Skip to main content
NCBI home page
Brain and Behavior logoLink to Brain and Behavior
. 2026 Mar 31;16(4):e71366. doi: 10.1002/brb3.71366

Integrative Multi‐Omics Mendelian Randomization Highlights Causal Autophagy‐Related Genes for Amyotrophic Lateral Sclerosis

Zheng Jiang 1,2, Yan‐Lin Ren 3, Xiao‐Jing Gu 4, Wei‐Ming Su 1, Qing‐Qing Duan 1, Kang‐Fu Yin 1, Bei Cao 1, Jing‐Yu Li 3,✉, Bo Yan 1,✉, Yong‐Ping Chen 1,✉
PMCID: PMC13112017  PMID: 41917768

ABSTRACT

Background

Autophagy dysregulation has been implicated in the toxic protein aggregates of amyotrophic lateral sclerosis (ALS). However, the causal relationship between impaired autophagy and ALS remains ambiguous, necessitating further elucidation.

Methods

This Mendelian randomization (MR) study employs a two‐sample design, utilizing genetic instruments to proxy autophagy dysregulation as the exposure and ALS as the outcome. It incorporates summary statistics of ALS (27,205 cases, 110,881 controls), along with data on DNA methylation, RNA splicing, gene expression, and protein abundance quantitative trait loci (QTLs) in both blood and brain tissues (mQTL, sQTL, eQTL, and pQTL, respectively) sourced from European cohorts. Cis‐variants situated proximal to or within the 604 autophagy‐related genes, exhibiting robust associations with molecular alterations in autophagy, are employed as instrumental variables. Their causal links with ALS are assessed via summary‐data‐based MR (SMR) analyses, followed by Bayesian colocalization, sensitivity analyses, brain cell‐specific MR analyses, protein–protein interaction (PPI), and druggable analyses.

Results

Consistent evidence supported the causal effects of two lysosome genes (FNBP1 and IDUA), one autophagy core gene (C9orf72), and one mitophagy gene (USP35) on ALS risk. Specifically, brain FNBP1 splicing level (OR = 1.18, p = 3.38E‐5) and blood USP35 expression level (OR = 1.17, p = 5.94E‐5) were positively associated with higher ALS risk. In contrast, we found strong causal evidence of brain IDUA methylation level (OR = 0.96, p = 8.36E‐6) and blood C9orf72 methylation level (OR = 0.55, p = 7.59E‐12) with lower ALS risk. Cell‐type‐specific MR analyses, PPI, and druggable analyses further nominated the key brain cell type (astrocytes), potential interaction with known causative genes (SQSTM1 and PFN1), and promising druggability for FNBP1 in ALS.

Conclusions

This multi‐omics MR study identified causal associations between the regulation of four autophagy‐related genes and ALS risk, shedding light on autophagy‐mediated mechanisms and offering early evidence of novel therapeutic targets for ALS.

Keywords: amyotrophic lateral sclerosis, autophagy dysregulation, causal genes, Mendelian randomization

Multi‐omics Mendelian randomization reveals causal roles of four autophagy‐related genes (FNBP1, IDUA, C9orf72, USP35) in ALS risk, with FNBP1 and USP35 increasing risk and IDUA and C9orf72 protective, shedding light on autophagy‐mediated mechanisms and offering early evidence of novel therapeutic targets for ALS.

graphic file with name BRB3-16-e71366-g002.jpg

1. Introduction

Autophagy is a cellular process that acts as a self‐cleaning mechanism, allowing cells to degrade and recycle their components. It is a highly regulated process essential for cell survival and has been implicated in various physiological and pathological conditions, including aging and neurodegenerative diseases (Klionsky et al. 2021). Autophagy dysfunction is a well‐known amyotrophic lateral sclerosis (ALS) feature, characterized by the accumulation of neurotoxic misfolded proteins. Defective autophagy may lead to the buildup of misfolded proteins and damaged organelles, contributing to neuronal dysfunction and the progression of ALS (Xu et al. 2021). Thus, understanding the precise roles of autophagy in ALS is essential for advancing our knowledge and developing potential therapeutic strategies for the disease.

It is worth noting that autophagy dysregulation represents a multifaceted cellular phenomenon manifesting across a spectrum of pathophysiological states, which pose challenges in accurately determining the timing and spatial localization of its impairment for specific diseases. Many autophagy‐related genes have been implicated in the autophagy‐lysosome process, and genetic variants in these genes may potentially lead to the occurrence of autophagy dysregulation (Kocak et al. 2022). Numerous genetic studies have endeavored to elucidate the causative association between autophagy dysregulation and ALS by exploring rare pathogenic variants and polymorphisms of autophagy‐related genes involved in ALS development (Deng et al. 2017). However, rare pathogenic variants usually account for only a very small fraction of ALS cases, and results from genetic association studies have often been inconsistent due to differences in sample size and ethnic origin. In contrast to these prior mutation‐ or candidate gene‐focused studies, our study adopts a systematic, multi‐omics Mendelian randomization (MR) approach to evaluate whether regulatory genetic variants affecting DNA methylation, RNA splicing, gene expression, or protein abundance in all autophagy‐related genes contribute causally to ALS risk, providing a broader and more unbiased assessment of autophagy's role in ALS pathogenesis.

MR is an analytical approach employing genetic variants as instrumental variables (IVs) to infer potential causal relationships between exposures and outcomes of interest, which mitigates biases arising from unobserved confounders and reverse causality (Skrivankova et al. 2021). Genome‐wide association studies (GWAS) leverage genetic associations with phenotypes utilizing single‐nucleotide polymorphisms (SNPs), while the combination of GWAS statistics with DNA methylation, RNA splicing, gene expression, or protein abundance data has facilitated the delineation of diverse quantitative trait loci (mQTL, sQTL, eQTL, or pQTL, respectively) (Bellenguez et al. 2022; Yang et al. 2021). Moreover, summary‐data‐based Mendelian randomization (SMR) has expanded and refined the concept of MR, enabling the utilization of independent GWAS summary statistics data and QTL data to prioritize potential causal genes derived from GWAS findings (Zhu et al. 2016). By employing SMR in conjunction with the heterogeneity in dependent instruments (HEIDI) test, potential causal relationships were discerned amid the genome's prevalent linkage disequilibrium (LD).

Currently, no MR study has investigated the potential causal association between autophagy dysregulation and ALS risk. The primary objective of this study was to investigate the potential causal association between autophagy dysregulation, as indicated by genetic predispositions in autophagy‐related genes, and ALS risk. In order to establish instruments for autophagy dysregulation from the autophagy‐related genome, we conducted a comprehensive and unbiased analysis that integrated multi‐omics QTL and GWAS data using a two‐sample MR approach. Furthermore, we performed various sensitivity analyses to strengthen the robustness of our findings.

2. Methods

Our research strictly followed the STROBE‐MR reporting guideline (see Supporting Information) (Skrivankova et al. 2021). Figure 1 provides an overview of the study's design and the methodology employed in selecting genetic variants and analytical processes. In order to ascertain autophagy dysregulation as indicated by the genetic predisposition in the autophagy‐related genome, we retrieved a list of 604 established autophagy‐related genes comprehensively summarized by an original publication (Bordi et al. 2021). The main categories included MTOR and upstream pathways (135 genes), autophagy core (197 genes), autophagy transcription factors (68 genes), mitophagy (80 genes), docking and fusion (22 genes), and lysosome (162 genes) and lysosome‐related genes (34 genes) (for details, see Table S1).

FIGURE 1.

FIGURE 1

Open in a new tab

Flow diagram of the MR analyses performed.

2.1. Data Source

2.1.1. Exposure Source

All SNPs included in the initial analysis had at least a suggestive p value < 5 × 10−8. SNPs located within the human major histocompatibility complex region at chromosome 6: 26–34 Mb (GRCh37/hg19) were excluded considering their complex LD structure (Zheng et al. 2020).

2.1.1.1. mQTL

Blood cis‐mQTL instruments robustly associated with autophagy gene methylation were extracted using summary data from a meta‐analysis of two cohorts (n = 1980) (McRae et al. 2018). As a result, 1601 SNPs were selected, corresponding to 1902 autophagy‐related DNA methylation CpG sites.

The brain cis‐mQTL database originated from a 2018 publication by Qi et al. (2018). In this study, the authors conducted a meta‐analysis of cis‐mQTLs using data from three different datasets: 468 cortical brain region samples from the ROSMAP study (Ng et al. 2017), 166 fetal brain samples by Hannon et al. (2016), and 526 frontal cortex region samples (Jaffe et al. 2016). This resulted in a final sample size of 1160 brain samples and nearly 6 million cis‐mQTLs. Consequently, we identified 1623 SNPs corresponding to 1929 autophagy‐related DNA methylation CpG sites.

2.1.1.2. sQTL

Blood cis‐sQTL instruments robustly associated with autophagy RNA splicing abundance were extracted using summary data from GTEx V8 (n = 338) (Garrido‐Martín et al. 2021). In total, 237 SNPs of 328 autophagy‐related splicing probes were selected as instruments for the MR.

Human brain cis‐sQTL summary data were obtained from Qi et al. (2022), which contains sQTL data from 2865 human brain transcriptomes. As a result, 1146 SNPs of 1591 autophagy‐related splicing probes were selected.

2.1.1.3. eQTL

Blood eQTL instruments for autophagy‐related genes were obtained using the cis‐eQTL summary statistics from the eQTLGen Consortium (https://www.eqtlgen.org/cis‐eqtls.html). The eQTLGen Consortium provides information on 10,317 trait‐associated SNPs from 31,684 individuals (Vosa et al. 2021). It is important to note that variants associated with gene expression levels of X and Y chromosomes were not included in the eQTLGen dataset. For the cis‐eQTL analysis, we selected 489 SNPs linked to the expression of 489 autophagy‐related transcripts.

To generate brain cis‐eQTL instruments for autophagy‐related genes, we used the BrainMeta v2 dataset (Qi et al. 2022), which contains eQTL data from 2865 human brain transcriptomes. As a result, 347 SNPs associated with the expression of 347 autophagy‐related transcripts were selected.

2.1.1.4. pQTL

Blood cis‐pQTL instruments for genetic variants associated with the expression of autophagy‐related proteins were selected from the study by Zheng et al. (2020), which integrated five published proteome datasets (Suhre et al. 2017; Yao et al. 2018; Folkersen et al. 2020; Sun et al. 2018; Emilsson et al. 2018), and 23 SNPs robustly associated with 23 autophagy‐related protein expressions were selected.

The brain cis‐pQTL data were derived from two datasets: the Religious Orders Study (ROS) or the Rush Memory and Aging Project (MAP) dataset (referred to as the ROSMAP dataset), consisting of dorsolateral prefrontal cortex (dlPFC) samples from 376 subjects (Beach et al. 2015), and the Banner dataset comprising dlPFC samples from 152 individuals collected by the Banner Sun Health Research Institute (Beach et al. 2015). Finally, 32 SNPs associated with autophagy‐related protein expressions were selected for analysis.

2.2. Outcome Source

The outcome GWAS were performed in 29,612 patients with ALS and 122,656 controls, but we only used the summary results conducted in European ancestry populations (27,205 cases, 110,881 controls) (van Rheenen et al. 2021).

Table 1 summarizes QTL and GWAS data used in this research, and details on the study design have been described in the original paper. The present study only involved publicly available GWAS summary statistics, so informed consent and ethical approval were not required.

TABLE 1.

Information on QTL and GWAS datasets.

Type of dataset Data subtype Resource Sample size Sample Age Group Brain region stratification Reference Download Site
QTL Brain cis‐mQTL Brain‐mMeta mQTL summary data 1160 Adults Yes Qi et al. (2018) https://yanglab.westlake.edu.cn/data/SMR/Brain‐mMeta.tar.gz
Blood cis‐mQTL McRae et al. mQTL summary data 1980 Adults (including older adults in LBC) / Y. Wu et al. (2018) https://yanglab.westlake.edu.cn/data/SMR/LBC_BSGS_meta_lite.tar.gz
Brain cis‐sQTL BrainMeta v2 sQTL summary data 2865 Adults Yes Qi et al. (2022) https://yanglab.westlake.edu.cn/data/SMR/BrainMeta_cis_sqtl_summary.tar.gz
Blood cis‐sQTL V8 release of the GTEx sQTL summary data 338 Adults (21–70 years) / GTEx Consortium (2020) https://yanglab.westlake.edu.cn/data/SMR/GTEx_V8_cis_sqtl_summary_lite.tar
Brain cis‐eQTL BrainMeta v2 eQTL summary data 2865 Adults Yes Qi et al. (2018) https://yanglab.westlake.edu.cn/data/SMR/BrainMeta_cis_eqtl_summary.tar.gz
Blood cis‐eQTL eQTLGen Consortium 31,684 Adults / Vosa et al. (2021) https://www.eqtlgen.org/cis‐eqtls.html
Brain cis‐pQTL ROSMAP Consortium 376 Older adults Only dorsolateral prefrontal cortex Robins et al. (2021) https://adknowledgeportal.org/
BANNER Consortium 149

Blood cis‐pQTL

 Five published proteome datasets integrated by Zheng et al. (2020) (PMID: 32895551) 1000 Adults (23–81 years) / Suhre et al. (2017) https://www.nature.com/articles/ncomms14357#MOESM1569
3200 Older adults (> 65 years) / Emilsson et al. (2018) https://www.science.org/doi/10.1126/science.aaq1327?url_ver=Z39.882003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%20%200pubmed
3301 Adults (blood donors) / Sun et al. (2018) http://www.phpc.cam.ac.uk/ceu/proteins/
6861 Adults (mean age 50) / Yao et al. (2018) https://www.nature.com/articles/s41467‐018‐05512‐x#additional‐information
21,758 Adults (middle and old age) / Folkersen et al. (2020) https://www.nature.com/articles/s42255‐020‐00287‐2#Sec35
cis‐eQTL of Eight Brain Cells Published paper 192 Adults No Bryois et al. (2022) https://zenodo.org/record/7276971
GWAS summary ALS Project MinE case: 27,205, control: 110,881 Adults / van Rheenen et al. (2021) https://www.projectmine.com/research/download‐data/
Open in a new tab

2.3. Statistical Methods

The main analytical process consisted of SMR, colocalization, and sensitivity analyses.

2.3.1. SMR Analysis

MR necessitates fulfilling three fundamental assumptions. Assumption 1: The genetic instruments should exhibit associations with the exposures, that is, QTL(s). Assumption 2: The genetic instruments should exclusively influence the outcomes, that is, ALS, through their effects on the exposures, thereby avoiding any instances of horizontal pleiotropy. Assumption 3: The genetic instruments should not be linked to any confounding factors that could impact the association between the exposure and outcome. To extend the concept of MR, the SMR method was devised to examine the pleiotropic association between exposure (DNA methylation, RNA splicing, gene expression, or protein abundance) and outcome (ALS risk) (Zhu et al. 2016). To satisfy the MR assumptions, we computed the causal association as β autodys‐ALS = β SNP‐ALS/β SNP‐autodys. β autodys‐ALS represents the estimated effect size of autophagy dysregulation on ALS, whereas β SNP‐autodys denotes the estimated effect size of the genetic variant on autophagy dysregulation (the association between the genetic variant and the exposure trait), and β SNP‐ALS denotes the estimated effect size of the genetic variant on ALS (the association between the same genetic variant and the outcome trait). The HEIDI test (p HEIDI > 0.01) was utilized to exclude significant SMR associations due to linkage rather than causality or pleiotropy (see Supporting Information).

In this study, we conducted SMR analyses utilizing the Linux version 1.3.1 of the SMR software by default settings. To assess the impact of autophagy dysregulation on ALS risk, we acquired odds ratio (OR) estimates by ORautodys‐ALS = exp(β autodys‐ALS), in which OR represents the odds ratio estimate per a 1‐ln increment in autophagy‐related molecular alteration levels, and exp denotes the base of the natural logarithm. We defined significance at a false discovery rate (FDR) of 0.01 for each SMR analysis (see Supporting Information).

2.4. Colocalization Analysis

The HEIDI test, which utilizes an external reference to estimate LD, is one method employed for colocalization analysis. To further refine the results, we conducted a Bayesian test to assess the colocalization of two traits and estimate the posterior probability of shared variants (Giambartolomei et al. 2014). We retrieved all SNPs within a 100 kb range upstream and downstream of the leading SNPs for colocalization analysis. This analysis aimed to evaluate the posterior probability of H4 (PP.H4), with a cutoff of PP.H4 > 0.8 commonly used to indicate strong evidence of colocalization between the GWAS and QTL associations (see Supporting Information).

2.5. Sensitivity Analysis

Following SMR analyses, sensitivity analyses were conducted using other MR methods, namely inverse variance weighting (IVW), weighted median, MR Egger, simple mode, and weighted mode, with default parameters. Each method originated from various assumptions on instrument validity estimates the causal effect, thereby providing robust evidence for primary findings. Additionally, Steiger analyses were used to confirm that the proposed instruments were directly associated with the outcome or effect estimate directionality (Hemani et al. 2017). To test for heterogeneity (Q test p < 0.05) and horizontal pleiotropy (MR‐Egger intercept p < 0.05), we employed Cochran's Q (Higgins et al. 2003) and MR Egger intercept test (Burgess and Thompson 2017), respectively. Furthermore, we conducted a leave‐one‐out analysis within the IVW method to assess the influence of individual variants on the observed association (Burgess et al. 2017). When only one genetic variant was included in the analysis, the simplest approach, the Wald ratio estimate, was used to obtain the causal estimate for the genetic variant representing the exposure on the outcome (Burgess et al. 2017). The F‐statistic was employed to measure the strength of the variants, with a value > 10 indicating a strong MR instrument (Skrivankova et al. 2021).

Colocalization and sensitivity analyses were mainly performed using the packages TwoSampleMR (0.5.6) and coloc (5.2.0) of the statistical software R (4.1.3).

2.6. Phenome‐Wide Scan of Identified Genetic Variants

To further evaluate horizontal pleiotropic effects, we conducted a comprehensive search for identified genetic variants surviving the above analyses using the PhenoScanner database (Kamat et al. 2019), which incorporates over 5000 GWAS at the time of this investigation. We performed a phenome‐wide scan (PheWAS) by considering all identified genetic variants related to disease traits, and an SNP was considered pleiotropic when satisfying the following criteria: (1) the SNPs exhibited the same effect allele as our findings, (2) the association reached genome‐wide significance (p < 5 × 10−8), (3) the absolute value of the effect size (β) was over 0.01, and (4) the ethnic population was of European descent.

2.7. Regulatory Relationships Between Epigenetic and Transcriptional Layers

DNA methylation and RNA splicing are known to influence gene expression and protein abundance through transcriptional and posttranscriptional mechanisms (J. Wang et al. 2013; Kornblihtt et al. 2013). To provide additional mechanistic context for the identified QTL–ALS associations, we conducted SMR analyses to examine whether ALS‐associated DNA methylation and RNA splicing signals were also associated with gene expression or protein abundance. These analyses were intended as exploratory investigations to characterize potential regulatory relationships between different molecular layers, rather than to establish formal causal hierarchies linking genetic variants to ALS risk.

2.8. Cell‐Type‐Specific MR Analyses of Identified Causal Autophagy Genes

To nominate the potential brain cell types for identified causal autophagy genes, we further performed MR analyses using cis‐eQTL data (Bryois et al. 2022) of eight brain cells (astrocytes, endothelial cells, excitatory neurons, inhibitory neurons, oligodendrocytes, OPCs, pericytes, and microglia). The IV selection criteria for identified genes were the same as those in the sensitivity analyses.

2.9. Protein–Protein Interaction Analysis

To investigate the interactions between the autophagy‐related genes in this study and the known ALS causative/risk genes extracted from the Amyotrophic Lateral Sclerosis Online Database (Abel et al. 2012), we explored the protein–protein interaction (PPI) network between them by using the Search Tool for the Retrieval of Interacting Genes (STRING) database version 11.5 (https://string‐db.org/) (Szklarczyk et al. 2019). We input gene symbols into the STRING database and set the minimum required interaction score at 0.4 to ensure confidence in the results.

2.10. Druggable Analysis

Genes with QTL signals in blood or brain that passed SMR, sensitivity, and colocalization analyses were interrogated for druggability in the Drug Gene Interaction Database (DGIdb) (version 4.2) (https://www.dgidb.org/) (Freshour et al. 2021). DGIdb employs various sources of information, including expert curation, text‐mining, and data on drug–gene interactions from databases such as DrugBank, PharmGKB, ChEMBL, and Drug Target Commons, to prioritize genes with potential druggability and assess gene function (Freshour et al. 2021).

3. Results

3.1. MR Analysis of Autophagic Cis‐mQTLs and ALS

For the causal association between the DNA methylation of the autophagy‐related genes and ALS, SMR analyses (FDR p SMR < 0.01) and HEIDI tests (p HEIDI > 0.01) were performed. After correction for multiple testing, we identified one association signal across one unique genetic locus for ALS in blood and two association signals across one unique genetic locus for ALS in brain (Figure 2 and Table S2). The sensitivity and colocalization analyses supported the SMR analyses (Tables S2 and S3). In concrete, we found strong causal evidence of blood C9orf72 methylation level (OR per SD, 0.55; 95% CI, 0.47–0.65; p = 7.59E‐12), as well as brain IDUA methylation level (OR per SD, 0.96; 95% CI, 0.94–0.98; p = 8.36E‐6) with lower ALS risk.

FIGURE 2.

FIGURE 2

Open in a new tab

Mendelian randomization results for the association between the autophagy‐related genes and ALS risk. (A) Volcano plot displaying the log2 Beta (x axis) against the −log10 FDR p value (y axis) for the association between the autophagy‐related QTLs and ALS risk. (B) Forest plot for SMR estimates for the outcomes reaching 1% FDR significance without evidence of heterogeneity as determined by the HEIDI test. aDifferent probes may detect the same SNP. bOdds ratios were calculated by the expectation of the causal estimate (β coefficient). cColocalization indicates PP.H4 between QTLs and ALS risk. PP.H4 > 0.8 is the well‐applied cutoff for the evidence of colocalization.

3.2. MR Analysis of Autophagic Cis‐sQTLs and ALS

For the causal association between the RNA splicing of the autophagy‐related genes and ALS, SMR analyses (FDR p SMR < 0.01) and HEIDI tests (p HEIDI > 0.01) identified two association signals across two unique genetic loci for ALS in the brain after correction for multiple testing (Figure 2 and Table S2). The sensitivity analyses supported the same associations as the SMR analyses (Table S3). Specifically, brain FNBP1 splicing level was positively associated with higher ALS risk (OR per SD, 1.18; 95% CI, 1.09–1.28; p = 3.38E‐5). However, colocalization analyses suggested that the causal effect of C9orf72 on ALS risk may be confounded by LD.

3.3. MR Analysis of Autophagic Cis‐eQTLs and ALS

For the causal association between the RNA expression of the autophagy‐related genes and ALS, SMR analyses (FDR p SMR < 0.01) and HEIDI tests (p HEIDI > 0.01) identified two association signals across two unique genetic loci for ALS in blood after correction for multiple testing (Figure 2 and Table S2). Specifically, blood USP35 expression level was positively associated with higher ALS risk (OR per SD, 1.17; 95% CI, 1.09–1.27; p = 5.94E‐5) (for details, see Figure 2, Table S2, and Table S3). However, the sensitivity and colocalization analyses failed to support the causal effect of NME4 expression abundance indicated by rs6600214 on ALS risk.

3.4. MR Analysis of Autophagic Cis‐pQTLs and ALS

For the causal association between the protein abundance of the autophagy‐related genes and ALS, SMR analyses (FDR p SMR < 0.01) and HEIDI tests (p HEIDI > 0.01) were performed. Only 23 proposed autophagy‐related SNPs in blood and 32 in the brain were extracted from cis‐pQTLs. However, no association signals survived the correction for multiple testing.

3.5. Summary of SMR, Sensitivity, and Colocalization Analyses

First, we identified potential causal relationships between five autophagy‐related genes and ALS after SMR analyses. Subsequently, four of them survived the predefined sensitivity and colocalization analyses. Finally, we found five QTL signals from two lysosome genes (FNBP1 and IDUA), one autophagy core gene (C9orf72), and one mitophagy gene (USP35) in either blood or brain that have causal effects on ALS risk (Figure 3). Overall, four identified autophagy‐related genes manifest their causal influence on ALS risk through mQTL, sQTL, and eQTL signals.

FIGURE 3.

FIGURE 3

Open in a new tab

Causal autophagy‐related genes for amyotrophic lateral sclerosis.

3.6. Phenome‐Wide Scan of Identified Genetic Variants

We performed a PheWAS of the included genetic variants using the PhenoScanner database to systematically identify potential associations with secondary phenotypes (Table S4). The RNA splicing variants of FNBP1 and C9orf72 that were causally associated with ALS were not found to be related to any secondary phenotype, supporting the robustness of their inferred causal effects on ALS risk.

In contrast, the DNA methylation variant of IDUA and the gene expression variant of USP35, both of which showed causal associations with ALS, were also associated with several additional phenotypes (Table S4). Specifically, the IDUA methylation variant was associated with Parkinson's disease (PD), while the USP35 expression variant was associated with whole‐body and limb impedance measures as well as age at menarche. These findings suggest the presence of potential pleiotropic effects for IDUA and USP35, which warrant further interpretation regarding their biological relevance and the nature of pleiotropy involved.

3.7. Regulatory Relationships Between Epigenetic and Transcriptional Layers

Furthermore, DNA methylation or RNA splicing can influence gene expression. Therefore, we also performed SMR analyses on the causal association between autophagy‐related gene methylation or RNA splicing and RNA expression by mapping the mQTLs or sQTLs to eQTLs through shared genetic variants. After correction for multiple testing and the HEIDI test, we obtained the gene list for the autophagy gene expression regulated by DNA methylation CpG or RNA splicing sites (Table S5). For the potential causal genes identified above, SMR results showed that the methylation of USP35, which was regulated by rs2512525, was associated with blood USP35 expression level (Table S5).

3.8. Cell‐Type‐Specific MR Analyses of Identified Causal Autophagy‐Related Genes

To support the role of identified causal autophagy genes in specific brain cells, cell‐type‐specific MR analyses were further performed. As a result, the expression level of FNBP1 in astrocytes was negatively associated with ALS risk (OR per SD, 0.88; 95% CI, 0.81–0.94; p = 3.99E‐4). In contrast, the expression level of C9orf72 in astrocytes was positively associated with ALS risk (OR per SD, 1.25; 95% CI, 1.15–1.36; p = 1.33E‐7) (Table S6). Due to the limited number of independent SNPs available in current cell‐type‐specific eQTL datasets, HEIDI and Bayesian colocalization analyses were not applicable; thus, these results were considered supportive rather than confirmatory.

3.9. PPI Analyses

The PPI network revealed interactions between two autophagy‐related genes (FNBP1 and C9orf72) and multiple ALS causative/risk genes (Figure 4 and Table S7). Specifically, FNBP1 interacts with SQSTM1 and PFN1, while the well‐known ALS causative C9orf72, also identified in this study, demonstrated various interactions with other ALS causative/risk genes, such as SOD1 and TARDBP (Figure 4).

FIGURE 4.

FIGURE 4

Open in a new tab

The protein–protein interaction network between the identified causal autophagy‐related genes and known ALS causative/risk genes. The nodes of red color mean the potential causal genes identified in this study, while the nodes of purple color represent the known ALS causal/risk genes interacting with the identified genes. Line thickness indicates the confidence of interaction.

3.10. Druggable Analyses

We next evaluated whether four autophagy‐related genes (FNBP1, IDUA, C9orf72, and USP35) with QTL signals robustly associated with ALS risk were potentially druggable by using DGIdb (Freshour et al. 2021). Although no evidence of drug–gene interactions was found, druggability analyses indicated that all of them were druggable (Table S8), which will contribute to prioritizing novel ALS drug targets.

4. Discussion

In this study, an integrative analytical pipeline combining SMR, colocalization, sensitivity analysis, regulatory relationship analyses involving eQTLs, cell‐type‐specific MR analysis, PPI analysis, and druggability analysis was employed to systematically explore the potential causal relationship between autophagy dysregulation and ALS risk. Specifically, we found two lysosome genes (FNBP1 and IDUA), one autophagy core gene (C9orf72), and one mitophagy gene (USP35) in either blood or brain that have causal effects on ALS risk. These genes may mediate the risk of ALS by modulating either RNA splicing, DNA methylation, or RNA expression. Especially, FNBP1 manifested no horizontal pleiotropy associated with any secondary traits and was also validated in cell‐specific MR analyses, and the PPI analyses further provided evidence for its gene‐gene interaction with the known causative/risk genes of ALS. In addition, this gene has been identified as clinically actionable through druggability analyses. Overall, our study is fundamental as an attempt to fill the gaps in our understanding of causally involved autophagy‐related genes and shed light on the critical components of autophagy in the pathogenesis of ALS.

Although a series of causative or risk genes for ALS have been discovered over time, they can only account for a small proportion of ALS cases (Chen et al. 2022). Currently, researchers have a minimal understanding of the underlying mechanisms driving the disease in most ALS patients. Prior investigations have implicated autophagy as a crucial player in ALS pathology; however, the exact mechanisms by which autophagy contributes to the disease remain incompletely elucidated. In the most recent GWAS (van Rheenen et al. 2021), the genetic loci associated with ALS were enriched with genes involved in autophagy and degradation of (misfolded) proteins. Using multi‐omics data from both blood and brain, our analysis identified four autophagy‐related genes potentially involved in the pathogenesis of ALS, including one known causative gene, C9orf72, as well as three promising genes, FNBP1, IDUA, and USP35.

The C9orf72 gene has been established as a known pathogenic gene associated with ALS, expanded repeats of which account for ∼40% of familial and ∼7% of sporadic ALS in Europeans (Akçimen et al. 2023). Three primary hypothesized mechanisms driving C9orf72‐related ALS involve C9orf72 protein haploinsufficiency and toxic gain of function through sequestration of RNA‐binding proteins and dipeptide repeat proteins formed through non‐canonical RNA translation (Gagliardi et al. 2020). Although the pathogenic mechanisms by which repeat expansions in the C9orf72 gene are causing ALS are not fully understood, many studies indicated that dysregulation of autophagy has been implicated in C9orf72‐related ALS (Beckers et al. 2021). Consistent with previous studies, our research reaffirmed the link between C9orf72 and ALS risk. Importantly, we indicated the protective mechanism by which C9orf72 influences ALS risk partially through blood DNA methylation, aligning with prior findings that C9orf72 hypermethylation may protect against ALS repeat expansion‐associated pathology (E. Y. Liu et al. 2014). In addition, cell‐type‐specific MR analyses validated the role of C9orf72 in astrocytes, which is an important brain cell type for ALS (Rajpurohit et al. 2020; Allen et al. 2019; Madill et al. 2017; Sugiyama et al. 2013). However, at the C9orf72 locus, we cannot fully exclude the possibility that some cis‐QTL variants are in LD with the pathogenic hexanucleotide repeat expansion, as previously demonstrated in ALS GWAS (e.g., van Rheenen et al. 2021). Nevertheless, our findings were based on QTL instruments that directly proxy regulatory changes (e.g., DNA methylation), which are less likely to be entirely explained by repeat tagging. Future studies incorporating long‐read sequencing or direct genotyping of C9orf72 expansions will be important to further clarify this issue.

FNBP1 encodes a member of the formin‐binding‐protein family, which mainly exists in the lysosome and cytosol. It plays a vital role in clathrin‐mediated endocytosis (Shimada et al. 2007). Although FNBP1 has been preliminarily reported as a risk gene for ALS by a multiethnic meta‐GWAS (Nakamura et al. 2020), its role was not clarified in ALS. A recent large‐scale transcriptomics analysis of human postmortem ALS spinal cords identified that oligodendrocytes were exclusively assigned as a putative cell type for FNBP1 (Humphrey et al. 2023). In this study, FNBP1 is proven to affect ALS through RNA splicing, which is consistent with our findings of a positive relationship between brain FNBP1 splicing level and ALS risk. More importantly, the sQTL data we analyzed are from the brain, which expands the potential pathophysiologic role of FNBP1 from the spinal cord to the brain, another key affected region in ALS. Furthermore, our study validates the protective role of FNBP1 expression level in astrocytes by cell‐type‐specific MR analyses, supported by the significant downregulation of this gene in animal models (W. Liu et al. 2020) as well as sporadic ALS patients (Andrés‐Benito et al. 2017). In the PPI analyses, FNBP1 is found to interact with an ALS‐associated gene SQSTM1 and PFN1, both of which may act in the formation of pathological protein aggregates in ALS models (Brady et al. 2011; C.‐H. Wu et al. 2012). Notably, the PheWAS of identified genetic variants for FNBP1 found no pleiotropic ones associated with the secondary traits, and the druggability analyses have assigned this gene to be clinically actionable. Overall, our results firstly indicated that FNBP1 may be a robust autophagy‐related gene with causal links with ALS in the brain.

IDUA and USP35 emerge as two novel genes discovered in our study, albeit with relatively limited supporting evidence for their involvement in ALS. IDUA encodes the enzyme alpha‐l‐iduronidase, which is responsible for the degradation of glycosaminoglycans in lysosomes. Mutations in the IDUA gene cause mucopolysaccharidosis type I, a lysosomal storage disease (Poletto et al. 2019). Although this gene has not been reported to be associated with ALS, our study provides preliminary evidence indicating a possible association between IDUA and ALS risk through DNA methylation in the brain. The dysregulation of IDUA‐mediated lysosomal degradation pathways may contribute to the accumulation of protein aggregates and impaired autophagic flux observed in ALS. The USP35 gene encodes a member of the peptidase C19 family of ubiquitin‐specific proteases, which catalyze the removal of ubiquitin proteins from target proteins. USP35 has been involved in several cellular processes, including the regulation of PARK2‐mediated mitophagy (Y. Wang et al. 2015) and Aurora B stability (Park et al. 2018). Our findings, which indicate a positive association between blood USP35 gene expression and ALS risk, differ from a recent study that observed downregulation of USP35 in ALS‐affected neurons (Farrawell et al. 2020), highlighting the complexity of its role across different cell types.

The PheWAS analysis revealed that genetic variants instrumenting IDUA and USP35 were associated with several secondary phenotypes, raising the possibility of pleiotropic effects. However, the presence of such associations does not necessarily imply horizontal pleiotropy that would violate MR assumptions. For USP35, the observed associations with whole‐body and limb impedance measures may reflect vertical pleiotropy, whereby altered gene expression influences body composition–related traits that could lie upstream of ALS risk. Such a mechanism would be compatible with a causal pathway linking USP35 expression to ALS rather than indicating an alternative confounding pathway. Similarly, the association between IDUA methylation and PD does not imply that PD mediates ALS risk. Instead, IDUA may exert independent effects on multiple neurodegenerative phenotypes through shared lysosomal or neurodegeneration‐related biological pathways. This pattern is consistent with biological pleiotropy rather than horizontal pleiotropy driven by distinct causal routes. Therefore, while pleiotropic associations were observed for IDUA and USP35, these findings should be interpreted cautiously and do not necessarily invalidate the inferred causal relationships with ALS. Future studies incorporating multivariable MR or pathway‐specific analyses may further clarify the underlying mechanisms.

The present study has several key strengths. First, it adopts a comprehensive multi‐omics MR approach to investigate the causal relationship between autophagy dysregulation and ALS. Additionally, integrating multiple molecular traits, that is, mQTL, sQTL, eQTL, and pQTL, provides a robust framework for causal inference. Second, the study includes various molecular traits specific to blood and brain tissues. This tissue‐specific analysis enables a more refined assessment of autophagy‐related molecular changes in tissues relevant to ALS pathology, while acknowledging that bulk brain and blood datasets may not fully capture disease‐relevant regulatory effects. Third, the study employs multiple MR methods, including the SMR approach, Bayesian colocalization, sensitivity analyses, regulatory relationship analyses involving eQTLs, cell‐type‐specific MR analyses, PPI, and druggable analyses. These rigorous analyses ensure the reliability, robustness, and consistency of the findings while offering valuable insights into potential pathways and therapeutic targets involved in the pathogenesis of ALS.

This study has several limitations as well. First, the relatively small number of autophagy‐related pQTLs likely reflects the limited coverage of current plasma proteomic datasets, which underrepresent low‐abundance intracellular proteins, rather than a true absence of protein‐level regulation. Moreover, a key limitation is the use of QTL data that sometimes conflates brain regions. While some datasets (e.g., GTEx Consortium 2020) distinguish partitions like the amygdala and cortex, others aggregate cortical areas or focus solely on the dlPFC (e.g., Robins et al., 2021). This may miss heterogeneity in gene expression, methylation, and splicing, potentially limiting insights into ALS mechanisms affecting motor neurons. We prioritized these datasets for sample size and statistical power, given the scarcity of ALS‐specific resources. Future studies incorporating single‐cell or motor neuron–focused QTL resources (e.g., PsychENCODE) may provide greater anatomical and cellular resolution. In addition, our analyses of regulatory relationships between DNA methylation, RNA splicing, and gene expression were not designed to establish definitive causal hierarchies from genetic variation to ALS risk. Ideally, such hierarchies would be evaluated by systematically testing whether mQTL or sQTL signals exert downstream effects on gene expression that subsequently mediate ALS risk. However, the limited number of variants instrumenting DNA methylation and RNA splicing for the identified genes constrained the statistical power for such analyses. Consequently, findings such as the association between a USP35 mQTL and USP35 expression, in the absence of a direct association between the mQTL and ALS risk, should be interpreted as providing a supportive mechanistic context rather than evidence of a complete causal pathway. These results suggest regulatory coherence across molecular layers but do not imply that upstream epigenetic variants independently drive ALS risk. Furthermore, for most autophagy‐related genes, only a single cis‐acting SNP was available as an IV, limiting our ability to perform multiple sensitivity analyses and reducing confidence in fully excluding horizontal pleiotropy. Some QTL datasets (e.g., ROSMAP) included both diseased and healthy individuals; while this increases power, it may also introduce potential bias due to disease‐related expression variation. Finally, the possibility of weak instrument bias cannot be completely ruled out, particularly for genes with low expression levels or limited QTL effect sizes. To mitigate this, we calculated F‐statistics for all instruments and excluded those with F < 10, but residual bias remains possible.

5. Conclusion

In summary, our results support the idea that autophagy dysregulation plays an essential role in the pathophysiology of ALS. Further work on the causal autophagy‐related genes for ALS, especially FNBP1, is needed to better understand the links between perturbed autophagy function and the neuronal death involved in ALS.

Author Contributions

Z.J. and Y.P.C. conceived and designed the study. Z.J. contributed to the statistical analysis and wrote the first draft of the manuscript. Z.J., Y.L.R., X.J.G., W.M.S., Q.Q.D., K.F.Y., and B.C. extracted the data and contributed to the statistical analysis. J.Y.L. and B.Y. revised and discussed the final edition. Y.P.C. supervised the project and revised and discussed the final edition. All authors read and approved the final version of the manuscript.

Funding

This study was supported by the National Key Research and Development Program of China (Grant No. 2022YFC2703101 to Y.P.C), the National Natural Science Fund of China (Grant No. 82371422 and 81971188 to Y.P.C.), the National Natural Science Fund of Sichuan (Grant No. 2022NSFSC0749 to B.C.), and the Science and Technology Bureau Fund of Sichuan Province (Grant No. 2023YFS0269 to Y.P.C).

Ethics Statement

This research involves analyzing publicly available data, for which ethical approval and individual consent were obtained from all original studies.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Materials: brb371366‐sup‐0001‐SuppMat.doc

BRB3-16-e71366-s001.doc (101.4KB, doc)

Supplementary Materials: brb371366‐sup‐0002‐SuppMat.xlsx

BRB3-16-e71366-s002.xlsx (1.3MB, xlsx)

Supplementary Materials: brb371366‐sup‐0003‐SuppMat.doc

BRB3-16-e71366-s003.doc (3.4MB, doc)

Acknowledgments

We thank all the investigators for making the summary data publicly available, and we are grateful to all the participants who contributed to those studies.

Contributor Information

Jing‐Yu Li, Email: jingyuli@scu.esu.cn.

Bo Yan, Email: yanbo96@163.com.

Yong‐Ping Chen, Email: yongping.chen@wchscu.cn.

Data Availability Statement

The GWAS summary statistics supporting this research are available from the corresponding GWAS consortium. The main paper and Supporting Information present all data supporting our findings. The code or algorithm used to generate results in this study is available from the corresponding authors upon reasonable request.

References

  1. Abel, O. , Powell J. F., Andersen P. M., and Al‐Chalabi A.. 2012. “ALSoD: A User‐Friendly Online Bioinformatics Tool for Amyotrophic Lateral Sclerosis Genetics.” Human Mutation 33, no. 9: 1345–1351. 10.1002/humu.22157. [DOI] [PubMed] [Google Scholar]
  2. Akçimen, F. , Lopez E. R., Landers J. E., et al. 2023. “Amyotrophic Lateral Sclerosis: Translating Genetic Discoveries Into Therapies.” Nature Reviews Genetics 24: 642–658. 10.1038/s41576-023-00592-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Allen, S. P. , Hall B., Castelli L. M., et al. 2019. “Astrocyte Adenosine Deaminase Loss Increases Motor Neuron Toxicity in Amyotrophic Lateral Sclerosis.” Brain 142: 586–605. 10.1093/brain/awy353. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Andrés‐Benito, P. , Moreno J., Aso E., Povedano M., and Ferrer I.. 2017. “Amyotrophic Lateral Sclerosis, Gene Deregulation in the Anterior Horn of the Spinal Cord and Frontal Cortex Area 8: Implications in Frontotemporal Lobar Degeneration.” Aging 9, no. 3: 823–851. 10.18632/aging.101195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Beach, T. G. , Adler C. H., Sue L. I., et al. 2015. “Arizona Study of Aging and Neurodegenerative Disorders and Brain and Body Donation Program.” Neuropathology 35, no. 4: 354–389. 10.1111/neup.12189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Beckers, J. , Tharkeshwar A. K., and Van Damme P.. 2021. “C9orf72 ALS‐FTD: Recent Evidence for Dysregulation of the Autophagy‐Lysosome Pathway at Multiple Levels.” Autophagy 17, no. 11: 3306–3322. 10.1080/15548627.2021.1872189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bellenguez, C. , Küçükali F., Jansen I. E., et al. 2022. “New Insights Into the Genetic Etiology of Alzheimer's Disease and Related Dementias.” Nature Genetics 54, no. 4: 412–436. 10.1038/s41588-022-01024-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bordi, M. , De Cegli R., Testa B., Nixon R. A., Ballabio A., and Cecconi F.. 2021. “A Gene Toolbox for Monitoring Autophagy Transcription.” Cell Death & Disease 12, no. 11: 1044. 10.1038/s41419-021-04121-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Brady, O. A. , Meng P., Zheng Y., Mao Y., and Hu F.. 2011. “Regulation of TDP‐43 Aggregation by Phosphorylation and p62/SQSTM1.” Journal of Neurochemistry 116, no. 2: 248–259. 10.1111/j.1471-4159.2010.07098.x. [DOI] [PubMed] [Google Scholar]
  10. Bryois, J. , Calini D., Macnair W., et al. 2022. “Cell‐Type‐Specific Cis‐eQTLs in Eight Human Brain Cell Types Identify Novel Risk Genes for Psychiatric and Neurological Disorders.” Nature Neuroscience 25, no. 8: 1104–1112. 10.1038/s41593-022-01128-z. [DOI] [PubMed] [Google Scholar]
  11. Burgess, S. , Bowden J., Fall T., Ingelsson E., and Thompson S. G.. 2017. “Sensitivity Analyses for Robust Causal Inference From Mendelian Randomization Analyses With Multiple Genetic Variants.” Epidemiology 28, no. 1: 30–42. 10.1097/EDE.0000000000000559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Burgess, S. , Small D. S., and Thompson S. G.. 2017. “A Review of Instrumental Variable Estimators for Mendelian Randomization.” Statistical Methods in Medical Research 26, no. 5: 2333–2355. 10.1177/0962280215597579. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Burgess, S. , and Thompson S. G.. 2017. “Interpreting Findings From Mendelian Randomization Using the MR‐Egger Method.” European Journal of Epidemiology 32, no. 5: 377–389. 10.1007/s10654-017-0255-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Chen, Y.‐P. , Yu S.‐H., Wei Q.‐Q., et al. 2022. “Role of Genetics in Amyotrophic Lateral Sclerosis: A Large Cohort Study in Chinese Mainland Population.” Journal of Medical Genetics 59, no. 9: 840–849. 10.1136/jmedgenet-2021-107965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Deng, Z. , Sheehan P., Chen S., and Yue Z.. 2017. “Is Amyotrophic Lateral Sclerosis/Frontotemporal Dementia an Autophagy Disease?” Molecular Neurodegeneration 12, no. 1: 90. 10.1186/s13024-017-0232-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Emilsson, V. , Ilkov M., Lamb J. R., et al. 2018. “Co‐Regulatory Networks of Human Serum Proteins Link Genetics to Disease.” Science 361, no. 6404: 769–773. 10.1126/science.aaq1327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Farrawell, N. E. , Mcalary L., Lum J. S., et al. 2020. “Ubiquitin Homeostasis Is Disrupted in TDP‐43 and FUS Cell Models of ALS.” Iscience 23, no. 11: 101700. 10.1016/j.isci.2020.101700. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Folkersen, L. , Gustafsson S., Wang Q., et al. 2020. “Genomic and Drug Target Evaluation of 90 Cardiovascular Proteins in 30,931 Individuals.” Nature Metabolism 2, no. 10: 1135–1148. 10.1038/s42255-020-00287-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Freshour, S. L. , Kiwala S., Cotto K. C., et al. 2021. “Integration of the Drug‐Gene Interaction Database (DGIdb 4.0) With Open Crowdsource Efforts.” Nucleic Acids Research 49, no. D1: D1144–D1151. 10.1093/nar/gkaa1084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gagliardi, D. , Costamagna G., Taiana M., et al. 2020. “Insights Into Disease Mechanisms and Potential Therapeutics for C9orf72‐Related Amyotrophic Lateral Sclerosis/Frontotemporal Dementia.” Ageing Research Reviews 64: 101172. 10.1016/j.arr.2020.101172. [DOI] [PubMed] [Google Scholar]
  21. Garrido‐Martín, D. , Borsari B., Calvo M., Reverter F., and Guigó R.. 2021. “Identification and Analysis of Splicing Quantitative Trait Loci Across Multiple Tissues in the Human Genome.” Nature Communications 12, no. 1: 727. 10.1038/s41467-020-20578-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Giambartolomei, C. , Vukcevic D., Schadt E. E., et al. 2014. “Bayesian Test for Colocalisation Between Pairs of Genetic Association Studies Using Summary Statistics.” PLOS Genetics 10, no. 5: e1004383. 10.1371/journal.pgen.1004383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. GTEx Consortium . 2020. “The GTEx Consortium Atlas of Genetic Regulatory Effects Across Human Tissues.” Science 369, no. 6509: 1318–1330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hannon, E. , Spiers H., Viana J., et al. 2016. “Methylation QTLs in the Developing Brain and Their Enrichment in Schizophrenia Risk Loci.” Nature Neuroscience 19, no. 1: 48–54. 10.1038/nn.4182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hemani, G. , Tilling K., and Davey Smith G.. 2017. “Orienting the Causal Relationship Between Imprecisely Measured Traits Using GWAS Summary Data.” PLOS Genetics 13, no. 11: e1007081. 10.1371/journal.pgen.1007081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Higgins, J. P. T. , Thompson S. G., Deeks J. J., and Altman D. G.. 2003. “Measuring Inconsistency in Meta‐Analyses.” BMJ 327, no. 7414: 557–560. 10.1136/bmj.327.7414.557. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Humphrey, J. , Venkatesh S., Hasan R., et al. 2023. “Integrative Transcriptomic Analysis of the Amyotrophic Lateral Sclerosis Spinal Cord Implicates Glial Activation and Suggests New Risk Genes.” Nature Neuroscience 26, no. 1: 150–162. 10.1038/s41593-022-01205-3. [DOI] [PubMed] [Google Scholar]
  28. Jaffe, A. E. , Gao Y., Deep‐Soboslay A., et al. 2016. “Mapping DNA Methylation Across Development, Genotype and Schizophrenia in the Human Frontal Cortex.” Nature Neuroscience 19, no. 1: 40–47. 10.1038/nn.4181. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kamat, M. A. , Blackshaw J. A., Young R., et al. 2019. “PhenoScanner V2: An Expanded Tool for Searching Human Genotype‐Phenotype Associations.” Bioinformatics 35, no. 22: 4851–4853. 10.1093/bioinformatics/btz469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Klionsky, D. J. , Abdel‐Aziz A. K., Abdelfatah S., et al. 2021. “Guidelines for the Use and Interpretation of Assays for Monitoring Autophagy (4th edition).” Autophagy 17, no. 1: 1–382. 10.1080/15548627.2020.1797280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kocak, M. , Ezazi Erdi S., Jorba G., et al. 2022. “Targeting Autophagy in Disease: Established and New Strategies.” Autophagy 18, no. 3: 473–495. 10.1080/15548627.2021.1936359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kornblihtt, A. R. , Schor I. E., Alló M., Dujardin G., Petrillo E., and Muñoz M. J.. 2013. “Alternative Splicing: A Pivotal Step Between Eukaryotic Transcription and Translation.” Nature Reviews Molecular Cell Biology 14, no. 3: 153–165. 10.1038/nrm3525. [DOI] [PubMed] [Google Scholar]
  33. Liu, E. Y. , Russ J., Wu K., et al. 2014. “C9orf72 hypermethylation Protects Against Repeat Expansion‐Associated Pathology in ALS/FTD.” Acta Neuropathologica 128, no. 4: 525–541. 10.1007/s00401-014-1286-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Liu, W. , Venugopal S., Majid S., et al. 2020. “Single‐Cell RNA‐Seq Analysis of the Brainstem of Mutant SOD1 Mice Reveals Perturbed Cell Types and Pathways of Amyotrophic Lateral Sclerosis.” Neurobiology of Disease 141: 104877. 10.1016/j.nbd.2020.104877. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Madill, M. , Mcdonagh K., Ma J., et al. 2017. “Amyotrophic Lateral Sclerosis Patient iPSC‐Derived Astrocytes Impair Autophagy via Non‐Cell Autonomous Mechanisms.” Molecular Brain 10, no. 1: 22. 10.1186/s13041-017-0300-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Mcrae, A. F. , Marioni R. E., Shah S., et al. 2018. “Identification of 55,000 Replicated DNA Methylation QTL.” Scientific Reports 8, no. 1: 17605. 10.1038/s41598-018-35871-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Nakamura, R. , Misawa K., Tohnai G., et al. 2020. “A Multi‐Ethnic Meta‐Analysis Identifies Novel Genes, Including ACSL5, Associated With Amyotrophic Lateral Sclerosis.” Communications Biology 3, no. 1: 526. 10.1038/s42003-020-01251-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Ng, B. , White C. C., Klein H.‐U., et al. 2017. “An xQTL Map Integrates the Genetic Architecture of the Human Brain's Transcriptome and Epigenome.” Nature Neuroscience 20, no. 10: 1418–1426. 10.1038/nn.4632. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Park, J. , Kwon M.‐S., Kim E. E., Lee H., and Song E. J.. 2018. “USP35 Regulates Mitotic Progression by Modulating the Stability of Aurora B.” Nature Communications 9, no. 1: 688. 10.1038/s41467-018-03107-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Poletto, E. , Pasqualim G., Giugliani R., Matte U., and Baldo G.. 2019. “Effects of Gene Therapy on Cardiovascular Symptoms of Lysosomal Storage Diseases.” Genetics and Molecular Biology 42, no. S1: 261–285. 10.1590/1678-4685-gmb-2018-0100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Qi, T. , Wu Y., Fang H., et al. 2022. “Genetic Control of RNA Splicing and Its Distinct Role in Complex Trait Variation.” Nature Genetics 54, no. 9: 1355–1363. 10.1038/s41588-022-01154-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Qi, T. , Wu Y., Zeng J., et al. 2018. “Identifying Gene Targets for Brain‐Related Traits Using Transcriptomic and Methylomic Data From Blood.” Nature Communications 9: 12. 10.1038/s41467-018-04558-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Rajpurohit, C. S. , Kumar V., Cheffer A., et al. 2020. “Mechanistic Insights of Astrocyte‐Mediated Hyperactive Autophagy and Loss of Motor Neuron Function in SOD1(L39R) Linked Amyotrophic Lateral Sclerosis.” Molecular Neurobiology 57, no. 10: 4117–4133. 10.1007/s12035-020-02006-0. [DOI] [PubMed] [Google Scholar]
  44. Robins, C. , Liu Y., Fan W., et al. 2021. “Genetic Control of the Human Brain Proteome.” American Journal of Human Genetics 108, no. 3: 400–410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Shimada, A. , Niwa H., Tsujita K., et al. 2007. “Curved EFC/F‐BAR‐domain Dimers Are Joined End to End Into a Filament for Membrane Invagination in Endocytosis.” Cell 129, no. 4: 761–772. 10.1016/j.cell.2007.03.040. [DOI] [PubMed] [Google Scholar]
  46. Skrivankova, V. W. , Richmond R. C., Woolf B. A. R., et al. 2021. “Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE‐MR Statement.” Jama 326, no. 16: 1614. 10.1001/jama.2021.18236. [DOI] [PubMed] [Google Scholar]
  47. Sugiyama, M. , Takao M., Hatsuta H., et al. 2013. “Increased Number of Astrocytes and Macrophages/Microglial Cells in the Corpus Callosum in Amyotrophic Lateral Sclerosis.” Neuropathology 33, no. 6: 591–599. 10.1111/neup.12027. [DOI] [PubMed] [Google Scholar]
  48. Suhre, K. , Arnold M., Bhagwat A. M., et al. 2017. “Connecting Genetic Risk to Disease End Points Through the Human Blood Plasma Proteome.” Nature Communications 8: 14357. 10.1038/ncomms14357. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Sun, B. B. , Maranville J. C., Peters J. E., et al. 2018. “Genomic Atlas of the Human Plasma Proteome.” Nature 558, no. 7708: 73–79. 10.1038/s41586-018-0175-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Szklarczyk, D. , Gable A. L., Lyon D., et al. 2019. “STRING v11: Protein‐Protein Association Networks With Increased Coverage, Supporting Functional Discovery in Genome‐Wide Experimental Datasets.” Nucleic Acids Research 47, no. D1: D607–D613. 10.1093/nar/gky1131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. van Rheenen, W. , van der Spek R. A. A., Bakker M. K., et al. 2021. “Common and Rare Variant Association Analyses in Amyotrophic Lateral Sclerosis Identify 15 Risk Loci With Distinct Genetic Architectures and Neuron‐Specific Biology.” Nature Genetics 53, no. 12: 1636–1648. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Võsa, U. , Claringbould A., Westra H.‐J., et al. 2021. “Large‐Scale Cis‐ and Trans‐eQTL Analyses Identify Thousands of Genetic Loci and Polygenic Scores That Regulate Blood Gene Expression.” Nature Genetics 53, no. 9: 1300–1310. 10.1038/s41588-021-00913-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wang, J. , Yu J.‐T., Tan M.‐S., Jiang T., and Tan L.. 2013. “Epigenetic Mechanisms in Alzheimer's Disease: Implications for Pathogenesis and Therapy.” Ageing Research Reviews 12, no. 4: 1024–1041. 10.1016/j.arr.2013.05.003. [DOI] [PubMed] [Google Scholar]
  54. Wang, Y. , Serricchio M., Jauregui M., et al. 2015. “Deubiquitinating Enzymes Regulate PARK2‐mediated Mitophagy.” Autophagy 11, no. 4: 595–606. 10.1080/15548627.2015.1034408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Wu, C.‐H. , Fallini C., Ticozzi N., et al. 2012. “Mutations in the Profilin 1 Gene Cause Familial Amyotrophic Lateral Sclerosis.” Nature 488, no. 7412: 499–503. 10.1038/nature11280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Wu, Y. , Zeng J., Zhang F., et al. 2018. “Integrative Analysis of Omics Summary Data Reveals Putative Mechanisms Underlying Complex Traits.” Nature Communications 9, no. 1: 918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Xu, X. , Shen D., Gao Y., et al. 2021. “A Perspective on Therapies for Amyotrophic Lateral Sclerosis: Can Disease Progression be Curbed?” Translational Neurodegeneration 10, no. 1: 29. 10.1186/s40035-021-00250-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Yang, C. , Farias F. H. G., Ibanez L., et al. 2021. “Genomic Atlas of the Proteome From Brain, CSF and Plasma Prioritizes Proteins Implicated in Neurological Disorders.” Nature Neuroscience 24: 1302–1312. 10.1038/s41593-021-00886-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Yao, C. , Chen G., Song C., et al. 2018. “Genome‐Wide Mapping of Plasma Protein QTLs Identifies Putatively Causal Genes and Pathways for Cardiovascular Disease.” Nature Communications 9, no. 1: 3268. 10.1038/s41467-018-05512-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Zheng, J. , Haberland V., Baird D., et al. 2020. “Phenome‐Wide Mendelian Randomization Mapping the Influence of the Plasma Proteome on Complex Diseases.” Nature Genetics 52, no. 10: 1122–1131. 10.1038/s41588-020-0682-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zhu, Z. , Zhang F., Hu H., et al. 2016. “Integration of Summary Data From GWAS and eQTL Studies Predicts Complex Trait Gene Targets.” Nature Genetics 48, no. 5: 481–487. 10.1038/ng.3538. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Materials: brb371366‐sup‐0001‐SuppMat.doc

BRB3-16-e71366-s001.doc (101.4KB, doc)

Supplementary Materials: brb371366‐sup‐0002‐SuppMat.xlsx

BRB3-16-e71366-s002.xlsx (1.3MB, xlsx)

Supplementary Materials: brb371366‐sup‐0003‐SuppMat.doc

BRB3-16-e71366-s003.doc (3.4MB, doc)

Data Availability Statement

The GWAS summary statistics supporting this research are available from the corresponding GWAS consortium. The main paper and Supporting Information present all data supporting our findings. The code or algorithm used to generate results in this study is available from the corresponding authors upon reasonable request.

Articles from Brain and Behavior are provided here courtesy of Wiley

RESOURCES