Long noncoding RNA GAS5 and miR-137 and two of their genetic polymorphisms contribute to acute ischaemic stroke risk in an Egyptian population

Objectives: To assess the serum expression levels of long non-coding ribonucleic acid growth arrest specific-5 and micro-ribonucleic acid-137, and the different genotypes of long non-coding ribonucleic acid growth arrest specific-5 rs2067079 (C>T) and micro-ribonucleic acid-137 rs1625579 (T>G) in acute ischaemic stroke patients. Methods: The case-control study was conducted at Cairo University, Cairo, Egypt, from January to August 2020, and comprised adult acute ischaemic stroke patients of either gender selected from the stroke unit of the Neurology Department at Kasr Alainy Hospital of Cairo University. Healthy individuals matched for age and gender were enrolled as controls. Quantitative real-time polymerase chain reaction was used to quantify serum expression levels of long non-coding ribonucleic acid growth arrest specific-5 and micro-ribonucleic acid-137, and to genotype long non-coding ribonucleic acid lncRNA growth arrest specific-5 rs2067079 and micro-ribonucleic acid-137 rs1625579 using TaqMan allelic discrimination. Data was analysed using SPSS 22. Results: Of the 100 subjects, 50(50%) were patients; 34(68%) males and 16(32%) females with mean age 60.4±10.0 years. The remaining 50(50%) were controls; 28(56%) males and 22(44%) females with mean age 56.9±12.2 years ( p >0.05). The patients had more smokers, more hypertensives and more diabetics than the controls ( p <0.05). Long non-coding ribonucleic acid growth arrest specific-5 expression levels were significantly increased, while micro-ribonucleic acid-137 expression levels were significantly reduced among the patients ( p <0.05). Acute ischaemic stroke risk was significantly higher in patients with growth arrest specific-5 rs2067079 (C>T) recessive model (homozygous minor TT genotype), while micro-ribonucleic acid-137 rs1625579 (T>G) was protective against acute ischaemic stroke in allelic (G minor allele), codominant (GT genotype), dominant (GT+GG), and over-dominant (GT genotype) models ( p <0.05). Conclusion: Long non-coding ribonucleic acid growth arrest specific-5 rs2067079 and micro-ribonucleic acid-137 rs1625579 may act as novel genetic markers of acute ischaemic stroke risk.


Introduction
Acute ischaemic stroke (AIS) contributes a significant proportion to global mortality and long-term disability. 1 It is believed to be a complex illness that results from both hereditary and environmental factors. 2 Identifying the genetic variants that predispose one to AIS would enable early screening of high-risk individuals and a better understanding of the disease's pathophysiology.
[5] During AIS, lncRNA growth arrest specific-5 (GAS5) has been linked to neuronal cell death. 6Also, lncRNA GAS5 neuronal expression was found to be increased in AIS experimental models in rats. 7As a result, lncRNA GAS5 is believed to contribute to the pathophysiology of AIS.
Micro-ribonucleic acid (miRNAs) are defined as short noncoding, single-stranded RNAs having length of around 22 nucleotides, regulating post-transcriptional gene expression.MiRNAs have been correlated with the pathophysiology of multiple human illnesses. 8e miRNA that is abundantly expressed in the brain is miR-137, and dysregulation in its expression has been linked to the development of AIS.During AIS and through inhibiting the Proto-oncogene c-Src (Src) -mediated mitogen-activated protein kinase (MAPK) signalling pathway, miRNA-137 has been shown to protect neurons from inflammation, oxidative stress and apoptosis. 9 addition, miR-137 has been proposed as a downstream target of lncRNA GAS5 in AIS, where the latter has been demonstrated to compete with miR-137 for the control of Notch1 signalling pathway. 10ngle nucleotide polymorphisms (SNPs) are the most common deoxyribonucleic acid (DNA) sequence variations in the human genome that can genotypically distinguish between individuals.In genetic association studies, SNPs are often employed as markers to evaluate their effect on the risk of development and pathogenesis of complex diseases. 11P rs2067079 in the lncRNA GAS5 gene has been linked to lncRNA GAS5 transcription and secondary structure stability, which suggest a possible role of this SNP in AIS lncRNA GAS5 expression dysregulation.12 Furthermore, the miR-137 gene's SNP rs1625579 has been demonstrated to influence miR-137 production levels.13 The current study was planned to assess the serum expression of the two non-coding RNAs (ncRNAs), miR-137 and lncRNA GAS5, and to genotype two of their SNPs, the lncRNA GAS5 rs2067079 and the miR-137 rs1625579, in AIS patients to evaluate their potential role in AIS risk and pathogenesis.

Patients and Methods
The case-control study was conducted at Cairo University, Cairo, Egypt, from January to August 2020.After approval from the institutional ethics review committee, the sample size was calculated using Epi-Info 3.01 14 with confidence level 95%, power of study 80% and based on previous findings 15 .The sample was raised from among adult AIS patients of either gender selected from the stroke unit of the Neurology Department at Kasr Alainy Hospital of Cairo University through non-probability convenient sampling technique.A computed tomography (CT) brain scan was used for AIS diagnosis.Pregnant or nursing women, cancer patients, those with concomitant infection, and those on long-term steroid treatment were excluded.Healthy individuals matched for age and gender without clinical indications or a family history of AIS were enrolled as controls from the same hospital.Written informed consent was obtained from all the participants.SNPs were selected based on the following criteria: global minor allele frequency (MAF) >0.1; SNPs previously reported to have a functional relationship to their gene products; and SNPs with previously reported correlation with neurological diseases.
Using the QIAamp DNA MiniKit (Qiagen, Valencia, CA, United States), according to the manufacturer's protocol, genomic DNA was extracted from whole ethylenediaminetetraacetic acid (EDTA) blood samples from each participant.The NanoDrop® (ND)-1000 spectrophotometer (NanoDrop Technologies, Inc. Wilmington, USA) was utilised to evaluate DNA quantitation and purity.
Genotyping was done using real-time polymerase chain reaction (RT-PCR) with TaqMan allelic discrimination assay (Applied Biosystems, USA) with a predesigned primer/probe sets for the 6 genotypes (Applied Biosystems, USA) as previously explained. 16DNA amplification was done in a total volume of 25μl containing 12.5µl TaqMan master mix, 1.25 primer/probe, 1µl DNA (~100ng), and 10.25 water by a Rotor Gene Q RT-PCR system (Qiagen, Valencia, CA, USA).Following a denaturation period of 10m at 95 o C, 45 cycles at 92 o C for 15s then at 60 o C for 90s for annealing and extension were performed and fluorescence was assessed at the end of every cycle and at the endpoint.
RT-PCR was used to quantify serum lncRNA GAS5 and miR-137 expression levels.First, total RNA was extracted from haemolysis-free serum utilising miRNeasy mini-kit (Qiagen, Valencia, CA, USA) complying with the manufacturer's guidelines.Then, by using the mi Script II RT kit (Qiagen, Valencia, CA, USA), reverse transcription of total RNA into complementary DNA was carried out in a final volume of 20µl RT reaction as per the manufacturer's protocols.The expression levels of lncRNA GAS5 and miR-137 were determined by RT-PCR using mi Script SYBR® Green PCR kit (Qiagen, Valencia, CA, USA) following the manufacturer's protocols.RT-PCR was performed using a 20μL final volume reaction mixture as previously explained. 17A Rotor Gene Q System was used to perform RT-PCR.The reaction conditions were 95°C for 15m then a total of 40 cycles of 95°C for 15m, 55°C for 30s and 70°C for 30s for miR-137, and 95°C for 15m then a total of 45 cycles of 95°C for 15m, and 60°C for 60s for lncRNA GAS5.
Melting curve analyses were done after the completion of the PCR cycles to confirm the precise production of the anticipated PCR product.Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as the internal reference of lncRNA GAS5 and the housekeeping miScript PCR control and miRNA small nucleolar RNA, C/D Box 68 (SNORD68) were used as the internal reference of miR-137.Gene expression relative to the internal control (2 −ΔCt ) was estimated, and then the fold change in lncRNA GAS5 or miR-137 expression was estimated using the equation 2 - ΔΔCt . 18ta was analysed using SPSS 22. Quantitative data was Egypt Suppl.: Climate Change and the need for One World One Health

S-185
J Pak Med Assoc (Suppl.4)     expressed as mean and standard deviation, or median and range, as appropriate.Categorical data was expressed as frequencies and percentages.Categorical data was compared using chi-square test.GAS5 and miR-137 data did not follow a normal distribution, so Mann-Whitney U tests or Kruskal-Wallis tests were used to compare them, as needed.Logistic regression analysis was done to identify the best predictors of AIS risk.Receiver operating characteristic (ROC) curve was generated with the area under curve (AUC) analysis to detect the best cut-off value of identification markers.P<0.05 was considered statistically significant.
The distribution of rs2067079 (C/T) and rs1625579 (G/T) genotypes in both the patients and the controls differed from in terms of Hardy-Weinberg equilibrium (HWE), but the difference was not significant (Table 4).
The MAF of both SNPs in the control group was >0.1 (G=0.23 for rs1625579 and T=0.26 for rs2067079).Regarding rs2067079, the major T and minor G allele frequencies non-significantly differed between the groups (p>0.05).The minor homozygous TT genotype was associated with a 16.5-fold higher risk of AIS in the recessive model (TT vs CC+CT; p=0.033) with adjustments for age, hypertension (HTN), diabetes mellitus (DM) and smoking.No significant change was found among cases and controls in the codominant, dominant, and over-dominant models (Table 5).
As regards rs1625579, the frequencies of major T and minor G alleles differed significantly between the cases and the controls, with a lower AIS risk in the minor G allele (G vs T; p=0.043).In the codominant model, the heterozygous GT  genotype significantly reduced AIS risk (GT vs TT; p=0.006), adjusted for age, HTN, DM and smoking.In addition, the dominant model showed a decreased AIS risk in the minor genotypes (GT+GG) compared to the major homozygous genotype (TT) (GT+TT vs TT; p=0.008) after adjustments for age, HTN, DM and smoking.In the over-dominant model, the GT heterozygous genotype was associated with a significant lower AIS risk (GT vs TT+GG; p=0.006) adjusted for age, HTN, DM and smoking (Table 6).
Univariate analysis for each predictor alone showed that serum miR-137 levels and rs1625579 GT+GG (dominant model) were significant negative predictors of AIS risk (p=0.041 and p=0.008, respectively).Serum GAS5 levels were significant positive predictors of AIS (p=0.003)(Table 7).
In multivariate analysis, serum miR-137 and lncRNA GAS5 levels were confirmed to be significant negative and positive predictors of AIS risk, respectively, (Table 8).

Discussion
Globally, AIS is considered a major cause of mortality and irreversible disability. 1 Despite extensive research to uncover the risk factors and pathophysiological mechanisms of AIS, the precise genetic components remain incompletely understood. 19covering the genetic predisposition to AIS may help to identify high-risk individuals and discover novel therapeutic targets, providing a better prognosis and promoting recovery after AIS. 20e ncRNAs are emerging as vital regulators of central nervous system (CNS) biological functions and pathological events, like apoptosis and inflammation. 21Therefore, the current study investigated two ncRNAs, lncRNA GAS5 and miR-137, and two of their genetic variants (rs2067079 for GAS5 and rs1625579 for miR-137) for a possible correlation with AIS.
LncRNA GAS5 is recognised to control cell survival negatively, causing a variety of cell types to arrest and/or undergo apoptosis. 22The current results showed significantly higher serum expression levels of lncRNA GAS5 in AIS cases than in the control group.
Similar results were reported by Wang et al. in 2019 as lncRNA GAS5 was highly expressed and contributed to apoptosis and inflammatory injury in mice who underwent middle cerebral artery occlusion (MCAO), and in cells treated with oxygen glucose deprivation (OGD) in vitro. 23spite the fact that the precise molecular pathways by which lncRNA GAS5 participates in the pathophysiology of AIS are still not fully understood, prior studies have suggested that lncRNA GAS5 may function in AIS as a competitive endogenous RNA (ceRNA) for certain miRNAs, like miRNA sponge.Zhou et al. in 2020 suggested that in hypoxia conditions, lncRNA GAS5 promotes neuronal apoptosis via sponging miR-221 which causes upregulation of p53 up-regulated modulator of apoptosis/ c-jun N-terminal kinase/ histone family member X (PUMA/JNK/H2AX) signalling, a pathway linked to brain ischaemic injury. 24In addition, lncRNA GAS5 served as a ceRNA for miR-9, increasing the forkhead box O (FOXO3) expression, a transcription factor linked to cerebral infarction and cell death in AIS. 23cently, miR-137 has been suggested to protect neurons in AIS.Zhang et al. in 2020 reported that miR-137 overexpression decreased apoptosis and alleviated inflammation in MCAO rats, which showed a lower infarct size and an enhanced neurological function score.They suggested that miR-137 improved AIS neuronal injury through blocking the Janus protein-tyrosine kinases/Signal transducer and activator of transcription 1 (JAK1/STAT1) pathway. 25Moreover, miR-137 may reduce AIS damage via inhibition of the Src/MAPK pathway which lowers oxidative stress and inflammation. 9 Some studies have suggested that miR-137 may act as a downstream target of lncRNA GAS5. 10, 26 In the present study, both lncRNA GAS5 and miR-137 serum expression levels significantly discriminated AIS cases from controls with high accuracy.Additionally, the study demonstrated that the serum lncRNA GAS5 and miR-137 levels may be used as positive and negative indicators, respectively, of AIS susceptibility.
Currently, much of the research is aimed at finding genetic variations that affect the incidence and pathogenesis of prevalent multifactorial polygenic disorders. 27SNPs are at the forefront of such studies because they are the most prevalent genetic variations in humans. 28cording to Guo et al. in 2017, SNP rs2067079 (C>T) in the lncRNA GAS5 gene may affect the stability as well as the secondary structure of lncRNA GAS5. 29In addition, Li et al. in 2020 demonstrated that the same SNP may be linked to a higher risk of ischaemic heart disease (IHD). 30Both AIS and IHD have many common risk factors and underlying pathophysiological events. 31Moreover, miR-137 SNP rs1625579 (T>G) has been found to influence the miR-137 S-188 expression in the brain. 13Together, these findings suggest that rs2067079 (C>T) and rs1625579 (T>G) may contribute to AIS risk and pathogenesis.
Interestingly, the current study revealed a significant correlation between rs2067079 (C>T) and a higher AIS risk in the recessive model (TT genotype).Moreover, there was a significant association between rs1625579 (T>G) and a lower AIS risk in allelic (G minor allele), codominant (GT genotype), dominant (GT+GG), and over-dominant (GT genotype) models.
To the best of the researchers' knowledge, no previous study has investigated GAS5 rs2067079 and miR-137 rs1625579 in AIS patients.Therefore, the current study is the first to introduce rs2067079 and rs1625579 as potential biomarkers for genetic susceptibility to AIS and as possible contributors to disease pathogenesis.
Other SNPs in GAS5 have been previously linked to AIS risk, like a correlation between rs145204276 del allele and increased AIS risk among a Chinese population. 32Similarly, rs2067079 has been reported to influence lncRNA GAS5 expression level in multiple sclerosis (MS) patients with rs2067079 TT genotype. 15However, the current study showed no similar effect of rs2067079 on lncRNA GAS5 expression which could be attributed to the relatively small sample size and the different clinical subtypes of AIS that the subjects had.
The current study also investigated whether any of the rs1625579 genotypes was associated with dysregulation in miR-137 expression in the serum of AIS patients.However, the results were statistically non-significant, which was in line with literature. 15In contrast, Guella et al. in 2013 suggested that rs1625579 can modulate miR-137 expression. 13e current study also has several limitations, like the subjects having different clinical AIS subtypes, which might render the statistical power insufficient.Besides, the molecular mechanisms by which rs2067079 and rs1625579 may affect AIS pathogenesis were not examined.Also, the study was limited to Egyptian population, and the findings need to be validated in other populations to warrant their generalisation.

Conclusions
The serum levels of lncRNA GAS5 and miR-137 could act as diagnostic biomarkers of AIS, and serum levels of both ncRNAs could positively and negatively predict AIS risk, respectively.The two polymorphisms, lncRNA GAS5 rs2067079 and miR-137 rs1625579, could act as new genetic markers of AIS risk.

Table - 5
: Genotype and allele frequencies of rs2067079 (C/T) in AIS patients and and controls.

Table - 6
: Genotype and allele frequencies of rs1625579 (G/T) in AIS patients and controls.