ArticlePDF Available

Dietary intervention impact on gut microbial gene richness

Authors:

Abstract and Figures

Complex gene–environment interactions are considered important in the development of obesity. The composition of the gut microbiota can determine the efficacy of energy harvest from food and changes in dietary composition have been associated with changes in the composition of gut microbial populations. The capacity to explore microbiota composition was markedly improved by the development of metagenomic approaches, which have already allowed production of the first human gut microbial gene catalogue and stratifying individuals by their gut genomic profile into different enterotypes, but the analyses were carried out mainly in non-intervention settings. To investigate the temporal relationships between food intake, gut microbiota and metabolic and inflammatory phenotypes, we conducted diet-induced weight-loss and weight-stabilization interventions in a study sample of 38 obese and 11 overweight individuals. Here we report that individuals with reduced microbial gene richness (40%) present more pronounced dys-metabolism and low-grade inflammation, as observed concomitantly in the accompanying paper. Dietary intervention improves low gene richness and clinical phenotypes, but seems to be less efficient for inflammation variables in individuals with lower gene richness. Low gene richness may therefore have predictive potential for the efficacy of intervention.
Content may be subject to copyright.
LETTER
doi:10.1038/nature12480
Dietary intervention impact on gut microbial
gene richness
Aure
´
lie Cotillard
1,2
*, Sean P. Kennedy
3
*, Ling Chun Kong
1,2,4
*, Edi Prifti
1,2,3
*, Nicolas Pons
3
*, Emmanuelle Le Chatelier
3
,
Mathieu Almeida
3
, Benoit Quinquis
3
, Florence Levenez
3,5
, Nathalie Galleron
3
, Sophie Gougis
4
, Salwa Rizkalla
1,2,4
,
Jean-Michel Batto
3,5
, Pierre Renault
5
, ANR MicroObes consortium{, Joel Dore
´
3,5
, Jean-Daniel Zucker
1,2,6
, Karine Cle
´
ment
1,2,4
& Stanislav Dusko Ehrlich
3
Complex gene–environment interactions are considered important
in the development of obesity
1
. The composition of the gut micro-
biota can determine the efficacy of energy harvest from food
2–4
and
changes in dietary composition have been associated with changes
in the composition of gut microbial populations
5,6
. The capacity to
explore microbiota composition was markedly improved by the
development of metagenomic approaches
7,8
, which have already
allowed productionof the first human gut microbial gene catalogue
9
and stratifying individuals by their gut genomic profile into diffe-
rent enterotypes
10
, but the analyses were carried out mainly in non-
intervention settings. To investigate the temporal relationships
between food intake, gut microbiota and metabolic and inflamma-
toryphenotypes, we conducted diet-induced weight-loss and weight-
stabilization interventions in a study sample of 38 obese and
11 overweight individuals. Here we report that individuals with
reduced microbial gene richness (40%) present more pronounced
dys-metabolism and low-grade inflammation, as observed concomi-
tantly in the accompanying paper
11
. Dietary intervention improves
low gene richness and clinical phenotypes, but seems to be less
efficient for inflammation variables in individuals with lower gene
richness. Low gene richness may therefore have predictive potential
for the efficacy of intervention.
To examine relationshipsbetween variations in gut microbiota com-
position and bioclinical parameters after dietary intervention, we used
the approach termed quantitativemetagenomics
11
. Forty-nine obese or
overweight subjects were recruited and subjected to a 6-week energy-
restricted high-protein diet followed by a 6-week weight-maintenance
diet (Methods); the compliance was good, as indicated by a principal
component analysis (PCA) of 35 nutrients over time (Supplementary
Fig. 1). Bioclinical characteristics and detailed qualitative and quant-
itative features of individuals’ food intake were obtained at baseline, 6
and 12 weeks (Supplementary Tables 1 and 2). The 35% decrease in
energy intake after the first 6 weeks was associated with a reduction in
body-fat mass, adipocyte diameter and improvements in insulin sensi-
tivity and markers of metabolism and inflammation (Supplementary
Tables 1 and 3). During the weight-maintenance phase, intake of
nutrients tended to return to baseline values, whereas dietary total
energy, carbohydrate and lipid intake remained lower than at begin-
ning of the intervention (Supplementary Tables 2 and 3). Serum lipid
variables also tended to return to their basal levels as well, while a pro-
gressive reduction occurred in systemic inflammation markers.
We first examined the gut microbial composition of the study popu-
lation at baseline (Methods). A bimodal distribution of bacterial gene
number was observed (Fig. 1a), similar to the one found in a cohort of
292 Danish individuals
11
, albeit less distinct, possibly owing to a lower
cohort size. At a threshold of 480,000 genes, corresponding to that
from the accompanying manuscript
11
, there were 18 (40%) low gene
count (LGC) and 27 (60%) high gene count (HGC) individuals, har-
bouring on average 379,436 and 561,499 genes respectively, a one-
third difference. A difference in diversity between lean and obese
individuals was reported previously
12
, but the difference among the
obese was not described.
We then examined the baseline phenotypes of the study population.
The LGC group had significantly higher insulin resistance and fasting
serum triglyceride levels, as well as a tendency towards higher LDL cho-
lesterol and inflammation than the HGC group (Fig. 2); as observed in
the accompanying paper
11
. Analysing gene richness as a quantitative
variablegavesimilar results (SupplementaryTable 4). We concludethat
in two European countries, the individuals of the LGC group present
phenotypes that expose them to an increased risk of obesity-associated
co-morbidities. Antibiotic treatments, which lower the diversity, have
*These authors contributed equally to this work.
{A list of authors and affiliations appears at the end of the paper.
1
Institut National de la Sante
´
et de la Recherche Me
´
dicale, U872, Nutriomique, E
´
quipe 7, Centre de Recherches des Cordeliers, Paris 75006, France.
2
Universite
´
Pierre et Marie-Curie-Paris 6, Nutriomique,
15 rue de l’Ecole de Medecine, Paris 75006, France.
3
INRA, Institut National de la Recherche Agronomique, Metagenopolis, Jouy en Josas78350, France.
4
Institute of Cardiometabolism and Nutrition,
Assistance Publique-Ho
ˆ
pitaux de Paris, CRNH-Ile de France, Pitie
´
-Salpe
ˆ
trie
`
re, Boulevard de l’Hopital, Paris 75013, France.
5
INRA, Institut National de la Recherche Agronomique, UMR 1319 Micalis, Jouy
en Josas 78350, France.
6
Institut de Recherche pour le De
´
veloppement, IRD, UMI 209, UMMISCO, France Nord, Bondy F-93143, France.
40
1.0
1.0
0.5
0.0
0.0 0.5 1.0
0.9
0.8
0.7
0.6
0.5
0 5 10 15
35
30
25
20
15
10
5
0
0 200,000
0 wk
MOHL-1
MOHL-10
4.8 × 10
–4
1.5 × 10
–3
1.5 × 10
–3
1.7 × 10
–4
4.8 × 10
–4
4.8 × 10
–3
2.6 × 10
–4
2.6 × 10
–4
4.6 × 10
–4
4.1 × 10
–5
4.8 × 10
–4
1.2 × 10
–3
2.6 × 10
–4
2.6 × 10
–4
4.1 × 10
–5
1.9 × 10
–5
1.9 × 10
–5
1.7 × 10
–4
F. Prau
MOHL-11
MOHL-12
MOHL-13
MOHL-15
MOHL-17
MOHL-18
MOHL-2
MOHL-3
MOHL-4
MOHL-5
MOHL-6
MOHL-7
MOHL-8
MOHL-9
6 wks 12 wks 0 wk 6 wks 12 wks
400,000 600,000 800,000
R. Inul
French AUC = 0.96
Danes AUC = 0.95
q
q
LGC
HGC
AUC
112,116 674,445 264,036 688,820 229,983 679,219
112,116 674,445 264,036 688,820 229,983 679,219
Gene number Gene number
a
c
b
Gene number Species
Figure 1
|
Gut microbial composition of LGC (
n
5 18) and HGC (
n
5 27)
subjects. a, Baseline gene count. b, Presence and frequency of 25 tracer genes
for species differentially abundant in LGC and HGC groups; Mann–Whitney
probability (q, false discovery rate (FDR) adjusted) is given. Genes are in rows,
frequency is indicated by colour gradient (white, not detected; red, most
abundant); individuals, ordered by increasing gene number, are in columns.
c, Highest AUC values for a combination of a given number of species in a ROC
analysis of 45 individuals of our cohort (red) and 292 individuals of the Danish
cohort
11
. Inset, AUC for the combination of six species.
29 AUGUST 2013 | VOL 500 | NATURE | 585
Macmillan Publishers Limited. All rights reserved
©2013
been reported to improve the hormonal, metabolic and inflammatory
status of obese mice; this apparent contradiction may be due to a res-
toration of a balance of the pro-inflammatory and inflammatory bac-
terial species in mice. Interestingly, LGC subjects seemed to consume
less fruits and vegetables and less fishery products than HGC subjects
(Supplementary Tables 4–6), raising the possibility that long-term
dietary habits may affect gene richness and the associated phenotypes,
as suggested for the elderly
13
.
We next searched for bacterial species differentially abundant in the
LGC and the HGC groups. To this aim, we first identified the genes
that had significantly different frequencies in the LGC and HGC
groups and then clustered the genes supposedly from the same species
by a frequency-based covariance analysis (Methods). We identified
6,230 genes that were different according to a Mann–Whitney test
(P , 0.0001); 4,462 (72%) were grouped into 112 clusters containing
at least 2 genes with a Spearman correlation coefficient r . 0.85. A vast
majority of these genes (3,966; 89%) were found in only 18 clusters,
which originate from species differentially abundant in the LGC and
HGC groups (Supplementary Table 7). The relative abundance of the
18 clusters in each individual was computed as a mean frequency of the
25 tracer genes for each cluster; all were significantly more abundant
among the HGC individuals (Fig. 1b and Supplementary Table 8).
To test whether the LGC and HGC individuals could be distingui-
shed using the 18 species represented by the tracer genes, we carried
out an exhaustive receiver operating characteristic (ROC) analysis of
all clusters combinations, with tenfold cross validation, using 90% of
individuals for computation and the remaining 10% for test (Methods).
The best area under the curve (AUC) values for combination of differ-
ent numbers of species are shown in Fig. 1c; they ranged between 0.96
and 0.99 for 2 to 9 species combinations, indicating an almost perfect
stratification of LGC and HGC individuals. Interestingly, 14 of the 18
species represented by the tracer genes (78%) were also identified as
differentially abundant among the LGC and HGC individuals in a
larger Danish cohort
11
. Not surprisingly, the combinations yielding
the best AUC values for our cohort also efficiently stratified LGC
and HGC Danes (Fig. 1c). This indicates that the LGC and HGC
individuals from two European countries differ in a similar way, not
only by their clinical phenotypes but also by specific features of their
gut microbiota.
Very interestingly, gene richness increased significantly in the LGC
group after the energy-restricted diet and remained after the stabiliza-
tion phase higher than at baseline even though a slight downwards
trend was apparent, whereas it didnot change significantly during inter-
vention in the HGC group (Fig. 3a). We conclude that a dietary inter-
vention can correct a putative loss of richness in the LGC group, albeit
partially, as the difference between the LGC and HGCgroups remained
significant at the end of the intervention.
To investigate the potential effect of the increase in gene richness on
patient status we analysed association of the changes of richness and of
bioclinical variables. Increase of gene richness was associated with a
significant decrease in adiposity measures (hip circumference and total
fatmass) and circulating cholesterolas wellas a trendtowards a decrease
in inflammation (highly sensitive C-reactive protein) (Supplementary
Table 9). These results suggest that the correction of a putative loss of
microbial richness is associated with an improvement of the systemic
metabolic status. However, although the inflammation was decreased
in all individuals, the difference between LGC and HGC individuals
was not attenuated (Fig. 2). Low basal gene richness was also associated
with increased adipose tissue inflammatory cells at 6 weeks and increased
–10
–8
–6
–4
–2
0
0 wk 6 wks 12 wks 0 wk 6 wks 12 wks
0 wk 6 wks 12 wks 0 wk 6 wks 12 wks
**
Disse index
0.0
0.5
1.0
1.5
*
HOMA−IR
mmol l
–1
0.0
0.5
1.0
1.5
2.0
2.5
***
**
*
Triglycerides
mg dl
–1
0
2
4
6
8
**
*
hsCRP
Figure 2
|
Differences between LGC and HGC subjects in bioclinical
variables. White and black bars refer to LGC (n 5 18) and HGC (n 5 27)
groups, respectively; error bars denote s.e.m. 0 weeks, baseline; 6 weeks, end of
the energy restriction period; and 12 weeks, end of stabilization period.
*P , 0.1, **P , 0.05, ***P , 0.01 by Mann–Whitney tests. ‘Disse index’ is
calculated by combining lipid and insulin values (see Supplementary
Information). HOMA-IR, homeostatic model assessment of insulin resistance;
hsCRP, highly sensitive C-reactive protein.
024681012
Time (weeks)
**** ** **
***
***
350,000
400,000
450,000
500,000
550,000
600,000
650,000
Gene richness
a
Cluster 61
**
*
*
b
Cluster 130
**
Cluster 213
**
Cluster 216
**
0612
0612
–14
–12
–10
–8
–6
Cluster 411
Time (weeks)
Mean log
–14
–12
–10
–8
–6
Mean log
–14
–12
–10
–8
–6
Mean log
–14
–12
–10
–8
–6
Mean log
–14
–12
–10
–8
–6
Mean log
–14
–12
–10
–8
–6
Mean log
–14
–12
–10
–8
–6
Mean log
–14
–12
–10
–8
–6
Mean log
**
Cluster 461
Time (weeks)
0612
Time (weeks)
0612
Time (weeks)
0612
Time (weeks)
0612
Time
(
weeks
)
0612
Time
(
weeks
)
0612
Time
(
weeks
)
**
Cluster 494
**
Cluster 640
***
Figure 3
|
Gene richness of LGC and HGC groups during the intervention.
Data are mean 6 s.e.m. Black line, HGC (n 5 27); grey line, LGC (n 5 18).
Differences between HGC and LGC groups were tested using Mann–Whitney
tests (black asterisks). *P , 0.1, **P , 0.05, ***P , 0.01, ****P , 0.001.
a, Overall pattern of variation. For each group, differences between one time
point and basal state were tested using Wilcoxon signed-rank tests (grey
asterisks). b, Variation of eight clusters that were significantly different at
baseline and modulated by the dietary intervention.
RESEARCH LETTER
586 | NATURE | VOL 500 | 29 AUGUST 2013
Macmillan Publishers Limited. All rights reserved
©2013
systemic inflammation at 12 weeks (Supplementary Table 4). Further-
more, higher gene richness at baseline was associated with a more mar-
ked improvement of adipose tissue and systemic inflammation (delta
changes at 6 and 12 weeks, respectively; Supplementary Table 10). Gene
richness may therefore help to predict the efficacy of dietary interven-
tion on inflammatory variables in overweight or obese individuals.
To further explore the effects of dietary intervention on gut micro-
bial species we used a gene clustering procedure similar to the one des-
cribed above for the comparison of LGC and HGC individuals. A set of
213,532 genes that varied significantly in frequency between different
time points (Wilcoxon signed-rank test, P , 0.05) was first identified.
To reduce the complexity of the data set, an entropy-filtering criterion
was then applied, removing the genes present in only a few individuals
(Supplementary Fig. 2). The remaining 58,109 genes were clustered by
frequency covariance (Methods and Supplementary Fig. 3). Some
34,920 genes (60%) were grouped in 39 clusters larger than 100 genes
(Supplementary Table 11); a large majority, 72%, were very compact,
with a clustering coefficient .0.75 (ref. 14) (Supplementary Informa-
tion, see cluster sheets for a more detailed description). Of the 39
clusters, 17 had $80% of their genes assigned to the same species and
19 to the same genus (the global distribution was 64% Firmicutes, 33%
Bacteroidetes, and 3% Actinobacteria; Supplementary Table 11), con-
firming a species-specific clustering (Methods).
The abundance of the potential species represented by the 39 clus-
ters was computed as the sum of the respective gene frequencies, and
variations over time and correlations withbioclinical variables and food
items were examined (Methods and Supplementary Tables 12 and 13).
We observed that the abundance of 26 clusters varied significantly with
time, indicating that a number of bacterial species can be modulated by
nutritional intervention; the remaining 13 were not studied further. Only
a few of our gene clusters decreased or showed a tendency to decrease
during the calorie restriction phase, but one of those was assigned to
Eubacterium rectale and another one to Bifidobacterium spp., in accord-
ance with previous results
6
(Supplementary Table 11 and Supplementary
Information).
The main trend after 6 weeks of energy-restricted diet was a signifi-
cant increase of abundance of most gene clusters (n 5 15), whereas the
trend was opposite after 6 weeks of weight-maintenance diet, as the
abundanceof 14 species decreased. A total of five different patterns was
observed, reflecting combinations of variation during the two periods
(Supplementary Table 11), but the overall tendency was to return close
to a baseline level by the end of the weight-maintenance phase (illu-
strated in Supplementary Informatino), suggesting a transient effect of
dietary intervention on gut microbiota, as described previously
15
.Interes-
tingly, for 8 of the 26 gene clusters that had a significantly lower abund-
ance in the LGC than HGC individuals at baseline (Supplementary
Table 14), the energy-restricted diet led to an increase of abundance in
the LGC individuals, bringing them close to the level found in the HGC
individuals (Fig. 3b); there was no significant abundance difference
between the LGC and HGC individuals upon the stabilization phase.
We conclude that the dietary intervention, in spite of its overall transient
effect, may lead to more persistent changes of some gut microbial species.
Quantitative metagenomics analysis ofthe gut microbiome in3 diffe-
rent samples for each of the 49 French (our study) and in 292 Danish
subjects
11
revealed the existence of a high proportion of individuals
(23–40%) with low microbial richness. In both study populations, a
detailed clinical analysis indicated that these individuals show adipo-
sity associated dyslipidaemia, higher insulin resistance and low-grade
inflammation when compared to their higher-gene-diversity counter-
parts. This deleteriousphenotypeis known to be associatedwith increa-
sed risk of pre-diabetes, type 2 diabetes, hepatic and cardiovascular
disorders as well as some forms of cancer
16–18
. In both study popula-
tions, abundance of many gut bacterial species in low-richness indivi-
duals was altered in a similar way relative to high richness individuals;
this alteration can be accurately detected by combinations of only a few
bacterial species. This indicates that simple diagnostic tests, based on
our ‘other genome could be developed to identify individuals at a
higher risk of obesity-associated co-morbidities. In the context of the
current global epidemics of obesity and metabolic disorders, such tests
could have a broad usefulness.
The concomitant improvement of gut microbial gene richness and
bioclinical variables in LGC individuals by a dietary intervention sug-
gests a possibility to advance from risk detection to risk alleviation,
under the assumption that the less rich microbiota are also less healthy
(see the accompanying paper
11
). Restoration of gene richness was not
achieved fully by our short-term intervention, but seems to be a desir-
able goal, as decreased gene richness was found to be associated with a
less efficient improvement of the inflammatory variables by dietary
intervention. Interestingly, increasedconsumptionof fruits and vegeta-
ble and thus higher fibre consumption before the intervention seemed
to be associated with high bacterial richness. This finding, although
exploratory in nature and requiring replication, supports a recently
reported link between long-term dietary habits and the structure of
gut microbiota
15
and suggests that a permanent change of microbiota
may be achieved by appropriate diet. Development of a two-pronged
approach, coupling early detection of an impending loss of gut bacterial
richness to appropriate nutritional recommendations, which is yet to
be established, may help to reach this goal and possibly contribute to
diminish the risk of the obesity-linked co-morbidities; stratification by
gene richness may have predictive value in respect to the efficacy of a
dietary treatment and even guide its choice. However, low-grade
inflammation, an important trait related to obesity but also common
to many chronic diseases, seemed relatively refractory to dietary inter-
vention in the LGC individuals, suggesting that specific therapeutic
actions, aiming at restoring gut microbiota richness and equilibrium
in obesity and altered metabolism, may need to be developed as well.
METHODS SUMMARY
Forty-nine obese or overweight subjects were recruited and subjected to a 6-week
energy-restricted high-protein diet followed by a 6-week weight-maintenance diet.
Bioclinical characteristics,physical activity scores and detailed qualitative andquan-
titative features of their food intake were obtained at baseline, 6 and 12 weeks
(Methods). The clinical trial was registered at http://www.ClinicalTrials.gov under
study number NCT01314690. The Ethical Committee of Ho
ˆ
tel Dieu Hos-
pital approved the clinical study and all subjects provided written informed con-
sent. Faecal samples were collected at each time point and analysed with the next
generation sequencing SOLiD System. After read mapping a frequency table of
microbial genes was obtained (Methods).
Two groups of patients with LGC and HGC were definedusing the gene-richness
distribution. Differences in terms of food, bioclinical variables and gene abundance
were identified by standard statistical methods (Methods). Focusing on the dietary
intervention and using a multi-criteria selection to narrow down the number of
genes to a few thousands, gene clusters of co-varying microbial genes were con-
structed. These resulting gene clusters were then analysed for changes over time
and correlations with bioclinical markers (Methods).
Full Methods and any associated references are available in the online version of
the paper.
Received 12 April 2012; accepted 17 July 2013.
1. Mutch, D. M. & Cle
´
ment, K. Unraveling the genetics of human obesity. PLoS Genet.
2, e188 (2006).
2. Ba
¨
ckhed, F. et al. The gut microbiota as an environmental factor that regulates fat
storage. Proc. Natl Acad. Sci. USA 101, 15718–15723 (2004).
3. Turnbaugh, P. J. et al. An obesity-associated gut microbiome with increased
capacity for energy harvest. Nature 444, 1027–1031 (2006).
4. Ba
¨
ckhed, F., Manchester, J. K., Semenkovich, C. F. & Gordon, J. I. Mechanisms
underlying the resistance to diet-induced obesity in germ-free mice. Proc. Natl
Acad. Sci. USA 104, 979–984 (2007).
5. Ley, R. E., Turnbaugh, P. J., Klein, S. & Gordon, J. I. Microbial ecology: human gut
microbes associated with obesity. Nature 444, 1022–1023 (2006).
6. Duncan, S. H. et al. Reduced dietary intake of carbohydrates by obese subjects
results in decreased concentrations of butyrate and butyrate-producing bacteria
in feces. Appl. Environ. Microbiol. 73, 1073–1078 (2007).
7. Riesenfeld, C. S., Schloss, P. D. & Handelsman, J. Metagenomics: genomic analysis
of microbial communities. Annu. Rev. Genet. 38, 525–552 (2004).
8. National Research Council. The New Science of Metagenomics: Revealing the
Secrets of Our Microbial Planet (The National Academies Press, 2007).
LETTER RESEARCH
29 AUGUST 2013 | VOL 500 | NATURE | 587
Macmillan Publishers Limited. All rights reserved
©2013
9. Qin, J. et al. A human gut microbial gene catalogue established by metagenomic
sequencing. Nature 464, 59–65 (2010).
10. Arumugam, M. et al. Enterotypes of the human gut microbiome. Nature 473,
174–180 (2011).
11. Le Chatelier, E. et al. Richness of human gut microbiome correlates with
metabolic markers. Nature http://dx.doi.org/10.1038/nature12506
(this issue).
12. Turnbaugh, P. J. et al. A core gut microbiome in obese and lean twins. Nature 457,
480–484 (2009).
13. Claesson, M. J. et al. Gut microbiota composition correlates with diet and health in
the elderly. Nature 488, 178–184 (2012).
14. Wasserman, S. & Faust, K. Social Network Analysis: Methods and Applications.
(Cambridge Univ. Press, 1994).
15. Wu, G. D. et al. Linking long-term dietary patterns with gut microbial enterotypes.
Science 334, 105–108 (2011).
16. Ouchi, N., Parker, J. L., Lugus, J. J. & Walsh, K. Adipokines in inflammation and
metabolic disease. Nature Rev. Immunol. 11, 85–97 (2011).
17. Shoelson, S. E., Lee, J. & Goldfine, A. B. Inflammation and insulin resistance. J. Clin.
Invest. 116, 1793–1801 (2006).
18. Renehan, A. G., Tyson, M., Egger, M., Heller, R. F. & Zwahlen, M. Body-mass index
and incidence of cancer: a systematic review and meta-analysis of prospective
observational studies. Lancet 371, 569–578 (2008).
Supplementary Information is available in the online version of the paper.
Acknowledgements We are grateful to O. Pedersen (Univ. Copenhagen) for helpful
comments on this manuscript and to the MetaHIT consortium for providing the gene
profiles of the Danish subjects used to test the ROC models in advance of publication
and the DNA samples sequenced on the SOLiD platform for comparison with the
Illumina platform used in the accompanying manuscript.We thank C. Baudoin,P. Ancel
and V. Pelloux who contributed to the clinical investigation study; S. Fellahi and J.-P.
Bastard for analyses of inflammatory markers; D. Bonnefont-Rousselot and R. Bittar for
help with the analysis of plasma lipid profile. This work was supported by Agence
Nationale de la Recherche (ANR MICRO-Obes, ANR, Nutra2sens, ANR-10-IAHU-05),
the Metagenopolis grant ANR-11-DPBS-0001, KOT-Ceprodi (Florence Massiera),
Danone Research (Damien Paineau) and the associations Fondacoeur, and
Louis-Bonduelle. Additional funding came from the European Commission FP7 grant
HEALTH-F4-2007-201052 and METACARDIS.
Author Contributions S.D.E., J.D. and K.C. designed the study; S.D.E., J.D., K.C. and P.R.
managed the study; K.C. and S.R. designed the clinical research; S.R. and L.C.K.
conducted the clinical research and clinical data management; A.C., S.R. and L.C.K.
conducted clinical and dietary data analysis; S.G. gave dietary counselling to the
patients and carried out analysis of dietary data; F.L. prepared the DNA for sequencing;
S.K. managed DNA sequencing, which B.Q. and N.G. carried out; N.P. and J.-M.B.
established the sequence analysis pipeline; A.C., J.-D.Z., E.P., N.P., E.L.C., M.A., J.-M.B.,
S.K. and S.D.E. carried out microbial data analysis; A.C., K.C., L.C.K. and S.D.E. wrote the
manuscript.
Author Information The raw solid read data for all samples has been deposited in the
European Bioinformatics Institute (EBI) European Nucleotide Archive (ENA) under the
accession number ERP003699. Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Readers are welcome to comment on the online version of the paper. Correspondence
and requests for materials should be addressed to S.D.E. (dusko.ehrlich@jouy.inra.fr)
or K.C. (karine.clement@psl.aphp.fr).
ANR MicroObes consortium members
Herve
´
Blottie
`
re
1,2
,MarionLeclerc
1
, Catherine Juste
1
, Tomas de Wouters
1
, Patricia
Lepage
1
, Charlene Fouqueray
1
, Arnaud Basdevant
3
, Cornelieu Henegar
3
,Cindy
Godard
3
, Marine Fondacci
3
, Alili Rohia
3
, Froogh Hajduch
3
, Jean Weissenbach
4
,Eric
Pelletier
4
, Denis Le Paslier
4
, Jean-Pierre Gauchi
5
, Jean-François Gibrat
6
, Valentin
Loux
6
, Wilfrid Carre
´
6
, Emmanuelle Maguin
1
, Maarten van de Guchte
1
, Alexandre
Jamet
1
, Fouad Boumezbeur
1
&Se
´
verine Layec
1
1
INRA, Institut National de la Recherche Agronomique, UMR 1319 Micalis, Jouy en Josas
78350, France.
2
INRA, Institut National de la Recherche Agronomique, Metagenopolis,
Jouy en Josas 78350, France.
3
Institute of Cardiometabolism and Nutrition, Assistance
Publique-Ho
ˆ
pitaux de Paris, CRNH-Ile de France, Pitie
´
-Salpe
ˆ
trie
`
re, Paris 75013, France.
4
Commissariat a
`
l’Energie Atomique, Genoscope, Evry 91000, France.
5
Institut National
de la Recherche Agronomique, Mathe
´
matiques et Informatique Applique
´
es, Jouy en
Josas 78350, France.
6
Institut National de la Recherche Agronomique, Mathe
´
matique,
Informatique et Ge
´
nome, Jouy en Josas 78350, France.
RESEARCH LETTER
588 | NATURE | VOL 500 | 29 AUGUST 2013
Macmillan Publishers Limited. All rights reserved
©2013
METHODS
Clinical investigation. Obese (n 5 38) and overweight (n 5 11) subjects, 8 men
and 41 women, were recruited for a 12-week controlled dietary intervention at the
Center of Research in Human Nutrition, Pitie
´
-Salpe
ˆ
trie
`
re Hospital, Paris, France.
The subjects included in the study had no chronic pathologies except excess body
weight.Their body weightwas stable within 3 months before the study. None of the
participants was undergoing chronic treatment or had been involved in weight-
loss programs in the preceding 12 months. No antibiotics or drugs were taken
within 2 months before or during the course of the study. The Ethical Committee
of Ho
ˆ
tel Dieu Hospital approved the clinical study and subjects provided written
informed consent. In thefirst6-week phase, subjects consumed an energy-restricted
high-protein diet (1,200 kilocalories (kcal) per day for women and 1,500 kcal for
men: 35% proteins, 25% lipids, 44% carbohydrates) with low glycaemic index
carbohydrates and enrichment with soluble fibres
19
. This phase was followed by
a second 6-week body weight stabilization period with 20% increase in total energy
intake, above their resting energy metabolic rate. At 0, 6 and 12 weeks, blood and
faecal samples were collected and anthropometric measurements were performed.
Subjects filled a 7-day dietary record and were interviewedby a registered dietician.
On the visit day, the dietician checked the information and clarified any ambiguit-
ies regarding detail of food consumed. All records were analysed by the registered
dietician using the computer software program PROFILE DOSSIER V3 (Audit
Conseil en Informatique Me
´
dicale), which has a dietary database initially made up
of 400 food items representative of the French diet as described previously
20
.A
nutrientanalysiswas generated foreach subject. Bodycomposition was determined
by dual-energy X-ray absorptiometry (DEXA). Blood samples were obtained after
12 h of fasting to measure total cholesterol, high-density lipoprotein (HDL) cho-
lesterol, triglycerides, insulin, glucose, and inflammatory markers (hsCRP and
interleukin 6 (IL-6)) as described previously
21
. Insulin resistance was estimated
using HOMA-IR and Disse index scores
22,23
. Subcutaneous abdominal adipose
tissue samples were obtained at all time points by needle biopsy from the perium-
bilical area under local anaesthesia (1% xylocaine) to measure the adipocytes
diameter
24
and for immunohistochemicalstudies (HAM561-stainedmacrophages
in adipose tissue). Whole faecal samples were self-collected in sterile boxes and
stored at 220 uC within 4 h, sampled (200-mg aliquots) and then stored at
280uC until analysis. Paired Wilcoxon tests were performed to analyse changes
in these variables between various time points (P , 0.05). P values were adjusted
for multiple testing using the Benjamini–Hochberg procedure
25
.
Metagenomic sequencing. Intestinal bacterial gene content of 49 obese and over-
weight individuals at 3 time-points (baseline, week 6 and week 12) was determined
by high-throughput ABI SOLiD sequencing technology of total faecal DNA. An
average of 76.5 million 6 36.5 million (mean 6 s.d.) 35-base-long single reads
were determined for each sample (a total of 393 Gb of sequence) (Supplementary
Table 15). By using corona_lite (v4.0r2.0), an averageof 24.8 million 6 14.3million
reads per individual were mapped on the reference catalogue of 3.3 million genes
9
with a maximum of 3 mismatches. Reads mapping at multiple positions were dis-
cardedand an average of14.2 million6 8.1 million uniquely mappedreads perindi-
viduals were retained for estimating the abundance of each reference gene by using
METEOR
26
software. Abundance of each gene in an individual was normalized
with METEOR by dividing the number of reads that uniquely mapped to a gene by
its nucleotide length. After that, normalized gene abundances were transformed in
frequencies by dividing them with the total number of uniquely mappedreads for a
given sample. The resulting set of gene frequencies, termed as a microbial gene
profile of an individual, was used for further analyses.
Comparison between SOLiD and Illumina sequencing technologies. Two pri-
mary short-read technologies currently exist for quantitative metagenomic analysis;
SOLiD and Illumina. To validate data set correspondences and comparisons between
results in this study and the accompanying paper
11
, 24 samples from the Danish
Inter99cohort,previously sequencedon an Illumina GAplatform,werealsosequen-
ced and analysed by SOLiD technology . Representative samples for cross-comparison
included 14 females and 10 males, 15 obese and 9 lean, and 15 HGC and 9 LGC
individuals. Hierarchical clustering demonstrated all samples self-clustered as
technology-independent pairs, with the average Pearson correlation coefficient of
0.87(computedupon logtransformation)betweenthe two technologiesand increa-
sing concordance associated with increased signal (Supplementary Fig. 4).
Gene-richness analysis. Gene richness was compared between subjects using the
same number of mapped reads. Data were downsized to adjust for technical vari-
ability linked to different sequencing depths. This downsizing was performed at
different levels by randomly selecting 4.5 or 7 million mapped reads for each
sample and then computing the mean number of genes over 30 drawings (Sup-
plementary Table 15). The 4.5-million-read downsizing allows keeping more than
90% of the individuals at each time point (required for the quantitative analysis of
gene richness), but shrinks the data distribution (Supplementary Fig. 5). The
7-million-read downsizing was used for the analysis of the gene count distribution
among the individuals and the enterotypes. The distribution of gene number
obtained with the two downsizings is quite similar as shown by Spearman cor-
relation (r . 0.99) (Supplementary Fig. 5).
Differentially abundant gene clusters between LGC and HGC. Two groups of
patients with LGC and HGC were defined using the 480,000-gene threshold,
consistent with the accompanying manuscript
11
(Fig. 1a, and main text). Genes
significantly different in groups of individuals were identified by Mann–Whitney
tests using a P-value threshold of ,0.0001. They were clustered by an abundance-
based binning strategy, using the covariance of their gene frequency profiles
among the individuals of the cohort, as described in the accompanying paper
11
.
Abundance of a given cluster in each individual was estimated as a mean abund-
ance of 25 arbitrarily selected ‘tracer’ genes for each cluster; these values were close
to those obtained by using all the genes of a cluster.
ROC analysis. The analyses were carried out to distinguish between HGC and
LGC individuals by a combination of gene clusters. For each combination, only a
single decision model was considered, computed as the sum of mean abundance of
clusters with greater abundance in HGC than in LGC minus the sum of those with
greater abundance in LGC than in HGC. As opposed to the infinite number of
regression models, such models are finite and can be exhaustively explored. To
select the best models, we used the cross-validated area under the ROC curve cross-
validated AUC criterion
27
well adapted to classification models for binary outcome
data.
Correlations between microbial gene clusters and clinical variables. Mann–
Whitney tests were used to compare bioclinical variables, food items and gene
clusters between LGC and HGC groups at each time point. Associations between
quantitative basal gene richness and bioclinical or food variables, or differences
(deltas) in bioclinical or food variables were investigated using linear models. For
the associations betweendeltas of bioclinicalparametersand deltas of generichness,
all pairs of deltas were computed (6 weeks–0 weeks, 12 weeks–6 weeks, 12 weeks
0 weeks). Linearmixed modelswere thenfitted using alldata.A P-valuethresholdof
0.05 was applied for statistical significance. Owing to the highly correlated biocli-
nical and food variables, adjustment for multiple testing is not really adequate, but
the false discovery rates (Benjamini–Hochberg
25
) are given for information pur-
poses in Supplementary Table 9.
Taxonomical annotation. The genes from clusters were mapped by BLASTN
(BLAST 2.2.24, default parameters) against a collection of 6,006 genomes (the
available reference genomes from NCBI and the set of draft gastrointestinal gen-
omes from the DACC and MetaHIT as of the 03.08.2012). Following taxonomical
assignation parameters described by Arumugam
10
, each gene was assigned with
the taxonomy of the best-hit covering $80% of the gene length and according to
the identity threshold for the taxonomic rank ($65% for phylum, $ 85% for genus
and $ 90% for species). To assess the taxonomy of clusters below these thresholds
we used BLASTP against the non-redundant sequences databases available at
NCBI. Based on the criterion of the homogeneity of the best hit taxonomic assign-
ment (at least 80% of tracer genes from a cluster having the same taxonomic best
hit assignment), 100% and 25% of the clusters could be assigned at a phylum and
genus level, respectively (Supplementary Table 7).
Gene clusters affected by the dietary intervention. The analysis was carried out
with genes with a potentially dietary linked signal. The first filtering step consisted
in selecting the genes whose frequency was modulated significantly by the nutri-
tional intervention during the dietary restriction or the stabilization period with a
Wilcoxon signed-rank test (P , 0.05). A subset of these genes, with high Shannon
entropy
28
, was selected in a second filtering step. The entropy distribution of the
filtered genes presented a bimodal distribution and the genes corresponding to the
highest mode were selected using a threshold estimation on an approximation of
its density function
29
(Supplementary Fig. 2). The genes with high entropy were
mostly shared among individuals of the cohort. Genes with significantly similar
frequency profiles (P divided by number of tests , 0.05) and high Spearman
correlation coefficient (r . 0.85), were clustered in a way similar to the LGC–
HGC clusters using single-linkage clustering (Supplementary Fig. 3). The 39
clusters with a size superior to 100 genes were kept for further analyses. The group
abundance of each cluster was computed as the sum of the frequencies of its genes,
and the data were log-transformed for parametric statistics.
Gene-cluster analysis. Gene clusters were analysed for changes over time and cor-
relations with bioclinical markers using linear mixed models were adjusted for age
and sex (Supplementary Tables 12 and 13). The highly correlated data induced
P values distributions not adapted to standard procedures for multiple testing adjust-
ments; nevertheless, we provide the false discovery rates using the Benjamini–
Hochberg method in Supplementary Tables 12 and 14. All statistical analyses were
performed using the R environment
30
.
19. Rizkalla, S. W. et al. Differential effects of macronutrient content in 2
energy-restricted diets on cardiovascular risk factors and adipose tissue cellsize in
LETTER RESEARCH
Macmillan Publishers Limited. All rights reserved
©2013
moderately obese individuals: a randomized controlled trial. Am. J. Clin. Nutr. 95,
49–63 (2012).
20. Bouche
´
,C.et al. Five-week, low-glycemic index diet decreases total fat mass and
improves plasma lipid profile in moderately overweight nondiabetic men.Diabetes
Care 25, 822–828 (2002).
21. Tordjman, J. et al. Structural and inflammatory heterogeneity in subcutaneous
adipose tissue: Relation with liver histopathology in morbid obesity. J. Hepatol. 56,
1152–1158 (2012).
22. Disse, E. et al. A lipid-parameter-based index for estimating insulin sensitivity and
identifying insulin resistance in a healthy population. Diabetes Metab. 34,
457–463 (2008).
23. Antuna-Puente, B. et al. Evaluation of insulin sensitivity with a new lipid-based
index in non-diabetic postmenopausal overweight and obese women before and
after a weight loss intervention. Eur. J. Endocrinol. 161, 51–56 (2009).
24. Prat-Larquemin, L. et al. Adipose angiotensinogen secretion, blood pressure, and
AGT M235T polymorphism in obese patients. Obes. Res. 12, 556–561 (2004).
25. Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical
and powerful approach to multiple testing. J. Roy. Stat. Soc. B 57, 289–300
(1995).
26. Pons, N. et al. METEOR, a platform for quantitative metagenomic profiling of
complex ecosystems. Journe
´
es Ouvertes en Biologie, Informatique et
Mathe
´
matiques http://www.jobim2010.fr/sites/default/files/presentations/
27Pons.pdf (2010).
27. Jiang, D., Huang, J. & Zhang, Y. The cross-validated AUC for MCP-logistic
regression with high-dimensional data. Stat. Methods Med. Res. http://dx.doi.org/
10.1177/0962280211428385 (28 November 2011).
28. Shannon, C. E. A mathematical theory of communication. Bell Sys. Tech. J. 27,
379–423 (1995), 623–656 (1948).
29. Silverman, B. W. Density Estimation for Statistics and Data Analysis (Chapman and
Hall, 1986).
30. R Development Core Team. R: A Language and Environment for Statistical Computing
http://www.R-project.org (R Foundation for Statistical Computing, 2011).
RESEARCH LETTER
Macmillan Publishers Limited. All rights reserved
©2013
... Thus, initial success is great, but long-term success is moderate in most cases, indicating the need for long-term maintenance strategies and reliable predictor variables of success. The analysis also shows that the outcome of such weight reduction programs is variable, and determinants of the outcome are of major interest, among which the intestinal microbiota might play a role [3,6]. ...
... It is known that the type of intervention and individual factors influence the outcome of weight-reductions means [2,3]. Among the individual factors, the intestinal microbiota is a highly interesting data source that might yield such predictive variables, since it has been shown that obesity and metabolic processes and disease are affected by individual microbiota patterns [6][7][8][13][14][15]. Here, we show for the first time that defined microbiota patterns can be related to individual weight loss success following a nonsurgical, multidisciplinary weight loss program performed over one year. ...
Article
Full-text available
Background: The long-term success of nonsurgical weight reduction programs is variable; thus, predictors of outcome are of major interest. We hypothesized that the intestinal microbiota known to be linked with diet and obesity contain such predictive elements. Methods: Metagenome analysis by shotgun sequencing of stool DNA was performed in a cohort of 15 adults with obesity (mean body mass index 43.1 kg/m2) who underwent a one-year multidisciplinary weight loss program and another year of follow-up. Eight individuals were persistently successful (mean relative weight loss 18.2%), and seven individuals were not successful (0.2%). The relationship between relative abundancies of bacterial genera/species and changes in relative weight loss or body mass index was studied using three different statistical modeling methods. Results: When combining the predictor variables selected by the applied statistical modeling, we identified seven bacterial genera and eight bacterial species as candidates for predicting success of weight loss. By classification of relative weight-loss predictions for each patient using 2–5 term models, 13 or 14 out of 15 individuals were predicted correctly. Conclusions: Our data strongly suggest that gut microbiota patterns allow individual prediction of long-term weight loss success. Prediction accuracy seems to be high but needs confirmation by larger prospective trials.
... The improvement in glucose metabolism, and the regulation of gut microbiota and plasma metabolites by FMT from lean donors are dependent on the reduction in fecal microbial diversity at baseline. Thus, pretreatment fecal microbiota characteristics may differ in response to bacterial species of lean donors, and thus pretreatment status may be correlated with the treatment effects (44,58,74). ...
Article
Full-text available
Type 2 diabetes mellitus (T2DM), one of the fastest growing metabolic diseases, has been characterized by metabolic disorders including hyperglycemia, hyperlipidemia and insulin resistance (IR). In recent years, T2DM has become the fastest growing metabolic disease in the world. Studies have indicated that patients with T2DM are often associated with intestinal flora disorders and dysfunction involving multiple organs. Metabolites of the intestinal flora, such as bile acids (BAs), short-chain fatty acids (SCFAs) and amino acids (AAs)may influence to some extent the decreased insulin sensitivity associated with T2DM dysfunction and regulate metabolic as well as immune homeostasis. In this paper, we review the changes in the gut flora in T2DM and the mechanisms by which the gut microbiota modulates metabolites affecting T2DM, which may provide a basis for the early identification of T2DM-susceptible individuals and guide targeted interventions. Finally, we also highlight gut microecological therapeutic strategies focused on shaping the gut flora to inform the improvement of T2DM progression.
... For determination of gene abundance, the reads were realigned with the gene catalog with Bowtie 2. Only genes with two or more mapped reads were deemed to be present in a sample [26]. The relative abundance of each gene in each sample was calculated according to the number of mapped reads and the length of the gene [27][28][29]. ...
Article
Full-text available
Hyalomma marginatum is an important arthropod vector in the transmission of various zoonoses. The aim of this study was to identify the tick-borne pathogens (TBPs) maintained in Hy. marginatum in Tibet and to estimate the risk of human tick-borne diseases. Adult Hy. marginatum ticks (n = 14) feeding on yaks were collected. The individual DNA samples of these ticks were sequenced with metagenomic next-generation sequencing to survey the presence of TBPs. TBPs in individual ticks were identified with nested polymerase chain reaction (PCR) combined with DNA sequencing. The presence of Rickettsia, Anaplasma, and Ehrlichia in individual ticks was indicated by the taxonomic profiles at the genus level, but only Rickettsia aeschlimannii (100%, 13/13) was further detected in the ticks by nested PCR. Conclusion: This study provides information on the microbial communities of Hy. marginatum in Tibet, China, and provides the first report of R. aeschlimannii found in Hy. marginatum in Tibet. The results of this study indicated that yaks in Tibet are exposed to R. aeschlimannii.
... Diet plays a chief role in articulating the chemostatic niche of the gut. Recently, various processes undertaken during food processing and preservation have manifested microbiome alteration imparting harmful health consequences, particularly in elderly population (Cotillard et al. 2013;Keenan et al. 2015). Alterations in physiological parameters like appetite reduction, masticatory abnormalities, and constipation have been associated with advancing age (Candela et al. 2014). ...
Chapter
Globally, the significance of gut microbial composition in alleviating several intestinal complications has gained tremendous clinical consideration. Increasing lines support the evidence that, healthy lipids such as omega fatty acids and short-chain fatty acids play a pivotal role in orchestrating the composition and shaping the host microbial population in the gastrointestinal system by producing protective derivatives, restoring intestinal gut barrier, modulating gut microbial community, nervous, and immune system, attenuating inflammation, insulin resistance, and fat accumulation. In the above context, the present chapter uncovers the recent advancements carried out in the recent past on the impact of long- and short-chain fatty acids on gut microbiome–brain communication.
... Clinical findings Some human studies have reported a possible link between gut microbiota and fasting plasma lipids in individuals with obesity, such as higher triglyceride levels in those with a low bacterial gene count, indicative of dysbiosis of the gut microbiota [77,78]; however, limited data are available on postprandial lipids. In a study of individuals with the metabolic syndrome, a diet and lifestyle intervention followed for 14 days induced significant modifications of gut microbiota, in parallel with significant reductions in plasma triglyceride levels and chylomicrons [79]. ...
Article
Full-text available
Postprandial hyperlipidaemia is an important feature of diabetic dyslipidaemia and plays an important role in the development of cardiovascular disease in individuals with type 2 diabetes. Postprandial hyperlipidaemia in type 2 diabetes is secondary to increased chylomicron production by the enterocytes and delayed catabolism of chylomicrons and chylomicron remnants. Insulin and some intestinal hormones (e.g. glucagon-like peptide-1 [GLP-1]) influence intestinal lipid metabolism. In individuals with type 2 diabetes, insulin resistance and possibly reduced GLP-1 secretion are involved in the pathophysiology of postprandial hyperlipidaemia. Several factors are involved in the overproduction of chylomicrons: (1) increased expression of microsomal triglyceride transfer protein, which is a key enzyme in chylomicron synthesis; (2) higher stability and availability of apolipoprotein B-48; and (3) increased de novo lipogenesis. Individuals with type 2 diabetes present with disorders of cholesterol metabolism in the enterocytes with reduced absorption and increased synthesis. The increased production of chylomicrons in type 2 diabetes is also associated with a reduction in their catabolism, mostly because of a reduction in activity of lipoprotein lipase. Modification of the microbiota, which is observed in type 2 diabetes, may also generate disorders of intestinal lipid metabolism, but human data remain limited. Some glucose-lowering treatments significantly influence intestinal lipid absorption and transport. Postprandial hyperlipidaemia is reduced by metformin, pioglitazone, alpha-glucosidase inhibitors, dipeptidyl peptidase 4 inhibitors and GLP-1 agonists. The most pronounced effect is observed with GLP-1 agonists, which reduce chylomicron production significantly in individuals with type 2 diabetes and have a direct effect on the intestine by reducing the expression of genes involved in intestinal lipoprotein metabolism. The effect of sodium–glucose cotransporter 2 inhibitors on intestinal lipid metabolism needs to be clarified. Graphical abstract
... Reduced diversity of microbiota was one of the main types of intestinal disease related to ecological imbalance (Requena et al., 2018) and was documented in many diseases, such as inflammatory bowel disease (IBD) (Frank et al., 2007), autoimmune hepatitis (Wei et al., 2020), and type 1 diabetes (Kostic et al., 2015). Researchers had shown that people with a reduction of microbial richness are more likely to develop chronic low-grade inflammation (Le Cotillard et al., 2013). Requena et al. (2018) once noticed that the microbiome diversity had close association with our health, and our result showed that the gut microbiota status has transformed significantly from MS to the development of MS with elevated GGT. ...
Article
Full-text available
It is predicted that by 2035, metabolic syndrome (MS) will be found in nearly more than half of our adult population, seriously affecting the health of our body. MS is usually accompanied by the occurrence of abnormal liver enzymes, such as elevated gamma-glutamyl transpeptidase (GGT). More and more studies have shown that the gut microbiota is involved in MS; however, the correlation between gut microbiota and MS with elevated GGT has not been studied comprehensively. Especially, there are few reports about its role in the physical examination of the population of men with MS and elevated GGT. By using the whole-genome shotgun sequencing technology, we conducted a genome-wide association study of the gut microbiome in 66 participants diagnosed as having MS accompanied by high levels of GGT (case group) and 66 participants with only MS and normal GGT level (control group). We found that the number of gut microbial species was reduced in participants in the case group compared to that of the control group. The overall microbial composition between the two groups is of significant difference. The gut microbiota in the case group is characterized by increased levels of “harmful bacteria” such as Megamonas hypermegale, Megamonas funiformis, Megamonas unclassified, Klebsiella pneumoniae, and Fusobacterium mortiferum and decreased levels of “beneficial bacteria” such as Faecalibacterium prausnitzii, Eubacterium eligens, Bifidobacterium longum, Bifidobacterium pseudocatenulatum, Bacteroides dorei, and Alistipes putredinis. Moreover, the pathways of POLYAMSYN-PWY, ARG+POLYAMINE-SYN, PWY-6305, and GOLPDLCAT-PWY were also increased in the case group, which may play a role in the elevation of GGT by producing amine, polyamine, putrescine, and endogenous alcohol. Taken together, there are apparent changes in the composition of the gut microbiome in men with MS and abnormal GGT levels, and it is high time to discover specific gut microbiome as a potential therapeutic target in that population. More in-depth studies of relevant mechanism could offer some new methods for the treatment of MS with elevated GGT.
... Specifically, metabolic inflammation represents the most important player in the way intestinal bacteria have to modulate energy homeostasis. A 40% reduction in microbiome diversity was associated with low-grade systemic inflammation and adipokines production, mainly mediated by LPS production [63]. Similarly, other bacterial products may elicit inflammation and insulin resistance, activating the toll-like receptor (TLR) pathway [64]. ...
Article
Full-text available
Non-alcoholic fatty liver disease (NAFLD) represents an increasing cause of liver disease, affecting one-third of the population worldwide. Despite many medications being in the pipeline to treat the condition, there is still no pharmaceutical agent licensed to treat the disease. As intestinal bacteria play a crucial role in the pathogenesis and progression of liver damage in patients with NAFLD, it has been suggested that manipulating the microbiome may represent a therapeutical option. In this review, we summarise the latest evidence supporting the manipulation of the intestinal microbiome as a potential therapy for treating liver disease in patients with NAFLD.
Chapter
The worldwide population of elderlies is rapidly accelerating and the health status of this elderly population warrants an immediate attention. Age-associated gut microbiota modulations are linked to immunosenescence and inflammaging which perturbs the gut microbiotal ecosystem. Several studies have correlated the human gut microbiota with the aging process and found that various factors are associated with variation in the gut microbial composition during aging. Among them, the crucial role players are diet, geographical origin, and intervention of pre/probiotics. To conclude with, the gut microbiota can be a potential target contributing toward the health status of the elderly population. In this chapter, we have attempted to highlight the updates regarding the possible factors associated with changes in GM composition during the aging process.
Chapter
A healthy diet provides the fluid, macronutrients, and adequate calories that help to maintain or improve overall health. Our gut harboring a variety of microbiota-a complex network of bacteria, viruses, protozoans, and fungi to sustain a symbiotic association with host body and regulate the immune system, metabolism, and gut physiology. Apart from normal physiology microbiota appears to contribute to behavioral and stress responses. Any modification in composition or absence of any specific species in gut microbiota could lead to the inflammatory response that causes or promotes cardiovascular, renal, neurodegenerative diseases, and many cancers. Alteration in gut microbiota is probably caused by lifestyle, diet, antibiotic treatment, environmental stress, and psychological stress. Growing evidence suggests that diet is a prominent factor in promoting the gut microbiota dysbiosis. In this chapter, we have focused on the critical links among gut microbiota, diet, and inflammation leading to neurodegenerative diseases. We discussed gut microbiome-based studies which include dysbiosis or absence of specific microorganisms induced neurodisorders, introduction of healthy diet-induced modification of gut microbiota to prevent neurological diseases.
Chapter
The immune system which evolved to tackle pathogenic microbes also tolerates the commensal microflora and they in turn help in shaping the immune system. Selection, tolerance, and proper compartmentalization of microflora are strictly maintained by the host immune system. The mechanisms involved in selection and maintenance of microflora seem to be dependent on genetics as well as environmental factors such as diet and antibiotic use. Such environmental treatments may disrupt the microflora, consequently causing imbalances in the host immune system. These microflora dysregulations and immune system imbalances can leave the host susceptible to infectious and noninfectious diseases. Microflora is, therefore, critical in protecting the host from various diseases. To avoid disease, the immune system is actively involved in maintenance of proper microflora. The highly regulated interplay between the microflora and the innate and adaptive immune system components, which is so crucial to host health, will be discussed in this chapter.
Conference Paper
Full-text available
The study of complex microbial ecosystems by a quantitative metagenomic approach has been made possible by advancements in high-throughput sequencing technologies. Quantitative metagenomics relies on deep sequencing to construct an ecosystem profile using gene and genome counts. Next generation sequencing (NGS) technologies such as SOLiD or Illumina produce millions of short sequences (35 to 75bp) which can be used as tags to establish gene profiles. This approach requires the use of a specific reference catalog which should be composed of genes present in the ecosystem of interest. The use of classical bioinformatic methods for the analysis of such large amounts of data is not feasible as we overpass the expected dataset size of common tools. We have therefore developed an integrated metagenomic analysis pipeline, METEOR, which includes the indexing of short reads to genomic objects. Data are indexed in an embedded database around the iMOMi framework [1] and organized in a dedicated file system. This optimization facilitates secondary analysis including gene/species abundance evaluation, cross-sample comparison, ecosystem metabolism reconstruction or gene/species diversity analysis. The METEOR pipeline has been implemented for several metagenomic projects such as MicroObes for characterizing the human intestinal microbiome of obese individuals following a restrictive diet or FoodMicrobiomes for studying the ecosystem of fermented food like French traditional cheeses. In MicroObes, we investigate the changes of gut microbiota in a human model of weight loss induced by restrictive diet in moderately obese subjects. DNA isolated from 195 faecal samples of 49 obese subjects collected at different date (start of the study, 6 weeks after a restrictive diet and 12 weeks) have been sequenced with SOLiD technology yielding about 300 gigabases. Short reads have been indexed against the 3.3 millions genes of the human gut microbial gene catalog (MetaHIT consortium [2]). Statistical analysis of the gene profiles generated indicate significant variations in gene and genome frequencies during the first 6 weeks of dieting and a subsequent stabilization after 12 weeks according to the observed success of patients dietary.
Article
Full-text available
We are facing a global metabolic health crisis provoked by an obesity epidemic. Here we report the human gut microbial composition in a population sample of 123 non-obese and 169 obese Danish individuals. We find two groups of individuals that differ by the number of gut microbial genes and thus gut bacterial richness. They contain known and previously unknown bacterial species at different proportions; individuals with a low bacterial richness (23% of the population) are characterized by more marked overall adiposity, insulin resistance and dyslipidaemia and a more pronounced inflammatory phenotype when compared with high bacterial richness individuals. The obese individuals among the lower bacterial richness group also gain more weight over time. Only a few bacterial species are sufficient to distinguish between individuals with high and low bacterial richness, and even between lean and obese participants. Our classifications based on variation in the gut microbiome identify subsets of individuals in the general white adult population who may be at increased risk of progressing to adiposity-associated co-morbidities.
Article
Full-text available
Alterations in intestinal microbiota composition are associated with several chronic conditions, including obesity and inflammatory diseases. The microbiota of older people displays greater inter-individual variation than that of younger adults. Here we show that the faecal microbiota composition from 178 elderly subjects formed groups, correlating with residence location in the community, day-hospital, rehabilitation or in long-term residential care. However, clustering of subjects by diet separated them by the same residence location and microbiota groupings. The separation of microbiota composition significantly correlated with measures of frailty, co-morbidity, nutritional status, markers of inflammation and with metabolites in faecal water. The individual microbiota of people in long-stay care was significantly less diverse than that of community dwellers. Loss of community-associated microbiota correlated with increased frailty. Collectively, the data support a relationship between diet, microbiota and health status, and indicate a role for diet-driven microbiota alterations in varying rates of health decline upon ageing.
Article
To understand the impact of gut microbes on human health and well-being it is crucial to assess their genetic potential. Here we describe the Illumina-based metagenomic sequencing, assembly and characterization of 3.3 million non-redundant microbial genes, derived from 576.7 gigabases of sequence, from faecal samples of 124 European individuals. The gene set, approximately 150 times larger than the human gene complement, contains an overwhelming majority of the prevalent (more frequent) microbial genes of the cohort and probably includes a large proportion of the prevalent human intestinal microbial genes. The genes are largely shared among individuals of the cohort. Over 99% of the genes are bacterial, indicating that the entire cohort harbours between 1,000 and 1,150 prevalent bacterial species and each individual at least 160 such species, which are also largely shared. We define and describe the minimal gut metagenome and the minimal gut bacterial genome in terms of functions present in all individuals and most bacteria, respectively.
Article
Bell System Technical Journal, also pp. 623-656 (October)