Published ahead of print on February 20, 2003, doi:10.1164/rccm.200210-1233OC
© 2003 American Thoracic Society Enhanced Expression of Vascular Endothelial Growth Factor in Pulmonary Arteries of Smokers and Patients with Moderate Chronic Obstructive Pulmonary Disease
Departments of Pulmonary Medicine and Pathology, Institut d'Investigacions Biomèdiques August Pi i Sunyer, Hospital Clínic Correspondence and requests for reprints should be addressed to Dr. Joan A. Barberà, Servei de Pneumologia, Hospital Clínic, Villarroel, 170, 08036 Barcelona, Spain. E-mail: jbarbera{at}clinic.ub.es
Chronic obstructive pulmonary disease (COPD) is associated with structural and functional changes in the pulmonary circulation that commence at an early stage. To investigate whether vascular endothelial growth factor (VEGF) might be implicated as a mediator in COPD-associated pulmonary vascular changes, we studied surgical specimens obtained from 19 nonsmokers, 21 smokers with normal lung function, 28 patients with moderate COPD, and 10 patients with severe emphysema. The expression of VEGF in pulmonary muscular arteries was evaluated by immunohistochemistry, its protein content in lung tissue by Western blot analysis, and VEGF mRNA and its isoforms were analyzed by reverse transcription-polymerase chain reaction. The immunohistochemical expression of VEGF was increased in pulmonary arteries of smokers (median, 68% [interquartile range, 60–88]) and patients with moderate COPD (77% [63–82]), compared with nonsmokers (53% [40–63]) (p < 0.05 each). The expression of VEGF in smooth muscle cells correlated with the thickness of the vessel wall (r = 0.38, p < 0.01). VEGF protein content in lung tissue was reduced in severe emphysema, where reverse transcription-polymerase chain reaction demonstrated a lower proportion of the VEGF189 isoform. In conclusion, the expression of VEGF varies according to the severity of COPD and might be involved in the pathogenesis of pulmonary vascular remodeling at early stages of the disease.
Key Words: chronic obstructive pulmonary disease • cigarette smoking • pulmonary artery • endothelium • vascular endothelial growth factor Pulmonary vascular remodeling leading to pulmonary hypertension and cor pulmonale is a characteristic feature of chronic obstructive pulmonary disease (COPD) (1, 2). Hypoxia has been classically considered the major pathogenic mechanism of these changes. However, structural abnormalities of pulmonary arteries are not exclusive of advanced COPD, as they have been shown also in patients with mild COPD without arterial hypoxemia and even in smokers with normal lung function (3). In agreement with this notion, recent studies suggest that the natural history of pulmonary hypertension in COPD might commence at moderate degrees of disease severity (4). In a previous study, we showed endothelial dysfunction in pulmonary arteries of patients with mild COPD, which was associated with an impaired release of endothelium-derived nitric oxide (NO) (5). Subjects with greater impairment in endothelial function were those with more pronounced vascular remodeling (intimal thickening) (5). Because the endothelium plays a central role in regulating vascular tone and controlling cell growth, we hypothesized that the impairment of endothelial function might promote pulmonary vessel remodeling at this early stage. At this time, it is not known which mediators are involved in this process. Growth factors are intercellular signaling molecules that regulate cell proliferation by autocrine and paracrine mechanisms. Vascular endothelial growth factor (VEGF) is abundantly expressed in cells of lung tissue. This molecule is mainly implicated in the maintenance of vascular endothelial cell function and in vascular cell proliferation (6, 7). Some of these functions are performed through NO-dependent mechanisms. VEGF has been involved in the vascular remodeling of primary pulmonary hypertension, which is characterized by endothelial and smooth muscle cell (SMC) proliferation (8). Cigarette smoking may upregulate VEGF, as suggested by an acute increase of VEGF plasma levels during smoking (9). Furthermore, it has been shown that components of cigarette smoke enhance VEGF mRNA and protein expression in cultured endothelial cells of carotid artery (10). Accordingly, we hypothesized that VEGF could play a role in the pathogenesis of the intimal cell proliferation shown in pulmonary arteries of smokers and patients with mild COPD (11). Interestingly, it has been recently suggested that VEGF might be involved in the pathogenesis of emphysema through apoptotic mechanisms (12). This concept has promoted a growing interest in clarifying the role of VEGF at different stages of COPD. With this background in mind, this study was addressed to investigate the potential role of VEGF in the pathogenesis of the vascular changes that take place in COPD. Accordingly, we evaluated the expression of VEGF protein and mRNA in pulmonary arteries and lung tissue of smokers and patients with different degrees of COPD severity. Preliminary results of the study have been previously reported in abstract form (13).
Lung tissue specimens from 78 patients were studied: 68 from lung resection for localized lung carcinoma and 10 from lung volume reduction surgery performed in patients with severe emphysema. Part of these tissue samples has been used in previous studies from our group (5, 14). Preoperative functional measurements were performed in all patients. Patients who underwent lung resection for lung carcinoma were divided according to their smoking habit and pulmonary function into three groups: nonsmokers, smokers with normal lung function, and moderate COPD. An additional fourth group included 10 patients with emphysema who underwent lung volume reduction surgery (15, 16). None of the patients with severe emphysema had pulmonary hypertension, as documented by right heart catheterization. General characteristics and lung function measurements in the four groups of patients are shown in Table 1 . The study was approved by the Committee on Human Research at our institution.
Morphometric characteristics of pulmonary muscular arteries with an external diameter of less than 500 µm and with complete elastic laminas were analyzed in sections stained with orcein using a computerized image analyzer, as previously reported (3). The areas occupied by the muscular layer, the intima and the lumen, were expressed as a percentage of the total area. Formalin-fixed paraffin-embedded tissue sections were immunostained with a mouse monoclonal antibody antihuman VEGF using the avidin–biotin complex/horseradish peroxidase method. A mean of 46 ± 26 arteries was analyzed in each subject. The number of pulmonary muscular arteries with positive immunoreactivity to VEGF in the vessel wall (either in endothelial or SMCs) was expressed as the percentage of the total number of arteries. The intensity of the immunoreaction was additionally graded as absent, 0; mild, 1; moderate, 2; or intense, 3. An average grading score was computed in each patient. Western blot analysis was performed from frozen samples of 36 patients (9 in each group). Some of these frozen tissues were collected after performing the immunohistochemical studies. Briefly, 100 µg of protein were loaded in each lane of a sodium dodecyl sulfate-polyacrylamide gradient gel (from 10 to 15%) and submitted to electrophoresis. Subsequently, proteins were transferred to PVDF membranes and incubated with an anti-VEGF antibody at a 1:200 dilution. After incubation with the secondary antibody, specific immunoreactivity was detected with the avidin–biotin complex/horseradish peroxidase method. Protein bands were visualized by enhanced chemiluminescence. Band intensity was quantitated by densitometry and was normalized for the expression of two housekeeping proteins and the mean density of all bands, as previously reported (14). Total RNA was extracted from frozen fresh lung tissue of 30 cases using Trizol Reagent. Reverse transcription-polymerase chain reaction (RT-PCR) for VEGF was performed as previously described (17). Under the exponential phase of PCR amplification (30 cycles), products of 403, 535, and 607 bp long were detected, corresponding to the 121-, 165-, and 189-amino acid VEGF isoforms, respectively. Data were expressed as mean ± SD or as medians and interquartile range according to whether the variable followed a normal distribution (Kolmogorov-Smirnov test). Comparisons between groups were performed using the analysis of variance or the Kruskall-Wallis test, when appropriate. Post hoc comparisons were performed using the Student's t test or the Mann-Whitney U-test for parametric and nonparametric tests, respectively. Correlations between variables were analyzed with Pearson's coefficient. Probability values lower than 0.05 were considered significant.
Functional and Morphometric Measurements The functional characteristics of the groups are reported in Table 1. Patients in the four groups were matched by age. Smokers were heavy cigarette consumers with a mean of 50 pack-years (range, 10 to 90 pack-years) without differences among them. Fifty-three percent (n = 31) were ex-smokers, and 47% (n = 28) were current smokers. In the moderate COPD group, arterial PO2 was normal or only slightly altered. Patients with severe emphysema had severe airflow obstruction, hyperinflation, and air trapping. All patients in this group were hypoxemic, and three had chronic respiratory failure. Morphometric measurements of pulmonary muscular arteries are summarized in Table 2 . We examined a similar number of arteries per patient in each group. According to morphometric analysis, the intimal layer was enlarged in smokers and in patients with moderate COPD, as compared with nonsmokers, in agreement with previous studies (3, 5). In the severe emphysema group, intimal enlargement was greater than in the smoker and moderate COPD groups. There were no differences in the thickness of the medial layer between groups.
Immunohistochemical Analysis In nonsmoker control cases, positive immunoreaction to VEGF antibody was observed in both endothelial and medial SMCs of the arterial wall, macrophages, and airway epithelial cells. Intense immunoreactivity to VEGF was observed in pulmonary arteries of the smoker and moderate COPD groups. In these patients, positive immunoreaction was shown in the endothelium, the medial layer, and also the cells (SMCs) proliferating in the enlarged intimas (Figure 1) (11). Compared with nonsmokers, semiquantitative analysis of the immunohistochemical data revealed that the proportion of pulmonary arteries showing positive immunoreaction to VEGF in the arterial wall was greater and the intensity of the immunoreaction stronger in both the smoker and the moderate COPD groups (Figure 2 and Table 3) . It should be noted that differences between groups were mainly determined by the expression in SMCs (Table 3). In severe emphysema, the expression of VEGF showed a trend to be lower than in moderate COPD.
The expression of VEGF was unrelated to the current smoking status, as the number of positive arteries in ex-smokers and current smokers was similar (median 71% [interquartile range, 56–91%] and 77% [59–86%], respectively; p = 0.64). The immunohistochemical expression of VEGF in SMCs correlated with the arterial wall thickness (r = 0.38, p < 0.01) (Figure 3) .
Western Blot Analysis of VEGF Protein Levels Protein extracts from lung tissue samples were examined for VEGF protein content by Western blot analysis. The monoclonal antibody to VEGF detected two protein bands at a molecular mass of 23 and 29 kD (Figure 4) that correspond to the VEGF165 and VEGF189 isoforms, respectively. Densitometric analysis of protein bands revealed a reduced VEGF protein content in patients with severe emphysema, which was due to a reduction in the VEGF189 isoform (Figure 4 and Table 3). No statistical differences were found among nonsmokers, smokers, and patients with moderate COPD. Protein signal of the VEGF165 isoform was similar in all groups. Total VEGF protein content was positively correlated with arterial PO2 (r = 0.43, p < 0.05).
VEGF mRNA Analysis by RT-PCR Total VEGF mRNA in lung tissue, measured by RT-PCR, did not differ between groups. Nevertheless, the relative proportion among VEGF isoforms differed in patients with severe emphysema (Figure 5 and Table 4) . The proportional contribution of the VEGF189 isoform to total VEGF mRNA was reduced in the severe emphysema group compared with the other three groups. This finding is consistent with the reduced expression of the 29-kD band in the severe emphysema group shown in the Western blot analysis. No differences in the PCR products of the VEGF121 and the VEGF165 isoforms were observed among the four groups (Table 4).
The results of this study show an increased expression of VEGF in pulmonary muscular arteries of patients with moderate COPD and also in smokers with normal lung function, as compared with nonsmokers, and that this expression is associated with the enlargement of the arterial wall. In contrast, in patients with severe emphysema, the immunohistochemical expression of VEGF in pulmonary arteries and its protein content in lung tissue tends to be low, despite intense vascular remodeling. This study was mainly addressed to investigate the potential role of VEGF in the pathogenesis of pulmonary vascular remodeling that takes place in the initial stage of COPD (3, 5). At this stage, remodeling is characterized by SMCs proliferation and extracellular matrix deposition in the intimal layer of pulmonary muscular arteries (11). A major finding of this investigation was an enhanced expression of VEGF in the wall of pulmonary arteries in patients with moderate COPD and in smokers with normal lung function, who also had enlarged intimas. In pulmonary vessels, VEGF can be expressed in both endothelial cells and SMCs (18, 19). In our study, differences among groups were essentially due to the expression of VEGF in SMCs, located either in the medial or in the intimal layer, whereas the expression in endothelial cells did not differ significantly (Table 3). Interestingly, the expression of VEGF in SMCs correlated with the thickening of the arterial wall (Figure 3), suggesting a potential role of VEGF in the pathogenesis of pulmonary vascular remodeling. The role of VEGF on SMC activity remains unsettled. In vitro studies indicate that VEGF may induce and modulate vascular SMC proliferation and migration through the upregulation of matrix metalloproteinases (7, 20) and that the VEGF receptors flt-1 and flk-1 are expressed in vascular SMCs (7, 21). Experimental studies in animal models show that VEGF may promote SMC proliferation and neointimal formation after vascular injury (22) and that recombinant VEGF enhances the progression of the atherosclerotic plaque (23, 24). In humans, VEGF has been related with the progression of coronary atherosclerosis (25). Consequently, we hypothesize that in the early stages of COPD, VEGF might contribute to the structural remodeling of pulmonary arteries, presumably by enhancing the proliferation and intimal migration of SMCs. Nevertheless, the expression and function of VEGF receptors involved in this process need further clarification to confirm the potential role of VEGF in mediating these vascular changes. Some of the biologic actions of VEGF are related to NO-dependent mechanisms (26). The inhibition of NO synthesis with N-nitro-L-arginine methyl ester has been shown to upregulate the gene expression of VEGF and VEGF receptor in a model of isolated perfused lung (27). It has been recently reported that long-term inhibition of NO by administering N-nitro-L-arginine methyl ester, induces coronary vascular arteriosclerosis (28). In this animal model, N-nitro-L-arginine methyl ester produced a significant increase of VEGF gene expression (29). Furthermore, when VEGF receptors were blocked, there was a decrease in the proliferative changes of the coronary artery (29), hence implicating this growth factor in the process. These results suggest that endothelial dysfunction and decreased NO synthesis may facilitate the upregulation of VEGF. In previous studies, we have shown endothelial dysfunction and decreased endothelial NO synthase expression in pulmonary arteries of smokers and patients with mild COPD (5, 14), changes that may likely promote VEGF activity. Indeed, in 25 cases of this series, in which we also had data on endothelium-dependent vascular relaxation (5), we have found an inverse relationship between VEGF expression and endothelial function, as assessed by the maximal relaxation of pulmonary artery rings induced by adenosine diphosphate (r = -0.52, p < 0.01). Accordingly, endothelial dysfunction and impaired NO synthesis, which are present in smokers and patients with mild COPD, may facilitate the upregulation of VEGF in pulmonary vessels. It could be speculated that upregulation of VEGF could be beneficial for preventing the development of pulmonary hypertension, as previously reported by Partovian and colleagues (30) in an hypoxic rat model. Nevertheless, it seems unlikely that VEGF could play a "vasodilator" role in our series because we found an inverse relationship between VEGF expression and endothelium-dependent vascular relaxation, suggesting that upregulation of VEGF occurred in the context of a vasoconstrictor/vasodilator imbalance that favors pulmonary hypertension. Hypoxia is a major factor involved in the induction of VEGF gene expression, being considered one of the major stimuli accounting for pulmonary vascular remodeling in chronic lung diseases (27, 31). However, the potential mechanisms that implicate VEGF in the early vascular remodeling of COPD do not seem to be related to hypoxia, as the majority of the patients in our series had normal arterial oxygen tension. Indeed, we found a weak although significant correlation between PaO2 values and protein content of VEGF in lung tissue. Therefore, it is necessary to consider other potential triggers for the increased VEGF activity at early stages of COPD. Inflammatory cell (CD8+ T-lymphocytes) infiltrate has been shown in the adventitial layer of pulmonary arteries in both smokers and patients with mild COPD (32). The degree of lymphocytic infiltrate correlated with the extent of intimal thickening (32). It is unknown, however, to what extent these recruited inflammatory cells are a cause or a consequence of vascular remodeling. Conceivably, an inflammatory stimulus could be a potential trigger for VEGF upregulation at initial stages of the process. Many inflammatory mediators, including prostaglandin E2, interleukin-6, and interleukin-1 have been shown to induce VEGF mRNA and protein expression (33, 34). In addition, it has been shown that macrophages and T-lymphocytes express VEGF receptors (35) and that VEGF induces monocyte activation and migration (36, 37). Accordingly, it might be speculated that increased VEGF activity in pulmonary arteries at early stages of COPD could be related, to some extent, with an underlying inflammatory process. It is conceivable that cigarette smoke products could be potential triggers for VEGF because its expression was increased in pulmonary arteries of smokers with normal lung function. This suggestion is supported by the findings of Conklin and colleagues (10), who demonstrated that nicotine and cotinine, in doses similar to those seen in the plasma of current smokers, upregulate VEGF expression in endothelial cells. Furthermore, a recent study by Wright and colleagues (38) demonstrates upregulation of VEGF gene expression and its receptor (flk-1) in pulmonary arteries of rats exposed to cigarette smoke. Finally, it can be argued that in our series the presence of a neoplasm could be a potential trigger for VEGF. However, we consider this unlikely because differences in VEGF expression were observed between nonsmokers, smokers, and patients with moderate COPD, all of them having a neoplasic process. Moreover, for the purposes of this study, we sampled areas of lung tissue with normal appearance, far from the neoplasic lesion, hence being doubtful the influence of the neoplasic lesion in these areas. Compared with patients with moderate COPD, the immunohistochemical expression of VEGF and its protein content was lower in patients with severe emphysema. These findings are consistent with recent observations by Kasahara and colleagues (12) demonstrating decreased VEGF expression in alveolar septal cells in emphysema. In our study, a quantitative analysis of VEGF expression in alveolar structures was not performed. Nevertheless, the fact that VEGF content in protein extracts of lung tissue was significantly reduced in patients with emphysema suggests that VEGF expression was diminished not only in pulmonary arteries but also in other lung structures, thus supporting the observation by Kasahara and colleagues (12). The reduced immunohistochemical expression of VEGF in pulmonary arteries occurred despite that patients with severe emphysema had marked remodeling in these vessels. A potential explanation for this finding could be related to the heterogeneous phenotype of SMCs proliferating in the intimal layer. In a recent study, we observed that a significant proportion of SMCs proliferating in the intimal layer of remodeled arteries of smokers and patients with mild COPD did not express contractile filaments, suggesting a less differentiated state and a synthetic phenotype (11). Inoue and colleagues (25) have suggested that nondifferentiated SMCs (synthetic phenotype) express more VEGF than differentiated SMCs (normal contractile phenotype). Thus, it can be speculated that pulmonary vascular remodeling in COPD might be an evolving process. At an early stage of the disease, vascular remodeling is produced by poorly differentiated SMCs with a high expression of VEGF, whereas at advanced stages, remodeling is mainly conformed by differentiated SMCs expressing less VEGF. RT-PCR studies revealed a reduced proportion of VEGF189 mRNA isoform in patients with severe emphysema. This finding is consistent with a reduced protein signal of the 29-kD molecular weight band in the Western blot analysis. The existence of multiple VEGF isoforms, which vary in localization, accessibility, receptor binding, and activity, raises the possibility that individual VEGF isoforms might mediate different aspects of vascular pathophysiology (39). The VEGF189 isoform appears to be involved in vessel development (40) and maintenance (41). Thus, a reduction of this isoform might be implicated in the apoptotic changes that take place in the alveolar septa of emphysematous lungs, as suggested by Kasahara and colleagues (12). In addition, VEGF189 is completely retained on the cell surface and in the extracellular matrix, from where it can be released as a result of the activity of proteases (6, 39). As a consequence, the decrease of the VEGF189 isoform might also be related to protease–antiprotease imbalance and/or lung tissue destruction. In summary, the results of this study show that the expression of VEGF in pulmonary arteries of patients with COPD may vary according to the severity of the disease. At an early stage, VEGF might be an important signaling molecule linking changes in endothelial function with vessel remodeling. In contrast, in patients with severe emphysema the vascular expression of VEGF decreases in agreement with changes that take place in the underlying lung parenchyma.
The authors thank M. Sánchez and M. Cerrillo for their technical work in immunohistochemical studies and the Department of Thoracic Surgery for their help in the collection of the lung specimens.
Supported by grants from the Fondo de Investigación Sanitaria (FIS 99/188 and 00/922) and the Sociedad Española de Neumología y Cirugía Torácica (SEPAR-1999). Dr. Santos is the recipient of a Research Fellowship award from SEPAR. Dr. Peinado is supported by the Comissió Interdepartamental de Recerca i Innovació Tecnològica and the Fundació Clínic per a la Recerca Biomèdica. Dr. Morales-Blanhir is the recipient of a Research Fellowship award from the European Respiratory Society. This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form October 29, 2002; accepted in final form February 17, 2003
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
