5 - ECTOMYCORHIZE!

1 - ENDO"MYCORHIZE!

ARBUSCULAIRE!

2 - ENDOMYCORHIZE A!PELOTONS (ORCHIDEES)!

3 - ENDOMYCORHIZE A!PELOTONS (ERICACEES)!

Pelotons!

Vési-cule!

Hyphes externes!

Réseau de Hartig!

Man-!teau!

Cylindre central! Peloton!

lysé!

Spororocyste!

Hyphes !externes!

Endoderme!

4 – ECTEN-!DOMYCORHIZE!

Mycorhizes

Nodosités

Nombremoyende

nodo

sités

parp

lante

Fig. 2. Distinct commensalniches control T cell cytokine pro-duction in the gut and skin. (A)Taxonomic classifications at thephylum level for 16S ribosomalRNA gene sequence data clus-tered at 97% identity from skintissue and fecal pellet of controlmice and mice treated with oralantibiotic cocktail (ATB) for 4 weeks.Each column represents an indi-vidual mouse. (B) Assessment ofIFN-g production in live CD45+

TCRb+ cells and IL-17A produc-tion in live CD45+ cells from skinand intestine of mice treated withoral antibiotic cocktail or water(Ctrl) for 4 weeks. Graphs showmeans T SEM of four mice (**P <0.005, ***P < 0.0005; ns, not sig-nificant). Results are representa-tive of two or three experiments.(C and D) Flow cytometric anal-ysis of IL-17A production in liveCD45+ TCRb+ cells from the gutand skin of SPF mice, GF mice,and GF mice monoassociated withS. epidermidis (GF + S.epi) for 2to 3 weeks. Graphs show means TSEM of three to five mice (**P <0.005). Results are representativeof two experiments.

Fig. 3. Cutaneous commensals drive immunity and pro-mote pathology in L. major infection. (A) Histopathologicalcomparison of ear pinnae skin lesions from L. major–infected SPF and GF mice. Scale bars, 500 mm. (B) Assess-ment of lesion size in SPF and GF mice. Each data pointrepresents an individualmouse (***P< 0.0005). (C andD)Flow cytometric analysis of Leishmania antigen-specificIFN-g and TNF-a production by TCRb+ CD4+ dermal cellsfrom L. major–infected SPF and GF mice. Each datapoint represents an individual mouse (**P < 0.005,***P < 0.0005). Results are representative of three ex-periments. (E) Number of L. major parasites per 1000nucleated cells from dermal lesions of infected SPF andGF mice. Each data point represents an individual mouse(***P< 0.0005). (F) Assessment of lesion size in SPFmice,GFmice, and GFmice monoassociated with S. epidermidis(S.epi). Each data point represents an individual mouse(**P < 0.005). Results are representative of two experi-ments. (G and H) Representative images of L. major skinlesions and analysis of IFN-g production by live TCRb+

CD4+ cells from SPF mice, GF mice, and GF mice mono-associated with S. epidermidis. Each data point representsan individual mouse (***P < 0.0005). Results are repre-sentative of two experiments.

www.sciencemag.org SCIENCE VOL 337 31 AUGUST 2012 1117

REPORTS

ContrôleAxéniqueAxénique + Staphylococcusepidermidis

Tube dig. Peau

Lym

phoc

yte

Protection du Tabac contre Botrytis cinerea

Le rumen

Fig. 2. Distinct commensalniches control T cell cytokine pro-duction in the gut and skin. (A)Taxonomic classifications at thephylum level for 16S ribosomalRNA gene sequence data clus-tered at 97% identity from skintissue and fecal pellet of controlmice and mice treated with oralantibiotic cocktail (ATB) for 4 weeks.Each column represents an indi-vidual mouse. (B) Assessment ofIFN-g production in live CD45+

TCRb+ cells and IL-17A produc-tion in live CD45+ cells from skinand intestine of mice treated withoral antibiotic cocktail or water(Ctrl) for 4 weeks. Graphs showmeans T SEM of four mice (**P <0.005, ***P < 0.0005; ns, not sig-nificant). Results are representa-tive of two or three experiments.(C and D) Flow cytometric anal-ysis of IL-17A production in liveCD45+ TCRb+ cells from the gutand skin of SPF mice, GF mice,and GF mice monoassociated withS. epidermidis (GF + S.epi) for 2to 3 weeks. Graphs show means TSEM of three to five mice (**P <0.005). Results are representativeof two experiments.

Fig. 3. Cutaneous commensals drive immunity and pro-mote pathology in L. major infection. (A) Histopathologicalcomparison of ear pinnae skin lesions from L. major–infected SPF and GF mice. Scale bars, 500 mm. (B) Assess-ment of lesion size in SPF and GF mice. Each data pointrepresents an individualmouse (***P< 0.0005). (C andD)Flow cytometric analysis of Leishmania antigen-specificIFN-g and TNF-a production by TCRb+ CD4+ dermal cellsfrom L. major–infected SPF and GF mice. Each datapoint represents an individual mouse (**P < 0.005,***P < 0.0005). Results are representative of three ex-periments. (E) Number of L. major parasites per 1000nucleated cells from dermal lesions of infected SPF andGF mice. Each data point represents an individual mouse(***P< 0.0005). (F) Assessment of lesion size in SPFmice,GFmice, and GFmice monoassociated with S. epidermidis(S.epi). Each data point represents an individual mouse(**P < 0.005). Results are representative of two experi-ments. (G and H) Representative images of L. major skinlesions and analysis of IFN-g production by live TCRb+

CD4+ cells from SPF mice, GF mice, and GF mice mono-associated with S. epidermidis. Each data point representsan individual mouse (***P < 0.0005). Results are repre-sentative of two experiments.

www.sciencemag.org SCIENCE VOL 337 31 AUGUST 2012 1117

REPORTS

Contrôle Axénique

Fig. 2. Distinct commensalniches control T cell cytokine pro-duction in the gut and skin. (A)Taxonomic classifications at thephylum level for 16S ribosomalRNA gene sequence data clus-tered at 97% identity from skintissue and fecal pellet of controlmice and mice treated with oralantibiotic cocktail (ATB) for 4 weeks.Each column represents an indi-vidual mouse. (B) Assessment ofIFN-g production in live CD45+

TCRb+ cells and IL-17A produc-tion in live CD45+ cells from skinand intestine of mice treated withoral antibiotic cocktail or water(Ctrl) for 4 weeks. Graphs showmeans T SEM of four mice (**P <0.005, ***P < 0.0005; ns, not sig-nificant). Results are representa-tive of two or three experiments.(C and D) Flow cytometric anal-ysis of IL-17A production in liveCD45+ TCRb+ cells from the gutand skin of SPF mice, GF mice,and GF mice monoassociated withS. epidermidis (GF + S.epi) for 2to 3 weeks. Graphs show means TSEM of three to five mice (**P <0.005). Results are representativeof two experiments.

Fig. 3. Cutaneous commensals drive immunity and pro-mote pathology in L. major infection. (A) Histopathologicalcomparison of ear pinnae skin lesions from L. major–infected SPF and GF mice. Scale bars, 500 mm. (B) Assess-ment of lesion size in SPF and GF mice. Each data pointrepresents an individualmouse (***P< 0.0005). (C andD)Flow cytometric analysis of Leishmania antigen-specificIFN-g and TNF-a production by TCRb+ CD4+ dermal cellsfrom L. major–infected SPF and GF mice. Each datapoint represents an individual mouse (**P < 0.005,***P < 0.0005). Results are representative of three ex-periments. (E) Number of L. major parasites per 1000nucleated cells from dermal lesions of infected SPF andGF mice. Each data point represents an individual mouse(***P< 0.0005). (F) Assessment of lesion size in SPFmice,GFmice, and GFmice monoassociated with S. epidermidis(S.epi). Each data point represents an individual mouse(**P < 0.005). Results are representative of two experi-ments. (G and H) Representative images of L. major skinlesions and analysis of IFN-g production by live TCRb+

CD4+ cells from SPF mice, GF mice, and GF mice mono-associated with S. epidermidis. Each data point representsan individual mouse (***P < 0.0005). Results are repre-sentative of two experiments.

www.sciencemag.org SCIENCE VOL 337 31 AUGUST 2012 1117

REPORTS

Fig. 2. Distinct commensalniches control T cell cytokine pro-duction in the gut and skin. (A)Taxonomic classifications at thephylum level for 16S ribosomalRNA gene sequence data clus-tered at 97% identity from skintissue and fecal pellet of controlmice and mice treated with oralantibiotic cocktail (ATB) for 4 weeks.Each column represents an indi-vidual mouse. (B) Assessment ofIFN-g production in live CD45+

TCRb+ cells and IL-17A produc-tion in live CD45+ cells from skinand intestine of mice treated withoral antibiotic cocktail or water(Ctrl) for 4 weeks. Graphs showmeans T SEM of four mice (**P <0.005, ***P < 0.0005; ns, not sig-nificant). Results are representa-tive of two or three experiments.(C and D) Flow cytometric anal-ysis of IL-17A production in liveCD45+ TCRb+ cells from the gutand skin of SPF mice, GF mice,and GF mice monoassociated withS. epidermidis (GF + S.epi) for 2to 3 weeks. Graphs show means TSEM of three to five mice (**P <0.005). Results are representativeof two experiments.

Fig. 3. Cutaneous commensals drive immunity and pro-mote pathology in L. major infection. (A) Histopathologicalcomparison of ear pinnae skin lesions from L. major–infected SPF and GF mice. Scale bars, 500 mm. (B) Assess-ment of lesion size in SPF and GF mice. Each data pointrepresents an individualmouse (***P< 0.0005). (C andD)Flow cytometric analysis of Leishmania antigen-specificIFN-g and TNF-a production by TCRb+ CD4+ dermal cellsfrom L. major–infected SPF and GF mice. Each datapoint represents an individual mouse (**P < 0.005,***P < 0.0005). Results are representative of three ex-periments. (E) Number of L. major parasites per 1000nucleated cells from dermal lesions of infected SPF andGF mice. Each data point represents an individual mouse(***P< 0.0005). (F) Assessment of lesion size in SPFmice,GFmice, and GFmice monoassociated with S. epidermidis(S.epi). Each data point represents an individual mouse(**P < 0.005). Results are representative of two experi-ments. (G and H) Representative images of L. major skinlesions and analysis of IFN-g production by live TCRb+

CD4+ cells from SPF mice, GF mice, and GF mice mono-associated with S. epidermidis. Each data point representsan individual mouse (***P < 0.0005). Results are repre-sentative of two experiments.

www.sciencemag.org SCIENCE VOL 337 31 AUGUST 2012 1117

REPORTS

Fig. 2. Distinct commensalniches control T cell cytokine pro-duction in the gut and skin. (A)Taxonomic classifications at thephylum level for 16S ribosomalRNA gene sequence data clus-tered at 97% identity from skintissue and fecal pellet of controlmice and mice treated with oralantibiotic cocktail (ATB) for 4 weeks.Each column represents an indi-vidual mouse. (B) Assessment ofIFN-g production in live CD45+

TCRb+ cells and IL-17A produc-tion in live CD45+ cells from skinand intestine of mice treated withoral antibiotic cocktail or water(Ctrl) for 4 weeks. Graphs showmeans T SEM of four mice (**P <0.005, ***P < 0.0005; ns, not sig-nificant). Results are representa-tive of two or three experiments.(C and D) Flow cytometric anal-ysis of IL-17A production in liveCD45+ TCRb+ cells from the gutand skin of SPF mice, GF mice,and GF mice monoassociated withS. epidermidis (GF + S.epi) for 2to 3 weeks. Graphs show means TSEM of three to five mice (**P <0.005). Results are representativeof two experiments.

Fig. 3. Cutaneous commensals drive immunity and pro-mote pathology in L. major infection. (A) Histopathologicalcomparison of ear pinnae skin lesions from L. major–infected SPF and GF mice. Scale bars, 500 mm. (B) Assess-ment of lesion size in SPF and GF mice. Each data pointrepresents an individualmouse (***P< 0.0005). (C andD)Flow cytometric analysis of Leishmania antigen-specificIFN-g and TNF-a production by TCRb+ CD4+ dermal cellsfrom L. major–infected SPF and GF mice. Each datapoint represents an individual mouse (**P < 0.005,***P < 0.0005). Results are representative of three ex-periments. (E) Number of L. major parasites per 1000nucleated cells from dermal lesions of infected SPF andGF mice. Each data point represents an individual mouse(***P< 0.0005). (F) Assessment of lesion size in SPFmice,GFmice, and GFmice monoassociated with S. epidermidis(S.epi). Each data point represents an individual mouse(**P < 0.005). Results are representative of two experi-ments. (G and H) Representative images of L. major skinlesions and analysis of IFN-g production by live TCRb+

CD4+ cells from SPF mice, GF mice, and GF mice mono-associated with S. epidermidis. Each data point representsan individual mouse (***P < 0.0005). Results are repre-sentative of two experiments.

www.sciencemag.org SCIENCE VOL 337 31 AUGUST 2012 1117

REPORTS

Con-trôle

Axé-nique

Axé.+Staphylococcusepidermidis

Lésion cutanées de souris attaquées par Leishmania major

Normales(blanc)

Axéniques(noir)

Zoneclose

Zoneouverte

ExtrémitéCentreTemps(secon

des)

normales axéniques

normalesouaxéniques

normalesouaxéniques

Nervegrowth factor-inducible cloneA,liéàl’anxiété

Brain-derived neurotrophic factor,liéàlaplasticitésynaptique

Recyclagedesneurotransmetteurs

Sérotonine(5-HT)-->5-HIAA

p<0.002 p<0.04

p<2e-18Prifti,LeChatelier etal.unpublished

canalauditif cuirchevelu bouche

narine œsophage

peau intestin

pénis vagin

Article

Sialylate

dMilk

Olig

osacch

aridesPro

mote

Micro

biota-D

ependentGro

wth

inModels

ofInfan

tUndern

utritio

n

Grap

hicalAbstract

Highlig

htsd

Malaw

ianmothers

with

underno

urishedinfants

have

decreased

milk

olig

osaccharid

es

dSialylated

milk

sugars

promote

growth

ofanim

alsco

lonized

with

infantmicro

biota

dGrowth

promotio

ndoes

notoccur

with

provisio

nofinulin

or

ingerm

-freemice

dSialylated

olig

osaccharid

esim

pact

liver,muscle,and

brain

metab

olism

Autho

rsMark

R.C

harbonneau,D

avidO’Donnell,

LauraV.B

lanton,...,C

arlitoLeb

rilla,

David

A.M

ills,JeffreyI.G

ordon

Corresp

ond

encejgordon@

wustl.ed

u

InBrief

Twopreclinicalm

odels

forinfant

undernutritio

nestab

lishalink

betw

een

consum

ptio

nofsialylated

milk

olig

osaccharid

es,thegutm

icrobiota,and

healthygrowth.

Charb

onneau

etal.,2016,C

ell164,859–871Feb

ruary25,2016

ª2016

Elsevier

Inc.http

://dx.d

oi.o

rg/10.1016/j.cell.2016.01.024

Juvénileaxénique

AdultenormalFlored’enfantbiennourri

Flored’enfantmalnourri

Flored’enfantmalnourri

Floreaméliorantlacroissance

OU

HMO

Adultenain

0204060

1050

910

0815

0720

0625

0530

0435

0340

0245

0150

00

Seq

uenc

es/s

ampl

e

Bacterial diversity (PD)

PC1

(16%

)

PC2

(6%

)

PC3

(3%

)

Yano

mam

i

Gua

hibo

Mal

awi

US

Pre

vote

llaE

ubac

teriu

mP

arap

revo

tella

ceae

Ana

erop

lasm

aB

ulle

idia

Lact

obac

illus

Cop

roco

ccus

Met

hano

mas

silii

cocc

acea

eP

revo

tella

(uns

p.)

Oxa

loba

cter

acea

eA

erom

onad

acea

e (u

nsp.

)S

piro

chae

tes

Hel

icob

acte

rD

esul

fovi

brio

Verr

ucom

icro

bia

Pha

scol

arct

obac

teriu

mM

ollic

utes

RF3

9B

acte

roid

ales

S24

−7

01

23

4LD

A s

core

0.00

0.25

0.50

0.75

1.00

43

21

0Lo

g 10 a

bund

ance

43

2 21

0Lo

g 10 a

bund

ance

Prevalence

5101520

1015

930

845

760

675

590

410

5312

0213

5115

00S

eque

nces

/sam

ple

Bacterial diversity (PD)

PC1

(16%

)

PC2

(7%

)

PC3

(6%

)

Yano

mam

iU

S

Ste

notro

phom

onas

(uns

p)S

phin

goba

cter

ium

Act

inom

yces

Veill

onel

laP

seud

omon

asA

cine

toba

cter

Ste

notro

phom

onas

Rot

hia

Pre

vote

lla n

ance

iens

isPo

rphy

rom

onas

Hae

mop

hilu

sP

revo

tella

[Pre

vote

llace

ae]

Gra

nulic

atel

laFu

soba

cter

ium

Veill

onel

la (u

nsp)

Pre

vote

lla [P

arap

revo

tella

ceae

]G

emel

laLe

ptot

richi

acea

e

01

23

45

LDA

sco

re

0.00

0.25

0.50

0.75

1.00

Prevalence

02040

1015

930

845

760

675

590

410

5312

0213

5115

00S

eque

nces

/sam

ple

Bacterial diversity (PD)

PC1

(14%

)

PC2

(5%

)

PC3

(4%

)

Yano

mam

iU

S

Pro

pion

ibac

teriu

mN

eiss

eria

ceae

Cor

yneb

acte

rium

Sta

phyl

ococ

cus

Mic

roco

ccus

Pse

udon

ocar

diac

eae

Ace

toba

cter

acea

eN

eiss

eria

Sol

ibac

tera

ceae

Aci

doba

cter

iace

aeC

onex

ibac

tera

ceae

Met

hylo

cyst

acea

eM

icro

cocc

acea

e (u

nsp)

Bra

chyb

acte

rium

(uns

p)K

ocur

iaLe

ucon

osto

cace

aeK

noel

lia

01

23

4LD

A s

core

0.00

0.25

0.50

0.75

1.00

−4−3

−2−1

0Lo

g 10 a

bund

ance

Prevalence

Pop

ulat

ion

Gua

hibo

Mal

awi

US

A

Yano

mam

i

Pop

ulat

ion

US

Yan

omam

i

Popu

latio

nU

SYa

nom

ami

A

CD

E

FG

H

IJ

KL

B

Fig.

1.Microbiotadiversity

infecal,oral,and

skin

sam-

ples

from

uncontactedYa

nomam

iinrelatio

nto

othe

rhu

man

grou

ps.(A)Faith

’sph

ylog

enetic

diversity

(PD)

(ave

rage

±SD

)of

fecalsamples

from

Yano

mam

ian

dGuahibo

Amerindians,Malaw

ians,and

U.S.sub

jects.OTU

tables

rarefiedat

5000

sequ

encespersam

ple.Interpop

ula-

tiondiffe

rences

weresig

nificant(P

<0.001,

ANOVA

with

Tukey’s

HSD

)for

allb

utGuahibo

-Malaw

icom

parison

(P=

0.73).(B)P

CoAplot

basedon

UniFrac

distancescalculated

ontheOTU

tableof

fecalsam

ples

rarefiedat

5000

se-

quencespe

rsam

ple.(C)Top

discrim

inativebacteriaam

ong

popu

latio

nsin

fecalsam

ples

asde

term

ined

bylineardis-

crim

inan

tan

alysis(LDA)

effect

size

(LEfSe)an

alysis.(D)

Normalized

prevalen

ce/abu

ndan

cecurves

forallOTU

sfoun

dat1%

abun

danceor

moreinfecalsam

ples.(E)

Faith’s

phylog

eneticdiversity

(average

±SD

)ofo

ralsam

ples

from

Yano

mam

iand

U.S.sub

jects.OTU

tables

rarefiedat

1500

sequ

encespe

rsample.

Interpop

ulationdiffe

rences

were

notsig

nificant(P

=0.296,

ANOVA

with

Tukey’s

HSD

).(F)

PCoA

plot

basedon

UniFrac

distancescalculated

onOTU

tables

oforalsamples

rarefiedat

1500

sequ

encespe

rsam

-ple.(G)Top

discrim

inativebacteriaam

ongpo

pulatio

nsinoralsamples

asdeterm

ined

byLEfSeanalysis.

(H)N

ormalize

dprevalence/abu

ndance

curves

fora

llOTU

sfoun

dat

1%ab

undanceor

morein

oralsamples.(I)Faith

’sph

ylog

eneticdiversity

(average

±SD

)ofskinsamples

from

Yano

mam

iand

U.S.sub

jects.

OTU

tables

rarefiedat

1500

sequ

encespe

rsam

ple.Interpop

ulationdiffe

rences

weresig

nificant(P<0.001,AN

OVA

with

Tukey’s

HSD

).(J)P

CoAplot

based

onUniFrac

distancescalculated

onOTU

tables

ofskinsamples

rarefiedat

1500

sequ

encespe

rsam

ple.(K)Top

discrim

inativebacteriaam

ongpo

pulatio

nsin

skin

samples

asde

term

ined

byLEfSeanalysis.

(L)N

ormalized

prevalence/abu

ndance

curves

forallO

TUsfoun

dat

1%abun

danceor

morein

skin

samples.

RESEARCH

ARTIC

LE

Clem

ente

etal.S

ci.A

dv.2

015;1:e1

5001

8317

April20

152of

12

Cle

men

te e

t al.

Sci.

Adv

. 201

5;1:

e150

0183

JF B

ach,

non

pub

l.

Fréq

uenc

es c

umul

ées

des

indi

vidu

s

Grain d’aleurone

CristalloïdeGloboïde

Phytates(dans les globoïdes)

Glucosinolates des Brassicaceae

Distribution cumulée des fréquences de nombre de copies d’AMY1 dans différentes populations (en bas) : celles qui ont un régime moyen riche en amidon (high starch) ont plus de copies en moyenne que celles ayant un régime pauvre en amidon (low starch, chasseurs-cueilleurs). D’après Perry et al., 2007.

6,56.96,8

5,55.35,85.2

Moyennes