INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: lung transplantation; gene therapy; fibrosis; chronic dysfunction

Lung transplantation is an established therapeutic modality for patients with end-stage lung disease. The early postoperative survival is excellent, with operative mortality in the range of 10%. However, more than 50% of patients who survive more than 1 yr after transplantation will develop fibrous obliteration of the small airways, or bronchiolitis obliterans (BO) (1), which is characterized by progressive shortness of breath and lung dysfunction that relentlessly progresses to death. This is the major factor that currently limits the long-term success of human lung transplantation.

BO is manifested as progressive fibroproliferation of small airways and thought to be a consequence of pathologic repair processes after graft injury. Transforming growth factor- (TGF), a potent immunosuppressive cytokine, promotes fibrosis by enhancing the production and deposition of extracellular matrix components. Several studies have demonstrated overexpression of TGF mRNA or increased production of TGF in the lungs of patients with chronic rejection after transplantation. For example, the gene expression of TGF in alveolar cells demonstrated marked peaks that preceded the diagnosis of rejection by several months (2). El-Gamel and colleagues (3) found that TGF₁ was heavily expressed in lung sections with fibrosis, and this expression correlated positively with the grade of fibrosis in lung transplants. TGF expression was greater in patients with BO in comparison to patients without BO. Positive staining of TGF preceded the histologic confirmation of BO by 6 to 18 mo (4). Similar observations have also been reported from other organ transplantation. For example, overexpression of TGF isoforms was found in patients with chronic rejection after kidney transplantation (5); TGF₁-expressing macrophages were found in fibrotic tissues from chronically rejected human liver allografts (6), and TGF staining in cardiac allografts was higher in patients with more severe graft vasculopathy (7). These results imply that TGF plays an important role in the process of organ rejection by mediating the fibrotic process. It has been suggested that strategies to inhibit the actions of TGF might improve the function and survival of lung (3, 4), cardiac (7), and other allografts.

TGF signals through a heteromeric complex of protein kinase receptors (types I and II receptors) that has a limited ability to bind ligand, which can be overcome by the action of betaglycan (TGF type III receptor), a membrane-anchored TGF-binding protein (8). Membrane betaglycan presents TGF directly to the type II signaling receptor, a transmembrane serine/threonine kinase, forming a high affinity ternary complex. Therefore, it enhances cell responsiveness to TGF, and eliminates marked biologic differences between TGF isoforms (8). The extracellular region of betaglycan can be shed by cells. It has been shown that recombinant soluble betaglycan acts as a potent inhibitor of TGF binding to membrane receptors, and thus it blocks TGF action (9). In the present study, gene encoding the extracellular domain of betaglycan was constructed into an adenoviral-mediated transfection vector, to test whether soluble TGF type III receptor (TGFBIIIR) functioning as a TGF antagonist could competitively inhibit the function of TGF in mediating allograft-induced fibrous airway obliteration.

A heterotopic tracheal transplant murine model developed by Hertz and colleagues (10) has been used to study the role of platelet-derived growth factor and basic fibroblast growth factor in the pathogenesis of BO (11, 12). Similarly, we have developed a rat heterotopic tracheal transplant model in which tracheal allografts develop a peak of lymphocytic infiltration on Day 7, followed by lumenal fibrous obliteration (13). Using this model we have demonstrated upregulation of mRNAs of Th1 cytokines and chemokines in allografts (14). Using adenovirus-mediated gene transfection of interleukin (IL)-10, or recombinant IL-10, we were able to prevent the development of fibrous airway obliteration in the tracheal allografts (15). We have also shown that an antibody against RANTES, a C-C chemokine that downregulates the presence of CD4⁺ cells and their function, also inhibits airway obliteration (16). Furthermore, the angiotensin system also plays a prominent role in the pathogenesis of fibrosis by activating TGF expression (17). Using an angiotensin-converting enzyme inhibitor, we were able to inhibit allograft transplant-induced fibrous airway obliteration (18). Therefore, in the present study, this model system was used to further investigate the role of TGF in the pathogenesis of BO, by targeting the action of TGF directly using adenoviral-mediated soluble TGFBIIIR gene transfer.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and Tracheal Transplant Operation

Male Brown-Norway and Lewis rats weighing 200 to 300 g were purchased from Harlan Sprague Dawley Inc. (Indianapolis, IN) and Charles River Canada Inc. (St. Constant, Quebec, Canada), respectively. Animal care was provided according to NIH guidelines and approved by the Toronto General Hospital Research Institute Animal Care Committee.

Heterotopic tracheal transplantation was carried out as previously described (13, 15, 16, 18). Briefly, entire trachea of Brown-Norway rat was excised, divided into two equal-sized segments, and then placed into a subcutaneous pouch made in the back of the recipient (Lewis rats). Grafts were removed on the designated days. The middle third of the tracheal segment was fixed with 10% buffered formalin for histologic examination and immunohistochemistry studies.

TGF Immunohistochemistry

Frozen graft specimens were collected on Days 2, 7, 14, and 21 after transplantation, and processed as described previously (16). Briefly, 5-µm sections were placed on poly-L-lysin-coated slides, air-dried, and fixed with acetone. After blocking with Protein Block Serum-Free solution, the sections were incubated with polyclonal rabbit anti-TGF IgG (Santa Cruz Biotechnology, Santa Cruz, CA) at 1:100 for 30 min. The secondary antibody and alkaline phosphatase conjugation steps and color reaction were performed as previously described (16). Negative control rats were incubated with PBS containing 0.1 % bovine serum albumin without the primary antibody, or with isotype-specific rabbit IgG.

Generation of Recombinant Adenovirus Ad2 TGFBIIIR

The TGFBIIIR gene was amplified from clone #7411 (19) using primers: 5'TGFB-3R-II GTAGAGCTCCACCATGACTTCCCATTAT GTGATTGCCAT and TGFBIII-3'GTGTCTAGACTAGTCCAGACC ATGGAAAATTGGTGG with Vent DNA polymerase (New England Biolabs, Beverly, MA). PCR products were fractionated on a 1% agarose gel, and a 2.2-kb product was purified, digested with Ecll36II-Xba I, and cloned into the Ecor V-Xba I sites of the pAdQUICK shuttle vector pSV2-ICEU I. Recombinant adenovirus was generated as previously described (Figure 1) (20). The transgene expression was confirmed by testing supernatants of 293 cells (ATCC, Rockville, MD) infected with Ad2TGFBIIIR with Western blotting, probed with goat antihuman TGFBIIIR antibody (R&D Systems, Minneapolis, MN).

View larger version (10K): [in this window] [in a new window]   Figure 1.   Graphic representation of recombinant adenovirus vector Ad2 TGFBIIIR E3 2.9. The expression cassette for human TGFBIIIR is inserted in place of the E1 region at the ICEU I site in the adenovirus expression vector pAdQUICK. This vector has wild-type E2 and E4 regions.

Experimental Groups for Gene Transfection Studies

In the first series of experiments, adenovirus-mediated soluble TGFBIIIR gene vector (5 × 10⁹ particles) was administered topically at the site of allograft at Days 0, 5, and 10 postoperatively. An untreated control group was included for comparison. In the second series of experiments, adenovirus containing soluble TGFBIIIR gene or empty vector (5 × 10⁹ particles) was administered either topically or by intramuscular injection on Day 5. Five study groups were designed: untreated control (CO), topical soluble TGFBIIIR gene (TG), topical empty vector (TV), intramuscular soluble TGFBIIIR gene (IMG), and intramuscular empty vector (IMV). All grafts were removed on Day 21 for histologic examination.

Morphometric Analysis

The formalin-preserved middle portion of the tracheal segment was cut to 4-µm sections for hematoxylin-eosin staining. Computerized morphometry (21) was performed in a blinded fashion as previously described (16, 18).

Statistical Analysis

Data are expressed as mean ± standard deviation of the means. Kruskal-Wallis one-way analysis of variance on ranks was used to analyze the differences between groups, because normality test was failed. All pairwise multiple comparisons were performed using the Student-Newman-Keuls test (16, 18). Data are considered statistically significant if p values are less than 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Expression of TGF in Rat Tracheal Allografts

The development of transplant-induced fibrous airway obliteration in the rat tracheal transplant model of BO has been previously characterized in detail (13). The fibrous airway obliteration that develops in allografts demonstrates a triphasic time course: an initial ischemic phase, followed by a marked cellular infiltrate phase with complete epithelial loss, and finally a fibrous obliterative phase of the allograft airway lumen (13). We have recently demonstrated that many of the infiltrating cells are CD4⁺ mononuclear cells (16). Using a murine tracheal model of BO, Neuringer and coworkers (22) have found other lymphocytes and macrophages in the allotracheas at early time points after transplantation. In the present study, the expression and distribution of TGF protein in allografted tracheal tissue was examined by immunohistochemistry staining at these three phases. As shown in Figure 2, the number of infiltrating mononuclear cells increased from Day 2 to Day 7, and these cells stained strongly with anti-TGF antibody (Figures 2A and 2B). At Day 14, the airway lumen was filled with fibrotic tissue, and few TGF-positive cells could be found (Figure 2C). At Day 21, no TGF-positive staining cells were found, but the fibrotic tissue was still positively stained (Figure 2D).

View larger version (163K): [in this window] [in a new window]   Figure 2.   Immunostaining of TGF in transplanted rat airway allografts. Immunohistochemical staining for TGF revealed positively stained infiltrating mononuclear cells (small arrows) on Day 2 (A) and Day 7 (B), and staining in the fibrous tissue (large arrows) (A- C) in the allograft (magnification: ×400). When the allograft sections were incubated without the specific anti-TGF antibody, or with isotype-specific IgG, as negative controls, no positive staining was found (data not shown).

Influence of Gene Transfer Time on Antifibrotic Effect

We have previously shown that delivery of an anti-inflammatory and immunosuppressive cytokine, IL-10, locally through an osmotic mini-pump, inhibited the fibrous airway obliteration in this rat tracheal transplant model. However, this protective effect was seen only when the delivery of recombinant IL-10 was started at Day 5, but not at Day 0, of allograft transplantation (15). The angiotensin system plays an important role in the pathogenesis of fibrotic diseases. In another of our studies, when allograft rats were treated with captopril, an angiotensin converting enzyme inhibitor, the inhibitory effect on airway obliteration was observed when the drug delivery was started 5 d before transplantation, or on postoperative Day 1, but not if started on postoperative Day 5 (18). Therefore, to test the effect of adenoviral-mediated gene transfer of soluble TGFBIIIR on fibrous airway obliteration, adenovirus (5 × 10⁹ particles) was initially administered by topical injection at three different time points (Day 0, Day 5, or Day 10 after allograft transplantation). Airway obliteration was inhibited when the adenovirus was injected on Day 5 postoperation (Figure 3). The inhibitory effect did not reach statistical significance (p = 0.06), which is likely due to the small sample sizes.

View larger version (40K): [in this window] [in a new window]   Figure 3.   Time-dependent inhibitory effect of adenoviral-mediated soluble TGF RIII gene transfer on the development of lumenal obliteration in allografts. Adenovirus containing soluble TGFBIIIR gene was injected at the site of transplants on Day 0 (D0), Day 5 (D5), or Day 10 (D10). Lumenal obliteration of allografts at Day 21 postoperation was quantified by computerized morphometric analysis. Gene administered on Day 5 inhibited the development of fibrous airway obliteration. p = 0.06 versus the control group. Each point represents the mean ± SD of four samples. CO = untreated control.

Influence of Gene Transfer Route on Antifibrotic Effect

To determine the specificity of the adenoviral-mediated soluble TGFBIIIR on the prevention of fibrous airway obliteration, adenovirus containing an empty vector was used as a negative control. We have previously shown that adenoviral-mediated IL-10 gene delivery through intramuscular injection effectively inhibited fibrous airway obliteration (15). To determine whether intramuscular injection of soluble TGFBIIIR gene could have similar inhibitory effects on airway obliteration, both empty vector and vector containing soluble TGFBIIIR gene were injected at Day 5 after allograft transplantation either topically, at the site of tracheal transplantation, or intramuscularly. Interestingly, only topical gene transfection of soluble TGFBIIIR preserved lumenal patency on Day 21, whereas all other groups showed almost complete fibrous lumenal obliteration (Figure 4). On examination of tracheal structure, however, although topical gene transfer prevented fibrous obliteration in the airway lumen, loss of the entire epithelial lining (which was the same in all groups) was not prevented by the gene transfer. In addition, minor degrees of fibroproliferation were observed in the subepithelial space, replacing the normal architecture in that location (Figure 5).

View larger version (34K): [in this window] [in a new window]   Figure 4.   Gene-specific and local administration of adenoviral-mediated soluble TGFBIIIR gene transfer prevents the development of lumenal obliteration in allografts. Adenoviruses containing soluble TGFBIIIR gene, or an empty vector, were injected at the site of transplants or intramuscularly on Day 5. Lumenal obliteration of allografts at Day 21 postoperation was quantified by computerized morphometric analysis. Each point represents the mean ± SD of 7 samples. *p < 0.05 versus other groups. CO = untreated control; TG = topical gene; TV = topical empty vector; IMG = intramuscular gene; IMV = intramuscular empty vector.

View larger version (145K): [in this window] [in a new window]   Figure 5.   Histology of effects of adenoviral-mediated soluble TGFBIIIR gene transfer on lumenal obliteration in allografts. Adenoviruses containing soluble TGFBIIIR gene, or an empty vector, were injected at the site of transplants or intramuscularly on Day 5. Representative histology sections of allografts on Day 21 are presented. CO = untreated control; TG = topical gene; TV = topical empty vector; IMG = intramuscular gene. Topical administration of soluble TGFBIIIR gene preserved lumenal patency.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Soluble TGFBIIIR Gene Transfer Inhibits Allograft-induced Fibrous Airway Obliteration

In the present study, we have demonstrated that adenoviral-mediated topical gene transfer of soluble TGFBIIIR significantly inhibits the development of allograft-induced fibrous airway obliteration in a rat tracheal transplant model of BO. This observation supports the hypothesis that clinically observed overexpression of TGF plays an important role in the development of airway fibrosis that further contributes to the chronic dysfunction of the grafted lung after transplantation. It also provides the first direct evidence that the therapeutic approaches designed to block the effect of TGF may be useful to inhibit chronic graft fibrosis. The strong anti-TGF staining of infiltrating mononuclear cells suggest that these cells could be one of the major sources of TGF. The staining in the fibrous tissue indicates that secreted TGF binds to matrix proteins.

The observation that the inhibitory effect of adenoviral-mediated delivery of soluble TGFBIIIR on fibrous airway obliteration depends on the time and location of gene delivery is very interesting and important. As we described previously in this model, at the end of the first week after graft transplantation, there is a peak of lymphocytic infiltration (13). The increased number of TGF-positive infiltrating cells seen at Day 7 in the present study suggests that this could also be the peak of TGF, which is produced from these cells and could be responsible for the subsequent development of fibroproliferation and fibrous deposition in the airway lumen. We have previously noted that adenoviral- mediated gene transfer takes 24 to 48 h to reach the peak of transgene expression (15, 23, 24). Therefore, adenovirus vector delivered on Day 5 may have a peak of transgene expression around Day 7, which coincides with the peak of the infiltrating cells, leading to effective inhibition of graft fibrosis induced by TGF.

In contrast to our previous work on adenoviral-mediated IL-10 gene transfer (15), in the present study we found that only topical injection of adenoviral vector containing soluble TGFBIIIR is effective in preventing lumenal obliteration. This suggests that although both soluble TGFBIIIR and IL-10 can inhibit allograft-induced fibrous obliteration, soluble TGFBIIIR functions locally to competitively block the effect of TGF on its target cells. IL-10, as an immunosuppressive cytokine, may be released from the location of gene transfer to function at remote organs; it may also function through circulatory lymphocytes to affect the entire immune response towards the allograft. The topical effect of soluble TGFBIIIR could be a specific advantage for clinical purposes. After lung transplantation, this protein or its gene could be delivered locally through the trachea to prevent the development of BO in the airway while minimizing its impact systemically. Thus, the amount of the agent required could be less and the potential for systemic side effects could be reduced.

Inhibition versus Overexpression of TGF after Lung Transplantation

In the literature there are reports of beneficial effects of overexpression of TGF in organ transplantation. For example, in a murine heterotopic cardiac transplant model, TGF gene transfer prolonged allograft survival with inhibition of donor-specific cytotoxic T cell and IL-2 producing helper T cell functions in graft-infiltrating cells (25). Endobronchial administration of naked plasmid DNA encoding TGF₁ was shown to reduce early lung allograft rejection in a rat model (26). This implies that TGF downregulates the cell-mediated immune response during acute rejection. It is well known that the TGF family of proteins is a set of pleiotropic-secreted signaling molecules with unique and potent immunoregulatory properties (27, 28). However, the application of TGF as a potential molecular therapy should be considered with caution. Isaka and colleagues (29) have shown that the introduction of TGF gene alone into the kidney induced glomerulosclerosis by affecting extracellular matrix accumulation. Using replication-deficient adenovirus vectors to transfer the cDNA of TGF₁ to rat lung, Sime and colleagues (30) have demonstrated that transient overexpression of active TGF₁ results in prolonged and severe interstitial and pleural fibrosis, which is characterized by extensive deposition of the extracellular matrix proteins: collagen, fibronectin, and elastin, and by emergence of cells with the myofibroblast phenotype. Therefore, although early overexpression of TGF in transplanted organs may ameliorate acute rejection and inflammationand thus may be beneficialuncontrolled prolonged expression of TGF in the targeted organs may be detrimental.

Potential Application of Adenoviral-mediated Anti-TGF Gene Therapy

In addition to the adenoviral-mediated gene transfer of soluble TGFBIIIR used in the present study, several related adenoviral-mediated vectors have also been described in the literature, which could be potentially used as TGF inhibitors. For example, a chimeric cDNA encoding an extracellular domain of the TGF type II receptor fused to the IgG Fc domain inhibited the action of TGF and suppressed extracellular matrix accumulation in a rat model of experimental glomerulonephritis (31). Adenoviral gene transfer vector of a dominant-negative mutant of TGF type II receptor (32), and an adenovirus-gene transfer vector of decorin (33), which is another TGF-binding proteoglycan, have also been developed. With these tools, the role of TGF in allograft-induced fibrous airway obliteration and other types of chronic organ dysfunction can be further investigated.

A major advantage of adenoviral-mediated gene transfer is its higher efficiency when compared with other gene-delivering modalities. However, in vivo gene transfer using adenoviral vectors as a therapeutic modality has been limited by the host immune response that induces inflammation, limits the amount and duration of transgene expression, and prevents effective retransfection. All transplant patients are routinely administered immunosuppressive therapy. We have recently shown that transplantation immunosuppression attenuates the post-transfection host immune response to adenoviral-mediated gene transfection and thereby increases and prolongs transgene expression (24). This approach also makes effective re-transfection of adenoviral vectors possible (23). Thus, clinical application of gene therapy in the setting of transplantation is certainly feasible. Although soluble TGFIIIR did not completely prevent fibrous airway obliteration in this study, it was effective in inhibiting the process. The duration and extent of such inhibition, particularly when combined with immunosuppressive medication, will be addressed in future studies.