Detection of mitochondrial DNA with 4977 bp deletion in leukocytes of patients with ischemic stroke

Background Coronary artery disease is associated with a common mitochondrial DNA alteration, a 4977 bp deletion (mtDNA4977). The role of mtDNA4977 in ischemic stroke is unknown. Methods Real-time quantitative PCR was performed to quantify total mtDNA and mtDNA4977 in leukocytes in 283 ischemic stroke cases and 135 controls. Ratios of mtDNA4977 to total-mtDNA and total-mtDNA to nuclear-DNA were calculated. Nested PCR and Sanger sequencing were used to confirm undetectable levels of mtDNA4977. Results For 191 patients and 74 control subjects in the male group and 92 patients and 61 control subjects in the female group, there were no significant between-group differences in age, cholesterol level, body mass index, stroke severity, or 4977 deletion. After adjusting for confounding factors, there was no correlation between mtDNA4977 amount and infarction risk, recurrent stroke, or stroke severity. However, mtDNA4977 was undetected in 6.94% subjects, and these individuals had a higher prevalence of stroke than those with detectable mtDNA4977 (OR: 0.181, 95% CI 0.041–0.798, p = 0.024). Additionally, mtDNA4977 status had no effect on stroke prognosis, including stroke severity and recurrent stroke. Conclusion In conclusion, there was no apparent association between mtDNA4977 deletion and cerebral infarction. Undetectable mtDNA4977 may be a marker or risk factor for ischemic stroke.


Results
For 191 patients and 74 control subjects in the male group and 92 patients and 61 control subjects in the female group, there were no significant between-group differences in age, cholesterol level, body mass index, stroke severity, or 4977 deletion. After adjusting for confounding factors, there was no correlation between mtDNA 4977 amount and infarction risk, recurrent stroke, or stroke severity. However, mtDNA 4977 was undetected in 6.94% subjects, and these individuals had a higher prevalence of stroke than those with detectable mtDNA 4977 (OR: 0.181, 95% CI 0.041-0.798, p = 0.024). Additionally, mtDNA 4977 status had no effect on stroke prognosis, including stroke severity and recurrent stroke.

Conclusion
In conclusion, there was no apparent association between mtDNA 4977 deletion and cerebral infarction. Undetectable mtDNA 4977 may be a marker or risk factor for ischemic stroke. PLOS

Introduction
Mounting evidence suggests that mitochondria dysfunction and the accumulation of mitochondria DNA (mtDNA) damage plays an important role in the development of atherosclerotic lesions. MtDNA damage not only correlates with the severity of atherosclerotic lesions but also precedes initial pathological processes [1][2][3][4][5]. The exact mechanism of this relationship is not well understood, but may be related to the central role of mitochondria in cellular energy homeostasis [5]. Evidence also suggests that somatic mtDNA alternations are associated with coronary artery disease (CAD) and atherosclerosis. Ballinger et al. [6] showed that atherosclerotic aortas had increased numbers of mtDNA oxidative lesions compared to normal aortas. Previous studies have shown that patients with CAD have an increased abundance of mtDNA 4977 bp deletions (mtDNA 4977 ) in the heart and in circulating leukocytes [7,8]. A significantly higher prevalence of mtDNA 4977 and higher relative amounts of the deletion were identified in CAD patients compared to healthy control subjects (26.2% versus 4.5%; p = 0.03 and 0.089 ± 0.02% versus 0.009 ± 0.009; p = 0.02) [9]. The accumulation of mtDNA 4977 has also been implicated in human aging [10]; in one study, the amount of mtDNA 4977 was positively associated with age, independent of traditional risk factors and clinical parameters [11].
Stroke is the third leading cause of death worldwide and is associated with a 30-day disability rate of 61.2% in Taiwan [12]. Atherothrombotic stroke, which is diagnosed when an ischemic lesion > 1.5 cm is identified on brain imaging and when there is arterial stenosis > 50% of any carotid/cerebral artery, accounts for 20% of all stroke events [13]. Atherothrombotic stroke and CAD share the same conventional risk factors, susceptibility genes, and pathophysiology [13]. Additionally, several major risk factors for ischemic stroke are associated with mitochondrial dysfunction such as that occurring in metabolic syndrome, cigarette smoke exposure, and hypercholesterolemia [14][15][16].
Based on the earlier work mentioned in the Introduction section above, we hypothesized that, similar to coronary artery disease, stroke is associated with a mtDNA 4977 deletion.
Recently, a genome-wide study reported an association between mitochondrial respiratory chain complex I/IV dysfunction and ischemic stroke, especially for small vessel stroke [2]. The common deletion mtDNA 4977 occurred between nucleotides 8470 and 13459, in which five genes for tRNA and seven genes for components of complex I (ND3, ND4, ND4L, and a part of NAD5), IV (COXIII), and V of the respiratory chain (ATPase 6 and a part of ATPase 8) were compromised [4]. Mitochondrial dysfunction, particularly involving complexes I and IV, has been associated with cardiovascular disease [4], stroke [3,8], Alzheimer's disease [17], Parkinson's disease [17][18][19][20], and psychiatric disorders [21]. While it is controversial how functions are altered by mtDNA 4977 deletion, electron movement in the respiratory chain of mtDNA 4977 cells is still likely to lead to reactive oxygen species (ROS) generation. Damaged mtDNA leads to the dysfunction of integral membrane protein complexes of the respiratory chain, which consequently leads to apoptotic death and cellular necrosis [3][4][5]15]. Therefore, we hypothesized that the pathology of mitochondrial damage in atherogenesis may be related to alterations in the generation of reactive oxygen species and in ATP synthesis.
In the present study, we thus hypothesized that mtDNA 4977 might play an important role in ischemic stroke. To date, little research has examined the presence of mtDNA 4977 deletions in the cells and tissues of patients with stroke. Therefore, we evaluated mtDNA 4977 in a cohort of patients with stroke and examined relevant associations.

Patients and control subjects
Study subjects were enrolled from Chang Gung Memorial Hospital. Each patient or her/his legally acceptable representative was informed of the aim of the study. This study has been cleared by Chang Gung Memorial Hospital Institution Ethics Review Board for human studies and patients or guardians provided written informed consent prior to study participation. Ischemic stroke and its subtypes were diagnosed and classified by two neurologists based on clinical presentation and the available brain imaging data. This study enrolled patients with atherothrombotic infarction or lacunar infarction. Subtypes of cerebral infarction were defined by the TOAST criteria with modifications as follows: (1) atherothrombotic infarction was diagnosed when there was an ischemic lesion > 1.5 cm on brain imaging and arterial stenosis > 50% of any carotid/cerebral artery; (2) lacunar infarction was diagnosed when the brainstem or subcortical hemispheric ischemia was < 1.5 cm in diameter on brain imaging, and when there was no clinical evidence of cerebral cortical or cerebellar dysfunction [22]. Patients with recent stroke (< 6 months) or a medical condition such as atrial fibrillation, acute coronary syndrome, chronic renal disease, infection or inflammatory disease, other neurodegenerative disease, or cardioembolism/undetermined stroke type were excluded. Healthy control subjects were recruited randomly from the community and had no history of stroke (including hemorrhagic or ischemic stroke), neurological disease, or overt medical disease such as chronic renal failure, myocardial infarction, atrial fibrillation, and cancer.

Evaluation of mtDNA 4977
Blood samples were collected and immediately (within 2 hours) processed for leukocyte isolation. Samples were then frozen at -80˚C and stored until analysis. Leukocyte DNA was extracted using a DNA Extraction Kit according to the manufacturer's specifications (Agilent Technologies, La Jolla, CA). Nested PCR was performed to detect the presence of mtDNA 4977 [23]. The two pairs of nested primers used for the detection of mtDNA 4977 are shown in Fig 1. The PCR condition was set as Table 1. The PCR condition was set as pre-denaturation at 94˚C for 5 min, 30 cycles at 94˚C for 10 secs, 58˚C for 45 secs and 72˚C for 50 secs, and a final extension at 72˚C for 10 min. We used the ratio of the absorbance at 260 and 280 nm (A260/280) to assess the purity of nucleic acids. The DNA samples were all with good purity (260/280 = 1.7-2.0). Then we run the DNA samples by standard protocol for performing 1% agarose gel electrophoresis to detect other possible contaminants free RNA or faint smear. The samples were run in triplicates. Products were sequenced by Sanger sequencing. The presence of mtDNA 4977 was indicated by the appearance of a 358-bp band and verified by sequencing analysis.

Quantification of mtDNA 4977
To quantify the relative ratio of mtDNA 4977 to total mitochondria DNA, real-time quantitative PCR (RTQ-PCR) was performed to quantify total mtDNA and mtDNA 4977 molecules in leukocytes. TaqMan 1 Gene Expression Assays and the Sequence Detection system (ABI Prism 7900, Applied Biosystems) were used for quantification (Table 2). β-actin gene expression was measured as an indicator of nuclear DNA concentration. The common mtDNA deletion of 4,977 bp had breaking points between nucleotides 8470 and 13347, with two direct 13-bp repeats. The ND1 gene used for reference was located at nucleotides 3312 to 4122 on mtDNA and provides one of the seven mitochondrial encoded subunits of NADH dehydrogenase.
Primers and TaqMan probes ordered with Applied Biosystems™ FAM ™ and VIC ™ dyes, used to amplify mtDNA 4977 , ND1 gene and β-actin gene. The Ct method (ΔCt) was used to calculate relative levels of mtDNA 4977 between patients and control subjects. First, we normalized the Ct endogenous control: ΔCt sample = Ct Target (4977 deletion)-Ct Endogenous Control

Assay ID Primers and VIC-probe
Forward Primer: The common mtDNA deletion of 4,977 bp had breaking points between nucleotides 8461 and 13347, with two direct 13-bp repeats. The fragment was the amplification target between 8461(1 st repeat:acctccctcacca) and 13347 bp (2 nd repeat:acctaaatcacca). The grey arrow at the outer circle shows mtDNA deletion of 4,977 bp, and the grey arrow at the inner circle demonstrates the product of the second nested PCR. In the presence of deletion, the amplification product is 358 bp in size, while the wild-type mtDNA yields a product of 5335 bp, which shall not be shown on sequencing. The left corner shows the Sanger sequence that proved the 2 nd nested PCR products and confirmed of the 4,977 bp deletions.

Statistical analysis
Pearson's χ 2 -tests or t-tests were utilized to compare demographic data between the patient and control groups; all significance tests were two-tailed. Association analyses were performed first stratified by sex and then combined. Co-variables included age, sex, hypertension, diabetes mellitus (DM), total cholesterol (TC) level, smoking, and alcohol use. A logistic regression was performed to assess the relationship between mtDNA 4977 and stroke outcome. All data analyses were performed using IBM SPSS Statistics software version 20. The threshold for statistical significance was p < 0.05.

Results
Study group characteristics are presented in Table 3. A total of 283 patients and 135 control subjects were included. In the male group, hypertension and smoking were significantly more common in patients with stroke than in control subjects (p = 0.008 and p = 0.016, respectively). In the female group, hypertension and DM were more common in patients with stroke than in control subjects (p = 0.006 and p < 0.001, respectively). When the groups were combined, patients with ischemic stroke showed higher prevalence of hypertension (p < 0.001), DM (p < 0.001), smoking (p < 0.001), and alcohol consumption (p = 0.046) compared to the control group. There were no significant differences in age, cholesterol level, body mass index, initial Glasgow coma scale (GCS) scores ( 9), ΔCt samples, and ΔCt reference values between the sexes. GCS scores were later dichotomized at 8, as this is the clinical cutoff point often used for coma [24]. We thus used an initial GCS score of 9 as an indicator of initial conscious level and stroke severity. Additionally, higher percentages of mtDNA 4977 were detected in the control group than in the ischemic stroke group (male group: control vs. stroke, 97.3% vs. 91.1%; female group: control vs. stroke, 100% vs. 89.1%; ischemic stroke group vs. control group, p = 0.002). The average intra -coefficient of variation (CV) and inter-CV of ΔCt (mtDNA 4977 -ND1) is 2.1% and 6.8%, respectively. Of the study cohort subjects (29 study subjects), 6.94% did not carry mtDNA 4977 , including 27 patients with infarction and two healthy control subjects. In these patients, deletion of 4977 was undetectable using TaqMan quantification and confirmed by Sanger sequencing. There was no difference in mean age, stroke risk factors, or ΔCt reference values between participants with and without detectable mtDNA 4977 (Table 4). Participants without detectable mtDNA 4977 had a higher rate of infarction than patients with detectable mtDNA 4977 (male group: 89.5% vs. 70.7%, p = 0.079; female group: 100% vs. 57.3%, p = 0.008; ischemic stroke group vs. control group, p = 0.002) ( Table 4).

Discussion
The mtDNA 4977 deletion is particularly relevant to the pathogenesis of atherosclerosis and has previously been reported as a potential predictor of coronary artery disease and stroke; yet, the relationship between mtDNA 4977 and ischemic stroke remains unclear. This study found no correlation between mtDNA 4977 amount and infarction risk, recurrent stroke, or stroke severity in a cohort of patients with acute ischemic stroke. Additionally, the presence or absence of mtDNA 4977 had no effect on stroke prognosis, including recurrent stroke and stroke severity. In our cohort of 283 patients and 135 control subjects, 6.94% of cohort subjects did not carry mtDNA 4977 , and there was a significant higher prevalence of stroke in these individuals compared to those with detectable mtDNA 4977 . In conclusion, there was no apparent association between mtDNA 4977 deletion and cerebral infarction. Undetectable mtDNA 4977 may be a marker or risk factor for ischemic stroke. MtDNA 4977 are related to ROS that arise as a by-product of oxidative phosphorylation in mitochondria [25]. MtDNA 4977 occur frequently in tissues of high oxygen demand and low mitotic activity, e.g. neuron cell or myocytes [26]. However, mtDNA 4977 can still be detected in fast replicating cells, such as blood leukocytes, in much lower amounts [25,26]. Botto et al. reported that mtDNA 4977 can be found in both samples of blood cells and atherosclerotic lesions from patients with CAD [9]. Although mtDNA 4977 in blood could not represent specifically for cerebral infarction, it is still a biomarker for systemic burden of oxidative stress. Because the brain tissue samples from stroke patients are difficult to access, biomarkers in blood is relatively more applicable.
MtDNA is more susceptible to oxidative damage than nuclear DNA [6] because of a limited mitochondrial capacity for DNA repair and the absence of histones for protection [8]. Numerous studies have implicated mtDNA 4977 in various forms of carcinogenesis [27] as well as in aging [28]. mtDNA 4977 is more frequent in the general population than previously thought; it was recently demonstrated that mtDNA 4977 was present in 98.3% of subjects in a cohort of unrelated Chinese participants aged 5 days to 91 years, with an exponential increase in the frequency of mtDNA 4977 with age [18]. This is consistent with the observation of mtDNA 4977 in 93.06% of subjects (389/418) in our study cohort. The discrepancy between earlier and these current findings are likely related to cohorts of a wider age range as well as technological improvements in DNA detection.
MtDNA 4977 in cardiac myocytes and smooth muscle cells in atherosclerotic lesions are important contributors to pathogenesis of coronary artery disease (CAD) [29,30]. Accumulation of mtDNA 4977 is particularly found in high oxygen demand and low mitotic activity tissues. Regional accumulation of mtDNA 4977 was higher in the left heart than in the right because the left heart has a higher workload and thus requires higher energy demand [31]. Regarding the hypothetical relationship between mitochondrial damage in atherogenesis and the generation of ROS and ATP synthesis, our data demonstrates that the presence of mtDNA 4977 did not vary in accordance with a history of ischemic stroke. Additionally, mtDNA 4977 levels were not significantly associated with recurrent stroke or stroke prognosis. Our results do not support the notion that mtDNA 4977 plays a major role in cerebral vascular arthrosclerosis and cerebral infarction. In one study, mtDNA dysfunction and damage preceded early atherosclerotic changes, while oxidative markers remained normal, suggesting that changes were probably independent of oxidative stress [5]. Moreover, stroke syndromes caused by defects of mtDNA and nuclear DNA may be different, and recent clinical reports suggest that small vessel diseases are associated with diverse nuclear genetic variants [2,32]. Because accumulation of mtDNA 4977 is found in high oxygen demand and low mitotic activity tissues, such as neuron cells and myocytes, prior CAD models demonstrated increasing mtDNA 4977 and ROS production in coronary artery, aorta, and cardiac myocytes. In contrast, leukocytes are fast replicating with high mitotic activity, which may cause insufficient time for mtDNA 4977 accumulation. Utilizing leukocytes from peripheral blood, our study showed that undetectable mtDNA 4977 in leukocytes may be a biomarker for ischemic stroke. Because the brain tissue samples from stroke patients are difficult to access to confirm the link between undetectable mtDNA 4977 and stroke, further animal model for confirmation is required before utilizing peripheral mtDNA 4977 as a biomarker for ischemic stroke.
A meta-analysis of 38 studies found that proportions of mtDNA 4977 were significantly decreased in cancerous tissue compared to adjacent non-cancerous tissue [33]. Another previous study showed that the frequency of mtDNA 4977 was significantly higher in normal tissue compared to paired cancerous tissues including breast [34,35], lung [36], esophageal squamous cell carcinoma, gastric [37], and colorectal cancers [23]. The possible mechanism underlying this relationship may be that mtDNA 4977 serves a protective function against the tumorpromoting effects of other somatic mutations [38] or clonal expansion dilution of mtDNA 4977 in tumor tissue during cancer progression. Moreover, mtDNA 4977 might confer a metabolic disadvantage to hyper-proliferating cells and thus eliminate these cells by leading to apoptosis [29]. This may explain our finding that only a low percentage of individuals in our cohort (6.94%) did not carry mtDNA 4977 , and the finding that stroke was more prevalent in these individuals.
A limitation of the present study is that patients with cardiovascular problems such as atrial fibrillation and acute coronary syndrome, as well as patients with severe stroke (initial GCS score 9; male: 6.1%, female: 14.7%) were excluded. Second, the small sample size is another limitation of this study. Based on the values of mtDNA 4977 in this study, at the level of 0.05, we will need approximate 1,200 case-control pairs to achieve an adequate power of 80% to detect differences in the level of mtDNA 4977 between the ischemic cases and the control group.
Future work should include a larger and more generalizable cohort to examine correlations between mtDNA 4977 and clinical stroke occurrence or stroke outcome.

Conclusions
MtDNA 4977 does not play a significant role in cerebral infarction. Inability to detect mtDNA 4977 may be a marker or risk factor for ischemic stroke.