Lung cancer is one of the most aggressive malignancies with poor prognosis. It is estimated that more than 160 000 and 65 000 lung cancer patients in the USA and Japan, respectively, die each year.(1,2) A wide variety of new chemotherapy medicines have been developed and introduced in clinical practice, but the mortality rate has not been improved.(1) Recently, the strategy of drug development has focused on targeting particular molecules that are supposed to be critical for cancer progression. Several molecules in the growth factor receptor pathway are specifically targeted because those molecules are well recognized as being aberrantly regulated in cancers. For example, epidermal growth factor receptor (EGFR) and its downstream molecules are often upregulated due to gene amplification or mutation;(3,4) therefore, targeting EGFR is a major therapeutic strategy for non-small-cell lung carcinoma (NSCLC).(5) Gefitinib is a well-known small molecule inhibitor that selectively suppresses EGFR tyrosine kinase activity(6) and has been applied in the treatment of NSCLC.(7) Several studies have shown that gefitinib treatment has a drastic antitumor effect in a subset of NSCLC which had acquired certain types of EGFR mutation.(8,9) Since the appearance of gefitinib, several selective EGFR inhibitors have been developed. However, these drugs only revealed a minimal effectiveness due to the aberrant regulation of molecules located downstream from the receptor tyrosine kinase pathways including Ras-Raf-MAPK and phosphatidylinositol 3′-kinase (PI3K)-Akt.(10,11) Among them, mammalian target of rapamycin (mTOR) is one of the major effectors regulated by the PI3K-Akt signaling pathway and plays a central role in this stimulated growth.(12,13) Moreover, there is an upregulation of mTOR activity in many types of cancers including NSCLC.(14,15) Therefore, several compounds that selectively inhibit mTOR activity have been developed for clinical use.(16,17) Temsirolimus (CCI-779), an analogue of rapamycin, was recently synthesized to specifically inhibit mTOR and has provided prolonged survival of patients with renal cell carcinoma. It was also reported that temsirolimus showed a certain antitumor effect on other types of cancers including breast cancer,(18) glioblastoma,(19) neuroendocrine carcinomas,(20) and mantle cell lymphoma.(21) Moreover, temsirolimus has antitumor effects in other diseases such as lymphangioleiomyomatosis.(22) Based on these observations, we questioned whether temsirolimus treatment could be a potential therapeutic option for NSCLC. In this study, we evaluated the antiproliferative and antitumor effects of temsirolimus in NSCLC in vitro and in vivo, with an assessment of its survival advantage in an animal model of advanced NSCLC.
Temsirolimus (CCI-779), a recently synthesized analogue of rapamycin, specifically inhibits mTOR and has been approved for clinical use in renal cell carcinoma. Recent reports have indicated the growth inhibitory effect of temsirolimus in some cancers including non-small-cell lung carcinoma (NSCLC). In this study, we aimed to explore the potential therapeutic use of temsirolimus as a treatment for NSCLC. Using cultured NSCLC cells (A549, H1299, and H358), we determined the effect of temsirolimus on cell proliferation and its antitumor effects on subcutaneous tumors, as well as its contribution to the survival of mice having pleural dissemination of cancer cells, mimicking advanced NSCLC. Temsirolimus suppressed proliferation of NSCLC cells in a dose-dependent manner, with an IC50 of <1 nM. Western blot analysis revealed that temsirolimus treatment specifically inhibited the phosphorylation of mTOR and its downstream effectors in 1 h, accompanied by an increased cell population in the G0/G1 phase, but according to flow cytometry, the cell population did not increase in the sub-G0 phase. When NSCLC subcutaneous tumor-bearing mice were treated with temsirolimus, tumor volume was significantly reduced (tumor volume on day 35: vehicle vs temsirolimus = 1239 vs 698 cm3; P < 0.05). Furthermore, prolonged survival was observed in pleural disseminated tumor-bearing mice with temsirolimus treatment (median survival: vehicle vs temsirolimus = 53.5 vs 72.5 days; P < 0.05). These results suggest that temsirolimus could be useful for NSCLC treatment, due to its antiproliferative effect, and could be a potential treatment for advanced NSCLC, giving prolonged survival. (Cancer Sci 2011; 102: 1344–1349)
Materials and Methods
Cell lines and cultures. Three cancer cell lines that were established from human NSCLC (A549, H1299, and H358) were used in this study. A549 was cultured in DMEM (Sigma-Aldrich, St. Louis, MO, USA) and H1299 and H358 were cultured in RPMI-1640 medium (Sigma-Aldrich) at 37°C in humidified air with 5% CO2 These media were supplemented with 10% FCS (Hyclone, Logan, UT, USA), 100 U/mL penicillin and 100 mg/mL streptomycin (Sigma-Aldrich).
Reagents. Temsirolimus, commercialized as Tricel by Wyeth K.K. (Madison, NJ, USA), was purchased from OZ International (Tokyo, Japan). The temsirolimus was diluted to the final concentration with culture media before an in vitro experiment. When temsirolimus was used in vivo, it was dissolved and diluted to a final concentration of 10 mg/kg with 0.9% sodium chloride.
Trypan blue exclusion assay. Cancer cells (5.0 × 103 per well) were plated directly in 24-well dishes with culture medium. After the cells entered into an exponential growth phase, they were treated with different concentrations of temsirolimus (0, 0.1, 1, 10, 100, or 1000 nM) for 48 h, stained with Trypan blue, and the number of viable cells was counted using a hemacytometer.
Apoptosis assay. Cells in apoptosis were determined by TUNEL assay using a MEBSTAIN Apoptosis kit II (MBL International, Woburn, MA, USA) according to the manufacturer’s protocol. Briefly, cells (1.0 × 104 per well) were seeded on Lab-Tek 8-well permanox chamber slides (Becton Dickinson, Franklin Lakes, NJ, USA) and were treated with 10 nM/L of temsirolimus or with an equivalent volume of diluted DMSO (final concentration, 0.005%) as a control for 48 h. The TUNEL-positive cells were counted with a fluorescence microscope.
Cell cycle analysis by flow cytometry. For cell cycle analysis, cancer cells were plated in six well tissue culture plates and treated with different concentrations of temsirolimus (0, 1, 10, or 100 nM/L). After a 24-h treatment, the cells were harvested and stained with 20 mg/mL propidium iodide. The DNA content was analyzed with a fluorescence-activated cell sorter (FACScan; Becton Dickinson) using CellQuest software (BD Biosciences, San Jose, CA, USA).
Western blot analysis. Whole cell lysates and nuclear protein were extracted using M-PER buffer (Thermo Fisher Scientific, Rockford, IL, USA) and NE-PER buffer (Thermo Fisher Scientific) supplemented with protease inhibitors and phosphatase inhibitors. The protein concentration of the collected supernatants was determined and equal amounts of protein were electrophoresed under a reducing condition in gradient polyacrylamide gels (ATTO, Tokyo, Japan) and were then transferred onto PVDF filter membranes (Millipore, Billerica, MA, USA). The membranes were incubated with primary antibodies at 4°C overnight, followed by incubation with secondary antibodies at room temperature for 1 h. An Amersham ECL Plus Western Blotting Detection System (GE Healthcare, Piscataway, NJ, USA) was used for signal detection. The antibodies used for Western blotting were phospho-mTOR (Ser2448), mTOR, phospho-p70 S6 kinase (Thr389), p70 S6 kinase, phospho-S6 ribosomal protein (Ser235/236), and hydroxy-HIF-1α (Pro564) (D43B5). All of them were obtained from Cell Signaling Technology (Beverly, MA, USA). β-Actin was obtained from Sigma-Aldrich. Horseradish peroxidase-conjugated rabbit anti-mouse IgG was obtained from Dako Cytomation (Glostrup, Denmark). Goat anti-rabbit IgG was obtained from American Qualex Antibodies (La Mirada, CA, USA).
Animal experiments. The protocol for the animal experiments was approved by the Ethics Review Committee for Animal Experiments of Okayama University (Okayama, Japan). Mice used in this study were purchased from Clea (Tokyo, Japan). A549 s.c. xenografts were produced on the backs of 6-week-old male BALB/c nu/nu mice by injecting 3 × 106 cells mixed with Matrigel (BD Biosciences) at a 1:1 ratio. After 7 days, the tumor-bearing animals were randomized into two groups that consisted of seven mice each: (i) temsirolimus (10 mg/kg given i.v. once/week for 5 weeks); and (ii) saline alone as a vehicle (given i.v. once/week for 5 weeks). Tumor volume was measured weekly (length × width × height). To create the A549 pleural dissemination model, 4 × 106 cancer cells were intrathoracically injected into the pleural cavity of 6-week-old male BALB/c nu/nu mice. After 7 days, the animals were randomized into two groups that consisted of eight mice each; (i) temsirolimus (10 mg/kg given i.p. once/week for 5 weeks); and (ii) saline alone as a vehicle. The drug was given once a week and lasted until the mice expired. Each animal experiment was repeated three times and the representative data is shown. The dose and schedule of temsirolimus treatment (10 mg/kg/week) in these animal experiments was decided based upon previous reports where the researchers used a range of 8–20 mg/kg/week.(23–25)
Immunohistochemistry. Surgically resected pleural membrane tissues from mice with disseminated pleural tumors from A549 cells were used for immunohistochemical study following procedures described previously.(26) Deparaffinized tissue sections were immersed in methanol containing 3% hydrogen peroxide to block endogenous peroxide activity. An autoclave pretreatment in citrate buffer was done for antigen retrieval. After incubation with a blocking buffer the sections were treated with an anti-phospho-mTOR rabbit mAb (Cell Signaling Technology) for 6 h at room temperature followed by immunobridging with Avidin DH-biotinylated HRP complex (Nichirei, Tokyo, Japan). Signal detection was done for 2–5 min using 3,3′-diaminobenzidine tetrahydrochloride dissolved to 50 mM/L Tris–HCl (pH 7.5) containing 0.001% hydrogen peroxide. The sections were counterstained with Mayer-hematoxylin. Monoclonal anti-human mouse Ki-67 antibody (MIB-1; Dako Cytomation) was used to calculate the Ki-67 labeling index by counting the number of positively stained cells per 1000 cancer cell nuclei for each section.
Statistical analysis. Student’s t-test was used to compare data between two groups. Data represent the mean ± SD. Overall survival was calculated using the Kaplan–Meier method and compared by the log–rank test. P < 0.05 was considered statistically significant.
Inhibition of mTOR by temsirolimus suppresses cell growth of NSCLC cells. First, we examined how effective temsirolimus was at inhibiting the proliferation of cultured NSCLC cells using a Trypan blue exclusion assay. As shown in Figure 1, temsirolimus suppressed the cell proliferation of A549, H1299, and H358 cells in a dose-dependent manner. The IC50 values were measured to examine the suppression of cell proliferation by temsirolimus in NSCLC cell lines. The IC50 for A549 cells was 0.76 nM and those for H1299 and H358 were 0.75 nM and 0.64 nM, respectively (Fig. 1). These data indicated that temsirolimus effectively inhibited the viability of NSCLC cells at a low concentration of <1 nM.
Temsirolimus inhibited mTOR pathway in a dose- and time-dependent manner. Next, in order to evaluate the effect of temsirolimus on regulating activation in the mTOR pathway, we examined the phosphorylation of mTOR and its downstream effectors by Western blot analysis. As expected, the treatment with temsirolimus suppressed the activations of mTOR, p70 ribosomal S6 kinase, and S6 in a dose-dependent fashion in A549 (Fig. 2A). This inhibitory effect occurred after 1 h, and lasted at least 4 h (Fig. 2B). Similar results were obtained in another NSCLC cell line, H1299 (Fig. 2C,D). Furthermore, we assessed the expression status of cell cycle markers including p21cip1, p27kip1, and cyclinD1, whose expression is often modified by the inactivation of p70 S6 kinase and S6.(27) Interestingly, p21cip1 was apparently induced by temsirolimus treatment, but we did not observe any change in cyclinD1 or p27kip1 (Fig. S1). Because the upregulation of p21cip1 is known to contribute to cell cycle arrest, these data suggest that the inhibition of the mTOR pathway using temsirolimus is a promising strategy to diminish the proliferation of NSCLC cells.
Temsirolimus treatment leads to G1 cell cycle arrest but not cell death. Our next question was whether temsirolimus treatment is lethal to NSCLC cells. In order to answer this question, we carried out flow cytometry to analyze the cell cycle distribution in A549 and H1299 cell lines under temsirolimus treatment. Interestingly, temsirolimus treatment increased the cell population in the G0/G1 phase, but not in the sub-G0 phase, which accounted for dead cells (Fig. 3). Furthermore, when we examined the amount of apoptosis by TUNEL assay, we did not observe a significant number of apoptotic cells (data not shown). Taken together, these results suggested that temsirolimus suppressed NSCLC cell proliferation by its cytostatic effect, not by cytotoxicity.
Temsirolimus reduces s.c. tumor growth of NSCLC cells. Next, we investigated the effect of temsirolimus on in vivo tumor growth. A549 s.c. xenografts were made. When 10 mg/kg temsirolimus was given weekly i.v. to the mice bearing the s.c. tumor, a significant delay of s.c. tumor growth was observed on day 35 (tumor volume: vehicle vs temsirolimus = 1239 vs 698 cm3; P < 0.05) (Fig. 4). None of the mice died of drug-induced toxicity and no other significant adverse events were observed. Moreover, during the observation period (up to 35 days after cell inoculation), there was no significant change in body weight in either group (data not shown).
Temsirolimus treatment prolonged survival of mice with disseminated pleural tumors of NSCLC cells. As shown above, we found that temsirolimus had a cytostatic effect on NSCLC cells and showed a delay of s.c. tumor growth. Based on these results, we predicted that a major advantage of temsirolimus treatment would be an improvement in the survival of patients bearing NSCLC tumors, similar to renal cell carcinoma.(28) To investigate the effect of temsirolimus on survival, we made a pleural dissemination animal model by injecting A549 cells into the intrapleural cavity, to mimic an advanced clinical stage of NSCLC. A weekly i.p. injection of 10 mg/kg temsirolimus significantly prolonged the survival period of these pleural disseminated tumor-bearing mice (median survival: vehicle vs temsirolimus = 53.5 vs 72.5 days; P < 0.05) (Fig. 5A,B).
Macroscopic observation by opening the thoracic cavity of the mice showed that temsirolimus treatment obviously reduced the number and the volume of pleural disseminated tumors on day 21 after the inoculation of A549 cells in the thoracic cavity (Fig. 5C,D), although tumors were recognized in a bilateral thoracic cavity regardless of temsirolimus treatment. This result led us to speculate that temsirolimus reduced the growth of pleural disseminated tumors, leading to the prolonged survival of the tumor-bearing mice. Furthermore, immunohistochemical analysis revealed that phosphorylation of mTOR was strongly suppressed in the tumor tissues of the temsirolimus-treated mice (Fig. 5E,F).
The immunohistochemical analysis for Ki-67 using disseminated pleural tumor tissues revealed a significant decrease in the number of proliferating cells (determined by calculating the Ki-labeling index, defined in Materials and Methods) in the tissues treated with temsirolimus (temsirolimus, 0.106 ± 0.019; control, 0.191 ± 0.044; P < 0.05) (Fig. S2A). However, temsirolimus treatment did not increase the incidence of apoptosis in the tumor tissues, as checked by immunohistochemistry for cleaved caspase-3 (temsirolimus, 0.004 ± 0.002; control, 0.004 ± 0.002; P > 0.05) (Fig. S2B). These results were similar to our in vitro data, supporting our conclusion that the primary effect of temsirolimus is antiproliferative rather than cytotoxic. Thus, the advantage of in vivo temsirolimus treatment was to provide prolonged survival in advanced NSCLC tumor-bearing mice by suppressing tumor growth.
Inhibition of mTOR by temsirolimus suppresses the action of hypoxia inducible factor 1α (HIF-1α). Finally, we assessed the inhibition of mTOR by temsirolimus in NSCLC cells and tumors. Because recent reports have shown that the action of HIF-1α, a major transcriptional activator for angiogenesis and oncogenes, is regulated by the mTOR pathway,(29) and is therefore inhibited by temsirolimus in vitro and in vivo,(25,30) we also determined the effect of temsirolimus on the expression status of HIF-1α in the nuclei, where activated HIF-1α normally translocates.(31) Temsirolimus treatment suppressed the translocation of HIF-1α to the nucleus in all of NSCLC cells (Fig. S3A). As HIF-1α is known to play a critical role in cell proliferation and angiogenesis,(32) this inhibition of HIF-1α action by temsirolimus should at least partially contribute to its antiproliferative effect.
Regarding the antiangiogenic effect of temsirolimus by negatively regulating HIF-1α, we additionally determined the expression of vascular endothelial cell growth factor (VEGF), a known transcriptional target of HIF-1α. In cultured NSCLC cells, the amount of VEGF protein secreted in the culture medium was suppressed by temsirolimus treatment in a dose-dependent manner (Fig. S3B,C). Similarly, the production of VEGF mRNA expression, especially the 572-bp form of VEGF, was decreased in the pleural disseminated tumors of the mice that had temsirolimus treatment (Fig. S3D). The inhibition of HIF-1α/VEGF-mediated angiogenesis might also contribute to slowing tumor growth by temsirolimus treatment.
Temsirolimus, an analogue of rapamycin, is a new molecular targeted agent and was first approved for the treatment of renal cell carcinoma. In terms of NSCLC, it was reported that inhibiting mTOR with rapamycin revealed a growth inhibitory effect in some NSCLC cell lines.(33) Temsirolimus was developed as an improved derivative of rapamycin,(34) and our data indicated its effectiveness by showing its potent inhibitory effect on cell proliferation of cultured NSCLC cells at a low concentration (as low as 1 nM). Concerning the antiproliferative effect of temsirolimus, our results reproduced the results of a previous report using rapamycin, which induced cell cycle arrest at the G1 checkpoint and inhibited cell proliferation of murine NSCLC without inducing apoptosis.(23) In this study, temsirolimus suppressed the phosphorylations of p70 S6 kinase and S6 (Fig. 2). As the action of p70 S6 kinase and S6 is critical for cell cycle progression,(27,35) the cytostatic effect of temsirolimus can be at least partially explained by the importance of p70 S6 kinase to cell cycle progression. Phosphatase and tensin homolog deleted on chromosome 10 (PTEN) and Akt are also interesting molecules related to cell proliferation signals. A recent study using rapamycin(33) showed that the inhibition of mTOR by temsirolimus appeared to regain Akt activity (Fig. S1). According to a previous report,(36) PTEN was lost in H1299 cells by its promoter methylation, whereas it remained intact in A549 cells. Regardless of their PTEN expression, our data indicated the similar potent antiproliferative effects of temsirolimus on those cell lines (Fig. 1).
Using an animal model of pleural dissemination, a condition for human lung cancer patients with one of the worst survival rates, we observed that temsirolimus reduced the growth of both s.c. tumors and pleural disseminated tumors of NSCLC cells, and that the treatment significantly prolonged the survival of mice bearing disseminated pleural tumors (Fig. 5). It is noteworthy that the dose and schedule of temsirolimus treatment in this study followed those currently in clinical use for renal cell carcinoma, with no apparent adverse effects in the mice. Because this regimen has also been tolerated in several clinical studies for other cancers,(18–21) temsirolimus treatment might safely provide prolonged survival for advanced NSCLC patients, possibly due to its cytostatic effect.
One immunohistochemical study showed that there were differences in mTOR signaling activation depending on histological type.(14) According to that study, adenocarcinoma had more frequent ctivation of phosphorylated mTOR than squamous cell carcinoma. However, it was unclear what histological type of NSCLC temsirolimus treatment would be effective in clinical use. mTOR is frequently activated in adenocarcinoma, but the outcome differs depending on the mTOR expression.(37) In a future study, it would be intriguing to establish a more effective combination therapy with temsirolimus,(38) because mTOR activity can be modified by other effectors, such as growth factors(39) and nutrition.(40,41)
In conclusion, our data suggests that temsirolimus, with a cytostatic effect on cell proliferation, may be useful for NSCLC treatment in general and could give prolonged survival to advanced NSCLC cases with pleural dissemination specifically.
We are grateful to Mr. Toru Tanida and Ms. Tae Yamanishi for their technical assistance and to Drs. Junji Matsuoka, Minoru Haisa, and Seishi Nishitani of Okayama University (Okayama, Japan), and Dr. Motowo Nakajima of Johnson and Johnson K.K. (Tokyo, Japan) for useful discussions.
The authors have no conflicts of interest to declare.