Targeting S100A9 Interactions in the Multiple Myeloma Bone Marrow Environment Reduces Angiogenesis and Tumor Growth

S100A9 belongs to a family of low-molecular-weight calcium-binding proteins and is involved in many biological processes including inflammation, cell migration and angiogenesis. It has been reported that S100A9 knockdown reduced myeloid-derived suppressor cell (MDSC) accumulation and Multiple Myeloma (MM) cell growth in MM models. MDSC are a heterogeneous population of immature myeloid cells and contribute to MM disease by immunosuppression, induction of angiogenesis and secretion of cytokines and growth factors. S100A9 is therefore proposed as an attractive drug target and compounds inhibiting the interaction of S100A9 with its receptors RAGE and TLR4 have been developed.

In this study we evaluated the therapeutic relevance of S100A9 inhibition in MM using the preclinical immunocompetent murine 5T33MM model and two different small molecule inhibitors for S100A9 interactions (Active Biotech AB, Sweden). The presence of S100A9 was investigated in different cell populations. S100A9 is expressed in monocytic and granulocytic MDSC in murine and MM patient derived bone marrow (BM) samples, as demonstrated by western blot analysis and flow cytometry. S100A9 levels were higher in granulocytic MDSC compared to monocytic MDSC. Human MM cell lines and patient derived MM cells demonstrated low S100A9 expression, while no expression could be observed in 5T33MM cells. Treatment of 5T33MM mice with 30 mg/kg/day ABR-215757 (paquinimod) (quinoline-3-carboxamide analog) or ABR-238901 (N-(heteroaryl)-sulfonamide derivate) significantly diminished tumor cell percentages in the BM with ABR-238901 being the most potent one (27% relative reduction, p<0.01). Interestingly, ABR-238901 in combination with bortezomib (0.7 mg/kg subcutaneously, 2 times/week) resulted in a higher reduction in tumor load compared to single treatments (55% relative reduction compared to bortezomib alone). We investigated the effect on MDSC accumulation and immunosuppressive activity after ABR-238901 treatment in vivo. Flow cytometry analysis demonstrated no differences in MDSC and CD8+ IFNy+ cell numbers in ABR-238901 treated mice compared to vehicle mice. In addition, blocking S100A9 did not inhibit the immunosuppressive activity of MDSC. Interestingly, a significant reduction in IL6 and IL10 was observed by PCR in MDSC from ABR-238901 treated mice compared to vehicle. Both cytokines are produced by MDSC and are involved in MM cell proliferation and survival. To confirm the effects of S100A9 on cytokine expression, we treated 5T33MM derived MDSC with recombinant S100A9 in vitro and observed an increased expression of IL-6 and IL-10. In addition, transwell migration assays demonstrated that recombinant S100A9 has a chemoattractive effect on 5T33MM cells with a 3.5-fold induction of migration compared to control. Since S100A9 is also involved in angiogenesis, microvessel density was analyzed in the BM by CD31 staining. 5T33MM diseased mice treated with ABR-238901 had a significant decreased angiogenesis compared to 5T33MM mice treated with vehicle (p<0.01). We plotted for each mouse the microvessel density versus the percentage of tumor cells in the BM. Interestingly, the effect of ABR238901 on angiogenesis seems independent of tumor load.

In conclusion, according to our study S100A9 is not involved in direct MDSC accumulation and activity; however it influences the expression of IL6 and IL10 by MDSC and act as a chemoattractant for MM cells. Interestingly, blocking S100A9 interactions in vivo affects angiogenesis and reduces tumor load especially in combination with bortezomib.