INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE MECHANISM of hypoxic pulmonary vasoconstriction (HPV) first observed by Euler and Liljestrand (18) remains unclear. Recent evidence (24, 34, 41) strongly suggests that HPV is elicited through direct modulation of K⁺ channels, leading to membrane depolarization. K⁺ channels have been identified in pulmonary vascular smooth muscle cells (13, 30, 37) and are reported to be involved in the regulation of vascular tone (15, 21). Activation of these channels causes an increase in K⁺ efflux, membrane hyperpolarization, inhibition of Ca²⁺ influx, and subsequent vascular smooth muscle relaxation. Recently, K⁺-channel inhibition has been implicated as a critical event in the initiation of HPV (33, 34, 41). Studies (34, 41) provided evidence that hypoxia inhibits whole cell macroscopic K⁺ currents that are sensitive to 4-aminopyridine (4-AP), causing depolarization of the resting membrane potential in isolated pulmonary vascular smooth muscle cells. Post et al. (33) reported that hypoxia inhibited the delayed rectifier K⁺ channel, and Yuan et al. (41) found that the ATP-sensitive K⁺-channel opener cromaklim inhibited hypoxic vasoconstriction in pulmonary arterial rings, an effect reversed by the ATP-sensitive K⁺-channel blocker glibenclamide (Glib). Wiener et al. (40) reported that ATP-sensitive K⁺ channels modulated pulmonary vasoconstriction to hypoxia in isolated ferret lungs, and Cornfield et al. (15) concluded that ATP-dependent K⁺ channels are present but not operative in the fetal pulmonary circulation, which is a low oxygen tension environment. Post et al. (34) observed that hypoxia inhibited the Ca²⁺-activated K⁺ channel in canine pulmonary vascular smooth muscle and suggested that K⁺-channel inhibition is a key event linking hypoxia to pulmonary vasoconstriction by causing membrane depolarization and subsequent Ca²⁺ entry. In addition, Albarwani and Nye (2) suggested that hypoxia inhibited both Ca²⁺-activated and ATP-sensitive K⁺ channels.

Although it has been established that hypoxia induces a rise in pulmonary vascular resistance in hypoxic regions of the lung, controversy exists regarding which vascular segments constrict to hypoxia. Hakim (19) reported that hypoxia constricted primarily the arterial segments, with a lesser effect on the venous segments, whereas Hillier et al. (22) observed that hypoxic venoconstriction occurred in the small postcapillary venules. Shirai et al. (36) concluded that alveolar hypoxia induced vasoconstriction in small pulmonary arteries and veins, with the larger effect occurring in the arteries, and Nagasaka et al. (27) suggested that hypoxia elicited constriction in small-artery segments. The importance in the identification of the specific vascular segments responding to hypoxia relates to 1) the effect on pulmonary capillary pressure (P_pc), which is determined by the distribution of vascular resistance in the pulmonary arteries and veins and 2) the maintenance of ventilation-perfusion matching by constriction of pulmonary arteries that direct blood flow away from hypoxic regions (22).

In light of these previous investigations that appear to establish a relationship between K⁺ channels and HPV, the present study was done to determine the role of K⁺-channel modulation on the effect of hypoxia on pulmonary vascular resistance and compliance in isolated blood-perfused dog lungs. The vascular occlusion technique was used to partition the pulmonary circulation into segmental resistances and compliances. Measurements generated by these occlusion techniques are based on the assumption that the pulmonary circulation is represented by a resistance-compliance circuit. In the present study, the compartmental model of pulmonary vascular resistance and compliance by Audi et al. (5) was used to determine the effect of hypoxia on segmental vascular resistance and compliance in the canine pulmonary circulation. In this particular model, which is relatively simple and more robust than other models (5), the pulmonary circulation is represented as a 3C2R circuit composed of precapillary (R₁) and postcapillary (R₂) resistances and arterial (C₁), middle-compartment (C₂), and venous (C₃) compliances. This model is based on the assumption that after arterial oclusion (AO) and before the arterial pressure curve [P_a(t)] falls below the equilibrium pressure obtained by simultaneous occlusion of both the arterial inflow and venous outflow cannulas [double-occlusion pressure (P_do)], the P_a(t) curve is determined primarily by the product of C₁ and R₁ (5). With the use of this assumption, equations were derived from the data obtained from all three occlusions (AO, venous occlusion, and double occlusion) that were used to calculate the pulmonary vascular resistances and compliances represented in the model (5). These occlusion techniques have previously been used to measure the pulmonary vascular resistance-compliance profile in normal lungs and in lungs challenged with hypoxia (25). Specifically, the role of ATP-sensitive K⁺ channels was investigated with cromakalim (Crom; an activator of ATP-sensitive K⁺ channels) and Glib (a blocker of ATP-sensitive K⁺ channels) to determine if ATP-sensitive K⁺ channels modulate the vasoconstrictor response to hypoxia. In addition, the Ca²⁺-activated K⁺-channel blocker tetraethylammonium (TEA) and the delayed rectifier K⁺-channel inhibitor 4-AP were used to determine whether these specific K⁺ channels also potentiate the pulmonary vascular response to hypoxia.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Adult, heartworm-negative mongrel dogs of either sex (15-19 kg) were anesthetized with pentobarbital sodium (30 mg/kg), intubated, and ventilated with a Harvard respirator with room air at a tidal volume of 15 ml/kg. A left thoracotomy was performed through the fifth intercostal space. The left upper and middle lobes of the lung were removed, and the lower left lobe was prepared for isolation by placing loose ligatures around the left main pulmonary artery and lower left bronchus. Each animal was then heparinized (10,000 U iv) and after 5-10 min was rapidly bled through a carotid arterial cannula. Three hundred milliliters of blood were used to prime the perfusion apparatus. After bleeding was completed, the pulmonary artery was ligated, and with the heart still beating, the lower left lobe with the attached left atrial appendage was rapidly excised and weighed. Plastic cannulas were secured in the lobar artery, lobar vein, and bronchus, and blood perfusion was started within 30 min of lung excision.

The isolated lung circuit has been previously described in detail (7-9, 17, 31) and is well established in this laboratory. Briefly, the lung was perfused at a constant flow by a roller pump (Master Flex, Cole-Parmer) that pumped blood from a venous reservoir through a heating coil encased in a water jacket (37.5 ± 0.5°C) to the rest of the closed circuit. The blood was continuously bubbled with a gas mixture of 95% O₂-5% CO₂ to maintain blood gases in a normal range (arterial PO₂ = 100-110 Torr and arterial PCO₂ = 30-40 Torr) as well as maintain a normal pH. After initial hyperinflation, airway pressure (P_aw) was set at 3 cmH₂O.

The perfused lobe was placed on a weighing pan that was counterbalanced by a strain-gauge transducer (Grass FT-10). Pulmonary arterial (P_pa) and venous (P_pv) pressures were measured by inserting catheters into the lobar artery and vein and connecting them to pressure transducers (Statham 23BC) positioned at the openings of the inflow and outflow cannulas. Pressures were zeroed at the level of the lung hilus. Blood flow (Q) was measured by an electromagnetic flow probe (Carolina Medical SF 300A) positioned in the venous outflow line that was connected to a digital flowmeter (Carolina Medical 701D). P_pa, P_pv, and lung weight were recorded on a Grass polygraph (model 7F). P_pa and P_pv were initially adjusted so that the lung lobe became isogravimetric, i.e., neither gaining nor losing weight in zone III conditions (P_pa > P_pv > P_aw).

P_pc. P_pc was determined with the double-occlusion technique (39). When both arterial and venous cannulas are simultaneously occluded, P_pa and P_pv quickly equilibrate to the same pressure (P_pc). If P_pa and P_pv did not equilibrate exactly to the same pressure on double occlusion, then the mean of both pressures was determined and defined as P_pc. The occlusion pressures were consistently within 1 cmH₂O of each other, and it has been shown that P_do is an excellent estimate of P_pc (39).

(1)

(2)

(3)

(4)

(5)

(6)

(7)

1

1

5

5

5

5

4

5

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Figure 1 shows the effect of Ca²⁺-activated K⁺-channel, ATP-sensitive K⁺-channel, and delayed rectifier K⁺-channel modulation on the P_pa response. Under normoxia (control), TEA, Glib, and 4-AP elicited a small but significant increase in P_pa. As shown in Fig. 2, hypoxia elicited a significant increase in P_pa, an effect potentiated when the Ca²⁺-activated K⁺ channels were blocked by TEA, the ATP-sensitive K⁺ channels were blocked by Glib, and the delayed rectifier K⁺ channels were inhibited by 4-AP. In contrast, opening the ATP-sensitive K⁺ channels with Crom inhibited the pressor response to hypoxia.

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Effect of tetraethylammonium (TEA), glibenclamide (Glib), and 4-aminopyridine (4-AP) on pulmonary arterial pressure response () under normoxia. Values are means ± SE; n = 5 lungs/group. * Significantly different from zero, P < 0.05.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Effect of TEA, Glib, 4-AP, and cromakalim (Crom) on pulmonary arterial pressure response to hypoxia (HYPOX). Values are means ± SE; n = 5 lungs for all groups except n = 6 lungs for HYPOX group. Significantly different (P < 0.05) from: * zero; ** all other treatment groups.

Figures 3 and 4 show the effects of the treatment groups on pulmonary arterial and venous resistances, respectively. Figure 3 shows that hypoxia significantly increased pulmonary arterial resistance, an effect potentiated by TEA, Glib, 4-AP, and TEA+Glib. In contrast, Crom inhibited the hypoxic increase in pulmonary arterial resistance during hypoxia. The effect of hypoxia on pulmonary venous resistance is shown in Fig. 4. Hypoxia significantly increased postcapillary resistance, which, in contrast to precapillary resistance, was not potentiated by either TEA, Glib, 4-AP, or TEA+Glib. However, pretreatment with Crom did block the initial pulmonary venoconstrictor response to hypoxia.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Effect of TEA, Glib, and Crom on pulmonary arterial resistance response to hypoxia. Control groups represent no treatment under normoxic conditions. Values are means ± SE; n = 5 lungs for all groups except n = 6 lungs for HYPOX group. Significantly different (P < 0.05) from: * respective control group; + HYPOX group; ** all other treatment groups.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Effect of TEA, Glib, 4-AP, and Crom on pulmonary venous resistance response to hypoxia. Control groups represent no treatment under normoxic conditions. Values are means ± SE; n = 5 lungs for all groups except n = 6 lungs for HYPOX group. * Significantly different from respective control group, P < 0.05.

The effect of hypoxia on P_pc, which is determined by the distribution of precapillary and postcapillary resistance, is presented in Fig. 5. Hypoxia significantly increased P_pc, a phenomenon that was blocked by Crom. Closing of the Ca²⁺-activated K⁺ channels with TEA, blocking the ATP-sensitive K⁺ channels with Glib, using TEA+Glib, or inhibiting the delayed rectifier K⁺ current with 4-AP did not significantly increase P_pc relative to the initial increase in P_pc observed during hypoxia. With respect to voltage-dependent Ca²⁺-channel function, Fig. 6 shows that verapamil inhibited the hypoxic pressor response on both the arterial and venous segments. In addition, a small number of experiments showed that verapamil also inhibited the potentiated hypoxic response to K⁺-channel inhibition by all three K⁺-channel blockers used in this study (data not shown). These results indicate that blockade of L-type voltage-dependent Ca²⁺ channels modulates hypoxic vasoconstriction.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effect of TEA, Glib, 4-AP, and Crom on pulmonary capillary pressure response to hypoxia. Control groups represent no treatment under normoxic conditions. Values are means ± SE; n = 5 lungs for all groups except n = 6 lungs for HYPOX group. * Significantly different from respective control group, P < 0.05.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Effect of verapamil on pulmonary arterial and venous resistances during hypoxia. Values are means ± SE; n = 5 lungs.

Table 1 summarizes the effect of hypoxia on pulmonary segmental vascular compliance. Hypoxia significantly decreased total vascular compliance by lowering both middle-compartment and large (arterial and venous)-vessel compliances, effects potentiated by TEA, Glib, or 4-AP but subsequently reversed by Crom. Specifically, TEA, Glib, or 4-AP significantly potentiated the decrease in middle-compartment compliance, and TEA and Glib together potentiated the effect on the middle-compartment, arterial, and venous compliances. In contrast, Crom or verapamil inhibited the effect of hypoxia on arterial, venous, and middle-compartment compliances.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

In this study, hypoxia increased the precapillary and postcapillary resistances and P_pc and decreased the microvascular and large-vessel compliances. Audi et al. (6) observed that hypoxic dog lungs elicited vasoconstriction in both precapillary and postcapillary pulmonary vessels, with preferential vasoconstriction in the arterial segments. In addition, studies using direct micropuncture measurements in isolated cat lungs suggested that hypoxia constricted small-artery vessels (27), and it has been proposed that the small distensible vessels in the capillary bed region constrict to hypoxia (20). Linehan and Dawson (25) observed that hypoxia decreased total vascular compliance, and Hofman and Ehrhart (23) reported that pulmonary vascular compliance under hypoxic conditions was further decreased in the presence of cyclooxygenase inhibition. In the present study, the decrease in vascular compliance that occurred when vascular pressure was increased reflected the relative indistensibility of the pulmonary vasculature on hypoxic stimulation. In addition, there was a tendency for large-vessel compliance (C₁ + C₃) to comprise a smaller percentage of total vascular compliance when vascular pressures were increased by hypoxia compared with control conditions (17 vs. 21%).

It appears that Ca²⁺-activated K⁺ channels, ATP-sensitive K⁺ channels, and delayed rectifier K⁺ channels play a role in maintaining normal pulmonary vascular tone. TEA, Glib, and 4-AP caused a small but significant increase in pulmonary vascular pressure under normoxic conditions, which was potentiated during hypoxia. Specifially, TEA, Glib, and 4-AP potentiated the response to hypoxia on the arterial but not on the venous segments and also further decreased pulmonary vascular compliance. In contrast, Crom inhibited the vasoconstrictor effect of hypoxia on both the arterial and venous vessels during hypoxia. The increase in precapillary resistance but not in postcapillary resistance on K⁺-channel blockade had the net effect of no further increase in P_pc, which is directly correlated to the distribution of pulmonary vascular resistance (1, 14, 16).

Recently, K⁺-channel inhibition has been implicated as a critical event in the initiation of HPV (33, 34, 41). Post et al. (33) reported that hypoxia inhibited the delayed rectifier K⁺ current, and Yuan et al. (41) found that the ATP-sensitive K⁺-channel opener Crom inhibited hypoxic vasoconstriction in pulmonary arterial rings, an effect reversed by the ATP-sensitive K⁺-channel blocker Glib. Wiener et al. (40) reported that ATP-sensitive K⁺ channels modulated pulmonary vasoconstriction to hypoxia in isolated ferret lungs. Post et al. (34) observed that hypoxia inhibited a Ca²⁺-dependent K⁺ channel in canine pulmonary vascular smooth muscle and suggested that K⁺-channel inhibition is a key event that associates hypoxia with pulmonary vasoconstriction by causing membrane depolarization and subsequent Ca²⁺ entry. In addition, Albarwani and Nye (2) suggested that hypoxia inhibited both Ca²⁺-activated and ATP-sensitive K⁺ channels.

Although not tested in this study, K⁺-channel activation by vasodilators that elevate cGMP or cAMP may also attenuate the pressor response to hypoxia. Cabell et al. (11) suggested that large-conductance Ca²⁺-activated K⁺ channels were important in the cAMP-mediated but not in the cGMP-mediated relaxation of coronary resistance arteries. In other species, it has been shown that the activity of high-conductance Ca²⁺-activated K⁺ channels is potentiated by the cGMP-dependent activation of protein kinase G (35), and Archer et al. (3) provided compelling evidence for the opening of Ca²⁺-activated K⁺ channels by increased cGMP levels in rat pulmonary vascular smooth muscle cells. In their experimental model, Archer et al. suggested that pulmonary vascular signal transduction is modulated by both cGMP and cAMP, which regulate K⁺-channel phosphorylation. Specifically, increased levels of these cyclic nucleotide second messengers promote the opening of Ca²⁺-activated K⁺ channels, which leads to membrane hyperpolarization, inhibition of voltage-gated Ca²⁺ channels, and subsequent vasorelaxation or an attenuation of vasoconstriction. Torphy (38), in a recent review, documented that the increases in cAMP and cGMP can simultaneously activate the protein kinase A and protein kinase G pathways to promote the opening of Ca²⁺-activated K⁺ channels and elicit membrane hyperpolarization. Therefore, vasodilatory mechanisms involving the stimulation of cAMP or cGMP may act to attenuate HPV.

Evidence indicates that hypoxia blocks voltage-gated K⁺ channels in pulmonary vascular smooth muscle cells (41), and hypoxia-induced membrane depolarization has been associated with the inhibition of whole cell K⁺ current, leading to an increase in pulmonary arterial tension (32). In contrast, ATP-sensitive K⁺-channel openers such as lemakalim, pinacidil, and minoxidil appear to attenuate hypoxia-induced effects through membrane hyperpolarization (12, 28). Yuan et al. (41) demonstrated that Crom inhibited hypoxia-induced contractions in isolated rat pulmonary arteries, which was antagonized by Glib. In addition, Post et al. (34) provided strong evidence for a direct role of Ca²⁺-activated K⁺-channel inhibition in hypoxic pulmonary vasoactivity.

Studies (33, 34) suggested that L-type voltage-dependent Ca²⁺ channels also modulate the potentiated hypoxic pressor response by K⁺-channel blockade. Post et al. (34) observed that the dihydropyridine Ca²⁺-channel blocker nisoldipine prevented hypoxic inhibition of K⁺ currents in pulmonary arterial smooth muscle cells. In addition, it has been shown that a decrease in oxygen from a normoxic to a hypoxic level causes depolarization of the resting membrane potential in pulmonary vascular smooth muscle and subsequent Ca²⁺ entry through voltage-gated Ca²⁺ channels (4, 26). In the present study, the L-type voltage-dependent Ca²⁺-channel blocker verapamil inhibited the hypoxic pressor response. In addition, a small number of experiments showed that verapamil also inhibited the potentiated hypoxic response to K⁺-channel inhibition by all three K⁺-channel blockers used in this study (data not shown).

In the pulmonary circulation, vascular smooth muscle tone is an important determinant of pulmonary vascular resistance, pulmonary vascular compliance, and lung blood flow. The membrane potential of vascular smooth muscle is regulated by K⁺ channels, which, in turn, modulate vascular smooth muscle tone and vasoconstriction. The importance in the identification of the specific vascular segments responding to hypoxia and the K⁺ channels that modulate the pressor response relates to the effect on P_pc, which is determined by the distribution of vascular resistance in the pulmonary arteries and veins and the maintenance of ventilation-perfusion matching by constriction of pulmonary arteries that direct blood flow away from hypoxic regions (22).

The results of this study show that hypoxia increased pulmonary arterial resistance, pulmonary venous resistance, and P_pc and decreased pulmonary vascular compliance. TEA, Glib, and 4-AP potentiated the response to hypoxia on the arterial but not on the venous segments and also further decreased pulmonary vascular compliance. In contrast, the ATP-sensitive K⁺-channel opener Crom and the L-type voltage-dependent Ca²⁺-channel blocker verapamil inhibited the vasoconstrictor effect of hypoxia on both the arterial and venous vessels. These results indicate that closure of the Ca²⁺-activated K⁺ channels, ATP-sensitive K⁺ channels, and delayed rectifier K⁺ channels potentiate the canine pulmonary arterial response under hypoxic conditions and that K⁺-channel inhibition may be a key event that links hypoxia to pulmonary vasoconstriction by eliciting membrane depolarization and subsequent Ca²⁺ influx through voltage-dependent Ca²⁺ channels.