Wei Zhengxiao, Liu Yingfen, Xiong Qingqing, Mei Xue, Li Jinghong, Wu Zhangjun
Department of Clinical Laboratory, Public Health Clinical Center of Chengdu, Chengdu, China.
Department of Science and Education Division, Public Health Clinical Center of Chengdu, Chengdu, China.
Front Med (Lausanne). 2024 Jul 2;11:1395526. doi: 10.3389/fmed.2024.1395526. eCollection 2024.
Blood metabolite abnormalities have revealed an association with cholestatic liver diseases (CLDs), while the underlying metabolic mechanisms have remained sluggish yet. Accordingly, the present evaluation aims to investigate the causal relationship between blood metabolites and the risk of two major CLDs, including primary biliary cholangitis (PBC) and primary sclerosing cholangitis (PSC).
Univariable and multivariable Mendelian randomization (MR) approaches were employed to uncover potential causal associations between blood metabolites and 2 CLDs, including PBS and PSC, through extracting instrumental variables (IVs) for metabolites from genome-wide association studies (GWAS) conducted on European individuals. The GWAS summary data of PBC or PSC were sourced from two distinct datasets. The initial analysis employed inverse variance weighted (IVW) and an array of sensitivity analyses, followed by replication and meta-analysis utilizing FinnGen consortium data. Finally, a multivariable MR analysis was carried out to ascertain the independent effects of each metabolite. Furthermore, the web-based tool MetaboAnalyst 5.0 was used to perform metabolic pathway examination.
A genetic causality between 15 metabolites and CLDs was recognized after preliminary analysis and false discovery rate (FDR) correction. Subsequently, 9 metabolites consistently represented an association through replication and meta-analysis. Additionally, the independent causal effects of 7 metabolites were corroborated by multivariable MR analysis. Specifically, the metabolites isovalerylcarnitine (odds ratio [OR] = 3.146, 95% confidence intervals [CI]: 1.471-6.726, = 0.003), valine (OR = 192.44, 95%CI: 4.949-7483.27, = 0.005), and mannose (OR = 0.184, 95%CI: 0.068-0.499, < 0.001) were found to have a causal relationship with the occurrence of PBC. Furthermore, erythrose (OR = 5.504, 95%CI: 1.801-16.821, = 0.003), 1-stearoylglycerophosphocholine (OR = 6.753, 95%CI: 2.621-17.399, = 7.64 × 10), X-11847 (OR = 0.478, 95%CI: 0.352-0.650, = 2.28 × 10), and X-12405 (OR = 3.765, 95%CI: 1.771-8.005, = 5.71 × 10) were independently associated with the occurrence of PSC. Furthermore, the analysis of metabolic pathways identified seven significant pathways in two CLDs.
The findings of the present study have unveiled robust causal relationships between 7 metabolites and 2 CLDs, thereby providing novel insights into the metabolic mechanisms and therapeutic strategies for these disorders.
血液代谢物异常已显示出与胆汁淤积性肝病（CLD）相关，但其潜在的代谢机制仍不明朗。因此，本评估旨在研究血液代谢物与两种主要胆汁淤积性肝病（包括原发性胆汁性胆管炎（PBC）和原发性硬化性胆管炎（PSC））风险之间的因果关系。
采用单变量和多变量孟德尔随机化（MR）方法，通过从针对欧洲个体进行的全基因组关联研究（GWAS）中提取代谢物的工具变量（IV），来揭示血液代谢物与两种胆汁淤积性肝病（包括PBS和PSC）之间的潜在因果关联。PBC或PSC的GWAS汇总数据来自两个不同的数据集。初始分析采用逆方差加权（IVW）和一系列敏感性分析，随后利用芬兰基因组联盟数据进行重复分析和荟萃分析。最后，进行多变量MR分析以确定每种代谢物的独立效应。此外，使用基于网络的工具MetaboAnalyst 5.0进行代谢途径检查。
经过初步分析和错误发现率（FDR）校正后，识别出15种代谢物与胆汁淤积性肝病之间存在遗传因果关系。随后，9种代谢物通过重复分析和荟萃分析始终显示出关联。此外，多变量MR分析证实了7种代谢物的独立因果效应。具体而言，发现异戊酰肉碱（比值比[OR] = 3.146，95%置信区间[CI]：1.471 - 6.726，P = 0.003）、缬氨酸（OR = 192.44，95%CI：4.949 - 7483.27，P = 0.005）和甘露糖（OR = 0.184，95%CI：0.068 - 0.499，P < 0.001）与PBC的发生存在因果关系。此外，赤藓糖（OR = 5.504，95%CI：1.801 - 16.821，P = 0.003）、1 - 硬脂酰甘油磷酸胆碱（OR = 6.753，95%CI：2.621 - 17.399，P = 7.64×10）、X - 11847（OR = 0.478，95%CI：0.352 - 0.650，P = 2.28×10）和X - 12405（OR = 3.765，95%CI：1.771 - 8.005，P = 5.71×10）与PSC的发生独立相关。此外，代谢途径分析在两种胆汁淤积性肝病中确定了7条显著途径。
本研究结果揭示了7种代谢物与两种胆汁淤积性肝病之间的紧密因果关系，从而为这些疾病的代谢机制和治疗策略提供了新的见解。
