Functional studies of the small pulmonary arteries

Structure and composition of pulmonary arteries, capillaries, and veins

The pulmonary vasculature comprises three anatomic compartments connected in series: the arterial tree, an extensive capillary bed, and the venular tree. Although, in general, this vasculature is thin-walled, structure is nonetheless complex. Contributions to structure (and thus potentially to function) from cells other than endothelial and smooth muscle cells as well as those from the extracellular matrix should be considered. This review is multifaceted, bringing together information regarding (i) classification of pulmonary vessels, (ii) branching geometry in the pulmonary vascular tree, (iii) a quantitative view of structure based on morphometry of the vascular wall, (iv) the relationship of nerves, a variety of interstitial cells, matrix proteins, and striated myocytes to smooth muscle and endothelium in the vascular wall, (v) heterogeneity within cell populations and between vascular compartments, (vi) homo- and heterotypic cell-cell junctional complexes, and (vii) the relation of the pulmonary vasculature to that of airways. These issues for pulmonary vascular structure are compared, when data is available, across species from human to mouse and shrew. Data from studies utilizing vascular casting, light and electron microscopy, as well as models developed from those data, are discussed. Finally, the need for rigorous quantitative approaches to study of vascular structure in lung is highlighted.

Regional variation in electrically-evoked contractions of rabbit isolated pulmonary artery

1. Electrically-evoked contractions in different regions of the rabbit isolated pulmonary artery have been investigated using stimulation parameters generally assumed to stimulate nerves selectively. 2. In extrapulmonary artery, trains of stimuli (10 Hz; pulse width 0.1 ms) evoked monophasic contractions. In contrast, a biphasic contraction was evoked in the intrapulmonary artery consisting of an initial fast component followed by a secondary very long-lasting component. 3. The contraction in the extrapulmonary artery was prazosin-sensitive (1 micro M) whereas that in the intrapulmonary artery was prazosin-resistant. 4. alpha,beta-Methylene ATP (1 micro M), atropine (1 micro M), losartan (1 micro M), BIBO3304 (1 nM) or nifedipine (1 micro M) had no effect on the biphasic contraction of the intrapulmonary artery. Bretylium (2 micro M) abolished the contraction of extrapulmonary artery but only partially inhibited the initial component in the intra region with no effect on the second component. 5. Tetrodotoxin (0.3-1 micro M), abolished the contraction of extrapulmonary artery but only partially reduced the electrically-evoked contraction of intrapulmonary artery. 6. Removal of the endothelium and application of sulphisoxazole (0.6-22 micro M) had no effect. 7. Varying the resting tone on the arteries, or applying gadolinium, had no effect on contractions. 8. Using confocal microscopy and calcium imaging, reproducible whole cell calcium transients were evoked in individual smooth muscle cells in intact preparations but only when direct muscle stimulation was used (pulse width of 5-10 ms). No detectable changes in calcium were elicited when brief pulse widths were used (0.1-2 ms). 9. Together, these data suggest that noradrenaline is the neurotransmitter inducing contraction in extrapulmonary artery. Noradrenaline and sympathetic nerves appear to play a less important role in the intrapulmonary artery. The tetrodoxin-resistant component is not mediated by ATP, NPY, acetylcholine, angiotensins, ET-1, stretch-activation or Ca(2+) influx through L-type Ca(2+) channels. Smooth muscle cells do not appear to be damaged by the stimulation protocol. The mechanism underlying the long lasting contraction of intrapulmonary artery evoked by brief electrical stimuli remains to be elucidated.

Evidence for specific regional patterns of responses to different vasoconstrictors and vasodilators in sheep isolated pulmonary arteries and veins

1. Responses of large (5-7 mm in diameter) and medium sized (3-4 mm in diameter) branches of sheep isolated intrapulmonary arteries and veins and three groups of small pulmonary arteries (200, 500 and 1000 microns diameter) to the vasoconstrictors endothelin-1, 5-hydroxytryptamine (5-HT), noradrenaline and the thromboxane A2 mimetic, U46619, were examined. Also, relaxation responses to the endothelium-dependent vasodilators, acetylcholine (ACh), bradykinin and ionomycin and the endothelium-independent vasodilator, sodium nitroprusside (SNP), were studied to determine their predominant site of action within the pulmonary vasculature. 2. Endothelin-1 was the most potent vasoconstrictor tested in all vessels. The maximum response to endothelin-1, expressed as a percentage of the maximum contraction to KC1 depolarization, did not differ significantly between the different vessels. By contrast, pulmonary arteries greater than 200 microns in diameter failed to contract to U46619, whereas U46619 was a potent constrictor of large and medium-sized veins. 3. 5-HT caused similar contractions in all arteries > 200 microns in diameter, but the maximum response was significantly diminished in smaller arteries. By contrast, the maximum response to noradrenaline was progressively attenuated with decreasing arterial diameter. Both 5-HT and noradrenaline caused poor contractions in veins. Pulmonary veins were less sensitive to 5-HT than arteries and at low concentrations 5-HT caused relaxation. No change in sensitivity to noradrenaline was noted between the arteries and veins. 4. Relaxation responses to bradykinin and ionomycin decreased progressively along the pulmonary vascular tree and were nearly absent in large veins. Also, ACh was a poor relaxing agent of large and medium-sized arteries and failed to mediate any relaxation response in other vessel segments. Surprisingly the smallest arteries examined (approximately 200 microns in diameter) failed to relax to ionomycin, bradykinin and SNP. However, both the sensitivity and maximum relaxation to SNP were similar in all other arterial and venous segments. 5. In conclusion, marked regional differences in reactivity to both vasoconstrictors and vasodilators occur in arterial and venous segments of the sheep isolated pulmonary vasculature. Such specialization may have important implications for the regulation of resistance in this low tone vascular bed.

Heterogeneity of 5-HT responsiveness in different segments of rabbit arteries

The variation in sensitivity to serotonin (5-HT) as assessed by pD2 was investigated in isolated 25 arterial segments of different vascular regions obtained from rabbits. The responses of the vascular rings which were opened by a transverse cut to form rectangular strips were recorded isometrically. pD2 values for 5-HT of thoracic and abdominal aortic segments were significantly greater than those of visceral arterial branches. Maximal contractions produced by 5-HT in visceral segments were higher than those of thoracic and abdominal aortic segments. There was a positive correlation between the pD2 values for 5-HT and the radii and the thickness of the tunica media layer (r = 0.62 and r = 0.54, respectively, P less than 0.05). The results indicate that the activity of 5-HT varies regionally in different segments of the rabbit arteries and the sensitivity to 5-HT of the segments decreases as the radii of the vessels diminish.

Analysis of pulmonary vascular responses in cats to sympathetic nerve stimulation under elevated tone conditions. Evidence that neuronally released norepinephrine acts on alpha 1-, alpha 2-, and beta 2-adrenoceptors

The influence of an increase in vascular tone on responses to sympathetic nerve stimulation and the receptors mediating these responses were investigated in the pulmonary vascular bed of the cat. Under conditions of controlled blood flow and constant left atrial pressure, stimulation of the sympathetic nerves to the lung elicited a biphasic response characterized by an initial increase in lobar arterial pressure followed closely by a decrease. The response to nerve stimulation was reproducible with respect to time and was not altered when a delay coil was added to the perfusion circuit, indicating that the response was directly mediated. The increase in pressure was reduced by prazosin and by yohimbine, whereas the decrease in pressure was blocked by propranolol or ICI 118551. These data suggest that the pressor component of the response is mediated by alpha 1- and postjunctional alpha 2-adrenoceptors, whereas the depressor response is mediated by beta 2-receptors. The pressor response was enhanced by propranolol or ICI 118551, whereas the depressor response was enhanced by prazosin or yohimbine, suggesting that the response to nerve stimulation represents the net effect of the actions of neuronally released norepinephrine on alpha- and beta-receptors. The pressor response to nerve stimulation was enhanced when tone was elevated with a prostaglandin endoperoxide analogue and when beta-receptors were blocked. The effects of an increase in tone and a passive increase in pressure on responses to sympathetic nerve stimulation were different.(ABSTRACT TRUNCATED AT 250 WORDS)

Vasodilator responses to alpha-human-atrial natriuretic peptide in isolated omental and pulmonary arteries from rabbit and man

In isolated vessels from the pulmonary and mesenteric/omental arterial circulations of rabbit and man, the effects of synthetic alpha-human atrial natriuretic peptide (alpha-hANP) were investigated. The vessels had an outer diameter of 0.6-1.2 mm and were contracted by noradrenaline (NA) and prostaglandin (PG) F2 alpha (mesenteric/omental arteries) or by 5-hydroxytryptamine (5-HT) and PGF2 alpha (pulmonary arteries). It was found that in rabbits, mesenteric vessels contracted by NA and PGF2 alpha were not significantly relaxed by alpha-hANP. Rabbit pulmonary vessels contracted by PGF2 alpha or 5-HT were concentration-dependently relaxed by alpha-hANP (maximum 70%). Human omental vessels contracted by NA and PGF2 alpha showed a moderate (30%) relaxation after addition of the peptide. Human pulmonary arteries contracted by 5-HT were relaxed by alpha-hANP almost to baseline, while PGF2 alpha-contracted vessels showed a maximum relaxation of 65%. It is concluded that alpha-hANP has a relatively selective effect on pulmonary arterial vessels from rabbit as well as man. The results suggest that the peptide may be involved in the regulation of pulmonary vascular tone.

The adrenergic innervation of pulmonary vasculature in the normal and pulmonary hypertensive rat

It would appear that susceptibility to chronic proliferative pulmonary hypertension in response to chronic alveolar hypoxia is most severe in species in which adrenergic innervation of pulmonary arteries is reduced or lacking. Intrapulmonary arteries of the rat have been reported to lack adrenergic innervation by some workers but not others. Since the rat develops severe proliferative pulmonary hypertension in response to prolonged alveolar hypoxia, the different divisions of the lung vasculature of Sprague-Dawley rats were thoroughly examined to determine the presence or absence of an adrenergic innervation. The degree of innervation in normal rats was compared with that of rats developing pulmonary hypertension. Both in normal and experimental pulmonary hypertensive rats the pulmonary arteries, all their branches and the small pulmonary veins with a smooth muscle media were found to be devoid of adrenergic innervation. In contrast, the cardiac-like muscle in the media of large pulmonary veins, the bronchial arteries and the vasa vasorum of larger vessels were richly innervated by adrenergic nerves. Thus the increase in medial smooth muscle which occurs in pulmonary arteries during chronic alveolar hypoxia is independent of a pre-existing adrenergic innervation or of such an innervation newly derived from that of adjacent vessels or structures. This is in contrast to systemic vessels where it has been suggested that increased adrenergic activity and density of innervation may augment hypertrophy of the media in hypertensive animals. Adrenergic nerves are suggested to have a protective action on pulmonary vessels.

Species dependent relaxation of intrapulmonary arteries (IPA) of rabbits, dogs and humans by prostacyclin

Helically cut strips of successive IPA segments of rabbits, dogs and human patients were set up for isometric recording in vitro. High tone was produced by norepinephrine (NE, 3 microM). This tone was markedly reduced by prostacyclin (PGI2) in the secondary, tertiary and quaternary branches of human and canine pulmonary trunk. The IC50 values for PGI2 ranged from 22 to 503 nM, the human vessels being more sensitive to prostacyclin than canine IPA. Under these conditions, the primary and secondary branches of the rabbit pulmonary trunk were not relaxed by PGI2. The contractile potency of NE was determined in each pulmonary vessel studied. The secondary segments of rabbit IPA were about ten times as sensitive to NE (EC50 for NE: 38 +/- 7 nM) as compared to the secondary IPA from dogs and humans (EC50 values: 370 +/- 84 and 440 +/- 50, respectively). When high tone was induced by equieffective contractile concentrations of NE (3 microM for canine and human IPA and 0.3 microM for rabbit vessels), PGI2 was still less effective (P less than 0.01) in relaxing secondary IPA of rabbits (IC25: 220 +/- 55) than the corresponding segments of dogs and humans (IC25: 51 +/- 12 and 17 +/- 4, respectively). The difference between canine and human vessels was also significant (P less than 0.02). These results indicate that there is an interspecies difference in the sensitivity of IPA to NE and PGI2.

An ultrastructural comparison of neuromuscular relationships in blood vessels with functional and 'non-functional' neuromuscular transmission

The neuromuscular relationships of the guinea-pig renal artery, which has previously been shown to be noradrenergically innervated but 'non-functional', i.e. not responsive to nerve stimulation in vitro, is compared with those of two arteries where functional neuromuscular transmission occurs: the carotid artery, where responses to nerve stimulation are slow and of medium amplitude, and the mesenteric artery, where the responses to nerve stimulation are relatively fast and of large amplitude. Substantial differences of ultrastructure were demonstrated. In the renal artery, there were no nerve varicosities closer than 400 nm to the smooth muscle layer, all varicosities had connective tissue interposed between them and the muscle wall, and there were significantly fewer varicosities (6 +/- 3.1 mm-1) compared with the carotid (33 +/- 8.3 mm-1; P less than 0.05) or mesenteric (105 +/- 15.8 mm-1; P less than 0.001) arteries. In the carotid artery, a small group of varicosities (17%), with little interposed connective tissue, had an average neuromuscular gap of 1.5 micron, whilst the remainder were separated by an average of 4.3 micron and by cellular and other connective tissue elements from the nearest smooth muscle cell. In the mesenteric artery, about half of the nerve varicosities were closer than 400 nm to the muscle layer and 25% of the varicosities had no connective tissue except basal lamina interposed in the neuromuscular space. Differences were also shown in the ultrastructural appearance of the nerves: analysis of the neuronal vesicles showed that the majority of varicosities were noradrenergic in all three vessels, with non-adrenergic varicosities forming 6% of the total in the carotid artery and 32% of the total in the mesenteric artery. Noradrenaline content was greater in the mesenteric artery (0.37 +/- 0.052 ng mm-2) compared with the carotid (0.17 +/- 0.017 ng mm-2; P less than 0.01) and renal (0.11 +/- 0.024 ng mm-2; P less than 0.01) arteries. These differences appear to correlate with the differences of neuromuscular activity in the three arteries.

The response of microvessels in rat lung explants to incubation with norepinephrine

Because of the difficulty in isolating precapillary or postcapillary segments of the pulmonary microcirculation, there is little information on their reactivity even though they are important in normal lung function and play a crucial role in the pathogenesis of pulmonary hypertension. In order to study the reactivity of microvessels in lung we have developed a method in which thin slices of tissue are incubated in vitro with a vasoactive drug and are then fixed, embedded, and sectioned at a thickness of 1 micron. Vessel profiles are analyzed on a computer graphics tablet, and their degree of constriction is determined morphometrically. In this system, muscular arteries less than 200 microns in diameter in adult rat lung explants are moderately constricted, but they constrict further when incubated with 1-norepinephrine (10(-4), 10(-5) M), an effect maintained for 30 min. Nonmuscular vessels (arteries and veins) less than 100 microns in diameter also constrict in response to similar doses of norepinephrine. Although the effector cells remain to be identified, it is likely that pericytes and precursor muscle cells in the walls of these vessels are at least partly responsible.

Studies on segmental differences in sensitivity to adrenergic agonists in pulmonary arteries isolated from the guinea-pig

This study was designed to examine the existence and possible mechanisms of segmental differences in sensitivity to adrenergic substances in the pulmonary vasculature of the guinea pig. The main pulmonary artery was found to be less sensitive to epinephrine, norepinephrine and phenylephrine than its left and right main branches. Because there were no differences in sensitivities between the left and right branches, most comparisons were made between the main and left branches. The difference in sensitivity was larger for adrenergic agonists which are known to be taken up by adrenergic neurons (epinephrine, norepinephrine, phenylephrine) than for an agonist which is not taken up by these neurons (methoxamine). The left artery from reserpine-treated animals was only slightly more sensitive (1.5-fold) to barium chloride than the main pulmonary artery. Cocaine potentiated to a greater extent in the main than in the left pulmonary artery the amines which are taken up by adrenergic neurons. The left arterial branch was less responsive to transmural electrical stimulation than the main artery in the absence and presence of cocaine. [3H]Norepinephrine uptake was larger in main arterial segments than in the branches and histochemical fluorescence studies demonstrated a greater degree of adrenergic innervation in the proximal segments. The results indicate that a major explanation for the differences in sensitivity between the vascular segments is a smaller degree of adrenergic neuronal uptake of agonist in the arterial branches.

Effects of SQ 14.225, an orally active inhibitor of angiotensin-converting enzyme, on hypoxic pulmonary hypertension and right ventricular hypertrophy in rats

Treatment with SQ 14.225, an inhibitor of the enzyme converting angiotensin I to angiotensin II and lowering of plasma renin activity by dietary salt overload, significantly decreased hypertrophy of the right ventricle in chronically hypoxic rats. The smaller degree of right ventricular hypertrophy registered in hypoxic rats with suppressed activity of the renin-angiotensin system was proportional to the decrease in hypoxic pulmonary hypertension noted in these animals. Lowering of pulmonary hypertension in chronically hypoxic rats by suppression of the renin-angiotensin system is consistent with the idea that angiotensin II may be involved in eliciting hypoxic pulmonary vasoconstriction. Inhibition of converting enzyme also lowered the systemic arterial pressure in hypoxic rats, but this finding could not be taken as evidence for the role of the renin-angiotensin system in homeostatic control of blood pressure in normal animals, since the plasma-renin activity in hypoxic rats is increased and the blood pressure higher than in rats kept under normoxic conditions.

Identification and characterization of adrenergic receptors and catecholamine-stimulated adenylate cyclase in hog pial membranes

In order to define the adrenergic receptors in small pial blood vessels, we studied [3H]dihydroalprenolol (DHA) binding, [3H]dihydroergocryptine (DHE) binding, and catecholamine-stimulated adenylate cyclase in membranes prepared from the hog pia. [3H]dha bound to a saturable number of sites (0.09 pmol/mg protein) with a high affinity to (10 mM). The ability of a small series of adrenergic agents to compete for [3H]DHA binding was as expected for binding to beta-adrenergic membrane receptors. The receptor binding appeared to be composed largely of beta 2-adrenergic sites but some sites with beta 1 properties were detected. The beta-adrenergic agonist, isoproterenol, stimulated adenylate cyclase activity by 20% in pial membranes in the presence of GTP. The potency of catecholamines in stimulating adenylate cyclase correlated well with theif sites (0.39 pmol/mg) on these membranes. alpha-Adrenergic agents were potent competitors for [3H]DHE binding but dopamine, serotonin and histamine had effects only at high concentrations. These data demonstrate the presence of alpha- and beta-adrenergic membrane receptors and beta-adrenergically stimulated adenylate cyclase activity in small pial blood vessels from the hog.

Quantitative analysis of the density and pattern of adrenergic innervation of blood vessels. A new method

Automated quantitative image analysis (QIAF) was used to measure and compare the adrenergic nerve plexuses of 4 blood vessels from the guinea pig, demonstrated by glyoxylic acid fluorescence (GAF). The results showed considerable quantitative variation of plexus density, size of bundles, and numbers of varicosities. A range of alternative procedural and anatomical sources of variability were investigated and assessed. The carotid artery was found to have a dense plexus with more nerves than that of the mesenteric artery; the mesenteric vein and abdominal aorta had sparse plexuses. The carotid artery plexus, despite the density of its nerves, possessed only half the number of varicosities of the mesenteric artery plexus. This sparse varicosity population was shown to have a similar density to the varicosities demonstrated by QIAF in the scattered nerves of the mesenteric vein and abdominal aorta. QIAF confirmed visual estimates of adrenergic plexus density, and was able to demonstrate less obvious differences of nerve density and size, and varicosity populations, between the different plexuses studied. The method is applicable to stretch preparations and transverse sections of many adrenergically innervated tissues.

Some bases of differences in vascular response to sympathetic activity

Basic patterns of neuroeffector organization vary widely in the vasculature, in general, with vessel diameter and type, and confer distinctive properties. The smaller the vessel, the more intimate the neuroeffector relationship, the more localized the action of the released transmitter, and the more important myogenic conduction compared to transmitter diffusion for the coordination of vascular effector response. Seemingly superimposed upon these basic general patterns are other variable features, conferring upon vessels of similar size and type diversity of function. These variables include sensitivity and magnitude and possible location of alpha- and beta-receptors and their subtypes, presence and nature of intrinsic vascular tone, and the density and pattern of adrenergic innervation to mention the more important. Functional diversity in neuroeffector characteristics can, to some extent, be understood in relation to embryological development, neurotrophic influences, effector regulation of innervation, and the mural response to an increase in intravascular pressure.