Cyclin D


Table of contents

Interaction with Retinoblastoma pocket proteins to enhance cell cycle progression

Complexes of the cyclin-dependent kinase, cdk4, and each of three different D-type cyclins phosphorylate the Retinoblastoma protein. In intact cells, Cyclins D2 and D3 (but not D1) bind Rb. Introduction of cdk4, together with Rb and D-type cyclins, induce Rb hyperphosphorylation and dissociation of Cyclins D2 and D3. The transcription factor E2F-1 also binds Rb, and coexpression of Cyclin D-cdk4 complex triggers Rb phosphorylation and prevents its interaction with E2F-1. Thus D-type cyclins play a dual role as cdk4 regulatory subunits and as adaptor proteins that physically target active enzyme complexes to particular substrates (Kato, 1993).

The growth suppression function of RB is dependent on its protein binding activity. RB contains at least three distinct protein binding functions: (1) the A/B pocket, which binds proteins with the LXCXE motif; (2) the C pocket, which binds the c-Abl tyrosine kinase; and (3) the large A/B pocket, which binds the E2F family of transcription factors. Phosphorylation of RB, which is catalyzed by cyclin-dependent protein kinases, inhibits all three protein binding activities. LXCXE binding is inactivated by the phosphorylation of two threonines (Thr821 and Thr826), while the C pocket is inhibited by the phosphorylation of two serines (Ser807 and Ser811). The E2F binding activity of RB is inhibited by two sets of phosphorylation sites acting through distinct mechanisms. Phosphorylation at several of the seven C-terminal sites can inhibit E2F binding. Phosphorylation of two serine sites in the insert domain can inhibit E2F binding, but this inhibition requires the presence of the RB N-terminal region. RB mutant proteins lacking all seven C-terminal sites and two insert domain serines can block Rat-1 cells in G1. These RB mutants can bind LXCXE proteins, c-Abl, and E2F even after they become phosphorylated at the remaining nonmutated sites. Thus, multiple phosphorylation sites regulate the protein binding activities of RB through different mechanisms, and a constitutive growth suppressor can be generated through the combined mutation of the relevant phosphorylation sites in RB (Knudsen, 1997).

In cycling cells, the retinoblastoma protein (pRb) is un- and/or hypo-phosphorylated in early G1 and becomes hyper-phosphorylated in late G1. The role of hypo-phosphorylation and identity of the relevant kinase(s) remains unknown. Hypo-phosphorylated pRb associates with E2F in vivo and is therefore active. Increasing the intracellular concentration of the Cdk4/6 specific inhibitor p15(INK4b) by transforming growth factor beta treatment of keratinocytes results in G1 arrest and loss of hypo-phosphorylated pRb with an increase in unphosphorylated pRb. Conversely, p15(INK4b)-independent transforming growth factor beta-mediated G1 arrest of hepatocellular carcinoma cells results in loss of Cdk2 kinase activity with continued Cdk6 kinase activity and pRb remains only hypo-phosphorylated. Introduction of the Cdk4/6 inhibitor p16(INK4a) protein into cells by fusion to a protein transduction domain also prevents pRb hypo-phosphorylation with an increase in unphosphorylated pRb. It is concluded that cyclin D:Cdk4/6 complexes hypo-phosphorylate pRb in early G1 allowing continued E2F binding (Ezhevsky, 1997).

Cyclin D1 expression is positively regulated by Rb. At the functional level, antibody-mediated Cyclin D1 knockout experiments demonstrate that the Cyclin D1 protein, normally required for G1 progression, is dispensable for passage through the cell cycle in cell lines whose Rb is inactivated through complex formation with T antigen, E1A, or E7 oncoproteins as well as in cells which have suffered loss-of-function mutations of the Rb gene. The requirement for Cyclin D1 function is not regained upon experimental elevation of Cyclin D1 expression in cells with mutant Rb, while reintroduction of wild-type Rb into Rb-deficient cells leads to restoration of the Cyclin D1 checkpoint. These results strongly suggest that Rb serves as a major target of Cyclin D1, while the Cyclin D1 cell cycle regulatory function becomes dispensable in cells lacking functional Rb. An autoregulatory feedback loop mechanism has been suggested, one that regulates both the expression of the Cyclin D1 gene and the activity of Rb, thereby contributing to a G1 phase checkpoint control in cycling mammalian cells (Lukas, 1994).

The E2F1 gene is subject to autoregulatory control during progression from G0 to S phase, primarily reflecting a negative control in G0 and early G1, when the majority of E2F exists as a complex with Rb family members. Deregulation of G1 cyclins in quiescent cells stimulates the E2F1 promoter and this is augmented by coexpression of cyclin-dependent kinases in an E2F-dependent manner. E2F1 gene is a downstream target for G1 cyclin-dependent kinase activity, most likely as a consequence of phosphorylation of Rb family members (Johnson, 1994). The presence of an E2F DNA-binding complex containing the Rb-related p130 protein (Rb2) correlates with E2F-1 gene expression. Overexpression of p130 inhibits transcription from the E2F-1 promoter. D-type cyclin-dependent kinase activity specifically activates the E2F-1 promoter by relieving E2F-mediated promoter (Johnson, 1995).

Ectopic expression of Cyclins D1 and E was previously shown to accelerate the G1/S-phase transition, indicating that both classes of G1 cyclins control an event(s) that is rate limiting for entry into S phase. In order to determine whether Cyclins D1 and E control the same or two different rate-limiting events, cell lines were created that express both cyclins in an inducible manner. Ectopic expression of both Cyclins E and D1 in the same cell has an additive effect on shortening of the G1 interval relative to expression of any single cyclin. While premature expression of either Cyclin alone advances the G1/S-phase transition to the same extent, premature expression of Cyclin D1 leads to immediate appearance of hyperphosphorylated pRb; premature expression of Cyclin E does not. Ectopic expression of both Cyclins E and D1 in the same cell has an additive effect on shortening of the G1 interval, while the effect on pRb phosphorylation is similar to the effect of Cyclin D1 alone. These results suggest that Cyclins E and D1 control two different events, both rate limiting for the G1/S-phase transition, and that pRb phosphorylation might be the rate-limiting event controlled by Cyclin D1 (Resnitzky, 1995).

Transcription of the human proto-oncogene MYC is repressed in quiescent or non-dividing cells. Upon mitogenic stimulation expression of MYC is rapidly and transiently induced, maintained throughout G1, and declines to a basal level throughout further cell cycle transitions. Regulation of MYC promoter activity critically depends on the presence of a binding site for transcription factor E2F. Transcription from the MYC P2 promoter is induced efficiently by E2F-1, but repressed by Rb. Furthermore, overexpression of Cyclin A strongly activates the MYC promoter. This effect is further enhanced by coexpression of E2F-1 and Cyclin A. Expression of G1-phase Cyclin D1 leads to an E2F binding site-dependent trans-activation of the MYC promoter and this activation can be abrogated by overexpression of Rb. The interaction of D-type G1 Cyclins with Rb resembles that of the adenovirus E1A protein with Rb: either can disrupt inhibitory E2F-Rb complexes. Intervention of distinct cyclins and their respective associated kinases appear to promote transcriptional activation of MYC throughout the cell cycle either by conversion of E2F within multimeric complexes into an active transcription factor or by liberation of free functional E2F (Oswald, 1994).

Cyclin E is the major rate-limiting target of Cyclin D

Two classes of cyclins are expressed in mammalian cells during the G1 phase of the cell cycle: D-type cyclins (cyclins D1, D2, and D3) and cyclin E. Although these proteins are collectively referred to as G1 cyclins, there are profound differences between the two G1 cyclin classes. While the three D-type cyclins share very similar amino acid sequences (over 70% identity), they display only weak sequence relatedness with cyclin E (38% within the most conserved 'cyclin box' domain). In fact, cyclin E shares more similarity with the S phase- and M phase-specific cyclins A, B1, and B2 than with the D-type cyclins. The expression of these two classes of G1 cyclins occurs at different points of cell cycle progression and is governed by distinct mechanisms. Thus, the expression of D-type cyclins is controlled largely and perhaps entirely by extracellular signals. Cyclin D1 in particular is rapidly induced following mitogen challenge; its levels rapidly decline when mitogens are withdrawn or when antimitogens are added. For this reason, D-cyclins are regarded as functional links between the extracellular environment and the cell cycle machinery. In contrast to the regulation of the D-type cyclins, the expression of cyclin E is controlled by an autonomous mechanism and peaks suddenly at the G1/S boundary. Once induced, D-cyclins bind and activate CDK4 and CDK6, while cyclin E associates with an entirely different catalytic partner, CDK2. Moreover, the cyclin D- and cyclin E-associated CDK complexes are activated at different times during cell cycle progression, cyclin D-CDK4/6 complexes being active during the mid-G1 phase, whereas active cyclin E-CDK2 complexes are found at the G1/S boundary. The activity of cyclin D-CDK4/6 complexes is negatively controlled by association with inhibitory molecules belonging to the INK4 family of CDK inhibitors. In contrast, cyclin E-CDK2 complexes appear to be immune to inhibition by members of this family of proteins. A mouse strain has been generated in which the coding sequences of the cyclin D1 gene (Ccnd1) have been deleted and replaced by those of human cyclin E (CCNE). In the tissues and cells of these mice, the expression pattern of human cyclin E faithfully reproduces that normally associated with mouse cyclin D1. The replacement of cyclin D1 with cyclin E rescues all phenotypic manifestations of cyclin D1 deficiency and restores normal development in cyclin D1-dependent tissues. Thus, cyclin E can functionally replace cyclin D1. These analyses suggest that cyclin E is the major downstream target of cyclin D1 (Geng, 1999).

What then is the molecular basis of the observed rescue of cyclin D1 functions by cyclin E? Two possibilities are considered. (1) Knockin cyclin E-CDK2 complexes replace the functions normally executed by cyclin D1-CDK4/6 complexes. (2) The cyclin ED1 allele results in ectopic expression of the major downstream target of cyclin D1, thereby obviating the need for cyclin D1 in cell cycle progression, revealing that cyclin E is the major rate-limiting target for cyclin D1. The evidence presented in this paper strongly favors the latter scenario. Thus, cyclin D1 is believed to play at least two functions in cell cycle progression: (1) it drives the phosphorylation of the pRB; (indeed, cells that have lost pRB no longer require cyclin D1 for their growth, indicating that pRB is the major downstream target of cyclin D1) and (2) cyclin D1 induction causes the redistribution of a cellular kinase inhibitor, p27Kip1, from the cyclin E-CDK2 pool to the cyclin D-CDK4/6 pool, thereby liberating cyclin E-CDK2 complexes from inhibition and resulting in their activation. Both functions of cyclin D1 were examined in the tissues of cyclin ED1 mice. It was found that neither of these cyclin D1-dependent functions is replaced by the ectopically expressed human cyclin E. Consequently, it is concluded that the ectopic cyclin ED1 allele rescues the proliferation of cyclin D1-dependent tissues by bypassing the functions of cyclin D1 rather than by replacing them. This indicates, in turn, that cyclin E activity is the major downstream rate-limiting target for cyclin D1 and that ectopic cyclin E expression, as achieved through the cyclin ED1 allele, bypasses the need for cyclin D1 in cell cycle progression (Geng, 1999 and references).

Currently available information on the mammalian cell cycle enables one to postulate how this functional replacement might occur. pRB is the major target for cyclin D-CDK4/6 complexes. When hypophosphorylated, pRB binds and sequesters several proteins, most notably transcription factors of the E2F family. The phosphorylation of pRB leads to the release of these captive proteins, which then proceed to transactivate target genes. Alternatively, the dissociation of pRB from E2Fs removes a transcription-repressing factor from E2Fs bound to control elements of certain promoters. One of the genes regulated by the E2Fs is cyclin E, leading to the hypothesis that cyclin D1 serves to control the activity of cyclin E via pRB and E2Fs. An elaboration of this model predicts that cyclin D1-CDK4/6 complexes bring about the initial, preparatory pRB phosphorylation, leading to the induction of small amounts of cyclin E. The end consequence of these steps is the activation of cyclin E, whose ectopic expression, as described here, appears to obviate the function of cyclin D1 (Geng, 1999 and references).

Apart from its role as a partner of CDK4 and CDK6 kinases, cyclin D1 has been reported to play a unique role as an Estrogen receptor (ER) coactivator. This might suggest that the observed defect in the mammary development of cyclin D1-/- mice derives from inadequate activation of the ER in this tissue. The evidence presented in this paper argues against this explanation: (1) this paper shows that cyclin E, which was reported not to bind the ER and not to serve as ER coactivator, can replace cyclin D1 in driving normal mammary epithelial development, and (2) this paper shows that the induction of a major mammary epithelial ER-responsive gene, the PR, proceeds normally in cyclin D1-deficient mice. These findings have led to the conclusion that the crucial role for cyclin D1 in mammary development reflects the activity of cyclin D1 toward other targets, such as pRB (Geng, 1999 and references).

The studies described in this paper strongly suggest that cyclin E is the major rate-limiting target for cyclin D1 in mammalian cells. This, together with the well-established fact that the D-type cyclins are controlled by extracellular signals, while E cyclins are controlled internally, might suggest that the cyclin D-pRB pathway has evolved in order to enable metazoan cells to connect extracellular signals with cyclin E activation, thereby controlling the entrance into S phase. Since this cyclin D-pRB pathway is subverted in essentially all human cancers, the elucidation of its role in driving normal cell proliferation is essential to understanding the molecular basis of malignant cell growth (Geng, 1999).

Other Cyclin D Interactions

Cyclin-dependent kinase-5 (cdk-5) is a serine/threonine kinase that displays neurone-specific activity. Experimental manipulation of cdk-5 expression in neurons has shown that cdk-5 is essential for proper development of the nervous system and, in particular, for outgrowth of neurites. Such observations suggest that cdk-5 activity must be tightly controlled during development of the nervous system. To identify possible regulators of cdk-5, the yeast two-hybrid system was used to search for proteins that interact with cdk-5. In two independent yeast transformation events, cyclin D2 interacted with cdk-5. Immunoprecipitation experiments confirm that cyclin D2 and cdk-5 interact in mammalian cells. Cyclin D2 did not activate cdk-5 as assayed using three different substrates, which was in contrast to a known cdk-5 activator, p35. However, cyclin D2 expression leads to a decrease in cdk-5/p35 activity in transfected cells. As cyclin D2 and cdk-5 are known to share overlapping patterns of expression during development of the CNS, the results presented here suggest a role for cyclin D2 in modulating cdk-5 activity in postmitotic developing neurons (Guidato, 1998).

The activities of cyclin D-dependent kinases serve to integrate extracellular signaling during G1 phase with the cell-cycle engine that regulates DNA replication and mitosis. Induction of D-type cyclins and their assembly into holoenzyme complexes depends on mitogen stimulation. Conversely, the fact that D-type cyclins are labile proteins guarantees that the subunit pool shrinks rapidly when cells are deprived of mitogens. Phosphorylation of cyclin D1 on a single threonine residue near the carboxyl terminus (Thr-286) positively regulates proteasomal degradation of D1. Glycogen synthase kinase-3beta (GSK-3beta) phosphorylates cyclin D1 specifically on Thr-286, thereby triggering rapid cyclin D1 turnover. Because the activity of GSK-3beta can be inhibited by signaling through a pathway that sequentially involves first Ras, then phosphatidylinositol-3-OH kinase (PI3K), and finally protein kinase B (Akt), the turnover of cyclin D1, like its assembly, is also Ras dependent and, hence, mitogen regulated. In contrast, Ras mutants defective in PI3K signaling, or constitutively active mitogen-activated protein kinase-kinase (MEK1) mutants that act downstream of Ras to activate extracellular signal-regulated protein kinases (ERKs), cannot stabilize cyclin D1. In direct contrast to cyclin D1, which accumulates in the nucleus during G1 phase and exits into the cytoplasm during S phase, GSK-3beta is predominantly cytoplasmic during G1 phase, but a significant fraction enters the nucleus during S phase. A highly stable D1 mutant (in which an alanine is substituted for the threonine at position 286 and is refractory to phosphorylation by GSK-3beta) remains in the nucleus throughout the cell cycle. Overexpression of an active, but not a kinase-defective, form of GSK-3beta in mouse fibroblasts causes a redistribution of cyclin D1 from the cell nucleus to the cytoplasm. Therefore, phosphorylation and proteolytic turnover of cyclin D1 and its subcellular localization during the cell division cycle are linked through the action of GSK-3beta (Diehl, 1998).

Overexpression of cyclin D1 is a common event in various forms of cancer. D1 can be overexpressed as a result of gene amplification or because it is targeted through chromosomal translocations. However, in certain tumors, high levels of cyclin D1 expression have not been explained by such mechanisms, and events affecting cyclin D1 turnover might play some role. Although the p16INK4a-cyclin D1-Rb pathway is disabled in many forms of human cancer, colon carcinomas provide a conspicuous exception. Inactivation of the adenomatous polyposis coli (APC) tumor suppressor is the single most common event in colon cancer. APC is a target of Wnt signaling, and it regulates the proteolytic turnover of beta-catenin in a manner that depends on phosphorylation of beta-catenin by GSK-3. beta-Catenin mutants that have lost GSK-3 phosphorylation sites remain constitutively active as coactivators of TCF/LEF-dependent transcription, and such mutations have now been found in the major fraction of colon cancers that lack mutated APC alleles. The fact that GSK-3 can also regulate cyclin D1 turnover suggests that deregulation of Wnt signaling in colon cancer may target cyclin D1 in addition to the APC-beta-catenin complex (Diehl, 1998 and references).

Sp1-mediated transcription is stimulated by Rb and repressed by cyclin D1. The stimulation of Sp1 transcriptional activity by Rb is conferred, in part, through a direct interaction with the TBP-associated factor TAF(II)250. This study investigated the mechanism(s) through which cyclin D1 represses Sp1. The ability of cyclin D1 to regulate transcription mediated by Gal4-Sp1 fusion proteins, which contain the Gal4 DNA-binding domain and Sp1 trans-activation domain(s), was examined. The domain of Sp1 sufficient to confer repression by cyclin D1 was mapped to a region important for interaction with TAF(II)110. TAF(II)250-cyclin D1 complexes can be immunoprecipitated from mammalian and baculovirus-infected insect cells and, recombinant GST-TAF(II)250 (amino acids 1-434) associates with cyclin D1 in vitro. Moreover, the overexpression of Rb or CDK4 reduces the level of TAF(II)250-cyclin D1 complex. The amino terminus of cyclin D1 (amino acids 1-100) is sufficient for association with TAF(II)250 and for repressing Sp1-mediated transcription. Taken together, the results suggest that cyclin D1 may regulate transcription by interacting directly or indirectly with TAF(II)250 (Adnane, 1999).

Cyclin D1 is overexpressed in a significant percentage of human breast cancers, particularly in those that also express the estrogen receptor (ER). Experimentally overexpressed cyclin D1 can associate with the ER and stimulate its transcriptional functions in the absence of estrogen. This effect is separable from the established function of cyclin D1 as a regulator of cyclin-dependent kinases. Cyclin D1 can also interact with the histone acetyltransferase p300/CREB-binding protein-associated protein (P/CAF), thereby facilitating an association between P/CAF and the ER. Ectopic expression of P/CAF potentiates cyclin D1-stimulated ER activity in a dose-dependent manner. This effect is largely dependent on the acetyltransferase activity of P/CAF. These results suggest that cyclin D1 may trigger the activation of the ER through the recruitment of P/CAF, by providing histone acetyltransferase activity and, potentially, links to additional P/CAF-associated transcriptional coactivators (McMahon, 1999).

Table of contents

Cyclin D: Biological Overview | Regulation | Developmental Biology | References

Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.