Remote hot spots mediate protein substrate recognition for the Cdc25 phosphatase

Abstract

Cdc25B is a phosphatase that catalyzes the dephosphorylation and activation of the cyclin-dependent kinases, thus driving cell cycle progression. We have identified two residues, R488 and Y497, located >20 A from the active site, that mediate protein substrate recognition without affecting activity toward small-molecule substrates. Injection of Cdc25B wild-type but not the R488L or Y497A variants induces germinal vesicle breakdown and cyclin-dependent kinase activation in Xenopus oocytes. The conditional knockout of the cdc25 homolog (mih1) in Saccharomyces cerevisiae can be complemented by the wild type but not by the hot spot variants, indicating that protein substrate recognition by the Cdc25 phosphatases is an essential and evolutionarily conserved feature.

Fig. 1.

Surface of Cdc25B showing the mutations. The image was generated with visual molecular dynamics (28).

Fig. 2.

Activity of Cdc25B mutants with pNPP, mFP, and protein substrate. Kinetic data for mutants of Cdc25B are shown as k_cat, k_cat/K_m, and Cdk2-pTpY/CycA substrate (k_cat/K_m) as normalized by activity with pNPP (k_cat/K_m).

Fig. 3.

K_m determination of Cdk2-pTpY/CycA for Cdc25B. The wild-type protein (open squares) yielded a K_m of 440 ± 80 nM, whereas both hot spot mutants, Y497A (open triangles) and R488L (filled circles), were linear up to the highest achievable concentration, as seen in Inset. Data shown are representative of three independent experiments.

Fig. 4.

Inhibition kinetics of catalytically inactive Cdc25B mutants. Activity of wild-type Cdc25B (3 nM) against Cdk2-pTpY/CycA (70 nM) was measured in the presence of varying concentrations (0.01–30 μM) of the catalytically inactive Cdc25B constructs [C473S (filled circles), C473D (filled squares), C473D-Y497A (open triangles), and C473S-Y497A (open circles)]. The percent activity observed at each given competitor concentration was fitted to percent activity = 100/(1 + [I]/IC₅₀) where [I] is the concentration of inhibitor and IC₅₀ values were calculated [C473S, 70 nM (apparent); C473D, 2.5 ± 0.5 μM; C473D/Y497A and C473S/Y497A, >150 μM). The data shown are mean values of at least three separate experiments.

Fig. 5.

Induction of GVBD and activation of Cdc2-histone kinase activity in Xenopus oocytes. (A) Cdc25 (1 μM final) was injected in stage-VI oocytes (14–18 replicates per enzyme form), and GVBD was monitored by visual inspection. (B) Histone kinase activity was tested from one to two pooled oocytes. (C) The presence of Cdc25B was demonstrated by Western blotting from six pooled oocytes.

Fig. 6.

Hot spot mutants do not rescue conditional mih1 yeast deletion strain. Yeast strain JMY1290 (hsl7Δ GAL-MIH1 leu2) transformed with a vector containing myc-tagged mih1 wild type, R488L, Y497A, or empty vector control as grown in media containing glucose (Upper) or galactose (Lower). Note the elongated, nondividing phenotype in the strains that do not express functional Mih1 (Cdc25). (B) Western blot demonstrating expression of Mih1 by means of the C-terminal Myc tag in all but the control strain.