A novel circular RNA hsa_circRNA_103809/miR-377-3p/GOT1 pathway regulates cisplatin-resistance in non-small cell lung cancer (NSCLC)

Background Cisplatin is the first-line chemotherapeutic drug for non-small cell lung cancer (NSCLC), and emerging evidences suggests that targeting circular RNAs (circRNAs) is an effective strategy to increase cisplatin-sensitivity in NSCLC, but the detailed mechanisms are still not fully delineated. Methods Cell proliferation, viability and apoptosis were examined by using the cell counting kit-8 (CCK-8) assay, trypan blue staining assay and Annexin V-FITC/PI double staining assay, respectively. The expression levels of cancer associated genes were measured by using the Real-Time qPCR and Western Blot analysis at transcriptional and translated levels. Dual-luciferase reporter gene system assay was conducted to validated the targeting sites among hsa_circRNA_103809, miR-377-3p and 3′ untranslated region (3’UTR) of GOT1 mRNA. The expression status, including expression levels and localization, were determined by immunohistochemistry (IHC) assay in mice tumor tissues. Results Here we identified a novel hsa_circRNA_103809/miR-377-3p/GOT1 signaling cascade which contributes to cisplatin-resistance in NSCLC in vitro and in vivo. Mechanistically, parental cisplatin-sensitive NSCLC (CS-NSCLC) cells were subjected to continuous low-dose cisplatin treatment to generate cisplatin-resistant NSCLC (CR-NSCLC) cells, and we found that hsa_circRNA_103809 and GOT1 were upregulated, while miR-377-3p was downregulated in CR-NSCLC cells but not in CS-NSCLC cells. In addition, hsa_circRNA_103809 sponged miR-337-3p to upregulate GOT1 in CS-NSCLC cells, and knock-down of hsa_circRNA_103809 enhanced the inhibiting effects of cisplatin on cell proliferation and viability, and induced cell apoptosis in CR-NSCLC cells, which were reversed by downregulating miR-377-3p and overexpressing GOT1. Consistently, overexpression of hsa_circRNA_103809 increased cisplatin-resistance in CS-NSCLC cells by regulating the miR-377-3p/GOT1 axis. Finally, silencing of hsa_circRNA_103809 aggravated the inhibiting effects of cisplatin treatment on NSCLC cell growth in vivo. Conclusions Analysis of data suggested that targeting the hsa_circRNA_103809/miR-377-3p/GOT1 pathway increased susceptibility of CR-NSCLC cells to cisplatin, and this study provided novel targets to improve the therapeutic efficacy of cisplatin for NSCLC treatment in clinic. Supplementary Information The online version contains supplementary material available at 10.1186/s12885-020-07680-w.

Background Non-small cell lung cancer (NSCLC) is a common malignancy with high morbidity and mortality, which has increased in incidence in the last decades [1,2]. According to the report in 2015, the crude and age-adjusted incidences of NSCLC in China are 54.20 per 100,000 people [3], and the 5-year rate for patients with metastatic NSCLC is less than 5% [1,2]. Currently, The efficacy of the current therapeutic strategies, which include surgical resection [4,5], chemotherapy [6], radiotherapy [7,8], and immunotherapy [9,10], are limited by advance-stage disease, chemoresistance and radio-resistance [11,12]. The chemotherapeutic drug cisplatin is currently used as first-line treatment for NSCLC [13,14]. Recent evidence show that continuous long-term stimulation of NSCLC cells by cisplatin caused alteration of multiple cancer associated Circular RNAs (circRNAs), resulting in a decrease in the effectiveness of the drug [13,14]. Uncovering the underlying mechanisms leading to this resistance might solve this problem. Based on the above information, by searching the online Pubmed database (https://pubmed.ncbi.nlm.nih.gov/), we selected hsa_circRNA_103809 for further investigations in this study, and the main reason is that hsa_circRNA_103809 acted as an oncogene to promote cancer development in colorectal cancer [15], breast cancer [16], hepatocellular carcinoma [17], gastric cancer [18] and lung cancer [19]. However, to date, the role of hsa_circRNA_103809 in drug resistance in cancer remains largely unknown and therefore important to investigate.
Glutamate oxaloacetate transaminase 1 (GOT1) mainly regulates cellular glutaminolysis, which converts glutamate (Glu) into a-ketoglutaric acid (a-KG) and is crucial for sustaining cancer progression [31,32]. Inhibition of GOT1 had been validated to be an effective strategy to impair cancer growth in pancreatic cancer [32] and lung cancer [31]. Notably, previous data suggests that cisplatin regulated mitochondrial GOT1 to induce nephrotoxicity in rats [33], which rendered the possibility that targeting GOT1 might help to increase the therapeutic efficacy of cisplatin in NSCLC. In addition, recent studies have suggested that miRNAs could bind to the 3′ untranslated regions (3'UTRs) of GOT1 mRNA, resulting in GOT1 degradation and downregulation [34,35], and Zhang K et al. found that miR-9 targeted GOT1 to regulate cell ferroptosis in melanoma [34]. By conducting the online miRDB software (http://mirdb.org/), we predicted that miR-377-3p potentially bound to the 3'UTR of GOT1 mRNA. Given the fact that hsa_circRNA_103809 sponged miR-377-3p in NSCLC cells, we speculated that hsa_circRNA_103809 might regulate GOT1 through miR-377-3p in a competing endogenous RNA (ceRNA)-dependent manner.
Based on the published literatures, by conducting in vitro and in vivo experiments, this study identified that the hsa_ circRNA_103809/miR-377-3p/GOT1 pathway regulated cisplatin-resistance in NSCLC cells, and targeting this pathway improved cisplatin-sensitivity in NSCLC, which provided potential avenues for improving NSCLC treatment in the clinic.

Cell culture and induction of cisplatin-resistant NSCLC (CR-NSCLC) cells
The parental CS-NSCLC cell lines, including A549 (ATCC® CCL-185™), H1299 (ATCC® CRL-5803™) and Calu-3 (ATCC® HTB-55™), were purchased from American Type Culture Collection (ATCC, USA) in Jan. 2019, and cultured in the incubator with standard culture conditions (37°C and 5% CO 2 humidified atmosphere). The cells were authenticated by STR profiling and were identified as mycoplasma-free by a commercial third-party company (Abace Biotechnology, Beijing, China). The Roswell Park Memorial Institute 1640 medium (RPMI-1640, HyClone, USA) containing 10% fetal bovine serum (FBS, Gibco, USA) was used for cell cultivation. According to the experimental procedures provided by the previous work [36,37] and our preliminary experiments (data not shown), the CS-NSCLC cells were exposed to continuous low-dose cisplatin treatment, ranged from 0.5 μg/ml to 5 μg/ml, for 80 days in a step-wise manner to generate descendent CR-NSCLC cells (A549/DDP, H1299/DDP and Calu-3/DDP). After that, the CR-NSCL C cells were stimulated with high-dose cisplatin (25 μg/ ml) for 0 h, 24 h, 48 h and 72 h, to validate the successful induction of CR-NSCLC cells.

Vectors transfection
The overexpression and downregulation vectors for hsa_ circRNA_103809 and GOT1, and miR-377-3p mimic and inhibitor were designed and synthesized by Sangon Biotech (Shanghai, China), and the above vectors were delivered into CS-NSCLC and CR-NSCLC cells to manipulate genes expressions by using the commercial Lipofectamine 2000 reagent (Invitrogen, USA), based on the experimental protocols provided by the producer. After that, Real-Time qPCR was conducted to validate the transfection efficiency of the above vectors. The sequence of siRNA for hsa_cir-

Cell counting kit-8 (CCK-8) assay
The NSCLC cells were pre-transfected with the above vectors, cultured in 96-well plates at standard culture conditions, and were subjected to cisplatin (25 μg/ml) stimulation for 0 h, 24 h, 48 h and 72 h, respectively. After that, the CCK-8 reaction solution (AbMole, USA) was incubated with the cells in the volume of 20 μl per well for 4 h at the incubator. Then, the plates were vortexed to thoroughly mix the cells with the solution, and were placed in a microplate reader (ThermoFisher Scientific, USA) to measure the optical density (OD) values at the wavelength of 450 nm, which could be used to represent relative cell proliferation in the cells.

Trypan blue staining assay
The CR-NSCLC and CS-NSCLC cells were pretransfected with different vectors, and stimulated by using the cisplatin. Then, the cells were prepared and stained with trypan blue staining solution obtained from Invitrogen (USA) for 20 min at room temperature. After that, a light microscope was used to observe and count the number for dead blue cells, which were used to evaluate cell viability according to the following formula: Cell viability (%) = (Total cells -Dead blue cells)/Total cells * 100%.

Annexin V-FITC/PI double staining assay
A apoptosis detection kit (Invitrogen, USA) was used to examine cell apoptosis in CS-NSCLC cells and CR-NSCLC cells, based on the protocols provided by the manufacturer. In brief, the cells were harvested and prepared, and subsequently stained with Annexin V-FITC and propidium iodide (PI) for 25 min at room temperature without light exposure. After that, a flow cytometer (FCM, ThermoFisher Scientific, USA) was used to examine the cell death ratio in NSCLC cells. Specifically, the early apoptotic cells were stained with Annexin V-FITC alone, the late apoptotic cells were stained with Annexin V-FITC and PI, and the necroptotic cells were stained with PI alone.

Real-time qPCR
The NSCLC cells were subjected to differential treatments, and the TRIzol reagent (Invitrogen, USA) was employed to extract the total RNA. Specifically, 5 × 10 6 cells were treated with 1 ml TRIzol solution for 5 min, and were subsequently treated with chloroform for 15 min at room temperature. Next, the upper water phase was collected, and was treated with 0.5 ml isopropyl alcohol for 10 min. After centrifugation with 12,000 g for 10 min, 75% ethyl alcohol was used to isolate and purify the total RNA. Next, the Real-Time qPCR was conducted to determine the expression levels of hsa_circRNA_103809, miR-377-3p and GOT1 mRNA, and the experimental procedures had all been documented in the previous publications [36,37]. Of note, to detect hsa_circRNA_103809 levels, the total RNA must be pre-treated with RNase R enzyme (3 U/μg) for 20 min at 37°C to eliminate linear RNA. The primer sequences for the involved genes are as follows: hsa_cir-cRNA_103809 (Forward: 5′-ACG CAT TCT TCG AGA  CCT

Western blot analysis
The RIPA lysis buffer was purchased from Solarbio (Beijing, China) to lyse the NSCLC cells/tissues and extract the total protein, according to the experimental procedures recorded in the previous publications [36,37], the expression levels of GOT1, β-actin, cyclin D1, CDK2, cleaved caspase-3 and Bax were examined by using the Western Blot analysis. Specifically, the 40 μg/lane protein lysates were separated by using the 10 -15% SDS-PAGE, and the target protein bands were transferred onto the PVDF membranes (Millipore, USA). Next, the membranes were incubated with 5% skim milk for 70 min at room temperature for blocking, and the membranes were probed with the primary antibodies against GOT1 (1: 1500, MW: 50 kDa, #PA5-24634, Thermo, USA), β-actin and Bax (1:1500, MW: 21 kDa, #ab32503, Abcam, UK) overnight at 4°C. After washing by PBS buffer for 3 times, the PVDF membranes were incubated with the secondary antibody (Abcam, UK) for 120 min at room temperature. Finally, the protein bands were visualized by ECL system (GE Healthcare Bio-science, USA) and quantified by using the Image J software.

Dual-luciferase reporter gene system assay
The binding sites of miR-377-3p with hsa_circRNA_ 103809 and 3′ UTR region of GOT1 mRNA were predicted by the online miRDB software (http://mirdb.org/), and validated by using the dual-luciferase reporter gene system, and the detailed experimental procedures had been well documented in the previous literatures [36,37]. Briefly, the targeting sites in hsa_circRNA_103809 and GOT1 were mutated, and named as Mut-circRNA and Mut-GOT1, respectively. Correspondingly, the original wild-type (Wt) genes were named as Wt-CircRNA and Wt-GOT1. The above sequences were cloned into the luciferase reporter vectors by Sangon Biotech (Shanghai, China), and the schematic image for the luciferase reporters was shown in Figure S3. The above vectors were delivered into NSCLC cells co-transfected with miR-377-3p mimic and inhibitor for 48 h. After that, the commercial dual-luciferase reporter assay kit (Promega, USA) was used to measure relative luciferase activities in the cells.

Xenograft tumor-bearing mice models
The CR-NSCLC cells were pre-transfected with different vectors, and were subcutaneously injected into the dorsal flank of male nude mice (N = 20), and the age of the mice ranged from 6 to 8 weeks. Each mouse was injected with 5 × 10 6 cells, at 7 days post-injection, the tumor were subjected to high-dose cisplatin (10 μg/ml) treatment every 3 days for 2 weeks. The above mice were equally divided into 4 groups, including Control, Cisplatin, KD-circRNA and Cisplatin + KD-circRNA, each group had 5 mice. The mice were sacrificed at 35 days post-injection. After that, the mice tumor tissues were collected, and the expression levels of proliferation associated proteins (cyclin D1 and CDK2) and apoptosis associated proteins (cleaved caspase-3 and Bax) were examined by using Western Blot analysis, and the expressions/localization of Ki67 protein in mice tissues were determined by Immunohistochemistry (IHC). All the animal experiments were approved by the Ethics Committee of The Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital), Institute of Basic Medicine and Cancer (IBMC), and the approval number was 2020-12-002.

Immunohistochemistry (IHC)
The mice tumor tissues were collected and spliced into sections of 5 μm thickness, and IHC assay was conducted to determine the expressions and localization of Ki67 protein in the mice tissues, the detailed experimental procedures can be found at the previous publications [36,37]. The antibody against Ki67 protein was bought from Abcam (UK), and was diluted at the ratio of 1:400.

Statistical analysis
Data analysis was conducted by using the SPSS 18.0 software, and the data was represented as Means ± Standard Deviation. The comparisons between two groups were performed by using the Student's t-test, and the comparisons among multiple groups were conducted by using one-way ANOVA analysis. Each experiment was repeated at least 3 times, *P < 0.05 could be regarded as statistical significance.

Results
The expression patterns of hsa_circRNA_103809, miR-377-3p and GOT1 in CS-NSCLC and CR-NSCLC cells The CS-NSCLC cell lines (A549, H1299 and Calu-3) were subjected to continuous low-dose cisplatin treatment to generate CR-NSCLC cells (A549/DDP, H1299/DDP and Calu-3/DDP), which simulated the realistic conditions of cisplatin-resistance in NSCLC patients in vitro. Next, the NSCLC cells were stimulated with high-dose cisplatin (25 μg/ml) for 0 h, 24 h, 48 h and 72 h, and cell proliferation was evaluated by the CCK-8 assay (Fig. 1a-c). The results showed that the proliferation abilities in CS-NSCLC cells but not in CR-NSCLC cells, were significantly inhibited by cisplatin treatment (Fold changes (72 Figure 1d-f). Next, the cells were stained with Annexin V-FITC and PI, and cell apoptosis was detected by using flow cytometry (FCM) (Fig. 1g). As expected, the data suggested that cisplatin induced apoptotic cell death in CS-NSCLC cells, compared to the CR-NSCLC cells (Fold changes: 9.07, 8.56 and 6.38 vs. CR-NSCLC cells, Fig. 1g), suggesting that CR-NSCLC cells were much more resistant to cisplatin treatment. Next, the expression status of hsa_circRNA_ 103809, miR-377-3p and GOT1 were examined in the NSCLC cells, and we found that hsa_circRNA_103809 (Fig.  1h) and GOT1 (Fig. 1j, k) were upregulated, while miR-377-3p (Fig. 1i) was downregulated in CR-NSCLC cells, suggesting that continuous low-dose cisplatin pressure altered the expression status of hsa_circRNA_103809, miR-377-3p and GOT1 in CR-NSCLC cells.
The regulatory mechanisms of hsa_circRNA_103809, miR-377-3p and GOT1 in NSCLC cells By using the online miRDB software (http://mirdb.org/), we predicted a relationship among hsa_circRNA_103809, miR-377-3p and GOT1 (Fig. 2). Mechanistically, binding sites for miR-377-3p with hsa_circRNA_103809 (Fig. 2a) and the 3′ untranslated region (3'UTR) of GOT1 mRNA (Fig. 2f) were predicted, which were validated by the subsequent dual-luciferase reporter gene system. The results showed that miR-377-3p mimic targeted the binding sites in hsa_circRNA_103809 (Fig. 2b-d) and 3'UTR of GOT1 mRNA ( Fig. 2g-i) to decrease the relative luciferase activities in CS-NSCLC cells, while miR-377-3p inhibitor had the opposite effects ( Fig. 2b-d, g-i). Additionally, the RNA pull-down assay verified that miR-337-3p could be enriched by biotin-labelled hsa_circRNA_103809 (Fig. 2e) and GOT1 mRNA (Fig. 2j) probes but not in the control probes. Next, the overexpression and downregulation vectors for hsa_circRNA_103809 were transfected into CS-NSCLC cells (Figure S2A), and the results showed that hsa_circRNA_103809 positively regulated GOT1 expressions in CS-NSCLC cells (Fig. 2k, l). In addition, the miR-337-3p mimic and inhibitor were delivered into CS-NSCLC cells ( Figure S2B), and the results showed that miR-337-3p inhibited GOT1 expressions in CS-NSCLC cells (Fig. 2m, n). Of note, the promoting effects of hsa_circRNA_103809 overexpression on GOT1 were abrogated by upregulating miR-337-3p (Fig. 2o, p). The above results indicated that hsa_cir-cRNA_103809 sponged miR-337-3p to upregulate GOT1 in CS-NSCLC cells. . d-f Trypan blue staining assay was conducted to evaluate NSCLC cell viability. g Cell apoptosis ratio was measured by using the Annexin V-FITC/PI double staining method. Real-Time qPCR was used to examine the expression levels of (h) hsa_circRNA_103809, i miR-377-3p and j GOT1 mRNA in NSCLC cells. k Western Blot analysis was employed to determine the protein levels of GOT1 in NSCLC cells, full-length blots/gels are presented in Supplementary Figure S4. Each experiment was repeated at least 3 times. *P < 0.05 Knock-down of hsa_circRNA_103809 sensitized CR-NSCLC cells to cisplatin by regulating the miR-377-3p/GOT1 axis Further experiments were conducted to investigate the regulating effects of hsa_circRNA_103809 on cisplatinresistance in NSCLC. To achieve this, the silencing vectors for hsa_circRNA_103809 were transfected into CR-NSCLC cells to knock down hsa_circRNA_103809 ( Figure S1A), and the CCK-8 results showed that either hsa_circRNA_ 103809 ablation or cisplatin treatment alone had little effects on cell proliferation abilities in CR-NSCLC cells, while silencing of hsa_circRNA_103809 enhanced the inhibiting effects of cisplatin on cell growth (Fig. 3a-c). Then, the miR-337-3p downregulation ( Figure S1B Figure 3g).

Targeting hsa_circRNA_103809 enhanced the inhibiting effects of cisplatin on CR-NSCLC cell growth in vivo
Next, we validated the above cellular results in vivo. To achieve this, the CR-NSCLC cells were pre-transfected with hsa_circRNA_103809 downregulation vectors, and the cells were subcutaneously injected into the dorsal flank of nude mice to establish xenograft tumor-bearing mice models. At 7 days post-injection, the tumor were subjected to high-dose cisplatin treatment every 3 days. The mice were sacrificed at day 35, and the tumor tissues were collected, prepared and analyzed by Western Blot analysis and immunohistochemistry (IHC) (Fig. 5). As shown in Fig. 5a-f, either cisplatin alone or hsa_cir-cRNA_103809 downregulation alone had little effects on the proliferation and apoptosis associated proteins, while knock-down of hsa_circRNA_103809 and cisplatin combination treatments downregulated Cyclin D1 and CDK2 to hamper cell cycle, and upregulated cleaved Caspase-3 and Bax to trigger apoptotic cell death in A549/DDP, H1299/DDP and Calu-3/DDP cells in vivo. Consistently, the expressions and localization of Ki67 protein were examined by IHC, and the images in Fig. 5g showed that cisplatin significantly decreased the expression levels of Ki67 protein in CR-NSCLC cells with hsa_circRNA_103809 downregulation in mice tumor tissues.

Discussion
Cisplatin is the first-line chemotherapeutic drug for nonsmall cell lung cancer (NSCLC) treatment in clinic [13,14], however, long-term cisplatin treatment causes cisplatinresistance in NSCLC cells, which seriously limits the therapeutic efficacy of this chemical drug, resulting in cancer recurrence and bad prognosis in NSCLC patients [13,14]. Based on the above information, recent studies focused on uncovering the underlying mechanisms of cisplatin-resistance in NSCLC, and managed to identify potential therapeutic targets to improve cisplatin-sensitivity in NSCL C cells [38,39]. Among all the cancer associated genes, researchers noticed that circular RNAs (circRNAs) were closely associated with cancer progression and drug resistance in NSCLC, and identification of novel circRNAs that regulated NSCLC pathogenesis and drug resistance became necessary and meaningful [13,14]. Therefore, in the present study, we identified a novel circRNA, hsa_circRNA_ 103809, that regulated cisplatin-resistance in NSCLC. Mechanistically, according to the previous publications [36,37], the cisplatin-resistant NSCLC (CR-NSCLC) cells were inducted from their corresponding parental cisplatinsensitive NSCLC (CS-NSCLC) cells, and we found that hsa_circRNA_103809 tended to be highly expressed in CR-NSCLC cells, compared to CS-NSCLC cells. Interestingly, further experiments evidenced that knock-down of hsa_cir-cRNA_103809 enhanced the inhibiting effects of cisplatin on cell proliferation and viability in CR-NSCLC cells. Furthermore, upregulation of hsa_circRNA_103809 increased Fig. 4 Upregulation of hsa_circRNA_103809 promoted cisplatin-resistance in CS-NSCLC cells. a-c Cell proliferation abilities were examined by using the CCK-8 assay. d-f Cell viability was evaluated by performing the trypan blue staining assay. g Cell apoptosis was determined by using the Annexin V-FITC/PI double staining assay. (Note: "Control: without vectors transfection and cisplatin treatment"). Each experiment was repeated at least 3 times. *P < 0.05 cisplatin-resistance in CS-NSCLC cells, implying that targeting hsa_circRNA_103809 could potentially improve cisplatin-sensitivity in NSCLC cells. Previous data suggested that hsa_circRNA_103809 acted as an oncogene to promote cancer progression [15][16][17][18][19], and this study evidenced that hsa_circRNA_103809 also modulated drug resistance in NSCLC, which broadened our knowledge in this field.
Glutamate oxaloacetate transaminase 1 (GOT1) is crucial for promoting cancer progression by regulating glutamate metabolism [31,32], and inhibition and silencing of GOT1 had been validated as an effective strategy to impair cancer growth [31,32]. Interestingly, previous data suggested that GOT1 could be regulated by cisplatin [33], and our experiments validated that hsa_cir-cRNA_103809 positively regulated GOT1 in NSCLC  Figure S6E-F). g IHC was performed to examine the expressions and localization of Ki67 protein in mice tumor tissues, the signal intensity in different groups were assessed as follows: Control (+++), Cisplatin (+++), KD-circRNA (+++) and Cis + KD-circRNA (+). (Note: "Control: without vectors transfection and cisplatin treatment"). Each experiment was repeated at least 3 times. *P < 0.05 cells through miR-377-3p. Mechanistically, there existed binding sites between miR-377-3p and 3'UTR of GOT1 mRNA, and miR-377-3p negatively regulated GOT1 in NSCLC cells at both transcriptional and translational levels. In addition, we noticed that upregulation of hsa_circRNA_ 103809 increased GOT1 expression levels, which were reversed by overexpressing miR-377-3p, implying that hsa_ circRNA_103809 sponged miR-377-3p to upregulate GOT1 in NSCLC cells. Furthermore, knock-down of hsa_ circRNA_103809 increased cisplatin-sensitivity in CR-NSCLC cells, while overexpression of hsa_circRNA_103809 increased cisplatin-resistance in CS-NSCLC cells, which were reversed by overexpressing and silencing GOT1, suggesting that hsa_circRNA_103809 upregulated GOT1 to modulate cisplatin-resistance in NSCLC cells. Finally, by performing the in vivo experiments, we evidenced that knock-down of hsa_circRNA_103809 triggered apoptotic cell death to inhibit tumorigenesis in the xenograft tumorbearing mice models.
Interestingly, recent data noticed that NRF2 mediated glutamine metabolism was closely associated with chemo-resistance in pancreatic cancers [40], given that GOT1 served as a crucial regulator for glutamate metabolism, we hypothesized that there might exist connections between NRF2 and GOT1 in regulating cisplatinresistance in NSCLC. However, the detailed mechanisms are still needed to be studied. In addition, since Kirsten rat sarcoma viral oncogene homolog (KRAS) is one of the driver gene of NSCLC [41], it was worthy to investigate the interplay between KRAS gene and the hsa_cir-cRNA_103809/miR-377-3p/GOT1 pathway in regulating NSCLC development in our future work.

Conclusions
Taken together, through in vitro and in vivo experiments, this study found that targeting the hsa_circRNA_103809/ miR-377-3p/GOT1 pathway inhibited cell proliferation and viability, and triggered cell apoptosis to increase cisplatin-sensitivity in NSCLC cells. Our work broadened our knowledge in this filed, and provided potential therapeutic targets to improve the therapeutic efficacy of current chemical drug for NSCLC.