Inhibitors of Fibroblast Activation Protein (FAP) Inhibit Primary Myeloma Growth and Osteoclastogenesis Ex Vivo and In Vivo.

Abstract

Fibroblast activation protein (FAP), a cell surface serine protease with both dipeptidyl peptidase and collagenase activity, is selectively expressed by tumor stroma and involved in tumor metastasis. We have reported that FAP is upregulated in myelomatous bone and is overexpressed in osteoclasts after coculture with myeloma (MM) cells. FAP is not expressed by MM cells and FAP siRNA reduced MM cell survival in cocultures (Ge et al., BJH 2006). The aim of the study was to investigate the effect of FAP inhibitors, PT-100 and PT-630 on MM cell growth and osteoclastogenesis using coculture system and the SCID-hu model for primary MM. PT-630 inhibits cell surface dipeptidyl peptidase activity while PT-100 also inhibits intracellular activity of these enzymes. MM cells from 6 patients were cocultured with osteoclasts and treated twice a day with PT-100 and PT-630 (0.1–100 μM) for 5–7 days. Whereas PT-100 effectively inhibited MM cell growth in all tested doses by 38%–62% (p<0.002 vs. 100 μM), PT-630 inhibited MM cell growth in a dose dependent manner reaching 45% growth inhibition with 100 μM (p<0.02). These compounds had no direct effect on MM cell survival. Moreover, recombinant FAP had no impact on MM cells cultured alone, suggesting that FAP-induced MM cell survival depends on close contact between MM cells and osteoclasts. The anti-MM effect of PT-100 in cocultures was mediated through downregulation of phosphorylated p38 in MM cells as detected by Phospho MAPK array and confirmed by Western blot. MMP-2 and MMP-9 have been associated with FAP activity. The level of MMP-2 but not MMP-9 was reduced in coculture conditioned media by 44±7% (p<0.04) following treatment with PT-100 while PT-630 had no significant effect on production of these matrix metalloproteinases. To test effect on osteoclastogenesis, osteoclast precursors were incubated with RANKL and M-CSF in the absence and presence of PT-100 (1 μM) and PT-630 (10 μM) for 5–7 days. PT-100 and PT-630 inhibited formation of multinucleated osteoclasts by 78±6% (p<0.001) and 56±6% (p<0.003), respectively. Culture of osteoclasts on dentine slices in the presence of PT-100 and PT-630 reduced resorption pit area by 92% (p<0.01) and 69% (p<0.04), respectively. The anti-osteoclastogenic effects were mediated through inhibition of phosphorylated p38 MAPK in osteoclastic cultures in a dose related manner. In vivo, SCID-hu mice engrafted with MM cells from 4 patients were orally treated for 4–5 weeks with PT-100 (20 mg/day) and PT-630 (200 mg/day). These agents inhibited MM growth in 2 experiments, delayed growth in one experiment and had no effect on MM in an additional experiment. Overall, final hIg levels in hosts treated with vehicle, PT-100 and PT-630 were 355±170, 183±78 and 76±27 mg/ml, respectively. Bone mineral density (BMD) of the myelomatous bone was increased in responding hosts (3% vs. -32% change from pretreatment level in control) and had reduced severity of bone loss in myelomatous bone of nonresponding hosts (−15% vs. −28% change from pretreatment level in control), suggesting that, as shown in vitro, these agents directly affect bone cell function in vivo. We conclude that FAP is critically involved in MM osteolysis and tumor growth and thus approaches to inhibit FAP activity in myelomatous bone may help control MM and its associated bone disease.