high-throughput diversity and functionality analysis of the gastrointestinal tract ... ·...
TRANSCRIPT
doi:10.1136/gut.2007.133603 2008;57;1605-1615 Gut
E G Zoetendal, M Rajilic-Stojanovic and W M de Vos
analysis of the gastrointestinal tract microbiotaHigh-throughput diversity and functionality
http://gut.bmj.com/cgi/content/full/57/11/1605Updated information and services can be found at:
These include:
References
http://gut.bmj.com/cgi/content/full/57/11/1605#BIBL
This article cites 105 articles, 58 of which can be accessed free at:
serviceEmail alerting
the top right corner of the article Receive free email alerts when new articles cite this article - sign up in the box at
Notes
http://journals.bmj.com/cgi/reprintformTo order reprints of this article go to:
http://journals.bmj.com/subscriptions/ go to: GutTo subscribe to
on 12 November 2008 gut.bmj.comDownloaded from
High-throughput diversity andfunctionality analysis of thegastrointestinal tract microbiotaE G Zoetendal,1,2 M Rajilic-Stojanovic,1 W M de Vos1,2,3
1 Laboratory of Microbiology,Wageningen University,Wageningen, The Netherlands;2 TI Food and Nutrition,Wageningen, The Netherlands;3 Department of Basic VeterinaryMedicine, University of Helsinki,Helsinki, Finland
Correspondence to:Dr E G Zoetendal, Laboratory ofMicrobiology, WageningenUniversity, Dreijenplein 10, 6703HB Wageningen, TheNetherlands; [email protected]
ABSTRACTThe human gastrointestinal (GI) tract microbiota plays apivotal role in our health. For more than a decade a majorinput for describing the diversity of the GI tract microbiotahas been derived from the application of small subunitribosomal RNA (SSU rRNA)-based technologies. Thesenot only provided a phylogenetic framework of the GI tractmicrobiota, the majority of which has not yet beencultured, but also advanced insights into the impact ofhost and environmental factors on the microbiotacommunity structure and dynamics. In addition, itemerged that GI tract microbial communities are host andGI tract location-specific. This complicates establishingrelevant links between the host’s health and the presenceor abundance of specific microbial populations and arguesfor the implementation of novel high-throughput technol-ogies in studying the diversity and functionality of the GItract microbiota. Here, we focus on the recent develop-ments and applications of phylogenetic microarrays basedon SSU rRNA sequences and metagenomics approachesexploiting rapid sequencing technologies in unravelling thesecrets of our GI tract microbiota.
The human gastrointestinal (GI) tract consists ofdifferent and connected organs that are involved insupplying the human body with nutrients andenergy sources by the conversion and absorption offood. The human GI tract has a well-knownanatomical architecture and is approximately 7 mlong with a surface area of approximately 300 m2
in adults.1 Since it is continuously exposed to theoutside environment, the GI tract has severalprotection systems. These include the low pH inthe stomach, the coverage of the complete GI tractwith a mucus layer, an enormous army of immunecells that lie beneath this mucus layer, and thepresence of commensal microbes that abundantlycolonise the GI tract. These microbial communitiesare collectively called commensal microbiota or GItract microbiota. This GI tract microbiota isdistributed along the entire GI tract, with thedensity and diversity increasing from the stomachto the colon. The acknowledged functions of theGI tract microbiota include the conversion ofindigestible food components2 3 and the produc-tion of essential vitamins and co-factors.However, our understanding of the GI tractmicrobiota is fragmented, which is due to thelimited accessibility of the different parts of theGI tract and the immensely complex and diversecommunity structure of the GI tract microbiota
which differs between individuals, intestinallocation and age groups.4
THE GASTROINTESTINAL TRACT MICROBIOTAThe myriad of microbial cells in the GI tract, whichoutnumber our body cells by a factor of at least 10,has a large species diversity and consists ofcultivated species as well as those that have notyet been cultured. A recent estimation of thecultivable fraction of the GI tract microbiotaincludes 442 bacterial, three archaeal, and 17eukaryotic species.5 However, it is evident thatany figures that report the known diversity of theGI tract microbiota are continuously outdated dueto the steady reporting of novel intestinal inhabi-tants. The majority of the currently availablecultivated representatives were discovered follow-ing the introduction of anaerobic cultivationtechnologies that are still being used as standardtools.6 Novel bacteria that are still being isolatedfrom the human GI tract include a wide variety ofbutyrate-producing bacteria belonging to the phy-lum Firmicutes7 8 and bacteria belonging to thephylum Bacteroidetes,9–12 which are abundant GItract phyla. In addition, the use of alternativecarbon sources brought the discovery of the first GItract representatives of the phyla Verrucomicrobiaand Lentisphearae, Akkermansia muciniphila andVictivallis vadensis, respectively.13 14 A muciniphila,especially, is a relevant isolate since it is a specialistin mucin degradation and a very common andabundant inhabitant of the GI tract which haspreviously not been noted because of its small sizeand specific carbon source requirements.13 15 16
Recently, novel culturing methods have beendeveloped which exploit microbeads or multi-plexed solid surfaces and allow unconventionaland high-throughput culturing approaches.17 18
Such methods enable the simultaneous single-cellcultivation of thousands of microbes, which isessential for an appropriate study of the complexand dense GI tract microbiota. Application of high-throughput culturing is expected to expand ourknowledge of this neglected yet important elementof the human body. However, in the course ofevolution many GI tract microbes have developedintimate relations with the host and with eachother, which makes microbes dependent on themetabolic activity of another member of theecosystem and, therefore, almost impossible togrow into pure culture.19 As an alternative, the use
Recent advances in basic science
Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603 1605
on 12 November 2008 gut.bmj.comDownloaded from
Tabl
e1
Ove
rvie
wof
phyl
ogen
etic
dist
ribut
ion
ofga
stro
inte
stin
altr
act
inha
bita
nts
(A–P
,se
eta
ble
foot
note
)th
atha
vebe
ende
tect
edby
cultu
ring
and
bysm
alls
ubun
itrib
osom
alR
NA
gene
libra
ryse
quen
cean
alys
is.
The
cut-
off
valu
eof
SS
UrR
NA
sequ
ence
sim
ilarit
yth
atis
used
for
phyl
otyp
ede
finiti
onis
97%
.
AB
CD
EF
GH
IJ
KL
MN
OP
Ref
eren
ce(f
irst
auth
or)
Cul
ture
d33
310
208
1341
111
06
674
182
10
Raj
ilic-
Sto
jano
vic5
Hea
lthy
tota
lm
icro
biot
a0
10
153
120
00
00
00
00
0Zo
eten
dal2
2
015
040
1324
00
00
00
00
00
Sua
u23
05
017
413
00
00
00
00
00
Bon
net2
4
91
08
05
00
00
20
00
00
Favi
er2
5
00
015
39
00
00
00
00
00
Wan
g26
08
025
415
00
26
50
00
00
Wan
g27
112
043
728
20
03
60
00
10
Wan
g28
818
011
526
580
00
22
10
00
0H
ayas
hi2
9
46
045
1221
00
00
00
00
10
Hay
ashi
30
016
055
1517
00
01
42
00
00
Hay
ashi
31
09
037
924
00
02
10
00
00
Hol
d32
1062
130
181
931
04
52
41
01
0Ec
kbur
g33
*
1659
025
110
084
00
04
54
00
20
Sau
nier
,un
publ
ishe
d
Ulc
erat
ive
colit
is0
10
299
140
00
00
00
00
0H
utso
n,un
publ
ishe
d
Cro
wn
423
011
134
491
00
13
00
01
0M
angi
n34
Spe
cific
prim
ers
00
020
00
00
00
00
00
00
Hei
lig3
5
30
00
00
00
00
00
00
00
Sat
okar
i36
00
014
29
00
00
07
00
02
Raj
ilic-
Sto
jano
vic5
Spe
cific
subs
trat
es/
activ
ity
10
024
320
00
00
00
00
00
Bar
ceni
lla7;
Dun
can3
7
00
022
119
00
00
21
00
10
Der
rien3
8
69
059
1137
00
10
30
00
00
Leitc
h39
*The
cut-
off
valu
eof
smal
lsu
b-un
itrR
NA
sequ
ence
sim
ilarit
yth
atis
used
for
phyl
otyp
ede
finiti
onis
99%
.A
,A
ctin
obac
teria
;B
,B
acte
roid
etes
;C
,C
yano
bact
eria
;D
,Fi
rmic
utes
(all)
;E,
Clo
strid
ium
clus
ter
IV;
F,C
lost
ridiu
mcl
uste
rX
IVa;
G,
Fuco
bact
eria
;H
,Le
ntis
phae
rae;
I,A
lpha
-Pro
teob
acte
ria;
J,B
eta-
Prot
eoba
cter
ia;
K,G
amm
a-Pr
oteo
bact
eria
;L,
Del
ta-P
rote
obac
teria
;M
,Ep
silo
n-Pr
oteo
bact
eria
;N
,S
piro
chae
tes;
O,
Ver
ruco
mic
robi
a;P,
TM7
phyl
um.
Recent advances in basic science
1606 Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603
on 12 November 2008 gut.bmj.comDownloaded from
of small subunit ribosomal RNA (SSU rRNA) andits corresponding gene has become established forthe classification and phylogenetic analysis ofmicrobes. The SSU rRNA gene includes approxi-mately 1500 bp, which is a sufficient size forcomparative sequence analysis. SSU rRNA ispresent in literally all organisms, and due to itsconserved function its sequence has remainedrelatively conserved throughout evolution.Nevertheless, it contains nine variable regions thatcan be used for the identification and differentia-tion of specific microbial species. In fact, all lifeforms can be classified using the SSU rRNAsequences and this has resulted in the discoveryof the Archaea as the third domain of life (Bacteriaand Eucarya are the other two domains) and theconstruction of the revolutionary ‘‘tree of life’’.20 21
SSU rRNA and its corresponding gene can beobtained directly from any environmental samplewithout cultivation procedures and this allows thedetection of basically all members of an ecosystem,including those that cannot be cultured yet.Currently, over 400 000 SSU rRNA sequences areavailable in DNA databases, which is far more thanfor any other gene (http://www.arb-silva.de(accessed 11 July 2008)). The first breakthroughdiscovery of the SSU rRNA based analyses wasthat cultivation represents only a small fraction(estimated to be between 10 and 50%, dependingon the study) of the true microbial diversity withinthe GI tract (table 1).
Exploring the diversity of an ecosystem bycomparative sequence analysis is based on identi-fication of species-level phylogenetic types, ie,phylotypes. Since variation of the SSU rRNA genesequence is present within the members of thesame species,40 41 and sometimes even betweendifferent copies of the SSU rRNA gene within thesame microbial genome,42 phylotypes are defined asgroups of SSU rRNA gene sequences with a certainlevel of similarity. However, the cut-off value ofSSU rRNA sequence similarity that is used forphylotype definition is not consistent betweendifferent studies and varies between 97 and 99%.The higher the cut-off value, the higher is thenumber of distinct phylotypes that will be foundin the same clone library.43 For diversity studies theSSU rRNA sequences can be targeted usingdifferent, often complementary, approaches.Commonly applied approaches include cloningand subsequent sequencing of SSU rRNA genes,fingerprinting approaches such as denaturing gra-dient gel electrophoresis (DGGE) and fluorescencein situ hybridisation (FISH), which allows visuali-sation of specific microbes. These approaches arerelatively low throughput and the choice ofapproach will mainly depend on the question tobe answered. Cloning of SSU rRNA genes deliversthe most detailed phylogenetic information, fin-gerprinting is most often used to compare micro-bial communities and monitor their dynamics,while FISH is the preferred approach to quantifyspecific microbial populations. The combinationof these approaches has greatly advanced the
phylogenetic framework of the enormous micro-bial diversity and novel insights into its ecology. Ithas to be realised that neither of the methods forstudying GI tract microbiota gives absolutelyaccurate information about the diversity of thisecosystem, including the ‘‘gold standard’’ cloningand sequencing-based studies. For example, theActinobacteria, which belong to the high G+CGram-positive bacteria, are frequently undetectedor under-represented in the clone libraries,although they represent a dominant fraction ofGI tract microbiota, as has been demonstrated byFISH using SSU rRNA-targeted oligonucleotideprobes.44 45
The composition of GI tract microbiota is mostoften studied by analysing faecal samples, but themicrobiota of other intestinal samples, such asspecimens from the human colon and ileum, havealso been characterised.27 32 33 46 47 Fingerprintingstudies based on DGGE analysis of SSU rRNAgene amplicons indicated that the dominantcommunity in faecal samples in about half of thesubjects studied does not necessarily represent thatfound in other parts of the GI tract, including thecolonic mucosa.48 49 This most likely reflects thefact that the number of microbes that areassociated with the mucosa is small compared tothe number of microbes present in the colonlumen. Although most likely primed by mucosa-associated microbes, the composition of the lattercan be affected by the growth on specific sub-strates that reach the colon. If these substratesbecome limited at the end of the colon, this couldalso explain why certain abundant microbialspecies in faeces are not viable any more and thisargues for studying samples derived from severalGI tract regions.50
Studies that employ SSU rRNA gene sequenceanalysis are rapidly expanding our knowledgeabout the diversity of the GI tract microbiota,and only a decade after their introduction, thenumber of molecularly detected GI tract phylo-types has by far outnumbered the cultivated GItract species (fig 1). An intensive survey of thepublications focusing on the diversity of thehuman GI tract microbiota demonstrated thatfrom more than 1200 microbes described, only12% were recovered by application of bothmolecular and cultivation-based approaches, whilethe vast majority (,75%) was detected solely as anSSU rRNA sequence.51 These values have to betaken with caution as it has also to be realised thatmost of the SSU rRNA sequences are retrievedfrom intestinal samples, mostly faeces, from only adozen individuals. Considering the location andindividual-specific microbial composition of the GItract, this indicates that the 1200 species describedonly reflect a fraction of the true microbialdiversity, which is estimated to consist of up to1000 microbes per individual and more than 5000microbes in total.23 33 51 52 This implies that we arestill at the beginning of describing the GI tractmicrobial diversity. Hence, there is a need todevelop and apply high-throughput approaches in
Recent advances in basic science
Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603 1607
on 12 November 2008 gut.bmj.comDownloaded from
further culturing studies, SSU rRNA sequencing andmicrobial diversity analysis, notably at differentlocations in the GI tract. These are essential toenable linking microbial populations to host-relatedfactors, such as host genotype, age, health status,geographical location, ethnic origin, and diet.
PHYLOGENETIC MICROARRAY ANALYSIS OFGASTROINTESTINAL TRACT MICROBESAs indicated above, the human GI tract contains amicrobiota, the diversity of which is beyond ourimagination given the total number of microbes,and their location- and individual-specific composi-tion. This complicates establishing links betweenmembers of the microbiota and GI tract disorders.At the moment, causal effects have been deter-mined only for the well-studied and culturedpathogens such as Helicobacter pylori, Listeriamonocytogenes, Clostridium difficile and members ofthe Enterobacteriaceae.53 In addition, correlationshave been suggested between the health status ofindividuals and the composition and activity oftheir microbiota. Irritable bowel syndrome (IBS)and intestinal bowel diseases (IBDs), such asCrohn’s disease and ulcerative colitis, have fre-quently been associated with a rather unstable anddisturbed microbiota composition in contrast tohealthy individuals.54–57 These differences have tobe taken with caution since they can be influencedby several factors, including the use of medicinethat may affect the microbiota. In addition, theunstable microbiota of patients with IBD can becaused by the disease status of the patients asdifferences in community structures were observedin patients who relapsed versus those who were inremission.54–57 Last but not least, the stability of themicrobiota in healthy individuals decreases whenthe time span between the sampling increases, anddiffers between microbial populations.5 It wasdemonstrated that a significant fraction of micro-bial phylotypes is continuously present in the GItract of a person over a 10 year time span, whichindicates that the microbiota consists of a stableindividual core of colonising microbes surrounded
by temporal visitors.5 Despite these individualdifferences, there are indications that some micro-bial phylotypes are shared by different people andthis led to the hypothesis that besides theindividual core of microbes representing the stablecolonisers in healthy individuals, humans also sharea common core of microbes in their GI tract.58 Thishypothesis expands a previously proposed hypoth-esis that the human gastrointestinal microbiota isdiverse but it is dominated by a limited number ofbacterial species in everybody.59–61
As it is not yet possible to define the microbiotaof a healthy intestinal tract, it is equally difficult todefine the microbiota associated with an intestinaldisorder. Another factor that complicates such anassociation is the fact that these GI tract disordershave a largely undefined, complex and possiblyheterogeneous aetiology in which, in addition tothe microbiota, host genetics and environmentalfactors also play a role. Moreover, our inability tocultivate all members of the microbiota makes itimpossible to formulate hypotheses about the roleof uncultured microbes in health and disease as a(partial) SSU rRNA sequence provides no informa-tion about the function of an organism. Last butnot least, the individual-specific composition of theGI tract microbiota is also an important factor thatcomplicates the establishment of links betweenmicrobes and the health status of the host.22 62
Since different people are recruited by the differentresearch groups, for which each of them has afavourite target microbe or methodology formicrobiota analysis, comparisons between differ-ent studies are basically impossible. Moreover, asthe number of persons tested is still low, it isevident that high-throughput analyses of themicrobiota using standardised methods are neededto make statistically relevant links between thepresence or quantity of uncultured bacterialpopulations and GI tract disorders.
The most commonly used high-throughputanalytical method is DNA microarrays. DNAmicroarrays are basically glass surfaces, each thesize of a microscopic slide, that are spotted withthousands of covalently linked DNA probes. TheseDNA microarrays can be hybridised with DNA orRNA and the most current applications includemonitoring gene expression (transcriptional profil-ing) and detecting DNA sequence polymorphismsor mutations in genomic DNA. However, it is alsopossible to use DNA microarrays for diversityanalysis (fig 2). Guschin and colleagues63 describedthe first phylogenetic microarray (or diversitymicroarray) in which oligonucleotides complemen-tary to SSU rRNA gene sequences of nitrifyingbacteria were used to detect and identify thesebacteria in environmental samples. Thereafter,DNA microarray technology has been implemen-ted in a variety of ecological studies in which notonly SSU rRNA genes, but also antibiotics resis-tance genes were used as targets.64–69 Recently, thegreat potential of using phylogenetic microarraysto detect thousands of microbes simultaneouslywas demonstrated by DeSantis and colleagues.70
Figure 1 Cumulative number of specific gastrointestinal(GI) tract phylotypes detected with culture-dependent andculture-independent approaches based on the dataprovided in table 1.
Recent advances in basic science
1608 Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603
on 12 November 2008 gut.bmj.comDownloaded from
This study illustrated that the use of phylogeneticmicroarrays is more powerful for the analysis ofmicrobial community structure than a canonicalclone library approach.70 Besides this generalphylogenetic microarray, there are also specificmicroarrays that focus on the microbial commu-nities in specific ecosystems5 71 72 (table 2). Theseinclude a phylogenetic microarray that targets themicrobiota of the oral cavity and two microarraysthat focus on the microbiota of the human GItract.
The first studies in which phylogenetic micro-arrays were used to characterise the GI tractmicrobiota have illustrated the power of such anapproach to gain insights into the structure andpopulation dynamics in the GI tract (box 1). In onestudy 14 newborn babies were monitored over ayear and the results showed that the microbiota ofthese infants is relatively simple and individual-specific, but chaotic in the early months of lifefollowed by a similar pattern of developmenttowards a more complex adult-like microbiota.77
These results confirm and expand earlier studiesperformed using DGGE analysis.25 The directphylogenetic identification enabled by the micro-array brought to light another remarkable observa-tion: bifidobacteria were found to be only a minorfraction of the total microbiota of the infantsanalysed. This contrasts with observations in severalprevious studies which showed that bifidobacteriaare the most dominant group in infants.78 79
Although such a surprising finding could be partiallyexplained by technical biases (eg, inadequate primersequence or inefficient cell lysis), a biologicalexplanation is also possible as infants from differentstudies originated from different continents andreceived different paediatric practices and diets.77
The disparity of findings of different studiesillustrates our present inability to define the normalGI tract microbiota even of its simplified form,which at present is in early infancy.
Besides focusing on the GI tract microbiota inhealthy individuals, the first phylogenetic micro-array studies on people with a GI tract disorderhave also been performed.5 As indicated before, thecommon GI tract disorders, such as IBD and IBS, arecomplex since they are influenced by the microbiotaand host genetic and environmental factors80–83 and,therefore, comprehensive and high-throughputapproaches are needed to gain insight into this
complexity. The phylogenetic microarray analysisdemonstrated that persons having a GI tract disorderhad distinct microbiotas compared to healthyindividuals and that the severity of the disease iscorrelated with the significance of the differencewith the healthy group.5 A striking observation wasthat patients with IBS had increased heterogeneityamong the microbiota composition compared tohealthy individuals.5 This could be explained by thefact that IBS is known as a heterogeneous disorderbased on varying clinical symptoms.84 Despite thisincreased heterogeneity of the GI tract microbiota, aremarkable observation is that, in IBS, members ofthe Firmicutes were affected the most dramatically.This is in line with the previous phylogeneticmicroarray studies which indicated that theFirmicutes are more strongly affected by environ-mental changes than are Bacteroidetes andActinobacteria.5 This suggests that the alterationsin the Firmicutes could be candidate biomarkers forIBS. The Firmicutes is the most diverse andabundant phylum within the GI tract microbiotaconsisting of a wide variety of uncultured organisms.As a result, the functionality of the majority ofFirmicutes is not known and it is difficult tohypothesise what it might be. Therefore, explainingthe observed trends is only possible for a fewmicrobial groups, such as Roseburia spp., which areamong the butyrate-producing isolates from thegut.85 For instance, Roseburia spp. were found to beincreased in patients suffering from diarrhoea-predominant IBS,5 which adds to the controversyabout the effect of butyrate on human health.86 87
With respect to these GI tract disorders it is evidentthat obtaining more insight into the function ofFirmicutes is crucial. To overcome the problem ofthe fact that the majority of Firmicutes will remainuncultured in the near future, culture-independentapproaches to study functionality are needed.
The first applications of phylogenetic micro-arrays have already provided novel insights into themicrobiota composition and dynamics in relationto health and disease and these approaches arepromising for future research. However, theapplication of phylogenetic microarrays also hasits limitations, as does any other microbiologicalapproach. Phylogenetic microarray analyses aredependent on the isolation of nucleic acids andsubsequent polymerase chain reaction (PCR)amplification of SSU rRNA genes, which are
Table 2 Overview of small subunit ribosomal RNA (SSU rRNA)-based phylogenetic microarrays which are fruitful to gain insight into the microbialdiversity of the human gastrointestinal (GI) tract microbiota
Target microbes Number of target organisms SSU rRNA database Number of probes Reference (first author)
Human commensal isolates 40 bacterial species GenBank 120 Wang61
Sulfate-reducing bacteria All recognised lineages ARB* 132 Loy65
All microbes 842 prokaryotic subfamilies Greengenes, 15 March 2002 version{ 297851 DeSantis70
Human oral cavity microbes NA NA NA Smoot72
Known GI tract commensals and medicallyrelevant microbes
1590 bacterial and 39 archaealspecies
prokMSA SSU rDNA sequence database,2004 version{
10500 Palmer73
Human GI tract microbiota 1140 bacterial species Human GI tract Microbiota database,2006 version1
4809 Rajilic-Stojanivic5
*Described by Ludwig et al.74 {Described by DeSantis et al.75 {Described by DeSantis et al.761Described by Rajilic-Stojanivic et al.51 NA, not available.
Recent advances in basic science
Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603 1609
on 12 November 2008 gut.bmj.comDownloaded from
vulnerable to technical biases. These are, unfortu-nately, general drawbacks of culture-independenttechnologies and should be minimised as much aspossible. Furthermore, phylogenetic microarrayshave a dynamic range that only covers the dominantmicrobes present in the GI tract. On the other hand,at present, there are numerous group- and species-specific quantitative PCR (qPCR) assays that areuseful for quantification of different phylogeneticgroups belonging to the human GI tract micro-biota.88–90 These specific PCR protocols can becombined with phylogenetic microarray analysis,which will allow determination of the diversity andrelative abundance within low abundant groups.This could be of special interest for studying thepopulation dynamics of pathogens or probiotics asthey are usually present in low numbers.
In conclusion, the first phylogenetic microarray-based studies provided novel insights into the GItract ecology by demonstrating correlations betweencertain microbial populations and host factors. Todetermine the significance of a certain correlation,the next questions to be addressed are whetherthis correlation reflects a causal relation and, if so,what mechanisms underlie the observed effects.Answering this type of question cannot be done bySSU rRNA gene-based approaches, but need theintegration of functional-based approaches, includ-ing the application of the so-called meta-‘‘omics’’approaches in GI tract research, as will be discussedin the next section.
METAGENOMICS AND OTHER COMMUNITY-BASED APPROACHESSSU rRNA-based approaches are fruitful fordescribing the microbial diversity in the humanGI tract and finding potential links betweenmicrobes and a certain health status. However,
the results obtained with such approaches cannotbe interpreted beyond the description of themicrobial diversity, since potential functions ofthese microbes cannot be extracted from SSUrRNA data. This means that a correlation betweena disease status and the presence of a microbecannot explain whether its presence is the result orthe cause of the disorder. This means that otherapproaches are needed to gain insight into thepotential roles of microbes in the human GI tractand how they are related to health and disease. Inthis respect, much insight has been gained fromisolates that are known as pathogens and havebeen well-characterised in the laboratory. Animaland in vitro models have shed some light on thestrategies these microbes have to interact with thehost and, moreover, the availability of theirgenome sequences allows the discovery of genesinvolved in these interactions. Besides pathogens,commensal bacteria and their role in the GI tracthave also been studied in a similar way. Thepioneering work of Gordon and co-workers91–94 andthat of others showed that microbial colonisationimproves nutritional and defensive functions ofhost. Moreover, the impact of the host on themicrobe was also studied.94 95 These so-calledreductionist approaches provide insight into newgenes and functions of individual organisms, whichserve as models for community based studies.
Model systems have provided detailed insightsinto mechanisms that underlie the communicationbetween host and microbe. However, there is still ahuge gap between understanding these interactionsin a model system and that in a complex ecosystem,such as the human GI tract. One way to gain insightinto potential functions and activities of microbeswithout the need of cultivation is by performingmetagenomics and other community approaches
Figure 2 Schematicrepresentation of high-throughput analysis ofhuman gastrointestinal (GI)tract microbiota via bruteforce sequencing andphylogenetic microarrayanalysis. SSU rRNA, smallsub-unit ribosomal RNA.
Recent advances in basic science
1610 Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603
on 12 November 2008 gut.bmj.comDownloaded from
(box 1). Metagenomics is a DNA-based approach togain insights into the genetic potential of microbialcommunities, while the other meta-‘‘omics’’approaches focus on activity biomarkers, such asmessenger RNA, proteins or metabolites (fig 3).Metagenomics is defined as the study of collectedgenomes from an ecosystem that can be used tostudy the phylogenetic, physical and functionalproperties of microbial communities.96
Metagenomics is performed by extracting DNAfrom the microbial community followed by cloningof the DNA fragments in a suitable host (usually Ecoli) using a vector, such as fosmid or bacterialartificial chromosome vectors. This results in ametagenomic library that can be used for sequence-driven or function-driven analysis. Sequence-drivenanalyses are basically performed to obtain a snap-shot of the genetic diversity of an ecosystem, whilefunction-driven analyses are done to screen thelibrary for novel enzymes or particular functions ofinterest.
Function-driven metagenomics offers great pos-sibilities to discover new classes of genes withspecific functions. The screening for these featuresrequires functional transcription and translation ofthe genes, which are located in the metagenomeclone, in the host that was used for the metagen-ome library construction. E coli is the mostcommonly used host for construction of themetagnomic library and it is predicted that it canexpress up to 40% of the functional potential fromrandomly cloned environmental DNA.97 Function-driven metagenomic studies based on enzymeactivity assays have already led to the discoveryof novel activities, such as b-glucanases in thecolon of mice,98 bacterial hydrolases in rumen,99
and antibiotic resistance in the oral cavity.100
Recently, a phenotyping approach was used toscreen for metagenomic clones that contain capa-cities to modulate the growth of epithelial cells invitro.101 This approach showed that the identifica-tion of potentially novel mechanisms of host–microbe interactions in the GI tract is possible.These types of screening methods can be very
useful to screen for beneficial or harmful functionsthat are present in the GI tract microbiota.Subsequently, the expression of the genes thatare responsible for the function under study can bemonitored in situ and serve as potential biomarkersfor intervention studies.
It has to be realised that functional screening isquite laborious since relatively large metagenomiclibraries have to be screened to obtain a handful ofpositives per enzyme screen.99 To overcome thisdrawback, a high-throughput screening approach,termed substrate-induced gene-expression screen-ing was developed, which allowed fluorescence-activated cell sorting, in which catabolic geneswere activated by various substrates.102 Althoughthis approach looks promising for high-throughputscreening of human GI tract microbiota to discovernovel enzyme-encoding genes, the positive identi-fication relies on several host requirements, whichinclude the recognition and transport of theinducing substrate and the ability of the host toexpress the corresponding genes located on thecloned insert.
Sequence-driven metagenomics approaches areused to create a catalogue of the genetic potentialthat is present in an ecosystem, which can befruitful to gain insight into the functionality of theparticular ecosystem. Sequence-driven metage-nomics has already been applied to study a varietyof ecosystems and the first metagenomic approachthat was performed on the human GI tractdescribed the diversity of the viral community inhuman faeces.103 This study revealed that the viralcommunity in faeces is very diverse consisting ofapproximately 1200 viral genotypes from whichthe majority of viral sequences had the highestsequence similarity to phages that are known toinfect Gram-positive bacteria. As mentioned pre-viously, Gram-positive bacteria, which include thephyla Firmicutes and Actinobacteria, were identi-fied as dominant members of the GI tractmicrobiota in SSU rRNA gene-based studies. Thefirst prokaryotic sequence-driven metagenomicapproach focused on the SSU rRNA sequencesthat are represented in the metagenomic librariesfrom pooled faecal samples from healthy indivi-duals and from patients with Crohn’s disease.104
This study demonstrated that Firmicutes aresignificantly reduced in complexity in patientswith Crohn’s disease compared to healthy sub-jects. This is in line with SSU rRNA gene-basedobservations, including those in which phyloge-netic microarrays were used.5 In another study the‘‘mobile metagenome’’ of the human GI tractmicrobiota was investigated. In this study trans-poson-aided capture was developed and applied tothe study of the plasmid-encoded genes that arepresent in the human GI tract.105 In addition togenes involved in replication and mobilisation ofthe plasmids, genes encoding for phosphoesteraseor phosphohydrolase enzymes could be detected.Besides these targeted sequence-driven metage-nomic approaches, large-scale sequence-drivenmetagenomic approaches have also been applied
Box 1 Novel insights that have been gained from high-throughputanalysis of the human gastrointestinal (GI) tract microbiota
c The human GI tract microbiota is composed of numerous uncultured microbesand predominated by Firmicutes, Bacteroidetes and Actinobacteria
c In healthy adults the human GI tract microbiota fluctuates around a stableindividual core of phylotypes that are affected by host genetics, environmentaland stochastic factors
c In infants the GI tract microbiota is succeeding from an unstable chaoticcommunity towards a stable adult community
c A reduction in the abundance and diversity of Firmicutes is frequentlyassociated with intestinal bowel diseases and irritable bowel syndrome
c The human GI tract microbiome is enriched in functions that are essential tothe human host
c The human GI tract microbiome from healthy human adults consists of afunctionally uniform core and is a hotspot for gene transfer
Recent advances in basic science
Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603 1611
on 12 November 2008 gut.bmj.comDownloaded from
to investigate the genetic potential in the GI tractecosystem.106 107 The first study demonstrated thatmany genes in the colonic microbiota representfunctions that are essential to the host, such asvitamin production, which have often beenascribed to the microbiota.106 Remarkably,sequences affiliated to Bacteroidetes, one of thedominant microbial groups in the colon, were notdetected in this library, which indicated that celllysis, DNA extraction and cloning procedures canhave a major impact on the genetic diversity thatis represented in a library. Recently, a comparativemetagenomics approach was described to compareand contrast the genetic diversity in the humancolon of 13 subjects, including adults andinfants.107 This study indicated that the humanmicrobiota is a hot spot for horizontal genetransfer. Moreover, this study demonstrated thatthe genomic features of infants were less complexand individual-specific, while those of adults weremore complex with high functional uniformitybetween individuals. The latter finding confirmsthe proposal that all human GI tract microbiotasshare a common core of microbes, despite theirindividual-specific composition.58
Sequence-driven approaches result in a wealthof scattered pieces of sequences and this meansthat special tools for post-metagenomic analysisare indispensable. Although software packageshave been developed to re-assemble the genomicfragments that are sequenced, a major limitationof metagenomics is that the genetic diversity inan ecosystem is enormous. Despite the contin-uous development of novel, relatively cheap,sequencing technologies, such as 454 pyrosequen-cing,108 which will result in increased numbers ofmetagenomic sequences, at the moment it isalmost impossible to obtain reasonable coverage
for complex ecosystems such as the human GItract microbiota. Nevertheless, this raw sequenceinformation can be analysed by comparativemetagenomics and these types of analyses canalready reveal the identification of genes that arespecific or enriched in a certain ecosystem orniche.109 110
It has to be realised that the detection of genes ina metagenomic library does not necessarily meanthat these are functionally important. Therefore,metagenomics should be considered more ascatalogues for activity-based approaches and thatother meta-‘‘omics’’ approaches, which use RNA,proteins and metabolites as targets, are betterapproaches to gain insight into the activity andfunctionally of the microbes in an ecosystem (fig 3).These meta-‘‘omics’’ approaches are still in theirinfancy,111 but we expect that they will beextensively used in the near future.
CONCLUDING REMARKS AND FUTUREPERSPECTIVESFor more than a decade it has been recognised thata major part of the human GI tract microbiota hasnot been characterised by cultivation. This has ledto the implementation of sequence analysis of SSUrRNA and its corresponding gene in studying thehuman GI tract microbiota diversity and this hasprovided tremendous expansion of our knowledgeabout the ecology of the GI tract. However, theseapproaches have indicated that the GI tractmicrobiota is individual- and location-specific,and that its diversity is enormous with thousandsof novel microbial species to be discovered.51 Thisargues for the introduction of novel high-through-put and comprehensive technologies for studyingthe microbiota in the human GI tract, as describedin this paper.
The implementation of high-throughput phylo-genetic microarrays allows the simultaneousanalysis of thousands of microbes in a singleexperiment and, therefore, is very attractive forstudying the population dynamics of the GI tractmicrobiota in health and disease. This resulted inthe discovery that the development of the micro-biota in infants is initially chaotic but stabilisestowards an adult-like community after 1 year.77 Inaddition, phylogenetic microarray analysis indi-cated that the human microbiota fluctuates aroundan individual core of stable colonisers.5 Last, butnot least, significant links have been establishedbetween the presence or abundance of specificgroups of microbes and GI tract disorders such asIBS and IBD.5 Therefore, it is already evident thatthe implementation of phylogenetic microarrays inGI tract research will increase our knowledge in thenear future. Nevertheless, the up-to-date status ofphylogenetic microarrays will always depend onthe discovery of novel GI tract inhabitants and,therefore, ongoing sequencing of SSU rRNA genelibraries and cultivation of the novel GI tractinhabitants are indispensable.
It is evident that phylogenetic microarrays willenable us to make correlations between microbial
Figure 3 Schematic representation of the metagenomics and other community-based‘‘omics’’ approaches. SSU rRNA, small subunit ribosomal RNA.
Recent advances in basic science
1612 Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603
on 12 November 2008 gut.bmj.comDownloaded from
groups and characteristics of the host. However,this will not lead to extrapolation of microbialfunctions as the majority of GI tract microbes areonly known as a partial SSU rRNA gene. Therefore,metagenomics and other meta-‘‘omics’’ approachesare needed to gain insight into the genetic potentialand activity of GI tract microbiota. The field of thesemeta-‘‘omics’’ is hardly 5 years old and theirapplication in the study of the human GI tract iseven younger. Therefore, the analysis and interpre-tation of data derived from meta-‘‘omics’’approaches are still in infancy. We expect manynovel technological procedures to be developed andimproved in the coming years, not only with respectto wetlab technologies, such as pyrosequencing andfunctional screening tools, but also in the field ofbioinformatics to analyse the enormous mass of datathat is obtained with such approaches. Nevertheless,the first applications of meta-‘‘omics’’ approacheshave already demonstrated their power as discussedin this paper and we expect this field to growexplosively in the near future.
Only an integration of all reductionist and meta-‘‘omics’’ approaches in the near future will provideadequate understanding of GI tract microbiota, asthese approaches complement each other bydelivering different pieces of the GI tract puzzle.For example, the analysis of different cell cultureswill result in the discovery of novel genes andfunctions of individual organisms and therefore,serve as milestones for meta-‘‘omics’’ approaches.This integration is essential to explain data derivedfrom meta-‘‘omics’’ analysis, since the limitationsin our predictive capacity was demonstrated in thefirst metaproteomics study of the human GI tractmicrobiota.112
Overall, it will be a challenging future for GItract researchers. With the introduction of thenovel high-throughput technologies that aredescribed here, it will be possible, for the firsttime, to obtain statistically relevant links betweenmicrobial phylotypes and activities, and humanhealth. Ultimately, this will lead to the discoveryof biomarkers that will help us to understand andpredict the microbial life in our intestine, which isthe dream of a GI tract microbial ecologist.
Competing interests: None.
REFERENCES1. Bengmark S. Ecological control of the gastrointestinal tract. The
role of probiotic flora. Gut 1998;42:2–7.2. Cummings JH, Macfarlane GT. Colonic microflora: Nutrition and
health. Nutrition 1997;13:476.3. MacFarlane GT, Cummings JH, Alison C. Protein degradation by
human intestinal bacteria. J Gen Microbiol 1986;132:1647–56.4. Zoetendal EG, Vaughan EE, de Vos WM. A microbial world
within us. Mol Microbiol 2006;59:1639–50.5. Rajilic-Stojanovic M. Diversity of the human gastrointestinal
microbiota: novel perspectives from high throughput analyses.Laboratory of microbiology. [PhD thesis]. Wageningen, TheNetherlands, Wageningen University, 2007:1–214.
6. Aranki A, Syed SA, Kenney EB, et al. Isolation of anaerobicbacteria from human gingival and mouse cecum by means of asimplified glove box procedure. Appl Microbiol 1969;17:568–76.
7. Barcenilla A, Pryde SE, Martin JC, et al. Phylogeneticrelationships of butyrate-producing bacteria from the human gut.Appl Environ Microbiol 2000;66:1654–61.
8. Pryde SE, Duncan SH, Hold GL, et al. The microbiology ofbutyrate formation in the human colon. FEMS Microbiol Lett2002;217:133.
9. Hayashi H, Shibata K, Bakir MA, et al. Bacteroides coprophilussp. nov., isolated from human faeces. Int J Syst Evol Microbiol2007;57(Pt 6):1323–6.
10. Hayashi H, Shibata K, Sakamoto M, et al. Prevotella copri sp.nov. and Prevotella stercorea sp. nov., isolated from humanfaeces. Int J Syst Evol Microbiol 2007;57(Pt 5):941–6.
11. Bakir MA, Kitahara M, Sakamoto M, et al. Bacteroidesintestinalis sp. nov., isolated from human faeces. Int J Syst EvolMicrobiol 2006;56:151–4.
12. Sakamoto M, Kitahara M, Benno Y. Parabacteroides johnsoniisp. nov., isolated from human faeces. Int J Syst Evol Microbiol2007;57(Pt 2):293–6.
13. Derrien M, Vaughan EE, Plugge CM, et al. Akkermansiamuciniphila gen. nov., sp. nov., a human intestinal mucin-degrading bacterium. Int J Syst Evol Microbiol 2004;54:1469–76.
14. Zoetendal EG, Plugge CM, Akkermans ADL, et al. Victivallisvadensis gen. nov., sp. nov., a sugar-fermenting anaerobe fromhuman faeces. Int J Syst Evol Microbiol 2003;53:211–5.
15. Collado MC, Derrien M, Isolauri E, et al. Intestinal integrity andAkkermansia muciniphila, a mucin-degrading member of theintestinal microbiota present in infants, adults, and the elderly.Appl Environ Microbiol 2007;73:7767–70.
16. Derrien M, Collado MC, Ben-Amor K, et al. The mucin-degraderAkkermansia muciniphila is an abundant member of the humanintestinal tract. J Appl Environ Microbiol 2008;74:1646–8.
17. Zengler K, Toledo G, Rappe M, et al. Cultivating the uncultured.Proc Natl Acad Sci U S A 2005;99:15681–5.
18. Ingham CJ, Sprenkels A, Bomer J, et al. The micro-Petri dish, amillion-well growth chip for the culture and high-throughputscreening of microorganisms. Proc Natl Acad Sci U S A2007;104:18217–22.
19. Stams AJM. Metabolic interactions between anaerobic bacteriain methanogenic environments. Antonie Van Leeuwenhoek1994;66:271–94.
20. Woese CR. Bacterial evolution. Microbiol Rev 1987;51:221–71.21. Woese CR, Kandler O, Wheelis ML. Towards a natural system of
organisms: proposal for the domains Archaea, Bacteria, andEucarya. Proc Natl Acad Sci U S A 1990;87:4576–9.
22. Zoetendal EG, Akkermans AD, de Vos WM. Temperaturegradient gel electrophoresis analysis of 16S rRNA from humanfecal samples reveals stable and host-specific communities ofactive bacteria. Appl Environ Microbiol 1998;64:3854–9.
23. Suau A, Bonnet R, Sutren M, et al. Direct analysis of genesencoding 16S rRNA from complex communities reveals manynovel molecular species within the human gut. Appl EnvironMicrobiol 1999;65:4799–807.
24. Bonnet R, Suau A, Dore J, et al. Differences in rDNA libraries offaecal bacteria derived from 10- and 25-cycle PCRs. Int J SystEvol Microbiol 2002;52:757–63.
25. Favier CF, Vaughan EE, de Vos WM, et al. Molecular monitoringof succession of bacterial communities in human neonates. ApplEnviron Microbiol 2002;68:219–26.
26. Wang M, Ahrne S, Antonsson M, et al. T-RFLP combined withprincipal component analysis and 16S rRNA gene sequencing: aneffective strategy for comparison of fecal microbiota in infants ofdifferent ages. J Microbiol Meth 2004;59:53.
27. Wang X, Heazlewood SP, Krause DO, et al. Molecularcharacterization of the microbial species that colonize human ilealand colonic mucosa by using 16S rDNA sequence analysis. J ApplMicrobiol 2003;95:508–20.
28. Wang M, Ahrne S, Jeppsson B, et al. Comparison of bacterialdiversity along the human intestinal tract by direct cloning andsequencing of 16S rRNA genes. FEMS Microbiol Ecol2005;54:219.
29. Hayashi H, Sakamoto M, Benno Y. Phylogenetic analysis of thehuman gut microbiota using 16S rDNA clone libraties and strictlyanaerobic culture-based methods. Microbiol Immunol2002;46:535–48.
30. Hayashi H, Sakamoto M, Benno Y. Fecal microbial diversity in astrict vegetarian as determined by molcular analysis andcultivation. Microbiol Immunol 2002;46:819–31.
31. Hayashi H, Sakamoto M, Benno Y. Molecular analysis of fecalmicrobiota in elderly individuals using 16S rDNA library and T-RFLP. Microbiol Immunol 2003;47:557–70.
32. Hold GL, Pryde SE, Russell VJ, et al. Assessment of microbialdiversity in human colonic samples by 16S rDNA sequenceanalysis. FEMS Microbiol Ecol 2002;39:33–9.
33. Eckburg PB, Bik EM, Bernstein CN, et al. Diversity of the humanintestinal microbial flora. Science 2005;308:1635–8.
Recent advances in basic science
Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603 1613
on 12 November 2008 gut.bmj.comDownloaded from
34. Mangin I, Bonnet R, Seksik P, et al. Molecular inventory of faecalmicroflora in patients with Crohn’s disease. FEMS Microbiol Ecol2004;50:25–36.
35. Heilig HG, Zoetendal EG, Vaughan EE, et al. Molecular diversityof Lactobacillus spp. and other lactic acid bacteria in the humanintestine as determined by specific amplification of 16S ribosomalDNA. Appl Environ Microbiol 2002;68:114–23.
36. Satokari RM, Vaughan EE, Akkermans ADL, et al. Bifidobacterialdiversity in human feces detected by genus-specific PCR anddenaturing gradient gel electrophoresis. Appl Environ Microbiol2001;67:504–13.
37. Duncan SH, Louis P, Flint HJ. Lactate-utilizing bacteria, isolatedfrom human feces that produce butyrate as a major fermentationproduct. Appl Environ Microbiol 2004;70:5810–7.
38. Derrien M, Adawi D, Ahrne S, et al. The intestinal mucosa as ahabitat of the gut microbiota and a rational target for probioticfunctionality and safety. Microbiol Ecol Health Dis 2004;16:137–44.
39. Leitch EC, Walker AW, Duncan SH, et al. Selective colonizationof insoluble substrates by human faecal bacteria. EnvironMicrobiol 2007;9:667–9.
40. Klappenbach JA, Saxman PR, Cole JR, et al. rrndb: theribosomal RNA operon copy number database. Nucl Acids Res2001;29:181–4.
41. Sacchi CT, Alber D, Dull P, et al. High level of sequence diversity inthe 16S rRNA genes of Haemophilus influenzae isolates is useful formolecular subtyping. J Clin Microbiol 2005;43:3734–42.
42. Acinas SG, Marcelino LA, Klepac-Ceraj V, et al. Divergence andredundancy of 16S rRNA sequences in genomes with multiple rrnoperons. J Bacteriol 2004;186:2629–35.
43. Venter JC, Remington K, Heidelberg JF, et al. Environmental genomeshotgun sequencing of the Sargasso Sea. Science 2004;304:66–74.
44. Harmsen HJM, Raangs GC, He T, et al. Extensive set of 16SrRNA-based probes for detection of bacteria in human feces. ApplEnviron Microbiol 2002;68:2982–90.
45. Lay C, Rigottier-Gois L, Holmstrom K, et al. Colonic microbiotasignatures across five Northern European countries. Appl EnvironMicrobiol 2005;71:4153–5.
46. Marteau P, Pochart P, Dore J, et al. Comparative study ofbacterial groups within the human cecal and fecal microbiota.Appl Environ Microbiol 2001;67:4939–42.
47. Ahmed S, Macfarlane GT, Fite A, et al. Mucosa-associatedbacterial diversity in relation to human terminal ileum and colonicbiopsy samples. Appl Environ Microbiol 2007;73:7435–42.
48. Zoetendal EG, von Wright A, Vilpponen-Salmela T, et al. Mucosa-associated bacteria in the human gastrointestinal tract are uniformlydistributed along the colon and differ from the community recoveredfrom feces. Appl Environ Microbiol 2002;68:3401–7.
49. Lepage P, Seksik P, Sutren M, et al. Biodiversity of the mucosa-associated microbiota is stable along the distal digestive tract inhealthy individuals and patients with IBD. Inflamm Bowel Dis2005;11:473–80.
50. Ben-Amor K, Heilig HG, Smidt H, et al. Genetic diversity ofviable, injured, and dead fecal bacteria assessed by fluorescence-activated cell sorting and 16S rRNA gene analysis. Appl EnvironMicrobiol 2005;71:4679–89.
51. Rajilic-Stojanovic M, Smidt H, de Vos WM. Diversity of thehuman gastrointestinal tract microbiota revisited. EnvironMicrobiol 2007;9:2125–36.
52. Moore WEC, Holdeman LV. Human fecal flora: the normal floraof 20 Japanese-Hawaiians. Appl Microbiol 1974;27:961–79.
53. Acheson DW, Luccioli S. Microbial–gut interactions in healthand disease. Mucosal immune responses. Best Pract Res ClinGastroenterol 2004;18:387–404.
54. Seksik P, Rigottier-Gois L, Gramet G, et al. Alterations of thedominant faecal bacterial groups in patients with Crohn’s diseaseof the colon. Gut 2003;52:237–42.
55. Andoh A, Sakata S, Koizumi Y, et al. Terminal restrictionfragment length polymorphism analysis of the diversity of fecalmicrobiota in patients with ulcerative colitis. Inflamm Bowel Dis2007;13:955–62.
56. Maukonen J, Satokari R, Matto J, et al. Prevalence andtemporal stability of selected clostridial groups in irritable bowelsyndrome in relation to predominant faecal bacteria. J MedMicrobiol 2006;55:625–33.
57. Matto J, Maunuksela L, Kajander K, et al. Composition andtemporal stability of gastrointestinal microbiota in irritable bowelsyndrome – a longitudinal study in IBS and control subjects.FEMS Immunol Med Microbiol 2005;43:213–22.
58. Dore J. A functional core of the human intestinal microbiota. In:11th International Symposium on Microbial Ecology – ISME-11;The hidden powers – microbial communities in action. 20–25August 2006, Vienna, Austria. Further information from [email protected]
59. Drasar BS, Barrow PA. Intestinal microbiology. Washington, DC:American Society for Microbiology, 1985.
60. Kleessen B. Culture-based knowledge on biodiversity,development and stability of human gastrointestinal microflora.Microb Ecol Health Dis 2000;12(Suppl 2):53–63.
61. Wang RF, Beggs ML, Robertson LH, et al. Design and evaluationof oligonucleotide-microarray method for the detection of humanintestinal bacteria in fecal samples. FEMS Microbiol Lett2002;213:175–82.
62. Ley RE, Turnbaugh PJ, Klein S, et al. Microbial ecology: Human gutmicrobes associated with obesity. Nature 2006;444:1022–23.
63. Guschin DY, Mobarry BK, Proudnikov D, et al. Oligonucleotidemicrochips as genosensors for determinative andenvironmental studies in microbiology. Appl Environ Microbiol1997;63:2397–402.
64. Small J, Call DR, Brockman FJ, et al. Direct detection of 16SrRNA in soil extracts by using oligonucleotide microarrays. ApplEnviron Microbiol 2001;67:4708–16.
65. Loy A, Lehner A, Lee N, et al. Oligonucleotide microarray for 16SrRNA gene-based detection of all recognized lineages of sulfate-reducing prokaryotes in the environment. Appl Environ Microbiol2002;68:5064–81.
66. El Fantroussi S, Urakawa H, Bernhard AE, et al. Direct profilingof environmental microbial populations by thermal dissociationanalysis of native rRNAs hybridized to oligonucleotidemicroarrays. Appl Environ Microbiol 2003;69:2377–82.
67. Peplies J, Glockner FO, Amann R. Optimization strategies forDNA microarray-based detection of bacteria with 16S rRNA-targeting oligonucleotide probes. Appl Environ Microbiol2003;69:1397–407.
68. Call DR, Bakko MK, Krug MJ, et al. Identifying antimicrobialresistance genes with DNA microarrays. Antimicrob AgentsChemother 2003;47:3290–5.
69. Volokov D, Chizhikov V, Chumakov K, et al. Microarray analysisof erythromycin resistance determinants. J Appl Microbiol2003;95:787–98.
70. DeSantis TZ, Brodie EL, Moberg JP, et al. High-density universal16S rRNA microarray analysis reveals broader diversity thantypical clone library when sampling the environment. Microb Ecol2007;53:371–83.
71. Palmer C, Bik EM, Eisen MB, et al. Rapid quantitative profiling ofcomplex microbial populations. Nucl Acids Res 2006;34:e5.
72. Smoot LM, Smoot JC, Smidt H, et al. DNA microarrays assalivary diagnostic tools for characterizing the oral cavity’smicrobial community. Adv Dent Res 2005;18:6–11.
73. Palmer C, Bik EM, Digiulio DB, et al. Development of the humaninfant intestinal microbiota. PLoS Biol 2007;5:e177.
74. Ludwig W, Strunk O, Westram R, et al. ARB: a softwareenvironment for sequence data. Nucl Acids Res 2004;32:1363–71.
75. DeSantis TZ, Hugenholtz P, Larsen N, et al. Greengenes, achimera-checked 16S rRNA gene database and workbenchcompatible with ARB. Appl Environ Microbiol 2006;72:5069–72.
76. DeSantis TZ, Dubosarskiy I, Murray SR, et al. Comprehensivealigned sequence construction for automated design of effectiveprobes (CASCADE-P) using 16S rDNA Bioinformatics2003;19:1461–8.
77. Palmer C, Bik EM, Digiulio DB, et al. Development of the humaninfant intestinal microbiota. PLoS Biol 2007;5:e177.
78. Harmsen HJM, Wildeboer-Veloo ACM, Raangs GC, et al.Analysis of intestinal flora development in breast-fed and formula-fed infants by using molecular identification and detectionmethods. J Pediatr Gastroenterol Nutr 2000;30:61–7.
79. Benno Y, Sawada K, Mitsuoka T. The intestinal microflora ofinfants: composition of fecal flora in breast-fed and bottle-fedinfants. Microbiol Immunol 1984;28:975–86.
80. Gwee KA. Irritable bowel syndrome in developing countries – adisorder of civilization or colonization? Neurogastroent Motil2005;27:317–24.
81. Sood A, Midha V, Sood N, et al. Incidence and prevalence ofulcerative colitis in Punjab, North India. Gut 2003;52:1587–90.
82. Bengtson MB, Ronning T, Vatn MH, et al. Irritable bowel syndromein twins: genes and environment. Gut 2006;55:1754–9.
83. Halfvarson J, Bodin L, Tysk C, et al. Inflammatory bowel diseasein a Swedish twin cohort: a long-term follow-up of concordanceand clinical characteristics. Gastroenterology 2003;124:1767.
84. Sartor RB. Mechanisms of disease: pathogenesis of Crohn’sdisease and ulcerative colitis. Nat Clin Pract Gastroenterol Hepatol2006;3:390–407.
85. Duncan SH, Hold GL, Barcenilla A, et al. Roseburiaintestinalis sp. nov., a novel saccharolytic, butyrate-producingbacterium from human faeces. Int J Syst Evol Microbiol2002;52:1615–20.
Recent advances in basic science
1614 Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603
on 12 November 2008 gut.bmj.comDownloaded from
86. Macfarlane GT, Gibson GR, Macfarlane S. Short chain fatty acidand lactate production by human intestinal bacteria grown in batchand continuous culture. In: Binder HJ, Cummings JH, Soergel KH,eds. Short chain fatty acids. London: Kluwer Publishing, 1994.
87. Scheppach W, Weiler F. The butyrate story: old wine in newbottles? Curr Opin Clin Nutr Metab Care 2004;7:563–7.
88. Matsuki T, Watanabe K, Fujimoto J, et al. Development of 16SrRNA-gene-targeted group-specific primers for the detection andidentification of predominant bacteria in human feces. ApplEnviron Microbiol 2002;68:5445–51.
89. Matsuki T, Watanabe K, Fujimoto J, et al. Use of 16S rRNAgene-targeted group-specific primers for real-time PCR analysis ofpredominant bacteria in human feces. Appl Environ Microbiol2004;70:7220–8.
90. Rinttila T, Kassinen A, Malinen E, et al. Development of anextensive set of 16S rDNA-targeted primers for quantification ofpathogenic and indigenous bacteria in faecal samples by real-timePCR. J Appl Microbiol 2004;97:1166–77.
91. Hooper LV, Gordon JI. Commensal host–bacterial relationshipsin the gut. Science 2001;292:1115–8.
92. Hooper LV, Xu J, Falk PG, et al. A molecular sensor that allows agut commensal to control its nutrient foundation in a competitiveecosystem. Proc Natl Acad Sci U S A 1999;96:9833–8.
93. Derrien M. Mucin utilisation and host interactions of the novelintestinal microbe Akkermansia muciniphila. Laboratory ofMicrobiology. [PhD thesis]. Wageningen, The Netherlands,Wageningen University, 2007:1–174.
94. Sonnenburg JL, Chen CT, Gordon JI. Genomic and metabolicstudies of the impact of probiotics on a model gut symbiont andhost. PLoS Biol 2006;12:e413.
95. de Vries MC. Analyzing global gene expression of Lactobacillusplantarum in the human gastrointestinal tract. Laboratory ofMicrobiology. [PhD thesis]. Wageningen, The Netherlands,Wageningen University, 2006:1–160.
96. Handelsman J. Metagenomics: application of genomics touncultured microorganisms. Microbiol Mol Biol Rev 2004;68:669–85.
97. Gabor EM, de Vries EJ, Janssen DB. Construction,characterization, and use of small-insert gene banks of DNAisolated from soil and enrichment cultures for the recovery ofnovel amidases. Environ Microbiol 2004;6:948–58.
98. Walter J, Mangold M, Tannock GW. Construction, analysis, andbeta-glucanase screening of a bacterial artificial chromosomelibrary from the large-bowel microbiota of mice. Appl EnvironMicrobiol 2005;71:2347–54.
99. Ferrer M, Golyshina OV, Chernikova TN, et al. Novel hydrolasediversity retrieved from a metagenome library of bovine rumenmicroflora. Environ Microbiol 2005;7:1996–2010.
100. Diaz-Torres ML, Villedieu A, Hunt N, et al. Determining theantibiotic resistance potential of the indigenous oral microbiota ofhumans using a metagenomic approach. FEMS Microbiol Lett2006;258:257–62.
101. Gloux K, Leclerc M, Iliozer H, et al. Development of high-throughput phenotyping of metagenomic clones from the humangut microbiome for modulation of eukaryotic cell growth. ApplEnviron Microbiol 2007;73:3734–7.
102. Uchiyama T, Abe T, Ikemura T, et al. Substrate-induced gene-expression screening of environmental metagenome libraries forisolation of catabolic genes. Nat Biotechnol 2005;23:88–93.
103. Breitbart M, Hewson I, Felts B, et al. Metagenomic analyses ofan uncultured viral community from human feces. J Bacteriol2003;185:6220–3.
104. Manichanh C, Rigottier-Gois L, Bonnaud E, et al. Reduceddiversity of faecal microbiota in Crohn’s disease revealed by ametagenomic approach. Gut 2006;55:205–11.
105. Jones BV, Marchesi J. Transposon-aided capture (TRACA) ofplasmids resident in the human gut mobile metagenome. NatMethods 2007;4:55–61.
106. Gill SR, Pop M, Deboy RT, et al. Metagenomic analysis of thehuman distal gut microbiome. Science 2006;312:1355–9.
107. Kurokawa K, Itoh T, Kuwahara T, et al. Comparativemetagenomics revealed commonly enriched gene sets in humangut microbiomes. DNA Res 2007;14:169–81.
108. Margulies M, Egholm M, Altman WE, et al. Genome sequencingin microfabricated high-density picolitre reactors. Nature2005;437:376–80.
109. Raes J, Korbel JO, Lercher MJ, et al. Prediction of effectivegenome size in metagenomic samples. Genome Biol 2007;8:R10.
110. Foerstner KU, von Mering C, Bork P. Comparative analysis ofenvironmental sequences: potential and challenges. PhilTrans R Soc Lond, B Biol Sci 2006;361:519–23.
111. Booijink CC, Zoetendal EG, Kleerebezem M, et al. Microbialcommunities in the human small intestine: coupling diversity tometagenomics. Future Microbiol 2007;2:285–95.
112. Klaassens ES, de Vos WM, Vaughan EE. A metaproteomicsapproach to study the functionality of the microbiota in the humaninfant gastrointestinal tract. Appl Environ Microbiol2007;73:1388–92.
Recent advances in basic science
Gut 2008;57:1605–1615. doi:10.1136/gut.2007.133603 1615
on 12 November 2008 gut.bmj.comDownloaded from