Mechanical loading in osteocytes induces formation of a Src/Pyk2/MBD2 complex that suppresses anabolic gene expression

Abstract

Mechanical stimulation of the skeleton promotes bone gain and suppresses bone loss, ultimately resulting in improved bone strength and fracture resistance. The molecular mechanisms directing anabolic and/or anti-catabolic actions on the skeleton during loading are not fully understood. Identifying molecular mechanisms of mechanotransduction (MTD) signaling cascades could identify new therapeutic targets. Most research into MTD mechanisms is typically focused on understanding the signaling pathways that stimulate new bone formation in response to load. However, we investigated the structural, signaling and transcriptional molecules that suppress the stimulatory effects of loading. The high bone mass phenotype of mice with global deletion of either Pyk2 or Src suggests a role for these tyrosine kinases in repression of bone formation. We used fluid shear stress as a MTD stimulus to identify a novel Pyk2/Src-mediated MTD pathway that represses mechanically-induced bone formation. Our results suggest Pyk2 and Src function as molecular switches that inhibit MTD in our mechanically stimulated osteocyte culture experiments. Once activated by oscillatory fluid shear stress (OFSS), Pyk2 and Src translocate to and accumulate in the nucleus, where they associate with a protein involved in DNA methylation and the interpretation of DNA methylation patterns -methyl-CpG-binding domain protein 2 (MBD2). OFSS-induced Cox-2 and osteopontin expression was enhanced in Pyk2 KO osteoblasts, while inhibition of Src enhanced osteocalcin expression in response to OFSS. We found that Src kinase activity increased in the nucleus of osteocytes in response to OFSS and an interaction activated between Src (Y418) and Pyk2 (Y402) increased in response to OFSS. Thus, as a mechanism to prevent an over-reaction to physical stimulation, mechanical loading may induce the formation of a Src/Pyk2/MBD2 complex in the nucleus that functions to suppress anabolic gene expression.

Figure 1. Anabolic protein and gene expression is effected in Pyk2−/− osteoblasts under static and OFSS conditions.

(A) Wild-type osteoblasts and Pyk2−/− osteoblasts were exposed to either static of one hour of OFSS. Representative of immunoblots for Cox-2, GAPDH, and Pyk2 show a OFSS-induced increase in Cox-2 protein expression. OFSS does not cause a change in the loading control, GAPDH. Graph represents the densitometry units of Cox-2, normalized to GAPDH. In wild-type osteoblasts, OFSS results in a 1.8 fold change in Cox-2 protein expression. Pyk2−/− osteoblasts have an elevated protein expression of Cox-2 under static conditions, and after OFSS the level of Cox-2 protein expression is highest compared to all other groups (static control, wild-type osteoblasts static control, and wild-type osteoblasts OFSS). Error bars represent standard error. *p<0.05 vs wild-type static control, #p<0.05 vs. wild-type osteoblasts. N = 3 in three separate trials. (B) Wild-type osteoblasts and Pyk2−/− osteoblasts were exposed to either static or 1 hour of OFSS conditions, using either an OFSS pump or orbital shaking platform. Static Pyk2−/− osteoblasts expressed significantly higher levels of osteopontin compared to static wild-type osteoblasts. In response to OFSS, using either method, the absence of Pyk2 further enhances the OFSS-induced osteopontin expression increase. Error bars represent standard error. *p<0.05 vs. wild-type static control, #p<0.05 vs wild-type osteoblasts. N = 3 in three separate trials.

Figure 2. Src kinase represses osteocalcin in static and OFSS conditions.

(A) Treating MC3T3 osteoblasts and MLO-Y4 osteocytes with Src Inhibitor 1 (SI1) (10 µM) for 1 hour inhibits Src kinase activity. Western blot analysis of phosphorylated-Akt (Ser308), Total Akt, and γ-tubulin under static and OFSS conditions (1 hour). (B) One hour of SI1 treatment significantly increases the expression of osteocalcin in MC3T3 osteoblasts and MLO-Y4 osteocytes (*p<0.05). (C) OFSS (1 hour) further enhanced the expression of osteocalcin in both MC3T3 osteoblasts and MLO-Y4 osteocytes compared to static controls treated with carrier (DMSO) (p<0.05). *Represents a statically significant increase compared to wild-type static control (p<0.05). #Represents a statically significant difference between treatment groups (#p<0.05) Error bars represent standard error. An n≥3 were used in all experiments and each experiment was repeated at least three times.

Figure 3. Pyk2 and Src accumulate in perinuclear and nuclear regions in response to OFSS.

(A) Immunofluorescence microscopy of osteoblast’s subjected to either static culture conditions or OFSS (30 minutes, 1 hour, or 1 hour+1 hour of rest). Slides were fixed immediately and processed for immunofluorescence using antibodies against Pyk2, followed by FITC-conjugated secondary antibodies. The nucleus was visualized using DAPI. Scale bars = 100 µm (B) OFSS induces accumulation of Src at perinuclear/nuclear regions in MLO-Y4 osteocytes. Immunofluorescence microscopy of MLO-Y4 osteocytes subjected to static culture conditions or OFSS for 20 minutes. Slides were fixed immediately and processed for immunofluorescence using antibodies against activated Src (Y416 followed by FITC-conjugated secondary antibodies). F-actin was visualized using Texas-Red Phalloidin and the nucleus was visualized using DAPI. White arrows indicate focal adhesions. Scale bars = 25 µm (B) Immunofluorescence microscopy of MLO-Y4 osteocytes subjected to static culture conditions or OFSS for 20 minutes. Slides were fixed immediately and processed for immunofluorescence using antibodies against total Src followed by FITC-conjugated secondary antibodies. F-actin was visualized using Texas-Red Phalloidin and the nucleus was visualized using DAPI. Scale bars = 25 µm.

Figure 4. OFSS causes an increase in nuclear Src activity.

(A) Composite phasor plot analysis of static Src biosensor lifetime and 20 minutes post-OFSS Src biosensor lifetime. (B) Graph represents n = 3 in which the average lifetime of each ROI (30) at each time point analyzed. Error bars represent standard error. (C) Graph of nuclear Src lifetimes. *Represents a statically significant increase compared to static control (*p<0.001). #Represents a statically significant difference between other OFSS time points examined (#p<0.05) Error bars represent standard error. (D, E) Lifetime maps of the same MLO-Y4 osteocyte under static, 10 minutes, 15 minutes or 20 minutes post-OFSS.

Figure 5. Src activation increases in the nucleus in response to OFSS.

(A) Western blot analysis of nuclear fractionation blotted for Src activation (Y416), total Src and lamin B in MLO-Y4 osteocytes exposed to 5 minutes of OFSS or static culture conditions. (B) Graph represents quantification of Src activation (Y416)/total Src in nuclear fractions. Error bars represent standard error. Statistically significant difference between static and 10 minutes post-OFSS (*p<0.05). An n≥3 were used in the repeated experiments.

Figure 6. A complex between MBD2 and Pyk2 forms under static conditions in MLO-Y4 osteocytes, while OFSS-induces the association of MBD2 and Pyk2 (Y402) and Src (Y418) and Pyk2 (Y402) in MLO-Y4 osteocytes.

(A) Co-immunoprecipitation between MBD2 and Pyk2 was performed from MLO-Y4 osteocytes harvested under static conditions. MBD2 was not associated with Pyk2 when control normal mouse Ig was used in the immunoprecipitation. Anti-MBD2 antibody was used to probe the blot. (B) Co-immunoprecipitation was performed in MLO-Y4 osteocytes harvested under static (S) or OFSS (F) conditions. Normal rat serum (NRS), MBD2, and Src (Y418) were the antibodies used for the immunoprecipitation. Anti-Pyk2 (Y402) specific antibody was used to probe the blot.