Role of single nucleotide polymorphisms (SNPs) in common migraine

Abstract

Single nucleotide polymorphisms (SNPs) may act as biological markers, as they can relate to the genes that are associated with various complex diseases such as heart diseases, diabetes, cancer, schizophrenia, blood pressure, migraine, and Alzheimer. These SNPs mostly locate within a gene or in a regulatory region near a gene and can affect the gene’s function to play a more direct role in disease. Hence, SNPs allow scientists to develop candidate drug therapy by evaluating an individual’s genetic makeup to develop a particular disease. Gene-gene interactions generally complicate migraine and its genetics and further gene-environmental interactions that often misguide the true defying causes of this disease. Due to its complex nature, it is difficult for scientists to reveal a complete list of SNPs or even all the genes that are related with the pathogenesis of this disease. Nowadays, much work has been done in this direction and new variants of migraine are being constantly identified. In this review article, the role of various SNPs reported to be disease-associated in published migraine GWAS has been discussed. To understand the molecular mechanisms of migraine attack by identifying new genetic variants of migraine can be a key to develop new therapeutic strategies in the future.

Introduction

Migraine is a common, chronic, polygenic neurovascular disorder in which both genetic and environmental factors play their roles. This disease affects about 15% of the population overall. The prevalence of migraine in women (18%) is higher when compared with men (6%), with attacks lasting from a few hours to some days [1]. The higher prevalence in women is typically attributed to hormonal fluctuations especially estrogen. Migraines typically begin during puberty or between the ages of 35 and 45 years [2, 3]. Migraine is a disabling disorder, which reduces the quality of life and significantly affects the ability to work and moreover results in higher cost and greater health care resource utilization [4,5,6]. Migraine can be classified into two types on the basis of aura: migraine without aura (MO) and migraine with aura (MA) and up to one-third of patients experience an aura prior or during the headache attacks. In migraine with aura (MA), patients can show transient facial neurological symptoms that mostly precede or in some cases accompany the headache. A premonitory phase having symptoms of depression, cravings for particular foods, fatigue, hyper or hypoactivity along with repetitive yawning, and neck stiffness is reported, which occur hours or days before the headache, and a headache resolution phase [7]. Migraine without aura (MO) is characterized by headache with specific features and associated symptoms such as unilateral location, pulsating quality, moderate, or severe pain intensity. During headache nausea, vomiting, photophobia, and phonophobia can occur [8]. The main criteria for the diagnosis of migraine with and without aura by the International Headache Society in 2013 [9] are given in Table 1.

Table 1 International classification of headache disorders 3rd edition [9]

Migraine and its genetics

Migraine has a direct impact on the lives of millions of people worldwide, and for most of them, there is still no effective treatment. Nowadays, genome-wide association studies have emerged as an important mean to uncover the genetic variants of complex disorders such as migraine. Whole exome sequencing is potentially an effective tool to identify coding variants underlying human diseases like migraine. In several GWAS, common SNPs related with migraine have proved to be very useful. These SNPs even help in understanding the role of the glutamatergic system, sensitization of nociceptive nerve endings, and cortical spreading depression in the pathogenesis of migraine. Despite this fact that GWAS have revealed new candidate genes and various SNPs related with them that could be the reason for migraine pathogenesis, still, the outcomes from previous findings have not brought us closer to understand its molecular and genetic bases. The reason behind this may be the multi-genetic and multi-factorial nature of this disorder. Further, more studies should be done for building a gene profile of the complex common migraine disorder which may provide new drug targets.

To better understand the genetics of migraine, approaches to study candidate genes and linkage studies are also widely used. For a large number of genes and its related SNPs, various replication studies have been done and these studies either confirmed or disproved the association. A criterion using the combination of physiological functions with genomic associations is widely used to find out the most possible variants related with common migraine pathogenesis. On this basis, three different groups of genes including neurological, hormonal, or vascular pathways have been identified for this disorder [10, 11]. Under the neurological group, genes that are involved with expression or control of ion channels (CACNA1A, KCCN3, Sodium-Potassium-ATPase), genes responsible for the synthesis/release and binding of neuropeptides (CGRP protein, glutamate), and the genes with functional effects in dopamine and serotonergic pathways have been reported. The second group of vascular genes is mostly functional in the regulation of blood pressure, endothelial cell expression, vasoconstriction, and vasodilation (MTHFR, ACE). Vascular genes are responsible for migraine particularly migraine with aura overlap with the genes that have a role in elevated risk of stroke and heart disease as a significant co-morbidity with risk for stroke and depression [12]. The third group of genes is typically related with women that can describe the sex-biased distribution of migraine. Genes that can regulate hormonal pathways in females and menstrual migraine include mostly those which can control estrogen and progesterone (ESR1, PGR) levels [13].

Migraine and its SNPs

Genome-wide association studies of migraine (MA and MO), family clustering, and twin studies have identified various migraine-associated common genetic variants [14, 15]. The first large migraine meta-analysis was performed by Anttila et al. [16] that had 29 studies with 23,285 migraineurs and 95,425 controls. In that study, 12 single nucleotide polymorphisms (SNPs) were revealed that were related with migraine susceptibility. Recently, a GWAS [17] had revealed 38 genomic variants mostly found in humans (with > 5% allele frequency) which can affect migraine risk. Previous studies [18, 19] have found additional variants that make the total number of genomic regions related with migraine as high as 47. These variants hint towards a remarkable progress in this direction and provide causes for optimism that the key mechanisms for this disorder can finally be understood. Thus, migraine-related GWAS and reported meta-analyses found different SNPs for this disease, which can be further classified into five pathways [20] on the bases of function:

• Genes involved in glutamatergic neurotransmission—rs1835740 in MTDH, rs11172113 in LRP1, and rs3790455 in MEF2D

• Genes for synapse development and neuroplasticity—rs6478241 in ASTN2, rs1320832 in FHL5

• Genes for pain sensitivity—rs10166942 in TRPM8

• Genes for metalloproteinases—rs10504861 near to MMP16, rs10915437 near to AJAP1, rs12134493 near to TSPAN2

• Genes responsible for vascular system and metabolism—rs4379368 in C7orf10, rs2651899 in PRDM16, rs9349379 in PHACTR1, rs7640543 near to TGFBR2

In this review, the most common genes and their variants involved as biomarkers for migraine have been discussed and listed in Table 2.

Table 2 Variants identified in common form of migraine and replicated in various studies

rs1835740

This SNP, present on 8q22.1 chromosomal region, is located in a 27 kb haplotype block between genes MTDH (metadherin, astrocyte elevated gene 1) and PGCP (plasma glutamate carboxypeptidase). This variant is mainly responsible for glutamate homeostasis and best known for harboring different genes involved in glutamate metabolism and transport. This SNP regulates the expression of MTDH gene and hence indirectly regulates the expression of the glutamate transporter gene SLC1A2 which is also known as EAAT2 or GLT-1 that further encodes a major glutamate transporter protein in the brain. It has been reported in various studies that by increasing the release of glutamate or reduction in uptake of glutamate increase the risk of migraine attacks [21,22,23]. In astrocytes, MTDH downregulates GLT-1, the gene encoding the major glutamate transporter protein. SNPs in MTDH gene thus can contribute towards understanding the role of glutamate in this disorder. Higher levels of glutamate increase the susceptibility towards cortical spreading depression. Glutamate is one of the major excitatory neurotransmitters in the CNS and therefore has an important role to play in the mediation of excitatory synaptic transmission [24,25,26]. In the migraine pain pathway, the anatomic structures contain glutamate-positive neurons which include the trigeminal ganglion (TG), thalamus, and the trigemino-cervical complex (TCC) [27, 28]. Migraineurs have higher levels of glutamate and glutamine in their cerebrospinal fluids on comparing with controls and also shown a positive correlation between glutamate levels and mean headache scores [29, 30]. This SNP in MTDH gene increases the interest in glutamatergic pathways in migraine but can explain only a small fraction of the overall genetic variance.

rs1801133 and rs1801131

The MTHFR gene, on chromosome 1p36, has 11 exons of ~ 17 kb in size [31]. A common SNP (C677T) in MTHFR gene encodes for amino acid valine (Val) instead of alanine (Ala) at position 222. Migraineurs with the valine instead of alanine amino acid due to this SNP may exhibit a significantly reduced MTHFR enzyme activity which is 30% in the Val/Val homozygous state as compared to 65% in the Ala/Val heterozygote [32]. If the patients further have a low intake of dietary folate, then this reduced activity of MTHFR gene can result in mild elevation in plasma homocysteine levels [33]. Mild hyper homo-cysteinemia has been associated with an increased risk of cardiovascular diseases (CVD) that include cerebral, coronary, and peripheral vascular diseases [34]. It is reported that homocysteine may act like an excitatory amino acid in migraine pathophysiology. Homocysteine can cause vasodilation or temporary thrombosis of cerebral blood vessels and thus leading to reduce the supply of oxygen into the brain [35, 36]. Interestingly, an increased risk of ischemic stroke is also associated with the migraine disorder itself, particularly with MA in some studies [37]. Further, cortical spreading depression (CSD) in MA, with changes in cerebral blood flow and headache, can also occur along with a stroke episode [38]. On the basis of the co-morbidity between migraine and stroke and with the role of homocysteine in disturbing the cerebro-circulatory system, it is possible that the MTHFR C677T SNP may represent an important genetic variant for the vascular pathophysiologies of migraine representing both neurovascular conditions.

Similarly, rs1801131 and A1298C SNP can cause a point mutation in MTHFR gene at position 1298 and are responsible for encoding glutamine instead of alanine residue [39]. This polymorphism A1298C may affect the enzyme activity of MTHFR as well as folate concentrations, although less than those reported in SNP C677T. Additionally, A1298C SNP is located within the regulatory domain of MTHFR gene. In the previous studies, it has been shown that individuals homozygous only for the A1298C SNP do not report higher homocysteine levels in serum than that of controls but those who are heterozygous both for the A1298C and C677T SNPs tend to have a similar biochemical profile with C677T homozygotes, showing an increase in homocysteine levels and a decrease in folate concentrations in serum [40]. Hence, the investigation of MTHFR polymorphisms and their effects on metabolic pathway of folate has been considered as an interesting topic for research in recent years.

rs11172113

rs11172113, a SNP on chromosome 12q13.3, is located in the first intron of LRP1 gene. This SNP is responsible for the lipoprotein receptor LRP1 [41], which further showed interaction with neuronal glutamate receptors in the brain. Hence, this variant also shares a common link with the glutamate pathway via interaction of LRP1 receptors with NMDA glutamate receptors in migraine. Both previous findings [15, 16] have discussed the role of glutamate in the pathophysiology of migraine and integrated well with various pharmacological approaches that can further target glutamate receptors for migraine treatment [26].

rs10166942

rs10166942, a SNP on chromosome 2q37, is located close to the gene TRPM8 at the transcription start site. TRPM8 gene encodes a sensor protein for a cold and cold-induced burning pain [42] that is primarily expressed in sensory neurons and the dorsal root ganglion in the brain [43]. It has been investigated that TRPM8 can be a target in various animal models with neuropathic pain. As both migraine and neuropathic pain share some features, the role for TRPM8 variant in this disease as well as its relation between both pain syndromes can be an interesting topic in future for research scientists [44, 45].

rs2651899

rs2651899 SNP, located on chromosome 1p36.32, is in the first intron of PRDM16 (PR domain containing 16) gene. This gene codes for a zinc finger transcription factor, supposed to be associated with differentiation of brown adipose tissue along with repression of TGF-β (transforming growth factor beta) signaling [46, 47], but the role of this gene is still unclear in migraine. More case-control studies with a reasonable number of individuals from different geographical regions are needed to explore the relationship of this SNP with migraine.

rs4379368

rs4379368 SNP, on chromosome 7p14, is found within an intron in the transcript of the C7orf10 gene and encodes enzyme succinic HMG coenzyme A transferase. It has been reported that mutations in c7orf10 can cause glutaricaciduria type III which is a rare metabolic abnormality in individuals leading to persistent excretion of glutaric acid [48]. Different SNPs in this gene may differentially control the transcription and/or RNA processing and as a result affecting the risk of migraine [49].

rs10504861

rs10504861, a SNP located on chromosome 8q21, is 200 kb away from the matrix metalloproteinase MMP16 gene. Metalloproteinase is an extended family of protease enzymes that are responsible for the breakdown of extracellular matrix in various normal physiological processes. This SNP has been linked with an elevated incidence of migraine without aura (MO) in genome-wide association studies (GWAS), as the protein encoded by MMP16 gene is associated with the cleavage of the low-density lipoprotein receptor protein (LRP1) [50]. A SNP within LRP1 (rs11172113) has already been revealed to be associated with migraine in a genome-wide association study (GWAS). Various studies have recently shown that MT3-MMP encoded by this gene is supposed to cause basal NgR1 (Nogo-66 receptor) shedding in cortical neurons and to increase axonal and synaptic plasticity [51]. Although for the mechanism showing association of MMP16 to migraines, no conclusive evidence exists so far, these two findings may offer plausible reasons for the biological implication of rs10504861 SNP in migraine pathology [14, 52]. More case-control studies are needed to associate this SNP with migraine.

rs10915437

The SNP rs10915437, on chromosome 1p36, is present approximately 500 kb telomeric from the adherens junction-associated protein 1 (AJAP1) gene and almost 300 kb centromeric from a gene cluster that encodes the apoptosis-related proteins DFFB and TP73 as well as centrosomal protein CEP104. AJAP1 has been related with tumor invasion and supposed to control the regulation of metalloproteinase activity and is expressed in the brain [53]. Hence, indirectly, this SNP may link with the migraine pathology and much work is required to relate this association.

rs12134493

The SNP rs12134493 is on chromosome 1p13 and located 87 kb 5′upstream of the tetraspanin 2 (TSPAN2) gene. This gene encodes a four trans-membrane protein which causes the modulation of development, activation, growth, and movement of cells via the mediation of signal transduction. Esserlind et al. reported another SNP (rs2078371) that is located near gene TSPAN2 in a GWAS and had shown the genome-wide significance of association with migraine in Danish and Icelandic populations and hence supported the indirect role of this SNP in migraine [14].

rs13208321

The SNP rs13208321, present on chromosome 6q16, is within the four and a half LIM domains protein 5 (FHL5) gene. This SNP is reported to influence protein function in cells. FHL5 gene can encode a transcription factor which further regulates cAMP-responsive elements CREM and CREB [54], and those have a role to play in synaptic plasticity associated with migraine and memory formation [55, 56].

Another study by Freilinger et al. [57] revealed associations of migraine with four new SNPs and also replicated the findings of previously related associations with TRMP8 and LRP1 genes. These new variants with the strongest association at the respective locus were rs3790455 (MEF2D) on chromosome1q22, rs7640543 (TGFBR2) on chromosome 3p24.1, rs9349379 (PHACTR1) on chromosome6p21, and rs6478241 (ASTN2) on chromosome 9q33.1. Myocyte enhancer factor 2D (MEF2D), a transcription factor, is found to promote the survival of newly formed neurons in the brain [58]. MEF2D is also responsible for influencing the number of excitatory synapses in an activity-dependent manner, thus indirectly neuronal excitability, which is a common factor for a migraine candidate gene. Phosphatase and actin regulator 1 (PHACTR1) gene is supposed to regulate the activity of PP1 (protein phosphatase 1), which further influences synaptic activity and morphology in migraine [59, 60]. Transforming growth factor β receptor 2(TGFBR2) is a serine-threonine kinase enzyme that is active in proliferation, and Astrotactin 2 (ASTN2) gene codes for a protein which is supposed to influence neuronal migration [61, 62]. In endothelial samples, only two gene products (TGFBR2 and MEF2D) exhibit moderate or greater expression.

In another study, which mainly reflects the known co-morbidity between migraine and cardiovascular diseases [63], it is reported that two of the 12 most proximal genes (TGFBR2 and PHACTR1) have also been related with cardiovascular traits: TGFBR2 mutations have been found to cause monogenic Marfan’s syndrome [64] and to be present in abdominal aortic aneurysms [65], while PHACTR1 is linked with early onset myocardial infarction [66]. Whereas TSPAN2 [67], MEF2D [68], TRPM8 [69], TGFBR2 [70], PHACTR1 [71], MMP16 [72], ASTN2 [62], and LRP1 [73] genes have been found to be functional in synaptic formation or its regulation. Mutations in PRDM16 have been linked to oxidative stress response in brain cells and those of AJAP1 in maintaining tissue borders.

Conclusions

Migraine is a common cause of headache, and early diagnosis and prompt treatment of migraine enhance the quality of life and prevent conversion of episodic migraine to chronic migraine. Single nucleotide polymorphism is a common form of allelic variation in human disease genes. SNPs may act as genomic biomarkers in search for variants that influence susceptibility to migraine. Diagnosis of migraine is still based on questionnaires and neuro-imaging mostly which should be replaced with various biomarkers. These biomarkers could easily be detected from saliva or blood samples of migraineurs by using biochemical or molecular methods. Various SNPs discussed in this review may reveal the phenotypic characters of migraineurs and have implications for functions of proteins coded by mutated genes. Each SNP may be genetically affected to a small group of migraineurs, but each one provides important information about migraine mechanism which can be further applicable to a wider population. These variants may further help in monitoring candidate drug response and in finding new targets for migraine pharmacotherapy. Hence, a better understanding of migraine genetics with rapid whole-genome sequencing in an individual could lead to personalize medicines in migraineurs.