Department of Neurology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China.
CNS Neurosci Ther. 2024 Jun;30(6):e14817. doi: 10.1111/cns.14817.
Proteome-wide Mendelian randomization studies have been increasingly utilized to identify potential drug targets for diseases. We aimed to identify potential therapeutic targets for migraine and its subtypes through the application of Mendelian randomization and co-localization analysis methods.
We utilized cis-protein quantitative trait loci data for 1378 plasma proteins available from two studies with 7213 individuals and 35,559 individuals, respectively. Summary data for migraine and its subtypes were obtained from a genetic study involving up to 1,339,303 individuals. Proteins that passed both the discovery and validation Mendelian randomization analysis, sensitivity analysis, heterogeneity test, and pleiotropy test, were associated with ≥2 outcomes, and received strong support from co-localization analysis (PP.H4.abf ≥0.80) and were classified as tier 1 proteins.
We identified three tier 1 proteins (LRP11, ITIH1, and ADGRF5), whose genes have not been previously identified as causal genes for migraine in genetic studies. LRP11 was significantly associated with the risk of any migraine (OR [odds ratio] = 0.968, 95% CI [confidence interval] = 0.955-0.981, p = 1.27 × 10) and significantly/suggestively associated with three migraine subtypes. ITIH1 was significantly associated with the risk of any migraine (OR = 1.044, 95% CI = 1.024-1.065, p = 1.08 × 10) and migraine with visual disturbances. ADGRF5 was significantly associated with the risk of any migraine (OR = 0.964, 95% CI = 0.946-0.982, p = 8.74 × 10) and suggestively associated with migraine with aura. The effects of LRP11 and ADGRF5 were further replicated using cerebrospinal fluid protein data. Apart from ADGRF5, there was no evidence of potential adverse consequences when modulating the plasma levels. We also identified another four proteins (PLCG1, ARHGAP25, CHGA, and MANBA) with no potential adverse consequences when modulating the plasma levels, and their genes were not reported by previous genetic studies.
We found compelling evidence for two proteins and suggestive evidence for four proteins that could be promising targets for migraine treatment without significant adverse consequences. The corresponding genes were not reported in previous genetic studies. Future studies are needed to confirm the causal role of these proteins and explore the underlying mechanisms.
全蛋白质组孟德尔随机化研究已越来越多地被用于鉴定疾病的潜在药物靶点。我们旨在通过应用孟德尔随机化和共定位分析方法，鉴定偏头痛及其亚型的潜在治疗靶点。
我们利用来自两项研究的 cis-蛋白质定量性状基因座数据，这两项研究分别涉及 7213 人和 35559 人，分析了 1378 种血浆蛋白。偏头痛及其亚型的汇总数据来自一项涉及多达 1339303 人的遗传研究。通过发现和验证孟德尔随机化分析、敏感性分析、异质性检验和多效性检验，对与≥2 个结果相关且得到共定位分析的有力支持（PP.H4.abf≥0.80）并被归类为 1 级蛋白的蛋白进行了鉴定。
我们鉴定出了三个 1 级蛋白（LRP11、ITIH1 和 ADGRF5），它们的基因在遗传研究中尚未被确定为偏头痛的因果基因。LRP11 与任何偏头痛的风险显著相关（OR[比值比] = 0.968，95%CI[置信区间] = 0.955-0.981，p = 1.27×10），并与三种偏头痛亚型显著/提示性相关。ITIH1 与任何偏头痛的风险显著相关（OR = 1.044，95%CI = 1.024-1.065，p = 1.08×10）和偏头痛伴视觉障碍。ADGRF5 与任何偏头痛的风险显著相关（OR = 0.964，95%CI = 0.946-0.982，p = 8.74×10），并提示与偏头痛伴先兆相关。使用脑脊液蛋白数据进一步复制了 LRP11 和 ADGRF5 的作用。除 ADGRF5 外，调节血浆水平时没有潜在不良后果的证据。我们还发现了另外四个蛋白（PLCG1、ARHGAP25、CHGA 和 MANBA），在调节血浆水平时没有潜在不良后果，而它们的基因在以前的遗传研究中没有报道。
我们发现了两个蛋白的有力证据和四个蛋白的提示性证据，这些蛋白可能是治疗偏头痛而没有明显不良后果的有希望的靶点。相应的基因在以前的遗传研究中没有报道。需要进一步的研究来确认这些蛋白的因果作用，并探索其潜在机制。
