Sevoflurane inhibits progression of glioma via regulating the HMMR antisense RNA 1/microRNA-7/cyclin dependent kinase 4 axis

ABSTRACT Sevoflurane (Sev) is a volatile anesthetic that can inhibit tumor malignancy. Glioma is a main brain problem, but the mechanism of Sev in glioma progression is largely unclear. This study aims to explore a potential regulatory network of long noncoding RNA (lncRNA)/microRNA (miRNA)/mRNA associated with the function of Sev in glioma progression. LncRNA HMMR antisense RNA 1 (HMMR-AS1), miR-7 and cyclin-dependent kinase 4 (CDK4) abundances were examined via quantitative reverse transcription polymerase chain reaction and western blot. Cell viability, invasion, and colony formation ability were analyzed via cell counting kit-8, transwell analysis, and colony formation. The target association was analyzed via dual-luciferase reporter analysis and RNA pull-down. The in vivo function of Sev was investigated by xenograft model. HMMR-AS1 abundance was increased in glioma tissues and cells, and reduced via Sev. Sev constrained cell viability, invasion, and colony formation ability via decreasing HMMR-AS1 in glioma cells. miR-7 expression was decreased in glioma, and was targeted via HMMR-AS1. HMMR-AS1 silence restrained cell viability, invasion, and colony formation ability by up-regulating miR-7 in glioma cells. Sev increases miR-7 abundance via decreasing HMMR-AS1. CDK4 was targeted via miR-7, and highly expressed in glioma. miR-7 overexpression inhibited cell viability, invasion, and colony formation ability via reducing CDK4 in glioma cells. CDK4 expression was reduced by Sev via HMMR-AS1/miR-7 axis. Sev suppressed cell growth in glioma by regulating HMMR-AS1. Sev represses glioma cell progression by regulating HMMR-AS1/miR-7/CDK4 axis.


Introduction
Glioma is the most common central nervous system malignancy with high recurrence and poor prognosis [1]. The use of anesthetic gases provides potential strategies for glioma treatment [2]. Sevoflurane (Sev) is a widely used anesthetic gas by interacting with γ-aminobutyric acid [3]. Multiple reports suggest Sev has an anti-glioma activity by decreasing cell proliferation, migration and invasion [4][5][6]. However, the mechanism underlying Sev in glioma treatment remains poorly understood.
The differential expressions of miRNAs (~22 nucleotides) are related to brain tumors, which are associated with glioma progression and treatment [15,16]. Previous studies report miR-7 can repress cell growth, migration, and invasion in glioma via regulating insulin-like growth factor 1 receptor and trefoil Factor 3 [17,18]. But it is uncertain whether miR-7 is interacted with HMMR-AS1. The cyclin-dependent kinases (CDKs) are important biomarkers for tumor treatment [19]. CDK4 is a key member in CDKs, which is relevant to human tumor progression [20]. Moreover, CDK4 contributes to glioma cell growth, which is mediated via miR-129 [21]. However, whether CDK4 is respond to Sev, and whether it is mediated by HMMR-AS1 and miR-7 are unknown.
The novelty of the present work is to explore new mechanism targeted by Sev in glioma progression. Here we hypothesized that Sev might target the HMMR-AS1/miR-7/CDK4 axis to regulate glioma progression. The purposes of our research were to explore the role of Sev in glioma cell viability, invasion, and colony formation ability, and analyze the interaction between Sev and the HMMR-AS1/miR-7/CDK4 axis in glioma cells. This study may provide a new insight into the effect of Sev on glioma treatment.

Bioinformatics analysis
The top 10 genes including CDK4 in glioma was explored according to Gene Expression Profiling Interactive Analysis (GEPIA) database (http:// gepia.cancer-pku.cn/) [22], and related information was displayed in Table 1.The targets of HMMR-AS1 were searched via LncBase V.2 (http://carolina.imis.athena-innovation.gr/diana_ tools/web/index.php?r= lncbasev2/index) [23]. The targets of miR-7 were predicted by microT-CDS (http://diana.imis.athena-innovation.gr/ D i a n a T o o l s / i n d e x . p h p ? r = m i c r o T _ C D S /index) [24].

Patient tissues
The glioma tissues were collected from 37 glioblastoma patients (17 cases at WHO grades I and II, and 20 cases at WHO grade III) who did not receive any other treatment prior to surgery from The Third Affiliated Hospital of Nanchang University. The normal brain tissues (n = 10) were collected from healthy donors who died in traffic accidents. The written informed consents were obtained from every patient. This study was authorized via the Ethics Committee of our university.

Quantitative reverse transcription polymerase chain reaction (qRT-PCR)
RNA was isolated via Trizol (Beyotime, Shanghai, China) according to the protocols in a previous report [25], and reversely transcribed to complementary DNA (cDNA) with the specific reverse transcription kit (Thermo Fisher Scientific). The cDNA together with SYBR Green (Thermo Fisher Scientific) and primers was used for qRT-PCR. The specific primers were synthesized via Sangon (Shanghai, China) and displayed in Table 3. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or U6 functioned as reference gene. The relative RNA level was calculated according to2 −ΔΔCt method [26].

Colony formation analysis
Colony formation assay was conducted as previously reported [29]. After the transfection or Sev exposure, 1 × 10 3 LN229 and T98 cells were inoculated into 6-well plates. Following incubation for 2 weeks, clones were fixed with 4% paraformaldehyde (Beyotime) and stained via 0.1% crystal violet (Solarbio, Beijing, China). The images of colony formation were photographed. The colony formation ratio was normalized to the control group× 100%.

Transwell analysis
Transwell invasion assay was performed according to a previous report with some modifications [30]. Cell invasion was analyzed with 24-well transwell chambers (Corning, Corning, NY, USA) precoated via Matrigel (Solarbio). After the transfection or Sev exposure, 3 × 10 5 LN229 and T98 cells in DMEM without serum were placed into upper chambers. The low chambers were added with 600 μL DMEM plus 10% serum. Following culture for 8 h, the invasive cells were stained by 0.1% crystal violet. The cells were imaged under 100 × magnification microscope (Olympus, Tokyo, Japan), and counted by Image J software (NIH, Bethesda, MD, USA).

Dual-luciferase reporter and RNA pull-down analyses
The dual-luciferase reporter and RNA pull-down assays were performed according to a previous report [30]. The wild-type sequence of HMMR-AS1 or CDK4 3 untranslated region (UTR) with miR-7 binding sequence was cloned into pGL3-Basic vector (Promega, Madison, WI, USA), generating HMMR-AS1-WT and CDK4-WT. The mutant-type luciferase reporter vectors HMMR-AS1-MUT and CDK4-MUT were formed via mutating the binding sites of miR-7. LN229 and T98 cells were transfected with these luciferase reporter vectors and miR-7 mimic or miR-con. The luciferase activity was detected via a dualluciferase assay kit (Promega) after 24 of posttransfection. RNA pull-down analysis was performed with Magnetic RNA-Protein Pull-Down kit (Thermo Fisher Scientific). The biotin labeled Bio-HMMR-AS1 probe and Bio-miR-7 were constructed via Viagene (Changzhou, China).1 × 10 7 LN229 and T98 cells were lysed and incubated with streptavidin magnetic beads overnight. RNA on the beads was isolated, and HMMR-AS1, the relative miRNAs and CDK4 mRNA levels were examined by qRT-PCR.

Xenograft experiment
The xenograft experiments were performed according to a previous report [31]. Four-weekold male BALB/c nude mice were purchased from Vital River (Beijing, China). LN229 cells stably transfected with HMMR-AS1 overexpression vector or pcDNA were exposed to 5.1% Sev or not, and then subcutaneously injected into mice. The mice were divided into control, Sev + pcDNA, or Sev + HMMR-AS1 group (n = 6). Tumor size was examined weekly, and calculated according to volume = 0.5 × length × width 2 . After 4 weeks, mice were euthanized using 5% isoflurane (Sigma, St. Louis, MO, USA). The tumors were dissected out and weighed, and then collected for measurement of HMMR-AS1, miR-7 and CDK4 levels.
The animal experiments were permitted via the Animal Ethics Committee of The Third Affiliated Hospital of Nanchang University.

Statistical analysis
The data from three independent experiments were displayed as mean ± standard deviation (SD). The linear correlation of gene level was tested by Pearson correlation analysis. The statistical analysis was conducted using GraphPad Prism 8 (GraphPad, La Jolla, CA, USA). The difference for two groups was detected by Student's t-test, and that for multiple groups was analyzed via analysis of variance (ANOVA) followed via Tukey post hoc test as appropriate. P < 0.05 indicated the significance.

HMMR-AS1 abundance is enhanced in glioma
Sev has a suppressive role in glioma development. The purpose of this work is to explore what role Sev plays and how it works in glioma. It was hypothesized that Sev could target the HMMR-AS1/miR-7/CDK4 axis to inhibit glioma progression. To explore whether HMMR-AS1 was related to glioma progression, HMMR-AS1 expression change was examined in glioma. HMMR-AS1 abundance was significantly higher in glioma tissues and serum samples (n = 37) than in normal samples (n = 10) (Figure 1a, supplementary  Figure S1). And HMMR-AS1 level was increased in patients at advanced stage (WHO grade III; n = 17) compared with those at early stage (WHO grades I and II; n = 20) (Figure 1b). Furthermore, patients were divided into low (n = 18) and high (n = 19) HMMR-AS1 level group according to the median value of HMMR-AS1 level in glioma patients. Patients with high HMMR-AS1 level had lower overall survival (P = 0.031) (Figure 1c). Moreover, HMMR-AS1 abundance was measured in glioma cell lines (LN229, T98 and A172) and control NHA cells. Results showed that HMMR-AS1 abundance was increased more than 2.8-fold in glioma cells   (Figure 1d). Taken together, increased HMMR-AS1 might be associated with glioma progression. LN229 and T98 cells with highest expression of HMMR-AS1 were selected for further experiments.

Sev inhibits cell viability, invasion, and colony formation by regulating HMMR-AS1 in glioma cells
To analyze whether HMMR-AS1 was associated with the function of Sev in glioma, the influence of Sev on glioma cell viability and HMMR-AS1 expression was assessed. Exposure to Sev significantly decreased viability of LN229 and T98 cells and HMMR-AS1 expression in a dose-dependent pattern (Figure 2(a,B)). The 5.1% Sev was used for further experiments. Furthermore, HMMR-AS1 abundance was up-regulated via transfection of HMMR-AS1 overexpression vector in the two cell lines under Sev treatment (Figure 2c). Additionally, HMMR-AS1 up-regulation attenuated Sev-induced viability inhibition in LN229 and T98 cells (Figure 2d). Moreover, Sev evidently suppressed invasion of LN229 and T98 cells, which was mitigated via HMMR-AS1 overexpression (Figure 2e). In addition, Sev significantly restrained colony formation ability of the two cell lines, and this effect was relieved by HMMR-AS1 restoration (figure 2f). These results suggested that Sev suppressed glioma progression by decreasing HMMR-AS1.

Sev up-regulates miR-7 expression via decreasing HMMR-AS1 in glioma cells
To study whether Sev could regulate miR-7 expression via HMMR-AS1, miR-7 expression was detected in LN229 and T98 cells with overexpression of HMMR-AS1 under Sev treatment. As exhibited in Figure 5(a,B), miR-7 abundance was significantly increased by Sev treatment, and this effect was mitigated via HMMR-AS1 overexpression. These data indicated Sev could increase miR-7 level by regulating HMMR-AS1.

CDK4 is targeted by miR-7 in glioma cells
To further underlie the regulatory network involved in HMMR-AS1/miR-7 axis in glioma, the potential downstream genes were searched.
The top 10 up-regulated genes in glioma tissues were predicted via GEPIA, and the related expression information is shown in Table 1. CDK4 was the mostly increased gene in glioma. microT-CDS analyzed and displayed there was binding site between miR-7 and CDK4 (Figure 6a). To validate their interaction, dual-luciferase reporter and RNA pull-down analyses were performed. miR-7 upregulation led to clear reduction of luciferase activity in the CDK4-WT group, but it did not change the luciferase activity in the CDK4-MUT group ( Figure 6(b,C)). Furthermore, RNA pull-down analysis exhibited miR-7 effectively enriched   (g and h), colony formation ration (i and j) were determined in LN229 and T98 cells transfected with si-con, si-HMMR-AS1, si-HMMR-AS1 + in-miR-con or in-miR-7. *P < 0.05.
CDK4 in LN229 and T98 cells (Figure 6d). High expression of CDK4 in glioma tissues was predicted via GEPIA (Figure 6e). Moreover, CDK4 mRNA level was detected in glioma tissues and cells, and results showed that CDK4 abundance was significantly increased (Figure 6(f,G)). miR-7 abundance in glioma tissues was negatively associated with CDK4 mRNA level (Figure 6h). Additionally, CDK4 protein expression was negatively regulated by miR-7 in LN229 and T98 cells (Figure 6i). These results indicated miR-7 could target CDK4 in glioma cells.

miR-7 overexpression restrains cell viability, invasion, and colony formation via reducing CDK4 in glioma cells
To test the function of miR-7/CDK4 axis in glioma progression, LN229 and T98 cells were transfected with miR-con, miR-7 mimic, miR-7 mimic + pcDNA or CDK4 overexpression vector. miR-7 overexpression obviously decreased cell viability, and CDK4 up-regulation mitigated this effect (Figure 7(a,B)). Moreover, miR-7 overexpression evidently inhibited invasion of the two cell lines, and this effect was abolished by CDK4 upregulation (Figure 7(c,D)). Additionally, miR-7 overexpression significantly constrained colony formation ability of LN229 and T98 cells, which was attenuated via CDK4 addition (Figure 7(e,F)). These data showed miR-7 overexpression constrained glioma progression by targeting CDK4.

Sev reduces tumor growth by decreasing HMMR-AS1 in glioma
To explore the anti-tumor role of Sev in glioma in vivo, we performed the xenograft experiments using LN229 cells stably transfected with HMMR-AS1 overexpression vector or pcDNA after exposure to 5.1% Sev or not. Tumor volume and weight were obviously decreased in Sev + pcDNA group compared with control group, which were reversed in Sev + HMMR-AS1 group (Figure 9(a-C)). These results suggested Sev inhibited glioma cell growth in vivo by regulation of HMMR-AS1.

Discussion
Glioma accounts for ~30% of brain tumors with high mortality [32]. This tumor has proved challenging to cure [33]. The exposure to anesthetic  gases like Sev opens promising option for glioma treatment by inhibiting glioma progression [2]. Exploring the mechanism allows Sev in glioma treatment helps understand the pharmacological effect of Sev. This study is the first to identify that Sev can play the anti-glioma role via regulating HMMR-AS1/miR-7/CDK4 axis.
Multiple studies suggested Sev could inhibit glioma cell migration and invasion by interacting with miRNAs, like miR-34a-5p, miR-146b-5p and miR-637 [4,34,35]. Additionally, Sev could repress cell proliferation and increase apoptosis of glioma cells [5,31,36]. These reports indicated the antiglioma role of Sev by reducing cell proliferation, migration and invasion and increasing apoptosis. Similarly, we also found this function of Sev in glioma cells. However, it was opposite to the study of Lai et al., which showed exposure to 4% Sev for 4 h contributed to cell migration, invasion and colony formation in glioma by regulating cell surface protein 44 [37]. We hypothesized that changed doses and exposure times of Sev might induce different response in glioma, and lower dose and shorter exposure time of Sev might not exhibit the anti-tumor role. Our study indicated the antiglioma function of Sev (5.1% for 6 h) in vitro.
LncRNAs are involved in the pathology of brain tumors, including glioma [38]. Moreover, lncRNA-mediated networks are responsible for understanding the activity of Sev [10], such as lncRNA metastasis-associated lung adenocarcinoma transcript 1 and small nucleolar RNA host gene 1 [39,40]. Here we were the first to confirm HMMR-AS1 was decreased via Sev in glioma cells.
Next, we wanted to explore a lncRNA/miRNA/ mRNA network mediated via HMMR-AS1 in glioma. Many brain miRNAs are dysregulated in response to Sev exposure [41]. Here we confirmed miR-7 was targeted by HMMR-AS1 and lowly expressed in glioma. Li et al. reported that miR-7 could repress proliferation, migration and invasion and trigger apoptosis of glioma cells through regulating neuro-oncological ventral antigen 2 [42]. Bhere et al. showed that miR-7 could enhance glioma cell apoptosis [43]. Moreover, miR-7 suppressed tumor angiogenesis and growth of glioma in a murine xenograft model [44]. Additionally, miR-7 is reported to inhibit growth of glioma cells and xenograft models by blocking the phosphoinositide-3-kinase (PI3K)/protein kinase B (ATK) and Raf/mitogen-activated protein kinase (MEK)/ extracellular signal-regulated kinase (ERK) pathways [45]. These reports suggested the antitumor function of miR-7 in glioma. In agreement with these reports, we also validated miR-7 inhibited glioma progression. Moreover, we found that Sev could regulate miR-7 via HMMR-AS1 to participate in glioma development.
We next identified CDK4 was targeted via miR-7 in glioma. Previous studies reported CDK4 could facilitate glioma cell proliferation and restrain apoptosis [21,46]. More importantly, CDK4 was reported to be regulated via Sev in glioma [5]. Similarly, we also found the oncogenic role of CDK4 in glioma by promoting colony formation. Moreover, multiple reports indicated that CDK4 could promote cell invasion in human tumors, such as non-small-cell lung cancer, gastric cancer and colon cancer [47][48][49]. However, there is no report in support of the role of CDK4 in glioma cell invasion. This study found that CDK4 might promote glioma cell invasion. Furthermore, we found that Sev could reduce CDK4 expression via regulating HMMR-AS1/miR-7 axis. In this way, Sev attenuated glioma progression in vitro. Additionally, we established the xenograft model, and further confirmed the anti-tumor role of Sev.
However, there are some limitations in this study. Previous reports indicated the serum lncRNAs might function as diagnostic and prognostic biomarkers for glioma patients [50,51]. Here we also found HMMR-AS1 expression was increased in serum of glioma patients, indicating the potential of HMMR-AS1 in diagnosis and prognosis of glioma patients. Hence, the clinical role of HMMR-AS1 would be explored in future. Moreover, only 37 patients were obtained in the current study. Hence, we need more samples to better investigate the clinical value of HMMR-AS1 to obtain more reliable conclusions. Additionally, microRNA, fungi and algae may have important roles in fighting cancer and COVID-19 [52][53][54]. Whether they could prevent glioma development remains further study in future.

Conclusion
Sev exhibits an anti-tumor role in glioma, possibly via regulating HMMR-AS1/miR-7/CDK4 axis. This study proposes a novel insight into the pharmacological function of Sev in glioma.