Mitochondrial DNA damage and vascular function in patients with diabetes mellitus and atherosclerotic cardiovascular disease

Type 2 diabetes mellitus affects an estimated 1 in 10 Americans and this number is expected climb with the current obesity epidemic [1]. Diabetes mellitus is a significant risk factor for cardiovascular disease; however, the mechanisms behind this increased risk are incompletely understood [1, 2]. Elevated oxidant levels have been shown to contribute to vascular dysfunction both in animal models and clinical studies [3‐5]. Mitochondria are an important source and target of oxidants that may contribute to vascular disease in diabetes mellitus [6‐8].

Mitochondrial DNA is more susceptible to oxidative damage compared to nuclear DNA due to multiple factors including a limited repair capacity and close proximity to the electron transport chain [9, 10]. Mitochondrial DNA damage has been closely associated with dysfunctional oxidative phosphorylation, which leads to further oxidative stress resulting in a positive-feedback cycle. In an animal model of atherosclerosis, excess mitochondrial DNA damage promoted atherosclerosis and plaque vulnerability through increased monocyte activation [11]. In a prior human study of patients with coronary artery disease, the extent of mitochondrial DNA damage in circulating white cells was associated with high risk plaque burden [12]. We have previously described altered mitochondrial oxidative phosphorylation, membrane potential and morphology in peripheral blood mononuclear cells which was associated with vascular dysfunction in patients with diabetes [7, 13]. The objective of the present study was to assess the relation of mitochondrial DNA damage in peripheral blood mononuclear cells to vascular function and the presence of diabetes mellitus and atherosclerotic cardiovascular disease.

In this study, we evaluated mitochondrial DNA damage in peripheral blood mononuclear cells and related the findings to vascular function in a large group of patients with diabetes mellitus, clinical atherosclerosis, and controls. We observed higher mitochondrial DNA damage in peripheral blood mononuclear cells in association with both diabetes mellitus and clinical atherosclerotic disease when accounting for concomitant cardiovascular risk factors including age. We observed similar levels of mitochondrial DNA damage in patients with diabetes mellitus compared to established atherosclerotic disease, which is consistent with the concept of diabetes as a coronary disease equivalent. No differences in mitochondrial DNA copy number were observed in any of the patient groups relative to the controls suggesting that the differences in mitochondrial DNA damage were not due to alterations in mitochondrial DNA copy number. Mitochondrial DNA damage was associated with higher baseline pulse amplitude consistent with an alteration of microvascular pulsatility. In contrast, there were no associations of mitochondrial DNA damage with measures of vasodilator responses in either conduit or microvessels. Taken together, our findings provide evidence that diabetes mellitus and atherosclerotic cardiovascular disease are associated with increased mitochondrial DNA damage that may be relevant to oxidative stress and microvascular dysfunction.

Prior work indicates a link between mitochondrial alterations, diabetes mellitus and atherosclerosis. In previous work by our group, we found that peripheral blood mononuclear cells from patients with diabetes mellitus had an abnormal mitochondrial phenotype characterized by higher mitochondrial oxidant production, mitochondrial membrane hyperpolarization, higher uncoupled respiration, and fragmented mitochondria [7, 13]. In the current study, we have extended our previous work by measuring mitochondrial DNA damage in circulating white cells. Mitochondrial DNA damage is postulated as a measure of environmental injury to the mitochondria that may mediate mitochondrial dysfunction. The ability to measure mitochondrial DNA damage in circulating white cells has potential utility as a biomarker of mitochondrial injury. Damage to mitochondrial DNA has been shown to mediate mitochondrial dysfunction including alterations in calcium handling, oxidant-mediated activation of cell stress pathways, and metabolic disturbances [5, 11, 12, 18]. Oxidative stress has been shown to induce mitochondrial DNA damage in vascular cells (endothelial and vascular smooth muscle cells) [19]. Additionally, mitochondrial DNA damage has been shown to be the result of replication errors and repair insufficiencies with the damaged mitochondrial DNA undergoing clonal expansion [20, 21]. Animal models with a genetic predisposition to increased mitochondrial DNA damage display mitochondrial dysfunction and accelerated atherosclerosis [11, 18]. Interestingly, enhanced mitochondrial DNA damage was associated with abnormal metabolic function including glucose intolerance consistent with a connection to diabetes mellitus [16]. Further, monocytes with increased mitochondrial DNA damage were pro-inflammatory and were associated with increased plaque vulnerability [17].

Several prior studies have measured specific mitochondrial DNA changes in blood cells from patients with atherosclerosis and diabetes mellitus. Patients with coronary artery disease were found to have higher levels of the common mitochondrial DNA 4977 bp deletion in circulating peripheral blood mononuclear cells compared to age matched, healthy controls [22]. In a study of patients with coronary artery disease, higher mitochondrial DNA damage was associated with vulnerable plaque characteristics and history of diabetes mellitus [18]. In an additional study, peripheral blood cells from diabetic patients had higher levels of a specific mitochondrial DNA point mutation (A3243G) compared to non-diabetic controls [23]. Using a sensitive quantitative PCR based technique to determine mitochondrial DNA damage, we demonstrate the presence of higher mitochondrial DNA damage in peripheral blood mononuclear cells both in patients with diabetes mellitus and with clinical atherosclerosis [10, 17]. This method allowed us to assess damage throughout the mitochondrial genome rather than at a single nucleotide position and is sensitive to strand-breaks, deletions, dimers, cross-linking, and bulky adducts allowing us to assess a number of different types of damage. Our findings suggest the independent association of both metabolic disease and atherosclerosis with mitochondrial DNA damage even when accounting for associated cardiovascular risk factors. The greater extent of mitochondrial DNA damage may drive increased production of mitochondrial reactive oxygen species and accentuate vascular damage in diabetes mellitus. The association of mitochondrial DNA damage with atherosclerosis suggests that mitochondrial DNA damage measures in a blood sample may serve as novel biomarkers of cardiovascular injury. Future studies are needed to evaluate the predictive value of mitochondrial DNA damage for cardiovascular events in a longitudinal study.

Several lines of evidence indicate the relevance of mitochondrial phenotype to vascular function [24]. Mitochondrial dysfunction in the endothelium disrupts nitric oxide signaling and impairs vasodilation [6, 25]. Relevant to the present work, studies of peripheral blood mononuclear cells have shown an association of lower mitochondrial mass, higher mitochondrial superoxide production, and altered oxidative phosphorylation with impaired vascular function in diabetes mellitus [7, 13]. We now show that mitochondrial DNA damage relates to higher baseline pulse amplitude measured in the fingertip circulation. Baseline pulse amplitude measures microvessel pulsatility and is governed by blood flow, vessel compliance and sympathetic tone. Higher small vessel pulsatility may lead to microcirculatory injury and has been associated with several cardiovascular risk factors [15, 26]. In one prior study of high risk patients, baseline pulse amplitude was not associated with large vessel cardiovascular events; however, the predictive value for microcirculatory events is not known [27]. The relation of mitochondrial DNA damage to higher pulse amplitude may reflect an impact on resting sympathetic tone or small vessel compliance.

In contrast, we did not observe associations of mitochondrial DNA damage with measures conduit and microvascular vasodilator function. The present finding differs from our prior studies showing impaired flow-mediated dilation and peripheral arterial tonometry hyperemic response in the presence of mitochondrial dysfunction. The lack of association may suggest that mitochondrial DNA damage is not an important mediator of endothelial dysfunction. It is also possible that mitochondrial DNA damage does not directly modulate mitochondrial reactive oxygen species generation in peripheral blood mononuclear cells. A recent study demonstrated that a novel regulator of kinases and phosphatases, prolyl-isomerase-1 (Pin1), mediates alterations in mitochondrial oxidant production and downstream vascular signaling pathways relevant to vascular tone and inflammation [28]. Pin1 levels were higher in the presence of diabetes and associated with impaired endothelial function and higher oxidative stress biomarkers, consistent with a connection between mitochondrial DNA damage and vascular function in patients [28]. Hence, it remains possible that greater mitochondrial DNA damage extent or injury in vascular cells would lead to vasodilator dysfunction. Additional studies will be necessary to define whether the microvascular changes observed with mitochondrial DNA damage in peripheral blood mononuclear cells contribute to target-organ consequences and the relevance of mitochondrial DNA damage in cells types important to vascular disease including leukocytes, platelets, endothelial cells, and smooth muscle cells.

Our study has a number of limitations. We assessed mitochondrial DNA damage in peripheral blood mononuclear cells and not a vascular tissue such as endothelial cells or vascular smooth muscle cells from patients. However, peripheral blood mononuclear cells are a readily accessible cell type in human subjects and are exposed to the same systemic risk factors. Additionally, due to the study design, we are limited to identifying correlations of mitochondrial DNA damage with vascular function rather than causal relations. Some of the types of mitochondrial DNA damage measured using the quantitative PCR technique may not affect mitochondrial function. However, we did observe an association of mitochondrial DNA damage with atherosclerosis and diabetes mellitus. One of the reasons for a lack of association of several vascular function measures with mitochondrial DNA damage may be due to the global assessment of mitochondrial DNA damage rather than measuring damage or mutation of a specific site of mitochondrial DNA. Further, mitochondria are responsible for multiple cellular activities including calcium storage, apoptosis, metabolism, regulation of cell growth, and heme biosynthesis. The optimal assessment of mitochondrial function is likely to combine multiple facets including mitochondrial oxidative stress, mitochondrial structure, mitochondrial complex activity, and mitochondrial DNA damage. Future work will be needed to assess the relationship of the multiple facets of mitochondrial function with vascular measures. We found an association of mitochondrial DNA damage with the presence of clinically evident atherosclerosis; however, we do not have measurements of subclinical atherosclerosis such as carotid-intima media thickness in our patient sample. Finally, further studies will be needed to define the mechanisms linking mitochondrial DNA damage to risk factors and to microvascular alterations. Strengths include the relatively large sample size, the comprehensive noninvasive assessment of vascular function, and the use of established quantitative PCR based methodology to measure mitochondrial DNA damage.

Our main findings indicate an association of mitochondrial DNA damage in peripheral blood mononuclear cells with microvascular pulsatility but not vasodilator function. Furthermore, both diabetes mellitus and clinical atherosclerotic disease associate with greater mitochondrial DNA damage consistent with prior work showing abnormalities in mitochondrial function. Additional work is required to elucidate the contribution of mitochondrial phenotype in microvascular disease and whether therapies aimed at protecting the mitochondria from DNA damage will have cardiovascular benefit.