Ecophysiology Intestinal

Microbial Eco-Physiology of the Human
Intestinal Tract:
A Flow Cytometric Approach

Promotor: Prof. dr. W. M. de Vos
Hoogleraar Microbiologie
Wageningen Universiteit
Co-promotoren: Dr. T. Abee
Universitair Hoofddocent bij de leerstoelgroep Levensmiddelen-
microbiologie
Dr. E. E. Vaughan
Lead Scientist
Unilever Research and Development, Vlaardingen
Promotiecommissie: Prof. dr. ir. M. H. Zwietering
Prof.dr. J. Bindels
Dr. G. W. Welling
Universiteit Groningen
Dr. J. Dor
Institut National de Recherche Agronomique
Centre de recherche de Jouy-en-Josas, France
Dit onderzoek is uitgevoerd binnen de onderzoekschool VLAG
Microbial Eco-Physiology of the Human
Intestinal Tract:
A Flow Cytometric Approach
Kaouther Ben Amor
Proefschrift
Ter verkrijging van de graad van doctor
op gezag van de rector magnificus
van Wageningen Universiteit,
Prof. dr. ir. L. Speelman,
in het openbaar te verdedigen
op vrijdag 10 september 2004
des namiddags te vier uur in de Aula
This work was carried out at Wageningen University, Department of Agro-Technology and
Food Sciences, Laboratory of Food Microbiology and Laboratory of Microbiology.
K. Ben Amor. Microbial Eco-physiology of the human intestinal tract: A flow cytometric
approach. PhD thesis. Wageningen University, The Netherlands, 2004. With summaries in
Dutch and English.
Key words: gastrointestinal tract, fecal microbiota, probiotics, 16S rRNA, Fluorescent In-Situ
Hybridization (FISH), flow cytometry, cell sorting, fluorescent probes, viability, microbial
physiology, Bifidobacteria, Inflammatory Bowel Disease (IBD).
ISBN: 90-8504-042-6

TABLE OF CONTENTS

Chapter 1:
General introduction .............................................................................................................................. 1
Chapter 2:
Application of flow cytometry in microbiology............................................................................... 21
Chapter 3:
Quantification of uncultured Ruminococcus obeum-like bacteria in human fecal
samples with fluorescent in situ hybridization and flow cytometry
using 16S ribosomal RNA targeted probes....................................................................................... 49
Chapter 4:
Mucosa-associated bacteria in the human gastrointestinal tract
are uniformly distributes along the colon and differ from
the community recovered from feces ................................................................................................ 67
Chapter 5:
Populations dynamics and diversity of fecal microbiota of patients
with ulcerative colitis participating in a probiotic trial .................................................................... 83
Chapter 6:
Multiparametric flow cytometry and cell sorting for the assessment of viable,
injured, and dead Bifidobacterium cells during bile salt stress ......................................................... 105
Chapter 7:
Genetic diversity of live, injured and dead fecal bacteria assessed by fluorescence
activated cell sorting and 16S RRNA gene analysis ....................................................................... 123
Chapter 8:
Summary and concluding remarks.................................................................................................... 149
Samenvatting......................................................................................................................................... 155
Acknowledgments ............................................................................................................................... 161
About the author.................................................................................................................................. 163
List of publications .............................................................................................................................. 164
Chapter 1
GENERAL INTRODUCTION:
Ecology of the human intestinal microbiota
A modified version of this chapter has been accepted for publication as a chapter in:
Gastrointestinal Microbiology
(Edited by Arthur Ouwehand and Elaine E. Vaughan and published by Marcel Dekker)
Molecular tools to analyze the composition of intestinal microbiota
Kaouther Ben Amor and Elaine E. Vaughan
1
The human gastro-intestinal (GI) tract is the home of a huge microbial assemblage, the
vast extent of which is only being revealed. The number of microorganisms (microbiota) greatly
exceeds human cells, resulting in one of the most diverse and dynamic microbial ecosystems,
where relationships amongst the microbes and between those and the host have a profound
influence on all concerned (33). This microbiota play essential roles in a wide variety of
metabolic and immunological processe and therefore significantly contribute to the well being
of the host (17). During the last decade, food-grade specific isolates, termed probiotics have been
extensively used in an attempt to modulate the composition and/or activity of the intestinal
microbiota so as to provide an advantage to the host. Despite certain haziness about the use of
probiotics as functional foods or as bio-therapeutic agents, today there is persuasive evidence
supporting their efficacy in the prevention or treatment of a number of intestinal disorders in
humans (54, 57). Nevertheless, in order to rationally use probiotics as functional foods or as
therapeutic agents, in-depth knowledge of the structure, dynamics and function of the bacterial
populations of the GI-tract microbiota is crucial.
Although the human intestinal microbiota have been extensively investigated by
culture-based methods more than any other natural ecosystem (19, 31, 46), our knowledge
about the culturable fraction of this community is limited (3, 75). The advent of molecular
techniques based on the 16S ribosomal RNA (rRNA) gene analysis is now allowing a more
complete assessment of this complex microbial ecosystem by unraveling the extent of the
diversity, abundance and population dynamics of this community. These techniques have
extended our view of those microorganisms that have proven difficult to culture and which
play an important role in the gut physiology. This huge intestinal microbial reservoir,
estimated to contain more than 1,000 bacterial species (82) and as much as 10
13
cells,
exhibits a highly diverse set of metabolic activities (17, 32). Hence, it is essential to identify
these microbes based upon their eco-physiological traits i.e. those that are functionally
active versus those that are effectively redundant and play little or no role at a particular
time or at a given site of the intestinal tract. It is therefore a major challenge to develop
approaches that monitor the activity of these microorganisms at the single cell level in their
natural habitat. This chapter will focus on these new insights, highlight newly developed
molecular methods to study the eco-physiology of the GI-tract, and culminates in an
overview of this thesis.
1.1 The uncultured GI-tract microbiota is identified by 16S rRNA
gene sequencing and phylogenetic analysis
The comparative analysis of environmentally retrieved nucleic acid sequences, most
notably of rRNA molecules and the genes encoding them, has become a standard for
cultivation-independent assessment of bacterial diversity in environmental samples (3).
Ribosomal RNA gene fragments are today routinely retrieved without prior cultivation of the
microbes by constructing 16S ribosomal DNA (rDNA) libraries. Large databases of especially
2
Chapter 1
16S rRNA gene sequence information for described as well as uncultured microorganisms are
available, and thus provide a high-resolution platform for the assignment of those new
sequences obtained in 16S rDNA libraries (41). The procedure is based upon PCR-mediated
amplification of 16S rRNA genes or gene fragments, using rRNA or rDNA isolated from the
environmental sample, followed by segregation of individual gene copies by cloning into
Escherichia coli. In this way a library of community 16S rRNA genes is generated, the
composition of which can be estimated by screening clones, full or partial sequence analysis
and comparing them with adequate reference sequences to infer their phylogenetic affiliation.
Sequencing of 16S rDNA clone libraries generated from various sites of the GI-tract
including terminal ileum, colon, mucosa and feces, obtained from healthy and diseased people
have confirmed that a relevant fraction of gut bacteria were derived from new, as yet
undescribed bacterial phylotypes (30, 53, 70, 80, 83, 85). Such studies revealed that the vast
majority of rDNA amplicons generated directly from fecal or biopsy samples were assigned to
three major phylogenetic lineages, namely the Clostridium coccoides, Clostridium leptum and
Bacteroides groups. Comprehensive phylogenetic analysis demonstrated that more than half of
the observed diversity was attributable to unknown dominant microorganisms within the
human gut. Additionally, Zoetendal et al. (85) demonstrated that the majority of predominant
bacterial species from an adult fecal sample did not correspond to known species, but that the
prominent bacteria were assigned to different Clostridium clusters namely, Ruminococcus obeum,
Eubacterium hallii and Fusobacterium prausnitzii. On the other hand, phylogenetic analysis of
16S rDNA clone libraries generated from mucosa-associated microbiota of patients with
inflammatory bowel disease (IBD), revealed a reduction in diversity due to a loss of normal
anaerobic bacteria especially those belonging to the Bacteroides, Eubacterium and Lactobacillus
species. Most of the sequenced clones retrieved from the biopsy samples (70%), obtained from
IBD patients, were assigned to known intestinal bacteria, but a significant number of the
cloned sequences were affiliated to normal residents of the oral mucosa such as Streptococcus
species (53). The authors suggested that alteration of the bacterial microbiota in mucosal
inflammation reflects a metabolic imbalance of the complex microbial ecosystem with
severe consequences for the mucosal barrier rather than disrupted defense to single
microorganisms (53).
Even though sequencing of cloned 16S rDNA amplicons provides relevant information
about the identity of uncultured bacteria, the data are not quantitative. Moreover, PCR and
cloning steps are not without biases (76), a recent comparative analysis of clone libraries from
a fecal sample pointed out that the number of PCR cycles may affect the diversity of the
amplified 16S rDNAs and thus should be minimized (8). More rapid culture-independent
options to the cloning procedures include examination of complex microbial populations using
a variety of fingerprinting methods.
3
General introduction

1.2 Fingerprinting techniques reveal the stability, uniqueness and
complexity of the GI-tract microbiota
The most commonly applied fingerprinting methods used to study the GI-tract
microbiota are denaturing temperature (DGGE) and temperature gradient gel electrophoresis
(TGGE) of PCR-amplified genes coding for 16S rRNA (75, 88). Other techniques such as
terminal restriction fragment length polymorphism (T-RFLP) and single strand conformation
polymorphism (SSCP) analysis have been applied but less frequently (50, 53). The common
principle of these methods is based on the separation of PCR-amplified segments of 16S rRNA
genes of the same length but with different sequence to visualize the diversity within the PCR
amplicons by a banding pattern. With DGGE/TGGE, separation is based on the decreased
electrophoretic mobility of partially melted double-stranded DNA molecules in
polyacrylamide gels containing a linear gradient of DNA denaturants (a mixture or formamide
and urea) or a linear temperature gradient, respectively. As a result mixed amplified PCR
products will form a banding pattern after staining that reflects the different melting behaviors
of the various sequences (49, 62). Subsequent identification of specific bacterial groups or
species present in the sample can be achieved either by cloning and sequencing of the excised
bands or by hybridization of the profile using phylogenetic probes (48). Furthermore,
complementation of the fingerprinting results with statistical analysis provides additional
information of the observed diversity by highlighting some putative correlations between
different sets of variables (20).
Since its application to study the intestinal microbiota, PCR-DGGE/-TGGE
fingerprinting has advanced our knowledge of the intestinal microbiota by unraveling the
complexity of this ecosystem and providing insight in the establishment and succession of the
bacterial community within the host (18, 85). In healthy adults, the predominant fecal
microbiota was shown to be host-specific, relatively stable in time and not significantly altered
following consumption of certain probiotic strains (72, 74, 84, 85). Furthermore, it revealed
that the predominant bacterial species associated with the colonic mucosa are uniformly
distributed along the colon, but significantly different from the predominant fecal community
(89). Under certain environmental circumstances and/or in genetically susceptible individuals,
there is persuasive evidence that the GI-tract microbiota may play a role in the pathogenesis
and aetiology of a number of inflammatory diseases such as ulcerative colitis (UC), and Crohns
disease (CD) (10, 66). Using DGGE, TTGE and SSCP fingerprinting analyses, it was
demonstrated that fecal and mucosa-associated microbiota of patients with UC and CD is
altered, less complex, and also unstable over time as compared to matched healthy people (53,
64 )(Chapter 4).
Although DGGE or TGGE were initially developed for total ecosystem communities,
the sensitivity of the method for detecting specific groups that are present in lower numbers in
the GI-tract such as bifidobacteria and especially lactobacilli has been considerably enhanced
4
Chapter 1
by using group- or genus-specific primers (29, 63, 72, 79). Consequently, it was possible to
monitor the effect of the administration of prebiotics and/or probiotics on the composition of
indigenous bifidobacterial species, and to track the probiotic strain itself (63). In the latter case,
DGGE profiles showed that the simultaneous administration of the prebiotic and probiotic
(symbiotic approach) did not improve the colonization of the probiotic strain in the gut of the
tested individuals. In another study, the DGGE profiles generated from fecal samples of healthy
individuals fed a probiotic strain Lactobacillus paracasei F19, allowed the tracing of the probiotic
and supported its presence as autochthonous within the intestinal community of a number of
individuals (29).
While the application of 16S rDNA-based fingerprinting are particularly well suited for
examining time series and population dynamics, a more quantitative approach is useful to
complement our knowledge about the composition and structure of this complex intestinal
ecosystem.
1.3 16S rRNA-targeted probes quantify the GI-tract microbiota
Hybridization with rRNA-targeted oligonucleotide probes has become the method of
choice for the direct cultivation-independent identification of individual bacterial cells in
natural samples. During the last decade, this technique has extended our view of bacterial
assemblage and population dynamics of complex microbial communities (3, 38, 47). The
most commonly used biomarker for hybridization techniques, either dot blot or fluorescent
in situ hybridization (FISH) is the 16S rRNA molecule because of its genetic stability, domain
structure with conserved and variable regions, and high copy number. Highly conserved
stretches may thus be used to design domain-specific probes such as EUB338/EUBII/EUBIII
which collectively target most of the bacteria, whereas specific probes for each taxonomic
level, between bacterial and archaeal, down to genus-specific and species-specific, can be
designed according to the highly variable regions of the 16S rRNA (3, 4, 43). The increasing
availability of 16S rRNA sequences contributed significantly to the development of the
hybridization methods and their application in different microbial ecosystems (41).
Unquestionably, the success of the implementation of 16S rRNA hybridization strategies
depends on different factors, among them a rational design and validation of newly designed
rRNA-targeted probes.
Probe design and validation
When designing new probes, one must consider specificity, sensitivity and accessibility
to the target sequence. Nucleic acid probes can be designed to specifically target taxonomic
groups at different levels of specificity (from species to domain) by virtue of variable
evolutionary conservation of the ribosomal rRNA molecules. Appropriate software such as the
ARB software package (40) and availability of large databases (http://rdp.cme.msu.edu/html/)
or the online resource for oligonucleotide probes Probe Base (39) are useful tools for a rapid
5

probe design and in silico specificity profiling. Additional experimental evaluation of the probes
with target and non-target microorganisms is necessary to ensure the specificity and the
sensitivity of the newly designed probe. It is important to notice that the validation of a newly
designed probe requires different procedures for the dot blot (15) and FISH format (12).
Moreover, the hybridization and washing conditions (temperature, salt concentration and
detergent) are also crucial for obtaining a detectable probe signal (69). The accessibility of the
probe to its target site is another factor to be considered when designing new probes. The
accessibility of probe target sites on the 16S and 23S rRNA of Escherichia coli has been mapped
systematically by flow cytomety (FCM) and FISH and it was shown that probe-conferred
signal intensities vary greatly among different targets sites (23, 24). More recently, it was
demonstrated that accessibility patterns of 16S rRNAs are more similar for phylogenetically
related organisms; these findings may be the first description of consensus probe accessibility
maps for prokaryotes (5).
Hybridization techniques
Nucleic acid probing of complex communities comprises two major techniques: dot
blot hybridization and fluorescent in situ hybridization (FISH). In the dot blot format, total
DNA or RNA is extracted from the sample and is immobilized on a membrane together with
a series of RNA from reference strains. Subsequently, the membrane is hybridized with a
radioactively labeled probe and after a stringent washing step the amount of target rRNA is
quantified. The membrane can be rehybridized with a general bacterial probe and the amount
of population-specific rRNA detected with the specific probe is expressed as a fraction of the
total bacterial RNA. Quantification of the absolute and relative (as compared to total rRNA)
amounts of a specific rRNA reflects the abundance of the target population and thus do not
represent a direct measure of cell number since cellular rRNA content varies with the current
environmental conditions and the physiological activity of the cells at the time of sampling
(45). Dot blot hybridization has been successfully used to quantify rRNA from human fecal
and cecal samples (44, 65). It was found that strict anaerobic bacterial populations represented
by the Bacteroides, Clostridium leptum and Clostridium coccoides groups were significantly lower
in the cecum (right colon) than in the feces, while the Lactobacillus group was significantly
higher in the feces than in the cecum (44).
In contrast to dot blot hybridization, FISH is applied to morphologically intact
cells and thus provides a quantitative measure of the target organism without the limitation
of culture-dependent methods (2, 3). Following fixation, bacteria from any given sample can
be hybridized with an appropriate probe or set of probes. The fixation allows permeabilization
of the cell membrane and thus facilitates the accessibility of the fluorescent probes to the
target sequence. For some Gram-positive bacteria, especially lactobacilli, additional pre-
treatments including the use of cell wall lytic enzymes e.g. lysozyme, mutanolysin, protease K
or a mixture is needed (6, 28). Prior to hybridization, the cells can be either immobilized on
6
Chapter 1
gelatine-treated glass slides or simply kept in suspension when analyzed by FCM. The
stringency, i.e. conditions of hybridization that increase the specificity of binding between the
probe and its target sequence, can be adjusted by varying either the hybridization temperature
or formamide concentration. Under highly stringent conditions oligonucleotide probes can
discriminate closely related target sites. Post-hybridization stringency can be achieved by
lowering the salt concentration in the washing buffer in order to remove unbound probe and
avoid unspecific binding.
Quantification of FISH signals
Over the past years, significant methodological improvements of the probe
fluorescent-conferred signal have been reported. These include the use of (i) brighter
fluorochromes i.e. Cy3 and Cy5 (25, 68), (ii) unlabeled helper oligonucleotide probes (22)
(iii) signal amplification with reporter enzymes (CARD-FISH) (55), and (iv) the use of
peptide nucleic acid (PNA) probes (52, 56). Commonly epifluorescence microscopy is the
standard method by which fluorescent-stained cells are enumerated, however the method is
time consuming and subjective (38, 47). Recently, this technique has been improved by
development of automated image acquisition and analysis software allowing accurate
microscopic enumeration of fecal bacteria cells (34). Alternatively, FCM offers a potential
platform for high-resolution, high throughput identification and enumeration of
microorganisms using fluorescent rRNA-targeted oligonucleotides with the possibility of cell
sorting (60, 77, 78, 87).
A FCM method for direct detection of the anaerobic bacteria in human feces
was first described by Van der Waaij et al. (73). They used a membrane-impermeant
nucleic acid dye propidium iodide (PI) in combination with the intrinsic scatter parameters of
the cells to discriminate the fecal cells from large particles. Coupling FCM results and image
analysis, the authors showed that most of the particles detected with a large forward scatter
value corresponded to aggregates most likely representing mucus fragments and indigested
dietary compounds. They confirmed by means of cell sorting that the PI-stained cells (fecal
cells) corresponded to a 2-D surface area of <1.5 m
2
while the unstained particles (aggregates)
were around 5.0 m
2
(73). The work highlighted the potential of FCM to study anaerobic fecal
bacteria without culturing. Despite this valuable work and to quote from Shapiro the subject
matter may stink, but the method is superb (67), the application of FCM to study the
intestinal microbiota is still in its infancy.
Recently, FISH-FCM was applied to detect and accurately quantify both fecal and
mucosa-associated bacteria and statistical analysis showed a high correlation between the
FCM counts and microscopic counts (87, 89). Using FCM, several thousands of cells can be
counted accurately in a few seconds. Following the hybridization step, fecal cells are stained
with a nucleic acid dye (i.e. propidium iodide, SYTO BC, TOTO-1) to detect the total cells
and subsequently spiked with standard beads of known size and concentration. The beads are
7

thus used as an internal standard to calibrate the measured volume and to determine the
absolute count of the probe-detected cells (Chapter 5). In addition to the determination of
the absolute cell counts, the fluorescence intensity signal can also be quantified using
fluorescent beads with known fluorescent intensities (67). This is of major importance for
determining optimal hybridization conditions for newly designed probes (16, 61). Definitely,
FCM will become the method of choice for high-resolution, high throughput identification
of microorganisms using fluorescent rRNA-targeted oligonucleotides.
Application of FISH to study the GI-tract ecosystem
During the last five years, hybridizations with rRNA-targeted probes have provided a
significant knowledge about the structure of the gut microbiota. A large panel of
oligonucleotide probes specific for various genera predominant in the GI-tract have been
designed and validated. These include Clostridium, Bacteroides, Eubacterium, Ruminococcus,
Bifidobacterium, Lactobacillus, Streptococcus, Fusobacterium, Collinsella, Atopobium and
Veillonella specific probes (Table 1) and which have been used intensively to study the
composition and structure of the intestinal microbiota. .
The uniqueness and complexity of the human gut microbiota revealed
by fingerprinting techniques was supported by results of analysis using nucleic-acid
probes based methods. Results of such studies revealed that the majority of fecal bacteria
belong to the Bacteroides-Prevotella, C. coccoides, C. leptum group, Atopobium group and
bifidobacteria (21, 26, 35, 60). These investigations showed that the genus Bacteroides and
members of C. coccoides and C. leptum constitute more than half of the fecal microbiota.
Among members of the C. coccoides group which equates to Clostridium rRNA cluster XIVa
(11), Ruminococcus, Eubacterium hallii, Lachnospira and Eubacterium cylindroides related
bacteria were found to be dominant members of the microbiota. However, Enterobacteriaceae,
Lactobacillus-Enterococcus group, Phascolarctobacterium and relatives, and Veillonella were less
dominant (26). However, differences in the occurrence of these bacterial groups have been
reported by different research groups. These deviations may be due the different methods or
probes used but it is also likely that the observed variance is due to the differences in the
genetic background, lifestyle, and diet in the human populations studied (60). The results of
two extensive studies, where an extensive array of oligonucleotide probes targeting the major
bacterial groups in the GI-tract was used, showed that 62-75% of the fecal bacteria could be
detected and identified (26, 36). The remainder (~ 30%) could either belong to members of
the Archaea, Eukarya or most likely to yet unknown bacteria. Furthermore, FISH-FCM
analysis of fecal microbiota of patients with ulcerative colitis revealed substantial temporal
variations in the major bacterial groups studied (i.e. Bacteroides, C. coccoides, Atopobium,
bifidobacteria and lactobacilli) (Chapter 5).
8
Chapter 1
Table 1: Major FISH probes used to study the GI-tract microbiota.
Probe Probe Sequence (5-3) Target organism % Formamide Reference
Eub338 GCTGCCTCCCGTAGGAGT Most bacteria 0 80 (4)
EubII GCAGCCACCCGTAGGTGT Planctomycetes 0- 60 (12)
EubIII GCTGCCACCCGTAGGTGT Verrucomicrobia 0- 60 (12)
Bac303 CCAATGTGGGGGACCTT Bacteroides/Prevotella 0 (43)
Erec482 GCTTCTTAGTCAR*GTACCG Clostridium coccoides cluster 0 (21)
Elgc01 GGGACGTTGTTTCTGAGT Clostridium leptum cluster 0 (21)
Fprau645 CCTCTGCACTACTCAAGAAAA Fusobacterium prausnitzii 15 (71)
Bif164 CATCCGGCATTACCACCC Bifidobacteria 0 (35)
Ato291 GGTCGGTCTCTCAACCC Atopobium group 0 (27)
Veil223 AGACGCAATCCCCTCCTT Veillonella 0 (26)
Ecyl1387 CGCGGCATTGCTGCTTCA Eubacterium cylindroides 20 (26)
Rbro729 AAAGCCCAGTAAGCCGCC
Ruminococcus group 20 (26)
Rfla730 TAAAGCCCAGY*AGGCCGC
Lach571 GCCACCTACACTCCCTTT Lachonospira group 40 (26)
Ehal1469 CCAGTTACCGGCTCCACC Eubacterium hallii group 20 (26)
Phasco741 TCAGCGTCAGACACAGTC Phascolarctobacterium group 0 (26)
Bdis656 CCGCCTGCCTCAAACATA Bacteroides distasonis 0 (21)
Bfra602 GAGCCGCAAACTTTCACAA Bacteroides fragilis 30 (21)
Bvulg1017 AGATGCCTTGCGGCTTACGGC Bacteroides vulgatus 30 (61)
Bfrag998 GTTTCCACATCATTCCACTG Bacteroides fragilis 30 (61)
Bdist1025 CGCAAACGGCTATTGGTAG Bacteroides distasonis 30 (61)
Lab158 GGTATTAGCAY*CTGT TTCCA Lactobacillus/Enterococcus 0 (28)
Urobe63a AATAAAGTAATTCCCGTTCG Uncultured Ruminococcus 20 (87)
Urobeb63b AATRAARTATTTCCCGTTCG obeum-like bacteria
Non338 ACATCCTACGGGAGGC Negative control (77)
* R and Y are the International Union of Pure and Applied Chemistry codes for ambiguous bases.
9
1.4 Inferring structure to metabolic activity
The aforementioned molecular techniques have greatly contributed to our
fundamental understanding of the biodiversity, establishment, succession and structure of the
intestinal microbiota; yet little is known about the in situ association between the microbial
diversity and the metabolic activity of a phylogenetic affiliated group. It is well recognized that
this highly diverse microbiota plays a significant role in the processing of undigested food to
the benefit of the host and contributes to the host defense by limiting colonization of the GI-
tract by pathogens (17, 32). For instance the generation of short chain fatty acids is a common
feature of the climax community, although many of the specific species responsible remain
undefined. It is therefore a major challenge to develop methods that allow monitoring of
microorganisms according to their eco-physiological traits in situ.
During the last years several innovative methods have been developed to resolve the
linkage between structure, activity and function in microbial communities. These include
methods where molecular techniques are coupled with substrate labeling such as stable isotope
probing (SIP) (58, 59), microautoradiography and FISH (MAR-FISH) (37, 51) or labeling
with fluorescent functional probes followed by flow cytometry and cell sorting analysis (7, 81,
Chapter 7). MAR-FISH allows monitoring of the radiolabeled substrate uptake patterns of
the probe-identified organisms under different environmental conditions (13, 37). This
method has been applied with high throughput DNA microarray analysis to study the
complex activated sludge ecosystem (1). In stable isotope probing (SIP), either lipid
biomarkers (9), DNA (58) or RNA (42) are extracted from microbial communities incubated
with
13
C-labeled substrates. If cells grow on the added compounds, their pool of
macromolecules will be isotopically enriched (heavy) compared to those of inactive organisms.
For DNA- or RNA-SIP, identification of the metabolically active organisms (heavy) is
achieved by separation of community DNA/RNA according to their buoyant density by
means of equilibrium density-gradient centrifugation, followed by PCR-amplification of 16S
rRNA genes in the isotopically heavy DNA/RNA pool, cloning and sequencing. The use of
RNA was proposed as a more responsive biomarker as its turnover is much higher than that
of DNA (42). Phospholipid fatty acids are also used as biomarker for
13
C enrichments, but
their resolution for diversity analysis is less powerful than for sequence analysis. On the other
hand, FCM has been viewed as a powerful technique to monitor the metabolic activity of
stressed and starved bacteria and identifies microorganisms in their natural habitat, with
potential for automation (Chapter 2). One major advantage of FCM is that it allows
monitoring of bacterial heterogeneity at the single cell level and provides a mean to sort sub-
populations of interest for further molecular analysis (14). This approach has been ultimately
applied to fecal microbiota, the results provided relevant ecological information related to the
diversity and activity of different affiliated phylogenetic groups and highlighted the
physiological heterogeneity of this complex ecosystem (Chapter 7). The application of
10
Chapter 1
cytometric protocols using fluorescent probes in combination with molecular techniques
opens the potential for examining key microbial processes and community function in
complex microbial ecosystems.
1.5 Outline of the thesis
Throughout this thesis, the potential of flow cytometry (FCM) and fluorescence
activated cell sorting (FACS) for the analysis of the complex intestinal microbiota will be
demonstrated, with the ultimate aim to provide insight into the biodiversity of the intestinal
ecosystem coupled with the global in situ activity of these microbes.
Chapter 2 reviews the potential of FCM and FACS as an analytical and preparative tool
to analyze microorganisms in different environmental settings.
Chapters 3 describes the application of FCM in combination with FISH (FISH-FCM)
to identify and enumerate an uncultured group of fecal bacteria, which have only been detected
by PCR-based approaches.
Chapter 4 describes the distribution of the predominant and Lactobacillus group bacterial
community along different sites of the colon of different individuals some of which are diagnosed
with ulcerative colitis or polyposis. The results demonstrate the ability of FCM for studying not
only fecal bacteria (suspended cells) but also mucosa-associated microbiota (attached cells).
Chapter 5 describes the application of FISH-FCM and denaturing gradient gel
electrophoresis (DGGE) as a high throughput platform to evaluate the effect of two probiotic
strains on the population dynamics of fecal microbiota of patients with ulcerative colitis during
a probiotic trial.
Chapter 6 describes a new approach based on the use of functional probes to assess the
viability of Bifidobacterium adolescentis and Bifidobacterium lactis during bile salt stress and
highlights the importance of multiparametric FCM as a powerful technique to monitor
physiological heterogeneity including live, dead and injured cells within stressed populations at
the single cell level.
Chapter 7 illustrates a novel approach where functional probes and FACS are combined
with 16S rRNA gene analyses to get insight into the genetic diversity of live, dead and injured
fecal bacteria.
The summary and concluding remarks are presented in Chapter 8.
11

1.6 References
1. Adamczyk, J., M. Hesselsoe, N. Iversen, M. Horn, A. Lehner, P. H. Nielsen, M.
Schloter, P. Roslev, and M. Wagner. 2003. The isotope array, a new tool that employs
substrate-mediated labeling of rRNA for determination of microbial community
structure and function. Appl. Environ. Microbiol. 69:6875-6887.
2. Amann, R., B. M. Fuchs, and S. Behrens. 2001. The identification of microorganisms
by fluorescence in situ hybridisation. Curr. Opin. Biotechnol. 12:231-236.
3. Amann, R., W. Ludwig, and K. Schleifer. 1995. Phylogenetic identification and
in situ detection of individual microbial cells without cultivation. Microbiol.
Rev. 59:143-169.
4. Amann, R. I., B. Binder, R. Olson, S. W. Chisholm, R. Devereux, and D. Stahl. 1990.
Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for
analyzing mixed microbial populations. Appl. Environ. Microbiol. 56:1919-1925.
5. Behrens, S., B. M. Fuchs, F. Mueller, and R. Amann. 2003. Is the in situ accessibility
of the 16S rRNA of Escherichia coli for Cy3-labeled oligonucleotide probes predicted by
a three-dimensional structure model of the 30S ribosomal subunit. Appl. Environ.
Microbiol. 69:4935-4941.
6. Beimfohr, C., A. Krause, R. Amann, W. Ludwig, and K. H. Schleifer. 1993. In situ
identification of lactococci, enterococci and streptococci. Syst. Appl. Microbiol. 16:450-
456.
7. Bernard, L., C. Courties, C. Duperray, H. Schafer, G. Muyzer, and P. Lebaron. 2001.
A new approach to determine the genetic diversity of viable and active bacteria in aquatic
ecosystems. Cytometry 43:314-321.
8. Bonnet, R., A. Suau, J. Dore, G. R. Gibson, and M. D. Collins. 2002. Differences in
rDNA libraries of faecal bacteria derived from 10- and 25-cycle PCRs. Int. J. Syst. Evol.
Microbiol. 52:757-763.
9. Boschker, H. T. S., and J. J. Middelburg. 2002. Stable isotope and biomarkers in microbial
ecology. FEMS Microbiol. Ecol. 40:85-95.
10. Campieri, M., and P. Gionchetti. 2001. Bacteria as the cause of ulcerative colitis. Gut
48:132-135.
11. Collins, M. D., P. A. Lawson, A. Willems, J. J. Cordoba, J. Fernandez-Garayzabal, P.
Garcia, J. Cai, H. Hippe, and J. A. Farrow. 1994. The phylogeny of the genus
Clostridium: proposal of five new genera and eleven new species combinations. Int.
J. Syst. Bacteriol. 144:812-826.
12
Chapter 1
12. Daims, H., A. Bruhl, R. Amann, K. H. Schleifer, and M. Wagner. 1999. The
domain-specific probe EUB338 is insufficient for the detection of all bacteria:
development and evaluation of a more comprehensive probe set. Syst. Appl.
13. Daims, H., J. L. Nielsen, P. H. Nielsen, K. H. Schleifer, and M. Wagner. 2001. In situ
characterization of Nitrospira-like nitrite-oxidizing bacteria active in wastewater
treatment plants. Appl. Environ. Microbiol. 67:5273-5284.
14. Davey, H. M., and M. K. Winson. 2003. Using flow cytometry to quantify microbial
heterogeneity. Curr. Issues Mol. Biol. 5:9-15.
15. de los Reyes, F. L., W. Ritter, and L. Raskin. 1997. Group-specific small-subunit rRNA
hybridization probes to characterize filamentous foaming in activated sludge systems.
Appl. Environ. Microbiol. 63:1107-1117.
16. Derrien, M., K. Ben-Amor, E. E. Vaughan , and W. M. de Vos. 2004. Validation of 16S
rRNA probe specific for the novel intestinal mucin-degrader Akkermansis muciniphila.
p. 60. PROEUHEALTH: The Food, GI-tract Functionality and Human Health Cluster.
VTT Biotechnolgy (http://www.vtt.fi/inf/pdf ). Sitges, Spain.
17. Falk, P. G., L. V. Hooper, T. Midtvedt, and J. I. Gordon. 1998. Creating and maintaining
the gastrointestinal ecosystem: what we know and need to know from gnotobiology.
Microbiol. Mol. Biol. Rev. 62:1157-1170.
18. Favier, C. F., E. E. Vaughan, W. M. de Vos, and A. D. L. Akkermans. 2002. Molecular
monitoring of succession of bacterial communities in human neonates. Appl. Environ.
19. Finegold, S. M., V. L. Sutter, and G. E. Mathisen.1983. Normal indigenous flora.,
p. 3-31. In D. J. Hentges (ed.), Human intestinal microflora in health and disease.
Academic Press, New York, N.Y.
20. Formin, N., S. Hamelin, S. Tarnawski, D. Roesti, K. Jourdain-Miserez, N. Foresties,
S. Teyssier-Cuvelle, F. Gillet, M. Aragno, and P. Rossi. 2002. Statistical analysis of
denaturing gel electophoresis (DGE) fingerprinting patterns. Environ. Microbiol. 4:634-
643.
21. Franks, A. H., H. J. M. Harmsen, G. C. Raangs, G. J. Jansen, F. Schut, and G. W.
Welling. 1998. Variations of bacterial populations in human feces measured by
fluorescent in situ hybridization with group-specific 16S rRNA-targeted oligonucleotide
probes. Appl. Environ. Microbiol. 64:3336-3345.
22. Fuchs, B. M., F. O. Glockner, J. Wulf, and R. Amann. 2000. Unlabeled helper
oligonucleotides increase the in situ accessibility to 16S rRNA of fluorescently labeled
oligonucleotide probes. Appl. Environ. Microbiol. 66:3603-3607.
13

23. Fuchs, B. M., K. Syutsubo, W. Ludwig, and R. Amann. 2001. In situ accessibility of
Escherichia coli 23S rRNA to fluorescently labeled oligonucleotide probes. Appl. Environ.
24. Fuchs, B. M., G. Wallner, W. Beisker, I. Schwippl, W. Ludwig, and R. Amann. 1998.
Flow cytometric analysis of the in situ accessibility of Escherichia coli 16S rRNA for
fluorescently labeled oligonucleotide probes. Appl. Environ. Microbiol. 64:4973-4982.
25. Glockner, F. O., R. Amann, A. Alfreider, J. Pernthaler, R. Psenner, K. Trebesius, and K.
H. Schleifer. 1996. An in situ hybridization protocol for detection and identification of
planktonic bacteria. Syst. Appl. Microbiol. 19:403-406.
26. Harmsen, H. J. M., G. C. Raangs, T. He, J. E. Degener, and G. W. Welling. 2002.
Extensive set of 16S rRNA-based probes for detection of bacteria in human feces. Appl.
Environ. Microbiol. 68:2982-2990.
27. Harmsen, H. J. M., A. C. M. Wildeboer-Veloo, J. Grijpstra, J. Knol, J. E. Degener, and
G. W. Welling. 2000. Development of 16S rRNA-based probes for the Coriobacterium
group and the Atopobium cluster and their application for enumeration of
Coriobacteriaceae in human feces from volunteers of different age groups. Appl. Environ.
Microbiol. 66:4523-4527.
28. Harmsen, J. H. M., P. Elfferich, F. Schut, and G. W. Welling. 1999. A 16 S rRNA-
tageted probe for detection of lactobacilli and enterococci in faecal samples by
fluorescent in situ hybridization. Micriol. Ecol. Health Dis. 11:3-12.
29. Heilig, H. G. H. J., E. G. Zoetendal, E. E. Vaughan, P. Marteau, A. D. L. Akkermans,
and W. M. de Vos 2002. Molecular diversity of Lactobacillus spp. and other lactic acid
bacteria in the human intestine as determined by specific amplification of 16S ribosomal
DNA. Appl. Environ. Microbiol. 68:114-123.
30. Hold, G. L., S. E. Pryde, V. J. Russell, E. Furrie, and H. J. Flint. 2002. Assessment of
microbial diversity in human colonic samples by 16S rDNA sequence analysis. FEMS
Microbiol. Ecol. 39:33-39.
31. Holdeman, L. V., I. J. Good, and W. E. Moore. 1976. Human fecal flora: variation in
bacterial composition within individuals and a possible effect of emotional stress. Appl.
32. Hooper, L. V., T. Midtvedt, and J. I. Gordon. 2002. How host-microbial interactions
shape the nutrient environment of the mammalian intestine. Annu. Rev. Nutr. 22:283-
307.
33. Hooper, L. V., M. H. Wong, A. Thelin, L. Hansson, P. G. Falk, and J. I. Gordon. 2001.
Molecular analysis of commensal host-microbial relationships in the intestine. Science
291:881-884.
14
Chapter 1
34. Jansen, G. J., A. C. M. Wildeboer-Veloo, R. H. J. Tonk, A. H. Franks, and G. W.
Welling. 1999. Development and validation of an automated, microscopy-based method
for enumeration of groups of intestinal bacteria. J. Microbiol. Methods. 37:215-221.
35. Langendijk, P., F. Schut, G. Jansen, G. Raangs, G. Kamphuis, M. Wilkinson, and G.
Welling. 1995. Quantitative fluorescence in situ hybridization of Bifidobacterium spp.
with genus-specific 16S rRNA-targeted probes and its application in fecal samples. Appl.
36. Lay, C., L. Rigottier-Gois, K. Holmstrom, M. Rajilic, E. Vaughan , M. D. Collins, R.
Thiel, P. Namsolleck, M. Blaut, and J. Dore. 2004. Assessment of human faecal
microbiota composition using FISH combined with flow cytomety, Pan-European
comparison. p. 75. PROEUHEALTH: The food, GI-tract functionality and human
health cluster 3rd workshop. VTT Biotechnology (http://www.vtt.fi/inf/pdf ). Stiges,
Spain.
37. Lee, N., P. H. Nielsen, K. H. Andreasen, S. Juretschko, J. L. Nielsen, K.H. Schleifer, and
M. Wagner. 1999. Combination of fluorescent in situ hybridization and
microautoradiography - a new tool for structure-function analyses in microbial ecology.
38. Lipski, A., U. Friedrich, and K. Altendorf. 2001. Application of rRNA-targeted
oligonucleotide probes in biotechnology. Appl. Microbiol. Biotechnol. 56:40-57.
39. Loy, A., M. Horn, and M. Wagner. 2003. ProbeBase: An online resource for rRNA-
targeted oligonucleotide probes. Nucleic Acids Res. 31:514-516.
40. Ludwig, W., O. Strunk, R. Westram, L. Richter, H. Meier, Yadhukumar, A. Buchner, T.
Lai, S. Steppi, G. Jobb, W. Forster, I. Brettske, S. Gerber, A. W. Ginhart, O. Gross, S.
Grumann, S. Hermann, R. Jost, A. Konig, T. Liss, R. Lussmann, M. May, B. Nonhoff,
B. Reichel, R. Strehlow, A. Stamatakis, N. Stuckmann, A. Vilbig, M. Lenke, T. Ludwig,
A. Bode, and K. H. Schleifer. 2004. ARB: a software environment for sequence data.
Nucleic Acids Res. 32:1363-1371.
41. Maidak, B. L., J. R. Cole, T. G. Lilburn, C. T. Parker, Jr, P. R. Saxman, R. J. Farris,
G. M. Garrity, G. J. Olsen, T. M. Schmidt, and J. M. Tiedje. 2001. The RDP-II
(Ribosomal Database Project). Nucleic Acids Res. 29:173-174.
42. Manefield, M., A. S. Whiteley, R. I. Griffiths, and M. J. Bailey. 2002. RNA stable
isotope probing, a novel means of linking microbial community function to phylogeny.
43. Manz, W., R. Amann, W. Ludwig, M. Vancanneyt, and K. Schleifer. 1996. Application
of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria
of the phylum cytophaga-flavobacter-bacteroides in the natural environment. Microbiol.
142:1097-1106.
15

44. Marteau, P., P. Pochart, J. Dore, C. Bera-Maillet, A. Bernalier, and G. Corthier. 2001.
Comparative study of bacterial groups within the human cecal and fecal microbiota.
45. Molin, S., and M. Givskov. 1999. Application of molecular tools for in situ monitoring
of bacterial growth activity. Environ. Microbiol. 1:383-391.
46. Moore, W. E., and L. V. Holdeman. 1974. Human fecal flora: the normal flora of 20
Japanese-Hawaiians. Appl. Microbiol. 27:961-979.
47. Moter, A., and U. B. Gobel. 2000. Fluorescence in situ hybridization (FISH) for direct
visualization of microorganisms. J. Microbiol. Methods. 41:85-112.
48. Muyzer, G., E. de Waal, and A. Uitterlinden. 1993. Profiling of complex microbial
populations by denaturing gradient gel electrophoresis analysis of polymerase chain
reaction-amplified genes coding for 16S rRNA. Appl. Environ. Microbiol. 59:695-700.
49. Muyzer, G., and K. Smalla. 1998. Application of denaturing gradient gel electrophoresis
(DGGE) and temperature gradient gel electrophoresis (TGGE) in microbial ecology.
Antonie Van Leeuwenhoek 73:127-41.
50. Nagashima, K., T. Hisada, M. Sato, and J. Mochizuki. 2003. Application of new
primer-enzyme combinations to terminal restriction fragment length polymorphism
profiling of bacterial populations in human feces. Appl. Environ. Microbiol.
69:1251-1262.
51. Nielsen, J. L., D. Christensen, M. Kloppenborg, and P. H. Nielsen. 2003.
Quantification of cell-specific substrate uptake by probe-defined bacteria under in situ
conditions by microautoradiography and fluorescence in situ hybridization. Environ.
52. Oliveira, K., S. M. Brecher, A. Durbin, D. S. Shapiro, D. R. Schwartz, P. C. De
Girolami, J. Dakos, G. W. Procop, D. Wilson, C. S. Hanna, G. Haase, H. Peltroche-
Llacsahuanga, K. C. Chapin, M. C. Musgnug, M. H. Levi, C. Shoemaker, and H.
Stender. 2003. Direct identification of Staphylococcus aureus from positive blood culture
bottles. J. Clin. Microbiol. 41:889-891.
53. Ott, S. J., M. Musfeldt, D. F. Wenderoth, J. Hampe, O. Brant, U. R. Folsch, K. N.
Timmis, and S. Schreiber. 2004. Reduction in diversity of the colonic mucosa associated
bacterial microflora in patients with active inflammatory bowel disease. Gut
53:685-693.
54. Ouwehand, A. C., S. Salminen, and E. Isolauri 2002. Probiotics: an overview of
beneficial effects. Antonie van Leeuwenhoek 82:279-289.
16
Chapter 1
55. Pernthaler, A., J. Pernthaler, and R. Amann. 2002. Fluorescence in situ hybridization
and catalyzed reporter deposition for the identification of marine bacteria. Appl.
56. Perry-O'Keefe, H., S. Rigby, K. Oliveira, D. Sorensen, H. Stender, J. Coull, and J. J.
Hyldig-Nielsen. 2001. Identification of indicator microorganisms using a standardized
PNA FISH method. J. Microbiol. Methods. 47:281-292.
57. Rachmilewitz, D., K. Katakura, F. Karmeli, T. Hayashi, C. Reinus, B. Rudensky, S.
Akira, K. Takeda, J. Lee, K. Takabayashi, and E. Raz. 2004. Toll-like receptor 9 signaling
mediates the anti-inflammatory effects of probiotics in murine experimental colitis.
Gastroenterol. 126:520-528.
58. Radajewski, S., P. Ineson, N. R. Parekh, and J. C. Murrell. 2000. Stable-isotope
probing as a tool in microbial ecology. Nature 403:646-649.
59. Radajewski, S., I. R. McDonald, and J. C. Murrell. 2003. Stable-isotope probing of
nucleic acids: a window to the function of uncultured microorganisms. Curr. Opin.
Biotechnol. 14:296-302.
60. Rigottier-Gois, L., A. G. Le Bourhis, G. Gramet, V. Rochet, and J. Dore. 2003.
Fluorescent hybridisation combined with flow cytometry and hybridisation of total
RNA to analyse the composition of microbial communities in human faeces using 16S
rRNA probes. FEMS Microbiol. Ecol. 43:237-245.
61. Rigottier-Gois, L., V. Rochet, N. Garrec, A. Suau, and J. Dore. 2003. Enumeration of
Bacteroides species in human faeces by fluorescent in situ hybridisation combined with flow
cytometry using 16S rRNA probes. Syst. Appl. Microbiol. 26:110-118.
62. Rosenbaum, V., and D. Riesner. 1987. Temperature-gradient gel electrophoresis.
Thermodynamic analysis of nucleic acids and proteins in purified form and in cellular
extracts. Biophys. Chem. 26:235-246.
63. Satokari, R. M., E. E. Vaughan, A. D. Akkermans, M. Saarela, and W. M. de Vos 2001.
Polymerase chain reaction and denaturing gradient gel electrophoresis monitoring of
fecal Bifidobacterium populations in a prebiotic and probiotic feeding trial. Syst. Appl.
64. Seksik, P., L. Rigottier-Gois, G. Gramet, M. Sutren, P. Pochart, P. Marteau, R. Jian,
and J. Dore. 2003. Alterations of the dominant faecal bacterial groups in patients with
Crohn's disease of the colon. Gut 52:237-242.
65. Sghir, A., G. Gramet, A. Suau, V. Rochet, P. Pochart, and J. Dore. 2000. Quantification
of bacterial groups within human fecal flora by oligonucleotide probe hybridization.
66. Shanahan, F. 2003. Probiotics: a perspective on problems and pitfalls. Scand. J.
Gastroenterol. Suppl. 237:34-46.
17

67. Shapiro, H. M. 1995. Practical Flow Cytometry., 3d ed. Wiley-Liss Inc, New York.
68. Southwick, P. L., L. A. Ernst, E. W. Tauriello, S. R. Parker, R. B. Mujumdar, S. R.
Mujumdar, H. A. Clever, and A. S. Waggoner. 1990. Cyanine dye labeling reagents-
carboxymethylindocyanine succinimidyl esters. Cytometry 11:418-430.
69. Stahl, D. A., and R. Amann. 1991. Development and application of nucleic acid probes.
In E. Stackerbrandt, and M. Goodfellow (eds), Nucleic acid techniques in bacterial
systematics. John Wiley & Sons Ltd, Chichester.
70. Suau, A., R. Bonnet, M. Sutren, J. J. Godon, G. R. Gibson, M. D. Collins, and J. Dore.
1999. Direct analysis of genes encoding 16S rRNA from complex communities reveals many
novel molecular species within the human gut. Appl. Environ. Microbiol. 65:4799-4807.
71. Suau, A., V. Rochet, A. Sghir, G. Gramet, S. Brewaeys, M. Sutren, L. Rigottier-Gois,
and J. Dor. 2001. Fusobacterium prausnitzii and related species represent a dominant
group within the human fecal flora. Syst. Appl. Microbiol. 24:139-145.
72. Tannock, G. W., K. Munro, H. J. M. Harmsen, G. W. Welling, J. Smart, and P. K. Gopal.
2000. Analysis of the fecal microflora of human subjects consuming a probiotic product
containing Lactobacillus rhamnosus DR20. Appl. Environ. Microbiol. 66:2578-2588.
73. van der Waaij, L. A., G. Mesander, P. C. Limburg, and D. Wan der Waaij. 1994. Direct
flow cytometry of anaerobic bacteria in human feces. Cytometry 16:270-279.
74. Vaughan, E. E., H. G. H. J. Heilig, E. G. Zoetendal, R. M. Satokari, J. K. Collins,
A. D. Akkermans, and W. M. de Vos. 1999. Molecular approaches to study probiotic
bacteria. Trends Food Sci. Technol. 10:400-404.
75. Vaughan, E. E., F. Schut, H. G. H. J. Heilig, E. G. Zoetendal, W. M. de Vos, M, and
A.D.L. Akkermans. 2000. A molecular view of the intestinal ecosystem. Curr. Issues
Intest. Microbiol. 1:1-12.
76. Von Wintzingrode, F., U. B. Gbel, and E. Stackebrandt. 1997. Determination of
microbial diversity in environmental samples: pitfalls of PCR-based rRNA analysis.
FEMS Microbiol. Rev. 21:213-229.
77. Wallner, G., R. Amann, and W. Beisker. 1993. Optimizing fluorescent in situ
hybridization with rRNA-targeted oligonucleotide probes for flow cytometric
identification of microorganisms. Cytometry 14:136-143.
78. Wallner, G., B. Fuchs, S. Spring, W. Beisker, and R. Amann. 1997. Flow sorting of
microorganisms for molecular analysis. Appl. Environ. Microbiol. 63:4223-4231.
79. Walter, J., C. Hertel, G. W. Tannock, C. M. Lis, K. Munro, and W. P. Hammes. 2001.
Detection of Lactobacillus, Pediococcus, Leuconostoc, and Weissella species in human feces by
using group-specific PCR primers and denaturing gradient gel electrophoresis. Appl.
18
Chapter 1
80. Wang, X., S. P. Heazlewood, D. O. Krause, and T. H. J. Florin. 2003. Molecular
characterization of the microbial species that colonize human ileal and colonic mucosa by
using 16S rDNA sequence analysis. Appl. Microbiol. 95:508-520.
81. Whiteley, A. S., R. I. Griffiths, and M. J. Bailey. 2003. Analysis of the microbial
functional diversity within water-stressed soil communities by flow cytometric analysis
and CTC
+
cell sorting. J. Microbiol. Methods. 54:257-267.
82. Whitfield, J. 2004. Features-Science and health: microbial soup of life is sieved for
treasure, Financial Times.
83. Wilson, K., and R. Blitchington. 1996. Human colonic biota studied by ribosomal
DNA sequence analysis. Appl. Environ. Microbiol. 62:2273-2278.
84. Zoetendal, E. G., A. D. L. Akkermans, W. M. Akkermans-van Vliet, J. A. G. M. de
Visser, and W. M. de Vos. 2001. The host genotype affects the bacterial community in
the human gastrointestinal tract. Microbial. Ecol. Health Dis. 13:129-134.
85. Zoetendal, E. G., A. D. L. Akkermans, and W. M. De Vos. 1998. Temperature gradient
gel electrophoresis analysis of 16S rRNA from human fecal samples reveals stable and
host-specific communities of active bacteria. Appl. Environ. Microbiol. 64:3854-3859.
86. Zoetendal, E. G., K. Ben-Amor, A. D. L. Akkermans, T. Abee, and W. M. de Vos. 2001.
DNA isolation protocols affect the detection limit of PCR approaches of bacteria in samples
from the human gastrointestinal tract. Syst. Appl. Microbiol. 24:405-410.
87. Zoetendal, E. G., K. Ben-Amor, H. J. M. Harmsen, F. Schut, A. D. L. Akkermans, and
W. M. de Vos. 2002. Quantification of uncultured Ruminococcus obeum-like bacteria in
human fecal samples by fluorescent in situ hybridization and flow cytometry using 16S
rRNA-targeted probes. Appl. Environ. Microbiol. 68:4225-4232.
88. Zoetendal, E. G., C. T. Collier, S. Koike, R. I. Mackie, and H. R. Gaskins. 2004.
Molecular ecological analysis of the gastrointestinal microbiota: a review. J. Nutr.
134:465-472.
89. Zoetendal, E. G., A. von Wright, T. Vilpponen-Salmela, K. Ben-Amor, A. D. L.
Akkermans, and W. M. de Vos. 2002. Mucosa-associated bacteria in the human
gastrointestinal tract are uniformly distributed along the colon and differ from the
community recovered from feces. Appl. Environ. Microbiol. 68:3401-3407.
19

.
Chapter 2
FLOW CYTOMETRIC ANALYSIS OF
MICROOGRANISMS
Kaouther Ben Amor, Willem M. de Vos and Tjakko Abee
Abstract: Flow cytometry analysis of fluorescently-labeled microorganisms has a
wide range of applications including detection, identification, viability assessment,
analysis of cellular function, as well as heterogeneity assessment of cell
populations. This chapter seeks to review the recent applications of flow cytometry
and fluorescent probes in microbial ecology and physiology and highlights the
progress made in developing new strategies for use in microbiological
investigations
21
2.1 Introduction
The continuous improvement of the sensitivity and the performance of flow
cytometric instruments has resulted on a wide range of applications to characterize bacteria,
yeast, fungi and even viruses (76, 89, 113). Until the late 1970s, applications of flow
cytometry (FCM) in the field of microbiology were rather limited due to the fact that most
of the flow cytometers available at that time were not suited for measurement of bacteria due
to their small size compared to that of mammalian cells. The first applications of flow
cytometry in the field of microbiology were published by Paau et al. (99) and Bailey et al. (11)
who studied the cell cycle of three bacterial species with different growth rates (Escherichia
coli, Rhizobium meliloti and Rhizobium japonicum) using a combination of light scattering and
ethidium bromide fluorescence signals. Hutter et al. (59) published a series of pioneering flow
cytometric studies demonstrating the suitability of the technique to determine the DNA and
protein content of several types of microorganisms and to discriminate live and dead cells on
the basis of their light scattering behaviour. However, Steen and co-workers (120, 121) were
the first to design a flow cytometer well suited for the analysis of bacteria and this led to a
breakthrough in the field of microbial FCM. The power of FCM stems from the ability to
perform multiparametric analysis at the single cell level, the high throughput capacity, and the
option of cell sorting. In this chapter we will discuss the principle of FCM and highlight a
number of applications in microbial ecology and physiology.
2.2 How does it work?
Flow cytometry is a mean of measuring specific physical and chemical characteristics
of cells or particles as they flow single-filed in a liquid stream through the focus of a laser
beam(s). At the sensing or measuring point, the stained cells will scatter light in different
directions and emit fluorescence. These light pulses are then collected by an array of detectors,
which in turn translate these signals into electrical pulses (voltage) (Fig. 1). The voltage level
of each detector, either a photo multiplier tube (PMT) or photodiode, can be adjusted to
optimize signal amplification. Logarithmic amplification is used to provide a wide dynamic
range so that both weak and strong signals can be recorded in the same scale. The analog
signal is then converted to a digital value, which is stored in a list mode data files where each
event (i.e. presence of a microbial cell) with the corresponding data for each parameter is
recorded sequentially. One of the main aims in analyzing flow cytometric data is to distinguish
individual target cells among the total population. This is accomplished in a first step by
setting an electronic threshold or by using electronic gating in order to minimize the
background noise or exclude non-targeted cells, respectively. Statistical analysis is then used to
generate representing cell counts including, the median, the mean, the standard deviation and
the coefficient of variance of the measured parameters. For visualization purposes, data are
displayed either as a frequency distribution where the magnitude of the parameter measured
is expressed as a function of number of cells, or two- or three-parameter dot plots or density
22
Chapter 2
plots (Fig. 2). For multiparametric analysis, more advanced multivariate statistical methods
such as principal component analysis, cluster analysis or neural networks can be used in order
to extract useful information from the large data sets (38). For a more detailed discussion on
the principles of FCM and data analysis the monograph by Shapiro (113) is recommended.
Figure 1: Standard optical detector array of a FACSCalibur cytometer equipped with a dual- laser (a blue
and red-diode lasers emitting at 488nm diode 623 nm, respectively). A cell is intercepted at the focused
laser beam (s) within the sensing region of the flow cell. Light is scattered by the cell in the forward angle
and detected by a photodiode (FSC). Light scattered at right angles to the cell passes through a dichroic
filter than splits the light wavelengths > 560 nm and < 560 nm. Fluorescence > 560 nm is subsequently
split again with a 640 nm long pass filter. The wave lengths > 650 nm are detected by the red
fluorescence PMT (FL3) and the range of wavelengths within approximately 560-545 nm is detected by
the orange fluorescence (FL2) PMT. Fluorescence within the range 515-545 nm is collected by the green
fluorescence (FL1) PMT. (With permission from Beckton and Dickinson, The Benelux).
23
Flow Cytometry
Figure 2: FCM data display. The data obtained from the FCM analysis can be displayed in different ways:
the one-parameter or frequency histogram (A), the two-parameter dot plot (B), and a three dimensional
representation of the data (C), generated from the analysis of Bifidobacterium adolescentis during pH, heat
and bile salt stress exposure, respectively. The plots were generated with WinMDI software available at
http//:facs.Scripps.edu/software.html.
2.3 What does it Measure?
A flow cytometer measures the light scattered by and the fluorescence emitted from
particles or cells upon excitation with a light source (laser) as they pass in liquid fluid. The light
scatter parameters measured by the FCM known as forward scatter (FSC) and side scatter
(SSC) provide information about the intrinsic cell properties. As a general rule, FSC is used to
estimate cell size and volume while the SSC parameter is a rough estimate of the internal cell
structure and granularity (1, 37, 39, 112, 113). Taken together, FSC and SSC can distinguish
cells in a mixed sample according to their morphological fingerprints, thus allowing exclusion
of aggregates or debris from the cell of interest. A FCM method for direct detection of
anaerobic bacteria in human feces and colon biopsies was described using propidium iodide
(PI), a nucleic acid dye, and scatter parameters to discriminate fecal and mucosa-associated
bacteria from non-bacterial aggregates (127, 128, 144). Combining cell sorting and image
24
Chapter 2
analysis, van der Waaij et al. (128) showed that particles with high value in the FSC represented
aggregated particles presumably food particles or mucus fragments. These parameters were also
used to determine bacterial cell biomass, size and volume (145). Sincock et al. (116)
discriminated populations of closely related Gram-positive spores based on their scattering
profiles.
Figure 3: Different cellular target sites for physiological and taxonomic fluorescent probes (See text for
further explanation).
While measurements of FSC/SSC parameters can be useful to characterize
bacterial cells and exclude background, it is the capability of the flow cytometer to
measure particle-associated fluorescence that makes the technique extremely attractive.
Commonly, the fluorescence emitted from the stained cell represents the expression
of an intracellular marker or a reporter molecule attached to an oligonucleotide or to an
antibody. The advance in fluorescent technology allowed the development of a wide range
of fluorescent probes (Table 1) targeting a large array of cellular site parameters (Fig. 3)
and well suited for FCM (53, 112). These include stains that have a high specificity for nucleic
acids, total proteins, or lipids. Indicators that reflect enzyme activity such as esterases,
galactosidases or dehydrogenases are also available and have been used for a wide range
25
Flow Cytometry
26
Chapter 2
T
a
b
l
e

1
:

C
o
m
m
o
n
l
y

u
s
e
d

f
l
u
o
r
e
s
c
e
n
t

p
r
o
b
e
s

t
o

s
t
u
d
y

m
i
c
r
o
o
r
g
a
n
i
s
m
s

b
y

F
C
M

(
3
9
,

5
3
)
.
F
l
u
o
r
e
s
c
e
n
t

p
r
o
b
e
E
x

(
m
a
x
)
a
(
n
m
)
E
m

(
m
a
x
)

(
n
m
)
L
i
g
a
n
d

o
r

s
u
b
s
t
r
a
t
e
A
p
p
l
i
c
a
t
i
o
n
s
P
r
o
p
i
d
i
u
m

I
o
d
i
d
e

(
P
I
)
5
3
6
6
1
7
D
N
A
,

R
N
A
V
i
a
b
i
l
i
t
y
,

D
N
A

c
e
l
l

c
y
c
l
e
T
O
T
O
-
1
5
1
4
5
3
3
D
N
A
,

R
N
A
D
e
t
e
c
t
i
o
n
,

e
n
u
m
e
r
a
t
i
o
n
S
Y
T
O

1
3
4
8
8
5
0
9
D
N
A
,

R
N
A
V
i
a
b
i
l
i
t
y
,

G
r
a
m

s
t
a
i
n
i
n
g
S
Y
B
R

G
r
e
e
n

I
4
9
4
5
2
1
D
N
A
,

R
N
A
D
N
A

q
u
a
n
t
i
f
i
c
a
t
i
o
n
,

V
i
a
b
i
l
i
t
y
T
O
P
R
O
-
3
6
4
2
6
6
1
D
N
A
,

R
N
A
V
i
a
b
i
l
i
t
y
D
A
P
I
3
5
8
4
6
1
D
N
A
/
R
N
A
D
e
t
e
c
t
i
o
n
,

e
n
u
m
e
r
a
t
i
o
n
H
o
e
s
h
s
t
3
3
2
5
8
/
2
2
3
4
2
3
4
0
4
5
0
D
N
A

(
G
C

p
a
i
r
s
)
D
e
t
e
r
m
i
n
a
t
i
o
n

o
f

%

G
C

c
o
n
t
e
n
t
,

H
e
x
i
d
i
u
m

i
o
d
i
d
e

(
H
I
)
5
1
8
6
0
0
D
N
A
,

R
N
A
G
r
a
m

s
t
a
i
n
i
n
g
F
I
T
C
4
9
5
5
2
5
P
r
o
t
e
i
n
D
e
t
e
c
t
i
o
n
,

S
i
z
e
c
F
D
A
-
S
E
5
1
9
5
4
2
P
r
o
t
e
i
n
C
e
l
l

t
r
a
c
k
i
n
g
N
i
l
e

R
e
d
5
5
1
6
3
6
L
i
p
i
d
s
P
o
l
y
-

-
h
y
d
r
o
x
y

b
u
t
y
r
a
t
e

p
r
o
d
u
c
t
i
o
n
I
n
d
o
-
1
3
3
0
-
3
5
0
3
9
0
-
4
8
5
C
a
2
+
C
a
l
c
i
u
m

c
o
n
c
e
n
t
r
a
t
i
o
n
F
l
u
o
-
3
5
0
6
5
2
6
C
a
2
+
C
a
l
c
i
u
m

c
o
n
c
e
n
t
r
a
t
i
o
n
R
h
o
d
a
m
i
n
e

1
2
3
5
1
0
5
8
0
M
e
m
b
r
a
n
e

p
o
t
e
n
t
i
a
l
V
i
a
b
i
l
i
t
y

o
f

G
r
a
m
+

b
a
c
t
e
r
i
a
,

O
x
o
n
o
l

[
D
i
B
A
C
4
(
3
)
]
4
8
8
5
2
5
M
e
m
b
r
a
n
e

p
o
t
e
n
t
i
a
l
V
i
a
b
i
l
i
t
y
,

A
n
t
i
b
i
b
i
o
t
i
c

s
u
s
c
e
p
t
i
b
i
l
i
t
y
B
C
E
C
F
4
6
0
-
5
1
0
5
2
0
-
6
1
0
p
H

(
6
,

9
)
C
e
l
l

i
n
t
e
r
n
a
l

p
H
,

V
i
a
b
i
l
i
t
y

S
N
A
R
F

-
1
4
9
0
-
5
4
0
5
8
7
-
6
3
5
p
H
C
e
l
l

i
n
t
e
r
n
a
l

p
H
C
F
D
A
4
9
2
5
1
7
E
s
t
e
r
a
s
e
s
V
i
a
b
i
l
i
t
y
C
a
l
c
e
i
n
4
9
4
5
1
7
E
s
t
e
r
a
s
e
s
V
i
a
b
i
l
i
t
y
F
D
G
4
9
1
5
1
4
-
G
a
l
a
c
t
o
s
i
d
a
s
e
R
e
p
o
r
t
e
r

g
e
n
e

e
x
p
r
e
s
s
i
o
n
,

C
T
C
5
3
0
-
5
5
0
v
a
r
i
e
s
D
e
h
y
d
r
o
g
e
n
a
s
e
s
R
e
s
p
i
r
a
t
o
r
y

a
c
t
i
v
i
t
y
,

v
i
a
b
i
l
i
t
y
F
u
n
-
1
5
0
8
5
2
5
-
5
9
0
Y
e
a
s
t

v
a
c
u
o
l
a
r

e
n
z
y
m
e

a
c
t
i
v
i
t
y
Y
e
a
s
t

m
e
t
a
b
o
l
i
c

a
c
t
i
v
i
t
y
C
a
l
c
o
f
l
u
o
r
3
4
7
4
3
6
C
h
i
t
i
n

a
n
d

o
t
h
e
r

c
a
r
b
o
h
y
d
r
a
t
e
F
u
n
g
a
l

d
e
t
e
c
t
i
o
n
C
y
3

5
5
0
5
7
0
N
u
c
l
e
o
t
i
d
e

s
e
q
u
e
n
c
e
,

A
n
t
i
b
o
d
i
e
s
I
d
e
n
t
i
f
i
c
a
t
i
o
n

C
y
5
6
5
1
6
7
4
N
u
c
l
e
o
t
i
d
e

s
e
q
u
e
n
c
e
,

A
n
t
i
b
o
d
i
e
s
I
d
e
n
t
i
f
i
c
a
t
i
o
n
a
E
x

(
m
a
x
)

a
n
d

E
m

(
m
a
x
)

a
r
e

t
h
e

w
a
v
e
l
e
n
g
t
h
s

o
f

m
a
x
i
m
a
l

e
x
c
i
t
a
t
i
o
n

a
n
d

e
m
i
s
s
i
o
n
,

r
e
s
p
e
c
t
i
v
e
l
y
.

I
t

s
h
o
u
l
d

b
e

n
o
t
e
d

t
h
a
t

t
h
e
s
e

p
a
r
a
m
e
t
e
r
s

a
r
e

d
e
p
e
n
d
e
n
t

t
o

a

v
a
r
i
a
b
l
e

d
e
g
r
e
e

o
n

t
h
e

c
o
n
d
i
t
i
o
n
s

u
s
e
d

f
o
r

m
e
a
s
u
r
e
m
e
n
t
.
of applications. Other fluorochromes are useful because their properties change as a function
of pH or because they are accumulated or extruded as a response to cell energization. The use
of fluorescently-labeled oligonucleotide probes, the use of fusions to fluorescent reporter
proteins such as GFP and the detection of molecular interaction by Fluorescent Resonance
Energy Transfer (FRET) offer other means to study microorganisms and their components by
FCM (36, 98, 132, 139, 143).
2.4 What makes flow cytometry a powerful technology?
2.4.1 Multiparametric measurements
Flow cytometry is the most effective single-cell analysis technology available today,
since each cell can be characterized with 5-10 parameters allowing a multiparametric
characterization of a given cell population (38, 39, 112). Indeed, combinations of scatter
parameters with marker-conferred fluorescence are used to simultaneously characterize two or
more cellular properties i.e. DNA content, protein content, enzyme activity or simply to
identify target cells using oligonucleotides or antibodies (39). The combination of functional
probes and FCM has been applied to study the stress response of a number of lactic acid
bacteria and bifidoabcteria during heat, acid, ethanol and bile salt stress (15, 25, 34). The
multiparametric assay allowed to discriminate subpopulations with different physiological
status: live, dead and injured-damaged cells (Fig. 4). With the advent of FCM, it became
increasingly clear that even clonal populations derived from a single cell are far from
homogeneous. The development of complementary technologies using multiplexed bead-
based arrays will offer even more possibilities for studying microbial eco-physiology (62, 96,
118, 119).
2.4.2 High throughput analysis.
The utility of FCM as a high throughput approach stems from the ability to perform
quantitative assays on a large number of cells, with single cell resolution. Analyses are
commonly performed at a flow rate of 10-100 l/minute thus enabling tens of thousands of
cells to be collected from one sample in less than one minute and statistics are then generated
in real-time. Recent development such as serial-sample delivery using micro-well plates will
extend the ability of FCM resulting in higher throughput and automation analysis of a large
number of samples (48, 72).
2.4.3 Cell sorting.
The most attractive feature of FCM is that cells of interest can be physically separated for
subsequent molecular analysis (16, 133) or functional assays (56, 93), the so-called fluorescence
activated cell sorting (FACS). The cells of interest can be separated from a complex mixture even
though it may be a minor subpopulation and thus allowing for physical enrichments. A
morphologically conspicuous bacterium that accounted for less than 0.01% of the total microbial
27
Flow Cytometry

community of activated sludge was physically sorted using the fluorescent signal conferred by a
specific oligonucleotide probe and forward scatter as sorting criteria. Cell sorting resulted in a 30%
enrichment of the target bacteria while traditional microbiological enrichment had failed (117).
Multiparameter FCM combined with cell sorting unquestionably adds an extra dimension to the
information one gains from the results of the post-sorting assays (36, 126).
Figure 4: Dual parameter dot plot of B. adolescentis DSM 20083 representing carboxyfluorescein (cF)
versus propidium iodide (PI) fluorescence. Cultures were exposed for 10 min to different concentrations
of deconjugated bile salts (0.1 %, left and 0.25 %, right) and subsequently stained with cFDA and PI in
order to monitor the esterase activity and membrane permeability, respectively. Three main subpopulations
can be readily differentiated, corresponding to viable cF-stained cells, injured cells double stained with PI
and cF and dead PI-stained cells. The number of colony forming units corresponded to the number of cF-
stained cells, while the injured cells were not detected by the plate count method.
2.5 Applications of FCM in microbiology
Excellent reviews describing applications of FCM in the field of general (21, 39, 116,
137), clinical (3), food (8, 116) and environmental microbiology (65, 104) have been published.
In this section we will focus more specifically on viability assessment, detection and identification
of microorganisms, and the use of gene reporter systems in combination with FCM.
2.5.1 Assessment of bacterial viability and metabolic activity
The most important application of FCM in the field of microbiology is the assessment
of cell viability. Although viability is one of the most basic properties of a cell, its definition still
remains a topic of discussion among microbiologists (13, 19, 68). The gold standard method
28
Chapter 2
for determination of cell viability is the plate count method, which measures the ability of cells
to replicate under the conditions provided for growth. However, this does not implicate that
cells that cannot grow on an agar plate are not viable. Indeed, it is well recognized that under
stress conditions or limitation of nutrient availability, some cells can enter a non-culturable state
(dormant cells or injured cells), yet they can still exhibit metabolic activity (13). Fluorescent
techniques together with FCM offer an additional tool to characterize the physiological
heterogeneity of a cell population at the single cell level and provide an encompassing view of
the response of a cell to environmental changes. A number of criteria for viability assessment by
FCM have been suggested; including membrane permeability/membrane integrity, enzyme
activity and /or maintenance of a membrane potential (93).
Membrane integrity
The cytoplasmic membrane is the primary barrier that allows the cells to selectively
interact with their environment. The loss of membrane integrity represents significant damage
for the cell due to multiple functions linked to it (selective permeability, active transport,
motility, etc). Therefore integrity of the cell plasma membrane has been exploited as an
indicator of cell viability in both prokaryotes and eukaryotes. The measurement of membrane
integrity is based on the capacity of a bacterial cell either to exclude or retain certain
fluorescent dyes. Most of the membrane integrity assays are based on the dye exclusion test
using the membrane-impermeant nucleic acid stains. These dyes can passively enter cells with
compromised membranes, resulting in an enhanced fluorescence upon nucleic acid binding
(Fig. 3). A number of membrane-impermeant dyes, including propidium iodide (PI),
ethidium bromide (EB), SYTOX Green, TOTO-series and TO-PRO series, have been used as
membrane damage or as cell death indicators, however PI exclusion remains the method of
choice for viability assessment of microorganisms (40, 49, 54, 58, 91, 112, 134). The dye
exclusion test is not without limitations, indeed transient permeability to PI and TOPRO-3
was observed in antibiotic-treated Staphylococcus aureus (97), starved Micrococcus luteus (66)
and bile salt-stressed bifidobacteria (15). In our studies of bacterial viability, one interesting
finding was that the plasma membrane of the injured cells was apparently permeable to the
membrane impermeant dye PI. An early stage of the recovery/resuscitation process involved
repair of the damaged membrane, and these cells could actually be scored dead by traditional
criteria (15, 24, 35). Therefore, membrane-permeant dyes such as the SYTO-series, SYBR I
Green, that cross the cytoplasmic membranes of both dead and living cells can be used in
combination with the above mentioned dyes for a better resolution between live, injured and
dead cells (12, 49). The commercial live/dead containing the impermeant dye PI and the
permeant stain SYTO 9 is now widely used for the assessment of bacterial viability (20, 53).
29
Flow Cytometry

Membrane potential
Membrane potential plays a critical role in bacterial physiology. As a component of the
proton motive force, it is intimately involved in the generation of ATP, but it has also been
implicated in various energy-requiring processes, such as nutrient transport, chemotaxis,
survival at low pH, and bacterial autolysis. Membrane potential analysis is based on the
selective permeability and active transport of charged molecules through intact membranes.
Cells with a trans-membrane potential actively take up lipophilic cationic dyes such as
Rhodamine 123 and 3, 3-dihexyloxacarbocyanine (DiOC6(3)) or actively exclude lipophilic
anionic dyes such as the negatively charged bis-(1,3-dibutylbarbutiric acid trimethine oxonol)
DiBAC4(3) known as oxonol (Fig. 3). The specific accumulation of Rhodamine 123, which is
taken up by polarized cells, has been used together with FCM to assess the viability of starved
M. luteus (67), to monitor the survival of starved Escherichia coli and Salmonella populations
in seawater (77), and to assess mitochondrial membrane potential in Zygosaccharomyces bailii
and Saccharomyces cerevisiae (81). However, the limited applicability of Rhodamine 123
especially for Gram-negative bacteria, which need to be permeabilized prior to the staining,
has led to the extensive use of oxonol for rapid assessment of the microbial response to
antibiotics (28, 64, 85, 122, 123) as well as for cell viability (9, 54, 77, 78, 86). Correcting
the fluorescence conferred by oxonol for bacterial cell size, a better discrimination was
obtained between viable and depolarized/dead cells in bile salt-stressed bifidobacteria (15).
Enzyme activity
Monitoring the cellular esterase activity is another approach to determine the viability
of cell populations by means of FCM. Assessment of the esterase activity is determined using
a lipophylic, uncharged and non-fluorescent precursor. A number of fluorochromes such as 5,
(and-6)-carboxyfluorescein diacetate (cFDA), calcein-AM, 5, (and-6)-carboxyfluorescein
succinimidyl ester cFDASE; 2,7bis (2-carboxyethl)-5-(and-6)-caboxyfluorescein acetoxy
methyl ester (BCECF-AM) have been extensively used to monitor the viability of a wide range
of microorganisms (Table 1) (9, 15, 22, 25, 29, 44, 84, 105). These fluorogenic molecules can
diffuse across the membrane of viable cells where they are cleaved by non-specific esterases and
converted to polar fluorescent products such as fluorescein or fluorescein derivatives (Fig. 3).
The staining capacity and subsequent retention of the fluorescent substrate by intact cells are
good indicators for metabolic activity and membrane integrity of the cell. However, the major
limitations encountered with the fluorogenic esterases are related to a poor dye uptake
especially by Gram-negative bacteria and active dye-extrusion resulting in non-stained viable
cells (23). On the other hand, since cFDA-cleaving reactions are typically not energy
dependent, some authors argue that cFDA alone cannot be used as a viability indicator. Da
Silveira et al. (34) demonstrated that in addition to cF-labeling, the subsequent extrusion of
the probe provided a high sensitive indicator of stress response in ethanol-stressed and
ethanol-adapted Oenococcus oeni cells during malolactic fermentation. The combination of
30
Chapter 2
cFDA with other viability indicators such as PI, TOTO-1 or oxonol provided more solid
information to assess the viability of stressed bacteria (9, 15, 24, 34, 141).
The respiration activity in aerobic bacteria can be detected using the substrate 5-cyano-
2, 3-ditolyl-tetrazolium chloride (CTC). CTC acts as an electron acceptor in the electron
transport system and can be reduced by a variety of dehydrogenases to an insoluble fluorescent
formazan, which accumulates inside the cell (Fig. 3). Since electron transport is directly related
to cellular energy metabolism in respiring cells, the ability of cells to reduce tetrazolium
compounds can be considered an indicator of bacterial activity. The CTC assay in combination
with FCM has been extensively used as a measure of cellular metabolic activity of different
bacterial species (66, 77, 108, 122, 123). Furthermore, its applicability to a number of anaerobic
bacteria including the sulphate reducers and methanogens has been demonstrated (18). The
redox CTC dye has gained plenty of applications in the study of microbial activity in different
ecosystems including drinking water (111), seawater (16, 114), activated sludge (94), soil (136),
food matrices (51, 138) and biofilms (7, 83). However, the universality of CTC to detect
respiring cells in natural samples remains arguable, due to its possible toxic effect on bacterial
cells (129).
2.5.2 Reporter gene expression systems
FCM has been used to monitor gene expression by reporter genes in yeasts (10, 32),
bacteria (2, 31, 126) and to study microbe-host interaction (126). The LacZ gene encoding
-galactosidase and the green fluorescent protein (GFP) from the jellyfish (Aequorea victoria) are
particularly useful for such studies because they enable gene expression in individual cells to be
examined non-destructively and in real time (137)
The sensitive fluorogenic substrate fluorescein di--galactopyranoside (FDG) has been
proven effective for monitoring -galactosidase expression levels in single cells of bacteria and
yeast using FCM. The non-fluorescent FDG substrate is hydrolyzed by cellular -galactosidases
first to fluorescein monogalactosidase and then to the highly fluorescent fluorescein (Fig. 3).
A ready to use kit is available and widely used for mammalian cells, however its application for
bacteria is somewhat limited due the poor substrate uptake and retention of the fluorescent
product in active cells (53, 103). Different approaches have been used to overcome
these problems such as the development of lipophylic derivatives of FDG, use of hypertonic
shock and encapsulation of single cells in agarose microbeads. Nir et al. (95) showed that
encapsulation of E. coli and Candida pseudotropicalis in agar beads allowed the diffusion of
FDG into the microcolonies without permeabilization, and viable cells were then analyzed and
sorted by FACS on the basis of their native -galactosidase activity (110). In another study
it was shown that exposure to osmotic shock resulted in the uptake of FDG by Myxococcus
xanthus strains containing transcriptional fusions to lacZ and allowed sorting of
31
Flow Cytometry

subpopulations according to their level of -galactosidase expression. A lipophilic derivative of
FDG (C8-FDG), which yields a product that is better retained by cells, has been used to study
the gene expression in sporulating cultures of Bacillus subtilus (31, 52).
Advances in engineering of the green fluorescent protein (GFP) from Aequorea victoria
into mutants with improved properties and altered colours have provided the basic tools that
allow the investigation of complex processes in live cells (101, 124, 125, 140). A major
advantage of using GFP is that it can be used to stain living cells and monitor them in real time.
GFP in combination with FCM has been used to detect and enumerate starved bacteria (30,
79), to investigate the heterogeneity of stress gene expression in S. cerevisiae during heat
treatment (10) and to track Campylobacter jejuni, a food-borne pathogen, in a mouse model
(90). Gunasekera et al. (50) used GFP production and PI uptake to monitor the gene expression
and viability of Pseudomonas putida, a psychotrophic milk spoilage bacterium, following
pasteurization. Their results showed that a substantial fraction of cells that were incapable of
forming colonies as a result of the heat stress, were metabolically active since they were able to
transcribe and translates genes, which in turn may affect milk quality and safety. The interaction
between bacteria and host cells has been also studied in vivo and ex-vivo using GFP in
conjunction with fluorescence activated cell sorting (46, 126).
A pioneering assay using -lactamase as reporter system has recently been reported by
Zlokarnik et al. (142). The novelty resides on the design of a fluorogenic substrate, derivative
of cephalosporins (coumarin cephalosporin fluorescein (CCF2)) that is well retained in active
cells. The nonfluorescent, esterified substrate CCF2/AM is relatively nonpolar and can
passively diffuses across the cell membrane. Once inside the cell, the ester groups are
hydrolyzed by nonspecific esterases to release and retain within the cells the substrate for
-lactamase. CCF2 is composed of a fluorescein and coumarin linked by a cephalosporin
bridge, which is cleaved in the presence of -lactamase. If the cells are not expressing the
-lactamase reporter enzyme, the intact CCF2 fluoresces green under ultraviolet excitation, as
a result of the fluorescence resonance energy transfer (FRET) between coumarin and the
green-emitting fluorescein. In the presence of -lactamase, substrate cleavage allows the
coumarin to emit blue fluorescence while the fluorescein is quenched. Using the ratio of
intensities at the two wavelengths (blue/green) allows more accurate signal quantification
since it improves signal-to-noise due to cell size variation. The -lactamase reporter system
(CCF2) was demonstrated to exhibit higher sensitivity and specificity than the -galactosidase
FDG, and it is well adapted for flow cytometric analysis (27, 70). Although this method was
designed to measure gene expression in mammalian cells, the principle of FRET and trapping
the substrate would also be applicable in analysis of microbial gene expression. The method
could be used without modification for the rapid analysis of cells expressing native
-lactamase activity and for screening for inhibitors (137).
32
Chapter 2
2.5.3 Identification of microorganisms
Flow cytometry offers numerous possibilities for identification and enumeration of
microorganisms in their natural habitat. Hence it provides an extensive toolbox for
microbiologists to detect, isolate and enumerate microorganisms of interest in natural samples
in an accurate and rapid way.
Nucleic acid dyes: non-specific detection
A large number of fluorescent dyes with high affinity binding to DNA and /or RNA are
used to discriminate the target cells from the background. The combination of scatter
parameters and the nucleic acid dye fluorescence is becoming the method of choice to
discriminate target cells from the background sample and is widely employed to detect and
enumerate bacteria in their natural environments (17, 20, 51, 65, 74, 109). A FCM method for
direct detection of the anaerobic bacteria in human feces and biopsies was described using PI for
discrimination the fecal cells and excluding large particles by FSC and SSC (128, 143, 144).
The range of fluorescent stains, with different spectral characteristics and high
quantum yield, appropriate for flow cytometry analysis is continuously expanding. One
further development in this area is the availability of some commercial kits well suited for
rapid detecting and counting of total and viable bacterial cells (53). The BacLight viability kit
has been used to monitor the viability of microorganisms in different environmental settings
such as in drinking water (20), seawater (45) and food products such as cheese (26) or milk
(51). Gram staining kits for unfixed and viable microorganisms have been developed (53) and
have recently been applied to study the viability of bacteria in sewage water (41). Using
hexidium iodide (HI), and SYTO13, Mason et al. (87) could correctly predict the Gram status
of 45 strains of clinically relevant organisms, including several known to be Gram variable.
Recently, a FCM-based Gram staining protocol was used to monitor milk contamined with
Staphylococcus aureus and E. coli demonstrating the suitability of this technique for detecting
bacteria in a food matrix by culture-independent means (55).
Antibodies: Immunodetection
Flow cytometry in conjunction with fluorescent antibodies has been used to
detect surface antigens in a number of bacteria including Salmonella (33, 88), Legionella (61)
and E. coli (138). Antibodies can be made fluorescent by covalently attaching them to
fluorescent organic compounds such as fluorescein isothiocyanate (FITC),
tetramethylrhodamine isothiocyanate (TRITC), texa red, phycobiliproteins, the cyanine dyes
such as indocarbocyanine dyes (Cy3, Cy5 and Cy7) (113) and the ALEXA dyes
(53). Recently, Barbesti el al. (12) developed a multi-color assay where a Cy5-labeled
monoclonal antibody, SYBR Green I and PI were used simultaneously to identify
immunolfluorescent viable cells of E. coli and B. subtilis. The threecolor assay allowed the
33
Flow Cytometry

simultaneous identification and viability assessment in a single assay. A rapid and novel FCM
assay combining immunomagnetic separation and a fluorescent stained bacteriophage was used
to identify E. coli O157:H7 by FCM with a detection limit of 104 bacteria per ml (47).
Furthermore, the conventional enzyme-linked immunoadsorbant assay (ELISA) is being
replaced by the so-called multiplexed immunoassays because of its high sensitivity and high
throughput capability (62). More recently, a 15-multiplexed-flow cytometric immunoassay was
developed for serotyping Streptococcus pneumonia isolates; the assay correctly identified the 86
S. pneumoniae with 100% sensitivity and specificity (100). Immunofluorescence techniques are
not only limited to surface antigens but can also be readily applied to intracellular proteins
provided that the cells are permeabilized. Koch et al. (71) reported on an immunofluorescence
flow cytometric approach to investigate the response of Pseudomonas fluorescens to carbon
limitation in soil by the analysis of the expression of -galactosidase from the
s
-dependent
E. coli promoter to carbon limitation in soil.
Oligonucleotides and fluorescent in situ hybridization (FISH)
During the last decade, hybridizations with rRNA-targeted probes have provided a
unique insight into the structure and spatio-temporal dynamics of complex microbial
communities (4). Nucleic acid probes can be designed to specifically target taxonomic groups at
different levels of specificity (from species to domain) by virtue of variable evolutionary
conservation of the ribosomal rRNA molecules. Appropriate software-environments such as the
ARB software package (82) and availability of large databases (http://rdp.cme.msu.edu/html/)
or the online resource for oligonucleotide probes ProbeBase (80) offer a powerful platform for
rapid probe design and in silico specificity profiling. The increasing availability of small subunit
(16S rRNA) rRNA sequences, estimates to contain over 100,000 sequences
(http://rdp.cme.msu.edu/html/analyses preview.html) has contributed significantly to the
development of the method. At present, a considerable number of validated domain-, division,
genus- and species-specific probes are available for the identification of the a large number of
microorganisms within their natural habitat (5, 80). FISH is applied to morphologically intact
cells and thus provides a quantitative measure of the target organism without the limitation of
culture-dependent methods. Recently, significant methodological improvements of the FISH
technique have been reported, which overcome many of the limitations due to the low
fluorescent signal discussed by Wagner et al. (130). Using FCM in combination with FISH, it
has been demonstrated for E. coli that the accessibility of probe target sites on the 16S and 23S
molecules differ strongly (42, 43). More recently, it was demonstrated that accessibility patterns
of 16S rRNAs are more similar for phylogenetically related organisms; these findings may be the
first description of consensus probe accessibility maps for prokaryotes (14)
Commonly epifluorescence microscopy is the standard method by which fluorescent-
stained cells are enumerated, however the method is time consuming and subjective (75, 92).
Recently, this technique has been improved by development of automated image acquisition
34
Chapter 2
and analysis software allowing accurate microscopic enumeration of bacterial cells (63, 102).
Alternatively, FCM offers a potential platform for high-resolution, high throughput
identification and sorting of microorganims using fluorescent rRNA-targeted oligonucleotides
(6, 131, 133). A number of studies have shown that fluorescently labeled rRNA probes can
be used to study microbial consortia in situ, even in matrices as difficult as activated sludge
(132), or feces (106, 107, 143, 144) and for targets as difficult as the very small picoplankton
(115). FISH in conjunction with FCM has been used to study the microbial dynamics and
structure of the human fecal microbiota. The results of such studies revealed the suitability of
FCM to study the ecology of uncultured bacteria, amongst others of the human
gastrointestinal tract (143).
Genome typing
Several molecular typing techniques have been developed during the past decade for
the identification and classification of bacteria at the species and strain level. The most
powerful of these are genetic based molecular methods known as DNA-fingerprinting
techniques such as pulsed-field gel electrophoresis (PFGE) of rare-cutting restriction
fragments, ribotyping, randomly amplified polymorphic DNA (RAPD), and amplified
fragment length polymorphism (AFLP).
Restriction fragment polymorphism (RFLP) analysis of bacterial DNA involves the
digestion of genomic DNA with rare-cutting restriction enzymes to yield a few relatively large
fragments. The restriction fragments are then size fractionated using PFGE that allows
separation of large genomic fragments. The generated DNA fingerprint obtained depends on
the specificity of the used restriction enzyme and the sequence of the bacterial genome, and
is therefore characteristic of a particular species or strain of bacteria. This fingerprint
represents the complete genome and thus can detect specific changes (DNA deletion,
insertion or rearrangements) within a particular strain over time.
Recently, a flow cytometry-based technique for bacterial genome fingerprinting
analysis has been described. This new approach combines the reproducibility of RFLP and the
sensitivity of FCM (57, 69, 73). Restriction fragments of different bacterial strains obtained
by rare-cutting restriction endonucleases were stained with either TOTO-1 or PicoGreen dyes
and analyzed by FCM. A histogram of burst sizes from the restriction fragments, linearly
related to fragment length in base pairs, resulted in a DNA fingerprint that could distinguish
among species and strains of E. coli, Staphylococcus aureus and Bacillus globigii. These studies
showed that the DNA fragment sizing analysis by FCM was approximately 100 times faster
and approximately 200,000 times more sensitive than PFGE. When sample preparation time
is included, the total DNA fragment analysis time was approximately 8 h by FCM and
approximately 24 h by PFGE (73).
35
Flow Cytometry

Bead-based microarray: the suspension array
One of the most powerful tools in molecular biology is the DNA microarray
technology, which represents a high throughput multiplexed platform for the detection of
microorganisms in their natural habitat. The unique ability of FCM to make quantitative,
homogeneous, multiparameter measurements has led to the development of a new class of
molecular analysis that uses microspheres as solid support (96). Each element in the array is
comprised of a subpopulation of beads, which has a unique distribution of fluorescent
intensities and each array bead bears a surface receptor. Oligonucleotide capture probes or
antibodies can be coupled to the beads to create an oligonucleotide hybridization assay or
immunoassay assay, respectively. The target PCR products or secondary antibodies are labeled
and serve as the reporter (60, 62, 135). Recently, a bead-based method for multiplexed
detection and quantification of DNA sequences isolated from ground water bacteria was
described (118, 119).
2.5 Conclusions
In this chapter we highlighted the potential applications of FCM in studying the
microbial eco-physiology in different environmental settings. Multiparameter cell sorting
unquestionably adds an extra dimension to the information one gains from the results of the
post-sorting assays. Hence, FCM should be regarded not only as diagnostic tool to analyze
functional or constituent parameters but also a preparative tool for further molecular,
biochemical and omics-based analysis. Despite many advantages that FCM offers to analyze
microorganisms, the main barriers to its general usage have been the expense and complexity
of the equipment and the fact that many of the methodologies and staining techniques have
been designed principally for studies on mammalian cells. However, the continuous
development of fluorescent technology, user-friendly and reasonably priced equipments will
make FCM and FACS a well-established method in microbiological laboratories. Moreover,
with the availability of a large number of microbial genomes, and the omics-based approaches,
FCM will play an important role in studying microbial genetics and eco-physiology at the
single cell level.
2.6 References
1. Allman, R., R. Manchee, and D. Llyod. 1993. Flow cytometric analysis of
heterogeneous bacterial populations, p. 27 - 48. In D. Llyod (ed.), Flow cytometry in
microbiology. Spring-Verlag, New York.
2. Alvarez, A. M., M. Ibanez, and R. Rotger. 1993. Beta-galactosidase activity in bacteria
measured by flow cytometry. Biotechniques 15:974-976.
36
Chapter 2
3. Alvarez-Barrientos, A., J. Arroyo, R. Canton, C. Nombela, and M. Sanchez-Perez. 2000.
Applications of flow cytometry to clinical microbiology. Clin. Microbiol. Rev. 13:167-195.
4. Amann, R., B. M. Fuchs, and S. Behrens. 2001. The identification of microorganisms
by fluorescence in situ hybridisation. Curr. Opin. Biotechnol. 12:231-236.
5. Amann, R., W. Ludwig, and K. Schleifer. 1995. Phylogenetic identification and in situ
detection of individual microbial cells without cultivation. Microbiol. Rev. 59:143-169.
7. Araya, R., K. Tani, T. Takagi, N. Yamaguchi, and M. Nasu. 2003. Bacterial activity
and community composition in stream water and biofilm from an urban river
determined by fluorescent in situ hybridization and DGGE analysis. FEMS Microbiol.
Ecol. 43:111-119.
8. Attfield, P. V., T. Gunasekera, A. E. Boyd, D. Deere, and D. A. Veal. 1999.
Applications of flow cytometry to microbiology of food and beverage industries.
Australias. Biotechnol. 9:159-166.
9. Attfield, P. V., D. Kletsas, A, D. A. Veal, R. van Rooijen, and P. L. Bell. 2000. Use of
flow cytometry to monitor cell damage and predict fermentation activity of dried
yeasts. J. Appl. Microbiol. 89:207-214.
10. Attfield, P. V., H. Y. Choi, D. A. Veal, and P. J. Bell. 2001. Heterogeneity of stress
gene expression and stress resistance among individual cells of Saccharomyces cerevisiae.
Mol. Microbiol. 40:1000-1008.
11. Bailey, J. E., J. Fazel-Madjlessi, D. N. McQuitty, L. Y. Lee, J. C. Allerd, and J. A. Oro.
1977. Characterization of bacterial growth by means of flow cytometry. Science
198:1175-1176.
12. Barbesti, S., S. Citterio, M. Labra, M. D. Baroni, M. G. Neri, and S. Sgorbati. 2000.
Two and three-color fluorescence flow cytometric analysis of immunoidentified viable
bacteria. Cytometry 40:214-218.
13. Barer, M. R., and C. R. Harwood. 1999. Bacterial viability and culturability, p. 94-126.
In R. K. Poole (ed.), Advances in microbial physiology, vol. 41. Academic Press, New York.
14. Behrens, S., B. M. Fuchs, F. Mueller, and R. Amann. 2003. Is the in situ accessibility
of the 16S rRNA of Escherichia coli for Cy3-labeled oligonucleotide probes predicted by
a three-dimensional structure model of the 30S ribosomal subunit. Appl. Environ.
Microbiol. 69:4935-4941.
15. Ben-Amor, K., P. Breeuwer, P. Verbaarschot, F. M. Rombouts, A. D. L. Akkermans,
W. M. De Vos, and T. Abee. 2002. Multiparametric flow cytometry and cell sorting
37
Flow Cytometry

for the assessment of viable, injured, and dead Bifidobacterium cells during bile salt
stress. Appl. Environ. Microbiol. 68:5209-5216.
16. Bernard, L., C. Courties, C. Duperray, H. Schafer, G. Muyzer, and P. Lebaron. 2001.
A new approach to determine the genetic diversity of viable and active bacteria in aquatic
17. Bernard, L., C. Courties, P. Servais, M. Troussellier, and M.L. Petit, P. Lebaron. 2000.
Relationships among bacterial cell size, productivity, and genetic diversity in aquatic
environments using cell sorting and flow cytometry. Microb. Ecol. 40:148-158.
18. Bhupathiraju, V. K., M. Hernandez, D. Landfear, and L. Alvarez-Cohen. 1999.
Application of a tetrazolium dye as an indicator of viability in anaerobic bacteria.
J. Microbiol. Methods. 37:231-243.
19. Bogosian, G., and E. V. Bourneuf. 2001. A matter of bacterial life and death. EMBO
Rep. 2:770-774.
20. Boulos, L., M. Prvost, B. Barbeau, J. Coallier, and R. Desjardins. 1999. LIVE/DEAD
BacLight: application of a new rapid staining method for direct enumeration of viable
and total bacteria in drinking water. J. Microbiol. Methods. 37:77-86.
21. Breeuwer, P., and T. Abee. 2000. Assessment of viability of microorganisms
employing fluorescence techniques. Int. J. Food. Microbiol. 55:193-200.
22. Breeuwer, P., J. Drocourt, F. Rombouts, and T. Abee. 1996. A novel method for
continuous determination of the intracellular pH in bacteria with the internally
conjugated fluorescent probe 5 (and 6-)-carboxyfluorescein succinimidyl ester. Appl.
23. Breeuwer, P., J. L. Drocourt, F. M. Rombouts, and T. Abee. 1994. Energy-dependent,
carrier-mediated extrusion of carboxyfluorescein from Saccharomyces cerevisiae allows rapid
assessment of cell viability by flow cytometry. Appl Environ Microbiol. 60:1467-1472.
24. Bunthof, C. J., and T. Abee. 2002. Development of a flow cytometric method to
analyze subpopulations of bacteria in probiotic products and dairy starters. Appl.
25. Bunthof, C. J., K. Bloemen, P. Breeuwer, F. M. Rombouts, and T. Abee. 2001. Flow
cytometric assessment of viability of lactic acid bacteria. Appl. Environ. Microbiol.
67:2326-2335.
26. Bunthof, C. J., S. van Schalkwijk, W. Meijer, T. Abee, and J. Hugenholtz. 2001.
Fluorescent method for monitoring cheese starter permeabilization and lysis. Appl.
27. Campbell, R. E. 2004. Realization of beta-lactamase as a versatile fluorogenic reporter.
Trends Biotechnol. 22:208-211.
38
Chapter 2
28. Carter, E. A., E. P. Frank, and P. A. Hunter. 1993. Cytometric evaluation of antifungal
agents. In D. Lloyd (ed.), Flow cytometry in microbiology. Springer-Verlag, London,
United Kingdom.
29. Chitarra, L. G., P. Breeuwer, R. W. Van Den Bulk, and T. Abee. 2000. Rapid
fluorescence assessment of intracellular pH as a viability indicator of Clavibacter
michiganensis subsp. michiganensis. J. Appl. Microbiol. 88:809-816.
30. Cho, J. C., and S. J. Kim. 1999. Viable, but non-culturable, state of a green
fluorescence protein-tagged environmental isolate of Salmonella typhi in groundwater
and pond water. FEMS Microbiol. Lett. 170:257-264.
31. Chung, J. D., S. Conner, and G. Stephanopoulos. 1995. Flow cytometric study of
differentiating cultures of Bacillus subtilis. Cytometry 20:324-333.
32. Cid, V. J., A. M. Alvarez, A. I. Santos, C. Nombela, and M. Sanchez. 1994. Yeast exo-
beta-glucanases can be used as efficient and readily detectable reporter genes in
Saccharomyces cerevisiae. Yeast 10:747-756.
33. Clarke, R. G., and A. C. Pinder. 1998. Improved detection of bacteria by flow cytometry
using a combination of antibody and viability markers. J. Appl. Microbiol. 84:577-584.
34. Da Silveira, G. M. 2003. Effect of ethanol on Oenococcus oeni: stress response,
adaptation and performance. PhD thesis. Wageningen University, The Netherlands.
35. Da Silveira, G. M., M. San Romao, M., M. C. Loureiro-Dias, F. M. Rombouts, and
T. Abee. 2002. Flow cytometric assessment of membrane integrity of ethanol-stressed
Oenococcus oeni cells. Appl. Environ. Microbiol. 68:6087-6093.
36. Daugherty, P. S., B. L. Iverson, and G. Georgiou. 2000. Flow cytometric screening of
cell-based libraries. J. Immunol. Methods. 243:211-227.
37. Davey, H. M., C. L. Davey, and D. Kell. 1993. On the determination of the size of
microbial cells using flow cytometry, p. 49-65. In D. Lloyd (ed.), Flow cytometry in
microbiology. Springer-Verlag, London, United Kingdom.
38. Davey, H. M., A. Jones, A. D. Shaw, and D. B. Kell. 1999. Variable selection and
multivariate methods for the identification of microorganisms by flow cytometry.
Cytometry 35:162-168.
39. Davey, H. M., and D. B. Kell. 1996. Flow cytometry and cell sorting of heterogeneous
microbial populations: the importance of single-cell analysis. Microbiol. Rev. 60:641-696.
40. Deere, D., J. Shen, G. Vesey, P. Bell, P. Bissinger, and D. Veal. 1998. Flow cytometry
and cell sorting for yeast viability assessment and cell selection. Yeast 14:147-160.
41. Forster, S., J. R. Snape, H. M. Lappin-Scott, and J. Porter. 2002. Simultaneous
fluorescent Gram staining and activity assessment of activated sludge bacteria. Appl.
39
Flow Cytometry

42. Fuchs, B. M., K. Syutsubo, W. Ludwig, and R. Amann. 2001. In situ accessibility of
Escherichia coli 23S rRNA to fluorescently labeled oligonucleotide probes. Appl.
43. Fuchs, B. M., G. Wallner, W. Beisker, I. Schwippl, W. Ludwig, and R. Amann. 1998.
Flow cytometric analysis of the in situ accessibility of Escherichia coli 16S rRNA for
fluorescently labeled oligonucleotide probes. Appl. Environ. Microbiol. 64:4973-4982.
44. Fuller, M. E., S. H. Streger, R. K. Rothmel, B. J. Mailloux, J. A. Hall, T. C. Onstott,
J. K. Fredrickson, D. L. Balkwill, and M. F. DeFlaun. 2000. Development of a vital
fluorescent staining method for monitoring bacterial transport in subsurface
environments. Appl. Environ. Microbiol. 66:4486-4496.
45. Gasol, J. M., U. L. Zweifel, F. Peters, J. A. Fuhrman, and A. Hagstrom. 1999.
Significance of size and nucleic acid content heterogeneity as measured by flow
cytometry in natural planktonic bacteria. Appl. Environ. Microbiol. 65:4475-4483.
46. Geoffroy, M. C., C. Guyard, B. Quatannens, S. Pavan, M. Lange, and A. Mercenier.
2000. Use of green fluorescent protein to tag lactic acid bacterium strains under
development as live vaccine vectors. Appl. Environ. Microbiol. 66:383-391.
47. Goodridge, L., J. Chen, and M. Griffiths. 1999. Development and characterization
of a fluorescent-bacteriophage assay for detection of Escherichia coli O157:H7. Appl.
48. Graves, S. W., R. C. Habbersett, and J. P. Nolan. 2001. A dynamic inline sample
thermoregulation unit for flow cytometry. Cytometry 43:23-30.
49. Gregori, G., S. Citterio, A. Ghiani, M. Labra, S. Sgorbati, S. Brown, and M. Denis.
2001. Resolution of viable and membrane-compromised bacteria in freshwater and
marine waters based on analytical flow cytometry and nucleic acid double staining.
50. Gunasekera, T. S., A. Sorensen, P. V. Attfield, S. J. Sorensen, and D. A. Veal. 2002.
Inducible gene expression by nonculturable bacteria in milk after pasteurization. Appl.
51. Gunasekera, T. S., D. A. Veal, and P. V. Attfield. 2003. Potential for broad
applications of flow cytometry and fluorescence techniques in microbiological and
somatic cell analysis of milk. Int. J. Food Microbiol. 85:269-279.
52. Harry, E., K. Pogliano, and R. Losick. 1995. Use of immunofluorescence to visualize
cell-specific gene expression during sporulation in Bacillus subtilis. J. Bacteriol.
177:3386-3393.
53. Haugland, R.P. 2002. Molecular Probes - Handbook of fluorescent probes and
research products, 9th ed. www.probes.com.
40
Chapter 2
54. Hewitt, C. J., and G. Nebe-Von-Caron. 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.
55. Holm, C., and L. Jespersen. 2003. A flow-cytometric Gram-staining technique for
milk-associated bacteria. Appl. Environ. Microbiol. 69:2857-2863.
56. Howlett, N. G., and S. V. Avery. 1999. Flow cytometric investigation of
heterogeneous copper-sensitivity in asynchronously grown Saccharomyces cerevisiae.
FEMS Microbiol. Lett. 176:379-386.
57. Huang, Z., J. H. Jett, and R. A. Keller. 1999. Bacteria genome fingerprinting by flow
cytometry. Cytometry 35:169-175.
58. Humphreys, M. J., R. Allman, and D. Lloyd. 1994. Determination of the viability of
Trichomonas vaginalis using flow cytometry. Cytometry 15:343-348.
59. Hutter, K. J., and H. E. Eipel. 1978. Flow cytometric determinations of cellular substances
in algae, bacteria, moulds and yeasts. Van Leeuwenhoek J. Microbiol. 44:269-282.
60. Iannone, M. A., J. D. Taylor, J. Chen, M. S. Li, F. Ye, and M. P. Weiner. 2003.
Microsphere-based single nucleotide polymorphism genotyping. Methods Mol. Biol.
226:123-134.
61. Ingram, M., T, T. Cleary, J, B. Price, J, and A. Castro. 1982. Rapid detection of
Legionella pneumophila by flow cytometry. Cytometry 3:134-147.
62. Jani, I. V., G. Janossy, D. W. Brown, and F. Mandy. 2002. Multiplexed immunoassays
by flow cytometry for diagnosis and surveillance of infectious diseases in resource-poor
settings. Lancet Infect. Dis. 2:243-250.
63. Jansen, G. J., A. C. M. Wildeboer-Veloo, R. H. J. Tonk, A. H. Franks, and G. W.
Welling. 1999. Development and validation of an automated, microscopy-based method
for enumeration of groups of intestinal bacteria. J. Microbiol. Methods. 37:215-221.
64. Jepras, R. I., F. E. Paul, S. C. Pearson, and M. J. Wilkinson. 1997. Rapid assessment
of antibiotic effects of Escherichia coli by bis-(1,3-dibutylbarbituric acid) trimethine
oxonol and flow cytometry. Antimicrob. Agents Chemother. 41:2001-2005.
65. Joux, F., and P. Lebaron. 2000. Use of fluorescent probes to assess physiological
functions of bacteria at single-cell level. Microb. Infect. 2:1523-1535.
66. Kaprelyants, A., and D. Kell. 1993. Dormancy in stationary-phase cultures of
Micrococcus luteus: flow cytometric analysis of starvation and resuscitation. Appl.
67. Kaprelyants, A., G. Mukamolova, H. Davey, and D. Kell. 1996. Quantitative analysis
of the physiological heterogeneity within starved cultures of Micrococcus luteus by flow
cytometry and cell sorting. Appl. Environ. Microbiol. 62:1311-1316.
41
Flow Cytometry

68. Kell, D. B., A. S. Kaprelyants, D. H. Weichart, C. R. Harwood, and M. R. Barer.
1998. Viability and activity in readily culturable bacteria: a review and discussion of
practical issues. Antonie van Leeuwenhoek 73:169-187.
69. Kim, Y., J. H. Jett, E. J. Larson, J. R. Penttila, B. L. Marrone, and R. A. Keller. 1999.
Bacterial fingerprinting by flow cytometry: bacterial species discrimination.
70. Knapp, T., E. Hare, L. Feng, G. Zlokarnik, and P. Negulescu 2003. Detection of
beta-lactamase reporter gene expression by flow cytometry. Cytometry 51A:68-78.
71. Koch, B., J. Worm, L. E. Jensen, O. Hojberg, and O. Nybroe. 2001. Carbon
limitation induces sigma(S)-dependent gene expression in Pseudomonas fluorescens in
soil. Appl. Environ. Microbiol. 67:3363-3370.
72. Kuckuck, F. W., B. S. Edwards, and L. A. Sklar. 2001. High throughput flow
cytometry. Cytometry 44:83-90.
73. Larson, E. J., J. R. Hakovirta, H. Cai, J. H. Jett, S. Burde, R. A. Keller, and B. L.
Marrone. 2000. Rapid DNA fingerprinting of pathogens by flow cytometry.
74. Lebaron, P., N. Parthuisot, and P. Catala. 1998. Comparison of blue nucleic acid dyes
for flow cytometric enumeration of bacteria in aquatic systems. Appl. Environ.
Microbiol. 64:1725-1730.
75. Lipski, A., U. Friedrich, and K. Altendorf. 2001. Application of rRNA-targeted
oligonucleotide probes in biotechnology. Appl. Microbiol. Biotechnol. 56:40-57.
76. Lloyd, D. 1993. Flow cytometry in microbiology. Spring-Verlag, London, United
Kingdom.
77. Lopez-Amoros, R., S. Castel, J. Comas-Riu, and J. Vives-Rego. 1997. Assessment of
E. coli and Salmonella viability and starvation by confocal laser microscopy and flow
cytometry using rhodamine 123, DiBAC4(3), propidium iodide, and CTC. Cytometry
29:298-305.
78. Lopez-Amoros, R., J. Comas, and J. Vives-Rego. 1995. Flow cytometric assessment of
Escherichia coli and Salmonella typhimurium starvation-survival in seawater using
rhodamine 123, propidium iodide, and oxonol. Appl. Environ. Microbiol. 61:2521-2526.
79. Lowder, M., A. Unge, N. Maraha, J. K. Jansson, J. Swiggett, and J. D. Oliver. 2000.
Effect of starvation and the viable-but-nonculturable state on green fluorescent protein
(GFP) fluorescence in GFP-tagged Pseudomonas fluorescens A506. Appl. Environ.
Microbiol. 66:3160-3165.
80. Loy, A., M. Horn, and M. Wagner. 2003. ProbeBase: an online resource for rRNA-
targeted oligonucleotide probes. Nucleic Acids Res. 31:514-516.
42
Chapter 2
81. Ludovico, P., F. Sansonetty, and M. Corte-Real. 2001. Assessment of mitochondrial
membrane potential in yeast cell populations by flow cytometry. Microbiol. 147:3335-
3343.
82. Ludwig, W., O. Strunk, R. Westram, L. Richter, H. Meier, Yadhukumar, A. Buchner,
T. Lai, S. Steppi, G. Jobb, W. Forster, I. Brettske, S. Gerber, A. W. Ginhart, O.
Gross, S. Grumann, S. Hermann, R. Jost, A. Konig, T. Liss, R. Lussmann, M. May,
B. Nonhoff, B. Reichel, R. Strehlow, A. Stamatakis, N. Stuckmann, A. Vilbig, M.
Lenke, T. Ludwig, A. Bode, and K. H. Schleifer. 2004. ARB: a software environment
for sequence data. Nucleic Acids Res. 32:1363-1371.
83. Luppens, S. B. 2002. Suspensions or biofilms and other factors that affect disinfectant
testing on pathogens. PhD Thesis. Wageningen University.
84. Luppens, S. B., B. Barbaras, P. Breeuwer, F. M. Rombouts, and T. Abee. 2003.
Selection of fluorescent probes for flow cytometric viability assessment of Listeria
monocytogenes exposed to membrane-active and oxidizing disinfectants. J. Food. Prot.
66:1393-1401.
85. Mason, D. J., R. Allman, J. M. Stark, and D. Lloyd. 1994. Rapid estimation of
bacterial antibiotic susceptibility with flow cytometry. J. Microsc. 176:8-16.
86. Mason, D. J., R. Lopez-Amoros, R. Allman, J. M. Stark, and D. Lloyd. 1995. The
ability of membrane potential dyes and calcafluor white to distinguish between viable
and non-viable bacteria. J. Appl. Bacteriol. 78:309-315.
87. Mason, D. J., S. Shanmuganathan, F. C. Mortimer, and V. A. Gant. 1998. A
fluorescent Gram stain for flow cytometry and epifluorescence microscopy. Appl.
88. McClelland, R., G, and A. Pinder. 1994. Detection of Salmonella typhimurium in
dairy products with flow cytometry and monoclonal antibodies. Appl. Environ.
Microbiol. 60:4255-4262.
89. McSharry, J. J. 1994. Uses of flow cytometry in virology. Clin. Microbiol. Rev. 7:576-
604.
90. Mixter, P. F., J. D. Klena, G. A. Flom, A. M. Siegesmund, and M. E. Konkel. 2003.
In vivo tracking of Campylobacter jejuni by using a novel recombinant expressing green
fluorescent protein. Appl. Environ. Microbiol. 69:2864-2874.
91. Mortimer, F. C., D. J. Mason, and V. A. Gant. 2000. Flow cytometric monitoring of
antibiotic-induced injury in Escherichia coli using cell-impermeant fluorescent probes.
Antimicrob. Agents. Chemother. 44:676-681.
92. Moter, A., and U. B. Gobel. 2000. Fluorescence in situ hybridization (FISH) for
direct visualization of microorganisms. J. Microbiol. Methods. 41:85-112.
43
Flow Cytometry

93. Nebe-von-Caron, G., P. J. Stephens, C. J. Hewitt, J. R. Powell, and R. A. Badley.
2000. Analysis of bacterial function by multi-colour fluorescence flow cytometry and
single cell sorting. J. Microbiol. Methods. 42:97-114.
94. Nielsen, J. L., M. Aquino de Muro, and P. H. Nielsen. 2003. Evaluation of the redox
dye 5-cyano-2, 3-tolyl-tetrazolium chloride for activity studies by simultaneous use of
microautoradiography and fluorescence in situ hybridization. Appl. Environ.
95. Nir, R., Y. Yisraeli, R. Lamed, and E. Sahar. 1990. Flow cytometry sorting of viable
bacteria and yeasts according to beta-galactosidase activity. Appl. Environ. Microbiol.
56:3861-3866.
96. Nolan, J. P., and L. A. Sklar. 2002. Suspension array technology: evolution of the flat-
array paradigm. Trends Biotechnol. 20:9-12.
97. Novo, D. J., N. G. Perlmutter, R. H. Hunt, and H. M. Shapiro. 2000. Multiparameter
flow cytometric analysis of antibiotic effects on membrane potential, membrane
permeability, and bacterial counts of Staphylococcus aureus and Micrococcus luteus.
Antimicrob. Agents. Chemother. 44:827-834.
98. Olsen, M. J., D. Stephens, D. Griffiths, P. Daugherty, G. Georgiou, and B. L.
Iverson. 2000. Function-based isolation of novel enzymes from a large library. Nat.
99. Paau, A. S., J. R. Cowles, and J. Oro. 1977. Flow micro-fluoremetric analysis of
Escherichia coli, Rhizobium meliloti and Rhizobium japonicum at different stages of the
growth cycle. Can. J. Microbiol. 23:1165-1169.
100. Park, M. K., D. E. Briles, and M. H. Nahm. 2000. A latex bead-based flow
cytometric immunoassay capable of simultaneous typing of multiple pneumococcal
serotypes (multibead assay). Clin. Diagn. Lab. Immunol. 7:486-489.
101. Patkar, A., N. Vijayasankaran, D. W. Urry, and F. Srienc. 2002. Flow cytometry as a
useful tool for process development: rapid evaluation of expression systems.
J. Biotechnol. 93:217-229.
102. Pernthaler, J., A. Pernthaler, and R. Amann. 2003. Automated enumeration of
groups of marine picoplankton after fluorescence in situ hybridization. Appl. Environ.
Microbiol. 69:2631-2637.
103. Plovins, A., A. Alvarez, M. Ibanez, M. Molina, and C. Nombela. 1994. Use of
fluorescein-di-beta-D-galactopyranoside (FDG) and C12-FDG as substrates for beta-
galactosidase detection by flow cytometry in animal, bacterial, and yeast cells. Appl.
104. Porter, J., D. Deere, M. Hardman, C. Edwards, and R. Pickup. 1997. Go with the flow -
use of flow cytometry in environmental microbiology. FEMS Microbiol. Ecol. 24:93-101.
44
Chapter 2
105. Porter, J. P., and C. Edwards. 1994. The use of fluorogenic esters to detect viable bacteria
by flow cytometry. J. Appl. Bacteriol. 77:221-228.
Fluorescent hybridisation combined with flow cytometry and hybridisation of total RNA
to analyse the composition of microbial communities in human faeces using 16S rRNA
probes. FEMS Microbiol. Ecol. 43:237-245.
107. Rigottier-Gois, L., V. Rochet, N. Garrec, A. Suau, and J. Dore. 2003. Enumeration of
Bacteroides species in human faeces by fluorescent in situ hybridisation combined with flow
cytometry using 16S rRNA probes. Syst. Appl. Microbiol. 26:110-118.
108. Rodriguez, G., D. Phipps, K. Ishiguro, and H. Ridgway. 1992. Use of a fluorescent
redox probe for direct visualization of actively respiring bacteria. Appl. Environ.
Microbiol. 58:1801-1808.
109. Roth, B., M. Poot, S. Yue, and P. Millard. 1997. Bacterial viability and antibiotic
susceptibility testing with SYTOX green nucleic acid stain. Appl. Environ. Microbiol.
63:2421-2431.
110. Russo-Marie, F., M. Roederer, B. Sager, L. Herzenberg, and D. Kaiser. 1993. (Beta)-
galactosidase activity in single differentiating bacterial cells. Proc. Natl. Acad. Sci. USA.
90:8194-8198.
111. Schaule, G., H. Flemming, and H. Ridgway. 1993. Use of 5-cyano-2,3-ditolyl tetrazolium
chloride for quantifying planktonic and sessile respiring bacteria in drinking water. Appl.
112. Shapiro, H. M. 2000. Microbial analysis at the single-cell level: tasks and techniques. J.
Microbiol. Methods. 42:3-16.
113. Shapiro, H. M. 1995. Practical Flow Cytometry, 3d ed. Wiley-Liss Inc, New York.
114. Sieracki, M. E., T. L. Cucci, and J. Nicinski. 1999. Flow cytometric analysis of 5-cyano-
2,3-ditolyl tetrazolium chloride activity of marine bacterioplankton in dilution cultures.
115. Simon, N., N. LeBot, D. Marie, F. Partensky, and D. Vaulot. 1995. Fluorescent in situ
hybridization with rRNA-targeted oligonucleotide probes to identify small phytoplankton
by flow cytometry. Appl. Environ. Microbiol. 61:2506-2513.
116. Sincock, S. A., and J. P. Robinson. 2001. Flow cytometric analysis of microorganisms.
Methods Cell Biol. 64:511-537.
117. Snaidr, J., B. Fuchs, G. Wallner, M. Wagner, K.-H. Schleifer, and R. Amann. 1999.
Phylogeny and in situ identification of a morphologically conspicuous bacterium,
Candidatus Magnospira bakii, present at very low frequency in activated sludge. Environ.
118. Spiro, A., and M. Lowe. 2002. Quantitation of DNA sequences in environmental PCR
products by a multiplexed, bead-based method. Appl. Environ. Microbiol. 68:1010-1013.
45
Flow Cytometry

119. Spiro, A., M. Lowe, and D. Brown. 2000. A bead-based method for multiplexed
identification and quantitation of DNA sequences using flow cytometry. Appl. Environ.
Microbiol. 66:4258-4265.
120. Steen, H. B. 1980. Further developments of a microscopic-based flow cytometry: light
scatter detection and excitation intensity compensation. Cytometry 1:26-31.
121. Steen, H. B., and T. Lindm 1979. Flow cytometry a high resolution for everyone. Science
204.
122. Suller, M. T., and D. Lloyd. 1998. Flow cytometric assessment of the postantibiotic effect
of methicillin on Staphylococcus aureus. Antimicrob. Agents. Chemother. 42:1195-1199.
123. Suller, M. T., and D. Lloyd. 1999. Fluorescence monitoring of antibiotic-induced bacterial
damage using flow cytometry. Cytometry 35:235-241.
124. Tombolini, R., A. Unge, M. E. Davey, F. J. de Bruijn, and J. K. Jansson. 1997. Flow
cytometric and microscopic analysis of GFP-tagged Pseudomonas fluorescens bacteria.
FEMS Microbiol. Ecol. 22:17-28.
125. Tsien, R. Y. 1998. The green fluorescent protein. Annu. Rev. Biochem. 67:509-544.
126. Valdivia, R. H., and S. Falkow. 1997. Fluorescence-based isolation of bacterial genes
expressed within host cells. Science 277:2007-2011.
127. van der Waaij, L., P. Limburg, Mesander G, and D. van der Waaij. 1996. In vivo IgA
coating of anaerobic bacteria in human faeces. Gut 38:348-354.
128. van der Waaij, L. A., G. Mesander, P. C. Limburg, and D. van der Waaij. 1994. Direct
flow cytometry of anaerobic bacteria in human feces. Cytometry 16:270-279.
129. Vives-Regoa, J., P. Lebaron, and G. Nebe-von Caron. 2000. Current and future
applications of flow cytometry in aquatic microbiology. FEMS Microbiol. Rev. 24:429-
448.
130. Wagner, M., M. Horn, and H. Daims. 2003. Fluorescence in situ hybridisation for the
identification and characterisation of prokaryotes. Curr. Opin. Microbiol. 6:302-9.
131. Wallner, G., Amann R, and W. Beisker. 1993. Optimizing fluorescent in situ hybridization
with rRNA-targeted oligonucleotide probes for flow cytometric identification of
microorganisms. Cytometry 14:136-143.
132. Wallner, G., R. Erhart, and R. Amann. 1995. Flow cytometric analysis of activated sludge
with rRNA-targeted probes. Appl. Environ. Microbiol. 61:1859-1866.
134. Wickens, H. J., R. J. Pinney, D. J. Mason, and V. A. Gant. 2000. Flow cytometric
investigation of filamentation, membrane patency, and membrane potential in Escherichia
coli following ciprofloxacin exposure. Antimicrob. Agents. Chemother. 44:682-687.
46
Chapter 2
135. Wilkins, S. P., M. R. Henry, and D. M. Kelso. 1999. DNA hybridization on
microparticles: determining capture-probe density and equilibrium dissociation constants.
Nucleic Acids Res. 27:1719-1727.
136. Winding, A., S. Binnerup, and J. Sorensen. 1994. Viability of indigenous soil bacteria
assayed by respiratory activity and growth. Appl. Environ. Microbiol. 60:2869-2875.
137. Winson, M. K., and H. M. Davey. 2000. Flow cytometric analysis of microorganisms.
Methods 21:231-240.
138. Yamaguchi, N., M. Sasada, M. Yamanaka, and M. Nasu. 2003. Rapid detection of
respiring Escherichia coli O157:H7 in apple juice, milk, and ground beef by flow cytometry
Cytometry 54A:27-35.
139. Zahavy, E., M. Fisher, A. Bromberg, and U. Olshevsky. 2003. Detection of frequency
resonance energy transfer pair on double-labeled microsphere and Bacillus anthracis spores
by flow cytometry. Appl. Environ. Microbiol. 69:2330-2339.
140. Zhang, J., E. C. Robert, A. Y. Ting , and R. Y. Tsien. 2002. Creating new fluorescent
probes for cell biology. Nat. Rev. Mol. Cell Biol. 3:906 -918.
141. Ziglio, G., G. Andreottola, S. Barbesti, G. Boschetti, L. Bruni, P. Foladori, and R. Villa.
2002. Assessment of activated sludge viability with flow cytometry. Water. Res. 36:460-
468.
142. Zlokarnik, G., P. A. Negulescu, T. E. Knapp, L. Mere, N. Burres, L. Feng, M. Whitney,
K. Roemer, and R. Y. Tsien. 1998. Quantitation of transcription and clonal selection of
single living cells with beta-lactamase as reporter. Science 279:84-88.
145. Zubkov, M. V., B. M. Fuchs, H. Eilers, P. H. Burkill, and R. Amann. 1999.
determination of total protein content of bacterial Cells by SYPRO staining and flow
cytometry. Appl. Environ. Microbiol. 65:3251-3257.
47
Flow Cytometry

.
Chapter 3
QUANTIFICATION OF UNCULTURED
RUMINOCCUS OBEUM-LIKE
BACTERIA IN HUMAN FECAL SAMPLES WITH
FLUORESCENT IN SITU HYBRIDIZATION
AND FLOW CYTOMETRY USING 16S
RIBOSOMAL RNA TARGETED PROBES
Erwin G. Zoetendal, Kaouther Ben-Amor, Hermie J. M. Harmsen, Frits Schut, Antoon D. L.
Akkermans, and Willem M. de Vos
Abstract: A 16S rRNA-targeted probe was designed and validated in order to quantify
the number of uncultured Ruminococcus obeum-like bacteria using fluorescent in situ
hybridization (FISH). These bacteria have frequently been found in 16S rDNA clone
libraries from bacterial communities in the human intestine. Thirty-two reference
strains from the human intestine, including a phylogenetically related and some other
Ruminococcus species were used as negative control and did not hybridize with the
newly designed probe. Microscopic and flow cytometric analysis revealed that a group
of morphologically similar bacteria in feces did hybridize with this probe. Moreover,
it was found that all hybridizing cells also hybridized with a probe specific for the
Clostridium coccoidesEubacterium rectale group, a group that included the uncultured
R. obeum-like bacteria. Quantification of the uncultured R. obeum-like bacteria
and the Clostridium coccoidesEubacterium rectale group by flow cytometry and
microscopy revealed that these groups comprised approximately 2.5% and 16% of
the total community in fecal samples, respectively. The uncultured R. obeum-like
bacteria comprise about 16% of Clostridium coccoides Eubacterium rectale group.
These results indicate that the uncultured R. obeum-like bacteria are numerically
important in human feces. Statistical analysis revealed no significant difference
between the microscopic and flow cytometric counts and the sampling time of the
feces while a significant host-specific effect on the counts was observed. Our data
demonstrates that the combination of FISH and flow cytometry is a useful approach
for studying the ecology of uncultured bacteria in the human gastrointestinal tract.
Applied and Environmental Microbiology 68:4225-4232 (2002)
Reprinted with permission of the American Society for Microbiology
49
3.1 Introduction
The human gastrointestinal (GI) tract harbors a diverse microbial community which has
an important metabolic and protective function in the GI tract (reviewed by 18). Recent studies
indicate that interactions between the host and the bacterial community are of considerable
importance, but very complex and just starting to be understood (4, 9, 10, 25). Most knowledge
about the bacterial community in the human GI tract has been obtained by selective cultivation
of microbes from fecal samples. The past 5 years, culture-independent approaches using the
sequence variability of the 16S rRNA genes have demonstrated that most of the predominant
bacteria in human fecal samples have not yet been obtained in culture, illustrating the limitation
in our knowledge of these predominant members (16, 23, 24). In addition, denaturing and
temperature gradient gel electrophoretic (DGGE/TGGE) analysis of fecal 16S rDNA and rRNA
amplicons has demonstrated to be powerful culture-independent approaches in determining and
monitoring the bacterial community in feces (24, 25). Such studies revealed that the
predominant bacterial community in human feces is relatively stable in time, host-specific, and
not significantly altered following consumption of certain probiotic strains (19, 20, 24).
Although the application of 16S rDNA-directed DGGE, cloning, and sequencing has
provided new insights into the bacterial composition of the human GI tract, these approaches
are all based on the use of PCR amplification methods and hence cannot be accurately converted
to real numbers of bacteria. Fluorescent in situ hybridization (FISH) using 16S rRNA targeted
oligonucleotide probes has demonstrated to be very powerful in detecting and quantifying
uncultured bacteria in environmental samples (for a review see reference 3). FISH analysis of
fecal populations has demonstrated that these were found to be relatively stable in time (5).
Various genera, like Bacteroides, Bifidobacterium, Streptococcus, Lactobacillus, Collinsella,
Eubacterium, and Clostridium could be quantified accurately in feces using the FISH approach
(5, 6, 7, 13). In addition, a group of Fusobacterium prausnitzii-like bacteria, which was
predominantly found in several fecal clone libraries (16, 23, 24), was also found as a
predominant member using the FISH approach (5, 17).
In several studies, a group of closely related uncultured bacteria that have Ruminococcus
obeum as closest cultivable relative was found regularly in fecal 16S rDNA clone libraries (16,
24) (Fig 1). These were also found to be present in fecal TGGE profiles of dominant and active
bacteria and could be an important member of the bacterial community in the human GI tract
(24). Since these closely related cloned amplicons all have less than 97% sequence similarity to
R. obeum they represent a group of hitherto unknown species. Up to now no quantitative data
have been obtained and the morphology of these bacteria remains unknown, since their presence
is only derived from PCR-based data.
50
Chapter 3
In many studies, FISH analysis is combined with microscopic analysis. Although less
frequently used, also the combination of FISH with flow cytometry has been applied successfully in
analyzing different microbial communities (2, 14, 21, 22). Major advantages of this combination
are the multiparametric analysis of samples, and the relatively fast and sensitive quantification of
populations, even those that make up only about 1% of the total community.
In the present study, we have developed a set of two probes to detect and quantify the
group of uncultured Ruminococcus obeum-like bacteria using FISH in combination with direct
microscopy and flow cytometry. After validation, the probe was used to quantify the number of
hybridized cells in fecal samples from different individuals by microscopic and flow cytometric
analysis. In addition, the Eubacterium rectaleClostridium coccoides group, that includes the
uncultured R. obeum-like bacteria, and the total number of bacteria were quantified. Variations
over time, among individuals and between the approaches were statistically analyzed.
Figure 1: Rooted neighbor-joining tree showing the phylogenetic relationships between the 16S rDNA
sequences of the uncultured Ruminococcus obeum-like bacteria (indicated in bold) and some closely and
distantly related bacteria. The phylogenetic tree was generated from the tree-of-life of the ARB software
package. The Bacteroides group was used as outgroup. Accession numbers of the 16S rDNA sequences are
indicated. The bar represents 10% sequence divergence. The target sequences of the Urobe63 probe and
Erec482 probe are indicated.
51
FISH detection of uncultured R. obeum-like bacteria in feces
3.2 Materials and Methods
Design and validation of the oligonucleotide probes.
Sequences of closely related uncultured R. obeum-like bacteria from human fecal
samples were obtained from two studies in which fecal 16S rDNA clone libraries were
constructed and analyzed (Fig. 1) (16, 24). The identity and accession number of these
sequences are fecal clone A14 (AF052415), fecal clone A20 (AF052417), fecal clone A57
(AF052422), adhufec171 (AF132243), and adhufec35.25 (AF153853). These sequences were
aligned using the ARB software package (15) and probes targeting these sequences were
designed. The newly designed probe was screened for specificity based on comparative analysis
of 16S rRNA sequences from the ARB database and ribosomal database project (RDP, 11) and
newly deposited sequences from GenBank using the ARB and RDP software (Table. 1). The
CHECK-PROBE analysis function of the RDP and the Probe Match function of ARB were
used to screen the 16S and 23S rRNA sequence databases for target sequences for the newly
developed probe.
Thirty-two reference strains (Table 2) from various phylogenetic groups in the GI tract,
including R. obeum, were used as negative control to evaluate the specificity of the newly
designed probe. This probe encoded as Urobe63 (the term Urobe is an abreviation for
Uncultured Ruminococcus obeum-like bacteria) is a 1:1 combination of the oligonucleotide
probes S-*-Urobe-0063-a-A-20a and S-*-Urobe-0063-b-A-20a (Table 1). The Eub338 probe
(2) was used as positive control for the hybridization of the negative reference strains. Since the
positive target of the Urobe63 probe constitutes a 16S rDNA sequence of an uncultured group
of R. obeum-like bacteria, no positive reference strain could be used. This means that the probe
specificity could only be determined by FISH analysis using bacteria from a reference fecal
sample as positive control. Therefore, bacteria from a reference fecal sample were added to the
reference strains in order to verify hybridization signals of the Urobe63 probe. The optimal
hybridization condition was determined by performing hybridizations at a constant
temperature of 50C and gradually increasing the formamide concentration in the
hybridization buffer and adjusting the salt concentration in the corresponding washing buffer
as described previously (12). In addition, fecal samples were hybridized with the Urobe63
probe and the Erec482 probe (5), which targets the Clostridium coccoides - Eubcaterium rectale
group that includes the uncultured R. obeum-like bacteria (Fig. 1), to determine whether the
Urobe63-positive cells appeared double-labeled. Furthermore, competitive experiments with a
probe which targets only the cultured species R. obeum (5 AAT GAA ARG TTT CCC GTT
CG) were performed to verify if cross reaction occurred with the non-target organism having
the least mismatches with the Urobe63 probe.
52
Chapter 3
Table 1. Alignments of probe sequences and their target sequences having maximal two mismatches
a
.
Sequence ID Sequence
S-*-Urobe-0063-a-A-20
b
3 GC- TTG- CCC- TTA- ATG- AAA- TAA
S-*-Urobe-0063-b-A-20
b
3 GC- TTG- CCC- TTT- ATR- AAR- TAA
Target 5 CG AAC GGG AAW UAY UUY AUU
Fecal clones A14, A20, A57 .. ... ... ... ... ... ...
Adhufec35.25 .. ... ... ... ... ... ...
Adhufec171 NN NNN NNN NN. ... ... ...
Ruminococcus obeum .. ... ... ... CY. ... ...
Rumen clone RFN27 .. ... ... ... ... ... U.A
a
N, R, W, and Y are the International Union of Pure and Applied Chemistry codes for ambiguous bases.
Dots indicate bases that are identical to bases in the target site. Differences between the probes are indicated
by boldface types. Mismatches of target sequences with the probe target site are underlined.
b
According to the ODP nomenclature (1). The Urobee63 probe is a 1:1 mixture of S-*-Urobe-0063-a-A-
20 and S-*-Urobe-0063-b-A-20.
Reference strains, culture conditions, and fixation
The thirty-two reference strains used in this study were obtained from various sources as
indicated: Deutsche Sammlung von Mikroorganismen und Zellcultures [DSM; Braunschweig,
Germany], American Type Culture Collection [ATCC; Rockville, Md], and Laboratory for
Medical Microbiology [MMB; Groningen, The Netherlands (5)] (Table 2). The ATCC and
DSM strains were cultivated as described in the respective catalogues. All MMB stains are
clinical or human fecal isolates from local and regional public health laboratories that were
identified and cultivated by using standard procedures (8). Exponentially grown cells were
harvested at 5,000 X g for 10 min, washed with 0.2-m-pore-size-filtered phosphate-buffered
saline (PBS; per liter: 8 g NaCl, 0.2 g KCl, 1.44 g Na2HPO4, and 0.24 g KH2PO4; pH7.2), and
diluted 1:3 with 4% (wt/vol) paraformaldehyde in PBS. After fixation at 4C for 16 h cells were
stored in 50% ethanol-PBS until FISH analysis (2).
53

Table 2. Reference strains used to validate the probe hybridization conditions.
Strain Origin
a
Strain Origin
a
Bacteroides fragilis (DSM 2151) Clostridium putrificum (DSM 1734)
Bacteroides distasonis (DSM 20701) Clostridium sporogenes (DSM 795)
Bacteroides ovatus (MMB) Collinsella aerofaciens (DSM 13713)
Bacteroides thetaiotaomicron (MMB) Eubacterium cylindroides (MMB)
Bacteroides uniformis (MMB) Eubacterium moniliforme (MMB)
Bacteroides vulgatus (DSM 1447) Eubacterium tenue (DSM 20695)
Bifidobacterium dentium (ATCC 27678) Eubacterium ventriosum (DSM 3988)
Bifidobacterium infantis (ATCC 15697) Lachnospira multipara (DSM 3073)
Bifidobacterium longum (MMB) Peptostreptococcus micros (DSM 20468)
Clostridium beijerinckii (MMB) Ruminococcus albus (ATCC 27210)
Clostridium butyricum (MMB) Ruminococcus bromii (ATCC 27255)
Clostridium carnis (DSM 1293) Ruminococcus callidus (ATCC 27760)
Clostridium innocuum (MMB) Ruminococcus obeum (ATCC 29174)
Clostridium nexile (MMB) Streptococcus intermedius (DSM 20573)
Clostridium perfingens (MMB) Succiniclasticum ruminis (DSM 9236)
Clostridium polysaccharolyticum (DSM 1801) Veillonella parvula (DSM 20373)
a
DSM, Deutsche Sammlung von Mikroorganismen und Zellkulturen, Braunschweig, Germany; MMB,
Laboratory for Medical Microbiology, Groningen, The Nethelands; ATCC, American Type Culture
Collection, Rockville, Md.
Fecal sample processing
Fecal samples were collected from 3 healthy Dutch male volunteers (25-32 years) and
were processed within 30 min. These volunteers provided three samples within a 4 weeks
period and had not been subjected to any feeding trial, specific diet, or antibiotic treatment
for the last 3 years. Additional fecal samples were collected from six volunteers (2 men and 4
women, age 25 40 years) who originate from different countries (Netherlands, Germany,
54
Chapter 3
China, Portugal, Italy, and Tunisia) but were living for at least 3 months in the Netherlands.
Fecal samples were processed as described previously (5). In short, 0.5 g of fecal sample was
resuspended in 4.5 ml PBS and vortexed with addition of 5-10 glass beads for 5 min to
homogenize the sample. After centrifuging at 700 X g for 1 min, 1 ml of supernatant was
added to 3 ml of 4% paraformaldehyde in PBS and stored overnight at 4C. After washing
twice with PBS the fixed cells were stored in 50% Ethanol-PBS at 20 C until further use
(for at least 1 h). Weighted portions of the remains of the fecal samples were lyophilized to
determine their dry weights.
FISH analysis of fecal samples by microscopy
For microscopic analysis, fixed cells were spotted on gelatin-coated glass slides and
dried for 20 min at 45C. Dilution series of fecal samples were performed in order to
determine the optimal cell concentration for counting using the different probes. After drying
of the slides, the cells were dehydrated for 2 to 3 min in a graded ethanol series with
concentrations increasing from 50% to 75% and finally 96% ethanol/H2O. Ten microliters of
hybridization buffer (0.9 M NaCl, 20 mM Tris-HCl (pH 7.5), 0.1% (wt/vol) SDS)
containing 3 ng/l of Cy3-labeled Urobe63 probe or 5 ng/l FITC-labeled Erec482 probe was
added to each well, followed by incubation at 50C for 3 h. After hybridization the slides were
washed in 50 ml hybridization buffer without SDS for 10 to 20 min. For the Urobe63 probe
a 20% formamide (vol/vol) containing hybridization buffer and a low salt washing buffer
(0.225 M NaCl, 20 mM Tris-HCl (pH 7.5), 10 mM EDTA) was used. For total counts 4,6-
diamidino-2-phenylindole (DAPI) was added to the wash buffer at a final concentration of
100 ng/ml. After rinsing the slides in water they were immediately air-dried and mounted in
Vectashield (Vector Labs, Burlingame, CA). Digital images of the slides, viewed with a Leica
(Wetzlar, Germany) DMRXA epifluorescence microscope, were taken with a Kodak Megaplus
1.4 charge-coupled device camera. These images were analyzed and fluorescent cells were
counted by using Quantimet HR550 image analysis software (Leica). For each analysis 25
microscopic fields were counted.
FISH analysis of fecal samples by flow cytometry
For each hybridization, 50 l of fixed cells were centrifuged for 3 min at 9,000 X g and
resuspended in 20 l of hybridization buffer (0.9 M NaCl, 20 mM Tris-HCl (pH8.0), 0.1%
(wt/vol) SDS). After addition of 2 l of the Cy5- and/or double (at 5 and 3 end) FITC-
labeled probes (30 ng/l and 50 ng/l respectively) the samples were incubated for 16 hours
at 50C in the dark. The Urobe63 and Erec482 probes were both Cy5-labeled for counting.
An FITC-labeled Erec482 probe was used to validate the specificity of the Urobe63 probe.
After hybridization, 980 l of prewarmed washing buffer was added and the samples were
incubated at 50C for 5 min. The hybridization buffer without SDS was used as washing
buffer after hybridization with the Erec482 probe, and the low salt washing buffer consisting
55

of 0.225 M NaCl, 20 mM Tris-HCl (pH8.0), 10 mM EDTA was used after hybridization with
the Urobe63 probe. The cells were centrifuged at 9,000 X g for 5 min and resuspended in 1 ml
of ice-cold PBS (pH 8.4). To avoid losing the signal intensity the hybridized cells were kept in the
dark on ice until flow cytometric analysis. Cells hybridized with the labeled probes were compared
to cells, which were incubated in hybridization buffer only, in order to optimize the settings for
flow cytometric analysis. Afterwards, the unlabeled cells were incubated with 10 l/ml of
propidium iodide (PI; 1 mg/ml) at 37C for 20 min to count the total number of bacteria.
Different cell concentrations of each sample were analyzed to check for potential clumps. The
bacterial concentration was adjusted to keep the count lower than 1,000 events/sec in order to
avoid coincidence. During analysis of fecal samples, 0.7 m-yellow green fluorescent beads (for
counting PI-labeled cells) or PC fluorescent beads (for counting Cy5-labeled cells) with known
concentration (Polysciences, Inc) were added according to the manufacturer's instructions in
order to determine cell numbers. Samples were analyzed using a Becton and Dickinson
FACScalibur flow cytometer. An air-cooled argon ion laser (488 nm) and a red-diode laser (635
nm) were used for excitation and the green, red, and far red signals of the bacteria and the beads
were collected in the FL1 (515 to 545 nm), FL3 (>600 nm long pass filter) and FL4 (653 to 669
nm) detectors, respectively. The system threshold was set on forward scatter signals and all
bacterial analyses were performed at the low flow rate settings (12 l/min). Data were collected
in list mode as pulse height signals (four decades in logarithmic scale each) and 10,000 cells were
acquired for further analysis which was performed using CellQuest software (Beckton Dickinson)
and/or WinMDI version 2.8 software (http//:facs.Scripps.edu/software.html). The whole
hybridization and counting analysis was performed three times for each probe and fecal sample.
Statistical analysis
The analyses performed on the various samples included the variables of methodology
(microscopy or flow cytometry), probe type (DAPI, Erec482 probe, or Urobe63 probe),
individual, and time. Therefore, the coefficient of variation (i.e. the standard variation divided
by the mean, termed CV) for the time, for the reproducibility of the counts, and for the
different individuals were compared. In addition, regression analysis on the mean counts of
fecal samples taken at different time intervals was performed for each individual. Since it
appeared that time had no significant effect on the counts, these counts were used as replicates
in a three-factor-ANOVA-test to statistically analyze the effect of the other variables.
3.3 Results
Probe design and validation
Probes targeting the uncultured R. obeum-like bacteria were designed based on
comparative analysis of 16S rRNA sequences from the ARB and RDP database and newly
deposited sequences from GenBank. Despite the small sequence variation in the hyper-variable
V1 region between the different sequences of this uncultured group, a combination of two
56
Chapter 3
probes was sufficient to discriminate between this uncultured group and the rest of the
sequences in all databases (Table 1). These two probes hybridize at the same target site and were
used in a a 1:1 mixture during all Urobe63 probe hybridizations. Analysis of the 16S rRNA
sequence databases showed that the Urobe63 probe has only two mismatches with the targets
of the R. obeum sequence and a sequence of a 16S rDNA clone from a rumen sample (Table
1). All other sequences have three or more mismatches with the probe target site. No match
was found with the adhufec171 sequence since the first non-discriminative bases of the
Urobe63 probe target were not present in this sequence (16). Furthermore, no match was
found with any of the 23S rRNA sequences. Thirty-two reference strains (Table 2), including
R. obeum, did not hybridize with the Urobe63 probe, while all of them showed strong
hybridization with the Eub338 probe as determined by microscopic and flow cytometric
analysis. For the microscopic analysis, hybridization buffer containing 20% formamide was
chosen since a few cells (~1% of Eub338 positive cells) of the Peptostreptococcus micros culture
and about 10-20% of Eub338 positive R. obeum cells hybridized with the Urobe63 probe when
no formamide was added. For flow cytometric analysis Cy5-labeled probes, which give
emission in the far red, were used since no adverse autofluorescence was observed in this
detector channel (FL4) when fecal samples or reference strains were analyzed. However,
autofluorescence of unlabeled fecal samples was observed in all other detectors (FL1, FL2,
FL3), which detect green, yellow, and red fluorescent signals, respectively (data not shown). It
appeared that the addition of formamide in the hybridization buffer could not be used during
flow cytometric analysis when Cy5-labeled probes were used. No hybridization signals could be
detected when formamide was added even when the reference strains were hybridized with the
Eub338 probe. However, washing of Urobe63 probe-hybridized cells with the low-salt washing
buffer was sufficient to render all negative reference strains unlabeled while maintaining
specific hybridization in fecal samples. Only the cells of R. obeum showed some weak positive
hybridization with the Urobe63 probe under these conditions, although the signal was more
than ten-fold lower than positively hybridized cells in a fecal sample.
Since no isolate is available to serve as a positive control strain for the Urobe63 probe, a
double hybridization of fecal samples was performed using the Urobe63 and Erec482 probes.
This latter probe targets the Clostridium coccoides - Eubacterium rectale group to which the
uncultured group of R. obeum-like bacteria belongs. Microscopic analysis revealed that all cells
that hybridized with the Urobe63 probe also hybridized with the Erec482 probe with similar
signal intensities (Fig. 2), indicating that Urobe63 probe hybridized the right target group. All
Urobe63-positive cells were morphologically similar in all fecal samples examined and appeared
as a relatively large single or duplococcus (Fig. 2A), while the Erec482-positive cells exhibited
many different morphotypes (Fig. 2B). The intensity of the Urobe63 probe and Erec482 probe
signals varied between different cells, which could be due to the difference in ribosome content
between the cells. As expected a variety of morphotypes was detected by using DAPI stain and
phase contrast microscopic analysis (Fig. 2C, D). Flow cytometric analysis revealed similar results
as obtained with microscopic examination (Fig. 3). Although the Erec482 probe was double
57

FITC-labeled, it could be used in combination with the Cy5-labeled Urobe63 probe, since the
Urobe63-positive cells did not show autofluorescence in the FL1 detector, which detects the
FITC-labeled cells. The Cy5-labeled Urobe63-positive cells were also detected as Erec482-positive
cells and could clearly be discriminated from unlabeled cells (Figs. 3A and 3B). When these
double labeled cells were gated, they appeared as a very small cluster in the flow cytometric dot
plots of the forward-angle light scatters (FSC) versus side-angle light scatters (SSC), indicating
that these cells exhibited limited morphological variability (Fig 3C).
Figure 2: Photographs of fecal cells in one microscopic field hybridized with the Cy3-labeled Urobe63
probe (A), the FITC-labeled Erec482 probe (B), and DAPI (C). (D) Phase-contrast photograph is
represented. Bar = 5 m.
To ensure that the right target bacteria are detected, cross hybridization experiments were
performed using the Urobe63 probe and a probe targeting R. obeum, which showed the least
mismatches with the Urobe63 probe. No cross hybridizations were found when using a pure
culture of R. obeum, fecal samples, or mixtures of R. obeum and fecal samples. These and the
double labeling results indicate that the Urobe63 probe specifically detects the uncultured
R. obeum-like bacteria.
58
Chapter 3
Figure 3. Flow cytometric analysis of fecal samples showing the dot plots representing the Urobe63-Cy5
fluorescence intensity and Erec482-FITC fluorescence intensity of (A) unhybridized bacteria and (B)
Urobe63 plus Erec482 hybridized bacteria in feces. Cells labeled with both probes (as indicated by the
arrow in plot B) were marked with a dark color and appeared as a small cluster in dot plots of the forward-
angle light scatters (FSC) versus side-angle light scatters (SSC) of the total community (C), indicating
limited morphological variability among these bacteria.
Comparison between microscopic and flow cytometric analysis
Microscopic and flow cytometric counts were determined and compared. About 60%
of the DAPI stained cells hybridized with the Eub338 probe as observed by microscopy, while
this was about 80% when Eub338 and PI positive cells were compared using flow cytometry.
This difference is probably due to dividing cells which containing two or more chromosomes.
These cells will be counted by microscopy as two cells when they are DAPI stained, while they
will only be counted as a single cell when hybridized with the Eub338 probe. Using flow
cytometry, a cell containing two or more chromosomes will be counted as a single cell.
To visualize the difference between the two approaches the ratio of the mean counts
for each probe per sample was determined to determine the total number of bacteria, the
number of uncultured R. obeum bacteria and the Clostridium coccoides - Eubacterium rectale
group in feces (Table 3). The ratios varied between 0.64 and 1.47 and only one relatively high
ratio (1.93), which was found for an Urobe63 count. The mean percentage of the Clostridium
coccoides Eubacterium rectale group varied between 16.9% and 13.5% as determined by
microscopy and flow cytometry, respectively. Similarly, the percentage of the uncultured R.
obeum group was found to be 2.6% and 2.2% of the total community, and 15.6% and 16.7%
of the Clostridium coccoides Eubacterium rectale group. This indicated that both groups were
numerically important in fecal samples and that the uncultured R. obeum-like bacteria
comprised a significant fraction of the Clostridium coccoides Eubacterium rectale group.
59
60
Chapter 3
T
a
b
l
e

3
.

D
A
P
I

a
n
d

P
I

t
o
t
a
l

c
e
l
l

c
o
u
n
t
s
,

E
r
e
c
4
8
2

p
r
o
b
e

h
y
b
r
i
d
i
z
a
t
i
o
n

c
o
u
n
t
s
,

a
n
d

U
r
o
b
e
6
3

p
r
o
b
e

h
y
b
r
i
d
i
z
a
t
i
o
n

c
o
u
n
t
s

o
f

f
e
c
a
l

s
a
m
p
l
e
s

f
r
o
m

t
h
r
e
e

i
n
d
i
v
i
d
u
a
l
s

(
A
,
B
,

C
)

t
a
k
e
n

o
v
e
r

a

4
-
w
e
e
k
s

p
e
r
i
o
d

a
s

d
e
t
e
r
m
i
n
e
d

b
y

m
i
c
r
o
s
c
o
p
y

(
M
i
c
)

a
n
d

f
l
o
w

c
y
t
o
m
e
t
r
y

(
F
C
M
)
.

D
A
P
I

o
r

P
I

c
o
u
n
t
s
E
r
e
c
4
8
2

c
o
u
n
t
s
U
r
o
b
e
6
3

c
o
u
n
t
s
I
n
d
i
v
i
d
u
a
l
T
i
m
e
(
w
e
e
k
s
)
M
i
c

(
1
0
1
1
)
a
F
C
M

(
1
0
1
1
)
a
R
a
t
i
o
b
M
i
c

(
1
0
1
0
)
a
F
C
M

(
1
0
1
0
)
a
R
a
t
i
o
b
M
i
c

(
1
0
1
0
)
a
F
C
M

(
1
0
1
0
)
a
R
a
t
i
o
b
A
0
1
.
8
7

0
.
1
7
2
.
4
0

0
.
5
4
0
.
7
8
3
.
9
5

0
.
6
3
2
.
8
0

1
.
1
6
1
.
4
1
1
.
0
6

0
.
1
7
0
.
5
5

0
.
3
6
1
.
9
3
2
3
.
2
7

0
.
4
0
4
.
9
8

0
.
5
2
0
.
6
6
4
.
3
2

0
.
7
7
5
.
6
9

0
.
7
3
0
.
7
6
1
.
0
6

0
.
3
0
1
.
3
3

0
.
3
1
0
.
7
9
4
4
.
0
6

0
.
4
7
5
.
2
0

2
.
6
0
0
.
7
8
4
.
9
0

1
.
2
2
5
.
1
6

0
.
7
5
0
.
9
5
1
.
0
6

0
.
2
5
0
.
7
2

0
.
1
2
1
.
4
7
B
0
3
.
5
5

0
.
4
1
4
.
7
2

0
.
7
1
0
.
7
5
8
.
3
5

1
.
0
0
7
.
1
6

1
.
7
6
1
.
1
7
0
.
5
6

0
.
0
7
0
.
8
8

0
.
3
0
0
.
6
4
2
3
.
4
2

0
.
3
7
3
.
3
0

0
.
5
9
1
.
0
4
5
.
9
3

0
.
6
9
5
.
1
5

1
.
4
5
1
.
1
5
0
.
8
8

0
.
2
0
1
.
3
1

0
.
3
7
0
.
6
7
4
3
.
2
4

0
.
3
3
3
.
1
5

0
.
9
5
1
.
0
3
8
.
3
0

1
.
9
0
4
.
9
8

1
.
5
1
1
.
6
7
0
.
7
1

0
.
1
7
0
.
7
5

0
.
1
2
0
.
9
4
C
0
2
.
2
7

0
.
4
1
2
.
0
6

0
.
6
4
1
.
1
0
3
.
5
5

0
.
5
8
4
.
4
5

1
.
9
1
0
.
8
0
0
.
7
1

0
.
2
5
0
.
5
7

0
.
1
9
1
.
2
5
2
2
.
4
3

0
.
2
5
2
.
4
1

0
.
9
7
1
.
0
0
2
.
7
5

0
.
4
6
2
.
9
4

0
.
9
1
0
.
9
4
0
.
3
8

0
.
0
6
0
.
3
1

0
.
0
1
1
.
2
2
4
2
.
2
3

0
.
4
5
2
.
2
4

1
.
5
0
1
.
0
0
2
.
4
9

0
.
6
2
2
.
9
6

1
.
0
0
0
.
9
3
0
.
5
4

0
.
1
5
0
.
4
1

0
.
1
6
1
.
3
0
a
T
h
e

v
a
l
u
e
s

a
r
e

m
e
a
n
s

s
t
a
n
d
a
r
d

d
e
v
i
a
t
i
o
n
s

a
n
d

a
r
e

e
x
p
r
e
s
s
e
d

i
n

c
e
l
l
s

p
e
r

g
r
a
m

d
r
y

w
e
i
g
h
t

o
f

f
e
c
e
s
.
b
T
h
e

R
a
t
i
o

w
a
s

c
a
l
c
u
l
a
t
e
d

b
y

d
i
v
i
d
i
n
g

t
h
e

m
e
a
n

m
i
c
r
o
s
c
o
p
i
c

c
o
u
n
t

b
y

t
h
e

m
e
a
n

f
l
o
w

c
y
t
o
m
e
t
r
i
c

c
o
u
n
t
.

Table 4: Three-factor analysis of variance test of the counts obtained by microscopy and flow cytometry
a
Source df Mean Square F ratio P value
M 1 2.61 X 10
21
1.03 0.316
P 2 5.05 X 10
23
199.82 <0.001
b
I 2 1.73 X 10
22
6.85 0.003
b
M X P 2 3.44 X 10
21
1.36 0.269
M X I 2 1.93 X 10
21
0.76 0.474
P X I 4 9.85 X 10
21
3.90 0.010
b
M X P X I 4 1.96 X 10
21
0.77 0.549
Error 36 2.53 X 10
21
- -
a
The degree of freedom, the mean square, the F ratio, and the p value of the variables, methodology
(microscopy and flow cytometry), probe (P) (DAPI or PI, Erec482, and Urobe63 probe), and individual (I)
are shown.
b
The variable had a significant effect (p < 0.05)
Regression analysis was performed on the mean of the total and specific counts separately
as determined by microscopy and flow cytometry from the three time samples taken from each
individual. None of the eighteen analyses resulted in a significant relationship between time and
counts (P>0.05). In addition, the CV for time was found to be comparable to the CV for
reproducibility of the counts (data not shown). In only half of the counts the CV for time was
higher than the CV for reproducibility. These results indicated that the bacterial composition is
relatively stable over time. Since no significant differences between the time samples were found
within each individual, these counts were used as replicates in a three-factor-ANOVA-test in
order to determine the effect of the other variables (Table 4). As may have been expected, a
significant difference in the counts was found when DAPI (or PI), the Erec482 probe or the
Urobe63 probe was used (Table 4). In addition, a significant effect of the individual on the
different counts was observed. Furthermore, a significant effect on the interaction between the
different probes and individuals was observed, which indicated that the different probe counts
varied significantly per individual. These individual differences were most clearly when the CV
among individuals was calculated for each probe at each sampling time. In 17 out of the 18
cases, the CV among individuals was higher than the CV for the assay (data not shown).
Furthermore, it was observed that the different counting approaches did not have a significant
influence on the outcome of the counts.
61

To determine how common the uncultured R. obeum-like bacteria are in different
individuals, the number of these bacteria was determined in feces from six additional individuals
(adults of both sexes) by microscopic analysis. The percentages of the uncultured R. obeum-like
bacteria varied between 1 and 5% of the total bacteria, which is similar to the number in feces
from the previous individuals.
3.4 Discussion
This study describes the development, validation and application of the Urobe63 probe
to detect, examine, and quantify uncultured R. obeum-like bacteria in human fecal samples. We
used a microscopic and flow cytometric approach in order to characterize the specificity of the
probe, the morphology of the cells, and the number of cells in different fecal samples. All
reference strains used as negative control in this study did not hybridize with the Urobe63 probe
and all Urobe63 probe-labeled fecal cells also hybridized with the Erec482 probe that is specific
for the Clostridium coccoides Eubacterium rectale group (5). These results indicate that the right
target group was detected. While many different morphotypes of bacteria hybridized with the
Erec482 probe, only one morphotype hybridized with the newly developed Urobe63 probe in
all fecal samples. The Urobe63 probe did not hybridize with the type strain R. obeum, which has
of all known sequences the least (only two) mismatches at the target site. In addition, no cross
hybridizations were observed when Urobe63 probe was combined with a probe specific for R.
obeum. The Urobe63 probe was designed and validated using currently available 16S rRNA
sequences including several hundreds of 16S rRNA sequences derived from cultured and
uncultured GI tract bacteria. Since various R. obeum-like sequences have been found frequently
in fecal clone libraries from different human individuals (16, 24), we conclude from the present
data that the probe can specifically detect a new and numerically important group of intestinal
bacteria.
microscopic analysis and flow cytometric analysis of FISH samples different protocols
were required due to the difference in handling procedures. Despite these different approaches
our data showed that the counts obtained for the Urobe63 and the Erec482 probes with either
method were similar (Table 3). This indicates that both approaches will give similar results when
quantifying the number of specific bacteria in environmental samples. In addition, the 16S
rRNA of 60 to 80% of the bacterial cells in feces was accessible as determined by comparing
Eub338 probe and DAPI counts, which is in line with previous observations (5, 19). In general,
about 16% of the total bacterial community in the fecal samples belonged the Clostridium
coccoides Eubacterium rectale group, which is slightly lower than reported by Franks and
colleagues (5), but similar to numbers found by Tannock and colleagues (19). This difference in
cell numbers found in a variety of fecal samples may be due to host-specific factors as observed
previously (24, 25). The percentage of the uncultured R. obeum-like bacteria in the investigated
samples was in average about 2.5% and varied between 1 and 6%. This indicates that they
comprise a significant fraction of the predominant bacterial community in feces as observed
62
Chapter 3
previously using PCR-based approaches (16, 24). Based on statistical analysis, we could observe
significant differences between the individuals, which are in line with previous observations
based on PCR and TGGE analysis of fecal samples (24). However, the number of individuals
used in this statistical test is too small for detailed comparisons.
The results described in this study reveal that a group of uncultured R. obeum-like
bacteria, whose presence was only suggested repeatedly by 16S rDNA cloning studies of fecal
samples, comprises a significant fraction of the fecal community as determined by FISH analysis
using a newly developed specific 16S rRNA targeted probe. Although the power of the 16S
rRNA probe hybridization has been demonstrated in many independent studies, its definitive
value in studying the ecology of uncultured bacteria requires additional systematic research. The
application of FISH combined with flow cytometry was demonstrated and offers a possibility to
sort the R. obeum-like bacteria and other uncultured populations in order to study them in
detail. Such studies may ultimately be helpful in refining our knowledge about the ecology of
the microorganisms in the human GI tract.
3.5 Acknowledgments
We thank the volunteers that provided fecal samples for this study. We thank Chantal
Doeswijk for lyophilizing the fecal samples and Patrick Verbaarschot for technical assistance
during the flow cytometric analysis. Dr. Arjan de Visser was very helpful in the statistical analysis
of the data. In addition, we thank Dr. Michael Wagner and co-workers (TU Munich) for their
initial help and advice on FISH techniques.
3.6 References
1. Alm E. W., D. B. Oerther, N. Larsen, D. A. Stahl, and L. Raskin. 1996. The
oligonucleotide probe database. Appl. Environ. Microbiol. 62:3557-3559.
2. Amann, R. I., B. J. Binder, R. J. Olsen, S. W. Chrisholm, R. Devereux, and D. A.
Stahl. 1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow
cytometry for analyzing mixed populations. Appl. Environ. Microbiol. 56:1919-1925.
3. Amann, R. I., W. Ludwig, and K-H. Schleifer. 1995. Phylogenetic identification and in
situ detection of individual cells without cultivation. Microbiol. Rev. 59:143-169.
4. Bry, L., P. G. Falk, T. Midtvedt, and J. I. Gordon. 1996. A model of host-microbial
interactions in an open mammalian ecosystem. Science 273:1381-1383.
5. Franks, A. H. , H. J. M. Harmsen, G. C. Raangs, G. J. Jansen, F. Schut, and G. W.
63

6. Harmsen, H. J. M., P. Elfferich, F. Schut, and G. W. Welling. 1999. A 16S rRNA-
targeted probe for the detection of lactobacilli and enterococci in faecal samples by
fluorescent in situ hybridization. Microbial Ecol. Health Dis. 11:3-12.
group and the Atopobium cluster and their application for enumeration of
Microbiol. 66:4523-4527.
8. Holdeman, L. V., E. P. Cato, and W. E. C. Moore. 1977. Anaerobe laboratory manual,
4
th
ed. Virginia Polytechnic Institute and State University, Blacksburg.
9. Hooper L. V., J. Xu, P. G. Falk, T. Midtvedt, and J. I. Gordon. 1999. A molecular
sensor that allows a gut commensal to control its nutrient foundation in a competitive
ecosystem. Proc. Natl. Acad. Sci. USA 96:9833-9838.
10. Hooper, L. V., M. H. Wong, A. Thelin, L. Hansson. P. G. Falk, and J. I. Gordon. 2000.
Molecular analysis of host-microbial relationships in the intestine. Science 291:881-884.
11. Maidak, B. L., J. R. Cole, T. G. Lilburn, C. T. Parker Jr, P. R. Saxman, R. J. Farris, G.
M. Garrity, G. J. Olsen, T. M. Schmidt, and J. M. Tiedje. 2001. The RDP-II
(Ribosomal Database Project). Nucleic Acids Res. 29:173-174.
12. Manz, W., R. Amann, W. Ludwig, M. Wagner, and K.-H. Schleifer. 1992. Phylogenetic
oligodesoxynucleotide probes for the major subclasses of proteobacteria: problems and
solutions. Syst. Appl. Microbiol. 15:593-600.
13. Schwiertz, A., G. Le Blay, and M. Blaut. 2000. Quantification of different Eubacterium
spp. in Human fecal samples with species-specific 16S rRNA-targeted oligonucleotide
14. Simon N., N. LeBot, D. Marie, F. Parentsky, and D. Vaulot. 1995. Fluorescent in situ
hybridization with rRNA-targeted oligonucleotide probes to identify small
phytoplankton by flow cytometry. Appl. Environ. Microbiol. 61:2506-2513.
15. Strunk, O., and W. Ludwig. 1995. ARB - a software environment for sequence data.
Department of Microbiology, Technical University of Munich, Munich, Germany.
(http://www.mikro.biologie.tu.Muenchen.de/pub/ARB/documentation/arb.ps)
16. Suau A., R. Bonnet, M. Sutren, J-J. Godon, G. R. Gibson, M. D. Collins, and J. Dor.
1999. Direct analysis of genes encoding 16S rRNA from complex communities reveals
many novel molecular species within the human gut. Appl. Environ. Microbiol.
65:4799-4807.
64
Chapter 3
17. Suau A., V. Rochet, A. Sghir, G. Gramet, S. Brewaeys, M. Sutren, L. Rigottier-Gois,
and J. Dor. 2001. Fusobacterium prausnitzii and related species represent a dominant
group within the human fecal flora. Sys. Appl. Microbiol. 24:139-145.
18. Tannock G. W. 1995. Normal microflora. An introduction to microbes inhabiting the
human body. Chapman and Hall, London.
19. Tannock, G. W., K. Munro, H. J. M. Harmsen, G. W. Welling, J. Smart, and P. K.
Gopal. 2000. Analysis of the fecal microflora of human subjects consuming a probiotic
product containing Lactobacillus rhamnosus DR20. Appl. Environ. Microbiol. 66:2578-
2588.
20. Vaughan E. E., G. H. J. Heilig, E. G. Zoetendal, R. Satokari, J. K. Collins, A. D. L.
Akkermans, and W. M. de Vos. 1999. Molecular approaches to study probiotic bacteria.
Trends Food Sci. Technol. 10: 400-404.
23. Wilson, K. H., and R. H. Blitchington. 1996 Human colonic biota studied by
ribosomal DNA sequence analysis. Appl. Environ. Microbiol. 62:2273-2278.
24. Zoetendal, E. G., A. D. L. Akkermans, and W. M. de Vos. 1998. Temperature gradient
gel electrophoresis analysis from human fecal samples reveals stable and host-specific
communities of active bacteria. Appl. Environ. Microbiol. 64:3854-3859.
25. Zoetendal E. G., A. D. L. Akkermans, W. M. Akkermans van-Vliet, J. A. G. M. de
the human gastrointestinal tract. Microbial Ecol. Health Dis. 13:129-134.
65

.
Chapter 4
MUCOSA- ASSOCIATED BACTERIA IN THE
HUMAN GASTROINTESTINAL TRACE ARE
UNIFORMLY DISTRIBUTED ALONG THE COLON
AND DIFFER FROM THE COMMUNITY
RECOVERED FROM FECES
Erwin G. Zoetendal, Atte von Wright, Terttu Vilpponen-Salmela, Kaouther Ben-Amor, Antoon
D. L. Akkermans, and Willem M. de Vos
Abstract: The human gastrointestinal (GI) tract harbors a complex community of
bacterial cells in the mucosa, lumen, and feces. Since most attention has been
focused on bacteria present in feces, knowledge about the mucosa-associated
bacterial communities in different parts of the colon is limited. In this study, the
bacterial communities in feces and biopsy samples from the ascending, transverse,
and descending colons of 10 individuals were analyzed by using a 16S rRNA
approach. Flow cytometric analysis indicated that 10
5
to 10
6
bacteria were present
in the biopsy samples (~0,5 mg). To visualize the diversity of the predominant and
the Lactobacillus group community, denaturing gradient gel electrophoresis
(DGGE) analysis of 16S rRNA gene amplicons was performed. DGGE analysis
and similarity index comparisons demonstrated that the predominant mucosa-
associated bacterial community was host specific and uniformly distributed along
the colon but significantly different from the fecal community (P < 0.01). The
Lactobacillus group-specific profiles were less complex than the profiles reflecting
the predominant community. For 6 of the 10 individuals the community of
Lactobacillus-like bacteria in the biopsy samples was similar to that in the feces.
Amplicons having 99% sequence similarity to the 16S ribosomal DNA of
Lactobacillus gasseri were detected in the biopsy samples of nine individuals. No
significant differences were observed between healthy and diseased individuals. The
observed host-specific DGGE profiles of the mucosa-associated bacterial
community in the colon support the hypothesis that host-related factors are
involved in the determination of the GI tract microbial community.
Applied and Environmental Microbiology 68:3401-3407 (2002)
Reprinted with permission of the American Society for Microbiology
67
4.1 Introduction
The human gastrointestinal (GI) tract harbors a diverse community of obligate and
facultative anaerobic bacteria. These bacteria have an important metabolic and protective
function in the GI tract (23). The complex interactions between the host and the bacterial
community are of considerable importance but are just starting to be understood (3, 9, 10). Most
of the knowledge about bacterial diversity in the human GI tract has been obtained by selective
cultivation of microbes from fecal samples. Recently, culture-independent approaches using the
sequence variability of the 16S rRNA genes have shown that most of the predominant bacteria in
human fecal samples have not been obtained in culture yet, which indicates that our knowledge
of these predominant members is very limited (22, 26, 28). In addition, denaturing gradient gel
electrophoresis (DGGE) and temperature gradient gel electrophoresis (TGGE) analyses of fecal
16S ribosomal DNA (rDNA) and rRNA amplicons have shown to be powerful approaches in
determining and monitoring the bacterial community in feces (28, 29). Such studies revealed that
the predominant bacterial community in mammalian feces is stable in time, host specific, affected
by ageing, and not altered after consumption of certain probiotic strains (11, 21, 24, 25, 28, 29).
Furthermore, DGGE has been used to compare bacterial communities in fecal samples from
infants with and without necrotizing enterocolitis, although no differences associated with this
disease were observed (17).
Our present knowledge of the bacterial diversity associated with the human GI tract is
based mainly on analysis of fecal samples, and in a few cases samples that originated from different
parts of the intestine have been characterized. Most of these analyses with contents from sudden-
death victims (15) or with biopsy samples from living individuals involved a culturing approach
and focused on the attachment of certain probiotic strains (1, 2, 12), the presence of sulfate
reducers (5, 27), and/or on bacterial population levels in diseased persons (6). Since biopsy
samples are very small in size and therefore more easily exposed to oxygen during sampling, the
number of viable strict anaerobes might be reduced easily. Not surprisingly, relatively high levels
of facultative anaerobes were reported to be present in intestinal biopsy samples. So far, there have
been no studies in which the bacterial composition of biopsy samples has been analyzed at the
species level. Molecular approaches based on the sequence variability of 16S rRNA genes could
be instrumental in analyzing the composition of bacterial communities in intestinal biopsy
samples. Recently, such an approach was used to study the bacterial diversity within the human
subgingival crevice (13). In another recent study, temporal TGGE analysis of 16S rDNA
fragments was successfully used to compare the bacterial compositions in gastric biopsy samples
and showed that Helicobacter is detectable in samples from healthy individuals and those suffering
from gastritis (18). In addition, Marteau and colleagues reported significant differences in
community structure between samples from feces and contents from the cecum by using
culturing techniques and dot blot hybridization (16).
68
Chapter 4
The aim of our research was to determine whether the bacterial composition in colonic
biopsy samples was significantly different from that in fecal samples and to investigate whether
differences in composition could be detected at different locations in the colon. We used DGGE
approaches to characterize the 16S rDNA sequence variability of the predominant bacterial
composition and that of the Lactobacillus-like species with general and specific PCR primers (8,
28). We focused on lactobacilli as a subgroup because of their potential probiotic effects in the
human GI tract. The compositional variability in feces and colonic biopsy samples from the
ascending, transverse, and descending colons of 10 volunteers was studied.
4.2 Materials and Methods
Experimental approach.
To describe the bacterial diversity in fecal and biopsy samples, a 16S rRNA approach was
used. DNA was isolated from these samples derived from the same individual, and the V6-to-V8
regions were PCR amplified with general primers and analyzed by DGGE. After scanning of the
gels, similarity indices of DGGE profiles were compared and statistically analyzed. In addition, a
specific PCR was performed to amplify the V2-to-V4 region of the Lactobacillus group that
subsequently was separated by DGGE. To quantify the number of bacteria per biopsy sample, a
flow cytometric approach was used.
Volunteers
Fecal and biopsy samples as fresh as possible were collected from 10 adult human
volunteers. The 10 volunteers donating biopsy and fecal samples were patients undergoing
routine diagnostic colonoscopies. The procedure normally includes biopsies, and so the study did
not cause any extra risk, pain, or discomfort to the participants. Informed consent was obtained
from each volunteer before the sampling. The group consisted of five men and five women (Table
1). With the exception of various GI symptoms (including pains, bloating, and, in patients with
ulcerative colitis, bouts of diarrhea) for which they underwent the examination, the volunteers
considered themselves healthy. They did not follow any special dietary regimen, and none had
recently received any antibiotic treatment.
Colonoscopy, fecal sample collection, and treatment
The colonic evacuation before the colonoscopy was performed by using a laxative
(Colonsteri; Orion Oy, Espoo, Finland) according to the instructions of the manufacturer. The
instrument used for the actual colonoscopy and biopsies was Pentax EC-3801 L. Biopsy samples
(~0.5 mg) were obtained from the ascending (A), transverse (T), and descending (D) parts of the
colon (two parallels per location). One of the parallel samples was stored in 0.05 M potassium
phosphate buffer (pH 7.0), and the other was stored in phosphate-buffered saline (PBS)
(containing, per liter, 8 g of NaCl, 0.2 g of KCl, 1.44 g of Na2HPO4, and 0.24 g of KH2PO4
69
Bacterial communities in colon and fecal samples

[pH 7.2]) with 4% paraformaldehyde. To minimize contamination during sampling, the
colonoscope jaws were carefully washed in tap water after each biopsy was performed. Fecal
samples were obtained before the colonic evacuation. They were stored in the home freezers of
the volunteers and collected immediately prior to the colonoscopy. Both fecal and biopsy samples
were subsequently deeply frozen at -70C, shipped in dry ice, and if appropriate, stored at -70C.
Samples were thawed in ice-water prior to further analysis.
Table 1: Characteristics of the volunteers in this study.
Volunteer Age Sex
a
Diagnosed Bacterial cell counts
c
SI
d
ID (years) illeness
b
A T D F-D A-D
1 59 F None ND 2.0 x 10
5
1.4 x 10
5
83.1 97.6
2 70 M None ND ND ND 91.3 98.1
3 79 M P ND ND ND 67.1 95.9
4 36 F UC* 8.6 x 10
4
1.1 x 10
5
1.1 x 10
5
82.9 56.5
5 43 M UC 6.4 x 10
5
6.4 x 10
5
1.3 x 10
5
22.9 91.2
6 63 M None ND ND ND 25.8 98.0
7 45 F None 5.7 x 10
5
3.8 x 10
5
1.2 x 10
5
24.7 ND
8 30 F UC 6.5 x 10
5
6.9 x 10
5
2.5 x 10
5
40.7 95.8
9 43 M None 1.7 x 10
5
1.2 x 10
5
2.0 x 10
5
13.6 81.6
10 51 F None 2.9 x 10
5
9.0 x 10
5
7.7 x 105 30.0 82.0
a
F, Female; M, Male.
b
P, polyposis; UC, ulcerative colitis, UC* remission of UC.
c
A, T and D, biopsy samples from ascending, transverse and descending parts of the colon, respectively; ND,
not determined.02
d
SI, similarity index; -, comparison ; F, fecal ample.
Bacterial counts in biopsy samples
The paraformaldehyde-fixed biopsy samples were washed twice with PBS and
resuspended in 50% ethanol-PBS. After incubation for at least 1 h at -20C, the biopsy samples
were sonicated in an ultrasonic water bath for 2 min to separate the bacterial cells from the biopsy
material. This treatment has shown to be optimal to separate viable cells from each other without
damaging them (19). After centrifugation at 700 X g for 1 min to remove host cells and debris,
70
Chapter 4
the supernatant was centrifuged at 9,000 X g for 5 min to pellet the bacteria. The bacteria were
resuspended in 490 l of PBS (pH 8.4) and incubated with 5 l of propidium iodide (PI) (1
mg/ml) at 37C for 20 min so that the total number of cells could be determined. Before flow
cytometric counts, 5 l of 0.7-m yellow-green (YG) beads with known concentration
(Polysciences, Inc) was added according to the manufacturer's instructions in order to determine
cell numbers. Samples were analyzed by a FACScalibur flow cytometer (Becton Dickinson).
Illumination of the samples was done with an argon ion laser (488 nm), and fluorescence of the
YG beads and PI were collected in the FL1 (515 to 545 nm) and FL3 (>600-nm long pass)
detectors, respectively. The system threshold was set on forward scatter signals, and all bacterial
analyses were performed at the low rate settings (12 l/min). Collection and analysis of the data
were performed as reported previously (30).
DNA isolation, PCR, and DGGE analysis
Before DNA isolation, fecal samples were resuspended in 0.05 M potassium phosphate.
DNA was isolated from the fecal and unfixed biopsy samples by using the bead beating method
as described previously (30). In short, samples were incubated at 55C for 1 h after addition of
50 l of 10% sodium dodecyl sulfate and 10 l of proteinase K (20 mg/ml), followed by addition
of 150 l of phenol (pH 7.5) and mechanical disruption at 5,000 rpm for 3 min. Phenol-
chloroform extractions and one chloroform extraction were performed to remove impurities.
Before ethanol precipitation at -20C was performed, 1 l of glycogen solution (20 mg/ml) was
added. After washing of the pellets, DNA was resuspended in 100 l of Tris-EDTA buffer.
DNA isolated from biopsy and fecal samples (<10 ng) was subsequently used as a template
to amplify the V6-to-V8 regions of 16S rDNA with primers F-0968-GC and R-1401 (20). The
amplification (35 cycles) and the analysis of 5 l of amplicons on ethidium-stained 1.2% agarose
gels were performed as described previously (28). DGGE analysis of the amplicons was performed
on 8% polyacrylamide gels containing a urea-formamide gradient from 38 to 48% (a 100% urea-
formamide solution consists of 7 M urea and 40% [vol/vol]) formamide). Electrophoresis and
staining of the gels were performed as reported previously (29). Stained gels were scanned at 400
dots per inch and analyzed with the software of Molecular Analyst 1.12 (Bio-Rad). The
similarities between the DGGE profiles were determined by calculating similarity indices of the
densitometric curves of the profiles compared by using the Pearson product-moment correlation
(7, 29). Unweighted pair group method using arithmetic averages (UPGMA), Ward's, and
neighbor-joining algorithms were performed, and corresponding dendrograms showing the
relationships between the DGGE profiles were constructed. Scanning and analysis of the gels
were performed three times.
Amplification of 16S rDNA fragments from the Lactobacillus group population was
performed by a nested-PCR approach. First, the complete 16S rDNA was amplified with the
canonical primers Bact-0011f and Bact-1492r (14). After purification with the Qiaquick PCR
purification kit (Qiagen, Hilden, Germany), the Lactobacillus group-specific PCR was performed
71

with primers Bact-0124-GCf and Lab-0677r followed by DGGE analysis on 8% polyacrylamide
gels containing a urea-formamide gradient from 30 to 60% (8). For cloning and sequence
analysis, the Lactobacillus group amplicons were purified, cloned, and sequenced as described
previously (8).
Paired and Student's t tests were used for statistical analysis of comparisons between the
cell numbers and between similarity indices from the scanned DGGE profiles, respectively.
Nucleotide sequence accession numbers
Sequences determined in this study were deposited in the GenBank database under
accession numbers AY027791 and AY027792.
4.3 Results
Colonoscopic examination of the volunteers
Three of the 10 individuals (numbers 4, 5, and 8 [Table 1]) had previously diagnosed
ulcerative colitis. The disease was in remission both clinically and histologically in individuals 4
and 8, while individual 5 was having a relapse at the time of the study. Polyposis was diagnosed
for individual 3 (Table 1). These four individuals are subsequently indicated as individuals having
a diagnosed illness. No intestinal disease could be detected in the remaining six individuals.
Bacterial numbers in the biopsy samples
The bacteria in biopsy samples of approximately 0.5 mg were counted by a flow
cytometric approach in order to quantify them in a culture-independent way. Since the bacteria
were released from the biopsy material by a mild treatment and since it is difficult to determine
how many cells were still attached after sonication, the total count of bacteria was determined as
the minimal number per biopsy sample. PI-stained bacterial cells could be accurately counted
when beads with known concentration were added, as illustrated in Fig. 1. The different biopsy
samples revealed bacterial quantities that varied between 8.6 x 10
4
and 6.9 x 10
6
cells depending
on the location in the colon and the individual (Table 1), with a mean count of 1.1 x 10
6
bacteria
per sample. The detection limit for accurate counting was found to be 3.7 x 10
4
( standard
deviation [SD] of 1.4 x 10
4
) cells per sample. The numbers of bacteria in specimens from the
ascending colon seem to be slightly lower than from the other locations, although no significant
differences in bacterial numbers at these locations (the lowest P2-tail was 0.075) were found by
paired t test analyses.
72
Chapter 4
Figure 1: Flow cytometric dot blots showing the discrimination between the PI-stained cells and the YG
beads. The different biopsy locations for one individual with colitis ulcerosa (individual 5) are illustrated. A,
T, and D, ascending, transverse, and descending parts of the colon, respectively.
Spatial distribution of the predominant bacterial community
Following DNA isolation from the fecal and biopsy samples of the 10 individuals (Table
1), PCR was performed to amplify the V6-to-V8 regions of 16S rDNA. Amplicons were detected
in all samples with the exception of the biopsy samples from the ascending and transverse colon
of individual 7. DGGE analysis of the fecal and biopsy samples showed an enormous difference
in the diversity of the amplicons in the profiles from the different individuals (as illustrated in Fig.
2). Dilution of biopsy specimen DNA (10 times) did not result in a change in the profile,
indicating that the number of cells per biopsy sample was sufficient to obtain reliable and
reproducible DGGE profiles. Remarkably, the predominant community in biopsy samples from
all locations in the colon gave very similar profiles in each individual, despite the difference in
diversity and diagnosed illness of the individuals (Fig. 2). In contrast, the fecal profiles were in
most cases different from those obtained with the biopsy samples, indicating that it is very
unlikely that fecal contamination took place during the colonoscopy. Since the biopsy samples
were taken after evacuation of the colon it is very plausible that the bacteria detected in these
specimens are mucosa-associated and therefore in close contact with the host cells.
To determine whether communities from feces and biopsy samples were significantly
different in single individuals, similarity indices of the DGGE profiles were calculated. It was
observed that within comparisons between a fecal sample and one of the biopsy samples the
individual variation was relatively high compared to the comparisons between different biopsy
samples from the same individual. For example, the similarity indices for comparisons between
feces and descending colon biopsy specimens varied from 13.6 to 91.3, while similarity indices
between 56.5 and 98.1 were found when descending colon biopsy specimens were compared to
ascending colon specimens (Table 1). Overall, indices for the similarity indices of comparisons
between all biopsy samples from the same individual were very high (91.6 9.6 [SD]), close to
those calculated for the reproducibility of the procedures (93.4 3.6). To rule out that the
73
A B
D
diagnosed illness of some of the individuals had an influence on the observed findings, the mean
and standard variation of each similarity index within the healthy individuals and those with
diagnosed illness were compared separately. Student's t test revealed that there was no significant
difference between the similarity indices of both groups for each comparison, since the lowest P2-
tail observed was 0.065 (7 df ) for the similarity indices for comparison between ascending and
transverse colonic biopsy samples.
Figure 2: Silver-stained DGGE gel showing profiles which represent the predominant community of feces
(F) and biopsy samples (A, T, D) of individuals 1, 5, and 10. M represents the marker for DGGE analysis.
The physiological conditions of the individuals are indicated.
The similarity indices between a fecal sample and one biopsy sample were compared with
those between the remaining biopsy samples in order to obtain independent comparisons for
statistical analysis (Fig. 3). All combinations of comparisons showed that the bacterial
composition in fecal samples was significantly different from that in the biopsy samples. The
highest P2-tail was 0.0012 (16 df ) for comparison between the similarity indices of feces and
transverse colon biopsy samples with ascending and descending colon biopsy samples.
Spatial distribution of the Lactobacillus community
A nested-PCR approach was used to specifically amplify the V2-to-V4 regions of the 16S
rDNA of the Lactobacillus group community, since no amplicons were retrieved by a direct
specific-PCR approach. In contrast to the DGGE profiles of the predominant bacterial
community, the Lactobacillus group-specific profiles were lower in diversity, as illustrated in Fig. 4.
74
Chapter 4
Because of this low diversity, similarity indices for the DGGE profiles cannot be determined. In
contrast to the predominant bacterial community, the Lactobacillus communities in fecal and
biopsy samples were very similar in 6 of the 10 individuals. In these individuals, only one
amplicon was dominating (see, for example, the data for individuals 1 and 5 in Fig. 4). In the
other individuals one of the fecal amplicons was the only predominant one in the biopsy samples
or vice versa (such as for individual 10 in Fig. 4). Furthermore, for 3 of the 10 individuals some
minor differences in the Lactobacillus group compositions between the biopsy samples were
found. These small differences could not be explained by the physiological condition, age, or
gender of the host since they were found in individuals 2 (healthy), 8 (remission of ulcerative
colitis), and 10 (healthy).
Figure 3: Comparison between fecal and biopsy samples. F, fecal samples; A, T, and D, biopsy samples from
the ascending, transverse, and descending parts of the colon, respectively; -, comparison. The means and SD
of similarity indices are indicated.
Comparison between healthy individuals and individuals with diagnostic illeness
DGGE profiles of biopsy samples from the descending colons of individuals with and
without a diagnosed illness were compared to see if the presence or absence of specific bacteria
could be correlated to the illeness. The descending colon was chosen, since all diagnosed illnesses
were observed at least in this part. The profiles of the predominant bacterial community appeared
to be unique for each individual, and no specific amplicon could be assigned to the presence or
absence of a colonic illness (Fig. 5A). To analyze the predominant communities, similarity indices
of the comparisons between the DGGE profiles were calculated. Repetitive comparisons between
the UPGMA, Ward, and neighbor-joining algorithms were performed, and dendrograms were
constructed. Only two clusters were found in all dendrograms, while the position of the others
branches in the dendrogram changed depending on the clustering method. The large error bars
of the nodes in the UPGMA tree (Fig. 5B) could be seen as an indication that the corresponding
branches of these nodes may vary between the different algorithms. One of the repetitive clusters
consisted of four healthy individuals (i.e., individuals 1, 2, 6, and 7), and the other consisted of
75
the two individuals diagnosed with an active form of a GI tract disorder (i.e., individuals 3 and
5). This preliminary observation suggests that there might be differences in the predominant
bacterial composition between healthy and diseased individuals, although the group of
individuals in this study is too small to allow a definite conclusion.
Figure 4: Silver-stained DGGE gel showing profiles which represent the Lactobacillus group community
of fecal (F) and biopsy (A, T, D) samples of individuals 1, 5, and 10. The physiological conditions of the
individuals are indicated.
For the Lactobacillus group community also no specific differences could be found
between the individuals with and those without diagnosed illness (Fig. 6). A striking observation
was the presence of an amplicon with identical DGGE band positions for 9 of the 10
individuals. Since it appeared to be a Lactobacillus gasseri-like species (see below), we tested
whether its predominance might be a result of preferential amplification. Our nested-PCR
approach did not show any preference in favor of L. gasseri when mixtures of its DNA with that
from Lactobacillus acidophilus and Lactobacillus paracasei were used as template DNA for PCR.
Cloning and sequence analysis of the amplicons from the DGGE profiles of individual 1
(healthy) and individual 3 (with polyposis) showed that both sequences had 99% similarity with
L. gasseri. Alignment of the two sequences showed that they differ by only one base (adenine in
one and thymine in the other). This indicates that L. gasseri is likely to be a predominant
Lactobacillus species in the biopsy samples.
4.4 Discussion
In this study we have used a culture-independent approach based on the 16S rDNA
sequence variability to analyze bacterial communities in different parts of the colon. Fecal and
biopsy samples were taken from people with and without a diagnosed illness. Since the colon
was evacuated before biopsies were performed, it is very likely that the bacteria in the biopsy
samples are mucosa associated. The minimum number of cells per biopsy sample as measured
76
Chapter 4
by flow cytometry is comparable to numbers found by cultivation of bacteria from biopsy
samples which were obtained by a similar procedure (1). Considerable variation was found in
the bacterial numbers from different biopsy samples. Factors that may cause this variation
include the evacuation and sampling procedures, the sonication procedure, and individual
differences. On the other hand, this variation in bacterial number may explain why no PCR
product could be obtained from two biopsy samples.
Figure 5: (A) Silver-stained DGGE gel showing profiles which represent the predominant communities of
the descending colon biopsy samples from individuals 1 to 10. M represents the marker for DGGE
analysis. The physiological conditions of the individuals are indicated. (B) UPGMA dendrogram
illustrating the correlation between the different DGGE profiles of panel A. Cluster I and II represent the
repetitive clusters obtained using different algorithms. The black bars represent the error bars.
Figure 6: Silver-stained DGGE gel showing profiles which represent the Lactobacillus group communities
of the descending colon biopsy samples from individuals 1 to 10. The physiological conditions of the
individuals are indicated. The arrowheads indicate the amplicons which have been identified by cloning
and sequencing.
77
DGGE analysis of 16S rDNA amplicons was used to determine, compare, and
visualize the compositions of the predominant bacterial and of the Lactobacillus group
communities. The DGGE profiles reflecting the predominant bacterial communities in
biopsy samples from different locations in the GI tract were highly similar to each other, while
they differed significantly from those of fecal samples (Fig. 3). Therefore, it seems that the
mucosa-associated bacteria are equally distributed along the complete colon and that different
populations are dominating in the mucosa and the feces. Recently, differences in the
structures of communities in feces and cecal contents, observed through a dot blot
hybridization and culturing approach, have been reported (16). Culture-dependent studies of
contents from different parts of the colon (including the ascending, transverse, and
descending parts) of sudden-death victims have revealed that the conditions, for example, pH
and concentration of fermentation products, in these parts differ considerably from one
another (15). This suggests that the uniform distribution of the attaching bacterial
composition along the colon is very likely due to host-bacterium interactions at the mucosa.
Several studies have already suggested that the bacterial community in the GI tract has a
strong effect on the host and that signaling between host and bacterium is very important (3,
9, 10). In a recent study, a significant positive relationship between the genetic relatedness of
the hosts and the similarity between their bacterial communities was found (29). However, it
is not clear yet what the nature of these host-related factors is.
With a nested-PCR approach using group-specific primers (8), the Lactobacillus
group-specific composition was analyzed. In contrast to those of the predominant community,
the profiles of biopsy and fecal samples were quite similar for 6 of the 10 individuals.
Furthermore, for 3 of the 10 individuals some minor differences between the biopsy samples
were found in the Lactobacillus group composition. This suggests that the changing conditions
in the GI tract influence the presence or absence of certain species belonging to the
Lactobacillus group. Another explanation could be the detection limit of these bacteria in the
biopsy specimens. Since we are focusing on a subpopulation in a community which contains
approximately 10
6
bacteria, a small difference in the number of organisms of a certain species
might have a large impact on its detection. Remarkably, one amplicon with the highest
sequence similarity (99%) to L. gasseri was found in descending colon biopsy samples of 9 of
the 10 individuals. Moreover, it was the most predominant one in most biopsy specimens.
Since the 16S rDNA of this species was not preferentially amplified by the nested-PCR
approach, L. gasseri may be regarded as a general mucosa-associated bacterium in humans.
Because colonic illnesses were observed during colonoscopy, samples from healthy
individuals and those with a diagnosed illness were compared to each other. Since the illnesses
are found especially in the descending colon, we compared the samples from these regions and
compared them with those from healthy individuals. No significant difference could be
detected with respect to the number of bacteria per biopsy specimen, the composition of the
predominant bacterial community, and that of the Lactobacillus group community. This is
78
Chapter 4
supported by the observation that the profiles reflecting the predominant community were
highly similar along the complete colon for both groups. Lactobacilli were detected in both
feces and biopsy samples from all individuals, and no differences in the Lactobacillus group
populations between healthy and diseased tissues were found.
The molecular approach used in this study can be influenced by preferential
amplification and difference in DNA isolation efficiency of different species. However, DNA
from biopsy samples could be diluted 10 times without changing the profiles. Furthermore,
for two individuals the similarity between fecal and biopsy samples was very high, while the
number of bacteria in the fecal samples was at least 10
3
times higher than that in the biopsy
samples. These data indicate that it is very unlikely that procedures such as sampling, storage,
and transport or preferential lysis of specific groups of bacteria have a major impact on our
observations.
In conclusion, using a culture-independent approach we were able to clearly
demonstrate that mucosa-associated bacterial communities in the colon are significantly
different in composition from those in feces. A strikingly high similarity between bacterial
communities from different locations in the colon was observed. This observation suggests
strongly that host-related factors are important in the colonic ecosystem, which is in line with
previous observations (3, 9, 10, 29). A culture-independent approach was also used to
characterize subpopulations of the Lactobacillus genus. Similar approaches using group-
specific primers can also be applied to study (sub)populations of other bacteria such as those
implicated in the initiation and maintenance of ulcerative colitis (4, 5, 27). Hence, systematic
culture-independent approaches could be instrumental in determining the roles of various GI
tract subpopulations in the pathogenesis of colonic diseases.
4.5 Acknowledgments
This work was partly supported by the Wageningen Centre for Food Sciences. We thank
G. H. J. Heilig, P. Verbaarschot, M.-L. Keklinen, and M. Rekola for technical assistance and J.
A. G. M. de Visser for advice on statistical analysis. In addition, we thank all volunteers for
providing fecal and biopsy samples.
4.6 References
1. Alander, M., R. Korpela, M. Saxelin, T. Vilpponen-Salmela, T. Mattila-Sandholm, and
A. von Wright. 1997. Recovery of Lactobacillus rhamnosus GG from human colonic
biopsies. Lett. Appl. Microbiol. 24:361-364.
2. Alander, M., R. Satokari, R. Korpela, M. Saxelin, T. Vilpponen-Salmela, T. Mattila-
Sandholm, and A. von Wright. 1999. Persistence of colonization of human colonic
79

mucosa by a probiotic strain, Lactobacillus rhamnosus GG, after oral consumption. Appl.
3. Bry, L., P. G. Falk, T. Midtvedt, and J. I. Gordon. 1996. A model of host-microbial
interactions in an open mammalian ecosystem. Science 273:1381-1383.
48:132-135.
5. Gibson, G. R., J. H. Cummings, and G. T. Macfarlane. 1991. Growth and activities of
sulphate-reducing bacteria in the gut contents of healthy subjects and patients with
ulcerative colitis. FEMS Microbiol. Ecol. 86:103-112.
6. Gillian Hartley, M., M. J. Hudson, E. T. Swarbrick, M. J. Hill, A. E. Gent, M. D.
Hellier, and R. H. Grace. 1992. The rectal mucosa-associated microflora in patients with
ulcerative colitis. J. Med. Microbiol. 36:96-103.
7. Hne, B. G., K. Jger, and H. Drexler. 1993. The Pearson product-moment correlation
coefficient is better suited for identification of DNA fingerprint profiles than band
matching algorithms. Electrophoresis 14:967-972.
and W. M. de Vos. Molecular diversity of Lactobacillus spp. and other lactic acid bacteria
in the human intestine as determined by specific amplification of 16S ribosomal DNA.
9. Hooper, L. V., J. Xu, P. G. Falk, T. Midtvedt, and J. I. Gordon. 1999. A molecular sensor
that allows a gut commensal to control its nutrient foundation in a competitive
ecosystem. Proc. Natl. Acad. Sci. USA 96:9833-9838.
10. Hooper, L. V., M. H. Wong, A. Thelin, L. Hansson, P. G. Falk, and J. I. Gordon. 2000.
Molecular analysis of host-microbial relationships in the intestine. Science 291:881-884.
11. Hopkins, M. J., R. Sharp, and G. T. Macfarlane. 2001. Age and disease related changes
in intestinal bacterial populations assessed by cell culture, 16S rRNA abundance, and
community cellular fatty acid profiles. Gut 48:198-205.
12. Johansson, M.-L., G. Molin, B. Jeppsson, S. Nobaek, S. Ahrn, and S. Bengmark. 1993.
Administration of different Lactobacillus strains in fermented oatmeal soup: in vivo
colonization of human intestinal mucosa and effect on the indigenous flora. Appl.
13. Kroes, I., P. W. Lepp, and D. A. Relman. 1999. Bacterial diversity within the human
subgingival crevice. Proc. Natl. Acad. Sci. USA 96:14547-14552.
80
Chapter 4
14. Lane, D. J. 1991. 16S/23S rRNA sequencing, p. 115-175. In. E. Stackebrandt and M.
Goodfellow (ed.), Nucleic acid techniques in bacterial systematics. J. Wiley & Sons,
Chichester, United Kingdom.
15. Macfarlane, G. T., G. R. Gibson, and J. H. Cummings. 1992. Comparison of
fermentation reactions in different regions of the human colon. J. Appl. Bacteriol.
72:57-64.
16. Marteau, P., P. Pochart, J. Dor, C. Bra-Maillet, A. Bernallier, and G. Corthier. 2001.
Comparative study of bacterial groups within the human cecal and fecal microbiota. Appl.
17. Millar, M. R., C. J. Linton, A. Cade, D. Glancy, M. Hall, and H. Jalal. 1996.
Application of 16S rRNA gene PCR to study bowel flora of preterm infants with and
without necrotizing enterocolitis. J. Clin. Microbiol. 34:2506-2510.
18. Monstein, H. J., A. Tiveljung, C. H. Kraft, K. Borch, and J. Jonasson. 2000.
Profiling of bacterial flora in gastric biopsies from patients with Helicobacter pylori-
associated gastritis and histologically normal control individuals by temperature
gradient gel electrophoresis and 16S rDNA sequence analysis. J. Med. Microbiol.
49:817-822.
19. Nebe-von-Caron, G., P. Stephens, C. H. Hewitt, J. R. Powel, and R. A. Badley. 2000.
Analysis of bacterial function by multi-colour fluorescence flow cytometry and single cell
sorting. J. Microbiol. Methods 42:97-114.
20. Nbel, U., B. Engelen, A. Felske, J. Snaidr, A. Wieshuber, R. I. Amann, W. Ludwig, and
H. Backhaus. 1996. Sequence heterogeneities of genes encoding 16S rRNAs in
Paenibacillus polymyxa detected by temperature gradient gel electrophoresis. J. Bacteriol.
178:5636-5643.
21. Simpson, J. M., V. J. McCracken, H. R. Gaskins, and R. I. Mackie. 2000. Denaturing
gradient gel electrophoresis analysis of 16S ribosomal DNA amplicons to monitor
changes in fecal bacterial populations of weaning pigs after introduction of Lactobacillus
reuteri strain MM53. Appl. Environ. Microbiol. 66:4705-4714.
22. Suau, A., R. Bonnet, M. Sutren, J. J. Godon, G. R. Gibson, M. D. Collins, and J. Dor.
many novel molecular species within the human gut. Appl. Environ. Microbiol.
65:4799-4807.
23. Tannock, G. W. 1995. Normal microflora. An introduction to microbes inhabiting
the human body. Chapman and Hall, London, United Kingdom.
Gopal. 2000. Analysis of the fecal microflora of human subjects consuming a
81

probiotic product containing Lactobacillus rhamnosus DR20. Appl. Environ.
Microbiol. 66:2578-2588.
25. Vaughan, E. E., G. H. J. Heilig, E. G. Zoetendal, R. Satokari, J. K. Collins, A. D.
L. Akkermans, and W. M. de Vos. 1999. Molecular approaches to study probiotic
bacteria. Trends Food Sci. Tech. 10:400-404.
26. Wilson, K. H., and R. H. Blitchington. 1996. Human colonic biota studied by
ribosomal DNA sequence analysis. Appl. Environ. Microbiol. 62:2273-2278.
27. Zinkevich, V., and I. B. Beech. 2000. Screening of sulfate-reducing bacteria in
colonoscopy samples from healthy and colitic human gut mucosa. FEMS Microbiol.
Ecol. 34:147-155.
28. Zoetendal, E. G., A. D. L. Akkermans, and W. M. de Vos. 1998. Temperature
gradient gel electrophoresis analysis from human fecal samples reveals stable and host-
specific communities of active bacteria. Appl. Environ. Microbiol. 64:3854-3859.
29. Zoetendal, E. G., A. D. L. Akkermans, W. M. Akkermans van-Vliet, J. A. G. M. de
Visser, and W. M. de Vos. 2001. The host genotype affects the bacterial community
in the human gastrointestinal tract. Microb. Ecol. Health Dis. 13:129-134.
30. Zoetendal, E. G., K. Ben-Amor, A. D. L. Akkermans, T. Abee, and W. M. de Vos.
2001. DNA isolation protocols affect the detection limit of PCR approaches of
bacteria in samples from the human gastrointestinal tract. Syst. Appl. Microbiol.
24:405-410.
82
Chapter 4
Chapter 5
POPULATION DYNAMICS AND DIVERSITY OF
FECAL MICROBIOTA OF PATIENTS WITH
ULCERATIVE COLITIS PARTICIPATING IN A
PROBIOTIC TRIAL
Kaouther Ben-Amor, Suzanne Verhaegh, Ineke Heikamp-de Jong, Willem M. de Vos and
Elaine E. Vaughan
Abstract: The gastrointestinal microbiota of patients with ulcerative colitis were
monitored comprehensively over time, while investigating its response to
ingestion of probiotics, either Lactobacillus salivarius subsp salivarius UCC118 or
Bifidobacterium infantis 35624, or a placebo. The 13 patients are a subset
partaking in a current double-blind trial to study the efficiency of the probiotics
to maintain the patients in remission. The microbiota of fecal samples prior to
the probiotic trial as well as 1, 3, 6, 9, and 12 month samples were analyzed by
fluorescent in situ hybridization (FISH) in combination with flow cytometry
(FCM), and the molecular fingerprinting technique PCR-denaturing gradient
gel electrophoresis (DGGE). A set of 6 FISH probes were used for enumeration
that covered major fecal bacterial groups, namely, Clostridium coccoides -
Eubacterium rectale, Bacteroides - Prevotella, Bifidobacterium species, Atopobium,
and E. coli and related species, as well as the lactic acid bacteria in order to
determine the Lactobacillus population. The numbers of bifidobacteria and
lactobacilli increased (p=0.04 and 0.07, respectively) over time in all patients
samples, while the Atopobium group decreased significantly (p=0.02). However
these changes were not related to the treatment type. Notably, the C. coccoides - E.
rectale group decreased significantly (p=0.02) in the placebo group at month 3.
The FISH analysis revealed substantial temporal variation in the microbiota of
each patient. The latter was supported by the PCR-DGGE profiles that further
revealed a relatively low complexity, both of which are in sharp contract to the
very stable and highly complex profiles of healthy subjects. Cluster analysis of the
profiles and calculation of the similarity indices confirmed the high temporal
shifts, but indicated that the microbiota were stabilizing later in the trial as from
month 6.
83
5.1 Introduction
The human gastrointestinal (GI) tract is colonized by a large and complex community of
microorganisms, an estimated half of which are still not culturable by conventional methods (38,
47). The GI-tract microbiota play essential roles in a wide variety of metabolic and
immunological processes and hence promote the health of the host (4). However, under certain
environmental circumstances and/or in genetically susceptible individuals, there is strong
evidence that the GI-tract microbiota may be involved in the pathogenesis and aetiology of a
number of inflammatory bowel diseases (IBD) such as ulcerative colitis (UC) and Crohns disease
(2). Indeed there are accumulating indications that the luminal and mucosa-associated microbiota
play a critical role in the onset and perpetuation of colitis (34, 40). This has led to attempts to
modulate the GI-tract microbiota by means of an alternative therapeutic approach consisting of
the use of live probiotic bacteria in order to restore a balanced microbiota and potentially
downregulate the inflammation (20, 36, 37).
Probiotics are defined as live microorganisms which, when administered in adequate
amounts confer a health benefit on the host (5). Lactobacillus and Bifidobacterium species are
the most commonly used probiotic bacteria and both are normal inhabitants of the healthy
intestine. Despite certain haziness about the use of probiotics as health promoting factors or as
therapeutic agents, today there is persuasive evidence supporting their efficacy in the treatment
or prevention of a number of intestinal disorders such as lactose intolerance, acute gastro-
enteritis, and food allergy (27, 29, 42). Recently, experimental and clinical studies have revealed
that certain probiotics may be useful in treatment and/or prevention of IBD. Successful results
have been reported with Escherichia coli Nissle 1917 in patients with UC (16, 30), and with a
multispecies probiotic containing eight bacterial strains (VSL#3) in the maintenance of
remission of UC patients (45) and in the prevention of pouchitis onset and relapse (8). More
recently. oral administration of two strains, namely Bifidobacterium infantis UCC35624
combined with Lactobacillus salivarius subsp. salivarius UCC118, significantly reduced the
effect of IBD severity in animal models as assessed by reduced weight loss, improvement of
colon pathology and markedly improved appearance of mice over a six-week period (3, 22).
The modulation of the enteric microbiota with probiotics appears to contribute to the
management of IBD. Thus, knowledge of the structure and dynamics of the bacterial
populations of the indigenous microbiota that interact with the gut are necessary for
understanding the impact of probiotics on this community.
During the last decade, the advent in molecular techniques, especially those based on
16S ribosomal RNA genes, allowed the evaluation of the total GI-tract microbial community
despite their poor cultivability (44). Fingerprinting techniques, such as PCR-denaturing
gradient gel electrophoresis (PCR-DGGE), and in-situ hybridization with a range of
fluorescent oligonucleotide probes (FISH) that monitor the microbiota, in absence of
cultivation have been applied to both feces and biopsy samples (14, 33, 39, 49, 51). These
84
Chapter 5
studies have advanced our knowledge of the intestinal microbiota by unraveling the
complexity of this ecosystem and strengthen our insight in the establishment and succession
of the bacterial community within the host (6, 9). The application of these methods
demonstrated that the predominant bacterial community in human feces is host-specific,
relatively stable in time and not significantly altered following consumption of certain
probiotic strains (41). Furthermore, it revealed that the predominant bacterial species
associated with the colonic mucosa are significantly different from the predominant fecal
community (51). Recent investigation using FISH coupled to FCM or microscopic analysis
have shown that the majority of fecal bacteria of healthy individuals belong to the Bacteroides-
Prevotella group, Eubacterium rectale-Clostridium coccoides cluster, Clostridium leptum group,
Atopobium group and bifidobacteria (7, 10, 31). Differences in the occurrence of these
bacterial groups have been reported and may possibly be related to differences in the genetic
background, lifestyle, and diet in the human populations studied.
However, there is sparse data on the intestinal microbiota of individuals with IBD such
as UC. The initial aim of this study was to investigate the response of the gut microbiota of
UC patients upon consumption of Lactobacillus salivarius subsp. salivarius UCC118,
Bifidobacterium infantis UCC35624 strains or a placebo. Thus the population dynamics and
structure (composition) of the microbiota in feces of patients with UC were characterized
using PCR-DGGE fingerprinting and FISH combined with flow cytometry. Our findings
indicated that microbiota of UC patients was less complex and unstable over time and quite
different to that of healthy individuals.
5.2 Materials and methods
Subjects
Patients with ulcerative colitis (UC) were recruited at the sites of the European project
PROGID (Probiotics and GastroIntestinal Disorders) including University College Cork,
Ireland; Hospital General Vall dHebron, Barcelona Spain; and Harjula Hospital of Kuopio,
Finland. These subjects were patients whose conditions were in clinical and endoscopic
remission (score <4) and who had experienced regular episodes of relapse despite steroid
treatment. They had to be younger than 75 year, and be on 5-aminosalicylic acid treatment
(5-ASA). These patients were randomized to receive either a probiotic product (P-1 or P-2) or
a placebo 2 to 4 weeks after stopping steroid use. During the trial, patients who had relapsed
(score >7) were excluded from the trial and went back to their normal treatment. In our study
a subgroup of 13 patients were followed to monitor the effect of the probiotic intake on the
fecal microbiota. Patients who partook 3 months of the trial by the time we carried out this
research were included. The gender and age of the subjects are presented in Table 2.
85
Population dynamics of fecal microbiota of UC patients

Probiotic clinical trial
The study was a large-scale, multi-centred, randomized, double blind, placebo-
controlled feeding trial within a subset of a European population with ulcerative colitis. The
duration of the study was 12 months for each patient. The probiotic was supplied in a sachet
containing 7.5 g of freeze dried yoghurt carrier powder containing Streptococcus thermophilus
UCC948, to which concentrated freeze-dried Bifidobacterium infantis UCC35624 or
Lactobacillus salivarus subsp. salivarus UCC118 was added. The placebo consisted of the dried
yoghurt carrier without probiotic. Patients were asked to consume one sachet per day, which
contained 10
10
viable probiotic cells.
Fecal sample preparation
Feces from patients were collected at the project sites. Samples intended for DNA
extraction were immediately frozen at 20C whereas samples to be analysed by FISH were
fixed within a few hours after the collection of the sample at the project sites. To extract fecal
microbial cells, 0.5 g of fecal sample was resuspended in 4.5 ml phosphate buffered saline
(PBS, pH 7.0) and homogenized by vortexing the fecal suspension for 3 min at high speed in
the presence of 5 to 10 glass beads of 3-m diameter. After centrifuging at 700 X g for 1 min,
1 ml of supernatant was added to 3 ml of 4% para-formaldehyde (PFA), and incubated at 4C
for 4 h and then stored at 80C. All samples were transported to our laboratory on dry ice.
Upon reception the fixed cells were washed twice with PBS and stored in 50% (vol/vol)
ethanol-PBS at 20 C until further use. Total cell count was performed upon reception of
the samples as described below.
Fluorescent in situ hybridization
The oligonucleotides probes used in this study, their sequence, their targeted bacteria
as well as the hybridization conditions are presented in Table 1. The Eub338 probe was used
as the positive control and the Non-Eub as a negative control to monitor the non-specific
binding. All oligonucleotide probes were purchased from MWG (Ebersgerg, Germany) and
were mono- labeled at the 5 end with Cy3 or Cy5. The nucleic acid stain TOTO-1 iodide
was purchased from Molecular Probes (Leiden, The Netherlands) and used for total cell
counts. TOTO-1 is a membrane-impermeant nucleic acid with a very high fluorescence
enhancement and quantum yield upon binding to the nucleic acids of a cell. When excited
with a blue light (488 nm), the stained cells fluoresce yellow-green and give a positive single
in both FL1 and FL2 detectors of the FCM. Hybridization was carried out as described
previously (50). Briefly, 50 to 100 ml of ethanol-fixed fecal cell suspension was centrifuged at
8,000 X g for 3 min and washed with 1 X PBS. The cell pellet was suspended in a volume of
40 ml hybridization buffer (0.9 M NaCl, 20 mM Tris-HCl (pH 7.2), 0.1% (wt/vol) SDS) to
which 5 l of the Cy5-labeled specific probe (30 ng/l) and 5 l of Cy3-labeled Eub338 was
added. The samples were incubated according to the conditions indicated in Table 1.
86
Chapter 5
Subsequently, the cells were centrifuged at 9,000 X g for 3 min and resuspended in a washing
solution (0.9 M NaCl and 20 mM Tris-HCl, pH 7.2) and incubated for 20 min at the
appropriate temperature (Table 1). Cells were finally pelleted and resuspended in 1 ml ice-
cold 1 X PBS (pH 7.0) and kept in the dark on ice until flow cytometric analysis was
performed within 1 h. Prior to hybridization with the Lab158 probe, cells were treated in 500
ml PBS buffer containing 10 mg of lysozyme/ml for 60 min at 37C to aid the
permeabilization of the cell membrane.
Table 1: List of oligonucleotide probes used in the present study.
Probe Probe sequence Target organism Hybridization Reference
name
Temp
1
Time
(C) (h)
Eub338 GCTGCCTCCCGTAGGAGT Most bacteria 50/50 3-16 (1)
NonEub ACATCCTACGGGAGGC 50/50 16 (46)
Erec482 GCTTCTTAGTCA(A/G)GTACCG Clostridium coccoides /
Eubacterium rectale
cluster 50/50 16 (7)
Bac303 CCAATGTGGGGGACCTT Bacteroides /
Prevotella group 46/48 3 (19)
Lab158 GGTATTAGCA(C/T)CTGT TTCCA Lactobacillus /
Enterococcus group 50/50 16 (12)
Ato291 GGTCGGTCTCTCAACCC Atopobium cluster 50/50 16 (11)
Bif164 CATCCGGCATTACCACCC Bifidobacteria 50/50 16 (17)
Ec1531 CACCGTAGTGCCTCG TCATCA Escherichia coli and
related species 37/37 3 (28)
1
The values correspond to the hybridization and washing temperature, respectively.
Flow cytometry
Samples were analyzed with a FACSCalibur flow cytometer (Beckton Dickinson
Immunocytometry Systems, San Jose, Calif.) equipped with an air-cooled argon ion laser
providing 15 mW at 488 nm combined with a 670 nm red-diode laser with the standard filter
set-up as described previously (50). All parameters were collected on logarithmic scale and
control samples were used for the instrument settings (voltage of the detectors and the
compensation). These controls consisted of unlabeled fecal cells and/or cells hybridized with
the Non-Eub probe, cells stained with TOTO-1 (FL1/FL2), cells stained with TO-PRO-3,
and/or hybridized with Eub338-Cy5 (FL4). CaliBRITE beads (Becton Dickinson) were used
to check the instrument sensitivity over time. Cells were discriminated using a double
threshold set on both side scatter (SSC) and forward scatter (FSC), with FSC set on E01 and
SSC on 400V. All samples were sonicated (2 X 60 sec) and thoroughly vortexed prior to FCM
87
analysis. A dual dot plot FSC versus SSC in combination with one parameter histogram
representing the TOTO-1 fluorescence (FL1) was used to back gate fecal cells and distinguish
them from the background. For total cell enumeration, samples were incubated in the
presence of 1 nM TOTO-1 for 5 min at room temperature. Unlabeled beads of 6.0-m
diameter provided from the Bacteria Counting Kit (Molecular probes BV, Leiden, The
Netherlands) were added to each sample stained with TOTO-1 at a final concentration of 10
6
beads/ml and served as an internal standard to calibrate the sample volume. The ratio of
TOTO-1 stained cells to the number of beads was used to calculate the absolute total cell
counts. The relative abundance (proportion) of bacterial groups was estimated as the ratio of
the number of cells hybridized with the Cy5-labeled group-specific probe to the number of
TOTO-1 stained cells.
DNA extraction and PCR amplification
The frozen feces was thawed and DNA extraction was performed with the FastDNA
kit, QIAmp mini stool kit (QIAGEN GmbH, Germany) from 0.2 g fecal material. DNA
isolated from fecal samples was used as a template to amplify the V6 to V8 regions of 16S
rDNA with primers 968-GC-f and 1401-r (24). PCRs were performed using a Taq DNA
polymerase kit from Life Technologies (Gaithersburg, Madison, USA). The reaction mixture
consisted of 20 mM Tris-HCl (pH 8.4), 3 mM MgCl2, 0.2 mM deoxynucleotide triphosphate
(dNTP), 0.2 M concentration of each primer, 1.25 U of Taq polymerase, and 1 l of
appropriately diluted template DNA in a final volume of 50 l. Samples were amplified in a
PE Applied Biosytems GenAmp PCR system 9700 (Foster City, California, USA) using the
following program: pre-denaturation at 94C for 5 min; 35 cycles of denaturation 94C for
30 sec, annealing temperature of 56C for 20 sec, extension at 68C for 40 sec, and final
extension at 68C for 7 min. The integrity of the nucleic acids was determined visually after
electrophoresis on a 1.2 % agarose gel containing ethidium bromide.
DGGE of PCR amplicons
PCR amplicons were separated by DGGE based on the protocol of Muyzer et al. (23)
using the Decode system (Bio-Rad Laboratories, Hercules, California, USA) with the
following modifications. Polyacrylamide gels consisted of 8% (vol/vol) polyacrylamide (ratio
of acrylamide-bisacrylamide, 37.5:1) and 0.5 x Tris-acetate-EDTA (pH 8.0) (TAE) buffer.
Denaturing acrylamide of 100% was defined as 7 M urea and 40% formamide. The
polyacrylamide gels were made with denaturing gradient ranging from 38 to 48%. The gels
were poured from the top using a gradient maker and a pump (Econopump; Bio-Rad
Laboratories, Hercules, California, USA.) set at a speed of 4.5 ml/min. Prior to the
polymerization of the denaturing gel (gradient volume, 28 ml), a 7.5-ml stacking gel without
denaturing chemicals was added, and the appropriate comb was subsequently inserted.
Electrophoresis was performed first for 5 min at 200 V and then for 16 h at 85 V in 0.5 x
88
Chapter 5
TAE buffer at a constant temperature of 60C. The gels were stained with AgNO3 as described
by Sanguinetti et al (32) and dried overnight at 60C.
The changes in the bacterial counts determined by FISH and FCM were analyzed by
Repeated Measures ANOVA using the SPSS statistical software (SPSS Inc, Chicago, USA).
The processing of the DGGE gels was performed using the Bionumerics software 2.0 (Applied
Maths, Kortrijk, Belgium). The Pearson correlation coefficient for each pair of lanes within a
gel was calculated as a measure of similarity between the community fingerprints, and the
clustering of patterns was calculated using the unweighted-pair group method using average
linkages (UPGMA). Dendrograms of each DGGE gel of one patient and a combination of all
gels were thus generated.
5.3 Results
Assessment of the fecal microbiota by using FISH and FCM
A set of 6 probes was used to enumerate the major fecal bacterial groups of the 13 UC
patients. These were (i) Bif164 to detect the genus Bifidobacterium, (ii) Erec482 for most
members of the Clostridium coccoides - Eubacterium rectale, (iii) Bac303 for Bacteroides -
Prevotella group, (iv) Ato291 to detect the Atopobium group with Collinsella aerofaciens as the
predominant species, (v) Lab158 for lactobacilli and enterococci, and (vi) Ec1531 for
Esherichia coli and related species. The Eub338 probe was used as a positive control since it
targets almost all bacteria and the NonEub as a negative control to monitor unspecific
hybridization.
FCM analysis of the hybridized samples gave a shift in signal of 1 log unit or more
compared to the non-hybridized cells, allowing the specific detection and enumeration of the
different bacterial groups (Fig. 1). Total cell enumeration was performed using a ratiometric
counting method and TOTO-1 staining. Fecal samples were stained with the membrane-
impermeant nucleic acid dye TOTO-1, and subsequently 10
6
unlabeled beads were added to
1 ml sample. The sample volume analyzed was determined from the number of beads
measured while the sample was run, and the amount of the cells was then determined by
dividing the number of cells counted by this volume. Figure 2 shows a typical FCM result for
cell counting. In the SSC versus FL1 dot plot (Fig. 2 C) a region R1 was used to gate the
TOTO-1-stained fecal cells and, a second region R2 was drawn around the bead population.
In the FL1 versus FL4 dot plot (Fig. 2 D), region R3 shows the Cy-5 labeled cells that were
hybridized with the specific probe. Total bacteria and specific bacterial groups can thus be
accurately counted by calculating the ratio of the cell events to the bead events and corrected
to the dilution factor. The fecal samples from the 13 patients revealed bacterial cell
concentrations that varied between 9.5 x 10
9
and 6.5 x 10
10
cells/g wet feces (Table 2).
89

No significant changes in the total counts were observed over time. The same sample was
analyzed at different time points to check the sensitivity of the method and the coefficient of
variance was found to be as low as 15% (data not shown). The relative abundance of the
specific bacterial groups was estimated as a ratio of the number of hybridized cells to the
number of TOTO-1 stained cells.
Figure 1: Analysis of the composition of fecal microbiota using the six 16S rRNA targeted probes. The
SSC versus FL4 dot plots display the positive Cy5 fluorescence signal conferred by all the Cy5-labeled
probes. A shift of at least 1 log scale is obtained upon hybridization (upper right quadrant) compared to
the non-hybridized cells (lower right quadrant). The signal in the lower left quadrant represents debris.
Figure 2: Gating strategy for absolute count of fecal cells using TOTO-1 staining and microspheres of
known concentration as described in materials and methods.
90
Chapter 5
91
T
a
b
l
e

2
:

C
h
a
r
a
c
t
e
r
i
s
t
i
c
s

o
f

t
h
e

s
u
b
j
e
c
t
s

s
t
u
d
i
e
d

a
n
d

t
h
e

t
o
t
a
l

f
e
c
a
l

c
e
l
l

c
o
u
n
t
s

a
s

d
e
t
e
r
m
i
n
e
d

b
y

T
O
T
O
-
1

s
t
a
i
n
i
n
g

a
n
d

F
C
M
.
M
e
a
n

o
f

t
o
t
a
l

b
a
c
t
e
r
i
a
l

c
o
u
n
t
s

(
n
u
m
b
e
r

o
f

c
e
l
l
s
/
g

w
e
t

f
e
c
e
s
)

2
P
a
t
i
e
n
t
G
e
n
d
e
r
A
g
e
G
r
o
u
p

1
M
o
n
t
h

0
M
o
n
t
h

1

M
o
n
t
h

3
M
o
n
t
h

6
M
o
n
t
h

9
M
o
n
t
h

1
2
U
C
-
1
0
F
e
m
a
l
e
4
1
P
-
1
3
.
5
1

x

1
0
1
0
5
.
3
2

x

1
0
1
0
7
.
9
6

x

1
0
1
0
1
.
3
5

x

1
0
1
1
7
.
6
1

x

1
0
1
0
4
.
2
8

x

1
0
1
0
U
C
-
1
1
M
a
l
e
6
7
P
-
1
4
.
9
8

x

1
0
1
0
4
.
7
8

x

1
0
1
0
2
.
9
9

x

1
0
1
0
2
.
6
9

x

1
0
1
0
n
.
d
n
.
d
U
C
-
1
2
F
e
m
a
l
e
2
7
P
-
2
6
.
5
2

x

1
0
1
0
2
.
5

x

1
0
1
0
3
.
4
6

x

1
0
1
0
n
.
d
n
.
d
n
.
d
U
C
-
1
3
M
a
l
e
3
2
P
-
2
5
.
9
8

x

1
0
1
0
1
.
6
6

x

1
0
1
1
4
.
2
9

x

1
0
1
0
n
.
d
n
.
d
n
.
d
U
C
-
1
4
M
a
l
e
5
3
P
l
a
c
e
b
o
2
.
6
6

x

1
0
1
0
1
.
7
7

x

1
0
1
0
4
.
8
6

x

1
0
1
0
2
.
5
7

x

1
0
1
0
n
.
d
n
.
d
U
C
-
1
5
F
e
m
a
l
e

3
4
P
l
a
c
e
b
o
3
.
6
8

x

1
0
1
0
4
.
6
0

x

1
0
1
0
2
.
9
4

x

1
0
1
0
4
.
4
9

x

1
0
1
0
5
.
2
6

x

1
0
1
0
n
.
d
U
C
-
1
6
F
e
m
a
l
e
5
0
P
-
1
n
.
d
1
.
3
3

x

1
0
1
1
1
.
0
9

x

1
0
1
1
n
.
d
n
.
d
n
.
d
U
C
-
1
7
M
a
l
e
6
0
P
-
2
3
.
1
6

x

1
0
1
0
1
.
4
6

x

1
0
1
0
8
.
9
5

x

1
0
9
1
.
2
2

x

1
0
1
0
n
.
d
n
.
d
U
C
-
1
8
M
a
l
e
3
3
P
l
a
c
e
b
o
1
.
5
2

x

1
0
1
0
8
.
6
6

x

1
0
9
n
.
d
n
.
d
n
.
d
n
.
d
U
C
-
5
7
M
a
l
e
4
5
P
-
2
3
.
8
7

x

1
0
1
0
7
.
8
5

x

1
0
1
0
4
.
1
5

x

1
0
1
0
n
.
d
2
.
6
1

x

1
0
1
0
n
.
d
U
C
-
6
1
M
a
l
e
5
6
P
-
1
n
.
d
1
.
7
8

x

1
0
1
0
3
.
4
2

x

1
0
1
0
1
.
0
9

x

1
0
1
1
n
.
d
n
.
d
U
C
-
1
4
7
F
e
m
a
l
e
6
1
P
-
1
1
.
4
7

x

1
0
1
0
1
.
4
7

x

1
0
1
0
4
.
0
7

x

1
0
1
0
n
.
d
n
.
d
n
.
d
U
C
-
1
4
8
F
e
m
a
l
e
5
7
P
l
a
c
e
b
o
9
.
4
6

x

1
0
9
7
.
8
2

x

1
0
9
2
.
7
1

x

1
0
1
0
n
.
d
n
.
d
n
.
d
M
e
a
n

S
D
4
7

1
3
3
.
5
1

(
1
.
7
9
)

x

1
0
1
0

5
.
2
7

(
5
.
0
9
)

x

1
0
1
0
3
.
5
8

(
1
.
8
4
)

x

1
0
1
0
5
.
9
0

(
5
.
0
)

x

1
0
1
0
5
.
1
6

(
2
.
5
0
)

x

1
0
1
0
1
P
-
1

a
n
d

P
-
2

c
o
r
r
e
s
p
o
n
d

t
o

t
h
e

p
r
o
b
i
o
t
i
c
-
1

a
n
d

p
r
o
b
i
o
t
i
c
-
2

g
r
o
u
p
,

r
e
p
e
c
t
i
v
e
l
y
.
2
V
a
l
u
e
s

a
r
e

m
e
a
n
s

o
f

6

r
e
p
l
i
c
a
t
e
s
.
n
.
d
;

n
o
t

d
e
t
e
r
m
i
n
e
d

(
s
a
m
p
l
e

w
a
s

n
o
t

y
e
t

r
e
c
e
i
v
e
d

f
o
r

F
I
S
H

a
n
a
l
y
s
i
s

o
r

p
a
t
i
e
n
t

h
a
d

r
e
l
a
p
s
e
d
)
.
Composition of the microbiota of UC patients prior to the trial
The mean percentages of all bacterial groups measured prior to the trial (month 0) are
presented in Table 3. The results show that at the start of the trial bacteria belonging to the
C. coccoides - E. rectale cluster were the most abundant and accounted for 35.9 7.5% of the total
TOTO-1-stained cells with a range of 20 to 65 %. This was followed by the Bacteroides/Prevotella
group which represented 12.2 7.5% of cells detected by TOTO-1 (ranging from 2.2 to 22.5%).
The Bifidobacterium group came third and corresponded to 9.7 5.4% of the total cells (with a
range of 0.4 - 17.3%). The Lactobacillus group showed a high variation between subjects covering
0.2 0.3% of the total cells. Lactobacillus and related species were not detected in all patients,
presumably because of the detection limit of the method which is estimated to be 10
5
cells (48).
Finally, the Atopobium and the Enterobacteriacae represented 4.1 2.4% and 4.7 10.9% of the
total cells, respectively. The E. coli / Klesbiella probe signal was positive in the feces from five
patients two of who, UC-17 and UC-18, had very high counts comprising 18 and 30% of the
total cells, respectively. When the percentages of all bacterial groups were added together a mean
of 67.4 17.2% was obtained, indicating that approximately 33% of the dominant microbiota
belonged to as yet undefined phylogenetic groups. The mean count of all analysed samples
revealed that 70 9.0%, of the total cells (TOTO-1 counts) hybridized with the Eub338 probe.
Table 3: Composition of the fecal microbiota of UC patients prior to the trial (month 0).
1
Mean of three replicates (three different hybridisations.
2
Additivity of the six probes used to illustrate the % coverage of these probes.
n.d, not determined (sample was not received).
92
Chapter 5
Effect of the probiotic intake on the patient faecal microbiota
The 13 subjects studied were a subgroup of a larger number of UC patients who have been
recruited by the PROGID project and have been randomized to receive either a placebo, or a
probiotic-1 or probiotic-2 (L. salivarius subsp. salivarius UCC118 or B. infantis 35624) during 1
year. Since the trial is double blind and the project is still ongoing, the codes are not yet available.
Figure 3: Effect of the probiotic intake on the alterations of the major fecal bacteria of UC patients during
the first 3 months of the trial as assessed by FISH and FCM. Results are expressed as mean (SD) percentage
of the total TOTO-1-stained.
93
FISH in combination with FCM was used to monitor the changes in the major fecal
bacteria of the 13 patients during the test period at month 1, 3, 6, 9 and 12. When patients
relapsed, samples were no longer obtained and occasionally samples were not received due to
FISH in combination with FCM was used to monitor the changes in the major fecal bacteria of
the 13 patients during the test period at month 1, 3, 6, 9 and 12. When patients relapsed, samples
were no longer obtained and occasionally samples were not received due to other reasons,
consequently, statistical analysis was carried out until month 3 as all patients were still in the trial.
The response in relation to modulation of the gut microbiota to the probiotic supplementation
was studied by comparing changes in the fecal microbial composition during the trial, and
between the intervention and the placebo groups (Fig. 3). The statistical analysis showed that the
fecal concentrations of bifidobacteria increased significantly (p=0.04) over time with an increase
to a near level of significance (p=0.07) in the Lactobacillus population compared to their baseline
level (pre-treatment or month 0) as shown in Figure 3. Indeed, the Bifidobacterium group was the
second most represented with 11.1 6.2% of total cells detected by TOTO-1 (ranging from 2.7
to 21.1%) at three-month post-treatment (data not shown). The Atopobium cluster count
decreased significantly (p=0.02) over time in all patients. The between subject analysis revealed
that the changes in the bifidobacteria, lactobacilli and atopobia were not related to the
administration of the probiotic. Notably the Bacteroides - Prevotella decreased to a near level of
significance (p=0.07) in the patient who took probiotic-2. In month 3 of the test period, a
significant decrease of C. coccoides - E. rectale populations was observed in the placebo group
(p=0.03) (Fig. 3).
Interestingly, the relapsed patients (with active disease), when considered as a group,
showed lower numbers of bifidobacteria, Atopobium and Bacteroides and a higher proportion of
clostridia/eubacteria then the patients who were still in remission (data not shown). Patient UC-
10 is the only patient who completed the trial until month 12 by the end of this study. This
patient showed significant (p<0.01) increase in bifidobacteria and lactobacilli, a decrease in the
proportion of bacteria belonging to the Bacteroides - Prevotella and C. coccoides E. rectale clusters,
while the Atopobium group stayed stable over time (data not shown). At month 6, the
bifidobacteria population increased spectacularly to reach a proportion of 71% of the total fecal
cells. At the end of the trial, this patient had a highly significant level (p<0.01) of bifidobacteria
(24%) and lactobacilli (20%) as compared to the pre-treatment period values of 16 and 0.6%,
respectively.
Biodiversity of the fecal microbiota as assessed by PCR-DGGE
DGGE analysis of the PCR amplified fragments of the V6-V8 regions of fecal bacteria
16S rRNA gene of the 13 patients was performed. The DGGE profiles representing the diversity
and the shift in the dominant bacterial community of the 12 patients are presented in Figure 4.
94
Chapter 5
Figure 4: PCR-DGGE profiles of the V6 -V8 regions of 16S rDNA from faecal samples of 12 patients taken
at month 0, 1, 3, 6, 9 and 12, reflecting the predominant bacterial communities.
95
These patterns demonstrate that each subject harbors a unique microbiota that is in
agreement with previous studies demonstrating that the host genotype influences the microbiota
(49). On the other hand, the total bacterial PCR-DGGE fingerprints revealed significant
temporal alterations as estimated by the number, position and intensity of the bands. Statistical
analysis of the profiles using the Bionumerics software allowed the calculation of a distance matrix
presenting the similarity indices of all possible gel tracks within the DGGE patterns by using the
Pearson correlation coefficient. Based on the values of the resulting matrix, a cluster analysis was
performed and samples were visualized in dendrograms. The cluster analysis of patients samples
supported the significant temporal variation, examples of 4 patients samples are presented in
Figure 5. In general, the average similarity index (SI) of healthy adult microbiota over a period
time of at least two years is approximately 76% (48), however the similarity indices for the UC
patients were somewhat lower (data not shown). In addition, the complexity of the microbiota
for a number of UC patients as estimated by the increase in the number of the bands seemed to
increase during the course of the trial (Fig. 4). Interestingly, the cluster analysis showed that the
patients microbiota appeared to be stabilizing in the months later in the trial i.e. month 6 to
month 12 (Fig. 5). The bacterial profiles of the patients UC-147, -148, -57 and -61 at month 6,
9 and 12 had higher similarity indices than the profiles at month 0, 1 and 3, this is illustrated in
Figure 5. For patient UC-57, the similarity indices was 26, 38, 27, 64 and 72% for month 0, 1,
3, 6 and 9, respectively, when compared to the profile of month 12, indicating that the profiles
at month 9 and 12 are becoming more similar.
5.4 Discussion
The present study involved a subset of UC patients who partook in an European Union
clinical trial to study the efficiency of 2 previously selected probiotic stains, L. salivarius subsp.
salivarius UCC118 and B. infantis UCC35624, in the maintenance of remission. The fecal
microbiota of these patients was monitored comprehensively for the first time over a period of
3 to 12 months using both the fingerprinting PCR-DGGE method and FISH combined with
FCM. The fingerprinting analysis revealed that the UC patients had a host specific fecal
microbiota, which is in agreement with previous studies, although the complexity was
somewhat lower than that of healthy subjects (48, 49). The uniqueness of the patients
microbiota observed by PCR-DGGE analysis (Fig. 4, 5) was supported by the results of FISH
and FCM (Table 3).
Furthermore, the wide variation in bacterial counts of the microbiota between each
patient supported the uniqueness of each individual microbiota (Table 4). Additionally, over
the course of the trial, significant alterations were observed both in the total bacterial DGGE
profiles (Fig. 4) as well as in the major bacterial group counts (Fig. 3), indicating that each
patients microbiota composition was unstable over time. Recently, Seksik et al. (35) showed
that the fecal microbiota composition in patients with Crohns disease was also unstable over
time and differed significantly compared to that of healthy people. Using temporal temperature
96
Chapter 5
gel electrophoresis (TTGE) and quantitative dot blot hybridization techniques, they
demonstrated that the fecal microbiota of diseased people was altered but retained a high degree
of diversity (35).
Figure 5: UPGMA cluster analysis of fecal bacteria DGGE profiles of four UC patients before (month 0)
and during the trial (month 1, 3, 6, 9 and 12) illustrated as M0, M1, M3, M6, M9 and M12.
The PCR-DGGE profiles of UC patients did not show any explicit band that could
indicate the presence of specific bacteria that may attributable to the disease. Remarkably, when
we compared the total facal bacterial profiles of six healthy individuals to that of UC patients at
month 0, a dendrogram with 2 clusters that clearly separated the patterns of healthy to those of
UC patients was obtained (Fig. 6). This suggests that the dominant microbiota of UC patients is
atypical. More recently Ott et al. (26) showed the colonic mucosa-associated microbiota of patient
with UC and Crohns disease (CD) differed significantly from healthy subjects. Their findings
revealed that the alterations of the fecal microbiota observed in IBD patients were associated with
a reduction in the microbial diversity especially in relation to the Bacteroides as well as
Lactobacillus species.
97
Figure 6: UPGMA cluster analysis of faecal bacteria profiles as assessed by PCR-DGGE fingerprinting of 10
UC patients (UC-11, 13, 14, 15, 16, 59, 147 and 148) at month 0 compared to 6 healthy adults (H1, 2,
3, 4 5 and 6)
While some culture studies reported a decrease in the number of Bifidobacterium species
and an increase in Bacteroides counts in UC patients (13, 43), our results revealed that
bifidobacteria were detected at the pre-treatment period in all patients at a level of 9.6 5.4 % of
the total fecal microbiota and that the Bacteroides - Prevotella cluster accounted only for 12.6
7.6%. Using the same technique, Harmsen et al. (10) and Rigottier-Gois et al. (31) reported that
in healthy people bifidobacteria accounted for 2.7 to 4.1% of the total microbiota while the
Bacteroides - Prevotella group represented 9.6 to 22.0%. In patients with active IBD, numbers of
Bacteroides detected in fecal and colonic mucosa-associated microbiota were reduced as compared
to patients with inactive disease or healthy subject (25, 35). Moreover, we found that bacteria
belonging to the C. coccoides - E. rectale group were detected in all UC patients at a relatively high
level, 20 to 65% of the total microbiota, in comparison to healthy people where the C. coccoides-
E. rectale group was reported to vary between 14 to 29% of the total fecal microbiota (10, 18). E.
coli and related species were detected in only three patients at high numbers.
Today, there is persuasive evidence that the intestinal microbiota play an important role in
the onset and maintenance of colitis arising from both experimental IBD animal models and
observations in humans (20). This has provided the stimulus to test probiotics as a therapeutic
agent in order to maintain remission in UC patients. Although, the molecular mechanisms by
which probiotics exert their beneficial effects are not fully elucidated, there is growing evidence
that the modulation of the intestinal microbiota contributes to the reduction of exacerbation of
UC. Recently, clinical trials were performed to test the benefits of probiotics for the treatment of
UC using a multispecies probiotic VSL#3 containing 3 strains of bifidobacteria, 4 strains of
lactobacilli and one strain of Streptococcus salivarius subsp. thermophilus at a concentration of
98
Chapter 5
5 x 10
11
cells/g (45). The ingestion of the probiotic preparation resulted in an increase in the
numbers of the lactobacilli, bifidobacteria and S. salivarius subsp. thermophilus, while those of
other genera (Bacteroides, clostridia, coliforms, and total aerobes and anaerobes) did not change.
Ishikawa et al. (15) showed that the intake of a bifidobacteria fermented milk, containing strains
of Bifidobacterium breve, Bifidobacterium bifidum and Lactobacillus acidophilus, did not result in a
significant difference in Bacteroides or bifidobacteria counts of UC patients.
During the course of the probiotic trial, a significant increase in both bifidobacterial
(p=0.04) and Lactobacillus (p=0.07) populations and a decrease in the Atopobium group (p=0.02)
were observed for all patients. These changes were not related to the type of treatment and were
observed in the three groups: placebo, probiotic-1 and probiotic-2 groups. However, Bacteroides
counts decreased significantly in the probiotic-2 group, while clostridia decreased in the placebo
group. In a previous study using the same L. salivarius strain UC118 and a placebo product
(yoghurt with S. thermophilus) on 80 healthy individuals, it was found that the yoghurt group
showed more pronounced responses to modulation of the gut microbiota and revealed a marked
reduction in the number of enterococci and clostridia (3, 21). These findings suggested that
products of fermentation may also play a role in in vivo effects, and that consideration must be
given to the delivery vehicle in comparisons to the efficiency between probiotic strains. On the
other hand, the wide variation in bacterial counts of bacterial groups between UC patients
resulted in high standard deviations, and may have limited the sensitivity of statistical
comparisons between treatment groups to detect subtle changes due to the probiotic intake (Table
3, Fig. 3). It is possible that speciation may reveal more subtle changes within this microbiota with
respect to the disease activity and the effect of probiotics. Ishikawa et al.(15) showed that although
Bacteroides did not change significantly during the supplementation of bifidobacteria fermented
milk to UC patients, the relative number of Bacteroides vulgatus was significantly reduced in the
probiotic groups as compared to the control group.
The findings of this study revealed that fecal microbiota of UC patients is host specific
like that of healthy subjects, but less complex and shows significantly more temporal instability.
The results obtained by both molecular methods supported the uniqueness of the bacterial
composition for UC patients. During the course of the trial, the bacterial profiles of a number of
patients tended to stabilize at the end of the trial. These observations suggest that the
administration of probiotics may have resulted in a normalization of the total microbiota. The
use of a greater variety of probes for FISH to cover the total fecal community, and more detailed
comparisons with samples from healthy individuals may reveal further relevant information that
can impact on probiotic therapy for UC patients.
99

5.5 Acknowledgments
This study was supported by the EU Quality of life and Management of Living Resources,
PROGID project (QLK1-2000-00563) coordinated by Professor Fergus Shanahan (Cork,
Ireland). It does not necessary reflect its views and in no way anticipates the commissions future
policy in this area. We thank Dr. Edwin Martens for his assistance for the statistical analysis and
Dr. Atte von Wright for providing samples.
5.6 References
48:132-135.
3. Dunne, C., L. Murphy, S. Flynn, L. O'Mahony, S. O'Halloran, M. Feeney, D.
Morrissey, G. Thornton, G. Fitzgerald, C. Daly, B. Kiely, E. M. Quigley, G. C.
O'Sullivan, F. Shanahan, and J. K. Collins. 1999. Probiotics: from myth to reality.
Demonstration of functionality in animal models of disease and in human clinical trials.
Antonie Van Leeuwenhoek 76:279-92.
4. Falk, P. G., L. V. Hooper, T. Midtvedt, and J. I. Gordon. 1998. Creating and
maintaining the gastrointestinal ecosystem: what we know and need to know from
gnotobiology. Microbiol. Mol. Biol. Rev. 62:1157-1170.
5. FAO/WHO. 2002. Guidelines for the evaluation of probiotics in food. Joint Food and
Agriculture Organization of the United Nations and World Health Organization
Working group report. FAO/WHO.
6. Favier, C. F., E. E. Vaughan, W. M. De Vos, and A. D. L. Akkermans. 2002. Molecular
7. Franks, A. H., H. J. M. Harmsen, G. C. Raangs, G. J. Jansen, F. Schut, and G. W.
8. Gionchetti, P., F. Rizzello, A. Venturi, P. Brigidi, D. Matteuzzi, G. Bazzocchi, G.
Poggioli, M. Miglioli, and M. Campieri. 2000. Oral bacteriotherapy as maintenance
treatment in patients with chronic pouchitis: a double-blind, placebo-controlled trial.
9. Harmsen, H. J., A. C. Wildeboer-Veloo, G. C. Raangs, A. A. Wagendorp, N. Klijn,
J. G. Bindels, and G. W. Welling. 2000. Analysis of intestinal flora development in
100
Chapter 5
breast-fed and formula-fed infants by using molecular identification and detection
methods. J. Pediatr. Gastroenterol. Nutr. 30:61-67.
Extensive set of 16S rRNA-based probes for detection of bacteria in human feces.
Group and the Atopobium cluster and their application for enumeration of
Microbiol. 66:4523-4527.
12. Harmsen, J. H. M., P. Elfferich, F. Schut, and G. W. Welling. 1999. A 16S rRNA-tageted
probe for detection of lactobacilli and enterococci in faecal samples by fluorescent in situ
hybridization. Micriol. Ecol. Health Dis. 11:3-12.
13. Hartley, M. G., M. J. Hudson, E. T. Swarbrick, M. J. Hill, A. E. Gent, M. D. Hellier,
and R. H. Grace. 1992. The rectal mucosa-associated microflora in patients with
ulcerative colitis. J. Med. Microbiol. 36:96-103.
and W. M. de Vos. 2002. Molecular diversity of Lactobacillus spp. and other lactic acid
15. Ishikawa, H., I. Akedo, Y. Umesaki, R. Tanaka, A. Imaoka, and T. Otani. 2002.
Randomized controlled trial of the effect of bifidobacteria-fermented milk on ulcerative
colitis. J. Am. Coll. Nutr. 22:56-63.
16. Kruis, W., E. Schutz, P. Fric, B. Fixa, G. Judmaier, and M. Stolte. 1997. Double-blind
comparison of an oral Escherichia coli preparation and mesalazine in maintaining
remission of ulcerative colitis. Aliment. Pharmacol. Ther. 11:853-858.
18. Lay, C., L. Rigottier-Gois, K. Holmstrom, M. Rajilic, E. Vaughan , M. D. Collins, R.
Thiel, P. Namsolleck, M. Blaut, and J. Dore. 2004. Assessment of human faecal
microbiota composition using FISH combined with flow cytomety, Pan-European
comparison., p. 75. PROEUHEALTH: The food, GI-tract functionality and human
health cluster 3
rd
workshop. VTT Biotechnolgy (http://www.vtt.fi/inf/pdf ). Stiges, Spain.
19. Manz, W., R. Amann, W. Ludwig, M. Vancanneyt, and K. Schleifer. 1996. Application
of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria of
101

the phylum cytophaga-flavobacter-bacteroides in the natural environment. Microbiol.
142:1097-1106.
20. Marteau, P., P. Seksik, and F. Shanahan. 2003. Manipulation of the bacterial flora in
inflammatory bowel disease. Best. Pract. Res. Clin. Gastroenterol. 17:47-61.
21. Mattila-Sandholm, T., S. Blum, J. K. Collins, R. Crittenden, W. M. de Vos, C. Dunne,
R. Fondn, G. Grenov, E. Isolauri, B. Kiely, P. Marteau, A. Morelli, A. C. Ouwehand,
R. Reniero, M. Saarela, S. J. Salminen, M. Saxelin, E. Schiffrin, F. Shanahan, E.
Vaughan, and A. von Wright. 1999. Probiotics: towards demonstrating efficacy. Trends
Food Sci. Technol. 10:393-399.
22. McCarthy, J., L. O'Mahony, L. O'Callaghan, B. Sheil, E. E. Vaughan, N. Fitzsimons, J.
Fitzgibbon, G. C. O'Sullivan, B. Kiely, J. K. Collins, and F. Shanahan. 2003. Double
blind, placebo controlled trial of two probiotic strains in interleukin 10 knockout mice
and mechanistic link with cytokine balance. Gut 52:975-980.
24. Nubel, U., B. Engelen, A. Felske, J. Snaidr, A. Wieshuber, R. Amann, W. Ludwig, and
178:5636-5643.
bacterial microflora in patients with active inflammatory bowel disease. Gut 53:685-693.
26. Otte, J. M., and D. K. Podolsky. 2004. Functional modulation of enterocytes by gram-
positive and gram-negative microorganisms. Am. J. Physiol. Gastrointest. Liver. Physiol.
286:G613-626.
27. Ouwehand, A. C., S. Salminen, and E. Isolauri. 2002. Probiotics: an overview of
beneficial effects. Antonie van Leeuwenhoek 82:279-289.
28. Poulsen, L. K., T. R. Licht, C. Rang, K. A. Krogfelt, and S. Molin. 1995. Physiological
state of Escherichia coli BJ4 growing in the large intestines of streptomycin-treated mice.
J Bacteriol. 177:5840-5845.
29. Reid, G., M. E. Sanders, H. R. Gaskins, G. R. Gibson, A. Mercenier, R. Rastall, M.
Roberfroid, I. Rowland, C. Cherbut, and T. R. Klaenhammer. 2003. New scientific
paradigms for probiotics and prebiotics. J. Clin. Gastroenterol. 37:105-118.
102
Chapter 5
30. Rembacken, B. J., A. M. Snelling, P. M. Hawkey, D. M. Chalmers, and A. T. Axon.
1999. Non-pathogenic Escherichia coli versus mesalazine for the treatment of ulcerative
colitis: a randomised trial. Lancet 354:635-639.
32. Sanguinetti, C. J., E. Dias Neto, and A. J. Simpson. 1994. Rapid silver staining and
recovery of PCR products separated on polyacrylamide gels. Biotechniques 17:914-21.
33. Satokari, R. M., E. E. Vaughan, A. D. L. Akkermans, M. Saarela, and W. M. de Vos.
2001. Bifidobacterial diversity in human feces detected by genus-specific PCR and
denaturing gradient gel electrophoresis. Appl. Environ. Microbiol. 67:504-513.
34. Schultsz, C., F. M. Van Den Berg, F. W. Ten Kate, G. N. Tytgat, and J. Dankert. 1999.
The intestinal mucus layer from patients with inflammatory bowel disease harbors high
numbers of bacteria compared with controls. Gastroenterol. 117:1089-1097.
35. Seksik, P., L. Rigottier-Gois, G. Gramet, M. Sutren, P. Pochart, P. Marteau, R. Jian, and
J. Dore. 2003. Alterations of the dominant faecal bacterial groups in patients with
36. Shanahan, F. 2003. Probiotics: a perspective on problems and pitfalls. Scand. J.
Gastroenterol. Suppl. 237:34-36.
37. Shanahan, F. 2002. Probiotics and inflammatory bowel disease: from fads and fantasy to
facts and future. Br. J. Nutr. 88:5-9.
many novel molecular species within the human gut. Appl. Environ. Microbiol. 65:4799-
4807.
39. Suau, A., V. Rochet, A. Sghir, G. Gramet, S. Brewaeys, M. Sutren, L. Rigottier-Gois,
and J. Dore. 2001. Fusobacterium prausnitzii and related species represent a dominant
group within the human fecal flora. Syst. Appl. Microbiol. 24:139-145.
40. Swidsinski, A., A. Ladhoff, A. Pernthaler, S. Swidsinski, V. Loening-Baucke, M. Ortner,
J. Weber, U. Hoffmann, S. Schreiber, M. Dietel, and H. Lochs. 2002. Mucosal flora in
inflammatory bowel disease. Gastroenterol. 122:44-54.
Gopal. 2000. Analysis of the fecal microflora of human subjects consuming a probiotic
product containing Lactobacillus rhamnosus DR20. Appl. Environ. Microbiol. 66:2578-
2588.
103

42. Teitelbaum, J. E., and W. A. Walker. 2002. Nutritional impact of pre- and probiotics as
protective gastrointestinal organisms. Annu. Rev. Nutr. 22:107-138.
43. van der Wiel-Korstanje, J. A., and K. C. Winkler. 1975. The faecal flora in ulcerative
colitis. J. Med. Microbiol. 8:491-501.
44. Vaughan, E. E., F. Schut, H. G. H. Heilig, E. G. Zoetendal, W. M. de Vos, and A. D.
L. Akkeremans. 2000. A molecular view of the intestinal ecosystem. Curr. Issues Intest.
Microbiol. 1:1-12..
45. Venturi, A., P. Gionchetti, F. Rizzello, R. Johansson, E. Zucconi, P. Brigidi, D.
Matteuzzi, and M. Campieri.1999. Impact on the composition of the faecal flora by a
new probiotic preparation: preliminary data on maintenance treatment of patients with
ulcerative colitis. Aliment. Pharmacol. Ther. 13:1103-1088.
47. Wilson, K., and R. Blitchington. 1996. Human colonic biota studied by ribosomal DNA
sequence analysis. Appl. Environ. Microbiol. 62:2273-2278.
the human gastronintestinal tract. Microbial Ecol. Health Dis. 13:129-134.
49. Zoetendal, E. G., A. D. L. Akkermans, and W. M. De Vos. 1998. Temperature gradient
community Recovered from Feces. Appl. Environ. Microbiol. 68:3401-3407.
104
Chapter 5
Chapter 6
MULTIPARAMETRIC FLOW CYTOMETRY AND
CELL SORTING FOR THE ASSESSMENT OF
VIABLE, INJURED AND DEAD BIFIDOBACTERIA
CELLS DURING BILE SALT STRESS
Kaouther Ben Amor, Pieter Breeuwer, Patrick Verbaarschot, Frank. M. Rombouts, Antoon D.
L. Akkerman, Willem M. de Vos, and Tjakko Abee
Abstract: Using a flow cytometry-based approach, we assessed the viability of
Bifidobacterium lactis DSM 10140 and Bifidobacterium adolescentis DSM 20083
during exposure to bile salt stress. Carboxyfluorescein diacetate (cFDA), propidium
iodide (PI) and oxonol (DiBAC4 (3)) were used to monitor the esterase activity,
membrane integrity and membrane potential, respectively, as indicators of bacterial
viability. Single staining with these probes rapidly and noticeably reflected the
behavior of the two strains during stress exposure. However, the flow cytometry
results tended to overestimate the viability of the two strains as compared to plate
counts, which appeared to be related to the unculturability of a fraction of the
population as a result of the sublethal injury caused by bile salts. When the cells
were simultaneously stained with cFDA and PI, flow cytometry and cell sorting
revealed a striking physiological heterogeneity within the stressed bifidobacteria
population. Three subpopulations could be identified based on their differential
uptake of the probes: cF-stained, cF and PI double-stained and PI-stained
subpopulations, representing viable, injured and dead cells, respectively. Following
sorting and recovery, a significant fraction of the double-stained sub-population
(40%) could resume growth on agar plates. Our result show that in situ assessment
of the physiological activity of stressed bifidobacteria using multiparametric flow
cytometry and cell sorting may provide a powerful and sensitive tool to assess the
viability and stability of probiotics.
Applied and Environmental Microbiology 68: 5209- 5216(2002)
Reprinted with permission from the American Society for Microbiology
105
6.1. Introduction
Bifidobacteria represent one of the most important bacterial groups of the human
gastrointestinal tract. They are believed to play a beneficial role in maintaining the balance of
the intestinal microbiota and have been implicated in a number of health promoting effects
(26, 36). Despite their increasing use as probiotics, scientific evidence is still lacking with
regard to the mechanisms to which these bacteria can contribute to the health promotion of
the host. Thus, knowledge of their metabolic activities and ecology represents an important
step in understanding the beneficial effects of these microorganisms and consequently may
form a basis to rationally select probiotic strains (17). This is partly due to the fact that some
of the current methodologies lack the resolving power to analyze the composition and the
metabolic activity of these microorganisms in situ. The assessment of active or viable
microorganisms is often difficult, since no single analytical method identifies all physiological
characteristics of a bacterium under a certain condition (1). Although the plate count
approach is employed as the standard method to measure bacterial viability, it only indicates
how many of the cells can replicate under the conditions provided for growth. It allows, in
principle, to determine the viability in a retrospective manner and there is no explicit evidence
that the failure of a bacterial cell to reproduce is an indication that the cell was dead at the
time of sampling (6, 14). The ability to reproduce might be repressed or blocked in a certain
cell type, or reproduction might be limited to a certain set of conditions. In addition, cell
populations that have been exposed to stress can enter an unculturable state while they still
can maintain metabolic activity. Moreover, there is increasing evidence that a significant
proportion of metabolically active bacteria in the environment can not be cultured and are
known as viable but not culturable (VNBC) or as others prefer to define as active but not
culturable (ABNC) (1, 16, 38).
Fluorescent techniques in combination with flow cytometry (FCM) have been
extensively used for the assessment of viability of microorganisms from different
environmental samples (2, 7, 28). FCM offers a powerful tool to analyze a cell population at
the single cell level, since it can be used to both identify and enumerate bacterial population
from environmental samples, and characterize functional properties of the individual cell (14,
31, 34). Moreover, it allows simultaneous measuring of different physical and biochemical
parameters and hence offers substantial information on the dynamics and physiological
heterogeneity of a bacterial population (10, 11, 23). In addition, FCM offers the facility to
physically separate selected cells using cell sorting for further molecular and physiological
analysis (15, 37).
The most widely used dyes developed for the assessment of cell viability involves
carboxy-fluoresceine diacetate (cFDA) a non-fluorescent precursor that readily diffuses across
the cell membranes. Once inside the cell it is converted by non-specific esterases to a
membrane-impermeant fluorescent compound. Retention of the dye by the cell indicates
106
Chapter 6
membrane integrity and functional cytoplasmic enzymes, while dead cells do not stain because
they lack enzyme activity or the carboxyfluorescein (cF) diffuses freely through the damaged
membranes (5, 8). In addition, membrane potential-sensitive probes such as bis-(1,3-
dibutylbarbutiric acid) trimethine oxonol (DiBAC4(3)) have been used extensively to assess
bacterial antibiotic susceptibility (13, 21, 35) and cell viability by flow cytometry (18, 19, 20).
Oxonol is a negatively charged molecule, which enters depolarized and dead cells and binds
to lipid-rich compounds and results in a high green fluorescence. Another group of probes
for viability studies consists of nucleic acid dyes, such as propidium iodide (PI) that are
excluded by viable cells with intact membranes but can enter into cells with compromised
membranes and bind to the DNA and RNA. The fluorescence conferred by these probes
indicates the degree of cell damage, cell permeability and ultimately cell death (9, 12, 29, 32).
The aim of this study was to assess the viability of the probiotic bacteria
Bifidobacterium lactis DSM 10140 and Bifidobacterium adolescentis DSM 20083 during
deconjugated bile salt stress (dBS) using a rapid method based on fluorescent probes and flow
cytometry. In a first step, we evaluated the possibility of using oxonol, PI, and cFDA in single
staining assays to monitor the changes in membrane potential, membrane permeability, and
enzyme activity, respectively of the two strains during stress conditions and compare these
parameters with the plate count method. In a second step, a multiparametric flow cytometry
assay was used in combination with cell sorting in order to determine the contribution of each
single cell to the overall physiological status of the bacterial population.
6.2. Materials and methods
Bacterial strains and media
Bifidobacterium lactis DSM 10140 and Bifidobacterium adolescentis DMS 20083 were
purchased from the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH
(Braunschweig, Germany). The strains were grown in MRS (Oxoid) supplemented with
0.05% L-cysteine-HCl (wt/vol) and adjusted to pH 6.8 in an anaerobic chamber with an
atmosphere of 10% CO2, 10% H2 and 80% N2. Stock cultures of the two strains were
maintained at 80C in MRS containing 0.05% L-cysteine-HCl (wt/vol) and 30% glycerol
(vol/vol).
Stress conditions
One milliliter of the stock culture was diluted in 10 ml MRS broth supplemented with
0.05% L-cysteine-HCl (wt/vol) and incubated at 37C in the anaerobic chamber for 15-16 h.
The overnight culture was then diluted 10-fold in a fresh MRS broth containing 0.05% L-
cysteine HCl at 37C, and the sub-cultured cells were allowed to grow anaerobically to reach
the mid-exponential phase corresponding to a concentration of approximately 10
8
cells/ml
and an optical density at 620 nm (OD620) of 0.6 to 0.7. The bacterial culture was then
107
Viability Assessment of Bifidobacteria by Flow Cytometry

centrifuged in a Mistral 3000 centrifuge (3,000 X g for 10 min at 4C), the pellet was washed
twice with anaerobic potassium phosphate buffer (50 mM; pH 7) containing 1mM
dithiothreitol (DTT). The cells were resuspended in the same buffer to obtain the desired
bacterial density. Cell suspensions of approximately 10
8
cells/ml were exposed to dBS,
consisting of 50% sodium cholate and 50% sodium deoxycholate (Sigma-Aldrich, Steinheim,
Germany) to a final concentration of 0.05, 0.1, 0.2, 0.25 and 0.3 % (wt/vol) for 10 min at
37C in anaerobic potassium phosphate buffer (50 mM; pH 7; containing 1mM DTT).
Untreated and heat-treated cells (70C for 30 min) served as control samples.
Probes
Carboxyfluorescein diacetate (cFDA), Propidium iodide (PI) and bis- (1,3-
dibutylbarbutiric acid) trimethine oxonol (DiBAC4(3)) were obtained from Molecular Probes
Europe BV, Leiden, The Netherlands.
DiBAC4
(3)
Staining.
A stock solution (1 mM) of the membrane potential probe DiBAC4(3) was made up
in dimethyl sulfoxide and kept at -20C. A working solution of 250 M was prepared in
ethanol and stored at 4C. The staining buffer contained 0.06 M Na2HPO4 and 0.06 M
NaH2PO4, mixed to produce a solution of pH 7 and supplemented with 5 mM KCl, 130 mM
NaCl, 1.3 mM CaCl2, 0.6 mM MgCl2 and 10 mM glucose (24). Samples were diluted in the
above mentioned buffer to approximately 10
6
to 10
7
cells/ml and were incubated for 4 min at
37C in the presence of 1 M DiBAC4(3). When appropriate 15 M carbonyl cyanide m-
chlorophenylhydrazone (CCCP) was added as a depolarizing agent. Heat-treated cells (70C
for 30 min) were used as a positive control for DiBAC4(3) staining.
PI-staining.
PI was supplied by the manufacturer as a 1-mg/ml solution in distilled water and was
used as the working solution and stored in the refrigerator in the dark. Ten l of each sample
was added to 985 l anaerobic potassium phosphate buffer (50 mM; pH 7; containing 1 mM
DTT) in the presence of 5 l of PI. The mixture was incubated for 15 min at 37C in a water
bath to allow staining of the cells. Samples were kept in the dark on ice and used within 1h
for FCM analysis.
cFDA staining
Stock solution (10 mM) of cFDA was prepared by dissolving 4.6 mg/ml in acetone and
stored at -20C in the dark. The stock solution was further diluted in acetone to 1 mM and
served as the working solution. Samples containing approximately 10
6
to 10
7
cells/ml were
incubated in anaerobic potassium phosphate buffer (50 mM; pH 7; containing 1 mM DTT)
108
Chapter 6
in the presence of 10 M cFDA for 30 min at 37C in a water bath. Stained samples were
kept on ice in the dark not longer than 1h, until flow cytometry analysis was performed.
Double staining
When dual labeling was performed, the same dye concentrations and incubation times,
described above, were used. Mixtures of heat-killed (70C for 30 min) and freshly harvested
cells were stained with cFDA and PI both in single staining and in multistaining assays. The
mixed cultures along with the unstained ones served as controls to set the flow cytometer
detectors and compensation. Cells electroporated in the presence of PI and subsequently
stained with cFDA were used as a control for the double stained cells. Overnight cultures of
B. lactis and B. adolescentis were used to inoculate fresh MRS supplemented with 0.5%
cysteine-HCl and the suspensions were incubated anaerobically at 37C for 3 to 4 h, until an
OD620 of approximately 0.6 to 0.7 was reached. The cells were then harvested by
centrifugation and washed twice with 1 mM HEPES buffer supplemented with 0.5 M
sucrose, and the pellets were resuspended in the same buffer. PI (5 M) was added to 200 L
bacterial suspensions (OD620= 10) in a precooled Gene Pulser disposable cuvette (inter-
electrode distance 0.2 cm; Bio-Rad). An electrical pulse of 1, 1.2, 1.4, 1.8 and 2.0 kv was
delivered with a Gene Pulser apparatus (Bio-Rad) by using the 25 F capacitor and setting the
pulse collector at 200 parallel resistance. Subsequently, the bacteria were diluted with 800l
MRS containing 0.05 % L- cysteine-HCl and incubated anaerobically for an additional 2 h
for recovery. Finally, the PI-labeled cells were centrifuged and washed twice with 50 mM
anaerobic potassium phosphate buffer (50 mM; pH 7; containing 1mM DTT) and stained
with cFDA as described above. These cells were used to adjust the FCM detectors and check
for PI toxicity after plating the sample onto MRS agar plates.
Flow cytometry analysis
Samples were analyzed with a FACSCalibur flow cytometer (Beckton Dickinson
Immunocytometry Systems, San Jose, Calif.) equipped with an air-cooled argon ion laser
emitting 15 mW of blue light at 488 nm and with the standard filter set-up. The Side scatter
signal was used as trigger signal. The green fluorescence from cF- and DiBAC4(3)-stained cells
was detected through a 530-nm, 30-nm-bandwidth band-pass filter (FL1 channel) and the red
fluorescence of PI signal was collected in the FL3 channel (>600 nm long pass filter). FACSFlow
solution (Beckton Dickinson) was used as sheath fluid. All bacterial analyses were performed at
the low rate settings (12 l/min), and the sample concentration was adjusted to keep the count
lower than 1,000 events/s. Data were collected in list mode as pulse height signals (4 decades in
logarithmic scale each) and analyzed using the computer program Windows Multiple Document
Interface (WinMDI, Joseph Totter, Salk Institute for Biological Studies, La Jolla, Calif.; available
at http//:facs.Scripps.edu/software.html). The machine was checked weekly for alignments
using 0.7 m green yellow fluorescent beads (Polyscience, Eppelheim, Germany).
109

Cell sorting.
B. lactis cultures were first exposed at 0.1% dBS for 10 min. at 37C, and then
simultaneously stained with 10 M/ml cFDA and 5g/ml PI, as described above. Cells were
analyzed by FCM, and sort gates were defined on an FL1-versus-FL3 dot plot of cFDA- and PI-
stained cells. The sorter was set to single-cell mode, and sorted cells were collected in 50-ml sterile
Greiner tube. Sorting was stopped after 2 min, which corresponded to the acquisition of 16,000
to 20,000 events for the untreated samples and 4,000 to 5,000 events for stressed cells. Filter-
sterilized phosphate buffered saline; pH 7 was used as sheath fluid. To determine the purity and
the recovery rate of sorted cells from the defined gates, the samples were reanalyzed in the FCM.
The sorted cells were centrifuged at 4,000 X g for 20 min, the supernatant was carefully removed,
and approximately 0.5 ml was left in the bottom of the tube. To this remaining volume, which
contained the sorted cells, 1 ml MRS containing 0.05% L-cysteine-HCl was added and the tubes
were then incubated for 2 to 3 h at 37C in the anaerobic chamber. Afterwards, the samples were
plated anaerobically onto MRS agar plates containing 0.05% L-cysteine-HCl and incubated at
37 for 72 h in anaerobic jars containing the Oxoid Gas Pack Anaerobic system.
Plate counts.
Prior to plating, samples were washed twice in anaerobic potassium phosphate buffer and
then diluted with saline solution (0.8%) containing 1 g peptone (Oxoid) and 0.05% cysteine-
HCl. Portions (100 l) of the appropriate dilutions were spread onto MRS supplemented with
0.05% HCl-cysteine and containing 1.5-% agar under anaerobic conditions. The plates were
incubated in anaerobic jars containing the Oxoid Gas Pack Anaerobic system for 3 days at 37C.
6.3. Results
Single staining and viability assessment
Membrane potential
Distinct DiBAC4(3)-fluorescence distributions of viable and dead bifidobacteria cells
could be measured in mixed cultures of heat killed (70C for 30 min) and freshly harvested cells
(fig.1A and B). Upon addition of 15 M of CCCP to freshly harvest cells, the green fluorescence
increased significantly as compared to the untreated cells indicating, that indeed DiBAC4(3)
fluorescence is a membrane potential-dependent dye (fig. 1C). Furthermore, the dual-parameter
dot plots show that DiBAC4(3) fluorescence and side scatter signal are closely correlated for all
cells suggesting that DiBAC4(3) fluorescence is influenced by cell size. Thus, a new parameter
proportional to the log ratio of the green-fluorescence was calculated for each cell using the
FSCPress software (Ray Hiks, Department of Medicine, University of Cambridge, United
Kingdom K; available at http://www.fcspress.com).
110
Chapter 6
Figure 1. Dual-parameter dot plot of the side scatter intensity (SSC) versus DiBAC4(3)-fluorescence of
B. lactis. Untreated cells (A), cells incubated with 15M/ml CCCP (B), and cells heat killed at 70C for
30 min (C) were stained with 1M DiBAC4(3) and analysed by FCM . Data show the effect of the
ionophore CCCP on membrane potential of B. lactis and a high correlation between DiBAC4(3)
fluorescence and side scatter signal. Control cells are gated on region R1 and display little green
fluorescence, while cells gated on region R2 show a marked green fluorescence increase as a result of the
addition of 15 M CCCP or heat treatment (70C for 30 min).
Figure 2. Fluorescence histogram overlays of B. lactis DSM 10140 (A) and B. adolescentis DSM 20083 (B)
stained with 1 M DiBAC4(3) for 4 min at 37C. The x axis represents the corrected DiBAC4(3)
fluorescence for the cell size, obtained by calculating the log ratio of the green fluorescence of DiBAC4(3)
to the side scatter intensity. Results are shown for control untreated cells (a), for cells exposed to 0.05%
dBS (b), 0.1% (c), 0.2% (d), 0.25% (e) and heat-treated at 70C for 30 min (f ).
The ratio derived from the oxonol fluorescence allowed a clear discrimination between
the different ratio distributions for each treatment for both strains (Fig.2). The median
fluorescence intensity of treated cells was shifted towards higher channel numbers for both
strains as the concentrations of bile salts increased, revealing a gradual depolarization of the cell
population related to cell death. Exposure to 0.05, 0.1 and 0.2% dBS for 10 min, resulted in
populations with a membrane potential intermediate between those observed for intact and
heat-treated cells for both strains. When cells of B. lcatis were exposed to 0.25 % dBS,
111
DiBAC4(3) staining revealed a fully depolarized (dead) population with a ratio similar to the
heat-cells (Fig. 2A), while for B. adolescentis the ratio was higher in heat-treated cells than in
cells treated with 0.25% dBS (Fig.2B).
Membrane integrity
PI is commonly used as a cell death marker since it is excluded by intact plasma
membranes; thus, the fluorescence conferred by the probe is generally associated with cells that
have lost their membrane integrity. When bifidobacterium cells were heat-treated (70C for 30
min), the whole population was permeable to PI. The red-fluorescent events were detected above
channel values of 258 and 520 of the FL3 detector for B. lactis and B. adolescentis respectively, while
a small fraction of the untreated cells (5 to 8%) was PI-positive. Following bile salt treatment, a
second high peak appeared above channel 255 or 520 (PI-positive subpopulation), representing the
damaged, PI-permeable cells of B. lactis and B. adolescentis respectively as shown in Fig.3. At dBS
concentrations of 0, 0.05, 0.1, and 0.2 % the percentage of PI-permeable cells was 5, 21, 74 to
98%, repectively for B. adolescentis and 7, 23, 56 and 91%, respectively for B. lactis.
Figure 3: Fluorescence histograms of B. lactis and B. adolescentis stained with 5 g/ml PI or 10 M/ ml cFDA
following exposure to different concentrations of deconjugated bile salts.
112
Chapter 6
Esterase activity
The percentage of cF-stained cells was used to estimate the viability of dBS-treated cells.
Retention of cF by the cells indicates enzyme activity as well as membrane integrity. Figure 3
shows fluorescence histograms of dBS-treated B. lactis and B. adolescentis cells stained with
cFDA. A clear separation between stained and unstained subpopulations was observed with all
treatments and for both strains. With increasing concentrations of dBS, the fraction with a
fluorescent intensity above channel 255 of the FL1 detector decreased for B. lactis and
B. adolescentis, and a cF-negative subpopulation could be observed. For B. lactis the cF-positive
population decreased from 95, 73, and 40% to reach 1% after exposure to 0, 0.05, 0.1, and
0.2% dBS, respectively. At the same concentrations of dBS, B. adolescentis showed a lower
percentage of cF-stained cells than B. lactis.
Figure 4 shows the numbers of permeable (damaged) cells (PI positive), depolarized cells
(DiBAC4(3) positive), and intact (active) cells (cF positive) and the number of CFU per
milliliter for both strains. The number of PI-stained bacteria correlated highly with the number
of DiBAC4(3)-stained bacteria and displayed an inverse relation with the percentage of survival
of the cells as determined by plate counts. The percentage of cF-stained cells showed the same
trend as the results obtained by plate counts; however, it gave a higher estimation of the
viability of B. lactis and B. adolescentis, by 20 to 30%, respectively, for all treatments.
Figure 4. Viability assessment of bile salt stressed cells of B. lactis DSM 10140 (A), and B. adolescentis DSM
20083 (B) by FCM and plate counts. Cells were harvested at the mid-exponential phase and then exposed
to 0, 0.05, 0.1, 0.2 and 0.25% deconjugated bile salts for 10 min at 37C in anaerobic potassium phosphate
buffer (50 mM , pH 7 and 1mM DTT). Results are expressed as percentage of cF-stained cells (), PI-
stained cells (s ), oxonol-stained cells (n) and the percentage survival as determined by plate counts ( ).
113
Multiparametric flow cytometry.
In order to validate the multiparameter assay, it was necessary to establish adequate
controls. Unstained, cF-stained, PI-stained, and cF and PI double-stained cells served as
controls by which to set the FCM detectors and the compensation settings. The quadrants of
the dot plots were set so that the unstained cells appeared in the lower left quadrant. Cultures
of both strains were exposed to 0, 0.05, 0.1, 0.2, or 0.25% dBS and then simultaneously
stained with cFDA and PI. The dual-parameter dot plots of Fig. 5 indicate the existence of
three main subpopulations of bile salt-treated B. lactis and show dynamic changes with
increases in bile salt concentration. These subpopulations are identified based on their
differential staining characteristics with PI and cF. These subpopulations consisted of cells that
stained only with cF (upper left quadrant; cF
+
PI
-
), cells that stained only with PI (lower right
quadrant; cF
-
PI
+
), and a double-stained population (upper right quadrant; cF
+
PI
+
). A
fourth population that did not exceed 3% appeared in the lower left quadrant (cF
-
PI
-
); it
most likely represented debris or lysed cells. The cF-stained subpopulation (cF
+
PI
-
) decreased
with increasing bile salt concentrations and accounted for 92, 67, 38, 4, and 3% of the total
population after exposure to 0, 0.05, 0.1, 0.2, and 0.25% dBS, respectively. The cF
+
PI
-
cells
corresponded to the number of CFU and thus were scored as the viable subpopulation (Fig.
6A). The fraction that stained only with PI (cF
-
PI
+
) increased from 2, 13, 31, and 84% to
reach 90% after exposure to 0, 0.05, 0.1, 0.2, and 0.25% dBS, respectively, and was scored as
the dead population. The percent double-stained B. lactis cells (cF
+
PI
+
) showed a subtle
fluctuation during the course of the stress exposure. This fraction constituted 5, 16, 28, 11,
and 9% of the total population after treatment with 0, 0.05, 0.1, 0.2, and 0.25% dBS,
respectively. Similar trends were obtained with B. adolescentis (Fig. 6B), showing the
succession of the three physiological states within the stressed population of B. adolescentis.
Cell Sorting
To ascertain that double-positive (cF
+
PI
+
) cells corresponded to bacteria with only
partially damaged cell membranes and therefore represented the injured cells; we sorted cells
from different gates. The gates were defined in the PI fluorescence-versus-cF fluorescence dot
plots as follows: cells from gate 1 (G1) were stained with cF only; cells from gate 2 (G2) were
double stained; cells from gate 3 (G3) were stained with PI. The cell fractions sorted from
each gate were mapped by FCM reanalysis in those regions that had been defined by the
respective sort gates. The FCM reanalysis demonstrated high purity (95%), while the recovery
rate did not exceed 73 to 77%, as shown in Table 1. Higher sorting recovery resulted in both
lower purity and reduced growth of the cells (data not shown). Sorting of bacteria from
different functional stages and subsequent incubation in MRS broth revealed that all types of
cells except the PI
+
cF
-
cells are capable of resuming growth on agar plates. A significant
number of the double-stained, injured subpopulation (40%) regained the capacity for cell
proliferation (Table 1).
114
Chapter 6
Figure 5. Multiparametric dot plots of B. lactis DSM 10140 representing PI-fluorescence versus cF-
fluorescence. Cultures were exposed for 10 min to 0, 0.05, 0.1, 0.2 and 0.25 % dBS in anaerobic
potassium phosphate buffer (50 mM, pH 7, containing 1mM DTT) for 10 min at 37C, and for 30
min at 70C.
115
Table 1. Results of sorting and recovery of B. lactis cells treated with 0.1% dBS for 10 min at 37C
a
.
a
Following exposure to bile salts, cells were stained simultaneously with 10 M cFDA and 5 g PI and
then sorted into sterile tubes.
b
Sort gates of differentially stained target population .
c
Calculated as [number of total sorted events/number of events reanalyzed after sorting)] X 100
d
Calculated as [number of colony forming units of the total sorted sample/number of sorted events after
recovery] X 100.
6.4. Discussion
In this paper, we report on the use of PI, an exclusion dye, cFDA, an intracellular
enzymatic stain, and DiBAC4(3), a membrane potential probe, used in single and/or
multiparameter FCM analysis to assess the viability of B. lactis and B. adolescentis during bile salt
stress. Clear discrimination between viable intact, permeabilized, and depolarized cells was
achieved by monitoring the extents to which they retained cF and accumulated PI or DiBAC4(3)
during bile salt exposure. The results of the single-staining approach provide extensive evidence
that the three fluorescent probes do reflect the responses of B. lactis and B. adolescentis to the
damaging effects of bile salts and thus reflect the degree of cell viability (3, 6, 12, 19).
Examining the membrane potential provided an additional means of characterizing the
physiological status of bile salt-stressed cells. Correction of DiBAC
4
(3) fluorescence for bacterial
size variations allowed for better discrimination between viable cells and cells that were depolarized
(dead) as a result of bile salt stress than the initial DiBAC
4
(3) fluorescence distribution. We found
that the ratio was higher for heat-killed B. adolescentis cells than for CCCP-treated cells or cells
treated with a lethal dose of dBS, e.g., 0.25% (Fig. 2B). These results suggest that the fluorescence
of DiBAC4(3) was affected differently in cells treated with dBS and CCCP than in heat-treated cells
(24, 25, 31). The ratiometric method used in our study will offer an accurate approach to
measuring bacterial membrane potential and assessing the viability of gram-positive bacteria.
116
Chapter 6
Figure 6. Viability assessment of B. lactis DSM 10140 (A), and B. adolescentis DSM 20083 (B) using
multiparametric flow cytometry and the plate count method. Results illustrate the different
physiological states of bile salts stressed cells consisting of viable active cells stained only with cF (n),
injured cells stained with both cF and PI ( ), dead cell stained only with PI (s ) and culturable cells
determined by the plate count method ().
The validity of cFDA for reflecting viability during stress has been reported for
lactic acid bacteria and a number of other microbes, and high correlations between
plate counts and FCM counts have been obtained (4, 8, 12, 27). In this study we showed
that, for both strains and for all stress conditions, single staining with cFDA always
gave a higher estimation of the number of viable cells than plate counts. This discrepancy was
higher with B. adolescentis, possibly because this strain is more oxygen sensitive than
B. lactis, and probably plating exerted an additional stress on this microorganism (22).
It has been reported that bifidobacterium cells when exposed to oxygen could
ferment carbohydrates, even though they could not increase in number by cell division
due to oxygen toxicity (33). The difference observed between the FCM and plate count
results suggests the presence in the stressed population of cells that could maintain
cell metabolic activity, as determined by the fluorescent dyes, yet were not able
to form colonies on agar plates. Indeed, stressed and starved cells can enter a
nonculturable state, most likely due to "sublethal-injury" mechanisms including
damage to the cell membrane, protein, and/or DNA, and can recover by repairing or replacing
those damaged molecules (1, 14, 16). Temporary nonculturability has been reported
for starved Micrococcus luteus (15) and Escherichia coli (27) cells by use of FCM and
cell sorting.
117
Multiparametric FCM has been successfully used to assess physiological heterogeneity
within bacterial populations during bacterial fermentations (10, 11) and to gain insight into
the mechanism of action of antibiotics on Staphylococcus aureus and M. luteus (25). Our
multiparameter FCM results clearly illustrated the succession of cell changes that occurred in
a bile salt-stressed bifidobacterium population and revealed physiological heterogeneity within
the cell population (23). Cell sorting confirmed that the bile salt-treated cell populations
contained a mixture of viable cells, dead cells, and an injured (stressed) subpopulation stained
with PI and cF. Regrowth of the injured cells following sorting confirmed that a fraction of
the stressed cells adopted a latent state in which they could not reproduce but could be
induced to a physiologically active state after recovery (1, 6, 15, 16). The percent recovery of
injured cells (40%) and the percent viable cells (47%) did not reach higher values presumably
because of the additional stress caused by sorting and plating (11, 23). The electroporated PI-
and cF-stained bifidobacterium cells were able to grow on agar plates (data not shown),
showing that the concentration of PI (5 g/ml) used in our experiment was not toxic for
bifidobacteria. Multiparametric results show that cell permeability as monitored by PI is a
sensitive marker of cell damage, yet it is a poor indicator of cell death of stressed bacteria (14,
25, 31). Therefore, we assume that bile salts induced sublethal injury within the
bifidobaterium 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. The precise mechanism of action of
bile salts is unknown, but these compounds act as detergents for the digestion of fats in the
intestinal tract and are reported to have inhibitory effects on a number of bacteria. Recently,
it was shown that bile acids induce expression of specific stress response genes in E. coli,
possibly in response to membrane perturbation, oxidative stress, or DNA damage (2). Bile
salt-stressed B. adolescentis NCC481 showed a remarkably increased resistance to lethal
concentrations of bile salts, most likely through induction of a mechanism allowing the cells
to build up a protection against the solubilization of their membrane proteins (30).
In this study we showed that multiparameter FCM combined with cell sorting
provides a way not only to distinguish between live and dead cells but also to discriminate
between different physiological states of a stressed bifidobacterium population. Moreover, it
clearly illustrates the dynamics in the physiology of microbial populations during dBS
treatment. In conclusion, this method may provide a novel tool for assessing the viability and
stability of bacteria during the processing and storage of probiotic products. Furthermore, we
aim to use this approach, along with molecular techniques such as fluorescent in situ
hybridization, to analyze the activity and stability of these microorganisms within the complex
ecosystem of the gastrointestinal tract.
118
Chapter 6
6.5. References
20. Barer, M. R., and C.R. Harwood. 1999. Bacterial viability and culturability. Advan.
Microbial. Physiol, p 94-137. In Poole, R.K. (ed.), Academic Press.
2. Breeuwer, P., and T. Abee. 2000. Assessment of viability of microorganisms employing
fluorescence techniques. Int. J. Food. Microbiol. 55:193-200
3. Bernstein, C., H. Bernstein, C. M. Payen, S. E. Beard, and J. Schneider. 1999. Bile salt
activation of stress response promoters in Escherichia coli. Current. Microbiol. 39:68-72.
4. Bunthof, C.J., S van Schalkwijk, W. Meijer, T. Abee, and J. Hugenholtz. 2001.
Fluorescent method for monitoring cheese starter permeabilization and lysis. Appl.
5. Bunthof, C. J., K. Bloemen, P. Breeuwer, F. M. Rombouts, and T. Abee. Flow cytometry
assessment of the viability of lactic acid bacteria. 2001. Appl. Environ. Microbiol.
67:2326-2335
6. Davey, H. M., D. H. Welchart, D. B. Douglas, and A. S. Kaprelyants. 1999. Estimation
of microbial viability using flow cytometry. p11.3.11-11.3.20. In P. Robinson,
Z. Darzynkiewics, P. Dean, A. Orfao, P. Rabinovitch, H. Tank, and L. Wheeless (ed.),
Current protocols in cytometry.
7. Davey, H. M., and D. B. Kell. 1996. Flow cytometry and cell sorting of heterogeneous
microbial populations: the importance of single-cell analyses. Microbiol. Rev. 60:641-
696.
8. Diaper J. P., and C. Edwards. 1994. The use of fluorogenic esters to detect viable bacteria
by flow cytometry. J. Appl. Bacteriol. 77:221-228.
9. Ericsson, M., D. Hanstrop. P. Hagberg,. J. Enger, and T. Nystrm. 2000. Sorting out
bacterial viability with optical tweezers. J. Bacteriol. 182:551-555.
10. Hewitt, C. J., and G. Nebe-von-Caron. 2001. An industrial application of multiparameter
flow cytometry: assessment of cell physiology state and its application to the study of
microbial fermentations. Cytometry 44:179-187.
11. Hewitt, C. J., G. Nebe-von-Caron, A. W. Niennow, and C. M. McFarlane. 1999. The
use of multi-parameter flow cytometry to compare the physiological response of
Escherichia coli W3110 to glucose limitation during batch, fed-batch and continuous
culture cultivations. J. Biotecnol. 75:251-264.
12. Humphreys, M. J., R. Allman, and D. Lloyd. 1994. Determination of the viability of
Trichomonas vaginalis using flow cytometry. Cytometry 15:343-348.
119

13. Jepras, R. I., F. E. Paul, and M. J. Wilkinson. 1997. Rapid assessment of antibiotic effects
of Escherichia coli by bis- (1,3-dibutylbarbituric acid) trimethine oxonol and flow
cytometry. Antimicrob. Agents Chemother. 41:2001-2005.
14. Joux, F., and P. Lebaron. 2000. Use of fluorescent probes to assess physiological functions
of bacteria at single-cell level. Microbes. Infect. 2:1523-1535.
15. Kaprelyants, A. S., G. V. Mukamolova, H. M. Davey and D. B. Kell. 1996. Quantitative
analysis of the physiological heterogeneity within starved cultures of Micrococcus luteus by
flow cytometry and cell sorting. Appl. Environ. Microbiol. 62:1311-1316.
16. Kell, D. B., A. S. Kaprelyants, D. H. Weichart, C. R. Harwood, and M. R. Barer. 1998.
Viability and activity in readily culturable bacteria: a review and discussion of practical
issues. Antonie van Leeuwenhoek 73:169-187.
17. Klaenhammer, T, R., and M. J. Kullen. 1999. Selection and design of probiotics. Int. J.
Food. Micrbiol. 50:45-57.
18. Lopez-Amaros, R., J. Comas, and J. Vives-Rego. 1995. Flow cytometry assessment of
Escherichia coli and Salmonella typhimurium starvation-survival in seawater using
rhodamine 123, propidium iodide, and oxonol. Appl. Environ. Microbiol. 61:2521-2526.
19. Lopez-Amaros, R., S. Castel, J. Comas-Riu, and J. Vives-Rego. 1997. Assessment of E.coli
and Salmonella viability and starvation by confocal laser microscopy and flow cytometry
using rhodamine 123, DiBAC4(3), propidium iodide, and CTC. Cytometry 29:298-605.
20. Mason, D.J., R. Lopez-Amoros, R. Allman, J.M. Strak, and D. Lloyd. 1995. The ability
of membrane potential dyes and calcafluor white to distinguish between viable and non-
viable bacteria. J. App. Bacteriol. 78:309-315.
21. Mason, D. J., R. Allaman, J. M. Stark, and D. Lloyds. 1994. Rapid estimation of
bacterial antibiotic susceptibility with flow cytometry. J. Microsc. 176:8-16.
22. Meile, L., W. Ludwig, U. Rueger, C. Gut, P. Kaufmann, G. Dasen, S. Wenger, and M.
Teuber. 1997. Bifidobacterium lactis sp. nov., a moderately oxygen tolerant species isolated
from fermented milk. Sys. Appl. Microbiol. 20:56-64.
23. Nebe-von-Caron., G., P. J. Stephens, C. H. Hewitt, J. R. Powell and R. A. Badley.
2000. Analysis of bacterial function by multi-color fluorescence flow cytometry and single
cell sorting. J. Microbiol. Methods. 42:97-114.
24. Novo, D., N. G. Perlmutter, R. H. Hunt, and H. M. Shapiro. 1999. Accurate flow
cytometric membrane potential measurement in bacteria using diethyloxacarbocyanine
and ratiometric technique. Cytometry 35:55-63.
25. Novo, D. J, N. G. Perlmutter, R. H. Hunt, and H. M. Shapiro. 2000. Multiparametric
flow cytometric analysis of antibiotic effects on membrane potential, membrane
120
Chapter 6
permeability and bacterial counts of staphylococcus aureus and Micrococcus luteus.
Antimicrob. Agents Chemother. 44:827-834.
26. Ouwehand, A. C., P. V. Kirjavainen, C. Shortt, and S.Salminen. 1999. Probiotics:
mechanisms and established effects. Int. Dairy. J. 9:43-52
27. Porter, J., C. Edwards, and R. W. Pickup. 1995. Rapid assessment of physiological status
in Escherichia coli using fluorescent probes. J. Appl. Microbiol. 79:399-408.
28. Porter, J., D. Deere, R. Pickup. and C. Edwards. 1996. Fluorescent probes and flow
cytometry: new insights into environmental bacteriology. Cytometry 23: 1-96.
29. Ritz, M., J. L. Tholozan, M. Fedeeighi, and M. F. Pilet. 2001. Morphological and
physiological characterization of Listeria monocytobenes subjected to high hydrostatic
pressure. Appl. Environ. Microbiol. 67:2240-2247.
30. Schmidt, G., and R. Zink. 2000. Basic features of the stress response in three species of
bifidobacteria: B. longum, B. adolescentis, and B. breve. Int. J. Food. Microbiol. 55:41-45.
31. Shapiro, H. M. 2000. Microbial analysis at the single-cell level: tasks and techniques.
J. Microbiol. Methods. 42:3-16.
32. Shapiro, H. M. 1995. Practical flow cytometry. 3
d
ed. Wiley-Liss, Inc. New York.
33. Shimamura, S., F.Abe, N. Ishibashi, H. miyakawa, T. Yaeshima, T. Araya, and M.
Tomita. 1992. Relationship between oxygen sensitivity and oxygen metabolism of
bifidobacterium species. J. Dairy. Sci. 75:3296-3306.
34. Sincock, S. A., and J. P. Robinson. 2001. Flow cytometric analysis of microorganisms.
Methods Cell. Biol. 64:511-537.
35. Suller, M. T. E., and D. Lloyd. 1999. Fluorescence monitoring of antiobiotic-induced
bacterial damage using flow cytometry. Cytometry 35:235-241.
36. Tannock, G.W. 1999. Microecology of Lactobacilli and Bifidobacteria inhabiting the
digestive tract: essential knowledge for successful probiotic research, p 17-31. In: Hanson,
L. A, and Yolken, R. H. (ed.), Probiotics, other nutritional factors, and intestinal
microflora. Nestl Nutrition Services.
37. Wallner, G., B.Fuchs, S. Spring, W. Beisker, and R. Amann. 1997. Flow sorting of
38. Ward. D., R. Weller, and M. M. Bateson. 1990. 16S rRNA sequences reveal numerous
uncultured microorganisms in a natural community. Nature 345:63-65.
121

.
Chapter 7
GENETIC DIVERSITY OF VIABLE, INJURED AND
DEAD FECAL BACTERIA ASSESSED BY FLUORESCENCE
ACTIVATED CELL SORTING AND 16S rRNA
GENE ANALYSIS
Kaouther Ben Amor, Hans Heilig, Hauke Smidt, Elaine E. Vaughan, Tjakko Abee and
Willem M. de Vos
Abstract: A novel approach combining a flow cytometric in situ viability assay with
16S ribosomal RNA gene analysis was used to study the relation between diversity
and activity of the fecal microbiota. Simultaneous staining with propidium iodide
(PI) and SYTO BC provided a clear discrimination between intact cells (49 %),
injured or damaged cells (19%), and dead cells (32 %). The three subpopulations
were sorted and characterized by denaturing gradient gel electrophoresis (DGGE) of
16S rRNA gene amplicons obtained from the total and bifidobacterial communities.
This analysis revealed that not only the total community but also the distinct
subpopulations are characteristic for each individual. Cloning and sequencing of the
dominant bands of the DGGE patterns showed that most of clones retrieved from
the live, injured and dead fractions belonged to Clostridium coccoides, Clostridium
leptum and Bacteroides. We found that some of the butyrate-producing related
bacteria such as Eubacterium rectale and Eubacterium hallii were obviously viable at
the time of sampling. However, amplicons affiliated with Bacteroides, Ruminococcus
obeum-, Eubacterium biforme-like bacteria and Butyrivibrio crossotus were especially
obtained from the dead population. Furthermore, some bacterial clones
were recovered from all sorted fractions this was especially noticeable for the
Clostridium leptum cluster. The bifidobacterial phylotypes identified in total samples
and sorted fractions were assigned to Bifidobacterium adolescentis, Bifidobacterium
longum, Bifidobacterium infantis, Bifidobacterium pseudocatenulatum, and
Bifidobacterium bifidum. Phylogenetic analysis of the live, dead and injured cells
revealed a remarkable physiological heterogeneity within these bacterial populations,
B. longum - B. infantis were retrieved from all sorted fractions, while B. adolescentis
was mostly recovered from the sorted-dead fraction. The combination of FCM, cell
sorting and subsequent 16S rRNA gene analysis that we have used here for the first
time for fecal microbiota provided relevant ecological information related to the
diversity and activity of the fecal microbiota in situ and highlights the physiological
heterogeneity of this complex ecosystem.
123
7.1 Introduction
The human gastrointestinal (GI)-tract harbors a complex and dynamic microbial
ecosystem where relationships amongst bacteria, and between these and the host are significant
(18). It is well recognized that this GI-tract microbiota play an important role in the gut
physiology and accordingly has a great influence on the health of the host. Indeed the large
number of bacterial species, estimated to be more than 1,000 (45), and the high number of
microorganisms, more than 10
13
cells that inhabit the GI-tract represent an enormous
biological potential for metabolic conversions (17). These include production of short chain
fatty acids, vitamin synthesis, deconjugation of bile salts, and degradation of mucin (11).
During the last decade, the application of cultivation-independent molecular techniques based
on 16S rRNA genes has provided new insights into the microbial ecology of the GI-tract.
Fingerprinting techniques, such as PCR-denaturing gradient gel electrophoresis (PCR-DGGE)
(16, 38, 49), and in situ hybridization (FISH) have been applied to both feces and biopsy
samples (13, 36, 50, 51). The results of such studies have advanced our knowledge of the
intestinal microbiota by unraveling the complexity of the ecosystem and providing insight in
the establishment and succession of the bacterial community within the host (12, 14, 42, 49).
However, little is known about the relation between the diversity and metabolic activity
of the GI-tract microbiota. Understanding this relation is important since not all members of
the ecosystem contribute similarly to the physiological function that can be influenced by
factors, such as nutritional changes, pathogens, stress, and drug intake (17). Therefore,
comprehensive in situ analytical approaches that provide simultaneous information about the
identity and activity of a microbial cell in its natural environment are essential. From an
ecological point of view, at least three categories of cells within microbial communities can be
distinguished (6): (i) viable and active cells that play a functional role and participate in the
production of biomass at the time of sampling, (ii) viable and inactive cells (dormant or
injured) that might play a role in the future, and (iii) dead cells that play no active or potential
role in the cycling of chemical elements and represent only particulate organic carbon. In the
GI-tract ecosystem, the latter category may still have some functions as shown by studies with
dead probiotic bacteria (35).
Several techniques have been used to identify metabolically active bacteria in complex
natural ecosystems. These methods address different physiological functions, such as DNA
synthesis, enzyme activities, membrane potential, and membrane permeability as viability
markers (9, 33, 40). With the advent of flow cytometry (FCM), it became clear that even pure
cultures derived from a single cell are far from homogeneous (10, 19, 30, 39) Moreover, the
recent developments in fluorescent technology allowed the use of a wide set of probes to analyze
the physiological activity of stressed or starved bacteria at a single cell level (9, 15, 40, 47). The
fluorogenic substrate, carboxyfluorescein diacetate (cFDA), is often used for the detection of
enzymatically active or viable cells by FCM. Furthermore, the dye exclusion assay based on the
124
Chapter 7
use of membrane-impermeant stains such as propidium iodide (PI), SYTOX Green, TOTO-1
and TO-PRO-3, has been used extensively as membrane damage or as cell death indicator.
Membrane-permeant dyes, such as SYTO 9, SYTO13, SYBR I Green, that cross the
cytoplasmic membranes of both dead and living cells can be used in combination with the
above mentioned dyes for a better resolution between live and dead cells. A so-called dead/live
kit, containing the impermeant dye PI and the permeant stain SYTO 9 is now widely used for
the assessment of bacterial viability (15). Recently, we described a FCM-based approach to
assess the viability of bifidobacteria, a common species of the GI-tract, during bile salt stress
using cFDA, oxonol and PI to monitor esterase activity, membrane integrity and membrane
potential, respectively (5). This multiparametric FCM method allowed to simultaneously
distinguish, within the stressed bifidobacteria population, three subsets of cells with different
physiological status: live, dead and injured or damaged cells.
In this study we report on the application of a viability assessment approach using FCM
to monitor the activity of human GI-tract microbiota using functional probes. Fecal samples
were subject to fluorescence activated cell sorting (FACS) followed by DGGE analysis of 16S
rRNA amplicons to obtain insight into the diversity of the different physiological fractions.
Cloning and sequencing of the most abundant DGGE bands from total, viable, dead and
injured cells was performed and phylogenetic affiliations of the different metabolic fractions
were assigned.
7.2 Materials and methods
Fecal samples
Fresh fecal samples were collected from 4 healthy adults (3 females, and 1 male, age 25-
55 years). These volunteers had not been subjected to any feeding trial, specific diet, or antibiotic
treatment for the last year. Samples were processed immediately after collection in an anaerobic
chamber with an atmosphere of 10% CO2, 10% H2 and 80% N2. To extract fecal microbial cells,
0.5 g of feces were suspended in 4.5 ml of anaerobic phosphate buffer saline (PBS), containing 1
mM dithiothreitol and homogenized by vortexing the fecal slurry for 3 min at high speed in the
presence of 5 to 10 glass beads of 3-m diameter. After centrifugation at 700 X g for 1 min, 1 ml
of supernatant was carefully recovered into an eppendorf tube and centrifuged at 6,000 X g for 3
min. The supernatant was discarded and the pellet was washed twice with 1ml anaerobic PBS.
The cell suspensions were sonicated twice for 20 sec in a water bath. The cells suspensions were
then diluted 10- to 100-fold and kept anaerobically on ice until further analysis.
Viability Staining
The following fluorescent probes were obtained from Molecular Probes BV, Leiden, The
Netherlands: cFDA, PI, SYTOBC, SYTO9, and SYTO13. To assess the viability of fecal bacteria
based on their membrane integrity, 1 ml of diluted fecal sample (1 x 10
8
to 5 x 10
8
cells) was
125
Genetic diversity of live, injured and dead fecal bacteria

incubated for 15 min at room temperature in the dark with 14.5 M/ml PI in the presence of
one of the SYTO dyes. The final concentrations of SYTO 9, SYTO 13, were 3.34 M, and 5
M, respectively; SYTO BC was used according to the manufacturers instructions. When
excited with a 488-nm argon laser, SYTO dyes will fluoresce in the green (maximum at 521 nm)
and PI in the red (maximum at 617 nm) wavelength ranges. The fluorogenic substrate cFDA
was used to monitor enzymatic activity as described previously (5). In short, 1 x 10
7
to 1 x 10
8
of fecal cells were incubated in the presence of 10 M cFDA for 30 min at 37C in a 50 mM
potassium buffer containing 1 mM dithiothreitol. The stained cells were then washed with
anaerobic potassium buffer. All labeled samples were kept on ice not longer than 1 h before
FCM analysis. All the staining steps were carried out under anaerobic conditions.
Flow cytometry and cell sorting
Flow cytometry was performed with a FACSCalibur (Becton Dickinson, San Jose,
California, USA), equipped with an air-cooled 15-mW Argon-ion laser, operating at 488 nm.
SYTO dyes and cFDA green fluorescence (FL1) was collected using a 530 30 bandpass filter;
the red fluorescence emitted from PI (FL3) was collected by a 650 13 bandpass filter. The data
were analyzed with the CellQuest software from Becton Dickinson; all parameters were
measured using logarithmic amplification. Each sample was acquired between 20 to 30 sec at
the low rate and the cell concentration was adjusted to keep a count of 500 to 600 events/s.
Control samples were used for the instrument settings (voltage of the detectors and the
compensation) and consisted of unlabeled fecal cells, heat killed fecal cells stained with PI (FL3)
and cells labeled with SYTO BC (FL1). CaliBRITE beads obtained from Becton Dickinson
were used to check the instrument sensitivity over time. Fecal cells were discriminated from
electronic noise using a double threshold set on both side scatter (SSC) and forward scatter
(FSC), with FSC set on E01 and SSC on 400V. All samples were sonicated (2 x 20 sec) and
thoroughly vortexed prior to FCM analysis. A dual dot plot FSC versus SSC in combination
with one parameter histogram representing the SYTO BC fluorescence (FL1) was used to back
gate fecal cells and distinguish them from the background. For total cell enumeration, samples
were incubated in the presence of 1 l SYTO BC for 5 min at room temperature. Unlabeled
beads of 6.0-m diameter provided from the Bacteria Counting Kit (Molecular probes BV,
Leiden, The Netherlands) were added to each sample at a final concentration of 106 beads/ml
sample and served as an internal standard to calibrate the sample volume. The ratio of SYTO BC
stained cells to the number of beads was used to calculate the absolute total cell counts. Data
were collected in list mode as pulse height signals (four decades in logarithmic scale each)
and analyzed using the computer program Windows Multiple Document Interface (WinMDI,
Joseph Totter, Salk Institute for Biological Studies, La Jolla, Calif.; available at
http//:facs.Scripps.edu/software.html).
For the sorting experiment, a FACSVantage flow cytometer (Becton Dickinson) was
used and sorting criteria were defined by drawing gates in two bivariate dot plots representing
126
Chapter 7
FSC versus SSC and red fluorescence (PI) versus green fluorescence (SYTO BC). Filtered,
autoclaved PBS (pH 7.2) was used as a sheath fluid, the pressure of sheath fluid was 45 PSI and
the nozzle size was 70 m. The drop drive frequency was set at 67 kHz/sec. Each event was
sorted in a sort envelop of 1.2 drops at a speed of 8 x 10
3
to 1 x 10
4
events /s. In order to achieve
high purity and recovery, the R mode was used to sort 1 x 10
6
to 5 x 10
6
cells from each
subpopulation into separate sterile polystyrene tubes. Sorted cells were then collected on 0.2-
m polycarbonate filters using a standard 25 mm glass fritted filtration base (Millipore BV,
Amsterdam, The Netherlands). The filters were stored at 20 C until further analysis.
Fluorescent in situ hybridization
The bifidobacteria specific probe Bif164 (24) and the general bacteria probe Eub338 (2)
were purchased from (MWG, Ebersgerg, Germany) and were mono-labeled at the 5 end with
Cy5 and Cy3, respectively. Hybridization was carried out as described previously (50). Briefly,
50 to 100 ml of ethanol-fixed fecal cell suspension was centrifuged at 9,000 X g for 3 min and
washed with 1 X PBS. The cell pellet was suspended in a volume of 40 ml hybridization buffer
(0.9 M NaCl, 20 mM Tris-HCl [pH 7.2], 0.1% [wt/vol] SDS) to which 5 l of the Cy5-labeled
Bif164 and 5 l Cy3-labeled Eub338 were added to a final concentration of 3 ng/l. The
samples were incubated at 50C for 16 h. Subsequently, the cells were centrifuged at 9,000 X g
for 3 min and resuspended in a washing solution (0.9 M NaCl and 20 mM Tris-HCl, pH 7.2)
and incubated for 20 min at 50C. Cells were finally pelleted and resuspended in 1 ml ice-cold
PBS and kept in the dark on ice until FCM analysis was performed within 1 h.
DNA isolation.
Bacterial DNA from fecal samples and sorted cells was isolated using the QIAamp DNA
stool mini kit (QIAgen, Hilden, Germany) with some modifications. In short, 0.220 g of feces
and the sorted filters were placed in different bead-beating tubes filled with 0.3 g of 0.1 mm
zirconia/silica beads and 1.4 ml of ASL solution from the stool mini kit was added. The tubes
were then agitated for 3 min at a speed of 5,000 rpm in a mini bead beater (Biospec Products,
Bartlesville, Oklohoma, USA). The protocol was then continued from step 3 as described by
the manufacturer. DNA samples were stored at -20C in a TE buffer.
PCR amplification
All primers used in this study are listed in Table 1. DNA isolated from total fecal
community and the sorted fractions was used as a template to amplify the V6 to V8 regions of
16S rRNA gene with primers 968-GC-f and 1401-r. PCRs were performed using a Taq DNA
polymerase kit from Life Technologies (Gaithersburg, Madison, USA). The reaction mixture
consisted of 20 mM Tris-HCl (pH 8.4), 3 mM MgCl2, 0.2 mM deoxynucleoside triphosphate
(dNTP), a 0.2 M concentration of each primer, 1.25 U of Taq polymerase, and 1 l of
appropriately diluted template DNA in a final volume of 50 l. Samples were amplified in a
127

PE Applied Biosytems GenAmp PCR system 9700 (Foster City, California, USA) using the
following program: pre-denaturation at 94C for 5 min; 35 cycles of denaturation 94C for 30
sec, annealing temperature of 56C for 20 sec, extension at 68C for 40 sec, and final extension
at 68C for 7 min. The integrity of the nucleic acids was checked visually after electrophoresis
on a 1.2-% agarose gel containing ethidium bromide.
Table 1: Primers used in this study.
Primer Sequence (5-3) Specificity Reference
1401-r CGGTGTGTACAAGACCC Bacterial 16S (32)
968-f AACGCGAAGAACCTTA Bacterial 16S (23)
Bact0011-f AGAGTTTGAT(C/T) (A/C)TGGCTCAG Bacterial 16S (20)
Lab0159-f GGAAACAG(A/G)TGCTAATACCG Lactobacillus 16S (16)
Lab0677-r CACCGCTACACATGGAG Lactobacillus 16S (16)
Univ0515-r ATCGTATTACCGCGGCTGCTGGCA Universal (23)
Lm26-f GATTCTGGCTCAGGATGAACG Bifidobacterium 16S (21)
Lm3-r CGGGTGCTICCCACTTTCATG Bifidobacterium 16S (21)
Bif164-f GGGTGGTAATGCCGGATG Bifidobacterium 16S (22)
Bif662-r CCACCGTTACACCGGGAA Bifidobacterium 16S (22)
GC-clamp CGCCGGGGGCGCGCCCCGGGCGGGGCG
GGGGCACGGGGGG (29)
T7 TAATACGACTCACTATAGG Sequencing Promega
Sp6 GATTTAGGTGACACTATAG Sequencing Promega
To investigate the diversity of Bifidobacterium group, genus-specific PCR was
performed using a nested PCR using lm3 and lm26 primers (Table 1) to amplify most of the
16S rRNA gene. Following purification of the PCR products by QIAgen kit (QIAgen, Hiden,
Germany), a second PCR was carried out with 10-fold diluted samples as DNA template with
16S rRNA gene-targeted primers Bif164-f and Bif662-GC-r as described by Satokari et al.
(38).
DGGE analysis
PCR amplicons were separated by DGGE based on the protocol of Muyzer et al. (29)
using the Decode system (Bio-Rad Laboratories, Hercules, California, USA) with the following
modifications. Polyacrylamide gels consisted of 8% (vol/vol) polyacrylamide (ratio of
acrylamide-bisacrylamide, 37.5:1) and 0.5 x Tris-acetate-EDTA (pH 8.0) (TAE) buffer.
Denaturing acrylamide of 100% was defined as 7 M urea and 40% formamide. The
polyacrylamide gels were made with denaturing gradient ranging from 38 to 48% and 45 to
55% to separate the generated amplicons of the total bacteria and Bifidobacterium
communities, respectively. The gels were poured from the top using a gradient maker and
128
Chapter 7
a pump (Econopump; Bio-Rad Laboratories, Hercules, California, USA.) set at a speed of
4.5 ml/min. Prior to the polymerization of the denaturing gel (gradient volume, 28 ml), a
7.5-ml stacking gel without denaturing chemicals was added, and the appropriate comb was
subsequently inserted. Electrophoresis was performed first for 5 min at 200 V and then for 16 h
at 85 V in 0.5 X TAE buffer at a constant temperature of 60C. The gels were stained with
AgNO
3
and dried overnight at 60C.
Cloning of the PCR-amplified products
PCR amplicons generated with the set of primers for the total (968-GC-f cand 1401-r)
and bifidobacterial (Bif164-f and Bif662-GC-r) communities were purified with the Qiaquick
PCR purification kit (Qiagen, Hilden, Germany) according to the manufacturers instructions.
PCR products were cloned into E. coli JM109 using the Promega pGEM-T vector system
(Promega, Madison, USA). Colonies of ampicillin-resistant transformants were transferred
with a sterile toothpick to 50 l of TE and were incubated at 95C for 15 min to lyse the cells.
PCR was performed on cell lysates using pGEM-T-specific primers T7 and SP6 (Table 1) to
check the size of the inserts. The plasmids containing inserts with the right size were used to
screen the transformant with the V6-V8 primers (968f-GC-f and 1401r) and the bifidobacteria
specific primers (Bif164-f and Bif662-GC-r). Clones for subsequent sequence analysis were
selected according to the migration position of the clone PCR fragment in the DGGE in
comparison to the fragments in the original DGGE profile. Insert PCR amplicons of selected
transformants were purified by the QIAquick PCR purification kit (Qiagen, Hilden,
Germany) and were subjected to DNA sequence analysis (Westburg, Leusden, The
Netherlands).
Sequence similarity were analyzed using BLAST tool (1) at the NCBI database
http://www.ncbi.nlm.nih.gov/BLAST). Alignment and further phylogenetic analysis of the
sequences was done using the ARB software package (25) and the tree was constructed using
the neighbor joining method (37).
7.3 Results
Optimization of fluorescent probes applied to fecal microbiota
In a previous study we have reported on an approach based on a combination of
fluorescent probes and FCM to assess the enzymatic activity, membrane integrity and
membrane potential of bifidobacteria, a common inhabitant of the GI-tract (5). This allowed
discriminating between viable, injured and dead bifidobacterial cells exposed to bile salts. In
this study, we extended this approach to a more complex and heterogeneous microbial
ecosystem: the fecal microbiota.
129

Figure 1: FCM cytograms of fecal sample stained with cFDA (A), and stained with cFDA and PI (B). In
each dot plot 3 regions containing cells with different fluorescent intensity in the FL1 and FL3 detectors are
identified.
In a first step, cFDA was used as viability marker to monitor the esterase activity.
When taken up by active cells, the non-fluorescent cFDA is hydrolyzed by intracellular non-
specific esterases to produce a green fluorescent substrate carboxyfluorescein (cF) that is
retained by cells having an intact membrane. Cytometric analysis of the cF-stained fecal
sample allowed 3 different subsets of cells to be distinguished based on their fluorescence
distribution and hence their differential uptake of the dye (Fig. 1A). In the dual dot plot
displaying cF-fluorescence (FL1) versus SSC signal, regions R1 and R2 represent cells that
took up the probe but showed a 10-fold difference in their fluorescence intensity (Fig. 1A).
The low signal conferred by the cells in R2 is most likely due to a low esterase activity,
presence of an efflux pump resulting in an active dye extrusion, or leakage of cF from a
number of cells as a result of a membrane damage. These metabolically active subpopulations
R1 and R2 accounted for 39.53 % and 25.23 % of the total cells, respectively. Cells in region
R3 corresponded to the unstained dead cells that did not take up the fluorescent dye and
accounted for 34.03% of the total cell counts. However, the simultaneous staining of the fecal
cells with cFDA and PI did not allow an unambiguous separation of the 3 subpopulations
(Fig. 1B), most likely due to the large heterogeneity in the physiology, cell envelope
architecture and diversity of fecal microbes. Hence, the use of the cFDA probe was abandoned
and an alternative approach based on simultaneous staining of bacterial nucleic acids by
permeant SYTO dyes and the impermeant fluorescent dye PI was used.
130
Chapter 7
Figure 2: FCM analysis of fecal sample stained with SYTO BC and PI. The two-color dot plot
discriminated between SYTO BC - stained viable cells (upper left quadrant), STYO BC and PI double
stained injured cells (upper right quadrant) and PI-stained dead cells (lower right quadrant). A fourth
population was detected in the lower left quadrant presenting electronic and sample background and did
not exceed 10% of the total population.
The SYTO dyes are capable of staining all cells, living or dead, whereas PI stains only
damaged or dead cells. In dead cells the simultaneous presence of a SYTO dye and PI activates
the energy transfer phenomena, the fluorescence emission spectrum of SYTO is absorbed by
PI and no longer detectable. Thus it is possible to distinguish viable cells (green fluorescence)
from dead ones (red fluorescence). In damaged cells with transient membrane permeability,
the energy transfer is not complete because of the lower PI concentrations and accordingly
damaged cells emit both green and red fluorescence (4). A number of SYTO dyes (SYTO BC,
SYTO 9 and SYTO 13) were used in combination with PI in order to optimize monitoring
the bacterial viability of fecal microbiota of 4 healthy adults (A, B, C and D). The best
discrimination between live damaged and dead cells was obtained with the fluorescent probes
SYTO BC and PI (data not shown). Figure 2 shows the cytograms of the green (FL1) versus
the red fluorescence (FL3) conferred by a fecal sample stained with both SYTO BC and PI.
Three different cell subpopulations could be identified: SYTO BC-stained intact or viable
cells, SYTO BC- and PI-stained injured cells, and PI-stained dead cells. The signal in the
lower left quadrant of the dot plot is due to the instruments electronic noise and the
background signal from non-fecal cells present in the sample. Table 2 shows the distribution
of the different cell counts of the fecal microbiota. Total cell numbers ranged between 1.68 x
10
11
to 4.64 x 10
10
cells/g wet feces for the four adults. Cells displaying an intact membrane
(live bacterial cells) accounted for 49.0 8.2 % of the total bacterial counts, while dead cells
represented 31.7 2.2 % of the total cell count for the 4 adults. The double stained fraction
representing the injured fecal cells showed a large variation and amounted to 19.4 8.7 %.
131
Table 2: Viability assessment of fecal microbiota using FCM and membrane integrity probes (PI and SYTO
BC).
Adult Total Count
1
EUB338 count
2
Viability count (%)
3
Viable Dead Injured
A 1.68 x 10
11
1.24 x 10
11
42.9 31.3 25.8
B 6.06 x 10
10
3.94 x 10
10
51.7 28.8 19.5
C 6.88 x 10
10
5.44 x 10
10
59.4 33.5 7.1
D 4.64 x 10
10
2.97 x 10
10
41.8 33.2 25.0
Mean (SD) 8.60(5.55) x 10
10
6.15(4.20) x 10
10
49.0 8.2 31.7 2.2 19.4 8.7
1
Number of fecal cells/ g wet feces cells determined by SYTO BC staining.
2
Number of cells hybridized by the bacterial probe Eub338 labelled with Cy5.
3
Percentage of viable, dead and injured cells as monitored by the combination of the fluorescent dyes: PI and SYTO BC.
Fluorescence-activated cell sorting
Following PI/SYTO BC staining, fecal samples were analyzed with a FACSVantage
flow sorter. Cells were sorted based on both scatter parameters and fluorescence signals in FL1
and FL3 by gating the sub-population of interest. This is illustrated by the analysis of the
samples of Adult A (Fig. 3). Region R4 was defined around the whole cell population
according to the scatter parameters (Fig 3A) and then a sort window was drawn around the
cells of interest (Fig. 3B). Subsequently, a gate was defined whereby any cell present within R4
and R1, R4 and R2, or R4 and R3 was sorted as live, damaged and dead cells into separate
sterile tubes, respectively. Approximately 1 x 10
6
to 5 x 10
6
cells were recovered from each
subpopulation. For adult C, the injured cells represented only 7.1% of the total fecal cells and
could not be recovered quantitatively during the sorting experiment. The major fractions in
this sample were the live (59.4%) and dead bacteria (33.5%). The sorted fractions were also
checked for purity by reanalyzing an aliquot of 1 ml from each sorting tube in both the
FACSVantage and FACSCalibur (Fig. 4). Sorted cells were then concentrated on filters that
were immediately checked by epifluorescent microscopy for purity, and subjected to DNA
extraction, in parallel with the non-sorted, fecal input samples.
132
Chapter 7
Figure 3. FACSVantage dual-parameter dot plots of a fecal sample from adult A stained with SYTO BC
and PI. Regions were defined around the whole cell population (R4) in dot plot A. In dot plot B regions
were set around target population as follows: SYTO BC-stained cells (R1), double-stained cells (R2) and
PI-stained cells (R3). A gate was defined whereby any particle present within R4 and R1, R4 and R2, R4
and R3 was sorted as live, injured and dead cells, respectively. Sorted cells were recovered into separate
sterile tubes to a range of 1 x 10
6
- 5 x 10
6
.
Figure 4: Sorting purity of recovered fractions. Following the sorting procedure, one ml of each recovered
sub-population was rerun in the FCM as illustrated in the panels shown above.
133
Chapter 7
Genetic diversity of dead, live and injured fecal bacteria assessed by PCR-DGGE
Following DNA isolation from the fecal samples and the sorted fractions, PCR was
performed to amplify the V6-V8 regions of the 16S rRNA genes followed by their separation
by DGGE (Fig. 5). The resulting PCR-DGGE banding patterns of the different amplicons
obtained from the total fecal community (before sorting) and the sorted viable, dead and
injured cells revealed different profiles in terms of number of bands, position and intensity.
Each subpopulation produced a complex fingerprint that in all cases was less complex than that
of the total fecal community. Variations in the profiles were observed between the adults,
indicating that not only the total fecal community but also the distinct subpopulations are
characteristic for each individual. This is in agreement with earlier studies where it was
demonstrated that the GI-tract microbiota is host-specific (48).
Figure 5: DGGE of PCR amplicons of the V6 to V8 regions of the 16S rRNA gene, obtained from total
fecal sample (lane 1) and sorted viable (lane 2), dead (lane 3) and injured (lane 4) subpopulations from
adult A to D. Lane M represents the marker. Bands that were sequenced are indicated by numbered
arrows and numbers (see also Fig. 8).
To assess changes in the genetic diversity of bacterial communities within the different
sorted cell fractions, a similarity index (SI) was calculated based on the Pearson (product-
moment) correlation coefficient. Subsequently, cluster analysis was performed using the
unweighted-pair group method using arithmetic averages (UPGMA), and dendrograms were thus
generated (Fig. 6). The average SI of the total microbiota in healthy adults over time is
approximately 76% (48). The PCR-DGGE patterns from adult B and D showed nearly identical
profiles between the injured and dead cells (SI of 89 and 83 % for adult B and D, respectively).
This indicates that the injured cells may represent the dying cells. For adult C, which contained
only a small injured fraction that could not be recovered (see above), the SI of the total bacteria
134
A B
C D
versus live and dead communities was 77 %, and 52 %, respectively. This may reflect the fact that
adult C contains the highest number of viable cells (see Table 2). However, for adult A, the SI for
the total compared to live fecal bacteria was 55 %, while the SI for comparison between the total
and dead communities was 71 %, suggesting that most phylotypes in this individual belong to
dead bacteria. For each individual, few bands were unique in each profile of the sorted viable and
dead fractions, indicating that these phylotypes are either live or dead, whereas other bands were
found in all profiles, suggesting that these bacterial groups occur as both live and dead in the fecal
sample. In contrast, only a very limited number of bands were either found in the live, dead or
injured profiles were not detected at the community level.
Figure 6: Dendrograms of the DGGE gel of each individual are generated by the Pearson
correlation index for each pair of lanes within a gel and was used as a measure of similarity
between the community fingerprints. The clustering of patterns was calculated using the
unweighted-pair group method using average linkages (UPGMA).
Analysis of the Bifidobacteria by specific PCR-DGGE in the sorted populations
To determine whether the results from the total bacterial population matched with that of
a specific bacterial group, we focused on the bifidobacterial population in the different adults
since they represent one of the major groups of the fecal microbiota. Fluorescent in situ
hybridization with the Bif164 probe, specific for bifidobacteria, revealed that this group
accounted for 10.10 0.39; 1.50 0.09; 2.84 0.11 and 3.93 0.15% of the total fecal cells for
adult A, B, C, and D, respectively. DGGE analysis of rRNA gene amplicons obtained with
Bifidobacterium-specific primers revealed that the bifidobacterial community in the four
135
Chapter 7
individuals was relatively simple (Fig. 7). Interestingly, bifidobacteria amplicons were detected in
the total fecal microbiota of the four adults as well as in the different sorted fractions isolated from
their feces. In general, the variations in DGGE patterns of the Bifidobacterium-specific amplicons
matched those observed with the total bacterial amplicons in the live, dead and injured fecal
microbiota of the four adults (Fig. 5).
Figure 7: DGGE profiles of bifidacterial PCR products of fecal samples from the 4 healthy adults (A, B, C,
and D). Lane 1: total fecal bacteria before sorting, lanes 2: live fecal bacteria, lane 3: dead fecal bacteria and
lane 4: injured cells, respectively. The dominant fragments indicated by numbered arrows were sequenced
and compared to known sequences in Genbank as illustrated in the phylogenetic tree (Fig. 9).
Phylogenetic analysis of dominant bands in the PCR-DGGE patterns
In order to gain insight into the genetic diversity of the live, dead and injured fecal
bacteria, the 16S rRNA genes from the sorted fecal fractions of adults A and C were amplified
and cloned in E. coli. The V6 to V8 regions of the cloned 16S rRNA gene fragments from the
cell lysates of transformants was amplified, and their migrations after DGGE were compared to
the rDNA-derived patterns of both adults (Fig. 5). In this way, 40 clones were each assigned to
the major bands in the DGGE profiles of both adults. The plasmid DNA from the corresponding
clones was purified and, subsequently the nucleotide sequence inserts was determined and
compared to the 16S rRNA databases. Finally, a phylogenetic tree of the partial 16S rRNA
sequences retrieved from the different sorted fractions (viable, injured, and dead bacteria) was
constructed (Fig. 8).
136
A
B C D
Figure 8: Phylogenetic tree of partial 16S rRNA sequences based on E. coli positions 968 to 1376 retrieved
from sorted viable (L), injured (I), and dead (D) fractions from fecal samples of adult A and adult C, and
full length reference sequence. Alignment and phylogenetic analyses were performed with ARB (25)
software, and the tree was constructed using the neighbor joining method (37) based on alignment positions
conserved among 50% of sequences analyzed. The reference bar indicates the branch length that represents
10% diversity. GenBank accession numbers of reference sequences are given. Suffixed are A: adult A, C:
adult C, L: sorted live cells, I: sorted injured cells and D: sorted dead cells.
137
Chapter 7
Among the 40 sequenced clones retrieved from all sorted bacterial fractions of both
individuals, 19 shared 95% or less 16S rRNA sequence identity with their nearest relatives. This
indicates that the majority of the sequences were derived from new, as yet undescribed bacterial
phylotypes. Most cloned sequences analyzed could be assigned to the major phylogenetic lineages
commonly encountered in human fecal clone libraries (Fig. 8) (41, 43). These included the
Clostridium coccoides (Clostridium XIVa cluster), Clostridium leptum (Clostridium cluster IV) (7)
and Bacteroides group. In general, it was Clostridium cluster XIVa that contained most of
sequences (27 clones) derived from both clone libraries. Analysis of the positioning of the clones
(Fig. 8) revealed that, the phylogenetic distribution of sequences from the clone library from adult
A was remarkably different from that of adult C. Furthermore the allocation in the phylogenetic
tree of the cloned sequences retrieved from the live, dead and injured fractions of both libraries
(adult A and C) was also different. These observations were supported by the PCR-DGGE
analysis, which revealed that the viable and dead communities showed different profiles (Fig. 5).
For the clone library of adult C, six of the Clostridiumcluster XIVa clones designated CL7,
CL8, CL9, CL10, CL12 and CL13 were affiliated with Eubacterium rectale subgroup with a
sequence similarity ranging from 94 to 100% and they were all recovered from the live fraction.
Furthermore, two clones (CL5 and CL1) were retrieved from the live fraction and were affiliated
with a butyrate producing bacterium L2-21 (AJ270477) with 96% sequence similarity and to
uncultured bacterium clone HuCA2 (AJ408958) with 94% sequence similarity, respectively.
Thus, they may belong to novel species. Two clones CL11, and CL6, retrieved from the live
fraction were related to Dorea longicatena (98% sequence similarity), a newly reclassified species,
and to Eubacterium formicigenerans (95% sequence similarity), respectively. In addition, two
clones CL3 and CL2 identified from the live fraction have sequences closest to Eubacterium hallii
(96% similarity) and butyrate-producing bacterium SM4/1 (AY305314) with 96% similarity,
repectively. However, all three clones (CD1, CD2 and CD4) that fell in the Eubacterium biforme
group (Clostridium cluster XVI) were only retrieved from the dead fraction. An additional clone
designated CD6 was also identified in the dead fraction and showed a perfect match with
Butyrivibrio crossotus. Ruminicoccus obeum-like species represented by clones CD5 and CD7 were
also found in the dead fraction of the fecal sample of adult C. Finally, one clone (CD3) retrieved
from the dead sorted fraction showed only 92% sequence similarity with Clostridium
cellulosolvens. Noticeably, the Bacteroides group was not represented in the clone library for adult C
neither the C. leptum subgroup.
The composition of the clone libraries of adult A differed from that of adult C although
Clostridium XIVa cluster contained most of the clones as for adult C. However the distribution of
the clones within this cluster differed between the two individuals, indeed no clones were
identified as E. rectale for example from the clone library of adult A. Two clones from the live
fraction AL3 and AL6 were closely related to E. formicigenerans (96% sequence similarity) and to
a butyrate producing bacterium SS3/4 (AY305316) (95% sequence similarity), respectively.
138

Five clones (AD1, AD3, AL7, AI2, and AD4) were positioned in the C. leptum subgroup
(Clostridium cluster IV), with Fusobacterium prausnitzii as the major fecal species (7). These clones
were retrieved from the live, injured and dead fecal fractions of adult A. Two clones (AL5 and
AD6), were more than 98,7% similar to each other, were closely related (99% sequence similarity)
to uncultured bacterium adhufec295 isolated by Suau et al. (41), and shared only 92 to 94%
similarity to the closest cultured relative that is Clostridium nexile, respectively. These two clones
were found both in the dead and live fraction; furthermore they had the same band position in
the PCR-DGGE pattern (Fig. 5, band 1A live, and band 1A dead). Remarkably, the clones that
clustered in the Bacteroides phylum were isolated only from the dead fraction with one clone
(AD8) being affiliated with the B. vulgatus subgroup and another (AD9) with the B. distasonis
subgroup. Finally one clone was closely related to the gram-negative bacterium Sutterella
wadsworthensis (99% sequence similarity), known to be associated with intestinal infections, was
recovered from the live fecal cells (44).
Phylogenetic analysis of the Bifidobacteria community
To identify the major bifidobacterial fragments in the viable, dead and injured populations
of fecal microbiota of adults A and C, cloning of approximately 550-bp PCR products generated
with Bif164-f and Bif662-GC-r primers was performed. Twenty-three clones that corresponded to
the major bands in the PCR-DGGE patterns were sequenced (Fig. 6). Analysis of these sequences
revealed that all clones retrieved were identified as Bifidobacterium species commonly found in the
human GI-tact (Fig. 9). Most clones (15 out of 23) were positioned in the B. adolescentis group,
five clones showed a sequence similarity of 96-99% to the B. longum -B. infantis group, two clone
was identified as B. bifidum (95-99% sequence similarity) and one clone had a high sequence
similarity (99% sequence similarity) with B. pseudocatenulatum.
Clones, designated B-AT4, B-AL1, B-AD3, B-AI1 and B-AI2, were identified as members
of B. infantis - B. longum group (98-99% sequence similarity). These clones were obtained from
the fecal sample of adult A and were detected in all cell fractions studied that were total, live, dead
and injured subpopulations. This suggests that these bacteria were metabolically active at the time
of sampling in the feces of adult A.
Cloned sequences closely related to B. adolescentis were identified from fecal samples of
both subjects. Seven sequences (B-AT1, B-AT2, B-AD2, B-AD4, B-CT3, B-CD2 and B-CD3),
although not identical, were more than 97% similar to each other and were 97-99% similar to
a B. adolescentis human fecal isolate. These sequences were retrieved from the total and dead fecal
cells of both adults, indicating that these bacteria were dead at the time of sampling. Other clones
closely related to B. adolescentis (B-CT1, B-CT4, B-CT5, B-CL2, B-CD1) were obtained
only from adult C; these clones had a high sequence similarity with an uncultured fecal
Bifidobacterium sp. 9C retrieved from a fecal human clone library (38). These clones were isolated
from the live and dead fecal fractions of adult C and were not found in the samples of adult
A. Two identical clones designated B-CT7 and B-CL1 were retrieved from the total community
139
Chapter 7
as well as the live fecal bacteria of adult C. These sequences were affiliated with uncultured fecal
Bifdobacterium sp. PL1 with 1% sequence divergence. However, these two later clones did not
migrate at the same position in the DGGE gel (Fig. 7 band 14C and 18C, respectively).
An additional clone (B-AD1) recovered from the dead fraction of adult A had a 98.82% similarity
with B-CT7 and B-CL1 clones, however the corresponding amplicons showed a different
migration behaviour in the DGGE gel (Fig. 7, band 8A).
Clone B-AL2 corresponding to band 5 in the PCR-DGGE gel (Fig 7) with 99% sequence
similarity to B. pseudocatenulatum, was identified in the live fraction of fecal sample of adult A.
Finally two clones (B-CT2 and B-CT6), recovered from the total fecal microbiota of adult C, were
identified as B. bifidum with 99 and 95% sequence similarity, respectively. These clones were
recovered from the total fecal bacterial community of adult C and were not detected in any other
sorted fractions.
Figure 9: Phylogenetic tree of bifidobacterial sequences based on E. coli positions 183 to 676. Alignment
and phylogenetic analysis were performed with the ARB software (25) and the tree was constructed using
the neighbor joining method (37) based on alignment positions conserved among 50% of sequences
analyzed. The reference bar indicates the branch length that represents 10% diversity. GenBank accession
numbers of reference sequences are given. Suffixed are A: adult A, C: adult C, T: total cells, L: sorted live
cells, I: sorted injured cells and D: sorted dead cells.
140
7.4 Discussion
In this study, we developed a new approach where flow cytometric in situ viability
assessment using functional probes (cFDA, SYTO BC and PI) was coupled with 16S rRNA gene
analysis to unravel the genetic diversity of viable, injured and dead fecal bacteria. We have
previously demonstrated the efficiency of a FCM assay based on cFDA, oxonol and PI as viability
marker to monitor the effect of bile salts on B. adolescentis, a common member of the GI-tract,
and B. lactis a widely used probiotic strain (5). The FCM assay was applied to monitor the
viability of fecal microbiota of four healthy individuals. The cFDA assay showed a striking
heterogeneity of the fecal bacteria and enabled to distinguish two metabolically active fractions
according to their differential uptake of the probes. These two metabolically active populations
accounted for 64.76% of the total microbiota, indicating that the remainder, approximately one
third, represent dead cells. When cFDA was used simultaneously with PI, although three
subpopulations could be separated according to their fluorescence intensity (Fig. 3B), the
resolution into live, injured and dead cells was not evident. This may be linked to the complexity,
physiological heterogeneity and the Gram status of the fecal bacteria. It had been reported that
cFDA is more effective at labeling Gram-positive than Gram-negative bacteria (9). In an
alternative approach, using SYTO BC and PI probes, which addresses membrane integrity,
discrimination of the fecal cell populations into viable intact, injured and dead cells was obtained.
The cytometric analysis revealed that the viable fecal populations accounted for approximately
half and the dead for almost a third of the total number of cells in the 4 volunteers. The remainder
of cells represented the injured ones but their population size showed some variation among the
tested individuals (19.4 8.7%). Recently, in another study, fecal samples from 10 healthy adults
were analyzed using PI and reported to contain a variable fraction of dead bacteria ranging from
17 to 34% (3).
The three different populations identified by the PI/SYTO BC assay were sorted and their
16S rDNA was amplified and, subsequently analyzed by PCR-DGGE. The PCR-DGGE
banding pattern of each individual revealed remarkable differences between the distinct
physiological fractions of the fecal microbiota. This was reflected by the similarity indices (SI)
based on the Pearson correlation coefficient. The SIs for comparison between the total
community and the sorted-live, -dead and -injured populations were lower (63, 67 and 58%,
respectively) than the average SI reported for stable microbiota of healthy adults over time (48).
However, for 2 different adults (B and D) the banding patterns of the dead and injured fractions
were quite similar (SI > 80%). The presence of common bands in the dead and injured fractions
suggests that some populations were composed of stressed and dead cells, reflecting a
physiological heterogeneity within some fecal bacterial populations. The death or injury of certain
fecal bacteria may be due to a variety of factors including, depletion of the readily metabolized
carbohydrates, low availability of free water, high ammonia concentration as digesta moves from
the proximal towards the distal colon, or sensitivity to antimicrobial products produced by
bacteria or the host. Consequently, some bacterial populations may tolerate these challenging
141
Chapter 7
conditions in the distal colon better than others (8). Interestingly, some bands were detected in
both the total community and in the fraction of metabolically active cells but were not identified
in the dead fraction. These observations suggest that these phylotypes were viable at the time
and the site of sampling. Inversely, some bands were found only in the dead sorted-fraction
indicating that these cells were dead at the time and site of sampling, but they might have
been active in the upper part of the colon (27).
The 40 sequenced clones obtained from dominant bands in the DGGE profiles of
the sorted viable, dead and injured populations of two selected adults were assigned to three
major phylogenetic lineages, namely Bacteroides, Clostridium coccoides and Clostridium leptum,
the two later equate to Clostridium rRNA cluster XIVa and IV, repectively (7). This is in
agreement with previous studies where comprehensive fecal clone libraries were made from
healthy adult fecal samples to get insight into the diversity of the GI-tract microbial ecosystem
(41, 43, 46). Additionally, Zoetendal et al. (49) demonstrated that the majority of
predominant bacterial species from an adult fecal sample did not correspond to known
species, but that the prominent bands of the fingerprint were assigned to different Clostridium
clusters. The result of such studies showed that a great majority (>50%) of the observed
diversity was attributable to unknown dominant microorganisms within the human gut.
In our study among the 40 sequenced clones retrieved from all sorted bacterial fractions of
both individuals, 19 shared 95% or less 16S rDNA sequence identity with their nearest
relatives.
The phylogenetic distribution of sequences from the adult A clone library differed in
terms of prevalence and species diversity from that of adult C as well as the allocation in the
phylogenetic tree of the cloned sequences retrieved from the live, dead and injured fractions
(Fig. 8). The phylogenetic analysis of the predominant bacteria obtained from the two adult
fecal samples showed that some bacterial groups were metabolically active at the time of
sampling like those belonging to the E. rectale subgroup, E. formicigenerans, D. longicatena and
E. hallii. This indicates that these bacteria have an important function at the end of the
gastrointestinal tract. Some members of Eubacteria are known to have a fermentative
metabolism with butyrate as a major end product (34). Alternatively, phylotypes belonging to
the Bacteroides group, Ruminococcus obeum- and E. biforme- like bacteria were obviously dead
at the time and site of sampling suggesting that these bacteria may have been active in the
upper part of the GI-tract i.e. proximal colon or terminal ileum. Bacteroides are known to be
one of the predominant bacterial groups in the gut; they are saccharolytic and play an
important role in the metabolism of bile salts and mucin degradation. Macfarlane et al. (26)
showed that during growth of fecal bacteria under carbon limiting conditions in an in vitro
model mimicking the distal colon, the number of Bacteroides declined significantly especially
with high retention time in the fermentor. Furthermore, some fecal populations identified in
the clone library of adult A were composed of viable and non-viable bacteria, this was
142

noticeably for the C. leptum cluster. The differences observed between the two adults in terms
of genetic diversity of the total, live and dead fecal bacterial populations might be attributable
to not only the host-genotype effect but also to other factors, such as diet and host health.
We also included the analysis of bifidobacteria in this study since they represent a major
group of the gut microbiota and have been implicated in a number of health promoting effects.
The PCR-DGGE and phylogenetic analysis were in agreement with earlier studies that adult
fecal bifidobacterial populations are host specific at the strain level (38). Most of the
bifidobacterial phylotypes detected in all samples and sorted fractions were identified as
B. adolescentis, B. longum, B. infantis, B. pseudocatenulatum, and B. bifidum. Matsuki et al.(28)
frequently detected B. longum, B. adolescentis and B. catenulatum by direct species- and group-
specific PCR in fecal samples, while B. infantis and B. breve were detected less frequently in
adult human feces (28). Furthermore, analysis of the live, injured and dead bifidobacterial
populations by FCM in combination with PCR-DGGE, cloning and sequencing revealed a
striking physiological heterogeneity within these populations. Indeed, the cloned sequences
that belonged to the B. longum B. infantis group were retrieved from all sorted fractions
(viable, injured and dead), whereas B. adolescentis related phylotypes were mostly retrieved from
the dead fractions. Although bifidobacteria are able to grow in different culture media, certain
species such as B. adolescentis are difficult to recover from fecal samples by culturing methods
while they could be detected by culture-independent methods (3, 31). This may be explained
simply by the fact that a number of B. adolescentis related species are dead or injured at the time
they are recovered from the feces, and hence cannot grow in culture media. This suggests that
one should analyze biopsies in order to get a good picture of the living microbiota.
Current GI-tract research has focused upon techniques such as the use of 16S rRNA
gene-based molecular tools to examine community structure and biodiversity. However, equally
important is the classification of microbes upon their physiology i.e. those that are functionally
active in the ecosystem versus those that are effectively redundant and play little or no role at
a particular time. FCM has been viewed as a a possible alternative to current assay techniques
to monitor the viability of stressed and starved bacteria and identify microorganisms in their
natural habitat, with potential for automation. The combination of FCM cell sorting and
subsequent 16S rRNA gene analysis that we have used here for the first time for fecal
microbiota provides relevant ecological information related to the diversity and activity of the
fecal microbiota in situ and highlights the physiological heterogeneity of this complex
ecosystem. FCM and cell sorting will unquestionably offer novel applications to study the GI-
tract microbial ecosystem. For instance, fecal bacteria can be sorted on the basis of their taxa
by means of fluorescent in situ hybridization (FISH) or immunofluorescence, followed by
characterization using omics-based approaches to further investigate the responses of these
microbes to changes in environmental conditions.
143

7.5 References
38. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local
alignment search tool. J. Mol. Biol. 215:403-410.
3. Apajalahti, J. H. A., A. Kettunen, P. H. Nurminen, H. Jatila, and W. E. Holben. 2003.
Selective plating underestimates abundance and shows differential recovery of
bifidobacterial species from human feces. Appl. Environ. Microbiol. 69:5731-5735.
4. Barbesti, S., S. Citterio, M. Labra, M. D. Baroni, M. G. Neri, and S. Sgorbati. 2000.
Two and three-color fluorescence flow cytometric analysis of immunoidentified viable
bacteria. Cytometry 40:214-218.
5. Ben-Amor, K., P. Breeuwer, P. Verbaarschot, F. M. Rombouts, A. D. L. Akkermans, W.
M. De Vos, and T. Abee. 2002. Multiparametric flow cytometry and cell sorting for the
assessment of viable, injured, and dead Bifidobacterium cells during bile salt stress. Appl.
6. Bernard, L., C. Courties, C. Duperray, H. Schafer, G. Muyzer, and P. Lebaron. 2001. A
new approach to determine the genetic diversity of viable and active bacteria in aquatic
7. Collins, M. D., P. A. Lawson, A. Willems, J. J. Cordoba, J. Fernandez-Garayzabal, P.
Garcia, J. Cai, H. Hippe, and J. A. Farrow. 1994. The phylogeny of the genus
Clostridium: proposal of five new genera and eleven new species combinations. Int. J. Syst.
Bacteriol. 44:812-826.
8. Cummings, J. 1995. The large intestine in nutrition and disease, p. 155, Danone chair
monograph, Bruxelles.
9. Davey, H., and D. Kell. 1996. Flow cytometry and cell sorting of heterogeneous
microbial populations: the importance of single-cell analyses. Microbiol. Rev. 60:641-
696.
10. Davey, H., and M. Winson. 2003. Using flow cytometry to quantify microbial
heterogeneity. Curr. Issues Mol. Biol. 5:9-15.
11. Falk, P. G., L. V. Hooper, T. Midtvedt, and J. I. Gordon. 1998. Creating and
maintaining the gastrointestinal ecosystem: what we know and need to know from
gnotobiology. Microbiol. Mol. Biol. Rev. 62:1157-1170.
144
Chapter 7
12. Favier, C. F., E. E. Vaughan, W. M. de Vos, and A. D. L. Akkermans. 2002. Molecular
13. Harmsen, H. J., and G. W. Welling. 2002. Fluorescent in situ hybridization as a tool in
intestinal bacteriology, p. 41-57. In G. W. Tannock (ed.), Probiotics and Prebiotics:
Where are we going. Caister Academic Press, Norfolk, England.
Extensive set of 16S rRNA-based probes for detection of bacteria in human feces. Appl.
15. Haugland, R. P. 2002. Molecular Probes - Handbook of fluorescent probes and research
products, 9
th
ed. www.probes.com.
and W. M. de Vos. 2002. Molecular diversity of Lactobacillus spp. and other lactic acid
17. Hooper, L. V., T. Midtvedt, and J. I. Gordon 2002. How host-microbial interactions shape
the nutrient environment of the mammalian intestine. Annu. Rev. Nutr. 22:283-307.
18. Hooper, L. V., M. H. Wong, A. Thelin, L. Hansson, P. G. Falk, and J. I. Gordon 2001.
Molecular analysis of commensal host-microbial relationships in the intestine. Science
291:881-884.
19. Joux, F., and P. Lebaron. 2000. Use of fluorescent probes to assess physiological functions
of bacteria at single-cell level. Microb. Infect. 2:1523-1535.
20. Kane, M., L. K. Poulsen, and D. Stahl. 1993. Monitoring the enrichment and isolation
of sulfate reducing bacteria by using oligonucleotide hybridization probes designed from
environmentally derived 16S rRNA sequences. Appl Environ Microbiol. 59:682-686.
21. Kaufmann, P., A. Pfefferkorn, M. Teuber, and L. Meile. 1997. Identification and
quantification of Bifidobacterium species isolated from food with genus-specific 16S
rRNA-targeted probes by colony hybridization and PCR. Appl. Environ. Microbiol.
63:1268-1273.
22. Kok, R., A. de Waal, F. Schut, G. Welling, G. Weenk, and K. Hellingwerf. 1996. Specific
detection and analysis of a probiotic Bifidobacterium strain in infant feces. Appl. Environ.
Microbiol. 62:3668-3672.
23. Lane, D. J. 1991. 16S/23S rRNA sequencing, p. 115-175. In E. Stackebrandt, and M.
Goodfellow (eds), Nucleic acid techniques in bacterial systematics. John Wiley & Sons,
Chichester, UK.
145

25. Ludwig, W., O. Strunk , Westram R, L. Richter, H. Meier, Y, adhukumar, A. Buchner,
T. Lai , S. Steppi, Jobb G, W. Forster, I. Brettske, S. Gerber, A. W. Ginhart , O. Gross,
S. Grumann , S. Hermann , R. Jost , A. Konig, T. Liss, R. Lussmann, M. May , B.
Nonhoff, B. Reichel, R. Strehlow , A. Stamatakis, N. Stuckmann, A. Vilbig, M.
Lenke, T. Ludwig, A. Bode, and K.-H. Schleifer. 2004. ARB: a software environment
for sequence data. Nucleic Acids Res. 32:1363-1371.
26. Macfarlane, S., M. E. Quigley, M. J. Hopkins, D. F. Newton, and G. T. Macfarlane.
1998. Polysaccharide degradation by human intestinal bacteria during growth under
multi-substrate limiting conditions in a three-stage continuous culture system. FEMS
Microbiol. Ecol. 26:231-243.
27. Marteau, P., P. Pochart, J. Dor, C. Bera-Maillet, A. Bernalier, and G. Corthier. 2001.
Comparative study of bacterial groups within the human cecal and fecal microbiota.
Appl. Enviro. Microbiol. 67:4939-4942.
28. Matsuki, T., K. Watanabe, J. Fujimoto, Y. Kado, T. Takada, K. Matsumoto, and R.
Tanaka. 2004. Quantitative PCR with 16S rRNA-gene-targeted species-specific
primers for analysis of human intestinal bifidobacteria. Appl. Environ. Microbiol.
70:167-173.
30. Nebe-von-Caron, G., P. J. Stephens, C. J. Hewitt, J. R. Powell, and R. A. Badley.
2000. Analysis of bacterial function by multi-colour fluorescence flow cytometry and
single cell sorting. J. Microbiol. Methods. 42:97-114.
31. Nielsen, D. S., P. L. Moller, V. Rosenfeldt, A. Paerregaard, K. F. Michaelsen, J, and M.
Jakobsen. 2003. Case study of the distribution of mucosa-associated Bifidobacterium
species, Lactobacillus species, and other lactic acid bacteria in the human colon. Appl.
32. Nubel, U., B. Engelen, A. Felske, J. Snaidr, A. Wieshuber, R. Amann, W. Ludwig, and
178:5636-5643.
33. Porter, J., D. Deere, M. Hardman, C. Edwards, and R. Pickup. 1997. Go with the
flow-use of flow cytometry in environmental microbiology. FEMS Microbiol. Ecol.
24:93-101.
146
Chapter 7
34. Pryde, S. E., S. H. Duncan, G. L. Hold, C. S. Stewart, and H. J. Flint. 2002. The
microbiology of butyrate formation in the human colon. FEMS Microbiol. Lett.
217:133-139.
35. Rachmilewitz, D., K. Katakura, F. Karmeli, T. Hayashi, C. Reinus, B. Rudensky, S.
Akira, K. Takeda, J. Lee, K. Takabayashi, and E. Raz. 2004. Toll-like receptor 9 signaling
mediates the anti-inflammatory effects of probiotics in murine experimental colitis.
36. Rigottier-Gois, L., A. G. Le Bourhis, G. Gramet, V. Rochet, and J. Dor. 2003.
37. Saitou N., and M. Nei. 1987. The neighbor-joining method: a new method for
reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425.
38. Satokari, R. M., E. E. Vaughan, A. D. L. Akkermans, M. Saarela, and W. M. de Vos.
2001. Bifidobacterial diversity in human feces detected by genus-specific PCR and
denaturing gradient gel electrophoresis. Appl. Environ. Microbiol. 67:504-513.
39. Shapiro, H. M. 2000. Microbial analysis at the single-cell level: tasks and techniques. J.
Microbiol. Methods. 42:3-16.
40. Shapiro, H. M. 1995. Practical Flow Cytometry, 3
d
ed. Wiley-Liss Inc, New York.
4807.
42. Vaughan, E. E., H. G. H. J. Heilig, E. G. Zoetendal, R. Satokari, J. K. Collins, A. D. L
Akkermans, and W. M. de Vos. 1999. Molecular approaches to study probiotic bacteria.
Trends Food Sci. Technol. 10:400-404.
43. Wang, X., S. P. Heazlewood, D. O. Krause, and T. H. Florin. 2003. Molecular
characterization of the microbial species that colonize human ileal and colonic mucosa by
using 16S rDNA sequence analysis. J. Appl. Microbiol. 95:508-520.
44. Wexler, H. M., D. Reeves, P. H. Summanen, E. Molitoris, M. McTeague, J. Duncan,
K. H. Wilson, and S. M. Finegold. 1996. Sutterella wadsworthensis gen. nov., sp. nov.,
bile-resistant microaerophilic Campylobacter gracilis-like clinical isolates. Int. J. Syst.
Bacteriol. 46:252-258.
45. Whitfield, J. 2004. Features-Science and health: microbial soup of life is sieved for
treasure., Financial Times.
46. Wilson, K. H., and R. B. Blitchington. 1996. Human colonic biota studied by ribosomal
DNA sequence analysis. Appl. Environ. Microbiol. 62:2273-2278.
147

47. Winson, M. K., and H. M. Davey. 2000. Flow Cytometric Analysis of Microorganisms.
Methods 21:231-240.
the human gastronintestinal tract. Microbial. Ecol. Health Dis. 13:129-134.
49. Zoetendal, E. G., A. D. L. Akkermans, and W. M. de Vos. 1998. Temperature gradient
148
Chapter 7
Chapter 8
SUMMARY AND CONCLUDING REMARKS
149
This thesis describes a multifaceted approach to further enhance our view of the complex
human intestinal microbial ecosystem. This approach combines the advantages of flow
cytometry (FCM), a single cell and high-throughput technology, and molecular techniques that
have proven themselves to be invaluable tools in studying the microbial diversity and structure
of the intestinal microbiota. The ultimate aim was to relate the genetic biodiversity of the
intestinal microbiota with their in situ physiological activity. The major findings of this thesis are
summarized below and discussed in a broader perspective:
In chapter 1, the insights and newly developed molecular methods to study the eco-
physiology of the gastrointestinal (GI)-tract are presented with a brief outline of the thesis.
In chapter 2, the principle of FCM was reviewed as well as its applications in microbial
ecology and physiology in different environmental settings. The power of FCM stems essentially
from the ability to perform multiparametric analysis at the single cell level, the high-throughput
capacity, and the option of cell sorting.
In chapter 3 the technique to detect fecal bacteria using FCM and fluorescent in situ
hybridization (FISH) was reported for the first time. The initial focus was to design and validate
a 16S rRNA oligonucleotide probe (Urobe63), targeting uncultured Ruminnococcus obeum-like
bacteria whose presence was only suggested by 16S rDNA cloning studies of fecal samples.
Subsequently, a flow cytometric-based method was developed to ascertain the specificity of the
probe, and to determine the abundance of this bacterial group in fecal samples using FISH.
A hierarchical set of probes was used including the general bacterial probe Eub338, the
Clostridium coccoides Eubacterium rectale group-specific probe Erec482, that encompasses the
uncultured R. obeum-like bacteria, and the newly designed probe Urobe63. Combining the
scatter parameters and the probe-conferred fluorescence of the detected cells revealed that a
group of morphologically similar bacteria in feces hybridized with the Urobe63 probe.
Quantification of the uncultured R. obeum-like bacteria by FCM-FISH, demonstrated that this
phylogenetic group comprises a significant fraction of the fecal community (2.2%). Statistical
analysis showed that FCM and fluorescent microscopy counts were in good agreement. These
FISH results substantiate previous studies of phylogenetic investigation of fecal clone libraries
that foreshadowed the high prevalence of this uncultured group of bacteria in fecal samples of
different individuals (8,9). The present study demonstrates that combination of FCM and
16S rRNA-targeted probes can indeed be successfully applied to enumerate those
microorganisms that have proven difficult to culture and which presumably play an important
role on the gut physiology.
The knowledge about the mucosa-associated bacterial communities in different parts of
the colon is limited, since most attention has been focused on bacteria present in feces. In
chapter 4, the spatial distribution of the intestinal microbiota was investigated by comparing the
bacterial communities in feces and biopsy samples taken from the ascending, transverse,
and descending colon using a 16S rRNA approach. Flow cytometric analysis indicated that
150
Chapter 8
10
5
to 10
6
bacteria were present in the biopsy samples (~ 0.5 mg) with a detection limit of
10
4
cells/sample. These results were in contrast to a previous report using a microscopic
method whereby hardly any bacteria were detected in biopsy specimens of healthy persons (6).
Differences to these results could be a result of the detection limit of both methods or to the
biopsy preparation. The numbers of bacteria counted in the ascending colon were slightly lower
than those from other locations of the intestine. Denaturing gradient gel electrophoresis
(DGGE) analysis of 16S rRNA gene amplicons and similarity index comparisons
demonstrated, that the predominant mucosa-associated bacterial community was host-specific
and differed significantly from the fecal community. However, no association of biopsy
localization with bacterial diversity was observed demonstrating that the mucosa-associated
microbiota is uniformly distributed along the colon. The cluster analysis of the PCR-DGGE
profiles pointed out a potential difference between the bacterial communities of healthy and
diseased individuals. However, it was not clear whether the differences observed were in fact the
consequence of the health status of the individuals, due to the low number of samples studied.
On the other hand, recent studies demonstrated that indeed the fecal and mucosa-associated
microbiota significantly differed from that of patients with ulcerative colitis (UC) and Crohns
disease (2, 7) see also Chapter 5.
It has been well established that the predominant fecal bacterial communities of healthy
individuals is host-specific, relatively stable in time, differs from mucosa-associated bacteria and
not significantly altered following consumption of certain probiotic strains. In Chapter 5 the
structure and population dynamics of the intestinal microbiota of UC patients over time and
as a response to the intake of probiotics (Lactobacillus salivarius subsp. salivarius UCC118 or
Bifidobacterium infantis 35624), was monitored using FISH-FCM and PCR-DGGE analysis.
The results obtained by both molecular approaches indicated that the microbiota of
UC patients is host specific, but revealed marked alterations in the major bacterial groups of
each patient. During the course of the probiotic trial, a significant increase in both
bifidobacterial and Lactobacillus populations and a decrease in the Atopobium group were
observed. However, these changes were not attributable to the type of treatment since they were
observed in the three cohorts: placebo, probiotic-1 and probiotic-2. On the other hand, the
between group statistical analysis revealed that Bacteroides counts decreased significantly in the
probiotic-2 group, while clostridia decreased in the placebo (a fermented milk containing
Streptococcus thermophilus) cohort. The observed placebo effect may suggest that products of
fermentation play a role in in vivo effects, and that consideration must be given to the delivery
vehicle in comparisons to the efficiency between probiotic strains. Interestingly, bifidobacteria
were detected at the pre-treatment period in all patients at a relatively high level. The PCR-
DGGE profiles did not show any explicit band that could indicate the presence of specific
bacteria that may be related to the disease. Statistical analysis of the DGGE profiles showed that
UC patients clustered separately from healthy controls. This suggests that the dominant
microbiota of UC patients is atypical and that an imbalance of the microbial ecosystem rather
than specific bacteria might play a role in the aetiology of the disease. Interestingly, the bacterial
151
Summary and concluding remarks

profiles of a number of patients tended to stabilize at the end the trial as revealed by the
fingerprints. These observations suggest that the administration of live bacteria may have
resulted in a normalization of the total microbiota.
The application of 16S rRNA-based techniques has greatly contributed to our
understanding of the diversity of the intestinal microbiota, however a pertinent question still
remains to be answered: which members of the ecosystem are functionally active at a particular
time or at a specific site? It is therefore a major challenge to develop methods that allow
monitoring of microorganisms according to their eco-physiological traits in situ. FCM has been
viewed as a possible alternative to current assay techniques to monitor the viability of stressed
and starved bacteria. In chapter 6 the efficiency of an FCM-based approach to monitor the effect
of bile salt stress on the viability of Bifidobacterium adolescentis, a common member of the GI-
tract, and Bifidobacterium lactis a widely used probiotic strain, was demonstrated.
Carboxyfluorescein diacetate, propidium iodide (PI) and oxonol (DiBAC4 (3)) were used to
assess the esterase activity, membrane integrity and membrane potential, respectively, as
indicators of bacterial viability. FCM analysis revealed a striking heterogeneity within the
stressed bifidobacterial population and allowed to discriminate between viable, injured and dead
sub-populations based on their differential uptake of the probes. One interesting finding was
that the plasma membrane of the injured cells was apparently permeable to PI. An early stage of
the recovery-resuscitation process involved repair of the damaged membrane, these cells could
actually be scored dead by traditional criteria. Bile salt may have induced a sublethal-injury
within bifidobaterial population possibly through a reversible and transient membrane
permeabilization which resulted in a loss of viability, as defined by plate counts but could regain
growth after being sorted and resuscitated. Preconditioning with bile salts showed a remarkably
increased resistance of B. adolescentis NCC251 to lethal concentrations of bile salts, most likely
through the induction of a mechanism allowing the cells to build up a protection against the
solubilization of their membrane proteins (5). The findings of this study showed that
multiparametric FCM is a very effective means for quantifying physiological heterogeneity of
stressed bacteria, and thus provides a sensitive tool to assess the viability and stability of
probiotics.
The FCM approach has been extended and complemented with 16S ribosomal RNA
gene analysis to study the relation between diversity and activity of the fecal microbiota
(chapter 7). Simultaneous staining with PI and SYTO BC provided a clear differentiation
between viable, injured, and dead fecal cells. Following sorting, the three subpopulations were
characterized by DGGE of 16S rRNA gene amplicons obtained from the total and
bifidobacterial communities. This analysis revealed that not only the total community but also
the distinct subpopulations are characteristic for each individual. Phylogenetic analysis
demonstrated that a number of amplicons, closely related to likely butyrate-producing bacteria,
were obviously viable at the time and site of sampling. However, amplicons affiliated with
Bacteroides, Ruminococcus obeum-, Eubacterium biforme-like bacteria were especially obtained
152
Chapter 8
from the dead population at the time of sampling. Furthermore, some bacterial clones were
recovered from all sorted fractions, this was especially noticeable for the Clostridium leptum
cluster. This may indicate that nutrient availability and environmental conditions help
determine the niche that a given microbe is able to occupy. Likewise, phylogenetic analysis of
the live, dead and injured cells of bifidobacteria revealed a remarkable physiological
heterogeneity within these bacterial populations. Basically, bacteria related to the B. longum -
B. infantis group were retrieved from all sorted fractions, indicating that these bacteria are
active in the lower part of the GI-tract. Recently, the genome sequence of B. longum indicated
its huge capability for untilization of complex carbohydrates derived from plants and which are
believed to be a major component of the lower part of the colon (4). However, B. adolescentis-
related bacteria were mostly recovered from the sorted-dead fraction and these may have been
active in the upper part of the colon. Finally, B. pseudocatenulatum, and B. bifidum-like bacteria
were identified only in the live fraction. The differences observed between the individuals in
terms of genetic diversity of the total, live and dead fecal bacterial populations might be
attributable to not only the host-genotype effect but also to other factors, such as diet and host
health.
In conclusion, the combination of FACS and 16S rRNA approaches that have been
used in this thesis, contributed to consolidate our understanding of the large spatial diversity,
structure, and the contribution to the global microbial activity to the intestinal ecosystem, as
related to the host-genotype as well as the host-health status. The striking eco-physiological
heterogeneity observed within the total and the bifidobacterial communities, may indicate that
the source and availability for nutrients, metabolic interdependencies of microbes, as well as the
challenging condition of the gut may play an essential role in shaping this complex ecosystem.
The challenge set before us now, is to figure out how these microbes interact with each other
and with the host to make their living and how the host designs its own community structure
in order to maintain the homeostasis of the ecosystem. Novel approaches are being developed
to answer such questions and to better grasp the function of gut-related bacteria in the
intestinal tract and help for a rational design of probiotics and other functional foods. These
include the use of omics-based approaches, i.e. global investigations of gene expression profiles
or in-vivo gene expression identification strategies, and the combination of molecular
techniques with substrate labeling such as stable isotope probing (SIP) (1, 3). These novel
approaches will obviously shed light on the metabolic functions of the gut microbiota and its
relation to the host health. With the expansion of bacterial genomes and the innovation of
omics-based approaches, FCM will have also an important role to play in resolving the
interactions between microbial genetics and physiology.
153
Summary and concluding remarks

References
1. de Vos, W. M., P. A. Bron, and M. Kleerebezem. 2004. Post-genomics of lactic acid
bacteria and other food-grade bacteria to discover gut functionality. Curr. Opin.
bacterial microflora in patients with active inflammatory bowel disease. Gut 53:685-693.
3. Radajewski, S., I. R. McDonald, and J. C. Murrell. 2003. Stable-isotope probing of
nucleic acids: a window to the function of uncultured microorganisms. Curr. Opin.
4. Schell, M. A., M. Karmirantzou, B. Snel, D. Vilanova, B. Berger, G. Pessi, M. C.
Zwahlen, F. Desiere, P. Bork, M. Delley, R. D. Pridmore, and F. Arigoni. 2002. The
genome sequence of Bifidobacterium longum reflects its adaptation to the human
gastrointestinal tract. Proc. Natl. Acad. Sci. USA 99:14422-14427.
5. Schmidt, G., and R. Zink. 2000. Basic features of the stress response in three species of
bifidobacteria: B. longum, B. adolescentis, and B. breve. Int. J. Food. Microbiol. 55:41-45.
6. Schultsz, C., F. M. Van Den Berg, F. W. Ten Kate, G. N. Tytgat, and J. Dankert. 1999.
The intestinal mucus layer from patients with inflammatory bowel disease harbors high
numbers of bacteria compared with controls. Gastroenterol. 117:1089-1097.
7. Seksik, P., L. Rigottier-Gois, G. Gramet, M. Sutren, P. Pochart, P. Marteau, R. Jian, and
J. Dore. 2003. Alterations of the dominant faecal bacterial groups in patients with
8. Suau, A., R. Bonnet, M. Sutren, J.-J. Godon, G. R. Gibson, M. D. Collins, and J. Dore
4807.
9. Zoetendal, E. G., A. D. L. Akkermans, and W. M. de Vos 1998. Temperature gradient
154
Chapter 8
SAMENVATTING
155
Het onderzoek beschreven in dit proefschrift richt zich op de ontwikkeling van
geavanceerde fluorescentietechnieken ten behoeve van de detectie, identificatie en de bepaling
van de activiteit van micro-organismen aanwezig in het menselijk maagdarmkanaal, de
zogenaamde darmflora. Deze micro-organismen spelen een belangrijke rol bij het omzetten
van voedselcomponenten (de vertering van het voedsel) en de bescherming tegen ongewenste
micro-organismen (de invasie van pathogene bacterin). Het mag duidelijk zijn dat de
darmflora een zeer belangrijke rol speelt in het functioneren van de mens (hoofdstuk 1).
Bij veel darmflora onderzoeken wordt gebruik gemaakt van zogenaamde kweekmethoden,
waarbij, na groei in het laboratorium, bacterin worden gedentificeerd en gekarakteriseerd.
Omdat men niet precies weet wat de verschillende bacterin nodig hebben voor de groei,
kunnen niet alle groepen gekweekt worden. Er wordt aangenomen dat ongeveer 50% van alle
darmbacterin niet in het laboratorium is te kweken, en zodoende via deze weg niet kan
worden bestudeerd. Men dient zich hierbij te realiseren dat de soortenrijkdom enorm is; er
wordt aangenomen dat het aantal bacteriesoorten in de darm de duizend overstijgt.
Recentelijk heeft men dan ook andere methoden gentroduceerd waarbij de bacterin niet
hoeven te worden gekweekt, maar waarbij deze worden gedetecteerd en gedentificeerd met
behulp van zogenaamde moleculaire genetische technieken welke zich richten op specifieke
kenmerken van onder andere het DNA. Hierop staan de erfelijke eigenschappen beschreven
in de vorm van genen, waarbij het gen coderend voor het 16S ribosomaal RNA (16S rRNA)
dat een onderdeel is van de ribosomen, de eiwitfabrieken, een centrale rol speelt in deze
technieken. Elke bacteriesoort wordt gekenmerkt door een specifiek 16S rRNA molecuul,
zodat met behulp van gegevens in een grote databank, welke meer dan 97.000 16S rRNA
sequenties bevat, kan worden onderzocht of het om bekende of nieuwe bacterin gaat. De
genen welke coderen voor 16S rRNA moleculen kunnen worden gesoleerd uit de bacterin,
vervolgens worden vermeerderd met behulp van de zogenaamde PCR techniek, en van elkaar
worden gescheiden met behulp van TGGE en DGGE technieken. Hierbij wordt een
streepjescode gemaakt waarbij ieder streepje afkomstig is van een 16S rRNA van een specifieke
bacterie in de darmflora. De toepassing van dergelijke kweek-onafhankelijke technieken heeft
nieuwe inzichten opgeleverd betreffende de samenstelling van de darmflora, en de variatie
hierin welke bestaat tussen mensen.
Een belangrijke vraag die daarna dient te worden beantwoord betreft het functioneren
van de verschillende soorten bacterin in de darm, oftewel: wat doen ze daar precies en hoe
doen ze dit? Het onderzoek beschreven in dit proefschrift heeft zich met name op dat aspect
gericht, gebruik makend van geavanceerde fluorescentie technieken, in het bijzonder de flow
cytometrie (hoofdstuk 2). Hierbij wordt gebruik gemaakt van een flowcytometer, een apparaat
waarmee kleine deeltjes, in dit geval bacterin, n voor n kunnen worden geanalyseerd.
Hierbij worden de bacterin blootgesteld aan licht met een vastgestelde golflengte, dat wil
zeggen van een bepaalde kleur, en vervolgens kunnen aan de hand van de lichtverstrooiing,
bepaalde kenmerken van de bacterin worden bepaald zoals hun grootte. Doordat de
stroomsnelheid waarmee de bacterin langs het meetpunt worden gevoerd kan worden
156
Samenvatting

gemeten, kan ook het aantal bacterin dat aanwezig is in een volume, worden bepaald.
Daarnaast kunnen nog veel meer parameters worden bestudeerd door de bacterin te kleuren
met fluorescente probes, welke informatie kunnen geven over de identiteit, activiteit en
levensvatbaarheid van een bacterie. Zo zijn er fluorescente bacteriesoortspecifieke DNA
probes die de onderzoeker, na flow cytometrische analyse van bijvoorbeeld een fecesmonster,
precies vertellen hoeveel bacterin er van die soort in de feces (aantal bacterin per gram)
aanwezig zijn. De detectie van bacterin met deze methode vereist dat de beschermende lagen
rond bacterin (celwand en celmembraan) permeabel worden gemaakt, door gebruik te maken
van bepaalde chemicalin. Via de ontstane gaten is het mogelijk fluorescente DNA probes de
bacterie in te transporteren zodat ze kunnen binden met het 16S rRNA. Anderzijds, kan de
mate van beschadiging van genoemde lagen, een indicatie geven over de intactheid van de
bacterie. Er zijn bijvoorbeeld probes waaronder propidiumjodide (PI), welke alleen gaan
fluoresceren (dat wil zeggen licht uitzenden met een iets andere kleur dan waaraan ze zijn
blootgesteld in het meetpunt van de flowcytometer), als ze een binding zijn aangegaan met het
DNA van de bacterie. Analyse van een fecesmonster met PI geeft dus informatie over het
aantal, in dit geval roodgekleurde, en dus dode cellen. Een andere veelgebruikte probe,
carboxyfluorescene diacetaat (cFDA), kan ook bij levende cellen de bacteriecelwand en
membraan passeren. In de cel wordt dit gesplitst in acetaat en het groen-fluorescerende
carboxyfluorescene (cF), dat alleen in actieve cellen ophoopt met een intacte celwand en -
membraan. Er zijn ook probes met zodanige eigenschappen, dat ze elke bacteriecelwand en
membraan kunnen passeren en vervolgens binden aan het DNA en dan gaan fluoresceren.
Op deze manier kan men alle bacterin tellen in een bepaald monster. Door nu slimme
combinaties van probes te gebruiken kunnen allerlei parameters tegelijk worden bepaald, de
zogenaamde multiparameter-aanpak.
Het mag duidelijk zijn dat fluorescente technieken voor de detectie, identificatie en de
bepaling van de activiteit van micro-organismen aanwezig in het menselijk maagdarmkanaal
(de darmflora) allerlei mogelijkheden bieden. Daarnaast zou deze aanpak gebruikt kunnen
worden om de activiteit en effectiviteit van zogenaamde probiotica en prebiotica vast te
stellen. Probiotica zijn levende micro-organismen die, wanneer ze worden geconsumeerd, een
gezondheidsbevorderend effect op de gastheer hebben. Dit effect kan een gunstig effect op de
darmflora zijn. Om levend de dikke darm te bereiken, moeten probiotica bestand zijn tegen
de omstandigheden in de maag en de dunne darm. De meest bekende probiotica behoren tot
de lactobacillen of de bifidobacterin. Prebiotica zijn voedingscomponenten die niet verteerd
worden en de dikke darm ongeschonden bereiken en daar selectief de groei of activiteit
bevorderen van bepaalde soorten gezondheidsbevorderende bacterin. Bekende prebiotica zijn
niet verteerbare oligo- en polysacchariden zoals inuline en galactooligosaccharides.
De toepassing van flowcytometrie in combinatie met verschillende probes, heeft
geresulteerd in zeer interessante nieuwe informatie over de menselijke darmflora. Zo wordt in
hoofdstuk 3 het voorkomen beschreven van een bepaalde bacteriesoort, een zogenaamde
157
Samenvatting

Ruminococcus soort, welke in het laboratorium niet is te kweken. Met deze nieuwe methode
kon worden vastgesteld dat deze bacterie ongeveer 2% uitmaakt van het totaal aantal bacterin
in de feces. Hoofdstuk 4 laat zien dat fluorescente technieken ook kunnen worden gebruikt
bij de analyse van zogenaamde biopten, stukjes darmwand verkregen uit patinten. De
darmwand is bedekt met een slijmlaag (mucus) welke grote aantallen bacterin blijkt te bevatten.
Vergelijking van de bacteriesamenstelling in de biopten met die van de feces laat grote verschillen
zien. Daarnaast liet dit onderzoek zien dat elk individu zowel een specifieke darmwandflora en
specifieke fecale flora bezit. Door in de toekomst verder onderzoek te doen aan de
darmwandbacterin kan men nog meer te weten komen over het functioneren van de darmflora.
In hoofdstuk 5 wordt het effect onderzocht van probiotica consumptie op de darmflora
samenstelling van mensen die lijden aan een chronische ontsteking van de dikke darm. Het blijkt
dat de gebruikte fluorescentietechnieken hierbij uitstekend voldoen en zodoende bij kunnen
dragen aan een kritische evaluatie van de effectiviteit van probiotica en in de toekomst mogelijk
ook van prebiotica.
Om levend de dikke darm te bereiken, moeten probiotica bestand zijn tegen de
omstandigheden in de maag en de dunne darm. Galzouten in het darmkanaal vormen een
belangrijke stress-factor voor deze bacterin, en er wordt verondersteld dat deze de celwand en
membraan beschadigen zodanig dat de bacterie dood gaat. Er bestaan echter grote verschillen in
galzoutgevoeligheid tussen bacterin. Om dit in detail te kunnen bestuderen werden twee soorten
bifidobacterin, Bifidobacterium adolescentis en Bifidobacterium lactis, blootgesteld aan oplopende
concentraties galzouten en werden vervolgens de effecten op de bifidobacterin met behulp van
flowcytometrie onderzocht. De combinatie van PI met cF liet verrassenderwijs zien, dat er drie
groepen fluorescent gelabelde bacterin konden worden aangetoond, namelijk PI-gekleurde rode
cellen, PI-cF gekleurde oranje cellen, en cF gekleurde groene cellen. De in het onderzoek
gebruikte flowcytometer kan zodanig worden ingesteld, dat de drie verschillend gekleurde
groepen bacterin kunnen worden gesorteerd. Vervolgens is onderzocht of de gesorteerde
bacterin konden worden gekweekt in het laboratorium. Hieruit bleek, dat de PI gekleurde cellen
niet, en de cF gekleurde cellen wel konden groeien. De PI-cF gekleurde cellen bleken alleen in
staat tot groei na een herstelperiode. Hieruit werd geconcludeerd dat de drie groepen
respectievelijk bestaan uit levende (cF gekleurde), dode (PI gekleurde) en beschadigde (cF-PI
gekleurde) cellen. Deze aanpak geeft dus gedetailleerde informatie over de mate van beschadiging
van bacterin na blootstelling aan stress en liet in dit geval zien dat B. lactis meer resistent is tegen
galzouten dan B. adolescentis. Dergelijke methoden zouden kunnen worden gebruikt bij de
selectie van probiotica, maar ook bij de analyse van de effectiviteit van prebiotica, omdat nu niet
alleen de aantallen bacterin worden bepaald, maar ook het percentage actieve cellen.
Het ultieme doel van dit onderzoek staat beschreven in hoofdstuk 7, namelijk het
detecteren, identificeren en bepalen van de activiteit van micro-organismen aanwezig in het
menselijk maagdarmkanaal. De verschillende technieken die zijn ontwikkeld en beschreven in de
158
Samenvatting

voorgaande hoofdstukken zijn hier samengebracht. Zo werden na een combinatiekleuring van
fecale monsters, de genoemde drie groepen gesorteerd, en werd vervolgens met behulp van
DGGE vastgesteld welke soorten bacterin aanwezig waren in de drie gekleurde groepen. De
levende groep bleek voornamelijk te bestaan uit bacterin die verwant zijn aan beschreven
butyraat producerende soorten, terwijl de dode groep voornamelijk Bacteroides, Ruminococcus,
en Eubacterium soorten bevatte. Deze aanpak liet ook grote verschillen zien bij de bifidobacterin.
B. adolescentis werd vooral gevonden in de PI groep, terwijl B. longum werd gevonden in alledrie
de groepen. Dit geeft aanwijzingen over de activiteit van genoemde bifidobacterin op
verschillende locaties in het darmkanaal, dat sommige bacteriesoorten zijn actief in het eerste deel
van het darmkanaal, terwijl andere (meer) actief zijn in het laatste deel, en in dit geval dus in fecale
monsters worden geclassificeerd als levend en beschadigd.
De ontwikkelde fluorescentietechnieken en de verkregen resultaten zoals beschreven in dit
proefschrift hebben nieuwe inzichten opgeleverd over de samenstelling en de activiteit van de
darmflora, inclusief de bacterin in de mucus. De ontwikkelde technieken vormen tevens een
uitstekende basis voor verder onderzoek aan de menselijke darmflora.
159
Samenvatting

.
ACKNOWLEDGEMENTS
This thesis has been 6 years in the making while my research and writing tried to catch
up with subject matters expanding faster than my working capacity. That I have been able to reach
some form of completion, however tentative, is due to the cooperation, help and support of a
number of people and institutions.
My deepest gratitude goes to my promoter Professor Willem de Vos for giving me the
opportunity to carry out the major part of my research in his laboratory. Willem, you have been
a nonstop source of energy and challenge through these last years. Your support and constructive
criticism have been crucial for the development of this thesis. A special thanks goes to my co-
promoter Dr. Tjakko Abee with whom the ideas of this thesis were discussed in depth. Thank you
Tjakko for the fruitful discussions, your guidance, your trust and also for your moral support
throughout all these years. I am also indebted to my co-promoter Dr. Elaine Vaughan whose
contribution has been important in bringing order and clarity towards the end of this thesis.
I would like to express a special gratitude to Professor Frans Rombouts without whom this
project would not have been started in the first place. Also I extend my thanks to Dr. Antoon
Akkermans who was involved in the start of the second phase of this project.
I am indebted to Prof. Abdelkader Chrif, my former director, for giving me the
opportunity and encouraging me to continue my PhD. Si Abdelkader you have been a real source
of motivation and strength during all the years I was working at ESIAT, especially during the very
difficult moments. Merci beaucoup.
During the first phase of this project, the interaction with my colleagues of the Food
Microbiology Laboratory was very beneficial for my adaptation to the Dutch way of working.
I will always remember the good times and friendly atmosphere of the lab and the very nice
working ambiance. I would like to thank in particular Pieter Breeuwer, Patrick Verbaarschot and
Christine Bunthof for sharing their expertise in fluorescent techniques and flow cytometry
with me. Also thanks to Suzanne Luppens, Jeroen Wouters, Jeroen Kiers, Birgit Hasenack,
Willem van Schaik, Ralf Hatemink, Chantal Kandhai and Aarieke de Jong for making a joyful
working environment in the lab, even during the most difficult periods. Bedankt allemaal voor
de hulp en gezelligheid.
Moving to the Microbiology Laboratory, I was very surprised by the diversity and multi-
cultural environment. It made it very easy for me to feel home. Many thanks to all the MoleEco
folks. I am very appreciative to Erwin Zoetendal, Hauke Smidt, Ineke Heikamp-de Jong and my
student Suzanne Verhaegh for their productive collaboration.
Hans and Wilma, I am very delighted that you are my paranimphs. Hans, I really enjoyed
working with you and thanks again for all you patience. Wilma, more than a colleague you
became a real friend, thanks for listening to me and sharing good momentum.
161
For hospitality, I would like to thank the Family Nienhuis for making me feel home. Eppy
you have been my second mother in Holland. Bedankt voor alles. Kier you were very enthusiastic
about my promotion, unfortunatelly you have left us so quickly. I wish you rest in peace.
Living in Wageningen was not always easy for me without the support of my dear friends
Ana Costa, Graa da Silveira Gilma Chitara Francesca OKane and Sara Martins Thanks girls for
the wonderful time, your continuous support ans cheeriness during all the ups and downs.
My special appreciation goes to my friend Dick Merx who stood by me throughout all these years.
Bedankt voor al je steun, geduld and liefde.
Last but not least, I would like to thank my family for all their love and support across the
miles. You have been always there whenever I needed you. Merci beaucoup. I would like to add
a special word of gratefulness to my sister Monia.
Tunis, July 23
th
, 2004
162
ABOUT THE AUTHOR
Kaouther Ben Amor was born on the 6
th
of December 1962 in Ksibet El
Medouini, Tunisia. She received her Baccalaureate (high school diploma) in
Biology and Mathematics from Lyce Bourguiba of Monastir in 1980. In
June 1984, she graduated from Institut National Agronomique de Tunis
(INAT), with an Engineering degree in Agronomy. From October 1984 until
December 1986, she worked at the Ministry of Agriculture in Tunis as
agronomist in the framework of sustainable rural development projects. In 1987 she received a
scholarship to pursue her MSc studies in Food Sciences at the University of Missouri, in the
United States. Upon her graduation in December 1989, she returned to Tunisia where she got a
position as a Food Technologist at the Department of Food Industries, Ministry of Agriculture.
In 1992 she joined the University 7 Novembre Carthage, where she holds a position as
assistant professor at Ecole Suprieure des Industries Alimentaires de Tunis. From 1992 until
1997, she was teaching food fermentation and took part in the research project Production of
Starter Cultures for Traditional Fermented Products in collaboration with the Centre Wallon de
Biologie Industrielle (CWBI, Lige, Belgium). In November 1997 she was accepted to pursue a
PhD program in collaboration between the Laboratory of Microbiology and the Laboratory of
Food Microbiology at the Wageningen University, The Netherlands. In 1999, she received a
Scholarship from the Islamic Development Bank and the Tunisian Ministry of Higher Education
to continue her PhD research under the supervision of Prof. Willem de Vos.
163
LIST OF PUBLICATIONS
Ben Amor K., and E. E. Vaughan. (2004) Methods to analyze the composition of intestinal
microbiota. In: Gastrointestinal microbiology. Ouwehand A., and E. E. Vaughan (eds). Marcel
Dekker, Inc, New York. In Press.
B Ben Amor K., and en Amor K., and W W. M de . M de V Vos. os. (2003) Molecular tools for the identification and quality
control of probiotics. Proceedings of the International Dairy Federation World Summit. Bruges,
Belgium, September.
Ben Amor K., I. Heikamp-de jong , S. Verhaegh, W. M. de Vos, and E. E. Vaughan. (2003)
Assessment and evaluation of the effects of two probiotic strains on fecal microbiota of patients
with inflammatory bowel disease. Poster. 2
nd
PROEUHEALTH workshop.
Derrien, M., K. Ben-Amor, E. E. Vaughan, and W. M. de Vos. (2004) Validation of a 16S rRNA
probe specific for the novel intestinal mucin-degrader Akkermansis muciniphila. Poster 3
rd
PROEUHEALTH workshop.
Ben Amor K., P. Breeuwer, P. Verbaarschot, F. M. Rombouts, A. D. L. Akkermans, W. M. de
Vos, and T. Abee (2002). Multiparametric flow cytometry and cell sorting for the assessment of
viable, injured and dead cells of bifidobacteria during bile salt stress. App. Env. Microbiol.
68: 5209-5216.
Vaughan E. E., M. C. de Vries, E. G. Zoetendal, K. Ben Amor, A. D. L. Akkermans, and W.
M. de Vos. (2002) Intestinal LABs. Antonie van Leeuwenhoek. 82:341-352.
Zoetendal E. G., K. Ben-Amor, H. J. M. Harmsen, Schut F., Akkermans A. D. L., and de Vos
W. M. (2002) Quantification of uncultured Ruminococcus obeum-like bacteria in human fecal
samples by fluorescent in situ hybridization and flow cytometry using 16S ribosomal RNA-
targeted probes. Appl. Envir. Microbiol. 68: 4225-4232.
Zoetendal E.G., A. von Wright, T. Vilpponen-Salmela, K. Ben-Amor, A. D. L. Akkermans, and
W. M. de Vos. (2002) Mucosa-associated bacteria in the human gastrointestinal tract are
uniformly distributed along the colon and differ from the community recovered from feces. Appl.
Envir. Microbiol. 68: 3401-3407.
Zoetendal E.G., K. Ben-Amor K., A. D. L. Akkermans, T. Abee, and W. M. de Vos. (2001)
DNA isolation protocols affect the detection limit of PCR approaches of bacteria in samples from
the human gastrointestinal tract. System. Appl. Micribiol. 24:405-410.
Ben Amor K., C. Cornelieus, A. Mahjoub, and P. Thonart. (1998) Identification de la flore
lactique du lait ferment (lben) et evaluation des composs aromatisants. Microbiol. Hyg. Ali.
10: 31.
164
Ben Amor K., and M. L. Fields. (1990) Production of niacin by microorganisms isolated from
fermenting corn meal. J. Food. Prot. 53:881.
RESEARCH WORK NOT PUBLISHED
Dallouli A., K. Ben Amor. (1997) Technologie de fabrication du yaghourt: tude comparative de
cinq ferments lactiques.
Douniama S.V., K. Ben Amor K. (1997) Suivi de la qualit de la Bire Cltia la SFBT.
Ayed L., C. Cornelieus, K. Ben Amor , P. Thonart, and A. Mahjoub. (1996) Contribution
ltude des potentialits aromatisantes des lactocoques isols du Lben.
Nouai R., K. Ben Amor. (1995) Controle de la qualit physico-chimique et microbiologique du
Lait UHT la S.L.P de Tunis.
Fezzani F. and K. Ben Amor. (1995) Controle microbiologique et physico-chimique des vins
tunisiens.
165
The research described in this thesis was part of the research program of the Graduate School
VLAG (Food Technology, Agrobiotechnology, Nutrition and Health Sciences)
Editing and printing : SERVICED S.A. (www.serviced.com.tn)
Financial Support: Islamic Development Bank, Wageningen University & EU Quality of life and
Management of Living Resources, PROGID project (QLK1-2000-00563).
Part of the printing costs of this thesis was supported by a grant from Dr Judith Zwartz
Foundation, Wageningen, The Netherlands.
166

Ecophysiology Intestinal

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ecophysiology Intestinal

Uploaded by

Copyright:

Available Formats

Microbial Eco-Physiology of the Human

You might also like