BACKGROUND Children with refractory epilepsy (RE) are associated with increased mortality rate, nonfatal injuries, disability, and diminished quality of life. Biomarkers for the early prediction of RE is still an unmet need. METHODS Eighteen children with RE and six age-matched unrelated controls were included in this study. Plasma samples were prefractionated by the optimized thermal treatment before proteomic analysis using 2DE-LC-MS/MS. Bioinformatic analysis was carried out using STRING protein network. Immunoassay of unprocessed plasma was applied to confirm changes of proteins of interest. P-value < 0.05 was considered statistically significant. RESULTS Proteomic analysis (n = 6 each group) revealed nine differentially expressed proteins, i.e., haptoglobin, S100A9, serpin B1, apolipoprotein A-I, apolipoprotein A-IV, apolipoprotein C-II, alpha-1-acid glycoprotein 1 and 2, and transthyretin. Western immunoblotting confirmed haptoglobin upregulation in the RE group. STRING protein network predicted the inflammatory cytokines, i.e., interferon gamma (IFN-γ), interleukin-1 beta (IL-1β) and tumor necrosis factor alpha (TNF-α), play roles in pathophysiology in RE patients. Cytokine immunoassay (n = 24, 18 RE vs. 6 controls) exhibited plasma IFN-γ was upregulated in RE patients as compared to the healthy individuals (median [IQR]; 2.9 [2.9, 4.9] vs. 1.32 [0.8, 1.5] pg/mL, p = 0.0013), and plasma IL-1β was significantly downregulated in patients (1.0 [0.2, 1.9] vs. 4.5 [1.9, 11.0] pg/mL, p = 0.01). TNF-α had no difference between groups. The results suggest that haptoglobin may be associated with oxidative brain damage, while IFN-γ and IL-1β may be involved with neuroinflammation. CONCLUSIONS Alterations in plasma haptoglobin, IFN-γ, and IL-1β were associated with RE patients. Future studies using a combination of these candidate biomarkers may help predict the intractability of epilepsy in pediatric populations.

中文翻译：

---

蛋白质组学分析显示血浆结合珠蛋白、γ-干扰素和白细胞介素-1β 是儿童难治性癫痫的潜在生物标志物

背景 患有难治性癫痫 (RE) 的儿童与死亡率增加、非致命伤害、残疾和生活质量下降有关。用于早期预测 RE 的生物标志物仍然是一个未满足的需求。方法 本研究包括 18 名 RE 儿童和 6 名年龄匹配的无关对照。在使用 2DE-LC-MS/MS 进行蛋白质组学分析之前，通过优化的热处理对血浆样品进行预分馏。使用STRING蛋白质网络进行生物信息学分析。应用未加工血浆的免疫测定来确认感兴趣的蛋白质的变化。P 值 < 0.05 被认为具有统计学意义。结果 蛋白质组学分析（每组 n = 6）揭示了 9 种差异表达的蛋白质，即触珠蛋白、S100A9、serpin B1、载脂蛋白 AI、载脂蛋白 A-IV、载脂蛋白 C-II、α-1-酸性糖蛋白 1 和 2，以及转甲状腺素蛋白。蛋白质免疫印迹证实了 RE 组中的触珠蛋白上调。STRING 蛋白网络预测炎症细胞因子，即干扰素 γ (IFN-γ)、白细胞介素 1 β (IL-1β) 和肿瘤坏死因子 α (TNF-α)，在 RE 患者的病理生理学中发挥作用。细胞因子免疫测定（n = 24, 18 RE 与 6 个对照）显示，与健康个体相比，RE 患者的血浆 IFN-γ 上调（中位数 [IQR]；2.9 [2.9, 4.9] 与 1.32 [0.8, 1.5] pg/mL，p = 0.0013），并且患者的血浆 IL-1β 显着下调（1.0 [0.2, 1.9] vs. 4.5 [1.9, 11.0] pg/mL，p = 0.01）。TNF-α 组间无差异。结果表明，触珠蛋白可能与氧化性脑损伤有关，而 IFN-γ 和 IL-1β 可能与神经炎症有关。结论 血浆结合珠蛋白、IFN-γ 和 IL-1β 的变化与 RE 患者相关。使用这些候选生物标志物组合的未来研究可能有助于预测儿科人群癫痫的难治性。