Src-like adaptor protein regulates osteoclast generation and survival

Abstract

Src-like adaptor protein (SLAP) is a hematopoietic adaptor containing Src homology (SH)3 and SH2 motifs and a unique carboxy terminus. Unlike c-Src, SLAP lacks a tyrosine kinase domain. We investigated the role of SLAP in osteoclast development and resorptive function. Employing SLAP-deficient mice, we find lack of the adaptor enhances in vitro proliferation of osteoclast precursors in the form of bone marrow macrophages (BMMs), without altering their survival. Furthermore, osteoclastogenic markers appear more rapidly in SLAP-/- BMMs exposed to RANK ligand (RANKL). The accelerated proliferation of M-CSF-treated, SLAP-deficient precursors is associated with enhanced ERK activation. SLAP's role as a mediator of M-CSF signaling, in osteoclastic cells, is buttressed by complexing of the adaptor protein and c-Fms in lipid rafts. Unlike c-Src, SLAP does not impact resorptive function of mature osteoclasts but induces their early apoptosis. Thus, SLAP negatively regulates differentiation of osteoclasts and proliferation of their precursors. Conversely, SLAP decreases osteoclast death by inhibiting activation of caspase 3. These counterbalancing events yield indistinguishable bones of WT and SLAP-/- mice which contain equal numbers of osteoclasts in basal and stimulated conditions.

Fig. 1

SLAP deficiency increases osteoclastogenesis. (A) WT BMMs were cultured in M-CSF (10 ng/ml) and RANKL (100 ng/ml) with time. SLAP and SLAP-2 mRNA was measured by RT-PCR. T cell lysate serves as positive control for expression of both isoforms. TRAP serves as positive control for osteoclastogenesis and GAPDH as loading control. (B) WT and SLAP−/− (KO) BMMs were cultured with 100 ng/ml of RANKL and the indicated concentrations of M-CSF. After 4 days, cells were stained for TRAP activity. (C) Statistical analysis of the number of WT and SLAP−/− TRAP positive multinucleated cells/well at day 4. (*p < 0.005)

Fig. 2

Expression of osteoclastogenic markers is accelerated in SLAP−/− cells. (A) RT-PCR analysis of osteoclastogenic markers in WT and SLAP−/− cells in culture with M-CSF (10 ng/ml) alone (day 0) or with RANKL (100 ng/ml) for 2 or 4 days. (GAPDH serves as loading control) (B) Quantitative analysis of data presented in A.

Fig. 3

SLAP regulates the proliferation but not survival of osteoclast precursors. (A) Equal numbers of WT and SLAP−/− BMMs were cultured with increasing concentrations of M-CSF for 3 days. Incorporation of BrdU during the last 4 hrs of culture was determined. (*p < 0.001). (B) WT and SLAP−/− BMMs were cultured in M-CSF alone (10 ng/ml) (day 0) or M-CSF and RANKL (100 ng/ml) (day 3). Apoptosis was determined as a function of DNA fragmentation.

Fig. 4

M-CSF-induced ERK phosphorylation is enhanced in SLAP−/− BMMs. BMMs were cultured with 10 ng/ml (A) or 25 ng/ml (B) of M-CSF for the indicated times. Akt and ERK phosphorylation was determined by immunoblot. Total ERK and Akt levels serve as loading controls. (C) Cytokine and serum starved WT and SLAP−/− BMMs were exposed to M-CSF (50ng/ml) with time. c-Fms immunoprecipitates were immunoblotted with antibodies to c-Fms. Total cell lysate was immunoblotted for tyrosine phosphorylated proteins and phospho-PLCγ2. Actin serves as loading control (D) BMMs were treated with RANKL (100 ng/ml) with time. Lysates were immunoblotted with indicated antibodies. Actin and total p38, JNK and IκB serve as loading controls.

Fig 5

c-Fms and SLAP associate in lipid rafts. WT BMM lysates were isolated by density gradient centrifugation. (A) Fractions were immunoblotted for c-Fms and SLAP. Flotillin serves as a lipid raft marker. (B) c-Fms immunoprecipitates of pooled fractions 12 and 13 were immunoblotted for SLAP, flotillin and c-Fms. Immunoblotting with mAb recognizing c-Src, which is absent in BMMs, serves as negative control. Immunoblot of lysate (TCL) derived from c-Src-expressing osteoclasts documents antibody efficacy.

Fig. 6

SLAP-deficient osteoclasts are predisposed to apoptosis. WT and SLAP−/− BMMs were cultured in M-CSF and RANKL. At day 4, cells were incubated with (+) or without (−) cytokines for the last 4 hours. (A) The magnitude of apoptosis was measured as a function of DNA fragmentation (B) Activated caspase-3 was measured by immunoblot on day 3 and day 4. (*p < 0.05, **p < 0.005).

Fig. 7

SLAP does not regulate function of individual osteoclasts in vitro. WT and SLAP−/− BMMs were cultured on bone with M-CSF (10 ng/ml) and RANKL (100 ng/ml) for 5 days. (A) Resorptive lacuna formation was examined on day 5. (B) Bone resorptive activity of the cultures in A was determined by collagen type I fragment ELISA of culture media on day 5. (*p < 0.05). (C) WT and SLAP−/− BMMs were cultured on plastic in the presence of M-CSF and RANKL. After 3 days, the cells were lifted and replated on bone in M-CSF and RANKL. One day later, resorptive activity was determined by collagen type I fragment ELISA.

Fig. 8

SLAP deficiency does not alter osteoclast number in vivo. WT and SLAP−/− mice were injected, supracalvarially, with PBS or RANKL (100 μg) daily, for 7 days. (A) TRAP-(red reaction product) stained histological sections of calvaria. (B) Percentage of osteoclast surface per unit bone volume.