INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Chronic hypoxia causes pulmonary hypertension and pulmonary vascular remodeling in most animal species (1), but the magnitude of pulmonary hypertension and the degree of vascular remodeling are species (1) and strain (8) dependent. The development of pulmonary hypertension and the muscularization of the precapillary pulmonary arteries can be inhibited by treatment of animals with a calcium entry blocker (9), by treatment with a platelet-activating factor (PAF) antagonist that does not affect hypoxic pulmonary vasoconstriction (10), or in transgenic mice that overexpress the prostacyclin (PGI₂) synthase gene in a lung-specific manner (11). However, the molecular and cellular mechanisms of vascular remodeling in hypoxic pulmonary hypertension are poorly understood.

PGI₂ is a potent pulmonary vasodilator and chronic intravenous infusion of PGI₂ is being used successfully in the treatment of patients with severe pulmonary hypertension (12, 13). Chronic PGI₂ treatment can reduce the pulmonary vascular resistance in patients with primary pulmonary hypertension (14) and PGI₂ attenuates the proliferation of cultured vascular smooth muscle cells (15) through signal transduction following ligand binding to its receptor. We found that patients with severe pulmonary hypertension have reduced PGI₂ receptor (PGI-R) expression in the remodeled pulmonary arterial smooth muscle (16). The question of whether endogenously produced PGI₂ modulates pulmonary vascular remodeling has been of interest for some time (17, 18) but could not be rigorously addressed because of the lack of specific PGI-R antagonists. Instead of applying pharmacological tools we employed the recently developed PGI-R knockout (KO) mouse model (19) to address the question of PGI-R-dependent lung vascular remodeling. Here we report on comparative studies with wild-type (WT) and PGI-R KO mice and show that after exposure of the animals to chronic hypobaric hypoxia, PGI-R KO mice develop a greater degree of pulmonary hypertension and pulmonary arterial media thickening than WT mice. Our data underscore the importance of PGI₂ in the modulation of the pulmonary vascular response to chronic hypoxia.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Animals

Prostacyclin receptor (PGI-R) knockout (KO) mice (19) were first backbred into the C57BL/6 background for six generations in Kyoto, Japan. Upon arrival in Denver, the animals were back-crossed four more generations to maintain a congenic strain. For all experiments, homozygous null mice were used.

Hypobaric Hypoxic Exposure

For exposure to hypobaric hypoxia, 8-wk-old male PGI-R KO mice (19) (n = 10) and WT C57BL/6 (n = 18) mice were housed in a hypobaric chamber for 3 wk at a simulated altitude of 17,000 ft (Pb = 410 mm Hg; inspired PO₂ = 76 mm Hg). Ten KO and 10 WT animals were kept at Denver altitude (5,280 ft; Pb = 630 mm Hg, inspired PO₂ = 122 mm Hg) for more than 5 wk.

Acute Hypoxic Exposure

To compare the effect of acute hypoxia on right ventricular pressure of the KO mice (n = 8) with that of WT animals (n = 8), male mice 8 wk old were placed in an acrylic chamber (fraction of inspired oxygen [FI_O2] = 0.1) at Denver altitude for 15 min (11). During pressure measurement, the animals were maintained in the hypoxic environment with flow-by oxygen at FI_O2 = 0.1.

Measurement of Right Ventricular Pressure

Hemodynamic assessment was performed as previously described by direct measurement of right ventricular systolic pressure (RVSP) (11, 20). After RVSP measurement, lung tissue was inflated with low-melt agarose and fixed with 10% buffered formalin.

Hematocrit Analysis

Spun hematocrit was measured collecting 0.2 ml of blood by way of ventricular puncture.

Assessment of Right Ventricular Hypertrophy

The hearts were fixed with 10% buffered formaldehyde for 24 h. The right ventricular free wall (RV) and the left ventricle together with the septum (LV + S) were weighed separately, and RV/(LV + S) was calculated.

Histology and Morphometric Analysis

A representative hematoxylin and eosin-stained lung section through the hilum was coded and blindly evaluated for remodeling of pulmonary arteries in the range of 30-50 µm as previously described (11). The wall thickness of each artery was expressed as percentage of the vessel diameter by the formula (Ed  Id)/Ed (VWT) (Ed, external diameter; Id, internal diameter; VWT, vessel wall thickness). For the conditions of control (Denver altitude) and chronic hypoxia, each experimental animal was examined (n = 10 for KO kept at Denver altitude; n = 10, WT at Denver altitude; n = 9, KO exposed to chronic hypobaric hypoxia; n = 13, WT, chronic hypobaric hypoxia). Between 10 and 20 measurements were made per animal and the average was calculated.

Immunohistochemistry

Histological sections were heated in 1× citrate buffer for 15 min, followed by H₂O₂ block of endogenous peroxidase for 15 min. The sections were then incubated with anti-smooth muscle -actin antibody at 37° C for 30 min, followed by Quickkit (Vector, Burlingame, CA) for 10 min at room temperature.

Statistical Analysis

All values were expressed as means ± SEM. Means of several groups were compared by unpaired t tests using Microsoft Excel 2000. For correlation data, a Pearson r correlation coefficient was calculated using GraphPad Prism 3.01 for Windows 2000 (GraphPad Software Inc., San Diego, CA). Two-tailed parametric analysis was considered significant when p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PGI-R Knockout Does Not Cause Pulmonary Hypertension at Denver Altitude

At baseline (Denver altitude; Pb = 630 mm Hg, inspired PO₂ = 122 mm Hg) PGI-R KO (n = 10) and WT (n = 10) mice displayed no difference in RVSP (31.6 ± 1.3 mm Hg versus 28.0 ± 1.0 mm Hg; not significance [ns]) (Figure 1).

View larger version (13K): [in this window] [in a new window]   Figure 1.   Right ventricular systolic pressure (RVSP) at baseline, during acute hypoxia, and after chronic exposure to hypobaric hypoxia. At baseline (control, Denver altitude), knockout (n = 10) and wild-type (n = 10) mice show similar RVSP. During acute hypoxia, knockout (n = 8) and wild-type (n = 8) mice showed comparable pressures. After 3 wk of exposure to hypobaric hypoxia, the knockout mice (n = 9) had significantly higher RVSP than did the wild-type animals (n = 13) (*p = 0.002).

PGI-R Knockout and WT Mice Exhibit a Similar Hypertensive Response to Acute Hypoxia

During acute hypoxia, KO (n = 8) and WT (n = 8) mice show a comparable RVSP (32.7 ± 0.9 mm Hg versus 29.8 ± 1.0 mm Hg; ns) (Figure 1).

PGI-R KO Mice Develop More Severe Pulmonary Hypertension after Chronic Hypobaric Hypoxia Than WT Animals

After 3 wk of exposure to hypobaric hypoxia, KO animals (n = 9) developed significantly more severe pulmonary hypertension, with the RVSP averaging 44.9 ± 2.4 mm Hg when compared with the WT animals (n = 13), which had an average RVSP of 35.6 ± 1.5 mm Hg (Figure 1).

Both PGI-R KO and WT Animals Demonstrate a Comparable Polycythemic Response with Exposure to Chronic Hypobaric Hypoxia

At Denver altitude, the PGI-R KO animals (n = 10) had a slightly lower hematocrit than did WT animals (n = 10) (48.8 ± 0.4% versus 51.5 ± 1.1%; p = 0.02). In response to chronic hypobaric hypoxia, both PGI-R KO (n = 9) and WT animals (n = 13) developed polycythemia when compared with Denver altitude control animals (57.7 ± 1.4% versus 48.8 ± 0.4% for PGI-R KO, p = 0.0001, and 55.0 ± 1.0% versus 51.5 ± 1.1% for WT, p = 0.02). The extent of polycythemia was the same for KO and WT animals after chronically hypobaric hypoxic exposure (57.7 ± 1.4% versus 55.0 ± 1.0%, ns).

PGI-R KO Mice Have More Severe Right Ventricular Hypertrophy after Chronic Hypobaric Hypoxia Than Do WT Animals

At Denver altitude, KO (n = 10) and WT (n = 10) mice had similar RV/(LV + S) values (0.26 ± 0.01 versus 0.26 ± 0.003; ns). After 3 wk of hypobaric hypoxic exposure, KO animals (n = 9) had significantly higher RV/(LV + S) than did WT animals (n = 13) (0.40 ± 0.01 versus 0.31 ± 0.01; p = 0.0001) (Figure 2).

View larger version (12K): [in this window] [in a new window]   Figure 2.   Ratio of right ventricular weight and combined weight of left ventricle + septum [RV/(LV + S)] at baseline and after exposure to hypobaric hypoxia. Wild-type (WT; n = 10) and knockout (KO; n = 10) mice did not differ in RV/(LV + S) at baseline (control, Denver altitude) (0.26 versus 0.26; ns.). After 3 wk of exposure to hypobaric hypoxia, both WT (n = 13) and KO (n = 9) animals developed significantly elevated RV/(LV + S) when compared with Denver altitude control animals (*p = 0.0001). RV/(LV + S) of KO mice was significantly higher than that of WT animals at high altitude (**p = 0.0001).

PGI-R KO Mice Have Pulmonary Vascular Remodeling after Hypobaric Hypoxia

At Denver altitude, KO and WT animals showed no morphological differences in pulmonary artery medial wall thickness in the vessels that expressed smooth muscle cell -actin (Figure 3A, WT at Denver altitude; B, KO at Denver altitude). After chronic exposure to hypobaric hypoxia, WT animals developed no medial wall hypertrophy (Figure 3C; WT, hypoxia), whereas KO animals had significant medial wall thickening (Figure 3D; KO, hypoxia). The statistical analysis of the lung vascular morphometry also revealed that at baseline in Denver, KO (n = 10) and WT (n = 10) animals did not differ in vessel wall thickness of the pulmonary arteries in the size range of 30-50 µm in external diameter (0.172 ± 0.010 versus 0.164 ± 0.004; ns), and that chronic hypobaric hypoxia caused no vessel wall hypertrophy in WT animals (n = 13, 0.165 ± 0.005) whereas significant vessel wall thickening occurred in KO animals (n = 9, 0.228 ± 0.010; p = 0.0002) (Figure 4).

View larger version (108K): [in this window] [in a new window]   Figure 3.   Smooth muscle cell -actin immunostaining of the pulmonary arteries from a WT mouse kept at Denver altitude (A), a PGI-R KO mouse at Denver altitude (B), a WT mouse exposed to hypobaric hypoxia for 3 wk (C ), and a chronically hypoxic KO mouse (D). The medial smooth muscle cells are strongly positive for -actin. Note the marked increase in the media wall thickness of the chronically hypobaric hypoxic KO mouse, whereas neither the Denver altitude KO mouse nor the hypoxic WT mouse developed the pulmonary media wall thickening.

View larger version (12K): [in this window] [in a new window]   Figure 4.   Comparison of the vessel wall thickness of small- to medium-sized (30-50 µm) pulmonary arteries. Vessel wall thickness was calculated by external diameter (Ed) minus internal diameter (Id) divided by the external diameter: (Ed  Id)/Ed. At baseline (control, Denver altitude), knockout (KO) (n = 10), and wild-type (WT) (n = 10) animals had similar vessel wall thickness (0.172 versus 0.164; ns). After chronic hypobaric hypoxia, KO animals (n = 9) had significantly more vessel wall thickening than did WT mice (n = 13) (0.228 versus 0.165; *p = 0.0002).

RVSP and RV/(LV + S) of All Animals Demonstrate a Significant Correlation

Plotting RV/(LV + S) against RVSP demonstrates the correlation of right ventricular hypertrophy to elevated RVSP. Both types of animals (WT and KO) have a significant correlation between RVSP and RV/(LV + S) (data not shown). When plotted as a group, Pearson r and two-tailed p value of the correlation were 0.75 and less than 0.0001 (Figure 5).

View larger version (17K): [in this window] [in a new window]   Figure 5.   Relationship between right ventricular systolic pressure (RVSP) and ratio of right ventricular weight and combined weight of left ventricle + septum [RV/(LV + S)] of PGI-R KO and WT mice both at baseline and after chronic hypobaric hypoxia. Note the significant correlation between RVSP and RV/(LV + S) at Pearson r of 0.75 and two-tailed p value of less than 0.0001.

Vessel Wall Thickening Correlates with RVSP Only in the PGI-R KO Mice, But Not in the WT Mice

When RVSP was plotted against VWT, there was a significant correlation only in the PGI-R KO mice (Pearson r, 0.60; two-tailed p value, 0.0067; Figure 6b). There was no significant correlation between vessel wall thickening and RVSP in the WT animals (Pearson r, 0.03; two-tailed p value, 0.90, ns; Figure 6a). The statistical analysis also revealed that the correlation coefficient of the KO mice between vessel wall thickening and RVSP was significantly different from that of the WT mice (p < 0.05).

View larger version (28K): [in this window] [in a new window]   Figure 6.   Relationship between vessel wall thickness of pulmonary arteries (VWT) and right ventricular systolic pressure (RVSP) in WT mice (a) and PGI-R KO mice (b) kept at Denver altitude and exposed to hypobaric hypoxia. (a) In the WT animals, RVSP is not correlated with VWT. Pearson r of the correlation is 0.03. (b) The PGI-R KO animals demonstrate a correlation between increased RVSP and pulmonary vessel wall thickening (Pearson r, 0.60; two-tailed p value, 0.0067). The covariance analysis also revealed that the correlation coefficient of the KO mice between VWT and RVSP was significantly different from that of the WT mice (p < 0.05).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The principal findings of this study are that PGI-R KO mice developed a greater degree of pulmonary hypertension, more right ventricular hypertrophy, and pulmonary vascular remodeling after chronic exposure to hypobaric hypoxia than WT mice. Our data indicate that PGI₂ and its receptor play an important role in the regulation of hypoxia-induced pulmonary vascular remodeling and that the absence of a functional PGI₂ receptor worsens hypoxic pulmonary hypertension.

PGI₂ reduces pulmonary vascular resistance (14) and attenuates proliferation of cultured vascular smooth muscle cells (15) through signal transduction following ligand binding to its receptor. We previously found that patients with severe pulmonary hypertension had a reduced PGI-R expression in the remodeled pulmonary arterial smooth muscle (16), and we hypothesized that PGI-R KO animals would have a greater degree of pulmonary hypertension and more pronounced lung vascular remodeling when compared with WT animals. The results of this study are consistent with this hypothesis. At Denver altitude, the PGI-R KO did not develop pulmonary hypertension, right ventricular hypertrophy, or lung vascular wall thickening when compared with their WT counterparts. However, after 3 wk of hypobaric hypoxia, PGI-R KO mice developed more severe pulmonary hypertension and right ventricular hypertrophy when compared with the WT animals (Figures 1 and 2). Chronic hypobaric hypoxia caused pulmonary vascular remodeling in the PGI-R KO mice but not in the WT mice (Figures 3 and 4). The finding that the C57BL/6 WT strain that was used in this study does not show pronounced pulmonary vascular remodeling in response to chronic hypoxia (20) has been confirmed by our present study and thus the vascular remodeling triggered by chronic hypobaric hypoxia in the C57BL/6 PGI-R KO animals is remarkable. Wild-type C57BL/6 mice show no correlation between RVSP and vessel wall thickness (a measurement of remodeling), whereas PGI-R KO mice show a significant correlation between these two parameters (Figure 6a and 6b).

A generally accepted paradigm in rat models of hypoxic pulmonary hypertension is the correlation between mean pulmonary artery pressure (measured by catheterization) and right ventricular hypertrophy, as assessed by RV/(LV + S) (5). This has not, to our knowledge, been previously demonstrated in murine models. In the current work, we demonstrate that for both WT and PGIR-KO mice, a good correlation exists between RVSP (an indirect measure of pulmonary artery pressure) and the development of right ventricular hypertrophy (Figure 5).

Narumiya and associates have previously characterized the PGI-R KO mice and shown their increased susceptibility to thrombosis and reduced inflammatory and pain responses (19, 21). We now add new data to characterize this PGI-R KO phenotype further.

The lack of PGI-R signaling as demonstrated in these animals previously (19, 21) could augment chronic hypoxic vasoconstriction, impair mechanisms that protect against pulmonary vascular remodeling, or facilitate vascular cell growth. The acute hypoxia-induced pulmonary presser response was not enhanced in the PGI-R KO animals, and the hematocrit in both the PGI-R and the WT animals did not differ following exposure to hypobaric hypoxia. Therefore, a change in hypoxic vasoconstriction or a change in blood viscosity is unlikely to account for the greater degree of pulmonary hypertension and vascular remodeling in the PGI-R KO mice. The work presented here is important because it shows for the first time that there is clearly a membrane prostacyclin receptor-dependent element to modulation of the lung vascular remodeling. A rigorous assessment of the importance of PGI-R in modulating lung vascular structure has previously not been possible due to the lack of a specific prostacyclin receptor antagonist. In addition to membrane prostacyclin receptor signaling, there may be additional prostacyclin ligand binding to nuclear receptors like peroxisome proliferator-activated receptor  (PPAR ) (22, 23). Should lung vascular cells utilize prostacyclin nuclear receptors (such as PPAR ), then we would expect that signaling through this mechanism is not altered in our current model of membrane PGI₂ receptor knockout mice. If nuclear PGI₂ receptor-dependent signaling also results in inhibition of lung vascular remodeling, then our data in the PGI-R KO mice could underestimate the total impact of PGI₂ on hypoxia-induced pulmonary vascular remodeling.