INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pathological studies of human hypertensive pulmonary arteries suggest that vascular remodeling is a common abnormality regardless of etiology. Generally, vascular remodeling in the large elastic pulmonary artery is thought to be an adaptive response to increased pressure in human severe pulmonary hypertension (1). The specific mechanisms, however, underlying both large and small vessel remodeling in human primary pulmonary hypertension (PPH) are still unknown. In chronic pulmonary thromboembolism, on the other hand, it is known that a varying degree of precapillary pulmonary hypertension is involved. In case of chronic thromboembolic pulmonary hypertension (CTEPH), the prognosis is quite poor if pulmonary hypertension is severe enough to resist medical treatment (4, 5). Recently, pulmonary thromboendarterectomy was reported as a beneficial surgical treatment that clearly improves mortality in CTEPH (6). However, the pathogenesis and molecular mechanisms of vascular remodeling causing CTEPH have so far not been well clarified.

It has been reported that chemokines or proinflammatory cytokines play a pivotal role in the pathogenesis of PPH (10). Monocyte chemoattractant and activating factor /monocyte chemoattractant protein-1 (MCAF/MCP-1), for examples, is a C-C chemokine that possesses potent chemotactic and activating effects for monocytes/macrophages in vitro (11, 12). MCP-1 acts as a pathogenic mediator in the development of congestive heart failure (CHF) in humans as well as in experimental animal models (13, 14). Aukrust and coworkers reported that the plasma level of MCP-1 was elevated in patients suffering from CHF, and that MCP-1 levels were inversely correlated with left ventricular ejection fractions (13). In addition, they suggested that monocytes/macrophages migrating into the heart are crucial sources for producing MCP-1 protein (13). Shioi and coworkers also reported that left ventricular hypertrophy and heart failure caused by pressure overload induce macrophage infiltration and MCP-1 mRNA gene expression as well as protein synthesis at the interstitium around hypertrophied cardiac myocytes in rats (14).

It has been elucidated that interleukin-1 (IL-1) and macrophage inflammatory protein-1 (MIP-1) play an important role in vascular remodeling of PPH (15, 16). Elevated circulating levels of interleukin-1 (IL-1) and IL-6 were also observed in patients with pulmonary hypertension associated with mixed connective tissue disease (MCTD), suggesting that they may act as chemical mediators in vascular remodeling caused by autoimmune-induced inflammatory reaction (17).

Furthermore, we also demonstrated that MCP-1 protein synthesis takes place in monocytes/macrophages around pulmonary arterioles in the lung parenchyma in the monocrotaline-induced pulmonary hypertension model (18). We proposed that MCP-1 acts as a mediator via autocrine and/or paracrine mechanisms to induce migration and activation of monocytes/macrophages in the lung (18). It is conceivable that monocytes/macrophages activated by excessive secretion of MCP-1 participate in pulmonary vascular remodeling, in which various kinds of growth factors such as transforming growth factor- (TGF-) (19), MIP-1 (16), and vascular endothelial growth factor (VEGF) (10) are also involved. Moreover, it is known that MCP-1 and IL-1 possess actions to induce muscular hypertrophy and angiogenesis by themselves (20, 21). It is widely recognized that patients with a diagnosis of antiphospholipid syndrome with positive lupus anticoagulants occasionally suffer from CTEPH (22, 23). These findings indicate that immunological and inflammatory mechanisms play roles in the pathogenesis of CTEPH. We hypothesized that the severity of pulmonary hypertension may be related to MCP-1, which is thought to be upregulated by blood pressure or shear stress in pulmonary vasculature as well as by immunological and inflammatory reactions in CTEPH. In our hypothesis, monocytes/macrophages activated by excessive secretion of MCP-1 may induce vascular cell proliferation and matrix protein synthesis via autocrine or paracrine mechanisms. To test this hypothesis, we performed a hemodynamic and pathological study in combination with the measurement of various kinds of immunological parameters or cytokines/chemokines including MCP-1, which may play a role in vascular remodeling in CTEPH.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

CTEPH was diagnosed in 14 patients (2 males and 12 females) aged from 28 to 66 yr (50.5 ± 10.6 yr; mean ± SD) by right cardiac catheterization during a clinically stable period. CTEPH was defined as mean pulmonary artery pressure () of more than 25 mm Hg and pulmonary capillary wedge pressure (Ppc, we) of less than 12 mm Hg, indicating precapillary pulmonary hypertension. Clinical features of the patients are summarized in Table 1. All patients had a history of dyspnea on exertion for more than 6 mo prior to admission.

Pulmonary perfusion scan with ^99mTc-macroaggregated albumin (MAA) revealed segmental or larger defects. Ventilation scan with krypton-81m (^81mKr), on the other hand, revealed no abnormality in any of the patients. Pulmonary angiography following right cardiac catheterization also supported the clinical diagnosis. No patient in this study was medicated with thrombolytic therapy (urokinase or tissue plasminogen activator), antiinflammatory drugs (corticosteroid, etc.), or vasodilating drugs including orally administered prostacyclin analogue (Veraprost). Anti-coagulant warfarin potassium (Warfarin) was discontinued for 72 h before right cardiac catheterization. One female patient (Case 3) underwent right cardiac catheterization twice before and after pulmonary thromboendarterectomy.

Patients were prospectively divided into two groups as follows: central predominant type (Cpred.) and peripheral predominant type (Ppred.), according to location of thromboemboli in pulmonary angiography (24) and imaging by spiral computed tomography (CT) scan with a bolus injection of contrast medium. Cpred. and Ppred. were defined as thromboemboli located in the lobar arteries or proximal in Cpred. and in segmental arteries or distal in Ppred. Pulmonary thromboendarterectomy was performed in six patients, four of Cpred. and two of Ppred. Representative cases of Cpred. and Ppred. are presented in Figures 1 and 2, respectively.

View larger version (52K): [in this window] [in a new window]   Figure 1.   A representative case defined as a central predominant (Cpred.) type is shown. Pulmonary perfusion scan with 99mTc-MAA reveals large defects in both lungs (A). Spiral CT scan demonstrates thromboemboli in the right main (arrow) and left lower lobar (double arrow) pulmonary arteries (B). Pulmonary angiography shows thromboemboli at bilateral lobar pulmonary artery levels (arrows) (C ).

View larger version (60K): [in this window] [in a new window]   Figure 2.   A representative case defined as peripheral predominant (Ppred.) type is shown. Pulmonary perfusion scan reveals multiple segmental defects (A). Spiral CT scan shows no thromboembolus between the main and lobar pulmonary arteries (B). Pulmonary angiography, on the other hand, reveals thromboemboli with pruning and intimal irregularities in either segmental or more peripheral pulmonary arteries (arrows) (C ).

Informed consent was obtained from all patients, and the protocols were approved by the institutional review board.

Hemodynamics

To evaluate the pulmonary hemodynamics, and cardiac output (CO) were measured using the thermodilution method from patients breathing room air. Cardiac index (CI) was calculated as CO divided by body surface area. Pulmonary vascular resistance (PVR) was obtained by the following equation:

Pulmonary function tests and arterial blood gas analyses were also performed in all patients (Table 1).

Cytokine/Chemokine Measurement

Levels of MCP-1 in plasma, and IL-1 and tumor necrosis factor- (TNF-) in serum were measured by sandwich enzyme-linked immunosorbent assay (ELISA) using the samples from peripheral venous blood. The detecting limits of MCP-1, IL-1 and TNF- were 25.0, 2.0, and 15.6 pg/ml, respectively. First, to analyze correlation between the cytokine/chemokine level and pulmonary hemodynamics, a regression line was drawn using the least-squares method, and the coefficient of correlation was calculated in each cytokine. Second, the cytokine/chemokine level was compared with serological immunoglobulins (IgG, IgM, and IgA) and complements (C3, C4, and CH50). Each sample was extracted from peripheral blood on the same day.

Histology and Immunohistochemistry

Pathological specimens of pulmonary arterial thrombi and lung tissue were obtained from two patients with Cpred. undergoing pulmonary thromboendarterectomy at Chiba University Hospital. Immediately following surgical resection, thrombus and lung tissue were fixed in 4% neutro-buffered formaldehyde for 24 h at room temperature and subsequently dehydrated in sequential 30%, 50%, and 70% ethanol washes. Tissues were embedded in paraffin and transverse slices of specimens were cut into 5-µm-thick sections. Hematoxylin and eosin (HE) and Elastica-van Gieson (EVG) stain were used for pathological observation. The same specimens were prepared for immunoperoxidase staining using a Histamine SAB-PO (R, M) kit (Nichirei Co. Ltd., Tokyo, Japan) as described previously (25). Endogenous peroxidase was blocked with 0.3% (vol/vol) H₂O₂ in methanol for 20 min at room temperature. Nonspecific immunoglobulin binding sites were blocked with 10% goat serum. Sections were subsequently incubated for 2 h at 4° C with rabbit polyclonal anti-human MCP-1 antibody (26). The specimens were also incubated for 2 h at 4° C with mouse monoclonal anti-human alveolar macrophage (HAM56) or anti-human -smooth muscle actin (-SMA) antibodies (DAKO Co. Ltd., Carpinteris, Denmark). Appropriate preimmune 10% serum served as negative controls. Sections were then incubated for 20 min with affinity-purified biotin-conjugated goat anti-rabbit immunoglobulin G (IgG) or rabbit anti-mouse IgG, IgM complex (10 µg/ml), washed, and incubated for 20 min with horseradish peroxidase-streptavidin (100 g/ml). Immunoglobulin complexes were visualized by incubation with 3,3'-diaminobenzidine (0.03 mmol/L in Tris-HCl, pH 7.4), 0.6% H₂O₂ and NiCl₂. Sections were washed, counterstained with Meyer's hematoxylin, dehydrated, mounted in Permount, and examined by light microscopy.

Statistical Analysis

All data are presented as mean ± SD. Spearman's correlation was used to analyze the relationship between cytokines/chemokines and hemodynamic parameters as well as cytokines/chemokines and immunologic parameters. A p value less than 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Hemodynamics

was elevated (42.5 ± 13.6 mm Hg) and CI was decreased (2.86 ± 0.64 L/min/m²) in patients with CTEPH. PVR was elevated, as a result, ranging from 302 to 1467 (mean PVR; 733.4 ± 330.3) dyne · s/cm⁵, suggesting increased pulmonary hemodynamics in CTEPH. PVR in Cpred. (792 ± 374 dyne · s/cm⁵; n = 8) tended to be elevated rather than that in PPred. (655 ± 273 dyne · s/cm⁵; n = 6) (Table 1).

Pulmonary Function and Arterial Blood Gases

Spirometry demonstrated normal pulmonary function in patients with CTEPH except for one case of restrictive ventilatory disturbance (%VC; 75.0%) and two cases (FEV₁%; 64.3% and 56.1%) of obstructive ventilatory disturbance. Hypoxemia combined with hypocapnia was observed in almost all subjects; mean Pa_O2 and Pa_CO2 were 57.1 ± 11.8 and 35.3 ± 4.2 mm Hg, respectively. The acid-base balance was slightly but significantly shifted to alkalosis (pH; 7.436 ± 0.026) in patients with CTEPH (Table 1).

Cytokine/Chemokine Level

The plasma MCP-1 level ranged widely from 25 to 799 (mean 173.4 ± 202.9 pg/ml). Mean concentrations of IL- and TNF- were 164.5 ± 279.3 (2.0-932.0 pg/ml) and 40.1 ± 44.8 (15.6- 183.0 pg/ml), respectively. Regression analysis revealed no significant coefficient of correlation between MCP-1 and (0.268; p > 0.1) and between MCP-1 and CI (0.393; p > 0.1). On the other hand, the coefficient of correlation was significant between MCP-1 and PVR (0.692; p < 0.01) (Figure 3). When it was analyzed in the subpopulation of Cpred. patients, the coefficient of correlation became much higher (r = 0.805), suggesting that the correlation became more prominent between MCP-1 and PVR in response to the type of thromboemboli. Case 3, a case of Cpred. in which pulmonary hemodynamics were evaluated before and after pulmonary thromboendarterectomy, revealed that PVR was dramatically decreased (1023 to 292 dyne · s/cm⁵) in conjunction with a parallel change in the MCP-1 plasma level (118 to 50 pg/ml). No significant relationship was demonstrated in the coefficients of correlation between CI and MCP-1 or and MCP-1. Similarly, no relationships were observed between pulmonary hemodynamics and other cytokines.

View larger version (29K): [in this window] [in a new window]   Figure 3.   Correlation between plasma MCP-1 and PVR in patients with CTEPH. Linear regression analysis shows positive correlation with significance (n = 14, r = 0.692, p < 0.01). Pulmonary hemodynamics in case 3 were evaluated both before (closed circle) and after pulmonary thromboendarterectomy (open circle). Plasma MCP-1 changes in parallel with PVR along the regression line. *Patients whose thromboemboli were detected in a peripheral predominant (Ppred.) pattern; PVR, pulmonary vascular resistance; MCP-1, monocyte chemoattractant protein-1.

The mean values of serum immunoglobulins IgG, IgA, and IgM were 1666.6 ± 443.4 (1064-2660), 379.4 ± 246.5 (194- 1105), and 199.8 ± 132.7 (66-500) mg/dl. C3, C4, and CH50 exhibited 81.6 ± 19.4 (59-128), 29.6 ± 9.4 (15-48) mg/dl, and 41.9 ± 8.3 (32.7-64.7) U/ml, respectively. The coefficients of correlation between circulating cytokines/chemokines and serum immunoglobulins or complements are summarized in Table 2. There was no significant correlation between cytokine level and serum immunoglobulins/complements.

Pathology and Immunohistochemistry

Neointimal formation was observed in an endarterectomized specimen, an inner part of the hypertensive large elastic pulmonary artery (Figure 4). Neointimal formation was also observed in small pulmonary arterioles in the lung parenchyma (Figure 5). Immunoreactivity of MCP-1 was demonstrated not only in endothelial cells (Figure 4C and 4D) but also in mononuclear cells (Figure 4D) and smooth muscle cells (Figure 4E) in a fibrinous portion adjacent to the vascular lumen in endarterectomized tissue. Mononuclear cells producing MCP-1 were confirmed as macrophages by HAM56 anti-human alveolar macrophage antibody (Figure 4F). In a preliminary study before this project, no immunoreactivity of MCP-1 was demonstrated in either endothelial cells or smooth muscle cells in the pulmonary arteries obtained from the patient undergoing pneumonectomy (data not shown).

View larger version (117K): [in this window] [in a new window]   Figure 4.   Histology of specimens obtained from the patient undergoing thromboendarterectomy. (A) Neointimal formation is observed in the endarterectomized specimen, an inner part of the hypertensive large elastic pulmonary artery. Several layers of elastin (e) and collagen (c) bundles associated with angiogenetic small vessels (arrowheads) suggest that active remodeling occurred repeatedly within neointima during the clinical course of CTEPH (original magnification: ×20). (B) High-power view of (A). Mononuclear cell infiltration is observed in fibrinous products adjacent to the vascular lumen (original magnification: × 100). (C ) Immunohistochemistry using anti-human MCP-1 antibody. Immunoreactivity of MCP-1 is partially demonstrated in endothelial cells (original magnification: ×200). ( D) Immunoreactivity of MCP-1 is also demonstrated in mononuclear cells (macrophages) migrating into the neointima as well as in endothelial cells (original magnification: ×400). (E ) MCP-1 is also demonstrated in spindle-shaped cells (myofibroblasts) in a fibrinous portion adjacent to the vascular lumen (original magnification: ×400). The cells were confirmed as smooth muscle cells by anti-human -SMA antibody (data not shown). (F  ) Mononuclear cells shown in (D) were confirmed as macrophages by HAM56 (original magnification: ×400). MCP-1, macrophage chemoattractant protein-1; -SMA, -smooth muscle actin; HAM56, human alveolar macrophage 56.

View larger version (122K): [in this window] [in a new window]   Figure 5.   Histology of the specimen by lung biopsy from the patient undergoing thromboendarterectomy. Hematoxylin and eosin (HE) stain in the lung parenchyma in which thromboemboli were not involved (original magnification: ×400) (A). Elastica-van Gieson (EVG) stain also reveals excessive elastin fragments within neointima even in small pulmonary arteriole (original magnification: ×400) (B).

Both physiological and pathological results suggest that increased pulmonary hemodynamics potentially activate pulmonary arterial endothelial cells or monocytes/macrophages to upregulate MCP-1 protein synthesis and induce neointimal formation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We hypothesized that MCP-1 may participate in the vascular response to injury caused by pressure overload or shear stress in the pulmonary vasculature in CTEPH. In our hypothesis, monocytes/macrophages activated by excessive secretion of MCP-1 may induce endothelial or smooth muscle cell growth and differentiation factors, which act via autocrine or paracrine mechanisms to stimulate vascular cell proliferation and matrix protein synthesis. In our data, the circulating level of MCP-1 was well correlated with PVR in patients with CTEPH. The fact that MCP-1 decreased in parallel with the decrease of PVR after pulmonary thromboendarterectomy in one case also supported our hypothesis. However, the plasma MCP-1 level was not correlated with pulmonary arterial pressure or cardiac index. These results indicate that the elevated pulmonary vascular resistance per se may cause MCP-1 secretion from the vascular wall in the diseased site but myocardial stretch by right ventricle pressure overload may not. MCP-1 secreted from macrophages induced macrophage migration and activation via an autocrine or paracrine mechanism to cause pulmonary vascular remodeling as we previously reported (18).

It is likely that MCP-1 secretion from endothelial cells, smooth muscle cells, and macrophages increases in response to shear stress caused by increased PVR. The immunoreactivity of MCP-1 demonstrated in vascular endothelial cells, smooth muscle cells, and macrophages also supported this paradigm. The coefficient of correlation between MCP-1 and pulmonary vascular resistance became greater when the subjects were limited to the subgroup exhibiting the Cpred. pattern, suggesting that patients with the Ppred. pattern seemed to have difficulty in responding to shear stress, and that intact endothelial cell functions at the level of the peripheral portion of pulmonary arteries/arterioles are considered to be essential to secrete MCP-1. It is also conceivable that MCP-1 synthesis is a kind of compensatory phenomenon that maintains pulmonary circulation normally through negative inotropic action of the myocardium. We are hardly able to discuss the involvement of a systemic inflammatory factor in vascular remodeling of CTEPH because no chemical mediators except MCP-1 seem to be correlated with hemodynamic parameters as far as could be determined in the present study.

Endothelial cells of hypertensive pulmonary arteries/arterioles must be consistently exposed to high blood pressure sustained with strong shear stress in the patients with severe pulmonary hypertension. Shear stress itself plays a key role in inducing various kinds of cytokine gene expressions including MCP-1 (27, 28). The same mechanisms may be involved in the upregulation of MCP-1 protein synthesis in CTEPH. It is reported that macrophages participate in pulmonary vascular remodeling in the hypertensive large elastic pulmonary artery in human PPH (29). This mechanism is also supported by evidence that macrophages produce TGF- and colocalize with extracellular matrix mRNA in the atherosclerotic large elastic pulmonary artery in human PPH (30, 31).

It has been proposed that local regulatory factors in the failing myocardium could clarify the pathogenesis of congestive heart failure (13, 14, 32, 33). This concept suggests another possibilitythat MCP-1 elevation is ascribed to right ventricular hypertrophy, which is derived from the same process as reported previously (14). Shioi and coworkers (14) demonstrated that chemokine gene expression and protein synthesis such as MCP-1 and IL-1 take place in close association with macrophage infiltration in the left ventricular myocardium in the animal model of cardiac hypertrophy chronically induced by mechanical overload. Leukocyte activation and migration from blood to the injured area appear to be of major importance in congestive heart failure (CHF) (32, 33). It is likely that the same mechanism is present in patients with CTEPH. Furthermore, the elevation of circulating levels of C-C chemokines was reported in patients with CHF by Aukrust and coworkers (13). They clearly demonstrated that patients with CHF exhibited higher concentrations of MCP-1 and MIP-1 compared with normal subjects and that the cytokine level was inversely correlated with the left ventricular ejection fraction. The pathological roles of proinflammatory cytokines, such as IL-1 and TNF-, have been reported as negative inotropic action in the perfused heart and cultured myocytes (34, 35). Our data, however, showed no significant relationship between IL-1 or TNF- and other hemodynamic parameters or serological data. Moreover, there was no significant relation between MCP-1 and IL-1 or between MCP-1 and TNF- (data not shown). The discrepancy between the report of Aukrust and coworkers (13) and our data may be explained, in part, by differences in the mechanism of MCP-1 protein secretion, although the true mechanism remains unknown.

In conclusion, this is the first report to demonstrate that MCP-1 is upregulated in response to increased pulmonary vascular resistance in CTEPH. We showed evidence of specific site of MCP-1 production by immunohistochemistry. The increase of MCP-1 appears to reflect the deterioration of pulmonary hemodynamics in CTEPH. In addition, MCP-1 protein synthesis seems to be enhanced by mechanical factors such as shear stress, although the involvement of other inflammatory cytokines is not totally ruled out. To explain the true mechanism of vascular response to injury in CTEPH, further study will be necessary including a study of other secretory components such as circulating monocytes in the blood, vascular smooth muscle cells, and cardiac myocytes.