INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: obliterative bronchiolitis; experimental lung transplantation; collagens I and III

Chronic rejection in the form of obliterative bronchiolitis (OB) is the major cause of long-term morbidity and mortality after lung transplantation (1). The exact pathogenesis of OB is unknown, but the respiratory epithelium is considered to be a potential allogeneic target for immunological effector mechanisms (2). Progressive damage and loss of airway epithelium are accompanied by increased production of cytokines and fibrogenic growth factors indicated as having a role in inducing the fibroproliferative events leading to OB (3). Histologically, OB is characterized by inflammation of the small airways and gradually progressing occlusion of the airway lumen by fibrous tissue, permanently obliterating the bronchioli (7).

The main constituents of lung extracellular matrix are collagens I and III (8). Fibroblasts are the major producers of type I and III collagens in the lung, but endothelial, epithelial, alveolar type II, and smooth muscle cells also synthesize these collagens (9). Both collagen types occur in bronchial mucosa and subintima (10). Changes in the distribution and increase in the production of interstitial collagens I and III exist in fibrotic pulmonary diseases (11). The presence of collagens I and III in subepithelial fibrotic lesions has been a consistent finding in immunohistochemical studies of the airways of patients with OB (12, 13). In transbronchial biopsies from lung transplant patients, increased amounts of type III collagen deposition appear in clinically manifest OB (14).

Investigations of the mechanisms of OB have used several experimental animal models with either ortho- or heterotopic transplantation of lung, bronchial, or tracheal structures (15- 17). Our group has developed a porcine heterotopic bronchial transplantation model exhibiting histological changes similar to those of human OB following lung transplantation (18). Total epithelial destruction and permanent luminal obliteration occur rapidly in allografts without immunosuppression (18). In lung allografts, prevention of the obliterative process is achieved (19) with combination therapy of cyclosporine A (CsA), methylprednisolone (MP), and RAD, that is, 40-O- (2-hydroxyethyl)-rapamycin (20).

Although collagen types I and III are found in OB lesions (12, 13), their transcriptional activity and the tissue distribution of cells expressing these procollagen messenger RNAs (mRNAs) have not been characterized. For further understanding of the deposition of these extracellular matrix proteins in the obliterative process, we investigated the changes in the temporal and spatial expression of procollagen I and III mRNA in our pig heterotopic bronchial allograft model. We also compared the activation of these procollagen genes during the rapid and prevented obliteration.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Experimental Model

Nonrelated domestic pigs served as donors and recipients. The animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals, NIH Publication No. 86-23, revised 1985. Heterotopic bronchial transplantations were performed as described (18). Briefly, for anesthesia, intramuscular ketamine sulfate (10-15 mg/kg), azaperone (10-15 mg/kg), atropine sulfate (0.05 mg/kg), and intravenous sodium pentobarbital (6-12 mg/kg), diazepam (0.25 mg/kg), and pancuronium bromide (2-4 mg), plus inhaled enflurane were used. In the operative procedure, bronchial implants were transplanted subcutaneously into the ventral side of the recipient. Three groups, with four pigs in each, were formed: autografts, nontreated allografts, and allografts treated with a daily oral dose of CsA 10 mg/kg, RAD 1.5 mg/kg, and MP 20 mg to prevent obliteration (19). The grafts were removed on postoperative Days 3, 7, 10, 14, 21, 30, and 60. Postoperative pain was controlled with intramuscular diclophenic acid (37.5 mg). At the end of the follow-up, the animals were euthanized with intravenous sodium pentobarbital.

Histological Evaluation

Epithelial destruction, luminal obliteration, fibrosis (defined as a pathological increase of connective tissue composed of fibroblasts and extracellular matrix) in the bronchial wall (considered as the area beneath the epithelium to the cartilage), total fibrosis (bronchial wall fibrosis and fibrosis in the pericartilaginous area surrounding the cartilaginous structures), bronchial wall inflammation (defined as the numbers of infiltrating inflammatory cells in the tissue), and cartilage destruction and new cartilage formation were graded on a semiquantitative scale of 0 to 3 in hematoxylin-eosin-stained sections. In fibrotic areas, the relation between cellular and extracellular matrix components was scored as equal, < 1, or > 1.

Determination of Total Tissue Collagen

The hydroxyproline content of total tissue samples was analyzed by high-pressure liquid chromatography. Total collagen content was estimated, assuming that hydroxyproline comprises 13.7% collagen by weight (21).

Construction of cDNA Clones for Porcine pro1(I) Collagen and pro1(III) Collagen mRNAs

For detection of porcine-specific mRNAs, complementary DNA (cDNA) clones were constructed for porcine pro(I) and pro1(III) collagen mRNAs. Briefly, total RNA was extracted from experimental granulation tissue by the method of Chomczynski and Sacchi (22), and synthesized into cDNA by reverse transcription. Aliquots of cDNA were amplified by polymerase chain reaction with primers based on existing human, mouse, and rat sequences (23).

Northern Hybridization

Total RNA was extracted from the bronchial samples by an established method (22). Northern hybridization was performed as previously described (28) by use of constructed cDNA clones as hybridization probes. The bound probe was detected by autoradiography, and the relative intensity of the bands analyzed by a computer-linked densitometer. Results were corrected for minor variations in the amount of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA in the respective samples.

In Situ Hybridization

Digoxigenin-labeled antisense and sense cRNA probes were created from the above-described cDNA clones. Paraffin-embedded bronchial samples were cut into sections and used for in situ hybridization performed as described previously (28). Staining seen only with the antisense probe was considered positive. The distribution and number of positive cells for pro1(I) and pro1(III) collagen mRNA were analyzed separately in the obliterative plug, in the bronchial wall, and in the pericartilaginous area. Positive cells were counted in five randomly chosen microscopic fields from each area at ×100 objective magnification. The number of positive cells was related to the area of negative, pale staining microscopic field.

Statistical Analysis

All data are expressed as mean + SEM. Variation between the groups was analyzed with the nonparametric Kruskal-Wallis one-way analysis by ranks. The rank sums were then used for Dunn's test at a significance level of 5%. Values of p < 0.05 were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Histopathology

The histopathological findings are shown in detail in Table E1 in the online data supplement.

Epithelium. In nontreated allografts, the respiratory epithelium was completely destroyed by Day 10. In autografts and in immunosuppressed allografts, after initial ischemic damage, normal ciliated epithelium was maintained throughout follow-up. Significance (p < 0.05) in epithelial destruction was reached between nontreated and immunosuppressed allografts at early assessment points and between autografts and nontreated allografts at later points.

Obliteration. No luminal obliteration was evident in autografts (Figure 1A) and in allografts receiving immunosuppressive therapy (Figure 1C). In nontreated allografts, fibroproliferative tissue protruding into the bronchial lumen was first seen on Day 7. Luminal obliteration was complete by Day 21 (Figure 1B) (p < 0.05 when compared with autografts and immunosuppressed allografts); whereafter, in the luminal fibrous tissue, the extracellular matrix component gradually increased. In the course of the follow-up, the degree of lymphocytic inflammation in the obliterative plug decreased to nonexistent levels.

View larger version (131K): [in this window] [in a new window]   View larger version (146K): [in this window] [in a new window]   View larger version (139K): [in this window] [in a new window]   Figure 1.   Photomicrographs of an autograft, a nontreated allograft, and an allograft immunosuppressed with RAD, CsA, and MP on Day 21 (HE staining; original magnification: ×10). (A) Normal epithelium and bronchial structures in autograft. (B ) Nontreated allograft exhibiting total luminal obliteration by fibroproliferative tissue (arrows), bronchial wall inflammation, and fibrosis (arrowhead ) and destruction of cartilaginous structures (double arrows). (C ) Open bronchial lumen with well preserved epithelium and bronchial wall structures in immunosuppressed allograft.

Bronchial wall. Bronchial wall fibrosis in autografts and in nontreated allografts consisted of fibroblasts and extracellular matrix components in equal proportions. In autografts, fibrosis remained generally mild. In nontreated allografts, fibrosis increased from mild to moderate by Day 10, persisting thereafter. In immunosuppressed allografts, fibrosis remained mild, and on Days 14, 21, and 30 only sparse areas of fibrosis (p < 0.05 when compared with nontreated allografts), consisting mainly of cellular components were present in the bronchial wall. In all groups, a highly fibrous capsule had formed around the implants, contributing to total fibrosis.

The bronchial wall in autografts showed only slight lymphocytic inflammation. In nontreated allografts, by Day 7, infiltration of inflammatory cells was moderate, and remained moderate to severe, with a significant (p < 0.05) difference from that of autografts. In immunosuppressed allografts, inflammation was milder than in nontreated allografts.

Cartilaginous structures. In autografts and in immunosuppressed allografts, bronchial cartilage remained viable. In nontreated allografts, nearly all cartilaginous structures were destroyed by Day 14, with a significant (p < 0.05) difference from autografts on Day 14 and thereafter. Formation of new cartilage was apparent in autografts and in immunosuppressed allografts.

Total Collagen Content

Data on total bronchial collagen content are presented in Table 1. In nontreated allografts, total collagen content decreased below the level of total collagen in native bronchial tissue. An increased total collagen content was evident in autografts and in immunosuppressed allografts at all assessment points. The difference between autografts and nontreated allografts reached significance (p < 0.05) on Days 7, 21, and 60, with total collagen content higher in autografts.

Quantification of Procollagen I and III mRNA Expression

Clone pPCol1a1-1 detected two mRNAs of approximately 5.0 and 5.9 kb, and clone pPCol3a1-1 one mRNA of approximately 5.1 kb in size in Northern hybridization. Under the stringent washing conditions used, no cross-hybridization to pro1(I) and pro1(III) collagen mRNAs was observed (Figure 2). The transcriptional activity of procollagen I increased in all groups by Day 7 (Figure 3A). On Days 14 and 21, type I collagen mRNA expression was induced in immunosuppressed allografts with no obliteration. Quantification of procollagen III mRNA showed no significant differences between groups (Figure 3B). In immunosuppressed allografts, induction of type III mRNA expression was seen on the same days as of type I collagen mRNA.

View larger version (40K): [in this window] [in a new window]   Figure 2.   Northern hybridization of total RNAs extracted from bronchial autografts, nontreated allografts, and allografts immunosuppressed with RAD, CsA, and MP for pro1(I) collagen, pro1(III) collagen, and GAPDH mRNAs on Days 7 and 21.

View larger version (30K): [in this window] [in a new window]   View larger version (24K): [in this window] [in a new window]   Figure 3.   Summary of Northern analyses of bronchial autografts, nontreated allografts, and allografts immunosuppressed with RAD, CsA, and MP (n = 4 in each group) for mRNA levels of pro1(I) collagen (A) and pro1(III) collagen (B). The mRNA levels expressed as densitometric units (mean + SEM) corrected for GAPDH mRNA levels in the same samples. *No data.

Localization of Procollagen I and III mRNA Expression

In situ hybridization for procollagen I and III mRNA revealed a positive signal with antisense probes but not with sense probes. In native tissue, only sparse positivity for procollagen III was visible in blood vessels in the bronchial wall. Almost no positivity for procollagen I mRNA was detectable. In auto- and allografts, the expression of type I and III mRNA was detected nearly exclusively in fibroblast-like cells and in occasional smooth muscle cells.

Obliterative plug. In nontreated allografts obliterating within 21 d, a positive label for procollagen I and III mRNA appeared in fibroblast-like cells in the connective tissue plugs invading the bronchial lumen. In early obliterative lesions on Day 7, cells positive for procollagen I and III mRNA existed in equal numbers (Figure 4). On Day 14, the majority of the positive cells were expressing type III mRNA, which was the case also at subsequent assessment points (Figures 4, 5C, and 5E).

View larger version (15K): [in this window] [in a new window]   Figure 4.   Number of positive cells per microscopic field for type I and III procollagen mRNA detected by in situ hybridization. Nontreated allografts show a preponderance of and increase in fibroblast-like cells expressing type III mRNA in the obliterative plug and in the bronchial wall. In autografts and in immunosuppressed allografts, no similar trend occurs between the numbers of type III and type I procollagen mRNA-positive cells in the bronchial wall. Cells were counted at ×100 objective magnification in five microscopic fields. All data expressed as mean + SEM, n = 4 in each group.

View larger version (136K): [in this window] [in a new window]   View larger version (131K): [in this window] [in a new window]   View larger version (155K): [in this window] [in a new window]   View larger version (149K): [in this window] [in a new window]   View larger version (159K): [in this window] [in a new window]   View larger version (152K): [in this window] [in a new window]   View larger version (143K): [in this window] [in a new window]   View larger version (138K): [in this window] [in a new window]   Figure 5.   Photomicrographs of expression of type I and type III collagen mRNA detected by in situ hybridization on Day 21 in autografts, nontreated allografts, and allografts immunosuppressed with RAD, CsA, and MP (sections counterstained with Mayer's hemalum). (A) A few positive fibroblast-like cells for type I collagen mRNA in the bronchial wall and pericartilaginous area, and (B) no positivity for type III collagen mRNA in autografts (original magnification: ×40). Only a few positive cells for type I collagen mRNA in the obliterative plug (C ) and in the bronchial wall (D) in nontreated allografts when compared with the increased number of cells expressing type III collagen mRNA in the respective areas (E ) and (F  ) (original magnification: ×20). No marked differences between expression of type I (G) and type III (H ) collagen mRNA existed in the bronchial wall in immunosuppressed allografts (original magnification: ×40).

Bronchial wall. In autografts, the number of cells expressing type I and III mRNA transcripts decreased by Day 14 and remained low thereafter (Figures 4, 5A, and 5B). In nontreated allografts, only a few cells positive for type I and III mRNA appeared on Days 3, 7, and 10 (Figure 4). Fibroblast-like cells expressing procollagen III mRNA started to increase on Day 14 and were significantly (p < 0.05) augmented on Days 21 and 30 compared with numbers of autografts. A slight decrease in the number of positive cells followed thereafter. In nontreated allografts with rapid obliteration, dominance of fibroblast-like cells showing a positive signal for type III collagen mRNA was evident at all assessment points, being most prominent from Day 14 onward (Figure 5D and 5F). In immunosuppressed allografts, no similar increase appeared in the number of procollagen III mRNA-positive cells compared with procollagen I mRNA-positive cells (Figures 4, 5G, and 5H). The pattern of fibroblast-like cell numbers expressing type I and III collagen mRNAs resembled that of autografts.

Pericartilaginous area. The number of positive cells for procollagen I and III mRNA in pericartilaginous areas is expressed in Table 2 as a percentage of total cell count. In all groups, the majority of positive cells were detected in this area at nearly all assessment points. Comparison of absolute cell numbers in autografts and in immunosuppressed allografts gave a preponderance of procollagen I mRNA to type III mRNA expression in fibroblast-like cells, whereas nontreated allografts showed the opposite finding.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Development of irreversible fibrotic changes in OB is a complex process. To date, only a few studies have explored extracellular matrix deposition in OB, and none has focused on collagen gene expression in this disorder. In our large animal model of OB, pathological changes in walls and lumina of small bronchi are presumably triggered by epithelial injury (19). Our model provides an opportunity to study chronic rejection in repeated adequate-size samples in tissue areas where OB is known to occur in human lung transplant recipients (7).

In studies of human OB, extracellular matrix deposition has been analyzed in endobronchial biopsy specimens (14), which represent only a small fraction of bronchial wall or in postmortem samples (12, 13), in which no development of the process can be followed. In this study, we assessed the cells expressing procollagen I and III genes during the development of OB in bronchial samples consisting of the full circumference of the bronchus. We demonstrated that development of OB was associated with an increase in fibroblast-like cells expressing procollagen III mRNA. The predominance of cells expressing procollagen III mRNA increased parallel to progressing luminal obliteration, together with a histologically identified increment in the amount of fibroblasts and extracellular matrix in the bronchial wall.

In previous studies of bleomycin-induced pulmonary fibrosis, the cells in fibrotic lesions expressing procollagen I and III mRNA were primarily derived from cells resembling fibroblasts and were located in the tissue underlying the airway epithelium (29). Increases in the numbers of cells expressing procollagen I and III genes and enhanced mRNA expression by individual cells were both suggested to contribute to the development of pulmonary fibrosis (30). However, these studies did not compare the expression of the procollagen genes. In fibrotic processes, an increase in type III collagen synthesis is suggested to occur in proliferating fibroblasts as a response to various stimuli including inflammation (9, 10).

Increased amounts of aminoterminal propeptide of type III procollagen have been detected in bronchoalveolar lavage fluid of patients with fibrosing alveolitis (33), adult respiratory distress syndrome (34), and pulmonary edema (35), predicting poor prognosis. In OB following lung transplantation, an increased deposition of type III collagen in the bronchial submucosa correlates with poor lung function (14). In silica-induced pulmonary fibrosis, an increased ratio of type III:type I collagen has also occurred (36). Similarly, in our study, the dominance of type III collagen expression in fibroblast-like cells in the obliterative plug and in the bronchial wall of rapidly obliterating allografts supports the hypothesis of a positive association between increased synthesis and deposition of procollagen III and severity of the disorder.

In our autografts, minor fibrotic changes with fibroblast-like cells expressing procollagen I and III mRNA in the bronchial wall indicate that the nonimmunological injury caused by our heterotopic method activates procollagen gene expression. In immunosuppressed allografts in which obliteration was prevented, histological findings and cell numbers producing type I and III procollagen mRNA resembled those of autografts. Although in both groups fibroblast-like cells expressing I and III procollagen mRNAs were detectable, only insignificant pathological alterations were evident. In the bronchial wall, an important finding in autografts and immunosuppressed allografts was the unaltered number of cells expressing procollagen III mRNA.

The quantitation of total collagen and procollagen I and III mRNA did not directly correlate with the increase in positive cells detected by in situ hybridization. The changes in the collagen type II derived from cartilage probably affected more the total collagen content than the relatively small fibrous areas in the bronchial wall or lumina. Quantification of mRNAs by Northern hybridization does not identify the tissue origin of mRNA. At some assessment points, cells positive for procollagen I and III mRNA were abundant in the pericartilaginous area, even in the total absence of positivity in the bronchial wall. This finding may explain the high expression of procollagen genes in immunosuppressed allografts at some assessment points when only a few positive cells were evident in the bronchial wall.

In conclusion, mechanisms leading to extracellular matrix deposition in OB are still in part unknown. This study shows for the first time that expression of procollagen I and III mRNA in fibroblast-like cells is found in OB lesions, along with a preponderance of and increase in cells expressing type III mRNA. In autografts and in allografts with immunosuppression adequate to prevent obliterative airway disease, only a minor degree of procollagen gene activation occurred, and no changes in the genetic activity between type I and III collagen similar to those in rapidly obliterating allografts. This observation indicates that one of the mechanisms in the fibrotic process of OB is an increase in fibroblast-like cells expressing procollagen III mRNA.