Protein Kinase C α Signaling Inhibits Cyclin D1 Translation in Intestinal Epithelial Cells*

Cyclin D1 is a key regulator of cell proliferation, acting as a mitogen sensor and linking extracellular signaling to the cell cycle machinery. Strict control of cyclin D1 levels is critical for maintenance of tissue homeostasis. We have reported previously that protein kinase C α (PKCα), a negative regulator of cell growth in the intestinal epithelium, promotes rapid down-regulation of cyclin D1 (Frey, M. R., Clark, J. A., Leontieva, O., Uronis, J. M., Black, A. R., and Black, J. D. (2000) J. Cell Biol. 151, 763–778). The current study explores the mechanisms underlying PKCα-induced loss of cyclin D1 protein in non-transformed intestinal epithelial cells. Our findings exclude several mechanisms previously implicated in down-regulation of cyclin D1 during cell cycle exit/differentiation, including alterations in cyclin D1 mRNA expression and protein turnover. Instead, we identify PKCα as a novel repressor of cyclin D1 translation, acting at the level of cap-dependent initiation. Inhibition of cyclin D1 translation initiation is mediated by PKCα-induced hypophosphorylation/activation of the translational suppressor 4E-BP1, association of 4E-BP1 with the mRNA cap-binding protein eIF4E, and sequestration of cyclin D1 mRNA in 4E-BP1-associated complexes. Together, these post-transcriptional effects ensure rapid disappearance of the potent mitogenic molecule cyclin D1 during PKCα-induced cell cycle withdrawal in the intestinal epithelium.

Tight control of cell proliferation is essential for maintenance of normal homeostasis in self-renewing tissues such as the intestinal epithelium. Previous studies have identified protein kinase C (PKC) 4 signaling as an important negative regulator of cell growth/cell cycle progression in intestinal epithelial cells (1)(2)(3). Several members of the PKC family are predominantly expressed/activated in non-proliferating and terminally differentiated intestinal cells (4 -6), pointing to a major function in regulation of post-mitotic events in this tissue. Consistent with this role, we have demonstrated that PKC␣ signaling triggers a program of cell cycle withdrawal in non-transformed intestinal crypt cells, paralleling the membrane translocation/activation of this enzyme precisely at the point of growth arrest within intestinal crypts in situ (3,4,7,8). Although the extracellular signals that trigger growth arrest in the intestine in situ remain poorly characterized, a possible physiological activator of PKC␣ in this system is transforming growth factor ␤. This potent growth inhibitory factor is known to promote G 0 /G 1 arrest in intestinal epithelial cells (9 -11) and to activate PKC␣ signaling in other systems (12).
One of the earliest events following PKC␣ activation in intestinal epithelial cells is down-regulation of cyclin D1 (3,7,8), indicating that this molecule is an important target of PKC␣ control. Cyclin D1 is a key regulator of cell proliferation, acting as a mitogen sensor and linking extracellular signaling to the cell cycle machinery (13). Progression through early G 1 involves the activity of holoenzymes consisting of cyclin D (D1, D2, or D3 depending on the cell type) in association with cdk4 or cdk6. Cyclin D-dependent kinases promote cell cycle progression by initiating phosphorylation/inactivation of the retinoblastoma growth suppressor protein in mid G 1 , a process that is completed later in G 1 by cyclin E/cdk2. Cyclin D/cdk complexes also have an important non-catalytic function that involves sequestration of cdk-inhibitory proteins of the Cip/Kip family, thus relieving repression of cyclin E-and cyclin A-cdk2 activity.
Precise regulation of cyclin D1 accumulation is of critical importance. Increased expression of the molecule shortens the G 1 interval in many cell types and can reduce/overcome dependence on physiological growth stimuli (14). Decreased levels of the protein, on the other hand, lengthen G 1 phase and reduce proliferation (14,15). Thus, cyclin D1 expression is subject to strict control at multiple levels, including transcription, message stability and nucleocytoplasmic transport, protein synthesis, and protein turnover (14, 16 -21). Notably, aberrant overexpression of cyclin D1 is a key component of tumor development in various tissues, including the intestine (14,22,23), and both transcriptional and post-transcriptional mechanisms have been implicated in deregulation of cyclin D1 expression in tumors (14).
In this study, we have explored the mechanisms underlying PKC␣induced down-regulation of cyclin D1 in IEC-18 non-transformed intestinal crypt cells. Our analysis of cross-talk between PKC signaling and cyclin D1 control unveiled a novel function for PKC␣ as a negative regulator of cyclin D1 translation initiation. PKC␣ modulates the activity of key translational regulators, including eukaryotic translation initiation factor eIF4E and eIF4E-binding protein 1 (4E-BP1), to repress cyclin D1 protein synthesis during intestinal epithelial cell cycle withdrawal. Importantly, the engagement of translational rather than transcriptional mechanisms ensures a rapid effect (24), with disappearance of cyclin D1 protein preceding other hallmark events of cell cycle withdrawal, including induction of Cip/Kip cyclin-dependent kinase inhibitors and activation of the growth suppressor function of pocket proteins (3,7,8).
Northern Blot Analysis-Total cellular and cytoplasmic RNA were isolated using the RNeasy system (Qiagen) as recommended by the manufacturer. Nuclear RNA was similarly isolated from the nuclear pellet generated during the cytoplasmic RNA isolation procedure. Northern blot analysis of RNA samples (10 g) was performed using randomly primed, 32 P-labeled probe corresponding to mouse cyclin D1 cDNA, and specific hybridization was detected by phosphorimaging as we have described (25).
Analysis of Protein Stability-Cells were incubated in methioninefree Dulbecco's modified Eagle's medium containing 4 mM glutamine, 5 g/ml of insulin, and 5% dialyzed fetal bovine serum for 30 min and then labeled with 94 Ci of [ 35 S]Met/Cys Expre 35 S 35 S protein-labeling mixture (PerkinElmer Life Sciences) for 1 h, prior to the addition of PMA or vehicle (ethanol) for 1 h. [ 35 S]Met/Cys was chased for various times in label-free complete medium containing PMA or vehicle, and cells were harvested for cyclin D1 immunoprecipitation using sc-753 anti-cyclin D1 antibody. Immunoprecipitates were resolved on 20% SDS-PAGE gels, transferred to nitrocellulose membrane, and 35 S-labeled protein was detected and quantified using the Storm/ImageQuant Phospor-Imaging system (Amersham Biosciences). For analysis of cyclin D1 decay following protein synthesis inhibition, cells were treated with 50 g/ml of cycloheximide after 1 h of PMA/vehicle treatment and harvested for Western blot analysis at various times thereafter.
Biosynthetic Labeling-IEC-18 cells were incubated in methioninefree medium as described above and treated with PMA or ethanol for 30 min. Cells were then incubated in [ 35 S]Met/Cys and PMA/vehicle for the indicated times as described (19). Cyclin D1 was immunoprecipitated using sc-450 antibody, and immunoprecipitates were analyzed as above.
Plasmids and Transfection-FLAG-tagged wild-type and T286A mutant cyclin D1 constructs were gifts from Dr. C. Sherr (St. Jude Children's Research Hospital, Memphis, TN). Destruction box mutants (R29Q and L32A) were generated from the wild-type cyclin D1 construct using the QuikChange site-directed mutagenesis kit (Stratagene) and appropriate primers (sense strands: GGGTGCTGCAGGCCAT-GCTCAAG; GCGAGCCATGGCCAAGACGGAG). Cells in 60-mm plates were transfected with 5 g of each construct (0.5 g for the T286A mutant) using Lipofectamine 2000 (Invitrogen).
In Vitro Cap Affinity Assay-Cap affinity chromatography was performed essentially as described (27). PMA-or vehicle-treated IEC-18 cells were harvested in lysis buffer (50 mM Tris, pH 7.4, 100 mM NaCl, 1 mM EDTA, and 1% Triton X-100, supplemented with protease/phosphatase inhibitor cocktails), and particulate material was removed by centrifugation. Cell lysates (350 -400 g of protein) were incubated with 50 l of 7-methyl-GTP-Sepharose 4B slurry (Amersham Biosciences) for 16 h at 4°C. Beads were washed three times with lysis buffer, and cap-bound protein was eluted with Laemmli sample buffer and subjected to immunoblot analysis.
Immunoprecipitation-RT-PCR Assay-PMA-or vehicle-treated cells were harvested in detergent-free lysis buffer (50 mM Tris-HCl, pH 7.5, 100 mM NaCl, 0.1 mM EDTA, 1 mM MgCl 2 , and 10% glycerol, containing protease/ phosphatase inhibitors and 56 units/ml of RNase inhibitor (Sigma)) (28). Following removal of particulate material by centrifugation, lysates were supplemented with Tween 20 (0.5%), RNase inhibitor (53 units/ml), and dithiothreitol (0.1 mM), and 4E-BP1-containing complexes were immunoprecipitated using rabbit anti-4E-BP1 antibody (1.8 g) and Protein A Dynabeads (Dynal). Beads were then washed twice (5 min) with lysis buffer containing 0.5% Tween 20 and once with lysis buffer without Tween 20 and resuspended in 45 l of lysis buffer. 25 l were taken for RNA isolation as described (28), and 20 l were boiled in 2ϫ Laemmli sample buffer for immunoblot analysis. Levels of cyclin D1 mRNA were assessed by RT-PCR in the presence of 50 Ci of [␣ 32 P]CTP using the Titanium one-step RT-PCR kit (Clonetech) and specific primers (GTCTTCCCGCTGGCCATGAACTAC; AAGAAAGT-GCGTTGTGCGGTAGCA). PCR cycle number was empirically determined to be in the linear range of amplification. PCR products were resolved on 5% PAGE gels and visualized by phosphorimaging. Quantitative RT-PCR analysis of cyclin D1 mRNA levels in immunoprecipitates was performed using the Full Velocity SYBR Green QRT-PCR Master Mix (Stratagene) and a 7300 real-time PCR system (Applied Biosciences).

PKC/PKC␣ Signaling Promotes Down-regulation of Cyclin D1 in Intestinal Epithelial Cells
We have previously shown that activation of PKC signaling in IEC-18 non-transformed intestinal crypt cells results in cell cycle withdrawal (3,8). Maintenance of cell cycle arrest requires sustained activation of PKC␣, and the effect is reversed coincident with down-regulation of the enzyme. As shown in Fig. 1, PKC-induced cell cycle blockade is associated with marked changes in the levels of cyclin D1. Activation of PKC␣, -␦, and/or -(the only phorbol ester-responsive PKC isozymes expressed in IEC-18 cells) (3,8) by PMA results in rapid down-regulation of cyclin D1 protein, detectable by 30 min and maximal by ϳ2 h (Fig. 1A). Down-regulation of cyclin D1 is also seen in response to PKC activation by the phorbol ester PdBu, the macrocyclic lactone Bryo, and the diacylglycerol analog DiC 8 (Fig. 1B), excluding potential nonspecific effects of PMA treatment (29). PKC agonist-induced down-regulation of cyclin D1 is blocked by the general PKC inhibitor bisindolylmaleimide I and by PKC␣, -␦, and -depletion with 1 M PdBu (Fig. 1C), confirming that the effect is PKC dependent. Notably, the PKC␣-selective inhibitor Gö6976 also prevents the effect of PKC agonists, pointing to a key role for PKC␣ in regulating cyclin D1 levels in intestinal epithelial cells (Fig. 1D).

PKC/PKC␣-induced Down-regulation of Cyclin D1 Does Not Involve Alterations in mRNA Levels/Subcellular Distribution or Protein Stability
Effects of PKC Signaling on Cyclin D1 mRNA-To investigate the molecular mechanisms underlying PKC-induced loss of cyclin D1, the effects of PMA on cyclin D1 mRNA were examined by Northern blotting. As shown in Fig. 2, PMA did not affect total cyclin D1 mRNA levels at 1 h of treatment, a time when cyclin D1 protein was significantly down-regulated (Fig. 1A), although a slight decrease in mRNA levels was observed at later times (e.g. 2 h, data not shown). Analysis of nuclear and cytosolic mRNA further excluded the possibility that PKC activation affects cyclin D1 mRNA nucleocytoplasmic distribution (Fig. 2). Thus, PKC-induced reduction in cyclin D1 protein does not involve effects on the levels or distribution of cyclin D1 mRNA, at least during the initial stages of PMA treatment.
PKC/PKC␣ Does Not Alter Cyclin D1 Protein Turnover-Cell cycle exit (e.g. under conditions of serum starvation or cell differentiation) has been linked to accelerated turnover of cyclin D1 protein (20,30). Therefore, [ 35 S]methionine pulse-chase experiments were performed to determine whether PKC activation alters cyclin D1 stability in IEC-18 cells. As expected, PMA treatment resulted in lower recovery of 35 S in cyclin D1 immunoprecipitates, reflecting the reduced cyclin D1 protein levels at the time points examined (1-2.5 h of PMA treatment) (Fig. 3A). However, PKC activation had no effect on the rate of decay of cyclin D1 protein (t1 ⁄ 2 ϳ 30 min), a result that was confirmed by examining cyclin D1 disappearance following inhibition of protein synthesis with cycloheximide (data not shown).
Degradation of cyclin D1 occurs via ubiquitin/proteasome-mediated proteolysis, triggered by its phosphorylation on Thr-286 by GSK-3␤ (20). An N-terminal (amino acids 29 -32) destruction box (RXXL) has also been implicated in regulating proteasomal degradation of cyclin D1 (18). Consistent with the absence of alterations in cyclin D1 stability, PKC activation did not decrease the levels of inhibitory phosphorylation of GSK-3␤ on Ser-9 (Fig. 3B, upper panel), and inhibition of GSK-3␤ activity with Li ϩ did not block PMA-induced loss of cyclin D1 (although its accumulation following Li ϩ treatment confirmed a role for GSK-3␤ in normal turnover of cyclin D1 in IEC-18 cells) (middle panel ). These data, together with the fact that PKC activation promoted the downregulation of degradation-resistant cyclin D1 mutants, including the GSK-3␤ phosphosite mutant T286A (TA, lower panel ) and the destruction box mutants R29Q (RQ) and L32A (LA) (Fig. 3C), confirmed that PKC-induced loss of cyclin D1 protein does not involve acceleration of cyclin D1 turnover.

PKC/PKC␣ Signaling Inhibits Translation of Cyclin D1 in IEC-18 Cells
Activation of PKC/PKC␣ Inhibits Cyclin D1 Synthesis-[ 35 S]Methionine biosynthetic labeling was performed to determine whether PKC signaling inhibits translation of cyclin D1. A reduction in radiolabeled cyclin D1 was detected in PMA-treated cells relative to vehicle-treated cells at each time point examined (Fig. 4). These data, together with the  PKC/PKC␣ Activates 4E-BP1, Inhibits Cap-dependent Translation Initiation, and Leads to Sequestration of Cyclin D1 mRNA in 4E-BP1associated Complexes-To gain insight into the molecular basis for PKC-mediated inhibition of cyclin D1 synthesis, we examined the effects of PKC activation on key regulators of translation initiation, including eIF4E and eIF2␣. eIF4E binds the 5Ј-cap structure (m 7 GpppN, where N is any nucleotide) found in the majority of eukaryotic mRNAs and associates with eIF4G and eIF4A to form eIF4F, a translation initiation complex that mediates recruitment of ribosomes to mRNA (31). A major mechanism for control of eIF4E function is through its interaction with a family of translational inhibitory proteins, the eIF4E-binding proteins (4E-BPs), the best characterized of which is 4E-BP1 (31). The hypophosphorylated ␣ and ␤ forms of 4E-BP1 compete with eIF4G for binding to eIF4E, thereby preventing eIF4F assembly and inhibiting capdependent translation. In contrast, the hyperphosphorylated ␥ form is unable to bind eIF4E. Immunoblot analysis revealed that PMA treatment leads to rapid hypophosphorylation/activation of 4E-BP1 (Fig.  5A), evident by the appearance of the faster migrating ␣/␤ forms within 15 min. 4E-BP1 hypophosphorylation was also seen with other PKC agonists including PdBu, Bryo, and DiC 8 (Fig. 5B), and the PKC dependence of the effect was confirmed using PKC␣-, PKC␦-, and PKC-depleted cells (Fig. 5C). Use of cells depleted of PKC␦ and -, but not PKC␣, by PMA pulse treatment (7) demonstrated that PKC␣ can mediate PMA-induced hypophosphorylation of 4E-BP1 (Fig. 5D), and a requisite role for PKC␣ signaling was confirmed using the PKC␣-selective inhibitor Gö6976 (Fig. 5E).
To determine whether the PKC-induced accumulation of 4E-BP1 ␣/␤ phosphoforms reflected an increase in 4E-BP1 activity, the interaction of 4E-BP1 with eIF4E was examined using cap affinity chromatography. Extracts from IEC-18 cells treated with vehicle or PMA were incubated with Sepharose 4B-immobilized 7-methyl-GTP cap analog to capture eIF4E and its binding partners. Serum-stimulated and serumstarved cells were included as negative and positive controls for 4E-BP1/  It should be noted that methionine starvation leads to decreased levels of cyclin D1 protein in control cells, an effect that is associated with inhibition of translation (data not shown and Refs. 16 and 54). Thus, although data from metabolic labeling analysis clearly demonstrate that PKC activation leads to reduced synthesis of cyclin D1, they are likely to underestimate the true extent of this effect. Data are representative of three independent experiments. eIF4E interaction, respectively. Immunoblot analysis of cap analogbound proteins revealed a striking increase in levels of cap-bound 4E-BP1 ␣/␤ phosphoforms in PKC agonist-treated cells, comparable with those in serum-starved cells (Fig. 5F ). Consistent with inhibition of cap-dependent translation initiation, metabolic labeling studies revealed that PKC agonist treatment results in a reduction in overall protein synthesis in IEC-18 cells (supplemental Fig. S1).
To examine the relevance of changes in 4E-BP1 activity to cyclin D1 translation, an immunoprecipitation RT-PCR protocol was used to assess sequestration of cyclin D1 mRNA in 4E-BP1-associated complexes. 4E-BP1 was immunoprecipitated from lysates of IEC-18 cells treated with PMA or vehicle. Immunoblot analysis confirmed immunoprecipitation of 4E-BP1 phosphoforms and increased association of eIF4E with 4E-BP1 in PMA-treated and serum-starved cells (Fig. 5G,  upper panel). Semiquantitative RT-PCR analysis of 4E-BP1 immunoprecipitates revealed that PKC activation leads to accumulation of cyclin D1 mRNA in 4E-BP1-associated complexes, comparable with levels seen in serum-starved cells (middle panel ). The specificity of the interaction was confirmed by the absence of cyclin D1 mRNA in PKC␣ immunoprecipitates. Quantitative real-time RT-PCR (lower panel) determined that PKC activation leads to an ϳ3-fold increase in the association of 4E-BP1 with cyclin D1 mRNA. As seen in cap binding assays, this effect was comparable with that induced by serum starvation. Thus, the increased activity of 4E-BP1 resulting from PKC activation directly impacts cap-dependent translation of cyclin D1 mRNA in intestinal epithelial cells.
Translation initiation is also regulated by eIF2␣, a factor involved in recruitment of initiator Met-tRNA to the 40 S ribosomal subunit (32). The activity of eIF2␣ is negatively regulated by phosphorylation on Ser-51. Immunoblot analysis revealed that PMA treatment has no effect on eIF2␣ phosphorylation (Fig. 6), arguing against its involvement in the inhibitory effects of PKC signaling on cyclin D1 in IEC-18 cells.

Loss of PKC/PKC␣ Signaling Results in Recovery of the Translational Machinery and Restoration of Cyclin D1 Levels
Members of the PKC family undergo down-regulation following activation (33). In IEC-18 cells, PMA treatment results in progressive down-regulation of phorbol ester-responsive PKC isozymes, albeit with different kinetics (Fig. 7A). PKC␦ is down-regulated first, followed by PKC␣ and then PKC. Activation-induced depletion/ desensitization of PKC isozymes was associated with a gradual restoration of cyclin D1 protein expression, first noted at ϳ3 h of PMA treatment (Fig. 7B). Comparison of the timing of cyclin D1 recovery with the kinetics of PKC isozyme down-regulation demonstrated that the reappearance of cyclin D1 coincides more closely with loss of PKC␣ (see 3-h time point), a finding that further supports the role of PKC␣ as a key negative regulator of cyclin D1 accumulation in IEC-18 cells. As shown in Fig. 7, A and B, PKC␣ expression remains steady for ϳ2 h of PMA treatment, paralleling the loss of cyclin D1 protein, and levels of the enzyme are considerably reduced by 3 h, coincident with restoration of cyclin D1 levels. In contrast, PKC␦ is markedly down-regulated by 1 h of PMA addition, and PKC expression is maintained for longer than 4 h in these cells. Loss of PKC/ PKC␣ signaling was also accompanied by reversal of PMA-induced hypophosphorylation/activation of 4E-BP1 (Fig. 7C, upper panel) and decreased association of the protein with cap analog (lower panel). Thus, PKC down-regulation is accompanied by recovery of the translational machinery and restoration of cyclin D1 levels in IEC-18 cells.

DISCUSSION
Cyclin D1 levels are low in proliferating intestinal crypt cells and are markedly decreased in post-mitotic cells (3, 34 -36). This expression profile points to the existence of mechanisms for tight control of the accumulation of this key G 1 progression factor in normal intestinal epithelial cells. Here we have identified PKC␣, an important mediator of growth inhibitory signaling in a variety of biological systems (e.g. Refs. 1,37,38), as a negative regulator of cyclin D1 expression in the intestinal epithelium. The involvement of PKC␣ in mediating down-regulation of cyclin D1 was established by demonstrating that (a) PKC agonists promote loss of cyclin D1 in IEC-18 cells expressing PKC␣ as the only phorbol ester-responsive isozyme (3), (b) the PKC␣-selective inhibitor Gö6976 blocks PMA-induced cyclin D1 down-regulation (Fig. 1), and (c) recovery of cyclin D1 levels closely correlates with loss of PKC␣ expression/activity in this system (Fig. 7).
Down-regulation of cyclin D1 is an early consequence of PKC␣ activation in IEC-18 cells, evident by 15-30 min of PKC agonist treatment (Figs. 1, 7). Analysis of the molecular mechanisms underlying the effect excluded several pathways previously implicated in the disappearance of cyclin D1 during cell cycle arrest/cell differentiation. Northern blot analysis revealed no change in cyclin D1 mRNA levels or subcellular distribution during the initial phase of the response (Fig. 2), indicating that PKC-induced down-regulation of cyclin D1 does not involve altered availability of cyclin D1 mRNA (21,39). Pulse-chase analysis further excluded a role for alterations in protein turnover (20), a conclusion that was substantiated by the ability of PKC/PKC␣ to promote down-regulation of GSK-3␤ phosphorylation site (T286A) and destruction box (R29Q and L32A) cyclin D1 mutants in IEC-18 cells (Fig. 3).
Instead, our findings provide the first evidence for the ability of  PKC␣ signaling to repress cyclin D1 synthesis by targeting cap-dependent initiation, a key point of translational control (Fig. 5). Inhibition of cap-dependent initiation is accomplished by rapid, PKC␣-dependent hypophosphorylation/activation of the translational repressor 4E-BP1 and increased association of the ␣/␤ phosphoforms of the protein with eIF4E (Fig. 5), thus preventing assembly of the eIF4F complex that mediates recruitment of ribosomes to mRNA (31). Accumulation of cyclin D1 mRNA in 4E-BP1-associated complexes provides support for a direct effect on translation of cyclin D1, consistent with findings that active 4E-BP1 represses cyclin D1 synthesis in other systems (40). The marked effects of 4E-BP1 on cyclin D1 synthesis are also consistent with evidence that limited availability of eIF4F complexes selectively impairs translation of several strong growth promoters, including cyclin D1, whose mRNAs are weak for translation due to lengthy and highly structured 5Ј-untranslated regions (41).
A number of studies have reported effects of PKC signaling on translation initiation factors and modulators, including 4E-BP1 (42)(43)(44). However, in direct contrast to our findings, these studies have linked PKC signaling to enhanced translation initiation and phosphorylation/ inactivation of 4E-BP1. This discrepancy likely reflects different functions of individual PKC family members. For example, in 293T and glioblastoma cells, inactivation of 4E-BP1 was attributed to the novel PKCs, PKC␦ and PKC, respectively (42,44). The finding that calcium ionophore can modulate the effects of PMA on the translational machinery in lymphocytes (45) suggests that cross-talk between signaling pathways may also underlie some of this discrepancy.
Taken together, the data demonstrate that, during cell cycle withdrawal, PKC␣ promotes down-regulation of cyclin D1 protein by engaging translational rather than transcriptional mechanisms, thus ensuring a rapid effect (24,46,47). Indeed, loss of cyclin D1 precedes other PKCinduced cell cycle-related events such as induction of p21 WAF1/CIP1 and p27 KIP1 and down-regulation of cyclin A (3, 7), arguing that targeting cyclin D1 is a key mechanism by which PKC␣ signaling promotes cell cycle exit in these cells. Cyclin D1 down-regulation is also an early event during intestinal epithelial growth arrest in situ. Although cyclins D1, E, and A are all expressed in proliferating intestinal crypt cells, only cyclin D1 is down-regulated near the point of growth arrest in the crypt region; changes in the levels of cyclins E and A occur only after the cells have migrated onto the villus (3,34,36,48). Because PKC␣ undergoes membrane translocation/activation precisely at the point within intestinal crypts at which cells cease dividing (3,4,7,8), the enzyme is appropriately positioned to promote cyclin D1 loss in vivo.
The finding that PKC␣-induced cell cycle exit is associated with translational repression of cyclin D1 is consistent with data from studies with other G 1 arresting signals. For example, blockade of cyclin D1 synthesis is associated with transforming growth factor ␤-induced G 0 /G 1 arrest in IEC-6 and RIE non-transformed intestinal cells (11). Given that transforming growth factor ␤ can activate PKC␣ (12), it is tempting to speculate that effects of transforming growth factor ␤-signaling on cyclin D1 expression are mediated by PKC␣ in the intestinal epithelium. The relevance of translational regulation of cyclin D1 to intestinal homeostasis is highlighted by the fact that overexpression of eIF4E, together with associated up-regulation of cyclin D1 (21,49), has been implicated in the development of colon tumors (14,23,35,41,50,51). In this regard, it is noteworthy that PKC␣ is frequently down-regulated/inactivated in colon adenomas and adenocarcinomas as well as in other epithelial cancers (5,37,52,53). Thus, loss of PKC␣-mediated negative regulation of critical translational regulators and cell cycle control molecules is likely to play an important role in epithelial tumorigenesis.