An allele of rs619586 polymorphism in MALAT1 alters the invasiveness of meningioma via modulating the expression of collagen type V alpha (COL5A1)

Abstract The rs619586 polymorphism has been shown to alter the expression of MALAT1, which act as a competing endogenous RNA (ceRNA) against miR‐145. And miR‐145 was found to target COL5A1, the interaction between which was shown to be involved in the pathogenesis of invasive meningioma. In this study, we aimed to explore the effect of rs619586 polymorphism and its underlying molecular mechanism in invasive meningioma. Real‐time PCR and Western Blot analysis were used to study the differentiated expression of miR‐145, MALAT1 (metastasis‐associated lung adenocarcinoma transcript 1) and COL5A1 (collagen alpha‐1(V) chain) in tumour/serum samples genotyped as rs619586 AA, AG and GG. Computational analysis and luciferase reporter assay were also conducted to identify the regulatory relationship between miR‐145 and MALAT1/COL5A1. Meanwhile, expression of miR‐145 and COL5A1 in different cell treatment groups was measured to validate the results obtained from earlier experiments. As shown by the results and in tumour/serum samples genotyped as AA, AG and GG, the expression of both MALAT1 and COL5A1 was down‐regulated in a stepwise fashion, while the expression of miR‐145 was increased, suggesting a potential negative relationship between MALAT1/COL5A1 and miR‐145. Meanwhile, miR‐145 was shown to bind to MALAT1, while COL5A1 was identified as a virtual target gene of miR‐145. As a consequence, a MALAT1/miR‐145/COL5A1 molecular pathway was established based on the above results. In particular, with the presence of rs619586 A>G polymorphism, the expression of MALAT1 and COL5A1 was both reduced, leading to reduced invasiveness of meningioma.


| INTRODUC TI ON
Meningioma is a type of intracranial tumour with a high incidence in adulthood. 1 While meningioma usually has a good prognosis after a surgical operation, some meningioma patients are associated with a higher risk of recurrence and a poorer prognosis. 2 Penetration of meningioma into adjacent brain parenchyma is an important factor associated with its recurrence and leads to even poorer prognosis. 3 Although meningioma has benign characteristics in general, brain invasive meningiomas have been considered as grade II tumours by the World Health Organization (WHO). 2 Ranging in size from 200 nt to >100 kb, long noncoding RNAs (lncRNAs) are the most complex and the largest family of noncoding RNAs (ncRNAs). In addition, the expression of many lncRNAs is specific to cell types. The expression of a majority of lncRNAs is quite low, with some only expressed at one copy per cell. 4 However, lncRNAs are involved in various molecular functions, such as regulating protein activities, controlling transcriptional activities, playing organizational or structural roles, modifying RNA processing or acting as small RNA precursors. 5 As an lncRNA exerting oncogenic effects in many cancers, MALAT1 participates in a wide range of cellular processes. 6 For example, MALAT1 has been shown to regulate the expression of miR-145, a molecule related to cardiomyocyte injury. 7 Moreover, miR-145 expression in HL-1 cells was negatively regulated by MALAT1.
Furthermore, the expression of BCL2/adenovirus E1B 19 kD protein-interacting protein 3 (Bnip3), a target gene of miR-145, is reduced with the decreasing miR-145 expression. 8 It was demonstrated that the reduction in miR-145 expression during aggressive meningiomas is related to increased cancer invasiveness and motility, while only minor reduction in the expression of miR-145 targets has been observed, with COL5A1 most significantly down-regulated in miR-145-positive cells. Furthermore, in vitro studies also detected a higher level of COL5A1 mRNA in primary meningioma showing poor miR-145 expression. 9 It was found that the rs619586A> G single nucleotide polymorphisms (SNP) provided a protective effect on pulmonary arterial hypertension (PAH). Other functional bioassays also showed that the change from rs619586A to G in MALAT1 could promote its role as a competing endogenous RNA (ceRNA) of miR-214 and consequentially up-regulate the expression of X-box binding protein 1 (XBP1), thus suppressing the migration and proliferation of vascular endothelial cells. In addition, it was shown that the rs619586A> G SNP in MALAT1 was related to a decreased risk of PAH. Animal experiments also indicated that the alteration from rs619586 A to G affected the function of MALAT1 in accelerating cell migration and proliferation.
The rs619586 polymorphism has been shown to alter the expression of MALAT1. 10 In addition, MALAT1 was found to act as a competing endogenous RNA (ceRNA) against miR-145, while COL5A1 was found to be a target of miR-145. In fact, the deregulation of both miR-145 and COL5A1 was shown to be involved in the pathogenesis of invasive meningioma. 9,11 In this study, we studied the effect of rs619586 on the signalling pathway of MALAT1/miR-145/COL5A1 and investigated the association between rs619586 polymorphism and the invasiveness of meningioma.

| Human subjects and sample collection
A total of 829 patients diagnosed with meningioma by histological examination and treated by radical resection at our institute were enrolled in this study. These patients were divided to two groups, that is, an invasive meningioma group (N = 427) and a non-invasive meningioma group (N = 402) based on the invasiveness of their malignancy. Invasiveness of meningioma is defined as those meningioma that is invasive into the neighbouring brain parenchyma and evaluated by imaging tests. The serum samples were collected from all patients and were immediately frozen in liquid nitrogen at −80°C after collection. In addition, among all these 427 patients enrolled at the beginning of our study, we further selected 74 invasive meningioma patients according to the inclusion criteria listed as below: 1. no history of other malignant tumour; 2. primary meningioma; 3. no history of chemotherapy, radiotherapy or other treatment before the operation described in this study; 4. all sample sections examined by at least 2 pathologists were consistent with the diagnostic criteria of meningioma; 5. with complete clinicopathological and follow-up data. Accordingly, the rs619586 polymorphism in 74 patients was genotyped and the patients were divided to the following groups according to their rs619586 polymorphism genotypes: AA (N = 42), AG (N = 25) and GG (N = 7). We collected surgically resected meningioma and serum samples from these patients for subsequent analyses. This study was approved by the Institutional Review Board of our institute, and written informed consent was obtained from all patients.

| Genotyping by Taqman assay
Genotyping of rs619586 polymorphism was conducted using a TaqMan assay (Applied Biosystems, Foster City, CA). In brief, the reactions between the TaqMan probes and PCR primers were carried out in a 96-well plate on a thermal cycler. Subsequently, the intensity of fluorescence was detected on an ABI 7500 qPCR machine (Applied Biosystems, Foster City, CA) and analysed using the manufacturer's software.

| RNA isolation and real-time PCR
Total RNA was extracted from tissues using a Trizol Kit

| Cell culture and transfection
KNS-89 and SNB-19 cells were purchased from Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences. The cells were cultured in a DMEM culture medium (Invitrogen) supplemented with 10% foetal bovine serum (FBS), 100 U/mL penicillin and 100 mg/mL streptomycin. The cells were incubated in an incubator under saturated humidity, 5% CO 2 and 37°C. The medium was replaced every 1 or 2 days according to the status of cell growth.
When the cell confluence reached 70%, the medium was replaced by a serum-free medium and the cells were seeded in a 6-well plate at 24 hours before transfection. When the cells reached 50% confluence, they were transfected with MALAT1, miR-145 mimics, anti-miR-145 or a miR-145 negative control using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol. After transfection, the cells were cultured at 37°C with 5% CO 2 for 18 hours. The medium was then replaced with a complete medium, and the cells were collected 48 hours after transfection. Each measurement was repeated three times.

| Vector construction and mutagenesis
The putative miR-145 binding site within the 3′-UTR of MALAT1 was identified using TargetScan (targetscan.org). Subsequently, the 3′-UTR of the MALAT1 gene was amplified by PCR, followed Similarly, to study the effect of miR-145 silencing, an anti-miR-145 sequence was cloned into pcDNA Luciferase Reporter Vector to generate pcDNA-anti-miR-145.

| Statistical analysis
Statistical analysis was carried out using the SPSS 21.0 software (IBM). Data were presented as mean ± standard deviations.
Differences between two groups were compared by t test, while

| Rs619586 polymorphism is associated with the risk of meningioma
At the beginning of this study, we enrolled a total of 829 patients diagnosed with invasive meningioma (N = 427) or non-invasive meningioma (N = 402) in this study, the demographic data of which were presented in Table 1. The genotype of rs619586 polymorphism was utilized as the basis to divide these meningioma patients for genotype and allele analysis, as presented in Table 2. In Table 2, we showed that heterozygote genotype (AG) is borderline associated with reduced risk of invasive meningioma, and the risk of invasive meningioma GG genotype is significantly lower compared with AA group. Meanwhile, the risk in AG+GG group was lower than the reference group (AA) for invasive meningioma patients. In addition, we also showed that G allele is significantly associated with the reduced risk of invasive meningioma.

| Possible negative relationship between MALAT1 and miR-145 in tumour/serum samples of meningioma
Serum and tumour samples were collected from 74 invasive meningioma patients selected from 829 meningioma patients according to our inclusion criteria. Real-time PCR was then conducted to detect the relative expression of MALAT1 and miR-145 in these samples.
As shown in Figure 1A

| Differentiated expression of COL5A1
The expression of COL5A1 mRNA and protein was compared among AA, AG and GG groups using real-time PCR and Western Blot assays.
As shown in Figure 2A,B, the relative expression of both COL5A1 mRNA and protein was the lowest in the samples genotyped as GG and the highest in the samples genotyped as AA, suggesting a negative correlation between COL5A1 and miR-145.

| MiR-145 was able to bind to MALAT1
Computational analysis and luciferase assay were conducted to further explore the regulatory relationship between miR-145 and MALAT1. As shown in Figure 3A, miR-145 was able to bind to MALAT1 on its 3′UTR. Furthermore, the relative luciferase activity in KNS-89 and SNB-19 cells co-transfected with the wild-type MALAT1 and miR-145 mimics was apparently reduced, while the relative luciferase activity in KNS-89 and SNB-19 cells co-transfected with the mutant MALAT1 and miR-145 mimics showed no significant difference (Figure 3), suggesting that miR-145 acted as a direct inhibitor of MALAT1 by binding to its 3′UTR.

| COL5A1 was a target gene of miR-145
Computational analysis of miR-145 and COL5A1 mRNA has identified COL5A1 mRNA as a potential target of miR-145. In addition,

| D ISCUSS I ON
MALAT1 is an lncRNA with a length of >8000 nt and is located in chromosome 11q13. 12 MALAT1 is highly conserved in mammals, and previous studies have shown that played a critical role in the tumorigenesis of many cancers. In addition, MALAT1 was shown to participate in a wide range of biological processes, including cell migration, cell proliferation and cell apoptosis. 12 and metastatic cells in transition cell carcinoma, 23 and up-regulated COL5A1 was also identified to contribute to lung adenocarcinoma metastasis by promoting cell invasion and proliferation while suppressing cell apoptosis. 24 Also, in primary ovarian tumours, the highly expressed COL5A1 was found to be associated with poor prognosis and high risk of metastasis. 25 Since the SNPs in protein coding genes can affect gene expression and function, they have been implicated in complex diseases. 26,27 Similarly, SNPs in functional lncRNAs might also alter susceptibility to human diseases. In this study, we found that the rs619586A>G SNP significantly altered PAH risk. In this study, the relative mRNA expression of MALAT1 and miR-145 was measured in tumour/serum samples genotyped as AA, AG and GG. As the results indicated, the level of MALAT1 decreased in a stepwise fashion with the increasing level of miR-145 in tumour/serum samples genotyped as AA, AG and GG, respectively. Therefore, the differentiated expression of MALAT1 and miR-145 indicated a negative correlation between MALAT1 and miR-145. Furthermore, the expression of COL5A1 mRNA and protein also decreased in a stepwise fashion with the increasing level of miR-145 in tumour/serum samples genotyped as AA, AG and GG, respectively, suggesting a negative correlation between COL5A1 and miR-145. All above results indicated that the rs619586A>G SNP in the MALAT1 gene could reduce its expression, thus impairing the effect of miR-145 on COL5A1. Consistent with previous findings, we observed that the meningioma cells carrying the rs619586G genotype showed a higher expression of COL5A1 than those carrying the AA genotype. Therefore, based on these results, we have been suggested that rs619586A>G SNP conferred protection against meningioma invasion by inhibiting the activation of genes downstream of COL5A1. 28

| CON CLUS ION
In this study, the MALAT1/miR-145/COL5A1 molecular pathway was established. In particular, with the presence of rs619586 A>G polymorphism, the expression of MALAT1 and COL5A1 was both reduced, leading to reduced invasiveness of meningioma.

CO N FLI C T O F I NTE R E S T
None.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.