Mechanism of Nitric Oxide–Induced Vasodilatation
Refilling of Intracellular Stores by Sarcoplasmic Reticulum Ca2+ ATPase and Inhibition of Store-Operated Ca2+ Influx
Abstract—The precise mechanisms by which nitric oxide (NO) decreases free [Ca2+]i, inhibits Ca2+ influx, and relaxes vascular smooth muscle are poorly understood. In rabbit and mouse aorta, agonist-induced contractions and increases in [Ca2+]i were resistant to nifedipine, suggesting Ca2+ entry through non–L-type Ca2+ channels. Relaxations to NO were inhibited by thapsigargin (TG) or cyclopiazonic acid (CPA) indicating the involvement of sarcoplasmic reticulum ATPase (SERCA). Studies of the effect of NO on [Ca2+]i and the rate of Mn2+ influx with fura-2 fluorometry in rabbit aortic smooth muscle cells in primary culture were designed to test how SERCA is involved in mediating the response to NO. When cells were stimulated with angiotensin II (AII), NO accelerated the removal of Ca2+ from the cytoplasm, decreased [Ca2+]i, and inhibited Ca2+ and Mn2+ influx. Inhibition of SERCA abolished all the effects of NO. In contrast, inhibition of the Na+/Ca2+exchanger or the plasma membrane Ca2+ ATPase had no influence on the ability of NO to decrease [Ca2+]i. NO maximally decreased [Ca2+]i within 5 s, whereas significant inhibition of AII-induced Ca2+ and Mn2+ influx required more than 15 s. The inhibition of cation influx strictly depended on [Ca2+]o and functional SERCA, suggesting that during the delay before NO inhibits Ca2+ influx, the influx of Ca2+ and the uptake into intracellular stores are required. In the absence of [Ca2+]o, NO diminished the AII-induced [Ca2+]i transient by a SERCA-dependent mechanism and increased the amount of Ca2+ in the stores subsequently released by ionomycin. The present study indicates that the initial rapid decrease in [Ca2+]i caused by NO in vascular smooth muscle is accounted for by the uptake of Ca2+ by SERCA into intracellular stores. It is proposed that the refilling of the stores inhibits store-operated Ca2+ influx through non–L-type Ca2+ conducting ion channels and that this maintains the decrease in [Ca2+]i and NO-induced relaxation.
The endothelial cell production of nitric oxide (NO) accounts for endothelium-dependent vasodilatation.1 Despite a growing understanding of the mechanisms by which NO is released from the endothelium, the precise molecular mechanisms by which NO relaxes vascular smooth muscle are not fully understood. A variety of mechanisms have been proposed by which NO participates in the regulation of smooth muscle free [Ca2+]i, which is the primary determinant of contractile tone.2 Among them, several different mechanisms for NO-induced inhibition of Ca2+ influx through L-type Ca2+ channels have been proposed, including their inhibition by cGMP-dependent mechanisms3 or by membrane hyperpolarization due to direct4 or indirect cGMP-dependent5 6 activation of Ca2+-dependent K+ channels. NO also has been proposed to reduce [Ca2+]i by inhibiting the agonist-induced release of Ca2+ from intracellular stores7 or by increasing Ca2+ removal from the cytoplasm by accelerating the plasma membrane Ca2+ ATPase8 or the Na+/Ca2+ exchanger.9 The sarcoplasmic(endo)plasmic reticulum Ca2+ ATPase (SERCA), which may be accelerated by the NO-induced rise in cGMP,10 is also thought to be involved.11 Thus, several potential Ca2+ regulatory targets have been proposed for NO in vascular smooth muscle, but their relative importance in decreasing [Ca2+]i and their individual contribution to smooth muscle relaxation is still unclear.
Supporting previous observations,12 the present study indicates that agonist-induced increases in [Ca2+]i and contractions of rabbit and mouse aortic smooth muscle cells are highly resistant to inhibitors of L-type Ca2+ channels, indicating that Ca2+ influx through different channels can mediate the responses. In a variety of cell types, including smooth muscle, Ca2+ entry is thought to be initiated by agonist-induced Ca2+ release and depletion of [Ca2+]i stores.13 Activation of store-operated Ca2+ entry is known to be triggered not only by receptor-dependent agonists but also by store depletion that occurs after inhibiting Ca2+ uptake into the stores by thapsigargin (TG) or cyclopiazonic acid (CPA).13 14 15
Increases in [Ca2+]i and contractions stimulated by agonists including phenylephrine, angiotensin II (AII), or 5-hydroxytryptamine (5-HT) are shown in the present study to be highly sensitive to NO, but those caused by inhibitors of SERCA are demonstrated to be resistant to NO. This important finding led us to further investigate the effect of NO on [Ca2+]i stores and store-operated Ca2+ entry. Our results suggest a simple model whereby NO initiates its response by accelerating sequestration of Ca2+ into intracellular stores via SERCA, thereby rapidly decreasing [Ca2+]i. The resulting refilling of Ca2+ stores would indirectly inhibit store-operated Ca2+ influx, thereby maintaining the decrease in [Ca2+]i and mediating smooth muscle cell relaxation. This coordinated effect on [Ca2+]i homeostasis represents a novel mechanism for NO-induced vasodilatation.
Materials and Methods
Relaxation of Intact Rabbit and Mouse Aorta
Rings of New Zealand White rabbit descending thoracic aorta were cleaned of adherent fat and connective tissue and were suspended in organ chambers filled with physiological salt solution (PSS) for tension measurements as previously described.16 Rings (3 mm long) of the thoracic aorta of mice (8 to 11 weeks of age) were suspended on horizontal wire myographs (Kent Scientific). After equilibration at the optimal resting tension (mouse aorta, 1.5 g; rabbit aorta, 6 g), the smooth muscle was contracted with phenylephrine hydrochloride, and after achieving a stable contraction, NO was added in increasing concentrations. When needed, the contractile tone caused by SERCA inhibitors was supplemented with phenylephrine to match the tone achieved by phenylephrine in the absence of the inhibitors.
[Ca2+]i Measurement in Intact Mouse Aorta
The mouse aorta was slid over a Teflon filament to fix it and to destroy the endothelial layer. In a Petri dish containing ice-cold PSS (in mmol/L: NaCl 140, KCl 2.8, MgCl2 1, CaCl2 2, glucose 11, and HEPES 10 [pH 7.4]), the aorta was cleaned of fat and connective tissue and cut into 2-mm-long rings. Each ring was transferred into an experimental chamber (ΔTC3 dish, Bioptechs, Inc) between 2 small plaques of neutral silicon grease. Here the ring was cut open, fixed by a piece of woven nylon mesh (Spectra/Mesh nylon, Spectrum) and stored in filtered PSS solution at 4°C until use, but not longer than 4 hours. Fura-2 loading of the strip was done in 10−5 mol/L fura-2 AM in PSS with 0.02% Pluronic F-127 (Molecular Probes) for 1 hour at 37°C. Strips were then washed in prewarmed PSS for 15 minutes at 37°C, and the chamber was mounted into a ΔTC3 temperature control system (Bioptechs, Inc) on the stage of the microscope (Olympus, ×20 fluorescence objective). The solution was warmed to 37°C. The microscope was focused on the elastin and collagen structure of the tissue beneath the nylon mesh. NO was applied as a 50-μL drop of 1 μmol/L solution close to the aortic strip. At the end of each measurement, digitonin and MnCl2 were applied in several steps to avoid contraction and dislocation of the strip, reaching final concentrations of 10−4 mol/L and 2×10−2 mol/L, respectively. Fluorescence background was automatically subtracted. Because of the well-known difficulties in calibrating fura-2 in intact tissues and obtaining precise values, [Ca2+]i changes are reported as ratios after background subtraction.
The thoracic aorta was removed from New Zealand White rabbits and cleaned of adherent fat and connective tissue. Strips of rabbit thoracic aorta were aseptically removed from the media, dispersed with collagenase (4 mg/mL) and elastase (1 mg/mL), and grown in medium M199 with 20% heat-inactivated FBS as previously described.17 Cells were seeded onto 9×22 mm No. 1 glass coverslips for [Ca2+]i studies. On reaching confluence, cells were growth-arrested by placing them in M199 with 1% FBS for 3 to 4 days before the experiments. All experiments were done on primary cultures that stain uniformly for smooth muscle α-actin.
[Ca2+]i Measurement in Cultured Smooth Muscle
As described previously,17 cells in culture on coverslips were loaded for 45 minutes with fura-2 AM (5 μmol/L) in PSS at 37°C and washed for 15 minutes to allow complete ester cleavage. The PSS was of the following composition (mmol/L): NaCl 119, NaHEPES 20, KCl 4.6, CaCl2 1.2, MgSO4 1.0, Na2HPO4 0.15, KH2PO4 0.4, and NaHCO3 5.0, supplemented with 0.1% albumin. Nominally Ca2+ free PSS was composed of the same solution without CaCl2 or EGTA added. Fura-2 fluorescence was measured in a photomultiplier-based fluorometer at 37°C or in an imaging system at 22°C to 24°C (Figure 3⇓) (both from IonOptix Inc). Sampling rate for the ratio of 510-nm emitted light from excitation at 340 and 380 nm was 2 Hz (photomultiplier system) or 5 Hz (imaging system). [Ca2+]i in cultured cells was estimated by previously published methods,17 and the concentration is reported as nanomoles per liter (nmol/L).
Measurement of Divalent Cation Influx Rate
Mn2+-induced quenching of fura-2 fluorescence was used to estimate rates of divalent cation influx.17 18 Fluorescence of fura-2–loaded cells was measured at the isosbestic point wavelength of 360 nm. Data are expressed as the Mn2+ influx rates, which were calculated from the slope of the linear decline in fura-2 fluorescence during the 30 s immediately after addition of MnCl2 (100 μmol/L), and were normalized to the level of fluorescence at the time of Mn2+ addition. When the rate of Mn2+ influx in the presence and absence of [Ca2+]o was compared, the rates were normalized to that measured under basal conditions.
Measurement of Inositol 1,4,5-Trisphosphate
Rabbit aortic smooth muscle cells were grown to confluence in 35-mm culture dishes and made quiescent by reducing the supplemental serum to 1% during the last 4 days before the experiment. Quiescent cells were incubated for 1 hour in serum-free PSS and then stimulated with 0.1 μmol/L AII for 15 s to 70 s. The extracellular solution was removed, and 250 μL of cold 1 mol/L trichloroacetic acid was added. The lysate was collected and centrifuged for 1 minute at 12 000g at 4°C. The trichloroacetic acid was extracted with a 3:1 solution of 1,1,2-trichloro-1,2,2-trifluoroethane-triotylamine, and the remaining inositol 1,4,5-trisphosphate (IP3) was quantified according to the methodology provided with the IP3 (3H) radioreceptor assay kit (Dupont NEN).
NO gas was obtained from Matheson. Saturated NO solution was prepared at 4°C and diluted in deoxygenated sealed tubes of aqueous solution as described.17 Responses of intact rings and cultured cells to NO (10−10 to 10−6 mol/L) were entirely prevented by deoxyhemoglobin (10−5 mol/L). Fura-2 AM was from Molecular Probes. All other chemicals were from Sigma Chemical Co.
All data are expressed as mean±SE. Evaluation of statistics was done by Student t test. Values of P<0.05 indicated significant differences and are designated in the figures by an asterisk.
Effect of Nifedipine and Inhibitors of SERCA on NO-induced Relaxations and Decreases in [Ca2+]i in Intact Rabbit and Mouse Aorta
The role of L-type Ca2+ channels in agonist-induced contractions and NO-induced relaxations of rings of rabbit thoracic aorta denuded of endothelium was determined. Figure 1A⇓ (left) shows that phenylephrine (10−8 to 3×10−6 mol/L) caused concentration-dependent contractions, and NO (10−9 to 10−5 mol/L) caused concentration-dependent relaxations (Figure 1A⇓, right), neither of which was significantly affected by pretreatment of the aorta with nifedipine (10−6 mol/L, 30 minutes; n=5 to 6, P>0.5). This concentration of nifedipine abolished contractions to potassium chloride (30 mmol/L, not shown).
The role of calcium uptake into intracellular stores in the relaxations to NO was determined in rabbit aortic rings pretreated with the SERCA inhibitors TG and CPA (Figure 1B⇑). TG (10−4 mol/L) or CPA (2×10−4 mol/L) caused small, transient, and inconsistent contractions of resting rabbit thoracic aortic rings denuded of endothelium. The concentration of phenylephrine required to contract the rings in the presence of TG or CPA was unchanged (Figure 1B⇑). Relaxations caused by NO (10−10 to 10−5 mol/L) were significantly inhibited by both TG and CPA (P<0.05, n=5; Figure 1B⇑).
In rings of mouse aorta contracted with phenylephrine, NO caused rapid, transient, and concentration-dependent relaxations (Figure 2A⇓). In the mouse aorta, TG consistently caused slowly developing contractions, whereas CPA caused small and variable contractions (Figure 2B⇓ and 2C⇓). The tone induced in the presence of the inhibitors of SERCA, with or without supplemental phenylephrine, was not significantly affected by NO (Figure 2B⇓ through 2D).
To determine whether in the above experiments contraction and relaxation reflected changes in [Ca2+]i, intact mouse aortic strips were loaded with fura-2, and the effects of NO on agonist- and TG-induced increases in [Ca2+]i were measured. 5-HT, used in these experiments because it was the most potent agonist, caused a rapid transient increase in [Ca2+]i followed by a plateau (Figure 2E⇑). NO significantly decreased the plateau level from a ratio of 0.045±0.011 to 0.014±0.010 (P<0.02, n=9). Treatment of the mouse aorta with nifedipine 10 μmol/L did not significantly affect either the plateau (0.038±0.016) or the decrease caused by NO (0.016±0.009, n=6). The NO-induced decrease in [Ca2+]i was 86±26% in the absence and 64±14% in the presence of nifedipine (P>0.1). TG caused a slow increase in [Ca2+]i to a plateau value of 0.104±0.016, and NO did not significantly affect this response (0.099±0.013, P>0.1, n=4; Figure 2F⇑).
SERCA Inhibitors Prevent the Decrease in [Ca2+]i Caused by NO in Rabbit Aortic Smooth Muscle Cells in Culture
The experiments discussed above showing that TG or CPA inhibited decreases in [Ca2+]i and relaxations to NO in intact aortic tissue suggest that calcium uptake into intracellular stores by SERCA is involved in mediating the response to NO. Therefore, additional experiments of the effect of TG and CPA on [Ca2+]i regulation by NO were performed in fura-2–loaded rabbit aortic smooth muscle cells in primary culture. In the presence of [Ca2+]o (1.2 mmol/L), TG (100 nmol/L) caused a rise in [Ca2+]i similar to that observed in intact mouse aorta. NO (10−6 mol/L) had no effect on the TG-induced increase in [Ca2+]i when it was applied after [Ca2+]i reached maximum level (Figure 3A⇓). If NO was added 1 minute before TG, the rise in [Ca2+]i caused by the SERCA inhibitor was slowed but reached a similar maximum level (Figure 3B⇓). If NO was added 1 minute after addition of TG, it caused a small transient decrease in [Ca2+]i (Figure 3C⇓). Thus, in the presence of TG, there is a time-dependent disappearance of the response to NO.
The effect of NO was studied further on AII-induced increases in [Ca2+]i. In these experiments, AII was used because, similar to phenylephrine and 5-HT, it activates IP3-induced Ca2+ release and influx, and it gave the most consistent response in the smooth muscle cells in culture. The AII-induced elevation of [Ca2+]i was separated into the transient rise due to the release of Ca2+ from intracellular stores (shown to occur in the nominal absence of [Ca2+]o) and the sustained elevation in [Ca2+]i after the addition and influx of [Ca2+]o (Figure 4A⇓). Similar to phenylephrine-induced contractions and 5-HT induced increases in [Ca2+]i in the intact rabbit and mouse aorta, the Ca2+ influx–induced sustained elevation of [Ca2+]i was insensitive to nifedipine (5 μmol/L) but was sensitive to NO (10−10 to 10−9 mol/L), as well as to blocking Ca2+ influx by Ni2+ (Figure 4A⇓).
NO (10−6 mol/L) applied during the elevation of [Ca2+]i caused by AII-induced Ca2+ influx (Figure 4B⇑) produced a very rapid and sustained decrease in [Ca2+]i. Because the response to NO could result not only from inhibition of Ca2+ influx but also from acceleration of Ca2+ extrusion from the cytoplasm via the Na+/Ca2+ exchanger and the plasma membrane Ca2+ ATPase (PMCA) or by sequestration of Ca2+ into the stores by SERCA, the effects of inhibition of these systems were determined. Inhibition of the Na+/Ca2+ exchanger (by replacing extracellular Na+ with choline) and/or PMCA (by elevating pHo to 8.8 and Mg2+ concentration to 20 mmol/L8 ) increased the sustained elevation of [Ca2+]i but did not affect the NO-induced decrease in [Ca2+]i (Figure 4C⇑; Table⇓). In contrast, a similar Ca2+ elevation evoked by inhibition of SERCA with TG or CPA (Figure 4D⇑; Table⇓) was completely insensitive to NO (1 μmol/L). Blocking Ca2+ influx with Ni2+ decreased [Ca2+]i when either TG (Figure 1D⇑) or PMCA and Na+/Ca2+ exchanger (Figure 1C⇑) were inhibited (Table⇓). However, when SERCA, PMCA, and Na+/Ca2+ exchanger were all inhibited (Figure 1E⇑), blocking Ca2+ influx with Ni2+ was not followed by a decrease in [Ca2+]i. This indicates that these 3 mechanisms are together responsible for eliminating Ca2+ from the cytoplasm and that they were effectively inhibited. These studies indicate that SERCA activity, but not that of the Na+/Ca2+ exchanger or PMCA, is required for NO to decrease [Ca2+]i.
NO-Induced Inhibition of Ca2+ and Mn2+ Influx Requires SERCA Activity and [Ca2+]o
After stimulation with AII in the absence of [Ca2+]o, treatment of cells with NO (1 μmol/L) significantly inhibited the rise in [Ca2+]i caused by the addition of [Ca2+]o (Figure 5A⇓). The initial rise in [Ca2+]i started similarly in the absence or presence of NO, but in the presence of NO, [Ca2+]i stopped increasing and reached a stable plateau of 214±60 nmol/L after 21±4 s (n=3), compared with a peak of 560±89 nmol/L (P<0.05, n=3) reached in 18±2 s in the absence of NO. In contrast, NO did not affect the rise in [Ca2+]i caused by addition of Ca2+ in the presence of TG (Figure 5B⇓; 551±20 and 531±28 nmol/L, n=3, with or without NO, respectively). These studies suggest that NO inhibits AII- but not TG-induced Ca2+ influx.
To study cation influx while excluding the influence of mechanisms that remove Ca2+ from the cytoplasm, AII-induced Mn2+ influx was measured. Figure 5C⇑ shows that AII increases the rate at which Mn2+ quenched fura-2 fluorescence. NO (1 μmol/L), applied 1 minute before Mn2+, significantly inhibited cation influx caused by AII. Interestingly, in the nominal absence of [Ca2+]o, NO had no effect (Figure 5E⇑). TG (100 nmol/L) increased Mn2+ influx to similar levels as AII, but NO had no effect on Mn2+ influx in either the presence or absence of [Ca2+]o (Figure 5D⇑ and 5F⇑).
Effect of NO on [Ca2+]i Precedes Its Effect on Cation Influx
Having investigated the effects of NO on cation influx, studies were conducted to determine the temporal relationship between the inhibition of cation influx and the decrease in [Ca2+]i caused by NO. The effect of NO was compared with that of Ni2+, which is known to directly block cation influx. When [Ca2+]i was increased by AII, NO (1 μmol/L) decreased [Ca2+]i more rapidly than did Ni2+ (Figure 6A⇓ and 6B⇓). The exponential half-time, τ, was significantly shorter with NO than with Ni2+ (1.4±0.1 versus 3.1±0.2 s, respectively; P<0.001, n=4). In contrast, when Mn2+ influx was measured in cells stimulated with AII, the initial rate of cation influx was inhibited faster by Ni2+ than by NO (Figure 6C⇓ and 6D⇓). When added simultaneously with, or at any time before the addition of Mn2+, Ni2+ fully and immediately blocked AII-induced Mn2+ influx (Figure 6D⇓), consistent with its blocking cation-conducting channels directly. In contrast, when NO was added simultaneously with Mn2+, the cation influx (estimated over the next 30 s) was not significantly different from that obtained with AII alone (Figure 6C⇓). Only if NO was added >15 s before the addition of Mn2+ was there sufficient time for NO to inhibit AII-induced Mn2+ influx. This delay in the action of NO is similar to that needed for inhibition of Ca2+ influx, observed in Figure 5A⇑, and suggests why after addition of [Ca2+]o, [Ca2+]i begins to rise normally, despite treatment with NO. These results indicate that a considerable delay occurs before NO inhibits the influx of extracellular cations and that the decrease in [Ca2+]i caused by NO occurs more rapidly than can be accounted for by a direct inhibition of cation channels. This finding is consistent with the initial rapid decrease in [Ca2+]i by NO being primarily due to stimulation of [Ca2+]i removal mechanisms.
NO Accelerates the Removal of Ca2+ From the Cytoplasm via SERCA and Promotes Refilling of [Ca2+]i Stores
Because the intracellular action of NO precedes its effect on Ca2+ influx, the effect of NO on the rate of removal of Ca2+ from the cytoplasm was further studied. When AII-induced Ca2+ influx was stopped by rapidly eliminating Ca2+ from the extracellular solution, [Ca2+]i decreased with a half-time for exponential decay, τ, of 21±4.6 s (n=5; Figure 7A⇓, left). The NO donor S-nitroso-N-acetylpenicillamine (SNAP, 10−4 mol/L) added at the same time that [Ca2+]o was eliminated significantly increased the rate at which [Ca2+]i declined, indicating that NO accelerates the removal of Ca2+ from the cytoplasm (τ=8.1±1.1 s, n=5, P<0.04; Figure 7A⇓, right). When Ca2+ was readmitted to the extracellular solution in the presence of SNAP, [Ca2+]i initially rose rapidly and then after a delay of 10 s to 15 s, reached a new plateau that was lower than in the absence of SNAP (Figure 7A⇓, right).
Ca2+ influx stimulated by TG elevated [Ca2+]i to a level similar to that caused by AII. Elimination of [Ca2+]o resulted in a decline in [Ca2+]i at a rate that was not significantly different from that observed in the presence of AII (τ=27±1.5 s, n=4, P>0.3; Figure 7B⇑, left). This suggests that in the absence of Ca2+ influx, [Ca2+]i removal mechanisms other than SERCA are primarily responsible for decreasing [Ca2+]i in the presence of either AII or TG. However, inhibition of SERCA with TG completely prevented the increase in the rate of [Ca2+]i removal caused by SNAP (τ=27±2.6 s, P>0.7; Figure 7B⇑, right). This result supports the concept that NO rapidly decreases [Ca2+]i in the presence of AII primarily by accelerating the removal of cytosolic Ca2+ via SERCA.
To determine if NO accelerates the removal of cytoplasmic Ca2+ by increasing its uptake into intracellular stores, the effect of NO was determined on the Ca2+ content of the stores. Ionomycin (10 μmol/L) alone or in combination with AII (0.1 μmol/L) was used to release Ca2+ from the stores in the nominal absence of [Ca2+]o. Ionomycin alone caused a peak increase in [Ca2+]i from a basal value of 92±9 to 835±72 nmol/L (Figure 8A⇓). Application of NO (1 μmol/L) 4 minutes before had no significant effect on the subsequent response to ionomycin (−7±9% compared to without NO, not shown in figure). AII (0.1 μmol/L) caused a peak transient increase in [Ca2+]i to 591±33 nmol/L and significantly reduced the release caused by the subsequent application of ionomycin (442±22 nmol/L, n=6, P<0.003; Figure 8A⇓). When NO (1 μmol/L) was given 1 minute before AII, the [Ca2+]i transient caused by AII was significantly reduced (434±40 nmol/L, P<0.007), whereas the subsequent ionomycin-induced release was increased by 42±7% compared with that in the absence of NO (631±47 nmol/L, P<0.003; Figure 8B⇓). Even if NO was added 20 s after AII (after the peak increase in [Ca2+]i caused by the agonist), the ionomycin-induced release was 26±9% greater than in the absence of NO (564±63, P<0.04; Figure 8C⇓), consistent with an NO-induced augmentation of the refilling of Ca2+ stores.
Inhibition of the peak increase in [Ca2+]i caused by AII could be achieved by either inhibition of Ca2+ release from the stores7 19 or by accelerated sequestration of Ca2+ into the stores. The role of uptake by SERCA was determined by releasing Ca2+ from the stores with AII in the absence or presence of TG. In the absence of [Ca2+]o, NO (1 μmol/L), added before AII, inhibited the transient increase in [Ca2+]i caused by AII by 44±7% (Figure 9A⇓ and 9B⇓). The peak increase in [Ca2+]i was 528±57 nmol/L compared with 319±52 nmol/L in the presence of NO. Two minutes after treatment with TG, AII caused a transient Ca2+ increase to a level that was not significantly different from that in the absence of TG (Figure 9C⇓ and 9D⇓). However, NO failed to inhibit the AII response when SERCA was inhibited by TG (Figure 9C⇓ and 9D⇓). Similarly, NO inhibited the Ca2+ influx-induced rise in [Ca2+]i caused by releasing the stores with AII but not with AII and TG. This suggests that NO inhibits the peak increase in [Ca2+]i caused by AII primarily by accelerating the sequestration of Ca2+ into the stores via SERCA, rather than by interfering with IP3-induced Ca2+ release.
NO Does Not Affect IP3 Production Caused by AII
The effect of NO on AII-induced IP3 production was measured by radioimmunoassay. Basal levels of IP3 were 0.9±0.1 pmol/plate. During the first 45 s after AII (0.1 μmol/L), IP3 levels were significantly increased (3.1±0.5, 3.1±0.7, and 3.1±0.5 pmol/plate at 15, 30, and 45 s after AII, respectively, n=5) but were not significantly different after 1 minute of treatment with NO (3.4±0.6, 2.7±0.4, 2.2±0.6 pmol/plate, respectively). Furthermore, IP3 levels 60 s and 70 s after AII were 2.3±0.4 and 2.1±0.3 pmol/plate, respectively. When NO was administered at 60 s after AII, the level of IP3 10 s later was not significantly decreased compared with that in the absence of NO (2.0±0.4 pmol/plate). Thus, changes in AII-induced IP3 levels do not explain the effect of NO on [Ca2+]i.
According to the capacitative Ca2+ entry model, Ca2+ influx can be activated not only by emptying the stores with agonists but also by depletion of the stores after inhibition of SERCA with TG or CPA.13 14 15 An important role of SERCA in the physiological response to NO in agonist-stimulated smooth muscle cells is suggested by the present study in which these 2 structurally different, highly specific inhibitors of SERCA20 21 interfered with the actions of NO. The present study also demonstrates that the initial action of NO in agonist-stimulated smooth muscle cells is to rapidly sequester [Ca2+]i via SERCA and secondarily to inhibit Ca2+ influx. This temporal sequence suggests the proposal that NO regulates agonist-induced Ca2+ influx by a 2-step process: (1) NO initiates a rapid decrease in [Ca2+]i by sequestering cytoplasmic Ca2+ into intracellular stores via SERCA, and (2) the refilling of intracellular stores inhibits store-operated Ca2+ entry, which in turn maintains the decrease in [Ca2+]i and relaxation of vascular smooth muscle.
Several pieces of evidence support sequestration of [Ca2+]i by SERCA as the primary mechanism by which NO initiates the decrease in [Ca2+]i. First, NO causes [Ca2+]i to fall more rapidly than can be explained by either directly inhibiting Ca2+ influx with Ni2+ (Figure 6⇑) or by removing [Ca2+]o (Figure 7⇑). Second, NO does not decrease [Ca2+]i when SERCA is inhibited by TG or CPA (Figures 2 through 4⇑⇑⇑; Table⇑). The ability of the inhibitors of SERCA to prevent the response to NO is time-dependent (Figure 3⇑), presumably indicating that the inhibitors require sufficient time to completely block uptake of Ca2+ into the stores. Third, studies in which PMCA or Na+/Ca2+ exchanger was inhibited (Figure 4⇑) indicate that although these mechanisms actively remove [Ca2+]i from the cytoplasm when it is elevated by either AII or TG, they are not essential for NO to decrease [Ca2+]i.
The ability of NO to increase the content of [Ca2+]i stores was made evident by the increased size of the response to ionomycin after NO was administered before the partial emptying of the stores caused by AII (Figure 8⇑). Ionomycin, at the concentrations used, is known to permeabilize membranes to Ca2+, to release all Ca2+ from stores, and to make Ca2+ pumps and transporters ineffective.22 In the absence of [Ca2+]o, an increased amount of Ca2+ released by ionomycin in the presence of NO can only be explained by increased uptake from the cytoplasm or decreased release from the stores. NO increased the ionomycin-releasable store even when NO was administered after Ca2+ release induced by AII was nearly complete, suggesting that NO increased reuptake, rather than inhibited the IP3-mediated release of Ca2+ from the stores. Indeed, the effect of NO on the AII-induced transient increase in [Ca2+]i in the absence of [Ca2+]o was entirely prevented by pretreatment with TG (Figure 9⇑). The latter experiment suggests that NO has little or no direct effect on the production of, or response to, IP3 initiated by AII in these cells. This finding is also supported by the lack of an effect of NO on AII-stimulated IP3 levels. Thus, our results indicate that NO initiates a decrease in [Ca2+]i primarily as a result of its sequestration into intracellular stores by SERCA.
The hypothesis that the inhibition of Ca2+ influx by NO is secondary to its initial effects on sequestration of Ca2+ into the stores by SERCA is based on several observations. The inhibition by NO of AII-induced Mn2+ influx was strictly dependent on [Ca2+]o (Figure 5⇑). This result suggests the apparent paradox that Ca2+ influx is required for NO to inhibit divalent cation influx. Indeed, when [Ca2+]o was added after AII in the presence of NO (Figures 5⇑, 7⇑, and 9⇑), [Ca2+]i initially rose rapidly but then equilibrated at a lower level compared with control. This requirement for Ca2+ influx for NO to inhibit Ca2+ influx is also evident from the delay observed in the NO-induced inhibition of Mn2+ influx (Figure 6⇑). Together with the lack of effects of NO when SERCA is inhibited, these experiments support a requirement for SERCA-dependent uptake of Ca2+ into the stores for divalent cation influx to be inhibited by NO.
Thus, our results can best be explained by proposing that NO indirectly inhibits capacitative or store-operated, agonist-activated Ca2+ influx in vascular smooth muscle cells by increasing Ca2+ in the stores. The observation that inhibitors of SERCA can prevent the actions of NO on [Ca2+]i was made previously in studies on platelets23 and HEK 293 cells,24 but the finding was interpreted as meaning that NO is not capable of regulating capacitative Ca2+ influx. However, the present findings suggest that functional SERCA is an essential requirement for NO to regulate store-operated Ca2+ influx. Although we have not measured a direct effect of NO on SERCA itself, others have shown that nitrovasodilators stimulate the ATPase activity of SERCA by a mechanism dependent on cGMP and protein kinase G.10 Furthermore, cGMP-dependent phosphorylation of phospholamban is induced by NO in vascular smooth muscle.10 25 Dissociation of phosphorylated phospholamban from SERCA would provide a molecular mechanism for NO to regulate SERCA activity and thus to regulate store-operated Ca2+ influx. cGMP-independent effects of NO on SERCA via redox-sensitive thiols are potentially possible, similar to those described recently for K+ channels4 and the ryanodine receptor.26 Indeed, the effects of NO on vascular smooth muscle [Ca2+]i and relaxation persist despite effective inhibition of cGMP and protein kinase G.27
Ca2+ entry in vascular smooth muscle is known to occur via 2 distinct pathways, only one of which is sensitive to inhibitors of L-type Ca2+channels.12 In the present study, nifedipine had little effect on agonist-induced increases in [Ca2+]i and contractions of rabbit or mouse aortic smooth muscle, indicating that non–L-type Ca2+ channels are responsible for the majority of Ca2+ entry in these conduit arteries. Thus, the mechanism proposed in the present study can explain how NO regulates the portion of Ca2+ influx and contractions induced by agonists and growth factors that occurs independently of L-type Ca2+ channels. The voltage-dependent activation of smooth muscle L-type Ca2+ channels also can be regulated by NO either directly3 or indirectly by hyperpolarization mediated by K+ channels.4 5 6 Thus, the proposal that NO regulates agonist-activated, store-operated Ca2+ influx widens the potential mechanisms by which NO causes vasodilatation.
This study was supported by NIH grants HL31607, HL54150, and HL55620.
- Received September 25, 1998.
- Accepted November 3, 1998.
- © 1999 American Heart Association, Inc.
Lincoln TM, Cornwell TL. Intracellular cyclic GMP receptor proteins. FASEB J. 1993;7:328–338.
Robertson BE, Schubert R, Hescheler J, Nelson MT. cGMP-dependent protein kinase activates Ca-activated K channels in cerebral artery smooth muscle cells. Am J Physiol. 1993;265:C299–C303.
Archer SL, Huang JM, Hampl V, Nelson DP, Shultz PJ, Weir EK. Nitric oxide and cGMP cause vasorelaxation by activation of a charybdotoxin-sensitive K channel by cGMP-dependent protein kinase. Proc Natl Acad Sci U S A. 1994;91:7583–7587.
Hirata M, Kohse KP, Chang C-H, Ikebe T, Murad F. Mechanism of cyclic GMP inhibition of inositol phosphate formation in rat aorta segments and cultured bovine aortic smooth muscle cells. J Biol Chem. 1990;265:1268–1273.
Furukawa K, Tawada Y, Shigekawa M. Regulation of the plasma membrane Ca2+ pump by cyclic nucleotides in cultured vascular smooth muscle cells. J Biol Chem. 1988;263:8058–8065.
Furukawa K, Ohshima N, Tawada-Iwata Y, Shigekawa M. Cyclic GMP stimulates Na+/Ca2+ exchange in vascular smooth muscle cells in primary culture. J Biol Chem. 1991;266:12337–12341.
Cornwell TL, Pryzwansky KB, Wyatt TA, Lincoln TM. Regulation of sarcoplasmic reticulum protein phosphorylation by localized cyclic GMP-dependent protein kinase in vascular smooth muscle cells. Mol Pharmacol. 1991;40:923–931.
Luo D-L, Nakazawa M, Ishibashi T, Kato K, Imai S. Putative, selective inhibitors of sarcoplasmic reticulum Ca+(+)-pump ATPase inhibit relaxation by nitroglycerin and atrial natriuretic factor of the rabbit aorta contracted by phenylephrine. J Pharmacol Exp Ther. 1993;265:1187–1192.
Xuan YT, Wang OL, Whorton AR. Thapsigargin stimulates Ca2+ entry in vascular smooth muscle cells: nicardipine-sensitive and -insensitive pathways. Am J Physiol. 1992;262:C1258–C1265.
Cowan CL, Palacino JJ, Najibi S, Cohen RA. Potassium channel mediated relaxation to acetylcholine in rabbit arteries. J Pharmacol Exp Ther. 1993;266:1482–1489.
Weisbrod RM, Griswold MC, Du Y, Bolotina VM, Cohen RA. Reduced responsiveness of hypercholesterolemic rabbit aortic smooth muscle cells to nitric oxide. Arterioscler Thromb Vasc Biol. 1997;17:394–402.
Chen XL, Rembold CM. Cyclic nucleotide-dependent regulation of Mn2+ influx, [Ca2+]i, and arterial smooth muscle relaxation. Am J Physiol. 1992;263:C468–C473.
Plenge-Tellechea F, Soler F, Fernandez-Belda F. On the inhibition mechanism of sarcoplasmic or endoplasmic reticulum Ca2+-ATPases by cyclopiazonic acid. J Biol Chem. 1997;272:2794–2800.
Fasolato C, Pozzan T. Effect of membrane potential on divalent cation transport catalyzed by the “electroneutral” ionophores A23187 and ionomycin. J Biol Chem. 1989;264:19630–19636.
Xu L, Eu JP, Meissner G, Stamler JS. Activation of the cardiac calcium release channel (ryanodine receptor) by poly-S-nitrosylation. Science. 1998;279:234–237.