Bone Marrow Monocyte / Macrophage Derived Activin A Mediates the Osteoclastogenic Effects of IL-3 in Myeloma,

Abstract 3933

Multiple myeloma (MM) is characterized by bone destruction with suppressed new bone formation that is mediated by multiple factors including MIP-1α (CCL3), RANKL, IL-3, DKK1 and recently, Activin A (ActA). IL-3 is a bifunctional cytokine that indirectly increases osteoclastogenesis and suppresses osteoblastogenesis via CD14⁺ bone marrow monocytes (BMM). Additionally, IL-3 levels are elevated in the BM plasma of MM patients compared with normals, however the mediators of IL-3's effects on myeloma bone disease are unknown. To investigate this we performed gene expression profiling using Affymetrix GeneChip® analysis of IL-3 treated BMM from MM patients and found ActA gene expression was increased 180-fold and confirmed this finding at the protein level by ELISA. ActA is a negative regulator of bone mass that promotes osteoclastogenesis and is overproduced in MM patients. Interestingly, treatment of MM cells or bone marrow stromal cells (BMSC) with IL-3 did not induce secretion of ActA.

We found that ActA was produced by MM and MGUS patient CD14+ cells treated with IL-3 to a significantly higher degree compared to healthy subjects. (Median ActA levels for MM and MGUS ActA levels were increased 66.57 and 51.6 fold respectively over untreated cells, while IL-3 treatment of normal cells increased ActA 8.5 fold.) ActA levels were also increased in freshly isolated marrow plasma of a cohort of patients with active MM as compared to patients with smoldering MM (SMM), MGUS, or healthy subjects (median ActA levels: active MM 453 pg/ml, SMM 328 pg/ml, MGUS 332 pg/ml, normal 286 pg/ml). ActA levels in MM patients with and without bone disease were not significantly different. (Median value with bone disease 463pg/ml vs. 407 pg/ml without bone disease.)

ActA has also been reported to have a role in the differentiation and polarization of CD14+ tumor associated macrophages (TAMs), which are osteoclast precursors and can block MM cell apoptosis. Therefore, we examined the potential role of ActA in IL-3 mediated osteoclast (OCL) formation. Culture of normal BMM with ActA or IL-3 significantly enhanced osteoclastogenesis compared with control (mean number of OCL / 1×10⁵ normal marrow non-adherent cells plated in IL-3 (100pg/ml) treated cultures 73; ActA (1ng/ml) treated cultures 123; cultures with vehicle alone, 8), and ActA enhanced RANKL-induced osteoclastogenesis. Osteoprotegerin treatment of normal donor BMM stimulated with ActA failed to block the osteoclastogenic effects of ActA, demonstrating that ActA's osteoclastogenic effects were RANKL independent. Importantly, the osteoclastogenic effect of IL-3 was dose-dependently inhibited by anti-ActA, and IL-3 induced ActA expression by BMM decreased during OCL differentiation. In support of early OCL precursors as the source of IL-3 induced ActA, we did not identify IL-3 receptors on mature OCL by flow cytometry.

These results demonstrate that IL-3 induction of osteoclastogenesis is mediated by ActA produced by CD14+ BMM and is RANKL independent. Thus, we hypothesize that therapies targeting the ActA receptor, such as the recently developed ActA receptor antagonist, should block both IL-3 and ActA, and thereby significantly impact MM bone disease via their effects on TAMs.