Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Research Collection
Doctoral Thesis
Novel Technologies for detection, production and screening ofstress tolerant bifidobacteria
Author(s): Reimann, Sebastian
Publication Date: 2009
Permanent Link: https://doi.org/10.3929/ethz-a-005939078
Rights / License: In Copyright - Non-Commercial Use Permitted
This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.
ETH Library
Diss. ETH Nr. 18627
NOVEL TECHNOLOGIES FOR DETECTION,
PRODUCTION AND SCREENING OF STRESS
TOLERANT BIFIDOBACTERIA
A dissertation submitted to
Swiss Federal Institute of Technology Zurich (ETH ZURICH)
for the degree of
Doctor of Sciences
presented by
SEBASTIAN REIMANN
Dipl. Natw. ETH
Born April 24, 1979
Citizen of Oberhof AG
Accepted on the recommendation of
Prof. Dr. Christophe Lacroix, examiner
Prof. Dr. Leo Meile, co-examiner
Prof. Dr. Paul Ross, co-examiner
2009
Table of contents
I
Table of contents
TABLE OF CONTENTS.................................. ................................................................. I
ABBREVIATIONS...................................... ..................................................................... V
SUMMARY..................................................................................................................... VI
ZUSAMMENFASSUNG .................................... ............................................................. IX
1. INTRODUCTION...................................................................................................... 1
1.1 Human gastrointestinal tract microbiota .................................................................................................... 1
1.2 Bifidobacterium ............................................................................................................................................. 3 1.2.1 General characteristics ................................................................................................................................ 3 1.2.2 Metabolism of Bifidobacteria ..................................................................................................................... 4 1.2.3 Role of Bifidobacteria in the GIT ............................................................................................................... 5 1.2.4 Transcriptomics of Bifidobacterium longum NCC2705 ............................................................................. 6
1.3 Probiotic concept ........................................................................................................................................... 7
1.4 Health promoting effects related to probiotics............................................................................................ 9
1.5 Potential risks associated with probiotics .................................................................................................... 2
1.6 Technological requirements of probiotic bacteria .................................................................................... 13
1.7 Technologies to improve cell viability and yields ..................................................................................... 15 1.7.1 Production technologies........................................................................................................................... 15
1.7.1.1 Optimization of growth medium ..................................................................................................... 15 1.7.1.2 Stress adaptation.............................................................................................................................. 16 1.7.1.3 Immobilized cell (IC) technology in continuous cultures................................................................ 20
1.7.1.3.1 Principle of cell immobilization ................................................................................................. 20 1.7.1.3.2 Techniques of cell immobilization ............................................................................................. 21 1.7.1.3.3 Physiological changes of immobilized cells ............................................................................... 22 1.7.1.3.4 Advantages of continuous IC cultures ........................................................................................ 23
1.7.2 Stabilization of probiotics ......................................................................................................................... 26 1.7.2.1 Microencapsulation ......................................................................................................................... 26 1.7.2.2 Spray drying .................................................................................................................................... 27
1.7.2.2.1 Principle of spray drying............................................................................................................ 28 1.7.2.2.2 Parameters affecting cell viability during spray drying .............................................................. 31 1.7.2.2.3 Parameters affecting storage of dried powders ........................................................................... 36 1.7.2.2.4 Rehydration ................................................................................................................................ 38
1.7.3 Enumeration of viable probiotic cells ....................................................................................................... 39
Table of contents
I
1.8 Hypothesis and background of this study.................................................................................................. 41 1.8.1 Background and general objective ............................................................................................................ 41 1.8.2 Specific objectives .................................................................................................................................... 42 1.8.3 Hypotheses................................................................................................................................................ 43
2 IMPROVED TOLERANCE TO BILE SALTS OF AGGREGATED B. LONGUM NCC2705 PRODUCED DURING CONTINUOUS CULTURE WITH IMMOBILIZED CELLS.............................................. ............................................................................. 44
2.1 Abstract ........................................................................................................................................................ 45
2.2 Introduction ................................................................................................................................................. 46
2.3 Material and Methods ................................................................................................................................. 48 2.3.1 Strain and medium .................................................................................................................................... 48 2.3.2 Cell immobilization .................................................................................................................................. 48 2.3.3 Continuous fermentations ......................................................................................................................... 49 2.3.4 Batch fermentation.................................................................................................................................... 51 2.3.5 Viable cells enumeration........................................................................................................................... 51 2.3.6 Dry biomass determination ....................................................................................................................... 52 2.3.7 Microscopic observation........................................................................................................................... 53 2.3.8 Metabolite analysis ................................................................................................................................... 53 2.3.9 Fatty acid analysis..................................................................................................................................... 53 2.3.10 Heat test................................................................................................................................................54 2.3.11 Bile salt test .......................................................................................................................................... 54 2.3.12 Tolerance to antibiotics ........................................................................................................................ 55 2.3.13 Statistical analyzes ............................................................................................................................... 56
2.4 Results........................................................................................................................................................... 56 2.4.1 Biomass production .................................................................................................................................. 56 2.4.2 Metabolic Activity .................................................................................................................................... 60 2.4.3 Cell sensitivity to heat and porcine bile salts ............................................................................................ 63 2.4.4 Tolerance to antibiotics............................................................................................................................. 64 2.4.5 Cell membrane fatty acid composition ..................................................................................................... 65
2.5 Discussion ..................................................................................................................................................... 67
3 DEVELOPMENT OF A REAL-TIME RT-PCR METHOD FOR ENUMER ATION OF VIABLE BIFIDOBACTERIUM LONGUM CELLS IN DIFFERENT MORPHOLOGIES . 72
3.1 Abstract ........................................................................................................................................................ 73
3.2 Introduction ................................................................................................................................................. 74
3.3 Materials and Methods................................................................................................................................ 76 3.3.1 Strain and growth conditions .................................................................................................................... 76 3.3.2 Preparation of bacterial samples for calibration curves ............................................................................ 76 3.3.3 Production of aggregated cells .................................................................................................................. 77 3.3.4 Heat stress ................................................................................................................................................. 77 3.3.5 Rifampicin test ..........................................................................................................................................78 3.3.6 RNA isolation and reverse transcription reaction ..................................................................................... 78 3.3.7 Determination of RNA recovery rate using internal reference ................................................................. 79 3.3.8 Quantitative PCR amplification ................................................................................................................ 80 3.3.9 Statistical analyzes .................................................................................................................................... 81
Table of contents
I
3.4 Results........................................................................................................................................................... 81 3.4.1 Morphology of B. longum NCC2705........................................................................................................ 81 3.4.2 Calibration curves ..................................................................................................................................... 82 3.4.3 Heat test .................................................................................................................................................... 84 3.4.4 Rifampicin test ..........................................................................................................................................85 3.4.5 Viable cells quantification of cell aggregates............................................................................................ 86
3.5 Discussion ..................................................................................................................................................... 87
4 DEVELOPMENT OF A RAPID SCREENING PROTOCOL FOR SELEC TION OF BIFIDOBACTERIUM LONGUM STRAINS RESISTANT TO SPRAY DRYING AND POWDER STORAGE.................................................................................................... 90
4.1 Abstract ........................................................................................................................................................ 91
4.2 Introduction ................................................................................................................................................. 92
4.3 Material and Methods ................................................................................................................................. 94 4.3.1 Bacterial strains and culture conditions .................................................................................................... 94 4.3.2 Preparation of bacterial suspensions used for survival tests ..................................................................... 95 4.3.3 Preparation of strain mix........................................................................................................................... 95 4.3.4 Spray drying.............................................................................................................................................. 96 4.3.5 Storage test................................................................................................................................................ 96 4.3.6 Determination of water activity and moisture content .............................................................................. 97 4.3.7 Viable cell counts in spray dried powders ................................................................................................ 97 4.3.8 Identification of best surviving strains using RAPD................................................................................. 98
4.4 Results........................................................................................................................................................... 99 4.4.1 Strain survival and selection during the first selection batches................................................................. 99 4.4.2 Strain survival and selection during the second selection round............................................................. 104 4.4.3 Validation of selection procedures using single strain preparations ....................................................... 106
4.5 Discussion ................................................................................................................................................... 107
5 GENERAL CONCLUSIONS AND OUTLOOK.................... ................................. 111
5.1 General conclusions ................................................................................................................................... 111
5.2 Outlook and perspectives .......................................................................................................................... 114
6 APPENDIX ........................................................................................................... 117
6.1 Supplementary material from Chapter 4 ................................................................................................ 117
6.2 Genome expression analysis of B. longum NCC2705 produced during continuous culture with immobilized cells and batch culture ....................................................................................................................... 118
6.2.1 Background............................................................................................................................................. 118 6.2.2 Material and methods.............................................................................................................................. 118 6.2.3 Results and discussion ............................................................................................................................ 121
6.2.3.1 Comparison of cells from ICC with batch cultures ....................................................................... 121 6.2.3.2 Comparison between ICC1 and ICC2 and within ICC2................................................................ 122
Table of contents
I
7 REFERENCES..................................................................................................... 128
ACKNOWLEDGEMENTS ................................... ........................................................ 154
Abbreviations
V
Abbreviations
bp Base pair
cDNA Complementary DNA (reverse transcribed RNA)
CFU Colony forming units
cysS Gene coding for cysteinyl-tRNA synthetase
Da Dalton
DNA Deoxyribonucleic acid
GIT Gastrointestinal tract
HPLC High pressure liquid chromatography
IC Immobilized cells
LAB Lactic acid bacteria
mRNA Messenger ribonucleic acid
MRS Man Rogosa and Sharpe medium
MRSC MRS supplemented with 0.5% L-cysteine
NCC Nestlé Culture Collection
OD Optical density
ORF Open reading frame
PCR Polymerase chain reaction
purB Gene coding for adenylosuccinate lyase
qPCR Quantitative PCR (real-time PCR)
qRT-PCR Quantitative reverse transcription real-time PCR
RAPD Randomly amplified polymorphic DNA
RNA Ribonucleic acid
rpm Revolutions per minute
RT Reverse transcription
SFA Saturated fatty acid
USFA Unsaturated fatty acid
w/v Weight per volume
Summary
VI
Summary
Probiotics are defined as “live organisms which when administered in adequate amounts confer a
health benefit on the host”. The delivery of live cultures t adequate levels at the target site in the
host represents a challenging task in the production of probiotic food products. Probiotic bacteria,
generally isolated from the gastrointestinal tract of humans, are fastidious microorganisms and
susceptible to environmental stresses that occur during production, storage and transit through the
gastro intestinal tract. Different strategies have been proposed to produce probiotics with
enhanced stress resistance such as efficient screening methods to identify probiotic strains with
intrinsic resistance to environmental stresses, genetic engineering, the pre-treatment of bacterial
cells with sub-lethal stresses or the use of immobilized cells (IC) and continuous culture.
The dissertation consisted of three hypotheses: The first hypothesis claimed that B. longum
NCC2705 cells produced as IC and continuous culture exhibits improved tolerance to
environmental stresses which increases with culture age. The second hypothesis states that qRT-
PCR can be used to quantify viable B. longum NCC2705 cells independently of morphological
state. The third hypothesis is that spray drying can be used for preparation of bifidobacteria, and
that random amplified polymorphic DNA (RAPD) technology can be used to identify the best
survivor strain.
In a first approach, the probiotic strain Bifidobacterium longum NCC2705 was produced with
(IC) technology in combination with continuous culture (Chapter 2). The production of
immobilized B. longum NCC2705 was carried out in a two-stage fermentation system consisting
of a first small reactor (R1) containing cells entrapped in gellan/xanthan gel beads for high cell
productivity and a second larger reactor (R2) inoculated with bacteria produced from the first
reactor. Reactor volumes of 360 ml (R1) and 1800 ml (R2) were chosen to yield a shorter
Summary
VII
residence time in R1 than in R2, in order to obtain high cell productivity in R1 (high immobilized
cell concentration at high dilution rates (2.25 h-1)) followed by a final growth in R2 with longer
residence time at a low dilution rate (0.2 h-1). Two continuous cultures were produced for 20 days
with and without glucose limitation. Both fermentations exhibited formation of macroscopic cell
aggregates. Cells in the effluent from the second reactors showed enhanced tolerance to porcine
bile salts and to aminoglycosidic antibiotics which were associated with modifications in the cell
membrane composition. However, there was no significant difference in terms of culture age and
heat resistance when compared to cells from free cell batch cultures.
Formation of large cell aggregates resulted in an underestimation of viable cell counts when
measured by the traditional plate counting method. This complicated the characterization of
physiological properties. Hence, we developed a new method to enumerate viable B. longum
NCC2705 which is not influenced by the morphological state of the cells (Chapter 3). The new
method was based on real time PCR technology and enabled the accurate quantification of cells
in the aggregated state. To achieve this, we measured the level of a constitutively expressed
housekeeping gene and calculated viable cell concentration using calibration curves from 7.0 ×
103 to 4.7 × 108 CFU/ml. This quantification of IC samples containing aggregates resulted in up
to 125-fold higher viable cell counts compared to traditional plate counts. The method was
further used for analyzes of tolerance to bile salts of IC samples (see above).
In the last part of the thesis, a different strategy was evaluated to produce robust bifidobacteria.
Instead of improving stress tolerance of one specific B. longum strain, we developed a high
throughput screening method to identify bifidobacterial strains with intrinsic resistance to stresses
occurring during spray drying and storage (Chapter 4). In this approach, twenty-two B. longum
strains were mixed in separate batches and subjected to spray drying and storage experiments
during two selection rounds. Unique identification of the twenty-two B. longum strains in the
Summary
VIII
mixed bacterial preparation was achieved using molecular-based methods. Within the mixed
strain preparation, a robust strain was identified and stability was confirmed by performing single
strain spray drying and storage tests. The selected strain exhibited a 250-fold better survival after
60 days of storage than the control strain. The method allowed faster screening of resistance of B.
longum strains to spray drying and storage compared to traditional screening performed on
individual strains.
In this thesis, IC and continuous cultures used to produce probiotic cells with improved stress
resistance resulted in the formation of large cell aggregates. Auto-aggregation is a known
phenotype in LAB and bifidobacteria and was associated with enhanced ability to adhere to host
epithelial cells. Therefore it would be of interest to test adhesion ability of the cells produced with
IC and continuous culture. Additionally, we designed and experimentally validated a high
throughput screening method to select for probiotic strains with intrinsic resistance to stresses
occurring during spray drying and storage. The same approach could be applied for selection of
robust strains in response to other stress responses.
Zusammenfassung
IX
Zusammenfassung
Die gesundheitsfördernde Wirkung probiotischer Bakterien u.a. Wiederherstellung und
Stabilisierung einer ausgewogenen Darmflora, Verdrängung von Krankheitserregern und
unerwünschten Bakterien im Darm, Beeinflussung des Immunsystems und Verbesserung der
natürlichen Abwehrkräfte des Körpers, Senkung der Konzentration gesundheitsschädigender
Stoffwechselprodukte und krebsfördernder Enzyme im Dickdarm und Verhinderung bzw.
Verkürzung von Durchfallerkrankungen, wird nach momentanem Wissenstand hauptsächlich mit
einer bestimmter Anzahl lebender Bakterien erzielt. Einzelne Stämme der Gattung
Bifidobacterium sind bekannt für ihre probiotische Wirkung und Produkte, welche diese Stämme
beinhalten, sind im Handel erhältlich. Da sich Bifidobakterien natürlicherweise im Darmtrakt von
Säugetieren (u.a.) befinden, sind sie anaerobe Mikroorganismen und sensibel gegenüber
Sauerstoff und anderen Umwelteinflüssen, welche bei der Produktion, Lagerung und Verzehr von
probiotischen Produkten auftreten können. In dieser Hinsicht erfreut sich die Entwicklung von
neuen Technologien, welche die Robustheit von probiotischen Bakterien gegenüber diesen
Umwelteinflüssen stärken, einer grossen Nachfrage von Produzenten probiotischer
Lebensmitteln.
Diese Dissertation befasst sich in erster Linie mit der Entwicklung neuer Strategien, welche die
Robustheit des probiotischen Modelbakteriums Bifidobacterium longum NCC2705 gegenüber
Umweltfaktoren stärken sollen. Vorangehende Studien haben gezeigt, dass kontinuierliche
Zellkulturen mit immobilisierten Bakterien in Polysaccharidgel-Kügelchen physiologische
Änderungen hervorrufen können, welche sich positiv auf die Toleranz gegenüber verschiedenen
Umweltfaktoren auswirken können. Diese Änderungen werden den speziellen Bedingungen,
welche innerhalb der Polysaccharid-Matrix und an der Oberfläche dieser Kügelchen
Zusammenfassung
X
vorherrschen, zugeschrieben. In dieser Arbeit, wurde der Stamm B. longum NCC2705 in
Polysaccharidgel-Kügelchen bestehend aus einer Mischung aus den Polysacchariden Gellan und
Xanthan immobilisiert und über eine Zeitdauer von 20 Tagen kontinuierlich kultiviert (Kapitel
2). Hierzu wurde ein bewährtes System verwendet, welches aus zwei Bioreaktoren verschiedener
Grössen besteht, die miteinander verbunden sind. Im ersten, kleineren Reaktor (360 ml) befinden
sich die immobilisierten Zellen, welche kontinuierlich mit einer hohen Durchflussrate (2.25 h-1)
frischer Nährlösung versorgt werden. Der zweite grössere Reaktor (1800 ml) ist in Serie mit dem
ersten Reaktor geschaltet und wird fortwährend mit Zellen angeimpft, welche im ersten Reaktor
produziert wurden. Das grössere Volumen des zweiten Reaktors bewirkt eine kleinere
Durchflussrate (0.2 h-1) und gewährt den Zellen eine längere Verweilzeit, so dass sie ihr
Wachstum im zweiten Reaktor abschliessen können. Proben wurden in zeitlichen Abständen dem
System entnommen, verschiedene Parameter getestet und mit einer Standard Referenz-Kultur
verglichen. Es konnte eine signifikant erhöhte Resistenz gegenüber Gallensalzen und gewissen
Antibiotika, ausgemacht werden, welche unabhängig vom Zeitpunkt der Probennahme waren und
welche dem veränderten Aufbau der Zellmembran bzw. deren Fettsäurestruktur zugeschrieben
wird. Diese Veränderung der Membranstruktur ist höchstwahrscheinlich auch der Grund, warum
sich nach einer gewissen Fermentationszeit makroskopische Zellklumpen von 0.5 – 1 mm
Durchmesser im Überstand beider Bioreaktoren gebildet haben. Verklumpung von Bakterien
kann durchaus interessante Eigenschaften für probiotische Bakterien bergen (physischer Schutz,
erhöhte Adhäsion an Epithelialzellen des Wirtes), führt jedoch zur Unterschätzung der Lebend-
Zell-Konzentrationen mit der traditionellen Methode des Verdünnens und Ausplattierens. Eine
physiologische Charakterisierung der verklumpten Bifidobakterien basierend auf dieser Methode
ist folglich erschwert oder gar unmöglich.
Zusammenfassung
XI
Um dennoch eine physiologische Analyse der immobilisierten Zellen durchzuführen, wurde im
zweiten Teil der Dissertation eine neue Methode entwickelt, um die Anzahle lebender B. longum
NCC2705 Zellen unabhängig von ihrer Morphologie bestimmen zu können (Kapitel 3). Die
Methode nutzt quantitative „echt Zeit“ Polymerasen Kettenreaktion (engl. qRT PCR) als ein
Mittel, um die Kopienzahl eines sogenannten Haushaltsgens zu messen, um anhand einer
Standardkurve, welche Zellkonzentrationen über eine Spannweite von 7.0 × 103 bis 4.7 × 108
Kolonien formende Einheiten pro ml abdeckt, auf die ursprüngliche Zellkonzentration in der
Zellprobe zu schliessen. Die Kopien eines Haushaltsgens wurden ausgewählt, weil zum einen die
Anzahl der Kopien pro Zelle konstant ist, unabhängig von der Umgebung und des
physiologischen Zustandes der Zelle, und zum andern weil die Kopien eines Gens (mRNA)
instabil sind und kurz nach der Bildung wieder abgebaut werden. Ein konstantes Niveau der
Kopienanzahl eines bestimmten Gens kann deshalb nur durch konstante Erneuerung und folglich
nur von metabolisch aktiven, also lebenden Zellen erreicht werden. Eine gewisse Anzahl
gemessener Kopien eines Haushaltgens kann deshalb mit einer bestimmten Anzahl lebender
Bakterien gleichgesetzt werden, was z.B. bei der stabileren DNA nicht der Fall ist, welche auch
Stunden oder gar nach Tage nach dem Zelltod nachgewiesen werden kann.
Die Methode wurde experimentell validiert und zur Konzentrationsbestimmung von Zellen,
welche in makroskopischen Zellklumpen vom der kontinuierlichen Kultur mit immobilisierten
Zellen stammen verwendet. Die erhaltenen Konzentrationen lebender Bakterien wurden dann mit
den Werten, welche mit der traditionellen Methode des Ausplattierens erhalten wurden,
verglichen. Das Resultat offenbarte, dass Lebend-Zell-Konzentrationen bestimmt mittels qRT-
PCR bis zu 125 Mal höher waren als jene, welche mittels Ausplattierens erhalten wurden. Mit
Hilfe dieser Methode war es nun möglich, die reale Konzentration von lebenden Zellen in
Zusammenfassung
XII
Zellklumpen, nachdem sie mit Gallensalzen oder Hitze behandelt wurden, zu berechnen und so
die Toleranz von diesen Zellen gegenüber Gallensalzen oder Hitze zu testen (siehe oben).
In der letzten Phase der Dissertation wurde ein neuer Ansatz gewählt. Anstatt robuste
Bifidobakterien zu produzieren, entwickelten wir ein effizientes Auswahlprüfverfahren, welches
die Identifizierung von B. longum Stämmen ermöglicht, die eine natürliche Toleranz gegenüber
Stressbedingungen aufweisen, die während der Sprühtrocknung und Lagerungskonditionen
auftreten können. Um zeitraubende Wiederholungen der Lagerungstests zu vermeiden, wurden
mehrere Stämme gleichzeitig getestet. Als notwendige Bedingung für ein solches simultanes
Verfahren gilt die Unterscheidbarkeit der einzelnen Stämme untereinander. Hierzu wurde ein
Verfahren verwendet, welches sich auf Polymerasen Kettenreaktion (engl. PCR) stützt und die
polymorphe DNA Sequenz von Stämmen derselben Unterklasse ausnutzt. Die Polymerasen
Kettenreaktion mit drei „Universalprimern“ vervielfältigt spezifische Sequenzen der bakteriellen
DNA (engl. RAPD). Diese spezifische Vervielfältigung variiert von Stamm zu Stamm und
ermöglicht die Fertigung von Stamm-spezifischen molekularen Vervielfältigungsmustern mittels
Gel-Elektrophorese, dem sogenannten „molekularen Fingerabdruck“. Die so identifizierten 22 B.
longum Stämme wurden in Ansätzen von 7 – 8 Stämmen sprühgetrocknet und das resultierende
Pulver mit den lebenden Bakterien einem Lagerungstest bei erhöhten Temperaturen unterzogen
und wöchentlich ausplattiert, um die Lebend-Zellzahl im Pulver zu bestimmen. Die DNA der
verbleibenden Kolonien (bestehend aus unzähligen, genetisch identischen Klonen) von den
Nährlösung-Agar-Platten (100 – 200 Kolonien) wurde extrahiert und mit RAPD Technologie
identifiziert. Mit diesem Verfahren konnte am Ende des Lagerungstests derjenige Stamm
(NCC572) bestimmt werden, von welchem die meisten Kolonien übrig geblieben sind, respektive
welcher die Bedingungen von Sprühtrocknung und beschleunigter Lagerungstest am besten
überlebt hat. Die Methode wurde experimentell verifiziert, indem Stamm NCC572 einzeln
Zusammenfassung
XIII
angezüchtet, sprühgetrocknet und gelagert und sein die Überlebensrate mit einem Kontrollstamm
(NCC2916) verglichen wurde. Tatsächlich stellte sich heraus, dass Stamm NCC572 nach 60
Tagen Lagerung in Pulverform rund 250 Mal besser überlebte als der Kontrollstamm NCC2916.
Die entwickelte Methode erlaubt die zeit- und kosteneffiziente Anwendung eines Auswahl-
prüfverfahrens, mit welcher probiotische Bakterienstämme mit inhärenter Stresstoleranz
identifiziert werden können.
Die Methoden, welche in der vorliegenden Dissertation entwickelt wurden, können mannigfaltig
angewendet werden. So kann z.B. das beschleunigte Auswahlprüfverfahren auf beliebig andere
Konditionen angewendet werden. Ferner wäre es interessant, die verklumpten Bifidobakterien
aus der kontinnuierlichen Kultur mit immobilisierten Zellen, auf ihre Adhäsionsfähigkeit an
menschlichen Epithelialzellen zu testen.
Chapter 1 - Introduction
- 1 -
1. Introduction
1.1 Human gastrointestinal tract microbiota
The human gastro intestinal tract (GIT) is a complex system composed of different
compartments: oral cavity, stomach, small intestine (duodenum, jejunum and ileum), and large
intestine (caecum, colon and rectum) exhibiting very different environmental conditions. The
intestinal epithelium has a combined surface area of 400 m2 and provides ecological niches for
the three domains of life: prokarya, archaea and eukarya. The GIT is one of the most dense
microbial ecosystems on earth (Whitman et al., 1998), and its population number of up to 1014
cells exceeds that of the human cells by a factor of 10 (Backhed et al., 2004). Despite this
tremendous number, the gut microbiota forms an exclusive community underlying a strict
selection pressure. Gut microbes must be able to react rapidly to environmental changes, survive
the harsh conditions during gastrointestinal transfer and during re-colonization of a new host,
evade bacteriophages, and be capable of growth under GIT conditions to avoid disappearance
(Ley et al., 2006). Gut inhabitants may also harbor a collection of enzymes for nutrient
degradation and contain a cell-surface composition which might enable attachment to host
epithelial cells (Ley et al., 2006).
The microbial colonization of the GIT varies with age and is characterized by temporary changes.
The sterile GIT of newborns is colonized mainly by enterobacteria, bifidobacteria, and
streptococci. The percental distribution depends on the environment, mode of birth, the diet and
possibly on the host genetic background (Edwards and Parret 2002; Vaughan et al., 2002; Fanaro
et al., 2003; Zoetendal et al., 2004). At older age, the microbial diversity increases to more than
1000 different subspecies of bifidobacteria, enterobacteria, streptococci, bacteroides, clostridiae
and cyanobacteria. However, a satisfying general representation of the ecosystem has not yet
Chapter 1 - Introduction
- 2 -
been established because only a limited number of human gastrointestinal microflora has been
analyzed and each subject harbors a unique microbial community (Rajilic-Stojanovic et al.,
2007). The diversity of microflora develops with host maturation and gets more unstable with
increasing age (Harmsen et al., 2000, Egert et al., 2006). Diversity of microflora is generally host
specific and depends on factors such as digestion characteristics, gender, age and genotype, as
well as on environmental factors like nutritional diet, ingestion of microorganisms and drugs
(Zoetendal et al., 2006).
Host and gut microflora live in a symbiotic relationship; the host provides the microflora a
nutrition-rich and protective habitat, while the microorganisms ferment non-digestible dietary
substrates and indigenous mucus produced by epithelial cells, resulting in the production of short
chain fatty acids (SCFA) which are absorbed by the host (Shanahan, 2002). The generated plus in
dietary energy from fibre fermentation favors the adsorption of ions (Mg2+, Ca2+, Fe2+/3+) in the
caecum (Guarner et al., 2006). Additionally, the commensal flora provides a barrier against
colonization of exogenous and pathogenic bacteria by outcompeting them for nutrients and
binding sites (Guarner et al., 2006) and by producing antimicrobials such as bacteriocins
inhibiting (a.o.) growth of pathogens (Shanahan, 2002). Besides protection, the gut microbes are
involved in developmental and immunological processes of the host by modulating immune
responses and influencing proliferation and differentiation of epithelial cells (Hooper et al., 2001;
Falk et al., 1998, Guarner and Magdalena, 2003). Endogenous bacteria are also involved in the
production of vitamin B12, vitamin K, biotin, folic acid, in the synthesis of amino acids from
ammonium and urea, and in the inactivation of dietary carcinogens (Wollowski et al., 2001,
Hooper et al., 2002). However, the complex microflora with its huge diversity of different
species and subspecies is a major obstacle in the prediction of specific bacterial function within
the GIT.
Chapter 1 - Introduction
- 3 -
The development of molecular techniques allowed the exploration of the human microflora and
revealed new insights during the last decade. In the future, advanced molecular tools will confer
on understanding of how single or groups of bacteria interact with the host, thereby providing
new leads for additional investigations. One of the key players in the transient microflora of
newborns and also for the development of probiotic products is the genus Bifidobacterium,
displaying an appearance ranging from 40 – 90 % of total bacteria counts in the intestine of
infants (Harmsen et al., 2000).
1.2 Bifidobacterium
1.2.1 General characteristics
Bifidobacteria were first isolated from faeces of breast-fed infants by Tissier in 1906. They are
among the first species to colonize the sterile GIT of newborns. In adults, Bifidobacterium is the
third most common genus, after bacteroides and eubacteria (Bezkorovainy, 2001). Bifidobacteria
are Gram-positive, heterofermentative, non-motile, non-sporulating bacteria. They vary in size
and exhibit rods of various shapes, often in ‘V’ or ‘Y’ patterns (Latin bifidus: cleft, divided).
Although considered strictly anaerobic, some strains exhibited poor growth in presence of
oxygen (Simpson et al., 2005). Most strains isolated from humans grow optimally at
temperatures of 36 – 38 °C. Bifidobacteria are often included in the group of LAB due to their
metabolic capacities and the sharing of ecological niches (Klein et al., 1998). However,
phylogenetically the genus Bifidobacterium belongs to the family of actinomycetaceae, a family
including Corynebacterium, Mycobacterium, and Streptomyces. Actinomycetaceae share the
common feature of having a high G+C content ranging from 42 to 67 %, which is sufficiently
higher to differentiate them from lactic acid bacteria (Biavati et al., 2000). The genus
Chapter 1 - Introduction
- 4 -
Bifidobacterium is relatively small with ca. 32 identified species to date, mainly of human or
animal origin and has a low level of phylogenetic and genomic diversity (Ventura et al., 2006).
The availability of genome sequences of microorganisms from the human GIT and progress in
molecular tools will lead to a better understanding of the behavior and survival of these bacteria
in the GIT. Up to today, the two B. longum strains NCC2705 (Schell et al., 2002) and DJO10A
and two strains of B. longum subsp. infantis (B. longum subsp. infantis ATCC55813 and
ATCC15697) have been sequenced (Lee et al., 2008; Sela et al., 2008).
1.2.2 Metabolism of Bifidobacteria
Bifidobacteria can utilize a large variety of carbohydrate sources. Utilization of complex
carbohydrates from dietary compounds confers a competitive advantage for bifidobacteria in the
lower GIT (Hooper et al., 1999; Klijn et al., 2005). Bifidobacteria metabolize carbohydrates
exclusively and characteristically in the so-called “bifido-shunt” via the key enzyme fructose-6-P
phosphoketolase (F6PPK). This enzyme converts hexoses mainly to lactic and acetic acid
(Scardovi and Tovatelli, 1965). However, sugar availability and consumption rate affects
production of other metabolites, such as formic acid and ethanol, thereby altering the molecular
ratio of these acids (van der Meulen et al., 2006).
Progressive genome sequencing technologies have disclosed an insight in encoded enzymes
involved in metabolic pathways. Genome annotations of bifidobacteria revealed that 8 % of the
genes code for enzymes involved in the carbohydrate metabolism which exceeds the percentage
of other GIT inhabitants such as Enterococcus faecium and Echerichia coli or allochthonous
Lactococcus lactis (Ventura et al., 2007). Sequence in combination with physiological analyzes
confirmed that bifidobacteria utilize a broad range of indigestible polysaccharides like hog gastric
Chapter 1 - Introduction
- 5 -
mucin, pectin, plant oligosaccharides and fructo-oligosaccharides (Schell et al., 2002;
MacConaill et al., 2003, Backhed et al., 2004, Matsuki et al., 2004; Parche et al., 2007). Taken
together, bifidobacteria have a unique potential to metabolize indigestible components, which
helps to prolong residence time in the competitive environment of the human gut.
1.2.3 Role of Bifidobacteria in the GIT
Each compartment of the GIT is a habitat for a specific bacterial community. Bifidobacteria can
be found all along the GIT and the main species present in humans are B. adolescentis, B.
bifidum, B. infantis, B. breve and B. longum (Crociani et al., 1996; Marteau et al., 2001; Nielsen
et al., 2003; Champagne et al., 2005). The preferred location of bifidobacteria inside the GIT
cannot be easily determined since numbers and species vary between individuals (Nielsen et al.,
2003). Furthermore, the majority of studies on human microbiota comes from analyzes of fecal
contents and are thus limited in reflecting the distribution of species in the different habitats of
the GIT (Tannock 1999; Marteau et al., 2001). Bifidobacteria can be autochthonous (found where
they grew) or allochthonous (found at places other than where they grew). The first colonizers of
the sterile GIT of newborns are mostly strains that reside autochthonously after colonization
(Vaughan et al., 2005).
In adults, species-diversity of microbiota increases and percentage of bifidobacteria decreases
with age depending on the health of the individuals (Hopkins and MacFarlane, 2002; Vaughan et
al., 2002; Hebtuerne et al., 2003; Woodmansey et al., 2004). This decrease was associated with
the fact that allochthonous bifidobacteria do not naturally occur in food bifidobacteria but
remains subject of discussions. However, bifidobacteria are added to food for their claimed health
benefits and their ingestion results in a temporary (elevated) appearance in the GIT with average
Chapter 1 - Introduction
- 6 -
retention time believed to be less than a week (Kullen et al., 1996, Fukushima et al., 1998), while
other studies claim it to be more than one week (Alander et al., 2001, Ouwehand et al., 2004,
Bartosch et al., 2005, Su et al., 2005). Ouwehand and coworkers summarized in their review in
2004 that several factors to affect the residence time of allochtonous bacteria such as duration of
the consumption, amounts of the administered doses, and the methods of detection used.
1.2.4 Transcriptomics of Bifidobacterium longum NCC2705
The transcriptome is defined as the set of all messenger RNA (mRNA) molecules, also called
"transcripts," produced in a population of cells. Since it includes all mRNA transcripts in the cell,
the transcriptome reflects the genes that are being actively expressed. Therefore, this relatively
new molecular tool can give insights into global cellular mechanisms under different
environmental and/or growing conditions. However, transcriptome analyzes require the
availability of DNA based arrays, thereby complete genome sequence of the tested organism. The
strain B. longum NCC2705 was the first completely sequenced genome within the genus
Bifidobacterium (Schell et al., 2002) and a microarray covering its whole genome is available at
Nestlé Research Center. Transcriptomic analyzes of this strain have been used to characterize
metabolism and growth on various sugars (Parche et al., 2007) as well as cellular response to
environmental stresses. The exposure of B. longum NCC 2705 cells to oxygen and starvation
induced expression of genes that are known to participate to the common stress response in other
bacteria, whereas other up-regulated genes still remain unidentified (Klijn et al., 2005). In 2007,
Rezzonico and co-workers tested the gene expression patterns of B. longum NCC 2705 after
exposure to different heat periods and reported a 46 % of differentially expressed genes,
including common bacterial stress-responses such as the induction of chaperones and the down-
Chapter 1 - Introduction
- 7 -
regulation of the translation, cell division and chromosome-partitioning machineries. In a recent
study, the effect of subinhibitory concentrations of bile on the expression levels of three different
bifidobacterial strains including B. longum NCC2705 were tested and lead to the identification of
a new multidrug resistance transporter in strain NCC2705 (BL0902) conferring increased
resistance to bile salts (Gueimonde et al., 2009).
These results emphasize the importance of transcriptome analyzes which enable new insights into
general cellular mechanisms and may help to design probiotic strains with enhanced functional
properties (Corcoran et al., 2008). Moreover, transcriptomics may also support the analysis of
complex physiological or morphological changes of bacteria as a result of changing
environments.
1.3 Probiotic concept
In 1907 the Russian scientist Elie Metchnikoff was the first to mention the beneficial effects of
bacteria in his book “The prolongation of life” and presented the concept of “replacement of
harmful microbes by useful microbes”. In 1906, Henry Tissier claimed that bifidobacteria are the
dominant species in the microflora of breast-fed infants and recommended the administration of
bifidobacteria to out-compete the putrefactive bacteria in infants suffering diarrhea. The term
“probiotic” (Greek: pro life) was introduced by Lilly and Stillwell (1965) and after several
redefinitions, the Joint Food and Agricultural Organization and World Health Organization
(FAO/WHO) working group defined the term “probiotic” in 2001 as “live microorganisms that
when administered in adequate amounts confer a health benefit on the host”. Probiotic bacteria
exert beneficial effects like inhibition of colonization of potential pathogens of the gut,
immunomodulation, alleviation of constipation, reduction of problems associated with
Chapter 1 - Introduction
- 8 -
assimilation of lactose, and alleviation of antibiotic-associated diarrhea (reviewed by DeVrese
and Schrezenmeir, 2008; see “1.4 Health promoting properties related to probiotics”).
Growing scientific interest in probiotic bacteria is reflected in the increasing number of scientific
publications on the subject (Figure 1).
Figure 1 Number of publications obtained with the key word “probiotic” in the Pubmed
database as a function of time. Dashed bar is an extrapolated value based on number of
publication obtained after 6 months in 2009.
Nowadays, the prospering probiotic food market in Western Europe generates more than 1.4
billion Euros, one billion of which is exclusively from yogurt and desserts. Annual growth of
sales is forecasted to be 7 – 8 % over the next five years (Saxelin, 2008). Besides dairy products,
probiotics are also administered in fruit juices, berry soups, ice cream, soy- and cereal-based
fermented products as well as cheese. Probiotic foods are manufactured either by adding the
desired strains simultaneously with the starter culture in the fermentation tank or by running the
fermentation in a separate reaction and adding the probiotics afterwards in frozen liquid or
formulated state (Champagne et al., 2005). The increase of probiotic consumption raises a
Chapter 1 - Introduction
- 9 -
demand by the food industry for more scientific research to develop new methods to improve
quality, safety, and functionality of probiotics.
1.4 Health promoting effects related to probiotics
Potential probiotic strains are screened for their health promoting qualities based on the following
established selection criteria (reviewed by DeVrese and Schrezenmeir, 2008)
• Origin from the intestinal tract of healthy persons, as such microorganisms are regarded
safe for humans and best adapted to the ecosystem of the gut
• Safe for humans, free of pathogenic and toxic effects
• Tolerance and resistance to gastrointestinal conditions and digestive enzymes enabling the
survival during the passage through stomach and upper intestinal tract eventually resulting
in health-promoting effects in the bowel
• Possess properties associated with a (positive) influence on the intestinal flora, like adhesion
to intestinal epithelial cells, survival and reproducing capacity in the human large intestine,
or production of antimicrobial substances
The most common strains used in probiotic food products are either lactobacilli or bifidobacteria.
Other strains exerting probiotic effects belong to the genus of Propionibacterium which is not
typical for human microbiota but strains of this genus were shown to produce vitamin B12, and
bacteriocins in addition to organic acetic and propionic acids from sugar metabolism.
Furthermore, yeasts from the genus of Saccharomyces have been show to exhibit antagonistic
effects against enteric pathogens (Klein et al., 1993) or are used as preventive and therapeutic
agents for the treatment of a variety of diarrheal disease.
Chapter 1 - Introduction
- 10 -
The beneficial effects of strains of bifidobacteria have been attested in vitro in numerous studies.
However, probiotic properties have to be confirmed also in vivo before claiming them as
probiotic strains (Reid, 2008) and therefore a number of clinical studies on the therapeutic effects
of supplementation of bifidobacterial strains in mouse models and humans have been conducted
in the recent years as illustrated in Table 1.
Table 1 A selection of studies with beneficial effects on humans or mice of products containing bifidobacteria Strain(s) Benefit Hosta Reference B. longum B6 and ATCC 15708 Reduction of lactose intolerance Humans (15) Jiang et al., 1996
B. infantis Reduced risk of necrotizing enterocolitis (NEC) Rats, neonates Caplan et al., 1999
B. bifidum Protection against S. enteridis ssp. typhimurium Mouse, healthy and gnotobioic Silva et al.,1999
B. lactis HN019 Enhance natural immunity in healthy elderly subjects Human (25) Arunachalam et al., 2000
B. lactis Bb-12 Decrease of extent and severity of atopic eczema Human, infants (27) Isolauri et al.. 2000
Strain mix: 4 Lactobacillus; 3 Bifidobacterium and 1 Streptococcus
Prevent flare-ups of chronic pouchitis Human (40) Gionchetti et al., 2003
B. lactis Bb 12 Protective effect against acute diarrhea Humans, children (90) Choraqui et al., 2004
B. pseudocatenulatum DSM 20439, B. breve
Inhibit Shiga toxin-production of E. coli (STEC) 0157:H7
Mice Ashara et al., 2004
Strain mix: Lb. confusus; Lb. fermentum, Lb. plantarum; B. infantis PL9506
Suppression of Th2 cytokines during antigen sensitization
Mice Lee et al., 2004
Strain mix: Lb. gasseri PA 16/8; B. longum SP 07/3; B. bifidum MF 20/5
Reduced severity and duration of cold episodes Humans (479) de Vrese et al., 2005
Strain mix: B. infantis, S. thermophilus; B. bifidus
Reduction of incidences and severity of necrotizing enterocolitis (NEC)
Human, neonates (145) Bin-Nun et al., 2005
B infantis 35624 Alleviates symptoms of irritable bowel syndrome (IBS)
Humans with IBS (77) O'Mahony et al., 2005
B. lactis (Bb-12)
Fewer and shorter episodes of diarrhea Humans (200) Weizman et al., 2005
B. thermacidophilum RBL 71
Reduced severity of E. coli O157:H7 infection Mice Gagnon et al., 2006
B. lactis HN019
Enhanced aspects in cellular immunity in the elderly Humans, elderly (30) Ahmed et al., 2007
Strain mix: Lb.acidophilus NCFM, B. animalis ssp. lactis Bi-07
Reduced fever, rhinorrhea, and cough incidence and duration and antibiotic prescription incidence
Humans, children (326) Leyer et al., 2009
Strain mix: Lb. rhamnosus GG, B. lactis Bb-12
Reduced risk of early acute otitis media and recurrent respiratory infections during the first year of life
Humans, children (81) Rautava et al., 2009
a in brackets: number of persons tested
Chapter 1 - Introduction
- 12 -
The particular molecular mechanisms underlying most of the probiotic effects of bifidobacteria
are still not completely understood (Leahy et al., 2005). Bifidobacterial strains exhibited
antimicrobial activities against other gut inhabitants which are attributed to production of
bacteriocins or organic acids that lower the colonic pH thereby inhibiting acid sensitive
pathogenic bacteria (Klijn et al., 2005; Cheikhyoussef et al., 2008). Bifidobacteria can modulate
the immune response by regulating the production of anti- and proinflammatory cytokines in
peripheral blood mononuclear cells of humans (Medina et al., 2007). Strains of bifidobacteria
were shown to produce linolenic acid in vitro (Coakley et al., 2009) or vitamins (Tamine et al.,
1995), both substances that can contribute a health benefit to the host. Genome analyzes revealed
the complete pathway for production of folic acid, thiamine, and nicotinate in B. longum
NCC2705 (Schell et al., 2002). Metabolic activity may moreover reduce serum cholesterol level
through bile salt hydrolase activity (De Smet et al., 1998).
1.5 Potential risks associated with probiotics
There are three theoretical concerns regarding the safety of probiotics (Snydman, 2008).
1. They may lead to disease, such as bacteremia or endocarditis.
2. Toxic or metabolic effects on the gastrointestinal tract
3. Transfer of antibiotic resistance in the gastrointestinal flora
In rare cases, Bifidobacterium species were shown to be involved in human dental caries,
pulmonary infections, bacteremia, abscesses, and blood stream infections which were not related
to consumption of probiotic products (Green, 1978; Gasser, 1994; Saarela et al., 2002; Modesto
et al., 2006). A tetracycline resistance gene (tet(W)) was reported on the chromosome of B. lactis
DSM 10140 (Kastner et al., 2006) which is problematic due to possible horizontal gene transfer
Chapter 1 - Introduction
- 13 -
to pathogens in the gut. In patients with short small bowel syndrome, deconjugated bile acid
metabolites produced by bifidobacteria can accumulate leading to malabsorption of food nutrients
(Snydman, 2008). However, bifidobacteria and lactobacilli (with exception of L. rhamnosus)
used in probiotic food products are “generally accepted as safe” (GRAS) by the Food and Drug
Administration of the USA and classified as “class I” (absolutely safe) in Germany. Furthermore,
there is no evidence for a higher risk of bacteremia or fungemia due to the ingestion of probiotic
products in comparison with conventional fermented food products (Snydman, 2008; Salminen et
al., 2002).
1.6 Technological requirements of probiotic bacteri a
In order to provide a therapeutic effect to the consumer probiotic food products should contain a
critical concentration of viable bacteria per gram of product (Tamime et al., 1995). Studies with
non-viable cells or cell-components of probiotic strains also showed positive effects on health or
immunomodulatory effect (Salminen et al., 1999; Lammers et al., 2003). However, the average
recommended level of viable probiotic bacteria is suggested between 106 to 107 CFU per gram of
product at the time of consumption (Talkamar et al., 2004; Ishibashi and Shimamura, 1993).
Additionally, the probiotic food product should be regularly consumed in sufficient quantities to
deliver the relevant “dose” of live bacteria to the gut. Nevertheless, increasing knowledge of
strain survival during GIT transit and strain specific therapeutic effect will probably modify these
numbers in the near future (Champagne et al., 2005). The achievement of sufficient viable
probiotic cells at the time of consumption and at target site remains a challenge for the food
industry.
Chapter 1 - Introduction
- 14 -
The preparation of a critical level of viable cells in probiotic food products and activity at the
target site depends on conditions during fermentation, downstream processing, storage and
during gastrointestinal transit (Figure 2). Exposure of microorganisms to different stresses
decreases bacterial viability and causes large fluctuations in viable cell counts of probiotic
products. In fact, numerous probiotic food products failed to meet the minimal level of viable
bacteria in probiotic food products (Masco et al., 2005, Huys et al., 2006). The selection criteria
of probiotic candidate strains are thus mainly focused on their technological properties, which
means that a lot of strains with promising health properties are probably missed (Lacroix and
Yildirim, 2007).
Figure 2 Main factors affecting the viability of probiotics from production to the
gastrointestinal tract (Lacroix and Yildirim, 2007).
Economic reasons raise the demand of food industries for new technologies that enable high
viable cell yield at large scale production. Major drawbacks in propagation and scale-up process
are sensitivity to oxygen and acidic conditions of numerous strains of intestinal origin.
Technologies which ensure probiotic activity and stability in food products are thus of highest
Chapter 1 - Introduction
- 15 -
interest. Improved survival of strains during manufacturing may enable production of
technologically less favorable candidate strains but with more relevant probiotic properties,
consequently resulting in higher product efficacy. Therefore, the selection of adequate strains and
the improvement of technologies for production of probiotics is crucial in the development of
new probiotic food products.
1.7 Technologies to improve cell viability and yie lds
1.7.1 Production technologies
1.7.1.1 Optimization of growth medium
Probiotic strains are preferentially added to dairy products after fermentation because the fast
growth and acidification rate of typical starter cultures are disadvantageous for probiotic culture
development and maintenance (Champagne et al., 2005). Probiotics are delivered as frozen liquid
or as dry state cultures produced with freeze- or spray drying technology (see “Stabilization of
probiotics”). The cultivation conditions prior to freezing or drying have an impact on growth,
culture stability and activity as well as on drying and subsequent storage (Reilly and Gilliland
1999; Desmond et al., 2002; Carvalho et al., 2004). Modulation of the culture conditions can
therefore improve stability and activity of probiotics in food products.
Traditional large scale production of cells is performed in large batch fermentations. Approaches
to enhance yields in biomass and enhance cell stability have ranged from designer growth
medium to alternative fermentation technologies. In research laboratories, expensive media such
as MRS broth (de Man, Rogosa, and Sharpe, 1960) is commonly used for cultivation of LAB or
bifidobacteria. Biomass production from ultra-filtered skim milk with different protein
concentrations or supplementation of milk with nitrogenous substrates such as whey and casein
Chapter 1 - Introduction
- 16 -
fractions from human or cow milk resulted in higher cell counts of bifidobacterial strains
compared to growth in skim milk only (Ventling and Mistry, 1993; Petschow and Talbott, 1990).
In contrast, bifidobacteria grown in soymilk or animal-product-free vegetable medium (based on
soy peptone, glucose and yeast extract (YE)) displayed lower yields or lower viability during
storage conditions due to the low buffer capacity of vegetable medium (Heenan et al., 2002,
Shimakava et al., 2003). Recently, a study showed a significantly increased yield of cell after 24
h growth of a B. animalis strains or a B. longum strain in modified vegetable medium compared
to standard MRS medium (Mättö et al., 2006). Growth medium for bifidobacteria is generally
supplemented with redox-reducing compounds such as cysteine to improve growth (Doleyres and
Lacroix, 2005). However, the growth promoting properties of such supplements lose the effect
when the disulfide bonds are reduced, for example by performic acid oxidation or reduction-
alkylation (Poch and Bezkorovainy, 1991; Ibrahim et al., 1994). Summarized, higher biomass
yields of bifidobacteria can be achieved by modulation of the medium components and opens
possibilities to develop low cost production methods.
1.7.1.2 Stress adaptation
The capacity of microorganisms to adapt to adverse environments is associated with the
expression of stress proteins and the development of cross resistances to numerous stresses (Ross
et al., 2005; Lindner et al., 2006). Besides medium composition, growth conditions like pH,
fermentation time or addition of protectants can induce cell responses that particularly affect
stability during downstream processing, storage, and transit through stomach and duodenum. The
enhanced tolerance to environmental stresses of cells from stationary growth phase compared to
exponentially growing cells is well established (Kolter, 1999) and reflects the bacterial ability for
Chapter 1 - Introduction
- 17 -
rapid adaptation to changing conditions. For example, improved tolerances to acidic conditions in
L. acidophilus could be linked to an over-expression of a F1F0-ATPase, an enzyme which is
actively involved in proton extrusion out of the cytoplasm (Kullen and Klaenhammer, 1999).
Exposure of oxygen tolerant B. longum strains to oxygen caused an induction of an Osp-protein
expression as well as changes in the bacterial membrane compositions such as an increase of
short- and cyclopropane fatty acids in the cell membrane to protect the cell from damage (Ahn et
al., 2001).
Using a proteomic approach, several up-regulated stress associated enzymes have been identified
in B. longum cells upon exposure to bile salts (Sanchez et al., 2005; Savijoki et al., 2005) or high
temperatures (Savijoki et al., 2005). These included general stress-related chaperones, proteins
involved in transcription and translation and some proteins from different metabolic pathways. In
a recent study of global gene expression, stress response upon heat treatment of B. longum
NCC2705 showed that 46 % of genes exhibited altered gene-expression. These included the
classical heat shock stimulon including chaperones DnaK, DnaJ or GroEL/ES which assure
correct folding of proteins (Sugimoto et al., 2008) and enzymes involved in repression of cell
division process or proteases to digest heat-induced misfolded proteins (Rezzonico et al., 2007).
Proteomic analyzes of B. longum and B. adolescentis upon bile treatment showed higher levels of
chaperones and enzymes involved in fatty acid synthesis, resulting in changes in membrane fatty
acid composition (Sanchez et al., 2007). These changes may favor tolerance to bile salts by
protecting cells from passive diffusion of bile salt into cytoplasm (Begley et al., 2005, Ruiz et al.,
2007). A model of a number of adaptation mechanisms to bile salt stress of bifidobacteria was
recently developed by Sanchez et al. (2008) based on data acquired from proteomic analyzes
(Figure 3).
Chapter 1 - Introduction
- 18 -
The ability of microbes to adapt to stress conditions resulting in improved survival has been
widely reported and exploited in yogurt starter culture Streptococcus thermophilus or probiotic
cultures (Schmidt and Zink 2000; Wouters et al., 1999; Saarela et al., 2004). Bifidobacteria
treated with various stress conditions such as low pH, heat or combined stresses resulted in
increased stress responses to homologous stresses or cross-protection (Saarela et al., 2004).
Increasing sublethal treatment with NaCl, bile salts or heat exposure enhanced survival of several
bifidobacterial strains to lethal heat stress or freeze-thawing (Schmidt and Zink, 2000). Starvation
of B. longum for 30 or 60 min or exposure of B. lactis to pH 5.2 under starvation conditions
increased survival to prolonged cold storage in growth medium and lethal acidic conditions,
respectively (Maus and Ingham, 2003). However, cellular responses to pre-conditioning and the
subsequent tolerance to lethal stress depend on the sublethal stress (duration, severity,
combination) and the particular strain used (Simpson et al., 2005). The molecular mechanisms
involved in stress resistance have still to be clarified in detail with the help of global stress
response analyzes such as proteomics and transcriptomics (Sanchez et al., 2008). Consequently,
optimization of sublethal treatment has to be assessed strain by strain which is time consuming
and tedious. Recently, a system has been developed in our laboratory using continuous culture in
combination with a two-stage fermentation system which allows the efficient screening of a
number of sublethal stresses under controlled conditions (Mozzetti, 2009). Nevertheless, bacterial
stress response is complex and the number of molecular mechanisms rendering stress resistance
remains to be explored. An alternative to sublethal stress adaptation of probiotics is the use of
immobilized cell technology. In previous studies it was shown that immobilized cells during
continuous culture can be used to produce probiotic strains with improved functional and
technological properties (Lacroix and Yildirim, 2007).
Chapter 1 - Introduction
- 19 -
Figure 3 A model of main physiological mechanisms involved in bile salt tolerance of
bifidobacteria. Conjugated bile acids/salts diffuse to the bifidobacterial cytoplasm (1), and are
cleaved by the bile salt hydrolase (BSH; bile adaptation and acid adaptation protein) and
rendering one amino acid (glycine or taurine) and one deconjugated bile acid moiety (2).
Unconjugated bile salts can also enter the cytoplasm by passive diffusion (3), being deprotonated
at the slightly acid pH of the cytoplasm (4). Ionized bile acids are non-permeable, and must be
excreted by the action of certain transporters, such as the cholate transporter “Ctr” of B. longum
(5) (Price et al., 2006). In addition, the process of tolerance to bile salts is associated with an
increase in the synthesis of molecular chaperones (6) and a decrease in the synthesis of long-
chain-fatty-acid CoA ligase (bile response and adaptation protein) together with a shift in the
fatty acid composition (7). Bile acid deprotonation cause cytoplasm acidification, which is
counteracted with mechanisms such as ammonia production from glutamine deamination (8) or
proton pumping by the F1F0-ATPase (acid response and bile adaptation protein) (9). The amounts
of ATP needed for feeding these mechanisms are provided through the bifid shunt (10) (adapted
from Sanchez et al., 2008).
Abbreviations: AA: amino acid; BCAA: branched chain amino acid; Ctr: cholate transporter
Chapter 1 - Introduction
- 20 -
1.7.1.3 Immobilized cell (IC) technology in continuous cultures
Fermentation technologies, such as continuous culture and immobilized cell systems, are not
widely established in food industries, yet they have potential for enhancing the performance of
fastidious probiotic organisms. These technologies might be employed to develop strains with
improved physiology and functionality when introduced into the gut and to enlarge the range of
commercially available probiotics, as well as expanding product applications (Lacroix and
Yildirim, 2007). In continuous cultures, bacteria grow in a bioreactor with continuous feed of
fresh medium and removal of fermented broth at a given dilution rate. Steady conditions in
continuous cultures allow analyzes of metabolism, growth rate and gene expression of bacteria
under constant conditions during long time period (Hoskisson and Hobbs, 2005). Continuous
culture of B. longum SH2 exhibited increased volumetric biomass productivity compared to
traditional batch cultures (Kim et al., 2003). Recently, a B. longum NCC2705 culture grown in
prolonged chemostat culture for 200 h showed metabolic and transcriptomic stability throughout
entire culture time enabling to apply these cells in a new method to screen for sublethal
treatments (Mozzetti, 2009). Nevertheless, a substantial drawback of continuous cultures is the
increased susceptibility to contamination. The permanent feeding of fresh medium must be
maintained at dilution rates lower than the maximum specific growth rate to prevent wash out of
active biomass. Low dilution rate on the other hand increases probability of a competitive
contaminant strain to establish in the bioreactor.
1.7.1.3.1 Principle of cell immobilization
The principle of immobilized cell fermentation is based on the retention of microorganisms in a
discrete location of the fermentation to yield high biomass and/or to protect bacteria from an
Chapter 1 - Introduction
- 21 -
antagonistic environment. Several immobilization techniques have been applied in dairy
fermentations to produce starter or probiotic cultures and metabolites, including attachment or
adsorption to a preformed carrier, membrane entrapment, microencapsulation and physical
entrapment in polymeric beads (Lacroix et al., 2005).
1.7.1.3.2 Techniques of cell immobilization
A promising technology using immobilized cells to produce probiotic cultures is the entrapment
of microorganisms in a food-grade porous polymeric matrix (Lacroix et al., 2005; Lacroix and
Yildirim, 2007). Cell entrapment is assessed by inoculating liquid polymeric matrix with fresh
cell culture followed by droplet formation of the polymer-cell mixture using extrusion or
emulsification. Spherical polymer gel beads of diameters between 0.3 – 3 mm are formed either
by thermal gelation (κ-carrageenan, locust bean gum (LSB), gellan, agarose, gelatin) or
ionotropic hardening (alginate, chitosan). After gelation, polymeric beads with the entrapped
viable cells are incubated in growth medium to colonize the beads (Lacroix et al., 2005).
When immobilized cells are incubated in growth medium, diffusion limitations occurring in gel
beads for both substrates and inhibitory products, mainly lactic and acetic acid in the case of
LAB, confer a more favorable environment for cell growth close to the bead surface than at the
bead center (Arnaud et al., 1992; Masson et al., 1994; Lamboley et al., 1997; Cachon et al.,
1998; Doleyres et al., 2002a). The very high density of cells at the bead surface can be compared
to bacterial biofilm structures (Lamboley et al., 1997). Colonized polymeric gel-beads incubated
in stirred reactors release large numbers of cells from their peripheral layers to the growth
medium upon pressure due to cell expansion, collision and shearing forces (Sodini et al., 1997;
Lamboley et al., 1999).
Chapter 1 - Introduction
- 22 -
1.7.1.3.3 Physiological changes of immobilized cells
Several studies have shown that immobilized and released bacteria exhibit changes in growth,
morphology and physiology compared with cells produced in conventional free-cell cultures for
LAB and probiotics (Lacroix and Yildirim, 2007). For instance, immobilized Lb. plantarum
showed changes in cell morphology from rod to coccoid shape along with a shift from homo- to
heterofermentative metabolism (Krishnan et al., 2001). Saccharomyces cerevisiae and
Acetobacter aceti exhibited enhanced tolerance to ethanol and acetic acid upon cell
immobilization (Krisch and Szajani, 1997). Similar effects were reported for immobilized
Escherichia coli to phenol and antibiotics (Heipieper, et al., 1991; Jouenne et al., 1994) and
Lactococcus and Leuconostoc strains to quaternary ammonium sanitizers (Trauth et al., 2001).
These alterations in cell physiology or metabolism may have different origins. The biofilm-like
growth on bead surface induces changes in cell response. Natural biofilms comprise of a mixture
of bacteria, proteins, nucleotides and exopolysaccharides which offers protection from
restrictional environment in the host (Stanley and Lazazzera, 2004). Biofilm-embedded cells
exhibit resistance to environmental stresses like high temperatures, low pH, osmolarity and
biocides including antibiotics due to metabolic slowdown and physical protection by the cell
community (Costerton et al., 1999; Fux et al., 2005). When biofilms reach a certain size,
complex structures are formed composed of pillars of water channels to allow nutrient influx and
waste efflux (Davey and O’Toole, 2000). Such a complex community requires some kind of
inter-bacterial communication known as quorum sensing. Bacteria accumulate specific
compounds (peptides, bacteriocins) to a certain threshold, thereby controlling expression of
various genes involved in regulation and adaptation to the exterior environment within the
biofilm (Stanley and Lazazzera, 2004). Although gel beads are artificially produced, the high cell
density on the bead surface are comparable to biofilm environment and could induce quorum-
Chapter 1 - Introduction
- 23 -
sensing responses, leading to improvement in physiological and technological characteristics that
are important for industrial applications of probiotics. The adaptation process might provide a
competitive advantage for the population, more effective adherence properties and rapid
responses to changing environmental conditions when introduced to a host (Sturme et al., 2002).
Alternatively, changes in cell morphology and physiology can be triggered by non-specific stress
adaptation during maturation of cells caused by the steep gradients of pH, nutrients and inhibitors
(organic acids) dominating the interior of the colonized gel beads (Doleyres et al., 2002b).
1.7.1.3.4 Advantages of continuous IC cultures
Continuous IC systems are usually processed at high dilution rates because physical retention of
cells prevents wash-out of active biomass and provide efficient biomass production. High
dilution rate ensures high substrate supply for cell growth and metabolism and rapid removal of
metabolites to prevent product inhibition, and diminishes risk of contamination. Continuous
culture with IC might also favor selection of stress tolerant subpopulations during prolonged
fermentation time (Doleyres et al., 2004a). Other advantages over free cell fermentations are the
reuse of biocatalysts, less sensitivity to contamination and bacteriophage attack, enhanced
plasmid stability, and physical and chemical protection (Doleyres and Lacroix, 2005).
Immobilized starter or probiotic bacteria have been produced using repeated batch or continuous
culture and exhibited various benefits compared to traditional batch free cell cultures (Table 2).
The complex cell response upon growth in polymeric gel beads remains to be elucidated. Further
knowledge about the mechanism responsible for the stress adaptation on a physiological and
global cell response level (transcriptomics, proteomics, metabolomics) may lead to the
production of probiotic strains with enhanced technological and biological properties during cell
production, gut transition and colonization in the human GIT (Lacroix and Yildirim, 2007).
Table 2 Fermentation systems with immobilized LAB or bifidobacteria and the corresponding bacterial phenotype Fermentation system/ immobilization matrix/ growth medium
Species Benefit of IC Reference
1-stage continuous/ polyacrylamid lattice/ Milk
Lb. casei
Investigate effect of IC on cell metabolism - Stable production of metabolites - Bead stability
Diviès and Siess, 1976
2-stage continuous/ fed batch/ alginate/ milk or cream
Lb. bulgaricus St. thermophilus St. diacetylactis; St. lactis; St. cremoris Lc. Lactis; Lc. diacetylactis
Yogurt/Cheese production and cream fermentation - Constant yogurt production, small volume - Cheese with constant properties, small volume - Cream with improved storage properties
Prévost et al., 1985; Prévost and Diviès, 1987 and 1992
1-stage continuous/ κ-carrageenan, locust soy bean (LSB)/ reconstituted skim milk, yeast extract (YE)
B. infantis Testing applicability of IC in milk fermentation - high bead stability - high biomass/cell volumetric productivity
Ouellette et al., 1994
2-stage continuous/ κ-carrageenan, LSB/ whey permeate, YE
Lb. helveticus Kinetic study of lactic acid production with IC - high lactic acid production - high biomass/cell volumetric productivity
Norton et al., 1994
1-stage continuous/ κ-carrageenan, LSB/ milk
Lc. lactis, Lc. diacetylactis, Lc. lactis; Leuconostoc mesenteroides
Continuous fermentation for fresh cheese production - feasible system for constant milk pre-fermentation in fresh cheese manufacture
Sodini et al., 1997
1-stage continuous/ alginate/ MRS (no acetate, lactose)
Lc. diacetylactis Comparison of chemostat IC vs free cell batch - high cell productivity - change of redox state
Cachon et al., 1998
1-stage continuous/ κ-carrageenan, LSB/ whey permeate, YE
2 strains Lc. lactis and 1 Lc. diacetylactis
Production of mixed starter culture - long-term biological stability for mixed starter culture
Lamboley et al., 1999
Repeated batches/ chitosan treated polypropylene/ medium of glucose, YE minerals
Lb. plantarum Physiological analysis of IC - shift from homo- to heterofermentative metabolism - changes in cell morphology
Krishnan et al., 2001
Repeated batches/ κ-carrageenan, LSB/ MRSC + CaCl2
B. longum; Lc diacetylactis Study effect of IC on co-cultivation - successful co-cultivation of mixed culture containing a non-competitive B. longum strain
Doleyres et al., 2002a
Table continued
1-stage continuous/ gellan/ MRS and cysteine (MRSC), whey permeate (WP)
B. longum Comparison of chemostat IC vs free cell batch growth - higher volumetric productivity
Doleyres et al., 2002b
2-stage continuous κ-carrageenan, LSB/ MRSC, lactose, CaCl2
B. longum; Lc. diacetylactis Effect of co-culture in chemostat IC on cell physiology - increased stress tolerance to freeze drying, H2O2 and gastrointestinal conditions
Doleyres et al., 2004a
2-stage continuous/ silicone rubber solid carriers/ whey permeate, YE
Lb. rhamnosus Biomass production chemostat IC vs free cells - induction of cell aggregation
Bergmaier et al., 2005
2-stage continuous/ κ-carrageenan, LSB/ whey permeate, YE
Lb. helveticus Modelling lactic acid production with chemostat IC - high lactic acid production - physiological adaptation
Schepers et al., 2006
1-stage continuous/ κ-carrageenan, LSB/ milk
Lb. rhamnosus; Lc. diacetylactis; P. acidilactici; Lc. cremoris
Test effect of mixed cultures with antagonistic properties in chemostat IC - enhanced acidification and tolerance to nisin
Grattepanche et al., 2007
2-stage continuous/ gellan, xanthan/ MRS
Lb. delbrüeckii Study the effect of chemostat IC and culture time - changes in cell morphology/aggregation - enhanced survival to freeze drying
Koch, 2006
1-stage continuous/ gellan, xanthan/ MRS + H2O2
B. longum System to apply selective pressure to continuous IC - selection of H2O-resistant mutant strain(s)
Mozzetti, 2009
Chapter 1 - Introduction
- 26 -
1.7.2 Stabilization of probiotics
1.7.2.1 Microencapsulation
Microencapsulation is a technology for packing solids, liquids or gaseous material in small sealed
capsules, which can release their contents at controlled rates under specific conditions (Shahidi
and Han, 1993). The principle of microencapsulation has initiated numerous efforts in the last
two decades to embed sensitive metabolically active probiotic strains in microcapsules or
microparticles in order to increase stability during shelf-life and consumption. Encapsulation was
shown to protect probiotic bacteria from oxygen, acidic environments in food products,
refrigerated storage, and simulated gastro-intestinal conditions (Doleyres and Lacroix, 2005;
Anal and Singh, 2007).
Different techniques and materials have been used for microencapsulation of probiotic cultures.
Cells encapsulated in gel beads are achieved by entrapping probiotic cells within a polymeric
matrix such as pectin, gellan gum, κ-carrageenan-locust bean gum and alginate. Gel capsules
containing probiotics can be produced by various methodologies such as emulsion or extrusion
techniques (Champagne and Fustier, 2007; Talwalkar and Kailasapathy, 2003). Several studies
reported protective effects of microencapsulation and high survival of sensitive bifidobacteria and
lactobacilli in plant polymer matrices (Krasaekoopt et al., 2003), particularly against oxygen
(Talwalkar and Kailasapathy, 2003), acidic environment (Sun and Griffiths, 2000), freezing
(Shah and Ravula, 2000), refrigerated storage of ice cream and milk (Hansen et al., 2002) and
yoghurt (Adhikari et al., 2003; Adhikari et al., 2000; Sultana et al., 2000) and during simulated
gastro-intestinal tests (Lee and Heo 2000; Ding and Shah, 2007; 2009). Although promising on a
laboratory scale, the technologies developed to produce gel beads present serious difficulties for
large-scale production such as low production capacity and large bead diameters (2-5 mm), that
Chapter 1 - Introduction
- 27 -
can affect food texture (Hansen et al., 2002). Moreover, the addition of these polysaccharides is
not permitted in yogurts or fermented milks in some European countries (Picot and Lacroix,
2004). Beside the limitations in up-scaling and interference with textural and sensorial properties
of food products (Anal and Singh, 2007), the proper release of viable cells at the target site is also
an obstacle in the probiotic approach. In their recent review, de Vrese and Schrezenmeir (2008)
reported that alginate capsules containing probiotics did not release their content after
administration to human and pigs and considered this type of encapsulation although frequently
used, as unsuitable as carrier for probiotic bacteria.
Another technique to encapsulate probiotics is the spray-coating technology for large scale
production, which is widespread in industries for diverse commercial products (Anal and Singh,
2007; Champagne and Fustier, 2007). But since the technology is difficult to master and most
information on the technology is proprietary to industry, there is only limited access for academic
research (Champagne and Fustier, 2007). Spray coating consists of suspending, or fluidizing,
particles of a core material in an upward stream of air and applying an atomized coating material
(lipid-based, proteins or carbohydrates) to the fluidized particles (Augustin and Sanguansri,
2008). One particular patent claims improved stability of spray-coated cells to simulating gastric
conditions, high temperature and during storage in powdered milk compared to uncoated freeze
dried cells. In addition the coated cells had better resistance to compression (US patent
2003/0109025).
1.7.2.2 Spray drying
Dairy starter and probiotic cultures are mainly preserved and distributed in frozen or dried form.
Frozen cultures occupy large volume and require storage at sub-zero temperatures causing high
Chapter 1 - Introduction
- 28 -
costs of storage, shipping and energy and hence preservation in dried form is sometimes preferred
(Johnson and Etzel, 1995). Transformation of bacteria to a dried form through the physical
removal of water is a critical step since bacteria require a water activity (aw) of about 0.98 in the
product matrix for survival and growth. Preservation of a viable state requires either high aw for
metabolic activity or low aw to survive in dormant state in powders (Paul et al., 1993). Drying of
bacterial suspension is mainly assessed by freeze- or spray drying.
Freeze drying is widely used for formulation of starter and probiotic cultures but it is an
expensive process with low yields, and as such spray drying offers an alternative inexpensive
approach yielding higher production rates (Zamora et al., 2006). Spray drying is energy-efficient
and an established tool in food industries for the production of milk powders and instant coffee.
Despite restrictive process conditions for microorganisms (inlet reaching ≥ 180 °C), the rapidity
of drying combined with the possibility to dry large amounts of bacterial cultures has caught the
attention of research and industry in recent years (Zamora et al., 2006; Chàvez and Ledeboer,
2007; Meng et al., 2008).
1.7.2.2.1 Principle of spray drying
During spray drying, feed solution is transformed from a fluid state into a dried form by spraying
the feed into a hot drying medium (Figure 4). The spray drying process can be divided into two
phases. First, the constant drying rate period just after the feed solution is dispersed by a fluid
nozzle or atomizer and encounters the hot air in the drying chamber. The interior of the droplets
is not challenged to heat exposure due to the cooling effect of evaporation. During the subsequent
falling drying period, the temperature of the particles increases but does not reach values of the
inlet temperature, because the remaining moisture content compensates exposure to heat until the
Chapter 1 - Introduction
- 29 -
entire water is evaporated. Hence, the residence time of the droplets in the drying chamber drives
the moisture content of the particles, as well as the survival of the microorganisms within the
particles (Santivarangkna et al., 2007). It is generally accepted that bacteria in the dried particle
are exposed to no higher temperatures than the outlet temperature, which in turn constitutes the
crucial factor for bacterial survival to spray drying (Gardiner et al., 2000; Lian et al., 2002). The
construction of the spray cylinder is designed to allow adequate residence time and droplet
trajectory distance for achieving the heat and mass transfer. After the drying of the spray in the
chamber, the majority of the dried product falls to the bottom of the chamber. The dried product
can either be discharged continuously from the bottom of the spray cylinder or passed into a
solid-gas separator (cyclone) where the solids are recovered from the gas stream. The powder is
then collected at the bottom of the glass container. Very fine particles which do not precipitate in
the glass container are usually collected in an outlet filter connected in series to the cyclone. The
operating conditions and dryer design are selected according to the drying characteristics of the
product and powder specification. The main parameters for powders are the residual moisture
content and the particle size. Moisture content depends mainly on evaporation rate and the dryer
∆ T (inlet air temperature minus the outlet air temperature), which in turn dictates the amount of
drying air needed and ultimately the sizing and cost of almost all of the system components. The
particle size requirement affects the choice of atomization method and can also affect the size of
the dryer.
The system allows many heat sensitive products (enzymes, microorganisms, volatile aroma
compounds, etc.) to be spray dried and transformed to powders with low moisture content.
Chapter 1 - Introduction
- 30 -
Figure 4 Pathways of air and feed during spray drying
(Left) Path of the drying air in the spray dryer Büchi B290. Cold air is aspirated through the air
inlet tunnel (1) and then electrically heated (2) prior to the atomization of the feed solution by the
two-fluid nozzle (3). Dispersed feed enters the spray cylinder (4), in which drying of the droplets
into solid particles takes place. Dried powder is separated from fine particles in the cyclone (5)
and collected in the glass container (6). The outlet filter (7) is placed in series with the cyclone to
remove fine particles and to prevent them from entering aspirator (8) which generates the air
flow. Temperatures are measured in the entrance of spray cylinder (9, inlet temperature) and in
the intermediate piece between spray cylinder and cyclone (10, outlet temperature).
(Right) Path of the feed solution and pressurized air in the spray dryer in co-current mode. Feed
solution (A) conveyed by peristaltic feed pump (B) and atomizing air (D, inlet) are passed
separately to the nozzle head (C), where the atomization of the feed solution into fine droplets
takes place. The co-current two-fluid nozzle is located at the centre of the upper part of the spray
cylinder. Atomization is created by compressed hot air at a pressure of 0.5 to 2 bar. Powders
produced with this adjustment have particle size ranged from 5 to 15 µm on average.
Chapter 1 - Introduction
- 31 -
1.7.2.2.2 Parameters affecting cell viability during spray drying
The drying process causes cell damage and decreased viability due to mechanical, heat, osmotic
stress, exposure to oxygen, and the removal of bound water (Meng et al., 2008; To and Etzel,
1997). The decrease of water availability inside or in the vicinity of the dried cells affects the
physiology of bacteria. Removal of water eventually results in slow down of bacterial
metabolism or stops it entirely after which bacteria enter a dormant state. Possible damages
caused by the dehydration are:
• increased permeability of cell wall/ cytoplasmic membrane to extracellular or leakage of
cytoplasmic compounds due to lesions in cell envelope (Texeira et al., 1996).
• transformation of liquid crystalline structure of phospholipid bilayer to gel phase and
incomplete reverse rehydration may cause leaking and disturbance of molecule transport
systems (Oliver et al., 1997).
• peroxidation of lipid cause membrane leakage causing loss of plasmids or increased influx of
DNase into cells affecting DNA integrity (Meng et al., 2008).
• conformational changes of enzymes may block or modify active binding sites resulting in
inactivation of the enzyme (Prestrelski et al., 1993; Burin et al., 2004).
• changes in the lipid bilayer of cell membrane causing displacement, denaturation or loss of
membrane-bound proteins (Texeira et al., 1996).
Poor survival of LAB and bifidobacteria during drying are substantial draw-backs for this
preservation method and strategies to optimize cell viability are thus required (Meng et al., 2008;
Santivarangkna et al., 2007). Survival of heat sensitive LAB and bifidobacteria during spray
drying depends on several factors such as the drying carrier, bacterial growth phase, the particular
Chapter 1 - Introduction
- 32 -
probiotic strain used and mainly on the outlet temperature of the drier. Adjustment and
optimization of these parameters affects bacterial survival to spray drying as shown in Table 3.
Stabilizing compounds for drying and shelf live are hydrophilic polyhydroxy substrates like
sugars which partly substitute missing water molecules in membranes and proteins (Leslie et al.,
1995, Oliver et al., 1997, Corcoran et al., 2004). Reconstituted skim milk (RSM) as a carrier
contains lactose as a possible water substituent and was shown to be suitable for drying of
probiotics (Corcoran et al., 2004; Ananta et al., 2005). Besides lactose, skim milk proteins in
parallel with calcium may prevent cellular injury by stabilizing phospholipid bilayers and
shielding the membrane-bound proteins by formation of a protective coating (King and Su, 1993;
Castro et al., 1995).
In recent years, numerous studies on the optimization of spray drying of probiotics have been
performed to either increase cell robustness or to modify the drying parameters to enhance the
yields of viable cells in spray dried powders (Table 3).
Table 3 Survival rates of lactobacilli and bifidobacteria during spray drying
Microorganism Survival (%) a T outlet (Tout)
T inlet (T in) Feed solution Reference Special remarks
Lb. acidophilus Tout 75, 80, 85 °C RSM (25%) 8.2, 2.4, 0.5 RSM (40%) 3.7, 0.5, 0.3
75, 80, 85 170 25 % solid milk 40 % solid milk
Espina and Packard, 1979
Moisture content of powder (M.C.) 3.5 - 6.1%
S. salivarius ssp. thermophilus, Lb. delbrueckii ssp. bulgaricus
2 - 0.1 60, 70, 80
140 – 180
yoghurt, 13,63% solid non fat
Kim and Bhownik, 1990
Lb. helveticus 15 82, 120 220 19% maltodextrin Johnson and Etzel, 1995
Tout adjustment using feed flow rate
Lb. bulgaricus n. i. 62-105 200 skim milk Texeira et al., 1995 Lb. bulgaricus n. i. 80 200 40% maltodextrin in
water or RSM 11% Texeira et al., 1995
Lb. bulgaricus
n. i. 70 200 RSM Texeira et al., 1996
Lc. lactis, Lb. casei, S. thermophilus
3.0 – 0.4 14.7 – 13.0 34 – 5.0
70 – 90 220 25% (w/w) solid maltodextrin
To and Etzel, 1997
Lb. lactis 14.5 – 0.6 68 160 – 200
20% RSM Mauriello et al., 1999
Flow rate feed: 10, 13, 17 ml/min
S. thermophilus, Lb. delbrueckii
54.7 – 51.6 15.8 – 13.7
60 – 80 180 yoghurt Bielecka and Majkowska, 2000
Water evaporation capacity 1645 kg/h
Lb. paracasei Lb. salivarius
65.0 1.0
60-120 170 20% RSM Gardiner et al., 2000
M.C. ca. 4%
Lb. paracasei Heat: Control 4 Adapted 23 Salt: Control 7 Adapted 33
85 – 105 170 20% RSM Desmond et al., 2001
M.C. 1.7 - 3.3 % Tout adjusted with feed flow rate
Bifidobacteria 16.6 45 100 10% (encapsulation) O’Riordan et al., 2001
B. infantis (2 strains), B. longum (3 strains)
82.6 – 0.1 50-60 100 Gelatin, gum arabic, skim milk, soluble starch, 10% RSM
Lian et al., 2002 M.C. 7-9% Carrier dependent survivalc: RSM> Gelatin> gum> starch
Table continued
Lb. salivarius, Lb. sakei
100 41.1
70 200 11% milk Silva et al., 2002
Lb. paracasei NFBC338 Tout 100, 105 °C 20% RSM: 1.7, 1.4 10% RSM+gum: 0.9,< 0.01
105 100
170°C 20% RSM or 10% RSM with 10% gum accacia
Desmond et al., 2002
M.C. 2.5 - 3.2% Tout adjusted with feed flow rate
Lb. acidophilus B. lactis Bb12
Tin 130 °C, Tout 75 °C Lb. acido. 1.9 B. lactis 72.2
75 – 125
130 – 190
Cellulose acetate phthalate, Glycerol, Maltodextrin, Raftilose, milk
Favaro-Trindale and Grosso, 2002
Avg. particle size 22 µm
Lb. rhamnosus GG, Lb. rhamnosus E800, Lb. salivarius
RSM, RSM + polydextrose 50, 30b 25, 41b 0.7, 80.1b
85-90 170 RSM 20% or 10% RSM with 10% polydextrose
Corcoran et al., 2004
M.C. <4%, Tout adjustment using feed flow rate
B. brevis B. longum
B. brevis R070 25.7 B. longum R023 1.4
80 160 10 % Whey protein Picot and Lacroix, 2004
M.C.: B. brevis 2.0% B. longum 2.1%
Lb. delbrueckii ssp. bulgaricus
n.i. 70 200 11% RSM Silva et al., 2004 Sucrose in growth medium: anti-oxidant for powder storage
St. thermophilus Lb. acidophilus B. infantis, B. longum
Tout 75 °C B. infantis mixed with - Lb.acido 2.9 - St. thermo. 3.4 B. longum mixed with - Lb.acido. 11.1 - St. thermo. 13.1
60 – 90 100 Fermented soy milk Wang et al., 2004 M.C. 4.2 – 10.6 %; Study about influence of mixed cultures on survival to spray drying
Lb. rhamnosus GG RSM 20% ca. 65 b RSM/ ca. 55 b Polydextrose RSM/ ca. 65 b Raftilose
70 – 100 varying RSM 20% modified with oligofructose and polydextrose
Ananta et al., 2005 M.C. 2.7 – 4.5 % Tout adjusted by changing Tin,
Table continued
Bifidobacterium subsp. 12 – 102 85 – 90 170 RSM 20% (w/v) or RSM (10%, w/v) and gum acacia (10%, w/v)
Simpson et al., 2005
M.C. 2.7 – 4.2 %
Lb. salivarius BetL+ 1.4 BetL- 0.3
80 – 85 170 RSM Sheehan et al., 2006
“BetL” plasmid containing listerial betaine uptake system BetLd
Lb. acidophilus Tin 120 °C, Tout 79 °C
1.5
58 – 79 100 – 120
RSM, whey permeate with guarana gum
Riveros et al., 2009 M.C. 6.3 %
Lb. rhamnosus GG; Lb. rhamnosus E-97800
MSG., RSM 69, 75 23, 55
60 – 75 varying Trehalose (20%, w/v) supplemented with 12.5 g/l monosodium glutamate (MSG)
Sunny-Roberts and Knorr, 2009
M.C. 3.6 – 4.4 %
a Different survival rates correspond to different spray drying parameters (e.g. strain, carrier, Tout etc.) b Pproximate values estimated from data in figures. c Series of descending survival for each carrier for the majority of tested strains d Genetically modified organism n.i.: not indicated
Chapter 1 - Introduction
- 36 -
1.7.2.2.3 Parameters affecting storage of dried powders
During powder storage, microbial stability is affected by parameters like temperature, oxygen
content, light exposure and storage materials. Moreover, powder characteristics such as moisture
content, powder composition and relative humidity can influence the glassy state of the carrier
medium and eventually the viability of microorganisms during shelf life. Glass transition
temperature (Tg) is the critical temperature at which a glass-forming liquid transforms from a
rubbery to a glassy state. The specific Tg of sugar containing powders plays an important role in
the stability of powders containing microorganisms and is dependent on storage temperature and
moisture content of the powder (Figure 5). High viscosity of sugar glasses below or around their
glass transition temperature retards molecular mobility and reactivity resulting in a stabilizing
effect of biological systems (Buitink et al., 2000). Consequently, the storage of dried cultures at
temperatures below the glass transition temperatures of sugars in powder mixtures enhance shelf-
life stability (Vega and Roos, 2006; Santivarangkna et al., 2008). High storage temperature and
remaining water in dried powders have thus a detrimental effect on cell viability, as shown for
dried lactobacilli and bifidobacteria in skim milk (Gardiner et al., 2000; Silva et al., 2002; Zayed
and Roos, 2004; Simpson et al., 2005).
Cell damage during storage may be caused by oxidation of membrane lipids. The presence of
allyl-groups in unsaturated fatty acids causes instable membrane properties during storage since
double-bonds are targets to oxidation and lead to formation of cytotoxic hydroperoxides (Texeira
et al., 1996). Progressive changes of lipid composition were observed during storage of
lactobacilli (Castro et al., 1996) and cells with damaged bacterial membranes became
consequently more sensitive to NaCl (Texeira et al., 1996). The detrimental effects on cell
viability during storage require treatments to stabilize and improve stability of spray dried
probiotic powders to make spray drying a cost efficient alternative to traditional freeze drying.
Chapter 1 - Introduction
- 37 -
Figure 5 A simplified state diagram shows the glass transition curve which relates the glass
transition temperature and moisture content. Molecular mobility and deleterious reaction rates in
the glassy state are extremely low, while they increase at the storage conditions above the curve
(rubbery state) (Santivarangkna et al., 2008).
Studies on optimization of storage stability of bacteria in dried powders focused on
supplementation or substitution of the carrier medium (usually RSM) by oligosaccharides or
oxygen scavenger compounds. Although sugars are believed to be the major components
affecting glass formation, it has also been shown that polypeptides can significantly alter the
glass properties of sugars (Buitink et al., 2000) and therefore play the important role in glass
formation. This could explain why skimmed milk powder containing a large number of milk
proteins is considered as an efficient desiccation protectant (Hubalek, 2003). Powder stability of
lactobacilli dried in RSM supplemented with prebiotic substances or polydextran did not improve
Chapter 1 - Introduction
- 38 -
bacterial viability during storage compared to RSM alone (Corcoran et al., 2004; Ananta et al.,
2005). Addition of anti-oxidants such as ascorbic acid and monosodium glutamate to powders of
Lb. delbrueckii ssp. bulgaricus stored at different temperatures showed ambiguous results and
even destabilizing effect at 20 °C (Texeira et al., 1995) concluding that protective agents during
the dehydration process do not necessarily support viability of cells during storage (Crowe et al.,
1987). Stability of probiotic powders can be improved by proper packaging. The use of
aluminium-based packaging material that protects the powder from exposure to light, moisture
and oxygen supports storage stability of dried probiotics (Nagawa et al., 1988; Ishibashi and
Shimamura, 1993). Shelf-life stability of probiotic bacteria is strain dependent and not related to
the survival during drying (Gardiner et al., 2000; Simpson et al., 2005; Sunny-Roberts and
Knorr, 2009). Hence, a careful selection of robust strains should be performed with a view to
obtained strains which have stable shelf life properties.
Variation of stability during drying and storage among strains of the same species can impede the
ability to extrapolate results across entire subspecies or species. Predictions are thus of limited
significance. Consequently, studies on the effect on survival of changing parameters in drying
and storage applications have to be performed strain by strain. This is laborious and time
consuming, particularly during storage tests, and hence new technologies for efficient
simultaneous screening methods are required to increase output of such experiments.
1.7.2.2.4 Rehydration
The rehydration of probiotic powders is the final critical step during stabilization of probiotic
powders and revitalization conditions can have significant influence on the recovery of viable
cells since bacterial membranes become more susceptible to environmental stresses (Texeira et
al., 1996). Revitalization of cells and can be divided into four steps: wetting, submersion,
Chapter 1 - Introduction
- 39 -
dispersion and dissolving (Freudig et al., 1999). Wetting of the particles is very often the
reconstitution controlling step and the rate of recovery depends on various factors including
osmolarity, pH, and nutritional energy as well as on temperature and volume (Carvalho et al.,
2004, Vega and Roos, 2006). Slow rehydration under controlled conditions was shown to
improve cell viability compared to immediate rehydration (Poirier et al., 1999). The medium to
rehydrate cells also affects recovery rate. For instance, complex media such as 10% (w/v) RSM
and PTM media consisting of peptone, tryptone and meat extract positively affected bacterial cell
recovery compared to minimal media such as phosphate buffer, sodium glutamate and water
(Costa et al., 2000). In other studies the use of the same cryopreservation and rehydration
solution resulted in increased viability (Ray et al., 1971; Abadias et al., 2001). A possible
explanation is that such a solution provides a high osmotic pressure environment which controls
the rate of hydration, and thus avoids osmotic shock (Morgan et al., 2006).
1.7.3 Enumeration of viable probiotic cells
Apart from advanced production and downstream processing technologies, the criterion of
minimal levels of viable cells within probiotic products requires also adequate quantification
methods to detect the viable probiotic strains in functional food products. The application of
rapid and reliable methods for enumeration of probiotics in mono- and mixed cultures is
important to validate quality of products containing probiotics. Cell quantification is generally
performed by plate counting on selective media to distinguish the microorganisms. However, this
method has several limitations including cell clumping, inhibition by neighboring cells, and it is
time consuming (Auty et al., 2001; Breeuwer and Abee, 2000; Bergmaier et al., 2005). Another
aspect concerns the lack of a standard medium to distinguish bifidobacteria from other LAB and
Chapter 1 - Introduction
- 40 -
expressed thus the need for alternate, non-culture based technologies for adequate cell
enumeration (Talkalwar and Kailaspathy, 2003; Masco et al., 2005). The use of fluorescent stains
alone or in combination with enzymatic assays enabled the differentiation between viable,
metabolic active, damaged, dormant, viable but not cultivable, and dead bacterial cells and were
used to quantify viable bacteria (Breeuwer and Abee, 2000; Alakomi et al., 2005). Molecular
tools like fluorescence in situ hybridization (FISH) and flow cytometry were successfully applied
to estimate viable cells in probiotic products (Maukonen et al., 2006). Moreover, flow cytometry
can distinguish between subpopulations of stressed versus unstressed bifdobacteria (Ben Amor et
al., 2002). However, culture-based and fluorescent methods demand single cells suspension
because cell in clusters or chains lead to biased quantification of viable cell counts (Daley et al.,
1977; Nebe-von-Caron et al., 2000; Bibiloni et al., 2001).
There exists however alternative methods to detect or quantify bifidobacteria in fecal samples and
pharmaceutical or probiotic products. Technologies to detect bifidobacteria are based on DNA
fingerprinting approaches such as denaturing gradient gel electrophoresis (DGGE), pulsed-field
gel electrophoresis (PFGE). Quantitative approaches relate to the use of quantitative real time
PCR (qRT-PCR) (Vitali et al., 2003; Gueimonde et al., 2004; Masco et al., 2005; Matsuda et al.,
2007; Masco et al., 2007). But these methods are unable to provide information on the metabolic
activity or viability of the microorganisms since DNA or rRNA are too stable to be used as
viability markers (Josephson et al., 1993; Sheridan et al., 1998). mRNA on the other hand with
short half-life times can be considered as accurate viable cell marker (Belasco et al., 1986;
Hellyer et al., 1999). The quantification of mRNA using qRT-PCR has been recently applied in
medical applications to enumerate viable pathogenic microorganisms (Birmingham et al., 2008,
Coutard et al., 2005), but to date no such approach with mRNA has been applied to probiotic
bacteria. Growing numbers of accessible genome sequences of food related microorganisms and
Chapter 1 - Introduction
- 41 -
progressing microarray technology will enlarge the knowledge about gene expression profiles of
bacteria in different environments which will raise the number adequate target mRNAs to
accurately quantify viable bacteria in various environments.
1.8 Hypothesis and background of this study
1.8.1 Background and general objective
Bifidobacteria are fastidious bacteria and sensitive to various environmental stresses during
production, downstream processing, incorporation in probiotic food products, storage and transit
through upper GIT. There is a general demand by food industries for advanced technologies to
enhance cell robustness and to improve selection procedure of candidate strains to meet minimal
viable cell numbers in products of prospering probiotic food market. Furthermore, the limitations
of accurate enumeration of viable cells require alternative methods to accurately quantify viable
cells independent of their morphological state.
The aim of the first part of the dissertation was to produce and characterize stress tolerant B.
longum NCC2705 strains using IC during continuous culture (Chapter 2). Continuous IC culture
induced formation of large cell aggregates which made physiological characterization with
traditional plate count method very difficult. Hence, we developed a new method to enumerate
viable B. longum NCC2705 irrespective of the morphological state (Chapter 3). The method
based on the quantification of the mRNA of a constitutively expressed housekeeping gene with
real-time PCR technology. With the help of this method, we were able to characterize survival of
aggregated cells produced with IC technology to lethal bile and heat stresses (Chapter 2).
In the last part of the doctoral thesis, we chose a different approach. Instead of producing stress
tolerant B. longum strains, we developed a high throughput screening method to identify B.
Chapter 1 - Introduction
- 42 -
longum strains with intrinsic resistance to stresses occurring during spray drying and storage
(Chapter 4). Molecular biological tools allowed a differentiation of 22 B. longum strains of which
we were able identify the best survivor strain after spray drying and storage. The method was
then validated by comparing survival of the selected strain with a poor surviving strain.
1.8.2 Specific objectives
a. Characterization of B. longum NCC2705 produced with ICT in continuous culture:
� Perform a long-term fermentation with immobilized cells in a two-stage fermentation
system for 20 days
� Characterize biomass production, cell metabolism, and membrane composition
� Analyze the effect of cell immobilization and continuous culture on cell physiology:
tolerance to antibiotics, heat and porcine bile compared to free cell batch culture
b. Development of a real time PCR method to quantify viable B. longum NCC2705 cells in large
aggregates:
� Generate calibration curves with level of transcript of two housekeeping genes as a
function of cell concentration
� Validate target fragments of transcripts as viability marker
� Apply the method to quantify viable cells in large aggregates from continuous IC culture
Chapter 1 - Introduction
- 43 -
c. Development of a fast screening method for selection of Bifidobacterium longum strains
resistant to spray drying and storage in dry powder:
� Establish strain specific molecular profiles using randomly amplified polymorphic DNA
(RAPD)
� Perform spray drying and storage experiments of mixtures of B. longum strains
� Validate the selection of resistant strain with single strain spray drying and storage
experiment
1.8.3 Hypotheses
� Immobilization and continuous culture of Bifidobacterium longum NCC2705 improves
volumetric productivity and tolerance to bile salt, heat and antibiotics.
� Quantification of transcript levels of a housekeeping gene with qRT-PCR allows accurate
enumeration of viable Bifidobacterium longum NCC2705 cells of different morphologies.
� The simultaneous spray drying and storage of mixed B. longum strains allows the rapid
identification of the most robust strains using RAPD PCR technology.
Chapter 2 - Immobilized cells
- 44 -
2 Improved tolerance to bile salts of aggregated
B. longum NCC2705 produced during continuous
culture with immobilized cells
Chapter 2 - Immobilized cells
- 45 -
2.1 Abstract
The effect of cell immobilization and continuous culture was studied on selected physiological
and technological characteristics of Bifidobacterium longum NCC2705. B. longum NCC2705 was
entrapped in gellan/xanthan gel beads and cultivated for 20 days with and without glucose
limitation in a two stage continuous fermentation system, with a first reactor (240ml) containing
immobilized cells, and a second reactor (1800ml) inoculated with free cells produced in the first
reactor. Continuous immobilized cell (IC) cultures with and without glucose limitation exhibited
formation of macroscopic cell aggregates after 12 and 9 days, respectively. Auto-aggregation
resulted in underestimation of viable cell counts by plate counts by more than 2 log units CFU/ml
compared to cell counts obtained with a cell-aggregate-independent method using real time PCR
technology. Cells in the effluent of continuous IC culture showed enhanced tolerance to porcine
bile salts and to aminoglycosidic antibiotics compared to stationary-phase cells produced during
free-cell batch fermentations. However, tolerance to lethal heat treatment was not different than
that from batch control culture. Changes in the cell membrane composition of continuously
produced cells such as the decrease of the ratio of unsaturated divided by saturated fatty acid
content of 0.57 ± 011 were observed compared to 1.74 ± 0.00 of membranes from batch control
culture.
Cells produced during continuous immobilized-cell cultures exhibited changes in cell membranes
which may induce formation of cell aggregates and tolerance to bile salts and aminoglycosidic
antibiotics. Cell immobilization in combination with continuous culture can be used to produce
probiotic bacteria with improved stress tolerance.
Chapter 2 - Immobilized cells
- 46 -
2.2 Introduction
Probiotics are “living micro-organisms which upon ingestion in certain numbers exert health
benefits beyond inherent basic nutrition” (WHO/FAO 2001). Bifidobacteria are widely used as
probiotics in functional food products (Leahy et al., 2005). Bifidobacteria are sensitive to various
factors during production and downstream processing and need to adapt to competitive and
changing environment in the human gastrointestinal tract to maintain viability (Lacroix and
Yildirim, 2007). Apart from safety issues, selection of potential probiotic candidate strains relies
on technological as well as functional properties. Among others, resistance to oxygen, heat and
gastro-intestinal conditions during production and consumption as well as the ability to adhere to
intestinal mucosa are thought to be important to provide beneficial health effects for the host
(Ouwehand et al., 2002; Salminen et al., 1999).
Different strategies such as pre-treatment of bacterial cells with sub-lethal stresses (Desmond et
al., 2002; Maus and Ingham, 2003) or physical protection using whole cell encapsulation
(Champagne et al., 2007) have been proposed to enhance stability of probiotics to stressful
conditions. Doleyres et al. (2004a) reported for continuous co-culture of immobilized B. longum
and L. lactis strains that cells produced in the effluent medium of the immobilized cell reactor
exhibited significantly increased tolerance to hydrogen peroxide, freeze-drying, and simulated
gastrointestinal conditions compared to free cells cultured in batch mode. Enhanced physiological
properties upon cell immobilization were also reported for lactobacilli where tolerance to nisin Z
and higher acidification activity were observed for continuously produced cells in mixed culture
(Grattepanche et al., 2007). The enhanced stress tolerance of immobilized cells were tentatively
attributed to changes in the microbial cell membranes of LAB and/or to non-specific stress
adaptations due to steep gradients of inhibitory products, pH and biomass within gel beads
Chapter 2 - Immobilized cells
- 47 -
(Masson et al., 1994; Trauth et al., 2001; Lacroix et al., 2005; Schepers et al., 2006). However,
the exact mechanism conferring increased stress tolerance remains unclear.
In this work, we studied the production of B. longum NCC2705 and time stability of continuous
cultures with immobilized cells (IC) in a two-stage fermentation system similar to that used by
Doleyres et al. (2004a) for mixed cultures with a first small reactor containing cells immobilized
on gel beads and a second large reactor connected in series and continuously inoculated from
cells produced in the first reactor. B. longum NCC2705 was chosen as model strain because its
genome has been sequenced (Schell et al., 2002). Furthermore, the probiotic characteristics,
technological properties, and stress tolerance and adaptation of this strain have been thoroughly
analyzed in the review of Klijn et al. (2005). The effect of cell immobilization and fermentation
time on cell production, metabolism, and selected physiological characteristics (morphology,
tolerance to bile salt, heat and antibiotics and membrane fatty acid composition) in the effluent
broth of continuous IC cultures were measured and data were compared to that of stationary-
phase cells produced from traditional free cell batch fermentations. Growth medium not
supplemented with glucose induced starving conditions in both reactors and thus a second
continuous fermentation was carried out in growth medium supplemented with extra-glucose.
Chapter 2 - Immobilized cells
- 48 -
2.3 Material and Methods
2.3.1 Strain and medium
Bifidobacterium longum NCC2705 was provided by the Nestlé Culture Collection (NCC; Nestlé
Research Centre, Lausanne, Switzerland). Cells were grown under anaerobic conditions using
atmosphere generation system packs (AnaeroGen, Oxoid, Pratteln, Switzerland) at 37 °C in de
Man, Rogosa and Sharp medium (MRS; de Man et al., 1960) broth (BioLife, Milan, Italy)
supplemented with 0.5 % (w/v) L-cysteine (Sigma-Aldrich, Buchs, Switzerland) (MRSC) and
inoculated at 1 % (v/v) with an overnight pre-culture before use.
2.3.2 Cell immobilization
Gel beads were produced in a two phase dispersion process using gellan and xanthan gum
according to Cinquin et al. (2004). Briefly, a polysaccharide mix was prepared containing gelrite
gellan (2.5 % (w/v), xanthan (0.25 % (w/v)) powders and sodium citrate (0.2% (w/v)) (Sigma-
Aldrich) resuspended in 500 ml pre-heated distilled water (90 °C) and mixed in a blender until
dissolution. The polymer solution was then autoclaved for 15 min at 121 °C just prior to use for
immobilization. The autoclaved polymer solution was cooled to 43 °C and inoculated aseptically
with B. longum NCC2705 pre-culture (2%; v/v). Gel beads were formed using a 2 phase
dispersion process by pouring the polymer solution into sterile hydrophobic phase (commercial
sunflower oil) at 43 °C. Beads were hardened by cooling the suspension to 25 °C and by
replacing the hydrophobic phase with 0.1 M solution of CaCl2 (Sigma-Aldrich). Beads with
diameters in the 1.0–2.0 mm range were selected by wet sieving and used for fermentation. The
entire process was completed in aseptic conditions within 1 h.
Chapter 2 - Immobilized cells
- 49 -
2.3.3 Continuous fermentations
Two IC continuous cultures were performed for 20 days. Before starting the continuous
fermentation, a bead colonization step was carried out in batch mode in a 500 ml bioreactor
(Sixfors, Infors-HT, Bottmingen, Switzerland) containing 370 ml MRSC and 80 ml of inoculated
gel beads. The colonization step was carried out at 37 °C for 16 h with CO2 flushed in the
headspace to ensure anaerobic conditions and agitation rate at 150 rpm by an inclined flat blade
impeller. pH was maintained at 6.0 by addition of 5 M NaOH. Colonized gel beads were then
aseptically wet sieved and transferred to a 500 ml bioreactor (R1) (Sixfors) containing 160 ml
medium for continuous fermentation for a total volume of 360 ml. Flow rate of fresh medium
was set at 360 ml/h using peristaltic pumps (Ismatec Reglo, Ismatec, Glattbrugg, Switzerland)
during the 20 days of culture, corresponding to a dilution rate of 2.25 h-1. Stirring was set at 200
rpm. A second reactor (R2) (Bioengineering, Wald, Switzerland) with a working volume of 1800
ml (dilution rate 0.2 h-1) was operated in series with R1 and continuously inoculated by
fermented broth from R1 (Figure 6). Stirring with two rushton type impellers in R2 was set at 250
rpm. The reactor volumes were chosen to yield a shorter residence time in R1 than in R2 (ca. 27
min and 5 h, respectively) for a high cell productivity in R1 containing a high immobilized cell
concentration and operated at high dilution rate and final growth in R2 with longer residence time
(Doleyres et al., 2004b). The pH in both reactors was maintained at 6.0 by addition of 5 M
NaOH. Temperature was set at 37 °C and CO2 was flushed in the headspace of both reactors to
provide anaerobic conditions.
Fermented broth samples were taken daily from the two reactors for microscopy, dry biomass,
HPLC analyzes, and viable cell enumeration by plate counts. Samples for stress tolerance assays
were taken every 2 days, centrifuged (6,000 × g for 15 min at 4 °C), and pellets were resuspended
Chapter 2 - Immobilized cells
- 50 -
pH 6.0Temp. 37 °CVol. 1800 mlDil. rate 0.2 h -1
MRSCor
MRSC +40 g/l glucose
pH 6.0Temp. 37 °CVol. 240 mlDil. rate 2.25 h -1
pH 6.0Temp. 37 °CVol. 1800 mlDil. rate 0.2 h -1
MRSCor
MRSC +40 g/l glucose
pH 6.0Temp. 37 °CVol. 240 mlDil. rate 2.25 h -1
in an equal volume of MRSC containing 10 % (w/v) glycerol (Sigma-Aldrich) prior to freezing in
liquid nitrogen and storage at -80 °C. This step was required to test samples in blocks and correct
for assay variability. HPLC samples were prepared by centrifuging 2 ml samples (10,000 × g for
1 min) and 1.5 ml of the supernatant was stored at -20 °C until analysis.
The first culture (ICC1) was run with unsupplemented MRSC leading to complete sugar
consumption in R1 and R2 whereas the second culture (ICC2) was operated with MRSC
supplemented with glucose (final concentration of 60 g/l) and anhydrous CaCl2 at 1.5 g/l (Sigma-
Aldrich). CaCl2 was added to ensure gel bead-stability throughout the entire fermentation time
because gel beads appeared soaked and consistency was weak at the end of ICC1. After
sterilization, initial glucose concentration measured in fresh medium for ICC1 and ICC2 was
13.6 and 52.8 g/l, respectively.
Figure 6 Diagram of the two-stage reactor setup for continuous fermentation with
immobilized cells including tubings for feed inlet, reactor connection, medium outlet, emergency
out, sampling ports, NaOH addition (pH: 6.0) and CO2 injection for anaerobic conditions.
Chapter 2 - Immobilized cells
- 51 -
2.3.4 Batch fermentation
Batch fermentation was carried out in a glass bioreactor (Bioengineering) with a working volume
of 2000 ml similar to R2 of the continuous fermentation. Unsupplemented MRSC was inoculated
at 2 % (v/v) with twice sub-cultured B. longum NCC2705 and incubated for 16 h at 37 °C under
anaerobic conditions by flushing CO2 in the headspace. Mixing was set at 250 rpm and pH
controlled at 6.0 by addition of 5 M NaOH. Growth was monitored by OD600 and growth phases
were set as follows: end exponential (9.5 h, OD600 4.3 ± 0.1), early stationary (12.5 h, OD600 of
5.7 ± 0.4) and mid stationary growth (16 h, OD600 5.6 ± 0.4). Samples were collected and
proceeded as described above. Batch fermentations were carried out in triplicate.
2.3.5 Viable cells enumeration
Viable cells in the fermented broth were enumerated by serially diluting fresh effluent of both
reactors in 12 mM phosphate buffer saline with 0.05 % cysteine (w/v) (PBSC). Appropriate
dilutions were drop plated in duplicate on MRS agar (1.5 %) (Becton Dickinson, Basel,
Switzerland) supplemented with 0.05 % (w/v) cysteine. Plates were incubated anaerobically
using atmosphere generation system packs (AnaeroGen, Oxoid, Pratteln, Switzerland) at 37 °C
for 48 h.
Viable cell enumeration using qRT-PCR method was only performed for samples from R2 of
ICC2 and bile- and heat tests of the same samples by quantifying levels of a 400 bp mRNA
fragment of a housekeeping gene, purB, (see Chapter 3.3). This method was used to estimate
viable cell counts in aggregated cell samples. Briefly, 0.5 ml of washed cell samples were put in
2-mL screw-cap tubes containing a mix of glass beads, phenol/chloroform 1:1 (v/v), sodium
dodecyl sulfate 10 % (w/v) and 3 M sodium acetate (pH 5.2) (Sigma-Aldrich) and immediately
Chapter 2 - Immobilized cells
- 52 -
frozen in liquid nitrogen. Cells were disrupted by bead-beating in a homogenizer (FastPrep FP
120, Q-Biogene, Carlsbad CA, USA), four times for 40 sec at speed 4.0 intercalated with cooling
periods on ice. Samples were then centrifuged to remove cell debris and glass beads. The aqueous
phase containing RNA was transferred to a new tube and supplemented with a known amount of
exogenous luciferase mRNA (Promega, Dübendorf, Switzerland) as a reference. Traces of phenol
were removed by extraction with pre-chilled chloroform. The resulting aqueous phase was mixed
with an equal volume of 70%-ethanol and applied to RNA purification column using RNeasy
Mini Kit (Qiagen, Basel, Switzerland). Further purification was performed according to the
manufacturer’s protocol including on-column DNase I digestion for 30 min followed by elution
in 100 µl H2O. Purified RNA (max 100 ng) was applied to reverse transcription using the high
capacity cDNA reverse transcription commercial kit (Applied Biosystems, Rotkreuz,
Switzerland). cDNA was then applied to quantitative PCR using an ABI PRISM 7700 equipment
with SYBR® Green PCR Master Mix (Applied Biosystems). The normalized levels of the 400 bp
amplicon were then measured to determine viable cell concentrations along with prepared
calibration curves.
2.3.6 Dry biomass determination
Dry biomass was determined according to Bergmaier et al. 2005 with slight modifications. 6 ml
of thawed sample were centrifuged at 10,000 × g for 15 min at 4 °C. The pellet was washed twice
with distilled water and transferred to a pre-weighed aluminum dish. The cells were dried for 20
h in an oven at 99 °C and weighed after cooling to room temperature (25 °C) in an exsiccator
(Duran Group GmbH, Mainz, Germany). Values are single determinations.
Chapter 2 - Immobilized cells
- 53 -
2.3.7 Microscopic observation
Pictures of fresh cells from the fermentation were obtained using an optical microscopy at 1000-
fold magnification (Leica DM, Leica Microsystems, Wetzlar, Germany) by an attached camera
(Leica DFC280) without any further treatment.
2.3.8 Metabolite analysis
Consumption of glucose and concentration of main metabolites, lactic and acetic acids, were
determined by HPLC (Merck Hitachi, Darmstadt, Germany). Samples were diluted in MilliQ
water and filtered through a 0.45 µm nylon membrane filter (Infochroma AG, Zug, Switzerland).
Separation was carried out in an Aminex HPX-87-H column (Bio-Rad Laboratories, Reinach,
Switzerland) with 10 mM H2SO4 as mobile phase at a flow rate of 0.6 ml/min and a temperature
of 40 °C. Compounds were detected using a refractive index detector (Merck Hitachi, Darmstadt,
Germany). Reported data are means of duplicate analysis.
2.3.9 Fatty acid analysis
The fatty acid composition of cell membrane was determined by gas chromatography at the
DSMZ external service (German Collection of Microorganisms and Cell Cultures, Braunschweig,
Germany) using the Microbial Identification System (MIDI, Newark, DE; http://www.midi-
inc.com). Frozen cell samples of end exponential growth phase cells produced with batch cultures
and from R2 of ICC2 at day 4, 12 and 20 were thawed at room temperature and washed (6,000 ×
g for 3 min) with PBSC and pellets were frozen using liquid nitrogen. Fatty acids were converted
to methyl esters and extracted in a five-step procedure according to the DSMZ protocols.
Analyzes were performed in duplicate.
Chapter 2 - Immobilized cells
- 54 -
2.3.10 Heat test
Heat survival tests were carried out according to Simpson et al. (2005) with some modifications.
Briefly, frozen cell samples of ICC2 were thawed at room temperature and centrifuged at 6,000 ×
g for 2 minutes. Supernatant was discarded and pellet resuspended in the same volume of fresh
MRSC and incubated for 4 h at 37 °C to reactivate cells. This treatment was selected based on
preliminary screening tests to obtain sufficient cell activity for RNA degradation. Cell
suspensions were split in 1 ml aliquots and mixed with either 9 ml pre-warmed MRSC at 56 °C
or room temperature (control) and incubated in a water bath at the corresponding temperature for
30 min. After the stress, cells were immediately cooled to room temperature in a water bath for
three minutes, washed in fresh MRSC and pellets suspended in 1 ml MRSC. To quantify viable
cells, 30 µl of cell solution was applied to a dilution row in PBSC and appropriate dilutions were
plated on MRSC agar.
To enumerate viable cells from macroscopic aggregates containing samples with qRT-PCR, 0.5
ml of the suspended cells solution was shock frozen in liquid nitrogen and stored at -80 °C for
RNA-extraction and qRT-PCR analysis as described above. For all tests, controls consisted of
cells produced during batch cultures. Experiments were performed in duplicate.
2.3.11 Bile salt test
The effects of bile salts were tested as previously described by Saarela et al. (2004) with some
modifications. Samples were prepared as for the heat test. After a reactivation step at 37 °C for 4
h, cell samples from ICC1 were split in aliquots of 1 ml and immediately added to 9 ml 1.5 %
(w/v) porcine bile salt (Sigma-Aldrich) dissolved in PBSC or in 9 ml PBSC without bile salts and
incubated for 17 min at 37 °C. After the stress, cells were cooled to room temperature in a water
Chapter 2 - Immobilized cells
- 55 -
bath for three minutes and washed twice in fresh MRSC to remove bile salts. To quantify viable
cells, 30 µl of cell solution was applied to a dilution row in PBSC and appropriate dilutions were
plated on MRSC agar.
Bile tolerance of samples from ICC2 was evaluated as described above, but lethal bile treatment
was carried out with 1.0 % (w/v) bile solution for 12 min at 37 °C. Less restrictive stress
conditions were chosen, because the detection limit of the qRT-PCR method (ca. 4.4 × 104
CFU/ml) was higher than for plate counts and could not detect viable after bile stress at 1.5 %
(w/v) for 20 min of control cells. Survival was determined with standard plate counts and qRT-
PCR as for the heat test. Experiments were performed in duplicate.
2.3.12 Tolerance to antibiotics
Antibiograms were performed using an antimicrobial disc susceptibility assay according to the
National Commitee for Clinical Laboratory Standards (1991) with some modifications. Cell
samples were diluted in MRSC to an OD600 of 0.0125 and streaked out with a cotton tip on
MRSC agar plates. Cells from IC cultures were only tested before formation of macroscopic
aggregates because OD values of samples containing cell clusters could not be accurately
measured and were not representative of cell counts. Discs of 6 mm diameter containing
antibiotics were then placed on agar plates. The used antibiotic-discs were: Penicillin G (P 10),
Norfloxacine (NOR 10), Nalidixic acid (NA 30), Ampicilline (AM 10), Cefotaxime (CTX 30),
Erythromicin (E 15), Gentamicin (GM 10), Tetracycline (TE 30), Vancomicin (VA 30),
Chloramphenicol (C 30) (BioMérieux Suisse SA, Geneva, Switzerland) and Neomicin (N 10),
Bacitracin (B 10), Teicoplanin (TEC 30) (Oxoid AG, Switzerland)). Plates were incubated
Chapter 2 - Immobilized cells
- 56 -
anaerobically at 37 °C for 48 h and the inhibition zone was measured in mm. Experiments were
performed in duplicate.
2.3.13 Statistical analyzes
Means of log10-transformed cell counts for heat- and bile tests as well as inhibition diameters of
antibiograms and molecular acetate /lactate ratios were compared with Student’s t-test using JMP
In Software version 6.0 (SAS Institute Inc., Cary, NC) with a level of significance at 0.05.
2.4 Results
Two continuous IC cultures were carried out in a two-stage fermentation system (Figure 6) for 20
days and effluents of both fermentations and reactors were tested for cell morphology, cell and
metabolites concentration, and cell survival to lethal heat and bile salt stresses to assess the
effects of immobilization in combination with continuous culture time and different glucose
concentrations.
2.4.1 Biomass production
Continuous culture of immobilized B. longum NCC2705 induced formation of large cell
aggregates in both reactors of ICC1 after 12 days of fermentation and to an even higher extent in
ICC2 where macroscopic aggregates of ca. 0.5 – 1 mm diameter were observed after only 9 days
of continuous culture (Figure 7).
Chapter 2 - Immobilized cells
- 57 -
Figure 7 Optical micrographs of fresh cells collected from fermented broth (1000 ×
magnification) of B. longum NCC2705 produced in batch culture (A) and in ICC2 from day 8
(B), day 12 (C) and day 20 (D).
Free cell population for ICC1 with limiting glucose tested in the fermented broth medium with
plate counts were very similar in both reactors and decreased progressively with culture time
from 1.4 ± 0.1 × 109 to 3.4 ± 0.7 × 108 CFU/ml between day 2 and 20 (Figure 8). Cell dry weight
(CDW) of ICC1 also decreased progressively but to a lower extent with time in R1 from 1.9 at
day 4 to 1.3 g/l at day 20, but remained constant in R2 at 1.7 ± 0.2 g/l throughout the entire 20
days (Figure 8). High maximal volumetric productivities were reached after 2 days of culture,
with 6.2 × 1011 for R1 and 2.5 × 1011 CFU/l h for the two stage culture (Table 4).
In ICC2 operated with glucose supplementation, populations of both reactors tested with plate
counts showed large fluctuations with fermentation time. From day 1 to 7 cell counts increased
steadily from 1.2 × 107 to 1.0 × 109 and from 1.7 × 107 to 2.3 × 109 CFU/ml for R1 and R2,
respectively (Figure 8). After detection of macroscopic cell aggregates at day 9, cell counts
decreased continuously until the end of fermentation to 2.4 × 108 and 6.3 × 108 CFU/ml for R1
and R2, respectively (Figure 8). In contrast, viable cell concentration determined for R2 with
aggregation independent qRT-PCR method increased steadily from 1.1 ± 0.4 to 8.6 ± 4.4 × 109
CFU/ml between day 4 and 12 and remained stable afterwards (Figure 8 B). Similarly, CDW
increased steadily from 0.3 to 2.7 g/l and from 0.3 to 4.6 g/l in R1 and R2, between day 2 and 20,
respectively (Figure 8). CDW correlated well with viable cell counts tested with qRT-PCR but
Chapter 2 - Immobilized cells
- 58 -
not with plate counts. Maximal volumetric productivities reached 5.9 × 1011 for R1 and 4.2 × 1011
for the two stage culture determined by plate counts or 1.5 × 1012 CFU/l h for the two stage
culture determined with qRT-PCR method (Table 4).
The maximal cell concentration for batch cultures was 1.6 ± 1.1 × 109 CFU/ml after 12.5 h
growth in MRSC at an OD600 of 5.7 ± 0.4 (early stationary growth phase) corresponding to a
volumetric productivity of 1.2 × 1011 CFU/l h (Table 4).
Table 4 Maximal cell counts and volumetric productivities of B. longum NCC2705 in
different fermentation systems. Values obtained by plate counts or by qRT-PCR method.
Fermentation systems Fermentation time
Dilution rate (h-1)
Cell counts (CFU/ml)
max. vol. productivity (CFU/l h)
Batch 13.5 h - 1.6 × 109 1.2 × 1011
ICC1 R1 2 days 2.25 1.4 × 109 6.2 × 1011
ICC1 R1 + R2 2 days 0.18 1.4 × 109 2.5 × 1011
ICC2 R1 6 days 2.25 1.3 × 109 5.9 × 1011
ICC2 R1 + R2* 7 days 12 days (qRT-PCR)
0.18 2.3 × 109
(7.9 × 109) 4.2 × 1011
(1.5 × 1012)
Continuous IC1 n.i. 0.5 4.9 × 109 2.5 × 1012
Free cell continuous2 105 h 0.1 1.1 × 109 1.1 × 1011
* values in brackets from qRT-PCR 1 for B. longum ATCC15707 in Doleyres et al. (2002) 2 Mozzetti (2009)
Chapter 2 - Immobilized cells
- 59 -
Figure 8 Viable cell concentrations and cell dry weight (CDW) in samples from R1 (A) and
R2 (B) during continuous immobilized cell cultures. Cell concentrations determined by plate
counts (full line, □, ■) or by qRT-PCR (semi dashed line, ●) and cell dry weight (dashed line, ∆,
▲). Data from ICC1 (open symbols) and ICC2 (solid symbols) are means of two determinations.
Chapter 2 - Immobilized cells
- 60 -
2.4.2 Metabolic Activity
Glucose consumption in R1 of ICC1 was stable throughout the entire fermentation at 12.5 ± 0.2
g/l and resulted in almost no sugar in the effluent of R1 at a mean residual concentration of 1.1 ±
0.2 g/l (Figure 9). Lactic and acetic acid production were stable during 20 days, with 5.4 ± 0.4
and 6.2 ± 0.6 for R1 and 0.3 ± 05 and 0.8 ± 0.9 g/l for R2, respectively. Therefore the molar ratio
of main metabolites acetic and lactic acids in R1 remained also constant during the entire
fermentation at 1.72 ± 0.07. The low sugar concentration in the effluent from R1 of ICC1 caused
starving conditions in R2 and hence formation of metabolites was not relevant.
In ICC2, glucose consumption varied with time in both reactors. In R1, glucose consumption
increased steadily from 16.0 ± 1.3 at day 6 to 39.7 ± 0.2 g/l at day 13 after an adaptation phase
during the first five days. Afterwards, consumption decreased to reach 29.9 ± 0.6 g/l at day 20. In
contrast, glucose consumption in R2 increased steadily between day 1 and 7 from 1.8 ± 0.0 to
23.0 ± 0.5 g/l, then decreased sharply to 4.1 ± 0.2 g/l at day 12. From day 13 until the end of
fermentation, consumption increased again to reach 19.5 ± 0.0 g/l. Residual glucose
concentrations in ICC2 were never lower than 13.1 ± 0.2 and 1.8 ± 0.1 in R1 and R2,
respectively, indicating no sugar limitation. The acetic and lactic acid production in R1 increased
from 7.0 ± 0.1 and 5.0 ± 0.1 at day 3 to 15.3 ± 0.2 and 14.3 ± 0.3, respectively, and then
decreased until the end of fermentation to 11.4 ± 0.4 and 11.0 ± 0.4 g/l, respectively. The
production of acetic and lactic acid in R2 varied like the sugar consumption and ranged between
a maximum of 9.4 ± 0.9 and 7.9 ± 0.8 at day 7 and a minimal value of 1.7 ± 1.1 and 0.9 ± 1.0 g/l
at day 12, respectively. The molar ratio of acetic to lactic acids showed a slight but significant
decrease in R1 after an adaptation phase in the first 3 days, from 1.75 at day 4 to 1.56 at day 20.
Chapter 2 - Immobilized cells
- 61 -
R2 showed large fluctuations over time in the range of 1.58 to 2.85, mirroring glucose
consumption (Figure 9 B).
Glucose in batch culture was entirely consumed when cells reached early stationary growth phase
after 12.5 h growth. The levels of acetic and lactic acid reached the maximum at mid stationary
growth phase (16 h) with 7.3 ± 0.0 and 5.6 ± 0.0 g/l, respectively, corresponding to a molar ratio
of acetic to lactic acids of 1.93.
Chapter 2 - Immobilized cells
- 62 -
Figure 9 Consumption of glucose and molar ratio of acetic to lactic acid in samples from
R1 (A) and R2 (B) during continuous immobilized cell cultures. Glucose consumption (full line;
□, ■) and molar ratio of acetate and lactate (dashed line; ∆, ▲). Data from ICC1 (open symbols)
and ICC2 (solid symbols) are means of two determinations.
Chapter 2 - Immobilized cells
- 63 -
2.4.3 Cell sensitivity to heat and porcine bile sal ts
Cell tolerance to porcine bile salt stress was first tested by exposing cells from ICC1 and control
batch for 20 min to 1.5 % (w/v) porcine bile salt solution and measured by plate count
enumeration. Samples from ICC1 at day 4, 12 and 20 showed significantly less loss of viable
cells of 3.5 ± 0.4, 2.8 ± 0.1 and 3.0 ± 0.1 for R1 and 2.6 ± 0.3, 2.6 ± 0.4 and 1.8 ± 0.2 log10
CFU/ml for R2, respectively, compared to batch stationary phase cells with 5.7 ± 1.2 log10 loss of
CFU/ml (Table 5). Indeed, the viable cell counts of batch control samples after exposure to 1.5 %
(w/v) bile salt solution were too low to be detected with the qRT-PCR method (detection limit ca.
4.5 × 104 CFU/ml) and hence the conditions for the bile assay were set at 1.0 % (w/v) bile salt
solution to test samples from ICC2 and batch control cells. Viability loss of ICC2 cells was tested
for cells from R2 at day 4, 12 and 20, using plate counts and qRT-PCR method. The bile assay
was carried out for 15 min at 1.0 % (w/v) bile salt solution, Viability loss of cell samples
evaluated with plate counts was 3.3 ± 0.0, 1.7 ± 0.1, 3.0 ± 0.2 log10 CFU/ml for cells from R2 day
4, 12, 20 and significantly smaller than 4.0 ± 0.2 log10 CFU/ml of batch control cells. When
tested with qRT-PCR, viability losses for cells from day 4 and 20 were 0.6 ± 0.2 and 0.5 ± 0.2
log10 CFU/ml, respectively, and also significantly smaller than 1.0 ± 0.1 log10 CFU/ml for batch
control cells, whereas no significant difference was detected for cells from day 12 (Table 5).
Heat tolerance was tested by heating cell samples from ICC2 R2 at 56 °C for 30 min. Viability
losses tested with plate counts of cells collected at day 4, 12 and 20 range between 4.1 – 3.4 log10
CFU/ml and were not significantly different to 3.4 ± 0.4 log10 CFU/ml of batch control cells.
Similarly, no significant differences between cells from continuous and batch were observed
using qRT-PCR method (Table 5).
Chapter 2 - Immobilized cells
- 64 -
Table 5 Cell viability loss after exposure to porcine bile salt and heat lethal stresses of
cells from batch, and continuous immobilized cell cultures at day 4, 12 and 20. Values are given
as loss of log10 (CFU/ml) determined by plate counts or qRT-PCR. Data are means of two
independent repetitions.
Treatment Detection method
Batch ICC1*
day 4 Day 12 Day 20
20 min in 1.5 % porcine bile
Plate counts
5.7 ± 1.2a
R1 3.5 ± 0.4b R2 2.6 ± 0.3b
2.8 ± 0.1b 2.6 ± 0.4b
3.0 ± 0.1b 1.8 ± 0.2C
Batch ICC2 (R2)
day 4 Day 12 Day 20
30 min at 56°C Plate counts
qRT-PCR
3.4 ± 0.4a
2.3 ± 0.3a
4.1 ± 0.1b
2.5 ± 0.2a
3.1 ± 0.3a
2.0 ± 0.3a
3.4 ± 0.0ab
2.5 ± 0.3a
15 min in 1 % porcine bile §
Plate counts
qRT-PCR
4.0 ± 0.2a
1.0 ± 0.1a
3.3 ± 0.0b
0.6 ± 0.2b
1.7 ± 0.1c
0.7 ± 0.1ab
3.0 ± 0.2b
0.5 ± 0.2b
Different letters on each line (p ≤ 0.05). * Values from R1 and R2
2.4.4 Tolerance to antibiotics
Tolerances to antibiotics were tested using disc-assay and only applied to samples from
continuous cultures before formation of aggregates. No inhibition zone was observed for cells
from ICC1 and 2 and batch cultures to norfloxacine and nalidixic acid (Table 6). Cells from ICC1
collected at day 2 and 10 showed significant increased tolerance to aminoglycosidic gentamicin
and neomycin (p ≤ 0.05) compared to batch control cells, and a slight but not significant
increased tolerance with culture time for both antibiotics (Table 6).
A significant increase in tolerance of cells from ICC2 was detected between day 2 and 8 for
penicillin, erythromycin, gentamicin, tetracycline, vancomycin, neomycin, bacitracin and
Chapter 2 - Immobilized cells
- 65 -
teicoplanin. Furthermore, cells collected at day 8 showed significantly smaller diameters (p ≤
0.05) for gentamicin and neomycin, compared to cells from control batch (Table 6).
Table 6 Inhibition diameters measured by the disc assay method of different antibiotics on
cells from control batch cultures and produced by continuous immobilized cell cultures ICC1 and
ICC 2 tested at different days of fermentation before formation of macroscopic cell aggregates.
Values are means of two repetitions.
Antibiotics Batch ICC1 ICC2
Day 2 Day 10 Day 2 Day 8
Penicillin G [10] 42.0 ± 1.3 a - - 52.5 ± 0.7b 47.5 ± 0.7c Norfloxocine [10] 6.0 ± 0.0* a - - 6.0 ± 0.0* a 6.0 ± 0.0* a Nalidixic acid [30] 6.0 ± 0.0* a - - 6.0 ±0.0* a 6.0 ± 0.0* a Amplicillin [10] 41.0 ± 2.3 a - - 48.0 ± 2.8b 42.5 ± 0.7ab Cefotaxime [30] 41.0 ± 2.9 a 40.0 ± 1.3 a 38.8 ± 1.0a 45.5 ± 3.5b 37.5 ± 4.9a Erythromycin [15] 45.0 ± 1.4 a - - 51.5 ± 2.1b 44.5 ± 0.7a Gentamicin [10] 16.0 ± 1.3 a 10.2 ± 2.4b 9.3 ± 2.3b 23.0 ± 1.4c 8.0 ± 2.8b Tetracycline [30] 41.0 ± 1.3 a - - 46.5 ± 3.5b 41.0 ± 0.0a Vancomycin [30] 31.0 ± 1.0 a - - 38.5 ± 0.7b 32.0 ± 0.0a Chloramph. [30] 44.0 ± 1.3 a - - 50.0 ± 4.2b 46.0 ± 0.0ab Neomycin[10] 14.0 ± 1.3a 9.4 ± 1.7b 8.6 ± 1.5b 16.5 ± 0.7a 6.0 ± 0.0b Bacitracin[10] 23.0 ± 2.2 a - - 31.0 ± 2.8b 24.0 ± 0.0a Teicoplanin [30] 30.0 ± 0.0 a - - 36.5 ± 2.1b 29.0 ± 1.4a *6 mm diameter refers to no inhibition (diameter of antibiotic disc). - : not tested Different letters on each line (p ≤ 0.05).
2.4.5 Cell membrane fatty acid composition
Fatty acid composition of membranes of cells produced in ICC2 exhibited no differences between
R1 and R2 (data not shown) but showed large fluctuations compared to membranes of batch
culture cells (Table 7). Major changes in fatty acid composition were observed in saturated 14 : 0
and 16 : 0, as well as in unsaturated 18 : 1w9c fatty acids which account for more than 60 % of
the total fatty acids of membranes from cells of all tested samples (Table 7). The ratio of
Chapter 2 - Immobilized cells
- 66 -
unsaturated to saturated fatty acids (USFA/SFA) of cells collected during continuous IC culture
was 0.58 ± 0.09 and significantly lower than 1.74 ± 0.00 for cells from free cells batch culture.
Table 7 Fatty acid membrane composition in % of total fatty acids of cells from ICC2
collected at day 4, 12, 20 and batch culture cells. Data are means of two determinations.
* Fatty acids also including DMA forms. n.d., not detected; USFA, unsaturated fatty acids; SFA, saturated fatty acids; BCFA, branched chain fatty acids; DMA, dimethyl acetal fatty acids
Fatty acid Batch ICC2 R2
Day 4 Day 12 Day 20
12:0 0.33 ± 0.01 0.77 ± 0.09 0.40 ± 0.06 0.66 ± 0.04
14:0* 2.91 ± 0.04 14.36 ± 0.82 10.38 ± 0.92 13.30 ± 1.89
16:0 17.57 ± 0.08 27.73 ± 2.81 29.74 ± 1.15 32.72 ± 0.29
18:0 10.16 ± 0.06 6.94 ± 0.51 9.14 ± 0.12 6.02 ± 0.70
19: 0 cyclo* 1.95 ± 0.00 10.34 ± 0.39 5.65 ± 0.14 3.02 ± 0.72
19 cycloprop 1.36 ± 0.04 6.01 ± 0.90 1.89 ± 0.25 4.20 ± 0.48
13:1 none 0.21 ± 0.01 3.25 ± 0.43 1.91 ± 0.09 1.46 ± 0.28
16:1 w9c 0.42 ± 0.01 1.60 ± 0.02 0.80 ± 0.02 1.37 ± 0.01
17:1w9c 5.59 ± 0.10 1.13 ± 0.04 2.98 ± 0.05 1.93 ± 0.27
18.1 w9c* 50.25 ± 0.13 18.89 ± 0.43 29.02 ± 0.15 29.30 ± 2.82
18:1w7c* 5.70 ± 0.03 1.71 ± 1.55 1.76 ± 0.14 2.07 ± 0.42
18:1w6c n.d. 3.48 ± 0.83 1.54 ± 0.11 1.32 ± 0.05
18:0 OH 5.10 ± 0.03 2.91 ± 1.20 0.95 ± 0.40 0.91 ± 0.27
Total BCFA (ISO) 1.46 ± 0.01 0.73 ± 0.23 0.44 ± 0.03 0.57 ± 0.09
Total DMA 32.63 ± 0.01 25.15 ± 0.89 28.61 ± 0.99 20.26 ± 3.92
USFA 62.17 ± 0.08 30.05 ± 1.33 37.99 ± 0.09 37.11 ± 2.43
SFA 35.72 ± 0.07 66.88 ± 0.73 57.41 ± 0.42 60.49 ± 1.22
USFA/SFA 1.74 ± 0.00 0.45 ± 0.02 0.66 ± 0.01 0.61 ± 0.05
Labeled 97.89 ± 0.15 96.93 ± 0.63 95.41 ± 0.35 97.60 ± 1.39
Not labeled 2.12 ± 0.15 3.07 ± 0.63 4.60 ± 0.35 2.40 ± 1.39
Chapter 2 - Immobilized cells
- 67 -
2.5 Discussion
Immobilized cell technology (ICT) in combination with continuous culture has previously been
suggested for the production of starter or probiotic cultures with enhanced functional and
technological properties and high volumetric productivities (Doleyres and Lacroix, 2005). In the
present study, two continuous IC cultures were carried out for 20 days with and without glucose
limiting conditions. The supplementation of additional glucose resulted in a high maximal cell
production of 7.9 × 109 CFU/ml and volumetric productivity of 1.5 × 1012 CFU/l h for the two
stage culture measured with qRT-PCR was reached in ICC2 at day 12. This value is similar to
that reported by Doleyres et al. 2002 during continuous culture with immobilized B. longum
ATCC15707 in MRS medium supplemented with whey permeate at a dilution rate of 0.5 h-1 and
approximately ten fold higher than that of continuous culture with free cells of B. longum
NCC2705 in MRSC operated at a dilution rate of 0.1 h-1 as reported in a by Mozzetti (2009) and
controlled batch cultures (Table 4).
Formation of large cell aggregates was observed in both continuous cultures leading to
considerable underestimation of cell concentration with plate counts. Indeed, accurate viable cell
enumeration of B. longum NCC2705 using the recently developed qRT-PCR method which is
independent of cell aggregates (see Chapter 3) detected higher viable cell counts compared to
traditional plate counts. Extensive formation of cell aggregates has already been reported during
continuous cultures with immobilized lactobacilli (Bergmaier et al., 2005; Koch, 2006), whereas
no cell aggregates were reported for the B. longum NCC2705 grown for more than 8 days in
continuous mode with free cells in MRS (Mozzetti, 2009). Aggregation of bacterial cells can be
induced by changing environmental factors for instance pH, level of divalent cations in the
Chapter 2 - Immobilized cells
- 68 -
medium and/or nutrients availability (Koop et al., 1989; Canzi et al., 2005, De Schryver et al.,
2008).
In this study, formation and variation of size of cell aggregates in both IC cultures may have
different origins. Although the pH in the bioreactors was constant at 6.0 during all fermentations
(ICC and batch), the effect of acidic conditions in gel beads could be involved in cell
aggregation. Indeed, steep pH gradients can develop within gel beads (Masson et al., 1994;
Schepers et al., 2006). Larger size of aggregates in ICC2 may be a result of the supplementation
of growth medium with CaCl2 because Ca2+ ions in the growth medium were previously shown to
induce auto-aggregation in bifidobacteria (Ibrahim et al., 2005). Divalent cations may act as a
bridging compound for negatively charged functional groups on the cell surface and may promote
adhesion between bacterial cells (Higgins and Novak, 1997). Another factor affecting size of
aggregates is the supplementation of growth medium with additional glucose. Higher sugar
concentration in fermentation broth yielded higher biomass concentration and consequently more
cell growth and larger sized cell aggregates. The sugar limiting conditions of ICC1 may also
explain a metabolic shift towards production of acetic acid resulting in elevated molecular ratio
of acetic and lactic acid. Enhanced acetic acid production was shown to improve the ATP-yield
in B. animalis (Ruas-Madiedo et al., 2005), thereby generating higher energy resources for
starving cells from equal amount of glucose. The molar ratio in batch cultures was higher than in
continuous cultures, but comparison of the ratio is limited due to metabolite accumulation in
batch fermentation systems.
Physicochemical characteristics of the cell surface such as hydrophobicity may also affect auto-
aggregation and adhesion of bacterial cells to different surfaces (Perez et al., 1998; Del Re et al.,
2000; Zavaglia et al., 2002). Changes in membrane characteristics upon growth in adverse micro-
environment created within gel structures were already reported for Lactococcus lactis ssp. lactis
Chapter 2 - Immobilized cells
- 69 -
SL03 and Lactococcus lactis ssp. cremoris SC09 (Trauth et al., 2001) or resulted in lower
USFA/SFA ratio in fatty acid composition of membranes of immobilized Saccharomyces
cerevisiae (Jirku et al., 1999). In the present study, immobilized B. longum exhibited notable
modifications in fatty acid profiles (e.g. lower USFA/SFA ratio) of cell membranes compared to
free cells grown in batch cultures. Alterations in membrane composition may occur as a result of
an adaptation to changing environment and can lead to changes in membrane hydrophobicity as
shown for starved or acid-adapted Listeria innocua cells (Moorman et al., 2008).
The observed changes in membrane composition of cells released from gel beads may also play
an important role in their improved resistance to bile salts and the aminoglycosidic antibiotics
gentamycin and neomycin. A similar decrease of the ratio of unsaturated to saturated fatty acids
of cell membrane from IC cultures was reported for a bile adapted mutant strain of B. animalis
compared to its mother strain (Ruiz et al., 2007). These changes might be involved in restricted
diffusion of bile salts into the cytoplasm (Begley et al., 2005). The higher cell loss tested with
aggregated cells and plate counts compared to qRT-PCR were not expected because in theory one
viable cell in aggregate can give a colony, leading to an overestimation of cell survival after the
bile test. The observed difference in viability-loss between plate count and qRT-PCR method can
be a result of a bile induced sublethal injury within the Bifidobacterium population, possibly
through a reversible and transient membrane permeabilization which resulted in a loss of
viability, as defined by plate counts, but these cells could regain growth after being sorted and
resuscitated as reported by Ben-Amor et al. (2002).
Modifications in the membrane properties might also prevent passive diffusion of
aminoglycosides into the cell (Bryan et al., 1984). The integrity of membranes of B. longum was
shown to be involved in the tolerance to aminoglycosides, where susceptibility to
Chapter 2 - Immobilized cells
- 70 -
aminoglycosides increased when cells were previously subjected to membrane corrupting oxgall
or H2O2 stress (Kheadr et al., 2007).
Doleyres et al. (2004a) reported a progressive increase of survival of B. longum ATCC 15707 to
simulated intestinal conditions and tolerance to nisin Z with fermentation time induced by cell
immobilization in continuous co-culture with Lc. lactis MD. The increase of bile tolerance with
time correlated with increased tolerance to nisin Z, which targets bacterial cell membranes. These
findings support the observations in this study of increased tolerance to bile salts due to
membrane modifications induced by IC and continuous culture. The effect of culture age on
stress tolerance can be strain dependent and/or may originate from an adaptation process with
fermentation time of B. longum to restrictive environments produced by the competitive Lc. lactis
strain (Doleyres et al., 2004a). In a recent study, a co-culture of B. longum and B. breve was
cultivated in a compartmentalized system where the cells grew separately but the supernatants
were mixed together (Ruiz et al., 2009). The authors reported an increased expression of stress
response proteins such as chaperones in B. breve which may protect bacterial cells from
environmental stresses.
Cells from ICC2 did not exhibit detectable changes in tolerance to heat compared to batch cells.
Although, changes in membrane composition can influence bacterial membrane fluidity and
hence cell tolerance to heat (Guillot et al., 2000; Denich et al., 2003), the increase in shorter fatty
acids (acyl-chains of less than 18 carbons) and the decrease of unsaturated fatty acids in
membranes of cells from ICC2 did not affect survival to lethal heat stress. It is noteworthy to
mention that preliminary results from gene expression analyzes of cells from ICC2 compared
with control batch cells showed that genes coding for stress related proteins were not
differentially expressed as reported for heat stressed B. longum NCC2705 cells (Rezzonico et al.,
2007).
Chapter 2 - Immobilized cells
- 71 -
The use of aggregated cells in probiotic food industries is still unfavorable in product applications
due to underestimation of viable cell counts with standard plate counts that is the standard
method to verify product quality (see Chapter 3). However, the ability to form aggregates could
be a beneficial functionality for probiotic strains. Indeed, aggregation ability of Bifidobacterium
as well as Lactobacillus cells has been associated with enhanced adhesion to epithelial cells from
human or pigs (Del Re et al., 2000; Kos et al., 2003), coaggregation with pathogenic bacteria
(Schachtsiek et al., 2004; Collado et al., 2007) and improved protective effect against colitis in
mice (Castagliuolo et al., 2005). Further experiments are required to test whether observed
morphological and physiological changes of cells from IC cultures affect survival during GIT
transit and adhesion-ability to epithelial cells.
Chapter 3 - qRT-PCR
- 72 -
3 Development of a real-time RT-PCR method for
enumeration of viable Bifidobacterium longum cells in
different morphologies
Data presented in this chapter were published in 2009 in Food Micriobiology with the following authors: Reimann S., Grattepanche F., Rezzonico E., and Lacroix C.
Chapter 3 - qRT-PCR
- 73 -
3.1 Abstract
Viability of probiotic bacteria is traditionally assessed by plate counting which has several
limitations, including underestimation of cells in aggregates or chains morphology. We describe a
quantitative PCR (qPCR)-based method for an accurate enumeration of viable cells of
Bifidobacterium longum NCC2705 exhibiting different morphologies by measuring the mRNA
levels of cysB and purB, two constitutively expressed housekeeping genes. Three primer-sets
targeting short fragments of 57-bp of cysS and purB and one 400-bp fragment of purB were used.
Cell quantification of serially diluted samples showed a good correlation coefficient of R2 0.984
± 0.003 between plate counts and qRT-PCR for all tested primer sets. Loss of viable cells
exposed to a lethal heat stress (56 °C, 10, 20 and 30 min) was estimated by qRT-PCR and plate
count. No significant difference was observed using qRT-PCR targeting the 400-bp fragment of
purB compared to plate count indicating that this fragment is a suitable marker of cell viability. In
contrast, the use of the 57-bp fragments led to a significant overestimation of viable cell counts
(18 ± 3 and 7 ± 2 fold for cysB and purB, respectively). Decay of the mRNA fragments was
studied by treatment of growing cells with rifampicin prior qRT-PCR. The 400-bp fragment of
purB was faster degraded than the 57-bp fragments of cysB and purB. The 400-bp fragment of
purB was further used to enumerate viable cells in aggregate state. Cell counts were more than 2
log10 higher using the qRT-PCR method compared to plate count.
Growing interest in probiotic characteristics of aggregating bacteria cells make this technique a
valuable tool to accurately quantify viable probiotic bacteria exhibiting heterogeneous
morphology.
Chapter 3 - qRT-PCR
- 74 -
3.2 Introduction
Probiotics are defined as “Live microorganisms which when administered in adequate amounts
confer a health benefit on the host” (FAO/WHO 2001). It is generally accepted that probiotic
food products should contain a minimal level of viable cells of 106 per gram or milliliter of
product (Ouwehand et al., 1998), although this value is relative since beneficial effect depends on
the strain and targeted health benefit (Lacroix and Yildirim, 2007). Viability of probiotic bacteria
is traditionally assessed by plate counting which has several limitations, including
underestimation of cells in aggregates or chains morphology. Auto-aggregation trait of
Bifidobacterium longum strains isolated from human gastric juice and intestine has been
associated with their ability to adhere to epithelial cells (Del Re et al., 2000) which is considered
as an important functionality of probiotic bacteria for some beneficial health effects such as to
prevent attachment of pathogens to gut mucosa (Resta-Lenert and Barrett, 2003; Cesena et al.,
2001; Castagliuolo et al., 2005).
However, bacterial cell aggregates lead to underestimation of cell counts with traditional, culture-
dependent methods (Davey et al., 1977; Auty et al., 2001). Enumeration techniques with
fluorescent stains targeting bacterial membrane potentials or enzymatic activities were proposed
as alternative methods to plate counts (Breeuwer and Abee, 2000; Hewitt and Nebe-Von-Caron,
2001; Alakomi et al., 2005). The accuracy of these methods using fluorescence probes, generally
combined with flow cytometry can be affected by clustering and chaining of cells (Wallner et al.,
1995; Nebe-von-Caron et al., 2000; Bibiloni et al., 2001).
The use of real-time PCR techniques to quantify bifidobacteria and lactic acid bacteria by
specific DNA or rRNA targeting showed good correlation with plate counts (Masco et al., 2007;
Friedrich and Lenke, 2006; Matsuda et al., 2007). However, their use as accurate viability
Chapter 3 - qRT-PCR
- 75 -
markers is limited since rRNA and DNA can be detected hours or even days after cell death
(TolkerNielsen et al., 1997; Sheridan et al., 1998; Hellyer et al., 1999; Aellen et al., 2006; Keer
and Birch, 2003; Lahtinen et al., 2008). Molecules of mRNA in contrast have short half-life
times ranging between 0.5 and 50 min (Belasco et al., 1986; Hellyer et al., 1999; Takayama and
Kjelleberg, 2000) and are therefore a more suitable viability marker for bacterial cells. Yeasts and
pathogenic bacteria were quantified using mRNA in combination with quantitative reverse
transcription PCR (qRT-PCR) (Bleve et al., 2003; Jacobsen and Holben, 2007; Coutard et al.,
2007; Gonzalez-Escalona et al., 2009). The use of instable mRNA as target molecule however
requires special attention to minimize losses during RNA extraction and purification (Bustin
2002, Johnson et al., 2005).
In this study, we describe the development of a qRT-PCR-based method for the quantification of
viable B. longum NCC2705 cells. Serial dilutions of bacterial culture aliquots were used to
generate qRT-PCR calibration curves targeting three mRNA fragments (two short amplicons of
57-bp of cysS or purB and one of 400-bp of purB) and plotted against plate counts. Spiked
internal reference mRNA was used to monitor mRNA loss during sample processing and thus to
provide accurate mRNA comparison across different experimental conditions (Johnson et al.,
2005). Stability of the three mRNA fragments was tested by comparing cell counts obtained by
qRT-PCR and plate counts of culture exposed to lethal heat stress. Instability of the different
mRNA fragments was also assessed by measuring mRNA decay after rifampicin treatment. The
mRNA fragment selected was then used to estimate viable cell numbers in culture samples
containing macroscopic cell aggregates.
Chapter 3 - qRT-PCR
- 76 -
3.3 Materials and Methods
3.3.1 Strain and growth conditions
Bifidobacterium longum NCC2705 (Nestlé Culture Collection, Nestlé Research Center,
Lausanne, Switzerland) was grown at 37 °C in rubber-lid flasks containing 100 ml MRS broth
(de Man et al., 1960) (BioLife, Milan, Italy) supplemented with 0.5 g/l L-cysteine (Sigma-
Aldrich, Buchs, Switzerland) (MRSC) and inoculated at 2 % (v/v) with an overnight MRSC pre-
culture. The headspace of flasks was flushed with CO2 to ensure anaerobic conditions. Colony
forming units (CFU) were determined by plating serial dilutions of culture aliquots in duplicate
on MRSC agar plates 1.5 % (w/v) (Becton Dickinson, Basel, Switzerland) and incubating at
37 °C for 48 h in anaerobic jars containing atmosphere generation system packs (AnaeroGen,
Oxoid, Pratteln, Switzerland).
3.3.2 Preparation of bacterial samples for calibrat ion curves
Cultures were grown at 37 °C in a 2000 ml reactor (Bioengineering, Wald, Switzerland)
containing 1800 ml MRSC inoculated at 0.5% (v/v) with pre-culture. pH was maintained at 6.0
by addition of NaOH (5 M), agitation rate was set at 200 rpm using a flat blade impeller and
anaerobic conditions were maintained by flushing CO2 in the head space of the reactor. Cells
were harvested in end-exponential growth phase and centrifuged at 6,000 × g for 2 min. Pellets
were suspended in an equal volume of a cryoprotectant solution composed of MRSC
supplemented with glycerol 10% (w/v) (Fluka, Buchs, Switzerland), shock frozen in liquid
nitrogen and stored at -80 °C before RNA extraction and plate counts. Standards with cell
concentrations ranging from about 3.2 × 104 to 4.7 × 108 CFU/ml determined by plate counts
were prepared by serial dilution in MRSC before RNA extraction.
Chapter 3 - qRT-PCR
- 77 -
3.3.3 Production of aggregated cells
B. longum NCC2705 cells aggregates were produced during continuous culture with immobilized
cells grown in a two-stage reactor system as described by Doleyres et al. (2004b), with some
modifications. Briefly, the first reactor (R1) was a 500 ml glass reactor (Sixfors, Infors-HT,
Bottmingen, Switzerland) with a working volume of 240 ml, containing 80 ml of gel beads pre-
colonized with B. longum NCC2705 as described by Cinquin et al. (2004) and stirred between
150 – 250 rpm. The second 2000 ml reactor (Bioengineering) was run in series at a volume of
1800 ml, agitated at 250 rpm and inoculated with cells produced in R1. CO2 was injected in the
headspace of the two reactors to maintain anaerobic conditions during culture. The continuous
system was fed with MRSC medium supplemented with 1.5 g/l CaCl2 (Sigma-Aldrich) and 40
g/l glucose at a flow rate of 360 ml/h. Continuous fermentation was run for 20 days at 37 °C and
pH was maintained at pH 6.0 by addition of NaOH (5 M) in both reactors. Culture samples from
the effluent of both reactors were periodically collected and processed as described above.
3.3.4 Heat stress
A 100-ml volume of MRSC was inoculated at 1% (v/v) with B. longum NCC2705 and incubated
in a rubber-lid flask at 37 °C for 16 h. Cells were collected by centrifugation at 6,000 × g for 2
min. Supernatant was discarded and pellet suspended in the same volume of fresh MRSC. Cell
suspensions were split in aliquots of 1 ml and immediately added to 9 ml MRSC either pre-
warmed at 56 °C for 10, 20 and 30 min or kept at room temperature. After heating, cells were
immediately cooled to room temperature in a waterbath for 3 min, collected by centrifugation,
washed and suspended in the initial volume of MRSC for plate counting. 0.5 ml of the cell
Chapter 3 - qRT-PCR
- 78 -
suspension was added to prepared screw cap tubes as described below and immediately frozen in
liquid nitrogen for RNA extraction. Experiments were performed in triplicate.
3.3.5 Rifampicin test
MRSC (100 ml) in a rubber-lid flask was inoculated at 5% (v/v) with preculture and incubated at
37 °C until the culture reached mid-exponential phase (OD600 0.5 – 0.8). Rifampicin (Sigma-
Aldrich) was then added to a final concentration of 400 µg/ml. Samples were taken 2.5, 7.5, 14,
20, 25 and 45 min after rifampicin-addition. Cells were collected by centrifugation, washed in
fresh MRSC and suspended in the initial volume of MRSC for RNA extraction. The half-life time
(T½) of an mRNA was determined from the degradation rate constant (k) corresponding to the
slope of a semi-logarithmic plot of mRNA amount as a function of time with the relation: T½ =
ln2/k. Experiments were performed in duplicate.
3.3.6 RNA isolation and reverse transcription react ion
RNA extraction was performed according to Stevens et al. (2008). Briefly, 0.5 ml of fresh or
stored cell samples was centrifuged and the pellet was washed once in MRSC, transferred to 2-ml
screw-cap tubes containing a mix of 500 mg glass beads of 100 µm diameter (Sartorius,
Goettingen, Germany), 500 µl of a 1:1 solution of phenol/chloroform (Sigma-Aldrich), 30 µl of
10% (w/v) sodium dodecyl sulfate (Sigma-Aldrich) and 30 µl of 3 M sodium acetate (pH 5.2) and
immediately frozen in liquid nitrogen. Frozen cells were then disrupted by bead-beating in a
homogenizer (FastPrep FP 120, Q-Biogene, Carlsbad CA, USA), four times for 40 sec at speed
4.0 intercalated with cooling periods on ice. Afterwards, the mixtures were centrifuged (21,000 ×
g, 3 min at 4 °C) to remove glass beads and cell debris. The aqueous phase was transferred to a
Chapter 3 - qRT-PCR
- 79 -
new tube and traces of phenol were removed by extraction with pre-chilled chloroform. The
resulting aqueous phase was mixed with an equal volume of 70%-ethanol and applied to RNA
purification column using RNeasy Mini Kit (Qiagen, Basel, Switzerland). Further purification
was performed according to the manufacturer’s protocol including on-column DNase I digestion
for 30 min followed by elution in 100 µl H2O. Total RNA quantity and purity was determined
using Nanodrop ND-1000 (Peqlab, Erlangen, Germany).
Reverse transcription (RT) was performed using the high capacity cDNA reverse transcription
commercial kit (Applied Biosystems, Rotkreuz, Switzerland) according to manufacturer’s
protocol. A maximum amount of 100 ng of purified total RNA was applied per RT reaction.
Reaction was carried out in a thermocycler (Biometra T-personal, Göttingen, Germany) using the
following cycling conditions: 10 min at 25 °C, 120 min at 37 °C, 5 sec at 85 °C followed by a
cooling step to 4 °C.
3.3.7 Determination of RNA recovery rate using inte rnal reference
Luciferase mRNA (Promega, Dübendorf, Switzerland) was used as internal reference to estimate
loss of RNA during extraction and purification procedures. 2.03 × 108 copies of luciferase mRNA
were added to the aqueous phase, after centrifugation of lysed cells. Copy numbers of luciferase
mRNA added per volume were calculated using transcript sizes of 1.8 kb and an average
molecular mass of 330 Da/nucleotide. The number of luciferase mRNA copies recovered from
spiked samples after RNA extraction was determined with qRT-PCR and standard curves
prepared from serial dilutions of luciferase mRNA. Recovery rate was evaluated by dividing the
measured level by the initial amount of spiked reference mRNA and used as normalization factor
to correct for loss of mRNA during the RNA isolation procedure.
Chapter 3 - qRT-PCR
- 80 -
3.3.8 Quantitative PCR amplification
Primers used for quantitative PCR amplification (Table 7) were designed with PrimerExpress
software Version 3.0 (Applied Biosystems). Genes cysS and purB were selected for their
constitutive expression under different culture conditions (Rezzonico et al., 2007). Quantitative
PCR was performed with an ABI PRISM 7700 equipment using SYBR® Green PCR Master Mix
(Applied Biosystems) according to the manufacturer’s protocol. Each 25 µl PCR reaction
contained 12.5 µl of qPCR master mix, 5 µl of template from the RT-reaction, and 200 nM
forward and reverse primers. PCR cycling conditions were as follows: 2 min at 50 °C, 10 min at
95 °C and 40 cycles of 15 sec at 95 °C and 1 min at 60 °C. Measurement of SYBR green
fluorescence was performed at the end of each amplification step and continuously during the
melt-curve analysis to control for amplicon specificity. The qPCR results and melting curve were
analyzed using the SDS 2.1 Software (Applied Biosystems). Control reactions from non-reverse
transcribed samples were included to check for residual genomic DNA in the RNA samples.
Difference between cycle threshold of reverse transcribed and non reverse transcribed samples
was typically higher than 3 cycles indicating that residual DNA accounted for less than 10% in
the final result.
Efficiency of PCR amplification was measured using serial dilutions of cDNA from luciferase
mRNA or total bacterial RNA after plotting the cycle threshold values versus the log10-
transformed number of copies of luciferase mRNA or cells per reaction. Efficiency of PCR
amplification was typically in the range of 94 to 99%.
Chapter 3 - qRT-PCR
- 81 -
3.3.9 Statistical analyzes
Means of log10-transformed cell counts determined by plate counts and qRT-PCR of heat- and
rifampicin tests were compared with Student’s t-test using JMP In Software version 6.0 (SAS
Institute Inc., Cary, NC, USA) with a level of significance at 0.05.
Table 8 Primers used in qRT-PCR to quantify mRNA fragments of cysS and purB
transcripts
Gene or target (locus)
Primer sequence (5’ to 3’) Amplicon length
Transcript length (location of fragment)
Reference
cysS (BL0301) 5’-CAACCGCCGCGATCTTC-3’ 5’-CCAGCTGTGAAAGCAACGTATT-3 57 bp 1641 (1271 – 1327) bp
Rezzonico et al., 2007
purB (BL1800) 5’-CATGGGCGGCCTTGAGT-3’ 5’-TCAAGCTCACGCTCGATGAC-3’ 57 bp 1446 (254 – 310) bp
Rezzonico et al., 2007
purB (BL1800) 5’-AGCAGCGGCATATCCTTGAA-3’ 5’-TTCTGGCCAACGGCTTTG-3’ 400 bp 1446 (893 – 1292) bp This study
Luciferasea 5’-TACAACACCCCAACATCTTCGA-3’ 5’-GGAAGTTCACCGGCGTCAT-3’ 67 bp 1751 (257 – 323) bp
Johnson et al., 2005
a reference mRNA
3.4 Results
3.4.1 Morphology of B. longum NCC2705
Light microscopy of B. longum NCC2705 cells from batch cultures exhibited rods in various
shapes with the typical V- or Y-pattern. Cells appeared in single cell state and few cell clusters of
small numbers were observed (Figure 10 A). Cells originating from continuous cultures with
immobilized cells displayed similar shapes but aggregation increased with fermentation time
from small aggregates at day 4 (Figure 10 B) to macroscopic aggregates of uncountable cells with
a diameter of 0.5 – 1 mm after 9 days of fermentation and remained stable for 20 days of
continuous culture (Figure 10 C). Calibration curves for qRT-PCR and stability tests were carried
out with free cells from batch cultures.
Chapter 3 - qRT-PCR
- 82 -
Figure 10 Optical micrographs of fresh cells collected from fermented broth (1000 ×
magnification) of B. longum NCC2705 produced by batch culture (A), and continuous
immobilized cell culture at day 4 (before macroscopic aggregation) (B) and day 16 (after
macroscopic aggregation) (C).
3.4.2 Calibration curves
Calibration curves of cysS 57-, purB 57- amd purB 400-bp mRNA fragments were generated by
serially diluting cell samples with concentrations ranging from 7.0 × 103 to 4.7 × 108 CFU/ml,
corresponding to 9 to 1.7 × 105 cells per qPCR reaction volume (25 µl). The internal reference of
luciferase mRNA was used to account for the variability induced by the RNA extraction
procedure with RNA recovery rates ranging between 6 and 70 %.
CT values for the three amplicons were plotted as a function of the log10-transformed CFU/ml
determined by plate counting multiplied by the recovery rates of spiked luciferase mRNA (Figure
11). Linear amplification of target mRNAs was measured over a range between 4 and 5.5 log10
values, with detection limits of 9 and 20 cells/reaction volumes for short amplicon length of purB
and cysS, respectively, and of 40 cells/reaction volumes for purB 400-bp fragments.
Chapter 3 - qRT-PCR
- 83 -
Figure 11 Calibration curves obtained by plotting CT values for cysS 57-bp (A), purB 57-bp
(B) and purB 400-bp (C) primer sets as a function of log10 transformed cell concentrations
determined by plate counts. Data points are from three independent repetitions.
Chapter 3 - qRT-PCR
- 84 -
3.4.3 Heat test
Cells were exposed to lethal heat stress and bacterial survival assessed with plate counts was
compared with survival obtained with qRT-PCR method. Plate counts showed a reduction of
viable cells of 2.7 ± 0.2, 3.2 ± 0.1 and 3.7 ± 0.2 log10 CFU/ml after 10, 20 and 30 min exposure at
56 °C, respectively (Figure 12). Cell numbers after heat stress determined by qRT-PCR targeting
the short amplicons of cysS and purB mRNAs were significantly higher by an average factor of
18.0 ± 3.3 and 7.0 ± 2.0, respectively, compared to plate counts (p ≤ 0.05). In contrast, cell
numbers measured with qRT-PCR and purB 400-bp fragments were the same as plate counts for
all three tested heat stress periods (p > 0.05).
Figure 12 Viable cell counts of B. longum NCC2705 cells measured with qRT-PCR using 3
primer sets and plate counts after heat treatment at 56 °C for 10 (dashed), 20 (grey), 30 min
(white) compared to untreated cells (black). Standard deviations of each column are represented
by error bars. Different letters indicate significantly different values for each treatment (p ≤ 0.05).
Chapter 3 - qRT-PCR
- 85 -
3.4.4 Rifampicin test
Cells were grown in batch mode until cultures reached OD600 0.5 - 0.8 at which point rifampicin
was added to stop transcription The decay of transcripts encoding fragments of cysS and purB of
B. longum growing at 37 °C was followed at 2.5, 7.5, 14, 20, 25 and 45 min after arrest of
transcription (Figure 13).
Figure 13 Decay of mRNA fragments of cysS 57-bp (▲), purB 57-bp (■) and purB 400-bp (●)
as a function of time after addition of rifampicin at 400 µg/ml to exponentially growing cells.
Values are given as corresponding viable cell concentrations derived from calibration curves.
Standard deviations of each point are represented by error bars.
The levels of cysS and purB fragments in treated samples were translated into viable cell
concentrations using calibration curves. Amplicon levels for cysS 57- and purB 57-bp decreased
rapidly within the first 2.5 min after addition of rifampicin (log10 loss 1.19 ± 0.14, T½ 38.4 ± 4.4
sec) followed by a slower decline until 15 min (1.61 ± 0.25). In comparison, the levels of purB
400-bp amplicons decreased significantly faster (p ≤ 0.05) (log10 loss 1.47 ± 0.03; T½ 30.7 ± 0.7
Chapter 3 - qRT-PCR
- 86 -
sec) and remained at a lower level until 15 min after addition of rifampicin (log10 loss 1.86 ±
0.25). After 15 min, the levels of all three fragments increased steadily. The long fragment of
purB reached similar level as its shorter counterpart 20 min after addition of rifampicin.
3.4.5 Viable cells quantification of cell aggregate s
The qRT-PCR method targeting purB 400-bp fragments was used to quantify viable B. longum
NCC2705 cells of samples containing macroscopic cell aggregate. Formation of macroscopic cell
aggregates (0.5 – 1 mm) was observed after ca. 9 day continuous culture and remained until the
end of fermentation (Figure 10 C). Viable cell numbers obtained by measuring purB 400-bp
fragment level were compared to plate counts method from samples before and after extensive
cell aggregation. Plate counts from samples taken before aggregation (day 4) were lower by 0.5
log10 (approximately 3.2 fold) compared to cell numbers determined with qRT-PCR (Table 8).
Samples taken after formation of large aggregates at day 12, 16, and 20 showed larger differences
ranging between 2.1, 1.3 and 1.2 log10 units, respectively.
Table 9 Cell concentrations of B. longum NCC2705 during continuous immobilized cell
culture determined by plate counts and qRT-PCR (purB 400-bp). Values from day 4 are before
formation of macroscopic cell aggregates. Reported values are means of two independent
repetitions ± standard deviation.
Fermentation day
Cell count (plate count)a [CFU/ml]
Cell count (qRT-PCR)a [Cells/ml]
Ratio qRT-PCR/plate counts
4a 3.3 ± 0.3 × 108 1.1 ± 0.4 × 109 3.2 ± 1.0
12 6.6 ± 1.9 × 107 8.6 ± 4.4 × 109 124.8 ± 26.1
16 1.8 ± 0.7 × 108 4.3 ± 2.2 × 109 22.5 ± 5.7
20 5.0 ± 2.3 × 108 7.4 ± 3.3 × 109 14.7± 0.2 a before formation of macroscopic aggregates
Chapter 3 - qRT-PCR
- 87 -
3.5 Discussion
Levels of mRNA fragments of housekeeping genes cysS and purB were used for quantification of
viable cells of B. longum NCC2705 in preparations displaying heterogeneous morphology. These
two genes are constitutively expressed in various conditions and used to normalize gene
expression analyzes (Rezzonico et al., 2007). The use of unstable mRNA as viability marker is
difficult and strongly dependent on the efficiency of RNA extraction because loss during sample
preparation significantly biases the output of the experiment (Bustin, 2000). Addition of an
exogenous reference mRNA circumvents this bias by monitoring the losses during sample
processing. The importance of normalizing mRNA levels with the recovery of exogenous
references is highlighted by the large variation in extraction yields ranging from 6 – 70 %. High
regression coefficients were obtained for all three mRNA fragments using normalized data with
recovery rates. A mean detection limit of 2.7 ± 1.7 × 104 cells per ml of culture was achieved for
all three mRNA used. This low level is comparable to that reported for bifidobacteria quantified
in fecal samples using real-time PCR targeting DNA (Gueimonde et al., 2004). A detection limit
of 5 × 104 cells of Salmonella per gram of seeded soil or chicken manure can be achieved using
real-time PCR method targeting mRNA extracted with magnetic capture hybridization (Jacobsen
and Holben, 2007).
The short mRNA fragments of both housekeeping genes were too stable to be used as accurate
viability markers and yielded an overestimation of viable cell counts after the three heat
treatments when compared to standard plate counts. In contrast, results obtained with long
fragments of purB coincided well with plate counts irrespective the duration of the stress
treatment. Differential stability of specific mRNA fragments was confirmed by treating B.
longum NCC2705 with rifampicin, an inhibitor of RNA-transcription (Campbell et al., 2001).
Chapter 3 - qRT-PCR
- 88 -
The 400-bp fragments of purB were degraded faster than short fragments of both genes. Similar
inverse relation between length of RNA or DNA molecules and their stability was recently
reported in other bacterial species (Aellen et al., 2006, Soejima et al., 2008). Quantification of
rRNA levels after antibiotic treatment in Streptoccus gordonii showed better correlation between
qRT-PCR values and plate counts when targeting a 427-bp compared to a 119-bp fragment which
remained stable after drug killing (Aellen et al., 2006). DNA was used as a viability marker of
Listeria monocytogenes in combination with a DNase-assay and showed that long DNA
fragments (894-bp) of heat killed cells were not detectable after treatment, whereas short
fragments (113-bp) were still evident causing false positive results (Soejima et al., 2008).
The rapid decrease of levels of the three fragments in the first minutes after addition of rifampicin
yield very small half-life times (33.6 ± 3.4 sec) compared to that reported for Escherichia coli,
Bacilllus subtilis and Lactococcus lactis of 3-8 min (Bernstein et al., 2002; Hambraeus et al.,
2003; Redon et al., 2005). However, mRNA half-lives in these studies were calculated from the
degradation rate constant obtained from microarray analyzes. The use of sensitive qRT-PCR
technique can affect the determination of RNA levels compared to microarray results. Decay
rates of transcripts determined with microarray technology were shown to be slower than values
obtained by Northern blots or real time PCR (Hambraeus et al., 2003; Hundt et al., 2007).
The increase of transcript levels shortly after the start of the treatment can be attributed to rapid
bacterial adaptation to rifampicin or selection for resistant subcultures within the same culture as
reported for bifidobacteria from human samples treated with this antibiotic (Mangin et al., 1994).
Proliferating cells regenerate new intact mRNA transcripts resulting in equal levels of de novo
synthesized transcripts of short and long purB fragments as observed 20 min after beginning of
treatment.
Chapter 3 - qRT-PCR
- 89 -
Although auto-aggregation might improve adhesion of probiotic bacteria to epithelial cells (Del
Re et al., 2000), application of aggregated cells in functional food products suffers from
underestimation of viable cell counts obtained with standard plate count since the method cannot
detect individual cells in aggregates. The use of cell aggregates is thus still unfavorable in
product applications with plate counts being the standard method to verify product quality. In
fact, the use of the qRT-PCR method with aggregated cells showed that plate counts largely
underestimated viable cell numbers with factors ranging from 15 to 125.
In conclusion, we developed an accurate real time PCR method determining viable cell counts of
B. longum NCC2705 with different morphologies. The careful selection of the target gene and
fragments length as viability markers is essential to avoid biases in viable cell counting.
Comparison of primer sequences of purB 400-bp with sequenced genomes showed direct hits
with genomes of B. longum DJO10A and B. longum subsp. infantis ATCC 55813, CCUG 52486
and ATCC 15697, an auto-aggregating strain (Rahman et al., 2008) and could therefore be used
as viability marker in these strains. qRT-PCR may be used for rapid quantification of viable cells
in mixed cultures provided that primer specificity and sample processing is not affected by mixed
culture conditions. Moreover, the method may be developed in combination with DNA
quantification to evaluate the proportions of dead cells within cell aggregates which may provide
a protection layer for viable cells inside the core of aggregates during transit through the GIT.
Growing interest in probiotic characteristics of aggregating bacterial cells make this technique a
valuable tool to accurately quantify viable probiotic bacteria exhibiting heterogeneous
morphology.
Chapter 4 - Spray drying
- 90 -
4 Development of a rapid screening protocol for
selection of Bifidobacterium longum strains resistant
to spray drying and powder storage
Data presented in this chapter were submitted and accepted with minor revision in 2009 in
Beneficial Microbes with the following authors: Reimann S., Grattepanche F., Baggenstos C.,
Rezzonico E., Berger B., and Lacroix C.
Chapter 4 - Spray drying
- 91 -
4.1 Abstract
An efficient screening method for selection of Bifidobacterium longum strains resistant to spray
drying and storage was developed based on Randomly Amplified Polymorphic DNA (RAPD) for
identification of the best survivors in mixed strains bacterial preparations. Three different primers
were used to generate RAPD profiles of 22 B. longum strains. All strains were distinguished
according to their RAPD profiles except for the strain NCC2705 and its H2O2 resistant derivative
variant. The 22 strains were grouped in 3 batches of 7, 7 and 8 strains and subjected to spray
drying and storage at 30 and 37 °C under anaerobic conditions. Batch survival rates after spray
drying reached 17.1 ± 4.4 %. Strains showing the highest prevalence and/or resistance to storage
at 37 °C were selected from individual batches for subsequent spray drying and storage testing.
After 67 days of storage, NCC572 was identified as dominant strain in powder. The stability of
strain NCC572 was confirmed by performing single spray drying and storage tests. Out of 22 B.
longum strains, a robust strain was identified by combining RAPD with a simultaneous screening
test for survival to spray drying and storage. The method allowed a fast screening of B. longum
strains in mixture for resistance to spray drying and storage compared to traditional screening
procedures carried out with individual strains, in the same conditions. This approach could be
applied to other stress conditions.
Chapter 4 - Spray drying
- 92 -
4.2 Introduction
Bifidobacteria are widely used as probiotics in functional food products. Industrial standards
currently used for products containing viable probiotic strains require presence of minimal
bacterial viable counts in order to confer health benefits to consumers. Probiotic bacteria in
functional food products are usually incorporated as frozen concentrates or freeze dried powders
(Meng et al., 2008; Saarela et al., 2006). Freezing and freeze drying are however expensive
stabilisation technologies due to the infrastructure and energy consumption during production,
transport and storage (Knorr, 1998). Spray drying has gained popularity over the last decade for
large scale production of starter cultures and probiotics (Corcoran et al., 2004; Desmond et al.,
2002; Gardiner et al., 2000; Lian et al., 2002; Meng et al., 2008; O’Riordan et al., 2001; Simpson
et al., 2005). Although spray drying is an economical process, between 5 – 10 times cheaper than
freeze drying (Santivarangkna et al., 2007), bacterial cells suffer lethal damages attributed to the
effect of exposure to heat and dehydration (Ross et al., 2005; Teixeira et al., 1995a). These have
deleterious effects on survival of sensitive bacteria such as probiotics. Survival rates of probiotic
bacteria during spray drying is strain-dependent and influenced by several factors, the most
important being the drying conditions applied, in particular outlet temperature (Gardiner et al.,
2000; Lian et al., 2002), the carrier media used (Corcoran et al., 2004; Johnson and Etzel, 1993;
Lian et al., 2002;), presence of protective compounds and the growth phase of cells (Meng et al.,
2008).
Loss of bacterial viability during storage of dry cultures is also a major limiting factor for many
cultures (Ross et al., 2005; Texeira et al., 1995). Parameters such as temperature, moisture
content of spray dried powders, water activity, carrier media, exposure to light and oxygen have
been shown to impact bacterial survival during storage in powder form (Ananta et al., 2005;
Chapter 4 - Spray drying
- 93 -
Chàvez and Ledeboer, 2007; Gardiner et al., 2000; Meng et al., 2008; Simpson et al., 2005;
Teixeira et al., 1996). Spray drying and storage survival rates do not concur and vary from strain
to strain (Corcoran et al., 2004; Gardiner et al., 2000; Simpson et al., 2005) and therefore careful
strain selection is important to ensure high survival during spray drying and subsequent storage
of dried cultures. The underlying molecular mechanisms responsible for viable cell losses during
drying and storage remain unclear and thus empirical approaches for efficient strain screening are
of utmost value to develop probiotic strains with good technological and functional properties
and to avoid labour-intensive, lengthy and costly strain-by-strain screening procedures.
Molecular typing techniques such as restriction fragment length polymorphism, pulse field gel
electrophoresis, repetitive element sequence-based PCR fingerprinting and randomly amplified
polymorphic DNA (RAPD) allow the differentiation of bacteria at strain level without prior
knowledge of genome sequence (Ben Amor et al., 2007; Collado et al., 2006; Gevers et al., 2001;
Masco et al., 2003; Rossetti and Giraffa, 2005; Vincent et al., 1998).
In this study, we applied an efficient method using RAPD-based strain identification to identify
the best survivor strains after spray drying and storage of mixed bacterial preparations during two
selection rounds followed by a final confirmation step (Figure 14). Twenty two B. longum were
divided into three batches of 7 or 8 strains and exposed to a first selection consisting of a spray
drying and storage experiment. Best survivors of each batch preparation were identified by
RAPD after 4 log10 viability loss during storage and applied to a second selection round to
identify the most resistant B. longum strain. The survival of the selected strain was tested
separately and compared to a low resistant strain for validation.
Chapter 4 - Spray drying
- 94 -
Figure 14 Flow chart of experimental procedure. The 22 B. longum strains were divided into
3 batches for spray drying and storage. After the first selection round, 2 - 3 best survivor strains
were selected for a subsequent spray drying and storage batch. After the second selection round,
the best survivor strain and a control strain were selected for a single strain spray drying and
storage confirmation batch.
4.3 Material and Methods
4.3.1 Bacterial strains and culture conditions
The 22 B. longum strains used in this study were provided by Nestlé Culture Collection (NRC,
Lausanne, Switzerland) (Table 10). Strains were routinely cultured in 20 ml-McCartney bottles
containing de Man, Rogosa, Sharpe (MRS) broth (BioLife, Milan, Italy) supplemented with 0.05
% (w/v) L-cysteine (Sigma-Aldrich, Buchs, Switzerland) (MRSC), and incubated for 16 h at 37
22 B. longum strains
First selection round
Second selection round
Single strainconfirmation batches
8 strains
1 strain 1 strain
22 B. longum strains
First selection round
Second selection round
Single strainconfirmation batches
8 strains
1 strain 1 strain
Chapter 4 - Spray drying
- 95 -
°C in anaerobic conditions using atmosphere generation system packs (AnaeroGen, Oxoid,
Pratteln, Switzerland). Cultures from frozen stock were sub-cultured twice before use as
inoculum at 1% (v/v). Cell counts of pure liquid cultures were determined by serial dilutions of
samples in 12 mM phosphate buffer saline with 0.05 % (w/v) L-cysteine (PBSC) followed by
plating in duplicate on MRSC agar 1.5% (w/v) (Becton Dickinson, Basel, Switzerland). Plates
were incubated anaerobically at 37 °C for 72 hours.
4.3.2 Preparation of bacterial suspensions used for survival tests
Individual strains were grown separately at 37 °C in rubber-lid flasks containing 250 ml MRSC
inoculated at 2 % (v/v). The headspace of flasks was flushed with CO2 to ensure anaerobic
conditions. Growth was monitored by measuring OD600 and cells were harvested 3 h after
reaching stationary growth phase. Volumes of 40 ml of cell broth were centrifuged (15,000 × g,
15 min at 4 °C), and cell pellets washed once with PBS-C and resuspended in 100 ml ice-cold 20
% (w/v) reconstituted skim milk (RSM) (Oxoid Ltd, Hampshire, United Kingdom). A sample of
1 ml was used for plate counts.
4.3.3 Preparation of strain mix
For the first and second selection batches, 100 ml of each bacterial suspension (7 or 8 strains per
batch) in RSM were mixed together in a 1 l-bottle containing a magnetic stirrer (Table 10). The
solution was kept at 4 °C in an ice bath and stirred for at least one hour before spray drying. For
single strain confirmation tests, a similar procedure was used with pure cultures to produce 400
ml ice-cold RSM cell suspension used for spray drying.
Chapter 4 - Spray drying
- 96 -
4.3.4 Spray drying
Spray drying was conducted using a laboratory scale spray dryer (Büchi model B-290, Büchi,
Flawil, Switzerland). The feed suspension was pneumatically atomized into a vertical, co-current
drying chamber using a two-fluid nozzle. The inlet air temperature was maintained at 170 °C and
the outlet air temperature was adjusted to 78.2 ± 1.9 °C by levelling the feed pump rate. A
maximum volume of 300 ml of bacterial RSM mix was used per batch corresponding to a
maximum of 3 h operation time. The spray dried powder was collected in a single cyclone
separator.
Spray drying was performed once for the first selection batches whereas second selection batch
and confirmation experiments using single strain preparations were carried out in duplicate.
4.3.5 Storage test
Spray dried powder was aliquoted in 2-ml polypropylene tubes (Eppendorf, Hamburg, Germany)
sealed with sterile cotton wool. Tubes were then placed in 2.7 l desiccators (DURAN Group
GmbH, Mainz, Germany) containing 150 ml of 5 M saturated potassium acetate as desiccant and
kept in the dark. Storage temperatures of 30 and 37 °C were tested for first selection batches
whereas powders from second selection and confirmation batches were stored at 37 °C only.
Desiccators were evacuated and then flushed with nitrogen to ensure anaerobic conditions.
Powders from selection batches were sampled 2-3 times per week or once a week towards the
end of storage or in confirmation batches for cell counts and water activity measurement. Storage
was stopped when a threshold loss in cell counts of ca. 4 log10 of number of colony forming units
(CFU) per gram of powder was reached. Storage experiments were performed once per spray
Chapter 4 - Spray drying
- 97 -
drying batch for the first selection batches and in duplicate per spray drying experiment for the
second selection batches.
4.3.6 Determination of water activity and moisture content
Water activity of powder was measured using a hygrometer (AquaLab, Gerber Instruments SA,
Effretikon, Switzerland) according to the manufacturer’s recommendations. Values are means of
two determinations per sample taken at a reference temperature of 25 °C.
Moisture content expressed as percentage of spray dried powder weight was determined by
measuring the difference of powder weight before and after heating 10 g of powder overnight at
102 °C. Dry matter of bacterial suspension in RSM was determined by weighing 10 ml of
suspension oven-dried for 36 hours at 80 °C. Determination of moisture content and dry matter
was performed in duplicate.
4.3.7 Viable cell counts in spray dried powders
Survival of B. longum strains to spray drying and storage was performed according to Gardiner et
al. (2000). Briefly, 0.1 g powder was resuspended in 9.9 ml PBS-C, homogenized using a
stomacher 80 (Müller and Krempel AG, Bülach, Switzerland) for 30s and stored at room
temperature for 1 hour to allow rehydration. Cell suspensions were serially diluted in PBSC and
appropriate dilutions were plated in duplicate on MRS-C agar. Percentage of survival after spray
drying was calculated as the CFU per gram powder after spray drying divided by the initial
number of CFU per gram dry matter in RSM bacterial suspensions before spray drying.
Chapter 4 - Spray drying
- 98 -
4.3.8 Identification of best surviving strains usin g RAPD
Genomic DNA was extracted from pure cultures grown in liquid media according to
Goldenberger et al. (1997). Briefly, 0.5 ml of culture were centrifuged and resuspended in 0.2 ml
digestion buffer containing 50 mM Tris-base (Sigma-Aldrich), 1 mM EDTA (pH 8.5) (Sigma-
Aldrich), 0.5 % SDS (Sigma-Aldrich), 200 µg/ml proteinase K (AppliChem GmbH, Darmstadt,
Germany). After incubation at 55 °C for 3 h under agitation (Thermomixer, Eppendorf,
Hamburg, Germany), the mixture was held at 95 °C for 10 minutes to inactivate proteinase K,
cooled to 4°C and centrifuged (12,000 × g, 10 min, at 4 °C). The supernatant was collected and
nucleic acid concentration was determined using Nanodrop ND-1000 (Peqlab, Erlangen,
Germany). Samples were diluted in distilled water to reach a final nucleic acid concentration of
25 ng/µl and stored at – 20 °C until PCR amplification.
RAPD analysis was carried out according to Vincent et al. (1998), i.e., one µl of DNA sample
(25 ng/µl) was mixed with 24 µl mastermix containing 0.2 mM dNTP (GE Healthcare Europe,
Glatttbrugg, Switzerland), 0.2 pmol of each primer, 10 x Taq buffer (Euroclone, Siziano, Italy),
1.5 mM MgCl2 (Sigma-Aldrich), 2 % (w/v) Tween 20 (Sigma-Aldrich) and 2.5 U Taq
Polymerase (Sysmex Digitana, Horgen, Switzerland). Three random single primers were used:
OPA-02 (5’ TGCCGAGCTG 3’), OPL-07 (5’ AGGCGGGAAC 3’) and OPL-16 (5’
GGGAACGTGT 3’) (Microsynth, Balgach, Switzerland). These three primers were selected
because they permit differentiation of all 22 strains except for NCC2705 and its H2O2-resistant
variant NCC2916 (data not shown) which could be differentiated according to physiological
H2O2 resistance tests (Figure S1). These two strains were distributed in two different batches
during the first selection batch experiments to avoid identification problems (Table 1). PCR
amplification was run in a Biometra®TGradient thermocycler (BioLabo, Châtel-St. Denis,
Chapter 4 - Spray drying
- 99 -
Switzerland) under the following cycling conditions: 94 °C for 3 min, 45 cycles of 94 °C for 1
min, 30 °C for 1 min and 72 °C for 2 min. PCR products were run on 2 % (w/v) agarose gel
stained with ethidium bromide. The RAPD profiles were recorded using AlphaImager™ system
(Alpha Innotech Corporation, California, USA).
Survival of individual strains after spray drying and storage of dry powder containing mixed
strains was measured after reaching 4 log10 loss in total bifidobacterial counts. A total of at least
hundred colonies from 2 - 4 plates harboring between 5 and 100 colonies were then randomly
picked and separately cultivated in MRSC broth at 37 °C for 24 h under anaerobic conditions and
stored at -20 °C until DNA extraction.
4.4 Results
4.4.1 Strain survival and selection during the firs t selection batches
22 B. longum strains were divided in three batches each consisting of 7 - 8 strains (Figure 14) for
spray drying and storage experiments (Table 10). Mean initial cell concentration of cell
suspensions in RSM for the three batches before spray drying was 1.5 ± 0.4 × 109 CFU/g dry
matter. Viable total cell counts of spray dried powders measured immediately after spray drying
were 2.2, 3.6 and 1.8 × 108 CFU per gram of powder, corresponding to survival rates of 12.6,
21.4 and 17.4 % for Batch 1, 2 and 3, respectively. Average moisture content of powders was 5.4
± 0.7 %. Aliquots of dried powder were then stored in two desiccators at 30 and 37 °C with a
water activity of 0.23 ± 0.01.
Chapter 4 - Spray drying
- 100 -
Table 10 Distribution of the 22 B. longum strains in the three batches of the first selection
round and origin of the strains.
Straina Batch number Source NCC284 1 Infant feces NCC293 1 Infant feces
NCC 305 1 Infant feces
NCC 324 1 Infant feces
NCC 344 1 Infant feces
NCC 2705 1 Infant feces
NCC 3001 1 Infant feces
NCC 417 (ATCC 15707) 2 Fermented milk (yogurt) NCC 435 (ATCC 15708) 2 Adult feces
NCC 444 (NCIMB 8809) 2 Infant feces
NCC 450 (DSM 20097) 2 Infant feces
NCC 461 2 Calf feces
NCC 469 2 Infant feces
NCC 4012 2 Commercial probiotic product
NCC 490 3 Infant feces NCC 510 3 Infant feces
NCC 521 3 Adult feces
NCC 552 3 Adult feces
NCC 572 3 Infant feces
NCC 585 3 Adult feces
NCC224 3 Infant feces
NCC 2916 3 H2O2 resistant mutant of NCC2705 a Alternative strain denotation indicated in brackets. ATCC: American Type Culture Collection; DSM: Deutsche Sammlung von Mikroorganismen; NCIMB: National Collection of Industrial, Marine and Food Bacteria; NCC: Nestlé culture collection A non-linear decrease of cell counts (log10) was observed in spray dried powders during storage
at both temperatures (Figure 2). Powder storage at 37 °C resulted in faster viability loss with a 4
log10 loss threshold in cell counts reached after 46, 40 and 49 days at 37 °C (Figure 2 B)
compared to 64, 65 and 53 days at 30 °C (Figure 2 A) for Batch 1, 2 and 3, respectively. After
this storage period, at least 100 individual colonies isolated from different plates of the highest
plating dilutions were collected and identified using RAPD.
Chapter 4 - Spray drying
- 101 -
Figure 15 Cell counts in spray dried powders of batches from first selection round during
storage at 30 °C (A) and 37 °C (B) for Batch 1 (full line, ●), 2 (semi-dashed line, ▲) and 3
(dashed line, ■).
Chapter 4 - Spray drying
- 102 -
Proportion of individual strains found within the three batches before and after spray drying and
storage at 37 °C are shown in Figure 16. Different fractions were already measured for individual
strains before spray drying as a result of different growth of cells in the culture used for
preparation of the bacterial strain mixtures in RSM. After spray drying and storage of Batch 1 at
37 °C, all seven strains present in the initial cell suspensions were detected at the end of the
storage period. NCC3001 and NCC293, which were predominant before spray drying showed
only a small decrease of their relative proportion after storage, from 32 and 45 % to 30 and 40 %,
respectively (Figure 16 A). NCC284 was the only strain showing increased relative proportion
after storage (from 4 to 10 %). Only three of the seven strains present in the initial cell
suspension were detected after storage at 30 °C. The relative percentage of strain NCC3001
increased from 32 to 56 % whereas NCC293 remained stable at about 44 %. NCC284 was not
detected at the end of storage at 30 °C (Figure 16 A).
Similarly to Batch 1, strain diversity in Batch 2 and 3 was also better conserved after storage at
37 °C compared to 30 °C. In Batch 2, strains NCC417 and NCC435 were already over-
represented before spray drying and displayed the highest stability since their relative proportion
increased after storage at both 30 and 37 °C, from 29 to 36 and 50 % for NCC417 and from 23 to
30 and 43 % for NCC435, respectively (Figure 16 B). Surprisingly, NCC461 doubled its relative
presence from 13 to 29 % in the powder stored at 30 °C, but its fraction decreased to only 1 % at
37 °C. In Batch 3, strain NCC572 initially accounting for 22 % of total bacterial counts, largely
dominated after storage, reaching 95 and 98 % at 30 and 37 °C, respectively (Figure 16 C).
Chapter 4 - Spray drying
- 103 -
Figure 16 Relative presence (%) and corresponding cell counts (CFU per gram) of B.
longum strains in RSM before spray drying and in spray dried powders after end of storage at 30
and 37 °C for the first selection Batches 1 (A), 2 (B) and 3 (C). Storage was stopped after 4 log10
loss of CFU counts in powder (in brackets: storage time). For RSM, distribution-profiles were
calculated using cell concentration of each strain in pure culture before mixing. Individual
populations in spray dried powders were estimated using their relative abundance (%) determined
by RAPD and total cell counts in powder.
Chapter 4 - Spray drying
- 104 -
In order to reduce the number of strains for the second selection round, only 2 or 3 strains from
each batch were selected based on their high survival in powders stored at 37 °C. Strains with
highest percentage (NCC293) or increased relative percentage (NCC 284) were selected among
the 7 strains detected in the powder for Batch 1 (Figure 16 A). For Batch 2, among the 5 strains
detected in the powder, strains NCC417 and NCC435 were selected due to their high fraction in
the RSM suspension (indicating good growth characteristics in MRSC) and their increased
relative proportion after storage. NCC450 was also selected because its fraction in the RSM cell
suspension and in the powder after storage remained stable (Figure 16 B). For Batch 3, among 3
strains detected in the powder after storage at 37 °C strains NCC572 and NCC585 were chosen
due to their dominant appearance (NCC572) and stable percentage (NCC585) (Figure 16 C).
Strain NCC2916 which showed low survival in Batch 3, was chosen as a “negative control” for
inclusion in the second selection round.
4.4.2 Strain survival and selection during the seco nd selection round
Selected strains were spray dried and stored at 37 °C. The moisture content and water activity in
spray dried powders from the second selection round with 8 strains were 5.4 ± 0.2 % and 0.21 ±
0.01, respectively. Mean initial cell concentration before spray drying in RSM was 1.2 ± 0.2 ×
109 CFU/g dry matters. Survival during spray drying was 16.2 ± 9.1 % and cell counts in the
powder before storage were 2.0 ± 1.3 × 108 CFU/g. The log10 transformed cell counts in spray
dried powder decreased non-linearly with storage time (Figure 17), but at a slower rate compared
to the first selection batches at 37 °C (Figure 2 B). Therefore, cell counts did not reach the 4 log10
loss threshold per gram of powder after 67 days of storage (Figure 17).
Chapter 4 - Spray drying
- 105 -
Figure 17 Cell counts in spray dried powders of second selection batch (8 strain mix) (full
line, ○) and confirmation batches NCC572 (semi-dashed line, ■) and NCC2916 (dashed line, ▲)
during storage at 37 °C. Data are means of two independent experiments. Standard deviations of
each point are represented by error bars.
Individual strains accounted for 6 to 20 % of total cell counts in the RSM mix (Figure 18 A).
RAPD identification of 239 colonies picked after 67 days storage at 37 °C indicated strain
NCC572 as best survivor after spray drying and storage, with a large proportional increase in
powder from 12 % in RSM before spray drying to 88 % in the powder (Figure 18 B). NCC585
also exhibited good survival in powder with constant proportion of total cells of about 9 ± 2 % in
the RSM mix and powder. All the other strains decreased in percentage or were not detected in
the powder (Figure 18 B).
Chapter 4 - Spray drying
- 106 -
Figure 18 Relative presence (%) and corresponding cell counts (CFU per gram) of the B.
longum strains of the second selection batch in RSM before spray drying (A) and in spray dried
powders after storage for 67 days at 37 °C (B). For RSM, distribution-profiles were calculated
using cell concentration of each strain in pure culture before mixing. After storage, individual
populations in spray dried powders were estimated using their relative abundance (%) determined
by RAPD and total cell counts. Data are means of two trials ± standard deviation. a No standard deviation since only one colony was picked.
4.4.3 Validation of selection procedures using sing le strain
preparations
To confirm NCC572 as good spray drying and storage survivor, NCC572 and NCC2916 (low
survivor control) were tested in pure culture spray drying and storage experiments. Average
moisture content and water activity of powders was 5.5 ± 0.6 % and 0.17 ± 0.01, respectively.
Survival rate of NCC572 and NCC2916 during spray drying was 18.8 ± 9.6 and 4.8 ± 3.9 % with
an initial viable cell count of 2.3 ± 2.7 × 109 and 9.7 ± 2.0 × 108 CFU/g dry matter, respectively.
Strain NCC2916 showed a 4 log10 loss threshold in cell counts after 32 days of storage whereas
Chapter 4 - Spray drying
- 107 -
strain NCC572 exhibited only 2 log10 viability loss after 34 days and 3.4 log10 after 69 days
(Figure 17). Viable cell counts in powders of NCC572 after 62 days of storage were 1.5 ± 0.5 ×
105 CFU/g powder and exceeded counts of NCC2916 after 60 days by a factor of 269.
4.5 Discussion
Spray drying and storage stability is an important bacterial characteristic for the use in food
application at industrial scale because spray drying is a low cost alternative to freeze drying.
Survival to drying and storage is strain dependent and selection of robust strains with good
functional properties is required for successful application. Screening of optimum process
conditions and strains with good resistance is laborious, expensive and limited by variability
between experiments. Simultaneous screening of multiple strains can be used to increase
screening throughput. RAPD is a fast, practical, easy to perform and inexpensive technique to
distinguish bacterial strains. Furthermore, its application does not require a high level of technical
skills (Rossetti and Giraffa, 2005). RAPD analysis was successfully used in this study to
discriminate 22 B. longum strains, except strain NCC2705 and its H2O2 resistant derivative
NCC2916, in mixtures of up to 8 strains. It is theoretically possible to apply all 22 strains in one
single spray drying and storage batch provided that uniform treatment for each strain is
guaranteed.
Spray drying parameters were chosen from previous studies with bifidobacteria and lactic acid
bacteria and kept constant for all batches. RSM as a carrier medium is commonly used for spray
drying of lactic acid bacteria due to its protective effect against heat (Ananta et al., 2005;
Desmond et al., 2002). The inlet and outlet temperatures were set at 170 °C and ca. 80 °C,
respectively, according to Gardiner et al., (2000). Survival rate of B. longum cells determined
Chapter 4 - Spray drying
- 108 -
after spray drying in mix or pure cultures ranged from 5 to 21 % under these conditions and
concurred well with survival rate of 20 ± 1 % reported for B. longum biotype longum NCIMB
8809 in 20 % (w/v) RSM solution (Simpson et al., 2005), a strain also used in this study
(NCC444). The different survival rates between batches are most likely caused by different strain
compositions. Strain dependent survival to spray drying was previously reported for
bifidobacteria ranging between 12 – 102 % with an outlet temperature of 85 – 90 °C (Simpson et
al., 2005). Conditions were set to achieve a moisture content of the powder close to the optimal
value of 4 % for stability during storage (Gardiner et al., 2000; Simpson et al., 2005). However, a
slightly higher moisture value of 5.4 ± 0.7 % of powders was obtained in our study, partly
explained by a lower outlet temperature of 78 °C. Increasing the outlet temperature to produce a
cell powder with 4 % moisture content would likely decrease survival rate after spray drying as
reported by Gardiner et al. (2000) and Lian et al. (2002) for spray drying of Lactobacillus
paracasei, Lactobacillus salivarius and bifidobacteria. Water activity of 0.23 ± 0.01 during
storage for all batches containing strain-mixes lies within the recommended range for stability of
dried lactic acid bacteria (Champagne et al., 1996; Teixeira et al., 1995b).
The log10 transformed cell counts in powder decreased non-linearly as shown for lactobacilli
stored in RSM powder at 30 °C (Gardiner et al., 2000). Storage temperatures of 30 and 37 °C
selected to accelerate the loss in cell viability caused a faster decrease in cell viability at 37 than
at 30 °C and reached a 4 log10 units loss within 5 - 10 weeks. This effect of temperature is in
agreement with other studies on survival of spray dried probiotic bacteria during storage
(Corcoran et al., 2004; Desmond et al., 2002; Gardiner et al., 2000; Simpson et al., 2005;
Teixeira et al., 1995b).
Survival of the single strains varied among powders stored at 30 and 37 °C and resulted in
differences of strain distribution in the two powders. Remarkably, a higher number of different
Chapter 4 - Spray drying
- 109 -
strains was detected after storage at 37 compared to 30 °C for Batch 1 and 2, and to lesser extent
for Batch 3. The time period of storage may have a noticeable impact on survival of each strain in
dried powders because time duration between collection of samples at 30 and 37 °C differed
between 18 and 25 days for Batch 1 and 2, respectively compared to Batch 3 where samples from
storage at 30 °C where taken after only 53 days of storage, 4 days after the sampling from storage
at 37 °C. These findings suggest that longer storage periods at lower temperatures may raise
selection pressure to strains in powder state leading to a lower variety of strains detectable at the
end of storage at 30 °C.
The second selection round aimed at identifying the best survivors among strains selected from
the first round. The assorted strains in the second selection batch exhibited as expected prolonged
survival in powder and did not reach the 4 log10 loss threshold after 67 days of storage compared
to first selection batches where thresholds were reached within less than 50 days. Survival to
spray drying and storage for 67 days at 37 °C varied largely among single strains and led to
identification of strain NCC572 as best survivor strain among 22 tested B. longum strains. Its
high survival was confirmed by single strain drying and storage tests.
Successful application of spray drying of probiotic strains as an economic alternative to freezing
or freeze drying depends on the careful selection of appropriate strains harbouring intrinsic
resistance to drying and powder storage in addition to aspects of safety and functionality. Our
study showed the potential of RAPD technology to screen strains for survival to drying and
storage in combination with a time-effective mixed strain screening approach. A similar method
combining strain mixture testing and RAPD detection could be used to screen probiotic strains
simultaneously for resistance to different lethal stresses related to technological and functional
properties such as gastric, bile, oxygen, low pH, heat, and osmotic stress. Moreover, this
approach allows direct comparison of stress response of single strains and avoids variation
Chapter 4 - Spray drying
- 110 -
between experiments and time consuming repetitions. Further analyses of the strains exhibiting
intrinsic tolerance to drying and storage using comparative genomics and transcriptomics tools
could be followed in order to try to elucidate mechanisms conferring microbial stability during
drying and storage.
Chapter 5 - Conclusion and Outlook
- 111 -
5 General conclusions and outlook
5.1 General conclusions
The PhD thesis “Novel technologies for detection, production and screening of stress tolerant
bifidobacteria” consisted of three hypotheses. The first hypothesis claimed that B. longum
NCC2705 produced with immobilized cell technology and continuous culture had an improved
stress resistance progressing with culture age. In previous studies, cell immobilization in
combination with continuous probiotic culture showed positive effects on environmental stress
tolerance improving with culture time. Our studies confirmed the hypothesis partly, since we
observed better tolerance of immobilized cells (IC) of B. longum NCC2705 to porcine bile salts
and some antibiotics compared to free cells from batch culture independent of the culture age,
whereas tolerance to heat stress did not change, neither with culture age nor compared to batch
control cells. However, a time dependent change of cell morphology was detected and resulted in
formation of macroscopic cell aggregates of B. longum NCC2705. This extensive morphological
changes lead to great difficulties in accurately determining viable cell numbers which are
prerequisite for testing cell stress tolerance. Continuous immobilized cell fermentation in a two-
stage process lead to high cell production, similar to bacterial batch cultures and increased cell
volumetric productivity by approximately ten fold.
The immobilization of B. longum NCC2705 did not concur with the observations of Doleyres et
al. (2004a) where immobilized cells of B. longum ATCC 15707 produced continuously in co-
culture with Lc. lactis showed increased tolerance to various environmental stresses and no
formation of cell aggregates. The differences in stress resistance as well as the induction of auto-
aggregation may be attributed to the strain specificity in addition to difficult viable cell
determination. Moreover, the enhanced robustness of B. longum ATCC 15707 may be a result of
Chapter 5 - Conclusion and Outlook
- 112 -
growth in the immobilized matrix in combination with an adaptation process to co-culture
conditions rather than only from cell immobilization. In fact, a recent study showed that only the
supernatant of co-culture in a compartmentalized fermentation with a B. longum and B. breve
strain affected bacterial physiology (Ruiz et al., 2009). Improved tolerance to porcine bile salts
and auto-aggregation of B. longum NCC2705 produced in our study could confer advantages in
the probiotic approach. Large cell aggregates might provide physical protection from adverse
environmental parameters during transit through GIT and could improve adhesion abilities to
mucosal epithelium from the host. Although the acquired characteristics could be partly
explained by the observed changes in the fatty acid composition, other factors may also
contribute to altered cell physiology. Preliminary results from microarray analyzes showing
transcriptional regulation of membrane associated transporter proteins as well as ribosomal
proteins may reflect a deep impact of the culture conditions on transport mechanisms and
translation machinery (see Appendix 6.2).
The acquired physiological and morphological characteristics of cells from ICC should be tested
for reversibility by repeated batch cultures and the predicted advantage in adhesion capacity
could be testing using in vitro assays with intestinal cells such as Caco-2 by testing large cell
aggregates versus batch control cells. However, physiological analyzes of bacteria in large
aggregates are impossible to perform with traditional plate count method.
In the second part of the thesis, we developed thus a new protocol to measure viable B. longum
NCC2705 cells using real time PCR technology. The method was based on measurement of
transcript levels of a housekeeping gene using quantitative real time PCR (qRT-PCR). The
development of the protocol is based on the detection of mRNA of a housekeeping gene whose
constitutive expression was confirmed with microarray analyzes (data from Nestlé Research
Center Lausanne, Switzerland). The hypothesis was verified in that an adequate viability marker
Chapter 5 - Conclusion and Outlook
- 113 -
was identified as an amplicon of 400 bp length of purB gene of B. longum NCC2705. Batch free
cells quantified with qRT-PCR showed good correlation with plate counts before and after a
lethal heat stress treatment. The quantification of cells in the aggregated state with qRT-PCR
confirmed that the traditional plate count method resulted in major underestimation of viable cells
counts.
Quantification of viable cells in macroscopic aggregates revealed the limitation of the traditional
plate count method and emphasized the need of new technologies to quantify viable bacteria. The
qRT-PCR method developed in this method is not limited to quantify viable B. longum NCC2705
cells in pure cultures only. The method could be applied to other bifidobacterial strains or used to
quantify Bifidobacterium subsp. in mixed cultures with bacteria belonging to different genera,
provided that enzyme activity and primer specificity remains unaffected. Drawbacks of the
method is the requirement of well-trained labor as well as the high cost of enzymes, fluorescent
stains and RNA-extraction which limit the extensive application of the method as a standard
protocol. Progressive use of such consumables for gene expression analyzes such as RT-PCR or
microarray analyzes might lower the cost in the near future and therefore paves the way for broad
application. Beyond the quantification of probiotic bacteria, the method could be a valuable tool
to examine the viability of pathogenic microorganism in biofilms. The growing number of
sequenced microorganisms will increase the knowledge of gene expression patterns and broaden
the applicability of rapid and accurate molecular quantification tools which will lead to
sophisticated quantification of viable bacteria independent of the morphological state.
In the last part of the thesis, we used a different approach to identify stress tolerant B. longum
strains. We aimed at developing an efficient strategy to screen a number of B. longum strains
with intrinsic resistance to conditions of spray drying and accelerated powder storage. We
hypothesized the possibility to apply a number of B. longum strains to spray drying and storage
Chapter 5 - Conclusion and Outlook
- 114 -
experiment simultaneously and identify the best survivor strain(s) using random amplified
polymorphic DNA (RAPD) method at the end of the storage test. Twenty-two strains of
bifidobacteria belonging to B. longum species could be distinguished using this approach that
allowed identifying one strain with enhanced survival rate to drying and storage. This higher
resistance was confirmed by single strain spray drying and storage experiment and by comparing
the results with a poor survivor strain.
The rapid screening protocol developed for the spray drying and storage approach can be
projected to various other stress tests relevant for production and consumption of probiotic food
products. An additional empirical approach with the same 22 B. longum strains applied to freeze
drying and subsequent accelerated storage might deliver important insights about parallels and
differences of survival of each strain to spray- and freeze-drying followed by storage.
5.2 Outlook and perspectives
The technologies and protocols applied in the thesis can be also applied alone or combined in
new approaches to produce or select stress tolerant probiotic strains. Continuous culture with
immobilized cells for instance can be applied to other probiotic strains to test whether induction
of aggregation and bile tolerance is strain dependent or not. The high throughput screening
protocol could be used to monitor survival to simulated gastrointestinal conditions or even in
vivo, with humans or gnotobiotic mice being administered with a number of probiotic strains
which are distinguishable by their RAPD-fingerprint and samples of feces or biopsies from the
GIT of mice could be analyzed to identify the strain with the longest residence time within the
host. If administered to individuals with an intact microflora, a pre-selection step with selective
media in combination with morphological analyzes would be necessary to select B. longum
Chapter 5 - Conclusion and Outlook
- 115 -
strains before applying picked colonies to RAPD analyzes. Another approach exploits the
potential of probiotic in co-cultures. Co-cultures with bifidobacteria were shown to induce stress
tolerance or production of chaperones in B. longum and thus the effect of different co-culture
combinations or ratios could be analyzed with immobilized cell technique and continuous
culture. Biomass of probiotic strains with different probiotic properties could be produced in
separate continuous reactors containing bacteria in gel beads and produced under controlled
conditions. Biomass produced in immobilized chemostat cultures would be used for controlled
inoculation of reactors run in batch mode in co-culture and simultaneously for monoculture in
batch mode as reference (Figure 19 A). The system would allow efficient analysis of effect of co-
cultures on cell physiology or on gene-/ protein expression. One important benefit of cell
immobilization is the ability to control and stability of different strains within a mixed
fermentation process (Doleyres et al., 2004b; Lacroix and Yildirim, 2007).
Another experiment would be the exposure of sub-lethal stress additional to cell immobilization
in a two-stage fermentation system and cell continuous culture (Figure 19 B). The first reactor is
used to efficiently produce active biomass with cells immobilized in gel beads. The released cells
from gel beads are pumped in a second reactor where they are exposed to sub-lethal stress
treatment similar as used with continuous free cell system recently described and validated by
Mozzetti (2009). The advantage of using two-stage continuous system with immobilized cells is
the stability and robustness of the system combined with an adjustable (high) dilution rate
without running the risk of washing out active biomass. The different stresses cells encounter
during growth in gel beads and sub-lethal treatment may lead to a synergistic stress response and
result in cross-protection.
Chapter 5 - Conclusion and Outlook
- 116 -
Figure 19 Possible fermentation systems to produce probiotic bacteria with enhanced
technological and functional characteristics. (A) A number strains are produced using continuous
culture with IC in separate systems and effluent used to inoculate mixed batch co-cultures at
different proportions for fast screening of strain interaction in the second reactor stages. (B) One
strain is continuously produced in a first reactor containing IC and used to inoculate a second
reactor, where cells are exposed to sub-lethal stresses (Lacroix and Yildirim, 2007).
Chapter 6 - Appendix
- 117 -
6 Appendix
6.1 Supplementary material from Chapter 4
Figure S1 RAPD patterns of 22 B. longum strains. Patterns obtained with the primers OPA-
02; OPL-07 and OPL-16. ML: Low mol. weight DNA marker; MH: High mol. weight DNA
marker.
Chapter 6 - Appendix
- 118 -
6.2 Genome expression analysis of B. longum NCC2705
produced during continuous culture with immobilized cells
and batch culture
6.2.1 Background
Global gene expression or comparison of protein profiles are valuable tools in the field of
comparative analysis between suspended and immobilized bacteria. Numerous studies compared
differences of gene expression profiles from cells within biofilms and in planktonic state and
reported changes only in small numbers of genes (Whiteley et al., 2001, Schembri et al., 2003).
However, only few studies describe the comparison of artificially immobilized microbial cells
and cells from free cell culture. Global proteomic analyzes of Gram positive and Gram negative
bacteria of sessile (immobilized) cells on various carriers compared to free (planktonic) cells
revealed considerable alterations with different expression of detected protein spots ranging from
3 to more than 50 % (Junter and Jouenne, 2004).
In our study, transcriptomic analyzes of samples from both ICC cultures and free cells batch
cultures were compared among each other to characterize the effect of 2-stage continuous IC
cultures with respect to sugar content of the feed, to culture age and to growth in immobilized
state. Gene expression patterns were compared with morphological and physiological
modifications of cells from IC continuous culture.
6.2.2 Material and methods
Samples were collected from both ICC as described in Chapter 2. Genome wide expression
analysis between continuously produced IC cells from R2 was carried only as single experiments
Chapter 6 - Appendix
- 119 -
for the following hybridizations: “ICC1 day 8 - Batch (ES)”; “ICC1 R2 day 8 –ICC2 R2 day 8”
and “ICC2 R2 day 20 – ICC2 R2 day 8” (Figure S2).
Figure S2 Hybridization scheme. Each arrow represents one hybridization experiment.
Abbreviations: “ES” early stationary and “MS” mid stationary growth phase
Microarray design and RNA extraction DNA based arrays, produced by Agilent Technologies
(www.agilent.com), were obtained by in situ synthesis of 60 mer oligonucleotides on glass slides
(Wolber et al., 2006). For each gene, 3 to 6 different probes were randomly distributed on the
array. Total RNA was extracted with the Macaloid method and purified as previously described
(Parche et al., 2006).
For each hybridization, cDNA was synthesized starting from 4 µg of total RNA and
subsequently labeled using the Array 900MPX Genisphere kit (Genisphere Inc., Hatfield, PE,
USA), following the protocol provided by the supplier. Luciferase and kanamycin control
ICC2day 8
ICC2day 20
BatchES
ICC1day 8
BatchMS
ICC2day 8ICC2day 8
ICC2day 20ICC2day 20
BatchES
BatchES
ICC1day 8ICC1day 8
BatchMS
BatchMS
Chapter 6 - Appendix
- 120 -
mRNA (Promega, Zurich, Switzerland) at 1 and 10 ng, respectively, were mixed with total RNA
before labeling to allow balancing of the two channels during scanning. After the hybridization
procedure, array slides were scanned at 10 µm using a Scanarray 4000 (Packard Biochip
Technologies, Billerica, MA, USA). Parameters (laser power and photomultiplier tube gain) were
set in order to prevent saturation of any spot, except the probes corresponding to rRNA.
Data extracted with Imagene 5.6 (Biodiscovery, El Segundo, CA, USA) were treated with
homemade scripts in Python language (www.python.org) and a local installation of the ArrayPipe
web server (Hokamp et al., 2004). Probes showing a signal smaller than twice the standard
deviation of the local background were considered without signal. Probes showing no signal or
saturated signals in both channels were discarded from the analysis. Assuming an intensity-
dependent variation in dye signal, (limma) loess global normalization was applied on signal
ratios. To calculate average gene expression values, data from different probes and hybridization
duplicates (with dye swap) were combined as follows. Within each hybridization data set, gene
fold changes were calculated from the median of the corresponding probes values. The
expression value of a gene was retained if a signal was detected in at least 50 % of its probes. The
average gene expression values were then calculated by combining data from two independent
hybridizations. Genes were considered to be differentially expressed if their log2-transformed
signal ratios were higher than 1.0 or smaller than -1.0. B. longum genes organized in operons
were identified according to Price et al. (2006) (http://www.microbesonline.org/operons/
OperonList.html).
Chapter 6 - Appendix
- 121 -
6.2.3 Results and discussion
All hybridization experiments were only carried out once without dye swap confirmation and
hence data are of limited significance, but general trends can be elucidated.
6.2.3.1 Comparison of cells from ICC with batch cultures
Cells from the second reactor produced during continuous cultures with immobilized cells were
compared with free cells batch cultures from early- and mid stationary growth phase. These
comparisons were carried out to find a correlation between gene expression profiles and
physiological changes including fatty acid compositions of cell membrane. Gene expression of
cells from ICC1 collected at day 8 were compared to cells from early stationary phase and
showed a relative expression of 1587 (92 %) ORFs of which 229 (14 %) genes were differently
expressed. When compared with cells from mid stationary phase a relative expression of 1347
ORFs (78 %) was reported among which 149 (11 %) were differentially expressed (Table S1).
Relative changes in gene expression of cells from continuous IC culture were more pronounced
when compared with cells from early stationary batch cells than when compared with cells from
mid stationary growth phase (Tables S2 a, b).
Gene expression of cells from ICC1 exhibited less differently expressed ORFs when compared
with batch cells from mid- than with cells from early stationary growth phase. This was expected
both cultures, from R2 of ICC1 and from batch mid stationary growth phase, face comparable
time periods in starvation conditions. ICC cultures indeed showed an increased expression of
proteins involved in the ribosomal complex compared to batch free cell cultures and may be a
result of sensing environmental changes and lead to modulated synthesis of metabolic enzymes
affecting cell physiology (Chen et al., 2003, Wilson and Nierhaus, 2005). Bile tolerant ICC cells
Chapter 6 - Appendix
- 122 -
did not show increased transcript levels of neither BL1102 (Na+ dependent nucleoside
transporter), a putative cholate transporter, nor BL0796 (choloylglycine hydrolase), a bile salt
hydrolase. Bile tolerance is therefore most likely attributed to changes in membrane composition.
These first gene expression profile comparisons showed that cells from ICC did neither show up-
regulation of ORFs related to stress response as reported for heat adapted B. longum NCC2705
(Rezzonico et al., 2007), nor different expression of (identified) enzymes involved in fatty acid
production (Schell et al., 2002). However, it is known that transcript level and proteins content
do not necessarily correlate since their relationship depends strongly on time, cellular location,
stability of the molecules (Ghigo, 2003) and thus an enzyme-mediated bile resistance in ICC cells
cannot be excluded.
Table S1 Summary of hybridization experiments. Absolute and relative number of totally
expressed, under- and over expressed genes.
a ES: early stationary growth phase; MS: mid stationary growth phase * Differentlly expressed, when over-/underexpression ≥ 2 fold.
6.2.3.2 Comparison between ICC1 and ICC2 and within ICC2
Gene expression profiles of cells from continuous cultures with immobilized cells with and
without glucose limitation collected at day 8 were compared and a relative expression value of
approximately 58 % (1007 genes) of the ORFs was found with 170 genes (17 %) differently
Hybridization a Control Treated Expr. Expr. (%)
Over-expr.
under-expr.
Diff. expr.*
ICC1 day 8 - Batch (ES) Batch (ES) ICC1 day8 1587 91.9 102 127 229
ICC1 day 8 - Batch (MS) Batch (MS) ICC1 day8 1347 78.0 85 64 149
ICC2 day 8 –ICC1 day 8 ICC1 day 8 ICC2 day 8 1007 58.3 69 101 170
ICC2 day 20 – ICC2 day 8 ICC2 day 8 ICC2 day 20 1541 89.2 52 47 99
Chapter 6 - Appendix
- 123 -
expressed (Table S1). The most over- and underexpressed genes were both hypothetical protein
BL1055a (hypothetical protein) and BL0981 (hypothetical protein with helix turn helix motif),
respectively, (Table S3). Among the 10 most repressed genes of cells from ICC2, 4 genes are
involved in sugar transport or metabolism. The sugar limitation in ICC1 may induce a
considerable increase of sugar transporter proteins compared to cells which are not sugar limited
(ICC2) in order to open gates for possible energy resources in the environment.
DNA micro array analysis between cells from ICC2 collected at day 20 containing large cell
aggregates were compared to cells collected before aggregation at day 8 and enabled assessment
of relative expression value of approximately 89% (1541) of ORFs (Table S1). In total, 99 genes
(6 %) were differentially expressed. The most over- and underexpressed genes were BL1040
(hypothetical protein) and BL1694 (probable sugar binding protein of ABC transporter for
pentoses), respectively (Table S4). Comparison of aggregated cells from ICC2 day 20 showed a
down regulation of transporter proteins of pentoses. The membrane proximity of high cell density
environment in macroscopic cell aggregates may induce quorum sensing (Sturme et al., 2002)
and result in a down-regulation of sugar transport proteins.
Although gene-expression experiments are only preliminary results, the data support some
physiological observations and may confer the understanding of molecular mechanisms
underlying increased bile tolerance and formation of auto-aggregation. However, further
hybridizations are required (including dye-swap experiments) to confirm obtained global gene
expression patterns of cells from ICC compared to cells from batch control cultures.
Chapter 6 - Appendix
- 124 -
Table S2a The 10 most over- and underexpressed genes represented by the factor of
induction/repression of cells from ICC1 Day 8 compared to Batch ES.
Fold change
Locus Product Name Operon
Over-expressed 9.3 BL1631 D-Glucose-proton symporter BL1630-1631
7.5 BL0105 beta-fructofuranosidase (sucrase/invertase); possible inulinase BL0105-0107
6.7 BL0106 sucrose transport protein BL0105-0107 4.9 BL1493 possible WhiB-like transcription factor BL1493-1494 4.9 BL1494 hypo. protein BL1494 BL1493-1494
4.6 BL1492 hypo. protein with similarity to the Par protein of B. breve plasmid pcibb1
-
4.3 BL1100 30S ribosomal protein S12 BL1097-1100 3.7 BL0624 hypo. protein BL0624 BL0624-0625 3.7 BL0573 hypo. protein BL0573 - 3.5 BL1586 50S ribosomal protein L16 BL1577-1604
Under-expressed -9.8 BL0036 probable ABC transporter permease protein for sugars BL0033-0036 -9.8 BL0188 solute binding protein of ABC transporter system BL0188-0190 -9.8 BL0035 probable ABC transport system permease protein for sugars BL0033-0036 -8.0 BL1329 alpha-mannosidase BL1330-1333 -7.5 BL1330 probable solute-binding protein of ABC transporter system BL1330-1333 -7.0 BL0034 ATP binding protein of ABC transporter BL0033-0036 -6.5 BL1331 probable sugar permease of ABC transporter system BL1330-1333 -6.5 BL0813 hypo. protein BL0813 -
-6.1 BL1335 probable endo-beta-N-acetylglucosaminidase; di-N-acetylchitobiosyl beta-N-acetylglucosaminidase BL1334-1335
-5.7 BL1165 probable solute binding protein of ABC transporter system for sugars -
Chapter 6 - Appendix
- 125 -
Table S2b The 10 most over- and underexpressed genes represented by the factor of
induction/repression of cells from ICC1 Day 8 compared to Batch MS.
Fold change
Locus Product Name Operon
Over-expressed 6.6 BL1631 D-Glucose-proton symporter BL1630-1631 6.1 BL0555 possible DO serine protease -
5.8 BL0105 beta-fructofuranosidase (sucrase/invertase); possible inulinase
BL0105-0107
5.4 BL0106 sucrose transport protein BL0105-0107 5.1 BL1594 30S ribosomal protein S8 BL1577-1604 4.9 BL1595 50S ribosomal protein L6 BL1577-1604 4.8 BL1598 50S ribosomal protein L30 BL1577-1604 4.6 BL1596 50S ribosomal protein L18 BL1577-1604 4.2 BL0358 ATP synthase gamma chain BL0356-0363 4.2 BL1586 50S ribosomal protein L16 BL1577-1604
Under-expressed -9.3 BL0036 probable ABC transporter permease protein for sugars BL0033-0036 -9.3 BL0034 ATP binding protein of ABC transporter BL0033-0036 -8.0 BL0035 probable ABC transport system permease protein for sugars BL0033-0036
-6.3 BL0033 probable solute binding protein of ABC transporter system possibly for sugars
BL0033-0036
-6.1 BL1340 probable transcriptional repressor in the Rok (NagC/XylR) family
BL1339-1340
-4.3 BL1772 sugar kinase in PfkB family BL1772-1773 -3.9 BL1771 C4-dicarboxylate transporter BL1770-1771
-3.6 BL1325 narrowly conserved hypo. protein with possible DNA interaction motif
-
-3.2 BL1770 widely conserved hypo. protein in the inosine-uridine preferring nucleoside hydrolase family
BL1770-1771
-3.2 BL1773 hypothetical protein BL1773 BL1772-1773
Chapter 6 - Appendix
- 126 -
Table S3 The 10 most over- and underexpressed genes represented by the factor of
induction/repression of cells from ICC2 Day 8 compared to ICC1 Day 8.
Fold change
Locus Product Name Operon
Over-expressed 6.2 BL1055a hypo. protein BL1055a BL1054-1055a 5.1 BL1161 solute bind. protein of ABC transp. system for peptides BL1157-1161 4.1 BL0945 ATP-dependent Clp protease proteolytic subunit 1 BL0943-0945 4.1 BL1183 FtsE-like ATP binding protein involved in cell division BL1182-1184 4.0 BL0813 hypo. protein BL0813 - 4.0 BL1182 FtsX-like protein involved in cell division BL1182-1184 3.5 BL0065 elongation factor P BL0065-0066 3.5 BL1062 acetylglutamate kinase BL1058-1068
3.0 BL1714 solute bind. protein of ABC transporter for branched-chain amino acids BL1714-1718
2.9 BL1071 narrowly conserved hypo. protein BL1069-1071
Under-expressed -39.4 BL0981 hypo. protein with helix turn helix motif - -22.3 BL0106 sucrose transport protein BL0105-0107
-17.4 BL0105 beta-fructofuranosidase (sucrase/invertase); possible inulinase BL0105-0107
-13.8 BL0980 hypo. protein BL0980 -
-11.6 BL1164 prob. solute binding protein of ABC transporter system for sugars BL1163-1164
-8.2 BL1445 narrowly conserved hypo. protein ? -8.0 BL0428 hypo. protein BL0428 BL0428-0431 -8.0 BL0595 hypo. protein BL0595 -
-7.8 BL1694 probable sugar binding protein of ABC transporter for pentoses
BL1694-1696
-7.2 BL1664 widely conserved protein in universal stress protein family -
Chapter 6 - Appendix
- 127 -
Table S4 The 10 most over- and underexpressed genes represented by the factor of
induction/repression of cells from ICC2 Day 20 compared to ICC2 Day 8.
Fold change
Locus Product Name Operon
Over-expressed 19.3 BL1040 hypo. protein BL1040 - 17.4 BL1041a possible GntR-family transcriptional regulator BL1041-1041a 13.6 BL1041 ATP binding protein of ABC transporter BL1041-1041a 7.4 BL0435 possible ammonium ion transporter BL0433-0435 6.9 BL1076 glutamine synthetase 1 - 5.4 BL1177 permease protein of ABC transporter system BL1176-1179 5.3 BL0011 cytosine deaminase - 4.0 BL1499 isocitrate dehydrogenase - 3.3 BL0002 chaperonin GroEL - 3.3 BL1404 hypothetical protein BL1404 -
Under-expressed
-5.8 BL1694 probable sugar binding protein of ABC transporter for pentoses
BL1694-1696
-5.3 BL1695 ATP binding protein of ABC transporter for pentoses BL1694-1696 -3.8 BL0923 hypo. protein BL0923 - -3.8 BL0932 possible ATP binding protein of ABC transporter BL0931-0932 -3.7 BL1055a hypo. protein BL1055a BL1054-1055a -3.6 BL0924 hypo. protein BL0924 - -3.4 BL0950 pyruvate formate-lyase 1 activating enzyme BL0947-0951 -3.2 BL1696 probable ABC transport system permease protein for sugars BL1694-1696 -3.2 BL0925 hypo. protein BL0925 - -3.1 BL0980 hypo. protein BL0980 -
Chapter 7 - References
- 128 -
7 References
Abadias, M., Teixido, N., Usall, J., Benabarre, A. and Vinas, I., 2001. Viability, efficacy, and storage stability of freeze-dried biocontrol agent Candida sake using different protective and rehydration media. Journal of Food Protection 64, 856-861.
Adhikari, K., Mustapha, A., Grun, I. U. and Fernando, L., 2000. Viability of microencapsulated
bifidobacteria in set yogurt during refrigerated storage. Journal of Dairy Science 83, 1946-1951.
Adhikari, K., Mustapha, A. and Grun, I. U., 2003. Survival and metabolic activity of
microencapsulated Bifidobacterium longum in stirred yogurt. Journal of Food Science 68, 275-280.
Aellen, S., Que, Y.A., Guignard, B., Haenni, M. and Moreillon, P., 2006. Detection of live and
antibiotic-killed bacteria by quantitative real-time PCR of specific fragments of rRNA. Antimicrobial Agents and Chemotherapy 50, 1913-1920.
Ahmed, M., Prasad, J., Gill, H., Stevenson, L. and Gopal, P., 2007. Impact of consumption of
different levels of Bifidobacterium lactis HN019 on the intestinal microflora of elderly human subjects. Journal of Nutrition, Health and Aging 11, 26-31.
Ahn, J. B., Hwang, H. J. and Park, J. H., 2001. Physiological responses of oxygen-tolerant
anaerobic Bifidobacterium longum under oxygen. Journal of Microbiology and Biotechnology 11, 443-451.
Alakomi, H. L., Mättö, J., Virkajärvi, I. and Saarela, M., 2005. Application of a microplate scale
fluorochrome staining assay for the assessment of viability of probiotic preparations. Journal of Microbiological Methods 62, 25-35.
Alander, M., Mättö, J., Kneifel, W., Johansson, M., Kogler, B., Crittenden, R., Mattila-
Sandholm, T. and Saarela, M., 2001. Effect of galacto-oligosaccharide supplementation on human faecal microflora and on survival and persistence of Bifidobacterium lactis Bb-12 in the gastrointestinal tract. International Dairy Journal 11, 817-825.
Anal, A. K. and Singh, H., 2007. Recent advances in microencapsulation of probiotics for
industrial applications and targeted delivery. Trends in Food Science & Technology 18, 240-251.
Ananta, E., Volkert, M. and Knorr, D., 2005. Cellular injuries and storage stability of spray-dried
Lactobacillus rhamnosus GG. International Dairy Journal 15, 399-409. Arnaud, J. P., Lacroix, C. and Castaigne, F., 1992. Counterdiffusion of lactose and lactic acid in
κ-carrageenan locust bean gum gel beads with or without entrapped lactic acid bacteria. Enzyme and Microbial Technology 14, 715-724.
Chapter 7 - References
- 129 -
Arunachalam, K., Gill, H. S. and Chandra, R. K., 2000. Enhancement of natural immune function by dietary consumption of Bifidobacterium lactis (HN019). European Journal of Clinical Nutrition 54, 263-267.
Asahara, T., Shimizu, K., Nomoto, K., Hamabata, T., Ozawa, A. and Takeda, Y., 2004. Probiotic
bifidobacteria protect mice from lethal infection with shiga toxin-producing Escherichia coli O157 : H7. Infection and Immunity 72, 2240-2247.
Augustin, M.A., and L. Sanguansri, 2008. Encapsulation of Bioactives. In Food Materials
Science, Principles and Practice, edited by J. M. Aguilera and P. J. Lillford. New-York: Springer, pp.577-601.
Auty, M. A. E., Gardiner, G. E., McBrearty, S. J., O'Sullivan, E. O., Mulvihill, D. M., Collins, J.
K., Fitzgerald, G. F., Stanton, C. and Ross, R. P., 2001. Direct in situ viability assessment of bacteria in probiotic dairy products using viability staining in conjunction with confocal scanning laser microscopy. Applied and Environmental Microbiology 67, 420-425.
Backhed, F., Ding, H., Wang, T., Hooper, L. V., Koh, G. Y., Nagy, A., Semenkovich, C. F. and
Gordon, J. I., 2004. The gut microbiota as an environmental factor that regulates fat storage. Proceedings of the National Academy of Sciences of the United States of America 101, 15718-15723.
Bartosch, S., Woodmansey, E. J., Paterson, J. C. M., McMurdo, M. E. T. and Macfarlane, G. T.,
2005. Microbiological effects of consuming a synbiotic containing Bifidobacterium bifidum, Bifidobacterium lactis, and oligofructose in elderly persons, determined by real-time polymerase chain reaction and counting of viable bacteria. Clinical Infectious Diseases 40, 28-37.
Begley, M., Gahan, C. G. M. and Hill, C., 2005. The interaction between bacteria and bile. FEMS
Microbiology Reviews 29, 625-651. Belasco, J. G., Nilsson, G., Vongabain, A., Cohen, S. N., 1986. The stability of Escherichia coli
gene transcripts is dependent on determinants localized to specific messenger-RNA segments. Cell 46, 245-251.
Ben Amor, K., Breeuwer, P., Verbaarschot, P., Rombouts, F. M., Akkermans, A. D. L., De Vos,
W. M. and Abee, T., 2002. Multiparametric flow cytometry and cell sorting for the assessment of viable, injured, and dead Bifidobacterium cells during bile salt stress. Applied and Environmental Microbiology 68, 5209-5216.
Ben Amor, K., Vaughan, E.E., and de Vos, W.M. 2007. Advanced molecular tools for the
identification of lactic acid bacteria. Journal of Nutrition 137, 741S-747S. Bergmaier, D., Champagne, C. P. and Lacroix, C., 2005. Growth and exopolysaccharide
production during free and immobilized cell chemostat culture of Lactobacillus rhamnosus RW-9595M. Journal of Applied Microbiology 98, 272-284.
Chapter 7 - References
- 130 -
Bernstein, J. A., Khodursky, A. B., Lin, P. H., Lin-Chao, S., and Cohen, S. N. (2002) Global analysis of mRNA decay and abundance in Escherichia coli at single-gene resolution using two-color fluorescent DNA microarrays. Proceedings of the National Academy of Sciences of the United States of America 99, 9697–9702.
Bezkorovainy, A., 2001. Probiotics: determinants of survival and growth in the gut. American
Journal of Clinical Nutrition 73, 399S-405S. Biavati, B., Vescovo, M., Torriani, S. and Bottazzi, V., 2000. Bifidobacteria: history, ecology,
physiology and applications. Annals of Microbiology 50, 117-131. Bibiloni, R., Zavaglia, A. G. and De Antoni, G., 2001. Enzyme-based most probable number
method for the enumeration of Bifidobacterium in dairy products. Journal of Food Protection 64, 2001-2006.
Bielecka, M. and Majkowska, A., 2000. Effect of spray drying temperature of yoghurt on the
survival of starter cultures, moisture content and sensoric properties of yoghurt powder. Food 44, 257-260.
Bin-Nun, A., Bromiker, R., Wilschanski, M., Kaplan, M., Rudensky, B., Caplan, M. and
Hammerman, C., 2005. Oral probiotics prevent necrotizing enterocolitis in very low birth weight neonates. Journal of Pediatrics. 147, 192-196.
Birmingham, P., Helm, J. M., Manner, P. A. and Tuan, R. S., 2008. Simulated joint infection
assessment by rapid detection of live bacteria with real-time reverse transcription polymerase chain reaction. Journal of Bone and Joint Surgery 90A, 602-608.
Bleve, G., Rizzotti, L., Dellaglio, F., Torriani, S., 2003. Development of reverse transcription
(RT)-PCR and real-time RT-PCR assays for rapid detection and quantification of viable yeasts and molds contaminating yogurts and pasteurized food products. Applied and Environmental Microbiology 69, 4116-4122.
Breeuwer, P. and Abee, T., 2000. Assessment of viability of microorganisms employing
fluorescence techniques. International Journal of Food Microbiology 55, 193-200. Bryan, L. E., Ohara, K. and Wong, S., 1984. Lipopolysaccharide changes in impermeability-type
aminoglycoside resistance in Pseudomonas aeruginosa. Antimicrobial Agents and Chemotherapy 26, 250-255.
Buitink, J., van den Dries, I. J., Hoekstra, F. A., Alberda, M. and Hemminga, M. A., 2000. High
critical temperature above T-g may contribute to the stability of biological systems. Biophysical Journal 79, 1119-1128.
Burin, L., Jouppila, K., Roos, Y. H., Kansikas, J. and Buera, M. P., 2004. Retention of beta-
galactosidase activity as related to Maillard reaction, lactose crystallization, collapse and glass transition in low moisture whey systems. International Dairy Journal 14, 517-525.
Chapter 7 - References
- 131 -
Bustin, S. A., 2000. Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. Journal of Molecular Endocrinolology 25, 169-193.
Bustin, S. A., 2002. Quantification of mRNA using real-time reverse transcription PCR (RT-
PCR): trends and problems. Journal of Molecular Endocrinolology 29, 23–39. Cachon, R., Anterieux, P. and Diviès, C., 1998. The comparative behavior of Lactococcus lactis
in free and immobilized culture processes. Journal of Biotechnology 63, 211-218. Campbell, E. A., Korzheva, N., Mustaev A., Murakami K., Nair S., Goldfarb A., Darst S. A.,
2001. Structural mechanism for rifampicin inhibition of bacterial RNA polymerase. Cell 104, 901-912.
Canzi, E., Guglielmetti, S., Mora, D., Tamagnini, I. and Parini, C., 2005. Conditions affecting
cell surface properties of human intestinal bifidobacteria. Antonie Van Leeuwenhoek International Journal of General and Molecular Microbiology 88, 207-219.
Caplan, M. S., Miller-Catchpole, R., Kaup, S., Russell, T., Lickerman, M., Amer, M., Xiao, Y.
and Thomson, R., Jr., 1999. Bifidobacterial supplementation reduces the incidence of necrotizing enterocolitis in a neonatal rat model. Gastroenterology 117, 577-583.
Carvalho, A. S., Silva, J., Ho, P., Teixeira, P., Malcata, F. X. and Gibbs, P., 2004. Relevant
factors for the preparation of freeze-dried lactic acid bacteria. International Dairy Journal 14, 835-847.
Castagliuolo, I., Galeazzi, F., Ferrari, S., Elli, M., Brun, P., Cavaggioni, A., Tormen, D.,
Sturniolo, G. C., Morelli, L. and Palu, G., 2005. Beneficial effect of auto-aggregating Lactobacillus crispatus on experimentally induced colitis in mice. FEMS Immunology and Medical Microbiology 43, 197-204.
Castro, H. P., Teixeira, P. M. and Kirby, R., 1995. Storage of lyophilized cultures of
Lactobacillus bulgaricus under different relative humidities and atmospheres. Applied Microbiology and Biotechnology 44, 172-176.
Castro, H. P., Teixeira, P. M. and Kirby, R., 1996. Changes in the cell membrane of
Lactobacillus bulgaricus during storage following freeze-drying. Biotechnology Letters 18, 99-104.
Cesena, C., Morelli, L., Alander, M., Siljander, T., Tuomola, E., Salminen, S., Mattila-Sandholm,
T., Vilpponen-Salmela, T., von Wright, A., 2001. Lactobacillus crispatus and its nonaggregating mutant in human colonization trials. Journal of Dairy Science 84, 1001-1010.
Champagne, C. P., Mondou, F., Raymond, Y. and Roy, D., 1996. Effect of polymers and storage
temperature on the stability of freeze-dried lactic acid bacteria. Food Research International 29, 555-562.
Chapter 7 - References
- 132 -
Champagne, C. P., Gardner, N. J. and Roy, D., 2005. Challenges in the addition of probiotic cultures to foods. Critical Reviews in Food Science and Nutrition 45, 61-84.
Champagne, C. P. and Fustier, P., 2007. Microencapsulation for the improved delivery of
bioactive compounds into foods. Current Opinion in Biotechnology 18, 184-190. Chavez, B. E. and Ledeboer, A. M., 2007. Drying of probiotics: Optimization of formulation and
process to enhance storage survival. Drying Technology 25, 1193-1201. Cheikhyoussef, A., Pogori, N., Chen, W. and Zhang, H., 2008. Antimicrobial proteinaceous
compounds obtained from bifidobacteria: From production to their application. International Journal of Food Microbiology 125, 215-222.
Chen, H. C., Teplitski, M., Robinson, J. B., Rolfe, B. G. and Bauer, W. D., 2003. Proteomic
analysis of wild-type Sinorhizobium meliloti responses to N-acyl homoserine lactone quorum-sensing signals and the transition to stationary phase. Journal of Bacteriology 185, 5029-5036.
Chouraqui, J.P., Van Egroo, L.D. and Fichot, M.C., 2004. Acidified milk formula supplemented
with Bifidobacterium lactis: Impact on infant diarrhea in residential care settings. Journal of Pediatric Gastroenterology and Nutrition 38, 288-292.
Cinquin, C., Le Blay, G., Fliss, I. and Lacroix, C., 2004. Immobilization of infant fecal
microbiota and utilization in an in vitro colonic fermentation model. Microbial Ecology 48, 128-138.
Coakley, M., Banni, S., Johnson, M. C., Mills, S., Devery, R., Fitzgerald, G., Ross, R. P. and
Stanton, C., 2009. Inhibitory Effect of Conjugated alpha-Linolenic Acid from Bifidobacteria of Intestinal Origin on SW480 Cancer Cells. Lipids 44, 249-256.
Collado, M. C., Moreno, Y., Cobo, J. M. and Hernandez, M., 2006. Microbiological evaluation
and molecular characterization of bifidobacteria strains in commercial fermented milks. European Food Research and Technology 222, 112-117.
Collado, M. C., Meriluoto, J. and Salminen, S., 2007. Measurement of aggregation properties
between probiotics and pathogens: In vitro evaluation of different methods. Journal of Microbiological Methods 71, 71-74.
Corcoran, B. M., Ross, R. P., Fitzgerald, G. F. and Stanton, C., 2004. Comparative survival of
probiotic lactobacilli spray-dried in the presence of prebiotic substances. Journal of Applied Microbiology 96, 1024-1039.
Corcoran, B.M., Stanton, C., Fitzgerald, G., and Ross, R.P., 2008. Life under stress: The
probiotic stress response and how it may be manipulated. Current Pharmaceutical Design 14, 1382-1399.
Chapter 7 - References
- 133 -
Costa, E., Usall, J., Teixido, N., Garcia, N. and Vinas, I., 2000. Effect of protective agents, rehydration media and initial cell concentration on viability of Pantoea agglomerans strain CPA-2 subjected to freeze-drying. Journal of Applied Microbiology 89, 793-800.
Costerton, J. W., Stewart, P. S. and Greenberg, E. P., 1999. Bacterial biofilms: A common cause
of persistent infections. Science 284, 1318-1322. Coutard, F., Pommepuy, M., Loaec, S. and Hervio-Heath, D., 2005. mRNA detection by reverse
transcription-PCR for monitoring viability and potential virulence in a pathogenic strain of Vibrio parahaemolyticus in viable but nonculturable state. Journal of Applied Microbiology 98, 951-961.
Coutard, F., Lozach, S., Pommepuy, M., Hervio-Heath, D., 2007. Real-time reverse transcription-
PCR for transcriptional expression analysis of virulence and housekeeping genes in viable but nonculturable Vibrio parahaemolyticus after recovery of culturability. Applied and Environmental Microbiology 73, 5183-5189.
Crociani, F., Biavati, B., Alessandrini, A., Chiarini, C. and Scardovi, V., 1996. Bifidobacterium
inopinatum sp. nov and Bifidobacterium denticolens sp nov, two new species isolated from human dental caries. International Journal of Systematic Bacteriology 46, 564-571.
Crowe, J. H., Crowe, L. M., Carpenter, J. F. and Wistrom, C. A., 1987. Stabilization of dry
phospholipid-bilayers and proteins by sugars. Biochemistry Journal 242, 1-10. Daley, R.J., 1977. Direct epifluorescence enumeration of native aquatic bacteria: uses,
limitations, and comparative accuracy. In: Costerton J. W., Colwell. R. R., Native Aquatic Bacteria: Enumeration, Activity and Ecology. American Society for Testing and Materials, Philadelphia, pp. 29–45.
Davey, M. E. and O'Toole, G. A., 2000. Microbial biofilms: from ecology to molecular genetics.
Microbiology and Molecular Biology Reviews 64, 847-867. Del Re, B., Sgorbati, B., Miglioli, M. and Palenzona, D., 2000. Adhesion, autoaggregation and
hydrophobicity of 13 strains of Bifidobacterium longum. Letters in Applied Microbiology 31, 438-442.
de Man, J. D., Rogosa, M., Sharpe, M. E., 1960. A medium for the cultivation of Lactobacilli.
Journal of. Applied Bacteriology 23, 130–135. Denich, T. J., Beaudette, L. A., Lee, H. and Trevors, J. T., 2003. Effect of selected environmental
and physico-chemical factors on bacterial cytoplasmic membranes. Journal of Microbiological Methods 52, 149-182.
De Schryver, P., Crab, R., Defoirdt, T., Boon, N. and Verstraete, W., 2008. The basics of bio-
flocs technology: The added value for aquaculture. Aquaculture 277, 125-137.
Chapter 7 - References
- 134 -
De Smet, I., De Boever, P. and Verstraete, W., 1998. Cholesterol lowering in pigs through enhanced bacterial bile salt hydrolase activity. British Journal of Nutrition 79, 185-194.
Desmond, C., Stanton, C., Fitzgerald, G. F., Collins, K. and Ross, R. P., 2001. Environmental
adaptation of probiotic lactobacilli towards improvement of performance during spray drying. International Dairy Journal 11, 801-808.
Desmond, C., Ross, R. P., O'Callaghan, E., Fitzgerald, G. and Stanton, C., 2002. Improved
survival of Lactobacillus paracasei NFBC 338 in spray-dried powders containing gum acacia. Journal of Applied Microbiology 93, 1003-1011.
de Vrese, M., Winkler, P., Rautenberg, P., Harder, T., Noah, C., Laue, C., Ott, S., Hampe, J.,
Schreiber, S., Heller, K., and Schrezenmeir, J., 2005. Effect of Lactobacillus gasseri PA 16/8, Bifidobacterium longum SP 07/3, B. bifidum MF 20/5 on common cold episodes: a double blind, randomized, controlled trial. Clinical Nutrition 24, 481-491.
de Vrese, M. and Schrezenmeir, J.: Probiotics, Prebiotics, and Synbiotics. In Food
Biotechnology. Edited by; 2008:1-66. Advances in Biochemical Engineering / Biotechnology, vol 111.]
Ding, W. K. and Shah, N. P., 2007. Acid, bile, and heat tolerance of free and microencapsulated
probiotic bacteria. Journal of Food Science 72, M446-M450. Ding, W. K. and Shah, N. P., 2009. Effect of Various Encapsulating Materials on the Stability of
Probiotic Bacteria. Journal of Food Science 74, M100-M107. Diviès, C. and Siess, M. H., 1976. Study on L-malic acid catabolism by Lactobacillus casei cells
immobilized into polyacrylamide-gel lattice. Annales De Microbiologie B127, 525-539. Doleyres, Y., Paquin, C., LeRoy, M. and Lacroix, C., 2002a. Bifidobacterium longum ATCC
15707 cell production during free- and immobilized-cell cultures in MRS-whey permeate medium. Applied Microbiology and Biotechnology 60, 168-173.
Doleyres, Y., Fliss, I. and Lacroix, C., 2002b. Quantitative determination of the spatial
distribution of pure- and mixed-strain immobilized cells in gel beads by immunofluorescence. Applied Microbiology and Biotechnology 59, 297-302.
Doleyres, Y., Fliss, I. and Lacroix, C., 2004a. Increased stress tolerance of Bifidobacterium
longum and Lactococcus lactis produced during continuous mixed-strain immobilized-cell fermentation. Journal of Applied Microbiology 97, 527-539.
Doleyres, Y., Fliss, I. and Lacroix, C., 2004b. Continuous production of mixed lactic starters
containing probiotics using immobilized cell technology. Biotechnology Progress 20, 145-150.
Doleyres, Y. and Lacroix, C., 2005. Technologies with free and immobilised cells for probiotic
bifidobacteria production and protection. International Dairy Journal 15, 973-988.
Chapter 7 - References
- 135 -
Edwards, C. A. and Parrett, A. M., 2002. Intestinal flora during the first months of life: new perspectives. British Journal of Nutrition 88, S11 – S18.
Egert, M., de Graaf, A. A., Smidt, H., de Vos, W. M. and Venema, K., 2006. Beyond diversity:
functional microbiomics of the human colon. Trends in Microbiology 14, 86-91. Espina, F. and Packard, V. S., 1979. Survival of Lactobacillus acidophilus in a spray –drying
process. Journal of Food Protection 42, 149-152. Falk, P. G., Hooper, L. V., Midtvedt, T. and Gordon, J. I., 1998. Creating and maintaining the
gastrointestinal ecosystem: What we know and need to know from gnotobiology. Microbiology and Molecular Biology Reviews 62, 1157-+.
Fanaro, S., Chierici, R., Guerrini, P. and Vigi, V., 2003. Intestinal microflora in early infancy:
composition and development. Acta Paediatrica 92, 48-55. FAO/WHO (2001) Evaluation of health and nutritional properties of powder milk with live lactic
acid bacteria. Report from FAO/WHO Expert Consultation, 1–4 October 2001, Cordoba, Argentina.
Favaro-Trindale, C. S. and Grosso, C. R. F., 2002. Microencapsulation of L-acidophilus (La-05)
and B-lactis (Bb-12) and evaluation of their survival at the pH values of the stomach and in bile. Journal of Microencapsulation 19, 485-494.
Freudig, B., Hogekamp, S. and Schubert, H., 1999. Dispersion of powders in liquids in a stirred
vessel. Chemical Engineerin and Processing 38, 525-532. Friedrich, U., Lenke, J., 2006. Improved enumeration of lactic acid bacteria in mesophilic dairy
starter cultures by using multiplex quantitative real-time PCR and flow cytometry-fluorescence in situ hybridization. Applied and Environmental Microbiology 72, 4163-4171.
Fukushima, Y., Kawata, Y., Hara, H., Terada, A. and Mitsuoka, T., 1998. Effect of a probiotic
formula on intestinal immunoglobulin A production in healthy children. International Journal of Food Microbiology 42, 39-44.
Fux, C. A., Costerton, J. W., Stewart, P. S. and Stoodley, P., 2005. Survival strategies of
infectious biofilms. Trends in Microbiology 13, 34-40. Gagnon, M., Kheadr, E. E., Dabour, N., Richard, D. and Fliss, I., 2006. Effect of Bifidobacterium
thermacidophilum probiotic feeding on enterohemorrhagic Escherichia coli O157 : H7 infection in BALB/c mice. International Journal of Food Microbiology 111, 26-33.
Gardiner, G. E., O'Sullivan, E., Kelly, J., Auty, M. A. E., Fitzgerald, G. F., Collins, J. K., Ross,
R. P. and Stanton, C., 2000. Comparative survival rates of human-derived probiotic Lactobacillus paracasei and L-salivarius strains during heat treatment and spray drying. Applied and Environmental Microbiology 66, 2605-2612.
Chapter 7 - References
- 136 -
Gasser, F., 1994. Safety of lactic acid bacteria and their occurrence in human clinical infections. Bulletin De L Institut Pasteur 92, 45-67.
Gevers, D., Huys, G. and Swings, J., 2001. Applicability of rep-PCR fingerprinting for
identification of Lactobacillus species. FEMS Microbiological Letters 205, 31-36. Ghigo, J. M., 2003. Are there biofilm-specific physiological pathways beyond a reasonable
doubt? Research in Microbiology 154, 1-8. Gionchetti, P., Rizzello, F., Helwig, U., Venturi, A., Lammers, K. M., Brigidi, P., Vitali, B.,
Poggioli, G., Miglioli, M. and Campieri, M., 2003. Prophylaxis of pouchitis onset with probiotic therapy: a double-blind, placebo-controlled trial. Gastroenterology 124, 1202-1209.
Goldenberger, D., Kunzli, A., Vogt, P., Zbinden, R. and Altwegg, M., 1997. Molecular diagnosis
of bacterial endocarditis by broad-range PCR amplification and direct sequencing. Journal of Clinical Microbiology 35, 2733-2739
Gonzalez-Escalona, N., Hammack, T. S., Russell, M., Jacobson, A. P., De Jesus, A. J., Brown, E.
W., Lampel, K. A., 2009. Detection of live Salmonella spp. cells in produce by a TaqMan(R) based quantitative reverse transcriptase real-time PCR (qRT-PCR) targeting invA mRNA. Applied and Environmental Microbiology 75, 3714-3720.
Grattepanche, F., Audet, P. and Lacroix, C., 2007. Enhancement of functional characteristics of
mixed lactic culture producing nisin Z and exopolysaccharides during continuous prefermentation of milk with immobilized cells. Journal of Dairy Science 90, 5361-5373.
Grattepanche, F. and Lacroix, C. (in press). Production of high quality probiotics using novel
fermentation and stabilization technologies. In D. Bagchi, F. C. Lau and D. K. Ghosh (Eds.), Biotechnology in Functional Foods and Nutraceuticals. Taylor & Francis Group. Abingdon, UK.
Green, S. L., 1978. Case-report – Fatal anaerobic pulmonary infection due to Bifidobacterium
eriksonii. Postgraduate Medicine 63, 187-192. Guarner, F. and Malagelada, J. R., 2003. Gut flora in health and disease. Lancet 361, 512-519. Guarner, F., Bourdet-Sicard, R., Brandtzaeg, P., Gill, H. S., McGuirk, P., van Eden, W.,
Versalovic, J., Weinstock, J. V. and Rook, G. A. W., 2006. Mechanisms of disease: the hygiene hypothesis revisited. Nature Clinical Practice Gastroenterology & Hepatology 3, 275-284.
Gueimonde, M., Tolkko, S., Korpimaki, T. and Salminen, S., 2004. New real-time quantitative
PCR procedure for quantification of bifidobacteria in human fecal samples. Applied and Environmental Microbiology 70, 4165-4169.
Chapter 7 - References
- 137 -
Gueimonde, M., Garrigues, C., van Sinderen, D., de los Reyes-Gavilan, C.G., and Margolles, A., 2009. Bile-inducible efflux transporter from Bifidobacterium longum NCC2705, conferring bile resistance. Applied Environmental Microbiology 75, 3153-3160.
Guillot, A., Obis, D. and Mistou, M. Y., 2000. Fatty acid membrane composition and activation
of glycine-betaine transport in Lactococcus lactis subjected to osmotic stress. International Journal of Food Microbiology 55, 47-51.
Hambraeus, G., von Wachenfeldt, C., Hederstedt, L., 2003. Genome-wide survey of mRNA half-
lives in Bacillus subtilis identifies extremely stable mRNAs. Molecular Genetics and Genomics 269, 706-714.
Hansen, L. T., Allan-Wojtas, P. M., Jin, Y. L. and Paulson, A. T., 2002. Survival of Ca-alginate
microencapsulated Bifidobacterium spp. in milk and simulated gastrointestinal conditions. Food Microbiology 19, 35-45.
Harmsen, H. J. M., Wildeboer-Veloo, A. C. M., Raangs, G. C., Wagendorp, A. A., Klijn, N.,
Bindels, J. G. and Welling, G. W., 2000. Analysis of intestinal flora development in breast-fed and formula-fed infants by using molecular identification and detection methods. Journal of Pediatric Gastroenterology and Nutrition 30, 61-67.
Hebuterne, X., 2003. Gut changes attributed to ageing: effects on intestinal microflora. Current
Opinion in Clinical Nutrition and Metabolic Care 6, 49-54. Heenan, C. N., Adams, M. C., Hosken, R. W. and Fleet, G. H., 2002. Growth medium for
culturing probiotic bacteria for applications in vegetarian food products. Food Science and Technology 35, 171-176.
Heipieper, H. J., Keweloh, H. and Rehm, H. J., 1991. Influence of phenols on growth and
membrane-permeability of free and immobilized Escherichia coli. Applied and Environmental Microbiology 57, 1213-1217.
Hellyer, T. J., DesJardin, L. E., Hehman, G. L., Cave, M. D., Eisenach, K. D., 1999. Quantitative
analysis of mRNA as a marker for viability of Mycobacterium tuberculosis. Journal of Clinical Microbiology 37, 290-295.
Hewitt, C. J., Nebe-Von-Caron, G., 2001. An industrial application of multiparameter flow
cytometry: Assessment of cell physiological state and its application to the study of microbial fermentations. Cytometry 44, 179-187.
Higgins, M. J. and Novak, J. T., 1997. Characterization of exocellular protein and its role in
bioflocculation. Journal of Environmental Engineering 123, 479-485. Hokamp, K., Roche, F. M., Acab, M., Rousseau, M. E., Kuo, B., Goode, D., Aeschliman, D.,
Bryan, J., Babiuk, L. A., Hancock, R. E., et al., 2004. ArrayPipe: a flexible processing pipeline for microarray data. Nucleic Acids Research 32, W457-459.
Chapter 7 - References
- 138 -
Hooper, L. V., Xu, J., Falk, P. G., Midtvedt, T. and Gordon, J. I., 1999. A molecular sensor that allows a gut commensal to control its nutrient foundation in a competitive ecosystem. Proceedings of the National Academy of Sciences of the United States of America 96, 9833-9838.
Hooper, L. V., Wong, M. H., Thelin, A., Hansson, L., Falk, P. C. and Gordon, J. I., 2001.
Molecular analysis of commensal host-microbial relations hips in the intestine. Science 291, 881-884.
Hooper, L. V., Midtvedt, T. and Gordon, J. I., 2002. How host-microbial interactions shape the
nutrient environment of the mammalian intestine. Annual Review of Nutrition 22, 283-307.
Hopkins, M. J. and MacFarlane, G. T., 2002. Changes in predominant bacterial populations in
human faeces with age and with Clostridium difficile infection. Journal of Medical Microbiology 51, 448-454.
Hoskisson, P. A. and Hobbs, G., 2005. Continuous culture - making a comeback? Microbiology
151, 3153-3159. Hubalek, Z., 2003. Protectants used in the cryopreservation of microorganisms. Cryobiology 46,
205-229. Hundt, S., Zaigler, A., Lange, C., Soppa, J., Klug, G., 2007. Global analysis of mRNA decay in
Halobacterium salinarum NRC-1 at single-gene resolution using DNA Microarrays. Journal of Bacteriology 189, 6936-6944.
Huys, G., Vancanneyt, M., D'Haene, K., Vankerckhoven, V., Goossens, H. and Swings, J., 2006.
Accuracy of species identity of commercial bacterial cultures intended for probiotic or nutritional use. Research in Microbiology 157, 803-810.
Ibrahim, S. A. and Bezkorovainy, A., 1994. Growth promoting factors fro Bifidobacterium
longum. Journal of Food Science 59, 189-191. Ibrahim, S. A., Hassan, O. A., Salameh, M. M. and Shahbazi, A., 2005. Effect of media
composition and incubation temperatures on autoaggregation behavior of bifidobacteria. Milk Science International 60, 127-129.
Ishibashi, N. and Shimamura, S., 1993. Bifidobacteria: Research and Development in Japan.
Food Technology 47, 129-134. Isolauri, E., Arvola, T., Sutas, Y., Moilanen, E. and Salminen, S., 2000. Probiotics in the
management of atopic eczema. Clinical and Experimental Allergy 30, 1604-1610. Jacobsen, C. S., Holben, W. E., 2007. Quantification of mRNA in Salmonella sp. seeded soil and
chicken manure using magnetic capture hybridization RT-PCR. Journal of Microbiological Methods 69, 315-321.
Chapter 7 - References
- 139 -
Jiang, T., Mustapha, A. and Savaiano, D. A., 1996. Improvement of lactose digestion in humans by ingestion of unfermented milk containing Bifidobacterium longum. Journal of Dairy Science 79, 750-757.
Jirku, V., 1999. Whole cell immobilization as a means of enhancing ethanol tolerance. Journal of
Industrial Microbiology and Biotechnology. 22, 147-151. Johnson, J.A.C. and Etzel, M.R., 1993. Inactivation of lactic acid bacteria during spray drying.
In: Barbosa-Cánovas, G.V. and Okos, M.R. (ed.) Food Dehydration. American Institute of Chemical Engineers, New York, USA, pp. 98-107.
Johnson, J. A. C. and Etzel, M. R., 1995. Properties of Lactobacillus helveticus CNRZ-32
attenuated by spray drying, freeze drying, or freezing. Journal of Dairy Science 78, 761-768.
Johnson, D. R., Lee, P. K. H., Holmes, V. F., Alvarez-Cohen, L., 2005. An internal reference
technique for accurately quantifying specific mRNAs by real-time PCR with application to the tceA reductive dehalogenase gene. Applied and Environmental Microbiology 71, 3866-3871.
Josephson, K. L., Gerba, C. P. and Pepper, I. L., 1993. Polymerase chain-reaction detection of
nonviable bacterial pathogens. Applied and Environmental Microbiology 59, 3513-3515. Jouenne, T., Tresse, O. and Junter, G. A., 1994. Agar-entrapped bacteria as an in-vitro model of
biofilms and their susceptibility to antibiotics. FEMS Microbiology Letters 119, 237-242. Junter, G. A. and Jouenne, T., 2004. Immobilized viable microbial cells: from the process to the
proteome... or the cart before the horse. Biotechnology Advances 22, 633-658. Kastner, S., Perreten, V., Bleuler, H., Hugenschmidt, G., Lacroix, C. and Meile, L., 2006.
Antibiotic susceptibility patterns and resistance genes of starter cultures and probiotic bacteria used in food. Systematic and Applied Microbiology 29, 145-155.
Keer, J. T., Birch, L., 2003. Molecular methods for the assessment of bacterial viability. Journal
of Microbiological Methods 53, 175-183. Kheadr, E., Dabour, N., Le Lay, C., Lacroix, C. and Fliss, I., 2007. Antibiotic susceptibility
profile of bifidobacteria as affected by oxgall, acid, and hydrogen peroxide stress. Antimicrobial Agents and Chemotherapy 51, 169-174.
Kim, S. S. and Bhowmik, S. R., 1990. Survival of lactic acid bacteria during spray drying of
plain yogurt. Journal of Food Science 55, 1008-1010. Kim, T. B., Song, S. H., Kang, S. C. and Oh, D. K., 2003. Quantitative comparison of lactose and
glucose utilization in Bifidobacterium longum cultures. Biotechnology Progress 19, 672-675.
Chapter 7 - References
- 140 -
King, V. A. E. and Su, J. T., 1993. Dehydration of Lactobacillus acidophilus. Process Biochemistry 28, 47-52.
Klein, S. M., Elmer, G. W., McFarland, L. V., Surawicz, C. M. and Levy, R. H., 1993. Recovery
and elimination of the biotherapeutic agent, Saccharomyces boulardii, in healthy human volunteers. Pharmaceutical Research 10, 1615-1619.
Klein, G., Pack, A., Bonaparte, C. and Reuter, G., 1998. Taxonomy and physiology of probiotic
lactic acid bacteria. International Journal of Food Microbiology 41, 103-125. Klijn, A., Mercenier, A. and Arigoni, F., 2005. Lessons from the genomes of bifidobacteria.
FEMS Microbiology Reviews 29, 491-509. Knorr, D., 1998. Technology aspects related to microorganisms in functional foods. Trends in
Food Science and Technology 9, 295-306. Koch SR: Effects of fermentation conditions on viability, physiological and technological
characteristics of autolytic dried direct vat set lactic starter cultures. PhD Dissertation ETH Nr.16909. Zurich: ETH Zurich; 2006
Kolter, R., 1999. Growth in studying the cessation of growth. Journal of Bacteriology 181, 697-
699. Koop, H. M., Valentijnbenz, M., Amerongen, A. V. N., Roukema, P. A. and Degraaff, J., 1989.
Aggregation of 27 oral bacteria by human whole saliva – Influence of culture medium, calcium, and bacterial-cell concentration, and interference by autoaggregation. Antonie Van Leeuwenhoek Journal of Microbiology 55, 277-290.
Kos, B., Suskovic, J., Vukovic, S., Simpraga, M., Frece, J. and Matosic, S., 2003. Adhesion and
aggregation ability of probiotic strain Lactobacillus acidophilus M92. Journal of Applied Microbiology 94, 981-987.
Krasaekoopt, W., Bhandari, B. and Deeth, H., 2003. Evaluation of encapsulation techniques of
probiotics for yoghurt. International Dairy Journal 13, 3-13. Krisch, J. and Szajani, B., 1997. Ethanol and acetic acid tolerance in free and immobilized cells
of Saccharomyces cerevisiae and Acetobacter aceti. Biotechnology Letters 19, 525-528. Krishnan, S., Gowda, L. R., Misra, M. C. and Karanth, N. G., 2001. Physiological and
morphological changes in immobilized L-plantarum NCIM 2084 cells during repeated batch fermentation for production of lactic acid. Food Biotechnology. 15, 193-202.
Kullen, M. J., Amann, M. M., O’ Shaughnessy, M. J., O’ Sullivan, D. J., Busta, F. F. and Brady,
L. J., 1996. Differentiation of ingested and endogenous bifidobacteria by DNA fingerprinting demonstrates the survival of an unmodified strain in the gastrointestinal tract of humans. Journal of Nutrition 127, 89-94.
Chapter 7 - References
- 141 -
Kullen, M. J. and Klaenhammer, T. R., 1999. Identification of the pH-inducible, proton-translocating F1F0-ATPase (atpBEFHAGDC) operon of Lactobacillus acidophilus by differential display: gene structure, cloning and characterization. Molecular Microbiology 33, 1152-1161.
Lacroix, C., Grattepanche, F., Doleyres, Y., & Bergmaier, D. (2005). Immobilized cell
technology for the dairy industry. In V. Nevidovic, and R. Willaert (Eds.), Cell immobilization biotechnology. focus in biotechnology. Springer: Heidelberg and Berlin, Germany.
Lacroix, C. and Yidirim, S., 2007. Fermentation technologies for the production of probiotics
with high viability and functionality. Current Opinion in Biotechnology 18, 176-183. Lahtinen, S. J., Ouwehand, A. C., Reinikainen, J. P., Korpela, J. M., Sandholm, J. and Salminen,
S. J., 2006. Intrinsic properties of so-called dormant probiotic bacteria, determined by flow cytometric viability assays. Applied and Environmental Microbiology 72, 5132-5134.
Lahtinen, S. J., Ahokoski, H., Reinikainen, J. P., Gueimonde, M., Nurmi, J., Ouwehand, A. C.,
Salminen, S. J., 2008. Degradation of 16S rRNA and attributes of viability of viable but nonculturable probiotic bacteria. Letters in Applied Microbiology 46, 693-698.
Lamboley, L., Lacroix, C., Champagne, C. P. and Vuillemard, J. C., 1997. Continuous mixed
strain mesophilic lactic starter production in supplemented whey permeate medium using immobilized cell technology. Biotechnology and Bioengineering 56, 502-516.
Lamboley, L., Lacroix, C., Artignan, J. M., Champagne, C. P. and Vuillemard, J. C., 1999. Long-
term mechanical and biological stability of an immobilized cell reactor for continuous mixed-strain mesophilic lactic starter production in whey permeate. Biotechnology Progress 15, 646-654.
Lammers, K. M., Brigidi, P., Vitali, B., Gionchetti, P., Rizzello, F., Caramelli, E., Matteuzzi, D.
and Campieri, M., 2003. Immunomodulatory effects of probiotic bacteria DNA: IL-1 and IL-10 response in human peripheral blood mononuclear cells. FEMS Immunology and Medical Microbiology 38, 165-172.
Leahy, S. C., Higgins, D. G., Fitzgerald, G. F. and van Sinderen, D., 2005. Getting better with
bifidobacteria. Journal of Applied Microbiology 98, 1303-1315. Lee, K. Y. and Heo, T. R., 2000. Survival of Bifidobacterium longum immobilized in calcium
alginate beads in simulated gastric juices and bile salt solution. Applied and Environmental Microbiology 66, 869-873.
Lee, H. Y., Park, J. H., Seik, S. H., Cho, S. A., Baek, M. W., Kim, D. J. and Lee, Y. H., 2004.
Dietary intake of various lactic acid bacteria suppresses type 2 helper T cell production in antigen-primed mice splenocyte. Journal of Microbiology and Biotechnology 14, 167-170.
Chapter 7 - References
- 142 -
Lee, J.H., Karamychev, V.N., Kozyavkin, S.A., Mills, D., Pavlov, A.R., Pavlova, N.V., Polouchine, N.N., Richardson, P.M., Shakhova, V.V., Slesarev, A.I., Weimer, B. and O'Sullivan, D.J., 2008. Comparative genomic analysis of the gut bacterium Bifidobacterium longum reveals loci susceptible to deletion during pure culture growth. BMC Genomics 9, 247.
Leslie, S. B., Israeli, E., Lighthart, B., Crowe, J. H. and Crowe, L. M., 1995. Trehalose and
sucrose protect both membranes and proteins in intact bacteria during drying. Applied and Environmental Microbiology 61, 3592-3597.
Ley, R. E., Peterson, D. A. and Gordon, J. I., 2006. Ecological and evolutionary forces shaping
microbial diversity in the human intestine. Cell 124, 837-848. Leyer, G. J., Li, S. G., Mubasher, M. E., Reifer, C. and Ouwehand, A. C., 2009. Probiotic effects
on cold and influenza-like symptom incidence and duration in children. Pediatrics 124, E172-E179.
Lian, W. C., Hsiao, H. C. and Chou, C. C., 2002. Survival of bifidobacteria after spray-drying.
International Journal of Food Microbiology 74, 79-86. Lilly, D. M. and Stillwel.Rh, 1965. Probiotics – Growth promoting factors produced by
microorganisms. Science 147, 747- 748. Lindner, J. D. D., Canchaya, C., Zhang, Z., Neviani, E., Fitzgerald, G. F., van Sinderen, D. and
Ventura, M., 2006. Exploiting Bifidobacterium genomes: The molecular basis of stress response. International Journal of Food Microbiology120, 13-24.
MacConaill, L.E., Butler, D., O'Connell-Motherway, M., Fitzgerald, G.F. and van Sinderen, D.,
2003. Identification of two-component regulatory systems in Bifidobacterium infantis by functional complementation and degenerate PCR approaches. Applied and Environmental Microbiology 69:4219-4226.
Mangin, I., Bourget, N., Bouhnik, Y., Bisetti, N., Simonet, J. M., Decaris, B., 1994. Identification
of Bifidobacterium strains by rRNA gene restriction patterns. Applied and Environmental Microbiology 60, 1451-1458.
Marteau, P., Pochart, P., Dore, J., Bera-Maillet, C., Bernalier, A. and Corthier, G., 2001.
Comparative study of bacterial groups within the human cecal and fecal microbiota. Applied and Environmental Microbiology 67, 4939-4942.
Masco, L., Huys, G., Gevers, D., Verbrugghen, L. and Swings, J., 2003. Identification of
Bifidobacterium species using rep-PCR fingerprinting. Systematic and Applied Microbiology 26, 557-563.
Masco, L., Huys, G., De Brandt, E., Temmerman, R. and Swings, J., 2005. Culture-dependent
and culture-independent qualitative analysis of probiotic products claimed to contain bifidobacteria. International Journal of Food Microbiology 102, 221-230.
Chapter 7 - References
- 143 -
Masco, L., Vanhoutte, T., Temmerman, R., Swings, J. and Huys, G., 2007. Evaluation of real-time PCR targeting the 16S rRNA and recA genes for the enumeration of bifidobacteria in probiotic products. International Journal of Food Microbiology 113, 351-357.
Masson, F., Lacroix, C. and Paquin, C., 1994. Direct measurement of pH profiles in gel beads
immobilizing Lactobacillus helveticus using a pH sensitive microelectrode. Biotechnology Techniques 8, 551-556.
Matsuda, K., Tsuji, H., Asahara, T., Kado, Y. and Nomoto, K., 2007. Sensitive quantitative
detection of commensal bacteria by rRNA-targeted reverse transcription-PCR. Applied and Environmental Microbiology 73, 32-39.
Matsuki, T., Watanabe, K., Fujimoto, J., Takada, T. and Tanaka, R., 2004. Use of 16S rRNA
gene-targeted group-specific primers for real-time PCR analysis of predominant bacteria in human feces. Applied and Environmental Microbiology 70, 7220-7228.
Mättö, J., Alakomi, H.-L., Vaari, A., Virkajärvi, K. and Saarela, M., 2006. Influence of
processing conditions on Bifidobacterium animalis subsp lactis functionality with a special focus on acid tolerance and factors affecting it. International Dairy Journal 16, 1029-1037.
Maukonen, J., Alakomi, H. L., Nohynek, L., Hallamaa, K., Leppamaki, S., Mättö, J. and Saarela,
M., 2006. Suitability of the fluorescent techniques for the enumeration of probiotic bacteria in commercial non-dairy drinks and in pharmaceutical products. Food Research International 39, 22-32.
Mauriello, G., Aponte, M., Andolfi, R., Moschetti, G. and Villani, F., 1999. Spray-drying of
bacteriocin-producing lactic acid bacteria. Journal of Food Protection 62, 773-777. Maus, J. E. and Ingham, S. C., 2003. Employment of stressful conditions during culture
production to enhance subsequent cold- and acid-tolerance of bifidobacteria. Journal of Applied Microbiology 95, 146-154.
Medina, M., Izquierdo, E., Ennahar, S. and Sanz, Y., 2007. Differential immunomodulatory
properties of Bifidobacterium longum strains: relevance to probiotic selection and clinical applications. Clinical and Experimental Immunology 150, 531-538.
Meng, X. C., Stanton, C., Fitzgerald, G. F., Daly, C. and Ross, R. P., 2008. Anhydrobiotics: The
challenges of drying probiotic cultures. Food Chemistry 106, 1406-1416. Modesto, M., Biavati, B. and Mattarelli, P., 2006. Occurrence of the family Bifidobacteriaceae in
human dental caries and plaque. Caries Research 40, 271-276. Moorman, M. A., Thelemann, C. A., Zhou, S. Y., Pestka, J. J., Linz, J. E. and Ryser, E. T., 2008.
Altered Hydrophobicity and membrane composition in stress-adapted Listeria innocua. Journal of Food Protection 71, 182-185.
Chapter 7 - References
- 144 -
Morgan, C.A., Herman, N., White, P.A. and Vesey, G., 2006. Preservation of micro-organisms by drying; A review. Journal of Microbiological Methods 66, 183-193.
Mozzetti, V. Novel technological approaches to en chance stress tolerance of Bifidobacterium
longum NCC2705 cells using continuous cultures. PhD Dissertation ETH Nr.18419. Zurich: ETH Zurich; 2009.
Nagawa, M., Nakabayashi, A. and Fujino, S., 1988. Preparation of the bifidus milk powder.
Journal of Dairy Science 71, 1777-1782. National Committee for Clinical Laboratory Standards (1991) Antimicrobial susceptibility
testing. 3rd edn. Villanova, PA: National Committee for Clinical Laboratory Standards. Nebe-von-Caron, G., Stephens, P. J., Hewitt, C. J., Powell, J. R. and Badley, R. A., 2000.
Analysis of bacterial function by multi-colour fluorescence flow cytometry and single cell sorting. Journal of Microbiological Methods 42, 97-114.
Nielsen, D. S., Moller, P. L., Rosenfeldt, V., Paerregaard, A., Michaelsen, K. F. and Jakobsen,
M., 2003. Case study of the distribution of mucosa-associated Bifidobacterium species, Lactobacillus species, and other lactic acid bacteria in the human colon. Applied and Environmental Microbiology 69, 7545-7548.
Norton, S., Lacroix, C. and Vuillemard, J. C., 1994. Kinetic –study of continuous whey permeate
fermentation by immobilized Lactobacillus helveticus for lactic-acid production. Enzyme and Microbial Technology 16, 457-466.
O'Mahony, L., McCarthy, J., Kelly, P., Hurley, G., Luo, F. Y., Chen, K. S., O'Sullivan, G. C.,
Kiely, B., Collins, J. K., Shanahan, F., et al., 2005. Lactobacillus and Bifidobacterium in irritable bowel syndrome: Symptom responses and relationship to cytokine profiles. Gastroenterology 128, 541-551.
O'Riordan, K., Andrews, D., Buckle, K. and Conway, P., 2001. Evaluation of microencapsulation
of a Bifidobacterium strain with starch as an approach to prolonging viability during storage. Journal of Applied Microbiology 91, 1059-1066.
Oliver, A. E., Crowe, L. M. and Crowe, J. H., 1997. Methods for dehydration-tolerance:
Depression of the phase transition temperature in dry membranes and carbohydrate vitrification. Seed Science Research 8, 211-221.
Ouellette, V., Chevalier, P. and Lacroix, C., 1994. Continuous fermentation of a supplemented
milk with immobilized Bifidobacterium infantis. Biotechnology Techniques 8, 45-50. Ouwehand, A. C., Salminen, S. J., 1998. The health effects of cultured milk products with viable
and non-viable bacteria. International Dairy Journal 8, 749-758. Ouwehand, A. C., Salminen, S. and Isolauri, E., 2002. Probiotics: an overview of beneficial
effects. Antonie Van Leeuwenhoek 82, 279-289.
Chapter 7 - References
- 145 -
Ouwehand, A. C., Kurvinen, T. and Rissanen, P., 2004. Use of a probiotic Bifidobacterium in a dry food matrix, an in vivo study. International Journal of Food Microbiology 95, 103-106.
Parche, S., Beleut, M., Rezzonico, E., Jacobs, D., Arigoni, F., Titgemeyer, F. and Jankovic, I.,
2006. Lactose-over-glucose preference in Bifidobacterium longum NCC2705: glcP, encoding a glucose transporter, is subject to lactose repression. Journal of Bacteriology 188, 1260-1265.
Parche, S., Amon, J., Jankovic, I., Rezzonico, E., Beleut, M., Barutcu, H., Schendel, I., Eddy,
M.P., Burkovski, A., Arigoni, F., and Titgemeyer, F., 2007. Sugar transport systems of Bifidobacterium longum NCC2705. Journal of Molecular Microbiology and Biotechnology 12, 9-19.
Paul, E., Fages, J., Blanc, P., Goma, G. and Pareilleux, A., 1993. Survival of alginate-entrapped
cells of Azospirillum lipoferum during dehydration and storage in relation to water properties. Applied Microbiology and Biotechnology 40, 34-39.
Perez, P. F., Minnaard, Y., Disalvo, E. A. and De Antoni, G. L., 1998. Surface properties of
bifidobacterial strains of human origin. Applied and Environmental Microbiology 64, 21-26
Petschow, B. W. and Talbott, R. D., 1990. Growth promotion of Bifidobacterium species by
whey and casein fractions from human and bovine-milk. Journal of Clinical Microbiology 28, 287-292.
Picot, A. and Lacroix, C., 2004. Encapsulation of bifidobacteria in whey protein-based
microcapsules and survival in simulated gastrointestinal conditions and in yoghurt. International Dairy Journal 14, 505-515.
Poch, M. and Bezkorovainy, A., 1991. Bovine-milk κ-casein trypsin digest is a growth enhancer
for the genus Bifidobacterium. Journal of Agricultural and Food Chemistry 39, 73-77. Poirier, I., Maréchal, P. A., Richard, S. and Gervais, P., 1999. Saccharomyces cerevisiae viability
is strongly dependant on rehydration kinetics and the temperature of dried cells. Journal of Applied Microbiology 86, 87-92.
Prestrelski, S. J., Tedeschi, N., Arakawa, T. and Carpenter, J. F., 1993. Dehydration-induced
conformational transitions in proteins and their inhibition by stabilizers. Biophysical Journal 65, 661-671.
Prévost, H., Diviès, C. and Rousseau, E., 1985. Continuous yogurt production with Lactobacillus
bulgaricus and Streptococcus thermophilus entrapped in calcium alginate. Biotechnology Letters 7, 247-252.
Chapter 7 - References
- 146 -
Prévost, H. and Diviès, C., 1987. Fresh fermented cheese production with continuous pre-fermented milk by a mixed culture of mesophilic lactic streptococci entrapped in calcium alginate. Biotechnology Letters 9, 789-794.
Prévost, H. and Diviès, C., 1992. Cream fermentation by a mixed culture of lactococci entrapped
in 2-layer calcium alginate gel beads. Biotechnology Letters 14, 583-588. Price, C. E., Reid, S. J., Driessen, A. J. M. and Abratt, V. R., 2006. The Bifidobacterium longum
NCIMB 702259(T) ctr gene codes for a novel cholate transporter. Applied and Environmental Microbiology 72, 923-926.
Rahman, M. M., Kim, W. S., Kumura, H. and Shimazaki, K., 2008. Autoaggregation and surface
hydrophobicity of bifidobacteria. World Journal of Microbiology and Biotechnology 24, 1593-1598.
Rajilic-Stojanovic, M., Smidt, H. and de Vos, W. M., 2007. Diversity of the human
gastrointestinal tract microbiota revisited. Environmental Microbiology 9, 2125-2136. Rautava, S., Salminen, S., and Isolauri, E. 2009. Specific probiotics in reducing the risk of acute
infections in infancy - a randomised, double-blind, placebo-controlled study. British Journal of Nutrition 101, 1722-1726.
Ray, B., Jezeski, J. J. and Busta, F. F., 1971. Effect of rehydration on recovery, repair, and
growth of injured freeze-dried Salmonella anatum. Applied Microbiology 22, 184-&. Redon, E., Loubiere, P., Cocaign-Bousquet, M., 2005. Role of mRNA stability during genome-
wide adaptation of Lactococcus lactis to carbon starvation. Journal of Biological Chemistry 280, 36380-36385.
Reid, G., 2008. Probiotics and prebiotics - Progress and challenges. International Dairy Journal
18, 969-975. Reilly, S. S. and Gilliland, S. E., 1999. Bifidobacterium longum survival during frozen and
refrigerated storage as related to pH during growth. Journal of Food Science 64, 714-718. Resta-Lenert, S., Barrett, K. E., 2003. Live probiotics protect intestinal epithelial cells from the
effects of infection with enteroinvasive Escherichia coli (EIEC). Gut 52, 988-997. Rezzonico, E., Lariani, S., Barretto, C., Cuanoud, G., Giliberti, G., Delley, M., Arigoni, F. and
Pessi, G., 2007. Global transcriptome analysis of the heat shock response of Bifidobacterium longum. FEMS Microbiology Letters 271, 136-145.
Riveros, B., Ferrer, J. and Borquez, R., 2009. Spray Drying of a Vaginal Probiotic Strain of
Lactobacillus acidophilus. Drying Technology 27, 123-132.
Chapter 7 - References
- 147 -
Ross, R. P., Desmond, C., Fitzgerald, G. F. and Stanton, C., 2005. Overcoming the technological hurdles in the development of probiotic foods. Journal of Applied Microbiology 98, 1410-1417.
Rossetti, L. and Giraffa, G., 2005. Rapid identification of dairy lactic acid bacteria by M13-
generated, RAPD-PCR fingerprint databases. Journal of Microbiological Methods 63, 135-144.
Ruas-Madiedo, P., Hernandez-Barranco, A., Margolles, A. and de los Reyes-Gavilan, C. G.,
2005. A bile salt-resistant derivative of Bifidobacterium animalis has an altered fermentation pattern when grown on glucose and maltose. Applied and Environmental Microbiology 71, 6564-6570.
Ruiz, L., Sanchez, B., Ruas-Madiedo, P., de los Reyes-Gavilan, C. G. and Margolles, A., 2007.
Cell envelope changes in Bifidobacterium animalis ssp. lactis as a response to bile. Fems Microbiology Letters 274, 316-322.
Ruiz, L., Sanchez, B., de los Reyes-Gavilan, C. G., Gueimonde, M. and Margolles, A., 2009.
Coculture of Bifidobacterium longum and Bifidobacterium breve alters their protein expression profiles and enzymatic activities. International Journal of Food Microbiology 133, 148-153
Saarela, M., Lahteenmaki, L., Crittenden, R., Salminen, S. and Mattila-Sandholm, T., 2002. Gut
bacteria and health foods--the European perspective. Int J Food Microbiol 78, 99-117. Saarela, M., Rantala, M., Hallamaa, K., Nohynek, L., Virkajärvi, I. and Mättö, J., 2004.
Stationary-phase acid and heat treatments for improvement of the viability of probiotic lactobacilli and bifidobacteria. Journal of Applied Microbiology 96, 1205-1214.
Saarela, M., Virkajärvi, I., Alakomi, H. L., Sigvart-Mattila, P. and Mättö, J., 2006. Stability and
functionality of freeze-dried probiotic Bifidobacterium cells during storage in juice and milk. International Dairy Journal 16, 1477-1482.
Salminen, S., Ouwehand, A., Benno, Y. and Lee, Y. K., 1999. Probiotics: how should they be
defined? Trends in Food Science and Technology 10, 107-110. Salminen, M. K., Tynkkynen, S., Rautelin, H., Saxelin, M., Vaara, M., Ruutu, P., Sarna, S.,
Valtonen, V. and Jarvinen, A., 2002. Lactobacillus bacteremia during a rapid increase in Probiotic use of Lactobacillus rhamnosus GG in Finland. Clinical Infectious Diseases 35, 1155-1160.
Sanchez, B., Champomier-Verges, M. C., Anglade, P., Baraige, F., Reyes-Gavilan, C. G. D.,
Margolles, A. and Zagorec, M., 2005. Proteomic analysis of global changes in protein expression during bile salt exposure of Bifidobacterium longum NCIMB 8809. Journal of Bacteriology 187, 5799-5808.
Chapter 7 - References
- 148 -
Sanchez, B., Champomier-Verges, M. C., Stuer-Lauridsen, B., Ruas-Madiedo, P., Anglade, P., Baraige, F., de los Reyes-Gavilan, C. G., Johansen, E., Zagorec, M. and Margolles, A., 2007. Adaptation and response of Bifidobacterium animalis subsp. lactis to bile: a proteomic and physiological approach. Applied and Environmental Microbiology 73, 6757-6767.
Sanchez, B., Ruiz, L., de los Reyes-Gavilan, C. G. and Margolles, A., 2008. Proteomics of stress
response in Bifidobacterium. Frontiers in Bioscience 13, 6905-6919. Santivarangkna, C., Kulozik, U. and Foerst, P., 2007. Alternative drying processes for the
industrial preservation of lactic acid starter cultures. Biotechnology Progress 23, 302-315. Santivarangkna, C., Kulozik, U. and Foerst, P., 2008. Inactivation mechanisms of lactic acid
starter cultures preserved by drying processes. Journal of Applied Microbiology 105, 1-13.
Savijoki, K., Suokko, A., Palva, A., Valmu, L., Kalkkinen, N. and Varmanen, P., 2005. Effect of
heat-shock and bile salts on protein synthesis of Bifidobacterium longum revealed by [S-35]methionine labelling and two-dimensional gel electrophoresis. FEMS Microbiology Letters 248, 207-215.
Saxelin, M., 2008. Probiotic formulations and applications, the current probiotics market, and
changes in the marketplace: A European perspective. Clinical Infectious Diseases 46, S76-S79.
Schachtsiek, M., Hammes, W. P. and Hertel, C., 2004. Characterization of Lactobacillus
coryniformis DSM 20001(T) surface protein Cpf mediating coaggregation with and aggregation among pathogens. Applied and Environmental Microbiology 70, 7078-7085.
Schell, M. A., Karmirantzou, M., Snel, B., Vilanova, D., Berger, B., Pessi, G., Zwahlen, M. C.,
Desiere, F., Bork, P., Delley, M., Pridmore, R.D. and Arigoni, F., 2002. The genome sequence of Bifidobacterium longum reflects its adaptation to the human gastrointestinal tract. Proceedings of the National Academy of Sciences of the United States of America 99, 14422-14427.
Schembri, M. A., Kjaergaard, K. and Klemm, P., 2003. Global gene expression in Escherichia
coli biofilms. Molecular Microbiology 48, 253-267. Schepers, A. W., Thibault, J. and Lacroix, C., 2006. Continuous lactic acid production in whey
permeate/yeast extract medium with immobilized Lactobacillus helveticus in a two-stage process: Model and experiments. Enzyme and Microbial Technology 38, 324-337.
Schmidt, G. and Zink, R., 2000. Basic features of the stress response in three species of
bifidobacteria: B. longum, B. adolescentis, and B. breve. International Journal of Food Microbiology 55, 41-45.
Chapter 7 - References
- 149 -
Sela, D.A., Chapman, J., Adeuya, A., Kim, J.H., Chen, F., Whitehead, T.R., Lapidus, A., Rokhsar, D.S., Lebrilla, C.B., German, J.B., Price, N.P., Richardson, P.M., and Mills, D.A., 2008. The genome sequence of Bifidobacterium longum subsp. infantis reveals adaptations for milk utilization within the infant microbiome. Proceedings of the National Academy of Sciences 105, 18964-18969.
Shah, N. P. and Ravula, R. R., 2000. Microencapsulation of probiotic bacteria and their survival
in frozen fermented dairy desserts. Australian Journal of Dairy Technology 55, 139-144. Shahidi, F. and Han, X. Q., 1993. Encapsulation of food ingredients. Critical Reviews in Food
Science and Nutrition 33, 501-547. Shanahan, F., 2002. The host-microbe interface within the gut. Best Practice and Research in
Clinical Gastroenterology 16, 915-931. Sheehan, V. M., Sleator, R. D., Fitzgerald, G. F. and Hill, C., 2006. Heterologous expression of
BetL, a betaine uptake system, enhances the stress tolerance of Lactobacillus salivarius UCC118. Applied and Environmental Microbiology 72, 2170-2177.
Sheridan, G. E. C., Masters, C. I., Shallcross, J. A. and Mackey, B. M., 1998. Detection of
mRNA by reverse transcription PCR as an indicator of viability in Escherichia coli cells. Applied and Environmental Microbiology 64, 1313-1318.
Shimakawa, Y., Matsubara, S., Yuki, N., Ikeda, M. and Ishikawa, F., 2003. Evaluation of
Bifidobacterium breve strain Yakult-fermented soymilk as a probiotic food. International Journal of Food Microbiology 81, 131-136.
Silva, A. M., Bambirra, E. A., Oliveira, A. L., Souza, P. P., Gomes, D. A., Vieira, E. C. and
Nicoli, J. R., 1999. Protective effect of bifidus milk on the experimental infection with Salmonella enteritidis subsp. typhimurium in conventional and gnotobiotic mice. Journal of Applied Microbiology 86, 331-336.
Silva, J., Carvalho, A. S., Teixeira, P. and Gibbs, P. A., 2002. Bacteriocin production by spray-
dried lactic acid bacteria. Letters in Applied Microbiology 34, 77-81. Silva, J., Carvalho, A. S., Pereira, H., Teixeira, P. and Gibbs, P. A., 2004. Induction of stress
tolerance in Lactobacillus delbrueckii ssp bulgaricus by the addition of sucrose to the growth medium. Journal of Dairy Research 71, 121-125.
Simpson, P. J., Stanton, C., Fitzgerald, G. F. and Ross, R. P., 2005. Intrinsic tolerance of
Bifidobacterium species to heat and oxygen and survival following spray drying and storage. Journal of Applied Microbiology 99, 493-501.
Snydman, D. R., 2008. The safety of probiotics. Clinical Infectious Disease 46, 104–111.
Chapter 7 - References
- 150 -
Sodini, I., Boquien, C. Y., Corrieu, G. and Lacroix, C., 1997. Use of an immobilized cell bioreactor for the continuous inoculation of milk in fresh cheese manufacturing. Journal of Industrial Microbiology and Biotechnology 18, 56-61.
Soejima, T., Iida, K. I., Qin, T., Taniai, H., Seki, M., Yoshida, S. I., 2008. Method to detect only
live bacteria during PCR amplification. Journal of Clinical Microbiology 46, 2305-2313. Stanley, N. R. and Lazazzera, B. A., 2004. Environmental signals and regulatory pathways that
influence biofilm formation. Molecular Microbiology 52, 917-924. Stevens, M. J. A., Wiersma, A., de Vos, W. M., Kuipers, O. P., Smid, E. J., Molenaar, D.,
Kleerebezem, M., 2008. Improvement of Lactobacillus plantarum aerobic growth as directed by comprehensive transcriptome analysis. Applied and Environmental Microbiology 74, 4776-4778.
Sturme, M. H. J., Kleerebezem, M., Nakayama, J., Akkermans, A. D., Vaughan, E. E. and de
Vos, W. M., 2002. Cell to cell communication by autoinducing peptides in gram-positive bacteria. Antonie Van Leeuwenhoek 81, 233-243.
Su, P., Henriksson, A., Tandianus, J. E., Park, J. H., Foong, F. and Dunn, N. W., 2005. Detection
and quantification of Bifidobacterium lactis LAFTI (R) B94 in human faecal samples from a consumption trial. FEMS Microbiology Letters 244, 99-103.
Sugimoto, S., Abdullah Al, M. and Sonomoto, K., 2008. Molecular Chaperones in Lactic Acid
Bacteria: Physiological Consequences and Biochemical Properties. Journal of Bioscience and Bioengineering 106, 324-336.
Sultana, K., Godward, G., Reynolds, N., Arumugaswamy, R., Peiris, P. and Kailasapathy, K.,
2000. Encapsulation of probiotic bacteria with alginate-starch and evaluation of survival in simulated gastrointestinal conditions and in yoghurt. International Journal of Food Microbiology 62, 47-55.
Sun, W. R. and Griffiths, M. W., 2000. Survival of bifidobacteria in yogurt and simulated gastric
juice following immobilization in gellan-xanthan beads. International Journal of Food Microbiology 61, 17-25.
Sunny-Roberts, E. O. and 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. International Dairy Journal 19, 209-214.
Takayama, K., Kjelleberg, S. A., 2000. The role of RNA stability during bacterial stress
responses and starvation. Environmental Microbiology 2, 355-365. Talwalkar, A. and Kailasapathy, K., 2003. Effect of microencapsulation on oxygen toxicity in
probiotic bacteria. Australian Journal of Dairy Technology 58, 36-39.
Chapter 7 - References
- 151 -
Tamime, A. Y., Marshall, V. M. E. and Robinson, R. K., 1995. Microbiological and technological aspects of milks fermented by bifidobacteria. Journal of Dairy Research 62, 151-187.
Tannock, G. W., 1999. Analysis of the intestinal microflora: a renaissance. Antonie Van
Leeuwenhoek 76, 265-278. Teixeira, P. C., Castro, M. H. and Kirby, R. M., 1995a. Death kinetics of Lactobacillus
bulgaricus in spray drying process. Journal of Food Protection 58, 934-936. Teixeira, P. C., Castro, M. H., Malcata, F. X. and Kirby, R. M., 1995b. Survival of Lactobacillus
delbrueckii ssp. bulgaricus following spray drying. Journal of Dairy Science 78, 1025-1031.
Teixeira, P., Castro, H. and Kirby, R., 1996. Evidence of membrane lipid oxidation of spray-
dried Lactobacillus bulgaricus during storage. Letters in Applied Microbiology 22, 34-38. To, B. C. S. and Etzel, M. R., 1997. Survival of Brevibacterium linens (ATCC 9174) after spray
drying, freeze drying, or freezing. Journal of Food Science 62, 167-170. TolkerNielsen, T., Larsen, M. H., Kyed, H., Molin, S., 1997. Effects of stress treatments on the
detection of Salmonella typhimurium by in situ hybridization. International Journal of Food Microbiology 35, 251-258.
Trauth, E., Lemaitre, J. P., Rojas, C., Diviès, C. and Cachon, R. M., 2001. Resistance of
immobilized lactic acid bacteria to the inhibitory effect of quaternary ammonium sanitizers. Food Science and Technology 34, 239-243.
US Patent 2003/0109025. Particles containing coated living microorganisms, and method for
producing same. Van der Meulen, R., Adriany, T., Verbrugghe, K. and De Vuyst, L., 2006. Kinetic analysis of
bifidobacterial metabolism reveals a minor role for succinic acid in the regeneration of NAD(+) through its growth-associated production. Applied and Environmental Microbiology 72, 5204-5210.
Vaughan, E. E., de Vries, M. C., Zoetendal, E. G., Ben-Amor, K., Akkermans, A. D. L. and de
Vos, W. M., 2002. The intestinal LABs. Antonie Van Leeuwenhoek 82, 341-352. Vaughan, E. E., Heilig, H. G., Ben-Amor, K. and de Vos, W. M., 2005. Diversity, vitality and
activities of intestinal lactic acid bacteria and bifidobacteria assessed by molecular approaches. FEMS Microbiology Reviews 29, 477-490.
Vega, C. and Roos, Y. H., 2006. Invited review: Spray-dried dairy and dairy-like - emulsions
compositional considerations. Journal of Dairy Science 89, 383-401.
Chapter 7 - References
- 152 -
Ventling, B. L. and Mistry, V. V., 1993. Growth characteristics of bifidobacteria in ultrafiltered milk. Journal of Dairy Sciences 76, 962-971.
Ventura, M., Canchaya, C., Del Casale, A., Dellaglio, F., Neviani, E., Fitzgerald, G. F. and van
Sinderen, D., 2006. Analysis of bifidobacterial evolution using a multilocus approach. International Journal of Systematic and Evolutionary Microbiology 56, 2783-2792.
Ventura, M., O'Connell-Motherway, M., Leahy, S., Moreno-Munoz, J. A., Fitzgerald, G. F. and
van Sinderen, D., 2007. From bacterial genome to functionality; case bifidobacteria. International Journal of Food Microbiology 120, 2-12.
Vincent, D., Roy, D., Mondou, F. and Dery, C., 1998. Characterization of bifidobacteria by
random DNA amplification. International Journal of Food Microbiology 43, 185-193. Vitali, B., Candela, M., Matteuzzi, D. and Brigidi, P., 2003. Quantitative detection of probiotic
Bifidobacterium strains in bacterial mixtures by using real-time PCR. Systematic and Applied Microbiology 26, 269-276.
Wallner, G., Erhart, R., Amann, R., 1995. Flow cytometric analysis of activated-sludge with
ribosomal-RNA-targeted probes. Applied and Environmental Microbiology 61, 1859-1866.
Wang, Y. C., Yu, R. C. and Chou, C. C., 2004. Viability of lactic acid bacteria and bifidobacteria
in fermented soymilk after drying, subsequent rehydration and storage. International Journal of Food Microbiology 93, 209-217.
Weizman, Z., Asli, G. and Alsheikh, A., 2005. Effect of a probiotic infant formula on infections
in child care centers: comparison of two probiotic agents. Pediatrics 115, 5-9. Whiteley, M., Bangera, M. G., Bumgarner, R. E., Parsek, M. R., Teitzel, G. M., Lory, S. and
Greenberg, E. P., 2001. Gene expression in Pseudomonas aeruginosa biofilms. Nature 413, 860-864.
Whitman, W. B., Coleman, D. C. and Wiebe, W. J., 1998. Prokaryotes: The unseen majority.
Proceedings of the National Academy of Sciences of the United States of America 95, 6578-6583.
Wilson, D. N. and Nierhaus, K. H., 2005. Ribosomal proteins in the spotlight. Critical Reviews in
Biochemistry and Molecular Biology 40, 243-267. Wolber, P. K., Collins, P. J., Lucas, A. B., De Witte, A. and Shannon, K. W., 2006. The Agilent
in situ-synthesized microarray platform. Methods in Enzymology 410, 28-57. Wollowski, I., Rechkemmer, G. and Pool-Zobel, B. L., 2001. Protective role of probiotics and
prebiotics in colon cancer. American Journal of Clinical Nutrition, 73, 451–455.
Chapter 7 - References
- 153 -
Woodmansey, E. J., McMurdo, M. E. T., Macfarlane, G. T. and Macfarlane, S., 2004. Comparison of compositions and metabolic activities of fecal microbiotas in young adults and in antibiotic-treated and non-antibiotic-treated elderly subjects. Applied and Environmental Microbiology 70, 6113-6122.
Wouters, J. A., Rombouts, F. M., de Vos, W. M., Kuipers, O. P. and Abee, T., 1999. Cold shock
proteins and low-temperature response of Streptococcus thermophilus CNRZ302. Applied and Environmental Microbiology 65, 4436-4442.
Zamora, L. M., Carretero, C. and Pares, D., 2006. Comparative survival rates of lactic acid
bacteria isolated from blood, following spray-drying and freeze-drying. Food Science and Technology International 12, 77-84.
Zavaglia, A. G., Kociubinski, G., Perez, P., Disalvo, E. and De Antoni, G., 2002. Effect of bile
on the lipid composition and surface properties of bifidobacteria. Journal of Applied Microbiology 93, 794-799.
Zayed, G. and Roos, Y. H., 2004. Influence of trehalose and moisture content on survival of
Lactobacillus salivarius subjected to freeze-drying and storage. Process Biochemistry. 39, 1081-1086.
Zoetendal, E. G., Collier, C. T., Koike, S., Mackie, R. I. and Gaskins, H. R., 2004. Molecular
ecological analysis of the gastrointestinal microbiota: a review. Journal of Nutrition 134, 465-472.
Zoetendal, E. G., Vaughan, E. E. and de Vos, W. M., 2006. A microbial world within us.
Molecular Microbiology 59, 1639-1650.
Curriculum vitae
- 154 -
Acknowledgements My thanks go to Christophe Lacroix, Leo Meile, Enea Rezzonico and Bernard Berger for the
supervision during this project. Thank you also Paul Ross who accepted to co-examine this thesis
and provide his comments. A special thank to Franck Grattepanche for patient supervision and
proof reading my manuscripts and also for sharing taste of (most) of my music library. Many
thanks to my diploma students Rafael Benz, who spend many hours beside continuous culture
and Claudia Baggenstos for precise work and endurance during spray drying and storage
experiments. My thanks go also to Valeria Mozzetti, for good cooperation during 3 ½ years, I
wish you and your young family all the best for the future.
Many thanks to past and present colleagues of the Laboratory of Food Biotechnology who
contributed with discussions, knowledge or only by mere presence to this work, especially to
Mark Stevens, Alfonso Die and of course to Frau Venakis: Danke Kiki, für die Geduld mit mir!
A heartfelt thanks goes to my family and friends, especially to my mother Sybille who supported
me always in every situation. A very special thanks to my father, Alois, who left us so
unexpectedly 3 days after I started my PhD project and with whom I would have liked to share
good and bad times during the last four years.
At the very end I want to express my deepest thanks to my wife, Maria, for your love and
confidence and patience to share your life with a “rab lat”.
This study was carried out thanks to the financial support of the Commission of Technology and
Innovation of Switzerland (CTI-Project Nr. 75272 LSPP-LS), Nestlé (Switzerland) and the
Walter Hochstrasser Foundation.