Mitochondria control functional Ca_V1.2 expression in smooth muscle cells of cerebral arteries

Abstract

Rationale

Physiological functions of mitochondria in contractile arterial myocytes are poorly understood. Mitochondria can uptake calcium (Ca²⁺), but intracellular Ca²⁺ signals that regulate mitochondrial Ca²⁺ concentration ([Ca²⁺]_mito) and physiological functions of changes in [Ca²⁺]_mito in arterial myocytes are unclear.

Objective

Identify Ca²⁺ signals that regulate [Ca²⁺]_mito, examine the significance of changes in [Ca²⁺]_mito, and test the hypothesis that [Ca²⁺]_mito controls functional ion channel transcription in myocytes of resistance-size cerebral arteries.

Methods and Results

Endothelin-1 (ET-1) activated Ca²⁺ waves and elevated global Ca²⁺ concentration ([Ca²⁺]_i) via inositol 1,4,5-trisphosphate receptor (IP₃R) activation. IP₃R-mediated sarcoplasmic reticulum (SR) Ca²⁺ release increased [Ca²⁺]_mito and induced mitochondrial depolarization, which stimulated mitochondrial reactive oxygen species (mitoROS) generation that elevated cytosolic ROS. In contrast, a global [Ca²⁺]_i elevation did not alter [Ca²⁺]_mito, mitochondrial potential, or mitoROS generation. ET-1 stimulated nuclear translocation of nuclear factor kappa B (NF-κB) p50 subunit and ET-1-induced IP₃R-mediated mitoROS elevated NF-κB-dependent transcriptional activity. ET-1 elevated voltage-dependent Ca²⁺ (Ca_V1.2) channel expression, leading to an increase in both pressure (myogenic tone)- and depolarization-induced vasoconstriction. Baseline Ca_V1.2 expression and the ET-1-induced elevation in Ca_V1.2 expression were both reduced by IP₃R inhibition, mitochondrial electron transport chain block, antioxidant treatment, and NF-κB subunit knockdown, leading to vasodilation.

Conclusions

IP₃R-mediated SR Ca²⁺ release elevates [Ca²⁺]_mito, which induces mitoROS generation. MitoROS activate NF-κB, which stimulates Ca_V1.2 channel transcription. Thus, mitochondria sense IP₃R-mediated SR Ca²⁺ release to control NF-κB-dependent Ca_V1.2 channel expression in arterial myocytes, thereby modulating arterial contractility.

Introduction

Intracellular calcium (Ca²⁺) signals modulate a wide variety of physiological functions in arterial myocytes, including contractility and gene expression ¹. The spatial location of mitochondria nearby Ca²⁺ channels can expose these organelles to domains of elevated intracellular Ca²⁺ concentration ([Ca²⁺]_i), leading to mitochondrial Ca²⁺ uptake ^2,3. Mitochondrial Ca²⁺ sequestration can reduce the elevation and diffusion of cytosolic Ca²⁺ signals and feedback to alter the activity of Ca²⁺ channels that generate the signal ^2–4. Mitochondria can also generate signaling molecules, including reactive oxygen species (ROS), which modulate local and global intracellular Ca²⁺ signals, leading to changes in arterial contractility ^2,3,5. Although it is recognized that mitochondria can sequester Ca²⁺, intracellular Ca²⁺ signals that regulate mitochondrial Ca²⁺ concentration ([Ca²⁺]_mito) and physiological functions of changes in [Ca²⁺]_mito in arterial myocytes are poorly understood.

Arterial myocytes generate several distinct local and global Ca²⁺ signals ^1,6. An elevation in global [Ca²⁺]_i occurs in response to plasma membrane Ca²⁺ influx and sarcoplasmic reticulum (SR) Ca²⁺ release and directly regulates vascular contractility ¹. Voltage-dependent Ca²⁺ (Ca_V1.2) channels are the major contributor to global [Ca²⁺]_i in arterial myocytes and are essential for diameter regulation in resistance-size arteries that regulate blood pressure and regional blood flow ^1,7,8. Ca²⁺ sparklets are subsarcolemmal Ca²⁺ signals generated by Ca²⁺ influx through Ca_V1.2 channels ⁶. Ca²⁺ sparklets contribute directly to global [Ca²⁺]_i, thereby regulating contractility ⁶. Ca²⁺ sparks are local Ca²⁺ transients generated by SR ryanodine-sensitive Ca²⁺ release (RyR) channels ¹. Ca²⁺ sparks activate large-conductance Ca²⁺-activated potassium channels, leading to membrane hyperpolarization and vasodilation ¹. Ca²⁺ waves are propagating Ca²⁺ transients that can occur due to the activation of SR inositol 1,4,5-trisphosphate receptor (IP₃R) channels and RyR channels ⁹. Physiological functions of Ca²⁺ waves are less clear with studies reporting that these Ca²⁺ signals either directly stimulate contraction or do not alter contractility ^10–12.

Here, we investigated Ca²⁺ signaling mechanisms that regulate [Ca²⁺]_mito in myocytes of resistance-size cerebral arteries and tested the hypothesis that changes in [Ca²⁺]_mito control the expression of Ca_V1.2 channels, an ion channel whose transcriptional regulation in arterial myocytes is unclear. Our data indicate that mitochondria sense IP₃R-mediated SR Ca²⁺ release to control mitochondrial ROS (mitoROS) generation, nuclear factor kappa B (NF-κB) activity, and functional Ca_V1.2 expression in arterial myocytes.

Methods

Tissue preparation

Animal protocols were reviewed and approved by the Animal Care and Use Committee of the University of Tennessee Health Science Center. All experiments were performed using Sprague-Dawley rat (~250 g) resistance-size (~50–200 µm diameter) cerebral arteries or myocytes isolated from these arteries as previously described ⁷.

Laser-scanning confocal Ca²⁺ imaging

Intracellular Ca²⁺ signals were imaged in myocytes of cerebral arteries using fluo-4 AM and a Noran Oz laser-scanning confocal microscope, as previously described ¹³.

Imaging of genetically-encoded fluorescent indicators

Vectors encoding 2mt8CG2, mt-cpYFP, or HyPer-CYTO were inserted into myocytes of intact cerebral arteries using reverse permeabilization. Expressed fluorescent indicators were imaged using a Zeiss LSM5 confocal microscope. Indicator localization was determined in myocytes isolated from arteries through colocalization with MitoTracker Orange CMTMRos using weighted colocalization.

Tetramethylrhodamine, Methyl Ester (TMRM) Imaging

Isolated arterial myocytes were loaded with TMRM and excited with 535 nm light. Background corrected TMRM fluorescence was collected at 610 nm using a Dage MTI iCCD camera.

NF-κB immunofluorescence

Paraformaldehyde-fixed arteries were incubated with antibodies against NF-κB p50 subunit (p50), followed by Cy3-conjugated secondary antibody. YOYO-1 was used to counterstain nuclei. Images were obtained using a Zeiss LSM5 confocal microscope. Weighted colocalization was used to quantify p50 nuclear localization.

NF-κB-dependent luciferase reporter gene activity

Vectors that express firefly luciferase under the control of an NF-κB promoter (NF-κB-p-Luc) were inserted into myocytes of intact arteries using reverse permeabilization. Promoter-deficient pGL3 Basic control vector containing a luciferase reporter gene was used as a control. Luciferase activity was quantified using a luminometer.

NF-κB p105 subunit (p105) knockdown

Two siRNAs directed against p105 (p105siRNA1 and p105siRNA2) were used, with scrambled siRNA (p105scrm) as control. siRNA was introduced into myocytes of intact arteries using reverse permeabilization.

Pressurized artery diameter measurements

Diameter was measured in endothelial-denuded arteries using edge-detection myography. Myogenic tone (%) was calculated as 100×(1-active diameter/passive diameter).

Statistical analysis

OriginLab and GraphPad InStat software were used for statistical analyses. Values are expressed as mean±SEM. Student’s t-test was used for comparing paired and unpaired data from two populations, and ANOVA with Student–Newman–Keuls post-hoc test used for multiple group comparisons. P<0.05 was considered significant. Power analysis was performed where P>0.05 to verify that sample size was sufficient to give a power value >0.8.

Detailed Methods is provided as Supplemental Material.

Results

ET-1 modifies local and global Ca²⁺ signals in cerebral artery myocytes

The regulation of local and global Ca²⁺ signals by endothelin-1 (ET-1), a phospholipase C (PLC)-coupled receptor agonist and vasoconstrictor, was studied in myocytes of endothelium-denuded cerebral artery segments. ET-1 reduced mean Ca²⁺ spark frequency to ~32% of control, elevated mean Ca²⁺ wave frequency to ~135% of control, and increased mean global [Ca²⁺]_i to ~127% of control (Fig. 1A–D). Xestospongin C (XeC), an IP₃R inhibitor, blocked ET-1-induced Ca²⁺ wave activation and reduced ET-1-induced global [Ca²⁺]_i elevation, but did not alter ET-1-induced Ca²⁺ spark inhibition (Fig. 1D).

Figure 1.

ET-1 regulates local and global Ca²⁺ signals in arterial myocytes. A, Confocal images illustrating average fluo-4 fluorescence in myocytes in the same artery in control and ET-1 (10 nmol/L). White boxes illustrate locations where sparks occurred during 10 seconds of imaging. Colored boxes illustrate locations from where normalized fluorescence (F/F₀) over time traces shown in C were determined. B, Two representative Ca²⁺ sparks that occurred at locations labeled in A for each condition. C, F/F₀ over time traces illustrate ET-1-induced Ca²⁺ wave activation. D, Mean data (n=7 for each condition). P<0.05: * vs control; # vs ET-1.

ET-1-induced IP₃R-mediated SR Ca²⁺ release elevates [Ca²⁺]_mito and depolarizes mitochondria in cerebral artery myocytes

Regulation of [Ca²⁺]_mito by ET-1-induced Ca²⁺ signals was measured in myocytes of intact arteries using 2mt8CG2 (K_d for Ca²⁺ ~5 µmol/L), a genetically-encoded, mitochondria-targeted, fluorescent Ca²⁺ indicator ^14,15. Confocal imaging revealed punctate 2mt8CG2 fluorescence that exhibited ~95% pixel colocalization with Mito Tracker Orange, indicating mitochondrial localization (Fig. 2A). In myocytes of intact arteries, ionomycin, a Ca²⁺ ionophore, and CCCP, a protonophore that disrupts mitochondrial potential, caused fluorescence changes in 2mt8CG2 consistent with the range of this indicator ¹⁵ (Fig. 2C). ET-1 increased mean 2mt8CG2 fluorescence to ~140% of control, and this elevation was blocked by thapsigargin, a SR Ca²⁺ ATPase inhibitor, XeC, or Ru360, a mitochondrial Ca²⁺ uniporter blocker (Fig. 2B,C). Thapsigargin, XeC, or Ru360 alone, or membrane depolarization (60 mmol/L K⁺), which elevates global [Ca²⁺]_i, did not alter 2mt8CG2 fluorescence (Fig. 2C).

Figure 2.

ET-1-induced IP₃R-mediated SR Ca²⁺ release elevates [Ca²⁺]_mito in arterial myocytes. A, Confocal images illustrate colocalization of punctate 2mt8CG2 fluorescence with MitoTracker Orange in a myocyte. Colocalization quantified using weighted colocalization was 94.8±0.8% (n=10, p<0.001). B, Confocal images of 2mt8CG2 fluorescence in myocytes in the same area of a cerebral artery in control and ET-1. Scale bars=10 µm. C, Mean data for ionomycin (10 µmol/L, n=7), ET-1 (n=8), thapsigargin (100 nmol/L, n=5), ET-1+thapsigargin (n=6), XeC (20 µmol/L, n=5), ET-1+XeC (n=6), Ru360 (10 µmol/L, n=4), ET-1+Ru360 (n=4), CCCP (10 µmol/L, n=4), and 60 mmol/L K⁺ (n=6). ET-1 concentration was 100 nmol/L in all experiments. Thapsigargin was applied 15 minutes prior to ET-1, a time course sufficient to deplete SR Ca²⁺ load ²¹. P<0.05: * vs control; # vs ET-1.

Mitochondria potential regulation by ET-1 was studied in isolated myocytes using TMRM. ET-1 caused reproducible concentration-dependent mitochondrial depolarization with an EC₅₀ of 29 nmol/L (Fig. 3A–C). Thapsigargin and XeC abolished ET-1-induced mitochondrial depolarization, but did not alter TMRM fluorescence when applied alone (Fig. 3A,C). CCCP depolarized mitochondria, whereas 60 mmol/L K⁺ did not alter mitochondrial potential (Fig. 3A,C). These data indicate that ET-1-induced IP₃R-mediated SR Ca²⁺ release elevates [Ca²⁺]_mito and depolarizes mitochondria in arterial myocytes. In contrast, a global [Ca²⁺]_i elevation does not alter [Ca²⁺]_mito or mitochondrial potential.

Figure 3.

ET-1-induced IP₃R-mediated SR Ca²⁺ release depolarizes mitochondria in arterial myocytes. A, ET-1 caused reproducible mitochondrial depolarization which was blocked by XeC (20 µmol/L). B, Concentration-dependent mitochondrial depolarization by ET-1 (n=4–7). C, Mean data for ET-1 (n=37), second ET-1 application (n=12), thapsigargin (100 nmol/L, n=18), ET-1+thapsigargin (n=18), XeC (20 µmol/L, n=12), ET-1+XeC (n=12), 60 mmol/L K⁺ (n=10), and CCCP (10 µmol/L, n= 65). ET-1 concentration was 30 nmol/L in all experiments. P<0.05: * vs control; # vs ET-1.

ET-1-induced IP₃R-mediated SR Ca²⁺ release stimulates mitoROS generation

We tested the hypothesis that a [Ca²⁺]_mito elevation may alter mitoROS generation. MitoROS was measured using mt-cpYFP, a genetically-encoded mitochondria-targeted fluorescent O₂⁻ indicator ¹⁶. mt-cpYFP exhibited ~95% pixel colocalization with MitoTracker Orange, indicating mitochondrial localization (Fig. 4A). In myocytes of intact arteries, ET-1 elevated mt-cpYFP fluorescence to ~199% of control, and this was blocked by XeC, Ru360, rotenone, a mitochondrial electron transport chain (ETC) complex I inhibitor, and micromolar CCCP (Fig. 4B, Supplemental Fig. I). Rotenone and micromolar CCCP reduced mt-cpYFP fluorescence when applied alone (Fig. 4B). In contrast, Ru360, XeC, and 60 mmol/L K⁺ did not alter mt-cpYFP fluorescence (Fig. 4B). 1 nmol/L CCCP, which induces a small mitochondrial depolarization in cerebral artery myocytes ⁵, elevated mt-cpYFP fluorescence (Fig. 4B).

Figure 4.

ET-1-induced IP₃R-mediated SR Ca²⁺ release elevates mitoROS generation, leading to an increase in cytosolic ROS in arterial myocytes. A, Confocal images illustrating mt-cpYFP and MitoTracker Orange fluorescence in the same myocyte. Colocalization quantified using weighted colocalization was 94.7±1.2% (n=10, p<0.001). Scale bar=10 µm. B, Mean mt-cpYFP fluorescence changes in myocytes of intact arteries. ET-1, XeC (20 µmol/L), ET-1+XeC, Ru360 (10 µmol/L), ET-1+Ru360, rotenone (10 µmol/L), ET-1+rotenone, CCCP (10 µmol/L), ET-1+CCCP, CCCP (1 nmol/L), and 60 mmol/L K⁺. C, Mean HyPer-CYTO fluorescence. H₂O₂ (100 µmol/L), ET-1 (endothelium-intact), ET-1 (endothelium-denuded), rotenone (10 µmol/L), ET-1+rotenone, MnTMPyP (10 µmol/L), and ET-1+MnTMPyP. ET-1 concentration was 100 nmol/L in all experiments. n=5 for each condition. P<0.05: * vs control; # vs ET-1.

Cytosolic ROS was measured using HyPer-CYTO, a genetically-encoded fluorescent cytosolic hydrogen peroxide (H₂O₂) indicator (Supplemental Fig. IIA) ¹⁷. Exogenous H₂O₂ elevated HyPer-CYTO fluorescence (Fig. 4C, Supplemental Fig. IIB). ET-1 similarly increased HyPer-CYTO fluorescence in myocytes of endothelium-intact and -denuded arteries (Fig. 4C, Supplemental Fig. IIC). Rotenone and MnTMPyP, a superoxide dismutase and catalase mimetic, reduced HyPer-CYTO fluorescence when applied alone, and blocked ET-1-induced elevations in HyPer-CYTO fluorescence (Fig. 4C).

Similar results were also obtained using CM-H₂DCFDA, an inorganic fluorescent ROS indicator (Supplemental Fig. IIIA–D). Oxypurinol, a xanthine oxidase inhibitor, 17-octadecanoic acid (17-ODYA), a cytochrome P450 blocker, or gp91ds-tat, a NADPH oxidase inhibitor, did not alter baseline dichlorofluorescin (DCF) fluorescence or ET-1-induced DCF fluorescence elevations (Supplemental Fig. IIIC,D). gp91ds-tat inhibited DCF fluorescence elevations induced by tumor necrosis factor-alpha (TNF-α), which stimulates NADPH oxidase-derived ROS in cerebral artery myocytes ¹⁸ (Supplemental Fig. IIID). Scrambled gp91-tat did not alter baseline DCF fluorescence, or ET-1- or TNF-α-induced DCF fluorescence elevations (Supplemental Fig. IIID).

These data indicate that mitochondria generate ROS in the absence of ET-1, and that ET-1-induced IP₃R-mediated SR Ca²⁺ release elevates mitoROS generation, leading to an increase in cytosolic ROS in cerebral artery myocytes.

ET-1-induced IP₃R-mediated mitoROS stimulate NF-κB

To examine physiological functions of an IP₃R-mediated mitoROS elevation, we tested the hypothesis that mitochondria regulate the expression of genes that modulate myocyte contractility. Therefore, immunofluorescence was performed to study cellular localization of the p50 subunit of NF-κB, a ROS-sensitive transcription factor ¹⁹, in arterial myocytes. ET-1 increased nuclear translocation of p50, measured as an increase in mean pixel colocalization with YOYO-1 from ~12 to 22% (Fig. 5A–C).

Figure 5.

ET-1 stimulates p50 nuclear translocation and NF-κB-dependent transcription through SR Ca²⁺ release and mitoROS elevation in arterial myocytes. A, Immunofluorescence images of myocytes in arteries illustrating YOYO-1 (nuclear stain, green), p50 (red), overlay, and DIC (lumenally-inserted rectangular glass cannula can be seen). B, Enlarged images indicated by boxes in A illustrate ET-1-induced elevation in p50 and YOYO-1 pixel colocalization (purple). Scale bars=20 µm. C, Mean data (n=10 for each). D, Average NF-κB-p-Luc luciferase activity with TNF-α (100 ng/ml, n=4), ET-1 (n= 5), thapsigargin (100 nmol/L, n=5), ET-1+thapsigargin (n=5), XeC (20 µmol/L, n=5), ET-1+XeC (n=5), rotenone (1 µmol/L, n=5), ET-1+rotenone (n=5), MnTMPyP (10 µmol/L, n=4), ET-1+MnTMPyP (n=4), H₂O₂ (100 µmol/L, n=4), and H₂O₂+rotenone (n=4). ET-1 concentration was 100 nmol/L in all experiments. P<0.05: * vs control; # vs ET-1; § vs rotenone or MnTMPyP.

To investigate the regulation of NF-κB-dependent transcriptional activity, assays were performed using vectors that express firefly luciferase under the control of an NF-κB promoter. As a control, regulation of luciferase activity by TNF-α, which activates NF-κB was measured. TNF-α increased mean luciferase activity to ~547% of control (Fig. 5D). ET-1 increased luciferase activity to ~372% of control, and this was blocked by thapsigargin and XeC (Fig. 5D). Rotenone and MnTMPyP reduced the ET-1-induced elevation in luciferase activity (Fig. 5D). Rotenone also blocked the ET-1-induced ROS elevation over the 24-hour period used for these transcriptional measurements (Supplemental Fig. IV). Exogenous H₂O₂ elevated luciferase activity, and this effect was not altered by rotenone, indicating that rotenone did not cause general inhibition of luciferase expression (Fig. 5D). When applied alone, thapsigargin, XeC, rotenone, and MnTMPyP reduced luciferase activity by ~16, 22, 23, and 20%, respectively (Fig. 5D). TNF-α, ET-1, thapsigargin, XeC, rotenone, MnTMPyP, and H₂O₂ did not alter luciferase activity in arteries in which a promoter-deficient control vector was inserted (Supplemental Fig. V). These data indicate that IP₃R-mediated SR Ca²⁺ release stimulates NF-κB primarily by elevating mitoROS generation, but also via a secondary mitochondria- and ROS-independent pathway in arterial myocytes.

IP₃R-mediated SR Ca²⁺ release stimulates NF-κB and Ca_V1.2 expression

We sought to examine the functional significance of IP₃R-mediated, mitochondrial-dependent NF-κB activation. Ca_V1.2 channels are the principal Ca²⁺ influx pathway in myocytes of resistance-size arteries and are essential for contractility regulation by a wide variety of stimuli, including intravascular pressure and membrane potential ⁸. However, mechanisms that regulate Ca_V1.2 gene (CACNA1C) transcription in arterial myocytes are unclear. Real-time PCR data indicated that ET-1 (6 hrs) increased Ca_V1.2 channel mRNA expression to ~244% of control in cerebral arteries (Fig. 6A, Supplemental Fig VI). Western blotting indicated that ET-1 (24 hrs) increased Ca_V1.2 channel (190 and 240 kD bands) protein to ~141% of control (Fig. 6B,C). XeC, rotenone, and MnTMPyP alone reduced basal Ca_V1.2 expression to ~87, 89 and 80% of control, respectively (Fig. 6B,C). XeC blocked, and rotenone and MnTMPyP reduced the ET-1-induced elevation in Ca_V1.2 expression (Fig. 6B,C). Exogenous H₂O₂ increased Ca_V1.2 expression, and this elevation was not altered by rotenone, indicating that rotenone did not induce non-specific inhibition of Ca_V1.2 expression (Fig. 6C). These data indicate that ET-1-induced IP₃R-mediated SR Ca²⁺ release and mitoROS elevation stimulate Ca_V1.2 expression in cerebral arteries.

Figure 6.

ET-1-induced IP₃R-mediated SR Ca²⁺ release and mitoROS elevation stimulate Ca_V1.2 expression in cerebral arteries. A, ET-1 elevated mean Ca_V1.2 mRNA (n=7). B, Western blot indicating that XeC (20 µmol/L) blocked ET-1-induced elevation in Ca_V1.2 expression. C, Mean data for ET-1 (n=12), XeC (20 µmol/L, n=7), ET-1+XeC (n=7), rotenone (1µmol/L, n=6), ET-1+rotenone (n=6), MnTMPyP (10 µmol/L, n=5), ET-1+MnTMPyP (n=5), H₂O₂ (100 µmol/L, n=5), and H₂O₂+rotenone (n=5). ET-1 concentration was 10 nmol/L in all experiments. P<0.05: * vs untreated; # vs ET-1; § vs rotenone or MnTMPyP.

Two different siRNAs were used to knockdown expression of p105, the p50 precursor. p105siRNA1, p105siRNA2, and a combination of both siRNAs (p105siRNAs) reduced mean p105 protein to ~73, 80, and 57%, respectively, of that in arteries treated with scrambled siRNA (p105scrm) (Supplemental Fig. VIIA,B). p105siRNAs also reduced p50 expression to ~53% of p105scrm (Supplemental Fig. VIIIC,D). p105siRNAs reduced Ca_V1.2 expression to ~64% of p105scrm (Fig. 7A,B). In p105scrm-treated arteries, ET-1 increased Ca_V1.2 protein to ~136% of p105scrm (Fig. 7A,B). p105siRNAs reduced the ET-1-induced elevation in Ca_V1.2 expression by ~71% (Fig. 7A,B). These data indicate that NF-κB controls basal Ca_V1.2 expression and that IP₃R-mediated SR Ca²⁺ release stimulates Ca_V1.2 expression via mitochondria-dependent NF-κB activation.

Figure 7.

NF-κB controls basal and ET-1-induced elevation in functional Ca_V1.2 expression in cerebral arteries. A, Western blot illustrating effects of p105siRNAs (10 µg/ml each) on basal and ET-1-induced Ca_V1.2 expression. B, Mean data. n=5 for each. C, Representative traces illustrating diameter in arteries pressurized to 60 mm Hg. Nimodipine (1 µmol/L) fully dilated p105scrm- and p105siRNAs-treated arteries. D, Mean myogenic tone at 60 mm Hg, in 60 mmol/L K⁺ at 60 mm Hg, and nimodipine at 60 mm Hg. n=5–6 for each. ET-1 concentration was 10 nmol/L in all experiments. P<0.05: * vs p105scrm; # vs p105siRNAs.

The p105 gene promoter contains an NF-κB-binding sequence, and p50 activation can elevate p105 expression ²⁰. Therefore, mechanisms that regulate NF-κB subunit expression in cerebral arteries were investigated. XeC reduced basal p105 and p50 expression, whereas rotenone or MnTMPyP did not alter basal p105 or p50 expression (Supplemental Fig. VIIIA,B). ET-1 elevated p105 and p50 expression and XeC blocked this effect (Supplemental Fig. VIIIA,B). In contrast, rotenone or MnTMPyP had no effect on the ET-1-induced elevation in p105 and p50 subunit expression (Supplemental Fig. VIIIB). Exogenous H₂O₂ applied alone or in the presence of rotenone did not change p105 and p50 expression (Supplemental Fig. VIIIB). p105siRNAs did not alter the ET-1-induced relative increase in p105 and p50 expression (Supplemental Fig. VIIIC,D). These data indicate that IP₃R-mediated SR Ca²⁺ release activates p105/p50 expression via a mitochondria-, ROS-, and NF-κB-independent pathway. Data also indicate that rotenone, MnTMPyP, and p105siRNAs do not cause general inhibition of gene transcription.

NF-κB stimulates functional Ca_V1.2 expression in cerebral artery myocytes

NF-κB regulation of functional Ca_V1.2 expression was examined in pressurized cerebral arteries. Arteries were treated with either p105scrm or p105siRNAs and then exposed to either no further treatment or to ET-1 for 24 hours. At 60 mm Hg, control arteries (p105scrm) developed ~21% myogenic tone and membrane depolarization with 60 mmol/L K⁺ elevated tone to ~45%. p105 knockdown (p105siRNAs) reduced myogenic tone and depolarization-induced vasoconstriction (Fig. 7C,D). A 24 hour exposure to ET-1 elevated mean myogenic tone and depolarization-induced tone in control arteries (Fig. 7C,D). p105 knockdown attenuated the ET-1-induced elevation in myogenic tone and depolarization-induced vasoconstriction (Fig. 7C,D). Nimodipine fully dilated arteries regardless of treatment, indicating that vasoconstriction occurred due to Ca_V1.2 channel activation (Fig. 7C,D). These data indicate that NF-κB is essential for functional basal Ca_V1.2 expression in cerebral artery myocytes and that ET-1-induced NF-κB activation elevates Ca_V1.2-dependent vasoconstriction.

Discussion

Here, we investigated physiological signaling mechanisms that regulate [Ca²⁺]_mito and functional consequences of changes in [Ca²⁺]_mito in myocytes of resistance-size cerebral arteries. A schematic diagram summarizing major findings of this study is provided in Figure 8. Our data indicate that ET-1-induced IP₃R-mediated SR Ca²⁺ release increases Ca²⁺ wave frequency, elevates [Ca²⁺]_mito and depolarizes mitochondria, leading to an increase in mitoROS generation. ET-1-induced IP₃R-mediated mitoROS elevate cytosolic ROS that increase NF-κB nuclear translocation and transcriptional activity. ET-1-induced NF-κB activation leads to an increase in Ca_V1.2 transcription and Ca_V1.2-dependent pressure- and depolarization-induced vasoconstriction. IP₃R-mediated SR Ca²⁺ release, mitoROS, and NF-κB control basal Ca_V1.2 expression and NF-κB knockdown reduces Ca_V1.2 expression, leading to vasodilation. Collectively, these data indicate that mitochondria sense IP₃R-mediated SR Ca²⁺ release to control functional Ca_V1.2 channel transcription in arterial myocytes, thereby regulating arterial contractility.

Figure 8.

Proposed signaling pathways that control basal and ET-1-induced Ca_V1.2 expression in cerebral artery myocytes. Δψ_m indicates mitochondrial potential.

PLC-coupled receptor agonists elevate IP₃, which stimulates IP₃Rs, and diacylglycerol, which activates protein kinase C (PKC). Vasoconstrictor-induced IP₃R activation elevates [Ca²⁺]_i in arterial myocytes via two distinct mechanisms: 1) SR Ca²⁺ release, which produces Ca²⁺ waves ^9,13, and 2) SR Ca²⁺ release-independent plasma membrane TRPC3 channel activation, which leads to depolarization-induced Ca_V1.2 channel activation and a global [Ca²⁺]_i elevation ^13,21,22. In contrast, vasoconstrictor-activated PKC inhibits Ca²⁺ sparks, leading to membrane depolarization, Ca_V1.2 channel activation, and a global [Ca²⁺]_i elevation ⁹. The regulation of arterial myocyte [Ca²⁺]_mito by these different intracellular Ca²⁺ signals was unclear. Electron microscopy indicates that mitochondria can be located in close proximity (~20 nm) to the SR membrane in many cell types, including cultured arterial myocytes ^23,24. Such localization could place mitochondria within the vicinity of SR Ca²⁺ release channels that generate local micromolar Ca²⁺ transients necessary for mitochondrial Ca²⁺ uptake via the uniporter ²⁵. Inorganic Ca²⁺ indicators, which cannot be targeted specifically to organelles, have been used to measure [Ca²⁺]_mito in cultured vascular myocytes. In these studies, SR Ca²⁺ release elicited a [Ca²⁺]_mito elevation ^24,26. Ca²⁺ influx via the plasma membrane Na⁺/Ca²⁺ exchanger also elevated [Ca²⁺]_mito in cultured vascular myocytes ²⁷. In non-cultured colonic myocytes, uncaging IP₃ leads to mitochondrial Ca²⁺ uptake that feeds back to regulate IP₃R activity ²⁸. To our knowledge, our study is the first to use a genetically-encoded mitochondria-targeted fluorescent indicator to measure [Ca²⁺]_mito in contractile arterial myocytes. Our data indicate that in resting arterial myocytes, mitochondria contain Ca²⁺ and generate low levels of ROS through mechanisms that are independent of IP₃R-mediated SR Ca²⁺ release. ET-1-induced IP₃R-mediated SR Ca²⁺ release stimulated Ca²⁺ waves and elevated [Ca²⁺]_mito, leading to mitochondrial depolarization and mitoROS generation. In contrast, a global [Ca²⁺]_i elevation did not alter [Ca²⁺]_mito, mitochondrial potential or mitoROS generation. These findings also raise the possibility that Ca²⁺ waves specifically regulate [Ca²⁺]_mito in arterial myocytes. Given that Ca²⁺ waves are propagating Ca²⁺ signals, [Ca²⁺]_mito may also oscillate. Here, [Ca²⁺]_mito was simultaneously measured within multiple myocytes in the arterial wall. Asynchronous Ca²⁺ oscillations within individual mitochondria would have been averaged out by the imaging protocol. Future studies should be designed to examine spatial and temporal relationships between Ca²⁺ waves and mitochondrial Ca²⁺ signals within individual mitochondria in arterial myocytes.

The mechanism by which mitochondrial Ca²⁺ uptake induces mitochondrial depolarization and elevates ETC-generated mitoROS was not determined, but several possibilities exist. Mitochondrial depolarization has been demonstrated to increase or decrease ROS in different cell types, including vascular myocytes ^5,16,17,29. Here, a small mitochondrial depolarization increased mitoROS generation, whereas a large mitochondrial depolarization inhibited mitoROS production, consistent with an earlier report in cerebral artery myocytes ⁵. A [Ca²⁺]_mito elevation may stimulate mitoROS generation through multiple mechanisms, including stimulation of the tricarboxylic acid cycle ²⁹. In addition, a [Ca²⁺]_mito elevation and mitochondrial depolarization can open the mitochondrial permeability transition pore (mPTP) ^29,30, which in turn can enhance depolarization ²⁹. mPTP opening elevates mitoROS generation via several pathways, including dissipation of chemical gradients across the mitochondrial membrane and diversion of electrons in the ETC to ROS generation ¹⁶. Therefore, a [Ca²⁺]_mito elevation and mitochondrial depolarization may stimulate mitoROS generation by opening mPTP. MitoROS production is also regulated by multiple additional factors, including redox status of respiratory substrates, and proton pumping by ETC complexes ³¹. Future investigations should examine the mechanisms by which a vasoconstrictor-induced [Ca²⁺]_mito elevation and mitochondrial depolarization stimulate mitoROS generation.

Regulation and physiological functions of NF-κB in contractile arterial myocytes are poorly understood. Data here indicate that ET-1-induced NF-κB activation occurs primarily via IP₃R-mediated mitoROS generation in cerebral artery myocytes. MitoROS activate NF-κB in several cell types, including cultured vascular myocytes ^19,32. Alternate ROS-mediated mechanisms can also activate NF-κB in arterial myocytes. For example, following balloon injury NAD(P)H oxidase-derived ROS stimulate NF-κB in arterial myocytes ³³. The mechanisms by which ROS activate NF-κB are unclear, with reports suggesting that ROS stimulate IκB kinase (IKK), leading to phosphorylation and proteasomal degradation of I kappa B (IκB) ³². Other studies indicate that ROS activate kinases other than IKK to phosphorylate IκB ³⁴. Additional studies will be necessary to identify the specific ROS involved and the mechanisms by which ROS activate NF-κB in contractile arterial myocytes. Data here indicate that IP₃R-mediated SR Ca²⁺ release also activates NF-κB via a secondary mitoROS-independent pathway. Supporting bi-modal activation, redox-dependent and -independent mechanisms activate NF-κB in U937 cells ³⁵. Redox-independent NF-κB activation mechanisms may be mediated by calcineurin, PI3K/Akt, and/or protein kinase C, as demonstrated in neurons ³⁶.

Ca_V1.2 channels are the principal functional Ca²⁺ influx pathway in myocytes of resistance-size arteries ^1,7,8. Our data indicate that basal Ca_V1.2 expression is controlled through both an IP₃R-mediated mitochondria-independent pathway and an IP₃R-independent mitoROS pathway acting through NF-κB. ET-1-induced IP₃R activation stimulates an NF-κB-dependent elevation in Ca_V1.2 expression primarily via a mitoROS-dependent pathway and via a secondary mitochondria-independent pathway. In the absence of PLC-coupled receptor ligands, intracellular IP₃ concentration ([IP₃]_i) and thus, IP₃R activity should be low. However, in the intact artery preparation studied here endothelial cell release of receptor ligands, including ET-1, may generate low levels of [IP₃]_i in myocytes. In our experiments, the degree of p105 knockdown and the reduction in basal Ca_V1.2 expression were similar, indicating that NF-κB is a major Ca_V1.2 gene transcription activator in arterial myocytes. In contrast, in colonic myocytes NF-κB p50 and p65 subunit activation reduced Ca_V1.2 expression ³⁷. Opposing regulation by NF-κB in these different myocyte types may occur through interaction with different κB binding motifs, of which there are several upstream of the Ca_V1.2 gene ³⁷. Furthermore, the presence or absence of additional transcriptional activators and/or repressors may explain differential regulation of Ca_V1.2 expression by NF-κB. Our data also indicate that ET-1-induced IP₃R-mediated SR Ca²⁺ release stimulates NF-κB subunit expression through a mitochondria-, ROS- and NF-κB-independent pathway. The ET-1-induced elevation in NF-κB expression may serve to amplify Ca_V1.2 expression. In rat renal arteries, membrane depolarization increased Ca_V1.2 protein ³⁸, whereas here IP₃Rs were necessary for ET-1-induced elevation in Ca_V1.2 expression. Data from these studies raise several possibilities, including that local and global Ca²⁺ signals regulate Ca_V1.2 expression by different mechanisms in cerebral and renal artery myocytes. Conceivably, this could occur at many levels, including that global Ca²⁺ may regulate [Ca²⁺]_mito in renal artery myocytes, leading to ROS generation and NF-κB activation.

Data here and previous studies suggest that vasoconstrictors cause a similar shift in local and global Ca²⁺ signals in myocytes of anatomically different arteries ^9,39. Therefore, IP₃R regulation of Ca²⁺ waves, [Ca²⁺]_mito and mitoROS generation may be a common mechanism by which vasoconstrictors regulate myocyte NF-κB activity and functional Ca_V1.2 expression. IP₃R-mediated SR Ca²⁺ release and local Ca²⁺ influx through Ca_V1.2 channels also stimulates calcineurin-dependent nuclear translocation of nuclear factor of activated T cell c3 in arterial myocytes ^40,41. Data here not only indicate that IP₃Rs control physiological Ca_V1.2 expression via a mitoROS/NF-κB pathway, but raise the possibility that disease-associated alterations in ion channel expression may also occur through activation of this pathway. Many vasoconstrictors, including ET-1, are elevated in cardiovascular diseases, including systemic hypertension ⁴². Hypertension is also associated with an increase in vascular ROS, NF-κB activity, and Ca_V1.2 protein and currents ^43–45. Therefore, targeting this transcriptional pathway may be beneficial in treating cardiovascular diseases.

In summary, this study indicates that mitochondria sense IP₃R-mediated SR Ca²⁺ release to control the activity of NF-κB, which stimulates functional Ca_V1.2 expression in cerebral artery myocytes.

Novelty and Significance.

What is known?

• Mitochondria can sequester calcium (Ca²⁺) when exposed to elevated levels of intracellular Ca²⁺ ([Ca²⁺]_i).

• Mitochondria generate reactive oxygen species (ROS), which can regulate the activity of a variety of downstream targets, including several transcription factors.

• Voltage-dependent Ca²⁺ (Ca_V1.2) channels are a major contributor to arterial smooth muscle cell global [Ca²⁺]_i, a Ca²⁺ signal that regulates blood pressure and regional blood flow.

What new information does this article contribute?

• In contractile arterial smooth muscle cells, sarcoplasmic reticulum (SR) Ca²⁺ released by inositol 1,4,5-trisphosphate receptors (IP₃R) specifically elevates mitochondrial Ca²⁺ concentration ([Ca²⁺]_mito), leading to mitochondrial depolarization.

• An IP₃R-mediated elevation in [Ca²⁺]_mito induces the generation of mitochondrial ROS (mitoROS), which activate nuclear factor kappa B (NF-κB), a transcription factor.

• NF-κB controls basal Ca_V1.2 expression and IP₃R-mediated mitoROS-induced NF-κB activation elevates Ca_V1.2, leading to vasoconstriction.

Physiological functions of mitochondria in contractile arterial smooth muscle cells are poorly understood. Arterial smooth muscle cells generate several distinct local and global Ca²⁺ signals. Mitochondria can sequester Ca²⁺, but Ca²⁺ signals that regulate [Ca²⁺]_mito and the significance of changes in [Ca²⁺]_mito are unclear. We show that IP₃R-mediated SR Ca²⁺ release, but not a global [Ca²⁺]_i elevation, elevates [Ca²⁺]_mito. An IP₃R-mediated [Ca²⁺]_mito elevation stimulates the generation of mitoROS, which activate NF-κB. IP₃R-mediated mitoROS-induced NF-κB activation elevates functional Ca_V1.2 expression. This study indicates that mitochondria sense changes in IP₃R-mediated SR Ca²⁺ release to alter NF-κB-mediated Ca_V1.2 expression, thereby modulating arterial contractility. Many cardiovascular diseases, including hypertension, are associated with altered ROS generation, NF-κB activity, Ca_V1.2 expression, and vascular contractility. The identification of this novel mitochondrial signaling pathway controlling arterial contractility may lead to the development of new approaches to treat cardiovascular diseases.

Non-standard Abbreviations and Acronyms

Ca²⁺

Calcium

[Ca²⁺]_i

Intracellular Ca²⁺ concentration

[Ca²⁺]_mito

Mitochondrial Ca²⁺ concentration

Ca_V1.2

Voltage-dependent Ca²⁺ channel

CCCP

Carbonyl cyanide 3-chlorophenylhydrazone

CT

fluorescence threshold

DCF

Dichlorofluorescin

ET-1

Endothelin-1

ETC

Mitochondrial electron transport chain

F

fluorescence

F₀

baseline fluorescence

gp91ds-tat

gp91^phox docking sequence peptide conjugated to tat

gp91scrm-tat

scrambled gp91^phox peptide conjugated to tat

H₂O₂

Hydrogen peroxide

IκB

I kappa B

IKK

IκB kinase

IP₃

Inositol 1,4,5-trisphosphate

[IP₃]_i

Intracellular IP₃ concentration

IP₃R

IP₃ receptor

mitoROS

Mitochondrial reactive oxygen species

MnTMPyP

Manganese (III) tetrakis(1-methyl-4-pyridyl)porphyrin

mPTP

Mitochondrial permeability transition pore

NF-κB

Nuclear factor kappa B

NF-κB-p-Luc

Vectors that express firefly luciferase under the control of an NF-κB promoter

17-ODYA

17-octadecanoic acid

p105

NF-κB p105 subunit

p105siRNA

siRNA directed against NF-κB p105 subunit

p105scrm

Scrambled siRNA

p50

NF-κB p50 subunit

PBS

phosphate-buffered saline

PKC

Protein kinase C

PLC

Phospholipase C

PSS

physiological salt solution

ROS

Reactive Oxygen Species

Rps5

ribosomal protein S5

RyR

Ryanodine-sensitive Ca²⁺ channel

SR

Sarcoplasmic reticulum

TMRM

Tetramethylrhodamine Methyl Ester

TNF-α

Tumor necrosis factor-alpha

XeC

Xestospongin C