Glucose Metabolism: Focus on Gut Microbiota, the Endocannabinoid System

Available online at ScienceDirect

www.sciencedirect.com

Diabetes & Metabolism 40 (2014) 246–257

Review

Glucose metabolism: Focus on gut microbiota, the endocannabinoid system

and beyond

∗,1 1 1 1 1

P.D. Cani , L. Geurts , S. Matamoros , H. Plovier , T. Duparc

Université catholique de Louvain, Louvain Drug Research Institute, WELBIO (Walloon Excellence in Life sciences and BIOtechnology), Metabolism and Nutrition

research group, Avenue E. Mounier, 73 Box B1.73.11, 1200 Brussels, Belgium

Received 28 November 2013; received in revised form 5 February 2014; accepted 5 February 2014

Available online 14 March 2014

Abstract

The gut microbiota is now considered as a key factor in the regulation of numerous metabolic pathways. Growing evidence suggests that

cross-talk between gut bacteria and host is achieved through speciﬁc metabolites (such as short-chain fatty acids) and molecular patterns of

microbial membranes (lipopolysaccharides) that activate host cell receptors (such as toll-like receptors and G-protein-coupled receptors). The

endocannabinoid (eCB) system is an important target in the context of obesity, type 2 diabetes (T2D) and inﬂammation. It has been demonstrated

that eCB system activity is involved in the control of glucose and energy metabolism, and can be tuned up or down by speciﬁc gut microbes (for

example, Akkermansia muciniphila). Numerous studies have also shown that the composition of the gut microbiota differs between obese and/or

T2D individuals and those who are lean and non-diabetic. Although some shared taxa are often cited, there is still no clear consensus on the precise

microbial composition that triggers metabolic disorders, and causality between speciﬁc microbes and the development of such diseases is yet to be

proven in humans. Nevertheless, gastric bypass is most likely the most efﬁcient procedure for reducing body weight and treating T2D. Interestingly,

several reports have shown that the gut microbiota is profoundly affected by the procedure. It has been suggested that the consistent postoperative

increase in certain bacterial groups such as Proteobacteria, Bacteroidetes and Verrucomicrobia (A. muciniphila) may explain its beneﬁcial impact

in gnotobiotic mice. Taken together, these data suggest that speciﬁc gut microbes modulate important host biological systems that contribute to the

control of energy homoeostasis, glucose metabolism and inﬂammation in obesity and T2D.

Keywords: Gut microbiota; Obesity; Endocannabinoid; Akkermansia

1. Introduction with germ-free mice, which lack microbiota [5]. These animals

exhibit a reduced fat mass compared with their conventionally

The gut microbiota is now considered a separate organ that raised littermates in spite of an increased food intake. More-

is involved in the regulation of numerous physiological path- over, transplantation of faecal microbiota from obese mice into

ways by impacting different functions of the host [1]. Among lean germ-free mice has demonstrated the capacity of the bac-

these regulatory actions, the inﬂuence of gut microbes on energy terial ecosystem to alter host phenotype independently of either

metabolism is of particular interest, as it has been proposed to genotype or diet [6]. Several studies have conﬁrmed this abil-

be a driving force in the pathogenesis of metabolic diseases, ity, including a recent one showing that the transplantation of

especially obesity. Intestinal microbes have developed a mutu- microbiota from twins discordant for obesity partially replicated

ally beneﬁcial relationship with their host, and can inﬂuence the donor’s phenotype. Transferring gut microbes from obese

physiological systems by modulating gut motility, intestinal twin donors to germ-free mice increased their weight gain com-

barrier homoeostasis, nutrient absorption and fat distribution pared with mice transplanted with microbiota from lean twin

[2–4]. The proof of concept for the involvement of gut bacte- donors [7]. These experiments conﬁrm the inﬂuence of micro-

ria in the processes of energy homoeostasis was demonstrated biota on the development of obesity, although causality between

the observed changes and metabolic symptoms is still unclear, as

well as the applicability of this knowledge to the diagnosis, pre-

∗ vention and treatment of the metabolic syndrome. Nevertheless,

Corresponding author. Tel.: +32 2 764 73 97; fax: +32 2 764 73 59.

E-mail address: [email protected] (P.D. Cani). they do suggest a relationship between nutrition, gut microbiota

All authors contributed equally to this work. and energy homoeostasis.

P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257 247

The present review discusses recent evidence supporting a strong relationship between gut microbiota, inﬂammation and

the hypothesis that gut microbiota can inﬂuence the whole of metabolic perturbations.

host metabolism through various mechanisms and that changes There is also growing interest in the study of the intesti-

in microbiota composition can trigger changes in metabolic nal mucus layer and its interactions with microbiota. Recently,

behaviour. Signalling and molecular pathways involved in the we have demonstrated the key role played by the gut micro-

regulation of energy homoeostasis by the gut microbiota are biota and its interaction with the mucus layer in the context

varied [including short-chain fatty acids (SCFAs), the endo- of diet-induced obesity and T2D, where the numbers of Akker-

cannabinoid (eCB) system and gut peptides], with many others mansia muciniphila, mucin-degrading bacteria that reside in and

that are yet to be identiﬁed or clariﬁed. abundantly colonize the mucus layer, were negatively correlated

with body weight and decreased under HFD conditions (Fig. 1)

[23]. Moreover, daily administration of A. muciniphila to HFD-

2. Gut microbiota and metabolism: putative actors and induced obese mice for 4 weeks improved their metabolic proﬁle

pathways by decreasing weight gain, restoring mucus-layer thickness, and

counteracting metabolic endotoxaemia and insulin resistance

Among the most studied bacterial metabolites that can inter- [23]. A recent study by Shan et al. [24] conﬁrmed that the

fere with host metabolism are the SCFAs. These products of mucus layer is not only a non-speciﬁc physical barrier, but also

microbiota-mediated fermentation of polysaccharides modu- a complex organized structure that can deliver immunoregula-

late levels of several gut hormones involved in glucose and tory signals that participate in gut homoeostasis. On the same

energy homoeostasis, including glucagon-like peptide (GLP)-1 topic, Kashyap et al. [25] recently revealed that modiﬁcation of

and ghrelin [8,9]. These metabolites circulate in the blood and, mucus carbohydrate composition also inﬂuences the microbiota.

thus, can act on peripheral targets to modulate insulin sensitivity In a gnotobiotic mouse model that presented a severe decrease

and the whole of the host’s energy metabolism [10,11]. Unfor- in mucus fucosylated glycans, they observed a decrease in fae-

tunately, most of the pathways underlying these effects are still cal microbiota diversity compared with a control group that was

largely unknown, although several studies have suggested a link associated with major changes in faecal and urinary metabolites,

with members of a recently identiﬁed G-protein-coupled recep- suggesting modiﬁcation of the entire metabolism of the animals

tor family that includes G-protein-coupled receptors 43 (GPR43) [25]. These data were further conﬁrmed by Sommer et al. [26],

and 41 (GPR41) [12,13]. An elegant study by De Vadder et al. who found alterations in gut microbiota composition and intesti-

[14] recently showed that SCFAs activate intestinal neogluco- nal architecture in mice with a defect in mucus glycosylation.

genesis through a cAMP-dependent mechanism and a gut–brain These ﬁndings support an important role for the mucus layer

neural circuit involving GPR41, thereby pointing to intestinal in energy homoeostasis, although the key mechanisms linking

neoglucogenesis as a novel actor in gut microbiota ability to intestinal mucus production and whole-body energy homoeosta-

host interactions. sis have yet to be fully elucidated.

In addition to these speciﬁc metabolites, the gut bacte- The gut microbiota interact with other organs. Its interactions

ria are also able to interact with the host through speciﬁc with the liver through LPS-induced inﬂammation in obesity have

cell membranes and other related molecules that activate already been alluded to, and other reports strongly suggest inter-

pattern-recognition receptors (PRRs). PRRs recognize molec- actions with the brain. The presence of cross-talk between the gut

ular patterns unique to bacteria and other microorganisms microbiota and the brain was demonstrated nearly 10 years ago,

(pathogen-associated molecular patterns, or PAMPs). The most when we found that changing the gut microbiota with prebiotics

studied PRRs are the toll-like receptors (TLRs) which, when reduced food intake, body weight and fat-mass development in

stimulated, result in an inﬂammatory response, cytokine produc- rodents; the phenomenon was also associated with an increase

tion and chemokine-mediated recruitment of acute inﬂammatory in the endogenous production of anorexigenic peptides, such as

cells [15]. Of the PAMPs, we discovered that lipopolysaccha- GLP-1 and peptide tyrosine tyrosine (PYY), and a decrease in

rides (LPSs), components of the cell walls of Gram-negative the orexigenic peptide ghrelin [27–29]. Differences in the central

bacteria, contribute to the development of inﬂammation and expression of peptides involved in energy homoeostasis between

insulin resistance in both obesity and type 2 diabetes (T2D) germ-free and conventional mice have also been demonstrated

in a condition known as “metabolic endotoxaemia” (Fig. 1) [30]. The latter animals exhibited decreased central expression

[16,17]. This is associated with altered gut microbiota compo- of proglucagon (precursor of GLP-1), and less brain-derived

sition as well as increased intestinal permeability, resulting in neurotrophic factor (BDNF) and its receptor, compared with

increased plasma LPS levels (Fig. 1) [16–19]. Such increases the germ-free mice. The authors also found that the expres-

lead to CD14/TLR4 activation, which in turn induces an sion of food intake regulatory neuropeptides was modiﬁed, with

inﬂammatory response that results in perturbations of energy decreases in neuropeptide Y (NPY) and Agouti-related pep-

homoeostasis as well as increased inﬂammation, hepatic steato- tide (AgRP) expression, and increases in proopiomelanocortin

sis, hyperinsulinaemia and insulin resistance [16]. Speciﬁcally, (POMC) and cocaine- and amphetamine-regulated transcript

LPS acts ﬁrst on the liver by inducing hepatic insulin resis- (CART) expression compared with conventional mice [30].

tance [16] and decreasing hepatic inﬂammatory responses by These studies suggest the involvement of a neural pathway

depleting Kupffer cells prevent high-fat-diet (HFD)-induced that allows an exchange of information between gut microbiota

metabolic effects [20–22]. Taken together, these data highlight and hypothalamic nuclei involved in energy homoeostasis. In

248 P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257

Fig. 1. Cross-talk between gut microbiota and host. Obesity and type 2 diabetes (T2D) are associated with changes in the composition of gut microbiota, leading

to increases in some genera and decreases in others (arrow direction indicates either increased or decreased abundance). Some bacteria are positively or negatively

associated with obesity and T2D, depending on the study (indicated by question marks). Increased mucus degradation is associated with increased gut permeability

and metabolic endotoxaemia, triggering inﬂammation, macrophage inﬁltration of adipose tissue and insulin resistance (adapted from Delzenne et al. [111]).

addition to interactions at the level of energy homoeostasis, and different links between the gut microbiota and the eCB system

to further underscore the close relationship between the gut and prompting an assessment of the physiology of the eCB system

the brain, particular pathologies such as autism and depression and its links to diabetes.

have also been linked to the gut microbiota [31,32]. Indeed,

earlier studies showed that depression in women is associated

with increased fermentation of carbohydrates [33], suggesting 3. Changing eCB system activity: a therapeutic target?

a relationship between changes in the composition or metabolic

activity of the gut microbiota and brain-related pathologies. The eCB system is complex, comprising several bioac-

Independently, recent data have also shown that jejunal pro- tive lipids and the enzymes that regulate their production and

teins from either obese mice or insulin-resistant human subjects degradation, and different types of nuclear and cell mem-

impaired muscle insulin signalling, thereby promoting insulin brane receptors. eCBs are synthesized on demand from cell

resistance [34]. From this it may be speculated that gut micro- membrane phospholipids and immediately released from the

biota and its byproducts might also interact with muscles and fat cell to target their receptors [37]. Anandamide (AEA) and 2-

depots (Fig. 1) [16,35]. However, in spite of this wide range of arachidonoylglycerol (2-AG) are the best-characterized eCBs so

actors and interactions, it is important to remember that energy far. Their principal receptors are the Gi/o-coupled receptors CB1

homoeostasis is a highly complex integrated system involving a and CB2, which are also targeted by the principal active compo-

9 9

number of pathways, and that gut microbes are neither the only nent of Cannabis sativa, -tetrahydrocannabinol ( -THC)

nor the most important factors involved in energy homoeostasis. [38,39]. Several pathways are involved in both AEA and 2-

A study by Harley et al. [36] of two sets of C57BL/6 mice that AG synthesis and degradation, making this a relatively complex

differed in their obesogenic responses to HFDs revealed that the signalling system [37,40]. Monoacylglycerol lipase (MAGL) is

differences were not due to different gut microbiota composi- thought to be the primary enzyme responsible for 2-AG degra-

tion. Given this result, the authors reported uncertainty as to the dation, whereas fatty acid amide hydrolase (FAAH) is thought

involvement of gut microbiota in obesity through the expression to be the main AEA degrading enzyme [41]. Although eCB

of species-independent metagenomic pathways. metabolism has mostly been studied at the level of FAAH and

Of the various pathways involved in the regulation of glu- MAGL, another degrading enzyme – ␣/␤-hydrolase domain 6

cose metabolism and energy homoeostasis, numerous studies (ABHD6) – has recently been implicated in 2-AG hydrolysis

have been done on the eCB system. We have demonstrated that [42].

P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257 249

Fig. 2. Schematic overview of endocannabinoid (eCB) synthesis and degradation. N-acylethanolamine (NAE), anandamide (AEA), N-palmitoylethanolamine (PEA)

and N-oleoylethanolamine (OEA) are synthesized on demand from cell membrane phospholipids through N-acylphosphatidyl-ethanolamine-speciﬁc phospholipase

D (NAPE-PLD). 2-Arachidonoylglycerol (2-AG) is also produced from cell membrane phospholipids through the action of diacylglycerol lipase (DAGL). These

bioactive lipids activate cannabinoid receptors 1 and 2 (CB1 and CB2), also targeted by 9-tetrahydrocannabinol (9-THC), the principal active component of

Cannabis sativa, or other non-cannabinoid receptors such as PPAR␣, GPR55, GPR119 and TRPV1. These lipids are hydrolyzed in the cell by several lipases. NAE is

mainly hydrolyzed by fatty acid amide hydrolase (FAAH) and N-acylethanolamine-hydrolyzing acid amidase (NAAA) into ethanolamine and a fatty acid (depending

on the NAE hydrolyzed). 2-AG is hydrolyzed by two serine hydrolases, monoacylglycerol lipase (MAGL) and ␣/␤-hydrolase domain 6 (ABHD6), into glycerol and

arachidonic acid (AA).

Other eCB-related lipids considered part of the eCB system [47]. However, we also found that eCB-related molecules such

include palmitoylethanolamine (PEA) and oleoylethanolamine as PEA, OEA and stearoyl ethanolamine (SEA) are inversely

(OEA). These lipids activate non-cannabinoid recep- related to AEA in obesity, T2D and inﬂammation (Geurts L.,

tors – namely, peroxisome proliferator-activated receptor Muccioli G.G. and Cani P.D., personal communication). CB1

alpha (PPAR␣), GPR55, GPR119 and transient receptor poten- receptor stimulation increases food intake (Fig. 3) [48], whereas

tial cation channel, subfamily V, member 1 (TRPV1) – and pharmacological inhibition of CB1 decreases fat-mass expan-

share biosynthesis and degradation pathways with AEA and sion and reduces body weight [49–52]. The role of the CB1

2-AG [40,43,44]. A schematic overview of eCB synthesis and receptor was conﬁrmed several years ago using CB1 knockout

degradation is shown in Fig. 2. (CB1-/-) mice. Without the CB1 receptor, mice were resistant to

Over the past decade, research into eCBs has conﬁrmed that diet-induced obesity (DIO) [48–50]. However, the CB1 receptor

the eCB system is a key signalling system implicated in the reg- is not the only receptor incriminated in these metabolic dis-

ulation of energy homoeostasis and metabolism [45]. In fact, turbances, as FAAH-/- mice have also shown augmented body

eCBs are widely produced in organs that contribute to energy weight and food motivation compared with their wild-type lit-

homoeostasis (pancreas, muscle, gut, adipose tissue, liver and termates [53]. More recently, the crucial role of ABHD6 was

hypothalamus) and normally facilitate energy intake and stor- detailed by Thomas et al. [54], who found that ABDH6 was

age in particular (Fig. 3). As eCB levels in tissues are tightly up-regulated by HFD, while ABHD6-/- mice were resistant to

controlled by a balance between synthesis and degradation, dys- DIO. These data suggest that the eCB system as a whole plays

regulation of this control can lead to pathological conditions such a central role in the control of body weight and central appetite

as obesity and T2D (Fig. 3). Indeed, the eCB system is severely regulation.

altered in obesity, which may result in increased eCB levels eCBs have also been implicated in the development of T2D

and CB1 activity in the brain, liver, adipose tissue and skeletal and insulin resistance, as CB1-/- mice do not display the distur-

muscles [46], and decreased levels of enzymes and receptors in bances associated with DIO. The ﬁrst studies describing CB1-/-

other organs, such as the stomach, kidneys and heart (Fig. 3) mice and pharmacological inhibition of CB1 also reported less

250 P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257

Fig. 3. The endocannabinoid (eCB) system and metabolism. This system plays a central role in the regulation of glucose homoeostasis and insulin sensitivity in

several peripheral organs. CB1 receptor activation induces gut permeability. Gut microbiota composition is associated with intestinal eCB content and CB1 receptor

mRNA expression.

insulin resistance in DIO mice [49,50] and more muscle glu- under physiological conditions and exacerbates glucose intol-

cose uptake [55]. The role of eCBs in the control of glucose erance under HFD challenge, highlighting the speciﬁc role of

homoeostasis has been reported by several studies showing the the eCB system in controlling glucose homoeostasis [35]. The

involvement of the eCB system in pancreatic islets in rodents and CB2 receptor has also been implicated in glucose homoeosta-

humans; however, it is still not clear whether CB1 receptor acti- sis, as CB2 receptor activation improves glucose tolerance in

vation controls the insulin signalling pathways or, more directly, rats receiving a glucose load, mimicking the effect of CB1

insulin release by ␤ cells (Fig. 3) [56–60]. Insulin is a potent receptor blockade [70]. These ﬁndings suggest that the coordi-

regulator of eCB metabolism in normal adipocytes [61]. Acute nated actions of both CB1 and CB2 receptors modulate glucose

administration of AEA and eCB agonists in mice inﬂuences homoeostasis. However, the exact role of the CB2 receptor still

insulin sensitivity, induces glucose intolerance, and modulates needs further clariﬁcation.

insulin-induced glucose uptake and glucose transporter type 4 Recent studies have unravelled the underlying mechanisms

(GLUT4) translocation in differentiated adipocytes and skele- of eCB-mediated glucose metabolism dysregulation. Acute and

tal muscle; these effects can be reversed by treatment with a chronic treatment of adipocytes with a CB1 receptor antago-

CB1 receptor antagonist [62–65]. Blocking the CB1 receptor nist upregulates GLUT-4 expression, and this regulation has

improves insulin sensitivity and increases whole-body insulin- a transcriptional effect on both nuclear factor (NF)-␬B and

dependent glucose utilization by stimulating glucose uptake in sterol regulatory element-binding protein (SREBP)-1 [71]. Fur-

white and brown adipose tissue (WAT and BAT, respectively). thermore, hepatic CB1 activation leads to accumulation of

eCB-induced modulation of BAT function could be one of long-chain ceramides in the liver that appear to mediate eCB-

the mechanisms by which eCBs exert their insulin-sensitizing induced hepatic insulin resistance [72]. An essential feature

effects [66]. Interestingly, liver-speciﬁc CB1-/- mice become in the progression of insulin resistance may be islet inﬂam-

obese under HFD conditions, but are resistant to hepatic steatosis mation, although the primary proinﬂammatory signal remains

and insulin resistance. These results highlight the key role of the unidentiﬁed. In addressing this issue, a recent study revealed

hepatic CB1 receptor [67], and were recently conﬁrmed by Liu that peripheral blockade of the CB1 receptor delays progression

et al. [68], who showed that mice with a speciﬁc overexpression of T2D and the underlying ␤-cell loss in Zucker diabetic fatty

of the CB1 receptor in hepatocytes became hyperinsulinaemic (ZDF) rats. This study showed that the CB1 receptors mediating

as a result of reduced insulin clearance. Furthermore, FAAH- these effects are located on inﬁltrating macrophages, and their

/- mice exhibit ectopic lipid storage in the liver, leading to an activation induces the Nlrp3 inﬂammasome [73]. Nevertheless,

insulin-resistant state. the exact role of the CB1 receptor is complicated, as demon-

These results demonstrate a role for eCB players beyond strated by Li et al. [74], who found that genetic loss of the CB1

the CB1 receptor and AEA [69]. Recently, we have shown receptor in leptin-resistant (ob/ob) mice failed to suppress Ob

that chronic stimulation with a potent CB1/CB2 receptor ago- phenotype and even worsened abnormal glucose metabolism

nist (HU210) induces glucose intolerance in wild-type mice in ob/ob mice, thereby highlighting the complex relationship

P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257 251

between the CB1 receptor and leptin. Thus, focusing on the eCB an ecosystem, whereas diversity takes into account the num-

system and diabetes reveals the essential roles of eCB mediators ber of bacterial phyla identiﬁed (richness) in addition to their

in insulin sensitivity, glycaemia and glucose uptake, although relative abundance. Despite being useful descriptors of the bac-

discrepancies can still be found in the literature regarding the terial ecosystem in general, neither term appears to be a reliable

precise mechanisms. indicator of host diabetic status [84,86]. Although animal experi-

ments show a clear separation between diabetic and non-diabetic

4. Cross-talk between eCBs and gut microbes: a key subjects based on their microbiota proﬁles [76,87], the interindi-

link in obesity and T2D? vidual variability in humans most likely masks these large-scale

differences [84]. Moreover, the results are not always concordant

To ascertain the role of eCBs as essential molecules in the between cohorts. For example, Danish and Chinese T2D patients

metabolic alterations observed in obesity and T2D, we recently are clustered separately based on metagenomic data [88]. This

proposed that the eCB system mediates communication between suggests that the extrapolation of results, such as the predic-

adipose tissue and the gut [75,76]. These links have recently been tion of diabetic status from microbiota data, is difﬁcult across

reviewed [77,78]. In brief, we revealed that eCBs are impli- populations and illustrates probable population-based speciﬁcity

cated in gut barrier function and that speciﬁc changes in the [88]. The concept of enterotypes was initially proposed to clas-

tone of the eCB system in obesity and T2D lead to increased sify human populations according to their dominant intestinal

gut permeability (Figs. 1 and 3) [75]. These ﬁndings have been microbial features [89], but no correlations were found between

conﬁrmed by two other recent studies [79,80]. More precisely, enterotypes and diabetic status in the Chinese cohorts [84,90].

we found that modulating the gut microbiota of obese and T2D While the taxonomy of the microbiota alone has failed to

mice profoundly affected the tone of the intestinal and adipose explain their inﬂuence on the onset of the metabolic syndrome,

tissue eCB system. Interestingly, changing the gut microbiota description of bacterial metabolic functions (at the genetic

of ob/ob mice with prebiotics also reduced AEA content and level) using metagenomic shotgun sequencing in humans and

CB1 mRNA expression in adipose tissue, and improved adipose mice has proved to be a reliable, complementary tool, and has

tissue metabolism, a phenotype similar to that observed follow- revealed shifts in metabolic function related to both obesity

ing CB1 receptor blockade [75]. Moreover, it was revealed that and T2D [6]. The most prominent features of T2D-associated

the lower fat mass and CB1 mRNA levels were associated with metagenomes are the enriched pathways related to carbohydrate

markers of differentiation and lipogenesis in adipose tissue [75]. metabolism and transport, branched-chain amino-acid transport

Recently, we demonstrated that the administration of and response to oxidative stress. On the other hand, pathways

A. muciniphila in mice fed either normal chow or HFD related to ﬂagellar assembly, butyrate biosynthesis and vitamin

increased levels of 2-AG and associated acylglycerols [2- metabolism were reduced in the Danish and Chinese cohorts

oleoylglycerol (2-OG) and 2-palmitoylglycerol (2-PG)] in the [88,90].

small intestine (Figs. 1 and 3) [23], thereby contributing to Following the same reasoning, several studies have identiﬁed

the anti-inﬂammatory effects and improved gut barrier func- individual taxa as important markers of obesity and diabetes

tion observed following treatment with these bacteria. Also, onset, although the exact roles of some of these species are

obese and T2D db/db mice showed changes in gut microbiota, not currently known. The Bacteroides, Roseburia and Akker-

which is consistent with an increased eCB system tone and mansia genera, as well as Faecalibacterium prausnitzii, were

inﬂammation in adipose tissue (Fig. 3) [76]. Chronic stimula- depleted in the T2D Chinese patients, whereas Dorea, Prevotella

tion of the eCB system increased adipogenesis in lean mice, and Collinsella were found in relatively greater abundance

and induced the metabolic endotoxaemia related to the low- (Fig. 1) [84]. In both humans and mice, Prevotella, Akker-

grade inﬂammation associated with obesity [16,75]. Recently, mansia and enterobacteria have previously been shown to vary

it has been proposed that adipocytes produce more AEA dur- signiﬁcantly between obese and lean subjects (Fig. 1) [91–95].

ing adipocyte differentiation, and this AEA may be converted The abundance of A. muciniphila is lower in diabetic mice and,

to prostamides by neighbouring preadipocytes, thus inhibiting as mentioned earlier, treatment with this bacterium by gavage

adipogenesis [81]. All these data demonstrate that the eCB sys- improved metabolism [23]. Also, a reduction in a cluster of

tem is closely involved in adipose tissue metabolism, and also genes belonging to Roseburia and F. prausnitzii has been iden-

acts as a mediator between gut microbiota and adipose tissue. tiﬁed as a discriminant marker for the prediction of diabetic

status in European women [88] and, in an obese French cohort,

5. Gut microbiota and T2D: a lack of consensus F. prausnitzii abundance was less than in control subjects, while

an increase in this bacterium correlated with an improved inﬂam-

One of the earliest pieces of evidence for obesity- and matory status [96]. However, these bacteria were increased in

diabetes-related dysbiosis in mice and humans was the increase obese Indian children compared with the lean controls, high-

in Firmicutes-to-Bacteroidetes ratio that correlated with an lighting once again the speciﬁcity of population, age and diets

increase in body mass index (BMI; Fig. 1) [6,82,83]. How- in phenotype–taxonomy associations [97]. In the same cohort,

ever, a number of recent studies, including some with human Lactobacillus abundance (species and gene clusters) was greater

cohorts, reported no variations in this ratio between diabetic in T2D patients and positively correlated with blood glucose

or obese patients and controls [84,85]. Abundance is a mea- levels (Fig. 1) [97]. Metagenomic analysis of the microbiota

sure of the relative proportion of each bacterial phylum within in an obese diabetic Chinese patient undergoing a 23-week

252 P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257

dietary intervention showed a sharp decrease in the abundance of 2

BMI ≥ 35 kg/m . In this procedure, the stomach is reduced to a

Enterobacteriaceae family members along with a simultaneous

small gastric pouch, and the small intestine is bypassed through

reduction in LPS synthesis pathways and inﬂammation [98].

the creation of a Roux limb. The rest of the stomach and duode-

Enterobacteria are known LPS producers and contribute to the

num thus constitute a blind duct. In addition to sustained weight

metabolic endotoxaemia observed in the metabolic syndrome

loss and BMI reduction, RYGB causes a weight-independent

[16,99–101]. Thus, despite their very low numbers relative to

increase in insulin sensitivity, leading to remission of T2D in

the immensity of the whole ecosystem, some taxa nevertheless

around 80% of patients [119,121].

appear to have a strong inﬂuence on the development of obesity

Several mechanisms can explain the impact of RYGB on T2D

and its associated disorders [102].

and they have been regrouped under the acronym BRAVE: bile

A major characteristic of T2D is the systemic increase in

acid ﬂow modiﬁcation; reduction of stomach size; anatomical

oxidative stress [103]. One way to counterbalance the increase

modiﬁcations and altered nutrient inﬂux; vagal manipulation;

of reactive oxygen species and depletion of antioxidants is to

and enteroendocrine secretion modulation [122]. Interestingly,

supplement the latter through diet. Several natural substances,

these factors have all been shown to inﬂuence or be inﬂu-

such as polyphenols, have antioxidant effects in addition to

enced by gut microbes. By affecting O2, pH and nutrient

prebiotic effects [104,105]. Intestinal bacteria such as Biﬁdobac-

levels, RYGB changes the gastrointestinal ecology, leading to

terium species also exhibit antioxidant capacities in vivo [106],

remodelling of the gut microbial population [123,124]. In addi-

and dietary interventions with probiotics containing Lactobacil-

tion, the gut bacteria themselves affect the pool of bile acids

lus acidophilus LA5 and B. animalis subsp. lactis BB12 have

[125] and secretion of gut hormones, and may regulate host

improved glucose tolerance and total antioxidant status in T2D

metabolism through the gut–brain axis [30,126], as detailed

patients [107]. Thus, this mechanism may also be contribut- above.

ing to the reported effectiveness of prebiotics and probiotics in

Over the past decade, many groups have investigated whether

the reduction of metabolic syndrome symptoms [95,108–112];

RYGB has an impact on gut microbiota through either next-

nevertheless, further investigations with different strains of Lac-

generation sequencing or quantitative polymerase chain reaction

tobacillus are still warranted (Fig. 1) [113].

(PCR) testing. Microbial communities were compared in obese

In addition to the use of prebiotics and probiotics, there are

subjects that had been treated or not with RYGB [91] and in the

contradictory data to suggest that antibiotic-induced changes in

same individuals before and after the operation [96,127,128].

gut microbiota composition may either counteract or worsen

These studies all found an increase in Proteobacteria after

obesity and T2D. For example, we discovered that broad-

surgery, mostly members of the Enterobacteriaceae family of

spectrum antibiotic treatments abolished diet-induced body

Gammaproteobacteria. This increase has been correlated with

weight gain, metabolic endotoxaemia, inﬂammation, oxidative

metabolic parameters such as reductions in BMI, adipose tissue

stress and insulin resistance in mice [17], and these data have

mass, serum leptin and, in some studies, inﬂammation markers

been conﬁrmed by others [114,115]. Such ﬁndings may sup-

such as C-reactive protein [1,96,127].

port the hypothesis that dysbiosis of the intestinal microbiota is

Another common feature is a change in Firmicutes-to-

associated with obesity. Furthermore, several experiments have

Bacteroidetes ratio. Although the above-mentioned bacterial

demonstrated the ability of antibiotics to increase the abundance

phyla were not always affected in the same way, this bacterial

of A. muciniphila in mice while reducing the incidence of dia-

ratio was always systematically decreased after RYGB. Vari-

betes (Fig. 1) [116]. On the other hand, it has been suggested that

ous levels of Proteobacteria, Firmicutes and Bacteroidetes were

early life exposure to antibiotics may contribute to body weight

conserved, regardless of patient gender, age and medication,

gain [117]. Thus, more studies are needed to elucidate the poten-

together with changes in BMI after surgery. However, while dif-

tial of antibiotics to control microbiota to improve health, as well

ferences in calorie intakes were mentioned in some studies [96],

as to demonstrate their role in the aetiology of obesity and T2D.

there was no information on any qualitative changes in the foods

Thus far, no treatments using one or several gut microbes

ingested before and after the procedure. As with HFD and/or pre-

have been used to treat T2D in humans. Although the transplan-

biotics, the composition of the diet can modify gut microbiota

tation of faecal microbiota from lean donors to obese subjects

[17,18,95], and RYGB has been shown to reduce the preference

improved insulin sensitivity [118], to the best of our knowl-

for fatty meals [129]. Thus, these effects on food choices need to

edge, this experiment has yet to be duplicated in either humans

be taken into account when investigating the impact of bariatric

or mice. Thus, ﬁnding a treatment for T2D remains a chal-

surgery on gut microbes.

lenge. Exercise and dietary interventions often yield only limited

Other studies have shown that RYGB was accompanied by

results, while pharmacological treatments are often associated

a reduction in the low-grade inﬂammation linking obesity with

with deleterious side effects.

insulin resistance [96,127]. As mentioned above, the effects of

metabolic endotoxaemia on inﬂammation and insulin resistance

6. Gastric bypass and gut microbiota: towards a

are associated with obesity [16]. Monte et al. [110] demonstrated

consensus?

that plasma LPS was reduced by 20% in patients 6 months after

RYGB and correlated with weight loss, while Trøseid et al.

Thus far, bariatric surgery is the best treatment for T2D

[130] found a reduction in endotoxaemia 1 year after the oper-

[119,120]. The most common technique is the Roux-en-Y gas-

ation, which was associated with an improvement in glycaemic

tric bypass (RYGB), which is performed on individuals with a control.

P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257 253

These data show a correlation between microbiota modiﬁ- 7. Conclusion

cation and T2D resolution following RYGB. Changes in gut

microbes were observed from 3 to 15 months after surgery, These data demonstrate that several relationships can

depending on the study [91,96,127,128]. This suggests that mod- be found between gut microbiota and glucose and energy

ulation of the gut microbial community after RYGB stabilizes homoeostasis. However, causality between the observed varia-

over time. However, it is not possible to specify the cause(s) of tions and metabolic symptoms is still unconﬁrmed. In addition,

the changes in gut microbes, or to determine whether they are the use of prebiotics as well as gastric bypass surgery has

causes or consequences of changes in host physiology. Recent revealed several putative metabolic targets (such as SCFAs, OEA

work by Liou et al. [131] in a murine model addressed this issue and PEA) and bacterial species that may be viewed as puta-

by comparing diet-induced obese mice undergoing either RYGB tive future therapeutic targets (F. prausnitzii, A. muciniphila).

or a sham operation. Initially, food consumption remained the Thus, investigations of the gut microbiota–host interactions and

same between the two groups while energy intakes decreased deciphering their symbiotic biological interactions constitute

in the RYGB mice, leading to weight loss after the surgery. important areas of research towards ﬁnding new ways to treat

Moreover, the RYGB mice exhibited improved glucose toler- obesity and its related diseases.

ance and insulin sensitivity compared with the weight-matched

sham-operated mice, conﬁrming the weight-independent effect Disclosure of interest

of RYGB on insulin resistance.

In addition, changes in gut microbiota due to RYGB are The authors declare that they have no conﬂicts of interest

conserved, with an increase in Proteobacteria (mainly Entero- concerning this article.

bacteriaceae) and Bacteroidetes, and a decrease in Firmicutes.

Moreover, Verrucomicrobia are also increased, mainly of the Acknowledgments

Akkermansia genus, which is consistent with previous ﬁndings

of the impact of A. muciniphila on glucose metabolism [1,23]. P.D. Cani is a research associate from theFRS-FNRS (Funds

Furthermore, colonization of germ-free mice with microbiota for Scientiﬁc Research) and a recipient of grants from the FNRS

from RYGB-treated animals led to losses in weight and fat mass, and FRSM (Funds for Scientiﬁc Medical Research, Belgium),

with an increasing trend towards insulin sensitivity, compared the ARC (Concerted Research Activities–French Community of

with mice colonized with microbiota from sham-operated ani- Belgium convention: 12/17-047) and the SFD (French-speaking

mals. These effects may have been due to modiﬁcation of SCFA Diabetes Society, France). Moreover, P.D. Cani is a recipient of

production in the RYGB-colonized mice [131]. These experi- the European Research Council Starting Grant 2013 (336452-

ments show that, at least in mice, the microbiota contribute to ENIGMO).

the impact of RYGB on host physiology.

The roles of Gammaproteobacteria and Enterobacteriaceae in

Appendix A. Supplementary data

the outcomes of RYGB are still controversial. Their abundance

is vastly increased by the procedure in every study thus far,

Supplementary data (French abstract) associated with this

although recent observations suggest that these bacteria might

article can be found, in the online version, at http://dx.doi.org/10.

also have a role in the development of metabolic endotoxaemia 1016/j.diabet.2014.02.004.

through their highly toxic LPSs [98,132]. The most striking evi-

dence comes from a study of colonization of germ-free mice References

with Enterobacter cloacae, the main component of the micro-

biota in a morbidly obese patient, which was sufﬁcient to induce [1] Cani PD. Metabolism in 2013: the gut microbiota manages host

obesity and insulin resistance after exposure to HFD. More- metabolism. Nat Rev Endocrinol 2014;10:74–6.

[2] Kashyap PC, Marcobal A, Ursell LK, Larauche M, Duboc H, Earle KA,

over, the E. cloacae-colonized mice had high levels of serum

et al. Complex interactions among diet, gastrointestinal transit, and gut

LPS and inﬂammation [98]. In a follow-up study, the same ani-

microbiota in humanized mice. Gastroenterology 2013;144:967–77.

mals demonstrated that the weight loss and improved insulin

[3] Everard A, Cani PD. Diabetes, obesity and gut microbiota. Best Pract Res

sensitivity after treatment with prebiotics were associated with Clin Gastroenterol 2013;27:73–83.

a decrease in Enterobacteriaceae and metabolic endotoxaemia [4] Zhao L. The gut microbiota and obesity: from correlation to causality.

[133]. Nat Rev Microbiol 2013;11:639–47.

[5] Backhed F, Ding H, Wang T, Hooper LV, Koh GY, Nagy A, et al. The

Paradoxically, RYGB is associated with both greater

gut microbiota as an environmental factor that regulates fat storage. Proc

Gammaproteobacteria abundance and less endotoxaemia. These

Natl Acad Sci U S A 2004;101:15718–23.

discordant results might be explained by the fact that differ- [6] Turnbaugh PJ, Ley RE, Mahowald MA, Magrini V, Mardis ER, Gordon JI.

ent bacterial species of the same family can have diametrically An obesity-associated gut microbiome with increased capacity for energy

harvest. Nature 2006;444:1027–31.

opposed effects on host physiology. However, any deleterious

[7] Ridaura VK, Faith JJ, Rey FE, Cheng J, Duncan AE, Kau AL, et al. Gut

long-term effects due to the increase in Gammaproteobacteria

microbiota from twins discordant for obesity modulate metabolism in

following RYGB cannot be ignored. Further studies are therefore

mice. Science 2013;341:1241214.

required to more precisely investigate the mechanisms linking [8] Tarini J, Wolever TM. The fermentable ﬁbre inulin increases postprandial

changes in gut microbiota and the improvement in metabolic serum short-chain fatty acids and reduces free-fatty acids and ghrelin in

healthy subjects. Appl Physiol Nutr Metab 2010;35:9–16.

parameters following bariatric surgery.

254 P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257

[9] Tolhurst G, Heffron H, Lam YS, Parker HE, Habib AM, Diakogiannaki E, glycemia-lowering effect of oligofructose in streptozotocin diabetic rats.

et al. Short-chain fatty acids stimulate glucagon-like peptide-1 secretion Regul Pept 2004;122:12–3.

via the G-protein-coupled receptor FFAR2. Diabetes 2012;61:364–71. [29] Delzenne NM, Cani PD, Daubioul C, Neyrinck AM. Impact of inulin

[10] Lin HV, Frassetto A, Kowalik Jr EJ, Nawrocki AR, Lu MM, Kosinski and oligofructose on gastrointestinal peptides. Br J Nutr 2005;93(Suppl.

JR, et al. Butyrate and propionate protect against diet-induced obesity 1):S157–61.

and regulate gut hormones via free fatty acid receptor 3-independent [30] Schele E, Grahnemo L, Anesten F, Hallen A, Backhed F, Jansson JO.

mechanisms. PLoS One 2012;7:e35240. The gut microbiota reduces leptin sensitivity and the expression of the

[11] Gao Z, Yin J, Zhang J, Ward RE, Martin RJ, Lefevre M, et al. Butyrate obesity-suppressing neuropeptides proglucagon (Gcg) and brain-derived

improves insulin sensitivity and increases energy expenditure in mice. neurotrophic factor (Bdnf) in the central nervous system. Endocrinology

Diabetes 2009;58:1509–17. 2013;154:3643–51.

[12] Brown AJ, Goldsworthy SM, Barnes AA, Eilert MM, Tcheang L, Daniels [31] Clayton TA. Metabolic differences underlying two distinct rat urinary

D, et al. The Orphan G protein-coupled receptors GPR41 and GPR43 are phenotypes, a suggested role for gut microbial metabolism of phenylala-

activated by propionate and other short chain carboxylic acids. J Biol nine and a possible connection to autism. FEBS Lett 2012;586:956–61.

Chem 2003;278:11312–9. [32] Park AJ, Collins J, Blennerhassett PA, Ghia JE, Verdu EF, Bercik P, et al.

[13] Kimura I, Ozawa K, Inoue D, Imamura T, Kimura K, Maeda T, et al. Altered colonic function and microbiota proﬁle in a mouse model of

The gut microbiota suppresses insulin-mediated fat accumulation via the chronic depression. Neurogastroenterol Motil 2013;25, 733–e575.

short-chain fatty acid receptor GPR43. Nature Commun 2013;4:1829. [33] Ledochowski M, Widner B, Sperner-Unterweger B, Propst T, Vogel W,

[14] De Vadder F, Kovatcheva-Datchary P, Goncalves D, Vinera J, Zitoun C, Fuchs D. Carbohydrate malabsorption syndromes and early signs of men-

Duchampt A, et al. Microbiota-generated metabolites promote metabolic tal depression in females. Dig Dis Sci 2000;45:1255–9.

beneﬁts via gut-brain neural circuits. Cell 2014;156:84–96. [34] Salinari S, Debard C, Bertuzzi A, Durand C, Zimmet P, Vidal H, et al.

[15] Beutler B. Inferences, questions and possibilities in Toll-like receptor Jejunal proteins secreted by db/db mice or insulin-resistant humans

signalling. Nature 2004;430:257–63. impair the insulin signaling and determine insulin resistance. PLoS One

[16] Cani PD, Amar J, Iglesias MA, Poggi M, Knauf C, Bastelica D, et al. 2013;8:e56258.

Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes [35] Geurts L, Muccioli GG, Delzenne NM, Cani PD. Chronic endo-

2007;56:1761–72. cannabinoid system stimulation induces muscle macrophage and lipid

[17] Cani PD, Bibiloni R, Knauf C, Waget A, Neyrinck AM, Delzenne NM, accumulation in type 2 diabetic mice independently of metabolic endo-

et al. Changes in gut microbiota control metabolic endotoxemia-induced toxaemia. PLoS One 2013;8:e55963.

inﬂammation in high-fat diet-induced obesity and diabetes in mice. Dia- [36] Harley IT, Giles DA, Pﬂuger PT, Burgess SL, Walters S, Hembree J,

betes 2008;57:1470–81. et al. Differential colonization with segmented ﬁlamentous bacteria and

[18] Cani PD, Neyrinck AM, Fava F, Knauf C, Burcelin RG, Tuohy KM, Lactobacillus murinus do not drive divergent development of diet-induced

et al. Selective increases of biﬁdobacteria in gut microﬂora improve high- obesity in C57BL/6 mice. Molecular Metabolism 2013;2:171–83.

fat-diet-induced diabetes in mice through a mechanism associated with [37] Di Marzo V. Endocannabinoids: synthesis and degradation. Rev Physiol

endotoxaemia. Diabetologia 2007;50:2374–83. Biochem Pharmacol 2008;160:1–24.

[19] Cani PD, Possemiers S, Van de WT, Guiot Y, Everard A, Rottier O, et al. [38] Mechoulam R, Gaoni Y. The absolute conﬁguration of delta-

Changes in gut microbiota control inﬂammation in obese mice through 1-tetrahydrocannabinol, the major active constituent of hashish.

a mechanism involving GLP-2-driven improvement of gut permeability. Tetrahedron Lett 1967 Mar;12:1109–11.

Gut 2009;58:1091–103. [39] Devane WA, Dysarz FA, 3rd, Johnson MR, Melvin LS, Howlett AC.

[20] Neyrinck AM, Cani PD, Dewulf EM, De BF, Bindels LB, Delzenne Determination and characterization of a cannabinoid receptor in rat brain.

NM. Critical role of Kupffer cells in the management of diet- Mol Pharmacol 1988;34:605–13.

induced diabetes and obesity. Biochem Biophys Res Commun 2009;385: [40] Muccioli GG. Endocannabinoid biosynthesis and inactivation, from sim-

351–6. ple to complex. Drug Discov Today 2010;15:474–83.

[21] Lanthier N, Molendi-Coste O, Horsmans Y, van RN, Cani PD, Leclercq [41] Hillard CJ. Biochemistry and pharmacology of the endocannabinoids

IA. Kupffer cell activation is a causal factor for hepatic insulin resistance. arachidonylethanolamide and 2-arachidonylglycerol. Prostaglandins

Am J Physiol Gastrointest Liver Physiol 2010;298:G107–16. Other Lipid Mediat 2000;61:3–18.

[22] Lanthier N, Molendi-Coste O, Cani PD, van Rooijen N, Horsmans Y, [42] Blankman JL, Simon GM, Cravatt BF. A comprehensive proﬁle of brain

Leclercq IA. Kupffer cell depletion prevents but has no therapeutic effect enzymes that hydrolyze the endocannabinoid 2-arachidonoylglycerol.

on metabolic and inﬂammatory changes induced by a high-fat diet. Chem Biol 2007;14:1347–56.

FASEB J 2011;25:4301–11. [43] Ryberg E, Larsson N, Sjogren S, Hjorth S, Hermansson NO, Leonova J,

[23] Everard A, Belzer C, Geurts L, Ouwerkerk JP, Druart C, Bindels et al. The orphan receptor GPR55 is a novel cannabinoid receptor. Br J

LB, et al. Cross-talk between Akkermansia muciniphila and intestinal Pharmacol 2007;152:1092–101.

epithelium controls diet-induced obesity. Proc Natl Acad Sci U S A [44] Hoareau L, Buyse M, Festy F, Ravanan P, Gonthier MP, Matias I,

2013;110:9066–71. et al. Anti-inﬂammatory effect of palmitoylethanolamide on human

[24] Shan M, Gentile M, Yeiser JR, Walland AC, Bornstein VU, Chen K, adipocytes. Obesity (Silver Spring) 2009;17:431–8.

et al. Mucus enhances gut homeostasis and oral tolerance by delivering [45] Pagotto U, Marsicano G, Cota D, Lutz B, Pasquali R. The emerging role of

immunoregulatory signals. Science 2013;342:447–53. the endocannabinoid system in endocrine regulation and energy balance.

[25] Kashyap PC, Marcobal A, Ursell LK, Smits SA, Sonnenburg ED, Costello Endocr Rev 2006;27:73–100.

EK, et al. Genetically dictated change in host mucus carbohydrate land- [46] Engeli S. Dysregulation of the endocannabinoid system in obesity. J

scape exerts a diet-dependent effect on the gut microbiota. Proc Natl Acad Neuroendocrinol 2008;20(Suppl. 1):110–5.

Sci U S A 2013;110:17059–64. [47] Iannotti FA, Piscitelli F, Martella A, Mazzarella E, Allara M, Palmieri V,

[26] Sommer F, Adam N, Johansson ME, Xia L, Hansson GC, Back- et al. Analysis of the “endocannabinoidome” in peripheral tissues of obese

hed F. Altered mucus glycosylation in core 1 o-glycan-deﬁcient mice Zucker rats. Prostaglandins Leukot Essent Fatty Acids 2013;89:127–35.

affects microbiota composition and intestinal architecture. PLoS One [48] Jamshidi N, Taylor DA. Anandamide administration into the ven-

2014;9:e85254. tromedial hypothalamus stimulates appetite in rats. Br J Pharmacol

[27] Cani PD, Dewever C, Delzenne NM. Inulin-type fructans modulate 2001;134:1151–4.

gastrointestinal peptides involved in appetite regulation (glucagon-like [49] Cota D, Marsicano G, Tschop M, Grubler Y, Flachskamm C, Schubert

peptide-1 and ghrelin) in rats. Br J Nutr 2004;92:521–6. M, et al. The endogenous cannabinoid system affects energy balance

[28] Cani PD, Daubioul CA, Neyrinck AM, Reusens B, Remacle C, via central orexigenic drive and peripheral lipogenesis. J Clin Invest

Catillon G, et al. Involvement of GLP-1 (7-36) amide (GLP-1) in 2003;112:423–31.

P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257 255

[50] Ravinet Trillou C, Arnone M, Delgorge C, Gonalons N, Keane P, Maffrand [69] Brown WH, Gillum MP, Lee HY, Camporez JP, Zhang XM, Jeong JK,

JP, et al. Anti-obesity effect of SR141716, a CB1 receptor antagonist, et al. Fatty acid amide hydrolase ablation promotes ectopic lipid storage

in diet-induced obese mice. Am J Physiol Regul Integr Comp Physiol and insulin resistance due to centrally mediated hypothyroidism. Proc

2003;284:R345–53. Natl Acad Sci U S A 2012;109:14966–71.

[51] Ravinet TC, Delgorge C, Menet C, Arnone M, Soubrie P. CB1 [70] Bermudez-Silva FJ, Sanchez-Vera I, Suarez J, Serrano A, Fuentes E, Juan-

cannabinoid receptor knockout in mice leads to leanness, resistance to Pico P, et al. Role of cannabinoid CB2 receptors in glucose homeostasis

diet-induced obesity and enhanced leptin sensitivity. Int J Obes Relat in rats. Eur J Pharmacol 2007;565:207–11.

Metab Disord 2004;28:640–8. [71] Furuya DT, Poletto AC, Freitas HS, Machado UF. Inhibition of

[52] Van Gaal LF, Rissanen AM, Scheen AJ, Ziegler O, Rossner S. Effects cannabinoid CB1 receptor upregulates Slc2a4 expression via nuclear

of the cannabinoid-1 receptor blocker rimonabant on weight reduction factor-kappaB and sterol regulatory element-binding protein-1 in

and cardiovascular risk factors in overweight patients: 1-year experience adipocytes. J Mol Endocrinol 2012;49:97–106.

from the RIO-Europe study. Lancet 2005;365:1389–97. [72] Cinar R, Godlewski G, Liu J, Tam J, Jourdan T, Mukhopadhyay B, et al.

[53] Tourino C, Oveisi F, Lockney J, Piomelli D, Maldonado R. FAAH deﬁ- Hepatic cannabinoid-1 receptors mediate diet-induced insulin resistance

ciency promotes energy storage and enhances the motivation for food. Int by increasing de novo synthesis of long-chain ceramides. Hepatology

J Obes (Lond) 2009. 2014;59:143–53.

[54] Thomas G, Betters JL, Lord CC, Brown AL, Marshall S, Ferguson D, [73] Jourdan T, Godlewski G, Cinar R, Bertola A, Szanda G, Liu J, et al.

et al. The serine hydrolase ABHD6 is a critical regulator of the metabolic Activation of the Nlrp3 inﬂammasome in inﬁltrating macrophages by

syndrome. Cell Rep 2013;5:508–20. endocannabinoids mediates beta cell loss in type 2 diabetes. Nat Med

[55] Liu YL, Connoley IP, Wilson CA, Stock MJ. Effects of the cannabi- 2013;19:1132–40.

noid CB1 receptor antagonist SR 141716 on oxygen consumption and [74] Li Z, Schmidt SF, Friedman JM. Developmental role for endocannabi-

soleus muscle glucose uptake in Lep(ob)/Lep(ob) mice. Int J Obes (Lond) noid signaling in regulating glucose metabolism and growth. Diabetes

2005;29:183–7. 2013;62:2359–67.

[56] Juan-Pico P, Fuentes E, Bermudez-Silva FJ, Javier Diaz-Molina F, [75] Muccioli GG, Naslain D, Backhed F, Reigstad CS, Lambert DM,

Ripoll C, Rodriguez de Fonseca F, et al. Cannabinoid receptors regulate Delzenne NM, et al. The endocannabinoid system links gut microbiota

Ca(2+) signals and insulin secretion in pancreatic beta-cell. Cell Calcium to adipogenesis. Mol Syst Biol 2010;6:392.

2006;39:155–62. [76] Geurts L, Lazarevic V, Derrien M, Everard A, Van Roye M, Knauf C,

[57] Matias I, Gonthier MP, Orlando P, Martiadis V, De PL, Cervino C, et al. et al. Altered gut microbiota and endocannabinoid system tone in obese

Regulation, function, and dysregulation of endocannabinoids in models and diabetic leptin-resistant mice: impact on apelin regulation in adipose

of adipose and beta-pancreatic cells and in obesity and hyperglycemia. J tissue. Front Microbiol 2011;2:149.

Clin Endocrinol Metab 2006;91:3171–80. [77] Cani PD, Osto M, Geurts L, Everard A. Involvement of gut microbiota

[58] Starowicz KM, Cristino L, Matias I, Capasso R, Racioppi A, Izzo AA, in the development of low-grade inﬂammation and type 2 diabetes asso-

et al. Endocannabinoid dysregulation in the pancreas and adipose tissue ciated with obesity. Gut Microb 2012;1:3.

of mice fed with a high-fat diet. Obesity (Silver Spring) 2008;16:553–65. [78] Geurts L, Neyrinck AM, Delzenne NM, Knauf C, Cani PD. Gut

[59] Bermudez-Silva FJ, Suarez J, Baixeras E, Cobo N, Bautista D, Cuesta- microbiota controls adipose tissue expansion, gut barrier and glucose

Munoz AL, et al. Presence of functional cannabinoid receptors in human metabolism: novel insights into molecular targets and interventions using

endocrine pancreas. Diabetologia 2008;51:476–87. prebiotics. Benef Microbes 2014;5:3–17.

[60] Anderson RL, Randall MD, Chan SL. The complex effects of cannabi- [79] Alhamoruni A, Wright KL, Larvin M, O’Sullivan SE. Cannabinoids

noids on insulin secretion from rat isolated islets of Langerhans. Eur J mediate opposing effects on inﬂammation-induced intestinal permeabil-

Pharmacol 2013;706:56–62. ity. Br J Pharmacol 2012;165:2598–610.

[61] D’Eon TM, Pierce KA, Roix JJ, Tyler A, Chen H, Teixeira SR. The [80] Zoppi S, Madrigal JL, Perez-Nievas BG, Marin-Jimenez I, Caso JR, Alou

role of adipocyte insulin resistance in the pathogenesis of obesity-related L, et al. Endogenous cannabinoid system regulates intestinal barrier func-

elevations in endocannabinoids. Diabetes 2008;57:1262–8. tion in vivo through cannabinoid type 1 receptor activation. Am J Physiol

[62] Bermudez-Siva FJ, Serrano A, Diaz-Molina FJ, Sanchez Vera I, Juan- Gastrointest Liver Physiol 2012;302:G565–71.

Pico P, Nadal A, et al. Activation of cannabinoid CB1 receptors induces [81] Silvestri C, Martella A, Poloso NJ, Piscitelli F, Capasso R, Izzo A, et al.

glucose intolerance in rats. Eur J Pharmacol 2006;531:282–4. Anandamide-derived prostamide F2alpha negatively regulates adipogen-

[63] Gasperi V, Fezza F, Pasquariello N, Bari M, Oddi S, Agro AF, et al. esis. J Biol Chem 2013;288:23307–21.

Endocannabinoids in adipocytes during differentiation and their role in [82] Ley RE, Backhed F, Turnbaugh P, Lozupone CA, Knight RD, Gordon

glucose uptake. Cell Mol Life Sci 2007;64:219–29. JI. Obesity alters gut microbial ecology. Proc Natl Acad Sci U S A

[64] Pagano C, Pilon C, Calcagno A, Urbanet R, Rossato M, Milan G, et al. 2005;102:11070–5.

The endogenous cannabinoid system stimulates glucose uptake in human [83] Ley RE, Turnbaugh PJ, Klein S, Gordon JI. Microbial ecology: human

fat cells via phosphatidylinositol 3-kinase and calcium-dependent mech- gut microbes associated with obesity. Nature 2006;444:1022–3.

anisms. J Clin Endocrinol Metab 2007;92:4810–9. [84] Zhang X, Shen D, Fang Z, Jie Z, Qiu X, Zhang C, et al. Human gut

[65] Song D, Bandsma RH, Xiao C, Xi L, Shao W, Jin T, et al. Acute cannabi- microbiota changes reveal the progression of glucose intolerance. PLoS

noid receptor type 1 (CB1R) modulation inﬂuences insulin sensitivity One 2013;8:e71108.

by an effect outside the central nervous system in mice. Diabetologia [85] Patil DP, Dhotre DP, Chavan SG, Sultan A, Jain DS, Lanjekar VB, et al.

2011;54:1181–9. Molecular analysis of gut microbiota in obesity among Indian individuals.

[66] Bajzer M, Olivieri M, Haas MK, Pﬂuger PT, Magrisso IJ, Foster MT, J Biosci 2012;37:647–57.

et al. Cannabinoid receptor 1 (CB1) antagonism enhances glucose uti- [86] Wu X, Ma C, Han L, Nawaz M, Gao F, Zhang X, et al. Molecular charac-

lisation and activates brown adipose tissue in diet-induced obese mice. terisation of the faecal microbiota in patients with type II diabetes. Curr

Diabetologia 2011;54:3121–231. Microbiol 2010;61:69–78.

[67] Osei-Hyiaman D, Liu J, Zhou L, Godlewski G, Harvey-White J, Jeong WI, [87] Roesch LF, Lorca GL, Casella G, Giongo A, Naranjo A, Pionzio AM,

et al. Hepatic CB1 receptor is required for development of diet-induced et al. Culture-independent identiﬁcation of gut bacteria correlated with

steatosis, dyslipidemia, and insulin and leptin resistance in mice. J Clin the onset of diabetes in a rat model. ISME J 2009;3:536–48.

Invest 2008;118:3160–9. [88] Karlsson FH, Tremaroli V, Nookaew I, Bergstrom G, Behre CJ, Fagerberg

[68] Liu J, Zhou L, Xiong K, Godlewski G, Mukhopadhyay B, Tam J, et al. B, et al. Gut metagenome in European women with normal, impaired and

Hepatic cannabinoid receptor-1 mediates diet-induced insulin resistance diabetic glucose control. Nature 2013;498:99–103.

via inhibition of insulin signaling and clearance in mice. Gastroenterology [89] Arumugam M, Raes J, Pelletier E, Le PD, Yamada T, Mende DR, et al.

2012. Enterotypes of the human gut microbiome. Nature 2011;473:174–80.

256 P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257

[90] Qin J, Li Y, Cai Z, Li S, Zhu J, Zhang F, et al. A metagenome-wide associ- [112] Yadav H, Lee JH, Lloyd J, Walter P, Rane SG. Beneﬁcial metabolic effects

ation study of gut microbiota in type 2 diabetes. Nature 2012;490:55–60. of a probiotic via Butyrate-induced GLP-1 hormone secretion. J Biol

[91] Zhang H, DiBaise JK, Zuccolo A, Kudrna D, Braidotti M, Yu Y, et al. Chem 2013;288:25088–97.

Human gut microbiota in obesity and after gastric bypass. Proc Natl Acad [113] Million M, Maraninchi M, Henry M, Armougom F, Richet H, Carrieri

Sci U S A 2009;106:2365–70. P, et al. Obesity-associated gut microbiota is enriched in Lactobacillus

[92] Santacruz A, Collado MC, Garcia-Valdes L, Segura MT, Martin-Lagos reuteri and depleted in Biﬁdobacterium animalis and Methanobrevibacter

JA, Anjos T, et al. Gut microbiota composition is associated with body smithii. Int J Obes (Lond) 2012;36:817–25.

weight, weight gain and biochemical parameters in pregnant women. Br [114] Carvalho BM, Guadagnini D, Tsukumo DM, Schenka AA, Latuf-Filho

J Nutr 2010;104:83–92. P, Vassallo J, et al. Modulation of gut microbiota by antibiotics improves

[93] Karlsson CL, Onnerfalt J, Xu J, Molin G, Ahrne S, Thorngren-Jerneck K. insulin signalling in high-fat fed mice. Diabetologia 2012;55:2823–34.

The microbiota of the gut in preschool children with normal and excessive [115] Murphy EF, Cotter PD, Hogan A, O’Sullivan O, Joyce A, Fouhy F, et al.

body weight. Obesity (Silver Spring) 2012;20:2257–61. Divergent metabolic outcomes arising from targeted manipulation of the

[94] Collado MC, Isolauri E, Laitinen K, Salminen S. Distinct composition gut microbiota in diet-induced obesity. Gut 2013;62:220–6.

of gut microbiota during pregnancy in overweight and normal-weight [116] Hansen CH, Krych L, Nielsen DS, Vogensen FK, Hansen LH, Sorensen

women. Am J Clin Nutr 2008;88:894–9. SJ, et al. Early life treatment with vancomycin propagates Akkermansia

[95] Everard A, Lazarevic V, Derrien M, Girard M, Muccioli GM, Neyrinck muciniphila and reduces diabetes incidence in the NOD mouse. Dia-

AM, et al. Responses of gut microbiota and glucose and lipid metabolism betologia 2012;55:2285–94.

to prebiotics in genetic obese and diet-induced leptin-resistant mice. Dia- [117] Cho I, Yamanishi S, Cox L, Methe BA, Zavadil J, Li K, et al. Antibiotics

betes 2011;60:2775–86. in early life alter the murine colonic microbiome and adiposity. Nature

[96] Furet JP, Kong LC, Tap J, Poitou C, Basdevant A, Bouillot JL, et al. Dif- 2012;488:621–6.

ferential adaptation of human gut microbiota to bariatric surgery-induced [118] Vrieze A, Van Nood E, Holleman F, Salojarvi J, Kootte RS, Bartelsman JF,

weight loss: links with metabolic and low-grade inﬂammation markers. et al. Transfer of intestinal microbiota from lean donors increases insulin

Diabetes 2010;59:3049–57. sensitivity in individuals with metabolic syndrome. Gastroenterology

[97] Balamurugan R, George G, Kabeerdoss J, Hepsiba J, Chandragu- 2012;143:913–6.

nasekaran AM, Ramakrishna BS. Quantitative differences in intestinal [119] Buchwald H, Avidor Y, Braunwald E, Jensen MD, Pories W, Fahrbach

Faecalibacterium prausnitzii in obese Indian children. Br J Nutr K, et al. Bariatric surgery: a systematic review and meta-analysis. JAMA

2010;103:335–8. 2004;292:1724–37.

[98] Fei N, Zhao L. An opportunistic pathogen isolated from the gut of an [120] Maggard-Gibbons M, Maglione M, Livhits M, Ewing B, Maher AR,

obese human causes obesity in germfree mice. ISME J 2013;7:880–4. Hu J, et al. Bariatric surgery for weight loss and glycemic control in

[99] Basu S, Haghiac M, Surace P, Challier JC, Guerre-Millo M, Singh K, nonmorbidly obese adults with diabetes: a systematic review. JAMA

et al. Pregravid obesity associates with increased maternal endotoxemia 2013;309:2250–61.

and metabolic inﬂammation. Obesity (Silver Spring) 2011;19:476–82. [121] Laferrere B, Reilly D, Arias S, Swerdlow N, Gorroochurn P, Bawa B, et al.

[100] Laugerette F, Vors C, Geloen A, Chauvin MA, Soulage C, Lambert- Differential metabolic impact of gastric bypass surgery versus dietary

Porcheron S, et al. Emulsiﬁed lipids increase endotoxemia: possible intervention in obese diabetic subjects despite identical weight loss. Sci

role in early postprandial low-grade inﬂammation. J Nutr Biochem Transl Med 2011;3:80re2.

2011;22:53–9. [122] Ashraﬁan H, Athanasiou T, Li JV, Bueter M, Ahmed K, Nagpal K, et al.

[101] Pussinen PJ, Havulinna AS, Lehto M, Sundvall J, Salomaa V. Endotox- Diabetes resolution and hyperinsulinaemia after metabolic Roux-en-Y

emia is associated with an increased risk of incident diabetes. Diabetes gastric bypass. Obes Rev 2011;12:e257–72.

Care 2011;34:392–7. [123] Aron-Wisnewsky J, Dore J, Clement K. The importance of the gut

[102] Hajishengallis G, Darveau RP, Curtis MA. The keystone-pathogen microbiota after bariatric surgery. Nat Rev Gastroenterol Hepatol

hypothesis. Nat Rev Microbiol 2012;10:717–25. 2012;9:590–8.

[103] Rains JL, Jain SK. Oxidative stress, insulin signaling, and diabetes. Free [124] Cani PD, Delzenne NM. Beneﬁts of bariatric surgery: an issue of

Radic Biol Med 2011;50:567–75. microbial-host metabolism interactions? Gut 2011;60:1166–7.

[104] Neyrinck AM, Van Hee VF, Bindels LB, De Backer F, Cani PD, Delzenne [125] Sayin SI, Wahlstrom A, Felin J, Jantti S, Marschall HU, Bamberg K, et al.

NM. Polyphenol-rich extract of pomegranate peel alleviates tissue inﬂam- Gut microbiota regulates bile acid metabolism by reducing the levels of

mation and hypercholesterolaemia in high-fat diet-induced obese mice: tauro-beta-muricholic acid, a naturally occurring FXR antagonist. Cell

potential implication of the gut microbiota. Br J Nutr 2013;109:802–9. Metab 2013;17:225–35.

[105] Yang CS, Lambert JD, Sang S. Antioxidative and anti-carcinogenic activ- [126] Osto M, Abegg K, Bueter M, le Roux CW, Cani PD, Lutz TA. Roux-en-

ities of tea polyphenols. Arch Toxicol 2009;83:11–21. Y gastric bypass surgery in rats alters gut microbiota proﬁle along the

[106] Shen Q, Shang N, Li P. In vitro and in vivo antioxidant activity of intestine. Physiol Behav 2013;119C:92–6.

Biﬁdobacterium animalis 01 isolated from centenarians. Curr Microbiol [127] Li JV, Ashraﬁan H, Bueter M, Kinross J, Sands C, Le Roux CW, et al.

2011;62:1097–103. Metabolic surgery profoundly inﬂuences gut microbial-host metabolic

[107] Ejtahed HS, Mohtadi-Nia J, Homayouni-Rad A, Niafar M, Asghari- cross-talk. Gut 2011;60:1214–23.

Jafarabadi M, Moﬁd V. Probiotic yogurt improves antioxidant status in [128] Graessler J, Qin Y, Zhong H, Zhang J, Licinio J, Wong ML, et al.

type 2 diabetic patients. Nutrition 2012;28:539–43. Metagenomic sequencing of the human gut microbiome before and

[108] Geurts L, Everard A, le RP, Delzenne NM, Cani PD. Ripened dairy prod- after bariatric surgery in obese patients with type 2 diabetes: correla-

ucts differentially affect hepatic lipid content and adipose tissue oxidative tion with inﬂammatory and metabolic parameters. Pharmacogenomics J

stress markers in obese and type 2 diabetic mice. J Agric Food Chem 2012. 2012.

[109] Qiao Y, Sun J, Ding Y, Le G, Shi Y. Alterations of the gut microbiota in [129] le Roux CW, Bueter M, Theis N, Werling M, Ashraﬁan H, Lowenstein

high-fat diet mice is strongly linked to oxidative stress. Applied Micro- C, et al. Gastric bypass reduces fat intake and preference. Am J Physiol

biology and Biotechnology 2013;97:1689–97. Regul Integr Comp Physiol 2011;301:R1057–66.

[110] Monte SV, Caruana JA, Ghanim H, Sia CL, Korzeniewski K, Schentag JJ, [130] Troseid M, Nestvold TK, Rudi K, Thoresen H, Nielsen EW, Lappegard

et al. Reduction in endotoxemia, oxidative and inﬂammatory stress, and KT. Plasma lipopolysaccharide is closely associated with glycemic con-

insulin resistance after Roux-en-Y gastric bypass surgery in patients with trol and abdominal obesity: evidence from bariatric surgery. Diabetes

morbid obesity and type 2 diabetes mellitus. Surgery 2012;151:587–93. Care 2013;36:3627–32.

[111] Delzenne NM, Neyrinck AM, Backhed F, Cani PD. Targeting gut micro- [131] Liou AP, Paziuk M, Luevano Jr JM, Machineni S, Turnbaugh PJ, Kaplan

biota in obesity: effects of prebiotics and probiotics. Nat Rev Endocrinol LM. Conserved shifts in the gut microbiota due to gastric bypass reduce

2011;7:639–46. host weight and adiposity. Sci Transl Med 2013;5:178ra41.

P.D. Cani et al. / Diabetes & Metabolism 40 (2014) 246–257 257

[132] Le Chatelier E, Nielsen T, Qin J, Prifti E, Hildebrand F, Falony G, et al. GPR119 (G-protein-coupled receptor 119): receptor for some eCB-related lipids

Richness of human gut microbiome correlates with metabolic markers. such as OEA and 2-OG

Nature 2013;500:541–6. FAAH (fatty acid amide hydrolase): NAE hydrolyzing enzyme

[133] Xiao S, Fei N, Pang X, Shen J, Wang L, Zhang B, et al. A gut microbiota- HU210: Synthetic cannabinoid, CB1/CB2 agonist

targeted dietary intervention for amelioration of chronic inﬂammation MAGL (monoacylglycerol lipase): 2-AG hydrolyzing enzyme

underlying metabolic syndrome. FEMS Microbiol Ecol 2013. Metagenome: genetic material recovered directly from an ecosystem without

previous distinction between organisms

Glossary Microbiota: bacterial ecosystem colonizing the human body

NPY (neuropeptide Y)/AgRP (Agouti-related peptide): two neurotransmitters

mostly known for their potent orexigenic effects

2-AG (2-arachidonoylglycerol): ester formed from arachidonic acid and glyc-

OEA (N-oleoyl ethanolamine): amide formed from oleic acid and ethanolamine,

erol; endogenous agonist of CB1 and CB2 receptors

an eCB-related lipid

␣ ␤

ABHD6 ( / -hydrolase domain 6): 2-AG hydrolyzing enzyme

PEA (N-palmitoyl ethanolamine): amide formed from palmitic acid and

AEA (N-arachidonoyl ethanolamine or anandamide): amide formed from

ethanolamine, an eCB-related lipid

arachidonic acid and ethanolamine; endogenous agonist of CB1 and CB2

POMC (proopiomelanocortin)/CART (cocaine- and amphetamine-regulated

receptors

transcript): two neurotransmitters mostly known for their potent anorex-

CB1 (Gi/o-coupled cannabinoid receptor 1): G-protein-coupled cannabinoid

igenic effects

receptor activated by endocannabinoids such as 2-AG and AEA

Prebiotics: selective stimulation of growth and/or action(s) of one or a limited

CB2 (Gi/o-coupled cannabinoid receptor 2): G-protein-coupled cannabinoid

number of microbial genus(era)/species in the gut microbiota conferring

receptor activated by endocannabinoids such as 2-AG and AEA

health beneﬁts to the host

Dysbiosis: imbalance in the microbiota due to the disturbance of several bacterial

Probiotics: microorganisms that some have claimed provide health beneﬁts

groups, causing a shift away from a healthy status

when consumed

Enterotype: classiﬁcation of a host organism based on the taxonomic structure

SCFAs (short-chain fatty acids): organic acids with two to six carbon atoms

of its microbiota [89]

produced by bacterial fermentation of indigestible ﬁbres

Germ-free mice: gnotobiotic animals harbouring no microorganisms

SEA (N-stearoyl ethanolamine): amide formed from stearic acid and

Ghrelin: gut hormone with orexigenic properties

ethanolamine, an eCB-related lipid

GLP-1 (glucagon-like peptide 1): potent gut peptide inducing glucose-

TLRs (toll-like receptors): receptors of the innate immune system that recognize

dependent stimulation of insulin secretion while suppressing glucagon

molecules broadly shared by microorganisms, but distinguishable from host secretion

molecules

Gnotobiotics: study and development of animal models characterized by deﬁned

TRPV1 (transient receptor potential cation channel, subfamily V, member

microbial communities

1): non-selective cation channel activated by endogenous ligands such as

GPR55 (G-protein-coupled receptor 55): receptor for some endocannabinoid AEA

(eCB)-related lipids