“…Also, as above one is not meaningful (and likely due to small sample size noise), we included GWASs with 0 < resulting in 362 blood protein GWASs (362/4625 = 7.8%; 4 × 10 –9 < ). Compared to our study of blood metabolome where 41.67% (405/972) of metabolites showed significant 12 this finding also suggests that blood protein levels tend to be more attributable to local heritability than polygenicity (many trans -pQTLs spread across the genome). Fig.…”
Section: Resultscontrasting
confidence: 71%
“…Comparing the heritability results from the present study to our recent study of blood metabolome 12 suggests a lower polygenicity for blood proteome than blood metabolome. First, polygenic SNP heritability was not associated with the number of included SNPs for the 4625 blood proteins, whereas it was significantly associated in our study of 972 blood metabolites.…”
Section: Discussionsupporting
confidence: 67%
“…Only six proteins showed evidence for pleiotropy at multiple levels, including genome-wide and genes levels (MMP7), genome-wide and SNPs levels (KITLG and VEGFA), and genes and SNPs levels (MMP1, RBBP5 and B3GNT2). Therefore, as reported by us 12 and others 17 , 20 , findings from pleiotropy analyses at different levels are mostly distinct and thus complementary. Our pleiotropy at the genome-wide level, a method similar to LDSC genetic correlation, provides a direction for the relationship (a signed correlation).…”
Section: Discussionsupporting
confidence: 57%
“…We observed that Z scores of are correlated with sample size ( r = 0.30, 95% confidence interval [CI] = 0.27–0.32, P < 2 × 10 –16 ), but not the number of included SNPs in LDSC analysis ( P = 0.87). Conversely, we recently showed that estimated for 972 blood metabolites is significantly associated with the number of included SNPs 12 , suggesting a lower (genome-wide) polygenicity for blood proteome than blood metabolome. This might highlight that a larger fraction of blood proteome heritability is attributable to local signals when the association is either at or near the gene locus that encodes the tested protein ( cis -pQTLs) or is far from the encoding gene ( trans -pQTLs).…”
Migraine is a common complex disorder with a significant polygenic SNP heritability ($${h}_{{SNP}}^{2}$$
h
S
N
P
2
). Here we utilise genome-wide association study (GWAS) summary statistics to study pleiotropy between blood proteins and migraine under the polygenic model. We estimate $${h}_{{SNP}}^{2}$$
h
S
N
P
2
for 4625 blood protein GWASs and identify 325 unique proteins with a significant $${h}_{{SNP}}^{2}$$
h
S
N
P
2
for use in subsequent genetic analyses. Pleiotropy analyses link 58 blood proteins to migraine risk at genome-wide, gene and/or single-nucleotide polymorphism levels—suggesting shared genetic influences or causal relationships. Notably, the identified proteins are largely distinct from migraine GWAS loci. We show that higher levels of DKK1 and PDGFB, and lower levels of FARS2, GSTA4 and CHIC2 proteins have a significant causal effect on migraine. The risk-increasing effect of DKK1 is particularly interesting—indicating a role for downregulation of β-catenin-dependent Wnt signalling in migraine risk, suggesting Wnt activators that restore Wnt/β-catenin signalling in brain could represent therapeutic tools against migraine.
“…Also, as above one is not meaningful (and likely due to small sample size noise), we included GWASs with 0 < resulting in 362 blood protein GWASs (362/4625 = 7.8%; 4 × 10 –9 < ). Compared to our study of blood metabolome where 41.67% (405/972) of metabolites showed significant 12 this finding also suggests that blood protein levels tend to be more attributable to local heritability than polygenicity (many trans -pQTLs spread across the genome). Fig.…”
Section: Resultscontrasting
confidence: 71%
“…Comparing the heritability results from the present study to our recent study of blood metabolome 12 suggests a lower polygenicity for blood proteome than blood metabolome. First, polygenic SNP heritability was not associated with the number of included SNPs for the 4625 blood proteins, whereas it was significantly associated in our study of 972 blood metabolites.…”
Section: Discussionsupporting
confidence: 67%
“…Only six proteins showed evidence for pleiotropy at multiple levels, including genome-wide and genes levels (MMP7), genome-wide and SNPs levels (KITLG and VEGFA), and genes and SNPs levels (MMP1, RBBP5 and B3GNT2). Therefore, as reported by us 12 and others 17 , 20 , findings from pleiotropy analyses at different levels are mostly distinct and thus complementary. Our pleiotropy at the genome-wide level, a method similar to LDSC genetic correlation, provides a direction for the relationship (a signed correlation).…”
Section: Discussionsupporting
confidence: 57%
“…We observed that Z scores of are correlated with sample size ( r = 0.30, 95% confidence interval [CI] = 0.27–0.32, P < 2 × 10 –16 ), but not the number of included SNPs in LDSC analysis ( P = 0.87). Conversely, we recently showed that estimated for 972 blood metabolites is significantly associated with the number of included SNPs 12 , suggesting a lower (genome-wide) polygenicity for blood proteome than blood metabolome. This might highlight that a larger fraction of blood proteome heritability is attributable to local signals when the association is either at or near the gene locus that encodes the tested protein ( cis -pQTLs) or is far from the encoding gene ( trans -pQTLs).…”
Migraine is a common complex disorder with a significant polygenic SNP heritability ($${h}_{{SNP}}^{2}$$
h
S
N
P
2
). Here we utilise genome-wide association study (GWAS) summary statistics to study pleiotropy between blood proteins and migraine under the polygenic model. We estimate $${h}_{{SNP}}^{2}$$
h
S
N
P
2
for 4625 blood protein GWASs and identify 325 unique proteins with a significant $${h}_{{SNP}}^{2}$$
h
S
N
P
2
for use in subsequent genetic analyses. Pleiotropy analyses link 58 blood proteins to migraine risk at genome-wide, gene and/or single-nucleotide polymorphism levels—suggesting shared genetic influences or causal relationships. Notably, the identified proteins are largely distinct from migraine GWAS loci. We show that higher levels of DKK1 and PDGFB, and lower levels of FARS2, GSTA4 and CHIC2 proteins have a significant causal effect on migraine. The risk-increasing effect of DKK1 is particularly interesting—indicating a role for downregulation of β-catenin-dependent Wnt signalling in migraine risk, suggesting Wnt activators that restore Wnt/β-catenin signalling in brain could represent therapeutic tools against migraine.
“…One study reported abnormally low serum levels of ALA in the majority of migraine patients [ 55 ]. Another study in which migraine GWAS data were investigated with respect to genetically associated blood metabolic traits, DHA and DPA were identified as fatty acids that were genetically associated with a higher risk for migraine [ 56 ]. Of note, lower dietary intake of DHA and EPA are associated with a higher attack frequency in migraine patients [ 18 ]; vice versa, increasing dietary intake of EPA and DHA in chronic and episodic migraine patients resulted in a reduction in headache frequency and severity [ 57 ] and physical pain [ 58 ], underscoring the involvement of these lipids in migraine pathophysiology.…”
Metabolite levels in peripheral body fluids can correlate with attack features in migraine patients, which underscores the potential of plasma metabolites as possible disease biomarkers. Migraine headache can be preceded by an aura that is caused by cortical spreading depolarization (CSD), a transient wave of neuroglial depolarization. We previously identified plasma amino acid changes after CSD in familial hemiplegic migraine type 1 (FHM1) mutant mice that exhibit increased neuronal excitability and various migraine-related features. Here, we aimed to uncover lipid metabolic pathways affected by CSD, guided by findings on the involvement of lipids in hemiplegic migraine pathophysiology. Using targeted lipidomic analysis, we studied plasma lipid metabolite levels at different time points after CSD in wild-type and FHM1 mutant mice. Following CSD, the most prominent plasma lipid change concerned a transient increase in PGD2, which lasted longer in mutant mice. In wild-type mice only, levels of anti-inflammatory lipid mediators DPAn-3, EPA, ALA, and DHA were elevated 24 h following CSD compared to Sham-treated animals. Given the role of PGs and neuroinflammation in migraine pathophysiology, our findings underscore the potential of monitoring peripheral changes in lipids to gain insight in central brain mechanisms.
The co-occurrence of migraine and glycemic traits has long been reported in observational epidemiological studies, but it has remained unknown how they are linked genetically. We used large-scale GWAS summary statistics on migraine, headache, and nine glycemic traits in European populations to perform cross-trait analyses to estimate genetic correlation, identify shared genomic regions, loci, genes, and pathways, and test for causal relationships. Out of the nine glycemic traits, significant genetic correlation was observed for fasting insulin (FI) and glycated haemoglobin (HbA1c) with both migraine and headache, while 2-h glucose was genetically correlated only with migraine. Among 1703 linkage disequilibrium (LD) independent regions of the genome, we found pleiotropic regions between migraine and FI, fasting glucose (FG), and HbA1c, and pleiotropic regions between headache and glucose, FI, HbA1c, and fasting proinsulin. Cross-trait GWAS meta-analysis with glycemic traits, identified six novel genome-wide significant lead SNPs with migraine, and six novel lead SNPs with headache (Pmeta < 5.0 × 10–8 and Psingle-trait < 1 × 10–4), all of which were LD-independent. Genes with a nominal gene-based association (Pgene ≤ 0.05) were significantly enriched (overlapping) across the migraine, headache, and glycemic traits. Mendelian randomisation analyses produced intriguing, but inconsistent, evidence for a causal relationship between migraine and headache with multiple glycemic traits; and consistent evidence suggesting increased fasting proinsulin levels may causally decrease the risk of headache. Our findings indicate that migraine, headache, and glycemic traits share a common genetic etiology and provide genetic insights into the molecular mechanisms contributing to their comorbid relationship.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
