Efatutazone and T0901317 exert synergistically therapeutic effects in acquired gefitinib‐resistant lung adenocarcinoma cells

Abstract The development of acquired EGFR‐TKI therapeutic resistance is still a serious clinical problem in the management of lung adenocarcinoma. Peroxisome proliferator activated receptor gamma (PPAR γ) agonists may exhibit anti‐tumor activity by transactivating genes which are closely associated with cell proliferation, apoptosis, and differentiation. However, it remains not clear whether efatutazone has similar roles in lung adenocarcinoma cells of gefitinib resistant such as HCC827‐GR and PC9‐GR. It has been demonstrated by us that efatutazone prominently increased the mRNA and protein expression of PPAR γ, liver X receptor alpha (LXR α),as well as ATP binding cassette subfamily A member 1 (ABCA1). In the presence of GW9662 (a specific antagonist of PPAR γ) or GGPP (a specific antagonist of LXR α), efatutazone (40 μmol/L) restored the proliferation of both HCC827‐GR and PC9‐GR cells and obviously inhibited the increased protein and mRNA expression of PPAR‐gamma, LXR‐alpha, and ABCA1 induced by efatutazone. LXR α knockdown by siRNA (si‐LXR α) significantly promoted the HCC827‐GR and PC9‐GR cells proliferation, whereas incubation efatutazone with si‐LXR α restored the proliferation ability compared with the control group. In addition, combination of efatutazone and LXR α agonist T0901317 showed a synergistic therapeutic effect on lung adenocarcinoma cell proliferation and PPAR gamma, LXR A and ABCA1 protein expression. These results indicate that efatutazone could inhibit the cells proliferation of HCC827‐GR and PC9‐GR through PPAR γ/LXR α/ABCA1 pathway, and synergistic therapeutic effect is achieved when combined with T0901317.


Introduction
Lung cancer is the most common malignant tumor in the world, and it is the number one killer of cancer [1]. Lung cancer consists of two types: small cell lung cancer and nonsmall cell lung cancer (NSCLC). Lung adenocarcinoma (LAC), which is the most common subtype of NSCLC, accounts for 85% of the cancer [2]. Patients with activated-type epidermal growth factor receptors (EGFRs) mutations have covered a new chapter in molecularly targeted therapies and have been specifically targeted by tyrosine kinase inhibitor (TKI) [3]. Gefitinib is recommended for the treatment of patients with EGFR mutations in NSCLC patients. About 80% of advanced NSCLC

ORIGINAL RESEARCH
Efatutazone and T0901317 exert synergistically therapeutic effects in acquired gefitinib-resistant lung adenocarcinoma cells patients benefit from the clinical application of EGFR TKI, which has led to the revolution in the lung cancer therapy [4]. Nevertheless, acquired resistance to TKI is the main reason for the failure of the treatment and most patients do not respond to the first generation of TKI after 10-16 months of treatment [5]. In previous articles, PTEN deletion, MET amplification, T790M EGFR 20 exon mutations, and consequent AKT over-activation may be the potential mechanisms of acquired resistance to gefitinib [6][7][8]. Therefore, there is an urgent need to develop effective new chemotherapeutic drugs.
Nuclear hormone receptors, such as pregnant X receptors (PXRs), liver X receptors (LXRs), as well as peroxisome proliferator activated receptors (PPARs), induce transcriptional activity through binding with lipophilic hormones such as thyroid hormones and steroids [9]. PPARγ, which is a ligand-activated transcription factor and also known as NR1C3, is a subtype of the PPAR family and the agonists of PPARγ have developed into a new type of anticancer drugs [10]. PPARγ may be activated by some synthetic thiazolidinediones (TZD), such as antidiabetic drug pioglitazone, rosiglitazone, and troglitazone.
Similarly, PPARγ may also be activated by fatty acids, eicosanoid derivatives, as well as prostaglandins [11,12]. Several studies have confirmed that troglitazone, the firstgeneration PPARγ ligand of thiazolidinedione, has shown significant antiproliferative effects in prostate cancer [13] and cervical cancer [14]. In respects of activation of PPAR response-element and inhibition of cancer cell growth, efatutazone, the new third-generation PPARγ agonist of thiazolidinedione, is 500 times higher effects than that of troglitazone [15]. In advanced cancer patients, the results of efatutazone I phase clinical trial show disease control effect and acceptable tolerability [16]. Therefore, this preparation shows great potential for cancer treatment.
Another class of nuclear hormone receptors-LXRs has two subtypes, namely LXRa and LXRβ. It is reported that LXRs are potential targets for the prevention and treatment of breast cancer, prostate cancer, liver cancer, ovarian cancer, skin cancer, lung cancer, and colorectal cancer, [17,18]. Both the premalignant lesions in the gallbladder in LXRb knockout animals and the elevated proliferation markers expression in colon tissue further confirmed the role of LXRs and their ligands in the initiation and progression of cancers [19,20]. Previous studies have shown that the activation of nuclear receptors PPARγ and LXRα is related to the growth inhibition of prostate cancer [9]. However, it is not clear whether efatutazone shows similar proliferation inhibition effect in acquired gefitinib resistant lung adenocarcinoma cells. This problem has been studied in this study. Moreover, we explored whether PPARγ agonist efatutazone and the LXRα agonist T0901317exert similar synergistic effects on proliferation in lung cancer cells.

Reagents
Efatutazone was purchased from MedChemExpress (New Jersey). T0901317 was purchased from SigmaeAldrich (St. Louis). Before added to cell cultures, efatutazone and T0901317 were prepared in dimethyl sulfoxide (DMSO) in the vitro analyses.
Cell culture HCC827 cells that carry EGFR exon 19 deletion (Del E746-A750) are lung adenocarcinoma cell lines, which were purchased from the cell bank of the Shanghai Academy of life sciences, Chinese Academy of Sciences. Lung adenocarcinoma cell line PC9 was derived from an untreated Japanese patient with an EGFR 19 exon deletion (Del E746-A750) [21], which is provided by Dr. Zhou (Guangxi Medical University, China). These cells were kept in the environment of 5% CO 2 at 37°C in RPMI-1640 medium supplemented with 10% FBS.

Establishment of the gefitinib-resistant HCC827-GR and PC9-GRsubline cells from HCC827 and PC9 cells
According to previously reported method, HCC827 and PC9 cells were exposed to increasing the concentration of gefitinib in order to establish gefitinib-resistant subline cells [22]. Ultimately, HCC827 and PC-9 cells produced stable gefitinib resistance: isolated HCC827-GR and PC9-GR cell lines were confirmed to resistant to gefitinib independently. These gefitinib-resistant cell lines were passed more than 25 times with gefitinib, and the resistance was verified by Cell counting kit-8(CCK-8) (Dojindo, Japan).

Analysis of publicly available datasets
The Oncomine (https://www.oncomine.org) database [23] was used to determine the gene expression of PPARG in lung adenocarcinoma. We used the Oncomine to query PPARG gene and filter the results by selecting lung adenocarcinoma and cancer vs. normal analysis. Three publically available GEO (Gene Expression Omnibus) datasets (http://www.ncbi.nlm.nih.gov/geo/) GSE74575, GSE38302, GSE59239 and GSE83666 were used to analyze PPARG expression with respect to lung adenocarcinoma EGFR-TKI-resistance. The Kaplan-Meier plotter (http://kmplot. com/analysis/) database was used to assess the effect of 54,675 genes on survival using 2437 lung cancer samples on the HGU133 Plus 2.0 array, Which was used to analyze the correlation between PPARG expression and overall survival (OS) in lung adenocarcinoma [24]. The log-rank P-value and hazard ratio with 95% confidence intervals were also computed.
To analyze the relationship between PPARG mRNA and LXRα as well as ABCA1 mRNA levels in lung adenocarcinomas, we acquired and analyzed the data from TCGA dataset using a tool in www.cbioportal.org. Specifically, select "Query" on the home page of the websitewww. cbioportal.org, selects "Lung Adenocarcinoma (TCGA, Provisional)" from Select CancerStudy. In the "Select Genomic Profiles," select "mRNA Expression z-Score (RNA Seq V2RSEM)" and "protein/phosphoprotein level (RPPA)." In "Enter Gene set," input "PPARG: EXP < 0," then click "Submit." On the next page, click "Protein Change" tab, then change Antibody Type to "mRNA Expression", click LXRα and ABCA1, the corresponding figure will show.
To analyze the relationship between PPARG and LXRα, or ABCA1 mRNA levels in lung adenocarcinomas, we obtained the data from TCGA, Provisional using www.cbioportal.org. Specifically, on the home page of the website, select "download data," then, select "LungAdenocarcinoma (TCGA, Provisional)," click "mRNA expression Z-score (microarray)" from Select Genomic Profiles, and enter gene set "PPARG, LXRα, or ABCA1," select "Transposedata matrix." Click "Submit," the PPARG, LXRα, and ABCA1 mRNA Z-scores of 522 cases will appear. The correlation between PPARG and other genes Z-scores was then analyzed by Pearson's correlation and plotted using GraphPad Prism 7.

Measurement of cell viability
The cells were plated in 96-well plates in various gefitinib concentrations with or without different concentrations of efatutazone after 24-h incubation, after additional 48 h, each wells were determined by CCK-8 reagents. The absorbance was measured at 450 nm with an ELISA plate reader. The IC 50 (50% inhibitory concentrations) values [10,25] were determined according to the percentages.

Sequencing the EGFR gene
We used the ABI 3500 sequencer (ABI, Massachusetts) to determine the EGFR sequence of HCC827-GR and PC9-GR cell, and primer sequences are displayed in Table 1.

Quantitative real-time PCR (qRT-PCR)
The quantitative PCR was fulfilled using SYBR Green mix (Life Technologies, Massachusetts). GAPDH was used as an internal control to normalize the amount of total RNA in each sample, and primer sequences are displayed in Table 1. The relative expression was calculated and normalized using the 2 −ΔΔCt method [26].

Western blotting analysis
HCC827-GR and PC-9GR cells were lysed with radio immunoprecipitation assay (RIPA) buffer (Invitrogen, Massachusetts) having proteinase inhibitors. After separation in the SDS-PAGE gel, the protein was transferred on a polyvinylidene fluoride (PVDF) membrane. Membranes were blocked in 5% BSA in TBS-T for 1.5 h and then incubated overnight (4°C) with antibodies against β-Actin (Cell Signaling Technology, Massachusetts), PPARγ, LXRα, and ABCA1 (Invitrogen, Massachusetts). After being washed in TBS-T, membranes were hatched with goat anti-rabbit HRP-conjugated secondary antibody for 2 h at room temperature. The blots were visualized by ECL detection (Invitrogen, Massachusetts).

Colony formation assay
HCC827-GR and PC-9GR cell were seeded at a density of 400 cells per well in flat-bottomed 6-well plates. After 24 h of incubation, cells were treated with efatutazone (40 μmol/L) alone diluted with the medium to appropriate concentrations. After 14 days, cells were fixed with 4% paraformaldehyde and had 0.1% crystal violet. Visible colonies [27,28] were counted.

Statistical analyses
Comparisons between treatment groups were made using two-tailed unpaired or paired Student's t-tests. Values are presented as means ± SD and analyzed using one-way analysis of variance, followed by least significant difference (LSD) test for comparisons of group means. The log-rank test was used for survival analysis, and the Kaplan-Meier method was used to assess survival time distribution. Statistical significance was defined as a P value <0.05. The synergy of data is calculated as [(efatutazone + T0901317)−control] ÷ [(efatutazone−control) + (T0901317−control)]. According to this formula, a value less than 0.5 is antagonistic, a value greater than 1.0 is synergistic, while a value of 0.5-1.0 is additive [10].

PPARG expressed in gefitinib-resistant lung cancer cells and lung adenocarcinoma
To investigate the expression of PPARG gene in lung adenocarcinoma, we employed the Oncominedatabase (http://www.oncomine.org) to compare the different levels of PPARG mRNA between cancer and normal tissues (Fig. 1A). These eight representative datasets revealed that PPARG gene expression levels were decreased in lung adenocarcinoma (Fig. 1A). To identify the survival time of patients with lung adenocarcinoma, we applied the Kaplan-Meier plotter and log rank analysis. The results showed that PPARG was positively associated with survival time in lung adenocarcinoma (HR = 0.59, P = 0.00078) (Fig. 1B). Then, we investigate levels of PPARG, LXRα, and ABCA1 in gefitinib-resistant lung cancer cells in datasets downloaded from GEO database (http://www.ncbi. nlm.nih.gov/geo/). The results showed that PPARG, LXRα, and ABCA1 down-regulated in gefitinib-resistant LAC cells: HCC827-GR and PC9-GR cells (Fig. 1C).
We therefore hypothesized that PPARG, LXRα, and ABCA1 were related to tumor-suppressive effects against acquired gefitinib-resistant lung adenocarcinoma cells, and efatutazone, the novel third-generation PPARG agonist, could be a potentially useful choice for patients with lung adenocarcinoma.
Establishment of the gefitinib-resistant HCC827-GR and PC9-GR subline cells from HCC827 and PC-9 cells To establish gefitinib-resistant sublines from HCC827 and PC9 cells harboring EGFR activating mutations, we cultured the cells in increasing concentrations of gefitinib within 6 months, as described in Results section. We used CCK-8 assay to confirm that the new cell line, HCC827-GR and PC9-GR, did not show growth suppression in response to exposure to gefitinib ( Fig. 2A  and B). This cell line in the absence of gefitinib medium serially passaged over 15 generations, no changes in sensitivity to gefitinib [6]. The IC50 values of gefitinib for HCC827, HCC827-GR, PC-9, and PC-9-GR cells were 0.048 ± 0.004 μmol/L, 17.70 ± 1.3 μmol/L, 0.037 ± 0.003 μmol/L, 16.78 ± 1.1 μmol/L, respectively (Fig. 2C). Compared with their parental cell lines, the HCC827-GR and PC9-GR cell lines were, respectively, 368.8 and 453.5 folds more resistant to gefitinib, implying the sensitivity of the parental cells and the drug resistance of the progeny cells.
Previous studies reported the association between the mutation of EGFR T790M and acquired resistance to gefitinib in patients [21]. We performed DNA sequencing to examine genetic alterations of EGFR at exons 18-22, including the well-known T790M mutation, and mutation of EGFR T790M at exon 20 was not observed in the HCC827-GR and PC9-GR cells. The general mechanism of acquired resistance to HCC827-GR and PC9-GR cells was ruled out (Fig. 2D).
We compared levels of PPARG, LXRα, and ABCA1 mRNA in parental-sensitive cell lines, and offspring resistant cell lines by qRT-PCR. The expressions of PPARG, LXRα, and ABCA1 decreased in offspring resistant cell lines compared with their parental-sensitive cells (Fig. 2E). The PPARG, LXRα, and ABCA1 protein expression levels were consistent with mRNA expression levels (Fig. 2F). These findings revealed that downregulation of PPARγ might be involved in acquired resistance to EGFR-TKI in lung adenocarcinoma.

Efatutazone inhibited the proliferation of HCC827-GR and PC9-GR cells
We explored whether efatutazone could inhibit proliferation of gefitinib-resistant lung cancer cell lines: HCC827-GR and PC9-GR cells. As expected, the proliferation of HCC827-GR cells was markedly decreased after incubation with efatutazone (0~40 μmol/L) for 12, 24, and 48, and the inhibitory effect of efatutazone increased with increasing concentration (Fig. 3A). Similar results were observed in PC9-GR cells (Fig. 3B).
Moreover, the colony forming assays demonstrated that efatutazone could inhibit the proliferation of gefitinibresistance cells (Fig. 3C and D). These results suggested that efatutazone inhibited the proliferation of HCC827-GR and PC9-GR cells in a dose-dependent manner.
Efatutazone regulated expressions of PPARγ, LXRα, and ABCA1 in HCC827-GR and PC9-GR cells Our data indicated that PPARγ affected cell growth of lung cancer cells expressing LXRα and ABCA1, we attempted to examine whether PPARγ regulated the expression of LXRα and ABCA1. Analysis for TCGA data by cBioPortal (http://www.cbioportal.org/public-portal/) elucidated a positive correlation between the expression of PPARγ mRNA and LXRα as well as ABCA1 mRNA levels in NSCLC (Fig. 4A and B). To determine whether PPARγ regulates LXRα and ABCA1 expression, we first performed cBioPortal to analyze the publically available TCGA data. A significantly positive correlation between PPARγ and LXRα as well as ABCA1 mRNA levels was revealed by the Spearman correlation analyses (Fig. 4C and D). Next, we confirmed this correlation by analyzing qRT-PCR and Western blot. Based on the results of the proliferation experiment, we chose the concentration of efatutazone with the highest inhibitory (40 μmol/L) to examine effect of efatutazone on expressions of PPARγ, LXRα, and ABCA1 in HCC827-GR cells for 12, 24, and 48 h. Results demonstrated that efatutazone significantly increased expressions of expression of PPARγ, LXRα, and ABCA1 at protein and mRNA levels during 48-h incubation, with the strongest effect at 48 h ( Fig. 4E and G). Similar results were observed in PC9-GR cells ( Fig. 4F and G).
Based on the results of time course, we selected the most effective point in time (48-h incubation) to explore the effect of efatutazone (0-40 μmol/L) on expressions of PPARγ, LXRα, and ABCA1 in HCC827-GR and PC9-GR cells. We observed that efatutazone significantly increased expressions of PPARγ, LXRα, and ABCA1 at the protein and mRNA levels ( Fig. 4H and J). Using LXR agonist T0901317 as a positive control, there was no significant change in the expression of PPARγ in HCC827-GR cells, but the expression of LXR and ABCA1 was significantly increased (Fig. 4H and J). Similar results were obtained in PC9-GR cells (Fig. 4I and J).

Inhibition of PPARG and LXRα diminished the synergistic effects
To confirm synergistic inhibition, we incubated HCC827-GR cells with efatutazone (40 μmol/L) in the presence or absence of GW9662 10 μmol/L (a selective antagonist of PPARγ) or GGPP 10 μmol/L (a selective antagonist of LXRα) for 48 h. As shown in Figure 5A, incubating HCC827-GR cells with efatutazone and GW9662 or GGPP, the anti-proliferative effect of efatutazone was effectively reversed and expression of PPARγ and LXRα induced by efatutazone was restored ( Fig. 5C and E). Similar results were obtained in PC9-GR cells (Fig. 5B, D, and E).

Efatutazone inhibited proliferation of HCC827-GR and PC9-GR cells via PPARγ-LXRα-ABCA1 pathway
We further proved the role of LXRα in the antiproliferative effect of efatutazone by the RNA silencing technique. Inhibition expression of LXRα in HCC827-GR and PC9-GR cells decreased levels of LXRα and ABCA1 without change of PPARγ (Fig. 6A and B). To explore the involvement of LXRα in the antiproliferative effects of efatutazone in HCC827-GR and PC9-GR cells, we transfected with HCC827-GR and PC9-GR cells with si-LXRα for 48 h and then treated the two sublines with efatutazone for 48 h. The results suggested that knockdown of LXRα significantly increased cell proliferation. In addition, treatment with si-LXRα+efatutazone restored the cell proliferation to control levels ( Fig. 6C and D), indicating that LXRα might be involved in the antiproliferative effects of efatutazone in HCC827-GR and PC9-GR cells.

Discussion
EGFR-TKIs, such as gefitinib and erlotinib, exert an excellent effect in NSCLCs patients with EGFR mutations. Unfortunately, almost all patients of NSCLCs succumb to relapse due to drug resistance [5]. Consequently, searching for novel and effective chemotherapeutic approaches is imminent. It is exciting that using PPARγ agonist efatutazone can not only re-sensitize the treatment of EGFR-TKIs, but also may overcome the problem of drug resistance with the increase in dose [10].
It has been shown that LXRα and PPARγ heterodimerize with retinoid X receptor (RXR) and activating the two heterodimers (LXRα/RXR and PPARγ/RXR) could inhibit the prostate cancer cells proliferation [9].However, it is still not clear whether efatutazone has a similar effect on the gefitinib-resistant lung adenocarcinoma cells (HCC827-GR and PC9-GR). In this article, we confirmed that the antiproliferative activity of efatutazone in HCC827-GR and PC9-GR cells may be due to up-regulating the PPARγ/LXRα/ABCA1 pathway. Meanwhile, synergetic effects of efatutazone and T0901317 on lung cancer cells proliferation inhibition and PPARγ/LXRα/ABCA1 pathway activation were also proved.
It was indicated by a great deal of evidence that PPARγ-LXRα pathway could mediate the expression of ABCA1 [30]. In the process of initiation and development of lung adenocarcinoma, the expressions of PPAR, LXR, and ABCA1 were decreasing. On the other hand, the lung adenocarcinoma cells proliferation could be inhibited by the over expression of the three genes above [10,22]. Our findings that efatutazone elevated the expression of PPARγ, LXRα, and ABCA1 further proved that efatutazone participates in PPARγ/LXRα/ABCA1 pathway in HCC827-GR and PC9-GR cells.
Afterward, antagonists and siRNA against LXRα were used to affirm that the PPARγ/LXRα/ABCA1 pathway Figure 6. PPARγ, LXRα, and ABCA1 mRNA expression levels (A and B) and cell proliferation (C and D) after knockdown of LXRα. HCC827-GR and PC9-GR cells were transfected with si-RNA against LXRα for 48 h, and mRNA expression was determined using qRT-PCR. n = 3, **P < 0.01; ***P < 0.001; ns, not significant. participated in inhibition effect by efatutazone on HCC827-GR and PC9-GR cells proliferation. It was proved by us that in HCC827-GR and PC9-GR cells, the antiproliferative effect of efatutazone was reversed by PPARγ antagonist (GW9662) and LXRα antagonist (GGPP) and the expression of PPARγ, LXRα, and ABCA1induced by efatutazone attenuated. Moreover, in HCC827-GR and PC9-GR cells, we also showed that with PPARγ expression stable, si-LXRα depressed LXRα and ABCA1 expression levels, which approved that LXRα is a downstream target gene of the PPARγ, consistent with previous reports [31]. There was another report that the lung adenocarcinoma cells proliferation could be restrained by LXRα activation [22]. Therefore, cancer cells proliferation may be increased followed by LXRα and ABCA1 inactivated. In addition, we demonstrated that the proliferation ability of HCC827-GR and PC9-GR cells was increased with si-LXRα treatment, whereas subsequent incubation of HCC827-GR and PC9-GR cells with efatutazone reversed si-LXRα-induced cell proliferation compared with the control groups. The results here obtained using antagonists (GW9662 and GGPP) and by knocking down LXRα confirmed that increased expression of LXRα was associated with higher sensitivity of lung adenocarcinoma cell proliferation inhibition [29].
Although clinical studies had affirmed the acceptable tolerability of efatutazone, about 51.6% patients still suffered from peripheral edema, which is a widely recognized adverse reaction under efatutazone therapy. In addition, the tumor suppressor effect of efatutazone was also confirmed in advanced cancer patients [16]. Besides, in a multicenter phase 1 trial, the combination of Efatutazone and paclitaxel is safe, tolerable, and biologically active [32].
T0901317, a synthetic LXR agonist and a positive control in our research, has been proved to restrained the gefitinibresistant lung adenocarcinoma cells growth, and this growth inhibition effect was related to the activated expression of LXR target gene [33]. Synergetic proliferation inhibition effect on HCC827-GR and PC9-GR cells and increased expression of PPARγ, LXRα, and ABCA1 proteins were observed in combination with of the efatutazone (40 μmol/L) and T0901317 (5 μmol/L). This synergistic action further verified that the PPARγ/LXRα/ABCA1 pathway played an important role in the anti-tumor proliferative effect of efatutazone.
In conclusion, our research demonstrated that in lung cancer EGFR-TKI-resistant HCC827-GR and PC9-GR cells, combination treatment with efatutazone and T0901317 produced a synergistic effect. The PPARγ/LXRα/ABCA1 signaling pathway was validated to be involved in the synergistic effect. Our results suggested that the combination of efatutazone with T0901317 could reverse the acquired TKI resistance in HCC827-GR and PC9-GR cells, providing a potential therapeutic approach for lung cancer.