Distinct roles for Crk adaptor isoforms in actin reorganization induced by extracellular signals

Abstract

Crk family adaptors, consisting of Src homology 2 (SH2) and SH3 protein-binding domains, mediate assembly of protein complexes in signaling. CrkI, an alternately spliced form of Crk, lacks the regulatory phosphorylation site and C-terminal SH3 domain present in CrkII and CrkL. We used gene silencing combined with mutational analysis to probe the role of Crk adaptors in platelet-derived growth-factor receptor beta (PDGFbetaR) signaling. We demonstrate that Crk adaptors are required for formation of focal adhesions, and for PDGF-stimulated remodeling of the actin cytoskeleton and cell migration. Crk-dependent signaling is crucial during the early stages of PDGFbetaR activation, whereas its termination by Abl family tyrosine kinases is important for turnover of focal adhesions and progression of dorsal-membrane ruffles. CrkII and CrkL preferentially activate the small GTPase Rac1, whereas variants lacking a functional C-terminal SH3 domain, including CrkI, preferentially activate Rap1. Thus, differences in the activity of Crk isoforms, including their effectors and their ability to be downregulated by phosphorylation, are important for coordinating dynamic changes in the actin cytoskeleton in response to extracellular signals.

Fig. 1.

Crk family adaptors are required for formation of circular dorsal-membrane ruffles and increased cell migration after PDGF stimulation. (A) The domain structure of Crk family adaptors. CrkI and CrkII have a PxxP motif in the SH2 domain that is absent in CrkL. Y represents the regulatory tyrosine phosphorylation site located at position 221 for human CrkII and position 207 for human CrkL. Numbers below domains show percent identity between CrkII and CrkL for each domain, and numbers in parentheses show similarity. (B) Expression levels of Crk family proteins determined by immunoblotting for NIH-3T3 cells infected with retrovirus carrying pSUPER vector only, or one with shRNA targeting CrkL, CrkI/II or all Crk family genes. Numbers to the left of blots indicate relative molecular mass in kDa. (C) Serum-starved cells were fixed before or 10 minutes after stimulation with 24 ng/ml PDGF. Fixed cells were stained with Hoechst 33342 and phalloidin–Texas-red for DNA (blue) and F-actin (red), respectively. Scale bar: 50 μm. (D) Number of cells forming circular dorsal membrane ruffle(s) and total cells and were counted for multiple fields. Graph represents mean ± s.e.m. of percentage of cells with circular dorsal-membrane ruffle(s). ^‡P<0.05 (n=15). (E) Wounds were introduced on the surface of confluent dishes of serum-starved cells. After washing, 0.1% CS-containing medium with or without 24 ng/ml PDGF was added to the dish. Images were taken immediately or 12 hours after wounds were introduced. Scale bar: 200 μm. (F) Cells migrating more than 200 μm from the initial edge of wound were counted 12 hours after wound introduction. The number was normalized relative to control cells under serum starvation. This value is referred to as the migration index. Graph represents mean ± s.e.m. of migration index. ^‡P<0.05 (n=9).

Fig. 2.

Tyrosine phosphorylation of CrkII and CrkL by Abl family proteins disrupts interactions with p130Cas after PDGF stimulation. (A) NIH-3T3 cells were infected with retrovirus carrying vector only or expressing shRNA targeting Abl, Arg or both. Cells were harvested before and at various time points after PDGF stimulation. Whole-cell lysates (WCL) were immunoblotted to detect tyrosine phosphorylated CrkII and CrkL, and expression of c-Abl and Arg. (B) Graphs represent mean ± s.e.m. of CrkII and CrkL tyrosine phosphorylation in wt, and in Abl-, Arg- or Abl/Arg-knockdown cells before and at various time points after PDGF stimulation. For CrkII, pY-CrkII was divided by total CrkII. For CrkL, pY-CrkL was normalized by pY-CrkL values in wt cells under serum starvation. ^‡P<0.05 (n=3). (C) NIH-3T3 cells were harvested before and at various time points after PDGF stimulation. WCL were subjected to immunoprecipitation (IP) with antibodies against CrkII, CrkL and p130Cas. WCL and immunoprecipitates were immunoblotted with various antibodies as indicated.

Fig. 3.

Role of domains and regulatory tyrosine of Crk adaptors in FA formation and PDGF-stimulated actin rearrangement. NIH-3T3 cells were infected with retrovirus carrying pSUPER vector only or one expressing shRNA targeting all Crk family genes. Cells were super-infected with MSCV-puro-derived retrovirus carrying EYFP or EYFP fused with wt or mutant Crk insensitive to shRNA. (A,B) Serum-starved cells were fixed before or 7.5 minutes after stimulation with 18 ng/ml PDGF. Fixed cells were stained with anti-vinculin–AlexaFluor 647, phalloidin–Texas-red, and Hoechst 33342 for vinculin, F-actin, and DNA, respectively. (A) Cell images are shown with vinculin (green), F-actin (red) and DNA (blue). Scale bar: 50 μm. (B) These images were enlarged from A. EYFP or EYFP fusion proteins are depicted in green and vinculin in red. Scale bar: 20 μm. (C) The area occupied by FAs (discrete spots visualized by vinculin staining) and total number of cells were quantified from fluorescent cell images. Graph represents mean ± s.e.m. of the relative change of area occupied by FAs in cells before and after PDGF stimulation. ^‡P>0.05 (n=3). (D) The numbers of total cells and cells forming circular dorsal-membrane ruffle(s) were counted from fluorescent cell images. Graph represents mean ± s.e.m. of percentages of cells with circular dorsal-membrane ruffle(s). ^‡P<0.05 (n=12).

Fig. 3.

Role of domains and regulatory tyrosine of Crk adaptors in FA formation and PDGF-stimulated actin rearrangement. NIH-3T3 cells were infected with retrovirus carrying pSUPER vector only or one expressing shRNA targeting all Crk family genes. Cells were super-infected with MSCV-puro-derived retrovirus carrying EYFP or EYFP fused with wt or mutant Crk insensitive to shRNA. (A,B) Serum-starved cells were fixed before or 7.5 minutes after stimulation with 18 ng/ml PDGF. Fixed cells were stained with anti-vinculin–AlexaFluor 647, phalloidin–Texas-red, and Hoechst 33342 for vinculin, F-actin, and DNA, respectively. (A) Cell images are shown with vinculin (green), F-actin (red) and DNA (blue). Scale bar: 50 μm. (B) These images were enlarged from A. EYFP or EYFP fusion proteins are depicted in green and vinculin in red. Scale bar: 20 μm. (C) The area occupied by FAs (discrete spots visualized by vinculin staining) and total number of cells were quantified from fluorescent cell images. Graph represents mean ± s.e.m. of the relative change of area occupied by FAs in cells before and after PDGF stimulation. ^‡P>0.05 (n=3). (D) The numbers of total cells and cells forming circular dorsal-membrane ruffle(s) were counted from fluorescent cell images. Graph represents mean ± s.e.m. of percentages of cells with circular dorsal-membrane ruffle(s). ^‡P<0.05 (n=12).

Fig. 4.

cSH3 domain and tyrosine phosphorylation of Crk are crucial for coordinated activation of Rac1 and Rap1 after PDGF stimulation. NIH-3T3 cells infected with retrovirus carrying pSUPER vector only or one with shRNA targeting all Crk family genes were super-infected with MSCV-puro-derived retrovirus expressing EYFP or EYFP fused with various types of Crk. Cells were harvested before and at 3 and 15 minutes after PDGF stimulation. WCL were subjected to GTPase assays. The amounts of Rac1 and Rap1 in pulldown fractions and their corresponding WCL were detected by immunoblotting. The amounts of GTP-bound relative to total GTPase were normalized to levels in wt cells under starvation. These values were used to obtain relative GTPase activities, and the means ± s.e.m. are represented by heat map graph. 1^‡ vs 2^‡ and 1^¶ vs 2^¶ in each column represent P<0.05 and P=0.065, respectively (n=3).