Identifying novel small molecule integrin agonists to mimic the favourable effects of the RGD domain of IGFBP-1 to treat cardiometabolic disease

Abstract

Abstract Funding Acknowledgements Type of funding sources: Other. Main funding source(s): British Heart Foundation (BHF) 4-Year PhD Programme Background Insulin-like growth factor binding protein (IGFBP)-1 has previously been shown to be associated with reduced type 2 diabetes (T2D) progression and preferable cardiovascular profiles in patients. Our laboratory previously demonstrated that IGFBP-1 is an insulin sensitiser in mouse models, in addition to having favourable vascular effects, independently of IGFs. IGFBP-1 has an Arg-Gly-Asp (RGD) domain which is thought to facilitate the direct interaction with the heterodimeric cell-surface receptor, integrin α5β1, inducing downstream cell signalling. We then showed the positive effects of IGFBP-1 could be mimicked with an RGD-containing linear peptide, GRGDTP, however this peptide is not a suitable therapeutic as it is rapidly metabolised in vivo. To our knowledge, RGD-based therapeutics have not been explored for cardiometabolic disease. Purpose There are currently few therapeutics which have been specifically designed to address both insulin resistance and the high risk of macrovascular disease in T2D. Therefore, exploring RGD-based integrin agonists will be essential in developing new insulin sensitising therapeutics which prevent cardiovascular complications. Methods The ROCS high speed ligand-based virtual screening tool was used to screen commercially available small molecule libraries mimicking the chemical structure of the RGD sequence of IGFBP-1. Additionally, we performed in silico modelling to predict the binding interactions of these molecules with the integrin α5β1 headpiece to select candidates. We then used a LIVE/DEAD cell viability assay to initially profile any cytotoxic effects of our selected compounds in integrin-expressing human umbilical vein endothelial cells (HUVECs) and C2C12 skeletal muscle myotubes. We further assessed the effects of compounds on relevant HUVEC functions using a high-throughput wound closure assay. Results The structure of the RGD domain of IGFBP-1 can be mimicked with small molecules containing amino acid moieties. These molecules are predicted to bind with micromolar affinity to the integrin, promoting beneficial interactions for integrin agonism. We selected 17 structurally diverse small molecules with the best predicted binding interactions. Initial in vitro profiling demonstrates that the molecules are not cytotoxic. In preliminary analyses, three of the selected compounds show a trend towards inhibition of HUVEC migration and proliferation, while six show a trend towards increased migration and proliferation. Conclusion The RGD domain of IGFBP-1 can be mimicked with novel small molecules, of which we have selected 17 ligands. Our compounds do not affect cell viability in relevant cell types, so can be further profiled. Our initial findings indicate that the compounds have different functional effects on HUVECs which allows us to prioritise hits having favourable effects in the context of T2D, such as enhancing angiogenic properties of endothelial cells.