Mitochondrial Oxidative Stress Enhances Vasoconstriction by Altering Calcium Homeostasis in Cerebrovascular Smooth Muscle Cells under Simulated Microgravity

Exposure to microgravity results in postflight cardiovascular dysfunction and orthostatic intolerance in astronauts, which poses a threat to the health of the astronauts and safety during spaceflight. Although much progress has been made in this field, the underlying mechanism remains to be established.

Studies on ground-based rodent animals have reported region-specific vascular structural and functional remodeling in hindlimb unweighting (HU) rat arteries^[1-4]. Both nitric oxide (NO) and endothelin-1 in endothelial cells are modified in a simulated microgravity setting^[5]. The levels of endothelial nitric oxide synthase and nitrate/nitrite content in cerebral arteries increase^{[6, 7]}, which was also confirmed in human umbilical vein endothelial cells and human microvessel endothelial cells exposed to microgravity simulated by a random positioning machine and rotating wall vessel^{[8, 9]}. Calcium influx and efflux are modulated by large-conductance calcium-activated K⁺ channels (BK_Ca) and L-type Ca²⁺ channels in HU rat cerebral vascular smooth muscle cells (VSMCs), which regulate vascular tension^{[10, 11]}. Plasma membrane calcium-regulating channels, sarco/endoplasmic reticulum Ca²⁺ ATPase, ryanodine receptors, and the inositol 1,4,5-trisphosphate receptor (IP₃R) regulate intracellular Ca²⁺ homeostasis in VSMCs^[12]. The influx of extracellular Ca²⁺ and the release of Ca²⁺ from intracellular calcium stores, such as the sarcoplasmic reticulum (SR) and mitochondria regulate cellular Ca²⁺ homeostasis^[13]. However, the mechanism of regulating Ca²⁺ homeostasis in VSMCs exposed to microgravity or simulated microgravity remains unclear.

Reactive oxygen species (ROS) increase cytoplasmic Ca²⁺ concentration ([Ca²⁺]) by facilitating Ca²⁺ release from the endoplasmic reticulum (ER)/SR through the IP₃R^{[14, 15]}. Mitochondrial-derived ROS enhance angiotensin II-triggered vascular contraction by elevating Ca²⁺ and IP₃ levels^[16]. Mitochondria regulate Ca²⁺ through fission and fusion mediated by mitofusion 1/2 (MFN1/2), dynamin-related protein 1 (DRP1), and fission protein 1 (FIS1)^[17-20]. Our previous studies have detected cellular oxidative stress and mitochondrial oxidative injury in HU rat cerebral VSMCs^[21-23] and inhibiting NADPH oxidase improves cerebrovascular reactivity in HU rats^[7]; however, whether these factors are associated with the regulation of Ca²⁺ homeostasis and vascular remodeling is unclear. To better understand the molecular mechanisms of vascular remodeling during microgravity, the present study was designed to investigate the roles and mechanism of mitochondrial oxidative stress during Ca²⁺ homeostasis and the regulation of cerebrovascular contraction in HU rat cerebral arteries.

The handling and treatment of animals were according to the Guiding Principles for the Care and Use of Animals in the Physiological Sciences and were approved by the Chinese guidelines for experimental animals. The care and use of experimental rats were supervised and approved by the Animal Ethical Committee of Chinese PLA General Hospital.

HU was used to simulate microgravity in rats as described in our previous studies^{[6, 24]}. Briefly, male Sprague-Dawley rats were randomly assigned to four groups: control (CON), HU, mitoTEMPO-treated HU (HU + MT), and mitoTEMPO-treated control (CON + MT). To simulate the cardiovascular effect of microgravity, the HU rats were maintained in about a −30° head-down tilt position and housed individually with their hindlimbs unloaded under a 12:12-h light-dark cycle at 23 ± 1 °C with food and water available ad libitum. The control rats were housed in identical Plexiglas cages, except that the tail suspension device was removed. HU + MT and CON + MT rats received distilled water containing mitoTEMPO (Alexis Biochemicals, San Diego, CA, USA) provided at a rate of 0.7 mg/(kg·day) by gavage. The rats in the other groups received an equal volume of vehicle (distilled water). The rats were killed by exsanguination via the abdominal aorta 3 weeks later. The cerebral arteries were rapidly removed and placed in cold Krebs buffer solution containing 118.3 mmol/L NaCl, 14.7 mmol/L KCl, 1.2 mmol/L KH₂PO₄, 1.2 mmol/L MgSO₄·7H₂O, 2.5 mmol/L CaCl₂·2H₂O, 25 mmol/L NaHCO₃, 11.1 mmol/L dextrose, and 0.026 mmol/L EDTA (pH 7.4). VSMCs were dissociated from cerebral arteries as described previously^[11].

The samples were homogenized in 10 mmol/L HEPES lysis buffer (320 mmol/L sucrose, 1 mmol/L EDTA, 1 mmol/L DTT, 10 μg/mL leupeptin, and 2 μg/mL aprotinin, pH 7.40) at 0–4 °C. The homogenate was centrifuged at 12,000 ×g for 10 min at 4 °C. Protein concentrations were determined with a bicinchoninic acid assay kit (Pierce, Rockford, IL, USA). The extracts were fractionated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane. The membranes were incubated with antibodies against IP₃R, MFN1, MFN2, DRP1, FIS1, or GADPH at 4 °C overnight (Abcam, Cambridge, UK). Then, the membranes were incubated with horseradish peroxidase-conjugated anti-mouse or anti-rabbit antibodies for 2 h. The blots were washed and developed with an enhanced chemiluminescent system (Amersham Biosciences, Uppsala, Sweden) according to the manufacturer’s protocol.

Cerebral VSMCs were washed three times in 0.1 mmol/L phosphate buffered solution (PBS), followed by fixation in 1% osmium tetroxide at 4 °C for 3 h. The samples were washed three times in 0.1 mmol/L PBS again and gradually dehydrated in alcohol at different concentrations at 4 °C. After replacing the solution with propylene oxide and embedding using the SPI-Chem Embedding Kit (SPI, Chicago, IL, USA), the samples were fixed at 70 °C for 8 h. The sections (70 nm) were prepared with a Leica EM UC6 ultramicrotome (Leica Microsystems, Wetzlar, Germany), placed on EM grids, and stained with 3% uranyl acetate-lead citrate solution (SPI), followed by washing, drying, and imaging using a JEM1230 TEM (JEOL, Tokyo, Japan).

Paraffin-embedded sections (3 μm) of the basilar arteries were mounted on Thermo Scientific SuperFrost™ plus slides and dried overnight. Briefly, the arterial sections were incubated with primary IP₃R antibody (Abcam). After washing with PBS, the horseradish peroxidase was developed with 3,3-diaminobenzadine (Roche Diagnostics, Mannheim, Germany) as the chromogen substrate. The sections were rinsed, dehydrated in ethanol, cleared in xylene, and mounted. IP₃R staining in the cerebrovascular sections was observed under a microscope.

Cytoplasmic, mitochondrial, and SR Ca²⁺ concentrations were determined using a flow cytometry system with the Ca²⁺ indicators Fluo-4 AM, X-Rhod-1 AM, and Fluo-5N, respectively. Isolated cerebral VSMCs were placed in Hank’s balanced salt solution (Sigma-Aldrich, St. Louis, MO, USA) and incubated with Fluo-4 AM (5 μmmol/L), X-Rhod-1 AM (5 μmmol/L), and Fluo-5N (10 μmmol/L) for 60 min at 37 °C. The Ca²⁺ content in individual groups of VSMCs was analyzed with the BD LSRFortessa^TM flow cytometer (BD Biosciences, Franklin Lakes, NJ, USA).

The patch-clamp technique and whole-cell patch-clamp recordings were used to analyze the K_V and BK_Ca currents. Extracellular fluid (150 mmol/L choline chloride, 5 mmol/L KCl, 2 mmol/L CaCl₂, 1 mmol/L MgCl₂, 10 mmol/L HEPES, 1 mmol/L CdCl₂, and 10 mmol/L D-glucose; adjusted to pH 7.4 with KOH; 320 mOsm) was continuously perfused at a rate of 0.2 mL/min. The tip of the patch-clamp electrode was about 1–2 μm, and resistance was 6–10 MΩ. The internal fluid in the glass electrode was 120 mmol/L potassium gluconate, 20 mmol/L KCl, 2 mmol/L MgCl₂, 10 mmol/L EGTA, 10 mmol/L HEPES, 5 mmol/L Na₂ ATP, and 1 mmol/L CaCl₂; adjusted to pH 7.2 with KOH; osmotic pressure 320 mOsm). The clamping voltage was set to −80 mV, depolarization voltage was in steps of 10 mV from −70 mV to 70 mV (pulse duration of 400 ms), and whole-cell K_V current was recorded every 2s. The signals were processed and recorded with an EPC-10 amplifier (HEKA Elektronik, Lambrecht, Germany). To separate the BK_Ca and K_V currents from the total current, the BK_Ca channel inhibitors TEA (1 mmol/L) and CTX (100 nmol/L) were added to the extracellular fluid. The current densities and open probabilities of BK_Ca and K_V were calculated.

Cerebral vascular reactivity was measured as described previously^[25]. Briefly, basilar arteries (2 mm long) were carefully dissected free from the brain and mounted on two 40 μm stainless wires in the jaws of the Dual Wire Myograph System (Danish Myo Technology A/S, Hinnerup, Denmark) to record changes in isometric force. After the arteries were mounted and prepared, they were challenged with KCl (10–100 mmol/L) or cumulative concentrations (10⁻⁹–10⁻⁵ mol/L) of 5-hydroxytryptamine (5-HT). Cumulative concentration response curves were generated.

The results are expressed as mean ± standard error. GraphPad Prism 5.0 software (GraphPad Software Inc., La Jolla, CA, USA) was used for the statistical analysis. Protein expression and mRNA levels were compared using two-way analysis of variance (ANOVA) followed by the unpaired Student’s t-test. The isometric force measurement data were analyzed by two-way ANOVA. A P-value of < 0.05 was considered significant.

Microgravity simulated by HU resulted in a significantly lower soleus muscle mass (P < 0.001). Soleus muscle-to-body mass ratios decreased significantly (P < 0.001) in the HU and HU + MT rats, which confirmed the efficacy of simulated microgravity by HU. The data are summarized in Table 1.

Cytoplasmic, mitochondrial, and SR Ca²⁺ distribution and content in cerebral VSMCs are shown in Figure 1. After HU, cytoplasmic Ca²⁺ content significantly increased (P < 0.001) (Figure 1A, 1D) with a significant decrease of Ca²⁺ in mitochondria (Figure 1B, 1E) and the SR (Figure 1C, 1F) (P < 0.001) compared with CON rat cerebral VSMCs. The chronic treatment with mitoTEMPO restored cytoplasmic (P < 0.001), mitochondrial (P < 0.01), and SR (P < 0.001) Ca²⁺ distribution and content in HU + MT rat cerebral VSMCs.

Figure 1.  The effects of mitoTEMPO on cytoplasmic (A, D), mitochondrial (B, E), and sarcoplasmic reticulum (C, F) Ca²⁺ distribution in rat cerebral vascular smooth muscle cells. CON, control; HU, hindlimb unweighting; MT, mitoTEMPO; CON + MT, mitoTEMPO-treated control; HU + MT, mitoTEMPO-treated HU. Values are mean ± standard error. ^**P < 0.01 and ^***P < 0.001.

We analyzed mitochondrial fusion and fission to investigate the mechanism of cytoplasmic, mitochondrial, and SR Ca²⁺ redistribution (Figure 2). TEM (Figure 2A) showed more long and narrow mitochondria in the HU rat cerebral VSMCs, while more elliptical mitochondria were observed in the CON, CON + MT, and HU + MT rat cerebral VSMCs (mitochondria are marked by white arrows), indicating that HU enhanced mitochondrial fission and that treatment with mitoTEMPO attenuates mitochondrial fission. The MFN1/2 protein and mRNA levels (Figure 2B, 2F and Figure 2C, 2G) in HU rat cerebral VSMCs decreased significantly compared to those in the CON rats (P < 0.001). The DRP1/FIS1 protein and mRNA levels (Figure 2D, 2H and Figure 2E, 2I) were significantly higher in HU rat cerebral arteries than those in the CON rats (P < 0.01 for protein and P < 0.001 for mRNA). Chronic treatment with mitoTEMPO significantly upregulated the expression of MFN1/2 (P < 0.05 for MFN1 mRNA; P < 0.001 for MFN1/2 protein and MFN2 mRNA) but decreased the expression of FIS1/DRP1 (P < 0.01 for protein and P < 0.001 for mRNA) compared to the HU.

Figure 2.  Effects of mitoTEMPO on mitochondrial fission and fusion (A) and the protein and mRNA levels of mitofusion 1 (MFN1) (B, F) and mitofusion 2 (MFN2) (C, G), dynamin-related protein 1 (DRP1) (D, H), and fission protein 1 (FIS1) (E, I) in rat cerebral vascular smooth muscle cells. CON, control; HU, hindlimb unweighting; MT, mitoTEMPO; CON + MT, mitoTEMPO-treated control; HU + MT, mitoTEMPO-treated HU. Values are mean ± standard error. ^*P < 0.05, ^**P < 0.01, and ^***P < 0.001.

To investigate whether cytoplasmic Ca²⁺ redistribution was IP₃R-dependent, we determined the abundance of the IP₃R in HU rat cerebral arteries (Figure 3). IP₃R protein (P < 0.001) and mRNA (P < 0.001) levels increased significantly (Figure 3A and 3B) in HU rat cerebral arteries compared to the CON. Immunohistochemical staining revealed that the IP₃R was more positive in HU rat cerebral VSMCs than that in the CON (Figure 3C). Chronic treatment with mitoTEMPO partially restored the enhanced expression of the IP₃R after HU.

Figure 3.  The effects of mitoTEMPO on protein (A) and mRNA (B) levels of the inositol 1,4,5-trisphosphate receptor (IP₃R) and immunohistochemistry for IP₃R (C) in rat cerebral vascular smooth muscle cells. CON, control; HU, hindlimb unweighting; MT, mitoTEMPO; CON + MT, mitoTEMPO-treated control; HU + MT, mitoTEMPO-treated HU. Values are mean ± standard error. ^***P < 0.001.

To investigate whether cytoplasmic Ca²⁺ redistribution induced by mitochondrial oxidative stress was associated with changes in plasma membrane K⁺ channels, we analyzed total K⁺ current, current densities, and open probabilities (Po) of the K_V and BK_Ca channels (Figure 4). Total K⁺ current decreased (Figure 4A), whereas current densities and open probabilities of K_V (Figure 4B and 4C) and BK_Ca (Figure 4D and 4E) decreased and increased, respectively, compared to control rats, in HU rat cerebral VSMCs, which were restored by chronic treatment with mitoTEMPO.

Figure 4.  Effects of mitoTEMPO on total K⁺ current (A), current activation, and opening probabilities of voltage-gated potassium (K_V) channels (B, C), and Ca²⁺-activated K⁺ (BK_Ca) channels (D, E) in rat cerebral vascular smooth muscle cells. CON, control; HU, hindlimb unweighting; MT, mitoTEMPO; CON + MT, mitoTEMPO-treated control; HU + MT, mitoTEMPO-treated HU. Values are mean ± standard error. ^*P < 0.05, ^**P < 0.01, and ^***P < 0.001.

To investigate whether the changes in cytoplasmic Ca²⁺ were associated with cerebrovascular contraction, we studied the cerebrovascular contraction to vasoconstrictors (Figure 5). Cumulative increases in KCl and 5-HT concentrations induced concentration-dependent vasoconstriction in basilar arteries from the four groups. Three-week HU significantly enhanced the maximal contractile responses to KCl and 5-HT in rat basilar arteries (P < 0.05) compared to the CON, which was attenuated by the chronic mitoTEMPO treatment (P < 0.05).

Figure 5.  Effects of mitoTEMPO on vasoconstriction to cumulative KCl or 5-hydroxytryptamine (5-HT) in rat basilar arteries. CON, control; HU, hindlimb unweighting; MT, mitoTEMPO; CON + MT, mitoTEMPO-treated control; HU + MT, mitoTEMPO-treated HU. Values are mean ± standard error. ^*P < 0.05 for HU vs. CON, ^▲P < 0.05 for HU vs. HU + MT.

The major findings of the present study are: (1) mitochondrial oxidative stress augmented HU rat cerebrovascular contraction to vasoconstrictors by regulating Ca²⁺ distribution and content in VSMCs; (2) mitochondrial oxidative stress regulated Ca²⁺ homeostasis by controlling IP₃R abundance, Ca²⁺ storage, mitochondrial fusion/fission, and changes in plasma membrane ion channels (K_V and BK_Ca) in HU rat cerebral VSMCs.

Exposure to microgravity results in postflight cardiovascular deconditioning and orthostatic intolerance in astronauts. Structural and functional remodeling of the cardiovascular system have been implicated in this process. We and other authors have found increased myogenic tone, enhanced vasoconstriction, and attenuated endothelium-dependent relaxation in HU rat cerebral arteries^{[3, 4, 6, 25]}. We previously reported mitochondrial oxidative stress and ER stress in rat cerebral VSMCs^{[21-23, 26]}. However, whether and how mitochondrial oxidative stress regulates HU rat cerebrovascular reactivity remains unclear. The dynamic changes in cytoplasmic Ca²⁺ are a critical mechanism regulating vascular tone and contraction. During microgravity, large-conductance calcium-activated K⁺ (BK_Ca) and L-type voltage-dependent Ca²⁺ (Ca_L) channels were activated in VSMCs, which are associated with apoptosis^{[10, 11]}. The Ca_L currents densities increased significantly in 4-week HU rat cerebral VSMCs, and therefore induced significantly higher cytoplasmic Ca²⁺ and enhanced vasoconstriction^[27]. K_V channels are expressed in blood vessels and are key regulators of vascular tension. An increase in cytoplasmic Ca²⁺ concentration inactivates K_V channels and induces vascular contraction^[28]. The BK_Ca channels in the vasculature located close to the IP₃Rs in the SR and the IP₃Rs stimulate the opening of BK_Ca channels dependent on Ca²⁺ released from IP₃Rs as in VSMCs^[29]. In the current study, K_V and BK_Ca current densities and open probabilities, which are both important Ca²⁺-regulating plasma ion channels, decreased and increased respectively; however, these changes were restored by the mitochondrial-targeted antioxidant mitoTEMPO. This means that mitochondrial oxidative stress regulates cytoplasmic Ca²⁺ concentration through the activities of plasma ion channels, which play roles in Ca²⁺ homeostasis in VSMCs exposed to microgravity or simulated microgravity.

The processes influencing cytoplasmic Ca²⁺ are important regulators of vascular function under physiological and pathophysiological settings. The mitochondria and SR are the major Ca²⁺ stores in VSMCs, which regulate Ca²⁺ homeostasis in intracellular organelles by Ca²⁺ sequestering activities^[13]. The close connection between mitochondria and the SR plays an important role in mitochondrial dynamics, ATP production, lipid synthesis, and Ca²⁺ signaling^{[30, 31]}. Mitochondria regulate intracellular Ca²⁺ content through Ca²⁺ uptake from SR via the SR-mitochondrial membrane^[32]. The rate of mitochondrial Ca²⁺ uptake is affected by the IP₃R and MFN2^{[33, 34]}, which maintain the balance between extracellular and intracellular [Ca²⁺]^[13]. The IP₃R is a pivotal Ca²⁺ release channel located perinuclearly and peripherally in the SR^[35]. ER/SR stress facilitates Ca²⁺ release from the SR through the IP₃R^[14], which causes Ca²⁺ uptake by mitochondria. Increased mitochondrial Ca²⁺ influx leads to mitochondrial Ca²⁺ overload and opening of the mitochondrial permeability transition pore, eventually triggering cell death^{[36, 37]}. Overloading of Ca²⁺ within the mitochondrial matrix inhibits ATP production but promotes mitochondrial ROS production^{[38, 39]}. In the current study, increased IP₃R protein abundance and decreased SR Ca²⁺ in HU rat cerebral VSMCs suggested that IP₃R-dependent Ca²⁺ release from the SR was activated and increased cytoplasmic Ca²⁺ contributed to enhance vasoconstriction.

The elevated ROS levels in mitochondria and the SR send feedback for SR membrane Ca²⁺ release by stimulating the IP₃R^{[40, 41]}, and increased ROS production enhances the rate of mitochondrial Ca²⁺ uptake and has a destructive effect on mitochondria, leading to changes in the state of mitochondria from fusion to fission by inducing the expression of DRP1 and FIS1^{[19, 42]}. Mitochondrial fusion provides the proper environment for Ca²⁺ regulation and cellular metabolism, which is important for Ca²⁺ homeostasis in VSMCs^{[19, 43]}. MFN1 and MFN2 play critical roles maintaining the correct mitochondrial and the SR-mitochondria connection, respectively^{[33, 44]}. MFN2, which is located in the outer mitochondrial membrane and SR membrane, tethers the SR to mitochondria and mediates Ca²⁺ signal transfer between the two organelles together with the IP₃R^[45]. Once the mitochondrial dynamics have shifted from fusion to fission, cell death occurred caused by over-expression of FIS1, which depends on SR-mitochondria; Ca²⁺ signal transfer^[46] and DRP1-induced mitochondrial fragmentation^[47]. The Ca²⁺ redistribution leads to increased cytoplasmic Ca²⁺ and an enhanced contractile response to vasoconstrictors. Consistent with these findings, our present study showed that microgravity simulated by HU changed the morphological characteristics of the mitochondria from elliptical to long and narrow, along with upregulated expression of the mitochondrial fission proteins (DRP1 and FIS1) and decreased levels of the fusion proteins (MFN1 and MFN2), which were reversed by a mitochondrial-targeted antioxidant. This finding suggests that mitochondrial oxidative stress during simulated microgravity shifted the mitochondrial state from fusion to fission. Together with our previous findings that microgravity simulated by HU is associated with oxidative stress and mitochondrial oxidative stress in cerebral VSMCs^{[21, 22, 48]}, we infer that Ca²⁺ and ROS overloading in mitochondria during simulated microgravity leads to mitochondrial fission by regulating the expression of mitochondrial fission and fusion proteins. Therefore, mitochondrial oxidative stress induced by 3-weeks of HU stimulated Ca²⁺ release from the SR by upregulating IP₃R expression, impairing mitochondrial Ca²⁺ uptake, and undergoing mitochondrial fission, which resulted in the accumulation of Ca²⁺ in the cytoplasm and enhanced vasoconstriction.

Interestingly, our recent study reported that ER stress induces activation of the iNOS/NO-NF-κB/IκB pathway and plays a key role inducing endothelial inflammation and apoptosis^[49]. Whether SR stress affects calcium signaling in cerebral VSMCs during microgravity simulation needs further investigation. Although we clarified that changes in calcium homeostasis caused by mitochondrial oxidative stress contributed to enhance vasoconstriction during simulated microgravity, we did not investigate the effects of mitochondrial oxidative stress on SR function. Further studies are needed to investigate functional alterations and the underlying mechanism of mitochondria and the SR during microgravity.

In conclusion, mitochondrial oxidative stress induced by simulated microgravity increased cytoplasmic Ca²⁺ by impairing mitochondrial fusion/fission-dependent Ca²⁺ uptake, enhancing IP₃R-dependent SR Ca²⁺ release, and regulating the functions of plasma K_V and BK_Ca channels. A mitochondrial-targeted antioxidant that alleviated mitochondrial oxidative stress restored Ca²⁺ homeostasis and vascular contraction to the agonists. The current results demonstrate that mitochondrial oxidative stress contributes to vasoconstriction by regulating calcium homeostasis in rat cerebrovascular smooth muscle cells exposed to simulated microgravity.

Guarantor of integrity of the entire study: LIU Zi Fan. Study concept and design: ZHANG Ran and LI Xin. Experimental studies: LIU Zi Fan, WANG Hai Ming, JIANG Min, WANG Lin, XIE Man Jiang, LIN Le Jian, ZHAO Yun Zhang, SHAO Jun Jie, and ZHOU Jing Jing. Data analysis: LIU Zi Fan, WANG Hai Ming, and ZHANG Ran. Manuscript preparation: LIU Zi Fan and WANG Hai Ming. Manuscript editing: LIU Zi Fan and WANG Hai Ming. Manuscript review: ZHANG Ran and LI Xin. All authors have read and approved the final version of the manuscript.

The authors declare no competing financial interests.