The genome of avian sarcoma virus CT10 encodes a fusion protein in which viral Gag sequences are fused to cellular Crk sequences containing primarily Src homology 2 (SH2) and Src homology 3 (SH3) domains. Transformation of chicken embryo fibroblasts (CEF) with the Gag-Crk fusion protein results in the elevation of tyrosine phosphorylation on specific cellular proteins with molecular weights of 130,000, 110,000, and 70,000 (p130, p110, and p70, respectively), an event which has been correlated with cell transformation. In this study, we have identified the 70-kDa tyrosine-phosphorylated protein in CT10-transformed CEF (CT10-CEF) as paxillin, a cytoskeletal protein suggested to be important for organizing the focal adhesion. Tyrosine-phosphorylated paxillin was found to be complexed with v-Crk in vivo as evident from coimmunoprecipitation studies. Moreover, a bacterially expressed recombinant glutathione S-transferase (GST)-CrkSH2 fragment bound paxillin in vitro with a subnanomolar affinity, suggesting that the SH2 domain of v-Crk is sufficient for binding. Mapping of the sequence specificity of a GST-CrkSH2 fusion protein with a partially degenerate phosphopeptide library determined a motif consisting of pYDXP, and in competitive coprecipitation studies, an acetylated A (p) YDAPA hexapeptide was able to quantitatively inhibit the binding of GST-CrkSH2 to paxillin and p130, suggesting that it meets the minimal structural requirements necessary for the interaction of CrkSH2 with physiological targets. To investigate the mechanism by which v-Crk elevates the tyrosine phosphorylation of paxillin in vivo, we have treated normal CEF and CT10-CEF with sodium vanadate to inhibit protein tyrosine phosphatase activity. These data suggest that paxillin is involved in a highly dynamic kinase-phosphatase interplay in normal CEF and that v-Crk binding may interrupt this balance to increase the steady-state level of tyrosine phosphorylation. By contrast, the 130-kDa protein was not tyrosine phosphorylated upon vanadate treatment of normal CEF and only weakly affected in the CT10-CEF, suggesting that a different mechanism may be involved in its phosphorylation.