Séchage par atomisation des bactéries probiotiques : des

mécanismes de protection à la production à l’échelle

pilote
Song Huang

To cite this version:

Song Huang. Séchage par atomisation des bactéries probiotiques : des mécanismes de protection à
la production à l’échelle pilote. Alimentation et Nutrition. Agrocampus Ouest; Soochow university
(Taipei), 2017. Français. �NNT : 2017NSARB298�. �tel-01661632�

HAL Id: tel-01661632

[Link]
Submitted on 12 Dec 2017

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
 RÉSUMÉ ABSTRACT

Séchage par atomisation des bactéries probiotiques : des mécanismes de protection à la production à l’échelle pilote
Séchage par atomisation des bactéries probiotiques : des méca- Spray drying of probiotic bacteria: From molecular mechanism
nismes de protection à la production à l’échelle pilote to pilot-scale production
Les probiotiques sont des microorganismes vivants qui, ingérés Probiotics are live microorganisms that, when administered in
en quantité suffisante, exercent des effets positifs sur la santé. adequate amounts, confer a health benefit on the host. Freeze
La lyophilisation est aujourd’hui questionnée quant à sa consom- drying, the reference drying method, is currently challenged
mation d’énergie et son caractère discontinu. S’il offre une al- because of its low energy-efficiency and productivity. Therefore,
ternative pour produire massivement des poudres probiotiques spray drying is expected to be an alternative and sustainable
à faible coût, le séchage par atomisation induit quant à lui des method for producing probiotic powders. The issue remains
stress thermiques et oxydatifs conduisant à des pertes de viabi- in the considerable inactivation of probiotics caused by high
lité rédhibitoires. temperature and dehydration during the process.
Dans ce travail, un procédé innovant de séchage par atomisation In this work, a novel spray-drying process for continuous
est proposé. Du lactosérum doux concentré (jusqu’à 30% p/p) production of probiotics was challenged. Concentrated sweet
est utilisé à la fois comme support de culture et de séchage de whey (up to 30% w/w dry matter) was used to both culture
P. freudenreichii et L. casei. Ce procédé élimine les étapes inter- and spray dry P. freudenreichii ITG P20 and L. casei BL23. This
médiaires à risque de contamination élevé, accroît la biomasse et process cuts down the steps between culturing and drying (e.g.
améliore la viabilité des bactéries. harvesting, washing), increases the cell population after growth
Les mécanismes sous-jacents ont été explorés au plan de la and improves spray drying productivity and probiotic viability.
résistance bactérienne et des conditions de séchage. Le milieu The mechanisms were explored from bacterial physiology and drying
concentré induit une osmoadaptation des bactéries par expres- process conditions. The hypertonic stress led to overexpression of
sion de protéines de stress et accumulation de solutés compa- key stress proteins and accumulation of intracellular compatible
tibles, conduisant à une tolérance accrue des probiotiques à solutes, which enhanced multistress tolerance. The presence of
différents stress. La présence d’agrégats et la concentration en protein aggregates and optimal concentration of Mg2+ in matrix
Mg2+ du milieu concentré pourraient également être impliquées. may also be involved.
Le scale-up du procédé a été étudié : un schéma technologique The feasibility of scaling up this process was validated. A multi-
semi-industriel impliquant séchage par atomisation, sur bande stage semi industrial drying process, coupling spray-drying with
et en lit fluidisé a permis d’atteindre une viabilité de 100% belt drying and fluid-bed drying, was applied to further improve
(> 109 CFU g-1). Par ailleurs, la fonctionnalité des poudres probio- the probiotic viability to 100% (> 109 CFU g-1). Moreover, the
tiques a été évaluée in vitro and in vivo sur modèle porcelet. Ce functionality of these probiotic powders was investigated in vitro
travail ouvre de nouvelles perspectives pour l’obtention à grande and in vivo with piglet model. This work opens new avenues
échelle et selon un procédé sobre et durable de préparations pro- for the sustainable development of new starter and probiotic Song HUANG • 30 mai 2017
biotiques. preparations with enhanced robustness.
Thèse AGROCAMPUS OUEST ÉCOLE DOCTORALE • Vie - Agro - Santé (VAS)
Mots-clés : probiotiques ; séchage par atomisation ; osmorégula- Keywords: Probiotics; spray drying; osmoregulation; stress sous le label de l’Université Bretagne Loire LABORATOIRE D’ACCUEIL • UMR INRA - AGROCAMPUS OUEST
tion ; tolérance ; viabilité tolerance; viability pour obtenir le grade de
Science et Technologie du Lait et de l’Œuf (STLO)
DOCTEUR D’AGROCAMPUS OUEST
Spécialité Science de l’aliment THÈSE EN CO-TUTELLE • School of Chemical and Environmental
Engineering, Soochow University, Suzhou

Séchage par atomisation MURIEL THOMAS

AGROCAMPUS OUEST • Institut supérieur des sciences agronomiques, Thèse B-298 — 2017-15 • HUANG Song Directrice de Recherche, INRA Micalis, France / présidente
agroalimentaires, horticoles et du paysage des bactéries probiotiques : CATHERINE STANTON
Professeure, University College Cork, Irlande / rapporteure
65 rue de Saint-Brieuc – CS84215 – F-35042 Rennes Cedex
Tél. : 02 23 48 50 00
[Link]
des mécanismes de protection à la CORDELIA SELOMULYA
Professeure, Monash University, Australie / rapporteure
production à l’échelle pilote MAARTEN SCHUTYSER
Professeur, Wageningen University, Pays-Bas / examinateur
XIAO DONG CHEN
Professeur, Soochow University, Chine / directeur de thèse
ROMAIN JEANTET
Professeur AGROCAMPUS OUEST, UMR INRA-AO STLO,
France / directeur de thèse
GWENAËL JAN
Chargé de recherche, UMR INRA-AO STLO, France /
co-directeur de thèse
 THESE / AGROCAMPUS OUEST
Sous le label de l’Université Européenne de Bretagne
pour obtenir le diplôme de :
DOCTEUR DE L'INSTITUT SUPERIEUR DES SCIENCES AGRONOMIQUES,
AGRO-ALIMENTAIRES, HORTICOLES ET DU PAYSAGE
Spécialité : Science de l'aliment
Doctoral College: VAS (Vie-Agro-Santé)

présentée par :
Song HUANG

Séchage par atomisation des bactéries probiotiques:

du mécanisme moléculaire à la production à l'échelle pilote
(Spray drying of probiotics:
From molecular mechanism to pilot-scale production)

Thèse préparée à l’UMR Sciences et Technologies du Lait et l’Œuf,

INRA-Agrocampus Ouest, Rennes
et
School of Chemical and Environmental Engineering,
Soochow University, Suzhou

Soutenue le 30 May 2017 devant la commission d’examen

Composition du jury :
Catherine Stanton, Professeure, University College Cork, Irlande Rapporteur
Cordelia Selomulya, Professeure, Monash University, Australie Rapporteur
Muriel Thomas, Directrice de Recherche, INRA Micalis, France Examinatrice
Maarten Schutyser, Professeur, Wageningen University, Pays-Bas Examinateur
Xiao Dong Chen, Professeur, Soochow University, Chine Directeur de thèse
Romain Jeantet, Professeur, Agrocampus Ouest, France Directeur de thèse
Gwenaël Jan, Scientific Director, INRA (STLO), France Directeur de thèse
This page is intentionally left blank.
 Foreword

欲穷千里目，更上一层楼。
- 王之涣 《登鹳雀楼》
Désirant embrasser toute l’étendue de ce paysage,
je gravis un étage supplémentaire de la tour.
De Zhihuan WANG (688-742)
 Acknowledgement I

Acknowledgement
Three years have gone by since the first time I arrived Rennes in 2014. France (Bretagne)
is the first country (region) I reach abroad, and Rennes is the first city I live out of China. It
seems not so far away from the first time I saw the amazingly huge green farms and clean blue
sky here. As the first and only child who got the “college-level” education in my big family, it
is hard to imagine this experience of doing a PhD in Europe. I am very grateful to this joint
PhD project supported by Soochow University and Agrocampus Ouest. I also thank the joint
research unit (STLO) between INRA and Agrocampus Ouest, and the Suzhou Key Lab of
Green Chemical Engineering for providing the world-class laboratories.

I would like to firstly thank Prof. Xiao Dong Chen, who teaches and helps me so much
during the 7 years as my supervisor and mentor. Because of you, I fell in love with scientific
research. You brought me to this “new world”, and because of this, everything happens now.
Thank you so much for changing my life and making my dream of studying oversea comes
true.

Prof. Romain Jeantet, my dear “father” in France. Thank you so much for taking care of me
in both study and life here. I appreciate this experience of working with you here, and the
memories at beautiful Plonéour and also in China. I admire and also appreciate the great
friendship between you and Dong. Because of it, I had this opportunity to come in France, to
see this big big world. Thank you so much for everything you did for my PhD project.

Dr. Gwénaël Jan, the super “Yoda”. I would like to thank you for your great supervison,
showing me (your young padawan) the magic of microbiology and molecular world. I admire
your enthusiasm in not only research, but also dancing and diving. Besides, I am very grateful
for the super “close” friendship we built in Suzhou, Beijing and Dublin.

During the time in STLO, I got great help and advice from Yves (Le Roi) and Pierre Schuck.
I am grateful for your participation in my PhD project. It makes me proud to announce my
super big “supervisor team” everywhere.

I would like to give my sincere gratitude to Isabelle Luron and Françoise Prigent for helping
the progress of my PhD project. I also appreciate the opportunities to learn from Julien Jardin
(my idol), Chantal Cauty (my part-time french teacher), Valérie Gagnaire, Gaelle Boudry and
Serge Mejean (drying master). I also sincerely thank the excellent technical support from
Paulette and Jessica, thank all my colleagues in Micro team and Aurélie, Nathalie and
 Acknowledgement II

Christophe from Standa for teaching and helping me when working in microbiology lab, thank
Anne Dolivet, Eric Beaucher, Gwénolé Le Maout for their help in powder analysis and Bionov
drying, thank Michel Ravailler, Laurent Fromont, Benoît Robert, Gilles Garric for their help in
the Dairy Platform, thank Laurence Fauvel for helping me to solve accommodation problems
(twice) in Rennes, thank Hui Zhang and Junxia Long for kind help and support in the lab in
Suzhou, thank Marie-Bernadette, Sandrine, Marie-Noëlle, Victoria, Fabienne Garnier-
Lambrouin, Pascaline Hamon, for their help in my experiments, and also thank Anne Giboulot,
Dominique Volland, Christophe Geneste, Nathalie Le Marre, Danielle Guilloux for their help
in the STLO lab.

As a Chinese student with zero French, honestly speaking, it is difficult at the beginning to
understand everything and enjoy life here. So I am very grateful for all the friends I made
during these three years. Because of you, I got great fun and wonderful memories in France. It
is touched everytime when I think about these warm moments: The first dinner in Rennes with
Peng Zhu and his family, and also the great help from him on my settle-down in Rennes; The
first party invitation from Lelia and Alran in the first weekend here, where I first time
recognized the “Samba killer” Gui and so-called “Rennes top-1 handsome guy” Carlos; The
first time searching accommodation and reviewing all kinds of apartments under the help of
Samira (also the endless help after moving in the same office); The first time visiting in Saint-
Malo and Mont Saint-Michael (both in rainy days) with my dear sister, “rain bringer”,
Oumaima (still owing me a couscous…); The first time falling in love with my “second french
home” E. Leclerc Cleunay with Meichun, Caiyun, Mabelyn and Ning; The first time travel in
Paris with Lelia, Elise (“saucisson queen”, my sister in S(ch)ong’s family), Guillaume, Andreza
and “Benzema”; The first birthday abroad with so-called “italian birthday cake” (actually a
pizza) from Natayme; The first time doing animal experiment (unforgettable in my life) with
“cheese queen” Houem; The first time seeing a cute European new-born baby, the daughter of
Jonathan (my idol); The first time eating snails with “my lord” Fellipe (who knows nothing);
The first meeting with French “president” Floriane (à gauche); The first “foodie” team
established for Marché des Lices with Ying Yao and Yue Li; The first time of ‘FRAZINA’
(see Figure without number) travel out of France in Berlin, and then further exploration in
Russia… There are so many nice memories with all of my friends here, Marilia, Ilham,
Federico, Adèle, Perrine, Juliana, Andreas, Xaxa, Sanda, Mathieu, Linda (Mother of dragons),
Lucie, Alexia (choupinou), Eve-Anne, Celine, Vincent, Coralie, Fanny, Fanny, Fanny... I am
happy to meet you here. I’m sorry for those friends who I couldn’t remember your names. Your
 Acknowledgement III

names are indeed new and complicated vocabularies for me. I am also sorry for those friends
who never talked to me because of the language problem. You have to imagine my humor, or
learn some Chinese...

最后，我想将最深的感激献给我的家人。在那个教育资源匮乏的家乡小镇，我的父
母在我身上付出难以想象的心血。你们竭尽所能地为我提供良好的成长与教育环境。
没有你们的鼓励与支持，我不会有机会走出那个偏远的家乡小镇。我也非常感谢我的
爱人小芬同志。我们相识相知已有八年，这八年里你总能在我最低落的时候鼓励我。
你的善良与阳光让我对生活，对未来都充满了期待。我很抱歉大部分的时间我都没能
在你身边照顾到你，这么多年让你独自一人对抗生活和工作的压力。非常感谢你对我
的理解与支持。我很开心很快我们就将相聚。

Figure without number. The logo of the office of Elise, Samira and Song, 2016-2017. “FRAZINA”
symbolizes the friendship among France, Brazil and China.
 Thesis Outputs IV

Thesis Outputs
Review article
Huang, S., Vignolles, M., Chen, X.D., Le Loir, Y., Jan, G., Schuck, P., and Jeantet, R., 2017.
Spray drying of probiotics and other food-grade bacteria: A review. Trends in Food
Sci. Technol. 63, 1-17.

Published or submitted research articles

Huang, S., Cauty, C., Dolivet, A., Le Loir, Y., Chen, X.D., Schuck, P., Jan, G., Jeantet, R.,
2016a. Double use of highly concentrated sweet whey to improve the biomass
production and viability of spray-dried probiotic bacteria. J. Funct. Foods 23, 453–
463.
Huang, S., Méjean, S., Rabah, H., Dolivet, A., Le Loir, Y., Chen, X.D., Jan, G., Jeantet, R.,
Schuck, P., 2017. Double use of concentrated sweet whey for growth and spray
drying of probiotics: Towards maximal viability in pilot scale spray dryer. J. Food
Eng. 196, 11–17.
Huang, S., Rabah, H., Jardin, J., Briard-Bion, V., Parayre, S., Maillard, M.-B., Loir, Y.L.,
Chen, X.D., Schuck, P., Jeantet, R., Jan, G., 2016b. Hyperconcentrated sweet whey, a
new culture medium that enhances Propionibacterium freudenreichii stress tolerance.
Appl. Environ. Microbiol. 82, 4641–4651.
Wang, J., Huang, S.*, Fu, N., Jeantet, R., Chen, X.D.*, 2016. Thermal aggregation of
calcium-fortified skim milk enhances probiotic protection during convective droplet
drying. J. Agric. Food Chem. 64, 6003–6010.
Yang, Y., Huang, S.*, Wang, J., Jan, G., Jeantet, R., Chen, X.D.*, 2017. Mg2+ improves the
thermotolerance of probiotic Lactobacillus rhamnosus GG, Lactobacillus casei Zhang
and Lactobacillus plantarum P-8. Lett. Appl. Microbiol. 64, 283-288.
Ferreira, A.A., Huang, S., Perrone, Í.T., Schuck, P., Jan, G., Carvalho, A.F., Tracking
Amazonian cheese microbial diversity: Development of an original, sustainable and
robust starter by spray-drying/freeze-drying. J. Dairy Sci. (Minor revision).
Zheng, X., Fu, N., Huang, S., Jeantet, R., Chen, X.D., 2016. Exploring the protective effects
of calcium-containing carrier against drying-induced cellular injuries of probiotics
using single droplet drying technique. Food Res. Int. 90, 226-234
 Thesis Outputs V

Rosa do Carmo, F.L., Rabah, H., Huang, S., Gaucher, F., Deplanche, M., Parayre, S.,
Dutertre, S., Jardin, J., Le Loir, Y., Azevedo, V., Jan, G., Propionibacterium
freudenreichii surface protein SlpB is involved in adhesion to intestinal HT-29 cells.
Front. Microbiol. (Under review)

Patents
Method for preparing a probiotic powder using a two-in-one whey-containing nutrient
medium. Patent application filed in Europe on 21 September 2015, EP no. 15
306465.4. Inventors: Jeantet, R., Huang, S., Jan, G., Schuck, P., Le Loir, Y., Chen,
X.D.

Papers in Preparation
Huang, S., et al., Concentrated sweet whey induces multistress tolerance of Lactobacillus
casei BL23.
Huang, S., et al., Long term survival of Lactobacillus casei BL23 in sweet whey medium.
Huang, S., et al., Probiotic efficacy of spray-dried Propionibacterium freudenreichii in piglet
model.

Oral Presentations
Huang, S., Le Loir, Y., Chen, X. D., Jan G., Schuck, P., & Jeantet, R. Armed probiotics with
2-in-1 medium: The whey is a way. The 2nd CHE-SIS, 11-13 October 2015, Suzhou,
China.
Huang, S., Cauty, C., Dolivet, A., Mejean, S., Le Loir, Y., Chen, X. D., Schuck, P., Jan G. &
Jeantet, R. From growth to spray drying: two-in-one use of sweet whey to improve the
biomass production and spray drying viability of probiotics. IDF Concentration and
Dried Milk Products Symposium, 11-13 April 2016, Dublin, Ireland.
Huang, S., Schuck, P., Jeantet, R., Jan, G. & Chen, X. D. Major breakthrough in probiotic
production: The two-in-one use of sweet whey affords yet unknown probiotic
viability and stability. International Scientific Conference on Probiotics and Prebiotics
- IPC2016, 21-13 June 2016, Budapest, Hungary.
Jeantet R, Huang S, Chen XD, Jan G. 2017. Smart drying of probiotics: from molecular
mechanisms to pilot scale production. Eurodrying’17, June 19th- 21st, Liège,
Belgium.
 Thesis Outputs VI

Jeantet R, Huang S, Jan G, Le Floch-Fouéré C, Chen XD. 2017. Drying research to overcome
dairy industry challenges: from single drop to drying of probiotics. 9th Asian Drying
Conference 2017, September 24th-26th, Wuxi, China.
Poster Presentations
Huang, S., Rabah, H., Jardin, J., Briard-Bion, V., Parayre, S., Maillard, M.-B., Le Loir, Y.,
Chen, X. D., Jan G., Schuck, P., Jeantet, R., & Jan, G. Double use of
hyperconcentrated sweet whey: a novel strategy for production of Propionibacterium
freudenreichii. The 4th International Symposium on Propionibacteria and
Bifidobacteria, 20-23 September 2016, Cork, Ireland (Best poster prize)
Huang, S., Jan, G., Le Loir, Y., Schuck, P., Jeantet, R., & Chen, X. D. A novel strategy to
increase energy efficiency and productivity for probiotic production: Combination of
molecular adaptation, protectant application and processing optimization. The 5th
International Congress on Sustainability Science & Engineering, 24-27 October 2016,
Suzhou, China
Chen, X. D., Huang, S., Jeantet, R., Wu, P., Liu, M., Chen, L., Zhang, X., Liao, Z.
Progressive construction of more realistic species specific in vitro digestion bio-
mechatronic models. The 5th International Conference on Food Digestion, 04-06
April 2017, Rennes, France

Supervision
Lab-supervision of two master students (Yi Yang, Xiamen University, from
01/06/2014 to 01/06/2015; Juan Wang, Soochow University, from 01/06/2014 to
01/06/2017)
 Table of Contents VII

Table of Contents
Acknowledgement .................................................................................................................................. I
Thesis Outputs...................................................................................................................................... IV
Table of Contents ................................................................................................................................ VII
List of Abbreviations ............................................................................................................................ X
Abstract.................................................................................................................................................. 1
General Introduction ............................................................................................................................ 3
Scientific and Economic Context of Probiotics .................................................................................. 3
Spray Drying of Probiotics: New Opportunities and Challenges ....................................................... 6
Chapter 1. Literature Review......................................................................................................... 10
1.1 Spray drying of probiotics: pre-drying stage.................................................................. 11
1.1.1 Selection of bacterial Strains ......................................................................................... 11
1.1.2 Response to cellular stress ............................................................................................. 18
1.1.3 Growth conditions and preparation of feed concentrate ............................................... 21
1.2 Spray drying probiotics: During drying ......................................................................... 25
1.2.1 Bacteria may be injured during spray drying ................................................................ 25
1.2.2 Drying Devices ............................................................................................................... 27
1.2.3 Protective agents ............................................................................................................ 29
1.2.4 Drying parameters ......................................................................................................... 33
1.2.5 “In-process” investigations ........................................................................................... 35
1.3 Spray drying probiotics: Post-drying stage .................................................................... 38
1.3.1. Storage stability of bacteria in powder form ................................................................. 38
1.3.2. Applications of dried bacteria........................................................................................ 42
1.4 Context, aim and innovation of the PhD Project ............................................................ 48
1.4.1 Context of the PhD project ............................................................................................. 48
1.4.2 Aim of the PhD project ................................................................................................... 49
1.4.3 Innovation of the PhD project ........................................................................................ 50
Chapter 2. A novel 2-in-1 Process: Double Use of Highly Concentrated Sweet Whey ............. 51
2.1 Introduction .................................................................................................................... 52
2.2 Materials and Methods ................................................................................................... 57
2.2.1 Strains and culture conditions ....................................................................................... 57
2.2.2 Growth in sweet whey culture media ............................................................................. 57
 Table of Contents VIII

2.2.3 Monitoring of bacterial growth and acidification curves .............................................. 58

2.2.4 Spray drying ................................................................................................................... 59
2.2.5 Size distribution measurement ....................................................................................... 59
2.2.6 Scanning electron microscopy ....................................................................................... 60
2.2.7 Storage ........................................................................................................................... 60
2.2.8 Enumeration and quantification of bacteria viability .................................................... 60
2.2.9 Statistical analysis.......................................................................................................... 61
2.3 Results and Discussion................................................................................................... 62
2.3.1 Effect of sweet whey concentration on bacterial biomass production ........................... 62
2.3.2 Long-term survival of [Link] BL23 during growth in sweet whey media ..................... 65
2.3.3 Improved probiotic viability in spray drying application .............................................. 67
2.3.4 Improved storage stability of probiotic powders ........................................................... 73
2.3.5 Additional results ........................................................................................................... 75
2.4 Conclusion ..................................................................................................................... 76
Chapter 3. What are the Underlying mechanisms of the enhanced Protection on Probiotics?77
3.1 Introduction .................................................................................................................... 78
3.2 Materials and Methods ................................................................................................... 81
3.2.1 Strains, culture conditions and bacteria-milk matrix preparation................................. 81
3.2.2 Stress challenges ............................................................................................................ 81
3.2.3 Physico- physicochemical properties of bacteria-milk matrix and particle .................. 84
3.2.4 P. freudenreichii proteomic analysis and indentification .............................................. 85
3.2.5 P. freudenreichii cellular compatible solutes ................................................................ 87
3.2.6 Statistical Analysis ......................................................................................................... 89
3.3 Results and Discussion................................................................................................... 90
3.3.1 Osmotic adaptation induces multistress tolerance of P. freudenreichii ........................ 90
3.3.2 Calcium-protein aggregates protect probiotics against droplet drying ...................... 100
3.3.3 Contribution of minerals: a preliminary exploration with magnesium effect .............. 111
3.4 Conclusion ................................................................................................................... 118
Chapter 4. Scale-up of the 2-in-1 Process in Semi-Industrial Scale Spray Drying ................. 119
4.1 Introduction .................................................................................................................. 120
4.2 Materials and Methods ................................................................................................. 123
4.2.1 Bacterial strains and starter preparation .................................................................... 123
4.2.2 Fermentation with sweet whey ..................................................................................... 123
4.2.3 Spray drying ................................................................................................................. 123
 Table of Contents IX

4.2.4 Analysis of physical and chemical properties .............................................................. 125

4.2.5 Storage of powders....................................................................................................... 125
4.2.6 In vitro simulated digestion .......................................................................................... 125
4.2.7 Enumeration and quantification of bacteria ................................................................ 126
4.2.8 Statistical analysis........................................................................................................ 126
4.3 Results and Discussion................................................................................................. 127
4.3.1 Validation the feasibility of scaling up the 2-in-1 process ........................................... 127
4.3.2 Improvement of probiotic viability through multi-stage drying process ...................... 129
4.3.3 Storage stability of probiotic powders ......................................................................... 131
4.3.4 Resistance of rehydrated probiotics to simulated digestion......................................... 134
4.4 Conclusion ................................................................................................................... 136
General Conclusion........................................................................................................................... 137
Future Perspective ............................................................................................................................ 140
Reference ........................................................................................................................................... 146
List of Figures.................................................................................................................................... 165
List of Tables ..................................................................................................................................... 169
 List of abbreviations X

List of Abbreviations
CFU – Colony forming unit

CAGR – Compound annual growth rate

GA – Gum Arabic

GIT – Gastricintestinal tracts

LAB – Lactic acid bacteria

LCBL23 – Lactobacillus casei BL23

LCZ – Lactobacillus casei Zhang

LGG – Lactobacillus rhamnosus GG

LP – Lactobacillus plantarum P-8

Md – Droplet mass

MRS – de Man, Rogosa and Sharpe broth

PF129 – Propionibacterium freudenreichii CIRM-BIA129

PSD – Particle size distribution

RSM – Reconstituted skim milk

SGF – Simulated gastric fluid

SIF – Simulated intestinal fluid

SW – Sweet whey

TS – Total solid content

T2 – Transverse relaxation time

Td – Droplet temperature

Tg – Glass transition temperature

WPI - Whey protein isolate

YEL - Yeast Extract Sodium Lactate broth

SCFAS - Short chain fatty acids

 Abstract 1

Abstract

Probiotics are defined by FAO/WHO as ‘live microorganisms that, when administered in

adequate amounts, confer a health benefit on the host’. Having the dried, preserved and ready-
to-use form of probiotics offers a wider range of applications for a substantially longer period
of time as compared to its unpreserved liquid or slurry form. Probiotic ingredients market was
valued at 33.6 Billion USD in 2015, would reach 64.4 Billion USD by 2020. Freeze drying is
currently a popular drying method to produce probiotic products due to its reliability in term
of cell viability. However, its high production-cost, low energy-efficiency and productivity
have become a stumbling block with regard to the increasing market demand. Spray drying is
expected to be an alternative method for production of probiotic powders. The low production
cost and high productivity are highly desirable by industry. The challenge consists in the
considerable inactivation of probiotics caused by high temperature and dehydration during the
process.

In this thesis, we present a novel spray drying process for continuous production of
probiotics. Sweet whey is used as 2-in-1 highly concentrated medium (up to 30% w/w dry
matter) for both culturing and spray drying of probiotics. This process simplifies the operation
units between culturing and drying (e.g. harvesting, washing, and re-suspending steps),
increases the cell population yield during growth. It improves probiotic viability and
productivity during spray drying and probiotic stability during storage.

The mechanisms were explored from bacterial stress resistance (intrinsic) and drying
process conditions (extrinsic). The hypertonic stress led to overexpression of key stress proteins,
accumulation of intracellular compatible solutes, which contributed to the enhanced multistress
tolerance acquisition. The presence of protein aggregates and increased concentration of Mg2+
in the drying matrix may also be responsible for the enhanced probiotic protection in highly
concentrated sweet whey.

The feasibility of scaling up this process was validated with a semi industrial pilot scale
spray dryer. A multi-stage mild-conditions drying process, coupling spray drying with belt
drying and fluid-bed drying, was applied to further improve the probiotic viability to
 Table of Contents 2

approximately 100% (> 109 CFU g-1). The storage stability and digestion resistance of these
probiotic powders were investigated.

This work opens new perspectives for the sustainable development of new starter and
probiotic preparations with enhanced robustness.
 General introduction 3

General Introduction
Scientific and Economic Context of Probiotics

The term “probiotics” has been originally defined by FAO/WHO, and recently endorsed by
ISAPP (The International Scientific Association for Probiotics and Prebiotics) as ‘live
microorganisms that, when administered in adequate amounts, confer a health benefit on the
host’ (FAO/WHO, 2001; Hill et al., 2014). The overview of microorganisms considered as
probiotics is shown in Table 1.

Table 1. Microorganisms commonly proposed as probiotics or potential probiotics (Concluded from

Cani and Van Hul, 2015; Felis et al., 2009; Foligné et al., 2013b; Hill et al., 2014; Holzapfel et al.,
2001).
Lactobacillus Bifidobacterium Other LAB Non lactic acid Potential next
species species bacteria generation
probiotics
L. acidophilus B. adolescentis Enterococcus Bacillus cereus var. Akkermansia
Faecalis toyoi muciniphila
L. cmylovorus B. animalis Enterococcus Bacillus clausii Clostridium
faecium butyricum
L. casei B. bifidum Lactooccus. lactis Bacillus coagulans Eubacterium
hallii
L. crispatus B. breve Leuconstoc Bacillus Faecalibacterium
mesenteroides licheniformis prausnitzii
L. bulgaricus B. infantis Pediococcus Bacillus Roseburia spp.
acidilactici mesentericus
L. lactis B. lactis Pediococcus Bacillus subtillis
pentosaceus
L. fermentum B. longum Sporolactobacillus Clostridium
inulinus butyricum
L. gallinarum Streptococcus Escherichia coli
thermophilus strain nissle
L. gasseri Propionibacterium
freudenreichii
L. helveticus Saccharomyces
cerevisiae
L. johnsonii Saccharomyces
boulardii
L. kefiri
L. paracasei
L. plantarum
L. reuteri
L. rhamnosus
L. sakei
L. salivarius
 General introduction 4

The mechanisms by which probiotics provide these beneficial effects on the host may be
mainly classified in three categories: probiotic metabolic effects, microbiota composition
normalization and molecular interaction between probiotics and host. Based on these
underlying mechanisms, probiotic effects are mostly dependent on strain (Figure 1), albeit some
effects are widely spread at the species level or even across different taxonomic groups (Hill
et al., 2014).

Figure 1. Various health benefits of probiotics and its possible distribution of mechanisms. Some
benefits might be widespread among commonly studied probiotic genera; others might be frequently
observed among most strains of a probiotic species; others may be rare and present in only a few strains
of a given species (Concluded from Hill et al., 2014; Parvez et al., 2006).

The benefits of probiotics are also dependent on adequate dose administered (Johansson et
al., 2015; Mimura et al., 2004). Although it differs from different regulators and countries, the
recommended minimum dose of orally administered probiotics is typically ranging from 106
to 107 colony-forming units (CFU)/g of product and 109 CFU intake per day. However and
from a clinical point of view, no standardised dose of probiotic bacteria generalized to any
bacterial species can be recommended to date as ensuring a beneficial effect. The duration of
 General introduction 5

probiotic intake can range from days to years depending on different cases and probiotic strains
(Deshpande et al., 2011; Vouloumanou et al., 2009).

Since the increasing evidences on these benefits, products including alive probiotics are
highly demanded by current food and pharmaceutical market. According to the report of Global
Market Insights, Inc., (2016), the global probiotics market size was valued at USD 36.6 Billion
in 2015, in which the human-use probiotics was over USD 34 Billion. This market is expected
to grow continuously with over 7.4% CAGR (Compound Annual Growth Rate) from 2016 to
2023, and is projected to 64.6 Billion USD by 2020 due to the growing standard of living and
health awareness at a global scale (Global Market Insights, 2016; Markets and Markets, 2015).

Currently, the global market of probiotics is highly competitive and fragmented between
food companies. In 2015, the top five players in probiotics market, Danone, Yakult, Nestle,
Chr. Hansen and Dupont Danisco, occupied less than 40% of the global revenue (Global
Market Insights, 2016). This latter is expected to grow during the following 10 years in all the
world regions. Figure 2A shows the potential increase of probiotic market revenue in the United
States. Globally speaking, the Asia-Pacific is the largest probiotics market accounting
approximately for 38% in 2014, in which Japan forms the largest market with a share of 45%
(Mordor Intelligence, 2016). However, China appears likely to be the fastest growing market,
with a forecast at 13% CAGR from 2014 to 2020 (Mordor Intelligence, 2016). Driven by this
potential of Chinese and Indian market, the Asia-Pacific should remain the largest regional
industry for participants (Figure 2B) (Global Market Insights, 2016).

A B

Figure 2. (A) U.S. Probiotics market revenue by product, 2014 - 2023 (USD Million) and (B) probiotic
ingredients market size, by region, 2015 vs 2020 (USD Million) (Adapted from Global Market Insights,
2016; MarketsandMarkets, 2015).
 General introduction 6

Spray Drying of Probiotics: New Opportunities and Challenges

The first detailed description of spray drying was given in United States patent in 1872
(Anandharamakrishnan and S, 2015), as “the process of simultaneously atomizing and
desiccating fluid and solid substances, and its application to the purpose of the exhaustion of
moisture from such substances, and for the prevention of destructive chemical change.”
(Samuel R., 1872).With more than 140 years’ growth and development, the spray drying
process has nowadays evolved as one of the most widely-used and industry-friendly drying
technique that is spread in dairy industries all over the world.

Briefly, spray drying consists in spraying (or atomizing) the liquid feed in fine droplets (10
to 150 µm) into a flow of hot and dry air (usually 150°C to 250°C; Figure 3). The subsequent
increase in the air-liquid interface area dramatically increases the drying kinetics, and it is
generally admitted that drying occurs within a few seconds.

Figure 3. Schematic of a typical spray dryer and its operation process.

The atomizer is the heart of a spray dryer. To date, several types of atomizers have been
developed in order to deal with various feed material and thus achieve satisfactory atomization.
Among these atomizers, pressure nozzle atomizer, two-fluid nozzle atomizer and rotating
 General introduction 7

atomizer are the most frequently used. Besides, several innovative atomizers have gradually
been applied to obtain powders with distinct particle properties, piezoceramics-driven glass
nozzle and ultrasonic nozzle for instances (Tatar Turan et al., 2015; Wu et al., 2014).

Using these technically innovations and multidisciplinary knowledge, it is possible

nowadays to tailor via spray drying powders with various surface/internal morphology (Figure
4), having therefore various encapsulating abilities (Nandiyanto and Okuyama, 2011).

Figure 4. Various particle morphologies prepared using the spray drying method (adapted from
Nandiyanto and Okuyama, 2011).

On the other hand, the application of spray drying technique is expected to extend from
dairy industry to other emergent and increasing market of food ingredients, as probiotic bacteria
for instance. Indeed, drying is a widely-used process for probiotic preservation, given the
 General introduction 8

longer and stable shelf-life and lower transportation cost of dried probiotics. Nowadays, freeze
drying remains the preferred technique to preserve probiotic bacteria; but it is a time-
consuming and expensive process (Table 2). Among the drying techniques, spray drying is one
of the most promising methods in substitution to freeze drying with regards to drying of
probiotics (Schuck et al., 2016). When compared to freeze drying, spray drying exhibits a lower
specific energy cost and higher productivity (Table 2).

Table 2. General production pattern and specific energy consumption of commonly used drying
processes. (Modified from Huang et al., 2017)

Specific
energy Productivity
Processes Production pattern Disadvantages
consumption ([Link]-1)
([Link]-1 water)

Liquid High temperature involvement

Spray drying ↓ Continuous 5,300 ~50000a High investments
Solid Inapplicable for small sample size

Liquid High residence time (24-48h)

Freeze drying ↓ Batch 18,000 ~10000a High production cost
Solid Low energy efficiency

High residence time

Solid
Fluidized-bed High investments
↓ Continuous 11,400 n.a.
drying Low energy efficiency
Solid
Inapplicable for liquid sample

n.a.: not available

a
An approximate value for a large scale dryer from personal communication (The productivity actually
depends on equipment, production scale and market demand).

However, the use of spray drying to produce viable cultures, especially with “sensitive”
probiotic strains, is still challenging (Broeckx et al., 2016; Liu et al., 2015; Peighambardoust
et al., 2011). Extensive investigations have been carried out in the recent two decades all over
the world in order to find strategies to protect probiotics during spray drying (Figure 5), in
particular based on:

- Inducing bacterial stress adaptation to enhance cellular intrinsic tolerance upon spray
drying (Microbiological strategy)
- Application of protective agents to minimize cell damage during process (Material
strategy)
- Optimization of process to moderate the harsh conditions (Processing strategy).
 General introduction 9

Figure 5. Overview of main research groups focusing on the application of spray drying in probiotic
production.
 Chapter 1: Literature review 10

Chapter 1. Literature Review

This chapter aims at providing the literature review of the current strategies used to improve
probiotic viability upon spray drying process. These strategies were categorized into three parts
according to the proceeding of spray drying probiotics: the influence of bacterial culture (i.e.
feed material) preparation before spray drying, the drying parameters and operating conditions
during drying, and the storage and application of powders after drying. The studies and
investigations made it possible to introduce the strategies for improving quality of spray dried
probiotics, mainly based on knowledge from microbiology, material science, engineering and
processing disciplines.

Besides, the PhD project is introduced at the end of this chapter with its aim and innovation.
It has benefited from the multidisciplinary knowledge due to the joint direction between China
(Chemical Engineering) and France (Food Science and Technology), and between processing
(SMCF) and microbiology team (MICROBIO) in the INRA-Agrocampus Ouest joint
laboratory (UMR1253 STLO).

The literature review has been published in Trends in Food Science & Technology as:
Huang, S., Vignolles, M., Chen, X.D., Le Loir, Y., Jan, G., Schuck, P., and Jeantet, R., 2017. Spray
drying of probiotics and other food-grade bacteria: A review. Trends in Food Sci. Technol.
63, 1-17.
 Chapter 1: Literature review 11

1.1 Spray drying of probiotics: pre-drying stage

1.1.1 Selection of bacterial Strains

To date, most efforts have focused on the drying of Lactobacillus, Lactococcus and various
Bifidobacteria species. These probiotic bacteria generally do not survive well after spray drying
because of the harsh conditions which prevail during the process. The resistance characteristics
of a bacterial strain should thus constitute an important criterion when selecting probiotic
bacteria, in order to improve the final probiotic viability of the spray-dried powders.

Heat, osmotic, oxidative and desiccation stresses are usually considered to be the main
mechanisms which cause the inactivation of bacteria during and after spray drying
(Santivarangkna et al., 2008b). It has been shown that different bacterial species, or even
strains, may display variable tolerance towards such stresses.

By comparison with Lactobacillus, Lactococcus and Bifidobacteria, Propionibacteria

usually display higher tolerance due to their greater abilities for environmental adaptation,
either through their metabolism or a multitolerance response (Leverrier et al., 2004). However,
compared to Lactobacillus (Lb.) and Bifidobacteria, studies involving the drying of
Propionibacteria (P.) are still rare.

Streptococcus (S.) is usually more resistant than Lactobacillus to spray drying; for instance,
S. thermophilus was shown to survive better than Lb. delbrueckii ssp. bulgaricus in spray-dried
yoghurt (Kumar and Mishra, 2004), and S. thermophilus CCRC14085 survived better than Lb.
acidophilus CCRC 14079 in spray-dried fermented soymilk (Wang et al., 2004). The threshold
temperature at which damage is caused to microbial cells is usually within the range of the
upper limit of growth temperature of the microbial species (Foerst and Kulozik, 2011). Thus
the spray drying resistance of S. thermophilus is probably linked to its greater thermotolerance.
In another observation, Lb. paracasei NFBC 338 was however found to survive as successfully
as S. thermophilus (Kearney et al., 2009). This finding indicates that certain strains within a
usually fragile specie may be as resistant as bacteria from a generally robust specie.

When compared within the Lactobacillus genus, Lb. plantarum is a specie with relatively
robust stress tolerance (Ferrando et al., 2015). Mille, Beney, and Gervais (2005) showed that
the osmotic tolerance of Lb. plantarum was generally greater than that of Escherichia coli,
Bradyrhizobium japonicum and Lb. bulgaricus, and in some cases even higher than that of
yeasts (Saccharomyces cerevisiae and Candida utilis). Other species frequently studied in the
 Chapter 1: Literature review 12

context of spray drying include Lb. rhamnosus, Lb. casei, Lb. paracasei, Lb. salivarius and Lb.
acidophilus. However, the stress tolerance of these species appears to be strain-dependent
(Table 3).

Bifidobacterium longum ssp. longum B6 survived spray drying better than Bifidobacterium
longum ssp. infantis CCRC 14633 when fermented soymilk and reconstituted skim milk were
used as the drying media, while both B. longum. ATCC15708 and CCRC 14634 survived better
than B. infantis CCRC14661 when using gelatin as the spray drying medium (Lian et al., 2002;
Wang et al., 2004). In a comparison of 17 strains from the Bifidobacterium genus, it was found
that the greater the resistance to heat and oxidative stresses, the better was survival after spray
drying (Simpson et al., 2005). This relationship has also been reported in the Lb. rhamnosus
and Lactococcus lactis (Lc. Lactis) species (Dijkstra et al., 2014; Lavari et al., 2015).

Although the species may generally reflect the apparent resistance of different bacteria, the
robustness of probiotic bacteria during spray drying seems more likely to be strain-dependent.
Such variability is likely related to the environment of the original source, the presence of
specific genes, interactions with the extracellular matrix, or an ability for intracellular
polyphosphate accumulation and exopolysaccharide production, etc. (Lebeer et al., 2008;
Papadimitriou et al., 2016). Given that susceptibility to both process and environment is strain-
dependent, drying conditions need to be adapted while considering the latter and not directly
transposed from one strain to another.
 Chapter 1: Literature review 13

Table 3. Review of key factors determining viability of probiotic bacteria (Lb. refers to Lactobacillus, Lc. refers to lactococcus, S. refers to Streptococcus, B.
refers to bifidobacteria, P. refers to propionibacteria) upon spray drying (Modified from Huang et al., 2017).
Growth parameters Cell Drying parameters Powder
concentration moisture Survival
Bacteria T Tinlet Toutlet Devices Reference
Medium before drying Medium content (%)
(ºC) (CFU mL-1) (ºC) (ºC) (%)

S. thermophilus 60 10.2 69.5

Fermented “Luwa”
MK-10 Condensed non 80 4.4 12.7 (Bielecka and
growth medium type
fat milk (18% 42 9.5×108 n.a.1 Majkowska,
60 industrial 10.2 22.1
Lb. bulgaricus w/w) (18% w/w) 2000)
spray dryer
151 80 4.4 8.0
B. infantis
10.0 1.3
CCRC14633
B. infantis
0.2
CCRC14661 Buchi B-
MRS broth with 191 Mini (Lian et al.,
B. longum 37 n.a. Gelatin (10%) 100 50
0.05% cysteine ‡ dryer, 54.3 2002)
ATCC15708
Switzerland
B. longum
39.4
CCRC14634
B. longum B6 7.8 63.7
Lb. rhamnosus
3.2×109 * 3.8 ~30 *
E800
Buchi B-
Lb. rhamnosus 191 Mini (Corcoran et
MRS broth 37 1.6×109 * RSM (20% w/v) 170 85~90 3.7 ~50 *
GG dryer, al., 2004)
Switzerland
Lb. salivarius
6.3×109 * 2.7 ~1 *
UCC500
 Chapter 1: Literature review 14
Table 3 continued
17 strains of
Bifidobacterium 0.1~3.2×109 2.5~4.2 12~102 a
species Buchi B-
MRS broth with 191 Mini (Simpson et
B. animalis ssp. 37 RSM (20% w/v) 170 85~90
0.05% cysteine ‡ 5.0×108 dryer, 3.2 79 al., 2005)
lactis BB12 b
Switzerland
B. breve
6.3×108 3.2 38
NCMB8807 b
Trehalose (20%
69
w/w)
Lb. rhamnosus Trehalose (20% 3.8
GG w/w) +
80.8
MRS broth Monosodium Buchi B-
glutamate (Sunny-
supplemented 191 Mini
with trehalose 37 1.8×109 n.a. 65~70 Roberts and
Trehalose (20% dryer,
23 Knorr, 2009)
(1.25% w/v) ‡ w/w) Switzerland
Lb. rhamnosus Trehalose (20% 4.1
E800 w/w) +
89.3
Monosodium
glutamate
Lb. paracasei
4.0×108
A12 Buchi B-
290 mini (Páez et al.,
Lb. casei Nad MRS broth 37 4.0×108 RSM (20% w/v) 170 85 <4 ~100 *
spray drier, 2013) †
Lb. acidophilus Switzerland
5.0×109
A9
Acid whey
permeate (10% GEA pilot
P. w/v) with corn 10 e Bionov (Schuck et
30 10 SW (~50% w/w) 130 60 ~3 d 100
acidipropionici steep (2.5% w/v) spray dryer, al., 2013)
and yeast extract France
(1% w/v)
 Chapter 1: Literature review 15
Table 3 continued
Buchi B-
14 strains of Lc. M17 broth with 290 mini c (Dijkstra et
30 n.a. RSM (20% w/v) 200 100 <4
lactis species 0.5% w/v glucose spray drier, al., 2014)
Switzerland

RSM (30% w/w) SD-05 lab- 4.3 77.7

Lb. acidophilus 9 scale spray- (Maciel et al.,
MRS broth 37 ~10 * 180 85~95
La-5 dryer, Lab- 2014)
SW (30% w/w) Plant, UK 4.8 75.4

Buchi B-
Lb. plantarum Maltodextrin 290 mini (Perdana et
MRS broth 30 ~108 135 90 4~8 ~10 *
WCFS1 (20% w/w) spray drier, al., 2014)
Switzerland

Orange juice
Niro spray
Lb. plantarum (1.5% w/v) + (Barbosa et
MRS broth 37 n.a. 150 70 dryer, aw = 0.42 ~100 *
299v Maltodextrin al., 2015a)
Denmark
(2% w/v)

WPI and GA
8
5.0×10 (3:1 mixture, 2.90 37.6
~5% w/w) Buchi B-
290 mini (Eratte et al.,
Lb. casei 431 MRS broth 37 WPI, tuna oil 180 80
spray drier, 2015)
and GA (3:2:1 Switzerland
2.0×109 3.19 56.2
mixture, ~10%
w/w)
Lb. plantarum
~50 *
CNRZ 1997
MRS broth Phosphate- Buchi B-
Lb. zeae CNRZ 8 buffered saline 290 mini (Iaconelli et
37 ~10 145 70 aw = 0.3 ~1 *
2268 spray drier, al., 2015)
(~1% w/v) Switzerland
B. bifidum CIP MRS broth with
~10 *
56.7 0.05% cysteine ‡
 Chapter 1: Literature review 16
Table 3 continued

Sodium caseinate
and vegetable oil
3.25 ~15
(1:1 mixture,
20% w/w) Laboratory-
scale ADL (Liu et al.,
Lb. zeae LB1 MRS broth 37 ~1010 170 80
Sodium caseinate 310 spray 2015)
and low melting dryer, USA
point fat (1:1 3.68 ~65
mixture, 20%
w/w)
Lb. acidophilus
MRS broth 3.2×108 * ~1 *
LA-5
Buchi B-
B. animalis 37 Goat’s milk 290 mini (Ranadheera
195 85 2.9
subsp. lactis BB- RC medium ‡ 108 * (20% w/v) spray drier, < 1* et al., 2015)
12 Switzerland
P. jensenii 702 SL broth 30 109 * ~8 *
WPI (pH 7.0,
5.6 69.0
10% w/w)
Laboratory
WPI (pH 4.0, LabPlant
5.3 39.3
Lb. plantarum 9 10% w/w) SDBasic (Khem et al.,
MRS broth 37 3~4×10 110 68-70
A17 FT30MKIII, 2016)
Denatured WPI
spray dryer,
(pH 7.0, 78 ºC
UK 5.4 25.0
for 20 min, 10%
w/w)
B. infantis ATCC MRS broth with LabPlant 91.6
15679 0.05% cysteine ‡ Maltodextrin SD-05 spray (Bustamante
8 9
37 10 ~10 110 75-80 aw~0.24
Lb. plantarum (15% w/v) dryer, et al., 2017)
MRS broth England 83.0
ATCC 8014
 Chapter 1: Literature review 17
Table 3 continued
Micellar caseins
with denatured Anhydro
MRS (pH
Lb. rhamnosus whey proteins MicraSpray (Guerin et al.,
controlled at 6.8 37 ~3.2×107 e 195 85 6.3 ~50 *
GG (90:10 v/v 150, 2017)
by 6 M NaOH)
mixture, 12.5% Danmark
w/w)
Buchi B-
MRS broth with 10 290 mini (Burns et al.,
B. lactis Bb12 37 1.3×10 RSM (20% w/v) 137 75-90 4.2 ~98
0.1% cysteine ‡ spray drier, 2017) †
Switzerland
1
n.a.: not available
* means data were calculated from the figures or information from the paper
† In vivo study of effect of spray drying on probiotic functionality
‡ Growth under anaerobic condition
a
Survival ranged from 12% to 102%, depend on species and strain, closely related strains exhibiting superior heat and oxygen tolerance performed best after
spray drying
b
representitive examples from the 17 strains
c
Lc. lactis subsp. lactis strains generally displayed more robust phenotypes than Lc. lactis subsp. cremoris strains. The most robust strains displayed a more-
than-200-fold-better survival during spray drying than the most sensitive strains.
d
Data was calculated from total solid content of spray dried powders after delivery by a crystallizer
e
The unit of bacterial population is CFU g-1
 Chapter 1: Literature review 18

1.1.2 Response to cellular stress

Growth conditions are widely accepted as key determinants for the acquisition of bacterial
stress tolerance. Indeed, like other bacteria, probiotics are able to withstand and adapt to
different adverse environmental stresses (e.g. high temperatures or osmotic stress) by activating
the cellular stress-response system (Table 4), which determines their tolerance during spray
drying. This system is generally induced by exposure to a sub-lethal dose of the stress itself.
For instance, significant improvements in heat and spray drying tolerance were reported for
heat-adapted strains of Lb. salivarius (Zhang et al., 2016).

In addition to heat adaptation, the exposure of bacteria to sub-lethal osmotic, H2O2 (0.003
M for 30 min) or bile salts (0.1% w/v for 30 min) could also improve bacterial viability during
spray drying (Desmond et al., 2001). Edwin, Ananta and Knorr (2003) also found that Lb.
rhamnosus GG pre-exposed to high pressure (100 MPa) displayed greater resistance to heat.
Furthermore, starvation (mostly glucose starvation) can also induce bacterial tolerance of
osmotic and heat stresses (Guchte et al., 2002).

Table 4. Examples of genes and proteins involved in probiotics adaptation towards spray drying-related
stress (Adapted from Huang et al., 2017).
Example
Stress* Gene/Protein Description and function Reference
strain
Transcription initiation
factor that enables specific
E. coli Nissle (Coldewey et
General rpoS/σ38 or σS binding of RNA polymerase
1917 al., 2007)
to gene promoters at
stationary phase
dnaK, dnaJ,
Lb. casei (Wu et al.,
grpE/DnaK, DnaJ,
Zhang 2011)
GrpE
Heat shock proteins folding
chaperones to prevent
groL, Lb. paracasei (Corcoran et
misfolding and to promote
groS/GroEL,GroES NFBC 338 al., 2006)
the refolding or degradation
of unfolded polypeptides
Heat B. longum (Khaskheli et
hsp20/Hsp20
Osmotic NCC2705 al., 2015)

clp genes / ClpB, Clp ATPase family (Frees and

Lc. cremoris
ClpC, ClpE, ClpP members act as chaperones Ingmer,
MG1363
etc. and regulators of proteolysis 1999)
High temperature
requirement protein with Lb. plantarum (Wu et al.,
htrA/HtrA
dual chaperone-protease FS5-5 2016)
activities
 Chapter 1: Literature review 19
Table 4 continued
ATP-dependent zinc
metalloprotease with dual
Lb. plantarum (Bove et al.,
Heat ftsH/FtsH chaperone-protease activities
WCFS1 2012)
and involving in protein
quality control
ATP-binding ABC
opu genes/OpuA, Lc. cremoris (Romeo et al.,
transporter in charge of
OpuB, OpuC etc. MG1363 2003)
compatible solute transport
Phosphocarrier protein
family catalyzes the
Osmotic
phosphorylation of incoming
L. rhamnosus (Prasad et al.,
ptsH/HPr sugar substrates
HN001 2003)
concomitantly with their
translocation across the cell
membrane.
Alkyl hydroperoxide
reductase C22 protein
destroys toxic radicals B. longum (Zuo et al.,
ahpC/AhpC normally produced within NCC2705 2014)
cells to maintain cell redox
homeostasis
Catalase decomposes
hydrogen peroxide into water Lb. brevis (Lyu et al.,
katA/KatA
and oxygen, to maintain cell CGMCC1306 2016)
Oxidative redox homeostasis
NADH oxidase and NADH
peroxidase promote NADH Lb. plantarum (Zotta et al.,
nox,npr/NOX, NPR
cycling to maintain cell redox C17 2013)
homeostasis
Superoxide dismutase
destroys toxic radicals
Lb. casei (Lee et al.,
sodA/SodA normally produced within
BL23 2015)
cells to maintain cell redox
homeostasis
* Cross-protection possibly emerges between different stresses

Apart from the adaptation to environmental conditions, modest improvements to stress

tolerance can also be achieved through genetic engineering. For instance, a genetically
engineered Lb. paracasei strain was able to overproduce the heat shock protein GroESL, and
thereby exhibited a 10-fold higher thermotolerance compared to its parent wild-type strain
(Desmond et al., 2004). Given its heat robustness, the GroESL-overproducing Lb. paracasei
strain survived 10-fold better than the wild-type strain during spray drying (Corcoran et al.,
2006). Cells appear to be more robust and intact in GroESL-overproducing bacteria and are
sometimes huddled together. However, the overexpression of GroESL did not enhance survival
during storage of the lactic acid bacteria (LAB) powder. Moreover, the thermotolerance of this
GroESL overproducing strain can be further improved up to 50-fold after sub-lethal heat
 Chapter 1: Literature review 20

adaptation, which indicates the involvement of other heat stress response mechanisms (e.g.,
contributions from other heat shock proteins or general stress proteins) (De Angelis et al., 2016;
Lebeer et al., 2008). Hence, the induction of the whole battery of heat stress tolerance
mechanisms by sub-lethal heat treatment may lead to greater viability. A genetically engineered
Lb. salivarius UCC 118, which can overexpress the listerial betaine uptake system BetL, was
also reported to have a 5-fold improvement in survival following spray drying (Sheehan et al.,
2006). Indeed, betaine protects (LAB) from desiccation (Table 5).

To sum up, in most cases, bacterial spray drying tolerance may be affected by how the strain
or culture has been produced. Furthermore, Teixeira, Castro, Malcata and Kirby (1995)
reported that protein synthesis was not observed in Lb. bulgaricus during its recovery from
spray drying injury. As concluded by Morgan, Herman, White and Vesey (2006), this result
suggests that such injured bacteria do not need to synthesize new proteins during recovery from
drying injury, or that they are unable to synthesize proteins. In this case, the overproduction of
stress response proteins before drying may constitute an effective means of protecting bacteria
from drying injury, and particularly those proteins or chaperones responsible for preventing or
repairing misfolding polypeptides.
 Chapter 1: Literature review 21

1.1.3 Growth conditions and preparation of feed concentrate

[Link] Growth media

Only a few published studies have addressed how the growth medium affects the viability
of probiotics after drying. Most studies have focused on the media used in laboratories, such
as De Man Rogosa and Sharpe medium (MRS broth) which is non-food grade, and relatively
expensive for industrial application. As an example, the presence of 1 mol.L-1 or 1.25 mol.L-1
NaCl in MRS during the growth of Lb plantarum resulted in decreased residual activity after
drying in a fluidized bed or by convection. The viability of Lb. plantarum after drying was also
greater when the bacteria were grown in diluted MRS medium, despite lower betaine and
carnitine uptake in this case (Linders et al., 1997).

As well as these studies based on MRS medium, several others have explored the effects of
other media on growth and the subsequent drying or storage of bacteria. Dairy media, such as
skimmed milk and whey, were mostly used during these studies (Table 3). By comparison with
MRS, these media are inexpensive, easily used for mass production and edible. Moreover, they
may improve following production processes such as drying (Lavari et al., 2014).

When facing a reduction in water activity, bacteria tend to compensate for the concomitant
osmotic pressure and thus maintain their viability by accumulating compatible solutes such as
amino acids, quaternary amines or carbohydrates, etc. (Table 5) (Papadimitriou et al., 2016;
Wood, 2011). The accumulation of compatible solutes from media during growth thus can be
used as a strategy to increase the viability of probiotics during drying. However, it has been
suggested that the uptake of compatible solutes may not necessarily guarantee tolerance
towards drying, depending on the bacterial strains studied (Morgan et al., 2006). Furthermore,
it is worth noting that the synthesis of compatible solutes in LAB was seen to be dependent on
the presence of their precursors in the medium (Wood, 2011). The transport and accumulation
of these compatible solutes is also an energy-dependent process (Romantsov et al., 2009).
Because of the short timescale of the spray drying process, it is suggested that compatible
solutes or their precursors should be added during the growth of bacteria rather than before
spray drying.
 Chapter 1: Literature review 22

Table 5. Examples of compatible solute accumulation on improvement of bacterial stress tolerance

(Adapted from Huang et al., 2017).
Example bacterial Application in
Compatible solute Stress response Reference
strain spray drying

(Jewell and
Lb. acidophilus Enhance osmotic
Proline not to date Kashket,
IFO 3532 tolerance
1991)

Carnitine
Lb. plantarum Enhance osmotic (Kets and
Acetylcarnitine not to date
P743 tolerance Bont, 1997)
Propionylcarnitine

No significant effect Improved the (Ferreira et

Glutamate Lb sakei CTC 494
on heat tolerance survival al., 2005)

Glycine betaine Pantoea Enhance osmotic,

(Teixidó et
agglomerans CPA- water and heat not to date
Ectoine al., 2005)
2 tolerance

Enhance osmo-,
Lb. salivarius Improved the (Sheehan et
Listerial betaine cryo-, baro- and chill
UCC118 survival al., 2006)
tolerance

Enhance osmotic,
P. freudenreichii (Dalmasso et
Glycogen cold tolerance, and Not to date
ITG 20 al., 2012)
long-term survival

Enhance osmotic,
Lb. salivarius heat, desiccated Improved the (Zhang et
Trehalose
NRRL B-30514 tolerance and long- survival al., 2016)
term survival

Enhance heat,
oxidative tolerance (Alcántara et
Polyphosphate Lb. casei BL23 Not to date
and long-term al., 2014)
survival
 Chapter 1: Literature review 23

[Link] pH of growth

The effects of the pH of the growth medium on bacterial viability during drying remain a
matter of discussion, depending on the processing scheme implemented. Linders et al. (1997)
reported a two-fold increase in Lb. plantarum viability after drying when the pH was controlled
during growth. By contrast, the viability of Lb. bulgaricus during spray drying and heating was
found to be greater when the pH was not controlled during growth (Silva et al., 2005). It was
shown that in this case, enhanced protection was due to the overexpression of Hsp70, GroES
and GroEL, induced by acid stress. However, cells may lose their viability and activity as being
unable to maintain a near neutral intracellular pH in a low pH environment for a long period.
The ability to maintain intracellular pH homeostasis differs between strains (Baker-Austin and
Dopson, 2007). This may condition strain dependence relative to acid adaptation and thus
induced cross-protection against drying. Further, because acid stress influences cells in a
dynamic manner, the final pH of cultures, the time of harvest or feed storage before drying
should be taken into account when investigating the effects of pH of growth.

[Link] Growth phase

Probiotics are generally harvested during either the late exponential phase or the early
stationary phase, at which a maximal yield is reached. However, (early) stationary phase is
mentioned more frequently as being optimal and used than the (late) exponential phase,
because the cells collected also exhibit higher viability during drying (Peighambardoust et al.,
2011). The viability of Lb. rhamnosus was 2%, but over 14% and 50% when cells were
harvested in the lag, early exponential and stationary phases, respectively (Corcoran et al.,
2004). Indeed, resistance to various types of stress is increased during the stationary phase, as
its challenging conditions trigger a stringent response that leads to multitolerance which
requires general stress proteins and alternative sigma factor (Alcantara et al., 2011; Hussain et
al., 2009; Upadrasta et al., 2011).

[Link] Harvesting techniques

Centrifugation is still the most widely used harvesting technique. It has long been
considered to be efficient in concentrating cells. The usual harvesting temperature is 4°C.
However, the effects of these centrifugal conditions on bacterial viability after drying have
rarely been reported. As well as centrifugation, other harvesting techniques, such as membrane
 Chapter 1: Literature review 24

filtration, were mentioned in the review by Santivarangkna, Kulozik, & Foerst (2007), but to
the best of our knowledge, they have not been reported in the context of the spray drying of
bacteria.

Separating bacteria from growth medium can cause considerable wastage of materials
and energy. It is therefore preferable to keep the growth medium and directly spray-dry
the probiotic culture during large-scale production. Apart from drying probiotic cells,
some beneficial metabolites such as short chain fatty acids, vitamins and bacteriocin are
also retained, possibly improving the use value of powders. This process requires the dual
use of one medium for both bacterial growth and spray drying. A food grade medium is
highly desirable because of application considerations. However, a key problem of this
process is that the solid content of bacterial growth isotonic media is typically low
(5%~10%), while a high solid content in the feed concentrate is required to enable
productivity and energy savings. This is usually achieved by adding more powders to the
culture, or by implementing concentration techniques (Schuck et al., 2013). However,
these intermediate steps may cause a loss of viability and increase the risk of
contamination.
 Chapter 1: Literature review 25

1.2 Spray drying probiotics: During drying

1.2.1 Bacteria may be injured during spray drying

Damage to bacteria during spray drying is not only ascribed to the thermal effect, but also
to a loss of bound water at the cell surface. The cytoplasmic membrane is highly susceptible to
cellular injury, and this is exacerbated during spray drying (Santivarangkna et al., 2008b).
Indeed, the removal of water leads to a state transition of the phospholipid bilayer, from
lamellar to gel phase, or even a hexagonal one (Figure 6A). This results in the phospholipid
chains gaining a rigid and fully extended structure (Crowe et al., 1998; Wolfe and Bryant,
1999). It was indeed shown that fine holes at the surface of Lc. cremoris were observed by
scanning electron microscopy in droplet drying with skimmed milk at 90°C and 110°C, while
the cell surface remained intact when drying at 70°C (Figure 6B) (Fu et al., 2013a).

A B

Figure 6. (A): The strains produced by dehydration-induced stress (source from (Wolfe and Bryant,
1999)): (a) The lamellar fluid phase at high hydration. (b) The geometric strain produced at lower water
content. (c) Transition to gel phase at lower water contents. (d) The mixed phase of lipid species with
high (top) and low (bottom) hydration. (e) Separation of hydrophilic and hydrophobic molecules. (f)
tHE topological response emerged at very low hydration. (B): The surface morphology of Lc. cremoris
cells dried with different air temperature (Adapted from Fu et al., 2013a).
 Chapter 1: Literature review 26

Other dehydration stress targets include the nucleic acids (where the mechanism remains
unclear) and ribosomes, which are probably injured as a result of the escape of Mg2+ from the
heat-compromised cell membrane (O’Connor et al., 2009).
 Chapter 1: Literature review 27

1.2.2 Drying Devices

[Link] Spray drying scale

To date, little attention has been paid to the effects of spray drying devices on the viability
of probiotics in powders. The main influence of different devices on probiotic powders is
probably the residence time of particles in the drying chamber: The longer the residence time,
the longer the bacteria are exposed to stress and consequently the poorer the viability. Another
factor worth noting is that industrial scale spray dryers are normally suitably equipped with
pneumatic devices to enable the continuous collection and cooling of the powders, thus
maximizing viability.

Table 3 presents the drying equipment used during different studies, which mainly focused
on laboratory-scale spray dryers. It is worth noting that only two studies mentioned
experiments on a larger semi-industrial to industrial scale, i.e., with a water evaporation
capacity in the range 100 to 1645 kg h-1 (Bielecka and Majkowska, 2000; Schuck et al., 2013).
It is therefore still difficult to conclude as to the influence of scaling-up on the residual probiotic
viability of the resulting powders.

[Link] Atomization

The cell inactivation caused by the atomization process per se is usually suggested to be
negligible. Fu, Suen and Etzel (1995) reported a complete viability of Lc. lactis ssp. lactis after
the spraying step, despite the shear force experienced in the nozzle. By contrast, Riveros, Ferrer
and Bórquez (2009) found that the viability of Lb. acidophilus increased by around 2 log CFU
g-1 when lowering the two-fluid spray nozzle pressure from 1.5 bar to 1 bar. However, it should
be noted that a lower spray pressure might first of all increase the droplet size, and hence affect
the drying kinetics and subsequent temperature increase. Besides, Guerin et al. (2017) showed
that the increase in bacteria population after atomization might be caused by cell reorganization
during atomization, or more specifically, the Lb. rhamnosus GG distributed in linear chains
before atomization and in individual cells after atomization (Figure 7). In another study, a
significant reduction in survival was reported for a strain of Lc. lactis ssp. cremoris after
atomization using both a rotary wheel and two-fluid nozzle atomizer (Ghandi et al., 2012).
Furthermore, the addition of ascorbic acid (as an anti-oxidant) to the drying medium was found
to reduce the damage caused by atomization. These results indicate that a loss of cell viability
 Chapter 1: Literature review 28

during atomization may be strain-dependent, and the inactivation mechanism could be

triggered by both shear force and oxidative stress.

Figure 7. Effect of spraying-induced shear stress on cellular organization (Adapted from Guerin et al.,
2017).
 Chapter 1: Literature review 29

1.2.3 Protective agents

A strategy that is commonly employed to enhance cell protection during drying consists in
adding specific components, known for their protective properties, to the surrounding medium.

[Link] Carbohydrates

Among protective agents of a carbohydrate type, trehalose has by far been the one most
frequently investigated, given the established role of its accumulation in the acquired survival
of some microorganisms facing anhydrobiosis. It has been suggested that this mechanism is
due to the stabilizing effect of trehalose on membranes and proteins, by replacing the water
around polar residues within these macromolecular structures (i.e. water replacement
hypothesis), and thus depressing the membrane phase transition temperature (Tm) (Figure 8A)
(Crowe et al., 1998; Santivarangkna et al., 2008a). Moreover, Conrad, Miller, Cielenski and
De Pablo (2000) demonstrated a synergistic effect of trehalose and borate ions in protecting
bacteria: indeed, the glass transition temperature (Tg) of the dry medium is increased in this
case due to crosslinking of the trehalose molecules.

A B

Figure 8. (A) a simplified phase diagram of a phospholipid associated with drying and rehydration
processes: During drying at a given temperature until cells reach a relatively low moisture content
(closed circle), cell membrane may remain in liquid phase in the presence of a sugar at this same point.
(B) According to the water replacement hypothesis: Sugars depress the membrane phase transition
temperature (a) by specifically interacting with phospholipid headgroups, or (b) by vitrification of
sugars between the membranes. (c) Polymeric sugars with a very high molecular weight are not capable
of depressing the Tm because they have fewer osmotic effects and are excluded from the intermembrane
space during the removal of water. (Adapted from Santivarangkna et al., 2008a)
 Chapter 1: Literature review 30

In a recent work, carbohydrates that included trehalose, sorbitol, mannitol, xylose, glucose,
sucrose, maltose, lactose, maltodextrin with dextrose, inulin, fructo-oligosaccharides, galacto-
oligosaccharide and potato starch were used as drying matrices in order to compare the
protection they endowed on probiotic Lb plantarum WCFS1 during convective droplet drying
(Perdana et al., 2014). It was found that a carbohydrate-rich formulation with a low molecular
weight and high glass transition temperature (Tg) procured the highest degree of protection
during drying. It may be explained by the exclusion effect of bacterial membrane on the high
molecular weight sugars (Figure 8B).

[Link] Reconstituted skimmed milk

Reconstituted skimmed milk (RSM) appears to be another suitable medium to enable the
efficient spray drying of probiotic cultures (Table 3). In view of the water replacement
hypothesis, the lactose in RSM may play an important role in the same way as the non-reducing
disaccharides trehalose and sucrose in cell protection. However and unlike these latter, the
effect of the reducing disaccharide lactose in protecting bacterial cells during drying remains
doubtful, because of the possibility of its interaction with proteins (Maillard reaction) during
drying and storage. When using whey permeate (which mainly consisted of lactose) as a carrier
for the spray drying of Lb. acidophilus, its survival was only half of that achieved with RSM
under similar conditions (Riveros et al., 2009). Nevertheless, whether or not the lactose in RSM
plays the main role in stabilizing LAB during storage remains undetermined.

The proteins in RSM can prevent cellular injury by stabilizing cell membrane constituents.
Furthermore, they may form a protective coating on the bacterial cell wall, when interacting
with milk calcium (Huang and Chen, 2013; Zheng et al., 2016). The calcium in milk can also
cause milk protein aggregation during heat treatment, forming aggregates that protect by their
structure the microorganisms: indeed the protection of Lb. rhamnosus GG was higher in pre-
aggregated milk than that of untreated milk during heat treatment (Huang et al., 2014).
Interestingly, innovative spray-dried microparticles with different reconstitution behaviors
have also been developed based on milk protein aggregation controlled by chymosin (Guerin
et al., 2017). The microparticles displayed an encapsulation effect on Lb. rhamnosus GG when
rehydrating at 40°C, while a releasing effect was seen at 8°C. During another laboratory-scale
spray drying experiment, RSM (20% w/v) was found to confer superior protection on Lb.
paracasei Nad, Lb. casei A13 and Lb. acidophilus A9, with around 100% survival for all three
 Chapter 1: Literature review 31

strains under the same drying conditions(Páez et al., 2013). As well as this well-preserved
viability, the resistance to digestion and immunomodulation capacity of these three strains
within the powders was also improved significantly when compared to the fresh culture.

[Link] Other protective agents

Other protective agents have also been used in the spray drying of probiotics: as such, dairy-
based materials have often been reported because of their compatibility with probiotics and the
possibility to promote their efficiency (Lee et al., 2015b). Whey has also been reported as
being an efficient protective agent for the spray drying of probiotics (Maciel et al., 2014).
Its advantages are linked to its source as a by-product of cheese manufacturing and to its
potentially good powder quality (e.g. solubility, flowability, dispersibility, etc.) when
compared with casein-based products (Lavari et al., 2014).

Low melting point fat (LMF) has also been reported to protect Lactobacillus cells from
damage when it is added to sodium caseinate during spray drying (Liu et al., 2015). This
protection may be attributed to the ability of LMF to absorb some of the thermal energy present
during spray drying.

As well as dairy-based materials, other reported protective agents include gelatin, gum
arabic, fruit juice, etc. (Table 3), used either alone or in a mixture to achieve synergistic effects.
For instance, the spray drying of probiotics with fruit juice as a medium or additive was recently
reviewed by Barbosa and Teixeira (2016). The incorporation of prebiotics in such protective
agents is also of interest because of the possible “synbiotic” effect. For example, the addition
of Omega-3 fatty acids to a whey protein-gum Arabic complex significantly improved the
viability of Lb. casei 431 after spray drying (Eratte et al., 2015). The viability of Lb. plantarum
after spray drying was also higher when incorporating galacto-oligosaccharides (GOS) in
maltodextrins, or fructooligosaccharide (FOS) in whey protein isolate (WPI) (Rajam and
Anandharamakrishnan, 2015; Sosa et al., 2016). Recently, mucilage and soluble proteins from
chia seed and flaxseed were found to protect Lb. acidophilus, Lb. plantarum and B. infantis
during spray drying (Bustamante et al., 2017, 2015). However, Corcoran et al. (2004) and Pinto
et al. (2015) reported that the presence of the prebiotics inulin and polydextrose in whey
concentrate did not enhance probiotic viability during spray drying or powder storage.
 Chapter 1: Literature review 32

[Link] Total solids content

The total solids contents of the drying media reported in the literature are usually 20~30%
(w/v): this value has been considered as being optimal to ensure the high residual viability of
different LAB strains (Table 3). An increase in the feed concentration above this point using
gelatin, gum arabic and soluble starch resulted in lower viability of the bifidobacteria (Lian et
al., 2002). Indeed, it should be noted that the stress adaptation to hyper-osmolality that drives
the increase in viability after drying may not be acquired in the case of too short an exposure
to osmotic stress before spray drying, leading to controversial results on the effect of the total
solid contents.

Apart from residual bacterial viability, the process energy costs, powder quality and
subsequent product applications should also be taken into account. More specifically, a high
total solids content is usually desired by industry because of the benefits for drying process
productivity, lower energy costs and a better encapsulation effect. By contrast, a low total solids
content may make it possible to maintain an isotonic environment for the bacteria before drying
and obtain fine powders with a high cell count (a higher cell/medium ratio) after spray drying.
 Chapter 1: Literature review 33

1.2.4 Drying parameters

[Link] Drying temperature

Drying temperature, in combination with drying time, is the key factor that influences the
final probiotic viability of powders because of the general heat sensitivity of these bacterial
cells. As such, the drying kinetics mainly determines bacterial inactivation during drying, and
particular attention should be paid to the drying curve, i.e. temperature as a function of time.
Briefly, the drying curve generally exhibits two stages:

- During the first stage, the temperature of the droplets is limited to the wet bulb
temperature by an almost constant evaporation rate of water, and bacterial inactivation
is therefore limited.
- During the second falling rate stage, the temperature of droplets rises toward the outlet
air temperature (Toutlet), depending on the residence time. The latter depends on both
the drying temperatures applied and the expected moisture content and water activity
of the dried product.

It has thus been shown extensively that the lower the Toutlet, the higher the post-drying
viability (Table 3). Toutlet is therefore considered to be the principal drying parameter that affects
the viability of spray-dried LAB, and any lack of monitoring and control of the latter may be
markedly detrimental (Peighambardoust et al., 2011). For instance, it was seen that relatively
small changes in Toutlet appeared to have significant effects on the survival of Lb. salivarius
(Zhang et al., 2016).

However, apart from bacterial viability, the Toutlet can also influence powder quality: if the
Toutlet is too low, the residual water activity and moisture content in powders should exceed the
values required for prolonged powder storage, e.g. around 0.2 of water activity or 4% of wet
basis moisture content (Abe et al., 2009; Vesterlund et al., 2012). The reduction in viability
caused by an increased Toutlet may also vary with the drying medium used. A greater reduction
in B. longum and B. infantis viability was observed in soluble starch compared to other carriers
such as gelatin, gum arabic and skimmed milk (Lian et al., 2002). Further, industrial spray
drying at a lower Toutlet may also result in improved storage stability (Desmond et al., 2002).
Therefore, spray drying temperatures are of considerable significance for the preservation of
bacteria, and probably need to be optimized individually for any new application.
 Chapter 1: Literature review 34

[Link] Drying rate

The influence of the drying rate on the viability of post-drying probiotics is still a matter of
debate. At the laboratory scale, it was found that slow drying kinetics led to significant
inactivation of the dehydration of Lb. plantarum, while a rapid drying rate could instantly
stabilize the cells and thereby prevent this inactivation (Perdana et al., 2013). In addition, the
influence of a slow drying rate on the viability of Lc. lactis bv. diacetylactis may have arisen
from the detrimental reactions favored by higher water mobility above an a W of 0.84-0.88
(Santivarangkna et al., 2007). Conversely, and as already stated in this review, a high drying
rate during the first stage of drying, when facilitated by hydraulic membrane permeability, may
limit bacterial adaptation because of too short exposure to the gradual withdrawal of moisture
(Linders et al., 1996).

[Link] Process optimization

In order to obtain probiotic powders with high bacteria viability and good powder quality,
the process should be optimized in order to simultaneously achieve a reduction in drying
temperature and an acceptable moisture content in the powders. A multi-stage drying process
is the strategy most frequently employed for this purpose. For example, a process was proposed
by Schuck et al. (2013) for pilot-scale spray drying which enabled a reduction in Toutlet during
the spray drying of Propionibacterium acidipropionici, with sweet whey as the carrier. A
crystallizer was used after the spray drying chamber to enable a low Toutlet at 60°C along with
a high moisture content: indeed, lactose crystallization reduced the powder hygroscopicity and
improved the flowability of the powder. Finally, an external fluid-bed dryer was used to remove
the excess of moisture. However, the applicability of this multi-stage drying process should
be validated on more bacterial strains.
 Chapter 1: Literature review 35

1.2.5 “In-process” investigations

[Link] Single droplet drying

During a real-life spray drying process, billions of droplets are sprayed into a relatively
large chamber. It is therefore only possible to analyze the microbial viability and drying status
of samples at the start and/or at the end of the process, without being able to address the
complexity of ‘in-process’ droplet-particle conversion and the related mechanism of probiotic
inactivation. To circumvent these difficulties, single droplet drying (Figure 9) has therefore
been developed to map the drying behavior of bacteria at the droplet level during the drying
process (Schutyser et al., 2012).

Figure 9. (A) Schematic diagram of a glass-filament single droplet drying experimental set-up (droplet
diameter, weight, and temperature were measured in separate runs). (B) An example of the observation
of morphology evolution during the single droplet drying (Adapted from Wang et al., 2014).

Under such a protocol, and controlling the drying conditions, the protective effect of the
medium can easily be compared without running spray drying at a larger scale. On the other
 Chapter 1: Literature review 36

hand, the effect of some drying parameters on bacterial inactivation can also be correlated by
comparing drying kinetics and bacterial inactivation using the same drying medium (Fu et al.,
2013a; Perdana et al., 2013). Further, the drying rate may be relatively slow when applying
mild conditions, thus enabling the observation and analysis of phase transition phenomenon
such as crystallization, sol-gel transition or skin formation. These observations may offer new
avenues for the design of functional probiotic particles via spray drying. For example, lactose
crystallization during droplet drying was found to be related to bacterial inactivation, which is
probably dependent on the crystal shape formed by different media (Perdana et al., 2014). In
addition, the milk protein type and its denaturation degree are important factors in determining
powder morphology (Khem et al., 2015; Sadek et al., 2013). This formation of particle
morphology was recently found to be a factor that triggered bacterial inactivation, probably
due to the mechanical stresses in play (Khem et al., 2016).

Based on the single droplet drying technique and the development of an in situ analytical
technique, it is also possible to gain insights into the interaction between drying conditions and
the structure of bacterial cell components, thus suggesting specific strategies for the protection
of bacterial cells, or even cellular components at the molecular level.

[Link] Modelling

Based on the “in-process” data obtained using the single droplet drying technique,
mathematical models have been developed to predict the inactivation of probiotics during
thermal convective drying.

A first order kinetics model was reviewed by (Chen and Patel, 2007). The inactivation
kinetics of microorganisms during drying was expressed as:
d(𝑁⁄𝑁0 )
= −𝑘𝑑 (𝑁⁄𝑁0 ) (Eq 1)
dt

where N and N0 are the concentrations of live bacteria (CFU mL-1) after and before drying,
respectively, and kd is the inactivation rate constant (s-1 for fast drying).

The Arrhenius equation is used to correlate the temperature of the microbial medium with
the inactivation rate of the microorganisms. Therefore, kd is usually expressed as:
𝐸
𝑘𝑑 = 𝑘0 exp⁡(− 𝑅 𝑑𝑇) (Eq 2)
𝑔
 Chapter 1: Literature review 37

where k0 is the pre-exponential factor and Ed is the deactivation energy which represents the
energy required to deactivate living cells, and Rg is the universal gas constant (approximately
8.314 J mol-1K-1).

To incorporate the moisture content factor (X), Meerdink & Riet (1995) proposed an
attractive approach containing just four parameters (a, b, k0 and Ed) which must be determined
from experiments. The inactivation rate constant (kd) is then expressed as:
𝐸𝑑 +𝑏𝑋
𝑘𝑑 = 𝑘0 exp⁡(𝑎𝑋 − ) (Eq 3)
𝑅𝑔 𝑇

(Eq 3) was then further developed by incorporating the influence of average drying rate and
average heating rate as:
𝑑𝑋 𝐸
𝑘𝑑 = 𝑘0 (1 + 𝑏 ∙ | 𝑑𝑡 |)exp⁡(− 𝑅 𝑑𝑇) (Eq 4)
𝑔

𝑑𝑇 𝑑𝑋 𝐸
𝑘𝑑 = 𝑘0 (1 + 𝑎 ∙ | 𝑑𝑡 |)(1 + 𝑏 ∙ | 𝑑𝑡 |)exp(− 𝑅 𝑑𝑇) (Eq 5)
𝑔

𝑑𝑋 𝑑𝑋 2 𝐸
𝑘𝑑 = 𝑘0 (1 + 𝑎 ∙ | 𝑑𝑡 | + 𝑏 ∙ | 𝑑𝑡 | )exp⁡(− 𝑅 𝑑𝑇) (Eq 6)
𝑔

The prediction of these three equations was found to be more accurate when compared with
the traditional non-rate models (Eq 2) and (Eq 3) (Li et al., 2006).

As well as these first-order-kinetics-based models, a statistical model was also developed

based on the principle of Weibull distribution (Perdana et al., 2013). This hypothesized the
death of microbial cells during spray drying as probabilities rather than determinism. This led
therefore to a distribution of inactivation times, which means the survival curve of bacteria is
a cumulative form of the inactivation times. Temperature and moisture content were
incorporated as two factors determining the inactivation of probiotic Lb. plantarum WCFS1.

Overall, the predictions of these models agreed well with the experimental data on bacterial
inactivation during single droplet drying. However, the predictions cannot be considered
precise, particularly in the context of real-life spray drying. This may be due to the much faster
drying rate in spray drying than in droplet drying, and because factors such as oxidative stress,
osmotic stress and the protective capacity of the medium, must also be taken into account.
Furthermore, the intrinsic tolerance of bacteria varies significantly over different species and
strains, which may lead to problems in applying these models to different cases.
 Chapter 1: Literature review 38

1.3 Spray drying probiotics: Post-drying stage

1.3.1. Storage stability of bacteria in powder form

[Link] Injury-induced loss of viability and strain-dependent shelf-life

The numbers of viable bacterial cells tend to decrease during storage (generally less than 6
months), and particularly during the early stages (up to one month), possibly due to cell injury
during spray drying (Wang et al., 2004). For instance, good survival was found to depend on
selecting the best survivors following spray drying (G.E. Gardiner et al., 2000). The stability
of a freeze-dried culture was reported as being better than that of a spray-dried culture,
suggesting the damage caused by heat stress (Wang et al., 2004).

The storage conditions of powders may also have a significant influence on the survival of
probiotics. Key parameters which must be controlled include: storage temperature, moisture
content or water activity, exposure to oxygen and light, powder composition and storage
materials (Morgan et al., 2006). These conditions may cause heat, desiccation, oxidative and
starvation stress to bacteria, which can lead to a considerable loss of the viability of probiotics
during long-term storage. The intrinsic tolerance of bacterial strains is known to play a critical
role in overcoming their inactivation due to either spray drying injury or storage-related adverse
stresses. The shelf-life of probiotic powders is therefore dependent on the bacterial strains
concerned.

[Link] Storage temperature and glass transition temperature (Tg)

Probiotic viability in a powder is inversely related to storage temperature. This has been
demonstrated extensively in probiotic-containing powders obtained from spray drying
(Barbosa et al., 2015b; Wang et al., 2004). However, prolongation of the shelf-life of probiotics
by reducing the storage temperature is not favored by industry because of the cost of chilled or
frozen storage, even though storage at room temperature still poses an overwhelming challenge
to the stability of probiotic powders.

During storage, the viability of probiotics is affected in particular by temperature in relation

to the Tg of the dried sample, which in turn is mainly dependent on its moisture content (Passot
et al., 2012). Indeed, dried biological matrices stored at a temperature below their Tg have been
shown to display greater stability because of the low molecular mobility and reaction kinetics
induced by the high viscosity of the glassy state (Buitink et al., 2000). In this respect,
 Chapter 1: Literature review 39

monitoring the moisture content during storage may be of considerable value to controlling the
detrimental influence of aging on viability, although this has rarely been reported in the
literature. Another strategy consists in supplementing the growth medium with specific
components (e.g. high-molecular-weight maltodextrins) in order to increase the overall Tg of
the final dried product and thus better maintain viability during storage under given conditions
(Conrad et al., 2000).

[Link] Moisture content: the influence of water activity

It is well known that the lower aw limit for bacterial growth is around 0.6 (Beuchat, 1981).
This suggests that a high aw might lead to a reduction of probiotic viability and an increase in
the risk of contamination during storage. The survival rate is not related linearly to aw. In most
studies, the optimal range of aw values for the storage of probiotics was lower than 0.2. For
example, the relatively poorer survival of S. thermophilus CCRC 14085 and B. longum B6
detected in spray-dried fermented soymilk after storage could partially be attributed to the high
aw of this dried product (Wang et al., 2004). Spray-dried Lb. paracasei CRL 431 was found to
survive better when the aw was lower than 0.33 (Poddar et al., 2014). It was also found in the
context of vacuum-drying that Lb. paracasei maintained significantly higher survival at an a w
of 0.07 when compared with that seen at 0.22 and 0.33 following storage at 20°C and 37°C for
two months, respectively (Foerst et al., 2012). Another study performed on Lb. rhamnosus GG
also showed that the viability of bacteria in dried crushed flaxseed dropped rapidly, with an aw
at 0.43 and 0.22 during long-term storage, while a loss of viability of only 0.29 log10 units was
found with an aw at 0.11 (Vesterlund et al., 2012).

[Link] Exposure to oxygen: lipid oxidation

During storage, the oxidation and subsequent saturation of membrane lipids exert a negative
impact on viability (Teixeira et al., 1996). Changes to the degree of lipid unsaturation, which
increased over time, were shown to markedly affect the passive permeability of the membrane.
Furthermore, the products of lipid peroxidation have been shown to induce damage to the
bacterial cell wall, cell membrane and DNA during storage (Teixeira et al., 1995; Zotta et al.,
2017).

In order to limit these detrimental effects, the drying medium can be supplemented with an
antioxidant. However, contradictory results have been reported in the literature regarding this
 Chapter 1: Literature review 40

strategy. The addition of ascorbic acid and monosodium glutamate during spray drying
improved culture viability during powder storage (Sunny-Roberts and Knorr, 2009), although
other studies found that they had detrimental effects on culture stability during storage. For
example, the addition of ascorbic acid and monosodium glutamate protected Lb. bulgaricus
cells, but only during storage at 4°C. At 20°C, the death rate of the culture was even higher in
the presence of these compounds than in the control sample (Teixeira et al., 1995). This could
be explained by the pro-oxidant properties of ascorbic acid as a metal ions reducer, in addition
to its antioxidant function as a radical scavenger.

Barbosa, Borges and Teixeira (2015) also reported that the exposure of Lb. plantarum 299v
to sub-lethal thermal, acid or oxidative stress before spray drying could enhance bacterial
survival during storage for 180 days at room temperature. However, these sub-lethal stresses
did not improve the storage stability of Pediococcus acidilactici HA-6111-2 during the same
study, which indicates the strain-dependence of this strategy.

[Link] Powder composition and protective agents

Spray drying is a method used at an industrial scale to encapsulate various ingredients,

including microorganisms. Encapsulated bacterial cells have been widely reported as being
more stable than free cells (De Prisco and Mauriello, 2016). Song, Cho and Park (2003) showed
that the protective effect of encapsulation during storage increased as the storage temperature
rose. The effects of encapsulation on the shelf-life of bacteria are also closely linked to the
encapsulating carrier (drying medium).

The protective agents used during drying (i.e. efficient drying medium) may also protect
bacteria during storage. For example, non-reducing disaccharides (trehalose and sucrose)
provided good protection for Enterococcus faecium and Lb. plantarum during both fluidized-
bed drying and storage (Strasser et al., 2009). However, a medium suitable for drying may be
not effective in protecting bacteria during powder storage, and vice versa. For instance,
addition of polydextrose and oligofructose to skimmed milk was found to decrease the stability
of Lb. rhamnosus during powder storage when compared to skimmed milk alone, although this
did not affect protection during spray drying (Ananta et al., 2005). The addition of chitosan to
the drying medium was found to decrease viability of Lb. acidophilus NCIMB 701748 during
spray drying, but provided excellent protection for the bacterial cells during long-term storage
(Yonekura et al., 2014).
 Chapter 1: Literature review 41

[Link] Storage materials

Options for packaging include different types of barriers to the aforementioned reactive
agents such as humidity, oxygen and light. High barrier plastic bags and blister packs have
been reported (Morgan et al., 2006). It was found that laminated pouches ensured better
protection during the storage of S. thermophilus CCRC 14085 and B. longum B6, when
compared to glass or PET bottles.

Modifying the composition of the surrounding atmosphere by means of a vacuum or

nitrogen generally enhances the storage stability of foods. However, the improvement to the
storage stability of probiotic powders procured using this method appears to be slight or
insignificant (Clementi and Rossi, 1984; Espina and Packard, 1979). Ambiguous results have
thus been obtained regarding the effects of the storage atmosphere on bacterial stability; so
they are rarely reported.
 Chapter 1: Literature review 42

1.3.2. Applications of dried bacteria

[Link] Rehydration capacity

After drying, a number of functionalities are expected to be retained in probiotics in

powders, such as metabolic activity, tolerance towards human gastrointestinal juices,
adherence to epithelial surfaces, antagonistic activity against pathogens and immunoregulatory
capacities. To operate these probiotic functionalities, rehydration is generally the first step for
products in a powder form.

The recovery of spray-dried probiotics increased with increase in the rehydration

temperature, and was seen to be a strain-dependent process (Mille et al., 2004). At the same
time, this temperature should not be as high as the heat-lethal temperature (i.e. 50°C for most
LAB). Rehydration of spray dried S. thermophilus and B. longum resulted in enhanced recovery
of live cells as much as the temperature increased, within the range 5 to 50°C (Wang et al.,
2004). When the rehydration of Lb. bulgaricus was performed at 4°C, the cell concentration
was only 2.1 x 109 CFU g-1, whereas it reached 1.5 x 1010 CFU g-1 at 30°C or 37°C. Between
these values, the live bacteria recovery increased steadily in line with the rehydration
temperature: 3 x 109 CFU g-1, 3.5 x 109 CFU g-1, 5.2 x 109 CFU g-1, and 1.1 x 1010 CFU g-1 at
10°C, 15°C, 20°C and 25°C, respectively (Mille et al., 2004). Conversely, temperature had no
significant effect on the final bacterial concentration when dried Lb. plantarum samples were
rehydrated at 30°C or 37°C (1.5 x1010 CFU g-1 and 1.3 x 1010 CFU g-1, respectively). This
could be explained by the higher cellular resistance of Lb. plantarum.

Interestingly, the difference in the optimum rehydration temperature observed when

recovering freeze-dried (20°C) and spray-dried (35°C to 50°C) bacteria further demonstrated
that a physiological difference does exist between freeze-dried and spray-dried cells of S.
thermophilus CCRC 14085 and B. longum B6 (Wang et al., 2004). Furthermore, the
reconstitution behavior of freeze-dried yoghurt was better than that of spray-dried yoghurt,
which may be due to the thermal cellular injury of spray-dried bacteria (Kumar & Mishra,
2004).

Regarding the effect of the rehydration medium, Teixeira, Castro, & Kirby (1995), and
Teixeira, Castro, Malcata (1995) demonstrated that there were no significant differences
between the recovery rates of dried Lb. bulgaricus when using skimmed milk, MRS broth,
deionized water or phosphate buffer as the rehydration medium. Only minor differences (of a
maximum 0.5 log CFU mL-1) were observed between the recovery rates of spray-dried Lb.
 Chapter 1: Literature review 43

paracasei NFBC338 powders in various rehydration media, which included AM buffer (0.01
mol/L K2HPO4 and 0.01 mol/KH2PO4), AM buffer containing 20% (w/v) sucrose, maximum
recovery diluent (MRD), 10% (w/v) RSM, and sterile water (Desmond et al., 2001).

[Link] Changes to bacterial metabolism

Unlike freeze-drying, spray drying has been reported to delay lactic acid production in Lc.
cremoris, Lactobacillus casei ssp. pseudoplantarum and S. thermophilus (To and Etzel, 1997).
The higher the outlet air temperatures, the longer the lag time before the acid production began.
For example, the lag time for Lb. pseudoplantarum was shorter (2h) at Toutlet 65°C than at Toutlet
90°C (10h). Apart from its influence on lactic acid fermentation, more attention needs to be
paid to how lag time affects the beneficial effects of spray-dried probiotics. For instance, it has
been documented that bacteria retain their ability to produce bacteriocin peptides following
spray drying: this has been shown for nisin (Nisaplin®; Danisco A/S, Copenhagen, Denmark),
lacticin 3147 (Morgan et al., 2006), and other bacteriocins produced by both lactobacilli and
lactococci (Mauriello et al., 1999; Zhang et al., 2015). Pérez-Chabela, Lara-Labastida,
Rodriguez-Huezo and Totosaus (2013) reported that the incorporation of spray-dried LAB in
meat batters enhanced the initial LAB population and a concomitant reduction of
Enterobacteria levels in the meat was seen during storage. Lb. salivarius UCC 118 was also
reported to retain its ability to produce bacteriocin after spray drying, even at high outlet air
temperatures (up to 95°C) (G. E. Gardiner et al., 2000). In another study, the spray drying
process did not affect the antagonistic activity of Lb. sakei and Lb. salivarius against
Staphylococcus aureus, Listeria innocua and Listeria monocytogenes (Silva et al., 2002).

Overall, the residual viability of probiotics after spray drying is an important indicator when
characterizing probiotic activity within spray dried powders. Low viability is more a reflection
of the cellular injury of probiotics during the drying process. Severe cellular injury may
significantly influence the lag time of bacterial growth and primary metabolism-related
functionalities, such as lactic acid production. However, the culture contains a sufficient
population of new generation bacteria when regrowth reaches the late log phase or early
stationary phase. These new generation bacteria and their secondary metabolism-related
functionalities (such as bacteriocin-producing ability and enzyme activity) are not influenced
by the drying process.
 Chapter 1: Literature review 44

[Link] Resistance to food manufacturing processes

Due to the convenience of powder-form products, spray-dried bacteria can be expected to

be used as an ingredient to be added in other foods. Conversely, this necessitates the heat
resistance of spray-dried probiotics because of the presence of a thermal process in most food
manufacturing (Dianawati et al., 2015; Zhang et al., 2014). Spray-dried bacteria generally
display higher heat resistance when compared to free bacteria. For example, the survival of
spray-dried Lactobacillus reuteri DSM 17938 (with co-cross-linked alginate and chitosan as
drying medium) was significantly better than that of free bacteria during the baking of a
chocolate soufflé (Malmo et al., 2013). The reduction in the bacterial population of spray-dried
Lb. plantarum BM-1 was only 0.08 log CFU mL-1 after exposure to 60°C, whereas the
reduction in free cells reached 2.58 log CFU mL-1 (Zhu et al., 2016). This can be explained by
three mechanisms: first, the encapsulation effect of the drying medium (De Prisco and
Mauriello, 2016; Sabikhi et al., 2010); second, the lower water activity surrounding bacterial
cells (Laroche et al., 2005), and third, the possible induction of a heat shock response due to
the heat stress experienced during spray drying (Rossi et al., 2016). In addition to the thermal
process, probiotics or LAB cells (which are used as a starter) are often exposed to an acidic
environment in foods because of the drop in pH during fermentation. Spray-dried bacteria can
often survive better in an acidic environment during their shelf-life. For example, Dimitrellou
et al. (2016) showed that spray-dried Lb. casei ATCC393 had an increased survival rate in
fermented milk during refrigerated storage, by comparison with free bacteria.

However, it is also possible that the cellular injury caused during spray drying may lead to
a higher bacterial death rate during subsequent processing (Wu, 2008). Hence, the adequate
bacterial viability and activity in spray-dried powders constitute important criteria for food
applications.

[Link] Resistance to gastrointestinal conditions

Survival is essential for bacteria that are targeted to populate the human gut: this is one of
the most important issues when procuring health benefits with probiotics (De Prisco and
Mauriello, 2016).

The challenge is to maintain probiotic viability during delivery through the digestive tract.
More specifically, a considerable loss of viability was observed when bacterial cells were
 Chapter 1: Literature review 45

exposed to acid and bile stresses, gastric or intestinal enzymes, and a mechanical shearing force
during consumption.

In vitro studies have shown that an appropriate spray drying medium can protect probiotics
against stress during digestion. For example, a loss of viability was limited by using gum acacia
(GA) as a carrier during spray drying: the viability of GA-treated bacteria was 100-fold higher
than the control following exposure for 120 min to porcine gastric juice at 37°C (Desmond et
al., 2002). Such protection may be reliant on the resistance of GA to digestion (Arslan et al.,
2015). As well as GA, it has also been suggested that a dairy matrix might act as a buffering
agent, thereby protecting ingested bacteria during transit through the upper gastrointestinal
tract (Würth et al., 2015). The anionic polysaccharide alginate, and the cationic polysaccharide
chitosan, are also known for their biocompatible resistance to digestion and their mucoadhesive
properties (Kim et al., 2014). Rajam et al. (2012) showed that a combination of sodium alginate
and denatured WPI as the drying medium could improve the survival rate and controlled core
release behavior of spray-dried Lb. plantarum during simulated acidic and bile conditions. By
comparison with sodium alginate, chitosan was reported to achieve better protection of
Bifidobacterium breve during fluid-bed drying, and of Lb. acidophilus during simulated
digestion (Cook et al., 2011; Yonekura et al., 2014). This could be explained by the attractive
electrostatic interactions between chitosan (cationic polysaccharide) and bacterial cells
(negative surface charge) (Anselmo et al., 2016).

To date, only a few studies have been performed in vivo. It has been shown that milk-
protein-based microcapsules display a protective effect on probiotics during simulated in vitro
digestion, but not in the mouse gastrointestinal system (in vivo) (Würth et al., 2015). Therefore,
the results of simple in vitro assays cannot be extrapolated directly to living organisms. It is
problematic to perform numerous digestion trials in the context of ethically and technically
challenging animal and human studies. The novel dynamic gastrointestinal models now
available, which can better reflect the complex digestive systems of living organisms, now need
to be applied to testing the behavior during digestion and efficiency of delivery of probiotic
products (Chen et al., 2016; Cordonnier et al., 2015).

[Link] Adhesion to epithelial cells

The importance of the adhesion of probiotics to epithelial cells or the intestinal mucus has
been well documented. However, very few publications have reported the effects of spray
 Chapter 1: Literature review 46

drying on the adhesion capacity of probiotics to intestinal epithelial cells. Lb. plantarum 83114
and Lactobacillus kefir 8321 did not lose their ability to adhere to intestinal Caco-2/TC-7 cells,
while Lb. kefir 8348 displayed a significant loss of adhesion capacity after spray drying under
the same conditions (Golowczyc et al., 2011). This result is in agreement with the sensitivity
of bacterial strains to spray drying. Lb. kefir 8348 had the poorest survival after spray drying
(~30%), while the survival of Lb. plantarum 83114 and Lb. kefir 8321 was between 80% and
90%.

Improved survival (approximately higher than 50%) may guarantee the retention of the
adhesion capacity of probiotics because of the insignificant injury of cell surface structures.
Further, probiotics adhesion capacity may also be improved after spray drying by using
prebiotics in the drying medium (Brink et al., 2006).

[Link] Immunomodulation ability

Immunomodulation is one of the most important probiotic functionalities and has been
extensively investigated during the past decade. However, there have been very few studies
reporting the effects of spray drying on the efficacy of probiotic immunomodulation. The
effects of air drying, freeze-drying and spray drying on the immunomodulation ability of the
probiotics Lb. plantarum CNRZ 1997 and Lactobacillus zeae CNRZ 2268 and B. bifidum CIP
56.7 were compared with peripheral blood mononuclear cells (PBMC) in vitro. The probiotic
functionality was not directly linked to cell survival, and displayed strain-dependent sensitivity
to each of the drying methods. After different drying processes, the probiotic powders might
positively or negatively modify the bacterial immunomodulation capacity, and spray drying
appeared to be the best drying process because of its effect on decreasing the production of
PBMC IL-12 (Iaconelli et al., 2015). Furthermore, during an in vivo study performed with Lb.
acidophilus A9, Lb. paracasei A13 and Lb. casei Nad, spray drying was carried out using 20%
(w/v) skimmed milk as the drying medium, and the probiotic powders were administered to
mice for 5 and 10 days. A significantly higher number of Immunoglobulin A (IgA)-producing
cells in the small intestine were induced by spray-dried cultures when compared with fresh
cultures (Páez et al., 2013).

Immunomodulation by probiotics is mainly due to molecular interactions between the

surfaces of the bacterial cell and host cell (Bron et al., 2011). The spray-dried probiotics are
encapsulated by the drying medium. The barrier effect of the drying medium may therefore
 Chapter 1: Literature review 47

modify the interaction between the microbes and the host, and this effect is probably dependent
on the digestion behavior of the matrix. The advantage of spray drying probiotics may be linked
to the barrier effect of the drying medium in such way it can protect the bacterial surface
structure from the digestion process, including pH-induced conformation changes, enzyme
catalysis and the shearing force of gastrointestinal tracts. However, severe cell damage caused
by spray drying may also lead to a diminished adhesion capacity of bacteria, heat or osmotic-
induced conformation changes to the surface structure or a shield effect of the drying medium.
However, this remains a hypothesis to be addressed by future research.
 Chapter 1: Literature review 48

1.4 Context, aim and innovation of the PhD Project

1.4.1 Context of the PhD project

The food and pharmaceutical industries are currently faced to a new opportunity with
respect to the drying of probiotics by spray drying. By far, it is clear that the final quality of
spray dried probiotic powders is determined by several factors ranging from bacteria
preparation to final product application (Figure 10).

Figure 10. The most influential factors and adverse stresses experienced during the growth, spray
drying, storage and application of probiotics (Adapted from Huang et al., 2017).

In this thesis, we firstly introduce a novel simplified process used for spray drying of
probiotics (Chapter 2). The advantages of this process will be described from both
microbiological and processing perspectives. This novel process results in a high bacterial
viability in final spray-dried powders. The underlying mechanisms of the viability
improvement are then investigated and presented (Chapter 3). The scale-up feasibility of the
novel process is validated in a semi industrial scale dryer (Chapter 4). This novel process is
further coupled with a multi-stage drying process to present an industry-friendly approach for
production of high quality probiotic powders.
 Chapter 1: Literature review 49

1.4.2 Aim of the PhD project

The aim of this PhD project is to explore an industry-friendly process for spray drying of
high quality probiotic powders. The “quality” should not be restricted to the viability/survival
of probiotics in the final powders, but also their remaining activity and functionality. To
achieve this aim, it is necessary to have a better understanding of the influences of process
conditions (including growth, drying and storage conditions) and material properties (mainly
including growth and drying media) on the stress tolerance physiology of probiotics throughout
the whole process (growth-drying-storage).

Through this project, we attempt to solve the following pending issues in the field of spray
drying probiotics:

- Unsatisfactory bacterial viability after spray drying

- Complex operation steps in comparison to freeze drying
- High cost of effective protective agents when scale-up drying process
- Possibility of probiotic functionality loss

With the results presented in this thesis, we hope to add the value of spray drying in terms
of probiotic production when comparing to freeze drying.
 Chapter 1: Literature review 50

1.4.3 Innovation of the PhD project

Involvement of multidisciplinary knowledges is required when tackling with increasingly

scientific questions and technical challenges in spray drying of probiotics. This project has
benefited from the first joint PhD project between Soochow University (Suzhou, China) and
Agrocampus Ouest (Rennes, France). The collaboration between chemical engineering lab
(China) and UMR STLO lab (Processing & Microbiology teams, France) provided access to
more instruments and analytical tools, and also allowed us to understand fundamental and
experimental issues with a broader view.

As reviewed earlier in this chapter, different strategies to improving bacterial viability

during spray drying have been investigated to date. However, these studies are mostly limited
in the pre-drying unit (bacterial growth, stress response physiology, etc.) or the drying unit
(protective agents, drying parameters, etc.). In comparison with previous studies, the
innovation of this project mainly concludes:

- This project proposed a strategy with the merits throughout the entire production
process, particularly targeting at high bacterial growth before spray drying and high
remaining probiotic viability in the final powders
- The operation steps were simplified compared to the conventional spray drying
process in terms of probiotic production
- The mechanisms were investigated from different aspects including bacterial stress
response physiology, droplet drying behavior, droplet-particle transition, etc.
- The cost efficiency was targeted both from processing and material perspectives and
the scale-up feasibility was validated in a semi-industrial scale experiment
- Apart from the bacterial viability, the storage stability and digestion resistance of
probiotic powders were investigated.
 Chapter 2: A novel 2-in-1 process 51

Chapter 2. A novel 2-in-1 Process: Double Use

of Highly Concentrated Sweet Whey

In this chapter, a novel process is proposed in order to produce probiotics simply and
sustainably via spray drying. In this novel and now patented process, the highly concentrated
sweet whey was first used as a 2-in-1 medium for growth and spray drying of probiotics.
Probiotic strain Lactobacillus casei BL23 and Propionibacterium freudenreichii CIRM-
BIA129 were used as test strains. Compared to the usual process, the intermediate operation
steps from growth to spray drying (e.g. cell washing, harvesting and re-suspending) were
avoided in this novel process. Meanwhile, probiotic metabolites could be retained in the final
powders as the food-grade feature of the culture medium.

To achieve such a 2-in-1 process, the growth of bacteria is a prerequisite: given the
hypertonic condition in highly concentrated sweet whey, “Fight or Flight” of bacteria thus
become a determining factor with regard to the feasibility of the process.

The total solids (TS) of the sweet whey medium was increased from 5% to 40%, and the
optimal range of TS was explored in relation to its effects on biomass production after growth
and viability after spray drying and during storage. The physiology of bacteria under hypertonic
conditions in the highly concentrated sweet whey was shown to drastically influence both the
resistance to drying and the final quality of probiotic powders.

The main contents in this Chapter have been published1 or prepared2 as:
1
Method for preparing a probiotic powder using a two-in-one whey-containing nutrient medium.
Patent application filed in Europe on 21 September 2015, EP no. 15 306465.4. Inventors:
Jeantet, R., Huang, S., Jan, G., Schuck, P., Le Loir, Y., Chen, X.D.
1
Huang, S., Cauty, C., Dolivet, A., Le Loir, Y., Chen, X.D., Schuck, P., Jan, G., Jeantet, R., 2016a.
Double use of highly concentrated sweet whey to improve the biomass production and
viability of spray-dried probiotic bacteria. J. Funct. Foods 23, 453–463.
2
Huang, S., Rabah, H., Gaucher, Le Loir, Y., Schuck, P., Chen, X.D., Jeantet, R., Jan, G., Long term
survival of Lactobacillus casei BL23 in sweet whey medium.
 Chapter 2: A novel 2-in-1 process 52

2.1 Introduction

Why do we need a 2-in-1 drying process?

As reviewed in Chapter 1, the challenge of utilizing spray drying to produce probiotic

powders is mainly related to the use of high temperatures during the process. Extensive
investigations have been carried out in order to improve the remaining viability of probiotic
bacteria during spray drying and subsequent storage, in which the strategies mainly include
enhancement of cellular resistance, addition of protectants to the drying medium and process
optimization. However, these strategies have usually focused only on the drying stage, and
rarely considered the overall process from growth to drying of bacteria.

Standard laboratory culture media (e.g. the most frequently used de Man, Rogosa and
Sharpe–MRS broth) have been used to grow bacteria in most studies (Table 3). Since they are
non food-grade, the subsequent operations required additional harvesting, rinsing and
re-suspension of bacteria before reaching the drying unit (see Figure 11). In addition to the
waste of material and beneficial probiotic metabolites, the removal of these culture media
may lead to the loss of bacteria viability and increase the risk of contamination during
rinsing, centrifugation and re-suspension operations. The possible residual components
from the culture media on the bacteria pellets may also interfere with subsequent
operations and further applications.

A B

Figure 11. Diagrams of (A) the conventional process used in spray drying of probiotic bacteria and (B)
the innovative 2-in-1 drying process.
 Chapter 2: A novel 2-in-1 process 53

Why did we use sweet whey?

A proper food-grade medium is a key factor to achieve the one-step spray drying without
intermediate operations.

The food-grade culture media based on MRS medium have been proposed for biomass
production of probiotic bacteria (Hwang et al., 2012; Lavari et al., 2014; Sawatari et al., 2006,
2006). However, probably due to the high cost of MRS, the feasibility of double using these
MRS based media for both biomass production and spray drying has rarely been reported.

Whey is an abundant food grade byproduct of the cheese industry. On a global scale,
most of the cheese production is realized in Europe and United states, leading to large
byproduction of whey (Table 6): 40 to 50% is exported from the EU, currently at a relatively
low price after a continuous increase from 2010 due to infant formula gold rush. The large
amount of lactose and whey proteins in sweet whey, obtained from the manufacturing of
rennet-type cheese, makes it an ideal medium for growing dairy bacteria. Moreover, several
reports have demonstrated that the lactose, whey proteins and calcium could have protective
effects on probiotics against adverse stresses during spray drying, storage and digestion
(Huang et al., 2014; Huang and Chen, 2013; Mattila-Sandholm et al., 2002; Mäyrä-Mäkinen
and Bigret, 1993; Picot and Lacroix, 2004; Rajam et al., 2012). In view of all of this, sweet
whey can be considered as an excellent candidate to be used in the 2-in-1 process (Figure 11B).

Table 6. The production, exportation and price of whey in the European Union. (Data source: CNIEL,
2011~2016)
Year 2010 2011 2012 2013 2014 2015

Production (×1000 tons) 1939 1976 2097 2218 2229 -

Exportation (×1000 tons) 917 973 932 917 913 888

Exported fraction (%) 47 49 44 41 41 -

Whey 669 845 911 960 884 652

a
Price (Euros/ton)
Whole milk 2716 3143 2758 3541 3067 2422

Relative whey to
25 27 33 27 29 27
whole milk price (%)
a
Data were from the quotations of milk products recorded on the internal market in France
 Chapter 2: A novel 2-in-1 process 54

Why did we increase whey concentration (total solid content)?

To achieve one-step drying from a bacterial culture, the total solid content (TS) of the
double-used medium, i.e. the concentration of sweet whey in this project, has a critical role in
the overall process and the final product. Indeed, the TS values of MRS and Yeast Extract
Sodium Lactate (YEL) broth (Malik et al., 1968), respectively the standard culture media of
Lactobacillus and Propionibacterium, are around 5% (w/w). This low value, in the same range
as that of sweet whey, has several adverse consequences when using these culture media as
raw materials in industrial spray drying: in particular, low powder flow rate, large amount of
water to evaporate and subsequent high drying temperature needed, and potential occurrence
of undesirable caking or sticking in these relatively fine powders (Stoklosa et al., 2012; Wu et
al., 2014). Therefore, considering the energy efficiency and productivity of spray drying,
the feed with high TS is generally desired by industry (Figure 12). In the aim of improving
bacterial survival, the feed with high TS also has higher potential in term of moderating
the drying condition (especially temperature).

A B

Figure 12. (A) Evaporation power (PGev, power required to evaporate water) and heating power
(PGaq, the power required to heat the drying air), (B) Total drying time and (C) Total drying energy
as a function of feed moisture content. (adapted from Kajiyama and Park, 2011)
 Chapter 2: A novel 2-in-1 process 55

Why did we choose the two probiotic strains?

In this project, the well-documented probiotics Lactobacillus casei BL23 and

Propionibacterium freudenreichii CIRM-BIA129 (also known as ITG P20) were used as the
model probiotic strains.

L. casei is an important species of lactic acid bacteria which has been widely used in dairy
production, especially as the nonstarter bacteria in cheese ripening. Besides, L. casei species
also forms the lactobacillus flora inhabiting the human mouth, intestine and vagina, which
plays significant role in human health and disease (Ahrné et al., 1998; Guarner and Malagelada,
2003). Increasing L. casei strains have been reported to confer beneficial effects on the hosts
through different action mechanisms, hence being recognized as potential probiotics used as
functional foods or adjuncts in treatment or prevention of certain diseases (Aktas et al., 2016;
Aoyagi et al., 2016; Ca et al., 1999; Kato-Kataoka et al., 2016). Lactobacillus casei BL23 is an
anti-inflammatory strain, also known to attenuate colitis (Rochat et al., 2007; Watterlot et al.,
2010). The ability of attenuating colitis of the L. casei BL23 strain was found to depend on the
dairy delivery matrix (Lee et al., 2015b), which makes it an ideal target to be encapsulated by
the sweet whey medium used in this work.

P. freudenreichii, the main species of dairy propionibacteria, is consumed in high amounts

both in traditional fermented foods and in probiotic supplements. Its promising probiotic
properties include beneficial modulation of several gut parameters (Cousin et al., 2010).
Consumption of P. freudenreichii indeed leads to increased bifidobacterial intestinal
population in humans (D. Bouglé, 1999). In an animal model of carcinogenesis, such
consumption reduces proliferation while enhancing apoptotic depletion of colon cancer cells
(Lan et al., 2008) in accordance with the pro-apoptotic effect of its major metabolites, the short
chain fatty acids (SCFAs) propionate and acetate (Cousin et al., 2012a; Jan et al., 2002; Lan et
al., 2007). This cytotoxic effect, killing cancer but not healthy human colon epithelial cells,
relies on survival and activity of P. freudenreichii in the ingested product and in the gut, key
prerequisites for in situ release of SCFAs (Cousin et al., 2016). Selected strains of P.
freudenreichii CIRM-BIA129 furthermore modulate gut inflammation through strain-specific
surface layer proteins (Foligné et al., 2013a, 2010), which induce the production of the
immunomodulatory cytokine IL-10 by human immune cells and which are expressed in
fermented dairy products (Plé et al., 2015). This anti-inflammatory effect may be further
enhanced by P. freudenreichii ability to release folic acid and the menaquinone biosynthesis
precursor 1.4-dihydroxy-2-naphthoic acid (DHNA) (Mitsuyama et al., 2007; Rossi et al., 2011).
 Chapter 2: A novel 2-in-1 process 56

Similarly with L. casei BL23, the dairy matrix was also demonstrated to display protective
effects for the probiotic potential of P. freudenreichii (Cousin et al., 2012a; Gagnaire et al.,
2015; Mitsuyama et al., 2007; Plé et al., 2015).

Altogether, these two strains were selected as model strains in this work in terms of their
potent beneficial effects and their compatibilities with the sweet whey medium.

Based on all the above considerations, sweet whey was used for the first time as a double-
used medium in a simplified one-step spray drying process in this work.
 Chapter 2: A novel 2-in-1 process 57

2.2 Materials and Methods

2.2.1 Strains and culture conditions

The probiotic strain Lactobacillus casei BL23 was kindly provided by UMR1219
MICALIS, (INRA-AgroParisTech, Jouy-En-Josas, France) and Propionibacterium
freudenreichii subsp. shermanii CIRM-BIA 129 was maintained and pre-cultured by the
CIRM-BIA Biological Resource Center (Centre International de Ressources Microbiennes-
Bactéries d'Intérêt Alimentaire, INRA, Rennes, France). L. casei was activated by inoculation
(1% inoculum size) in MRS Broth and static cultivation at 37ºC for 16 h. P. freudenreichii was
inoculated (1% inoculum size) in YEL broth and cultivated statically at 30ºC for 50 h.

2.2.2 Growth in sweet whey culture media

Sweet whey powder (Lactalis ingredients, Mayenne, France) was used to prepare the sweet
whey culture media with different levels of TS. Composition of the sweet whey powder was
analyzed by the procedures described by Gaucher et al. (2008) (Table 7).

Table 7. Physical and chemical analysis of the sweet whey powder (composition expressed as w/w%).
Total solids (%) 94.81
pH 6.52
Total nitrogen (%) 11.81
Non-protein nitrogen (%) 3.11
Lactose (%) 67.91
Ash (%) 6.85
Calcium (%) 0.37
Magnesium (%) 0.11
Sodium (%) 0.59
Potassium (%) 2.27
Chloride (%) 1.72
Phosphate (%) 1.14
Citrate (%) 2.26

The sweet whey powder was dissolved in deionized water to obtain culture media with final
total solids content (TS, w/w) of 5%, 10%, 20%, 30% and 40%, respectively. The osmolality
of the medium (Figure 13) was measured with a freezing-point osmometer (Osmomat 030-D,
Gonotec, Berlin, Germany) calibrated with standard solutions.
 Chapter 2: A novel 2-in-1 process 58

3500

3000

Osmolality (m OSmol/kg)
2500

2000

1500

1000

500

0
MRS YEL 5% 10% 20% 30% 40%
Different TS (w/w %) of sweet whey and standard lab media

Figure 13. Osmolality of sweet whey media and standard lab media (MRS and YEL) as a function of
the TS (w/w %)

The culture media with casein peptone were prepared by adding casein peptone plus
(Organotechnie, France) to the above sweet whey culture media at a concentration of 0.5%
w/w (Cousin et al., 2012b). All of these culture media were autoclaved at 100 ºC for 30 min
before inoculation of probiotic bacteria. L. casei was inoculated at 1% inoculum size in the
different sweet whey culture media (i.e. with different TS of sweet whey with/without casein
peptone) with the MRS preculture. The inoculated culture media were incubated statically at
37 ºC for 48 h. P. freudenreichii was inoculated in the sweet whey culture media using a YEL
preculture and incubated statically at 30 ºC for 120 h (5 days).

2.2.3 Monitoring of bacterial growth and acidification curves

The growth curves and pH changes of L. casei were monitored for the cultures grown in
the MRS broth, 5% sweet whey with casein peptone supplementation and 30% sweet whey
without casei peptone supplementation. The growth curves were expressed by the bacterial
population in the culture. Bacterial population was determined by CFU counting on MRS agar
plates. The L. casei culture was diluted serially (1 mL to 9 mL) in peptone water (0.1% w/v)
prior to pouring onto MRS agar plates. The agar plates were then incubated at 37ºC for 48 h
(aerobic conditions) before counting the bacterial colonies.
 Chapter 2: A novel 2-in-1 process 59

2.2.4 Spray drying

The probiotic cultures grown in the casein peptone-supplemented culture media were used
for spray drying. In addition, further samples were prepared by growing bacteria in 5% casein
peptone-supplemented medium but increasing the TS value to 30% by adding the sweet whey
powder in order to dry these bacteria cultures directly.

Spray drying was carried out using a pilot-scale Mobile MinorTM spray dryer (GEA Niro
A/S, Denmark) at the maximum evaporation rate of 5 kg water h-1. A two-fluid spray nozzle
with an orifice diameter of 0.8 mm was used coupled to a peristaltic pump (Watson-Marlow,
France) for feeding and atomization. The absolute humidity of the inlet air was maintained at
around 1g kg-1 of water in the air by a dehumidifier (Munters, Sweden). Spray drying
parameters were monitored by SD2P® software (Schuck et al., 2009). All the sample-contact
parts of the dryer, including the nozzle, chamber etc., were washed with hot water at 90ºC. Inlet
temperature of 200ºC was used to dry the dryer and inactivate any possible contaminating
microorganisms 2 h before drying the probiotic culture.

All the culture media (1 L) were agitated moderately for 10 min before spray drying. In
view of the intrinsic resistance of both strains, the inlet temperature of spray drying was set at
140ºC for L. casei (fragile strain) and 180ºC for P. freudenreichii (robust strain). The outlet
temperature was 63 ± 2ºC for L. casei and 73 ± 2ºC for P. freudenreichii, and the relative
humidity of outlet air was maintained at 10 ± 1% by adjusting the feed rate. The water content
and water activity of the powders were tested according to the methods described by Schuck,
Dolivet, & Jeantet (2012).

2.2.5 Size distribution measurement

The size distribution of powders was measured by laser light scattering with a MasterSizer
2000 equipped with a 5-mW helium-neon laser (Malvern Instruments, Malvern, UK).The dry
powder feeder attachment was coupled to the standard optical model presentation for
dispersion of powders in the air. The D0.5 and span parameters were used to characterize the
size distribution of powders, D0.5 being the maximum diameter of 50% of the particles, and the
span of powders being calculated as:
𝑫𝟎.𝟗 −𝑫𝟎.𝟏
𝑺𝒑𝒂𝒏 = (Eq 7)
𝑫𝟎.𝟓
 Chapter 2: A novel 2-in-1 process 60

where D0.9 and D0.1 are the maximum diameters of 90% and 10% of the particles,
respectively.

The result represented the mean of two independent samplings, each measured three times
successively.

2.2.6 Scanning electron microscopy

The [Link] powders from culture media with 5% and 30% TS were fixed on carbon tape
and then sputter-coated with gold-palladium. These powder samples were observed by
scanning electron microscopy (JSM 7100F. JEOL. Tokyo. Japan) at 5 kV.

2.2.7 Storage

The powders were collected and sealed in sterilized polystyrene bottles (Gosselin, France),
stored at a controlled temperature of 4ºC and kept away from light. The samples were analyzed
at 30 day intervals for 120 days (4 months).

2.2.8 Enumeration and quantification of bacteria viability

The number of viable cells was first measured after growth (i.e. before spray drying) by
colony forming unit (CFU) counting as follows. After 10 min agitation, the bacteria culture
was diluted serially (1 mL to 9 mL) in peptone water (0.1% w/v). After spray drying or during
storage, the powder samples were rehydrated by dissolving 1 g of powder in 9 mL of peptone
water before serial dilutions. Each diluted sample of L. casei was poured onto MRS agar and
incubated at 37ºC for 48 h (aerobic conditions). P. freudenreichii dilutions were poured into
YEL agar and incubated at 30ºC for 6 days under anaerobic conditions (Anaerocult®, Merck
KgaA, Germany).

The dependency of bacteria final population on the casein peptone supplementation was
calculated according to:
𝑁+ −𝑁−
𝐷𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑐𝑦 = (Eq 8)
𝑁−

where N+ is the bacteria population (CFU/mL) in the medium with the casein peptone
supplementation, and N- the bacteria population without casein peptone supplementation.
 Chapter 2: A novel 2-in-1 process 61

The survival of bacteria after spray drying (%) was calculated according to:
𝑁𝑑
𝑆𝑢𝑟𝑣𝑖𝑣𝑎𝑙 = × 100 (Eq 9)
𝑁0

where Nd refers to the bacteria population (CFU/g) in powders after spray drying. The initial
population N0 (CFU/g) was calculated from N+ and the total solid content (TS) of medium as
follows:
1−𝑇𝑆
𝑁0 = 𝑁+ × (Eq 10)
𝑇𝑆

The log reduction of bacteria after 120 days’ storage was calculated as:

𝐿𝑜𝑔⁡𝑟𝑒𝑑𝑢𝑐𝑡𝑖𝑜𝑛 = 𝐿𝑜𝑔⁡(𝑁𝑑 ) − 𝐿𝑜𝑔⁡(𝑁120 ) (Eq 11)

where N120 is the bacteria population in the powders following storage for 120 days.

2.2.9 Statistical analysis

All the experiments were repeated at least three times. The results are presented as mean
value with standard error. Significant differences (p < 0.05) between the mean values were
determined by Tukey’s test. The statistical analysis was carried out using R 3.2.1 with the
‘Rcmdr’ package (R Development Core Team).
 Chapter 2: A novel 2-in-1 process 62

2.3 Results and Discussion

2.3.1 Effect of sweet whey concentration on bacterial biomass production

The growth of bacteria was compared in sweet whey culture media at different TS, with or
without casein peptone supplementation. As shown in Figure 14A, the final population of L.
casei in sweet whey culture media without casein peptone increased as TS increased from 5%
to 40%. Moreover, casein peptone supplementation significantly enhanced the final L. casei
population at 5% to 30% TS, but not that at 40%. In these casein peptone supplemented culture
media, a final L. casei population of approximately 2×109 CFU mL-1 was reached for culture
media with 20%, 30% TS, and 1×109 CFU mL-1 for that in MRS broth. The dependency of L.
casei final population on casein peptone decreased when the TS of sweet whey increased.

Propionibacterium freudenreichii subsp. shermanii CIRM-BIA 129, used in this study, is

able to use various substrates as carbon and energy sources for growth. Carbohydrates,
including sucrose, lactose, glucose, galactose, inositol, erythritol, adonitol, and esculine, may
be used (Loux et al., 2015), as well as amino acids, glycerol, or its preferred carbon source,
which is lactate (contained in the reference YEL growth medium). Within dairy products, it
can therefore utilize either lactose, the major milk carbohydrate, or lactate, resulting from lactic
acid fermentation. Sweet whey and acid whey, which may contain either lactose or lactate or a
mixture thereof, should thus sustain P. freudenreichii growth. The final population of P.
freudenreichii also showed a similar trend to that of L. casei when the TS of sweet whey
increased (Figure 14B). Specifically, the population increased as the TS increased from 5% to
30%, but with a considerable reduction at 40%. The optimal TS values also ranged from 20%
to 30%, with larger populations than those obtained in YEL broth (~ 1.0×109 CFU mL-1). The
casein peptone-supplemented medium with 30% TS produced the highest population of P.
freudenreichii (~ 2.5×109 CFU mL-1) among all the culture media tested in this study. However,
compared to L. casei, the final population of P. freudenreichii in sweet whey culture media
was less influenced by the casein peptone supplementation. The maximum improvement was
obtained when adding casein peptone to sweet whey with 20% TS. However, the dependency
was around 1.0, which means that the P. freudenreichii population was only doubled after
supplementation with casein peptone.
 Chapter 2: A novel 2-in-1 process 63

A
1 E+10 * 40
L. casei populations (CFU mL-1)
** **
*** ***
c c 30
bc
ab

Dependency
1 E+09 ab
ab ab ab 20

a 10
1 E+08 a
0
a
1 E+07 -10
5 10 20 30 40 MRS
Total solid content (w/w%) in sweet whey media

without casein peptone with casein peptone

dependency on casein peptone

B
4 E+09 *** 1,5
**
P. freudenreichii populations (CFU mL-1)

* * *
bc c 1,0
3 E+09

Dependency
bc
0,5
2 E+09
ab
0,0
a a
a
1 E+09
a a a -0,5
a
0 E+00 -1,0
5 10 20 30 40 YEL
Total solid content (w/w%) in sweet whey media

without casein peptone with casein peptone

dependency on casein peptone

Figure 14. Final bacteria populations (left Y axis) and the dependency (right Y axis) of bacteria final
population on casein peptone of (A) L. casei 48-h culture and (B) P. freudenreichii 120-h culture in the
sweet whey culture media with different TS, with or without casein peptone supplementation (compared
to the standard culture media, i.e. MRS broth for L. casei and YEL broth for P. freudenreichii).
Mean ± SEM, n = 6. Different letters mean significant difference between bacteria populations
(p < 0.05). Different number of * means significant difference in dependency (p < 0.05).

The results indicate that increasing the TS of sweet whey to values ranging between 20%
and 30% was able to improve the biomass production of both probiotic strains. As mentioned
above, these higher TS values would be beneficial for subsequent spray drying. The
improvement in the final bacteria populations may result from the richer nutrients in the culture
media with higher TS values. Although the osmotic pressure was also higher in these culture
 Chapter 2: A novel 2-in-1 process 64

media, the time of growth may have been long enough to trigger the stress response of these
two strains to adapt to the high-osmolality environments (Wood, 2011). For example, it has
been reported that these two strains are able to accumulate intracellular polyphosphate which
is related to the improvement in bacterial stress tolerance (Alcántara et al., 2014). The presence
of large amounts of phosphate in the culture media with 20% or 30% TS may facilitate the
accumulation of polyphosphate by uptake of phosphate from the extracellular environment.
However, when increasing the TS of sweet whey to 40%, the inhibition effect caused by the
extreme osmolality and related energy consumption in osmoregulation began to be deleterious
for the growth of both strains.
 Chapter 2: A novel 2-in-1 process 65

2.3.2 Long-term survival of [Link] BL23 during growth in sweet whey media

The growth curves of L. casei was monitor and expressed by the bacterial population in
Figure 15. They were significantly different in sweet whey medium from that in MRS broth,
either at 5% or 30% TS. The bacterial population in MRS peaked at 12 h with a population
around 2.5×109 CFU mL-1, then maintained constant for around 1 day prior to a gradient drop
from 109 to 103 CFU mL-1 during the following 4 days (i.e. 105 CFU drop from 48 h to 192 h).
Interestingly, the alive bacterial population in sweet whey media maintained constant after
peaking at 12 h for 8 days, albeit the highest population is around 1 log lower than that of MRS
culture (~ 5×108 CFU mL-1). Moreover, this long term survival of bacteria existed both in
isotonic (5% TS) and hypertonic (30% TS) sweet whey.

1 E+10
[Link] population (CFU mL-1)

1 E+09
1 E+08
1 E+07
1 E+06
1 E+05
1 E+04
1 E+03
0 24 48 72 96 120 144 168 192 216
Growth time (h)
MRS 5% SW 30% SW

Figure 15. Growth curve (CFU) of L. casei BL23 in MRS broth, 5% sweet whey with casein peptone
supplementation and 30% sweet whey without casein peptone supplementation.

Long term survival of food-grade bacteria within the food matrix is of interest from both
practical and scientific perspectives, especially for the probiotics with beneficial claims.
Although we cannot propose a confirmed mechanisms for the long term survival of L. casei in
sweet whey medium, we can offer a number of hypotheses.

Growth in MRS resulted in a higher bacterial population (~ 1 log more). It indicated that
nutrients in MRS may be consumed completely, leading to starvation stress on bacteria in the
late stationary phase. Similarly, the rapid fermentation in MRS may lead to faster and harsher
acid stress towards L. casei than in sweet whey medium (Figure 16). This limited acid stress in
 Chapter 2: A novel 2-in-1 process 66

sweet whey medium can also be caused by the buffering capacity of the whey proteins and
phosphate salts (Salaün et al., 2005). Overall, the severe starvation and acid stress may be the
reason of bacterial inactivation in MRS culture in the late stationary phase.

6
pH

3
0 4 8 12 16 20 24 28 32 36 40 44 48
Growth time (h)
MRS 5% SW 30% SW

Figure 16. Acidification curve of L. casei BL23 in MRS broth, 5% sweet whey with casein peptone
supplementation and 30% sweet whey without casein peptone supplementation.

Apart from the stress-induced bacterial death in MRS, it is also possible that some elements
in sweet whey medium suppress those lethal factors. The buffering capacity of sweet whey
medium on the decreasing of pH may affect the autolysis of L. casei to some extent (Kang et
al., 1998; Martínez-Cuesta et al., 1997), that is although an important matter for cheese ripening.
Besides, the lower bacterial population in sweet whey culture could also avoid quorum-
sensing-dependent programmed cell death which may emerge in the MRS culture when the
population reached up to 2.5×109 CFU mL-1 (Heurlier et al., 2005; You et al., 2004). However,
these hypotheses warrant further investigation.
 Chapter 2: A novel 2-in-1 process 67

2.3.3 Improved probiotic viability in spray drying application

The remaining viability of probiotics was expressed as the population and as the percentage
of survival of bacteria after spray drying (Figure 17). The water content and water activity
values of all powders were 6±1% and 0.2 ± 0.05, respectively.

A
3,0E+09 100
L. casei populations in powders

Survival after spray drying (%)

*** ***
** c
2,5E+09 80
*
2,0E+09 *
(CFU g-1)

bc 60
1,5E+09
40
1,0E+09
ab 20
5,0E+08 a a
0,0E+00 0
5 10 20 30 40
Total solid content (w/w%) in sweet whey media

bacteria population bacteria survival

B
1,0E+11 ** 100
P. freudenreichii populations in

Survival after spray drying (%)

* ab **
* a
a 80
1,0E+10
powders (CFU g-1)

a
***
60
1,0E+09
40
1,0E+08 c
20

1,0E+07 0
5 10 20 30 40
Total solid content (w/w%) in sweet whey media

bacteria population bacteria survival

Figure 17. Remaining bacteria populations (left Y axis) and survival (right Y axis) of (A) L. casei and
(B) P. freudenreichii in the sweet whey culture media with different TS after spray drying. Mean ±
SEM, n = 3. Different letters mean significant difference between bacteria populations (p < 0.05).
Different number of * means significant difference between survival rates (p < 0.05).
 Chapter 2: A novel 2-in-1 process 68

The survival of L. casei in powders after spray drying increased with TS from 5% to 40%,
with a maximum survival of 40% for TS values of 30% and 40%. In comparison with the
powder from the 5% TS medium, the survival in the powder from the 30% TS medium was
improved approximately 60-fold (~0.6% survival for 5% TS and ~40% survival for 30% TS).
The population of L. casei in powders also increased as the TS increased from 5% to 30%, but
decreased at 40% TS due to the relatively low L. casei population in the 40% TS culture before
drying. The maximum survival rates for the P. freudenreichii strain (ca. 70%) after spray drying
occurred in the culture media with 20% and 30% TS. Similarly to L. casei, the low TS culture
resulted in lower P. freudenreichii survival close to 40% after spray drying. The populations
of viable bacteria in the powders from the 5%, 10%, 20% and 30% TS cultures were all above
109 CFU g-1. The highest population was obtained for the 20% TS culture, which reached 1010
CFU g-1. The P. freudenreichii growing in 40% TS culture displayed the lowest survival and
remaining population after spray drying.

When comparing the two probiotic strains, the survival of P. freudenreichii was generally
higher than that of L. casei after spray drying, despite the higher drying temperature used. This
suggested that the P. freudenreichii strain was more tolerant to spray drying stress than the L.
casei strain. It is known that P. freudenreichii is generally a heat resistant species which is
often used for making Emmental-type cheeses in which the curd is heated at 50 ~ 55ºC. It has
been reported that P. freudenreichii accumulates intracellular trehalose and glycogen in
addition to polyphosphate, which are involved in improving bacterial tolerance of heat and/or
desiccation stress (Falentin et al., 2010b).

In relation to both the survival and final viable population of the probiotics in powders, the
optimal TS value of sweet whey was between 20% and 30%. In this range, both strains also
presented optimal biomass production. The improvement in survival may have resulted from
the robust cellular tolerance induced by higher osmolality in the 20% to 30% TS culture media.
However, it may also have been due to the higher levels of dry matter (i.e. TS values). Indeed,
relatively fine droplets might be formed from low TS culture media due to their lower viscosity
and solid content, leading to shorter evaporation time and potentially to longer residence time
of the resulting particles within the dryer (Jeantet et al., 2008). The consequence would be a
longer exposure to higher temperature.

Moreover, a lower solid content in the medium also indicates that less wall material can be
used to encapsulate the bacteria cells. In other words, the bacteria in the lower TS medium may
be more exposed to the hot air (Perdana et al., 2014).
 Chapter 2: A novel 2-in-1 process 69

Lower solid content in the feed seems therefore to be disadvantageous for production of
powders with live probiotic bacteria. This can also be deduced from the particle size
distribution (PSD) of the powders (Figure 18 and Table 8). For both strains, the powder PSD
right shifted as the TS in the culture media increased. This result was concordant with the
observation by scanning electron microscopy, shown in Figure 19. Powders dried from high TS
(30%) culture had a wider size distribution (Span > 2) coming, with D0.5 larger than 10 µm.
Powders from 5% TS culture were finer, with a narrow size distribution (Span between 1 and
1.5) peaking below 10 µm. However, it was difficult to find the bacteria cells on the surface of
powders in the SEM observation, even for the powders obtained from the medium with 5% TS.
This indicated that the 5%TS sweet whey was already sufficient for full microencapsulation of
bacteria during spray [Link] coincided with the results for spray dried powders of
Lactobacillus plantarum with trehalose as drying medium (Perdana et al., 2014).

A
12
10
Volume (%)

8 5%
6 10%
4 20%
2 30%
0 40%
1 10 100
Particle size (µm)

B
10
8
Volume (%)

5%
6
10%
4
20%
2 30%
0 40%
1 10 100
Particle size (µm)

Figure 18. Particle size volume of spray dried powders of (A) L. casei and (B) P. freudenreichii from
sweet whey culture media with different TS. Curves were obtained from the average values of duplicate
experiments. The size range was 0.1 to 2000 µm; the volume values in the range 0.1 to 1 µm and 200
to 2000 µm were all 0.
 Chapter 2: A novel 2-in-1 process 70

Table 8. Size distribution of spray dried powders from culture media with different TS content.
5% 10% 20% 30% 40%
L. casei
D0.5 (m) 6.1 ± 0.1 6.9 ± 0.1 8.5 ± 0.1 15.3 ± 0.5 17.1 ± 0.1
Span 1.1 ± 0.1 1.3 ± 0.1 1.5 ± 0.1 2.8 ± 0.1 2.8 ± 0.1
P. freudenreichii
D0.5 (m) 7.2 ± 0.1 8.2 ± 0.1 10.0 ± 0.1 10.7 ± 0.1 17.6 ± 0.1
Span 1.4 ± 0.1 1.5 ± 0.1 2.2 ± 0.1 2.0 ± 0.1 2.6 ± 0.1
Size distribution values are significantly different from each other in lines (p < 0·05).

A B

C D

Figure 19. SEM observation of spray dried powders of L. casei obtained from (A,B) sweet whey culture
media with 5% TS and (C,D) sweet whey culture media with 30% TS. D: the size of the object is 5×
bigger than in B (Image with same magnification as B inserted in D).
 Chapter 2: A novel 2-in-1 process 71

In order to investigate the contribution of increased TS on the improved survival of

probiotics after spray drying, a drying experiment was conducted by growing bacteria in 5%
TS casein peptone-supplemented medium but increasing the TS to 30% immediately before
drying (Figure 20). After increasing the TS, the survival of L. casei following spray drying was
improved to around 6% compared to that of growing and drying at 5% TS. However, it was
still significantly lower than that of growing and drying at 30% TS. In contrast, the survival of
P. freudenreichii decreased slightly (from 45% to 26%) when TS was increased from 5% to
30% before drying, while growth in 30% TS significantly enhanced spray drying survival to
70%.

80 ****
Survival after spray drying (%)

70
60
*** ***
50
40 ***
30
20
**
10 *
0
L. casei P. freudenreichii
Probiotic strains
growth with 5% TS
growth with 5% + adding 25% TS just before drying
growth with 30% TS

Figure 20. Survival of bacteria grown and spray dried in 5% medium, grown in 5% medium but 30%
added before drying, grown and spray dried in 30% medium. Mean ± SEM, n = 3. Different number of
* means significant difference between survival rates (p < 0.05).

These results indicate that the improvement in probiotic survival after spray drying was not
a direct consequence of higher TS levels (thus lower water content) at the time of drying. The
enhanced intrinsic bacterial tolerance triggered by high osmolality during growth may play a
decisive role in the enhanced survival following spray drying. It has been reported that salt-
adaptation of Lactobacillus paracasei with high osmolality before spray drying (by 0.3 M NaCl)
can improve bacteria survival following heat treatment and spray drying (Desmond et al., 2001).
In our study, the two bacteria strains were exposed to high osmolality during the growth time
(2 days for L. casei and 5 days for P. freudenreichii). It is known that reaching cellular
 Chapter 2: A novel 2-in-1 process 72

homeostasis by accumulation of compatible solutes is an energy-dependent bacterial

osmoregulation process (Papadimitriou et al., 2016; Wood, 2011). In other words, an induction
time is required for bacteria to adapt to the adverse environments and trigger the stress response.
In contrast to bacteria growing in normal osmolality but exposed to osmolality before drying,
growing bacteria in a medium with high osmolality may have enough time to accumulate the
compatible intracellular solutes either by synthesis de novo or by uptake from the environment.
 Chapter 2: A novel 2-in-1 process 73

2.3.4 Improved storage stability of probiotic powders

The changes in bacteria population in powders were monitored during storage at 4ºC for
120 days (Figure 21). The viability of both L. casei and P. freudenreichii strains in the powders
dried from the 20% ~ 40% TS culture remained constant, with a maximum log reduction of
0.89. After storage for 120 days, the L. casei population in the powders from 20% and 30% TS
culture remained the highest at around 7×108 CFU g-1, while the P. freudenreichii population
was around 2×109 CFU g-1. Since the initial bacteria populations in the powders from 40% TS
culture were relatively low, the L. casei and P. freudenreichii populations were around 4×107
and 1×107 CFU g-1 after 120 days, respectively. In comparison, the viable L. casei population
in the powder from 5% medium decreased significantly over the entire storage time, with a
final population below 106 CFU g-1 at the end of storage (~ 2.8 log reduction at 120 days).
Moreover, the L. casei in the powder from 10% medium remained unchanged during the first
90 days (viability loss < 1 log), but decreased considerably after 90 days (~ 2.2 log reduction
at 120 days). The P. freudenreichii powders from the 5% and 10% culture media exhibited
very steady viability during the first 60 days. Although there was subsequently reduction in
viability after storage (~1.5 and 1 log reduction for 5% and 10%, respectively), the final viable
bacteria population still remained above 108 CFU g-1 at the end of 120 days. The ability of P.
freudenreichii to accumulate polyphosphate, trehalose and glycogen as energy and carbon
storage compounds, as well as compatible solutes, might lead to higher survival rate throughout
long-term storage (Boyaval et al., 1999; Cardoso et al., 2007; Leverrier et al., 2004).

It was obvious that the two probiotic strains in the powders from the lower TS culture (i.e.
5% to 10%) less survived well than those in the powders from higher TS culture (20% to 40%).
Three reasons can be proposed. First, although the bacteria in powders from lower TS culture
were also encapsulated in the drying matrix (Figure 19), the lower solid content led to a thinner
encapsulating wall layer, which may not be effective enough to protect bacteria from oxidation
stress in long term storage (Elversson et al., 2003; Perdana et al., 2014). Secondly, the bacteria
in the powders from lower TS culture had lower survival levels after spray drying. This means
that bacteria in these powders may be initially more damaged from the spray drying step, and
might die upon storage if not able to recover from these damage (Ananta et al., 2005; Teixeira
et al., 1995). Thirdly, the bacteria in culture media with higher TS were more tolerant due to
the cellular response triggered by the high osmolality during growth (e.g., accumulation of
compatible intracellular solutes) (Kets et al., 1996; Selmer-Olsen et al., 1999; Sunny-Roberts
and Knorr, 2009).
 Chapter 2: A novel 2-in-1 process 74

A
1 E+10
[Link] populations in powders
1 E+09
1 E+08 *
*
(CFU g-1)

1 E+07 *
* **
1 E+06 **
**
1 E+05
0 30 60 90 120
Storage time (days)

5% 10% 20% 30% 40%

B
1 E+11
[Link] populations in

1 E+10
powders (CFU g-1)

1 E+09 * *
1 E+08 * *
1 E+07 * * * *

1 E+06
0 30 60 90 120
Storage time (days)

5% 10% 20% 30% 40%

C
3,0
Log reduction of bacteria

2,5
2,0
1,5
1,0
0,5
0,0
5 10 20 30 40
Powders dried from media with different total solid
content (w/w%)
L. casei P. freudenreichii

Figure 21. Change in bacteria populations in spray dried powders of (A) [Link] and (B)
[Link] from sweet whey culture media with different TS during storage at 4°C for 120 days.
Mean ± SEM, n = 3. (C) Final log reduction of two probiotic strains when stored for 120 days. Different
number of * means significant difference between survival rates (p<0.05)
 Chapter 2: A novel 2-in-1 process 75

2.3.5 Additional results

The bacterial population of other Lactobacillus species grown in 5% and 30% TS sweet
whey media were also tested (Table 9). Among the four species investigated, L. acidophilus, L.
plantarum and L. rhamnosus showed a higher bacterial population in 30% TS sweet whey
culture compared to the control 5% TS sweet whey culture. L. reuteri population in the 30%
TS culture was slightly lower (~0.25 log) than that in the 5% TS culture, but still with a final
bacterial population above 108 CFU mL-1. This result suggests that the 30% TS sweet whey
may be used for Lactobacillus species in a more general way.

Table 9. The bacterial population of different lactobacillus strains grown in 5% or 30% TS sweet whey
media for 48 hours.
L. acidophilus L. plantarum L. reuteri L. rhamnosus
CIRM-BIA CIRM-BIA CIRM-BIA CIRM-BIA
1674 466 522 607
5% sweet whey 7.96±0.19 8.61±0.19 8.32±0.03 9.24±0.01
(log CFU mL-1)
30% sweet whey
(log CFU mL-1) 8.38±0.10 9.00±0.21 8.07±0.10 9.31±0.10

Further, L. acidophilus CIRM-BIA 1674 cultures in 5%, 25%, 30% and 35% TS sweet
whey were submitted to spray drying (Figure 22). The 30% TS L. acidophilus culture survived
best (~15%) among all the four groups test in this study, with a survival approximately 7-folds
higher than that of the 5% TS culture. It suggests the potential genericity of the 2-in-1 process
to Lactobacillus species

20 b
Survival of L. acidophilus
after spray drying (%)

16

12
a
8
a
4 a

0
5 25 30 35
Different TS (% w/w) of sweet whey media

Figure 22. Survival of L. acidophilus CIRM-BIA 1674 grown in different TS sweet whey after spray
drying at inlet temperature 185°C and outlet temperature 75°C. Different letters above the column mean
significant difference between bacterial survivals (p < 0.05).
 Chapter 2: A novel 2-in-1 process 76

2.4 Conclusion

In this chapter, to the best of our knowledge, we proposed for the first time the feasibility
of simplifying the process of spray drying bacteria by utilizing a double-used medium with
increased total solid content.

From our results, it was clearly showed that culturing L. casei and P. freudenreichii in sweet
whey with 20% ~ 30% TS resulted in greater biomass production and sustained viability after
spray drying and following a long-term storage. Besides, it is interesting to note that the growth
of L. casei was less dependent on casein peptone supplementation in the sweet whey with high
TS. More remarkably, sweet whey medium allowed L. casei to survive in the liquid culture for
long term at 37°C.

The results add to the value of sweet whey and may indicate a new route to producing
probiotic powders with higher process efficiency and final probiotic viability. However, the
protective mechanisms of this process on probiotics during spray drying and storage warrant a
further investigation at a molecular level. Moreover, the feasibility of scaling-up this process
should also be explored on an industrial scale.
 Chapter 3: What underlies the protection 77

Chapter 3. What are the Underlying

mechanisms of the enhanced Protection on
Probiotics?

In chapter 2, we showed a 2-in-1 process for growth and drying of probiotics using highly
concentrated sweet whey. This process affords an exceptional protection on probiotics during
spray drying and storage. Therefore, we tried here to reveal the probiotic protection
mechanisms enhanced by the 2-in-1 process.

These latter were proposed and studied on three aspects: Bacterial stress response on the
one side, and specific composition and structure of drying matrix on the other side. The results
suggest that the protection on probiotics may result from the synergistic effects of these three
aspects during the process.

The main contents in this Chapter have been published as:

Huang, S., Rabah, H., Jardin, J., Briard-Bion, V., Parayre, S., Maillard, M.-B., Loir, Y.L., Chen, X.D.,
Schuck, P., Jeantet, R., Jan, G., 2016b. Hyperconcentrated sweet whey, a new culture medium
that enhances Propionibacterium freudenreichii stress tolerance. Appl. Environ. Microbiol.
82, 4641–4651.
Wang, J., Huang, S., Fu, N., Jeantet, R., Chen, X.D., 2016. Thermal aggregation of calcium-fortified
skim milk enhances probiotic protection during convective droplet drying. J. Agric. Food
Chem. 64, 6003–6010.
Yang, Y., Huang, S., Wang, J., Jan, G., Jeantet, R., Chen, X.D., 2017. Mg2+ improves the
thermotolerance of probiotic Lactobacillus rhamnosus GG, Lactobacillus casei Zhang and
Lactobacillus plantarum P-8. Lett. Appl. Microbiol. (in press)
 Chapter 3: What underlies the protection 78

3.1 Introduction

What could make bacteria stronger?

As reviewed in Chapter 1, the ability of bacteria to adapt to sub-lethal stresses may later
induce their tolerance towards the same type of stress (stress adaptation) or different types of
stresses (multistress response, also termed cross-protection). Extensive studies have shown that
such cross-protection could be induced in lactic acid bacteria and propionibacteria (Boyaval et
al., 1999; Guchte et al., 2002). Therefore, the cross-protection can be used as an effective
strategy to guarantee probiotic viability and activity. For examples, heat adaptation of bacteria
has been mostly employed to improve probiotic thermotolerance and thereby help probiotics
to better survive during spray drying (Desmond et al., 2001).

Apart from heat stress, osmotic stress (also described as dehydration stress elsewhere) is
another stressing factor for bacteria during drying. In our innovative process introduced in
Chapter 2, bacteria face osmotic stress during growth in the highly concentrated sweet whey
culture (see the osmolality of 30% sweet whey, Figure 13, 2.2.2). However, the large final
population (Figure 14, 2.3.1) suggests that this osmotic stress is probably at a sub-lethal level
for the two bacterial strains used in this work. In other words, bacteria adapt to the osmotic
stress during growth in 30% sweet whey, instead of being inhibited by this latter. This osmotic
adaptation per se has been reported to protect bacteria during drying (Prasad et al., 2003). It
could also induce multistress tolerance, i.e. cross-protection, against different stresses existing
in spray drying process (e.g. heat and oxidative stresses) (Wood, 2011). The investigations
aiming at confirmation of these hypotheses were carried out with proteomic analysis,
enzymatic and microscopic techniques in this Chapter.

Why did we study aggregation effect with Ca2+-induced aggregated milk as the model?

In the 2-in-1 process, the highly concentrated sweet whey was used as bacterial growth and
drying medium. The sterilization step is thus necessary in order to guarantee the quality of
bacterial culture and final powders. In comparison to the natural whey, the concentrated whey
induces mineral-protein aggregation during autoclave or pasteurization due to the increasing
concentration of mineral salts and whey proteins (Havea et al., 2002; Oldfield et al., 2005).
Moreover, during fermentation of lactobacilli, the pH of whey medium will decrease to a value
close to 4.5, which will lead to even more protein aggregation within the medium prior to
 Chapter 3: What underlies the protection 79

drying step (Alting et al., 2000; Britten and Giroux, 2001). Therefore, from both practical and
scientific perspectives, it is of importance to explore the potential influence of these aggregates
on probiotic protection by the medium during drying process.

However, in order to study the independent effect of the aggregates, the influence of dry
matter content should be exclude. In other words, it is necessary to design a comparison
between the media with and without aggregates, but with same level of TS content. To this
purpose, induction of aggregation in skim milk by very low concentration of calcium salts
could be a solution due to their well-known interaction with casein micelles. Indeed, among all
the mineral salts in milk, calcium (especially free Ca2+) is known to be the most efficient one
in terms of inducing thermal aggregation of milk proteins (Gaucheron, 2005; Varunsatian et
al., 1983). Besides, calcium-fortified milk is of value from both nutrition and marketing
perspectives. Even more remarkably, the combination of calcium supplementation with
probiotics has been reported to improve both the absorption of calcium and the efficacy of
probiotics (Scholz-Ahrens et al., 2007; Trautvetter et al., 2012). On the other hand, although
the characteristics of thermal aggregation of milk proteins induced by calcium supplementation
has been extensively studied, the feasibility of applying calcium-fortified milk to
microencapsulate probiotics by drying has rarely been reported.

Therefore, we employed the single droplet drying technique to map the drying behavior of
Ca2+-induced aggregated skim milk. This latter was used to protect commercial probiotic strain
Lactobacillus rhamnosus GG during droplet-particle conversion.

Why did we focus on Mg2+?

Food-grade carbohydrates and proteins are extensively employed as protective agents to

improve bacterial viability upon heat treatment and spray drying (Table 3). However, the effect
of inorganic salts on thermotolerance of probiotics has rarely been reported. Previously, Ca2+
was found to influence the thermotolerance of lactic acid bacteria (Huang and Chen 2013).
Apart from Ca2+, Mg2+ is generally present in most of dairy products including the sweet whey
ingredients used in this project (Table 7).

In bacteria, Mg2+ is the second-most abundant cation (Romani and Scarpa 2000). The roles
of bacterial Mg2+ in homeostasis, sensing and transport were extensively investigated (mostly
in gram-negative bacteria Salmonella enterica serovar typhimurium and Escherichia coli),
including acting as a cofactor in ATP-dependent phosphorylation and a variety of other
 Chapter 3: What underlies the protection 80

enzymatic reactions, stabilizing ribosome and membranes, influencing RNA folding, the
nucleic acid-proteins interactions and bacterial virulence (Groisman et al. 2013). O’Connor et
al. (2009) first reported the function of Mg2+ in the regulation of bacterial thermotolerance.
Increased expression of Mg2+ transport proteins was found to enhance survival of Salmonella
enterica upon heat treatment. However, the role of Mg2+ in the regulation of thermotolerance
in other organisms still remains unclear.

Therefore, we preliminarily investigated in this work the influence of Mg2+ on

thermotolerance of three well-documented probiotic lactobacilli strains, Lactobacillus
rhamnosus GG, Lactobacillus casei Zhang and Lactobacillus plantarum P-8.
 Chapter 3: What underlies the protection 81

3.2 Materials and Methods

3.2.1 Strains, culture conditions and bacteria-milk matrix preparation

P. freudenreichii CIRM-BIA 129 was cultivated in YEL broth or sweet whey media with
different TS described in Item 2.2.1 and 2.2.2. Lactobacillus strains were routinely activated
and cultivated statically at 37°C for 24 h in MRS broth prior to further experiment. Among
these strains, Lactobacillus rhamnosus GG (LGG) was obtained from a commercial product
(Culturelle, CVS Pharmacy). Lactobacillus casei Zhang (LCZ) and Lactobacillus plantarum
P-8 (LP) were provided by the Key Laboratory of Dairy Biotechnology and Engineering,
Ministry of Education, China.

The effect of milk aggregates on protective capacity of milk matrix was studied with LGG
strain. The Ca2+-aggregated milk was prepared as described previously (Huang et al., 2014).
Briefly, UHT skim milk (Devondale, Australia) was added by 10 mmol l-1 CaCl2 (Sinopharm,
Shanghai, China) and then heated at 90°C for 30 min to cause the protein aggregation. Apart
from the untreated UHT skim milk, the 10 mmol l-1 CaCl2-added milk (without heating) was
also used in this work as a control to exclude the effects of both CaCl2 addition and higher dry
matter (Ca-added milk, in brief). All of the milk matrices were fully vortexed and then placed
at 25°C for 30 min by a thermal mixer (Eppendorf, Germany) before being mixed with LGG
cells. LGG cells were harvested in the 2 mL Eppendorf tube from the 24 h culture by
centrifuging at 8000×g for 15 min (Heal Force, Hong Kong, China). The cell pellets were
washed by 0.5% (w/v) peptone water in the same centrifuging conditions and then suspended
in a 1:1 volume of the above milk matrix. All of the bacteria-suspended matrices were fully
vortexed and placed at 4°C before single droplet drying.

3.2.2 Stress challenges

(Multistress challenges to P. freudenreichii)

Acid, bile salts, and heat challenges were applied to stationary-phase cultures of P.
freudenreichii according to standardized procedures at pH 2 and 37°C for 60 min in YEL broth
adjusted to pH 2.0 using HCl (Jan et al., 2001, 2000) or in the presence of 1 g L-1 of bile salts
(an equimolar mixture of cholate and deoxycholate; Sigma Chemical, St. Louis, MO, USA) in
YEL at 37°C for 60 min (Leverrier et al., 2003) or at 60°C for 10 min (Leverrier et al., 2004)
prior to CFU counting on YEL agar to determine survival percentages. As a control, cultures
 Chapter 3: What underlies the protection 82

were left untreated for the same time in order to determine the population corresponding to
100% survival. The viable and cultivable population of bacteria was determined by CFU
counting on plates for maximal recovery of both treated and untreated cultures. Percent survival
was then determined by comparing stressed and unstressed cultures at the end of the same time
period.

(Heat challenge to Lactobacillus)

The effect of Mg2+ on thermotolerance of lactobacillus was investigated with LCZ, LGG
and LP strains. Preparation of bacterial suspension and heat challenge experiments were
performed as described previously (Huang and Chen, 2013). Lactose solution (10% w/v) was
used as the medium due to its excellent drying behavior and potential protection on bacteria
during spray drying. Briefly, 500 µL of bacterial culture after 24 h incubation was washed with
peptone water and then re-suspended in 500 µL 10% w/v lactose solution with 0, 5, 10, 20, 50,
100 and 500 mmol l-1 of MgCl2, respectively. Besides, the same concentrations of NaCl and
ZnCl2 were also used as a control to compare with the effects of Mg2+. The suspensions were
moderately shaken at 25oC for 30 minutes prior to a heat treatment at 70oC for 2 mins. The
viable and cultivable population of Lactobacillus was determined by CFU counting, on plates
of MRS medium, solidified by 10 g l-1 agar, for both heated and unheated cultures. The survival
was thus expressed as log N/N0, where N is the population following the heat treatment and N0
the population before the heat challenge.

(Recovery of Lactobacillus from heat injury)

Similarly as in the heat challenge experiment, 1 mL of bacterial suspension was prepared
by re-suspending bacterial pellets in 1 mL 10% w/v lactose solution. Heat treatment was
performed at 75oC for 1 min. After the heat treatment, the heated suspension was inoculated
(1% inoculum) into the MRS broth supplemented with different concentrations of inorganic
salts in the 48-wells plate. The bacterial regrowth was monitored by measurement of the optical
density change (OD600) at 37oC for 30 h using a microplate reader (SpectraMax M5, Molecular
Devices). The modified Gompertz model was used to fit the OD600 curves (Zwietering et al.
1990). The growth lag phase (λ) and asymptotic value (A, indicates the final biomass yield)
were calculated to characterize the regrowth activity of bacteria (an example of fitting is shown
in Figure 23 and Table 10). The R-square values between all experimental curves and fitting
equations were calculated by Origin Pro 8 SR02 (OriginLab, USA). The goodness of fitting of
all curves were found to be higher than 0.995. To compare the effects of salts on the regrowth
 Chapter 3: What underlies the protection 83

activity of bacteria, the difference between heat-induced changes of lag phase and asymptotic
value were compared by λ–λ0 and A-A0. λ0 and A0 refer to the lag phase and asymptotic value
of bacteria regrowth in MRS without inorganic salt supplementation after heat treatment, while
λ and A refer to that of in MRS with inorganic salt supplementation after heat treatment,
respectively.

Figure 23. An example to describe the growth curve of LGG at 37°C fitted with the Gompertz model.
The solid curve is experimental curve, while the dotted one represents the fitted curve.

Table 10. The values of λ and A from the curve fitted by Gompertz model in Figure 23
L. rhamnosus GG Equation Y=a×exp(-exp(-k×(x-x×c)))
a 0.67503
Fitted value x×c 7.48637
k 0.24523
=(k×xvc-1)/k =3.963933
Formula
A=a A=0.64727

(Drying challenge of L. rhamnosus GG via single-droplet-drying technique)

Single droplet drying was carried out with the SDPA MARK II equipment (Dong-Concept
New Material Technology Co. Ltd.). The working principles of this equipment have been
described in detail elsewhere (Fu et al., 2011; Lin and Chen, 2002). In brief, the system consists
of an air supply unit used to provide laminar air flow with controllable temperature, humidity,
and velocity and a droplet-drying unit in which an isolated droplet can be suspended at the tip
of a specially made glass filament for drying in the conditioned air stream. A total of three
 Chapter 3: What underlies the protection 84

droplet-suspension modules were equipped in the droplet-drying unit for measuring the
changes of droplet temperature, mass, and diameter, respectively. A high-resolution video
camera was coupled with the droplet-drying unit to monitor and record the change of droplet
morphology during drying. Droplet moisture content on a dry basis (Xd, kg kg−1) was calculated
with the following equation:

Xd = (Md − Ms)/Ms (Eq 12)

where Md is the mass of droplet, and Ms is the total solids mass of the matrix (mg).

The single droplet was generated from the above bacteria suspended milk matrixes with a
gas chromatograph microsyringe (SGE Analytical Science Pty Ltd.). The control consisted of
the drying of sterilized Milli-Q water droplets. For each drying run, the syringe was rinsed by
Milli-Q water, 75% (v/v) ethanol, and sterilized Milli-Q water subsequently. The remaining
LGG viability within droplets was measured as described before (Fu et al., 2013a; Zheng et al.,
2015). Briefly, the drying process was stopped at a defined time by blocking the air flow with
a cold, humid barrier. The droplet was then immediately rehydrated in 1 mL of 0.5% peptone
water and subsequently serially diluted for the measurement of the viable bacteria population
by pouring it on an MRS agar plate. The agar plates were incubated at 37°C for 48 h. The initial
bacteria population in three milk matrixes was also measured before drying (∼109 CFU mL−1
for all three matrixes). The drying conditions used in this work were set as follows: air
temperature, 65 ± 0.2°C; air velocity, 0.91 ± 0.02 m s−1; air moisture content, 1 ± 0.01 g kg−1;
and droplet volume, 2 ± 0.05 μL.

3.2.3 Physico- physicochemical properties of bacteria-milk matrix and particle

The size distribution of dispersed elements in milk matrixes were measured by laser-light
diffraction (Horiba LA-960). The size distribution of LGG cells suspending in the peptone
water (0.5% w/v) was also measured after washing twice by the peptone water.

The viscosity of milk matrixes was measured using an oscillatory rheometer (Kinexus Pro+,
Malvern, UK). The samples were first equilibrated at 25°C and then quickly pipetted onto the
bottom plate of the rheometer. A 40 mm cone (4°) geometry was used with the sheer rate at 50
s−1, and the operation temperature was controlled at 25 ± 0.5°C.

The content of milk whey proteins in the serum phase of different milk matrixes was
compared by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Milk
 Chapter 3: What underlies the protection 85

matrixes were first centrifuged at ×8000g for 15 min. The supernatant was collected and then
mixed with the LDS sample buffer (NuPGE, ThemoFisher Scientific) at the volume ratio of
1:1 prior to heat denaturation at 75°C for 10 min. One-dimensional polyacrylamide gel
electrophoresis was conducted on the XCell SureLock Mini-Cell system (ThermoFisher
Scientific) with 12% Bis−Tris Precast gel (NuPAGE, ThermoFisher Scientific). The gel was
stained by Bio-Safe Coomassie blue reagent (Biorad) prior to photograph with the imaging
system (ChemiDoc XRS+, Biorad).

To characterize the mobility of water in bacteria-suspended milk matrixes, the transverse

relaxation time (T2) was measured using a 23 MHz 1H nuclear magnetic resonance (NMR)
spectrometer (Suzhou Niumai Electronic Technology Co., Ltd.) at 32 ± 0.1°C. The equilibrated
samples were placed in the 15 mm diameter tube. The tube was sealed with parafilm to prevent
the water loss during measurement. The T2 distributions were measured using the Carr−Purcell
−Meiboom−Gill (CPMG) pulse sequence, with the 90−180° pulse spacing varied between 25
and 50 μs. The echo number was set to 4000, the echo time was 0.5 s, the waiting time was 2
s, and 16 scans were acquired for each measurement. The data were collected and analyzed by
the instrument MultiExp Invfit-T Analysis software.

The particles after drying at 65°C for 420 s (by single-droplet drying) were collected and
cut by a surgical knife blade. The particle pieces were placed on the carbon tape and then
sputter-coated by gold − palladium. These samples were observed by scanning electron
microscopy (Hitachi Ltd.).

3.2.4 P. freudenreichii proteomic analysis and indentification

In order to understand the bacterial stress response in highly concentrated sweet whey, a
proteomic analysis was performed with P. freudenreichii. Two-dimensional electrophoresis
was carried out as described previously (Leverrier et al., 2004). Briefly, propionibacteria were
harvested by centrifugation, together with medium-insoluble materials; washed in 0.1 M Tris
buffer, pH 8, containing 2% sodium citrate; centrifuged; washed twice in 0.1 M Tris buffer,
pH 8, containing 0.001 M EDTA and then in phosphate-buffered saline (PBS); pelleted; frozen;
and then resuspended in SDS lysis buffer prior to sonication and cell lysis using a Retsch
MM301 mixer mill. The resulting SDS extracts were recovered by centrifugation (21,000 × g;
20°C; 20 min) and analyzed by one-dimensional SDS-PAGE (27) prior to protein precipitation
using a two-dimensional (2D) cleanup kit (GE Healthcare Bio-Sciences AB, Uppsala,
 Chapter 3: What underlies the protection 86

Sweden). Protein pellets, as well as immobilized pH gradient (IPG) dry strips, were rehydrated
in destreaking solution (GE Healthcare Bio-Sciences AB) supplemented with IPG buffer, pH
4 or 7. For each two-dimensional gel, 300 µg of cellular proteins, whatever the treatment, were
loaded onto the acid end of a linear-gradient IPG dry strip (pH 4 to 7) prior to isoelectric
focusing on a MultiphorII up to 60,000 V·h according to a standardized procedure (Jan et al.,
2001). The second dimension was SDS-PAGE (12.5% polyacrylamide; 16 by 16 by 0.1 cm;
Protean II; Bio-Rad, Hercules, CA). The gels were stained using Bio-Safe Coomassie G-250
stain (Bio-Rad) prior to scanning on an ImageScanner III (GE Healthcare Bio-Sciences AB).
Image analysis, gel matching, and quantification of the protein amounts in individual spots,
calculated as average normalized volumes, were performed using Progenesis SameSpot
software 3.1 (Nonlinear Dynamics, Newcastle upon Tyne, United Kingdom). Molecular
weights and isoelectric points were calculated with the software using molecular markers
(LMW; GE Healthcare Bio-Sciences AB) and assuming a linear 4-to-7 pH gradient, as
indicated by the supplier. Gel subgroups corresponding to growth in 5% SW and in 30% SW
were compared to compute fold and P values of all spots using one-way analysis of variance
(ANOVA) and principal-component analysis (PCA), as described previously (Morris et al.,
2010). The spots of interest, significantly upregulated in 30% SW compared to 5% SW (above
a 1.2-fold change), were selected and spotted for further identification, as previously described
(Gagnaire et al., 2015). The fold changes are indicated.

Gel pieces were washed with acetonitrile and ammonium bicarbonate solution and then
dried under vacuum in a Speed-Vac concentrator (SVC100H-200; Savant; Thermo Fisher
Scientific, Waltham, MA, USA). In-gel trypsin digestion was performed overnight at 37°C and
stopped with spectrophotometric-grade trifluoroacetic acid (TFA) (Sigma-Aldrich, Saint
Quentin Fallavier, France) as described previously (Gagnaire et al., 2015). The supernatants
containing peptides were then vacuum dried in a Speed-Vac concentrator and stored at -20°C
until mass spectrometry (MS) analysis. Nanoscale liquid chromatography coupled online with
tandem mass spectrometry (NanoLC–MS-MS) analysis was performed with an on-line liquid
chromatography (MS-MS) setup using a Dionex U3000-RSLC NanoLC system fitted to a
QStar XL (MDS SCIEX, Ontario, Canada) equipped with a nano-electrospray ion (ESI) source
(Proxeon Biosystems A/S, Odense, Denmark). Samples were first concentrated on a PepMap
100 reverse-phase column (C18; 5-µm particle size; 300-µm inside diameter [i.d.] by 5-mm
length) (Dionex, Amsterdam, The Netherlands). Peptides were separated on a reverse-phase
PepMap 100 column (C18; 3-µm particle size; 75-µm i.d. by 150-mm length; Dionex) at 35°C,
 Chapter 3: What underlies the protection 87

using solvent A (2% [vol/vol] acetonitrile, 0.08% [vol/vol] formic acid, and 0.01% [vol/vol]
TFA in deionized water) and solvent B (95% [vol/vol] acetonitrile, 0.08% [vol/vol] formic
acid, and 0.01% [vol/vol] TFA in deionized water). A linear gradient from 10 to 40% solvent
B in 17 min was applied for elution at a flow rate of 0.3 µl min-1. The eluted peptides were
directly electrosprayed into the mass spectrometer operated in positive mode. A full continuous
MS scan was carried, out followed by three data-dependent MS-MS scans. Spectra were
collected in the selected mass range of 300 to 2,000 m/z for MS and 60 to 2,000 m/z for MS-
MS spectra. Charged ions (2+ to 4+) were considered for the MS-MS analysis when the ion
intensity was above 10 cps. The mass spectrometer was operated in data-dependent mode,
automatically switching between MS and MS-MS acquisition using Analyst QS 1.1 software.
The instrument was calibrated by multipoint calibration using fragment ions that resulted from
the collision-induced decomposition of a peptide from β-casein (β-CN) (193 to 209). To
identify peptides, all data (MS and MS-MS) were submitted to X! Tandem using the X!
Tandem pipeline developed by Plateforme d’Analyze Protéomique de Paris Sud-Ouest
(PAPPSO), INRA, Jouy-en-Josas, France ([Link] The search was performed
against a database composed of (i) a homemade database containing all the predicted proteins
of P. freudenreichii strain CIRM-BIA 129 used in this study and (ii) a portion of the UniProtKB
database corresponding to P. freudenreichii. Database search parameters were specified as
follows: trypsin cleavage was used, and the peptide mass tolerance was set to 0.2 Da for both
MS and MS-MS. Oxidation of methionine and deamidation of asparagine and glutamine
residues were included as variable modifications. Protein identifications were automatically
validated when they showed at least two unique peptides with an E value below 0.05.

3.2.5 P. freudenreichii cellular compatible solutes

Polyphosphate granule staining was done as recently described for lactobacilli (Alcántara
et al., 2014) for the detection of polyphosphate granules using the procedure of Neisser.
Bacterial cultures were smeared on microscope slides and heat fixed. The slides were covered
for 1 min with a freshly prepared mixture of 1 volume of Neisser’s methylene blue solution
(Fluka Analytical, France) and 2 volumes of 0.33% crystal violet solution in 10% ethanol.
Excess dye was absorbed using blotting paper, and the slides were covered for 1 min with a
0.3% solution of chrysoidin G solution (Sigma-Aldrich, Saint Quentin Fallavier, France) prior
to extensive rinsing with water. The slides were observed on an Olympus BX51 microscope at
a ×1,000 magnification. In addition, polyphosphate granules were visualized by DAPI (4’,6’-
 Chapter 3: What underlies the protection 88

diamidino-2-phenylindole) staining as previously described (Günther et al., 2009) and later

modified (Mukherjee et al., 2015) for cyanobacteria. Briefly, bacteria were washed in
McIlvaine’s buffer (Mukherjee et al., 2015), fixed in 4% formaldehyde, permeabilized in 0.3%
Triton X-100, and stained in 20 µg mL-1 DAPI in the same buffer. The stained cultures were
observed on an Olympus BX51 epifluorescence microscope equipped with a U-MWU2
fluorescence filter cube (excitation filter, 330 to 385 nm; emission filter, 480 to 800 nm) and
an Olympus plan 100×/1.25 oil objective. For maximal resolution, bacteria were further imaged
using a confocal Leica SP8 and a 63×/1.4 oil HC PL APO CS objective. Images were acquired
using LAS-AF (Leica, Wetzlar, Germany) software. DAPI was excited with a 405-nm laser.
Photomultiplier detector parameters were set to collect fluorescence between 412 and 450 nm
and between 553 and 598 nm. Polyphosphate granules were further visualized by transmission
electron microscopy as described previously for lactobacilli (Alcántara et al., 2014). Briefly,
propionibacteria, grown in different media until stationary phase, were washed in PBS and in
sodium cacodylate buffer (0.1 M, pH 7.3) prior to fixation with 2.5% glutaraldehyde for 3 h.
The bacterial pellet was embedded in agarose prior to being cut into 1-mm pieces and fixed
with 1% sodium tetroxide for 1 h. The agar pieces were rinsed with cacodylate buffer and
dehydrated in ethanol prior to inclusion in Epon-Araldite-DMP30 resin (polymerized at 60°C
for 48 h). Thin sections (90 nm) were cut (Leica ultra microtome Ultracut E), collected on
copper grids, and then stained with uranyl acetate. Samples were observed with a Jeol 1400
electron microscope (Jeol Co. Ltd., Tokyo, Japan), and images were digitally captured with a
Gatan Orius camera (Digital Micrograph Software). Cellular glycan detection was done by
using the periodic acid-Schiff (PAS) staining method, as previously described (Moats, 1959).
The fixed bacterial smears on slides were stained with 1%periodic acid solution for 5 min and
then transferred into absolute ethanol. The slides were rinsed with water, air dried, and
subsequently stained with Schiff’s reagent (Sigma-Aldrich, France), consisting of 1%
pararosaniline HCl and 4% sodium metabisulfite in 0.25 mol L-1 hydrochloric acid. The slides
were washed with flowing water for 5 min and then air dried before observation. For scanning
electron microscopy (SEM) examination of the powder grains, powders were attached to SEM
stubs using a two-sided carbon adhesive tape and then sputter coated with gold-palladium.
Samples were analyzed at 5 kV at room temperature with a scanning electron microscope (JSM
7100F; Jeol Co. Ltd., Tokyo, Japan).

Trehalose and glycogen were assayed as previously described for Saccharomyces

cerevisiae (Parrou and François, 1997) and adapted for propionibacteria (Dalmasso et al.,
 Chapter 3: What underlies the protection 89

2012). Briefly, propionibacteria were washed twice in PBS, resuspended in acetate buffer (40
mmol l-1, pH 5.2), heat inactivated for 5 min at 95°C, and disrupted using zirconium beads in
a Retsch MM301 mixer mill prior to centrifugation of cellular debris. The resulting extract was
divided into two parts, further digested using amyloglucosidase (from Aspergillus niger [no.
10115; Sigma-Aldrich]) or trehalase (from porcine kidney [no. T8778; Sigma-Aldrich]) for the
hydrolysis of glycogen and trehalose at 57°C and 37°C, respectively. Samples were quickly
frozen, and the resulting glucose was quantified using a glucose hexokinase assay kit (Sigma).
The results were expressed as the concentration of generated glucose per unit weight of cellular
wet matter.

3.2.6 Statistical Analysis

All experiments were independently repeated three times or performed with samples in
triplicate (biological replicates). The results are presented as means ± standard deviations.
Significant differences (P < 0.05) between the mean values were determined by Tukey’s test.
Statistical analysis was performed using R 3.2.1 with the Rcmdr package (R Development Core
Team). The features of bacterial and particle morphology, particle size distribution and SDS-
PAGE were repeatedly observed in triplicate samples.
 Chapter 3: What underlies the protection 90

3.3 Results and Discussion

3.3.1 Osmotic adaptation induces multistress tolerance of P. freudenreichii

[Link] Multitolerance response is triggered during growth

As shown in Chapter 2 (Figure 14B in Item 2.3.1), the use of highly concentrated sweet whey
sustained P. freudenreichii growth. The final populations of propionibacteria were around 109
CFU mL-1, whatever the growth medium, as determined by CFU counting. Hyperconcentrated
sweet whey, 20 and 30% TS, allowed growth yields above that of the normal YEL reference
medium, with a larger final population of P. freudenreichii, in accordance with the larger
amount of substrates present in the medium. In contrast, growth was inhibited at 40% dry
matter, suggesting inhibition at a concentration that was too elevated. As a matter of fact,
osmolality increased with the concentration (Figure 13 in Item 2.2.2) to levels far above that of
YEL medium; as hyperosmotic stress is a major stress for bacterial elongation and growth
(Empadinhas and da Costa, 2008). P. freudenreichii was thus able to adapt to osmolality up to
about 1,500 but not above 2,500 mosmol kg-1.

We then looked at the effect of such stress, undergone during growth, on tolerance for other
stresses, imposed after growth. Growth in 20 and 30% TS sweet whey led to significant
increases of stress tolerance in P. freudenreichii (Figure 24). The most striking example was
bile salts challenge survival, with 60.9% survival in 30% TS sweet whey compared to 2.6% in
5% TS sweet whey. This result strongly suggests that the growth conditions imposed sublethal
doses of stress responsible for stress adaptation; hyperosmotic stress being the main stimulus
in hyperconcentrated sweet whey and known to trigger bile salts tolerance in other bacteria
(Flahaut et al., 1996; Zhang et al., 2011).
 Chapter 3: What underlies the protection 91

A
40% b
b
30%

% survival
20%

10% a
a
a a
0%
5 10 20 30 40 YEL
Sweet whey concentration (w/w %)

B 70% c
60% b
50%
% Survival

40%

30%

20% a
a a
10% a

0%
5 10 20 30 40 YEL
Sweet whey concentration (w/w %)

C 70% c

60% c

50%
% Survival

40%
a
a
30%
ab
20%

10% b

0%
5 10 20 30 40 YEL
Sweet whey concentration (w/w %)

Figure 24. P. freudenreichii were cultivated for 72 h in the indicated growth media until stationary
phase, then subjected to (A) heat (60°C for 10 min), (B) acid (pH 2.0 for 1 h), or (C) bile salts (1 g l-1
for 1 h) challenge. Viable propionibacteria were enumerated by plate counting in treated and control
cultures, and then expressed as percent survival. Different letters above the columns indicate significant
differences (P < 0.05).
 Chapter 3: What underlies the protection 92

[Link] The multitolerance response involves key stress proteins

To investigate the molecular mechanisms responsible for such adaptation, we performed a

proteomic analysis in 5%versus 30% sweet whey culture media. As shown in Figure 25, at least
41 protein spots were enhanced in the pH range from 4 to 7 in 30% sweet whey. They were
also enhanced in this hypertonic medium compared to the reference rich and isotonic YEL
medium (data not shown).

A B

Figure 25. Hyperconcentrated sweet whey culture triggers overexpression of stress proteins.
Propionibacteria were cultivated for 72 h in isotonic 5% (A) or hyperconcentrated 30% (B) sweet whey.
The proteins accumulated in the cellular compartment were then analyzed by 2D electrophoresis (pH 4
to 7 isoelectric focusing and then SDS-12.5% PAGE) prior to differential analysis using the SameSpot
software. The proteins induced in 30% sweet whey are indicated by arrows. Their identities, revealed
by trypsinolysis and NanoLC–MS-MS, are shown in Table 11.

The induced protein spots were then picked and subjected to in-gel trypsinolysis as
previously described (Leverrier et al., 2004), and the resulting peptides were subjected to
NanoLC–MS-MS as described in Materials and Methods. For each protein spot, the major
identified protein is presented in Table 11.
 Chapter 3: What underlies the protection 93

Table 11. P. freudenreichii CIRM BIA 129 proteins overexpressed in hyperconcentrated 30% sweet
whey culture, compared to isotonic 5% sweet whey culture, as identified by NanoLC–MS-MS
Log E No. of
No.a Mexpb PIexpc Locus tag Description Cover %e Mtheorf PItheorg Foldi
valued Spectrah
PFCIRM129
1 93.2 5.71 _06355 Chaperone clpB 2 -524.4 75 94.2 5.33 220 1.5

Methylmalonyl-CoA
PFCIRM129
2 82.5 5.55 _07240
mutase large subunit -426.4 75 80.1 5.21 160 1.5
mutB
Myo-inositol
PFCIRM129
3 72.1 5.30 _10415 catabolism iolD -472.7 78 69.7 5.04 160 1.7
protein
PFCIRM129 FeS assembly protein
4 56.8 5.28 _11225
-471.8 86 53.8 5.02 214 1.5
SufB
Myo-inositol
PFCIRM129
4 56.8 5.28 _10420
catabolism IolA -369.2 77 52.7 5.00 132 1.4
protein
PFCIRM129 Aspartate ammonia-
5 55.5 5.38 _00340
-215.0 65 53.0 5.03 93 1.3
lyase
PFCIRM129
6 55.2 5.45 _11435 Pyruvate kinase 1 -432.7 83 53.9 5.17 202 1.4

PFCIRM129 Coenzyme A
7 54.6 6.07 _01440
-695.6 87 55.6 5.53 261 1.6
transferase
PFCIRM129 Coenzyme A
8 54.2 6.18 _01440
-526.4 81 55.6 5.53 177 1.4
transferase
Glycine
PFCIRM129
9 53.9 5.68 _10235 hydroxymethyltransfe -432.3 80 51.8 5.28 187 1.6
rase precursor
PFCIRM129 Xaa-Pro
10 53.5 5.56 _11790
-302.5 74 54.4 5.32 127 1.7
aminopeptidase I
ATP-binding protein
PFCIRM129 opuCA of
11 51.9 5.90 _07855
-352.3 70 47.9 5.51 134 1.3
osmoprotectant ABC
transporter
PFCIRM129 Dihydrolipoyl
12 49.7 5.33 _09770
-395.4 75 49.5 5.08 179 1.4
dehydrogenase
PFCIRM129 Fumarate hydratase.
13 49.1 5.63 _11895
-611.8 78 51.1 5.34 261 1.6
class-II
PFCIRM129 Fumarate hydratase.
14 48.6 5.74 _11895
-636.2 90 51.0 5.34 311 1.4
class-II
6-phosphogluconate
PFCIRM129
15 48.6 5.54 _07025 dehydrogenase. -398.4 84 52.5 5.25 191 1.5
decarboxylating
PFCIRM129
16 47.2 6.15 _06185 Galactokinase -328.8 63 41.8 5.57 99 1.4

Anaerobic glycerol-3-
PFCIRM129 phosphate
17 44.5 5.50 _04835
-321.2 89 45.8 5.27 118 1.9
dehydrogenase
subunit B
PFCIRM129
18 44.5 5.34 _08275 Elongation factor Tu -344.6 68 43.6 5.05 138 1.4

PFCIRM129 Tryptophan synthase

19 44.4 6.05 _04290
-250.1 72 44.5 5.47 72 1.4
beta chain (TrpB)
Glutamate-5-
PFCIRM129
20 43.6 5.63 _06815 semialdehyde -243.3 75 43.5 5.20 109 1.5
dehydrogenase
Putative
PFCIRM129
21 42.9 5.50 _07640 isocitrate/isopropylma -330.8 78 44.7 5.20 178 1.4
late dehydrogenase
PFCIRM129 Tryptophanyl-tRNA
22 42.5 5.70 _02990
-227.9 77 40.7 5.45 113 1.2
synthetase
 Chapter 3: What underlies the protection 94

Table 11 continued
PFCIRM129
23 42.2 6.06 _03745 Acetate kinase -600.6 89 42.4 5.50 303 1.5
PFCIRM129
24 40.6 5.23 _10390 Transaldolase 2 -300.4 78 39.3 5.10 150 1.7
PFCIRM129
25 39.9 5.32 _10390
Transaldolase 2 -295.4 79 39.3 5.10 173 2.8

PFCIRM129 zinc-binding
26 39.5 6.05 _07545
-522.9 83 37.1 5.52 201 1.2
dehydrogenase
PFCIRM129
27 39.1 5.78 _03400 Aldo/keto reductase -301.6 80 36.9 5.36 140 1.7
PFCIRM129
28 38.6 6.11 _05680 Oxidoreductase -274.2 89 38.7 5.57 142 1.6

Glyceraldehyde-3-
PFCIRM129
29 37.9 5.52 _11300 phosphate -424.3 85 36.0 5.46 306 1.2
dehydrogenase
Glyceraldehyde-3-
PFCIRM129
30 37.6 5.66 _11300
phosphate -412.2 78 37.7 5.46 313 1.4
dehydrogenase
PFCIRM129 NADPH:quinone
31 35.3 6.07 _01130
-171.3 74 33.6 5.59 64 1.9
reductase
Putative aldo/keto
PFCIRM129
32 32.6 5.98 _01465 reductase -233.0 81 31.1 5.48 135 1.4
(oxidoreductase)
Putative aldo/keto
PFCIRM129
33 32.3 5.78 _01465 reductase -182.0 79 30.4 5.48 122 1.4
(oxidoreductase)
PFCIRM129
34 31.2 5.45 _07790 Cysteine synthase 1 -175.6 87 31.6 5.11 62 1.8

NADH-quinone
PFCIRM129
35 30.1 5.61 _08515 oxidoreductase chain -227.1 80 28.6 5.36 138 1.8
C
PFCIRM129
36 28.3 5.91 _10185 Oxidoreductase -104.0 54 28.8 5.56 52 1.3

PFCIRM129 Superoxide dismutase

37 24.5 5.41 _08020
-452.1 89 22.8 5.28 209 1.4
[Mn/Fe] sodM
PFCIRM129 Superoxide dismutase
38 24.0 5.57 _08020
-533.7 93 22.8 5.28 319 1.6
[Mn/Fe] sodM
PFCIRM129 Methylmalonyl-CoA
39 19.1 5.79 _05095
-214.1 90 16.6 5.40 203 1.1
epimerase
Heat shock protein 20
PFCIRM129
40 18.0 4.71 _02555 1 (20 kDa chaperone -139.6 77 17.3 4.73 119 1.4
1)
PFCIRM129 Phosphocarrier. HPr
41 16.2 4.28 _10610
-146.0 95 8.9 4.22 106 1.8
family
a
The numbers correspond to the spots in Figure 25.
b
Mexp, molecular mass (kDa) evaluated from bidimensional electrophoresis migration.
c
pIexp, isoelectric point evaluated from bidimensional electrophoresis migration.
d
The E value is the probability that a given peptide score will be achieved by incorrect matches from a database search. Protein
identifications were automatically validated when they showed at least two unique peptides with E values below 0.05,
corresponding to a log(E value) of less than -1.3.
e
Cover %, the percentage of the protein amino acid sequence covered by tandem mass spectrometry identification of the
peptide.
f
Mtheor, protein molecular mass (kDa) automatically predicted from the corresponding gene on the Agmial annotation platform.
g
pItheor, isoelectric point automatically predicted from the corresponding gene on the Agmial annotation platform
([Link]
h
Number of MS-MS spectra with an individual E value of <0.05 matched to a peptide sequence from the protein.
i
Fold change of protein expression in 30% SW compared to 5% SW as determined using Progenesis SameSpot software.
j
CoA, coenzyme A.
 Chapter 3: What underlies the protection 95

Several of the identified proteins are known to participate in stress adaptation of P.

freudenreichii and/or in other microorganisms. As an example, the ClpB 2 chaperone (spot no.
1 in Table 11), involved in acid stress adaptation in P. freudenreichii (Leverrier et al., 2004), is
heat and salt inducible in Bifidobacterium breve (Ventura et al., 2005). The protein synthesis
elongation factor EF-Tu (spot no. 18) was also induced here. EF-Tu is known to alleviate the
negative effect of heat stress by acting as a molecular chaperone (Caldas et al., 1998). It is
induced by acid stress in P. freudenreichii (Leverrier et al., 2004) and in Lactobacillus
delbrueckii (Zhai et al., 2014) and plays a role in osmotic adaptation in Escherichia coli
(Berrier et al., 2000). Citrate synthase (spot no. 16) is involved in stress tolerance, and its
disruption leads to multistress susceptibility in Pseudomonas syringae (Sengupta et al., 2015),
while its overexpression leads to aluminum tolerance in alfalfa (Barone et al., 2008),
confirming a role in multistress tolerance response. We also reported the implication of
cysteine synthase (no. 34), superoxide dismutase (no. 37), methylmalonyl-coenzyme A (CoA)
mutase (no. 2), and aspartate ammonia-lyase (no. 5) in P. freudenreichii stress adaptation
(Leverrier et al., 2004).

[Link] The multitolerance response is linked with osmoadaptation

Our observations suggest that P. freudenreichii, in the 30% sweet whey culture medium,
adapted via osmoadaptation mechanisms leading to intracellular accumulation of compatible
solutes. Likewise, previous studies have evidenced that some strains of P. freudenreichii
possess efficient osmoadaptation mechanisms using glycine betaine,
dimethylsulfoniopropionate, and dimethylsulfonioacetate (Boyaval et al., 1999) and involving
the synthesis of trehalose via the OtsA-OtsB pathway, induced by osmotic, acid, and oxidative
stresses (Cardoso et al., 2007). Here, we observed the induction of the OpuCA compatible
solute transport system (no. 11 in Table 11) involved in glycine betaine uptake and
osmoadaptation (Du et al., 2011), which suggests adaptation of P. freudenreichii to
hyperconcentrated sweet whey by increased transport of such compatible solute molecules.
Compatible solutes, in addition to those transported by the OpuCA system, may also be
synthesized de novo, as suggested here by the induction of serine hydroxymethyltransferase,
also called glycine hydroxymethyltransferase (no. 9 in Table 11; EC [Link]), which is involved
in accumulation of choline precursors and thus in osmoadaptation in the halotolerant
cyanobacterium Aphanothece halophytica (Waditee-Sirisattha et al., 2012). Furthermore,
induced proteins also comprised Xaa-Pro aminopeptidase (no. 10), which is involved in
 Chapter 3: What underlies the protection 96

osmoprotection by proline-containing peptides in Bacillus subtilis (Zaprasis et al., 2013), and

gamma-glutamyl phosphate reductase (no. 20, encoded by proA), involved in osmoadaptive
proline production in B. subtilis (Brill et al., 2011). They also comprised glycerol 3-phosphate
dehydrogenase (no. 17), which is involved in osmoadaptation in yeasts, where glycerol plays
a key role, while its mutational inactivation results in reduced salt tolerance (Yan et al., 2008).
Altogether, these data strongly suggest that P. freudenreichii triggers an osmoadaptive
response during growth in hyperconcentrated sweet whey.

We also investigated on the accumulation of trehalose and glycogen, two carbohydrates

known to participate in stress tolerance and in long-term survival in P. freudenreichii (Cardoso
et al., 2007; Dalmasso et al., 2012). As shown in Figure 26, both compounds were below the
limit of detection when P. freudenreichii was grown in YEL medium. In contrast, both were
accumulated in 30%, and to a lesser extent in 5%, sweet whey cultures.

A B C

D E
60 c
Trehalose (glucose eq.

250
Glycogen (glucose eq.

c
50 200
b
[µmol/g])

40
[µmol/g])

150
30
20 100
b 50
10 a a
0 0
YEL 5%SW 30%SW YEL 5%SW 30%SW
Culture medium Culture medium

Figure 26. Hyperconcentrated sweet whey culture triggers intracellular accumulation of carbohydrates.
(A to C) Propionibacteria were cultivated for 72 h in YEL medium (A) or in isotonic 5% (B) or
hyperconcentrated 30% (C) sweet whey. Periodic acid-Schiff staining was applied to smears of the
cultures. (D and E) Glycogen (D) and trehalose (E) were then enzymatically quantified in whole-cell
extracts. The results are expressed as concentrations of generated glucose per unit weight of cellular
wet matter. Different letters above the bars indicate significant differences (P_0.05).
 Chapter 3: What underlies the protection 97

Such accumulation is known to confer enhanced osmotolerance and long-term survival

under cheese-ripening conditions (Dalmasso et al., 2012). Accordingly, periodic acid-Schiff
staining of carbohydrates in smeared cultures of P. freudenreichii evidenced the presence of
deeply stained round structures at both ends of the bacteria in 30% sweet whey that were not
present in YEL cultures.

Inorganic polyphosphate was also reported to play a key role in stress tolerance by acting
as a protein-protecting chaperone and by regulating sigma factors involved in stress response
in bacteria (Gray and Jakob, 2015). Polyphosphate accumulation in the form of volutin granules
has been evidenced in various Lactobacillus species (Alcántara et al., 2014), in
Corynebacterium glutamicum (Klauth et al., 2006), and in P. freudenreichii (Clark et al., 1986).
In our study, we revealed the characteristic intracellular dark-stained granules in P.
freudenreichii by the Neisser staining used by the above-mentioned authors (Figure 27A to C).
Granules were present in sweet whey cultures—more abundant in 30% than in 5% sweet
whey—while they were not detected in YEL-cultured propionibacteria. In addition to Neisser
staining, the fluorescent DAPI probe was also used to visualize polyphosphate (green) and
DNA (blue). Green-light-emitting bacteria were evidenced in sweet whey cultures using
epifluorescence microscopy, where emission wavelengths of 420 to 800 nm were seen together
(Figure 27D to F). This was further confirmed by confocal microscopy (Figure 27G to I), where
blue (412 to 450 nm) and green (553 to 598 nm) images were acquired and then merged.
Furthermore, transmission electron microscopy (Figure 27J to L) showed, in sweet whey
cultures, circular and oval holes that had already been described when these granules were
chipped or torn out during cutting of thin sections (Alcántara et al., 2014; Auling et al., 1991;
Bode et al., 1993).
 Chapter 3: What underlies the protection 98

Figure 27. Propionibacteria were cultivated in YEL (A, D, G, and J) or in 5% (B, E, H, and K) or 30%
(C, F, I, and L) sweet whey. (A to C) Polyphosphate metachromatic granules were evidenced by Neisser
staining and observed at ×1,000 magnification. (D to I) Polyphosphate was visualized by DAPI staining
prior to epifluorescence microscopy (D to F) or confocal microscopy (G to I) observation. Green
fluorescence indicates cytosolic poly(P), and blue indicates DNA. (J to L) Transmission electron
micrographs of propionibacteria. The positions that were occupied by granules appear as holes in the
bacterial cytoplasm (examples are indicated by the arrows).

Altogether, these data indicate that such granules are accumulated in P. freudenreichii
CIRM-BIA 129 under the growth conditions used in the present study. In lactobacilli,
accumulation of intracellular polyphosphate in the form of polyphosphate granules depends on
the species considered, the strain within the species, and the culture conditions (Alcántara et
 Chapter 3: What underlies the protection 99

al., 2014). The presence of inorganic phosphate in the culture medium is a prerequisite for
polyphosphate accumulation. In the sweet whey used here, inorganic phosphate accounts for
1.14% of the dry matter, so its concentration is close to 6 mM and 36 mM for 5% and 30%
sweet whey, respectively. This is close to the 37 mmol l-1 KH2PO4 concentration in malic
enzyme induction (MEI) medium (0.5% yeast extract, 0.5% tryptone, 0.4% KH2PO4, 0.5%
KH2PO4, 0.02% MgSO4·7H2O, 0.005% MnSO4, 1 mL Tween 80 liter-1, 0.05% cysteine, 0.5%
glucose), favoring polyphosphate accumulation in lactobacilli, which coincides with enhanced
stress tolerance, including salt and low pH (Alcántara et al., 2014). Interestingly, abolishment
of polyphosphate accumulation ability by mutational inactivation of the ppk gene leads to
reduced stress tolerance (Alcántara et al., 2014). We propose that osmotically induced
accumulation of polyphosphates, used as both energy storage molecules and compatible solutes
by P. freudenreichii, participates in osmoinduced enhanced multitolerance.

[Link] Summary

In summary, highly concentrated sweet whey, as used as a culture medium in the novel 2-
in-1 process, triggers different stress tolerance pathways, including overexpression of
chaperones and accumulation of key compatible solutes. It leads to live and stable probiotics
with induced multitolerance. This can partially explain the reason why bacteria survived better
during spray drying and among storage.

Considering these adaptation processes, we can even expect that probiotic powders
produced from the 2-in-1 process may display enhanced tolerance towards digestive stresses,
when compared to fresh culture. This would lead to better delivery efficiency of probiotic
health effects. However, this remains an interesting topic for further investigation.
 Chapter 3: What underlies the protection 100

3.3.2 Calcium-protein aggregates protect probiotics against droplet drying

A gentle reminder

In comparison to the natural whey, the concentrated whey induces more

aggregation during autoclave, pasteurization and fermentation. In order to
study the independent effect of the aggregates on probiotic protection
(excluding the influence of high TS content as already proven in the previous
sections), protein aggregation was induced in non-concentrated skim milk
(ca-aggregated milk in brief) by adding very low concentration of calcium
salts (10 mmol l-1). The untreated UHT skim milk (skim milk in brief) and the
10 mmol l-1 CaCl2-added skim milk (without heating, Ca-added skim milk in
brief) were used as controls to exclude the effects of CaCl2 addition and high
TS content.

[Link] Physical and chemical properties of aggregated milk

The size distribution of L. rhamnosus GG (LGG) and the dispersed elements in the three
LGG-suspended milk matrixes is shown in Figure 28. The size of LGG cells ranged from 1 to
10 µm (Figure 28A). This size range appears in all the three LGG-suspended milk matrixes
(Figure 28B). Specifically, the dispersed elements in untreated skim milk and Ca-added milk
showed a very similar size distribution, consisting mostly in elements ranging approximately
from 70 to 500 nm and, conversely, in a smaller proportion of elements with a large diameter
(from 3 to 10 µm). These later might correspond, respectively, to the casein micelles and LGG
cells in the milk (Kim et al., 2009; McGann et al., 1980). Therefore, the addition of 10 m mol-
1
CaCl2 had no influence on the solid size distribution in skim milk. By contrast, most of
dispersed elements in the Ca-aggregated milk had diameters ranging from 5 to 50 µm. In
addition, the Ca-aggregated milk contained a portion of elements with diameter larger than 100
µm. The protein aggregation and increasing of the casein micelle size emerged in the skim milk
with additional Ca2+ upon the heat treatment taking place.

Besides this, the content of proteins in the serum phase of Ca-aggregated milk was largely
decreased (Figure 28C) due to the formation of whey protein-casein micelle aggregates during
 Chapter 3: What underlies the protection 101

heat treatment in the presence of added Ca2+ (Anema and Li, 2003; Dalgleish, 1990;
Gaucheron, 2005).

A
12
LGG Milk serum phase
10
C SM CaM CaAggrM
Percentage (%)

8
6
4
2
0
0,01 0,1 1 10 100 1000
Diameter of solids (µm)

B
15 Skim milk with LGG
Ca-added milk with LGG
Percentage (%)

Ca-aggregated milk with LGG

10

0
0,01 0,1 1 10 100 1000
Diameter of solids in bacterial suspension (μm)

Figure 28. (A) The size distribution of dispersed LGG cells in the 0.5% w/v peptone water. (B) The
size distribution of dispersed elements in three LGG-suspended milk matrixes. (C) The milk serum
protein content of the three bacteria-suspended milk matrixes indicated by SDS-PAGE. SM: UHT skim
milk, CaM: Ca-added skim milk, and CaAggrM: Ca-aggregated skim milk.

The Ca-aggregated milk displayed a considerably higher viscosity when compared to that
of the skim milk and Ca-added milk (Table 12). The higher viscosity of Ca-aggregated milk
was caused by the heat treatment, leading to denaturation of whey proteins and further
aggregation, either in the soluble milk phase or with the casein micelles (Jeurnink and De Kruif,
1993). In addition, the untreated and Ca-added skim milk showed a similar T2 relaxation times
distribution, as shown in Figure 29 and Table 12. However, the Ca-aggregated milk displayed a
 Chapter 3: What underlies the protection 102

left shift of the T2 distribution with a smaller area under the peak, which indicated a decrease
of T2 (Li et al., 2012). It can be concluded that water had a lower mobility in the Ca-aggregated
milk than in untreated and Ca-added skim milk. It could result from the aggregation process,
leading to the entrapping of water into the protein network (Belloque and Ramos, 1999;
Lambelet et al., 1992).

1600
1400
1200
Amplitude

1000
800
600
400
200
0
50 70 90 110 130 150 170 190 210 230 250 270 290
Relaxation time (ms)

skim milk Ca-added milk Ca-aggregated milk

Figure 29. Distribution of transverse relaxation time (T2) of the three bacteria-suspended milk matrixes
measured by a low-field 1H nuclear magnetic resonance (NMR) spectrometer.

Table 12. Viscocity and distribution of T2 relaxation time of the three bacteria-suspended milk matrixes

T2 distribution
Viscocity
Milk matrixes Area proportion
(mPa·s) Start (ms) Peak (ms) End (ms) Area
(%)

Untreated skim 1.4±0.1a 75.6±0.1a 141.0±0.0a 265.6±0.1a 5440±110a 98.3±0.6a

milk

1.3±0.1a 75.6±0.0a 141.0±0.1a 265.6±0.0a 5339±27ab 98.5±0.2a

Ca-added milk

Ca-aggregated 28.9±0.5b 65.8±0.1b 131.2±0.1b 231.0±0.2b 5187±100b 98.2±1.0a

milk
a
Different roman letters (by column) refer to significant difference (p < 0.05).
 Chapter 3: What underlies the protection 103

[Link] Drying kinetics of milk matrixes in single droplet drying

The drying kinetics of three different bacteria-suspended milk matrixes during convective
drying are shown in Figure 30 through the description of temperature and mass changes. The
features of droplet temperature (Td) and mass (Md) were consistent with previous studies
(Sadek et al., 2013; Zheng et al., 2015). For the temperature curve, four stages can be identified
during drying: in the first stage, the Td was rapidly preheated to the wet-bulb range (~ 23°C);
in the second stage, the Td was maintained as nearly constant for tens of seconds; in the third
stage, the Td curve experienced a progressive increase with time and tended toward a final
plateau with a sigmoid shape; in the fourth and last stage, the Td approached the air temperature
at around 65°C. At the same time, the curve of droplet mass Md decreased progressively with
time. Considering the temperature-curve stages previously mentioned, the decrease of Md was
almost linear with time at the first and second stages. The mass loss rate slowed gradually from
the third stage until it approached zero at the fourth stage. The untreated skim milk and Ca-
added milk showed similar Td and Md behaviors during drying. The addition of 10 mmol l-1
CaCl2 without further thermal aggregation had no significant influence on the drying kinetics
of skim milk. However, Ca-aggregated milk displayed distinct drying kinetics, with a longer
second stage and a shorter time to reach the final fourth-stage plateau.

70
2,0
Droplet temperature (ºC)

60
Droplet mass (mg)

50 1,5
40
30 1,0

20
0,5
10
0 0,0
0 100 200 300 400 500
Time (s)

Figure 30. Changes of droplet temperature (solid curves, corresponding to the left ordinate) and mass
(dotted curves, corresponding to the right ordinate) in three bacteria-suspended milk matrixes (UHT
skim milk, Ca-added skim milk, and Ca-aggregated skim milk) and pure water during single-droplet
drying.
 Chapter 3: What underlies the protection 104

The features of Ca-aggregated milk were closer to those of pure water droplets under the
same drying conditions. It is known that the evaporation of pure water can be considered as
diffusion-controlled and has the highest dehydration rate under the same conditions compared
to those of other materials (Sadek et al., 2013). Thus, the Ca-aggregated milk experienced
longer time of diffusion-controlled evaporation in comparison with those of untreated and Ca-
added skim milk. The overall drying efficiency was thus the highest among the three milk
matrixes tested in this work. It can be assumed that the water in the serum phase of the Ca-
aggregated milk was bound to a lesser extent due to its lower protein content (Figure 28C). It
may therefore lead to the more water-like drying behaviour previously observed, with a longer
second stage and faster drying rate during final convective drying.

[Link] Droplet morphology evolution during drying

The evolution of surface morphology of droplets during drying was recorded. The droplet
morphology with the corresponding drying time for each milk matrix is shown in Figure 31.

Figure 31. Droplet morphology and the corresponding drying time of three bacteria-suspended milk
matrixes (UHT skim milk, Ca-added skim milk, and Ca-aggregated skim milk) at four representative
events during single-droplet drying.
 Chapter 3: What underlies the protection 105

Regardless of the milk matrixes, the droplets had a similar shape and size at the beginning
of drying. They shrank homogeneously from the beginning of drying to the end of second stage,
which can be assumed as diffusion-controlled (Arcamone et al., 2007). By the end of second
stage, the spherical shape of droplets were maintained, but the droplet size of Ca-aggregated
milk was smaller than that of untreated and Ca-added milk. When the drying proceeded to the
third stage, the droplet began to distort due to the skin formation on the surface and
progressively lost its spherical shape (Fu et al., 2013b; Pauchard and Allain, 2003). During this
droplet-particle conversion stage, the water diffused through the droplet surface because of the
porous properties of the skin. However, as the sol-gel transition took place over the free surface
of droplet, the droplet skin developed into a thickened shell that impeded the shape of droplet
from further change and progressively hindered the evaporation (Arai and Doi, 2012; Sadek et
al., 2013). In this stage, a vacuole progressively formed inside the particle, as indicated by the
transparent feature of droplets. This could be explained by the pressure change across the
droplet skin, which made it possible for the air to enter the droplet at the same time as water
diffused through the shell (Arai and Doi, 2012; Sadek et al., 2013). This transparent feature
was particularly apparent on the final skim milk and Ca-added milk particles.

Interestingly, the particle formed by the Ca-aggregated milk was hemispherical (lower
height) compared to that of from the skim milk and Ca-added milk during the droplet-particle
conversion. The morphology observed by the SEM displayed an even more remarkable
different structure of the Ca-aggregated milk dried particle (Figure 32). Specifically, this latter
presented a wrinkled surface, and the surface of skim milk and Ca-added milk particles were
relatively smooth (Figure 31). It has been reported in the convective droplet drying that the
particles formed by the whey proteins displayed a relatively smooth surface and spherical
shape, and the casein micelles would form preferentially wrinkled particles (Sadek et al., 2016,
2015, 2014). We suggest that the smooth and spherical particle surface of the untreated skim
milk and Ca-added milk could be attributed to the predominant participation of whey proteins
in the shell formation due to their small size and thereby faster diffusion rate compared to that
of the casein micelles. As discussed above, the whey proteins in Ca-aggregated milk were
associated with the casein micelles to form the large size of aggregates. Hence, the wrinkled
particle surface of the Ca-aggregated milk was due to the main participation of casein micelles
in the shell formation.

Besides, it is also interesting to note that in Chapter 2, we have displayed the surface
morphology of powders dried from 5% and 30% sweet whey (Figure 19 in Item 2.3.3). To some
 Chapter 3: What underlies the protection 106

extent, the features (smooth surface) displayed by the 5% sweet whey powders is similar with
that of the skim milk particle, while the features (wrinkled surface) displayed by the 30% sweet
whey powders is similar with that of the Ca-aggregated milk particle.

Figure 32. Micromorphology of particles dried from (A) UHT skim milk, (B) Ca-added skim milk, and
(C) Ca-aggregated skim milk. Pictures were observed by scanning electron microscopy. For each milk
matrix, four pictures were used to display the particle morphology from different focus sites (from left
to right): surface of intact particle at ×100 magnification, vacuole structure at ×100 magnification, shell
section at ×400 magnification, and shell section at ×10000 magnification.

Last, the inner-morphology of dried particles is also shown in Figure 32. The vacuole
structure was indeed formed inside the particles regardless of the matrix type. However, the
vacuole size of the Ca-aggregated milk particles was smaller than that of untreated and Ca-
added skim milk particles. In a comparison of the particle shell section, the skim milk and Ca-
added milk showed a similar compact and thin shell, and the Ca-aggregated milk displayed a
dense shell with the bubble-like structure. It supports our above hypothesis that the casein
micelles played the main role in the shell formation of the particles dried from Ca-aggregated
 Chapter 3: What underlies the protection 107

milk (Sadek et al., 2015, 2014). In the figures with ×10000 magnification, the network-like
structure in the shell was clearly observed in the particle of Ca-aggregated milk. It was
significantly different from that of matrix-filled structure displayed in the particle shell of
untreated and Ca-added skim milk. This porous properties of shell may also explain the distinct
drying kinetics of the Ca-aggregated milk. As evidence, it should be easier for water to diffuse
across a porous shell, leading thereby to the longer second stage and faster drying rate observed.

[Link] Bacteria viability during single droplet drying

The inactivation kinetics of LGG is shown in Figure 33A. The initial bacterial population
was around 1.0×109 CFU g-1 before drying, whatever the milk matrix. The viability of LGG in
the three milk matrixes maintained constant within the first 200 s of drying, corresponding to
the period from the first stage to the end of the third stage in Figure 30. Viability of bacteria
dropped when the drying proceeded to the fourth stage (~210 s); in skim milk and Ca-added
milk, it decreased dramatically to a level lower than 106 CFU g-1 within 240 s. In comparison,
the Ca-aggregated milk showed a remarkable protection on LGG. Specifically, the LGG
population in Ca-aggregated milk decreased significantly slower than that in the skim milk and
Ca-added milk from 210 s to 420 s. At the end of drying, the remaining bacterial population in
the particle of Ca-aggregated milk was still higher than the level of 107 CFU g-1, which made
the final viability around 2 logs higher than that of in the skim milk or Ca-added milk.

Inactivation of LGG in the droplets was also plotted as a function of droplet temperature
(Figure 33B). It was found that the viable LGG population in the three milk matrixes all
remained at the initial level of 109 CFU g-1 for temperatures lower than 50°C. However, as the
droplet temperature increased continually, the bacterial population in the skim milk and Ca-
added milk droplets decreased by approximately 1 log CFU g-1, from around 53 to 60°C, and
then dropped sharply when the temperature overcame 60°C. In contrast, the Ca-aggregated
milk-matrix better-protected LGG bacteria in the temperature range from 53 to 62°C. For this
matrix, a significant loss of bacterial population was only observed during the fourth stage,
when the droplet temperature approached the air temperature (~63°C). In addition, the bacterial
inactivation as a function of droplet moisture content is shown in Figure 33C. The reduction of
bacterial population cropped up when the droplet moisture content decreased to approximately
0.5 (kg kg-1). Similarly as in Figure 33B, the Ca-aggregated milk exhibited better LGG
 Chapter 3: What underlies the protection 108

protection at the end of drying, i.e., when the droplet moisture content was lower than 0.3 kg
kg-1.

A
Bacterial population (CFU g-1) 1,0E+10

1,0E+09

1,0E+08

1,0E+07

1,0E+06

1,0E+05

1,0E+04
0 50 100 150 200 250 300 350 400 450
Time (s)
Skim milk Ca-added milk Ca-aggregated milk

B 1,0E+10
Bacteria population (CFU g-1)

1,0E+09

1,0E+08

1,0E+07

1,0E+06

1,0E+05
20 30 40 50 60 70
Droplet temperature (ºC)
Skim milk Ca-added milk Ca-aggregated milk

C
5,0E+09
Bacterial population (CFU g-1)

5,0E+08
4,0E+09
Bacterial population

5,0E+07
(CFU g-1)

4,0E+08
5,0E+06

4,0E+07
5,0E+05 0,1 0,2 0,3 0,4 0,5
Droplet moisture content
5,0E+04
0 2 4 6 8 10 12
Droplet moisture content on dry basis (kg kg-1)
Skim milk Ca-added milk Ca-aggregated milk

Figure 33. Inactivation of LGG in three bacteria-suspended milk matrixes (UHT skim milk, Ca-added
skim milk, and Ca-aggregated skim milk) during single-droplet drying. (A) The bacteria population vs
the drying time, (B) the bacteria population vs the droplet temperature, and (C) the bacteria population
vs the droplet moisture content.
 Chapter 3: What underlies the protection 109

The inactivation of LGG during single droplet drying mainly rose when the droplet
moisture content was lower than 0.5 kg kg-1 and the temperature reached above 60ºC. The Ca-
aggregated milk also exerted its protective effects on LGG during this period. It is known that
the inactivation of bacteria during single droplet drying is due to the thermal and dehydration
stress (Schutyser et al., 2012). Previously, the Ca-aggregated milk has been found to protect
LGG under thermal stress due to the better interaction between milk solids and bacterial cells
and a more-stable extracellular environment during heat treatment (Huang et al., 2014). Such
a mechanism could explain the protection of LGG bacteria by Ca-aggregated milk observed
here at the end of thermal convective drying. In our previous work, we reported that the 10
mmol l-1 Ca2+ could improve the heat resistance of lactic acid bacteria (Huang and Chen, 2013).
However, the skim milk with 10 mmol l-1 Ca2+ addition did not increase the survival of LGG
in the present work. It should be noted that the Ca2+ added in milk was subjected to salt
equilibrium between the soluble phase and casein micelles, ruled by HPO4Ca solubility; it was
then partially transferred to the casein micelles calcium phosphate nanoclusters instead of
acting on the bacterial cells (Arifin et al., 2014; Mekmene et al., 2009).

The improved survival of LGG in the Ca-aggregated milk may also be attributed to its
physicochemical properties and distinct drying behavior. The Ca-aggregated milk matrix
before drying presented a higher viscosity compared to the untreated and Ca-added skim milk
(Table 12). It has been assumed that the high viscosity of media could retard molecular mobility
and reaction rate, hence stabilizing the biological system (Buitink et al., 2000). The Ca-
aggregated milk was also relatively inhomogeneous and consisted of a network of large and
insoluble aggregates that incorporated denatured whey proteins and entrapped part of water.
This matrix may provide a local protective environment for the embedded bacterial cells under
the dehydration and thermal stress (Hansen and Riemann, 1963; Potts, 1994). In addition, the
droplet of Ca-aggregated milk had a longer second stage during evaporation (Figure 30); that is
to say, it was maintained at a lower temperature. Conversely, it was faster to reach the final
fourth stage to form the dried particle. It has been reported that the fast drying may induce
instant fixation of the cells and thereby prevent the dehydration inactivation of bacteria
(Perdana et al., 2013). The bacteria cells hence accumulated less dehydration inactivation
within the droplet of Ca-aggregated milk in comparison to that of untreated and Ca-added skim
milk. Besides, it has been observed that the bacteria cells are more heat-resistant in the dry state
than in a hydrated environment (Xing et al., 2014; Zhang et al., 2014). When the temperature
reached the harmful range (mostly the upper limit of the growth temperature, i.e., ~40ºC for
 Chapter 3: What underlies the protection 110

lactobacillis), the Ca-aggregated milk provided a drier extracellular environment for LGG,
likely to explain the better survival of LGG at the high-temperature range. Besides this and
according to (Mezzenga et al., 2005; Sadek et al., 2015), the mechanical properties of the
porous structure in the Ca-aggregated particle (Figure 32) should be soft and ductile, which may
indicate that the bacteria has experienced less mechanical stress during the sol-gel transition
and shrinkage of droplet. It might also explain the reason of the probiotic protection by the
30% sweet whey in the 2-in-1 process due to the similar morphology between powders dried
from 30% sweet whey and the Ca-aggregated milk particle. This remains a hypothesis for
further research.

[Link] Summary

To conclude, the protein aggregation occurring in the Ca-aggregated milk led to lower
protein concentrations in the milk-serum phase and, conversely, protein enrichment of the
casein micelle phase. This microstructure resulted in a more-waterlike drying kinetics, different
feature of droplet-particle conversion (especially regarding the porosity of the shell formed),
and a higher protective capacity on L. rhamnosus GG during single droplet drying.

This work may open new avenues for development of probiotic products with high bacterial
viability and calcium enrichment, as well as a novel strategy for the optimization of drying
efficiency and powder morphology in the dairy industry. The rehydration and digestion
behaviour of powders from the Ca-aggregated milk await further investigation. On the other
hand, it also provides a possible mechanism to explain the protection of the 2-in-1 process on
probiotics by the presence of protein aggregates in highly concentrated sweet whey culture.
 Chapter 3: What underlies the protection 111

3.3.3 Contribution of minerals: a preliminary exploration with magnesium effect

A gentle reminder

In comparison to carbohydrates and proteins, the influence of inorganic

salts on thermotolerance of probiotics has rarely been reported. Mg2+ is
generally present in most of dairy products including the sweet whey
ingredients used in this project. In bacteria, Mg2+ is the second-most abundant
cation. It has been reported to influence thermotolerance of Salmonella
enterica serovar typhimurium. However, the role of Mg2+ in the regulation of
thermotolerance in other organisms still remains unclear. We thus address the
impact of Mg2+ on thermotolerance of Lactobacillus.

[Link] Mg2+ improves bacterial survival after heat challenge

The survival of three probiotic Lactobacillus strains, L. casei Zhang (LCZ), L. rhamnosus
GG (LGG) and L. plantarum P-8 (LP), after heat challenge, is shown in Figure 34. Treatment
of bacteria with low concentration of MgCl2 (5 ~ 100 mmol l-1) in a 10% lactose solution was
found to improve bacterial thermal tolerance. More specifically, 20 mmol l-1 MgCl2 was the
optimal concentration to enhance thermotolerance of LGG and LP (Figure 34A and C), while
10 mmol l-1 for LCZ (Figure 34B). In LGG, 20 mmol l-1 MgCl2 addition led to a 100 fold higher
survival rate upon heat challenge. However, a negative effect on thermotolerance of all three
probiotic strains was observed when the concentration of MgCl2 was higher than 500 mmol l-
1
. By contrast, treatment of bacteria with ZnCl2 and NaCl did not enhance the bacterial
thermotolerance despite the use of high or low concentration. Moreover, the bacterial
thermotolerance dropped sharply as a result of high concentrations of ZnCl2 (> 100 mmol l-1).
 Chapter 3: What underlies the protection 112

A
Concentrations of inorganic salts (mmol l-1)

Survival of LCZ (Log N/N0)

0

after heat treatment

-1 0 5 10 20 50 100 500 1000
**
-2
-3
-4 **
**
-5
***
-6 *** ***
***
-7
L. rhamnosus GG ***
-8

B Concentrations of inorganic salts (mmol l-1)

0
Survival of LGG (Log N/N0)

-1 0 5 10 20 50 100 500 1000

-2 *
after heat treatment

-3
-4
-5
-6 *** ***
-7
-8 ***
-9
L. casei Zhang ***
-10

C
Concentrations of inorganic salts (mmol l-1)
0
-1 0 5 10 20 50 100 500 1000
Survival of LP (Log N/N0)

*** ***
-2 *** ***
after heat treatment

**
-3 ***
*
-4 ** *
***
-5 *** ***
-6
-7 ***
-8
L. plantarum P-8
-9

Figure 34. Survival (log N/N0) of (A) L. rhamnosus GG, (B) L. casei Zhang and (C) L. plantarum P-8
in lactose solutions supplemented with different concentration of MgCl2 (-■-), ZnCl2 (-▲-) and NaCl
(-●-) after heat treatment. Significant difference was compared between the salt-treated samples and the
samples without adding salt, i.e. 0 mmol l-1 added (*: 0.01<P<0.05, **: 0.001<P<0.01, ***: P<0.001).
 Chapter 3: What underlies the protection 113

The results in Figure 34 show the different effects of MgCl2, ZnCl2 and NaCl on the survival
rate of three bacterial strains after heat challenge. The comparison suggests the effects of MgCl2
on bacterial thermotolerance should be attributed to the cation Mg2+, instead of the anion
chloride. The osmolality caused by salt supplementation (mostly by NaCl) may trigger
thermotolerance in bacteria in lactose solution (De Angelis and Gobbetti 2004). However, this
cross protection phenomenon was not induced by high concentration of salts in this work. It
may be explained by the insufficient duration of the 30-minute salt treatment to induce the
cellular stress response (Huang et al. 2016). Moreover, high concentration of salts (500 mmol
l-1) displayed negative effects on the survival of bacteria upon heat treatment. This may be
caused by the extra osmotic stress apart from heat stress. Since there was only lactose in the
suspension, the bacteria were not able to uptake the compatible solutes from the environment
to reach cellular homeostasis.

The enhanced thermotolerance of Lactobacillus strains in this work is in agreement with

the speculation proposed by O’Connor et al. (2009), who found that overproduction of the B.
subtilis MgtE protein lead to enhanced thermotolerance of Salmonella. The genes encoding
similar magnesium-transport proteins can be found in most of lactobacilli species and strains,
mgtA, mgtC and corA in LGG for instance (Morita et al. 2009). It indicates that the Mg2+ effect
on bacterial thermotolerance may be of interest for more other lactobacilli species and strains
used as starter and adjunct cultures for many food fermentations. However, the possible
interactions between the magnesium transport, homeostasis and heat shock response in
Lactobacillus require further transcriptomics and proteomics investigation.

Besides, the effect of Mg2+ on improving bacterial thermotolerance observed here was
similar to the Ca2+ effect reported previously regarding to the close range of effective
concentration (Huang and Chen 2013). Indeed, Mg2+ is similar to Ca2+ regarding their roles in
lactic acid bacteria, especially for their interactions with enzyme activity, cell division, surface
proteins and nucleic acids (Boyaval 1989). This indicates that the distinct physicochemical
properties of Mg2+ and Ca2+ may play significant roles in thermotolerance of Lactobacillus,
particularly their abilities in binding and stabilizing biomolecules such as nucleic acids,
proteins and phospholipid bilayer, and their influence on cell adhesion on environmental
matrices or cell mutual adhesion (Burgain et al. 2014). These physicochemical properties may
allow the effect of Mg2+ to be applicable to other bacterial species.
 Chapter 3: What underlies the protection 114

[Link] Mg2+ improves bacterial recovery of heat-injury

The regrowth activity of bacteria is characterized by lag phase duration (λ), and asymptotic
value (A) from the regrowth curves (Table 13 and Figure 35). The definition of regrowth
corresponds to the ability of Lactobacillus to grow after being exposed to a heat treatment of 1
min at 75oC. The methods of obtaining λ and A explained in the Materials and Methods (Figure
23 and Table 10, 3.2.2). Heat treatment caused extended λ and decreased A in the absence of
salt addition in MRS. Interestingly, as a reference, MRS supplementation with MgCl2 or NaCl
can significantly shorten the heat-induced extension of lag phase duration λ–λ0 (Table 13 and
Figure 35). The optimal concentrations differed depending on the strain. For ZnCl2
supplementation, only a low concentration (5 mmol l-1) showed a positive effect on λ–λ0 of LP
strain, but not for LGG and LCZ strains. Regarding the asymptotic value (A) in the regrowth
curves, the effects of inorganic salts differed largely depending on different strains.
Specifically, the MRS supplementation with 5 to 50 mmol l-1 MgCl2 and 10 to 20 mmol l-1
MgCl2 showed a significantly enhanced A-A0 values for LCZ and for LP, respectively, whereas
no effect of MgCl2 was shown on LGG. By contrast, low concentration of ZnCl2 increased A-
A0 values of LGG and LP, while the NaCl supplementation increased A-A0 values of LGG (5
mmol l-1) and LCZ (5, 10 and 100 mmol l-1).

Recovery of bacteria from sub-lethal injury requires de novo synthesis ribosomes and
membrane (Wu 2008). Mg2+ can influence the nitrogen metabolism and enzyme activity of
bacteria, which may relate to the protein synthesis after thermal injury (Boyaval 1989;
O'Connor et al. 2009). Besides, Mg2+ can stabilize bacterial cell morphology and membrane
integrity, which was found to relate to the heat shock response in L. plantarum WCFS1
(Capozzi et al. 2011). For instance, Rayman and MacLeod (1975), found that Mg2+ strengthen
the bacterial cell wall by influencing the structure of peptidoglycan.

Mg2+ was also reported to promote the division of Gram-positive bacteria at certain
concentrations, which may explain the better regrowth after being treated with Mg2+ (Hayek et
al. 2013). However, the effect of Mg2+ on bacterial regrowth was strain-dependent as shown in
Table 13. Moreover, certain concentrations of Na+ and low concentration of Zn2+ also displayed
a strain-dependent effect on promoting the regrowth of heat-injured bacteria. We speculated
that the reason may be the interactions between Na+ and Zn2+, and bacterial enzyme activity
and membrane integrity, which possibly affect the metabolism and transmembrane transport of
nutrients in bacteria as suggested by Korkeala et al. (1992) and Omburo et al. (1992).
 Chapter 3: What underlies the protection 115

Table 13. The heat-induced changes of lag time duration λ–λ0 and asymptotic value A-A0 between the
regrowth curves of L. rhamnosus GG, L. casei Zhang and L. plantarum P-8 in MRS with and without
inorganic salt supplementation.
Concentrations of mineral ions (mmol l-1)
0 5 10 20 50 100 500

L. rhamnosus GG
λ–λ0 (h)
MgCl2 0.1±0.1c 0.2±0.0cd -0.2±0.1d -0.5±0.1b 0.3±0.1a ND
ZnCl2 0 0.6±0.1a 5.0±0.1b ND ND ND ND
NaCl -0.5±0.1a 0.5±0.0d 0.2±0.1c -0.2±0.1b -0.3±0.1b 1.6±0.1e
A-A0
MgCl2 0.0±0.0b 0.0±0.0b 0.0±0.0b 0.0±0.0b -0.2±0.0a ND
ZnCl2 0 0.1±0.0b 0.0±0.0a ND ND ND ND
NaCl 0.1±0.0c -0.1±0.0a 0.0±0.0b 0.0±0.0b 0.0±0.0b -0.1±0.0a

L. casei Zhang
λ–λ0 (h)
MgCl2 -0.7±0.1b -0.6±0.1b -2.5±0.2a -0.8±0.1b 1.7±0.1c ND
ZnCl2 0 0.9±0.1a 2.5±0.1b ND ND ND ND
NaCl -1.2±0.2c -0.6±0.2d -2.4±0.3a -1.7±0.2b -1.4±0.2bc -0.6±0.2d
A-A0
MgCl2 0.2±0.0c 0.3±0.0d 0.1±0.0b 0.1±0.0b -0.1±0.0a ND
ZnCl2 0 0.0±0.0b -0.1±0.0a ND ND ND ND
NaCl 0.2±0.0e 0.2±0.0e -0.1±0.0a 0.0±0.0c 0.2±0.0d -0.1±0.0b

L. plantarum P-8
λ–λ0 (h)
MgCl2 0.0±0.1d -0.8±0.1a -0.5±0.1b -0.3±0.1c -0.1±0.1d ND
ZnCl2 -0.6±0.1a 5.2±0.1b
0
-
NaCl -0.5±0.1a -0.4±0.1ab -0.2±0.1c -0.5±0.1a 0.5±0.1d
0.3±0.1bc
A-A0
MgCl2 -0.1±0.0a 0.1±0.0c 0.2±0.0d 0.0±0.0b 0.0±0.0b ND
ZnCl2 0 0.1±0.0b 0.0±0.0a ND ND ND ND
NaCl 0.0±0.0b 0.0±0.0b 0.0±0.0b 0.0±0.0b 0.0±0.0b -0.1±0.0a

Different superscript letter represents significant difference from the values in the same lines, i.e. the
samples treated with the same salts (P < 0.05).
‘ND’ represents undeterminable due to non-growth of bacteria after heat treatment
Data in bold means positive effect in regrowth (i.e. less extended lag time and less decreased biomass yield
 Chapter 3: What underlies the protection 116

Figure 35. The regrowth curves of (A) L. rhamnosus GG, (B) L. casei Zhang and (C) L. plantarum P-
8 grown in MRS broth containing different concentrations of MgCl2 after heat challenge. The curve of
bacteria grown in MRS broth before heat challenge (i.e. -■- Unheated) is provided as a reference.
 Chapter 3: What underlies the protection 117

[Link] Summary

To conclude, Mg2+ ranging from 10 to 50 mmol l-1 exerted a general improvement on the
thermotolerance of the three lactobacilli strains. The reason may be the effect of Mg2+ on the
recovery of bacteria from heat injury. These results confirm the important roles of mineral salts
in probiotic physiology. It should be noticed that, apart from carbohydrates and proteins,
certain mineral salts existing in foods could also be used as potential protectants for probiotics
during thermal processing, such as spray drying. Besides, this work indicates that the effect of
Mg2+ may also attribute to the probiotic protection of 2-in-1 process because of the improved
concentration of Mg2+ in highly concentrated sweet whey culture (specifically, the
concentration of total magnesium in 30% sweet whey is around 20 mmol l-1 as calculated from
the value in Table 7).
 Chapter 3: What underlies the protection 118

3.4 Conclusion

In this Chapter, we attempted to get insight into the mechanisms of the probiotic protection
effect of the 2-in-1 process.

Firstly, the stress response of bacteria during growth in highly concentrated sweet whey
could contribute to the enhanced tolerance of bacteria against harsh conditions during spray
drying. More specifically, cellular osmoregulation occurs when bacteria adapt to the hypertonic
condition in highly concentrated sweet whey. This osmoregulation includes the uptake or
synthesis of compatible solutes inside cells and the induction of osmotic and general stress
proteins, leading to cross-protection. Hence, the bacteria in highly concentrated sweet whey
display a multistress tolerant phenotype in comparison to the bacteria grown in isotonic
medium, leading to enhanced survival under the heat, osmotic, oxidative and dehydration
stresses during spray drying. Besides, we also explored the effect of magnesium salt (MgCl2)
on the thermotolerance of lactobacilli. The concentration of Mg2+ ranging from 10 to 50 mmol
l-1 was found to improve the bacterial survival under heat stress and to promote the bacterial
recovery from heat injury. It suggests that the influence of mineral salts on bacterial physiology
should not be ignored during thermal processing. Given the range of Mg2+ concentration in
highly concentrated sweet whey, it may be another factor which contributes to the protection
in the 2-in-1 process.

In addition to the effect of matrix composition, the presence of protein aggregates in

medium also showed a positive effect on improving bacterial survival during thermal
convective single-droplet drying, which highlights the importance of matrix sturcture. These
aggregates may moderate the excellular environment of bacterial cells during droplet-particle
transition through influencing the drying kinetics and/or particle formation.

Overall, the 2-in-1 process introduced in Chapter 2 is a complex system combining bacterial
growth and thermal convective drying. Therefore, the final viability of probiotics in powders
is thus influenced by multiple factors including growth conditions, feed physiochemical
properties, droplet-particle transition.
 Chapter 4: Scale-up of the 2-in-1 process 119

Chapter 4. Scale-up of the 2-in-1 Process in

Semi-Industrial Scale Spray Drying

In this chapter, the feasibility of scaling up the 2-in-1 process was validated with a semi industrial
pilot scale spray dryer. A multi-stage mild-conditions drying process, coupling spray drying with belt
drying and fluid-bed drying, was therefore applied in this work, in which the final probiotic survival
was improved to approximately 100% (> 109 CFU g-1).

Probiotic viability in the powders was monitored during a 6-months storage, which indicated that
storage temperature and moisture content of powders both played crucial roles in the stability of
probiotic powders. Moreover, the functionalities of spray-dried probiotic powders were also
investigated. Spray drying afforded a strain-dependent enhancement of bacterial survival in simulated
intestinal fluid, in comparison with fresh cultures.

The main contents in this Chapter have been published as:

Huang, S., Méjean, S., Rabah, H., Dolivet, A., Le Loir, Y., Chen, X.D., Jan, G., Jeantet, R., Schuck,
P., 2017. Double use of concentrated sweet whey for growth and spray drying of probiotics:
Towards maximal viability in pilot scale spray dryer. J. Food Eng. 196, 11–17.
Huang, S., Cauty, C., Dolivet, A., Le Loir, Y., Chen, X.D., Schuck, P., Jan, G., Jeantet, R., 2016a.
Double use of highly concentrated sweet whey to improve the biomass production and
viability of spray-dried probiotic bacteria. J. Funct. Foods 23, 453–463.
 Chapter 4: Scale-up of the 2-in-1 process 120

4.1 Introduction

Why did we investigate the scale-up feasibility of the 2-in-1 process?

Scale-up is a term in Chemical Engineering, which generally means the migration of a

process from the laboratory scale to the pilot-scale, and further to the industrial scale. To
validate the scale-up feasibility an innovation process is of prime importance for its industrial
application and further development. As scaling factors and laws are generally poorly known,
this represents often the bottleneck of innovation.

The studies on spray drying of probiotics, that were mostly performed on laboratory-scale
spray dryers, have been reviewed in Chapter 1 (Table 3). To the best of our knowledge, two
studies mentioned experiments on a larger scale with a water evaporation capacity in the range
100 to 1645 kg h-1 (Bielecka and Majkowska, 2000; Schuck et al., 2013). In comparison to
laboratory-scale spray drying, spray drying at pilot-scale or industrial scale should pay more
attention to concerns about the mixing and homogeneity of feed, pumping, atomization,
contamination, etc. On the other hand, the conditions of large scale drying, such as the generally
longer residence time may also differ the results of probiotic viability in powders which are
obtained from lab-scale drying. However, it is still ambiguous to conclude on the influence of
scaling-up on probiotic viability since the very few studies at larger scale drying.

Regarding the 2-in-1 process, to investigate the scale-up feasibility will help to address the
following questions:

1. Do the bacteria grow in highly concentrated sweet whey in large scale (i.e. concerns on
the mass transfer of nutrients during static fermentation)?

2. Can bacteria maintain stable during large scale drying (as residence time distribution of
particles is dependent of the dyer type and scale)?

3. Does the presence of aggregates impede the pumping and atomization?

4. Overall, does scaling up influence the bacterial viability in the final powders?

Why did we incorporate multi-stage drying in the process?

In order to minimize the heat inactivation of bacteria during spray drying, the drying
temperature can be moderated through technical innovation or process optimization, multi-
 Chapter 4: Scale-up of the 2-in-1 process 121

stage drying process for instance. (Schuck et al., 2013) has previously designed a multi-stage
drying process for the production of Propionibacterium acidipropionici powders. In this
process, the bacteria were cultivated in 13.5% (w/v) acid whey medium and then concentrated
by membrane ultrafiltration and powder addition prior to the spray drying. The spray dryer was
coupled with a crystallizer belt followed by a vibro-fluidizer. Due to the contribution of the
crystallizer belt and vibro-fluidizer stages to the overall water elimination, the first drying stage
could be subsequently limited, which allowed the setting of spray drying inlet temperature at
130°C and outlet temperature at 60°C, thus minimizing bacteria death.

However, it is still a prerequisite to provide the concentrated feed material for the first spray
drying stage in order to form the semi-dried but shell-formed particles, which are further dried
on the crystallization belt and fluidized bed. Regarding the 2-in-1 process, the probiotics were
already cultivated in highly concentrated sweet whey with 30% TS. Use of this culture with
high dry matter provided the possibility of skipping the intermediate concentration step. In
other words, the 2-in-1 process allowed probiotic fermentation to be directly coupled with the
multi-stage drying process, which may achieve the sustainable continuous production of high-
viability probiotics.

Is viability the only criteria to assess the powder quality?

The main output and challenge of probiotics spray drying is generally considered as the
preservation of bacterial viability during the drying process. In most studies, the viability after
spray drying refers to the culturable cell counts on the growth agar medium within a certain
time (Papadimitriou et al., 2016). Although this definition of viability can well represent the
amount of residual viable probiotics in the powders, it should not be regarded as the only
criterion to assess the quality of probiotic powders.

Probiotics within the powders should be able to pass through a long-term storage at low
temperature (2~8°C for most of commercial dairy products), or even at more challenging
temperature (ambient temperature at 25°C for instance) (Tripathi and Giri, 2014). Last but not
least, the probiotics rehydrated from powders should be resistant to the harsh stress during
digestion for several hours from gastric to intestinal tract to guarantee the delivery efficiency
and expected functionalities of probiotics (Cook et al., 2012). Moreover, the probiotics
rehydrated from powders should be challenged on their effective beneficial properties to the
host before validation of a drying processing sheme for their obtention. This point is of outmost
 Chapter 4: Scale-up of the 2-in-1 process 122

importance, although rarely investigated and reported to date in the literature (only 2 papers
mentionning this point, see Perspectives).
 Chapter 4: Scale-up of the 2-in-1 process 123

4.2 Materials and Methods

4.2.1 Bacterial strains and starter preparation

Lactobacillus casei BL23 and Propionibacterium freudenreichii CIRM-BIA 129 were used
in this work. L. casei was activated in MRS Broth and static cultivation at 37ºC for 16 h. P.
freudenreichii was activated in YEL broth and cultivated statically at 30ºC for 50 h.

Sweet whey medium with 30 wt% total solid content was prepared by rehydration of sweet
whey powder (Lactalis ingredients, Mayenne, France) in deionized water. For preparation of
starter culture, the 5 L sweet whey was autoclaved at 100°C for 30 min before inoculation of
L. casei or P. freudenreichii. The inoculation (1 v/v% inoculum size) was made from the above-
mentioned pre-culture of L. casei in MRS broth or of P. freudenreichii in YEL broth. The
inoculated sweet whey culture of L. casei was incubated statically at 37°C for 30 h, and P.
freudenreichii at 30°C for 72 h. The obtained culture was used as starter for further
fermentation.

4.2.2 Fermentation with sweet whey

Sweet whey medium for fermentation at the semi industrial pilot scale was prepared in a
steel tank (Goavec, Alençon, France) by rehydrating 150 kg sweet whey powders in 350 kg
water to obtain 30 wt% total solid content. The medium was then pumped through a scraped
surface heat exchanger (HRS Heat Exchangers, France) for heat treatment. The heating
temperature and the residence time of sweet whey medium within the heat exchanger were
120°C and 1 min, respectively. The heated sweet whey medium was then transferred into a 500
L bio-reactor (Goavec, Alençon, France). The pipes and the bio-reactor were previously treated
by steam. The above-mentioned 5 L starter culture was inoculated after the temperature of
sweet whey had been cooled down to the setting growth temperature (i.e. 37°C for L. casei and
30°C P. freudenreichii). The sweet whey with L. casei was fermented statically at 37 °C for 48
h, and P. freudenreichii at 30°C for 96 h.

4.2.3 Spray drying

Before spray drying, the sweet whey probiotic culture was agitated moderately for 20 min.
Three processes of spray drying were performed for each probiotic strain (Figure 36).
 Chapter 4: Scale-up of the 2-in-1 process 124

(cooling)
(without drying)

Figure 36. The schematic diagram of three drying process: (A) Lab-scale spray drying (for obtaining
Powder 1), (B) Semi industrial pilot scale one-stage spray drying (for obtaining Powder 2) and (C) Semi
industrial pilot scale multi-stage spray drying (for obtaining Powder 3).

For the lab-scale spray drying (Figure 36A), 1 L sweet whey culture was pumped to a Mobile
Minor™ spray dryer (GEA Niro A/S, Denmark). A two-fluid spray nozzle with an orifice
diameter of 0.8 mm was used. The evaporation rate of this dryer was approximately 3 kg h-1.
The inlet air temperature was at 140 ± 1°C, and the outlet air temperature 60 ± 3°C. The one-
stage semi industrial pilot scale spray drying (Figure 36B) was performed in the Bionov spray
dryer pilot workshop (Niro Atomizer, GEA, Saint Quentin en Yvelines, France) based in
Rennes (France). A pressure nozzle with an orifice diameter of 0.73 mm was used. The
evaporation capacity was approximately 80 kg h-1. The drying temperatures were set to the
same values as in minor dryer, namely inlet air temperature at 140 ± 5°C and outlet air
temperature at 60 ± 3°C. A belt (GEA, Saint Quentin en Yvelines, France) and a vibro-fluidizer
(GEA, Saint Quentin en Yvelines, France) were used following the spray drying step (Figure
36B), but respectively for conveying (turn-off status) and powder cooling purposes in this
configuration. The multi-stage semi industrial pilot scale spray drying (Figure 36C) was also
performed using the Bionov spray dryer pilot workshop, as described before (Schuck et al.,
2013). Briefly, the inlet air temperature of spray dryingwas decreased to 127 ± 3°C, and the
outlet air temperature 47 ± 2°C. After spray drying, the partially dried powder was delivered
through the belt dryer at ambient temperature for crystallization purpose. The residence time
 Chapter 4: Scale-up of the 2-in-1 process 125

of powders on belt was approximately 5 min. Fluid-bed drying was then carried out in a vibro-
fluidizer (VF) with inlet temperature at 80 ± 2°C and outlet temperature at 40 ± 2°C.

4.2.4 Analysis of physical and chemical properties

For liquid samples, the pH value was measured with a pHmeter (Ecolab, Issy-l_es-
Moulineaux, France). Viscosity measurements were performed using an AR 2000 rheometer
(TA instruments, Guyancourt, France) equipped with coaxial cylindrical geometry (stator inner
radius: 25 mm; rotor outer radius: 23 mm; immersed cylinder height: 30 mm; bottom cap: 4000
mm). Apparent viscosity was determined at 20°C using the HerscheleBulkley model at a shear
rate of 1 s-1.

The water content of liquid or powder samples was measured according to the method
described by Schuck et al. (2012): the sample (respectively 5 g for liquid samples and 1 g for
powder samples) was mixed with 25 g pre-dried sand and then dried at 105°C for 7 h (liquid
samples) or 5 h (powder samples). The water activity of powders was determined at 25°C using
an aw-meter (Novasina, aw-center 92T0003) immediately after drying and cooling.

4.2.5 Storage of powders

The powders were collected and aliquoted in PA/PE plastic vacuum bags (La Bovida,
France). The bags were then sealed in the presence of air or under vacuum conditions (Britek,
France) respectively. These powders were stored at the controlled temperature of 4°C and 25°C
and kept away from light.

4.2.6 In vitro simulated digestion

The effect of spray drying on the probiotic tolerance against digestion stress was
investigated in the in vitro simulated digestion experiment. The powders obtained from multi-
stage drying process (Figure 36C; powder 3) were used to compare with the fresh probiotic
culture in 30% sweet whey. The powders were rehydrated to reconstitute the 30% probiotic
suspension. The simulated gastric fluid (SGF) and simulated intestinal fluid (SIF) were
prepared according to Minekus et al. (2014). In the gastric phase, the probiotic samples were
mixed with SGF (1:1 v/v) containing 2000 U mL-1 porcine pepsin, and incubated at 37°C for 3
h under agitation. In the intestinal phase, the gastric digesta were mixed with SIF (1:1 v/v)
 Chapter 4: Scale-up of the 2-in-1 process 126

containing 10 mmol l-1 porcine bile extract and porcine pancreatin with 100 U mL-1 trypsin
activity, and incubated at 37°C for 3 h under agitation. The pH for the gastric and intestinal
phases were adjusted to 3.0, and 7.0 respectively. The viability of probiotics during simulated
gastric or intestinal digestion were tested in 1 h interval.

4.2.7 Enumeration and quantification of bacteria

The enumeratioin and quantification of bacteria were performed as described in Item 2.2.8.
Briefly, the populations of probiotics in liquid samples were serial diluted with peptone water
(0.1% w/v) and then poured onto MRS or YEL agar plates. The powder samples were
rehydrated by dissolving 3 g of powder in 7 g of peptone water prior to serial dilution. L. casei
was incubated at 37 °C for 48 h and P. freudenreichii at 30 °C for 6 days under anaerobic
conditions for colony counting (Anaerocult®, Merck KgaA, Germany). Percent survival was
then determined by comparing bacteria population before and after spray drying or simulated
digestion treatment.

Besides, total thermophilic and coliform flora populations were evaluated according to the
methods described previously (Schuck et al., 2013).

4.2.8 Statistical analysis

All the results displayed in this work were obtained from triplicate samples. The results are
presented as mean value with standard deviation. The analysis of variance (ANOVA) followed
by Tukey test was performed using R software with the ‘Rcmdr’ package (R Development
Core Team). Differences between mean values were considered significant when p < 0.05.
 Chapter 4: Scale-up of the 2-in-1 process 127

4.3 Results and Discussion

4.3.1 Validation the feasibility of scaling up the 2-in-1 process

The bacterial populations of L. casei and P. freudenreichii after fermentation in 500 L 30%
sweet whey were (1.6 ± 0.1) × 109 and (5.0 ± 0.6) × 109 CFU mL−1, respectively. These results
are in agreement with our previous work at laboratory-scale described in Item 2.3.1 (1–2 L): it
validates that the two probiotic strains are able to yield high biomass production at a semi
industrial pilot scale when cultivated in 30% sweet whey without casein peptone
supplementation. Besides, the viscosity of sweet whey significantly decreased after being
fermented by both probiotic strains (Table 14), due to the lactose consumption during
fermentation, subsequent mass transfer to the biomass and last biomass decantation. More
specifically, the large amount of lactose, which contributes considerably to the viscosity of
sweet whey, was mainly metabolized into lactic acid and propionic acid by L. casei and P.
freudenreichii, respectively. The control of the viscosity of feed at a given total solid content
is generally desired by industries in order to facilitate atomization during spray drying (Cal and
Sollohub, 2010).

Table 14. The physical properties of feed concentrates and powders.

Viscosity Water Content
Sample pH Water activity
(mPa·s) w/w%
Heat treated media 6.3±0.1 33.0±0.2 70.7±0.1 -

L. casei culture 4.4±0.1 22.0±0.4 71.0±0.0 -

a
Powder 1 - - 5.3±0.2 0.27±0.02a
Powder 2 - - 5.2±0.2a 0.27±0.01a
Powder 3 - - 5.5±0.1a 0.28±0.01a

P. freudenreichii
5.7±0.0 23.0±0.1 71.2±0.0
culture
Powder 1 - - 5.0±0.2a 0.27±0.01a
Powder 2 - - 5.2±0.1a 0.27±0.01a
Powder 3 - - 6.0±0.2b 0.30±0.01a

The different superscript indicates the significant difference (p < 0.05).

 Chapter 4: Scale-up of the 2-in-1 process 128

Spray drying was carried out in three different processes as illustrated in Figure 36. The
powders obtained from the semi industrial pilot scale process and lab-scale drying were non-
significantly different regarding water content and water activity, except for the water content
of P. freudenreichii powder 3 (Table 14). It may be caused by the variation of drying parameters
during process (Písecký, 2012). Nevertheless, the water contents and water activities were at
the whole within a same range, so that the viability results could be compared between different
processing shemes.

For the two probiotic strains, the remaining viabilities after lab-scale and one-stage semi
industrial pilot scale spray drying (Figure 36A and B) were non significantly different, the
survival being at around 60% for L. casei (remaining population ∼1.0 × 109 CFU g−1) and
100% for P. freudenreichii (remaining population ∼4.5 × 109 CFU g−1) (Figure 37). The
comparison between two strains indicated that the robustness of P. freudenreichii upon spray
drying was higher than that of L. casei. It is known that P. freudenreichii can accumulate
intracellular trehalose, glycogen and polyphosphate, which are involved in improving bacterial
tolerance against heat and/or desiccation stress (Anastasiou et al., 2006; Leverrier et al., 2004).
This thermotolerance acquisition ability is in agreement with the ability of P. freudenreichii to
withstand harsh conditions of Swiss-type cheese making, including thermal treatment (52°C,
30–60 min) and saline stress caused by immersion (48–72 h) in saturated brine. These results
were fully consistent with the bacterial survival previously obtained at lab-scale spray drying
(Item 2.3.3), albeit survival was further increased due to the lower drying temperature used in
this work. Therefore, the feasibility of scaling up the 2-in-1 process (i.e., using highly
concentrated sweet whey to grow and spray probiotics) is validated by this work.

A B
5,E+09
[Link] population

1,E+10 120
[Link] population in
powders (CFU/g)

90
in powders (CFU/g)
Survival (%)

Survival (%)

1,E+09 100
70
5,E+08

50 1,E+08 80

5,E+07 30 1,E+07 60
P1 P2 P1 P2
Powders from different processes Powders from different processes
Bacterial population Survival Bacterial population Survival

Figure 37. Survival of probiotics after spray drying at lab-scale or semi industrial scale. Bacterial
population (refers to column, left Y-axis) and survival (refers to curve, right Y-axis) of (A) [Link]
BL23 and (B) [Link] ITG20 after spray drying. P1 refers to the powders from lab-scale spray
drying, and P2 refers to that of from semi industrial pilot scale.
 Chapter 4: Scale-up of the 2-in-1 process 129

4.3.2 Improvement of probiotic viability through multi-stage drying process

When using the multi-stage semi industrial pilot scale spray drying (Figure 36C), the
probiotic survival was significantly improved to approximately 100% (p < 0.05) for L. casei,
and slightly improved for P. freudenreichii (p < 0.2) (Figure 38). This process was previously
conducted on two P. acidipropionici stress-tolerant strains (Schuck et al., 2013). However and
given the robustness of the two strains, the survival of bacteria was shown to be 100%
throughout the whole process. In this work, the multi-stage semi industrial pilot scale spray
drying was performed for the first time on L. casei, which represent commonly commercialized
and more fragile probiotics. The survival of L. casei was increased from 60% to 100%.
Compared to the single-stage drying mode, the particles containing probiotics suffered here
less heat stress at the first stage of spray drying as the outlet temperature was lowered from
around 13°C. In accordance, the viability of bacteria was higher at the end of spray drying
(Zhang et al., 2016). Although the bacteria experienced another heat stress at the second stage
of fluid-bed drying (80°C inlet temperature and 40°C outlet temperature), the cells were already
embedded in the partially dried powders. Moreover, it has been reported that the heat resistance
of bacteria in a dry state, especially in the aw range of 0.3–0.5, was higher than that of the same
bacteria in a wet state (Laroche et al., 2005). Therefore, this process is more suitable for such
heat sensitive bacteria, in comparison with the one-stage semi industrial pilot scale spray
drying.

A B
1,E+10 140
[Link] population in

1,E+10 120
*
[Link] population in powders

100 120
Survival (%)

powders (CFU/g)

Survival (%)

100
1,E+09 80 1,E+09
80
(CFU/g)

60
60
1,E+08 40 1,E+08
40
20
20
1,E+07 0 1,E+07 0
P2 P3 P2 P3
Powders from different processes Powders from different processes
Bacterial population Survival Bacterial population Survival

Figure 38. Survival of probiotics after one-stage spray drying or multi-stage drying at semi industrial
scale. Bacterial population (refers to column, left Y-axis) and survival (refers to curve, right Y-axis) of
(A) [Link] BL23 and (B) [Link] ITG20 after spray dryingin the semi industrial pilot scale.
P2 refers to the powders from one-stage spray drying, and P3 refers to that of from multi-stage drying
process. * means significant difference between the survival rates (p<0.05)
 Chapter 4: Scale-up of the 2-in-1 process 130

In addition to the targeted probiotic strains, thermophillic and coliform flora populations
were also evaluated in the powders after spray drying as indicators of contamination.
Thermophilic and coliform floras represent common microbiological spoilage in dairy powders
and constitute a major concern in controlling the safety of foods at industrial scale production
(Breeuwer, 2014; Chandan et al., 2015). As shown in Table 15, low amount of thermophilic and
coliform floras initially contaminated the sweet whey medium, despite the heat treatment
applied and the previous disinfection of the equipment: the conditions may not have been strict
enough to achieve a fully aseptic medium. However, the bacterial population of thermophilic
and coliform floras kept constant without break out during probiotic fermentation, probably
due to the competitive inhibition (Verschuere et al., 2000; Wilderdyke et al., 2004). After spray
drying, the coliform flora was undetectable in the powders but the thermophilic flora still
remained. This indicates that the coliform flora is sensitive to spray drying stress, while the
thermophilic flora is more resistant during spray drying probably due to its higher heat
resistance. Therefore, more attention should be paid on controlling the thermophilic flora
before spray drying.

Table 15. The enumeration of contaminating flora (thermophillic and coliform floras)
Thermophillic flora Coliform flora
Sample
(log CFU g-1) (log CFU g-1)
Heat treated media 2.6±0.1 2.8±0.0

L. casei culture 2.4±0.0 2.8±0.1

Powder 1 2.6±0.0 ND
Powder 2 2.4±0.1 ND
Powder 3 2.2±0.2 ND

P. freudenreichii culture 1.9±0.6 2.0±0.2

Powder 1 1.4±0.6 ND
Powder 2 1.4±0.6 ND
Powder 3 0.9±0.2 ND
ND means undetectable in plate agar counting method
 Chapter 4: Scale-up of the 2-in-1 process 131

4.3.3 Storage stability of probiotic powders

Probiotics stability upon storage was compared among the powders obtained from different
drying processes. For L. casei at 4°C (Figure 39A), the viability kept constant during the first 2
months. A one log CFU g−1 reduction of viability was reached after 6 months storage. In
contrast, the L. casei viability devreased gradually after 7 days at 25°C (Figure 39B). Within
one month, the viability loss was around 2 logs CFU g−1 whatever the drying process and scale
considered. After 6 months storage, the L. casei viability remained at approximately 4 logs
CFU g−1. Moreover, the conditioning environment (vacuum or atmosphere) had no significant
effect on the stability of L. casei powders at both 4°C and 25°C (Figure 39).

A
1,0E+10
The [Link] population in

1,0E+09
powders (CFU g-1)

1,0E+08
1,0E+07
1,0E+06
1,0E+05
1,0E+04
0 7 30 60 180
Storage time (days)
P1 P2 P3 P2-vacuum P3-vacuum

B
1,0E+10
The [Link] population in

1,0E+09
powders (CFU g-1)

1,0E+08
1,0E+07
1,0E+06
1,0E+05
1,0E+04
1,0E+03
0 7 30 60 180
Storage time (days)

P2 P3 P2-vacuum P3-vacuum

Figure 39. Storage stability of [Link] BL23 powders from different drying processes (expressed as
reduction of survival) during 6 months storage at (A) 4°C and (B) 25°C. P1 refers to the powders from
lab-scale spray drying, P2 refers to that of from one-stage spray drying at semi industrial pilot scale,
and P3 refers to that of from multi-stage drying process at semi industrial pilot scale. P2-vacuum and
P3-vacuum mean the powders stored under vacuum.
 Chapter 4: Scale-up of the 2-in-1 process 132

Similarly, the viability of P. freudenreichii in powders also kept constant at 4°C for the first
2 months and reduced by around 1 log CFU g−1 after 6 months storage whatever the drying
process considered (Figure 40A). However, at this time, the powders from pilot scale one-stage
drying process (P2) displayed slightly better stability when conditioned under vacuum. At
25°C, a larger viability loss was also observed in all powders, reaching around 5 logs CFU g−1
after 6 months storage under these conditions (Figure 40B).

A
The [Link] population

1,0E+11
1,0E+10
in powders (CFU g-1)

1,0E+09
*
1,0E+08
1,0E+07
1,0E+06
1,0E+05
1,0E+04
0 7 30 60 180
Storage time (days)

P1 P2 P3 P2-vacuum P3-vacuum

B
The [Link] population

1,0E+11
1,0E+10 *
in powders (CFU g-1)

1,0E+09 **
1,0E+08
1,0E+07
1,0E+06
1,0E+05
1,0E+04
0 7 30 60 180
Storage time (days)

P2 P3 P2-vacuum P3-vacuum

Figure 40. Storage stability of P. freudenreichii ITG20 powders from different drying processes
(expressed as reduction of survival) during 6 months storage at (A) 4°C and (B) 25°C. P1 refers to the
powders from lab-scale spray drying, P2 refers to that of from one-stage spray drying at semi industrial
pilot scale, and P3 refers to that of from multi-stage drying process at semi industrial pilot scale. P2-
vacuum and P3-vacuum mean the powders stored under vacuum. Presence of * means significant
difference in population compared to other groups at the same storage time (p < 0.05). Different number
of * indicates significant difference between each other (p < 0.05).
 Chapter 4: Scale-up of the 2-in-1 process 133

This result highlights the predominant influence of the storage temperature on the stability
of probiotic powders. Moreover, the P. freudenreichii P2 powder stored under vacuum at 25°C
showed a better stability in the first 2 months, compared to same powder stored under regular
atmosphere: indeed, the corresponding viability losses at this time were lower than 1 log CFU
g−1 and approximately 2 logs CFU g−1, respectively. By comparison, the viability of P.
freudenreichii in powders was more stable than that of L. casei in powders. Such a difference
may be attributed to the better ability of P. freudenreichii cells to accumulate energy and carbon
storage compounds such as polyphosphate, trehalose, glycogen, glycine betaine (previously
discussed in Item 3.3.1), as well as the absence of their lysis during stationary either dormant
phases, which allows long-term survival even under carbon starvation (Aburjaile et al., 2016;
Falentin et al., 2010a).

Last, we observed a significant better stability of P. freudenreichii in P2 powders compared

to P3 powders at 6 months storage under 4°C and at 2 months storage under 25°C. This could
be explained by the detrimental effect of higher water content in P3 powders (Table 14)
(Vesterlund et al., 2012). Besides, storage under vacuum improved the stability of P.
freudenreichii powders at both 4 and 25°C, but not that of L. casei powders. The reason may
be the facultative anaerobic phenotype of L. casei, leading to oxidative stress tolerance.
However, storage under vacuum didn't make it possible to enhance the P. freudenreichii
viability in the P3 powder. It may indicate that the influence of moisture is more detrimental
than the oxidation on the stability of probiotic powders in long-term storage.
 Chapter 4: Scale-up of the 2-in-1 process 134

4.3.4 Resistance of rehydrated probiotics to simulated digestion

The simulated digestion of probiotic powders from pilot scale multi-stage drying process
(P3 powders) was investigated in comparison with the fresh 30 w/w % sweet whey culture
(Figure 41). The results showed that both strains were resistant to the simulated gastric fluid
(SGF) within 3 h regardless of the rehydrated or fresh culture type. It may be explained by the
large amount of whey proteins in concentrated sweet whey (around 2.6 w/w %), which can
protect bacteria from digestion stress through its buffer effect (Chen et al., 2006; Doherty et
al., 2011; Tavares et al., 2014). Besides, it has also been previously reported that the osmolality
of concentrated sweet whey could trigger the multi-stress tolerance of bacteria during growth,
leading to the higher survival of bacteria under acid and bile salt stresses (discussed in Item
3.3.1).

A B
140% 120%
Survival of L. casei in SIF

30% sweet whey culture

Survival of L. casei in SGF

120% 100% Powder 3

100%
80%
80%
60%
60%
40% 40%
30% sweet whey culture
20% 20%
Powder 3
0% 0%
0 1 2 3 0 1 2 3
Time (h) Time (h)

C D
140% 120%
freudenreichii in SGF

freudenreichii in SIF

120% * *
Survival of P.

100%
Survival of P.

100% *
80%
80%
60%
60%
40%
40%
30% sweet whey culture 30% sweet whey culture
20% 20%
Powder 3 Powder 3
0% 0%
0 1 2 3 0 1 2 3
Time (h) Time (h)

Figure 41. The survival of probiotic in 30 wt% fresh culture and powders from Multi-stage drying
process (P3), (A) [Link] BL23 in simulated gastric fluid (SGF), (B) [Link] BL23 in simulated
intestinal fluid (SIF), and (C) P. freudenreichii ITG20 in simulated gastric fluid (SGF), (D) P.
freudenreichii ITG20 in simulated intestinal fluid (SIF).
 Chapter 4: Scale-up of the 2-in-1 process 135

However, the viability of two probiotics both decreased gradually in the simulated intestinal
fluid (SIF) within 3 h. It may be caused by the effects of bile salt and trypsin, since there is
barely lipid in the sweet whey to buffer the bile salt and the whey proteins would be digested
to a large extent. Interestingly, P. freudenreichii powders showed a stronger tolerance,
compared to the fresh culture, during simulated intestinal digestion (Figure 41D). However, this
enhanced tolerance did not emerge on the L. casei strain (Figure 41B). It indicates that the heat,
osmotic or oxidative stress underwent during spray drying or the desiccate environment in the
spray-dried powders may further trigger the stress tolerance of bacteria and this cellular stress
response may be strain-dependent. This hypothesis however remains further research.
 Chapter 4: Scale-up of the 2-in-1 process 136

4.4 Conclusion

In this chapter, the 2-in-1 process (i.e., use of highly concentrated sweet whey for growth
and spray drying of probiotics) was successfully scaled up at a semi industrial pilot scale (500
L fermentation and spray drying) for two well-documented probiotic strains. Indeed, a high
viability of probiotics was obtained after spray drying.

The multi-stage drying process was shown to further improve the probiotic survival rate,
compared to conventional one-stage spray drying. This multi-stage drying process can
represent a win-win strategy with the 2-in-1 process due to the benefits of utilizing highly
concentrated sweet whey (30% TS). This couping could thus be used in protecting more heat
sensitive microorganisms in the future.

The storage temperature and powder moisture content were found to be the key factors
influencing the stability of probiotic powders. This may suggest that the viability of probiotics
after drying should not be considered as the only criteria driving the quality of probiotic
powders. From a long-term stability point of view, it may be worth to increase the drying
temperature to some extent, hence obtaining probiotic powders with ideal water
content/activity while sacrificing a minimum viability during drying.

In addition, spray drying was found to improve the tolerance of P. freudenreichii against
simulated intestinal fluid. It suggests the possibility of cross protection from spray drying stress
to digestion stress, which would be bacterial strain dependent.
 General Conclusion 137

General Conclusion

The market demand for functional probiotics-containing foods is facing an exponential

growth due to the growing standard of living and health awareness at a global scale. Spray
drying holds great promises for sustainable and large-scale production of probiotic powder
ingredients. Although it has been widely applied in dairy and pharmaceutical industries at a
global scale, its use for producing powders of probiotics or other food-grade microorganisms
remains a major challenge due to the concomitant undesired loss of cell viability. Improving
the quality of spray-dried probiotics thus warrants more investments in research and
development to gain better understanding of the process-probiotics compatibility and probiotic
physiology-functionality interaction. In this project, we attempted to design an industry and
environment-friendly process for spray drying of probiotics, and provide insights into the
interaction between bacterial physiology and processing. The obtained results can be
considered as the answers to the following questions:

What are the advantages of the novel 2-in-1 process?

In the present project, a simplified process from growth to spray drying of probiotics was
developed, using a dairy by-product (sweet whey) in a concentrated form (~30% dry matter)
as a 2-in-1 medium to both culture probiotic bacteria and protect them from spray drying injury.
This innovative process, briefly named 2-in-1 process, has been patent-protected.

In the 2-in-1 process, the usual intermediate steps between bacterial growth and spray
drying (such as harvesting, washing, re-suspending) were skipped and the probiotic metabolites
were retained. In fact, instead of being inhibited by the hypertonic conditions in 30% sweet
whey, probiotic L. casei and P. freudenreichii were found to yield higher biomass than that in
isotonic 5% sweet whey. The growth of L. casei was less dependent on casein peptone
supplementation when increasing the dry matter of sweet whey from 5% to 30%. As a liquid
culture medium, sweet whey also enabled longer survival of L. casei in stationary phase than
in MRS broth at 37°C. More importantly, the survival of probiotics after spray drying and their
stability during storage at 4°C were significantly improved after growth in concentrated
(hypertonic) sweet whey in the 2-in-1 process, in comparison to the 5% isotonic sweet whey.
 General Conclusion 138

What contributes to the improvement of probiotic survival in the 2-in-1 process?

Use of highly concentrated sweet whey as a bacterial growth medium is one of the most
important features for the 2-in-1 process. The stress response of bacteria during growth in this
hypertonic medium was hypothesized to be a main reason for the enhanced bacterial multistress
tolerance (i.e. multistress tolerance conferred by osmotic adaptation). In this work, we
conducted an experiment with P. freudenreichii to demonstrate this hypothesis. The
osmoregulation was proven in the P. freudenreichii sweet whey culture (30% dry matter) by
evidencing the accumulation of compatible solutes (trehalose, glycogen, polyphosphate) and
the upregulation of several proteins in charge of osmoregulation (e.g., ATP-binding protein
OpuCA of osmoprotectant ABC transporter). At the same time, the induction of multistress
tolerance was demonstrated through the improved bacterial survival upon heat, acid, bile-salt,
spray drying treatments, and the upregulation of universal stress proteins (e.g., heat shock
proteins 20, Chaperone ClpB 2).

Besides, Mg2+ ranging from 10 to 50 mmol l-1 also showed a positive effect on the
thermotolerance of bacteria, which may contribute to the protection of highly concentrated
sweet whey displayed in the 2-in-1 process.

In addition to cross protection and Mg2+ effect, the calcium-induced milk protein aggregates
also displayed a significant effect on improving the protective capacity of milk on bacteria
during thermal convective drying. This improved protection is suggested to link to the distinct
drying kinetics and droplet-particle transition of the aggregates-containing milk. It highlights
the contribution of matrix structure to probiotic protection during drying.

Although the correlation between Mg2+ or calcium-induced protein aggregates with the
highly concentrated sweet whey requires further investigations, these results clearly suggest
that, apart from the well-known protection effects of sugar and proteins, the mineral salts may
also play important roles in probiotic survival during drying, either through influencing the
bacterial physiology directly (i.e., composition effect), or through changing the protective
capacity of matrixes (i.e., structure effect).

How to scale up the 2-in-1 process, and for which powder quality?

The feasibility of scaling up the 2-in-1 process was validated at a semi industrial pilot scale
(500 L fermentation and spray drying). A high residual viability of probiotics was obtained in
the spray-dried powders. Besides, the 2-in-1 process is easy to couple with other industrial
 General Conclusion 139

processes for the purpose of decreasing drying temperature given the high solid content of the
probiotic sweet whey culture. For example, we showed that the 2-in-1 process can be well
coupled with a multi-stage drying process, resulting in a further improvement of probiotic
viability in the final powders. However, it is suggested that the viability after spray drying
should not be considered as the only criteria of the quality of probiotic powders. Given the
importance of moisture content and water activity for the probiotic storage stability, it may be
more reasonable to sacrifice a minimum viability of probiotics through increasing drying
temperature, in order to obtain powders with ideal moisture content (<4%) and water activity
(<0.2).

As another important property of probiotics, the digestion resistance of spray dried

probiotics was also investigated. P. freudenreichii rehydrated from spray dried powders was
found to be more resistant against the simulated intestinal fluid. However, this improvement
wasn’t found in L. casei. This suggests the potential strain-dependent cross protection between
spray drying stimuli and digestion stress tolerance.

Overall, this work will open a new perspective for large-scale production of probiotics in
the future. It offers an alternative to the currently used freeze drying with the following main
advantages:

- Continuous production of probiotics via spray drying

- Use of the food-grade dairy byproduct (sweet whey) throughout the process
- Further improved energy efficiency and productivity of spray drying probiotics
- Producing probiotic powders with high bacterial viability and resistance phenotype
- Avoid the intermediate steps between fermentation and drying
- Keep the probiotic metabolites in final powders
 Future Perspectives 140

Future Perspective

To achieve the aim of using spray drying for probiotic production/encapsulation, more
studies should be conducted to provide insights into the correlations of processing, bacterial
physiology and probiotic functionality.

Surrounding the 2-in-1 process (short-term perspective)

Does osmoregulation influence probiotic functionality?
Whether or not spray drying process influences the probiotics functionality remains a key
concern. To the best of our knowledge, only two in vivo studies investigated this potential
influence using mice model: A higher number of Immunoglobulin A (IgA)-producing cells in
the small intestine were induced in mice by spray-dried cultures, when compared with fresh
cultures (Páez et al., 2013). Another recent study also reported that spray drying process
preserved the immunomodulatory properties of B. lactis strain on acute and chronic colitis in
mice, as no significant differences were observed between fresh and spray-dried bacteria
(Burns et al., 2017). However, it should be noticed that the probiotic powders in these two
studies were both produced with a lab-scale spray dryer (Buchi B290, Flawil, Switzerland) with
probiotic survivals close to 100%. This high survival is generally unexpected in lab-scale spray
drying due to the intensive energy required to evaporate the moisture. It may not be
representative of the influence of real industrial spray drying process on bacterial physiology.
Besides, a closer to human animal model, porcine for instance, should be considered in order
to better simulate the human digestion.

We have highlighted in our study the enhanced acid and bile-salt resistance in the 30%
sweet whey culture in comparison to the 5%. It indicates that the probiotics grown in 30%
sweet whey may survive better during delivery to gastrointestinal tract. However, considering
the change of bacterial physiology during osmoregulation, more studies should be carried out
in vitro and in vivo to investigate in detail the potentially modified functionalities of probiotics
obtained from the 2-in-1 process. For example, does the 2-in-1 process influence the expression
of P. freudenreichii CIRM-BIA 129 surface proteins, hence affecting its immunomodulatory
effects (Le Maréchal et al., 2015)? Similarly, does it influence the expression and activity of L.
casei BL23 sortases, hence affecting its adhesive and immunomodulatory effects (Muñoz-
Provencio et al., 2012)?
 Future perspective 141

Free combination of elements to create a new 2-in-1 process

The applicability of the 2-in-1 process could be further extended. The key innovation of the
2-in-1 process is utilizing bacterial osmoregulation to link bacterial growth with drying process.
Four key elements (i.e., microorganisms, medium, stress, production) can be extracted from
the 2-in-1 process as shown in Figure 42:

Sweet Osmotic
Bacteria
Whey adaptation Spray
drying

Figure 42. The key four elements of the current 2-in-1 process. To further extend the applicability of
this process, each element is a matter of investigation to form a new technical route in terms of probiotic
production.

First, more probiotic species and strains should be tested with regard to this process in order
to extend its applicability and demonstrate its potential genericity. The challenge remains to
explore whether or not this process could be efficient in encapsulating more fragile bacteria
such as bifidobacteria. It would be interesting to understand the difference of osmoregulation-
pattern among different bacterial species/strains. Moreover, the possibility of cultivating multi-
species bacteria instead of single strain should be explored. The species-species interaction and
quorum sensing issue may exert significant effects on bacterial physiology during growth,
stress adaptation and drying.

Second, the 2-in-1 medium could be replaced by other food or pharmaceutical-grade

materials, the only constraint being the food-grade feature of the medium and the ability of its
nutrients to sustain the growth of bacteria. However, it is also possible to add some basic
nutrients in a given material to form a 2-in-1 medium. In order to support the choice and
formulation of medium, it is of prime importance to better understand the role of each
component: is it possible to identify a key component (or group of components) to sustain
bacterial growth? What component(s) is(are) able to trigger osmotic stress, overall to provide
protection during drying? On the contrary, what may induce negative effects on bacterial
biomass yield or survival during drying? A clear understanding of the role of each component
will provide great help in developing a chemically defined food grade medium, which may
 Future perspective 142

promote the improvement of biomass yield, bacterial resistance before drying, and maximize
the bacteria viability after drying.

Third, in the 2-in-1 process, concentrated medium promotes bacterial tolerance in the
growth unit, and also provides high dry matter and production yields for the drying unit. In
other words, osmotic adaptation links growth and drying on the basis of its benefits for both
sides. It would be interesting to explore the possibility of other cross benefits apart from
osmotic adaptation, for instance on thermal, oxidative or acid adaptation. Take thermal
adaptation as an example: Bacteria could be cultivated in an isotonic 2-in-1 medium with the
purpose of producing maximal bacterial population. Then this bacterial culture could be
delivered to a concentration unit with sublethal heating to trigger the thermal adaptation (falling
film evaporator for instance). Once dry matter achieves the desirable range (15~40%), this
thermal adapted culture could be delivered to spray drying. Moreover, exploring the possibility
to trigger oxidative stress resistance seems at this point of higher interest.

Last but not least, as the induced cross-protection is not specific to spray drying, the 2-in-1
process could be also efficient when applied to freeze drying for probiotic production purposes.
The high dry matter is also desirable for freeze drying from the energy efficiency point of view.
It is also interesting to explore the possibility of cross protection from osmotic stress to cold
stress, as well as the glass transition behavior of concentrated 2-in-1 medium during freeze
drying.

From probiotic-delivery to functionality-delivery (Long-term perspective)

Tailor the powders: Making every cell matter
Nowadays, extensive studies have been carried out in the aim of maximal delivery of viable
probiotics to the gastrointestinal tract (GIT). Spray drying is one of the methods used to
encapsulate probiotics within a drying matrix, hence protecting probiotics cells from digestion
stress. The protection here mostly refers to maintenance of bacterial viability within the GIT.
However, does the viability of probiotics guarantee its probiotic functionality?

It should be noted that, in most studies, the viability of probiotics refers to the metabolism
and cell-division abilities of bacteria in their favored growth medium (Papadimitriou et al.,
2016). These abilities may not completely reflect the successful delivery of probiotic
functionality. For example, most of immunomodulatory effects of probiotics are induced by
the interaction between probiotic surface molecules and host intestinal mucosa (Bron et al.,
 Future perspective 143

2011; Lebeer et al., 2010). The loss or inactivation of these molecules during digestion may
not destroy the metabolism and cell-division abilities of probiotics, but it would result in the
silencing of beneficial effects conferred by probiotics. In contrast, encapsulated probiotics may
also lose the interaction with host mucosa due to the shield effect of drying matrix. This latter
could also alleviate the beneficial effects of probiotic metabolites, such as bacteriocin, SCFAs,
vitamin and secreted proteins. Therefore, controlling the delivery of probiotic functionality is
the next key step in order to tailor the probiotics chain. To guarantee the successful delivery
of probiotic functionality, an encapsulation strategy should be designed on the basis of a
better understanding of the specific probiotic strains and of the action mechanisms of
their probiotic effects. For example, core-shell particles or porous particles could constitute
ideal vehicle to solve the above-mentioned concerns in probiotic-functionality delivery (Figure
43).

Figure 43. Schematic diagram of two hypothetic particles used to encapsulate and deliver probiotics
which have two different action mechanisms. (A) Layer-by-layer encapsulation for delivery of
probiotics that exert their beneficial effects through the interaction between their surface molecules and
host intestinal cells or mucosa (inspired by Anselmo et al., 2016): (1) The first layer of shell protects
the particle through gastric phase; (2) The second layer of shell is gradually digested in intestinal phase
to partly protect the particle core from the bile and gradually release the particle core; (3) The particle
core is collapsed and release the probiotics into the intestinal phase. (B) Encapsulation of porous core
particle for delivery of probiotics that exert their beneficial effects through secretion of beneficial
molecules (inspired by Mahbubani et al., 2014 and Nandiyanto and Okuyama, 2011): (1) The
encapsulated layer protects the particle core through gastric phase; (2) The porous particle core allows
water to penetrate inside to rehydrate and to activate probiotics, but hinders the bile salt to penetrate
inside; (3) The porous particle core allows the release of probiotic beneficial secretions.
 Future perspective 144

Although it is possible to produce powders with different particle morphologies and

structures (Figure 4) via spray dying (Nandiyanto and Okuyama, 2011), and thus theroretically
to optimize the encapsulation features with regard to the susceptibility of probiotics, tailoring
such powders in terms of probiotic encapsulation remains practically challenging. Besides, the
other difficulty relates to the range of available materials, which should be food-grade and
affordable. From this point of view, milk-protein based (eg, WPI) materials seem promising
(Tavares et al., 2014), but in the same time are expensive for large-scale production. Therefore,
this requires more innovative studies with assistance of multidisciplinary knowledge in the
future.

An ideal in vitro system for investigation of shaping microbiota

The gut microbiota, considered as our “forgotten organ”, plays a major role in human health
(Clemente et al., 2012). As the definition of “healthy microbiota” becomes increasingly clear
in the future, it is possible to shape the gut microbiota through oral delivery of probiotics in the
aim of improving human health (Gareau et al., 2010).

Due to the ethics issues, it is difficult to obtain the in vivo information on the interaction of
oral probiotic supplementation and gut microbiota (not fecal microbiota). An in vitro digestion
system that can reflect the real human GIT environment is expected (Figure 44).

Figure 44. Schematic diagram of expected development in vitro digestion system for study of
probiotics-microbiota interaction.
 Future perspective 145

X axis in Figure 44 shows the expected technological development of an in vitro digestion

system: First, simple static system refers to conventional beaker/tube-based system with the
use of simulated digestive juice, which is most commonly used in vitro system by far (Minekus
et al., 2014). Second, dynamic factor refers to the flow-rate and time-dependent changes of
digestive juice, which has been reported in increasing literatures in the recent decade (Minekus
et al., 2014). Third, spatial factor includes the morphological, mechanical and physico-
chemical properties of different sections of GIT. Yet simulating these complexity of real human
GIT remains a crucial challenge (Guerra et al., 2012). The final bionic system refers to the near
real digestive tracts that can well reflect lumen-side environment, mucosal penetration, GIT
peristaltic movements, hormonal and nervous control, feedback mechanisms, and involvement
of the local immune system (Guerra et al., 2012). It requires more interdisciplinary and
multidisciplinary endeavours in the future.

Y axis in Figure 44 is the expected increasing understanding of biologically influential

factors during probiotic delivery: First, delivery loss means the loss of bacteria due to the
physico-chemical effect of digestive juice, which nowadays mostly refers to the effects of
gastric juice (acid stress) and bile-salt (Cook et al., 2012). Second, adhesion factor mainly
includes the effects of GIT movement and dynamic flow of digestive juice on the bacteria
adhesion to GIT, which has been investigated in vitro in a few studies in the recent decade (Van
Tassell and Miller, 2011). However, it should be noted that most of these studies are conducted
in a simple static in vitro digestion model, which may not reflect the real dynamic and
mechanical effects of digestive flow, and the interaction between probiotics and mucosal layer.
Third, competition factor refers to the effects of existing gut microbiota on the adhesion and
colonization of the delivering bacteria at a targeted niche. Similarly with the second “adhesion
factor”, the investigations in this aspect is largely limited by the authenticity of in vitro model
(Payne et al., 2012). Besides, the unculturability of most gut bacteria in lab also himders a more
insightful understanding of the competition factor within the GIT (Browne et al., 2016). The
final homeostasis means colonization and proliferation of delivering bacteria result in a desired
homeostatic ecosystem. The developments of “omics” analytical approach, sequencing, and in
situ imaging technologies are expected to promote the breakthoughs in this axis in the future.
 Reference 146

Reference

Abe, F., Miyauchi, H., Uchijima, A., Yaeshima, T., Iwatsuki, K., 2009. Stability of bifidobacteria in
powdered formula. Int. J. Food Sci. Technol. 44, 718–724.
Aburjaile, F. f., Madec, M.-N., Parayre, S., Miyoshi, A., Azevedo, V., Le Loir, Y., Falentin, H., 2016.
The long-term survival of Propionibacterium freudenreichii in a context of nutrient shortage.
J. Appl. Microbiol. 120, 432–440. doi:10.1111/jam.13000
Ahrné, Nobaek, Jeppsson, Adlerberth, Wold, Molin, 1998. The normal Lactobacillus flora of healthy
human rectal and oral mucosa. J. Appl. Microbiol. 85, 88–94. doi:10.1046/j.1365-
2672.1998.00480.x
Aktas, B., Wolfe, T.J.D., Safdar, N., Darien, B.J., Steele, J.L., 2016. The impact of Lactobacillus casei
on the composition of the cecal microbiota and innate immune system is strain specific. PLOS
ONE 11, e0156374. doi:10.1371/[Link].0156374
Alcántara, C., Blasco, A., Zúñiga, M., Monedero, V., 2014. Accumulation of polyphosphate in
Lactobacillus spp. and its involvement in stress resistance. Appl. Environ. Microbiol. 80, 1650–
1659. doi:10.1128/AEM.03997-13
Alcantara, C., Revilla-Guarinos, A., Zuniga, M., 2011. Influence of two-component signal transduction
systems of Lactobacillus casei BL23 on tolerance to stress conditions. Appl. Environ.
Microbiol. 77, 1516–1519. doi:10.1128/AEM.02176-10
Alting, A.C., Hamer, R.J., de Kruif, C.G., Visschers, R.W., 2000. Formation of disulfide bonds in acid-
induced gels of preheated whey protein isolate. J. Agric. Food Chem. 48, 5001–5007.
doi:10.1021/jf000474h
Anandharamakrishnan, C., S, P.I., 2015. Spray Drying Techniques for food ingredient encapsulation.
John Wiley & Sons.
Ananta, E., Knorr, D., 2003. Pressure-induced thermotolerance of Lactobacillus rhamnosus GG. Food
Res. Int. 36, 991–997. doi:10.1016/[Link].2003.07.001
Ananta, E., Volkert, M., Knorr, D., 2005. Cellular injuries and storage stability of spray-dried
Lactobacillus rhamnosus GG. Int. Dairy J. 15, 399–409. doi:10.1016/[Link].2004.08.004
Anastasiou, R., Leverrier, P., Krestas, I., Rouault, A., Kalantzopoulos, G., Boyaval, P., Tsakalidou, E.,
Jan, G., 2006. Changes in protein synthesis during thermal adaptation of Propionibacterium
freudenreichii subsp. shermanii. Int. J. Food Microbiol. doi:10.1016/[Link].2005.11.015
Anema, S.G., Li, Y., 2003. Association of denatured whey proteins with casein micelles in heated
reconstituted skim milk and its effect on casein micelle size. J. Dairy Res. 70, 73–83.
doi:10.1017/S0022029902005903
Anselmo, A.C., McHugh, K.J., Webster, J., Langer, R., Jaklenec, A., 2016. Layer-by-Layer
encapsulation of probiotics for delivery to the microbiome. Adv. Mater. 28, 9486–9490.
doi:10.1002/adma.201603270
Aoyagi, Y., Park, S., Matsubara, S., Honda, Y., Amamoto, R., Kushiro, A., Miyazaki, K., Shephard, R.
j., 2016. Habitual intake of fermented milk products containing Lactobacillus casei strain
Shirota and a reduced risk of hypertension in older people. Benef. Microbes 1–8.
doi:10.3920/BM2016.0135
Arai, S., Doi, M., 2012. Skin formation and bubble growth during drying process of polymer solution.
Eur. Phys. J. E 35. doi:10.1140/epje/i2012-12057-2
 Future perspective 147

Arcamone, J., Dujardin, E., Rius, G., Pérez-Murano, F., Ondarçuhu, T., 2007. Evaporation of femtoliter
sessile droplets monitored with nanomechanical mass sensors. J. Phys. Chem. B 111, 13020–
13027. doi:10.1021/jp075714b
Arifin, M., Swedlund, P.J., Hemar, Y., McKinnon, I.R., 2014. Calcium phosphates in Ca 2+ -fortified
milk: Phase identification and quantification by Raman spectroscopy. J. Agric. Food Chem. 62,
12223–12228. doi:10.1021/jf503602n
Arslan, S., Erbas, M., Tontul, I., Topuz, A., 2015. Microencapsulation of probiotic Saccharomyces
cerevisiae var. boulardii with different wall materials by spray drying. LWT - Food Sci.
Technol. 63, 685–690. doi:10.1016/[Link].2015.03.034
Auling, G., Pilz, F., Busse, H.J., Karrasch, S., Streichan, M., Schön, G., 1991. Analysis of the
polyphosphate-accumulating microflora in phosphorus-eliminating, anaerobic-aerobic
activated sludge systems by using diaminopropane as a biomarker for rapid estimation of
Acinetobacter spp. Appl. Environ. Microbiol. 57, 3585–3592.
Baker-Austin, C., Dopson, M., 2007. Life in acid: pH homeostasis in acidophiles. Trends Microbiol.
15, 165–171. doi:10.1016/[Link].2007.02.005
Barbosa, J., Borges, S., Amorim, M., Pereira, M.J., Oliveira, A., Pintado, M.E., Teixeira, P., 2015a.
Comparison of spray drying, freeze drying and convective hot air drying for the production of
a probiotic orange powder. J. Funct. Foods 17, 340–351. doi:10.1016/[Link].2015.06.001
Barbosa, J., Borges, S., Teixeira, P., 2015b. Influence of sub-lethal stresses on the survival of lactic acid
bacteria after spray-drying in orange juice. Food Microbiol. 52, 77–83.
doi:10.1016/[Link].2015.06.010
Barbosa, J., Teixeira, P., 2016. Development of probiotic fruit juice powders by spray drying: a review.
Food Rev. Int. 0, null. doi:10.1080/87559129.2016.1175016
Barone, P., Rosellini, D., Lafayette, P., Bouton, J., Veronesi, F., Parrott, W., 2008. Bacterial citrate
synthase expression and soil aluminum tolerance in transgenic alfalfa. Plant Cell Rep. 27, 893–
901. doi:10.1007/s00299-008-0517-x
Belloque, J., Ramos, M., 1999. Application of NMR spectroscopy to milk and dairy products. Trends
Food Sci. Technol. 10, 313–320. doi:10.1016/S0924-2244(00)00012-1
Berrier, C., Garrigues, A., Richarme, G., Ghazi, A., 2000. Elongation factor Tu and DnaK are
transferred from the cytoplasm to the periplasm of Escherichia coli during osmotic downshock
presumably via the mechanosensitive channel mscL. J. Bacteriol. 182, 248–251.
Beuchat, L.R., 1981. Microbial stability as affected by water activity. Cereal Foods World 26, 345–349.
Bielecka, M., Majkowska, A., 2000. Effect of spray drying temperature of yoghurt on the survival of
starter cultures, moisture content and sensoric properties of yoghurt powder. Nahr.-Food 44,
257–260.
Bode, G., Mauch, F., Ditschuneit, H., Malfertheiner, P., 1993. Identification of structures containing
polyphosphate in Helicobacter pylori. Microbiology 139, 3029–3033. doi:10.1099/00221287-
139-12-3029
Bove, P., Capozzi, V., Garofalo, C., Rieu, A., Spano, G., Fiocco, D., 2012. Inactivation of the ftsH gene
of Lactobacillus plantarum WCFS1: Effects on growth, stress tolerance, cell surface properties
and biofilm formation. Microbiol. Res. 167, 187–193. doi:10.1016/[Link].2011.07.001
Boyaval, P., Deborde, C., Corre, C., Blanco, C., Bégué, É., 1999. Stress and osmoprotection in
propionibacteria. Le Lait 79, 59–69. doi:10.1051/lait:199914
Breeuwer, P., 2014. Adaptation of Pathogenic Microorganisms to Dry Conditions, in: Gurtler, J.B.,
Doyle, M.P., Kornacki, J.L. (Eds.), The microbiological safety of low water activity foods and
spices, Food microbiology and food safety. Springer New York, pp. 37–48. doi:10.1007/978-
1-4939-2062-4_3
 Future perspective 148

Brill, J., Hoffmann, T., Bleisteiner, M., Bremer, E., 2011. Osmotically controlled synthesis of the
compatible solute proline is critical for cellular defense of Bacillus subtilis against high
osmolarity. J. Bacteriol. 193, 5335–5346. doi:10.1128/JB.05490-11
Brink, M., Todorov, S. d., Martin, J. h., Senekal, M., Dicks, L. m. t., 2006. The effect of prebiotics on
production of antimicrobial compounds, resistance to growth at low pH and in the presence of
bile, and adhesion of probiotic cells to intestinal mucus. J. Appl. Microbiol. 100, 813–820.
doi:10.1111/j.1365-2672.2006.02859.x
Britten, M., Giroux, H.J., 2001. Acid-induced gelation of whey protein polymers: effects of pH and
calcium concentration during polymerization. Food Hydrocoll., 5th International Hydrocolloids
Conference 15, 609–617. doi:10.1016/S0268-005X(01)00049-2
Broeckx, G., Vandenheuvel, D., Claes, I.J.J., Lebeer, S., Kiekens, F., 2016. Drying techniques of
probiotic bacteria as an important step towards the development of novel pharmabiotics. Int. J.
Pharm. 505, 303–318. doi:10.1016/[Link].2016.04.002
Bron, P.A., van Baarlen, P., Kleerebezem, M., 2011. Emerging molecular insights into the interaction
between probiotics and the host intestinal mucosa. Nat. Rev. Microbiol.
doi:10.1038/nrmicro2690
Browne, H.P., Forster, S.C., Anonye, B.O., Kumar, N., Neville, B.A., Stares, M.D., Goulding, D., and
Lawley, T.D., 2016. Culturing of “unculturable” human microbiota reveals novel taxa and
extensive sporulation. Nature 533, 543–546.
Buitink, J., van den Dries, I.J., Hoekstra, F.A., Alberda, M., Hemminga, M.A., 2000. High critical
temperature above Tg may contribute to the stability of biological systems. Biophys. J. 79,
1119–1128. doi:10.1016/S0006-3495(00)76365-X
Burns, P., Alard, J., Hrdỳ, J., Boutillier, D., Páez, R., Reinheimer, J., Pot, B., Vinderola, G., Grangette,
C., 2017. Spray-drying process preserves the protective capacity of a breast milk-derived
Bifidobacterium lactis strain on acute and chronic colitis in mice. Sci. Rep. 7, 43211.
doi:10.1038/srep43211
Bustamante, M., Oomah, B.D., Rubilar, M., Shene, C., 2017. Effective Lactobacillus plantarum and
Bifidobacterium infantis encapsulation with chia seed (Salvia hispanica L.) and flaxseed
(Linum usitatissimum L.) mucilage and soluble protein by spray drying. Food Chem. 216, 97–
105. doi:10.1016/[Link].2016.08.019
Bustamante, M., Villarroel, M., Rubilar, M., Shene, C., 2015. Lactobacillus acidophilus La-05
encapsulated by spray drying: Effect of mucilage and protein from flaxseed (Linum
usitatissimum L.). LWT - Food Sci. Technol. 62, 1162–1168. doi:10.1016/[Link].2015.02.017
Ca, P., Ao, B., Er, P., Cf, B., P, R., 1999. The effect of supplementation with milk fermented by
Lactobacillus casei (strain DN-114 001) on acute diarrhoea in children attending day care
centres. Int. J. Clin. Pract. 53, 179–184.
Cal, K., Sollohub, K., 2010. Spray drying technique. I: Hardware and process parameters. J. Pharm.
Sci. 99, 575–586. doi:10.1002/jps.21886
Caldas, T.D., Yaagoubi, A.E., Richarme, G., 1998. Chaperone properties of bacterial elongation factor
EF-Tu. J. Biol. Chem. 273, 11478–11482. doi:10.1074/jbc.273.19.11478
Cani, P.D., Van Hul, M., 2015. Novel opportunities for next-generation probiotics targeting metabolic
syndrome. Curr. Opin. Biotechnol., Food Biotechnology • Plant Biotechnology 32, 21–27.
doi:10.1016/[Link].2014.10.006
Cardoso, F.S., Castro, R.F., Borges, N., Santos, H., 2007. Biochemical and genetic characterization of
the pathways for trehalose metabolism in Propionibacterium freudenreichii, and their role in
stress response. Microbiology 153, 270–280. doi:10.1099/mic.0.29262-0
Chandan, R.C., Kilara, A., Shah, N.P., 2015. Dairy processing and quality assurance. John Wiley &
Sons.
 Future perspective 149

Chen, L., Remondetto, G.E., Subirade, M., 2006. Food protein-based materials as nutraceutical delivery
systems. Trends Food Sci. Technol. 17, 272–283. doi:10.1016/[Link].2005.12.011
Chen, L., Xu, Y., Fan, T., Liao, Z., Wu, P., Wu, X., Chen, X.D., 2016. Gastric emptying and
morphology of a “near real” in vitro human stomach model (RD-IV-HSM). J. Food Eng. 183,
1–8. doi:10.1016/[Link].2016.02.025
Chen, X.D., Patel, K.C., 2007. Micro-organism inactivation during drying of small droplets or thin-
layer slabs - A critical review of existing kinetics models and an appraisal of the drying rate
dependent model. J. Food Eng. 82, 1–10. doi:10.1016/[Link].2006.12.013
Clark, J.E., Beegen, H., Wood, H.G., 1986. Isolation of intact chains of polyphosphate from
“Propionibacterium shermanii” grown on glucose or lactate. J. Bacteriol. 168, 1212–1219.
doi:10.1128/jb.168.3.1212-1219.1986
Clemente, J.C., Ursell, L.K., Parfrey, L.W., Knight, R., 2012. The impact of the gut microbiota on
human health: An integrative view. Cell. 148, 1258-1270. doi: 10.1016/[Link].2012.01.035
Clementi, F., Rossi, J., 1984. Effect of drying and storage-conditions on survival of Leuconostoc oenos.
[Link]. 35, 183–186.
CNIEL. L'économie laitière en chiffres. From 2011 to 2016 editions. Paris.
Coldewey, S.M., Hartmann, M., Schmidt, D.S., Engelking, U., Ukena, S.N., Gunzer, F., 2007. Impact
of the rpoS genotype for acid resistance patterns of pathogenic and probiotic Escherichia coli.
BMC Microbiol. 7, 21. doi:10.1186/1471-2180-7-21
Conrad, P.B., Miller, D.P., Cielenski, P.R., de Pablo, J.J., 2000. Stabilization and preservation of
Lactobacillus acidophilus in saccharide matrices. Cryobiology 41, 17–24.
doi:10.1006/cryo.2000.2260
Cook, M.T., Tzortzis, G., Charalampopoulos, D., Khutoryanskiy, V.V., 2012. Microencapsulation of
probiotics for gastrointestinal delivery. J. Controlled Release 162, 56–67.
doi:10.1016/[Link].2012.06.003
Cook, M.T., Tzortzis, G., Charalampopoulos, D., Khutoryanskiy, V.V., 2011. Production and
evaluation of dry alginate-chitosan microcapsules as an enteric delivery vehicle for probiotic
bacteria. Biomacromolecules 12, 2834–2840. doi:10.1021/bm200576h
Corcoran, B.M., Ross, R.P., Fitzgerald, G.F., Dockery, P., Stanton, C., 2006. Enhanced survival of
GroESL-overproducing Lactobacillus paracasei NFBC 338 under stressful conditions induced
by drying. Appl. Environ. Microbiol. 72, 5104–5107. doi:10.1128/AEM.02626-05
Corcoran, B.M., Ross, R.P., Fitzgerald, G.F., Stanton, C., 2004. Comparative survival of probiotic
lactobacilli spray-dried in the presence of prebiotic substances. J. Appl. Microbiol. 96, 1024–
1039.
Cordonnier, C., Thévenot, J., Etienne-Mesmin, L., Denis, S., Alric, M., Livrelli, V., Blanquet-Diot, S.,
2015. Dynamic in vitro models of the human gastrointestinal tract as relevant tools to assess
the survival of probiotic strains and their interactions with gut microbiota. Microorganisms 3,
725–745. doi:10.3390/microorganisms3040725
Cousin, F.J., Jouan-Lanhouet, S., Dimanche-Boitrel, M.-T., Corcos, L., Jan, G., 2012a. Milk fermented
by Propionibacterium freudenreichii induces apoptosis of HGT-1 human gastric cancer cells.
PLoS ONE 7, e31892. doi:10.1371/[Link].0031892
Cousin, F.J., Jouan-Lanhouet, S., Théret, N., Brenner, C., Jouan, E., Moigne-Muller, G.L., Dimanche-
Boitrel, M.-T., Jan, G., Cousin, F.J., Jouan-Lanhouet, S., Théret, N., Brenner, C., Jouan, E., Le
Moigne-Muller, G., Dimanche-Boitrel, M.-T., Jan, G., 2016. The probiotic Propionibacterium
freudenreichii as a new adjuvant for TRAIL-based therapy in colorectal cancer. Oncotarget 7,
7161–7178. doi:10.18632/oncotarget.6881
 Future perspective 150

Cousin, F.J., Louesdon, S., Maillard, M.-B., Parayre, S., Falentin, H., Deutsch, S.-M., Boudry, G., Jan,
G., 2012b. The first dairy product exclusively fermented by Propionibacterium freudenreichii:
A new vector to study probiotic potentialities in vivo. Food Microbiol. 32, 135–146.
doi:10.1016/[Link].2012.05.003
Cousin, F.J., Mater, D.D.G., Foligné, B., Jan, G., 2010. Dairy propionibacteria as human probiotics: A
review of recent evidence. Dairy Sci. Technol. 91, 1–26. doi:10.1051/dst/2010032
Crowe, J.H., Carpenter, J.P., Crowe, L.M., 1998. The role of vitrification in anhydrobiosis. Annu. Rev.
Physiol. 60, 73–103.
D. Bouglé, P.A., N. Roland, F. Lebeurrier, 1999. Effect of propionibacteria supplementation on fecal
bifidobacteria and segmental colonic transit time in healthy human subjects. Scand. J.
Gastroenterol. 34, 144–148. doi:10.1080/00365529950172998
Dalgleish, D.G., 1990. Denaturation and aggregation of serum proteins and caseins in heated milk. J.
Agric. Food Chem. 38, 1995–1999. doi:10.1021/jf00101a001
Dalmasso, M., Aubert, J., Even, S., Falentin, H., Maillard, M.-B., Parayre, S., Loux, V., Tanskanen, J.,
Thierry, A., 2012. Accumulation of intracellular glycogen and trehalose by Propionibacterium
freudenreichii under conditions mimicking cheese ripening in the cold. Appl. Environ.
Microbiol. 78, 6357–6364. doi:10.1128/AEM.00561-12
De Angelis, M., Calasso, M., Cavallo, N., Di Cagno, R., Gobbetti, M., 2016. Functional proteomics
within the genus Lactobacillus. PROTEOMICS 16, 946–962. doi:10.1002/pmic.201500117
De Prisco, A., Mauriello, G., 2016. Probiotication of foods: A focus on microencapsulation tool. Trends
Food Sci. Technol. 48, 27–39. doi:10.1016/[Link].2015.11.009
Deshpande, G.C., Rao, S.C., Keil, A.D., Patole, S.K., 2011. Evidence-based guidelines for use of
probiotics in preterm neonates. BMC Med. 9, 92. doi:10.1186/1741-7015-9-92
Desmond, C., Fitzgerald, G.F., Stanton, C., Ross, R.P., 2004. Improved stress tolerance of GroESL-
overproducing Lactococcus lactis and probiotic Lactobacillus paracasei NFBC 338. Appl.
Environ. Microbiol. 70, 5929–5936. doi:10.1128/AEM.70.10.5929-5936.2004
Desmond, C., Ross, R.P., O’Callaghan, E., Fitzgerald, G., Stanton, C., 2002. Improved survival of
Lactobacillus paracasei NFBC 338 in spray-dried powders containing gum acacia. J. Appl.
Microbiol. 93, 1003–1011.
Desmond, C., Stanton, C., Fitzgerald, G.F., Collins, K., Ross, R.P., 2001. Environmental adaptation of
probiotic lactobacilli towards improvement of performance during spray drying. Int. Dairy J.
11, 801–808.
Dianawati, D., Mishra, V., Shah, N.P., 2015. Survival of microencapsulated probiotic bacteria after
processing and during storage: A Review. Crit. Rev. Food Sci. Nutr. 0, 00–00.
doi:10.1080/10408398.2013.798779
Dijkstra, A.R., Setyawati, M.C., Bayjanov, J.R., Alkema, W., van Hijum, S.A.F.T., Bron, P.A.,
Hugenholtz, J., 2014. Diversity in robustness of Lactococcus lactis strains during heat stress,
oxidative stress, and spray drying stress. Appl. Environ. Microbiol. 80, 603–611.
doi:10.1128/AEM.03434-13
Dimitrellou, D., Kandylis, P., Petrović, T., Dimitrijević-Branković, S., Lević, S., Nedović, V.,
Kourkoutas, Y., 2016. Survival of spray dried microencapsulated Lactobacillus casei ATCC
393 in simulated gastrointestinal conditions and fermented milk. LWT - Food Sci. Technol. 71,
169–174. doi:10.1016/[Link].2016.03.007
Doherty, S.B., Gee, V.L., Ross, R.P., Stanton, C., Fitzgerald, G.F., Brodkorb, A., 2011. Development
and characterisation of whey protein micro-beads as potential matrices for probiotic protection.
Food Hydrocoll. 25, 1604–1617. doi:10.1016/[Link].2010.12.012
 Future perspective 151

Du, Y., Shi, W.-W., He, Y.-X., Yang, Y.-H., Zhou, C.-Z., Chen, Y., 2011. Structures of the substrate-
binding protein provide insights into the multiple compatible solute binding specificities of the
Bacillus subtilis ABC transporter OpuC. Biochem. J. 436, 283–289. doi:10.1042/BJ20102097
Elversson, J., Millqvist-Fureby, A., Alderborn, G., Elofsson, U., 2003. Droplet and particle size
relationship and shell thickness of inhalable lactose particles during spray drying. J. Pharm. Sci.
92, 900–910. doi:10.1002/jps.10352
Empadinhas, N., da Costa, M.S., 2008. Osmoadaptation mechanisms in prokaryotes: distribution of
compatible solutes. Int. Microbiol. Off. J. Span. Soc. Microbiol. 11, 151–161.
Eratte, D., McKnight, S., Gengenbach, T.R., Dowling, K., Barrow, C.J., Adhikari, B.P., 2015. Co-
encapsulation and characterisation of omega-3 fatty acids and probiotic bacteria in whey protein
isolate–gum Arabic complex coacervates. J. Funct. Foods, Omega-3 Lipids 19, Part B, 882–
892. doi:10.1016/[Link].2015.01.037
Espina, F., Packard, V.S., 1979. Survival of Lactobacillus acidophilus in a spray-drying process. J.
Food Prot. 42, 149–152.
Falentin, H., Deutsch, S.-M., Jan, G., Loux, V., Thierry, A., Parayre, S., Maillard, M.-B., Dherbécourt,
J., Cousin, F.J., Jardin, J., Siguier, P., Couloux, A., Barbe, V., Vacherie, B., Wincker, P., Gibrat,
J.-F., Gaillardin, C., Lortal, S., 2010a. The complete genome of Propionibacterium
freudenreichii CIRM-BIA1T, a hardy actinobacterium with food and probiotic applications.
PLoS ONE 5, e11748. doi:10.1371/[Link].0011748
Falentin, H., Postollec, F., Parayre, S., Henaff, N., Le Bivic, P., Richoux, R., Thierry, A., Sohier, D.,
2010b. Specific metabolic activity of ripening bacteria quantified by real-time reverse
transcription PCR throughout Emmental cheese manufacture. Int. J. Food Microbiol. 144, 10–
19. doi:10.1016/[Link].2010.06.003
FAO/WHO, 2001. Health and nutritional properties of probiotics in food including powder milk with
live lactic acid bacteria.
Felis, G.E., Dellaglio, F., Torriani, S., 2009. Taxonomy of probiotic microorganisms, in:
Charalampopoulos, D., Rastall, R.A. (Eds.), Prebiotics and probiotics science and technology.
Springer New York, pp. 591–637. doi:10.1007/978-0-387-79058-9_15
Ferrando, V., Quiberoni, A., Reinhemer, J., Suárez, V., 2015. Resistance of functional Lactobacillus
plantarum strains against food stress conditions. Food Microbiol. 48, 63–71.
doi:10.1016/[Link].2014.12.005
Ferreira, V., Soares, V., Santos, C., Silva, J., Gibbs, P.A., Teixeira, P., 2005. Survival of Lactobacillus
sakei during heating, drying and storage in the dried state when growth has occurred in the
presence of sucrose or monosodium glutamate. Biotechnol. Lett. 27, 249–252.
doi:10.1007/s10529-004-8351-x
Flahaut, S., Benachour, A., Giard, J.-C., Boutibonnes, P., Auffray, Y., 1996. Defense against lethal
treatments and de novo protein synthesis induced by NaCl in Enterococcus faecalis ATCC
19433. Arch. Microbiol. 165, 317–324. doi:10.1007/s002030050333
Foerst, P., Kulozik, U., 2011. Modelling the dynamic inactivation of the probiotic bacterium L.
paracasei ssp. paracasei during a low-temperature drying process based on stationary data in
concentrated systems. Food Bioprocess Technol. 5, 2419–2427. doi:10.1007/s11947-011-
0560-4
Foerst, P., Kulozik, U., Schmitt, M., Bauer, S., Santivarangkna, C., 2012. Storage stability of vacuum-
dried probiotic bacterium Lactobacillus paracasei F19. Food Bioprod. Process. 90, 295–300.
doi:10.1016/[Link].2011.06.004
Foligné, B., Breton, J., Mater, D., Jan, G., 2013a. Tracking the microbiome functionality: focus on
Propionibacterium species. Gut gutjnl-2012-304393. doi:10.1136/gutjnl-2012-304393
 Future perspective 152

Foligné, B., Daniel, C., Pot, B., 2013b. Probiotics from research to market: the possibilities, risks and
challenges. Curr. Opin. Microbiol., Ecology and industrial microbiology • Special Section:
Innate immunity 16, 284–292. doi:10.1016/[Link].2013.06.008
Foligné, B., Deutsch, S.-M., Breton, J., Cousin, F.J., Dewulf, J., Samson, M., Pot, B., Jan, G., 2010.
Promising immunomodulatory effects of selected strains of dairy propionibacteria as evidenced
in vitro and in vivo. Appl. Environ. Microbiol. 76, 8259–8264. doi:10.1128/AEM.01976-10
Frees, D., Ingmer, H., 1999. ClpP participates in the degradation of misfolded protein in Lactococcus
lactis. Mol. Microbiol. 31, 79–87. doi:10.1046/j.1365-2958.1999.01149.x
Fu, N., Wai Woo, M., Qi Lin, S.X., Zhou, Z., Dong Chen, X., 2011. Reaction engineering approach
(REA) to model the drying kinetics of droplets with different initial sizes—experiments and
analyses. Chem. Eng. Sci. 66, 1738–1747. doi:10.1016/[Link].2011.01.009
Fu, N., Woo, M.W., Selomulya, C., Chen, X.D., 2013a. Inactivation of Lactococcus lactis ssp. cremoris
cells in a droplet during convective drying. Biochem. Eng. J. 79, 46–56.
doi:10.1016/[Link].2013.06.015
Fu, N., Woo, M.W., Selomulya, C., Chen, X.D., 2013b. Shrinkage behaviour of skim milk droplets
during air drying. J. Food Eng. 116, 37–44. doi:10.1016/[Link].2012.11.005
Fu, W.-Y., Suen, S.-Y., Etzel, M.R., 1995. Inactivation of Lactococcus lactis ssp. lactis C2 and alkaline
phosphatase during spray drying. Dry. Technol. 13, 1463–1476.
doi:10.1080/07373939508917033
Gagnaire, V., Jardin, J., Rabah, H., Briard-Bion, V., Jan, G., 2015. Emmental cheese environment
enhances Propionibacterium freudenreichii stress tolerance. PLOS ONE 10, e0135780.
doi:10.1371/[Link].0135780
Gardiner, G.E., O’Sullivan, E., Kelly, J., Auty, M.A., Fitzgerald, G.F., Collins, J.K., Ross, R.P.,
Stanton, C., 2000. Comparative survival rates of human-derived probiotic Lactobacillus
paracasei and L. salivarius strains during heat treatment and spray drying. Appl. Environ.
Microbiol. 66, 2605–2612.
Gardiner, G.E., O’Sullivan, E., Kelly, J., Auty, M.A.E., Fitzgerald, G.F., Collins, J.K., Ross, R.P.,
Stanton, C., 2000. Comparative survival rates of human-derived probiotic Lactobacillus
paracasei and L. salivarius strains during heat treatment and spray drying. Appl. Environ.
Microbiol. 66, 2605–2612.
Gareau, M.G., Sherman, P.M., Walker, M.A., 2010. Probiotics and the gut microbiota in intestinal
health and disease. Nat. Rev. Gastroenterol. Hepatol. 7, 503-514.
Gaucher, I., Boubellouta, T., Beaucher, E., Piot, M., Gaucheron, F., Dufour, E., 2008. Investigation of
the effects of season, milking region, sterilisation process and storage conditions on milk and
UHT milk physico-chemical characteristics: a multidimensional statistical approach. Dairy Sci.
Technol. 88, 291–312. doi:10.1051/dst:2007022
Gaucheron, F., 2005. The minerals of milk. Reprod. Nutr. Dev. 45, 473–483. doi:10.1051/rnd:2005030
Ghandi, A., Powell, I.B., Howes, T., Chen, X.D., Adhikari, B., 2012. Effect of shear rate and oxygen
stresses on the survival of Lactococcus lactis during the atomization and drying stages of spray
drying: A laboratory and pilot scale study. J. Food Eng. 113, 194–200.
doi:10.1016/[Link].2012.06.005
Global Market Insights, 2016. Probiotics market size by end use (human, animal), by application
(functional foods & beverages [dairy, non-dairy, cereals, baked goods, fermented meat
products, dry foods], dietary supplements [food, nutritional, specialty, infant formula], animal
feed probiotics), industry analysis report, regional outlook (U.S., Germany, UK, China, Japan,
India, Brazil), application potential, price trends, competitive market share & forecast, 2016 -
2023. URL [Link] (accessed
9.15.16).
 Future perspective 153

Golowczyc, M.A., Silva, J., Teixeira, P., De Antoni, G.L., Abraham, A.G., 2011. Cellular injuries of
spray-dried Lactobacillus spp. isolated from kefir and their impact on probiotic properties. Int.
J. Food Microbiol. 144, 556–560. doi:10.1016/[Link].2010.11.005
Gray, M.J., Jakob, U., 2015. Oxidative stress protection by polyphosphate — new roles for an old
player. Curr. Opin. Microbiol., Cell regulation 24, 1–6. doi:10.1016/[Link].2014.12.004
Guarner, F., Malagelada, J.-R., 2003. Gut flora in health and disease. The Lancet 361, 512–519.
doi:10.1016/S0140-6736(03)12489-0
Guchte, M. van de, Serror, P., Chervaux, C., Smokvina, T., Ehrlich, S.D., Maguin, E., 2002. Stress
responses in lactic acid bacteria. Antonie Van Leeuwenhoek 82, 187–216.
doi:10.1023/A:1020631532202
Guerin, J., Petit, J., Burgain, J., Borges, F., Bhandari, B., Perroud, C., Desobry, S., Scher, J., Gaiani, C.,
2017. Lactobacillus rhamnosus GG encapsulation by spray-drying: Milk proteins clotting
control to produce innovative matrices. J. Food Eng. 193, 10–19.
doi:10.1016/[Link].2016.08.008
Guerra, A., Etienne-Mesmin, L., Livrelli, V., Denis, S., Blanquet-Diot, S., and Alric, M., 2012.
Relevance and challenges in modeling human gastric and small intestinal digestion. Trends
Biotechnol. 30, 591–600.
Günther, S., Trutnau, M., Kleinsteuber, S., Hause, G., Bley, T., Röske, I., Harms, H., Müller, S., 2009.
Dynamics of polyphosphate-accumulating bacteria in wastewater treatment plant microbial
communities detected via DAPI (4′,6′-Diamidino-2-Phenylindole) and tetracycline labeling.
Appl. Environ. Microbiol. 75, 2111–2121. doi:10.1128/AEM.01540-08
Hansen, N.-H., Riemann, H., 1963. Factors affecting the heat resistance of nonsporing organisms. J.
Appl. Bacteriol. 26, 314–333. doi:10.1111/j.1365-2672.1963.tb04783.x
Havea, P., Singh, H., Creamer, L.K., 2002. Heat-induced aggregation of whey proteins: Comparison
of cheese WPC with acid WPC and relevance of mineral composition. J. Agric. Food Chem.
50, 4674–4681. doi:10.1021/jf011583e
Heurlier, K., Dénervaud, V., Haenni, M., Guy, L., Krishnapillai, V., Haas, D., 2005. Quorum-sensing-
negative (lasR) mutants of Pseudomonas aeruginosa avoid cell lysis and death. J. Bacteriol.
187, 4875–4883. doi:10.1128/JB.187.14.4875-4883.2005
Hill, C., Guarner, F., Reid, G., Gibson, G.R., Merenstein, D.J., Pot, B., Morelli, L., Canani, R.B., Flint,
H.J., Salminen, S., Calder, P.C., Sanders, M.E., 2014. Expert consensus document: The
International Scientific Association for Probiotics and Prebiotics consensus statement on the
scope and appropriate use of the term probiotic. Nat. Rev. Gastroenterol. Hepatol. 11, 506–514.
doi:10.1038/nrgastro.2014.66
Holzapfel, W.H., Haberer, P., Geisen, R., Björkroth, J., Schillinger, U., 2001. Taxonomy and important
features of probiotic microorganisms in food and nutrition. Am. J. Clin. Nutr. 73, 365s–373s.
Huang, S., Cauty, C., Dolivet, A., Le Loir, Y., Chen, X.D., Schuck, P., Jan, G., Jeantet, R., 2016. Double
use of highly concentrated sweet whey to improve the biomass production and viability of
spray-dried probiotic bacteria. J. Funct. Foods 23, 453–463. doi:10.1016/[Link].2016.02.050
Huang, S., Chen, X.D., 2013. Significant effect of Ca2+ on improving the heat resistance of lactic acid
bacteria. FEMS Microbiol. Lett. 344, 31–38. doi:10.1111/1574-6968.12151
Huang, S., Vignolles, M.-L., Chen, X.D., Le Loir, Y., Jan, G., Schuck, P., Jeantet, R., 2017. Spray
drying of probiotics and other food-grade bacteria: A review. Trends Food Sci. Technol.
doi:10.1016/[Link].2017.02.007
Huang, S., Yang, Y., Fu, N., Qin, Q., Zhang, L., Chen, X.D., 2014. Calcium-aggregated milk: A
potential new option for improving the viability of lactic acid bacteria under heat stress. Food
Bioprocess Technol. 7, 3147–3155. doi:10.1007/s11947-014-1331-9
 Future perspective 154

Hussain, M. a., Knight, M. i., Britz, M. l., 2009. Proteomic analysis of lactose-starved Lactobacillus
casei during stationary growth phase. J. Appl. Microbiol. 106, 764–773. doi:10.1111/j.1365-
2672.2008.03961.x
Hwang, C.-F., Chang, J.-H., Houng, J.-Y., Tsai, C.-C., Lin, C.-K., Tsen, H.-Y., 2012. Optimization of
medium composition for improving biomass production of Lactobacillus plantarum Pi06 using
the Taguchi array design and the Box-Behnken method. Biotechnol. Bioprocess Eng. 17, 827–
834. doi:10.1007/s12257-012-0007-4
Iaconelli, C., Lemetais, G., Kechaou, N., Chain, F., Bermúdez-Humarán, L.G., Langella, P., Gervais,
P., Beney, L., 2015. Drying process strongly affects probiotics viability and functionalities. J.
Biotechnol. 214, 17–26. doi:10.1016/[Link].2015.08.022
Jan, G., Belzacq, A.-S., Haouzi, D., Rouault, A., Métivier, D., Kroemer, G., Brenner, C., 2002.
Propionibacteria induce apoptosis of colorectal carcinoma cells via short-chain fatty acids
acting on mitochondria. Cell Death Differ. 9, 179–188. doi:10.1038/[Link].4400935
Jan, G., Leverrier, P., Pichereau, V., Boyaval, P., 2001. Changes in protein synthesis and morphology
during acid adaptation of Propionibacterium freudenreichii. Appl. Environ. Microbiol. 67,
2029–2036. doi:10.1128/AEM.67.5.2029-2036.2001
Jan, G., Rouault, A., Maubois, J.-L., 2000. Acid stress susceptibility and acid adaptation of
Propionibacterium freudenreichii subsp. shermanii. Le Lait 80, 12. doi:10.1051/lait:2000128
Jeantet, R., Ducept, F., Dolivet, A., Méjean, S., Schuck, P., 2008. Residence time distribution: a tool to
improve spray-drying control. Dairy Sci. Technol. 88, 31–43. doi:10.1051/dst:2007006
Jeurnink, T.J.M., De Kruif, K.G., 1993. Changes in milk on heating: viscosity measurements. J. Dairy
Res. 60, 139–150. doi:10.1017/S0022029900027461
Jewell, J.B., Kashket, E.R., 1991. Osmotically regulated transport of proline by Lactobacillus
acidophilus IFO 3532. Appl. Environ. Microbiol. 57, 2829–2833.
Johansson, M.E.V., Jakobsson, H.E., Holmén-Larsson, J., Schütte, A., Ermund, A., Rodríguez-Piñeiro,
A.M., Arike, L., Wising, C., Svensson, F., Bäckhed, F., Hansson, G.C., 2015. Normalization
of host intestinal mucus layers requires long-term microbial colonization. Cell Host Microbe
18, 582–592. doi:10.1016/[Link].2015.10.007
Kajiyama, T., Park, K.J., 2011. Influence of feed parameters on spray drying energy consumption. J.
Food Process Eng. 34, 414–426. doi:10.1111/j.1745-4530.2009.00365.x
Kang, O.J., Vézinz, L.-P., Laberge, S., Simard, R.E., 1998. Some factors influencing the autolysis of
Lactobacillus bulgaricus and Lactobacillus casei. J. Dairy Sci. 81, 639–646.
doi:10.3168/jds.S0022-0302(98)75618-8
Kato-Kataoka, A., Nishida, K., Takada, M., Kawai, M., Kikuchi-Hayakawa, H., Suda, K., Ishikawa,
H., Gondo, Y., Shimizu, K., Matsuki, T., Kushiro, A., Hoshi, R., Watanabe, O., Igarashi, T.,
Miyazaki, K., Kuwano, Y., Rokutan, K., 2016. Fermented milk containing Lactobacillus casei
strain Shirota preserves the diversity of the gut microbiota and relieves abdominal dysfunction
in healthy medical students exposed to academic stress. Appl. Environ. Microbiol.
doi:10.1128/AEM.04134-15
Kearney, N., Meng, X.C., Stanton, C., Kelly, J., Fitzgerald, G.F., Ross, R.P., 2009. Development of a
spray dried probiotic yoghurt containing Lactobacillus paracasei NFBC 338. Int. Dairy J. 19,
684–689. doi:10.1016/[Link].2009.05.003
Kets, E.P.W., Bont, J.A.M.D., 1997. Effect of carnitines on Lactobacillus plantarum subjected to
osmotic stress. FEMS Microbiol. Lett. 146, 205–209. doi:10.1111/j.1574-6968.1997.tb10194.x
Kets, E.P.W., Teunissen, P.J.M., Debont, J.A.M., 1996. Effect of compatible solutes on survival of
lactic acid bacteria subjected to drying. Appl. Environ. Microbiol. 62, 259–261.
 Future perspective 155

Khaskheli, G.B., Zuo, F., Yu, R., Chen, S., 2015. Overexpression of small heat shock protein enhances
heat- and salt-stress tolerance of Bifidobacterium longum NCC2705. Curr. Microbiol. 71, 8–
15. doi:10.1007/s00284-015-0811-0
Khem, S., Bansal, V., Small, D.M., May, B.K., 2016. Comparative influence of pH and heat on whey
protein isolate in protecting Lactobacillus plantarum A17 during spray drying. Food Hydrocoll.
54, Part A, 162–169. doi:10.1016/[Link].2015.09.029
Khem, S., Woo, M.W., Small, D.M., Chen, X.D., May, B.K., 2015. Agent selection and protective
effects during single droplet drying of bacteria. Food Chem. 166, 206–214.
doi:10.1016/[Link].2014.06.010
Kim, B.J., Park, T., Moon, H.C., Park, S.-Y., Hong, D., Ko, E.H., Kim, J.Y., Hong, J.W., Han, S.W.,
Kim, Y.-G., Choi, I.S., 2014. Cytoprotective alginate/polydopamine core/shell microcapsules
in microbial encapsulation. Angew. Chem. Int. Ed. 53, 14443–14446.
doi:10.1002/anie.201408454
Kim, H., Park, K., Oh, S., Chang, I.S., 2009. Rapid detection of lactobacillus and yeast concentrations
using a particle size distribution analyser. J. Appl. Microbiol. 107, 1499–1504.
doi:10.1111/j.1365-2672.2009.04331.x
Klauth, P., Pallerla, S.R., Vidaurre, D., Ralfs, C., Wendisch, V.F., Schoberth, S.M., 2006.
Determination of soluble and granular inorganic polyphosphate in Corynebacterium
glutamicum. Appl. Microbiol. Biotechnol. 72, 1099–1106. doi:10.1007/s00253-006-0562-8
Kumar, P., Mishra, H.N., 2004. Yoghurt powder-a review of process technology, storage and
utilization. Food Bioprod. Process. 82, 133–142. doi:10.1205/0960308041614918
Lambelet, P., Berrocal, R., Renevey, F., 1992. Low-field nuclear magnetic resonance relaxation study
of thermal effects on milk proteins. J. Dairy Res. 59, 517–526.
doi:10.1017/S0022029900027187
Lan, A., Bruneau, A., Bensaada, M., Philippe, C., Bellaud, P., Rabot, S., Jan, G., 2008. Increased
induction of apoptosis by Propionibacterium freudenreichii TL133 in colonic mucosal crypts
of human microbiota-associated rats treated with 1,2-dimethylhydrazine. Br. J. Nutr. 100,
1251–1259. doi:10.1017/S0007114508978284
Lan, A., Lagadic-Gossmann, D., Lemaire, C., Brenner, C., Jan, G., 2007. Acidic extracellular pH shifts
colorectal cancer cell death from apoptosis to necrosis upon exposure to propionate and acetate,
major end-products of the human probiotic propionibacteria. Apoptosis 12, 573–591.
doi:10.1007/s10495-006-0010-3
Laroche, C., Fine, F., Gervais, P., 2005. Water activity affects heat resistance of microorganisms in
food powders. Int. J. Food Microbiol. 97, 307–315.
Lavari, L., Ianniello, R., Páez, R., Zotta, T., Cuatrin, A., Reinheimer, J., Parente, E., Vinderola, G.,
2015. Growth of Lactobacillus rhamnosus 64 in whey permeate and study of the effect of mild
stresses on survival to spray drying. LWT - Food Sci. Technol. 63, 322–330.
doi:10.1016/[Link].2015.03.066
Lavari, L., Páez, R., Cuatrin, A., Reinheimer, J., Vinderola, G., 2014. Use of cheese whey for biomass
production and spray drying of probiotic lactobacilli. J. Dairy Res. 81, 267–274.
doi:10.1017/S0022029914000156
Le Maréchal, C., Peton, V., Plé, C., Vroland, C., Jardin, J., Briard-Bion, V., Durant, G., Chuat, V.,
Loux, V., Foligné, B., Deutsch, S.-M., Falentin, H., Jan, G., 2015. Surface proteins of
Propionibacterium freudenreichii are involved in its anti-inflammatory properties. J.
Proteomics 113, 447–461. doi:10.1016/[Link].2014.07.018
Lebeer, S., Vanderleyden, J., De Keersmaecker, S.C.J., 2010. Host interactions of probiotic bacterial
surface molecules: comparison with commensals and pathogens. Nat. Rev. Microbiol. 8, 171–
184. doi:10.1038/nrmicro2297
 Future perspective 156

Lebeer, S., Vanderleyden, J., De Keersmaecker, S.C.J., 2008. Genes and molecules of lactobacilli
supporting probiotic action. Microbiol. Mol. Biol. Rev. 72, 728–764.
doi:10.1128/MMBR.00017-08
Lee, B., Tachon, S., Eigenheer, R.A., Phinney, B.S., Marco, M.L., 2015a. Lactobacillus casei low-
temperature, dairy-associated proteome promotes persistence in the mammalian digestive tract.
J. Proteome Res. 14, 3136–3147. doi:10.1021/[Link].5b00387
Lee, B., Yin, X., Griffey, S.M., Marco, M.L., 2015b. Attenuation of colitis by Lactobacillus casei BL23
is dependent on the dairy delivery matrix. Appl. Environ. Microbiol. AEM.01360-15.
doi:10.1128/AEM.01360-15
Leverrier, P., Dimova, D., Pichereau, V., Auffray, Y., Boyaval, P., Jan, G., 2003. Susceptibility and
adaptive response to bile salts in Propionibacterium freudenreichii: Physiological and
proteomic analysis. Appl. Environ. Microbiol. 69, 3809–3818. doi:10.1128/AEM.69.7.3809-
3818.2003
Leverrier, P., Vissers, J.P.C., Rouault, A., Boyaval, P., Jan, G., 2004. Mass spectrometry proteomic
analysis of stress adaptation reveals both common and distinct response pathways in
Propionibacterium freudenreichii. Arch. Microbiol. 181, 215–230. doi:10.1007/s00203-003-
0646-0
Li, X., Lin, S.X.Q., Chen, X.D., Chen, L., Pearce, D., 2006. Inactivation kinetics of probiotic bacteria
during the drying of single milk droplets. Dry. Technol. 24, 695–701.
doi:10.1080/07373930600684890
Li, X., Ma, L.Z., Tao, Y., Kong, B.H., Li, P.J., 2012. Low field-NMR in measuring water mobility and
distribution in beef granules during drying process. Adv. Mater. Res. 550–553, 3406–3410.
doi:10.4028/[Link]/AMR.550-553.3406
Lian, W.C., Hsiao, H.C., Chou, C.C., 2002. Survival of bifidobacteria after spray-drying. Int. J. Food
Microbiol. 74, 79–86. doi:10.1016/S0168-1605(01)00733-4
Lin, S.X., Chen, X.D., 2002. Improving the glass-filament method for accurate measurement of drying
kinetics of liquid droplets. Chem. Eng. Res. Des. 80, 401–410.
doi:10.1205/026387602317446443
Linders, L.J.M., Meerdink, G., Riet, K. van ’t, 1996. Influence of temperature and drying rate on the
dehydration inactivation of Lactobacillus plantarum. Food Bioprod. Process. Trans. Inst.
Chem. Eng. Part C 74, 110–114.
Linders, L.J.M., Meerdink, G., vantRiet, K., 1997. Effect of growth parameters on the residual activity
of Lactobacillus plantarum after drying. J. Appl. Microbiol. 82, 683–688.
Liu, H., Gong, J., Chabot, D., Miller, S.S., Cui, S.W., Ma, J., Zhong, F., Wang, Q., 2015. Protection of
heat-sensitive probiotic bacteria during spray-drying by sodium caseinate stabilized fat
particles. Food Hydrocoll. 51, 459–467. doi:10.1016/[Link].2015.05.015
Loux, V., Mariadassou, M., Almeida, S., Chiapello, H., Hammani, A., Buratti, J., Gendrault, A., Barbe,
V., Aury, J.-M., Deutsch, S.-M., Parayre, S., Madec, M.-N., Chuat, V., Jan, G., Peterlongo, P.,
Azevedo, V., Le Loir, Y., Falentin, H., 2015. Mutations and genomic islands can explain the
strain dependency of sugar utilization in 21 strains of Propionibacterium freudenreichii. BMC
Genomics 16, 296. doi:10.1186/s12864-015-1467-7
Lyu, C., Hu, S., Huang, J., Luo, M., Lu, T., Mei, L., Yao, S., 2016. Contribution of the activated catalase
to oxidative stress resistance and γ-aminobutyric acid production in Lactobacillus brevis. Int.
J. Food Microbiol. 238, 302–310. doi:10.1016/[Link].2016.09.023
Maciel, G.M., Chaves, K.S., Grosso, C.R.F., Gigante, M.L., 2014. Microencapsulation of Lactobacillus
acidophilus La-5 by spray-drying using sweet whey and skim milk as encapsulating materials.
J. Dairy Sci. 97, 1991–1998. doi:10.3168/jds.2013-7463
 Future perspective 157

Mahbubani, K.T., Slater, N.K.H., Edwards, A.D., 2014. Protection of dried probiotic bacteria from bile
using bile adsorbent resins. New Biotechnol. 31, 69–72. doi:10.1016/[Link].2013.09.001
Malik, A.C., Reinbold, G.W., Vedamuthu, E.R., 1968. An evaluation of the taxonomy of
Propionibacterium. Can. J. Microbiol. 14, 1185–1191.
Malmo, C., Storia, A.L., Mauriello, G., 2013. Microencapsulation of Lactobacillus reuteri DSM 17938
cells coated in alginate beads with chitosan by spray drying to use as a probiotic cell in a
chocolate soufflé. Food Bioprocess Technol. 6, 795–805. doi:10.1007/s11947-011-0755-8
Markets and Markets, 2015. Probiotic ingredients market by function (regular, preventative, therapy),
application (food & beverage, dietary supplements, & animal feed), end use (human & animal
probiotics), ingredient (bacteria & yeast), and by region - global trends & forecast to 2020.
URL [Link]
[Link] (accessed 7.26.16).
Martínez-Cuesta, M.C., Peláez, C., Juárez, M., Requena, T., 1997. Autolysis of Lactococcus lactis ssp.
lactis and Lactobacillus casei ssp. casei. Cell lysis induced by a crude bacteriocin. Int. J. Food
Microbiol. 38, 125–131.
Mattila-Sandholm, T., Myllärinen, P., Crittenden, R., Mogensen, G., Fondén, R., Saarela, M., 2002.
Technological challenges for future probiotic foods. Int. Dairy J. 12, 173–182.
doi:10.1016/S0958-6946(01)00099-1
Mauriello, G., Aponte, M., Andolfi, R., Moschetti, G., Villani, F., 1999. Spray-drying of bacteriocin-
producing lactic acid bacteria. J. Food Prot. 62, 773–777.
Mäyrä-Mäkinen, A., Bigret, M., 1993. Industrial use and production of lactic acid bacteria. 65–95.
McGann, T.C.A., Donnelly, W.J., Kearney, R.D., Buchhemm, W., 1980. Composition and size
distribution of bovine casein micelles. Biochim. Biophys. Acta BBA - Gen. Subj. 630, 261–
270. doi:10.1016/0304-4165(80)90430-4
Meerdink, G., Riet, K. van ’t, 1995. Prediction of product quality during spray drying.
Mekmene, O., Le Graët, Y., Gaucheron, F., 2009. A model for predicting salt equilibria in milk and
mineral-enriched milks. Food Chem. 116, 233–239. doi:10.1016/[Link].2009.02.039
Mezzenga, R., Schurtenberger, P., Burbidge, A., Michel, M., 2005. Understanding foods as soft
materials. Nat. Mater. 4, 729–740. doi:10.1038/nmat1496
Mille, Y., Beney, L., Gervais, P., 2005. Compared tolerance to osmotic stress in various
microorganisms: Towards a survival prediction test. Biotechnol. Bioeng. 92, 479–484.
Mille, Y., Obert, J.P., Beney, L., Gervais, P., 2004. New drying process for lactic bacteria based on
their dehydration behavior in liquid medium. Biotechnol. Bioeng. 88, 71–76.
Mimura, T., Rizzello, F., Helwig, U., Poggioli, G., Schreiber, S., Talbot, I.C., Nicholls, R.J., Gionchetti,
P., Campieri, M., Kamm, M.A., 2004. Once daily high dose probiotic therapy (VSL#3) for
maintaining remission in recurrent or refractory pouchitis. Gut 53, 108–114.
doi:10.1136/gut.53.1.108
Minekus, M., Alminger, M., Alvito, P., Ballance, S., Bohn, T., Bourlieu, C., Carrière, F., Boutrou, R.,
Corredig, M., Dupont, D., Dufour, C., Egger, L., Golding, M., Karakaya, S., Kirkhus, B.,
Feunteun, S.L., Lesmes, U., Macierzanka, A., Mackie, A., Marze, S., McClements, D.J.,
Ménard, O., Recio, I., Santos, C.N., Singh, R.P., Vegarud, G.E., Wickham, M.S.J., Weitschies,
W., Brodkorb, A., 2014. A standardised static in vitro digestion method suitable for food – an
international consensus. Food Funct. 5, 1113–1124. doi:10.1039/C3FO60702J
Mitsuyama, K., Masuda, J., Yamasaki, H., Kuwaki, K., Kitazaki, S., Koga, H., Uchida, M., Sata, M.,
2007. Treatment of Ulcerative Colitis with Milk Whey Culture with Propionibacterium
freudenreichii. J. Intest. Microbiol. 21, 143–147. doi:10.11209/jim.21.143
 Future perspective 158

Moats, W.A., 1959. Application of periodic acid-schiff type stains to bacteria in milk. J. Bacteriol. 78,
589–593.
Mordor Intelligence, 2016. Asia-Pacific probiotics market segmented by product type, applications,
ingredients and geography. URL [Link]
pacific-probiotics-market-industry?gclid=CN-Spcm1kc8CFRQ6Gwodxq8FlA (accessed
9.15.16).
Morgan, C.A., Herman, N., White, P.A., Vesey, G., 2006. Preservation of micro-organisms by drying;
A review. J. Microbiol. Methods 66, 183–193. doi:10.1016/[Link].2006.02.017
Morris, J.S., Clark, B.N., Wei, W., Gutstein, H.B., 2010. Evaluating the performance of new approaches
to spot quantification and differential expression in 2-dimensional gel electrophoresis studies.
J. Proteome Res. 9, 595–604. doi:10.1021/pr9005603
Mukherjee, C., Chowdhury, R., Ray, K., 2015. Phosphorus recycling from an unexplored source by
polyphosphate accumulating microalgae and cyanobacteria—a step to phosphorus security in
agriculture. Front. Microbiol. 6. doi:10.3389/fmicb.2015.01421
Muñoz-Provencio, D., Rodríguez-Díaz, J., Collado, M.C., Langella, P., Bermúdez-Humarán, L.G.,
Monedero, V., 2012. Functional analysis of the Lactobacillus casei BL23 sortases. Appl.
Environ. Microbiol. 78, 8684–8693. doi:10.1128/AEM.02287-12
Nandiyanto, A.B.D., Okuyama, K., 2011. Progress in developing spray-drying methods for the
production of controlled morphology particles: From the nanometer to submicrometer size
ranges. Adv. Powder Technol., 4th Asian Particle Technology Symposium (APT2009) - II 22,
1–19. doi:10.1016/[Link].2010.09.011
O’Connor, K., Fletcher, S.A., Csonka, L.N., 2009. Increased expression of Mg2+ transport proteins
enhances the survival of Salmonella enterica at high temperature. Proc. Natl. Acad. Sci. 106,
17522–17527. doi:10.1073/pnas.0906160106
Oldfield, D.J., Singh, H., Taylor, M.W., 2005. Kinetics of heat-induced whey protein denaturation and
aggregation in skim milks with adjusted whey protein concentration. J. Dairy Res. 72, 369–
378. doi:10.1017/S002202990500107X
Páez, R., Lavari, L., Audero, G., Cuatrin, A., Zaritzky, N., Reinheimer, J., Vinderola, G., 2013. Study
of the effects of spray-drying on the functionality of probiotic lactobacilli. Int. J. Dairy Technol.
66, 155–161. doi:10.1111/1471-0307.12038
Papadimitriou, K., Alegría, Á., Bron, P.A., Angelis, M. de, Gobbetti, M., Kleerebezem, M., Lemos,
J.A., Linares, D.M., Ross, P., Stanton, C., Turroni, F., Sinderen, D. van, Varmanen, P., Ventura,
M., Zúñiga, M., Tsakalidou, E., Kok, J., 2016. Stress physiology of lactic acid bacteria.
Microbiol. Mol. Biol. Rev. 80, 837–890. doi:10.1128/MMBR.00076-15
Parrou, J.L., François, J., 1997. A simplified procedure for a rapid and reliable assay of both glycogen
and trehalose in whole yeast cells. Anal. Biochem. 248, 186–188. doi:10.1006/abio.1997.2138
Parvez, S., Malik, K. a., Ah Kang, S., Kim, H.-Y., 2006. Probiotics and their fermented food products
are beneficial for health. J. Appl. Microbiol. 100, 1171–1185. doi:10.1111/j.1365-
2672.2006.02963.x
Passot, S., Cenard, S., Douania, I., Tréléa, I.C., Fonseca, F., 2012. Critical water activity and amorphous
state for optimal preservation of lyophilised lactic acid bacteria. Food Chem., 6th International
Conference on Water in Food 132, 1699–1705. doi:10.1016/[Link].2011.06.012
Pauchard, L., Allain, C., 2003. Buckling instability induced by polymer solution drying. Europhys. Lett.
EPL 62, 897–903. doi:10.1209/epl/i2003-00457-7
Payne, A.N., Zihler, A., Chassard, C., and Lacroix, C. (2012). Advances and perspectives in in vitro
human gut fermentation modeling. Trends Biotechnol. 30, 17–25.
 Future perspective 159

Peighambardoust, S.H., Golshan Tafti, A., Hesari, J., 2011. Application of spray drying for preservation
of lactic acid starter cultures: a review. Trends Food Sci. Technol. 22, 215–224.
doi:10.1016/[Link].2011.01.009
Perdana, J., Bereschenko, L., Fox, M.B., Kuperus, J.H., Kleerebezem, M., Boom, R.M., Schutyser,
M.A.I., 2013. Dehydration and thermal inactivation of Lactobacillus plantarum WCFS1:
Comparing single droplet drying to spray and freeze drying. Food Res. Int. 54, 1351–1359.
doi:10.1016/[Link].2013.09.043
Perdana, J., Fox, M.B., Siwei, C., Boom, R.M., Schutyser, M.A.I., 2014. Interactions between
formulation and spray drying conditions related to survival of Lactobacillus plantarum
WCFS1. Food Res. Int. 56, 9–17. doi:10.1016/[Link].2013.12.007
Pérez-Chabela, M.L., Lara-Labastida, R., Rodriguez-Huezo, E., Totosaus, A., 2013. Effect of spray
drying encapsulation of thermotolerant lactic acid bacteria on meat batters properties. Food
Bioprocess Technol. 6, 1505–1515. doi:10.1007/s11947-012-0865-y
Picot, A., Lacroix, C., 2004. Encapsulation of bifidobacteria in whey protein-based microcapsules and
survival in simulated gastrointestinal conditions and in yoghurt. Int. Dairy J. 14, 505–515.
doi:10.1016/[Link].2003.10.008
Pinto, S.S., Fritzen-Freire, C.B., Benedetti, S., Murakami, F.S., Petrus, J.C.C., Prudêncio, E.S., Amboni,
R.D.M.C., 2015. Potential use of whey concentrate and prebiotics as carrier agents to protect
Bifidobacterium Bb-12 microencapsulated by spray drying. Food Res. Int. 67, 400–408.
doi:10.1016/[Link].2014.11.038
Písecký, J., 2012. Handbook of milk powder manufacture. GEA Niro, Soeborg.
Plé, C., Richoux, R., Jardin, J., Nurdin, M., Briard-Bion, V., Parayre, S., Ferreira, S., Pot, B., Bouguen,
G., Deutsch, S.-M., Falentin, H., Foligné, B., Jan, G., 2015. Single-strain starter experimental
cheese reveals anti-inflammatory effect of Propionibacterium freudenreichii CIRM BIA 129
in TNBS-colitis model. J. Funct. Foods 18, 575–585. doi:10.1016/[Link].2015.08.015
Poddar, D., Das, S., Jones, G., Palmer, J., Jameson, G.B., Haverkamp, R.G., Singh, H., 2014. Stability
of probiotic Lactobacillus paracasei during storage as affected by the drying method. Int. Dairy
J. 39, 1–7. doi:10.1016/[Link].2014.04.007
Potts, M., 1994. Desiccation tolerance of prokaryotes. Microbiol. Rev. 58, 755–805.
Prasad, J., McJarrow, P., Gopal, P., 2003. Heat and osmotic stress responses of probiotic Lactobacillus
rhamnosus HN001 (DR20) in relation to viability after drying. Appl. Environ. Microbiol. 69,
917–925.
Rajam, R., Anandharamakrishnan, C., 2015. Microencapsulation of Lactobacillus plantarum (MTCC
5422) with fructooligosaccharide as wall material by spray drying. LWT - Food Sci. Technol.
60, 773–780. doi:10.1016/[Link].2014.09.062
Rajam, R., Karthik, P., Parthasarathi, S., Joseph, G.S., Anandharamakrishnan, C., 2012. Effect of whey
protein – alginate wall systems on survival of microencapsulated Lactobacillus plantarum in
simulated gastrointestinal conditions. J. Funct. Foods 4, 891–898.
doi:10.1016/[Link].2012.06.006
Ranadheera, C.S., Evans, C.A., Adams, M.C., Baines, S.K., 2015. Microencapsulation of Lactobacillus
acidophilus LA-5, Bifidobacterium animalis subsp. lactis BB-12 and Propionibacterium
jensenii 702 by spray drying in goat’s milk. Small Rumin. Res. 123, 155–159.
doi:10.1016/[Link].2014.10.012
Riveros, B., Ferrer, J., Bórquez, R., 2009. Spray drying of a vaginal probiotic strain of Lactobacillus
acidophilus. Dry. Technol. 27, 123–132. doi:10.1080/07373930802566002
Rochat, T., Bermúdez-Humarán, L., Gratadoux, J.-J., Fourage, C., Hoebler, C., Corthier, G., Langella,
P., 2007. Anti-inflammatory effects of Lactobacillus casei BL23 producing or not a
 Future perspective 160

manganese-dependant catalase on DSS-induced colitis in mice. Microb. Cell Factories 6, 1–10.

doi:10.1186/1475-2859-6-22
Romantsov, T., Guan, Z., Wood, J.M., 2009. Cardiolipin and the osmotic stress responses of bacteria.
Biochim. Biophys. Acta BBA - Biomembr. 1788, 2092–2100.
doi:10.1016/[Link].2009.06.010
Romeo, Y., Obis, D., Bouvier, J., Guillot, A., Fourçans, A., Bouvier, I., Gutierrez, C., Mistou, M.-Y.,
2003. Osmoregulation in Lactococcus lactis: BusR, a transcriptional repressor of the glycine
betaine uptake system BusA. Mol. Microbiol. 47, 1135–1147. doi:10.1046/j.1365-
2958.2003.03362.x
Rossi, F., Zotta, T., Iacumin, L., Reale, A., 2016. Theoretical insight into the heat shock response (HSR)
regulation in Lactobacillus casei and L. rhamnosus. J. Theor. Biol.
doi:10.1016/[Link].2016.04.029
Rossi, M., Amaretti, A., Raimondi, S., 2011. Folate production by probiotic bacteria. Nutrients 3, 118–
134. doi:10.3390/nu3010118
Sabikhi, L., Babu, R., Thompkinson, D.K., Kapila, S., 2010. Resistance of microencapsulated
Lactobacillus acidophilus LA1 to processing treatments and simulated gut conditions. Food
Bioprocess Technol. 3, 586–593. doi:10.1007/s11947-008-0135-1
Sadek, C., Li, H., Schuck, P., Fallourd, Y., Pradeau, N., Le Floch-Fouéré, C., Jeantet, R., 2014. To what
extent do whey and casein micelle proteins influence the morphology and properties of the
resulting powder? Dry. Technol. 32, 1540–1551. doi:10.1080/07373937.2014.915554
Sadek, C., Pauchard, L., Schuck, P., Fallourd, Y., Pradeau, N., Le Floch-Fouéré, C., Jeantet, R., 2015.
Mechanical properties of milk protein skin layers after drying: Understanding the mechanisms
of particle formation from whey protein isolate and native phosphocaseinate. Food Hydrocoll.
48, 8–16. doi:10.1016/[Link].2015.01.014
Sadek, C., Schuck, P., Fallourd, Y., Pradeau, N., Jeantet, R., Le Floch-Fouéré, C., 2016. Buckling and
collapse during drying of a single aqueous dispersion of casein micelle droplet. Food Hydrocoll.
52, 161–166. doi:10.1016/[Link].2015.06.016
Sadek, C., Tabuteau, H., Schuck, P., Fallourd, Y., Pradeau, N., Le Floch-Fouéré, C., Jeantet, R., 2013.
Shape, shell, and vacuole formation during the drying of a single concentrated whey protein
droplet. Langmuir 29, 15606–15613. doi:10.1021/la404108v
Salaün, F., Mietton, B., Gaucheron, F., 2005. Buffering capacity of dairy products. Int. Dairy J. 15, 95–
109. doi:10.1016/[Link].2004.06.007
Samuel R., P., 1872. Improvement in drying and concentrating liquid substances by atomizing.
US125406.
Santivarangkna, C., Higl, B., Foerst, P., 2008a. Protection mechanisms of sugars during different stages
of preparation process of dried lactic acid starter cultures. Food Microbiol. 25, 429–441.
doi:10.1016/[Link].2007.12.004
Santivarangkna, C., Kulozik, U., Foerst, P., 2008b. Inactivation mechanisms of lactic acid starter
cultures preserved by drying processes. J. Appl. Microbiol. 105, 1–13.
Santivarangkna, C., Kulozik, U., Foerst, P., 2007. Alternative drying processes for the industrial
preservation of lactic acid starter cultures. Biotechnol. Prog. 23, 302–315.
Sawatari, Y., Hirano, T., Yokota, A., 2006. Development of food grade media for the preparation of
Lactobacillus plantarum starter culture. J. Gen. Appl. Microbiol. 52, 349–356.
Scholz-Ahrens, K.E., Ade, P., Marten, B., Weber, P., Timm, W., Açil, Y., Glüer, C.-C., Schrezenmeir,
J., 2007. Prebiotics, probiotics, and synbiotics affect mineral absorption, bone mineral content,
and bone structure. J. Nutr. 137, 838S–46S.
 Future perspective 161

Schuck, P., Dolivet, A., Jeantet, R., 2012. Analytical methods for food and dairy powders: Analytical
methods for food and dairy powders. Wiley-Blackwell, Oxford, UK.
Schuck, P., Dolivet, A., Méjean, S., Hervé, C., Jeantet, R., 2013. Spray drying of dairy bacteria: New
opportunities to improve the viability of bacteria powders. Int. Dairy J., Novel technologies for
sustainable dairy products - International Dairy Federation World Dairy Summit 31, 12–17.
doi:10.1016/[Link].2012.01.006
Schuck, P., Dolivet, A., Méjean, S., Zhu, P., Blanchard, E., Jeantet, R., 2009. Drying by desorption: A
tool to determine spray drying parameters. J. Food Eng., Food Powder Technology 94, 199–
204. doi:10.1016/[Link].2008.08.014
Schuck, P., Jeantet, R., Bhandari, B., Chen, X.D., Perrone, Í.T., Carvalho, A.F. de, Fenelon, M., Kelly,
P., 2016. Recent advances in spray drying relevant to the dairy industry: A comprehensive
critical review. Dry. Technol. 34, 1773–1790. doi:10.1080/07373937.2016.1233114
Schutyser, M.A.I., Perdana, J., Boom, R.M., 2012. Single droplet drying for optimal spray drying of
enzymes and probiotics. Trends Food Sci. Technol. 27, 73–82. doi:10.1016/[Link].2012.05.006
Selmer-Olsen, E., Birkeland, S.-E., Sørhaug, T., 1999. Effect of protective solutes on leakage from and
survival of immobilized Lactobacillus subjected to drying, storage and rehydration. J. Appl.
Microbiol. 87, 429–437. doi:10.1046/j.1365-2672.1999.00839.x
Sengupta, D., Sangu, K., Shivaji, S., Chattopadhyay, M.K., 2015. Tolerance of an antarctic bacterium
to multiple environmental stressors. Curr. Microbiol. 71, 483–489. doi:10.1007/s00284-015-
0874-y
Sheehan, V.M., Sleator, R.D., Fitzgerald, G.F., Hill, C., 2006. Heterologous expression of BetL, a
betaine uptake system, enhances the stress tolerance of Lactobacillus salivarius UCC118. Appl.
Environ. Microbiol. 72, 2170–2177.
Silva, J., Carvalho, A.S., Ferreira, R., Vitorino, R., Amado, F., Domingues, P., Teixeira, P., Gibbs, P.A.,
2005. Effect of the pH of growth on the survival of Lactobacillus delbrueckii subsp bulgaricus
to stress conditions during spray-drying. J. Appl. Microbiol. 98, 775–782.
Silva, J., Carvalho, A.S., Teixeira, P., Gibbs, P.A., 2002. Bacteriocin production by spray-dried lactic
acid bacteria. Lett. Appl. Microbiol. 34, 77–81.
Simpson, P.J., Stanton, C., Fitzgerald, G.F., Ross, R.P., 2005. Intrinsic tolerance of Bifidobacterium
species to heat and oxygen and survival following spray drying and storage. J. Appl. Microbiol.
99, 493–501.
Song, S.-H., Cho, Y.-H., Park, J., 2003. Microencapsulation of Lactobacillus casei YIT 9018 using a
microporous glass membrane emulsification system. J. Food Sci. 68, 195–200.
doi:10.1111/j.1365-2621.2003.tb14139.x
Sosa, N., Gerbino, E., Golowczyc, M.A., Schebor, C., Gómez-Zavaglia, A., Tymczyszyn, E.E., 2016.
Effect of galacto-oligosaccharides: maltodextrin matrices on the recovery of Lactobacillus
plantarum after spray-drying. Food Microbiol. 584. doi:10.3389/fmicb.2016.00584
Stoklosa, A.M., Lipasek, R.A., Taylor, L.S., Mauer, L.J., 2012. Effects of storage conditions,
formulation, and particle size on moisture sorption and flowability of powders: A study of
deliquescent ingredient blends. Food Res. Int. 49, 783–791. doi:10.1016/[Link].2012.09.034
Strasser, S., Neureiter, M., Geppl, M., Braun, R., Danner, H., 2009. Influence of lyophilization,
fluidized bed drying, addition of protectants, and storage on the viability of lactic acid bacteria.
J. Appl. Microbiol. 107, 167–177.
Sunny-Roberts, E.O., Knorr, D., 2009. The protective effect of monosodium glutamate on survival of
Lactobacillus rhamnosus GG and Lactobacillus rhamnosus E-97800 (E800) strains during
spray-drying and storage in trehalose-containing powders. Int. Dairy J. 19, 209–214.
doi:10.1016/[Link].2008.10.008
 Future perspective 162

Tatar Turan, F., Cengiz, A., Kahyaoglu, T., 2015. Evaluation of ultrasonic nozzle with spray-drying as
a novel method for the microencapsulation of blueberry’s bioactive compounds. Innov. Food
Sci. Emerg. Technol. 32, 136–145. doi:10.1016/[Link].2015.09.011
Tavares, G.M., Croguennec, T., Carvalho, A.F., Bouhallab, S., 2014. Milk proteins as encapsulation
devices and delivery vehicles: Applications and trends. Trends Food Sci. Technol. 37, 5–20.
doi:10.1016/[Link].2014.02.008
Teixeira, P., Castro, H., Kirby, R., 1996. Evidence of membrane lipid oxidation of spray-dried
Lactobacillus bulgaricus during storage. Lett. Appl. Microbiol. 22, 34–38.
Teixeira, P.C., Castro, M.H., Malcata, F.X., Kirby, R.M., 1995. Survival of Lactobacillus delbrueckii
ssp. bulgaricus following spray-drying. J. Dairy Sci. 78, 1025–1031. doi:10.3168/jds.S0022-
0302(95)76718-2
Teixidó, N., Cañamás, T. p., Usall, J., Torres, R., Magan, N., Viñas, I., 2005. Accumulation of the
compatible solutes, glycine–betaine and ectoine, in osmotic stress adaptation and heat shock
cross-protection in the biocontrol agent Pantoea agglomerans CPA-2. Lett. Appl. Microbiol.
41, 248–252. doi:10.1111/j.1472-765X.2005.01757.x
To, B.C.S., Etzel, M.R., 1997. Spray drying, freeze drying, or freezing of three different lactic acid
bacteria species. J. Food Sci. 62, 576-.
Trautvetter, U., Ditscheid, B., Kiehntopf, M., Jahreis, G., 2012. A combination of calcium phosphate
and probiotics beneficially influences intestinal lactobacilli and cholesterol metabolism in
humans. Clin. Nutr. 31, 230–237. doi:10.1016/[Link].2011.09.013
Tripathi, M.K., Giri, S.K., 2014. Probiotic functional foods: Survival of probiotics during processing
and storage. J. Funct. Foods 9, 225–241. doi:10.1016/[Link].2014.04.030
Upadrasta, A., Stanton, C., Hill, C., Fitzgerald, G.F., Ross, R.P., 2011. Improving the stress tolerance
of probiotic cultures: Recent trends and future directions, in: Tsakalidou, E., Papadimitriou, K.
(Eds.), Stress responses of lactic acid bacteria, Food microbiology and food safety. Springer
US, pp. 395–438. doi:10.1007/978-0-387-92771-8_17
Van Tassell, M.L., and Miller, M.J., 2011. Lactobacillus adhesion to mucus. Nutrients 3, 613–636.
Varunsatian, S., Watanabe, K., Hayakawa, S., Nakamura, R., 1983. Effects of Ca++, Mg++ and Na+
on heat aggregation of whey protein concentrates. J. Food Sci. 48, 42–46. doi:10.1111/j.1365-
2621.1983.tb14785.x
Ventura, M., Kenny, J.G., Zhang, Z., Fitzgerald, G.F., van Sinderen, D., 2005. The clpB gene of
Bifidobacterium breve UCC 2003: transcriptional analysis and first insights into stress
induction. Microbiol. Read. Engl. 151, 2861–2872. doi:10.1099/mic.0.28176-0
Verschuere, L., Rombaut, G., Sorgeloos, P., Verstraete, W., 2000. Probiotic bacteria as biological
control agents in aquaculture. Microbiol. Mol. Biol. Rev. 64, 655–671.
doi:10.1128/MMBR.64.4.655-671.2000
Vesterlund, S., Salminen, K., Salminen, S., 2012. Water activity in dry foods containing live probiotic
bacteria should be carefully considered: a case study with Lactobacillus rhamnosus GG in
flaxseed. Int. J. Food Microbiol. 157, 319–321. doi:10.1016/[Link].2012.05.016
Vouloumanou, E.K., Makris, G.C., Karageorgopoulos, D.E., Falagas, M.E., 2009. Probiotics for the
prevention of respiratory tract infections: a systematic review. Int. J. Antimicrob. Agents 34,
197.e1-197.e10. doi:10.1016/[Link].2008.11.005
Waditee-Sirisattha, R., Sittipol, D., Tanaka, Y., Takabe, T., 2012. Overexpression of serine
hydroxymethyltransferase from halotolerant cyanobacterium in Escherichia coli results in
increased accumulation of choline precursors and enhanced salinity tolerance. FEMS
Microbiol. Lett. 333, 46–53. doi:10.1111/j.1574-6968.2012.02597.x
 Future perspective 163

Wang, Y., Che, L., Selomulya, C., Chen, X.D., 2014. Droplet drying behaviour of docosahexaenoic
acid (DHA)-containing emulsion. Chem. Eng. Sci. 106, 181–189.
doi:10.1016/[Link].2013.11.028
Wang, Y.C., Yu, R.C., Chou, C.C., 2004. Viability of lactic acid bacteria and bifidobacteria in
fermented soymilk after drying, subsequent rehydration and storage. Int. J. Food Microbiol. 93,
209–217. doi:10.1016/[Link].2003.12.001
Watterlot, L., Rochat, T., Sokol, H., Cherbuy, C., Bouloufa, I., Lefèvre, F., Gratadoux, J.-J., Honvo-
Hueto, E., Chilmonczyk, S., Blugeon, S., Corthier, G., Langella, P., Bermúdez-Humarán, L.G.,
2010. Intragastric administration of a superoxide dismutase-producing recombinant
Lactobacillus casei BL23 strain attenuates DSS colitis in mice. Int. J. Food Microbiol., The
16th CBL (Club des Bactéries Lactiques) Symposium, May 2009, Toulouse, France 144, 35–
41. doi:10.1016/[Link].2010.03.037
Wilderdyke, M.R., Smith, D.A., Brashears, M.M., 2004. Isolation, identification, and selection of lactic
acid bacteria from alfalfa sprouts for competitive inhibition of foodborne pathogens. J. Food
Prot. 67, 947–951.
Wolfe, J., Bryant, G., 1999. Freezing, drying, and/or vitrification of membrane– solute–water systems.
Cryobiology 39, 103–129. doi:10.1006/cryo.1999.2195
Wood, J.M., 2011. Bacterial Osmoregulation: A paradigm for the study of cellular homeostasis. Annu.
Rev. Microbiol. 65, 215–238. doi:10.1146/annurev-micro-090110-102815
Wu, R., Song, X., Liu, Q., Ma, D., Xu, F., Wang, Q., Tang, X., Wu, J., 2016. Gene expression of
Lactobacillus plantarum FS5-5 in response to salt stress. Ann. Microbiol. 66, 1181–1188.
doi:10.1007/s13213-016-1199-1
Wu, R., Zhang, W., Sun, T., Wu, J., Yue, X., Meng, H., Zhang, H., 2011. Proteomic analysis of
responses of a new probiotic bacterium Lactobacillus casei Zhang to low acid stress. Int. J.
Food Microbiol. 147, 181–187. doi:10.1016/[Link].2011.04.003
Wu, V.C.H., 2008. A review of microbial injury and recovery methods in food. Food Microbiol. 25,
735–744. doi:10.1016/[Link].2008.04.011
Wu, W.D., Liu, W., Gengenbach, T., Woo, M.W., Selomulya, C., Chen, X.D., Weeks, M., 2014.
Towards spray drying of high solids dairy liquid: Effects of feed solid content on particle
structure and functionality. J. Food Eng. 123, 130–135. doi:10.1016/[Link].2013.05.013
Würth, R., Hörmannsperger, G., Wilke, J., Foerst, P., Haller, D., Kulozik, U., 2015. Protective effect of
milk protein based microencapsulation on bacterial survival in simulated gastric juice versus
the murine gastrointestinal system. J. Funct. Foods 15, 116–125. doi:10.1016/[Link].2015.02.046
Xing, Y., Li, A., Felker, D.L., Burggraf, L.W., 2014. Nanoscale structural and mechanical analysis of
Bacillus anthracis spores inactivated with rapid dry heating. Appl. Environ. Microbiol. 80,
1739–1749. doi:10.1128/AEM.03483-13
Yan, H., Jia, L.-H., Lin, Y.-P., Jiang, N., 2008. Glycerol accumulation in the dimorphic yeast
Saccharomycopsis fibuligera: cloning of two glycerol 3-phosphate dehydrogenase genes, one
of which is markedly induced by osmotic stress. Yeast 25, 609–621. doi:10.1002/yea.1606
Yonekura, L., Sun, H., Soukoulis, C., Fisk, I., 2014. Microencapsulation of Lactobacillus acidophilus
NCIMB 701748 in matrices containing soluble fibre by spray drying: Technological
characterization, storage stability and survival after in vitro digestion. J. Funct. Foods 6, 205–
214. doi:10.1016/[Link].2013.10.008
You, L., Cox, R.S., Weiss, R., Arnold, F.H., 2004. Programmed population control by cell–cell
communication and regulated killing. Nature 428, 868–871. doi:10.1038/nature02491
Zaprasis, A., Brill, J., Thüring, M., Wünsche, G., Heun, M., Barzantny, H., Hoffmann, T., Bremer, E.,
2013. Osmoprotection of Bacillus subtilis through import and proteolysis of proline-containing
peptides. Appl. Environ. Microbiol. 79, 576–587. doi:10.1128/AEM.01934-12
 Future perspective 164

Zhai, Z., Douillard, F.P., An, H., Wang, G., Guo, X., Luo, Y., Hao, Y., 2014. Proteomic characterization
of the acid tolerance response in Lactobacillus delbrueckii subsp. Bulgaricus CAUH1 and
functional identification of a novel acid stress-related transcriptional regulator Ldb0677.
Environ. Microbiol. 16, 1524–1537. doi:10.1111/1462-2920.12280
Zhang, L., Huang, S., Ananingsih, V.K., Zhou, W., Chen, X.D., 2014. A study on Bifidobacterium lactis
Bb12 viability in bread during baking. J. Food Eng. 122, 33–37.
doi:10.1016/[Link].2013.08.029
Zhang, Q., Feng, Y., Deng, L., Feng, F., Wang, L., Zhou, Q., Luo, Q., 2011. SigB plays a major role in
Listeria monocytogenes tolerance to bile stress. Int. J. Food Microbiol. 145, 238–243.
doi:10.1016/[Link].2010.12.028
Zhang, W., Sun, Z., Bilige, M., Zhang, H., 2015. Complete genome sequence of probiotic Lactobacillus
plantarum P-8 with antibacterial activity. J. Biotechnol. 193, 41–42.
doi:10.1016/[Link].2014.11.011
Zhang, Y., Lin, J., Zhong, Q., 2016. Effects of media, heat adaptation, and outlet temperature on the
survival of Lactobacillus salivarius NRRL B-30514 after spray drying and subsequent storage.
LWT - Food Sci. Technol. 74, 441–447. doi:10.1016/[Link].2016.08.008
Zheng, X., Fu, N., Duan, M., Woo, M.W., Selomulya, C., Chen, X.D., 2015. The mechanisms of the
protective effects of reconstituted skim milk during convective droplet drying of lactic acid
bacteria. Food Res. Int. 76, 478–488. doi:10.1016/[Link].2015.07.045
Zheng, X., Fu, N., Huang, S., Jeantet, R., Chen, X.D., 2016. Exploring the protective effects of calcium-
containing carrier against drying-induced cellular injuries of probiotics using single droplet
drying technique. Food Res. Int. doi:10.1016/[Link].2016.10.034
Zhu, Z., Luan, C., Zhang, H., Zhang, L., Hao, Y., 2016. Effects of spray drying on Lactobacillus
plantarum BM-1 viability, resistance to simulated gastrointestinal digestion, and storage
stability. Dry. Technol. 34, 177–184. doi:10.1080/07373937.2015.1021009
Zotta, T., Guidone, A., Ianniello, R. g., Parente, E., Ricciardi, A., 2013. Temperature and respiration
affect the growth and stress resistance of Lactobacillus plantarum C17. J. Appl. Microbiol. 115,
848–858. doi:10.1111/jam.12285
Zotta, T., Parente, E., Ricciardi, A., 2017. Aerobic metabolism in the genus Lactobacillus: impact on
stress response and potential applications in the food industry. J. Appl. Microbiol. n/a-n/a.
doi:10.1111/jam.13399
Zuo, F., Yu, R., Khaskheli, G.B., Ma, H., Chen, L., Zeng, Z., Mao, A., Chen, S., 2014. Homologous
overexpression of alkyl hydroperoxide reductase subunit C (ahpC) protects Bifidobacterium
longum strain NCC2705 from oxidative stress. Res. Microbiol. 165, 581–589.
doi:10.1016/[Link].2014.05.040
 List of Figures 165

List of Figures

Figure 1. Various health benefits of probiotics and its possible distribution of mechanisms. Some
benefits might be widespread among commonly studied probiotic genera; others might be
frequently observed among most strains of a probiotic species; others may be rare and present in
only a few strains of a given species (Concluded from Hill et al., 2014; Parvez et al., 2006). ....... 4
Figure 2. (A) U.S. Probiotics market revenue by product, 2014 - 2023 (USD Million) and (B)
probiotic ingredients market size, by region, 2015 vs 2020 (USD Million) (Adapted from Global
Market Insights, 2016; MarketsandMarkets, 2015). ....................................................................... 5
Figure 3. Schematic of a typical spray dryer and its operation process. ................................................. 6
Figure 4. Various particle morphologies prepared using the spray drying method (adapted from
Nandiyanto and Okuyama, 2011). .................................................................................................. 7
Figure 5. Overview of main research groups focusing on the application of spray drying in probiotic
production. ...................................................................................................................................... 9
Figure 6. (A): The strains produced by dehydration-induced stress (source from (Wolfe and Bryant,
1999)): (a) The lamellar fluid phase at high hydration. (b) The geometric strain produced at
lower water content. (c) Transition to gel phase at lower water contents. (d) The mixed phase of
lipid species with high (top) and low (bottom) hydration. (e) Separation of hydrophilic and
hydrophobic molecules. (f) tHE topological response emerged at very low hydration. (B): The
surface morphology of Lc. cremoris cells dried with different air temperature (Adapted from Fu
et al., 2013a). ................................................................................................................................. 25
Figure 7. Effect of spraying-induced shear stress on cellular organization (Adapted from Guerin et al.,
2017). ............................................................................................................................................ 28
Figure 8. (A) a simplified phase diagram of a phospholipid associated with drying and rehydration
processes: During drying at a given temperature until cells reach a relatively low moisture
content (closed circle), cell membrane may remain in liquid phase in the presence of a sugar at
this same point. (B) According to the water replacement hypothesis: Sugars depress the
membrane phase transition temperature (a) by specifically interacting with phospholipid
headgroups, or (b) by vitrification of sugars between the membranes. (c) Polymeric sugars with a
very high molecular weight are not capable of depressing the Tm because they have fewer
osmotic effects and are excluded from the intermembrane space during the removal of water.
(Adapted from Santivarangkna et al., 2008a) ............................................................................... 29
Figure 9. (A) Schematic diagram of a glass-filament single droplet drying experimental set-up
(droplet diameter, weight, and temperature were measured in separate runs). (B) An example of
the observation of morphology evolution during the single droplet drying (Adapted from Wang
et al., 2014). .................................................................................................................................. 35
Figure 10. The most influential factors and adverse stresses experienced during the growth, spray
drying, storage and application of probiotics (Adapted from Huang et al., 2017). ....................... 48
Figure 11. Diagrams of (A) the conventional process used in spray drying of probiotic bacteria and
(B) the innovative 2-in-1 drying process. ..................................................................................... 52
Figure 12. (A) Evaporation power (PGev, power required to evaporate water) and heating power
(PGaq, the power required to heat the drying air), (B) Total drying time and (C) Total drying
energy as a function of feed moisture content. (adapted from Kajiyama and Park, 2011) ........... 54
Figure 13. Osmolality of sweet whey media and standard lab media (MRS and YEL) as a function of
the TS (w/w %) ............................................................................................................................. 58
 Future perspective 166

Figure 14. Final bacteria populations (left Y axis) and the dependency (right Y axis) of bacteria final
population on casein peptone of (A) L. casei 48-h culture and (B) P. freudenreichii 120-h culture
in the sweet whey culture media with different TS, with or without casein peptone
supplementation (compared to the standard culture media, i.e. MRS broth for L. casei and YEL
broth for P. freudenreichii). Mean ± SEM, n = 6. Different letters mean significant difference
between bacteria populations (p < 0.05). Different number of * means significant difference in
dependency (p < 0.05). .................................................................................................................. 63
Figure 15. Growth curve (CFU) of L. casei BL23 in MRS broth, 5% sweet whey with casein peptone
supplementation and 30% sweet whey without casein peptone supplementation. ....................... 65
Figure 16. Acidification curve of L. casei BL23 in MRS broth, 5% sweet whey with casein peptone
supplementation and 30% sweet whey without casein peptone supplementation. ....................... 66
Figure 17. Remaining bacteria populations (left Y axis) and survival (right Y axis) of (A) L. casei and
(B) P. freudenreichii in the sweet whey culture media with different TS after spray drying. Mean
± SEM, n = 3. Different letters mean significant difference between bacteria populations (p <
0.05). Different number of * means significant difference between survival rates (p < 0.05). .... 67
Figure 18. Particle size volume of spray dried powders of (A) L. casei and (B) P. freudenreichii from
sweet whey culture media with different TS. Curves were obtained from the average values of
duplicate experiments. The size range was 0.1 to 2000 µm; the volume values in the range 0.1 to
1 µm and 200 to 2000 µm were all 0. ........................................................................................... 69
Figure 19. SEM observation of spray dried powders of L. casei obtained from (A,B) sweet whey
culture media with 5% TS and (C,D) sweet whey culture media with 30% TS. D: the size of the
object is 5× bigger than in B (Image with same magnification as B inserted in D). .................... 70
Figure 20. Survival of bacteria grown and spray dried in 5% medium, grown in 5% medium but 30%
added before drying, grown and spray dried in 30% medium. Mean ± SEM, n = 3. Different
number of * means significant difference between survival rates (p < 0.05). .............................. 71
Figure 21. Change in bacteria populations in spray dried powders of (A) [Link] and (B)
[Link] from sweet whey culture media with different TS during storage at 4°C for 120
days. Mean ± SEM, n = 3. (C) Final log reduction of two probiotic strains when stored for 120
days. Different number of * means significant difference between survival rates (p<0.05) ........ 74
Figure 22. Survival of L. acidophilus CIRM-BIA 1674 grown in different TS sweet whey after spray
drying at inlet temperature 185°C and outlet temperature 75°C. Different letters above the
column mean significant difference between bacterial survivals (p < 0.05)................................. 75
Figure 23. An example to describe the growth curve of LGG at 37°C fitted with the Gompertz model.
The solid curve is experimental curve, while the dotted one represents the fitted curve. ............. 83
Figure 24. P. freudenreichii were cultivated for 72 h in the indicated growth media until stationary
phase, then subjected to (A) heat (60°C for 10 min), (B) acid (pH 2.0 for 1 h), or (C) bile salts (1
g l-1 for 1 h) challenge. Viable propionibacteria were enumerated by plate counting in treated and
control cultures, and then expressed as percent survival. Different letters above the columns
indicate significant differences (P < 0.05). ................................................................................... 91
Figure 25. Hyperconcentrated sweet whey culture triggers overexpression of stress proteins.
Propionibacteria were cultivated for 72 h in isotonic 5% (A) or hyperconcentrated 30% (B) sweet
whey. The proteins accumulated in the cellular compartment were then analyzed by 2D
electrophoresis (pH 4 to 7 isoelectric focusing and then SDS-12.5% PAGE) prior to differential
analysis using the SameSpot software. The proteins induced in 30% sweet whey are indicated by
arrows. Their identities, revealed by trypsinolysis and NanoLC–MS-MS, are shown in Table 11.
...................................................................................................................................................... 92
Figure 26. Hyperconcentrated sweet whey culture triggers intracellular accumulation of
carbohydrates. (A to C) Propionibacteria were cultivated for 72 h in YEL medium (A) or in
isotonic 5% (B) or hyperconcentrated 30% (C) sweet whey. Periodic acid-Schiff staining was
 Future perspective 167

applied to smears of the cultures. (D and E) Glycogen (D) and trehalose (E) were then
enzymatically quantified in whole-cell extracts. The results are expressed as concentrations of
generated glucose per unit weight of cellular wet matter. Different letters above the bars indicate
significant differences (P_0.05). ................................................................................................... 96
Figure 27. Propionibacteria were cultivated in YEL (A, D, G, and J) or in 5% (B, E, H, and K) or 30%
(C, F, I, and L) sweet whey. (A to C) Polyphosphate metachromatic granules were evidenced by
Neisser staining and observed at ×1,000 magnification. (D to I) Polyphosphate was visualized by
DAPI staining prior to epifluorescence microscopy (D to F) or confocal microscopy (G to I)
observation. Green fluorescence indicates cytosolic poly(P), and blue indicates DNA. (J to L)
Transmission electron micrographs of propionibacteria. The positions that were occupied by
granules appear as holes in the bacterial cytoplasm (examples are indicated by the arrows). ...... 98
Figure 28. (A) The size distribution of dispersed LGG cells in the 0.5% w/v peptone water. (B) The
size distribution of dispersed elements in three LGG-suspended milk matrixes. (C) The milk
serum protein content of the three bacteria-suspended milk matrixes indicated by SDS-PAGE.
SM: UHT skim milk, CaM: Ca-added skim milk, and CaAggrM: Ca-aggregated skim milk. ... 101
Figure 29. Distribution of transverse relaxation time (T2) of the three bacteria-suspended milk
matrixes measured by a low-field 1H nuclear magnetic resonance (NMR) spectrometer. ......... 102
Figure 30. Changes of droplet temperature (solid curves, corresponding to the left ordinate) and mass
(dotted curves, corresponding to the right ordinate) in three bacteria-suspended milk matrixes
(UHT skim milk, Ca-added skim milk, and Ca-aggregated skim milk) and pure water during
single-droplet drying. .................................................................................................................. 103
Figure 31. Droplet morphology and the corresponding drying time of three bacteria-suspended milk
matrixes (UHT skim milk, Ca-added skim milk, and Ca-aggregated skim milk) at four
representative events during single-droplet drying. .................................................................... 104
Figure 32. Micromorphology of particles dried from (A) UHT skim milk, (B) Ca-added skim milk,
and (C) Ca-aggregated skim milk. Pictures were observed by scanning electron microscopy. For
each milk matrix, four pictures were used to display the particle morphology from different focus
sites (from left to right): surface of intact particle at ×100 magnification, vacuole structure at
×100 magnification, shell section at ×400 magnification, and shell section at ×10000
magnification. ............................................................................................................................. 106
Figure 33. Inactivation of LGG in three bacteria-suspended milk matrixes (UHT skim milk, Ca-added
skim milk, and Ca-aggregated skim milk) during single-droplet drying. (A) The bacteria
population vs the drying time, (B) the bacteria population vs the droplet temperature, and (C) the
bacteria population vs the droplet moisture content. .................................................................. 108
Figure 34. Survival (log N/N0) of (A) L. rhamnosus GG, (B) L. casei Zhang and (C) L. plantarum P-8
in lactose solutions supplemented with different concentration of MgCl2 (-■-), ZnCl2 (-▲-) and
NaCl (-●-) after heat treatment. Significant difference was compared between the salt-treated
samples and the samples without adding salt, i.e. 0 mmol l-1 added (*: 0.01<P<0.05, **:
0.001<P<0.01, ***: P<0.001). .................................................................................................... 112
Figure 35. The regrowth curves of (A) L. rhamnosus GG, (B) L. casei Zhang and (C) L. plantarum P-
8 grown in MRS broth containing different concentrations of MgCl2 after heat challenge. The
curve of bacteria grown in MRS broth before heat challenge (i.e. -■- Unheated) is provided as a
reference...................................................................................................................................... 116
Figure 36. The schematic diagram of three drying process: (A) Lab-scale spray drying (for obtaining
Powder 1), (B) Semi industrial pilot scale one-stage spray drying (for obtaining Powder 2) and
(C) Semi industrial pilot scale multi-stage spray drying (for obtaining Powder 3). ................... 124
Figure 37. Survival of probiotics after spray drying at lab-scale or semi industrial scale. Bacterial
population (refers to column, left Y-axis) and survival (refers to curve, right Y-axis) of (A)
 Future perspective 168

[Link] BL23 and (B) [Link] ITG20 after spray drying. P1 refers to the powders from
lab-scale spray drying, and P2 refers to that of from semi industrial pilot scale. ....................... 128
Figure 38. Survival of probiotics after one-stage spray drying or multi-stage drying at semi industrial
scale. Bacterial population (refers to column, left Y-axis) and survival (refers to curve, right Y-
axis) of (A) [Link] BL23 and (B) [Link] ITG20 after spray dryingin the semi
industrial pilot scale. P2 refers to the powders from one-stage spray drying, and P3 refers to that
of from multi-stage drying process. * means significant difference between the survival rates
(p<0.05)....................................................................................................................................... 129
Figure 39. Storage stability of [Link] BL23 powders from different drying processes (expressed as
reduction of survival) during 6 months storage at (A) 4°C and (B) 25°C. P1 refers to the powders
from lab-scale spray drying, P2 refers to that of from one-stage spray drying at semi industrial
pilot scale, and P3 refers to that of from multi-stage drying process at semi industrial pilot scale.
P2-vacuum and P3-vacuum mean the powders stored under vacuum. ....................................... 131
Figure 40. Storage stability of P. freudenreichii ITG20 powders from different drying processes
(expressed as reduction of survival) during 6 months storage at (A) 4°C and (B) 25°C. P1 refers
to the powders from lab-scale spray drying, P2 refers to that of from one-stage spray drying at
semi industrial pilot scale, and P3 refers to that of from multi-stage drying process at semi
industrial pilot scale. P2-vacuum and P3-vacuum mean the powders stored under vacuum.
Presence of * means significant difference in population compared to other groups at the same
storage time (p < 0.05). Different number of * indicates significant difference between each other
(p < 0.05). .................................................................................................................................... 132
Figure 41. The survival of probiotic in 30 wt% fresh culture and powders from Multi-stage drying
process (P3), (A) [Link] BL23 in simulated gastric fluid (SGF), (B) [Link] BL23 in simulated
intestinal fluid (SIF), and (C) P. freudenreichii ITG20 in simulated gastric fluid (SGF), (D) P.
freudenreichii ITG20 in simulated intestinal fluid (SIF). ........................................................... 134
Figure 42. The key four elements of the current 2-in-1 process. To further extend the applicability of
this process, each element is a matter of investigation to form a new technical route in terms of
probiotic production. ................................................................................................................... 141
Figure 43. Schematic diagram of two hypothetic particles used to encapsulate and deliver probiotics
which have two different action mechanisms. (A) Layer-by-layer encapsulation for delivery of
probiotics that exert their beneficial effects through the interaction between their surface
molecules and host intestinal cells or mucosa (inspired by Anselmo et al., 2016): (1) The first
layer of shell protects the particle through gastric phase; (2) The second layer of shell is
gradually digested in intestinal phase to partly protect the particle core from the bile and
gradually release the particle core; (3) The particle core is collapsed and release the probiotics
into the intestinal phase. (B) Encapsulation of porous core particle for delivery of probiotics that
exert their beneficial effects through secretion of beneficial molecules (inspired by Mahbubani et
al., 2014 and Nandiyanto and Okuyama, 2011): (1) The encapsulated layer protects the particle
core through gastric phase; (2) The porous particle core allows water to penetrate inside to
rehydrate and to activate probiotics, but hinders the bile salt to penetrate inside; (3) The porous
particle core allows the release of probiotic beneficial secretions. ............................................. 143
Figure 44. Schematic diagram of expected development in vitro digestion system for study of
probiotics-microbiota interaction. ............................................................................................... 144
 List of Tables 169

List of Tables

Table 1. Microorganisms commonly proposed as probiotics or potential probiotics (Concluded from

Cani and Van Hul, 2015; Felis et al., 2009; Foligné et al., 2013b; Hill et al., 2014; Holzapfel et
al., 2001). ........................................................................................................................................ 3
Table 2. General production pattern and specific energy consumption of commonly used drying
processes. (Modified from Huang et al., 2017)............................................................................... 8
Table 3. Review of key factors determining viability of probiotic bacteria (Lb. refers to Lactobacillus,
Lc. refers to lactococcus, S. refers to Streptococcus, B. refers to bifidobacteria, P. refers to
propionibacteria) upon spray drying (Modified from Huang et al., 2017)................................... 13
Table 4. Examples of genes and proteins involved in probiotics adaptation towards spray drying-
related stress (Adapted from Huang et al., 2017).......................................................................... 18
Table 5. Examples of compatible solute accumulation on improvement of bacterial stress tolerance
(Adapted from Huang et al., 2017). .............................................................................................. 22
Table 6. The production, exportation and price of whey in the European Union. (Data source: CNIEL,
2011~2016) ................................................................................................................................... 53
Table 7. Physical and chemical analysis of the sweet whey powder (composition expressed as w/w%).
...................................................................................................................................................... 57
Table 8. Size distribution of spray dried powders from culture media with different TS content. ....... 70
Table 9. The bacterial population of different lactobacillus strains grown in 5% or 30% TS sweet
whey media for 48 hours............................................................................................................... 75
Table 10. The values of λ and A from the curve fitted by Gompertz model in Figure 23 .................... 83
Table 11. P. freudenreichii CIRM BIA 129 proteins overexpressed in hyperconcentrated 30% sweet
whey culture, compared to isotonic 5% sweet whey culture, as identified by NanoLC–MS-MS 93
Table 12. Viscocity and distribution of T2 relaxation time of the three bacteria-suspended milk
matrixes ....................................................................................................................................... 102
Table 13. The heat-induced changes of lag time duration λ–λ0 and asymptotic value A-A0 between the
regrowth curves of L. rhamnosus GG, L. casei Zhang and L. plantarum P-8 in MRS with and
without inorganic salt supplementation. ..................................................................................... 115
Table 14. The physical properties of feed concentrates and powders. ................................................ 127
Table 15. The enumeration of contaminating flora (thermophillic and coliform floras) .................... 130
 RÉSUMÉ ABSTRACT

Séchage par atomisation des bactéries probiotiques : des mécanismes de protection à la production à l’échelle pilote
Séchage par atomisation des bactéries probiotiques : des méca- Spray drying of probiotic bacteria: From molecular mechanism
nismes de protection à la production à l’échelle pilote to pilot-scale production
Les probiotiques sont des microorganismes vivants qui, ingérés Probiotics are live microorganisms that, when administered in
en quantité suffisante, exercent des effets positifs sur la santé. adequate amounts, confer a health benefit on the host. Freeze
La lyophilisation est aujourd’hui questionnée quant à sa consom- drying, the reference drying method, is currently challenged
mation d’énergie et son caractère discontinu. S’il offre une al- because of its low energy-efficiency and productivity. Therefore,
ternative pour produire massivement des poudres probiotiques spray drying is expected to be an alternative and sustainable
à faible coût, le séchage par atomisation induit quant à lui des method for producing probiotic powders. The issue remains
stress thermiques et oxydatifs conduisant à des pertes de viabi- in the considerable inactivation of probiotics caused by high
lité rédhibitoires. temperature and dehydration during the process.
Dans ce travail, un procédé innovant de séchage par atomisation In this work, a novel spray-drying process for continuous
est proposé. Du lactosérum doux concentré (jusqu’à 30% p/p) production of probiotics was challenged. Concentrated sweet
est utilisé à la fois comme support de culture et de séchage de whey (up to 30% w/w dry matter) was used to both culture
P. freudenreichii et L. casei. Ce procédé élimine les étapes inter- and spray dry P. freudenreichii ITG P20 and L. casei BL23. This
médiaires à risque de contamination élevé, accroît la biomasse et process cuts down the steps between culturing and drying (e.g.
améliore la viabilité des bactéries. harvesting, washing), increases the cell population after growth
Les mécanismes sous-jacents ont été explorés au plan de la and improves spray drying productivity and probiotic viability.
résistance bactérienne et des conditions de séchage. Le milieu The mechanisms were explored from bacterial physiology and drying
concentré induit une osmoadaptation des bactéries par expres- process conditions. The hypertonic stress led to overexpression of
sion de protéines de stress et accumulation de solutés compa- key stress proteins and accumulation of intracellular compatible
tibles, conduisant à une tolérance accrue des probiotiques à solutes, which enhanced multistress tolerance. The presence of
différents stress. La présence d’agrégats et la concentration en protein aggregates and optimal concentration of Mg2+ in matrix
Mg2+ du milieu concentré pourraient également être impliquées. may also be involved.
Le scale-up du procédé a été étudié : un schéma technologique The feasibility of scaling up this process was validated. A multi-
semi-industriel impliquant séchage par atomisation, sur bande stage semi industrial drying process, coupling spray-drying with
et en lit fluidisé a permis d’atteindre une viabilité de 100% belt drying and fluid-bed drying, was applied to further improve
(> 109 CFU g-1). Par ailleurs, la fonctionnalité des poudres probio- the probiotic viability to 100% (> 109 CFU g-1). Moreover, the
tiques a été évaluée in vitro and in vivo sur modèle porcelet. Ce functionality of these probiotic powders was investigated in vitro
travail ouvre de nouvelles perspectives pour l’obtention à grande and in vivo with piglet model. This work opens new avenues
échelle et selon un procédé sobre et durable de préparations pro- for the sustainable development of new starter and probiotic Song HUANG • 30 mai 2017
biotiques. preparations with enhanced robustness.
Thèse AGROCAMPUS OUEST ÉCOLE DOCTORALE • Vie - Agro - Santé (VAS)
Mots-clés : probiotiques ; séchage par atomisation ; osmorégula- Keywords: Probiotics; spray drying; osmoregulation; stress sous le label de l’Université Bretagne Loire LABORATOIRE D’ACCUEIL • UMR INRA - AGROCAMPUS OUEST
tion ; tolérance ; viabilité tolerance; viability pour obtenir le grade de
Science et Technologie du Lait et de l’Œuf (STLO)
DOCTEUR D’AGROCAMPUS OUEST
Spécialité Science de l’aliment THÈSE EN CO-TUTELLE • School of Chemical and Environmental
Engineering, Soochow University, Suzhou

Séchage par atomisation MURIEL THOMAS

AGROCAMPUS OUEST • Institut supérieur des sciences agronomiques, Thèse B-298 — 2017-15 • HUANG Song Directrice de Recherche, INRA Micalis, France / présidente
agroalimentaires, horticoles et du paysage des bactéries probiotiques : CATHERINE STANTON
Professeure, University College Cork, Irlande / rapporteure
65 rue de Saint-Brieuc – CS84215 – F-35042 Rennes Cedex
Tél. : 02 23 48 50 00
[Link]
des mécanismes de protection à la CORDELIA SELOMULYA
Professeure, Monash University, Australie / rapporteure
production à l’échelle pilote MAARTEN SCHUTYSER
Professeur, Wageningen University, Pays-Bas / examinateur
XIAO DONG CHEN
Professeur, Soochow University, Chine / directeur de thèse
ROMAIN JEANTET
Professeur AGROCAMPUS OUEST, UMR INRA-AO STLO,
France / directeur de thèse
GWENAËL JAN
Chargé de recherche, UMR INRA-AO STLO, France /
co-directeur de thèse