Differential Distribution of Electrophysiologically Distinct Myocytes in Conduit and Resistance Arteries Determines Their Response to Nitric Oxide and Hypoxia
Abstract The cellular mechanisms that determine differences in reactivity of arteries of varying size and origin are unknown. We evaluated the hypothesis that there is diversity in the distribution of K+ channels between vascular smooth muscle (VSM) cells within a single segment of the pulmonary arteries (PAs) and that there are differences in the prevalence of these cell types between conduit and resistance arteries, which contribute to segmental differences in the vascular response to NO and hypoxia. Three types of VSM cells can be identified in rat PAs on the basis of their whole-cell electrophysiological properties—current density and the pharmacological dissection of whole-cell K+ current (IK)—and morphology. Cells are referred to as “KCa, KDR, or mixed,” acknowledging the type of K+ channel that dominates the IK: the Ca2+-sensitive (KCa) channel, delayed rectifier (KDR) channel, or a mixture of both. The three cell types were identified by light and electron microscopy. KCa cells are large and elongated, and they have low current density and currents that are inhibited by tetraethylammonium (5 mmol/L) or charybdotoxin (100 nmol/L). KDR cells are smaller, with a perinuclear bulge, but have high current density and currents that are inhibited by 4-aminopyridine (5 mmol/L). Conduit arteries contain significant numbers of KCa cells, whereas resistance arteries have a majority of KDR cells and few KCa cells. NO rapidly and reversibly increases IK and hyperpolarizes KCa cells because of an increase in open probability of a 170-pS KCa channel. Hypoxia depolarizes KDR cells by rapidly and reversibly inhibiting one or more of the tonically active KDR channels (including a 37-pS channel) that control resting membrane potential. The effects of both hypoxia and NO on K+ channels are evident at negative membrane potentials, supporting their physiological relevance. The functional correlate of this electrophysiological diversity is that KDR-enriched resistance vessels constrict to hypoxia, whereas conduit arteries have a biphasic response predominated by relaxation. Although effective in both segments, NO relaxes conduit more than resistance rings, in both cases by a cGMP-dependent mechanism. We conclude that regional electrophysiological diversity among smooth muscle cells is a major determinant of segmental differences in vascular reactivity.
Segmental differences in vasomotor reactivity are well documented in the pulmonary vasculature,1 but the mechanism permitting local variations in response to common vasoactive stimuli, such as hypoxia and NO, are unknown. Hypoxia, for example, causes sustained constriction in resistance arteries2 3 4 5 while causing a biphasic response in conduit PAs, predominantly relaxation.6 7 HPV is caused in part by inhibition of a K+ channel in the PA VSM, resulting in membrane depolarization.8 9 The involvement of a K+ channel in HPV has been confirmed by several laboratories.10 11 12
NO causes pulmonary vasodilatation,13 14 in part, by increasing the levels of cGMP, which activates a cGMP-dependent protein kinase, opening KCa channels.15 Segmental differences in PA VSM reactivity to NO have not yet been reported. Existing studies using inhaled NO16 are inconclusive in this respect because of the different availability of inhaled NO to resistance and conduit PAs.
Recent evidence suggests that there are three or four types of VSM cells in the conduit PA.17 The functional consequence of cell diversity is unknown, although data from systemic arteries suggest that there are several types of arterial VSM cells that respond differently to agonists, such as endothelin-1 and α-thrombin.18
During previous experiments,15 we noted two types of whole-cell IK in PA VSM cells. Some cells had IK predominated by KCa channels. IK in these cells increased in response to NO and was inhibited by TEA or the specific KCa channel blocker CTX.19 In other cells, IK predominantly reflected the contribution of KDR channels and was inhibited by low doses of 4-AP but not TEA or CTX.15 In these cells, NO did not increase IK. We hypothesize that the regional differences in HPV and NO relaxation result from the existence of distinct populations of PA VSM cells in conduit and resistance vascular segments, with KCa- and KDR-predominant cells being more prevalent in conduit PAs and resistance PAs, respectively. The present study was undertaken to characterize these PA VSM cell subtypes and examine the functional consequences of regional diversity in K+ channel prevalence to HPV and NO relaxation in PA rings. The reference to cells as “KCa, KDR, or mixed” refers to the class of K+ channel predominating their whole-cell current and does not imply the cells have exclusively a single type of K+ channel.
Materials and Methods
Whole-cell electrophysiological characteristics (Em, capacitance, and current density) were assessed in enzymatically dispersed PA VSM cells from rat conduit and resistance PAs. Conduit PAs were defined as the main, right, or left PA (external diameter, 500 to 600 μm), and resistance PAs were intraparenchymal, fourth- and fifth-division PAs (external diameter, <300 μm). The sensitivity of Em and IK to K+ channel inhibitors, TEA, 4-AP, and CTX was determined using whole-cell current and voltage-clamp techniques, respectively.20 Morphological characteristics of cells from both arterial segments were determined by light and scanning electron microscopy. The prevalence of the three cell types was determined by a light microscopy survey of enzymatically dispersed cells from each segment. Single-channel characteristics of KCa and KDR channels were measured in amphotericin-perforated vesicles and cell-attached patches, and the effects of NO and hypoxia on the channels were studied in these preparations. Conductances of KCa and KDR channels were also calculated in excised patches, and the sensitivity of the KCa channel to Ca2+ was determined. To assess the functional significance of the electrophysiological diversity, we compared the effects of TEA, 4-AP, hypoxia, and NO on tension in conduit and resistance rings.
PA VSM cells were freshly dissociated each day. Male Sprague-Dawley rats (body weight, 300 to 400 g) were anesthetized with 50 mg/kg sodium pentobarbital. The heart and lungs were removed en bloc and perfused with Ca2+-free Hanks’ solution (see “Solutions”), as previously described.14 Conduit and resistance PAs were dissected free of adventitia, placed in Ca2+-free Hanks’ solution for 10 minutes, and then transferred to a papain solution (see “Solutions”) at 4°C for 20 minutes. Then the same papain solution was supplemented with dithiothreitol (0.85 mg/mL), and incubation was continued for ≈50 minutes at 4°C and then 15 minutes at 37°C (for all solutions, pH 7.4). Arteries were washed for 10 minutes and maintained in iced Hanks’ solution supplemented with glucose (1 mg/mL). Gentle trituration produced a suspension of single cells, which was then pipetted into a perfusion chamber on the stage of an inverted microscope for patch-clamp studies.20 After a brief period to allow partial adherence to the bottom of the recording chamber, cells were perfused via gravity with a bath solution (see “Solutions”) at a rate of 2 mL/min. Conventional whole-cell patch-clamp experiments were performed as previously described.15 Cells were proven to be VSM cells at the beginning of the study by immunohistochemistry (factor VIII antigen negative, smooth muscle–specific actin positive). All drugs were supplied by Sigma Chemical Co except LY83583 (Calbiochem).
Whole-Cell Amphotericin-Perforated Patch-Clamp Studies
Whole-cell recordings were performed using the amphotericin-perforated patch-clamp technique.21 Amphotericin B was included in the pipette at a final concentration of 120 μg/mL. Experiments were performed in low light intensity because of the light sensitivity of amphotericin B. Pipette resistances ranged from 1 to 4 MΩ after Sylgard coating and fire-polishing, with series resistance typically compensated by 90% to 95%. Membrane current was monitored on a digital oscilloscope. Cells were voltage-clamped at a holding potential of −70 mV, and currents were evoked by 10-mV steps to more positive potentials using test pulses of 200- to 400-millisecond duration at a rate of 0.033 to 0.1 Hz. Currents were filtered at 1 kHz and sampled at 4 kHz. For Em experiments, cells were held in current clamp at their resting Em, and control recordings were made for at least 1 minute before application of the drug, to ensure stability. All data were recorded and analyzed using pCLAMP 6.02 software (Axon Instruments). Whole-cell data are presented in the form of current-voltage plots, permitting a complete description of the channel activity over a wide range of Ems.
Whole-cell capacitance was measured using the manual whole-cell capacitance controls on the Axopatch amplifier. Whole-cell currents for each cell (at +70 mV) were divided by the cell’s capacitance, giving a measure of current density (in picoamperes per picofarad).
Three single-channel configurations were used: amphotericin-perforated vesicles,22 excised patches, and cell-attached patches.20 Bath and pipette solutions are given in the “Solutions” section. Pipette resistance for each configuration was 7 to 10 MΩ.
Amphotericin-perforated vesicles and cell-attached patches were used to study the responses to NO and hypoxia, as these configurations preserve intracellular organelles and second messengers that are thought to be important to O2 sensing9 and NO relaxation.15
The technique for amphotericin-perforated vesicles is similar to that for whole-cell perforated-patch recording, except that once a seal is formed on the cell and series resistance has been confirmed to decline slowly over 15 minutes, the pipette, containing the same solution as for the conventional whole cell, is withdrawn from the cell. This forms a perforated vesicle that envelops normal cytosolic and organellar contents.
Mean single-channel amplitudes and open times were calculated using preset detection levels in the events list analysis program of pCLAMP 6.02 (Axon Instruments). Currents were filtered at 1 kHz and sampled at 4 kHz. Channels were characterized by measuring their slope conductance in symmetrical or near symmetrical K+ (see “Solutions”). Since intracellular K+ can only be estimated in the cell-attached configuration, the slope was not forced through the origin. All single-channel recordings contained at least 2500 openings. Effects of NO and hypoxia on Po, NPo, and channel open dwell time (τ) were calculated using the following equations and pCLAMP 6.02 software:
where to and ti are total open time and time interval, respectively, and N is the number of channels. Fit for dwell time is given as follows:
where Pn is the fractional proportion of the nth component to the area under the curve, τ is open dwell time, and t is time.
Po and NPo were calculated from latency analysis and subsequent Po histogram generation for perforated vesicles, since single-channel data were obtained using Clampex. For cell-attached and excised patches, Po and NPo were calculated from continuous recordings obtained in Fetchex.
K+ Channel Pharmacology
To characterize the K+ channel subtypes, the cells/vesicles were treated with TEA (5 mmol/L, a preferential KCa inhibitor at this low dose), synthetic CTX (100 nmol/L, a relatively selective peptide KCa inhibitor19 ), or 4-AP (5 mmol/L, preferential KDR inhibitor).12 23 24 25 Recent studies show that although high doses of TEA and 4-AP are not “specific” for KCa and KDR channels, low doses do inhibit different components of the IK in PA VSM.11 Specifically, TEA (≤5 mmol/L) inhibits KCa channels, whereas 4-AP (≤5 mmol/L) inhibits CTX- and TEA-insensitive K+ channels.12 23 24 25 Although CTX can inhibit other K+ channels,26 low doses (≤100 nmol/L) appear to be fairly specific for KCa channels. In PA VSM, CTX inhibits the portion of the IK that is 4-AP insensitive.15
To study the effects of NO, 1 μL of a 2 mmol/L solution, prepared as previously described,27 was administered close to the cell/vesicle. This dose was chosen on the basis of preliminary experiments, which established that it was the minimum that could be reliably delivered manually and reproducibly activate K+ channels. In experiments that examined the effects of hypoxia, pipette solutions were supplemented with various channel blockers (ie, niflumic acid, CTX, and/or TEA). The divalent cation concentration was also varied slightly to optimize detection of KDR channels in these experiments (see “Solutions”). Experiments using NO were performed at room temperature.
Single-channel hypoxia experiments were performed on KDR cells from resistance arteries. All experiments involving hypoxia were performed at 32°C, as the O2-sensitive channel is relatively inactive at room temperature (data not shown).
Hanks’ solution contained (mmol/L) NaCl 140, KCl 4.2, KH2PO4 1.2, MgCl2 0.5, HEPES 10, and EGTA 0.1 (pH 7.4).
Papain solution contained Hanks’ solution without EGTA, papain (1 mg/mL), and bovine albumin (0.25 mg/mL).
Amphotericin B Perforated-Patch Whole-Cell and Vesicle Experiments
Pipette solution. The pipette solution contained (mmol/L) KCl 134, KH2PO4 1.2, MgCl2 1.0, HEPES 5, and 120 μg/mL amphotericin B, pH 7.25.
Bath solutions. For hypoxia experiments, the bath solution contained (mmol/L) NaCl 115, NaHCO3 25, KCl 4.2, KH2PO4 1.2, MgCl2 0.5, CaCl2 1.5, and HEPES 10, pH 7.4. For NO experiments, the bath solution contained (mmol/L) NaCl 140, KCl 4.2, KH2PO4 1.2, MgCl2 0.5, CaCl2 1.5, and HEPES 10, pH 7.4.
Cell-Attached Single-Channel Experiments
Pipette solution for KCa channels. This solution contained (mmol/L) KCl 134, KH2PO4 1.2, CaCl2 1.0, HEPES 5, and 4-AP 5, pH 7.30.
Pipette solution for KDR channels. This solution contained (mmol/L) KCl 134, KH2PO4 1.2, MgCl2 1.0, HEPES 5, pH 7.30, EGTA 10, TEA 7.5 or CTX 0.001, and niflumic acid 0.2.
Bath solution. This solution was the same as for amphotericin B perforated-patch vesicle experiments.
Excised-Patch Single-Channel Experiments
Pipette solution. This solution contained (mmol/L) KCl 134, KH2PO4 1.2, CaCl2 1.0, and HEPES 5, pH 7.30.
Bath solution. This solution contained (mmol/L) KCl 134, KH2PO4 1.2, CaCl2 varied as described below, MgATP 4, sodium creatine phosphate 2, and EGTA 10, pH 7.30.
To vary bath Ca2+ to 50 nmol/L, 5.7 mmol/L Ca2+ was added to nominally Ca2+-free solution (yielding a 200 nmol/L solution12 ) and then diluted 1:4.
The effects of hypoxia were studied by switching between normoxic and hypoxic perfusate reservoirs. Normoxic and hypoxic perfusate reservoirs were bubbled with 20% and 0% O2, respectively (plus 3.5% CO2/balance N2), producing Po2s in the cell chamber of 162 and 27 mm Hg, respectively. Pco2 was ≈40 mm Hg, and pH was 7.40 to 7.41 under both conditions.
Cell morphology was assessed by light microscopy in freshly dispersed cells immediately after trituration. In addition, identically handled cells were fixed in 4% gluteraldehyde for examination by scanning electron microscopy. Microscopic examination was performed by an independent electron microscopy service (Mark Sanders, PhD, University of Minnesota Imaging Center).
Rat conduit PA rings were isolated and mounted in 2-mL baths as previously described.28 Resistance PA rings were harvested and treated identically. The optimal resting tension to maximize constriction to KCl for both conduit and resistance PA rings was 600 mg. Endothelium was left intact, as confirmed by a normal relaxation response to acetylcholine (10−8 to 10−5 mol/L). The baths housing the rings were bubbled with 20% O2 (normoxia) and 0% O2 (hypoxia) (plus 5% CO2/balance N2), resulting in Po2 values of 124±2 and 35±3 mm Hg, respectively. Pco2 was 26 to 30 mm Hg, and pH was 7.39 to 7.47. The comparative effect of K+ channel blockers and hypoxia on conduit and PA rings was assessed by adding TEA (10 mmol/L) or 4-AP (10 mmol/L) to the bath. Although the Ki values of TEA for KCa channels and 4-AP for KDR channels are 0.25 and 0.7 mmol/L, respectively (in isolated renal VSM),29 30 vasoconstriction only occurs at doses of TEA and 4-AP above 1 mmol/L in isolated lungs31 and PA rings.15 This suggests that Ki in intact vascular tissue may be greater than that noted in dispersed cells.
To measure the relaxation response to NO (1 to 100 μL of 2 mmol/L solution), conduit and resistance PA rings were constricted with endothelin-1 (10−7 mol/L) in normoxia in the presence or absence of the nonspecific guanylate cyclase inhibitor LY83583 (10−4 mol/L).32 33
The data are mean±SEM. Changes in Em were analyzed by a factorial ANOVA. The current-voltage and dose-response curves were analyzed using repeated measures ANOVA (StatView 4.02, Abacus Concepts). Data from the PA rings were electronically filtered, digitized at 12 samples per minute, and evaluated with repeated measures ANOVA. A value of P<.05 was considered statistically significant.
All currents were completely eliminated by substituting Cs+ (145 mmol/L) for K+ in the patch pipette, indicating they were conducted by K+ (n=3, data not shown). The resting Em of PA VSM derived from resistance arteries was −38±4 mV. Outward KDR currents (inhibited by 4-AP) were activated positive to −45 mV, whereas KCa currents (inhibited by TEA) were activated positive to −30 mV (Fig 1A⇓ and 1B⇓).
VSM cells could be separated into three distinct populations based on current density (Fig 1C⇑) and whole-cell current pharmacology (Fig 1B⇑). In some cells, the whole-cell current was inhibited significantly by TEA with little effect from 4-AP; these cells were classified as KCa cells. In others, the KDR cells, the reverse was true. There was a third group of mixed type cells in which TEA and 4-AP were approximately equally effective in inhibiting IK. The average current densities were significantly different among KDR, KCa, and mixed cell types (Fig 1C⇑). Cell capacitance was highest in KCa cells (most abundant in the conduit PA) and lowest in the KDR cells (found predominantly in the resistance PA), reflecting the larger size of the KCa cells.
Hypoxia (Fig 2C⇓) and 4-AP (4 mmol/L, Fig 2B⇓) both reversibly depolarized PA VSM cells obtained from resistance arteries, whereas TEA (5 and 10 mmol/L) and CTX (100 nmol/L) had no effect, indicating that resting Em is largely controlled by KDR channels (Fig 2A⇓ and 2C⇓). The importance of the KDR or 4-AP–sensitive K+ conductance was evident in studies of tone in resistance PA rings. Hypoxia and 4-AP both constricted resistance arteries (Fig 2D⇓), whereas CTX (100 nmol/L) had no effect (data not shown).
Cell phenotype was different among KCa, KDR, and mixed cells (Fig 3⇓). Phenotypic categorization by light microscopy, before approaching the cell with the patch pipette, predicted the cells’ electrophysiological attributes when they were subsequently studied. At one end of the spectrum were KCa cells, which were elongated (72±5 μm) tapered cells (Fig 3⇓) with a small IK (440±11 pA at +70 mV, n=7). The marked inhibition of IK by 5 mmol/L TEA and the minimal effect of 4-AP indicated that the predominant K+ channel was the KCa type. Furthermore, TEA removed the noisy spontaneously spiking component of the current in KCa cells, leaving behind a low-noise slowly inactivating current, consistent with inhibition of KCa channels (Fig 1A⇑). At the other extreme were shorter cells (45±6 μm, P<.05 versus KCa cells) with a characteristic perinuclear bulge and significantly larger IK (1747±11 pA at +70 mV, n=20, P<.0001 versus KCa cells) despite their smaller size (Fig 3⇓). Inhibition of IK with low-dose 4-AP suggested predominance of a KDR-like channel (KDR cells). Mixed cells were identified, which physically resembled KDR cells (length, 50±4 μm) but lacked a distinct perinuclear bulge (Fig 3⇓). IK in these cells was of intermediate size (670±12 pA at +70 mV, n=5, P<.05 versus KCa or KDR) and was inhibited similarly by both TEA and 4-AP (Fig 1A⇑). Of note, 4-AP reduced currents ≈50%, leaving a noisy spiking current, consistent with residual KCa channel activity. Spherical cells were neither characterized morphologically nor used in electrophysiological studies, as they are judged likely to be injured.
On scanning electron micrographs, cells from conduit PAs had the same morphological appearance as seen on light microscopy. In general, their surfaces were smooth, with few cytoplasmic projections. KDR cells tended to have more convoluted membranes than did KCa cells. In general, cells from the resistance PA (regardless of shape) were more convoluted than cells from the conduit PA, with prominent cytoplasmic projections, despite similar isolation protocols (Fig 3⇑).
KCa cells were more prevalent in conduit than in resistance PAs (Fig 4⇓). In contrast, the resistance PAs had very few KCa cells, and KDR cells were most numerous (Fig 4⇓). The mixed cell type is found throughout the vascular tree.
NO Effects on KCa Channels
NO increased IK in KCa cells (Fig 5A⇓) by activating KCa channels (Fig 5B⇓). This caused reversible membrane hyperpolarization and relaxation of PA rings (Fig 5C⇓ and 5D⇓). NO-induced K+ channel activation was inhibited by TEA or CTX, both in whole-cell and single-channel recordings (Figs 5⇓ and 6⇓).
NO increased Po of a CTX-sensitive channel, while having no effect on channel amplitude (Fig 5B⇑). In single-channel cell-attached experiments, NO increased the NPo of KCa channels at a holding potential of −20 mV. There was minimal KCa activity until NO was applied (Fig 6A⇑). NO caused a rapid and reversible activation of KCa channels. The control value for open dwell time (1.6 milliseconds) was fitted with a single exponential curve. After application of NO, two channels appeared, and the open dwell time was fitted with a biexponential curve, resulting in open dwell times of 8.2 and 111.0 milliseconds.
In excised patches, increasing the bath Ca2+ concentration from nominally Ca2+-free to 50 nmol/L caused a proportionate increase in NPo from 0.004 to 0.156 (Fig 7A⇓). The slope conductance of the KCa channel in symmetrical K+ was similar in cell-attached and excised-patch configurations (170 and 176 pS, respectively; Fig 7B⇓). The KDR channel was also calculated to have similar conductance in these two configurations (37 versus 42 pS, respectively). The excised patch data were derived from mixed cells from the resistance PA, confirming the presence of both KCa and KDR channels in such cells.
Hypoxic Effects on K+ Channels
Hypoxia rapidly inhibited IK (Fig 8A⇓) and caused membrane depolarization (Fig 2⇑) in KDR cells. Inhibition of K+ channels occurred within 1 to 2 minutes in whole-cell experiments (Fig 8A⇓) and in <1 minute in single-channel experiments (Fig 8B⇓). The hypoxic inhibition of IK in whole cells was evident at physiologically relevant Ems (−30 mV; inset, Fig 8A⇓). In singlechannel studies in KDR cells from resistance PAs, hypoxia reversibly decreased NPo (Figs 8B⇓ and 9⇓) and minimally shortened open dwell time (from 6.2 to 4.9 milliseconds in the example shown in Fig 9⇓). In these cell-attached studies, there were several KDR channels, judged by their different amplitudes, which could be measured during normoxia, despite the presence of TEA (or CTX) and niflumic acid in the patch pipette. The openings of these channels were inhibited by hypoxia. The channel most consistently inhibited was a 37-pS channel (Fig 9⇓). Reversal of hypoxic K+ channel inhibition was almost universally present in single-channel experiments on reoxygenation but was only clearly demonstrated in two of seven whole-cell experiments. Hypoxia increased IK in KCa cells at +70 mV (Fig 8⇓).
TEA caused greater constriction in conduit PAs (+285±80 mg at 10 mmol/L, n=5) than in resistance PAs (+62±32 mg, n=5, P<.05), consistent with the greater prevalence of KCa cells in the former (Fig 4⇑). Constriction to 4-AP was similar (P=.07) in conduit rings (+376±77 mg at 10 mmol/L, n=5) and resistance rings (+157±73 mg, n=5), consistent with the ubiquitous distribution of mixed type cells (Fig 4⇑).
Response to NO
NO was an effective relaxant in both conduit and resistance PA rings. However, consistent with the greater prevalence of NO-activated KCa cells in the conduit PA, NO caused more relaxation in conduit than resistance PA rings (Fig 5D⇑). NO-induced relaxation of both segments was attenuated by the guanylate cyclase inhibitor LY83583 (Fig 5D⇑).
In control resistance rings (no K+ channel blocker present), hypoxic vasoconstriction was demonstrated without the use of pharmacological priming, which is commonly required to elicit hypoxic constriction in excised PAs (Fig 2D⇑). This attests to the excellent condition of the rings. In resistance arteries, hypoxia caused monophasic constriction, which was sustained for the 12-minute exposure to hypoxia (Fig 2D⇑). Hypoxic constriction of resistance rings was enhanced by pretreatment with TEA and to a greater extent by pretreatment with 4-AP (Fig 2D⇑).
Hypoxic Pulmonary Relaxation
Hypoxia caused a biphasic response in conduit arteries, an initial small constriction followed by relaxation to tones below the baseline. Hypoxic relaxation was not abolished by 4-AP or TEA (Fig 2D⇑).
The primary finding of the present study is the great diversity in K+ channel electrophysiology (judged by current density and whole-cell K+ channel pharmacology) among cells taken from the same segment of the PA. Morphological criteria allow identification of three different cell types and also reveal that their distribution changes longitudinally along the pulmonary vasculature, from conduit to resistance segments. This diversity explains, in part, the segmental differences in the response of the pulmonary circulation to hypoxia and NO. In addition, the “side-by-side” existence of electrophysiologically distinct cells may explain the conflicting results investigators sometimes obtain when studying the effects of stimuli such as NO and hypoxia on randomly selected cells.
Three other important findings are discussed: (1) The Em in PA VSM is primarily controlled by KDR channels. (2) NO activates a KCa channel at physiological Ems by increasing both Po and open dwell time. (3) Authentic hypoxia causes a time-dependent inhibition of a 4-AP–sensitive whole-cell current. Hypoxia decreases the NPo of one or more KDR channels.
Arterial VSM is not homogeneous in terms of cell morphology17 and electrophysiology.15 18 34 The present study shows that there are three distinct cell types in the PA based on pharmacological, morphological, and electrophysiological criteria. Cells can be separated by electrophysiological criteria based on their current density and the class of K+ channels conducting the majority of their whole-cell current. The use of whole-cell current density provides an objective measure of net current, corrected for cell size. Using a similar technique, Albarwani et al34 found that cells in resistance and conduit PAs had capacitances of 10 to 12 pF with current densities of 157±64 and 141±66 pA/pF, respectively, at a holding potential of +30 mV. The large standard error of their data and the fact that their values are similar to the mean obtained when the current densities of our three cell types are averaged (mean, 110 pA/pF; KCa, 15.6±2.2 pA/pF; KDR, 265.3±36.6 pA/pF; and mixed, 51.7±10.8 pA/pF; Fig 1C⇑) suggest that they may have had a similar, but unrecognized, diversity in cell types.
There are phenotypic similarities, on light and scanning electron microscopy, among cells of each electrophysiological type (Fig 3⇑); consequently, morphology can be used to predict which channels are predominant in a given cell. It is important to emphasize that the morphology does not reflect variations in digestion, as all three cell types can be seen in a single dispersion from a single ring, in which all cells are treated identically. Furthermore, cells from conduit and resistance arteries have marked structural differences, despite identical harvesting and dispersion protocols. Overdigested cells are identified by their rounded or “humped-up” appearance and are not included in the electrophysiological or morphological survey.
The nomenclature “KCa, KDR, or mixed cells” reflects the K+ channel type that contributes most to whole-cell current and is a convenient means of expressing the cell’s identity. It does not mean that a cell has only one type of K+ channel. In fact, in both cell-attached and cell-excised patches, single-channel studies show the presence of both KDR and KCa channels in a single KDR cell (from a resistance artery). Nonetheless, the classification is quite useful, as the cells have characteristic responses to NO and hypoxia. Coupled with the finding that the three cell types (particularly KCa and KDR cells) have a different frequency in different arterial segments, this suggests that K+ channel diversity may explain much of the segmental variation in the pulmonary vasculature’s response to NO and hypoxia.
Conduit arteries are the main repository of KCa cells, where they reside interspersed with mixed and KDR cell types (Fig 4⇑). In contrast, the resistance arteries, the site at which vascular tone is primarily regulated in the lung,2 3 have a more homogeneous cell population composed mainly of mixed and KDR cell types. NO increases IK in the KCa cells but not in KDR cells. Conversely, hypoxia inhibits IK in KDR but not KCa cells (Figs 8⇑ and 9⇑). This correlates well with the localization of HPV to the resistance PA and the somewhat greater relaxation caused by NO in the conduit PA (Fig 5D⇑).
The existence of cell diversity in arteries is a recent concept. Frid et al17 found at least three populations of PA VSM, based on immunological and morphological criteria, within the media of the calf main PA. They noted diversity in cells within the PA, based on the distribution of contractile and cytoskeletal proteins (α-actin, myosin, calponin, desmin, and metavinculin), as they surveyed from the lumen toward the adventitia. Furthermore, the prevalence of the cell types varied during development from fetus to adult. Unlike the present study, cells were studied in sections (rather than after isolation), and no data relating to cell function were acquired. Since our cells were dispersed, we do not know whether the three cell types in the main PA were arranged randomly or in a luminal to abluminal array.
In systemic arteries, there are two types of smooth muscle cells, based on myosin isoform distribution.35 Furthermore, in the adult rat, cells appear in the neointima after vascular injury that differ in morphology from normal cells in the media, and these have an ability to produce matrix and growth factors, reminiscent of VSM from immature rats.36 37 Neylon et al18 found two cell types in cultures of adult rat aortic VSM. There were elongated cells that depolarized in response to vasoactive agonists (endothelin-1 and α-thrombin), which elicited an influx of Na+ and Ca2+ and epithelium-like cells that hyperpolarized (in response to the same stimuli), probably as a result of the activation of KCa channels.18 Unlike the present study, they performed no patch-clamp studies and relied on the use of fluorescent dyes to measure Em and cytosolic Ca2+. The cell morphology they described in nondispersed cultured cells is not comparable to the morphology in our dispersed isolated cells.
Neither the present study nor the studies of Frid et al17 and Neylon et al18 provide a clear explanation for the diversity of cell types within the vessel wall. We speculate that the disparity in the proportion of VSM cell subtypes between conduit and resistance PAs could be related to the different embryological origin of the conduit PA (sixth aortic arch) and resistance PA (lung bud).38 Consequently, the conduit PA may have reactivity closer to that of systemic vessels (both dilate in response to hypoxia, whereas the resistance PAs constrict).
If regional differences in K+ channel distribution are of functional importance, one might expect that this would be evident when examining the segmental response of the pulmonary vasculature to hypoxia and NO, stimuli that accomplish vasoconstriction and dilatation in the pulmonary circulation, which are at least in part due to their effects on K+ channels.8 9 10 15
HPV and 4-AP–Sensitive KDR Channels
HPV is characterized by the rapid reversible constriction of intrapulmonary resistance arteries that, although modulated by many neurohumoral factors, is intrinsic to PA VSM4 5 39 (see Reference 40 for review). Our isolated vascular ring experiments confirm resistance PA as the site of more significant and sustained hypoxic constriction compared with conduit PA. This is in accord with previous reports from isolated PA VSM cells,4 PA rings,41 isolated lungs,42 and intact animals.2 43 The precise identity of the K+ channels involved in HPV has been uncertain, although whole-cell data suggest it may be a KDR-type channel.9 10 In the present study, 4-AP, a preferential KDR inhibitor, caused more constriction than did TEA, consistent with the finding that 4-AP, but not TEA or CTX, depolarizes PA VSM from resistance arteries (Fig 2B⇑). The greater effect of 4-AP on tone in rings is consistent with results of a previous study that compared the effects of K+ channel blockers on tone in the isolated rat lung.31 Together, these findings support the contention of Yuan25 and Post et al12 that the resting Em in PA VSM is controlled by the KDR class of K+ channels. Our finding of several TEA-resistant, CTX-resistant, and niflumic acid–resistant hypoxia-inhibitable K+ channels suggests that the KDR may not be a single channel but a family of channels (Fig 9⇑). Further biophysical studies are required to comment on this possibility.
It is expected that the segment most enriched in such KDR cells (the resistance PA, Fig 4⇑) should have the best hypoxic constrictor response, which is in fact the case (Fig 2D⇑). If resistance PAs constrict because of KDR channels, why do conduit arteries relax during hypoxia? One possible explanation is that hypoxia increases IK in KCa cells (Fig 8⇑), much as it does in cerebral VSM.44 It is possible that both KCa and KATP channels contribute to the hypoxic relaxation in conduit PA and systemic arteries.
The data in Figs 8⇑ and 9⇑ provide the first single-channel demonstration (using authentic hypoxia rather than dithionite) that hypoxia rapidly and reversibly inhibits KDR channels. The fact that this occurs at negative Ems (Fig 8⇑, inset) and that hypoxia depolarizes the membrane (Fig 2C⇑) confirms the physiological relevance of K+ channel inhibition to HPV. Dithionite, although a convenient way to reduce hemoproteins and lower Po2, does not mimic hypoxia created by alveolar ventilation with hypoxic gas or N2 bubbling of solutions.45 In fact, dithionite only lowers Po2 by means of an attendant and obligatory generation of O2 radicals and peroxides, which actually results in vasodilatation, rather than constriction, in the isolated lung.45
NO Activates KCa Channels
NO causes relaxation by several mechanisms, including sequestration of intracellular Ca2+,46 inhibition of the voltage-gated Ca2+ channel, and desensitization of the contractile apparatus to Ca2+.47 In addition, a variable portion of NO relaxation may relate to its ability to activate KCa channels, evident in coronary,48 cerebral,49 and pulmonary15 VSM. In the conduit PA, part of the dilator effect of NO is due to KCa activation, and relaxation is impaired by CTX.15 NO is an effective dilator in both conduit PAs and resistance PAs; thus, it might not appear that K+ channel distribution is important to the hemodynamic effects of NO. However, careful comparison of the relaxant effects of NO reveals a slight but significant increase in dilator efficacy in the conduit rings, relative to resistance rings (Fig 5D⇑). The difference in the potency of NO between segments may reflect the greater prevalence in conduit PAs of KCa cells, which respond to NO with an increase in IK and Po and hyperpolarization (Figs 5⇑ and 6⇑). There is controversy concerning whether NO acts directly on the KCa channel50 or via an intracellular pathway involving cGMP and cGMP-dependent protein kinase.15 48 49 In the present study, NO-induced relaxation was attenuated by the guanylate cyclase inhibitor LY83583, consistent with prior studies demonstrating that the effects of NO on K+ channels and vascular tone are mediated through cGMP.15 48 49 LY83583, like methylene blue, is a nonspecific inhibitor of guanylate cyclase and may have other unanticipated effects, although it does lower cGMP.32 33 Although direct cGMP-independent K+ channel activation by NO occurs in excised patch studies,50 this effect does not appear essential to the dilator efficacy of NO in arterial rings.
The present study demonstrates that NO increases both the Po and the open dwell time of the KCa channel. It is noteworthy that KCa channels were activated by NO at relatively negative Ems (−20 mV, Fig 6A⇑) and that NO causes membrane hyperpolarization, consistent with a physiological role for this effect.
We conclude that regional diversity in vascular reactivity to NO and hypoxia result, in large part, from differences in the prevalence of the three types of VSM cells, which possess electrophysiologically distinct types of K+ channels.
Selected Abbreviations and Acronyms
|HPV||=||hypoxic pulmonary vasoconstriction|
|KCa channel (cell)||=||Ca2+-sensitive K+ channel (cell)|
|KDR channel (cell)||=||delayed rectifier K+ channel (cell)|
|NPo||=||number of channels multiplied by Po|
|Po||=||channel open probability|
|VSM||=||vascular smooth muscle|
This study was supported by the Department of Veterans Affairs, National Institutes of Health grant HL-45735 (Dr Archer), and the American Heart Association, Minnesota Affiliate, Inc (Drs Reeve and Hampl). We thank Dennis Knapp for immunochemical staining of VSM cells. We acknowledge the assistance of Mark Sanders, PhD, in performing the electron microscopy.
- Received April 27, 1995.
- Accepted November 21, 1995.
- © 1996 American Heart Association, Inc.
Dawson CA, Linehan JH, Bronikowski TA, Rickaby DA. Pulmonary microvascular hemodynamics: occlusion methods. In: Will JA, Dawson CA, Weir EK, Buckner CK, eds. The Pulmonary Circulation in Health and Disease. Orlando, Fla: Academic Press; 1987:175-197.
Kato M, Staub NC. Response of small pulmonary arteries to unilobar alveolar hypoxia and hypercapnia. Circ Res. 1966;19:426-440.
Shirai M, Sada K, Ninomiya I. Effects of regional alveolar hypoxia and hypercapnia on small pulmonary vessels in cats. J Appl Physiol. 1986;61:440-448.
Madden JA, Vadula MS, Kurup VP. Effects of hypoxia and other vasoactive agents on pulmonary and cerebral artery smooth muscle cells. Am J Physiol. 1992;263:L384-L393.
Vadula MS, Kleinman JG, Madden JA. Effect of hypoxia and norepinephrine on cytoplasmic free Ca2+ in pulmonary and cerebral arterial myocytes. Am J Physiol. 1993;265:L591-L597.
Bennie RE, Packer CS, Powell DR, Jin N, Rhoades RA. Biphasic response of pulmonary artery to hypoxia. Am J Physiol. 1991;261:L156-L163.
Jin N, Packer S, Rhoades RA. Pulmonary arterial hypoxic contraction: signal transduction. Am J Physiol. 1992;263:L73-L78.
Post JM, Hume JR, Archer SL, Weir EK. Direct role for potassium channel inhibition in hypoxic pulmonary vasoconstriction. Am J Physiol. 1992;262:C882-C890.
Archer SL, Huang J, Henry T, Peterson D, Weir EK. A redox-based O2 sensor in rat pulmonary vasculature. Circ Res. 1993;73:1100-1112.
Yuan X-J, Goldman WF, Tod ML, Rubin LJ, Blaustein MP. Hypoxia reduces potassium currents in cultured rat pulmonary but not mesenteric arterial myocytes. Am J Physiol. 1993;264:L116-L123.
Smirnov SV, Robertson TP, Ward JPT, Aaronson PI. Chronic hypoxia is associated with reduced delayed rectifier K+ current in rat pulmonary artery muscle cells. Am J Physiol. 1994;266:H365-H370.
Post JM, Gelband CH, Hume JR. [Ca2+]i inhibition of K+ channels in canine pulmonary artery: novel mechanism for hypoxia-induced membrane depolarization. Circ Res. 1995;77:131-139.
Archer SL, Rist K, Nelson DP, DeMaster EG, Cowan N, Weir EK. Comparison of the hemodynamic effects of nitric oxide and endothelium-dependent vasodilators in intact lungs. J Appl Physiol. 1990;68:735-747.
Archer SL, Huang JM-C, 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.
Roos CM, Rich GF, Uncles DR, Daugherty MO, Frank DU. Sites of vasodilation by inhaled nitric oxide vs. sodium nitroprusside in endothelin-constricted isolated rat lungs. J Appl Physiol. 1994;77:51-57.
Frid MG, Moiseeva EP, Stenmark KR. Multiple phenotypically distinct smooth muscle cell populations exist in the adult and developing bovine pulmonary arterial media in vivo. Circ Res. 1994;75:669-681.
Neylon CB, Avdonin PV, Dilley RJ, Larsen MA, Tkachuk VA, Bobik A. Different electrical responses to vasoactive agonists in morphologically distinct smooth muscle cell types. Circ Res. 1994;75:733-741.
Hille B. Ionic Channels of Excitable Membranes. 2nd ed. Sunderland, Mass: Sinauer Associates Inc; 1992.
Gelband CH, Hume JR. [Ca2+]i inhibition of K+ channels in canine renal artery: novel mechanism for agonist-induced membrane depolarization. Circ Res. 1995;77:121-130.
Yuan X-J. Voltage-gated K+ currents regulate resting membrane potential and [Ca2+]i in pulmonary artery myocytes. Circ Res. 1995;77:370-378.
Garcia ML, Knaus H-G, Munujos P, Slaughter RS, Kaczorowski GJ. Charybdotoxin and its effects on potassium channels. Am J Physiol. 1995;269:C1-C10.
Archer SL, Shultz PJ, Warren JB, Hampl V, DeMaster EG. Preparation of standards and measurement of nitric oxide, nitroxyl, and related oxidation products. Methods Enzymol. 1995;7(suppl):21-34.
Archer SL, Cowan NJ. Measurement of endothelial cytosolic calcium concentration and nitric oxide production reveals discrete mechanisms of endothelium-dependent pulmonary vasodilatation. Circ Res. 1991;68:1569-1581.
Gelband CH, Hume JR. Ionic currents in single smooth muscle cells of the canine renal artery. Circ Res. 1992;71:745-758.
Gelband CH, Ishikawa T, Post JM, Keef KD, Hume JR. Intracellular divalent cations block smooth muscle cell K+ channels. Circ Res. 1993;73:24-34.
Kontos H, Wei E. Hydroxyl radical-dependent inactivation of guanylate cyclase in cerebral arterioles by methylene blue and LY83583. Stroke. 1993;24:427-434.
Schmidt MJ, Sawyer BD, Truex LL, Marshall WS, Fleisch JH. LY83583: an agent that lowers intracellular levels of cyclic guanosine 3′, 5′-monophosphate. J Pharmacol Exp Ther. 1985;232:764-769.
Walker LN, Bowen-Pope DF, Ross R, Reidy MA. Production of platelet derived growth factor-like molecules by cultured arterial smooth muscle cells accompanies proliferation after injury. Proc Natl Acad Sci U S A. 1986;83:7311-7315.
Majesky MW, Giachelli CM, Reidy MA, Schwartz SM. Rat carotid neointimal smooth muscle cells reexpress a developmentally regulated mRNA phenotype during repair of arterial injury. Circ Res. 1992;71:759-768.
Hislop A, Reid LM. Formation of the pulmonary vasculature. In: Hodson WA, ed. Development of the Lung. New York, NY: Marcel Dekker Inc; 1977:37-86.
Cornfield DN, Stevens T, McMurtry IF, Abman SH, Rodman DM. Acute hypoxia causes membrane depolarization and calcium influx in fetal pulmonary artery smooth muscle cells. Am J Physiol. 1994;266:L469-L475.
Weir EK, Archer SL. The mechanism of acute hypoxic pulmonary vasoconstriction: the tale of two channels. FASEB J. 1995;9:183-189.
Nagasaka K, Bhattacharya J, Nanjo S, Gropper MA, Staub NC. Micropuncture measurement of lung microvascular pressure profile during hypoxia in cats. Circ Res. 1984;54:90-95.
Archer SL, Hampl V, Nelson DP, Sidney E, Peterson DA, Weir EK. Dithionite increases radical formation and decreases vasoconstriction in the lung: evidence that dithionite does not mimic alveolar hypoxia. Circ Res. 1995;77:174-181.
Twort CHC, van Breemen C. Cyclic guanosine monophosphate–enhanced sequestration of Ca2+ by sarcoplasmic reticulum in vascular smooth muscle. Circ Res. 1988;62:961-964.
Raeymaekers L, Hofmann F, Casteels R. Cyclic GMP-dependent protein kinase phosphorylates phospholamban in isolated cardioplasmic reticulum from cardiac and smooth muscle. Biochem J. 1988;252:269-273.
Robertson BE, Schubert R, Hescheler J, Nelson M. cGMP-dependent protein kinase activates Ca-activated K channels in cerebral artery smooth muscle cells. Am J Physiol. 1993;265:C299-C303.