Detrimental effects of transient cerebral ischemia on middle cerebral artery mitochondria in female rats

Abstract

Mitochondrial numbers and dynamics in brain blood vessels differ between young male and female rats under physiological conditions, but how these differences are affected by stroke is unclear. In males, we found that mitochondrial numbers, possibly due to mitochondrial fission, in large middle cerebral arteries (MCAs) increased following transient middle cerebral artery occlusion (tMCAO). However, mitochondrial effects of stroke on MCAs of female rats have not been studied. To address this disparity, we conducted morphological, biochemical, and functional studies using electron microscopy, Western blot, mitochondrial respiration, and Ca²⁺ sparks activity measurements in MCAs of female, naïve or sham Sprague–Dawley rats before and 48 h after 90 min of tMCAO. Adverse changes in mitochondrial characteristics and the relationship between mitochondria and sarcoplasmic reticulum (SR) in MCAs were present on both sides. However, mitochondria and mitochondrial/SR associations were often within the range of normal appearance. Mitochondrial protein levels were similar between ipsilateral (ipsi) and contralateral (contra) sides. Nonrespiratory oxygen consumption, maximal respiration, and spare respiratory capacity were similar between ipsi and contra but were reduced compared with sham. Basal respiration, proton leak, and ATP production were similar among MCAs. Ca²⁺ sparks activity increased in sham and ipsi MCAs exposed to a mitochondrial ATP-sensitive potassium channel opener: diazoxide. Our results show that tMCAO has effects on mitochondria in MCAs on both the ipsi and contra sides. Mitochondrial responses of cerebral arteries to tMCAO in females are substantially different from responses seen previously in male rats suggesting the need for specific sex-based therapies.

NEW & NOTEWORTHY We propose that differences in mitochondrial characteristics of males and females, including mitochondrial morphology, respiration, and calcium sparks activity contribute to sex differences in protective and repair mechanisms in response to transient ischemia-reperfusion.

INTRODUCTION

Healthy and adaptable mitochondria are essential to satisfy the high energy demands of the neurovascular unit. Although neurons and glial cells have long been recognized as having extensive metabolic needs, the brain circulation also has steep energy requirements (beyond the maintenance of normal cellular structure and function of blood vessels) for the maintenance of tight junctions, transport of substances across the blood-brain barrier (BBB), and the constant resistance vessel adjustments to match metabolic demand while protecting the vulnerable microvasculature from upstream arterial pressure. Mitochondria are a major source of energy in the form of ATP for brain vascular cells (1–10). Brain circulation and mitochondrial numbers and function are influenced by multiple factors: sex, insulin resistance/diabetes, aging, and ischemic stress (5–9, 11–17). For example, we found that large arteries, such as the middle cerebral artery (MCA), have greater mitochondrial mass and numbers in females than males in young rats, and this greater abundancy of mitochondria is reflected in greater mitochondrial respiration in female arteries compared with male arteries (9). In addition, MCA dilation to mitochondrial ATP-sensitive potassium (mK_ATP) channel activator is greater in females than in males (9). The effects of female sex hormones, specifically estrogen, on the vasculature and mitochondria are well studied in both rodents and humans (reviewed in Ref. 18). One of the most important estrogenic actions on the vasculature is the endothelial nitric oxide-mediated vasodilation. Vasodilatory effects of estrogen were demonstrated by increased flow-mediated vasodilation in postmenopausal women in response to estrogen treatment (19, 20). Similarly, estrogen-mediated vasodilatory effects decreased myogenic tone of different vascular beds from rodents upon exposure to estrogen, resulting in a greater blood flow (21–23). Estrogen not only affects vascular function but has profound effects on mitochondria by increasing oxidative phosphorylation (reviewed in Ref. 18). The brain is a high-energy demand organ and brain endothelial cells, including microvessels that form the BBB, contain more mitochondria than those in the periphery (24, 25). Thus, disruption of mitochondrial structure and/or function has important effects on cellular energetics, vascular responsiveness, and transport mechanisms contributing to or underlying cerebrovascular pathologies. Moreover, in recent RNA sequencing and proteomic studies, we have shown that sex differences in arteries of young rats extend to the microvascular level (12). Specifically, energy producing pathway-related protein abundance (such as the tricarboxylic acid cycle or oxidative phosphorylation) were greater in female microvessels compared with males, suggesting increased energy demand (13). A possible synergistic interaction between greater ATP production and fuel source flexibility during reproductive years may contribute to greater female vasculature resilience compared with male. Mitochondria in MCAs are only modestly affected by insulin resistance or type 2 diabetes in rats, perhaps illustrating organelle ability to adapt to fluctuating blood glucose levels or other pathologies associated with the metabolic syndrome (6–8, 16).

Mitochondria in MCAs of male rats are also resilient to ischemic stress. We found that young, male, MCA mitochondria exhibited a robust response to transient cerebral ischemia, resulting in increased mitochondrial mass, mitochondrial respiration, and retained vascular responsiveness to mK_ATP channel activation (5). We and others have reported that pharmacological activation of mK_ATP channels in the inner mitochondrial membrane results in mitochondrial depolarization and superoxide anion production leading to vasodilation. Superoxide anions activate the ryanodine-sensitive Ca²⁺ channels in the sarcoplasmic reticulum (SR) leading to Ca²⁺ sparks, and the opening of large-conductance BK_Ca channels, allowing K⁺ efflux and vasodilation due to the subsequent decrease in the intracellular Ca²⁺ and hyperpolarization of vascular smooth muscle (VSM) cells via the inactivation of voltage-gated calcium channels. Thus, greater mitochondrial mass may indicate the presence of more mK_ATP channels (reviewed in Ref. 11). The opening of mK_ATP channels may result in stimulation of cellular oxygen consumption and a subsequent increase in proton pumping to maintain mitochondrial membrane potential (reviewed in Ref. 26). A surprising finding in male rats was that mitochondrial numbers and function in the MCAs on the nonischemic hemisphere were reduced compared with the ischemic side. Whether nonischemic changes in male, cerebral vascular mitochondria are also present in female rats following ischemic stress is unknown. Our study explores mitochondrial morphology and function in response to experimental stroke in young female rats. We performed morphological and functional studies using electron microscopy to document morphology, Western blot to determine mitochondrial protein levels, Agilent Seahorse Bioscience XFe24 Analyzer to measure mitochondrial respiration, and Ca²⁺ sparks activity to evaluate the functional mitochondrial-SR interactions in MCAs of female, Sprague–Dawley rats before and 48 h after 90 min of transient middle cerebral artery occlusion (tMACO) and in female rats undergoing sham surgery.

MATERIALS AND METHODS

Animals

Age-matched, female, Sprague–Dawley rats (n = 55, 9–12 wk old, Charles River Laboratories, Wilmington, MA) were housed with standard light/dark cycle and free access to food and water, ad libitum, in accordance with the approval of and the guidelines of the Institutional Animal Care and Use Committee of Tulane University and the National Institutes of Health Office of Laboratory Animal Welfare. Experiments were conducted according to the Animal Research: Reporting in Vivo Experiments, ARRIVE guidelines. Rats were exposed to 90 min of ischemia induced by transient middle cerebral artery occlusion followed by 48 h reperfusion (5). We used the same protocol as previously used in male rats for comparison purposes. Anesthesia was induced with an intraperitoneal ketamine and xylazine injection (80–110 mg/kg of 100 mg/mL KetaVed, St. Joseph, MO; 5–10 mg/kg of 20 mg/mL xylazine, Santa Cruz Biotechnology, Dallas, TX; respectively), whereas buprenorphine (0.05–0.1 mg/kg of 0.3 mg/mL MWI Veterinary Supply, Dallas, TX) was used for analgesia. Then, a silicon-coated monofilament (Doccol, Sharon, MA) was advanced to the origin of the right MCA via the common and internal carotid arteries. Body temperature was maintained via a thermometer-controlled heating pad. After 48 h reperfusion, animals were euthanized and decapitated using a rodent guillotine. A 1% solution of 2,3,5-triphenyl-2H-tetrazolium chloride staining (TTC; No. T8877, Sigma-Aldrich, St. Louis, MO) was used to visualize infarct, representing 26 ± 3% of the ipsilateral, ischemic hemisphere with no apparent infarcted tissue on the contralateral side.

Electron Microscopy

For electron microscopy, studies were carried out on 2% glutaraldehyde and 3% formaldehyde-fixed MCAs (5). Osmium tetroxide (1%) was used to post fix the samples followed by Spurr’s resin embedding. Ultrathin, 80–90-nm artery sections were mounted on formvar-coated copper grids (200 mesh), and then air-dried and stained using uranyl acetate and lead citrate (7 min and 7 min, respectively). Imaging was done at different magnifications using an FEI Tecnai BioTwin 120 keV transmission electron microscopy with a digital imaging setup (Wake Forest University Health Sciences, Winston-Salem, NC). Jpeg and tiff images were obtained at ×24,500 (Fig. 1, A, C, and E) and at ×29,000 (Fig. 1, B, D, and F) with a scale bar of 500 nm in length.

Figure 1.

Mitochondrial morphology of middle cerebral arteries (MCAs) in vascular smooth muscle (VSM) cells following experimental stroke and in sham and naïve female rats. A and B: high-magnification images of ipsilateral (ipsi) MCAs from two rats 48 h after transient middle cerebral artery occlusion (tMCAO). Mitochondria in VSM cells of ipsilateral (ipsi) arteries often appeared damaged (indicated by arrows), displaying condensed, dark appearance, or decreased numbers (circle). However, many post-transient middle cerebral artery occlusion (tMCAO) VSM cells with damaged mitochondria were found adjacent to neighboring populations of healthy mitochondria, and normal-looking mitochondria often occurred in VSM cells with damaged mitochondria. Although the VSM cells normally showed close associations between mitochondria and sarcoplasmic reticulum (SR), which is essential for the generation of calcium sparks, in some sections mitochondria appeared to be absent (A, circle) from areas where they normally would be present. Also, increased separation among mitochondrial and SR membranes was often present in VSM cells. Mitochondria showed various forms of damage, including localized areas where normal morphology of internal structures such as cristae were lost, generalized mitochondrial swelling, or increased density of staining, which is consistent with condensation of internal structures. C and D: high-magnification image of VSM cells from MCAs from two rats 48 h after tMCAO on the opposite, contralateral (contra) side. Although many VSM cells have normal-looking mitochondria, many areas of VSM cells showed damaged mitochondria (arrows) or shrunken organelles. However, as with the VSM cells on the ipsi side, many VSM cells showed normal-looking mitochondria, even in the same cell containing damaged mitochondria. E and F: high-magnification image of VSM cells from a sham-operated rat (E) and a naïve rat (F). Mitochondria in both cases had normal-looking morphology with densities similar to our previously published images. Mitochondria in VSM were largely arrayed in dense fields with interspersed SR in sham arteries. In sham and naïve rats, occasional mitochondria observed in VSM cells lacked internal structures indicating mitophagy, a characteristic present in otherwise normal cells under physiological conditions. Images A, C, and E were at ×24,500 and B, D, and F were at ×29,000. Bars are 500 nm in length.

Western Blot

The MCAs were isolated, cleaned from the surrounding connective tissue, and homogenized in ice-cold NP40 lysis buffer (No. FNN0021, Invitrogen, Frederick, MD) containing phosphatase and protease inhibitors (No. P0044 and No. P8340, Sigma-Aldrich, St. Louis, MO) and then centrifuged. The supernatant was used for Pierce bicinchoninic acid (BCA) protein assay (No. PI23225, Thermo Scientific, Waltham, MA) and Simple Wes (Protein Simple Western, WS3259, San Jose, CA) experiments to determine protein expression (27). Sample-specific conditions were titrated to achieve a linear range of detection using isolated cerebral artery homogenates and manufacturer suggested positive control (rat liver tissue lysate-mitochondrial extract provided with Abcam Cat. No. ab110414, RRID:AB_2687585). Separation module reagents (SM W004 and PS-ST01EZ, Protein Simple Western, San Jose, CA) were prepared as instructed by the manufacturer’s protocol. Briefly, 40 µL of dH₂O was added to the dithiothreitol (DTT) tube and mixed, resulting in a stock concentration of 400 mM. A 5× fluorescent master mix was prepared by adding 20 µL of 10× sample buffer (provided in the kit) and 20 µL of the 400 mM DTT solution, and then mixed. The biotinylated ladder was prepared by adding 20 µL of dH₂O to the biotinylated ladder tube, and then mixed. All samples, except for S2 ipsilateral, were diluted using 0.1× sample buffer to have a final concentration of 0.2 mg/mL, then 5× fluorescent master mix was added to the samples as 1:4, and mixed. Samples were then denatured at 85°C for 5 min in a dry heat block and kept on ice. The provided Luminol-S and H₂O₂ were mixed in a 1:1 ratio and kept on ice until use. Samples were normalized to total protein level using a total protein detection kit and biotin labeling reagents (DM-TP01-1 and No. 042-973; Protein Simple, San Jose, CA). Reagents were prepared according to the manufacturer’s instructions. Briefly, the biotin labeling reagent was prepared before plate loading using 150 µL of reconstitution agent 1, followed by mixing with a pipette, and an addition of 150 µL of reconstitution agent 2, followed by gently mixing with a pipette. Plate pipetting for biotinylated ladder, samples, antibody diluent, primary antibody, streptavidin horseradish peroxidase (HRP), secondary antibody, luminol-peroxidase mix, total protein labeling reagent, and wash buffer were done according to the manufacturer provided schematic plate layout using the manufacturer instructed volumes, alternating the total protein and immunoassay reagents (Fig. 2E). We used a membrane integrity antibody cocktail (Abcam Cat. No. ab110414, RRID:AB_2687585; 1:1,250) containing five mouse monoclonal antibodies against mitochondrial structure proteins, namely porin, cytochrome-c, complex Va and complex III Core 1, and cyclophilin D, and the appropriate anti-mouse secondary antibody from the Simple Wes kit (No. 042-205; Goat Anti-Mouse Secondary HRP Conjugate; Protein Simple, San Jose, CA: 10 µL/well according to manufacturer instruction). The assay was run on a Wes WS 3259 with firmware 2.5.30766 using the total protein assay protocol. Immunoassay results were normalized to total protein (Fig. 2D) following the manufacturer’s protocol by selecting capillary No. 2 as the total protein reference to calculate the normalization factor for each total protein capillary. The normalization factor was used to multiply the corresponding sample area to calculate the corrected immunoassay area. We used the Shapiro–Wilk test in GraphPad Prism 9 to assess normal distribution and results were analyzed using paired, two-tailed parametric t test.

Figure 2.

Western blots show no significant differences in select mitochondrial proteins between stroke and nonstroke hemispheres. Protein amounts of cytochrome-c (A), cyclophilin D (B), and complex III/V (C) proteins were similar. Data showed normal distribution when tested using the Shapiro–Wilk test. Data were expressed as means ± SD, n = 5 for each group, from five rats. No significant difference was detected using paired, parametric, two-tailed, t tests. Superimposed signals of total protein from individual samples (D), and the proteins of interest (E) with biotinylated ladder (marking 12, 40, 66, 116, 180, and 230 kDa) exported from the protein simple Compass for Simple Wes. Green color indicates cytochrome-c, pink denotes cyclophilin D, whereas light blue corresponds to the complex III/V signals. Contra, nonstroke side; ipsi, stroke side.

Mitochondrial Respiration

The mitochondrial oxygen consumption rate (OCR) of isolated MCAs was measured using a Seahorse Bioscience XFe24 Analyzer in a 24-well islet plate (No. 101122-100, Agilent Technologies, Santa Clara, CA) as described previously (5–7, 9). Each well contained 525 µL of Seahorse XF Assay medium (No. 102365-100, Seahorse Bioscience) with 5.0 mmol/L glucose and 2.0 mmol/L pyruvate, at pH 7.4 and 37°C. Eight measurement cycles were used for baseline OCR with a media injection after the third measurement, followed by the sequential injection of oligomycin (2 μmol/L), carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP, 1 μmol/L), and 1.5 μmol/L each of antimycin/rotenone (XF Cell Mito Stress Test Kit, No. 103015-100, Agilent Technologies, Santa Clara, CA). OCR was measured in the presence of the listed drugs throughout five measurement cycles. The components of mitochondrial respiration were calculated using protein normalized values as described previously (5–7, 9). Data distribution was tested using the D’Agostino and Pearson tests within GraphPad Prism 9. Statistical testing of data with normal distribution was done using parametric, one-way ANOVA with Tukey’s multiple comparison test.

Calcium Sparks Imaging

Zeiss laser scanning confocal system (LSM 510 and 7 Live, Jena, Germany) with a Zeiss C-Apochromat 63 × 1.2 numerical aperture (NA) water immersion objective was used to measure Ca²⁺ sparks activity in MCAs in the presence of dimethyl sulfoxide (DMSO) or 250 µM of diazoxide (D8418 and D9035, respectively, Sigma Aldrich, St. Louis, MO). MCAs were incubated in 5 µM of Fluo-4 AM (F-14201, Invitrogen, Frederick, MD) and 20% Pluronic F-127 (P-3000MP, Invitrogen, Frederick, MD) dissolved in HEPES (No. 16928, Affymetrix) for 1 h while protected from light. Afterward, samples were washed and kept in HEPES solution for 30 min at room temperature for deesterification (16, 28). ImageJ Fiji Software (NIH) with the xySpark plugin (University of Leeds, UK) was used to calculate and normalize sparks to 10 µm² cell area of 512 × 512-pixel dimension confocal images (5, 16). The normal distribution of data was tested using the D’Agostino and Pearson tests. Then, the nonparametric ANOVA with Kruskal–Wallis test, and Dunn’s multiple comparisons tests were used to compare groups.

Data Analysis and Statistics

Results were expressed as the means ± SD. The number of measurements is indicated by “n.” Data distribution was tested with GraphPad Prism 9 using the Shapiro–Wilk or D’Agostino and Pearson tests according to sample numbers. Data comparison with normal distribution was performed using parametric, two-tailed Student’s t test or one-way ANOVA with Tukey’s multiple comparison test, whereas nonparametric t test or one-way ANOVA with Dunn’s multiple comparisons were used for data without a normal distribution. P < 0.05 was considered statistically significant. Although blinding was not applied, the experiments were completed in the presence of two coauthors, and the data analysis and animal group assignment were done randomly.

RESULTS

Electron Microscopy

Mitochondria in VSM of MCAs from both stroke and nonstroke hemispheres showed evidence of damage 48 h after tMCAO. We focused on VSM cells, which contain large numbers of easily identified mitochondria and changes in morphology and mitochondria/SR relationships are easier to observe. Ipsilateral artery mitochondria often appeared damaged (indicated by arrows), displayed a condensed, dark appearance, or decreased numbers (circle) (Fig. 1, A and B). However, many post-tMCAO VSM cells with damaged mitochondria were found adjacent to populations of healthy, normal-looking mitochondria and normal-looking mitochondria often occurred in VSM cells with damaged mitochondria. Although VSM cells normally showed close associations between mitochondria and SR, essential for the generation of calcium sparks, mitochondria in some sections appeared to be absent (Fig. 1A, circle) from locations where they would normally be present. Although difficult to quantify without extensive staining approaches, it appeared that spatial separation occurred between mitochondrial and SR membranes. Damaged mitochondria showed various forms of mitochondrial damage, including localized areas where internal structures such as cristae were lost, generalized mitochondrial swelling occurred, or increased staining density was present. Although many mitochondria in VSM cells from the nonstroke side looked normal, many areas showed damaged (arrows) or shrunken mitochondria (Fig. 1, C and D). However, similar to the ipsilateral side, VSM cells on the contralateral side also showed normal-looking mitochondria, even in cells containing damaged mitochondria. Mitochondrial morphology in brain VSM cells following tMCAO contrasted dramatically with mitochondria of sham or naïve cells (Fig. 1, E and F). Mitochondria in both sham and naïve MCAs displayed normal-looking morphology with densities similar to our previously published images (5, 9). Notably, mitochondria in VSM cells in MCAs from naïve and sham female rats normally occur in clusters containing many mitochondria, have roughly similar morphology, and have well-defined cristae (Fig. 1, E and F). Moreover, the identified mitochondrial clusters are usually closely associated with the SR. In sham and naïve rats, occasional mitochondria observed in VSM cells lacked internal structures indicating mitophagy, a characteristic present in otherwise normal cells under physiological conditions.

Western Blot

The antibody cocktail was expected to detect five proteins; however, we were unable to separate complex III and V signals because of their proximity and the strong complex III signal (Fig. 2). Therefore, we attributed the peak area to the sum of complexes III (49 kDa) and V (55 kDa) signals. In addition, we observed a peak area at the expected porin location (39 kDa) but due to a poor signal-to-noise ratio we did not include this in our analysis. Data normality was tested in GraphPad Prism 9 using the Shapiro–Wilk test due to the sample size. Data showed normal distribution for all investigated proteins: cytochrome-c (detected at 16 kDa), cyclophilin D (detected at 25 kDa), complexes III and V (detected at 49–50 kDa). There were no significant differences between the total protein normalized immunoassay areas of the investigated proteins: cytochrome-c (ipsi: 321,578 ± 91,059; contra: 382,959 ± 47,402; paired, two-tailed parametric t test, P = 0.149, n = 5 rats), cyclophilin D (ipsi: 151,768 ± 31,996; contra: 144,046 ± 12,940; paired, two-tailed parametric t test, P = 0.638, n = 5 rats), complex III/V (ipsi: 596,040 ± 198,425; contra: 768,816 ± 136,929; paired, two-tailed parametric t test, P = 0.162, n = 5 rats) (Fig. 2).

Mitochondrial OCR Measurement

Mitochondrial respiration of MCAs and the individual components of OCR were similar between the ischemic and contralateral sides (Fig. 3) collected 48 ± 6 h reperfusion. Data normality was tested in GraphPad Prism 9 using D’Agostino and Pearson tests due to the sample size. Data showed normal distribution, and parametric, one-way ANOVA with Tukey’s multiple comparison test was used to compare the groups (n = 12–13) for nonmitochondrial respiration (ipsi: 44.18 ± 27.4; contra: 39.63 ± 32.10; sham: 116.3 ± 60.85), proton leak (ipsi: 110.5 ± 93.21; contra: 68.26 ± 47.58; sham: 127 ± 120.9), ATP production (ipsi: 17.25 ± 12.15; contra: 12.62 ± 10.47; sham: 19.86 ± 19.79), basal respiration (ipsi: 127.8 ± 99.15; contra: 80.88 ± 47.7; sham: 146.9 ± 120.4), maximal respiration (ipsi: 216.2 ± 162.5; contra: 207 ± 156.9; sham: 402.5 ± 176.8), and for spare respiratory capacity (ipsi: 88.46 ± 81.12; contra: 126.1 ± 129.7; sham: 255.7 ± 161.2) (Fig. 3, A and B). Nonrespiratory oxygen consumption, maximal respiration, and spare capacity in MCAs were also similar between ipsilateral and contralateral sides, both were reduced compared with sham rats. Basal respiration, ATP production, and proton leak were similar among MCAs from ipsilateral, contralateral, or sham hemispheres.

Figure 3.

Respiration of middle cerebral arteries (MCAs). A: continuous tracings of protein-normalized mitochondrial oxygen consumption rate (OCR) expressed in pmol/min/µg protein unit in each group. Arrows indicate Seahorse experimental design and show the sequence of agent injections to elicit different components of mitochondrial respiration. A/R, antimycin/rotenone; F, FCCP; M, media; O, oligomycin. B: calculated values of mitochondrial OCR with statistical analyses shown for nonmitochondrial respiration, proton leak, ATP production, basal respiration, maximal respiration, and spare respiratory capacity in ipsilateral (ipsi), contralateral (contra), and sham MCAs. Data are expressed as means ± SD. Data showed normal distribution when tested using the D’Agostino and Pearson tests. *ispi vs. sham; †contra vs. sham, one-way ANOVA with Tukey’s multiple comparisons test, P < 0.05. n = 12 or 13/group from 12 and 13 rats.

Calcium Sparks Measurement

The Ca²⁺ sparks activity was moderately affected by tMCAO (Fig. 4). Data normality was tested in GraphPad Prism 9 using D’Agostino and Pearson tests due to sample size. Normal distribution was not shown for all investigated parameters, therefore nonparametric ANOVA with Kruskal–Wallis test and Dunn’s multiple comparisons tests were used to compare groups. Outliers were identified by the ROUT method within GraphPad Prism 9, and a total of nine outliers were excluded from the entire data set (originating from 19 animals). Ca²⁺ sparks activity was significantly greater in ipsilateral arteries at baseline (0.08366 ± 0.06919; n = 49) in the presence of DMSO compared with the contralateral group (0.03436 ± 0.02781; n = 16). Diazoxide increased Ca²⁺ sparks activity in both groups that reached significance in the ipsi group (ipsi: 0.1377 ± 0.08737, n = 37; contra: 0.08966 ± 0.07680, n = 18). In addition, Ca²⁺ sparks activity was significantly increased in sham MCAs in the presence of diazoxide (DMSO: 0.06458 ± 0.05616, n = 32; DZ: 0.1306 ± 0.1018, n = 52) but not in the naïve group (DMSO: 0.2059 ± 0.1277, n = 64; DZ: 0.2184 ± 0.1094, n = 87).

Figure 4.

Calcium sparks activity in isolated middle cerebral arteries (MCAs). Ca²⁺ sparks activity was determined in the presence of DMSO (vehicle) and diazoxide (DZ) in isolated ipsilateral (ipsi), contralateral (contra), sham, and naïve MCAs. Data are expressed as means ± SD. Normality was tested using the D’Agostino and Pearson tests. ipsi DMSO, contra DMSO, sham DZ, and naïve DMSO groups did not pass normality testing, nonparametric Kruskal–Wallis, one-way ANOVA followed by Dunn’s multiple comparison test was used to compare vehicle and treatment within groups, *DMSO treatment vs. DZ, P < 0.05. n = 16–87/group, from 19 rats.

DISCUSSION

The major finding of this study is that tMCAO has adverse effects on mitochondria in MCAs distal to the occlusion as well as MCAs on the contralateral, nonstroke side in young female rats. Furthermore, the mitochondrial responses in female rat MCAs are substantially different than responses we observed previously in young male rats (5). In our current study, mitochondria often showed adverse morphological changes not only on the ipsilateral, stroke side but also on the nonstroke, contralateral hemisphere. Rather than enhanced mitochondrial protein mass and function seen previously in males on the side ipsilateral to stroke, female MCAs showed reduced mitochondrial respiration and similar calcium sparks activity in both hemispheres compared with sham rats. In contrast, male rats showed considerable differences between the ipsilateral and contralateral hemispheres in mitochondrial mass and function on the stroke side (5). However, a common feature in both males and females was the detrimental effect of tMCAO on mitochondria in MCAs on the side contralateral to the ischemic episode. Thus, the results of our current study support and extend our earlier findings showing large differences in mitochondrial dynamics between young male and female rats under normal conditions, which probably explain the observed sex-specific mitochondrial responses following ischemic challenge (Fig. 5). The findings warrant further study of chronological changes in mitochondrial morphology and function via the inclusion of middle-aged and aged rats.

Figure 5.

Potential consequences of sex differences in mitochondrial dynamics in brain arteries. Greater mitochondrial mass and function and the ability to use alternative fuels in females, but not males, may be advantageous during normal fluctuations in physiological status, and during moderate and chronic disease conditions. However, the rapid upregulation of mitochondrial dynamics seen in males but not females, may provide greater resiliency to brain arteries for recovery and/or for responding to additional stresses following acute, severe stresses such as ischemia-reperfusion.

Mitochondria are complex organelles composed of a permeable outer membrane, an impermeable inner membrane, the intermembrane space, inner membrane extensions called cristae, and the intra-cristae spaces: the matrix (reviewed in Refs. 11, and 29–31). Mitochondria possess their own DNA and protein synthesizing capability that produces approximately one-half of their proteins whereas the other proteins are encoded by nuclear DNA and then imported into mitochondria. The cytochrome-c oxidase complex, which is a major electron transport chain component located on the inner membrane, is a good example of the dual origin of proteins present in mitochondria since ten subunits are nuclear in origin and three are synthesized within the mitochondria. In addition to protein synthesis of mitochondrial components, mitochondrial numbers, shapes, and mitophagy are controlled by fission and fusion proteins. Thus, maintenance, repair, replacement, and enhanced numbers require many coordinating steps for full mitochondrial functionality, involving de novo protein synthesis, transport, and placement of mitochondrial elements, and adequate ATP availability from oxidative phosphorylation and glycolysis to accomplish these actions. Both our previous and current results indicate that responses to severe stress such as ischemia trigger a less robust recovery response in young, female rat MCAs compared with MCAs of young, males rats, which have lower mitochondrial numbers in MCAs under normal conditions (5). Recovery response results, originating from whole vascular samples containing both endothelial and smooth muscle cells, lead to two speculations: 1) female MCAs are primed to respond to moderate physiological levels of stress including but not limited to stress involving hormonal fluxes during the menses/estrus cycle and during pregnancy, whereas male brain blood vessels are equipped to respond to more severe forms of stress, and 2) our recent RNAseq findings in brain microvessels show a female advantage in that several important signaling pathways are present in females that provide greater resiliency and repair of blood vessels against stress and injury for females (12). Treatment modalities need to be specifically tailored to sex and hormonal status, which could involve target different signaling cascades as well as differential timing and dosing of drug treatments and physical therapy (32–34).

A surprising finding from our previous study on males (5), and now confirmed in female MCAs, is that adverse changes in mitochondrial structure and function occur at sites distant from ischemic or infarcted tissues. The nature of the signaling cascades from ischemic/infarcted tissue to distant mitochondria of otherwise healthy arteries is unclear. In our male MCAs study, we found that not only mitochondria-driven dilation but also dilation responses to acetylcholine and bradykinin were reduced in the nonischemic hemisphere, indicating widespread impairment of the brain vasculature distant from initial injury (17). Our results showing significant impairment of contralateral vascular reactivity were confirmed by two other laboratories in male rats (35–37), and thus support the use of sham or naïve MCAs in studies involving localized brain ischemia. Ours is the first report of contralateral vascular dysfunction, in this case, related to MCA mitochondrial characteristics, in female animals. Although mechanisms are unclear at this time, harmful factors, either free or contained in extracellular vesicles, from brain parenchyma impacted by ischemia have easy access to distant areas via the cerebrospinal fluid. The identity of harmful factors and mechanisms of action are unknown and is currently the subject of study. Nonetheless, the implication is that immediate stroke treatments as well as poststroke therapies need to consider how the insult affects blood vessels in different brain areas.

Transient ischemia not only directly affected mitochondrial structure and function but also disrupted mitochondrial interactions with other cellular organelles. Calcium sparks activity, which promotes VSM relaxation as well as other, cell-type specific activities, involves close coordination between mitochondria and SR in VSM. The precise mitochondrial mechanisms underlying the generation of Ca²⁺ sparks activity by SR are controversial. Some authors have indicated that mitochondria-derived reactive oxygen species (ROS) induce Ca²⁺ sparks activity by mitochondria (38), whereas we have shown that depolarization of mitochondrial alone is sufficient (16) to promote Ca²⁺ sparks activity. Diazoxide, the agent used in our studies, depolarized mitochondria as well as generated ROS, and was used to maximize the generation of Ca²⁺ sparks by SR. In male MCAs, we reported previously that Ca²⁺ sparks activity was impaired on both the stroke and nonstroke side, unlike in females, probably due to loss of functional mitochondria and/or disruption of the tight morphological coupling between mitochondria and SR as shown by electron microscopy (5). However, similar to female MCAs, Ca²⁺ sparks activity in male MCAs was reduced on the contralateral side in response to the mK_ATP activator. Thus, considering our previous and current data on vasodilator failure to several substances that act through different mechanisms, it appears that the detrimental effects of substances released by a damaged brain impact distant arteries at several, distinct levels.

Our finding of a smaller infarct area in the brains of female rats compared with male rats, despite the same duration of MCA occlusion, is consistent with reports by other laboratories. The reasons for sex-dependent differences in the damaged brain are complex but probably involve hormonal and nonhormonal factors that affect brain resiliency and recovery from severe stress (32–34, 39–42). We do not believe the smaller infarct size detracts from our findings for two reasons. First, we studied an MCA segment distal to the occlusion, which thus experienced the same duration of blood flow stoppage and local anoxia. Second, the smaller infarcted area, if this was the prevailing consideration, would be expected to prompt retained mitochondrial structure and function on both the ipsilateral and contralateral MCAs rather than the mitochondrial damage and dysfunction that we observed.

In conclusion, transient cerebral ischemia shows modest effects on mitochondrial characteristics of female cerebral arteries on the stroke side that contrasts with the pattern observed in male rats. Mitochondrial characteristics on the nonstroke side were altered in the MCAs of both sexes, probably via the transport of harmful agents released into the cerebrospinal fluid from distant infarcted tissue. The foundation of the underlying mechanisms for sex differences in the MCAs are unclear but are probably due to differences in baseline mitochondrial characteristics, and the relative prominence of the protective and repair mechanisms that differ between male and female rat blood vessels. The findings of this study underscore the need to consider both males and females in studies of the brain vasculature in health and disease.

GRANTS

This work was supported by National Institutes of Health Grants HL093554, AG063345, and HL148836 (to D. W. Busija) and NS094834, NS114286, and AG074489 (to P. V. Katakam). This research was also funded in whole or in part by the Louisiana Board of Regents Endowed Chairs for Eminent Scholars program (to D. W. Busija).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

I.R. and D.W.B. conceived and designed research; I.R., I.M., S.W., and C.M. performed experiments; I.R., I.M., S.W., and C.M. analyzed data; I.R. and D.B. interpreted results of experiments; I.R. and D.B. prepared figures; I.R. and D.B. drafted manuscript; I.R., I.M., S.W., C.M., P.K.C., P.V.K., and D.W.B. edited and revised manuscript; I.R., I.M., S.W., C.M., P.K.C., P.V.K., and D.W.B. approved final version of manuscript.