Abstract

   Chronic widespread musculoskeletal pain (CWP), a core symptom of
   fibromyalgia, is a complex condition with an unclear pathogenesis.
   Despite findings from genome-wide association studies (GWAS),
   translating genetic risk variants into therapeutic targets remains
   challenging due to limited understanding of their functional roles in
   CWP. In this study, we developed an integrative analytical pipeline to
   efficiently link genetic associations with novel risk genes for CWP. By
   combining high-throughput data from multiple sources, we integrated
   proteome-wide association studies (PWAS), transcriptome-wide
   association studies (TWAS), summary-based Mendelian randomization
   (SMR), and Bayesian co-localization analyses. This approach allowed us
   to prioritize genes that may increase the risk of CWP by altering
   protein abundance and gene expression in the brain. We further examined
   these genes using various advanced methodologies to validate their
   significance. We identified eight genes expressed in the brain, whose
   protein levels were associated with CWP. Four of these genes were
   confirmed through a subsequent PWAS, while three showed associations
   with cis-regulated mRNA expression. Only four genes, GMPPB, COMT, NME1,
   and GPX1, passed SMR and Bayesian colocalization analyses. These genes
   were expressed in pain-related brain regions and showed selective
   expression in oligodendrocytes, microglia, dopaminergic neurons, and
   interneurons. Additionally, COMT and NME1 were identified as potential
   druggable targets using the DGIdb and DrugBank databases. Our findings
   suggest that GMPPB, COMT, NME1, and GPX1 are potential risk genes for
   CWP, offering new insights into the molecular mechanisms underlying the
   condition. These genes represent promising targets for future research
   and therapeutic intervention development.

Supplementary Information

   The online version contains supplementary material available at
   10.1038/s41598-025-04379-5.

   Keywords: Chronic widespread musculoskeletal pain, Brain proteins,
   Proteome-wide association study, Transcriptome-wide association study

   Subject terms: Genetics, Neurology, Rheumatology

Introduction

   Chronic widespread musculoskeletal pain (CWP) is a hallmark symptom of
   fibromyalgia and a common, complex trait influenced by both genetic and
   environmental factors^[30]1,[31]2. The prevalence of CWP ranges from
   11.4 to 24% in the global population^[32]3, contributing to excess
   mortality and substantial societal costs. Unfortunately, current
   treatments for CWP are ineffective for a large subset of individuals,
   highlighting the urgent need for a deeper understanding of its
   pathogenesis to develop more effective therapies.

   Recent genome-wide association studies (GWAS) have identified numerous
   genetic loci associated with CWP^[33]4–[34]6. However, translating
   these findings into clinical interventions remains challenging due to
   the limited understanding of how these genetic variants contribute to
   the disease. To bridge this gap, it is critical to investigate the
   functional roles of these variants and identify proteins that mediate
   their effects on CWP. Proteins are attractive candidates for biomarkers
   and drug targets because they are the final products of gene expression
   and play central roles in cellular function and biological
   processes^[35]7,[36]8. Moreover, gene regulation occurs at the
   post-transcriptional, translational, and post-translational levels,
   underscoring the importance of directly examining proteins.

   Given the pivotal role of proteins in disease pathogenesis, various
   statistical genetics methods have been developed to identify
   disease-associated proteins. One such method is proteome-wide
   association studies (PWAS), which link protein levels to disease
   phenotypes^[37]9–[38]12. In addition, transcriptome-wide association
   studies (TWAS) have been used to explore the relationship between gene
   expression and disease^[39]13. Methods like summary-based Mendelian
   randomization (SMR) analysis^[40]14 and Bayesian co-localization
   (COLOC)^[41]15 have also emerged as powerful tools to determine whether
   proteins mediate the association between genetic variants and disease.
   These integrative approaches not only offer valuable insights into
   disease biology but also facilitate the identification of promising
   drug targets, potentially accelerating drug discovery and improving the
   success rates of clinical trials.

   In the present study, we apply these analytic approaches to identify
   potential causal brain proteins involved in CWP pathogenesis by
   integrating high-throughput brain proteomics with genetic data. The
   overall analytical pipeline is summarized in Fig. [42]1. We performed a
   PWAS using CWP GWAS data and reference human brain proteomes to
   identify potential CWP risk genes in both discovery and confirmatory
   datasets. To explore these genes from both transcriptomic and proteomic
   perspectives, we conducted a TWAS for CWP using reference human brain
   transcriptomes from the Common Mind Consortium (CMC). The identified
   genes underwent rigorous evaluation using a range of methodologies,
   including SMR and co-localization. Finally, we carried out GO-KEGG
   pathway enrichment analyses, as well as cell-type-specific and
   brain-region expression studies, alongside druggability evaluations.
   These analyses collectively shed light on the roles and potential
   mechanistic pathways of the identified genes in the context of CWP.

Fig. 1.

   [43]Fig. 1
   [44]Open in a new tab

   The flowchart of this study. CWP chronic widespread musculoskeletal
   pain, FDR false discovery rate, GWAS genome-wide association studies,
   pQTL protein quantitative trait locus, PWAS proteome-wide association
   study, SMR summary-based Mendelian randomization, TWAS
   transcriptome-wide association study.

Materials and methods

Data sources

Human brain protein abundance references for discovery and confirmation PWAS