QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 15 November 2005
doi: 10.1242/jcs.02655

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yaroslavskiy, B. B. Articles by Blair, H. C. Search for Related Content PubMed PubMed Citation Articles by Yaroslavskiy, B. B. Articles by Blair, H. C. Social Bookmarking                         What's this?

NO-dependent osteoclast motility: reliance on cGMP-dependent protein kinase I and VASP

¹ Departments of Pathology and of Cell Biology and Physiology, University of Pittsburgh and Veteran's Affairs Medical Center, Pittsburgh, PA 15243, USA
² Mount Sinai Bone Program, Department of Medicine, Mount Sinai School of Medicine, New York, NY 10029, USA

View larger version (69K): [in a new window]   Fig. 1. Osteoclast motility is accelerated by cGMP. (A) Motility in human osteoclasts with cGMP activation. This is shown as superimposed frames at 30 minute intervals of osteoclast preparations on glass with the blocking cGMP analog Rp-cGMPS (50 µM, top) or with the cGMP agonist 8-pCPT-cGMP (100 µM, bottom). Two phase images were subtracted with the first image deleting the red channel and the second deleting the green channel, so that net motion can be seen as red and green double images (yellow where moving cells overlap). There was only minor movement when cGMP was blocked, mainly redistribution of nuclei and vesicles within the cells but also some slight variation of membrane spreading. The cGMP agonist in contrast causes net cell motion. The fields shown are 225 by 290 µm. The NO donor sodium nitroprusside (100 µM) produced motility at essentially the same rate, with a rapid time course. For phase-contrast microcinematography of these fields showing the comparison with the NO donor sodium nitroprusside, see supplementary material. (B) High-resolution measurements of motility in rat osteoclasts. This was measured using the attached area, summing new attachment and retracted area as a function of time as described (Zaidi et al., 1992) over 1 minute frames for 10 minutes before (left bar, mean±s.d. of eight frames) and after (right bar, mean±s.d. of 20 frames) addition of 250 µM 8-Br-cGMP. The difference in the rate of motility was statistically significant (P<0.01) even over this short time frame. Cellular motility by geometric center of attachment showed that motion was accelerated by cGMP activation (shown in supplementary material).

View larger version (43K): [in a new window]   Fig. 2. VASP and vß3 overlap in stationary cells but distribute separately in motile cells. Cells were incubated for 1 hour with 50 µM Rp-cGMPS to inactivate endogenous cGMP activity (A,C) or for 1 hour with 100 µM 8-pCPT-cGMP to induce maximal motility (B,D). (A,B) When cGMP was blocked (top panels), osteoclasts on bone labeled for VASP (red) and vß3 (green) showed distribution consistent with the annular attachment site; merged images confirmed this (yellow labeling). After cGMP activation (bottom panels), VASP distribution was partially separated from vß3. It is seen in parallel rings (arrows, bottom merged image) on one side of the cell. Under these conditions, the cells are highly motile (see Fig. 1). (C,D) When grown on glass, the cells spread uniformly, facilitating labeling, and an analogous labeling pattern to that on bone was clearly seen. With PKG I suppressed by Rp-cGMPS (top), VASP was associated with the integrin forming a yellow, double-labeled ring in the merged image. With the nonhydrolyzable cGMP analog 8-pCPT-cGMP (bottom panels), vß3 and VASP were disassociated at one edge of the cell, now occurring in discrete rings. This polarization was consistent with cell motility. Bar, 10 µm (A,B); 20 µm (C,D).

View larger version (48K): [in a new window]   Fig. 3. Eliminating PKG I from human osteoclasts prevents cGMP-induced motility and eliminates cGMP-dependent reorganization of the attachment ring. (A) Gene silencing at 48 hours post siRNA transfection reduced PKG I 95% by densitometry of western analysis. Real-time PCR gave a similar result (tenfold reduction, not shown). The western blot shows osteoclast lysates from cells treated with siRNA for PKG I compared to lysate from cells after mock transfection using a noncoding siRNA sequence. (B) Effect on motility in transfected cells. Means±s.d. of eight cells are shown. Time-lapse photography over 2 hours after addition of 8-pCPT-cGMP (see Fig. 1). Inset, motion over 20 frames by digital difference (lighter image, outlined in yellow, new position; darker image, outlined in green, old position). Note that the siRNA-treated cell has spread slightly, without moving, whereas the mock-transfected cell has moved about one cell diameter in this period. The variable diameter in the moving mock-transfected cell is typical of motile osteoclasts (see also Movie 1 in supplementary material). (C) Effect on cGMP-induced detachment. In each frame, a single osteoclast on bone is shown, with actin labeled with Alexa-488 phalloidin (green) and siRNA labeled with Cy3 (red). In frames 1 and 2, the transfected siRNA is specific for PKG I. The cell in frame 2 was also treated with 8-pCPT-cGMP, 100 µM for 1 hour, but its attachment ring was unaffected. The cell in frame 3 (bottom frame) was also transfected, but with a noncoding siRNA. In this case, its response to 8-pCPT-cGMP was normal, and the attachment has broken up into discrete clumps (arrows). Each field is 80 µm square.

View larger version (27K): [in a new window]   Fig. 4. Western blot studies of VASP in human osteoclasts, VASP-associated proteins in VASP precipitates, the effect of PKG I suppression on VASP phosphorylation, and the time-course of VASP phosphorylation after cGMP or NO activation. (A) Osteoclast lysates probed for VASP. VASP is abundant and occurs in cGMP-inhibited (Ctl, 50 µM Rp-cGMPS) and cGMP-stimulated (cGMP, 100 µM 8-pCPT-cGMP) osteoclasts in similar amounts. Fifteen µg lysates of osteoclasts were separated on 10% polyacrylamide in Laemmli buffer, and transferred to PVDF for labeling. Std, 100 ng of recombinant VASP. (B) VASP immunoprecipitates in cGMP-inhibited and -activated osteoclasts, probed for migfilin, v integrin and VASP. Immunoprecipitation for VASP was done using 900 µg lysates of osteoclasts after treatment with 50 µM Rp-cGMPS for 1 hour (Inhib) or with 100 µM 8-pCPT-cGMP for 1 hour (cGMP). These conditions essentially eliminate (Rp-cGMPS) or strongly promote (8-pCPT-cGMP) PKG I activity. Note that VASP is phosphorylated by activated PKG I. This also changed the relative abundance of its associated proteins. Migfilin was co-precipitated with VASP mainly in cGMP-activated cells (top panel), whereas v integrin precipitation by anti-VASP was reduced by cGMP. Phospho-Vasp (p-VASP), as expected, was greatly increased by cGMP activation relative to VASP. Two or more blots gave similar results. (C) Western blots of PKG I, phospho-VASP, VASP and actin in human osteoclasts after siRNA transfection targeting PKG I relative to mock-transfected control cells. Each lane is a 15 µg osteoclast lysate separated as in A, using cells either transfected with siRNA targeting PKG I as in Fig. 3 (left lane), or a mock-transfected control (right lane). The efficiency of siRNA targeting (75%) was lower than in the motility study shown in Fig. 3, but nonetheless it markedly reduced phospho-VASP relative to the mock-transfected control cells. (D) Phospho-VASP in VASP immunoprecipitates as a function of time after treatment of osteoclasts with 8-pCPT-cGMP or S-nitroso-N-acetylpenicillamine. Polyclonal anti-VASP was used for immunoprecipitation of protein from 900 µg osteoclast lysates, either without treatment or after 100 µM 8-pCPT-cGMP or 60 µM S-nitroso-N-acetylpenicillamine for the times indicated. Each lane (except an isoimmune control for immunoprecipitation by the antibody and a medium control with no cell lysate) represents an osteoclast immunoprecipitate separated on 9% SDS-PAGE as in A, with immunolabeling using antibody to phospho-VASP (p-VASP) compared to total VASP. Phospho-VASP was detected using anti phospho-Ser239-VASP monoclonal antibody. Increases in phospho-VASP using 8-pCPT-cGMP typically peaked by 1 hour, whereas S-nitroso-N-acetylpenicillamine (SNAP, lower left) peaked at 2-3 hours, in keeping with the half-life of that agent. Sodium nitroprusside gave a similar response but p-VASP increased within 10 minutes in keeping with the short half life of sodium nitroprusside (not illustrated).

View larger version (46K): [in a new window]   Fig. 5. Distribution of PKG I, VASP and migfilin in cGMP-inhibited and cGMP-activated osteoclasts. In osteoclasts grown on glass transfected with siRNA specific for PKG 1 (A), VASP (green) was partially distributed in a peripheral ring (arrows) similar to vß3 (see Fig. 2) whereas migfilin (red) was cytoplasmic. In contrast, in cGMP-stimulated cells (B), migfilin clearly colocalized with the peripheral VASP (arrows). Migfilin distribution in cells on bone was consistent with that in A and B, but is seen better in the higher resolution possible on glass substrate, and only glass-attached cells are shown. (C) In osteoclasts on bone with the cGMP inhibitor Rp-cGMPS at 50 µM, PKG I (green) was distributed in the cytoplasm; very little PKG I occurred in the cell periphery, marked by phalloidin (red) to label actin (arrows). (D) With the addition of 100 µM 8-pCPT cGMP, a nonhydrolysable cGMP analog, PKG I and actin were redistributed and there was some overlap (arrows). In cGMP agonists at high concentrations over longer times, nuclear localization of PKG I also occurred (not illustrated); extensive nuclear labeling correlated with complete detachment and may be associated with cell death (see text). Bar, 20 µm (A,B); 10 µm (C,D).

View larger version (30K): [in a new window]   Fig. 6. Reducing VASP with siRNA impairs formation of the osteoclast attachment and eliminates motility in response to NO. (A) Western blot showing reduction of VASP in cells with VASP-specific sequences versus a noncoding siRNA construct. This transfection used a mixture of four siRNAs. The knockdown efficiency (80-90%) was sufficient to show clear differences. (B) Impaired attachment with altered vß3 distribution in VASP-targeted siRNA-treated cells on bone. The siRNA was labeled with Cy3 to allow identification of transfected cells. Approximately 85% of attached cells were labelled with Cy3 (red). In contrast to the effect of PKG I knockdown, where the attachment ring was intact (Fig. 3), VASP-knockdown cells had an abnormal attachment with a fragmented vß3 ring (green). After VASP knockdown, the attachment structure did not change significantly with the cGMP agonist 8-pCPT-cGMP (100 µM, 1 hour). Each field is 40 µm horizontally. (C) Effect of a cGMP agonist on motility in mock-transfected and VASP-inhibited cells. There was no measurable motility in VASP-transfected cells when cells were exposed to 50 µM 8-pCPT-cGMP. The effect on motility was essentially the same as the effect of PKG I knockdown, but effects on the cell attachment were not observed with PKG I knockdown. Results are the means±s.d. of ten separate experiments.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter