INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

IT IS WELL KNOWN that when mammalian species (including humans) that live at sea level are exposed to high altitude, certain species develop pulmonary hypertension with increased muscularization of the pulmonary arterioles (5, 36, 41). The degree of pulmonary hypertension varies among species (44) and is related to the duration of high-altitude residence and magnitude of the hypoxic stimulus. Mild pulmonary hypertension by acute hypoxia is an important adaptive mechanism that improves the matching of ventilation and perfusion, which minimizes arterial hypoxemia (20). However, severe and long-term pulmonary hypertension in humans can induce an increase in workload for the right ventricle, limit the cardiac output reserve, and possibly lead to right ventricular failure, i.e., the symptoms of "high-altitude heart disease," which can be fatal (50). The most striking feature of this disease at histological examination is severe medial hypertrophy of small pulmonary arteries with crenation of the elastic lamina, the incidence of which in Qinghai Province, People's Republic of China, was found to be 2.2% in adults and 5.1% in children, i.e., rather rare in Tibetan people (50).

Evidence from previous studies (3, 16, 29, 41), however, shows that species that are long-term residents at high altitudes, such as yaks, snow pigs, and llamas, maintain a low pulmonary arterial pressure with an absence of highly muscularized pulmonary arterioles, despite living at very high altitudes. This may be an adaptation to chronic hypoxia by genetic transmission (3).

The pika (Ochotona curzoniae), a member of the Ochotonidea family, is a small rodent that lives in remote mountain areas at very high altitudes (28). Pika fossil samples found on the north edge of the Qinghai-Tibetan plateau are ~37 million years old. The pika is thus an animal model for the study of the mechanism of adaptation to an hypoxic environment (14). Previous studies of pikas at high altitude (14, 15) have examined the effect of hypoxia on the liver and neuroendocrine system, but there is little information on the properties of the pulmonary circulation in pikas. In this study, we examined the hemodynamics, histology, immunohistochemistry, blood gas, and hematology of pikas and rats at a high altitude. Of particular interest was whether the structural changes of the pulmonary arteries in chronic hypoxia are mediated by transforming growth factor (TGF)-.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Ten male pikas and 10 male Wistar rats were used for this study. Pikas, weighing 100-180 g, were captured at an altitude of 4,300 m by traps and then transported to Xining (altitude 2,260 m), Qinghai, People's Republic of China, where the study was conducted from 2 to 4 days after their arrival. The Wistar rats, weighing 160-260 g, were born and raised at Xining.

The study was conducted in a large decompression chamber at an altitude of 2,260 m at the Qinghai High Altitude Medical Sciences Institute in Xining. The chamber, a large room (30 × 10 ft), was connected to a smaller room (10 × 9 ft). The temperature and relative humidity in the chambers were maintained between 18 and 25°C and 40 and 50%, respectively. After baseline measurements were made at 2,260 m (580 Torr), the animals were exposed to simulated altitudes of 4,300 (450 Torr) and 5,000 (405 Torr) m in the decompression chamber.

Hemodynamic study. Each animal was anesthetized with pentobarbital sodium (30-50 mg/kg). The right external jugular vein was isolated, and then a curved-tip polyvinylene catheter (ID 0.28 mm) was inserted and advanced to both the right ventricle and main pulmonary artery. Another catheter (PE-50, ID 0.58 mm) was inserted into the exposed left carotid artery. Correct placement was confirmed by the typical pressure wave tracing seen on an oscilloscope. The catheter was flushed and filled with heparinized saline and closed with a blunted wire plug, and then both the pulmonary and aortic catheters were exteriorized at the back of the neck of the animal by the method of Rabinovitch et al. (36). After 12 h were allowed for the animal's recovery, measurements of pulmonary and systemic arterial pressures were taken with calibrated pressure transducers (MPU-0.5, Nihon Koden, Tokyo, Japan) and logged with an electronic two-channel recorder (WT-685G). The zero reference for pressures was the level of the left atrium. Hemodynamic measurements were obtained with disposable transducers 1 h after arrival at each simulated altitude while the animals were in the conscious state and still in the chamber.

Blood gas and hematologic studies. Arterial blood samples were drawn from the systemic arterial catheter for the measurement of pH, arterial PCO₂ (Pa_CO2) and arterial PO₂ (Pa_O2) (model 1303, Instrumentation Laboratories) at baseline and at the simulated altitude of 5,000 m. Additionally, in venous blood, measurements were made of 1) concentration of Hb with cyanmethemoglobin, 2) hematocrit (Hct) with a microcentrifugation method, and 3) concentration of red cell 2,3-diphosphoglycerate (2,3-DPG) with a calorimetric assay (Boehringer Mannheim, Mannheim, Germany) using the DPG standard for quantification. These measurements were all done at the baseline altitude (2,260 m).

Histology and morphological analysis. After completion of the pressure and blood gas measurements at simulated 5,000-m altitude, the animals were anesthetized and killed by exsanguination. The chest was then opened, and the heart and lungs were removed en bloc. A block of tissue was taken from the right and left lungs and suspended in 10% formaldehyde solution for 1 wk. The tissues were embedded in paraffin and cut into 4-µm-thick sections. The sections were stained with hematoxylin and eosin, toluidine blue, and elastica-Van Gieson stain. The vessel diameter and wall thickness were measured by a digital analyzer. A minimum of 10 small arteries from each tissue were sampled.

Immunohistochemical staining. Paraffin-embedded lung tissues were cut at 4 µm and immunostained for mast cell tryptase and TGF-₁ by the avidin-biotin complex method (31, 46). The tissues were pretreated with 0.23% pepsin in PBS (hyaluronidase 1 mg/ml in PBS) for 30 min at 37°C after endogenous peroxidase was blocked with 0.5% H₂O₂ in methanol for 20 min at room temperature, and for demonstration of mast cell tryptase, the sections were treated with 1% trypsin (Sigma Chemical, St. Louis, MO) for 30 min at 37°C and then incubated in PBS containing monoclonal mouse anti-human mast cell tryptase (DAKO-Mast Cell, no. M7052; 1:100 dilution) and monoclonal mouse anti-TGF-₁ (no. 80-1835-03; 1:100 dilution) at 4°C for 18 h. After being rinsed in PBS, sections incubated with monoclonal antibodies were treated in solutions containing horseradish peroxidase-conjugated goat anti-mouse IgG antibody (diluted to 1:100 in PBS). After further rinsing in PBS, antibody or lectin binding sites were visualized with 0.05 M phosphate buffer, pH 7.4, containing 3,3'-diaminobenzidine (Wako Pure Chemicals, Tokyo) and hydrogen peroxide. These sections were counterstained with hematoxylin.

Statistical analysis. Data are expressed as means ± SE. Hemodynamic data at different altitudes were analyzed by means of a two-way analysis of variance with repeated measures. Comparisons between groups were made by Fisher's protected least significant difference test at the 95% significance level. Blood gas values at baseline and at simulated 5,000-m altitude were compared with a paired Student's t-test. Student's unpaired t-test was used for comparison of the pika and rat groups for other data. Linear regression analysis and correlation coefficients were used to assess the relationships between variables. Comparisons and correlations were considered significant when P < 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Hemodynamics. At baseline (2,260 m), the mean pulmonary arterial pressure (P_pa) of the pikas was significantly lower than that of the Wistar rats (11.5 ± 0.5 vs. 21.8 ± 1.1 mmHg; P < 0.01). At the simulated altitudes, P_pa in the pikas was not changed from baseline (12.4 ± 0.9 mmHg, not significant) at 4,300 m and slightly increased to 14.6 ± 1.2 mmHg at 5,000 m (P < 0.05). In contrast, P_pa of the rats was significantly increased with increasing altitude (21.8 ± 1.1 mmHg at 2,260 m, 32.1 ± 1.7 mmHg at 4,300 m, and 38.8 ± 2.5 mmHg at 5,000 m; Fig. 1). Changes in P_pa (P_pa) from 2,260 to 4,300 and 5,000 m were 1.48 ± 0.49 and 4.80 ± 0.67 mmHg in the pikas and 10.38 ± 3.36 and 19.10 ± 2.28 mmHg in the rats (Fig. 2). Systemic arterial pressure and heart rate in both groups were similar at all altitudes (Fig. 1).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Changes in mean heart rate (HR, A), systemic pressure (Psa, B), and pulmonary arterial pressure (Ppa, C) in pikas and rats at 2,260, 4,300, and 5,000 m. Values are means ± SE. * P < 0.05 vs. 2,300 m; ** P < 0.01 vs. 2,300 m; ++ P < 0.01 between pikas and rats.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Changes in mean Ppa (Ppa) in pikas and Wistar rats from 2,260 to 4,300 and 5,000 m. Values are means ± SE. * P < 0.01.

Heart weight. The comparison of the heart weights of pikas and rats is shown in Table 1. There was a marked right ventricular hypertrophy in the rats compared with the pikas. The calculated RV/(LV+S) was 0.22 in the pikas and 0.45 in the rats.

Lung histology. The results of the light microscopic examination of the pulmonary vessels are also shown in Table 1. The wall of the small pulmonary arteries of the pikas was extremely thin and without smooth muscle cells compared with those of the rats (Fig. 3, A and B). Figure 3 (inset) also shows that the smaller pulmonary arteries (100 µm) have a wall composed of only a single elastic lamina in pikas, whereas the corresponding arteries in the rats have a thick muscular wall composed of thick smooth muscle sandwiched between inner and outer elastic lamina (Fig. 3B, inset). The average percent thickness of the medial pulmonary arteries was 9.22% in the pikas and 27.21% in the rats, which was well correlated with mean P_pa, and the slopes of the regression lines were significantly different between groups (Fig. 4). In addition, mast cells were observed only in the rats; the pikas had no mast cells in the lung (Fig. 3, C and D). Among 10 rats, the number of mast cells ranged widely, from 10.4 to 23.1 cells/mm²; the mean was 7.1 ± 0.33 cells/mm².

View larger version (121K): [in this window] [in a new window]   Fig. 3.   A: lung tissue of a pika, elastica-Van Gieson staining, ×25. Wall of a pulmonary arteriole accompanying a bronchiolus was very thin. B: lung tissue of Wistar rat, elastica-Van Gieson staining, ×25. Pulmonary arteriole wall with dense elastic fibers was much thicker than that of pika. In higher-power magnifications of pulmonary smaller arteries (×400, insets in A and B), there is a very thin-walled arteriole (70 µm) with a wall consisting of a single elastic lamina in the pika (inset, A), and there is a thick muscular wall composed of thick smooth muscle sandwiched between inner and outer elastic lamina in the rat (inset, B). C: lung tissue of a pika, toluidine blue staining, ×100. No mast cells were found. D: lung tissue of a Wistar rat, toluidine blue staining, ×100. Many mast cells were observed in the connective tissue around a pulmonary artery.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Correlation between level of mean Ppa and percent wall thickness of pulmonary arterioles in pikas and rats (n = 10 each group). , y = 11.765 + 0.33743x, r = 0.678, P < 0.05; , y = 36.756 + 2.7900x, r = 0.769, P < 0.01.

Lung immunohistochemistry. There was intense immunoperoxidase staining for mast cell tryptase in the rats, and tryptase-positive mast cells were mainly seen around small vessels (Fig. 5B), whereas no such demonstrable immunoreaction was observed in the pikas (Fig. 5A). The distribution of immunostaining corresponded with that seen when adjacent serial sections were stained using a standard procedure with toluidine blue.

View larger version (141K): [in this window] [in a new window]   Fig. 5.   A: lung tissue of a pika. Immunostaining with anti-tryptase antibody, ×200. No mast cells were observed. B: lung tissue of a Wistar rat. Immunostaining with anti-tryptase antibody, ×200. Anti-tryptase-positive mast cells were found in the connective tissue around a small pulmonary artery. C: lung tissue of a pika. Immunostaining with anti-transforming growth factor (TGF)-1 antibody, ×200. No cells were stained. D: lung tissue of a Wistar rat. Immunostaining with anti-TGF-1 antibody, ×200. Neutrophils and mononuclear cells were stained in connective tissues between a pulmonary artery and a bronchiole. In higher-power magnifications (×400, inset in D), neutrophils and mononuclear cells are clearly indicated.

Blood gas and hematology. Tables 1 and 2 show the results of the arterial blood gas, Hb, Hct, and 2,3-DPG analyses. At 2,260 m, there was no significant difference in any blood gas parameter between the pikas and rats. Pa_O2 decreased significantly after the simulated altitude of 5,000 m, but this decrease was more conspicuous in the rats than in the pikas; Pa_CO2 also decreased and pH increased, but these parameters were not significantly different between the two groups. Both Hb and Hct of the pikas were markedly lower than those of the rats, and 2,3-DPG concentration was also lower in the pikas than in the rats.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The present studies indicate that pikas have lower hypoxic pulmonary vasoconstriction, thin-walled pulmonary arteries, and no right ventricular hypertrophy compared with Wistar rats. Both mast cell tryptase and TGF-₁ were intensely immunostained on the lungs of the rats but not in the pikas; Hct, Hb, and 2,3-DPG of the pikas were significantly lower than those of the rats.

The hemodynamic measurements used in the present study have the advantages of being direct readings of P_pa and systemic arterial pressure while the animals were still in the hypobaric chamber and of being uninfluenced by anesthesia. P_pa did not rise significantly in the pikas with the increase in altitude but rose significantly in the rats (Fig. 1). The findings presented here clearly demonstrate that the pulmonary vasoconstrictor response to acute hypoxia was much smaller in the pikas than in the rats. This observation is in agreement with reports on other indigenous mountain species such as the yak, snow pig, and mountain viscacha (16, 30, 41) but inconsistent with that on the llama by Banchero et al. (5). They found that the mean P_pa of three llamas born and bred at an altitude of 250 m rose from an average of 14 mmHg to an average of 23 mmHg after 10-wk residence at 3,400 m. Our present data, in contrast, are quite consistent with the observation by Harris et al. (23), who measured P_pa in 12 llamas at 4,200 m and found that mean P_pa was 14 mmHg, which is not substantially different from the figure of 12 mmHg that we found in pikas at 4,300 m. In these llamas, the muscular pulmonary arteries proved to be very thin walled, and the pulmonary arterioles had a wall that consisted of a single elastic lamina, as we found in the pikas. In humans, it has been shown that the native highlanders in the Himalayas have normal P_pa (similar to that of persons residing at sea level). Groves et al. (19) reported that the mean P_pa in Tibetans at 3,658 m was 15 mmHg at rest and did not significantly increase with further hypoxia.

The pulmonary hemodynamics in response to hypoxia are associated with extensive structural changes of the pulmonary arterial bed, which were well documented by Tucker et al. (44). The morphological analysis in the present study has shown that the pika maintains exceedingly thin-walled pulmonary arteries with only a single elastic lamina without smooth muscle cells, findings that are quite consistent with those in humans (21) and animals (3, 16, 29, 30) that have lived only at high altitude.

In the Wistar rats, we found that the walls of small pulmonary arteries were thick and composed of tightly packed, circularly oriented smooth muscle cells. This finding supports the previous concept that the impact of chronic alveolar hypoxia on pulmonary vessels in unadapted animals produces a complex remodeling with the development of vascular smooth muscle so that the pulmonary arterioles become muscularized. Stenmark et al. (40) studied newborn calves exposed to a simulated altitude of 4,300 m and found that these animals developed severe pulmonary hypertension, right ventricular hypertrophy, and markedly increased wall thickness of pulmonary arteries in just 2 wk at high altitude. Rabinovitch et al. (36) and Will et al. (47) reported that P_pa rose significantly with chronic exposure to hypobaric hypoxia, and the degree of pulmonary hypertension was well correlated with the wall thickness of the pulmonary small arteries. The factors responsible for the maintenance of thin-walled pulmonary vessels in animals at high altitude and for the increase in wall thickness in pulmonary hypertensive animals have not been identified. Durmowicz et al. (16) demonstrated that the endothelial cells of small pulmonary arteries in the yak were longer, wider, and more rounded than those of cows, suggesting that the endothelial regulation of pulmonary vascular tone may be different in these two species.

In the present study, we observed a number of mast cells around the pulmonary arterioles in the rats but not in the pikas. Previous studies showed that the increase in lung mast cells in sea level animals is related to hypoxic exposure (49). Kay et al. (32) found that the lung mast cell density was doubled in rats exposed for 20 days to a simulated altitude of 5,500 m. Haas and Bergofsky (26) and Ahmed (2) showed degranulation of perivascular mast cells in rats and sheep after acute alveolar hypoxia. The reason for the lack of mast cells in the pika is unclear. Pikas have lived at high altitude longer than any other animal species, which has induced some specific adaptations to hypoxic environments. We consider that long-term residents of high altitude are best served by an absence of both pulmonary hypertension and vascular remodeling, which in these animals may be related to few or no mast cells in the lungs. Thus it is likely that the lack of mast cells in the pika is the result of natural selection through many thousands of generations at high altitude.

The causal role of mast cells in hypoxic pulmonary hypertension remains controversial. At the outset, evidence suggested that alveolar hypoxia causes mast cells to liberate chemical mediators such as histamine that induce pulmonary vasoconstriction (2, 26). This concept has been challenged by findings that the accumulation of mast cells was more likely to be a response to the muscularity of pulmonary vessels. Heath (27) reported from La Paz, Bolivia, at 3,600 m that the native highlanders without pulmonary vascular remodeling had a normal density of mast cells, as found in normoxic lowlanders, and that the highlanders with vascular remodeling had increased mast cell density. Heath concluded that lung mast cell hyperplasia in the highlanders was associated with muscular remodeling of the pulmonary arteries. Tucker et al. (43) and Williams et al. (49) have also pointed out that the magnitude of mast cell hyperplasia in chronically hypoxic animals is well correlated with the degree of medial thickness of small pulmonary arteries, suggesting that the thickest muscular arteries have the most mast cells in the their adventitial coats. Evidence in opposition to these observations was reported by Williams et al. (48), who found that a high density of mast cells was present around the thin-walled pulmonary arteries of the llama at high altitude. This is also inconsistent with the present findings of thin-walled pulmonary arteries and no mast cells in the pikas. Both llamas and pikas are species indigenous to high altitudes; however, we do not know the reason(s) for the distinct difference between llamas and pikas in the effect of mast cells on pulmonary hypertension. Comparative studies are necessary to determine the differences regarding mast cells between these two species.

The histological findings of the present study suggest that the thickness of pulmonary arteries could be associated with some remodeling growth factors. To test this hypothesis, we investigated mast cell tryptase and TGF-₁ with a monoclonal antibody in the lung tissue of rats and pikas.

Tryptase is a multifunctional growth factor (24, 37) and has been used as a discriminating marker for mast cells (46). Our review of the literature suggested that it is likely that tryptase plays an important role in airway inflammation. The incubation of bronchial smooth muscle with tryptase caused smooth muscle hyperresponsiveness in dogs (25) and stimulated the proliferation of lung fibroblasts (37) and epithelial cells (12). More recent data obtained by Blair et al. (7) showed that mast cells act at sites of new vessel formation by secreting tryptase, which then functions as a potent and previously unrecognized angiogenic factor. We found that there was intense immunoperoxidase staining for tryptase in the rats, and tryptase-positive mast cells were seen around thick-walled rat pulmonary vessels, whereas in the pikas no such immunostaining was found, suggesting that mast cell tryptase may be related to pulmonary vessel thickness. It is not known, however, what function the mediators of mast cells are fulfilling. One of the most striking effects of chronic alveolar hypoxia is the muscularization of pulmonary arterioles, and this remodeling process involves the proliferation or migration of smooth muscle cells and the increased production of extracellular matrix component. We therefore speculate that in rats, mast cells may participate in the process of pulmonary vessel remodeling by secreting specific growth factors, and in the pikas there is no growth response in vessels even at very high altitude, which may be regarded as evidence of successful adaptation to high altitude.

It is interesting that TGF-₁ immunoreactivity was observed in the muscularized vessels in the rats but not in those of the pikas. TGF-₁ is a mediator of vascular remodeling (8). Botney et al. (9) found that anti-TGF- antibody staining was stronger in patients with primary pulmonary hypertension than in normal subjects. Increased TGF- is also observed in models of vascular remodeling including the development of pulmonary hypertension after air embolism (35), repair of arterial injury after carotid endarterectomy (33), and development of systemic hypertension after salt and mineralocorticoid administration (39). It was also reported by several investigators (17, 24) that hypoxia can stimulate an increased expression of TGF-₁ mRNA.

Previous studies indicated that TGF- can be released by many cells (4, 10, 34, 45), but the precise source by which cells release it is unclear. Taipale et al. (42) found that mast cell chymase and leukocyte elastase effectively released TGF-₁, which could specifically bind to and be activated at the surface of smooth muscle cells, and induced smooth muscle cell proliferation. The present results indicate that the neutrophils and mononuclear cells were strongly positive for TGF-₁, which seems to play an important role for modulating vascular remodeling. The mechanism by which tryptase and TGF- elicit a growth response to cells could be caused by the effect of proteolytic enzymes, because the granule proteinase of mast cells participates in the degradation of connective tissue and smooth muscle (38). The concept that enzymes have direct angiogenic properties was previously confirmed (22).

In agreement with previous observations at high altitude (1, 6), we observed that the levels of Hb and Hct of pikas were lower than those of rats. Lower Hb and Hct values at high altitude, which can be a feature of residence at high altitude and may be regarded as evidence of genetic adaptation to high altitude (6), may function to maintain blood viscosity and pulmonary vascular resistance at low levels. It is therefore possible that the maintenance of low levels of P_pa and the lack of RV hypertrophy in pikas are caused in part by low Hb and Hct caused by the higher Pa_O2. Our results indicate that Pa_O2 was decreased more at the simulated altitude of 5,000 m in the rats than the pikas, although there was no significant difference in Pa_O2 between the groups at the baseline altitude (2,260 m). The higher Pa_O2 in the pikas may indicate that the pulmonary gas exchange is more efficient and maintains adequate tissue oxygenation despite reduced oxygen tension, which may be caused in part by the higher hypoxic ventilatory response (HVR). However, the changes of ventilation in indigenous mountain animals are not fully understood. Previously it was indicated (28) that llamas living in the Andes exhibit lower HVR, and no blunted HVR was found in yaks in the Himalayas. Observations of HVR in the pika are sparse.

Another very characteristic feature of adaptation to high altitude in indigenous mountain animals is that the oxygen-Hb dissociation curve is shifted to the left (28), which is partially caused by a low concentration of 2,3-DPG. Bullard (11) pointed out that successful adaptation to high altitude in these animals does not involve an increase in the concentration of 2,3-DPG. The levels of 2,3-DPG in high-altitude animals are extremely low or absent in contrast to those of unadapted animals, as previously reported by Adams et al. (1) and Chiodi (13). Our results are similar to these observations in that the 2,3-DPG level in the pikas was significantly lower than that of the rats, indicating increased affinity of Hb for oxygen in the pikas, which may reflect successful adaptation to high altitude.

In summary, we observed that the acute hypoxic pressor response was smaller in pikas than in Wistar rats. The pikas had a very thin pulmonary vascular medial layer, virtually no smooth muscle in the small pulmonary arterioles, and no RV hypertrophy compared with the rats. A number of mast cells were observed around pulmonary vessels in the rats, whereas there were no mast cells in the pikas. Both TGF-₁ and mast cell tryptase demonstrated strongly positive immunoperoxidase staining in the pulmonary arteries of the rats but not in the pikas. Hct, Hb, and 2,3-DPG of the pikas were significantly lower than those of the rats. These results suggest that the pika has adapted genetically to life at high altitude by losing hypoxic pulmonary vasoconstriction. In rats, mast cells may play an important role in the remodeling of pulmonary vessels by activating some growth factors.