Science Magazine


1 APRIL 2011    VOL 332    SCIENCE    www.sciencemag.org 32

Five years ago, a team headed by Jeffrey 

Gordon of Washington University in St. 

Louis (WUSTL) in Missouri made a surpris-

ing discovery: The guts of obese mice and 

people harbor an array of microbes different 

from that of their lean counterparts. More 

provocatively, when they gave lean mice 

certain gut-dwelling microbes, the rodents 

became fat (Science, 29 May 2009, p. 1136). 

The fi ndings sparked headlines and fueled 

popular speculation that manipulating gut 

bacteria might keep weight down in people.

Already, Martin Blaser had been head-

ing down a similar track. Blaser, a microbi-

ologist at New York University in New York 

City, was struck by how successful farmers 

are at increasing the growth rates of livestock 

by adding low doses of antibiotics to their 

feed. “The earlier in life they start the antibi-

otic, the more profound the effect,” he points 

out. He began to wonder whether antibiotic 

use, particularly in children, might affect the 

long-term establishment of a balanced micro-

bial community in the human gut, eliminat-

ing bacteria there that could help ward off 

obesity. He started conducting mouse studies 

to examine the hypothesis. 

Since then, several other groups have 

joined in. A raft of intriguing obesity-

related fi ndings was presented at a meeting 

last month on the microbiome, the bacteria 

that live inside the guts and other tissues of 

animals. Yet many in the fi eld caution that 

it remains diff icult to determine whether 

changes in gut microbes drive or contribute 

to obesity or whether the excess weight itself 

triggers those changes. “The jury is still out 

[about] what the role of the gut microbiota 

may be in obesity in humans,” says Claire 

Fraser-Liggett, a microbiologist at the Uni-

versity of Maryland School of Medicine in 

Baltimore who has studied gut bacteria and 

obesity in the Amish. 

Case of the missing microbes
The farm animal–antibiotic connection 

was one clue that led Blaser to wonder 

about microbial causes of the obesity epi-

demic. Another was the fact that very few 

people now harbor the ulcer-causing bacte-

rium Helicobacter pylori in their stomachs. 

H. pylori, which has also been linked to 

stomach cancers, is one of up to 1000 differ-

ent microbes that call the human body home. 

Once ubiquitous in the human microbiome 

and still so in the guts of people from devel-

oping countries, it is now found in just 6% 

of U.S. children. That might seem like good 

news, as there should be fewer ulcers and 

cancers. But Blaser suspects that it is also 

bad news, as studies suggest that H. pylori’s 

presence in the gut helps regulate the stom-

ach’s production of the hormone ghrelin, 

which stimulates food intake. 

This bacterium may 

not be the only species 

disappearing from our 

microbiome. After a 

person takes antibiotics, 

“it has always been pre-

sumed that the micro-

biota will spring back,” 

Blaser says. But the fate 

of H. pylori suggests 

otherwise. Its vanishing 

act and other shifts in the 

microbiome may con-

tribute to an increased 

risk for weight gain, 

Blaser worries.

He has started to investigate 

this theory by giving mice either 

low doses of antibiotics over long 

periods, akin to what farm ani-

mals receive, or short-term, high 

doses, more like what a sick infant 

or adult would get. He then com-

pares the physiology and microbi-

omes of these treated rodents with 

those of mice raised under similar 

conditions but given no antibiot-

ics. In one set of studies, the mice 

fed low doses of antibiotics long-

term wound up with 15% more 

body fat than the control mice, 

Blaser reported last month at the 

International Human Microbiome 

Congress in Vancouver, Canada. 

The chubbier, antibiotic-fed mice 

also had about 25% more fat in 

their livers.  

The treated mice also had a different set 

of bacterial species inhabiting their guts. 

And several hundred bacterial genes, includ-

ing ones for fatty acid production, exhibited 

different levels of activity—some increas-

ing, others decreasing—in these mice com-

pared with the controls. Similar changes 

occur in the rodents given short pulses of 

antibiotics, he noted. 

Antibiotics “may be driving the gut 

microbiome to a place where it shouldn’t 

be,” Fraser-Liggett says. “We do not know 

the functional consequences, but with these 

miracle drugs now 60 years later, we may be 

seeing effects that change susceptibility to 

various diseases.” 

Blaser will examine the gut microbiomes 

of children to see whether his results are 

applicable to humans. If so, “that would be a 

remarkable connection that could have a sig-

nifi cant impact on medical care,” says genome 

scientist George Weinstock of WUSTL. But 

he’s cautious: “In a lot of cases, the micro-

biome in mice doesn’t translate into humans.”

Patterns in genes
S. Dusko Ehrlich has 

avoided that issue, 

bypassing mice and 

instead directly exam-

ining whether patterns 

in the microbiome of 

people relate to body 

mass index and obesity. 

A microbiologist at the 

INRA Microbiology 

and Food Chain Divi-

sion in Jouy-en-Josas, 

France, Ehrlich is part 

of a group, the Meta-

NEWSFOCUS

C
R

E
D

IT
S

 (
T

O
P

 T
O

 B
O

T
T

O
M

):
 R

A
D

E
 L

U
K

O
V

IC
/T

H
IN

K
S

T
O

C
K

; 
J
A

S
O

N
 H

E

Go with the gut. Antibiotic use may 
adversely affect the long-term makeup of 
the intestine’s bacterial communities.

Girth and the Gut (Bacteria)
Mouse, human studies begin to clarify gut bacteria’s role in obesityMouse, human studies begin to clarify gut bacteria’s role in obesity

M I C R O B I O L O G Y

Published by AAAS

 o
n
 J

a
n
u
a
ry

 1
6
, 
2
0
1
2

w
w

w
.s

ci
e
n
ce

m
a
g
.o

rg
D

o
w

n
lo

a
d
e
d
 f
ro

m
 

http://www.sciencemag.org/


www.sciencemag.org    SCIENCE    VOL 332    1 APRIL 2011 33

NEWSFOCUS
C

R
E

D
IT

: 
L
A

U
R

A
 K

Y
R

O
/
C

E
N

T
E

R
 F

O
R

 G
E

N
O

M
E

 S
C

IE
N

C
E

S
 &

 S
Y

S
T

E
M

S
 B

IO
L
O

G
Y

HIT consortium, investigating connections 

between microbial genes in human intestines 

and human health. By comparing such genes 

from obese and nonobese individuals, he and 

his colleagues have found that certain sets of 

bacterial genes and bacteria correlate with 

excess weight and insulin resistance.

The researchers fi rst sequenced all the bac-

terial genes in stool samples of 177 Danes, 55 

who were thin and 122 who were either over-

weight or obese. Although the researchers 

concluded that most participants in the study 

had roughly 600,000 distinct bacterial genes 

in their guts, almost one-third of the obese 

study participants had only about 360,000 

such genes, 30% to 40% fewer. A similar per-

centage of 36 obese French people had a com-

parable dearth of gut bacteria genes, Ehrlich 

reported at the Vancouver meeting. Moreover, 

the obese people “don’t have as great a bac-

terial diversity” in their 

guts, Ehrlich reported. One 

missing microbe in that 

group was a methane pro-

ducer, leading Ehrlich to 

wonder whether “the car-

bon that does not get out 

[of the body] as gas could 

be incorporated as fat.”   

When they looked 

at medical histories of 

all their study subjects, 

Ehrlich and his colleagues 

found that the obese peo-

ple with fewer gut bacte-

ria genes were more likely 

to be insulin resistant than 

were the obese people 

who had a typical tally of 

intestinal microbial genes. 

These obese people also 

tended to have higher than 

normal white blood cell counts, suggesting 

that they were in a state of low-level infl am-

mation, Ehrlich said. Some researchers have 

found evidence of a link between inflam-

mation and obesity (Science, 17 December 

2010, p. 1621).

Ehrlich and his colleagues have also tested 

whether the types of bacteria in a person’s gut 

can “diagnose” obesity. Using just six meta-

species, they were able to correctly predict 

whether a person was lean or obese more than 

80% of the time, he reported. When research-

ers try to make the same predictions by con-

sidering all of a person’s genetic risk factors 

for obesity, they are right only 58% of the 

time, Ehrlich pointed out. 

At this point, however, it’s unclear whether 

the differences in intestinal microbes are “the 

cause, a contribution to, or the consequence” 

of obesity, notes Ehrlich. “If we can provide 

evidence that they [at least] provide a contri-

bution, then we can go and fi nd a treatment.”

Help from the Amish
Other work presented at the microbiome 

meeting indicates that sorting out this cause-

and-effect puzzle will be tough. Frustrated by 

the inconsistent results others were getting 

when they looked for connections between the 

microbiome and obesity, Fraser-Liggett and 

her colleagues examined 400 adult Amish liv-

ing in Pennsylvania. Amish marry within their 

group and have very similar lifestyles, envi-

ronment, and eating habits—they even cook 

in communal kitchens. Thus, Fraser-Liggett 

hoped to eliminate some of the variables that 

might have confounded other studies. 

The body mass index of Amish ranged 

from 16 to 51 (30 is obese), and some of the 

obese ones also had metabolic syndrome. 

Fraser-Liggett and her colleagues captured 

a snapshot of the gut microbiome of each 

Amish by obtaining stool samples, sequenc-

ing the DNA in them, and using the 16S ribo-

somal subunit gene often used to tell bacteria 

apart, identifying any microbial components. 

Although the scientists did detect some dif-

ferences in certain bacteria between obese 

and lean Amish, they didn’t fi nd the dramatic 

shifts that Gordon had documented between 

lean and obese mice, Fraser-Liggett reported 

at the meeting. 

One problem may be that simply taking a 

census of the bacteria present in a person’s gut 

may not be enough. “16S [analysis] is not very 

informative,” Ehrlich says. “We need to go to 

more precise measures.” Increasingly, micro-

biome researchers are looking at what bacte-

rial genes are active in a person and not just at 

which bacteria are there. Scientists have found 

that although the species mix of the micro-

biome may vary signifi cantly from one person 

to the next, those individuals often still have 

equivalent complements of bacterial genes at 

work inside them. Through such gene analy-

ses, researchers can begin to better assess 

what the bacteria in the gut are really doing to, 

or for, their host, Fraser-Liggett notes. 

Gordon suggests that more clarity on the 

obesity-microbiome issue will also come 

from using the guts of mice as bioreactors 

for human microbes. His group has pio-

neered the study of germ-free mice, which 

are grown in a sterile environment from 

birth, and he is now exposing such mice 

to bacteria from human guts. Once those 

human microbes have established residence 

in the guts of the mice, Gordon then feeds 

the animals a variety of 

human diets. Using these 

rodent proxies, he can 

thus track how different 

diets affect the “human” 

microbiomes, assessing 

the bacteria as often as he 

needs to to get a dynamic 

picture. “You can sample 

many features between the 

microbial community and 

the host,” he says. 

Despite his role in ignit-

ing the study of obesity 

and microbiomes, Gordon 

resists the idea that our gut 

bacteria are the sole expla-

nation for the growing 

number of obese people. 

“It’s not the dominant part 

of the problem; excessive 

energy intake is,” he con-

tends. The simplistic notion of only chang-

ing one’s gut bacteria to lose weight has been 

“hyped a lot,” he complains. 

Others, such as microbial ecologist Lip-

ing Zhao of Shanghai Jiao Tong University 

in China, are more convinced that the micro-

biome will prove important when it comes 

to obesity. “Increased abundance of ‘bad 

genes’ and [a] decrease of ‘good genes’ in 

our diet-disrupted gut microbiome [may] 

be the primary driving force for this obesity 

epidemic,” Zhao says. 

Even if Zhao’s prediction proves right, 

the studies to date make clear that the con-

nection between the microbiome and excess 

weight is complex, Fraser-Liggett says. For 

those looking to bacteria to stem the obesity 

epidemic, she concludes, “there’s clearly no 

magic formula.” –ELIZABETH PENNISI

Personalized microbiota. 

Germ-free mice grown in 

sterile environments and 

given human gut microbes 

provide a way to test links 

between bacteria and obesity.

Published by AAAS

 o
n
 J

a
n
u
a
ry

 1
6
, 
2
0
1
2

w
w

w
.s

ci
e
n
ce

m
a
g
.o

rg
D

o
w

n
lo

a
d
e
d
 f
ro

m
 

http://www.sciencemag.org/