S6K1 Modulates STAT3 Activation to Promote Resistance to Radiotherapy in Lung Cancer

Abstract

Radiotherapy is a mainstay in the management of locally advanced lung cancer; however, intrinsic and acquired radioresistance contribute to poor prognosis. S6K1, a serine/threonine kinase, regulates cell growth, protein synthesis, and survival, and is increased in tumors, which is linked to enhanced survival under therapeutic stress, including radiation. The mechanisms, however, are not fully understood. This study investigates the role of S6K1 in lung cancer radioresistance and the mechanisms involved. Intrinsic radioresistance in lung cancer cells was associated with increased S6K1 activation. Pharmacologic inhibition or genetic deletion of S6K1 enhanced radiosensitivity both in vitro and in vivo, highlighting the therapeutic potential of targeting S6K1. Transcriptomic analysis revealed that S6K1 deletion significantly downregulated STAT3 expression, a transcription factor that promotes radioresistance. S6K1 deletion reduced STAT3 phosphorylation and transcriptional activity, thereby sensitizing lung cancer to radiation. Additionally, radiation exposure or overexpression of a constitutively active S6K1 isoform restored STAT3 activation in S6K1 knockout cells, underscoring the regulatory role of S6K1 in STAT3 signaling. Together, these findings establish a novel S6K1–STAT3 axis that drives radioresistance in lung cancer and suggest that targeting this pathway may enhance radiotherapy efficacy.

1. Introduction

Lung cancer is the leading cause of cancer-related deaths worldwide, with non-small cell lung cancer (NSCLC) accounting for the majority of cases [1]. In 2025, it is predicted that lung cancer will account for 20% of the total cancer-related deaths in the United States [2]. Despite advances in early detection and treatment such as immunotherapy which has shown some improvement in outcome, more than half of patients experience treatment failure. Therefore, the prognosis for lung cancer patients, especially those diagnosed in advanced stages, remains poor [3,4]. Radiotherapy is a cornerstone for lung cancer treatment, particularly for patients with locally advanced tumors. However, the emergence of radioresistance poses a significant challenge, limiting the effectiveness of radiotherapy and leading to treatment failure [5]. Furthermore, the underlying molecular mechanisms driving the development of radioresistance in lung cancer remain incompletely understood, though previous studies suggest that specific cell signaling pathways may contribute to this phenomenon. Therefore, identifying and targeting the molecular drivers of radioresistance is crucial to improving therapeutic efficacy.

S6K1 is a serine/threonine kinase encoded by the gene RPS6KB1. It is a key component of the mechanistic target of rapamycin (mTOR) signaling pathway, which regulates cell growth, protein synthesis, and survival in response to various stimuli, including nutrient availability and stress signals [6,7,8]. Full activation of S6K1 requires phosphorylation of multiple serine and threonine C-terminal residues, including T389, a residue whose phosphorylation is crucial for kinase activity [8,9]. S6K1 has been implicated in poor prognosis to therapies such as tyrosine kinase inhibitors (TKI), chemotherapy and radiotherapy, due to its critical role in promoting the survival of cancer cells under stressful conditions [10,11,12]. Besides, S6K1 is associated with poor survival in cancer patients, and cancer cells harboring RPS6KB1 gene amplification exhibit increased resistance to several DNA-damaging agents [13]. In this context, our previous work demonstrated that active S6K1 promotes radioresistance by modulating the activation of the MRN complex, a key DNA repair system activated in response to radiation-induced damage in lung cancer which suggests a broader role for this kinase in therapy resistance [14]. Notably, it has been suggested that S6K1 hyperphosphorylation correlates with worse survival in NSCLC patients [15]. In support of these findings, an inhibited S6K1 protein in lung cancer cells is associated with increased sensitivity to radiation, more apoptosis, enhanced G2/M checkpoint arrest, and more DNA damage [11]. Similarly, the pharmacological inhibition of S6K1 sensitized lung cancer cells to cisplatin, a compound widely used for the treatment of solid tumors [16] and promoted a higher therapeutic effect and overcoming resistance to TKI, a mainstay therapy in NSCLC with activating EGFR mutations [12]. Despite this evidence, the mechanisms through which S6K1 promotes resistance to these standard therapies in lung cancer remain unclear.

The STAT family of transcription factors plays a critical role in mediating cellular responses to various extracellular signals, including cytokines, growth factors, and radiation. STAT proteins are activated through phosphorylation by Janus kinases (JAKs) in response to receptor activation, leading to their dimerization and translocation to the nucleus, where they regulate the expression of target genes involved in cell survival, proliferation, and immune modulation [17]. In cancer, STAT3, a member of the STAT protein family, is activated in an aberrant manner inducing the expression of downstream proteins associated with the formation, progression, and metastasis of cancers [18]. STAT3 is activated by phosphorylation at serine 727 (S727), which triggers its transcriptional activity that is linked to tumor progression, epithelial-mesenchymal transition, therapy resistance, and poor survival [19,20,21]. Importantly, recent studies have linked STAT3 to promoting radiation-induced cancer stemness and evasion of DNA damage-mediated cell death leading to radioresistance in multiple malignancies [22,23,24,25].

A potential link between S6K1 and STAT3 activation has been suggested [26]. However, the role of this S6K1-STAT3 axis in mediating radioresistance in cancer has not been investigated. In this study, we hypothesize that S6K1 promotes radioresistance in lung cancer via STAT3 activation. To our knowledge, this is the first report to demonstrate that S6K1 promotes tumor growth and radioresistance in lung cancer by activating STAT3 signaling. Specifically, S6K1 knockout decreased STAT3 activation by decreasing phosphorylation at pS727, thereby sensitizing lung cancer cells to radiation both in vitro and in vivo. Furthermore, pharmacological inhibition of S6K1 suppresses STAT3 transcriptional activity, and treatment with a highly specific STAT3 inhibitor produces a radiosensitizing effect in lung cancer. Together, these findings identify the S6K1-STAT3 axis as a novel mediator of therapeutic resistance and a promising target to enhance radiation in lung cancer.

2. Results

2.1. S6K1 Activity Correlates with Intrinsic Radioresistance in Lung Cancer Cells

To examine the relationship between endogenous radioresistance and S6K1 activation, clonogenic assays were performed across a panel of lung cancer cell lines (Figure 1A). H23 cells were the most radiosensitive, showing a surviving fraction (SF) of 0.0001 at 4 Gy. In contrast, H226 (SF:0.11), H661 (SF:0.4), and A549 (SF:0.52) cells were more resistant exhibiting increasing SF (Figure 1A,C). Applying the linear-quadratic model to derive radiobiological parameters for quantitative comparison, H23 and H226 demonstrate greater intrinsic radiosensitivity (SF2 = 0.074 and 0.375, respectively) with H661 and A549 showing more radiation resistance (SF2 = 0.631 and 0.948, respectively). These findings were corroborated by MTT proliferation assays performed at short time points post-irradiation, which demonstrated a dose-dependent decrease in cell viability. At 6 Gy, H23 cells exhibited the greatest reduction in proliferation (55% of control), followed by H226 (43%), H661 (20%), and A549 (4%), consistent with their relative radiosensitivity profiles (Figure 1B). Plating efficiency (PE) calculations also showed that H661 (PE: 0.38) and A549 (PE: 0.45) have higher PE compared to H23 and H226 (≈0.15), confirming the differences in radioresistance levels for both groups of lung cancer cells (Supplementary Figure S1B). To correlate these radioresistance profiles with S6K1 pathway activity, levels of phosphorylated S6K1 and its downstream effector p-S6 were measured. Elevated phosphorylation levels for S6, the main target of S6K1, were detected in the most resistant cell lines, H661 and A549, showing a fold change of 1.7 and 1.8, respectively (p < 0.05), compared with the sensitive H23 cells. Besides, a significant increase in p-S6K1 expression was only found in the most resistant cells, A549 (fold change 1.3, p < 0.05), compared with the sensitive H23 cells (Figure 1D,E). Consistent with these findings, S6K1 expression was significantly elevated in lung tumor tissue (min-max: 8.9–11.5, range: 2.6) compared to adjacent normal tissue (min-max: 8.9–9.8, range: 0.8) in publicly available datasets (p < 0.0001) (Figure 1F).

Figure 1.

S6K1 signaling confers radioresistance to lung cancer cells. (A) Cells were radiated at the shown doses, and the surviving fractions were calculated for each cell line as explained in the Section 4. Note the dramatic decrease at 4 Gy of the surviving fraction (SF) in the sensitive cells H23 (SF: 0.0001) and H226 (SF: 0.11) compared to the most resistant H661 (SF: 0.4) and A549 cells (SF: 0.52). (B) Cells were irradiated at the indicated doses, and cell proliferation was evaluated 4 days after radiation using Alamar blue. Again, H23 and H226 cells showed a lower proliferation rate in our cell models compared to no-irradiated controls, against the most resistant cells, H661 and A549 (C) Clonogenic assays showing the colony formation after a 4 Gy dose of radiation. H23 is clearly the most sensitive cell to radiation, followed by H226, H661, and A549. (D) Western blot experiments showing higher phosphoactivation of S6 and S6K1 in most radioresistant cells A549 and H661. (E) Quantification of Immunoblots using ImageJ software (version 1.54r). The most radioresistance cells H661 and A549 showed an increase in the expression of pS6, the main target of S6K1, with a fold change of 1.7 and 1.8, respectively, compared to the most sensitive H23, used as an internal control. * Denotes a p value < 0.05. *** Denotes a p value < 0.0001. Statistical differences were determined using Tukey’s test as explained in the methods. (F) S6K1 expression levels from control patients (non-tumor tissue; n: 104) and lung tumor patients (n: 986) were downloaded from the Xena TCGA database (University of California).

2.2. Pharmacologic and Genetic Inhibition of S6K1 Sensitizes Resistant Cells to Radiation

To evaluate the functional contribution of S6K1 to radioresistance, cells were first treated with pharmacologic inhibition using the selective S6K1 inhibitor PF-4708671 on A549 and H661 resistant cells. Treatment with 5 µM PF-4708671 significantly reduced p-S6 levels by 50% in H661 and by 95% in A549 cells compared to controls, confirming the S6K1 pathway inhibition (Figure 2A). As reported previously, p-S6K1 levels increased following PF-4708671 treatment due to the mechanism of action of this inhibitor which involves a feedback phosphorylation [27] (Figure 2A; Supplementary Figure S1A). To evaluate the radiosensitive effects of PF-4708671 on radiation outcome, a smaller dose of radiation combined with 5 µM PF-4708671 was used in clonogenic assays. Our results revealed that combining PF-4708671 with radiation markedly decreased the surviving fraction in the most radioresistance cells H661 and A549 (SF: 0.12 for both) compared to either treatment alone (Figure 2B,C). To exclude off-target effects and test the role of S6K1 knock out, A549 cells with S6K1 genetic deletion (S6K1^−/−) were subjected to radiation. S6K1 knockout cells exhibited a significantly reduced colony formation (CF) (CF-KO1: 4; CF-KO2: 8) compared to wild-type controls (CF: 24) following 4 Gy irradiation (Figure 2D), confirming the radiosensitizing effect of S6K1 inhibition.

Figure 2.

Inhibition of S6K1 increases radiation sensitivity of lung cancer cells. (A) Top: Immunoblot assay showing that the pharmacological inhibition of S6K1 with PF-4708671 (5 μM) for 48 h reduces the phosphorylation of S6, a downstream target of S6K1. Bottom: quantitation of p-S6 using ImageJ. Note a reduction of 53% (H661) and 95% (A549) in the expression of pS6 in cells treated with PF-4708671 compared to controls. Statistical differences were determined using a Student’s t-test. (B,C) Colony formation in cells pre-treated with DMSO or PF-4708671 plus radiation. Then, cells were treated with low doses of radiation (2 Gy). PF-4708671 was kept until the end of the experiment. Surviving fraction was calculated for each condition compared to non-treated controls. Data showed that PF-4708671 dramatically sensitized the resistant cells H661 (SF: 0.12) and A549 (SF:0.12) to low doses of radiation. (D) S6K1 KO cells and control wild types were seeded as before for clonogenic assays, and the surviving colonies were stained and counted. S6K1 genetic deletion decreases the average colony formation (CF) (CF-KO1: 4; CF-KO2: 8) compared to control (CF: 24) after radiation. Statistical differences were determined using Tukey’s test as explained in the methods. Right panel: S6K1 KO was confirmed by Western blot. * Denotes a p value < 0.05. ** Denotes a p value < 0.001. # Denotes the number.

S6K1 regulates STAT3 activation to promote radioresistance. Transcriptomic profiling of S6K1^−/− A549 cells revealed differential expression of genes involved in extracellular matrix remodeling, tyrosine kinase signaling, and DNA repair (Scheme 1, Table 1). Furthermore, in our analysis, STAT3, a transcription factor playing a key role in radioresistance, emerged as significantly downregulated gene in both S6K1^−/− clones (H5 and D7) (Table 2), supporting a mechanistic link between S6K1 and STAT3 signaling.

Scheme 1.

Differentially expressed genes in S6K1 KO clones compared to control (wild type). S6K1 genes with a p value < 0.005- and a 1.5-fold change were detected in our KO cell clones.

Table 1.

Analysis of cell signaling pathways enrichment ratio in KO cells compared to controls. Genes detected in samples processed as differentially expressed were used to detect. Molecular function categories were analyzed using the database Reactome.

GeneSet	Description	Enrichment Ratio	FDR
R-HSA-1474244	Extracellular matrix organization	3.289873262	1.27 × 10−7
R-HSA-2022090	Assembly of collagen fibrils and other multimeric structures	5.9808136	4.45 × 10−5
R-HSA-1474228	Degradation of the extracellular matrix	3.908888889	4.45 × 10−5
R-HSA-8875878	MET promotes cell motility	6.991508582	9.05 × 10−5
R-HSA-3000171	Non-integrin membrane-ECM interactions	5.741870684	9.05 × 10−5
R-HSA-216083	Integrin cell surface interactions	4.59869281	1.97 × 10−4
R-HSA-202733	Cell surface interactions at the vascular wall	3.614057853	2.49 × 10−4
R-HSA-8874081	MET activates PTK2 signaling	7.817777778	2.49 × 10−4
R-HSA-1474290	Collagen formation	4.343209877	2.80 × 10−4
R-HSA-9006934	Signaling by Receptor Tyrosine Kinases	2.233650794	3.37 × 10−4

Table 2.

STAT3 expression was downregulated in S6K1 KO clones compared to (wild type) controls. Note the drastic down-regulation in the transcription of this gene in S6K1 KO clones (H5 and D7).

Gene	logFC	pValue	FDR	Clone
STAT3	−0.720830242	0.002357589	0.043844248	H5
STAT3	−0.063725188	0.732722128	1.000000000	D7

To assess this mechanistic link S6K1-STAT3, we test whether STAT3 activation is induced by radiation. H661 and A549 cells were irradiated with 10 Gy. STAT3 and p-STAT3 levels increased at 24- and 48-h post-radiation (STAT3: fold change 1.8 and 1.78; p-STAT3: fold change 1.25 and 1.5, respectively) in A549 cells, and total STAT3 levels also increased in H661 cells (fold change: 1.5 and 1.7). Expression of c-Myc, a STAT3 target, was also upregulated in both cells (H661 fold change: 1.48–1.7; A549 fold change: 1.8–1.8) when compared to no-irradiated controls (Figure 3A; Supplementary Figure S2A). Additionally, the re-expression of a constitutively active S6K1 in S6K1^−/− A549 cells restored the p-STAT3 levels (fold change: H5: 1.5 and D7: 1.85; p < 0.05), indicating that S6K1 controls STAT3 activation (Figure 3B; Supplementary Figure S2B). Furthermore, one dose of radiation further significantly increased the p-STAT3 levels (fold change: 3, p < 0.0001) in the D7 clone compared to no-irradiated S6K1^−/− controls (Figure 3B). To validate this hypothesis, we pharmacologically inhibited S6K1 with PF-4708671 and found that S6K1 inhibition antagonized the increases of p-STAT3 levels after radiation (fold change control: 1; RT: 3; RT + PF: 0.7). c-Myc expression was also reduced after PF treatment (Figure 3C; Supplementary Figure S2C). Furthermore, phosphorylation of STAT3 at Ser727, a critical site for transcriptional activation, was also significantly reduced in S6K1^−/− cells after radiation (50% compared to wild type controls) (Figure 3D; Supplementary Figure S2D). Additionally, across a cohort of 962 NSCLC patients, Pearson’s correlation analysis revealed that RPS6KB1 expression was positively correlated with STAT3 expression at the transcript level. Specifically, the Pearson correlation coefficient was approximately r ≈ 0.40, indicating a moderate positive linear relationship between the two genes. Importantly, this association was statistically significant (p < 0.05), suggesting that higher RPS6KB1 expression tends to accompany higher STAT3 expression in tumor tissues. These findings are consistent with the hypothesis of potential co-activation or shared regulatory mechanisms involving these pathways in NSCLC. Thus, this evidence indicates that S6K1 controls STAT3 activity.

Figure 3.

STAT3 activation increases after radiation to promote radioresistance modulated by S6K1. (A) Compared to non-irradiated controls, STAT3 and STAT3 phosphoactivation increases after a single dose of 10 Gy in H661 and A549 cells at 24 and 48 h. (B) A549 S6K1-KO cells, transfected with a plasmid expressing a constitutively active form of S6K1 protein, showed an increase in the phospho-activation of STAT3 before and after radiation. (C) PF-4708671 (5 µM) antagonizes the phospho-activation of STAT3 and the expression of c-myc in A549 cells after radiation. Protein expressions were studied by Western Blot. (D) Expression analysis showing the downregulation of STAT3 activation (p-Ser727) in S6K1 KO cells. S6K1 deletion decreases the p-STAT3 expression after radiation. (E) A549 cells treated with the inhibitor Stattic plus radiation showed the lowest number of colonies (CF: 1.3) compared to radiotherapy (CF: 10) or static (CF: 13) alone and control (DMSO; CF: 43) (* p < 0.05, ** p < 0.005, and **** p < 0.0001). (F) Transcriptional activity of STAT3 was measured in the presence of vehicle, PF-4708671 (5 μM), radiation (10 Gy) or the combination by using the Dual-Glo^® Luciferase Assay System. PF-4708671 decreased the STAT3 transcriptional activity compared to control, before and after radiation (* p < 0.05, *** p < 0.0005). Statistical differences were determined using Tukey’s test as explained in the methods.

2.3. Targeting STAT3 Enhances Radiosensitivity in Lung Cancer Cells

To directly test the role of STAT3 in radioresistance, clonogenic assays were performed in A549 cells treated with Stattic, a selective STAT3 inhibitor. Stattic or radiotherapy alone significantly reduced colony formation (CF: 13 and 10, respectively) compared to controls. However, the combined treatment resulted in a further significant suppression (CF: 1.3) (Figure 3E). Moreover, reporter assays showed that STAT3 transcriptional activity was elevated by radiation (RLU: 0.9; p < 0.005) but abolished by PF-4708671, both at baseline and post-irradiation (RLU: 0.25 p < 0.005) (Figure 3F).

2.4. S6K1 Deletion Sensitizes Lung Tumors to Radiation In Vivo

To validate these findings in vivo, nude mice were injected with wild-type or S6K1^−/− A549-Luc cells and treated with localized irradiation. Tumors derived from S6K1^−/− cells showed greater sensitivity to radiation, with significantly lower growth and weight compared to controls (tumor size: S6K1^−/−+ radiation: 244 mm³; radiation: 970 mm³; S6K1^−/−: 740 mm³; control: 2084 mm³) (Figure 4A,B). IVIS imaging confirmed reduced luminescence in S6K1^−/− tumors post-radiation (Figure 4C,E), and excised tumors demonstrated visibly reduced size in the combination group (Figure 4D).

Figure 4.

S6K1 Knockout sensitizes lung cancer cells to radiation by decreasing tumor growth in vivo. (A) 1 × 10⁶ wild type or S6K1^−/− cells were inoculated in the back leg of nude mice. Tumors were irradiated with a dose of 7.5 Gy for 2 consecutive days. At the last point evaluated (33 days after radiation), the S6K1^−/−+ radiation group showed the lowest tumor size (average: 244 mm³) compared to radiation (average: 970 mm³), S6K1^−/− (average: 740 mm³), or control alone (average: 2084 mm³). (B) At the end of the experiment, tumors were extirpated and weighed. (C,E) Before euthanizing the mice, tumors were imaged using an IVIS system as described in methods (* p < 0.05, ** p < 0.005, *** p < 0.0005, and **** p < 0.0001). Lower signals were observed in the combined treatments compared to controls. (D) Pictures of tumors surgically removed at the end of the experiment. There was an evident reduction in the tumor size after radiation in S6K1^−/− tumors compared to wild type. (F) Tumors from nude mice were extirped and fixed in formalin for immunohistochemistry staining. 7Compared to radiation alone, the KO of S6K1 decreases the expression of p-STAT3. Statistical differences were determined using Tukey’s test as explained in the methods.

S6K1–STAT3 signaling is conserved in murine and human models. IHC analysis of tumors confirmed reduced p-STAT3 staining in S6K1^−/− tumors with or without radiation (Figure 4F). Transcriptomic analysis of human lung cancer tissue treated with PF-4708671, radiation, or both, revealed enrichment of genes involved in DNA damage response, double-strand break repair, and cell cycle regulation in the combination group (Figure 5), supporting enhanced radiosensitivity upon S6K1 inhibition.

Figure 5.

S6K1 genetic deletion modifies the transcriptome profile of lung cancer cells. Next-generation sequencing data from lung cancer patient tissues treated with RT (10 Gy) + PF (5 µM) compared to PF alone. The figure shows gene ontology (GO) terms that are significantly enriched with an adjusted p-value less than 0.05 in the differentially expressed gene sets (up to 40 terms). Significantly expressed genes were clustered by their gene ontology, and the enrichment of gene ontology terms (biological themes that are over-represented in our list of genes) was tested using the Fisher exact test (GeneSCF v1.1-p2).

3. Discussion

Radiotherapy remains a cornerstone in the treatment of non-small cell lung cancer (NSCLC) [5]. Despite advances in therapeutic strategies, recurrence and treatment resistance continue to impede durable responses, particularly in advanced-stage disease [28]. This study identifies a previously unrecognized molecular axis—centered on S6K1 and the transcription factor STAT3—that drives intrinsic radioresistance in lung cancer. Targeting this pathway enhances radiation sensitivity, providing a strong mechanistic rationale for combined therapeutic strategies in NSCLC.

S6K1, a kinase frequently upregulated and hyperactivated in solid tumors [15,29,30,31], has been implicated in resistance to diverse anticancer therapies, including radiotherapy [10,11,12,31]. Our findings show that elevated phosphorylation of S6K1 and its downstream substrate S6 correlates with radioresistant phenotypes in lung cancer cell lines, suggesting that pathway activation—rather than gene expression alone—may serve as a functional biomarker. These observations support the potential of phospho-S6K1 as a predictive marker to stratify patients for radiation-sensitizing interventions.

Mechanistically, both pharmacological inhibition with PF-4708671 and CRISPR-mediated knockout of S6K1 re-sensitized resistant lung cancer cells to ionizing radiation, as evidenced by decreased clonogenic survival in vitro and reduced tumor burden in vivo. These data align with prior studies demonstrating that S6K1 inhibition impairs DNA repair processes and enhances radiosensitivity [13,14,31,32].

Transcriptomic and biochemical analyses revealed that S6K1 promotes radioresistance, at least in part, through activation of STAT3. Specifically, S6K1 inhibition led to a marked decrease in STAT3 phosphorylation at serine 727, a modification known to enhance its transcriptional activity [19]. This effect was accompanied by downregulation of canonical STAT3 targets such as c-Myc. Re-expression of constitutively active S6K1 in knockout cells restored STAT3 phosphorylation and activity, strongly supporting a functional S6K1–STAT3 axis. While our findings support a functional association between S6K1 activity and STAT3 phosphorylation, it remains unclear whether S6K1 directly phosphorylates STAT3 or whether the STAT3 regulation is carried out indirectly by mTORC1, MAPK or other intermediary kinases; therefore, future studies incorporating targeted kinase assays and phosphoproteomic approaches will be required to resolve this mechanism.

STAT3 is a well-established mediator of resistance to both radiation and chemotherapy in several malignancies, including lung, breast, and prostate cancers [21,22,23]. In NSCLC, ionizing radiation induces STAT3 phosphorylation and nuclear localization, driving the expression of pro-survival genes such as Bcl2 and Bcl-XL. These effects are reversed by STAT3 inhibitors like Niclosamide [33].

In our study, pharmacologic inhibition of STAT3 with Stattic—an SH2 domain-selective small molecule, reduced colony formation and enhanced radiosensitivity, closely mirroring the effects of S6K1 inhibition. Importantly, STAT3 transcriptional activity was also attenuated by PF-4708671 treatment, both at baseline and after radiation, further confirming that S6K1 contributes to STAT3-dependent survival signaling under genotoxic stress.

Although the tumor microenvironment (TME) was not directly investigated in this study, both S6K1 and STAT3 have been implicated in shaping immune responses, cytokine signaling, and stromal interactions. Exploring the role of this axis in modulating the TME may uncover additional therapeutic vulnerabilities and opportunities for synergy with immunotherapy.

Notably, S6K1 and STAT3 are implicated not only in radiation resistance but also in chemoresistance across multiple tumor types [34,35,36,37,38]. To our knowledge, this is the first report to demonstrate that S6K1 regulates STAT3 activation to promote radioresistance in lung cancer. In vivo, S6K1-deficient tumors exhibited substantially reduced growth after radiation compared to wild-type controls, an effect accompanied by diminished STAT3 phosphorylation as shown by immunohistochemistry. These findings underscore the physiological relevance of this axis and provide strong preclinical support for targeting S6K1–STAT3 signaling to overcome resistance in NSCLC.

While these findings are compelling, several limitations should be acknowledged. Although S6K1-mediated regulation of STAT3 phosphorylation at S727 is demonstrated, it remains unclear whether this is a direct or indirect interaction. Additional studies using phospho-mutants, kinase assays, or proximity ligation approaches will help clarify this mechanism. The use of a broader array of lung cancer models, including patient-derived xenografts and organoids, would also strengthen translational relevance. Furthermore, while PF-4708671 and Stattic were effective in vitro and in vivo, their pharmacologic profiles limit clinical utility. Next-generation inhibitors with greater specificity and bioavailability should be prioritized for future studies. Dose-rate differences between in vitro (350 cGy/min) and in vivo (80 cGy/min) experiments reflect equipment optimizations, as detailed in Methods. While dose-rate effects can influence radiation responses, with lower rates allowing more sublethal damage repair and potentially reducing radiosensitization efficacy compared to higher rates, the rates used here are within typical preclinical ranges (0.1–5 Gy/min). The higher in vitro rate may modestly enhance observed radiosensitization by limiting repair time, but the consistent trends across models strengthen the biological interpretation without necessitating matched-rate comparisons, which were not performed due to setup constraints.

In summary, this study identifies the S6K1–STAT3 signaling axis as a key driver of radioresistance in NSCLC. These findings not only provide mechanistic insights into how S6K1 contributes to therapeutic resistance but also support the dual inhibition of this pathway as a promising radiosensitization strategy. Ultimately, defining and targeting resistance circuits such as this axis may pave the way for precision radiotherapy approaches that improve outcomes in lung cancer and potentially other solid tumors.

4. Materials and Methods

4.1. Cell Culture and DNA Transfection

The NSCLC cell lines H23 (ATCC#CRL-5800; RRID# RRID: CVCL_1547), H226 (ATCC#CRL-5826; RRID# CVCL_1544), A549 (ATCC#CRM-CCL-185; RRID# CVCL_0023), the large cell lung cancer cell line H661 (ATCC#HTB-183; RRID# CVCL_1577), and the 293 T cells (ATCC# CRL-3216; RRID# CVCL_0063) were obtained directly from American Type Culture Collection (ATCC, Manassas, VA, USA). H23, H661, H226, and 293 T cells were maintained in RPMI-1640 (CORNING# 10–040-CV, Corning Inc., Corning, NY, USA) medium supplemented with 10% fetal bovine serum (CORNING# 35–011-CV) and incubated in a humid atmosphere in a cell incubator at 37 °C and 5% CO₂. A549 cells were maintained in a similar way using the F-12K Medium (ATCC#30–2004) with 10% fetal bovine serum. All cell lines were certified by ATCC, and a low passage number (less than 10) was used to perform all the experiments. For cell transfection, the FuGENE^® 4K Transfection Reagent was used according to the manufacturer’s instructions (PROMEGA E5912, Promega, Madison, WI, USA). For S6K1 re-expression in A549 KO cells, we used the pRK7-HA-S6K1-F5A-E389-R3A plasmid [9] (Addgene # 8991; RRID: Addgene_8991, Watertown, MA, USA). For S6K1 inhibition, we used the highly S6K1-specific inhibitor PF-4708671 (MCE MedChemExpress #HY-15773, Monmouth Junction, NJ, USA).

4.2. Cell Proliferation Assay

Cells were seeded at a low confluency 24 h before radiation in 100 μL of appropriate medium for each cell in 96-well plates. The next day, cells were irradiated at the indicated doses as explained below. Four days after radiation, 10 μL of the reagent of the CellTiter-Blue^® Reagent (PROMEGA# G8081) was added to the wells and incubated for 3 h at 37 °C and 5% CO₂. Fluorescence signal indicative for cell viability was measured at 560/590 nm in an INFINITE 200 PRO microplate reader (Tecan, Raleigh, NC, USA; RRID:SCR_020543). Three independent experiments were performed. All experiments were conducted in quadruplicate.

4.3. Clonogenic Assays

A clonogenic assay was performed as before [14]. Briefly, cell suspensions between 50–1000 cells/well were seeded in 6-well plates in triplicate. Twenty-four hours after seeding, cells were exposed to the specified treatments at the indicated times and doses. After incubation for 2–3 weeks, cells were fixed in methanol for 20 min and stained with Crystal violet (SIGMA C0775; 0.5% w/v+ 25% methanol in PBS) for 30 min. The stain was washed away by rising 3 times with tap water. Surviving colonies with more than 50 cells were counted, photographed, and recorded. Surviving fraction (SF) was calculated as follows: SF = colony number in experimental group/(seeded cells/plating efficiency). Plating efficiency was calculated as follows: (colony number in experimental group/seeded cells in control group).

4.4. Irradiation Protocol

Irradiation was performed using an XRad320 biological irradiator (Precision X-Ray Inc., Madison, CT, USA, Model: 320 kV; RRID:SCR_026275). For in vitro experiments, cells were irradiated at a dose rate of 350 cGy/min with a source-to-surface distance (SSD) of 40 cm and an aluminum filter to achieve efficient exposure suitable for thin cell monolayers. For in vivo experiments, mice were irradiated at 80 cGy/min with an SSD of 50 cm and an aluminum + copper filter to harden the beam, ensuring dose uniformity and penetration in tissues while adhering to training protocols for animal safety and dosimetry accuracy [manufacturer brochure for XRad320 specs]. These configurations are standard for the system to optimize biological relevance in each model. For animal experiments, mice receive 2 doses of 7.5 Gy for 2 consecutive days for a final dose of 15 Gy. The body was protected using a custom lead jig so that irradiation could be delivered only to the primary tumor in the back leg.

4.5. Western Blot

Cells were lysed with RIPA buffer (Thermo Fisher #89901; Waltham, MA, USA) supplemented with a phosphatase–protease inhibitor cocktail (Thermo Fisher #1861281; Waltham, MA, USA) and kept on ice for 30 min. Cells were sonicated for 2 min and cleared by centrifugation. Transfer to PDVF membranes was conducted using the Trans-Blot^® Turbo TM Transfer System (BIO-RAD model 1704150; RRID:SCR_023156) following the manufacturer’s instructions. Membranes were blocked for 1 h with 5% non-fat milk diluted in TBS-tween and exposed to antibodies diluted in 5% albumin in TBS-tween overnight at 4 °C in constant shaking. After washing 3 times, secondary antibody (Agilent Technologies, Santa Clara, CA, USA, Cat# P0448, RRID:AB_2617138) was added and exposed for 1 h in 5% non-fat milk. The signal was detected in a ChemiDoc™ MP Imaging System (BIO-RAD, Hercules, CA, USA, model 12003154; RRID:SCR_019037) using chemo-luminescence reagents (Thermo Scientific #34075; Waltham, MA, USA). Antibodies included p-S6 (Cell Signaling Technology, Danvers, MA, USA, Cat# 4858, RRID:AB_916156), S6 (Cell Signaling Technology Cat# 2217, RRID:AB_331355), p-S6K1 (Thermo Fisher Scientific Cat# 701064, RRID:AB_2532366), S6K1 (Cell Signaling Technology Cat# 2708, RRID:AB_390722), actin (Sigma-Aldrich, St. Louis, MO, USA, Cat# A2066, RRID:AB_476693), p-STAT3 (Cell Signaling Technology Cat# 9134, RRID:AB_331589), STAT3 (Cell Signaling Technology Cat# 4904, RRID:AB_331269), c-myc (Cell Signaling Technology Cat# 5605, RRID:AB_1903938). All antibodies were diluted according to the manufacturer’s recommendations.

4.6. Crispr-Cas9 Experiments

S6K1 was genetically deleted in A549-luc cells through Crispr-Cas9 tools from SYNTHEGO (Redwood City, CA, USA; Synthego (RRID:SCR_026304). The gRNA sequence (AAUGAAAGCAUGGACCAUGG) used was targeted against exon 2. Two knockout clones were selected for further experiments. Cells transfected with a mock gRNA were used for controls (wild type). H5 and D7 denote both cell clones with S6K1 genetic deletion.

4.7. TCGA Analysis

RNA-seq expression analysis for S6K1 from patients was conducted using the database Xena from the University of California (UCSC Xena; RRID:SCR_018938). A comparison between normal and lung cancer patients was performed. Statistical analyses were conducted using GraphPad Prism 10 software (RRID:SCR_002798).

4.8. RNA-Seq Experiments

RNA was isolated using the NucleoSpin RNA, Mini kit for RNA purification (MACHEREY-NAGEL 740955.50). Samples were processed using the Illumina NovaSeq 6000 Sequencing System (RRID:SCR_016387). Library preparation was performed with Illumina Stranded Total RNA Prep with Ribo-Zero Plus (Illumina, San Diego, CA, USA). The tools FastQC (RRID:SCR_014583) (version 0.12.1) and Trimmomatic (RRID:SCR_011848) (version 0.36) were used to check the quality of the data and to trim the adapter sequences of the sequencing reads, respectively. We aligned the sequencing reads to the human genome (GRCh38) using STAR (RRID:SCR_004463) (version 2.7.11b) and determined gene-level counts with quantMode GeneCounts setting. The tool edgeR (RRID:SCR_012802) (version 4.2.0) was used to normalize the count data and determine differentially expressed genes in the knockout A549 cells compared with the control cells (wild type). We conducted pathway or gene ontology (GO) enrichment analysis using KEGG, REACTOME, HALLMARK, and Gene Ontology databases, which are embedded into the web-server WebGestalt GEne SeT AnaLysis Toolkit (RRID:SCR_006786). Significantly enriched (p  <  0.05; after being adjusted using the Benjamini and Hochberg method) pathways were reported.

4.9. Tumor Patient Explant Culture

Tumors obtained on an IRB-approved protocol, the same day of surgery, were excised, and pieces of similar size wereplaced in gelatin dental sponges and hydrated in complete medium (5 mL of complete culturemedium RPMI 1640 supplemented with 10% FBS and Pen/Strep), and maintained at 37 °C. Tumors were treated either with RT (10 Gy) + PF-4708671 (5 μM) or the combination for 5 days. At the end of the experiment, samples were fixed in formalin and embedded in paraffin as before. Samples were sent to Azenta for RNA extraction and RNA-seq analysis.

4.10. Transcriptional Activity Assay

To evaluate the transcriptional activity of STAT3, 40.000 293 T cells/well were seeded in 96-well plates and reversely transfected with the reporter and control plasmids (Cignal STAT3 Reporter (luc) Kit (QIAGEN # 336841, QIAGEN, Hilden, Germany) using Attractene Transfection Reagent (QIAGEN # 301005) according to the manufacturer’s instructions. The next day, cells were pre-treated with 5µM PF-4708671 for 2 h and irradiated with a 10 Gy dose for 24 h. Luciferase activity was measured using the Dual-Glo^® Luciferase Assay System (PROMEGA #E2920). Luminescence was recorded in a INFINITE 200 PRO microplate reader (Tecan, Raleigh, NC, USA; RRID:SCR_020543).

4.11. Animal Experiments

Athymic nude mice (Crl:NU(NCr)-Foxn1nu, Charles River Laboratories; Strain code 490) from both sexes and 4–6 weeks old were acclimated for 1 week before experimental manipulation. Then, mice were subcutaneously injected with 1 × 10⁶ S6K1^−/− KO A549-Luc cells (ATCC # CCL-185-LUC2; RRID:CVCL_UR31) in the right back leg using a 27 G needle. When tumors reached a size between 50 and 150 cm³, mice were randomly split into 4 groups for treatment: (1) Control, (2) radiotherapy, (3) S6K1^−/− KO A549-Luc cells, (4) S6K1^−/− KO A549-Luc cells + Radiotherapy. Radiation doses consisted of 2 rounds of 7.5 Gy each for two consecutive days, for a final dose of 15 Gy per mouse. For radiation, we exposed only the leg of the mice, protecting the rest of the body using a custom lead jig so that irradiation could be delivered to the primary tumor itself. Tumors were monitored and measured with a caliper at the indicated times until day 33 after radiation, when they reached a volume of 2000 cm³ and were humanely euthanized. Tumors were then extracted, weighed, and photographed. 8 mice per group were used. All animal experiments were performed at the animal facility room at Thomas Jefferson University, Philadelphia, PA. All experimental procedures were previously approved by the Institutional Animal Care and Use Committee (IACUC) under the Protocol ID: 01859–1.

4.12. Immunohistochemistry (IHC) Assays

Tissues were fixed in 10% buffered formalin and embedded in paraffin. Slide sections (5 μm) were baked at 60 °C for 1 h, deparaffinized in xylene, and rehydrated in serial amounts of alcohol. Antigenic retrieval and unmasking were performed by submerging slides in sodium citrate buffer (10 mmol/L, pH 6.0) and then boiling for 10 min. Slides were treated with 5% H₂O₂ in methanol and then blocked with 50% goat serum in 1 × TBS in a humidified chamber for 1 h at room temperature. Antibody anti phospho-STAT3 (Thermo Fisher Scientific Cat# MA5–33199, RRID:AB_2812015) was incubated at 1:100 dilution in 10% goat serum in TBS overnight at 4 °C in humid chamber. For detection, HRP conjugated secondary anti-rabbit antibody (Agilent Technologies Cat# P0448, RRID:AB_2617138) diluted 1:200 in 10% goat serum in TBS was added and exposed for 1 h at room temperature and washed 3 times in TBS. Signal was developed using the ImmPACT^® DAB Substrate Kit, Peroxidase (HRP) (Vector laboratories #SK-4105, Newark, CA, USA) following the manufacturer’s recommendations. Images were scanned at 20× magnification in a Aperio CS2 Scanscope digital scanner (Leica Aperio CS2 scanner; RRID:SCR_025111) (Leica Biosystems, Nussloch, Germany). Slides were viewed by using the Aperio ImageScope—Pathology Slide Viewing Software (version 12,4,6; RRID:SCR_020993).

4.13. IVIS Imaging

Before euthanizing, mice were anesthetized with a mix of xylazine and buprenorphine injections. Later, 150 mg/kg of luciferin (D-Luciferin, Firefly potassium Salt, PerkinElmer #122799) was prepared according to the manufacturer’s instructions and intraperitoneally injected in 100 µL of sterile PBS. Ten minutes after luciferin injection, animals were imaged using an IVIS^® Lumina LT Series III In Vivo Imaging System (PerkinElmer, Waltham, MA, USA; RRID:SCR_025239) equipped with Isoflurane and O2 regulation and using default optics and resolution settings. The average and total photon counts in the defined ROI were calculated automatically by the software and analyzed using the Tukey test and ANOVA.

4.14. Statistical Analysis

All statistical analyses were conducted using GraphPad prism version 10.4.1 software (GraphPad Prism RRID:SCR_002798). Data are shown as mean values ± SD. Statistically significant differences between three or more groups were determined using one-way ANOVA. Differences between the two groups were calculated using a Student’s t-test. Differences between three or more groups were calculated using a Tukey’s test. p-values ≤ 0.05 were considered to be statistically significant between groups unless otherwise indicated. Experiments were repeated a minimum of three times or as otherwise indicated.

5. Conclusions

In conclusion, this study establishes a novel S6K1–STAT3 signaling axis as a critical driver of radioresistance in non-small cell lung cancer, demonstrating that S6K1-mediated phosphorylation of STAT3 at serine 727 promotes tumor cell survival under radiation stress. Both pharmacological inhibition and genetic knockout of S6K1 effectively sensitize resistant lung cancer cells to radiotherapy in vitro and in vivo, which occurs via STAT3. Collectively, these results position S6K1 as both a predictive biomarker and a therapeutic target capable of overcoming drug and radiation resistance, potentially improving survival outcomes and personalizing treatment strategies for patients with resistant lung cancer.

Abbreviations

The following abbreviations are used in this manuscript:

ATCC	American Type Culture Collection
cGy	centigray
DMSO	dimethyl sulfoxide
EGFR	epidermal growth factor receptor
IHC	immunohistochemistry
IRB	Institutional Review Board
IVIS	In Vivo Imaging System
JAK	Janus kinase
KO	knockout
mTOR	mechanistic target of rapamycin
mTORC1	mechanistic target of rapamycin complex 1
NCI	National Cancer Institute
NIH	National Institutes of Health
NSCLC	non-small cell lung cancer
ROI	region of interest
RPS6KB1	ribosomal protein S6 kinase B1 (gene encoding S6K1)
RT	radiotherapy
S6K1	ribosomal protein S6 kinase 1
SD	standard deviation
SF	surviving fraction
STAT3	signal transducer and activator of transcription 3
TCGA	The Cancer Genome Atlas
TKI	tyrosine kinase inhibitor
TME	tumor microenvironment

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijms27041915/s1.

Author Contributions

Conceptualization, A.C.-A., J.H. and N.L.S.; methodology, A.C.-A., N.F. and N.L.S.; validation, A.C.-A.; formal analysis, A.C.-A., S.X. and N.L.S.; resources, T.G. and O.O., data curation, A.C.-A., S.X. and N.F.; writing—original draft preparation, A.C.-A. and N.L.S.; writing—review and editing, A.C.-A., T.G., S.X., O.O., J.H. and N.L.S.; supervision, A.C.-A. and N.L.S.; funding acquisition, N.L.S. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board of Thomas Jefferson University (protocol 20D.826 approved 13 February 2023). The animal study protocol was approved by the Institutional Animal Care and Use Committee at Thomas Jefferson University (protocol 23–07-704 approved on 16 January 2024).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors on request. The sequence data generated in this study will be publicly available in Genbank and in Gene Expression Omnibus (GEO) at the time of publication.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

This was funded in part by NCI/NIH U54 CA273956 (NLS), R01 CA227479 (ACA, NLS) and 5P30CA056036–24 (ACA, NF, JH, NLS for shared resources).