Intestinal sodium/glucose cotransporter 3 expression is epithelial and downregulated in obesity.

AIM: We aimed to determine whether the sodium/glucose cotransporter family member SGLT3, a proposed glucose sensor, is expressed in the intestine and/or kidney, and if its expression is altered in mouse models of obesity and in humans before and after weight-loss surgery. MAIN METHODS: We used in-situ hybridization and quantitative PCR to determine whether the Sglt3 isoforms 3a and 3b were expressed in the intestine and kidney of C57, leptin-deficient ob/ob, and diabetic BTBR ob/ob mice. Western blotting and immunohistochemistry were also used to assess SGLT3 protein levels in jejunal biopsies from obese patients before and after weight-loss Roux-en-Y gastric bypass surgery (RYGB), and in lean healthy controls. KEY FINDINGS: Sglt3a/3b mRNA was detected in the small intestine (duodenum, jejunum and ileum), but not in the large intestine or kidneys of mice. Both isoforms were detected in epithelial cells (confirmed using intestinal organoids). Expression of Sglt3a/3b mRNA in duodenum and jejunum was significantly lower in ob/ob and BTBR ob/ob mice than in normal-weight littermates. Jejunal SGLT3 protein levels in aged obese patients before RYGB were lower than in lean individuals, but substantially upregulated 6 months post-RYGB. SIGNIFICANCE: Our study shows that Sglt3a/3b is expressed primarily in epithelial cells of the small intestine in mice. Furthermore, we observed an association between intestinal mRNA Sglt3a/3b expression and obesity in mice, and between jejunal SGLT3 protein levels and obesity in humans. Further studies are required to determine the possible role of SGLT3 in obesity.

A modeling approach for compounds affecting body composition.

Body composition and body mass are pivotal clinical endpoints in studies of welfare diseases. We present a combined effort of established and new mathematical models based on rigorous monitoring of energy intake (EI) and body mass in mice. Specifically, we parameterize a mechanistic turnover model based on the law of energy conservation coupled to a drug mechanism model. Key model variables are fat-free mass (FFM) and fat mass (FM), governed by EI and energy expenditure (EE). An empirical Forbes curve relating FFM to FM was derived experimentally for female C57BL/6 mice. The Forbes curve differs from a previously reported curve for male C57BL/6 mice, and we thoroughly analyse how the choice of Forbes curve impacts model predictions. The drug mechanism function acts on EI or EE, or both. Drug mechanism parameters (two to three parameters) and system parameters (up to six free parameters) could be estimated with good precision (coefficients of variation typically <20 % and not greater than 40 % in our analyses). Model simulations were done to predict the EE and FM change at different drug provocations in mice. In addition, we simulated body mass and FM changes at different drug provocations using a similar model for man. Surprisingly, model simulations indicate that an increase in EI (e.g. 10 %) was more efficient than an equal lowering of EI. Also, the relative change in body mass and FM is greater in man than in mouse at the same relative change in either EI or EE. We acknowledge that this assumes the same drug mechanism impact across the two species. A set of recommendations regarding the Forbes curve, vehicle control groups, dual action on EI and loss, and translational aspects are discussed. This quantitative approach significantly improves data interpretation, disease system understanding, safety assessment and translation across species.

Structural and pharmacological characterization of novel potent and selective monoclonal antibody antagonists of glucose-dependent insulinotropic polypeptide receptor.

Glucose-dependent insulinotropic polypeptide (GIP) is an endogenous hormonal factor (incretin) that, upon binding to its receptor (GIPr; a class B G-protein-coupled receptor), stimulates insulin secretion by beta cells in the pancreas. There has been a lack of potent inhibitors of the GIPr with prolonged in vivo exposure to support studies on GIP biology. Here we describe the generation of an antagonizing antibody to the GIPr, using phage and ribosome display libraries. Gipg013 is a specific competitive antagonist with equally high potencies to mouse, rat, dog, and human GIP receptors with a Ki of 7 nm for the human GIPr. Gipg013 antagonizes the GIP receptor and inhibits GIP-induced insulin secretion in vitro and in vivo. A crystal structure of Gipg013 Fab in complex with the human GIPr extracellular domain (ECD) shows that the antibody binds through a series of hydrogen bonds from the complementarity-determining regions of Gipg013 Fab to the N-terminal α-helix of GIPr ECD as well as to residues around its highly conserved glucagon receptor subfamily recognition fold. The antibody epitope overlaps with the GIP binding site on the GIPr ECD, ensuring competitive antagonism of the receptor. This well characterized antagonizing antibody to the GIPr will be useful as a tool to further understand the biological roles of GIP.

Interplay between T(h)1 and T(h)17 effector T-cell pathways in the pathogenesis of spontaneous colitis and colon cancer in the Gαi2-deficient mouse.

Gαi2-deficient mice spontaneously develop colitis. Using xMAP technology and RT-PCR, we investigated cytokine/chemokine profiles during histologically defined phases of disease: (i) no/mild, (ii) moderate, (iii) severe colitis without dysplasia/cancer and (iv) severe colitis with dysplasia/cancer, compared with age-matched wild-type (WT) littermates. Colonic dysplasia was observed in 4/11 mice and cancer in 1/11 mice with severe colitis. The histology correlated with progressive increases in colon weight/cm and spleen weight, and decreased thymus weight, all more advanced in mice with dysplasia/cancer. IL-1ß, IL-6, IL-12p40, IL-17, TNF-α, CCL2 and CXCL1 protein levels in colons, but not small intestines increased with colitis progression and were significantly increased in mice with moderate and severe colitis compared with WT mice, irrespective of the absence/presence of dysplasia/cancer. CCL5 did not change during colitis progression. Colonic IL-17 transcription increased 40- to 70-fold in all stages of colitis, whereas IFN-γ mRNA was gradually up-regulated 12- to 55-fold with colitis progression, and further to 62-fold in mice with dysplasia/cancer. IL-27 mRNA increased 4- to 15-fold during the course of colitis, and colonic IL-21 transcription increased 3-fold in mice with severe colitis, both irrespective of the absence/presence of dysplasia/cancer. FoxP3 transcription was significantly enhanced (3.5-fold) in mice with moderate and severe colitis, but not in mice with dysplasia/cancer, compared with WT mice. Constrained correspondence analysis demonstrated an association between increased protein levels of TNF-α, CCL2, IL-1ß, IL-6 and CXCL1 and dysplasia/cancer. In conclusion, colonic responses are dominated by a mixed T(h)1/T(h)17 phenotype, with increasing T(h)1 cytokine transcription with progression of colitis in Gαi2(-/-) mice.

The application and relevance of ex vivo culture systems for assessment of IBD treatment in murine models of colitis.

The aim of this study was to investigate the relevance of mouse ex vivo cultures as a first screening model for new therapeutic agents of Inflammatory Bowel Disease (IBD). Two murine models (dextran sodium sulphate (DSS)-induced colitis and Galphai2-deficient mice) and two anti-inflammatory agents (methyl-prednisolone and the proteasome inhibitor MG132) were evaluated. The in vivo effects of methyl-prednisolone were assessed in both models. Ex vivo colonic tissue from both mouse models were cultured in the presence or absence of the drugs and TaqMan Low-Density arrays were used to assess the regulation of inflammatory genes before and after drug treatment. Colitis induced a similar inflammatory gene profile in both mouse models in in vivo studies and in ex vivo cultures. The differences encountered reflected the different phases of colitis in the models, e.g. innate cytokine/chemokine profile in the DSS model and T cell related markers in Galphai2-deficient mice. After steroid treatment, a similar pattern of genes was suppressed in the two mouse models. We confirmed the suppression of inflammatory gene expression for IL-1beta, IL-6 and iNOS in ex vivo and in vivo colons from both mouse models by quantitative RT-PCR. Importantly, the inflammatory responses in the murine ex vivo culture system reflected the in vivo response in the inflamed colonic tissue as assessed by changes in inflammatory gene expression, suggesting that the murine culture system can be used for validation of future IBD therapies.