Published ahead of print on September 24, 2004, doi:10.1164/rccm.200406-698OC
© 2005 American Thoracic Society doi: 10.1164/rccm.200406-698OC
Allergen Exposure of Mouse Airways Evokes Remodeling of both Bronchi and Large Pulmonary VesselsDepartment of Physiological Sciences and Department of Clinical Pharmacology, Lund University Hospital, Lund, Sweden Correspondence and requests for reprints should be addressed to Kristina Rydell Törmänen, M.Sc., Department of Physiological Sciences, BMC F10, Lund University, 221 84 Lund, Sweden. E-mail: kristina.rydell{at}mphy.lu.se
Remodeling of airway structures is a well-documented feature of allergic airway inflammation. To investigate whether bronchial remodeling is associated with remodeling of adjacent pulmonary vessels, sensitized mice were subjected to repeated ovalbumin inhalations, and bronchi and pulmonary vessels were subjected to histologic analysis. Allergen challenges induced peribronchial as well as perivascular eosinophilia. Remodeling of systemic airway microcirculation, as studied in tracheal whole-mount preparations, revealed an allergen-induced increase in both the diameter and length of the airway microvessels. Immunostaining for -smooth muscle actin disclosed an increase in smooth muscle mass in both bronchi and large pulmonary vessels. Both bronchi and pulmonary vessels also displayed increased expression of procollagen I and procollagen III. Staining for proliferating cell nuclear antigen revealed increased proliferation of bronchial epithelial and smooth muscle cells as well as pulmonary vascular endothelial and smooth muscle cells. We conclude that central features of remodeling that take place in allergen-exposed airways are present also in the pulmonary vessels. The significance of this finding with respect to occurrence in disease, pathophysiologic importance, and involved mechanisms warrants further investigation.
Key Words: animal model asthma eosinophils vascular remodeling Allergic asthma involves airway hyperresponsiveness, persistent airway inflammation, and tissue remodeling (1). The airway remodeling includes thickened basement membrane (BM), subepithelial fibrosis (2), increased smooth muscle mass (3), and changes in airway mucosal vascularity (4). In patients with asthma, tissue samples are mainly obtained as bronchial biopsies, necessarily restricting the focus to large airways. However, observations of autopsies from patients with asthma indicate that inflammation and remodeling may involve airways of all sizes (5), and the inflammation and remodeling in asthma may even extend beyond the airways to affect also the pulmonary vasculature (6). In 1991, Saetta and colleagues (6), examining lungs from patients who died suddenly during an asthma attack, reported on the occurrence of inflammatory changes, including eosinophilia, in large pulmonary arteries adjacent to bronchi. Findings in lung tissues from patients with chronic obstructive pulmonary disease (7) may also support the possibility that in airway inflammatory diseases the pulmonary vasculature may be subjected to inflammation and remodeling. It has been suggested that interleukin-5 (8, 9) and eosinophil-derived proteins may have a role in remodeling of asthmatic bronchi, but the involved mechanisms as well as actual clinical consequences of bronchial remodeling have not as yet been fully established (10). Regarding pulmonary vessels in asthma and in animal models of the disease, almost nothing seems to be known even of basic aspects such as occurrence, inducement factors, and features of any remodeling. Mouse models of allergic airway inflammation, although they differ from human allergic asthma in several important aspects (11, 12), may exhibit tracheobronchial remodeling changes reminiscent of those occurring in the human disease (8, 13, 14). Also, both patients with asthma and allergic mice may exhibit prominent peribronchial as well as perivascular eosinophilia (10, 15). In this study, we have used a validated mouse model of allergic airway inflammation and compared bronchial remodeling features with those that potentially may occur in large pulmonary vessels. The results are intriguing. Virtually all structural changes occurring in allergen-exposed airways of allergic mice were found to be present also in the large pulmonary vessels of these animals. The results of this study have previously been reported in part in the form of abstracts (16, 17).
Model of Allergic Inflammation Female BALB/C mice were sensitized on Day 0 with ovalbumin (OVA; n = 10, 10 µg [grade III; Sigma, St. Louis, MO] + 1 mg AlOH3 intraperitoneally) or saline (n = 7), followed by daily 30 minutes of 1% OVA aerosol exposures on Days 1420 according to a validated protocol (18). At 24 hours after the last exposure, the animals were killed, and lung tissue specimens were processed for cryostat (eosinophil peroxidase and periodic acid-Schiff staining) and paraffin sectioning (immunohistochemistry and terminal deoxy RNase nick end labeling [TUNEL] staining) (18). For labeling of endothelial cells in situ, fluorescein isothiocyanatelabeled lectin (Table 1) was administered intravenously 3 minutes before killing (19). All protocols were approved by the local ethics committee (Malmo/Lund, Sweden, M16702).
Tissue Eosinophilia Development of an allergic airway inflammation was determined as lung tissue eosinophilia, (enzyme-histochemical staining for eosinophil peroxidase) (20).
Cell Turnover; Proliferation and Apoptosis
Remodeling Collagen synthesis was assessed by staining for procollagens using antibodies directed against the N-terminal of procollagen I and procollagen III. For each antibody, an optimized protocol was used (for details on primary and secondary antibodies, see Table 1).
Tracheal Blood Vessels
Quantification and Statistics
Characterization of the Allergic Inflammation After 7 days of saline treatment, 63 ± 31 eosinophils/mm2 tissue were detected in lung sections compared with 231 ± 42 eosinophils/mm2 in OVA-exposed animals (p < 0.05). The distribution of eosinophils was both peribronchial and perivascular (Figures 1A and 1B), and in OVA-exposed mice, a dense infiltration of eosinophils was present in the tissue between juxtapositioned bronchi and vessels. The overall cell turnover was detected as total number of proliferative (proliferating cell nuclear antigen positive) and apoptotic (TUNEL positive) cells/mm2 tissue. In animals exposed to OVA, the total numbers of proliferative and apoptotic cells were significantly increased compared with saline-treated control animals. The number of proliferating cells increased from 2.9 ± 0.6 cells/mm2 in control animals to 18.7 ± 2.8 in OVA-exposed animals (p < 0.05). The numbers of apoptotic cells were 12 ± 2.4 cells/mm2 in control animals and 26 ± 5.7 in OVA-exposed animals (p < 0.05).
The mean circumference of analyzed bronchi and vessels was 763 ± 40.5 µm and 438 ± 43 µm, respectively. No subdivision into size-dependent groups was made because of the uniformity in size.
Bronchial Remodeling
Bronchial smooth muscle remodeling. Sections from OVA-exposed lungs, stained for -SMA, displayed a significant increase in smooth muscle area (230 ± 56 µm2/µm BM; Figures 3A and 3B) compared with control animals (52 ± 11 µm2/µm BM, p < 0.05). There was also an increase in number of proliferating SMCs in OVA-exposed animals compared with control animals (Figure 2B). In OVA-exposed mice, the number of TUNEL-positive bronchial SMCs increased from 1.1 ± 0.28 cells/mm BM in saline control animals to 2.25 ± 0.45 in OVA-exposed animals (p < 0.05).
Remodeling of bronchial extracellular matrix. The pattern of procollagen expression differed between procollagen I and procollagen III. Procollagen I was foremost distributed within cells, showing a punctuate cellular staining, whereas procollagen III was distributed diffusely around the bronchi. The discrepancy in expression pattern prompted different modes of quantification. Both the number of procollagen Ipositive cells (Figure 4A) and the area of procollagen IIIpositive staining (Figure 4B) around bronchi increased significantly after OVA exposure compared with control animals.
Remodeling of tracheal microvessels. The mean inner diameter of tracheal capillaries increased from 7.5 ± 0.6 µm in saline-treated animals to 11.4 ± 0.3 µm in animals exposed to OVA (p < 0.05). The capillary mean length also increased significantly after OVA exposure from 330 ± 21 µm in saline-treated animals to 480 ± 21 µm (p < 0.05).
Pulmonary Vascular Remodeling
Vascular smooth muscle remodeling. The area of vascular SMCs (Figures 3C and 3D) increased significantly in OVA-exposed animals compared with control animals. Control animals displayed a mean -SMApositive area of 260 ± 32 µm2/µm BM compared with 1,090 ± 98 µm2/µm BM in OVA-exposed animals (p < 0.05). The number of proliferating vascular SMCs was significantly higher in OVA-exposed animals compared with saline control animals (Figure 5B). The number of TUNEL-positive SMCs did not differ between OVA-exposed and control animals.
Remodeling of vascular extracellular matrix.
This study demonstrates that week-long allergic airway inflammation in mice is associated with advanced tissue remodeling of tracheobronchial airways and, surprisingly, of large pulmonary vessels. The structural changes in both types of tissues include an increase in smooth muscle mass, increased proliferation of epithelial and endothelial lining cells, and extracellular remodeling. Hence, topical airway exposure to allergen may readily evoke inflammatory remodeling processes that extend beyond the bronchi causing structural changes of pulmonary vessels. In agreement with previous work in sensitized mice, this study demonstrated that significant remodeling of tracheobronchial airways may be produced after only a few days of OVA exposure (13, 14, 25). This brief time period makes it difficult to compare these findings with those occurring in human chronic airway diseases. In human subjects with asthma, most studies of bronchial remodeling have been performed many years after the onset of symptoms. This fact and the chronic nature of asthma have made it easy to overlook the more dynamic aspects of remodeling in this disease. Indeed, several findings suggest that also in asthma remodeling may develop early. Thus, bronchial remodeling can be detected already in childhood asthma (26). The dynamic nature of remodeling is further exemplified by findings that airway myofibroblasts increase in numbers within 24 hours after a single allergen challenge in individuals with asthma (27). Notably, the pulmonary vascular remodeling revealed in this study had a similar time course and magnitude as the bronchial remodeling. Thus, for both tissues, no significant changes were found after 3 days of OVA exposure (K. Rydell and colleagues, unpublished data), whereas significant remodeling was established 24 hours after 7 days of daily allergen exposures. Furthermore, also the nature of the remodeling in the two tissues was similar. Hence, apart from the airway-specific goblet cell hyperplasia and tracheal microcirculatory changes, all features of bronchial remodeling that were demonstrated in this study each had a corresponding remodeling change in pulmonary vessels, although the latter were not subjected to topical allergen challenges.
The increase in bronchial smooth muscle mass, as previously observed in human asthma and mouse models of allergic airway inflammation, has been suggested to result from hypertrophy of individual cells or hyperplasia (2, 23) or a combination of both. The increase in proliferating SMCs in this study indicates that hyperplasia has contributed to the present smooth muscle remodeling in both bronchi and vessels. Furthermore, considering the increased synthesis of procollagens after OVA exposure, it cannot be excluded that increased deposition of extracellular matrix (ECM) between SMCs may also have contributed to the present smooth muscle enlargement. However, detailed examination of procollagen staining within the smooth muscle area suggests that this effect is marginal. In previous studies of OVA-treated rats, it has been demonstrated that although the bronchial smooth muscle mass (i.e., volume) is increased, the amount of The multiple allergen exposures in this study induced a proliferation response in the bronchial epithelium as well as in the endothelium of pulmonary vessels. Increased proliferation of bronchial epithelial cells in mouse OVA models has been observed previously (31), although mechanisms explaining this phenomenon have not been presented. Foremost among acknowledged factors causing epithelial hyperproliferation are damage and repair processes (24). Similarly, vascular endothelial injury has been forwarded as a potent factor leading to endothelial proliferation (32). However, neither in the vascular endothelium nor in the bronchial epithelium could we detect any signs of ongoing damage and repair. The latter observation is in agreement with previous studies reporting a maintained epithelial integrity in OVA-exposed mouse airways (31). Hence, it is likely that the present proliferation response in vascular and bronchial epithelia is induced by more subtle processes than direct injury. Indeed, it has been suggested that the hyperproliferative response in OVA-exposed airways simply represents a response to increased levels of growth factors or other mitogenic mediators found in inflamed regions (31). It is possible that such bioactive proteins may be distributed some distance away from the allergen-induced airway mucosal processes. Involvement of such distribution mechanisms rather than any spread of injurious inflammatory processes would also agree with the present observation of increased endothelial cell proliferation occurring in adjacent pulmonary vessels where no endothelial damage was observed. Remodeling of the ECM in asthmatic airways includes several collagens. In this regard, immunohistochemical analysis has identified an increased occurrence of the collagen types I, III, and V (1). To study both localization and type of collagen changes in this study, we used specific antibodies, directed against procollagen I and procollagen III, detecting ongoing collagen synthesis rather than total content of collagen. Using this approach, we could show that the ECM changes include increased production of these procollagens. These data agree with earlier findings on occurrence of collagen I (13) and collagen III (8) in OVA-exposed mouse airways. In addition, we demonstrated that equally pronounced increases of these procollagens take place in large pulmonary vessels. The distribution of the increased collagen deposition was similar in bronchi and vessels and localized primarily outside the smooth muscle layer. Interestingly, Henderson and colleagues (33), who used the pan-collagen staining Masson's trichome to explore the role of cysteinyl leukotrienes in OVA-exposed mouse airways, found an increased collagen staining also in large pulmonary vessels. Hence, although no quantification, further analysis, or discussion of this observation was made (33), this previous observation in passing supports these findings of ECM remodeling in large pulmonary vessels. In small rodents, including mice, microcirculation carrying systemic blood is lacking in the intralobular airways but is present in the trachea, which makes this a relevant study object in search for allergen-induced changes in the systemic circulation (34). Here we could demonstrate significant inner enlargement and elongation of tracheal capillaries after OVA exposure. These novel data may reflect early remodeling but may in part also reflect allergen challenge-induced vascular engorgement. Similar alterations in mouse tracheal microvessels have previously been found in mice subjected to Mycoplasma pulmonis infection (19, 35). Because the method used to display tracheal vessels, in situ labeling with fluorescein isothiocyanatelabeled lectin, only visualizes the inner surface of vessels, we cannot tell whether additional structural changes occurred within the walls of the tracheal microvessels. This study demonstrates that during an allergic inflammation not only a cellular infiltration but also remodeling processes may spread from the site where the allergic response is originating (in this case the airway mucosa) to adjacent tissues. Spreading of remodeling processes between tissue compartments may be a general but little explained phenomenon in inflamed airways. For example, selective removal of epithelial cells (without damaging the BM or underlying tissue) from a small area of guinea pig trachea in vivo evokes an intense proliferation response that, apart from surrounding epithelial repair cells, also involves more remote tissue components including subepithelial SMCs and fibroblasts (24, 36). Currently, potential mechanisms behind the spreading phenomenon remain speculative. Saetta and colleagues (6), who observed eosinophilic inflammation in large pulmonary vessels in cases of fatal asthma, raised the possibility that pulmonary vascular inflammation may be induced by "spill over" of proinflammatory mediators from adjacent bronchi. In support of this notion, these authors found particularly pronounced eosinophilia in the vascular regions closest to bronchi. Also, in this study, a high density of eosinophils was observed in the tissue areas between bronchi and vessels (Figure 1B). This finding suggests the possibility of traffic of inflammatory cells between large pulmonary vessels and bronchi. The direction of this traffic is not obvious because inflammatory cells, including eosinophils, can enter the lung tissue through the large pulmonary vessels (37). It cannot be excluded that eosinophils are involved in the present vascular as well as bronchial remodeling changes. As with human airways (38), eosinophils present in OVA-exposed mouse lungs represent a major source of the potent profibrotic cytokine transforming growth factor-ß (8). Furthermore, both in mouse OVA models and human asthma antieosinophil treatment with neutralizing antiinterleukin-5 attenuates the airway remodeling. Eosinophils might also be involved in the effects on large pulmonary vessels, which display an equally extensive eosinophilia and a similar pattern of remodeling to that of the bronchi. Hopefully, future assessment of involved molecular mechanisms, including the generation of novel specific tools and drugs, can also shed light on the clinical importance of pulmonary vascular remodeling in asthma of which nothing is known at present. In conclusion, this study has demonstrated that allergic airway inflammation is associated with both tracheobronchial and pulmonary vascular remodeling. Thus, several well-established remodeling features of allergic airway inflammation, including increased smooth muscle mass, ECM deposition, and increased proliferation of lining cells, were present in both bronchi and large pulmonary vessels. The mechanisms behind this intriguing finding and to what degree the present pulmonary vascular remodeling may contribute to the pathophysiology of allergic airway inflammation warrant further studies.
The authors thank Saana Karttunen and Professor Juha Risteli, Oulu University, Finland, for generously providing the antibodies directed against the N-terminals of procollagen I and procollagen III. The authors also thank Karin Jansner for participation and technical assistance in the study.
Supported by Medical Faculty, Lund University, Sweden; the Swedish Medical Research Council; and the Heart and Lund Foundation, Sweden. Conflict of Interest Statement: K.R.T. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; L.U. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; C.G.A.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J.S.E. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form June 2, 2004; accepted in final form September 22, 2004
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||