INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

INTRACELLULAR AND EXTRACELLULAR pH vary in response to respiratory or metabolic impairment. In pulmonary circulation, pH is an essential factor for changes in vascular tone. The literature indicates that acidosis evokes vasoconstriction, whereas alkalosis causes vasodilatation in pulmonary circulation (12, 16, 17). In fact, hypocapnic alkalosis has been widely used to treat severe neonatal and pediatric pulmonary hypertension (4). Previous studies at our laboratory suggest that changes in the diameter of pulmonary vessels under conditions with abnormal partial pressures of CO₂ (PCO₂) are chiefly mediated by cyclooxygenase (COX)-associated vasoactive substances including prostacyclin (PGI₂) (22, 30). Although COX is known to be mainly expressed in the endothelial cells of pulmonary circulation and requires an increase in intracellular Ca²⁺ concentrations ([Ca²⁺]_i) for its activation in producing PGI₂ (8, 19), studies of the essential aspects of Ca²⁺ mobilization within pulmonary endothelial cells in response to CO₂-related environmental pH changes caused by hypercapnic acidosis and hypocapnic alkalosis have been few. In the present study, we attempted to assess the effects of hypercapnic acidosis and hypocapnic alkalosis on [Ca²⁺]_i mobilization in human pulmonary artery endothelial cells (HPAECs) by using fluorescent probes of 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein (BCECF) for intracellular pH (pH_i) and fura 2 for [Ca²⁺]_i indicators. We also studied the significance of Ca²⁺ mobilization from the endoplasmic reticulum and extracellular Ca²⁺ influx for [Ca²⁺]_i changes in response to hypercapnic acidosis and hypocapnic alkalosis. Finally, we estimated the extent of PGI₂ production in HPAECs that was mediated by hypercapnic acidosis and hypocapnic alkalosis.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Endothelial cell culture. Passage 4 HPAECs were purchased (Kurabou, Osaka, Japan) and grown to confluence to passage 7 before experimentation. Thus passage 8 HPAECs were used for analysis in the present study. HPAECs were cultured on coverslips (Matsunami, Tokyo, Japan) for measurement of [Ca²⁺]_i and pH_i. We used Humedia-II (Kurabou) containing 10% fetal calf serum, 10 U/ml penicillin, and 100 µg/ml streptomycin (GIBCO, Grand Island, NY) as the culture medium, which was equilibrated in a humidified atmosphere with 5% CO₂ at 37°C. All measurements were conducted with confluent cell monolayers.

[Ca²+]_i analysis. Transitional changes in [Ca²⁺]_i were measured with fura 2 as an appropriate indicator (14). Cultured cells were incubated with 2 µM fura 2-AM (Dojin Chemical, Kumamoto, Japan) and 0.08% Pluronic F-127 (Molecular Probes, Eugene, OR) in a standard isotonic solution adjusted to pH 7.4 (in mmol/l: 130 NaCl, 5.4 KCl, 0.8 MgSO₄, 1.8 CaCl₂, 1.0 NaH₂PO₄, and 26 HEPES) for 30 min at 37°C. After completion of dye loading, coverslips were washed three times with standard isotonic solution. Dye loading did not change HPAEC morphologies as assessed by light microscopy. Coverslips with fura 2-loaded cells were placed in a flow chamber and mounted on the stage of a microspectrometer connected to a fluorescence microscope (Diaphot T300, Nikon, Tokyo, Japan) with an excitation filter changer and a multipoint imaging system (Argus 50, Hamamatsu Photonics, Shizuoka, Japan). Coverslips were perfused with warmed buffer at 2 ml/min; the temperature of the measurement system was maintained at 37°C. After 20 min of perfusion, cells were alternately excited at two wavelengths, 340 and 380 nm, and resultant light emission was detected at 510 nm. Fluorescence intensity data, F₃₄₀ and F₃₈₀, were analyzed using customized software. The ratio (R) of fura 2 fluorescence intensities excited by two wavelengths (F₃₄₀/F₃₈₀) was calculated after subtraction of background fluorescence at every 15 s. Maximum and minimum fluorescence ratios (R_max and R_min) were determined by using 10 mmol/l of CaCl₂ and EGTA (Dojin Chemical), respectively. [Ca²⁺]_i was calculated based on a formula proposed by Grynkiewicz et al. (14)
where  is the ratio of emission fluorescence at 380 nm in the presence of 10 mmol/l CaCl₂ and EGTA. Measured  values averaged 12.4 in our Ca²⁺ measurement system (n = 5). Because the dissociation constant (K_d) of fura 2 with Ca²⁺ is significantly affected by pH_i changes, it was corrected with the following equation (5)

pH_i analysis. pH_i was measured with BCECF, a fluorescent pH indicator dye (23). BCECF-AM (Molecular Probes) was dissolved in DMSO initially at 2.5 mmol/l and was used at a final concentration of 5 µmol/l in standard isotonic solution. Cells were loaded with BCECF for 30 min at 37°C. To measure pH_i, we applied the excitation wavelengths of 490 and 450 nm and measured emission at 530 nm. The ratio of fluorescence intensity (F₄₉₀/F₄₅₀) was used to assess the change in pH_i. pH_i measurement was calibrated with nigericin (7 µmol/l) containing high-K⁺ solution (40 mmol/l) at different extracellular pH (pH_o) levels. Fluorescence ratio data thus obtained were analyzed by linear regression to calculate pH_i (26). The pH_i was measured at intervals of 15 s, thus allowing us to estimate the K_d value at every 15 s.

Hypercapnic acidosis and hypocapnic alkalosis. Cells were perfused at 2 ml/min on the microscope stage for 20 min with Krebs-Henseleit buffer solution (in mmol/l: 118 NaCl, 4.7 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄7H₂O, 2.5 CaCl₂6H₂O, 25 NaHCO₃, and 5.6 glucose) equilibrated with a gas mixture containing 21% O₂ and 5% CO₂. PCO₂ in the circulating buffer was changed by switching the equilibration gas from normocapnia (21% O₂ and 5% CO₂ in N₂) to either hypercapnia (21% O₂ and 10% CO₂ in N₂, n = 10) or hypocapnia (21% O₂ and 2% CO₂ in N₂, n = 10). Equilibration by hypercapnic gas and hypocapnic gas changed the pH_o from 7.4 to 7.0 and from 7.4 to 7.8, respectively. We examined the transitional changes in pH_i and [Ca²⁺]_i at every 15 s under each experimental condition.

Intracellular acidification and alkalinization. To determine the importance of intracellular acidification in the modulation of [Ca²⁺]_i in hypercapnic acidosis, endothelial cells were perfused with physiological HEPES buffer solution (in mmol/l: 137 NaCl, 5.0 KCl, 1.2 MgSO₄, 1.2 KH₂PO₄, 1.2 CaCl₂, 10 HEPES, and 16 glucose) in which pH was adjusted to 7.4, and then the circulating medium was switched to Krebs-Henseleit solution equilibrated with a gas mixture containing 21% O₂ and 5% CO₂, enabling pH_o to be maintained at 7.4 (n = 7).

Contribution of extracellular Ca²+ and intracellular Ca²⁺ stores. We studied Ca²⁺ sources involved in pH-dependent [Ca²⁺]_i mobilization in HPAECs. To determine the relative contribution of extracellular Ca²⁺, pH-induced [Ca²⁺]_i changes were measured in Ca²⁺-free Krebs-Henseleit solution containing 2 mmol/l of EGTA in hypercapnic acidosis (n = 6) and hypocapnic alkalosis (n = 8). To analyze the importance of intracellular Ca²⁺ stores, extracellular and intracellular Ca²⁺ stores were depleted simultaneously. Extracellular stores were decreased by EGTA and intracellular stores by introducing 100 nmol/l of thapsigargin (TG) (Wako, Osaka, Japan). Under these experimental conditions, we calculated [Ca²⁺]_i changes as a result of hypercapnic acidosis (n = 6) and hypocapnic alkalosis (n = 6).

PGI₂ measurement. HPAECs were cultured in 25-cm² tissue culture flasks and used at confluence. Experiments were conducted in a warm dish at 37°C. The culture medium was replaced with Krebs-Henseleit solution equilibrated with gas containing 5% CO₂ and 21% O₂ (pH_o = 7.4). After 1 h, the supernatant was removed and another 3 ml of Krebs-Henseleit buffer were equilibrated with normocapnic, hypercapnic, or hypocapnic gas; a gas mixture having the same gas composition as used for the buffer equilibration was continuously supplied to the dish. The pH of each supernatant was maintained at 7.38 ± 0.03 (normocapnia, n = 5), 7.01 ± 0.02 (hypercapnia, n = 5), or 7.82 ± 0.04 (hypocapnia, n = 5). Ten minutes later, 0.5 ml of the supernatant was removed to measure PGI₂. All samples were stored at 20°C, and 6-keto prostaglandin F₁ (6-keto-PGF₁), the stable hydrolysis product of PGI_2, was measured by radioimmunoassay (18).

Statistical analysis. Data are presented as means ± SE. We judged the statistical significance of the changes in pH_i and [Ca²⁺]_i when the gas mixture was altered from normocapnia to hypercapnia or hypocapnia by the paired t-test. Differences in pH_i and [Ca²⁺]_i before and after intracellular acidification or alkalinization was introduced were also judged by the paired t-test. Statistical differences in increments of [Ca²⁺]_i in hypercapnic acidosis and in hypocapnic alkalosis were determined by the unpaired t-test. Changes in [Ca²⁺]_i among groups with different medications were compared by applying ANOVA followed by Scheffé's F test. Differences in PGI₂ production among normocapnia, hypercapnia, and hypocapnia were also judged by ANOVA and Scheffé's F test. A P value of <0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of hypercapnia and hypocapnia on [Ca²+]_i. Exposure to hypercapnic acidosis rapidly decreased pH_i from 7.19 ± 0.07 to 7.04 ± 0.11, in association with a modest increase in [Ca²⁺]_i from 112 ± 24 to 157 ± 38 nmol/l (Fig. 1). Exposure to hypocapnic alkalosis increased pH_i from 7.18 ± 0.05 to 7.34 ± 0.10, leading to a marked increase in [Ca²⁺]_i from 114 ± 18 to 359 ± 72 nmol/l. [Ca²⁺]_i-to-pH_i ratios (where  means change) were much larger for hypocapnic alkalosis [(15 ± 4) × 10²] than hypercapnic acidosis [(3.0 ± 0.7) × 10²].

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Examples of intracellular pH (pHi) and intracellular Ca2+ concentration ([Ca2+]i) changes in hypercapnic acidosis and hypocapnic alkalosis. A: pHi changes in hypercapnic acidosis. B: [Ca2+]i changes in hypercapnic acidosis. C: pHi changes in hypocapnic alkalosis. D: [Ca2+]i changes in hypocapnic alkalosis. The increase in [Ca2+]i in hypocapnic alkalosis was significantly greater than that in hypercapnic acidosis (E). *P < 0.05 compared with the values obtained in hypercapnic acidosis.

Effect of intracellular acidification and alkalinization on [Ca²+]_i. Intracellular acidification (pH_i = 0.17 ± 0.03) at pH_o maintained at 7.4 significantly augmented [Ca²⁺]_i by 24 ± 8 nmol/l, whereas intracellular alkalinization (pH_i = +0.38 ± 0.08) at a constant pH_o of 7.4 had little influence on [Ca²⁺]_i (Fig. 2). The increase in [Ca²⁺]_i elicited by intracellular acidification did not differ from that obtained under conditions with hypercapnic acidosis.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effects of intracellular acidification and alkalinization on [Ca2+]i in human pulmonary artery endothelial cells (HPAECs). Intracellular acidification increased [Ca2+]i, but intracellular alkalinization did not. NS, not significant.

Contribution of extracellular Ca²+ and TG-sensitive intracellular Ca²⁺ pool to the increase in [Ca²⁺]_i. Hypercapnic acidosis increased [Ca²⁺]_i by 45 ± 14 nmol/l under control conditions in the absence of EGTA and TG, by 40 ± 13 nmol/l in the presence of EGTA (removal of Ca²⁺ from the perfusate), and by 35 ± 16 nmol/l in the presence of both EGTA and TG (depletion of both extracellular and intracellular Ca²⁺) (Fig. 3). There was no significant difference between the values.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Contribution of extracellular Ca2+ and thapsigargin (TG)-sensitive Ca2+ stores to the increase in [Ca2+]i with hypercapnic acidosis and hypocapnic alkalosis. A: hypercapnia-induced [Ca2+]i changes in the presence of EGTA only or both EGTA and TG. Removal of Ca2+ from the perfusate by EGTA and depletion of intracellular Ca2+ stores by TG did not significantly affect the hypercapnia-induced [Ca2+]i increase. B: hypocapnia-induced [Ca2+]i changes in the presence of EGTA only or both EGTA and TG. Treatment with both EGTA and TG almost abolished the [Ca2+]i increase. *P < 0.05 compared with the control group. #P < 0.05 compared with the EGTA group.

Effect of hypercapnia and hypocapnia on PGI₂ production. Hypercapnic acidosis did not augment endothelial PGI₂ production compared with that obtained under normocapnic conditions (Fig. 4). However, hypocapnic alkalosis markedly augmented PGI₂ production in pulmonary artery endothelial cells.

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Effect of hypercapnic acidosis and hypocapnic alkalosis on 6-keto-PGF1 production in HPAECs. Hypercapnia did not increase 6-keto-PGF1 production, whereas hypocapnia increased it significantly. *P < 0.05 compared with normocapnia.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Critique of methods. In the present study, we used passage 8 HPAECs for assessing the relationship between pH_i and [Ca²⁺]_i under various experimental conditions. To see whether differences in the passage of HPAECs would exert a significant influence on [Ca²⁺]_i mobilization, we preliminarily examined the hypercapnia- and hypocapnia-elicited [Ca²⁺]_i kinetics in HPAECs ranging from passage 8 to passage 12. [Ca²⁺]_i kinetics observed for these HPAECs with various passages did not differ quantitatively, indicating that the difference in passage would have little effect on [Ca²⁺]_i mobilization evoked by hypercapnic acidosis and hypocapnic alkalosis.

Differential effects of hypercapnic acidosis and hypocapnic alkalosis on intracellular Ca²+ mobilization in endothelial cells. Although many studies have focused on the significance of acidosis or alkalosis in Ca²⁺ mobilization in vascular endothelial cells, most of them addressed aortic endothelial cells rather than pulmonary endothelial cells (9, 11, 25, 27, 28, 32). Furthermore, findings obtained for aortic endothelial cells are mutually qualitatively inconsistent. For instance, Ziegelstein et al. (32) found that intracellular acidification enhanced [Ca²⁺]_i, whereas intracellular alkalinization reduced it in cultured rat aortic endothelial cells. On the other hand, Danthuluri et al. (9) reported findings opposing those of Ziegelstein et al., suggesting that intracellular alkalinization increased Ca²⁺ in bovine aortic endothelial cells. In regard to which one is important in mobilizing Ca²⁺ (pH_i or pH_o), some researchers demonstrated that the reduction of pH_o, but not pH_i, was the essential factor regulating [Ca²⁺]_i in canine coronary artery endothelial cells subjected to severe acidosis (11, 25). These researchers also showed that intracellular alkalinization increased [Ca²⁺]_i in bovine aortic endothelial cells (11, 25). On the other hand, Wakabayashi and Groschner (28) reported that the extracellular elevation of pH was the primary cause increasing [Ca²⁺]_i in vascular endothelial cell line ECV 304. Experimental findings reported in the literature thus do not provide a confident conclusion on the regulatory mechanism of [Ca²⁺]_i mobilization by pH even in endothelial cells originating from systemic circulation. No study has approached the interrelationship between pH_o and/or pH_i environments and [Ca²⁺]_i regulation in pulmonary endothelial cells, which have functions that are expected to be qualitatively different from those of endothelial cells in systemic circulation (22, 30). Nitric oxide synthase and COX expressed in endothelial cells are equivalently important enzymes for regulating vascular tone in systemic circulation in response to acidosis and alkalosis (1). Endothelial COX, however, appears to be much more important than endothelial nitric oxide synthase in maintaining vascular tone in pulmonary circulation when environmental pH is changed (22, 30). Based on these facts, we systematically analyzed the quantitative effects of acidosis and alkalosis implemented by changing CO₂ concentrations (i.e., hypercapnic acidosis and hypocapnic alkalosis) on [Ca²⁺]_i and its effects on yielding COX-related vasoactive substance of PGI₂ in pulmonary endothelial cells originally obtained from human pulmonary arteries. PCO₂ varies significantly from region to region even in normal lungs (29). The regional variation of PCO₂ modifies the constrictive state of pulmonary vessels there (13), firmly indicating that it is important to analyze the mechanisms involved in regulating [Ca²⁺]_i under conditions with hypercapnic acidosis and hypocapnic alkalosis in pulmonary endothelial cells.

Physiological significance. To assess the importance of endothelial [Ca²⁺]_i changes in modulating vascular tone in pulmonary circulation under hypercapnic and hypocapnic conditions, we estimated the effects of hypercapnia and hypocapnia on the production of PGI₂, which is expected to be an essential vasodilating prostaglandin yielded by COX (Fig. 4). As discussed above, we confirmed in previous studies that COX-mediated vasoactive substances play a significant role in modifying pulmonary vascular tone under conditions with varied pH induced by changes in CO₂ concentrations (22, 30). As presumed from the large difference in the increment of [Ca²⁺]_i (Fig. 1), hypocapnic alkalosis produced PGI₂ about 10 times as much as hypercapnic acidosis (Fig. 4). Enhanced PGI₂ production may support the hypocapnia-elicited vasodilatation consistently observed in pulmonary circulation (4, 12, 13). Several groups of investigators showed that hypercapnic acidosis generally evoked pulmonary vasoconstriction (6, 10, 16, 17, 31), which would not appear to be at variance with a small amount of PGI₂ produced on the stimulation of hypercapnic acidosis observed in the present study. However, the small amount of PGI₂ does not explain the positive vasoconstriction of pulmonary vasculature in hypercapnic acidosis. Ahn and Hume (3) and Berger et al. (7) ascertained that acidosis-elicited constriction of pulmonary arterial smooth muscle cells was mediated through the inhibition of voltage-dependent K⁺ channels. Combining our findings with the electrophysiological results reported by the above authors (3, 7), pulmonary vasoconstriction induced by hypercapnic acidosis is attributable to the direct effect of acidosis on smooth muscle cells, whereas the significance of endothelial function modified by acidosis may be trivial for hypercapnia-related pulmonary vasoconstriction.