Mitochondrial Reactive Oxygen Species Trigger Calcium Increases During Hypoxia in Pulmonary Arterial Myocytes
We hypothesized that mitochondria function as the O2 sensors underlying hypoxic pulmonary vasoconstriction by releasing reactive oxygen species (ROS) from complex III of the electron transport chain (ETC). We have previously found that antioxidants or inhibition of the proximal region of the ETC attenuates hypoxic pulmonary vasoconstriction in rat lungs and blocks hypoxia-induced contraction of isolated pulmonary arterial (PA) myocytes. To determine whether the hypoxia-induced increases in mitochondrial ROS act to trigger calcium increases, we measured changes in cytosolic calcium ([Ca2+]i) using fura 2-AM (fluorescence at 340/380 nm) during perfusion with hypoxic media (Po2 12 mm Hg). Hypoxia caused an increase in fura 2 fluorescence, indicating an increase in [Ca2+]i. In superfused PA myocytes, diphenyleneiodonium, rotenone, and myxothiazol, which inhibit the proximal region of the ETC, attenuated hypoxia-induced calcium increases. Antimycin A and cyanide, which inhibit the distal region of the ETC, failed to abolish hypoxia-induced [Ca2+]i increases. To test whether mitochondrial H2O2 is required to trigger [Ca2+]i increases, catalase was overexpressed in PA myocytes with the use of a recombinant adenovirus. Catalase overexpression attenuated hypoxia-induced increases in [Ca2+]i, suggesting that H2O2 acts upstream from calcium increases during hypoxia. These results support the conclusion that mitochondria function as O2 sensors during hypoxia and demonstrate that ROS generated in the proximal region of the ETC act as second messengers to trigger calcium increases in PA myocytes during acute hypoxia.
Hypoxic pulmonary vasoconstriction (HPV) diverts blood flow away from the lung during fetal development and optimizes lung gas exchange after birth by enhancing the matching of blood flow and ventilation. Excised lungs retain the HPV response,1–6⇓⇓⇓⇓⇓ as do rings of the pulmonary artery (PA)7,8⇓ even when they are denuded of endothelium.9,10⇓ Even isolated PA myocytes contract during hypoxia,11 indicating that an O2 sensor is intrinsic to those cells.
Although HPV has been well characterized, the underlying mechanism of O2 sensing is not established. Mitochondria have long been known to generate reactive oxygen species (ROS),12 although these oxidants have classically been viewed as toxic byproducts of the electron transport pathway, possibly contributing to the effects of aging.13 More recently, mitochondrial ROS have been implicated as intracellular signaling agents. We previously reported that mitochondria increase ROS generation during hypoxia and that site-specific inhibition of electron transport could attenuate hypoxia-induced ROS generation.14,15⇓ In isolated buffered-perfused rat lungs and isolated PA myocytes, we found that inhibition of the mitochondrial electron transport chain (ETC) upstream from complex III attenuated HPV.16 By contrast, inhibition downstream from complex III either had no effect or augmented HPV.16 The likely explanation for these findings is that O2-dependent ROS production occurs within complex III. In the Q cycle, a free radical (ubisemiquinone) is normally generated during the electron transport process. This radical can potentially donate its unpaired electron to O2, thereby generating superoxide. Our model suggests that the process of ROS generation from that site is amplified during hypoxia.
In support of this model, we found that antioxidants selectively abolish the HPV response in isolated lungs.16 However, that study did not test whether the calcium increases during hypoxia required ROS production. Accordingly, the present study sought to determine whether oxidant production from the mitochondria is responsible for triggering calcium increases and therefore myocyte contraction during HPV. This hypothesis was tested by measuring the effects of site-specific mitochondrial inhibitors and antioxidants on calcium signaling in primary cultured PA myocytes under hypoxic conditions.
Materials and Methods
Fura 2-AM was obtained from Molecular Probes; diphenyleneiodonium (DPI) was obtained from CalBiochem; and angiotensin II, rotenone, myxothiazol, antimycin A, and cyanide were obtained from Sigma Chemical Co. The pharmacological agents were dissolved in dimethyl sulfoxide (100%) as a 1000× stock solution so that when it was added to the media, the dimethyl sulfoxide concentration was 0.1%. Stock solutions were stored at −20°C and thawed on the day of the experiment. Unless otherwise noted, pharmacological agents were incubated with the cells for 10 minutes before the start of hypoxia or the administration of angiotensin II, H2O2, or cyanide.
Pulmonary Microvessel Myocyte Isolation
Myocytes were isolated using a modification of the method of Marshall et al.7 As described previously,16 freshly excised rat heart and lungs were rinsed with PBS containing penicillin and streptomycin (1%). The right ventricle was cannulated, and the pulmonary vasculature was flushed with the PBS solution (30 mL). By use of the PA cannula, growth medium 199 (M199, 30 mL) containing HEPES (25 mmol/L) along with penicillin and streptomycin (1%) plus low-melting-point agarose (0.5%) and iron particles (0.5%) was flushed through the pulmonary vasculature. The iron particles were too large to pass through the capillaries; therefore, only the arteries were filled with the agarose and iron particles. The airways were filled via the trachea with M199 (15 mL) containing low-melting-point agarose (1%) without iron. The lungs were placed in cold PBS to cause the agarose to gel. After 10 minutes, the lobes were dissected free and finely minced in a Petri dish. Lung fragments were resuspended and washed (3 times) with PBS by use of a magnet to retain the iron-containing fragments. The iron-containing pieces were resuspended in M199 (25 mL) containing collagenase (80 U/mL) and incubated at 37°C for 30 to 60 minutes. To remove extravascular tissue, fragments were first drawn through a 15-gauge needle and subsequently through an 18-gauge needle. The iron-containing fragments then were washed (3 times) with M199 containing FBS (20%) and drained. The resulting fragments were placed in a Petri dish containing collagen-coated (0.01%) glass coverslips and resuspended in M199 containing FBS (10%). The Petri dishes were incubated at 37°C with CO2 (5%) in air for 4 or 5 days, during which time the myocytes were observed to migrate and adhere to the coverslips. After 4 or 5 days, the media and iron-containing particles were transferred to a new dish containing fresh media. The adherent myocytes continued to propagate until the cells were 50% confluent. Cells were confirmed to be PA myocytes, as previously described.16
Cytosolic Calcium Measurements
PA myocytes on coverslips were incubated at 37°C with fura 2-AM (5 μmol/L) for 1 hour and then incubated for an additional 15 minutes in medium without the dye to ensure that all of the intracellular dye had the ester moiety cleaved off. Coverslips were then placed in a low-volume glass-covered recording chamber (Warner) on the microscope and perfused (1 to 2 mL/min) with M199 bubbled with 21% O2/5% CO2/74% N2 (normoxia). Cells were maintained at 34.5±0.2°C by heating both the M199 and the chamber itself. Because a buffered salt solution was being used to perfuse the cells, angiotensin II (10 μmol/L) was added to the perfusate 10 minutes before hypoxia to prime cells for hypoxia as previously described.16 Hypoxia was induced by switching the perfusing medium from normoxia to hypoxia (1.5% O2/5% CO2/73.5% N2). Experimental drugs were diluted in M199 in separate reservoirs connected to a manifold, and solutions were changed by computer-controlled solenoids. The fluid within the recording chamber was completely exchanged every 3 to 4 seconds.
PA myocytes were observed with an inverted microscope (Nikon) and imaged with a cooled CCD camera (Photometrics) controlled by computer (Universal Imaging). Images were obtained using a ×40 (numerical aperture 1.3) oil-immersion objective. The fura 2 was excited using a 75-W xenon source, the intensity of which was decreased with neutral density filters. For studies of PA myocyte morphology, differential interference contrast images were obtained with transmitted halogen illumination. Sets of three images were serially acquired at 20-second intervals: a pair of epifluorescence images, followed by a differential interference contrast image.
Fura 2 ratios were calibrated to cytosolic calcium ([Ca2+]i, in nanomoles per liter) in situ according to the method of Grynkiewicz et al.17 We used 224 nmol/L as the Kd for calcium in the following formula: Ca2+=Kd · F380max/F380min · (R−Rmin)/(Rmax−R), where F380 is fluorescence at 380 nm, and R is the fluorescence ratio.
Fura 2–loaded cells were incubated in calcium-free 3 mmol/L EGTA–containing buffer for 1 hour in the presence of the calcium ionophore 4-bromo-A23187 (10 μmol/L), and the resultant fura 2 fluorescence was used as Rmin. The solution was then switched to a EGTA-free buffer containing 10 mmol/L calcium and A23187, and the resultant fluorescence was used as Rmax.
Recombinant adenoviruses containing either LacZ or catalase were obtained from Dr Jingxiang Bai (Mount Sinai School of Medicine, New York, NY).18 For adenovirus production, small aliquots of adenovirus were administered to low-passage 293 cells. Virus production was determined to have taken place when the cells displayed a cytopathic effect (CPE), ie, when the cells round up and begin to detach. When ≈90% of the cells had detached because of CPE, all cells were harvested in their media and were then lysed by five rounds of freeze-thaw to release virus. Cell debris was spun down, and the supernatant was used to infect one 150-mm plate of 293 cells. When CPE was observed in these cells, the harvesting process was repeated, and the resulting supernatant was used to infect 20 to thirty 150-mm plates. Cells from this large-scale preparation were harvested, and the virus-containing supernatant was collected as described above in 10 mL complete medium. Infectious adenovirus was purified by two rounds of cesium chloride gradient ultracentrifugation, followed by dialysis against 10 mmol/L Tris pH 7.4, 1 mmol/L MgCl2, and 10% glycerol. Total viral yield was determined by absorbance at 280, and infectious viral titer was determined by plaque assay. PA myocytes were infected with two plaque-forming units per cell in complete medium (M199 with FBS) for 18 hours and then changed to complete medium without virus. Experiments were performed 36 to 48 hours after the initial infection. Experiments were also performed as a blind study, in that the experimenter had no knowledge of which virus was used to treat the cells.
H2O2 Degradation in PA Myocytes Overexpressing Catalase
To determine whether the catalase overexpressed in PA myocytes was functional, the rate of exogenous H2O2 degradation was determined using a xylenol orange assay.19 Cells on coverslips were incubated in medium (2 mL) to which a bolus of H2O2 was added (700 μmol/L). The medium was sampled at 15-minute intervals for the measurement of H2O2. Samples (900 μL) of medium were incubated at 27°C for 45 minutes with 100 μL of a 10× xylenol orange stock solution [10 μmol/L xylenol orange, 2.5 μmol/L Fe(NH4)2(SO4)2 · 6H2O, and 25 mmol/L H2SO4; Sigma], after which absorbance was read at 590 nm, and the concentration of H2O2 was determined from a standard curve (0.1, 0.5, 1, 5, 10, 50, 100, 500, and 1000 μmol/L H2O2). The H2O2 concentration was determined by spectrophotometric analysis at 240 nm using an extinction coefficient of 43.6 (mol/L)−1 · cm−1.
Changes in [Ca2+]i were analyzed using a t test to evaluate significant differences between changes in hypoxia-induced fura 2 ratio in the absence and presence of the experimental agents. To control for differences in the hypoxic responses of cultured myocytes, experimental studies and control experiments were always carried out on the same day. Statistical significance was set at P<0.05.20
Hypoxia Increases [Ca2+]i in PA Myocytes
Perfusion of the cells with medium containing angiotensin II (5 μmol/L) caused an increase in [Ca2+]i, which returned to baseline levels within 10 minutes (data not shown). Angiotensin II was used to confirm the ability of these cells to increase [Ca2+]i in response to a vasoconstrictor and to prime the cells to respond to hypoxia.16 Figure 1A depicts a photomicrograph of fura 2 fluorescence (ratio of F340/F380) in isolated cultured PA myocytes during perfusion with normoxic medium (Po2 140 mm Hg). Perfusion with hypoxic medium (Po2 12 mm Hg) for 2 minutes resulted in an increase in fura 2 fluorescence compatible with an increase in [Ca2+]i (Figure 1B). Perfusion of isolated cultured PA myocytes with hypoxic medium caused an increase in [Ca2+]i from a baseline level of 165±8 to 290±29 nmol/L (Figure 1, bottom). This increase in [Ca2+]i was stable over at least 9 minutes of hypoxia. These results confirm the suitability of this system to measure hypoxia-induced increases in [Ca2+]i in isolated cultured PA myocytes.
Inhibition of Mitochondrial Electron Transport Upstream From Complex III Decreases the Hypoxic Response
To determine the requirement for electron transport upstream from complex III in hypoxia-induced calcium increases, the flavoprotein inhibitor DPI was added to the medium perfusing the PA myocytes. To determine whether our experimental drugs had nonspecific effects on [Ca2+]i, we performed additional experiments measuring changes in [Ca2+]i in response to the vasoconstrictor angiotensin II and the ROS H2O2. DPI at 10 μmol/L, a concentration previously shown to inhibit HPV,16 had no effect on an angiotensin II (5 μmol/L)– or an H2O2 (50 μmol/L)–induced increase in [Ca2+]i (Figure 2A). However, DPI significantly decreased hypoxia-induced fura 2 fluorescence (0.02±0.02, peak ratio−baseline ratio) compared with control (0.20±0.03).
To clarify the role of the proximal region of the ETC, the complex I inhibitor rotenone was added to the medium during the incubation period with fura 2-AM. Rotenone at 5 μg/mL, a dose shown to irreversibly inhibit mitochondrial O2 consumption,16 significantly decreased hypoxia-induced fura 2 fluorescence (0.06±0.01) compared with control (0.13±0.02) (Figure 2B) without affecting the response to angiotensin II or H2O2.
Myxothiazol inhibits complex III by blocking electron transfer to the Rieske iron-sulfur center in the bc1 complex.21 Myxothiazol at 5 μg/mL, a dose shown to inhibit mitochondrial O2 consumption and HPV,16 had no effect on the response to angiotensin II or H2O2 (Figure 2C). However, myxothiazol significantly decreased hypoxia-induced fura 2 fluorescence (0.03±0.03) compared with control (0.20±0.03).
Inhibition of Mitochondrial Electron Transport Downstream From Complex III Does Not Alter the Hypoxic Response
Antimycin A inhibits the oxidation of cytochrome b562 at complex III downstream from ubisemiquinone.21 Antimycin A at 1 μg/mL, a dose shown to inhibit mitochondrial O2 consumption,16 had no effect on hypoxia-induced fura 2 fluorescence (0.17±0.01) compared with control (0.14±0.01) (Figure 3). In addition, antimycin A had no effect on the response to angiotensin II or H2O2 (data not shown). Cyanide inhibits complex IV of the mitochondria. Cyanide at 10 μmol/L, a dose shown to inhibit mitochondrial O2 consumption,16 had no effect on hypoxia-induced fura 2 fluorescence (0.10±0.01) compared with control (0.10±0.01) (Figure 3). Furthermore, cyanide had no effect on the angiotensin II or H2O2 response (data not shown).
Increased ROS Mediate the Calcium Signal During Hypoxia
Having determined that the calcium response depends on a mitochondrial site downstream from complex I but upstream from cytochrome b562, we sought to determine whether increases in H2O2 are required for calcium increases during hypoxia. Catalase was overexpressed in PA myocytes by infection with a recombinant adenovirus 36 to 48 hours before the study. Expression levels were verified by Western blot analysis (Figure 4, inset). Control cells were infected with an identical virus containing the LacZ gene instead of the catalase insert. Compared with wild-type PA myocytes or PA myocytes that were infected with an adenovirus containing LacZ, PA myocytes infected with the catalase adenovirus displayed an increase in catalase expression. Inspection of cells expressing β-galactosidase indicated virtually 100% expression efficiency of the transgene (data not shown). To verify that catalase overexpression amplified H2O2 degradation, PA myocytes were incubated with exogenous H2O2 (≈700 μmol/L), and the concentration of H2O2 in the media bathing the cells was measured as a function of time (Figure 4). H2O2 was stable in medium (without cells), whereas wild-type and LacZ-expressing cells slowly degraded the H2O2 via their endogenous antioxidant machinery. PA myocytes overexpressing catalase degraded H2O2 significantly faster than did the wild-type or LacZ-transfected cells. Thus, the adenovirus system resulted in the expression of the transgene with high efficiency at a level that could be adjusted by varying the titer of viral concentration.
Catalase overexpression in cells had no effect on angiotensin II–induced fura 2 fluorescence (Figure 5) compared with wild-type and LacZ transfected cells. However, it did decrease H2O2-induced fura 2 fluorescence (0.10±0.05) compared with wild-type (0.27±0.07) and LacZ-transfected (0.29±0.10) cells. Catalase overexpression in cells also decreased hypoxia-induced increases in fura 2 fluorescence (0.06±0.004) compared with wild-type (0.11±0.02) and LacZ-expressing (0.10±0.01) cells.
Cyanide Mimics Hypoxia-Induced Calcium Increases
Cyanide has been shown to mimic HPV during normoxia through an as-yet-unknown mechanism.2,16⇓ To determine whether cyanide mimics HPV by increasing ROS generation at complex III, we measured cyanide-induced [Ca2+]i during normoxia in PA myocytes (Figure 6). Cyanide at 10 μmol/L increased fura 2 fluorescence during normoxia (0.14±0.04). To determine whether inhibition of the mitochondrial ETC upstream from ubisemiquinone could block this response, cells were preincubated with myxothiazol. Myxothiazol (5 μg/mL) significantly attenuated cyanide-induced fura 2 fluorescence (0.04±0.02), suggesting that mitochondrial electron transport upstream from complex III is required for cyanided-induced [Ca2+]i increases. To determine whether increased H2O2 is required for cyanide-induced [Ca2+]i increases, PA myocytes overexpressing catalase were challenged with cyanide during normoxia. PA myocytes overexpressing catalase significantly decreased cyanide-induced fura 2 fluorescence increases (0.02±0.02), whereas PA myocytes expressing LacZ responded to cyanide in a manner similar to that in control cells (0.10±0.04). Thus, H2O2 generation is required for calcium increases in response to cyanide.
Mitochondria Mediate Calcium Increases During Hypoxia
To determine whether mitochondria function as the O2 sensors mediating calcium increases during hypoxia, we measured hypoxia-induced fura 2 fluorescence in isolated PA myocytes. In cells loaded with fura 2, hypoxia induced an increase in [Ca2+]i from a baseline of 165±8 to 290±28 nmol/L. This finding is in agreement with previously published results.22 Inhibitors of the proximal region of the mitochondrial ETC, including DPI, rotenone, and myxothiazol, attenuated the increase in calcium in response to hypoxia without affecting the response to angiotensin II or H2O2. By contrast, distal inhibitors, including antimycin A and cyanide, had no effect on hypoxia-induced calcium signaling. We conclude that the response to hypoxia requires electron transport but does not require mitochondrial ATP because all of the inhibitors block oxidative phosphorylation, yet only the proximal inhibitors selectively decrease the hypoxia-induced increase in fura 2 fluorescence.
These results are consistent with our previously reported observations in isolated buffer-perfused rat lungs.16 Recently, Leach et al23 reported a series of compatible observations in isolated PAs. They found that rotenone and myxothiazol attenuated hypoxia-induced increases in [Ca2+]i and subsequent vasoconstriction. By contrast, inhibition of complex IV with cyanide augmented the hypoxic response but had no effect on [Ca2+]i. Interestingly, in PAs that had been treated with rotenone, they were able to restore HPV by using succinate to shuttle electrons into complex III via complex II, effectively bypassing the rotenone inhibition of complex I.23 Our findings confirm and extend their conclusions that mitochondria function as the O2 sensors mediating calcium increases in PA myocytes during hypoxia.
ROS as Second Messengers in Hypoxia-Induced Calcium Increases
The present results suggest that the mitochondrial ETC acts as an O2 sensor during hypoxia by releasing ROS that function as signaling messengers (Figure 7). Superoxide generation is known to occur at the ubisemiquinone site of complex III via univalent electron transfer to O2.24 The overexpression of catalase decreased hypoxia- and H2O2-induced [Ca2+]i responses without affecting the response to angiotensin II, suggesting that ROS act as second messengers that regulate calcium increases during hypoxia.
These results are consistent with our previous observations in isolated buffer-perfused rat lungs16 and other studies that implicate increased ROS generation in the response to hypoxia. For example, Monaco et al25 showed that HPV was augmented when catalase was inhibited with aminotriazole. Also, Weissmann et al5 observed that superoxide dismutase and 4,5-dihydroxy-1,3-benzenedisulfonic acid (to accelerate H2O2 generation from superoxide) did not affect HPV, suggesting that H2O2, rather than superoxide, is involved. By contrast, nitro blue tetrazolium, which traps superoxide and prevents H2O2 formation, attenuated HPV.5 These findings are consistent with a role for increased H2O2 as a signaling molecule during hypoxia.
Mitochondria as the Source of ROS During Hypoxia
Mitochondrial ROS have been increasingly implicated as intracellular signaling agents. We previously reported that mitochondria increase ROS generation during hypoxia and that site-specific inhibition of electron transport could attenuate this response.14–16⇓⇓ During normoxia, mitochondrial inhibitors, such as cyanide, which block sites distal to complex III, tend to augment ROS generation (Figure 7). The likely explanation for this is that ROS production occurs primarily within complex III at the ubisemiquinone site. By contrast, when electron transport is inhibited at a more proximal site by rotenone, DPI, or myxothiazol, the ubiquinone pool becomes fully oxidized, and ROS generation at complex III is abrogated. A caveat to this is that myxothiazol has been reported in certain cases to increase ROS production at complex III during normoxia, although this effect was not evident in our studies.26 According to our model, the ROS production at complex III increases during hypoxia, and these oxidants trigger signal pathways culminating in [Ca2+]i increases and subsequent HPV. In accordance with this model, distal inhibitors of the ETC, such as cyanide, induced calcium increases during normoxia by triggering ROS production (Figure 6), and this response could be inhibited by catalase overexpression or myxothiazol. This is consistent with data from Archer et al,2 who found that cyanide caused vasoconstriction during normoxia and augmented HPV without affecting the response to angiotensin II or KCl. Their observations are consistent with our model, but the mechanism by which hypoxia amplifies ROS generation at complex III is not yet known.
Mitochondrial ROS-Induced Calcium Increases
Although H2O2 appears to act as a signaling messenger in the sequence leading to hypoxia-induced calcium increases, the details of that pathway are not fully understood. It has been suggested that hypoxia-induced calcium increases may involve 1 or more 4-aminopyridine (4-AP)–sensitive KV channels. Inhibition of these channels during hypoxia would promote membrane depolarization, the opening of voltage-gated L-type calcium channels, and the subsequent intracellular calcium increases.2,9–11,27–29⇓⇓⇓⇓⇓⇓ However, other studies have questioned the requirement for KV channels for this response. For example, HPV has been found to remain intact despite the inhibition of KV channels with 4-AP.30,31⇓ Furthermore, HPV was significantly blunted, but not completely abolished, in SWAP mice, which lack the KV1.5 channel.32 This suggests that KV channels may function to augment the HPV response rather than to trigger it directly.
An alternative explanation is that hypoxia causes the release of calcium from intracellular stores, a step that appears to be required for HPV.9,33⇓ Release of intracellular stores could then cause inhibition of 4-AP–sensitive KV channels, leading membrane depolarization.34 Such a mechanism would be consistent with a model of hypoxia-induced calcium increases in which hypoxia triggers the release of calcium from internal stores, resulting in KV channel inhibition, membrane depolarization, the opening of voltage-gated L-type calcium channels, and subsequent contraction.
Possible sources of intracellular calcium during hypoxia include mitochondria, inositol 1,4,5-triphosphate (IP3)-sensitive sarcoplasmic reticulum (SR) stores, and ryanodine-sensitive SR stores. Oxidative stress has been shown to promote mitochondrial depolarization, resulting in calcium efflux. However, our previous studies have demonstrated that these levels of hypoxia do not induce mitochondrial depolarization, which suggests that mitochondria are not the source of calcium in our experiments.16 Gelband and Gelband9 and Robertson et al35 found that hypoxic contraction of isolated PAs was inhibited by thapsigargin, which suggests that hypoxia triggers the release of calcium from IP3-sensitive SR stores. Along these lines, exogenous H2O2 has been shown to activate phospholipase C in a time- and concentration-dependent manner.36 Therefore, it is possible that mitochondrial ROS could trigger intracellular calcium release by activating phospholipase C, which in turn would metabolize phosphatidylinositol 4,5-bisphosphate to IP3 and diacylglycerol. IP3 could then trigger the release of calcium from IP3-sensative intracellular stores. However, studies demonstrating this mechanism have not been published. Hypoxia does appear to trigger the release of calcium from ryanodine-sensitive SR stores.9,22,33,37–40⇓⇓⇓⇓⇓⇓ Recently, Wilson et al40 reported that hypoxia increases cyclic ADP-ribose accumulation in isolated rabbit PA and calcium release from ryanodine-sensitive SR stores. Richter et al41 reported that H2O2 could cause oxidation of mitochondrial pyridine nucleotides, resulting in increased ADP-ribose production, mono-ADP ribosylation of mitochondrial proteins, and calcium release. By these mechanisms, mitochondrial ROS could trigger calcium release from IP3- and/or ryanodine-sensitive SR stores. The resulting increase in intracellular calcium could inhibit KV channels, and the increase in calcium resulting from membrane depolarization could trigger calmodulin-mediated activation of myosin light chain kinase, actin-myosin interaction, and contraction.
In the present study, we focused exclusively on the PA myocyte response to acute hypoxia. However, more prolonged hypoxia can result in the recruitment of additional components of HPV, including the endothelium, which can modulate HPV by releasing vasoconstricting substances that increase myofilament calcium sensitivity, or NO, which promotes relaxation. Recently, Leach et al23 demonstrated that glucose and glycolysis are required for the development of sustained HPV by supporting a hypoxia-induced endothelium-dependent calcium sensitization pathway. This suggests that an increase in [Ca2+]i is not the only requirement for the HPV response. Future studies are required to determine the mechanism of O2 sensing underlying these endothelial responses to hypoxia during HPV.
In summary, our results demonstrate that hypoxia-induced calcium increases require mitochondrial electron transport proximal to the ubisemiquinone site but do not require the entire ETC to be functional. ROS generated by mitochondria appear to function as second messengers during hypoxia and act to trigger the calcium-signaling process responsible for the contraction of PA myocytes during hypoxia.
This research was supported by NIH grants HL-35440, HL-66315, and HL-10405 and American Heart Association Grant 0235457Z. We thank Drs Jingxiang Bai and Arthur I. Cederbaum of Mount Sinai School of Medicine, New York, NY, for the generous gift of the recombinant adenoviruses containing LacZ and catalase.
Original received July 17, 2002; revision received August 26, 2002; accepted August 27, 2002.
- ↵Rounds S, McMurtry IF. Inhibitors of oxidative ATP production cause transient vasoconstriction and block subsequent pressor responses in rat lungs. Circ Res. 1981; 48: 393–400.
- ↵Archer SL, Huang J, Henry T, Peterson D, Weir EK. A redox-based O2 sensor in rat pulmonary vasculature. Circ Res. 1993; 73: 1100–1112.
- ↵McMurtry IF. Angiotensin is not required for hypoxic constriction in salt solution-perfused rat lungs. J Appl Physiol. 1984; 56: 375–380.
- ↵Gelband CH, Gelband H. Ca2+ release from intracellular stores is an initial step in hypoxic pulmonary vasoconstriction of rat pulmonary artery resistance vessels. Circulation. 1997; 96: 3647–3654.
- ↵Archer SL, Souil E, Dinh-Xuan AT, Schremmer B, Mercier JC, El Yaagoubi A, Nguyen-Huu L, Reeve HL, Hampl V. Molecular identification of the role of voltage-gated K+ channels, Kv1.5 and Kv2.1, in hypoxic pulmonary vasoconstriction and control of resting membrane potential in rat pulmonary artery myocytes. J Clin Invest. 1998; 101: 2319–2330.
- ↵Sohal RS, Weindruch R. Oxidative stress, caloric restriction, and aging. Science. 1996; 273: 59–63.
- ↵Chandel NS, Maltepe E, Goldwasser E, Mathieu CE, Simon MC, Schumacker PT. Mitochondrial reactive oxygen species trigger hypoxia-induced transcription. Proc Natl Acad Sci U S A. 1998; 95: 11715–11720.
- ↵Duranteau J, Chandel NS, Kulisz A, Shao Z, Schumacker PT. Intracellular signaling by reactive oxygen species during hypoxia in cardiomyocytes. J Biol Chem. 1998; 273: 11619–11624.
- ↵Waypa GB, Chandel NS, Schumacker PT. Model for hypoxic pulmonary vasoconstriction involving mitochondrial oxygen sensing. Circ Res. 2001; 88: 1259–1266.
- ↵Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985; 260: 3440–3450.
- ↵Bai J, Cederbaum AI. Adenovirus-mediated overexpression of catalase in the cytosolic or mitochondrial compartment protects against cytochrome P450 2E1-dependent toxicity in HepG2 cells. J Biol Chem. 2001; 276: 4315–4321.
- ↵Wallenstein S, Zucker CL, Fleiss JL. Some statistical methods useful in circulation research. Circ Res. 1980; 47: 1–9.
- ↵Dipp M, Nye PC, Evans AM. Hypoxic release of calcium from the sarcoplasmic reticulum of pulmonary artery smooth muscle. Am J Physiol. 2001; 281: L318–L325.
- ↵Boveris A, Oshino N, Chance B. The cellular production of hydrogen peroxide. Biochem J. 1972; 128: 617–630.
- ↵Sham JS, Crenshaw BR Jr, Deng LH, Shimoda LA, Sylvester JT. Effects of hypoxia in porcine pulmonary arterial myocytes: roles of KV channel and endothelin-1. Am J Physiol. 2000; 279: L262–L272.
- ↵Archer SL, London B, Hampl V, Wu X, Nsair A, Puttagunta L, Hashimoto K, Waite RE, Michelakis ED. Impairment of hypoxic pulmonary vasoconstriction in mice lacking the voltage-gated potassium channel Kv1.5. FASEB J. 2001; 15: 1801–1803.
- ↵Morio Y, McMurtry IF. Ca2+ release from ryanodine-sensitive store contributes to mechanism of hypoxic vasoconstriction in rat lungs. J Appl Physiol. 2002; 92: 527–534.
- ↵Post JM, Gelband CH, Hume JR. [Ca2+]i inhibition of K+ channels in canine pulmonary artery: novel mechanism for hypoxia-induced membrane depolarization. Circ Res. 1995; 77: 131–139.
- ↵Wang XT, McCullough KD, Wang XJ, Carpenter G, Holbrook NJ. Oxidative stress-induced phospholipase C-gamma 1 activation enhances cell survival. J Biol Chem. 2001; 276: 28364–28371.
- ↵Liu Q, Sham JS, Shimoda LA, Sylvester JT. Hypoxic constriction of porcine distal pulmonary arteries: endothelium and endothelin dependence. Am J Physiol. 2001; 280: L856–L865.
- ↵Dipp M, Evans AM. Cyclic ADP-ribose is the primary trigger for hypoxic pulmonary vasoconstriction in the rat lung in situ. Circ Res. 2001; 89: 77–83.
- ↵Wilson HL, Dipp M, Thomas JM, Lad C, Galione A, Evans AM. ADP-ribosyl cyclase and cyclic ADP-ribose hydrolase act as a redox sensor. a primary role for cyclic ADP-ribose in hypoxic pulmonary vasoconstriction. J Biol Chem. 2001; 276: 11180–11188.