Acute exposure to cell-free mitochondrial DNA induces pregnancy-specific aortic endothelial dysfunction and organ-selective inflammation in rats

Pregnancy complications such as preeclampsia are associated with circulating cell-free mitochondrial DNA (mtDNA), a damage-associated molecular pattern capable of activating Toll-like receptor 9 (TLR9). We hypothesized that acute mtDNA exposure induces maternal inflammation and endothelial dysfunction during pregnancy via TLR9 activation. Non-pregnant and pregnant rats (gestational days 14-15) were treated intravenously with saline or purified mtDNA and euthanized 4 h after treatment. mtDNA increased cytokine mRNA expression in lung and liver of non-pregnant and pregnant rats, with magnitude varying by pregnancy status and organ. Aortas from pregnant, but not non-pregnant, rats exhibited reduced acetylcholine (ACh)-induced relaxation following mtDNA treatment (Emax, saline: 90.1 ± 3.9 % vs. mtDNA: 62.1 ± 20.7 % KClmax, p<0.05), while uterine artery function was preserved, indicating vascular bed-specific effects. Ex vivo incubation of aortic rings with mtDNA ± white blood cells did not replicate in vivo findings, implicating systemic rather than direct vascular mechanisms. Nuclear DNA did not affect ACh-induced relaxation (p>0.05), confirming that the vascular effects were mtDNA-specific. Pharmacological antagonism of TLR9 with ODN2088 partially attenuated mtDNA-induced maternal endothelial dysfunction. Although overt vascular ROS increases were not detected, aortas from pregnant rats had reduced sod-1 expression (p<0.05) and increased eNOS protein abundance (p<0.05). Acute mtDNA exposure during pregnancy induces maternal organ inflammation and impairs endothelium-dependent vasodilation, with partial TLR9 involvement. In conclusion, aortic transcriptional changes in antioxidant pathways and 4 increased eNOS abundance were also observed, though their functional significance remains to be determined. 5 New & Noteworthy: To our knowledge, this is the first study to demonstrate that acute exposure to circulating mtDNA induces pregnancy-specific maternal endothelial dysfunction and organ-selective inflammatory responses. Our findings reveal pregnancy- and vascular-bed specific responses of the maternal vasculature to mitochondrial danger signals, with partial TLR9 involvement. Aortic transcriptional changes in antioxidant pathways and increased nitric oxide synthase abundance were identified as molecular correlates of this dysfunction. Keywords: Inflammation, preeclampsia, Toll-like receptor 9, mitochondrial DNA, endothelial dysfunction 6

Pregnancy is a critical period of immune adaptation characterized by tightly regulated co-operative interactions between the maternal, fetal, and placental immune systems. These dynamic immunological and inflammatory shifts support key pregnancy stages such as implantation, placental development, fetal growth, labor and delivery, while preserving the mother’s ability to respond to exogenous pathogens (1, 2). During pregnancy, the maternal immune system undergoes selective tolerance through compartmentalized rather than systemic, immune modulation, which accounts for greater maternal susceptibility to specific pathogens (3). As such, exposure to certain pathogen-associated molecular patterns (PAMPs; e.g., bacterial infections) or damage-associated molecular patterns (DAMPs; sterile inflammation) can elicit heightened maternal immune responses and are associated with adverse maternal and fetal outcomes (1). Maternal inflammation and endothelial dysfunction are common pathophysiological features of pregnancy complications (4), such as preeclampsia, and inflammatory signals can promote or exacerbate endothelial dysfunction. Specifically, maternal endothelial cells can undergo functional and morphological changes upon their interaction with circulating inflammatory molecules, resulting in endothelial dysfunction. Endothelial activation can be triggered by bacterial endotoxins and pro-inflammatory cytokines, as well as by direct activation of pattern recognition receptors (PRRs) expressed on endothelial cells following recognition of PAMPs or DAMPs (reviewed in (5)). In addition, endothelial cells can be activated indirectly through their interaction 7 with immune cells and immune cell-derived mediators, amplifying PRR-driven inflammatory signaling and promoting sustained vascular inflammation (5). Cell-free mitochondrial DNA (mtDNA) is released into the circulation through normal cellular turnover, but under conditions of cellular stress or tissue injury, it can act as a DAMP. Extracellular mitochondria and mitochondrial components can originate from multiple cellular sources and interact with the immune system through diverse mechanisms (6). mtDNA is able to promote immune activation via PRRs such as Toll-like receptor 9 (TLR9) (reviewed in (7, 8)), which recognizes unmethylated CpG motifs enriched in mtDNA. In pregnancy, increased circulating cell-free mtDNA concentrations have been reported in association with inflammatory states and adverse pregnancy outcomes, supporting its potential role as both a biomarker and a mediator of sterile inflammation in gestational complications (7, 9-13). Supporting a causal role for mtDNA as an inflammatory stimulus, Collins et al. demonstrated that intra-articular injection of purified mtDNA in mice promoted immune cell infiltration and increased production of pro-inflammatory mediators (14). Given that cellular responses to CpG DNA can be mediated by TLR9, TLR9 is a plausible innate immune sensor linking mtDNA exposure to downstream inflammatory signaling (8, 15). Consistent with this concept, previous studies have shown a negative impact of TLR9 stimulation on pregnancy outcomes (reviewed in (16)), with TLR9 agonists inducing fetal demise, preterm loss, maternal hypertension (17, 18), disrupted circadian blood pressure regulation (19), and augmented contractile responses in maternal resistance arteries (20). 8 However, whether purified cell-free mtDNA itself, rather than synthetic TLR9 agonists, can causally induce maternal vascular dysfunction and inflammation during pregnancy remains unknown. Therefore, in the current study, we aimed to evaluate the impact of cell-free mtDNA on maternal inflammation and vascular function during pregnancy. We hypothesized that acute exposure to circulating mtDNA induces maternal inflammation and endothelial dysfunction during pregnancy via TLR9 activation. To test this hypothesis, we acutely challenged pregnant and non-pregnant rats with purified mtDNA with and without pharmacological inhibition of TLR9 and assessed vascular function and inflammatory outcomes. We focused on the thoracic aorta as a model conduit vessel, with uterine arteries included as a pregnancy-adapted reference vascular bed to assess vascular bed specificity of responses. We used a single acute mtDNA exposure to capture early innate immune-mediated responses, consistent with the 3-6 hour window commonly used to assess DAMP-induced inflammatory responses in rodent models (19, 21). Materials and Methods Chemicals and Reagents Details of chemicals and reagents used in this study are provided in Supplementary Materials (Table S1). Animals 9 Animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC 22-003) of Loma Linda University, and all procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The data presented herein are part of a larger research project evaluating maternal vascular and fetoplacental responses to circulating mtDNA. Vascular outcomes reported here were derived from cohorts that overlap with animals from which placental tissue was also collected at the same terminal endpoint for assessments reported separately. Male, virgin female, and timed-pregnant Sprague-Dawley rats were purchased from Envigo (Indianapolis, IN). On arrival, males were ~400 g and 13-15-weeks old and virgin females were 190-230 g and 9-11-week-old. Purchased timed-pregnant rats arrived on gestational day (GD)5-8 (term=22-23 days) and were ~220 g and 11-12-week-old. Rats were pair-housed under 12:12-h light-dark cycle (lights on, 07:00 h; lights off, 19:00 h) in a temperature- and humidity-controlled environment and had free access to tap water and standard laboratory rodent chow. Rats were allowed to acclimate to housing conditions at Loma Linda University animal facilities for one week before handling and experimentation. In addition to purchasing timed-pregnant rats (Envigo), we bred rats in-house as previously described (19, 22). Briefly, female rats were pair-mated overnight, and the presence of spermatozoa in vaginal smears the following morning was used to designate GD1. Experiments were performed when rats were 12-24-weeks old. Pregnant rats were studied on GD14-15, with age-matched non-pregnant females used 10 as controls. This gestational age was selected for three reasons. First, earlier exposure to an immune stimulus such as mtDNA could affect placental development and pregnancy viability, which were not primary foci of this study and have been shown to be adversely affected by TLR9 activation at earlier gestational ages (18, 23). Second, GD14-15 represents a gestational period after which fetal growth acceleration is accompanied by substantial maternal cardiovascular adaptations, making it physiologically relevant window for studying maternal vascular responses (24, 25). Third, this gestational age is consistent with our previous work demonstrating that acute exposure to an innate immune stimulus at comparable gestational ages elicits placental inflammation and disruption of maternal blood pressure circadian rhythms, enabling cross-study comparison of maternal responses (19). In total, 135 female and 6 male rats were used for this project. Sample sizes for each experiment were based on prior studies and our published protocols (20, 22, 26-29). Male rats were only used for breeding purposes. Experimental design and timeline Three studies were conducted using separate cohorts. Study 1 evaluated the in vivo effects of mtDNA on organ inflammation and vascular function in pregnant and non-pregnant rats. Study 2 examined whether mtDNA directly affects the vascular function by testing the ex vivo effects of mtDNA on isolated arteries. Study 3 11 investigated the contribution of TLR9 signaling to mtDNA-induced vascular dysfunction through in vivo pharmacological antagonism of TLR9. Purified DNA mtDNA and nuclear DNA (nDNA) were isolated from liver of GD14 pregnant rats and age-matched non-pregnant rats. The liver was selected as the mtDNA source because hepatocytes are among the most mitochondria-rich cell types in mammals, providing high-yield mtDNA preparations, and because liver-derived mitochondrial DAMPs have been used in established rodent models of systemic inflammatory responses analogous to the current experimental design (21, 30). Donor pregnancy status was matched to recipient status to maximize the physiological relevance of the mtDNA preparation to the inflammatory context being modeled. Tissue inflammation was measured in lung and liver using reverse transcription quantitative polymerase chain reaction (RT-qPCR), as these organs represent key sites of innate immune activation in response to circulating DAMPs (21); the liver via sinusoidal endothelial cell- and resident Kupffer cell-mediated TLR9 signaling (31, 32) and the lung via its direct exposure to circulating immune stimuli through the pulmonary circulation. Inflammatory gene expression was also assessed in aortic tissue. Vascular function was assessed in isolated arteries using pin and wire myography. Rats were euthanized and tissues were collected 4 h after treatment to capture early, innate immune system-mediated responses while minimizing confounding effects of chronic systemic changes and compensatory mechanisms. Experiments were conducted only in female rats because this study assessed maternal physiology during pregnancy. 12 Animal treatments In Study 1 and Study 3, female rats received a tail-vein injection of sterile 0.9% saline (vehicle) or mtDNA (300 µg/kg body weight). The mtDNA dose was selected based on prior work showing acute systemic inflammatory response in rats at comparable doses (30). Injection volume was adjusted for body weight and mtDNA preparation concentration to achieve the target dose. In Study 1, a subset of animals received nuclear DNA (nDNA, 300 µg/kg body weight) to assess mtDNA specificity. All injections were performed under isoflurane anesthesia (5% for induction, 3% for maintenance, 100% oxygen). In Study 3, we evaluated the contribution of TLR9 to maternal vascular responses to mtDNA. For this study, female rats received an intravenous injection of either sterile 0.9% saline (Saline group), mtDNA (300 µg/kg body weight; mtDNA group), ODN2088 (TLR9 antagonist, 60 µg/kg body weight) (33) plus 0.9% saline (ODN2088 group), or ODN2088 (60 µg/kg body weight) plus mtDNA (300 µg/kg body weight) (ODN2088+mtDNA group). ODN2088, an inhibitory nucleotide serving as a TLR9 antagonist, was administered 5 min before saline or mtDNA injection. ODN2088 stock solutions (500 mM) were prepared by resuspending lyophilized ODN2088 with endotoxin-free water. The stock solution was then diluted in sterile saline for in vivo treatments. Before and during treatments, rats were maintained warm using a heating pad. Intravenous injections were administered via the tail vein and delivered within 1 min using a 1 mL syringe fitted with a 25G x 5/8’’ needle. All solutions were prepared and 13 administered under sterile conditions. After treatment, rats were returned to clean cages and were monitored during their recovery. Injections were administered between 8:00 – 9:00 am. DNA extraction for in vivo and in vitro treatments Purified mtDNA (in vivo injections and ex vivo artery incubations) and nDNA (in vivo injections) were isolated from donor rat liver as described below. mtDNA and nDNA were prepared from pregnant and/or age-matched non-pregnant donors as indicated for each experiment. Mitochondrial enrichment: Mitochondria were isolated from liver tissue using the Mitochondria Isolation Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. The resulting mitochondrial pellet was used for mtDNA extraction. Nuclear enrichment: Nuclei were isolated from liver tissue as previously described (34). Briefly, liver was gently homogenized and passed through a 40-mm cell strainer (Falcon®, Corning, NY, NY, USA, Cat. No. 352340) into ice-cold buffer A (250 mM sucrose, 5 mM MgCl2, and 10 mM Tris–HCl, pH 7.4). The homogenate was centrifuged at 600 × g for 10 min at 4 ℃, and the pellet was then resuspended in ice-cold buffer B (2.0 M sucrose, 1 mM MgCl2, and 10 mM Tris–HCl, pH 7.4) and centrifuged at 16,000 × g for 30 min at 4 ℃. The resulting nuclear pellet was used for DNA extraction. DNA isolation, quantification, and mtDNA quality control testing: DNA was isolated from mitochondrial and nuclear pellets using the QIAamp DNA Mini Kit (QIAGEN LLC) 14 according to the manufacturer’s instructions. DNA concentration was quantified by spectrophotometry (NanoDrop™ OneC; Thermo Fisher Scientific, Waltham, MA, USA), and DNA was stored at 4 ℃ and used within 1 month after extraction. Purity and integrity of mtDNA preparations were evaluated by spectrophotometry, qPCR-based assessment of mtDNA enrichment relative to nDNA, protein assay, and TapeStation-based sizing. A260/280 ratios were consistently between 1.8-2.1 qPCR indicated a low level of nDNA carryover in mtDNA preparations; therefore, nDNA injections were included as a specificity control. Protein contamination was assessed using a bicinchoninic acid (BCA) assay and was not detectable in mtDNA preparations. Endotoxin was measured using a chromogenic Limulus amebocyte lysate (LAL) assay (Thermo Fisher Scientific) and was below the assay’s limit of detection in a randomly selected subset of preparations. TapeStation analysis demonstrated a predominant high-molecular weight peak (~20 kb), consistent with near full-length mtDNA (~16.3 kb), with limited fragmentation evident as modest peak broadening that varied between samples. Representative quality control data are shown in supplementary materials (Figure S1). Euthanasia, tissue harvest, and processing Rats were anesthetized with isoflurane (5% for induction, 3% for maintenance, 100% oxygen) and euthanized by isoflurane overdose followed by bilateral thoracotomy and removal of their hearts. Whole blood was collected from the inferior vena cava 15 under deep anesthesia into EDTA-coated tubes (BD, Franklin Lakes, NJ; Cat No. 367856) and heparin-containing tubes for plasma isolation. After euthanasia, maternal aorta, uterus, lung, and liver were excised and placed in ice-cold Krebs solution of the following composition (in mM): 130 NaCl, 4.7 KCl, 1.18 KH2PO4, 1.8 MgSO4, 14.9 NaHCO3, 5.6 Dextrose, 1.56 CaCl2·H2O. Aortas and uterine arteries were cleaned from perivascular connective and adipose tissue before being used for vascular reactivity experiments. In addition, aortic segments were either stored in Tissue-Tek O.C.T Compound (Sakura, Torrance, CA, USA) for immunofluorescence staining or snap-frozen for further molecular assessment of protein and RNA levels with Western blotting and RT-qPCR, respectively. Plasma was isolated from whole blood collected in EDTA-coated tubes by centrifugation at 2,000 × g for 15 min at 4 ℃ for subsequent quantification of cell-free mtDNA. Vascular function studies To determine whether mtDNA-induced vascular effects reflect systemic mechanisms or direct actions on the vascular wall, vascular function experiments were performed (1) 4 h after in vivo treatments or (2) following 4 h after ex vivo arterial incubation of arteries with mtDNA or saline. Mounting, normalization, and viability protocols. Isolated thoracic aortas from pregnant and non-pregnant rats were the primary vessels of interest and were used 16 across all vascular studies. Main uterine arteries from pregnant rats were studied as a reference vascular bed to determine whether mtDNA-induced effects were vascular bed specific, given the significant structural and functional remodeling of uterine arteries during pregnancy (35). Accordingly, thoracic aortas and main uterine arteries were used for ex vivo assessment of vascular isometric tension as previously described by Wencelsau et al. with some modifications (36). Briefly, 2-mm aortic rings were mounted in a pin myograph and 2-mm uterine artery rings in a wire myograph, and resting tension was determined (Danish Myo Technology A/S, Aarhus, Denmark). Chambers were filled with 5 mL Krebs-Henseleit solution and continuously gassed with 95% O2, 5% CO2 at 37 ℃. Optimal baseline tension was then determined for each vascular segment using the DMT Normalization Module for LabChart Software (ADInstruments, Colorado Springs, CO) and vessels were equilibrated for 1 h at baseline tension. Vascular viability was confirmed by two contractions to high-K+ (120 mM KCl). Endothelial integrity was assessed in response to a bolus of acetylcholine (ACh, 3 mM) following pre-constriction with a bolus of phenylephrine (PE, 3 mM). Vascular isometric tension measurements after in vivo mtDNA challenge. For experiments with aorta, cumulative concentration-response curves (CCRCs) were generated for PE (0.1 nM - 30 mM), thromboxane A2 receptor agonist U46619 (0.001 nM - 10 mM), ACh (0.1 nM - 30 mM), sodium nitroprusside (SNP, 0.1 nM - 30 mM) and KCl (4.7 - 80 mM). Pre-constriction for ACh and SNP CCRCs were performed with a bolus of PE (1 mM). The force generated during pre-constriction period was similar between groups. 17 In some aortic rings, the endothelium was mechanically removed (“-E”) by gentle luminal rubbing with a pipette tip. Successful denudation was confirmed in vessels with less than 10% relaxation to ACh. To assess the contribution of NO, vessels were incubated with Nω-Nitro-L-arginine methyl ester hydrochloride (L-NAME, 100 mM) for 30 min (20) before PE pre-constriction and throughout the ACh CCRCs. For uterine arteries, we performed CCRCs to ACh, SNP, and KCl. Vascular measurements after ex vivo incubation with mtDNA. Thoracic aortas from untreated pregnant rats (GD14-15) were placed in a culture dish containing DMEM (supplemented with 1% of charcoal-stripped fetal bovine serum (FBS), 100 U/mL penicillin, 100 𝜇g/mL streptomycin, and L-glutamine) and incubated at 37 ℃ in a humidified incubator for 4 h, as previously described (37). For treatments, aortas were incubated with saline or mtDNA at various concentrations (0.02, 2, or 20 ng/µL) isolated from pregnant donor liver mitochondria as described above. mtDNA treatment concentrations were selected based on previously reported circulating mtDNA levels in inflammatory conditions (38-40). To assess the contribution of immune cells to aortic responses to mtDNA, aortic rings were incubated with mtDNA in the presence and absence of autologous white blood cells (WBCs, count: 1 × 106 cells/vessel). Isolation of total WBCs is detailed below. Following incubation, aortas were mounted on a pin myograph. Mounting, normalization, and vascular and endothelial integrity protocols were performed as described above. Vasorelaxation and vasoconstriction were assessed via CCRCs to 18 ACh (0.1 nM - 30 mM; pre-constriction was performed with 60 mM KCl solution) and KCl (4.7 - 80 mM). For ex vivo incubation experiments, 60 mM KCl was used for pre-constriction rather than PE, because prolonged incubation of isolated vascular segments reduces α1-adrenergic receptor-mediated constriction (41). KCl pre-constriction produced comparable contractile responses across all experimental groups. Vascular data analysis and outcomes. CCRCs were analyzed using sigmoidal nonlinear regression (Prism, version 10.0; GraphPad Software Inc., San Diego, CA, USA) and the following parameters were calculated: (i) EC50 (expressed as pEC50: negative logarithm of EC50) and (ii) Emax (maximum response to agonist). The area under the curve (AUC) was calculated from the CCRCs and defined as the cumulative contractile response. Contractile responses are expressed as a percentage of the maximum response to KCl (120 mM). Myography drug dilutions. Stock solutions of PE, ACh, SNP, and L-NAME were prepared in Milli-Q distilled water, whereas U46619 were dissolved in dimethyl sulfoxide (DMSO). Dilutions were prepared fresh on the day of the experiments. Biochemical analyses Plasma cell-free mitochondrial DNA quantification 19 DNA extraction. DNA was extracted from approximately 200 µL of plasma using the DNeasy Blood and Tissue kit (QIAGEN) according to the manufacturer’s instructions, starting at the lysis step. For samples with a starting volume < 200 µL, the entire sample was used, and the pre-binding reagents (proteinase K, AL buffer, and 100% ethanol) were adjusted proportionally before loading onto the column. After DNA binding, standard volumes of the remaining reagents were used. DNA was eluted in 200 µL of DNase-free water and concentrated to 20 µL using the Vacufuge vacuum concentrator (Eppendorf). Mitochondrial DNA quantification via qPCR. Quantification of both nDNA and mtDNA was performed through quantitative PCR (qPCR) using the 7500 Real-Time PCR System (Applied Biosystems™, Waltham, MA, USA). The qPCR was performed following the methodology published in Nicklas et al. (42) with some modifications, including the omission of the mitochondrial deletion target (26). mtDNA was quantified via the mitochondrial D-loop and nDNA was quantified via the β-actin gene. The qPCR master mix was prepared using 13 µL of TaqManTM Universal Master Mix II, no UNG (Applied Biosystems™) per sample along with the primers and probes listed in Table S2. The prepared master mix was added to the wells of a 96-well plate along with either 2 µL of DNA sample or DNase-free water to serve as a negative control. Absolute quantification of the mtDNA target was performed through running a mtDNA standard curve of known mtDNA copy numbers along with each batch of the unknown samples (Sequence: GGTTCTTACTTCAGGGCCATCAATTGGTTCATCGTCCATACGTTCCCCTTAAATAAG 20 ACATCTCGATGGTAACGGGTCTAATC; Cell-free mtDNA is expressed in copies/mL plasma). Circulating nitric oxide metabolite quantification Plasma was isolated from heparin-containing tubes by centrifugation at 9,600 × g for 30 s for subsequent measurements of circulating nitrate and nitrite concentrations (43). Freshly isolated plasma was snap-frozen in liquid nitrogen and stored at −80 ℃ for further analysis of nitric oxide metabolites. In this manuscript, the term nitrite refers to nitric oxide-related species detected by the assay, including nitrite, S-nitrosothiols, iron-nitrosyls, and free NO. Plasma nitrite levels were measured using the tri-iodide (I₃⁻) assay in combination with an ozone-based chemiluminescence NO analyzer (280i, Sievers, Boulder, CO, USA), as previously described (43-45). Nitrate was assessed using the NiR + I₃⁻ assay, which detects nitrate in addition to the targets of the I₃⁻ assay. For this method, samples were incubated with NiR (a mixture of nitrate reductase, FAD, and NADPH) at 37 ℃ for 45 min, followed by analysis using the I₃⁻ assay. Nitrite and nitrate concentrations are expressed as µM. Extraction of RNA from maternal lung, liver, and aortic tissues and cDNA synthesis Total RNA was isolated from maternal lung, liver, and aortic tissue using QIAzol® Lysis Reagent (QIAGEN), followed by purification with the miRNeasy Mini Kit 21 (QIAGEN). RNA purity and quantity were assessed by spectrophotometry (NanoDrop™ OneC). Reverse transcription was performed using Sensiscript RT Kit reagents (QIAGEN) with RiboGuard RNase inhibitor (LGC, Biosearch Technologies) and oligo-dT primers (QIAGEN), as we previously published (19, 46). Reactions were incubated at 37 ℃ for 1 h, and cDNA was stored at -20 ℃ until qRT-PCR experiments. Quantitative real-time polymerase chain reaction Expression of pro-inflammatory cytokines, anti-inflammatory cytokines, immune cell markers, and housekeeping genes in rat aorta, liver and lung tissues was determined using qRT-PCR. Primer sequences are listed in Table S3. qRT-PCR reactions were performed as previously described (19) using iQ SYBR Green Supermix (Bio-Rad) on a CFX96 Real-Time PCR Detection System with CFX Maestro Software v.2.3 (Bio-Rad). Cycling conditions consisted of an initial denaturation at 95 ℃ for 3 min, followed by 40 cycles of 95 ℃ for 10 sec, and 60 ℃ for 1 min. A dissociation melt curve analysis was performed at the end of each run to verify amplification specificity. Target gene expression was normalized to the reference genes 18s (for aortic tissue) or gapdh (for lung and liver tissues), which were selected based on their stable expression across samples. Relative gene expression was calculated using the 2-ΔΔCT method. 22 White blood cell isolation from whole blood White blood cells (WBCs) were isolated and immediately used for ex vivo artery incubations. Briefly, 10 mL aliquots of whole blood were subjected to red blood cell lysis buffer (per 1L: 155 mM NH4Cl, 10 mM KHCO3, 0.2 mL of 0.1 M EDTA diluted in distilled water; pH: 7.2-7.4; sterilized through a 0.2 µm filter) and incubated for 5 min in 37 ℃ water bath. Cells were centrifuged at 400 × g for 5 min, and the pellet was washed and resuspended in 1× RPMI solution containing 10% FBS (Gibco) and 1× antibiotic-antimycotic (Gibco). Cell numbers were determined using trypan blue (Sigma) exclusion, and 1 × 106 cells per aortic segment were used for ex vivo vessel incubation studies. Western blot analysis Western blot analysis was performed in rat aortic tissue, as previously described (26, 47). Proteins were extracted using T-PER Tissue Protein Extraction Reagent (Thermo Fisher Scientific) supplemented with protease inhibitor cocktail tablets (Sigma Aldrich), and concentrations were determined using a BCA assay. Samples were denatured with 𝛽-mercaptoethanol (Sigma Aldrich) and equal amounts of protein were separated by SDS-PAGE (5 μg for SOD-1; 20 μg for SOD-2, Catalase, eNOS). Proteins were transferred to nitrocellulose or polyvinylidene difluoride (PVDF) membranes using a Trans-Blot Turbo Transfer System (Bio-Rad) or by wet transfer, with membranes and transfer method selected according to the protein of interest (Table S5). Membranes 23 were blocked and incubated with primary and secondary antibodies as detailed in Tables S4-S5. Protein signals were detected using either Odyssey CLx imaging system (LI-COR Biosciences, Lincoln, NE, USA) or an AZURE 300 system (Biosystems), with analysis performed either with analysis in Image Studio (v5.2) or ImageJ software (v13.0.6, NIH, USA). Signals were normalized to total protein using Ponceau staining or Revert™ 700 Total Protein Stain (LI-COR Biosciences). Data are expressed relative to the saline control group. Oxidant- and NO-related fluorescence imaging in rat aorta Oxidant-sensitive and NO-reactive fluorescence was assessed in thoracic aorta from saline- and mtDNA-treated rats using dihydroethidium (DHE) and (4-Amino-5-Methylamino-2',7'-Difluorofluorescein Diacetate (DAF-FM) staining, respectively, as previously described (33, 48). Aortic segments were embedded in Tissue-Tek® O.C.T. Compound (Sakura) and snap-frozen in liquid nitrogen. Cross-sections (10 µm) were equilibrated in PBS (3 × 5 min; PBS replaced every 5 min) in a light-protected humidified chamber. For oxidant-sensitive staining, sections were incubated with DHE (10 µmol/L, Sigma-Aldrich) for 30 min. For NO-reactive staining, sections were incubated with DAF-FM Diacetate (20 µmol/L; Invitrogen™, Thermo Fisher Scientific) for 1 h. Negative control sections were processed in parallel using PBS without probe. Slides were imaged on an ECLIPSE Ts2 inverted microscope (Nikon Instruments Inc., 24 NY, USA) using a ×4 objective with TRITC (tetramethylrhodamine isothiocyanate) filter sets for DHE and FITC (fluorescein isothiocyanate) filter sets for DAF-FM. Fluorescence was quantified using ImageJ software (v1.54J). DHE fluorescence was used as an index of oxidant-dependent DHE oxidation within the aortic wall, whereas DAF-FM fluorescence was used as an index of NO-derived nitrosating activity/NO-related signaling following intracellular de-esterification. Statistical analysis Data distributions were assessed using the Shapiro-Wilk test. When data were not normally distributed, log-transformation was applied prior to analysis. Outliers were identified using the ROUT (Robust regression and Outlier) method with Q=1%. Comparisons were performed using unpaired t-tests for normally distributed data with equal variances, Mann-Whitney U tests for non-normally distributed data, one-way ANOVA with Tukey’s post hoc test, or two-way ANOVA with Sidak post hoc test, as appropriate. For RT-qPCR, statistical analyses were performed on ΔCt values. Data are presented as mean ± standard deviation (SD), except concentration-response curves which are presented as mean ± standard error of the mean (SEM). Statistical significance was set at α = 0.05. The specific statistical approach used for each dataset is specified in the corresponding figure legends, and exact p values are reported for all analyses. Analyses were conducted in GraphPad Prism (v10.0.1; GraphPad Software, San Diego, CA, USA). 25

Circulating levels of mtDNA At euthanasia, plasma cell-free mtDNA was quantified 4 h after injection of saline (control) or mtDNA in pregnant and non-pregnant rats. Plasma mtDNA copy number (copies/mL) was comparable across groups (non-pregnant saline: 497.1 ± 429.4, n=7 rats; non-pregnant mtDNA: 672.5 ± 800.3, n=9; pregnant saline: 315.6 ± 93.8, n=8; pregnant mtDNA: 469.1 ± 350.3, n=9). Statistical analyses were performed on log-transformed data using two-way ANOVA and showed no pregnancy-by-treatment interaction (F(1, 29)=0.0246, p=0.876) and no main effects of pregnancy (F(1, 29)=0.2270, p=0.637) or treatment (F(1, 29)=0.00041, p=0.984). These data indicate that circulating cell-free mtDNA levels were not elevated at 4 h after treatment, consistent with clearance/redistribution by this time point. Systemic markers of maternal inflammation To assess systemic maternal inflammatory activation 4 h following mtDNA exposure, we measured pro-inflammatory cytokines il-1β, tnf-α, il-6, and ifn-γ and immunomodulatory/anti-inflammatory cytokines il-10 and il-4 in liver and lung tissues from non-pregnant and pregnant rats. In addition, we measured tlr9 expression in these tissues to determine whether innate immune sensing pathways were engaged. 26 In the liver (Figure 1A-G, left panels), mtDNA exposure elicited a robust inflammatory transcriptional response in both non-pregnant and pregnant rats, with increased expression of il-1β, il-6, tnf-α, ifn-γ, and il-4 (two-way ANOVA with Sidak multiple comparisons test; all p ≤ 0.04). There were significant pregnancy-by-treatment interactions for il-1β (F(1, 23)=9.919, p=0.0045), il-6 (F(1, 23)=6.832, p=0.0156), and il-10 (F(1, 25)=7.8, p=0.01), indicating that the mtDNA effects differ by pregnancy status. Sidak tests demonstrated that the mtDNA-induced increase in il-1β was greater in non-pregnant rats compared with pregnant rats (predicted least squares means diff ± SEM, non-pregnant: 0.598±0.066; pregnant: 0.3001 ± 0.068, p=0.0004), whereas the increase in il-6 was greater in pregnant rats (non-pregnant: 2.06 ± 0.61, p=0.0051; and pregnant: 4.34 ± 0.63, p<0.0001). Liver il-10 increased in non-pregnant but did not change in pregnant rats in response to mtDNA (predicted least squares means difference ± SEM, non-pregnant: 0.14 ± 0.027, p<0.001; pregnant: 0.0250 ± 0.03, p=0.65). In the lung (Figure 1A-G, right panels), mtDNA exposure induced a pregnancy-dependent inflammatory transcriptional response. Two-way ANOVA showed significant (all p≤0.02) pregnancy-by-treatment interactions for il-1β (F(1, 20)=6.173, p=0.0219), tnf-α (F(1, 20)=11.16, p=0.0033), il-6 (F(1, 20)=21.41, p=0.0002), indicating that the pulmonary cytokine response to mtDNA differ by pregnancy status. Sidak post hoc tests revealed that il-1β increased only in non-pregnant rats (p=0.0002), ifn-γ increased only in pregnant rats (p<0.0001), whereas il-6 increased in non-pregnant rats (p=0.0019) and decreased in pregnant rats (p=0.0286). The mtDNA-induced increase in tnf-α was greater in the non-pregnant group compared to the pregnant group (mean diff ± SEM, 27 non-pregnant: 4.579 ± 0.543, p<0.0001; pregnant: 1.659 ± 0.543, p=0.0124). Interestingly, ifn-γ was the only cytokine that was increased only in lungs of pregnant rats (F(1, 20)=26.19, p0.05). mtDNA increased tlr9 mRNA in both liver (Figure 1G, left panel) and lung (Figure 1G, right panel) in non-pregnant (p<0.0001) and pregnant (p=0.03) rats, consistent with engagement of innate immune sensing in response to mtDNA exposure. Two-way ANOVA revealed a significant pregnancy-by-treatment interaction for lung tlr9 (F(1,20)=31.20, p<0.0001), indicating that mtDNA effects on tlr9 expression in this tissue differ by pregnancy status. Post hoc Sidak tests showed that the mtDNA-induced increase in lung tlr9 was greater in non-pregnant compared to pregnant rats (predicted least squares means difference ± SEM, non-pregnant: 4.876 ± 0.4626, p<0.0001; pregnant: 1.222 ± 0.462, p=0.031). This pregnancy-dependent pattern in lung tlr9 expression was consistent with the pregnancy-specific differences observed in pulmonary cytokine response to mtDNA. Together, these data indicate that mtDNA elicited an inflammatory response, as manifested by maternal liver and lung cytokine expression changes, and induced tlr9 transcription, with the magnitude of these responses varying with pregnancy status and differing across maternal organs. Vascular function 28 Given the systemic inflammatory activation and possible TLR9 engagement we observed, and the association between endothelial function and inflammation documented in previous investigations (7, 9-13), we next tested whether mtDNA exposure alters vascular reactivity, with emphasis on endothelial function. In vivo mtDNA challenge – contractile responses. To evaluate changes in aortic contractile function due to mtDNA exposure, we performed CCRCs to PE (Figure 2A-F) and U46619 (Figure 2G-L) in endothelium denuded (-E) or intact (+E) aortic rings in non-pregnant and pregnant rats at 4 h after injection. Two-way ANOVA demonstrated a main effect of denudation for PE pEC50 in both non-pregnant and pregnant rats (all p≤0.05; Figure 2C, F), with no treatment-by-denudation interaction for either constrictor (all p≥0.05). In non-pregnant rats, there was also a main effect of treatment for U46619 (p=0.047; Figure 2I). In Sidak-adjusted post hoc comparisons, denudation increased PE pEC50 in both saline- and mtDNA-treated non-pregnant groups (all p0.05; Figure 2E, F). Aortic contractile responses to non-receptor-mediated depolarization (KCl CCRC) were comparable between saline-treated and mtDNA-treated rats in both the non-pregnant (p=0.92) and pregnant (p=0.98) groups (two-way ANOVA, n=5-7 rats/group; Figure S2). Detailed outcomes from statistical assessment of the CCRC variables are presented in Table S6. Collectively, these data indicate that mtDNA does not alter overall aortic contractile responses but attenuates endothelial modulation of PE-induced constriction in pregnancy. 29 In vivo mtDNA challenge – endothelium-dependent relaxation responses. To determine whether mtDNA affects endothelium-dependent relaxations, we assessed aortic relaxation responses to ACh (Figure 3A-C). Two-way ANOVA showed a main effect of pregnancy (F(1, 21)=15.36, p=0.0008), a main effect of treatment (F(1, 21)=10.57, p=0.004), and a significant pregnancy-by-treatment interaction (F(1, 21)=5.410, p=0.03) for ACh Emax. Sidak adjusted post hoc analysis showed that mtDNA treatment reduced Emax in pregnant (p=0.0017; n=6-8 rats/group) but not in non-pregnant rats (p=0.96; n=5-6 rats/group). There were no differences between non-pregnant and pregnant saline rats in relaxation responses (Emax, p=0.76); however, in the presence of mtDNA, pregnant rats had lower relaxation responses compared to non-pregnant rats (Emax, p=0.0005). A similar pattern was observed for ACh pEC50 (Table S6). These results indicate that mtDNA impairs endothelium-dependent relaxations in the aorta in a pregnancy-specific manner. In vivo nDNA challenge – treatment specificity. Next, we assessed whether the mtDNA-induced vascular effects observed in pregnant rats were specific to mtDNA. In this set of experiments, we measured PE-induced contractile responses and ACh-induced dilatory responses in aortas from pregnant rats 4 h after intravenous injection of nDNA or saline. For these comparisons, the saline control group was derived from the corresponding saline dataset shown in the prior PE and ACh experiments (Figures 2, 3). Two-way ANOVA showed no significant main effects or interaction for PE Emax (p≥0.17; Figure 4A, B). There were significant effects of denudation (F(1,14)=35, 30 p<0.001) and treatment (F(1, 14)=9.1, p=0.009) for PE pEC50, with no denudation-by-treatment interaction (F(1, 14)=0.18, p=0.68) (Figure 4A, C). Sidak-adjusted paired comparisons showed that denuded aortas from both saline and nDNA treated rats had greater pEC50 compared with endothelium intact vessels (all p£0.004). In addition, ACh responses were similar between saline and nDNA groups, with no differences in Emax (p=0.61; Mann-Whitney U test) or pEC50 (p=0.32; unpaired t-test; n=4-6 rats/group; Figure 4D-F). Together, these data show that nDNA does not alter aortic contractile and dilatory functions in pregnant rats, suggesting that the endothelial dysfunction observed in this study was specific to mtDNA. In vivo mtDNA challenge – uterine artery reactivity. To determine whether the effects of mtDNA were vascular bed specific, we also assessed dilatory and contractile responses in uterine arteries from pregnant rats treated with mtDNA or saline. ACh-induced relaxations did not differ between groups (Emax, p=0.29, Mann-Whitney U test; pEC50, p=0.54, unpaired t-test; all n=6-7 rats/group; Figure S3). Similarly, SNP-induced relaxations were comparable between saline and mtDNA groups (Emax, p=0.18, Mann-Whitney U test; pEC50, p=0.19, unpaired t-test; all n=6-7 rats/group; Figure S3). KCl-mediated uterine artery constrictions were also similar between groups (AUC, p=0.14, unpaired t-test; Emax, p=0.23, unpaired t-test; all n=6-7 rats/group; Figure S3). These data indicate that despite a systemic inflammatory response and endothelial impairments in pregnant aortas, uterine artery function was not affected by exposure to mtDNA in pregnant rats. 31 In vivo TLR9 antagonism. To assess the contribution of TLR9 to impaired ACh-mediated relaxations in pregnant rats, we studied a separate cohort, in which pregnant rats received intravenous injections of saline, mtDNA, ODN2088, or ODN2088+mtDNA (Figure 5A-B). We observed a decrease in ACh pEC50 in the mtDNA group compared with saline (one-way ANOVA, F(3,19) = 5.645, p=0.0061; Tukey’s post hoc test, saline vs. mtDNA: p=0.0051). Post hoc multiple comparisons tests demonstrated that ACh pEC50 was comparable between mtDNA and ODN2088+mtDNA groups (p=0.24), and no differences were observed between saline and ODN2088+mtDNA (p=0.15) or between saline and ODN2088 (p=0.62) (5-7 rats/group; Figure 5A-B). These data suggest that pharmacological antagonism of TLR9 partially reduced mtDNA-induced maternal endothelial dysfunction, as ODN2088+mtDNA-treated rats did not differ significantly from saline controls, yet the mtDNA and ODN2088+mtDNA groups also did not differ significantly from each other. Because the magnitude of the mtDNA-induced effect was smaller in this cohort than in Study 1, we examined whether inter-cohort differences in animal characteristics or baseline vascular responses contributed to this difference. We compared maternal body weights, litter size, gestational day distribution, and ACh responses across all four groups using two-way ANOVA and Fisher’s exact test (Table 1). There were no differences in litter size and gestational day distribution within the GD14-15 window between the four groups with no cohort main effect or treatment × cohort interactions (p>0.05). Although maternal body weight differed between cohorts as a main effect (p=0.04), the absence of litter size difference and gestational day distribution difference 32 suggest that this may reflect batch variability rather than a difference in pregnancy state at the time of treatment. Two-way ANOVA revealed significant main effects of both treatment and cohort on ACH Emax (treatment: p=0.003; cohort: p=0.002) and ACh EC50 (treatment: p=0.003; cohort: p=0.02), with no significant treatment × cohort interactions for either outcome (p>0.05). The significant treatment main effects confirm that mtDNA consistently impaired maternal endothelium-dependent relaxation across cohorts, while the cohort main effects reflect differences in baseline vascular reactivity between Study 1 and Study 3 saline-treated groups. Ex vivo mtDNA challenge. To determine whether mtDNA alters vascular function through direct interactions with the vascular wall (as opposed to downstream systemic influences), we incubated ex vivo aortic rings from pregnant rats with mtDNA (0.2, 2, or 20 ng/µL) for 4 h in the presence or absence of WBCs (1 × 106 cells/vessel) prior to vascular function assessment. We observed no differences in ACh-mediated relaxation across groups (all p≥0.40; one-way ANOVA with Tukey's multiple comparisons test; n=4 rats/group; Figure S4). Molecular and biochemical correlates of vascular dysfunction Because mtDNA selectively impaired endothelium-dependent relaxations in aortas from pregnant rats, we next tested whether this phenotype reflected altered NO signaling and NO signaling modulators, including oxidative stress and inflammatory 33 molecules. Key assays are shown in Figure 6 (A-D), with additional NO/redox and inflammatory outcomes provided in Figures S5-S6). First, we sought to investigate if reduced ACh-induced relaxations reflect impaired vascular smooth muscle responsiveness to NO. Dilatory responses to NO donor SNP in endothelium-denuded aortas were comparable between saline and mtDNA groups (Figure 6A), suggesting preserved vascular smooth muscle sensitivity to NO. Second, we performed ACh CCRCs in the presence of a NOS inhibitor L-NAME to assess the effects of mtDNA on the relative contribution of NOS-derived NO to ACh-induced relaxations. NOS inhibition reduced ACh-induced relaxation in both saline and mtDNA-treated pregnant groups (Figure S5A). Accordingly, the NO contribution to ACh CCRC was comparable between groups (NO contribution to ACh CCRC (%), Median (IQR), saline: 105.7 (29.6), mtDNA: 90.70 (23.65), Mann-Whitney U test, p=0.20; Figure 6B), indicating that relative NO contribution to ACh remained unchanged following the treatment with mtDNA. We next tested predefined NO-, redox-, and inflammatory-related mechanisms that commonly modulate endothelial function. Despite reduced ACh-mediated dilation, eNOS protein abundance was increased in aortas from mtDNA-treated rats compared with saline-treated controls (p<0.007, n=5 rats/group, unpaired t-test; Figure 6C), suggesting potential upregulation of eNOS in response to mtDNA exposure. Plasma nitrate and nitrite concentrations (nitrate, p=0.54, Mann-Whitney U test, n=8-9 rats/group; nitrite, p=0.77, unpaired t-test, n=8-9 rats/group; Figure S5B) and DAF-FM mean fluorescence intensity in aortas were comparable between saline and mtDNA groups (DAF-FM: n=5-6 rats/group; p=0.82, unpaired t-test; Figure S5C), 34 suggesting no major systemic reduction in NO bioavailability and no changes in aortic bulk NO production in response to mtDNA. Aortic DHE mean fluorescence intensity was comparable between saline and mtDNA groups (DHE: n=5-7 rats/group; p=0.55, unpaired t-test; Figure 6D). These results indicate an increase in eNOS abundance in aortas from pregnant rats but not downstream measurable changes in NO signaling and vascular smooth muscle sensitivity to NO in response to mtDNA exposure. Aortic sod-1 mRNA expression was decreased in mtDNA-treated pregnant rats relative to saline controls (p=0.0358, n=4-5 rats/group, unpaired t-test), while there were no group differences in sod-2 and catalase mRNA, or SOD-1, and Catalase protein abundance (all p>0.05; Figure 7). SOD-2 protein content increased in aortas from mtDNA-treated pregnant rats (p=0.0556, n=5 rats/group, Mann-Whitney U test). Aortas from mtDNA-treated rats had reduced tnf-α expression compared to saline-treated controls (n=4-5 rats/group; p=0.008, unpaired t-test; Figure S6A), whereas mcp-1 and il-1β mRNA levels were unchanged (n=4-6 rats/group; p≥0.25, unpaired t-test; Figure S6B-C). All immunoblots are presented in supplementary materials (Figures S7-S10). Collectively, these data demonstrate that the mtDNA-induced impairment of endothelium-dependent relaxation is not explained by impaired smooth muscle NO responsiveness, NOS-dependent vasodilation, or overt vascular oxidative/inflammatory activation at this timepoint. Discussion 35 In the present study, we evaluated the inflammatory and vasoactive potential of purified mtDNA and determined how pregnancy modifies these responses. Acute mtDNA exposure induced early innate immune transcriptional activation in the liver and lung of both non-pregnant and pregnant rats, but the magnitude and pattern of cytokine gene upregulation were pregnancy- and organ-specific. Importantly, acute mtDNA exposure induced endothelial dysfunction in the aorta of pregnant but not non-pregnant rats, while uterine artery function was preserved, indicating vascular bed-specific effects. This aortic endothelial dysfunction was partially attenuated by TLR9 antagonism and was not explained by impaired smooth muscle responsiveness to NO, NOS-dependent vasodilation, or overt vascular oxidative and inflammatory activation at this early time point. The inability to replicate the in vivo vascular effects ex vivo suggests that systemic rather than direct vascular mechanisms mediate mtDNA-induced endothelial dysfunction during pregnancy. Extracellular mtDNA can act as a DAMP capable of activating innate immune pathways through multiple sensors, and it has been implicated in inflammatory states, including pregnancy complications (reviewed in (49)). To our knowledge, this is the first study to reveal that the magnitude of inflammatory transcriptional responses to purified mtDNA across lung and liver differ between pregnant and non-pregnant states, being relatively greater in the non-pregnant than pregnant group. These findings suggest that pregnancy shifts the immune system toward a more tolerant phenotype in response to pro-inflammatory stimuli such as mtDNA, in an organ- and pathway-specific manner, 36 with the exception of ifn-γ, a pro-Th1 inflammatory marker, which was upregulated in both liver and lung of pregnant, but not non-pregnant rats. Previous studies suggest that pregnancy modulates the expression of cytokines and associated signaling pathways that contribute to systemic tolerance and a local immunosuppressive state in various organs such as the liver (reviewed in (50)). For instance, Yang et al. showed a downregulation of key inflammatory cytokines such as IFNγ, IL-2, IL-4, IL-6 and IL-10 in the liver of pregnant ewes compared to non-pregnant controls (51). A potential factor that may be involved in the regulation of hepatic responses to pro-inflammatory stimuli is changes in the maternal hormonal profile during pregnancy. These include increased estrogen and progesterone levels (52-54) that could promote regulatory immune pathways and are associated with reduced inflammatory cytokine signaling in the maternal liver. These pregnancy-associated hormonal changes may therefore contribute to the attenuated hepatic transcriptional response to mtDNA observed in pregnant compared to non-pregnant rats in the current study. In contrast to the broadly attenuated hepatic response, the pulmonary transcriptional response to mtDNA during pregnancy showed a more selective pattern of activation. We showed an increased expression of pro-inflammatory tnf-α and ifn-γ in the lung following mtDNA exposure during pregnancy, whereas the expression of several other cytokines was reduced or unchanged. Notably, lung il6 expression was decreased in pregnant rats following mtDNA treatment. This finding was unexpected given the well-established role of IL-6 as a proinflammatory mediator in preeclampsia 37 and other pregnancy complications (55). This reduction may reflect pregnancy-specific regulatory mechanisms that limit IL-6 signaling in the pulmonary compartment, potentially involving IL-6 trans-signaling modulation or the anti-inflammatory actions of pregnancy hormones or pulmonary immune cells (56); however, the precise mechanism warrants further investigation. Pregnancy is associated with substantial structural, functional, and immunological changes in the pulmonary system (57). Vermillion et al. reported that pregnant female mice exhibit higher lung compliance, total lung capacity, and fixed lung volume compared with non-pregnant controls, and that these physiological changes help preserve pulmonary function during influenza infection (58). These findings suggest that pregnancy-induced changes in lung mechanics create a physiologically distinct pulmonary environment that modifies the organ’s response to immune challenges. By extension, similar pregnancy-induced adaptations in pulmonary structure and immune tone may determine the lung transcriptional response to sterile immune stimuli such as circulating mitDNA, contributing to selective cytokine pattern observed in our study. In addition, pregnancy alters susceptibility and severity of certain respiratory infections. Pneumonia may be more severe during pregnancy due in part to circulatory changes and reductions in functional residual lung capacity caused by increased abdominal pressure (3, 59), further underscoring that the pregnant lung operates under a different physiological baseline. The selective upregulation of tnf-α and ifn-γ in the lungs of pregnant rats followed mtDNA exposure may reflect a shift toward innate 38 antiviral-type immune surveillance in the pulmonary compartment during pregnancy, consistent with the known role of these cytokines in respiratory host defense. A key and novel finding of this study was the pregnancy-specific impairment of endothelium-dependent relaxation in aortas from pregnant but not non-pregnant rats. The contribution of the endothelium to PE-induced vasoconstriction was also impaired during pregnancy following mtDNA exposure, while overall aortic contractile capacity and smooth muscle sensitivity to NO were preserved. These findings demonstrate that acute systemic mtDNA exposure is sufficient to induce aortic endothelial dysfunction in a pregnancy-specific manner. Because endothelial function is regulated by interactions between NO signaling, redox balance, and inflammatory pathways, we examined components of these axes based on our previous work (20). Although we did not detect overt increases in vascular ROS production or changes in NO bioavailability at this early time point, we observed transcriptional changes in antioxidant pathways, including reduced sod-1 expression and a trend toward increased sod-2, alongside increased aortic eNOS protein abundance. Key inflammatory cytokines did not increase in the aorta in response to mtDNA. Taken together, these molecular findings are correlates of the observed endothelial dysfunction whose functional significance remains to be determined. The absence of overt oxidative stress or NO depletion at 4 hours suggests that the endothelial impairment may precede the development of detectable redox or inflammatory changes, consistent with an early and dynamic perturbation of endothelial homeostasis rather than established oxidative or inflammatory injury. These findings 39 contrast with our prior study, in which repeated stimulation of TLR9 with a synthetic agonist induced vascular oxidative stress and impaired NOS-dependent mechanisms in mesenteric arteries (20), suggesting that the temporal and mechanistic profile of vascular injury differs between acute purified mtDNA exposure and chronic synthetic TLR9 stimulation. Circulating cell-free mtDNA levels were not elevated at 4 hours post-injection, consistent with the known rapid clearance kinetics of circulating DNA (60) (61) and possible cellular uptake by immune or peripheral cells. These observations suggest that even transient exposure to circulating mtDNA may be sufficient to initiate inflammatory and vascular responses that persist beyond the period of detectable plasma increases. The mtDNA-induced upregulation of tlr9 in the lung and liver of both pregnant and non-pregnant states is consistent with engagement of CpG-sensing pathways, and pharmacological antagonism of TLR9 partially reduced the endothelial dysfunction, supporting a role for TLR9 in mediating the vascular effects of mtDNA during pregnancy. The incomplete reversal by TLR9 antagonism suggests that additional DNA-sensing pathways may contribute to the observed vascular phenotype, including cGAS-STING or NLRP3 inflammasome (62-64). Downstream TLR9 signaling mediators such as Myeloid Differentiation Primary Response 88 (MyD88) and regulatory factor 7 (IRF7) were not assessed in the current study and warrant further investigation. The smaller magnitude of mtDNA-induced endothelial dysfunction in Study 3 compared with Study 1 was addressed by formal between-cohort comparison of animal characteristics and baseline vascular reactivity (Table 1). Two-way ANOVA revealed 40 consistent treatment main effects on ACh Emax and pEC50 across both cohorts with no treatment × cohort interactions, demonstrating that mtDNA impaired endothelium-dependent relaxation similarly across cohorts. The cohort main effects for ACh Emax and pEC50 reflected differences in baseline vascular reactivity between cohorts rather than differential sensitivity to mtDNA. Litter size, gestational day distribution, and maternal weight did not interact with the effect of treatment. These findings confirm the reproducibility of the mtDNA-induced endothelial impairment and support the validity of the TLR9 antagonism data. To determine whether mtDNA acts directly on the vascular wall, we incubated aortic rings ex vivo with mtDNA with or without WBCs for 4 hours; however, vascular function remained intact under these conditions. These results suggest that the in vivo endothelial dysfunction requires systemic mediators, such as circulating immune mediators, neurohumoral signals, pr hemodynamic stimuli, rather than direct vascular actions of mtDNA. Finally, the absence of functional changes in uterine arteries despite systemic endothelial dysfunction in the aorta demonstrates that mtDNA-mediated vascular effects are vascular bed-specific. This pattern is consistent with our previous study showing that TLR9 stimulation did not alter uterine artery function, while impairing mesenteric resistance artery responses during pregnancy (20). The uterine vasculature undergoes substantial structural and functional remodeling during pregnancy to meet fetoplacental developmental demands (35, 65), and this remodeling may confer resilience to inflammatory stimuli, though the specific mechanisms protecting uterine artery function 41 from mtDNA-induced impairments were not directly examined in this study. Collectively, these findings support the concept that pregnancy differentially programs vascular beds, resulting in selective susceptibility of systemic vessels while preserving uterine perfusion. Strengths and Limitations A major strength of the present study is the use of a controlled in vivo model with rigorous mtDNA quality control to directly assess the effects of circulating mtDNA on maternal vascular function and inflammation during pregnancy. The inclusion of nuclear DNA as a specificity control, multi-organ inflammatory assessment across liver and lung, vascular bed comparisons between aorta and uterine arteries, and pharmacological TLR9 antagonism collectively provide a mechanistically informative dataset linking a physiologically relevant DAMP stimulus to maternal endothelial dysfunction. Several limitations should be considered. First, our experimental design involved a single systemic injection of mtDNA at one gestational time point within mid-gestation, which does not recapitulate the chronic inflammatory environment that characterizes pregnancy complications such as preeclampsia. Second, estrous cycle was not controlled in non-pregnant rats, which may have introduced variability in vascular and inflammatory outcomes. Third, donor pregnancy status was matched to recipient pregnancy status for mtDNA preparation, which introduces the possibility that biological 42 differences between pregnant- and non-pregnant-derived mtDNA contributed in part to the observed pregnancy-specific effects. Fourth, the ex vivo negative results implicate systemic mechanisms in the in vivo vascular phenotype, the specific circulating mediators or processes responsible were not identified. Fifth, only cytosolic ROS were assessed. Mitochondrial ROS, which may independently influence vascular function, were not directly measured. Finally, our study used exogenous purified mtDNA rather than endogenously released mtDNA, and the extent to which these findings reflect the pathophysiology of conditions involving endogenous sterile DAMP release warrants further investigation. Conclusions In summary, our findings demonstrate that acute exposure to circulating mtDNA during pregnancy induces systemic maternal inflammation and impairs maternal endothelium-dependent vasodilation in a pregnancy- and vascular bed-specific manner. These vascular changes are associated with transcriptional changes in antioxidant pathways and increased eNOS protein abundance in the aorta, with partial involvement of the TLR9 pathway. Because pregnancy complications such as preeclampsia are characterized by heightened inflammatory stress, these findings provide novel evidence that non-bacterial inflammatory stimuli can adversely affect maternal vascular function during pregnancy. Improved understanding of how mitochondrial danger signals influence maternal vascular physiology may help inform strategies aimed at protecting maternal cardiovascular health during complicated pregnancies. 43 Supplemental material Supplemental Tables: Table S1, Table S2, Table S3, Table S4, Table S5, Table S6 Supplemental Figures: Figure S1, Figure S2, Figure S3, Figure S4, Figure S5, Figure S6 Acknowledgments: We thank Loma Linda University Core Facility for providing access to cryostat, Western blot imaging, and qPCR instrumentation. Grants: This study was supported by NIH R01 HL146562 (SG), AHA 24POST1198395 (NH), AHA 24POST1198627 (RNOS), and the Dean’s Stipend Award from the School of Medicine, Loma Linda University (DE). Disclosures: No conflicts of interest, financial or otherwise, are declared by the author(s). Disclaimers: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. 44 Author Contributions: NH and SG conceived and designed research; NH, RNOS and SG analyzed data; NH, SG, RNOS, LL, TL, IG, GK, MR performed experiments; NH, RNOS, TL, EMG, AB, NRP, XQH, LZ and SG interpreted results of experiments; NH and SG prepared figures; NH and SG drafted manuscript; NH, RNOS, DE, LL, TL, EMG, AB, XQH, LZ and SG edited and revised manuscript; NH, RNOS, DE, LL, TL, IG, GK, MR, EMG, AB, NRP, XQH, LZ and SG approved final version of the manuscript. 45

1. Kumar M, Saadaoui M, Al Khodor S. Infections and Pregnancy: Effects on Maternal and Child Health. Front Cell Infect Microbiol. 2022;12:873253. 2. Mor G, Cardenas I. The immune system in pregnancy: a unique complexity. Am J Reprod Immunol. 2010;63(6):425-33. 3. Abu-Raya B, Michalski C, Sadarangani M, Lavoie PM. Maternal Immunological Adaptation During Normal Pregnancy. Front Immunol. 2020;11:575197. 4. Echeverria C, Eltit F, Santibanez JF, Gatica S, Cabello-Verrugio C, Simon F. Endothelial dysfunction in pregnancy metabolic disorders. Biochim Biophys Acta Mol Basis Dis. 2020;1866(2):165414. 5. Theofilis P, Sagris M, Oikonomou E, Antonopoulos AS, Siasos G, Tsioufis C, et al. Inflammatory Mechanisms Contributing to Endothelial Dysfunction. Biomedicines. 2021;9(7). 6. Tupin F, Gonzalez-Chapa JA, Chung JH, Lood C, Boilard E. Clotting the Gap Between Mitochondria-Mediated Immunity and Mitochondrial Transfer. Circ Res. 2026;138(8):e326987. 7. Bradshaw JL, Cushen SC, Phillips NR, Goulopoulou S. Circulating Cell-Free Mitochondrial DNA in Pregnancy. Physiology (Bethesda). 2022;37(4):0. 8. Hemmi H, Takeuchi O, Kawai T, Kaisho T, Sato S, Sanjo H, et al. A Toll-like receptor recognizes bacterial DNA. Nature. 2000;408(6813):740-5. 9. Hu XQ, Zhang L. Mitochondrial Dysfunction in the Pathogenesis of Preeclampsia. Curr Hypertens Rep. 2022;24(6):157-72. 46 10. Holland OJ, Cuffe JSM, Dekker Nitert M, Callaway L, Kwan Cheung KA, Radenkovic F, et al. Placental mitochondrial adaptations in preeclampsia associated with progression to term delivery. Cell Death Dis. 2018;9(12):1150. 11. Vishnyakova PA, Volodina MA, Tarasova NV, Marey MV, Tsvirkun DV, Vavina OV, et al. Mitochondrial role in adaptive response to stress conditions in preeclampsia. Sci Rep. 2016;6:32410. 12. Cushen SC, Ricci CA, Bradshaw JL, Silzer T, Blessing A, Sun J, et al. Reduced Maternal Circulating Cell-Free Mitochondrial DNA Is Associated With the Development of Preeclampsia. J Am Heart Assoc. 2022;11(2):e021726. 13. Mando C, De Palma C, Stampalija T, Anelli GM, Figus M, Novielli C, et al. Placental mitochondrial content and function in intrauterine growth restriction and preeclampsia. Am J Physiol Endocrinol Metab. 2014;306(4):E404-13. 14. Collins LV, Hajizadeh S, Holme E, Jonsson IM, Tarkowski A. Endogenously oxidized mitochondrial DNA induces in vivo and in vitro inflammatory responses. J Leukoc Biol. 2004;75(6):995-1000. 15. Kumagai Y, Takeuchi O, Akira S. TLR9 as a key receptor for the recognition of DNA. Adv Drug Deliv Rev. 2008;60(7):795-804. 16. Goulopoulou S, Matsumoto T, Bomfim GF, Webb RC. Toll-like receptor 9 activation: a novel mechanism linking placenta-derived mitochondrial DNA and vascular dysfunction in pre-eclampsia. Clin Sci (Lond). 2012;123(7):429-35. 17. Thaxton JE, Romero R, Sharma S. TLR9 activation coupled to IL-10 deficiency induces adverse pregnancy outcomes. J Immunol. 2009;183(2):1144-54. 47 18. He B, Yang X, Li Y, Huang D, Xu X, Yang W, et al. TLR9 (Toll-Like Receptor 9) Agonist Suppresses Angiogenesis by Differentially Regulating VEGFA (Vascular Endothelial Growth Factor A) and sFLT1 (Soluble Vascular Endothelial Growth Factor Receptor 1) in Preeclampsia. Hypertension. 2018;71(4):671-80. 19. Bradshaw JL, Cushen SC, Ricci CA, Tucker SM, Gardner JJ, Little JT, et al. Exposure to unmethylated CpG oligonucleotides disrupts blood pressure circadian rhythms and placental clock gene network in pregnant rats. Am J Physiol Heart Circ Physiol. 2023;325(2):H323-H37. 20. Goulopoulou S, Wenceslau CF, McCarthy CG, Matsumoto T, Webb RC. Exposure to stimulatory CpG oligonucleotides during gestation induces maternal hypertension and excess vasoconstriction in pregnant rats. Am J Physiol Heart Circ Physiol. 2016;310(8):H1015-25. 21. Zhang Q, Raoof M, Chen Y, Sumi Y, Sursal T, Junger W, et al. Circulating mitochondrial DAMPs cause inflammatory responses to injury. Nature. 2010;464(7285):104-7. 22. Osikoya O, Cushen SC, Ricci CA, Goulopoulou S. Cyclooxygenase-dependent mechanisms mediate in part the anti-dilatory effects of perivascular adipose tissue in uterine arteries from pregnant rats. Pharmacol Res. 2021;171:105788. 23. Scharfe-Nugent A, Corr SC, Carpenter SB, Keogh L, Doyle B, Martin C, et al. TLR9 provokes inflammation in response to fetal DNA: mechanism for fetal loss in preterm birth and preeclampsia. J Immunol. 2012;188(11):5706-12. 48 24. Slangen BF, Out IC, Verkeste CM, Peeters LL. Hemodynamic changes in early pregnancy in chronically instrumented, conscious rats. Am J Physiol. 1996;270(5 Pt 2):H1779-84. 25. Spradley FT, Ge Y, Granger JP, Chade AR. Utero-placental vascular remodeling during late gestation in Sprague-Dawley rats. Pregnancy Hypertens. 2020;20:36-43. 26. Gardner JJ, Oliveira da Silva RN, Bradshaw JL, Mabry S, Wilson EN, Hula N, et al. Gestational chronic intermittent hypoxia triggers maternal inflammation and disrupts placental stress responses. Am J Physiol Cell Physiol. 2025;329(2):C630-C45. 27. Bradshaw JL, Wilson EN, Gardner JJ, Mabry S, Tucker SM, Rybalchenko N, et al. Pregnancy-induced oxidative stress and inflammation are not associated with impaired maternal neuronal activity or memory function. Am J Physiol Regul Integr Comp Physiol. 2024;327(1):R35-R45. 28. Goulopoulou S, Hannan JL, Matsumoto T, Webb RC. Pregnancy reduces RhoA/Rho kinase and protein kinase C signaling pathways downstream of thromboxane receptor activation in the rat uterine artery. Am J Physiol Heart Circ Physiol. 2012;302(12):H2477-88. 29. Osikoya O, Cushen SC, Gardner JJ, Raetz MM, Nagarajan B, Raut S, et al. Exosomes facilitate intercellular communication between uterine perivascular adipose tissue and vascular smooth muscle cells in pregnant rats. Am J Physiol Heart Circ Physiol. 2022;323(3):H577-H84. 49 30. He J, Lu Y, Xia H, Liang Y, Wang X, Bao W, et al. Circulating Mitochondrial DAMPs Are Not Effective Inducers of Proteinuria and Kidney Injury in Rodents. PLoS One. 2015;10(4):e0124469. 31. Gao Y, Song B, Aoki S, Ito K. Role of Kupffer cells in liver injury induced by CpG oligodeoxynucleotide and flucloxacillin in mice. EXCLI J. 2020;19:387-99. 32. Martin-Armas M, Simon-Santamaria J, Pettersen I, Moens U, Smedsrod B, Sveinbjornsson B. Toll-like receptor 9 (TLR9) is present in murine liver sinusoidal endothelial cells (LSECs) and mediates the effect of CpG-oligonucleotides. J Hepatol. 2006;44(5):939-46. 33. McCarthy CG, Wenceslau CF, Goulopoulou S, Ogbi S, Baban B, Sullivan JC, et al. Circulating mitochondrial DNA and Toll-like receptor 9 are associated with vascular dysfunction in spontaneously hypertensive rats. Cardiovasc Res. 2015;107(1):119-30. 34. Nagata T, Redman RS, Lakshman R. Isolation of intact nuclei of high purity from mouse liver. Anal Biochem. 2010;398(2):178-84. 35. Allerkamp HH, Pole T, Boukham A, James JL, Clark AR. Pregnancy-specific uterine vascular reactivity: a data-driven computational model of shear-dependent, myogenic, and mechanical radial artery features. Am J Physiol Heart Circ Physiol. 2022;323(1):H72-H88. 36. Wenceslau CF, McCarthy CG, Earley S, England SK, Filosa JA, Goulopoulou S, et al. Guidelines for the measurement of vascular function and structure in isolated arteries and veins. Am J Physiol Heart Circ Physiol. 2021;321(1):H77-H111. 50 37. Hu XQ, Dasgupta C, Song R, Romero M, Wilson SM, Zhang L. MicroRNA-210 Mediates Hypoxia-Induced Repression of Spontaneous Transient Outward Currents in Sheep Uterine Arteries During Gestation. Hypertension. 2021;77(4):1412-27. 38. Wang L, Xie L, Zhang Q, Cai X, Tang Y, Wang L, et al. Plasma nuclear and mitochondrial DNA levels in acute myocardial infarction patients. Coron Artery Dis. 2015;26(4):296-300. 39. Valdes-Aguayo JJ, Garza-Veloz I, Vargas-Rodriguez JR, Martinez-Vazquez MC, Avila-Carrasco L, Bernal-Silva S, et al. Peripheral Blood Mitochondrial DNA Levels Were Modulated by SARS-CoV-2 Infection Severity and Its Lessening Was Associated With Mortality Among Hospitalized Patients With COVID-19. Front Cell Infect Microbiol. 2021;11:754708. 40. Kung C-T, Hsiao S-Y, Tsai T-C, Su C-M, Chang W-N, Huang C-R, et al. Plasma nuclear and mitochondrial DNA levels as predictors of outcome in severe sepsis patients in the emergency room. Journal of Translational Medicine. 2012;10(1):130. 41. Fan LL, Ren S, Zhou H, Wang Y, Xu PX, He JQ, et al. Alpha(1D)-adrenergic receptor insensitivity is associated with alterations in its expression and distribution in cultured vascular myocytes. Acta Pharmacol Sin. 2009;30(12):1585-93. 42. Nicklas JA, Brooks EM, Hunter TC, Single R, Branda RF. Development of a quantitative PCR (TaqMan) assay for relative mitochondrial DNA copy number and the common mitochondrial DNA deletion in the rat. Environ Mol Mutagen. 2004;44(4):313-20. 51 43. Liu T, Zhang M, Duot A, Mukosera G, Schroeder H, Power GG, et al. Artifacts Introduced by Sample Handling in Chemiluminescence Assays of Nitric Oxide Metabolites. Antioxidants (Basel). 2023;12(9). 44. Mukosera GT, Liu T, Ishtiaq Ahmed AS, Li Q, Sheng MH, Tipple TE, et al. Detection of dinitrosyl iron complexes by ozone-based chemiluminescence. Nitric Oxide. 2018;79:57-67. 45. Feelisch M, Rassaf T, Mnaimneh S, Singh N, Bryan NS, Jourd'Heuil D, et al. Concomitant S-, N-, and heme-nitros(yl)ation in biological tissues and fluids: implications for the fate of NO in vivo. FASEB J. 2002;16(13):1775-85. 46. Osikoya O, Ahmed H, Panahi S, Bourque SL, Goulopoulou S. Uterine perivascular adipose tissue is a novel mediator of uterine artery blood flow and reactivity in rat pregnancy. J Physiol. 2019;597(15):3833-52. 47. Gardner JJ, Cushen SC, Oliveira da Silva RN, Bradshaw JL, Hula N, Gorham IK, et al. Oxidative stress induces release of mitochondrial DNA into the extracellular space in human placental villous trophoblast BeWo cells. Am J Physiol Cell Physiol. 2024;326(6):C1776-C88. 48. Wenceslau CF, Rossoni LV. Rostafuroxin ameliorates endothelial dysfunction and oxidative stress in resistance arteries from deoxycorticosterone acetate-salt hypertensive rats: the role of Na+K+-ATPase/ cSRC pathway. J Hypertens. 2014;32(3):542-54. 49. West AP, Shadel GS. Mitochondrial DNA in innate immune responses and inflammatory pathology. Nat Rev Immunol. 2017;17(6):363-75. 52 50. Yang L, Meng Y, Shi Y, Fang H, Zhang L. Maternal hepatic immunology during pregnancy. Front Immunol. 2023;14:1220323. 51. Yang L, Bai J, Zhao Z, Li N, Wang Y, Zhang L. Differential expression of T helper cytokines in the liver during early pregnancy in sheep. Anim Reprod. 2019;16(2):332-9. 52. Xiong YH, Yuan Z, He L. Effects of estrogen on CD4(+) CD25(+) regulatory T cell in peripheral blood during pregnancy. Asian Pac J Trop Med. 2013;6(9):748-52. 53. Breous E, Somanathan S, Vandenberghe LH, Wilson JM. Hepatic regulatory T cells and Kupffer cells are crucial mediators of systemic T cell tolerance to antigens targeting murine liver. Hepatology. 2009;50(2):612-21. 54. Raghupathy R, Szekeres-Bartho J. Progesterone: A Unique Hormone with Immunomodulatory Roles in Pregnancy. Int J Mol Sci. 2022;23(3). 55. Prins JR, Gomez-Lopez N, Robertson SA. Interleukin-6 in pregnancy and gestational disorders. J Reprod Immunol. 2012;95(1-2):1-14. 56. Rose-John S, Jenkins BJ, Garbers C, Moll JM, Scheller J. Targeting IL-6 trans-signalling: past, present and future prospects. Nat Rev Immunol. 2023;23(10):666-81. 57. Hill CC, Pickinpaugh J. Physiologic changes in pregnancy. Surg Clin North Am. 2008;88(2):391-401, vii. 58. Vermillion MS, Nelson A, vom Steeg L, Loube J, Mitzner W, Klein SL. Pregnancy preserves pulmonary function following influenza virus infection in C57BL/6 mice. American Journal of Physiology-Lung Cellular and Molecular Physiology. 2018;315(4):L517-L25. 53 59. Sheffield JS, Cunningham FG. Community-acquired pneumonia in pregnancy. Obstet Gynecol. 2009;114(4):915-22. 60. Trumpff C, Michelson J, Lagranha CJ, Taleon V, Karan KR, Sturm G, et al. Stress and circulating cell-free mitochondrial DNA: A systematic review of human studies, physiological considerations, and technical recommendations. Mitochondrion. 2021;59:225-45. 61. Lo YM, Zhang J, Leung TN, Lau TK, Chang AM, Hjelm NM. Rapid clearance of fetal DNA from maternal plasma. Am J Hum Genet. 1999;64(1):218-24. 62. Chen KQ, Tang WR, Liu X. Research and progress of cGAS/STING/NLRP3 signaling pathway: a mini review. Front Immunol. 2025;16:1594133. 63. Shirasuna K, Karasawa T, Takahashi M. Role of the NLRP3 Inflammasome in Preeclampsia. Front Endocrinol (Lausanne). 2020;11:80. 64. Sun J, Yu M, Du W, Zhu S, Chen Z, Tao J, et al. The cGAS-STING pathway promotes the development of preeclampsia by upregulating autophagy: Mechanisms and implications. Int Immunopharmacol. 2024;128:111531. 65. James JL, Chamley LW, Clark AR. Feeding Your Baby In Utero: How the Uteroplacental Circulation Impacts Pregnancy. Physiology (Bethesda). 2017;32(3):234-45. 54 FIGURE LEGENDS Figure 1. Liver and lung cytokine mRNA expression 4 hours after mitochondrial DNA (mtDNA) treatment. Relative expression (2-∆∆Ct) of (A) interleukin (il) 1β (il-1β), (B) tumor necrosis factor α (tnf-α), (C) il-6, (D) interferon γ (ifn-γ), (E) il-10, (F) il-4, and (G) toll-like receptor 9 (tlr9) in liver (left panels) and lung (right panels) from non-pregnant (NP) and pregnant (Preg) rats 4 h after intravenous injection of saline (control) or purified mtDNA. Statistical tests were performed on ΔCt values. Data were analyzed by two-way ANOVA with Sidak post hoc test; individual data points are shown, with bars representing mean ± SD; n=6 rats/group. Figure 2. Aortic contractile responses in non-pregnant and pregnant rats after in vivo saline or mitochondrial DNA (mtDNA) challenge. Cumulative concentration-response curves (CCRCs) and corresponding pEC50 values for phenylephrine (PE; A-F) and U46619 (G-L) were assessed in thoracic aortic rings from non-pregnant (A-C, G-I) and pregnant rats (D-F, J-L) 4 h after intravenous injection of saline (black symbols) or mtDNA (red symbols). Endothelium-denuded aorta assays are shown in open symbols, saline-treated in black symbols, and mtDNA-treated in red symbols. pEC50 is the negative logarithm of the EC50, and area under the curve (AUC) reflects total contractile activity across the concentration range. Data were analyzed by two-way ANOVA with Sidak post hoc test. CCRCs are presented as mean ± SEM. pEC50 values are shown as individual data points with bars representing mean ± SD; n=6-9 rats/group. 55 Figure 3. Aortic endothelium-dependent relaxation responses in non-pregnant and pregnant rats after in vivo saline or mitochondrial DNA (mtDNA) challenge. Cumulative concentration-response curves (CCRCs) and corresponding maximum responses (Emax) for acetylcholine (ACh) were assessed in thoracic aortic rings from in non-pregnant (A, C) and pregnant (B, C) rats 4 h after intravenous injection of saline (black symbols) or mtDNA (red symbols). Data were analyzed by two-way ANOVA with Sidak post hoc test. CCRCs are presented as mean ± SEM. Emax values are shown as individual data points with bars representing mean ± SD; n=5-8 rats/group. Figure 4. Aortic contractile and relaxation responses after in vivo saline or nuclear DNA (nDNA) challenge. Cumulative concentration-response curves (CCRCs) and corresponding maximum responses (Emax) and pEC50 values to phenylephrine (PE; A-C) and acetylcholine (ACh; D-F) were assessed in thoracic aortic rings from pregnant rats 4 h after intravenous injection of saline (black symbols) or nDNA (red symbols). Open symbols show the assays in endothelium-denuded (-E) aorta rings and closed symbols in intact (+E) aorta rings. Emax (B, E) represents the maximum contractile or relaxation response and pEC50 (C, F) is the negative logarithm of the EC50. Data were analyzed by two-way ANOVA with Sidak post hoc test. CCRCs are presented as mean ± SEM. pEC50 and Emax values are shown as individual data points with bars representing mean ± SD; n=4-5 rats/group. For these comparisons, the saline control group was derived from the corresponding saline dataset presented in Figures 2, 3. 56 Figure 5. Aortic endothelium-dependent relaxation responses in pregnant rats following in vivo mitochondrial DNA (mtDNA) challenge with or without toll-like receptor 9 (TLR9) antagonism. Cumulative concentration-response curves to acetylcholine (ACh) and corresponding pEC50 values were assessed in thoracic aortic rings from pregnant rats 4 h after intravenous injection of saline (black symbols) or mtDNA (red symbols) in the absence (closed symbols) or presence (open symbols) of the TLR9 antagonist ODN2088. ODN2088 was administered intravenously 5 min before mtDNA or saline treatment. pEC50 is the negative logarithm of the EC50. Data were analyzed by one-way ANOVA with Tukey’s post hoc test. CCRCs are presented as mean ± SEM. pEC50 values are shown as individual data points with bars representing mean ± SD; n=4-7 rats/group. Figure 6. Molecular and biochemical correlates of mitochondrial DNA (mtDNA)-induced endothelial dysfunction in aortas from pregnant rats. (A) Relaxation responses to the NO donor sodium nitroprusside (SNP) were assessed in endothelium-denuded thoracic aortic rings. (B) Relative NO contribution to ACh-mediated relaxation was calculated from acetylcholine (ACh) cumulative concentration-response curves ± the NOS inhibitor L-NAME and expressed as percent NO contribution. (C) Aortic endothelial NO synthase protein content. Data were analyzed using Mann–Whitney U tests (B) or unpaired t-tests (C). SNP concentration-response curves are presented as 57 mean ± SEM. The remaining data are shown as individual data points with bars representing mean ± SD. Representative image for eNOS is shown in C. Figure 7. Antioxidant gene and protein expression in aortas from pregnant rats following in vivo saline or mitochondrial DNA (mtDNA) challenge. Thoracic aortas were collected from pregnant rats 4 h after intravenous injection of saline (control) or mtDNA. (A-C) Aortic mRNA expression of sod-1 (A; n=4-5 rats/group), sod-2 (B; n=6 rats/group), and catalase (C; n=4-5 rats/group). (D-F) Aortic protein content of SOD-1 (D), SOD-2 (E), and Catalase (F), with representative immunoblots shown for each protein (all n=5 rats/group). For mRNA analyses, statistical testing was performed on ΔCt values. Data were analyzed using unpaired t-tests (A, D) or Mann–Whitney U tests (B, C, E, F). Data are shown as individual data points with bars representing mean ± SD. 58 Table 1. Maternal characteristics and baseline vascular reactivity across treatment groups in Study 1 (in vivo mtDNA challenge) and Study 3 (TLR9 antagonism) cohorts. Parameter Study 1 – Saline (n=6) Study 1 – mtDNA (n=8) Study 3 – Saline (n=4) Study 3 – mtDNA (n=7) Treatment effect Cohort effect Interaction Maternal body weight (g) 285.7 ± 41.8 287.1 ± 47.5 244.1 ± 19.4 260.4 ± 26.1 p=0.57 p=0.04 p=0.63 Litter size 13.0 ± 3.3 12.8 ± 4.8 12.3 ± 1.0 11.4 ± 4.7 p=0.87 p=0.55 p=0.13 ACh Emax 90.08 ± 3.9 62.09 ± 20.7 97.8 ± 0.95 91.01 ± 5.1 p=0.003 p=0.002 0.052 ACh EC50 7.2 ± 0.29 5.9 ± 1.1 7.6 ± 0.32 6.88 ± 0.33 p=0.002 p=0.02 p=0.37 GD14/15, n 3/3 5/3 1/3 3/4 Fisher’s exact p=0.75 mtDNA, mitochondrial DNA; TLR9, Toll-like receptor 59 Figure 1 Liver Lungs NPPreg01234 501001502-ΔΔCt mtDNA pPregn x Trtm: 0.005pPregn: <0.0001pTrtm: <0.0001 Saline il-1β p=0.0004p<0.0001 NPPreg0 4 8 2030402-ΔΔCt mtDNA pPregn x Trtm: 0.02pPregn: <0.0001pTrtm: <0.0001p=0.005 Saline il-6p<0.0001 NPPreg0 10 20 30 402-ΔΔCt mtDNA pPregn x Trtm: 0.7pPregn: 0.008pTrtm: <0.0001 Saline tnf-α p<0.0001 p<0.0001 NPPreg0 20 40 602-ΔΔCt mtDNA pPregn x Trtm: 0.13pPregn: 0.08pTrtm: <0.0001 Saline ifn-γ p<0.0001 p<0.0001 NPPreg0 5 10 15 20 252-ΔΔCt mtDNA pPregn x Trtm: 0.01pPregn: 0.02pTrtm: <0.001 Saline il-10p<0.001 NPPreg0 10 20 302-ΔΔCt mtDNA pPregn x Trtm: 0.23pPregn: <0.0001pTrtm: <0.0001 Saline tlr9p<0.0001 p=0.0004 A. B. C. D. E. F. NPPreg0 2 4 62-ΔΔCt mtDNA pPregn x Trtm: 0.65pPregn: 0.0005pTrtm: 0.36 Saline il-4p=0.04p=0.007 G. NPPreg0246204060801002-ΔΔCt Saline pPreg x Trtm: <0.0001pPreg: 0.01pTrtm: <0.0001 mtDNA p<0.0001tlr9 p=0.03 NPPreg0 2 4 6 82-ΔΔCt Saline pPreg x Trtm: 0.28pPreg: 0.02pTrtm: 0.02 mtDNA p=0.03il-4 NPPreg0 1 2 3 42-ΔΔCt Saline pPreg x Trtm: 0.35pPreg: 0.001pTrtm: 0.33 mtDNA il-10 NPPreg0123204060801002-ΔΔCt Saline pPreg x Trtm: <0.0001pPreg: 0.9pTrtm: 0.0005 mtDNA p<0.0001 ifn-γ NPPreg0 20 40 602-ΔΔCt Saline pPreg x Trtm: 0.003pPreg: 0.14pTrtm: <0.0001 p=0.005 mtDNA tnf-α p=0.02 NPPreg0 10 20 30 402-ΔΔCt mtDNA pPreg x Trtm: 0.0002pPreg: 0.41pTrtm: 0.009 p=0.0001Salineil-6 NPPreg0 10 20 30 402-ΔΔCt Saline pPreg x Trtm: 0.02pPreg: 0.08pTrtm: 0.0004 mtDNA p=0.0002il-1β 60 Figure 2 -10-9-8-7-6-5 0 30 60 90 120 150 180 PE [log (M)] PE (% KClmax) Saline (+E)Saline (-E) -10-9-8-7-6-5 0 30 60 90 120 150 180 PE [log (M)] PE (% KClmax) mtDNA (-E)mtDNA (+E) SalinemtDNA0 6 7 8 9pEC50 PE (+E)(-E) p Trtm x D: 0.29pTrtm.: p=0.11pDenudation: <0.001 p < 0.001p < 0.001 -10-9-8-7-6-5 0 30 60 90 120 150 180 PE [log (M)] PE (% KClmax) Saline (+E)Saline (-E) -10-9-8-7-6-5 0 30 60 90 120 150 180 PE [log (M)] PE (% KClmax) mtDNA (+E)mtDNA (-E) SalinemtDNA0 6 7 8 9pEC50 (PE) (+E)(-E) pTrtm x D: 0.27pTrtm: 0.83pDenudation: 0.01 p = 0.05 -12-10-8-6 0 40 80 120 160 U46619 [log (M)] U46619 (% KClmax) Saline (+E)Saline (-E) -12-10-8-6 0 40 80 120 160 U46619 [log (M)] U46619 (% KClmax) mtDNA (-E)mtDNA (+E) SalinemtDNA0 200 400 600 800AUC U46619 (+E)(-E) pTrtm x D.: 0.05pTrtm.: 0.047pDenudation: 0.09 -12-10-8-6 0 40 80 120 160 U46619 [log (M)] U46619 (% KClmax) Saline (+E)Saline (-E) -12-10-8-6 0 40 80 120 160 U46619 [log (M)] U46619 (% KClmax) mtDNA (-E)mtDNA (+E) SalinemtDNA0 200 400 600 800 1000AUC U46619 (+E)(-E) pTrtm x D.: 0.89pTrtm.: 0.30pDenudation: 0.09 Phenylephrine - Non-pregnant Phenylephrine - Pregnant U46619 - Non-pregnant U46619 - Pregnant A. B. C. D. E. F. G. H. I. K. L.J. 61 Figure 3 -10-9-8-7-6-5-4 0 20 40 60 80 100 ACh [log (M)] % Relaxation (PE) Saline mtDNA -10-9-8-7-6-5-4 0 20 40 60 80 100 ACh [log (M)] % Relaxation (PE) Saline mtDNA NP Preg 0 20 40 60 80 100Emax ACh Saline mtDNA pPregn x Trtm.: 0.03 pPreg.: 0.0008 pTrtm.: 0.004 p = 0.0017 p = 0.0005Non-pregnant Pregnant A. B. C. 62 Figure 4 Saline nDNA 0 6 7 8 9 ACh pEC50 -10-9-8-7-6-5-4 0 20 40 60 80 100 ACh [log (M)] % Relaxation (PE) Saline nDNA SalinenDNA 0 70 80 90 100Emax (% PE) A. B. C. D. -10-9-8-7-6-5 0 30 60 90 120 150 180 PE [log (M)] PE (% KClmax) Saline (+E) nDNA (+E) Saline (-E) nDNA (-E) SalinenDNA 0 6 7 8 9pEC50 (PE) (+E) (-E) pTrtm x D: 0.68 pTrtm: 0.009 pDenudation: <0.001 p = 0.004p = 0.003 Saline nDNA 0 100 200 300Emax (% KClmax) (+E) (-E) pTrtm x D : 0.46 pTrtm : 0.17 pDenudation : 0.56 E. F. 63 Figure 5 -10-9 -8 -7 -6 -5 -4 0 20 40 60 80 100 ACh [log (M)] % Relaxation (PE) Saline mtDNA mtDNA+ODN2088 ODN2088 6.0 6.5 7.0 7.5 8.0 8.5 ACh pEC50 p = 0.0051 Saline: + - - + mtDNA: - + + - ODN2088: - - + + A. B. 64 Figure 6 65 Figure 7 SalinemtDNA 0.0 0.5 1.0 1.5 2.02-ΔΔCt 0.0358 sod-1 mtDNASmtDNAS 20 kDa SOD-1 SalinemtDNA 0 50 100 150 200 SOD-1/total protein (% average saline) 15 kDa SalinemtDNA 0 1 2 3 4 52-ΔΔCt sod-2 SalinemtDNA 0 1 2 3 42-ΔΔCt catalase SalinemtDNA 0 100 200 300 400 500 SOD-2/total protein (% average saline) SOD-2 20 kDa 25 kDa 37 kDa SmtDNASmtDNA p = 0.0556 SalinemtDNA 0 50 100 150 200 Catalase /total protein (% average saline) Catalase 50 kDa 75 kDa 100 kDa SmtDNASmtDNA A. B. D. E. F. C.