INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CHANGES IN OXYGEN TENSION (PO₂) profoundly affect the normal physiology of an organism. Systemic vascular smooth muscle (VSM) relaxes in response to acute hypoxia or decrease in PO₂ (6). Vasodilation to hypoxia in systemic VSM is likely a protective response to increase blood flow to areas of low PO₂. A variety of theories have been proposed to explain the mechanism of hypoxia-induced vasodilation, such as release of vasoactive substances (25, 29), activation of ATP-sensitive K⁺ (K⁺_ATP) channels (5, 18), and energy limitation due to a limited anaerobic metabolic capacity (4, 10, 16). The last is unlikely in light of evidence that suggests relaxation in response to hypoxia occurs without compromising energy metabolism (26). It has also been shown that hypoxic vasorelaxation is insensitive to various ion channel and metabolic antagonists (3, 26). Recently we showed that mechanisms involving reduction in ATP, decrease in pH, increased P_i, or activation of K⁺_ATP or calcium-activated potassium (K_Ca) channels cannot explain hypoxic relaxation in pig coronary artery (26). Moreover, both Ca²⁺-dependent and Ca²⁺-independent mechanisms were required to account for hypoxia-induced vasodilation.

The role of intracellular calcium concentration ([Ca²⁺]_i) handling in hypoxic sensitivity has been studied extensively. Hypoxia-induced pulmonary vasoconstriction has been linked to increases in [Ca²⁺]_i attributed to release from intracellular stores (17) and inhibition of sarcolemmal K⁺ currents (15, 22), both of which activate Ca²⁺ channels. Ca²⁺ modulation in hypoxic vasodilation of systemic VSM is less clear. Although several studies have shown an inhibition of L-type Ca²⁺ channels by hypoxia (8, 13, 24), our studies (26) and other (1) indicate that hypoxia-induced relaxation can occur independently of the decreases in [Ca²⁺]_i that are normally associated with smooth muscle vasodilation. Still others report dilation of arterial smooth muscle via K_Ca channel activation by release of Ca²⁺ from intracellular stores (21). Although Ca²⁺ modulation by hypoxia has been studied extensively, no consistent theory has emerged.

Hypoxia is known to cause an increase in coronary blood flow in vivo (7). Although many studies have demonstrated this effect of hypoxia on small coronary resistance vessels, hypoxia-induced vasodilation of larger coronary arteries has been shown both in vivo and in vitro (6, 6a, 7) and may be of physiological significance (30). This relaxation response to hypoxia can be demonstrated in isolated porcine coronary arterial rings (Fig. 1). Recently, we observed that organ culture severely inhibits coronary relaxation to hypoxia, whereas contractility appears normal (27). In this study, we characterize the effects of organ culture in detail to probe further the mechanisms of O₂ sensing. We demonstrate that 1) the inhibition of relaxation is specific for hypoxia, 2) loss in ability to respond to hypoxia is linked to changes in [Ca²⁺]_i handling properties after organ culture, and 3) Ca²⁺-independent components of vasorelaxation are also impaired. Thus organ culture may provide a novel way to investigate the mechanisms of hypoxic vasorelaxation.

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Typical recording of a freshly isolated porcine coronary arterial ring in response to KCl stimulation (40 mM) and hypoxia (95% N2-5% CO2). The right axis bar shows %O2 as determined by the output of an O2 electrode in the bath and calibrated as 100% = 760 mmHg.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Coronary Artery Culture Conditions

Organ Bath Studies

8

9

5

The hypoxic response was characterized in terms of the maximum hypoxic relaxation, expressed as a percentage of the initial isometric force, and the time course, quantified by time to half-maximal relaxation (t_1/2). All organ bath measurements were recorded using a digital data acquisition system (Biopac). Force was normalized to cross-sectional area [F/A = (change in force × circumference)/(2 × wet weight)].

Intracellular Calcium Imaging

[Ca²+]_i calibration. The fluorescent emission at 510 nm for intensity at 340 nm excitation was divided by that measured at 380 nm, and this ratio was used as an index of [Ca²⁺]_i. For statistical analysis, the ratio was assigned values of 0% for resting muscle and 100% for tissue stimulated with 40 mM KCl. This protocol was chosen as a general routine over absolute calibration of the fura 2 fluorescence in lieu of the major uncertainties in terms of absolute calibration (2). Using standard calibration procedures (9), we previously reported values of 50.4 ± 17.2 nM and 441 ± 163 nM for [Ca²⁺]_i under baseline and KCl simulation, respectively (2, 20). In this limited range, the 340/380 ratio is nearly linearly dependent on [Ca²⁺]_i as reported by others (19).

Statistical Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Hypoxia-induced Vasodilation in Fresh, Cold, Stored, and Cultured Arteries

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Effect of differential organ culture on porcine coronary artery. Cultured rings (bottom) have an attenuated hypoxia-induced relaxation relative to control (top, n = 24). Tracings show relaxation to hypoxia during 40 mM KCl (left) and 1 µM U-46619 (right) activation.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   A summary of the effects of organ culture on maximum isometric force generation and hypoxia-induced vasodilation for 40 mM KCl (top) and 1 µM U-46619 (bottom) contractions. B and D: maximum relaxation in response to acute hypoxia (15-20 min) is attenuated by 40-60% after organ culture at 37°C. A and C: the ability to generate isometric force is not significantly altered by this treatment. Results represent the means ± SE for 12 experiments. *P < 0.05 for differences between organ cultured and control.

The reduction in hypoxic relaxation after organ culture occurred without any significant change in maximum force developed in response to either KCl or U-46619 (Fig. 3, A and C). Figure 4 shows the average time course of isometric force after bubbling with N₂ from experiments such as that shown in Fig. 1. Organ culture can be seen to result in a significant decrease in the rate of hypoxia-induced vasodilation and increased t_1/2 compared with control.

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Average (n = 12) time course of isometric force after bubbling with N2 from experiments such as that shown in Fig. 1. The ordinate represents the amount of force remaining after imposition of hypoxia. Organ culture can be seen to result in a significant decrease in the rate of hypoxia-induced vasodilation and to increase the half-maximal time (t1/2) compared with control. Time course for KCl (A) and U-46619 (B) activation are both slowed by about 50% (n = 12). C: average t1/2 for hypoxic relaxation in control and cultured arteries for 40 mM KCl and 1 µM U-46619 activation. Bar graph shows the means ± SE for 8 (KCl) and 6 (U-46619) experiments. *Differences were significant at P < 0.05.

Figure 5 shows the effects of the duration of time under control or culture conditions on the hypoxic relaxation. Isometric force generating capability with depolarization or receptor-mediated activation in either group was not affected until 30 h of storage (Fig. 5, A and C). In contrast, reduced relaxation to hypoxia occurred as early as 6 h with KCl stimulation and 12 h with U-46619 stimulation in cultured rings. Control rings exhibited a small attenuation in relaxation to hypoxia at 30 h compared with shorter durations of storage when stimulated with 40 mM KCl (Fig. 5B). There was no loss of relaxation to hypoxia with U-46619 stimulation in control rings (Fig. 5D).

View larger version (38K): [in this window] [in a new window]   Fig. 5.   Effects of time in storage on control and cultured coronary arterial rings. The ability to generate isometric force was not significantly altered in response to 40 mM KCl (A) and 1 µM U-46619 (C) stimulation. Only after 30 h did control or cultured rings exhibit a decline in force. Relaxation response to hypoxia was significantly different between control and culture groups as early as 12 h (B and D). Bar graph shows the means ± SE for 8 (KCl) and 6 (U-46619) experiments. *Differences were significant at P < 0.05.

We investigated whether organ culture impaired relaxation in general by studying the response to various smooth muscle vasodilators. Figure 6 shows the average force remaining after increasing concentrations of SNP, isoproterenol, and forskolin. Sensitivity to increasing concentrations of the nitric oxide donor SNP was not significantly different between experimental groups (Fig. 6C). Similarly, organ culture did not significantly affect the sensitivity of coronary rings to receptor-mediated (isoproterenol) or direct (forskolin) activation of adenylate cyclase (Fig. 6, A and C). Thus the ability of porcine coronary arterial rings to relax to G- or A-kinase mediated pathways was not affected by organ culture at 37°C for 24 h. Analysis of the decrease in isometric force upon washout of agonist (Fig. 7) indicates that organ culture also does not alter this mechanism of relaxation as well.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Concentration-response curves for sodium nitroprusside (A), forskolin (B), and isoproterenol (C). The ordinate represents the average force remaining after treatment with increasing concentrations of each drug. Results indicate no change in G- or A-kinase-mediated smooth muscle relaxation pathways after organ culture at 37°C. There was no significant difference in sensitivity to these vasodilators at any concentration used (n = 8).

View larger version (11K): [in this window] [in a new window]   Fig. 7.   A: typical recording of a control and cultured arterial ring stimulated with 40 mM KCl and rinsed with fresh physiological salt solution under normoxic conditions. B: summary of t1/2 to relaxation for 8 experiments. The bar graph shows the means ± SE.

We recently showed that with high stimulus levels hypoxic vasorelaxation occurred without a decrease in [Ca²⁺]_i (26). The response to hypoxia under high and low stimulus conditions thus gives an index of the relaxation's dependence on [Ca²⁺]_i. Figure 8 illustrates that in all cases (low and high activation, U-46619 and KCl), organ-cultured arteries had a significantly reduced relaxation in response to hypoxia. The largest difference was seen with 20 mM KCl stimulation (lane 4), a low-level depolarization, where cultured arteries relaxed 60% less than control. Interestingly, the smallest difference between control and cultured arteries was observed during low-level receptor-mediated stimulation, 0.1 mM U-46619 (lane 3). These data suggest that both Ca²⁺-dependent and -independent components of hypoxia-induced relaxation are affected by organ culture.

View larger version (15K): [in this window] [in a new window]   Fig. 8.   Hypoxia-induced relaxation during high (1 µM U-46619 and 40 mM KCl) and low (0.1 µM U-46619 and 20 mM KCl) activation. Organ culture at 37°C for 24 h significantly reduces relaxation response to hypoxia regardless of stimulation type or level. Bar graph shows means ± SE; n = 8-12. *Differences were significant at P < 0.05.

Effects of Protein Synthesis Inhibition on Relaxation in Response to Hypoxia

View larger version (19K): [in this window] [in a new window]   Fig. 9.   The effects of 10 µM cycloheximide (chx) incubation on force (A) and hypoxic relaxation (B) for control and cultured porcine coronary arteries. Inhibition of protein synthesis causes an attenuated relaxation to hypoxia in control arterial rings but no further changes in cultured rings. Bar graph shows means ± SE; n = 8. Differences were significant at P < 0.05 (*control vs. cultured; §control + chx vs. control  chx).

Intracellular Calcium Measurements

View larger version (23K): [in this window] [in a new window]   Fig. 10.   Calcium tracings during hypoxia on 40 (A) and 20 (B) mM KCl contractures for a fresh coronary ring. Although hypoxia decreases force in either case (not shown in this figure), intracellular calcium concentration ([Ca2+]i) is decreased only at 20 mM (B).

View larger version (35K): [in this window] [in a new window]   Fig. 11.   A: simultaneous force and Ca2+ measurements showing a typical tracing of a coronary artery after organ culture that has lost the ability to reduce [Ca2+]i during hypoxia under low stimulus levels (20 mM KCl). B: consistent with freshly isolated vessels, there is a concomitant decrease in [Ca2+]i during hypoxia with low activation in control vessels.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this investigation we show that organ culture at 37°C for 24 h significantly reduces the relaxation response to hypoxia in porcine coronary artery. The ability to generate control levels of isometric force after organ culture suggests that significant smooth muscle cell alteration does not occur. Moreover, the ability to respond to both depolarization (KCl) and receptor-mediated (U-46619) contractile stimuli suggests that the machinery necessary for these two different excitation pathways is not altered by organ culture. These results are quite different from those of cultured smooth muscle cells where both contractility and activation pathways can be significantly altered (8, 13, 24).

The attenuation of the hypoxic relaxation was dependent on the duration of organ culture (Fig. 5), with a reduction compared with control seen as early as 6 h. In control rings, a significant reduction in the hypoxic relaxation did not occur until 30 h. The importance of this observation is less clear, as there was also a significant drop in force generation at 30 h as well, possibly simply reflecting tissue degradation. Although control rings showed some minor loss in relaxation in response to hypoxia at various time points compared with earlier ones, they always relaxed significantly more than cultured rings. These results suggest that the specific attenuation of hypoxic vasorelaxation depends both on time and storage condition. The most consistent results in terms of a maximal loss of relaxation response to hypoxia without a concomitant loss of force production were obtained with a duration of 24 h, which is when all other experiments were performed.

Importantly, relaxation in response to activation of cGMP and cAMP-mediated vasodilatory pathways was not affected by organ culture. Sensitivity to these pathways activated directly (SNP, forskolin) or through receptor stimulation (isoproterenol) was not significantly different (Fig. 6). This evidence implies that organ culture does not impair the innate ability of coronary arteries to respond to the major smooth muscle pathways of relaxation. Therefore, the loss in hypoxic relaxation in these tissues cannot be attributed to a generalized loss in inherent relaxation capability. Moreover, this also suggests that O₂ sensing does not involve the effector pathways downstream from A- or G-kinase activation. This is consistent with previous studies based on pharmacological approaches (3, 26). The idea that organ culture specifically reduces hypoxia-induced vasodilation is strengthened further by the fact that cultured vessels will relax when the stimulus is removed (Fig. 7).

We previously identified both Ca²⁺-independent and -dependent components of the mechanism(s) underlying hypoxic relaxation in coronary arteries (24, 26). The Ca²⁺-dependent component is visible at lower K⁺ stimulus levels such as 20 mM KCl (Figs. 8 and 9) or at low-level receptor-mediated stimulation with 0.1 µM U-46619 (Fig. 8). In response to hypoxia, both freshly dissected coronary artery and those stored at 4°C (control) decrease [Ca²⁺]_i when activated at low levels (Fig. 10). However, after 24 h in organ culture, both the decrease in [Ca²⁺]_i (Table 1) and the decrease in force (Fig. 8) in response to hypoxia are lost. A reduction in the capacity to lower [Ca²⁺]_i could directly translate into a reduced ability to relax in response to hypoxia in cultured arteries. The fact that relaxation in response to washout of agonist is not affected by organ culture (Fig. 7) suggests that this impairment of Ca²⁺ handling appears to be specific to hypoxia-induced vasodilation. Our results indicate that both Ca²⁺-independent and -dependent mechanisms are markedly reduced after organ culture (Fig. 8).

At high stimulus intensities, 40 mM KCl or 1 µM U-46619, little change in [Ca²⁺]_i occurs during acute hypoxic challenge in fresh or control arteries despite a significant decrease in force (Fig. 8, lanes 1 and 2). In cultured arteries under these high-intensity stimulus conditions, the relaxation response to hypoxia is severely inhibited (Fig. 8). Our results thus indicate that both Ca²⁺-independent and -dependent mechanisms of the relaxation response to hypoxia are markedly reduced after organ culture (Fig. 8).

Organ culture elicits a very specific reduction in the relaxation response to hypoxia, which may help delimit the underlying O₂-sensing mechanisms. As the mechanisms regulating smooth muscle contraction are complex, mechanisms of hypoxic relaxation would be expected to be dependent on the nature of the stimulus. It is of interest to note that with KCl stimulation, the hypoxia relaxation in control arteries is attenuated at high stimulus levels (40 mM) relative to low (20 mM). This suggests that K⁺ channels, some classes of which are known to be sensitive to oxygen tension (15, 22), may be involved, because their effects would be expected to decrease with increasing KCl depolarization. The loss of the hypoxia-induced decrease in [Ca²⁺]_i could reflect a loss of K⁺ channels or Ca²⁺ channels, which have also been shown to be oxygen sensitive (8a). This would be consistent with the reduction of the hypoxic relaxation after cycloheximide treatment in control arteries (Fig. 9).

With U-46619 stimulation, one might anticipate that both Ca²⁺-dependent mechanisms, as do K⁺ channel activation and Ca²⁺-independent mechanisms, could play a role in hypoxic relaxation. This would be expected particularly at low intensity (0.1 µM U-46619), for which decreases in [Ca²⁺]_i with hypoxia have been observed. But the magnitude of hypoxic relaxation at 0.1 and 1 µM was similar (Fig. 8), thus the effects are not additive in any simple fashion. This is not surprising, because relaxation is complex, particularly with U-46619, when Ca²⁺ sensitization of the contractile elements is likely (14).

There is also a change in the Ca²⁺-independent component of relaxation in response to hypoxia. Again, this could be due to an decrease or increase in some factor(s) not involved in Ca²⁺ handling, but the evidence here suggests the loss of function by degradation. Many genes and proteins change during 24 h of organ culture of VSM (28). Presumably an alteration in contractile proteins, heat shock proteins, or redox proteins could make the tissue less responsive to hypoxia, directly or indirectly. With the recent development of molecular biology techniques such as subtraction cDNA libraries and proteomics for identification of differential mRNA or protein changes, the specific loss of hypoxic relaxation may provide a new approach to elucidating the mechanism(s) of hypoxic vasorelaxation.

In this investigation summary, our results show that hypoxic vasorelaxation is lost after 24 h in organ culture at 37°C, while contractility and other relaxation pathways are unaffected. Thus organ culture may provide a tool for identifying new targets for understanding the elusive mechanisms underlying this important physiological response.