THE P130 POCKET PROTEIN: KEEPING ORDER AT CELL CYCLE EXIT/RE-ENTRANCE TRANSITIONS

Xavier Mayol and Xavier Grana

Fels Institute for Cancer Research and Molecular Biology and Department of Biochemistry and Temple University School of Medicine. 3307 North Broad St., Philadelphia, PA19140

Received 12/5/97 Accepted 12/9/97

2. INTRODUCTION

The E2F family of transcription factors includes five related proteins, namely E2F-1, 2, 3, 4, and 5, that heterodimerize with 3 more distantly related proteins, DP-1, 2, and 3 (figure 1), which are required for the DNA-binding capacity of E2F (1, 2). Most of the E2F-responsive genes so far identified are required for the G1 transition to the S phase of the cell cycle, being transcriptionally activated at a period of the G1 phase coincident with passage through the restriction point. Some genes that have been demonstrated or proposed to be under E2F control encode for either cell cycle regulatory proteins such as cyclins E (3-5) and A (6, 7), the cell cycle kinase CDC2 (8, 9), the CDC25C phosphatase (10), the protooncogenes B-Myb (11, 12), c-myc (13, 14), and N-Myc (2), the pocket protein p107 (15), and E2F-1 (16-19) and E2F-2 (17, 20) themselves, and some enzymes involved in DNA metabolism such as thymidylate synthase (TS) (21), DHFR (22, 23), DNA polymerase alpha (24), TK (25), RRM2 (26) and HsOrc1 (27). E2F/DP heterodimers function as transcriptional activators of E2F-responsive promoters placed upstream of reporter genes, whereas pocket protein complex formation turns E2F either into transcriptional repressors of some of these promoters or prevents E2F transcriptional activation in other cases (figure 2). Association of pocket proteins with E2F is controlled, at least in part, by the temporal activity of cyclin/CDK holoenzymes during the cell cycle, so that the E2F-binding capacity of pocket proteins is abrogated by cyclin/CDK phosphorylation. Other mechanisms that contribute to regulation of E2F activity include modulation of E2F and pocket protein levels (reviewed by 28), cyclin/CDK-mediated phosphorylation of E2F/DP heterodimers (28, 29), and subcellular localization of E2F complexes (30-33). Moreover, while p130 and p107 seem to have binding capacity only for E2F-4 and E2F-5, pRB associates in vivo with either E2F-1, E2F-2, E2F-3 or E2F-4 (figure 1). Furthermore, another putative function of p130 and p107 is to act directly as CDK inhibitors, which provides an additional independent growth suppressor function for these two proteins (34-37) and suggests that novel feedback loop pathways participate in the control of certain cell cycle transitions.

Figure 1. Pocket proteins interact with E2F/DP heterodimers. E2F family members seem to be able to heterodimerize with any DP protein. p130 associates with both E2F-4/DP and E2F-5/DP heterodimers. p107 associates primarily with E2F-4, *and has been found associated with E2F-5 in transfection assays. pRB associates with E2F-1 to 4/DP heterodimers.

As detailed below, in this review, many cell cycle genes are transcriptionally silent in G0. Analysis of a number of promoters in such genes indicates that these genes are actively repressed through E2F-like sites. Since most of the E2F proteins in quiescent cells are complexed with p130- E2F-4 being the predominant E2F family member in these cells-, it is assumed that p130 is responsible for this transcriptional silencing (figure 2A). Progression through the G1 phase involves a shift of E2F to pRB- and p107-containing complexes mostly due to complete hyperphosphorylation and downregulation of p130, to expression of pRB-associated E2F members, and to new synthesis of p107. Then, a balance between hypo- and hyper-phosphorylated pRB and p107 is thought to tightly regulate, at least in part, E2F activity during the late G1 and S phases. This scenario is summarized in Figure 3, and more details are given in Section 3 of this review.

Figure 2. Distinct models of E2F-mediated regulation of E2F-responsive promoters. (A) E2F-mediated negative regulation of E2F-responsive promoters in G0 by binding of pocket protein/E2F/DP complexes. Repression is released in mid- to late G1 by cyclin CDK phosphorylation (see text). (B) E2F-mediated positive regulation of E2F-responsive promoters in mid-to-late G1 and S phases by binding of E2F/DP complexes with transactivation activity. These promoters should be silent in G0 due to the absence of E2F complexes with transactivation activity. (C) E2F-mediated negative and positive regulation of E2F-responsive promoters. This cartoon depicts a dual model that integrates positive and negative regulation of transcription through E2F sites. Genes such as the DHFR gene of Chinese hamster might be regulated in a similar manner in ovarian cells (39). This type of regulation can be envisioned through a single or multiple E2F sites.

Figure 3. Cell cycle entry: regulation of E2F-dependent transcription by pocket proteins. High levels of phosphorylated p130 forms 1 and 2 are present in quiescent cells (G0 phosphorylation is represented by P). E2F-4 is the major E2F species in quiescent cells and it is bound to p130. p130/E2F-4/DP complexes repress the transcription of some genes whose products are required for cell cycle progression (G0/G1-repressed genes: indicated by G0/G1-rep. genes). Although low levels of pRB/E2F complexes are detected in quiescent cells, these complexes seem to participate in the repression of some genes (see text). Cell cycle re-entry results in the activation of G1 cyclin/CDK holoenzymes, which phosphorylate p130 to form 3 and abrogate p130/E2F-4 interaction (cyclin/CDK catalyzed phosphorylations are represented by boxed P). In addition, p130 levels abruptly decrease (represented by a crossed out molecule). G0 phosphorylation sites in p130 may not be phosphorylated from mid G1 to the remainder of the cell cycle (represented by (P) ). Among the genes containing E2F sites which are repressed in G0, there is E2F-1, E2F-2 and several other genes, including the p107 gene itself (see text). De-repression of transcription by disruption of p130/E2F-4 complexes results in the expression of E2F-1 and E2F-2 and allows then for the positive regulation of transcription of other genes containing E2F sites (G1 and S-transactivated genes: indicated by G1/S transact. genes). The free E2F-4/DP complexes released from p130 may also transactivate genes in mid to late G1 and S phases. p107 accumulates abruptly in mid G1 and is also phosphorylated by G1 cyclin/CDK holoenzymes. However, low levels of hypophosphorylated p107 associate with free E2F-4/DP complexes to repress the transcription of some genes during late G1 and S phases (G1/S repressed genes: indicated by G1/S rep. genes). pRB is hyperphosphorylated in mid G1, and the remaining low levels of hypophosphorylated pRB associate with newly synthesized E2Fs 1 through 4 and the E2F-4 released from p130/E2F-4 complexes by cyclin/CDK-mediated phosphorylation of p130. pRB/E2F complexes might repress genes containing E2F sites during mid-to-late G1 and S phases. In addition, promoters with E2F complexes bound to E2F sites, for instance E2F/DP/pocket or E2F/DP, might be turned on/off respectively (red reversible arrows) by cyclin /CDK phosphorylation of diverse complex components (not shown in this figure).

The specificity of interaction between individual pocket proteins and E2F family members and the timing of formation of particular pocket protein/E2F complexes during the cell cycle suggests that individual E2F complexes have distinguishable effects on gene transcription. Moreover, transcriptional regulation of specific genes by individual pocket protein-E2F complexes is suggested by recent experiments employing mouse fibroblasts deprived of individual and/or combinations of pocket proteins (38). In this respect, DNA-sequence specificity may exist for some E2F members (39, 40), and the presence or absence of certain consensus sites in some E2F-regulated promoters correlates with the response to E2F-mediated activation or repression (41). Moreover, the E2F-target gene specificity also seems to be determined to a great extent by the availability of individual E2F complexes. In other words, many E2F-responsive promoters would be activated or repressed depending on the presence of certain E2F complexes in the nucleus. This is the case for genes that are transcriptionally repressed during G0 by p130/E2F complexes, as well as the case for some genes regulated in late G1 and S phases where a balance between hypo- and hyper-phosphorylated forms of pRB and p107 is kept. It is conceivable that the biological significance of the existence of such diverse pocket and E2F protein families is to contribute in a precise manner to transcriptional regulation under a multitude of growth conditions in different cell types. In this respect, p130 has proved to be an example of how certain sets of genes are regulated at transitions involving cell cycle exit to, or cell cycle entrance from, a quiescent state. The regulation of p130 activity mainly with respect to its interaction with E2F complexes and the functional implications of this pathway on the physiology of the cell cycle and of differentiation processes will be discussed in this review.