SUN-347 Glucagon-like Peptide 1 (GLP-1) Acts Directly On Human Osteoclasts To Increase Differentiation And Bone Resorptive Activity

Abstract

Glucagon-like peptide 1 (GLP-1) is an intestinal hormone released in response to nutrient intake that promotes glucose-dependent insulin secretion by acting upon the pancreatic GLP-1 receptor (GLP-1R). GLP-1R agonists (GLP-1RAs) are widely used in treatment of type 2 diabetes. Preclinical data indicate that GLP-1RAs could be repurposed to treat low bone mass as GLP-1R-depleted mice have higher bone resorption and thinner cortical bones, while insulinopenic and insulin resistant rats have improved bone formation and reduced bone mass deterioration when treated with GLP-1 or GLP-1RAs. However, the effect of GLP-1 and GLP-1RAs on human bone cells remains undetermined. We aimed to elucidate the effect of GLP-1 on primary human osteoclast (OC) and osteoblast (OB) cultures. OCs were differentiated over 10 days from human blood-derived CD14+ monocytes and OBs over 4–6 weeks from human bone. Cells were seeded on bovine bone slices and studies performed using fetal bovine serum, MCSF and RANKL (OC monocultures) or MCSF only (OB monocultures and OB+OC co-cultures). We first investigated the effect of GLP-1 on bone resorptive activity of mature OCs on bovine bone slices. GLP-1 increased the eroded bone surface percentage compared to vehicle in both OC monocultures (1nM P=0.002; 10nM P=0.023; n=8 donors) and OC+OB co-cultures (1nM P=0.013; 10nM P=0.012; n=8 donors). We then tested the effects of GLP-1 on osteoblast activity in OC+OB co-cultures by measuring alkaline phosphatase (ALP). We found that GLP-1 increased ALP in OC+OB cultures (1nM, P=0.049; 10nM, P=0.019) and these effects were reversed by the GLP-1R antagonist exendin 9–39 (1nM, P=0.93, 10nM, P=0.64). However, in OB monocultures GLP-1 had no effects on ALP (1nM P=0.93, 10nM P=0.64) indicating a GLP-1-driven increase in osteoblast activity through osteoclast-osteoblast coupling. We then assessed the effect of GLP-1 on OC differentiation by assessing TRAcP activity. Although there was a trend towards increased TRAcP activity upon stimulation with GLP-1 on day 10 of osteoclastogenesis, this was not statistically significant (1nM P=0.12; n=8 donors; 10nM P=0.29, n=4 donors). Our studies indicated GLP-1 may have a direct effect on osteoclasts, and we therefore sought to characterise GLP-1-mediated signalling in these cells. We assessed the effect of GLP-1 on cAMP signalling using LANCE assays and assessed phosphorylation of ERK proteins by Western blot analysis in human OC cultures. OCs treated with 10nM GLP-1 for 30 minutes had increased cAMP signaling (P=0.004, n=12 bone slices from 2 donors) when compared to vehicle. Furthermore, 10nM GLP-1 induced rapid increases in phosphorylated ERK (P=0.03 following 2 minutes exposure, n=4 blots). In conclusion, our studies reveal that GLP-1 increases activity in primary mature human OCs, and OBs, via OCs. Our signaling studies in OCs indicate this is mediated by direct action of GLP-1 on human bone cells.