ISBN

ISBN 978-961-6379-26-7 (pdf)

Full title



The DREAM Project Book of Results

Publisher

University of Ljubljana, Biotechnical Faculty

Jamnikarjeva 101

1000 Ljubljana, Slovenia

Editorial Board

Monique Axelos, Institut National de la Recherche Agronomique,

France

Matthijs Dekker, Wageningen University, The Netherlands

Paola Lavermicocca, Consiglio Nazionale delle Ricerche - Istituto di

Scienze delle Produzioni

Alimentari, Italy

Alan Mackie, Institute of Food Research, United Kingdom

Nathalie Perrot, Institut National de la Recherche Agronomique,

France

Kaisa Poutanen, VTT Technical Research Centre of Finland

Andras Sebok, Campden BRI Magyarország Nonprofit Korlátolt

Felelősségű Társaság, Hungary

Peter Raspor, University of Ljubljana, Biotechnical Faculty, Slovenia

Carolina Realini, Institut de Recerca i Tecnologia Agroalimentàries,

Spain

Technical Editor

Lidija Baša, University of Ljubljana, Biotechnical Faculty,



Ljubljana, Slovenia

Peter Raspor, University of Ljubljana, Biotechnical Faculty,

Ljubljana, Slovenia



Cover

Lidija Baša, University of Ljubljana, Biotechnical Faculty,

Ljubljana, Slovenia



Printed By

Littera picta d.o.o., Ljubljana, Slovenia

Numbers of Copies

30 copies





Acknowledgment

“The research leading to these results has received funding from the European Community’s

Seventh Framework Programme (FP7/ 2007-2013) under the grant agreement n°FP7-222 654”





Chapters


FW


Foreword

6

Mathematical knowledge integration for food model

WP-1

numeric simulation

7

WP-2

Filled cellular solid model

20

WP-3

Proteinous cel ular network model

30

WP-4

Combined gel ed / dispersed / aerated systems model

39

WP-5

Open solid foams model

49

WP-6

Model food applicability

57

WP-7

Technology transfer

65

WP-0&8

Management & Dissemination

69

CC

Being curious? Consumers clip for you.

77

CONC

Conclusion

79

PROS

Perspectives

81

5

Foreword

DREAM for “Design and development of REAlistic food Models with well-characterized micro- and macrostructure and composition” was an EC-funded FP7 large-scale integrating collaborative project who started in 2009 for almost five years. DREAM embarked 18 partners from nine countries on a very stimulating scientific and technological adventure around the question: how could the consumer demand for safe, high quality food be supported by developing modeling approaches?

The DREAM ambition was to develop food models realistic enough to be used by the industry and



sufficiently versatile to be used as predictive tools to understand the impact of changing composition and/or processing conditions on food quality. The ultimate goal is to save time in the development of an innovation and to provide decision support for food professionals.

In this book we present the results obtained by the different teams on the four generic categories of food.

These categories were chosen according to their generic structures: filled cellular solid (fruit and vegetables), proteinous cellular network (meat), combined gelled/dispersed systems (dairy products and cheeses) and open solid foam (cereal products such as bread and biscuits). In practice, on each product, the DREAM

project has developed three types of models. First, generic models foods for which the relationships between the structure, chemical composition and functional properties are known. For these realistic physical models, several parameters can be varied (fiber content in cereal products, type of fat in dairy matrices…), leading to a series of samples representative of each type of food studied. Second, in-silico food models which allow simulating for example the role played by the temperature, pressure, chemical composition, etc. on the food structure and resulting material properties. Third, integrated knowledge models to bring together the technical expertise gained by professionals and the data obtained from research. This modeling uses the most advanced recent cognitive science and complex systems science to extend the methods developed in other food products. Such models also allow assessing the impact of processing conditions on the nutritional food properties and safety issues.

The applicability of the model foods and food models was assessed before transferring the protocols and disseminating the gained knowledge to industry and other stakeholders through demonstration training.

To support the practical application of realistic food models a Practical Guideline on use of models was developed. The main objective of the guideline was to provide an overview on different food models and on modeling tools/software for the industrial users in order they benefit from models to optimize their existing processes or to come up with new ones to develop innovative products.

Dissemination to the general public, students, scientists, industry, food authorities was made through DREAM

website, E-news releases, peer reviewed publications, workshops, training and congresses and at the DREAM

international conference. All results are available on the DREAM web site: http://dream.aaeuropae.org/

Finally, I would like to thank all colleagues and teams who have contributed to the DREAM project and con-gratulate them for their achievements. It was also a stimulating collaborative public-private research challenge and an incredible human adventure.

Monique Axelos

INRA - Head of the division for science and process engineering of agricultural products

DREAM project Coordinator

6



Mathematical knowledge

integration for food

model numeric

simulation

It this information important for me?

Consumers

numeric simulation

Food Industry

Regulatory Authorites

Scientific Community

mathematical knowledge

WP-1

Workpackage



7

Task 1.1

Generic structure for the modelling approach:

The objective of the task was to propose a structure

of the way to integrate knowledge that could be

available for all the cases studied in WP2 to 5.

Task 1.2





Introduction of WP1


Construction of the numeric food model and

uncertainty management: In deep interaction with

the researchers of WP3 to 6, mathematical concepts

and methods have been established and applied to

food models of DREAM. The main objective in this

task was to integrate the knowledge (generated

Nathalie Perrot, Workpackage Leader

by the experimental trials in WP3 to 6), the techni-

UMR 782 Génie et Microbiologie des Procédés Alimentaires GMPA,

cal expert knowledge extract from WP1T1.1 and the

nathalie.perrot@grignon.inra.fr

uncertainty on the knowledge as to reconstruct the

dynamic network at each time linking the scales of

the system under process conditions.

Aims and background

Task 1.3

The challenge of the WP1 was to develop or adapt

Reverse engineering using the numeric food

models: In this task, the challenge was to develop

new applied mathematical tools able to predict the

methods able to help to determine the sequences

emerging organization of a food model at different

of actions driving the food process along admissible

scales and the functions associated to it, so called

evolutions and requirements. The problem was turn

IKM’s (Integrated Knowledge Model). To do that,

into a theoretical adap Ttip

ati !on of a viability approach

initially developed Saint-Pierre (1994) for dimension

several authors had shown the necessity to resolve

lower than 6.

the enormous challenge of unifying complex and

dissimilar data, knowledge and models, specifically

to understand the dynamics of such a complex food

system (Perrot et al., 2011). It is particularly true

Main outcomes

when it is applied to real systems. Food structure

prediction and structure-function relationships are

Numeric IKM’s able to answer to the problematic fixed

only well established for simple – or “simplistic”

under the WPs have been developed. Manage the uncer-

– models of foods (gels, emulsions, dry foams…),

tainty and lack of knowledge on some phenomenon was

which are far from the foods they are aimed to rep-

a prerequisite. Approaches of modeling, coupling sto-

resent, at least for their composition and function,

chastic and mechanistic approaches in the form of equa-

and thus useless with respect to industry’s needs.

tions or graphical models has thus been developed and

Nevertheless the improvements of food processes

applied to food models. For example an IKM has been

by optimization methods are restricted to the few

developed and validated as to contribute to the under-

applications where the mathematical modeling is

standing of the structure and texture dynamics of a dairy

complete (Banga et al., 2003). In this context the

oil-in-water emulsion. It has been applied to a dairy dessert

development of mathematical approaches having

(Foucquier et al., Descamps et al.) and generalized to a

the ability to take into account the heterogeneous

cream cheese model (Jelinko, D.) in col aboration with

knowledge and simultaneously the uncertainty on

INRA Nantes and IFR Norwich. An IKM mimicking the

the system were promising and has been investi-

thermal treatment of a broccoli including quality degrada-

gated in DREAM.

tion and microbial Inactivation has also been developed

to be used for decision support purposes in col aboration

with the WUR Wageningen (Gaucel, et al) and the ADRIA

Prospects

Quimper. A graphical model of biscuit baking was also

Promising results have been reached to facilitate the

in test in INRA, GMPA in col aboration with UB. Theoriti-

task of numerical modeling in food science. Such

cal developments at the interface of applied mathematics

models can be of a precious help in a context of

and computing have in parallel been proposed (1) to help

sustainability, either through an help for a better

and simplified the modeling task (Tonda, Lutton et al.,

understanding of the emerging organization of the

2013). (2) to help the reverse engineering task by a cou-

food model during its elaboration or by optimization

pling between viability theory and methods of clustering

of the process based on numerical simulators.

(Alvarez et al.,2013) and otpimisation.

8



In silico comprehension and prediction

of the structure and texture of a

dairy dessert

Dairy products have been experimentally shown to behave like complex systems: Their resulting textures depend on various factors, including their composition and their processing conditions. Out of these processing conditions, the most influential are: the nature of heat treatment, processing parameters applied during acidification and during the homogenization process. Nevertheless being able to predict the texture upon process conditions is an interesting challenge for the industry. The work developed under the WP1 was to answer to this challenge by developing an in silico model able to integrate the available knowledge and the uncertainty on the domain. The originality of the approach is about the integration of recent theoretical developments crossing over applied mathematics and computing science. The predictions of the model have been validated on experimental data reached under the WP4 applied to different neutral dairy emulsions and generalized to a cream cheese model.

Research aims and background

Dairy products have been experimentally shown to

different domains (physical chemistry, microbiol-

behave like complex systems: Their resulting tex-

ogy, computer science, applied mathematics, etc.).

tures depend on various factors, including their

dairy dessert, the story starts at a lower scale :

composition and their processing conditions. Out

the nanoscale level (Descamps, 2013) (Figure 1). A

of these processing conditions, the most influen-

dairy dessert is an oil in water emulsion stabilized

tial are: the nature of heat treatment, processing

by milk proteins. From the behavior of each type

parameters applied during acidification and during

of the particles, whey denaturated proteins or na-

the homogenization process (Foucquier, 2011). From

tive whey proteins, agregates of proteins and casein

an industrial point of view texture of dairy products

micels, depend the organization of the macrostruc-

have crucial importance.

ture of the gel. Particles are indeed in competition to

Consumer appreciation of dairy desserts such as

colonize the interface of a fat globule. From this dy-

gels, yogurts or cream cheese variants is influ-

namic system is reached a more or less regular and

enced by the texture of the product. Due to diffi-

stabilized fat globule interface. From this individual

culties of thorough experimental product charac-

interface and the connexion between all of them,

terization along the production chain, mathematical

is going to emerge a connected organization at an

simulation and modeling approaches are well suited

upper level. We propose an integrated model that

tools to gain deeper understanding of how the com-

mimic in silico this competition. Knowledge of the

position and some of the processing parameters

laboratories working in the WP4 has been integrated

can be related to the final structure of the prod-

in the structure and some parameters of the model.

uct. Nevertheless the task is not easy because of

Last developments crossing over applied mathemat-

the complex interactions that can occur between

ics and computing science have been included in the

key variables at different scales (Perrot, 2011)

mathematical functions allowing to couple determin-

if we want to mimic a real system. The model



istic and stochastic algorithms.

proposed is based on knowledge originated from

9





Significance and benefits

The model is a coupling between a first order dif-

ferential kinetic model of protein denaturation (M1)

and a stochastic model of simulation of the fat drop-

lets interface colonisation (M2a) and bridge creation

between fat droplets (M2b) (Figure 2). Coupling M1

and M2 allows to simulate the emergence of a net-

work at a mesoscale level from local droplet consid-

erations. The inputs needed are the initial relative

concentration of each particle in the solution, the

droplet size and volume distribution and the thermal

denaturation rate. The fat droplet local interface or-

ganization is predicted through two dimensions: the

interface composition: % of the different particles

Figure 1: Interaction of casein micelles (only interacting

parts are shown), casein from heated milk with attached

fixed at the interface and interfacial concentration. A

whey protein/κ-casein complexes. These complexes

good prediction is observed for data reached during

prevent coagulation even if the κ-casein is removed and

provide points of attachment between micelles during acid

experiments led by the WP4 partners (Descamps,

gelation (Dalgleish & Corredig, 2012).

DREAM congress). The macroscopic structure is ex-

plorated upon the number and organization of the

links between the particles in the space. For exam-

ple for the three experiments represented figure 3,

the percentage of caseins has been evaluated by the

experts to have a strong influence on the perception

of the texture of the product which has been shown

in the literature to be in link with the number of fat

droplet connections. The prediction is in good ac-

cordance with this knowledge. Thus for a ratio 80/20

of casein/whey proteins, a weak viscosity is mea-

sured (40 Pa.s at 0.001s-1) and few local interfaces

Figure 2: an IKM in silico model of a dairy dessert includ-

are predicted to be connected. In the contrary for a

ing heat treatment and homogenization

ratio 5/95, the gel is structured with a viscosity of

2600 Pa.s at 0.001s-1) with a prediction of a highly

connected structure.

This approach has been generalized and applied in

collaboration with data coming from the industry

to a cream cheese model (Jelinko, 2013). The ex-

perimental data and model simulations show similar

tendencies, also demonstrated in previous studies,

where increasing homogenization pressure in cream

cheese production resulted in increased storage

modulus (‘G) in the interval of (0 – 50 Mpa) and in-

creased texture and number of links.

Figure 3: Example of the in silico IKM prediction at the

macroscopic level

10



Prospects and challenges

This work is a first contribution able to predict real

dairy dessert structures crossing over the scales.

The fertilization of the different disciplines: food,

computing and applied mathematics has led to a

model of understanding and comprehension that

can be used by the industry for optimization or sus-

tainability purposes.

Results and applications

A generalisation of this approach is needed to be

sufficiently generic to cover the wide range of dairy

desserts produced in the industry. For that a deeper

understanding of what it takes place at the interface

is necessary coupled to complex system tools devel-

oped specifically to handle this understanding and to

propose way of process optimization.

References

Descamps, E., Perrot, N., Trelea, C., Gaucel, S.,

Riaublanc, A., Mackie, A., Lutton, E. (2013). Coupling

deterministic and random sequential approaches for

structure and texture prediction of a dairy oil in water

interface. Oral presentation DREAM international

conference Juin 2013 from model food to food models.

Validated for submission to IFSET.

Foucquier Julie. PhD AgroParitech in food science.

December 2012. Modélisation mathématique multi-

échelles de la

prédiciton de la texture d’une émulsion laitière à travers

la description de sa structure.

Foucquier J, Gaucel S, Surel C, Riaublanc A, Baudrit C,

Perrot N. 2011. Modelling the formation of fat droplet

Authors

interface during homogenization in order to describe the

Descamps Etienne1, Foucquier Julie1,

texture. Procedia Food Science 1: 706-712.

Gaucel Sébastien1, Lutton lyne2,

Mackie Alan3, Perrot Nathalie1,

Foucquier, J., Chantoiseau, E., Flick, D., Le Feunteun S.,

Riaublanc Alain4, Trelea Ioan Cristian1

Gaucel S., Perrot, N. (2012) Toward an integrated

modeling of the dairy products transformations, a review

Organisation/Institute

of the existing mathematical models. Food Hydrocolloids,

1INRA, UMR GMPA AgroParisTech INRA, team

27, 1-13.

MALICES, Paris, France

2

Jelinko, D. Master thesis of AgroParisTech. Erasmus

INRIA, Team AVIZ, Paris, France

3

Mundus Master - Food Innovation and Product Design.

Food & Health programme, IFR Norwich,

Comprehension and prediction of acid milk gel texture via

Norwich, England

4

stochastic modeling.

INRA, UMR BIA, Nantes France

Perrot, N., Baudrit, C., Trelea, I.C., Trystram, G.,

Contact

Bourgine, P. (2011). Modelling and analysis of complex

Perrot Nathalie, UMR 782 Génie et Microbiologie des

food systems: state of the art and new trends. Trends in

Procédés Alimentaires GMPA, 78850 Thiverval Grignon

Food Science and Technology, 22(6), 304-314.

nathalie.perrot@grignon.inra.fr

11



Reduction of complexity of an in-silico

milk-gel model, using visualization and

optimization

Dairy products, complex system, in-silico model, reduction of complexity and

visualization

Replicating in-silico the structuring dynamics of food models is a relevant challenge for a better understanding of these systems. It is then important to simplify as much as possible the structure and the number of parameters of in-silico food models: given their intricate structure, however, it becomes hard for experts to thoroughly explore the behaviour of the system and search for meaningful correlations between parameters.

In this work, developed under the WP1, we combine visualisation with model exploration to search for correlations in an established computer model of a milk gel, following the subsequent steps: (a) data are collected during the computations of a learning algorithm, (b) data are made available via a multidimensional visualisation tool, (c) subset selection tools and navigation in the multidimensional parameters space help the expert to evaluate the behaviour of the model. Through this approach, we found a correlation between two parameters of the model, that we were able to support with a formal analysis.

Research aims and background

If the structural characteristics of pure protein ag-

parameters of complex models (Baeck 1993). Clas-

gregates submitted to heat treatment are widely

sical uses of EAs only consider the best individual of

studied (Rabe 2011), research on aggregates of ca-

the last population as an estimation of the optimum,

sein coupled to whey proteins (denatured or not)

but recent works points out the potential benefit of

is still in the initial stages (Morand 2012). Models

visualizing data collected during the execution of an

that are built become more and more complex, and

EA (Lutton 2011), and shows how a multidimension-

necessitate the use of robust and efficient algorith-

al visualisation tool, GraphDice (Bezerianos 2010),

mic techniques. This work is a contribution to the

can help to efficiently navigate inside the data set

design of such complex models: it addresses the

collected during the execution of an EA.

question of parameter learning using robust optimi-

zation techniques and visualisation. The issue here

is to show that the observation of the behaviour of

an optimization algorithm yield important informa-

tions about the optimization problem itself, and as a

consequence on the model under study. The experi-

ments conduced in this work were based on an evo-

lutionary algorithm (EA), a stochastic optimisation

technique that relies on the computer simulation

of natural evolution mechanisms. EAs are specially

well suited for the resolution of difficult optimisation

problems, and particularly for learning the optimal

12



Significance and benefits

Results and applications

Even for skilled scientists, it is often extremely hard

Data has been collected during two emulsification

to validate the behaviour of an in-silico model: due

experiences (respectively used as training and vali-

to complexity an extensive exploration of the search

dation sets), where the continuous phase of the

space is often impossible. The proposed methodol-

emulsion is formed by dissolving milk proteins in

ogy generates a limited amount of data that is likely

permeate. The analysis is based on a model previ-

to be of interest for the user.

ously developed (Foucquier 2011), that predicts the

structure characterised by the percentages of ad-

sorbed caseins and native whey proteins, and the

interfacial concentration. This model depends on 5

unknown parameters, that can be learned from ex-

perimental points (learning set), using an EA that

searches a 5 dimensional space. A visual exploration

(using GraphDice, see figure) of the set of points

visited during the optimisation process shows a con-

vergence toward a rather large area of values for a

couple of parameters (see red points of Figure be-

low), highligthing some evidence about a possible

Figure 1: In-silico model

dependence between these parameters. This evi-

dence has then been confirmed by a mathematical

rewriting of the differential equations of the model,

allowing to consider 4 unknown parameters instead

References

of 5. An optimisation run within this reduced search

M. Rabe, D. Verdes, and S. Seeger. Understanding

space yield a good fitting of the 4 parameters mod-

protein adsorption phenomena at solid surfaces.

Advances in Colloid and Interface Science,

el. These results have been confirmed on the valida-

162(1-2):87–106, 2011.

tion set.

M. Morand, A. Dekkari, F. Guyomarc’h, and MH. Famelart.

Increasing the hydrophobicity of the heat-induced whey

protein complexes improves the acid gelation of skim

Prospects and challenges

milk. International Dairy Journal, 25(2):103–111, AUG

2012

This work can be extended to a general methodolo-

Thomas Baeck and Hans-Paul Schwefel. An overview of

gy for model exploration. Such a technique could be

evolutionary algorithms for parameter optimization.

invaluable to assist experts in assessing the validity

Evolutionary computation, 1(1):1–23, 1993.

and the weak points of their in-silico models, both in

Evelyne Lutton and Jean-Daniel Fekete. Visual

the agri-food and other domains.

analytics of ea data. In Genetic and Evolutionary

Computation Conference, GECCO 2011, 2011. July 12-16,

2011, Dublin, Ireland

Anastasia Bezerianos, Fanny Chevalier, Pierre Dragicevic,

Authors

Niklas Elmqvist, and Jean-Daniel Fekete. Graphdice: A

Gaucel Sébastien1, Lutton Evelyne2, Perrot Nathalie1,

system for exploring multivariate social networks.

Riaublanc Alain3, Tonda Alberto1

Computer Graphics Forum (Proc. EuroVis 2010),

29(3):863–872, 2010.

Organisation/Institute

1

Julie Foucquier, Sebastien Gaucel, Claire Surel, Alain

INRA, UMR GMPA AgroParisTech INRA, team MAL-

Riaublanc, Cedric Baudrit, and Nathalie Perrot. Modelling

ICES, Paris, France)

2

the formation of fat droplet interface during homogenisa-

INRIA, Team AVIZ, Paris, France

3

tion in order to describe the texture. In Saravacos, G and

INRA, UMR BIA, Nantes France

Taoukis, P and Krokida, M and Karathanos, V and Laza-

rides, H and Stoforos, N and Tzia, C and Yanniotis, S, edi-

Contact

tor, volume 1 of Procedia Food Science, pages 706–712,

Nathalie Perrot, UMR 782 Génie et Microbiologie

ELSEVIER SCIENCE BV. 11th International Congress on

des Procédés Alimentaires GMPA, 78850 Thiverval

Engineering and Food (ICEF), Athens, GREECE, MAY 22-

Grignon

nathalie.perrot@grignon.inra.fr

13



Coupling viability theory and active learning

with kd-trees

Viability theory, kd-trees, controlled dynamical system

The mathematical viability theory is very useful when trying to confine controlled dynamical systems into a set of desirable states. It defines states area (the viability kernel) where the system can evolve safely.

Unfortunately the computation of the exact kernel is a complex and computationally intensive task. We propose a method coupling viability algorithm and active learning with kd-tree in order to provide a compact representation of the viability kernel and to limit the call to the model, which, in case of food models, are generally time consuming. Viability theory is a set of mathematical and algorithmic methods proposed for maintaining the evolutions of control ed dynamical systems inside a set of admissible states, cal ed the viability constraint set. This framework has proved in the recent past to be useful for food control in domain where the knowledge of what exactly is a good candidate objective function for optimization is not so clear [Sicard et al, 2012]. Unfortunately the algorithms that are available presently have a complexity which is exponential with the dimension of the state space. This is a severe limitation to the use of the method in real applications. The work we present aims at

providing a compact storage of the set being computed, in order to limit the number of cal s to the dynamical model. It aims also at being reusable.

Results and applications

An algorithm based on kd-tree has been developed in order to learn the boundary of an hyper-volume

[Rouquier et al, submitted], considering the viability kernel as a classification function [Alvarez et al, 2010].

The objective was to limit the number of call of the underlying model. A second algorithm has been developed in order to compute the viability kernel of a viability problem (a dynamical system defined as a black box and a set of desirable states), using the previous kd-tree storage algorithm. This new algorithm has been proved to converge to the true viability kernel [Alvarez et al, 2013].

Research aims and background

Prospects and challenges

This work is a crucial step in order to provide a com-

This method focuses on the boundary of the viability

plete set of tools to compute the viability kernel and

sets rather than on the sets themselves, so it allows

capture basin of viability problems. The code of the

to consider state space with one additional dimen-

algorithm is open source and will be freely available

sion. But it still suffers the curse of dimensionality.

at the ISC-PIF forge. Sustainability problems (such

Further work is in progress in order to parallelize the

as resilience study) and control problems which

refining part of the algorithm.

can be defined as viability problems could then be

Real application of the algorithm will also be



studied with the algorithm we provide.

implemented in order to validate the platform.

14



HIGH

LOW

Figure 1: Level sets of the capture basin of a viability kernel (bilinguism), to define the resilience of the system.

References

I. Alvarez, R. de Aldama, S. Martin and R. Reuillon: “As-

sessing the Resilience of Socio-Ecosystems: Coupling

Viability Theory and Active Learning with kd-Trees”, IJCAI

2013, Proc. of the 23rd International Joint Conference on

Artificial Intelligence, pp. 2776-2782 (2013).

http://ijcai.org/papers13/Papers/IJCAI13-409.pdf

Authors

J.B. Rouquier, I. Alvarez, R. Rreuillon, P-H Wuillemin: “A

De Aldama Ricardo1, Alvarez Isabelle2,

kd-tree algorithm to discover the boundary of a blackbox

Martin Sophie2, Rouquier J.B.1,

hypervolume”. Submitted to Annals of Mathematics and

Reuillon Romain1, P-H. Wuillemin3

Artificial Intelligence.

Organisation/Institute

M. Sicard, N. Perrot, R. Reuillon, S. Mesmoudi, I. Alvarez,

1ISCPIF, complex system institute of Paris,

S. Martin: “A viability approach to control food processes:

Paris, France

Application to a Camembert cheese ripening process”,

2IRSTEA, Team LISC, Clermont Ferrand, France

Food Control, vol. 23 (2), pp. 312-319 (2012).

3LIP6, the Paris computing institute, Paris France

I. Alvarez, S. Martin, S. Mesmoudi: “Describing the Result

of a Classifier to the End-User: Geometric-based Sensi-

Contact

tivity”, European Conference on Artificial Intelligence, vol.

Isabelle Alvarez, Laboratoire d’Ingénierie des Systèmes

215, Frontiers in Artificial Intelligence and Applications,

Complexes (LISC), IRSTEA, France.

pp. 835-840, (IOS Press) (2010)

http://motive.cemagref.fr/people/isabelle.alvarez

15



An integrative modelling of quality and

safety of vegetable during thermal

treatment: application to the cooking of

Broccoli

Mathematical modelling can be very useful in the food industry for control and design of processes. Thermal treatment is a widespread process and literature provides several models, from mechanistic to purely based on data. Unfortunately, they often focus on the process itself or on a single macroscopic indicator of the food product. As a consequence, model coupling is a required step to provide an overview of the food macroscopic characteristics. Process control so implies to find a tradeoff between complexity of the model and accuracy of the prediction, allowing to use the tools of control theory. In this work, we propose a model taking into account food quality, including consumer acceptability, and food safety, applied to the cooking of Broccoli.

The results show a satisfactory prediction, albeit improvable, of the experimental data. Furthermore, the low complexity of the model make it a good candidate for control application.

Research aims and background

Glucosinolates (GSs) are beneficial components

In addition, efficiency of the tools from control theo-

present in Brassica vegetables, which shown an

ry is directly related to the dimension of the model.

ability to reduce the risk of several cancers. The

The issue of this work is to show the feasibility of

concentrations of these compounds are strongly af-

well-predicting the dynamical behaviour of several

fected by the processing of vegetables, especially

macroscopic characteristics in a low complexity

the heat treatment. A model has been developed to

model.

describe the fate of GSs during thermal processing

(Sarvan et al. 2012). However, this model alone is

Results and successful application

not suitable for process control as maximisation of

GSs concentration corresponds to raw cabbage. In-

A model has been developed in order to reproduce

deed, optimisation of the control temperature only

the dynamics of the concentration of a single (or

makes sense for models requiring antagonistic con-

mean) GS in vegetable and cooking water, colour

straints on temperature , e.g. low temperature to

and texture of the vegetable and concentration of

preserve the GSs content and high temperature to

the biological pathogen. The single control variable

ensure food safety. Our aim is to extend the previ-

is the temperature which is assumed to be homo-

ous model to other macroscopic features of vegeta-

geneous in both vegetable and cooking water and

ble during heat processes, related to the consumer

equal to the temperature of the heating device. As

acceptability (colour, texture) and the food safety.

there is no available data on thermal inactivation of

Literature provides simple models for this purpose,

a specific biological pathogen for this study, a “vir-

dealing with texture (Rizvi and Tong 1997) and co-

tual” biological pathogen is considered, allowing to

lour (Tijskens et al. 2001). Inactivation of foodborne

test different thermal resistances. Finally, the en-

pathogens is extensively studied but the recent

zymatic degradation of GSs was neglected so that

works mainly focus on the Weibull frequency distri-

all sub-models (GSs, texture, colour and biological

bution model (Mafart et al. 2002).

pathogen) are independent, the only link being the

16



temperature dependency. The corresponding mathematical system (not shown) is a set of 7 ordinary differential equations with 16 parameters. Parameter estimation has been performed on data from the cooking of Broccoli. The fitting between experimental and simulated results, for some of the state variables, is shown in the Figure 1. Due to the low complexity of the mathematical system, analytical solving has been performed, leading to expression of direct function of time for all the state variables (result not shown).

Figure 1: Fitting between

experimental and simulat-

ed results for some of the

state variables for isto-

thermal heat treatments.

Hatched areas correspond

to targets at the end of the

process.

Prospects and challenges

While the presented model shows an acceptable fitting of the experimental data, it must be extended to a larger range of control temperatures, including kinetics. Further work is in progress to apply viability analysis tools in order to find control temperature kinetics leading to the identified targets.

References

Sarvan, I., Verkerk, R. and Dekker, M. Modelling the fate

of glucosinolates during thermal processing of

Authors

Brassica vegetables. LWT - Food Science and Technology

Dekker Matthijs1, Gaucel Sébastien2, Perrot Nathalie2

49 (2012). 178-183.

Postollec Florence3, Sarvan Irmela1, Verkerk Ruud1

Tijskens, L.M.M., Schijvens, E.P.H.M. and Biekman,

E.S.A. Modelling the change in colour of broccoli and

Organisation/Institute

green beans during blanching. Innovative Food Science &

1WUR, Wageningen University Research, Wageningen,

Emerging Technologies 2 (2001). 303-313

Netherland,

2INRA, MALICES Team, UMR782 AgroParisTech INRA,

Rizvi, A.F. and Tong, C.H. Fractional Conversion for

GMPA Genie et Microbiologie des Procedes Alimentaires,

Determining Texture Degradation Kinetics of Vegetables.

Thiverval Grignon, France.

Journal of Food Science 62-1 (1997).

3ADRIA Developpement, Quimper, France.

Mafart, P., Couvert, O., Gaillard, S. and Leguerinel, I. On

calculating sterility in thermal preservation methods:

Contact

application of the Weibull frequency distribution model.

Sébastien Gaucel, UMR 782 Génie et Microbiologie des

International Journal of Food Microbiology 72 (2002) 107-

Procédés Alimentaires GMPA, 78850 Thiverval Grignon.

113

sebastien.gaucel@grignon.inra.fr

http://www6.versailles-grignon.inra.fr/gmpa

17



Semi-supervised learning of a biscuit

baking model, using symbolic regression and

Bayesian networks

Machine learning methodologies can be an important aid to modeling experts in the food industry, allowing them to obtain reliable models with more efficiency. While human expertise cannot be replaced, it is possible to exploit automated techniques to obtain several candidate models from which an expert could later choose the best, or draw general conclusions on recurring patterns. In this work, developed under the WP1, we perform a feasibility study to learn the model of an industrial biscuit baking process, using two different machine learning paradigms. Symbolic regression is employed to obtain a set of equations, starting from raw data; and an interactive approach is used to learn a Bayesian network model, after discretizing the original dataset. The results show a good prediction capability, albeit further improvements must be studied in order to produce physically meaningful models.

Research aims and background

Machine learning techniques have gained increased

and they can be even manually modified. Several

popularity in recent years, mainly due to the in-

research lines work on the automatic and interactive

creasing complexity of problems faced in industry.

reconstruction of Batesian network starting from

Symbolic regression, an evolutionary technique

data (Tonda 2012) and several libraries have been

based on Genetic Programming (Koza 1992), is able

developed for the purpose (Druzdzel 1999). Our aim

to automatically reconstruct free-form equations

is to perform a feasibility study on the application

from data, uncovering hidden relationships between

of these techniques to the food domain, verifying

variables in a dataset. Commercial software using

whether it’s possible to produce reliable models for

symbolic regression is already available to the public

a specific industrial process.

domain (Schmidt 2009). While symbolic regression

can work on the original data with no modification,

models produced can be very complex and not ad-

Results and successful application

herent to the physical reality of a process, thus be-

Data has been collected during 16 runs of an in-

ing hard to understand for a human expert. Bayes-

dustrial baking process by the company United Bis-

ian networks are graphical probabilistic models that

cuits: 12 runs are used for training, 4 for valida-

work with discretized variables. They represent a set

tion. The variables measured are top and bottom

of variables and their conditional dependencies via a

flux of heat in the oven, color, height, and weight

directed acyclic graph, and are widely used to rep-

loss of the biscuits. Additional features include the

resent knowledge in many different domains, rang-

temperature in each zone of the oven, which might

ing from computational biology to decision support

vary between different experiences. Both machine

systems. While the discretization of variables might

learning techniques are trying to find models that

introduce further sources of error, Bayesian network

predict the considered variable at instant t+1, hav-

representations are intuitive for the end user; they

ing access only to observable values (such as top

can be validated by experts of a specific process

flux, bottom flux, heat in the different zones of the

with little to no knowledge of their inner working;

oven at time t, and initial value of the considered

18





variable at time t=0). Symbolic regression is able to find several good equations to predict color, height and weight loss, performing well also on the validation set.

The Bayesian network obtained through an interactive learning approach shows several relationships between variables that also appear among the best models produced by symbolic regression, thus supporting the findings of the previous step.

Significance and benefits

Prospects and challenges

Modeling experts usually need a long phase of tri-

While the obtained models show a good perfor-

al and error before finding a satisfying model of a

mance even on unseen data, the machine learning

complex process, such as is often the case in the

algorithms completely ignore the physical meaning

food industry. Semi-supervised learning techniques

of the models themselves. Further studies on semi-

can generate several candidate solutions with good

supervised approaches are needed, in order to get

fitting in a small amount of time, from which the

high-quality results coherent with the physical phe-

expert can then choose the most promising or

nomena underlying the processes.

physically sound. An added benefit is the possibil-

ity of examining the candidate models for recurring

patterns, which might unveil some unknown



relationships between variables of the process.

References

Druzdzel, Marek J. “SMILE: Structural Modeling,

Inference, and Learning Engine and GeNIe: a develop-

ment environment for graphical decision-theoretic

models.” AAAI/IAAI. 1999.

Koza, John R. Genetic Programming: vol. 1, On the

programming of computers by means of natural

selection. Vol. 1. MIT press, 1992.

Schmidt, Michael, and Hod Lipson. “Distilling free-form

natural laws from experimental data.” Science 324.5923

Authors

(2009): 81-85.

Gaucel Sébastien1, Lutton Evelyne 1,

Perrot Nathalie1, Tonda Alberto 1,

Tonda, Alberto, et al. “Bayesian network structure

learning from limited datasets through graph evolution.”

Organisation/Institute

Genetic Programming. Springer Berlin Heidelberg, 2012.

1INRA-AgroParisTech, Team GMPA,

254-265.

Thiverval-Grignon, France

Contact

Tonda Alberto

INRA Versailles-Grignon

alberto.tonda@grignon.inra.fr

19



Filled cellular solid

model

egetables egetables ed changes

It this information important for me?

Consumers

Food Industry

Regulatory Authorites

Scientific Community

WP-2

Workpackage



thermal impact on fruits & v

mechanical impact on fruits & v

mathematical model describing observ

20

Task 2.1

Selection of three plant food models taking into

account works done in FP6 projects, especially





Introduction of WP2


EU-SOL, ISAFRUIT, FLORA and FLAVO: (INRA,

WUR) In this task the specific varieties and cultivation

conditions to obtain the biological material on brassica

vegetables, apple and tomato for the further studies

have been defined.

Task 2.2

Processing of the three plant food models to

Matthijs Dekker, Workpackage Leader

represent mechanical and thermal treatments of

Wageningen University, Food Quality & Design Group, WUR

fruits and vegetables by industry: (INRA, WUR)

matthijs.dekker@wur.nl

For brassica vegetables the effects of thermal

processing on the content of glucosinolates, texture

and colour have been investigated. For tomato the

effect of hot/cold break on the content and accessibil-

Aims and background

ity of lycopene has been investigated. For apple the

texture and the characteristics of polyphenols in cell

To develop well-characterised, realistic, food models

wall polysaccharides were studied.

for plant foods able to serve as tools for integrating

Task 2.3

and harmonising food and nutrition research on plant

Characterisation of the macro- and microstruc-

food products. Fruit and vegetables are physiologi-

ture in the fruit and v T

e ip

ge ! table GMFs before and

cal active products that, even for the same variety,

after processing in terms of different kinetic,

mechanical, chemical, and enzyme events

show variation in many characteristics depending

leading to changes in the bioavailability of

on the cultivation conditions (location, soil, weath-

selected phytonutrients: (IFR) The changes in cell

er, light,...) and post-harvest conditions. Therefore

wall structures and composition during processing

even with standardised protocols variation will re-

have been investigated by imaging techniques.

main. Therefore there is a need to link processing

Task 2.4

behaviour with measurable characteristics that are

Mathematical modelling (in connection with

determining kinetics of nutritional changes. Mecha-

WP1) to describe the main mechanisms of chang-

nistic mathematical models are needed that are ro-

es in product structure, phytonutrient (content

and availability) and sensory properties during

bust to deal with the natural variation and still give

processing: (WUR) A mathematical model has been

meaningful results to be used for product and pro-

developed that describes the content of glucosinolates

cess optimisation with respect to nutritional, sen-

in brassica vegetables based on the various

sory and safety quality attributes.

mechanism that occur during thermal processing: cell

lysis, leaching, enzymatic degradation, enzyme

inactivation and thermal degradation in different

compartments. Kinetic texture and colour models have

been developed.

Main outcomes

Selected varieties of brassica vegetables, toma-

toes and apples have been studied for their chemi-

Prospects

cal, physical and nutritional properties. For brassica

vegetables the main focus has been on developing

The developed approaches to study fruit and veg-

a mathematical model for thermal processes that

etables, as illustrated on the selected varieties and

can describe the content of phytochemicals (gluco-

quality attributes in this workpackage, can serve

sinolates) in the products as a function of process-

as guidelines to study nutritional and sensory re-

ing conditions. In addition kinetic models have been

lated product properties in fruits and vegetables in

developed for the texture and colour of the prod-

a much broader perspective. With limited additional

ucts. For tomato the focus has been on relating the

experimental efforts, the parameters of the devel-

processing conditions with the content and acces-

oped thermal process model can estimated. With

sibility of lycopene. For apples the focus was on the

these specific parameters the model can be applied

distribution and absorption of polyphenols in the cell

to simulate and optimise the content of many phyto-

walls and on texture development.

chemicals within fruits and vegetable varieties.

21



Using mathematical modelling to optimise

processing of fruits and vegetables with

respect to nutritional and sensory quality

Mathematical modelling for improved food quality

The phytochemical content of processed fruit and vegetable products is highly variable and unpredictable.

By studying the underlying mechanisms of changes in the contents during processing, mathematical models have been developed to simulate and optimise the thermal processing conditions with respect to the phytochemical content in the final product. Combining these models with models describing texture and colour changes depending on the conditions allows multi-criteria optimisation of fruit and vegetable overall quality.

The modelling approach was illustrated with brassica vegetables and the phytochemicals glucosinolates. This case study can serve as a blueprint for the application of mathematical modelling to enhance the nutritional properties of plant foods, while respecting the sensory quality, in a much broader sense.

Research aims and background

Fruit and vegetables are an important part of our di-

ity to a wide range of phytochemicals in fruits and

ets. The intake of fruit and vegetables is associated

vegetables can be used by the food industry to

with reduced risk for many diseases like cancers,

efficiently improve the quality of plant foods with

cardio vascular diseases, diabetes type 2, etc. Phy-

respect to their health promoting and sensory



tochemicals in fruit and vegetables show important

qualities.

biological activities related to their health promot-

ing effects. The content of these phytochemicals in

current fruit and vegetable products is highly vari-

Results and applications

able. Next to breeding and cultivation, processing

As a case study the thermal processing of bras-

and preparation have been shown as main sources

sica vegetables (broccoli, cabbages and Brussels

of this observed variation.

sprouts) was studied in detail. By investigating the

Understanding the changes in the phytochemical

mechanisms that are responsible for the changes

content during processing and preparation allows

in the phytochemical (glucosinolates in the case of

the development of mathematical models to de-

brassica vegetables) composition during thermal

scribe the effects of processing conditions on the

processing models were developed to describe:

final level.

• Cell lysis kinetics

With these models product and process optimisation

• Glucosinolate and enzyme leaching kinetics

can be done in order to enhance the phytochemical

composition of the final product, while in the mean-

• Enzymatic conversion kinetics

time the sensory quality attributes of the products

• Enzyme inactivation kinetic

are respected.

• Thermal degradation kinetics of glucosinolates in

The results of this research in terms of developed

vegetable tissue

models and the approach to broaden the applicabil-

• Thermal degradation kinetics of glucosinolates in

22





processing water.

Significance and benefits

By specific experimental set-ups the parameters in

With the developed models and experimental ap-

these models can be efficiently estimated. In addi-

proaches to estimate model parameters it is now

tion to the phytochemical model, also semi-mech-

possible to efficiently describe and optimise the

anistic models te describe the kinetics of texture

phytochemical content of plant food products, while

and colour changes during processing have been

respecting their sensory quality. The results can

developed and their parameters have been esti-

be used by the food industry to efficiently improve

mated. With the developed set of models process

products, but also by nutritionists that want to have

conditions can be optimised with respect to product

an overview of the phytochemical contents of plant

quality (health promoting effects as well as sensory

foods after processing and/or preparation. Most nu-

quality). The developed approach was illustrated on

tritional data bases only mention the contents in

a specific group of phytochemicals in specific veg-

raw materials and perhaps one standard processed

etables. However the same approach and often even

or prepared product. With the developed model the

the same models can be used on a much broader

effects of many process and preparation conditions

range of plant foods and their specific phytochemi-

can be estimated and used to improve the quality of

cals. What is needed to apply this is to estimate the

intake data in e.g. epidemiological studies.

specific parameters in the models for each case. The

experimental procedures developed in this project

can serve as a guideline for that.

Figure 1: schematic representation of mechanisms that effect the content of phytochemicals during thermal processing of vegetable tissue. (green triangles: phytochemicals, red squares: active/inactive enzymes, beige circles: thermal breakdown products, green squares: enzymatic breakdown products).

23





Figure 2: simulation results of the effect of a sterilisation process on the content of phytochemicals (GS) and breakdown products (BDP).

Prospects and challenges

References

I. Sarvan, R. Verkerk, M. Dekker, (2012) Modeling the

The developed approaches to study fruit and veg-

fate of glucosinolates during thermal processing of Bras-

etables, as illustrated on the selected varieties and

sica vegetables. LWT: Food Science and Technology, 49,

quality attributes in this workpackage, can serve

178-183.

as guidelines to study nutritional and sensory re-

I. Sarvan, F. Valerio, S. L. Lonigro, S. de Candia, R.

lated product properties in fruits and vegetables in

Verkerk, M. Dekker, P.Lavermicocca (2013) Glucosino-

a much broader perspective.

late content of blanched cabbage (Brassica oleracea var.

capitata) fermented by the probiotic strain Lactobacillus

With limited additional experimental efforts, the



paracasei LMG-P22043. Food Research International,

parameters of the developed thermal process model

available online.

can estimated. With these specific parameters the

I. Sarvan, R. Verkerk, MAJS van Boekel, M. Dekker

model can be applied to simulate and optimise the

(2013) Comparison of the degradation and leaching

content of many phytochemicals within fruits and

kinetics of glucosinolates during processing of four Bras-

sicaceae (broccoli, red cabbage, white cabbage, Brussels

vegetable varieties.

sprouts). submitted.

M. Dekker, E. Dekkers, A. Jasper, C. Baár and R. Verkerk

(2013) Predictive modelling of vegetable firmness after

thermal pre-treatments and steaming. submitted

Authors

Matthijs Dekker, Irmela Sarvan, Ruud Verkerk

Organisation/Institute

Wageningen University, Food Quality & Design Group,

Wageningen, The Netherlands

Contact

Matthijs Dekker, Wageningen University,

Food Quality & Design Group

matthijs.dekker@wur.nl

24



Optimisation of mathematical model describ-

ing transfer of polyphenol from fruit to juice

by understanding non-covalent interaction

between procyanidins and pectins

Non-covalent interaction between procyanidins and pectins

Procyanidins are the main phenolics in cocoa and many Rosaceae fruit, and have a major role in their bitter-ness and astringency, as well as potential health benefits. Disruption of the natural matrix during processing and interaction between procyanidins, and cell walls may have a strong influence on the release, on the bioavailability, and on the biological activity of procyanidins. A mathematical model has been developed, which is already applicable to predict retention of procyanidins by cell walls, e.g. in juice extraction processes. However, it needs to be improved by adding the influence of pectin characteristics that are degree of methylation and neutral side chain composition, in order to be used by the food industry to enhance food quality.

Research aims and background

Polyphenolic compounds, including procyanidins,

affinity for procyanidins. However, the mechanism

are commonly perceived to be found mainly in the

by which procyanidins and pectins interact and the

vacuoles of plants where they are separated from

structural and compositional parameters that influ-

other cellular components. However, many may also

ence their association are not known.

be associated with cellular components, such as the

Understanding the influence of structural and com-

cell wall, especially after cell injury when vacuoles

positional parameters during procyanidin-pectin in-

rupture during processing. This results in the re-

teractions may allow optimising the initial model to

lease of phenolic compounds which may then asso-

better describe the effect of mechanical processing

ciate with cell wall polysaccharides through hydro-

on juice procyanidin concentration.

gen bonding and hydrophobic interactions. These

The model developed may be used by the food in-

interactions have a strong influence in the release

dustry to identify influenced parameters and to sim-

but also in the bioavailability, and in the biological

ulate them in order to optimise their process and to

activity of procyanidins.

enhance food quality.

The cell wall capacity to bind procyanidins depends

upon compositional and structural parameters, such

as contents and structure of the various cell wall

polymers, stereochemistry, conformational flexibili-

ty and molecular weight of procyanidins, and cell wall

and procyanidin concentrations. It also depends on

surrounding conditions, such as temperature, ionic

strength or ethanol content. A mathematical mod-

el has been developed to describe the transfer of

procyanidin from fruit to juice. Among the different

polysaccharides classes (cellulose, hemicellulose,

pectins), pectins are those that have the greatest

25





Results and applications

The interactions between procyanidins and pectins were studied in details by varying both the composition of procyanidins and pectins. Experiments confronted B-type procyanidin from apple with various degrees of polymerization to the different pectin substructures, such as homogalacturonan presenting different degree of methylation, rhamnoglacturonan I with different neutral sugar side chains and rhamnoglacturonan II as monomer or dimer.

The affinity constants of procyanidins-homogalactu-

ronans interactions in solution are the highest when

both the procyanidins degree of polymerization and

the homogalacturonan degree of polymerization are

the highest. Procyanidins interacted with high meth-

ylated homogalacturonan mainly through hydropho-

bic interactions.

Associations between rhamnogalacturonan I frac-

tions and procyanidins involved hydrophobic inter-

actions and hydrogen bonds. No difference in asso-

Figure 1: Neutral sugar side chains of pectins limit interac-

ciation constants between rhamnoglacturonan I with

tions with procyanidins

different neutral sugar side chains and procyanidins

of degree of polymerization of 9 was found. Nevertheless, rhamnogalacturonan I rich in long arabinan chains showed lower association constants, and rhamnogalacturonan I without neutral sugar side chains showed higher association with procyanidin of degree of polymerization of 30. Only very low affinities were obtained with rhamnogalacturonans II. It seems that the ramification state of rhamnogalacturonan I limits their association with procyanidins. This might explain some of the variation in transfer of procyanidins from fruit to juice with maturation, as one of maturation features is loss in galactose and/or arabinose from the cell walls.

The influence of degree of methylation of pectins could now be integrated in the mathematical model in order to optimise it and could be used by the food industry to optimise the process to enhance food quality.

Significance and benefits

Interactions between procyanidins and pectins have

been studied in details and could be now integrated

in the mathematical model describing the transfer

of procyandins from fruit to juice after mechanical

treatment. The result obtained could be used to the

food industry to improve food quality.

Moreover, the process of fining for the removal of

procyanidins for clarification and astringency reduc-

tion in wines traditionally uses protein extracts in

its application. The wine manufacturer could use the

result obtain here to optimise their fining treatment

by using fibers as an alternative to fining with pro-

teins in winemaking.

26



Successful applications

The impact of temperature and duration of pressing

that were identified in the initial model have been

validated by the Institut Français des Productions

Cidricoles for use in cider apple pressing.

Prospects and challenges

The two main research prospects at this point are:

- Application to understanding the impact of matu-

ration on procyanidin extraction, with a thesis start-

ing on that topic (applied to pears due to their simple

phenolic composition and existence of a large range

of degrees of polymerization);

- Validation of the impact of the interactions on the

colonic fermentation patterns for cell wall – procy-

anidin complexes.

References

Le Bourvellec C., Le Quéré J-M., Renard C.M.G.C. (2007).

Impact of non-covalent interactions between condensed

tannin and cell walls on their transfer from fruit to juice:

studies in model suspensions and application. J. Agric.

Authors

Food Chem., 55, 7896-7904.

Aude A. Watrelot, Carine Le Bourvellec, Catherine

Watrelot A.A., Le Bourvellec C., Imberty A., Renard

M.G.C. Renard.

C.M.G.C. 2013. Interactions between pectic compounds

and procyanidins are influenced by methylation degree

Organisation/Institute

and chain length. Biomacromolecules, 14, 709-718.

INRA, UMR408 Security and Quality of Plant Products,

Avignon, France

Watrelot A.A., Le Bourvellec C., Imberty A., Renard



C.M.G.C. (2014). Neutral sugars side chains of pectins

Contact

limit interactions with procyanidins. Carbohydrate poly-

Catherine M.G.C. Renard:

mers, 99, 527-536.

INRA, UMR408 Security and Quality of Plant Products

Catherine.renard@avignon.inra.fr

27



Tomato processing method modify the

bioaccessibility of its lycopene

Modulation of tomato carotenoid accessibility through appropriate processing

Tomato product is the main source of lycopene of our diet. Its accessibility, i.e. its liberation from the food matrix, and its bioavailability, i.e. the subsequent transfer to its cellular targets at the end of the digestive process, determine its true health benefit against several cancer or degenerative diseases. The accessibility is greatly influenced by the physical properties of the food matrix and especially, it is enhanced in cooked products. Hot break (HB) and Cold Break (CB) treatments used by industries to control the viscosity of tomatoes purees was shown to partially control the ability of the tomato lycopene to diffuse from puree to an oil phase. Food particle size and lycopene/matrix interactions were identified as the main factors affecting the diffusivity.

Research aims and background

Modelling the availability lycopene in response to processing was the objectives for the tomato model in the DREAM project. Lycopene has been already identified as a health benefit product. The initial step of the digestive process for such lipophilic micronutrients consists in diffusing from the plant matrix to reach the lipid phase of the emulsion of the bolus. For lycopene, the major tomato carotenoid, bioavailability is known to be enhanced in cooked products. Indeed, if most of industrial tomatoes processes do not much modify the overall lycopene content, its bioaccessibility can be greatly affected. The aims of the research were therefor to identify the factors limiting the lycopene diffusion and how these factors are modified by the processes applied to fruits. Controlling the carotenoid bioaccessibility by the process is a challenge for the fruit and vegetable industries in order to boost the nutritional value of their product.

Results and applications

The starting hypothesis to set up the model linked

break (HB) processing, which exhibited contrasted

particle sizes to the ability of lycopene to diffuse to

lycopene diffusion rate when they were mixed to

an oil phase. Tomato is a filled-cellular model, and

oil in a standardized protocol. Our first results indi-

then, 1- the smallest the particles generated by the

cated that no clear difference of particle sizes could

process (grinding, cooking temperature), the quick-

explain the contrast that we observed (Page et al.

est the diffusion of one nutrient, according to the

2012). Wet sieving experiments indicated that most

second Fick law’s and 2- depending of the numbers

of the HB/CB difference was explained by the be-

of intact barriers (i.e. intact cell-wall or membranes)

haviour of the smallest particles (Figure 1). We also

that may remind in the matrix between the lyco-

studied more into details, using one standard com-

pene and the intestinal membranes. The case of

mercial tomato juice, the other physicochemical pa-

study chosen for the project was two tomato pu-

rameters affecting the lycopene diffusion, including

rees obtained either using cold-break (CB) or hot-

pH effect, temperature and how much the diffusion

28





including pH effect, temperature and how much the diffusion is modified when tomato purees where mixed to an emulsion instead of pure oil. For this last, the rate of the diffusion rose, but the partition factor of the lycopene between puree and oil was not much affected. Even more, it was reduced by some interactions with the emulsifier (Degrou et al. 2013). These last results indicated that molecular interactions between lycopene and other components of the matrix may affect the availability of the lycopene, making the model more complex than expected, and making necessary a set of new experiments to recover variables that should be implement into the tomato model.

Figure 1: Threshold effect on particle

size. Most of the available lycopene

is included into the smallest class of

particle sizes (limit at 125µm, i.e.

roughly the average fruit cell size).

Prospects and challenges

At the end of the DREAM project, the main factors affecting the lycopene diffusion from the matrix have been identified. Modelling needs now to be achieved in order to set up a mathematical model that could predict the availability of the carotenoid from the processing parameters. And, moreover, the identification of the smallest particles as the main factor of the contrast between HB and CB puree make it a further research challenge to understand the biochemical change of the fruit tissues that lead to lycopene release. Structural description of these small particles is a challenge.

Significance and benefits

References

Degrou, A., S. Georgé, et al. (2013). “Physicochemical

From the Dream results, it is now clear that carot-

parameters that influence carotenoids bioaccessibility

from a tomato juice.” Food Chemistry 136(2): 435-441.

enoid bioaccessibility can be deeply modified by the

processing methods, and particularly if the very first

Page, D., E. Van Stratum, et al. (2012). “Kinetics of tem-

perature increase during tomato processing modulate

temperature ramping (i.e. during the first minutes

the bioaccessibility of lycopene.” Food Chemistry 135(4):

of heating, which correspond to the main difference

2462-2469.

between HB and CB purees) is efficiently applied,

leading to a quick rise of the initial temperature. In

this case a significant quantity of carotenoid is made

bioaccessible.

Authors

David Page1, Stephane George2, Catherine Renard3

Organisation/Institute

1, 3 Quality and safety of plant food, INRA, Avignon,

France

2Quality and processes, CTCPA, Avignon, France

Contact

David PAGE:

Quality and safety of plant food, INRA, Avignon, France

E-mail address:David.Page@avignon.inra.fr

29



Proteinous cellular

l

network model

odel tic ge issues

el m ime eat t

etic g n m f m

im t o s o

It this information important for me?

Consumers

pac

Food Industry

Regulatory Authorites

nalyse

Scientific Community

artificial m ath imhe age am

WP-3

opic i

Workpackage



icroscm

30

Task 3.1

Rationalisation of meat tissue selection.

Three meat categories were characterized in composi-

tion and structure and two distinct muscles were se-

lected (Longissimus thoracis and Masseter). Main raw

meat parameters associated with the nutritional target





Introduction of WP3


property (in vitro protein digestibility) were identified.

Guidelines to select categories of raw meats based on

in vitro digestion were developed.

Task 3.2

Prospective investigation to build artificial mimetic gel

Carolina Realini, Workpackage Leader

models representative of animal tissues.

Food Technology, IRTA

Meat tissue structure couldn’t be represented using

carolina.realini@irta.es

gel models. Thus, kinetic laws which govern changes

in proteins’ physicochemical state induced by heating

were measured using an in vitro model: myofibrillar

proteins in a buffer with the various compounds that

Aims and background

can be found in different types of meat: Fe, ascorbate,

oxidants and anti-oxidants. Heat treatments were ap-

Heating is one of the most important processes ap-

plied at 45°C, 60°C, 75°C and 90°C up to 120 min.

plied to animal tissues, since meat is usually con-

Task 3.3

sumed after cooking at industrial or domestic scale.

Microscopic image analyses of typical different meat

This physical process induces structural and chemi-

tissues to assess th

cal changes that can affect meat nutritional value.

T

ei ip

r s!tructural characteristics.

Methods and algorithms to semi-quantify in situ the

The work developed in WP3 was focused on meat

morphology and distribution of cell types in muscle

tissue from histological images were developed and

and the impact of heating on its protein physico-

applied to the meat categories. It was shown from

chemical changes and digestibility. Homogeneous

statistical analysis that connective tissue and fibre

muscle models are needed to evaluate the effect of

type distributions vary considerably with muscles and

different processing parameters on the nutritional

/ or collection areas. The overall structure was also

assessed from MRI images using a Diffusion Tensor

quality of meat. One objective was to define the best

Imaging technique.

experimental models representative of meat prod-

ucts by selecting and assessing meat tissues or by

Task 3.4

creating artificial mimetic samples. Another objec-

Use of artificial gel models to understand reaction

routes promoted by heating of animal tissues.

tive was to evaluate the effect of compositional and

While thermal denaturation can be considered as a

structural properties of proteinous foods on the re-

unique first order reaction oxidation is more com-

actions promoted by heating.

plex since many interactive chemical reactions are

involved. A mathematical model which accounts for

this complexity was successfully developed to analyze

Prospects

and predict the effect of (i) iron content which differs

depending on the meat type, and of (ii) the heating

Two models were developed: three meat catego-

conditions through the time-temperature couples.

ries (pork muscles) that differ in their structure and

composition, and a mimetic model (suspension of

myofibrillar proteins). The pork muscles allowed to

Main outcomes

characterize their structural variability by imag-

ing and histology, to set up a non-invasive noble

The physical and mathematical models developed in WP3

method (MRI), and to identify heat induced protein

proved to be useful tools to evaluate the impact of heat-

changes linked with in vitro digestion. The mimetic

ing on in vitro protein digestion. These models could be

model allowed to evaluate the effect of iron con-

used to evaluate the impact of other conditions or pro-

tent and temperature on the kinetics of myofibrillar

cesses applied to proteinous foods on nutritional, sensori-

protein oxidation and denaturation. A mathematical

al or technological target properties of scientific or indus-

‘stoichio-kinetic’ model was developed to investigate

trial relevance. Further research should focus on sensory

from calculations the main chemical reaction routes

properties using the defined meat categories and the

involved in protein changes.

mimetic and mathematical models will be progressively

completed to include all the possible compounds in meat.

31



Meat models and mathematical modelling

to investigate how heating affects

protein digestibility

Models to better analyse and control meat cooking

Meat cooking is an important process applied at industrial or domestic scale that has nutritional impact.

Two experimental models were developed to investigate how meat characteristics and heating conditions, respectively, affect the nutritional quality of meat: (1) meat categories that differ in their structure, composition and metabolic type, and (2) a mimetic model composed by suspension of myofibrillar proteins which composition can be modified by adjusting the concentrations of major meat compounds: iron, oxidant and antioxidant enzymes. In parallel, a mathematical model called ‘stoichio-kinetic’ was developed to integrate knowledge. The model is composed of differential equations that represent all elementary reactions.

Model predictions agree with the experimental measurements. This approach can be applied to other meat processes and sensorial properties.

Research aims and background

Apart from a few exceptions, animal tissues (meat

opment of experimental and mathematical models.

and fish) are eaten after being cooked. Heating is

We worked in two main directions:

therefore the most important process applied to this

• To define the best way to prepare samples repre-

type of foods either at industrial or domestic scale.

sentative of meat products either by selecting and

This physical process induces structural changes, at

assessing meat tissues or by creating artificial mi-

microscopic and macroscopic levels, and promotes

metic samples;

protein changes that can have nutritional impacts.

• To evaluate, using the above meat models and

A great number of works had been carried out on

mathematical modelling, the effect of meat tissue

meat and fish products to assess the impact of pro-

characteristics and cooking conditions on the reac-

duction parameters (genetic, breeding, feeding,

tions which are promoted by heating and can have

slaughter conditions, processing conditions…) on

nutritional consequences.

meat product qualities. This is usually done by com-

The tools that were set up during the DREAM project

paring samples that undergo a specific treatment

can help engineers to develop precooked or cooked

to a reference sample. However, in technological

industrial products with better nutritional proper-

or laboratory tests it is rather difficult to well con-

ties. They can be also used by scientists to investi-

trol and measure all the parameters that can have

gate a larger range of processing conditions: other

an impact on a specific output. This often prevents

meat tissue characteristics, other processes.

to generalize conclusions of a particular study and

leads to contradictions between studies because

of the interactions between sample properties and

phenomena involved.

With the aim to improve the generalization of labo-

ratory results, our work was focussed on the devel-

32





Results and applications

Parallel works have been carried out to develop two

complementary experimental models: (1) extreme

examples of meat tissues representing three “meat

categories” that differ in their structure, composi-

tion and metabolic type (Realini et al., 2013 a & b),

and (2) a mimetic model composed by a suspen-

sion of myofibrillar proteins (Promeyrat et al., 2013

a &b).

Changes in protein state and nutritional value were

related by applying two extreme time-temperature

heating couples (10 min, 75°C or 45 min, 90°C) to the

two extreme meat categories. The nutritional value

was evaluated from in vitro digestion tests using ei-

ther gastric pepsin or trypsin and α-chymotrypsin.

Figure 1: Two meat categories were cooked either “10

It is clear from Fig. 1 that the variables associat-

min, 75°C” or “45 min, 90°C”. A Principal Component

ed with protein changes in conformation and those

Analysis showed that protein denaturation (hydropho-

bicity) and oxidation were correlated to digestion rates

linked with oxidation are positively and negatively

measured in vitro.

correlated, respectively, with rates of in vitro pro-

tein digestion. The mimetic model was developed

to avoid the confusing effect due to uncontrolled

biological variability generally observed in animal

tissue. Its composition can be modified by adjust-

ing independently the concentrations of the major

chemical compounds in meat (Fig. 2): iron, oxidant

and antioxidant enzymes. This experimental model

allows to easily determining kinetic laws that govern

changes in protein state and to assess how much

the kinetic parameters are affected by initial prod-

uct characteristics and cooking conditions. Thermal

denaturation can be modeled by a unique first order

Figure 2: Kinetic laws of protein denaturation and oxida-

reaction. Oxidation is more complex (many interac-

tion were determined using a mimetic model: aqueous

suspension of pure myofibrillar proteins, control of both

tive chemical reactions that are differently affect-

physic-chemical environment (pH, ionic strength) and

ed by pH and temperature). Thus, a mathematical

compound contents to mimic many kinds of meat.

stoichio-kinetic model which accounts for this com-

Significance and benefits

plexity was successfully developed (Promeyrat et

al., 2012). Simulation calculations allow to analyze

Our collaborative work provides a sound basis to

and predict the effect of (i) iron content which dif-

build a mathematical tool, or simulator, that can pre-

fer from one meat to another and of (ii) the heating

dict the effects of various processes such as chilled

conditions through the time-temperature couples

storage, modified atmosphere conditioning, curing,

(Fig. 3). The predictions are accurate enough to be

heat treatments, etc on the sensorial and nutritional

used for practical purpose.

qualities of processed meat products.

This approach needs that new scientific knowledge

is progressively added to improve our simulator by

assessing parameters associated with each individ-

ual phenomenon or chemical reaction. The strategy

based on experiments with both realistic and mi-

metic models is certainly the best way to achieve

this goal.

33



References

Promeyrat, A., Gatellier, P., Broyart, B., Kondjoyan,

A., Daudin, J. D. 2012, Kinetic Modelling of Myofibrillar

Protein Denaturation and Oxidation during Heating, In

EFFoST Annual Meeting - A Lunch Box for Tomorrow : An

interactive combination of integrated analysis and spe-

cialized knowledge of food. November, Montpellier, France

Realini, C. E., Venien, A., Gou, P., Gatellier, P., Perez-Juan,

M., Danon, J., and Astruc, T. 2013 a, Characterization

of Longissimus thoracis, Semitendinosus and Masseter

muscles and relationships with technological quality in

pigs. 1. Microscopic analysis of muscles, Meat science 94,

408-416.

Realini, C. E., Perez-Juan, M., Gou, P., Diaz, I., Sarraga,

C., Gatellier, P., Garcia-Regueiro, J. A. 2013 b, Charac-

terization of Longissimus thoracis, Semitendinosus and

Masseter muscles and relationships with technological

quality in pigs. 2. Composition of muscles, Meat Sci. 94,

417-423.

Promeyrat, A., Daudin, J. D., Gatellier, P. 2013 a, Kinetics

of protein physicochemical changes induced by heating

in meat using mimetic models: (1) relative effects of heat

and oxidants, Food Chem. 138, 581-589.

Promeyrat, A., Daudin, J. D., Astruc, T., Danon, J.,

Gatellier, P. 2013 b, Kinetics of protein physicochemical

changes induced by heating in meat using mimetic mod-

els: (2) Effects of fibre type, peroxides and antioxidants,

Figure 3: A mathematical model based on differential

Food Chem. 138, 2283-2290.

equations representing numbers of interactive chemi-

cal reactions can predict kinetics of protein changes; for

example, carbonyl production according to Fe content and

heating conditions.

Prospects and challenges

We took a big step forward in predicting the impact of

practical cooking conditions on protein denaturation

and oxidation through the use of both experimental

and mathematical meat models. These phenomena

affect protein digestibility but also various other

technical and sensorial qualities. For example, the

Authors

kinetic of protein denaturation is certainly linked to

Astruc Thierry1, Danon Jeanne1,

cooking losses and oxidation of myoglobin is known

Daudin Jean-Dominique1, García-Regueiro José

to determine colour. Thus, our strategy could be ap-

Antonio2, Gatellier Philippe1, Gobert Mylène1,

Gou Pere2, Kondjoyan Alain1, Pérez-Juan María2,

plied to many other targets. Using a mathematical

Promeyrat Aurelie1, Realini Carolina2,

model is essential to extrapolate laboratory results

Santé-Lhoutellier Véronique1, Vénien Annie1

to industrial process conditions. For example, pro-

tein denaturation and oxidation can now be predict-

Organisation/Institute

1UR370 Qualité des Produits Animaux, INRA, F-63122

ed for time variable cooking conditions, since all the

Saint Genès Champanelle

reactions are represented by differential equations.

2IRTA, Finca Camps i Armet, E-17121 Monells,

Moreover, these calculations can be easily inserted

Girona, Spain

in other mathematical models that predict the time

Contact

evolution of the temperature distribution within

Daudin Jean-Dominique, UR370 Qualité des Produits

meat pieces during cooking (Heat transfer math-

Animaux, INRA

ematical model based on Finite Element technique).

jean-dominique.daudin@clermont.inra.fr

34



Microstructure characterization of muscle

tissue by quantitative imaging

in situ high resolution imaging of meat tissue microstructure by MRI using

diffusion tensor imaging (DTI)

Meat results from the post mortem transformation of animals skeletal muscles. These muscles are



mainly composed of water (75%), muscle cells of 3 different types (I, IIA and IIB), connective tissue and fat (3 to 5%). The muscle is strongly organized in fiber bundles more or less aligned depending on its type and functionality. The muscle structural architecture can be roughly modeled as a square lattice array of fluid filled cylinders surrounded by fluids. With the MRI technique used here, we look at diffusion: the physical parameter related to the water local motion. Diffusion in tissues differs from diffusion in free solutions because compartments made by fibers hinder and restrict water motion. The structural meat architecture is hereby assessed, not directly and destructively as done by tomography microimaging (microscopy) but indirectly and non-destructively.

Research aims and background

Even if diffusion quantitative magnetic resonance imaging is a generic method to assess Brownian random motion of molecules, it has been mostly applied to water which is abundant in tissues and visible by MRI.

Both muscular cells size and muscular cells shape as well as interactions within the different compartments of muscular cell can influence the diffusion properties. If the structure is anisotropic, as in muscle, water diffusivity also displays anisotropic behavior. As muscle, and then meat, is highly organized, the diffusion of water is facilitated in the fiber direction, meaning that, in a group of fibers having the same direction, the apparent diffusion coefficient is: maximum in the fiber direction, minimum orthogonally to the fiber direction.

We decided to make use of this anisotropy to indirectly and non-destructively characterize meat microstructure. The anisotropic water diffusion can be modeled voxelwise using ellipsoid by means of a first order tensor, according to Diffusion Tensor Imaging (DTI). Moreover, by varying the intensity of the magnetic gradient field, one can obtain the diffusion decay. This decay is then fitted by a bi-exponential curve, at each voxel, in order to obtain quantitative mapping at low and high gradient field values, exhibiting hindered and restricted diffusion, respectively. One first objective was to obtain high resolution imaging of meat tissue microstructure in situ by MRI using diffusion tensor imaging (DTI) at different b-values in order to assess by quantitative imaging the fiber structure of muscle tissue at the mesoscopic scale. Moreover, knowing that the muscle structure can be roughly modeled as a square lattice arrangement of fluid-filled cylinders surrounded by fluid, we can plot the theorical diffusion decay. This plot exhibits deep gaps in attenuation curves that have been directly correlated to the geometry of the cylinder lattice.

We have so attempted, in a second objective, to determine if such attenuation behavior can be found in the real meat structure made of aligned fibers. The tools that were set up in this part of the DREAM project can be used as input for creating artificial mimetic meat sample and to feed mathematical models.

35





Results and applications

Regardless of the sample orientation relative to the magnet, diffusion weighted signals were obtained with gradient fields applied in 6 directions and then calculation (tensor diagonalization) was made in order to consider diffusion in the meat fiber oriented frame. The huge advantage of this approach is that the diffusion (closely linked to the spatial fiber organization) is then assessed free of the effects of local fibers orientations.

After diagonalization, the first eigenvector corresponds to the main fiber axis direction. The second and third eigenvectors correspond to the 2 orthogonal directions of the fiber axis. The same process was applied to each 6-directions dataset at increasing b-values (ranging 100-20000 s/mm²).

Quantitative mapping

Parallel and perpendicular decays both deviate from the Gaussian diffusion, expressed by a mono-exponential decay since they follow a bi-exponential shape (Figure 1). This highlights hindered and/or restricted diffusion of water, out of and into muscle fibers, or exchanging between the two compartments. Diffusion maps reveal details due to spatial variations of structure at a resolution much below the acquisition resolution.

Figure 1: Diffusion-weighted intensity attenuation images from the trace of the diffusion tensor matrix. L image (left) is at low b value, H image (right) is at high b value, inhibiting respectively high diffusion corresponding to hindered diffusion and low diffusion corresponding to restricted diffusion.

Identification

of

structural

and

morphologic components

After MRI experiment, samples were frozen

at -180°C to avoid muscular cells degradation

and microscopic histological observations

were performed. After spatial registration,

histological images were first super-

imposed on high-resolution images

(susceptibility-weighted gradient-echo

images, with 300 × 300 voxels) to iden-

tify morphologic components (fat/col-

lagen network), and subsequently



superimposed on quantitative diffusion

maps (Figure 2).

Figure 2: Superimposition of histological and diffusion image.

36



Relationship between diffusion and muscle fibers types

To assess the architecture of the muscle, we investigated into the relationship between diffusion and muscle fibers types. Histological cuts (approximately 1mm x1mm) were observed after histoenzymological ATPase staining with pH 4.35 preincubation. At low b-values, diffusion parameters seem to correlate with metabolic characteristics of meat fibers, as highlighted by photomicrography of areas characterized by histoenzymological ATPase staining (Figure 3).

Figure 3: Superimposition of histoenzymological ATPase stained and diffusion images. On the histoenzymological images the three intensities are black (type I), white (IIA) and grey (IIX plus IIB). Diffusion is mapped with a blue/green to orange/red scale.

Variability of the dimensions of structural components

Less deep minima amplitudes on acquired data in homogeneous region are observed compared to the prediction of the mathematical models. The experimental data plot deviates from the theoretical plot mainly due to the heterogeneity in the fibers diameter, the no strictly parallelism of the fibers and also perhaps the no strictly impermeability of the fibers membranes.

37

Significance and benefits

Prospects and challenges

It is essential in the building of a mimetic model

The results of this research are a step ahead to-

where structural components are involved to pre-

wards the construction of realistic food models such

cisely access these components in situ on real ma-

as meat and will make possible to feed the models

terial. The tools that were set up in this part of the

built in the other workpackages. Promising results

DREAM project and results describing structural

have been obtained showing structural details cor-

meat components can be used as input for creat-

related with metabolic characteristics. To the best of

ing artificial mimetic meat sample and could be ap-

our knowledge, it is the first time that a scatter-like

plied to many other targets. We have showed that

behavior was observed in a biological matrix which

characterizing the behavior of water diffusion within

seems very promising for quantifying structural in-

a voxel provides a means for describing the inner

formation from the resulting diffusion attenuation

microstructure at a cellular scale, taking into ac-

plots. Future efforts will be put on the differentiation

count apparent diffusion coefficients measured in

within type of meat muscle and meat fibers types

muscle. Using the anisotropy of water diffusion due

and on the variability of the structural components.

to the highly fibrillar structure of meat, we modeled

Furthermore developed methods could be applied

this diffusion in three dimensions using tensors. We

on other food product.

used Diffusion Tensor Imaging (DTI) with different

b-values to obtain high resolution diffusion parame-

References

ter mapping of tissue which were registered to high-

Damez, J. L., Clerjon, S., & Bonny, J. M. (2010). Modeling

resolution susceptibility-weighted gradient-echo im-

of Diffusion Behavior In Muscle Tissue For

ages and histological images to determine if there

Microstructure Characterization by Quantitative Imaging.

In 10th International Conference on the Applications of

existed relationships between meat microstucture

Magnetic Resonance in Food Science). Clermont-Ferrand,

and diffusion observed at a meso-scale.

France.

Damez, J. L., Clerjon, S., Labas, R., Danon, J., Peyrin, F.,

& Bonny, J. M. (2011). Microstructure characterization

of muscle tissue by quantitative imaging. In 1st joint

french-spanish GERM-GERMN). Sitges, Spain.

Damez, J. L., Clerjon, S., & Bonny, J. M. (2012).

Microsctructure characterization of muscle food by quan-

titative MRI. In 11th International Conference on the

applications of Magnetic Resonance in Food, (pp. 60).

Wagenin gen. The Netherlands

Damez, J. L., Clerjon, S., Labas, R., Danon, J., Peyrin, F.,

& Bonny, J. M. (2012). Microstructure characterization of

meat by quantitative MRI In 58th International Congress

of Meat Science and Technology, (pp. p17 ).

Montreal, Canada.

Authors

Astruc Thierry, Bonny Jean-Marie, Clerjon Sylvie,

Damez Jean-Louis, Labas Roland, Vénien Annie

Organisation/Institute

UR370 Qualité des Produits Animaux, INRA, F-63122

Saint Genès Champanelle

Contact

Damez Jean-Louis

UR370 Qualité des Produits Animaux, INRA

E-mail address: jean-louis.damez@clermont.inra.fr

38



Combined gelled /

nt

dispersed / aerated

odel

systems model

velopme

l de essert m

It this information important for me?

Consumers

Food Industry

ode

Regulatory Authorites

Scientific Community

xisting d

sse m f e

che

WP-4

esearch o

Workpackage



additional r

39



Task 4.1

The objective was to produce and characterise a range

of gelled / dispersed / aerated generic model foods.

The primary model was a dairy dessert that could be

used for risk assessment and was tested for use in





Introduction of WP4


following lipid oxidation. The changes in properties of

the dessert were described as a function of the ratio of

caseins to whey proteins.

Task 4.2

Alan Mackie, Workpackage Leader

This task focused on producing a range of model

Institute of Food Research (IFR)

alan.mackie@ifr.ac.uk

cheese systems that could be used for assessing the

effect of changes in formulation or processing on in-

dustrially relevant parameters. In particular the effect

of salt reduction was assessed in the micro-cheese,

the generic cheese and the standardised pilot-scale

Prospects

cheese milk models.

At the start of the project there were no gelled

or dispersed model systems that could be



consistently reproduced and that could be made

Main outcomes

widely available for testing nutrient, allergen or toxi-

This workpackage was split into two tasks, each

cant release, microbiological safety etc.

looking at a different food group, namely dairy



Therefore we aimed to produce well characterised,

dessert and cheese. For the main dairy dessert

realistic, food models for these types of food sys-

model, three kinds of rheological behaviour and four

tems including cheeses and desserts that could

different interfacial structures were characterized.

serve as tools for harmonising food quality, safety

In the first group, the emulsions were liquid,

and nutrition research on these types of food prod-

showing no connections between droplets.

ucts.

In the second group the emulsions were struc-

We also aimed to produce mathematical



tured, the droplets being weakly connected by WP

models that could relate model composition such

aggregated.

as fat content, protein ratio and processing to

model functionality such as nutrient release and

In the third group the emulsions gelled, the droplets

texture.

being connected by CM/WP complex or in other case

by aggregated WP. In the same task the impact of

enzymatic cross linking of the protein was demon-

Aims and background

strated and in a slight variation to the basic model

the impact of homogenisation pressure on the tex-

The two main models produced from this WP both

tural properties of cream cheese was also demon-

have great potential for the future but the develop-

strated.

ment of the models also leaves a number of questions

unanswered. The standardised cheese milk model is

The studies on cheese models yielded a generic

already being used by Actalia in the TeRiFiQ project

cheese model for studying for studying the effect

and a simplified version has been developed by Sore-

of composition of texture and bacterial growth. The

dab for use with SMEs who do not have the expertise

use of a micro-cheese model for high throughput

or equipment to use the complex model. Additionally,

screening of bacterial growth was also demonstrat-

results from the dairy dessert model are being com-

ed. Finally the primary cheese model yielded a robust

bined with other data from the literature to build an

protocol for the standardisation of the milk used for

integrated knowledge model of protein based emul-

pilot scale production of experimental cheeses with

sion systems that will be much more predictive in

very low variability. The usefulness of the approach

nature.

was demonstrated for a wide range of cheese types

including Camembert, Brie and Swiss cheese.

40



The Dairy dessert: A model for designing

tailored interfaces in food systems

Composition and structure of interface impacts texture of emulsions

The main objective of this work was to understand the impact of interfacial composition and



organization on the connectivity between fat droplets and the microstructure of oil-in-water emulsions.

Four kinds of object were obtained by different processes applied to proteins solutions containing various protein ratios. Their different properties led to competition between these objects at the interface generating four types of structure with different connectivity leading to various rheological properties: liquid, structured or gel. In complex dairy desserts, only the presence of aggregated whey at the interface increased the firmness of these desserts. When emulsions and dairy desserts were enriched in PUFAs, the oxidation of lipids remained low after 2 months storage at 4°C.

Research aims and background

Most processed foods contain gels and dispersions

proteins. Consequently, the aim of this study was to

of some sort that consist of small particles such as

evaluate the combined effect of heat treatment and

fat or protein dispersed in another medium. There

CM/WP ratio change, on the structure of interfaces,

are currently no gelled/dispersed model systems

in relation with the impact on the texture of O/W

that can be consistently reproduced and that can

emulsions.

be made widely available for testing nutrient, al-

For that, our strategy was to generate emulsions

lergen, toxicant release, microbiological safety etc.

with a range of well-defined different interfaces and

The work described here addresses this issue with

combining different compositions (weight ratio of

a standardised dairy dessert model. The production

CM to WP: 80:20 to 12:88) and heat treatment of

of the dairy dessert is built on experience at IFR

milk proteins (60°C or 80°C).

(UK), INRA and Soredab (France) in studying and

understanding proteins/surfactant interactions in

emulsions and foams, fat composition, and on the

Results and applications

use of biopolymers such as starch or pectin as thick-

The results of the work on the dairy dessert have

ening/gelling agents. We hypothesized that changes

been published in a series of papers from INRA

in protein structure caused by heat treatment could

Nantes and INRA Grignon (Foucquier et al. 2011;

induce different structures of the interfacial layer

Surel et al. 2013). The way the dessert model is

(difference of thickness and of homogeneity) with or

made is shown in Figure 1. Initial measurements de-

without disulfide bridges.

termined the mean size of the casein micelles (CM)

These different morphologies would then have an

to be 140-160 nm regardless of processing tempera-

impact on the structure and the texture of emul-

ture (60°C or 80°C) whereas the whey protein (WP)

sion, that could moreover affect release and bioac-

increased in size from 6 to 100 nm as the processing

cessibility of nutrients contained in fat (PolyUnsatu-

temperature increased.

rated Fatty Acid ω-3, for example) or digestibility of

41





The emulsion droplet size was remarkably insensi-

tive to either CM/WP ratio or temperature.

The primary controlling factor of the model was the

interfacial composition as a function of CM/WP ra-

tio and processing temperature. When the ratio was

below 0.2, the interface was dominated by WP and

at 0.2-0.3 there was a minimum in the amount of

protein adsorbed to the interface. At CM/WP ratios

above 0.3 the interface was dominated by the CM

and the effect was more marked at the higher tem-

perature. These differences in interfacial composi-

tion also had a marked effect on the interactions

between the droplets and thus the rheological be-

haviour of the emulsions.

Figure 1: Flow diagram for the production of the Dairy

Dessert.

The overall effect is shown in Figure 2 in which the

first group with a liquid texture consists of emul-

sions whose proteins were treated at 60°C regard-

less of the CM/WP ratio and those with a ratio of 0.8

or more treated at 80°C. The second group produc-

ing a structured liquid with higher viscosity includes

emulsions with CM/WP ratios from 0.19-0.26 treated

at. The final group of gelled systems are all treated

at 80°C and have CM/WP ratios from below 0.15 or

from 0.3 to 0.5. In these last systems the emul-

sion droplets are all interconnected by aggregated

protein causing the system to gel. In parallel with

the experimental work undertaken to develop the

physical model, a mathematical model was devel-

oped that aimed at predicting the texture of the final

Figure 2: Schematic of the different structures induced by

composition and processing.

model based on a number of inputs, including the

amount of casein and whey, the size of the whey

aggregates, the processing temperature and the

homogenisation pressure. From this information a

number of internal variables are generated that pro-

vide information on the amount of adsorbed CM and

WP, the amount of aggregated WP adsorbed and the

emulsion droplet connectivity. From these variables

the structure of the resulting emulsion system can

be predicted.

Equipment required in order to produce the model is

relatively basic, comprising stirrers, heaters and an

Ultra Turrax to make the pre emulsion that is then

Figure 3: Schematic of the mathematical modelling ap-

passed through high pressure homogeniser capable

proached used to link formulation and processing to

texture.

of delivering 50 bar.

42



Significance and benefits

Prospects and challenges

This model offers the possibility of using a stan-

This model has successfully been used to determine

dardised dessert model with a range of textural

lipid oxidation rates after the incorporation of kiwi

properties. The texture of the dessert can be pre-

seed oil containing high levels of ω-3 polyunsaturat-

dicted from the amount of micellar casein and whey

ed fatty acids into the formulation. No statistically

protein added and from the thermal treatment and

significant differences were seen in the levels of oxi-

the homogenisation pressure used to make the

dation after 2 months of storage at 4 °C regardless

emulsion. The standardised and predictable nature

of formulation or processing temperature.

of the model makes it ideal for use in risk assess-

ment involving the growth of pathogenic or spoilage

Successful applications

organisms or involving toxin contamination. In ad-

dition it is also suitable for use in studies on bioac-

The development of mathematical models able

cessibility or bioavalability using in vitro or in vivo

to predict the formation of structure and thus

methods respectively.

the texture of dispersed systems represents an



interesting prospect for the future. It also of-

fers the possibility of broadening the range of



References

dispersed food systems that can be included in the

Foucquier, J., Gaucel, S., Surel, C., Riaublanc, A., Bau-

drit, C. and Perrot, N. 2011. Modelling the formation of

modelling.

fat droplet interface during homogenisation in order to

In particular, coordinated approaches to producing

describe the texture. Pages 706-712 in: 11th Interna-

healthier food systems with lower salt and/or fat

tional Congress on Engineering and Food. G. Saravacos,

P. Taoukis, M. Krokida, V. Karathanos, H. Lazarides, N.

and/or sugar are becoming increasingly important.

Stoforos, C. Tzia and S. Yanniotis, eds. Elsevier Science

The development of standardised models of signifi-

Bv: Amsterdam.

cant food types offers the possibility of designing

Surel, C., Foucquier, J., Perrot, N., Mackie, A. R., Garnier,

healthier foods with the desired textural properties

C., Riaublanc, A. and Anton, M. 2013. Composition and

and shelf-life.

structure of interface impacts texture of O/W emulsions.

Food Hydrocolloids in press.

Authors

Claire Surel1, Catherine Garnier1, Alain Riaublanc1,

Marc Anton1, Alan Mackie2

Organisation/Institute

1BIA, INRA, Nantes, France

2IFR, Norwich, UK

Contact

Marc Anton, BIA, INRA

Marc.anton@nantes.inra.fr

43



A pilot scale model for the reproducible

production of soft cheese

A pilot scale model for the reproducible production of soft cheese

A realistic cheese model (Brie cheese manufactured with industrial technologies, Figure 1) was built as tools for experimental studies on cheese. A great attention has been paid to the authenticity and repeatability of these small scale models (around 1kg). Twenty replicates per cheese model were made to determine the repeatability of about a hundred state-variables and control-variables: from the milk composition to the cheese qualities. Their coefficient of variation was around 1%-1.5%. For each of these variables, statistical analyses were performed to characterize the dispersion of the data and the origin of the variability, in order to improve the models. Reliability of the models was confirmed in experiments dealing with the influence of salt content in cheese on Bifidobacteria in Brie-cheese.

Research aims and background

Many problems in the field of dairy research and

utes in a part of the progress made in the cheese

development require the implementation of cheese

industry by the improvement of the knowledge in the

making trials. These trials include technological top-

dairy science. More particularly, experimental results

ics such as process modifications, yield, raw materi-

on food safety (e. g. growth and survival of patho-

al, spoilage microorganisms, sensory studies (effect

genic bacteria in cheese) or on the cheese milk quality

of starter or adjunct cultures), nutritional questions

are very important for the cheese makers and require

(reduction of salt or fat content, decrease of the pro-

suitable cheese models. The screening of the ingredi-

portion of saturated fatty acids in fat and the use of

ents for cheese making must also be performed using

probiotics and/or prebiotics) and food safety issues

viable models. Finally, the strategy of characterisation

(survival of pathogens, presence of harmful chemi-

of the representativeness and the repeatability can be

cals). However, cheese making experiments are ex-

applied to the pilot plant in the cheese factories.

pensive and time consuming, even on a pilot scale.

They require specific equipment and required envi-

Results and applications

ronmental conditions are difficult to control. There-

fore, several alternatives have been suggested for

Preparation of the cheese milk as outlined in the

experimental studies on cheese. The improvement

schematic diagram in Figure 2, starts with the

of cheese models and the proposition of a strategy

heat treatment of raw milk (88°C/1min). This is

for the development of cheese models, as proposed

to denature the whey proteins and to inactivate v

in our study, are useful for the cheese industry and

egetative cells of bacteria present in the milk. This is

its suppliers (e. g. enzymes, lactic and ripening cul-

followed by the preparation and addition of protein

tures, proteins, etc). Indeed, cheese models are too

concentrate to increase of the casein content of

often considered as black boxes in some laboratories

cheese milk. The next stage is to adjust the fat and

and so their improvement will increase the reliability

recoverable proteins content which in turn controls

of the results. Indeed, the dairy research contrib-

the fat-in-dry-matter of the final cheese.

44





Results and applications

The last step in the process is the microfiltration of the skim milk (1.4 µm) and heat treatment of the cream (120°C for 1 minute) in order to remove spores and thermo-resistant bacteria.

The cheese itself is prepared as follows: The milk is prepared with starters and acidifiers at 39°C for 30 min in order to standardize the pH at renneting. The milk is then coagulated with recombinant chymosin to gel the milk, which is then cut into 1.7x1.5x1.5 cm pieces to promote syneresis. The vat containing the milk gel is then drained and the cheese placed in moulds, which are in turn drained at 32°C for 3h then 18°C. After 1

day the cheese is placed in saturated brine at 12°C for 55 min and then ripened for 12 days (12°C, 96% RH) on grids before wrapping and storing. Mean value and standard deviation were measured for the composition of the cheese milk, day-1 cheeses and the ripened cheeses respectively. A good reproducibility was obtained for all state variables with coefficient of variation around 1-1.5%

The equipment needed for milk standardization is: Pilot scale microfiltration with 1.4 µm ceramic membranes, Pasteurizer, Tri-blender (or mixer) and a skimming centrifuge. Cheese manufacture requires a cheese vat, cutting blades or wires, cheese moulds and a thermostated cheese making room. Finally, cheese ripening requires wire grids and a ripening room (RH and temperature controlled).

Figure 1: Brie style soft cheese produced using the cheese

milk model

Figure 2: A flow chart for the standardisation of cheese

milk in the soft cheese model

Significance and benefits

A detailed, realistic and reproducible cheese model

was built and characterized. The model is currently

Prospects and challenges

used in our laboratory and could be used as a ba-

This model offers a way of producing pilot scale

sis for model development in research institutions

cheese with low variability. One of the main



or industrial laboratories. A simplified version of

limitations of the model is its complexity.



the model has been developed in collaboration with

Therefore, one task for further development is the

Soredab.

simplification of the model so that it may be used

more widely in the cheese making industry.

Successful applications

This model has been successfully applied to study

the influence of salt in moisture on the growth and

Authors

survival of Bifidobacterium lactis BB12. The experi-

Jean-René Kerjean1, Romain Richoux1, Alan Mackie2

mental data obtained from the model was in agree-

ment with previous studies, showing an inhibitory

Organisation/Institute

1

effect of the increase of the salt in-moisture on bu-

Actalia, Rennes, France

2IFR, Norwich, UK

tyric acid fermentation. Our results obtained in Brie

cheese also confirm previous findings which showed

Contact

little influence of salt content on the survival of B.

Jean-René Kerjean, Actalia

lactis BB12 in cheese.

jr.kerjean@actalia.eu

45



Coupled impact of product process and

composition in lipids as emulsifiers on the

structure and the texture of a cream cheese

Impact of product process and composition on cream cheese texture and oxidation.

A cream cheese is a dairy emulsion where proteins act both as emulsifiers at the fat globule interface and as protein network components. The objective of this work is to understand the impact of the process and the addition of lipid emulsifiers on the interface composition then on the model texture. The most discriminating step of the process on product’s rheological properties is the homogenization. Nevertheless, its influence is modulated by the heat treatment. The addition of lipid emulsifiers decreases the connectivity and hence decreases the product’s firmness. These molecules compete with proteins and replace them at the interface.

The cream cheese model is not sensitive to oxidation either after intense thermo-mechanical treatment, storage, or enrichment in polyunsaturated fatty acids.

Research aims and background

The problem was to have a cream cheese model easy to produce (Fig 1.), with well characterized composition to understand the impact of the process and its composition in lipid and protein emulsifying agents on its structure, texture and oxidation stability.

There are some previous works on the impact of product process on product texture (Sanchez et al., 1996).

However, there is a lack of information regarding the influence of each process step on final product.

Moreover, the emulsifying capacity of phospholipids is well known but the influence of its addition on fat globule interface, product structure and texture stays little known.

Results and applications

Regarding the impact of the process (described in

pH 4.9 (Coutouly et al., 2013).

Fig 2) on the structure and the texture of the cream

Regarding the impact of enrichment of the cream

cheese, the project showed that the final homog-

cheese with lipid emulsifiers, the project showed that

enization pressure has a large impact on the cream

the phospholipids proves to have a larger affinity for

cheese model texture. When the pressure increas-

fat globule interface than proteins. As low molecular

es, the fat globule size decreases. This leads to an

weight surfactants (Mackie et al., 1999), they replace

increase in the specific surface of fat, number of

proteins at the interface (Fig 3.). Consequently, the

interactions between particles and consequently in

number of interactions between particles decreases

the cream cheese model firmness.

and the cream cheese model firmness decreases

The pH at the end of acidification has a very limited

consequently. The presence of phospholipids results

impact on rheological characteristics of the model,

in partial coalescence of fat droplets which yields

but it has a significant one on its sensory properties:

softer and smoother final products.

products acidified to pH 5.2 are considered more

spreadable and brighter than products acidified to

46





Finally, regarding the stability of the cream cheese model, enriched or not with polyunsaturated fatty acids, and whatever the process to manufacture it, the cream cheese model proved to be very stable regarding the lipid oxidation, during 3 months of storage at 4°C.

This project allowed better understanding of how the interactions between particles and microstructure impact the final cream cheese texture.

Significance and benefits

The project was performed with cream cheese mod-

els analogous to actual products.

The cream cheese model manufacture integrated

a process and some ingredients commonly used in

dairy industries.

The different parameters studied here (process



parameters such as heat treatment temperature,

final acidification pH and homogenization pressure;

or ingredients such as proteins or phospholipids) can

be tested straightforwardly in dairy industries.

The project outsides can give direct applications /

solutions for industries.

Figure 1: cream cheese model

Thorough characterization of the product was

achieved at each stage in the process and at



different scales.

The consequence of each variation of the process

or of the formula has been characterized from



nanoscale through electron microscopy to mac-

roscale through rheology and sensory analysis.

The project allowed obtaining an overview of the

possible parameters that can be used to modulate

cream cheese texture. These parameters include

both technological and formulation factors.

Successful applications

The cream cheese model could be used in other

research projects.

The direct results provide information to link



process, formula and product structure and



texture.

Moreover, this work allowed the production of



various data which are currently used for mathemat-

ical modeling. Two studies are in progress: modeling

protein aggregates formation in tubular exchanger

during heat treatment and the modeling coloniza-

tion of the fat globule interface upon homogeniza-

Figure 2: Cream cheese process

tion and induced connectivity.

47





Figure 3 : Schemes (a) and transmission

electron miscrocopy images (b) of cream cheese

with or without phospholipids

References

Prospects and challenges

Coutouly A, Riaublanc A, Axelos M, Gaucher I. 2013.

Impact of heat treatment, homogenization pressure and

This work raises various perspectives. Firstly, the

acidification on the structure and texture of a cream

analysis of the fat globule interface has to be more

cheese model. J Dairy Sci Tech. DOI: 10.1007/s13594-

013-0148-z

specific. In order to reach this goal, a method for the

characterization of the fat globule interface in the

Mackie AR, Gunning AP, Wilde PJ, Morris VJ. 1999. Oro-

genic displacement of protein form the air/water interface

final product should be developed.

by competitive adsorption. J Colloids Interf Sci. 210:

Moreover, for further work, a source of purified

157-166.

phospholipids or other emulsifiers should be used.

Sanchez C, Beauregard JL, Chassagne ML, Bimbenet JJ,

Indeed, in this work we used buttermilk isolates as

Hardy J. 1996. Effects of processing on rheology and

source of polar lipids. Finally, it will be necessary to

structure of double cream cheese. Food Res Int. 28 (6):

547-552.

quantify the connectivity and link it practically with

rheological properties and the texture of product.

Authors

Aliénor Coutouly1,2, Isabelle Gaucher2,

Alain Riaublanc1

Organisation/Institute

1INRA, UR1268 BIA, Nantes, France

2Soredab, La Boissière Ecole, France

Contact

Isabelle Gaucher

SOREDAB

48

isabelle.gaucher@soredab.org





Open solid foams model

D&

erical R elationships

um

It this information important for me?

Consumers

Food Industry

or n

Regulatory Authorites

Scientific Community

ata f

cture-function r

input d tru

WP-5

Workpackage



hing cereal s

researc

49



Task 5.1

Definition of the working domain and relation

with WP1 and WP6: Food models and process and

recipe variables were selected. Processes required for

manufacturing the food models were detailed accord-

ing to the possibility of control and in-line measure-





Introduction of WP5


ment. They were transferred to WP6 for safety and

nutritional tests. The available numerical models and

the know-how for structure-properties and structure-

processing relationships (BKM) were collected.

Task 5.2

Cereal foam structure design and processing:

Kaisa Poutanen, Workpackage Leader

VTT Technical Research Centre of Finland

Laboratory scale processes, and in biscuit model also

Kaisa.Poutanen@vtt.fi

pilot processes, were investigated to elucidate the

most essential phenomena for cereal foam design.

Samples were taken during and after processing for

further analysis, or measured in-situ during process-

ing. The phase/state changes were assessed at dif-

Aims and background

ferent structural levels, by determining starch digest-

Open solid foams represent the structure of baked

ibility, rheological properties, porous architecture by

cereal products, the properties of which are closely

X-Ray tomography, electron and light microscopy, as

well as water distribution by NIR and calibrated colour

related to the intrinsic properties of the solid materi-

imaging.

al, its density and cellular structure. The solid phase

is a composite material blend of biopolymers and

Tip!

Task 5.3

components of lower molecular weight. The size and

Improving (numerical) models and generating

pathways for cereal model foams: Partners’

distribution of pores is also important in the prod-

know-how and expertise were combined for specific

uct behaviour. Fracture properties and dynamics of

process operations (mixing, bubble formation, bak-

water in the matrix are important for the sensory

ing) and product specifications in order to complete

quality of products based on solid foam structure.

BKMs, where differential equation (PDE) based models

are too complex or not available; the existing models

The micro and macro level structure is likely to have

were extended to the computation of other variables

an important effect on digestibility and delivery of

(colour during baking, for instance), in WP1. Integrat-

small molecular weight components in the human

ing BKMs will allow to design complete pathways for

gastrointestinal tract.

cereal food processing.

The aim was to develop models for open cereal solid

foam foods with special reference to effects of whole

grain and fibre on structure.

Main outcomes

Generic model foods (GMF) were developed for solid

cereal foams; bread and biscuit, and accurate meth-

Prospects

ods to characterize these were developed. A pre-

liminary cake model was also developed. Structural

The results showed the importance of models in un-

changes during processing were ascertained for

derstanding the role of food formulation and process

bread, with emphasis on the creation of the cellular

variables at different structural levels of the product

structure and its heterogeneity. Dough liquid phase,

and further on nutritional properties, stability and

including sugar and fat, was studied as a stabilizer

safety. In bread, important nutritional benefits in-

of the air bubbles in dough and bread matrix. We

clude high fibre content and glycemic response. In

have used and further specified the model for bread

the former, maintenance of good sensory texture is

in the ring test, and protocols for the bread model

critical, and in the latter, food structure has a de-

were validated and disseminated for end users. Ef-

cisive role. In biscuits, additionally fat quality and

fects of oven heat flux on biscuit moisture distribu-

amount of sugar should be considered. Fundamental

tion and colour were studied, and a model for bubble

understanding and modelling of structure formation

growth and biscuit thickness was made. The models

helps to develop solid foam foods with consumer

were used to study effects of dietary fibre /bran ad-

benefits.

dition in bread and biscuits, and to study effects of

fat and sugar in bread.

50





Structural challenges in nutritionally

improved biscuits

How to produce whole grain and high-fibre biscuits with slow starch digestibility

and high sensory quality?

Biscuits are an important cereal food category, made most often of white flour, fat and sugar and consequently typically have high calorie but low nutrient density. We wanted to understand effects of adding dietary fibre (whole grain flour and wheat bran) on biscuit microstructure, texture and in vitro starch digestibility.

Fine and coarse wheat bran were used to make five different biscuits with various dietary fibre content (5-15

% ). The biscuits with small bran particle size had the best sensory texture. The generally low biscuit starch digestibility in vitro was increased slightly by wheat bran addition.

Research aims and background

There is a large consumer and public health demand

The results are important for the baking industry

to increase the nutritional profile and expected

and biscuit manufacture in development of more nu-

health benefit of cereal based snacks. Biscuits, with

tritious products. The knowledge of interactions of

white flour, fat and sugar as major ingredients, are

dough components and bran particles benefits also

popular snacks and also breakfast items. The pro-

producers of other baked snacks and dry products.

duction steps during biscuit manufacturing do not

allow starch to gelatinize completely in spite of the

baking process, and the glycaemic response of bis-

cuits is typically lower than that of bread. On the

other hand, the nutritional profile of biscuits is not

optimal Making biscuits of whole grain flour or add-

ing dietary fibre in the recipe would improve their

nutritional profile, but poses challenges to the open

solid foam structure determining the mechanical

properties and sensory quality of biscuits.

We wanted to elucidate the role of dietary fibre in

the form of cereal bran, as well as that of whole

grain flour on biscuit structure and starch in vitro

digestibility of biscuits. The aim was to be able to

change the process conditions so as to maintain

good sensory perception of texture and low starch

digestibility while increasing the amount of cereal

Figure 1: Biscuit assessment methods: Texture (2mm rod

fibre and associated phytochemicals in the biscuit

penetrometry)

formulation.

51





Results and applications

Five types of biscuit were produced containing 5-15

% dietary fibre, including a standard recipe and

biscuits made with coarse (260-560 µm) and fine

(25-160 µm) wheat bran. Textural measurements

were made with a three point bending test. Fur-

ther measurements were made by a penetrometry

method for 40 types of biscuit with a wider range

of fibre variations including bran and soluble inulin

fibre. Starch crystallinity was analysed by

differential scanning calorimetry, and in vitro starch

digestibility rate was determined as hydrolysis index

Figure 2: Biscuit assessment methods: Colour

by an enzymatic method.

Biscuits became darker with increasing bran con-

tent. Bran particle size had little effect on average

colour, but the particles were visible as specks.

The mechanical properties of the bran-containing

biscuit matrix were affected by both by bran content

and particle size. Addition of wheat bran increased

the penetration energy (‘firmness’), only slightly

increased the elastic modulus, and reduced the

number of penetrometry force peaks (‘crunch’).



Inulin also increased firmness but, unlike bran,

resulted in a high initial penetration force. Bran

particle size reduction had little effect on firmness,

but increased the elastic modulus and hardness

measured by three point bending. Bran supple-

mentation level had a greater effect on penetrom-

etry tests than particle size, but particle size had a

Figure 3: Biscuit assessment methods: Mass

greater effect in three point bending. The failure

strain at maximum bran loading level was the low-

est among the coarse bran supplemented biscuits.

These samples had very low strain at failure which

varied between 1.1-1.4%, which is unique to brittle

materials.

Structural factors had more impact on in vitro starch

digestibility rate compared to the status of starch

which was measured by degree of gelatinization.

Biscuits with finely ground bran had a visually more

compact structure without any surface or internal

defects and were harder than those with a coarse

bran particle size. Increasing fibre content from 5

to 15 % increased the hydrolysis index by 20%.

The effect of bran particle size reduction in biscuit

formulations did not influence hydrolysis index.

Figure 4: Biscuit assessment methods: Dimensions

52



Significance and benefits

Prospects and challenges

It is a food engineering challenge to increase con-

We hope that the results will encourage biscuit man-

tent of dietary fibre in biscuits while retaining struc-

ufacturers in development of high-fibre products,

tural and sensory characteristics, including texture,

thereby diluting the amount of refined ingredients in

colour and taste. There are very few reports avail-

the product. The nutritional and health claim regula-

able about adding dietary fibre in biscuits, and ef-

tion by European Food Safety authority would allow

fects on these characteristics or starch digestibility.

a claim for fibre content of products with adequate

The use of whole grain flour or bran seems a logical

nutritional profile. The idea of bran pre-treatment

way of increasing fibre content of biscuits, but no

prior to incorporation in baking could also be devel-

reports of their use are available. Fundamental un-

oped further in collaboration with milling and baking

derstanding about structure- function relationships

industry.

in biscuit structure and fracture properties also is

limited. Current milling techniques enable efficient

particle size reduction of bran, delivering new types

of ingredients for the baking industry. Their applica-

tions have hitherto been reported to a limited ex-

tent. The approach of this work provided thus many

aspects on novelty.

The results, when applied, would encourage the use

of finely milled bran and whole grain flour as bis-

cuit raw materials, to produce biscuits with good

sensory texture and colour, and retain the naturally

low starch digestibility in biscuits. The benefit for

biscuit manufacturers would be the production of

new, healthier products, providing consumers with

healthier options. The use of bran in food production

would also increase sustainability, as a side stream

could be utilized in a mainstream food product.

References

Sozer N, Cicerelli L, Poutanen K (2013). Influence of

wheat bran on biscuit structure and starch digestibility.

Journal of Cereal Science Submitted

Sozer N, Cicerelli L, Poutanen K (2013). Addition of

wheat bran influences biscuit structure but not in vitro

Authors

starch digestibility. Poster presented at C&E Spring Meet-

Kaisa Poutanen1, Nesli Sozer1 , Martin Withworth2,

ing, May 29-31, Leuven, Belgium.

Guy della Valle3, Lucio Cicerelli4

Sozer N, Cicerelli L, Raija_Liisa Heiniö, Poutanen K

(2013). Wheat bran addition and particle size reduction

Organisation/Institute

have an influence on biscuit structure but not in in vitro

1VTT Technical Research Centre of Finland, POB 1000,

starch digestibility. Poster presented at IFT Annual meet-

FI 02044 VTT, Espoo, Finland

ing and expo, July 13-16, Chicago, US.

2Campden BRI, Chipping Campden, Gloucestershire,

GL55 6LD, U.K.

Whitworth MB, Chau A, Cicerelli L. (2011) Effects of fibre

3INRA, UR 1268 Biopolymères, Interactions & Assem-

and baking conditions on digestive biscuit properties.

blages (BIA), Nantes , France

Poster presented at AACC International Annual Meeting,

4United Biscuits, High Wycombe, Buckinghamshire,

December 16-19, Palm Springs, US.

HP12 4JX, U.K.

Whitworth MB, Gates FK (2013) Measuring and address-

ing texture challenges in healthy baked products, Cereal

Contact

Foods World 58(3), 120-125.

Kaisa Poutanen, VTT Technical Research

Centre of Finlande

Kaisa.Poutanen@vtt.fi

53



Creation of bread cellular structure

Designing bread texture thanks to the solid foam model

The growth of gas bubbles in viscous matrices is the main mechanism responsible for the structure of open solid foams, like bread.

The creation of the cellular structure in wheat flour dough and its heterogeneity was ascertained at different structural scales, focusing on proofing. Dough liquid phase, including sugar and fat, was studied as a stabilizer of the air bubbles in dough and bread matrix. The contribution of each scale was integrated by adapting a capillary number. We have used and further specified the model for bread in a ring test, in order to increase its fibre content, and protocols for the bread model were validated and disseminated for end users.

Research aims and background

The texture of bread is a fundamental element of its

which underlines the importance of minor compo-

acceptability by the consumers, and it has a strong

nents. Starting from the study of bubbles growth

impact on its nutritional properties. For instance,

and coalescence in model dough systems, the aim of

increasing the fibre content leads to denser, and

our work was to determine the role of the aqueous

less acceptable breads. Like solid foam mechanical

phase and the starch/gluten matrix on the mecha-

properties, texture depends on density and cellular

nisms which govern the creation of cellular structure

structure of bread. So, there is a need to better un-

at microscopic scale, and at macroscopic scale on

derstand the mechanisms of cellular structure cre-

the loss of stability at the end of fermentation.

ation during dough processing and provide models

The results are important for the baking industry,

to better control process and final properties.

first for manufacturing breads with increased fibres

Wheat four dough contains about 45% water (tot.

content, without loss of sensory quality. Secondly,

basis) and its cellular structure is created during

it strengthens the knowledge on the breadmaking

fermentation, or proofing; during this stage, poros-

chain by providing engineers with basic knowledge

ity increases from 0.1 to 0.7 and CO2 bubbles are

models.

connected together, although the dough does not

collapse. This stability could be due to the forma-

tion of a liquid foam, constituted by a liquid phase,

Results and applications

co-continuous to the starch/gluten visco-elastic ma-

About twelve recipes of dough were processed

trix. To determine the role of this phase in the cre-

with varying content of sugar (0-15%), fat (0-

ation of the cellular structure, the aqueous phase,

10%), fibres (0-15%), in a range for which a typi-

so-called dough liquor (LdP), has been considered

cal bread cellular structure was always obtained.

as a good model for these interfacial films. In com-

The elongational properties of starch/gluten ma-

plement, elongational properties of dough can be

trix, measured by lubricated squeezing flow, largely

determined by lubricated squeezing flow test (LSF),

influenced dough proofing stability, the evolution

54





of which was assessed by 2D image follow-up and adjusted by an exponential decay; this result could not be explained by the single bubble growth model. Porosity kinetics determined at macroscopic level were in good agreement with results determined at microscopic levels by X-ray microtomography (XRT) at ESRF (F38-Grenoble), and both followed a Gompertz model. Analysis of 3D-XRT images showed that most bubbles were connected, for highest porosity of dough (≥0.5). The homogeneity of the cellular structure was defined from the size distributions of gas cells and walls; it was characterized by a critical thickness of walls (≈ 1μm), below which the cells were separated by liquid films.

Figure 1: Two dimensional images of dough from beginning (66 minutes) to the end (178 minutes) of proofing, diameter 5 mm.

Images are issued from Turbin-Orger, A., Boller, E., Chaunier, L., Chiron, H., Della Valle, G., Reguerre, A-L. (2012a). Kinetics of bubbles growth in wheat flour dough during proofing studied by computed X-ray micro-tomography. J. Cereal Sci., 56, 676-683. Images have been obtained at the European Synchrotron Radiation Facility at Grenoble, France.

The fermented dough could thus be



considered as a three-phases medium: visco-

elastic matrix / gas cell / liquid phase.

Dough liquor was taken as a model of this liquid

phase, and extracted from dough; it behaved like

a macromolecular solution, and was characterized

by the surface tension (≈ 40mN/m), related to the

presence of polysaccharides-proteins complexes at

interfaces.

The contributions of the different levels of



organization of the dough were then integrat-

ed by defining a (dimensionless) capillary num-

ber, that ruled the overall behavior of the dough.

These results, including process specifications,

were used in a ring test to study the effect of fi-

Figure 2: Three dimensional images of dough at the begin-

bre addition, in different labs; although various

ning (66 minutes) and at the end (178 minutes) of proofing.

texture properties were obtained, they could all be

Images are issued from Turbin-Orger, A., Boller, E., Chau-

nier, L., Chiron, H., Della Valle, G., Reguerre, A-L. (2012a).

integrated in the same relation between texture and

Kinetics of bubbles growth in wheat flour dough during

density, which finally validated the open solid foam

proofing studied by computed X-ray micro-tomography. J.

Cereal Sci., 56, 676-683. Images have been obtained at

model.

the European Synchrotron Radiation Facility at Grenoble,

France.

55



Significance and benefits

References

Turbin-Orger A, Della Valle G, Doublier JL, Marion D,

Improving bread nutritional properties without de-

Novales B. Poster : Foaming and rheological properties

creasing its sensory properties and, mainly its tex-

of the soluble phase of wheat flour dough. EUFOAM 2010

Conference Borovets, Bulgaria, 14 – 16 July, 2010

ture, is a real challenge in baking industries. A bet-

Turbin-Orger A, Boller E, Chaunier L, Chiron H, Della Valle

ter understanding of various operations is necessary

G, Réguerre A-L. (2012), “Kinetics of bubbles growth in

to control the density and cellular structure of these

wheat flour dough during proofing studied by computed

products, which can, in turn be related to texture

X-ray micro-tomography”, Journal of Cereal Science, 56,

by available mechanical models. In this purpose, a

676-683.

common representation of these operations can be

Turbin-Orger A, Chaunier L, Chiron H, Della Valle G.

obtained by the definition of Basic Knowledge Models

Alveolar structure of bread dough and rheological proper-

ties of its constitutive phases. AACC, Orlando, Oct 1-3rd,

(BKMs), which captures the main physics of the phe-

2012.

nomena involved. For instance, the relation between

Turbin-Orger A, Chaunier L, Chiron H, Della Valle G.

the capillary number and stability can be considered

Envisioning wheat flour dough as a triphasic medium to

as a BKM; it suggests that the simple measurement

predict bubbles stability. An. Trans. Nordic Rheol. Soc.,

of dough elongational viscosity and the knowledge

21, 2013

of dough liquor surface tension can lead to the pre-

Della Valle G, Chiron H, Cicerelli L, Kansou K, Katina K,

diction of the cellular structure of the dough.

Ndiaye A, Whitworth M, Poutanen K. Basic knowledge

models for the design of bread texture. Submitted to

TIFST, 2013

Prospects and challenges

The integration of such BKM requires the upgrading of computer tools. Their use for designing cereal food products requires (a) to integrate the available know-how and expertise for specific process operations where the use of models based on differential equations is still difficult, and (b) to extend the existing models to a domain of composition, in order to cope for the necessary increase in dietary fibre, that will modify greatly the rheological behaviour of dough. The use of such integrated models may be thought of to design products with target nutritional and sensory properties, provided their porosity and cellular structure are precisely characterized, and defined the pathways for reaching them, according to the so-called reverse engineering approach.

Authors

Guy Della Valle1, Hubert Chiron1, Martin Withworth2,

Kati Katina3, Kaisa Poutanen3

Organisation/Institute

1INRA, UR 1268 Biopolymères, Interactions & Assem-

blages (BIA), Nantes , France

2Campden BRI, Chipping Campden, Gloucestershire,

GL55 6LD, U.K.

3VTT Technical Research Centre of Finland, POB 1000,

FI 02044 VTT, Espoo, Finland

Contact

Guy della Valle, INRA

Guy.Della-valle@nantes.inra.fr

56



Model food applicability

ion ent

lidatva ssessm

It this information important for me?

Consumers

Food Industry

Regulatory Authorites

Scientific Community

pplicability a

ts a

WP-6

resulD

Workpackage



&R

57



Task 6.1

In WP6, model foods used for nutritional studies as

well for microbiological food safety and quality assess-

ment. It is important to ascertain that GMF are repro-

ducible and well characterized. In this task, protocols

for the realization of seven model foods have been

standardized and cross-validated among the partners

and tested for their practicability by industries.





Introduction of WP6


Task 6.2

Models were further characterized by using analytical

methods and high level structural studies. Main physi-

co-chemical features, e.g. firmness, color, bioactive

Angelo Visconti, Workpackage Leader

CNR-ISPA

molecule content have been assessed in WP2-models.

angelo.visconti@ispa.cnr.it

Protein concentration and oxidation surface hydro-

phobicity were assessed to characterize WP3-mimetic

model. Deep investigation characterized WP4-models

for viscosity, fat globule size distribution, protein

content/oxidation, while hardness, texture profile, and

Aims and background

color have been monitored to describe WP5-models.

Change in composition or process conditions may

Task 6.3

deeply impact the microbiological quality and safety of

Through the development of appropriate methods

foods, as well as the bioaccessibility of toxic or bioac-

and using the most recent analytical technologies,

tive compounds and therefore their absorption during

the in vitro and in vivo studies have been focused at

determine how different processes, modifying the

digestion. To assess their reproducibility and indus-

molecular and physical structure of GMF, can influence

trial practicability, Generic Model Foods (GMF) were

the bioavailability of selected nutrients or bioactive

standardized and validated. Their physico-chemical

compounds and their kinetics of appearance in blood

characterization can provide information on bioacces-

of minipigs during digestion of a complete meal com-

posed by GMF.

sibility of nutrients and/or toxicants, fundamental for

understanding the effect of different processes on the

Task 6.4

bioavailability - in vitro and in vivo studies - of se-

In T6.4, the impact of environmental conditions on

lected nutrients and their impact on consumer health.

microbial behavior or mycotoxin production have been

obtained by inoculating GMF with molecularly char-

Additionally, GMF can represent a powerful tool for

acterized spoilage, pathogenic or beneficial strains.

microbiologists. According to microbial behaviour of

The use of experimental kinetics performed during

beneficial, spoilage and pathogen populations, the use

the challenge test has allowed the determination of

of GMF can help in optimizing formulations, processes

growth/destruction parameters specific for each com-

bination microorganism/GMF. Mathematical models

and storage - finally improving risk assessment - via

have been then used to simulate growth for other

the implementation of available decision making tools.

conditions.

Main outcomes

Protocol standardization, cross-validation and phys-

Prospects

ical-chemical characterization were achieved for



Administration of different GMF in vivo as a complete

seven generic model foods - WP2-Brassica and Tomato

meal can highlight the effect of processes in chang-

purée; WP3-Mimetic meat model; WP4-Dairy dessert

ing the food structure and consequently nutrient/

and Soft cheese; WP5-Bread and Biscuit. The in vitro

bio-molecule bioavailability giving a realistic evalua-

digestion and in vivo studies highlighted the effect of the

tion of possible interactions of foods within the diges-

different processes in changing the physical and



tive tract. A specific task includes the implementa-

molecular structure of food and consequently the bio-

tion and consolidation of microbiological data deriving

accessibility and bioavailability of nutrients or bioactive

from GMF exploitation in the Sym’Previus-decision-

compounds. The quantification of the impact of process

making-tool. The predictive module will assist scien-

and storage on microbial food safety and quality was

tific community and industrials to i) select food model

obtained after inoculating GMF, fol owing the microbial

formulations and optimize processes and storage;

behaviour and implementing the decision making tool to

ii) determine the appropriate shelf-life according to

simulate their performance in food.

microbial behaviour; iii) assess quantitative risk.

58



Acting on micro- and macrostructure of

meat, cooking conditions can modify the

nutritional potential of meat proteins

How to cook meat for obtaining the best health effects?

In addition to the composition in amino acids and the digestibility, new criteria are appearing to fully describe the nutritional potential of proteins: the kinetics of amino acid absorption, and the potential to release bioactive peptides during digestion. For meat, these parameters can be modified by cooking conditions, which act on the micro- and macrostructure of the product. Drastic cooking conditions lead to protein aggregation, which slows enzymatic digestion of proteins. However, this effect, related to the microstructure, is limited in comparison to the effect of the macrostructure, and the chewing efficiency of the consumer. Furthermore, a simple difference in cooking conditions significantly modified the postprandial plasma peptide profile of the consumer, and therefore the potential health effect of meat.

Research aims and background

The classic criteria for evaluating the quality of a

rameters of the meat micro and macrostructure and

protein source are based on amino acid composi-

mastication remain difficult to establish.

tion, and protein digestibility in the digestive tract.

Furthermore, it has been shown that meat diges-

It is now known that these basic criteria are not suf-

tion reproducibly releases peptides containing

ficient to fully describe the nutritional potential of

amino acid sequences with antihypertensive activ-

a protein. For instance, it has been shown that the

ity (Bauchart et al., 2007a), and that a significant

rate of protein digestion can regulate postprandial

amount of carnosine is released in blood after a

protein retention (Dangin et al., 2003). Thus, the

meat meal (Bauchart et al., 2007b). This dipeptide

ranking of protein sources according to their rate

has numerous health benefits such as prevention of

of digestion is necessary. Additionally, total digest-

pathologies related to oxidative damage, or protein

ibility is not a good predictor of amino acid bioavail-

glycosylation. Nothing was known on the effect of

ability Indeed, only digestion in the small intestine

the meat structure (micro and macro) on the bio-

is thought to supply amino acids to the body. Finally,

availability of meat derived peptides.

all dietary proteins are potential sources of peptides

In this context we used in vitro and in vivo ap-

with beneficial health effects.

proaches (using minipigs and rats as animal models)

Previous work on meat shows that modifications at

to investigate the effect of structural modifications,

the microscopic scale of the protein cellular net-

on the bioavailability (kinetics and quantity) of ami-

work, for instance by protein aggregation during

no acids and peptides by manipulating meat cooking

heat treatments, can slow down the enzymatic di-

conditions.

gestion of proteins (Bax et al., 2013). The rate of

meat protein digestion can also be modulated by

chewing efficiency (Rémond et al., 2007), probably

in conjunction with meat macrostructure. However,

the hierarchy between the effects on digestion pa-

59





Significance and benefits

We have shown that cooking pork meat for 45 min at

90°C accelerates protein digestion, compared to a

10 min cooking at 75°C, but has no effect on the to-

tal bioavailability of amino acids (Figure 1). Although

these results underlined the importance of the

structure of the ingested meat in the determination

of amino acid absorption kinetics, they seem to be in

disagreement with a previous study with beef , ac-

cording to which meat protein digestion was slowed

down by increasing cooking temperature from 75°C

(30 min) to 95°C (30 min) (Bax et al., 2012). Beside a

possible meat origin effect, this suggests that, more

than the temperature, the time-temperature couple

Figure 1: Amino acid absorption. Portal net flux of the

is important in the determination of protein diges-

sum of indispensable amino acids (IAA), in minipigs



(n = 5) receiving two meals differing by the meat cooking

tion rate. Furthermore, in the beef study, meat was

conditions.

minced before meal serving, whereas in our study,

meat was only sliced. An interaction between cook-

ing conditions, and the efficiency of oral and gastric

meat degradation, could therefore also explain the

apparent discrepancy between the 2 studies. Indeed

in the present study, prolonged cooking at 90°C

produced a meat whose structure was much more

sensitive to mechanical degradation in the mouth,

leading to increased digestion rate.

Cooking conditions did not modify meat carnosine

content, and in both animal models (rats and mini-

pigs), the carnosine bioavailability was not affected

by cooking.

In order to address the question of peptide bioavail-

Figure 2: Peptidome. OSC-PLS score plot of plas-

ability, we have developed an analytical approach,

ma peptidome of minipigs fed with pork meat cooked



according two different conditions (d1 = 10 min, 75°C, d3

using an LTQ-Orbitrap Velos mass spectrometer and

= 45 min, 90°C)

data extraction with XCMS, for the characterization

of the plasma peptidome and the identification of

selected peptides. The method developed enabled

us to discriminate the plasma peptidome of the

minipigs fed with the pork meat cooked according

to the 2 conditions (Figure 2). We identified up to

33 peptides discriminating the cooking conditions.

The identification of these peptides was validated by

analysis of their fragmentation. Furthermore the ki-

netics analysis allowed us to show the postprandial

trajectory of the plasma peptidome (Figure 3).

Figure 3: Peptidome. OSC-PLS score showing plasma pep-

tidome trajectory during the postprandial kinetics after in-

gestion of meat, in minipigs.

60



Results and applications

References

Bauchart C, Morzel M, Chambon C, Patureau Mirand P,

Our work highlighted the difficulty of predict-

Reynès C, Buffière C, Rémond D, 2007a. Peptides repro-

ing protein digestion rate. Although it is clear that

ducibly released by in vivo digestion of beef meat and

protein aggregation slows down the accessibility of

trout flesh in pigs. Br J Nutr 98:1187-1195.

digestive enzymes to their cleavage site within the

Bauchart C, Savary-Auzeloux I, Patureau Mirand P, Thom-

proteins. This parameter appears to have minor



as E, Morzel M, Rémond D, 2007b. Carnosine concentra-

tion in ingested meat affects carnosine net release into

effect when meat is not ground. Indeed in this case, the

the portal vein of minipigs. J Nutr 137, 589-593.

resistance of the meat structure to disruption by

Bax ML, Buffière C, Hafnaoui N, Gaudichon C, Dardevet

chewing activity could be the main determinant

D, Santé-Lhoutellier V, Rémond D, 2013. Effects of meat

of the digestion speed. As it is attributable to the

cooking, and of ingested amount, on protein digestion

chewing efficiency of the consumer, it seems difficult

speed and entry of residual proteins into the colon: a

to predict it only from meat derived measurements

study in minipigs. PlosOne 8(4):e61252.

(chemical composition, structure characterization,

Dangin M, Guillet C, Garcia-Rodenas C, Gachon P,

and in vitro digestion). We have clearly evidenced

Bouteloup-Demange C et al. (2003) The rate of protein

digestion affects protein gain differently during aging in

that for a given meat, a simple difference in cook-

humans. J Physiol 549: 635-644.

ing conditions significantly modifies the postprandial

Rémond D, Machebeuf M, Yven C, Buffière C, Mioche L,

plasma peptidome of the consumer, and therefore

Mosoni L, Patureau Mirand P, 2007. Postprandial whole-

the potential health benefit of meat proteins.

body protein metabolism after a meat meal is influenced

by chewing efficiency in elderly subjects. Am J Clin Nutr

85, 1286-1292.

Prospects and challenges

This work is continued through the study, in human,

of the interaction between meat structure (cohesive-

ness and tenderness), and chewing activity, in order

to assess the final impact on the level of degradation

of the swallowed bolus, and the consequences on

protein digestion rate and the peptide release.

If we identify peptides that are reproducibly re-

leased from meat proteins, the next challenge will

be to identify their potential biological activity (an-

tihypertensive, immunomodulator, antioxidant, etc)

Authors

Didier Rémond1, Marie-Agnès Peyron1,

Estelle Pujos1, Jean-Louis Sébédio1,

Louise Salt3, Emöke Szerdahelyi2,

Andràs Nagy2, Eva Gelencser2,

Alan Mackie3

Organisation/Institute

1 UMR1019-UNH, INRA, Clermont-Fd/Theix, France

2 CFRI, Budapest, Hungary

3 IFR, Norwich, UK

Contact

Didier Rémond, UMR1019-UNH, INRA

didier.remond@clermont.inra.fr

61



Industrial concerns and needs towards

fungal risk assessment

How to assess and take into account the ability of Alternaria alternata mould to

grow and produce mycotoxin during tomato purée shelf-life

Alternaria species are reported to be the most commonly fungi affecting either tomato fruit and plant causing the so called black mould of tomato. Rapid infection of Alternaria in tomato may occur on the crop or post harvest yielding high economical loss due to spoilage of industrialized products such as tomato purée. Moreover, under specific growth conditions, Alternaria spp. may also produce various mycotoxins which represent a serious risk for human consumption of tomato-based products. Within the frame of this collaboration, the ability to grow and produce mycotoxins were determined as a function of tomato purée pH and temperature of storage. These boundaries and growth simulations will help food industrials to further optimize tomato-based food formulation and shelf-life.

Research aims and background

Predictive modelling and microbial risk assessment

appearance time on pastries, an adaptation of the

have emerged as a comprehensive and systematic

gamma concept of Zwietering (1992) was used to

approach for addressing the risk of microbial patho-

characterize fungal strain and to model its growth.

gens and spoilers in specific foods and processes.

In tomato-based products, Alternaria alternata rep-

Within WP6.4, food shelf-life or the impact of physi-

resents a relevant microbial hazard since it is re-

co-chemical factors (pH, aw, Temperature) on mi-

sponsible for black mould spoilage (Figure 1) yield-

crobial behavior were determined after (i) artificial

ing high economical losses (Bottalico & Logrieco,

inoculation of microorganism of interest in devel-

1992). In order to model the effect of environmen-

oped model food, (ii) fitting of experimental kinetics

tal factors on the development of mould on tomato

in food and (iii) prediction of shelf-life for various

medium, deep characterization of the strain was

scenarii in static or dynamic conditions of storage.

performed for a wide range of temperature and pH

Challenge tests in DREAM model foods were per-

values at a given water activity of 0.99. Moreover, to

formed according to guidelines and standardized

further screen the conditions yielding to mycotoxin

protocols based on the current NF V01-009 on

production, a similar experimental set up was used

guidelines to conduct microbial challenge tests. Ex-

and analysed.

periment fitting and shelf-life determination were

performed with recognized mathematical models

available in Sym’Previus decision making tool (www.

symprevius.org). While simulation with a large vari-

ety of characterized strains of pathogenic bacteria

is possible, few spoilage fungal strains are available

in the database. Similarly to the modelling approach

reported by Huchet et al. 2013 predicting mould

62





Results and applications

A toxigenic strain of Alternaria alternata (ITEM8176)

isolated from tomato fruit affected by black mould

(Somma et al., 2011) and deposited at the ISPA col-

lection, Italy (ITEM accession: http://www.ispa.cnr.

it/Collection/) was used for growth and mycotoxin

production assessment.

Growth ability and mycotoxin production of this

strain were determined on similar samples, after

Figure 1: Alternaria species are reported to be the most

inoculating fungal ascospores (7day-old culture)

commonly fungi affecting either tomato fruit and plant

causing the so called black mould of tomato

on WP2 cold break tomato purée supplement-

(Bottalico & Logrieco, 1992).

ed with agar. Adequate controls and a complete

factorial design with a total of 6 levels of pH (2, 3,

4, 5, 6 and 7) and 10 levels of temperature (6.5, 10,

15, 20, 25, 30, 35 and 37°C) were used (60 condi-

tions in total), for 3 replicates, to define which con-

ditions of pH and temperature supported (i) fungal

development and (ii) mycotoxin production. Growth

ability was determined on tomato-based medium by

regular observations of fungal development and ad-

ditional experimental data in broth were performed

to determine the strain cardinal values, i.e. minimal,

optimal and maximal values of pH and temperature

allowing growth. When plates were covered by the

fungus, mycotoxin quantification (tenuazonic acid,

TeA; alternariol, AOH and alternariol methyl ether,

Figure 2: Impact of pH and temperature on growth abil-

ity of Alternaria alternata ITEM8176 strain on tomato

AME) was performed by HPLC coupled with UV/DAD

based medium. For tested conditions, physico-chemical

conditions supporting growth are indicated in red (>90%

detection according to an adapted protocol from

growth probability). For conditions close to growth/no

Solfrizzo et al. 2004.

growth boundaries, the impact of annex flora, texture or

complex formulation might impact growth ability and ex-

After a maximum incubation time of 1 month, growth

perimental testing or challenge test is needed.

was observed in 35 conditions. A pH 3 was the lower

pH values enabling growth. Lower and higher val-

ues of temperature allowing growth were further

validated with experimental data. Among the in-

vestigated mycotoxins, only TeA was produced by

Alternaria alternata ITEM8176 in studied conditions.

The impact of pH and temperature on growth and

TeA production of the strain on tomato based me-

dium are indicated respectively in figure 2 and figure

3. Interestingly, growth optimum (24.5°C, pH5.5)

and mycotoxin optimum production (15°C, pH3.5)

occurred for opposite conditions suggesting a strong

impact of stress conditions on the strain virulence

that may represent a health issue for non adapted

Figure 3: Impact of pH and temperature on TeA myco-

conditions of storage.

toxin production of Alternaria alternata ITEM8176 strain

on tomato based medium. For tested conditions, dark red

dots refer to environmental conditions of temperature and

pH yielding higher concentrations of mycotoxin (15°C, pH

3-4)

63



Prospects and challenges

References

Bottalico A, Logrieco A. 1992. Alternaria plant diseases in

To our knowledge this is the first time that

Mediterranean countries and associated mycotoxins. In:

boundaries representing physico-chemical condi-

Chelkowski J, Visconti A, editor. Alternaria: biology plant

tions yielding growth and mycotoxin production of

diseases and metabolites. Amsterdam: Elsevier. 209-32.

Alternaria black mould spoilage were defined on

Huchet V, Pavan S, Lochardet A, Divanac’h ML, Postol-

tomato-based products. These results were



lec F, Thuault D. 2013. Development and application of a

predictive model of Aspergillus candidus growth as a tool

obtained thanks to a fruitful

to improve shelf-life of bakery products. Food Microbiol

collaboration with complementary expertise and

36(2):264-6

know how. These data are brand new and are



Solfrizzo M, De Girolamo A, Vitti C, Visconti A, van den

currently being disseminated towards public,



Bulk R. 2004. Liquid chromatographic determination of

scientific and food industrial audiencefor future

Alternaria toxins in carrots. JAOAC int. 87(1):101-6

successful applications.

Somma S, Pose G, Pardo A, Mulè G, Fernandez Pinto

V, Moretti A, Logrieco AF. 2011. AFLP variability, toxin

production, and pathogenicity of Alternaria species from

Argentinean tomato fruits and puree. Int J Food Microbiol

Significance and benefits

145(2-3):414–6

Alternaria alternata is the major microbial contami-

Zwietering M, Wijtzes T, de Wit J, Van’t Riet K. 1992. A

nant causing tomato black spoilage yielding high

decision support system for prediction of the microbial

economical losses. Thanks to these results the pre-

spoilage in foods. Journal of Food Protection 55(12):973-7

diction of spoilage development is now possible for

static or dynamic conditions of pH and temperature

on tomato medium. Future work could address (i)

validation for various recipe of tomato-based prod-

ucts, (ii) characterization of a larger number of

strains to take into account biodiversity in growth

prediction or (iii) further characterization of myco-

toxin synthesis to predict health issue for various

industrial relevant conditions.

Authors

Florence Postollec1, Véronique Huchet1,

Noémie Desriac1, Annalisa De Girolamo2,

Francesca Valerio2, Paola Lavermicocca2,

Angelo Visconti2, Danièle Sohier1

Organisation/Institute

1 ADRIA food technical Institute, Food safety and qual-

ity department, Quimper, France

2 CNR-ISPA, Institute of Sciences of Food Protection –

National Research Council, Bari, Italy

Contact

Florence Postollec, Food safety and quality

department, ADRIA Quimper

florence.postollec@adria.tm.fr

64





Technology transfer

gin ent

train ssm y needs

sse tr

It this information important for me?

esults a

Consumers

r

Food Industry

D

Regulatory Authorites

& rching indus

R

Scientific Community

resea

WP-7

Workpackage



65



Task 7.1

Collecting and mapping the needs of the industry -

Identification of the needs of the industry and barriers

to using modelling methods in 4 sectors (vegetable

and fruit processing; meat, fish and poultry process-

ing; dairy and bakery industry) and 3 countries based





Introduction of WP7


on interviews and focus group discussions (partners

involved: ACTALIA / France, CCFRA / United Kingdom,

CCHU / Hungary).

András Sebők, Workpackage Leader

Task 7.2

Campden BRI Magyarország Nonprofit Kft., (CCHU, Hungary)

Assessing the practicability of the generic food models

a.sebok@campdenkht.com

and their protocols -

The next step of WP7 was the testing of the practica-

bility of the models developed by the relevant WPs.

Tests was carried out by the WP partners (SOREDAB,

Prospects

ACTALIA, ADRIA, CCFRA, CCHU). After the adjustment

the improved models and protocols were evaluated for

The use of existing modelling approaches as time

practicality by involving industry partners

saving and cost effective tools for assessing and

optimising processes and their impact on product

Task 7.3

Development of industry guides -

quality and supporting decisions during the process

To support the practical aTpip

pli!cation of the models and

and product development activities is currently rela-

to provide an overview on different food models and

tively limited in the industry, especially at SMEs.

on modelling tools/software for the potential users a

practical guideline was developed by involving indus-

Technology transfer was integrated into the project

try partners including SMEs from each country. CCHU,

to promote that the model development is conduct-

SOREDAB and ACTALIA were responsible to consult

ed in collaboration with food manufacturers through

national industry partners during the development of

industry needs and feedback approach.

the guide.

To ensure that the practical needs of the indus-

try, especial y those of the SMEs are consid-

ered during the development of the real model

Main outcomes

foods; To convert results of the research into

As it was emphasized for the model developer WPs

appropriate format that can be used by the in-

by WP7, models developed within DREAM should

dustry as simple process and product devel-

consider the following aspects:

opment tools, and To test their applicability at

industry level and provide feedback for further

• Models should have practically applicable outputs,

improvement.

clearly described for the user.

• Models should be quick and as product specific as pos-

sible, easy to handle.

Aims and background

• Results should be as reliable and precise as possible.

According to the industry

• The modelling activity shall not require special equip-

• the economical advantages of using models should

ment, but purchasing necessary materials and software

be emphasised and clearly explained

is not a barrier. Companies could ensure a food engineer

for operating the model.

• models should be practically well applicable

• Confidentiality should be considered, regarding inputs.

• development of models should be focusing on food

safety and the effect of addition or removal of compo-

• The main barrier of using models by the industry is the

nents/ingredients to the product or applying new pro-

lack of knowledge, therefore a significant effort should

cess conditions on shelf life, on texture and on flavour.

be taken to remedy those failings.

Industry guide for Food Modelling:

Industry guide for Food Modelling:

The guideline is intended to give a technical aid to

After the consultation of the draft version with



the users but also to create awareness and encour-

national industry partners the final version was

age the use of modelling in the food industry.

discussed and adjusted.

66





Industry guide for Food Modelling

A guideline which is intended to give a technical aid to the users but also to

create awareness and encourage the use of modelling in the food industry

This guideline can be used as a manual, in which the potential user can find advice on questions related to the use of specific models and also for general considerations on the application and design of food models.

The content of the guideline, including the descriptions of the models, is based mainly on the models that were developed within the DREAM FP7 project and moreover on some models outside the

DREAM project that are actual y available and frequently used in the food industry. Brief descriptions of some general examples of successful practices, and hints for avoiding typical traps and

failures are summarised in the guideline.

Research aims and background

To support the practical application of realistic food

models a Practical Guideline on use of models was

developed. The main objective of the guideline is to

provide an overview on different food models and

on modelling tools/software for the potential users.

Models can be used in many different activi-

ties in the food sector considering the complex-

ity of the food and their different applications.

As changes in needs and requirements related to

food products arise increasingly quickly and fre-

quently, dissemination of available and effective

models to the food industry and also to all the

sectors who deal with food is therefore of high



importance.

The target audience of this guideline includes sev-

eral stakeholders of the food sector, particularly

representatives of the food industry (including

SMEs) and R&D teams, and decision makers on food

safety, quality and nutritional questions. It is also

cess development, for the assessment of the safety

recommended for food safety and regulatory bod-

of product/process design, and can help in under-

ies, nutritionists, food scientists and for marketing

standing the impact of process parameters on final

specialists who are particularly responsible for the

characteristics of the food and yield. However, their

industrial development of food companies.

use requires an appropriate level of expertise, com-

Food models are useful tools for product and pro-

petence, skills and clear practical guidance.

67



Significance and benefits

One of the main advantages of the realistic food models is to mimic the behaviour of real food products.

Furthermore, models can predict the impact of changes to the ingredients, compositions and process parameters. Thus, they can reduce the number of necessary experiments in real conditions, which is particularly important in the case of experiments in factory environments. The use of models can save time and can reduce costs. Standardised physical modelling materials and calculations with mathematical models provide a more reproducible benchmark for the impact of different treatments on food properties than experiments with real foods. Because of the rapidly changing conditions and demands of the market, for the food producers it is required to have a safe but fast process for development. As experiments under real conditions can be expensive and time-consuming, frequently there are significant limitations for carrying out a large number of experiments

under such conditions. In these cases,

food models and software models can

definitely be good tools to screen

options at low cost and to enable



experiments with real foods to



focus on the most promising test



parameters. By using models, waste

of the valuable real food product can

be significantly reduced during the ex-

Figure 1: The scheme of the food modelling process driven by the consumer periments.

/ industry needs

Results and applications

Prospects and challenges

Although there is a wide range of models having

However, finding satisfactory answers is a time con-

different scope, the model development process is

suming process. Models as time-saving and cost-

typically divided into five phases:

effective tools provide fairly good support when a

i) Defining the goal of the model: developing a state-

company decides to develop a new product: food

ment of purpose

models can help to reduce the time needed to pro-

ii) Designing and developing the model

vide an initial protocol for a production process, and

mathematical models can support the simulation of

iii) Practical testing and verification

different processes and the changes of the param-

iv) Making the model available for the audience

eters (see figure 1).

v) Maintenance of the model

This guideline describes these five phases as a sys-

tematic procedure and provides a brief description

Authors

of those steps that are essential to be considered

András Sebők1, Ágnes Gyuró1, Csaba Baár1,

Isabelle Gaucher2, Oline Rusten2,

during the model development. The most impor-

Jean-René Kerjean3

tant facilities and requirements for application and

operation of the models are discussed to help their

Organisation/Institute

1

use and to raise awareness among potential users.

Campden BRI Magyarország Nonprofit Kft

(CCHU), Hungary

The model descriptions are grouped by four major

2SOREDAB SAS (SOREDAB), France

generic structure groups representing vegetable,

3Institute for Dairy Products and Food Safety

dairy, meat and cereal products. Furthermore, there

(ACTALIA) France

is an additional group for models with general ap-



Contact

plicability, in which these models are discussed ac-

András Sebők, Campden BRI Magyarország

cording to their function, such as predictive micro-

Nonprofit Kft (CCHU), Hungary

bial models and heat treatment models.

a.sebok@campdenkht.com

68



Management &

y

Dissemination

It this information important for me?

communication

aphical identit

Consumers

Food Industry

Regulatory Authorites

Scientific Community

planning and reporting project gr

WP-0

Workpackage 8

&

69





Introduction of WP0


Monique Axelos, Workpackage Leader

INRA, France

monique.Axelos@nantes.inra.fr

Management tasks

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Caroline Sautot, DREAM project manager

The Management tasks were aimed at strategic

INRA Transfert, France

steering, project monitoring, administrative and

caroline.sautot@paris.inra.fr

financial management of the project.

Management

The management of DREAM met the – exciting – challenges posed by the project that:

i) adopts a bold scientific approach consisting of applying state - of - the - art modeling techniques that have been so far rarely used in food sciences;

ii) gathers a truly multi-disciplinary and multi-sectorial consortium made up of physicists, chemists, process engineers, microbiologists,

mathematicians, bioinformaticians, etc. in terms of expertises, and academia, technical centres;

iii) aims to provide concrete answers to an industrial sector facing somehow contradictory societal demands –

taste and tradition, safety and nutritiveness, product affordability and sustainable production – by searching solutions in fundamental science. The management needed to be:

i) sufficiently flexible to enable its work programme to remain relevant throughout external events, whether scientific, technological, or even societal;

ii) transparent and focused enough to make the consortium work together in conformity with the work plan.

These main requirements have been achieved by a three-tiered management system integrating strategic steering (including survey of the societal background and the scientific state of the art),

project monitoring (including the development of a collaborative platform, a management guideline and a risk management plan) and consortium management (including financial and administrative issues).

,

Authors

1Monique Axelos, 2Caroline Sautot

Organisation/Institute

1INRA, France,

2INRA Transfert, France



Contact

Monique Axelos, INRA, France

monique.Axelos@nantes.inra.fr

70



Task 8.1





Introduction of WP8


Dissemination to the industry and food authorities:

Trainings covered potential applications, the practical

benefits, the functions and operation of the models,

their limitations, explanation of the practical steps,

practices and hints for their use, warnings about

typical traps and malpractices. Trainings at national

Peter Raspor, Workpackage Leader

level on national languages created awareness and

UL-BF

induce a multiplication effect (e.g. spread of use) also

peter.raspor@bf.uni-lj.si

through development of the industry culture. A core

training package was developed in English, covering

each group of models.

Dissemination

Task 8.2

Dissemination to the scientific community:

As DREAM meets important industrial and



The publication in scientific journals and participation

societal issues while addressing fundamental



to scientific meetings as well as industrial fairs were

scientific issues, we played a particular atten-

found as good opportunity to disseminate DREAM

developments and results through oral presentation,

tion to technology transfer since the consortium is

poster or discussion. We performed organisation of

convinced that the project results are to improve

scientific workshops and sessions at EEFoST, IUFoST,

competitiveness of the European agrifood sector,

CEFooD meetings and relevant ETPs aiming at

which was one of the rationales of the project.

exchange of best practices.

As the project was designed to respond to some of

Task 8.3

the most important issues listed by the European

Training and career development:

Technology Platform on Food for Life, which express

For the efficient technology transfer it is important

that young researchers understand the needs of the

industrial expectations from scientific research,

industry, the way how the industry thinks and oper-

industrialists will certainly be interested in



ates, and the requirements of daily operation in the

appropriating the project outcomes.

industry in terms of process, quality and food safety

management, and also the specific priorities, condi-

Another prioritised target of the project is



tions, resources and constraints of SMEs. We provided

constituted by food authorities. Last but not least,

a practical training session for young participants at-

European citizens will certainly be interested by the

tached to project events focused to the last advances

societal stakes of the project as it responds to some

in food model development, food process, industrial

quality management.

of the “hottest” societal topics that are food safety,

nutrition patterns and food supply. For that reasons,

Task 8.4

it was important to disseminate outcomes as soon

Communication with the general public:

as possible to relevant stakeholders.

The objective of this task was to inform the general

public about the context and achievements of DREAM,

Our vision was:

in terms of the benefits for society. European citizens

-Make the best possible use of the project results by

were informed about the project values, with special

attention to communication to public through invita-

the food industry and concerned authorities.

tion of the media and through the DREAM website,

-Ensure fruitful exchange with the scientific

project popularized publications in brochures, news-

community, including individual scientists and



letters, lectures.

initiatives.

-Inform European citizens about the societal stakes of the project and the way their money is used.

The intention of implementation of new concepts and techniques, which had not been applied by the industry widelly, such as modeling was enhanced significantly by practical training sessions, workshops provided to industry and to science. However, the special treatment for SMEs was done on their mother tongue. This was particularly important to achieve good understanding in the area of comprehensive dissemination activities, which is not as easy as it looks from the beginning since we had to induce awareness for this issue first with the media editors and finally with involved parties.

71



Figure 1: Active communication among partners Meeting of the chairs with coordinator dr. Monique Axelos fourth from the right in Nantes where project started in 2009.

The concept pushed to our research under terms Generic Model Foods, Basic Knowledge Models and

Integrated Knowledge Models was very difficult to share with general public.

It looks very simple and strait, however when you ask even professionals about the details you face some barriers which can not be applied straightforward in daily practices and daily life. Let’s analyse this particular case more in detail.

The term Generic Model Foods (GMFs) defined as realistic physical models in which several parameters can be varied represent structured statement but it has to be explained clearly that this type of model has to be established on particular food sample of the kind and is substantially different from the other which might look in consumer eyes and taste very similar. Additional explanation about structure and chemical composition makes this issue even more complex and complicated not just for general consumer, but sometimes even for practitioners in food supply chain when they are asked to explain some details in simply and transparent language. Additional obstacle is in research community. All this case development is based on lingua franca English which is not mother tongue for majority of researchers in the project and consequently we adopt this thinking for English, when we switch back to mother tongue we convert this to mother tongues and

suddenly it get different emphasis or even different meaning. More scientifically complex statement has more possibilities to be transposed to other language with deformation, not just linguistically since researchers in this area does not posses this knowledge, but also professionally.

If we go to more simplistic Basic Knowledge Models or more sophisticated models like Integrated Knowledge Models this problem remains, actually It becomes even stronger. Just take a closer look to Basic Knowledge Model which is considered to be simpler since it is describing specific aspects of the model, through heuristic or mathematical approaches, what is not clear to average professions in food supply chain. For example description of the role played by temperature, pressure, chemical composition, etc. in formulation of food structure and (food) material properties is far to be simple and has many different implications in different foods.

72



Finally we addressed Integrated Knowledge Models the dynamic networks based on software systems

integrating the operating rules of Basic Knowledge Models, technical expert knowledge, food properties and food processing data from the Generic Model Foods. This complex approach clustered initial experiments and simulations to improve models to reveal key parameters and material behaviour and help refine, optimise and establish the food model concept prior to the pilot stage.

All this issues ware elaborated carefully and we addressed potential professionals to create awareness about this new concepts and also to convey this message to general public. This was achieved through scientific meetings we participated (ie.: IUFoST in Brasil in the session with lectures at global meetings like.: “Food safety consequences arising from the Design and development of REAlistic food Models with wellcharacterised micro- and macro- structure and composition (DREAM)« or as organizers of some activities within the programe of regional meetings like CEFood in Serbia. Beside this we transferred DREAM experiences and messages also students and teachers in Ljubljana at joint workshop with EU USA ToBeSafe project.

Moreover we approached general public also with contribution on world food day 16 of October 2012 at national event marking this day with DREAM contribution to profound knowledge on food models and food modeling. This is just to mention the main streams of our activities with dissemination in the project, which will continue via our web site and book of results for 5 additional years as planed.

Figure 2: Joint Workshop DREAM (FW7) & TU_BE_SAFE (ATLANTIS) Ljubljana 17 may 2011 gave opportunity to disseminate models to teachers and students in food area.

73





Figure 3: Novi Sad 2012 CEFood congress, Dream was disseminated to general food public in Central Europe at the meeting with 500 participants from academia and industry.

Figure 4: IUFOST (international union of food science and technology) gave possibility to DREAM to present their achievements in the area of modeling in Brasil 2012 to few thousand patricians from all over the world.

74





Figure 5: Marking importance of world food day 16th October 2012 Ljubljana (second from the left prof. dr. Peter Raspor, Responsible for dissemination in DREAM project)

Figure 6: The DREAM Project International Conference “From Model Foods to Food Models” held from 24th until 26th June 2013 in Nantes. At the event we had combination of dissemination activates: lectures, poster presentation, round table discussion and interviews.

75



Figure 7: The DREAM project last meeting held on 15th and 16th October 2013 in Bari, Italy. The DREAM project partners (upper image) and DREAM project WP leaders (bottom image) – from the left: Caroline Sautot (management WP0), Peter Raspor (leader of WP8), Kaisa Poutanen (leader of WP5), Matthijs Dekker (leader of WP2), Monique Axelos (coordinator), Alan Mackie (leader of WP4), Nathalie Perrot (leader of WP1), Carolina Realini (leader of WP3), Paola Lavermicocca representing Angelo Visconti (leader of WP6), András Sebők (leader of WP7).

Authors

Peter Raspor, Lidija Baša

Organisation/Institute

UL-BF Slovenia

Contact

Peter Raspor, UL-BF, peter.raspor@bf.uni-lj.si

76





Being curious?

Consumers clip for you.

Food Modelling

Model Foods

77

Food Modelling

Modelling in food science is used to predict/simulate an outcome regarding quality of food, production process or food shelf life. For example, if you are developing a new food product and you are missing certain information on what is going to happen with this food product in certain conditions (heating, etc) the food models can help you to find right answer to the question. What makes models to answer to this question? Food models are records of numerous laboratory experiments made by food experts and scientist and collated in a user friendly form that offers you to reuse them when needed.

Food modelling or food models reduce number of experiments; time, costs and provide process optimisation, improve process automation and control, due to results measured in previous experiments or production conditions. What has the DREAM project contributed to food modelling? The DREAM project used for its studies real food products, for example: dairy products, bakery products, meat, fruit and vegetables. The research demonstrated that mathematical models can be successfully adapted to complex food matrices (i.e.: food structure, composition) which are profoundly changed (i.e.: visually, nutritionally,) from entering substrate (i.e.: milk, flavour) to final product (i.e. Cheese, Bread).

Model Food

Model food is a material that consists of or contains essential body nutrients, such as carbohydrates, fat, proteins, vitamins and minerals in exactly known amount. Model food is used for mimicking real food products.

For example: the soft cheese model mimics real soft cheese characteristics. This kind of food models are used in research and development circumstances. Why to use food models rather than real food? Let’s say that you would like to produce soft cheese with lower salt content. In the early stage of the new product development you can use model food to run the experiments. What means that you may use soft cheese model to mimic the real soft cheese characteristics. By using the model food you may reduce time to get comprehensive data for designing real new food product. This is achieved by reduction of number of necessary experiments, food material, food waste, packing material and consequently cost. Can model food completely replace experiments on real food? No. The model food cannot replace the real food it can only speed up the development of new products with designed nutritional and food safety characteristics; like less sugar, salt, fat, etc. What is the DREAM project contribution to the model food development? One of the main contributions of the DREAM

project to the model food is linking mathematical modelling to the physical model foods from different foods (pork meat, soft and hard cheese, bread and biscuits, tomato and brassicas). Linking physical model foods with mathematical models gives good support tool for industry and for research.

by Peter Raspor, UL-BF & Lidija Baša, UL-BF

Food Safety

Food Quality

Process optimization

Food safety aims to assure

a high level of food safety,

Food quality refers to the

minimum technological, From a practical point

of view, we define the

animal health, animal welfare

nutritional and gastronomical

optimization as approach

and plant health within food

standards for substances to

which find the best solution

supply chain through coherent

qualify as fit for human con-

for running particular process

farm-to-table measures and

sumption

within constraints.

adequate monitoring.

by Peter Raspor, UL-BF

78

Conclusion

by Peter Raspor, UL-BF

Since DREAM kick off meeting in May 2009 we have been working deeply on research but also on innovation aspects with permanent cooperation between different stakeholders (institutions, external industrial partners). DREAM as trans-disciplinary partnership in nine countries focused on development of realistic, physical and mathematical, food models for use as standards to be exploited across all major food categories to facilitate development of common approaches to risk/benefit assessment and nutritional quality in food research and industry.

Since DREAM kick off meeting in May 2009 we

On dairy dessert models, the impact of the interfa-

have been working deeply on research but also on

cial composition and organization on the connectivi-

innovation aspects with permanent cooperation



ty between fat droplets and the microstructure of oil

etween different stakeholders (institutions,



in water emulsions were characterized depending on

external industrial partners). DREAM as trans-dis-

the protein type and concentration (casein micelle

ciplinary partnership in nine countries focused on

and whey protein) and heat treatments. The repro-

development of realistic, physical and mathemati-

ducibility of a soft-cheese model and a hard-cheese

cal, food models for use as standards to be ex-

model developed is 3 times better than the repro-

ploited across all major food categories to facilitate

ducibility of classical platforms; it allows reducing

development of common approaches to risk/benefit

the number of trials by 3/5 depending on the sig-

assessment and nutritional quality in food research

nificant difference to prove and the alpha and beta

and industry.

risks. The generic bread and biscuit models were

Through 4 years, we have been constantly



developed to study the influence of added dietary

improving our capacity and skills in food models

fiber on product quality and to develop mathemati-

design and application. The multidisciplinary



cal models to describe dough formation and prod-

approach adopted by partners in the project



uct quality. All these models were standardized and

integrated the diverse disciplines engaged in food

cross-validated and fully characterized in structure

science and engineering, mathematics, physics,

and composition. They were also used to assess nu-

chemistry, biochemistry, microbiology, process



tritional properties as well as microbiological food

engineering, nutrition, etc. The knowledge gained

safety. The in vitro digestion and in vivo studies

on relationships between structure/formulation

highlighted the effect of the different processes in

property functions and structure process functions

changing the physical and molecular structure of

allowed tailoring processes according to desired end

food and consequently the bioaccessibility and bio-

properties such as safety, nutritive value, and sen-

availability of nutrients or bioactive compounds.

sory qualities.

The model development was conducted in close

The project delivered several prototypes of food

collaboration with food manufacturers through a

models covering the four major food categories:

specification-based approach. All these models have

bread, cheese, meat and vegetables. On bras-

been implemented thanks to the mathematical ap-

sica vegetables, the effect of thermal processing

proach. This approach realized a complete dynamic

on colour, texture parameters and the effects of

description of food processing using an innovative

processing on bioconversion and bioaccessibility of

strategy exploiting most recent advances in cogni-

glucosinolates were modelled. On apple, a model

tive and complex system sciences to allow the gen-

to predict the transfer of procyanidin from fruit to

eralised methodologies to be extended to other food

juice depending on processing conditions has been

products. Starting from the project outcomes, the

developed. On meat, a mathematical stoichio-kinetic

industry is able to develop food matrices tailored to

model which assesses how much the kinetic param-

requested functions like controlled nutrient delivery,

eters are affected by initial product characteristics

bioactive compound release as mentioned before.

and cooking conditions, has been developed.

During the project, we had good attendance at the

79





different DREAM meetings organized in France, in Spain, in Slovenia, in Hungary and in Italy and we got a very good feedback from the press and relevant technological and scientific

community. Respecting this, we did not focus only on research but also

we introduced dissemination of our findings to relevant environments

such as: science, industry, consumers and regulators. Models and

protocols developed within DREAM were efficiently dissemi-

nated by expert partners via existing scientific and research

channels (CEFood, EFFoST, IUFoST congress) R&D channels

(ETP and national platforms, CIAA and national federations,

EFSA, national regulatory bodies) to be used by scientists,

SMEs and multinationals to improve nutritional quality and

benefit-risk management of the food supply networks.

Finally, not just research but also industry is now able to

develop food matrices tailored to requested functions like

controlled nutrient delivery, bioactive compound release…

In short we can summarise our achievements as follows:

A) Industrial achievements: decision-support systems allowing



realization of tailored microstructure in the food industry; protocols to

produce operating procedures to standardize model foods and practical guidelines on the use of models for industry.

B) Technological achievements: realistic and easily-reproducible in-laboratory food models encompassing large structure variations; optimised methods to characterise relevant structural, chemical and biological characteristics and follow their changes during processing; critical assessment of the relevance of the developed models to determine relationships between matrix microstructure, biological and chemical composition, and activity, delivery and transfer of bioactive compounds (nutrients, phytochemicals, toxins etc) during static (storage) and dynamic conditions (processing, in vivo degradation, etc.); and critical assessment of the relevance of the developed models to valuate the impact of environmental changes on the microbial population of real products and conversely, the effect of microbial populations on food structures.

C) Scientific achievements: mathematical models linking food formulation, processing, resulting matrix tructure and material properties; cognitive mapping of technical know-how involved in construction of food matrices; unified integrated knowledge description of each selected food matrix, and description of the relationships linking molecular and macroscopic structures to functionality.

80

The DREAM Project International Conference “From Model Foods to Food Models” held from 24th until 26th June 2013 in Nantes.



Perspectives

by Monique Axelos, INRA (DREAM coordinator)

During the last four years DREAM partners joined their efforts to develop models with in mind the idea that modelling will allow to mimic complexity and make progress in the process-properties relationship to be able to design tailor made food. Even though food processing is a field in which computer modelling is just emerging, the mathematical knowledge integration approach developed in the project appeared to be relevant and useful in giving interesting results as illustrated in this book. Three major issues must be pointed out.

The approach used in DREAM allows i) to integrate multidisciplinary skills, ii) to take into account a large number of parameters which could be integrated in decision support tools, iii) to gather heterogeneous and complex data and to face to the lack of data. Even though some progress has been done during the project further works must be undertaken. For the future, our findings suggest important avenues for further exploration. There are three that I will mention here.

Addressing issues on the col aboration between science and industry is of major importance to be sure that the models developed meet the real needs of the industry. For this purpose, interactions with stakeholders should play a key role and should be held sufficiently upstream of the model development in the laboratory so that opinions from both sides can genuinely be taken into account. Beside direct relationships among project partners our work showed that the main barrier of using models by the industry is the lack of knowledge of what a model could do. A significant effort should be taken to create awareness and encourage the use of modelling in the food industry especially for SMEs proving that models, as time-saving and cost effective tools, provide good support for innovation or for improving existing processes to reach new objectives. Work should continue on improving the industry guide and training tools that have been so far developed.

The second major avenue of interest is the use of this approach to answer complex, multi objectives new societal challenges like food sustainability. We know that the food system makes a significant contribution to the greenhouse gas emissions and gives rise to major environmental impact like biodiversity loss, water extraction and pol ution, energy and material waste. In order to cope with this challenge, technologies are required to enable the management of elements (such as energy, conservation, waste reduction and efficient use of capacity) but other non-technical data such as consumer acceptance, economic impact, regulation modification, etc… must be taken into account. Knowledge integration as developed in DREAM appears a very adequate and realistic approach to take into account a large number of criteria in a dynamic way and to provide new way of conceptualizing this challenge.

The third one deals with the increase of knowledge acquisition and integration in laboratory at the interface between mathematics, cognitive and bio-technical sciences. Even inside the academic world this approach is still poorly developed and implemented in daily

work practices. It is a real challenge for public in-

stitutions to create awareness and encourage sci-

entists to exchange at an early stage with math-

ematician to determine the lack of knowledge and

to build together new approach of modelling. Food

science has really much to gain to go along that

route. Respecting new knowledge and skills in this

area will contribute much to better quality, greater

food safety and more economical food production,

preservation and distribution.

81



The DREAM Project Consortium



Project Facts

Acronym

DREAM

Project Title



Design and development of REAlistic food Models with well-characterised mi-



cro- and macro-structure and composition

Cooperation Work

7th Framework Programme; Food, Agriculture and Fisheries, and

Programme



Biotechnology

Funding Scheme

Large Collaborative Project

Project Number

222654

Number of Partners

18



Budget

8.639.415 €





Acting Coordinator

INRA (Institut National de la Recherche Agronomique)



EC Contribution

5.995.786 €



Duration

48 months (from 1st of May 2009 until 30 April 2013)

Extension: 6 months (31 October 2013)

Table of Content

Foreword

Monique Axelos

6

Mathematical knowledge integration for food model numeric simulation (WP1)

7





Introduction of WP1


Nathalie Perrot


8

In silico comprehension and prediction of the structure and texture of a dairy dessert

Descamps Etienne, Foucquier Julie, Gaucel Sébastien, Lutton lyne, Mackie Alan, Perrot Nathalie,

Riaublanc Alain, Trelea Ioan Cristian

9

Reduction of complexity of an in-silico milk-gel model, using visualization and

optimization

Gaucel Sébastien, Lutton Evelyne, Perrot Nathalie, Riaublanc Alain, Tonda Alberto

12

Coupling viability theory and active learning with kd-trees

De Aldama Ricardo, Alvarez Isabelle, Martin Sophie, Rouquier J.B., Reuillon Romain, P-H. Wuillemin 14

An integrative modelling of quality and safety of vegetable during thermal treatment:

application to the cooking of Broccoli

Dekker Matthijs, Gaucel Sébastien, Perrot Nathalie, Postollec Florence, Sarvan Irmela,

Verkerk Ruud

16

Semi-supervised learning of a biscuit baking model, using symbolic regression and

Bayesian networks

Dekker Matthijs, Gaucel Sébastien, Perrot Nathalie, Postollec Florence, Sarvan Irmela, Verkerk Ruud 18

Filled cellular solid (WP2)

20





Introduction of WP2


Matthijs Dekker


21

Using mathematical modelling to optimise processing of fruits and vegetables with re-

spect to nutritional and sensory quality

Matthijs Dekker, Irmela Sarvan, Ruud Verkerk

22

Optimisation of mathematical model describing transfer of polyphenol from fruit to juice

by understanding non-covalent interaction between procyanidins and pectins

Aude A. Watrelot, Carine Le Bourvellec, Catherine M.G.C. Renard

25

Tomato processing method modify the bioaccessibility of its lycopene

David Page, Stephane George, Catherine Renard

28

Proteinous cellular network model (WP3)

30





Introduction of WP3


Carolina Realini


31

Meat models and mathematical modelling to investigate how heating affects protein

digestibility

Astruc Thierry, Danon Jeanne, Daudin Jean-Dominique, García-Regueiro José Antonio, Gatellier

Philippe, Gobert Mylène, Gou Pere, Kondjoyan Alain, Pérez-Juan María, Promeyrat Aurelie, Realini Carolina, Santé-Lhoutellier Véronique, Vénien Annie

33

Microstructure characterization of muscle tissue by quantitative imaging

Astruc Thierry, Bonny Jean-Marie, Clerjon Sylvie, Damez Jean-Louis, Labas Roland, Vénien Annie

35

Combined gelled / dispersed / aerated systems model (WP4)

39





Introduction of WP4


Alan Mackie


40

The Dairy dessert: A model for designing tailored interfaces in food systems

Claire Surel, Catherine Garnier, Alain Riaublanc, Marc Anton1, Alan Mackie

41

A pilot scale model for the reproducible production of soft cheese

Jean-René Kerjean, Romain Richoux, Alan Mackie

44

Coupled impact of product process and composition in lipids as emulsifiers on the

structure and the texture of a cream cheese model in relationship with product oxidation

Aliénor COUTOULY, Isabelle GAUCHER, Alain RIAUBLANC

46

Open solid foams model (WP5)

49





Introduction of WP5


Kaisa Poutanen


50

Structural challenges in nutritionally improved biscuits

Kaisa Poutanen, Nesli Sozer, Martin Withworth, Guy della Valle, Lucio

51

Creation of bread cellular structure

Guy Della Valle, Hubert Chiron, Martin Withworth, Kati Katina, Kaisa Poutanen

54

Model food applicability (WP6)

57





Introduction of WP6


Angelo Visconti


58

Acting on micro- and macrostructure of meat, cooking conditions can modify the

nutritional potential of meat proteins

Didier Rémond, Marie-Agnès Peyron, Estelle Pujos, Jean-Louis Sébédio, Louise Salt,

Emöke Szerdahelyi, Andràs Nagy, Eva Gelencser, Alan Mackie

59

Industrial concerns and needs towards fungal risk assessment

Florence Postollec, Véronique Huchet, Noémie Desriac, Annalisa De Girolamo, Francesca Valerio,

Paola Lavermicocca, Angelo Visconti, Danièle Sohier

62

Technology transfer (WP7)

65





Introduction of WP7


András Sebők

66

Industry guide for Food Modelling

András Sebők, Ágnes Gyuró, Csaba Baár, Isabelle Gaucher, Oline Rusten, Jean-René Kerjean

67

Management (WP0) & Dissemination (WP8)

69

Introduction of WP0 & WP8

70

Management

Monique Axelos, Caroline Sautot

70

Dissemination

Peter Raspor, Lidija Baša

71

Being curious? Consumers clip for you.

77

Food Modelling & Model Food

Peter Raspor, Lidija Baša

78

Food Safety & Food Quality & Process optimisation

Peter Raspor

78

Conclusion

Peter Raspor

79

Perspectives

Monique Axelos

81

The DREAM Project Consortium

83

CIP - Kataložni zapis o publikaciji

Narodna in univerzitetna knjižnica, Ljubljana

664.011(082)

The DREAM Project book of results [Elektronski vir] / [editorial board Monique Axelos ... et al.]. - El.

knjiga. - Ljubljana : Biotechnical Faculty, 2013

Način dostopa (URL): http://dream.aaeuropae.org/AboutDREAM/tabid/56/Default.aspx

ISBN 978-961-6379-26-7 (pdf)

1. Axelos, Monique A. V.

270799360