Expression of nuclear factor kappa B-associated long non-coding RNAs in drug-naïve and valproic acid-treated epilepsy patients
Abstract
Serum lncRNAs/cytokines were measured in controls, drug-naïve, and valproic acid (VPA)-treated patients. Serum UCA1 and ANRIL were elevated in untreated patients versus controls. Bayesian regression and ROC analyses linked UCA1 upregulation to epilepsy. VPA normalized UCA1 expression towards healthy levels. Background Epilepsy has a complex etiopathogenesis and frequently shows resistance to conventional antiseizure medications, underscoring the need for a deeper mechanistic understanding of the disease. Accumulating evidence implicates NF-κB–mediated neuroinflammation in epileptogenesis, and several long non-coding RNAs (lncRNAs) are known to regulate this pathway. However, the expression patterns of the NF-κB–associated lncRNAs UCA1 , ANRIL , and NKILA in epilepsy, and whether antiseizure medications can modulate them, remain largely unknown. This study therefore assessed these lncRNAs, together with the inflammatory cytokines IL-1β and IL-6, in drug-naïve and valproic acid (VPA)-treated epileptic patients compared with healthy controls. Methods In a prospective, multicenter study with 120 adults, participants were divided into three groups: drug-naïve patients, VPA-treated patients, and healthy volunteers. Early morning blood samples were collected to measure serum levels of the selected lncRNAs and cytokines. Correlations between molecular markers and seizure parameters were examined. Results Serum IL-1β, IL-6, UCA1 , and ANRIL were significantly higher in drug-naïve patients than in healthy controls (p < 0.001, p < 0.001, p = 0.003, and p = 0.014, respectively), whereas NKILA showed no significant difference (p = 0.484). Compared with drug-naïve patients, VPA-treated patients showed no significant difference in IL-1β (p = 0.056) or IL-6 (p = 1.00); in the age- and sex-adjusted Bayesian model, UCA1 was significantly lower in the VPA group compared to the drug-naive group (p = 0.015) and no longer differed from controls (p = 0.424), whereas ANRIL remained largely unchanged. Serum IL-1β had a significant positive correlation with IL-6 (ρ = 0.58, p < 0.001) and ANRIL (ρ = 0.20, p = 0.028), and in drug-naïve patients seizure severity significantly correlate positively with UCA1 (ρ = 0.41, p = 0.009). On ROC analysis, IL-1β and IL-6 best discriminated untreated patients from controls (AUC = 0.84 and 0.96, respectively). These findings indicate heightened systemic inflammation in untreated epilepsy and support IL-1β and IL-6 as potential candidate biomarkers, with UCA1 as a potentially treatment-responsive lncRNA. Conclusion Our results identify serum UCA1 as the most notably altered lncRNA, significantly elevated in untreated epilepsy and potentially responsive to valproic acid. While IL-1β and IL-6 reflected the accompanying inflammation. These findings warrant further validation as candidate biomarker.