INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

HYPOXIC PULMONARY VASOCONSTRICTION (HPV) in mammalian pulmonary vascular smooth muscle is mediated by both endothelial and direct smooth muscle effects (4, 5, 11, 33), and it matches ventilation to perfusion by decreasing blood flow to underventilated alveoli. The calcium dependence of HPV in mammals has been well documented (15, 24). During HPV, a rise in free cytosolic calcium (Ca) accompanies a series of plasma membrane channel events that lead to full depolarization and vasoconstriction, although the order and relative importance of these events remain controversial (15, 23, 32, 34, 35).

Vasoconstriction appears to be prevalent in the respiratory circuits of most nonmammalian vertebrates during hypoxia (10, 12); however, the role of calcium and the intracellular mechanisms mediating these responses have not been examined. We have recently described for the first time a profound hypoxic vasoconstriction (HV) in postgill, but not pregill, systemic vessels from the New Zealand hagfish, the Pacific hagfish, and the sea lamprey (22). Although the physiological significance of this response in these animals is not presently known, we previously hypothesized that cyclostomes may be a useful model with which to study the intrinsic mechanisms underlying HPV (22).

The purpose of the present study was to characterize the role of calcium during HV in the primitive cyclostome model. We examined the utilization of stored intracellular calcium (Ca) and extracellular calcium (Ca) during the hypoxic response in postgill systemic arteries [dorsal aortas (DA)] from sea lamprey and New Zealand hagfish. Results indicate that in lamprey DA, 1) Ca is mobilized during HV and the rise in [Ca²⁺]_c during HV is temporally correlated with contractile force, 2) Ca is not required for HV, 3) L-type calcium channels are not activated for Ca influx during HV, 4) Ca²⁺ stores are used differently during HV than during adrenergic stimulation and are not easily depleted using standard pharmacological agents, and 5) Na⁺/Ca²⁺ exchange is present, but extracellular sodium (Na) is not required for relaxation following HV. The hypoxic response in hagfish DA is somewhat different in that Ca influx contributes significantly to the overall magnitude of HV, and Na is requisite for relaxation following HV.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals. Sea lamprey (Petromyzon marinus, 130-450 g) were captured by the U. S. Geological Survey, Biological Resources Division, in Michigan during the spring-summer spawning migration and airlifted to Notre Dame. At Notre Dame, they were housed in 500-liter rectangular tanks in aerated, flowing well water (15°C) and exposed to a 12:12-h light-dark photoperiod. They were not fed. Lamprey were anesthetized with benzocaine (ethyl-p-aminobenzoate; 1:5,000 wt:vol), and the vessels were dissected out and placed in lamprey HEPES-buffered saline (LHBS) at 4°C.

Vascular smooth muscle. DA from lamprey and hagfish were cut transaxially into 3- to 4-mm rings. Rings were hung on 280-µm stainless steel hooks and suspended in 20-ml water-jacketed (15°C) smooth muscle chambers (21) containing the appropriate saline. In experiments with lamprey, stainless steel hooks attached vascular rings to plastic diffusers inside the smooth muscle chambers. The diffusers were designed and fabricated in our lab to allow rapid mixing of chemicals and gasses while protecting the rings from direct gas contact, thereby reducing turbulence. The diffusers were attached to lids to minimize surface gas exchange with the atmosphere and facilitated the addition of drugs and bath changes. Hypoxia was administered by aerating the muscle chambers with 100% nitrogen gas (N₂), and normoxia was restored by aeration with room air. Tension was measured with Grass FTO3C force-displacement transducers and recorded on either a computer-interfaced Gould 8000 series or Grass model 8TC polygraph. Data were collected electronically using Labtech Notebook data-collection software (Laboratory Technologies, Andover, MA).

5

5

Fura 2-AM. The experimental apparatus for measuring [Ca²⁺]_c was similar to that described previously by Chen and Rembold (7). Helical DA strips from lamprey were stretched to optimal length and loaded with 5 µM fura 2-AM and 0.3 mM neostigmine in cold LHBS for 90 min. Fura 2 fluorescence is bright enough following this loading protocol to permit measurements at intracellular dye concentrations determined in rats to have no effect on calcium buffering or damping of calcium transients that might interfere with excitation-contraction coupling (7). Fluorophore leakage does not significantly contribute to the fluorescence measurements, because 1) bath perfusion rates of 1 ml/min diluted any leaking indicator, and 2) the illuminated area includes only a small volume of bath solution. The tissue was then mounted isometrically to a capacitive force transducer and bathed in a 3-ml water-jacketed tissue bath perfused with room temperature LHBS bubbled with room air. A bifurcated light guide was placed 0.5-1.0 mm from the lumenal surface of a 1.25- to 1.75-mm-wide smooth muscle strip. Excitation light was passed through one arm of the bifurcated light guide to illuminate a 1.5-mm diameter spot near the center of a smooth muscle strip that covered 75-100% of the light beam. Emitted light was passed retrograde through the other arm of the bifurcated light guide and then through a 525 ± 30-nm filter to a Throm EMI 9828 photomultiplier tube. The demodulated fluorescence outputs (University of Pennsylvania Biomedical Instrumentation Group) and the raw force signals were converted to digital signals by a Metrabyte Dash 16 AD board and stored on a personal computer. Ca was measured according to the method of Bruschi et al. (6). Briefly, changing calcium concentrations can be correlated with the fluorescence ratio at 340/380 nm excitation after subtracting background fluorescence at each excitation wavelength. Background fluorescence was determined at the end of each experiment by lysing cells with 0.2 mM MnCl₂ solution, followed by 5 mM MnCl₂ solution to quench fura 2 fluorescence, then measuring autofluorescence at pertinent wavelengths (340- and 380-nm excitation, 525-nm emission). Data were used from experiments where fluorescence at 380 nm was at least 2.5 times as great as the background signal, and the 340- and 380-nm fluorescence signals changed in opposite direction in response to the agonist. Simultaneous changes in isometric force and 340/380 nm fluorescence ratios were recorded in response to 90 mM KCl (maximum response) and in response to hypoxia. Changes in both tension and fluorescence ratio at 340/380 nm to hypoxia were expressed as a percentage of the maximum response.

Chemicals. The composition of LHBS was as follows (in g/l): 8.74 NaCl, 0.22 KCl, 0.29 CaCl₂ · 2 H₂O, 0.14 MgSO · 7 H₂O, 0.72 HEPES acid form, 1.8 HEPES sodium salt, and 0.9 glucose, pH 7.8. The composition of HHBS was as follows (in g/l): 27.70 NaCl, 0.60 KCl, 0.75 CaCl₂ · 2 H₂O, 0.75 MgSO · 7 H₂O, 0.72 HEPES acid form, 1.82 HEPES sodium salt, and 1.00 glucose, pH 7.8. All chemicals were purchased from Sigma Chemical (St. Louis, MO).

Calculations. At the end of an experiment, the vessel was blotted on paper toweling, weighed, and vessel tension was normalized to wet weight, i.e., milligrams of tension per grams wet weight. Because the hypoxic responses of individual vessels were reproducible (22), a vessel served as its own control and treatment effects were statistically examined by paired t-test or repeated-measures tests. Results are presented as means ± SE. Student's t-test and ANOVA were used for comparisons between vessels. The fiducial limit of significance was set at P  0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Sea lamprey. Contractile force and [Ca²⁺]_c increased simultaneously when DA were exposed to either 90 mM KCl or hypoxia (Fig. 1A). A temporal correlation between the active force generated by HV in lamprey DA and a rise in [Ca²⁺]_c was noted in all vessels treated. The active force, normalized as the ratio of the force generated by HV to the force generated by depolarization with 90 mM KCl, was linearly related to the change in intracellular calcium, normalized as the ratio of the calcium signal produced by HV to the calcium signal produced during 90 mM KCl (Fig. 1B).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   A: representative traces showing simultaneous measurement of fura 2 signal [cytosolic calcium concentration ([Ca2+]c); top trace] and force (bottom trace) during stimulation with 90 mM KCl or hypoxia (N2). B: correlation between active force (in g) and [Ca2+]c in lamprey dorsal aortas (DA). The hypoxia tension (N2) and Ca responses are expressed relative to a maximal KCl (80 mM) response; n = number of fish.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Effects of extracellular calcium (Ca) and extracellular sodium (Na) on hypoxic vasoconstriction (HV) in lamprey DA. A: HV in zero [Ca2+]o is not different from control HV in normal (2 mM) [Ca2+]o. Restoration of [Ca2+]o produces a slight but insignificant increase in tension. B: removal of Na significantly increases tension compared with a control HV (N2). HV in the absence of Na is unaffected. Return to normoxia (Air) relaxes vessel to prehypoxic level; however, full relaxation is not achieved until [Na+]o is restored. C: in the absence of Ca, removal of Na produces a transient contraction that is weakly sustained. Under these conditions, HV is augmented, prehypoxic tension is not restored on return to air, and full relaxation is achieved only when [Na+]o is restored.

4

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Effects of L-type Ca2+ channel antagonist methoxyverapamil (D600) and agonist BAY K 8644 (BAY) on HV in lamprey DA and the effect of the sarco/endoplasmic reticulum Ca2+ ATPase pump inhibitor cyclopiazonic acid (CPA) on HV in D600-treated vessels. Values expressed as a percentage of an initial control (dashed line) hypoxia (N2) response ± SE; (n) = number of fish. *Significantly different from control response (P  0.05). Inset: effect of CPA on HV in untreated DA from 1 fish.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Effects of Ca2+-specific pharmacological agents on lamprey HV in zero [Ca2+]o; CPA, ionomycin (IO), ryanodine (RY), thapsigargin (TH), and caffeine added to TH-treated vessels. Values are means ± SE and are given as a percentage of control. *Significant difference from an initial HV (P  0.05); (n) = number of fish.

5

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Effect of repeated hypoxia (A) and norepinephrine (NE) treatment (B) on lamprey DA in zero [Ca2+]o. Values are expressed as a percentage of an initial HV in 2 mM Ca (means ± SE). Black bars are NE responses, dotted bars are hypoxia (N2), and striped bars represent N2 response in NE-stimulated vessels. Dashed horizontal lines indicate 50 and 100% of control HV. *,Significant difference from like control. *Significant difference from control HV in 2 mM [Ca2+]o; significant difference from control NE response in 2 mM [Ca2+]o. P  0.05; n = number of fish. Experiments were conducted over 2 consecutive days.

New Zealand hagfish. HV in hagfish was significantly reduced by 38.1 ± 4.8% of the previous control HV (n = 6 fish) in the absence of Ca. Subsequent restoration of [Ca²⁺]_o to normal (5 mM) increased HV to 156.8 ± 15.6% of control (Fig. 6A).

View larger version (14K): [in this window] [in a new window]   Fig. 6.   A: myography of the typical contractile response of isolated hagfish DA to hypoxia in 5 mM [Ca2+]o. B: effect of restoring 5 mM [Ca2+]o to the bath during hypoxia. C: response of hagfish DA to [Na+]o replacement with sucrose and to hypoxia (N2) in the absence of Na. The vessel does not relax on restoration of normoxia, but it fully relaxes when [Na+]o is restored.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present experiments showed that HV in systemic arteries from two cyclostomes, the sea lamprey and the New Zealand hagfish, is a calcium-dependent response that uses intracellular (lamprey) or intracellular and extracellular calcium stores (hagfish). The active force generated during HV was temporally correlated with changes in [Ca²⁺]_c in lamprey DA, indicative of the direct involvement of Ca²⁺ in the stimulus-response coupling of hypoxia to the activation of this vascular smooth muscle. To our knowledge, this is the first time active force and Ca measurements have been taken simultaneously in any fish vessel. Lamprey HV was reproducible over multiple hypoxic exposures in the absence of Ca. Pretreatment of tissues with an L-type Ca²⁺ channel antagonist had no effect on Ca entry during hypoxic episodes in either species, yet HV was significantly increased in lamprey DA treated with an L-type Ca²⁺ channel agonist. The magnitude of HV in lamprey DA was significantly reduced in zero Ca by treatment with ionomycin or RY, or by the addition of caffeine to TH-treated vessels. NE more effectively depleted HV-used Ca than repeated hypoxia, and prestimulation with NE augmented HV after separate NE and hypoxia responses were nearly abolished. Removal of sodium from the extracellular milieu inhibited relaxation of hypoxia-stimulated vessels. However, these responses varied between the two cyclostomes. Results clearly show that, although the specifics of Ca²⁺ use may be different in cyclostomes and mammals, an increase in [Ca²⁺]_c is a requisite for HV in cyclostomes as it is for HPV.

Ca²+ dependence. The calmodulin-dependent cross- bridge cycling between actin and myosin filaments that produces contraction in mammalian smooth muscle is mediated by an increase in [Ca²⁺]_c (1, 14), and an increase in [Ca²⁺]_c is, for the most part, temporally coupled with active tension during HPV (24). The present study shows for the first time that HV in lamprey is also a calcium-mediated process.

Contributions of Ca and Ca during HV. Ca is required for HPV in most mammalian tissues (8, 9, 13, 16, 25, 34); however, the relative contributions of Ca and Ca vary among species. Rat (25) and fetal lamb pulmonary vascular smooth muscles (8) have been shown to be completely dependent on Ca, and HPV was markedly decreased by removal of Ca in perfused ferret lungs (13). In isolated pulmonary arteries from the cat, Harder et al. (16) found that raising or lowering [Ca²⁺]_o produced concomitant changes in the magnitude of HPV, which could be abolished by Ca blockade. Ca release was shown to be an initial step in HPV of isolated rat pulmonary arteries (15) and of cultured pulmonary arterial smooth muscle cells (26). These studies also showed HPV to be ultimately dependent on Ca. Experimental conditions during the present lamprey and hagfish DA studies were nearly identical, excluding differences in saline composition. However, the Ca and Ca requirements between these species during HV were quite different.

Ca²+ channels. Large-conductance (L-type) Ca²⁺ channels appeared to be present in lamprey DA, because HV was augmented by pretreatment with the L-type channel agonist BAY K 8644 (Fig. 3). However, L-type channels were either inactive or insignificant during HV in these vessels, because HV was independent of Ca and was unaffected by D600 in the presence of 2 mM Ca (Fig. 3). Treatment with the SERCA pump inhibitor CPA augmented HV in normal [Ca²⁺]_o (Fig. 3, inset), and this augmentation was not affected by the presence of D600 (Fig. 3). Additionally, CPA had no effect on HV in zero Ca (Fig. 4). These results suggest that either the entry of Ca independent of L-type channels was responsible, in part, for the augmented HV or that CPA blocked the export of Ca.

SR Ca²+ receptors. Two specific receptors mediate Ca release from the SR in mammalian vascular smooth muscle; those sensitive to the plant alkaloid RY and those sensitive to inositol 1,4,5-trisphosphate (IP₃). RY-sensitive receptors are opened by RY, Ca²⁺, or caffeine (19). IP₃-sensitive receptors on the SR may be opened directly by Ca²⁺ (19). Both receptor types can occur in the same vascular smooth muscle (17). Calcium stores sensitive to IP₃ and RY appear to be organized into spatially distinct compartments in pulmonary arteries but are conjoined in renal arteries, allowing differential, agonist-dependent release of calcium in the former but not in the latter (18). SR receptors in lamprey DA have not been previously characterized. Our experiments indicate that these vessels also possess more than one type and that they may be spatially or functionally distinct.

Ca²+ cycling during HV. Ca²⁺ cycling during hypoxia in lamprey DA may involve the release and/or reuptake of Ca²⁺ by intracellular stores (hypoxia-dependent stores) that are spatially or functionally different from the Ca²⁺ stores used by other contractile agonists. Ca appeared to be more efficiently recycled during HV than during adrenergic stimulation in these fish, because in zero Ca, repeated hypoxic exposures (8 times) were required compared with one NE treatment before a significant drop in HV was registered (Fig. 5, A and B). These results suggest that lamprey DA possess efficient mechanisms for uptake of Ca into these hypoxia-dependent stores. Inhibition of Na⁺/Ca²⁺ exchange during hypoxia may enhance Ca uptake into the hypoxia-dependent stores by reducing Ca efflux.

Na+/Ca²⁺ exchange during HV. The present study suggests that lamprey DA possess an active Na⁺/Ca²⁺ antiporter and that this exchange mechanism may be inhibited during HV. However, our results indicate that neither Ca influx nor Na⁺/Ca²⁺ exchange is important during HV or recovery from HV in these vessels.

Perspectives