Expression and pharmacological inhibition of thymidylate synthase and Src kinase in nonsmall cell lung cancer

Authors


  • Potential conflicts of interest: M.P. and G.V.S. received honoraria from Eli Lilly.

Abstract

The combination of cytotoxic chemotherapy with signaling pathway inhibitors represents a potential strategy to improve the treatment of nonsmall cell lung cancer (NSCLC). Thymidylate synthase (TS) is an enzyme essential for DNA synthesis, and its overexpression has been associated with the reduced sensitivity to antifolate agents. Src is a tyrosine kinase that modulates the cytotoxicity of cancer cells after drug treatment, and in vitro data indicate that its inhibition could revert the resistance to TS-inhibiting drugs. Our study investigated the significance of TS and Src expression in NSCLC tissues, and the effects of their pharmacological inhibition in cell lines. In tumor and normal tissues from 94 resected NSCLC patients, TS and Src transcript levels were found positively correlated (RS = 0.66), associated with patients smoking history and overall survival. At multivariate analysis, TS gene expression was an independent prognostic factor (relative risk (RR) = 1.78, from 1.16 to 2.72; p < 0.01). Immunohistochemical detection in tumor specimens confirmed that Src kinase activation, evaluated by phospho-specific antibody, was associated to a higher TS expression. In cell lines, dasatinib, a Src-inhibiting agent, synergistically enhanced pemetrexed-cytotoxicity of A549 cells, as evaluated by MTT and apoptosis assays. The biological explanation for this interaction was based on the upregulation of TS messenger RNA and protein levels induced by pemetrexed, which was significantly prevented by dasatinib cotreatment. The data of our study suggest that TS and Src may belong to a common pathway that bears prognostic significance in NSCLC, and that Src represents a potential target to improve the efficacy of TS-inhibiting agents.

Cytotoxic chemotherapy is the standard of care for locally advanced and metastatic nonsmall cell lung cancer (NSCLC) even though the efficacy of the currently available therapeutic armamentarium is limited,1, 2 and new therapeutic strategies are clearly needed to improve patients quality of life, response rates and overall survival.

Thymidylate synthase (TS) is the enzyme which catalyzes the reaction that provides the sole de novo intracellular source of deoxythymidine monophosphate, or thymidylate, which is essential for DNA replication and repair.3, 4 TS protein and messenger RNA (mRNA) levels have been found elevated in many human cancers, and correlated with poor prognosis in patients with colorectal,5 breast,6 head and neck,7 pancreatic8 and in NSCL cancers.9, 10 Transfection of catalytically active TS has been shown to induce a transformed phenotype in human cells, as manifested by foci formation, anchorage independent growth and tumor formation in nude mice, suggesting that TS itself could be considered as an oncogene.11 TS is also the cellular target of 5-fluorouracil (5-FU)4 and of pemetrexed.12 Pemetrexed is a multitargeted antifolate that prevents the formation of DNA and RNA, blocking the growth of cancer cells, by inhibiting the activity of four enzymes in the folate pathway, being TS the most important target.13–15 In lung cancer, it was previously reported an higher TS mRNA and protein level in squamous cell carcinoma, as opposed to adenocarcinoma,10 and a 10-fold higher TS expression in small-cell lung cancer (SCLC) as compared to NSCLC16 and to carcinoid tumors.17 More recently, it was shown a consistency between the aforementioned findings and TS enzymatic activity in lung cancer cell lines of different histotypes,18 and these findings led to a rationale interpretation of the outcome of the clinical trials which investigated the role of pemetrexed in NSCLC and SCLC.19–21 As molecular strategies aimed at TS inhibition, it has been shown to improve the in vitro effects of pemetrexed;13 in our study, we investigated one of the molecular mechanisms that could regulate TS expression in lung cancer cells. Src is a tyrosine kinase (TK) involved in many aspects of tumorigenesis, including proliferation, migration and angiogenesis; it is activated in lung tumors22 and it has been postulated to be a potential relevant therapeutic target.23 Dasatinib is an oral dual Bcr/Abl and Src family TK inhibitor approved for the use in chronic myelogenous leukemia.24 In lung cancer, dasatinib was shown to be selectively active in cells with epidermal growth factor receptor (EGFR) mutations in vitro,25 and a recent clinical trial reported the modest activity of single-agent administration in unselected NSCLC patients.26 When combined with cisplatin in NSCLC cell lines with wild-type EGFR, dasatinib showed to increase cisplatin-induced cell death by impairing the upmodulation of several critical DNA repair and synthesis genes, including TS.27 As it was recently reported that Src kinase inhibition could also revert the 5-FU-resistance of cancer cells through indirect TS regulation,28 our study aimed at elucidating the possible links between TS and Src expression in NSCLC, by the quantification of gene and protein expression in tumor tissues, and by investigating the in vitro effects of their pharmacological inhibition in NSCLC cell lines.

Material and Methods

Patients and samples

Surgical specimens of both tumor and corresponding normal lung tissues of 94 NSCLC patients completely resected between 2004 and 2005 at the San Luigi Hospital were consecutively collected. One part of each specimen was snap-frozen in liquid nitrogen for gene expression analysis, and the immediately adjacent part was fixed in formalin, embedded in paraffin and used for H&E staining and for immunohistochemical analyses. None of the patients received preoperative or postoperative treatments with chemotherapy and/or radiotherapy according to an institutional treatment policy implemented in the years of specimen collection. Ever smokers were defined as current or former smokers. Median survival time was 60.7 months, with a minimum follow-up time of 5 years. In total, 49 patients had died at the time of the analyses. All samples were reviewed by the pathology staff and tumors were classified according to the World Health Organization classification29; none of the researchers conducting gene expression and statistical analyses had access to disclosed clinical–pathological data. The study was approved by the institutional review board at the University of Torino, S. Luigi Hospital.

RNA isolation and cDNA synthesis from cells and fresh snap-frozen NSCLC specimens

Total RNA was isolated from both normal and tumor lung specimens using the RNeasy 96 Kit implemented on Biorobot 8000 (Qiagen, Hilden, Germany) according to the manufacturer's instructions. RNA was extracted and retrotranscribed to complementary DNA (cDNA) as described previously.10 Total RNA was isolated from cell lines with Qiazol lysis reagent (Qiagen) according to manufacturer's instructions.

Real-Time Polymerase Chain Reaction

Relative cDNA quantitation for each gene, using β-actin as internal reference gene, was performed by a fluorescence-based real-time detection method (ABI PRISM 7900 Sequence Detection System [Taqman]; Applied Biosystems, Foster, CA). The sequences of the primers and probes used for TS, Src and β-actin have been published previously.10, 27 All primers and probes were intron-spanning to avoid genomic DNA contamination. Polymerase chain reaction (PCR) mixture and cycling conditions were as previously described.10

Immunohistochemistry

Expression levels of phoshpo-Src and TS protein were detected by rabbit p-SrcTyr416 (Cell Signaling, Beverly, MA) and mouse TS106 (from Santa Cruz Biotechnology, Santa Cruz, CA) antibodies diluted 1/100 with overnight incubation at 4°C, followed by horseradish peroxidase (HRP)-conjugated secondary antibody incubation at room temperature for 30 min. Immunoreactions were revealed by a biotin-free dextran-chain detection system (Envision, DakoCytomation, Glostrup, DK) and developed using diaminobenzidine as the chromogen. Before primary antibody incubation, antigen retrieval was performed by Pascal pressure chamber (Dako) heating in ethylenediaminetetraacetic acid (EDTA) buffer solution pH 8.0 (5 min at 125°C). Slides were counterstained with hematoxylin. For each tumor, the percentage of expression of p-Src and TS was evaluated.

Cell lines, cultures and drugs

Three human NSCLC cell lines (H1395, A549 and H1299, where all are adenocarcinomas) were purchased from American Type Culture Collection (Manassas, VA). One primary NSCLC cell line (named “primary NSCLC”) was established in our laboratory as described in Ref. 30, after obtaining a specific informed consent, from a 59-year-old male patient with the pathological diagnosis of adenocarcinoma of the lung, as assessed by morphologic examination and by immunohistochemistry (IHC) profiling (thyroid transcription factor 1 (TTF-1+)/p63). All cell lines carried wild-type EGFR. Cells were maintained in RPMI-1640 supplemented with 10% fetal calf serum (FCS), 2 mM L-glutamine, penicillin (25 units/ml) and streptomycin (25 μg/ml, all from Sigma-Aldrich) in a humidified atmosphere containing 5% CO2 at 37°C. Dasatinib (BMS-354825, Bristol-Meyer Squibb) was prepared at 20 mM concentration and dissolved in dimethyl sulfoxide (DMSO) (Sigma-Aldrich, Milan, Italy) as in Ref. 27. Original stock solutions of pemetrexed (Eli Lilly, Indianapolis) at a concentration of 60 mM were stored at −20°C and freshly dissolved in culture medium before use.

Cell viability assay

Cells were plated in 96-well plates for 24 hr and then treated with increasing concentrations of pemetrexed (500 nM, 1 μM, 5 μM, 15 μM, 50 μM and 100 μM) or dasatinib (20 nM, 50 nM, 100 nM, 200 nM, 500 nM and 1 μM) or with the combination of both drugs for 72 hr. A 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Sigma-Aldrich) assay was performed according to the manufacturer's instruction. Absorbance was measured at 590 nm with a microplate reader (Biorad).

Apoptosis assay

Cells were seeded in six-well plates at appropriate density and then treated with 2 or 4 μM pemetrexed in the presence or absence of 100 or 200 nM dasatinib for 72 hr. Cells were then harvested and stained with annexin V and propidium iodide (Immunostep, Salamanca Spain) and subsequently analyzed on a cytofluorimeter by FACScan (BD Biosciences, San Jose, CA). propidium positive cells were considered as necrotic, annexin positive/propidium negative as apoptotic and the double negative as alive. Apoptotic rates were calculated as reported elsewhere.27

Western immunoblot analysis

After treatments, cells were lysed in 50 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.5% Nonidet-P40, 10 mM glycerol, 5 mM EDTA and 0.5% sodium deoxycholate and the lysates were treated with a mixture of protease and tyrosine phosphatase inhibitors (1 mM phenyl methane sulfonyl fluoride (PMSF), 1 μg/ml aprotinin, 1 μg/ml leupeptin and 1 mM Na-OrtoVanadate). Proteins were separated on 8% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and blotted onto polyvinylidene fluoride (PVDF) membrane (Hybond-P, Amersham, Milan, Italy). Filters were incubated with the following monoclonal antibodies against β-Actin, thymidylate synthase (TS106, all from Santa Cruz Biotechnology), c-Src and pSrcTyr416 (Cell Signaling, Beverly, MA). Detection was performed using enhanced chemiluminescence kit (Pierce, Rockford, IL).

SiRNA-mediated silencing of c-Src

A549 cells were seeded in six-well dishes at a density of 1.2 × 105 cells per well 24 hr before the experiment. FlexiTube small interfering RNA (siRNA) directed to human c-Src and negative controls (AllStars negative controls) were transfected at a 500-nM final concentration using RNAifect transfection reagent according to the manufacturer's instructions (all reagents from Qiagen). Down regulation of c-Src was assessed by Western blotting at 72 and 96 hr after SiRNA transfection, comparing to negative controls. Silencing efficiency was greater than 70%.

Statistical analyses

TS and Src mRNA levels in tumor and normal specimens were estimated with the ΔCt method, with β-actin as a reference gene. The differential TS and Src mRNA expression between tumors and corresponding normal lung tissues was analyzed by the ΔΔCt method. All data were converted into a linear scale and normalized to the median expression in normal tissues (considered as 1.00). The correlation between the level of TS and Src upregulation in tumor samples was tested by Spearman's rank correlation test. To test significant associations with clinical–pathological variables, the Mann–Whitney U-test and the Kruskal–Wallis test were used. Univariate analysis of survival was carried out by the method of Kaplan and Meier dividing patients according to median levels of upregulation, and the significance estimated with the log-rank test. Multivariate analysis was performed by Cox's proportional-hazards method including the variables with an impact on prognosis of patient at univariate. In cell line experiments, the t-test was performed comparing treated samples versus untreated controls. Nonlinear regression analyses for the calculation of all the effective concentrations, including the half maximal inhibitory concentration (IC50), were carried out by the Graphpad software version 5.0. The nature of the interaction between drugs was assessed by combination index (CI) method and by isobologram analysis as described previously.31 In Western blot experiments, integrated densitometry data were obtained using the ImageJ software (http://rsbweb.nih.gov\ij). In the IHC experiments, the association between Src phosphorylation and TS expression in tumors was estimated by the χ2 test. Phospho-Src-positive tumors were considered as those with ≥5% of positive cells, whereas negative cells were considered as those with no staining in tumor cells. The cutoff used to define TS expression as “low” or “high” was 10% of positive cells, as reported previously.10 In all tests, the statistical significance was set at p = 0.05.

Results

TS and Src gene expression levels are significantly correlated in NSCLC and are associated with patients smoking history

The main characteristics of patients are shown in the Supporting Information Table 1. The results of real-time PCR quantification showed a significant upregulation of TS transcripts in tumors, being 1.00 (from 0.09 to 20.82) the median level in non-neoplastic lung as compared to 13.9 (from 0.09 to 252, all unit-less ratios) for the tumor tissues (p < 0.0001, Mann–Whitney U-test). A significant upregulation of Src levels was also observed, being 1.00 (from 0.05 to 3.95) the median level in non-neoplastic as compared to 2.19 (from 0.03 to 8.35) for the tumor tissues (p < 0.01). Moreover, the levels of TS and of Src upregulation (expressed as tumor/normal ratios, −ΔΔCT) were found significantly correlated by Spearman correlation method (RS = 0.66, p < 0.0001; Fig. 1a). TS and Src gene regulation were significantly associated with smoking history of patients, being both genes more upregulated in active/former smoker patients, as compared to never-smoker patients (n = 8; p = 0.02 and 0.01, Figs. 1b and 1c, respectively). Never-smokers were mostly women (n = 6) and all adenocarcinomas. Moreover, TS expression, but not that of Src, was significantly lower in adenocarcinoma than squamous cell carcinoma (p < 0.01) and in well and moderate versus poorly differentiated tumors (p = 0.01).

Figure 1.

(a) Correlation between the levels of TS (y-axis) and SRC (x-axis) gene regulation (−ΔΔCt) in 94 NSCLC samples (Spearman rank correlation method). TS and Src gene expression levels are upregulated in ever smokers as compared to never smoker NSCLC patients (b, c; Mann-Whitney U-test).

Src phosphorylation correlates with an increased TS expression in lung tumor tissues

To investigate the association between Src kinase phosphorylation and TS protein expression in NSCLC, the first 50 consecutive tumor samples were evaluated by IHC. A total of 27 cases (54%) were p-Src positive (ranging from 5% to 50% of tumor cells) with a membrane pattern of localization, whereas the remaining 23 tumors were negative (absence of positive tumor cells). The median TS protein expression in tumor tissues was 15% (ranging from 0 to 70%), and 19 cases (38%) were totally negative. Figure 2 (panels a, b and c) shows two positive and one negative representative cases. Dividing tumors into two populations by phospho-Src positivity, a significant association between p-Src and increased TS protein level was detected, being the majority (∼ 70%) of p-Src positive tumors associated to a “high TS” expression (see methods) and vice versa (p < 0.01, χ2 test; Fig. 2d). Overall, out of the 19 TS totally negative tumors, 13 (of 23; 56%) were p-Src negative and six (of 27, 22%) were Src-positive.

Figure 2.

Immunohistochemical detection of phospho-Src (p-Src) and of TS in NSCLC samples. (a) and (b) represents two positive and (c) one negative case at ×100 magnification. (d) Association between p-Src and TS expression as evaluated by χ2 statistics.

Prognostic significance of TS and Src expression in NSCLC

To investigate the prognostic impact of TS and Src mRNA expression in NSCLC, all tumors were divided into “low” and “high” TS and Src expression according to the median value of upregulation for each gene and univariate analyses of survival were performed according to the Kaplan–Meier method. Lower TS (75.2 vs. 43.3 months, log-rank p = 0.01, Fig. 3a) and Src (74.2 vs. 49.7 months, log-rank p = 0.04, Fig. 3b) levels were significantly correlated with a longer overall survival time. According to tumor histotypes, the prognostic impact of both molecules expression was stronger in adenocarcinoma as compared to squamous carcinoma (see TS in Figs. 3c and 3d). Among the other characteristics, only disease stage was significantly associated with prognosis of patients (stage I, II vs. stage III, p = 0.001), and at multivariate analysis (which included only the variables with a significant impact at univariate), TS expression and stage were found to be independent predictors of survival (both p < 0.01; Supporting Information Fig. S1).

Figure 3.

Prognostic significance of TS and Src mRNA expression in NSCLC. Kaplan–Meier survival curves of all the NSCLC patients divided by (a) TS and (b) Src “low” and “high” expression, according to the median values. Survival curves of (c) adenocarcinoma and (d) squamous cell carcinoma patients divided by TS expression (p values are log-rank tests; ns = not significant).

Effects of the concomitant inhibition of TS and Src in NSCLC cell lines

The effects of the simultaneous in vitro inhibition of TS and Src, by pemetrexed and dasatinib, were measured by MTT assay in four cell lines of lung adenocarcinoma. The data obtained from the growth inhibition curves (as shown in Fig. 4) were used to perform nonlinear regression analyses and for the IC50 calculation. CI analysis indicated strong synergistic effects for the combination of pemetrexed and dasatinib in A549 cells (CI = 0.23 ± 0.12); whereas in h1299, the drugs showed a trend toward supra-additivity (CI = 0.88 ± 0.03), and additive effects (CI = 1) were estimated in both H1395 and the primary NSCLC cell line. The synergistic effects of the pemetrexed/dasatinib simultaneous administration were maintained (i.e., all CIs < 1) at each effective concentration tested in A549 cells only, while the combination was defined as additive for all the other cell lines tested (not shown). Supporting Information Figure S2 illustrates the results of the isobologram analysis at 50% of growth inhibition, indicating the synergistic nature of pemetrexed/dasatinib interaction in A549 cells. No relevant differences were observed when the two drugs were administered at different schedules, or for different times of incubation (not shown). To further confirm these observations, an apoptosis assay was performed. As shown in the Figure 5, A549 cells exhibited a poor induction of apoptosis after 72 hr of dasatinib treatment at 100 nM dose (annexin V positive/propidium iodide negative cells below 5% of total cells), whereas 2 μM pemetrexed as single agent induced apoptosis in 28.7% of the cells. This percentage strongly increased up to 52.3% when the two drugs where simultaneously administered, concordantly with the synergistic interaction demonstrated in the aforementioned analyses. Similarly, at the same drug concentrations and time intervals, a moderate supra-additivity was detected in H1299 cells (from 8 to 16%, when the dasatinib and pemetrexed were tested as single agents to 32.2% for the concomitant treatment), while dasatinib did not induce any increase in the pemetrexed sensitivity in the two cell lines tested (Fig. 5 shows the primary NSCLC cell line). When tested, higher concentrations of dasatinib failed to produce further enhancements in pemetrexed sensitivity in the A549 cell line (see Supporting Information Fig. S3).

Figure 4.

Cytotoxicity assay of four NSCLC cell lines treated with 6 concentrations of pemetrexed (500 nM, 1 μM, 5 μM, 15 μM, 50 μM and 100 μM) or dasatinib (20 nM, 50 nM, 100 nM, 200 nM, 500 nM and 1 μM) or with the combination of both drugs, for 72 hr. (d) Results are expressed as percentage of surviving cells as compared to untreated controls at each dose. PEM, pemetrexed; DAS, dasatinib.

Figure 5.

Annexin V/propidium iodide apoptosis assay of cells exposed to dasatinib (DAS) or pemetrexed (PEM) or the combination of both at the indicated concentrations. Propidium positive cells were considered as necrotic, annexin positive/propidium negative as apoptotic (bottom right) and the double negative as alive (bottom left).

Dasatinib blocks pemetrexed-induced TS upregulation in A549 cell line

The functional mechanisms behind the synergistic interaction of the combination of pemetrexed and dasatinib in the A549 cell line were further investigated by Western blot analysis. The results (Fig. 6a) showed that A549 cells underwent a detectable increase of Src phosphorylation at the catalytic site (tyrosine 416) on 24-hr treatment with pemetrexed, as compared to untreated cells, while coadministration with dasatinib was able to completely abrogate both the endogenous and pemetrexed-induced Src phosphorylation. Similarly, TS protein was upregulated by pemetrexed, and its expression was reduced with the concomitant treatment with dasatinib at 100 and 200 nM. The integrated densitogram analysis (shown in the Fig. 6b) quantified as +43% and +106% the pemetrexed-induced increase of p-Src and TS, respectively, as compared to untreated controls, whereas in the presence of dasatinib at 100 and 200 nM doses, the upregulation of TS protein was found to be limited to +61% and +41%, respectively. In the same experimental conditions, the real-time PCR quantification of TS transcript showed a significant (about twofold) increase in TS levels following pemetrexed exposure, that was partially prevented by cotreatment with dasatinib 100 nM (Fig. 6c, black columns), in line with the Western blotting results. H1299 cells exhibited a similar trend, but with lower enhancement of TS transcription (Fig. 6c, white columns). Finally, to specifically address the role of Src kinase, among the other known targets of dasatinib24 in the sensitization to pemetrexed observed in A549 cell line, a SiRNA-mediated silencing approach was used. Treatment of A549 cells with Src-specific SiRNA produced a ∼ 75% reduction in Src protein level, as compared to the cells treated with negative control SiRNA (scrambled, Fig. 6d), with poor and not significant changes on TS levels (not shown), and significantly enhanced pemetrexed sensitivity of A549 cells in terms of apoptotic rate, as evaluated by annexin V/propidium iodide staining (Fig. 6e).

Figure 6.

(a) Western blot analysis of A549 cells treated with the indicated concentrations of dasatinib (DAS) or pemetrexed (PEM) or the combination of both. Lysates were subjected to immunoblotting with phospho-Src, total Src, TS and β-Actin antibodies. (b) Densitogram analysis of western blot data. Protein amounts for Src and TS were calculated as ratios of P-Src/total Src and TS/β-Actin, respectively, and expressed comparing to untreated controls (±SD from two independent experiments). (c) Quantification of TS mRNA levels by real-time PCR in A549 and H1299 cells treated with the indicated drugs for 18 hr. Values are expressed as −ΔΔCT as compared to untreated controls (±SD from two independent experiments). (d) Src depletion in A549 cells by SiRNA silencing. (e) Apoptosis assay in A549 cells transfected with Src-SiRNA, or with negative controls (scrambled), and treated with pemetrexed (±SD from two independent experiments).

Discussion

Our study indicates that TS and Src kinase are coexpressed in NSCLC, where their expression is upregulated in tumors from former/current smokers, and is associated with an adverse prognosis. Moreover, the simultaneous TS and Src inhibition by pemetrexed and dasatinib synergistically increase the cytotoxicity in one NSCLC cell line.

Previously, it was reported the overexpression and the activation of Src kinase in NSCLC tissues and cell lines27, 32 and TS transcript upregulation has been extensively documented in lung tumors.9, 10, 16 To the best of our knowledge, this is the first report in which TS and Src are concomitantly investigated and detected to be closely associated. Moreover, in the present case-series, TS expression was found to be an independent prognostic factor for survival. This is in agreement with a previously published article, in which the transcript expression of 22 DNA repair genes in tissue samples of NSCLC and corresponding normal lung tissues was analyzed, and tumoral TS overexpression was significantly associated with an adverse prognosis at the univariate analysis.33 Concerning Src kinase, an increased activation has already been reported in squamous cell carcinoma and in active smoker patients, but no impact on NSCLC prognosis has so far been described,22, 32 although some retrospective studies have indicated a clear association between Src activity and poor clinical prognosis in other tumor types.34, 35

NSCLC in never-smokers has been referred to as a different disease in terms of prognosis and peculiar molecular profile,36 showing an increased rate of EGFR mutations37 and of EML4-ALK fusion gene.38 Consistently, among the eight never-smokers of our study, six are women and all of them are adenocarcinomas, two factors typically correlated with EGFR and EML4-ALK molecular changes. Even though the size of the never-smoker group in the present case-series is small as compared to the group of current/former smokers, a significant association between TS and Src expression and smoking status of patients has been found, being both genes not (or less) regulated in never-smoker patients. If confirmed by future investigations, these evidences could support a previously reported model in which the mitogenic effect of nicotine in NSCLC involves Src activation, induction of Rb-Raf-1 interaction and phosphorylation of Rb, in a pathway that leads to increased proliferation via E2F1,39 which is the transcription factor responsible for TS transcription.16, 40

Consistently with this model, a high level of direct correlation between TS and Src mRNA expression in tissues, as evaluated by real-time PCR quantification, was found. Moreover, as Src activation is mediated by phosphorylation, an immunohistochemical analysis of TS and phospho-Src in 50 NSCLC specimens was performed to reveal functional associations. The results showed that tumors presenting detectable Src kinase phosphorylation were more likely expressing high TS levels, thus suggesting that the two molecules could be functionally related. However, the existence of p-Src-positive/TS-negative tumors (and vice versa) indicates that other molecular mechanisms, other than Src kinase, could be involved in TS regulation.

In cancer cells, Src kinase can modulate the cytotoxicity of different chemotherapeutic drugs.27, 28, 41 Therefore, the effects of the concomitant Src and TS inhibition in NSCLC cell lines by dasatinib and pemetrexed were assessed in vitro in a limited panel of four cell lines with wild-type EGFR. Dasatinib synergistically enhanced pemetrexed cytotoxicity in one (A549 cell line) of four cell lines tested. Similarly, to the data recently reported in lung cancer cell lines with cisplatin plus dasatinib27 and erlotinib plus pemetrexed42 combinations, TS mRNA and protein levels were rapidly induced by pemetrexed, and this increase was significantly prevented by the coadministration of dasatinib. Noteworthy, treatment with dasatinib alone had poor effects on baseline TS levels further indicating that other mechanisms could be responsible for its transcription, and that Src is responsible for controlling the drug-induced levels, as recently shown.27 However, the other three cell lines investigated showed minimal supra-additive or additive interactions without any evident antagonistic effect. H1299 cells exhibited a significantly lower increase in TS mRNA levels on pemetrexed administration, and this may account for the limited response to the pemetrexed/dasatinib combination. In light of these data, further investigations describing the complete signaling pathway underlying the drug-mediated TS regulation in NSCLC, similar to what was recently reported in pancreatic carcinoma cells,28 are clearly warranted. Moreover, as the cell lines used in our study were all derived from patients with lung adenocarcinomas, for the treatment of which pemetrexed has been indicated, further studies should be addressed to investigate whether synergistic effects could be reached also in lung squamous carcinomas cell lines.

Finally, as dasatinib administered at higher doses can inhibit several other structurally similar kinases,24 a siRNA-mediated silencing approach was used, and confirmed the specific role of Src in modulating pemetrexed-sensitivity of A549 cells. Nevertheless, the lack of a clear dose-dependency of dasatinib in the regulation of TS levels (Fig. 6b) and in the sensitivity (Supporting Information Fig. S3) of pemetrexed-treated A549 cells also indicated the minimal impact of the other kinases potentially targeted by dasatinib at the highest dose.

In conclusion, our data demonstrate the coexpression and the prognostic impact of TS and Src kinase in NSCLC, and support the role of Src as a potential target to improve the efficacy of TS-inhibiting agents, by the interference with critical mechanisms of resistance. As recent clinical trials evaluating the combination of signaling pathway inhibitors with cytotoxic chemotherapy, in unselected population of patients, provided controversial and sometimes disappointing results,43, 44 future investigations should be addressed to identify the molecular predictors of the efficacy of the dasatinib/pemetrexed combination, in the attempt to optimize drug sensitivity in NSCLC patients.

Acknowledgements

This work was partially supported by a grant from Regione Piemonte, Ricerca Sanitaria Finalizzata (14631/DB20.01; to G.V.S.; May 2010).

Ancillary