Cks1 is required for G(1) cyclin-cyclin-dependent kinase activity in budding yeast

Abstract

p13(suc1) (Cks) proteins have been implicated in the regulation of cyclin-dependent kinase (CDK) activity. However, the mechanism by which Cks influences the function of cyclin-CDK complexes has remained elusive. We show here that Cks1 is required for the protein kinase activity of budding yeast G(1) cyclin-CDK complexes. Cln2 and Cdc28 subunits coexpressed in baculovirus-infected insect cells fail to exhibit protein kinase activity towards multiple substrates in the absence of Cks1. Cks1 can both stabilize Cln2-Cdc28 complexes and activate intact complexes in vitro, suggesting that it plays multiple roles in the biogenesis of active G(1) cyclin-CDK complexes. In contrast, Cdc28 forms stable, active complexes with the B-type cyclins Clb4 and Clb5 regardless of whether Cks1 is present. The levels of Cln2-Cdc28 and Cln3-Cdc28 protein kinase activity are severely reduced in cks1-38 cell extracts. Moreover, phosphorylation of G(1) cyclins, which depends on Cdc28 activity, is reduced in cks1-38 cells. The role of Cks1 in promoting G(1) cyclin-CDK protein kinase activity both in vitro and in vivo provides a simple molecular rationale for the essential role of CKS1 in progression through G(1) phase in budding yeast.

FIG. 1

Yeast extract activates baculovirus-expressed Cln2-Cdc28 protein kinase. Lysates from Sf9 cells (5 μg) coinfected with recombinant baculoviruses encoding Cln2^MH6 and Cdc28^HA were incubated either alone (lane 5) or in the presence of 18, 54, and 180 μg of extract from Cln-depleted cdc28^ts cells (lanes 6 to 8, respectively; relevant genotype: cdc28^ts cln1Δ cln2Δ cln3Δ GAL-CLN3). In a parallel experiment, 180 μg of yeast extract was incubated alone (lane 1) or in the presence of 2.5, 5, and 7.5 μg of lysate from baculovirus-infected Sf9 cells expressing Cln2^MH6 plus Cdc28^HA (lanes 2 to 4, respectively). Following incubation for 15 min at 24°C in the presence of an ATP-regenerating system, all reaction mixtures were immunoprecipitated with anti-HA monoclonal antibody 12CA5, assayed for histone H1 kinase activity, fractionated on an SDS-polyacrylamide gel, and quantitated with a PhosphorImager. Relative kinase activities in lanes 1 to 8 are 1, 9, 14, 15, 2, 4, 7, and 17, respectively.

FIG. 2

Cks1 activates Cln2-Cdc28 protein kinase. (A) Cks1 substitutes for yeast extract in the activation of Cln2^MH6-Cdc28^HA. Extracts prepared from Sf9 cells coinfected with Cln2^MH6- and Cdc28^HA-expressing baculoviruses were mixed with the indicated components, incubated at 24°C for 15 min, immunoprecipitated with anti-HA monoclonal antibody 12CA5, and assayed for histone H1 protein kinase activity. Relative kinase activities in lanes 1 to 5 are 1, 11, 14, 12, and 14, respectively. U and B, E. coli lysate (40 μg) containing Cks1 or yeast extract (100 μg) that was either untreated (U) or boiled for 5 min (B). (B) Purified Cks1. Cks1 (10 μg) purified from E. coli was fractionated on an SDS–15% polyacrylamide gel and stained with Coomassie blue. (C) Titration of Cks1. The indicated amounts of Cks1 were mixed with lysate (5 μg) prepared from Sf9 cells coinfected with recombinant baculoviruses encoding Cln2^MH6 and Cdc28^HA, and the mixtures were processed as described for panel A. Relative kinase activities in lanes 1 to 4 are 22, 16, 2, and 1, respectively.

FIG. 3

Cks1 binds to Cln2-Cdc28 and activates phosphorylation of multiple substrates. (A) Lysates from insect cells coinfected with baculovirus vectors encoding Cln2, GST-Cdc28^HA, and Cks1^His6 were adsorbed to glutathione resin, and specifically bound proteins were revealed by staining with Coomassie blue (CB) or immunoblotting with anti-His6 antibodies (IB). Arrowheads, migration of molecular mass markers (from top to bottom, 68, 46, 30, and 21.5 kDa). (B) Lysates (40 μg) of Sf9 cells infected with recombinant baculoviruses encoding Cln2^MH6 or Cdc28^HA, as indicated, were either mock treated or mixed with 390 ng of purified Cks1 for 15 min at 24°C. Reaction mixtures were immunoprecipitated (I.P.) with anti-HA monoclonal antibody 12CA5, and immunoprecipitates were divided in thirds and assayed for their quantity of histone H1 kinase (top), Cln2^MH6 kinase (middle), or Cln2^MH6 antigen (bottom; detected by immunoblotting with affinity-purified anti-Cln2 polyclonal antibody). (C) Lanes 1 to 4, immunoprecipitates prepared as described for panel B were directly assayed for Far1 or MBP-Sic1 or histone H1 kinase activity. Lanes 5 to 8, Cks1-free immunoprecipitates prepared from Sf9 lysates were subdivided into aliquots which were then either mock treated or supplemented with 390 ng of Cks1. After a brief incubation, kinase complexes were washed and assayed for Far1, MBP-Sic1, or histone H1 kinase activity as indicated. All samples were also evaluated by immunoblotting with an antimyc monoclonal antibody (bottom).

FIG. 3

Cks1 binds to Cln2-Cdc28 and activates phosphorylation of multiple substrates. (A) Lysates from insect cells coinfected with baculovirus vectors encoding Cln2, GST-Cdc28^HA, and Cks1^His6 were adsorbed to glutathione resin, and specifically bound proteins were revealed by staining with Coomassie blue (CB) or immunoblotting with anti-His6 antibodies (IB). Arrowheads, migration of molecular mass markers (from top to bottom, 68, 46, 30, and 21.5 kDa). (B) Lysates (40 μg) of Sf9 cells infected with recombinant baculoviruses encoding Cln2^MH6 or Cdc28^HA, as indicated, were either mock treated or mixed with 390 ng of purified Cks1 for 15 min at 24°C. Reaction mixtures were immunoprecipitated (I.P.) with anti-HA monoclonal antibody 12CA5, and immunoprecipitates were divided in thirds and assayed for their quantity of histone H1 kinase (top), Cln2^MH6 kinase (middle), or Cln2^MH6 antigen (bottom; detected by immunoblotting with affinity-purified anti-Cln2 polyclonal antibody). (C) Lanes 1 to 4, immunoprecipitates prepared as described for panel B were directly assayed for Far1 or MBP-Sic1 or histone H1 kinase activity. Lanes 5 to 8, Cks1-free immunoprecipitates prepared from Sf9 lysates were subdivided into aliquots which were then either mock treated or supplemented with 390 ng of Cks1. After a brief incubation, kinase complexes were washed and assayed for Far1, MBP-Sic1, or histone H1 kinase activity as indicated. All samples were also evaluated by immunoblotting with an antimyc monoclonal antibody (bottom).

FIG. 4

Cks1 is not required for protein kinase activity of Clb-Cdc28^HA complexes. Sf9 cells were coinfected with a recombinant baculovirus encoding Cdc28^HA plus an additional virus encoding either Cln2^MH6, Clb4, or Clb5, as indicated. Lysates of infected cells were either mock treated or supplemented with 390 ng of purified Cks1, incubated at 24°C for 10 min, and immunoprecipitated with anti-HA monoclonal antibody 12CA5. Immunoprecipitates were assayed for their content of histone H1 kinase activity, which was assessed by SDS-PAGE followed by phosphorimaging. In this experiment, the relative maximal H1 kinase activities obtained were 1.0 for Clb4-Cdc28, 0.69 for Cln2-Cdc28, and 0.42 for Clb5-Cdc28.

FIG. 5

Mutant cks1-38 cells fail to assemble active Cln2-Cdc28 protein kinase complexes. (A) Absence of active Cln2^3×HA-Cdc28 complexes in cks1-38 cell extract. Total cell extract proteins (3 mg) prepared from CLN2 CKS1, CLN2^3×HA CKS1, and CLN2^3×HA cks1-38 cells grown at either the permissive (P; 24°C) or restrictive (R; 38.5°C) temperature were immunoprecipitated (IP) with anti-HA monoclonal antibody 12CA5, and immunoprecipitates were divided in half and assayed for their content of histone H1 kinase activity (top) or Cdc28 antigen (bottom). Samples in even-numbered lanes were supplemented with 3.9 μg of purified Cks1 prior to the immunoprecipitation step. Relative kinase activities in lanes 1 to 12 (top) are 1, 1, 2, 1, 25, 29, 15, 23, 2, 20, 2, and 8, respectively. (B) cks1-38 extracts contain high levels of Clb2-Cdc28 kinase activity. Total cell extract proteins (1 mg) from the experiment described for panel A were incubated with anti-Clb2 affinity-purified polyclonal antibodies, and immunoprecipitates were assayed for their content of histone H1 kinase activity. MBP-Clb2, same as sample shown in lane 1, except 20 μg of purified MBP-Clb2 was added prior to the immunoprecipitation step; 3× α-Clb2, same as sample shown in lane 1, except threefold more antibody was used. These two controls indicate that immunoprecipitation was specific and that the antibody was used at saturating levels for the assays depicted in lanes 1 to 4. The relative protein kinase activities in lanes 1 to 6 are 18, 6, 24, 36, 1, and 13, respectively. (C) Accumulation of phosphorylated Cln2^3×HA is strongly delayed in cks1-38 cells released from a pheromone arrest. CKS1 (odd-numbered lanes) and cks1-38 (even-numbered lanes) cultures were arrested in G₁ phase with α-factor, shifted to 38.5°C for 1 h, and then released by being washed into medium lacking α-factor at 38.5°C. Lysates (100 μg) prepared from cells withdrawn at the indicated times were fractionated by SDS-PAGE and immunoblotted with anti-HA monoclonal antibody 12CA5 to detect Cln2^3×HA. Note that both the level and phosphorylation state of Cln2 are diminished in cks1-38 cells. (D) Cln2^3×HA-Cdc28 activity fails to accumulate in cks1-38 cells released from a pheromone arrest. Samples (3 mg) from the experiment shown in panel C were immunoprecipitated with 12CA5. Immunoprecipitates were assayed for their content of histone H1 kinase activity, which was evaluated by SDS-PAGE followed by phosphorimaging.

FIG. 6

Mutant cks1-38 cells fail to assemble active Cln3-Cdc28 protein kinase complexes. (A) GAL-CLN3^3×HA CKS1 and GAL-CLN3^3×HA cks1-38 cultures grown in galactose medium at 25°C were split, and half of each culture was shifted to 38.5°C for 3 h. Total cell extract proteins (10 mg) prepared from cells maintained at 25°C (P) or shifted to 37°C (R) were immunoprecipitated with anti-HA monoclonal antibody 12CA5, and immunoprecipitates were divided in thirds and assayed for their content of Cdc28 (B) and Cln3^3×HA antigens (not shown) by immunoblotting and for histone H1 kinase activity (A). Kinase activity shown in panel A is normalized to Cln3^3×HA levels. (C) Cln3^3×HA present in total cell lysate of untagged controls (lanes 1 and 4), GAL-CLN3^3×HA CKS1 cells (lanes 2 and 3), and GAL-CLN3^3×HA cks1-38 cells (lanes 5 and 6) was evaluated by immunoblotting with anti-HA.