergy of acesulfame-K with aspartame, cyclamate, thaumatin, stevioside, and sucralose. Stephens and Torres (1985) found synergy in a combination of acesulfame-K with alitame. Beyts and Latymer (1985) reported that sucralose shows synergy with saccharin, acesulfame-K, or stevioside, but not with sucrose or aspartame. Neotame shows isobole synergy with saccharin, acesulfame-K, aspartame, and sucrose (Pajor and Gibes, 2000). Not all combinations have shown synergy. Synergy was not found when fructose was blended with saccharin, aspartame, or acesulfame-K (Van Tornout et al., 1985). Lawless found that the combinations sucrose/aspartame and acesulfame-K/saccharin do not exhibit synergy (Ayya and Lawless, 1992).
15.3
Blending sweeteners to improve taste profile
Many of the studies described above under the heading of synergy also reported an improvement in taste profile. The earliest report of sweetener blending (Paul, 1921) noted, in addition to synergy, that the mixture of saccharin plus dulcin is more pleasant than saccharin alone. Helgren and coworkers noted substantially reduced off-tastes in blends of saccharin plus cyclamate (Vincent et al., 1955; Helgren, 1957). Two-way and three-way blends of saccharin, dulcin and cyclamate have decreased bitterness and other off-tastes (Ueno, 1964; Yamane, 1966). Blending aspartame with saccharin (Hill and La Via, 1972; Yamaguchi et al., 1972) or cyclamate (Yamaguchi et al., 1972) is also known to improve taste quality. Bakal (1983) reported improved taste quality for several blends of aspartame with other sweeteners.
346
Optimising sweet taste in foods
Acesulfame-K was reported to exhibit a more sucrose-like taste quality when blended with aspartame, cyclamate, saccharin, or neohesperidin dihydrochalcone (von Rymon Lipinski and LuÈck, 1979). Acesulfame-K/aspartame blends have found wide application in the food industry. Stephens and Torres (1985) reported that the combination of acesulfame-K with alitame results in less bitterness. Blending of saccharin, aspartame, or acesulfame-K with 2±3% fructose was found to improve the taste quality of soft drinks (Van Tornout et al., 1985). Not all blends improve taste quality. Horne et al. (2002) showed that the bitter tastes of saccharin and acesulfame-K are positively correlated among panelists, suggesting that they act at a common bitter receptor. This was confirmed by Kuhn et al. (2004), who showed that these two sweeteners activate the same two bitter taste receptors. Thus, it is not be surprising that this blend shows little or no taste quality improvement.
15.4
Blending sweeteners to improve temporal profile
Sucrose, the standard against which all other sweeteners are judged, has a characteristic time-intensity profile. Sweeteners which diverge from this profile are judged to be less acceptable in most food applications. Some high potency sweeteners, such as saccharin and acesulfame-K, have rapid onset of sweetness. Others, such as thaumatin and neotame, have slow onset and lingering sweetness. Blending sweeteners with differing temporal profiles is well known to produce sweetness with a temporal profile closer to that of sucrose (Bakal, 1991; Walters, 1993).
15.5
Using additives to improve taste quality of sweeteners
Numerous additives have been proposed to improve the taste quality or temporal profile of sweeteners and sweetener blends. For example, saccharin taste quality is reported to be improved by formulation as a series of amine salts such as the ethanolamine salt or various aminosugar salts (Vacek, 1967). The aluminum salt of saccharin is reported to exhibit a longer temporal profile in chewing gum (Rieger and Yang, 1989). The temporal profile of neotame can be modified by addition of hydrophobic organic acids (such as cinnamate) or hydroxyamino acids (such as serine and tyrosine) (Gerlat et al., 2002; Prakash et al., 2001). These additives, which are not themselves sweet, may compete for receptor binding sites, leading to less lingering sweetness. Burge and Nechutny (1978) reported that sugar acids can improve the sweetness of thaumatin, monellin, and saccharin when these sweeteners are used alone or in combination.
Improving the taste of sweeteners
15.6
347
Future trends
Sweetener blends offer numerous advantages to the food scientist. Synergy can significantly lower sweetener usage and cost. Blending can minimize off-tastes and tailor the temporal profile of sweetness to specific needs. The trend is clearly away from single sweetener usage in most applications.
15.7
Sources of further information and advice
Schiffman has published extensively on binary and ternary blends of sweeteners (Schiffman et al., 1995, 2000, 2003). This work has covered a range of sweetness levels and a large number of sweetener combinations.
15.8
References
and LAWLESS, HT (1992) Quantitative and qualitative evaluation of high-intensity sweeteners and sweetener mixtures. Chem. Senses, 17, 245±259. BAKAL, AI (1983) Functionality of combined sweeteners in several food applications. Chem & Ind. (London), 700±708. BAKAL, AI (1991) Mixed sweetener functionality. In Nabors, LO & Gelardi, RC (Eds.) Alternative Sweeteners. 2nd edn. New York, Marcel Dekker. BEYTS, PK and LATYMER, Z (1985) Sweetening agents containing chlorodeoxysugar. US Pat 4,495,170. BURGE, MLE and NECHUTNY, Z (1978) Sweetening compositions containing aldohexuronic acids. US Pat 4,096,285. EISENSTADT, ME (1981) Dipeptide sweetener compositions. US Pat 4,254,154. FRANK, RA, MIZE, SJS and CARTER, R (1989) An assessment of binary mixture interactions for nine sweeteners. Chem. Senses, 14, 621±632. GERLAT, PA, WALTERS, GC, BISHAY, IE, PRAKASH, I, JARRETT, TC, DESAI, N, SAWYER, HA and BECHERT, C-LT (2002) Use of additives to modify the taste characteristics of Nneohexyl-alpha-aspartyl-L-phenylalanine methyl ester. US Pat 6,368,651. HELGREN, FJ (1957) Sweetening compositions and method of producing the same. US Pat 2,803,551. HILL, JA and LA VIA, AL (1972) Sweetening compositions containing saccharin and dipeptides. US Pat 3,695,898. HORNE, J, LAWLESS, HT, SPEIRS, W and SPOSATO, D (2002) Bitter taste of saccharin and acesulfame-K. Chem. Senses, 27, 31±38. ISHII, K, TODA, J, AOKI, H and WAKABAYASHI, H (1972) Sweetening composition. US Pat 3,653,923. KRAVITZ, HL (1960) Sweetening agents. Canada Pat 602,572. AYYA, N
KUHN, C, BUFE, B, WINNIG, M, HOFMANN, T, FRANK, O, BEHRENS, M, LEWTSCHENKO, T, SLACK,
and MEYERHOF, W (2004) Bitter taste receptors for saccharin and acesulfame K. J. Neurosci., 24, 10260±10265. PAJOR, LL and GIBES, KM (2000) N-[N-(3,3-dimethylbutyl)-L-alpha-aspartyl]-Lphenylalanine 1-methyl ester synergistic sweetener blends. US Pat 6,048,999. JP, WARD, CD
348
Optimising sweet taste in foods
(1921) Der SuÈssungsgrad von Dulcin und Saccharin [The sweetness potencies of dulcin and saccharin]. Chem.-Ztg, 45, 38. PRAKASH, I, BISHAY, IE, DESAI, N and WALTERS, DE (2001) Modifying the temporal profile of the high-potency sweetener neotame. J. Agric. Food Chem., 49, 786±789. RIEGER, MM and YANG, RK (1989) Aluminum salt of saccharin. US Pat 4,855,145. SCHIFFMAN, SS, BOOTH, BJ, CARR, BT, LOSEE, ML, SATTELY-MILLER, EA and GRAHAM, BG (1995) Investigation of synergism in binary mixtures of sweeteners. Brain Res. Bull., 38, 105±120. SCHIFFMAN, SS, SATTELY-MILLER, EA, GRAHAM, BG, BOOTH, BJ and GIBES, KM (2000) Synergism among ternary mixtures of fourteen sweeteners. Chem. Senses, 25, 131± 140. SCHIFFMAN, SS, SATTELY-MILLER, EA, GRAHAM, BG, ZERVAKIS, J, BUTCHKO, HH and STARGEL, WW (2003) Effect of repeated presentation on sweetness intensity of binary and ternary mixtures of sweeteners. Chem. Senses, 28, 219±229. SCOTT, D (1973) Sweetening compositions and method for use thereof. US Pat 3,780,189. STEPHENS, CR, JR. and TORRES, A (1985) Synergistic sweetening compositions. Eur Pat Appl 84306007.0 UENO (1964) [A novel method for the utilization of artificial sweeteners]. Shokuhin to Kagaku [Food Science], 6, 39±43. VACEK, LC (1967) Organic amine salts of saccharin. US Pat 3,325,475. VAN TORNOUT, P, PELGROMS, J and VAN DER MEEREN, J (1985) Sweetness evaluation of mixtures of fructose with saccharin, aspartame or acesulfame K. J. Food Sci., 50, 469±472. VERDI, RJ and HOOD, LL (1993) Advantages of Alternative Sweetener Blends. Food Technol., 47, 94±101. VINCENT, HC, LYNCH, MJ, POHLEY, FM, HELGREN, FJ and KIRCHMEYER, FJ (1955) A Taste Panel Study of Cyclamate-Saccharin Mixture and of its Components. J. Amer. Pharm. Assoc., Sci. Ed., 44, 442±446. È CK, E (1979) Sweetener mixture. US Pat 4,158,068. VON RYMON LIPINSKI, G-W and LU WALTERS, DE (1993) High-Intensity Sweetener Blends: Sweet Choices. Food Product Design, 3, 83±92. WELLS, AG (1989) The use of intense sweeteners in soft drinks. In Grenby, TH (Ed.) Progress in Sweeteners. London, Elsevier Applied Science. YAMAGUCHI, S, SHIMIZU, A, KIRIMURA, J and NINOMIYA, T (1972) A method for improving the taste of artificial sweeteners. Japan Pat 43-90383. YAMANE, T (1966) Kanmiryo [Sweeteners]. Korin-Compendia. Tokyo, Korinshoin. PAUL, T
16 Analysing and predicting synergy in sweetener blends P. Laffort, Centre des Sciences du GouÃt, France
16.1
Introduction
The important increase in consumption of artificial sweeteners in Western countries in recent decades is, of course, due to our sedentary life style and to the concomitant need to limit calorie intake. In fact, as is well known, the adaptation of eating habits to suit people's lives that have become more and more sedentary is multifaceted and needs to be greatly improved, judging by the increased problem of obesity in the population. In this chapter we will confine our study to those low calorie sweeteners in the market with palatability as close as possible to those of natural sugars: sucrose, glucose and fructose.1 Artificial sweeteners are made up with low calorie substances, having sweetness power often one thousand times higher than natural sugars. Commercial products often include mixtures of several of these substances in order to gain the benefit of a mutual synergistic effect. The role played by the models for chemosensory mixtures is therefore important: it allows one to optimise the positive aspects (a pleasant sweet taste) and to minimise the negative ones (bitter or irritant off-tastes or after-tastes). Before starting the description of some published predictive models and their application to some concrete examples, let us specify the different meanings given to the concept of interaction.
1. Interest in this concerns the population as a whole, but, more especially, the increasing number of people struck down by diabetes.
350
Optimising sweet taste in foods
16.1.1 What is synergy? An important paper had this title in 1989 by M. C. Berembaum, considered one of the pioneers, in pharmacology, of interaction phenomena in general, that is to say: the suppression, the addition (or absence of interaction) and the synergy. Regarding the question asked, in spite of numerous models that are not always easy to understand in their mathematical presentation, the answer is straightforward: There are only two ways to consider the perception of sweetener mixtures. The first one considers the sweetness as a function of the component concentrations (psychophysical approach) and the second one as a function of the component responses taken separately (perceptive approach). The first approach is supposed to be more practical, but unfortunately the results obtained cannot generate simple general rules. The second approach, by contrast, even less practical, can generate indices of interaction which are constant, in principle, for all reciprocal concentrations of the components and for various levels of perceived intensities. Once the indices of interaction are established, it is possible to come back to the first approach by combining these indices with the parameters of the psychophysical function of the components. Before we successively develop these two approaches, we will first recall the ins and outs of sweet taste measurements, particularly those concerning the psychophysical function.
16.2
Sweet taste measurements
The several psychophysical techniques used to measure sweet perception are detailed in Chapter 6. We consider these descriptions as established and we will first focus here on the stimulus-response diagrams in chemical senses in general, and more particularly in the perception of sweet taste. This question is, as we will see, very important in the building of interaction models, and needs to be clarified in advance. 16.2.1 The stimulus-response diagrams Almost always in olfaction, and in many cases in taste, the psychophysical experimentation does not allow one to elicit the stimulus-response diagrams on the whole, from the very low perceptions up to the saturation plateaus. Some peripheral animal experimentation allows one to do that, with more often, in addition, better reproducibility. These techniques of animal experimentation are principally, on the small vertebrate, sums of spikes from whole sensitive nerves (olfactory nerve in olfaction, chorda tympani in taste) or recording of the global potentials (olfactory mucosa). It has been widely demonstrated that these peripheral global measurements are proportional to the perceived intensities. It is observed that the diagram response vs. the concentration presents a general sigmoid shape in semi-log coordinates, and a straight line ended by a plateau in log-log coordinates (see Fig. 16.1). The parameters of the schematic diagram in Fig. 16.1 have been set using four models:
Analysing and predicting synergy in sweetener blends 351
Fig. 16.1 General representation of the sensorial stimulus-response function, in log-log and semi-log coordinates, together with the parameters involved in four practical models of this function (see Equations 16.1, 16.2, 16.3 and 16.5). Note that Cx is not equivalent to K, as sometimes printed, but to K1/n (from Laffort, 1994a).
The Hill model (1913), adopted by Tateda in 1967, and expressed using the following equations: R
RM
C=Cx n 1
C=Cx n
or
R
RM C n =K 1 C n =K
16:1
in which R is the sensorial response, RM the maximal response, C the concentration, Cx the concentration at the inflexion point, n the slope of the rectilinear section in log-log coordinates and K a so-called dissociation constant.2
2. The Hill model, as well as the Beidler model presented later, lie on a theoretical approach based on the law of mass action between stimulus and receptors. We leave aside this controversial aspect and only keep the practical aspect of the models. Let us note that sometimes in the literature, the parameter K 0 1=K is applied instead of K.
352
Optimising sweet taste in foods
This model is easy to handle and very popular. Its only drawback is that it implies that Rx RM =2, and this relation is not always experimentally verified. It should also be noted that the Cartesian coordinates (i.e. without log transformations), sometimes used in taste literature for a graphical representation of the Hill model, are not suitable, as we will see in Section 16.2.3. The Laffort model (1966, 1977, 1994a), allows one to overcome the drawback of the Hill model, but it is less easy to handle (Patte et al., 1989). It is given by the equation: R0
16:2 R C0 =C
R0 =RM 1=n n where R0 and C0 are the coordinates of an anchorage point in the rectilinear part of the dose-response diagram (more often the minimum of perception). The Beidler model (1954),3 is a particular case of the two previous models, when n 1: R
RM C=K 1 C=K
16:3
Let us note that in the particular case where n 1, the way to go from the Hill to the Beidler model is evident, whereas going from the Laffort to the Beidler model needs the following equation: VM C0
16:4 R0 K The power law. In olfactory psychophysical experimentations (human verbal responses), the inflexion point is almost never reached: we have only the left part of the stimulus-response diagram (Fig. 16.1), that is to say a straight line in log-log coordinates. This function is called the power law or Stevens's law, from S. S. Stevens (1936), who first proposed it for several sensorial modalities. The power law (obtained with RM 1 in Equation 16.2) is expressed as follows: n R C or log R n
log C ÿ log C0 log R0
16:5 R0 C0 It is generally considered that R0 1, or, it amounts to the same thing, that R=R0 is the perceived intensity. It is observed in psychophysical olfactory experimentation, that the slopes (or exponents) values are more often in a range of 0.2±0.7 (e.g. Devos et al., 2002). In taste psychophysics, by contrast, the exponents are often nearer to 1 and, in addition, the plateau values are often reached. These double characteristics explain the initial success of the Beidler model in taste studies. 3. In fact, Lloyd Beidler adopted the Hill model in 1971.
Analysing and predicting synergy in sweetener blends 353 16.2.2 Comparison between the different techniques of sweetness measurement As shown by Nicklaus and Issanchou (Chapter 6, this volume), three principal psychophysical techniques are applied to evaluate sweetness: the magnitude estimation, the category scaling and the sucrose equisweet matching. Only the magnitude estimation can be considered as a direct measurement, supposed to best reflect the actual perceived intensity. However, as observed by Laffort et al. (2002) on the basis of some published examples, the sucrose equisweet appears as the more precise and reproducible, whereas the data obtained using the magnitude estimation are the most scattered, and those with category scaling fall in between. A similar observation is reported in olfaction by Devos et al. (2002) in a compilation of human power law exponents: among ten or so techniques, the three most used are the magnitude estimation, the category scaling and the 1butanol olfactory matching; the greatest scattering of measurements is observed for the magnitude estimation. The use of the sucrose equisweet method for sweetness measurement, the best method as we saw, needs a good knowledge of the power law exponent for sucrose, in order to transform relative sweetness perceived intensities into absolute perceived intensities. In numerous studies using magnitude estimation, the power law exponent for sucrose has always been found very near to 1. We will see in Section 16.2.3 that the value of 0.94 seems more precise. 16.2.3 Values of the anchorage point and of the power law exponent for several sweeteners Schiffman et al. (1995) have measured, using a reference scale of sucrose solutions, the sweetness intensity rating (SIR) of 14 sweeteners at six concentrations. The concentrations of the sweeteners under study have been chosen as the multiples (respective ratios of 1.5, 2.5, 3, 3.5, 5 and 7) of their concentrations equisweet to 2% sucrose. Because of the number of sweeteners studied and the good reproducibility of measurements, we will use the data of this paper in the present chapter to illustrate developments that are sometimes slightly abstract. We begin by drawing the stimulus response diagrams in log-log coordinates for the 14 sweeteners, each one with seven experimental points (6 1) (Fig. 16.2). Precise technical details on the procedures used by these authors and on our data processing are given in the appendix (Section 16.9). Several comments can be made about Fig. 16.2. Let us first note that, in eight cases out of 14, a plateau of saturation is reached. The values of the obtained slopes (exponents in Cartesian coordinates), which will be applied in Section 16.4.3 to calculate the interaction indices in binary mixtures, are therefore only applicable in the range of concentrations where the linearity in log-log coordinates is effective. We have not established the precise values of the plateaus and inflexion points using Equations 16.1 or 16.2, since the interaction models considered, as we will see, do not account for the whole stimulus-
Fig. 16.2 Sweetness intensity ratings (SIR) vs. seven relative concentrations for 14 sweeteners, in log-log coordinates, according to the experimental data of Schiffman et al. (1995). Relative concentrations are multiples of the concentrations equisweet to 2% sucrose, respectively 1, 1.5, 2.5, 3, 3.5, 5 and 7. Open circles indicate points not taken in the regression (see text).
356
Optimising sweet taste in foods
response sigmoid.4 It is sometimes possible to recover points of the stimulusresponse curvilinear drawing in log-log coordinates, using a change of variables (Laffort et al., 2002). We will not apply this procedure here, in order not to overcomplicate the presentation. Let us also note the shakiness of the point 6 (twice the concentration isointense with 3% sucrose), particularly for thaumatin and neo-DHC.5 We therefore discarded this point in the regression for these two sweeteners as well as for fructose and mannitol. These characteristics can probably be explained by the fact that the experimental procedures are not the same for the six points (see details in the appendix (Section 16.9)). The values of the power law exponents obtained, as well as the anchorage points of concentrations (according to Schiffman et al., 2000) are reported in Table 16.1. We observe in Table 16.1, the wide range of sweetness powers (C2 concentrations in a ratio of about 1 to 56,000). Let us also note the range of power law exponents n, quite different to those observed in olfaction, where the most frequent values stand around 0.35 (Devos et al., 2002). We can see the strong correlation between the values of n and log C2 (r 0:81, N 14, p < 0:001), in the same way as has been known in olfaction for a long time: the lower the thresholds the lower generally are the exponents. One generally explains this phenomenon as a recruitment, in the statistical meaning of the word, of populations of neuroreceptors kitted out with several types of molecular receptors and with different activation thresholds. Let us underline that these various observations are far from having only academic interest: every consumer of artificial sweeteners knows that to increase coffee sweetness by adding a half piece of sugar to a previous full piece is equivalent to adding a second piece of aspartame to the first one. Also on a practical point of view, we will see in Section 16.5.1 that Table 16.1 suggests an improvement of the equisweet techniques as they are currently applied.
16.3 Response of a mixture vs. responses to its components: the perceptive models 16.3.1 The vector model The response to a mixture vs. the response to its components or perceptive approach, allowed Berglund et al. to propose in 1973 a vector model, which turned out to be immediately very efficient and popular in olfaction. Its principle is the following: The perceived intensities combine according to the rule of the parallelogram of forces, as shown in Fig. 16.3. 4. We will see in Section 16.5.4 that a model which considers the whole sigmoid response of components gives rise to an interaction model with too many parameters. 5. Neohesperidin dihydrochalcone.
Analysing and predicting synergy in sweetener blends 357 Table 16.1 Experimental slopes n in log-log coordinates of the SIR (sweetness intensity ratings) vs. concentrations for 14 sweeteners, according to the experimental data of Schiffman et al. (1995). Also reported are the log of concentrations C2 iso-intense with 2% sucrose, according to Table 1 of Schiffman et al. (2000) Sweeteners Acesulfame K Alitame Aspartame Fructose Glucose Mannitol Na cyclamate Na saccharin Neo-DHC Rebaudioside A Sorbitol Stevioside Sucrose Thaumatin
log C2 (ppm)
n
R2
No. of points
1.99 0.65 1.90 4.19 4.53 4.57 2.88 1.59 1.13 1.70 4.71 2.02 4.30 ÿ0.05
0.60 0.62 0.69 0.92 1.03 0.95 1.00 0.66 0.55 0.57 1.57 0.72 0.94 0.58
0.98 0.98 0.99 1.00 1.00 1.00 0.96 0.95 0.98 0.99 0.99 0.96 1.00 0.99
6 7 6 6 7 5 5 6 6 6 5 4 7 5
The mathematical expression of this model is as follows: q RAB R2A R2B 2RA RB cos
16:6
An objection to this model is that it does not allow us to strictly speaking account for synergies: the upper limit of cos is, indeed, the value 1. In fact, this objection can be easily overcome in an imaginary space where cos > 1. Of course, in this latter case the phenomenon cannot be visualized anymore, as in Fig. 16.3. RB
R AB
RB
a
R AB
a RB
RB
Fig. 16.3 Principle of combination of sensory perceptions, according to Berglund et al. (1973), by using the so-called vector model. (from Laffort, 1994b).
358
Optimising sweet taste in foods
A more serious objection is that obtaining cos values from experimental data needs squares of these data, with a consequent amplification of the measurements' uncertainties. To overcome this difficulty, the authors of the vector model determine cos values from iso-intense components of the mixtures. In this case, using classical trigonometrically rules, Equation 16.6 becomes: RAB
RA RB cos =2
16:7
The constraint of iso-intensity of the components is real in most practical situations, but not in the present case of the data of Schiffman et al. (1995), which we have selected here to illustrate and test the described interaction models. 16.3.2 The U model An alternative model, also of the perceptive approach type, named the U model, has been proposed by Patte and Laffort (1979).6 By contrast to the vector model, the U model cannot be visually represented, but its index of interaction can be defined without any problem from all reciprocal relative concentrations of the components. The U model is expressed by the equation: p
16:8 RAB RA RB 2 RA RB cos u 16.3.3 Application of the perceptive models to experimental data In the paper by Schiffman et al. (1995), chosen to illustrate the present chapter, the authors have studied three mixtures for each of the 14 selected sweeteners: the sweetness intensity ratings (SIR) of the mixtures 3 3, 5 5 and 7 7 of sucrose equivalent values (SEV). The authors evaluate the interaction using two methods, respectively called I and II. Method I consists of verifying if the responses are significantly different from the physical additivity. As we can see in Table 16.2, this method is not suitable, the results at the point 7 7 being opposite to those at the point 3 3. The so-called method II counteracts the scale effect, comparing the intensities of mixtures to the intensities of components, in a similar way as the vector and U models, but in a less formal manner not here detailed. In fact, in the case of iso-intense components of the mixtures we have here, the vector and U models are equivalent (according to Equations 16.6 and 16.8, the ratio RAB =
RA RB is in this case a constant with both models, whatever the levels are).7 For personal preferences, we chose the U model. Usually we recommend 26 experimental points to establish a mean value of cos u : five concentrations 6. Its name comes from the shape of the drawing of vs. with RAB =
RA RB and RB =
RA RB . 7. Also equivalent is the Independent Component Index (ICI) of Hyman and Frank (1980), not defined here in order not to over-complicate the presentation. In fact, the ICI model is only interesting for mixtures with iso-intense components. For more details, see Laffort (1994c).
Analysing and predicting synergy in sweetener blends 359 Table 16.2 Summary of the interactions in 91 binary mixtures of sweeteners, obtained by Schiffman et al. (1995) using `method I' (see text)
No. of suppressions No. of additions No. of synergies
Level 3 3
Level 5 5
Level 7 7
3 27 36
10 48 8
48 18 0
No. of suppressions at the three levels: No. of additions at the three levels: No. of synergies at the three levels:
3 7 0
of each of the two components taken separately and 16 different mixtures (4 4). In the present case, because we have only three points for each pair of sweeteners, we have preferred to establish only one value of cos u for each sweeteners pair at the points 5 5, and applied it at the points 3 3 and 7 7. The results, as they can be seen in Fig. 16.4, are only partially satisfactory, with a slight over prediction at the level 7 7, possibly due to numerous plateau values observed at this level (see Fig. 16.2). The values of cos u for the points 5 5 are reported in Fig. 16.5. Let us remember that according to Equation 16.8, negative values correspond to suppressions and positive values to synergies, the bold and tinted values standing for significant interactions. The general aspect is not really very different from the original Table 4 in Schiffman et al. (1995), with numerous values of suppression when one sweetener is added to itself, which is, of course, not very
Fig. 16.4 Predicted vs. experimental sucrose intensity ratings (SIR) of the 91 binary mixtures of 14 sweeteners, by using the U model applied to the experimental data of Schiffman et al. (1995). Learning with the cos u of Fig. 16.5 established at the level 5+5 (not represented here); testing with the levels 3 3 (open circles) and 7 7 (full circles).
Fig. 16.5 Interaction cos u values for 91 binary sweeteners mixtures, according to the data of Schiffman et al., 1995 (points 5+5 SEV). Also reported are the 14 cos u values of the sweeteners added to themselves. Significant suppressions are in bold and significant synergies have tinted background. See comments in text.
Analysing and predicting synergy in sweetener blends 361 relevant. This phenomenon is due to the influence of the power law exponents, as can be easily demonstrated, but it can also be verified with the data of Schiffman et al. (1995). There is, indeed, a significant correlation between these cos u values (diagonal of Fig. 16.5) and the power law exponents of the 14 sweeteners, as reported in Table 16.1 (r 0:82, N 14, p 0:001). We will therefore refer to the interactions characterized by the U model as global. These global interactions include the interactions coming from the power law of components and also intrinsic interactions. The purpose of Section 16.4 will be to separate precisely the two types of interaction.
16.4 Response of a mixture vs. concentrations of its components: the ÿ (Gamma) family models The experimental procedure adopted by Schiffman et al. (1995) is the most practical: two sweet solutions are mixed (of course, avoiding a dilution phenomenon) and the resulting sweetness is estimated. This procedure does not, however, provide an answer to the question generally asked: Which are the various mixtures producing a given sweetness, for example iso-intense to 10% sucrose? The ÿ family models allow us to go from the most practical procedure to the question generally asked, in other words the sweetness vs. the concentrations of components. 16.4.1 The ÿ (Gamma) model Historically, this model, published by Laffort et al. in 1989, was the first to provide an index (the ÿ index) reflecting an interaction independent of the power law exponents of the components and valid for all their respective concentrations. The interaction due to the power law exponents is expressed by an index called cos UPL2 (UPL2 stands for the second version of the model, in which the power law is applied to the U model). The challenge of cos UPL2 is to transform exponential expressions into a multiplicative coefficient. Its definition requires the prior definitions of P (as proportion), and cos A and cos B (A and B characterizing the substances of the binary mixture): P cos A cos B cos UPL2
1=nB
RB
1=nA
RA
1=nB
RB
1 ÿ PnA
1 ÿ PnA 2PnA =2
1 ÿ PnA =2 1 ÿ PnB
1 ÿ PnB 2PnB =2
1 ÿ PnB =2 RA cos A RB cos B RA RB
16:9
362
Optimising sweet taste in foods
A priori established by Laffort et al. (1989), with only an a posteriori validation, the theoretical validity of Equations 16.9 was mathematically demonstrated by Callegari (1998). This demonstration can also be found in appendix B of Laffort et al. (2002). Once cos UPL2 established, the ÿ index is defined as follows: 1 cos u
16:10 ÿ 1 cos UPL2 From these definitions, values of ÿ > 1 correspond to synergies, ÿ < 1 to suppressions and ÿ 1 to absences of interaction. cos UPL2 and ÿ index are generalizations to all respective proportions of components of binary mixtures, of interaction indices previously defined by Laffort and Dravnieks (1982), respectively called cos UPL and index, and only valid for mixtures of iso-intense components. 16.4.2 Other models of the ÿ family The ÿ0 model includes the same equations' cascade 16.9 of the ÿ model, but with Equation 16.10 replaced by the following: ÿ 1 cos u ÿ cos UPL2
16:11
The ÿ-vector model has been proposed by Laffort (1994c). It is complementary to the vector model, like the ÿ model it is complementary of the U model. The relation (Equation 16.10) is replaced by the following: 1 cos VECT
16:12 ÿVECT 1 cos VCPL and the equations' cascade as follows: P
1=nB
RB
1=nA
RA
1=nB
RB
cos A
1 ÿ P2nA
1 ÿ P2nA 2PnA
1 ÿ PnA
cos B
1 ÿ P2nB
1 ÿ P2nB 2PnB
1 ÿ PnB
cos VCPL
RA cos A RB cos B RA RB
16:13
A variation of the ÿ-vector model, named ÿ0 -vector model, uses instead of Equation 16.12, the following: ÿ0VECTOR 1 cos VECT ÿ cos VCPL
16:14
A comparative study of these four variations of the ÿ model, on the basis of various experimental data published in olfaction, taste and pharmacology, has
Analysing and predicting synergy in sweetener blends 363 shown that the ÿ0 model produces the least scattering for the interaction index and therefore appears as the most suitable (SeÂreÂe and Laffort, 1997; ThomasDanguin and Laffort, 1998). The ÿ0 model will therefore be applied in the two following sections to the construction of iso-intensity diagrams. 16.4.3 A priori construction of iso-intensity diagrams The purpose is to build, from ÿ0 indices well established, iso-intensity curves in a diagram as represented in Fig. 16.6, in which Cartesian coordinates are relative concentrations to Cë concentrations producing a given effect (for example, in our case, SEV 10%).8 In this type of representation, the straight dotted line stands for an absence of interaction and separates the two areas of synergy and suppression (sometimes called in the literature respectively hyper-addition and hypo-addition). The equations' cascade (Equation 16.9 and Equation 16.11), which allow one to quite easily calculate the ÿ0 indices, are not, however, reciprocal: They generate implicit equations. In other words, once a ÿ0 index is established, it is not easy to come back to iso-intensity curves vs. relative concentrations according to Fig. 16.6. Laffort et al. (1989) used an iterative program for that. A user-friendly version of this program in EXCEL, named MIG (mixtures intensities generation), written by Pierre HeÂricourt (2005), has been used here.9 16.4.4 Application of the ÿ0 model to experimental data The ÿ0 model has been applied to the 91 mixtures and to the 14 sweeteners added to themselves, studied by Schiffman et al. (1995) at the level 5 5. We used for that the power law exponents reported in Table 16.1. The corresponding 91 14 ÿ0 indices have been reported in Fig. 16.7. Generally speaking, the ÿ0 indices for pure components (diagonal in Fig. 16.7) are near 1, each time the linearity is observed in the log-log drawings of Fig. 16.2, without reaching plateau values. By contrast, wrong inhibition values of ÿ0 indices (values in bold) appear in Fig. 16.7 for the four sweeteners presenting an early plateau in Fig. 16.2 and/or a greater scattering of the experimental points. This is particularly true for sorbitol therefore rendering the inhibition ÿ0 values doubtful for the sweetener mixtures containing sorbitol. These exceptions illustrate the need, already mentioned, of a change of variables when a linearity of the stimulus-response function in log-log coordinates is not effective in a given range of concentrations. That means, of course, taking some precautions, for example, making the same change of variables for the two sweeteners of the mixture and also coming back to the initial variables at the end of the procedure. Let us note in Fig. 16.7, that the strongest observed synergy (the highest value of ÿ0 ) concerns the mixture aspartame/acesulfame-K (ÿ0 1:45). We also
8. This type of diagram was first applied in chemical senses by E. P. KoÈster (1969). 9. The MIG program will soon be available on the website: http://paul.laffort.free.fr
364
Optimising sweet taste in foods
Fig. 16.6
Principle of representation of the chemosensory intensity interaction at a given level.
observe a good agreement for the mixture fructose/glucose between the value reported here (ÿ0 1:04) and the value established by Laffort et al. (2002) from the data of De Graaf and Frijters (1986): ÿ0 1:08. By contrast, the strong synergy found by Laffort et al. (2002) for the mixture sucrose/Na cyclamate from the data of Nahon et al. (1998) (ÿ0 1:32) are not in agreement with the value of Fig. 16.7 (ÿ0 1:08). The latter value is very probably wrong, because Na-cyclamate presents an early plateau in Fig. 16.2. All these applications to experimental data are not at all to suggest definitive indices of interaction, but to give examples of how the models described work. A question arises about the sweet taste interaction indices ÿ0 : if ÿ0 values between A and B, and between B and C are known, can we deduce the value between A and C? At first sight from Fig. 16.7 data, the answer seems to be no, and therefore this property, called transitivity, unfortunately does not apparently apply to the sweetness ÿ0 indices. A priori iso-intensity diagrams have been built using the MIG program, for two typical mixtures: one exhibiting a strong symmetrical synergy and the other one exhibiting a dissymmetrical suppression. These two examples are drawn in Fig. 16.8. The observed dissymmetry in the right side is due to the difference of power law exponent's n of the two components. It should be underlined that the isointensity curves generated using the ÿ0 model can differ when the range of levels Ä is wide. Here, the diagrams have been built for an SIR value equal 7 ( 7% SEV). It has been verified that for SIR 5 and SIR 9, the curves are practically superimposed. In addition to the experimental points 5 5, which, by construction, lie exactly on the curves, we plotted the experimental point 3 3 for the mixture acesulfame-K/aspartame. We saw the shakiness of the point 3 3 for thaumatin, and therefore it is not present in the right part of Fig.
Fig. 16.7 Interaction ÿ0 indices for 91 binary sweeteners mixtures, according to the data of Schiffman et al.,, 1995 (points 5 5 SEV). Also reported are the 14 ÿ0 values of the sweeteners added to themselves. Significant suppressions are in bold and significant synergies have tinted background. See comments in text.
366
Optimising sweet taste in foods
Fig. 16.8 Two typical iso-sweet curves for two sweetener mixtures, according to the ÿ0 model applied to experimental data of Schiffman et al. (1995) at the level 5+5: strong and almost symmetrical synergy on the left side; dissymmetrical suppression on the right side. Open circles stand for experimental points (5 5 upon the curves and 3 3 near to the curve on the left side; see text).
16.8, nor the two experimental points 7 7, both situated in plateau areas in Fig. 16.2. Anyway, the experimental points are too few to justify the curves of Fig. 16.8, which is just an illustration of how to go from perceptive approach experimentations to this type of presentation.
16.5
Other interaction models
16.5.1 The substitution model The MDI model (Mixture Discrimination Index), also called substitution model, has been defined by Hyman and Frank (1980) by the following equation: RAB
16:15 MDI 0 R in which R0 is `the response to either component at a concentration equal to the sum of two concentrations of one chemical equivalent in effect to the components of the mixture'. When applying the power function to this definition (Equation 16.5), two equations are obtained: 1=nA
RB
1=nB
RB
R0
RA R0
RA
1=nA nA
16:16
1=nB nB
16:17
Equations 16.16 and 16.17 are only equivalent when the exponents for the two mixtures are the same. That is, as one saw, a limited view of the actual reality. No rule is suggested when the exponents differ, but one could make an arithmetic mean of the two values of R0 . In the particular case where the two exponents are the same, it can be easily demonstrated that MDI is equivalent to the first expression of the ÿ family indices (called ÿ not ÿ0 ) for equally strong
Analysing and predicting synergy in sweetener blends 367 components (Laffort et al., 2002). In other words, the ÿ model appears as a generalization of the substitution model. By contrast, no equivalence is found between MDI and the ÿ0 index currently used and judged more suitable. 16.5.2 The equiratio model This model (Frijters and Oude Ophuis, 1983) was proposed for the study of taste mixtures. Even if the name was new, the method had been proposed previously (Sales, 1958) and used in the gas industry over a period of many years (Borelli and Angleraud, 1965; Blanchard, 1976). The principle of the equiratio model is quite simple: the concentration-response straight line in log-log coordinates for a series of dilutions of a mixture is expected to lie between the straight lines corresponding to the components of the mixture, based on the ratio of the mixture's components. Angleraud (1966) and later Blanchard (1976) demonstrated that this assumption was not experimentally verifiable in the general case of power function exponents not equal to 1 (more details can be seen in Laffort, 1994c, fig. 3). The non-applicability of this model to olfaction was confirmed by Schiet and Frijters (1988). The equiratio model has been successfully applied in some studies of sweet taste mixtures (Frijters et al., 1984; De Graaf and Frijters, 1986; Frijters and De Graaf, 1987), but failed in other cases (Frijters, 1987, present chapter in Fig. 16.8). In the particular case where the value of the exponents are very similar and near to 1, the equiratio model for iso-intensity mixtures (equisweet for sweet taste) is simplified as follows (Equation 9 in De Graaf and Frijters, 1986): CA CB
16:18 1 A B C C The graphic expression of Equation 16.18 is the dotted straight line in Fig. 16.7. Similarly, it can be demonstrated (Laffort et al., 2002) that in this particular case the ÿ0 model can be expressed as follows: CA CB 1ÿ C A C B r ÿ0 1 cos u 1
16:19 CA CB 2 C A C B That means that when Equation 16.18 is verified, ÿ0 1 (numerator equal zero in Equation 16.19), and therefore it becomes possible for the equiratio model to be successfully applied. In all other cases, the ÿ0 index value may provide an estimation of the divergence between the equiratio model and the experimental results (cases of Fig. 16.8). However, this formal estimation is only possible when the exponents values nA and nB are very close to 1. 16.5.3 The isobole approach applied in pharmacology According to the authors using this method (for example Berembaum, 1985) an index of interaction I can be calculated as follows:
368
Optimising sweet taste in foods I
CA CB 0 CA0 CB
16:20
in which CA0 and CB0 are the concentrations of A and B that would individually elicit the same response as the mixture. It can be applied to all types of stimulus± response relationships and can be generalized to any number of components of a mixture. Taking the hypothesis of the power function for single components (Equation 16.5), it becomes: RA 1=nA RB 1=nB
16:21 I RAB RAB This model has been applied to olfactory and taste binary mixtures by SuÈhnel (1993). Without entering into details, it can be said that the isobole interaction approach satisfies the condition that if one sweetener is added to itself, the interaction index I remains constant, equal to 1, like ÿ0 . By contrast, I values are not constants characterizing given pairs of odorants, each time the exponents are different. One can estimate values of I by adding x and y coordinates of iso-intensity curves in Fig. 16.8. This type of figure is called in pharmacological literature `isobologram', except that the coordinates used by Berembaum and SuÈ hnel are absolute concentrations, not relative concentrations. Strictly speaking, diagrams as in Fig. 16.8 should be named adapted isobolograms. It should also be noted that if the iso-intensity points are experimentally obtained, there is no objection to their isobologram representation. By contrast, when the experimental points come from estimations of responses by adding solutions, as in the Schiffman et al. (1995) paper, SuÈhnel used interpolation programs currently available in statistical packages. If the number of experimental points is few, as is generally the case in chemosensory experimentation, this type of interpolation may be hazardous in the absence of an underlying model (as, for example, the ÿ0 model). 16.5.4 The OBM model (Olfactory Binary Mixtures) The principle of this model (Thomas-Danguin and Chastrette, 2002) lies in a summation of the sigmoid stimulus-response curves observed in semi-log coordinates for components to be further mixed. The frequent occurrence of saturation plateaus in sweet taste studies, as shown in Fig. 16.2, could a priori make use of the OBM model suitable in this field. The implementation of this model needs the prior setting of parameters of the two sigmoid curves of the binary mixture, using the Hill model (Equation 16.1) modified by adding a supplementary parameter for each component, i.e. I0A and I0B, defined as `the perceived intensity without odorant for the component A (B)'. This model of interaction involves therefore eight parameters (four for each component) plus four interaction indices, that is to say 12 parameters in all. The nature and the number of parameters considered by these authors do not allow application of the OBM model to sweetener mixtures. However, another
Analysing and predicting synergy in sweetener blends 369 more appropriate type of summation of sigmoid curves could perhaps be a topic of further fruitful research.
16.6
Future trends
16.6.1 Testing a modified sucrose reference scale Taking into account the power law exponent of sucrose, equal to 0.935 (appendix, Section 16.9), allows one to suggest two reference scales of sweetness, respectively with five and seven equidistant steps, as described in Table 16.3. It should be interesting to test if one or another of these scales is more efficient than that of DuBois et al. (1991), used by Schiffman et al. (1995). 16.6.2 Interaction in mixtures with more than two components The models with more than two components in the field of chemical senses, whatever the approaches considered, are not satisfactory and need substantial improvements. One of the reasons comes from the small number of published experiments in this field until recent years. The numerous data published by Schiffman et al. (2000) on ternary mixtures could perhaps be in favour of a modelling improvement in a near future. 16.6.3 Sweet taste synergy and off-taste suppression The question of the off-tastes is certainly the most important to be improved. As mentioned in the introduction, the tools to be applied are the same models as those used for sweetness. More precisely we recommend the selection of 24 experimental points (four for each sweetener and 16 mixtures) in the range of concentrations previously studied for sweetness, and to establish for each point a profile of off-tastes (for example, initial bitterness, after sweet and after bitter
Table 16.3 Sweet taste reference scales used by Schiffman et al. (1995) and alternative scales (5 and 7 equidistant points) according to Equation 16.22 in the appendix (Section 22.8). DuBois scale % sucrose sweet 2.0 5.0 7.5 10.0 12.0 16.0
2.0 5.0 7.5 10.0 12.0 15.0
Alternative scale 1 % sucrose sweet 2.0 5.3 8.8 12.4 16.0
2 5 8 11 14
Alternative scale 2 % sucrose sweet 2.0 4.2 6.5 8.8 11.2 13.6 16.0
2 4 6 8 10 12 14
370
Optimising sweet taste in foods
Fig. 16.9
Schematic diagram illustrating the possible minimization of an off-taste for a given synergistic binary blend of sweeteners (see text).
tastes, and so on). Then, establishing a cos U value for each off-taste and using them to interpolate the corresponding values of all off-tastes in the whole range of sweetness considered. The texture sensitivity in the solid state and in solution could also be studied in a similar way. Unfortunately, we did not find any off-taste or texture sensitivity published data for sweetener mixtures. For example, in Ayya and Lawless (1992) there are several drawings with responses of sweetness and bitterness produced by several sweetener mixtures (time-intensity technique), but without either numerical data or confidence intervals. It is certainly a field of experimentation to be implemented in order to improve the sweet quality of blended sweeteners. Figure 16.9 is an example of how to manage together sweet taste and off-taste responses in mixtures, when the latter is available. Based on the data for the acesulfame-K/aspartame mixture already applied in Fig. 16.8, the X-axis is the proportion of relative concentrations of components eliciting an iso-sweet curve. In the general case, the Y-axis should be the interpolated responses of off-tastes using the U model applied to a limited number of experimental cases. In the absence of such values, we applied for the Y-axis, just for illustration purposes, the sum of relative concentrations in Fig. 16.8. We observe a minimum of supposed off-taste in a wide range of relative concentrations in the middle of the diagram. 16.6.4 Other properties of artificial sweeteners An important aspect of artificial sweeteners present in the market should be the persistence of their sensory qualities after cooking (sweetness, texture, absence of off-tastes). The same tools used for the initial sensory properties can be applied to the study of their persistence after cooking.
Analysing and predicting synergy in sweetener blends 371 The artificial sweeteners must also, of course, not present unhealthy side effects. For given components supposed totally safe, the harmlessness of the commercial products can in some way be improved via synergistic sweetness, which implies a diminution of intake doses (precaution principle).
16.7
Acknowledgements
The author is grateful to Dominique Valentin for fruitful discussions and critical reading, and to Pierre HeÂricourt for writing the MIG program in EXCEL applied in the present study (both from the Centre des Sciences du GouÃt of Dijon). He also sincerely thanks William Spillane for his help in writing the English text.
16.8
References
(1966), Pers Com, Gaz de France, Research Department, 361 avenue President Wilson, F-93210 Saint-Denis, France. AYYA N and LAWLESS H T (1992), `Quantitative and qualitative evaluation of highintensity sweeteners and sweetener mixtures', Chem Senses, 17, 245±259. BEIDLER L M (1954), `A theory of taste stimulation', J Gen Physiol, 38, 133±139. BEIDLER L M (1971), `Taste-receptor stimulation with salts and acids', in Beidler L M, Handbook of sensory physiology IV/2, Taste, New York, Springer-Verlag, 200± 220. BERENBAUM M C (1985), `The expected effect of a combination of agents: the general solution', J Theor Biol, 114, 413±431. BERENBAUM M C (1989), `What is synergy ?', Pharmacol Rev, 41, 93±141. BERGLUND B, BERGLUND U, LINDVALL T and SVENSSON L T (1973), `A quantitative principle of perceived intensity summation in odor mixtures', J Exper Psychol, 100, 29±38. BLANCHARD J (1976), `Relations between the odour of a gas and its mercaptan content', Gas (Rome), 26, 287±309. BORELLI F and ANGLERAUD O (1965), `Some practical investigations about odorizing. SmelI intenseness of the various gases', Gas (Rome) Suppl, 7±8, 27 pp. CALLEGARI P (1998), `Aspects conceptuels, expeÂrimentaux et calculatoires de la qualite olfactive', Thesis University of Bourgogne, Dijon, France, 201 pp. DE GRAAF C and FRIJTERS J E R (1986), `A psychophysical investigation of Beidler's mixture equation', Chem Senses, 11, 295±314. DEVOS M, ROUAULT J and LAFFORT P (2002), Standardized olfactory power law exponents in Man, Dijon (France), Editions Universitaires de Dijon, 128 pp. ANGLERAUD O
DUBOIS G E, WALTERS D E, SCHIFFMAN S S, WARWICK Z S, BOOTH B J, PECORE S D, GIBES K,
and BRANDS L M (1991), `Concentration-response relationships of sweeteners: A systematic study', in Walters D E, Orthoefer F T and DuBois G E, Sweeteners. Discovery, molecular design and chemoreception. ACS Symposium Series 450, Washington DC, American Chemical Society, 261±276. CARR B T
372
Optimising sweet taste in foods
(1987), Pers. Com, Department of Human Nutrition, Agricultural University, 6703 BC Wageningen, The Netherlands. FRIJTERS J E R and DE GRAAF C (1987), `The equiratio taste mixture model successfully predicts the sensory response to the sweetness intensity of complex mixtures of sugars and sugar alcohols', Perception, 5, 615±628. FRIJTERS J E R, DE GRAAF C and KOOLEN H C M (1984), `The validity of the equiratio taste mixture model investigated with sorbital-sucrose mixtures', Chem Senses, 9, 241± 248. FRIJTERS J E R and OUDE OPHUIS P A M (1983), `The construction and prediction of psychophysical power functions for the sweetness of equiratio sugar mixtures', Perception, 12, 753±767. HEÂRICOURT P (2005), Pers Com, Centre des Sciences du Gou à t, Dijon, France. HILL A V (1913), `The combinations of haemoglobin with oxygen and with carbon monooxide', Biochem J, 7, 471±480. HYMAN A M and FRANK M E (1980), `Effects of binary taste stimuli on the neural activity of the hamster chorda tympani', J Gen Physiol, 76, 125±142. È STER E P (1969), `Intensity in mixtures of odorous substances)', in Pfaffmann C, KO Olfaction and Taste III, New York, Rockefeller University Press, 142±149. LAFFORT P (1966), `Recherche d'une loi de l'intensite odorante supraliminaire, conforme aux diverses donneÂes expeÂrimentales', J Physiol Paris, 54, 551. LAFFORT P (1977), `Some aspects of molecular recognition by chemoreceptors', in Le Magnen J, Olfaction and Taste VI, London, IRL Press, 17±25. LAFFORT P (1994a), `Olfactory communication and facets of odorousness', in Martin G and Laffort P, Odors and deodorization in the environment, New York, VCH Publ, 105±141 (translated from French, Paris 1991, Tec-Doc Lavoisier, 97±130). LAFFORT P (1994b), `Synergy and inhibition in olfaction' in Martin G and Laffort P, Odors and deodorization in the environment, New York, VCH Publ, 185±211 (translated from French, Paris 1991, Tec-Doc Lavoisier, 169±194). LAFFORT P (1994c), `The application of synergy and inhibition phenomena to odor reduction', in Vigneron S, Hermia J and Chaouki J, Characterization and control of odors and VOC in the process industries, Amsterdam, Elsevier, 105±117. LAFFORT P and DRAVNIEKS A (1982), `Several models of supra-threshold quantitative olfactory interaction in humans applied to binary, ternary and quaternary mixtures', Chem Senses, 7, 153±174. LAFFORT P, ETCHETO M, PATTE F and MARFAING P (1989), `Implications of power law exponent in synergy and inhibition of olfactory mixtures', Chem Senses, 14, 11±23. LAFFORT P, WALSH R M and SPILLANE W J (2002), `Application of the U and ÿ0 models in binary sweet taste mixtures', Chem Senses, 27, 511±520. NAHON D F, ROOZEN J P and DE GRAAF C (1998), `Sensory evaluation of mixtures of sodium cyclamate, sucrose and an orange aroma', J Agric Food Chem, 46, 3426±3430. PATTE F and LAFFORT P (1979), `An alternative model of olfactory quantitative interaction in binary mixtures', Chem Senses Flavor, 4, 267±274. PATTE F, ETCHETO M, MARFAING P and LAFFORT P (1989), `Electroantennogram stimulusresponse curves for 59 odourants in the honey-bee Apis mellifica', J Insect Physiol, 35, 667±675. SALES M (1958), `Odour and odorization of gases', 7th Intern. Conf. Gases (Rome), 36, 166 pp. SCHIET F T and FRIJTERS J E R (1988), `An investigation of the equiratio-mixture model in olfactory psychophysics: a case study', Percept Psychophys, 44, 304±308. FRIJTERS J E R
Analysing and predicting synergy in sweetener blends 373 and GRAHAM B G (1995), `Investigation of synergism in binary mixtures of sweeteners', Brain Res Bull, 38, 105±120. SCHIFFMAN S S, SATTELY-MILLER E A, GRAHAM B G, BOOTH B J and GIBES K M (2000), `Synergism among ternary mixtures of fourteen sweeteners', Chem Senses, 25, 131±140. SEÂREÂE N and LAFFORT P (1997), `Fitting comparison of the ÿ (Gamma) and ÿ-Vectorial models applied to experimental data in olfactory, gustatory and pharmacological mixtures', Chem Senses, 22, 221±222. STEVENS S S (1936), `A scale for the measurement of a psychological magnitude loudness', Psychol Rev, 43, 405±416. SuÈHNEL J (1993), `Evaluation of interaction in olfactory and taste mixtures', Chem Senses, 18, 131±149. TATEDA H (1967), `Sugar receptor and -amino acid in the rat', in Hayashi T, Olfaction and Taste II, Oxford, Pergamon Press, 383±397. THOMAS-DANGUIN T and CHASTRETTE M (2002), `Odour intensities of binary mixtures of odorous compounds', Comptes Rendus Biologies, 325, 767±772. THOMAS-DANGUIN T and LAFFORT P (1998), `The ÿ0 model applied to sweet and other taste mixtures', TOSTQ workshop, Wageningen, The Netherlands, 11 December. SCHIFFMAN S S, BOOTH B J, CARR B T, LOSEE M L, SATTELY-MILLER E A
16.9
Appendix: precise technical details
Building of Fig. 16.2 Its principle is simple: relating a perceptive scale to a physical scale (relative concentrations). In practice it is not that simple, since the elements are dispersed into tables 2, 3, 4 and 5 in Schiffman et al. (1995) and table 1 of Schiffman et al. (2000). · The relative concentrations refer to the concentrations equisweet to 2% sucrose. The relative concentrations respectively equisweet to 3%, 5% and 7% of sucrose (we call the points 3, 5 and 7) are obtained combining the absolute concentrations equisweet to 3%, 5% and 7% of sucrose reported in table 2 in Schiffman et al. (1995) and the concentrations equisweet to 2% of sucrose reported in table 1 of Schiffman et al. (2000) (these later values are also reported in Table 16.1). · Twice these concentrations are added to the previous ones; we call the points 6, 10 and 14. · By definition, the perceived intensities at the points 3, 5 and 7 are also 3, 5 and 7 SIR. The perceived intensities at the points 6, 10 and 14 are given in the diagonals of tables 3, 4 and 5 of Schiffman et al. (1995) (sweeteners added to themselves). Whereas points 3, 5 and 7 have been established using the technique of sucrose equisweet strictly speaking, points 6, 10 and 14 have been established using `sweetness intensity rating', i.e. a memorized matching, after training of
374
Optimising sweet taste in foods
Fig. 16.10 Perceptive scale SIR (sucrose intensity rating) vs. physical scale (multiples of 2% of sucrose) in Cartesian coordinates. Open circles: experimental points; solid line: theoretical curve according to equation 22; dotted line: physical straight line. For SIR = 7, the two curves are practically identical. The two experimental points out of the dotted line are for SIR equals 9.12 and 12.30, corresponding to the relative concentrations 5 and 7 (i.e. the so-called points 10 and 14).
subjects on a physical scale of solutions of sucrose in the range 2%±12% and in a modified scale in the range 12%±16%.10 The fact that points 3, 5 and 7 of the sucrose scale are both physical and perceptive, and that points 6, 10 and 14 are only perceptive, is only apparently contradictory as it can be seen in Fig. 16.10. In this figure the seven experimental points for sucrose (6 1) are reported and the continuous curve corresponding to the power law exponent of sucrose, i.e. 0.935 (rounded-off to 0.94 in Table 16.1). Csucrose 0:935
16:22 SIRsucrose 2 C2sucrose The paradox of Fig. 16.10 is the observed disconnection from the curve of the experimental points 10 and 14, since from the training procedure of the subjects this is only expected this disconnection for point 14. One can question the interest of a training scale for subjects lying on Equation 16.22, rather than the method used by Schiffman et al. (1995) (see Section 16.6.1).
10. `Before evaluating the binary mixtures, the trained panelists tasted sweet taste references according to the method used by DuBois et al. (1991): 2 sweet (2% sucrose), 5 sweet (5% sucrose), 7.5 sweet (7.5% sucrose), 10 sweet (10% sucrose), 12 sweet (12% sucrose) and 15 sweet (16% sucrose).'
17 Bulk sweet tasting compounds in food product development M. Lindley, Lindley Consulting, UK
17.1
Introduction
High potency sweeteners are essentially mono-functional ingredients; they deliver sweetness at low concentration and provide no physical functionalities to foods and beverages. Although high potency sweeteners are very important and successful ingredients, their use is therefore largely restricted to those products that contain substantial amounts of water. This restriction is mainly because the physical functionalities traditionally delivered by bulk sweeteners are of limited importance to the physical development of such product systems. There are some exceptions to this generality with, for example, high potency sweeteners being used successfully in chewing gums and other selected very low moisture confectionery systems, but the great majority of applications for high potency sweeteners are in applications where the bulk normally supplied by sucrose or other carbohydrate sweeteners is replaced by water. Because high potency sweeteners, by definition, are used at very low concentrations in foods, they do not affect the physical characteristics of foods to any significant degree. They have no role to play in the development of structure or texture, nor do they influence such physical parameters as freezing points, boiling points or water activity. It is for these reasons, therefore, that this chapter is primarily concerned with the contribution of sugars to product development, discussing the influence of bulk carbohydrate and carbohydrate-derived sweet tasting compounds on the physical characteristics of food products. Physical characteristics of foods that are influenced by soluble carbohydrates, including appearance, colour, texture and those physical characteristics that control shelf-life in all its aspects, such as water activity, osmotic pressure and vapour pressure, are reviewed. Nutritive sweeteners discussed are sucrose, glucose syrups, fructose/fructose syrups and some sugars
376
Optimising sweet taste in foods
currently of minor commercial relevance such as isomaltulose and ,-trehalose. The physical characteristics of these sweeteners are described. Those properties of particular value in the development of specific foods are highlighted.
17.2
Characteristics of bulk sweeteners
All of the bulk sweeteners used by the developed food industry, including the free sugars and their reduced forms, the polyols, display a range of characteristics the magnitudes of which influence whether or not a particular bulk sweetener will be selected for any specific application. To exemplify, all of the soluble bulk sweeteners elicit sweetness, but their relative sweetness ranges from that of fructose, generally considered to be the sweetest of the un-modified carbohydrates, down to that of perhaps lactose, a low sweetness sugar. The sweetness potency of these sugars, relative to sucrose whose potency is normally taken to be 1, range from c. 1.2 for fructose down to c. 0.25 for lactose. In other words, there is a four- to five-fold change in relative sweetness between the sweetest and least sweet bulk sweetener. Similarly, all of the bulk sweeteners are soluble in water, but again absolute solubility covers a wide range of c. 85 g/ 100 g solution for fructose down to c. 25 g/100 g solution for lactose, both measured at 40ëC. Some of the polyols are even more soluble than fructose under these conditions. All sugars and polyols deliver many of the same qualitative functionalities although, as noted, the quantitative nature of these functionalities differs widely. Therefore, it seems sensible to discuss these functionalities in general terms and then to describe how key food industry sectors exploit the functionalities of individual sugars and polyols in their product development programmes. The characteristics of the bulk sweeteners of relevance to the food industry are summarised in Table 17.1. These characteristics include mainly sensory and physical functionalities, although reference is also made to microbiological characteristics. There is no detailed discussion of the nutritional characteristics of these ingredients because a review of the various metabolic routes whereby each different sugar and polyol is metabolised or handled in the body is outside the scope of this chapter. However, all sugars deliver c. 4 k.calories/g and the polyols, by definition in Europe, are all considered to deliver 2.4 k.calories/g. Actual caloric delivery of the polyols (excluding erythritol which delivers c. 0.2 k.calories.g) ranges between around 2.0 and 3.2 k.calories/g. The nutritional characteristics of the sugars are of importance largely because they offer justifications for the development of alternative ingredients. Whereas their energy delivery was historically important and valuable to Western populations, today this caloric delivery has provided impetus to the search for, identification and development of lower calorie alternatives, such as the high potency sweeteners and bulking agents, to permit the preparation of reducedcalorie foods. Obviously, the polyols have also been developed in response to `sugar-free' or `reduced-sugar' market drivers, partly because of the calorie
Bulk sweet tasting compounds in food product development Table 17.1
377
Characteristics of bulk sweeteners (from Nicol, 1979)
Category
Characteristic
Nutritional
Energy
Sensory properties
Sweetness Flavour Texture Appearance
Physical functionalities
Solubility and viscosity Osmotic and vapour pressures Hygroscopicity/humectancy/water activity
Microbiological properties
Preservation Fermentation
content of sugars, but also because of their potential to be metabolised by the oral microflora and hence be involved in the aetiology of dental caries.
17.3
Sensory properties
17.3.1 Sweetness There is a tendency to assume that all carbohydrates deliver similar, clean sweet tastes, probably because this is the taste elicited by sucrose. However, although those sugars that are used commercially elicit few, if any, significant off-tastes, not all carbohydrate sweeteners elicit pure sweetness; bitterness is actually often associated with simple sugars and this may be due to polar and lipophilic features as well as ring size and shape (Birch and Lee, 1976). Small molecular weight carbohydrates that elicit both sweetness and bitterness include the monosaccharide D-mannose in its -pyranose conformation and the disaccharides gentiobiose (6-O- -D-glucopyranosyl-D-glucopyranose) and , -trehalose. The similarity of the tastes delivered by a selection of small molecular weight, largely commercial carbohydrates has been studied (Schiffman et al., 1979). These workers demonstrated that most sugars and polyols are judged similar to one another. The sugars evaluated in this study were the monosaccharides glucose, fructose, sorbose and xylose, the di-saccharide maltose and the polyols sorbitol and xylitol. In general, these sugars were described as eliciting sweetness that developed relatively quickly with less aftertaste than the other sweet stimuli evaluated in the study. Their tastes were considered to be `good', `natural' and `syrupy'. 17.3.2 Flavour Empirical observations confirm that perceived sweetness enhances many flavours. Some high potency sweeteners exhibit clear flavour enhancing
378
Optimising sweet taste in foods
properties, with, for example, aspartame effectively enhancing fruit and coffee flavours (Beck, 1978), the protein sweetener thaumatin modifying many flavour types (Higginbotham, 1983) and neohesperidin dihydrochalcone being particularly effective in combination with fruit flavours (Lindley et al., 1993). Interestingly, there are claims that some of these effects are perceived when the sweetener is added at sub-threshold concentrations, thus implying that flavour modification is, at least in part, structurally rather than sensorially mediated. Sugars also modify and enhance flavours, with effects being noted at sub- and supra-threshold use levels for sucrose. Sucrose is claimed to improve the flavour of sauces, soups, cheese, mayonnaise and yoghurt at use levels below sweetness threshold and even finds application in cured meats where, through flavour enhancement, product acceptability is enhanced (Nicol, 1982). Fructose modifies and enhances many fruit flavours (Osberger, 1991). Isomaltulose and isomalt are said to modify flavours, probably through their ability to mediate bitterness (Irwin and Straeter, 1991), and similar functionality has also been claimed for the disaccharide ,-trehalose and the keto-hexose D-tagatose. Although some of the sugars do induce flavour enhancement in selected foods and beverages, their more important flavouring role is contributing to the generation of flavour by caramelisation or through participation in Maillard reactions. Since colour of foods and beverages is frequently the first attribute recognised by the senses, the relevance of sugars-induced browning reactions to food and beverage acceptability should be obvious. Caramel colours are amorphous, brown materials resulting from the carefully controlled heat treatment of sugars in the presence of small quantities of acids, alkalis or salts. Caramelisation has been carried out for as long as food has been cooked, caramel colours first gaining commercial importance as additives to brewery products. Main carbohydrates used in the manufacture of caramels are glucose, invert sugar, malt syrups, sucrose and high DE glucose syrups. Flavour may also be generated by the complex interactions between sugars and amino acids known as the Maillard reaction. This reaction is central to food chemistry because of a number of features, including the production of colour, potential reductions in nutritional value that result from removal of possibly essential amino acids, generation of potentially toxic compounds as well as the generation of flavours and off-flavours. Flavour compounds generated during the Maillard reaction are extremely complex and include sugar dehydration and fragmentation products such as furans, pyrones, aldehydes and ketones, acids as well as amino acid degradation products and flavours produced by further interactions. The prevailing pH influences the types of compounds produced when inverted sucrose, glucose and/or fructose are heated, as summarised in Table 17.2. A flavour effect often poorly understood, and one that can have a major impact on product quality, particularly of beverages, is the influence of the solute (sucrose, glucose or fructose syrups) on flavour volatility. In general, the partition coefficient of a volatile substance in solution increases with addition of solute due to salting out effects. However, sometimes other competing effects,
Bulk sweet tasting compounds in food product development Table 17.2
379
Products of heat and acid or alkali on glucose and/or fructose
Acid
Alkali
Acid and alkali
Hydroxymethyl furfural formation
Lactic acid Transformations Re-arrangements
Maillard reaction induced colours, flavours and condensation products
such as intermolecular associations between the volatile and the solute may decrease volatility (King, 1983). Flavour volatility is generally influenced by sucrose concentration, thus illustrating the potential qualitative impact of removing sucrose or other sweet carbohydrate from beverage formulations in the preparation of low-calorie, `light' or `diet' versions. In fact, a number of studies have shown that reducing sucrose contents will reduce flavour volatility of compounds such as n-hexylacetate and 2-heptanone (Nawar, 1971). Interestingly, the volatility of flavour compounds such as heptanal are increased with decreasing sucrose content, thus further demonstrating the complexity of the formulation challenge of removing sucrose from beverage formulations. The potential effects on flavour compound volatility that can result from removing sucrose are illustrated in Table 17.3 (Marinos-Kouris and Saravacos, 1975). 17.3.3 Texture The textures of many food products are influenced by soluble carbohydrates. Sucrose, in particular, contributes extensively to the creation of a variety of familiar food textures, ranging from the crunchiness of separate crystals in a decorative topping used in selected bakery and confectionery applications, through the grittiness of meringues, the pastiness of fudge and icing, the smooth glass of hard boiled confectionery products, the light alveolation of cakes, pectin gels in jams and jellies and mouthfeel in syrups, sauces and beverages. Sucrose is therefore clearly an example of a multi-functional ingredient and while many other sugars are able to influence product texture in similar manners, none is Table 17.3 Relative volatilities of organic compounds in aqueous sucrose solutions at 25 ëC (from Marinos-Kouris and Saravacos, 1975) Compound
Methyl anthranilate n-amyl alcohol n-hexanol 1-butanone 3-pentanone Ethyl acetate
Relative volatilities at sucrose concentration (%; w/v) 15.0
0.0
1.27 24.7 36.0 96.0 85.0 265.0
1.0 23.0 31.0 76.0 77.0 225.0
380
Optimising sweet taste in foods
able to display the extreme versatility of sucrose. As has been noted, however, this is not altogether surprising given the reality that formulations of all these foods and beverages were originally developed using sucrose, the most abundantly available small molecular weight carbohydrate. Now, sucrose is frequently blended with other, less expensive sugars such as glucose syrups and high fructose corn syrups is some of these applications. On occasion, blending glucose syrup with sucrose can lead to improvements in product performance. In hard-boiled confectionery products, for example, this is due to the ability of glucose syrup to inhibit or retard the crystallisation of sucrose. 17.3.4 Appearance Food is bought and eaten in part because of its appearance. Crystalline carbohydrates such as sucrose and fructose, sprinkled on bakery and confectionery items in particular, can add significantly to the visual appeal of products, thus enhancing acceptability. In solution, the participation of sugars in caramelisation and Maillard browning reactions has fundamental influences on the acceptability of many cooked foods, particularly baked goods.
17.4
Physical functionalities
17.4.1 Solubility and viscosity The fact that the sugars are generally quite soluble ingredients underpins many of their physical functionalities; it is their relationships with water that control the magnitudes of the physical effects observed. With the high solubility of most simple sugars, lack of solubility rarely creates technical problems surrounding the use of sugars in foods. However, when some food products are prepared with high solids contents, especially when containing high concentrations of sucrose, glucose or lactose, adequate precautions do need to be taken to prevent crystallisation occurring. In such circumstances, maltodextrins and low DE glucose syrups can be used to good effect. Generally speaking, the greater the number of molecular species in a sugars mixture, the harder it will be for any single molecular component to crystallise. Long chain oligomers have the added advantage of acting as a diluent matrix for simple sugars. Therefore, even with the high intrinsic solubility of sucrose, in an application such as hard boiled confectionery crystallisation can and does occur over time, resulting in the appearance of `graining'. In these applications, formulation with a low DE glucose syrup effectively inhibits sucrose re-crystallisation. An additional example of the importance of sugar solubility is lactose. Lactose is one of the sugars exhibiting relatively low aqueous solubility and this can lead to the sugar crystallising in dried milk powders and in products such as ice cream and sweetened condensed milk. The viscosities of pure solutions of simple sugars are Newtonian. This is of operational significance in the food industry for the movement of liquid sugars, both into and out of transport containers and within food and beverage
Bulk sweet tasting compounds in food product development
381
production facilities. In finished products and at practical use concentrations the mouthfeel generated in products such as soft drinks is invariably smooth. Viscosity also plays an important role in the regulation of crystallisation rate, particularly important in hard-boiled confectionery products where the degree of supersaturation can be so high that diffusion is severely limited and crystallisation rates are correspondingly slowed. 17.4.2 Colligative properties Collectively, the colligative properties of sugars influence food preparation and stability to a greater extent than do other sensory and functional properties. The presence of soluble carbohydrate molecules in solution gives rise to osmotic and vapour pressure phenomena. Sugars in solution generate high osmotic pressures and this is the major factor behind the use of sugars as preservatives in, for example, jams and jellies. Boiling point elevation and freezing point depression are influenced by the concentration of sugars in solution and their molecular weight. Closely allied with vapour pressure effects is the water activity or equilibrium relative humidity (ERH) generated over sugars in crystalline form or in solution. The osmotic pressures generated by sugars in solution are often used in food manufacture to control microbiological growth in, for example, jams and preserves (Flanyak, 1991). Osmotic pressures generated are directly proportional to the molecular concentration of sugars in solution and this infers that smaller molecular weight sugars such as the monosaccharides glucose and fructose, at equal weight concentrations, are more effective than larger molecular weight sugars such as the disaccharide sucrose. Similarly, high DE glucose syrups are more effective generators of osmotic pressure than low DE syrups. Interestingly, there is a linear relationship between osmotic pressure and DE (Kearsley and Birch, 1977) and hydrogenation of glucose syrups has no impact on the osmotic pressure it can generate, a factor utilised to allow the estimation of hypothetical DE of such syrups. Similarly, the colligative properties boiling point elevation and freezing point depression are also dependent on the concentration and molecular weight of the sugars in solution. For sucrose, the relationships between concentration and effect may be expressed as follows. At atmospheric pressure, 1 mole (342.3 g) sucrose dissolved in 1 kg water will depress freezing point by approximately 2ëC and, under the same atmospheric conditions, 1 mole of sucrose in 1 kg water will raise boiling point by approximately 0.5ëC. In practical terms, the relationships between concentration and the impact on boiling and freezing points are presented in Table 17.4. The relationships between molecular weight and boiling point elevation/ freezing point depression mirror those between molecular weight and osmotic pressure. Thus, at equal weight concentrations, lower molecular weight sugars are more effective than higher molecular weight sugars. These effects influence the selection of glucose syrups for use in both confectionery and frozen desserts.
382
Optimising sweet taste in foods
Table 17.4 Impact of sucrose on boiling point and freezing point Sucrose concentration (ëBrix) 10 20 30 40 50 60
Freezing point depression (ëC)
Boiling point elevation (ëC)
0.61 1.01 1.50 2.6 4.5 Not measured
0.15 0.25 0.40 0.70 1.20 2.0
In confectionery products, selection of the optimum blend of sucrose, invert sugar and glucose syrup can lower the boiling point of the mix significantly, thus leading to less heat-induced browning and also potentially a cheaper product due to lower energy utilisation. In contrast, in the formulation of ice creams and frozen desserts, lower DE glucose syrups are often preferred so as to minimise the depression of freezing point in these products (Dziedzic and Kearsley, 1984). 17.4.3 Hygroscopicity, humectancy and water activity A key property of small molecular weight carbohydrates is their ability to tie up water. A humectant is defined as an ingredient that has the ability to resist changes in moisture content, and hygroscopicity is the property of absorbing water from the atmosphere. At any given vapour pressure and temperature, sugars will neither attract nor lose moisture from their surroundings, and this is defined as the equilibrium relative humidity (ERH) of the sugar. Sugars in an atmosphere in which there is more moisture than there is in the sugar will attract/ absorb water; sugars at a higher ERH than their surroundings will lose moisture. Under fixed conditions, the ERH of a sugar solution is inversely proportional to the molecular concentration of solids in the aqueous phase. In other words, the higher the molecular concentration of sugars, the greater the reduction in ERH. Thus, smaller molecular weight sugars of higher solubilities are more effective humectants. Water activity (Aw) is defined as equal to the partial vapour pressure of the food or solution divided by the saturation vapour pressure of water. Of the sugars, sucrose frequently allows a much lower Aw for the same moisture content and this retards the rates of many reactions, including microbial spoilage. The impact of sugars on water activity can be very complex. For example, a 44ëBrix sucrose solution (containing 56% water) has the same Aw (0.80) as a starch/water mixture with a water content of only 20%. This explains why one major food use for sucrose is in food preservation. 17.4.4 Other properties Different carbohydrates exert varying effects on the sugar-acid-high methoxy pectin ratios used in the manufacture of jams and preserves. Glucose and
Bulk sweet tasting compounds in food product development
383
maltose enhance gelling properties whereas longer chain oligosaccharides interrupt pectin networks, thus having a detrimental effect on gel formation (Howling, 1979). Careful selection of glucose syrups can help to ensure that the resulting product will have enhanced stability against gel syneresis, thus explaining the selection of high maltose glucose syrups in jam and preserve manufacture. In contrast, low DE glucose syrups are more suitable for products such as instant whips, dream toppings and other foamed confections. These low DE syrups are able to entrap air, thus stabilising foam networks.
17.5
Bulk sweeteners
17.5.1 Sucrose Sucrose is often described as the purest bulk food commodity, with World production approximating 120 million tonnes per annum. So-called granulated sugars are produced at purities in excess of 99.95% sucrose, and even the `brown' sugars are produced to high levels of purity with Demerara sugar, for example, typically containing 97.5% sucrose. Sucrose is used as the normal reference for sweetness against which all other sweeteners are compared, both from the standpoint of assessing relative potency and assessing taste quality. This is partly because of its ready availability, but also because it elicits a singularly pure sweetness free from secondary flavours or aftertastes. In fact, sucrose is often described as delivering the most acceptable sweet taste quality of all the sugars and high potency sweeteners, but it still remains conjectural whether this is a cause or an effect of the predominance of sucrose as a sweetener in the diet (Nicol, 1979). Sucrose plays a multi-functional role in the food industry. Many of the food products consumed today as part of a normal diet were, at least to some degree, originally created around the physical functional properties delivered by sucrose. Whether this was actually a consequence of sucrose's ready availability, or its intrinsic sweetness : bulk ratio or due to the delivery of its many physical functionalities is something of a moot point; the reality is that a number of large and successful food industry sectors depend for their existence on the myriad properties of sucrose. Sucrose also contributes subtle effects in many food applications, particularly those related to flavour enhancement and retention. For example, there are a number of reports describing techniques to immobilise flavours using sucrose (Mitchell and Stahl, 1974; Malizia and Mitchell, 1979). All these techniques exploited the ability of sucrose to encapsulate volatile flavour compounds within so-called micro-regions by dehydration of a sucrose melt. In spite of this great versatility with respect to contributing macro and micro functionalities, and its general importance to today's food and beverage industries, sucrose is not a perfect sweetener. Although not an expensive sweetener, cheaper sources of sweetness are available and so this relatively high cost-in-use resulted in the development of sweeteners from starch, namely, glucose syrups
384
Optimising sweet taste in foods
and high fructose corn syrups. The cariogenicity of sucrose stimulated markets for the many polyol sweeteners and of course its energy delivery has prompted the development of many high potency, zero-calorie sources of sweetness. 17.5.2 Other disaccharides In addition to sucrose, the disaccharides lactose, maltose, ,-trehalose and isomaltulose are used to varying degrees as food ingredients. Lactose, or milk sugar, is only approximately one-third as sweet as sucrose and so is not used for its sweetening properties. It is extracted from milk and is used as a bulking agent in some applications and as an excipient in pharmaceutical products. Maltose is also not widely used as a pure sugar, but is normally added as a component of high maltose glucose syrups. It can inhibit the retrogradation of starch, prolonging the shelf life of baked goods, and it also can stabilise icings and glazes, but at lower sweetness delivery than sucrose (Johnson, 1976). ,-Trehalose (trehalose) is a naturally occurring disaccharide consisting of two glucose moities linked through their respective anomeric carbon atoms (1,1) by an -glycosidic bond. This sugar has been known for many years for its ability to protect organisms against complete dehydration during drought. It apparently protects both the structural and functional integrity of membrane vesicles against dehydration (Roser, 1991). The glycosidic linkage between glucose moieties in trehalose is stable. Thus, the sugar does not readily undergo thermal or hydrolytic breakdown over a broad pH range in food and beverage applications. As a non-reducing sugar, it is not subject to Maillard browning reactions. Under conditions where trehalose may hydrolyse, glucose is released. It can be used to replace some or all of the sucrose in products where it is desirable to reduce the level of sweetness, thus leading to a more balanced and improved flavour profile. Finally, isomaltulose is a reducing glucose-fructose disaccharide in which glucose and fructose are linked through their 1 and 6 carbon atoms, respectively. It is prepared by the enzymatic rearrangement of sucrose using the enzyme sucrose-glucosylmutase, itself obtained from the microorganism Protaminobacter rubrum. During the enzymatic conversion, the 1,2 linkage between glucose and fructose in sucrose is rearranged to a 1,6-glycoside linkage. It has been used commercially in Japan since 1985 where it finds significant markets in a range of products including yoghurt, chewing gum and a number of sugar confectionery products (Clarke, 1991). It has recently been accorded novel food status in Europe. 17.5.3 Glucose and fructose syrups The mono-saccharide glucose (or dextrose) is used in food systems, although not widely. Glucose syrups are, however, very commonly used sources of sweetness in both foods and beverages. They are prepared by acid, acid-enzyme or enzyme
Bulk sweet tasting compounds in food product development
385
conversion of starch and their compositions can be adjusted so as to suit a range of applications. High maltose syrups, in particular, are able to substitute for sucrose in many applications. In the 1960s, enzyme catalysed isomerisation of glucose to fructose made possible the development of high fructose syrups, with equivalent sweetening power to sucrose, but at significantly reduced cost. Glucose syrups are used in a variety of applications, including sugar confectionery, frozen desserts and ice creams, jams and preserves and many bakery applications. Another major application is in alcoholic beverages where its conversion to alcohol, through fermentation, means that it is a preferred ingredient. As has been noted, adjustment of glucose syrup composition through process development ensures its optimal suitability for this broad range of applications. In contrast to glucose syrups, which are largely used for the functionalities they deliver to foods, it is the sweetness of high fructose corn syrups that is of primary importance. With fructose syrups containing 55% fructose matching the sweetness of sucrose at lower cost, they comprise the major beverage sweetener in the USA. In Europe, however, glucose-fructose (high fructose) syrup markets are restricted due to the European Union quota system. Crystalline fructose is marketed as a speciality carbohydrate sweetener and is used in applications that capitalise on its `natural' and `healthy' image. It enhances flavours, particularly of the fruit variety and so is used in a number of fruit flavoured beverages and desserts. It has a low glass transition temperature which explains its susceptibility to moisture absorption because of its high hygroscopicity as well as its tendency to generate brown pigments in heat induced browning reactions. These phenomena mean that care must be exercised in using pure fructose in food systems.
17.6
Conclusions
Carbohydrate sweeteners deliver much more than sweetness to many important foods and beverages. They exert major influences on the structure of food, largely due to their use at relatively high concentrations in many products. While sucrose remains the most important of the bulk sweeteners, many other sugars make important physical contributions, delivering specific functional benefits or providing equivalent functionalities to sucrose, but at reduced cost. Soluble carbohydrates all deliver sweetness and that sweetness has an enhancing effect on many flavours. However, the carbohydrate sweeteners differ in their effective delivery of functional properties such as boiling point elevation, freezing point depression and humectancy, with many of the differences attributable to differences in molecular weight and solubility. The food industry has available a broad range of carbohydrate sweeteners with which to formulate. Although sucrose is generally considered to be the `gold standard' against which the performances of other sugars are compared, the other sugars discussed in this chapter all have established positions within
386
Optimising sweet taste in foods
the market. However, the development of yet more novel carbohydrate ingredients seems unlikely, partly because nature dictates that there are few other feasible structural options, but also because there are few if any obvious functionality gaps not filled by those sugars that are already commercially available.
17.7
Sources of further information and advice
During the last 30 years there have been a number of books published that cover this subject and various symposium proceedings also address aspects of the topic. Relevant academic symposia and their published proceedings that can be recommended include: Birch, G.G., Green, L.F. and Coulson, C.B. (1970) Glucose syrups and related carbohydrates, Amsterdam, Elsevier. Birch, G.G. and Green, L.F. (1973) Molecular structure and function of food carbohydrate, London, Applied Science Publishers. Birch, G.G. and Parker, K.J. (1979) Sugar: Science and technology, London, Applied Science Publishers. Birch, G.G. and Parker, K.J. (1982) Nutritive sweeteners, London and New Jersey, Applied Science Publishers. The following text-books are recommended for further study: Dziedzic, S.Z. and Kearsley, M.W. (1984) Glucose syrups: Science and technology, London, New York, Elsevier Applied Science. Honig, P. (1953) Principles of sugar technology, London, Elsevier. Marie, S. and Piggott, J.R. (1991) Handbook of sweeteners, Glasgow and London, Blackie. Mathlouthi, M. and Reiser, P. (1995) Sucrose: Properties and applications, Glasgow, Blackie Academic and Professional/Chapman and Hall. Nabors, L.O'B. and Gelardi, R.C. (1991) Alternative sweeteners, New York, Basel, Hong Kong, Marcel Dekker.
17.8
References
(1978) `Application potential for aspartame in low calorie and dietetic foods' in Dwivedi, B.K. Low calorie and special dietary foods, West Palm Beach, CRC Press, pp. 59±114. BIRCH, G.G. and LEE, C.-K. (1976) `Structural functions of taste in the sugar series: the structural basis of bitterness in sugar analogues', J. Fd Sci., 41, 1403±1407. CLARKE, M.A. (1991) `Non-sucrose carbohydrates' in Marie, S. and Piggott, J.R. Handbook of sweeteners, Glasgow and London, Blackie, pp. 52±71. DZIEDZIC, S.Z. and KEARSLEY, M.W. (1984) `Physicochemical properties of glucose syrups' BECK, C.I.
Bulk sweet tasting compounds in food product development
387
in Dziedzic, S.Z. and Kearsley, M.W. Glucose syrups: Science and technology, London, New York, Elsevier Applied Science, pp. 137±168. FLANYAK, J.R. (1991) `Effects of sucrose in confectionery processes', Manufact. Conf., October, 61±66. HIGGINBOTHAM, J.D. (1983) `Recent developments in non-nutritive sweeteners' in Grenby, T.H., Parker, K.J. and Lindley, M.G. Developments in sweeteners II, London, Applied Science, pp. 119±155. HOWLING, D. (1979) `The general science and technology of glucose syrups' in Birch, G.G. and Parker, K.J. Sugar: Science and technology, London, Applied Science, pp. 259±285. IRWIN, W.E. and STRAETER, P.J. (1991) `Isomaltulose' in Nabors, L.O'B. and Gelardi, R.C. Alternative sweeteners, New York, Marcel Dekker, pp. 229±308. JOHNSON, J.C. (1976) Specialized sugars for the food industry, Park Ridge, NJ, Noyes Data Corporation, 580 pp. KEARSLEY, M.W. and BIRCH, G.G. (1977) `Physical properties of hydrogenated glucose syrups', Staerke, 29, 425±427. KING, C.J. (1983) `Physical and chemical properties governing volatilisation of flavour and aroma compounds' in Peleg, M. and Bagley, E.B. Physical properties of foods, Westport, AVI Publishing Corp., pp. 399±421. LINDLEY, M.G., BEYTS, P.K., CANALES, I. and BORREGO, F. (1993) `Flavour modifying characteristics of the intense sweetener neohesperidin dihydrochalcone', J. Fd Sci., 58, 592±594 and 666. MALIZIA, P.D. and MITCHELL, W.A. (1979) `Sucrose volatile flavours', Canadian Patent 1,066,945. MARINOS-KOURIS, D. and SARAVACOS, G. (1975) `Volatility of organic compounds in aqueous sucrose solutions' Proceedings 5th Internat. Congress of Chem. Eng. Chem. Equip. Design and Automation, Prague, Czech Republic. MITCHELL, W.A. and STAHL, H.D. (1974) `Fixed volatile flavours and method for making same', United States Patent 3,787,592. NAWAR, W.W. (1971) `Some variables affecting composition of headspace aroma', J. Agric. Fd Chem., 19, 1057±1059. NICOL, W.M. (1979) `Sucrose and food technology' in Birch, G.G. and Parker, K.J. Sugar: Science and technology, London, Applied Science, pp. 211±230. NICOL, W.M. (1982) `Sucrose, the optimum sweetener' in Birch, G.G. and Parker, K.J. Nutritive sweeteners, London, Applied Science, pp. 17±35. OSBERGER, T.F. (1991) `Crystalline fructose' in Nabors, L.O'B. and Gelardi, R.C. Alternative sweeteners, New York, Marcel Dekker, pp. 219±248. ROSER, B. (1991) `,-Trehalose', Trends Fd Sci. Technol., 2, 166±169. SCHIFFMAN, S.S., REILLY, D.A. and CLARK, T.B. (1979) `Qualitative differences among sweeteners', Physiol. Behav., 23, 1±9.
18 Hydrocolloid±sweetener interactions in food products D. Cook, The University of Nottingham, UK
18.1
Introduction
Hydrocolloids are high molecular weight hydrophilic polymers which form colloidal dispersions in water. A wide range of naturally occurring and modified hydrocolloids are used by the food industry to control the physical properties of a multitude of products. Application areas include thickeners, emulsifiers, stabilisers, film-formers and moisture/fat barriers, encapsulants and fatreplacers. The most significant hydrocolloids in terms of market value are starches and modified starches, followed by gelatin. A number of non-starch hydrocolloids (`gums') are of increasing commercial value (e.g. xanthan, guar, carrageenans . . .). Since hydrocolloids are added to foods to modify their physical properties, it is not surprising that they can also affect the processes of flavour release and perception. Taking sweet taste as an example, the intensity of perceived sweetness is reduced if the rate of flux of sweetener to the taste buds drops. This may indeed be the case when hydrocolloids are added to thicken or gel systems. Furthermore, flavour perception may be directly modified by the perception of oral textural properties (e.g. mouthfeel, viscosity, creaminess); hence this chapter also considers `texture±sweetness' interactions which arise post-receptor (interaction in this context means that perception in one sensory modality is modified by the stimulation of another). The distinction between these two modes of action of hydrocolloid addition may be seen with reference to Fig. 18.1. In terms of sweetener±hydrocolloid interactions, the distal stimulus consists of the sweetener type(s) and concentration(s) together with the physical structure of the food. During eating, each of these stimuli is sensed by the respective
Hydrocolloid±sweetener interactions in food products
389
Fig. 18.1 A scheme of flavour perception from stimulus to perception illustrating the origins of hydrocolloid±sweetener interactions.
sensory systems, but the distal stimulus is modified by a range of factors before it is sensed at the receptors (proximal stimulus; Fig. 18.1). Food texture, inmouth mixing behaviour, mass transfer, human physiology and mastication patterns all combine to determine the rate at which sweetener reaches the taste buds. Likewise, food structure may be modified substantially in-mouth (indeed eating would not be particularly useful if it wasn't!). Hence the innate structure of a food material gives rise to a texture stimulus whilst it is being modified by factors such as mastication (oral shear), melting and mixing in-mouth and the actions of salivary enzymes. Subsequent to detection at the receptor level, signals from different sensory systems may interact at a neural or perceptual level, giving rise, for example, to texture±sweetness interactions. A key area of interest in sweetener±hydrocolloid interactions concerns the development of low-calorie foods using sugar replacement strategies. This area of research and development has been stimulated by health concerns related to conditions such as diabetes, weight gain, obesity-related disorders and dental caries. When formulating low sugar foodstuffs it is important to address the contribution of sugar to both sweetness and the physical body, or structure, of a product. Replacing each of these effects requires a delicate balance of ingredients and an understanding of the underlying sensory mechanisms. Thus a hydrocolloid is frequently used in addition to a high-potency sweetener to replace the physical or structural properties contributed by sugars.
390
18.2
Optimising sweet taste in foods
Hydrocolloid±sweetener interactions
18.2.1 Viscous liquid systems In view of the potential for multiple confounding effects when studying hydrocolloid±sweetener interactions (Fig. 18.1) the widely adopted scientific approach has been to investigate simple (model) systems consisting of a hydrocolloid solution (or gel) a sweetener and/or an aroma. Such systems have gradually evolved in complexity as knowledge and the necessary sensory and statistical methods have evolved. Whilst these systems may at times seem far removed from the real-world applications of hydrocolloids (when does anybody drink hydrocolloid solutions in isolation, other than in a sensory testing booth?) it is nonetheless important to grasp the scientific principles which have been learned from such studies and which form the basis of our understanding. When reading the subsequent paragraphs the reader should remember that in all experiments the hydrocolloid was added as a bulk thickener in the liquid (or gelled) state. Hence these conclusions may not apply directly to more specialist applications where hydrocolloids are used to form microstructures, e.g. in emulsions or as encapsulants. Fundamental research has demonstrated the suppression of sweetness intensity by food hydrocolloids. An early report of the phenomenon by Vaisey et al. (1969) concluded that gums which exhibited a smaller decrease in viscosity as shear rates increase tended to mask sweetness perception. Thus corn starch (which is `shear-thinning' in nature) was said to mask sweetness less than either guar gum or CMC (carboxymethylcellulose). Moskowitz and Arabie (1970) worked with just one thickener (NaCMC) and investigated the relationship between thickener concentration (i.e. viscosity) and the perceived intensities of a series of concentrations of representative sweet, bitter, sour and salty tastants. The sweetness of glucose (0.125±2 M) decreased with increasing viscosity of NaCMC solutions and for each glucose concentration sweetness was modelled using a power law with a negative exponent. Similar, but more extensive studies were conducted by Pangborn et al. (1973) who worked with several different hydrocolloids (NaCMC, alginate, xanthan and hydroxypropylcellulose) and sweeteners (sucrose and Na saccharin). The hydrocolloid concentrations and hence the viscosity range investigated (1±72 mPas) were lower than in many comparable trials because this research was targeted at beverage products. The perceived sweetness of sucrose was decreased by each of the thickeners tested. Thus the impact upon sweetness was sufficient to be detected even at relatively low hydrocolloid levels typical of beverage applications. An apparently contradictory result was observed with Na saccharin, which was judged to become sweeter at higher concentrations of Na CMC and Na alginate. To account for this increase, it was suggested that Na ions are involved in the sweet taste of Na saccharin and, indeed, the effect was not observed for thickeners which didn't contain Na counter-ions. As a last note on this paper, the authors concluded that, except for sucrose, modification of taste intensity was independent of viscosity and appeared to be related to the physicochemical properties of the hydrocolloid
Hydrocolloid±sweetener interactions in food products
391
and the taste compound. This finding may relate specifically to low viscosity systems where the effects of individual sweeteners on overall viscosity are more evident and can interact with taste perception. Thus the authors noted that hydrocolloid solution viscosity was decreased by the addition of all taste compounds other than sucrose, which increased the physical viscosity. Christensen (1980) used different molecular weight grades of NaCMC (and a much greater viscosity range than Pangborn et al., 1973) to show that sweetness suppression (of sucrose) was principally a function of solution viscosity, as opposed to the concentration of thickener added to achieve this viscosity. This finding was further verified by Izutsu et al. (1981). The effects of viscosity on the stimulus and recognition thresholds of sucrose were determined by Paulus and Haas (1980). Thickeners used in the study were methyl cellulose, guar seed flour, carob seed flour and tara seed flour. An increase in solution viscosity was shown to raise the sensory thresholds of all tastants used in the trial. The study did not find many differences in behaviour between hydrocolloids (or indeed tastants) which suggested that the increase in sensory threshold with viscosity was due to a masking or perceptual effect of the gums themselves, as opposed to specific physicochemical interactions. Cook et al. (2002) used sensory paired comparison tests to investigate differences in taste intensity between a low viscosity level of hydroxypropylmethylcellulose (HPMC; 2 g/L) and a high viscosity level (10 g/L). The perceived sweetness of all four sweet stimuli investigated (250 mg/L aspartame, 50 g/L sucrose, 4.5 g/L fructose and 39 mg/L neohesperidin dihydrochalcone) was greater at the lower viscosity level. A multiple-paired comparison test design was used to identify how much additional sucrose or aspartame was required to redress the decrease in sweetness which occurred in the more concentrated HPMC (10 g/L). This revealed that the magnitude of the effect was similar for both aspartame and sucrose, on a proportionate basis. Hence the underlying mechanism was operative for sweeteners used at considerably different concentrations and which are thought to have different taste transduction mechanisms (Kinnamon, 1996; Naim et al., 2002). Model studies employing more complete flavour systems (i.e. sweetener and aroma) have likewise noted the suppression of sweetness at increased concentrations of a range of hydrocolloids. A landmark study in the field was conducted by Baines and Morris (1987), who investigated both sweetness (of 100 g/L sucrose) and strawberry flavour perception in solutions thickened with guar gum and incorporating a strawberry aroma. Three different molecular weight samples of guar were employed at a range of concentrations whilst the level of flavouring remained constant. The authors noted that suppression of sweetness occurred progressively at hydrocolloid concentrations above c* (c* is the concentration of a particular thickener at which the polysaccharide chains begin to overlap and entangle with one another in solution, causing an abrupt increase in apparent viscosity above this concentration). The potential effects of sweetness suppression on flavour perception overall (see Section 18.3.3) were illustrated by the work of Hollowood et al. (2002). This study re-visited the fundamental approach used by Baines and Morris, but in addition was able to
392
Optimising sweet taste in foods
measure aroma release in the nose whilst samples were consumed, using a realtime mass spectrometry technique (Taylor et al., 2000). The significant finding, was that in-nose release of strawberry aroma from viscous hydrocolloid solutions was independent of thickener concentration and hence that the observed decrease in strawberry flavour intensity was not due to a physical effect of viscosity on aroma release. The authors concluded that the principal effect of increasing solution viscosity was a reduction in perceived sweetness and thus that strawberry flavour intensity was `driven' by a perceptual interaction between sweetness and aroma. This interaction was also observed by Cook et al. (2003) using isoamyl acetate (banana) flavoured sucrose-sweetened (50 g/L) solutions. This study employed five different thickeners (HPMC, MC, guar, xanthan and -carrageenan) and investigated the relationship between physical properties of hydrocolloid solutions and flavour perception. Perception of sweetness for all of the gum solutions could be predicted by an oral viscosity parameter calculated for each solution. This supported (but did not prove) the hypothesis that the textural attributes of hydrocolloid solutions interact directly with sweetness at a perceptual level. This review of hydrocolloid±sweetness interactions in liquid systems reveals some generic rules. Addition of a wide range of hydrocolloids has been shown to decrease the perceived intensities of a variety of sweet agents. Only when considering a low viscosity range, typical of beverages, were there significant effects of hydrocolloid or sweetener type. The decrease in sweetness is principally a function of solution viscosity and it has been suggested that highly shear-thinning hydrocolloids had less of an impact upon perceived sweetness. Sensory threshold concentrations of sweeteners are likely to be increased in the presence of food hydrocolloids. However, the rate of increase in perceived sweetness with stimulus concentration is the same in viscous solutions as in water (Moskowitz and Arabie, 1970). Lastly, it has been demonstrated that, whilst increased viscosity reduced the perceived sweetness of sucrose, it did not change the rate of sensory adaptation to the sweet stimulus delivered (Theunissen et al., 2000). 18.2.2 Gelled systems Sweetness suppression by hydrocolloids has also been reported in gelled food systems. The primary mechanism by which gels reduce sweetness intensity is by lowering the rate at which sweetener is transported to the taste receptors. This is determined principally by the breakdown and melting behaviour of the gel inmouth; a soft gel which breaks down more rapidly leads to relatively fast sweetener release whereas a firm gel retains its structural integrity for a longer period, resulting in lower rates of sweetener release (e.g. see Wilson and Brown, 1997). Whilst it is generally agreed that the extent of taste suppression depends on hardness in a particular hydrocolloid gel, when comparing systems gelled with different hydrocolloids this dependence is not clear (Bayarri et al., 2003). The reasons for this lie in the complexities of hydrocolloid gels, since there are a
Hydrocolloid±sweetener interactions in food products
393
range of factors which influence the rate of breakdown in-mouth. Gelatin, for example, is a special case due to its `melt-in-the-mouth' properties. This results in less inhibition of sweetener release than would be predicted purely on the basis of gel hardness measurements (Moritaka and Naito, 2002). A range of textural parameters of gels have been correlated with taste, and in particular sweetness, perception. These can help to explain apparent anomalies related to differences in taste intensity. For example, the brittle nature of gellan gels (and corresponding breakdown in-mouth) could explain why they have been reported to suppress sweetness less markedly, even at relatively high gel strengths (Costell et al., 2000). Top chef Heston Blumenthal employs gellan gels in his restaurant `The Fat Duck' at Bray (Berkshire, UK), citing their good flavour release properties. A typical dish using gellan is red pepper jelly, which he serves to garnish a hot soup. As with hydrocolloid solutions it is likely that the sensory perception of gel texture itself interacts with and may directly modify sweetness perception. Lethuaut et al. (2003) investigated the interactions between sweetness and texture in model dairy desserts which varied in terms of sucrose content and carrageenan composition (-, -, -carrageenans or an equal weight mixture of the three). Dairy desserts are a specific application area of carrageenans and different gel strengths may be formed by varying the blend which is used (carrageenan gives the hardest gels). In this study the dessert texture was assessed by both instrumental and sensory methods and this data was related to perceived sweetness and gel composition. It was concluded that the texture of the gels varied both with sucrose concentration and carrageenan-type, but that physical measurements of texture alone could not account for the observed sweetness of the desserts. Hence it was concluded that texture-sweetness interactions of a sensory (psychophysical) nature were also involved in determining the overall sweetness of the gels. As was noted in viscous liquid systems, starch gels may have superior flavour release properties in comparison with other hydrocolloids ± a suggestion first made by Hill et al. (1995) and based upon the perceived sweetness of lemon pie fillings gelled with corn starch. When sweetness data were plotted against the starch concentration relative to c*, the suppression of sweetness above c* was less marked than had previously been observed using hydrocolloid gelled systems. A further consideration when formulating gels is that their texture can be modified by the concentration of co-solutes such as sugars. For example, it was reported that sucrose influenced the compression parameters (rupture stress, strain and deformability modulus) of -carrageenan and gellan gum gels, whereas aspartame did not (Bayarri et al., 2003). Thus sugar replacement strategies must allow for changes in textural properties of gels and the consequent modification to gel breakdown and sweetener release.
394
Optimising sweet taste in foods
18.3
Mechanisms of hydrocolloid±sweetener interactions
From the foregoing discussion of the literature, it is clear that hydrocolloid± sweetener interactions are a complex field of study where multiple mechanisms may co-exist to account for observed sensory effects. However, I find it convenient to think in terms of three categories of effect: 1. 2. 3.
Specific binding effects Mass transport effects Multisensory interactions.
Placed in this (logical) order, hydrocolloid addition can impact upon sweetness by (1) reducing the amount of sweetener available for release, (2) lowering the rate at which sweetener can reach the taste receptors or (3) modifying the perception of sweetness at a cognitive level by stimulation of another sensory modality (e.g. texture). Each of these mechanisms is now considered in turn. 18.3.1 Specific binding between a hydrocolloid and sweetener This topic has received relatively little attention in the literature, in comparison to the well-documented field of aroma-binding to hydrocolloids. However, it is clear from consideration of the range of chemical structures presented by both sweeteners and hydrocolloids that specific binding is a possibility. This could result in a reduction in the pool of sweetener available for release and hence a reduced driving force for mass transfer to the receptors. For example, binding between a high-potency sweetener (present at a low concentration) and a macromolecule with multiple repeating binding sites might be envisaged to have a substantial effect upon sweetness perception. Hansen et al. (1989) used C-13 NMR to demonstrate physico-chemical interactions between starch and sucrose, although it is not clear from this work whether such interactions are sufficiently strong, or permanent in nature, to modify taste release. 18.3.2 Mass transport effects With regards to sweet taste, mass transport involves all of the processes which deliver sweetener from within (or on the surface of) a product to the taste buds. Hydrocolloids may influence this process at several levels because they change the physical characteristics of a food. First, they can change the rate at which sweet molecules can move through a food itself; this is usually expressed in terms of the molecular diffusion coefficient (D) for a sweetener within a particular food. This parameter has been included in many models of flavour release in vivo, although its influence may be less significant than has frequently been cited. For one reason, molecular diffusion coefficients have been measured for tastants in different hydrocolloid solutions and the variation has not proven considerable, in spite of demonstrable differences in taste perception between such systems. For example, the diffusion coefficient of sucrose (10%) in xanthan gum (an extremely viscous 1% w/w) was measured as 5:7 10ÿ10 m2.sÿ1,
Hydrocolloid±sweetener interactions in food products
395
whereas the corresponding figure in water was 5:8 10ÿ10 m2.sÿ1 (Basaran et al., 1999). The similarity between these figures is thought to reflect the high proportion of water in viscous hydrocolloid systems (i.e. at the molecular level, sucrose in a 1% xanthan solution is effectively diffusing through water for much of the time). Mass transport of sweetener within a foodstuff only influences perceived sweetness intensity if the rate of transfer to the product surface becomes rate-limiting during eating. Since foods and beverages are consumed under conditions of high mixing and shear, surface renewal of the bolus is rapid and it is rare for mass transfer within the food to become limiting. There are examples in the literature where the perception of sweetness has been correlated with measurements (or predictions) of the diffusion coefficients of sweet molecules (Kokini et al., 1982; Cussler et al., 1979). My interpretation of these papers would be that the diffusion coefficients co-varied with a physical parameter of the systems which determined the rate of tastant release in-mouth ± e.g. their bulk mixing behaviour. In-mouth mixing is the second mechanism by which hydrocolloids can influence the mass transport of sweetener. The physical qualities which they impart can alter the way in which a food mixes with saliva (e.g. in terms of the intimacy of mixing and the resultant contact area across which mass transfer occurs). A visual representation of this process was provided by Ferry et al. (2006), who filmed the transfer of red food colouring from different hydrocolloid solutions into a beaker of water following a brisk stir with a spoon (conditions of shear). Although this experiment was fairly crude, the differences which it highlighted were plain enough. Some hydrocolloids (e.g. HPMC) were particularly resistant to mixing with the aqueous phase and the red pigment remained within a discrete phase of un-mixed hydrocolloid. However, other equally viscous hydrocolloid solutions (e.g. cross-linked waxy maize starch) showed more rapid mixing behaviour indicated by dissolution of the pigment. It was hypothesized that this mixing behaviour ± which was assumed to extrapolate to mixing between gums and saliva in the mouth ± might explain differences in taste suppression between equi-viscous gum solutions. In the past, such differences have frequently been attributed to the different shear thinning behaviour of gums under oral shear conditions (this was incorporated in the Kokini oral shear stress which Cook et al. (2003) used to model flavour perception). A further example of the modification of sweetener release by hydrocolloids is in the melting and oral breakdown behaviour of gels, as discussed in Section 18.2.3. Naturally, this only applies where the hydrocolloid is present at a concentration which is sufficient to modify the bulk physical properties of a food. For example, gels containing an equivalent amount of sweetener to liquid hydrocolloid systems taste less sweet (Alley and Alley, 1998), predominantly because in-mouth mixing is less rapid/intimate for gelled materials. As a final note on mass transport processes it should be remembered that transfer into saliva is not in itself sufficient to elicit sweet taste; the tongue is far from flat and taste buds reside in narrow pores which extend beneath its surface. Sweet molecules are assumed to diffuse passively into the taste pores under a
396
Optimising sweet taste in foods
concentration gradient. Hydrocolloids might obstruct this process by forming a physical barrier on the tongue's surface and limiting access of sweetener to the taste pores. Such a mechanism is theoretically possible, although it is unlikely to be the main mode of action of hydrocolloids in modifying taste intensity; otherwise the substantial scientific literature indicating varying effects of hydrocolloids on the different taste modalities would be hard to explain! 18.3.3 Multisensory interactions Sweet taste provides the focus for this book, yet current knowledge of the multisensory nature of flavour perception (Delwiche, 2004) dictates that we should not consider the sensory modalities in isolation. When food is consumed, multiple sensory stimuli arise simultaneously and combine to produce the sensation which we identify as flavour. The relative contribution from each of the senses to flavour perception depends upon the food under consideration, as do the interactions between the senses which combine to produce the overall quality and intensity of flavour. Taste and smell, the primary flavour stimuli, are known to interact when the perceived stimuli are `congruent' with one another (in this context congruent implies a combination of taste and aroma which is commonly paired in foods and hence are associated together by the consumer). In particular, taste is known to be a key driver of overall flavour intensity, (e.g. see Davidson et al., 1999; Asquith and Swaine, 2004) whereas the accompanying aroma embodies the identity or character of a flavour. In the current context, the multisensory nature of flavour has two primary areas of significance. First, a reduction in perceived sweetness can change both the character and intensity of food flavour. Second, the modification of texture brought about by the addition of hydrocolloids (or the removal of sugars), may alter flavour at a perceptual level. The latter mechanism was proposed by SaintEve et al. (2004), to account for sweetness±texture interactions in yoghurt formulations which contained no added hydrocolloids. An increase in yoghurt viscosity resulted in a decrease in sensory sweetness; however, the texture of the yoghurts was modified purely by mechanical treatment, as opposed any change in formulation.
18.4
Implications for food product development
The implications of hydrocolloid±sweetener interactions for the food product development technologist are perhaps less complex than might at first appear. It is not necessary to have a detailed understanding of the full mechanisms of hydrocolloid±sweetener interactions in order to design quality food products. The product-dependent nature of many of the mechanisms means that there cannot be generic models to guide formulations for all product types, flavour systems and so-forth. Hence there is no substitution for in-house product
Hydrocolloid±sweetener interactions in food products
397
knowledge coupled with advice from an applications specialist from the hydrocolloid industry. Scientific principles may then be used to guide product (re-)formulation and a suggested flow sheet for the early stages of such a process is illustrated in Fig. 18.2. When adding hydrocolloids to food systems it should be borne in mind that hydrocolloids themselves may give rise to taste sensations, albeit that they are frequently described as tasteless. These sensations are often difficult to categorise and may in fact be of a somatosensory origin. For example, some hydrocolloids at higher concentrations leave a drying or puckering sensation inmouth. Others cause sensations closer to a taste stimulus and may have a pasty character. These factors are rarely problematic in food applications, due both to
Fig. 18.2 Flowchart for the early stages of a product development process highlighting consideration of hydrocolloid±sweetener interactions.
398
Optimising sweet taste in foods
masking effects in real food systems and the typically low levels of application. However, food grade hydrocolloids differ in purity and the product developer should remain watchful for taints or off-notes developed by the introduction of hydrocolloids. Clark (2002) described the bitterness or `bite' attributable to the addition of propylene glycol alginate to lemon-flavoured beverages. Interestingly, this flavour note was considered desirable in the commercial product Sunny DelightÕ! 18.4.1 Balancing functionality and flavour Sugar replacement strategies must address the contribution of sugar to both sweetness and the physical body, or mouthfeel, of a product (Kilcast, 2004a). If we take the formulation of a diet beverage product as an example, it is clear that substitution of nutritive sweeteners with a blend of high intensity sweeteners (e.g. acesulfame K/aspartame) produces a beverage which is deficient in terms of the mouthfeel contributed by 10±15% of sucrose or high fructose corn syrup. For this reason it is common practice to add a hydrocolloid thickener as a part of sugar-replacement. However, this may further modify the perception of sweetness and extensive sensory testing may be required to achieve the optimal combination of flavour intensity, quality and body of the beverage. The perceptual link between sweetness and the viscosity of beverages was demonstrated by Burns and Noble (1985) using vermouth of varying sucrose content, whose viscosity could be further modified using the glucose polymer Polycose. Samples in which viscosity was increased by the non-sweet Polycose were rated sweeter than vermouth solutions of the same sucrose concentration but lower physical viscosity (N.B. the viscosity of samples was typically 3 mPas and in this low viscosity range inhibition of sweetness is negligible. The observed enhancement of sweetness was probably caused by the brain being `tricked' into thinking that more sucrose was present than was actually the case, due to the enhanced body provided by Polycose). Conversely, when vermouths of identical viscosity were compared, the vermouths with higher sucrose concentration were judged to be more viscous. It is a well-known phenomenon of sweetness perception that there are distinct qualities and time-courses of perception associated with different sweeteners (e.g. see Walters and Roy 1996). Equally, some blends of sweeteners act synergistically when mixed together (e.g. acesulfame K and aspartame), whilst others do not (e.g. acesulfame K and saccharin). The complex nuances of sweetness perception are covered elsewhere in this book, but here it suffices to mention the fact that different sweetener blends are required to best enhance specific flavours. Thus the balance of a particular formulation may be substantially disrupted by adding a hydrocolloid which modifies the perception of a particular sweetener blend. By way of illustration, it has been reported that a blend of acesulfame K and aspartame (40 : 60) in orange flavoured beverages resulted in temporal profiles of sweetness and fruitiness which were similar to those produced by sucrose (Matysiak and Noble, 1991). However, in raspberry
Hydrocolloid±sweetener interactions in food products
399
flavoured beverages the optimum blend differed, depending upon the origins of the flavour (e.g. 25 : 75 for a natural flavouring, Baron and Hanger, 1998). Clearly the task of matching a particular flavour to a sweetener blend and attaining the required physical and functional properties is likely to be an iterative process! 18.4.2 Criteria for hydrocolloid selection in food product development Hydrocolloid selection for a particular application should meet the functional requirements and cost constraints whilst having the most desirable impact upon sensory quality (taste and texture are most relevant in the current context). Highly shear thinning hydrocolloids such as xanthan can be used to formulate products that combine a high resting viscosity with a relatively low oral viscosity (at mouth shear rates). This combination can provide functionality whilst minimising the direct sensory effects of texture on flavour (Cook et al., 2003), a feature which may account for xanthan's increasing popularity with chefs, in addition to its many industrial applications. Novel processing conditions may also be used to produce systems which combine a high low-shear viscosity with strong shear-thinning behaviour; e.g. the formation of sheared gels by applying shear as a hydrocolloid is undergoing gelation (Phillips and Williams, 2000). The influence of added hydrocolloid on the in-mouth mixing behaviour of a product should also be considered. For example, the in-mouth melting behaviour of gelatin gels has been linked to their superior taste release properties (Clark, 2002) and from a flavour viewpoint it is important that this melting behaviour is mimicked when seeking alternative gelling agents. Giannouli and Morris (2003) have reported the development of xanthan cryogels which melted in the desired range, although some restrictions upon possible co-solute concentrations were noted. In viscous liquid systems, the work of Ferry et al. (2006) suggests that starch or modified starch thickened foods should exhibit more rapid in-mouth mixing behaviour than those thickened with, e.g., modified celluloses. This should result in more rapid release of sweetener to the taste buds. However, for a desired viscosity level, starches are added at considerably higher concentrations than other food hydrocolloids and this together with the potential for aroma binding to amylose molecules may limit applications in some flavour systems (e.g. see Godshall and Solms 1992).
18.5
Future trends
The range of food applications of hydrocolloids has increased considerably in recent years and this trend seems likely to continue, driven by contemporary patterns of food production and consumption and increasing requirements for designer foods incorporating a range of functionalities. There is likely to be expansion both in terms of materials (there is considerable research into novel
400
Optimising sweet taste in foods
sources of hydrocolloids or novel modifications/blends of existing hydrocolloids) and in terms of applications (novel foods and functionalities). Hydrocolloid applications in so-called `functional foods', which exploit their non-digestible nature and physical bulk, are likely to be a big growth area. In spite of the high regulatory cost of developing novel food ingredients it is also likely that advances in nutrition and health sciences will continue to drive the development of novel sweeteners, fat and salt replacers and functional foods. To this extent, the future represents an extension of current trends and the scientific principles described herein will be applied in foods with increasing levels of complexity. As the range of hydrocolloids and sweeteners is enhanced, precise tailoring of sweetener-hydrocolloid blends to individual foodstuffs will become more sophisticated in order to optimise functionality and flavour. Lastly it is likely that advances in understanding of the multisensory nature of flavour perception will continue to unravel the complex nature of sweetness± texture interactions. In particular, the development of sophisticated brainimaging techniques will help us to understand how the brain interprets flavour information from the different senses and to elucidate the roles of other factors that impact upon perception (e.g. learning, attention, culture, environment etc.). With this we may build upon the pioneering work of Cerf-Ducastel et al. (2001) and Rolls et al. (2001) and finally understand the relative contributions of the sensory and physical processes which determine hydrocolloid±sweetener interactions.
18.6
Sources of further information and advice
If you are seeking a practically based guide to hydrocolloid functionality and applications, then look no further than the book Hydrocolloids by Andrew Hoefler. This book is pitched at an intermediate level from a scientific viewpoint but contains a host of useful information including functions and properties, compatibility issues between different hydrocolloids, selecting the right hydrocolloid for your application and example formulations. There is however no separate treatment of flavour development as an issue in its own right. For a more science-based text I recommend the Handbook of hydrocolloids by Phillips and Williams (2000). This book starts with a general overview of food hydrocolloids and continues with 25 chapters each devoted to a specific hydrocolloid (typically including sections on manufacture/structure/properties/ applications and regulatory status). If you wish to familiarise yourself with aspects of texture perception relating to foods, there are two recent texts which relate first to semi-solid and subsequently solid food systems (McKenna, 2003; Kilcast, 2004b). These texts include consideration of how texture is perceived by the consumer, how we can measure or predict the textural properties of foods and how texture is generated and/or modified in a range of food systems. Lastly, a comprehensive guide to alternative sweeteners can be found in the book edited by O'Brien Nabors (2001). The book commences with an overview of
Hydrocolloid±sweetener interactions in food products
401
sweeteners provided by the editor and continues with 22 chapters, each devoted to an individual sweetener. Thereafter, some mixed sweetener formulations and fat and oil replacers are considered. The following web-addresses might provide you with a rapid source of information with which to supplement the above texts: · http://www.hydrocolloid.com/ (the food hydrocolloid information centre, courtesy of IMR international). · http://www.lsbu.ac.uk/water/hydro.html (London South Bank University site containing useful information on structure, functionality and applications of food hydrocolloids).
18.7
References
and ALLEY, T. R. (1998), `The influence of physical state and color on perceived sweetness', Journal of Psychology, 132, 561±568. ASQUITH, T. N. and SWAINE, R. L. (Eds) (2004), Challenges in Taste Chemistry and Biology, Washington, American Chemical Society, 254±262. BAINES, Z. V. and MORRIS, E. R. (1987), `Flavour/taste perception in thickened systems: the effect of guar gum above and below c*', Food Hydrocolloids, 3, 197±205. BARON, R. F. and HANGER, L. Y. (1998), `Using acid level, acesulfame potassium aspartame blend ratio and flavor type to determine optimum flavor profiles of fruit flavored beverages', Journal of Sensory Studies, 13, 269±283. BASARAN, T. K., COUPLAND, J. N. and MCCLEMENTS, D. J. (1999), `Monitoring molecular diffusion of sucrose in xanthan solutions using ultrasonic velocity measurements', Journal of Food Science, 64, 125±128. BAYARRI, S., DURAN, L. and COSTELL, E. (2003), `Compression resistance, sweetener's diffusion and sweetness of hydrocolloid gels', International Dairy Journal, 13, 643±653. BURNS, D. J. W. and NOBLE, A. C. (1985), `Evaluation of the Separate Contributions of Viscosity and Sweetness of Sucrose to Perceived Viscosity, Sweetness and Bitterness of Vermouth', Journal of Texture Studies, 16, 365±381. CERF-DUCASTEL, B., VEN DE MOORTELE, P. F., MACLEOD, P., LE BIHAN, D. and FAURION, A. (2001), `Interaction of gustatory and lingual somatosensory perceptions at the cortical level in the human: a functional magnetic resonance imaging study', Chemical Senses, 26, 371±383. CHRISTENSEN, C. M. (1980), `Effects of solution viscosity on perceived saltiness and sweetness', Perception & Psychophysics, 28, 347±353. CLARK, R. (2002), `Influence of hydrocolloids on flavour release and sensory-instrumental correlations'. In Phillips, G. O. (Ed.), Gums and Stabilisers for the Food Industry 11, Cambridge, Royal Society of Chemistry, 217±225. COOK, D. J., HOLLOWOOD, T. A., LINFORTH, R. S. T. and TAYLOR, A. J. (2002), `Perception of taste intensity in solutions of random-coil polysaccharides above and below c*.' Food Quality and Preference, 13, 473±480. COOK, D. J., HOLLOWOOD, T. A., LINFORTH, R. S. T. and TAYLOR, A. J. (2003), `Oral shear stress predicts flavour perception in viscous solutions', Chemical Senses, 28, 11±23. COSTELL, E., PEYROLON, M. and DURAN, L. (2000), `Note. Influence of texture and type of ALLEY, R. L.
402
Optimising sweet taste in foods
hydrocolloid on perception of basic tastes in carrageenan and gellan gels', Food Science and Technology International, 6, 495±498. CUSSLER, E. L., KOKINI, J. L., WEINHEIMER, R. L. and MOSKOWITZ, H. R. (1979), `Food texture in the mouth', Food Technology, 33, 89±92. DAVIDSON, J. M., LINFORTH, R. S. T., HOLLOWOOD, T. A. and TAYLOR, A. J. (1999), `Effect of sucrose on the perceived flavor intensity of chewing gum', Journal of Agricultural and Food Chemistry, 47, 4336±4340. DELWICHE, J. (2004), `The impact of perceptual interactions on perceived flavor', Food Quality and Preference, 15, 137±146. FERRY, A.-L., HORT, J., MITCHELL, J. R., COOK, D. J., LAGARRIGUE, S. and VALLES PAMIES, B. (2006), `Viscosity and flavour perception: why is starch different from hydrocolloids?' Food Hydrocolloids, in press (published on-line 28 November 2005). GIANNOULI, P. and MORRIS, E. R. (2003), `Cryogelation of xanthan', Food Hydrocolloids, 17, 495±501. GODSHALL, M. A. and SOLMS, J. (1992), `Flavor and Sweetener Interactions with Starch', Food Technology, 46, 140±145. HANSEN, L. M., SETSER, C. S. and PAUKSTELIS, J. V. (1989), `Investigations of sugar±starch interactions using C-13 nuclear magnetic resonance. 1. sucrose', Cereal Chemistry, 66, 411±415. HILL, M. A., MITCHELL, J. R. and SHERMAN, P. A. (1995), `The relationship between the rheological and sensory properties of a lemon pie filling', Journal of Texture Studies, 26, 457±470. HOEFLER, A. C. (2004), Hydrocolloids, Eagen Press, St. Paul, Minn., USA. HOLLOWOOD, T. A., LINFORTH, R. S. T. and TAYLOR, A. J. (2002), `The effect of viscosity on the perception of flavour', Chemical Senses, 27, 583±591. IZUTSU, T., TANEYA, S. I., KIKUCHI, E. and SONE, T. (1981), `Effect of viscosity on perceived sweetness intensity of sweetened sodium carboxymethylcellulose solutions', Journal of Texture Studies, 12, 259±273. KILCAST, D. (2004a), `Satisfying consumer demands for sweetness quality', Agro Food Industry Hi-Tech, 15, 36±38. KILCAST, D. (Ed., 2004b), Texture in food. Volume 2: Solid foods, Woodhead Publishing Limited, Cambridge. KINNAMON, S. C. (1996), `Taste transduction: Linkage between molecular mechanisms and psychophysics', Food Quality and Preference, 7, 153±159. KOKINI, J. L., BISTANY, K., POOLE, M. and STIER, E. (1982), `Use of mass-transfer theory to predict viscosity±sweetness interactions of fructose and sucrose solutions containing tomato solids', Journal of Texture Studies, 13, 187±200. LETHUAUT, L., BROSSARD, C., ROUSSEAU, F., BOUSSEAU, B. and GENOT, C. (2003), `Sweetnesstexture interactions in model dairy desserts: effect of sucrose concentration and the carrageenan type', International Dairy Journal, 13, 631±641. MATYSIAK, N. L. and NOBLE, A. C. (1991), `Comparison of Temporal Perception of Fruitiness in Model Systems Sweetened with Aspartame, an Aspartame + Acesulfame K Blend, or Sucrose', Journal of Food Science, 56, 823±826. MCKENNA, B. M. (Ed., 2003), Texture in food: Volume 1: Semi-solid foods, Woodhead Scientific and CRC Publications. MORITAKA, H. and NAITO, S. (2002), `Agar and gelatin gel flavor release', Journal of Texture Studies, 33, 201±214. MOSKOWITZ, H. R. and ARABIE, P. (1970), `Taste intensity as a function of stimulus concentration and solvent viscosity', Journal of Texture Studies, 1, 502±510.
Hydrocolloid±sweetener interactions in food products
403
and SAMUELOV-ZUBARE, M. (2002), In Paredes, D. (Ed.), Chemistry of Taste, Washington DC, American Chemical Society, 2±17. O'BRIEN NABORS, L. (Ed., 2001), Alternative Sweeteners, Marcel Dekker, Inc., New York. PANGBORN, R. M., TRABUE, I. M. and SZCZESNIAK, A. S. (1973), `Effect of hydrocolloids on oral viscosity and basic taste intensities.' Journal of Texture Studies, 4, 224±241. PAULUS, K. and HAAS, E. M. (1980), `The influence of solvent viscosity on the threshold values of primary tastes', Chemical Senses, 5, 23±32. PHILLIPS, G. O. and WILLIAMS, P. A. (Eds.) 2000), Handbook of hydrocolloids, Woodhead Publishing, Cambridge. ROLLS, E. T., VERHAGEN, J. V. and KADOHISA, M. (2003), `Representations of the texture of food in the primate orbitofrontal cortex: Neurons responding to viscosity, grittiness, and capsaicin', Journal of Neurophysiology, 90, 3711±3724. SAINT-EVE, A., KORA, E. P. and MARTIN, N. (2004), `Impact of the olfactory quality and chemical complexity of the flavouring agent on the texture of low fat stirred yogurts assessed by three different sensory methodologies', Food Quality and Preference, 15, 655±668. TAYLOR, A. J., LINFORTH, R. S. T., HARVEY, B. A. and BLAKE, A. (2000), `Atmospheric pressure chemical ionisation mass spectrometry for in vivo analysis of volatile flavour release', Food Chemistry, 71, 327±338. THEUNISSEN, M. J. M., KROEZE, J. H. A. and SCHIFFERSTEIN, H. N. J. (2000), `Method of stimulation, mouth movements, concentration, and viscosity: Effects on the degree of taste adaptation', Perception & Psychophysics, 62, 607±614. VAISEY, M., BRUNON, R. and COOPER, J. (1969), `Some sensory effects of hydrocolloid sols on sweetness', Journal of Food Science, 34, 397±400. WALTERS, D. E. and ROY, G. (1996), `Taste interactions of sweet and bitter compounds'. In Leland, J. V. (Ed.), Flavor±Food Interactions, Washington DC, American Chemical Society, 130±142. WILSON, C. E. and BROWN, W. E. (1997), `Influence of food matrix structure and oral breakdown during mastication on temporal perception of flavor', Journal of Sensory Studies, 12, 69±86. NAIM, M., NIR, S., SPIELMAN, A. I., NOBLE, A. C., PERI, I., RODIN, S.
19 Future directions: using biotechnology to discover new sweeteners, bitter blockers and sweetness potentiators R. McGregor, Linguagen Corp., USA
19.1
Introduction
Throughout human evolution, our ancestors have been faced with a constant struggle to find and consume enough nutrients for survival. This has driven the development of the sense of taste, the purpose of which is to discern nutrient content of potential foods prior to consumption using a chemosensory mechanism which links chemical interaction at the input end of the digestive system, the mouth, with determination of preference in the choice centers of the brain. When foods with high energy content interact with this system it triggers hedonic responses in the brain telling us to consume the food. During the second half of the twentieth century farming methods developed to such an extent that food became readily available for much of the world's population. Human physiology, on the other hand, has not evolved at anywhere near the same rate as our ability to feed ourselves, and hence we are faced at the start of the twentyfirst century with a world population with not only the hard-wired desire to consume calories, but also the opportunity. The result of this is accelerating rates of obesity worldwide, which is increasingly threatening not only people's quality of life and life expectancy, but also the healthcare systems and economies of many nations. What are needed are replacements for high calorie food ingredients that supply the hedonic sensation of high energy food without the calories. The average weight of an adult in the United States increased by 25 lb between 1960 and 2002 (Centers for Disease Control press release 27 October 2004) and by 2002, 60% of the US population was overweight, with 30% being
Fig. 19.1 The proportion of overweight and obese individuals in the US as measured by Body Mass Index (BMI) has increased rapidly since 1991 (Behavioral Risk Factor Surveillance System, Centers for Disease Control).
406
Optimising sweet taste in foods
obese (with a body mass index over 30) (National Center for Health Statistics NHANES IV Report, 2002). Figure 19.1 demonstrates the rapid rise in obesity in the US since 1991. Diseases related to obesity include non-insulin dependent diabetes, hypertension, gallbladder disease, hyperlipidaemia, some cancers (e.g. breast/colorectal), sleep apnea, and degenerative joint disease, with new diseases in which obesity appears to be a contributing factor being identified all the time. Growth in obesity spending accounted for 27% of growth in inflation-adjusted per capita healthcare spending between 1987 and 2001 (Thorpe et al., 2004). The individual costs of obesity are also significant, with obese individuals paying 37% more in healthcare costs than individuals of normal weight (Finkelstein et al., 2003). Staggeringly, a morbidly obese individual (BMI greater than 40 at age 25) is expected to die 12 years earlier than a normal weight person. Obesity can no longer be considered an American problem, as figures released in March 2005 by the International Obesity Task Force (IOTF) show over 200 million adults in the European Union alone are overweight or obese, and the World Health Organization reporting over 1.7 billion overweight individuals worldwide, 300 million of whom are obese. With these statistics in mind, it is evident that now is the time for science to identify substitutes for high calorie ingredients with palatability, safety and cost that are acceptable to the worldwide population. Over the past 15 years, great strides have been made in understanding the biochemical mechanisms of taste at the molecular level. This chapter discusses how this knowledge is being used to enable modern biotechnological approaches in the discovery of novel taste modifying compounds for the optimization of sweet taste in foods without high caloric content. The first section outlines developments in the understanding of the science of taste perception, and how this knowledge enables the approaches now being used to discover novel sweeteners and sweetness enhancers. The next section discusses how modifiers of bitter taste can be used to improve the taste profile of sweeteners and includes an example of a bitter taste blocker used to improve the taste of a currently marketed sweetener. The chapter closes with a discussion of future trends in sweetener science and development, including how further elucidation of the molecular structures involved in sweet taste transduction will enable the discovery of the next generation of sweeteners and sweetness enhancers.
19.2 Understanding taste at the molecular level enables discovery of novel sweeteners The discovery of the major non-caloric sweeteners currently on the market was, by and large, a serendipitous process. The folklore surrounding the discoveries of saccharin, aspartame, and sucralose by scientists accidentally tasting these sweeteners, as a result of either spillage or misunderstanding, is well known. With the elucidation at the molecular level of the cellular pathways involved in
Future directions
407
sweet taste, scientists now have the opportunity to take a systematic approach to identifying non-nutritive sweeteners. In humans and other animals, perception of taste occurs in specialized structures called taste buds. Taste buds are groups of elongated taste receptor cells, whose purpose is to bind tastants present in the mouth, and transduce this physical binding into an electrochemical signal which is relayed to the brain. The taste proteins that have been characterized in recent years are integral to this process as they convert the physical binding of tastant at the apical surface of the cell (exposed to the mouth cavity) into release of neurotransmitter at the basolateral surface of the cell (exposed to the peripheral nervous system), resulting in a nerve response. Taste is a sense that has evolved to enable an organism to determine important information regarding the quality of food. Of the five currently recognized taste modalities: sweet, salty, sour, bitter and umami (savory), three are associated with positive attributes of food and two with negative ones. Sweet tasting foods are typically high in carbohydrates, salty food contains important minerals, and umami taste is indicative of the presence of amino acids. The perception of sour taste is a protective mechanism against ingestion of spoiled food, while bitter taste is associated with poisons, such as plant alkaloids. Other tastes, such as the taste of polysaccharides, may be accepted as distinct taste modalities in the future (Sclafani, 2004). Taste science entered the molecular era in the early 1990s with the discovery in taste responsive cells of the protein, gustducin (McLaughlin et al., 1992). Gustducin was determined to interact with receptors in the taste cell membrane upon binding of certain tastants, and its activation results in taste cell activation. The subsequent decade and a half has seen the identification of other taste transduction components including the receptors responsible for sweet taste. Gustducin is a member of a family of proteins called G proteins, and hence the receptor proteins it interacts with are known as G protein coupled receptors (GPCRs). A number of G proteins have subsequently been identified in the taste system (McLaughlin et al., 1994). Upon tastant binding, a change in conformation of the receptor leads to activation of G protein. G protein activation in turn results in the switching on of intracellular proteins known as effector enzymes. These effector enzymes modulate the concentration of molecules within the cell called second messengers. In the resting state, taste cells are polarized due to the active maintenance of concentration gradients of ions across the cell membrane. As second messenger levels change, various ion channels are activated both in the cell membrane and in intracellular membranes, resulting in the depolarization of the taste cell as ions flow down their respective concentration gradients. Depolarization results in release of neurotransmitter from the taste cell into the synaptic cleft and subsequent activation and depolarization of the adjacent neuron. This electrical signal is transmitted through the nervous system to the brain where it is interpreted as taste. Sweet and umami tastes are transduced through GPCR heterodimers consisting of two receptor subunits. T1R1 and T1R3 receptors act in concert
408
Optimising sweet taste in foods
to detect umami compounds, whilst T1R2 and T1R3 form the functional sweet taste receptor (Kitagawa et al., 2001; Max et al., 2001; Montmayeur et al., 2001; Sainz et al., 2001; Nelson et al., 2001; Bachmanov et al., 2001; Li et al., 2002). The T1Rs are members of the class C family of GPCRs, which all have a large extracellular domain containing the binding site for ligands. The structure of the T1R2±T1R3 heterodimer, and how this receptor complex is believed to bind sweeteners will be discussed in more detail shortly. In studies on mice lacking expression of one of the subunits of the receptor, it has been shown that the mice are responsive to certain sweeteners and umami compounds, hence raising the possibility that other receptors or mechanisms are involved in sweet and umami taste transduction in mice (Damak et al., 2003; Zhao et al., 2003). In addition to the T1R1±T1R3 heterodimer, a GPCR called mGluR4 has been hypothesized as an umami receptor (Chaudhari et al., 2000). There are at least 25 GPCRs proposed to be involved in bitter taste (Adler et al., 2000; Matsunami et al., 2000). These GPCRs, known as the T2Rs, allow for the detection of the wide variety of structural classes of compounds that are bitter. This large family of receptors also speaks to the etiological importance of bitter taste detection. If only one receptor was involved in bitter taste, then a non-functioning mutation of this receptor would likely result in death as there would be no first line detection apparatus for many poisons. It appears that not all bitter tastants exert their bitter taste by activating T2R receptors. It has been reported that certain amphiphilic substances, including H1receptor antagonists, can directly activate G proteins (Naim et al., 1994; Burde et al., 1996). It is possible that the GPCR-independent mechanism of G protein activation represents a bitter-sensing mechanism for many bitter tastants present at high concentrations. Other bitter compounds, such as the methyl xanthines (e.g. caffeine in coffee, theophylline in tea and theobromine in cocoa) are known to interact with phosphodiesterase (PDE), which is one of the effector enzymes present in taste cells. Studies have shown that in response to caffeine and theophylline, taste cell levels of the substrate for PDE rises (Rosenweig et al., 1999). Indeed, it can be predicted that the taste of any compound that modulates the level of activation of a protein involved in the taste transduction cascade could be determined by such an interaction. In contrast to bitter, sweet, and umami taste, salt and sour taste appear to be transduced by pathways beginning with interaction of tastant with ion channels. For salty taste, evidence suggests that the epithelial sodium channel (ENaC) functions as a salt receptor, as amiloride hydrochloride, a diuretic drug that blocks ENaC, also reduces sensitivity to salt (Lindeman et al., 1998; Kretz et al., 1999; Lin et al., 1999). The amiloride-insensitive salt taste transduction mechanism appears to be mediated at least in part by a variant of the transient receptor potential V1 (TRPV1) channel, as modulators of TRPV1 have activating and blocking effects on the nerve responses to salts, and mice with TRPV1 genetically ablated lack an amiloride-insensitive nerve response (Lyall et al., 2004, 2005). Both mechanisms appear to be involved in human taste perception (Feldman et al, 2003). Whatever the relative contributions of these
Future directions
409
pathways to salty taste transduction, the effect of sodium ions on taste cells is depolarization and neurotransmitter release. Sour taste detection appears to be even more complex with a number of pathways possibly involved. Hydrogen ions in acids enter taste cells through proton channels and also interact with a variety of channels including potassium, sodium, calcium and chloride channels (DeSimone et al., 2001 for review). The overall effect is again a depolarization of the taste cell and the generation of a nerve impulse. With this knowledge in hand, it is now possible to use the powerful techniques of biotechnology to search for taste modifiers that would be unlikely to be discovered using traditional approaches. By using recombinant DNA technology, proteins involved in sweet taste transduction can be expressed in cellular systems and their modulation by sweeteners and test compounds quantified. These engineered cell lines can be formatted to enable the rapid screening of thousands of molecules to determine their taste modifying activity. This throughput is many times greater than is possible with more traditional approaches and means that sweeteners with suitable characteristics of taste profile, stability and safety will be identified in the coming years. Recombinant DNA techniques can also be used to determine the structure activity relationship (SAR) between taste modifiers and the proteins with which they interact. Mutagenesis of receptor proteins enables determination of essential amino acid residues and regions of the receptor involved in binding of tastants and other taste cell components. This approach is being used to characterize sweet taste. A total of four distinct binding sites for sweeteners have so far been postulated. By creating variations of the T1R subunits, Xu et al. (2004) identified the C-terminal transmembrane domain of T1R3 as being essential for the binding of the sweetener cyclamate and lactisole (an inhibitor of sweet taste). Medicinal chemistry facilitates a similar resolution of the important moieties of tastants that govern their properties. Once SAR has been determined, it allows for the efficient selection or synthesis of libraries of test compounds for tastant identification. To this end, homology models have been developed for the extracellular domain of T1R homodimers (Max et al. 2001; Tancredi et al., 2004), and mouse T1R2±T1R3 heterodimers (Temussi et al., 2002) proposing that the ligand free active form of the sweet receptor is stabilized by sweet proteins. Recently, significant progress has been made in determining the SAR of the human T1R2 and T1R3. A model has been published identifying key binding sites for a variety of sweet tasting compounds ranging from carbohydrates, amino acids and sweet molecules such as saccharin, to peptides and proteins (Morini et al., 2005). The model, built using homology to the similar metabotropic glutamate receptor (mGluR1), which has been previously crystallized (Kunishima et al., 2000; Tsuchiya et al., 2002), identifies a number of binding sites on the sweet receptor complex. The so-called wedge site, at which sweet proteins bind the extracellular region formed by the amino terminal domains, was confirmed in this model. These extracellular domains are also the site of binding pockets for small molecular weight sweeteners, one each on T1R2 and T1R3.
410
Optimising sweet taste in foods
These modeling studies are also shedding light on the phenomenon of sweetness potentiation. Certain mixtures of sweeteners are sweeter in combination than the sum of their individual sweetnesses (Schiffman et al., 2000). Sweeteners that have synergistic effects with other sweeteners, such as saccharin and aspartame are all modeled to bind to one of the extracellular domains on the T1R2±T1R3 receptor complex. It has been suggested that binding of these sweeteners to this site, along with concomitant binding of sweetener to the other extracellular site results in an increased sweet response.
19.3
The use of bitter blockers to improve sweet taste
The biotechnological approaches being used for sweetener discovery are also being used to identify modifiers of other taste modalities that can help optimize sweet taste in foods. A number of the sweeteners currently on the market have taste characteristics that limit their acceptance. One such characteristic which is a major problem is that of bitter aftertaste present in sweeteners such as saccharin, acesulfam potassium and stevia. With acesulfam potassium, formulation options are limited due to the narrow range of concentration at which it is perceived as sweet without a bitter taste. Intelligent molecular design and high throughput screening is being used to identify bitter blockers that can be formulated with current and future sweeteners to improve their overall taste profile. In the coming years, as currently available sweeteners become commoditized, using bitter blockers with these sweeteners should allow for formulation of cost effective non-nutritive sweeteners with good taste. One such bitter blocker that has shown efficacy at improving the taste of nonnutritive sweeteners is Linguagen Corp.'s first generation bitter blocker, adenosine-50 -monophosphate (AMP). AMP was discovered using the very same biotech approaches that are now being used to identify novel non-nutritive sweeteners (Ming et al., 1999). In carbonated soft drink formulations containing saccharin, AMP has been demonstrated in human sensory studies to reduce the bitter taste of saccharin and improve the overall taste of the drink (Fig. 19.2). Although AMP decreases the bitterness of sweeteners, it has other taste characteristics, notably an umami taste. This limits the use of AMP in sweetener applications and makes it more appropriate for savory formulations, particularly in salt reduction, where it decreases the bitter taste of the salt substitute potassium chloride. However, more suitable bitter blockers are being sought and should come to market in the coming years.
19.4
Future trends
With the characterization of sweet receptors at the atomic level, by first modeling and subsequently crystallization, the full structure activity relationship for sweeteners will be determined. This will enable predictive modeling of the
Future directions
Fig. 19.2
411
The bitter blocker, adenosine 50 -monophosphate, improves the taste profile of cola flavored carbonated soft drink sweetened with saccharin.
properties of molecules by computer and enable ultra-high throughput screening in silico. Computer generated hits will only be synthesized once their sweetener properties have been determined. As has been described elsewhere in this book, the taste profile of a sweetener consists of a number of factors, not just overall sweetness, but also time for onset, duration of effect and off-tastes. Full knowledge of SAR will enable sweeteners to be designed for specific purposes. For example, a sweetener could be designed with high affinity of binding to the sweet receptor complex. This sweetener, once bound to the receptor would trigger sweet taste perception for an extended period of time before finally disengaging from the receptor, and would be ideal for gum applications, where sweetness is required for at least 20 minutes. The food industry will be able to draw on numerous sweeteners and sweetness enhancers to satisfy the requirements of specific application. With the worldwide rise in obesity, the most important contribution that will be made by advances in sweetener science in the coming years will be in providing consumers with safe and functional non-nutritive sweeteners and sweetness enhancers that can replace nutritive sweeteners without compromising taste.
19.5
Sources of further information and advice
For an introduction to taste, Smith and Margolskee (2001) is a useful primer. A more detailed review of the genetics of taste perception is Kim et al. (2004).
412
Optimising sweet taste in foods
Murphy and Johnson (2003) review the science behind the guidance given on sugar intake in the US governments' Dietary Guidelines for Americans (http:// www.health.gov/dietaryguidelines). For those who don't know the history of high potency sweeteners discovery, Eric Walters has a short history on his website (http://www.rosalindfranklin.edu/cms/biochem/walters/sweet/ history.html).
19.6
References and ZUKER, (2000) A novel family of mammalian taste receptors. Cell 100: 693±702
ADLER, E., HOON, M.A., MUELLER, K.L., CHANDRASHEKAR, J., RYBA, N.J.P.
C.S.
BACHMANOV A.A., LI X., REED D.R., OHMEN J.D., LI S., CHEN Z., TORDOFF M.G., DE JONG P.J., WU
and BEAUCHAMP G.K. (2001) Positional cloning of the mouse saccharin preference (Sac) locus. Chem. Senses 26: 925±933. BURDE, R., DIPPEL, E. and SEIFERT, R. (1996) Receptor-independent G protein activation may account for the stimulatory effects of first generation H1-receptor antagonists in HL-60 cells, basophils and mast cells. Biochem. Pharmacology 51: 125±131. CHAUDHARI, N., LANDIN, S.D. and ROPER, S.D. (2000) A metabotropic glutamate receptor variant functions as a taste receptor. Nature Neurosci. 3: 113. C., WEST D.B., CHATTERJEE A., ROSS D.A.
DAMAK, S., RONG, M., YASUMATSU, K, KOKRASHVILI, Z., VARADARAJAN, V., ZOU, S., JIANG, P., NINOMIYA, Y. and MARGOLSKEE, R. F. (2003) Detection of sweet and umami taste in the absence of taste receptor T1r3. Science 301: 850±853. DESIMONE, J.A., LYALL, V., HECK, G.L. and FELDMAN, G.M. (2001) Acid detection by taste receptor cells. Respir. Physiol. 129(1±2): 231±245. FELDMAN, G. M., MOGYOROSI, A., HECK, G. L., DESIMONE, J. A., SANTOS, C. R., CLARY, R. A. and LYALL, V. (2003) Salt-evoked lingual surface potential in humans. J. Neurophysiol. 90: 2060±2064. FINKELSTEIN, E. A., FIEBELKORN, I. C. and WANG, G. (2003) National medical spending attributable to overweight and obesity: How much, and who's paying? Health Affairs July±Dec.; Suppl Web Exclusives: W3-219±W3-226. KIM, U. K., BRESLIN, P. A. S., REED, D. and DRAYNA, D. (2004) J. Dent. Res. 83(6): 448±453. KITAGAWA, M., KUSAKABE, Y., MIURA, H., NINOMIYA, Y. and HINO, A. (2001) Molecular genetic identification of a candidate receptor gene for sweet taste. Biochem. Biophys. Res. Commun., 283: 236±242. KRETZ, O., BARBRY, P., BOCK, R. and LINDEMANN, B. (1999) Differential expression of RNA and protein of the three pore-forming subunits of the amiloride-sensitive epithelial sodium channel in taste buds. J. Histochem. Cytochem. 47: 51±64. KUNISHIMA, N., SHIMADA, Y., TSUJI, Y., SATO, T., YAMAMOTO, M., KUMASAKA, T., NAKANISHI, S.,
JINGAMI, H. and MORIKAWA, K. (2000) Structural basis of glutamate recognition by a dimeric metabotropic glutamate receptor. Nature 407: 971±977. LI, X., STASZEWSKI, L., XU, H., DURICK, K., ZOLLER, M. and ADLER, E. (2002) Human receptors for sweet and umami taste. Proc. Natl. Acad. Sci. USA 99(7): 4692±4696. LIN, W., FINGER, T. E., ROSSIER, B. C. and KINNAMON, S. C. (1999) Epithelial Na+ channel subunits in rat taste cells: localization and regulation by aldosterone. J. Comp. Neurol. 405: 406±420. LINDEMANN, B., BARBRY, P., KRETZ, O. and BOCK, R. (1998) Occurance of ENaC subunit mRNA and immunocytochemistry of the channel subunits in taste buds of the rat vallate papilla. Ann. N.Y. Acad. Sci. 855: 116±127.
Future directions
413
LYALL, V., HECK, G. L., VINNIKOVE, A. K., GHOSH, S., PHAN, T.-H. T., ALAM, R. I., RUSSELL, O, F., MALIK, S. A., BIGBEE, J. W. and DESIMONE, J. A. (2004) The mammalian amilorideinsensitive non-specific salt taste receptor is a vanilloid receptor-1 variant. J. Physiol. 558: 147±159. LYALL, V., HECK, G. L., VINNIKOVA, A. K., GHOSH, S., PHAN, T.-H. T. and DESIMONE, J. A. (2005) A novel vanilloid receptor (VR-1) variant mammalian salt taste receptor. Chem. Senses 30: i42±i43. MATSUNAMI, H., MONTMAYEUR, J-P. and BUCK, L. B. (2000) A family of candidate taste receptors in human and mouse. Nature 404: 601±604 MAX, M., SHANKER, Y.G., HUANG, L., RONG, M., LIU, Z., CAMPAGNE, F., WEINSTEIN, H., DAMAK, S.
and MARGOLSKEE, R.F. (2001) Tas1r3, encoding a new candidate taste receptor, is allelic to the sweet responsiveness locus Sac. Nat. Genet. 28: 58±63. MCLAUGHLIN, S. K., MCKINNON, P. J. and MARGOLSKEE, R. F. (1992) Gustducin is a taste-cellspecific G protein closely related to the transducins. Nature 357: 563±569. MCLAUGHLIN, S. K., MCKINNON, P. J., SPICKOFSKY, N., DANHO, W. and MARGOLSKEE, R. F. (1994) Molecular cloning of G proteins and phosphodiesterases from rat taste cells. Physiol. Behav. 56(6): 1157±1164. MING, D., NINOMIYA, Y. and MARGOLSKEE, R. F. (1999) Blocking taste receptor activation of gustducin inhibits gustatory responses to bitter compounds. Proc. Nat. Acad. Sci. USA 96: 9903±9908. MONTMAYEUR, J. P., LIBERLES, S. D., MATSUNAMI, H. and BUCK, L. B. (2001) A candidate taste receptor gene near a sweet taste locus. Nat. Neurosci. 4: 492±498. MORINI, G., BASSOLI, A. and TEMUSSI, P. A. (2005) From small sweeteners to sweet proteins: Anatomy of the binding sites of the human T1R2±T1R3 receptor. J Med. Chem. 48(17): 5520±5529. MURPHY, S. P. and JOHNSON, R. K. (2003) The scientific basis of recent US guidance on sugar intake. Am. J. Clin. Nutr. 78(suppl): 827S±833S. NAIM, M, SEIFERT, R., NURNBERG, B., GRUNBAUM, L. and SCHULTZ, G. (1994) Some taste sunstances are direct activators of G proteins. Biochem. J. 297: 451±454. NELSON, G., HOON, M. A., CHANDRASHEKAR, J., ZHANG, Y., RYBA, N. J. P. and ZUKER, C. S. (2001) Mammalian sweet taste receptors. Cell 106: 381±390. ROSENWEIG, S., YAN, W., DASSO, M. and SPIELMAN, A. I. (1999) Possible novel mechanism for bitter taste mediated through cGMP. J. Neurophysiol. 81(4): 1661±1665. SAINZ, E., KORLEY, J.N., BATTEY, J.F. and SULLIVAN, S.L. (2001) Identification of a novel member of the T1R family of putative taste receptors. J. Neurochem. 77: 896±903. SCHIFFMAN, S. S., SATTELY-MILLER, E. A., GRAHAM, B. G., BOOTH, B. J. and GIBES, K. M. (2000) Synergism among ternary mixtures of fourteen sweeteners. Chem. Senses 25: 131± 140. SCLAFANI, A. (2004) The sixth taste? Appetite 43(1): 1±3. SMITH, D. V. and MARGOLSKEE, R. F. (2001) Making sense of taste. Sci. Am. 284(3): 32±39. TANCREDI, T., PASTORE, A., SALVADORI, S., ESPOSITO, V. and TEMUSSI, P. A. (2004) Interaction of sweet proteins with their rereceptor. A conformational study of peptides corresponding to loops of brazzein, monellin and thaumatin. Eur. J. Biochem. 271: 2231±2240. TEMUSSI, P. A. (2002) Why are sweet proteins sweet? Interaction of brazzein, monellin and thaumatin with the T1R2±T1R3 receptor. FEBS Lett. 526: 1±3. THORPE, K. E., FLORENCE, C. S., HOWARD, D. H. and JOSKI, P. (2004). The impact of obesity on rising medical spending. Health Affairs July±Dec.; Suppl Web Exclusives: W4480±W4-486.
414
Optimising sweet taste in foods
and MORIKAWA, K. (2002) Structural views of the ligand binding cores of a metabotropic glutamate receptor complexed with an antagonist and both glutamate and Gd3+. Proc. Natl. Acad. Sci. 99: 2660± 2665. XU, H., STASZEWSKI, L., TANG, H., ADLER, E., ZOLLER, M. and LI, X. (2004) Different functional roles of T1R subunits in the heteromeric taste receptors. Proc. Natl. Acad. Sci. 101: 13972±13073. ZHAO, G. Q., ZHANG, Y., HOON, M. A., CHANDRASHEKAR, J., ERLENBACH, I. and RYBA, N. J. (2003) The receptors for mamalian sweet and umami taste. Cell 115: 255±266. TSUCHIYA, D., KUNISHIMA, N., KAMIYA, N., JINGAMI, H.
Index
abrusoside A 294 acceptable daily intake (ADI) polyols 168±70 low-calorie sweeteners 177±213 acesulfame 309, 319 acesulfame K 5, 93, 178, 207, 292, 319, 354, 357, 410 advantages, disadvantages and applications 199 bitterness reduction 88 blending 167, 345, 346 discovery 309 safety, metabolic and physiological properties 193 sensory, chemical and physical properties 186 acetalic isovanillic derivatives 333 acids acid-vanilla mixture 73 interactions with sweeteners 88±92 carbohydrate sweetener±acid interactions 89±90 high potency sweetener±acid interactions 90±2 active conformation 333±5 adaptation 232 adapted isobolograms 368 added sugars 30±1 additives 346 adenosine monophosphate (AMP) 229, 410, 411 adenylyl cyclase (AC) 8, 14±15, 16 ADMET profiling 313
adolescents 56±7, 115±17 affination 139 Africa 225 age development of sweet taste preference 55±7 and sweet taste perception 57±8, 115±18 ageing 117±18 AH-B model 285, 313, 330 AH-B-X model 285, 313, 330 albiziasaponin B 294 alitame 178, 207, 309, 354, 357 advantages, disadvantages and applications 200 development as a commercial product 320 safety, metabolic and physiological properties 193 sensory, chemical and physical properties 186±7 American Dental Association 161 American Dietetic Association (ADA) 31, 160, 217, 227 Americas 223±4 amine salts 346 amino acids 293, 295 amino-terminal domain (ATD) 10 amphipathic tastants 16±17 permeation of taste cells in vivo and inhibition of receptor±related kinases in vitro 19±23 amyloglucosidase 267 analgesic effects 59±60
416
Index
analytical tasting strategies 71±2 anchorage point 352 values 353±6, 357 anethole 294, 299 antibody-based fluorescence polarisation assay 316 apical oral route 19±20 appearance 380 arabinogalactan 235 arylureas 287 ASA sweeteners 314±15 Asia 224 aspartame 179, 207, 292, 316, 354, 357, 378 advantages, disadvantages and applications 200 development into a commercial product 319 discovery 309, 406 individual differences in sweet hedonics 35 in sweetness intensity 33±4 interactions with acid 90 market economics 223 modification to make neotame 317 pharmacore models 286 PROP and sweetness intensity 38±40 QSAR studies 284 SAC gene and 47 safety, metabolic and physiological properties 193±4 sensory, chemical and physical properties 187 synergy 93, 345, 346 aspartame-acesulfame salt 179, 187±8, 194, 201, 207, 228, 234 associative learning 76±7 attentional strategies 71±2 automatic response 56 auxoglucs 283 AVI 37 baiyunoside 209, 293 bakery goods 147±8, 236 balanced incomplete block test designs 105 basal (progenitor) cells 7 behavioural effects 221 Beidler model 352 benzoindenopyrans 293, 295 beverages 145±6, 149, 234, 236, 256 diet beverages 226, 398 bimodal neurons 77±8 binary QSAR 312 binding sites 285±6, 340±1, 409 biotechnology 404±14 bitter blockers 410, 411 development of new low-calorie sweeteners 228±9
discovery of novel sweeteners 406±10 future trends 410±11 biscuits 147±8 bitter blockers 229, 235, 410, 411 bitter taste 3, 121, 407 bulk sweeteners 377 genetic differences in sweet taste perception 45±6 interaction of bitter compounds with sweeteners 86±8 bitterness reduction 88 mixture suppression 86±7 PROP see propylthiouracil PTC 36, 86, 87 bitter taste receptors 10, 12, 18±19, 228±9, 408 see also T2R38 gene blending sweeteners see sweetener blends blood glucose levels 220 body weight 119, 217±20 see also obesity boiled sweets 147 boiling point elevation 381±2 brain 77±8 brain-imaging techniques 78, 400 brazzein 209, 295 breakdown in-mouth 392±3, 395 breakfast cereal 236 brown sugars 141, 383 bulk sweeteners 375±87 characteristics 376±7 fructose syrups 253, 266, 267, 269, 271, 384, 385 gelling properties 382±3 glucose syrups see glucose syrups isomaltulose 253, 259±61, 275, 378, 384 lactose see lactose maltose 253, 271±2, 384 nutritional characteristics 376±7 physical functionalities 377, 380±3 colligative properties 381±2 hygroscopicity, humectancy and water activity 382 solubility and viscosity 380±1 sensory properties 377±80 appearance 380 flavour 377±9 sweetness 377 texture 379±80 sucrose see sucrose trehalose 253, 262±4, 275, 377, 378, 384 caffeine 20±1, 22, 408 cakes 147±8 calcium chloride 235 calming effect 56, 59 caloric alternatives 252±4, 259±75
Index fructose see fructose glucose see glucose hydrolysed lactose and galactose 264±5, 275 isomaltulose 253, 259±61, 275, 378, 384 lactose see lactose leucrose 253, 265±6 maltose 253, 271±2, 384 properties 253 starch hydrolysates 266±9, 275 trehalose 253, 262±4, 275, 377, 378, 384 cAMP 14±15 cancer preventive effects 302 capsaicin 121 caramelisation 144±5, 378 carbohydrate sweeteners interactions with acids 89±90 see also bulk sweeteners carbonation 89, 139 carboxyl-terminal tail (C-tail) 10, 11 carboxymethylcellulose (CMC) 390, 391 carrageenans 393 categorisation 76 category (discrete) scales 99±100, 102, 106, 353 central integration 77±8 cerebral blow flow (CBF) changes 78 cGMP 15 chemical irritation 121 chewing gum 168, 236 children 54±65, 115±17 experience and sweet taste preferences 58±9 ontogeny of sweet taste preferences 55±8 adolescents and children 56±7 foetus 55 individual differences 57±8 infants 55±6 physiological properties of sweet tastes 59±60 sweetness intensity measurement 101±2 T2R38 genotype 43±4 China 224 chiral isovanillic derivatives 335±6 chocolate 146±7 chorda tympani nerve 46 cinnamaldehyde 294, 299 cinnamic acid 91±2 circumvallate (CV) papillae 6, 7 citral 67 citric acid 89 click chemistry 312 cluster analysis 108 coating agents 165 Coca-Cola 319 Codex Alimentarius Commission 214 cold pressor test 60
417
colligative properties 381±2 colour 120, 121 sucrose and colour formation 144±5 combinatorial chemistry 312, 316 combinatorial libraries 312 commercial sweeteners discoveries 309±10 from discovery to exploitation 318±20 low-calorie sweeteners 177, 178±206, 207, 208 molecular design in development of 317 natural sweeteners 296±8 concentration concentrations of components and synergy 361±6 production of sugar from sugar beet 138 of sweetener and liking 107, 108±9, 112, 113 condiments 237 confectionery 146±7, 163, 237, 256, 381±2 congruency, taste-odour 72±3 consumer perceptions low-calorie sweeteners 226±7 see also sweet taste perception consumer preferences 226 consumption patterns (low-calorie sweeteners) 223±5 context effects 102, 106±7 continuous (line) scales 99 continuous response hedonic scale 106 cooling effects 158, 159, 163 corn starch 390 cos 357±8 cos u 358±61 cos UPL2 361±2 costs 222, 232 cream of tartar 235 cross-adaptation 32 cryoprotectants 165 crystallisation prevention of 380 regulation of crystallisation rate 381 sucrose 138, 139±40, 142 culture 74, 80, 114±15 cultured dairy products 261 curculin 209 cyclamate 180, 207, 284, 316, 354, 357 advantages, disadvantages and applications 202 blending 234, 345 development as a commercial product 318 discovery 309 safety, metabolics and physiological properties 194±5 sensory, chemical and physical properties 188
418
Index
cyclic nucleotide-gated channel (CNGgust) 15 cyclocarioside A 294 cysteine rich domain (CRD) 10, 11 cytoplasmic G proteins 14 D-tagatose see tagatose dairy desserts 237, 393 dairy products 237, 261 decolourisation 139 dental caries 161, 221±2 dental health 221±2 dental plaque 221±2, 265±6 deoxy monatin derivatives 338±9 deoxypentenosides 257 depolarisation 409 dereplication 299 descriptive sensory method 103±4 desserts 237 frozen 381±2 detection threshold 98±9 developing countries 225 dextrose equivalent (DE) value 266, 267, 269 diabetes 4, 216, 220±1, 302 diet beverages 226, 398 dietary behaviour 41 Dietary Reference Intakes 30, 216 differentiation threshold 98 diffusion coefficient 394±5 dihydroisocoumarins 293, 294 dihydroquercetin acetate methyl ether 295 dihydroxybenzoic acid 235 dimethylhexahydrofluorenedicarboxylic acid 293 dimethylphenylsulphonylalkanoic acid 314 disaccharides 299, 383±4 see also under individual names discrete (category) scales 99±100, 102, 106, 353 discretionary calories 31 dissimilarity scaling 104 distal stimulus 388±9 diterpenes 293, 302 docking software 288 dressings 237 dried products 263 drug-likeness concept 313 drugs 119, 307 drying 140 dual mechanism of taste±odour interactions 78±9 dulcin 213, 345 dulcoside A and B 212 dumping effect 75, 78±9 dusting agents 165 dynamic combinatorial chemistry 312
effector enzymes 407 elderly people 117±18 electrophile sites 285±6 electrostatic potential, molecular 286 energy bulk sweeteners 376±7 intake and sugar 30±1 enzymatic hydrolysis 252, 264, 267 epithelial sodium channel (EnaC) 408 equilibrium relative humidity (ERH) 382 equiratio model 367 erythritol 153±70 passim ethnicity 58, 60 Europe 224±5 European Food Safety Authority 214 European Union (EU) 224 Scientific Committee on Food 168±70, 214 Sweeteners Directive 215 evaporation 138, 139±40 excipients 148±9 experience and sweet taste perception 114±15 and sweet taste preferences 58±9 extracellular domains 10, 11, 409±10 extralingual papillae 6 Fahlberg, C. 318 fermentation 146, 237 field work 298±9 filiform papillae 6 fillings 237 filtration 139 first order neurons 7 first position bias 107 flavonoids 292, 293, 294±5 flavour balancing with functionality in food product development 398±9 bulk sweeteners 377±9 design and modification and low-calorie sweeteners 234±5 release and perception 388±9 role of sucrose in flavour formation 144±5 taste, smell and 66 flavour enhancement 81 high potency sweeteners 377±8 polyols 163±5 fluorescence polarisation assay 316 foetus 55 foliate papillae 6, 7 Food Chemistry 321 Food and Drug Administration (FDA) (US) 161, 168±70, 214 food intake 218 low-calorie sweeteners and 218±19
Index food product design genetic differences and 47±9 low-calorie sweeteners 230±5, 236±8 food product development bulk sweeteners in 375±87 hydrocolloid±sweetener interactions and 396±9 and sweetness perception 122 taste±odour interactions and 79±80 four-dimensional QSAR (4D QSAR) 311, 312 fragment approach 315 freezing point depression 381±2 frozen desserts 381±2 fructose 256±7, 272±3, 354, 357, 376, 378, 385 applications 273 excess dietary fructose 220, 221 high fructose corn syrups 253, 266, 267, 269, 271, 384, 385 interactions with acid 89±90 inversion of sugar 143±4 physiology 273 production 272 properties 253, 272 fructose intolerance 273 fruit 237 fudges 147 functional foods 400 functional magnetic resonance imaging (fMRI) 78 fungiform papillae 6, 7 density 37 sweet hedonics 42±5 sweetness intensity and 37±41 G protein coupled receptors (GPCRs) 7±14, 287±8, 340, 407±8 class C 9, 10, 408 signalling pathways 14±17 see also sweet taste receptors G proteins 8, 14, 407, 408 galactose 253, 256±7 hydrolysed 264±5 Gamma (ÿ) family models 361±6 ÿ index 361±2 ÿ model 361±2 ÿ0 model 362±6, 367 application to experimental data 363±6 iso-intensity diagrams 363, 364 ÿ-vector model 362 ÿ0 -vector model 362 gelatin 388, 399 gellan gels 393 gelled systems 392±3 gelling properties 382±3 gender 118
419
gene cloning 313 general labelled magnitude scale (gLMS) 100, 102±3 general purpose sweetener 215 generally recognised as safe (GRAS) 168±70, 215 genetic differences 30±53, 226 implications for designing foods 47±9 individual differences in sweet hedonics 35±6 in sweet taste intensity 31±4 PTC/PROP tasting 36±7 relating genetic taste markers with dietary sweet behaviours 41 relation of sweet liking to PROP, fungiform papillae density and T2R38 42±5 relation of sweetness to PROP, fungiform papillae density and T2R38 37±41 SAC gene 46±9 variation in other bitter markers 45±6 genetic modification 229 genetically evolved receptor models 284±5 genomics 229, 313 Gentiana lutea 88 gentiopicrin 88 geometric descriptors 334±5 global interactions 358±61 global liking, evaluation of 104±8 glucophore mapping 313±14 glucophores 283, 310, 330, 338 identification and role in designing new analogues 331±3 glucose 34, 252, 253, 256±7, 270±1, 354, 357 applications 271 inversion of sugar 143±4 occurrence, chemical and physical properties 270 physiology 270 production 270 glucose syrups 147, 269, 380, 383±5 colligative properties 381±2 gelling properties 382±3 hydrogenated glucose syrup (HGS) 155±6 low DE syrups and foam networks 383 glutamic acid 336 glycaemic index 161±2, 255 glycogen 270 glycosidic bond 384 glycosylation 300 glycyrrhizin 209, 235, 292, 294, 297, 300, 302, 328 GPCR kinases (GRKs) 17±19 inhibition by amphipathic tastants in vitro 19±23 granulated sugar 140±1, 383
420
Index
growth 57 guanidine-acetic acids 286 guanidines 213 guar gum 390, 391 gums non-starch hydrocolloids 388 sweets 147 gustducin (Ggust) 14, 15±16, 407 Gymnema sylvestra 86 haematoxylin 295 haemolysis 258 hardness of gels 392±3 harmony, tastant±odorant 72±3 health low-calorie sweeteners and 216±22 spending on 406 trends 216±17 see also diabetes; obesity heating, effect of 159, 378±9 hedonic dimension 104±11 evaluation of global liking 104±8 evaluation of ideal sweetness 108±11 heptahelical transmembrane domain (HD) 10, 11 hernandulcin 210, 293, 328 heteroatoms 331±3 heterotrimeric G proteins 14 high boil sweets 147 high fructose corn syrups 253, 266, 267, 269, 271, 384, 385 high intensity (high potency) sweeteners 149, 292, 375 bitterness reduction 88 flavour enhancing properties 377±8 interactions with acids 90±2 Hill model 351±2 homology modelling 287±8, 316, 409 honey 308 humectancy 382 humectants 165 hunger 218 hydrocolloid±sweetener interactions 388±403 future trends 399±400 gelled systems 392±3 implications for food product development 396±7 balancing functionality and flavour 398±9 criteria for hydrocolloid selection 399 flowchart 397 mechanisms 394±6 viscous liquid systems 390±2 hydrogenated disaccharides 154, 155 hydrogenated glucose syrup (HGS) 155±6 hydrogenated monosaccharides 154, 155
hydrogenated saccharides and polysaccharides mixture 154, 155 hydrogenated starch hydrolysate (HSH) 153±70 passim hydrolysed galactose 264±5 hydrolysed lactose 264±5, 275 hydrolysis 143±4 enzymatic 252, 264, 267 hydrophobic binding site 285 hydrophobic organic acids 346 hydroxyamino acids 346 hydroxyl group 338±9 hydroxypropylmethylcellulose (HPMC) 391 hygroscopicity 158±9, 160, 382 hyperactivity 221 ice creams 237, 382 icing sugars 141 ideal sweetness, evaluation of 108±11 impact factor (IF) of journals 321 in-mouth mixing 395, 399 in vitro assays with cloned receptors 329, 340 Index Kewensis 298 indices of interaction 350, 367±8 individual experience 115 individual variations 226 sweetness perception 112±20, 226 age 115±18 experience 114±15 gender 118 physiological factors 118±20 psychological factors 120 sweet hedonics 35±6 sweetness intensity 32±4 sweetness preferences and age 57±8 taste±odour interactions 74±5 taste perception 226, 229 infants 55±6, 115 inflexion point 351, 352 Institute of Medicine (US) 216 insulin 4 intellectual property protection 317 intensity, sweetness see sweetness intensity internal preference mapping 107±8 inversion 143±4 invert sugar (invert) 143±4, 145±6 invert syrups 141, 144 investment in R&D 320±1 isobole approach 367±8 isobolograms 368 iso-intensity diagrams 363±6 isomalt 153±70 passim, 378 isomaltase±sucrase complex 260 isomaltulose 253, 259±61, 275, 378, 384 applications 261
Index occurrence, chemical and physical properties 259±60 physiology 260±1 production 260 regulation 261 isoprenoids 292, 293 isovanillic derivatives 331±6 jams 238 Japan 224 jellies 147 Joint Expert Committee on Food Additives (JECFA) (FAO/WHO) 168±70, 214 Journal of Agricultural and Food Chemistry 321 Journal of Medicinal Chemistry 321 just-about-right (JAR) response scales 109, 110, 111 L-sugars 210, 256±8 applications 257, 258 occurrence, chemical and physical properties 256±8 physiology 257, 258 production 257, 258 labelled magnitude scale (LMS) 100±1, 102±3 lactisole 316 suppression 11±12 lactase 274 lactitol 153±70 passim lactose 253, 273±5, 376, 380, 384 applications 275 hydrolysed 264±5, 275 occurrence, chemical and physical properties 273±4 physiology 274 production 274 lactose intolerance 274 lactulose 253, 258±9 applications 259 occurrence, chemical and physical properties 258 physiology 259 production 259 Laffort model 352 Latin America 224 laxation threshold 162 laxative effects 255, 259 lead structures 312 learned synaesthesia 76±7 lemon 74, 79 leucrose 253, 265±6 applications 266 occurrence, chemical and physical properties 265
421
physiology 265±6 production 265 ligand based design 310±12 line (continuous) scales 99 lingering aftertaste 5, 19, 22±3 liquid sugars 141 literature search 298 low-calorie products consumption in US 160, 161 most popular products in US 165±6 low-calorie sweeteners 4, 153, 175±251 advantages, disadvantages and applications 199±206 commercially available 177, 178±206, 207, 208 consumer perceptions and attitudes 226±7 consumer preferences 226 consumption patterns 223±5 development of new low-calorie sweeteners 227±9 food product design 230±5 blends containing low-calorie sweeteners 230±4 flavour design and modification 234±5 functionality 231, 235, 236±8 future market trends 235±9 health-related developments 216±22 history and overview 178±85 market economics 222±3 metabolic and physiologic properties 193±8 natural intense sweeteners 177, 209±12 properties of perfect low-calorie sweeteners 176 regulation 177, 178±85, 214±15 safety 177±213 sensory, chemical and physical properties 186±92 structures 207, 208 synthetic intense sweeteners 177, 213 technological advances 239 uses 176 luzindole 13 mabinlin I and II 210 mageu 225 magma 139 magnitude estimation 100, 102, 353 Maillard reaction 144±5, 378 malic acid 90 maltitol 156±70 passim maltooligosaccharides 271 maltose 253, 271±2, 384 applications 271±2 occurrence, chemical and physical properties 271 physiology 271
422
Index
production 271 mannitol 153±70 passim, 354, 357 mannose 377 market trends 235±9 markets for low-calorie sweeteners 223±5 marmalades 238 mass transport effects 394±6 massecuite 140 MDI (Mixture Discrimination Index) model (substitution model) 366±7 medicinal products 119, 148±9, 238, 307 melanoidins 144, 145 melting 139 behaviour of gels 392±3, 395 melting range 253 methyl xanthines 408 mGluR1 receptor 340, 409 mGluR4 receptor 408 mid-cal products 239 MIG programme 363, 365 milling 147 mixing, in-mouth 395, 399 mixture suppression 85 sweet and bitter compounds 86±7 mixtures 40±1 sweetener blends see sweetener blends modified starches 388, 399 modified sucrose reference scale 369 mogroside IV 210 mogroside V 210, 293, 297±8, 302 moisture sorption isotherms 158±9, 160 molasses 138 molecular design 307±26 commercial alternative sweetener discoveries 309±10 in commercial sweetener development 317 development of commercial products 318±20 future trends 320±1 novel approaches and chances 310±13 screening and visualising candidates for new sweeteners 313±16 monatin 211, 295, 328 synthesising new sweeteners from 336±9 monellin 8, 211, 295, 301 monosaccharides 299 monoterpenes 293 mothers' milk 54, 59 mouthfeel 398 multidimensional sensory evaluation methods 107±8 multidimensional similarity (MDS) analysis 4 multiple effect evaporation 138 multiple paired comparison tests 105
multiple sweet receptor mechanisms, evidence for 32 multipoint attachment (MPA) model 4, 313±14, 330 multisensory interactions 80±1, 121, 396 N-carbamoyl dipeptides 286 naloxone 4 naltrexone 119 naringin 20±1, 22 natural (sweet) compounds 327±43 future trends 340±1 importance of developing new sweeteners from 327±9 methods of designing new sweeteners from 329±36 active conformation 333±5 identification and role of potential glucophores 331±3 role of stereochemistry 335±6 synthesising new sweeteners from 336±9 natural sweeteners 177, 209±12, 292±306 approaches to discovery 298±300, 302 commercially used 296±8 development of new natural sweeteners 229 future trends 301±2 improvement of sweet taste 300±1 from plants 292±6 polyols 153, 154 negative feelings 221 neohesperidin dihydrochalcone (NHDC) 13, 20±1, 181, 207, 292, 294, 297, 331, 354, 357 advantages, disadvantages and applications 202±3 and flavour enhancement 378 safety, metabolic and physiological properties 195 sensory, chemical and physical properties 188±9 neosmin 235 neotame 181, 208, 292, 309, 316, 345, 346 advantages, disadvantages and applications 203 development into a commercial product 319 interactions with acids 91±2 modification of aspartame 310, 317 safety, metabolic and physiological properties 196 sensory, chemical and physical properties 189 neural networks 311±12, 314, 316 neurons 7 bimodal 77±8
Index neurotransmitter release 407, 409 nine-point hedonic scale 106 nitroanilines 284, 314 no-observed-adverse-effect-level (NOAEL) 177 non-taste receptors in taste cells 13±14, 23 novel foods 400 novel sweeteners, discovery of 406±10 nucleophile sites 285±6 nutrigenomics 229 Nutritional Task Force (UK) 217 obesity 4 diseases related to 406 global problem 149, 404, 406 low-calorie sweeteners and 217±20 US 216, 404±6 OBM (Olfactory Binary Mixtures) model 368±9 odour-induced taste enhancement 67±9 odour-induced taste suppression 67 odour±taste interactions see taste±odour interactions off-taste suppression 344, 369±70 older people 117±18 olfactory referral 67 opioid antagonists 119 oral shear 395 orbitofrontal cortex 77±8 organoleptic testing 299 orthobenzoyl sulphimide 309 osladin 211, 294 osmotic pressure 381 oximes 213 pain reduction 59±60 paired comparison test 99 paired preference tests 104±6 palatability 218 Panel on Food Additives, Flavourings, Processing Aids and Materials in Contact with Food (AFC) 214 patents 317 PAV 37 pentadin 211 perceptions of sweet taste see sweet taste perception perceptive approach 350 relating perceptive scales to physical scales 353±6, 373±4 perceptive models 356±61 application to experimental data 358±61 U model 358 vector model 356±8 periandrins 211 perillartine 213, 293 peripheral gustatory system 6±7
423
persistence after cooking 370 pH 378±9 see also acids pharmaceuticals 119, 148±9, 238, 307 pharmacogenomics 313 pharmacology 367±8 pharmacore mapping 310±12 pharmacore models 285±6 phenylpropanoids 293, 294, 299 phenylthiocarbamide (PTC) 36, 86, 87 phosphodiesterase (PDE) 408 phosphorylation 17 phyllodulcin 211, 294, 298, 328, 331 physiology effects of sweet tastes 59±60 impact of physiological factors on perceptions of sweetness 118±20 properties of low-calorie sweeteners 193±8 properties of polyols 160±2 PKA 17 PKC 17 plants approaches to discovery of natural sweeteners 298±9 natural sweeteners of plant origin 292±6 PLC 2 16 Polycose 398 polyols 149, 153±74, 235, 299 advantages and disadvantages 162±3 applications 163±6 future trends 170±1 information resources on the web 171, 172 mixed sweetener potential 166±8 physicochemical and functional properties 156±60 physiological properties 160±2 regulatory status 168±70 types, chemical structure and manufacture 154±6 polypodoside A 211, 294 positron emission tomography (PET) 78 power law 352 values of power law exponent 353±6, 357 prebiotic effects 161, 255 precaution principle 371 predictive modelling 283±91, 410±11 pharmacore models 285±6 QSAR models 284±5, 311, 312, 314 receptor models 287±8, 338 preserves 238 Principal Component Analysis (PCA) 335 proanthocyanidins 293, 295 product development see food product development
424
Index
progenitor (basal) cells 7 propylene glycol alginate 398 propylthiouracil (PROP) 36, 87 discordance between PROP and quinine bitterness 45±6 relation of sweetness intensity to 37±41 supertasting and 37 sweet hedonics 42±5 protein kinases 17±19 proteins 292, 293, 295 proximal stimulus 389 pseudoreceptor models 287, 316, 338±9 psychohedonic functions 112, 113 psychological factors 120 psychophysical functions 99, 112, 113, 350 sweet taste measurements 350±6 pterocaryoside A 212, 294 pterocaryoside B 212 pullulanase 267 purification 138 qualitative sweetness synergy 232 quality, sweetness 103±4 quantitative structure-activity relationship (QSAR) models 284±5, 311, 312, 314 quantitative sweetness synergy 232 quinine 43, 86±7 discordance between PROP and quinine bitterness 45±6 racemic monatin 337 ranking method 99 rapid taste onset 346 rating conditions 70±1 rating scales 75 rebaudioside A 212, 292, 293, 297, 300, 355, 357 rebaudioside A-enriched stevia extract 301 rebaudioside B, C, D, E 212 receptor binding assays 302 receptor-independent mechanisms 16±17 receptor mapping 310±12 receptor models 287±8, 338 receptors, taste see sweet taste receptors; taste receptors recognition threshold 99 recombinant DNA technology 409 recommended intake of sugars 30±1, 60, 216±17 reduced-calorie sweeteners 153, 253, 254±9, 275 L-sugars 210, 256±8 L-sorbose 253, 257±8 lactulose 253, 258±9 tagatose see tagatose
regulation isomaltulose 261 low-calorie sweeteners 177, 178±85, 214±15 petitions for new sweeteners 215 polyols 168±70 specifications in sweetener regulations 215 tagatose 256 trehalose 263±4 regulators of G protein signalling (RGS) proteins 14 relative concentrations 353±6, 373±4 relative-to-ideal procedure 109±10, 111 Remsen, I. 318 research and development (R&D), investment in 320±1 response alternatives 70±1 response bias 75±6 response scales hedonic measurement 106±7 sweetness intensity 99±100 reversion 267 rhodopsin 20±2 rigid isovanillic derivatives 334 roller refining 147 rubusoside 212 SAC gene 46±9 Sac locus 8±9, 46 saccharin 5, 20±1, 181±2, 208, 292, 354, 357, 406 advantages, disadvantages and applications 203±4 bitterness reduction 88, 410 blends 234, 345, 346 development as a commercial product 318 discovery 309, 314 interactions with hydrocolloids 390±1 market economics 223 safety, metabolic and physiological properties 196±7 sensory, chemical and physical properties 190 stimulation of non-taste receptors 13 safety 177±213 saliva 7 in-mouth mixing 395 salivary flow 118±19 salt 235 salty taste 121, 407, 408±9 interactions of sweet and salty compounds 92 satiety 119±20 Scientific Committee on Food of the European Union 168±70, 214
Index scotch 40, 41 Scottish Intercollegiate Network (SIGN) 161 screened sugars 141 screening 313±16 second messengers 407 second order neurons 7 self-organising neural networks (SOM) 311±12 selligueain A. 295 Senomyx 316, 320, 321 senses multisensory interactions 80±1, 121, 396 taste±odour interactions see taste±odour interactions see also taste sensitivity, sweetness 98±9 sensory evaluation 299±300, 328 sweet taste perception see sweet taste perception training for 71±2, 80 separation 140 serendipity 308, 309±10, 330 sesquiterpenes 293 setts 138 Shallenberger theory 283 shear thinning 399 signal termination mechanisms 17±23 signalling pathways 14±17 similarity, tastant±odorant 72±3 similarity paradox 311 single point mutation experiments 288 `slow carbs' 149±50 slow taste onset 5, 346 smell 66, 396 see also taste±odour interactions smelled taste 72±3 sodium carboxymethylcellulose (NaCMC) 390, 391 sodium saccharin 390±1 solubility 253 bulk sweeteners 376, 380±1 sucrose 142±3, 380 sorbitol 153±70 passim, 355, 357 sorbose 253, 257±8 applications 258 occurrence, chemical and physical properties 257±8 physiology 258 production 258 sour taste 121, 407, 408±9 specific binding effects 394 stability 233 starch gels 393 starch hydrolysates 266±9, 275 applications 269 composition 268
425
hydrogenated starch hydrolysate (HSH) 153±70 passim physiology 269 production 267±9 properties 266±7 starches 388, 399 modified 388, 389 stereochemistry 335±6 stereoisomers of monatin 337±8 steric barrier 285 steric descriptors 334±5 STERIMOL parameters 284, 314 steroidal saponins 294 Stevens's law see power law stevia extract 301 stevia sweeteners 300±2, 410 see also under individual names steviolbioside 212 stevioside 183, 208, 227, 292, 293, 297, 302, 328, 355, 357 advantages, disadvantages and applications 204 improvement of sweet taste 300 safety, metabolic and physiological properties 197 sensory, chemical and physical properties 190 stimulus±response diagrams 350±2, 353±6 storage of sugar 138, 140 strawberry flavour 74, 391±2 structural modification 300±1 structure±activity relationships (SAR) 310±11, 329, 334±5, 336, 409, 410±11 see also quantitative structure±activity relationship (QSAR) models structure based design 310±13, 320 structure±taste relationships 283±91 future trends 288 pharmacore models 285±6 QSAR models 284±5, 311, 312, 314 receptor models 287±8, 338 substitution model (MDI model) 366±7 sucralose 149, 184, 208, 292, 336, 345, 406 advantages, disadvantages and applications 205 discovery 309±10 interactions with acids 90±1 safety, metabolic and physiological properties 198 sensory, chemical and physical properties 191 sucrose (sugar) 34, 135±52, 253, 328, 355, 357 bulk sweetener 383±4, 385 demand 252 and diabetes 220
426
Index
flavour modification and enhancement 378 flavour volatility and concentration of 379 functionality in food products 145±9 future trends 149±50 history 135±7, 308 manufacture 137±40 mixtures with citric acid 89 polyols compared with 156±8, 159 prices 308 production and consumption 136, 137 products 140±1 properties 142±5 colligative properties 381, 382 colour and flavour formation 144±5 crystallisation 142 functional properties 230, 231 inversion 143±4 solubility 142±3, 380 sweetness perception individual differences in hedonics 35±6 individual differences in intensity 33±4 PROP and sweetness intensity 38±40 SAC gene, intensity and hedonics 47, 48 synthesising new sweeteners from 336 and texture 379±80 UK usage of 145, 146 water activity 382 sucrose equisweet matching 353 sugar beet 135, 136±7, 252, 308 processing 137±8 sugar manufacture from 137±8 sugar cane 135±6, 252, 308 sugar manufacture from 138±40 sugar `doctors' 147 sugar-free chewing gum 168, 236 sugar-free products consumption in US 160, 161 most popular products in US 165±6 sugar-transferred stevia extract 301 sulphamates 284 Sunnet 319 see also acesulfame K Sunnet Nutrinova Multi-Sweetener Concept 319 Sunny Delight 398 superadditivity effects 78 supertasters, PROP 37 supply constraints 239 `Sweet 'n' Low' 88 sweet taste perception 97±131, 350±6 comparison of techniques 353 factors related to food 120±1 future trends 121±2 individual variations 112±20
methods to determine 97±111 hedonic dimension 104±11 sweetness intensity 99±103 sweetness quality 103±4 sweetness sensitivity 98±9 temporal dimension 103 stimulus±response diagrams 350±2, 353±6 values of anchorage points and power law exponent 353±6, 357 sweet taste receptors 7±14, 287±8, 316, 340±1, 407±8, 409±10 development of new low-calorie sweeteners 228±9, 239 discovery of 8±12 discrimination of taste modalities at taste bud cells 12±13 mediation of sweet taste by G protein coupled receptors 7±8 stimulation of non±taste receptors by sweet tastants 13±14 sweet taste reference scales 369 sweet tooth 120 sweet water aftertaste 5 sweetener blends 93±4, 344±8 additives 346 containing low-calorie sweeteners 230±4 future trends 347 hydrocolloids and 398±9 polyols 166±8 synergy see synergy taste profile 345±6 temporal profile 346 Sweeteners Directive 215 sweetening power relative to sucrose (tonnes sucrose equivalent) 158, 159, 222±3 sweetness bulk sweeteners 377 historical development of consumption 308±9 palatability and food intake 218 sweetness intensity (potency) 376 individual differences in 31±4 measurement 99±103 PROP, fungiform papillae density and T2R38 37±41 sweetness intensity ratings (SIR) 353±6, 357, 373±4 application of perceptive models to experimental data 358±61 sweetness preference/liking 3±4, 104±11 children see children and concentration 107, 108±9, 112, 113 evaluation of global liking 104±8 evaluation of ideal sweetness 108±11 individual differences 35±6, 57±8
Index relation to PROP, fungiform papillae density and T2R38 42±5 SAC gene 46±9 sweetness potentiation 410 sweetness quality measurement 103±4 sweetness sensitivity 98±9 SweetScreenHT 316 synaesthesia, learned 76±7 synapses 7 synergy 4, 32, 92±4, 120, 349±74, 410 blending sweeteners to provide 344±5 blends containing low-calorie sweeteners 230±4 defining 350 equiratio model 367 future trends 369±71 interaction in mixtures with more than two components 369 modified sucrose reference scale 369 off-taste suppression 369±70 other properties of artificial sweeteners 370±1 Gamma family models (response of a mixture vs concentrations of its components) 361±6, 367 isobole approach 367±8 OBM model 368±9 perceptive models (response of a mixture vs responses to its components) 356±61 polyols and sweetener blends 166±8 substitution model 366±7 sweet taste measurements 350±6 comparison between techniques 353 stimulus±response diagrams 350±2 values of anchorage point and power law exponent 353±6, 373±4 synthetic accessibility of derivatives 329 synthetic intense sweeteners 177, 213 T1R receptors see sweet taste receptors T2R receptors see bitter taste receptors T2R38 gene 37, 57±8 sweet hedonics 42±5 sweetness intensity and 37±41 table-top sweeteners 238 tagatose 185, 208, 253, 254±6, 275, 378 advantages and disadvantages 205 applications 205, 256 chemical and physical properties 191±2, 254 physiology 198, 255±6 production 255 regulation 256 safety and metabolic properties 198 sensory properties 191±2 tannic acid 90±1, 235
427
target protein expression 313 taste 404, 407 blending sweeteners to improve taste profile 345±6 hydrocolloids and taste sensations 397±8 interactions between tastes 120, 121 peripheral organisation of 6±7 structure±taste relationships see structure± taste relationships taste buds 6±7, 407 discrimination of taste modalities at taste bud cells 12±13 taste cells 3±29 cellular mechanisms for signal termination 17±23 downstream signalling components 14±17 permeation by amphipathic tastants in vivo 19±23 stimulation of non-taste receptors by sweet tastants 13±14 sweet taste receptors see sweet taste receptors taste±ingredient interactions 85±96 future trends 94 interactions between sweet compounds 92±4 sweet and salty compounds 92 sweeteners and acids 88±92 sweeteners and bitter compounds 86±8 taste±odour interactions 66±84, 120 factors affecting 69±75 stimuli-driven factors 72±3 subject-driven factors 73±5 task-driven factors 70±2 future trends 80±1 implications for food product development 79±80 literature review 67±9 mechanisms of 75±9 viscous hydrocolloid solutions 391±2 taste papillae 6 taste pores 7, 395±6 taste receptors 7, 228±9, 239 bitter 10, 12, 18±19, 228±9, 408 sweet see sweet taste receptors tuning taste receptor cells 32 see also receptor mapping; receptor models taste stripes 6 tastomics 229 technological advances 239 technological constraints 239 telosmoside A15 294 temperature solubility of sucrose and 142, 143 and sweet taste perception 121 temporal profiles 5, 103
428
Index
sweetener±acid interactions 90±2 sweetener blends 346 termination mechanisms 17±23 texture 120, 121, 388±9, 393, 396 bulk sweeteners and 379±80 texture sensitivity 370 thaumatin 8, 185, 227, 235, 292, 295, 298, 355, 357, 378 advantages, disadvantages and applications 206 safety, metabolic and physiological properties 198 sensory, chemical and physical properties 192 theophylline 22, 408 thioacetalic isovanillic derivatives 333, 335±6 third order neurons 7 three-alternative forced choice (3-AFC) test 98 three-dimensional QSAR (3D QSAR) 311 time±intensity methodology 103 see also temporal profiles tomato 92 topological models 330 toppings 237 toxicity testing 299, 328±9 training 71±2, 80 trans-anethole 294, 299 trans-cinnamaldehyde 294, 299 transient receptor potential V1 (TRPV1) channel 408 trehalase 263 trehalose 253, 262±4, 275, 377, 378, 384 applications 263 occurrence, chemical and physical properties 262 physiology 263 production 262 regulation 263±4 triacylglycerol 221 trigeminal system 67 trihalogenated benzamides 213 triterpenes 293±4, 302 TRPM5 (transient receptor potential ion channel) 16 tryptophan 13, 235, 336 derivatives 213 two-alternative forced choice (2-AFC) test 98 Type II receptor cells 7
U model 358 application to experimental data 358±61 umami taste 407 receptors 10, 12, 407±8 United Nations Convention of Biological Diversity 298 United States (US) consumption of low-calorie, sugar-free products 160, 161 consumption of low-calorie sweeteners 223±4 Department of Agriculture (USDA) 216±17 Food and Drug Administration (USFDA) 161, 168±70, 214 health trends 216 Institute of Medicine 216 most popular low-calorie, sugar-free products 165±6 obesity 216, 404±6 trends in health-related behaviour 216 urea derivatives 213 vanilla 73, 74, 79 vapour pressure 381, 382 vector model 356±8 vegetables 237 Venus flytrap module (VFTM) 10, 11 vermouth 398 viagra 313 viscosity 121, 398 bulk sweeteners 380±1 viscous liquid systems 390±2 visualisation 311±12 volatility, flavour 378±9 von Ebner's salivary glands 7 water activity 382 wedge site 409 weight control 217±20 weight status 119 whey 274 white sugar 140±1 World Health Organization (WHO) 216 xanthan 399 xanthan cryogels 399 xylitol 153±69 passim yoghurts 237, 396 yohimbine 13