Lung squamous cell carcinoma (LUSC) accounts for one of three of non–small cell lung carcinoma (NSCLC) and 30% of LUSC patients present with locally advanced, unresectable/medically inoperable disease, who are commonly treated with definitive chemoradiation. However, disease relapse in the radiation fields occurs in one of three cases. We aim to explore the underlying molecular mechanisms of chemoradiation resistance of LUSC. Patient-derived xenograft (PDX) models of LUSC were established in immunodeficient mice, followed by treatment with cisplatin in combination with clinically relevant courses of ionizing radiation (20, 30, and 40 Gy). The recurrent tumors were extracted for functional proteomics using reverse phase protein analysis (RPPA). We found that phospho-AKT-S473, phospho-AKT-T308, phospho-S6-S235/6, and phospho-GSK3β-S9 were upregulated in the chemoradiation-resistant 20 Gy + cisplatin and 40 Gy + cisplatin tumors compared with those in the control tumors. Ingenuity pathway analysis of the RPPA data revealed that AKT–mTOR signaling was the most activated signaling pathway in the chemoradiation-resistant tumors. Similarly, elevated AKT–mTOR signaling was observed in stable 40 Gy and 60 Gy resistant HARA cell lines compared with the parental cell line. Accordingly, pharmacologic inhibition of mTOR kinase by Torin2 significantly sensitized LUSC cell lines to ionizing radiation. In conclusion, using chemoradiation-resistant PDX models coupled with RPPA proteomics analysis, we revealed that deregulation of AKT–mTOR signaling may contribute to the chemoradiation resistance of LUSC.

Implications:

Clonal selection of subpopulations with high AKT–mTOR signaling in heterogeneous tumors may contribute to relapse of LUSC after chemoradiation. mTOR kinase inhibitors may be promising radiosensitizing agents in upfront treatment to prevent acquired resistance.

Lung cancer is the leading cause of cancer-related death worldwide with a five-year survival rate of ∼19%. Non–small cell lung carcinoma (NSCLC) represents 85% of lung cancer, with the majority of cases consisting of lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), and less frequent histologies (1–5). PD1/PD-L1 immune-checkpoint blockade has improved the clinical outcome of lung cancer patients; however, the therapeutic window is still narrow. For example, the response rate to immunotherapy is only ∼20% in NSCLC (6–8). Targeted therapies have improved outcomes for certain subtypes of LUAD with mutations in oncogenic drivers (e.g., EGFR, ALK, ROS, RET, and BRAF), but most patients ultimately develop therapy resistance. Although multiple genome sequencing studies have established the driving genetic alterations in LUSC, there are no FDA-approved therapies to target these, and the majority of the lower frequency genetic aberrations (2). Therefore, there is an urgent need to develop effective treatment strategies for NSCLC, including LUSC, given high rates of disease recurrence.

LUSC is a distinct subtype of NSCLC, accounts for ∼25% to 30% of all lung cancers, and is associated with poor prognosis (1–5). Recent genomic, transcriptomic, and proteomic profiling studies have revealed four major hallmarks in LUSC. First, the majority of LUSCs have TP53 mutations (90%), as well as CDKN2A and RB1 mutations and silencing (80%), leading to loss of the G0 and G1 cell-cycle restriction points resulting in subsequent premature S phase entry and replication stress (9, 10). Second, LUSCs have aberrantly high mTOR signaling activity, mainly due to PTEN mutation and PIK3CA mutation and amplification (11). Third, there is cell differentiation deficiency in LUSC resulting from amplification of SOX2 and TP63 (12, 13). Fourth, there is a high level of oxidative stress response resulting from mutations of KEAP1 and NFE2L2 (9). How these characteristic events are coordinated to promote LUSC tumorigenesis and therapy resistance is still unknown. Moreover, how to target these characteristic genetic alterations is a current challenge for the management of LUSC patients.

Radiation with chemotherapy remains a major treatment option for LUSC (2). Common therapies for stage I–III LUSC include surgery (when possible), chemotherapy, and radiation, but many patients relapse despite these systemic therapies. About one of three LUSC patients present with locally advanced, unresectable/medically inoperable disease. Patients who present with locally advanced NSCLC are commonly treated with definitive chemoradiation (14). Even though an initial response to concurrent chemoradiation is commonly observed, disease relapse in the radiation fields occurs in up to one of three cases. However, the underlying molecular mechanisms of chemoradiation resistance of NSCLC remain to be determined.

In the present study, we have performed functional proteomics analyses using reverse phase protein array (RPPA) of our established chemoradiation-resistant patient-derived xenograft (PDX) models of LUSC. We found that AKT–mTOR signaling is the most upregulated signaling pathway in chemoradiation-resistant tumors. Accordingly, molecular targeting of mTOR kinase sensitized LUSC cells to radiation.

Antibodies, chemicals, and cell culture

HARA and HCC15 cell lines were from American Type Culture Collection (ATCC) and maintained at 37°C in 5% CO2 in RPMI-1640 medium (#11875093) supplemented with 10% fetal bovine serum (#11550, Sigma), 1% penicillin/streptomycin (Life Technologies). The authentication of HARA and HCC15 cell lines was confirmed by short tandem repeat genotyping (Identifier Kit, Applied Biosystems). The Universal Mycoplasma Detection Kit (ATCC) was used to detect mycoplasma in cell culture. The cells were kept in culture minimum two passages prior and maximum 20 passages when the experiments were performed. Cisplatin (NSC 119875, #S1166) and Torin2 (#S2817) were obtained from Selleck Chemicals and were dissolved in dimethyl formamide (DMF) and dimethyl sulfoxide (DMSO; Sigma), respectively, and added to medium with a final concentration of no more than 0.1% DMF/DMSO. ATM (#2873), phospho-ATM (Ser1981, #5883), CHK2 (#6334), phospho-CHK2 (Thr68, #2197), CHK1 (#2360), phospho-CHK1 (Ser345, # 2348), phospho-H2AX (S139, #9718), ATR (#13934), phospho-ATR (Ser428, #2853), AKT (#4691), phospho-AKT (Ser473, #4060), phospho-AKT (Thr308, #13038), S6 (#2217), phospho-S6 (Ser235/236, #4858), S6K1 (#2708), phospho-S6K1 (Thr389, #9234), GSK3β (#12456), phospho-GSK3β (Ser9, #5558), PTEN (#9188), and GAPDH (#2118) primary antibodies, and anti-rabbit (#7074) and anti-mouse (#7076) secondary antibodies were purchased from Cell Signaling Technology.

Immunoblotting

Immunoblotting was performed as described previously (15). Briefly, cell lysates were prepared using RIPA buffer (#89900, Thermo Fisher) supplemented with 1× protease inhibitors (cOmplete, #4693132001, Roche) and phosphatase inhibitors (PhosSTOP, @4906845001, Roche) followed by protein quantification with the Dc protein assay kit (Bio-Rad). Equal amounts of protein were loaded and resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes. Membranes were incubated in 5% fat-free milk in Tris buffered saline with 0.1% Tween-20 (TBST) blocking buffer for 1 hour at room temperature. Primary antibodies with dilution of 1:200–4,000 were allowed to bind overnight at 4°C, or for 2 hours at room temperature. After washing in TBST, the membranes were incubated with horse radish peroxidase conjugated secondary antibodies at a 1:5,000 dilution for 1 hour at room temperature. Membranes were washed with TBST and allowed to ECL detection, followed by X-ray film recording.

Radiation clonogenic assay

Radiation clonogenic assay was performed as previously described (16). Briefly, cells were trypsinized to generate a single-cell suspension and seeded in 60-mm tissue culture plates in triplicate. Cells were incubated with DMSO or Torin2 for 3 hours and then irradiated with various doses (0–8 Gy). Radiation was performed with 160 kV, 25 mA at a dose rate of approximately 113 cGy/min using an RS-2000 biological irradiator with a 0.3-mm copper filter (RadSource). Twenty-four hours after radiation, the medium was replaced with fresh medium without DMSO or Torin2. Seven to 10 days after seeding, colonies were fixed with methanol/acetic acid, and stained with 0.5% crystal violet. The number of colonies containing at least 50 cells was counted using a dissecting microscope (Leica Microsystems, Inc). Experiments were repeated at least three independent times, and a representative experiment was chosen.

In vivo studies

Animal studies were conducted in accordance with an approved protocol adhering to the IACUC policies and procedures at The Ohio State University. The LUSC PDX model in Fig. 1 was derived from the primary tumor of one treatment-naïve patient who underwent lung resection at The Ohio State University Medical Center. The additional LUSC CD1168 PDX model was obtained from the NCI PDX repository (NCI tumor model #997726-040-R). Six- to eight-week-old male NOD-SCID gamma (NSG) or athymic nude mice (Taconic Farms Inc.) were caged in groups of five or fewer and fed with a diet of animal chow and water ad libitum. PDX tumor tissues were surgically implanted subcutaneously into the flanks of each mouse, and the wound was closed with a wound clip. Mice were randomized to start different treatment regimens once tumors reached 200 to 250 mm3 in size. Cisplatin (2 mg/kg) in 0.9% saline was administered via intraperitoneal injection two times a week. Using a custom shielding apparatus to block nontargeted areas, 2 Gy/day of radiation was administered directly to tumors daily for five days a week. For combination treatment, cisplatin was injected one to three hours before radiation. For the 20-, 30-, and 40-Gy groups, the mice were treated for two, three, and four weeks, respectively. To obtain a tumor growth curve, perpendicular diameter measurements of each tumor were measured one to two times every week from the start of treatment with digital calipers, and volumes were estimated by the formula (L × W × W)/2 (mm3).

Figure 1.

Establishment of chemoradiation-resistant LUSC PDX models. A, Schema of experimental flow. B, The growth of tumors in each individual mouse in each group (n = 5 mice per group). *, ERC, early removal criterion or end of study; **, found dead; ***, died during treatment. Arrows indicate time for sacrifice and tumor collection for RPPA.

Figure 1.

Establishment of chemoradiation-resistant LUSC PDX models. A, Schema of experimental flow. B, The growth of tumors in each individual mouse in each group (n = 5 mice per group). *, ERC, early removal criterion or end of study; **, found dead; ***, died during treatment. Arrows indicate time for sacrifice and tumor collection for RPPA.

Close modal

RPPA

When tumors reached early removal criterion, the tumors in the parental (untreated) and chemoradiation-resistant groups were removed from mice and processed for protein extraction with RIPA buffer. Two tumors each from parental, 20-Gy group and 40-Gy groups were selected. Total protein (200 μg per sample) was made following standard RPPA protein extraction protocols and then shipped in dry ice for RPPA at MD Anderson Cancer Center (Houston, TX). Three replicates per sample were sent for RPPA profiling. The data were analyzed at the Bioinformatics Department of OSUMC (A.W.).

Differential protein abundance and ingenuity pathway analysis (IPA)

The normlinear RPPA abundance values were log2 normalized with voom and limma was used to compare 20 Gy versus control, and 40 Gy versus control using “duplicateCorrelation” to fit a random effect for the blocking variable to account for technical sample replicates. Dendograms were generated to show relationships between samples using all measured values. Protein name, phospho log2 ratio, and phospho site for each significant gene list (20 Gy vs. control FDR < 0.05, 20 Gy vs. control P < 0.05, 40 Gy vs. control FDR < 0.05, 40 Gy vs. control P < 0.05) were analyzed using IPA core analysis.

Establishment of stable cell lines

With the parental and chemoradiation-resistant PDX tumors, primary cells were established with the methods using ROCK inhibitors (17). Radiation-resistant HARA cell lines were established by sequential treatment of the cells with 4, 6, 8, 10, and 12 Gy radiation to obtain the HARA 40 Gy–resistant cell line, which was further treated with additional 20 Gy radiation to obtain the HARA 60 Gy–resistant cell line (18).

Statistical analysis

Data are presented as the mean ± standard error of the mean (SEM) for clonogenic survival. Statistical comparisons were made between the control and experimental conditions using the unpaired two-tailed Student t test (e.g., individual timepoints), or two-way ANOVA for the whole groups with significance assessed at P < 0.05.

Establishment and validation of chemoradiation-resistant LUSC PDX tumors

To establish chemoradiation-resistant LUSC PDX tumors, parental PDX tumors were implanted in the flanks of NSG mice. When PDX tumors reached 200 to 250 mm3 (typically six weeks after implantation), the mice were divided into control group (0-Gy), 20-, 30-, and 40-Gy groups, and were treated with cisplatin and radiation for 2, 3, or 4 weeks, respectively (Fig. 1A). In the control group, the tumors grew very rapidly, and all mice were sacrificed within four months. There were no apparent recurrent tumors in the 30-Gy group; the stochastic nature of tumor chemoradiotherapy resistance and recurrence, as well as the limited number of mice in the experiments (5–6 mice/group), may have contributed to the lack of the recurrent tumors in this group. Several mice developed recurrent tumors in the 20- and 40-Gy groups (Fig. 1B), with a total of 4 of 15 (27%) mice (2 each from 20 Gy and 40 Gy groups) recurring after chemoradiation. To validate the chemoradiation resistance of the recurrent PDX tumors, the tumors from the control and 20-Gy groups were implanted in athymic mice with five mice in each group. When the tumors reached 200 to 250 mm3, all the mice were treated with cisplatin and radiation for two weeks (Fig. 2A). Within the first four weeks, the tumors in the parental (radiation-naïve) group steadily regressed after chemoradiation, whereas the tumors in the recurrent group did not regress after chemoradiation. By seven to eight weeks or so, the majority of tumors in both groups had recurred (Fig. 2B).

Figure 2.

Validation of chemoradiation-resistant LUSC PDX models in athymic mice. A, Schema of experimental flow. B, The growth of tumors in each group (n = 5 mice per group). ERC, early removal criterion. *, P < 0.05 (individual timepoints); **, P < 0.001 (whole group).

Figure 2.

Validation of chemoradiation-resistant LUSC PDX models in athymic mice. A, Schema of experimental flow. B, The growth of tumors in each group (n = 5 mice per group). ERC, early removal criterion. *, P < 0.05 (individual timepoints); **, P < 0.001 (whole group).

Close modal

AKT–mTOR signaling is elevated in chemoradiation-resistant LUSC tumors

The recurrent tumors in control, 20- and 40-Gy groups were collected and subjected to functional proteomics using RPPA at MD Anderson Cancer Center (Fig. 1B). IPA core analysis was done on each significant protein lists. Based on P < 0.05, there were 52 significantly altered proteins in 20 Gy versus control, and 32 significantly altered proteins in 40 Gy versus control (Fig. 3A). After adjusting P values for false discovery rate (FDR) < 0.05, we found 15 significantly altered proteins in 20 Gy vs. control, and two significantly altered proteins in 40 Gy versus control (Fig. 3B). Despite some overlap, there were apparent difference of protein expression patterns between 20- and 40-Gy groups. These results demonstrated distinct protein expression profiles between parental and chemoradiation-resistant LUSC tumors. Our results also indicate that there are different protein expression patterns in recurrent LUSC tumors treated with different radiation dosages. A dendrogram showed that technical replicates clustered together and control samples clustered closer to 40-Gy samples than 20-Gy samples (Supplementary Fig. S1). This may explain why there were fewer proteomic significant differences detected between 40 Gy and control than between 20 Gy and control.

Figure 3.

Protein expression heat maps from RPPA. A, Protein expression profiles between 20 Gy recurrent and control models, and between 40 Gy recurrent and control models for proteins with P < 0.05. B, Protein expression profiles between 20 Gy recurrent and control models, and between 40 Gy recurrent and control models for proteins with adjusted P (FDR) < 0.05. Protein expression is voom log2 normalized and scaled across samples. Red: above average; blue: below average.

Figure 3.

Protein expression heat maps from RPPA. A, Protein expression profiles between 20 Gy recurrent and control models, and between 40 Gy recurrent and control models for proteins with P < 0.05. B, Protein expression profiles between 20 Gy recurrent and control models, and between 40 Gy recurrent and control models for proteins with adjusted P (FDR) < 0.05. Protein expression is voom log2 normalized and scaled across samples. Red: above average; blue: below average.

Close modal

We performed pathway analysis by IPA using the altered protein data set with P < 0.05. Table 1 shows the top-10 altered signaling pathways in 20 Gy– and 40 Gy–resistant groups compared with the control group. Sirtuin signaling was downregulated, whereas nine signaling pathways were upregulated in the 20 Gy– and 40 Gy–resistant groups compared with the control group. We found PTEN (downregulation) and PI3K/AKT (upregulation) signaling pathways were among the top-5 altered canonical pathways in the 20 Gy–resistant group compared with the control group (Table 2). Our analyses revealed that PI3K/AKT/mTOR signaling was the most upregulated signaling pathway in 20 Gy–resistant group compared with the control group (pathway shown in Supplementary Fig. S2). IPA of molecular and cellular functions (Supplementary Table S1) showed that most of the altered proteins are involved in cell growth, proliferation, and survival.

Table 1.

The top-10 commonly altered signaling pathways in 20 Gy– and 40 Gy–resistant groups compared with the control group based on P < 0.05.

Canonical pathwaysLimma results 20 Gy vs. controlLimma results 40 Gy vs. control
mTOR signaling 2.53 2.236 
Regulation of eIF4 and p70S6K signaling 2.449 
JAK/Stat signaling 2.121 
Acute myeloid leukemia signaling 2.828 
FLT3 signaling in blood progenitor cells 2.828 1.633 
ILK signaling 1.265 1.414 
IL15 signaling 2.714 1.342 
IL3 signaling 2.449 1.342 
p70S6K signaling 2.333 1.342 
Sirtuin signaling pathway −2.828 −1.342 
Canonical pathwaysLimma results 20 Gy vs. controlLimma results 40 Gy vs. control
mTOR signaling 2.53 2.236 
Regulation of eIF4 and p70S6K signaling 2.449 
JAK/Stat signaling 2.121 
Acute myeloid leukemia signaling 2.828 
FLT3 signaling in blood progenitor cells 2.828 1.633 
ILK signaling 1.265 1.414 
IL15 signaling 2.714 1.342 
IL3 signaling 2.449 1.342 
p70S6K signaling 2.333 1.342 
Sirtuin signaling pathway −2.828 −1.342 

Note: Value reports z score calculated based on the direction of protein abundance fold change and the known effect that this type of change should make to a given pathway. Positive number means upregulation; negative number means downregulation.

Table 2.

The top-5 altered canonical pathways in 20 Gy resistant group compared with control group based on P < 0.05.

NameP valueOverlap
PTEN signaling 3.87e−20 10.3% 13/126 
IL15 signaling 6.92e−19 14.7% 11/75 
PI3K/AKT signaling 3.14e−18 7.4% 13/175 
B-cell receptor signaling 6.56e−18 7.0% 13/185 
Estrogen receptor signaling 8.15e−18 4.6% 15/328 
NameP valueOverlap
PTEN signaling 3.87e−20 10.3% 13/126 
IL15 signaling 6.92e−19 14.7% 11/75 
PI3K/AKT signaling 3.14e−18 7.4% 13/175 
B-cell receptor signaling 6.56e−18 7.0% 13/185 
Estrogen receptor signaling 8.15e−18 4.6% 15/328 

Note: P value calculated based on likelihood of observing the number of molecules in the provided gene list based on the number of proteins in pathway and how likely this could happen by chance. The overlap reports number of proteins from 20 Gy list divided by the total number in pathway.

The above RPPA and IPA demonstrated that mTOR signaling is the most activated pathway in chemoradiation-resistant LUSC. Further analysis of commonly altered protein expression among 20 and 40 Gy recurrent models revealed that markers of AKT–mTOR signaling were among the top hits, including pAKT-S473, pAKT-T308, pGSK3β-S9, pRICTOR-T1135, pS6-S240/244, and pS6-S235/236 (Fig. 3). We next validated mTOR signaling upregulation in chemoradiation-resistant LUSC tumors by immunoblotting. We confirmed increased activation of the AKT–mTOR pathway through elevated pAKT-S473, pAKT-T308, pGSK3β-S9, and pS6K1-T389 in 20 and 40 Gy recurrent tumors compared with control tumors, but no alteration of PTEN levels (Fig. 4A). Similar upregulation of pAKT-S473, pAKT-T308, pGSK3β-S9, and pS6K1-T389 signals was observed in tumors surviving chemoradiation in another LUSC PDX model (CD1168), which was treated with cisplatin + 20 Gy over two weeks (Supplementary Fig. S3). mTOR kinase forms two distinct complexes, mTORC1 and mTORC2. mTORC2 is upstream of AKT and directly phosphorylates AKT at S473, whereas pGSK3β-S9 is a direct target of AKT kinase. mTORC1 is downstream of AKT and phosphorylate S6K1 at T389, which in turn phosphorylates S6 at S235/236 (19, 20). Our results demonstrated that both mTORC1 and mTORC2 signaling are elevated in chemoradiation-resistant LUSC tumors.

Figure 4.

AKT–mTOR signaling is elevated in chemoradiation-resistant LUSC tumors. A, Validation of the RPPA data by immunoblotting. Tumors from untreated, 20 Gy and 40 Gy chemoradiation-resistant groups were prepared for immunoblotting of the indicated proteins. B, Primary cells from the tumors of untreated and 20 Gy chemoradiation-resistant groups were prepared for immunoblotting of the indicated proteins. C, Parental (radiation-naïve) HARA and 40 Gy and 60 Gy radiation–resistant cell lines were prepared for immunoblotting of the indicated proteins. RS, radioresistant. D, Model of mTOR signaling in resistance to chemotherapy and radiotherapy.

Figure 4.

AKT–mTOR signaling is elevated in chemoradiation-resistant LUSC tumors. A, Validation of the RPPA data by immunoblotting. Tumors from untreated, 20 Gy and 40 Gy chemoradiation-resistant groups were prepared for immunoblotting of the indicated proteins. B, Primary cells from the tumors of untreated and 20 Gy chemoradiation-resistant groups were prepared for immunoblotting of the indicated proteins. C, Parental (radiation-naïve) HARA and 40 Gy and 60 Gy radiation–resistant cell lines were prepared for immunoblotting of the indicated proteins. RS, radioresistant. D, Model of mTOR signaling in resistance to chemotherapy and radiotherapy.

Close modal

mTOR signaling is upregulated in radiation-resistant HARA cells

To support that mTOR signaling is upregulated in chemoradiation resistance PDX tumors, primary cells were established from the control and 20-Gy groups using the ROCK inhibitor (17). Compared with the cells from parental tumors, we again observed elevated pAKT-S473, pAKT-T308, and pGSK3β-S9 levels without PTEN alteration (Fig. 4B). We next established 40 Gy and 60 Gy ionizing radiation–resistant HARA cell lines. In comparison with parental HARA cells, there was increased phosphorylation of pAKT-S473, pAKT-T308, and pGSK3β-S9 without PTEN alteration in both 40 Gy and 60 Gy ionizing radiation–resistant HARA cell lines. Of note, there were higher levels of pAKT-S473, pAKT-T308, and pGSK3β-S9 in 60 Gy than those in 40 Gy ionizing radiation–resistant HARA cells, suggesting a dose-dependent effect on AKT–mTOR activation (Fig. 4C). This finding is consistent with the upregulation of pAKT-S473 and pAKT-T308 in 40 Gy recurrent tumors with FDR < 0.05 when compared with control tumors (Fig. 3B). These results indicate that AKT–mTOR signaling is increased in radiation-resistant LUSC cells independently of PTEN.

mTOR kinase inhibitor Torin2 sensitizes LUSC cells to radiotherapy

Mutations of PTEN or PIK3CA lead to enhanced activity of AKT–mTOR signaling, which is one of the main cell growth-promoting signaling pathways (21). mTOR signaling is commonly deregulated across human cancer types, and inhibition of mTOR signaling is one of the main strategies being attempted in targeted cancer therapy (22, 23). Moreover, it has been well documented that mTOR signaling promotes DNA damage response to maintain cell survival in response to genotoxins (24–26). Our finding of mTOR signaling upregulation in chemoradiation-resistant LUSC indicates enhanced mTOR signaling may lead to tumor relapse, suggesting mTOR kinase inhibitors in combination with chemoradiation may result in better tumor control of LUSC patients (Fig. 4D). To test this hypothesis, we treated parental HARA cells with different concentrations of Torin2 (second generation of mTOR kinase inhibitor) alone or in combination with 4 Gy radiation. At two hours after radiation, the cells were processed for immunoblotting. We found that Torin2 potently inhibited pAKT-S473 and pS6-S235/236 at 10 nmol/L. Torin2 has also been shown to inhibit ATM and ATR at high concentrations. Indeed, radiation induced robust pATM-S1981 (and its downstream target pCHK2-T68), but was only reduced at 100 nmol/L and higher of Torin2. Similarly, Torin2 reduced pATR-S428 only at 100 nmol/L and higher (Fig. 5A). To further support that low Torin2 concentration inhibits mTOR but not ATM and ATR, we treated HARA cells with different concentrations of Torin2 alone or in combination with 300 μmol/L hydroxyurea to induce replication stress (and activate ATR/CHK1). At two hours after hydroxyurea, the cells were processed for immunoblotting. Consistently, we found that Torin2 inhibited hydroxyurea-mediated pCHK1-S345 (ATR downstream target) and pATM-S1981 only at concentrations of 100 nmol/L and higher (Fig. 5B). Finally, we performed radiation clonogenic assays with 10 nmol/L Torin2 in HARA and HCC15 cells. Torin2 potently sensitized both HARA and HCC15 cells to radiation (Fig. 5C).

Figure 5.

mTOR inhibitor Torin2 sensitizes LUSC cells to radiotherapy. A, B, Immunoblotting of HARA cells treated with Torin2 alone or in combination with radiation (A) or hydroxyurea (B). C, Radiation clonogenic assays of HARA and HCC15 cells treated with 10 nmol/L Torin2. D, Radiation clonogenic assays of HARA parental, HARA 40 Gy radioresistant and HARA 60 Gy radioresistant cells treated with 10 nmol/L Torin2. E, Cell viability of HARA parental, HARA 40 Gy radioresistant and HARA 60 Gy radioresistant cells treated with Torin2 using AlamarBlue assay. F, Immunoblotting of HARA 40 Gy and HARA 60 Gy radioresistant cells treated with increasing concentrations of Torin2 alone or in combination with 4 Gy of radiation. IR, ionizing radiation; HU, hydroxyurea; RT, radiotherapy; DER, dose enhancement ratio.

Figure 5.

mTOR inhibitor Torin2 sensitizes LUSC cells to radiotherapy. A, B, Immunoblotting of HARA cells treated with Torin2 alone or in combination with radiation (A) or hydroxyurea (B). C, Radiation clonogenic assays of HARA and HCC15 cells treated with 10 nmol/L Torin2. D, Radiation clonogenic assays of HARA parental, HARA 40 Gy radioresistant and HARA 60 Gy radioresistant cells treated with 10 nmol/L Torin2. E, Cell viability of HARA parental, HARA 40 Gy radioresistant and HARA 60 Gy radioresistant cells treated with Torin2 using AlamarBlue assay. F, Immunoblotting of HARA 40 Gy and HARA 60 Gy radioresistant cells treated with increasing concentrations of Torin2 alone or in combination with 4 Gy of radiation. IR, ionizing radiation; HU, hydroxyurea; RT, radiotherapy; DER, dose enhancement ratio.

Close modal

To test whether inhibition of mTOR signaling sensitizes radioresistant LUSC cells to radiation, we first assayed whether the 40 and 60 Gy radioresistance cell lines were indeed radioresistant. Our clonogenic assays confirmed enhanced radioresistance in the 40 Gy– and 60 Gy–selected tumor cells compared with that of parental HARA cells (Fig. 5D). We next assessed Torin2 sensitivity in parental, 40 Gy, and 60 Gy HARA cell lines with an AlamarBlue assay. Our results demonstrated that radioresistant 40 and 60 Gy cells were relatively more sensitive to Torin2 compared with parental HARA cells; however, there was no statistically significant difference in IC50 values (Fig. 5E). Furthermore, when the 40 and 60 Gy radioresistant cell lines were treated with 10 nmol/L Torin2, we found that Torin2 sensitized these radioresistant cell lines to radiation (Fig. 5D). We next treated radioresistant HARA 40 and 60 Gy radioresistant cells with different concentrations of Torin2 alone or in combination with 4 Gy ionizing radiation. Our immunoblotting results in Fig. 5F (two hours after radiation) showed that Torin2 at 10 nmol/L potently inhibited pAKT-S473 and pS6-S235/236 in both 40 Gy– and 60 Gy–resistant cells without IR treatment. Interestingly, 10 nmol/L of Torin2 only partially reduced pAKT-S473 signaling in both 40 Gy– and 60 Gy–resistant cells treated with IR, whereas pS6-S235/236 was completely inhibited. These results suggest potential differential alteration of mTORC1 and mTORC2 signaling in the acquired radioresistant 40 Gy and 60 Gy cells, although the underlying mechanisms are still not clear.

Using chemoradiation-resistant PDX models coupled with RPPA proteomics analysis, we revealed distinct patterns of protein expression between parental and chemoradiation-resistant LUSC tumors. Among these altered proteins, components of AKT–mTOR signaling were the most upregulated in recurrent tumors. Moreover, the upregulation of the AKT–mTOR signaling pathway was also observed in radiation-resistant LUSC cells in vitro. Consistent with these findings, targeting mTOR kinase potently sensitized LUSC cells to ionizing radiation.

Lung cancer remains the leading cause of cancer-related death worldwide with a five-year survival rate of ∼19% (3). LUSC is a distinct subtype of NSCLC and accounts for ∼25% to 30% of all lung cancers (2). Although targeted therapies have improved outcomes for subtypes of LUAD with certain mutations in oncogenic drivers (e.g., EGFR, ALK, ROS, RET, and BRAF), few targeted therapies exist for patients with LUSC (2). Though multiple genome sequencing studies have established that the most mutated genes in LUSC are TP53, CDKN2A, PTEN, PIK3CA, KEAP1, MLL2, HLA-A, NFE2L2, NOTCH1, and RB1, there are no FDA-approved therapies to target these genetic aberrations (9). Current therapy options for stage I–III LUSC are surgery (when possible), chemotherapy, and radiation, but many patients relapse despite these therapies (9). Chemoradiation resistance is likely driven by a subpopulation of therapy-resistant tumor cells including cancer stem cells (CSC; Fig. 4D; ref. 27). To explore the underlying molecular mechanisms of chemoradiation resistance of LUSC, we have performed functional proteomics analyses using RPPA of our established chemoradiation-resistant PDX models of LUSC. In doing so, we found that AKT–mTOR signaling was dramatically upregulated in the chemoradiation-resistant tumors when compared with the control tumors. Similar upregulation of AKT–mTOR signaling was found in radiation resistant 40 and 60 Gy LUSC cells, with even higher AKT–mTOR signaling activation in 60 Gy– compared with 40 Gy–resistant cells. As previously noted, one genetic hallmark of LUSC is the aberrantly high AKT–mTOR signaling activity due to high rates of PTEN mutation and PIK3CA mutation and amplification (11). Our data suggest an important role for AKT–mTOR signaling in LUSC chemoradiation resistance.

Almost all LUSC have somatic mutations of TP53 and inactivation of CDKN2A/RB1 to render LUSC cells dependent on S phase and G2–M phase checkpoints for survival (9, 10). The highly constitutive activation of replication stress response and DNA damage repair signaling coupled with loss-of-function of G1 checkpoint indicates that LUSC cell survival may rely on ATR/ATM signaling, the critical DNA replication and damage checkpoints in response to chemotherapy and radiotherapy (28, 29). It has been well established that mTOR signaling maintains cancer cell survival by suppressing endogenous DNA damage and replication stress via promoting DNA damage repair (24–26, 30, 31), supporting that pharmacologic inhibition of mTOR signaling leads to DNA damage and sensitizes cancer cells to chemotherapy and radiotherapy (25, 32). Consistent with that, we find that pharmacologic inhibition of mTOR kinase significantly sensitizes LUSC cell lines to ionizing radiation. Therefore, it is likely that in a heterogeneous LUSC tumor, there are subpopulations of cells with high mTOR signaling activity that increases and maintains DNA damage repair capacity, enabling clonal selection.

Furthermore, emerging data have shown that mTOR signaling maintains CSC biology in various cancer types (33, 34). It has been well documented that LUSC has CSC contributing to relapse after chemoradiotherapy (35–38). It is, therefore, possible that in a heterogeneous LUSC tumor, there is a subpopulation of CSC with high activity of mTOR signaling that facilitates survival after chemoradiation (39). It will be important to compare CSC biomarkers between chemoradiation-resistant tumors and the untreated control tumors. Finally, it has been well documented that CSC possess enhanced DNA repair capacity (40). Thus, our findings suggest that it is possible that elevated mTOR signaling maintains LUSC CSC by promoting DNA repair leading to chemoradiation resistance, and targeting mTOR signaling may sensitize LUSC to chemoradiation by eradicating CSC (41, 42).

In conclusion, our studies point to important roles of AKT–mTOR signaling in promoting the resistance of LUSC to DNA damaging–based therapies and provide strong scientific premise to develop and test novel targeted therapies in LUSC utilizing agents that target mTOR signaling.

No disclosures were reported.

C. Shen: Conceptualization, formal analysis, supervision, investigation, methodology, writing–original draft, writing–review and editing. D.-L. Shyu: Formal analysis, investigation, methodology. M. Xu: Validation, investigation, methodology. L. Yang: Formal analysis, investigation, methodology, writing–review and editing. A. Webb: Resources, data curation, software, formal analysis, methodology, writing–review and editing. W. Duan: Conceptualization, resources, writing–review and editing. T.M. Williams: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, writing–original draft, project administration, writing–review and editing.

Research reported in this publication was supported by The Ohio State University Comprehensive Cancer Center (OSU-CCC) and the NIH under grant number P30 CA016058. Additional funding (to T.M. Williams) from RSG-17-221-01-TBG, award number grant KL2TR001068 from the National Center for Advancing Translational Sciences.

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

1.
Morgensztern
D
,
Devarakonda
S
,
Govindan
R
.
Genomic landscape of squamous cell carcinoma of the lung
.
Am Soc Clin Oncol Educ Book
2013
;
348
53
.
2.
Gandara
DR
,
Hammerman
PS
,
Sos
ML
,
Lara
PN
Jr
,
Hirsch
FR
.
Squamous cell lung cancer: from tumor genomics to cancer therapeutics
.
Clin Cancer Res
2015
;
21
:
2236
43
.
3.
Siegel
RL
,
Miller
KD
,
Jemal
A
.
Cancer statistics, 2019
.
CA Cancer J Clin
2019
;
69
:
7
34
.
4.
Thomas
A
,
Liu
SV
,
Subramaniam
DS
,
Giaccone
G
.
Refining the treatment of NSCLC according to histological and molecular subtypes
.
Nat Rev Clin Oncol
2015
;
12
:
511
26
.
5.
Sanchez-Danes
A
,
Blanpain
C
.
Deciphering the cells of origin of squamous cell carcinomas
.
Nat Rev Cancer
2018
;
18
:
549
61
.
6.
Topalian
SL
,
Hodi
FS
,
Brahmer
JR
,
Gettinger
SN
,
Smith
DC
,
McDermott
DF
, et al
.
Safety, activity, and immune correlates of anti-PD-1 antibody in cancer
.
New Engl J Med
2012
;
366
:
2443
54
.
7.
Brahmer
JR
,
Tykodi
SS
,
Chow
LQM
,
Hwu
WJ
,
Topalian
SL
,
Hwu
P
, et al
.
Safety and activity of anti-PD-L1 antibody in patients with advanced cancer
.
New Engl J Med
2012
;
366
:
2455
65
.
8.
Ferris
RL
,
Blumenschein
G
,
Fayette
J
,
Guigay
J
,
Colevas
AD
,
Licitra
L
, et al
.
Nivolumab for recurrent squamous-cell carcinoma of the head and neck
.
New Engl J Med
2016
;
375
:
1856
67
.
9.
Cancer Genome Atlas Research Network
.
Comprehensive genomic characterization of squamous cell lung cancers
.
Nature
2012
;
489
:
519
25
.
10.
Macheret
M
,
Halazonetis
TD
.
Intragenic origins due to short G1 phases underlie oncogene-induced DNA replication stress
.
Nature
2018
;
555
:
112
6
.
11.
Rekhtman
N
,
Paik
PK
,
Arcila
ME
,
Tafe
LJ
,
Oxnard
GR
,
Moreira
AL
, et al
.
Clarifying the spectrum of driver oncogene mutations in biomarker-verified squamous carcinoma of lung: lack of EGFR/KRAS and presence of PIK3CA/AKT1 mutations
.
Clin Cancer Res
2012
;
18
:
1167
76
.
12.
Kim
BR
,
Van de Laar
E
,
Cabanero
M
,
Tarumi
S
,
Hasenoeder
S
,
Wang
D
, et al
.
SOX2 and PI3K cooperate to induce and stabilize a squamous-committed stem cell injury state during lung squamous cell carcinoma pathogenesis
.
Plos Biol
2016
;
14
:
e1002581
.
13.
Justilien
V
,
Walsh
MP
,
Ali
SA
,
Thompson
EA
,
Murray
NR
,
Fields
AP
.
The PRKCI and SOX2 oncogenes are coamplified and cooperate to activate hedgehog signaling in lung squamous cell carcinoma
.
Cancer Cell
2014
;
25
:
139
51
.
14.
Bradley
JD
,
Paulus
R
,
Komaki
R
,
Masters
G
,
Blumenschein
G
,
Schild
S
, et al
.
Standard-dose versus high-dose conformal radiotherapy with concurrent and consolidation carboplatin plus paclitaxel with or without cetuximab for patients with stage IIIA or IIIB non-small-cell lung cancer (RTOG 0617): a randomised, two-by-two factorial phase 3 study
.
Lancet Oncol
2015
;
16
:
187
99
.
15.
Wolfe
AR
,
Robb
R
,
Hegazi
A
,
Abushahin
L
,
Yang
L
,
Shyu
DL
, et al
.
Altered gemcitabine and nab-paclitaxel scheduling improves therapeutic efficacy compared with standard concurrent treatment in preclinical models of pancreatic cancer
.
Clin Cancer Res
2021
;
27
:
554
65
.
16.
Robb
R
,
Yang
L
,
Shen
C
,
Wolfe
AR
,
Webb
A
,
Zhang
X
, et al
.
Inhibiting BRAF oncogene-mediated radioresistance effectively radiosensitizes BRAF(V600E)-mutant thyroid cancer cells by constraining DNA double-strand break repair
.
Clin Cancer Res
2019
;
25
:
4749
60
.
17.
Liu
XF
,
Krawczyk
E
,
Suprynowicz
FA
,
Palechor-Ceron
N
,
Yuan
H
,
Dakic
A
, et al
.
Conditional reprogramming and long-term expansion of normal and tumor cells from human biospecimens
.
Nat Protoc
2017
;
12
:
439
51
.
18.
Li
G
,
Liu
Y
,
Su
Z
,
Ren
S
,
Zhu
G
,
Tian
Y
, et al
.
MicroRNA-324–3p regulates nasopharyngeal carcinoma radioresistance by directly targeting WNT2B
.
Eur J Cancer
2013
;
49
:
2596
607
.
19.
Saxton
RA
,
Sabatini
DM
.
mTOR signaling in growth, metabolism, and disease
.
Cell
2017
;
169
:
361
71
.
20.
Shaw
RJ
,
Cantley
LC
.
Ras, PI(3)K and mTOR signalling controls tumour cell growth
.
Nature
2006
;
441
:
424
30
.
21.
Zoncu
R
,
Efeyan
A
,
Sabatini
DM
.
mTOR: from growth signal integration to cancer, diabetes and ageing
.
Nat Rev Mol Cell Biol
2011
;
12
:
21
35
.
22.
Meric-Bernstam
F
,
Gonzalez-Angulo
AM
.
Targeting the mTOR signaling network for cancer therapy
.
J Clin Oncol
2009
;
27
:
2278
87
.
23.
LoRusso
PM
.
Inhibition of the PI3K/AKT/mTOR pathway in solid tumors
.
J Clin Oncol
2016
;
34
:
3803
3815
.
24.
Shen
C
,
Lancaster
CS
,
Shi
B
,
Guo
H
,
Thimmaiah
P
,
Bjornsti
MA
.
TOR signaling is a determinant of cell survival in response to DNA damage
.
Mol Cell Biol
2007
;
27
:
7007
17
.
25.
Shen
C
,
Oswald
D
,
Phelps
D
,
Cam
H
,
Pelloski
CE
,
Pang
Q
, et al
.
Regulation of FANCD2 by the mTOR pathway contributes to the resistance of cancer cells to DNA double-strand breaks
.
Cancer Res
2013
;
73
:
3393
401
.
26.
Zhou
X
,
Liu
W
,
Hu
X
,
Dorrance
A
,
Garzon
R
,
Houghton
PJ
, et al
.
Regulation of CHK1 by mTOR contributes to the evasion of DNA damage barrier of cancer cells
.
Sci Rep
2017
;
7
:
1535
.
27.
Steinbichler
TB
,
Dudas
J
,
Skvortsov
S
,
Ganswindt
U
,
Riechelmann
H
,
Skvortsova
II
.
Therapy resistance mediated by cancer stem cells
.
Semin Cancer Biol
2018
;
53
:
156
67
.
28.
Macheret
M
,
Halazonetis
TD
.
DNA replication stress as a hallmark of cancer
.
Annu Rev Pathol-Mech
2015
;
10
:
425
48
.
29.
Hills
SA
,
Diffley
JF
.
DNA replication and oncogene-induced replicative stress
.
Curr Biol
2014
;
24
:
R435
444
.
30.
Ma
YX
,
Vassetzky
Y
,
Dokudovskaya
S
.
mTORC1 pathway in DNA damage response
.
Bba-Mol Cell Res
2018
;
1865
:
1293
311
.
31.
Lamm
N
,
Rogers
S
,
Cesare
AJ
.
The mTOR pathway: implications for DNA replication
.
Prog Biophys Mol Bio
2019
;
147
:
17
25
.
32.
Horn
D
,
Hess
J
,
Freier
K
,
Hoffmann
J
,
Freudlsperger
C
.
Targeting EGFR-PI3K-AKT-mTOR signaling enhances radiosensitivity in head and neck squamous cell carcinoma
.
Expert Opin Ther Targets
2015
;
19
:
795
805
.
33.
Meng
D
,
Frank
AR
,
Jewell
JL
.
mTOR signaling in stem and progenitor cells
.
Development
2018
;
145
:
dev152595
.
34.
Dogan
F
,
Avci
CB
.
Correlation between telomerase and mTOR pathway in cancer stem cells
.
Gene
2018
;
641
:
235
9
.
35.
Batlle
E
,
Clevers
H
.
Cancer stem cells revisited
.
Nat Med
2017
;
23
:
1124
34
.
36.
Eun
K
,
Ham
SW
,
Kim
H
.
Cancer stem cell heterogeneity: origin and new perspectives on CSC targeting
.
BMB Rep
2017
;
50
:
117
25
.
37.
Masciale
V
,
Grisendi
G
,
Banchelli
F
,
D'Amico
R
,
Maiorana
A
,
Sighinolfi
P
, et al
.
Isolation and identification of cancer stem-like cells in adenocarcinoma and squamous cell carcinoma of the lung: a pilot study
.
Front Oncol
2019
;
9
:
1394
.
38.
Prabavathy
D
,
Swarnalatha
Y
,
Ramadoss
N
.
Lung cancer stem cells-origin, characteristics and therapy
.
Stem Cell Investig
2018
;
5
:
6
.
39.
Pajonk
F
,
Vlashi
E
,
McBride
WH
.
Radiation resistance of cancer stem cells: the 4 R's of radiobiology revisited
.
Stem Cells
2010
;
28
:
639
48
.
40.
Soteriou
D
,
Fuchs
Y
.
A matter of life and death: stem cell survival in tissue regeneration and tumour formation
.
Nat Rev Cancer
2018
;
18
:
187
201
.
41.
Zeng
Z
,
Wang
RY
,
Qiu
YH
,
Mak
DH
,
Coombes
K
,
Yoo
SY
, et al
.
MLN0128, a novel mTOR kinase inhibitor, disrupts survival signaling and triggers apoptosis in AML and AML stem/progenitor cells
.
Oncotarget
2016
;
7
:
55083
97
.
42.
Xia
P
,
Xu
XY
.
PI3K/Akt/mTOR signaling pathway in cancer stem cells: from basic research to clinical application
.
Am J Cancer Res
2015
;
5
:
1602
9
.

Supplementary data