Regulation of Cell Growth and Cyclin D1 Expression by the Constitutively Active FRAP-p70^s6K Pathway in Human Pancreatic Cancer Cells¹ Free

Pancreatic cancer constitutes the fourth leading cause of cancer death in Western countries for both sexes and has a dismal prognosis (1). The growth of pancreatic cancer is driven by multiple factors such as activating mutations in the small GTPase Ki-ras and various growth factors acting in an autocrine or paracrine manner (2). However, the downstream targets of these proteins and the arising signal transduction pathways involved in autocrine/paracrine human pancreatic cancer cell growth are poorly understood. Defining these pathways could give rise to novel therapeutic approaches that are urgently needed.

The serine/threonine kinase p70^s6K is a highly conserved element in a wide array of cellular processes including the mitogenic response to growth factors (3). This enzyme is activated in vivo by phosphorylation mediated in part by a phosphatidyl kinase-related kinase, FRAP³ or mTOR (mammalian target of rapamycin; Ref. 4). The immunosuppressant rapamycin has emerged as a useful tool to elucidate the cellular function of FRAP and its downstream target, p70^s6K (3, 5, 6): rapamycin inhibits FRAP by forming a stable complex with the immunophilin FK506-binding protein, which binds to FRAP. As a result of this interaction, rapamycin induces dephosphorylation of several sites (Thr²²⁹, Thr³⁸⁹, and Ser⁴⁰⁴) on p70^s6K, leading to its rapid inactivation (7). Interestingly, rapamycin blocks the proliferation of a variety of cells that have not entered the cell cycle (8, 9). Furthermore, we have recently shown that rapamycin inhibits constitutive p70^s6K phosphorylation and cell growth in classical small cell lung cancer cell lines (10). Consequently, there has been considerable interest in the downstream targets of rapamycin and p70^s6K that include 5′-terminal oligopyrimidine tract mRNA translation and protein synthesis (11). The translation inhibitor 4E-BP1 is also phosphorylated by the FRAP-p70^s6K pathway. This leads to the dissociation of 4E-BP1 from initiation factor eIF-4E, permitting increased protein translation and mitogenesis (12). In contrast, dephosphorylated 4E-BP1 interacts with eIF-4E and thereby inhibits cap structure-dependent protein synthesis and cell growth (13). Interestingly, rapamycin induces dephosphorylation of 4E-BP1, leading to inactivation of eIF-4E (6, 14). In contrast to previous studies, more recent reports suggest that 4E-BP1 is not a direct downstream target of p70^s6K but is regulated by FRAP in a parallel manner (14). This view is supported by results obtained in p70^s6K −/− cells, demonstrating that rapamycin can still prevent phosphorylation of 4E-BP1 and inhibit growth (15). Thus, rapamycin most likely inhibits growth via both p70^s6K-dependent and -independent pathways. The role of these rapamycin-sensitive pathways in the growth of pancreatic cancer cells is unknown.

In the present study, we demonstrate that the FRAP-p70^s6K pathway is constitutively phosphorylated/active in MiaPaCa-2 and Panc-1 human pancreatic cancer cell lines and a pancreatic cancer tissue sample. Rapamycin induced p70^s6K dephosphorylation and inactivation of constitutively active p70^s6K in serum-starved MiaPaCa-2 and Panc-1 cells. Rapamycin also inhibited constitutive phosphorylation of the translation inhibitor 4E-BP1 in these cells. Furthermore, proliferation and colony formation of MiaPaCa-2 and Panc-1 human pancreatic cancer cells were markedly reduced in the presence of rapamycin. Finally, rapamycin profoundly inhibited expression of cyclin D1. Thus, our results suggest that the rapamycin-sensitive FRAP-p70^s6K pathway could serve as a novel target for therapeutic intervention in pancreatic cancer.

Human pancreatic cancer cell lines MiaPaCa-2 and Panc-1 were purchased from the American Type Culture Collection (Manassas, VA). Stocks were maintained in DMEM supplemented with 10% (v/v) FBS in a humidified atmosphere of 5% CO₂: 95% air at 37°C. The cells were passaged every 3 days.

Pancreatic carcinoma tissue samples were obtained from a patient undergoing a surgical operation for pancreatic cancer at the Department of General Surgery, University of Ulm. Informed consent was obtained from the patient before surgery. The tissue was collected after surgical removal, snap-frozen immediately in liquid nitrogen, and stored at −80°C.

MiaPaCa-2 and Panc-1 cells were washed twice in serum-free DMEM and incubated in fresh DMEM for an additional 24 h. Cells were then treated with rapamycin as indicated in the figure legends and lysed in 50 mm Tris-HCl, 5 mm EDTA, 100 mm NaCl, 40 mm β-glycerophosphate, 50 mm NaF, 1 mm Na₃VO₄, 1% Triton X-100, 1 mm phenylmethylsulfonyl fluoride, 10 μg/ml aprotinin, and 10 μg/ml leupeptin (pH 7.6; lysis buffer). For immunoprecipitations, lysates were incubated with a polyclonal anti-4E-BP1 antibody for 2 h at 4°C on a rotating wheel with protein A-Sepharose beads added for the second h. Beads were washed twice in lysis buffer and resuspended in 2× SDS-PAGE sample buffer. Proteins were further analyzed by SDS-PAGE, followed by Western blotting using a monoclonal anti-eIF-4E antibody or a polyclonal anti-4E-BP1 antibody with immunoreactive bands visualized by enhanced chemiluminescence detection. For p70^s6K− and 4E-BP1 mobility shift assays and detection of cyclin D1, cyclin E, and p27^kip1, cells were treated as indicated in the figure legends and lysed in SDS-PAGE sample buffer, and samples were further analyzed by SDS-PAGE and Western blotting with specific antisera to these proteins essentially as described above. Pancreatic tissues were lysed and dissociated by sonication. The lysates were subsequently boiled in SDS-PAGE sample buffer for 5 min and further analyzed by SDS-PAGE and Western blotting as described above.

Serum-starved MiaPaCa-2 and Panc-1 cells were incubated with rapamycin, PD 098059, and TGF-α as indicated in the figure legends. Controls received an equivalent amount of solvent. Cells were then lysed at 4°C in 1 ml of lysis buffer. Immunoprecipitations were performed at 4°C using an anti-p70^s6K antibody or an anti-ERK2 antibody for 2 h, with protein A-agarose added for the second hour. Immune complexes were washed three times in lysis buffer and once with p70^s6K kinase buffer [20 mm HEPES (pH 7.4), 10 mm MgCl₂, 1 mm DTT, and 10 mm β-glycerophosphate] or ERK kinase buffer [15 mm MgCl₂, 15 mm Tris-HCl (pH 7.4)], respectively. Kinase reactions were performed by re-suspending the protein A-Sepharose pellets in 25 μl of kinase assay mixture containing the appropriate kinase buffer with 0.2 mm S6 peptide (RRRLSSLRA) or myelin basic protein, 20 μm ATP, 5 μCi/ml [γ-³²P]ATP, 2 μm cyclic AMP-dependent protein kinase inhibitor peptide, and 100 nm microcystin LR. Incubations were performed under linear assay conditions at 30°C for 20 min and terminated by spotting 25 μl of the supernatant onto Whatman p81 chromatography paper. Papers were washed four times for 5 min in 0.5% o-phosphoric acid, immersed in acetone, and dried before scintillation counting. The average radioactivity of two blank samples containing no immune complex was subtracted from the result of each sample.

MiaPaCa-2 and Panc-1 cells were serum-starved for 24 h and then incubated with various additions of rapamycin as described in the figure legends. Control cells received solvent or DMEM/10% FBS to determine maximum DNA synthesis. After 18 h of incubation at 37°C, [³H]thymidine (0.25 μCi/ml; 1 μm) was added to the cultures, and cells were incubated for another 6 h at 37°C. Cells were then washed with PBS and incubated in 5% trichloroacetic acid at 4°C for 30 min, washed with ethanol, and solubilized in 1 ml of 2% Na₂CO₃, 0.1 m NaOH, and 1% SDS. The acid-insoluble radioactivity was determined by Cerenkov counting in 6 ml of Ultima Gold (Packard). For detection of BrdUrd incorporation into cellular DNA, MiaPaCa-2 and Panc-1 cells were serum-starved in fresh DMEM for 24 h. Cells were then incubated with 20 ng/ml rapamycin or solvent for 24 h at 37°C, with 10 μm BrdUrd added for the last 6 h. Cultures were subsequently washed with PBS, fixed in 70% ethanol for 20 min, and incubated with anti-BrdUrd monoclonal antibody, followed by labeling with an antimouse IgG-fluorescein antibody. Cells were examined using a Zeiss Axiophot immunofluorescence microscope. Data are expressed as the percentage of BrdUrd-labeled nuclei.

Three days after passage, MiaPaCa-2 and Panc-1 cells were washed in serum-free DMEM, trypsinized, and resuspended in DMEM/1% FBS. Cells were plated at a density of 1 × 10⁴ cells in 1 ml of DMEM/1% FBS in the presence or absence of 20 ng/ml rapamycin in duplicate. At the times indicated in the figure legends, the cell number was determined using a cell counting chamber.

MiaPaCa-2 and Panc-1 cells were washed, trypsinized, and resuspended in DMEM. The cell number was determined using a cell counting chamber. Cells (3 × 10⁴ ) were mixed with DMEM/1% FBS containing 0.3% agarose in the presence or absence of rapamycin at the concentrations indicated and layered over a solid base of 0.5% agarose in DMEM/1% FBS in the presence or absence of rapamycin at the same concentrations in 33-mm dishes. The cultures were incubated in humidified 5% CO₂: 95% air at 37°C for 14 days and then stained with the vital stain nitroblue tetrazolium. Colonies of >120 μm in diameter (20 cells) were counted using a microscope.

Rapamycin was obtained from Calbiochem-Novabiochem. Antibodies against p70^s6K, 4E-BP1, p27^KIP1, cyclin D1, and cyclin E were obtained from Santa Cruz Biotechnology. The monoclonal anti-eIF-4E antibody was from Transduction Laboratories. The NH₂-terminally directed anti-p70^s6K polyclonal antibody used to determine p70^s6K activity was obtained from Upstate Biotechnology, Inc. The polyclonal anti-ERK2 antibody was a kind gift of Dr. Jo van Lint (Katholieke Universiteit, Leuven, Belgium). PD 098059 was obtained from New England Biolabs. Protein A-Sepharose and the BrdUrd labeling and detection kit were from Boehringer Mannheim. Enhanced chemiluminescence reagents, [³H]thymidine, and [γ-³²P]ATP, were from Amersham. All other reagents were of the purest grade available.

Activation of p70^s6K by mitogens can be determined by the appearance of slower-migrating forms in SDS-PAGE due to the phosphorylation of p70^s6K on Thr²²⁹, Thr³⁸⁹, and Ser⁴⁰⁴, which are not basally phosphorylated in quiescent cells (3). The phosphorylation of these sites can be prevented or reversed by treatment with rapamycin, which specifically inhibits the p70^s6K activator FRAP (3, 4). To examine the status of p70^s6K phosphorylation in human pancreatic cancer cells, cultures of Panc-1 and MiaPaCa-2 cells were incubated in the absence or presence of 20 ng/ml rapamycin. Cells were lysed, and the lysates were analyzed by immunoblotting. p70^s6K exhibited a retarded electrophoretic mobility characteristic of the phosphorylated form of this enzyme in Panc-1 and MiaPaCa-2 cells. In control experiments, there was no retarded electrophoretic mobility of p70^s6K in serum-starved mouse or human fibroblasts (data not shown), in accordance with previous results (3). Treatment of cells with rapamycin induced a striking dephosphorylation of p70^s6K, as demonstrated by the increase in the electrophoretic mobility (Fig. 1,A, top panels). The effect of rapamycin on constitutive p70^s6K phosphorylation was concentration dependent; maximum effects were achieved at 6 and 20 ng/ml rapamycin in Panc-1 and MiaPaCa-2 cells, respectively (Fig. 1,A, bottom panels). Dephosphorylation of p70^s6K by rapamycin in Panc-1 and MiaPaCa-2 cells was first visible after 5 min and reached a maximum 10 min after the addition of rapamycin to the cells (data not shown). Next we performed immune complex kinase assays to directly assess p70^s6K activity. Treatment of cells with rapamycin substantially reduced the basal kinase activity of p70^s6K by 92% and 99% in serum-starved Panc-1 and MiaPaCa-2 cells, respectively (Fig. 1,B, top panels). The effect of rapamycin on p70^s6K activity was specific because rapamycin did not affect basal and TGF-α-stimulated activation of the mitogen-activated protein kinase ERK2 in both cell lines. In contrast, the mitogen-activated protein/ERK kinase 1 inhibitor PD 098059 markedly inhibited basal and TGF-α-stimulated ERK2 activation (Fig. 1 B, bottom panels).

It has been suggested that phosphorylation and thus inactivation of the translation inhibitor 4E-BP1 are also mediated by a rapamycin-sensitive pathway (6, 14). Therefore, we examined the phosphorylation status of 4E-BP1 in serum-starved human pancreatic cancer cells in the absence or presence of rapamycin. To determine 4E-BP1 phosphorylation, a Western blot analysis was performed using a polyclonal antibody against 4E-BP1. Three forms of 4E-BP1 were present in serum-starved Panc-1 and MiaPaCa-2 cells, suggesting constitutive phosphorylation of 4E-BP1 in these cells. Treatment with rapamycin leads to a decreased intensity of the upper two bands and to an increased intensity of the fastest migrating lower band (Fig.1,C, left, top and bottom panels). According to previous studies, the lowest band represents the unphosphorylated form of 4E-BP1 (16, 17). We next examined whether the constitutive phosphorylation of 4E-BP1 was sufficient to decrease the interaction of eIF-4E with 4E-BP1. Serum-starved Panc-1 cells were incubated with solvent or 20 ng/ml rapamycin. Cell lysates were immunoprecipitated with a polyclonal anti-4E-BP1 antibody and analyzed further by Western blotting with a monoclonal anti-eIF-4E antibody. As shown in Fig. 1,C (middle, top panel), a small amount of eIF4E coimmunoprecipitated with 4E-BP1 in nontreated cells. Upon treatment with rapamycin, the amount of eIF-4E that could be detected in 4E-BP1 immunoprecipitates increased strikingly. When blots were stripped and reprobed with anti-4E-BP1 antibody, similar amounts of 4E-BP1 protein could be detected in rapamycin-treated and untreated cells (Fig. 1 C, middle, bottom panel). In addition, using 7-methyl-GTP-Sepharose that specifically binds eIF-4E, an increased amount of 4E-BP1 was found in 7-methyl-GTP-Sepharose immunoprecipitates of lysates of Panc-1 cells treated with rapamycin as compared to untreated cells (data not shown). Thus, the level of constitutive phosphorylation of 4E-BP1 in serum-starved pancreatic cancer cells is sufficient to prevent interaction with eIF-4E.

Constitutive p70^s6K- and 4E-BP1 phosphorylation was not restricted to pancreatic cancer cells exhibiting activating Ki-ras mutations such as MiaPaCa-2 and Panc-1 cells but could also be detected in the human pancreatic cancer cell lines SW 850 and SW 979 that exhibit wild-type Ki-ras (data not shown). Thus, constitutive phosphorylation of p70^s6K and 4E-BP1 is not due to the activating Ki-ras mutation in these cells. Interestingly, constitutive phosphorylation of p70^s6K and 4E-BP1 could also be observed in a human pancreatic carcinoma tissue sample (Fig. 1 C, right panels).

We first examined the effect of rapamycin on basal DNA synthesis in both cell lines to determine the role of the constitutively active FRAP-p70^s6K pathway for pancreatic cancer cell growth. Basal DNA synthesis in serum-free DMEM as assessed by [³H]thymidine incorporation was 35% of maximum stimulation in response to 10% FBS in Panc-1 and MiaPaCa-2 cells (Fig. 2,A). Upon treatment with rapamycin, basal [³H]thymidine incorporation decreased substantially in both cell lines. The effect of rapamycin was concentration dependent, with half-maximal effects at 1 and 3 ng/ml in Panc-1 and MiaPaCa-2 cells, respectively, and maximal effects at 20 ng/ml in both cell lines. At this concentration, basal DNA synthesis was reduced by 77% and 66% in Panc-1 and MiaPaCa-2 cells, respectively (Fig. 2,A). These results are in good agreement with the data obtained on rapamycin-induced dephosphorylation of p70^s6K (Fig. 1,A). To further substantiate our observations, we applied a distinct technique in which DNA synthesis was determined using an immunofluorescence assay to detect BrdUrd incorporation into cell nuclei. As shown in Fig. 2,B, BrdUrd incorporation into cell nuclei was markedly inhibited in the presence of rapamycin. At a concentration of 20 ng/ml rapamycin, a maximum 62% reduction in BrdUrd incorporation could be detected in both cell lines (Fig. 2 C).

Next we examined the effect of rapamycin on actual cellular proliferation. MiaPaCa-2 and Panc-1 cells were incubated in the absence or presence of 20 ng/ml rapamycin, and cell numbers were determined over a period of up to 7 days. In the presence of rapamycin, cell numbers were reduced by a maximum of 66% and 68% at day 7 in Panc-1 and MiaPaCa-2 cells, respectively (Fig. 3 A). Thus, the constitutively active, rapamycin-sensitive FRAP-p70^s6K signaling pathway is likely to participate in sustaining anchorage-dependent growth of human Panc-1 and MiaPaCa-2 cells. In contrast to recent results obtained in rhabdomyosarcoma cells (18), rapamycin did not induce apoptosis in pancreatic cancer cells, as judged by an in situ assay to detect DNA fragmentation (data not shown).

Tumors and transformed cells, including human pancreatic cancer cells, are able to grow in an anchorage-independent manner by forming colonies in agarose medium. There is even a positive correlation between the cloning efficiency of tumor cells in soft agar and the histological involvement and invasiveness of the tumor in specimens taken from different carcinomas (19). Consequently, we determined the ability of MiaPaCa-2 and Panc-1 cells to form colonies in semisolid medium in the presence or absence of rapamycin. As shown in Fig. 3 B, rapamycin markedly inhibited the formation of colonies by MiaPaCa-2 and Panc-1 cells in a concentration-dependent manner. Half-maximal and maximal effects were achieved at 6 and 20 ng/ml rapamycin in both cell lines.

Progression from G₁ to the S phase of the cell cycle is regulated by the expression of cyclins D and E, which modulate the activities of the cyclin-dependent kinases. Rapamycin can block G₁ to S-phase cell cycle progression in a number of cell types by blocking growth factor-stimulated elimination of the cyclin-dependent kinase inhibitor p27^KIP1 (20, 21). We have recently shown that rapamycin strikingly reduces bombesin-induced expression of cyclins D1, D3, and E in Swiss 3T3 fibroblasts (22). These effects of rapamycin may be a consequence of the inhibition of p70^s6K or may be due to a distinct pathway(s) regulated by the FRAP-p70^s6K pathway. To further explore the nature of the rapamycin-sensitive and -insensitive mechanisms regulating the basal growth of human pancreatic cancer cells, we determined the expression of cyclin D1, cyclin E, and p27^KIP1 and correlated these with the activation of p70^s6K. As shown in Fig. 4, treatment of cells with rapamycin leads to a sustained dephosphorylation of p70^s6K for up to 36 h. Similar results were obtained for 4E-BP1 phosphorylation (data not shown). Rapamycin treatment of MiaPaCa-2 cells at a concentration that inhibited p70^s6K− and 4E-BP1 phosphorylation as well as basal cell growth resulted in a marked and sustained reduction in the expression of cyclin D1, compared to the levels observed in cells treated with vehicle alone. Reduced cyclin D1 expression was first visible after 6 h of incubation and was still detectable after 36 h of incubation with rapamycin. In contrast, expression of cyclin E and levels of p27^KIP1 expression did not change in response to rapamycin treatment (Fig. 4). Similar data were obtained in Panc-1 cells (data not shown). These results suggest that the inhibitory effect of rapamycin on basal pancreatic cancer cell growth is mediated by inhibition of cyclin D1 expression rather than by inhibition of p27^KIP1 elimination.

This is the first report demonstrating that the FRAP-p70^s6K signaling pathway is constitutively active in serum-starved human pancreatic cancer cells. The two major downstream targets of FRAP kinase, p70^s6K and 4E-BP1, were found to be constitutively phosphorylated in pancreatic cancer cell lines as well as a human pancreatic cancer tissue sample. Activation of both could be prevented by treating cells with the selective FRAP inhibitor rapamycin. The effect of rapamycin on constitutive p70^s6K− and 4E-BP1 phosphorylation was indeed selective, because rapamycin did not prevent basal or TGF-α-induced ERK2 activation and hence tyrosine kinase activity of the epidermal growth factor receptor, which is required for the activation of ERK2 by TGF-α. Interestingly, rapamycin substantially inhibited anchorage-dependent and -independent growth of MiaPaCa-2 and Panc-1 cells. p70^s6K mediates 5′-terminal oligopyrimidine tract mRNA translation (11), and 4E-BP1 is a major regulator of eIF-4E mediated 5′ cap mRNA translation (12). Thus, these mechanisms are likely to sustain protein synthesis and growth in human pancreatic cancer cells. The precise contribution of the two FRAP targets to pancreatic cancer cell growth is difficult to determine. Recent data suggest that p70^s6K and 4EB-P1 are regulated by FRAP in a parallel rather than sequential manner (14). Furthermore, 4E-BP1 obviously plays an independent role in the rapamycin-sensitive regulation of cell growth: rapamycin can still prevent 4E-BP1 phosphorylation and cell growth in p70^s6K−/− cells (15). The antiproliferative effect of rapamycin in p70^s6K−/− cells could possibly be due to the inhibition of a novel, rapamycin-sensitive p70^s6K isoform, which has recently been described in these cells (23). However, dephosphorylated 4E-BP1 has clearly been demonstrated to inhibit cell growth (13). Thus, constitutively phosphorylated p70^s6K and 4EB-P1 are likely to contribute to both anchorage-dependent and -independent growth of human pancreatic cancer cells. Rapamycin also blocks constitutive p70^s6K phosphorylation and proliferation in classical small cell lung cancer cells (10). Therefore, if cancer cells exhibit constitutive activation of FRAP/p70^s6K, this signaling pathway is likely to play a role in the autonomous growth of these tumors.

Because p70^s6K regulates progression from G₁ to the S phase of the cell cycle, we reasoned that the molecular mechanism underlying growth inhibition by rapamycin could be related, at least in part, to the cell cycle machinery.

Indeed, the inhibitory effect of rapamycin on pancreatic cancer cell growth was associated with a reduction in cyclin D1 expression. This was not due to a general inhibition of protein synthesis by rapamycin because we could not detect any change in the expression of cyclin E or p27^KIP1. It is presently not known whether the effects of rapamycin on p70^s6K phosphorylation/activation, 4E-BP1 phosphorylation, and cyclin D1 expression in human pancreatic cancer cells are mediated by a linear, split, or parallel pathway(s). However, because cyclin D1 expression is rapamycin-sensitive in human pancreatic cancer cells, its regulation is likely to involve a FRAP-dependent pathway including the activation of p70^s6K and the phosphorylation of 4E-BP1. This is an important observation because regulation of cyclin D1 expression can also be rapamycin-insensitive and hence independent of FRAP (24). Recently, Hashemolhosseini et al. (25) reported that inhibition of cyclin D1 expression by rapamycin in serum-stimulated NIH 3T3 cells is due mainly to the effects of rapamycin on cyclin D1 mRNA and protein stability. However, the upstream, rapamycin-sensitive signaling pathways involved in this process remain to be established.

Cyclins and cyclin-related genes are important regulators of cell cycle progression and are amplified and/or overexpressed in a major fraction of human tumors at a relatively early stage in the carcinogenic process (26, 27). In particular, cyclin D1 is overexpressed in pancreatic cancer, and its accumulation in pancreatic cancer tissues is associated with poor prognosis (28). Moreover, inhibition of cyclin D1 in Panc-1 cells using antisense oligonucleotides is associated with loss of tumorigenicity, decreased growth, and enhanced sensitivity to chemotherapeutic agents (29). Thus, rapamycin indirectly targets a key tumor promoter in pancreatic cancer. In summary, our data strongly suggest that inhibitors of the FRAP-p70^s6K signaling pathway constitute novel antiproliferative agents for human pancreatic cancer.

We thank Jo van Lint for the anti-ERK2 antibody and Michael J. Seckl (Charing Cross Hospital, London, United Kingdom) for critical reading of the manuscript.