Genetic Basis of Stroke Occurrence, Prevention, and Outcome

Key Points

Fabry disease is an X-linked recessive disorder caused by reduced α-galactosidase activity for which enzyme replacement therapy is available.
Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy is a small-vessel stroke disorder caused by mutations in the NOTCH3 gene.
Cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy is a stroke disorder caused by mutations in the HTRA1 gene.
Homocysteinemia, commonly due to cystathionine-β-synthase deficiency, can be treated with restriction in dietary methionine and vitamin B ₆ supplementation.
Mitochondrial mutations can lead to a syndrome of encephalopathy, lactic acidosis, and stroke-like episodes known as mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes.
In children with sickle cell disease, transcranial Doppler ultrasound can define populations at high risk of stroke and transfusion therapy can dramatically reduce the risk of stroke.
Mutations in COL4A1 can cause small-vessel stroke, cerebral hemorrhage, and the hereditary angiopathy, nephropathy, aneurysms, and cramps syndrome.
Several single-gene disorders cause cerebral amyloid angiopathy.
Several single-gene disorders cause cerebral cavernous malformations.
Genome-wide association studies have identified risk loci for large-vessel and cardioembolic stroke.

Introduction

The basic terms and concepts of modern genetic epidemiology, reviewed in this introductory section, provide a framework for the chapter. One of the most important accomplishments of the late 20th century was completion of the Human Genome Project. With the identification of millions of genetic variations comes the possibility of identifying genetic risk associations for numerous complex traits.

A simple trait refers to a disease (or phenotype) that has a Mendelian (i.e., autosomal recessive, autosomal dominant, X-linked recessive, or X-linked dominant) or mitochondrial pattern of inheritance. Typically, a mutation in a single gene is both necessary and sufficient for a Mendelian disorder like Huntington disease to occur. A complex trait, however, refers to a disease (or phenotype) that does not follow a simple pattern of inheritance and is the result of multiple factors, including genes. Stroke, cancer, and many other common diseases are highly complex, with multiple interacting genetic and environmental risk factors. Other features of complex traits include genetic heterogeneity in which different genes lead to the same disease and phenotypic variation in which the same disease gene causes different phenotypes.

The central dogma of genetics is that DNA is transcribed into RNA, which is then translated into proteins. The genome refers to all the genetic information of an organism stored in DNA. Analogously, the transcriptome refers to all the messenger RNA transcribed from DNA; the proteome, to all the expressed proteins. Autosomes (non-sex chromosomes) are characterized by size, from largest (chromosome 1) to smallest (chromosome 22).

At any given location (locus) in the genome, a genotype may refer to the sequences on the paternal and maternal chromosomes. A polymorphism indicates a genetic variation, and the different variations are called alleles. Numerous types of polymorphisms occur, including single base pair changes called single nucleotide polymorphisms (SNPs) as well as insertions, deletions, or translocations of a single base pair or sections of DNA. Other types of polymorphisms include variable numbers of tandem repeats (VNTRs), which refer to consecutive repeats of the same nucleotide sequence. Examples of VNTR polymorphisms are trinucleotide repeats such as those seen in Huntington disease. These are also called microsatellites. Although most microsatellites are nonpathogenic, they occur frequently and have multiple alleles. Thus microsatellites are useful both for forensic identification of individuals and as markers for linkage analysis.

Copy number variations refer to sections of DNA that are copied onto the same section for variable numbers of times. A classic example in neurology occurs at the peripheral myelin protein 22 (PMP22) gene, in which the loss of one copy of the gene leads to hereditary neuropathy with liability to pressure palsies; the gain of one copy, Charcot-Marie-Tooth disease. Several hundred thousand copy number variations have been identified in the genome, although their use as markers for disease has not yet been established.

Two classic methods of identifying risk genes are association studies and linkage studies. Genetic association refers to the co-occurrence of one or more genotypes and a particular phenotype more often than expected by chance alone. Association is not synonymous with causation, however, and a genetic marker may be associated with a disease due to chance, confounding, or linkage with a causal variant. The major advantages of association studies are the ability to identify risks of small effect, and the ease of collecting unrelated cases and controls relative to collecting families.

The other major method for identifying risk genes is through linkage analysis ( Fig. 19.1 ). Mendel’s law of independent assortment states that the inheritance pattern of one trait is independent of the inheritance of another. This is true for traits that are on different chromosomes or appreciably separated in physical space on the same chromosome. However, loci that are physically close to one another on the same chromosome will not segregate independently and will instead be inherited together more often than expected by chance.

Fig. 19.1, Parental chromosomes (top) and chromosomes after recombination (bottom) are shown. The assumption in this figure is that a stroke mutation occurs on the purple band, but it has not yet been identified. The locations of markers A and B are known. Because B is so much closer to the disease-causing locus than A , it will be inherited along with the trait of stroke more often than A will. Only a recombination event between B and the stroke mutation will lead to the marker not being linked with stroke. A classic example of linkage of traits is hemophilia and color blindness. Neither trait causes the other to occur, but they occur in the same individual more often than expected by chance because the genes responsible for each trait are physically adjacent to one another on the same chromosome. Using this basic phenomenon, one can examine a series of markers across the genome. Even if the marker itself does not lead to disease, if it is close enough to the actual disease-causing gene, it will be inherited more often than expected by chance alone.

Evidence for Heritability of Stroke and its Subtypes

Many lines of evidence support a familial component to stroke risk. A meta-analysis of 53 independent studies found that monozygotic twins were 65% more likely to be concordant for stroke (i.e., both twins having stroke) than dizygotic twins. However, the confidence intervals were broad, and twin studies often failed to differentiate ischemic from hemorrhagic stroke. Case-control studies showed that a positive family history of stroke increased the risk of stroke by 76%. One limitation of case-control studies is that patients with severe strokes may not have survived long enough to be included in the study (survivor bias). Furthermore, patients might have had recall bias in which experiencing a stroke might trigger memories about family members who had a similar affliction (information bias). Finally, there may have been a tendency to mainly publish positive results (publication bias). Cohort studies have the advantage of not having the same risk of survivor or information biases as case-control studies. It is thus reassuring that cohort studies show that family history of stroke increases the risk of stroke by about 30%.

Probands have a nearly linear increased risk of having a sibling history of stroke with age of the proband at the time of stroke from 55 to 80 years. However, family history of stroke seems to be a greater risk factor for stroke in individuals younger than 70 years of age. A positive maternal history of stroke increases the risk in women by nearly 50% relative to a positive paternal history. Whether these observations are attributable to genetic, epigenetic, or nongenetic factors is uncertain.

Inherited risk may differ depending on the type of ischemic stroke. A review of two population-based studies from Oxfordshire and three hospital-based studies found that family history of stroke was least frequent in patients with cardioembolic stroke compared with patients with large- or small-vessel stroke or stroke of unknown etiology. Heritability seemed comparable among non-cardioembolic etiologies of ischemic stroke. Notably, however, comparable rates of family history of conditions like large-vessel stroke and small-vessel stroke give insight into the magnitude of the heritable component to stroke risk, but these rates do not address whether genetic factors for the various types of ischemic stroke might differ qualitatively; that is, different genetic variants might predispose to different types of ischemic stroke.

Heritability estimates using genome-wide genotype data for ischemic stroke is 42% for young stroke (age <55 years) and 34% for old stroke. Heritability estimates using genome-wide genotype data for intracerebral hemorrhage have been estimated to 15% for the apolipoprotein E (APOE) locus and 29% for non-APOE loci.

Disorders Associated with Ischemic or Hemorrhagic Stroke

Several disorders have cerebrovascular disease as a prominent feature. Stroke is common and typically not associated with an obvious genetic disorder. This, combined with the fact that genetic disorders are so diverse, makes the task of precise diagnosis seem daunting. Targeted gene testing should be done after considering the constellation of physical findings ( Table 19.1 ) and considering patterns of inheritance consistent with the findings of a detailed family history ( Table 19.2 ).

TABLE 19.1

Physical Examination Findings That Are Clues to Inherited Disorders Associated With Cerebrovascular Diseases.

System	Disease	Findings
Ophthalmologic findings	MELAS Fabry disease RVCL Homocystinuria Moyamoya disease	Bilateral cataracts Whorl-like corneal dystrophy Retinal vascular malformations Ectopia lentis, glaucoma Morning glory optic disc
Dermatologic	Fabry disease	Angiokeratoma (“swim trunk” distribution)
	Vascular Ehlers–Danlos	Thin translucent skin and excessive bruising
Otologic	MELAS	Progressive bilateral sensorineural hearing loss

MELAS, Mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes; RVCL, retinal vasculopathy with cerebral leukodystrophy.

TABLE 19.2

Patterns of Inheritance for Diseases or Conditions Associated With Cerebrovascular Diseases.

Patterns of Inheritance	Disease/Condition
Autosomal dominant	CADASIL
	CARASAL
	RVCL
	Vascular Ehlers–Danlos syndrome
	Cerebral cavernous malformations
	CCM1
	CCM2
	CCM3
	Cerebral amyloid angiopathies
	HCHWA-Dutch type
	HCHWA-Icelandic
	FAP
	Polycystic kidney disease
	ADPKD 1
	ADPKD 2
Autosomal recessive	Sickle cell disease
	CARASIL
	Homocystinuria
X-linked recessive	Fabry disease (can affect females)
Mitochondrial	MELAS

ADPKD, Autosomal dominant polycystic kidney disease; CADASAL, cathepsin A-related arteriopathy with subcortical infarcts and leukoencephalopathy; CADASIL, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy; CARASIL, cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy; CCM, cerebral cavernous malformation; FAP, familial amyloid polyneuropathy; HCHWA, hereditary cerebral hemorrhage with amyloidosis; MELAS, mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes; RVCL, retinal vasculopathy with cerebral leukodystrophy.

Fabry Disease

Fabry disease (Anderson-Fabry disease or angiokeratoma corporis diffusum) is an X-linked recessive disorder caused by reduced activity of the enzyme α-galactosidase. This enzyme is essential to the biodegradation of lipids, and its decreased activity leads to accumulation of lipids in lysosomes in endothelial and vascular smooth muscle cells, where cellular damage may lead to stroke.

Early in life, patients with Fabry disease may present with burning pain or acroparesthesia due to small-fiber sensory neuropathy, corneal clouding (cornea verticillata), or angiokeratomas. Because peripheral neuropathy typically involves only small fibers, it may be missed on nerve conduction studies and electromyography. Later in life, the major sequelae of Fabry disease are symptoms of stroke, heart disease, and kidney disease from blood vessel ectasia. Grewal reported a stroke prevalence of 24% in patients with cerebrovascular complications, and six of eight patients had strokes occur before 40 years of age.

Although Fabry disease has traditionally been classified as an X-linked recessive disorder in which males showed complete penetrance and women were carriers, the epidemiology of the disease may be more complex than originally thought. In particular, heterozygous females exhibit a wide range of clinical manifestations, from asymptomatic throughout a normal life span to as severe as many affected males. This variation has been attributed to X-chromosome inactivation. Beyond α-galactosidase A (GLA) gene mutations, it is likely that there are other genetic and nongenetic modifiers. Among 721 German adults aged 18–55 years with cryptogenic stroke, 4.9% of male patients and, surprisingly, 2.4% of female patients had a biologically significant mutation of the GLA gene. Among the male patients with stroke, 38.1% demonstrated the typical dolichoectatic vertebrobasilar vessels. Fewer than half the patients had angiokeratomas, acroparesthesia, or cornea verticillata, which suggests wide phenotypic variation. In a study of 103 young patients with cryptogenic stroke in a Belgian population, only three patients had low α-galactosidase activity, and none was found to have a GLA mutation.

Diagnosis of Fabry disease can be made by measurement of α-galactosidase activity less than 35% of the mean in male patients, but the gold standard for diagnosing the condition in female patients is through sequencing the α-galactosidase gene. Enzyme replacement therapy is commercially available, but efficacy may be mitigated by neutralizing antidrug antibodies, which are much more commonly seen in males likely due to complete absence of native enzyme.

CADASIL

Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is caused by mutations of the NOTCH3 receptor gene on chromosome 19. NOTCH3 expression occurs almost exclusively in smooth muscle cells, and mutations of the receptor lead to accumulation of the protein. Pathologically, granular osmiophilic material accumulates in the smooth muscle of vessels, which leads to smooth muscle degeneration and, ultimately, subcortical leukoencephalopathy. Magnetic resonance imaging (MRI), positron emission tomography, and transcranial Doppler (TCD) studies have demonstrated reduced cerebral blood flow in the white matter.

The mean age of onset of symptoms is in the later part of the 4th decade, although the condition may be detectable by MRI years earlier. More than 85% of patients have migraine headaches, and transient ischemic attacks (TIAs) and mood disorders are also common at presentation. Classic lacunar stroke syndromes may occur in two-thirds of patients. A patient with progressive leukoencephalopathy, particularly a young patient without hypertension, should prompt a search for CADASIL. In northeast England, CADASIL has a minimum prevalence of 1–25 in 50,000 in the general population. Brain MRI scans show T2-weighted hyperintensities occurring symmetrically in the white matter and deep gray nuclei. Small lacunar lesions immediately subcortical in the anterior temporal lobes (O’Sullivan sign) have been reported to be 100% specific and 59% sensitive for CADASIL.

Genetic testing is available but may require extensive sequencing of the large NOTCH3 gene, which contains 33 exons. Identification of granular osmiophilic material in vascular tissue from biopsy specimens from skin, muscle, or peripheral nerves in the appropriate clinical setting of premature white matter ischemic strokes, dementia, and migraine may be sufficient for diagnosis.

Donepezil at a daily dose of 10 mg causes significant improvements in Trail-Making Tests, parts A and B, and on the Executive Interview 25 (EXIT25), but the clinical significance of the modest improvements is limited. Anecdotally, acetazolamide at daily doses of 125–500 mg may help prevent migraines.

CARASIL

Cerebral autosomal recessive arteriopathy with subcortical infarcts and leukoencephalopathy (CARASIL), also called Maeda syndrome, is a distinct clinical entity that predominantly affects individuals of Japanese ancestry, although Chinese and Caucasian pedigrees have been reported. CARASIL is caused by mutations in the HTRA1 gene encoding HtrA serine peptidase/protease 1. Subcortical encephalopathy leads to psychomotor deterioration when patients are in their 20s and 30s. Migraine is not a feature. Distinctive extracerebral manifestations of the disease that often predate the neurologic presentation include premature alopecia and bouts of lumbago with herniation of vertebral disks and spondylosis deformans. T2-weighted MRI shows extensive areas of hyperintensity in the hemispheric white matter and less dramatic changes in the thalami and pons. The white matter changes occur without significant hypertension. Pathologically, there is severe widespread loss of arterial medial smooth muscle cells. Sclerotic changes are infrequent compared with CADASIL. Also, unlike CADASIL, abnormal vessels are not periodic acid-Schiff-stain positive.

Homocystinuria

The most common inherited form of elevated homocystine is attributable to cystathionine β-synthase deficiency and is referred to as homocystinuria. The trait is inherited in an autosomal recessive pattern; thus both parents are typically asymptomatic carriers. It was first described in 1962 when elevated homocystine levels were identified in the serum and urine of mentally retarded patients; the enzymatic defect was reported several years later.

Patients with homocystinuria are typically divided between those who respond to vitamin B ₆ (pyridoxine) supplementation and those who do not. In 1985, Mudd et al. reported on 629 patients with homocystinuria; by the age of 10 years, 55% of vitamin-B ₆ –responsive patients and 82% of vitamin-B ₆ –unresponsive patients had dislocated lenses. Along with Marfan syndrome, syphilis, and trauma, homocystinuria is one of the major diagnostic considerations in a patient with dislocation of the lens of the eye. Also similar to patients with Marfan syndrome, those with homocystinuria may develop skeletal abnormalities such as long limbs, tall height, pectus excavatum (caved chest) or pectus carinatum (pigeon chest), and arachnodactyly. Classically, homocystinuria patients have the dermatologic features of thin blonde hair, a malar flush, and livedo reticularis. Affected young adults with stroke may lack the classic phenotype. Without treatment, mental retardation may occur in two-thirds of patients. Elevated homocystine levels are associated with both thrombotic and embolic strokes, which occur by the age of 15 years in 12% of vitamin-B ₆ –responsive patients and in 27% of vitamin-B ₆ –unresponsive patients. Stroke occurs in more than 60% of patients with homocystinuria by the age of 40 years.

Homocystinuria is screened for by measuring plasma homocysteinemia and confirmed by gene testing. Routine testing at birth can identify patients at an early age. Restriction of dietary methionine and supplementation of vitamin B ₆ are recommended.

Mitochondrial Encephalopathy, Lactic Acidosis, and Stroke-Like Episodes

The syndrome of mitochondrial encephalopathy, lactic acidosis, and stroke-like episodes (MELAS) is caused by mutations in mitochondrial DNA. As with other mitochondrial disorders, it is inherited in a maternal pattern. Clinical criteria for making the diagnosis include stroke before the age of 40 years, encephalopathy characterized by seizures or dementia, and blood lactic acidosis or ragged red fibers on skeletal muscle pathologic examination. Brain MRI typically shows lesions involving the occipital lobes. The lesions often do not respect the boundaries of named vascular territories. MELAS mutations cause impairment in respiratory chain enzymes, particularly complex I. Eighty percent of cases are caused by a substitution mutation (A3243G) in the gene encoding for transfer RNA (tRNA) ^Leu(UUR) . Other substitution mutations and deletions have also been described. ^,

Favorable responses of stroke-like episodes to nitric oxide precursors (L-arginine and L-citrulline), antiepileptic drugs, antioxidants, the ketogenic diet, and steroids have been reported anecdotally, but controlled trials are lacking. Seizures should not be treated with valproic acid because patients can have a paradoxical reaction.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here