Intestinal targeting of small molecule MMP and TGR5 ligands for inflammatory bowel disease

The aetiology of inflammatory bowel disease (IBD) is incompletely understood but it is generally accepted to involve a dysregulated intestinal immune response to commensal microbiota and an increase in intestinal permeability. Matrix metalloproteinases (MMPs), MMP-9 (gelatinase B) in particular, are implicated in this compromise of intestinal barrier structure and have further roles in the potentiation and augmentation of disease. Takeda G-protein coupled receptor 5 (TGR5), a bile acid receptor, is also involved in maintenance of intestinal barrier integrity and there is evidence to suggest that both these targets are accessible to modulators confined to the gut lumen. Expressed in many tissues and cell types and involved in many physiological processes, the potential of MMP and TGR5 modulators has been limited by unwanted off-target effects. We proposed that small molecule ligands designed to be confined to the gastrointestinal lumen through judicious modification of physicochemical properties could achieve non-systemic effects. Suppression of oral absorption potential of an established gelatinase inhibitory scaffold was pursued by synthetic strategies to increase molecular weight and/or polarity including preparation of a dimer linked by a short PEG chain or introduction of a permanent charge. Bulky conjugates of an MMP inhibitor and lithocholic acid (LCA), the most potent endogenous bile acid agonist at TGR5, were also prepared. By virtue of their high molecular weight, such conjugates would be expected to have limited absorption potential and we hoped to retain activity at both targets. Compound 53, a permanently positively charged barbiturate-based gelatinase inhibitor was a potent inhibitor of recombinant human MMP-9 as measured by fluorogenic assay and an uptake assay in Caco-2 cells indicated that it was a poor candidate for passive uptake. It was chosen for evaluation in a DSS mouse model of colitis to test the hypothesis that disease modifying effects can be achieved by inhibition of apically secreted gelatinases. Compound 53 treatment reduced the severity of DSS-induced colitis in mice and the disease activity index (DAI) scores of the treated group were significantly lower than the DSS group. PCR analysis of the retained colon samples indicated that the pro-inflammatory cytokines TNF-?, IL-1? and IL-6 were downregulated at gene level in the treated group compared to the DSS control group and these downregulations were statistically significant. The MMP-9 inhibitor ? TGR5 agonist conjugates retained potent MMP-9 inhibition but activation of TGR5 showed tighter structural specificity than expected. Some compounds of the series caused accumulation of cAMP, the downstream second messenger of TGR5 activation, in human enterendocrine NCI-H716 cells but potency was much lower than LCA. It is yet unknown if this level of activity could exert a functional effect in the context of a gut confined agent in appropriate animal models. This work contributed to our knowledge of binding at the human TGR5 receptor, its SAR and capacity for chemical manipulation to achieve gut confinement. TGR5 agonists that are expected to be poor candidates for absorption by virtue of their polarity were also synthesised from LCA. These candidates retained activity at TGR5 and they are promising candidates for further study