microRNA-301b-3p downregulation underlies a novel inhibitory role of long non-coding RNA MBNL1-AS1 in non-small cell lung cancer

Background Non-small cell lung cancer (NSCLC) is the second most prevalent cause of cancer-related fatality. Long non-coding RNAs (lncRNAs) have been observed to exercise functions in NSCLC. Here, the current study aimed to explore the potential mechanism of lncRNA MBNL1-AS1 in NSCLC. Methods Microarray analysis was performed to screen the differentially expressed lncRNA associated with NSCLC and its potential mechanism. The lncRNA MBNL1-AS1 expression was quantified in 56 paired NSCLC and adjacent normal tissue samples. In an attempt to outline the function of lncRNA MBNL1-AS1 in NSCLC and to identify the interaction among lncRNA MBNL1-AS1, microRNA-301b-3p (miR-301b-3p) and TGFBR2, ectopic expression, depletion, and reporter assay experiments were conducted to detect CSC proliferation, migration, invasion, drug resistance, and sphere formation in NSCLC. Results Initially, the intersection among lncRNA MBNL1-AS1, miR-301b-3p, and TGFBR2 was observed in NSCLC. While a poor expression of lncRNA MBNL1-AS1 and TGFBR2, along with a high expression of miR-301b-3p was observed in NSCLC tissues. A demonstration of lncRNA MBNL1-AS1 restoration significantly decreased CSC proliferation, migration, invasion, drug resistance, and sphere formation in NSCLC. LncRNA MBNL1-AS1 functioned as a sponge of miR-301b-3p, which inverted the inhibitory role of lncRNA MBNL1-AS1 in CSC proliferation, migration, invasion, drug resistance, and sphere formation in NSCLC. LncRNA MBNL1-AS1 positively regulated TGFBR2 which was a target gene of miR-301b-3p. At last, upregulated lncRNA MBNL1-AS1 or depleted miR-301b-3p suppressed the xenograft tumor formation in vivo. Conclusion Collectively, the present study suggests an inhibitory role of lncRNA MBNL1-AS1 in CSC drug resistance of NSCLC by upregulating miR-301b-3p-targeted TGFBR2.


Introduction
Non-small cell lung cancer (NSCLC) has been a persistent cause of cancer-related mortality worldwide, accompanied by a comparatively low 5-year survival rate [1]. NSCLC metastatic progression is correlated with the high mortality rates associated with NSCLC [2]. Early diagnosis of NSCLC has become a clinical essential in planning and providing effective treatment, which markedly improves the overall survival rate of patients with NSCLC [3]. However, the prognosis of patients suffering from NSCLC remains to be unfortunate [4]. In recent years, the theory of cancer stem cells (CSCs), that cancers are maintained by subpopulations of tumor cells with features of stem cells and progenitors, has served as a focus of in-depth investigations in cancer research [5]. A research reported the principal involvement of CSCs in tumor progression, recurrence, and drug resistance [6]. CSCs due to their high tumorigenicity can be observed in multiple human cancers, with great significance in NSCLC [7]. Targeting CSCs has been therapeutically proven as a curative strategy for NSCLC treatment [8], while the specific mechanisms still remain as a topic of research. Thus, it is necessary to develop new treatment strategies against CSCs as a mode of better clinical intervention in NSCLC.
Long non-coding RNAs (lncRNAs), comprising of over 200 nucleotides in length, with vital functionality as precarious modulators of cancer biology with significant impacts on cell proliferation, metastasis, and apoptosis [9]. MBNL has been proven as a participant in the modulation of selective splicing of pre-mRNA [10]. MBNL1 plays a vital role in the initiation of colorectal cancer (CRC) by reducing the expression of microRNAs (miRNAs/miRs) [11]. LncRNA MBNL1-AS1 has been elucidated to be downregulated in NSCLC [12]. Existing literature has implicated the functionality of various miRNAs like miR-504 as facilitators along the progression of NSCLC [13], thereby shedding light on the urgency for identification of novel valuable therapeutic targets for NSCLC. A clinically proven correlation has been implicated by miR-301b with vasoconstriction in pulmonary hypertension by regulating a target gene [14]. The participation of miR-301b-3p was proven to be vital in myogenic differentiation by targeting Rb1-inducible coiled-coil 1 [15]. Wu et al. have confirmed the presence of an aberrantly expressed miR-301b along the progression of lung cancer [16]. Furthermore, microarray analysis from the current study revealed and verified the transforming growth factor beta receptor II (TGFBR2) as a target of miR-301b-3p. The TGFBR2 belongs to the serine/threonine protein kinase family and the TGF-β receptor subfamily, and cases of disrupted TGF-β pathway have been demonstrated to stimulate tumor progression [17]. Accumulating evidence from a previous study revealed that TGFBR2 depletion also implicates along the progression of carcinogenesis of NSCLC [18]. With the abovementioned literature serving as a hypothesis, we obtained NSCLC tissues and adjacent normal tissues to explore and develop tumorigenicity assay in nude mice in order to verify the conceivable effects of lncRNA MBNL1-AS1 on the cellular processes of NSCLC CSCs via miR-301b-3p and TGFBR2.

Ethics statement
The study was performed with the approval of the Ethics Committee of The Affiliated Cancer Hospital of Zhengzhou University, Henan Cancer Hospital. All patients signed informed consents prior to tissue collection, which was in compliance with the Declaration of Helsinki [19]. Moreover, the animal experiment procedures were performed in accordance with the protocols issued by the Institutional Animal Research Committee of The Affiliated Cancer Hospital of Zhengzhou University, Henan Cancer Hospital. Measures were taken to minimize animal suffering.

Microarray analysis
Chip data relative to NSCLC, including GSE101929, GSE102286, and GSE33532, were retrieved from the Gene Expression Omnibus (GEO) database (https://www.ncbi. nlm.nih.gov/geo/), from which GSE102286 emerged as the miRNA expression chip. The Affy package of R language (http://www.bioconductor.org/packages/release/bioc/html/ affy.html) was employed for standard pretreatment of the chip data, and the limma package of R language (http:// master.bioconductor.org/packages/release/bioc/html/ limma.html) was employed so as to identify the differentially expressed genes (DEGs). The heat maps were drawn using the pheatmap package (https://cran.r-project.org/ web/packages/pheatmap/index.html).

Specimen collection
The NSCLC tissues and adjacent normal tissues samples (at least 5 cm from the edge of NSCLC mass) were collected from 56 patients (Table 1) with NSCLC who underwent surgical intervention in The Affiliated Cancer Hospital of Zhengzhou University, Henan Cancer Hospital from 2016 to 2017. None of the patients were administered chemotherapy or radiotherapy prior to the study.

Flow cytometry
According to a previous study [20], the NSCLC cells were stained with 5 μg/mL Hoechst 33342 (B2261, Sigma, St. Louis, MO, USA) irrespective of the presence or absence of 50 μM verapamil (V105, Sigma-Aldrich Chemical Company, St Louis, MO, USA), followed by staining with 2 μg/mL propidium iodide (PI, P4170, Sigma-Aldrich Chemical Company, St Louis, MO, USA). The cells were sorted in a FACS-Aria III flow cytometer (BD Biosciences, San Jose, CA, USA). The low Hoechst Red and low Hoechst Blue and the missing area of the verapamil group were set and prepared for the supplementary population (gate of SP cells). The cells were initially gated so as to sort the SP + cells and then analyzed for CD133 expression.

RNA isolation and quantitation
Total RNA was extracted using the Trizol Kit (10296010, Invitrogen, Carlsbad, CA, USA). The primer sequences were synthetized by BGI Co., Ltd. (Shenzhen, China) ( Table 2). Next, for RT-qPCR of miRNAs, 100 ng of total RNA was reverse-transcribed and subjected to Taqman® miRNA assay (Applied Biosystems), and for RT-qPCR of mRNAs, cDNA synthesis was performed with 1 μg of total RNA according to the instructions of the EasyScript First-Strand cDNA Synthesis SuperMix (AE301-02, Beijing TransGen Biotech Co., Ltd., Beijing, China). The ABI7500 quantitative PCR instrument was employed for RT-qPCR following the instructions of the SYBR®Premix Ex TaqTM II kit. The level of miR-301b-3p was normalized to the housekeeping gene U6, while the expression level of lncRNA MBNL1-AS1 and TGFBR2 was normalized to the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The expression ratio of the target gene between the experimental and control groups was calculated using the 2 −ΔΔCt method. The experiment was conducted three times.

G418 screening
The preliminary experiment was conducted to verify the extermination concentration of antibiotics (the minimum cell death concentration was regarded as the best screening concentration). After 24 h of transfection, G418 with the best screening concentration was added for the screening of stable transgenic plants. Briefly, upon termination of a large number of cells after 6 days of culturing, the cells were further incubated with increasing serum concentration. After 10 days of incubation, the concentration of G418 had reduced by half to maintain the optimum screening pressure. On the 14th day after the screening, the resistant clone occurred. After selection and large amplification of the monoclone using the limiting dilution assay, the total RNA was extracted in order to determine the expression of the target gene using RT-qPCR.

Fluorescence in situ hybridization (FISH)
The subcellular localization of lncRNA MBNL1-AS1 was detected using the FISH kit (BIS-P0001, Guangzhou Bersin Biotechnology Co., Ltd., Guangzhou, China). The NSCLC CSC slide was added with the lncRNA MBNL1-AS1 probe hybridization solution labeled by Digoxigenin, while the antagonistic lncRNA MBNL1-AS1 probe served as a negative control (NC). The slide was hybridized at 42°C for 16 h and immersed in 2 × SSC, followed by subsequent immersion in 70% ethanol for 3 min and staining with 4′6-diamidino-2-phenylindole (DAPI) for 5-10 min. The slide was photographed under a confocal laser-scanning microscope to document the observations. The experiment was conducted three times. All images were obtained under the Zeiss LSM880 NLO (2 + 1 with BIG) confocal microscope (Leica Microsystems, Mannheim, Germany).

Cell counting kit-8 (CCK-8)
Cells were seeded into 96-well plates at a density of 2 × 10 3 cells/well and incubated with 5% CO 2 at 37°C for cell adherence. Each well was added with 10 uL of the CCK-8 reagent, and then, the optical density (OD) value was measured at a wavelength of 450 nm at variable time points of 24, 48, 72, and 96 h. Then, the cell resistance was evaluated following procedures from existing research [21]. A total of 5 × 10 3 cells were seeded into the 96-well plates and incubated for 24 h after transfection. Next, the cells were further incubated in variable concentrations of gefitinib (SML1657, Sigma-Aldrich Chemical Company, St Louis, MO, USA) or cisplatin (P4394, dissolved in 0.15 M NaCl, Sigma-Aldrich Chemical Company, St Louis, MO, USA) for 48 h. The cells were incubated with the CCK-8 reagent at 37°C for 4 h. The OD value was measured at a wavelength of 570 nm using a microplate reader (Sunrise Microplate Reader, TECAN, Switzerland).

Clone formation assay in vitro
Transfected cells were seeded into 6-well plates at a density of 400 cells/well and incubated in DMEM supplemented with 10% FBS and 5% CO 2 at 37°C for 2 weeks. The clone cells were fixed in 95% alcohol and stained with 0.1% crystal violet for 10 min. Finally, the total number of clones (a clone was regarded as > 50 cells) were counted.

Transwell assay
For transwell assay, 24-well transwell plates (diameter of 8 μm, Corning Incorporated, Corning, NY, USA) were employed so as to measure the cell migratory and invasive ability. The 5 × 10 4 cells were seeded into an apical chamber covered with 200 mg/mL of Matrigel (1:8, Yepsen Company, Shanghai, China). Then, the medium containing the serum was added to the basolateral chamber. After incubation for 24 h, the cells invading into the basolateral chamber through Matrigel were fixed in 100% methanol for 10 min, stained with 0.1% crystal violet for 10 min, and then observed and counted under a microscope (× 200, Leica Inc., Wetzlar, Germany) with 5 random fields selected.

Scratch test
NSCLC CSCs were seeded into 6-well plates at a density of 2.5 × 10 4 cells/cm 2 , and the 10 μL scratch was prepared. Subsequently, the samples were washed with phosphate-buffered saline (PBS) two times and incubated in the DMEM containing 10% FBS in a 5% CO 2 incubator with saturated humidity at 37°C. Images were obtained at 0 h and 24 h under an inverted microscope and analyzed using ImageJ software. The distance between the cells on both sides of the scratch at each time point (μm) was recorded. The migration distance of cells was calculated as an equivalent obtained by subtracting the distance between the scratch edge at 0 h from the migration edge at 24 h, which indicated the migratory ability of cells. Three replicates were set for each group.
Tumorigenicity assay in nude mice About 1 × 10 7 cells were subcutaneously injected into the armpits of 5 specific-pathogen-free female BALB/c nude mice (aging 4-6 weeks and weighing 18-22 g, Hunan SJA Laboratory Animal Co., Ltd., Changsha, China) without thymus. During the 4th week after injection, the nude mice were euthanatized and the tumor was resected and fixed in 4% paraformaldehyde.

Statistical analysis
Statistical analysis was performed using the SPSS 21.0 software (IBM Corp. Armonk, NY, USA), and the measurement data were expressed as mean ± standard deviation (SD). All experiments were conducted three times to obtain mean values. Comparisons between NSCLC and adjacent normal tissues were assessed by the paired t test, and comparisons between two groups were analyzed using the unpaired t test. Comparisons among multiple groups were highlighted by one-way analysis of variance (ANOVA), and pairwise comparisons of mean values among multiple groups were analyzed using the Tukey post hoc test. Correlation of lncRNA MBNL1-AS1 with a prognosis of patients with NSCLC was analyzed by Kaplan-Meier curve, and the data at different time points were analyzed by repeated measurement ANOVA. A value of p < 0.05 was considered to be statistically significant.

Results
LncRNA MBNL1-AS1 may affect NSCLC through miR-301b-3p-targeted TGFBR2 The vital step in calculating our experimental results, the objective of understanding whether lncRNA MBNL1-AS1 may, or may not, affect NSCLC through regulating the miR-301b-3p-targeted TGFBR2, began with screening out of respective DEGs the NSCLC-related chip dataset GSE101929, which comprised of the lncRNA expression data, based on the GEO database. Subsiding the DEG screening, the heat maps of the 50 DEGs were drawn (Fig. 1a), which were representative of a significantly downregulated lncRNA MBNL1-AS1 in NSCLC tissues. Moreover, lncRNA MBNL1-AS1 expression was higher in the cytoplasm with respect to LncATLAS (Fig. 1b). LncRNA MBNL1-AS1 was speculated to affect the NSCLC disease by absorbing miRNA. Thus, we predicted the miRNAs that were regulated by lncRNA MBNL1-AS1 in the RAID database, which highlighted 33 miRNAs. On comparing various parameters among the 33 miRNAs and differential miRNAs in GSE102286, 4 crucial intersected miRNAs were chosen for subsequent experimentation (Fig. 1c), including hsa-miR-301b-3p, hsa-miR-30d-5p, hsa-miR-218-5, and hsa-miR-30a-5p. The expression profiles of the 4 miRNAs in GSE102286 are presented in Fig. 1d, which revealed that miR-301b was upregulated, whereas the other 3 miRNAs exhibited a contradictory trend, while lncRNA MBNL1-AS1 was aberrantly downregulated (Fig. 1a). Therefore, our team of researchers speculated the functionality of lncRNA MBNL1-AS1 as a molecular sponge of miR-301b-3p to influence NSCLC. Then, the target genes of miR-301b-3p were predicted in the mirDIP, miRDB, miRSearch, RNA22, and miRTarBase databases. After comparing the results with DEGs in GSE33532 in order to obtain the intersected gene, TGFBR2 was selected (Fig. 1e). TGFBR2 was downregulated in NSCLC based on GSE33532 (Fig. 1f). Overall, these results served as evidence supporting that lncRNA MBNL1-AS1 absorbing miR-301b-3p could potentially affect NSCLC by targeting TGFBR2.

LncRNA MBNL1-AS1 is downregulated in NSCLC tissues and cells
Following our speculation if lncRNA MBNL1-AS1 could serve as a potential target to affect NSCLC by regulating miR-301b-3p and TGFBR2, with a supplementary aim to calculate the expression rate of lncRNA MBNL1-AS1 in NSCLC tissues; foremost, RT-qPCR was conducted in order to examine the expression of lncRNA MBNL1-AS1 in NSCLC tissues and adjacent normal tissues from 56 patients, tissues with and without lymph node metastasis, NSCLC cell lines (A549, H1299, and SK-MES-1), and human normal lung epithelial cell line BEAS-2B. The results showed that lncRNA MBNL1-AS1 expression in the NSCLC tissues and tissues with lymph node metastasis were lower than that observed in the adjacent normal tissues and tissues without lymph node metastasis, respectively (p < 0.05). The patients were divided based on the value among high expression (≥ 0.437) and low expression (< 0.437) (p < 0.05) with a mean value (0.437) in NSCLC tissues as the dividing line in between them (Fig. 2a and Fig. 3e). CCK-8 showed that the cell viability of NSCLC CSCs decreased significantly after treatment with an overexpression of lncRNA MBNL1-AS1 (Fig. 3f), along with a significantly decreased 50% inhibition concentration (IC50) value of gefitinib and cisplatin in cells treated with oe-lncRNA MBNL1-AS1 (Fig. 3g), suggesting that lncRNA MBNL1-AS1 overexpression increased the sensitivity of A549 CSCs to anti-tumor drugs. Clone formation assay in vitro, scratch test, transwell assay, and sphere formation highlighted the evidently reduced proliferation, u A B C D Fig. 2 LncRNA MBNL1-AS1 is poorly expressed in NSCLC tissues and cells. a lncRNA MBNL1-AS1 expression in NSCLC tissues and adjacent normal tissues determined using RT-qPCR (n = 56). b lncRNA MBNL1-AS1 expression in tissues with and without lymph node metastasis determined using RT-qPCR. c lncRNA MBNL1-AS1 expression in NSCLC cell lines (A549, H1299, and SK-MES-1) and human normal lung epithelial cell line BEAS-2B determined using RT-qPCR. d Correlation of lncRNA MBNL1-AS1 with a prognosis of NSCLC patients analyzed by Kaplan-Meier curve; *p < 0.05 vs. the adjacent normal tissues, tissues without lymph node metastasis, or BEAS-2B cells. Measurement data were depicted as mean ± standard deviation; data in Fig. 2a was analyzed using a paired t test. Comparisons between two groups were analyzed using the unpaired t test, and comparisons among multiple groups were analyzed using one-way ANOVA with Tukey post hoc test used. The experiment was repeated three times. RT-qPCR, reverse transcription quantitative polymerase chain reaction; ANOVA, analysis of variance migration, invasion, and sphere formation of NSCLC CSCs in cells treated with oe-lncRNA MBNL1-AS1 ( Fig. 3h-k). In addition, western blot analysis revealed that the protein levels of endogenous stem genes of NSCLC CSCs (Oct4 and ABCG2) in cells treated with oe-lncRNA MBNL1-AS1 were lower than the protein levels observed in cells treated with oe-NC (p < 0.05, Fig. 3l). These results suggested that restored lncRNA MBNL1-AS1 inhibited CSC proliferation, migration, invasion, drug resistance, and sphere formation in NSCLC.
Restored lncRNA MBNL1-AS1 or depleted miR-301b-3p inhibits xenograft tumor formation in nude mice With findings supporting the regulation of the biological activity of A549-SP + via the TGF-β pathway due to lncRNA MBNL1-AS1 and miR-301b-3p via the TGF-β pathway, our last objective was to evaluate whether the restored lncRNA MBNL1-AS1 or depleted miR-301b-3p affected the xenograft tumor formation in nude mice. This experiment was commenced by the transportation of oe-lncRNA MBNL1-AS1, miR-301b-3p-inhibitor, and A B C Fig. 7 LncRNA MBNL1-AS1 and miR-301b-3p regulate the TGF-β pathway to modulate the biological activity of A549-SP + . a, b Levels of TGFBR2, smad2/3, and p-smad2/3 determined using western blot analysis. c Levels of Oct4 and ABCG2 in A549-SP + assessed using immunofluorescence staining; *p < 0.05 vs. A549 cells treated with NC. Measurement data were depicted as mean ± standard deviation. Comparisons among multiple groups were analyzed using one-way ANOVA with Tukey post hoc test used. The experiment was repeated three times oe-lncRNA MBNL1-AS1 + miR-301b-3p-mimic into A549-SP + following the instructions of Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Next, the transfected A549-SP + were subcutaneously injected into the armpits of female BALB/c nude mice, which indicated that miR-301b-3p depletion or lncRNA MBNL1-AS1 elevation reduced the mean tumor volume and weight, and upregulation of both lncRNA MBNL1-AS1 and miR-301b-3p contributed to normal tumor formation (Fig. 8a-c). Western blot analysis indicated that miR-301b-3p depletion or lncRNA MBNL1-AS1 restoration elevated the TGFBR2 protein level (Fig. 8d). With the aforementioned results serving as the theoretical basis, it was speculated that restored lncRNA MBNL1-AS1 or depleted miR-301b-3p inhibited the xenograft tumor formation in nude mice and increased the TGFBR2 protein level.

Discussion
In recent years, CSCs with higher tumorigenicity have prevailed as an efficacious approach to treat NSCLC, while the understanding of the characteristics of CSCs in NSCLC therapies is still a withstanding topic of discussion [6,8]. Accumulated evidences have indicated the involvement of lncRNAs in vital processes of NSCLC via functioning as mediators of CSCs [24]. The current study was performed with an aim to explain the suppressive role of lncRNA MBNL1-AS1 in NSCLC in an attempt to further elucidate the underlying mechanism associated with miR-301b-3p-targeted TGFBR2. The key observations of the current study were primarily centered around the protection of CSC cells from the degrading nature of proliferation, invasion, drug resistance, and sphere formation in NSCLC by activating TGFBR2 via a blockade of the miR-301b-3p upon restoration of lncRNA MBNL1-AS1 expression. Initially, the current study revealed that lncRNA MBNL1-AS1 is downregulated in NSCLC tissues and cells. The dysregulation of lncRNA expression has been acknowledged as an implication in the tumorigenesis of NSCLC [9]. Upon detection of the expression of several A B C D Fig. 8 Overexpressed lncRNA MBNL1-AS1 or downregulated miR-301b-3p suppresses xenograft tumor formation in nude mice. a Xenograft tumor formation in nude mice. b Tumor volume of nude mice. c Tumor weight of nude mice. d TGFBR2 protein level determined using western blot analysis. NC group: mice treated with NC. miR-301b-3p-inhibitor group: mice treated with miR-301b-3p inhibition. miR-301b-3p-mimic group: mice treated with miR-301b-3p overexpression. miR-NC + oe-lncRNA MBNL1-AS1 group: mice treated with lncRNA MBNL1-AS1 overexpression and miR-NC. miR-301b-3p-mimic + oe-lncRNA MBNL1-AS1 group: mice treated with lncRNA MBNL1-AS1 overexpression and miR-301b-3p overexpression; *p < 0.05 vs. nude mice treated with NC. Measurement data were depicted as mean ± standard deviation. Comparisons among multiple groups were analyzed using one-way ANOVA with Tukey post hoc test used. The experiment was repeated three times lncRNAs, the lncRNA HMlincRNA717 and lncRNA taurine-upregulated gene 1 (TUG1) were clinically proven to be poorly expressed in NSCLC and associated with an inferior prognosis [25,26]. Our findings were consistent with the findings of an existing research, which demonstrated the lncRNA MBNL1-AS1 to be poorly expressed in NSCLC [12]. Moreover, the current study disclosed the functionality of lncRNA MBNL1-AS1 as a molecular sponge of miR-301b-3p, due to its persistently high expression in NSCLC tissues and cells. A previous study demonstrated the liberating property of lncRNAs on various RNA transcripts by functioning as ceRNAs to implicitly regulate miRs, and lncRNAs participate in ceRNA networks and mRNA-miRNA-lncRNA crosstalk, which is implicated in human disease [27]. An existing research observed that lncRNA MBNL1 was bound with C allelic pre-miR-1307, which led to a reduced expression of miR-1307-3p, thus increasing the susceptibility to colorectal cancer [11]. An aberrant expression of miRNAs has been observed in the pathogenesis of several diseases, including various human cancers [28]. Shi et al. indicated an overexpression of miR-301a in NSCLC tissues [29]. Furthermore, a research observed an upregulated miR-301b expression in lung cancer tissues and cells [16]. The results of biological analysis and relevance analysis further verified TGFBR2 as a target gene of miR-301b-3p, which has a positive correlation with lncRNA MBNL1-AS1. A previous study speculated a consistently poor expression of TGFBR2 in NSCLC [30].
A study elucidated miR-301b to be a member of the pan-cancer oncogenic miRNA superfamily, which targets TGFBR2 [31]. The aforementioned evidences support the speculation that lncRNA MBNL1-AS1 is downregulated while miR-301b-3p is upregulated in NSCLC, with a positive correlation of lncRNA MBNL1-AS1 with TGFBR2 by functioning as a sponge of miR-301b-3p. In addition, the data from this current study supports the speculation that upregulated lncRNA MBNL1-AS1 or inhibited miR-301b-3p suppresses CSC proliferation, invasion, migration, drug resistance, sphere formation, and tumor formation in NSCLC by upregulating TGFBR2. An existing research confirmed a stimulated proliferation in combination with suppressed apoptosis of the skeletal muscle cells upon lncRNA MBNL1-AS1 upregulation [32]. Liu et al. also proved that lncRNA maternally expressed gene 3 (MEG3) restoration led to highly stimulated drug sensitivity both in vitro and in vivo, with emphasis on the inhibitory impact of elevated MEG3 on drug resistance [33]. A research demonstrated Fig. 9 Potential mechanism of lncRNA MBNL1-AS1 involved in the biological activity of NSCLC CSCs via regulation of TGFBR2. In NSCLC CSCs, lncRNA MBNL1-AS1 sponged to miR-301b-3p that negatively regulated TGFBR2, and downregulation of lncRNA MBNL1-AS1 weakened the adsorption capacity, increased miR-301b-3p expression, and reduced TGFBR2 level, thus promoting the proliferation, migration, invasion, drug resistance, and sphere formation yet suppressing the activation of TGF-β pathway. By contrast, lncRNA MBNL1-AS1 overexpression strengthened the adsorption capacity of miR-301b-3p, accompanied by elevated TGFBR2 expression, hence resulting in a decline of the proliferation, migration, invasion, drug resistance, and sphere formation yet an increase in activation of TGF-β pathway miR-301b to stimulate cell invasion and facilitate drug resistance in pancreatic carcinoma [34]. Moreover, silencing of miR-301a could suppress cell proliferation, migration, and invasion in NSCLC, which implied to the functionality of miR-301a as a potential therapeutic strategy in NSCLC treatment [35]. Previous studies have proven that excised miR-301b could consequently inhibit bladder cancer cell proliferation, migration, and invasion [36]. Li et al. have elicited that miR-9-5p facilitated the proliferation and invasion of NSCLC cells by inhibiting the TGFBR2 expression [37], which compelled to the involvement of TGFBR2 silencing in the mechanism of miR-9-5p in regard to the proliferation and invasion of NSCLC cells. A previous study observed and credited the suppressive role of TGFBR2 in the carcinogenesis in NSCLC [17]. The abovementioned findings as the basis for this study suggest that lncRNA MBNL1-AS1 elevation suppresses the proliferation, invasion, drug resistance, and sphere formation of CSCs in NSCLC by inhibiting miR-301b-3p via upregulation of TGFBR2.

Conclusions
To conclude, the current study serves as evidence supporting that lncRNA MBNL1-AS1 restoration could decelerate the occurrence and progression of NSCLC, thereby highlighting the functionality of lncRNA MBNL1-AS1 restoration as a sponge of miR-301b-3p to suppress the proliferation, invasion, drug resistance, and sphere formation of CSC cells in NSCLC via upregulation of TGFBR2 (Fig. 9). Thus, the lncRNA MBNL1-AS1-miR-301b-3p network could facilitate as a novel aspect in NSCLC treatment. This study may serve as a potential insight for developing efficacious therapeutic treatment strategies for NSCLC treatment.