human intestinal microbiota in health and disease - a...

Human intestinal microbiota in health and disease- a microbiomics perspective -

Joël DoréINRA, Micalis, Jouy-en-Josas, France

Metagenome MetaprotomemRNADiversity, composition

Integrated microbiomics view:

Metabolome

Species diversity profiling indicates high inter-individual variations …

Zoetendahl et al. 1998 ; Sutren et al. 2000Seksik et al. 2003 ; Vanhoutte et al. 2004

14 healthy adults

…and stability over time

2 healthy adults

months

Only a fewbacterialspeciesappearconserved

Dominant microbiota

Sub-dominant microbiota

TransitingBacteria

1012

1011

1010

109

108

107

106

105

Bacteria/gfeces

L.rham

nosu

s

Lc lactis

L.plan

tarumL. c

aseiS.

therm

ophil

us

Phylogenetic groups and spp(qPCR assessment)

Enterococcus

E. coli

LactobacillusLeuconostoc

S. salivarius

BacteriaC. Leptumgroup

C. Coccoidesgroup

BacteroidesPrevotella

group

Bifidobacterium

J-P Furet et al., 2009

• Colonisation and immune imprinting - early times –

• Early colonisation and geographical location

w = weeks Mq = marqueur

Molecular signatures shared by different biological fractions Potential transfer from mother to newborn via maternal milk Streptococcus salivarius, Staphilococcus epidermidis, Bif. longum

P. Perez et al. 2007 Pediatrics

Maternal milk, source of active immune imprinting(n=7 mother-newborn couples)

Mq 2w 3w 4w 1w 2w 3w 4w1w 2w 3w 4w 1w 2w 3w 4w Mq Mq

Feces Blood cells Milk cells Feces

Mother NewbornBlood cellsBlood cells

Human milk cellHuman milk cell

Infant fecal microbiota differs from that of adults, until 2nd year of life.

Genus Bifidobacterium is most representedBacteroides and relatives are early colonizers following enterobacteria.Adult phylogenetic groups C. coccoides and C. leptum are not ‘yet’ dominant.

M. Fallani et al., JPGN 2010

0,42,4

0,8 1,83,5

42,2

5,6

11,1

74,0

6,1

0

20

40

60

80

100

Bifido

bacte

riaC.

cocc

oides

C.lep

tum

Atop

obium

Bacte

roide

sEn

tero

bact

eria

Lact

obac

illae

Stre

ptoc

occi

C.pe

rf+

C.dif

sp.

Tota

l%

n=606 ; 5 EU countries

Fecal microbiota of 6 weeks-old European infants FISH-FCM

Somme

Bifidobacterium

Nord-South gradient from Stockholm, Sweden to Granada, Spain : more bifids in the north ; early diversification in the south

(n=606)

Fecal microbiota of 6 weeks-old European infants

Spain

Sweden

Birth location has an incidence on composition of the fecal microbiota.

M. Fallani et al., JPGN 2010

Montecarlo testP = 0.024

age-related Bacteria

70 100Elena Biagi – Patrizia Brigidi –Claudio Francesci - University of Bologna

Lotta Nylund , Reetta Satokari & Willem M. de Vos University of Helsinki & Wageningen

the dominant fecal microbiota of elderly - specific alterations at the level of dominant species -

Under-represented:

Faecalibacterium sp.

Over-represented:

OxalobacterAkkermansia

Bifodobacterium

Metagenome MetaprotomemRNADiversity, composition

Core species Core genes

2% of non- 9% of non-redundant redundant

Core microbiome of the « normal » microbiota of healthy subjects ; a schematic view

Metagenome MetaproteomemRNADiversity, composition

inter-individual redundancy within theintestinal ecosystem at the level of

environmental proteome ?

Cy3 Cy5

Cy3 Cy5

Cy5Cy3

1092 proteins commonto both subjects

397 proteinsspecific to subject 1

376 proteinsspecific to subject 2

1865 validated spots~ 59 % comigrating spots *

for 2 metaproteomes

subject 1

subject 2

Juste et al, unpublished

Core-proteins

Metaproteome of human intestinal microbiotaPilot study in 6 volonteers / cytosolic proteins of fecal bacteria

* up to 15% may be host proteins

Metagenome MetaprotomemRNADiversity, composition

Core microbiome ; a schematic view

Core species Core genes Core proteins

66/>1000 1,000’s / 3M 100’s />2000

Defining “normal” microbiota - Eubiosis - : density, diversity, composition, core-species, dynamics (stability and resilience) of structure and functions (core-microbiome),

Hence, it becomes possible to explore Dysbiosis of the dominant intestinal microbiota in patients compared to healthy subjects.

This opens novel fields of exploration : – rationality for strategies to restore Eubiosis / homeostasis.

– mechanisms by which dysbiosis of the intestinal microbiota acts in chronic, immune, metabolic or degenerative pathologies ; especially in the context of bacteria-cell crosstalk…

Eubiosis versus Dysbiosis

Frailty in seniors Van Tongeren et al., 2005Crohn Seksik et al., 2003; Sokol et al., 2006, 2008, 2009; Qin et al., 2010Ulcerative colitis Sokol et al., 2008; Martinez et al., 2008 Hutson unpublishedPouchitis Lim et al., 2009, Kühbacher et al., 2006Obesity Ley et al., 2007; Kalliomäki et al., 2008; Furet et al. 2010Type-2 diabetis Cani and Delzenne, 2009 ; Sun et al., 2010Type-1 diabetis Dessein et al., 2009; Wen et al., 2008Coeliac disease Nadal et al., 2007; Collado et al., 2009Allergy Kirjavainen et al., 2002; Björkstén, 2009Autism Finegold et al., 2002; Parracho et al., 2005

Disease suspicion for an involvement of the commensal microbiota(Animal studies, antibiotics, probiotics or clinical studies)

Indicators of dysbiosis in disease versus health

Bach JF, N Eng J Med 2002

Importance of microbiota inimmune diseases increasing in prevalence ?

INCIDENCE OF CROHN’S DISEASE

0

5

10

15

1920 1940 1960 1980 2000

US & CanadaSwedenUKSouth AfricaFranceItalySpainJapan

Cases per 100.000 persons-year

Data from Ekbom et al, in McDermott & Stenson ‘IBD’

PEDIATRIC CROHN’S DISEASE IN SCOTLAND

0

5

10

1920 1940 1960 1980 2000

UK

Cases per 100.000 persons-year

Sawcenko et al, Lancet 2001

In children <16 years

EPIDEMIOLOGY OF IBD:GENETIC SUSCEPTIBILITY

• First-degree relatives: RR 10.7 for IBD

• Concordance rates for IBD in twin pairsfor CD monozygotic twins: 33-50%dizygotic twins: 0-10%

for UCmonozygotic twins: 13-19%dizygotic twins: 0-5%

Orholm et al, N Eng J Med 1991; Thompson et al, BMJ 1996; Orholm et al, Scand J Gastroenterol 2000; Halfvarson et al, Gastroenterology 2003

Inflammatory Bowel Diseases- Crohn’s disease- Ulcerative colitis

Dysbiosis of the Gut Microbiota ?

Deregulation of the Immune Response

Genetic Predisposition

MUCOSA ASSOCIATED MICROBIOTA IN IBD

Ott et al,

GUT 2005

Number of spp.Diversity index

Num

bero

fspp

.

Div

ersi

tyin

dex

Quantitative distortion in the density of mucosal microbiota(Swidsinski et al., Gastroenterology 2002)

Higher bacterial concentrations in mucosa associated microbiota of CD patients

Mucosa associated microbiota in IBD

Reduced Microbial Gene Diversity in IBD

Bacteroidetes371 clones33 OTUs

Firmicutes89 clones43 OTUs

20 (6)

79 Bifidobacterium longum (7)78 Bifidobacterium sp. (6)

81 Corynebacterium lipophiloflavum (3)80 Brachybacterium paraconglomeratum (1)

84(15)82 Coriobacterium sp. (17)

83(1)19 HuCC28 (14)

21 Adhufec61 (51) 22 Adhufec41 (9)

25 Bacteroides splanchnicus (2) 16 Adhufec367 (22)

18 Bacteroides vulgatus (77) 17 HuCB3 (5)

15 Adhufec151 (10)12 (33)

13 (10)14 (4)7 HuCA21 (1)

8 Bacteroides caccae (1)4 (1)

3 (1)1 Bacteroides thetaiotaomicron (4)

6 HuCC30 (3) 5 Bacteroides ovatus (3)

10 Bacteroides stercoris (16) 9 Bacteroides uniformis (56) 11 Bacteroides eggerthi (4)

24 Bacteroides merdae (3) 23 Baceroides distasonis (17) 30 (1)

34 (2) 33 (1)

31(1)32(1)

26 (1) 27 (2)

29 Bacteroides putredinis (4) 28 HuCB23 (5)

89 Bilophila wadsworthia (4)88 Escherichia coli K12 (16)

87 (1)86 (3)85(1)

71 Cadhufec079h7 (1)72 Dialister invisus (1)

70 Eubacterium biforme (1)44 (1)

45 (2)46 Cadhufec14h7 (11)

48 (2)47 Ruminococcus sp. (1)

49 (1)51 ckcm323-B (1)50 (1)

39 p-5459-2Wb (3)40 HuCB5 (8)

41 p-2559-9F5 (2)42 butyrate producing M12/2 (5) 43 Adhufec365 (2)

35 Adhufec311 (1)36 p-2053-s95 (1)37 Adhufec269 (2)

38 p-953-s962 (1)75 (1)

77 (1)76 (1)

73 HuCA6 (1)74 (1)

69 Clone p-2722 (1)54 Adhufec406 (1)

52 (1)53 Butyrovibrio crossotus (4)

62 C. catus (1)58 Uncultured bacterium A20 (5)60 Uncultured bacterium G36 (1)59 HuCB12 (2)

57 HuCA22 (1)55 p-3075-SwA (1)

56 (2)61 Cadhufec001h7 (2)66 p-2431-55G (1)68 HuCB10 (2)

67 (1)64 butyrate producing M104/1 (8)

65 butyrate producing M72/1 (1)63 Lachnospira pectinoschiza(1)

68

52

58

68 98

92

52

100

56

72

45 clones17 OTUs

35 clones17 OTUs

Proteobacteria25 clones5 OTUs

Actinobacteria50 clones7 OTUs

Clostridiumleptumgroup

Clostridiumcoccoides

group

Metagenomicclone name(Adhufec-x)

clone number

Healthy library

Bacteroidetes556 clones33 OTUs

Firmicutes37 clones13 OTUs

14 (3)15 (1)

51 Coriobacterium sp. (1) 56 Desulfovibrio piger (1)

49 Bifidobacterium pseudolongum (28) 50 (1)

40(5)39 (4)

41 (3)42 (2)43 (1)47 (3)

38 (1)36 (4)

34 HuCB5 (3) 35 butyrate producing M21/2 (2) 37 (1)

45 (1)44 (7)

52 Escherichia coli K12 (27) 55 (1)

54 (1)53 (1)

29 (1)23 Bacteroides putredinis (38) 31 (3)

30 (1)22 (1) 28 (1)

24 (6)25 (57)26 (30)

27 (1)21 Bacteroides splanchnicus (9)

19 Bacteroides merdae (34) 18 Bacteroides distasonis (10)

20 (1)33 (37)

32 (12)16 (2)11 HuCC21 (2)

17 (7)13 (1)

12 Adhufec61 (68) 9 (5)6 Bacteroides vulgatus (91)5 Adhufec367 (18)

7 (1)8 (3)

4 Bacteroides stercoris (15) 10 (7)

1 Bacteroides ovatus (2) 2 HuCC30 (17) 3 Bacteroides uniformis (71)

5194

30

87

100

35

14 clones5 OTUs

15 clones6 OTUs

31 clones5 OTUs

Proteobacteria

30 clones7 OTUs

Actinobacteria

Clostridium coccoidesgroup

Clostridium leptumgroup

Metagenomicclone name(Cadhufec-x)

clone number

Crohn library

Manichanh et al, Gut 2006

Fecal microbiota dysbiosis in Crohn’s disease

Lesser proportion of bacteria from the Clostridium leptum group in CD

In Feces In Mucosa

Fecal and mucosal microbiota dysbiosis in Crohn’s disease

Harry SokolManichanh et al., Gut 2006, Sokol et al., IBD 2006

FISH analysis of biopsies

M0surgical resection

M6colonoscopy

• Eub338 (Eubactia)• Bac303 (Bacteroides-Prevotella)• Ent1458 (Enterobacteria)• Erec482 (Clostridium coccoides)• Lab158 (Lactobacillus-Enterococcus)• Bif164 (Bifidobacterium)• Fprau645 (F. prausnitzii and relatives)

Still in remission

or

Endoscopic relapse

20 patients with active CD, requiring ileo-caecal resection :

Mucosal Dysbiosis in Crohn’s DiseasePh. Marteau & Ph. Pochart

Harry Sokol, Philippe Langella et al. PNAS 2008

FISH analysis of biopsies

M0surgical resection

M6colonoscopy

• Eub338 (Eubactia)• Bac303 (Bacteroides-Prevotella)• Ent1458 (Enterobacteria)• Erec482 (Clostridium coccoides)• Lab158 (Lactobacillus-Enterococcus)• Bif164 (Bifidobacterium)• Fprau645 (F. prausnitzii and relatives)

Still in remission

or

Endoscopic relapse

F. prausnitzii at M0 (p=0.027)3.3% Remission at M60.3% Relapse at M6

Faecalibacterium prausnitzii is associated with protection from endoscopicinflammation relapse 6 months after surgery => anti-inflammatory properties

20 patients with active CD, requiring ileo-caecal resection :

Mucosal Dysbiosis in Crohn’s DiseasePh. Marteau & Ph. Pochart

Harry Sokol, Philippe Langella et al. PNAS 2008

A commensal bacterium with anti-inflammatoryproperties in vitro and in vivo

Sokol, et al. PNAS 2008Willing et al. IBD 2009 [dysbiosis : F prausnitzii - ; E coli + in ICD ]

Faecalibacterium prausnitzii ;

Anti-inflammatory properties of F. prausnitzii

Anti-inflammatory properties of F. prausnitzii A2-1651. In vitro experiments:• High IL10/IL12 cytokine release by Peripheral blood mononucleated cells

• Reduction of IL-1β induced IL-8 secretion by Caco-2 cells

• Supernatant abolished TNFalpha induced NF-kB activity in HT-29 cells

2. In vivo experiments:• Both F. prausnitzii and its supernatant reduced scores and blood parameters of

inflammation in TNBS-induced colitis in Balb/c mice • Administered IP, its supernatant protected mice from death induced by TNBS.

Sokol et al. PNAS 2008

TNBS induced Colitis :intraperitoneal administration of F. prausnitzii fractions

F prausnitziiSupernatant

DexamethasoneTNBS + PBS

0%

20%

40%

60%

80%

100%

D5 D10 D15 D20D2

Survival rate

Faecalibacterium prausnitzii supernatant contains “bioactives” that confer protection against a lethal condition caused by TNBS induced gut inflammation

Seungha Kang, Chris McSweeney, Mark Morrisson

qPCR validation of fecal microbiota dysbiosis (phylogenetic µarray) in Crohn vs healthy subjects

-30

-25

-20

-15

-10

-5

0

5

10R.albus

F.prausnitzii

specific PCR Systems

Fact

or(h

ealth

y/cr

ohn)

C.difficile

E.faecalis

Profiling-based assessment of fecal dysbiosis (PCR-DGGE) in Crohn patients vs healthy subjects

Several candidate marker-species differed in prevalence and relative intensities (P<0.05) and F. prausnitzii, B. adolescentis and R. gnavus were validated by qPCR

M. Joossens et al. submitted

RT-qPCR validation of fecal microbiota dysbiosis (metagenome) in Crohn patients vs healthy subjects

Crohn patients

Healthy controls

O. valericigenes is detected in 12% of Crohn patients vs 93% of controls, and with a higher abundance in controls than in CD patients (ΔCt = 17.33, P<0.001)

Subdoligranulum sp. R. bromii B. vulgatusP. distasonis O. valericigenes E.coli

S. Mondot et al. IBD 2010

S. Mondot et al. IBD 2010

Faecalibactrerium prausnitzii

Fecal microbiota dysbiosis (µarray) in Crohn patients vs healthy subjects (n=14+16)

d = 5

Scores and c lasses

N :N

Y :C D

Y :U C

Ax is3

d = 2

Y :U C

p-value: 0.031

Fecal Dysbiosis in Crohn’s Diseaseat the level of metagenome

Crohn Patients

UC Patients

HealthyControls

Francisco Guarner (Val d’Hebron Hospital, Barcelona)Wang Jun and colleagues (BGI-Shenzen) Junjie Qin et al Nature, March 2010Dusko Ehrlich, Patricia Lepage, Julien Tap (INRA)

• A cohort of 21 Spanish individuals• patients: n = 8• healthy: n = 13•Ranksum search for genes different in frequency(hits/kb/30 million reads) between the two groups in a 3.3 million gene catalog

3802 “CD related genes” found at p< 3*10-4

Ehrlich et al.

Fecal Dysbiosis in Crohn’s Diseaseat the level of metagenome : CD related genes can be identified.

Phenotype CD genes (median)

Healthy 3038   out of  562 271CD 643 out of 425 397 

“Gene loss” in CDp<2*10-13

(one‐tailed t test)

HealthyPatients

Ehrlich et al. => diagnostic model with a limited set ofmarker-species (5 Firmicutes candidates)

Fecal Dysbiosis in Crohn’s Diseaseat the level of metagenome : CD related genes can be identified.

12 participants (age & sex matched)6 Crohn’s Disease patients (CD) vs 6 healthy controls (HC)

Catherine Juste & Saint-Antoine Hospital INRA platforms PICT & PAPSSO

-2.0 -1.5 -1.0 -0.5 0. 0 0.5 1.0 1.5 2.0

Log2 Fold Change

threshold

39 under-expressed in CD 41 over-expressed in CD

B s

tatis

tic

unchanged

Differential protein expression from 2D-DIGE experiments

Protein spot identification using LC-MS/MS, in progress:- 66 spots identified out of 84 already analysed (~78% yield)- over-expressed proteins associated with Bacteroides (several known as “immunoactive”, and/or with eucaryotic homologs)- under-expressed proteins mainly associated with Firmicutes

2007 spots

Participants (duplicates)

HCCD

Prot

eins(

80 d

iffer

entia

llyex

pres

sed)

Metaproteomic signatures in Crohn

PLS-DA Scores plot, R2Y=89.7%, Q2Y=61.1%

Control

Crohn’s

PCA Scores plot (Paretto Scaling)

Highlights several discriminating metabolites (under-represented in Crohn)

Metabolomic signatures in Crohn12 volonteers : 6 healthy vs 6 Crohn

Jansson et al (PLoS ONE July 2009) : metabolites discriminating ICD from CCD and health

Tyrosine metabolism, amino-acids, bile acid metabolism, …

Bile acid metabolism in ICD > CCD > Healthy (P value concn. ICD vs others)Glycocholate 5/6 4/8 0/10 .001Taurocholate 5/6 2/8 2/10 .002Glycochenodeoxycholate 5/6 2/8 1/10 .002

Dysbiosis in Crohn’s disease => vicious circle favoring aggravation and chronicity• increased bacterial density at mucosal level (Swidzinski 2002)• increased proportion of immuno-agressive commensals

(Gram-negative, subdominant in health?) (Darfeuille Michaud 2004, Sokol 2009, Willing 2009)• reduced proportion of anti-inflammatory commensals

(Gram-positive Firmicutes?, Actinobacteria?) (Sokol 2009, Willing 2009)• increased proportion of proteins potentially promoting auto-immunity (Juste submitted)

Further contributions of Dysbiosis : disruption of tolerance, alteration of mucosal healing, secretion of defensins, … ?

Dysbiosis of the Gut Microbiota

Deregulation of the Immune Response

Genetic Predisposition

Complex interactions underlying polygenic obesity

NutritionExercise Viruses

AllergiesSocial Status

Food Abundance

Peer pressure

PollutionTechnologicalProgress

Psychology

NutritionExercise Viruses

AllergiesSocial Status

Food Abundance

Peer pressure

PollutionTechnologicalProgress

PsychologyPsychology

Mutch D & Clement K, Plos genet 2006

GUTMicrobiota

Endo

crin

egl

and

Transcriptional interaction networks in human adipose tissue

Overexpressed in obese AT Underexpressed in obese AT

Centrality (%) Henegar, Genome biology, 2008Prifti, Bioinformatics, 2008

Altered metabolism

Increased inflammation

Bypass surgery Switch of functions / weight loss (3mo)

Overexpressed themes Underexpressed themes

Transcriptional domain cover

Improvedinflammatorycontext

ImprovedMetabolismTissue remodeling

Metabolism & inflammation improvement after GBPMetabolism & inflammation improvement after GBP

Wilcoxon pairé; * p<0.05 vs M0 ; *entre M3 et M6

Before GBP (M0)3m after GBP (M3)6mois after GBP (M6)

Calorie intake (kcal)

** *

Glycemia (mmol/l)

**

hsCRP (mg/dl)* *

Weight (kg)

** *

-22%-50%

0

10

20

30

40

50

60

1 2 3

Leptin (ng/ ml)

0 3m 6m

-50%

* *

0 3m 6m 0 3m 6m

Bacteria groups increasing with weight lossBacteria groups increasing with weight loss

Variation associates (- neg) with corpulence (BMI, FM, Leptin)

But dependant on calorie ( LME).

Wilcoxon, * p<0.05 vs C; Wilcoxon pairé, * vs M0

*

* *

Amount of bacteria = log10 (bacteria) - log10 (total bacteria)

Bacteroides/PrevotellaBacteroides/Prevotella

* *

Associations (-ve) with BMI, FM, Leptin (R -0.53 p < 0.001)

Remained if calorie adjusted

E. coliE. coliPhyllum: ProtéobactériesProtéobactéries

Furet JP, Kong LC et al Diabetes 2010

Positive correlations (+pos) with BMI, % fat mass, Leptin, Insulin (R 0.30, P<0.05) / depend on calorie intake (ajustments)

Wilcoxon paired, * vs M0

* *

BifidobacteriumPhyllum: Actinobactéries

* *

LactobacillusPhyllum: Firmicutes

Bacteria groups decreasing with weight lossBacteria groups decreasing with weight loss

Furet JP, Kong LC et al Diabetes 2010

F.prausnitzii is lower in diabetics Negative correlations with inflammatory markers : basal & kinetics

hsCRP (R -0.39), Interleukin 6 (R -0.35), p< 0.0001

Wilcoxon; * p<0.05 D vs C/ND

Bacteria groups associate with low-grade inflammation F.prausnitzii

C.leptumC.leptum;; Phyllum: Firmicutes

C : control leanD: diabeticsND: non diabetiics*

Furet JP, Kong LC et al Diabetes 2010

ConclConcl (1) Gut microbiota differentially adapts after GBP(1) Gut microbiota differentially adapts after GBP

**differential adaptationdifferential adaptation

Bacteroides(-neg)E.coli(-neg)

Metabolism

BMI, fat mass, Leptin,Insulin, etc

Bifido, Lacto(+pos)

Inflammation

CRPus, IL-6, etc

F. prausnitzii(-neg)

**links (links (BactBact--MetaMeta--InflInfl))

**Food intakeFood intakeModel effectSequencing

? ? ? ? ?

Microbiota-host interactions in obesity & related complications

Cani PD and Delzenne NM. Curr Opin Clin Nutr Metab Care 2007; 10:729–734

Obesity state

Gut Microbiota

?

Conserved traits at various levels of integration :Core microbiome

Criteria to qualify normobiosis/eubiosis

Microbiome specificities in diseases, towards : identification of predictive biomarkers, new targets and strategies for nutritional and/or therapeutic applications

in intestinal disorders