Relationship between MTHFR C677T, homocysteine, and ischemic stroke in a large sample of the Han Chinese population

Abstract

Ischemic stroke, one of the prevalent causes of death and disability worldwide, is linked to environmental and genetic factors, including polymorphisms in the methylenetetrahydrofolate reductase (MTHFR) gene involved in homocysteine metabolism. The present study aimed to explore the relationship between the MTHFR C677T variant, plasma homocysteine, and risk of developing large-artery atherosclerotic ischemic stroke (LAAIS) among Han Chinese.

A population-based case-control study, which included 1810 patients with LAAIS and 1765 unrelated control subjects, was conducted. Compared to the controls, LAAIS patients had a significantly higher prevalence of hypertension, diabetes mellitus, smoking, and alcohol consumption (P < .001), as well as significantly higher mean fasting blood glucose, triglyceride, total cholesterol, and plasma homocysteine levels (P < .001). The TT homozygous genotype correlated with increased risk of developing LAAIS, as indicated by a significantly higher odds ratio (OR) compared to the CT and CC genotypes, in both additive (OR = 3.215, P = .01) and recessive models (OR = 3.265, P = .01). The plasma homocysteine level was genotype-dependent according to the following trend: TT > CT > CC.

In conclusion, our data demonstrate that, in spite of its low prevalence in both patients and controls (1.5% vs 0.8%), the MTHFR C677T variant could, at least in part, affect homocysteine levels and this, either alone or in combination with other factors, increases the risk of LAAIS.

1. Introduction

Ischemic stroke is a relatively heterogeneous multifactorial disorder caused by interruption of blood supply to a part of the brain, which results in sudden loss of function. Ischemic stroke accounts for at least 80% of all stroke cases,^[1] though in China this value is around 70%,^[2] indicating that the proportion of acute stroke cases is population-dependent.

The risk of suffering an ischemic stroke depends on a number of environmental and genetic factors.^[3–6] The role of genetic determinants in ischemic stroke pathogenesis has been demonstrated by studies on twins,^[7] as well as case-control and cohort studies of familial aggregation of stroke risk.^[8,9] The heritable component is estimated at 37–40%^[5]; however, the magnitude of the genetic contribution differs for each subtype. Bevans et al reported that heritability accounted for 40% in large-vessel stroke, 33% in cardioembolic stroke, and 16% in small-vessel stroke.^[10] Holliday et al estimated a higher contribution of heritability in the Australian population: 66% (P < .001) for large-artery atherosclerotic stroke, 60% (P < .001) for cardioembolic stroke, and 10% (P = .26) for small-vessel stroke.^[11] Confirmed by other studies, these findings support the key role of genetic factors in the pathogenesis of ischemic stroke, particularly of the large-vessel type.

At present, it is unclear which genes are actually involved in ischemic stroke and how their function may contribute to it. Similarly, little is known about the interaction among candidate genes, and between them and environmental factors.

So far, most studies that tried to identify causative genes of stroke have focused on those involved in atherosclerosis, thrombosis,^[12–14] homocysteine metabolism,^[15–19] and hypertension.^[17,20–22] Large-scale studies have identified some genetic variants associated with ischemic stroke. The gene encoding for methylenetetrahydrofolate reductase, a key rate-limiting enzyme in the homocysteine metabolic pathway, is one of the most common susceptibility genes identified in recent years.^[23,24] Elevated plasma homocysteine levels represent an independent and potentially modifiable risk factor for ischemic stroke in different ethnic groups.^[25] Homocysteine is a reactive sulfur-containing amino acid formed in vivo as an intermediate in the metabolism of methionine and cysteine. Elevated homocysteine levels may result from both acquired and genetic determinants.

MTHFR catalyzes the conversion of 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate, which serves as a methyl donor in the remethylation of homocysteine to methionine in vivo. MTHFR is important for the cell’s folate metabolism, which is integral to Deoxyribonucleic acid (DNA) and Ribonucleic acid (RNA) synthesis, as well as protein methylation. A common missense mutation in the MTHFR gene, C677T, leads to the substitution of alanine for valine at position 222, creating a thermolabile enzyme with 50% lower activity at 37°C. Studies on different ethnic populations have failed to reveal a consistent role of C667T in ischemic stroke. This inconsistency can be likely attributed to small sample size, different types of stroke examined, and geographic disparities, highlighting the need for comparisons using larger populations.

In the present study, we aimed to explore the relationship between the MTHFR C677T variant, plasma homocysteine levels, and the risk for ischemic stroke in a large sample of the Han Chinese population. Moreover, the study focused on a geographic location with the highest incidence of stroke in China. We believe our study can provide valuable insights on stroke etiology at a population and regional level.

2. Materials and Methods

2.1. Study population

This project was reviewed and approved by the Medical Ethics Committee of Qiqihar Medical University ([2016]06) and informed consent was obtained from each participant. The study was conducted in accordance with the Declaration of Helsinki guidelines.

The design was that of a case-control study. The study population consisted of 3575 unrelated subjects, including 1810 patients with ischemic stroke (males/females = 965/845) enrolled through the Second and Third Affiliated Hospital of Qiqihar Medical University and Mohe City Hospital in Heilongjiang Province between July 2016 and December 2020. All participants resided within a geographic area spanning 47° to 53° N and 124° to 128° E. The subjects identified themselves as Han Chinese, with no history of intermarriage for at least 3 generations.

Patient diagnosis was based on the occurrence of a new and abrupt focal neurological deficit, with signs persisting for > 24 h, followed by confirmation via computed tomography and/or magnetic resonance imaging of the head. The specific genotypic subtype was determined according to the Trial of ORG1-72 in acute stroke treatment (TOAST) classification criteria. Patients with LAAIS were included in the study. Exclusion criteria for the ischemic stroke group included patients with severe cardiac disorders, renal or hepatic disease, and cancer; patients with cardioembolic stroke; and patients with small-artery occlusion or other undermined etiology.

Control individuals were recruited from the general population following a free physical examination. They included unrelated healthy volunteers with no clinical or radiological evidence of cardio-cerebrovascular diseases, and with matching living area and ethnic origin as ischemic stroke patients. Exclusion criteria for the control group included subjects aged < 40 years, as well as individuals with hypertension, diabetes mellitus, renal and liver insufficiency, and cancer.

Demographic and risk factor information was collected using a structured questionnaire. Anthropometric data and clinical parameters were obtained through a physical examination and clinical tests. The body mass index was calculated as the body weight (kg) divided by the squared weight (m²). Blood pressure was measured with a mercury sphygmomanometer on 2 to 3 occasions, with at least 5-min intervals in between. The percentage of oxygen saturation (SaO₂) of arterial hemoglobin was measured by index finger loop transcutaneous pulse oximetry (GE Healthcare, Shanghai, China). The reported SaO₂ was the average of 3 readings taken over 10 seconds. Plasma homocysteine levels were determined by a double-reagent enzymatic cycling method using an automatic biochemical analyzer (Au 54000; Olympus, Tokyo, Japan).

2.2. Genotyping

Genomic DNA was extracted from peripheral blood using a DNA isolation kit (Beijing Kangwei Century Biotechnology Co., Beijing, China) according to the manufacturer’s protocol. The DNA concentration was measured using a Nanodrop 2000 microspectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) and the purity was deemed satisfactory at an absorbance ratio A260/A280 ≥ 1.80. The MTHFR C677T polymorphism was analyzed using the Sequenom Massarray IPLEX platform (Sequenom, San Diego, CA, USA). Allele detection was completed using MALDI TOF mass spectrometry. All samples were assessed in duplicate and their genotyping displayed 100% concordance.

2.3. Statistical analysis

Statistical analysis was performed using SPSS 20.0 software (IBM Inc., Chicago, IL, USA). Continuous variables were presented as the mean ± standard deviation; whereas categorical variables were presented as percentages. Comparisons between patients and controls were performed using an unpaired Student t-test. Allele frequencies were estimated by the gene counting method. The χ2 test was used to assess the goodness-of-fit between observed allele frequencies and the expected frequencies based on Hardy-Weinberg equilibrium, as well as to evaluate differences across genotypes. Logistic regression was used to analyze the association of MTHFR C677T polymorphism with ischemic stroke risk. ANOVA was used to estimate the association of MTHFR C677T polymorphism with plasma homocysteine levels. Each genotype was assessed by logistic regression analysis, assuming additive, dominant, and recessive models of inheritance. The odds ratio (OR) and 95% confidence intervals (95%CI) were calculated for each polymorphism, with OR ≥ 1.50 considered to indicate ischemic stroke susceptibility. GraphPad Prism 8.0 (GraphPad Software, La Jolla, CA, USA) was used to draw the histogram representing homocysteine levels of different genotypes. P < .05 was considered to indicate statistically significant differences.

3. Results

3.1. Baseline characteristics of the study population

The overall demographic, biophysical, and biochemical parameters of patients and controls are summarized in Table 1a. Patients exhibited significantly higher mean values for body mass index, systolic blood pressure, diastolic blood pressure, total cholesterol, triglycerides, uric acid, and plasma homocysteine compared to control subjects (P < .001) (Table 1b). Among patients, SaO₂ was lower in women than in men (93.13 ± 20.57 vs 94.90 ± 16.15, P = .04) (Table 1c). Considering traditional risk factors, patients had a significantly higher incidence of hypertension (71.7% vs 22.84%), diabetes mellitus (29.2% vs 10.03%), familial history of coronary atherosclerotic heart disease (26.9% vs 18.65%), as well as smoking and alcohol consumption (37.7% vs 29.23%, and 33.5% vs 18.41%, respectively, all P < .001) (Table 1a). These results point to more suspected risk factors in the ischemic stroke group.

Table 1A.

Baseline characteristics of the study population.

	Controls (n,%)	Patients (n,%)	t/χ2	P
N	1765	1810
Age (yr)	50.82 ± 8.87	62.71 ± 11.86	27.521	<0.001
Sex (male/female,%)	62.27%/37.73%	53.31%/46.69%	29.343	<0.001
Smoking (n,%)	516 (29.24%)	682 (37.68%)	28.599	<0.001
Driking (n,%)	325 (18.41%)	606 (33.48%)	105.328	<0.001
Hyperteinsion (n,%)	403 (22.83%)	1298 (71.71%)	856.022	<0.001
Diabetes mellitus (n,%)	177 (10.03%)	528 (29.17%)	206.845	<0.001
History of CAD (n,%)	329 (18.64%)	487 (26.91%)	34.660	<0.001
Hyperhomocysteine-mia (n,%)	–	179 (9.89%)

Abbreviations: CAD = coronary heart disease.

Table 1B.

Baseline characteristics of the study population($\overline{x}\pm s$).

	Controls (1765)	Patients (1810)	t	P
BMI (kg/m2)	23.57 ± 3.11	25.15 ± 4.31	9.903	<0.001
SBP (mm Hg)	118.32 ± 13.16	139.10 ± 16.49	33.582	<0.001
DBP (mm Hg)	77.46 ± 8.95	86.85 ± 10.81	22.880	<0.001
BP (mm Hg)	40.86 ± 15.13	52.25 ± 12.14	20.002	<0.001
SaO2 (%)	94.86 ± 16.69	94.15 ± 18.16	1.216	0.22
TC (mmol/L)	5.04 ± 0.97	5.42 ± 1.14	10.743	<0.001
TG (mmol/L)	1.52 ± 2.16	2.05 ± 4.23	4.735	<0.001
FPG (mmol/L)	4.95 ± 1.95	5.32 ± 2.97	4.414	<0.001
BUN (mmol/L)	6.56 ± 24.51	4.77 ± 1.71	3.099	<0.01
UA (µmol/L)	322.9 ± 109.25	374.95 ± 128.98	13.031	<0.001
CR (µmol/L)	64.57 ± 9.41	64.13 ± 8.58	1.460	0.14
Hcy (µmol/L)	12.49 ± 4.36	13.67 ± 6.62	6.309	<0.001

Abbreviations: BMI = body mass index, BP = SBP-DBP, BP = Pulse pressure difference, BUN = blood urea nitrogen, CR =creatinine, DBP = diastolic blood pressure, FPG = fasting plasma glucose, Hcy = homocysteine, SaO₂ = percentage of oxygen saturation, SBP = systolic blood pressure, TC = total cholesterol, TG = triglyceride, UA = uric acid.

Table 1C.

Baseline characteristics of the study population ($\overline{x}\pm s$).

Characteristic	Male			Female			P control	P case
	Controls	Patients	P	Controls	Patients	P
No of subjects	570	965		1195	845
Age (yr)	51.75 ± 9.23	61.56 ± 11.97	<0.001	50.38 ± 8.65	64.04 ± 11.6	<0.001	<0.01	<0.001
BMI (kg/m2)	23.74 ± 3.06	25.44 ± 4.66	<0.001	23.49 ± 3.13	24.82 ± 3.82	<0.001	0.11	<0.01
SBP (mm Hg)	121.7 ± 10.95	141.29 ± 17.33	<0.001	116.71 ± 13.79	136.41 ± 15	<0.001	<0.001	<0.001
DBP (mm Hg)	78.61 ± 7.15	88.77 ± 11.47	<0.001	76.91 ± 9.64	84.72 ± 9.61	<0.001	<0.001	<0.001
SaO2 (%)	95.54 ± 15.09	93.13 ± 20.57	0.009	94.52 ± 17.49	94.90 ± 16.15	0.61	0.21	0.04
TC (mmol/L)	1.51 ± 1.33	1.93 ± 1.58	<0.001	1.52 ± 2.47	2.20 ± 0.23	<0.001	0.91	<0.001
TG (mmol/L)
FPG (mmol/L)	4.89 ± 0.81	5.10 ± 1.30	<0.001	4.98 ± 2.32	5.50 ± 4.27	0.001	0.23	0.01
BUN (mmol/L)	63.72 ± 8.95	63.0 ± 8.56	0.118	64.99 ± 9.62	64.41 ± 8.63	0.15	0.01	0.001
UA (µmol/L)	326.98 ± 104.43	381.01 ± 120.76	<0.001	320.88 ± 111.66	366.74 ± 139.17	<0.001	0.26	0.02
CR (µmol/L)	64.38 ± 8.64	64.29 ± 8.78	0.845	64.83 ± 9.40	64.56 ± 8.87	0.51	0.32	0.52
Hcy (µmol/L)	11.93 ± 5.46	13.86 ± 6.74	<0.001	12.86 ± 5.74	14.45 ± 6.26	<0.001	0.001	0.05

Abbreviations: BMI = body mass index, BUN = blood urea nitrogen, CR = creatinine, DBP = diastolic blood pressure, FPG = fasting plasma glucose, Hcy = homocysteine, SBP = systolic blood pressure, SaO2 = percentage of oxygen saturation, TC = total cholesterol, TG = triglyceride, UA = uric acid.

3.2. Distribution of the MTHFR c677t genotype and allele frequency

As shown in Table 2, the distribution of MTHFR C677T genotype frequencies did not deviate significantly from the Hardy-Weinberg equilibrium. Univariate analysis demonstrated that the frequency of the MTHFR C677T polymorphism differed between patients and controls. The prevalence of ischemic stroke was higher in subjects with the T allele, either in heterozygous (CT) or homozygous (TT) form (OR = 1.104 vs 2.015, unadjusted), compared to the CC homozygous genotype. In particular, the TT genotype carried a more than 2-fold higher relative risk than the CC genotype (OR = 2.015, 95%CI = 1.099–3.763). To assess whether the C677T single nucleotide polymorphism was independently associated with the occurrence of ischemic stroke after adjustment for age, sex, and age + sex, we performed logistic regression analysis. The results demonstrated that MTHFR C677T correlated with susceptibility to ischemic stroke, with the TT homozygous genotype showing significant association with increased risk for ischemic stroke (OR₂ = 3.036, 95%CI = 1.364–6.757; OR₃ = 2.768, 95%CI = 1.390–5.512; and OR₄ = 3.215, 95%CI = 1.385–7.463;). The existence of a statistically significant association was confirmed in 2 models out of 3: additive (OR = 3.215, 95%CI = 1.385–7.463, P = .01) and recessive (OR = 3.265, 95%CI = 1.409–7.569, P = .01), after adjustment for age and sex, respectively (Table 3). These findings indicate that MTHFR C677T might be related to the occurrence of cerebral infarction.

Table 2.

Distribution of allele and genotype frequencies in patients and controls.

		Controls	Patients	OR1 (95%CI)	OR2 (95%CI)	OR3 (95%CI)	OR4 (95%CI)
rs1801133	CC	1330 (75.4)	1320 (72.9)	1.0	1.0	1.0	1.0
	CT	420 (23.8)	460 (25.4)	1.104 (0.947,1.286)	0.925 (0.747,1.144)	0.984 (0.815,1.188)	0.934 (0.752,1.161)
	TT	15 (0.8)	30 (1.7)	2.015 (1.079,3.763)*	3.036 (1.364,6.757)*	2.768 (1.39,5.512)*	3.215 (1.385,7.463)*
	C	3080 (87.3)	3100 (85.6)	1.0
	T	450 (12.7)	520 (14.4)	1.148 (1.002,1.315)*

Abbreviations: OR = odds ratio, OR₁ = not adjusted any factor, OR₂ = adjusted age, OR₃ = adjusted sex, OR₄ = adjusted age and sex.

*P < 0.05.

Table 3.

Logistic regression analysis adjusted for age, sex and (age + sex) according to the additive (co-dominent), dominant and recessive genetic modes.

	OR (95%CI)	P+
Additive
TT VS CC(ref)
Crude	2.015 (1.079, 3.763)	0.028
Adjusted	3.215 (1.385, 7.463)	0.007
Dominant
CT + TT VS CC(ref)
Crude	1.135 (0.977, 1.319)	0.098
Adjusted	0.993 (0.804, 1.228)	0.950
Recessive
TT VS CC + CT(ref)
Crude	1.966 (1.054, 3.667)	0.033
Adjusted	3.265 (1.409, 7.569)	0.006

Abbreviations: OR = odds ratio, CI = confidence interval.

P⁺values, adjusted for age and sex.

3.3. Plasma homocysteine levels and the MTHFR c677t genotype

The relationship between plasma homocysteine levels and the MTHFR C677T polymorphism was compared using ANOVA. A significant association was observed between elevated homocysteine and genotype. Specifically, plasma homocysteine levels were highest in the TT genotype (22.13 ± 7.76 μmol/L), intermediate in the TC genotype (19.26 ± 9.65 μmol/L), and lowest in the CC genotype (11.96 ± 4.13 μmol/L) (P < .001; Table 4, Figure 1, the detailed data are shown in appendix 1, Supplemental Digital Content 1, http://links.lww.com/MD/H318). This result suggests that the MTHFR C677T variant correlates with LAAIS, but only the TT homozygous genotype presents a significantly increased risk of developing LAAIS. The underlying mechanism may be associated, at least in part, with elevated homocysteine levels.

Table 4.

Relationship between the homocysteine levels and genotype for MTHFR C677T in the patients with LAAIS.

genotype	Plasma Hcy levels(umol/L)	P 1	P 2
CC	11.96 ± 4.13
CT	19.26 ± 9.65	<0.001
TT	22.13 ± 7.76	<0.001	0.15

Abbreviations: MTHFR = methylenetetra hydrofolate reductase.

P₁: CC genetype as a reference.

P₂: CT genetype as a reference.

Figure 1.

Graphic display regarding the association of plasma tHcy levels with MTHFR C677T polymorphism according to each genotype.

4. Discussion and conclusions

To the best of our knowledge, this is the first study to clearly demonstrate a significant association of the MTHFR C677T polymorphism with the risk of ischemic stroke in a large sample of the Han Chinese population with the highest incidence of stroke in China. These findings have a 2-fold importance. First, they further corroborate that the MTHFR C677T functional variant is a reliable genetic biomarker for ischemic stroke, regardless of ethnicity, gender or ischemic stroke subtype. Second, they confirm how the prevalence of the MTHFR C677T allele and its genotypes varies depending on ethnicity and geographic area.

In the past 40 years, China has experienced rapid economic development. This has brought a change of lifestyle, increasing the proportion of elderly people in the population and with it a shift in disease patterns such as an increase in the number of stroke patients. Previous epidemiological studies have suggested a marked geographic variation in stroke incidence, mortality, and prevalence among the Chinese population. The highest incidence of stroke has been recorded in Heilongjiang Province, where it is 6 times higher than in Guangxi Province, while mortality is 9 times higher. Acute ischemic stroke is significantly more prevalent than hemorrhagic stroke (P < .01), and both are more prevalent in males than in females (P < .01). Interestingly, ischemic stroke cases increase during the winter and spring seasons in Heilongjiang Province. Previous reports have shown that hypertension is the most important risk factor for stroke, giving extra independent significance to both systolic and diastolic blood pressure readings.

The diastolic blood pressure reading was a significant risk factor for stroke in the current study. Indeed, existing epidemiological evidence suggests that the 40% of geographic variation in stroke incidence and mortality in China could be attributed mainly to differences in the prevalence of hypertension.^[26]

In our study population, the prevalence of hypertension reached 71.71%. To determine why the incidence of stroke was so high in Heilongjiang Province, we took into account traditional cerebrovascular disease risk factors and genetic factors. By focusing on the modifiable cerebrovascular disease risk factors (Table 1a, b), we found that particularly hypertension and alcohol consumption could explain such a high incidence of stroke. Patients were characterized by a significantly higher alcohol consumption compared to controls (OR = 2.230, 95%CI = 1.910–2.604, P < .001).

We found that the MTHFR C677T variant clearly correlated with LAAIS, with the TT genotype showing significantly increased risk for LAAIS. This finding implies that, even at low frequency, the MTHFR C677T variant exerts a large influence on the occurrence of stroke in this population. At the same time, we detected a gradual increase in plasma homocysteine levels with increasing T allele dosage. This result is consistent with previous reports.^[27–29] One possible explanation is that the MTHFR C677T variant increases the risk of developing ischemic stroke, at least in part, by elevating homocysteine levels.

There are also some limitations to our study. First, we did not determine folate and vitamin B levels, neither in patients nor in controls, which prevented us from evaluating the role of nutritional status in our case population. Second, we only assessed MTHFR, whose most common mutation is a C > T substitution at position 677 (rs1801133),^[30,31] which leads to the replacement of alanine with valine at position 222,^[32] and correlates with decreased enzyme activity.^[33] The enzyme activity of CT and TT genotypes is <35% and 70% of that of the CC genotype, respectively, eventually leading to elevated plasma homocysteine levels.^[34] Therefore, this variant has been considered an ideal candidate for identifying any predisposition to ischemic stroke. The prevalence of MTHFR C677T varies depending on ethnicity and location. The global prevalence of the TT homozygous genotype varies between 8% and 18% in European and Northern American populations,^[35,36] but is much lower among Blacks (1.45%).^[37] In Asian populations, the frequency of the TT genotype is 15% among the Japanese,^[38] 14% among Koreans, and 0–3.7% among Indians.^{[6,7,19–21,37,38]}

In the Chinese population, the prevalence of the MTHFR C677T variant changes widely across case-control studies. Yuan et al reported a prevalence of up to 32.8% in the control population^[39]; whereas Mao et al reported a value of only 8.08% in controls (n = 200).^[31] These inconsistent results may be complicated by the susceptibility of the functional variant to common diseases.

Ischemic stroke is a late-onset, complex, polygenic disease, and plasma homocysteine levels are influenced by both genetic and environmental factors (particularly, dietary intake of folic acid and vitamin B). Accordingly, a specific genetic predisposition and traditional risk factors interact with each other, conferring a combined significant effect to the susceptibility of ischemic stroke. Generally, genetic risk factors are thought immutable; whereas many traditional risk factors are modifiable or avoidable (e.g., smoking and drinking). Therefore, current and future strategies should focus on identifying the genes associated with stroke and, at the same time, investigating the effect of environmental risk factors. A comparison with similar diseases such as acute myocardial infarction should be also carried out. Such data are important mostly to expound the underlying pathogenesis of stroke and to plan a possible preventive intervention.

Author contributions

KZ, JL, and CQ conceived and designed the experiments; MJ, HZ, and YH performed the experiments; JZ, XL, JN, CL, and NW were responsible for diagnosis and sample collection; MC and NW analyzed and interpreted the laboratory data; JL and CQ wrote the manuscript. All authors approved the final version of the manuscript.

Acknowledgements

We are very grateful to the doctors and nurses in the participating hospitals for their help with collecting the data and patients’ samples. We would also like to thank all the study staff and participants for their contribution. This work was supported by a grant from the National Nature Science Foundation of China (NO. 31171146, 31371208) and The Specific Research Foundation of Qiqihar Medical University (NOs. QY2016GJ-02 and QY2017KZDZ-01).