Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Activation of the ras/raf signal-transduction pathway has been shown to contribute to transformation and tumorigenicity in many types of cancers (1). However, small-cell lung cancer (SCLC) cell lines, and SCLC in patients, rarely have ras mutations (5). In these cancers, ras mutations may not confer a growth advantage. Instead, in SCLC cell lines, introduction of an activated ras gene resulted in cell differentiation accompanied by slower cell growth; this has also been seen in pheochromocytoma and medullary thyroid carcinoma (MTC) cell lines (6). Ras- induced signaling often involves activation of the c-raf-1 gene, a cytosolic serine/threonine protein kinase. Activated raf-1 phosphorylates mitogen-activated protein kinase kinase (MEK), which activates downstream mitogen-activated protein kinases (MAPKs) (1). Several reports have shown that direct activation of raf-1, with resultant MAPK activation, can result in growth arrest and cell differentiation (9). We have recently shown that activation of raf-1 can result in growth arrest in two retinoblastoma gene (Rb)-negative SCLC cell lines, NCI-H209 and NCI-H510, with induction of the cyclin-dependent kinase (cdk) inhibitor p27^KIP1 (12). In the study reported here we examined the effect of raf-1 activation in DMS53 cells. The DMS53 SCLC cell line is unusual because, in contrast to almost all SCLC patient specimens and cell lines, it retains a functional retinoblastoma susceptibility protein (pRb), as well as expressing a mature neuroendocrine (NE) phenotype and secreting the polypeptide hormone calcitonin (CT) (13). Previously, we showed that transfection of a viral Harvey-ras gene into the DMS53 SCLC cell line resulted in cell differentiation (8). In the study reported here we showed that activation of Raf-1:ER can cause DMS53 cells to cease proliferation. This growth arrest is accompanied by induction of cdk inhibitor p16^INK4 and by reduced expression of all three Rb-family proteins. Thus, in different SCLC cell lines, raf-1 activation appears to induce growth arrest via different pathways.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture and Cell Lines

DMS53 SCLC cells were cultured in RPMI-1640 medium without phenol red, 9% fetal bovine serum (FBS) (Sigma Chemical Co., St. Louis, MO), 100 U/ml penicillin, and 100 µg/ml streptomycin sulfate (GIBCO, Grand Island, NY) in a humidified atmosphere of 5% CO₂ at 37°C. DMS53 cells were infected with equal volumes of retroviral supernatant from PA317 producer cells transfected with the retroviral vector pLNCRaf-1:ER, containing a Raf-1:ER fusion construct (16). This construct is an estrogen receptor-raf fusion molecule, and contains the ligand-binding domain of the estrogen receptor fused to the raf kinase domain of c-raf-1. Infection of DMS53 cells was augmented by 2 µg/ml polybrene (Sigma) in the medium. After 48 h the medium was replaced by selection medium containing 0.5 mg/ml of G418. Pooled cultures of G418-resistant DMS53 cells were grown and were analyzed by Northern blotting for expression of the Raf-1:ER construct. DMS53:Raf-1:ER cells were treated with 1 µM -estradiol to activate the Raf-1:ER fusion molecule. Parental DMS53 cells, parental cells exposed to 1 µM -estradiol, and untreated DMS53:Raf-1:ER cells were used as controls. No effects of Raf-1:ER transduction without estradiol administration were observed.

Soft-Agar Cloning Assay

Soft-agar cloning assays were performed in 35-mm dishes over a bottom layer of 0.8% low-melting agarose in growth medium. In these assays, 1.0 × 10⁴ cells were plated in growth media containing 0.4% (wt/vol) agarose in the presence or absence of 1 µM -estradiol. After 3 wk of incubation in a humidified atmosphere containing 5% CO₂ at 37°C, colonies with more than 30 cells were scored and the percent cloning efficiency was calculated.

Cell Cycle Analysis

Cells were pulse labeled with 1 µM bromodeoxyuridine (BrdU; Sigma) for 40 min at 37°C, and were then fixed in 70% ethanol/phosphate buffered saline (PBS) and their nuclei prepared by denaturation. Extracted nuclei were stained with fluorescein isothiocyanate (FITC)-labeled anti-BrdU antibodies (Becton Dickinson, San Jose, CA) and propidium iodide. Flow cytometric analysis was performed with an EPICS 752 flow cytometer (Coulter Electronics, Hialeah, FL), with a gate that selects single nuclei within a normal size range. The cell cycle parameters from 10,000 gated nuclei were determined with multicycle software (Phoenix Flow Systems, San Diego, CA).

Protein Analysis

Cells were lysed in PBS, 1% Triton X-100, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 10 µg/ml pepstatin, and 1.0 mM phenylmethyl sulfonyl fluoride (PMSF). Protein concentrations were determined with the Bradford protein assay (Bio-Rad Laboratories, Hercules, CA). After protein concentrations were determined, 50-100 µg of proteins were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to Immobilon-P polyvinylidene difluoride (PVDF) membrane (Millipore, Bedford, MA). Membranes were probed with appropriate dilutions of primary anti-Raf-1 (C-12), anti-pRb (IF8), anti-p130 (C-20), anti-p107 (C-18), anti-cyclin E (HE12), anti-cdk2 (M2), anti-cdk4 (C-22), anti-cdk6 (C-21), E2F-1 (C-20), and anti-p34^cdc2 (C-17) antibodies from Santa Cruz Biotechnology (Santa Cruz, CA); anti-cyclin D1 (17A6-4), anti-cyclin D2 (34B1-3) antibodies from Oncogene Science (Cambridge, MA); anti-p16^INK4 (DCS-50.2) antibody from Lab Vision Corporation (Fremont, CA); anti-MAPK (phospho-specific MAPK antibody from New England Biolabs, Beverly, MA); and anti-cyclin A antibody, which was obtained by immunizing rabbits with the bacterially expressed glutathione-S-transferase (GST)- cyclin A fusion protein (17). Immunoreactive protein complexes were detected through enhanced chemiluminescence (Amersham, Arlington Heights, IL).

Immunoprecipitations

Whole-cell lysates were prepared in 50 mM Tris-HCl, pH 7.5; 137 mM NaCl; 5 mM MgCl₂; 1% Triton X-100; 50 mM Na -glycerophosphate; 2 mM ethylenediamine tetraacetic acid (EDTA); 10 mM ethylene glycol-bis-(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA); 1 mM dithiothreitol (DTT); 1 mM Na₃VO₄; 10 µg/ml aprotinin; 10 µg/ml leupeptin; 10 µg/ml pepstatin; and 1 mM PMSF. From 300 to 500 µg of whole-cell lysates were incubated for periods ranging from 2 h to overnight at 4°C with 1 µg/ ml primary antibody. The immune complexes were immunoprecipitated for 1 h with protein A-Sepharose beads (Pharmacia, Piscataway, NJ), and the immune complexes bound to the beads were washed three times with the same lysis buffer and twice with buffer containing 10 mM Tris-HCl, pH 7.5. The beads were boiled with 25 µl of 2× Laemmli buffer, and the complexes resolved on SDS- PAGE gels and transferred to Immobilon-P PVDF membranes (Millipore). The membranes were Western blotted with the appropriate antibodies, and immunoreactive protein complexes were detected through enhanced chemiluminescence (Amersham).

In Vitro Kinase Assays

Whole-cell lysates were prepared as described earlier. Aliquots of 100 µg of whole-cell lysates were incubated for 2 h with 1 µg/ml anti-cdk2, anti-cdk4, or anti-MAPK antibody. The immune complexes were immunoprecipitated for 1 h with protein A-Sepharose beads, and the immune complexes bound to the beads were washed twice with the lysis buffer described earlier and three times with kinase buffer. Kinase assays were done with histone H1 for cdk2 and p130-associated kinase activity; GST-Rb2/p130 (18) for cdk2, GST-pRb (Santa Cruz Biotechnology) for cdk4, and myelin basic protein (MBP) (Sigma) for MAPK. The immunoprecipitates and substrates were incubated in a volume of 25 µl containing 50 mM Tris-HCl, pH 7.5; 10 mM MgCl₂; 1 mM EGTA; 1 mM DTT; and 0.125 µCi/µl [-³²P]adenosine triphosphate ([-³²P]ATP) at 30°C for 20 min. The reactions were stopped by the addition of 25 µl of 2× Laemmli buffer, and the reaction mixtures were boiled for 3 min and resolved on 12.5% SDS-PAGE gels and transferred to Immobilon-P PVDF membranes. Phosphorylated proteins were visualized by autoradiography and quantitated on a Phosphorimager (Molecular Dynamics, Sunnyvale, CA).

Northern Blotting

Total RNA was extracted with an acid phenol-guanidinium isothiocyanate method (19). Poly(A)⁺ RNA was isolated by oligo deoxythymidine (oligo[dT])-cellulose chromatography. Poly(A)⁺ RNA (5 µg/lane) was separated on 1.2% agarose/2.2 M formaldehyde denaturing gels and transferred to a positively charged nylon membrane (Zeta-Probe; Bio-Rad). Probes used in Northern blot analysis were a 0.6-kb EcoR1 fragment of gastrin releasing peptide (GRP); a 0.4-kb EcoR1-Nco1 fragment of Ret complementary DNA (cDNA), and a 2.2 kb BamH1 fragment of human -actin gene. These probes were labeled with [-³²P]deoxycytidine triphosphate ([-³²P]dCTP) (Dupont-New England Nuclear, Boston, MA) by random primer labeling (Boehringer Mannheim, Indianapolis, IN). Hybridizations were done with radiolabeled probe (1 × 10⁶ cpm/ml) at 42°C for 16 to 18 h and the products were rinsed twice at room temperature and washed once for 30 min at 65°C with 1× standard saline citrate (SSC) and 1% SDS. Membranes were then exposed to X-ray film (X-OMAT; Kodak, Rochester, NY) at 80°C with intensifying screens. Autoradiograms were quantitated on a Phosphorimager.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of Raf-1:ER Activation on Morphology of DMS53 Cells

DMS53 cells and DMS53 cells transfected with Raf-1:ER (DMS53:Raf-1:ER) were morphologically identical and grew as tightly adherent, flat cells (14, 15). Because -estradiol was used to activate the Raf-1:ER fusion molecule, DMS53 cells, DMS53 cells exposed to estradiol and DMS53:Raf-1:ER without estradiol were used as controls for the study. In DMS53:Raf-1:ER cells, activation of Raf-1:ER by estradiol resulted in the phosphorylation of raf-1, generation of MAPKs, and activation of MAPKs (Figure 1a). Activation of MAPK was sustained for at least for a week after raf activation (data not shown). MAPK was not phosphorylated in any of the control cells. Estradiol treatment did not cause any morphologic changes in DMS53 parental cells (Figure 1c). However, when the Raf-1:ER protein was activated by estradiol, DMS53:Raf-1:ER cells became rounded within 48 h (Figure 1e). Proliferation of DMS53:Raf-1:ER cells was markedly reduced after Raf-1:ER activation (Figure 2a). We also measured the soft-agarose cloning efficiency of DMS53 and DMS53:Raf-1:ER cells in the presence or absence of estradiol. The soft-agarose cloning efficiency of DMS53:Raf-1:ER cells in the presence of estradiol was reduced by 90% as compared with that of control cells (Figure 2b). There was no increase in the proportion of apoptotic cells after Raf-1:ER activation, suggesting that the raf-induced growth suppression might have been due to growth arrest or differentiation of cells. Because DMS53:Raf-1:ER cells proliferated poorly after raf activation, we examined the cell cycle progression from the G₁ through the S phase by measuring BrdU incorporation. Cell cycle analysis using BrdU labeling showed that Raf-1:ER activation reduced the proportion of cells in the S phase to 39.4% of parental DMS53 cells, with an associated accumulation of cells in the G₁ and G₂ phases (Figures 2c and 2d). No effect of the added estradiol on the cell cycle was observed in parental DMS53 cells (data not shown). Thus, these data suggest that raf pathway activation reduced the growth and cell cycle progression of DMS53 cells.

View larger version (68K): [in this window] [in a new window]   Figure 1.   Activation of Raf-1:ER caused morphologic changes, accompanied by phosphorylation of raf-1 and MAP kinase, in DMS53 cells. (a) Lysates from parental DMS53 cells, DMS53 cells treated with estradiol (+E) for 48 h, and Raf-1:ER-transduced DMS53 cells either untreated or treated with estradiol for 48 h were immunoblotted with antibodies to phosphorylated MAPK and raf-1. The same cell lysates were immunoprecipitated with anti-MAPK antibody, and kinase activity was measured with myelin basic protein as a substrate, as described in MATERIALS AND METHODS. (b-e) Morphologic effects of activated Raf-1:ER in DMS53 SCLC cells. Cells were grown in the presence of 1 µM estradiol (+E) for 6 d. Photographs of parental cells and transduced cells were taken with a phase-contrast light microscope at a magnification of ×100. (b) DMS53 cells, (c) DMS53 cells +E, (d) Raf-1:ER-transduced DMS53 cells, (e) Raf-1:ER-transduced DMS53 cells +E. No morphologic changes were observed in parental DMS53 cells grown in the absence or presence of estradiol. Activation of Raf-1:ER in DMS53 cells resulted in rounded cells.

View larger version (44K): [in this window] [in a new window]   Figure 2.   Activated Raf-1:ER inhibited growth and soft-agar cloning of DMS53 cells. (a) Proliferation of DMS53:Raf-1:ER cells after raf activation. DMS53 and DMS53:Raf-1:ER cells were seeded in six-well plates and treated with estradiol (E). Cells were trypsinized and counted at indicated time points. Each point represents a mean of triplicate wells. (b) Soft-agar cloning of: (1) DMS53 cells; (2) DMS53 cells +E; (3) Raf-1:ER-transduced DMS53 cells; (4) Raf-1:ER-transduced DMS53 cells +E. Data represent the mean of three independent experiments with the indicated standard errors. Colonies of more than 50 cells were scored and the percent of cloning efficiency was calculated and shown on the y axis. Cloning efficiency for estradiol-treated control cells was considered as 100%. (c, d) Cell cycle distribution of DMS53 cells and their Raf-1:ER-transduced cells. BrdU labeling of DMS53 (c) and DMS53:Raf-1:ER (d) cells in the presence of 1 µM estradiol (+E).

Activation of Raf-1:ER Reduced cdk4 and cdk2 Kinase Activity

The mammalian cell cycle is regulated by cyclins and their associated cdks (20, 21). The Rb-family proteins (pRb, p107, and p130) are phosphorylated by cyclin/cdk complexes, and their phosphorylation is necessary for G₁/S transition (22, 23). In cells with functional Rb-family molecules, members of the cyclin D family complex with either cdk4 or cdk6 and phosphorylate pRb in the early G₁ phase of the cell cycle. The cyclin D/cdk4 complex phosphorylates p107 in mid-G₁. In late G₁, the cyclin E/cdk2 complex associates with p130 and controls progression beyond the restriction point. Cyclin A, which is required for transit through the S phase of the cell cycle, complexes with cdk2 in the S phase and with p34^cdc2 near the G₂/M portion of the cell cycle. For cells to pass from G₂ to M in the cell cycle, p34^cdc2/cyclin B activity is needed (24, 25). We determined whether Raf-1:ER activation altered any of the cyclin cdk complexes in DMS53 cells. Activation of Raf-1:ER did not have any effect on the expression of cyclin D family members, cyclin E, or cyclin B (data not shown), or on cdk2, cdk4, or cdk6 protein levels (Figure 3a). However, the Rb-family proteins pRb, p130, p107, and E2F-1 were reduced after Raf-1:ER activation in DMS53 cells (Figure 3b).

View larger version (64K): [in this window] [in a new window]   Figure 3.   Effect of Raf-1:ER activation on cell cycle proteins. (a) Parental DMS53 cells and DMS53:Raf-1:ER cells were grown in the presence (+E) or absence of estradiol for 3 d, and cell lysates were made as described in MATERIALS AND METHODS. Immunoblot analysis showed that raf activation reduced pRb, p130, p107, E2F-1, cyclin A, and cdc2 proteins. The cdk inhibitor p16INK4 was induced after Raf-1:ER activation. (b) Effects of activated Raf-1:ER on cdk2 and cdk4 activity. Activation of Raf-1:ER led to reduced cdk2 activity in DMS53 cells. cdk4 kinase activity was measured with GST-Rb as substrate. Phosphorylation of pRb by cdk4 diminished by 61% in DMS53 cells after Raf-1:ER activation. The phosphorylation of GST-Rb2/p130 was also determined in cdk2 immunocomplexes. Phosphorylation of p130 was reduced to 42% of that in parental cells after Raf-1:ER activation.

Even though cdk4 and cdk2 protein levels were not reduced, their kinase activities were reduced after Raf-1:ER activation in DMS53 cells. In vitro kinase assays, using histone H1 and pRb as substrates for cdk2 and cdk4, respectively, showed that cdk2 activity was reduced to 39% of control parental levels and that cdk4 activity was reduced to 68% of control parental levels after Raf-1:ER activation in DMS53 cells. Reduced Rb phosphorylation could result in a cell cycle block in mid G₁. Because cyclin E/cdk2 complexes with another Rb-family protein, Rb2/ p130, in late G₁, and its activity is necessary for cells to exit from G₁ to S (26), we measured the phosphorylation of Rb2/p130 by cyclin E/cdk2. Cyclin E/cdk2-mediated phosphorylation of Rb2/p130 was reduced to 42% of the parental levels after Raf-1:ER activation (Figure 3b). Rb2/ p130, like pRb, has growth-suppressive properties when ectopically expressed, and is phosphorylated by cdks at G₁/S transition (18). As consistent with the reduced fraction of cells in the S phase, Raf-1:ER activation in DMS53 cells reduced the expression of cyclin A and p34^cdc2 protein (Figure 3b). Our data suggest that raf activation may block cell cycle progression by reducing both the expression of Rb-family proteins and the phosphorylation of Rb and Rb2/p130 by cdks in DMS53 cells.

Activation of Raf-1:ER Induced cdk Inhibitor p16^INK4

The activity of cdks and Rb function are controlled by cdk inhibitors, and decreased cdk activity can result from the induction of cdk inhibitors (29). In Rb-negative cells, cell cycle arrest can be induced by cdk inhibitors of the CIP/ KIP class (p21^WAF1/CIP1, p27^KIP1, and p57^KIP2), which inhibit cdk2 (29). In Rb-positive cells, cell cycle inhibitors of the INK class, such as p16^INK4, can induce cell cycle arrest, by inhibiting cdk4 and cdk6 activity, which are necessary to phosphorylate Rb prior to entry into the S phase (30, 31). We have previously shown that raf-1 induces p27^KIP1 in growth arrest of cells of the Rb-negative SCLC cell lines H209 and H510. Therefore, we investigated whether Raf-1:ER activation had any influence on the cdk inhibitors p21^WAF1/CIP1, p27^KIP1, p57^KIP2, or p16^INK4 in Rb-positive DMS53 cells. Western blot analysis showed an increase in p16^INK4 protein expression after Raf-1:ER activation (Figure 3a). The induction of p16^INK4 protein was not associated with increased levels of p16^INK4 messenger RNA (mRNA) (data not shown), suggesting that the increase in p16^INK4 is due to increases in p16 protein synthesis or stability. Raf-1:ER activation in DMS53 cells did not induce p21^WAF1/CIP1, p27^KIP1, or p57^KIP2. These data suggest that Raf-1:ER activation induces cell cycle arrest by inducing p16^INK4, inhibitor of the cyclin D-pRb cell cycle control pathway in these SCLC cells.

Effect of Raf-1:ER Activation on Neuroendocrine Markers

We have previously shown that introduction of an activated Harvey-ras oncogene can influence the NE phenotypic features of SCLC cells including DMS53 (8). We investigated whether Raf-1:ER activation modifies the expression of these NE features in DMS53 cells. GRP, a 27-amino-acid human equivalent of the amphibian tetradecapeptide bombesin, is an important NE growth factor for the proliferation of normal bronchial epithelium and for tumors associated with this tissue (32). Several studies have shown that GRP can function as mitogen for normal bronchial epithelial cells and SCLC cell lines (33). In our study, Northern blot analysis showed that GRP mRNA was markedly decreased (by 6.4-fold) after raf activation in DMS53:Raf-1:ER cells (Figure 4). This decrease in GRP mRNA was reflected in lower levels of biologically active GRP protein in DMS53:Raf-1:ER cells; Raf-1:ER activation in DMS53 cells led to decreased (by 8.5-fold) levels of immunoreactive GRP protein as compared with that of the parental DMS53 cells (Table 1). In contrast, expression of another endocrine marker, calcitonin, was not altered by Raf-1:ER activation (data not shown). In MTC cells, raf-induced differentiation was accompanied by silencing of the ret tyrosine kinase (9). Similarly, in DMS53:Raf-1:ER cells, ret mRNA was markedly decreased within 48 h after Raf-1:ER activation (Figure 4). The ret gene produces mRNA transcripts of four sizes (7.0, 6.0, 4.5, and 3.9 kb), and quantities of all four mRNA transcripts were decreased following raf activation in DMS53:Raf-1:ER cells. These data suggest that the ras/raf/MAPK signal transduction pathway can mediate NE changes in SCLC cells.

View larger version (44K): [in this window] [in a new window]   Figure 4.   Northern blot analysis of GRP and Ret expression in DMS53 cells after Raf-1:ER activation. Poly(A+) RNA was made from both the control cells and activated Raf-1:ER DMS53 cells, and 5 µg of this RNA was electrophoresed through a 1.2% formaldehyde gel and transferred to a nylon membrane. The membrane was hybridized to a 32P-labeled probe and was then stripped and hybridized with another probe. Expression of GRP and Ret were reduced after Raf-1:ER activation in DMS53 cells. Expression of -actin was used as a loading control.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Activation of the ras/raf/MAPK pathway is often associated with cellular transformation and proliferation in primary cells and with tumor progression in established cancers (5). In contrast to those findings, our data suggest that raf-induced p16^INK4 suppresses the growth of DMS53 cells by interacting with G₁ cyclin/cdks and the Rb pathway. Activation of the ras/raf/MAPK pathway has been shown to result in growth arrest in several cell types (9- 12). Our findings are reminiscent of possible differentiation effects of ras, raf, and MAPK, as has been seen previously in NE cells including PC12 murine pheochromocytoma cells (6), SCLC cells (8, 12), human MTC cells (7, 9), and hippocampal neuronal cells (10). Recently, Serrano and colleagues (34) showed that ras-induced G₁ arrest and cell senescence in primary human or rodent cells was accompanied by accumulation of p16 and p53. Similarly, Woods and coworkers (11) and Sewing and associates (35) showed that depending on the level of raf kinase activity, mouse fibroblasts can progress or arrest in the cell cycle, and that the raf-induced cell cycle arrest was associated with induction of the cdk inhibitor p21^WAF1/CIP1. Activation of raf also caused growth arrest in primary rat Schwann cells through induction of the cdk inhibitor p21 (36). Ras-regulated hypophosphorylation of pRb and G₁ arrest accompany growth inhibition and neuronal differentiation in PC12 cells (37). Thus, the ras/raf/MAPK pathway can promote growth arrest by inducing the production of several cdk inhibitors, including p16^INK4, p21^WAF1, and p27^KIP1, in a cell-specific manner. The mechanisms by which the ras/raf/MAPK pathway induces expression of cdk inhibitors are not fully characterized for any of these cell types. A single mechanism that might apply to all of these cells, such as induction of a transcription factor for all cdk inhibitors, probably does not apply; p16^INK4 and p27^KIP1 appear to be regulated at the level of protein synthesis or stability (the present study and Reference 12), and p21^WAF1 may be regulated transcriptionally (11, 35) or at the level of mRNA stability (our unpublished findings). One can speculate that MAPK phosphorylation may activate factors specific for each of these cell types, which alter the transcription, mRNA stability, or protein stability of cdk inhibitors. The identification of these factors will provide important links in the mechanism of ras/raf/MAPK-mediated growth arrest in SCLC and other cell types.