Colonic Proteome Signature in Immunoproteasome-Deficient Stressed Mice and Its Relevance for Irritable Bowel Syndrome.

A role for immunoproteasome in the regulation of intestinal permeability has been previously suggested both in mice during water avoidance stress (WAS) and in patients with irritable bowel syndrome (IBS). Here, we provide evidence that the ubiquitin-proteasome system (UPS) contributes to the pathophysiology of IBS. Indeed, we report that colonic proteome is altered in WAS mice and that ß2i subunit deficiency modifies the proteome response that is associated with a limitation of colonic hyperpermeability. Interestingly, we show specific alterations of proteins involved in UPS, mitochondrial, and energy metabolism. We also report changes in the pattern of colonic ubiquitome in diarrhea-predominant IBS (IBS-D) patients and particularly a reduced expression of ubiquitinated proteins involved in the nuclear factor-kappa B (NF-κB) inflammatory signaling pathway. All these data suggest that immunoproteasome targeting may represent a new therapeutic strategy for the treatment of IBS patients with increased intestinal permeability.

Colonic Mucosal Proteome Signature Reveals Reduced Energy Metabolism and Protein Synthesis but Activated Autophagy during Anorexia-Induced Malnutrition in Mice.

Anorexia nervosa is an eating disorder often associated with intestinal disorders. To explore the underlying mechanisms of these disorders, the colonic proteome was evaluated during activity-based anorexia. Female C57Bl/6 mice were randomized into three groups: Control, Limited Food Access (LFA) and Activity-Based Anorexia (ABA). LFA and ABA mice had a progressive limited access to food but only ABA mice had access to an activity wheel. On colonic mucosal protein extracts, a 2D PAGE-based comparative proteomic analysis was then performed and differentially expressed proteins were identified by LC-ESI-MS/MS. Twenty-seven nonredundant proteins that were differentially expressed between Control, LFA, and ABA groups were identified. ABA mice exhibited alteration of several mitochondrial proteins involved in energy metabolism such as dihydrolipoyl dehydrogenase and 3-mercaptopyruvate sulfurtransferase. In addition, a downregulation of mammalian target of rapamycin (mTOR) pathway was observed leading, on the one hand, to the inhibition of protein synthesis, evaluated by puromycin incorporation and mediated by the increased phosphorylation of eukaryotic elongation factor 2, and on the other hand, to the activation of autophagy, assessed by the increase of the marker of autophagy, form LC3-phosphatidylethanolamine conjugate/Cytosolic form of Microtubule-associated protein 1A/1B light chain 3 (LC3II/LC3I) ratio. Colonic mucosal proteome is altered during ABA suggesting a downregulation of energy metabolism. A decrease of protein synthesis and an activation of autophagy were also observed mediated by mTOR pathway.

Glutamine enema regulates colonic ubiquitinated proteins but not proteasome activities during TNBS-induced colitis leading to increased mitochondrial activity.

Ubiquitin proteasome system contributes to the regulation of intestinal inflammatory response as its inhibition is associated with tissue damage improvement. We aimed to evaluate whether glutamine is able to limit inflammation by targeting ubiquitin proteasome system in experimental colitis. Colitis was induced in male rats by intrarectal instillation of 2-4-6-trinitrobenzen sulfonic acid (TNBS) at day 1. From day 2 to day 6, rats daily received either an intrarectal instillation of PBS (TNBS/PBS group) or glutamine (TNBS/Gln). Rats were euthanized at day 7 and colonic samples were taken to evaluate ubiqutinated proteins by proteomic approach combining 2D electrophoresis and immunoblots directed against ubiquitin. Results were then confirmed by evaluating total expression of proteins and mRNA levels. Survival rate, TNFα, and IL-1ß mRNA were improved in TNBS/Gln compared with TNBS/PBS (p < 0.05). Proteasome activities were affected by TNBS but not by glutamine. We identified eight proteins that were less ubiquitinated in TNBS/PBS compared with controls with no effect of glutamine. Four proteins were more ubiquitinated in TNBS/PBS group and restored in TNBS/Gln group. Finally, 12 ubiquitinated proteins were only affected by glutamine. Among proteins affected by glutamine, eight proteins (GFPT1, Gapdh, Pkm2, LDH, Bcat2, ATP5a1, Vdac1, and Vdac2) were involved in metabolic pathways. In conclusion, glutamine may regulate ubiquitination process during intestinal inflammation.

Enteral glutamine infusion modulates ubiquitination of heat shock proteins, Grp-75 and Apg-2, in the human duodenal mucosa.

Glutamine, the most abundant amino acid in the human body, plays several important roles in the intestine. Previous studies showed that glutamine may affect protein expression by regulating ubiquitin-proteasome system. We thus aimed to evaluate the effects of glutamine on ubiquitinated proteins in human duodenal mucosa. Five healthy male volunteers were included and received during 5 h, on two occasions and in a random order, either an enteral infusion of maltodextrins alone (0.25 g kg(-1) h(-1), control), mimicking carbohydrate-fed state, or maltodextrins with glutamine (0.117 g kg(-1) h(-1), glutamine). Endoscopic duodenal biopsies were then taken. Total cellular protein extracts were separated by 2D gel electrophoresis and analyzed by an immunodetection using anti-ubiquitin antibody. Differentially ubiquitinated proteins were then identified by liquid chromatography-electrospray ionization MS/MS. Five proteins were differentially ubiquitinated between control and glutamine conditions. Among these proteins, we identified two chaperone proteins, Grp75 and hsp74. Grp75 was less ubiquitinated after glutamine infusion compared with control. In contrast, hsp74, also called Apg-2, was more ubiquitinated after glutamine. In conclusion, we provide evidence that glutamine may regulate ubiquitination processes of specific proteins, i.e., Grp75 and Apg-2. Grp75 has protective and anti-inflammatory properties, while Apg-2 indirectly regulates stress-induced cell survival and proliferation through interaction with ZO-1. Further studies should confirm these results in stress conditions.

Sex-dependent circadian alterations of both central and peripheral clock genes expression and gut-microbiota composition during activity-based anorexia in mice.

RATIONALE: Patients with anorexia nervosa (AN) often present sleep disorders and circadian hormonal dysregulation. The role of the microbiota-gut-brain axis in the regulation of feeding behavior has emerged during the last decades but its relationships with the circadian rhythm remains poorly documented. Thus, we aimed to characterize the circadian clock genes expression in peripheral and central tissues in the activity-based anorexia mouse model (ABA), as well as the dynamics of the gut-microbiota composition. METHODS: From day 1 to day 17, male and female C57Bl/6 mice were submitted or not to the ABA protocol (ABA and control (CT) groups), which combines a progressive limited access to food and a free access to a running wheel. At day 17, fasted CT and ABA mice were euthanized after either resting (EoR) or activity (EoA) phase (n = 10-12 per group). Circadian clock genes expression was assessed by RT-qPCR on peripheral (liver, colon and ileum) and central (hypothalamic suprachiasmatic nucleus or SCN) tissues. Cecal bacterial taxa abundances were evaluated by qPCR. Data were compared by two-way ANOVA followed by post-tests. RESULTS: ABA mice exhibited a lower food intake, a body weight loss and an increase of diurnal physical activity that differ according with the sex. Interestingly, in the SCN, only ABA female mice exhibited altered circadian clock genes expression (Bmal1, Per1, Per2, Cry1, Cry2). In the intestinal tract, modification of clock genes expression was also more marked in females compared to males. For instance, in the ileum, female mice showed alteration of Bmal1, Clock, Per1, Per2, Cry1, Cry2 and Rev-erbα mRNA levels, while only Per2 and Cry1 mRNAs were affected by ABA model in males. By contrast, in the liver, clock genes expression was more markedly affected in males compared to females in response to ABA. Finally, circadian variations of gut-bacteria abundances were observed in both male and female mice and sex-dependent alteration were observed in response to the ABA model. CONCLUSIONS: This study shows that alteration of circadian clock genes expression at both peripheral and central levels occurs in response to the ABA model. In addition, our data underline that circadian variations of the gut-microbiota composition are sex-dependent.

Anorexia nervosa is an eating disorder with a female predominance. However, the underlying pathophysiological mechanisms are still incompletely understood. Patients with anorexia nervosa often show alterations in circadian rhythm, including sleep disorders and modifications in hormone circadian rhythm. The circadian rhythm is controlled in the central nervous system, particularly in the suprachiasmatic nucleus, but clocks have also been described in peripheral tissues. To better understand the putative role of circadian rhythm in the pathophysiology of anorexia nervosa, we have conducted an experimental study in a rodent model of anorexia nervosa called "activity-based anorexia" on both males and females. Interestingly, we observed that the expression of genes involved in the circadian rhythm is affected by the activity-based anorexia model in both the suprachiasmatic nucleus and peripheral tissues, such as the small intestine and liver. In addition, gut­microbiota also shows circadian variation. Interestingly, the anorexia-induced alterations of circadian variations (clock genes expression and gut­microbiota composition) are sex- and tissue-dependent. For instance, female mice exhibited more marked alterations in the ileum, whereas, in males, modifications were more pronounced in the liver. This study highlights sex-dependent alterations of circadian clock genes expression and of gut­microbiota in response to the anorexia rodent model. Further experiments should be performed to investigate the contribution of these mechanisms in the etiology of anorexia nervosa and the higher prevalence in females.

Evaluation of ubiquitinated proteins by proteomics reveals the role of the ubiquitin proteasome system in the regulation of Grp75 and Grp78 chaperone proteins during intestinal inflammation.

The ubiquitin proteasome system (UPS) is the major pathway of intracellular protein degradation and may be involved in the pathophysiology of inflammatory bowel diseases or irritable bowel syndrome. UPS specifically degrades proteins tagged with an ubiquitin chain. We aimed to identify polyubiquitinated proteins during inflammatory response in intestinal epithelial HCT-8 cells by a proteomic approach. HCT-8 cells were incubated with interleukin 1ß, tumor necrosis factor-α, and interferon-γ for 2 h. Total cellular protein extracts were separated by 2D gel electrophoresis and analyzed by an immunodetection using antiubiquitin antibody. Differential ubiquitinated proteins were then identified by LC-ESI MS/MS. Seven proteins were differentially ubiquitinated between control and inflammatory conditions. Three of them were chaperones: Grp75 and Hsc70 were more ubiquitinated (p < 0.05) and Grp78 was less ubiquitinated (p < 0.05) under inflammatory conditions. The results for Grp75 and Grp78 were then confirmed in HCT-8 cells and in 2-4-6-trinitrobenzen sulfonic acid induced colitis in rats mimicking inflammatory bowel disease by immunoprecipitation. No difference was observed in irritable bowel syndrome like model. In conclusion, we showed that a proteomic approach is suitable to identify ubiquitinated proteins and that UPS-regulated expression of Grp75 and Grp78 may be involved in inflammatory response. Further studies should lead to the identification of ubiquitin ligases responsible for Grp75 and Grp78 ubiquitination.

Regulation of intestinal protein metabolism by amino acids.

Gut homeostasis plays a major role in health and may be regulated by quantitative and qualitative food intake. In the intestinal mucosa, an intense renewal of proteins occurs, at approximately 50% per day in humans. In some pathophysiological conditions, protein turnover is altered and may contribute to intestinal or systemic diseases. Amino acids are key effectors of gut protein turnover, both as constituents of proteins and as regulatory molecules limiting intestinal injury and maintaining intestinal functions. Many studies have focused on two amino acids: glutamine, known as the preferential substrate of rapidly dividing cells, and arginine, another conditionally essential amino acid. The effects of glutamine and arginine on protein synthesis appear to be model and condition dependent, as are the involved signaling pathways. The regulation of gut protein degradation by amino acids has been minimally documented until now. This review will examine recent data, helping to better understand how amino acids regulate intestinal protein metabolism, and will explore perspectives for future studies.

Intestinal Epithelial Toll-like Receptor 4 Deficiency Modifies the Response to the Activity-Based Anorexia Model in a Sex-Dependent Manner: A Preliminary Study.

The role of microbiota in eating disorders has recently emerged. Previous data reported that lipopolysaccharides induce anorexia and a decrease of body weight through the activation of toll-like receptor 4 (TLR4). In the activity-based anorexia (ABA) mouse model, an increase of TLR4 expression in intestinal epithelial cells (IEC) has been described. We thus aimed to characterize the role of TLR4 in IEC in the ABA model in male and female mice. For this purpose, Vill-CreERT2-TLR4 LoxP, which are depleted for TLR4 in IEC in response to 4-OH tamoxifen, were submitted (ABA) or not (CT) to the ABA procedure that combined free access to a running wheel and progressive time-limited access to food. We thus compared CT and ABA TLR4IEC-/- mice to CT and ABA TLR4IEC+/+ mice. In response to the ABA model, TLR4IEC+/+ male and female mice exhibited a body weight loss associated to a decrease of lean mass. In TLR4IEC-/- male mice, body weight loss was delayed and less pronounced compared to TLR4IEC+/+ male mice. We did not observe a difference of body weight loss in female mice. The body composition remained unchanged between TLR4IEC-/- and TLR4IEC+/+ mice in both sexes. In both sexes, ABA TLR4IEC+/+ mice exhibited an increase of food-anticipatory activity, as well as an increase of immobility time during the open field test. However, female TLR4IEC-/- mice showed a decrease of the time spent at the centre and an increase of the time spent at the periphery of the open field area, whereas we did not observe differences in the male mice. In conclusion, the invalidation of TLR4 in IEC modified the response to the ABA model in a sex-dependent manner. Further studies should decipher the underlying mechanisms.

Dietary salt exacerbates intestinal fibrosis in chronic TNBS colitis via fibroblasts activation.

Intestinal fibrosis is a frequent complication in inflammatory bowel diseases (IBD). It is a challenge to identify environmental factors such as diet that may be driving this risk. Intestinal fibrosis result from accumulation of extracellular matrix (ECM) proteins secreted by myofibroblasts. Factors promoting intestinal fibrosis are unknown, but diet appears to be a critical component in its development. Consumption of salt above nutritional recommendations can exacerbate chronic inflammation. So far, high salt diet (HSD) have not been thoroughly investigated in the context of intestinal fibrosis associated to IBD. In the present study, we analyze the role of dietary salt in TNBS chronic colitis induced in rat, an intestinal fibrosis model, or in human colon fibroblast cells. Here, we have shown that high-salt diet exacerbates undernutrition and promoted ECM-associated proteins in fibroblasts. Taken together, our results suggested that dietary salt can activate intestinal fibroblasts, thereby contributing to exacerbation of intestinal fibrosis. Dietary salt may be considered as a putative environmental factor that drives intestinal fibrosis risk.

Magnetic resonance colonography assessment of acute trinitrobenzene sulfonic acid colitis in pre-pubertal rats.

Pre-pubertal murine models of acute colitis are lacking. Magnetic resonance colonography (MRC) is a promising minimally invasive tool to assess colitis. We aimed to: 1/ Adapt a model of acute experimental colitis to pre-pubertal rats and determine whether MRC characteristics correlate with histological inflammation. 2/ Test this model by administering a diet supplemented in transforming growth factor ß2 to reverse inflammation. Twenty-four rats were randomized at weaning to one of 3 groups: Trinitrobenzene Sulfonic Acid (TNBS) group (n = 8) fed a standard diet, that received an intra-rectal 60 mg/kg dose of TNBS-ethanol; Control group (n = 8) fed standard diet, that received a dose of intra-rectal PBS; TNBS+MODULEN group (n = 8) that received a dose of TNBS and were exclusively fed MODULEN-IBD® after induction of colitis. One week after induction of colitis, rats were assessed by MRC, colon histopathology and inflammation markers (Interleukin 1ß, Tumor necrosis factor α, Nitric Oxide Synthase 2 and Cyclooxygenase 2). TNBS induced typical features of acute colitis on histopathology and MRC (increased colon wall thickness, increased colon intensity on T2-weighted images, target sign, ulcers). Treatment with MODULEN-IBD® did not reduce signs of colitis on MRC. Inflammatory marker expression did not differ among study groups.

Gut microbiota depletion affects nutritional and behavioral responses to activity-based anorexia model in a sex-dependent manner.

BACKGROUND & AIMS: In the last decade, the role of the microbiota-gut-brain axis in eating behavior and anxiety-depressive disorders has gained increasing attention. Although a gut microbiota dysbiosis has been reported in anorectic patients, its pathophysiological role remains poorly understood. Thus, we aimed to characterize the potential role of gut microbiota by evaluating the effects of its depletion in the Activity-Based Anorexia (ABA) mouse model both in male and female mice. METHODS: Male and female C57Bl/6 mice were submitted (ABA group) or not (CT group) to the ABA protocol, which combines access to a running wheel with a progressive limited food access. Gut microbiota was previously depleted or not by a cocktail of antibiotics (ATB) delivered by oral gavages. We monitored body composition, anxiety-like behavior, leptin and adiponectin plasma levels, hypothalamic and hippocampal neuropeptides mRNA levels, as well as dopamine (DRD) and serotonin (5HT1 and 4) receptors mRNA expression. RESULTS: In response to the ABA model, the body weight loss was less pronounced in ATB-treated ABA compared to untreated ABA, while food intake remained unaffected by ATB treatment. ATB-treated ABA exhibited increased fat mass and decreased lean mass compared to untreated ABA both in male and female mice, whereas but plasma adipokine concentrations were affected in a sex-dependent manner. Only male ABA mice showed a reduced anticipatory physical activity in response to ATB treatment. Similarly, anxiety-like behavior was mainly affected in ATB-treated ABA male mice compared to ATB-treated ABA female mice, which was associated with male-specific alterations of hypothalamic CRH mRNA and hippocampal DRD and 5-HT1A mRNA levels. CONCLUSIONS: Our study provides evidence that ATB-induced gut microbiota depletion triggers alterations of nutritional and behavioral responses to the activity-based anorexia model in a sex-dependent manner.

Influence of Glutamine and Branched-Chain Amino Acids Supplementation during Refeeding in Activity-Based Anorectic Mice.

BACKGROUND: Optimizing the refeeding of patients with anorexia nervosa remains important to limit somatic complications of malnutrition, as well as to avoid disease relapses by targeting persistent mood and intestinal disorders. We aimed to evaluate the effects of glutamine (Gln) and branched-chain amino acids (BCAA) supplementation during refeeding in activity-based anorectic (ABA) mice. METHOD: Male C57Bl/6 mice were randomized in control and ABA groups. Once ABA-induced malnutrition was established, mice were progressively refed or not. Refed mice had free access to drinking water supplemented or not with 1% Gln or 2.5% BCAA for 10 days. RESULTS: A progressive refeeding was associated with a partial restoration of body weight and lean mass, while a fat mass rebound was observed. In addition, refeeding restored glucose and leptin. Gln did not affect these parameters, while BCAA tended to increase body weight, fat mass, and glycaemia. In the colon, refeeding improved total protein synthesis and restored the LC3II/LC3I ratio, a marker of autophagy. Gln supplementation enhanced colonic protein synthesis, which was associated with an increased p-p70S6kinase/p70S6kinase ratio, whereas these effects were blunted by BCCA supplementation. CONCLUSIONS: In ABA mice, Gln and BCAA supplementations during a progressive refeeding fail to restore body weight and lean mass. However, Gln supplementation improves total colonic protein synthesis conversely to BCAA. Further studies are needed to decipher the underlying mechanisms involved in these opposite results.

Stress-induced intestinal barrier dysfunction is exacerbated during diet-induced obesity.

Obesity and irritable bowel syndrome (IBS) are two major public health issues. Interestingly previous data report a marked increase of IBS prevalence in morbid obese subjects compared with non-obese subjects but underlying mechanisms remain unknown. Obesity and IBS share common intestinal pathophysiological mechanisms such as gut dysbiosis, intestinal hyperpermeability and low-grade inflammatory response. We thus aimed to evaluate the link between obesity and IBS using different animal models. Male C57Bl/6 mice received high fat diet (HFD) for 12 weeks and were then submitted to water avoidance stress (WAS). In response to WAS, HFD mice exhibited higher intestinal permeability and plasma corticosterone concentration than non-obese mice. We were not able to reproduce a similar response both in ob/ob mice and in leptin-treated non-obese mice. In addition, metformin, a hypoglycemic agent, limited fasting glycaemia both in unstressed and WAS diet-induced obese mice but only partially restored colonic permeability in unstressed HFD mice. Metformin failed to improve intestinal permeability in WAS HFD mice. Finally, cecal microbiota transplantation from HFD mice in antibiotics-treated recipient mice did not reproduce the effects observed in stressed HFD mice. In conclusion, stress induced a more marked intestinal barrier dysfunction in diet-induced obese mice compared with non-obese mice that seems to be independent of leptin, glycaemia and gut microbiota. These data should be further confirmed and the role of the dietary composition should be studied.

Characterizing the metabolic perturbations induced by activity-based anorexia in the C57Bl/6 mouse using 1H NMR spectroscopy.

BACKGROUND & AIMS: Anorexia nervosa (AN) is a severe psychological and potentially life-threatening eating disorder. The activity-based anorexia (ABA) mouse model is commonly used to investigate physiological abnormalities associated with this disorder. Characterizing the holistic biochemical alterations induced by anorexia is essential to understanding AN pathophysiology as well as to define biomarkers for prognosis. METHODS: To unravel the adaptive biochemical mechanisms occurring in this model in response to self-starvation, the urinary, plasma and fecal metabolic phenotypes of mice under different experimental conditions were compared. This included control mice with and without physical activity (CT and CTPA mice), a group with limited food access (LFA), and a group with both limited food access and physical activity (ABA). Using 1H nuclear magnetic resonance (NMR) spectroscopy, several biochemical perturbations were observed. RESULTS: Physical activity altered the abundance of 14 fecal metabolites, including those involved in gut microbial metabolism and proteolysis. Food restriction disrupted a wide range of metabolic pathways including gut microbial metabolism, proteolysis and fatty acid breakdown (24 urinary and 6 plasma metabolites). The combined impact of food restriction and physical activity resulted in the same pattern of metabolic disruption (24 urine, 6 plasma). CONCLUSIONS: This work defined the metabolic signatures of ABA mice and provides novel insights into biological adaptations of mice in response to both food restriction and physical activity. These results should be further confirmed in AN patients.

Glutamine, but not Branched-Chain Amino Acids, Restores Intestinal Barrier Function during Activity-Based Anorexia.

BACKGROUND: During activity-based anorexia (ABA) in mice, enhanced paracellular permeability and reduced protein synthesis have been shown in the colon while the gut-brain axis has received increasing attention in the regulation of intestinal and mood disorders that frequently occur during anorexia nervosa, a severe eating disorder for which there is no specific treatment. In the present study, we assessed the effects of oral glutamine (Gln) or branched-chain amino acids (BCAA) supplementation during ABA to target intestinal functions, body composition and feeding behavior. METHODS: C57BL/6 male mice were randomized in Control (CTRL) and ABA groups. After ABA induction, mice received, or not, either 1% Gln or 2.5% BCAA (Leu, Ile, Val) for one week in drinking water. RESULTS: Neither Gln nor BCAA supplementation affected body weight and body composition, while only Gln supplementation slightly increased food intake. ABA mice exhibited increased paracellular permeability and reduced protein synthesis in the colonic mucosa. Oral Gln restored colonic paracellular permeability and protein synthesis and increased the mucin-2 mRNA level, whereas BCAA did not affect colonic parameters. CONCLUSION: In conclusion, oral Gln specifically improves colonic response during ABA. These data should be further confirmed in AN patients.

Alterations of proteome, mitochondrial dynamic and autophagy in the hypothalamus during activity-based anorexia.

Restrictive anorexia nervosa is associated with reduced eating and severe body weight loss leading to a cachectic state. Hypothalamus plays a major role in the regulation of food intake and energy homeostasis. In the present study, alterations of hypothalamic proteome and particularly of proteins involved in energy and mitochondrial metabolism have been observed in female activity-based anorexia (ABA) mice that exhibited a reduced food intake and a severe weight loss. In the hypothalamus, mitochondrial dynamic was also modified during ABA with an increase of fission without modification of fusion. In addition, increased dynamin-1, and LC3II/LC3I ratio signed an activation of autophagy while protein synthesis was increased. In conclusion, proteomic analysis revealed an adaptive hypothalamic protein response in ABA female mice with both altered mitochondrial response and activated autophagy.

Sex differences in response to activity-based anorexia model in C57Bl/6 mice.

Anorexia nervosa is a severe eating disorder often associated with physical hyperactivity and is more frequently observed in female sex. Activity-Based Anorexia (ABA) model combines physical activity (PA) and reduced food intake and thus allows a better understanding of the mechanisms underlying anorexia nervosa. We aimed to assess sex differences in response to ABA model in C57Bl/6 mice. Twenty four male and 16 female C57BL/6 mice were studied. ABA mice were placed in individual cages with a continuously recorded activity wheel. ABA mice had a progressive limited food access from 6h/day (day 6) to 3h/day (day 9) until the end of the protocol (day 17). Body weight and food intake were daily measured. We studied physical activity during 24h, during the dark phase (D-PA) and the light phase (L-PA). We also evaluated the feeding anticipatory physical activity (A-PA), the physical activity during food intake period (FI-PA) and the post-prandial physical activity (PP-PA). We observed 16.7% of mortality in males (4 out of 24 mice) during ABA protocol while no female mice died (p=0.09). At day 17, food intake was significantly higher in females than in males (p<0.05) that was associated with a lower body weight loss than in females (p<0.05). Before limited food access, no gender differences in wheel running activity were observed. From day 9, A-PA significantly increased over time in males (p<0.05 vs females) while females exhibited higher FI-PA and PP-PA (p<0.05 vs males). Correlations between wheel running activities and, respectively, food intake and body weight loss showed gender differences, in particularly for L-PA and A-PA. Our results suggest a greater susceptibility of male mice to develop ABA, males and females exhibit different patterns of physical activity after limitation of food access. Underlying mechanisms should be further investigated.

A role for intestinal TLR4-driven inflammatory response during activity-based anorexia.

Anorexia nervosa (AN) is associated with low-grade systemic inflammation and altered gut microbiota. However, the molecular origin of the inflammation remains unknown. Toll-like receptors are key regulators of innate immune response and their activation seems also to be involved in the control of food intake. We used activity-based anorexia (ABA) model to investigate the role of TLR4 and its contribution in anorexia-associated low-grade inflammation. Here, we found that ABA affected early the intestinal inflammatory status and the hypothalamic response. Indeed, TLR4 was upregulated both on colonic epithelial cells and intestinal macrophages, leading to elevated downstream mucosal cytokine production. These mucosal changes occurred earlier than hypothalamic changes driving to increased levels of IL-1ß and IL-1R1 as well as increased levels of plasma corticosterone. Paradoxically, TLR4-deficient mice exhibited greater vulnerability to ABA with increased mortality rate, suggesting a major contribution of TLR4-mediated responses during ABA-induced weight loss.

Maintaining physical activity during refeeding improves body composition, intestinal hyperpermeability and behavior in anorectic mice.

A role of gut-brain axis emerges in the pathophysiology of anorexia nervosa and maintaining adapted physical activity during refeeding remains discussed. We aimed to assess gastrointestinal protein metabolism and investigate the contribution of physical activity during refeeding in C57BL/6 mice with activity-based anorexia (ABA). ABA mice exhibited lower body weight and food intake with increase of lean mass/fat mass ratio and fat oxidation. Colonic permeability was increased in ABA. Ad libitum food access was then restored and ABA group was divided into two subgroups, with access to running wheel (ABA-PA) or not (ABA-NPA). After refeeding, fat free mass was completely restored only in ABA-PA. Colonic permeability was enhanced in ABA-NPA. Finally, muscle kynurenine conversion into kynurenic acid was lower in ABA-NPA who also exhibited altered behavior. Maintaining physical activity during refeeding may thus limit colonic hyperpermeability and improve behavior in anorectic mice.

Gut Commensal E. coli Proteins Activate Host Satiety Pathways following Nutrient-Induced Bacterial Growth.

The composition of gut microbiota has been associated with host metabolic phenotypes, but it is not known if gut bacteria may influence host appetite. Here we show that regular nutrient provision stabilizes exponential growth of E. coli, with the stationary phase occurring 20 min after nutrient supply accompanied by bacterial proteome changes, suggesting involvement of bacterial proteins in host satiety. Indeed, intestinal infusions of E. coli stationary phase proteins increased plasma PYY and their intraperitoneal injections suppressed acutely food intake and activated c-Fos in hypothalamic POMC neurons, while their repeated administrations reduced meal size. ClpB, a bacterial protein mimetic of α-MSH, was upregulated in the E. coli stationary phase, was detected in plasma proportional to ClpB DNA in feces, and stimulated firing rate of hypothalamic POMC neurons. Thus, these data show that bacterial proteins produced after nutrient-induced E. coli growth may signal meal termination. Furthermore, continuous exposure to E. coli proteins may influence long-term meal pattern.