(Received for publication, August 28, 1996, and in revised form, November 15, 1996)
From the Molecular Oncology Group, Nippon Roche Research Center, 200 Kajiwara, Kamakura, Kanagawa 247, Japan and the § First Department of Internal Medicine, Saitama Medical Center, Saitama Medical School, 1981 Tsujido, Kawagoe, Saitama 350, Japan
Cyclin-dependent kinase (Cdk) inhibitory proteins are involved in cell cycle arrest induced by antiproliferating factors or chemicals. High cell density also induces cell cycle arrest in which the genomic DNA is unreplicated, even in the presence of a mitotic dose of growth factors; this is termed contact inhibition. Although the cell cycle of the rat fibroblast cell line, 3Y1, was arrested in quiescence by contact inhibition, the Cdk4 bound to its regulatory subunit, cyclin D1 or D3. However, these complexes were enzymatically inactive. Phosphorylation of the cyclin D1-bound Cdk4 by the Cdk-activating kinase could convert the inactive cyclin D1-Cdk4 complex into its active form in vitro, suggesting that threonine 172 of the Cdk4, of which phosphorylation is required for its activation, was in part unphosphorylated in contact-inhibited 3Y1 cells. Although MO15 was active in cell extracts prepared from the arrested 3Y1 cells, activation of bacterially produced Cdk4 in the cell extracts was inhibited. Removal of p27kip1 from the cell extracts allowed the MO15 holoenzyme to phosphorylate the Cdk4 and in turn activate it, indicating that p27kip1 plays a role in inhibiting the phosphorylation of Cdk4 by MO15 in the contact-inhibited 3Y1 cells.
Sequential activation and inactivation of cyclin-dependent kinases (Cdks)1 govern the progression and the transitions of the cell cycle in eukaryotic cells. Their activity is regulated through phosphorylation of tyrosine or threonine residues and by the binding of inhibitors or regulatory subunits, called cyclins. In turn, the cyclins are regulated by their expression and degradation. In mammalian cells, as quiescent cells enter the cell cycle in response to growth factor stimulation, D-type cyclins (D1, D2, and D3) synthesized early in the G1 phase of the cell cycle form enzymatically active complexes with cyclin-dependent kinase 4 (Cdk4) or Cdk6 (1, 2). These complexes catalyze phosphorylation of the retinoblastoma gene product, pRb (2, 3), whose phosphorylation is necessary for cells to enter the S phase. Deprivation of growth factor during the G1 phase results in a quiescent state or apoptosis. Microinjection of antibody or antisense DNA to cyclin D1 into G1 phase cells prevents them from entering the S phase, whereas injection into G1/S transition cells shows no such effect (4, 5), suggesting that the cyclin D-Cdk4 complex exerts its effect during the middle to late G1 phase. To become a fully active holoenzyme, Cdk4 must bind a D-type cyclin as well as undergo phosphorylation at Thr-172 (6). The Cdk-activating kinase (CAK), composed of cyclin H and MO15 (alias Cdk7) (7-11), phosphorylates not only Thr-161 in Cdc2 and Thr-160 in Cdk2 (10, 11) but also Thr-172 in Cdk4 (12).
Cdk-inhibitory proteins (CKIs), induced by various antimitotic signals,
prevent cells from progressing through the cell cycle; overexpression
of the CKIs induces cell cycle arrest, suggesting that the induction of
those CKIs triggers the arrest. Two families of CKIs have been
identified in mammalian cells. The first family, called INK4, includes
p16INK4A (13), p15INK4B (14), p18INK4C (15,
16), and p19INK4D (16, 17). It is composed of a repeat
ankyrin-like sequence and selectively inhibits Cdk4 and Cdk6 by binding
to the Cdk subunit alone. The second family includes
p21cip1/waf1/sdi1 (18-22), p27kip1 (23, 24), and
p57kip2 (25, 26), which can inhibit a broad range of cyclin-Cdk
complexes. p27kip1 plays a physiologically important role in
the regulation of cell proliferation in many tissues (27-29) and is
induced by various antimitotic signals, such as transforming growth
factor- (30-32), cAMP (33), rapamycin (34), serum deprivation (35,
36), and contact inhibition (30). In quiescent cells, the level of p27kip1 protein is elevated. Accumulation of p27kip1 is
correlated with inactivation of the G1 cyclin-Cdk complexes and results in G1 arrest. In cyclic AMP-treated
macrophages, an increased amount of p27kip1 prevents the
activation of the cyclin D-Cdk4 complex and induces G1
arrest (33). During T cell mitogenesis, antigen stimulation promotes
the synthesis of cyclins and Cdks, after which interleukin-2 allows for
the activation of the cyclin-Cdk complexes by decreasing the
p27kip1 level. This is prevented by rapamycin (34). High cell
density also induces cell cycle arrest with unreplicated genomic DNA. This phenomenon, termed contact inhibition, is generally observed in
normal or untransformed cells, and loss of it is one of the remarkable
properties of transformed cells. In the contact-inhibited rat
fibroblast cell line, 3Y1, we showed that the cyclin
D-dependent kinase was inactivated and that its
inactivation was in part associated with the inhibition of the access
of the Cdk-activating kinase to Cdk4 by p27kip1. During
resumption of the G1 phase through the release of the cells
from contact inhibition, we showed that timing of the activation of the
cyclin D-dependent kinase correlated well with the
reduction of the p27kip1.
Exponentially growing 3Y1 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (FBS), 4 mM L-glutamine, and 40 µg/ml kanamycin. Confluent cells (described in our experiments as arrested cells by contact inhibition) were maintained for 4 more days. Cells released from the contact inhibition by trypsinization (growing cells) were diluted 7-10-fold with Dulbecco's modified Eagle's medium containing 10% FBS and maintained for 48 h. For analysis of the DNA content, monolayer cells were trypsinized and suspended in a 1-ml solution containing 0.1% sodium citrate, 0.1% Triton X-100, and 50 µg/ml of propidium iodide and treated for 30 min at 37 °C with 10 µg of RNase A. The stained cells were analyzed by a FACScan flow cytometer (Becton Dickinson, San Jose, CA) and the distribution of cells at each stage of the cycle was determined.
AntibodiesAntisera to murine p27kip1, murine Cdk4,
and human cyclin H were prepared by immunizing rabbits with a
bacterially produced glutathione S-transferase
(GST)-p27kip1, GST-Cdk4, or GST-cyclin H fusion protein,
respectively. Neither murine p21cip1/waf1/sdi1 nor murine
p57kip2 transcribed/translated in vitro was
immunoprecipitated by the antibody to p27kip1. Mouse monoclonal
antibody (mAb) to Cdk4 was raised to a synthetic oligopeptide
corresponding to the C-terminal 20 amino acids of murine Cdk4 by a
standard procedure. Antiserum to murine MO15 was prepared by immunizing
rabbits with a multiple antigenic peptide-conjugated synthetic
oligopeptide corresponding to the 15 amino acids at the C terminus of
murine MO15. Mouse mAb to cyclin D1, rat mAb to cyclin D3, and an
antiserum to a synthetic C-terminal peptide of murine Cdk4 (Rz) were
generous gifts from C. J. Sherr (St. Jude Children's Research
Hospital, Memphis, TN). Insect cells, Spodoptera frugiperda
(Sf9) (1 × 106 cells), infected with recombinant
baculoviruses encoding murine MO15 or human cyclin H alone,
respectively, or infected with the MO15 baculovirus together with the
cyclin H baculovirus for 48 h with a multiplicity of infection 3 were metabolically labeled for 2 h with 50 µCi of
[35S]methionine (ICN, Irvine, CA) in 1 ml of
methionine-free medium supplemented with 10% dialyzed FBS. Cells were
harvested by centrifugation and lysed with 1 ml of Nonidet P-40 lysis
buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl,
1% Nonidet P-40, 0.5% sodium deoxycholate) containing 5 µg/ml of
aprotinin, 0.1 mM phenylmethylsulfonyl fluoride, 0.1 mM NaF, 10 mM -glycerophosphate, and 0.1 mM sodium orthovanadate. Clarified cell lysates were
precipitated for 12 h at 4 °C with antiserum to cyclin H or
MO15. Immunoprecipitates were brought down with protein A-Sepharose
beads (Pharmacia Biotech AB, Uppsala, Sweden), washed three times with
Nonidet P-40 lysis buffer, boiled in sample buffer, separated on
polyacrylamide gels containing SDS, and detected by
autoradiography.
Cells were rinsed three times in ice-cold
phosphate-buffered saline and lysed in Tween 20 lysis buffer (50 mM HEPES (pH 7.5), 150 mM NaCl, 1 mM EDTA, 2.5 mM EGTA, 1 mM
dithiothreitol, 0.1% Tween 20, 10% glycerol) containing protease and
phosphatase inhibitors. Sonicated extracts were clarified by
centrifugation, and the supernatants were precipitated for 6 h at
4 °C with the indicated rabbit antiserum or mAb. Immunoprecipitates
were brought down with protein A- or protein G-Sepharose beads
(Pharmacia Biotech AB) for 45 min at 4 °C, and the beads were washed
with Tween 20 lysis buffer. For immunoblotting, the beads were boiled
in sample buffer, separated on SDS-polyacrylamide gels, transferred to
nitrocellulose, and immunoblotted with rabbit antisera
(anti-p27kip1 antibody) or mAbs (anti-cyclin D1, D3, or Cdk4
antibody). Immunoblotting analyses were performed using ECL enhanced
chemiluminescence (Amersham International Plc, Little Chalfont, UK),
according to the manufacturer's instructions. The assay for pRb kinase
activity was described previously (3). In brief, cells were lysed in
Tween 20 lysis buffer, and lysates were incubated with the indicated
antibody. Recovered cyclin D-Cdk4 complexes on protein A- or protein
G-Sepharose were incubated for 30 min at 30 °C in 20 µl of kinase
buffer (50 mM HEPES (pH 8.0), 10 mM
MgCl2, 1 mM dithiothreitol, 2.5 mM
EGTA, 10 mM -glycerophosphate, 0.1 mM NaF,
0.1 mM sodium orthovanadate) supplemented with 20 µM ATP, 10 µCi of [
-32P]ATP
(Amersham), and 0.5 µg of purified GST-pRb substrate. Reactions were
stopped by boiling in sample buffer and then separated on a 10%
polyacrylamide gel. Phosphorylated proteins were visualized by
autoradiography. For Fig. 2d, recovered cyclin D-Cdk4
complexes on protein G-Sepharose were resuspended with kinase buffer
supplemented with 20 µM ATP, incubated for 15 min at
30 °C with 5 µl of insect cell lysates infected with the indicated
baculoviruses for 48 h, washed three times with Tween 20 lysis
buffer and once with 50 mM HEPES (pH 8.0), and assayed for
pRb kinase activity.
CAK Assay
The immune complex CAK assay was performed as
previously reported (12). In brief, cells were lysed by sonication in
CAK buffer (80 mM sodium--glycerophosphate, 20 mM EGTA, 15 mM MgCl2, 5 mM dithiothreitol) containing protease and phosphatase
inhibitors and were clarified for 20 min in a microcentrifuge.
Supernatants were incubated for 6 h at 4 °C with antiserum to
MO15, and protein A-Sepharose beads were added. After rocking the
mixture for 30 min at 4 °C, we washed the beads twice with Tween 20 lysis buffer and twice with CAK buffer. They were then resuspended in
50 µl of CAK buffer with 1 mM ATP containing bacterially
produced cyclins and GST-Cdks and incubated for 1 h at 23 °C.
After centrifugation, cyclin-Cdk complexes in the supernatants were
recovered with glutathione-Sepharose beads (Pharmacia Biotech AB),
washed four times with Tween 20 lysis buffer and once with HEPES (pH
8.0), and assayed for pRb kinase activity. The cells described in Fig.
3d (0.2 × 106 cells/lane) were lysed in
CAK buffer, clarified by centrifugation, and the supernatants were
mixed and incubated with the indicated amounts of recombinant
His-tagged cyclin D2 and GST-Cdk4 in the presence of 1 mM
ATP. The cyclin D2-Cdk4 complex bound to glutathione-Sepharose beads
was collected by centrifugation and assayed for pRb kinase activity.
The arrested cells described in Fig. 4b (0.5 × 106 cells/lane) were lysed and immunodepleted by incubation
with protein A-Sepharose beads precoated with saturating amounts of antiserum to p27kip1 or preimmune serum. After centrifugation,
the supernatants were incubated with both cyclin H-MO15 complexes
recovered on protein A-Sepharose beads from Sf9 cells producing the
cyclin H-MO15 holoenzyme and bacterially produced His-cyclin
D2/GST-Cdk4 (0.6 µg each). Then the immune complex CAK assay was
performed.
In the presence of 10% FBS, the
rat fibroblast cell line, 3Y1, grows exponentially and forms a
monolayer on culture plates until the cells come in contact with each
other. These cells then enter the quiescent phase of the cell cycle
with no replication of the genomic DNA (Fig.
1a). This phenomenon is termed contact inhibition. When these arrested cells were released from their contact
inhibition by trypsinization and then dilution in 10% FBS, the cells
resumed their cell cycle progression. 48 h after the release, flow
cytometric measurements of their DNA contents revealed growing cells in
various phases of the cell cycle (Fig. 1a). Asynchronously
growing 3Y1 expressed cyclin D1 and D3, and both of the D-type cyclins
formed enzymatically active complexes with Cdk4 (Fig. 1, b
and c). Although cyclin D1 and D3 were expressed and bound
to Cdk4 in the arrested 3Y1 cells as well as in the asynchronously
growing cells, neither of the complexes was active (Fig. 1,
b and c).
Activation of the Cyclin D-dependent Kinase in Vitro by the Reconstituted Cyclin H-MO15 Holoenzyme
The cyclin D-Cdk4 complex is catalytically inactive until it undergoes phosphorylation at threonine 172 in the cyclin D-bound Cdk4 (6); this raises the possibility that phosphorylation of the Cdk4 could be inhibited in the arrested 3Y1 cells. Since the CAK, composed of cyclin H and MO15, is known to phosphorylate threonine 172 in the Cdk4 (12), we made polyclonal antibodies to cyclin H and MO15. Antiserum to MO15 precipitated the MO15 as well as the cyclin H-MO15 complex produced in insect cells (Sf9) infected with recombinant cyclin H and MO15 baculoviruses. The antiserum to cyclin H precipitated only cyclin H (Fig. 2a). Given the failure of the cyclin H antiserum to recognize the complex, the immune complex kinase assay was performed with the antiserum to MO15. The immunoprecipitated cyclin H-MO15 complex was subjected to the kinase assay with bacterially produced Cdk2 or Cdk4, fused to GST. GST-Cdk2 was phosphorylated by the cyclin H-MO15 complex immunoprecipitated with antisera to MO15 (Fig. 2b). By contrast, GST-Cdk4 was not appreciably phosphorylated by the immunoprecipitated cyclin H-MO15 complex in both the absence and the presence of His-tagged cyclin D2 (negative data not shown). Since GST-Cdk4 could be activated by the cyclin H-MO15 complex as a pRb kinase (see below), we assumed that a major fraction of the bacterially produced GST-Cdk4 was denatured and that a minor fraction of the GST-Cdk4 was phosphorylated and activated by the cyclin H-MO15 complex.
To assess the activity of the cyclin H-MO15 complex produced in Sf9 cells as CAK, we performed an immune complex CAK assay by using bacterially produced GST-cyclin A plus GST-Cdk2 or His-tagged cyclin D2 plus GST-Cdk4 for the first step substrate and histone H1 or GST-pRb for the second step substrate, respectively (12). The cyclin-Cdk complex yielded a significantly higher level of histone H1 or pRb kinase activity (Fig. 2c), indicating that the cyclin H-MO15 complex reconstituted in Sf9 cells was bona fide CAK and that the antiserum to MO15 could precipitate active CAK. To examine whether the cyclin D-Cdk4 complex in the contact-inhibited cells was inactive due to the threonine unphosphorylation in cyclin D-bound Cdk4, we incubated the immunoprecipitated cyclin D-Cdk4 complex from the arrested cells with insect cell lysates producing CAK or with suitable control cell lysates in CAK kinase buffer before running the pRb kinase assay. We then incubated the cyclin D-Cdk4 complex on protein A-Sepharose beads for a short period of time at a high temperature (see "Experimental Procedures") to reduce an effect of endogenous CAK activity in insect cells. Preincubation of the immunoprecipitated cyclin D-Cdk4 complex with the cyclin H-MO15 holoenzyme only could convert the inactive form into the active form in vitro (Fig. 2d). In agreement, [32P]orthophosphate labeling of the contact-inhibited cells showed an undetectable level of threonine phosphorylation on Cdk4 (negative data not shown). These results suggest that the Cdk4 activation was in part suppressed in the contact-inhibited cells. Since it might be possible that Cdk4-CAK is reduced or inactivated in the arrested cells, the CAK activity in contact-inhibited and growing cells was measured.
CAK Activity in 3Y1When MO15 was precipitated from lysates of growing cells or from those arrested by contact inhibition, no significant differences in its Cdk2 kinase activities were detected (Fig. 3a). Moreover, immunoprecipitated MO15 from contact-inhibited cells as well as from growing cells could activate bacterially produced Cdk2 and Cdk4 (Fig. 3, b and c), indicating that CAK was active when precipitated from contact-inhibited cells, whereas when bacterially produced His-tagged cyclin D2 plus GST-Cdk4 proteins were directly added to lysates, recovered on glutathione-Sepharose, and then tested for pRb kinase activity, the CAK-mediated activation was significantly lower in the contact-inhibited cells than in the growing cells (Fig. 3d). Although the CAK activity was equally abundant in extracts of both contact-inhibited and growing cells, in the cell lysates prepared from contact-inhibited cells, the activation of bacterially produced Cdk4 was inhibited. These results suggest that MO15 could not access Cdk4 in contact-inhibited cells.
The inhibitory activity of the Cdk4 activation was gradually diluted out by the addition of an increased amount of bacterially produced D-type cyclin and Cdk4. However, the inhibitory activity in the contact-inhibited cells was more than that in the growing cells (Fig. 3d). The inhibitory activity was eliminated by adding an excess amount of the bacterially produced D-type cyclin and Cdk4, indicating that the inhibitory factor, which interrupted the CAK-mediated Cdk4 activation, directly interacted with the cyclin D-Cdk4 complex.
p27kip1 and Cyclin D1-dependent Kinase Activity in Contact-inhibited Cells and Cycling CellsCKIs have been reported to associate with cyclin D-Cdk4 complex directly and to inhibit its kinase activity (37, 38). One of the CKIs, p27kip1, has been known to increase in contact-inhibited cells (30) and to be regulated posttranscriptionally (35, 39) or posttranslationally (35). In fact, the p27kip1 mRNA level did not fluctuate in either the contact-inhibited or growing 3Y1 cells (data not shown). By contrast, immunoblotting with an antibody to p27kip1 (see "Experimental Procedures" for its specificity) showed that 10-fold more p27kip1 protein was present in the contact-inhibited cells than in the growing cells (Fig. 4a). To examine whether p27kip1 was involved in the inactivation of Cdk4 in contact-inhibited cells, p27kip1-deprived or nondeprived cell lysates were subjected to inhibition of the activation of bacterially produced Cdk4. Although the cell lysates were precleared with preimmune serum-coated protein A-Sepharose beads, somehow, a slight Cdk4 activation was observed, and the p27kip1 depletion allowed for an effective activation of Cdk4 in the contact-inhibited cell lysates (Fig. 4b). These findings indicate that p27kip1 was in part involved in the regulation of Cdk4 activation in which the MO15 cannot access Cdk4 in contact-inhibited cells.
When contact-inhibited 3Y1 cells were trypsinized and subsequently
diluted 7-10-fold in the presence of a mitotic dose of serum (10%
FBS), the cultured cells synchronously entered S phase 18-22 h later
(Fig. 5c). Throughout the cell cycle, the
protein levels of cyclin D1, Cdk4, and cyclin D1-bound Cdk4 did not
fluctuate dramatically (Fig. 5a). As cells progressed toward
the S phase, p27kip1 decreased and cyclin
D-dependent kinase activity appeared. The timing of the
activation of cyclin D-dependent kinase correlated well
with the reduction of p27kip1, suggesting that p27kip1
also regulated cyclin D-dependent kinase activity in the
cycling cells (Fig. 5, a and b).
Mammalian cells can enter the cell cycle in response to growth factor-induced signals. Although macrophages transiently deprived of colony-stimulating factor 1 halt in the early G1 phase and fibroblasts transiently deprived of serum halt in quiescence, both cells reenter the cell cycle upon restimulation by growth factors. When the growth of either cell types is arrested by deprivation of the growth factor, the expression of D-type cyclins is undetectable. However, fibroblasts constitutively expressing both D-type cyclins and Cdk4 undergo growth arrest by serum starvation. In these cells, overproduced D-type cyclins and Cdk4 are detected even when the overexpessing cells are quiescent; however, the D-type cyclins and Cdk4 do not interact with each other (3). In contact-inhibited cells, in the presence of a mitotic dose of serum, the amounts of D-type cyclins and Cdk4 were identical with those of growing cells. Moreover, D-type cyclins bound Cdk4 despite the quiescence of the cells. Together, serum stimulation induced not only synthesis of D-type cyclins but also an assembly factor for D-type cyclins and Cdk4 even when the cells were in a quiescent state, whereas complexes composed of D-type cyclins and Cdk4 were enzymatically inactive in contact-inhibited cells. We reasoned that in the contact-inhibited cells Thr-172 of the Cdk4 was unphosphorylated, because phosphorylation is necessary for Cdk4 activation mediated by CAK (6).
CAK activity is detected in the growth factor-deprived macrophages and maintained throughout the cell cycle (12). Therefore, we examined whether CAK activity was detected in the contact-inhibited cells. Although the activity was not detected in the contact-inhibited cell lysates, the immunoprecipitated CAK was active, and its kinetics measured as Cdk2 kinase activity did not show significant differences between contact-inhibited cells and growing cells. The addition of increasing amounts of bacterially produced cyclin D and Cdk4 diluted out the CAK inhibition, suggesting that an inhibitor that could associate with cyclin D-Cdk4 complexes was present.
Our prediction is as follows. An inhibitor associates with cyclin D-Cdk4 complexes and prevents their kinase activity. Once the amounts of cyclin D-Cdk4 complexes override an inhibitory activity, their kinase activity first makes an appearance. A candidate for an inhibitor that can block the activation of Cdk4 as well as the activity of cyclin D-Cdk4 complexes is p27kip1, because p27kip1 has been known to have such dual inhibitory effects on cyclin-Cdk complexes (23, 30, 33).
In our experiments, a 10-fold greater amount of the p27kip1 protein was detected in contact-inhibited 3Y1 cells than in growing cells, whereas kip1 mRNA levels did not fluctuate. These findings suggest that the regulation of the p27kip1 accumulation was attributable to posttranscriptional control in the contact-inhibited cells. In quiescent cells, it has been reported that ubiquitin-regulated degradation (35) and translational control contribute to the p27kip1 level (39). We examined the mechanism by which the p27kip1 accumulation is regulated in the contact-inhibited cells. The half-life of the p27kip1 was much longer in the contact-inhibited cells (>4 h) than in the growing cells (approximately 2.5 h), although [35S]methionine incorporation into p27kip1 is identical,2 indicating that the degradation is predominantly responsible for the regulation of the p27kip1 accumulation in the contact-inhibited cells.
An increased amount of p27kip1 in contact-inhibited cells blocked the access of CAK to unphosphorylated cyclin D-bound Cdk4. The same manner of p27kip1-induced Cdk4 inactivation has been also observed in cAMP treated macrophages. The macrophage cell line, Bac1.2F5, requires colony-stimulating factor 1 to proliferate and is sensitive to cAMP-induced growth arrest. Cdk4 recovered from Bac1.2F5 arrested by cAMP treatment lacks threonine phosphate (33). These results suggest that inhibition of Cdk4 phosphorylation by p27kip1 may play a crucial role in maintaining cell cycle arrest. As cells were released from the contact inhibition and progressed toward the S phase, the amounts of D-type cyclins, Cdk4, and their complexes did not fluctuate dramatically. p27kip1 decreased gradually, and then cyclin D-dependent kinase activity appeared. Timing of the activation of the cyclin D-dependent kinase correlated well with the reduction of p27kip1.
An increased amount of p27kip1 is expressed in various cell lines arrested in quiescence by growth factor deprivation, such as colony-stimulating factor 1 in macrophages (33), interleukin-2 in T-cells (34), and serum in fibroblasts (35, 36). As these cells are released from quiescence by growth factor stimulation and cell cycle progression is resumed, the amount of p27kip1 gradually decreases through G1 phase. This reduction of the p27kip1 level supports the view that p27kip1 also regulates the kinase activity of cyclin D-Cdk4 complexes in cycling cells released from quiescence. D-type cyclins and Cdk4 are key regulators for G1 progression, and their kinase activities increase as cells approach the G1/S boundary.
In our experiments, p27kip1 plays an important role in the activation of Cdk4 by CAK during contact inhibition and in the regulation of cyclin D-dependent kinase activity in the G1 phase of cycling cells released from contact inhibition. Although p27kip1 can regulate a broad range of cyclin-Cdk complexes in vitro (23, 24), inhibition of Cdk4 may be one of the crucial roles for p27kip1 in cell cycle control as a Cdk inhibitor in contact-inhibited cells and in released cycling cells.
We are grateful to Dr. C. J. Sherr (St. Jude Children's Research Hospital, Memphis, TN) for monoclonal antibodies to cyclin D1 and D3 and antiserum to Cdk4 C terminus and to Dr. M. Matsuoka (Institute of Medical Science, University of Tokyo, Tokyo, Japan) for His-tagged cyclin D2. We thank M. Y. Nishikawa for critical reading of the manuscript.