GPCR kinase 2 interacting protein 1 (GIT1) regulates osteoclast function and bone mass

Abstract

G-protein-coupled receptor (GPCR) kinase 2 interacting protein-1 (GIT1) is a scaffold protein expressed in various cell types including neurons, endothelial, and vascular smooth muscle cells. The GIT1 knockout (KO) mouse has a pulmonary phenotype due to impaired endothelial function. Because GIT1 is tyrosine phosphorylated by Src kinase, we anticipated that GIT1 KO should have a bone phenotype similar to Src KO. Microcomputed tomography of the long bones revealed that GIT1 KO mice have a 2.3-fold increase in bone mass compared to wild-type controls. Histomorphometry showed increased trabecular number and connectivity suggesting impaired bone remodeling. Immunoblot analysis of GIT1 expression showed that it was expressed in both osteoclasts and osteoblasts. Osteoblast activity and function assayed by alkaline phosphatase, mineral nodule formation, and in vivo calcein labeling were normal in GIT1 KO mice suggesting that the observed increase in bone mass was due to an osteoclast defect. GIT1 KO bone marrow cells differentiated into multinucleated osteoclasts, but had defective bone resorbing function on dentin slices. This defect was likely caused by loss of podosome belt based on immunofluorescence analysis and previous studies showing that GIT1 is required for podosome formation. Furthermore, we found that GIT1 was a regulator of receptor activator of NFκB (RANK) signaling since it was tyrosine phosphorylated in a Src-dependent manner and was required for phospholipase C-γ2 phosphorylation. These data show that GIT1 is a key regulator of bone mass in vivo by regulating osteoclast function and suggest GIT1 as a potential target for osteoporosis therapy.

Fig. 1. GIT1 KO mice have increased bone mass

(A) Three dimensional microCT analysis of the long bones from 10-12 week old sex matched GIT1 WT and KO mice. Wt n=9; KO n=8. (B-E) Bone morphometric analysis of the long bones from GIT1 WT and KO mice. BV/TV%: percentage of bone volume (BV)/total volume (TV); Conn. Density (1/mm³): Connectivity density per mm³. All values are expressed as mean ± SEM. Data analysis was done by two-tailed unpaired student’s t-test. A p<0.05 was considered to statistically significant. (F) Longitudinal sections of femur bones of GIT1 WT and KO mice stained with H&E. Trabecular bone number and connectivity are greater in GIT1 KO femurs compared to WT controls as indicated by arrows. (Scale bar: 20 μm) n=3 per group.

Fig. 2. GIT1 KO mice have normal osteoclast number

(A) BM cells from GIT1 WT were differentiated into OC with MCSF (20 ng/ml) and RANKL (50 ng/ml). Cell lysates were harvested on day 7 and protein expression probed using GIT1 antibody. GIT1 KO OC did not show GIT1 protein expression. Actin serves as loading control. (B,C) TRAP stain in the femur sections of GIT1 WT and KO mice. OC per mm² were quantified using light microscopy. WT= 10±1; KO= 11± 2. Values are expressed as mean ± SEM (n=3) Scale bar: 50μm. (D,E) GIT1 WT and KO femur sections were stained with alcian blue to visualize cartilage and counterstained with Orange O-eosin for bone. Note increased presence of cartilage remnants (blue) in the trabecular bone (orange) in GIT1 KO mice (E, arrows) (Scale bar: 20 μm, n=3).

Fig. 3. GIT1 deficiency impairs osteoclast function

GIT1 WT and KO BM cells were differentiated into OC for 7 days on (A,B) culture dish and (C,D) dentin slices. Cells were fixed, TRAP stained and TRAP positive cells greater than 3 nuclei (arrows) were quantified using a light microscope (WT= 80±10, KO=75 ±12, p>0.05). (E, F) Resorbtion pits on dentine slices were examined by toludiene stain after removing OCs. (G) Quantitative analysis of the pit area volume as percentage of control showed a 65% decrease in GIT1 KO OC. (H, I) GIT1 WT and KO OC were fixed on day 7 and stained for F-actin rings using rhodamine phalloidin. In WT OC, F-actin (red) was organized in podosome belts (H) while GIT1 KO OC showed loss of podosome belts (I). Nuclei (blue) were stained with DAPI to identify multinucleated cells. (Scale bar: 15 μm) (J) Quantitative analysis of number of normal podosome belts in GIT1 WT and KO OC showed a 60% decrease in GIT1 KO OC. A total of 70 OC were counted from three individual experiments. All values are expressed as mean ± SEM. (n=3). Data analysis was done by two-tailed unpaired student’s t-test. A p<0.05 was considered statistically significant.

Fig. 4. GIT1 functions downstream of RANK signaling

(A) GIT1 WT and KO BM cells were differentiated into OC by treatment with RANKL (50 ng/ml) and MCSF (20 ng/ml) for 7 days. Cells were serum starved for 6 hrs, stimulated with 100 ng/ml RANK for the indicated times. Cell lysates were harvested and probed with 4G10 phosphotyrosine antibody to visualize GIT1 phosphorylation (97 kd, arrow). Blots were then probed for GIT1 expression. (B) WT BM OC were treated with Src inhibitor PP2 (10 μM for 1 hr), stimulated with 100 ng/ml RANKL for 5 mins. GIT1 phosphorylation was probed using 4G10 antibody. Blots were reprobed for GIT1 expression. (C) GIT1 WT and KO OC on day 7 were starved and treated with RANKL (100 ng/ml) for indicated times. Lysates were immunoblotted with phoshorylated specific PLCγ2 (Y759) antibody (p-PLCγ2 Y759). Fold induction of normalized, phosphorylated protein vs time 0 of the control are shown.

Fig. 5. Src, GIT1 and PLCγ2 localizes in the podosome belts

WT KO BM cells were differentiated into OC by treatment with MCSF (20 ng/ml) and RANKL (50 ng/ml) for 7 days, fixed and stained for (A) Src, (B) GIT1 and (C) PLCγ2. (D) WT OCs were serum starved for 4 hrs and stimulated with RANKL (100ng/ml) for 5 mins, fixed and stained for phosphorylated form of PLCγ2 using phospho- PLCγ2 (Y759) antibody. Rhodamine-phalloidin was used to visualize actin rings. (Scale bar: 15 μm)