Identification of immune-inflammation targets for intracranial aneurysms: a multiomics and epigenome-wide study integrating summary-data-based Mendelian randomization, single-cell-type expression analysis, and DNA methylation regulation

Abstract

Background: Dysfunction of the immune system and inflammation plays a vital role in developing intracranial aneurysms (IAs). However, the progress of genetic pathophysiology is complicated and not entirely elaborated. This study aimed to explore the genetic associations of immune-related and inflammation-related genes (IIRGs) with IAs and their subtypes using Mendelian randomization, colocalization test, and integrated multiomics functional analysis.

Methods: The authors conducted a summary-data-based Mendelian randomization (SMR) analysis using data from several genome-wide association studies of gene expression (31 684 European individuals) and protein quantitative trait loci (35 559 Icelanders), as well as information on IAs and their subtypes from The International Stroke Genetics Consortium (IGSC) for discovery phase and the FinnGen study for replication. This analysis aimed to determine the causal relationship between IIRGs and the risk of IAs and their subtypes. Further functional analyses, including DNA methylation regulation (1980, European individuals), single-cell-type expression analysis, and protein-protein interaction, were conducted to detect the specific cell type with enriched expression and discover potential drug targets.

Results: After integrating multiomics evidence from expression quantitative trait loci (eQTL) and protein quantitative trait loci (pQTL), the authors found that tier 1: RELT [odds ratio (OR): 0.14, 95% CI: 0.04-0.50], TNFSF12 (OR: 1.24, 95% CI: 1.24-1.43), tier 3: ICAM5 (OR: 0.89, 95% CI: 0.82-0.96), and ERAP2 (OR: 1.07, 95% CI: 1.02-1.12) were associated with the risk of IAs; tier 3: RELT (OR: 0.11, 95% CI: 0.02-0.54), ERAP2 (OR: 1.08, 95% CI: 1.02-1.13), and TNFSF12 (OR: 1.24, 95% CI: 1.05-1.47) were associated with the risk of aneurysmal subarachnoid hemorrhage (aSAH); and tier 1: RELT (OR: 0.04, 95% CI: 0.01-0.30) was associated with the risk of unruptured intracranial aneurysms (uIAs). Further functional analyses showed that RELT was regulated by cg06382664 and cg18850434 and ICAM5 was regulated by cg04295144 in IAs; RELT was regulated by cg06382664, cg08770935, cg16533363, and cg18850434 in aSAH; and RELT was regulated by cg06382664 and cg21810604 in uIAs. In addition, the authors found that H6PD (OR: 1.13, 95% CI: 1.01-1.28), NT5M (OR: 1.91, 95% CI: 1.21-3.01), and NPTXR (OR: 1.13, 95% CI: 1.01-1.26) were associated with IAs; NT5M (OR: 2.13, 95% CI: 1.23-3.66) was associated aSAH; and AP4M1 (OR: 0.06, 95% CI: 0.01-0.42) and STX7 (OR: 3.97, 95% CI: 1.41-11.18) were related to uIAs. STX7 and TNFSF12 were mainly enriched in microglial cells, whereas H6PD, STX7 , and TNFSF12 were mainly enriched in astrocytes.

Conclusions: After integrating multiomics evidence, the authors eventually identified IIRGs: RELT, TNFSF12, ICAM5 , and ERAP2 were the novel therapy targets for IAs. These new results confirmed a vital role of immune and inflammation in the etiology of IAs, contributing to enhance our understanding of the immune and inflammatory mechanisms in the pathogenesis of IAs and revealing the complex genetic causality of IAs.

Figure 1

Flowchart of study design. IIRGs, immune-inflammation related genes; IAs, intracranial aneurysms; ISCG, The International Stroke Genetics Consortium; QTL, quantitative trait loci; SNP, single nucleotide polymorphism; aSAH, aneurysmal subarachnoid hemorrhage; uIAs, unruptured intracranial aneurysms; PPH4, posterior probability of H4.

Figure 2

(A) Volcano plots of IAs in eQTL and pQTL level. (B) Volcano plots of aSAH in eQTL and pQTL level. (C) Volcano plots of uIAs in eQTL and pQTL level. OR, odds ratio; IAs: intracranial aneurysms; aSAH: aneurysmal subarachnoid hemorrhage; uIAs: unruptured intracranial aneurysms.

Figure 3

HyPrColoc plot of RELT and TNFSF12 variants in IAs and uIA. (A) and (B) showed the causal associations between two candidate genes and IAs. (C) showed the causal associations between RELT and uIA.

Figure 4

Single-cell type expression in IAs for the candidate genes of proteins identified by SMR analysis. (A) A total of 20 cell clusters and 14 cell types were identified. (B) and (C) show the expression of protein coding genes in each cluster. (D) Four protein-coding genes had evidence of enrichment in a cell type at average Log2FC >0.5 and FDR <0.05 level. (E) T-SNE plots show cells expression from different samples. (F) Heatmap of gene expression between different clusters.

Figure 5

(A) and (C) showed the construction of PPI network in pQTL level (P <0.05). (B) showed the Chromosomal localization of candidate IIRGs in IAs.