LncRNA MIR31HG fosters stemness malignant features of non-small cell lung cancer via H3K4me1- and H3K27Ace-mediated GLI2 expression

Non-coding RNAs are responsible for oncogenesis and the development of stemness features, including multidrug resistance and metastasis, in various cancers. Expression of lncRNA MIR31HG in lung cancer tissues and peripheral sera of lung cancer patients were remarkably higher than that of healthy individuals and indicated a poor prognosis. Functional analysis showed that MIR31HG fosters stemness-associated malignant features of non-small cell lung cancer cells. Further mechanistic investigation revealed that MIR31HG modulated GLI2 expression via WDR5/MLL3/P300 complex-mediated H3K4me and H3K27Ace modification. In vivo MIR31HG repression with an antisense oligonucleotide attenuated tumor growth and distal organ metastasis, whereas MIR31HG promotion remarkably encouraged cellular invasion in lung and liver tissues. Our data suggested that MIR31HG is a potential diagnostic indicator and druggable therapeutic target to facilitate multiple strategic treatments for lung cancer patients.

for 15 min before optical imaging using an Olympus IX53 inverted fluorescence microscope paired with an Olympus DP73 color camera (Olympus, Tokyo, Japan).
Molecular imager (ChemiDocXRS+, Bio-Rad, Hercules, CA, USA) was used to screen and analyze the densitometry of each visual field.

Wound healing assay
Cells were seeded into 12-well plate at a density of 1 × 10 5 cells per well for 24 h.When cells reached 90% confluence, a gap was created using a 200 µL tip of a pipette and washed with phosphate-buffered saline (PBS).Cells were incubated with serum-free medium for 24 h before photomicrographs were taken with the IX53 inverted microscope and DP73 color camera (Olympus).Relative distance of the remaining wound area was calculated as previously described [1].Experiments were performed in triplicate.

Transwell assay
A transwell migration assay was performed as previously described [2].Briefly, 1 × 10 5 cells were suspended in 200 μL of RPMI 1640 medium (Biological Industries, Beit-Haemek, Israel) without FBS and seeded on the top chamber of transwell inserts (3422,Corning).The lower chambers contained RPMI1640 medium with FBS.After incubation for 48 h, the chamber was rinsed three times with PBS, then dyed with crystal violet for 30 s and rinsed again with PBS five times to obtain a transparent background.Plates were imaged using the IX53 inverted fluorescence microscope and the DP73 color camera (Olympus).

Immunofluorescence staining
Cells were seeded in 24-well plates at 5 × 10 4 per well.After 24 h, cells were fixed as previously described [2].Cells were incubated with primary antibodies overnight at 4°C.As a negative control, slides incubated with PBS were stained in parallel.Then cells were incubated with the secondary antibody Alexa Fluor 594 AffiniPure Goat 33112ES60,Yeasen Biotechnology Co.,Ltd.,Shanghai,China)and diluted with PBS for 2 h at 37 °C.Cells were counterstained with 4',6-diamidino-2-phenylindole (DAPI) and visualized with an advanced fluorescence microscope (Olympus BX63).

Co-IP
H1299 and H520 cells were collected and lysed for 30 min on ice.A part of the centrifuged lysate was used as an input sample.The other part of the lysate was mixed with 5 μg antibody and 30 μL rProtein A/G MagPoly Beads (SM015005, Changzhou Smart-Lifesciences Biotechnology Co., Ltd.) and vortexed for 3 h at 4°C.The mixture was washed five times with TBST.Subsequently, the mixture was inoculated with 25 µL protein loading buffer and boiled for 10 min at 95 °C.Samples were centrifuged and stored at -80 °C for further analysis.Antibodies against WDR5 (#13105, Cell Signaling Technology, Danvers, Massachusetts, USA), mono-methyl-histone H3 (#5326, Cell Signaling Technology), MLL3 (28437-1-AP, Proteintech, Wuhan, China) and P300 (A13016, Abclonal, Woburn, MA, USA) were used for protein immunoprecipitation.

RIP assay
H1299 and H520 cells were either treated with 20 μM WDR5 inhibitor OICR-9429 (HY-16993/CS-5776, MedChemExpress) or left untreated for 48 h; the procedure was identical to that subsequent for the control and experimental groups.All cell precipitates were chilled in polysome lysis buffer for 5-10 min on ice.All lysates were centrifuged at 13,000 g for 10 min; some supernatant was preserved as an input sample, and the rest was treated with antibody-rProtein A/G MagPoly Beads (SM015005, Changzhou Smart-Lifesciences Biotechnology Co., Ltd., Changzhou, China) mixture, which was preincubated at 25℃ for 30 min and vortexed at 4 °C overnight.Samples were washed six times with NT-2 buffer and further treated with proteinase K buffer (MedChemExpress; HY-108717).After heating in a water bath at 55 °C for 30 min, the RNA was extracted at 5,000 g for 15 s using a mini centrifuge (D1008E, DLAB Scientific, Beijing, China) and by collecting the supernatant of the NT-2 buffer.The antibodies used in the above experiments are those against WDR5 (#13105, Cell Signaling Technology), MLL3 (28437-1-AP, Proteintech) and P300 (A13016, Abclonal, Wuhan, China) that were used for MIR31HG immunoprecipitation.
Membranes were incubated with primary antibodies at 4 °C for 12 h, then incubated with secondary antibodies for 2 h at 37 °C.Enhanced chemiluminescence signals were detected using Molecular imager (ChemiDocXRS+, Bio-Rad).Antibody against tubulin (#5568, Cell Signaling Technology) was used in parallel as an internal control.

RNA isolation and real-time qPCR (qRT-PCR)
Total RNA of cultured cells was isolated using a Total RNA Isolation Kit (Shanghai Promega Biological, Beijing, China) following manufacturer's instructions.Blood samples from healthy and NSCLC patients were obtained from Binzhou Medical University Hospital in accordance with a protocol approved by the Review Board (Approval No.: KYLL-2021-05).lncRNA was extracted from sera using a BIOG cfRNAEasy Kit (51028, Changzhou Bio-generating Biotechnology Corp, (Guangzhou, China).A RevertAid First Strand cDNA Synthesis Kit (K1622, Thermo Fisher Scientific) was used for cDNA synthesis.The concentrations of cDNA in different groups were measured and adjusted for qPCR.Real-time qPCR was performed on a CFX96™ Real-Time PCR System (1855195, Bio-Rad) with a 20 μL reaction pool, including 10 μL FastStart Universal SYBR Green Master (Novoprotein), 1 μL forward primer (Sangon Biotech), 2 μL cDNA, 1 μL reverse primer, and 6 μL deionized DEPC water.PCR conditions were as follows: 95 °C, 10 s; 95 °C, 5 s, 54 °C, 34 s, 40 cycles.
Gene expression was normalized to that of GAPDH in the quantification analysis.All primer sequences used in this study are shown in Table S3.

Stably transfected MIR31HG overexpression and knockdown cells, GLI2
overexpression and knockdown cells were harvested, filtered, and centrifuged.anti-CD34 (A23107) and anti-CD133 (A0219) (ABclonal, Wuhan, China) antibodies were used for surface staining overnight at 4 °C.After washing with phosphate-bufered saline (PBS) for three times, adding fluorescein secondary antibody and incubating at room temperature for 1 h.After washing with PBS buffer for three times and finally resuspended in PBS buffer for surface marker detection On the BD C6 Flow C6 flow meter and analyzed in FlowJo software.

Tissue microarray and probe hybridization
cDNA from 15 pairs of human NSCLC tissues and adjacent lung specimens were obtained from Shanghai Outdo Biotech (cDNA-HLugA030PT01) for further real-time qPCR analysis.A tissue microarray (HLug-Squ150Sur-01) was purchased from Shanghai Outdo Biotech Co., Ltd.(Shanghai, China), who also performed probe hybridization to detect the transcription levels of lncRNA-MIR31HG in 75 paired NSCLC tissues and adjacent specimens.Probes detecting MIR31HG were synthesized according to manufacturer's instructions for the Enhanced Sensitive ISH Detection Kit Ⅰ(MK1030, Boster Bio, Wuhan, China).

Tumor-bearing mouse model construction and treatments
Four-week-old BALB/c nude mice (female, 15-18 g) were purchased from Ji'nan Pengyue Laboratory Animal Breeding Co., Ltd.(Jinan, China) and kept in a specific pathogen-free laboratory animal room.The use of animals was approved by the Animal Experiments Ethical Committee of Binzhou Medical University Hospital and performed according to the Guide for Care and Use of Laboratory Animals.A tumorbearing mouse model was developed via subcutaneous injection of 5 × 10 6 H520 cells per mouse.When tumors reached approximately 5 mm in diameter, mice were randomly divided into an H520 control group, an H520 MIR31HG overexpression group, an ASO control group, and an ASO-MIR31HG group for subsequent treatment (n = 5 in each group).To evaluate the therapeutic potential of targeting MIR31HG with ASO, 5 nmol of ASO-hMIR31HG-001 (Ribobio, Guangzhou, China; lnc6211215095815) were dissolved in 50 μL RNase free ddH2O and administered via intra-tumoral injection every 2 d for 21 d.ASO control group mice were injected with saline.The tumor size and body weight of mice were measured every alternate day.
Tumor volume was calculated as previously described [1].All mice were euthanized after 4 weeks of treatment and tumors were excised for further analysis.

Histological analysis
Hematoxylin and eosin and immunohistochemistry staining of paraffin sections were performed as previously reported [1].Images were observed via bright field microscopy (BX63, Olympus) and quantified using Image J software.For immunofluorescence staining, tissues were immersed in 30% sucrose solution overnight and embedded in Tissue Tek OCT compound (4583, Sakura Finetek, Tokyo, Japan).Subsequently, frozen sections (5 μm) were prepared using a cryostat (Leica CM1950, Leica, Wetzlar, Germany).Frozen sections were incubated with 5% normal goat serum (Zhongshan Golden Bridge Biotechnology, Beijing, China) for 40 min at 37°C.Primary antibodies against SOX2 (A0561, Abclonal) and Ki67 (A16919, Abclonal) were diluted 1:100 in PBS, applied to cover sections, and incubated at 4°C for 12 h.Sections incubated with only PBS were used as negative controls.Sections were then incubated with secondary antibody Alexa Fluor 594 AffiniPure Goat Anti-Rabbit IgG (H+L; 33112ES60, Yeasen Biotechnology Co., Ltd.) in PBS for 2 h at 37°C.DAPI was used to co-stain the cell nuclei.Images were acquired under an advanced fluorescence microscope (BX63, Olympus BX63) and quantified using Image J software.

Quantification and statistical analysis
Data are presented as mean ± standard error of mean (SEM).Graphpad Prism Software 8.0 (San Diego, CA, USA) was used for statistical analysis.Comparisons of two groups were analyzed with two-tailed Student's t-test.Multiple comparisons were performed using one-way analysis of variance (ANOVA) or two-way ANOVA.Differences were considered significant at p < 0.05.
Fig. S1 MIR31HG promotes migration capability in NSCLC cells.A, B Wound healing assay (left)

Fig
Fig. S6 GLI2 promotes stemness features of NSCLC cells.A, B Immunofluorescent staining of H1299 and H520 cells upon forced GLI2 overexpression or repression, compared to pc3.1 and shv empty vector separately.C, D The expression of cancer stemness markers CD133 and CD34 were detected by FACS analysis upon GLI2 overexpression or knockdown compared to levels in empty vector.E Schematic map of the primer locations covering promoter of SOX2.F Interaction between GLI2 and SOX2 promoter regions (-2000 to +300 around transcriptional starting site) in NSCLC cells explored by chromatin Immunoprecipitation RT-qPCR.Data are presented as mean ± SEM; significance abundance: * p < 0.05, ** p < 0.01, *** p < 0.001.
Fig. S7 Repression of GLI2 retards stemness features of NSCLC cells caused by MIR31HG OE.A