This article is mentioned in:

Introduction

Lung cancer is the most common cause of cancer-associated mortality worldwide (1). The majority (80–85%) of all lung cancers are non-small cell lung cancer (NSCLC) (2), which is characterized by multiple mutations in the epidermal growth factor receptor (EGFR) gene (3). Since mutations in EGFR constitutively act as an active receptor tyrosine kinase in NSCLC cells, several tyrosine kinase inhibitors, including gefitinib, have been developed and used as chemotherapeutic drugs to treat patients with NSCLC (4). Despite the initial clinical success of tyrosine kinase inhibitors (TKIs), acquired resistance to TKIs has developed in numerous patients with NSCLC (5). Much of the acquired resistance to TKIs has been associated with a secondary T790M mutation in EGFR (6,7). To overcome resistance to TKIs, several combined NSCLC treatments, such as erlotinib and cetuximab (8), afatinib and cetuximab (9), yuanhuadine and gefitinib (10), and metformin and gefitinib (11) have been proposed and studied). However, these therapeutic approaches often result in renewed drug resistance by activating alternative survival pathways (12–14).

Hypoxia is a characteristic of solid tumors, including NSCLC, that directly stimulates the malignant properties of cancer (15). In this tumor microenvironment, hypoxia-inducible factors (HIFs) are activated, and activated HIFs induce the expression of multiple genes associated with angiogenesis, metabolic regulation, cell apoptosis and tumor survival (16). The essential roles of HIFs in blood vessel formation and the recovery of the tumor blood supply make tumors difficult to treat, leading to resistance to radiotherapy, chemotherapy and immunotherapy (17). HIFs are heterodimeric transcription factors consisting of a HIF-α (HIF-α) and a constitutive β (HIF-β) subunits (18). The HIF-1α protein is strictly regulated by oxygen concentration in the tumor microenvironment (19). Due to hypoxia-specific expression and activity of HIF-1α in tumorigenesis and escape from cancer therapy, HIF-1α may be a promising therapeutic target (20).

Gefitinib showed substantial efficacy in patients with NSCLC and active EGFR mutations (21). However, almost all patients who experience a marked response to gefitinib eventually develop progressive disease (21). This type of resistance has been observed in clinical trials, and includes primary and acquired resistance (22). Primary resistance usually occurs in patients with wild-type EGFR and other gene mutations downstream of the EGFR signaling pathway, such as the KRAS mutation (23). The acquired mutation is mainly due to the mutation T790M in the tyrosine kinase functional domain of EGFR (21). The T790M mutation is located in the ATP-/drug-binding cleft and triggers resistance by blocking the binding of gefitinib and the kinase domain (21).

In the present study, gefitinib was selected as the representative EGFR-TKI, and the differences between normal NSCLC and gefitinib-resistant (GR) NSCLC cell lines were investigated, focusing on HIF-1α protein expression and hypoxia-induced angiogenesis. The results showed that HIF-1α protein expression level and hypoxia-induced angiogenesis were increased in GR NSCLC cell lines. These results suggested that HIF-1α directly or indirectly regulated gefitinib resistance in NSCLC cell lines and could be a novel therapeutic target for combination treatment with EGFR-TKIs.

Materials and methods

Materials

Gefitinib was purchased from Sigma Aldrich; Merck KGaA. The HIF-1α antibodies were purchased from BD Pharmingen (BD Biosciences) and NOVUS Biologicals LLC.. The CD31 antibody was obtained from BD Pharmingen (BD Biosciences). The antibody against α-tubulin, MTT and cobalt chloride (CoCl2) were all purchased from Sigma Aldrich; Merck KGaA. Matrigel was purchased from BD Pharmingen; BD Biosciences.

Cell culture, hypoxia treatment and transfection

The human NSCLC PC9 and HCC827 cell lines were purchased from Asan Medical Center and the American Type Culture Collection, respectively. The PC9/GR and HCC827/GR cell lines were generated as previously described (24,25). All the NSCLC cell lines were maintained in RPMI-1640 medium (Welgene, Inc.) supplemented with 10% FBS (HyClone; Cytiva) and 1% penicillin/streptomycin (Corning Life Sciences). The human dermal microvascular endothelial cell line (HMEC-1) was kindly provided by Dr Fransisco Candal Center for Disease Control and Prevention and maintained in MCDB131 medium (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% FBS (HyClone; Cytiva), 1% penicillin/streptomycin, 10 mM L-glutamate (Gibco; Thermo Fisher Scientific, Inc.), 10 ng/ml epidermal growth factor (Sigma Aldrich; Merck KGaA) and 1 µg/ml hydrocortisone in a 37°C humidified incubator with 5% CO2 or in a hypoxic chamber (1% O2, 5% CO2 and 94% N2; InvivO2; The Baker Company). The cell lines were inspected for mycoplasma contamination using a light microscope.

The vectors, pGIPZ, pGIPZ-V2LHS-132152 and pGIPZ-V2LHS-236718 were purchased from Open Biosystems Inc., (Thermo Fisher Scientific Inc.) and the latter 2 were used for short hairpin (sh)RNA vectors against human HIF-1A. pcDNA-HIF-1α for the overexpression of human HIF-1α and the pcDNA3.1 vector were a kind gift from Professor Gregg L. Semenza (Department of Medicine, Johns Hopkins School of Medicine, Baltimore, MD, USA). The PC9 cell line, in a 60-mm culture dish were transfected with 5 µg pGIPZ, pGIPZ-V2LHS-132152, pGIPZ-V2LHS-236718, pcDNA3.1 or pcDNA- HIF-1α using 4 µg/ml polyethyleneimine (Sigma Aldrich; Merck KGaA) at 37°C for 6 h. pGIPZ or pcDNA3.1 was transfected and used as the control or mock. The media were replaced with fresh complete media, 6 h after transfection. The cells were used for subsequent experimentation 24 h after transfection.

MTT assay

The PC9, PC9/GR, HCC827 and HCC827/GR cells (1×104 cells per well) were plated in a 24-well plated and incubated with 0.1 µM gefitinib for 48 or 72 h in a 37°C humidified incubator with 5% CO2. MTT (0.1 mg/ml) was added to each well and incubated at 37°C for 2 h, then dimethyl sulfoxide was added. The absorbance was measured at 560 nm using an iMark microplate absorbance reader (Bio-Rad Laboratories). All the data are presented as the mean ± SEM from three wells and from 3 independent experiments.

Colony formation assay

A total of 50 PC9 cells were plated on 60-mm dishes and were treated with increasing concentrations of gefitinib (10–200 nM) for 2 weeks at 37°C in a humidified incubator with 5% CO2. Following the incubation, RPMI-1640 medium (Welgene, Inc.) with 200 nM gefitinib and 10% FBS (HyClone; Cytiva) was removed, the cells were rinsed with PBS, fixed in acetic acid:methanol (1:7, vol/vol) for 5 min, and stained with crystal violet staining solution for 10 min, both at room temperature (RT). The number of colonies was counted under a light microscope. The data are presented as the mean ± SEM from 3 independent experiments.

Western blot analysis

The PC9, PC9/GR, HCC827 and HCC827/GR cell lines were incubated in a hypoxic chamber (37°C; 1% O2, 5% CO2 and 94% N2; InvivO2; The Baker Company) for 2, 8, 24 and 48 h. The 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 BCA method was used to determine the concentrations of the cell extracts. The cell extracts (30 µg/lane) were separated using 9% SDS-PAGE, then transferred to PVDF membranes (EMD Millipore). The membrane was blocked with 5% skimmed milk in TBS containing 0.1% Tween-20 for 1 h at room temperature, then incubated overnight at 4°C with the appropriate primary antibodies. Next, the membrane was incubated with a HRP-conjugated secondary antibody (1:10,000; cat. no. PI-2000; Vector Laboratories; Maravai Lifesciences) for 1 h at room temperature. The signal was developed using an ECL western detection reagent (Thermo Fisher Scientific, Inc.). The following primary antibodies were used: Anti-HIF-1α (1:3,000; cat. no. 610958) and anti-α-tubulin (1:10,000; cat. no. T5168).

Conditioned media preparations

For preparation of conditioned medium (CM), the medium from the PC9 and PC9/GR cells was changed with 1% FBS (HyClone; Cytiva)-containing RPMI-1640 (Welgene, Inc.) and further incubated for 16 h in a hypoxic chamber (37°C; 1% O2, 5% CO2 and 94% N2; InvivO2; The Baker Company). The CM was collected and filtered through a 0.22-µm pore membrane (EMD Millipore).

Wound healing assay

The HMEC-1 cell line (1×105) were plated on 24-well plates, cultured, then the monolayer was wounded with a micropipette tip and images of the cells were captured at 0 h time point. The attached HMEC-1 cells were incubated with growth medium mixed with 1 mM thymidine and CM collected from hypoxia-stimulated PC9 or PC9/GR cells for 16 h. Then, the cells were rinsed with PBS, fixed in absolute methanol for 5 min, stained with Giemsa (Sigma-Aldrich; Merck KGaA) for 10 min, both at RT, then images of each wound were captured at the same location. The migrated cells that moved beyond the reference line were counted and the number of migrated cells was divided by the number of unmigrated cells. The data are presented as the mean ± SEM from 3 independent experiments. The average migrated HMEC-1 cells, treated with CM from normoxic PC9 cells, was set to 100%.

Tube formation assay

A total of 200 µl Matrigel was polymerized on 24-well plates at 37°C for 30 min. The HMEC-1 cell line (1×105) was seeded on the polymerized Matrigel and incubated with CM from either the hypoxia-stimulated PC9 or PC9/GR cell lines. After 16 h, morphological changes were observed and the areas of the tube branches were measured using ImageJ software v1.52 (National Institutes of Health). The data are presented as the mean ± SEM from 3 independent experiments. The average areas of the tube branches treated with CM from normoxic PC9 cells was set to 100%.

Rat aortic ring sprouting assay

The rat experiments were approved by the Committee for Care and Use of Laboratory Animals at the Kyung Hee University (KHUASP-20-289). A 6-week-old male Sprague Dawley rat (150–180 g) was purchased from Daehan Biolink and anesthetized with 30 mg/kg Zoletil and 10 mg/kg Rompun (26,27) by intraperitoneal injection. The aorta was extracted and the peri-aortic fibroadipose tissue was carefully removed. The rings were sliced at a thickness of 1 mm and randomly divided into 4 groups (PC9+normoxia, PC9/GR+normoxia, PC9+hypoxia and PC9/GR+hypoxia; n=3). Each ring was placed on polymerized Matrigel in each well of a 24-well plate, then covered with an additional 50 µl Matrigel. CM from hypoxia-stimulated PC9 or PC9/GR cell lines were added to each well for 5 days. Average sprouting was measured with ImageJ software v1.52 (National Institutes of Health) after images of the plates were captured. The data are presented as the mean ± SEM from 3 independent experiments. The average sprouting with CM from normoxic PC9 cells was set to 100%.

Mouse xenograft tumor model

All the mouse experiments were approved by the Committee for Care and Use of Laboratory Animals at the Kyung Hee University [KHUASP(SE)-17-144]. A total of 24 6-week-old male BALB/c nude mice were purchased from Daehan Biolink. The mice were housed 2–4 per cage and maintained under a controlled temperature (23±0.5°C), humidity (50±10%) and a 12:12 h light:dark cycle, and food, drinking water and litter were changed every 2 days. The 7-week-old male mice (20–22 g) were randomly divided into 2 groups (PC9 and PC9/GR; n=12), anesthetized with 30 mg/kg Zoletil and 10 mg/kg Rompun (28,29), by intraperitoneal injection, then injected at a dorsal flank site with the PC9 or PC9/GR cell lines (5×106 cells per mouse), suspended in Matrigel, to perform subcutaneous tumorigenesis analysis. Their body weights and tumor sizes were measured every day for 3 weeks with a vernier caliper (Mitutoyo Corporation) and a digital balance, respectively. The tumor volume was calculated using the following formula: Tumor volume (mm3) = 0.5 (width × length × height). After 3 weeks, the mice were euthanized with sodium pentobarbital (100 mg/kg), by intraperitoneal injection and the death of the mice was verified 10 min later by loss of movement, breath, heartbeat, corneal reflex and muscular tension. Before tumor extraction, subcutaneous tissue attached to each tumor was examined and images were captured under a light microscope. Microvessels could be detected by the naked eye. Extracted tumors were frozen with optimal cutting temperature (OCT) compound (Sigma-Aldrich; Merck KgaA) at −20°C for 2 h and stored at −80°C until use.

Immunohistochemical and immunofluorescent staining

OCT compound-frozen tumor tissues were sliced at a thickness of 10-µm and placed on gelatin-coated glass slides. The sectioned tissues were fixed with 4% paraformaldehyde at RT for 15 min and incubated with methanol containing 3% hydrogen peroxide for 20 min. To increase tumor permeability, the tissues were incubated with 0.3% Triton X-100 in PBS at RT for 20 min. The tissues were blocked with blocking solution [0.1% BSA (Sigma-Aldrich; Merck KgaA), 0.3% Triton X-100 and 1.5% FBS (HyClone; Cytiva)] at RT for 1 h, then incubated with the appropriate antibodies overnight at 4°C. The tissues were subsequently washed with PBS and incubated with biotinylated anti-rabbit (cat. no. BA-1000) and anti-rat (cat. no. BA-4000) IgG (H+L) antibodies (1:500; Vector Laboratories; Maravai LifeSciences) or fluorescent anti-rabbit (cat. no. A11008) and anti-rat (cat. no. A21471) IgG (H+L) antibodies (1:500; Invitrogen; Thermo Fisher Scientific, Inc.) labeled secondary antibodies.

For biotinylation, the tissues were incubated with an Elite ABC kit (Vector laboratories, Maravai LifeSciences), and immunodetected by incubation with 3′3 diaminobenzidine (Sigma-Aldrich; Merck KgaA) solution at RT for 5 min. For fluorescence detection, the nuclei were stained with Hoechst 33342 (1:10,000; cat. no. 62249; Thermo Fisher Scientific, Inc.) at RT for 15 min. The following primary antibodies were used: Anti-HIF-1α (1:300; cat. no. NB100-479) and anti-CD31 (1:200; cat. no. 550274). Three random fields of view in each section were calculated using ImageJ software (National Institutes of Health) and the relative protein expression level of each protein was quantified according to integrated optical intensity from 3 independent experiments.

Statistical analysis

All the experiments were performed 3 times independently and the data are presented as the mean and SEM using SPSS software (v25; IBM Corp). Differences between 2 groups were evaluated using an unpaired Student's t-test differences between 3 groups were evaluated by one-way ANOVA followed by a Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant significance.

Results

HIF-1α protein expression levels were increased in GR NSCLC cell lines

To investigate the characteristics of GR NSCLC cell lines, PC9/GR and HCC827/GR were used in the present study. The GR cell lines were verified by analyzing the viabilities of these cells following treatment with gefitinib (Fig. 1A). Since HIF-1α is a key factor in tumor progression (30), HIF-1α protein expression levels in the PC9 and HCC827 cell lines were compared with that in the PC9/GR and HCC827/GR cell lines, respectively. As shown in Fig. 1B, the HIF-1α protein expression level in the PC9/GR cell line was higher compared with that in the PC9 cell line at all hypoxic exposure times tested. The same experimental results were observed between the HCC827 and HCC827/GR cell lines (Fig. 1C), indicating that hypoxia-induced HIF-1α protein expression levels were upregulated in GR NSCLC cell lines, and the PC9 cell line was selected for further analysis.

Figure 1. HIF-1α protein expression levels are increased in gefitinib-resistant non-small cell lung cancer cells lines (A) PC9, PC9/GR, HCC827 and HCC827/GR cell lines were treated with 0.1 µM gefitinib for 48 and 72 h. Cell viability was examined using a MTT assay and the data are presented as the mean ± SEM. (B) PC9, PC9/GR and (C) HCC827 and HCC827/GR cell lines were treated with gefitinib under hypoxic conditions for the indicated times, then the protein expression level of HIF-1α and α-tubulin was measured. **P<0.01 vs. PC9. ##P<0.01 and ###P<0.001 vs. HCC827.

GR cell lines induce angiogenesis under hypoxic conditions

As HIF-1α mainly regulates hypoxia-induced angiogenesis during tumorigenesis (30,31), the angiogenic properties of GR NSCLC cell lines were analyzed using several angiogenesis assays and endothelial cells. The PC9 and PC9/GR cells were incubated under hypoxic conditions for 16 h, and CM was collected and administered to human microvascular endothelial cells, HMEC-1. As shown in Fig. 2A and B, the CM from hypoxia-stimulated PC9 cells induced the migration of the HMEC-1 cells. Furthermore, the induced migratory activity of the HMEC-1 cells was increased by CM from hypoxia-stimulated PC9/GR cells. To examine the tube-forming activities of the endothelial cells, a tube formation assay was performed using the HMEC-1 cells, treated with the aforementioned CM. Consistent with the migration assay, CM from hypoxia-stimulated PC9/GR cells increased hypoxia-induced tube formation (Fig. 2C and D). To confirm the increase in hypoxia-induced angiogenic activities in the HMEC-1 cells treated with PC9/GR CM, an ex vivo rat aortic ring sprouting assay was performed. As shown in Fig. 2E and F, the number of microvessels sprouting from the aortic ring was increased in the cells treated with CM from hypoxia-stimulated PC9/GR cells compared with that in cells treated with CM from hypoxia-stimulated PC9 cells. Furthermore, the hypoxic induction of angiogenic activities in the HMEC-1 treated with PC9/GR CM were higher compared with that in the HMEC-1 cells treated with PC9 CM (Fig. 2B, D and F). These results suggested that hypoxia-induced angiogenesis was increased by the PC9/GR cell line compared with that in the PC9 cell line, and that GR NSCLC cell lines could stimulate angiogenesis during tumorigenesis.

Figure 2. Gefitinib-resistant cells induced angiogenesis under hypoxic conditions. PC9 and PC9/GR cells were incubated under normoxic or hypoxic conditions for 24 h and CM was collected, and added to the HMEC-1 cells for further analysis (A) Wound healing assays were performed using HEMC-1 cells treated with CM from the PC9 and PC9/GR cells and (B) analyzed statistically. (C) Tube formation assays using the HMEC-1 cells were performed to determine the tube-forming abilities following treatment with CM from the PC9 and PC9/GR cells and the results were (D) quantified using ImageJ software and analyzed statistically. (E) Rat aortic ring sprouting assays were performed on aortic rings treated with CM from PC9 and PC9/GR cells and the results were (F) quantified and analyzed statistically. The data are presented as the mean ± SEM from 3 independent experiments. *P<0.05 and **P<0.01 vs. CM from PC9 cells under normoxic conditions; #P<0.05 and ##P<0.01 vs. CM from hypoxia-stimulated PC9 cells. nor, normoxia; hy, hypoxia.

Tumor angiogenesis is increased in the GR cell lines

Tumor angiogenesis in GR tumors was subsequently investigated, as angiogenesis is an essential event in tumor progression (32). Tumor formation was induced by subcutaneously injecting PC9 and PC9/GR cells into BALB/c nude mice. As shown in Fig. 3A, the PC9 and PC9/GR tumors had a similar growth pattern and the body weight of the mice is shown in Fig. 3B. When the isolated tumors, 3 weeks after injection, were examined, there were no significant differences in tumor volume and weight between the PC9 and PC9/GR groups (Fig. 3C and D). The microvessels in the subcutaneous tissue attached to the tumors were also analyzed and the number of microvessels near the PC9/GR-derived tumors was increased compared with that near the PC9-derived tumors (Fig. 3E). To evaluate vessel formation within the tumors, it was examined whether the endothelial cells were present inside the tumors by staining for CD31, which is a specific marker of endothelial cells. As shown in Fig. 3F, the CD31 signal was increased in the PC9/GR tumors. Furthermore, it was found that the HIF-1α signal was also increased in the PC9/GR tumors (Fig. 3G). To confirm the association between HIF-1α expression and blood vessel formation, fluorescent double-staining was performed in the PC9 and PC9/GR tumors. In the PC9/GR tumor tissues, CD31 expression was increased around the HIF-1α-expressing region (Fig. 3H and I), indicating that HIF-1α-induced angiogenesis was stimulated in GR tumors.

Figure 3. PC9 and PC9/GR tumors were generated in BALB/c nude mice and tumor angiogenesis was increased in PC9/GR tumors. BALB/c nude mice were injected with PC9 cells or PC9/GR cells, then (A) mouse weight and (B) tumor volume were measured using a digital balance and vernier caliper, respectively, every day for 3 weeks. (C) The mice were sacrificed, 3 weeks after the injection of the PC9 or PC9/GR cells and the tumors were isolated and images were captured, and (D) tumor volume and weight were measured. The data are presented as the mean ± SEM of 3 independent experiments with 4 mice per group (total 12 mice per group). (E) Blood vessels under the skin adjacent to the tumor tissues were examined. Arrowheads indicate visible blood vessels. Tumor sections were immunostained with specific antibodies for (F) CD31 and (G) HIF-1α. The squares with full black lines are the 2-fold magnified images for the dotted squares. (H) Double immunofluorescence staining was performed using specific antibodies against HIF-1α and CD31. Nuclei were stained with Hoechst 33342. (I) Signals for HIF-1α and CD31 were quantified using ImageJ software. The data are presented as the mean ± SEM from 3 independent experiments. **P<0.01 vs. PC9 tumors.

Inhibiting HIF-1α attenuates the acquisition of GR in the NSCLC cell lines

To demonstrate the association between HIF-1α expression and gefitinib resistance, HIF-1α-regulated PC9 cells treated with CoCl2, and transfected with shHIF-1A vector or HIF-1α overexpression vector were treated with increasing concentrations of gefitinib (from 10 to 200 nM) for 2 weeks in the colony formation assay. The PC9 cells were treated with CoCl2 to induce the upregulation of HIF-1α expression (Fig. 4A) and a colony formation assay was performed following CoCl2 and gefitinib treatment. As shown in Fig. 4B and C, CoCl2 increased the number of PC9 colonies following gefitinib treatment. Next, shRNA vectors against HIF-1A were transfected into the PC9 cells (Fig. 4D) and were subsequently treated with CoCl2 following which a colony formation assay was performed under CoCl2 and gefitinib treatment. As shown in Fig. 4E and F, the colony forming activity was significantly inhibited by HIF-1α knockdown. On the contrary, HIF-1α overexpression vector was transfected into the PC9 cells (Fig. 4G) and a colony formation assay was performed under gefitinib treatment. As shown in Fig. 4H and I, number of colonies was significantly increased by HIF-1α overexpression, suggesting that HIF-1α induced gefitinib resistance in NSCLC cell lines.

Figure 4. Inhibiting HIF-1α attenuates the acquisition of gefitinib resistance in the PC9 cell line. (A) PC9 cells were treated with CoCl2 (20 µM) for 24 h and HIF-1α protein expression level was detected using western blot analysis. (B) Colony formation assay was performed to evaluate cell proliferation in the PC9 cells treated with CoCl2 (20 µM) and increasing concentrations of gefitinib (from 10 to 200 nM). After 2 weeks, the colonies were stained, images were captured and (C) the colony numbers were analyzed from 3 independent experiments. *P<0.05 vs. control. (D) Western blot analysis was used to detect HIF-1α protein expression level in the PC9 cells transfected with shHIF-1A vectors. (E) Colony formation assay was performed in the shHIF-1A vectors-transfected PC9 cells treated with CoCl2 (20 µM) and increasing concentrations of gefitinib (from 10 to 200 nM) and (F) the colony numbers were analyzed from 3 independent experiments. **P<0.01 and ***P<0.001 vs. Mock. pGIPZ was used as the mock. (G) HIF-1α protein expression level was determined using western blot analysis following transfection with HIF-1α overexpression vector. (H) Colony formation assay was performed in pcDNA-HIF-1α-transfected PC9 cells treated with increasing concentrations of gefitinib (from 10 to 200 nM) and (I) the colony numbers were analyzed from 3 independent experiments. The data are presented as the mean ± SEM. *P<0.05 vs. Mock. pcDNA3.1 was used as the mock. HIF-1α, pcDNA-HIF-1α; sh, short hairpin; shHIF-1A#1, pGIPZ-V2LHS-132152; shHIF-1A#2, pGIPZ-V2LHS-236718.

Discussion

Chemotherapy using EGFR-TKIs has successfully improved the survival time in patients NSCLC; however, there are limitations to the use of EGFR-TKIs due to the acquisition of resistance (33). Gefitinib is a first-line treatment that has been used for patients with NSCLC, and acquired resistance inevitably develops (34). GR NSCLC promoted cell proliferation and exhibited more aggressive clinical progression (35). HIF-1 is a primary transcription factor that is activated by hypoxia, and induces the expression of various genes associated with angiogenesis, proliferation and survival during tumor progression (36). In addition, the accumulation of HIF-1α correlated with radiotherapy resistance and drug resistance to various cytotoxic agents (37–42). In the present study, it was found that GR tumors had increased HIF-1α protein expression level and tumor angiogenesis in NSCLC. This suggested that inhibition of HIF-1α may inhibit angiogenesis in GR NSCLC. Furthermore, in the PC9 cells treated with gefitinib, the regulation of HIF-1α controlled the acquisition of resistance. These findings indicate that HIF-1α could also be a potential target for overcoming acquired gefitinib resistance in NSCLC.

Due to signaling complexities and an increasing trend in anticancer drug resistance, a molecular target that plays a central role in diverse oncogenic signaling pathways has been investigated (43). A promising anticancer drug target, is HIF-1, which promotes physiological changes associated with therapeutic resistance, including the limitation of drug accumulation within cells and the regulation of the cellular response to chemotherapeutic agents (44–46). Previous studies have demonstrated that activation of HIF-1α inhibited doxorubicin-mediated apoptosis in human osteosarcoma (47), was associated with cisplatin resistance through the regulation of the glutamate-cysteine ligase modifier subunit and multidrug resistance (MDR)-associated proteins in lung cancer (48), and mediated resistance to cetuximab by the induction of glycolysis in head and neck squamous cell carcinoma cells (49,50). Numerous studies have suggested that HIF-1α was associated with the induction of the MDR1 gene, which encodes for and results in the overexpression of phospho-glycoprotein, a predominant membrane transporter associated with chemotherapy resistance (51). HIF-1α also induced the expression of pyruvate dehydrogenase kinase (PDK)-1 and −3, and the upregulation of PDK3 was associated with drug resistance in colon cancer (52). However, the mechanism by which HIF-1α mediates drug resistance in cancer is not fully understood. It was previously reported that HIF-1α expression was decreased in NSCLC cell lines following gefitinib treatment (53,54), and activation of HIF-1α in the HCC827 cell line stimulated tumorigenic activities, including proliferation and migration, even though gefitinib was administered for ~48 h (54). YC-1 enhanced the antitumor activity of gefitinib and reversed sensitivity to gefitinib in GR HCC827 cell lines (55,56). In the present study, the characteristics of GR NSCLC cell lines were analyzed and it was found that HIF-1α expression and angiogenic activities were increased in GR NSCLC cell lines and GR tumors compared with that in the parent NSCLC cell lines. Furthermore, overexpression of HIF-1α increased the number of colonies formed by gradually increasing concentrations of gefitinib-treated PC9 cells for 2 weeks; in contrast, knockdown of HIF-1α expression by shRNA vectors decreased the colony number. From the results, we hypothesized that regulation of HIF-1α may modulate the acquisition of gefitinib resistance in NSCLC cell lines. In further studies, the related mechanism by which HIF-1α regulates the acquisition of gefitinib resistance should be examined.

Several studies support an association between EGFR activation and HIF-1α expression, which could be important in tumor progression (57–60). The EGF and EGFR signaling pathways might induce the translation of HIF-1α (57) and HIF-1α mediated angiogenesis by upregulating angiogenic factors, such as VEGF in tumor cells (61). In accordance with previous studies, the results in the present study indicated the induction of hypoxia-mediated tumor angiogenesis in GR NSCLC cell lines (Fig. 2). Previous studies have demonstrated that gefitinib reduced HIF-1α and VEGF protein expression levels in EGFR-sensitive NSCLC cell lines (57,62). HIF-1α expression is reduced by gefitinib; however, HIF-1α enhanced gefitinib resistance (54). We propose a possible mechanism to explain how downregulated HIF-1α controls gefitinib resistance. First, PI3K-AKT-FRAP signaling increases the rate of HIF-1α synthesis by other receptor and non-receptor tyrosine kinases, including HER2 and VSRC (63). Second, HIF-1α may increase c-Jun protein expression level, and a positive feed-forward loop exists between HIF-1α and c-Jun, in which constitutive activation of the JNK-c-Jun pathway can upregulate HIF-1α protein levels (64). Third, recent studies have revealed that HIF-1α was a direct transcriptional regulator of EGFR (59). EGFR enhances HIF-1α expression via a positive feed-forward loop (59). The induction of HIF-1α promotes EGFR transcription, and ultimately (59), EGFR mutations may be generated during the EGFR transcriptional process.

In conclusion, the results from the present study demonstrated that HIF-1α was increased in GR NSCLC cell lines and lead to induction of angiogenic effects in a mouse xenograft model and in angiogenesis assays in vitro. In addition, the regulation of HIF-1α controlled the proliferative ability of the PC9 cells under gefitinib treatment for two weeks. Therefore, HIF-1α could be a potential target for overcoming acquired gefitinib resistance using other EGFR-TKIs.

Acknowledgements

Not applicable.

Funding

This study was supported by the National Research Foundation of Korea grant funded by the Korea government (NRF-2021R1A2C1003297) and by a grant from the Basic Research Program through the National Research Foundation funded by the Ministry of Science and ICT (2019R1F1A1041812) of Republic of Korea.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

SHL and JWJ designed the experiments. JEC, WYB and JSC performed the experiments and analyzed the data. WYB and JWJ wrote the paper. JEC, WYB, JSC, and JWJ confirm the authenticity of the data in the present manuscript. All authors read and approved the final version of the manuscript.

Ethics approval and consent to participate

All animal experiments were approved by the Committee for Care and Use of Laboratory Animals at the Kyung Hee University [KHUASP(SE)-17-144, 11-24-2017].

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Glossary

Abbreviations

Abbreviations:

References