New Links Between Blood Proteins and Migraine

A genetic study reveals that five of the proteins show a causal relationship with migraine risk.

By Neil Andrews

Numerous studies of migraine show that particular genetic variants are associated with an increased risk of the disorder. But a basic tenet of genetics is that a single gene can affect multiple traits, a phenomenon known as pleiotropy. A new study from the International Headache Genetics Consortium (IHGC) now examines pleiotropy in the context of migraine, with a focus on the relationship between blood proteins and migraine.

The research, led by Dale Nyholt, chair of the IHGC and a professor and geneticist at Queensland University of Technology, Brisbane, Australia, and Hamzeh Tanha, his PhD student, uncovers a link between the risk of migraine and levels of dozens of blood proteins. The investigators report that five of those proteins have a causal effect on migraine, including a protein called DKK1 that has previously been linked to Alzheimer's disease and is a target for treatment of that condition.

Along with illuminating possible therapeutic avenues to treat migraine, the new work provides novel insight into the complex biology of migraine.

“I think this work moves the field forward and will spark new research that could give us an even better understanding of migraine,” said Thomas Folkmann Hansen, Copenhagen University, Denmark.

Hansen was particularly impressed with the authors' use of a statistical tool, called a latent causal variable model, that allowed the researchers to determine causal relationships between the blood proteins and risk of migraine.

“The way they approach causality is more thorough than in previous studies, and I also think it's more objective. It's a very fresh take on how to determine causality,” said Hansen, who is on the executive committee of the IHGC but was not involved with the current work.

The research appeared May 11, 2022, in Nature Communications.

Multiple levels

The new study builds upon previous work from the same group. In November of last year, the authors reported a genetic overlap between migraine and blood metabolites. A handful of those metabolites, primarily lipids and lipid-like molecules, were causally related to migraine.

In their current investigation, the investigators shifted their focus from blood metabolites to blood proteins. But the goal was similar: to identify pleiotropic genes with effects on both migraine risk and, in this case, blood proteins. As in their earlier work, the authors were also interested in causality – here, whether a causal relationship existed between the blood proteins and the risk of migraine.

To address all of these questions, the researchers looked for pleiotropic effects at three different levels: at the level of single nucleotide polymorphisms, known as SNPs (a type of DNA sequence variation at a single nucleotide, the basic DNA building block), at the gene level, and across the entire genome.

“When you're looking for relationships across traits, there could be differences in whether the actual DNA variant – the SNP – is associated with both traits,” said Nyholt. “But it could also be that low or high expression of a gene is actually the thing that is in common between those two traits. ?

“And then you can look across the whole genome – you can look at all the genetic factors that, on the whole, are causing an increase in risk for migraine, and at all the genetic factors that, on the whole, seem to be related to an increase or a decrease in blood protein levels. So there are multiple levels, and I think it's important to look at those multiple levels,” according to Nyholt.

Searching for pleiotropy

With their strategy in hand, the investigators got to work, using data from a 2016 migraine genome-wide association study (GWAS) to do their analysis. (In a GWAS, the entire genome is scanned to identify SNPs associated with migraine risk; see MSC related news story).

Starting at the whole-genome level, the researchers found that increased risk of migraine was correlated with increased levels of six blood proteins. Meanwhile, increased risk of migraine was correlated with decreased levels of seven blood proteins.

The group next moved to the gene level, looking for pleiotropy at over 18,000 protein-coding genes. This led to the identification of pleiotropic genes that influenced both migraine risk and levels of 15 blood proteins. Finally, the investigators examined pleiotropy at the SNP level. They found that pleiotropic SNPs influenced migraine risk and levels of 36 blood proteins.

Of note, the genes for the majority of blood proteins identified in these analyses were located far from the genetic variants identified by migraine GWAS. This means that the analysis of blood proteins in conditions like migraine can provide information that GWAS may miss.

“Hit the ground running”

The investigators were then interested in learning if any of the dozens of blood proteins they identified played a causal role in migraine. For this purpose, they used the latent causal variable model, a statistical tool that has not been used much in genetic investigations of migraine so far.

The model allowed the researchers to determine whether a correlation between blood protein levels and migraine occurred simply because each trait has shared genetic influences, or occurred because there was a causal effect, with specific blood proteins contributing to migraine.

The results showed that blood levels of five proteins had a causal effect on migraine. One of those proteins that especially caught the eye of the study authors is DKK1. This protein is an inhibitor of a signaling pathway, called the Wnt signaling pathway, which had previously been linked to neuropathic pain as well as to familial migraine, and also to Alzheimer's disease (AD). In the case of AD, increased levels of DKK1 have been found in the brains of AD patients. DKK1 is also under investigation as a drug target to treat AD.

In the current study, higher levels of DKK1 were causally linked to migraine, just as higher levels of DKK1 have been linked to AD. This is encouraging to Nyholt from a therapeutic perspective.

“The DKK1 protein that we've found associated with migraine is in the same direction – increased levels – as with Alzheimer's disease. So there's high probability that if you take those drugs that are being developed for Alzheimer's disease, which work by the same action, then those drugs could have the same effect of upregulating the Wnt signaling pathway during migraine, making people less sensitive to migraine pain,” said Nyholt. ?

“So essentially you could take a drug that could be repurposed – you could hit the ground running,” Nyholt continued. “If it's shown to be safe with Alzheimer's disease, for example, then you could fast-track it into some type of clinical trial where you could then see if there's a benefit to reduce an individual's migraine pain.”

Nyholt added that he hopes the current study results will spur more basic science investigations of DKK1/Wnt signaling in the context of migraine.

Looking ahead

For his part, Hansen was particularly intrigued by another blood protein shown to have a causal relationship with migraine, called platelet-derived growth factor subunit B (PDGFB). This protein has previously been reported to play a role in inflammatory and neuropathic pain in rats.

Considering its prior link to pain, “I think PDGFB is worth revisiting; I would dig more into it,” Hansen said. He added that it would be interesting to study how the blood proteins identified in the study act during and after a migraine attack. “That could be a natural, very obvious next step.” The role that the blood proteins play in different subtypes of migraine is another interesting aspect for future studies to consider, Hansen said.

As for Nyholt, he is interested in doing another study of blood proteins and migraine, but using data from the most recent migraine GWAS published earlier this year, rather than from the 2016 GWAS upon which the current study is based. “We want to repeat what we've just published but using that more powerful, and more informative polygenic risk profile,” he said.

In the end, Nyholt said, it's all about “trying to pull everything together using genetics as the master key, to understand the relationships between all these different biological measures, how they all relate and interact with each other, to start to provide some really important biological insight into what's happening in migraine.”

Neil Andrews is a science journalist and Executive Editor of the Migraine Science Collaborative.

Reference: Genetic analyses identify pleiotropy and causality for blood proteins and highlight Wnt/β-catenin signalling in migraine. Tanha et al. Nat Commun. 2022 May 11;13(1):2593.

This article was originally published on the Migraine Science Collaborative.