Western blots were quantified by NIH ImageJ software version 1.45. Human DKK1 levels in the conditioned medium from HPSE-low and HPSE-high CAG cells were measured using hDKK1 Quantikine ELISA kit (R&D Systems). Mouse DKK1 levels

in conditioned medium from primary murine osteoblastic progenitor cells, C3H10T1/2 pre-osteoblastic cells and ST2 stromal selleck inhibitor cells cultured in the absence or presence of rHPSE were measured by mDKK1 Quantikine ELISA (R&D Systems). All assays were performed according to the protocols provided by the manufacturers. Statistical comparisons between two experimental groups were analyzed by Student’s t test. ANOVA was employed for statistical analyses among multiple groups, followed by a post-hoc Bonferroni test. The correlations between heparanase and osteocalcin expression in MM patients’ samples were assessed using Spearman correlation coefficient. P < 0.05 was considered statistically significant and is reported as such. To determine the effects of heparanase expression by myeloma cells on mesenchymal lineage cells, we first

examined the effect of heparanase on osteoblastogenesis and parameters of bone formation by evaluating osteoblast parameters in SCID-hu mice [36]. We measured osteocalcin expression, a marker of mature osteoblast differentiation, in histologic sections of engrafted bone [23]. The analysis revealed that the number of osteocalcin-positive this website osteoblasts was significantly diminished in both primary (injected with tumor cells) and contralateral (not injected with tumor cells) engrafted bones of the mice bearing

HPSE-high tumor, compared to animals bearing tumors formed by HPSE-low cells (Figs. 1A–B). These data provided the first direct experimental evidence that osteoblastogenesis was inhibited by heparanase in vivo. Interestingly, immunohistochemical staining for human kappa light chain, a specific marker of Interleukin-3 receptor CAG myeloma cells, revealed no detectable tumor cells in uninjected contralateral grafts (data not shown), suggesting that the inhibition of osteoblastogenesis was humoral and not due to myeloma tumor metastasis to the contralateral graft. We next investigated the correlation between heparanase expression and osteocalcin expression in myeloma patients, using primary bone marrow core biopsy specimens obtained from myeloma patients. Specific immunohistochemical staining for heparanase and osteocalcin was performed on 28 myeloma patient bone marrow core biopsy specimens. We identified a significant negative correlation (r = − 0.62, P < 0.001) between heparanase expression by myeloma cells and osteocalcin expression by bone marrow cells (Fig. 2 and Supplementary Table 1).