Phosphorylation site mutants of PAK6-5A and PAK6-5E exhibit altered ability to interact with AR and androgen-stimulated kinase activation.

<p>(A) Interactions between AR and PAK6-5A, PAK6-5E are assessed by a mammalian two-hybrid assay. Full-length AR is cloned in-frame downstream of VP16 transactivation domain with a HA-tag using a pACT vector. PAK6-wt, -5A and 5E were cloned in-frame downstream of GAL4 DBD with a HA tag using a pBIND vector. A mammalian two-hybrid assay was performed by co-transfecting HEK-293cells with pACT-AR or AR truncated mutants, as indicated, with pBIND-PAK6 and pG5-Luc vector. Cells were treated with vehicle control or DHT (10 nM) for 20 h followed by dual luciferase assay. A renilla luciferase reporter activity was used as a transfection internal control. The luciferase activity was normalized by the internal control and the vector basal control groups. Data represent the mean ± SEM (n=3, *<i>p</i><0.01 compared to their respective vehicle controls; Student’s <i>t</i>-test). (B) A reciprocal immunoprecipitation assay was performed using AR and PAK6-WT, -5A and -5E mutants. HEK-293 cells were transfected with equal amount of AR and PAK6 expression vectors in the presence or absence of 10 nM DHT. Twenty hours post-transfection either AR or PAK6 was immunoprecipitated using antibodies specifically for either molecule then reciprocally probed by western blot for presence of co-precipitated interacting proteins as depicted. (C) A representative experiment of androgen-stimulated activation of PAK6-WT, PAK6-5A and PAK6-5E in an <i>in </i><i>vitro</i> kinase assay. The upper phosphorylation band is PAK6 autophosphorylation. The lower bands are the phosphorylation of the exogenous substrate histone H4. (D) Quantification of relative kinase activities of in vitro kinase assays from (C). Data represent the mean ± SEM (n=4, *<i>p</i><0.01 compared to their respective vehicle controls; Student’s <i>t</i>-test).