Immunodepletion of p35Nck5a reduces Cdk5-associated kinase activity.
(A) Cdk5-associated kinase activity on 2 in vitro substrates histone H1 (H1) and the retinoblastoma protein (pRb) is shown for lysates of undifferentiated HL60 cells (lanes 1, 3, and 5), and lysates of HL60 cells treated with 10−7M 1,25D3 for 96 hours (lanes 2, 4, 6, and 7). The first 2 lanes show Cdk5-associated kinase activity of the cell lysates, lanes 3 and 4 the Cdk5-associated kinase activity after immunodepletion of the lysate with an antibody to cyclin D1, and lanes 5 and 6 after immunodepletion of the lysate with an antibody to p35Nck5a. Lane 7 shows a nonspecific signal that was obtained by subjecting the lysates of HL60 cells differentiated with 1,25D3 to the immunoprecipitation procedure but omitting the Cdk5 antibody from the IP protocol. Although the lower panel (pRb substrate) shows no signal in the scanned image, a weak signal was evident on the autograms in lanes 5 and 6. (B) Four autoradiograms from replicate experiments were scanned by a phosphoimager to quantitate Cdk5-associated H1 kinase activity (left panel), and pRb kinase activity (right panel). T indicates 1,25D3-treated cells; C, untreated control cells. The differences between values in lanes 2 versus 4 (the pairs identified by an asterisk) were not statistically significant (P < .05) for both substrates (histone H1 and pRb), whereas the differences between pairs identified with the symbol # (lanes 2 and 6) were highly significant (P > .01). (C) Lane 1, Cdk5-associated kinase activity from untreated HL60 cells using histone H1 as the substrate; lane 2, Cdk5-associated kinase activity from HL60 cells treated with 10−7M 1,25D3 for 48 hours; lane 3, as lane 2, but in the presence of 2 μg Ab to p35Nck5a; lane 4, as lane 2, but in the presence of 2 μg Ab to cyclin D1; lane 5 background, determined as in the last lane of Figure 3A. (D) Quantitation of experiments illustrated in Figure 3C, performed as described for Figure 3B.
