ERK phosphorylation predicts synergism between gemcitabine and the epidermal growth factor receptor inhibitor AG1478

Lung Cancer 73 (2011) 274–282

Contents lists available at ScienceDirect

Lung Cancer journal homepage: www.elsevier.com/locate/lungcan

ERK phosphorylation predicts synergism between gemcitabine and the epidermal growth factor receptor inhibitor AG1478 Peter P. Luk a,b,∗ , Peter Galettis a,b , Matthew Links a,b a b

Cancer Pharmacology and Therapeutic Laboratory, Medical Oncology, St. George Hospital, Sydney, NSW, Australia St. George Clinical School, University of New South Wales, Sydney, NSW, Australia

a r t i c l e

i n f o

Article history: Received 2 September 2010 Received in revised form 29 October 2010 Accepted 18 December 2010 Keywords: Lung cancer Gemcitabine EGFR Synergy ERK

a b s t r a c t Background: Clinical trials combining epidermal growth factor receptor (EGFR) inhibitors with gemcitabine-based chemotherapy in non-small cell lung cancer (NSCLC) have not produced a survival advantage. This may be caused by antagonism between the two drugs or mutations that promote such, possibly RAS mutation. Furthermore, ERK, a critical growth regulator downstream of RAS, may play a role. This study aimed to explore the relationship between ERK, synergy/antagonism and cell cycle arrest in combination treatment. Methods: A549 (mutant KRAS), H322 (wildtype KRAS) and siRNA-mediated KRAS knockdown A549 were treated with gemcitabine and/or the EGFR inhibitor AG1478 and analyzed with median effect analysis. Cell cycle distribution and ERK phosphorylation were assessed using flow cytometry and ELISA, respectively. Effect on cytotoxicity after ERK inhibition by U0126 was also assessed. Results: Cytotoxic interaction was dose dependent with antagonism at high dose AG1478. G1 arrest was observed with both high dose AG1478 and high dose gemcitabine and therefore was inconsistently associated with antagonism. Furthermore, ERK phosphorylation was increased by gemcitabine and its suppression by AG1478 was related to antagonism particularly in H322. ERK’s effect in antagonism was further confirmed by using U0126. Greater antagonism was observed in the KRAS mutant cell line and KRAS knockdown by siRNA resulted in increased sensitivity to AG1478 as well as combination treatment. Conclusion: Our findings are consistent with a model in which ERK phosphorylation favors synergy and the outcome depends on the balance between gemcitabine-induced and AG1478-inhibited ERK phosphorylation. KRAS mutation confers resistance to AG1478 as well as combination treatment. © 2011 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Anti-cancer drugs are commonly given in combination with the aim of increasing efficacy and overcoming resistance. The development of targeted therapy further allows selection of therapy for individual patients based on tumor genotype and clinical evidence [1]. The combination between epidermal growth factor receptor (EGFR) inhibitors and gemcitabine (as well as other chemotherapy agents) has been intensely studied in the treatment of non-small cell lung cancer (NSCLC) and pancreatic cancer.

Abbreviations: CI, combination index; EGFR, epidermal growth factor receptor; ELISA, enzyme-linked immunosorbent assay; NSCLC, non-small cell lung cancer; SRB, sulforhodamine B; TKI, tyrosine kinase inhibitor. ∗ Corresponding author at: Ground Floor, WR Pitney Building, St. George Hospital, Gray St, Kogarah, NSW 2217, Australia. Tel.: +61 2 91132434; fax: +61 2 91132960. E-mail addresses: [email protected], [email protected] (P.P. Luk). 0169-5002/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.lungcan.2010.12.016

EGFR plays a key role in promoting oncogenesis by activating downstream pathways such as RAS and phosphoinositide 3-kinase (PI3K) [2]. EGFR tyrosine kinase inhibitors (TKI) such as erlotinib (Tarceva) and gefitinib (Iressa) block EGFR autophosphorylation and downstream activation. EGFR TKIs have been shown to have antiproliferative and proapoptotic effect in vitro [3,4]. Furthermore, erlotinib improves survival in NSCLC as shown by the BR.21 phase III clinical trial [5]. Gemcitabine (Gemzar, dFdC) is a deoxycytidine analog which is incorporated during DNA synthesis resulting in inhibition of further DNA synthesis and apoptosis [6]. A phase III trial (TALENT) combining erlotinib with the cisplatin/gemcitabine doublet (as well as a similar trial (INTACT1) using gefitinib) failed to demonstrate an improvement in survival in patients with advanced NSCLC [7,8]. Similarly negative results were seen in phase III trials combining erlotinib (TRIBUTE) or gefitinib (INTACT2) with the carboplatin/paclitaxel doublet [9,10]. The lack of benefit when adding an agent with a demonstrated increase in survival suggested clinical antagonism. Indeed, there is in vitro evidence suggesting that G1 arrest caused by EGFR inhibitors antagonize the S and M phase effects of chemotherapy

P.P. Luk et al. / Lung Cancer 73 (2011) 274–282

[11]. This is the currently accepted model of antagonism which has led to clinical trials exploring pharmacodynamic separation (i.e. sequential treatment rather than simultaneous treatment) [12]. Another hypothesis for the lack of observed benefit is that molecular heterogeneity masked the benefit in a small subgroup [13]. The TALENT and TRIBUTE trials both showed that the chemotherapy plus erlotinib combination increased overall survival in the never-smoker subpopulation [7]. Although molecular factors such as activation of EGFR (mutation, overexpression or amplification) have been associated with response to single agent EGFR TKI, clinical studies have not identified the molecular characteristics of tumors that clearly benefit from combination therapy [10,14,15]. A similar situation exists for pancreatic cancer in which although there is a very small survival advantage with the combination of erlotinib and gemcitabine, there is a small subgroup who achieved a longer term benefit [16]. While there is a great deal of interest in characterizing molecular determinants of response to the single agents, little data exist concerning molecular determinants of a synergistic or antagonistic interaction. RAS is a GTPase that functions downstream of EGFR and the mutation of one isoform, KRAS, causes constitutive downstream activation of the ERK pathway and leads to resistance to EGFR inhibitors in vitro [17] as well as clinically [18]. Furthermore, Eberhard et al. [14] found that KRAS mutation confers poorer clinical outcome when treated with combined chemotherapy and erlotinib. Therefore the effect of KRAS mutation on combination therapy is worth further investigation. Although ERK is generally associated with proliferation and survival signaling, intriguingly it has also been reported to promote gemcitabine-induced apoptosis [19]. Therefore ERK appears to have dual roles and may act as a molecular switch that affects synergy/antagonism by determining whether the cell survives or dies. Exploring the changes in ERK signaling in combination therapy will also improve our fundamental understanding of the interaction between DNA damage and EGFR pathways. AG1478 is a quinazoline with a similar chemical structure to erlotinib and gefitinib. It also has the same mechanism of action which is inhibition of EGFR phosphorylation through competing with ATP binding [20,21]. According to enzyme inhibition data, AG1478 is at least as selective for EGFR as clinically approved EGFR TKIs [22,23]. Therefore models using AG1478 should be applicable to clinical EGFR TKIs. AG1478 has been used by other preclinical studies as models of EGFR inhibition [24,25]. The objective of the current study was to explore the determinants of the in vitro interaction between gemcitabine and the EGFR inhibitor AG1478. We examined ERK phosphorylation, cell cycle changes and KRAS mutation in relation to cytotoxic interaction. Furthermore, the MEK inhibitor U0126 and KRAS siRNA were used to inhibit ERK and KRAS function respectively to assess their effect on synergy.

2. Methods

275

2.2. Mutation analysis of KRAS gene Genomic DNA was isolated from each cell line using DNeasy kit (Qiagen). The primers for amplification of KRAS exon 2 (containing codons 1–37) were 5 -TTCTTAAGCGTCGATGGAGGAG-3 and 5 -GGACCCTGACATACTCCCAAGG-3 (Invitrogen). After 35 cycles of PCR amplification, PCR products were purified by QIAqiuck PCR Purification Kit (Qiagen) and sequenced at the University of New South Wales Ramaciotti Centre for Gene Function Analysis using an ABI 3730 Capillary Sequencer (Applied Biosystems).

2.3. Growth inhibition studies by SRB assay Exponentially growing cells were seeded in 96-well flat bottom plates: 2000 cells/well for A549 and 7000 cells/well for H322. After 24 h of incubation, drugs were added and cells were cultured for another 72 h. Cells were treated with gemcitabine alone, AG1478 alone or their combination simultaneously. At the end of the culture period, cell growth was assessed using the sulforhodamine B (SRB) assay as previously described [26]. Values were normalized against untreated controls. Cytotoxic interaction between AG1478 and gemcitabine were analyzed by median effect analysis [27] using Calcusyn software (Biosoft, Inc.) and expressed as combination index (CI): CI < 1 indicates synergism, CI = 1 indicates additivity and CI > 1 indicates antagonism.

2.4. Analysis of cell cycle by flow cytometry Exponentially growing cells were seeded in 25 cm2 culture flasks and treated with AG1478 and gemcitabine alone or in combination simultaneously. Cells were trypsinized and fixed with 70% cold ethanol at −20 ◦ C overnight. The fixed cells were washed with PBS and stained with 50 ␮g/ml propidium iodide and 10 ␮g/ml RNase A for at least 1 h at room temperature. Analyses of 30,000 events were acquired on a BD LSR flow cytometer (Becton Dickinson Biosciences) and cell cycle distribution was analyzed using FlowJo 7.2.5 (Tree Star, Inc.).

2.5. Enzyme-linked immunosorbent assay (ELISA) Phospho-ERK and total-ERK were detected using Cellular Activation of Signaling ELISA kits (SuperArray Bioscience Corporation). Briefly, cells were cultured in 96-well plates and treated with gemcitabine and AG1478 alone or simultaneously for 72 h. The cells were fixed with 4% formaldehyde and then processed according to manufacturer’s instructions. Absorbance was read at 450 nm. After that, wells were stained with Cell Staining Buffer, solubilized with 1% SDS and absorbance read at 595 nm. This allowed normalization of readings against differing cell density. Phosphorylated ERK relative to total ERK protein for each treatment condition was calculated by the formula: (phospho-ERK OD450 /OD595 )/(total-ERK OD450 /OD595 ).

2.1. Cell culture, experimental reagents and chemicals

2.6. siRNA and transfections

NSCLC cell line A549 was obtained from the American Type Culture Collection while NCI-H322 (abbreviated to H322) was obtained from the European Collection of Cell Cultures. All cell lines were grown in RPMI 1640 supplemented with 10% fetal calf serum, 2 mM l-glutamine, 100 IU/ml pencillin and 100 ␮g/ml streptomycin. AG1478 was purchased from Sigma, gemcitabine was a gift from Eli Lilly and U0126 was purchased from Selleck Chemicals.

KRAS siRNA and negative control siRNA were purchased from Ambion. The optimal cell density for A549 and siRNA concentration were determined in pilot studies. 1.6 × 105 cells in 25 cm2 culture flasks or 2000 cells/well in 96-well plates were transfected with negative control siRNA or KRAS siRNA at a final concentration of 5 nM using Lipofectamine 2000 (Invitrogen). Knockdown of KRAS protein was analyzed at day 4 after transfection by immunoblotting.

276

P.P. Luk et al. / Lung Cancer 73 (2011) 274–282

Fig. 1. CI plots depicting the cytotoxic interaction between AG1478 and gemcitabine for 72 h simultaneous treatment for (A) A549 and (B) H322. Each point is shown as the mean ± SD (n = 3). CI < 1 indicates synergism, CI = 1 indicates additivity and CI > 1 indicates antagonism.

2.7. Immunoblotting Whole cell lysates were denatured in sample buffer containing SDS and equal amounts of total protein from each sample (50 ␮g) were separated by 12% SDS-PAGE followed by transfer to PVDF membranes. After blocking with 5% non-fat milk, membranes were probed with primary antibodies overnight at 4 ◦ C and then with appropriate HRP-conjugated secondary antibodies. Antibodies used were: anti-phospho-ERK, anti-ERK, anti-␤-actin (all from Cell Signaling) and anti-KRAS (Santa Cruz). Secondary antibodies were obtained from both Cell Signaling and Santa Cruz. Detection was performed using enhanced chemiluminescence reagent (Amersham). Films were scanned into .tif format using a Canon CanoScan LiDE 70 flatbed scanner and blots were quantitated in Adobe Photoshop following the method outlined at http://www.lukemiller.org/journal/2007/08/quantifyingwestern-blots-without.html. 2.8. ERK inhibition by U0126 Cells were treated for 72 h with gemcitabine/AG1478 combination with or without 2.5 ␮M U0126 and then assessed using SRB assay. This was used to calculate 3-way CI. ERK phosphorylation was assessed using immunoblotting. 2.9. Statistical analysis Statistics was evaluated using SPSS 17 (SPSS Inc.). One-way ANOVA was used to assess single drug effect while two-way ANOVA was used to determine interaction between two drugs. Differences at a level of p < 0.05 were considered statistically significant. Pearson’s correlation coefficient was used to assess correlation. 3. Results 3.1. Molecular characteristics and cytotoxicity of AG1478 and gemcitabine in human NSCLC cell lines Both A549 and H322 have previously been reported to have wildtype EGFR, while A549 harbors a KRAS mutation [28]. To confirm KRAS mutation in our A549 cell line, direct DNA sequencing of

KRAS exon 2 was performed. Wildtype sequence was detected in H322 while A549 had a point mutation (GGT to AGT) at codon 12, thus confirming its G12S mutation status. Clear differences were observed in the sensitivity for both drugs in the cell lines: A549 was sensitive to gemcitabine (IC50 = 4.9 nM) but H322 was resistant (IC50 = 51.6 ␮M) on the other hand A549 was resistant to AG1478 (IC50 = 13.0 ␮M) but H322 was sensitive (IC50 = 0.8 ␮M) (Supplementary Fig. I). 3.2. Interaction between AG1478 and gemcitabine is dose dependent To evaluate the cytotoxic interaction between AG1478 and gemcitabine, the growth inhibition of combining varying doses of AG1478 and gemcitabine was analyzed using median effect analysis. The tested gemcitabine concentration range for H322 was much greater than A549 due to their differing sensitivity. The cytotoxic interaction was concentration dependent in both cell lines (Fig. 1). In A549 (Fig. 1A), 0.5 ␮M AG1478 was synergistic but increasing AG1478 concentration resulted in increase in CI values with the greatest antagonism seen at 10 ␮M. On the other hand, increasing gemcitabine concentration resulted in convergence towards additivity. In H322 (Fig. 1B), 0.5 ␮M and 2.5 ␮M AG1478 were synergistic but 5 ␮M and 10 ␮M were antagonistic. Increasing gemcitabine concentration changed this antagonism to synergism. Overall, the KRAS wildtype cell line H322 showed a more synergistic picture than the KRAS mutant cell line A549. 3.3. Cell cycle change is dose dependent Cell cycle distribution was analyzed by flow cytometry after single agent and concurrent drug treatment. In A549 cells (Fig. 2A), 0.5 ␮M (IC20 ), 2.5 ␮M (IC20 ), 5 ␮M (IC25 ) and 10 ␮M (IC40 ) AG1478 only caused minor increase in G1 cells (p = 0.08). 5 nM (IC50 ) gemcitabine also had little effect. However, 10 nM (IC80 ) induced a significant reduction in G1 cells (p < 0.01) with accumulation of cells in S and G2/M phase, as well as an increase in sub G1 cells. Simultaneously treating A549 with a constant dose of 10 nM gemcitabine but varying AG1478 concentrations resulted in attenuation of gemcitabine-induced S phase arrest, with statistically significant interaction (p < 0.01). In contrast, adding AG1478 to 5 nM

P.P. Luk et al. / Lung Cancer 73 (2011) 274–282

277

Fig. 2. Cell cycle distribution following single agent as well as combination treatment in (A) A549 (B) H322. Columns represent mean (n = 3), error bars represent SD.

gemcitabine did not change cell cycle distribution. Similarly, 2.5 nM (IC20 ) gemcitabine had no effect (data not shown). Treating H322 cells (Fig. 2B) with 0.5 ␮M (IC30 ), 2.5 ␮M (IC60 ), 5 ␮M (IC65 ) and 10 ␮M (IC70 ) of AG1478 induced a significant dosedependent increase in G1 cells in H322 (p < 0.01) with reduction in S and G2/M phases. In contrast, low dose gemcitabine (5 nM, IC10 ) induced S phase arrest in H322 (p < 0.01) while 50 ␮M (IC50 ) induced G1 arrest and increase in sub G1 cells. Adding AG1478 simultaneously to 5 nM gemcitabine significantly attenuated the gemcitabine-induced S arrest (p < 0.01) to a greater extent compared to A549. With 50 ␮M gemcitabine, gemcitabine-induced G1 arrest was unchanged by AG1478 and the same was seen with 500 nM (IC30 ) (data not shown). 3.4. Antagonism is inconsistently related to G1 cell cycle arrest In A549, a positive correlation was seen between G1 percentage and CI at 10 nM gemcitabine (r = 0.99), in other words, antagonism was associated with G1 arrest. At 5 nM gemcitabine, changes in CI values were associated with little change in percentage of G1 cells, suggesting that cell cycle change does not explain the observed synergism or antagonism under these conditions. In H322, a positive correlation also exists at 5 nM gemcitabine, albeit less strongly (r = 0.78). At higher gemcitabine concentration, this relationship was absent as G1 arrest was seen regardless of AG1478 concentration (Supplementary Fig. II). 3.5. Gemcitabine upregulates phosphorylation of ERK which is correlated with synergy ERK is activated by phosphorylation as a downstream event of EGFR activation. The changes in phosphorylation of ERK were investigated with ELISA. In A549 (Fig. 3A), gemcitabine at

2.5 nM and 5 nM had no effect while 10 nM markedly increased ERK phosphorylation. Single agent AG1478 caused a minor reduction in ERK phosphorylation which was also seen when combined with 2.5 nM or 5 nM gemcitabine. A more noticeable reduction in ERK phosphorylation was seen when AG1478 was combined with 10 nM gemcitabine. In H322 (Fig. 3B), treatment with AG1478 caused a dose dependent inhibition of ERK phosphorylation. On the other hand, gemcitabine induced a dose-dependent increase in ERK phosphorylation. In combination treatment, AG1478 attenuated the gemcitabine-induced ERK phosphorylation in a dose-dependent manner at a greater extent than observed in A549. The relationship between CI and ERK phosphorylation was explored. In A549 (Fig. 3C), there was an association between synergism and ERK phosphorylation at 10 nM gemcitabine but not for lower gemcitabine doses due to the unchanged ERK phosphorylation profile at these doses. In H322 (Fig. 3D), a clearer relationship between synergism and ERK phosphorylation was observed across all gemcitabine concentrations. This basically reflects that increasing AG1478 dose reduces ERK phosphorylation and results in antagonism at the same time. 3.6. Inhibition of ERK phosphorylation attenuates gemcitabine cytotoxicity The MEK inhibitor U0126 was used to study the effect of inhibiting gemcitabine-induced ERK phosphorylation on cytotoxicity. In A549, both U0126 and the combination of U0126 with 0.5 ␮M AG1478 effectively inhibited gemcitabine-induced ERK phosphorylation. However, adding U0126 to 10 ␮M AG1478 did not appear to have further effect on gemcitabine-induced ERK phosphorylation. Furthermore, the combination of U0126 and AG1478 did not affect baseline ERK phosphorylation. In H322, 2.5 ␮M U0126

278

P.P. Luk et al. / Lung Cancer 73 (2011) 274–282

Fig. 3. Phosphorylation of ERK following combination treatment in (A) A549 and (B) H322. Y-Axis is the phosphorylated versus total ERK ratio calculated as described in methods. ERK phosphorylation is further plotted against CI in (C) A549 and (D) H322. Each set of data points represents a constant gemcitabine concentration combined with variable AG1478 concentration. Each point represents the mean (n = 3) and error bars represent SD.

prevented gemcitabine-induced upregulation of ERK phosphorylation but adding U0126 to AG1478 did not have a further effect. U0126 and AG1478 together were more effective at inhibiting gemcitabine-induced ERK phosphorylation than adding AG1478 only to gemcitabine at both doses (Supplementary Fig. III). When 2.5 ␮M U0126 was added to the gemcitabine/AG1478 combination in A549 (Fig. 4A), reduction in cytotoxicity was observed for 0.5 ␮M AG1478 but this effect became less apparent with increasing AG1478 concentration. This is represented as increased CI (greater antagonism) at low concentrations of AG1478 (Fig. 4B). Similarly in H322 (Fig. 4C), addition of U0126 attenuated cytotoxicity at 0.5 ␮M and 2.5 ␮M AG1478 but this effect was lost at higher AG1478 concentrations. This was seen as reduced synergism on the CI curve (Fig. 4D). The reduced synergism was not associated with increased G1 arrest (Supplementary Fig. IV). 3.7. KRAS knockdown increased cytotoxicity but not synergism KRAS protein level was assessed with immunoblotting after 96h siRNA treatment (Fig. 5A). The double-band appearance is due to non-prenylated RAS running slower than the prenylated form on SDS-PAGE [29]. KRAS protein level was observed to be decreased following siRNA treatment compared to controls as indicated by ␤-actin loading control.

KRAS knockdown did not significantly affect sensitivity to gemcitabine (IC50 5.8 ± 1.0 nM for negative control versus 5.4 ± 0.9 nM for KRAS knockdown; p = 0.74). In contrast, KRAS knockdown restored AG1478 sensitivity with reduction of IC50 from 14.4 ± 2.2 ␮M for negative control to 5.7 ± 2.1 ␮M for KRAS knockdown (p < 0.01). A constant gemcitabine concentration of 10 nM was combined with varying AG1478 concentrations with and without KRAS knockdown (Fig. 5B). In both negative control and KRAS knockdown, adding 0.5 ␮M AG1478 to gemcitabine increased cytotoxicity while increasing the AG1478 concentration to 10 ␮M gradually reduced cytotoxicity. Although higher cytotoxicity was observed with KRAS knockdown throughout all AG1478 doses, the cytotoxic interaction was almost unchanged. 0.5 ␮M AG1478 resulted in synergism which gradually became antagonistic as the AG1478 dose was increased to 10 ␮M regardless of KRAS knockdown (Fig. 5C).

4. Discussion In this study, we showed that the interaction between gemcitabine and AG1478 can be either synergistic or antagonistic and that the concentration of the drugs and the characteristics

P.P. Luk et al. / Lung Cancer 73 (2011) 274–282

279

Fig. 4. Dose response curves and cytotoxic CI plots for A549 in 72 h combined treatment of gemcitacine, AG1478 and U0126 are shown in (A) and (B), respectively. These are also shown for H322 in (C) and (D). Gemcitabine concentration was set constant at 10 nM for A549 and 500 nM for H322. U0126 concentration was 2.5 ␮M and AG1478 concentrations as indicated on graph. Each point is shown as the mean ± SD (n = 3).

of the cells are important determinants of synergy. Our data also challenges the prevailing model that antagonism is primarily determined by cell cycle arrest preventing S-phase associated cell death with gemcitabine [30]. It identifies the balance between upregulation of phospho-ERK by gemcitabine and down-regulation by EGFR inhibition as a critical determinant of the interaction. The finding of a survival advantage with erlotinib for refractory disease [5] and the lack of such an advantage in combination therapy with EGFR inhibitors [7–10] strongly suggest antagonism is clinically important. It has been suggested that G1 arrest caused by EGFR inhibition prevents the cytotoxicity of S phase specific drugs [11,28,31,32]. This has led to attempts to overcome this problem by temporal or pharmacodynamic separation [33,34]. We therefore examined the relationship between cell cycle arrest and synergy in our model. We found that G1 arrest correlated inconsistently with antagonism and that the effect was concentration dependent. With some concentrations there was a clear linear relationship and with others there was significant variation in the combination index with no change in cell cycle. This demonstrates that the prevailing model of cell cycle dependent antagonism is insufficient to explain synergy or antagonism. We further explored downstream signaling and showed, for the first time, that synergism is related to phosphorylation of ERK in combination therapy. Two isoforms of ERK, ERK1 and ERK2, are phosphorylated by MEK1/2 which is in turn phosphorylated by Raf. Raf is a direct effector of RAS. Mutation of KRAS renders it

constitutively active due to insensitivity to inactivation by GTPase activation proteins and leads to aberrant downstream signaling implicated in oncogenesis [35]. ERK is generally thought of as a survival pathway [36] and its downstream effectors mediate cell growth and promotes G1/S transition [37]. EGFR inhibitors downregulate phosphorylation of ERK as well as other downstream effectors such as AKT [17] which leads to both antiproliferative and proapoptotic effects. The present study showed that EGFR inhibition results in suppression of ERK phosphorylation in KRAS wildtype but not KRAS mutant cell line, a finding which is supported by others [17]. Interestingly, gemcitabine induced ERK phosphorylation in both KRAS wildtype and mutant cell line. Suppressing ERK activation using U0126 resulted in reduced cytotoxicity and increased antagonism, though a plateau effect was seen as the effect of U0126 was minimal with high dose AG1478. This indicates that ERK plays a role in mediating gemcitabine cytotoxicity and that AG1478 causes antagonism through its inhibitory effect on ERK. Others have also reported that gemcitabine induces ERK phosphorylation in breast [38] and NSCLC cell lines [19]. ERK plays a role in mediating apoptosis since its pharmacological inhibition was found to attenuate Bcl-2 downregulation and cytochrome c release [19]. Phosphorylation of ERK has been reported to be upregulated by other chemotherapy agents including cisplatin [39], paclitaxel [40] and etoposide [41], indicating that it is not simply an effect of DNA damage but rather a more general effect of chemotherapy-induced cell death. The molecu-

280

P.P. Luk et al. / Lung Cancer 73 (2011) 274–282

Fig. 5. (A) Immunoblot for KRAS protein using ␤-actin as loading control. (B) Dose response curves for 72 h of gemcitacine (10 nM) combined with AG1478 in A549 with and without KRAS knockdown. (C) CI plot. MW, molecular weight standards.

lar mechanism for gemcitabine-induced ERK phosphorylation is unclear, however our current understanding is that under these conditions ERK plays a role in promoting cell death via the mitochondrial pathway [19]. Furthermore, ERK is also known to mediate non-apoptotic programmed cell death via neurokinin-1 receptor [42]. This data is best explained by a model (Fig. 6) where there are phospho-ERK dependent and independent pathways for both gemcitabine and AG1478. Gemcitabine induced cell death is primarily a result of direct incorporation into DNA and termination of replication and DNA repair [6] supplemented by indirect effects on nucleotide synthesis via ribonucleotide reductase [43]. Phospho-ERK fulfills a secondary role and indeed inhibition of ERK phosphorylation did not completely reverse cytotoxicity, indicating other ERK independent pathways at work. EGFR inhibitors such as AG1478 inhibit downstream EGFR signaling which appears to cross over gemcitabine-related death pathways. Our data suggests that synergy is primarily determined by the balance between phospho-ERK upregulation by gemcitabine and downregulation by EGFR inhibition. Interestingly, the persistent ERK activation caused by KRAS mutation did not result in increased synergy possibly due

to RAS activation of offsetting survival pathways which need to be further investigated. This study highlights the need for further work in elucidating the function of ERK as molecular switch determining death or survival of the cell ERK’s potential ability to act as a molecular switch that can determine cell survival or death. There is growing evidence that the function of ERK is determined by subcellular localization. Nuclear translocation is essential for its proliferative signals, however scaffold proteins such as PEA-15 and Sef can cause cytosolic retention of ERK. Furthermore, cytosolic ERK potentiates the proapoptotic effects of proteins such as death associated protein kinase [44]. Further experiments to study ERK localization post gemcitabine treatment using immunohistochemistry would shed light on this aspect. Apart from antagonism, unrecognized molecular heterogeneity can also result in overall negative clinical trials in which the benefit in one molecular subtype is negated by a lack of benefit (or a detrimental effect) in another [13]. An important subgroup of patients is those with KRAS mutations. There is increasing evidence that KRAS mutation is a poor predictive factor for single agent EGFR inhibitor therapy [18,45–48]. Therefore the next important ques-

P.P. Luk et al. / Lung Cancer 73 (2011) 274–282

281

Fig. 6. Schematic diagrams representing a possible model of interaction between gemcitabine and EGFR inhibition. (A) High dose gemcitabine favoring synergism and (B) high dose EGFR inhibitor favoring antagonism. + represents activation while − represents inhibition.

tion will be whether KRAS affects cytotoxic interaction between EGFR inhibitor and chemotherapy. The present study demonstrated that KRAS mutation does not preclude synergy but the effect is more pronounced in the KRAS wildtype. Furthermore, the restoration of sensitivity to AG1478 by KRAS knockdown is consistent with the study by Uchida et al. [49], who found that transfection of mutant KRAS resulted in resistance to gefitinib. Although this led to greater combination cytotoxicity, synergy was not increased, indicating that the factor affecting cytotoxic interaction is, in fact, not as simple as a single gene. Nevertheless, this provides further rationale for focusing combination therapy on KRAS wildtype patients. This also raises the question regarding whether other predictive markers of EGFR TKI could affect combination therapy. For example, EGFR mutant cells are highly sensitive to EGFR TKI [50] and it would be interesting to investigate whether this has an effect on synergy. A limitation of the current study is the use of only two cell lines due to limited time frame and manpower. Indeed, performing a screening study involving a large number of cell lines would better correlate KRAS mutation with antagonism and minimize the effect of unknown cell characteristics. However, the objective of the current study was to establish a model of synergy for further investigation and therefore emphasis was placed on the siRNA model rather than a cell line panel. Future studies will need to include multiple cell lines for comparison as well as other chemotherapy agents such as paclitaxel, irinotecan and vinorelbine. In conclusion, this study has confirmed that KRAS wildtype would be more beneficial for combined EGFR inhibitor and gemcitabine treatment. More importantly, this study offers novel insight into the molecular interaction between EGFR TKI and gemcitabine with the finding that ERK plays a role in determining the fate of the cell following gemcitabine treatment, including cytotoxic interaction with EGFR inhibitor. Our understanding of cell signaling and cell death pathways continues to increase rapidly, allowing the discovery of novel therapeutic approaches to improve patient outcomes. These strategies will most likely be combined with conventional chemotherapy and optimizing these treatment

modalities based on lessons learnt from the past will no doubt pave the way to individualized therapy in oncology. Conflict of interest None declared. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.lungcan.2010.12.016. References [1] Decker S, Sausville EA. Preclinical modeling of combination treatments: fantasy or requirement? Ann N Y Acad Sci 2005;1059:61–9. [2] Yarden Y, Sliwkowski MX. Untangling the ErbB signalling network. Nat Rev Mol Cell Biol 2001;2:127–37. [3] Moyer JD, Barbacci EG, Iwata KK, Arnold L, Boman B, Cunningham A, et al. Induction of apoptosis and cell cycle arrest by CP-358,774, an inhibitor of epidermal growth factor receptor tyrosine kinase. Cancer Res 1997;57:4838–48. [4] Ciardiello F, Caputo R, Bianco R, Damiano V, Pomatico G, De Placido S, et al. Antitumor effect and potentiation of cytotoxic drugs activity in human cancer cells by ZD-1839 (Iressa), an epidermal growth factor receptor-selective tyrosine kinase inhibitor. Clin Cancer Res 2000;6:2053–63. [5] Shepherd FA, Rodrigues Pereira J, Ciuleanu T, Tan EH, Hirsh V, Thongprasert S, et al. Erlotinib in previously treated non-small-cell lung cancer. N Engl J Med 2005;353:123–32. [6] Plunkett W, Huang P, Xu YZ, Heinemann V, Grunewald R, Gandhi V. Gemcitabine: metabolism, mechanisms of action, and self-potentiation. Semin Oncol 1995;22:3–10. [7] Gatzemeier U, Pluzanska A, Szczesna A, Kaukel E, Roubec J, De Rosa F, et al. Phase III study of erlotinib in combination with cisplatin and gemcitabine in advanced non-small-cell lung cancer: the tarceva lung cancer investigation trial. J Clin Oncol 2007;25:1545–52. [8] Giaccone G, Herbst RS, Manegold C, Scagliotti G, Rosell R, Miller V, et al. Gefitinib in combination with gemcitabine and cisplatin in advanced non-small-cell lung cancer: a phase III trial–INTACT 1. J Clin Oncol 2004;22:777–84. [9] Herbst RS, Giaccone G, Schiller JH, Natale RB, Miller V, Manegold C, et al. Gefitinib in combination with paclitaxel and carboplatin in advanced nonsmall-cell lung cancer: a phase III trial–INTACT 2. J Clin Oncol 2004;22: 785–94. [10] Bell DW, Lynch TJ, Haserlat SM, Harris PL, Okimoto RA, Brannigan BW, et al. Epidermal growth factor receptor mutations and gene amplification in non-

282

[11]

[12]

[13] [14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23] [24]

[25]

[26]

[27]

[28]

[29]

[30]

P.P. Luk et al. / Lung Cancer 73 (2011) 274–282 small-cell lung cancer: molecular analysis of the IDEAL/INTACT gefitinib trials. J Clin Oncol 2005;23:8081–92. Chun PY, Feng FY, Scheurer AM, Davis MA, Lawrence TS, Nyati MK. Synergistic effects of gemcitabine and gefitinib in the treatment of head and neck carcinoma. Cancer Res 2006;66:981–8. Gandara DR, Davies AM, Gautschi O, Mack PC, Lau DH, Lara Jr PN, et al. Epidermal growth factor receptor inhibitors plus chemotherapy in non-small-cell lung cancer: biologic rationale for combination strategies. Clin Lung Cancer 2007;8(Suppl. 2):S61–67. Betensky RA, Louis DN, Cairncross JG. Influence of unrecognized molecular heterogeneity on randomized clinical trials. J Clin Oncol 2002;20:2495–9. Eberhard DA, Johnson BE, Amler LC, Goddard AD, Heldens SL, Herbst RS, et al. Mutations in the epidermal growth factor receptor and in KRAS are predictive and prognostic indicators in patients with non-small-cell lung cancer treated with chemotherapy alone and in combination with erlotinib. J Clin Oncol 2005;23:5900–9. Gatzemeier U, Heller A, Foernzler D, Moecks J, Ward C, Rosa Fd, et al. Exploratory analyses EGFR, kRAS mutations and other molecular markers in tumors of NSCLC patients (pts) treated with chemotherapy +/−erlotinib (TALENT). J Clin Oncol 2005;23(Suppl.):abstr 7028. Moore MJ, Goldstein D, Hamm J, Figer A, Hecht JR, Gallinger S, et al. Erlotinib plus gemcitabine compared with gemcitabine alone in patients with advanced pancreatic cancer: a phase III trial of the National Cancer Institute of Canada Clinical Trials Group. J Clin Oncol 2007;25:1960–6. Janmaat ML, Rodriguez JA, Gallegos-Ruiz M, Kruyt FA, Giaccone G. Enhanced cytotoxicity induced by gefitinib and specific inhibitors of the Ras or phosphatidyl inositol-3 kinase pathways in non-small cell lung cancer cells. Int J Cancer 2006;118:209–14. Zhu CQ, da Cunha Santos G, Ding K, Sakurada A, Cutz JC, Liu N, et al. Role of KRAS and EGFR as biomarkers of response to erlotinib in national cancer institute of canada clinical trials group study BR.21. J Clin Oncol 2008;26:4268–75. Chang GC, Hsu SL, Tsai JR, Wu WJ, Chen CY, Sheu GT. Extracellular signalregulated kinase activation and Bcl-2 downregulation mediate apoptosis after gemcitabine treatment partly via a p53-independent pathway. Eur J Pharmacol 2004;502:169–83. Fry DW, Kraker AJ, McMichael A, Ambroso LA, Nelson JM, Leopold WR, et al. A specific inhibitor of the epidermal growth factor receptor tyrosine kinase. Science 1994;265:1093–5. Ward WH, Cook PN, Slater AM, Davies DH, Holdgate GA, Green LR. Epidermal growth factor receptor tyrosine kinase. Investigation of catalytic mechanism, structure-based searching and discovery of a potent inhibitor. Biochem Pharmacol 1994;48:659–66. Wakeling AE, Guy SP, Woodburn JR, Ashton SE, Curry BJ, Barker AJ, et al. ZD1839 (Iressa): an orally active inhibitor of epidermal growth factor signaling with potential for cancer therapy. Cancer Res 2002;62:5749–54. Levitzki A, Gazit A. Tyrosine kinase inhibition: an approach to drug development. Science 1995;267:1782–8. Busse D, Doughty RS, Ramsey TT, Russell WE, Price JO, Flanagan WM, et al. Reversible G(1) arrest induced by inhibition of the epidermal growth factor receptor tyrosine kinase requires up-regulation of p27(KIP1) independent of MAPK activity. J Biol Chem 2000;275:6987–95. Lenferink AE, Busse D, Flanagan WM, Yakes FM, Arteaga CL. ErbB2/neu kinase modulates cellular p27(Kip1) and cyclin D1 through multiple signaling pathways. Cancer Res 2001;61:6583–91. Skehan P, Storeng R, Scudiero D, Monks A, McMahon J, Vistica D, et al. New colorimetric cytotoxicity assay for anticancer-drug screening. J Natl Cancer Inst 1990;82:1107–12. Chou TC. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies. Pharmacol Rev 2006;58:621–81. Li T, Ling YH, Goldman ID, Perez-Soler R. Schedule-dependent cytotoxic synergism of pemetrexed and erlotinib in human non-small cell lung cancer cells. Clin Cancer Res 2007;13:3413–22. Khwaja A, Dockrell ME, Hendry BM, Sharpe CC. Prenylation is not necessary for endogenous Ras activation in non-malignant cells. J Cell Biochem 2006;97:412–22. Davies AM, Ho C, Lara Jr PN, Mack P, Gumerlock PH, Gandara DR. Pharmacodynamic separation of epidermal growth factor receptor tyrosine kinase inhibitors and chemotherapy in non-small-cell lung cancer. Clin Lung Cancer 2006;7:385–8.

[31] Xu JM, Azzariti A, Colucci G, Paradiso A. The effect of gefitinib (Iressa ZD1839) in combination with oxaliplatin is schedule-dependent in colon cancer cell lines. Cancer Chemother Pharmacol 2003;52:442–8. [32] Azzariti A, Xu JM, Porcelli L, Paradiso A. The schedule-dependent enhanced cytotoxic activity of 7-ethyl-10-hydroxy-camptothecin (SN-38) in combination with Gefitinib (Iressa ZD1839). Biochem Pharmacol 2004;68:135– 44. [33] Riely GJ, Rizvi NA, Kris MG, Milton DT, Solit DB, Rosen N, et al. Randomized phase II study of pulse erlotinib before or after carboplatin and paclitaxel in current or former smokers with advanced non-small-cell lung cancer. J Clin Oncol 2009;27:264–70. [34] Mok TS, Wu YL, Yu CJ, Zhou C, Chen YM, Zhang L, et al. Randomized, placebo-controlled, phase II study of sequential erlotinib and chemotherapy as first-line treatment for advanced non-small-cell lung cancer. J Clin Oncol 2009;27:5080–7. [35] Roberts PJ, Der CJ. Targeting the Raf-MEK-ERK mitogen-activated protein kinase cascade for the treatment of cancer. Oncogene 2007;26:3291– 310. [36] Chang L, Karin M. Mammalian MAP kinase signalling cascades. Nature 2001;410:37–40. [37] Torii S, Yamamoto T, Tsuchiya Y, Nishida E. ERK MAP kinase in G cell cycle progression and cancer. Cancer Sci 2006;97:697–702. [38] Nelson JM, Fry DW. Akt MAPK (Erk1/2), and p38 act in concert to promote apoptosis in response to ErbB receptor family inhibition. J Biol Chem 2001;276:14842–7. [39] Wang X, Martindale JL, Holbrook NJ. Requirement for ERK activation in cisplatin-induced apoptosis. J Biol Chem 2000;275:39435–43. [40] Bacus SS, Gudkov AV, Lowe M, Lyass L, Yung Y, Komarov AP, et al. Taxol-induced apoptosis depends on MAP kinase pathways (ERK and p38) and is independent of p53. Oncogene 2001;20:147–55. [41] Tang D, Wu D, Hirao A, Lahti JM, Liu L, Mazza B, et al. ERK activation mediates cell cycle arrest and apoptosis after DNA damage independently of p53. J Biol Chem 2002;277:12710–7. [42] Castro-Obregon S, Rao RV, del Rio G, Chen SF, Poksay KS, Rabizadeh S, et al. Alternative, nonapoptotic programmed cell death: mediation by arrestin 2, ERK2, and Nur77. J Biol Chem 2004;279:17543–53. [43] Bepler G, Kusmartseva I, Sharma S, Gautam A, Cantor A, Sharma A, et al. RRM1 modulated in vitro and in vivo efficacy of gemcitabine and platinum in nonsmall-cell lung cancer. J Clin Oncol 2006;24:4731–7. [44] Mebratu Y, Tesfaigzi Y. How ERK1/2 activation controls cell proliferation and cell death: Is subcellular localization the answer? Cell Cycle 2009;8:1168– 75. [45] Linardou H, Dahabreh IJ, Kanaloupiti D, Siannis F, Bafaloukos D, Kosmidis P, et al. Assessment of somatic k-RAS mutations as a mechanism associated with resistance to EGFR-targeted agents: a systematic review and meta-analysis of studies in advanced non-small-cell lung cancer and metastatic colorectal cancer. Lancet Oncol 2008;9:962–72. [46] Allegra CJ, Jessup JM, Somerfield MR, Hamilton SR, Hammond EH, Hayes DF, et al. American Society of Clinical Oncology provisional clinical opinion: testing for KRAS gene mutations in patients with metastatic colorectal carcinoma to predict response to anti-epidermal growth factor receptor monoclonal antibody therapy. J Clin Oncol 2009;27:2091–6. [47] Karapetis CS, Khambata-Ford S, Jonker DJ, O’Callaghan CJ, Tu D, Tebbutt NC, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med 2008;359:1757–65. [48] Brugger W, Triller N, Blasinska-Morawiec M, Curescu S, Sakalauskas R, Manikhas G, et al. Biomarker analyses from the phase III placebo-controlled SATURN study of maintenance erlotinib following first-line chemotherapy for advanced NSCLC. J Clin Oncol 2009;27(Suppl.):abstr 8020. [49] Uchida A, Hirano S, Kitao H, Ogino A, Rai K, Toyooka S, et al. Activation of downstream epidermal growth factor receptor (EGFR) signaling provides gefitinib-resistance in cells carrying EGFR mutation. Cancer Sci 2007;98:357–63. [50] Helfrich BA, Raben D, Varella-Garcia M, Gustafson D, Chan DC, Bemis L, et al. Antitumor activity of the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor gefitinib (ZD1839 Iressa) in non-small cell lung cancer cell lines correlates with gene copy number and EGFR mutations but not EGFR protein levels. Clin Cancer Res 2006;12:7117–25.