Introduction
In the past decades, targeted therapy has shown a remarkable prolongation of survival in patients with advanced non-small cell lung cancer (NSCLC) harboring driver genetic alterations, including epidermal growth factor receptor (EGFR) mutations and anaplastic lymphoma kinase (ALK) translocation [1-3]. Treatment with first-generation EGFR-tyrosine kinase inhibitors (TKIs) is associated with better overall survival (OS) than treatment with cytotoxic chemotherapeutics; further, newer generation EGFR-TKIs, including afatinib, osimertinib, and lazertinib have been shown superior clinical benefit over first-generation drugs [4-6]. Recent data demonstrate that the median OS of patients with EGFR-mutant NSCLC who received first-line afatinib reached up to 49 months [7]. Similarly, sequential use of frontline second- and third-generation ALK inhibitors in patients with ALK-rearranged lung cancer, revealed a marked prolongation of OS up to 54.1 months [8,9]. However, the degree and duration of response to TKIs vary widely among patients and the emergence of resistance is inevitable in most patients, which remains a clinically unmet need in current practice [2].
Resistance can be acquired in an oncogene-dependent or independent manner; the former includes acquisition of resistant mutations (i.e., EGFR T790M or C797X), and the latter includes activation of bypass tracks, including mesenchymal epithelial transition (MET) amplification or histologic transformation [10]. Based on these findings, numerous studies have been conducted to overcome TKI resistance, and novel therapeutic approaches, including fourth-generation TKIs and bispecific antibodies blocking both EGFR and MET, have been developed that have exhibiting promising results [11,12]. Therefore, understanding the biology of TKI resistance is critical for identifying novel therapeutic targets and prolonging the survival of patients with driving genetic alterations.
Kelch-like ECH-associated protein 1 (KEAP1)- nuclear factor erythroid-2-related factor 2 (NRF2) pathway is a crucial modulator of cellular homeostasis that enables cells to tolerate oxidative and metabolic stresses [13]. Under unstressed conditions, NRF2, encoded by nuclear factor, erythroid 2-like 2 (NFE2L2), a transcription factor, and KEAP1, a negative regulator of NRF2, binds to NRF2 at the DLG and ETGE degron motifs in the Neh2 domain [14]. Oxidative and electrophilic cellular stress induce the modification of critical KEAP1 Cys residues, producing conformational changes and resulting in KEAP1 release from NRF2 [15]. NRF2 translocates to the nucleus and induces the expression of target genes containing antioxidant response elements in their promoters. Cumulative evidence has demonstrated that this pathway is involved in tumorigenesis and in the resistance of cancers to cytotoxic chemotherapy and immunotherapy [13,16]. However, the therapeutic implications of modulating the KEAP1-NRF2 pathway in oncogene-addicted cancers have scarcely been investigated.
We performed this study to investigate whether KEAP1-NRF2 pathway is involved in the TKI resistance, and the modulation of NRF2 can affect TKI-induced growth inhibition in EGFR-mutant lung cancer.
Materials and Methods
1. Reagents and antibodies
NRF2 antibody and horseradish peroxidase (HRP)-conjugated secondary antibodies were obtained from Abcam (Cambridge, United Kingdom). β-Actin antibody, dimethyl sulfoxide (DMSO), and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide salt (MTT) were obtained from Millipore Sigma (Burlington, MA, USA). Matrigel was obtained from BD Pharmingen (Franklin Lakes, NJ, USA). Gefitinib, osimertinib, and brusatol were obtained from Selleck Chemical (Houston, TX, USA).
2. Cell culture
EGFR-mutant cell lines (PC9 and HCC827) were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). The cell lines with acquired gefitinib resistance (PC9/GR and HCC827/GR) were provided by Professor Jae Cheol Lee (Department of Internal Medicine, Asan Medical Center, Seoul, Korea). Gefitinib resistant cells were isolated by exposing them to increasing doses of gefitinib for more than 8 months, using a previously described procedure [17]. The cell lines were grown in Roswell Park Memorial Institute medium (RPMI-1640, Corning Life Science, Corning, NY, USA) supplemented with 10% heat-inactivated fetal bovine serum and 1% penicillin/streptomycin (Corning Life Science).
3. Western blotting
EGFR-mutant cell lines were incubated in a hypoxic chamber (37°C; 1% O2, 5% CO2 and 94% N2; InvivO2; The Baker Company, Sanford, ME, USA) for 2, 8, 24, and 48 hours. Cells were harvested and lysed in lysis buffer (50 mM Tris, 150 mM NaCl, and 1% NP‑40) supplemented with a protease inhibitor cocktail and phosphatase inhibitors (1 mM sodium orthovanadate and 10 mM sodium fluoride). The bicinchoninic acid assay was used to determine protein concentration in the cell extracts. Proteins (30 μg/lane) were separated using 9% sodium dodecyl-sulfate polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes (EMD Millipore, Burlington, MA, USA). The membrane was blocked with 5% skim milk prepared in Tris-buffered saline containing 0.1% Tween‑20 for 1 hour at room temperature and then probed overnight with the appropriate primary antibodies at 4°C. The membrane was then incubated with a HRP-conjugated secondary antibody for 1 hour at room temperature. The signal was developed using ECL Western detection reagent (Thermo Fisher Scientific Inc., Waltham, MA, USA). Primary antibodies against NRF2 and β-actin were used. The band intensities were normalized to those of β-actin using Scion Image software (Scion Corp., Frederick, MD, USA).
4. Determining cell viability
Cell viability was determined using the MTT assay. Toward this, cells were seeded at a density of 3×103 cells per well in 96-well plates and treated with the indicated concentration of osimertinib and brusatol either alone or in combination for 72 hours at 37°C in a humidified incubator with 5% CO2. Then, 10 μL MTT (0.5 mg/mL) was added to the cells in each well and incubated at 37°C for 2 hours to allow for the formation of blue formazan crystals. Residual MTT was carefully removed, and the crystals were dissolved by incubation with 150 μL DMSO for 30 minutes. The plate was agitated for 1 hour, absorbance was measured at 560 nm using a Spark microplate absorbance reader.
5. Tumor xenograft mouse model
All animal protocols were approved by the Institutional Animal Care and Use Committee of Kyung Hee University Medical Center (KHMC-IACUC-24-001). A total of 15 female 6‑week‑old BALB/c nude mice were obtained from Daehan Biolink (Eumseong, Korea). PC9/GR cells (5.0×106 cells in 50 μL of phosphate-buffered saline) were mixed with cold Matrigel (1:1 v/v) and injected subcutaneously into the mouse flanks. The mice were randomly divided into three groups, i.e., vehicle, osimertinib, and osimertinib/brusatol combination. After the tumor volumes reached 200 mm3, osimertinib (1 mg/kg, three times a week, orally) and brusatol (4 mg/kg, three times a week, intraperitoneally) were administered either alone or in combination. Tumor sizes were measured three times a week using a Vernier caliper (Mitutoyo Corporation, Kawasaki, Japan) until the tumor volume in the control mice reached 1,000 mm3, at which point the mice were euthanized. The tumor volume was calculated using the following formula: tumor volume (mm3)=0.5 (width×length×height). The mice were euthanized with sodium pentobarbital (100 mg/kg) by intraperitoneal injection, and death was verified 10 minutes later by loss of movement, breathing, heartbeat, corneal reflex, and muscular tension. The tumors were then harvested and their weights were measured using a digital balance.
6. Next-generation sequencing
Detailed methods of sample preparation, sequencing, quality assessment, and variant calling are described in the Supplementary Materials and Methods. Briefly, DNA was extracted from PC9/GR cells using a RecoverAll Multi-Sample Isolation Kit (Thermo Fisher Scientific Inc.) according to the manufacturer’s instructions. Oncomine Comprehensive Assay Plus (Thermo Fisher Scientific Inc.) and Ion Reporter Software version 5.20 (Thermo Fisher Scientific Inc.) were used for next-generation sequencing (NGS) library preparation and data analysis.
7. RNA sequencing
PC9/GR cells were incubated with 40 nM brusatol for 24 hours. Total mRNA was isolated using the TRIzol reagent (Thermo Fisher Scientific Inc.). NGS analysis of the RNA was conducted by Macrogen (Seoul, Korea). Integrity of the total RNA was assessed using RNA ScreenTape System (2200 TapeStation system, Agilent, Santa Clara, CA, USA). High-quality RNA preparations with an RNA integrity number greater than 7.0, were used for RNA library construction. The libraries were independently constructed using 1 µg of total RNA using Illumina TruSeq Stranded mRNA Sample Prep Kit (#RS-122-2101, Illumina, San Diego, CA, USA), and NovaSeq 6000 platform (Illumina) was used for RNA sequencing. Ensembl release GRCm38 (mm10) was used as a reference. Transcript-level quantification was aggregated to the gene expression level using the exact test (R package, R Foundation for Statistical Computing, Vienna, Austria). The false discovery rate was controlled by adjusting the p-value using the Benjamini-Hochberg algorithm.
8. Statistical analysis
Data are expressed as mean±standard of at least three independent experiments. Differences between the two groups were evaluated using an unpaired Student’s t-test, and differences among the three groups were evaluated using one‐way analysis of variance (ANOVA) followed by Tukey’s post hoc test. A p<0.05 was considered to be statistically significant. Statistical analyses were performed using the SPSS version 20.0 for Windows (IBM Co., Armonk, NY, USA).
Results
1. Identifying gefitinib resistance in EGFR-mutant cell lines
Figure 1A depicts representative photographs of the four EGFR-mutant cell lines. Cell viability was measured using the MTT assay to confirm whether the two GR cell lines retained resistance to EGFR-TKIs. The half-maximal inhibitory concentration (IC50) of HCC827/GR and PC9/GR was 14 and 17 μM, respectively, that was approximately 1,000-fold higher than that of their parental cells, suggesting the maintenance of gefitinib resistance in these GR cell lines (Figure 1B).
2. Comparison of NRF2 expression in EGFR-mutant cell lines
We analyzed NFR2 expression in various lung cancer cells to investigate whether TKI resistance is related to NRF2 overexpression. NRF2 expression was subtle in HCC827 and HCC827/GR cells but was enhanced in both PC9 and PC9/GR cells. Notably, NRF2 expression was more prominent in PC9/GR cells than in the parental cells (Figure 2A). These results suggest that the dysregulation of NRF2 may be involved in drug resistance in PC9/GR cells; therefore, we selected PC9 and PC9/GR cells for further studies. To determine whether NRF2 activation was caused by KEAP1 inactivation, we performed NGS on PC9/GR cells. NGS revealed several mutations, including EGFR exon 19, EGFR T790M, and tumor protein p53 (TP53) mutations. However, we did not detect mutations related to KEAP1 or NRF2 (Table 1).
3. Cell viability following inhibition of NRF2 activity
Brusatol was used as an NRF2 inhibitor based on previous studies [18,19]. Inhibitory effect of brusatol on NRF2, was determined using a dose of 20, 40, 80, and 100 nM, respectively. As depicted in Figure 2B, brusatol downregulated NRF2 at concentrations of 20 nM and above in both the cell lines.
Next, we evaluated the effect of osimertinib, a third-generation EGFR-TKI, and brusatol, either alone or in combination, on cell viability using MTT assay. Figure 2C shows the effects of a 24 hours treatment with brusatol and osimertinib at varying concentrations on cell viability. In PC9 cells, 10 nM osimertinib reduced viability by 40%, a reduction that was further increased in a dose-dependent manner with increasing concentrations (Figure 2C, left panel, black bars). In PC9/GR cells, 10 nM osimertinib reduced viability by 10%, and this reduction was further enhanced dose-dependently. However, this reduction in viability was less prominent than that observed in PC9 cells (Figure 2C, right panel, black bars). Considering the different NRF2 expression levels in the cell lines, these results suggest that the reduced sensitivity of PC9/GR cells to osimertinib could be partly attributed to NRF2 overexpression.
In terms of brusatol treatment, approximately 55% reduction in cell viability was observed at a concentration of 20 nM in PC9 cells, and a further reduction in cell viability occurred in a dose-dependent manner (Figure 2C). In PC9/GR cells, the viability was reduced by 25% at a brusatol concentration of 20 nM, and further dose-dependent reduction in viability was observed. However, this decrease in viability was less prominent than that observed in PC9 cells. For the combination treatment, the addition of brusatol induced more prominent, dose-dependent reduction in cell viability regardless of osimertinib concentrations in PC9 cell. In addition, similar results were observed in PC9/GR cell. These data suggest that NRF2 inhibition can enhance susceptibility to TKI in EGFR-mutant lung cancers even in those harboring TKI resistance. We further evaluated whether there is a synergism in the cotreatment by calculating Loewe model-based synergy score using Combenefit software (https://sourceforge.net/projects/combenefit; SourceForge, San Diego, CA, USA) [20]. The result showed that the synergy matrix scores were over 18 in high dose of osimertinib in PC9 cell, indicating the synergistic growth inhibition in this cell (Supplementary Figure S1).
4. Combined effect of NRF2 inhibitor and EGFR-TKI in vivo
Combined effects on cell viability observed in vitro were confirmed via an in vivo study using a PC9/GR mouse xenograft model. Schematic diagram of the experimental schedule was summarized in Figure 3A. As presented in Figure 3B, osimertinib inhibited tumor growth, and the combination treatment with brusatol showed significantly more prominent inhibition of tumor growth than osimertinib alone (p<0.01). Tumor weight was also significantly reduced in the combination treatment than osimertinib (p<0.01) (Figure 3C, D). These results confirm the in vitro data and suggest that NRF2 inhibition can enhance osimertinib-induced tumor growth inhibition even in TKI-resistant tumor.
5. Comparison of mRNA expression before and after NFR2 inhibition
Subsequently, we performed mRNA sequencing before and after brusatol treatment to determine the intracellular changes in gene transcription induced by NRF2 inhibition. When comparing differentially expressed genes, 4,920 (12%) and 4,627 (11.3%) genes were found to be upregulated and downregulated, respectively, after brusatol treatment (Figure 4). Enriched gene ontology (GO) analysis of genes using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) showed that up- and downregulated genes were mainly involved in cell migration, response to hypoxia, apoptosis, and cell proliferation (Figure 5). The GO scatter plot showed that upregulated genes were significantly related to lipid metabolism, response to stimulus, or regulation of nuclear factor κB signaling, whereas the downregulated genes were mainly related to developmental processes, response to stress, macromolecular metabolic processes, and regulation of biological processes (Figure 6). Enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis showed that genes involved in the tumor necrosis factor signaling pathway, C-type lectin receptor signaling pathway, and herpes simplex virus 1 infection were upregulated, whereas genes involved in the Rap1 and mitogen-activated protein kinase (MAPK) signaling pathways were downregulated (Figure 7).
Discussion
In this study, we demonstrated that NRF2 can be upregulated in EGFR-mutant lung cancer cells after the acquisition of TKI resistance, and that NRF2 overexpression may be related to reduced susceptibility to TKI treatment. Additionally, NRF inhibition enhanced TKI-mediated cell death and growth inhibition in NRF-overexpressed EGFR-mutant tumors. RNA sequencing data have shown that NRF2 inhibition results in alterations in intracellular gene expression, which is involved in a variety of biological processes.
KEAP1-NRF2 pathway represents the major system protecting cells from oxidative and electrophilic stress [21]. Mutations in the KEAP1-NRF2 pathway are common in NSCLC. Mutations in KEAP1 and NRF2 occur in approximately 20% of the lung adenocarcinomas and 25%-30% of the lung squamous cell carcinomas. KEAP1 alterations often co-occur with serine/threonine kinase 11 (STK11) and KRAS in lung adenocarcinomas, whereas NFE2L2 and TP53 mutations coexist in lung squamous cell carcinomas [22]. Additionally, previous studies indicate that loss-of-function mutations in KEAP1 and gain-of-function mutations in NRF2 confer resistance to chemotherapy and radiotherapy [23]. Notably, recent data reveals that KEAP1 mutations are associated with poor response and unfavorable clinical outcomes in patients with NSCLC treated using immunotherapy [24,25].
Alterations in the KEAP1-NRF2 pathway are less common in EGFR-mutant lung cancer than in EGFR-wild-type, with frequencies of approximately 2% and 10% in untreated and TKI-resistant tumors, respectively [22]. Although their prevalence is not negligible, the significance of alterations in this pathway in EGFR-mutant tumors is scarcely reported. Using genetically engineered mouse models, Foggetti et al. [22], have reported that KEAP1 inactivation reduced the sensitivity to osimertinib in EGFR-driven tumor. In that study, they also demonstrated that genetic alterations in the KEAP1-NFR2 pathway, including KEAP1 mutations and NFR2 amplifications, were found either prior to TKI treatment or after the emergence of resistance, and these alterations were associated with a decreased duration of TKI treatment in patients with EGFR-mutant lung cancer [22]. Moreover, alterations in the KEAP1 pathway were significantly associated with unfavorable clinical outcomes after adjusting for multiple compounding factors [22].
In this study, we used two EGFR-mutant cell lines; however, NRF2 was overexpressed in PC9/GR cells. Considering the aforementioned low frequency of mutations in this pathway in EGFR-mutant tumors [22], NRF2 overexpression is probably not a common phenomenon during TKI resistance. As KEAP1 is a key regulator of NRF2, we initially hypothesized that KEAP1 mutations might be involved in the upstream mechanism of NRF2 activation in PC9/GR cells. However, NGS did not reveal any mutations in KEAP1 or NRF2 genes (Table 1). The exact mechanism of NRF2 overexpression in the present study is not clear; however, gain of copy number, or KEAP1-independent alterations in NRF2 might be a possible explanation.
Our data showed a dose-dependent decrease in the viability induced by brusatol treatment in both cell lines. As presented in Figure 2A, NRF2 expression was more enhanced in PC9 than HCC827 cell lines, which can explain the growth inhibition of PC9 cell by NRF2 inhibition. This finding is in line with previous data [18,19] and suggests an antiproliferative effect of NRF2 inhibition in EGFR-mutant lung cancer. Additionally, the reduction in the viability of PC9/GR cells was less prominent with osimertinib treatment than that in the parental cells. These data suggest that NRF2 overexpression can be a one of the mechanisms of TKI resistance in EGFR-mutant tumors. Although, the intracellular processes were not further evaluated in this study, a previous study by Park et al. [19], demonstrated that NRF2 upregulation by KEAP1 inactivation can induce increased viability, proliferation, migration, and invasiveness in EGFR-mutant cells. In that study, cotreatment of NFR2 inhibitor and EGFR-TKI synergistically inhibited cell proliferation and tumor growth, which is consistent with our data [19]. NRF2 activation confers enhanced progression, resistance to therapy, and increased antioxidant capacity, leading to the development of “NRF2 addiction” in lung cancer cells. This activation is involved in a variety of intracellular process including (1) transcription, translation, post-translational regulation of numerous genes; (2) metabolic reprogramming; (3) cell proliferation and differentiation; (4) ferroptosis; (5) angiogenesis; and (6) epithelial-mesenchymal transition [26]. Thereby, we postulated that the constitutional activation of NRF2 could activate various target genes and potentiate cell proliferation independent of EGFR signaling. Overall, the dysregulation of the KEAP1-NRF2 pathway may be involved in TKI resistance; thus, modulating this signaling pathway might be a potential therapeutic target to overcome TKI resistance. Further studies are required regarding the exact mechanism by which NRF2 overexpression is involved in TKI resistance to develop safe and effective NRF2 inhibitors that can be translated into clinical trials.
In this study, we performed RNA sequencing to elucidate the genes and pathways affected by NRF2 inhibition. NRF2 inhibition resulted in the upregulation or downregulation of numerous genes involved in cell migration, response to hypoxia, apoptosis, and cell proliferation. Although we could not identify other specific target pathways or molecules that are critically involved in TKI resistance, we assume that our sequencing data will provide fundamental information for future investigations on the NRF2-targeted therapies.
This study has several limitations. First, although we provided preclinical data on the role of NRF2 overexpression in TKI resistance and NRF2 inhibition in overcoming resistance, we could not provide relevant clinical data, including the prevalence and prognostic impact of KEAP1-NRF2 alterations in patients with EGFR-mutations. Second, despite the data on transcriptional changes after NRF2 inhibition, we could not identify other mechanisms that might be involved in TKI resistance in our preclinical model. Further investigations are needed on the clinical significance of the disruption in KEAP1-NRF2 pathway, and optimal treatment strategy for EGFR-mutant tumors which have co-alterations in this cell-protecting pathway.
In conclusion, this study demonstrated that NRF2 overexpression is involved in reduced sensitivity to TKI treatment and that NRF2 inhibition can enhance susceptibility to TKI in EGFR-mutant tumors. Although further studies on the exact mechanism of NRF2 involvement in TKI resistance and its clinical implications are required, our data highlights that the KEAP1-NRF2 pathway may be a novel therapeutic target and provide insight into the complex interplay between this pathway and the driving genetic alterations in a variety of human malignancies.
