DLX6-AS1 activated by H3K4me1 enhanced secondary cisplatin resistance of lung squamous cell carcinoma through modulating miR-181a-5p/miR-382-5p/CELF1 axis

Cisplatin (CDDP) based chemotherapy is widely used as the first-line strategy in treating non-small cell lung cancer (NSCLC), especially lung squamous cell carcinoma (LUSC). However, secondary cisplatin resistance majorly undermines the cisplatin efficacy leading to a worse prognosis. In this respect, we have identified the role of the DLX6-AS1/miR-181a-5p/miR-382-5p/CELF1 axis in regulating cisplatin resistance of LUSC. qRT-PCR and Western blot analysis were applied to detect gene expression. Transwell assay was used to evaluate the migration and invasion ability of LUSC cells. CCK-8 assay was used to investigate the IC50 of LUSC cells. Flow cytometry was used to test cell apoptosis rate. RNA pull-down and Dual luciferase reporter gene assay were performed to evaluate the crosstalk. DLX6-AS1 was aberrantly high expressed in LUSC tissues and cell lines, and negatively correlated with miR-181a-5p and miR-382-5p expression. DLX6-AS1 expression was enhanced by H3K4me1 in cisplatin resistant LUSC cells. Besides, DLX6-AS1 knockdown led to impaired IC50 of cisplatin resistant LUSC cells. Furthermore, DLX6-AS1 interacted with miR-181a-5p and miR-382-5p to regulate CELF1 expression and thereby mediated the cisplatin sensitivity of cisplatin resistant LUSC cells. DLX6-AS1 induced by H3K4me1 played an important role in promoting secondary cisplatin resistance of LUSC through regulating the miR-181a-5p/miR-382-5p/CELF1 axis. Therefore, targeting DLX6-AS1 might be a novel way of reversing secondary cisplatin resistance in LUSC.


Discussion
Hitherto, the standard chemotherapy regimen for LUSC is still based on platinum-containing alkylating agents 30 .
Although most LUSC patients initially respond well to chemotherapy, the clinical outcome of LUSC remains extremely poor due to inevitably recurrence and development of secondary chemoresistance 31 . In recent years, a large number of genome-wide studies verified that lncRNAs played a vital role in the chemoresistance of tumor cells 32 . For instance, lncRNA RAD51-AS1 reduced the activity of DNA damage repair pathway involved in RAD51 to increase the etoposide sensitivity of HCC cells 33 . LncRNA LINC00461 negatively regulated the expression of miR-411-5p to mediate docetaxel resistance in breast cancer 34 . In present research, we found that lncRNA DLX6-AS1 was highly expressed in LUSC and local recurrence tissues, and overexpressed in LUSC cell lines. Additionally, survival analysis indicated that a high level of DLX6-AS1 was closely correlated with worse cancerspecific survival. Thus, DLX6-AS1 was selected to deeply study its role in secondary chemoresistance of LUSC.
Recently, increasing studies demonstrated that lncRNAs sponged the specific miRNAs to regulate the initiation, progression, and therapeutic response of lung cancer [35][36][37] . For example, lncRNA MALAT1 sponged miR-374b-5p to regulate the proliferation, migration, and invasion of NSCLC cells by upregulating SRSF7 38 . LncRNA MEG3 was overexpressed in NSCLC tissues and positively regulated SOX7 expression to induce cisplatin sensitivity of NSCLC cells by targeting miR-21-5p 39 . Our study found that DLX6-AS1 had a higher expression in LUSC tissues and cell lines, while miR-181a-5p and miR-382-5p expression were at a lower level in LUSC tissues and cell lines. Furthermore, DLX6-AS1 knockdown restrained proliferation, invasion, migration, and promoted apoptosis of LUSC cells. Recent studies of DLX6-AS1 involved in cell growth and apoptosis of breast cancer and renal cell carcinoma uncovered similar results 40,41 . The further assay showed that miR-181a-5p and miR-382-5p were up-regulated after DLX6-AS1 inhibition. Accordingly, Dual luciferase reporter gene assay and RNA pull-down assay further demonstrated that DLX6-AS1 directly interacted with miR-181a-5p and miR-382-5p. Therefore, our results indicated that DLX6-AS1 negatively modulated the expression of miR-181a-5p and miR-382-5p.
Then we explored the potential effect of DLX6-AS1 in LUSC secondary chemoresistance and found that DLX6-AS1 expression was increased in CDDP resistance LUSC cells, while its expression was reversed after treated with CDDP. Interestingly, further experiments revealed that H3K4me1 was enriched at the DLX6-AS1 promoter in CDDP resistance LUSC cells, and this trend was partially reversed by CDDP treatment. This finding was important because the methylation and demethylation of histone precisely regulated specific regulatory processes by activating or inhibiting the expression of certain cancer-related genes 42 . For example, histone methylation around the PTEN promoter inhibited its protein expression in colon cancer and melanoma 43 . The JMJD2D histone demethylase was recruited at the p21 promoter, which led to up-regulated p21 expression in liver cancer 44 . The present results indicated that H3K4me1 was abundant at the promoter of DLX6-AS1 and was inhibited by CDDP treatment, suggesting that histone methylation modification played a crucial role in mediating the transcription of DLX6-AS1 in the CDDP resistance process of LUSC. Then we suppressed DLX6-AS1 expression in CDDP resistance LUSC cells and found that the CDDP resistance index was significantly decreased, indicating the promoting role of DLX6-AS1 in LUSC chemoresistance. Moreover, miR-181a-5p and miR-382-5p expression were lower in CDDP resistance LUSC cells, and their overexpression significantly enhanced the cisplatin sensitivity of CDDP resistance LUSC cells, whereas their knockdown partially reversed this trend. Previous studies highlighted the tumor suppressor function of miR-181a-5p and miR-382-5p in regulating the progression and chemoresistance of several malignant tumors. Yang et al. revealed that miR-181a-5p enhanced the sensitivity of esophageal adenocarcinoma cells to cisplatin 45 . Zheng et al. suggested that miR-382-5p inhibited proliferation and invasion of breast cancer through regulating SNHG1 expression 46 . Similarly, our research also illustrated that miR-181a-5p and miR-382-5p were down-regulated in LUSC tissues and cell lines, and miR-181a-5p and miR-382-5p overexpression remarkably restrained the effect of platinum on cell viability of LUSC, which was reversed by the depletion of miR-181a-5p and miR-382-5p.
Generally, miRNAs could bind to the 3′ UTR region of a specific mRNA to restrain its transcription, thereby affecting its function 47 . In present study, CELF1 was predicted as the potential target of miR-181a-5p and miR-382-5p, and the interaction relationship was verified by qRT-PCR, Western blot, and Dual luciferase reporter gene assay. Former investigations revealed that CELF1 could enhance cell migration, invasion, and chemoresistance in colorectal cancer by targeting ETS2 28 . CELF1 was abundantly expressed in the glioma tissues, and CELF1 expression was repressed by miR-330-3p to inhibit proliferation and migration of glioma cells 27 . Similarly, our results also indicated that CELF1 was overexpressed in CDDP resistance LUSC cells, and CELF1 attenuation led to the decline of CDDP resistance in LUSC cells.
In conclusion, this research was the first to illustrate that DLX6-AS1 was highly expressed in LUSC tissues and cell lines, and the aberrant expression of DLX6-AS1 was tightly associated with proliferation, apoptosis, and local recurrence of LUSC. Moreover, our results demonstrated that DLX6-AS1 induced by H3K4me1 facilitated platinum-based chemoresistance in LUSC through interacting with miR-181a-5p/miR-382-5p/CELF1 axis. Hence, these findings identified the pivotal role of DLX6-AS1 in the progression and secondary chemoresistance in LUSC and provided a novel therapeutic target for patients with LUSC. Cell culture. SK-MES-1, NCI-H226, and MRC-5 cell lines were purchased from American type culture collection (ATCC, USA). DMEM medium (HyClone, USA) supplemented with 10% fetal bovine serum (Gibco, Rockville, MD) and 1% penicillin-streptomycin (HyClone, USA) in a 5% CO 2 incubator at 37 °C. The SK-MES-1 cells were then treated with continuous low-dose of cisplatin in a stepwise manner to developed cisplatin resistant SK-MES-1 (SK-MES-1-resistance) cells. This process was repeated until cells could stably survive in 1 µg/mL cisplatin medium. Then we increased the concentration of cisplatin to 2, 4, 6, 8 and 10 µg/mL. Resistance Index (IC50 of resistant cells/IC50 of parental cells) was used to evaluate the resistance ability. Resistance Index larger than 5 was considered resistant.
qRT-PCR. RNeasy Mini Kit (Qiagen, Leusden, Netherlands) was used to extract the total RNA extraction.
PrimeScript™ RT reagent Kit (Takara Biotechnology Ltd., China) was used to reverse transcribe the extracted the RNA into cDNA. As for miRNAs, One Step PrimeScript™ miRNA cDNA Synthesis Kit was used for reversetranscription. SYBR Premix Ex Taq II kit (TaKaRa, Shiga, Japan) was used for the qRT-PCR process. 2−ΔΔCt method was used to quantify the results. U6 and GAPDH were used as the internal control for miRNAs and genes respectively. The primers are as follows:

Isolation of cytoplasmic and nuclear RNA. Cytoplasmic & Nuclear RNA Purification Kit (Norgen,
Canada) was used to isolate cytoplasmic and nuclear RNA. According to the manufacturer's guidance, lysis buffer J was used to lyse the target cells. Then the cell lysis was centrifuged; the supernatant was used to extract the cytoplasmic fraction and a pellet was used to extract the nuclear fraction. Buffer SK and ethanol were used to purify and extract the cytoplasmic and nuclear RNA.
Chromatin immunoprecipitation (ChIP) assay. ChIP assay was conducted by using an EpiQuik ChIP kit (Epigentek, USA) according to the instruction of the manufacturer. Briefly, LUSC cells were fixed with 1% formaldehyde at room temperature for 10 min. Then cells were added with 1/10 volume of 1.25 M glycine to stop fixation and incubated for 5 min. Then added 5 mL cell lysis buffer to resuspend the cell pellet and lysed on ice for 10 min. After centrifugation, cells were added with 1 ml ChIP buffer (added 12 μL PMSF and 10 μL protease inhibitor for every 1 mL ChIP buffer) for sonication (10 min, 10 s on, 10 s off). Then 200 µg of the protein-chromatin complex was extracted for each immunoprecipitation, and the pre-blocked dynabeads protein G (Invitrogen) was used to capture antibody-protein complexes. The eluted product was purified with a DNA purification kit (Invitrogen) to obtain ChIP DNA. Then the ChIP DNA was analyzed by qPCR. The antibodies included: H3K4me1 (Cell Signaling Technology, USA), and normal mouse IgG (Epigentek, USA).

Statistics analysis.
All data involved in this study were processed and visualized by using the SPSS 18.0 software and the Graphpad Prism 8.2 software. The results presented were expressed as Mean ± Standard Deviation (SD). A paired Student's t-test was used for comparison between two groups. One-way ANOVA was conducted to compare the differences among multiple groups. Survival analysis was performed by applied Kaplan-Meier analysis. Each experiment in this research was independently repeated at least three times. A P value of less than 0.05 was considered to indicate a statistically significant difference.