Introduction

   A stroke is a permanent neurological deficit caused by inadequate perfusion of a region of the brain or brain stem. Ischemic strokes comprise the majority of strokes, between 70-80% of all strokes. Stroke risk factors include both genetic and environmental factors. Stroke risk factors that can be treated include hypertension, heart disease, smoking, transient ischemic attacks, and high red cell blood cell count. Risk factors for stroke that cannot be changed include age, gender, race, diabetes mellitus, prior stroke, and family history of strokes. Other controllable risk factors, secondary risk factors, for stroke that also contribute to heart disease, include high blood LDL-cholesterol and lipids, physical inactivity, and obesity.⁶²⁾ 
   Ischemic stroke is a heterogeneous disease caused by different pathogenic mechanisms, and it is important that these are understood in interpreting genetic studies. There is growing evidence to suggest important interplays between levels of modifiable risk factors and the genetic susceptibility to stroke.¹⁷⁾

1. Subtyping ischemic stroke and genetic research
   Various mechanistic and syndromic systems have been developed for subtyping ischemic stroke. Many ischemic stroke subtyping systems have been developed for use in randomized clinical trials or epidemiologic studies. Some of these systems, or modified versions, are being used in stroke genetics research to address heterogeneity of the ischemic stroke phenotype (Table 1). The mechanistic approach to ischemic stroke, as advocated in the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) study,¹⁾ is the best suited to genetic research. This classifies ischemic stroke into five mutually exclusive categories：1) large-artery atherosclerosis；2) cardioembolism；3) small-vessel occlusion；4) stroke of other determined etiology；and 5) stroke of undetermined etiology. Jerrard-Dunne and colleagues³⁶⁾ used this approach in a family history study of 1000 individuals with ischemic stroke and 800 controls. The authors found that a family history of vascular disease was a risk factor in both small-vessel disease and large-vessel atherosclerosis, but not in cardioembolic stroke or stroke of undetermined etiology. This suggests that genetic research may be most fruitful when concentrating on these two particular subtypes.

2. Small-vessel occlusion stroke
   Small-vessel disease was defined as a clinical lacunar syndrome with a relevant infarct of <1.5 cm in the absence of cardioembolic source or carotid stenosis >50%. Ischemic change in the subcortical white matter, manifest as leukoaraiosis on CT or MR imaging, results from impaired blood flow in the distribution of the long penetrating arterioles of the brain. Occlusion of the short penetrating arterioles produces lacunar infarction in the deep gray nuclei. Both regions are supplied by end arterioles with little collateral circulation. Ischemic change in the subcortical white matter and lacunar infarctions tend to co-occur in the same patient and probably share etiologic and pathogenic mechanisms. The most consistently identified risk factors for small-vessel disease are age and hypertension.⁶⁾ However, relatively few studies have attempted to identify genes contributing to manifestations of arteriolosclerotic cerebrovascular disease. To date, efforts to identify genes contributing to leukoaraiosis have relied exclusively on candidate gene association approaches. In particular, polymorphisms in genes influencing interindividual variation in blood pressure level and diagnostic category (hypertension versus normotension) have been investigated for their possible relationship with the presence and severity of leukoaraiosis.

3. Large artery stroke
   Large-vessel disease was defined as >50% stenosis or occlusion of an appropriate major brain artery or branch cortical artery in the absence of sources of cardiac embolism. Large artery stroke is caused by carotid or vertebral artery stenosis, which in most cases is secondary to atherosclerosis. The exposure of the endothelium to single or multiple injurious factors causes the initiation of an inflammatory state leading to atherosclerosis.⁶⁸⁾ This inflammatory process includes the increased expression of adhesion molecules, cytokines, chemokines, metalloproteinases, and antigen mediated activation of macrophages and T-lymphocytes. Platelet aggregation and clot formation with or without plaque rupture ensue.
   Atherosclerosis of the carotid arteries is measured by carotid intimal-medial wall thickness(IMT) and is a strong predictor of future stroke.⁶⁰⁾ IMT differs by sex and is predicted by both family history of vascular disease and IMT of first-degree relatives.⁷⁹⁾⁹⁴⁾ Thirty percent of IMT variability is attributable to genetic factors.⁸⁶⁾
   It is of interest that although IMT is widely used as a marker for atherosclerosis, different heritable factors may be involved in plaque formation compared with IMT.⁹⁶⁾ Furthermore, some investigators have concluded that although IMT and plaque formation are age-dependent phenomena, they have distinct pathologic characteristics.²⁸⁾
   Although the current review of the literature supports an association between the presence of IMT and the development of atherosclerotic plaque, preliminary studies that examine genetic profiling related to the pathophysiology of atherosclerosis based solely on IMT may need to be confirmed in populations with frank carotid atherosclerosis. 

4. Candidate genes
   Apart from the novel loci, the likely candidate genes associated with stroke are those that are associated with matrix deposition(stromelysin-1, MMP3), inflammation(IL-6), and lipid metabolism(hepatic lipase, APOE, CETP, PON1) and clotting(factor V Leiden, fibrinogen). 

1) Stromelysin-1(matrix metalloproteinase 3, MMP3)
   Vascular remodeling is an essential phenomenon in the development of atherosclerotic changes in the arterial wall. Matrix metalloproteinases(MMP) and their endogenous tissue inhibitors(eg, TIMP-1) regulate the accumulation of extracellular matrix(ECM) during tissue injury, and thus the growth of the atherosclerotic plaque.⁹¹⁾ Stromelysin 1 is a key member of the MMP family and has wide substrate specificity. There is a common variant in the promoter sequence of this gene,⁹³⁾ in which one allele has a run of six adenosines(6A) whereas the other has only five(5A). In vitro studies of promoter strength showed that the 5A allele expressed higher activity than the 6A allele.⁹³⁾ The 6A allele has been associated with higher IMT,⁶⁶⁾⁶⁹⁾ presumably as these individuals are genetically predisposed to produce less stromelysin-1 mRNA. However, in a recent study looking at 237 Italians with ischemic stroke²¹⁾ an association was found with the 5A/5A genotype. This inconsistency in the relationship between the 5A/6A genotype and stroke requires more studies to resolve.

2) Interleukin-6(IL-6)
   IL-6 is an inflammatory cytokine, and three common variants have been identified in the IL-6 gene promoter, of which -174G>C has been most widely studied.²⁰⁾⁸³⁾ Recent in-vivo studies have reported that in healthy individuals there is little or no difference in plasma IL-6 levels by genotype, but after coronary artery bypass surgery⁷⁾ and in abdominal aortic aneurysm patients,⁴⁰⁾ individuals with the -174CC genotype consistently have higher IL-6 protein levels. There is discrepancy in studies examining IMT and IL-6 genotype, with some showing the -174GG genotype to be significantly associated with higher carotid IMT⁶⁶⁾⁶⁹⁾ and in others the -174C allele.¹¹⁾ The CC genotype was found to modulate the effects of alcohol on carotid atherosclerosis in a large community population(Carotid Atherosclerosis Progression Study)³⁸⁾ and the C allele was associated with increased IMT in smokers.³⁷⁾ Two recent studies have examined the risk of ischemic stroke with IL-6 genotype,²¹⁾²⁶⁾ and both have shown that the -174 GG genotype is associated with ischemic stroke in patients. Clearly, until further studies have resolved this discrepancy, the association between IL-6 genotype and stroke must be taken as unconfirmed.

3) Hepatic lipase
   The hepatic lipase -480C>T polymorphism results in low hepatic lipase activity and higher HDL levels.³⁴⁾ High hepatic lipase activity is associated with an increase in small dense LDL particles, and as such, an increased risk of the development of atherosclerosis and vascular events. The hepatic lipase genotype CC has been shown to increase IMT in one study,⁶⁹⁾ but a recent study has shown the opposite effect, with the TT genotype having a significantly higher carotid IMT in male patients with familial combined hyperlipidemia.⁹²⁾ Until further studies have resolved this discrepancy, the association between hepatic lipase genotype and IMT must be taken as unconfirmed.

4) Apolipoprotein E(APOE)
   The gene for apolipoprotein E(APOE) has a common variation that creates the E2, E3, and E4 isoforms, which have been demonstrated to be an important predictor of plasma lipid levels as well as having a small but consistent effect on the risk of coronary artery disease(CAD).⁷⁸⁾ A number of studies have examined this gene with respect to IMT, with most showing an association between the E4 allele and higher IMT,¹⁰⁾⁸⁴⁾ although some have failed to find a significant effect.³⁾¹⁹⁾ Framingham Offspring Study were recently studied and it was found that the E2 allele was associated with lower carotid atherosclerosis in women, and the E4 allele was associated with higher carotid IMT in diabetic men.¹⁸⁾ Several studies have also suggested an association between the E4 allele and the risk of stroke, and this is supported by a meta-analysis.⁵⁷⁾ A recent study in young Italian adults showed that the APOE epsilon4 allele and cigarette smoking act synergistically, increasing an individual's propensity to have a cerebral ischemic event.⁶⁴⁾ Szolnoki and colleagues⁸⁰⁾ also found the presence of E4 acted synergistically with hypertension, smoking, diabetes and drinking in both large and small-vessel ischemic stroke. However, recent studies have also showed no association.^16)54)76)85)

5) Angiotensin converting enzyme(ACE)
   The ACE gene has an insertion(I)/deletion(D) polymorphism that has been associated with risk of MI.⁷⁰⁾ Several studies have examined this gene with respect to IMT, with some studies showing an association²⁹⁾⁴³⁾ and others failing to find an effect.³⁰⁾³²⁾ In the recent Rotterdam study,⁷³⁾ the D allele was significantly associated with IMT in smokers. However, as with APOE, the association of higher IMT with the ACE D allele has not been contradicted, and the failure to find an association may be because of small sample sizes or confounding. A meta-analysis of published data also supported an association between the D allele and the risk of stroke.⁷⁵⁾

6) Angiotensinogen
   Angiotensinogen(the precursor protein in the renin-angiotensin system) has a methionine to threonine substitution at amino acid 235(M235T), in which the T allele has been associated with an increased risk of CAD.⁹⁰⁾ This is an complete linkage disequilibrium with the promoter polymorphism AGT G-6A, and therefore previous analysis referring to the AGT 235T allele can be extrapolated to the AGT-6A allele. The AGT-6A allele has been significantly associated with increased mean carotid IMT in women,¹²⁾ and more recently the AGT TT genotype,⁵⁾ but not in two other studies,³⁵⁾⁷¹⁾ and thus currently the relationship between the AGT genotype and IMT is not clear. In a recent study of 365 Korean patients with cerebral infarction,⁸⁵⁾ the authors found that the TT genotype was higher in patients with cerebral infarction, and furthermore that the AGT/TT genotype increased the relative risk of stroke in individuals with the ACE/DD genotype. Before that study there were no compelling data supporting an association of the AGT genotype with the risk of stroke.⁴⁾⁷⁴⁾

7) Paraoxonase
   Paraononase(PON1) is a glycoprotein that prevents the oxidation of LDL and may play an important role in the development of atherosclerosis.⁵⁵⁾ Two common variants within the coding region of the PON1 gene have been reported, L55M and Q192R and more recently a -107T>C variant in the PON1 promoter has been identified.⁵⁰⁾ The ability of paraoxonase to prevent lipid oxidation and the fact that smoking reduces paraoxonase activity directly suggests that the PON1 gene may play an important role in smoking-associated CAD risk and in the interaction with those proteins(eg, APOE) in which oxidation may modulate their function. Several studies have examined these variants with respect to IMT, with some showing an association,⁴⁹⁾⁵⁶⁾ and others failing to find an effect or finding the opposite effect of the allele with IMT.¹⁴⁾⁴²⁾ The 192R allele has been associated with the risk of stroke,⁸⁷⁾ as has the 107T allele,⁸⁸⁾ but because of these conflicting results there are currently no enough data to support the role of the paraoxonase gene in stroke.

8) Cholesteryl ester transfer protein(CETP)
   CETP facilitates the exchange of neutral lipids among plasma lipoproteins and induces a transfer of cholesteryl esters from HDL to triglyceride-rich lipoproteins in exchange for triglyceride. The protective role of HDL cholesterol against atherosclerosis is well established. In human beings, the association of CETP polymorphism with risk for atherosclerosis depends on an interaction between environmental factors, such as alcohol intake, and lipoprotein profile.²⁵⁾⁹⁵⁾ Several polymorphisms in the human CETP gene have been studied：in particular, there is one common variable TaqI site(TaqIB), which is a silent base change affecting nucleotide 277 in the first intron of the CETP gene.²⁾ The TaqI rare allele is associated with low CETP activity in the plasma, and high concentrations of HDL, as is a second gene variant that causes an isoleucine to valine substitution at residue 405(I405V). CETP genotype also affects plasmal concentrations of small dense LDL,⁸²⁾ which itself is known to be an important determinant of IMT.³¹⁾ Two studies have examined the I405V CETP polymorphism with respect to IMT, and the reported association is inconsistent.⁴¹⁾⁶¹⁾ No published studies indicate a role for these genotypes in stroke risk.

9) 5,10-methylenetetrahydrofolate reductase(MTHFR)
   MTHFR is the enzyme required for the conversion of dietary folate to 5-methyltetrahydrofolate, the methyl group donor required for the remethylation of homocysteine to methionine. Elevated homocysteine levels have been associated with atherothrombotic complications. Elevated homocysteine levels have been associated with atherothrombotic complications.⁵⁸⁾ A single base pair(677