- Email: [email protected]
Recipient intramuscular gene transfer of active transforming growth factor-β1 attenuates acute lung rejection
Recipient Intramuscular Gene Transfer of Active Transforming Growth Factor-1 Attenuates Acute Lung Rejection Takashi Suda, MD, Franco D’Ovidio, MD, Niccolo Daddi, MD, Jon H. Ritter, MD, Thalachallour Mohanakumar, PhD, and G. Alexander Patterson, MD, FRCS(C) Division of Cardiothoracic Surgery, Department of Pathology and Department of Immunology, Washington University School of Medicine, Barnes Jewish Hospital, St. Louis, Missouri
Background. Gene transfer into the donor graft has been demonstrated to be feasible in reducing ischemiareperfusion injury and rejection in lung transplantation. This study was undertaken to determine whether intramuscular gene transfer into the recipient can also reduce subsequent lung graft rejection. Methods. Brown Norway rats served as donors and F344 rats as recipients. Recipient animals were injected with 1010 plaque-forming units of adenovirus encoding active transforming growth factor 1 (group I, n ⴝ 6), -galactosidase as adenoviral controls (group II, n ⴝ 6), or normal saline without adenovirus (group III, n ⴝ 6) into both gluteus muscles 2 days before transplantation. Gene expression was confirmed by enzyme-linked immunosorbent assay. Graft function was assessed on postoperative day 5. Results. Successful gene transfection and expression
were confirmed by the presence of active transforming growth factor 1 protein in muscle and plasma. Oxygenation was significantly improved in group I (group I vs II and III, 353.6 ⴞ 63.0 mm Hg vs 165.7 ⴞ 39.9 and 119.1 ⴞ 41.5 mm Hg; p ⴝ 0.02 and 0.004). The muscle transfected with the transforming growth factor 1 showed granulation tissue with fibroblast accumulation. Conclusions. Intramuscular adenovirus-mediated gene transfer of active transforming growth factor 1 into the recipients attenuates acute lung rejection as manifested by significantly improved oxygenation in transplanted lung allografts. This intramuscular transfection approach as a cytokine therapy is feasible in transplantation and may be useful in reducing rejection as well as reperfusion injury. (Ann Thorac Surg 2001;71:1651– 6) © 2001 by The Society of Thoracic Surgeons
A
that the transgene products of muscle in recipients enter into the circulation, resulting in beneficial effects on the transplanted organs. Transforming growth factor 1 (TGF-1) is one of a number of closely related, multifunctional molecules that play central roles in embryonic development, tumorigenesis, wound healing, fibrosis, and immunoregulation [7]. The immunomodulator function is expressed by suppressing the proliferation of B and T cells; antagonizing inflammatory cytokines such as interleukin-1, tumor necrosic factor-alpha (TNF␣), or interferon-gamma (IFN␥); and inhibiting natural killer cells [8 –10]. We hypothesized that the inhibition of immune response by intramuscular TGF1 gene transfection into the recipient could suppress acute allograft rejection. The aims of the present study were: (1) to utilize an intramuscular gene transfection strategy to achieve gene overexpression in muscle and the bloodstream; and (2) to investigate the feasibility of gene transfection with active TGF-1 gene into the recipient by using an adenoviral vector to study its effects on acute rejection in a lung transplant model.
lthough lung transplantation has proven to be a valid therapeutic option for a variety of end-stage pulmonary disease, lung allograft ischemia-reperfusion injuries, as well as acute and chronic rejection, continue to be challenges to the success of lung transplantation. Gene transfer into the lung has the potential to be a powerful treatment for pulmonary disease by creating a means of providing local immunosupression [1]. In experimental lung transplantation, gene transfer into the donor lung graft is feasible and may be useful in reducing ischemia-reperfusion injury and rejection. Recently, our laboratory has demonstrated the effectiveness of gene transfer in the ex vivo (graft) or in vivo (donor before harvest) rat lung graft model [2, 3]. Gene transfer into the recipient is another potential therapeutic option. Recipient intramuscular gene transfections of interleukin-10 [4], major histocompatibility complex class I antigen [5], and granulocyte-macrophage colony-stimulating factor (GM-CSF) [6] have been demonstrated to decrease experimental graft rejection. These studies suggested
Presented at the Thirty-seventh Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 29 –31, 2001. Address reprint requests to Dr Patterson, Division of Cardiothoracic Surgery, Washington University School of Medicine, One Barnes Jewish Hospital Plaza, 3108 Queeny Tower, St. Louis, MO 63110; e-mail: [email protected].
© 2001 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
Material and Methods The first groups of experiments (nontransplant setting, groups Ex-I, Ex-II, and Ex-III) were aimed to demonstrate 0003-4975/01/$20.00 PII S0003-4975(01)02528-0
1652
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
in vivo intramuscular gene delivery and to confirm gene expression in muscle and plasma. The second groups of experiments (transplant setting, groups I, II, and III) were conducted to evaluate the effect of TGF-1 gene transfection into the recipient in a rat lung transplant model of acute rejection.
Animals F344 rats and Brown Norway rats (Harlan Sprague Dawley Inc, Indianapolis, IN), weighing 250 to 270 g, were used in all experiments. All animal procedures were approved by the Animal Studies Committee at Washington University. Animals received humane care in compliance with “Principles of Laboratory Animal Care” formulated by the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH publication 85-23, revised 1985).
Adenoviral Vectors Adenovirus encoding rat TGF-1 (AdCMVTGF1 provided by Dr Debra A. Hullett, Department of Surgery, University of Wisconsin, Madison, WI) is a replicationdeficient adenoviral vector encoding the mutated “bioactive” TGF-1 gene, driven by the cytomegalovirus immediate early promoter. This contains a mutation of cystine to serine at positions 223 and 225, rendering the expressed TGF-1 biologically active. Site-specific mutation of codons encoding for Cys-223 and Cys-225 was performed by altering the Cys-223 codon from TGC to AGC and altering the Cys-225 codon from TGT to TCT. Complementary changes were also made in the 3⬘ direction. Therefore, in the resulting protein sequence, these two cystein residues were substituted by serine residues. First-generation replication-deficient adenovirus serotype 5 carrying the Eshcherichia coli LacZ gene encoding for -galactosidase and driven by the constitutive cytomegalovirus promoter (AdCMVgal) served as controls. It was provided as a gift from the Gene Therapy Center at the University of North Carolina (Chapel Hill, NC). Adenoviral amplification was achieved by propagation in 293 cells to obtain high-titer stocks, as determined by the plaque assay (courtesy of Dr R. Jude Samulski and Dr Douglas McCarty, Gene Therapy Center Vector Core Facility, University of North Carolina, Chapel Hill, NC). Purified viral aliquots were stored at ⫺80°C in 10% glycerol buffered with 10 mmol/L tris, 140 mmol/L NaCl, and 1 mmol/L MgCl2. Immediately before use, these stocks were thawed and diluted in 1 mL of sterile normal saline.
Experiment 1 (Nontransplant Setting): Expression of Active TGF-1 Gene (Groups Ex-I, Ex-II, and EX-III) These experiments were performed to demonstrate in vivo intramuscular gene delivery and to confirm gene expression. F344 rats were divided into three groups (n ⫽ 3 per group). In groups Ex-I and Ex-II, animals received intramuscular injection of 1010 plaque-forming units
Ann Thorac Surg 2001;71:1651– 6
(pfu) of adenovirus encoding active TGF-1 (group Ex-I) or -galactosidase (group Ex-II) in two equally divided doses into both gluteus muscles of the animal. In group Ex-III, rats were injected with normal saline. Animals were sacrificed 1, 2, and 7 days after injection and both gluteus muscles were harvested. Blood samples were obtained from the ascending aorta. The expression of recombinant active TGF-1 gene was determined by enzyme-linked immunosorbent assay (ELISA) as described below.
Experiment 2 (Transplant Setting): Effects of Active TGF-1 Gene Transfection (Groups I, II, and III) Animals were randomly divided into three groups. Brown Norway rats (RT1n) served as donors and F344 rats (RT11v1) as recipients. This strain combination was chosen because of the strong major and minor histocompatibility locus mismatch that results in well-documented complete lung graft rejection that is known to occur within the fifth postoperative day in control recipients without immunosupression. Recipient animals received 1010 pfu of adenovirus encoding active TGF-1 (group I, n ⫽ 6), 1010 pfu of adenovirus encoding -galactosidase as adenoviral controls (group II, n ⫽ 6), or normal saline without adenovirus (group III, n ⫽ 6). Forty-eight hours after gene transfection into the recipients, donor lungs were harvested as described below. Briefly, after general anesthesia, mechanical ventilation, and systemic heparinization, donor rat lungs were flushed through the main pulmonary artery with 20 mL of cold (4°C) lowpotassium dextran-1% glucose (LPDG) solution at 20-cm H2O pressure. The heart-lung block was then removed with the lungs inflated at end-tidal volume. The left lung was stored at 4°C in LPDG until implantation. Recipient animals were anesthetized and intubated, and underwent a left thoracotomy. The pulmonary vessels were anastomosed using a modification of the previously described “cuff technique” [2, 3]. Bronchial anastomosis was performed using a running 8-0 Prolene suture (Ethicon, Somerville, NJ). Ventilation and perfusion were restored and a temporary chest tube was placed, which was removed after recovery from anesthesia. In all groups, no immunosuppressive drugs were used and recipients were sacrificed on the fifth postoperative day based on previous studies [3]. Recipient animals were reanesthetized using the donor technique described above, then mechanically ventilated with 100% oxygen. Median laparotomy-sternotomy was performed and the contralateral right hilum was clamped. Animals were then ventilated for 5 minutes using a tidal volume of 1.5 mL, respiratory rate of 100 breaths/minute, and PEEP 1.0 H2O in order to assess the function of the lung isograft by arterial blood gas analysis using blood samples obtained from the ascending aorta. After the animals were euthanized, the lung graft was flushed with cold saline solution and the grafts were subjected to ELISA testing to investigate the endogenous TGF-1 expression in transplanted lungs.
Ann Thorac Surg 2001;71:1651– 6
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
1653
Histologic Assessment Lungs were perfused through the pulmonary arterial trunk with 20 mL of normal saline and 20 mL of Histochoice (Amresco, Solon, OH). The specimens were fixed in Histochoice for 24 hours at 4°C and embedded in paraffin wax. Tissue sections 7 m thick were cut on a microtome and mounted on slides. Lungs were stained with hematoxylin and eosin (H-E). A blinded observer (pathologist: J.H.R.) scored rejection according to the 1995 revision of the working formulation for the classification of pulmonary allograft rejection [11]. Vascular and airway rejection scores ranged from 0 (no rejection) to 4 (complete destruction of the allograft). The muscle injected with vectors or saline were stained with H-E or Elastic van Gieson (EVG), a specific histochemical stain for collagen and elastin, to investigate the influences of active TGF-1 transfection.
ELISA for Active TGF-1 Blood samples (3 mL) were collected into ethylenediaminetetraacetic acid-containing tubes with pepstatin A (5 g/mL) and protease inhibitor cocktail (Complete Mini Tabs; Boehringer-Mannheim, Mannheim, Germany). Plasma was obtained by centrifugation at 1,000 g for 30 minutes and clarified by centrifugation at 15,000 g for 15 minutes. The platelet-poor plasma was then stored at ⫺80°C until ELISA assessment. The active TGF-1 protein was extracted by homogenizing muscle in lysis solution containing 100 mmol/L potassium phosphate (pH 7.8), 0.2% triton X-100 with pepstatin A (5 g/mL), and protease inhibitor cocktail (Complete Mini Tabs; Boehringer-Mannheim). The homogenate was then centrifuged at 15,000 rpm for 15 minutes after extraction at room temperature for 15 minutes, and the supernatant was stored at ⫺80°C until ELISA assessment. The TGF-1 ELISA kit used in this study (R&D Systems, Minneapolis, MN) is cross-reactive for human, rat, and mouse. The procedure to activate latent TGF-1 in the ELISA kit was not used for the detection of active TGF-1.
Statistical Analysis Values are reported as mean ⫾ standard error of the mean. In the pathologic rejection score, Kruskal-Wallis rank test was used to compare groups. In other assessments, one-way analysis of variance (ANOVA) with pairwise comparison by Fisher’s PLSD method was used to compare overall differences among groups. Differences were considered significant if the p value was less than 0.05.
Results Experiment 1 Increased active TGF-1 protein expression in group Ex-I muscle was evident from day 1, peaked at day 2, and decreased at day 7. The protein production in the muscle was statistically different from groups injected with AdCMVgal or normal at day 1 ( p ⬍ 0.0001 for group
Fig 1. Level of active TGF-1 in muscle (A) and plasma (B) of rats injected with AdCMVTGF1, AdCMVgal, or normal saline. Muscles were collected at various time points after injection, and were assayed by ELISA. *p ⬍ 0.05, **p ⬍ 0.005.
Ex-I vs Ex-II, p ⬍ 0.0001 for group Ex-I vs Ex-III), and day 2 ( p ⫽ 0.0030 for group Ex-I vs Ex-II, p ⫽ 0.0032 group Ex-I vs Ex-III) (Fig 1A). Increased active TGF-1 protein expression in group Ex-I plasma was evident from day 2, and decreased at day 7. The protein levels in plasma was statistically different from groups injected with AdCMVgal or normal saline at day 2 ( p ⫽ 0.0307 for group Ex-I vs Ex-III), and day 7 ( p ⫽ 0.0028 for group Ex-I vs Ex-II, p ⫽ 0.0158 group Ex-I vs Ex-III) (Fig 1B).
Experiment 2 Left lungs from group I, which had been transfected with TGF-1, had superior arterial oxygen (PaO2) levels compared with groups II or III. This difference between group I versus groups II or III was statistically significant, with p ⫽ 0.02 for group I vs II and p ⫽ 0.004 for group I versus III (Fig 2). Mean arterial carbon dioxide tensions were not significantly different in any of the groups (43.1 ⫾ 12.2, 33.2 ⫾ 1.2, and 47.3 ⫾ 9.5 mm Hg, for groups I, II, and III, respectively, p ⬎ 0.5). Vascular rejection scores were (median): 2.83 (range 2 to 3.5) in group I (TGF-1); 2.91 (range 2.5 to 3) in group
1654
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
Ann Thorac Surg 2001;71:1651– 6
Fig 2. The mean arterial oxygenation (PaO2) after mono-ventilation of the allograft at the time of sacrifice on the fifth postoperative day. PaO2 in group I was superior in comparison with group II and group III. *p ⬍ 0.05, **p ⬍ 0.005.
II (-galactosidase); and 2.83 (range 2 to 3) in group III (saline solution). No statistical differences were noted among the groups. Airway rejection scores were (median): 2.33 (range 2 to 3) in group I; 2.92 (range 2.5 to 3) in group II; 2.42 (range 2 to 3) in group III. No statistical differences were noted among the groups. H-E staining in muscle injected with AdCMVTGF1 showed granulation tissue with fibroblast accumulation (Fig 3A). This was not seen in muscles transfected with either -galactosidase or injected with saline solution alone; the control cases injected with AdCMVgal showed only mild interstitial chronic inflammation. The EVG staining showed modest collagen deposition in the granulated muscle (Fig 3B). Mean active TGF-1 protein expressions (pg/100 g total protein) in transplanted lungs were not significantly different in any of the groups (9.555 ⫾ 2.222, 9.402 ⫾ 4.277, and 5.560 ⫾ 1.033, for groups I, II, and III, respectively; p ⬎ 0.5).
Comment
Fig 3. H-E staining (A) and Collagen red and Elastin black EVG staining (B) in muscle with transfected active TGF-1. (A) There is some focal myocyte necrosis, while other myofibers are trapped within and compressed by granulation tissue with fibroblast accumulation (⫻100). (B) The EVG-stained section of the AdCMVTGF1 injected highlights the areas of myocyte destruction, and also indicates the cellular nature of the fibrosis, with accumulation of fibroblasts and only modest collagen deposition. Increased elastin was not clear (⫻100).
In the present study, we transfected the gluteus muscles of the recipients to determine whether gene transfer into recipient can reduce acute rejection in transplanted lung grafts. Expression of the gene products entered the circulation and had beneficial effects on the distant transplanted lung grafts, as manifested by significantly improved oxygenation. The targets of gene therapy are both intracellular and extracellular. Intracellularly, the gene delivered to the cell affects the cell by replacing a defective or missing gene or by providing a product that kills or inhibits the growth of cells. Gene delivered to the cell may also protect the cells. Extracellularly, the cell releases the product of the new gene, which can act locally on neighboring cells or enter the circulation for delivery to distant cells [12]. The present study used the extracellular approach for delivery to distant lung grafts by
intramuscular transfection. The advantages of this intramuscular approach are: (1) the procedure is easy in the clinical setting; (2) the cytokine therapy can be performed simply; and (3) the graft, which was injured by storage, ischemia-reperfusion, and rejection, does not exhibit the inflammation usually associated with transfection. The adenovirus vector is known as a cause of local inflammation and the transfected gene may have local side effects such as the fibrosis due to the TGF-1. The graft, located at a distant site, is spared the local inflammatory events associated with transfection. Direct gene transfer into muscle was investigated by Wolff and colleagues using naked plasmid [13]. The nuclei of mature muscle do not undergo division; thus,
Ann Thorac Surg 2001;71:1651– 6
naked plasmid may be useful for this approach [5, 6, 14]. However, the disadvantage of naked plasmid is that relatively few cells take up the DNA (0.64% to 3%). Thus, only a small amount of the encoded protein is produced. In contrast, LacZ gene expression after intramuscular gene transfer with an adenoviral vector was observed in 21.8% of cells by immunohistochemical staining [12, 15]. Therefore, if a large amount of gene products is required to reach distant grafts, naked plasmid transfection might be not suitable. Adenovirus injected intramuscularly remained localized at the site of injection. We are investigating the organ distribution of transfection by different approaches, such as intravenous, airway, and intramuscular administration. The study showed that when the adenovirus was injected into skeletal muscle, the adenovirus was transfected into the muscle, spleen, and liver. However, the expression in liver and spleen was very weak and significantly lower than the intravenous transfection group (unpublished data). In the present study, successful gene transfection and expression was confirmed by active TGF-1 protein in muscle and plasma. However, the levels decreased by day 7. The duration of the effects from the adenovirus intramuscular approach has been shown to vary in previous studies using mouse, rat, rabbit, and primate experimental models [16, 17]. Because the duration of adenovirus vector is shortened by an antiviral cellular immune response to viral proteins expressed on the cell surface, this approach may be beneficial mainly in acute rejection or ischemiareperfusion injury in the transplant setting. Attenuation of lung allograft acute rejection as assessed by arterial oxygenation was noted after intramuscular administration of AdCMVTGF1. Transforming growth factors encompass polypeptides that regulate the growth and differentiation of both normal and transformed cells. One of these, TGF-1, has been isolated and purified from various normal and neoplastic cells [18]. TGF-1 is produced naturally as a latent 390-amino acid dimeric precursor that is converted into a mature bioactive 112-amino acid dimer after cleavage and dissociation of the amino-terminal portion, termed the latency associated peptide (LAP). This cleavage and dissociation can be readily achieved in vitro nonphysiologically by extremes of pH, temperature, and chaotropic agents. However, the activation mechanisms have not yet been fully elucidated in vivo, although thrombospondin and the proteases plasmin and cathepsin have been implicated. By using the active form of TGF-1 in the present study, it was easy to control the blood level of active TGF-1, which has a mature bioactivity. There is a possibility that if the overexpressed latent form TGF-1 is achieved by transfection, it may not be converted to enough of the active form of TGF-1 in vivo. H-E and EVG staining in the present study showed that transfection of the active form of TGF-1 induced severe fibrosis, granulation, and collagen deposition in the muscle where it was injected. Sime and colleagues demonstrated that transient overexpression of active, but not latent, TGF-1 resulted in prolonged and severe interstitial and pleural fibrosis characterized by extensive deposition of extracellular
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
1655
matrix proteins collagen, fibronectin, and elastin, and by emergence of cells with the myofibroblast phenotype after active TGF-1 transfection intratracheally [19]. In regard to prevention of side effects such as fibrosis, because the latent form of TGF-1 can be regulated in vivo, transfection of latent rather than active TGF-1 may provide a better route to deliver this powerful immunosuppressive agent in vivo. Therefore, further investigation regarding comparison of latent and active TGF-1 using this approach is warranted. In terms of the rejection score, it was not apparent that amelioration compared with the oxygenation. This suggests that histologic examination may be a crude method for analyzing the rejection process. However, histologic examination in the TGF-1-transfected group showed mild to moderate acute rejection. If cotransfer of immunosuppressive cytokine genes such as interleukin-10 are used, more efficient reduction in acute rejection may be accomplished. The goal of our study was to verify the applicability of gene therapy by intramuscular injection of adenovirus encoding for the TGF-1 in an acute rejection model. We chose to deliver the adenovirus well in advance of transplant in order to have optimal transgene expression at the time of lung allograft implantation. Although the time frame of 48 hours does not mimic current clinical settings, our choice was based on data in the preliminary experiments described in this report. In previous studies in our laboratory, we have documented amelioration of acute rejection by ex vivo TGF-1 lung allograft transfection at the time of harvest by using nonviral vectors and other delivery routes [20]. We plan a follow-up study on the best timing of gene therapy administration of TGF-1, whether before harvest, at reperfusion, or in conjunction with the onset of acute rejection, thus aiming the therapy as a precise prophylactic or rescue strategy. In conclusion, active TGF-1 intramuscular gene transfer into recipients has beneficial effects as a cytokine therapy in the rat lung transplantation setting. This intramuscular transfection approach is feasible in transplantation of other organs and may be useful for not only reducing graft rejection but also reperfusion injury. We thank Dr Debra A. Hullett (Department of Surgery, University of Wisconsin, Madison, WI) for kindly providing adenovirus encoding TGF-1. We also thank Kathleen Grapperhaus for technical assistance, Dawn Schuessler and Mary Ann Kelly for secretarial support, and Diane Toeniskoetter for her assistance. Statistical advice was obtained from Richard B. Schuessler, PhD. This work was supported by National Institutes of Health (NIH) grant 1 R01 HL-41281; T.M. is supported by NIH grant HL-56643.
References 1. Hullett DA. Gene therapy in transplantation. J Heart Lung Transplant 1996;15:857– 62. 2. Suda T, Mora BN, D’Ovidio F, et al. In vivo adenovirusmediated endothelial nitric oxide synthase gene transfer ameliorates lung allograft ischemia-reperfusion injury. J Thorac Cardiovasc Surg 2000;119:297–304. 3. Yano M, Mora BN, Ritter JM, et al. Ex vivo transfection of transforming growth factor 1 gene to pulmonary artery
1656
4.
5. 6. 7. 8. 9.
10. 11.
12.
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
segments in lung grafts. J Thorac Cardiovasc Surg 1999;117: 705–13. Boehler A, Chamberlain D, Xing Z, et al. Adenovirusmediated interleukin-10 gene transfer inhibits posttransplant fibrous airway obliteration in an animal model of bronchiolitis obliterans. Hum Gene Ther 1998;9:541–51. Geissler EK, Wang J, Fechner JH, Burlingham WJ, Knechtle SJ. Immunity to MHC class I antigen after direct DNA transfer into skeletal muscle. J Immunol 1994;152:413–21. Zhai Y, Hong X, Wang J, et al. Modulation of alloimmunity to major histocompatibility complex class I by cotransfer of cytokine genes in vivo. Transpl Immunol 1998;6:169 –175. Qin L, Chavin KD, Ding Y, et al. Gene transfer for transplantation: Prolongation of allograft survival with transforming growth factor-1. Ann Surg 1994;220:508–19. Ruscetti FW, Palladino MA. Transforming growth factor-beta and the immune system. Prog Growth Fact Res 1991;3:159–75. Rook AH, Kehrl JH, Wakefield LM, et al. Effects of transforming growth factor-beta on the functions of natural killer cells: depressed cytolitic activity blunting of interferon responsiveness. J Immunol 1989;136:3916–20. Wahl SM, McCarteney-Francis N, Mergenhagen SE. Inflammatory and immunomodulatory roles of TGF-1. Immunol Today 1989;10:258– 61. Yousem SA, Berry GJ, Cagle PT, et al. Revision of the 1990 working formulation for classification of pulmonary allograft rejection: Lung Rejection Study Group. J Heart Lung Transplant 1996;15:1–15. Blau HM, Springer ML. Gene therapy—A novel form of drug delivery. N Engl J Med 1995;2:1204– 07.
Ann Thorac Surg 2001;71:1651– 6
13. Wolff JA, Malone RW, Williams P, et al. Direct gene transfer into mouse muscle in vivo. Science 1990;247:1465– 8. 14. Donnelly JJ, Ulmer JB, Liu MA. DNA vaccines. Life Sci 1997; 60:163–72. 15. Couffinhal T, Kearney M, Sullivan A, Silver M, Tsurumi Y, Isner JM. Histchemical staining following LacZ gene transfer underestimates transfection efficiency. Hum Gene Ther 1997; 8:929–34. 16. Tripathy SK, Goldwasser E, Lu MM, Barr E, Leiden JM. Stable delivery of physiologic levels of recombinant erythropoietin to systemic circulation by intramuscular injection of replication-defective adenovirus. Proc Natl Acad Sci USA 1994;91:11557– 61. 17. Svensson EC, Black HB, Dugger DL, et al. Long-term erythropoietin expression in rodents and non human primates following intramuscular injection of a replication-defective adenoviral vector. Hum Gene Ther 1997;8:1797– 806. 18. Wallick SC, Figari IS, Morris RE, Levinson AD, Palladino MA. Immunoregulatory role of transforming growth factor  (TGF-) in development of killer cells: comparison of active and latent TGF-1. J Exp Med 1990;172:1777– 84. 19. Sime PJ, Xing Z, Graham FL, Csaky KG, Gauldie J. Adenovector-mediated gene transfer of active transforming growth factor-1 induces prolonged severe fibrosis in rat lung. J Clin Invest 1997;100:768–76. 20. Boasquevisque CHR, Mora BN, Boglione M, et al. Liposomemediated gene transfer in rat lung transplantation: a comparison between the in vivo and ex vivo approaches. J Thorac Cardiovasc Surg 1999;117:8–15.
DISCUSSION DR JOSEPH B. ZWISCHENBERGER (Galveston, TX): In your abstract you presented your Pao2 data.
DR ZWISCHENBERGER: How do you think TGF-, the growth factor, impacted directly on lung function?
DR SUDA: Yes. DR ZWISCHENBERGER: Was this study on 100% oxygen? DR SUDA: Yes. DR ZWISCHENBERGER: You performed a statistical comparison on partial pressure of oxygen in the blood. In terms of calculated oxygen content due to the saturation oxygen in the plasma, there is virtually no difference between those three numbers. I would suggest that you do an estimated shunt fraction where you’ll find that your data show that you have a 30% shunt, a 27% shunt, but in your treated group you have a 17% shunt. I would encourage you to include that in your manuscript. Is there any proposed mechanism of this improvement? DR SUDA: Regarding mechanisms, we did not investigate mechanisms in this study. We currently are investigating mechanism of this approach using TGF-1, however.
DR SUDA: TGF-1 alters the cytokine balance favoring an antiinflammatory environment and inhibits chemokine induced T lymphocyte recruitment and proliferation. Furthermore, other experiments found in the literature indicate that TGF-1 contributes to a TH2 type immune response, which is responsible for improved graft survival.
DR HARVEY I. PASS (Detroit, MI): Why do you have to use gene therapy? Why can’t you use recombinant TGF- and do the same thing? There may be differences with regard to lymphocyte recruitment depending upon whether you use DNA versus protein. Have you done protein comparisons?
DR SUDA: Because the half-life of TGF-1 is very short, it is very difficult to use recombinant TGF-1 using this approach.
Background. Gene transfer into the donor graft has been demonstrated to be feasible in reducing ischemiareperfusion injury and rejection in lung transplantation. This study was undertaken to determine whether intramuscular gene transfer into the recipient can also reduce subsequent lung graft rejection. Methods. Brown Norway rats served as donors and F344 rats as recipients. Recipient animals were injected with 1010 plaque-forming units of adenovirus encoding active transforming growth factor 1 (group I, n ⴝ 6), -galactosidase as adenoviral controls (group II, n ⴝ 6), or normal saline without adenovirus (group III, n ⴝ 6) into both gluteus muscles 2 days before transplantation. Gene expression was confirmed by enzyme-linked immunosorbent assay. Graft function was assessed on postoperative day 5. Results. Successful gene transfection and expression
were confirmed by the presence of active transforming growth factor 1 protein in muscle and plasma. Oxygenation was significantly improved in group I (group I vs II and III, 353.6 ⴞ 63.0 mm Hg vs 165.7 ⴞ 39.9 and 119.1 ⴞ 41.5 mm Hg; p ⴝ 0.02 and 0.004). The muscle transfected with the transforming growth factor 1 showed granulation tissue with fibroblast accumulation. Conclusions. Intramuscular adenovirus-mediated gene transfer of active transforming growth factor 1 into the recipients attenuates acute lung rejection as manifested by significantly improved oxygenation in transplanted lung allografts. This intramuscular transfection approach as a cytokine therapy is feasible in transplantation and may be useful in reducing rejection as well as reperfusion injury. (Ann Thorac Surg 2001;71:1651– 6) © 2001 by The Society of Thoracic Surgeons
A
that the transgene products of muscle in recipients enter into the circulation, resulting in beneficial effects on the transplanted organs. Transforming growth factor 1 (TGF-1) is one of a number of closely related, multifunctional molecules that play central roles in embryonic development, tumorigenesis, wound healing, fibrosis, and immunoregulation [7]. The immunomodulator function is expressed by suppressing the proliferation of B and T cells; antagonizing inflammatory cytokines such as interleukin-1, tumor necrosic factor-alpha (TNF␣), or interferon-gamma (IFN␥); and inhibiting natural killer cells [8 –10]. We hypothesized that the inhibition of immune response by intramuscular TGF1 gene transfection into the recipient could suppress acute allograft rejection. The aims of the present study were: (1) to utilize an intramuscular gene transfection strategy to achieve gene overexpression in muscle and the bloodstream; and (2) to investigate the feasibility of gene transfection with active TGF-1 gene into the recipient by using an adenoviral vector to study its effects on acute rejection in a lung transplant model.
lthough lung transplantation has proven to be a valid therapeutic option for a variety of end-stage pulmonary disease, lung allograft ischemia-reperfusion injuries, as well as acute and chronic rejection, continue to be challenges to the success of lung transplantation. Gene transfer into the lung has the potential to be a powerful treatment for pulmonary disease by creating a means of providing local immunosupression [1]. In experimental lung transplantation, gene transfer into the donor lung graft is feasible and may be useful in reducing ischemia-reperfusion injury and rejection. Recently, our laboratory has demonstrated the effectiveness of gene transfer in the ex vivo (graft) or in vivo (donor before harvest) rat lung graft model [2, 3]. Gene transfer into the recipient is another potential therapeutic option. Recipient intramuscular gene transfections of interleukin-10 [4], major histocompatibility complex class I antigen [5], and granulocyte-macrophage colony-stimulating factor (GM-CSF) [6] have been demonstrated to decrease experimental graft rejection. These studies suggested
Presented at the Thirty-seventh Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 29 –31, 2001. Address reprint requests to Dr Patterson, Division of Cardiothoracic Surgery, Washington University School of Medicine, One Barnes Jewish Hospital Plaza, 3108 Queeny Tower, St. Louis, MO 63110; e-mail: [email protected].
© 2001 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
Material and Methods The first groups of experiments (nontransplant setting, groups Ex-I, Ex-II, and Ex-III) were aimed to demonstrate 0003-4975/01/$20.00 PII S0003-4975(01)02528-0
1652
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
in vivo intramuscular gene delivery and to confirm gene expression in muscle and plasma. The second groups of experiments (transplant setting, groups I, II, and III) were conducted to evaluate the effect of TGF-1 gene transfection into the recipient in a rat lung transplant model of acute rejection.
Animals F344 rats and Brown Norway rats (Harlan Sprague Dawley Inc, Indianapolis, IN), weighing 250 to 270 g, were used in all experiments. All animal procedures were approved by the Animal Studies Committee at Washington University. Animals received humane care in compliance with “Principles of Laboratory Animal Care” formulated by the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences and published by the National Institutes of Health (NIH publication 85-23, revised 1985).
Adenoviral Vectors Adenovirus encoding rat TGF-1 (AdCMVTGF1 provided by Dr Debra A. Hullett, Department of Surgery, University of Wisconsin, Madison, WI) is a replicationdeficient adenoviral vector encoding the mutated “bioactive” TGF-1 gene, driven by the cytomegalovirus immediate early promoter. This contains a mutation of cystine to serine at positions 223 and 225, rendering the expressed TGF-1 biologically active. Site-specific mutation of codons encoding for Cys-223 and Cys-225 was performed by altering the Cys-223 codon from TGC to AGC and altering the Cys-225 codon from TGT to TCT. Complementary changes were also made in the 3⬘ direction. Therefore, in the resulting protein sequence, these two cystein residues were substituted by serine residues. First-generation replication-deficient adenovirus serotype 5 carrying the Eshcherichia coli LacZ gene encoding for -galactosidase and driven by the constitutive cytomegalovirus promoter (AdCMVgal) served as controls. It was provided as a gift from the Gene Therapy Center at the University of North Carolina (Chapel Hill, NC). Adenoviral amplification was achieved by propagation in 293 cells to obtain high-titer stocks, as determined by the plaque assay (courtesy of Dr R. Jude Samulski and Dr Douglas McCarty, Gene Therapy Center Vector Core Facility, University of North Carolina, Chapel Hill, NC). Purified viral aliquots were stored at ⫺80°C in 10% glycerol buffered with 10 mmol/L tris, 140 mmol/L NaCl, and 1 mmol/L MgCl2. Immediately before use, these stocks were thawed and diluted in 1 mL of sterile normal saline.
Experiment 1 (Nontransplant Setting): Expression of Active TGF-1 Gene (Groups Ex-I, Ex-II, and EX-III) These experiments were performed to demonstrate in vivo intramuscular gene delivery and to confirm gene expression. F344 rats were divided into three groups (n ⫽ 3 per group). In groups Ex-I and Ex-II, animals received intramuscular injection of 1010 plaque-forming units
Ann Thorac Surg 2001;71:1651– 6
(pfu) of adenovirus encoding active TGF-1 (group Ex-I) or -galactosidase (group Ex-II) in two equally divided doses into both gluteus muscles of the animal. In group Ex-III, rats were injected with normal saline. Animals were sacrificed 1, 2, and 7 days after injection and both gluteus muscles were harvested. Blood samples were obtained from the ascending aorta. The expression of recombinant active TGF-1 gene was determined by enzyme-linked immunosorbent assay (ELISA) as described below.
Experiment 2 (Transplant Setting): Effects of Active TGF-1 Gene Transfection (Groups I, II, and III) Animals were randomly divided into three groups. Brown Norway rats (RT1n) served as donors and F344 rats (RT11v1) as recipients. This strain combination was chosen because of the strong major and minor histocompatibility locus mismatch that results in well-documented complete lung graft rejection that is known to occur within the fifth postoperative day in control recipients without immunosupression. Recipient animals received 1010 pfu of adenovirus encoding active TGF-1 (group I, n ⫽ 6), 1010 pfu of adenovirus encoding -galactosidase as adenoviral controls (group II, n ⫽ 6), or normal saline without adenovirus (group III, n ⫽ 6). Forty-eight hours after gene transfection into the recipients, donor lungs were harvested as described below. Briefly, after general anesthesia, mechanical ventilation, and systemic heparinization, donor rat lungs were flushed through the main pulmonary artery with 20 mL of cold (4°C) lowpotassium dextran-1% glucose (LPDG) solution at 20-cm H2O pressure. The heart-lung block was then removed with the lungs inflated at end-tidal volume. The left lung was stored at 4°C in LPDG until implantation. Recipient animals were anesthetized and intubated, and underwent a left thoracotomy. The pulmonary vessels were anastomosed using a modification of the previously described “cuff technique” [2, 3]. Bronchial anastomosis was performed using a running 8-0 Prolene suture (Ethicon, Somerville, NJ). Ventilation and perfusion were restored and a temporary chest tube was placed, which was removed after recovery from anesthesia. In all groups, no immunosuppressive drugs were used and recipients were sacrificed on the fifth postoperative day based on previous studies [3]. Recipient animals were reanesthetized using the donor technique described above, then mechanically ventilated with 100% oxygen. Median laparotomy-sternotomy was performed and the contralateral right hilum was clamped. Animals were then ventilated for 5 minutes using a tidal volume of 1.5 mL, respiratory rate of 100 breaths/minute, and PEEP 1.0 H2O in order to assess the function of the lung isograft by arterial blood gas analysis using blood samples obtained from the ascending aorta. After the animals were euthanized, the lung graft was flushed with cold saline solution and the grafts were subjected to ELISA testing to investigate the endogenous TGF-1 expression in transplanted lungs.
Ann Thorac Surg 2001;71:1651– 6
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
1653
Histologic Assessment Lungs were perfused through the pulmonary arterial trunk with 20 mL of normal saline and 20 mL of Histochoice (Amresco, Solon, OH). The specimens were fixed in Histochoice for 24 hours at 4°C and embedded in paraffin wax. Tissue sections 7 m thick were cut on a microtome and mounted on slides. Lungs were stained with hematoxylin and eosin (H-E). A blinded observer (pathologist: J.H.R.) scored rejection according to the 1995 revision of the working formulation for the classification of pulmonary allograft rejection [11]. Vascular and airway rejection scores ranged from 0 (no rejection) to 4 (complete destruction of the allograft). The muscle injected with vectors or saline were stained with H-E or Elastic van Gieson (EVG), a specific histochemical stain for collagen and elastin, to investigate the influences of active TGF-1 transfection.
ELISA for Active TGF-1 Blood samples (3 mL) were collected into ethylenediaminetetraacetic acid-containing tubes with pepstatin A (5 g/mL) and protease inhibitor cocktail (Complete Mini Tabs; Boehringer-Mannheim, Mannheim, Germany). Plasma was obtained by centrifugation at 1,000 g for 30 minutes and clarified by centrifugation at 15,000 g for 15 minutes. The platelet-poor plasma was then stored at ⫺80°C until ELISA assessment. The active TGF-1 protein was extracted by homogenizing muscle in lysis solution containing 100 mmol/L potassium phosphate (pH 7.8), 0.2% triton X-100 with pepstatin A (5 g/mL), and protease inhibitor cocktail (Complete Mini Tabs; Boehringer-Mannheim). The homogenate was then centrifuged at 15,000 rpm for 15 minutes after extraction at room temperature for 15 minutes, and the supernatant was stored at ⫺80°C until ELISA assessment. The TGF-1 ELISA kit used in this study (R&D Systems, Minneapolis, MN) is cross-reactive for human, rat, and mouse. The procedure to activate latent TGF-1 in the ELISA kit was not used for the detection of active TGF-1.
Statistical Analysis Values are reported as mean ⫾ standard error of the mean. In the pathologic rejection score, Kruskal-Wallis rank test was used to compare groups. In other assessments, one-way analysis of variance (ANOVA) with pairwise comparison by Fisher’s PLSD method was used to compare overall differences among groups. Differences were considered significant if the p value was less than 0.05.
Results Experiment 1 Increased active TGF-1 protein expression in group Ex-I muscle was evident from day 1, peaked at day 2, and decreased at day 7. The protein production in the muscle was statistically different from groups injected with AdCMVgal or normal at day 1 ( p ⬍ 0.0001 for group
Fig 1. Level of active TGF-1 in muscle (A) and plasma (B) of rats injected with AdCMVTGF1, AdCMVgal, or normal saline. Muscles were collected at various time points after injection, and were assayed by ELISA. *p ⬍ 0.05, **p ⬍ 0.005.
Ex-I vs Ex-II, p ⬍ 0.0001 for group Ex-I vs Ex-III), and day 2 ( p ⫽ 0.0030 for group Ex-I vs Ex-II, p ⫽ 0.0032 group Ex-I vs Ex-III) (Fig 1A). Increased active TGF-1 protein expression in group Ex-I plasma was evident from day 2, and decreased at day 7. The protein levels in plasma was statistically different from groups injected with AdCMVgal or normal saline at day 2 ( p ⫽ 0.0307 for group Ex-I vs Ex-III), and day 7 ( p ⫽ 0.0028 for group Ex-I vs Ex-II, p ⫽ 0.0158 group Ex-I vs Ex-III) (Fig 1B).
Experiment 2 Left lungs from group I, which had been transfected with TGF-1, had superior arterial oxygen (PaO2) levels compared with groups II or III. This difference between group I versus groups II or III was statistically significant, with p ⫽ 0.02 for group I vs II and p ⫽ 0.004 for group I versus III (Fig 2). Mean arterial carbon dioxide tensions were not significantly different in any of the groups (43.1 ⫾ 12.2, 33.2 ⫾ 1.2, and 47.3 ⫾ 9.5 mm Hg, for groups I, II, and III, respectively, p ⬎ 0.5). Vascular rejection scores were (median): 2.83 (range 2 to 3.5) in group I (TGF-1); 2.91 (range 2.5 to 3) in group
1654
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
Ann Thorac Surg 2001;71:1651– 6
Fig 2. The mean arterial oxygenation (PaO2) after mono-ventilation of the allograft at the time of sacrifice on the fifth postoperative day. PaO2 in group I was superior in comparison with group II and group III. *p ⬍ 0.05, **p ⬍ 0.005.
II (-galactosidase); and 2.83 (range 2 to 3) in group III (saline solution). No statistical differences were noted among the groups. Airway rejection scores were (median): 2.33 (range 2 to 3) in group I; 2.92 (range 2.5 to 3) in group II; 2.42 (range 2 to 3) in group III. No statistical differences were noted among the groups. H-E staining in muscle injected with AdCMVTGF1 showed granulation tissue with fibroblast accumulation (Fig 3A). This was not seen in muscles transfected with either -galactosidase or injected with saline solution alone; the control cases injected with AdCMVgal showed only mild interstitial chronic inflammation. The EVG staining showed modest collagen deposition in the granulated muscle (Fig 3B). Mean active TGF-1 protein expressions (pg/100 g total protein) in transplanted lungs were not significantly different in any of the groups (9.555 ⫾ 2.222, 9.402 ⫾ 4.277, and 5.560 ⫾ 1.033, for groups I, II, and III, respectively; p ⬎ 0.5).
Comment
Fig 3. H-E staining (A) and Collagen red and Elastin black EVG staining (B) in muscle with transfected active TGF-1. (A) There is some focal myocyte necrosis, while other myofibers are trapped within and compressed by granulation tissue with fibroblast accumulation (⫻100). (B) The EVG-stained section of the AdCMVTGF1 injected highlights the areas of myocyte destruction, and also indicates the cellular nature of the fibrosis, with accumulation of fibroblasts and only modest collagen deposition. Increased elastin was not clear (⫻100).
In the present study, we transfected the gluteus muscles of the recipients to determine whether gene transfer into recipient can reduce acute rejection in transplanted lung grafts. Expression of the gene products entered the circulation and had beneficial effects on the distant transplanted lung grafts, as manifested by significantly improved oxygenation. The targets of gene therapy are both intracellular and extracellular. Intracellularly, the gene delivered to the cell affects the cell by replacing a defective or missing gene or by providing a product that kills or inhibits the growth of cells. Gene delivered to the cell may also protect the cells. Extracellularly, the cell releases the product of the new gene, which can act locally on neighboring cells or enter the circulation for delivery to distant cells [12]. The present study used the extracellular approach for delivery to distant lung grafts by
intramuscular transfection. The advantages of this intramuscular approach are: (1) the procedure is easy in the clinical setting; (2) the cytokine therapy can be performed simply; and (3) the graft, which was injured by storage, ischemia-reperfusion, and rejection, does not exhibit the inflammation usually associated with transfection. The adenovirus vector is known as a cause of local inflammation and the transfected gene may have local side effects such as the fibrosis due to the TGF-1. The graft, located at a distant site, is spared the local inflammatory events associated with transfection. Direct gene transfer into muscle was investigated by Wolff and colleagues using naked plasmid [13]. The nuclei of mature muscle do not undergo division; thus,
Ann Thorac Surg 2001;71:1651– 6
naked plasmid may be useful for this approach [5, 6, 14]. However, the disadvantage of naked plasmid is that relatively few cells take up the DNA (0.64% to 3%). Thus, only a small amount of the encoded protein is produced. In contrast, LacZ gene expression after intramuscular gene transfer with an adenoviral vector was observed in 21.8% of cells by immunohistochemical staining [12, 15]. Therefore, if a large amount of gene products is required to reach distant grafts, naked plasmid transfection might be not suitable. Adenovirus injected intramuscularly remained localized at the site of injection. We are investigating the organ distribution of transfection by different approaches, such as intravenous, airway, and intramuscular administration. The study showed that when the adenovirus was injected into skeletal muscle, the adenovirus was transfected into the muscle, spleen, and liver. However, the expression in liver and spleen was very weak and significantly lower than the intravenous transfection group (unpublished data). In the present study, successful gene transfection and expression was confirmed by active TGF-1 protein in muscle and plasma. However, the levels decreased by day 7. The duration of the effects from the adenovirus intramuscular approach has been shown to vary in previous studies using mouse, rat, rabbit, and primate experimental models [16, 17]. Because the duration of adenovirus vector is shortened by an antiviral cellular immune response to viral proteins expressed on the cell surface, this approach may be beneficial mainly in acute rejection or ischemiareperfusion injury in the transplant setting. Attenuation of lung allograft acute rejection as assessed by arterial oxygenation was noted after intramuscular administration of AdCMVTGF1. Transforming growth factors encompass polypeptides that regulate the growth and differentiation of both normal and transformed cells. One of these, TGF-1, has been isolated and purified from various normal and neoplastic cells [18]. TGF-1 is produced naturally as a latent 390-amino acid dimeric precursor that is converted into a mature bioactive 112-amino acid dimer after cleavage and dissociation of the amino-terminal portion, termed the latency associated peptide (LAP). This cleavage and dissociation can be readily achieved in vitro nonphysiologically by extremes of pH, temperature, and chaotropic agents. However, the activation mechanisms have not yet been fully elucidated in vivo, although thrombospondin and the proteases plasmin and cathepsin have been implicated. By using the active form of TGF-1 in the present study, it was easy to control the blood level of active TGF-1, which has a mature bioactivity. There is a possibility that if the overexpressed latent form TGF-1 is achieved by transfection, it may not be converted to enough of the active form of TGF-1 in vivo. H-E and EVG staining in the present study showed that transfection of the active form of TGF-1 induced severe fibrosis, granulation, and collagen deposition in the muscle where it was injected. Sime and colleagues demonstrated that transient overexpression of active, but not latent, TGF-1 resulted in prolonged and severe interstitial and pleural fibrosis characterized by extensive deposition of extracellular
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
1655
matrix proteins collagen, fibronectin, and elastin, and by emergence of cells with the myofibroblast phenotype after active TGF-1 transfection intratracheally [19]. In regard to prevention of side effects such as fibrosis, because the latent form of TGF-1 can be regulated in vivo, transfection of latent rather than active TGF-1 may provide a better route to deliver this powerful immunosuppressive agent in vivo. Therefore, further investigation regarding comparison of latent and active TGF-1 using this approach is warranted. In terms of the rejection score, it was not apparent that amelioration compared with the oxygenation. This suggests that histologic examination may be a crude method for analyzing the rejection process. However, histologic examination in the TGF-1-transfected group showed mild to moderate acute rejection. If cotransfer of immunosuppressive cytokine genes such as interleukin-10 are used, more efficient reduction in acute rejection may be accomplished. The goal of our study was to verify the applicability of gene therapy by intramuscular injection of adenovirus encoding for the TGF-1 in an acute rejection model. We chose to deliver the adenovirus well in advance of transplant in order to have optimal transgene expression at the time of lung allograft implantation. Although the time frame of 48 hours does not mimic current clinical settings, our choice was based on data in the preliminary experiments described in this report. In previous studies in our laboratory, we have documented amelioration of acute rejection by ex vivo TGF-1 lung allograft transfection at the time of harvest by using nonviral vectors and other delivery routes [20]. We plan a follow-up study on the best timing of gene therapy administration of TGF-1, whether before harvest, at reperfusion, or in conjunction with the onset of acute rejection, thus aiming the therapy as a precise prophylactic or rescue strategy. In conclusion, active TGF-1 intramuscular gene transfer into recipients has beneficial effects as a cytokine therapy in the rat lung transplantation setting. This intramuscular transfection approach is feasible in transplantation of other organs and may be useful for not only reducing graft rejection but also reperfusion injury. We thank Dr Debra A. Hullett (Department of Surgery, University of Wisconsin, Madison, WI) for kindly providing adenovirus encoding TGF-1. We also thank Kathleen Grapperhaus for technical assistance, Dawn Schuessler and Mary Ann Kelly for secretarial support, and Diane Toeniskoetter for her assistance. Statistical advice was obtained from Richard B. Schuessler, PhD. This work was supported by National Institutes of Health (NIH) grant 1 R01 HL-41281; T.M. is supported by NIH grant HL-56643.
References 1. Hullett DA. Gene therapy in transplantation. J Heart Lung Transplant 1996;15:857– 62. 2. Suda T, Mora BN, D’Ovidio F, et al. In vivo adenovirusmediated endothelial nitric oxide synthase gene transfer ameliorates lung allograft ischemia-reperfusion injury. J Thorac Cardiovasc Surg 2000;119:297–304. 3. Yano M, Mora BN, Ritter JM, et al. Ex vivo transfection of transforming growth factor 1 gene to pulmonary artery
1656
4.
5. 6. 7. 8. 9.
10. 11.
12.
SUDA ET AL INTRAMUSCULAR GENE TRANSFER OF ACTIVE TGF-1
segments in lung grafts. J Thorac Cardiovasc Surg 1999;117: 705–13. Boehler A, Chamberlain D, Xing Z, et al. Adenovirusmediated interleukin-10 gene transfer inhibits posttransplant fibrous airway obliteration in an animal model of bronchiolitis obliterans. Hum Gene Ther 1998;9:541–51. Geissler EK, Wang J, Fechner JH, Burlingham WJ, Knechtle SJ. Immunity to MHC class I antigen after direct DNA transfer into skeletal muscle. J Immunol 1994;152:413–21. Zhai Y, Hong X, Wang J, et al. Modulation of alloimmunity to major histocompatibility complex class I by cotransfer of cytokine genes in vivo. Transpl Immunol 1998;6:169 –175. Qin L, Chavin KD, Ding Y, et al. Gene transfer for transplantation: Prolongation of allograft survival with transforming growth factor-1. Ann Surg 1994;220:508–19. Ruscetti FW, Palladino MA. Transforming growth factor-beta and the immune system. Prog Growth Fact Res 1991;3:159–75. Rook AH, Kehrl JH, Wakefield LM, et al. Effects of transforming growth factor-beta on the functions of natural killer cells: depressed cytolitic activity blunting of interferon responsiveness. J Immunol 1989;136:3916–20. Wahl SM, McCarteney-Francis N, Mergenhagen SE. Inflammatory and immunomodulatory roles of TGF-1. Immunol Today 1989;10:258– 61. Yousem SA, Berry GJ, Cagle PT, et al. Revision of the 1990 working formulation for classification of pulmonary allograft rejection: Lung Rejection Study Group. J Heart Lung Transplant 1996;15:1–15. Blau HM, Springer ML. Gene therapy—A novel form of drug delivery. N Engl J Med 1995;2:1204– 07.
Ann Thorac Surg 2001;71:1651– 6
13. Wolff JA, Malone RW, Williams P, et al. Direct gene transfer into mouse muscle in vivo. Science 1990;247:1465– 8. 14. Donnelly JJ, Ulmer JB, Liu MA. DNA vaccines. Life Sci 1997; 60:163–72. 15. Couffinhal T, Kearney M, Sullivan A, Silver M, Tsurumi Y, Isner JM. Histchemical staining following LacZ gene transfer underestimates transfection efficiency. Hum Gene Ther 1997; 8:929–34. 16. Tripathy SK, Goldwasser E, Lu MM, Barr E, Leiden JM. Stable delivery of physiologic levels of recombinant erythropoietin to systemic circulation by intramuscular injection of replication-defective adenovirus. Proc Natl Acad Sci USA 1994;91:11557– 61. 17. Svensson EC, Black HB, Dugger DL, et al. Long-term erythropoietin expression in rodents and non human primates following intramuscular injection of a replication-defective adenoviral vector. Hum Gene Ther 1997;8:1797– 806. 18. Wallick SC, Figari IS, Morris RE, Levinson AD, Palladino MA. Immunoregulatory role of transforming growth factor  (TGF-) in development of killer cells: comparison of active and latent TGF-1. J Exp Med 1990;172:1777– 84. 19. Sime PJ, Xing Z, Graham FL, Csaky KG, Gauldie J. Adenovector-mediated gene transfer of active transforming growth factor-1 induces prolonged severe fibrosis in rat lung. J Clin Invest 1997;100:768–76. 20. Boasquevisque CHR, Mora BN, Boglione M, et al. Liposomemediated gene transfer in rat lung transplantation: a comparison between the in vivo and ex vivo approaches. J Thorac Cardiovasc Surg 1999;117:8–15.
DISCUSSION DR JOSEPH B. ZWISCHENBERGER (Galveston, TX): In your abstract you presented your Pao2 data.
DR ZWISCHENBERGER: How do you think TGF-, the growth factor, impacted directly on lung function?
DR SUDA: Yes. DR ZWISCHENBERGER: Was this study on 100% oxygen? DR SUDA: Yes. DR ZWISCHENBERGER: You performed a statistical comparison on partial pressure of oxygen in the blood. In terms of calculated oxygen content due to the saturation oxygen in the plasma, there is virtually no difference between those three numbers. I would suggest that you do an estimated shunt fraction where you’ll find that your data show that you have a 30% shunt, a 27% shunt, but in your treated group you have a 17% shunt. I would encourage you to include that in your manuscript. Is there any proposed mechanism of this improvement? DR SUDA: Regarding mechanisms, we did not investigate mechanisms in this study. We currently are investigating mechanism of this approach using TGF-1, however.
DR SUDA: TGF-1 alters the cytokine balance favoring an antiinflammatory environment and inhibits chemokine induced T lymphocyte recruitment and proliferation. Furthermore, other experiments found in the literature indicate that TGF-1 contributes to a TH2 type immune response, which is responsible for improved graft survival.
DR HARVEY I. PASS (Detroit, MI): Why do you have to use gene therapy? Why can’t you use recombinant TGF- and do the same thing? There may be differences with regard to lymphocyte recruitment depending upon whether you use DNA versus protein. Have you done protein comparisons?
DR SUDA: Because the half-life of TGF-1 is very short, it is very difficult to use recombinant TGF-1 using this approach.