SEARCH

SEARCH BY CITATION

Keywords:

  • p38;
  • lung cancer;
  • cisplatin;
  • ERCC1;
  • never smokers

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

BACKGROUND:

Expression of DNA-repair proteins and activated mitogen-activated protein kinases (MAPKs) may differ according to smoking status. The authors investigated whether p38 MAPK activity contributed to the viability of cisplatin in lung cancer cell lines from never or light smokers and to ERCC1 mRNA expression.

METHODS:

Activated p38 MAPK was tested as a predictor for ERCC1 levels in 117 lung adenocarcinomas. Cell viabilities of NCI-H1975, NCI-H1793, NCI-H1650, and NCI-H1651 cell lines, derived from never or light smokers, were measured after treatment with the p38 MAPK inhibitor SB202190 and cisplatin. The role of p38α (MAPK14) and p38β (MAPK11) isoforms and ERCC1 was evaluated using RNA interference.

RESULTS:

ERCC1 protein-level expression was predicted by activated p38 MAPK in lung adenocarcinoma tissues. The p38-specific inhibitor SB202190 strongly decreased cell viability (43%-63%). SB202190 plus cisplatin significantly decreased cell viability in every cell line, including cisplatin-resistant NCI-H1793. Genetic inhibition, targeting both MAPK11 and MAPK14, reduced the viability of the different cell lines: down-regulation of p38β accounted for most of this effect. Cisplatin's effect was greater after MAPK11 down-regulation for NCI-H1651, and MAPK14 down-regulation for NCI-H1650. In addition, both SB202190 and MAPK11 inhibition reduced excision repair cross-complementing 1 mRNA levels.

CONCLUSIONS:

Lung cancer cells from never or light smokers rely on p38 MAPK signaling for survival. MAPK11 is involved in that pathway and might contribute to ERCC1 expression. Sensitization to cisplatin can be achieved by pharmacological inhibition of p38 MAPK signaling. Cancer 2012. © 2012 American Cancer Society.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

Lung cancer is the leading cause of cancer death worldwide. Smoking is, by far, the main contributor to lung cancer. Common treatments include surgery, cisplatin-based chemotherapy, and radiation therapy. In recent years, various targeted molecular therapies, alone or combined with cisplatin, have been developed to treat advanced lung cancer. Drugs that target epidermal growth factor receptors (EGFR) are now used in the clinical setting. Lung cancer patients with no or minimal history of tobacco use show markedly better clinical outcomes when treated with these agents.1-3 Sensitivity to EGFR inhibitors has been associated with EGFR mutations, which make cancer cells dependent on continued EGFR signaling for proliferation and survival.4 Other differences in the genetic makeup of lung cancer cells from lung cancer patients who have never smoked (never smokers) are important; however, nonsmoking status is the strongest predictor of benefit from EGFR tyrosine kinase inhibitors.5

p38 mitogen-activated protein kinases (MAPKs), a class of serine/threonine kinases, are responsive to stress stimuli, such as cytokines, ultraviolet irradiation, heat shock, and osmotic shock, and are involved in cell differentiation and apoptosis.6 Four isoforms of p38 MAPK, p38α (MAPK14), p38β (MAPK11), p38γ (MAPK12), and p38δ (MAPK13), have been identified. MAPK14 and MAPK11 are ubiquitously expressed, whereas MAPK12 and MAPK13 have more restricted patterns and are likely to have specialized functions.

The role of p38 MAPK in inflammatory disease has led to the development of a large number of small-molecule p38 inhibitors. Most of these compounds target MAPK14. Theoretically, they could be used to treat tumors that rely on MAPK14 for progression, or could be combined with DNA-damaging chemotherapy to trigger the death of cancer cells by impairing cell cycle arrest and DNA repair mechanisms.7

It is not clear whether p38 MAPK activity favors or inhibits cancer cell growth. A role for MAPK14 in increasing cancer cell migration, tumor invasion, and metastasis has been supported by experiments that have used cancer-cell lines where expression of matrix metalloproteinases and angiogenic factors are induced by p38 MAPK signaling.8-13 Conversely, direct evidence of a tumor suppressor role for MAPK14 has been provided by knockout mice. Ventura et al reported that deletion of MAPK14 in adult mice resulted in increased proliferation and defective differentiation of lung stem and progenitor cells, both in vivo and in vitro.14 As a consequence, the inactivation of MAPK14 leads to immature and hyperproliferative lung epithelium that is highly sensitized to K-RASG12V-induced tumorigenesis.

Chemotherapy induces p38 MAPK activity, which contributes to the cytotoxic effect of cisplatin and other chemotherapeutic agents.15 The effect of combining p38 inhibitors with cisplatin appears to depend to a large extent on cell type. Studies are needed to determine which types of cancer are likely to respond to therapies that target the p38 MAPK pathway.

Previously, we studied the expression of the p38 MAPK pathway in a large cohort of patients with lung cancer.16 The results from our study showed that the p38 MAPK pathway was activated in a substantial proportion of lung tumors (55%), a finding consistent with the results reported by Greenberg et al.17 Importantly, by correlating the expression of activated p38 MAPK in tumor cells with smoking status, we showed that it was strongly associated with life-long nonsmoking, independent of age, sex, and histological type.16

Many preclinical studies of cisplatin resistance have reported the importance of cells' abilities to repair DNA damage. We and others have reported that the expression of the ERCC1 (excision repair cross-complementation group 1) protein is a key predictor of the benefit of cisplatin-based chemotherapy in lung cancer.18, 19 We have also reported that the expression of ERCC1 in lung cancer cells was higher in never smokers compared with smokers.20

Herein, we have studied the inhibition of the p38 MAPK pathway on the growth of lung tumor cells from never smokers who had various mutational profiles and sensitivities to cisplatin. We also explored whether ERCC1 expression was related to p38 MAPK signaling in both tissues and cell lines, and whether inhibition of p38 MAPK effected tumor cell sensitivity to cisplatin.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

Cell Lines and Culture

All cell lines were purchased from the American Tissue Culture Collection. H1793 and H1651 cells were grown in DMEN/F12 medium, containing 2.5 mM L-glutamine, 0.02 mg/mL insulin, and 0.01 mg/mL transferrin, supplemented with 10% fetal calf serum (FCS), 100 U/mL penicillin G sodium, and 100 μg/mL streptomycin sulfate. H1650 cells and H1975 cells were grown in RPMI-1640 with 2 mM L-glutamax, supplemented with 10% FCS and antibiotics, as above. Media and supplements for cell cultures were purchased from Gibco-Invitrogen (Carlsbad, Calif).

Cell Proliferation and Viability Assays

Cells were seeded at 3 × 103 per well, in 96-well microplates, in a final volume of 100 μL/well of culture medium and then grown for 24 hours. The cells were then treated with 20 μM SB202190 (Calbiochem, Darmstadt, Germany) for 30 minutes or with 20 μM cisplatin (Merck, Frankfurt, Germany) for 3 hours, with or without 30 minutes of pretreatment with 20 μM SB202190. Alternatively, cells were transfected with small interfering (si)RNA (see below). Transfected cells were grown alone or were treated 24 hours later with 20 μM cisplatin for 3 hours. After the treatments, cells were washed with phosphate-buffered saline and supplemented with fresh complete medium. Measurements of viable cell mass were performed 5 days later using a colorimetric-based reaction by adding 10 μL/well of Cell Proliferation Reagent WST-1 for 2 hours at 37°C, in accordance with the manufacturer's protocols (Roche, Basel, Switzerland).

Thereafter, the absorbance was measured at 450 nm with an enzyme-linked immunosorbent assay plate reader. Cell viability was expressed as viable cell mass after a given treatment that was normalized to that of parallel cultures of untreated cells (viable cell mass, %). Each experiment was performed with at least 3 measurements for each condition. The results shown are the mean values of 3 such experiments.

siRNA Transfections

SiRNAs for the down-regulation of the different p38 MAPK isoforms (Hs_MAPK14_5_HP and Hs_MAPK11_5_HP validated siRNAs), ERCC1 (Hs_ERCC1_4_HP), and negative control siRNA (AllStars Negative Control) were purchased from Qiagen (Hilden, Germany). The sequences of the siRNAs used for this study were as follows:

MAPK14 sense 5′-CUGCGGUUACUUAAACAUATT-3′, antisense 5′-UAUGUUUAAGUAACCGCAGTT-3′; MAPK11 sense 5′-GGAUGGAGCUGAUCCAGUATT-3′, antisense 5′-UACUGGAUCAGCUCCAUCCTG-3′; ERCC1 sense 5′-CGUGAAGUCAGUCAACAAA-3′, antisense 5′-UUUGUUGACUGACUUCACG-3′.

Before transfections, cell lines were seeded in 96-well plates at a concentration of 3 × 103 cells/well. After 24 hours, cells were transfected with the different siRNAs to make final concentrations of 5 to 50 nM using Hiperfect transfection reagent (Qiagen), following the manufacturer's instructions. Control experiments were done in parallel by transfecting the cells with the nonsilencing control, siRNA (Qiagen), at equivalent concentrations to the target siRNA. Vehicle containing Hiperfect alone was applied to cells, and viable cells were considered as having 100% survival. Assays were performed in triplicate with at least 3 measurements for each condition.

Quantitative Reverse Transcriptase Polymerase Chain Reaction

For real-time quantitative reverse transcriptase polymerase chain reaction (PCR) analysis, cells were seeded at 3 × 104 cells per well, in 12-well plates, in a final volume of 1 mL/well of culture medium. At 24 hours later, cells were transfected with p38 siRNA or ERCC1 siRNA, or the negative control siRNA, or were exposed to the p38 MAPK-specific inhibitor (SB202190) alone or combined with cisplatin, as described above. Cultures were stopped after different time points (24, 48, and 72 hours) by adding 400 μL of RLT buffer (Qiagen). RNAs were extracted using a RNeasy-micro kit (Qiagen). cDNAs were synthesized using the M-MlV Rev transcriptase enzyme with random hexamers (Applied Biosystem, Foster City, Calif). Template cDNAs were added to a Taqman PCR Universal Master Mix with specific primers (described below) and a probe for each gene, to make a final volume of 25 μL (all reagents from Applied Biosystems).

Primers for ERCC1 were GGGAATTTGGCGACGTAATTC (forward) and GCGGAGGCTGAGGAACAG (reverse), and 6FAM (carboxyfluorescein) 5′-CACAGGTGCTCTGGCCCAGCACATA-3′, as described previously.21 The endogenous reference genes were PPIA (peptidylprolyl isomerase A) and RPLPO (ribosomal protein P0). Primers for MAPK14 (assay ID, Hs00176247_m1) and MAPK11 (assay ID, Hs00177101_m1) were purchased from Applied Biosystems. Quantification of gene expression was performed using the ABI Prism 7900HT Sequence Detection System (Applied Biosystems). Relative gene expression values were calculated by the 2−DDCt method using Sequence Detection System 2.3 software (Applied Biosystems). The amount of target gene was normalized to an endogenous reference gene (PPIA, RPLPO RNAs) and was relative to a calibrated sample (Human Ref RNA; Stratagene, La Jolla, Calif).

Immunohistochemistry

The study initial cohort included a total of 188 chemonaive patients with primary lung adenocarcinoma, for which formalin-fixed, paraffin-embedded tumor samples were available.16, 20 The immunohistochemical data had been previously recorded separately for nucleotide excision repair protein ERCC1 and the phosphorylated forms of MAPKs, including ERK, JNK, and p38. The present analyses were limited to the subgroup of cases for which ERCC1 and MAPK values were available. ERCC1 expression was categorized into high (positive expression) and low levels (negative expression), and scores were attributed to phosphorylated MAPK expression, using the median as a cutoff point, as previously described.16, 20 The protocols for the immunostaining of the tumors samples are shown in Table 1.

Table 1. Protocols for the Immunostaining of the Samples
ProteinAntigen RetrievalPrimary Antibody DilutionIncubationCloneSourceAntibodyExternal Positive ControlReferenceTechnology
  1. Abbreviation: p, phosphorylated.

pERKCitrate buffer 10 mM, pH 6.0, 30 minutes in a water bath1/400Overnight 4°CRabbitPolyclonalProstate9101Cell Signaling
pP38Citrate buffer 10 mM, pH 6.0, 30 minutes in a water bath1/800Overnight 4°C12F8RabbitMonoclonalBreast4631Cell Signaling
pJNKCitrate buffer 10 mM, pH 6.0, 30 minutes in a water bath1/1600Overnight 4°CG-7MouseMonoclonalLungsc-6254Santa Cruz
ERCC1Citrate buffer 10 mM, pH 6.0, 30 minutes in a water bath1/30060 minutes at room temperatureA8F1MouseMonoclonalTonsilMS-671-P1Neomarker

Statistical Analyses

Differences between subgroups were assessed using Fisher exact test or the Mann-Whitney U test. The Kendall tau statistic was used to measure the correlation between immunohistochemical variables. Differences between subgroups were assessed using Fisher exact test or the Mann-Whitney U test. The Kendall tau statistic was used to measure the correlation between immunohistochemical variables. The logistic regression model was developed with the dichotomized levels of proteins to determine whether ERCC1 expression was explained by the status of phosphorylated (p)ERK, pJNK, and pp38 after adjustment for clinical or pathological variables linked to the smoking status.16

The means of viable cell masses or ERCC1 relative expression values were compared using 1-way or 2-way analysis of variance, which was followed when appropriate by Tukey-Kramer tests to compare 2 means, and by simple linear regression when the means were associated with ordered variables, such as increasing concentrations of siRNAs.

All tests were 2-sided, and a P value of <.05 was considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

ERCC1 Protein Expression and p38 MAPK Activation State in Tumor Tissues

Among the 188 validated cases, 160, 155, 159, and 148 cases were assessable for ERCC1, pJNK, pERK, and pp38, respectively, and 135, 137, and 130 cases were assessable for ERCC1 with an H score value for activated protein JNK, ERK, or p38, respectively and were included in the univariate analysis. Among the initial cases, 117 (62%) had a value for each protein (ERCC1, pJNK, pERK, and pp38) and were included in the multivariate analysis. Patients who were included in the multivariate analysis (n = 117) and validated cases based on our population, (n = 188) were comparable in terms of demographic, clinical, and histological characteristics. The median age at diagnosis of the included patients was 66 years (interquartile range, 51-72 years). Patients were well balanced between men and women (58 of 59). The majority of patients were smokers (84 of 117, 72%).

Examples of tissue immunostaining with phosphorylated p38 MAPK and ERCC1 antibodies are shown in Figure 1A. The distribution of ERCC1-negative and ERCC1-positive cases, according to activated p38 MAPK expression scores in tissues, is shown in Figure 1B. The median and interquartile range of the H scores for pERK, pJNK, and pp38 according to the ERCC1 status are shown in Figure 2. By using logistic regression, activated p38 MAPK (P = .003) and activated JNK (P = .02), but not activated ERK (P = .37), predicted high ERCC1 expression levels in univariate models (Table 2). Activated p38 MAPK (adjusted odds ratio, 1.70; 95% confidence interval, 1.10-2.63; P = .02), but not activated JNK (adjusted odds ratio, 1.11; 95% confidence interval, 1.69-1.78, P = .68), predicted high ERCC1 expression levels in multivariate models that were adjusted for the smoking status and smoking status-related variables, including age, sex, and histological subtype (Table 2).

thumbnail image

Figure 1. ERCC1 immunohistochemical positivity is shown according to p38 mitogen-activated protein kinase (MAPK) activation state in 117 lung adenocarcinomas. (A) Examples show tumor tissues expressing high levels of ERCC1 or activated p38 MAPK. (B) ERCC1-positive cases distributed according to their activated p38 MAPK expression scores were grouped into 3 categories (0-0.5, 1-1.5, and 2-3). The P value is for the predicted ERCC1 positivity by activated p38 MAPK scores using a multivariate logistic regression model that was adjusted according to smoking status and smoking status-related variables, including age, sex, and histological subtype.

Download figure to PowerPoint

thumbnail image

Figure 2. Distribution of phosphorylated (p)ERK, pJNK, and pp38 H score is shown according to ERCC1 status. Lower and upper limits of boxes indicate the first and third quartiles, circles within boxes indicate the median values, and the lines extending from the boxes indicate the range of values. E0, ERCC1 negative; E1, ERCC1 positive.

Download figure to PowerPoint

Table 2. Univariate and Logistic Regression Analysis of pERK, pJNK, and pP38 Expression According to ERCC1 Status
ModelERCC1, No.Overall Model Fit, PPredictor Variable
LowHighTotalPOdds Ratio95% CI
  1. Abbreviations: CI, confidence interval; p, phosphorylated.

Separate univariate logit models       
 pERK8057137.37.371.140.86-1.50
 pJNK7857135.02.021.551.07-2.25
 pp387357130.003.0031.691.19-2.40
Multivariate logit model6453117.001   
 pERK    .871.030.74-1.41
 pJNK    .681.110.69-1.78
 pp38    .021.701.10-2.62

These results suggest that ERCC1 expression might depend on p38 MAPK signaling in lung cancer. Because p38 MAPK signaling was generally higher in never smokers than in smokers, we selected 3 cancer cell lines (NCI-H1975, NCI-H1793, NCI-H1651) from never smokers and 1 cancer cell line (NCI-H1650) from a light smoker (10 pack-years) for the following experiments.

Pharmacological Inhibition of p38 MAPK Signaling and Cell Survival

p38 MAPK inhibition with the p38 MAPK-specific inhibitor, SB202190 (INH190), resulted in decreased viability of NCI-H1975, NCI-H1793, NCI-H1650, and NCI-H1651 cell lines (referred to herein as H1975, H1793, H1650, and H1651). The decrease in cell viability, expressed as a percentage of the control cell's viability (Fig. 3), was strong and significant (P < .01) for H1975 and H1650 (40% and 63%, respectively), and was modest for H1793 (11%; P < .05) and H1651 (12%; not significant).

thumbnail image

Figure 3. Cell viability after treatment with the p38 mitogen-activated protein kinase inhibitor SB202190 is shown. H1975, H1793, H1650, and H1651 cells were incubated either with SB202190 at 20 μM, or with SB202190 at 20 μM, followed by cisplatin (CDDP) at 20 μM (SB202190, CDDP 20 μM). Control cells were grown without any drug (Control, cisplatin 0 μM) or with cisplatin at 20 μM (Control, cisplatin 20 μM). Cell viability was assessed using a WST-1 assay and was expressed as percentage ± standard error of the mean with respect to the controls. The control without any drug (Control, cisplatin 0 μM) was used as a reference (100%). Three independent experiments were performed in triplicate. *P < .05 versus control without cisplatin; **P < .01 versus control without cisplatin.

Download figure to PowerPoint

Parallel experiments showed that cells were sensitive to exposure to cisplatin at 20 μM, with the exception of H1793 (decrease in cell viability of 24%, 7%, 48%, and 56% for H1975, H1793, H1650, and H1651, respectively). Cisplatin was only superior to SB202190 for H1651.

When treatment with SB202190 was followed by exposure to cisplatin, a significant decrease (P < .01) in cell viability, compared with the control, was observed for every cell line (52%, 29%, 76%, and 83% for H1975, H1793, H1650, and H1651, respectively). The decrease in cell viability using this drug combination significantly exceeded (P < .01) the corresponding decrease in cell viability with cisplatin alone for H1793, H1650, and H1651.

Concomitant Genetic Inhibition of MAPK14 and MAPK11, and Cell Survival

To verify the effect of SB202190 to target both MAPK14 and MAPK11, equimolar mixtures of siRNAs (sip38) that targeted MAPK11 and MAPK14 mRNA isoforms were used (Fig. 4). Although cells transfected with a scrambled siRNA control at 50 nM did not show reduced cell viability compared with cells treated with vehicle alone, treatment with an equimolar mixture of sip38 at 5 nM was sufficient to significantly (P < .01) reduce cell viability in every cell line (12% for H1975, 43% for H1793, and 42% for H1650 and H1651) compared with the si control. Decreased expression of p38 protein after the exposure of cell lines to p38 siRNAs was confirmed in H1975, H1793, H1650, and H1651 cells by Western blot.

thumbnail image

Figure 4. Cell viability after transfection with p38-specific small interfering (si)RNAs is shown. H1975, H1793, H1650, and H1651 cells (A-D, respectively) were incubated with an equimolar mixture of p38-specific siRNAs that targeted α and β isoforms (sip38) at 5 nM, 25 nM, and 50 nM, or with an siRNA-negative control at 50 nM (si control). Transfected and control cells (control) were grown for 5 days with and without cisplatin (CDDP) at 20 μM. The results of the WST-1 assay are expressed as percentage of vehicle-treated control (considered as 100%), and represent the mean ± standard error of the mean of at least 3 independent experiments performed in triplicate. When there was no interaction between p38 siRNA and cisplatin, the P value for the main effect of p38 siRNA (ie, independent of exposure to cisplatin) is indicated above the corresponding bars (over braces). When there was an interaction between p38 siRNA and cisplatin, the means for each condition (with or without cisplatin) were compared with their respective si controls (with or without cisplatin) to separately assess the effect of p38 siRNA in each of the 2 conditions. **P < .01 versus si control.

Download figure to PowerPoint

At 25 nM of sip38 for H1793 and H1650, and at 50 nM sip38 for H1651, cell viability was more significantly decreased (P < .01) than after exposure to 20 μM cisplatin alone (H1793, 67% vs 35%; H1650, 55% vs 38%; and H1651, 73% vs 55%). The decrease in cell viability was still greater at 50 nM of sip38 for H1793 and H1650 (74% and 58%, respectively).

There was a dose-response relationship with increasing concentration of sip38 for H1975, H1793, and H1651 (R2 > 0.95 and P < .03 for each dose-response analysis).

The interaction test between sip38 and cisplatin was significant (P < .05) for H1793, H1650, and H1651. In these cells, the addition of cisplatin, with increasing concentrations of sip38, decreased the plateau at the higher concentrations of sip38.

Separate Genetic Inhibition of MAPK11 and MAPK14, and Cell Survival

To determine the respective contribution of MAPK11 and MAPK14, genetic inhibition of each isoform with MAPK11-specific (si11) or MAPK14-specific (si14) siRNAs was performed separately (Fig. 5). Specific inhibition of mRNA MAPK11 and MAPK14 by the 2 validated siRNAs was previously confirmed by reverse transcriptase PCR in H1975, H1793, H1650, and H1651 cells.

thumbnail image

Figure 5. Cell viability after transfection with small interfering (si)RNAs targeting p38α (MAPK14) and p38β (MAPK11) is shown. H1975, H1793, H1650, and H1651 cells (A-D, respectively) were incubated either with siRNAs targeting p38α (MAPK14, Si14) or p38β (MAPK11, Si11) at 5 nM, 25 nM, and 50 nM, or with an siRNA-negative control at 50 nM (si control). Transfected and control cells with vehicle alone (control) were grown with or without cisplatin (CDDP) at 20 μM. The results of the WST-1 assay are expressed as percentage of vehicle-treated control (considered as 100%), and represent the mean ± standard error of the mean of at least 3 independent measurements, performed in triplicate. When there was no interaction between siRNA and cisplatin, the P value for the main effect of the corresponding siRNA (ie, independently of exposure to cisplatin) is indicated above the corresponding bars (over braces). When there was an interaction between siRNA and cisplatin, the means for each condition (with or without cisplatin) were compared with their respective si controls (with or without cisplatin) to separately assess the effect of siRNA in each of the 2 conditions. *P < .05 versus si control; **P < .01 versus si control.

Download figure to PowerPoint

si11 significantly reduced (P < .01 for each cell line) cell viability at 5 nM for H1975, H1650, and H1651 (47%, 36%, and 46%, respectively), and at 25 nM for H1793 (44%). There was dose-response activity with increasing concentrations of si11 for H1793 (R2 = 0.99, P = .003). The interaction test between si11 and cisplatin was significant for H1651 (P = .001), in which the effect of cisplatin was greater with 5 nM si11 compared with cisplatin plus the si control (P < .05); however, this response plateaued at higher concentrations of si11 (decrease in cell viability at 66%, 78%, 86%, and 85% at 0, 5, 25, and 50 nM si11, respectively).

si14 did not reduce the cell viability of H1975, H1793, and H1650, whereas it significantly reduced (P < .01) the cell viability of H1651 at 25 nM and at 50 nM (45% and 44%, respectively). The sensitivity of H1650 to cisplatin was significantly (P < .01) greater when cells were pretreated with si14 at 5 nM compared with when they were not pretreated, although the effect plateaued at higher si14 concentrations (reduction in cell viability of 49%, 70%, 67%, and 68% at 0, 5, 25, and 50 nM si14, respectively; P for the interaction = .02).

p38 MAPK Inhibition and ERCC1 Gene Expression

After treatment with SB202190, ERCC1 mRNA levels decreased significantly (P < .05) for every cell line compared with the control. The decreased mRNA levels occurred at 24 hours after treatment with SB202190 for H1793 and H1651, and at 48 hours for H1975 and H1650 (Fig. 6).

thumbnail image

Figure 6. ERCC1 mRNA expression after inhibition of p38 mitogen-activated protein kinase activity with SB202190 is shown. H1975, H1793, H1651, and H1650 (A-D, respectively) cells were incubated with 20 μM SB202190 or without SB202190 (control cells). Treated and control cells were grown without cisplatin. Cells were lysed for mRNA extraction at 24, 48, and 72 hours (H24, H48, and H72, respectively). The results of the reverse transcriptase quantitative polymerase chain reaction assays are expressed using untreated cultured cells as a reference, are arbitrarily marked as 1, and represent the mean ± standard error of the mean of at least 3 independent measurements, performed in triplicate. Significant differences are stated in the text.

Download figure to PowerPoint

The effects of si11 to target MAPK11 and si14 to target MAPK14 on ERCC1 mRNA levels were also recorded at 24, 48, and 72 hours. Transfection with si11 resulted in significantly decreased ERCC1 mRNA levels for every cell line after 24 hours (H1651), or 48 hours (H1975, H1793, H1650) depending on the cell line and the siRNA concentration. The data for 25 nM and 50 nM si11 are displayed in Figure 7 (data for 5 nM si11 are available on request). The decrease occurred at 24 hours after transfection with 50 nM si11, and at 48 hours with 5 nM and 25 nM si11 for H1651. It occurred at 48 hours with 5 nM si11 for H1650, and with 25 nM si11 for H1975 and H1793.

thumbnail image

Figure 7. ERCC1 mRNA expression after inhibition of p38 mitogen-activated protein kinase (MAPK) activity by transfection with small interfering (si)RNA targeting p38α (MAPK14), p38β (MAPK11), and ERCC1 is shown. H1975, H1793, H1650, and H1651 cells (A-D, respectively) were incubated either with siRNAs to target p38α (MAPK14, si14) or p38β (MAPK11, Si11) at 5 nM, 25 nM, and 50 nM, or with siRNA to target ERCC1 at 50 nM (siERCC1), or with a negative siRNA control at 50 nM (si control). Transfected and control cells with vehicle alone (control) were grown without cisplatin. Cells were lysed for mRNA extraction at 24, 48, and 72 hours (H24, H48, and H72, respectively). The results of the reverse transcriptase quantitative polymerase chain reaction assays are expressed using control si as a reference, are arbitrarily marked as 1, and represent the mean ± standard error of the mean of at least 3 independent measurements, performed in triplicate. Significant differences are stated in the text.

Download figure to PowerPoint

si14 had no significant effect on ERCC1 mRNA expression for H1975 and H1793. However, a significant decrease in ERCC1 mRNA levels occurred with 25 nM si14 at 24 hours for H1651, and at 48 hours for H1650.

As a control experiment, transfection with siRNA targeting ERCC1 (siERCC1) induced ERCC1 down-regulation in every cell line after 24 hours, which was maximal after 48 or 72 hours. The ERCC1 mRNA levels at 24 hours with siERCC1 were close to those recorded at the same time as down-regulation with SB202190 for H1793 and H1951, and with 50 nM of si11 for H1951. The reduction in cell viability at 5 days after transfection, which resulted in ERCC1 down-regulation, was similar to that observed at the same time after transfection with si11 at 25 nM for H1975, H1793, and H1651, and with si11 at 5 nM for H1650.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

The results of the present study provide evidence that p38 MAPK signaling can contribute positively to the viability of lung cancer cells derived from never or light smokers. The effect on cell viability was observed after pharmacological and genetic inhibition of p38 MAPK, although it varied in magnitude with the type of inhibitor, its concentration, and the cell line. Although every cell line was sensitive to low concentrations of mixtures of siRNAs that targeted both isoforms, the cell lines H1793 and H1651 were less sensitive to SB202190. Comparison between the effects of cisplatin and p38 MAPK inhibition may be informative, as cisplatin was used at a concentration that could discriminate between sensitive and resistant cell lines. In most experiments, the effect of p38 MAPK inhibition was equal to or superior to that observed in cells that were exposed to cisplatin alone. To our knowledge, this is the first report of the reliance of lung cancer cell survival on the p38 MAPK pathway.

SB202190 targets both MAPK11 and MAPK14.22 Interestingly, MAPK11 down-regulation reduced cell viability in every cell line, whereas a similar effect after MAP14 down-regulation was only recorded for H1651, suggesting that MAPK11 signaling is the main contributor to lung cancer cell survival in never smokers. Previous studies have reported that MAPK11 and MAPK14 may have opposite effects on cell differentiation, proliferation, and survival. In cardiomyocytes, Jurkat cells, and HeLa cells, MAPK14 induces apoptosis, whereas MAPK11 enhances survival.23-25 Other reports indicate that MAPK14 can positively regulate proliferation and enhance survival, for instance in hematopoietic cells and several cancer cell lines.26 The effect of MAPK14 activity seems to depend on cell-specific differences. Therefore, the role of MAP14 or the interplay between the MAPK11 and MAPK14 isoforms in favoring survival of particular lung cancer cells, such as H1651, should not be dismissed, although it seems at variance with the tumor-suppressing action that has been demonstrated for MAPK14 in mouse models.14, 27 The selectivity of the negative regulation of tumorigenesis may be an important aspect.28 MAPK14 antagonizes malignant transformations by N-Ras in fibroblasts or mutated K-RAS in colon cancer cells.29, 30 The proliferation of immature lung stem cells in MAPK14 knockout mice also facilitates K-RASG12V tumorigenesis.14 Of note, lung adenocarcinomas in patients who have never smoked infrequently harbor K-RAS mutations, which are common in smokers.31, 32

Cisplatin is a major cytotoxic drug that is used to treat lung cancer. However, not every patient with lung cancer benefits from cisplatin-based chemotherapy. Interestingly, in 2 cell lines (H1793 and H1651), pretreatment with SB202190 sensitized cells to cisplatin, that is, the effect of their combination was superior to that of their separate effects, indicating a synergy. As a result, the cisplatin-resistant cell line, H1793, became sensitive to cisplatin when pretreated with SB202190. In addition, sensitivity to cisplatin was higher after MAPK11 down-regulation of H1651, or after MAPK14 down-regulation of H1650. These observations have potential clinical relevance, as current treatments for lung cancer with cytotoxic drugs are only modestly efficacious. More studies are needed to validate the combination of p38 MAPK pathway inhibition and cisplatin, especially because in 2 other cell lines their combination has been inferior when the drugs are used separately. However, our findings should be validated by in vivo studies. Further efforts to uncover the molecular alterations associated with p38 MAPK-mediated sensitization to cisplatin are warranted.

Some p38 inhibitor drugs are currently under clinical assessment, such as ARRY-614 from Array BioPharma (Boulder, Colo; or ARRY-797, a potent inhibitor of p38α enzyme), which is being currently tested in hematological malignancies. In patients who are treated by surgery, expression of the ERCC1 protein in tumor cells has been shown to predict the benefit of cisplatin-based chemotherapy.18, 19 We have reported that activated p38 MAPK and ERCC1 are frequently expressed at high levels in lung cancer in those who had never smoked, which was a reason for exploring the consequences of p38 MAPK inhibition for ERCC1 expression in lung cancer cell lines derived from never smokers.16, 20 Here, the additional analysis showed that ERCC1 expression actually depended on p38 MAPK activation state, independently of smoking status. Although more frequent in tumors from nonsmokers, activated p38 MAPK expression is not specific to smoking status.

The ERCC1 gene contains AP-1 sites that are bound by the transcription factors JUN and ATF2.33 In breast cancer cells, Hayakawa et al reported that JNK was required to activate the transcription of several DNA damage repair genes, including ERCC1, ERCC3, MSH2, XPA, and XPC, in response to cisplatin.34 In hepatocarcinoma cells, however, Andrieux et al reported that ERCC1 was up-regulated during proliferative signals in response to stimulation with epidermal growth factor, which involved ERK activity and binding of the transcription factor GATA-1.35 We examined whether ERCC1 expression could be down-regulated by inhibiting MAPK p38 signaling. Our results suggest that ERCC1 mRNA levels can be regulated by p38 MAPK, notably by MAPK11 in cells from never or light smokers. However, we could not demonstrate clear modifications in ERCC1 protein levels. Furthermore, down-regulation of ERCC1 mRNA expression with ERCC1-specific siRNA was sufficient to reduce cell viability. Thus, down-regulation of ERCC1 expression may account, in part, for the effect of p38 MAPK inhibition on cell viability. Crosstalk between p38 and JNK pathways should be considered, as both MAPK pathways share several upstream regulators, and multiple stimuli simultaneously activate both pathways. The mechanistic basis of the relation between p38 MAPK signaling and sensitivity to cisplatin remains unknown.

In conclusion, the results of our study support the hypothesis that lung cancer cells derived from never or light smokers rely on p38 MAPK signaling for their survival. Inhibition of p38 MAPK can also synergize with cisplatin in different models and results in decreased ERCC1 mRNA expression. These findings highlight the potential of focused analysis of the cellular and molecular characteristics of lung cancers in nonsmokers to uncover new therapies for lung cancer.

FUNDING SOURCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

Financial support was received from INCa Projet Libre 2006 P086.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES