Hypoxia and the Pulmonary Microvasculature

were prepared fur histologic processing and stained with hematoxylin and eosin and elastic stains. The presence ofPPA was determined by the mean muscle area (MMA) ratio, which is [(total vessel area - luminal area)/ total vessel area]. Areas were determined by planimetry on elastic stained slides of30-100-J.L diameter vessels. Multiple measurements were made in each lobe. An analysis of variance was used to test the differences between the left and right lung MMA ratios. As we found previously, the mean pulmonary artery pressure rose only slightly, from 15±1.5 mm Hg to 16.4±2 mm Hg (p = .0466). The Pa02 remained normal throughout (86 ± 4 mm Hg at baseline; 95 ± 10 mm Hg at 6 weeks). The normal PTT in dogs is 12 seconds. The mean PTTwas slightly elevated 1 and 4 hours after a dose ofheparin (14.7 ± .12 sand 15.1± .1s) but returned to the normal value of12.1± .09 sjust before the next dose. Thus, the animals had a minimal elevation of the PTT but not in the antithrombotic range (1.5N). In contrast to the animals we studied previously, who had undergone only left pulmonary artery ligation, the MMA ratio was not significantly different in the left and right lungs in these animals who received heparin after pulmonary artery ligation (.486± .0259 vs .513± .033; p= .362.) We conclude that heparin appears to inhibit the development of PPA which is a nonhypoxic model of pulmonary vascular remodeling. With the data mentioned earlier, this finding suggests that heparin may be a common regulator of pulmonary vascular smooth muscle growth. It also suggests that endothelial cell injury, allowing derepression of heparinmediated endothelial cell control over smooth muscle cell growth, may occur in this model. REFERENCES 1 Shure D, Swain JA, Abraham JL, Moser KM. Pulmonary arterial hypertensive changes distal to a central vascular obstruction in the dog. Am Rev Respir Dis 1984; 129:331 2 Shure D, Dockweiler DW, Peters RM. Post-obstructive pulmonary arteriopathy occurs without elevation of distal (obstructed) pulmonary artery pressure. Am Rev Respir Dis 1985; 131:402 3 Shure D, Dockweiler DW, Lammers RJ, Bloor CM. Post-obstructive pulmonary arteriopathy is related to the extent of bronchopulmonary anastomotic fOrmation in an animal model. Am Rev Respir Dis 1986; 133:161A 4 Hoover RL, Rosenberg R, Haering W, Kamovsky MJ. Inhibition of rat arterial smooth muscle cell proliferation by heparin: in vitro studies. Circ Res 1980; 47:578-83 5 Guyton JR, Rosenberg RD, Clowes AW, Karnovsky MJ. Inhibition of rat arterial smooth muscle cell proliferation by heparin: in vivo studies with anticoagulant and nonanticoagulant heparin. Circ Res 1980; 46:625-34 6 Hales CA, Kradin RL, Brandstetter RD, Zhu Y-J. Impairment of hypoxic pulmonary artery remodeling by heparin in mice. Am Rev Respir Dis 1983; 128:747-51

Hypoxia and the Pulmonary Microvasculature* Physiology and Pathophysiology Sidney S. Sobin, M.D., Ph.D.

interest in the effect of hypoxia on the pulmonary Ourmicrocirculation was its use as a tool to answer the

question: what is the origin of the vascular smooth muscle (VSM) of the pulmonary arterioles in adult pulmonary hypertension? Although "true" pulmonary arterioles are not present in the adult mammalian lung, 1 traces ofVSM persist even in the aged lung arterioles. 2 We wished to distinguish between possible recall or activation ofVSM and the de nova formation ofVSM and believed that hypoxic challenge was an appropriate tool. With 10% 0 2 in nitrogen or Y2 atm exposure, we found rapid changes in the prealveolar (acinar) arterioles of less than 30 J.L diameter. 3 Blebs developed between the endothelium and its basal lamina, and edema occurred in the arteriolar wall connective tissue. These were present after 1 h of hypoxia (Fig 1); 5% 0 2 produced similar changes after only 15 min. Fibroblasts, normally present in the arteriolar wall at the intercept with the alveolar wall, tripled in number by 24 h, migrated subendothelially, and during the next 24-48 h developed a basal lamina, myofibrils, lost their rich endoplasmic reticulum, and completed their transformation into typical VSM by 3-4 days. The endothelium of these muscularized vessels appeared normal after 48 h. These changes of subendothelial blebbing, wall edema, and increased number of fibroblasts were not present in the capillary bed, small venules, or proximal small arteries. The number of platelets per arteriolar cross-section increased by 24 h and peaked at 48 h, with an ocsasional platelet adherent to the endothelium. The rapid onset of these sequential changes in a limited segment of the pulmonary microvasculature with hypoxia suggested an anatomic cascade triggered by a biochemical sequence. We thought an endothelium-derived arachidonic acid (AA) release could initiate such a cascade of events. Therefore, in the anesthetized rat we infused mepacrine IV and recorded pressure from the pulmonary artery prior to and during 10% 0 2 challenge. This was effective in blocking the acute rise in pulmonary pressure. In parallel experiments with mepacrine blockage of hypoxic response, AA, and PGF..,each produced an increase in pulmonary artery pressure, indicating persistent vasoactive response. Blockage of hypoxic pulmonary hypertension by mepacrine for up to 4 h markedly reduced the immediate anatomic changes in the arteriolar wall seen during hypoxia when examined by light microscopy.• Further review of EMs now shows an accumulation of platelets at the endothelium, with the presence of membrane bound densely staining bodies, circular in section, and without the characteristic structure ofWeibel-Palade bodies. They accumulate rapidly and are profuse at 1 h of 10% oxygen, persist at 24 h, and begin to disappear at 48 h. *From the Department of Physiology, University of Southern CalifOrnia School of Medicine, Los Angeles, and Department of Biology, University of California, San Diego, La/olla. Reprint requests: Dr. Sobin, 336 13th Street, De Mar; California

92014

CHEST I 93 I 3 I MARCH, 1988 I Supplement

1558

N

eutrophils (PMN) may contribute to vascular injury under ischemic conditions. For example, in studies in dogs, PMN were found adherent to endothelium of blood vessels within ischemic areas of myocardium after 60 minutes of experimental coronary artery occlusion. Within 30 minutes ofsubsequent reperfusion, PMN migrated into ischemic myocardium. 1 However, the stimulus for accumulation of PMN in ischemic tissue is unknown. We have previously shown that cultured endothelial cells release neutrophil chemoattractant activity in response to a variety of vasoactive agents, such as angiotensin II• and histamine. 3 In addition, others have shown that endothelial cell metabolic activities are altered under hypoxic conditions. • Therefore, we sought to determine whether endothelial cells are capable of contributing to PMN accumulation under conditions which might exist in ischemic tissues where there is stasis of blood

How.

In these experiments we determined whether cultured endothelial cells from a variety of vascular beds released chemoattractant activity for PMN after exposure to hypoxia. In addition, we assessed whether hypoxia or hypercarbia caused changes in endothelial cell surface substances or altered adherence of PMN to the endothelial cell surface. METHODS

FIGURE 1. Pulmonary arteriole after 1 h10'11 oxygen exposure, 22 1.1. diameter. Extensive edema of the wall and dissolution of some of the cellular elements, probably fibroblasts. The endothelium is structurally Intact. (original magnification, x 1,650, E-M section). Platelets are abundant in that same period. These dense bodies are not present in the endothelium of systemic arteries, and their nature is not known. We believe the acute hypoxic response may be triggered by an endothelium-derived AA release, which in turn may react with a variety of other factors, such as complement C~. platelet, and polymorphonuclear leukocyte factors, to set in motion the anatomic cascade we have described. REFERENCES 1 Brenner 0. Pathology of the vessels of the pulmonary circulation. Arch Intern Med 1935; 56:211-337 2 Sobin SS, Lindal RG, Bernick S. The pulmonary arteriole. Microvasc Res 1977; 14:227-39 3 Sobin SS, 1remer HM, Hardy JH, Chiodi HP. Changes In arteriole In acute and chronic hypoxic pulmonary hypertension and recovery In rat. J Appl Physioll983; 55:1445-55 4 Jiang HX, Sobin SS, 1remer HM, Watkins GL. The effect of mepacrine on acute hypoxic pulmonary hypertension. Fed Proc 1986; 45:556

Effects of Hypoxia and Hypercarbla on Cultured Endothelial Cells* Sharon Rounds, M.D.; Haf'T'ilon W Farber, M.D.; Marta L. Render, M.D. ; and Faith A Barnard, M.D. *From The Pulmonary Center, Boston University School of Medicine, Boston. Supported by grants from the NHLBI (HL 34009, 07035, 19717). 1588

Endothelial cells were isolated from bovine aorta, pulmonary arte114 and coronary arte114 and from human umbilical vein and maintained in culture using standard methods.U Cultured monolayers were incubated at 37"C fur 4 hours In varying oxygen concentrations-21%, 10'11, 3%, and 0'11 0 1; 5% C01 ; balance, N1• Culture supernatants were then removed and assessed fur PMN chemoattractant activity using a modified Boyden chamber assay and human PMN as target cells.L3 To assess the effects of hypoxia and hypercarbia on endothelial cell surfilce substances, other monolayers of bovine aortic and pulmonary artery cells were incubated at 37"C In either 0'11 0 1 , 5% CO,; 21% 0 1 , 5% C01 ; 0% 0 1 , 10% C01 ; or 21% 0 1 , 10% CO,. After 4 hours, surface substances were radiolabeled using 1311 and lactoperoxidase-glucose oxidase conjugated to Latex beads. The labeled monolayers were extracted with detergent (NOG~ the resulting extract separated using 15% SDS-PAGE gel electrophoresis, and labeled substances were identified on lluorograms of the gels. Other monolayers of bovine aortic cells were incubated fur 4 hours in varying 0 1 and C01 concentrations as described above, and the adherence to the monolayers of 51Cr-labeled human PMN assessed. RESULTS

We found that hypoxic cultured endothelial cells released chemoattractant activity for PMN into culture supernatants in a variable fashion, dependent on 0 1 concentration and cell origin (Fig 1~ Supernatants from bovine aortic and coronary artery and human umbilical vein endothelial cells released chemoattractant activity in response to 10% and 3% 0 1 , while supernatants from cells incubated in 0% 0 2 did not stimulate PMN migration. In contrast, supernatants from bovine pulmonary artery cells only stimulated PMN migration when the 0 1 concentration was lowered to 0%. We found that this release of PMN chemoattractant activity under hypoxic conditions was prevented by incubation of endothelial cells with diethylcarbazine or eicosatetraynoic