Abstract
Inflammatory infiltrates and endothelial cell proliferation have been appreciated in plexiform and concentric lesions, which characterize the vascular remodeling in primary pulmonary hypertension (PPH). Leukotriene production by perivascular and alveolar macrophages relies on activation of 5-lipoxygenase (5-LO), with translocation of the enzyme to the nuclear membrane, and association with the 5-LO activating protein (FLAP). Using immunohistochemical staining, we localized and semi-quantitatively estimated the abundance of 5-LO and FLAP in lungs obtained from patients with PPH, patients with interstitial lung disease (ILD), and normal control subjects. Expression of 5-LO and FLAP was prominent in alveolar macrophages in both the normal and PPH lungs; however, alveolar macrophages were more frequently clustered in the vicinity of remodeled blood vessel in PPH. Medium- and small-size pulmonary arteries in PPH showed more abundant FLAP expression than in control and ILD lungs. 5-LO expression in small arteries in PPH was more intense than in control and ILD patients. Endothelial cells in plexiform and concentric lesions in PPH expressed both 5-LO and FLAP. In situ hybridization confirmed the presence of 5-LO transcripts in macrophages and endothelial cells of the remodeled vessels in PPH. We propose that the overexpression of 5-LO and FLAP represents evidence for the participation of inflammation in the process of PPH vasculopathy or, alternatively, that the overabundance of the enzymes involved in generation of inflammatory mediators may themselves be related to vascular cell proliferation and cell growth.
Primary pulmonary hypertension (PPH) is an unexplained, progressive disease of the pulmonary vasculature characterized by increased vascular resistance and pressure overload of the right ventricle. Chronic pulmonary hypertension results in remodeling of lung vessels. Histologically, PPH lungs demonstrate medial hypertrophy of small pulmonary arteries, progressive intimal thickening, aneurysmal dilatation of muscular pulmonary arteries, and proliferation of endothelial cells resulting in plexiform lesions (1). The plexiform lesions, found in 28 to 80% of patients with PPH (2, 3), are often associated with concentric endothelial cell proliferation ultimately causing complete obliteration of the vessel lumen.
The endothelial cell component of both plexiform and concentric vascular lesions (3), as well as the presence of perivascular inflammatory cell infiltrates, have recently been described in PPH and in pulmonary hypertension associated with CREST syndrome and HIV infection (3-5). Alveolar macrophages and vascular endothelial cells produce a variety of vasoactive substances that regulate blood flow and vascular remodeling (6-8); however, the factors that initiate the perivascular accumulation of inflammatory cells and stimulate endothelial cell proliferation are unknown. Intuitively, one might consider that stress-induced activation of endothelial cells results in inflammatory cytokine and growth factor release. Alveolar macrophages in the immediate vicinity of pulmonary vessels may also become activated and play a role in vascular remodeling, contributing to the development of medial hypertrophy and plexiform lesions in PPH.
Agents that cause pulmonary inflammation in animals (e.g., Crotolaria [9], carageenan [10], monocrotaline [11], and endotoxin [12]) and physical forces such as hypoxia-induced shear stress (13) produce some, but not all, of the structural vascular alterations seen in human pulmonary hypertension. Histologically, in addition to medial hypertrophy, animals with chronic pulmonary hypertension show accumulation of macrophages (11) and alterations in endothelial cell function (14).
Eicosanoids are markers and mediators of lung damage and leukotrienes are generated from arachidonic acid during acute lung injury (15-17). 5-Lipoxygenase (5-LO), which catalyzes the first two steps in the synthesis of leukotrienes, is abundantly present in alveolar macrophages (18-20). Activation of 5-LO generates the highly reactive intermediate LTA4, which is rapidly converted to the potent inflammatory compounds LTB4, LTC4, and LTD4 (16). Activation of macrophages also results in translocation of 5-LO from the cytosol to the nuclear membrane, (21, 22). The interaction of 5-LO with the membrane-bound 5-LO activating protein (FLAP) results in leukotriene synthesis (15, 21). We have recently shown that alveolar macrophages from rats exposed to chronic hypoxia demonstrate increased 5-LO expression and that, in chronically hypoxic pulmonary hypertensive lungs, both 5-LO and FLAP are immunohistochemically localized to vascular endothelial cells (14). In addition, the expression of 5-LO appears to be pathogenetically relevant in this model since the FLAP inhibitor, MK-886, inhibits the development of pulmonary hypertension, and 5-LO–deficient (knockout) mice develop less pulmonary hypertension than 5-LO competent mice (14).
It is our hypothesis that inflammation is an important aspect in the pathogenesis of PPH and, specifically, that key inflammatory enzymes participating in leukotriene synthesis are more abundantly present in the pulmonary vessels and alveolar macrophages in PPH patients than in normal lung. To examine this hypothesis, we performed immunohistochemical staining and in situ hybridization of lung tissue from patients with plexogenic pulmonary hypertension and, for comparison, from patients with interstitial lung disease (ILD) and normal control subjects and assessed the presence and localization of 5-LO and FLAP.
Archival human lung samples of patients with PPH (n = 8), with ILD (n = 5), and without known lung disease (n = 8) were selected by review of surgical open lung biopsies or autopsy specimens. Review of the available clinical data showed that patients with PPH had pulmonary hemodynamic parameters that placed them in a severe pulmonary hypertension category (mean pulmonary artery pressures: > 44 mm Hg; range: 44 to 112). Smoking status was available on two patients; both of whom were smokers. Hematoxylin and eosin (H&E)– stained slides were reviewed for confirmation of the pathologic diagnosis and for selection of representative blocks for sectioning. A single block from each patient with PPH was selected; the selection was based solely on the presence of plexiform lesions. Serial sections from each block were stained with the respective immune markers. Since normal tissues were selected from autopsy material, these sections occasionally showed mild pathologic abnormalities ranging from focal pulmonary edema to mild emphysema. The selected formalin-fixed, paraffin-embedded tissues were then cut into 5-μm sections. Sections were placed onto Superfrost Plus™ microscope slides. Some samples were dehydrated under vacuum for 30 min to 1 h before staining to improve the adherence of samples to the slides.
The immunohistochemical staining was performed using the immunoperoxidase system (modified avidin-biotin-peroxidase) (14) with polyclonal rabbit anti-human 5-LO antiserum (LO-32), 5-LO activating protein (FLAP) (kindly provided by Dr. J. F. Evans, Merck-Frost and Pointe Claire-Dorval, Quebec, Canada [23]), and Factor VIII related antigen (Factor VIII r.Ag) (Dako Corp., Carpinteria, CA). After deparaffinization and dehydration, slides were placed in an 800-w microwave oven (full power) for 10 min, with 10 mM citric acid monohydrate buffer (pH 6.0), to enhance antigen retrieval (24). Deionized water (50 ml) was added at 5 min to replace evaporative losses. The samples were cooled for 20 min, digested with pronase protease (0.05% wt/vol) (Calbiochem-Novabiochem Co., La Jolla, CA) for 3 min, and the washed in phosphate-buffered saline (Dulbecco's formula [PBS]). Endogenous peroxidase activity was quenched with 3% (vol/vol) H2O2 for 30 min followed by tap water and PBS washes. Nonspecific antibody binding was blocked using 2% normal goat serum incubated for 20 min. Optimal dilutions of primary antibodies and incubation times were determined using multiple dilutions, incubation times, and temperatures and selected based on the higher specific signal over background staining. Primary antibody dilutions (all vol/vol) in 1% bovine serum albumin in deionized water were as follows: 5-LO (1: 100), FLAP (1:600), and Factor VIII r.Ag (1:125). The slides were incubated in a humid chamber at room temperature for 1 h and rinsed in PBS. Normal rabbit serum immunoglobulin (1:100) was used for negative controls. Antibody binding was detected with diaminobenzidine tetrahydrochloride (Amresco). The slides were counterstained for 2 min with Gill's hematoxylin III diluted 1:10, followed by a 2-min tap water wash. Samples were dehydrated and then mounted in Permount. The CD-68 macrophage immunostaining (Dako) was handled in a similar fashion as the 5-LO and FLAP samples; however, the CD-68 antibody was diluted 1:50 and incubated for 30 min at room temperature.
Factor VIII r.Ag immunostaining was used to classify and define vascular lesions as shown in studies that evaluated angiogenesis in breast cancer (25). The similarly intense intracellular Factor VIII r.Ag staining seen in endothelial cells of both normal as well as PPH and ILD lungs was defined as 4 (maximum staining intensity), while normal rabbit serum samples defined the zero staining intensity (similar to the method of Giaid and colleagues [6]). Poorly defined, weak Factor VIII r.Ag immunostaining was seen in the subendothelial layer and in the matrix between the subintimal smooth muscle cells. This pattern probably represents Factor VIII r.Ag present in the subendothelial matrix after release from endothelial cell stores (3). Grading of immunostained tissue samples involved examination of macrophages and endothelial cells of small- (< 500 μm) and medium-sized (> 500 μm) blood vessels. Ten vascular structures or cell clusters per slide, identified under high power (×40), were scored for intensity of staining, and the mean of the 10 intensities for each tissue structure was recorded. The macrophage number was determined by examining 10 high power fields per patient and averaging the number of macrophages stained. Localization of macrophages was determined grossly as diffuse or clustered, and related to the adjacent structures (bronchial or vascular) to which they clustered.
After initial quantitation by one of us (L.W.), a second observer (C.C.) using the same criteria graded each slide but was blinded to the hypothesis or study design. Mean values per case were averaged and SEM computed for each diagnostic group and for each structure evaluated (macrophage, small-sized vessels, and medium-sized vessels). Comparisons between groups were by one-factor analysis of variance using Statview 512+ computer software, and intergroup differences were determined by the Fisher's protected least significant difference. Representative photographs were taken to illustrate the findings.
The details of the methodology applied for the detection of 5-LO mRNA are described elsewhere (14). Hybridization control consisted of an unrelated cRNA sequence (sense Vascular Endothelial Growth Factor cRNA) (14) labeled with digoxigenin and hybridized under similar conditions as the anti-sense 5-LO cRNA probe. mRNA in situ hybridization was performed as previously described (14). Because in situ hybridization for 5-LO required fresh frozen lung tissue, we obtained lung tissue from three additional patients with PPH and three normal control subjects (not included in the immunohistochemical sample). The lung morphology of these three patients with PPH was similar to the remaining eight PPH lungs subjected to immunohistochemistry for 5-LO and FLAP. Frozen sections from four patients with PPH and three normal control subjects were examined for the presence of 5-LO gene expression. In situ hybridization was performed three times on each of these tissues, with similar results.
The immunohistochemical staining for 5-LO and FLAP was most prominent in alveolar and perivascular macrophages. In addition, increased immunostaining intensity for 5-LO and FLAP was found in endothelial cells of the plexiform lesions and concentric lesions in the samples from patients with PPH, with light immunoreactivity in the vascular structures of control subjects and patients with ILD. Perivascular clustering of CD-68–positive macrophages was a predominant feature in PPH, as reported previously (3). A cluster of these perivascular macrophages is shown in Figure 1. 5-LO expression within macrophages was most prominent within the nuclei, with light reactivity in the cytoplasm. FLAP staining was predominantly cytoplasmic but also occurred in nuclei. Because of the size of a typical alveolar macrophage (12 μm) and the thickness of the sections (5 μm), it was not possible to follow the immunostaining pattern in the exact same tissue cluster of macrophages on multiple serial sections. However, macrophage perivascular clustering was observed in all samples from PPH lungs as described by us previously (3, 5).
Fig. 1. Macrophage expression of 5-LO and FLAP in lungs from patients with PPH. Numerous alveolar macrophage clusters were observed in PPH located in perivascular as well as remote parenchymal regions. (a) A cluster of macrophages that stained intensely for 5-LO and a vessel with endothelial cell expression of 5-LO (original magnification: ×400). The intracellular localization of 5-LO reactivity is most intense in the macrophage nuclei (arrow). (b) A similar cluster of cells identified by anti-FLAP antibody. In contrast to 5-LO, the staining is predominantly cytoplasmic. Note also the small-vessel endothelial cell staining for FLAP (arrow). (c) We confirmed the macrophage nature of these cell clusters using the macrophage-specific marker CD-68 which highlights the perivascular localization of these clusters ( V = vessel) (immunohistochemistry; original magnification: ×800).
[More] [Minimize]Endothelial cell expression of 5-LO and FLAP was observed in all PPH patients in a variety of remodeled vascular structures. A pulmonary artery with preserved vascular architecture, but with medial thickening, is seen in serial section in Figure 2. Factor VIII r.Ag staining was intense in the thin endothelial layer, yet faint staining (probably due to Factor VIII r.Ag present in the subendothelial matrix) (3) was also noted in the inner region of smooth muscle proliferation and hypertrophy. Of interest, 5-LO and FLAP similarly showed endothelial cell immunoreactivity. Inflammatory cells, present both intravascularly and perivascularly, and the adventitia stained intensely for 5-LO and were also FLAP reactive.
Fig. 2. Expression or 5-LO and FLAP in serial sections of a muscular pulmonary artery in PPH. (a) Intense Factor VIII r.Ag (original magnification: ×400) staining of vascular endothelial cells. Inset (original magnification: ×2,000) shows the pattern of intense Factor VIII r.Ag staining of the endothelium and a weak immunoreactivity present in the subendothelial matrix. The smooth muscle cells are negative for Factor VIII r.Ag. (b) 5-LO (original magnification: ×400) immunoreactivity is predominantly seen in perivascular and intravascular inflammatory cells (arrows) as well as in vascular endothelium and adventitia. There is a less intense, diffuse media staining. Similar reactivity is also noted in the endothelium of the small surrounding vessels. Inset (original magnification: ×2,000) highlights the degree of endothelial staining (arrowheads). Note the 5-LO expression, albeit at a lower level than in endothelial cells and macrophages, in medial smooth muscle cells (arrow). (c) Vascular endothelial cell and perivascular inflammatory cell expression of FLAP was somewhat less intense than staining with 5-LO (arrows). Inset (original magnification: ×2,000) with intravascular macrophage and endothelial cell staining. (d ) Normal rabbit serum control processed in an identical manner to 5-LO and FLAP (original magnficiation: ×400). Note the absence of inflammatory cell and vascular endothelial cell immunoreactivity, which underscores the lack of nonspecific staining and defines zero intensity for semi-quantitative analysis.
[More] [Minimize]Plexiform lesions were present in all lung samples obtained from patients with PPH. Serial sections of a plexiform lesion (Figure 3) exhibited intense intracellular Factor VIII r.Ag staining (confirming the endothelial cell nature of this cellular proliferation). 5-LO and FLAP immunostaining was localized to identical structures within the plexiform lesion. In the syncitium of cells of the remodeled vessel, 5-LO was expressed predominantly in the cytoplasm of endothelial cells, however, 5-LO was also present in the nuclei of endothelial cells lining the open vascular channels. FLAP was similarly present in the endothelial cells of the syncitium and in the cells lining the vascular channels, as well as in the infiltrating inflammatory cells. FLAP localization within endothelial cells was predominantly cytoplasmic; however, there was also a perinuclear distribution.
Fig. 3. Expression of 5-LO and FLAP in serial sections of a plexiform lesion in PPH. (a) Hematoxylin and eosin stain (original magnfication: ×800) shows the hallmark pathologic lesion of severe pulmonary hypertension. Vascular remodeling has resulted in a glomeruloid structure with multiple vascular channels within the lumen of the hypertensive vessel. (b) Factor VIII r.Ag (original magnification: ×800) of the same plexiform lesion, showing intense endothelial cell staining lining the vascular channels as well as endothelial cell proliferation within the glomeruloid structure. (c) 5-LO (original magnification: ×800) also shows endothelial cell staining within the vascular channels of the glomeruloid structure. (d ) Higher-power (original magnification: ×2,000) view of 5-LO. The majority of the Factor VIII r.Ag–positive cells express cytoplasmic immunoreactivity for 5-LO (arrows); however, a few flattened endothelial cells lining the open vascular channels express 5-LO in the nucleus (arrowhead). Note the intense immunoreactivity for 5-LO in perivascular and intravascular mononuclear cells, probably macrophages (open arrows). (e) FLAP (original magnification: ×1,200) and ( f ) FLAP (original magnification: ×2,000). FLAP is expressed by the same cells that express 5-LO. The endothelial cells show predominantly cytoplasmic FLAP expression (arrows); however, an occasional perinuclear reaction is noted. In the intravascular macrophages (open arrows), cytoplasmic FLAP reactivity is predominant.
[More] [Minimize]Concentric lesions (Figure 4), seen in three of eight patients (two additional patients demonstrated vessels with pathology intermediate between plexiform and concentric lesions), showed that the multiple layers of cells filling the vascular lumina were intensely Factor VIII r.Ag positive. 5-LO and FLAP immunoreactivity was more intense in the endothelial cells lining the lumen than in the abluminal endothelial cells. The inflammatory cell infiltrations at the outer boundary of concentric lesions consisted of macrophages and perivascular lymphocytic cells. The Factor VIII r.Ag–negative, CD-68–positive macrophages surrounding and infiltrating the concentric vascular proliferative lesions also expressed 5-LO and FLAP.
Fig. 4. Expression of 5-LO and FLAP in serial sections of a concentric lesion in PPH. (a) Factor VIII r.Ag (original magnification: ×400) in concentric endothelial cells (arrow), with nearly complete obliteration of the vascular lumen. (b) Serial section of the same concentric lesion (original magnfication: ×400) stained for 5-LO shows reactivity of endothelial cells lining the luminal remnants (arrow), with faint reactivity in the remainder of the endothelial cell layers. (c) Similarly, FLAP immunoreactivity (original magnification: ×200) shows faint endothelial cell immunoreactivity and more intensely stained perivascular inflammatory cells (arrow). (d ) Under higher power (original magnification: ×400), the luminal endothelial cell immunoreactivity is appreciated (arrow). (e) Macrophage-specific CD-68 immunostaining (original magnification: ×200) confirms the presence of macrophages in the outer limits of the concentric lesion (open arrow) as well as in the surrounding parenchyma (arrow). ( f ) Normal rabbit serum (original magnfication: ×400) as negative control with complete absence of staining.
Normal tissue samples are shown in Figure 5. Intense endothelial staining by Factor VIII r.Ag was observed. The endothelium demonstrated patchy low-intensity 5-LO staining, while the small muscular pulmonary arteries were essentially devoid of FLAP expression.
Fig. 5. Expression of 5-LO and FLAP in serial sections of a normal pulmonary vessel. (a) Factor VIII r.Ag. (b) 5-LO. (c) FLAP (d ) Normal rabbit serum (original magnification: ×200).
The semiquantitative analyses of consecutive sections from the eight patients with PPH, eight normal control subjects, and five patients with ILD for 5-LO and FLAP confirmed the initial microscopic observations (Figure 6). Staining of various intensities was observed in all samples, in all groups, for 5-LO and FLAP. The 5-LO and FLAP staining intensities for macrophages, when normalized to Factor VIII r.Ag expression on endothelial cells, were the same for all groups; however, the macrophage number (as identified by 5-LO staining positivity and confirmed in several cases by CD-68 staining) was significantly increased in patients with PPH versus normal control subjects (26 ± 3 macrophages per high power field versus 4 ± 1). Macrophage clustering in perivascular regions primarily accounted for the increased macrophage number in patients with PPH. Medium-sized blood vessels showed no significant difference in 5-LO staining intensity (2.1 ± 0.2 PPH versus 2.0 ± 0.2 normal versus 1.4 ± 0.4 ILD), although small vessels had significantly higher 5-LO staining intensity in PPH samples compared with normal or ILD samples (2.0 ± 0.2 PPH versus 1.4 ± 0.1 normal versus 0.7 ± 0.2 ILD; p < 0.01). FLAP staining intensity in macrophages was uniform for all groups, although the macrophage staining intensity was less than that of the same tissue stained for 5-LO. The staining intensity was significantly greater in PPH samples than in normal or ILD samples for both medium (2.3 ± 0.3 PPH versus 1.4 ± 0.2 normal versus 1.2 ± 0.2 ILD; p < 0.01) and small (2.0 ± 0.2 PPH versus 1.0 ± 0.1 normal versus 1.2 ± 0.1 ILD; p < 0.01) blood vessel endothelial cells, although there was staining in all tissues.
Fig. 6. Semi-quantitative analysis of 5-LO and FLAP staining intensity in the lungs obtained from normal control subjects or from patients with ILD or PPH. (Left panel ) 5-LO intensities for macrophages (m.phage), medium (> 500 μm) muscular pulmonary arteries, and small (< 500 μm) pulmonary arteries. Intensities are normalized to Factor VIII r.Ag staining (see text). (Right panel ) FLAP staining intensities. (*p < 0.01; bar = SD).
[More] [Minimize]Expression of 5-LO mRNA (Figure 7) was prominent in the macrophage clusters, in the monolayer endothelium of vessels with medial hypertrophy, in endothelial cells lining the plexiform and concentric lesions, as well as in the glomeruloid proliferation of endothelial cells in the plexiform lesions of patients with PPH (n = 4). Similar reactivity was absent in serial sections treated with a sense-strand, control probe. Normal vessel endothelium (n = 3) showed less pronounced 5-LO mRNA expression when compared with PPH vessels. No hybridization signal was detected in vascular smooth muscle cells.
Fig. 7. In situ hybridization for 5-LO mRNA in PPH and normal lungs. (a) Vessel showing medial hypertrophy (original magnfication: ×200) with abundant endothelial cell 5-LO mRNA expression. Note the absence of 5-LO transcripts in the smooth muscle layers. (b) The control probe of a serial section of the vessel in panel a lacks hybridization signal. (c) Plexiform lesion (original magnfication: ×600) demonstrating 5-LO gene expression in cells lining the vascular channels (open arrows). Note also the presence 5-LO transcripts scattered within the cellular proliferation (arrows). (d ) Macrophage cluster (original magnfication: ×400) from a patient with PPH, demonstrating intense reactivity for the 5-LO mRNA. (e) Concentric lesion (original magnfication: ×400) from a patient with PPH. Reactivity for 5-LO transcripts is most prominent in the luminal endothelial cells (arrow) but is also present in adjacent endothelial cell layers. Weak hybridization signal is present in the reactive pneumocytes surrounding the plexiform and concentric lesions. ( f ) Negative control (sense strand), in situ hybridization of the same concentric lesion as panel e, demonstrating lack of reactivity. (g) Vessel from a normal control subject, demonstrating a normal small pulmonary artery (original magnfication: ×200) with weak endothelial cell positivity to 5-LO (arrow). (h) Sense-strand, negative control of the same vessel as in panel g.
We have shown that the expression of 5-LO, the key enzyme of the arachidonic acid cascade required for leukotriene biosynthesis, and of its activating protein (FLAP), is increased in the lungs from patients with plexogenic (primary) pulmonary hypertension. In addition to the increased proteins by immunohistochemistry, we also demonstrated increased mRNA for 5-LO by in situ hybridization in macrophages and vascular endothelial cells. Although both 5-LO and FLAP were detectable by immune histology in lung tissue obtained from normal subjects and patients with ILD, these proteins were more abundant in PPH lungs. The PPH lungs also showed increased numbers of perivascular macrophages that expressed 5-LO and FLAP.
Recent data from this laboratory demonstrated for the first time 5-LO and FLAP staining of vascular endothelial cells in lungs from chronically hypoxic rats with pulmonary hypertension (14). In this rat model of pulmonary hypertension, endothelial cells remain in a monolayer, whereas in PPH many arteries also demonstrate dramatic endothelial cell proliferation as highlighted by staining against the endothelium marker Factor VIII r.Ag. The localization of 5-LO and FLAP in the human tissue in the current study is similar to that observed in the rat.
Based on the previous animal data (14) and the fact that the endothelium produces vasoactive mediators under both physiologic and pathologic conditions, we predicted that most of the cells from the human PPH lung samples, which were Factor VIII r.Ag positive (i.e., endothelial cells), would also stain with the antibodies directed against 5-LO and FLAP. Although this prediction proved to be correct, we did not anticipate the faint staining of smooth muscle cells of the media. Our in situ hybridization findings of increased steady-state levels of 5-LO transcripts in endothelial cells of remodeled pulmonary vessels, but not within the smooth muscle cell layer, suggest that 5-LO protein may have diffused into neighboring vascular smooth muscle cells.
The staining pattern for 5-LO in macrophages was predominantly nuclear, suggesting a state of cell activation (20, 22). Increased localization of 5-LO to the nuclear membrane of macrophages has been recently described in lung tissue from patients with idiopathic pulmonary fibrosis compared with lung tissue from normal control subjects (26). Thus, enzyme translocation resulting in macrophage activation, as demonstrated by perinuclear 5-LO distribution, may be a common theme in inflammatory diseases of the lung. In the present study, FLAP immunoreactivity was predominantly cytoplasmic in endothelial cells, yet occasionally demonstrated a perinuclear distribution. We can think of several reasons for a cytosolic localization of the FLAP polypeptide in pulmonary hypertensive cells. First, in pulmonary cells exposed to chronic hypertensive stress, protein trafficking may be deranged, resulting in misplacement of the FLAP polypeptide into a cytosolic membranous compartment rather than its ‘typical' placement within the lipid bilayer of the nuclear envelope (16). Second, within the pulmonary hypertensive endothelial cell, processes that govern FLAP post-translational protein modification may be abnormal relative to the nondiseased state, leading to a ‘novel' FLAP polypeptide prone to cytosolic compartmentalization. A third possibility rests upon the processes of membrane biogenesis, turnover, and stability within pulmonary cells exposed to chronic hypertension. Changes in critical cellular equilibria associated with lipid bilayers may have unanticipated consequences for protein-lipid interactions that lead to a prominent non-nuclear FLAP localization. Presently, it is not clear if cytosol FLAP is functional with respect to the current model of 5-LO activation.
It is possible that the augmented staining of vascular wall cells for 5-LO and FLAP is a marker of inflammatory activity. If this staining pattern indeed indicates inflammation, it is a form of inflammation certainly different from the classic vasculitic changes that can occur in PPH (1). The overexpression of endothelin (6) and the decreased nitric oxide synthase (27) in PPH and, as reported here, the overexpression of 5-LO and FLAP, may all characterize the phenotypically altered hypertensive and dysfunctional endothelial cell that is part of the remodeling process. Given that endothelin and 5-LO are associated with inflammatory activities—endothelin can have chemotactic activity (6) and the 5-LO product LTB4 can attract lymphocytes (15)—the overexpression of these proteins in PPH vessels indicates to us that, in PPH (in addition to the ‘professional' inflammatory cells), smooth muscle and endothelial cells may participate in inflammatory activity.
An alternative explanation to the concept of endothelial cells and smooth muscle cells participating in inflammation is that overexpression of 5-LO and FLAP proteins in PPH may have a role in cell proliferation process independent of production of leukotrienes. For example, downregulation of the 12-lipoxygenase gene, by specific antisense oligonucleotides or by inhibition of 12-LO enzyme in lung cancer cell lines, results in enhanced apoptosis (28), and specific antagonists of 5-LO metabolism cause significant growth reduction in lung cancer cell lines. Interruption of 5-LO signalling also results in enhanced levels of apoptosis (29, 30). 5-LO may also promote cell growth nonenzymatically via binding to SH-3 regions of target proteins involved in signal transduction in vascular cells (31). Regardless, whether vascular 5-LO and FLAP overexpression in PPH are part of an inflammatory or proliferative program, the overexpression of these proteins indicates cell activation in the diseased vessels rather than the presence of inert scar tissue.
In contrast to the plexiform lesions, only the luminal endothelial cells of the concentric lesions stain intensely for 5-LO and FLAP and express the 5-LO gene, whereas the remaining endothelial cell layers are only weakly immunopositive. We hypothesize that the concentric lesions are further advanced in the remodeling process (i.e., older) and that abluminal endothelial cells lose their immunoreactivity. If so, then this would also suggest that the role of 5-LO is perhaps to support cell growth and proliferation.
At the present time, neither 5-LO inhibitors nor FLAP inhibitors are available for treatment of patients with pulmonary hypertension, and one can only speculate that increased lung vascular (and macrophage) production of leukotrienes may contribute to increased pulmonary artery pressure, vascular cell growth, or recruitment of inflammatory cells. As much as data from animal models of pulmonary hypertension (14) can be translated to the human condition of severe pulmonary hypertension one would expect that the 5-LO/FLAP system contributes to the pathogenesis of severe pulmonary hypertension. Unfortunately, lung tissue from patients with severe pulmonary hypertension seldom becomes available for direct assessment of lung homogenate leukotriene release.
In summary, we have shown an overexpression of 5-LO and FLAP proteins by immunohistology, and of 5-LO mRNA by in situ hybridization, in the small- and medium-sized pulmonary arteries of PPH, relative to comparable vessels in normal or ILD tissue. Inasmuch as the vascular alterations in ILD are primarily caused by vascular medial hypertrophy and not by endothelial cell proliferation, the vascular overexpression of 5-LO and FLAP in PPH, and not in ILD, is consistent with the concept of PPH as an inflammatory/proliferative vasculopathy. In addition, activated macrophages may be involved in the formation of the concentric and plexiform lesions and thus may play a role in the pathogenesis of PPH.
The writers wish to thank Mr. Kelly Wade for his expert technical assistance.
| 1. | Wagenvoort C. A., Wagenvoort N.Primary pulmonary hypertension: a pathologic study of the lung vessels in 156 clinically diagnosed cases. Circulation42197011631184 |
| 2. | Pietra G. G., Edwards W. D., Kay J. M., Rich S., Kernis J., Schloo B., Ayres S. M., Bergofsky E. H., Brundage B. H., Detre K. M., Fishman A. P., Goldring R. M., Groves B. M., Levy P. S., Reid L. M., Vreim C. E., Williams G. W.Histopathology of primary pulmonary hypertension: a qualitative and quantitative study of pulmonary blood vessels from 58 patients in the National Heart, Lung, and Blood Institute, Primary Pulmonary Hypertension Registry. Respiration5219878185 |
| 3. | Tuder R. M., Groves B., Badesch D., Voelkel N. F.Exuberant endothelial cell growth and elements of inflammation are present in plexiform lesions of pulmonary hypertension. Am. J. Pathol.1441994275285 |
| 4. | Caslin A. W., Heath D., Madden B., Yacoub M., Gosney J. R., Smith P.The histopathology of 36 cases of plexogenic pulmonary arteriopathy. Histopathology161990919 |
| 5. | Cool, C. D., D. Kennedy, N. F. Voelkel, and R. M. Tuder. 1997. Pathogeneis and evolution of plexiform lesions in pulmonary hypertension associated with scleroderma and human immunodeficiency virus infection. Hum. Pathol. 434–442. |
| 6. | Giaid A., Yanagisawa M., Langleben D., Michel R. P., Levy R., Shennib H., Kimura S., Masaki T., Duguid W. P., Stewart D. J.Expression of endothelin-1 in the lungs of patients with pulmonary hypertension N. Engl. J. Med.328199317321739 |
| 7. | Botney M. D., Parks W. C., Crouch E. C., Stenmark K., Mecham R. P.Transforming growth factor-B1 is decreased in remodeling hypertensive bovine pulmonary arteries. J. Clin. Invest.89199216291635 |
| 8. | Tuder R. M., Flook B. E., Voelkel N. F.Increased gene expression for VEGF and VEGF receptors KDR/Flk and Flt in lungs exposed to acute or chronic hypoxia: modulation of gene expression by nitric oxide. J. Clin. Invest.95199517981807 |
| 9. | Meyrick B., Reid L.Development of pulmonary arterial changes in rats fed Crotalaria spectabilis. Am. J. Pathol.9419793750 |
| 10. | Herget J., Palecek F., Preclik P., Cermakova M., Vizek M., Petrovicka M.Pulmonary hypertension induced by repeated pulmonary inflammation in the rat. J. Appl. Physiol. Respir. Environ. Exercise Physiol.511981755761 |
| 11. | Wilson D. W., Segall H. J., Pan L. C., Dunston S. K.Progressive inflammatory and structural changes in the pulmonary vasculature of monocrotaline-treated rats. Microvasc. Res.3819895780 |
| 12. | Meyrick B., Brigham K. L.Repeated Escherichia coli endotoxin-induced pulmonary inflammation causes chronic pulmonary hypertension in sheep: structural and functional changes. Lab. Invest.551986164176 |
| 13. | Ono S., Wescott J. Y., Voelkel N. F.PAF antagonists inhibit pulmonary vascular remodeling induced by hypobaric hypoxia in rats. J. Appl. Physiol.73199210841092 |
| 14. | Voelkel N. F., Tuder R. M., Wade K., Hoper M., Lepley R. A., Goulet J. L., Koller B. H., Fitzpatrick F.Inhibition of 5-lipoxygenase-activating protein (FLAP) reduces pulmonary vascular reactivity and pulmonary hypertension in hypoxic rats. J. Clin. Invest.97199624912498 |
| 15. | Goetzl E. J., An S., Smith W. L.Specificity of expression and effects of eicosanoid mediators in normal physiology and human diseases. FASEB J.9199510511058 |
| 16. | Ford-Hutchinson A. W., Gresser M., Young R. N.5-Lipoxygenase. Annu. Rev. Biochem.631994383417 |
| 17. | Samuelsson B.Leukotrienes: mediators of immediate hypersensitivity reactions and inflammation. Science2201983568575 |
| 18. | MacDermot J., Kelsey C. R., Waddell K. A., Richmond R., Knight R. K., Cole P. T., Dollery C. T., Landon D. N., Blair A. I.Synthesis of leukotriene B4 and prostanoids by human alveolar macrophages: analysis by gas chromatography/mass spectrophotometry. Prostaglandins271984163177 |
| 19. | Coffey M. J., Wilcoxen S. E., Peters-Golden M.Increases in 5-lipoxygenase activating protein expression account for enhanced capacity for 5-lipoxygenase metabolism that accompanies differentiation of peripheral blood monocytes into alveolar macrophages Am. J. Respir. Cell Mol. Biol.111994153158 |
| 20. | Woods J. W., Coffey M. J., Brock T. G., Singer I. I., Peters-Golden M.5-Lipoxygenase is located in the euchromatin of the nucleus in resting human alveolar macrophages and translocates to the nuclear envelope upon cell activation. J. Clin. Invest.95199520352046 |
| 21. | Rouzer C. A., Kargman S. J.Translocation of 5-lipoxygenase to the membrane in human leukocytes challenged with ionophore A23187. Biol. Chem.26319881098010988 |
| 22. | Coffey M., Peters-Golden M., Fantone J. C., Sporn P. H. S.Membrane association of active 5-lipoxygenase in resting cells: evidence for novel regulation of the enzyme in the rat alveolar macrophage. J. Biol. Chem.2671992570576 |
| 23. | Kargman J., Vickers P. J., Evans J. F.A23187-induced translocation of 5-lipoxygenase in osteosarcoma cells. J. Cell Biol.119199217011709 |
| 24. | Merz H., Malisius R., Mannweiler S., Zhou R., Hartmann W., Orscheschek K., Moubayed P., Feller A. C.Methods in laboratory investigation. Immunomax: a maximized immunohistochemical method for the retrieval and enhancement of hidden antigens. Lab. Invest.731995149156 |
| 25. | Weidner N.Intratumor microvessel density as a prognostic factor in cancer. Am. J. Pathol.1471995919 |
| 26. | Wilborn J., Baille M., Coffey M., Burdick M., Strieter R., Peters-Golden M.Constitutive activation of 5-lipoxygenase in the lungs of patients with idiopathic pulmonary fibrosis. J. Clin. Invest.97199618271836 |
| 27. | Giaid A., Saleh D.Reduced expression of endothelial nitric oxide synthase in the lungs of patients with pulmonary hypertension. N. Engl. J. Med.3331995214221 |
| 28. | Tang D. G., Chen Y. Q., Honn K. V.Arachidonate lipoxygenases as essential regulators of cell survival and apoptosis. Proc. Natl. Acad. Sci. U.S.A.93199652415246 |
| 29. | Avis I. M., Jett M., Boyle T., Vos M. D., Moody T., Treston A. M., Martinez A., Mulshine J. L.Growth control of lung cancer by interruption of 5-lipoxygenase-mediated growth factor signalling. J. Clin. Invest.971996806813 |
| 30. | Anderson K. M., Levin J., Jajeh A.Induction of apoptosis in blood cells from a patient with acute myelogenous leukemia by SC41661A, a selective inhibitor of 5-lipoxygenase. Prostaglandin Leukot. Essen. Fatty Acids481993323 |
| 31. | Lepley R. A., Fitzpatrick F. A.5-Lipoxygenase contains a functional Src homology 3-binding motif that interacts with the Src homology 3 domain of Grb2 and cytoskeletal proteins. J. Biol. Chem.26919942416324168 |
