Expression Analysis of lncRNAs in Refractory and Non-Refractory Epileptic Patients

Abstract

Long non-coding RNAs (lncRNAs) have been demonstrated to be involved in the pathogenesis of neuropsychiatric disorders such as epilepsy. In the current study, we evaluated expression of eight lncRNAs in 80 epileptic patients (40 refractory and 40 non-refractory ones) and 40 normal individual using quantitative real-time PCR. Bayesian regression model showed significant higher expression of UCA1 in both refractory and non-refractory groups compared with controls (posterior beta of relative expression (RE) = 2.03, P value = 0.003, and posterior beta of RE = 4.05, P value < 0.0001, respectively). Besides, expression of UCA1 was higher in non-refractory patients compared with refractory ones (posterior beta of RE = 2.008, P value = 0.019). When repeating statistical analyses in a gender-based manner, differences in expression of UCA1 were significant in all subgroup analyses except for male non-refractory vs. refractory subgroups analysis. Expression levels of NKILA and ANRIL were higher in both refractory and non-refractory groups compared with controls (posterior beta of RE = 1.565, P value = 0.018, and posterior beta of RE = 1.902, P value = 0.006 for NKILA; posterior beta of RE = 1.304, P value < 0.0001, and posterior beta of RE = 1.603, P value = 0.019 for ANRIL, respectively). However, expression levels of these two lncRNAs were not different between refractory and non-refractory groups. Gender-based analysis for these two lncRNAs revealed similar results except for lack of difference in ANRIL expression between male refractory group and controls. Expression of THRIL was significantly lower in both refractory and non-refractory groups compared with controls (posterior beta of RE = - 0.842, P value = 0.044 and posterior beta of RE = - 1.969, P value < 0.0001, respectively). Furthermore, expression of this lncRNA was lower in non-refractory patients compared with refractory ones (posterior beta of RE = - 1.129, P value = 0.002). However, no significant difference was detected between non-refractory and refractory patients either in males or females. The interactions between gender and relative expressions of PACER, DILC, and MALAT1 were significant, so the results were assessed in gender-based manner. In females, expression of DILC was higher in non-refractory patients compared with refractory ones (posterior beta of RE = 0.959, P value = 0.044). Expression of MALAT1 was lower in female non-refractory patients compared with controls and in female non-refractory patients compared with refractory ones (posterior beta of RE = - 1.35, P value = 0.002, and posterior beta of RE = - 0.942, P value = 0.045, respectively). Finally, expression of PACER was higher in refractory patients vs. controls and non-refractory patients vs. controls in both male and female subgroups. However, comparison between non-refractory and refractory patients revealed significant results only among females. Expression of none of the assessed lncRNAs was correlated with age of study participants. There were robust correlations between expression levels of lncRNAs. The most robust correlations were detected between UCA1 and PACER (r = 0.84, P < 0.0001) and between UCA1 and ANRIL (r = 0.75, P < 0.0001). Taken together, our study demonstrated dysregulation of lncRNAs in peripheral blood of epileptic patients and potentiated them as biomarkers for this neurologic condition.