The majority of ovarian cancer patients are treated with platinum-based chemotherapy, but the emergence of resistance to such chemotherapy severely limits its overall effectiveness. We have shown that development of resistance to this treatment can modify cell signaling responses in a model system wherein cisplatin treatment has altered cell responsiveness to ligands of the erbB receptor family. A cisplatin-resistant ovarian carcinoma cell line PE01CDDP was derived from the parent PE01 line by exposure to increasing concentrations of cisplatin, eventually obtaining a 20-fold level of resistance. Whereas PE01 cells were growth stimulated by the erbB receptor-activating ligands, such as transforming growth factor-α (TGFα), NRG1α, and NRG1β, the PE01CDDP line was growth inhibited by TGFα and NRG1β but unaffected by NRG1α. TGFα increased apoptosis in PE01CDDP cells but decreased apoptosis in PE01 cells. Differences in extracellular signal-regulated kinase and phosphatidylinositol 3-kinase signaling were also found, which may be implicated in the altered cell response to ligands. Microarray analysis revealed 51 genes whose mRNA increased by at least 2-fold in PE01CDDP cells relative to PE01 (including FRA1, ETV4, MCM2, AXL, MT3, TRAP1, and FANCG), whereas 36 genes (including IGFBP3, TRAM1, and KRT4 and KRT19) decreased by a similar amount. Differential display reverse transcriptase-PCR identified altered mRNA expression for TCP1, SLP1, proliferating cell nuclear antigen, and ZXDA. Small interfering RNA inhibition of FRA1, TCP1, and MCM2 expression was associated with reduced growth and FRA1 inhibition with enhanced cisplatin sensitivity. Altered expression of these genes by cytotoxic exposure may provide survival advantages to cells including deregulation of signaling pathways, which may be critical in the development of drug resistance.

Platinum-based chemotherapy is a standard first-line treatment for ovarian cancer (1). In many instances, however, resistance to chemotherapy develops, limiting its use. The changes in cellular signaling after drug treatment are currently attracting a great deal of interest and this is of particular relevance when cytotoxic therapy is combined with signaling inhibitors. Among the key regulators of growth in this disease is the group 1 receptor tyrosine kinase family of erbB receptors. Cisplatin is known to interact with the epidermal growth factor receptor (EGFR) pathway (2) either promoting (3, 4) or inhibiting apoptosis and cell death (5).

The erbB receptor family consists of the EGFR (erbB-1 and HER-1), erbB-2 (HER-2), erbB-3 (HER-3), and erbB-4 (HER-4; refs. 6, 7). Ligands of the EGF family, including transforming growth factor-α (TGFα), activate the EGFR, whereas members of the neuregulin (NRG)/heregulin family activate erbB-3 and erbB-4 (6). ErbB-2 is activated via interaction with other ligand-stimulated family members. Ligand binding to EGFR promotes either dimerization with another EGFR (homodimerization) or binding to another erbB receptor family member (heterodimerization) in which case, erbB2 is the preferred option (7). The erbB receptor family are widely reported to have key roles in regulating a network of signaling pathways, including the Ras/Raf/mitogen-activated protein kinase kinase (MEK)/extracellular signal-regulated kinase (ERK) pathway (8), the phosphatidylinositol 3-kinase (PI3K)/Akt pathway (9), and the PLCγ cascades (10). These are implicated in the translation of ligand-mediated signaling at the cell membrane to the nucleus where induction of gene expression provides the basis for the cell response. In response to external stimuli, cells may be directed to grow, differentiate, migrate, or apoptose; outcomes which will ultimately determine the rate and the manner in which the disease will progress and also how it will respond to treatment.

The erbB receptors and their ligands play key roles in the growth and progression of several cancers including ovarian cancer (6, 11). Overexpression of both EGFR and erbB2 separately have been linked to poor survival in ovarian cancer (1215) and EGFR activators are mitogens in ovarian cancer cell lines in culture (1618). Several EGFR-targeted inhibitors are currently being considered for evaluation in advanced ovarian cancer and the effect of prior treatment with cisplatin on the signaling pathways may be important in influencing tumor response. These agents include the tyrosine kinase inhibitors gefitinib (“Iressa”; ZD 1839; refs. 19, 20) and erlotinib (“Tarceva”; OSI-774; ref. 21).

To understand further the changes in EGF signaling after the onset of cisplatin resistance, we have developed an in vitro model wherein the parent cell line PE01 has been made resistant to cisplatin by exposure to increasing levels of drug. Coupled with this change in drug sensitivity is a modified responsiveness to ligands of the erbB receptor family. In this report, we have identified a number of changes that might be significant in explaining the altered phenotype.

Cell lines and culture conditions. The PE01 cell line was derived from a poorly differentiated human ovarian adenocarcinoma (22). The PE01CDDP variant was established by in vitro exposure of the PE01 line to cisplatin commencing at 25 nmol/L cisplatin and increasing to 1 μmol/L over a period of 4 months. The resistant cell line thus obtained was then passaged in the absence of cisplatin for a further 4 months to ensure that the resistance remained stable. Both cell lines were maintained in RPMI 1640 (Life Technologies, Paisley, Scotland) containing 10% heat-inactivated FCS, penicillin (100 units/mL), and streptomycin (100 μg/mL). Cells were maintained routinely at 37°C in a humidified atmosphere of 5% CO2 in air.

Growth assays. Cells in log-phase growth were seeded in 24-well plates (Falcon, Franklin Lakes, NJ) at a density of 2.5 × 104 cells per well in quadruplicate in RPMI 1640 containing 10% heat-inactivated FCS, penicillin (100 units/mL), and streptomycin (100 μg/mL). After 24 hours, medium was replaced by phenol red-free RPMI 1640 containing 5% double charcoal-stripped FCS (DCS-FCS), penicillin (100 units/mL), streptomycin (100 μg/mL), and glutamine (2 mmol/L). After a further 24 hours, medium was removed and replenished. Ligand additions were made at this time point, designated day 0, and medium plus additions was replaced on day 2. In certain experiments, inhibitors were also added alongside ligands. Cells were harvested on days 0, 2, and 5 and counted using a cell counter (Coulter Electronics, Ltd., Luton, England). TGFα and EGF were obtained from Boehringer Mannheim (Indianapolis, IN), NRG1α from Sigma Ltd. (St. Louis, MO), and NRG1β from Neomarkers (Fremont, CA). Gefitinib was a gift from AstraZeneca (Macclesfield, United Kingdom) and LY294002 (#40202) and PD98059 (#513000) were obtained from Calbiochem (La Jolla, CA).

Clonogenic assay. Cells in logarithmic growth were plated in 35-mm 6-well plates (Costar, Cambridge, MA) at densities that produced 100 to 200 colonies in untreated wells (2 × 103/mL for PE01 and 103/mL for PE01CDDP cells). The medium described above was used but with the addition of 1 mmol/L sodium pyruvate to enhance plating at low cell densities. After 48 hours, drug was added at appropriate concentrations to the wells and left for 24 hours. Medium was changed every 2 to 3 days for 12 to 14 days until colonies (>50 cells) were obtained in the untreated control wells. The surviving fraction in each of the drug-treated samples was then counted using an inverted microscope and the colony counts obtained for drug-treated wells expressed as a percentage of the counts in untreated control wells. Dose-response curves were plotted for each drug-cell line combination and IC50 values extrapolated from summed data obtained from at least three experiments in every case. Cytotoxic drugs were obtained from the following sources: Carboplatin (CBDCA) and iproplatin (CHIP) were gifts from Dr. Mervyn Jones (Institute of Cancer Research, Sutton, United Kingdom); cisplatin and JM40 were obtained from Bristol Myers Pharmaceuticals; chlorambucil and melphalan from Sigma; doxorubicin from Farmitalia Carlo Erba Ltd.; and prednimustine from Aktiebolaget Leo (Helsingborg, Sweden).

Apoptosis assay. PE01 and PE01CDDP cells were grown to 80% confluence in the presence of RPMI 1640 containing 10% FCS. Following an overnight incubation in phenol red–free RPMI 1640 containing 5% DCS-FCS, the medium was replenished before the addition of TGFα. Apoptosis was measured using the TACS Annexin V-FITC kit (R&D Systems, Minneapolis, MN) following the prescribed protocol.

mRNA extraction and reverse transcriptase-PCR. Total cellular RNA was extracted from cells in log-phase growth using TRI reagent (Sigma, Poole, United Kingdom), following the standard protocol provided. Samples were treated with 20 units DNase 1 (Roche, Nutley, NJ) to remove genomic DNA contamination. RNA was then re-extracted using a phenol/chloroform protocol. Reverse transcription was done with a first-strand cDNA Synthesis kit (Roche) using the oligo dT primer provided. One microgram of RNA yielded 20 μL of cDNA of which 2 μL were used for each subsequent PCR reaction with each primer pair. PCR reactions were done in a final volume of 20 μL containing the following: 1× PCR buffer, 1.5 mmol/L MgCl2, 0.2 mmol/L deoxynucleotide triphosphate mixture, 2.5 units Taq polymerase (Cancer Research UK, Clare Hall, South Mimms, United Kingdom), 400 μmol/L each primer. The amplification reaction was carried out for 35 cycles using the following variables: denaturation, 94°C for 5 minutes; cycling, 94°C for 30 seconds, 55°C for 30 seconds, 72°C for 45 seconds (repeated for 35 cycles); extension, 72°C for 5 minutes. For semiquantitative reverse transcriptase-PCR (RT-PCR), 2 μL cDNA were amplified using the standard PCR protocol but reactions were terminated at 15, 20, or 25 cycles to fall within the linear region of PCR product synthesis. Activator protein (AP-1) primers (designed using Primer 3 software) were as follows: JUNB: TCTCTCAAGCTCGCCTCTTC, ACGTGGTTCATCTTGTGCAG; JUN: GGAGTGTCCAGAGAGCCTTG, GAAAGGCTTGCAAAAGTTCG; JUND: TTCTACTCGGGGAACAAACG, GGCGAACCAAGGATTACAAA; FOSB: GACTCAAGGGGGTGACAGAA, AAAATGTCACAGCCCCTCAC; FOS: TTTATAGTGGGCGGAAGTGG, ACGTCCTGGACAAAGGTCAC; fos-related antigen 2 (FRA2): GGAGCTGGAGGAGGAGAAGT, GGGCTCCTGTTTCACCACTA; FRA1: CCCTGCCGCCCTGTACCTTGTATC, AGACATTGGCTAGGGTGGCATCTGCA.

PCR products were visualized after electrophoresis on polyacrylamide gels and sized using a 100-bp ladder (Life Technologies). PCR reactions were also carried out on RNA which had not been reverse transcribed to check for genomic DNA contamination but these were routinely negative.

Quantitative reverse transcriptase-PCR. RNA extraction, DNase treatment, and cDNA synthesis were conducted as described above; 2.5 μL of cDNA was analyzed by real-time PCR using a LightCycler (Idaho Technology). Reactions included LC Master Mix (Biogene, Kimbolton, United Kingdom), SYBR green dye at a final concentration of 1:20,000 (1765; Biogene), magnesium chloride at 4 mmol/L, and forward and reverse primers at 0.5 μmol/L. The primer pairs used are were as follows: Fanconi anemia complementation group G (FANCG): CTGTAGCTGCCACGTTTTGA, GGTGGTGGCAGAGATTGTTT; tumor necrosis factor type 1 receptor–associated protein (TRAP1): TGCGAGATGTGGTAACGAAG, CGGTGCGTCCGTCTTATAGT; Axl oncogene (AXL): TTTCCTGAGTGAAGCGGTCT, CATCTGAGTGGGCAGGTACA; minichromosome maintenance 2 (MCM2): ACCAGGACAGAACCAGCATC, CAGGATGTCAAAGCGTGAGA; ETV4: GCAGATCCCCACTGTCCTAC, GTTCTCTGTGGTTGGGGAAA; metallothionein 3 (MT3): TGTGAGAAGTGTGCCAAGGA, GTCATTCCTCCAAGGTCAGC; T-complex protein 1 (TCP1): CCCAGGTTCTCAGAGCTCA, GGATGACACACAAAGCATCG; proliferating cell nuclear antigen (PCNA): CGGGGGAATGTTAAGAGGAT, CCAGCCACGAAAGTGAAAGT; insulin-like binding protein 3 (IGFBP3): CGTCAACGCTAGTGCCGTCAGCCG, GACCATATTCTGTCTCCCGCTTGGACT; KRT19: GGTCAGTGTGGAGGTGGATT, TCAGTAACCTCGGACCTGCT; translocating chain-associating membrane protein (TRAM): ACAGCTGGCTTACTGGCTTC, CGGGAAATGTGGAAAAGAAA; KRT4: GGCAGCAGAAGCCTCTACAA, CCACCCTTACCACTGAAGGA; ZXDA: GGACTCTTTGGCCATGAAAA, ACTGGGTTTCTCCCTCCTGT; secretory leukoprorease inhibitor 1 (SLP1): GGGAAGTGCCCAGTGACTTA, AAAGGACCTGGACCACACAG; β-actin: CTACGTCGCCCTGGACTTCGAGC, GATGGAGCCGCCGATCCACACGG.

The standard protocol used was as follows: premelt: 95°C for 10 seconds; PCR: 95°C for 0 second, ramp at 20°C/s, 55°C for 0 second, ramp at 20°C/s, 72°C for 6 seconds, ramp at 5°C/s, X°C for 0 second, ramp at 20°C/s (where X was an empirically determined temperature high enough to melt primer dimer but leave PCR product intact) followed by a melt step from 72°C to 95°C at 0.1°C/s.

Standard curves were obtained by performing reactions with predetermined amounts of target template DNA for each primer pair. Contamination of RNA by genomic DNA was excluded by doing reactions on RNA, which had not been reverse transcribed.

Determination of epidermal growth factor receptor, erbB2, and erbB3 by immunofluorescence. Expression of EGFR, erbB2, and erbB3 proteins were measured in PE01 and PE01CDDP cells by immunofluorescence using a flow cytometer. The following antibodies were used: EGFR, clone EGFR1 (Imperial Cancer Research Fund, Clare Hall, London, United Kingdom); erbB2, clone CB11 (Novocastra); and erbB3, clone RTJ1 (Novocastra). Cells were harvested by trypsinisation, washed in cold PBS containing 5% FCS, and aliquots of ∼106 cells were then incubated for 60 minutes with antibody. For erbB2 and erbB3 staining, 1% saponin (BDH, Poole, Dorset, United Kingdom) was added to the cells before antibody addition (23).

Cells were then washed in PBS/FCS and incubated with sheep-antimouse FITC (1:20) for 60 minutes and washed twice with PBS/FCS. Cells were resuspended in PBS and analyzed on the FACScan flow cytometer.

Western blotting. Western analysis was undertaken as previously described (24). The following antibodies were used: phosphotyrosine (PY20; Santa Cruz Biotechnology, Santa Cruz, CA), phospho-ERK (9101; New England Biolabs), phospho-AKTS473 (9271; New England Biolabs), ERK I/II (9102; New England Biolabs), AKT (9272; New England Biolabs), actin (CP01; Oncogene Research Products), FRA1 (N-17, sc-183; Santa Cruz Biotechnology), PCNA (NA03; Oncogene Research Products), ETV4 (sc-114; Santa Cruz Biotechnology), TRAP1 (616480; Calbiochem), and IGFPB3 (GF60; Oncogene Research Products). Antibodies were used at 1/1,000 dilution apart from PY20, which was used at 1/200.

The following inhibitors were used: gefitinib (“Iressa”; ZD1839; AstraZeneca; 1 μmol/L) is an EGFR-tyrosine kinase inhibitor and was added to cells 5 minutes before growth factor; LY294002 (440202; Calbiochem; 10 μmol/L) inhibits PI3K and was added to cells 30 minutes before growth factor; PD98059 (513000; Calbiochem; 50 μmol/L) inhibits MEK and was added 60 minutes before growth factor. For the signaling study, 1 nmol/L TGFα was added for 15 minutes.

Extracellular signal-regulated kinase activity. ERK activity was measured using a Biotrak mitogen-activated protein kinase activity assay (RPN 84; Amersham Life Science, Amersham, United Kingdom) following the recommended protocol.1

Differential display. RNA extraction, DNase treatment, and re-extraction were as described above. Forty-cycle PCR reactions were undertaken with three single base anchored primers (AAGCTTTTTTTTTTTA, AAGCTTTTTTTTTTTC, and AAGCTTTTTTTTTTTG) each paired with seven random primers (AAGCTTTCTACCC, AAGCTTTGGCTCC, AAGCTTATACAGG, AAGCTTGTCATAG, AAGCTTCAAGTCC, AAGCTTCTGACAC, and AAGCTTCAATCGC) in a mix which contained 32P-dATP. Samples were run on a 6% sequencing gel, bands of interest were revealed by exposure to film, excised, amplified by PCR, sequenced, and BLAST searched to identify genes.

Small interfering RNA transfection. Small interfering RNA (siRNA) was used to inhibit the expression and function of target mRNAs. Cells (30-50% confluence) were treated with 100 nmol/L siRNA which had been precomplexed with oligofectamine (Invitrogen, San Diego, CA). Transfection was done in Opti-MEM (Invitrogen) as per manufacturer's guidelines for 4 hours. After 4 hours, medium was supplemented with RPMI 1640 and DCS-FCS (to a final concentration of 5%), 1 nmol/L TGFα was added as appropriate. RNA was harvested after 24 hours and cell counts obtained after 2 or 5 days. Sense sequences used were as follows: FRA1, GCAUCAACACCAUGAGUGGTT; AXL, GGUACAUUGGCUUCGGGAUTT; TCP1, GGCCCUCAAGUCUCAUAUATT; MCM2, GGAUGGAGAGGAGCUCAUUTT: FRA1 (second sequence), GUAUCCCACAUCCAACUCCTT. Antisense sequences used were the complementary sequences and sense sequences were annealed to antisense sequences before use. Negative control siRNA (Ambion, Austin, TX) was also included and showed nonsignificant effects on growth. For the Western analysis of FRA1 inhibition, a second negative control siRNA (Ambion) was used.

Clontech array. Atlas human cancer cDNA expression arrays (Human cancer 1.2; Clontech, Palo Alto, CA) were used to analyze differential gene expression between PE01 and PE01CDDP cell lines, with and without 48 hours treatment with TGFα (1 nmol/L). Isolation of total RNA, cDNA synthesis, labeling of cDNA, and hybridization of cDNA probes to the Atlas Array filters were done according to the manufacturer's protocol (Clontech). To normalize the signal intensity between a pair of arrays, the global (sum) normalization method was used.

Differential growth responses to ErbB ligands after development of cisplatin resistance. The PE01CDDP ovarian cancer cell line was derived from the parental wild-type PE01 cell line by exposure to increasing concentrations of cisplatin and eventually showed a 20-fold increase in resistance to this agent as measured by clonogenic assay (Fig. 1A). Cross-resistance to a range of other platinum analogues, such as carboplatin, iproplatin, and JM40, and also certain cytotoxics, including prednimustine and melphalan, was observed (Fig. 1A). In addition to becoming chemoresistant, PE01CDDP cells showed a reduced dependency on serum-containing hormones and growth factors as indicated by similar growth rates in either complete or charcoal-stripped serum, whereas PE01 cells showed a dramatically reduced growth rate when cultured in charcoal-stripped conditions (Fig. 1B).

These two cell lines exhibited markedly different responses to erbB receptor activating ligands. Whereas the parent PE01 cell line was growth stimulated by TGFα (10 nmol/L), the PE01CDDP cell line was growth inhibited (Fig. 1C). EGF produced similar effects (Fig. 1C). Both NRG1α and NRG1β stimulated the growth of the PE01 cell line at 1 nmol/L with the latter producing a more potent effect. Against the PE01CDDP line, NRG1β inhibited growth to 20% of control cell number, whereas NRG1α had no effect (Fig. 1C).

The effect of TGFα on growth over an 8-day period is shown in Fig. 1B. PE01 cells failed to grow in 5% DCS-containing medium, whereas the PE01CDDP cells achieved log-phase growth under these conditions. Addition of TGFα (0.1 mmol/L) markedly stimulated the growth of PE01 cells but reduced the cell counts of PE01CDDP cells relative to untreated controls after 4 days exposure. Detachment of TGFα-treated PE01CDDP cells was observed at this and later time points (Fig. 2A) and increased apoptosis was shown by an increase in Annexin V positivity (Fig. 2B).

Altered response is not associated with erbB receptor expression or activation. Immunofluorescence antibody measurement of the EGFR, erbB2, and erbB3 by FACscan indicated that both cell lines have similar levels of each receptor (Fig. 3A). Western blot analysis and mRNA transcript expression levels (normalized to γ-actin) also indicated no major differences in receptor expression between these cell lines (data not shown). Treatment with TGFα (10 nmol/L) showed that phosphorylation of EGFR and erbB2 proteins were not significantly different in PE01CDDP cells compared with PE01 cells (Fig. 3B). Both EGFR and erbB2 were tyrosine phosphorylated by TGFα at 30 minutes and 1 hour followed by a reduced level of phosphorylation of erbB2 at 6 hours and a return to control value at 24 hours. The identity of these bands had been confirmed in previous experiments using antibodies specific for the EGFR and erbB2 (data not shown). A similar level of EGFR down-regulation was shown in both cell lines after 2 and 5 days exposure to TGFα (Fig. 3C).

Activation of extracellular signal-regulated kinase and phosphatidylinositol 3-kinase pathways in PE01 and PE01CDDP cells by transforming growth factor-α. Because no differences in the erbB receptor levels were observed between the wild-type and resistant lines in either the presence or absence of TGFα, modifications in intracellular signaling pathways were next investigated. ERK activity was measured in both cell lines at intervals between 5 and 60 minutes after TGFα addition and in untreated controls (Fig. 3D). PE01CDDP cells had a higher basal level of ERK activity compared with that found in PE01 cells. Both cell lines were stimulated to a similar extent over basal levels by the addition of TGFα with maximal stimulation between 5 and 15 minutes and increased activity sustained over the duration of the 60-minute period measured. The increased basal activation of ERK in PE01CDDP cells relative to PE01 cells was also observed on Western blot analysis and this is consistent with an enhanced growth rate (Fig. 4A). Whereas the EGFR inhibitor gefitinib could eliminate TGFα-stimulated ERK activation in PE01 cells, it could only partially reduce its activation in PE01CDDP cells suggesting either constitutive activation of ERK- or EGFR-independent activation in the resistant line (Fig. 4A).

Activation of PKB/Akt by 1 nmol/L TGFα for 15 minutes was observed in PE01 cells consistent with enhanced survival/reduced apoptosis but not in PE01CDDP cells (Fig. 4A). This was reversed by either the PI3K inhibitor LY294002 or gefitinib.

Gefitinib reduced basal growth of PE01 and PE01CDDP cells and was able to reverse both the TGFα-induced growth of PE01 cells and the TGFα-induced growth inhibition of PE01CDDP cells (Fig. 4B). Basal growth of PE01 cells was reduced by both LY294002 and the MEK inhibitor PD98059 suggesting growth dependency on both PI3K/Akt and MEK/ERK pathways (Fig. 4B). TGFα-stimulated growth was reduced by both LY294002 and gefinitib consistent with enhanced survival. Basal growth of PE01CDDP cells was again decreased by both LY294002 and PD98059 (Fig. 4B). The TGFα reduction of growth was reversed by PD98059 consistent with the MEK/ERK pathway being associated with modified growth rates/survival.

Differential gene expression in PE01CDDP cells. Differential display RT-PCR and microarray analysis were used to identify differences in gene expression between PE01 and PE01CDDP cells. Comparisons were made in the presence and absence of TGFα (1 nmol/L). Differential display RT-PCR identified a number of mRNAs that were differentially expressed when run on a 6% sequencing gel. Bands of interest were excised, amplified by PCR, sequenced, and BLAST searched. Putative changes were then confirmed by quantitative RT-PCR. Expression of mRNAs for TCP1 and PCNA were increased over 4-fold in PE01CDDP cells relative to PE01, whereas expression of ZXDA and SLP1 were decreased by >4-fold (Fig. 5A).

Differential mRNA expression was also analyzed using the Clontech Atlas human Cancer 1.2 array. Of 1,185 genes analyzed, 51 increased by at least 2-fold in PE01CDDP relative to PE01, whereas 36 decreased by a similar amount (Table 1). Confirmation of at least a 2-fold increase in mRNA levels in PE01CDDP cells relative to PE01 cells has been verified by real-time PCR for the following genes: FRA1, FANCG/XRCC9, TRAP1, AXL, MCM2, ETV4, and MT3 (Fig. 5A). In contrast, expression of the following genes was reduced in PE01CDDP cells by over 2-fold: IGFBP3, TRAM1, and keratins 4 and 19 (Fig. 5A).

Expression of these genes was further modified by TGFα (Fig. 5A). TGFα increased expression in PE01CDDP cells of seven of the nine genes already increased relative to PE01. In contrast, TGFα tended to decrease expression in PE01 cells in 11 of these genes (Fig. 5A).

Selected genes with known associations to cell proliferation were explored further. Western analysis confirmed the increased expression of FRA1, PCNA, E1AF, and TRAP1 in PE01CDDP cells relative to PE01 and reduced expression of IGFBP3 (Fig. 5B). To investigate the roles of signaling cascades downstream of EGFR in the regulation of these genes, specific signaling inhibitors were added to cells with TGFα before RNA extraction. Gefitinib blocked both TGFα-induced FRA1 expression and IGFBP3 inhibition (Fig. 4C). LY294002 and PD98059 partially blocked the FRA1 modulations, implicating the involvement of both PI3K/Akt and MEK/ERK pathways in regulating these gene responses (Fig. 4C). Similar data to that obtained for FRA1 was obtained with TCP1 and MCM2 (data not shown).

To explore the functionality of the genes associated positively with proliferation, siRNA knockdown of expression was used. Specific siRNAs targeted to FRA1, MCM2, AXL, and TCP1 expression selectively reduced expression of these mRNAs by ∼50% with minimal effects on the expression of the other targets (Fig. 6A). Protein reduction for FRA1 was shown after 48 hours relative to either no treatment or with negative control siRNAs (Fig. 6B). Growth was significantly reduced after targeting MCM2, TCP1, and FRA1, but only minimal effects were obtained by targeting AXL (Fig. 6C).

Effects of transforming growth factor-α on activator protein transcription complex expression. Cisplatin resistance has frequently been shown to be associated with changed expression of AP-1 family members, in particular c-fos (25, 26). Furthermore, because expression of FRA1 was markedly elevated in PE01CDDP cells relative to PE01 cells, it was of interest to compare FRA1 against other members of the AP-1 family.

Semiquantitative RT-PCR was done on RNA extracted from the two cell lines at time points between 15 minutes and 24 hours after TGFα addition. Results showed different patterns of expression of AP1 complex members in the two cell lines both in the presence and absence of TGFα (Fig. 7A). The most dramatic difference between the two cell lines was in the expression of FRA1 mRNA. Nonstimulated PE01 cells had undetectable expression. FRA1 mRNA was induced by TGFα after 60 minutes and continued to increase up to 24 hours. PE01CDDP cells had higher FRA1 expression under nonstimulated conditions and mRNA induction was much stronger at time points between 60 minutes and 24 hours. Differences were also observed in other family members; both JUN and JUND were down-regulated by TGFα at 24 hours in PE01 relative to control, whereas FRA2 seemed sharply elevated at 30 minutes in PE01 compared with PE01CDDP cells. Differences in FRA1 expression were quantified by real-time PCR and are shown in Fig. 7B.

Because FRA1 was elevated in the cisplatin-resistant PE01CDDP line, we determined to see if FRA1 reduction (by siRNA) could influence cisplatin sensitivity of this line. PE01CDDP cells were treated with FRA1 siRNA for 48 hours to reduce FRA1 protein expression (cf. Fig. 6B) and treated with either 12 or 24 μmol/L cisplatin for 1 hour. Cells were plated on plastic (at equivalent numbers to allow for FRA1 siRNA reduction of growth) and colonies counted after 14 days (Fig. 8). Reduction of FRA1 was associated with enhanced sensitivity to cisplatin suggesting that the elevated level of FRA1 expression in PE01CDDP cells was indeed associated with cisplatin resistance.

The erbB receptors are major regulators of cell signaling and function in ovarian cancer cells. Most published data supports the view that EGFR, erbB2, and erbB3 positively regulate cell growth and invasion, whereas erbB4 mediates an inhibitory growth function (1618, 27, 28). Furthermore, the level of erbB receptor expression may also influence the type of response (29). In this report, we describe changes that occur after treatment with cisplatin, one of the cytotoxic agents most widely used in the treatment of ovarian cancer. Resistance to platinum-based therapy occurs frequently in the course of disease management and when this occurs, relapsed tumors often respond poorly to other cytotoxic agents as well. As such, there is a need to understand the signaling changes that can occur in resistant tumor cells to better develop new approaches to their treatment.

We observed three major sets of differences in this cancer cell line model after prolonged exposure to cisplatin: the development of platinum resistance, the ability to grow in charcoal-stripped serum conditions, and differential growth responses to erbB activation. These differences could not be explained by changed expression of the erbB receptors or by initial upstream signaling effects in response to ligands. However, TGFα activated the PI3K/Akt pathway in PE01 cells but not in PE01CDDP cells and this could be linked with reduced apoptosis observed in these cells. Conversely, ERK activity was increased in PE01CDDP cells relative to PE01 cells and this would be consistent with an enhanced rate of proliferation. Whereas the EGFR inhibitor gefitinib could completely block TGFα-stimulated ERK activation in PE01 cells, it could only partially reduce ERK activation in PE01CDDP cells. This may suggest constitutive activation of ERK- or EGFR-independent activation, which could contribute to the changed functionality. There is accumulating evidence that the duration and intensity of ERK signaling is critical in determining the type of biological response (30, 31). Transient activation of ERK is generally associated with proliferation, whereas sustained activation links with growth arrest (30, 31). Sustained stimulation through high levels of activated Ras or Raf can similarly produce growth inhibition (32, 33).

Differential display and microarray analyses identified several gene expression changes that can be linked to platinum resistance or insensitivity (MT3, XRCC9, PCNA) or with cell proliferation and invasion (ETV4, MCM2, TCP1, IGFBP3, SLP1). Significantly, the gene showing the largest change in expression between the two cell lines, FRA1, a member of the AP-1 transcription complex family of proteins, is implicated in the regulation of expression of many genes and may be critical in determining the cell response to external stimuli.

MT3 was elevated 5.5-fold in PE01CDDP cells and this is consistent with previous reports associating increased expression of metallothioneins with exposure to (34) and resistance to platinum (35). The XRCC9/FANCG gene which is linked with Fancomi anemia was also increased (by 3.6-fold) in PE01CDDP cells relative to PE01 cells. Interestingly, TGFα increased expression 2-fold in PE01CDDP while inducing a 10-fold reduction in PE01 cells. Knockout of this gene results in hypersensitivity to cross-linking agents; hence, increased expression might account for reduced sensitivity to cisplatin (36). PCNA was increased 5.4-fold in the resistant line and this is in line with data for lung cancer wherein PCNA expression was higher in cisplatin-resistant tumors (37).

ETV4 (E1AF or PEA3) is a member of the Ets-related transcription factor family and has been associated with both rapid cell growth (38) and invasion (39) in ovarian cancer. Platinum treatment has been reported to up-regulate ETV4 (40) and this change seems to have been maintained in the PE01CDDP line with a 12-fold increase in expression compared with PE01. We observed increased ERK activity in the PE01CDDP cells relative to PE01 cells and ERK is reported to regulate ETV4 (41).

Expression of the tyrosine kinase receptor AXL, which acts as a receptor for GAS6, was increased in PE01CDDP cells. GAS6 is expressed in both PE01 and PE01CDDP cells (data not shown) and acting through AXL is reported to induce cell cycle division in serum-starved cells and activate the ERK pathway (42, 43). This may be responsible for the ability of PE01CDDP cells to thrive in charcoal-stripped serum conditions and could also contribute to an enhanced basal level of phospho-ERK in PE01CDDP cells; however, reduction of AXL by RNA interference (RNAi) had little effect on growth.

Several expression changes reflect the enhanced growth of PE01CDDP cells relative to PE01. The MCM2 protein is required for DNA replication and cell division and has been proposed as a proliferation marker in esophageal cancer (44), and in PE01CDDP cells, it was found to be up-regulated. The TCP1 chaperonin is also up-regulated in PE01CDDP cells. Expression of this chaperonin is particularly linked with the S to G2-M phase transition of the cell cycle and again is likely to be associated with proliferation (45). siRNA reduction of both these targets was associated with growth inhibition in the PE01CDDP cell line suggesting a role for these molecules.

In PE01CDDP cells, IGFBP3 is dramatically down-regulated (54-fold). Numerous studies have described its growth inhibitory effects either through blockade of IGF-stimulated mitogenesis and cell survival or via antiproliferative activity unrelated to its ability to bind to IGFs (43). Recently, it has also been shown in breast epithelial cells that IGFBP3 can potentiate cell proliferation stimulated by EGF (46). This ability may assist growth stimulation in PE01 cells but not in PE01CDDP cells. SLP1 (or antileukoproteinase) has been reported to be overexpressed in ovarian cancer relative to normal ovary (47, 48). Expression of this is reduced 12-fold in PE01CDDP cells relative to PE01 cells.

FRA1 expression is dependent on ERK activity (49) and can also be induced by Akt (50). In untreated PE01CDDP cells, FRA1 expression was 18-fold higher compared with PE01 cells and exposure to TGFα resulted in a further 2-fold increase. FRA1 expression under both basal conditions and TGFα regulation was associated with both Akt and ERK signaling in PE01 cells but with only ERK signaling in PE01CDDP cells. The functional AP-1 complex is a dimer composed of JUN (JUN, JUNB, and JUND) and FOS (FOS FOSB, FRA1, and FRA2) family members (51). The composition of the AP-1 dimers has been shown to confer promoter specificity (51) suggesting that the choice of dimerization partner will affect both the genes up-regulated and the cell response. Furthermore, it is reported that in control of Cyclin D1 (and cell cycle), a system of chromatin trafficking exists whereby early expression and recruitment of c-fos is superseded by later and prolonged FRA1 recruitment on ERK-1- and ERK-2-dependent promoter sites (52). The elevated FRA1 in the PE01CDDP cell line is consistent with rapid growth of this cell line even in reduced serum conditions but the further increase we observed after treatment with TGFα may serve to promote AP-1 dimers that favor an apoptotic response. In the C6 glioma cell line, overexpression of FRA1 was shown to both inhibit growth and increase apoptosis post treatment with hexamethylene bisacetamide (53). The observed temporal changes in expression seen not only with FRA1 but also FRA2, JUN and JUN D suggest that in these two cell lines the AP-1 dimer composition may be different which could explain their differing responses to the same erbB ligands.

FRA1 has been shown to have a role in cell motility and attachment in fibroblastoid L929 cells (54) and in the proliferation, invasiveness, and motility of breast cancer cells (55). PC-12 pheochromocytoma cells transfected with FRA1 showed a significant enhancement of proliferation and were able to proliferate in low-serum–containing medium (56). This is in agreement with our observation of a more rapid growth rate of PE01CDDP cells relative to PE01 and their ability to proliferate in charcoal-stripped conditions. Reduced expression of FRA1 following RNAi inhibited growth consistent with this view. Similarly, reduced expression of FRA1 also enhanced cisplatin sensitivity suggesting a role for FRA1 in cisplatin resistance as has been reported for FOS (25, 26).

In conclusion, these results indicate that exposure to cisplatin can lead not only to the onset of cisplatin resistance but also modified signaling with consequent changes in growth regulation. In this report, we have identified a number of potential gene changes, which might be significant in explaining the changed phenotype. In particular, we have highlighted differences in the AP-1 transcription factor family members, which may contribute to differential gene regulation and responses within this model system. Further studies are required to explore the prevalence of these changes in primary ovarian cancers.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Astrazeneca for supplies of gefitinib.

1
McGuire WP, Markham M. Primary ovarian cancer chemotherapy: current standards of care.
Br J Cancer
2003
;
89
:
S3
–8.
2
Benhar M, Engelberg D, Levitzki A. Cisplatin-induced activation of the EGF receptor.
Oncogene
2002
;
21
:
8723
–31.
3
Christen RD, Hom DK, Porter DC, et al. Epidermal growth factor regulates the in vitro sensitivity of human ovarian carcinoma cells to cisplatin.
J Clin Invest
1990
;
86
:
1632
–40.
4
Cenni B, Aebi S, Nehme A, Christen RD. Epidermal growth factor enhances cisplatin-induced apoptosis by a caspase 3 independent pathway.
Cancer Chemother Pharmacol
2001
;
47
:
397
–403.
5
Frankel A, Mills GB. Peptide and lipid growth factors decrease cis-diamminedichloroplatinum-induced cell death in human ovarian cancer cells.
Clin Cancer Res
1996
;
2
:
1307
–13.
6
Yarden Y, Slikowski MX. Untangling the ErbB signalling network.
Nat Rev Mol Cell Biol
2001
;
2
:
127
–37.
7
Olayioye MA, Neve RM, Lane NE, Hynes NE. The ErbB signaling network: receptor heterodimerization in development and cancer.
EMBO J
2000
;
19
:
3159
–67.
8
Egan SE, Weinberg RA. The pathway to signal achievement.
Nature
1993
;
365
:
781
–2.
9
Verbeek BS, Adriaansen-Slot SS, Vroom TM, Beckers T, Rijksen G. Overexpression of EGFR and c-erbB2 causes enhanced cell migration in human breast cancer cells and NIH3T3 fibroblasts.
FEBS Lett
1998
;
425
:
145
–50.
10
Chen P, Xie H, Sekar MC, Gupta K, Wells A. Epidermal growth factor receptor-mediated cell motility: phospholipase C activity is required, but mitogen-activated protein kinase activity is not sufficient for induced cell movement.
J Cell Biol
1994
;
127
:
847
–57.
11
Yarden Y. The EGFR family and its ligands in human cancer. Signalling mechanisms and therapeutic opportunities.
Eur J Cancer
2001
;
37
:
S3
–8.
12
Bartlett JMS, Langdon SP, Simpson BJB, et al. The prognostic value of epidermal growth factor receptor mRNA expression in primary ovarian cancer.
Br J Cancer
1996
;
73
:
301
–6.
13
Berchuck A, Kamel A, Whitaker R, et al. Overexpression of HER-2/neu is associated with poor survival in advanced epithelial ovarian cancer.
Cancer Res
1990
;
50
:
4087
–91.
14
Berchuck A, Rodriguez GC, Kamel A, et al. Epidermal growth factor receptor expression in normal ovarian epithelium and ovarian cancer. I. Correlation of receptor expression with prognostic factors in patients with ovarian cancer.
Am J Obstet Gynecol
1991
;
164
:
669
–74.
15
Slamon DJ, Godolphin W, Jones LA, et al. Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer.
Science
1989
;
244
:
707
–12.
16
Morishige K, Kurachi H, Amemiya K, et al. Evidence for the involvement of transforming growth factor α and epidermal growth factor receptor autocrine growth mechanism in primary human ovarian cancers in vitro.
Cancer Res
1991
;
51
:
5322
–8.
17
Scambia G, Benedetti-Panici P, Battaglia F, Ferrandina G, Gaggini C, Mancuso S. Presence of epidermal growth factor (EGF) receptor and proliferative response to EGF in six human ovarian carcinoma cell lines.
Int J Gynecol Cancer
1991
;
1
:
253
–8.
18
Crew AJ, Langdon SP, Miller EP, Miller WR. Mitogenic effects of epidermal growth factor and transforming growth factor-α on EGF-receptor positive human ovarian carcinoma cell lines.
Eur J Cancer
1992
;
28
:
337
–41.
19
Sewell JM, Macleod KG, Ritchie A, Smyth JF, Langdon SP. Targeting the EGF receptor in ovarian cancer with the tyrosine kinase inhibitor ZD 1839 (“Iressa”).
Br J Cancer
2002
;
86
:
456
–62.
20
Fujimura M, Hidaka T, Saito S. Selective inhibition of the epidermal growth factor receptor by ZD1839 decreases the growth and invasion of ovarian clear cell adenocarcinoma cells.
Clin Cancer Res
2002
;
8
:
2448
–54.
21
Finkler N, Gordon A, Crozier M, et al. Phase 2 evaluation of OSI-774, a potent oral antagonist of the EGFR-TK in patients with advanced ovarian carcinoma.
Proc Am Soc Clin Oncol
2001
;
20
:
208a
(abs 831).
22
Wolf CR, Hayward IP, Lawrie SS, et al. Cellular heterogeneity and drug resistance in two ovarian adenocarcinoma cell lines derived from a single patient.
Int J Cancer
1987
;
39
:
695
–702.
23
Brotherick I, Shenton BK, Angus B, Waite IS, Horne CHW, Lennard TWJ. A flow cytometric study of c-erbB-3 expression in breast cancer.
Cancer Immunol Immunother
1995
;
41
:
280
–6.
24
Mullen P, McPhillips F, MacLeod K, Monia B, Smyth JF, Langdon SP. Antisense oligonucleotide targeting of Raf-1:importance of Raf-1 mRNA expression levels and Raf-1 dependent signaling in determining growth response in ovarian cancer.
Clin Cancer Res
2004
;
10
:
2100
–8.
25
Moorehead RA, Singh G. Influence of the proto-oncogene c-fos on cisplatin sensitivity.
Biochem Pharmacol
2000
;
59
:
337
–45.
26
Scanlon KJ, Jiao L, Funato T, et al. Ribozyme-mediated cleavage of c-fos mRNA reduces gene expression of DNA synthesis enzymes and metallothionein.
Proc Natl Acad Sci U S A
1991
;
88
:
10591
–5.
27
Gilmour LM, Macleod KG, McCaig A, Gullick WJ, Smyth JF, Langdon SP. Expression of erbB-4/HER-4 growth factor receptor isoforms in ovarian cancer.
Cancer Res
2001
;
61
:
2169
–76.
28
Gilmour LM, Macleod KG, McCaig A, et al. Neuregulin expression, function, and signaling in human ovarian cancer cells.
Clin Cancer Res
2002
;
8
:
3933
–42.
29
Xu F, Yu Y, Le XF, Boyer C, Mills GB, Bast RC Jr. The outcome of heregulin-induced activation of ovarian cancer cells depends on the relative levels of HER-2 and HER-3 expression.
Clin Cancer Res
1999
;
5
:
3653
–60.
30
Marshall CJ. Specificity of receptor tyrosine kinase signaling: transient versus sustained extracellular signal-regulated kinase activation.
Cell
1995
;
80
:
179
–85.
31
Cowley S, Paterson H, Kemp P, Marshall CJ. Activation of MAP kinase kinase is necessary and sufficient for PC12 differentiation and for transformation of NIH 3T3 cells.
Cell
1994
;
77
:
841
–52.
32
Sewing A, Wiseman B, Lloyd AC, Land H. High-intensity Raf signal causes cell cycle arrest mediated by p21Cip1.
Mol Cell Biol
1997
;
9
:
5588
–97.
33
Woods D, Parry D, Cherwinski H, Bosch E, Lees E, McMahon M. Raf-induced proliferation or cell cycle arrest is determined by the level of Raf activity with arrest mediated by p21Cip1.
Mol Cell Biol
1997
;
17
:
5598
–611.
34
Ferguson PJ. Mechanisms of resistance of human tumours to anticancer drugs of the platinum family: a review.
J Otolaryngol
1995
;
24
:
242
–52.
35
Koo MS, Kwo YG, Park JH, Choi WJ, Billiar TR, Kim YM. Signaling and function of caspase and c-jun N-terminal kinase in cisplatin-induced apoptosis.
Mol Cells
2002
;
13
:
194
–201.
36
Koomen M, Cheng NC, van de Vrugt HJ, et al. Reduced fertility and hypersensitivity to mitomycin C characterize Fancg/Xrcc9 null mice.
Hum Mol Genet
2002
;
11
:
273
–81.
37
Ogawa J, Iwazaki M, Inoue H, Koide S, Shohtsu A. Immunohistochemical study of glutathione-related enzymes and proliferative antigens in lung cancer. Relation to cisplatin sensitivity.
Cancer
1993
;
71
:
2204
–29.
38
Nishikawa A, Iwasaki M, Akutagawa N, et al. Expression of various matrix proteases and Ets family transcriptional factors in ovarian cancer cell lines: correlation to invasive potential.
Gynecol Oncol
2000
;
79
:
256
–63.
39
Kaya M, Yoshida K, Higashino F, Mitaka T, Ishii S, Fujinaga K. A single ets-related transcription factor, E1AF, confers invasive phenotype on human cancer cells.
Oncogene
1996
;
12
:
221
–7.
40
Funaoka K, Shindoh M, Yoshida K, et al. Activation of the p21(Waf1/Cip1) promoter by the ets oncogene family transcription factor E1AF.
Biochem Biophys Res Commun
1997
;
236
:
79
–82.
41
Ito E, Sweterlitsch LA, Tran PB, Rauscher FJ III, Naryanan R. Inhibition of PC-12 cell differentiation by the immediate early gene fra-1.
Oncogene
1990
;
5
:
1755
–60.
42
Goruppi S, Ruaro E, Schneider C. Gas6, the ligand of Axl tyrosine kinase receptor, has mitogenic and survival activities for serum starved NIH3T3 fibroblasts.
Oncogene
1996
;
12
:
471
–80.
43
Allen MP, Zeng C, Schneider K, et al. Growth arrest-specific gene 6 (Gas6)/adhesion related kinase (Ark) signaling promotes gonadotropin-releasing hormone neuronal survival via extracellular signal-regulated kinase (ERK) and Akt.
Mol Endocrinol
1999
;
13
:
191
–201.
44
Kato H, Miyazaki T, Fukai Y, et al. A new proliferation marker, minichromosome maintenance protein 2, is associated with tumor aggressiveness in esophageal squamous cell carcinoma.
J Surg Oncol
2003
;
84
:
24
–30.
45
Dittmar G, Schmidt G, Kopun M, Werner D. Mapping of G2/M-phase prevalences of chaperon-encoding transcripts by means of a sensitive differential hybridization approach.
Cell Biol Int
1997
;
21
:
383
–91.
46
Martin JL, Weenink SM, Baxter RC. Insulin-like growth factor-binding protein-3 potentiates epidermal growth factor action in MCF-10A mammary epithelial cells. Involvement of p44/42 and p38 mitogen-activated protein kinases.
J Biol Chem
2003
;
278
:
2969
–76.
47
Shigemasa K, Tanimoto H, Underwood LJ, et al. Expression of the protease inhibitor antileukoprotease and the serine protease stratum corneum chymotryptic enzyme (SCCE) is coordinated in ovarian tumors.
Int J Gynecol Cancer
2001
;
11
:
454
–61.
48
O'Hagan RC, Tozer RG, Symons M, McCormick F, Hassell JA. The activity of the Ets transcription factor PEA3 is regulated by two distinct MAPK cascades.
Oncogene
1996
;
13
:
1323
–33.
49
Tiwari G, Sakaue H, Pollack JR, Roth RA. Gene expression profiling in prostate cancer cells with Akt activation reveals Fra-1 as an Akt inducible gene.
Mol Cancer Res
2003
;
6
:
475
–84.
50
Vial E, Marshall C. Elevated ERK-MAP kinase activity protects the FOS family member FRA-1 against proteasomal degradation in colon carcinoma cells.
J Cell Sci
2003
;
116
:
4957
–63.
51
Bakiri L, Matsuo K, Wisniewska M, Wagner EF, Yaniv M. Promoter specificity and biological activity of tethered AP-1 dimers.
Mol Cell Biol
2002
;
13
:
4952
–64.
52
Burch PM, Yuan Z, Loonen A, Heintz NH. An extracellular signal-related kinase 1- and 2-dependent program of chromatin trafficking of c-Fos and Fra-1 is required for cyclin D1 expression during cell cycle reentry.
Mol Cell Biol
2004
;
11
:
4696
–709.
53
Shirsat NV, Shaikh SA. Overexpression of the immediate early gene Fra-1 inhibits proliferation, induces apoptosis, and reduces tumourigenicity of c6 glioma cells.
Exp Cell Res
2003
;
291
:
91
–100.
54
Tkach V, Tulchinsky E, Lukanidin E, Vinson C, Bock E, Berezin V. Role of the Fos family members, c-Fos, Fra-1 and Fra-2 in the regulation of cell motility.
Oncogene
2003
;
22
:
5045
–54.
55
Belguise K, Kersual N, Galtier F, Chalbos D. FRA-1 expression level regulates proliferation and invasiveness of breast cancer cells.
Oncogene
2005
;
24
:
1434
–44.
56
Shaulian E, Karin M. AP-1 in cell proliferation and survival.
Oncogene
2001
;
20
:
2390
–2400.