Retinoic Acid Inhibition of Cell Cycle Progression in MCF-7 Human Breast Cancer Cells

EXPERIMENTAL CELL RESEARCH ARTICLE NO.

234, 293–299 (1997)

EX973589

Retinoic Acid Inhibition of Cell Cycle Progression in MCF-7 Human Breast Cancer Cells Wei-Yong Zhu, Carol S. Jones, Andras Kiss, Karen Matsukuma, Sonal Amin, and Luigi M. De Luca1 Laboratory of Cellular Carcinogenesis and Tumor Promotion, National Cancer Institute, Bethesda, Maryland 20892-4255

Cell cycle analysis indicates that retinoic acid (RA) inhibition of MCF-7 cell growth occurs through induction of G1 arrest with a concomitant reduction in the proportion of cells in S and G2 / M phases. RA did not affect cyclins D1, A, and E and cyclin-dependent kinase 2 (CDK2) expression, but significantly reduced cyclin D3 and CDK4 expression after 24 h. RA also inhibited cyclin B1 and CDC2 expression, possibly responsible for the reduction of the proportion of cells in G2 / M and S phases. RA did not induce p16 and p27 expression, but obviously reduced p21 level in MCF-7 cells. The retinoid markedly reduced pRB protein level and abrogated pRB phosphorylation after 48 h; it also reduced transcription factor E2F1 expression at both the mRNA and protein levels. E2F1 promoter activity was reduced by 60%, which is probably responsible, at least in part, for the reduction of E2F1 expression in RAtreated MCF-7 cells. These observations demonstrate a marked effect of RA on some of the key cell cycle regulatory proteins in MCF-7 cells. Cyclin D3 and CDK4 are likely the early targets of RA, followed by reduced pRB expression and phosphorylation, as well as by the inhibition of the E2F1 transcription factor which controls progression from G1 to S phase. Most of these events precede the observed reduction in MCF-7 cell growth, which begins at Day 3 of RA treatment. q 1997 Academic Press

INTRODUCTION

Retinoic acid (RA), a natural derivative of vitamin A, plays a pivotal role in development, differentiation, and cell growth [1]. Many of the actions of RA are generally mediated through two families of nuclear receptors, the retinoic acid receptors and the retinoid X receptors. These receptors act as ligand-inducible transcription factors that increase the transcription of target genes by binding to cis-acting retinoic acid-response elements on DNA [2–4]. 1 To whom reprint requests should be addressed at Laboratory of Cellular Carcinogenesis and Tumor Promotion, National Cancer Institute, Building 37, Room 3A17, 37 Convent Drive, Bethesda, MD 20892-4255. Fax: (301) 496-8709. E-mail: [email protected].

RA has been shown to inhibit the growth of estrogen receptor-positive human breast cancer cells in a number of studies, but little is known about its effects on cell cycle regulatory proteins. Moreover, RA inhibition of cell growth does not seem to be due to general cytotoxicity, because cell division continues after RA treatment [5]. It also appears that RA-treated human breast cancer MCF-7 cells do not undergo typical differentiation [6]. Cell cycle control is the major regulatory mechanism of cell growth [7]. In recent years, considerable advances have been made in the understanding of the role of cyclins, cyclin-dependent kinases (CDKs), CDK inhibitors (CDIs), retinoblastoma protein (pRB), and the E2F transcription factor family in cell cycle progression. This process is regulated by the coordinated action of CDKs in association with their specific regulatory cyclin proteins. Primary regulators of G1 progression are the D-type cyclins (D1, D2, D3), cyclin E, and CDK2, 4, and 6 [8, 9]. CDK4 and CDK6 are activated early in G1 by interactions with D-type cyclins [10, 11], whereas CDK2 is activated later in G1 by interactions with cyclin E [12, 13] and at the G1/S boundary and throughout S phase by interactions with cyclin A [14, 15]. Functional activation of both D-cyclin- and Ecyclin-dependent kinases is certainly required for G1 to S transition [16, 17]. CDIs, including p21 (WAF1/ CIP1), p27 (Kip1), and p16 (Ink4), also contribute to the regulation of cell cycle progression by controlling CDK activity. p21, as well as p27, inhibits a wide variety of cyclin–CDK complexes in vitro, including CDK4 and CDK2 complexes, and overexpression of these proteins blocks progression of cells through G1 [18]. In addition to p21 and p27, a distinct family of 15- to 20kDa inhibitors called p16 (Ink4) can also associate with CDK4 and specifically inhibit CDK4 activity. The retinoblastoma gene product (pRB) is a nuclear protein which functions as a tumor suppressor. The phosphorylation of pRB is important for the regulation of cell cycle progression [19, 20]. pRB phosphorylation and inactivation of its growth suppressive function appear to be necessary for S phase entry. pRB is maintained in an underphosphorylated active state through much of G1 phase and becomes inactivated by further phosphorylation in late G1 phase, releasing seques-

293

AID

ECR 3589

/

6i23$$$521

07-01-97 13:00:00

0014-4827/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

ecal

294

ZHU ET AL.

tered transcription regulators which are members of the E2F family of transcription factors and enable cells to progress to S phase [19, 21]. E2F was originally identified as a DNA-binding activity that is stimulated by the adenovirus E1A protein [22]. At least five different E2F family members (E2F1–5) and at least three DP family members (DP1–3) have been cloned [23]. Considerable evidence suggests that the E2Fs and DPs constitute a family of heterodimeric transcription factors that play a critical role in cell growth control [23] and E2F appears to directly regulate the transcription of genes implicated in cell growth control. Although overexpression of E2F1 in tissue culture cells can stimulate cell proliferation and be oncogenic, loss of E2F1 in mice results in tumorigenesis, indicating that E2F1 may also function as a tumor suppressor [24, 25]. Wilcken et al. [26] have recently reported that RA does not affect cyclin D1 gene expression nor significantly change the mRNA or protein levels of cyclin D3, cyclin E, CDK2, or CDK4 in human breast cancer T47D cells. In our study, a low concentration of serum (0.5%) was used to reduce the impact of serum albumin on RA availability [27, 28]. These conditions permit greater availability of medium RA to MCF-7 cells, which could grow steadily in the medium with low concentration of serum. We found profound effects of RA on the expression of specific cyclins and related CDKs. RA reduced pRB protein expression as well as pRB phosphorylation and decreased the expression of E2F1 transcription factor and E2F1 promoter activity. Therefore, this study demonstrates important RA-induced changes in cell cycle control proteins, which appear to be responsible for the RA inhibition of cell growth. MATERIALS AND METHODS Cell culture and RA treatment. MCF-7 human breast cancer cells were obtained from the American Type Culture Collection (Rockville, MD) and maintained in Dulbecco’s MEM/Ham’s F12, 1:1 (DMEM/ F12) supplemented with antibiotics and 10% FBS. All-trans-retinoic acid was obtained from Sigma Chemical Co. (St. Louis, MO). RA was stored as a 1000-fold concentration in DMSO in brown glass ampules protected from light and oxygen in an inert atmosphere. MCF-7 cells were plated in 100-mm dishes and grown in DMEM/F12 medium with 10% FBS for 16–24 h. At this time, the cultures are about 50– 60% confluent, and the medium was changed to 0.5% FBS containing DMEM/F12 medium and cultured for an additional period of 16–24 h. 1 mM concentration of RA (final concentration) or the solvent DMSO was added to the medium to start treatment and kept for the indicated time; medium was replaced daily. Cell growth experiments. Subconfluent cultured cells were harvested by treating the cells with trypsin (0.05% trypsin and 0.2% EDTA; Biofluids, Inc., Rockville, MD). Cells were seeded in a series of 60-mm-diameter tissue culture dishes at 1 1 105 cells/dish in the medium with 10% FBS, and after incubation for 12 h the medium was changed to the 0.5% FBS-containing DMEM/F12. The cells were treated with RA or DMSO vehicle as described above and cultured at 377C in a humidified atmosphere composed of 95% air and 5% CO2 . Dishes were removed from the incubator at each of the indicated times (24–96 h) and rinsed twice with 5 ml PBS (pH 7.4); cells

AID

ECR 3589

/

6i23$$$521

07-01-97 13:00:00

FIG. 1. Effect of RA on MCF-7 cell growth. MCF-7 cells were seeded in a series of 60-mm-diameter tissue culture dishes at 1 1 105 cells/dish in the medium with 10% FBS, and after incubation for 12 h the medium was changed to a 0.5% FBS-containing DMEM/ F12 medium. The cells were treated with 0.01–1 mM RA or DMSO vehicle for the indicated time. The result for each time point shown in the growth curve represents means of triplicate cultures. A, DMSO; B, 0.01 mM RA; C, 0.1 mM RA; D, 1 mM RA.

were detached after a brief exposure to 0.05% trypsin and suspended repeatedly to give a single-cell suspension. The number of cells was measured using an electronic Coulter counter (Coulter Electronics, Inc., Hialeah, FL). The result of each time point shown in the growth curve (Fig. 1) represents averages from triplicate cultures. Cell cycle analysis. MCF-7 cells were plated on 100-mm cell culture dishes and treated with RA or DMSO vehicle as described above for the indicated time. Cultures were trypsinized, washed in cold PBS, and fixed overnight in cold 70% ethanol. Fixed cells were treated with 100 mg/ml RNase A in PBS for 1 h, followed by staining with 50 mg/ml propidium iodide in PBS. Flow cytometric analysis of stained cells was performed with a Becton–Dickinson FACScan (Becton–Dickinson, San Jose, CA). Data analysis was performed with Becton–Dickinson cellFIT cell-cycle analysis, version 2.01.2, to determine percentages of cells in the G1, S, and G2 / M phases of the cell cycle. Western blot analysis. Cell monolayers were washed twice in icecold PBS, and Laemmli buffer without reducing agent and bromphenol blue was added. Whole cell lysates were boiled for 5 min and centrifuged to remove insoluble cell debris. Protein concentration was determined by the bicinchoninic acid method (Pierce). b-Mercapthanol and a saturated solution of bromphenol blue were added to the samples to 1% final concentration. Equal amounts of protein were loaded onto polyacrylamide 4–15% gradient gels (for pRB protein detection, 10% polyacrylamide gel was used). The proteins were transferred to a supported nitrocellulose membrane (Bio-Rad) on a Bio-Rad electroblot apparatus. The blots were blocked at 47C using 5% nonfat milk overnight in TBS (50 mM Tris–HCl, pH 7.4, 150

ecal

RETINOIC ACID AND THE CELL CYCLE mM NaCl) and incubated 2 h at room temperature with the following primary antibodies: monoclonal anti-cyclin A (BF683), E2F1 (KH95), polyclonal anti-cyclin D3 (C-16), cyclin E (C-19), CDC2 (PSTAIRE), CDK2 (M2), and CDK4 (C-22) from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA); monoclonal anti-p21 (CIP1/WAF1), and p27 (kip1) from Transduction Laboratories (Lexington, KY); polyclonal anti-p16 (Ink4) from Upstate Biotech., Inc. (Lake Placid, NY); and monoclonal anti-pRB (14001A) from PharMingen (San Diego, CA). The blots were then washed three times and incubated 2 h with HRP-conjugated secondary antibodies at a 1:2000 dilution in TBS. Proteins were visualized using the ECL detection system (Amersham). Northern blot analysis. Isolation of total RNA was performed with TRIzol LS reagent from GIBCO BRL (Gaithersburg, MD). The open reading frame fragment of human E2F1 was excised from the plasmid PSP72-E2F1 and the probes were labeled with [32P]dCTP using random primer labeling methods. Total RNA (20 mg) was fractionated on a 1% agarose gel and blotted overnight onto Schleicher & Schuell nitrocellulose. The membranes were prehybridized for 5 h at 427C in a prehybridization solution of 61 SSC, 51 Denhardt’s reagent, 0.5% SDS, 100 mg/ml denatured, fragmented salmon sperm DNA, 50% formamide. The probes (5 1 106 cpm/ml) were boiled and added to the prehybridization buffer and the membranes were hybridized for 24 h at 427C. After washing, autoradiography on Kodak X-Omat AR film was performed on double intensifying screens. In order to reprobe the membranes with labeled GAPDH, the blots were stripped using boiled 0.1% SDS solution and the solution was allowed to cool to room temperature. Transient transfection and chloramphenicol acetyltransferase (CAT) assay. MCF-7 cells were seeded at a density of 1 1 105/100mm cell culture dish. One day later, cotransfections were performed by the calcium phosphate method using 10 mg E2F1-CAT reporter plasmid or pCAT-Basic vector along with 5 mg of pCH110 containing the b-galactosidase gene. The E2F1 promoter–CAT constructs were kindly provided by Dr. Joseph Nevins (Duke University Medical Center), and pCH110 (Pharmacia, Piscataway, NJ) was used for internal control to normalize the transfection efficiency. All transfections were performed in triplicate and repeated two times. After transfection, the media were replaced with fresh media containing 0.5% FBS and incubated for 16 h, and 1 mM RA (or vehicle DMSO) was added. The transfected cells were treated with RA for 48 h. Cells were harvested and lysed with lysis buffer (Boehringer-Mannheim, Germany); protein quantitation was performed using the Bio-Rad protein assay according to the manufacturer’s instructions. Chloramphenicol acetyltransferase or b-galactosidase (b-Gal) activity was determined on equal amounts of protein (100 mg/well) from each sample, using the CAT ELISA kit or b-Gal ELISA kit from Boehringer-Mannheim. Substrate enhancer with POD substrate ABTS was used to increase the sensitivity of the CAT assay.

RESULTS

RA Inhibits MCF-7 Cell Growth and Increases the Proportion of Cells in G1 Phase The inhibitory effect of RA on MCF-7 cells grown in 0.5% serum was concentration-dependent over the range 0.01–1 mM (Fig.1). Growth inhibition was observed at Day 3 in cells grown in 1 mM RA. Cells grown in 10% serum showed growth inhibitory effects starting 24–48 h later (not shown), possibly because RA is mostly found in complex with serum albumin under these conditions. Cell cycle analysis showed that little change was apparent over the first 24 h of exposure to RA, but the proportion of cells in G1 phase started to increase after

AID

ECR 3589

/

6i23$$$521

07-01-97 13:00:00

295

FIG. 2. Effect of RA on the cell cycle phase distribution of MCF7 cells. MCF-7 cells were plated on 100-mm cell culture dishes in 10% FBS and kept for 16 h. The medium was then changed to 0.5% FBS-containing DMEM/F12 and incubation proceeded for another 16 h. The cells were treated with RA (1 mM) or DMSO vehicle for the indicated time and cell cycle analysis was performed. Each point showing the percentage of cells in the G1, S, and G2 / M phase represents the mean of triplicate samples. Solid lines, RA-treated cells; broken lines, DMSO-treated cells.

RA treatment for 48 h and reached approximately 130% of the control value by 72 h. These increases were mirrored by decreases in the proportion of cells in S phase and to a lesser extent in the proportion of cells in G2 / M phase (Fig. 2). Effects of RA on Cyclin and CDK Gene Expression RA treatment produced a significant decrease in cyclin D3 expression (Fig. 3) which was detected 24 h after treatment, with over 80% reduction at 72 h after RA treatment, while no obvious change in cyclin D1 expression was observed after RA treatment (data not shown). Figure 3 also shows that the expression of cyclins A and E and the expression of CDK2 were unaffected by RA treatment. CDC2 expression was reduced by 50% by 72 h of RA treatment. Cyclin B1 was reduced by 66% between 48 and 72 h of RA treatment, and CDK4 was reduced after 24 h. These results suggest that cyclin D3 and CDK4 are major early targets of RA action in MCF-7 cells. Effects of RA on CDK Inhibitors A number of proteins bind stoichiometrically to CDKs or cyclin/CDK complexes and render the com-

ecal

296

ZHU ET AL.

FIG. 5. Effects of RA on pRB and its phosphorylation. Total cell lysates from MCF-7 cells treated with RA (1 mM) or DMSO vehicle were electrophoretically separated on a 10% polyacrylamide gel and immunoblotted with anti-pRB antibody. The upper band, ppRB, represents the hyperphosphorylated form of RB, and the lower band, pRB, represents the hypophosphorylated form.

FIG. 3. Effects of RA on the expression of cyclins and CDKs in MCF-7 cells. Total cell lysates from MCF-7 cells treated with RA (1 mM) or DMSO vehicle were analyzed by SDS–PAGE, and Western blots were detected with antibodies to cyclins A, E, B1, and D3, CDC2, CDK2, and CDK4.

plexes inactive or unavailable for activating phosphorylation [29]. These inhibitory proteins play a negative regulatory role in G1, and disappearance of these inhibitory activities is associated with G1/S progression induced by a number of mitogens [30, 31]. It was of interest to determine whether CDK inhibitors, including p27, p21, and p16, might be targets of RA action. After RA treatment, Western blot analysis showed that the levels of p27 and p16 proteins did not change in RAtreated MCF-7 cells, while, unexpectedly, the level of p21 was markedly reduced after RA treatment for 12 h (Fig. 4). Regulation of RB Gene Expression and pRB Phosphorylation by RA Since pRB functions as a regulator of cell cycle progression at the late G1 phase [19, 21], and G1 cyclin/ CDKs can regulate the activity of pRB, we studied pRB expression and its phosphorylation. Western blot analysis of cell lysates with anti-RB antibodies showed two separate but closely migrating bands; the upper one represents the more highly phosphorylated pRB and the lower one the unphosphorylated or hypophosphory-

FIG. 4. Effect of RA on CDK inhibitors. Total cell lysates from MCF-7 cells treated with RA (1 mM) or DMSO vehicle were analyzed by SDS–PAGE, and Western blots were detected with antibodies to cyclins p16, p27, and p21.

AID

ECR 3589

/

6i23$$$521

07-01-97 13:00:00

lated pRB. Treatment with RA resulted in a decrease both in pRB protein expression and in pRB phosphorylation. This decrease could already be observed at 24 h. RA treatment for 48 h significantly reduced pRB expression and abrogated its phosphorylation (Fig. 5), consistent with the consequent inhibition of cell growth (Fig. 1). RA Reduced E2F1 Expression and E2F1 Transcriptional Activity Western blot analysis showed a marked reduction of E2F1 protein after RA treatment for 72 h and showed that this reduction starts from the 48-h time point (Fig. 6B). E2F1 mRNA was also reduced (Fig. 6A). Next we tested the possibility that RA regulates E2F1-mediated transcriptional activity, presumably through inhibition of pRB phosphorylation. RA (1 mM, 48 h)-treated cells showed a 60% reduction in E2F1 promoter-driven CAT expression compared to the DMSO (Fig. 6C). A control pCAT-Basic vector expressed a very low to undetectable level of CAT, and RA did not show any obvious effect on the CAT expression in pCAT-Basic transfected cells (data not shown). DISCUSSION

We chose a 0.5% serum medium to maximize the effectiveness of RA, since higher serum concentrations are known to reduce RA effectiveness, probably because of complex formation with albumin [27]. We confirmed the growth inhibitory activity of RA under these conditions and the inhibition of cell cycle progression with resulting G1 arrest. The results show that cyclin D3 and CDK4 are early (24 h) targets of RA in MCF7 cells. In succession, RA can also significantly reduce pRB protein expression and abrogate its phosphorylation at 48 h. Finally, RA showed marked inhibitory effects on E2F1 mRNA and protein levels at 72 h, as well as on E2F1 promoter activity. This sequence of events (Table 1) is consistent with a fundamental inhibitory role of RA on general transcriptional events under these conditions. D cyclin/CDK4 and E cyclin/CDK2 complexes govern the rate of G1/S progression in mammalian cells [8]. CDKs require complex formation with appropriate

ecal

297

RETINOIC ACID AND THE CELL CYCLE

cers and breast cancer cell lines [33, 34]. Cyclin D1 (0/0) knockout mice show disrupted development of mammary gland epithelium [35]. Proliferating MCF-7 cells overexpress cyclin D1 mRNA [33], and microinjection studies have shown that cyclin D1 and cyclin E are required for cell cycle progression in MCF-7 cells. In our assay system, cyclin D1 was expressed in MCF7 cells at a moderate level and did not show obvious response to RA treatment. This result is similar to the recent report on T-47D human breast cancer cells in which RA failed to affect cyclin D1 mRNA and protein levels [26]. An important finding of our study is that RA treatment results in significant reduction of cyclin D3 and CDK4 expression in MCF-7 cells. RA was reported to inhibit CDK4 activity in MCF-7 cells after treatment for 24 h, but could not inhibit expression of cyclin D3 and CDK4 in 5% serum cultures [26]. Serum concentration might be the major reason for these different results. We showed in this study that RA can reduce the expression of cyclin B1 and CDC2, which may be responsible for RA-induced reduction of cell proportion in the G2/M phase. Our results demonstrate for the first time that RA can reduce the expression of cyclin D3/CDK4, but not cyclin E/CDK2. Therefore cyclin D3 and CDK4 appear to mediate RA effects on cell cycle progression. CDK inhibitors, which can negatively control CDK activity and potentially act as tumor suppressors [18], have been examined for their response to some extracellular positive or negative stimuli of cell growth in a number of studies. p27 was recently shown to be strongly downregulated by estradiol in MCF-7 cells [36], and p27 as well as p21 was found to be induced FIG. 6. Effect of RA on E2F1 transcription factor. (A) RA reduced the E2F1 mRNA level in MCF-7 cells. Cells treated with RA (1 mM) or DMSO vehicle were harvested at the indicated time for total RNA extraction. 20 mg of RNA per lane was analyzed and Northern blots were probed for E2F1 mRNA level in MCF-7 cells. The same blot was stripped and reprobed with 32P-labeled GAPDH. (B) RA reduced the E2F1 protein level in MCF-7 cells. Total cell lysates from RA (1 mM) or DMSO vehicle-treated cells were electrophoretically separated, and Western blots were probed with monoclonal anti-E2F1 antibody. (C) Effect of RA on the E2F1 promoter activity. E2F1 promoter-CAT constructs or pCAT-Basic vector (as a control) was cotransfected with pCH110 into MCF-7 cells. After calcium phosphate transfection, the medium was replaced with fresh medium containing 0.5% FBS and incubated for 16 h, and then 1 mM RA (or vehicle DMSO) was added. The transfected cells were treated with RA for 48 h. 100 mg of each protein sample in the triplicate group was subject to the measurement of CAT activity or b-Gal activity with CAT or b-Gal ELISA kit. The level of CAT activity was corrected for the activity derived from the internal control of pCH110 encoding the b-galactosidase gene.

cyclin proteins for activation in addition to phosphorylation by CDK-activating kinase [8, 32]. A number of studies have demonstrated that cyclin genes are amplified and overexpressed at high frequency in breast can-

AID

ECR 3589

/

6i23$$$521

07-01-97 13:00:00

TABLE 1 Summary of RA Effects on Cell Cycle Regulatory Proteins (Summary of Densitometry Analysis) 12 h

24 h

48 h

72 h

99%a 94% 106% 101%

105% 98% 102% 81%

88% 102% 98% 38%

71% 99% 32% 17%

CDC2 CDK2 CDK4

102% 87% 82%

102% 84% 43%

71% 102% 41%

43% 99% 37%

p27 p16 p21

102% 98% 50%

95% 100% 42%

91% 110% 45%

88% 109% 48%

pRB

99%

74%

21%

19%

E2F1

87%

67%

41%

15%

Cyclin Cyclin Cyclin Cyclin

A E B D3

Note. Densitometric evaluation of Western blot was performed with a Molecular Dynamic Research densitometer. a [Densitometry arbitrary units (DAU) obtained from RA-treated cell/DAU from DMSO-treated cells at the same time point] 1 100%.

ecal

298

ZHU ET AL.

by antiestrogen ICI 182780 in MCF-7 cells [37], suggesting that estrogen or antiestrogen growth inhibition of MCF-7 cells is at least partially mediated by regulation of p27 or p27/p21. Our present results show that RA does not obviously affect p27 or p16 expression, indicating that p27 and p16 are not involved in RA inhibition of cell growth in MCF-7 cells. Unexpectedly, reduced p21 expression was found after RA treatment for 12 h. p21 is known to bind and inhibit a wide variety of cyclin/CDK complexes, including cyclin D/CDK4, cyclin E/CDK2, and cyclin A/CDK2. p21 is also a target of p53 and is regulated by p53 [18, 32]; p21 expression can be induced by wt (but not mutant) p53 via its p53-binding site [38]. We are not able, at present, to explain the significance and mechanism of p21 reduction by RA. Our study further shows that pRB is an important RA target in MCF-7 cells. RA dramatically reduced pRB expression as well as pRB phosphorylation. The effect of RA on pRB is more obvious than the effect recently observed in T-47D cells [26]. As we observed in this study, the obvious reduction of pRB expression and its phosphorylation was seen starting at 48 h of RA treatment. pRB serves as a main substrate for G1 CDKs and its major role is to act as a signal transducer, connecting the cell cycle clock with the transcriptional machinery [19]. Transition between hyper- and hypophosphorylated pRB controls cell cycle progression. Physiologic signals that favor cell proliferation were reported to enhance pRB phosphorylation. For example, TGF-b, as a growth stimulator in mouse fibroblasts C3H/10T1/2 cells, induces the phosphorylation of pRB, as does EGF. TGF-b was also shown to increase pRB expression [39]. On the other hand, treatment for 18 h with an antiestrogen compound could greatly reduce the pRB phosphorylation and cause a decrease in total pRB protein in MCF-7 cells [37]. It therefore seems that both RA and antiestrogen compounds may inhibit MCF-7 cells through inhibiting pRB protein expression and its phosphorylation. Our results also support the idea that the differentiation effects may not be responsible for RA inhibition of breast cancer cell growth, because several lines of evidence suggest that pRB levels strongly increase upon differentiation of basal keratinocytes and colonic crypt cells [37]. Although pRB has an important role in regulating the G1 to S phase transition, available evidence suggests that E2Fs are downstream effectors of pRB or critical targets through which pRB and related proteins inhibit cell growth [23]. Among the E2F family members, E2F1 is the best studied and is thought to be preferentially associated with pRB and also primarily regulated by pRB [23]. In our study, RA significantly reduced the E2F1 mRNA and protein expression in MCF-7 cells treated for 72 h. CAT assay further demonstrated that E2F1 promoter is one of the targets for RA and its activity was inhibited by 60% after 72 h of RA

AID

ECR 3589

/

6i23$$$521

07-01-97 13:00:00

treatment, which is at least in part responsible for the reduction in RA-induced E2F1 protein expression. E2F1 is regulated by a number of proteins, including pRB, CDKs, and their inhibitors [40]. RA-induced reduction of E2F1 expression may result from its effects on cyclin D/CDK4 and pRB (Table 1). Numerous in vitro studies have shown the growth stimulating effects of E2F1. For example, TGF-b, a growth stimulator in C3H/10T1/2 mouse fibroblasts, induced pRB expression and its phosphorylation; TGF-b can also stimulate wild-type adenoviral E2 promoter activity 12-fold [39]. On the other hand, while TGF-b acts as a growth suppressor in mink lung epithelial cells, E2F1 RNA levels are drastically reduced and overexpression of E2F1 can overcome this TGF-b-mediated effect [41]. Our results are in line with these in vitro findings showing E2F1 growth stimulating effects, although recent in vivo studies have shown that E2F1 may also function as a tumor suppressor [24, 25]. A number of cellular genes implicated in the control of cell cycle progression have been identified to contain E2F1-binding sites. These include the proto-oncogenes c-myc and B-myb; genes encoding enzymes required for DNA synthesis such as DHFR, thymidine kinase, thymidylate synthase and DNA polymerase; and genes encoding components of the basic cell cycle clock, such as CDC2, cyclin A, and cyclin D1 [23]. RA can obviously reduce CDC2 expression and, to a lesser extent, cyclin A, but not cyclin D1. In conclusion, we have demonstrated a downregulatory sequential effect of RA on cyclin D3/CDK4, pRB and its phosphorylation, and E2F1 transcription factor. Since these effects precede or occur during the inhibition of cell growth by RA, they may be responsible for some of the effects of RA on cell growth and cell cycle regulation. We thank Dr. William Kaelin of the Dana-Farber Cancer Institute for a generous gift of plasmid PSP72-E2F1, and we also thank Dr. Joseph R. Nevins for the human E2F1 promoter construct pE2F1CAT.

REFERENCES 1. Lotan, R., Lotan, D., and Sacks, P. G. (1990) Methods Enzymol. 190B, 100–110. 2. Chambon, P. (1994) Semin. Cell Biol. 5, 115–125. 3. Gudas, L. J. (1994) J. Biol. Chem. 269, 15399–15402. 4. Mangelsdorf, D. J. (1994) in The Retinoids, (Anonymous), 2nd Ed., pp. 319–348, Raven Press, New York. 5. Garcia, M., Derocq, D., Freiss, G., and Rochefort, H. (1992) Proc. Natl. Acad. Sci. USA 89, 11538–11542. 6. Guilbaud, N. F., Gas, N., Dupont, M. A., and Valette, A. (1990) J. Cell Physiol. 145, 162–172. 7. Rittling, S. R., Brooks, K. M., Cristofalo, V. J., and Baserga, R. (1986) Proc. Natl. Acad. Sci. USA 83, 3316–3320. 8. Sherr, C. J. (1993) Cell 73, 1059–1065. 9. Sherr, C. J. (1994) Cell 79, 551–555.

ecal

RETINOIC ACID AND THE CELL CYCLE 10. Matsushime, H., Ewen, M. E., Strom, D. K., Kato, J. Y., Hanks, S. K., Roussel, M. F., and Sherr, C. J. (1992) Cell 71, 323–334. 11. Meyerson, M., and Harlow, E. (1994) Mol. Cell Biol. 14, 2077– 2086. 12. Dulic, V., Lees, E., and Reed, S. I. (1992) Science 257, 1958– 1961. 13. Koff, A., Giordano, A., Desai, D., Yamashita, K., Harper, J. W., Elledge, S., Nishimoto, T., Morgan, D. O., Franza, B. R., and Roberts, J. M. (1992) Science 257, 1689–1694. 14. Elledge, S. J., Richman, R., Hall, F. L., Williams, R. T., Lodgson, N., and Harper, J. W. (1992) Proc. Natl. Acad. Sci. USA 89, 2907–2911. 15. Rosenblatt, J., Gu, Y., and Morgan, D. O. (1992) Proc. Natl. Acad. Sci. USA 89, 2824–2828. 16. Pagano, M., Pepperkok, R., Lukas, J., Baldin, V., Ansorge, W., Bartek, J., and Draetta, G. (1993) J. Cell Biol. 121, 101–111. 17. Tsai, L. H., Lees, E., Faha, B., Harlow, E., and Riabowol, K. (1993) Oncogene 8, 1593–1602. 18. Hunter, T., and Pines, J. (1994) Cell 79, 573–582. 19. Weinberg, R. A. (1995) Cell 81, 323–330. 20. Whyte, P. (1995) Semin. Cancer Biol. 6, 83–90. 21. Riley, D. J., Lee, E. Y., and Lee, W. H. (1994) Annu. Rev. Cell Biol. 10, 1–29. 22. Kovesdi, I., Reichel, R., and Nevins, J. R. (1986) Cell 45, 219– 228. 23. Adams, P. D., and Kaelin, W. G., Jr. (1996) Curr. Top. Microbiol. Immunol. 208, 79–93. 24. Yamasaki, L., Jacks, T., Bronson, R., Goillot, E., Harlow, E., and Dyson, N. J. (1996) Cell 85, 537–548. 25. Field, S. J., Tsai, F. Y., Kuo, F., Zubiaga, A. M., Kaelin, W. G., Jr., Livingston, D. M., Orkin, S. H., and Greenberg, M. E. (1996) Cell 85, 549–561. 26. Wilcken, N. R., Sarcevic, B., Musgrove, E. A., and Sutherland, R. L. (1996) Cell Growth Differ. 7, 65–74.

27. Avis, I., Mathias, A., Unsworth, E. J., Miller, M. J., Cuttitta, F., Mulshine, J. L., and Jakowlew, S. B. (1995) Cell Growth Differ. 6, 485–492. 28. Takatsuka, J., Takahashi, N., and De Luca, L. M. (1996) Cancer Res. 56, 675–678. 29. Sherr, C. J., and Roberts, J. M. (1995) Genes Dev. 9, 1149–1163. 30. Toyoshima, H., and Hunter, T. (1994) Cell 78, 67–74. 31. Xiong, Y., Hannon, G. J., Zhang, H., Casso, D., Kobayashi, R., and Beach, D. (1993) Nature 366, 701–704. 32. Peeper, D. S., Van der Eb, A. J., and Zantema, A. (1994) Biochim. Biophys. Acta 1198, 215–230. 33. Buckley, M. F., Sweeney, K. J., Hamilton, J. A., Sini, R. L., Manning, D. L., Nicholson, R. I., deFazio, A., Watts, C. K., Musgrove, E. A., and Sutherland, R. L. (1993) Oncogene 8, 2127– 2133. 34. Keyomarsi, K., and Pardee, A. B. (1993) Proc. Natl. Acad. Sci. USA 90, 1112–1116. 35. Sicinski, P., Donaher, J. L., Parker, S. B., Li, T., Fazeli, A., Gardner, H., Haslam, S. Z., Bronson, R. T., Elledge, S. J., and Weinberg, R. A. (1995) Cell 82, 621–630. 36. Foster, J. S., and Wimalasena, J. (1996) Mol. Endocrinol. 10, 488–498. 37. Watts, C. K., Brady, A., Sarcevic, B., deFazio, A., Musgrove, E. A., and Sutherland, R. L. (1995) Mol. Endocrinol. 9, 1804– 1813. 38. El-Deiry, W. S., Tokino, W. S., Velculescu, V. E., Levy, D. B., Parsons, R., Trent, J. M., Lin, D., Mercer, W. E., Kinzler, K. W., and Vogelstein, B. (1993) Cell 75, 817–825. 39. Kim, T. A., Ravitz, M. J., and Wenner, C. E. (1994) J. Cell Physiol. 160, 1–9. 40. Guy, C. T., Zhou, W., Kaufman, S., and Robinson, M. O. (1996) Mol. Cell Biol. 16, 685–693. 41. Schwarz, J. K., Bassing, C. H., Kovesdi, I., Datto, M. B., Blazing, M., George, S., Wang, X. F., and Nevins, J. R. (1995) Proc. Natl. Acad. Sci. USA 92, 483–487.

Received February 6, 1997 Revised version received April 2, 1997

AID

ECR 3589

/

6i23$$$521

07-01-97 13:00:00

299

ecal