Circulating Oxidized mtDNA is Associated Broadly with Cardiovascular Disease in a Longitudinal Cohort Study of Psoriasis

Psoriasis (PSO) is a chronic and systemic inflammatory autoimmune disease associated with atherosclerosis and myocardial infarction. Given that atherosclerosis is both inflammation and immune driven, we sought to expand on known immune and inflammatory biomarkers in a PSO cohort. In this study, we focus on oxidized mtDNA (ox-mtDNA), a product of cells undergoing pyroptosis, including keratinocytes, which was quantified in patients with PSO and individuals without PSO by ELISA. Patients with PSO had significantly higher ox-mtDNA levels than healthy subjects (mean ± SD = 9246 ± 2518 pg/ml for patients with PSO vs 7382 ± 2506 pg/ml for those without; P = .006). Importantly, ox-mtDNA was positively associated with IL-17a (β = 0.25; P = .03) and low-density granulocytes (β = 0.37; P = .005) but negatively associated with high-density lipoprotein-cholesterol (β = −0.29; P = .006). After adjusting for traditional cardiovascular risk factors, we found that ox-mtDNA was associated with noncalcified coronary burden, which was measured by coronary computed tomography angiography (β = 0.19; P = .003). Biologic-naïve patients with PSO receiving anti–IL-17a therapy had a 14% decrease in ox-mtDNA (mean ± SD: 10540 ± 614 pg/ml at baseline to 9016 ± 477 pg/ml at 1 year; P = .016) and a 10% reduction in noncalcified coronary artery burden (mean ± SD: 1.06 ± 0.45 at baseline, reducing to 0.95 ± 0.35 at 1 year; P = .0037). In summary, levels of ox-mtDNA in PSO are associated with measures of coronary plaque formation, indicating that this biomarker may be an autoimmune-driven early atherosclerotic feature.


INTRODUCTION
Myocardial infarction is the leading cause of death worldwide (Global Burden of Metabolic Risk Factors for Chronic Diseases, 2014).Atherosclerosis, the primary cause of myocardial infarction, is a disease driven by lipid infiltration, inflammatory cell activation, and metabolic derangement (Libby et al, 2009).To better understand the link between chronic inflammatory states and atherosclerosis, we have utilized psoriasis (PSO) as a human model for the study of coronary artery disease (CAD).PSO is a chronic inflammatory disease with increased cardiovascular risk, including subclinical atherosclerosis (Elnabawi et al, 2019b;Mehta et al, 2010).Despite being younger than patients with traditional cardiovascular risk factors, patients with PSO have a higher prevalence of atherosclerotic plaque (Lerman et al, 2017).Furthermore, atherosclerosis, particularly the development of CAD, is associated with PSO severity (Baumer et al, 2018;Elnabawi et al, 2019a;Lerman et al, 2017); yet, the underlying mechanisms are not completely understood.
We previously reported that in PSO, a unique subpopulation of neutrophils (low-density granulocytes [LDGs]) is upregulated in circulation compared with that in controls, and the frequency of LDGs correlated with CAD as assessed by coronary computed tomography angiography (CCTA) (Teague et al, 2019).LDGs have elevated mitochondrial superoxide production, and the DNA released by neutrophil extracellular traps (NETs) is highly enriched in oxidized mtDNA (ox-mtDNA), which enhances the formation of NETs and triggers the inflammasome, a macromolecular complex that directs innate immune activation, within neutrophils (Lood et al, 2016).Notably, ox-mtDNA is implicated in the pathogenesis of atherosclerosis (Nakayama and Otsu, 2018).In atherogenic mouse models, inflammasome activation is associated with NET formation within nascent atherosclerotic plaques (Westerterp et al, 2018).
Taken together, we hypothesized that elevated circulating ox-mtDNA would be (i) observed in patients with PSO, (ii) associated with measures of immune dysfunction (as assessed by LDGs), and (iii) associated with noncalcified coronary plaque burden (a CCTA marker of subclinical atherosclerosis).
Values are reported as mean AE SD or median (IQR) for continuous data and n (%) for categorical data.Continuous data were compared using t-test for parametric and Wilcoxon rank-sum for nonparametric observations.Groups containing categorical data were compared using Pearson's chi-square test.
1 Patients self reported.
SS Lateef et al.
ox-mtDNA Associated with CVD in Psoriasis ox-mtDNA and apolipoprotein B (Table 2) (b ¼ 0.30; P ¼ .004),an atherogenic lipid species.Taken together, these data demonstrate that ox-mtDNA tends to be positively correlated with increased early predictors of high cardiovascular risk.
ox-mtDNA was associated with measures of inflammatory and immune dysfunction In addition to lipid dysfunction, markers of immune dysfunction were also associated with ox-mtDNA (Table 2).
We noted positive associations between ox-mtDNA and IL-17a (b ¼ 0.25; P ¼ .03)as well as with LDGs, a proinflammatory subpopulation of neutrophils (b ¼ 0.37; P ¼ .005).In contrast, an association between ox-mtDNA and normal-density granulocytes was not observed (b ¼ 0.04; P ¼ .78),suggesting that the increased ox-mtDNA in PSO may be related to the well-established increase in neutrophils, potentially related to NLRP3-mediated NETosis (Shao et al, 2019).
ox-mtDNA was directly associated with subclinical atherosclerosis Among patients with PSO, ox-mtDNA was directly associated with CAD as assessed by noncalcified coronary plaque burden (Table 2) (b ¼ 0.27; P < .001).This relationship was maintained in models adjusted for traditional cardiovascular risk factors (eg, Framingham risk score, statin therapy, anthropomorphic measure of obesity) and systemic inflammation (eg, PSO severity as assessed by PASI score and highsensitivity C-reactive protein) (b ¼ 0.19; P ¼ .003).

Longitudinal analysis of ox-mtDNA by biologic treatment type
In our PSO cohort, 13 patients were biologic naı ¨ve at baseline but undertook antieIL-17a therapy (secukinumab or ixekizumab) for 1 year.We postulated that patients with PSO undergoing antieIL-17a biologic therapy for 1 year would also exhibit a reduction in circulating ox-mtDNA (Figure 1).Indeed, the patients receiving a year of antieIL-17a biologic therapy had a 14%  Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein.
Data are represented as standardized b-coefficient and P-value.
1 Patients self reported.
SS Lateef et al.
ox-mtDNA Associated with CVD in Psoriasis decrease in circulating ox-mtDNA (Figure 1) (mean AE SD ¼ 10540 AE 614 pg/ml at baseline to 9016 AE 477 pg/ml at 1 year; P ¼ .016).A decrease in circulating ox-mtDNA was also observed in patients undertaking antieTNF-a therapeutics for 1 year (Figure 1) (mean AE SD ¼ 8554 AE 781 pg/ml at baseline vs 8561 AE 497 pg/ml) at 1-year follow-up, but this decrease was not significant (P ¼ .50).In summary, we have discovered that ox-mtDNA is an IL-17aedependent biomarker in patients with PSO that is associated with CAD.

DISCUSSION
In this cross-sectional study of PSO, we present 4 key findings: (i) that circulating ox-mtDNA is higher in patients with PSO than in a control subpopulation; (ii) a positive association between ox-mtDNA and both IL-17a and LDG levels; (iii) an inverse association between ox-mtDNA and antiatherogenic markers high-density lipoprotein cholesterol and apolipoprotein A1 but a positive association with markers known to be proatherogenic such as high triglycerides and apolipoprotein B; and (iv) in adjusted models for cardiovascular risk factors and systemic inflammation, ox-mtDNA is strongly associated with noncalcified coronary plaque burden, an imaging marker for CAD.Although, to our knowledge, previously unreported as a biomarker for PSO, ox-mtDNA has been demonstrated as a biomarker for other inflammatory diseases such as lupus and metabolic syndrome, which have both been recognized as risk factors for cardiovascular diseases (CVDs) (Lood et al, 2016;Ye et al, 2023).Elevated circulating ox-mtDNA has also been seen in malignancies such as myelodysplastic syndrome, which is distinct among other myeloid malignancies for exhibiting elevated inflammation in the bone marrow microenvironment (Ward et al, 2021).This elevated inflammation is driven by NLRP3 inflammasome assembly and pyroptotic cell death (Ward et al, 2021).Furthermore, the authors also reported a positive association between ox-mtDNA and members of the alarmin family of proteins, S100A8 and A9.(Ward et al, 2021).In myelodysplastic syndrome, ox-mtDNA contributes to feed-forward bone marrow failure through toll-like receptor 9, an endosomal DNA-sensing pattern recognition receptor known to prime and activate the inflammasome propagating the type 1 IFNeinduced inflammatory response in neighboring healthy hematopoietic stem and progenitor cells (Ward et al, 2023).
Although we have yet to investigate inflammasome assembly and pyroptosis in PSO, enhanced inflammasome activity has been reported by other groups (Garshick et al, 2019;Verma et al, 2021).In addition, dysregulation of both pyroptotic and ferroptotic pathways has been observed in psoriatic patients (Zhang et al, 2022).The inflammasome is activated in PSO as measured through transcriptomic studies of caspase-5 and IL-1b and proteomic studies of IL-6 within vascular endothelial cells (Garshick et al, 2019).We have previously demonstrated that the S100A8/A9 dimer (calprotectin) is increased in the serum of patients with PSO (Naik et al, 2015), and in the same cohort, we have also found elevations of circulating S100A7, A8, and A12 (Playford MP, unpublished data).An association between plasma ox-mtDNA/glucose ratio has also been observed with S100A9, S100A12, and the S100A8/A9 dimer (Florida et al, unpublished data).
LDGs are a subpopulation of neutrophils that increase NET release (NETosis) and subclinical atherosclerosis as measured by noncalcified coronary plaque burden in inflammatory diseases (Lood et al, 2016;Teague et al, 2019).In this study, we identified an association between ox-mtDNA, IL-17a, and LDGs, potentially linking IL-17a signaling with the release of oxidized nucleic acid species into NETs produced by LDGs.IL-17a also promotes expression of S100A8 and S100A9 during the inflammatory response of keratinocytes (Christmann et al, 2021).We also show that antieIL-17a intervention significantly reduces circulating ox-mtDNA (Figure 1); further preliminary data demonstrate that similar therapy also reduces LDGs (Teague et al, unpublished data).
The mechanism by which IL-17a increases oxidation on mtDNA is not known.Although not a cell of myeloid origin, we propose that the HaCaT keratinocyte cell line may be used to assess such a mechanism.To this end, we have pilot studies that show that a modest increase in mitochondrialoxidized nucleoids was observed with IL-17a but not with TNF-a treatment (Pantoja, unpublished data).
Recent data have proposed an IL-17aeextracellular signaleregulated kinase/NLRP3/caspase-1 signaling pathway leading to pyroptosis in nasal epithelial cells (Li et al, 2022).Pyroptosis is a putative component of the inflammatory response leading to cardiovascular injury, including coronary artery pathology.In previous studies, we have associated CCTA markers of subclinical atherosclerosis, such as ruptureprone noncalcified coronary artery burden and high-risk plaque features, with PSO severity (Lerman et al, 2017).In this study, we noted an association between ox-mtDNA and noncalcified coronary artery burden.Our findings are consistent with those of previous studies of the chronic inflammatory disease PSO and show that ox-mtDNA may be an informative marker to identify patients with PSO who will develop immune, lipid, and metabolic dysfunction as well as early atherosclerosis in PSO.
Our study is limited because it is cross-sectional and does not track serial measurements of the blood marker ox-mtDNA.Our assessment of cardiovascular outcomes would be enhanced by the addition of hard cardiovascular event data.Tracking the incidence of major cardiovascular events such as revascularization, transient ischemic attack, or myocardial infarction would lend to longitudinal analysis but is difficult in our cohort owing to the relatively short duration of the study and the relatively young age of our study population.To further characterize this chronic and systemic inflammatory disease, detailed studies of immune cell populations, including LDGs, and inflammatory processes, particularly the release of NETs and ox-mtDNA, in PSO are needed.
In summary, we have characterized the systemic release of ox-mtDNA and subsequent subclinical atherosclerosis in a large cohort of patients with PSO.ox-mtDNA relates to immune, lipid, and cytokine markers at baseline and modulates with targeted biologic therapy.This work has implications for the inflammatory pathogenesis of autoimmunedriven CVD.Future mechanistic studies must further SS Lateef et al.
ox-mtDNA Associated with CVD in Psoriasis elucidate these pathways and potential therapeutic inhibitors to target them.

Demographic and clinical characteristics of PSO and control cohorts
Our study included a cohort of consecutively recruited patients with PSO (n ¼ 89) from January 2013 to November 2019 as part of the Psoriasis, Atherosclerosis, and Cardiometabolic Disease Initiative (NCT01778569) at the National Institutes of Health (NIH) Clinical Center.Participants aged !18 years with PSO consisting of typical skin findings and associated joint, nail, and/or hair findings clinically diagnosed by a physician were included in this study.Participants were excluded if they had severe renal excretory dysfunction; estimated glomerular filtration rate <30 ml/min/1.73m2;existing CVD; and any comorbid condition known to be associated with CVD or systemic inflammation, such as uncontrolled hypertension, internal malignancy within 5 years, HIV, active infection within 72 hours of baseline, major surgery within 3 months, and/or pregnancy or lactation.Physician consent was required for donors to enroll in this program, including that they must be healthy per the physician's standards.
For comparison, a cohort of healthy volunteers (NCT01934660) (n ¼ 14) was recruited from 2013 to 2019 (Figure 2).Patients were excluded if they had any comorbid condition known to promote CVD or systemic inflammation, such as known CVD, uncontrolled 89 patients with psoriasis total for baseline assessment ox-mtDNA and by CCTA.hypertension, internal malignancy within 5 years, HIV, active infection within the past 72 hours of baseline, and major surgery within 3 months.This was a pilot study where consideration of sample size or power was not a priority.To the best of our knowledge, no subjects lacked any of the data points presented.All patients provided written informed consent, and all study protocols were approved by the institutional review board at the NIH.All study protocols are in compliance with the Declaration of Helsinki.
Patients were asked to complete survey-based questionnaires regarding smoking, previous CVD, family history of CVD, and previously established diagnoses of hypertension and diabetes.Patient responses were then confirmed during history and physical examination by the study provider.CVD included acute coronary syndrome comprising both myocardial infarction and unstable angina pectoris; angina pectoris; cerebrovascular event; transient ischemic attack; peripheral vascular disease; and revascularization procedures, including both coronary artery bypass grafting and percutaneous interventional procedures.Diabetes was defined as fasting glucose !126 mg/dl, glycated hemoglobin >6.5%, or use of diabetic medication.Hypertension was defined as systolic blood pressure !140 mmHg, diastolic blood pressure !90 mmHg, or use of antihypertensive medication.Hyperlipidemia was defined as total cholesterol >5.18 mmol/l, low-density lipoprotein cholesterol !4.14 mmol/l, or high-density lipoprotein cholesterol 1.04 mmol/ l.Hypertriglyceridemia was not included in the assessment of hyperlipidemia.A 10-year Framingham risk score was assessed for all patients.
All patients underwent fasting blood draws for the assessment of the lipid panel at baseline, including high-and low-density lipoprotein cholesterol.A dermatologist or physician evaluated all patients with PSO for the assessment of PSO skin disease severity through the PASI score.

CCTA
All patients underwent blood draw (plasma collection) and CCTA on the same day in the same computed tomography scanner (320detector row Aquilion ONE ViSION).Guidelines established by the NIH Radiation Exposure Committee were followed.Scans were performed with retrospective gating at 120 kV and tube current of 750e850 mA, with a gantry rotation time 420 ms.Imageacquisition characteristics included slice thickness of 0.75 mm and pitch of 0.2e0.4.Coronary artery phenotyping was performed using QAngio CT (Medis Medical Imaging) with high (interclass correlation > 0.95) intraobserver and interobserver agreement and with readers blinded to patient characteristics as previously published (Lerman et al, 2017).Manual adjustment of inner lumen and outer vessel wall delineations was performed if needed.Measurements of noncalcified coronary artery burden (mm 2 ) were determined by dividing noncalcified volume (mm 3 ) by vessel length (mm) and were attenuated for luminal intensity measures for accuracy.Consecutive patients with PSO were then assessed for change in subclinical atherosclerosis by CCTA at 1-year follow-up.

Measurements of inflammatory markers
Oxidized DNA was quantified in human plasma using the DNA/ RNA oxidative damage (high sensitivity) ELISA kit (Cayman Chemical Company; item number 589320).Plasma glucose concentration, measured by colorimetric assay (Cayman Chemical Company), was performed to control for hyperglycemia-induced inflammasome activation.The indicated cytokines were measured using a custom multiplex assay (Uplex, Meso Scale Diagnostic) following the manufacturer's instructions.
330 patients with psoriasis consecutively recruited for baseline assessment fromJanuary 2013 until November 2019.psoriasis consecutively returned for one-year follow-up.20excluded: patients lacked had an uninterpretable CCTA at one-year followup plasma.37patients with psoriasis total for one-year follow-up assessment of ox-mtDNA and by CCTA.32 excluded: patients with psoriasis who have not yet arrived for scheduled one-year follow-up visits.

Figure 2 .
Figure 2. Recruitment schematic for patients with psoriasis at the National Institutes of Health.CCTA, coronary computed tomography angiography; ox-mtDNA, oxidized mtDNA.

Table 1 .
Baseline Characteristics of Patients with Psoriasis and Controls Figure1.Change in ox-mtDNA levels over 1 year stratified by biologic treatment type.Data are presented as median AE interquartile range.Outliers are represented by a single dot (eg, within the 1-year data for the antieIL-17a group).Comparisons were made using paired t-test.There was a significant decrease in ox-mtDNA levels over 1 year in the group receiving antieIL-17a treatment for 1 year (P ¼ .016)but not in the group receiving antieTNF-a treatment for 1 year (P ¼ .50).ox-mtDNA, oxidized mtDNA.

Table 2 .
Association of Oxidized mtDNA with Other Covariates in Psoriasis (n [ 89)