Review nature publishing group

Understanding the role of gut microbiome in metabolic

disease risk
Yolanda Sanz1, Marta Olivares1, Ángela Moya-Pérez1 and Carlo Agostoni2

The gut microbiota structure, dynamics, and function result fact that gut microbiota shows certain plasticity, particularly in
from interactions with environmental and host factors, which response to the diet, also makes it possible to develop interven-
jointly influence the communication between the gut and tion strategies that promote a healthy gut ecosystem to reduce
peripheral tissues, thereby contributing to health program- disease risk (3). Breast milk-associated bifidobacteria in
ming and disease risk. Incidence of both type-1 and type-2 infants are a hallmark of the host–gut microbiome response to
diabetes has increased during the past decades, suggesting diet, promoting a particular microbial community structure,
that there have been changes in the interactions between and presumably contributing to the role of breastfeeding in
predisposing genetic and environmental factors. Animal stud- reducing disease risk (e.g., infections, obesity, and type-2 dia-
ies show that gut microbiota and its genome (microbiome) betes) (4). Although microbiota seems to be more sensitive to
influence alterations in energy balance (increased energy environmental factors in infants than in adults, the latter also
harvest) and immunity (inflammation and autoimmunity), respond to dietary intervention with variations of at least 10%
leading to metabolic dysfunction (e.g., insulin resistance and (2). There is also evidence that gut microbiota and their prod-
deficiency). Thus, although they have different origins, both ucts have large and diverse effects on immunity and immune-
disorders are linked by the association of the gut microbiota mediated disorders (5). Furthermore, bacterial interactions
with the immune–metabolic axis. Human studies have also (mutualism and antagonism) that define the gut ecosystem are
revealed shifts in microbiome signatures in diseased subjects known to be indirectly mediated by the host immune system
as compared with controls, and a few of them precede the and particularly by its innate components (6,7). These findings
development of these disorders. These studies contribute to reveal the two-way communication between the gut microbi-
pinpointing specific microbiome components and functions ota and the host, which might be shifted by diet and influence
(e.g., butyrate-producing bacteria) that can protect against the risk of developing immune-mediated disorders, particu-
both disorders. These could exert protective roles by strength- larly at early stages of development.
ening gut barrier function and regulating inflammation, as In the light of this evidence, the increased incidence of diet-
alterations in these are a pathophysiological feature of both associated inflammatory diseases (obesity and type-2 diabe-
disorders, constituting common targets for future preventive tes) and autoimmune disorders (e.g., type-1 diabetes) could be
approaches. partially explained by changes in gene–environment interac-
tions, including diet-induced changes in microbiota. Systemic
inflammation and autoimmunity are detectable years prior to

T he bacterial ecosystem (microbiota) living in our intestine

plays a fundamental role in the normal functioning of both
metabolic and immune systems, beyond genetic determinants.
the onset of overt disease, promoting the development of met-
abolic and autoimmune diseases (8). This suggests that sub-
jects predisposed to disease might benefit from interventions
Consequently, imbalances in gut microbiota (dysbiosis) are a targeting the immune system directly or indirectly before the
possible causal factor of metabolic and autoimmune diseases. disease manifests. Thus, new biomarkers of disease progres-
The initial colonization of the newborn intestine is particu- sion, including those related to the microbiome, must be iden-
larly relevant to the proper development of the host’s immune tified to discover the pathophysiological determinants of these
and metabolic functions and to determine disease risk in early disorders and develop preventive interventions in the form of
and later life (1). Gut microbiota signatures seem to be highly personalized health-care and nutrition-based strategies.
specific for each individual, resulting in large interindividual Here, we will consider two diseases, type-1 and type-2
variations that depend on both host genetics and environmen- diabetes, as both disorders are linked to the gut microbiota
tal factors (1,2). This specificity has hindered our understand- association with the immune–metabolic axis (Figure 1). We
ing of the microbiota’s role in health and disease. However, the will provide a comprehensive review of the extent to which

1
Microbial Ecology, Nutrition & Health Research Group, Institute of Agrochemistry and Food Technology, National Research Council (IATA-CSIC), Valencia, Spain; 2Pediatric
Clinic, Department of Clinical Sciences and Community Health, University of Milan, Fondazione IRCCS Ca’ Granda-Ospedale Maggiore Policlinico, Milan, Italy.
Correspondence: Yolanda Sanz ([email protected])
Received 3 June 2014; accepted 2 October 2014; advance online publication 12 November 2014. doi:10.1038/pr.2014.170

236 Pediatric Research      Volume 77 | Number 1 | January 2015 Copyright © 2015 International Pediatric Research Foundation, Inc.

Licensed to Natalia Sabrina Ortiz - [email protected]

 Gut microbiome and metabolic disease Review
Diet and other environmental factors

Immunity Microbiota Metabolism

Dysbiosis

Epithelial cell Tight junctions

activation

Leaky gut
HLA-DQ

LPS Dendritic cell

T cell

Autoreactive T cells

Cytokine release

B cell
Inflammatory
response Autoantibodies

Type 2 diabetes Type 1 diabetes

Figure 1. Schematic representation of the interaction of the gut microbiota with the immune–metabolic axis and the different mechanisms proposed to
explain its implication in health and disease risk. In type-1 diabetes, the intestinal dysbiosis and the increased gut permeability and altered immunoregu-
latory mechanisms seem to trigger the autoimmune response leading to the destruction of β cells in the pancreatic islets. In type-2 diabetes, saturated fat
and dysbiosis due to “obesogenic” diets cause inflammation and alterations in gut permeability contributing to disease onset.

available evidence supports a role for gut microbiota compo- regulation (9). Nevertheless, gut microbiota alterations associ-
sition and function in disease risk. We will also identify the ated with obesity were also proposed to be a secondary con-
gaps in the understanding of the respective roles of gut micro- sequence of the diets inducing obesity or of the genotype (6).
biota, dietary habits, and host factors in disease progression However, comparison of germ-free and conventionally colo-
and prevention. nized mice has demonstrated that gut microbiota colonization
leads to impaired glucose metabolism and increased macro-
GUT MICROBIOTA AND RISK OF DEVELOPING OBESITY phage accumulation in white adipose tissue not only in mice
AND TYPE-2 DIABETES fed a high-fat diet but also fed a standard diet, suggesting that
Obesity results from a positive imbalance between energy microbiota effects are partly independent of diet (10). Also,
intake and expenditure and is also associated with low-grade animals with a common genotype and fed the same diet can
inflammation leading to chronic metabolic disease (type-2 develop different metabolic phenotypes (either diabetic or
diabetes). The steady increase in obesity prevalence has mainly nondiabetic) as a function of their specific gut microbiota pro-
been attributed to socio-economic factors, dietary changes, file, suggesting that gut microbiota per se determines the risk of
and sedentary lifestyle. Nevertheless, recent evidence suggests developing metabolic dysfunction to some extent (11).
that our gut microbiota and microbiome also contribute to the Figure 2 provides a summary of evidence for the possible
host genetic make-up and to the whole metabolic activity of mode of action of the microbiota in obesity and the associ-
the body and immune function, and therefore, could play a ated comorbidities, including type-2 diabetes. Initially, com-
role in these disorders. parisons between germ-free and conventional colonized mice
indicated that the microbiota, as a whole, increases our ability
Mechanisms Behind the Role of Gut Microbiota in Obesity and to extract energy from the diet and store this energy in periph-
Type-2 Diabetes in Animals eral tissues (liver, adipose tissue, etc.), leading to impairment
Alterations in gut microbiota structure and function detected of insulin sensitivity (12). The fact the microbiota breaks down
in both genetically and diet-induced obesity models led to indigestible dietary components (mainly plant-derived poly-
initially establishing a role of microbiota in body-weight saccharides) is considered one of the mechanisms improving

Copyright © 2015 International Pediatric Research Foundation, Inc. Volume 77 | Number 1 | January 2015      Pediatric Research  237

Licensed to Natalia Sabrina Ortiz - [email protected]

 Review Sanz et al.

DIET
(fiber vs. fat)

Eubiosis/symbiosis Dysbiosis

SCFAs
LPS Cytokine H2S cytotoxic
(extra energy)
production
Butyrate Propionate/acetate

PPARγ M cell
TLR 4/2

NF-kB Gpr41/43
GLP-2

Bacterial components
Inhibitor of
HDAC 6/9 Inflammatory mediators
LPS
(TNFα, IFNγ, IL6, IL18, MCP1, etc.)
DNA
iNOS, COX2
Treg
Portal vein Macrophages
Immune
homeostasis
Dendritic cells
PYY GLP-1

Inflammation
Gut Satiety
Insulin resistance
motility Insulin
sensitivity Steatosis Larger adipocytes

Figure 2. Main mechanisms of action of gut microbiota and derived metabolites in obesity and associated metabolic dysfunctions (insulin resistance and
type-2 diabetes). Gut microbiota contributes to the hydrolysis of complex polysaccharides from dietary fiber, and thereby might contribute to increasing
energy harvest and to generating short-chain fatty acids (SCFAs; acetic propionic and butyric acid) that affect the host’s metabolism in different ways. SCFAs
might activate the G-protein coupled receptor (Gpr) 41 inducing the expression of peptide YY, an intestinal hormone that influences gut motility, increases
intestinal transit rate, and reduces energy harvest from the diet. SCFAs might also activate Gpr43 and Gpr41 inducing glucagon-like peptide-1 (GLP-1)
secretion, increasing insulin sensitivity, and inducing satiety. Butyrate provides energy to enterocytes, exerting a trophic effect and inducing the synthesis
of GLP-2, thereby strengthening the gut barrier function. Butyrate may also promote the formation of peripheral regulatory T cells (Treg) by its ability to
inhibit the histone deacetylases 6 and 9 that leads to acetylation of histone H3, which promotes the expression of the Treg-specific forkhead transcription
factor FoxP3. Intestinal dysbiosis, which could be partly caused by “obesogenic” diets rich in saturated fats, may lead to the growth of potential pathogens
(Gram-negative bacteria and derived lipopolysaccharide (LPS)) with proinflammatory effects via generation of cytotoxic compounds (H2S) or interaction
with innate immune receptors (TLR4, TLR2) and contribute to inflammatory cytokine production, attraction of inflammatory cells, and translocation of
bacterial products (LPS, DNA) by transcellular and paracellular pathways that activate inflammation in peripheral tissues. NF, nuclear factor.

the host’s ability to extract energy from the diet. This is con- exert beneficial effects in the context of obesity and associated
sistent with evidence that the microbiome of human feces is comorbidities. One of the proposed modes of action of SCFAs
enriched in genes involved in the utilization of complex dietary (specially of butyrate) is related to their ability to increase sati-
polysaccharides (13). This activity leads to the generation of ety and decrease calorie intake and postprandial glycemia via
short-chain fatty acids (SCFAs; butyric, acetic, and propionic) modification of gut peptide production (glucagon-like peptide
and gases (e.g., hydrogen), which can also be further metabo- (GLP)-1 and gastric inhibitory peptide). In addition, butyrate
lized, activating the overall colonic fermentation and increas- is the main energy source for enterocytes, and therefore, regu-
ing the efficiency of energy extraction from ingested nutrients lates cell proliferation and differentiation and induces GLP-2
(14). Although the energy contribution of complex plant production, which altogether strengthens the gut barrier func-
polysaccharides is by far lower than that produced by digest- tion. Butyrate also reduces oxidative damage and inflamma-
ible carbohydrates, their metabolism can constitute a survival tion by inhibiting histone deacetylases and the activation of
strategy mediated by our symbiotic microbiota. However, the the transcription factor nuclear factor-κB and the associated
role of the gut microbiota in supplying extra energy from the cytokine production (15,16; Figure 2).
diet could be less prominent in the context of Western diets Inflammation seems to be one of the major pathophysio-
rich in fats and simple carbohydrates. This theory of “increased logical factors leading to insulin resistance and progressively
energy harvest” is somehow contradictory to the benefits on to type-2 diabetes. Gut microbiota alterations can also con-
metabolic health attributed to high dietary fiber intake and the tribute to this inflammatory condition. Obesogenic diets may
SCFAs generated. Some of the SCFAs resulting from intake of promote the growth of pathobionts (potential pathogens),
high fiber diets, and particularly butyrate, are considered to which could trigger an inflammatory response via local

238 Pediatric Research      Volume 77 | Number 1 | January 2015 Copyright © 2015 International Pediatric Research Foundation, Inc.

Licensed to Natalia Sabrina Ortiz - [email protected]

 Gut microbiome and metabolic disease Review
activation of innate immune receptors (e.g., TLR4, TLR2) Proteobacteria or enterobacteria were positively associated
with production of proinflammatory cytokines (17,18) or with obesity in a few studies, including one in preschool chil-
via production of toxic compounds (e.g., hydrogen sulphide) dren, as well as Staphylococcus spp. in pregnant women (22–24).
(19). Both mechanisms could also lead to a leaky gut, translo- By contrast, Akkermansia spp. was associated with a lean phe-
cation of microbial molecules, and overall, promote systemic notype in two human studies, one conducted in children (22).
inflammation. Also, saturated fatty acids of obesogenic diets Diet is a major driver of gut microbiota composition, which
might favor a leaky gut by inducing expression and activa- may partly explain why microbiota alterations are associated
tion of innate immune receptors (Toll-like receptors) and with an obese phenotype in humans. However, the micro-
increasing lymphocyte flux and proliferation; altogether this biota structure resulting from unhealthy dietary habits is
may lead to the production of cytokines (e.g., tumor necro- also thought to contribute to body-weight regulation (25).
sis factor-alpha, interferon-gamma) that increase intestinal In this context, higher energy intake leading to an increased
permeability, facilitating translocation of bacterial prod- ratio of Firmicutes to Bacteroidetes has also been related to
ucts regardless of possible alterations in the gut microbiota increased energy absorption from diet (25). Human interven-
composition. Lipopolysaccharide can also translocate via a tion studies have also revealed that the individual’s microbiota
transcellular epithelial pathway together with chylomicrons composition and richness could influence the effectiveness of
formed to incorporate dietary long-chain fatty acids in the dietary interventions in improving the metabolic phenotype
form of triglycerides or through intestinal-epithelial micro- (26,27). Furthermore, transplantation experiments of human
fold (M) cells (9). Murine models of high-fat diet-induced microbiota in mouse models reveal that the influence of diet–
obesity have also demonstrated that live Gram-negative com- microbiota interactions on the host metabolic phenotype is
mensal intestinal bacteria (Escherichia coli) can translocate transmissible. A recent study has shown that when micro-
to the blood and adipose tissue and that this process is medi- biota from subjects discordant for obesity (uncultured fecal
ated by dendritic cells and depends on innate immunity pat- communities and the corresponding fecal bacterial cultures)
tern-recognition receptors (TLR4 and Nod1) (20). However, is transplanted in germ-free mice, these mice develop the
intervention in the gut ecosystem with specific prebiotics or corresponding phenotype when fed a low in fat and high in
potential probiotics reduces the high-fat diet-induced meta- plant polysaccharides chow diet (28). The same study showed
bolic endotoxemia, inflammatory tone, bacterial product cohousing mice harboring the obese human microbiota with
translocation, and metabolic dysfunction, demonstrating mice containing the lean human microbiota prevented the
that these effects are partly mediated by gut microbiota- development of an obesity-associated metabolic phenotype
induced changes in animal models (20,21). due to the invasion of the obese mouse intestine with specific
members of Bacteroidetes (e.g., Bacteroides uniformis, B. cac-
Evidence Supporting a Role of Gut Microbiota in Obesity in cae, B. cellulosilyticus, etc.) from the lean microbiota. However,
Humans this effect was diet-dependent and positive only in the context
In humans, a large number of studies have established rela- of a chow diet representing low saturated fat, high fruit, and
tionships between alterations in gut microbiota structure vegetable intake, but not in the context of a chow diet, rep-
and function, and obesity, although there is no consensus on resenting high saturated fat, low fruit, and vegetable intake
the key players in this disorder yet. Numerous studies report (28). These findings were also supported by administration of
reduced proportions of Bacteroidetes or its subgroups (e.g., a selected Bacteroides strain (B. uniformis CECT 7771) from
Bacteroides) in obese subjects compared with lean subjects, humans to a mouse model of diet-induced obesity in an inde-
parallel to increased proportions of Firmicutes or its sub- pendent study (29). Furthermore, recent fecal transplants to
groups (22,23). However, other studies have not found con- type-2 diabetic subjects beneficially influenced glucose metab-
sistent results (22,24). These include a recent metagenomic olism and insulin resistance, proving that microbiota replace-
study reporting that subjects with low bacterial richness (low ment strategies could help to protect from metabolic disease
gene count) gained more weight and had increased inflam- in humans (30).
matory tone (C reactive protein and leptin), insulin resistance Despite all this evidence, prospective epidemiological stud-
and dyslipidemia as compared with subject with high bacte- ies are necessary to establish whether specific microbiota
rial gene counts (24). This finding supports the idea that a less features constitute risk factors and predict obesity and the
diverse microbiota is less resilient to the invasion of unhealthy associated metabolic disorders; however, only a couple of stud-
microbes, which may contribute to disease. Bacteroides ies have been published to date. A follow-up study including
and Ruminococcus spp. were more abundant in subjects 49 infants has reported that differences in fecal microbiota
with low gene counts and metabolic dysfunction, whereas composition at 6 and 12 mo of age precede subsequent over-
Faecalibacterium prausnitzii, Bifidobacterium, Lactobacillus, weight in children at 7 y of age. Children maintaining normal
Alistipes, Akkermansia, among others, were more dominant in weight showed a greater number of bifidobacteria, while chil-
subjects with high gene counts and healthier metabolic phe- dren who became overweight harbored a greater number of
notypes. Nonetheless, not all subjects with either a high or low Staphylococcus aureus during infancy (31). A more recent and
gene counts have a similar metabolic phenotype, suggesting larger study of 330 healthy Danish infants from 9 to 36 mo of
that findings could be influenced by confounding factors. age reported a positive correlation between the increase in BMI

Copyright © 2015 International Pediatric Research Foundation, Inc. Volume 77 | Number 1 | January 2015      Pediatric Research  239

Licensed to Natalia Sabrina Ortiz - [email protected]

 Review Sanz et al.

and the increase in SCFA-producing clostridia (the Clostridum Evidence Supporting the Role of Gut Microbiota in Type-2
leptum group and Eubacterium hallii) (32). Taking into con- Diabetes in Humans
sideration the positive associations established between rapid Studies reporting associations between shifts in gut microbiota
infant weight gain and later-life obesity, these microbiological composition and function and type-2 diabetes in humans are
differences could play a role in obesity; however, direct evi- summarized in Table 1. Some of the findings suggest that meta-
dence must be established. Santacruz et al. (33) detected posi- bolic markers of disease could be relevant for defining the rela-
tive correlations between mother’s E. coli numbers and birth tionships between obesity and gut microbiota (35). A recent
body weight, but data for the effects on offspring were not metagenomic study also led to the development of a mathemati-
followed-up. Another study concluded that microbiota in 79 cal model to identify metagenomic markers for type-2 diabetes
infants born large for gestational age differed significantly com- and diabetes-like metabolism (38). Nevertheless, the discrimi-
pared with microbiota in infants born with an average weight nant metagenomic markers were shown to differ between dif-
for gestational age. Gram-negative Proteobacteria were more ferent cohorts of subjects, revealing that our understanding
abundant in neonates born large for gestational age, whereas is insufficient to enable prediction of an individual’s disease
Gram-positive Firmicutes were more abundant in neonates risk based solely on the gut microbiome. Notwithstanding,
born average for gestational age (34). Due to the fact that high a decreased abundance of some butyrate-producing bacteria
birth weight is a risk factor for the development of metabolic (e.g., Roseburia, F. prausnitzii) and an increased abundance
disorders, these findings suggest that early microbiota could be of opportunistic pathogens (e.g., Clostridium clostridioforme,
a factor programming immune and metabolic health, but data E. coli) could be a potential microbiome signature of reduced
from follow-up studies were not reported. glucose tolerance and type-2 diabetes. However, prospective

Table 1. Changes in the intestinal microbiota associated with type-2 diabetes in human
Microbiota changea
Study group Methodology Phylum or class Genus or species Reference
Diabetics vs. nondiabetics Real-time PCR ↑Betaproteobacteria 35
(n = 18 vs. n = 18)
↓Firmicutes (Clostridia)
Diabetics vs. nondiabetics Real-time PCR ↓Bifidobacterium 36
(n = 16 vs. n = 12)
↓Bacteroides vulgatus
Diabetics vs. nondiabetic Shotgun sequencing ↑Bacteroides caccae
(n = 71 vs. n = 74)
↑Clostridium hathewayi
↑Clostridium ramosum
↑Clostridium symbiosum
↑Eggerthella lenta
↑Escherichia coli 37
↑Akkermansia muciniphila
↑Desulfovibrio
↓Clostridiales sp. SS3/4
↓Eubacterium rectal
↓Faecalibacterium prausnitzii
↓Roseburia intestinalis
↓Roseburia inulinivorans
Diabetics vs. nondiabetics Shotgun sequencing ↑Lactobacillus spp. 38
(n = 53 vs. n = 43)
↓Clostridium spp.
↑Clostridium clostridioforme
↓Roseburia
Prediabetics vs. nondiabetics 16S rDNA sequencing ↓Akkermansia muciniphila 39
(n = 64 vs. n = 44) ↓Faecalibacterium prausnitzii
Diabetics vs. prediabetics 16S rDNA sequencing ↓Bacteroides 39
(n = 13 vs. n = 64)
a
Arrows indicate increases or decreases of each bacterial group in the disease subject group compared with the control group (either nondiabetic or prediabetic).

240 Pediatric Research      Volume 77 | Number 1 | January 2015 Copyright © 2015 International Pediatric Research Foundation, Inc.

Licensed to Natalia Sabrina Ortiz - [email protected]

 Gut microbiome and metabolic disease Review
epidemiological studies in well-controlled populations are protected from developing type-1 diabetes (5). This protec-
needed to provide sounder evidence of the microbiome fea- tion was lost when Myd88−/− NOD mice were housed under
tures that constitute risk factors for these disorders, as well as germ-free conditions, suggesting that the microbiota exerted
dietary intervention studies to establish causality. a protective role. The absence of MyD88 in NOD mice led to
In relation to the mechanisms of action, it has been pro- changes in microbiota composition, characterized by increases
posed that the microbiota could contribute to triggering the in bacteria of the phylum Bacteroidetes, which could suppress
chronic-low grade inflammation underlying insulin resis- the development of diabetes, presumably through the produc-
tance and type-2 diabetes. This hypothesis is supported by the tion of an immunomodulatory compound (5). A more recent
detection of lipopolysaccharide from Gram-negative bacteria study demonstrated that the protective functions of the micro-
in the blood of subjects with metabolic syndrome and type-2 biota against type-1 diabetes development could be transferred
diabetes (40). Moreover, bacterial DNA (mostly belonging to (7). Gut bacteria from MyD88-deficient mice, administered
the phylum Proteobacteria) was detected in the blood of sub- over a 3-wk period, stably altered the composition of the gut
jects before diabetes onset, and was higher in those who had microbiome (increasing Lachnospiraceae and Clostridiaceae
abdominal adiposity (20). and decreasing Lactobacillaceae), reducing insulitis and delay-
ing the onset of diabetes. This affected the mucosal immune
GUT MICROBIOTA AND RISK OF DEVELOPING TYPE-1 system, increasing the concentration of IgA and transform-
DIABETES ing growth factor beta in the lumen and of CD8+CD103+ and
Type-1 diabetes is an autoimmune disease caused by destruc- CD8αβ T cells in the lamina propria of the large intestine,
tion of insulin-producing β cells in the pancreatic islets of possibly delaying the development of autoimmune diabetes.
Langerhans by immune mediated mechanisms that lead to Administration of the probiotic product VSL#3 (a combina-
insulin deficiency in genetically predisposed individuals (41). tion of strains of the genera Bifidobacterium, Lactobacillus,
Although the etiology of the disease is not completely under- and Streptococcus) to NOD mice also attenuated destructive
stood, increased intestinal permeability, aberrant immune insulitis, and preserved beta cells. This was presumably due to
responses, and intestinal dysbiosis have been proposed as the the induction of interleukin-10 producing lymphocytes that
“perfect storm” that triggers the development of type-1 diabe- recirculated from the gut to pancreatic islet (49). Recently, a
tes (42). Insufficient regulation of immune attacks on β cells, gluten-free diet has been demonstrated to reduce the diabetic
due to genetic and other modifiable factors (e.g., microbiota) is outcomes in NOD mice due to changes in the mouse micro-
characteristic of disease (43). Increased intestinal permeability biome (increased Bacteroides and Akkermansia numbers), via
has also been described in type-1 diabetes patients (44–46), stimulation of a higher percentage of CD4+CD25+Foxp3 regu-
their relatives (45), and type-1 prediabetic patients (45,46). latory cells (50).
In this context, it is unclear whether the altered immune sta-
tus causes intestinal integrity disruption or whether intestinal Evidence Supporting a Role of Gut Microbiota in Type-1
integrity disruption by an environment trigger (e.g., infections Diabetes in Humans
or dysbiosis) causes an abnormal immune response. Possibly Studies reporting associations between shifts in gut microbiota
both processes are involved. composition and function and type-1 diabetes in humans are
summarized in Table 2. Although results on the gut micro-
Mechanisms Behind the Role of Gut Microbiota in Type-1 biota components and functions contributing to type-1 dia-
Diabetes in Animals betes are not fully consistent, butyrate producing bacteria and
Animal studies support the involvement of intestinal micro- bifidobacteria seem to be protective, whereas Proteobacteria/
biota in the early events that precede and lead to type-1 diabe- enterobacteria seem to constitute risk factors for disease. From
tes, and some suggest that interactions with the host immune a functional point of view, the protective role of the commen-
system mediate the effects of microbiota in this disorder. Some sal intestinal microbiota against this disorder could be medi-
studies described changes in microbiota composition preced- ated by the fact the microbiota and its metabolites (SCFAs) can
ing its onset in diabetic-prone BioBreeding rats (47). Others strengthen mucosal integrity, modulate gut hormone produc-
describe that germ-free mice have increased diabetes incidence tion improving glucose metabolism, and reduce inflammation
or that specific monocolonization of germ-free nonobese dia- and prevalence of potential proinflammatory bacteria (e.g.,
betic (NOD) mice delays the onset and reduces the incidence enterobacteria) (51,53).
of diabetes (48). A number of animal studies also demonstrate Other environmental factors with an impact on the gut
that exposure to bacterial antigens or infection in the early microbiota and immune system have also been linked to the
neonatal period prevents type-1 diabetes, thereby supporting risk of developing type-1 diabetes, including viral infections,
the notion that microbial immunostimulation may beneficially type of delivery (natural birth or caesarian), and infant feed-
affect the maturation of the postnatal immune system and pro- ing practices (42). A meta-analysis showed that birth by cae-
tect against type-1 diabetes (48). The role of the microbiota sarean section increased the risk of developing the disease by
and its interaction with the innate immune system in type-1 20% regardless of the gestational age, weight, maternal age,
diabetes was demonstrated in a study showing that NOD breastfeeding practices, and maternal diabetes (55). The lack
mice deficient in the innate signaling molecule MyD88 were of breastfeeding has also been associated with type-1 diabetes

Copyright © 2015 International Pediatric Research Foundation, Inc. Volume 77 | Number 1 | January 2015      Pediatric Research  241

Licensed to Natalia Sabrina Ortiz - [email protected]

 Review Sanz et al.

Table 2. Changes in the intestinal microbiota associated with type-1 diabetes (T1D) in humans
Microbiota changea
Study group Methodology Phylum or class Genus or species Reference
Autoimmune children (positive in at least two Shotgun ↑Actinobacteria ↑Bifidobacterium 51
autoantibodies) vs. healthy children (n = 4 vs. n = 4) sequencing
↑Bacteroidetes ↑Bacteroides
↑Proteobacteria ↑Lactobacillus
↓Firmicutes ↑Lactococcus
↓Fusobacteria ↑Streptococcus
↓Tenericutes ↑Veillonella
↓Verrucomicrobia ↑Alistipes
↓Prevotella
↓Akkermansia
↓Eubacterium
↓Fusobacterium
↓Anaerostipes
↓Roseburia
↓Subdoligranulum
↓Faecalibacterium
Autoimmune children (positive in at least two 16S rDNA ↑Bacteroidetes ↑Bacteroides 52
autoantibodies) vs. healthy children (n = 4 vs. n = 4) sequencing
↑Bacteroides ovatus
↓Firmicutes ↓Eubacterium
↓Faecalibacterium
↓Clostridia ↓Bacteroides vulgatus
↓ Bacteroides fragilis
Children with T1D vs. healthy children (n = 16 vs. n = 16) Real-time PCR ↑Bacteroidetes ↑Bacteroides 53
↑Veillonella
↑Clostridium
↓Firmicutes ↓Prevotella
↓Lactobacillus
↓Actinobacteria ↓Bifidobacterium
↓Blautia coccoides–
Eubacterium rectale
Children with T1D vs. healthy children (n = 35 vs. n = 35) Plate counting ↑Candida albicans 54
↑Enterobacteriaceae other
than Escherichia coli
↓Bifidobacterium
a
Arrows indicate increases or decreases of each bacterial group in the disease subject group compared with the control group (healthy subjects).

(reviewed in Pereira et al., 2014 (56)). Thus, it is plausible been described (1,60). However, the influence of the HLA-DQ
that differences in the acquisition and evolution of the new- genotype on the microbiota in subjects at risk of type-1 diabe-
born microbiota in caesarian vs. vaginally born infants, and tes has not been specifically studied, although they share some
in formula vs. breastfed infants, may also influence the risk genetic determinants of CD.
of developing type-1 diabetes (57). The association of these There are also speculations on the role of infections in the
factors with other autoimmune disorders such as celiac dis- risk of developing type-1 diabetes, although no clear conclu-
ease (CD) has been reported (58). Type-1 diabetes and CD sions can be drawn from existing human data. For instance,
are autoimmune disorders with common genetic determi- the reduction in the antibody levels against Helicobacter pylori,
nants (HLA-DQ) that frequently coexist. In fact, CD affects at cytomegalovirus, Epstein–Barr virus, and Toxoplasma in type-1
least 10% of diabetic patients at some point in their life (59). diabetic patients supports the “hygiene hypothesis” according
The influence of the HLA-DQ2/8 genotype on the intestinal to which reduced exposure to microbial stimulus increases the
microbiota composition of infants at family risk of CD has disease risk (61). Infections by two coxsackieviruses (B3 and

242 Pediatric Research      Volume 77 | Number 1 | January 2015 Copyright © 2015 International Pediatric Research Foundation, Inc.

Licensed to Natalia Sabrina Ortiz - [email protected]

 Gut microbiome and metabolic disease Review
B6) have also been linked to a reduced risk of type-1 diabetes, 2. Salonen A, Lahti L, Salojarvi J, et al. Impact of diet and individual variation
which was considered to be a result of immunological cross- on intestinal microbiota composition and fermentation products in obese
protection (62). By contrast, intestinal infections by rotavi- men. ISME J 2014 (doi:10.1038/ismej.2014.63).
3. Maslowski KM, Mackay CR. Diet, gut microbiota and immune responses.
rus (63) and other enterovirus (62,64) have been associated Nat Immunol 2011;12:5–9.
with the onset or with the risk of disease, suggesting that they 4. Sela DA, Mills DA. The marriage of nutrigenomics with the microbiome:
may contribute to triggering type-1 diabetes. In this vein, an the case of infant-associated bifidobacteria and milk. Am J Clin Nutr
observational study involving type-1 diabetics under 15 y of 2014;99:697S–703S.
age described that perinatal infections were associated with an 5. Wen L, Ley RE, Volchkov PY, et al. Innate immunity and intestinal micro-
biota in the development of Type 1 diabetes. Nature 2008;455:1109–13.
increased risk in the autoimmune disorder; conversely, atten- 6. Vijay-Kumar M, Aitken JD, Carvalho FA, et al. Metabolic syndrome
dance to preschool day-care, which promotes microbial expo- and altered gut microbiota in mice lacking Toll-like receptor 5. Science
sure, decreased type-1 diabetes risk (65). Overall, the findings 2010;328:228–31.
suggest that the type of microbial agent, as well as time of 7. Peng J, Narasimhan S, Marchesi JR, Benson A, Wong FS, Wen L. Long term
exposure, can determine its role in disease risk, although con- effect of gut microbiota transfer on diabetes development. J Autoimmun
2014;53:85–94.
founding factors could also be a major source of variation in 8. Musso G, Gambino R, Cassader M. Obesity, diabetes, and gut microbiota:
the results. the hygiene hypothesis expanded? Diabetes Care 2010;33:2277–84.
9. Sanz Y, Moya-Pérez A. Microbiota, inflammation and obesity. In:
Conclusions Lyte Mark, Cryan John F, eds. Microbial Endocrinology: The Micro-
Gut microbiota has recently been recognized to play a major biota–Gut–Brain Axis in Health and Disease. New York: Springer,
2014:291–317.
role in regulating the metabolic–immune axis, and conse- 10. Caesar R, Reigstad CS, Bäckhed HK, et al. Gut-derived lipopolysaccha-
quently, changes in gut microbiota related to lifestyle (type of ride augments adipose macrophage accumulation but is not essential for
delivery, hygienic measures, diet, etc.) may contribute to the impaired glucose or insulin tolerance in mice. Gut 2012;61:1701–7.
increased incidence of autoimmune and metabolic diseases 11. Serino M, Luche E, Gres S, et al. Metabolic adaptation to a high-fat diet is
in developed countries. Type-1 and type-2 diabetes represent associated with a change in the gut microbiota. Gut 2012;61:543–53.
12. Bäckhed F, Manchester JK, Semenkovich CF, Gordon JI. Mechanisms
two examples of diseases mediated by interactions between
underlying the resistance to diet-induced obesity in germ-free mice. Proc
the gut microbiota and the immune system with subsequent Natl Acad Sci USA 2007;104:979–84.
development of specific organ autoimmunity and/or meta- 13. Moore AM, Munck C, Sommer MO, Dantas G. Functional metage-
bolic dysfunction. Knowledge of the role gut microbiota plays nomic investigations of the human intestinal microbiota. Front Microbiol
in these two diseases could be used to develop intervention 2011;2:188.
14. Rey FE, Faith JJ, Bain J, et al. Dissecting the in vivo metabolic potential of
strategies to prevent and/or treat imbalances that lead to the
two human gut acetogens. J Biol Chem 2010;285:22082–90.
inflammation preceding overt manifestations of autoimmune 15. Cani PD, Possemiers S, Van de Wiele T, et al. Changes in gut microbiota
and metabolic disorders. The specific microbiome compo- control inflammation in obese mice through a mechanism involving GLP-
nents promoting or protecting against these disorders and 2-driven improvement of gut permeability. Gut 2009;58:1091–103.
the possible molecular targets for intervention are still under 16. Fernandes CA, Fievez L, Neyrinck AM, Delzenne NM, Bureau F, Van-
bever R. Sirtuin inhibition attenuates the production of inflammatory
debate. Nonetheless, mechanistic insights from human and
cytokines in lipopolysaccharide-stimulated macrophages. Biochem Bio-
animal studies point to the protective role of the intestinal phys Res Commun 2012;420:857–61.
bacteria that produce butyrate, which may strengthen the gut 17. Caricilli AM, Picardi PK, de Abreu LL, et al. Gut microbiota is a key
barrier function, and that induce immunoregulatory mecha- modulator of insulin resistance in TLR 2 knockout mice. PLoS Biol
nisms, commonly altered in both type-2 and type-1 diabetes. 2011;9:e1001212.
18. Kim KA, Gu W, Lee IA, Joh EH, Kim DH. High fat diet-induced gut micro-
Nevertheless, well-controlled prospective human studies are
biota exacerbates inflammation and obesity in mice via the TLR4 signaling
still necessary to gain a better understanding of the contribu- pathway. PLoS One 2012;7:e47713.
tion of specific gut microbiota and its response to environ- 19. Devkota S, Wang Y, Musch MW, et al. Dietary-fat-induced taurocholic
mental factors. Such information could be used to identify acid promotes pathobiont expansion and colitis in Il10−/− mice. Nature
effective preventive strategies targeting specific component of 2012;487:104–8.
the gut ecosystem. 20. Amar J, Serino M, Lange C, et al. Involvement of tissue bacteria in the onset of
diabetes in humans: evidence for a concept. Diabetologia 2011;54:3055–61.
STATEMENT OF FINANCIAL SUPPORT 21. Cano PG, Santacruz A, Trejo FM, Sanz Y. Bifidobacterium CECT 7765
This review was supported by public grants AGL2011-25169 from the Span- improves metabolic and immunological alterations associated with obesity
ish Ministry of Economy and Competitiveness (MINECO; Spain) and EC in high-fat diet-fed mice. Obesity 2013;21:2310–2122.
Project No. 613979 (MyNew Gut) from the 7th Framework Program. The 22. Sanz Y, Rastmanesh R, Agostoni C. Understanding the role of gut
scholarships to M. Olivares from CSIC (Spain) and to A. Moya-Pérez from microbes and probiotics in obesity: how far are we? Pharmacol Res
the Ministry of Economy and Competitiveness (MINECO; Spain) are fully 2013;69:144–55.
acknowledged. 23. Verdam FJ, Fuentes S, de Jonge C, et al. Human intestinal microbiota com-
Disclosure: There are no disclosures related to financial ties to products in position is associated with local and systemic inflammation in obesity.
the study or potential/perceived conflicts of interest. Obesity 2013;21:E607–15.
24. Le Chatelier E, Nielsen T, Qin J, et al. Richness of human gut microbiome
References correlates with metabolic markers. Nature 2013;500:541–6.
1. Olivares M, Neef A, Castillejo G, et al. The HLA-DQ2 genotype selects for 25. Jumpertz R, Le DS, Turnbaugh PJ, et al. Energy-balance studies reveal
early intestinal microbiota composition in infants at high risk of develop- associations between gut microbes, caloric load, and nutrient absorption
ing celiac disease. Gut 2014 (doi:10.1136/gutjnl-2014–306931). in humans. Am J Clin Nutr 2011;94:58–65.

Copyright © 2015 International Pediatric Research Foundation, Inc. Volume 77 | Number 1 | January 2015      Pediatric Research  243

Licensed to Natalia Sabrina Ortiz - [email protected]

 Review Sanz et al.
26. Santacruz A, Marcos A, Warnberg J, et al. Interplay between weight loss gut flora involved in the development of type 1 diabetes? Diabetologia
and gut microbiota composition in overweight adolescents. Obesity 2006;49:2105–8.
2009;17:1906–15. 48. King C, Sarvetnick N. The incidence of type-1 diabetes in NOD mice
27. Cotillard A, Kennedy SP, Kong LC, et al. Dietary intervention impact on is modulated by restricted flora not germ-free conditions. PLoS One
gut microbial gene richness. Nature 2013;500:585–8. 2011;6:e17049.
28. Ridaura VK, Faith JJ, Rey FE, et al. Gut microbiota from twins discordant 49. Calcinaro F, Dionisi S, Marinaro M, et al. Oral probiotic administration
for obesity modulate metabolism in mice. Science 2013;341:1241214. induces interleukin-10 production and prevents spontaneous autoimmune
29. Gauffin Cano P, Santacruz A, Moya Á, Sanz Y. Bacteroides uniformis CECT diabetes in the non-obese diabetic mouse. Diabetologia 2005;48:1565–75.
7771 ameliorates metabolic and immunological dysfunction in mice with 50. Marietta EV, Gomez AM, Yeoman C, et al. Low incidence of spontaneous
high-fat-diet induced obesity. PLoS One 2012;7:e41079. type 1 diabetes in non-obese diabetic mice raised on gluten-free diets is asso-
30. Kootte RS, Vrieze A, Holleman F, et al. The therapeutic potential of manip- ciated with changes in the intestinal microbiome. PLoS One 2013;8:e78687.
ulating gut microbiota in obesity and type 2 diabetes mellitus. Diabetes, 51. Brown CT, Davis-Richardson AG, Giongo A, et al. Gut microbiome
Obes Metab 2012;14:112–20. metagenomics analysis suggests a functional model for the development of
31. Kalliomäki M, Collado MC, Salminen S, Isolauri E. Early differences in autoimmunity for type 1 diabetes. PLoS One 2011;6:e25792.
fecal microbiota composition in children may predict overweight. Am J 52. Giongo A, Gano KA, Crabb DB, et al. Toward defining the autoimmune
Clin Nutr 2008;87:534–8. microbiome for type 1 diabetes. ISME J 2011;5:82–91.
32. Bergström A, Skov TH, Bahl MI, et al. Establishment of intestinal micro- 53. Murri M, Leiva I, Gomez-Zumaquero JM, et al. Gut microbiota in children
biota during early life: a longitudinal, explorative study of a large cohort of with type 1 diabetes differs from that in healthy children: a case–control
Danish infants. Appl Environ Microbiol 2014;80:2889–900. study. BMC Med 2013;11:46.
33. Santacruz A, Collado MC, García-Valdés L, et al. Gut microbiota composi- 54. Soyucen E, Gulcan A, Aktuglu-Zeybek AC, Onal H, Kiykim E, Aydin A.
tion is associated with body weight, weight gain and biochemical param- Differences in the gut microbiota of healthy children and those with type 1
eters in pregnant women. Br J Nutr 2010;104:83–92. diabetes. Pediatr Int 2014;56:336–43.
34. Karlsson CL, Molin G, Cilio CM, Ahrné S. The pioneer gut microbiota 55. Cardwell CR, Stene LC, Joner G, et al. Caesarean section is associated with
in human neonates vaginally born at term—a pilot study. Pediatr Res an increased risk of childhood-onset type 1 diabetes mellitus: a meta-anal-
2011;70:282–6. ysis of observational studies. Diabetologia 2008;51:726–35.
35. Larsen N, Vogensen FK, van den Berg FW, et al. Gut microbiota in human 56. Pereira PF, Alfenas Rde C, Araújo RM. Does breastfeeding influence the
adults with type 2 diabetes differs from non-diabetic adults. PLoS One risk of developing diabetes mellitus in children? A review of current evi-
2010;5:e9085. dence. J Pediatr (Rio J) 2014;90:7–15.
36. Wu X, Ma C, Han L, et al. Molecular characterisation of the faecal micro- 57. Dominguez-Bello MG, Costello EK, Contreras M, et al. Delivery mode
biota in patients with type II diabetes. Curr Microbiol 2010;61:69–78. shapes the acquisition and structure of the initial microbiota across multi-
37. Qin J, Li Y, Cai Z, et al. A metagenome-wide association study of gut ple body habitats in newborns. Proc Natl Acad Sci USA 2010;107:11971–5.
microbiota in type 2 diabetes. Nature 2012;490:55–60. 58. Decker E, Engelmann G, Findeisen A, et al. Cesarean delivery is associated
38. Karlsson FH, Tremaroli V, Nookaew I, et al. Gut metagenome in Euro- with celiac disease but not inflammatory bowel disease in children. Pediat-
pean women with normal, impaired and diabetic glucose control. Nature rics 2010;125:e1433–40.
2013;498:99–103. 59. Rewers M, Liu E, Simmons J, Redondo MJ, Hoffenberg EJ. Celiac disease
39. Zhang X, Shen D, Fang Z, et al. Human gut microbiota changes reveal the associated with type 1 diabetes mellitus. Endocrinol Metab Clin North Am
progression of glucose intolerance. PLoS One 2013;8:e71108.39. 2004;33:197–214, xi.
40. Pussinen PJ, Havulinna AS, Lehto M, Sundvall J, Salomaa V. Endotoxemia 60. Palma GD, Capilla A, Nova E, et al. Influence of milk-feeding type and
is associated with an increased risk of incident diabetes. Diabetes Care genetic risk of developing coeliac disease on intestinal microbiota of
2011;34:392–7. infants: the PROFICEL study. PLoS One 2012;7:e30791.
41. American Diabetes Association. Diagnosis and classification of diabetes 61. Krause I, Anaya JM, Fraser A, et al. Anti-infectious antibodies and auto-
mellitus. Diabetes Care 2014;37:Suppl 1:S81–90. immune-associated autoantibodies in patients with type I diabetes mellitus
42. Vaarala O, Atkinson MA, Neu J. The “perfect storm” for type 1 diabetes: and their close family members. Ann NY Acad Sci 2009;1173:633–9.
the complex interplay between intestinal microbiota, gut permeability, and 62. Laitinen OH, Honkanen H, Pakkanen O, et al. Coxsackievirus B1 is associ-
mucosal immunity. Diabetes 2008;57:2555–62. ated with induction of β-cell autoimmunity that portends type 1 diabetes.
43. Roep BO, Tree TI. Immune modulation in humans: implications for type 1 Diabetes 2014;63:446–55.
diabetes mellitus. Nat Rev Endocrinol 2014;10:229–42. 63. Honeyman MC, Coulson BS, Stone NL, et al. Association between rotavi-
44. Secondulfo M, Iafusco D, Carratu R, et al. Ultrastructural mucosal altera- rus infection and pancreatic islet autoimmunity in children at risk of devel-
tions and increased intestinal permeability in non-celiac, type I diabetic oping type 1 diabetes. Diabetes 2000;49:1319–24.
patients. Dig Liver Dis 2004;36:35–45. 64. Lönnrot M, Korpela K, Knip M, et al. Enterovirus infection as a risk
45. Sapone A, de Magistris L, Pietzak M, et al. Zonulin upregulation is associ- factor for beta-cell autoimmunity in a prospectively observed birth
ated with increased gut permeability in subjects with type 1 diabetes and cohort: the Finnish Diabetes Prediction and Prevention Study. Diabetes
their relatives. Diabetes 2006;55:1443–9. 2000;49:1314–8.
46. Bosi E, Molteni L, Radaelli MG, et al. Increased intestinal permeability 65. Levy-Marchal C, Patterson CC, Green A, EURODIAB ACE Study Group.
precedes clinical onset of type 1 diabetes. Diabetologia 2006;49:2824–7. Europe and Diabetes. Geographical variation of presentation at diagnosis
47. Brugman S, Klatter FA, Visser JT, et al. Antibiotic treatment partially pro- of type I diabetes in children: the EURODIAB study. European and Dibe-
tects against type 1 diabetes in the Bio-Breeding diabetes-prone rat. Is the tes. Diabetologia 2001;44:Suppl 3:B75–80.

244 Pediatric Research      Volume 77 | Number 1 | January 2015 Copyright © 2015 International Pediatric Research Foundation, Inc.

Licensed to Natalia Sabrina Ortiz - [email protected]