Genetic basis of stroke: An overview
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.63780
Source of Support: None, Conflict of Interest: None
Stroke or "brain attack" is a complex disease caused by a combination of multiple risk factors. It has major social and economic consequences. Various epidemiological studies in families and twins have revealed that there is a genetic component to stroke risk. Stroke may be the outcome of single gene disorders or more commonly, a polygenic multifactorial disease. Mutations in several candidate genes have been found to be associated with stroke. However, association studies in population-based samples have failed to identify reliable disease markers. The publication of the "Human Genome Project" has indeed improved our knowledge about the potential role of genetics in complex disorders including stroke. Rapidly expanding field of genetics is in a state of transforming medicine into a new kind in future, the individualized medicine, using tailor made drugs according to the genetic makeup of the individuals. However, this involves integrating genome wide genetic information with medical information. The first genome wide association study on ischemic stroke has been published recently. Further studies will hopefully tell us how far the genetic information will assist us to tailor clinical and therapeutic decisions to an individual's genotype.
Keywords: Genetics, genome wide association, single gene disorders, stroke, multifactorial stroke
Stroke or brain attack is a complex disease, comprising of a mix of clinically different risk profiles, incidence rates, management and outcomes.  It is the third largest killer in the world after heart attack and cancer. , Genetic predisposition to stroke does occur and has been documented in both animal models and human beings. However, a precise definition of genetic factors responsible for stroke is still lacking because research into genetic basis of stroke presents some unique challenges. A small portion of stroke cases are caused by single gene disorders. More commonly, it seems to be a multifactorial polygenic disorder.  Mutations in some candidate genes are likely to predispose or give protection against stroke. Several mutations in various genes have found to be associated with stroke. However, we have a long way to go before we can accurately pinpoint the genes responsible for multifactorial stroke. It is hoped that identification of individuals with a high genetic risk may lead to the development of a strategy to effectively prevent the disease.
Studies involving twins, siblings and families have detected significant evidence for heritability, indicating that the genetic determinants are important but the extent of predisposition is not known.  In recent years there have been several parallels of research to establish the functional variants of some candidate genes and the risk of stroke.
A small number of twin studies have been carried out to assess the heritability of stroke. Classical twin studies look for differences among a cohort of monozygotic twins and dizygotic twins. The National Academy of Sciences- National Research Council (NAS-NRC) Twin Registry has been used to study the heritability of stroke. This registry is a cohort of 15,924 twin pairs of white men who served in the US Armed Services.  The study provided evidence that a genetic component of stroke risk does exist. Another analysis of the Danish Twin Registry further suggests that genetic factors increase the risk of stroke but the magnitude of the effect is moderate. 
The basic assumption of twin studies is that relatedness of monozygotic twins to each other is similar to that of dizygotic twins in all respects except genetic relatedness. However, this assumption may not be always valid. There may be differences in environmental sharing according to zygosity.  In case control studies, a family history of stroke has been shown to increase the risk of stroke by about 75%. 
If there are genetic variants that predispose to stroke, a positive family history of stroke should be a risk factor for stroke. Family history studies of stroke have been reported since at least the 1960s.  However, these studies vary considerably in design, methodological rigor and scope. Despite these methodological concerns, family history studies generally support a genetic component of stroke risk. A prospective cohort of 789 men living in Gothenburg, Sweden, was followed for up to 18.5 years.  Men whose mothers had died of stroke had a threefold increased incidence in the risk of stroke compared with men without a maternal history of stroke. Interestingly, the study did not find paternal history of fatal stroke to be a risk factor for stroke in cohort members. In a systematic review, Flossmann and colleagues have reported that first degree relatives of stroke patients are at an increased risk of developing the disease (OR=1.76). When the studies were adjusted for conventional risk factors for stroke such as smoking and hypertension, this risk was still evident. 
A large case-control family history study of 727 consecutive patients with ischemic stroke recruited from a single referral clinic in Southeast London, compared with 623 age, sex, and ethnicity matched controls, was performed by Hassan et al.  When age, sex, smoking, hypertension, diabetes mellitus, and number of siblings were controlled for, a family history of early-onset stroke remained a risk factor for both ischemic stroke at all ages, and young ischemic stroke.
Genetic predisposition differs depending on stroke subtype and age, both twin and family-history studies document a stronger genetic component in stroke patients aged younger than 70 years in comparison with those who are older. ,, Genetic factors seem to be more important in large- vessel stroke and small vessel stroke. There is no epidemiological evidence for a genetic component in cardioembolic stroke. ,, Genetic factors might affect stroke at various levels [Figure 1]. They could contribute to conventional risk factors such as hypertension, diabetes or homocysteine concentrations, which in turn have a known genetic component. , They can interact with environmental factors  or contribute directly to an established stroke mechanism such as atherosclerosis. , Further latency to stroke, infarct size after vessel occlusion or stroke outcome, might also be affected by genetic factors. 
Many studies have been carried out to investigate potential risk genes for multifactorial stroke. However, most robust findings have been on single gene disorders in stroke.
In young stroke patients without known risk factors Mendelian conditions are important causes of stroke. Identification of rare mendelian forms has been the most successful approach to map stroke related genes. When referring to stroke a distinction must be made between disorders where stroke is the prevailing manifestation and others where it is a part of wider spectrum. CADASIL or cerebral autosomal dominant arteriopathy with subcortical infarcts and leucoencephalopathy (a rare form of small vessel disease) is the only form of isolated stroke displaying heritable patterns of inheritance caused by mutations in Notch 3 gene (OMIM*600276).  Almost 100 mutations have been recorded in this gene. It is quite relevant in clinical practice. The clinical phenotype consists of recurrent strokes and transient ischemic attacks, progressive cognitive impairment and psychiatric disturbance with onset usually in the third to sixth decade. About a third of patients have been reported to develop migraine and aura. , Notch 3 gene is a large transmembrane receptor which has a role in arterial development and is expressed on vascular smooth muscle. The mutation spectrum is broad, 50 different mutations have been reported. However, no clear genotype-phenotype correlations have emerged. The phenotype is variable even within families.
There are many other single gene disorders like Fabry's disease, sickle cell disease, homocystinurea, the syndrome of mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS) and Moyamoya disease associated with stroke [Figure 2].
The genetic contribution to common multifactorial stroke is polygenic, and identification of individual causative mutations is problematic due to the complexity of such a condition. Probably there are many alleles with small effects.  Significant research is being conducted to establish the relationship between the functional variants of a variety of genes and the risk of stroke in different ethnic groups across the world.  These include angiotensin converting enzyme gene, , the endothelial nitric oxide synthase gene,  genes associated with inflammation [e.g. interleukin - (IL-6)], thrombosis and coagulation (e.g. factor V Leiden, fibrinogen, prothombin) lipid metabolism [e.g.paraoxonase-1 (PONI)], Apolipoprotein E(APOE), 5.10, Methylenetetrahydrofolate reductase (MTHFR) and Protein Z gene [Figure 3]. Very few attempts have been made to study the role of genetic variation in development of stroke in Indian population. However, we are conducting research to establish the relationship between functional variants of various candidate genes in the South Indian population from Andhra Pradesh. We have found that the angiotensin converting enzyme gene and PDE4D gene and eNOS variants are significant risk factors while variants in the genes encoding coagulation factors like prothrombin, Pro and anti-inflammatory genes (MMP3 and IL-10 genes) donot seem to be associated with stroke in this population. ,,, In addition to these we are studying many other candidate genes for a possible association with ischemic stroke in this population. Homocysteine has been recognized as an independent risk factor for cardiovascular and neurological diseases. Sharma et al, from Institute of Genomics and Integrative Biology, New Delhi, have identified 135 genes in 1137 abstracts that either modulate the levels of homocysteine or get modulated directly or indirectly by an elevated level of homocysteine. Mapping these genes to their respective pathways revealed that elevated levels of homocysteine lead to atherosclerosis. Elevated levels of homocysteine also decrease the bioavailability of nitric oxide and modulate the levels of other metabolites, which may result in cardiovascular or neurological disorders including stroke. 
A more recent and exciting discovery in the field is the identification of a gene that appears to contribute to the risk of stroke in carriers independent of other genetically based risk factors like hypertension or diabetes. The first locus, STRK1 was mapped to 5q12 using a genome wide search for susceptibility genes in the common forms of stroke.  This was done in Icelanders by decode group.  The Icelandic population is relatively small and genealogy can be traced over 10 centuries. The analysis was conducted among 468 patients from 179 families (2.6 individuals per family) by DECODE group (Reykjavik, Iceland). They used a broad but rigorous definition of the phenotype including patients with ischemic stroke, transient ischemic stroke, and hemorrhagic stroke. The logarithm of differences (LOD) score after adding a higher density of markers (one marker every 1CM) was 4.40 at the marker D5S2080. Once the hemorrhagic patients were removed, the LOD score increased to 4.9, suggesting that the gene at the locus is primarily important for ischemic stroke. Ultimately the most promising region harboring a stroke susceptibility gene was narrowed down to a segment <6 CM (approximately 3.8 Mb), from D5S1474 to D5S398, as defined by a decrease of one in LOD score (one-LOD interval). Subsequently the region was extensively fine mapped and tested for association to stroke. The group found the strongest association in the one-LOD interval, with in the phosphodiesterase 4D gene (PDE4D; OMIM 600129*). However, the strongest signal observed at PDE4D was the two major subtypes of stroke, carotid and cardiogenic stroke. Within this gene haplotypes or specific sets of genetic markers were identified, corresponding either to significantly increased or significantly decreased risk of stroke. First the association was identified using microsatellite markers and then the microsatellite data were supplemented with a denser set of single nucleotide polymorphism (SNPs). Two hundred and sixty PDE4D gene SNPs were examined. Of these, six were significantly associated with stroke after adjustment for multiple comparisons. 
Interestingly, the originally described haplotypes that have been linked to stroke do not affect the coding sequence; rather they are within the noncoding region of the protein. A protective haplotype which included it one microsatellite marker and an SNP was also described, that conferred a relative risk of 0.68.  It has been suggested that the haplotypes alter the transcription of PDE4D and may affect transcription levels or the ratio of alternatively spliced PDE4D isoforms. 
The gene encoding 5-Lipoxinase activating protein (ALOX5AP; OMIM 603700*) is another gene that has been discovered through genome-wide linkage analysis by decode in 2004. A four SNPs haplotype (Hap A) has been associated with 1.7 fold increased risk of stroke in the Icelandic population.  5 / lipoxygenase protein (FLAP) encoded by ALOX5AP, is an important component of the leukotreine pathway. In a Scottish study involving 450 ischemic stroke patients and 710 controls, Hap A imparted a stroke OR of 1.36 under a multiplicative model assumption. 
In addition to these there is a long list of candidate genes that have been studied for a possible association with ischemic stroke.
An extension of the classical patient-control studies, are genome wide association studies. These provide a non biased approach to examine the effect of a genetic variation in a specific trait. These studies are now quite common in the scientific literature. The remarkable advances in genotyping and DNA sequencing technologies give researchers the potential to pinpoint a disease trait.  These high throughput technologies provide geneticists with an opportunity to dissect multiple genetic loci influencing the phenotype of complex disorders such as stroke. Matarin et al, published the first genome wide association study in ischemic stroke;  250 ischemic stroke patients and equal number of controls from the US White population were included in the study. The authors identified positive mutations across 12 chromosomes and a number of genes were highlighted. The study has generated more than 200 million individual genotypes and a number of potentially interesting SNPs were nominated. Although a major locus has not been highlighted, it was pointed out that two of the parallel characterized genes that were nominated are involved in potassium transport (KCNIP4 on chr4p15 and KCNK17 on chr6p21) and two are located at chr9p21. , Two SNPs (rs 10757274 and rs 2383206) nominated a region adjacent to two candidate genes cDKN2A and CDKN2B. ,,, This seems to be the first genetic risk consistently associated with stroke and also highlights strong links between cardiovascular and cerebrovascular disorders.
Since the mode of inheritance of stroke seems to be quite complex, it causes several major difficulties for the identification of the stroke causing genes in humans. An important tool for investigating the genetic basis of stroke has been provided by available experimental animal models. Fully inbred animals avoid the problem of genetic heterogeneity and moreover, a much more precise control of environmental factors is possible in animal experimentation than in clinical studies.
The most widely used animal model is the stroke prone spontaneously hypertensive rat (SHR-sp) obtained by selective inbreeding from the spontaneously hypertensive rat (SHR). Experimental crossbreeding in the strain of stroke-prone rats has led to the identification of three major genetic loci that affect stroke risk in these animals. , Three loci are Str 1, Str2 and Str 3. However, the genetic variants underlying these loci remain to be identified.
Another genetic approach is the targeted disruption of genes in these animal models; for example in one study mice deficient in nitric oxide synthase develop smaller nitric oxide synthase infarcts and less severe functional deficits than control animals following occlusion of the middle cerebral artery.  Certainly these studies have helped in understanding the mechanism of ischemic stroke and also in defining candidate genes for genetic association studies in humans. However, it is often not an easy task to apply genetic experimental data to humans. Nevertheless, the use of animal models may still provide useful advances in understanding genetics of stroke in a number of ways.
The genetic variation contributing to the risk of stroke is now being studied extensively. However, much progress has been made in the identification of genes for Mendelian condition associated with stroke. Comparatively little is known about the genes involved in the multifactorial stroke. The identification of candidate genes like PDE4D, associated with stroke give some hope that genetic studies might have a direct impact on the treatment of the disease. The selective inhibition of candidate genes using drugs might ultimately lead to the prevention of a stroke. For example PDE4D and other members of PDE4D gene family are characterized by selective inhibition by Rolipram.  Once drugs are available to interfere with PDE4D pathways, the variants within this gene might be useful to decide which patients would be most likely to benefit from which drugs. Pharmacogenomics holds the promise that drugs might one day be tailor made for individuals affected with stroke. There have been a number of recent advances in understanding the implications of genetic polymorphisms in various cardiovascular diseases. 
At present, although a genetic predisposition may be a risk factor in individual patients, identifying the underlying genetic basis is usually impossible, and is clinically irrelevant. Currently, no gene test is mandated as part of the routine assessment of patients with ischemic stroke.  However, in a minority of stroke patients, abnormalities in a single gene are responsible for stroke. In this group, it is quite realistic to identify the underlying genetic abnormality. This may have important implications for both clinical management, and genetic testing of other family members.
There seems to be more than one way by which stroke genetics may alter our current management of stroke. Further studies will hopefully tell us how far the genetic information will assist to tailor clinical and therapeutic decisions to an individual genotype. This will also enable the identification of presymptomatic at risk individuals. Ultimately, the healthcare costs and the social burden associated with stroke might be lowered with this kind of information.
[Figure 1], [Figure 2], [Figure 3]