In the etiology of a significant number and diverse range of human tumors has been recently characterized as a protein that inhibits the invasive migration of human tumor cells through the control of MAL/SRF signaling. To gain further insight into how SCAI impacts on gene transcription, we have performed a screen for SCAI-interacting proteins. We show here that SCAI is functionally linked to SWI/ SNF complexes to promote changes in gene expression that may be critical for tumor cell invasion. However, the question of how SCAI impacts on gene 
regulation at a molecular level as well as the link between SCAI expression and tumor development is at present not fully understood. SCAI does neither share any sequence homologies to other known proteins nor any predicted domain architecture or intrinsic catalytic activity. Therefore, it seemed tempting to speculate that SCAI could serve as an adapter protein that recruits chromatin modifying enzymes, like Histone Deacetylases, Methytransferases, or ATP-utilizing chromatin remodeling enzymes to specific genomic regions and thereby controls expression of target genes. To identify the molecular link between SCAI and the dynamic regulation of the chromatin architecture we have performed an affinity screen for SCAI-interacting proteins. A fragment of SCAI comprising amino acids 35�C280 was used as bait protein. SCAI-interacting proteins in high salt fraction of mouse brain lysate were separated and analyzed by mass spectroscopy analysis. The data showed proteins, mainly involved in histone modifications and having ATPase and DNA helicase activities. Among these, 6 subunits of the SWI/SNF complex associated with SCAI. We were able to further confirm this potential interaction by coimmunoprecipitation experiments. SCAI and BRM, the central core ATPase subunit of the human SWI/SNF complex, were expressed in HEK 293 cells. SCAI was immunoprecipitated and the precipitates were analyzed for the presence of BRM. Interestingly, the N-terminal fragment comprising amino acids 1�C212, a region that we have previously characterized as a critical region for its biologically activity, was sufficient and required for interaction with BRM, whereas a construct lacking the N-terminus did not co-immunoprecipitated with BRM. We were also able to map the N-terminal 360 amino acids of BRM as the region required and sufficient to interact with SCAI by co-immunoprecipitation experiments. However, we have not been able to see association of endogenous BRM and SCAI, indicating that SCAI could be a substoichiometric, nonobligate partner for BRM and that this complex is only operative at certain promoters. Our data further indicate that SCAI requires the presence of a AbMole Octinoxate functional SWI/SNF complex to suppress promoter activity. We performed SRF-dependent reporter gene assays in SW13 cells, a human adrenal adenocarcinoma cell line that lacks expression of BRM and the closely related BRG1 protein.