| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2002;124:259-269
© 2002 The American Association for Thoracic Surgery
Cardiothoracic Transplantation (TX) |
From the Division of Cardiothoracic Surgery,a Department of Pathology and Immunology,b Washington University School of Medicine, St Louis Mo; Università degli Studi di Bologna,c Bologna, Italy; and Genzyme Co,d Framingham Mass.
This work was supported by National Institutes of Health (NIH) grant 1 R01 HL-41281. Dr T. Mohanakumar is supported by NIH grant HL-56643. Dr S. A. Kanaan is supported by individual NRSA-NIH grant 1 F32 HL-68401-01.
Received for publication Sept 14, 2001. Revisions requested Nov 19, 2001; revisions received Dec 6, 2001. Accepted for publication Dec 7, 2001. Address for reprints: G. Alexander Patterson, MD, 3108 Queeny Tower, One Barnes-Jewish Hospital Plaza, St Louis, MO 63110 (E-mail: pattersona{at}msnotes.wustl.edu).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
, interferon 
, and interleukin 2) were measured, and immunohistochemistry was performed. 
.03). The wet/dry ratio, reflecting lung edema, was reduced in the cotransfected group compared with that in control animals (nontreated, P < .02; chloramphenicol acetyltransferase, P < .03; chloramphenicol acetyltransferase plus ß-galactosidase, P < .01). Myeloperoxidase, which measures neutrophil sequestration, was also reduced with cotransfection compared with that seen in control animals (P .03). All proinflammatory cytokines were decreased in the cotransfected group compared with those in all other groups (interleukin 1ß, P < .04; tumor necrosis factor , P < .002; interferon , P < .0001; interleukin 2, P < .03). These results indicate that cotransfection provides a synergistic benefit in graft function versus either cytokine alone, neutrophil sequestration, or inflammatory cytokine expression. Immunohistochemistry showed positive staining of transforming growth factor ß1 plus interleukin 10 in type I and II pneumocytes and localized edema fluid. |
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
1 and malignant tumors. 2 It also represents a new modality to modulate ischemia-reperfusion (I/R) injury in lung transplantation. Severe graft dysfunction caused by I/R injury occurs in 10% to 20% of lung transplant recipients and results in progressive hypoxemia, decreased pulmonary compliance, and high permeability pulmonary edema, which often are associated with a need for prolonged intensive care. 3,4 I/R injury leading to severe graft dysfunction is also the most common cause for mortality within the first month after lung transplantation. 3
The inflammatory cascade of I/R injury is a dynamic combination of events. During the period of ischemia before reperfusion, the classical complement pathway is activated, and macrophages are attractedand release preformed cytokines, such as tumor necrosis factor (TNF-). 5 These cytokines in turn increase the expression of neutrophil adhesion molecules on endothelial cells. 5,6 Neutrophils then contribute to I/R injury by generating oxygen-derived free radicals, releasing proteases, and secreting arachidonic acid metabolites and other proinflammatory mediators. 7
Interleukin 10 (IL-10) is a potent inhibitor of inflammatory cytokine production. It is normally produced by macrophages and T lymphocytes. 8 Two targets of IL-10 inhibition are TNF- and interferon (IFN-. Because macrophage cytotoxicity is induced by IFN- and requires TNF- as a costimulatory signal, IL-10 modulates macrophage activity by decreasing TNF-. 9 Furthermore, IL-10 strongly inhibits natural killer cells that produce IFN-, leading to decreased IFN- production. Thus, by decreasing TNF- and IFN-, IL-10 reduces macrophage cytotoxicity and plays a role in decreasing I/R injury. 9
Transforming growth factor ß1 (TGF-ß1) is a pleiotropic protein with unique and potent immunoregulatory properties produced by several cells, including lymphocytes and macrophages. 10 It works in concert with IL-10 to block the activation of macrophages and T helper cells. 11,12 For example, both TGF-ß1 and IL-10 enhance production of the interleukin 1 (IL-1) receptor antagonist, a potent anti-inflammatory mediator. 10
Several approaches using the delivery of gene vectors have been successful by producing transgene expression in experimental studies for lung transplantation. 13-19 Transfection of viral vectors encoding heat shock protein 70, IL-10, and endothelial nitric oxide synthase has been shown to decrease lung graft I/R injury. 14-16 Intramuscular injection of nonviral vectors encoding TGF-ß1 has also been shown to attenuate the proinflammatory response in I/R injury. 17
In this study we demonstrate successful transgene expression after intramuscular naked plasmid gene transfer of viral IL-10 (vIL-10) and TGF-ß1 and show that their cotransfection results in amelioration of lung graft I/R injury.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
Plasmids
The plasmid viral IL-10 (pCMVievIL-10), kindly provided by Dr Jonathan Bromberg (Mount Sinai School of Medicine, New York, NY), consists of a human cytomegalovirus immediate-early promoter and enhancer followed by the tripartite leader from Epstein-Barr virus, a hybrid intron constituted by IL-10 (BCRF1-vIL10), and a polyadenylation signal from bovine growth hormone. Endotoxin units tested were lower than 1 EU/mL (EU < 1).
The plasmid TGF-ß1 (pMP6A-TGFß1-Active) was created in our laboratory, encoding the rodent active fragment of TGF-ß1 by mutated cysteine in serine in position 223 and 225 of the TGF-ß1 cDNA. The mutated TGF-ß1 cDNA, kindly provided by Dr Debra A. Hullett (Department of Surgery, University of Wisconsin, Madison, Wis), was amplified by means of polymerase chain reaction with Ampli-TAQ polymerase (Perkin Elmer Applied Biosystems, Foster City, Calif) from an adenovirus encoding his sequence. Primers for the amplification were constructed with NheI terminals. The amplified cDNA was then cloned in the NheI cloning site of the adeno-associated plasmid pMP6A with a cytomegalovirus initial promoter and an SV40 polyadentylation signal sequence, also kindly provided by Dr Jonathan Bromberg. Correct orientation was checked by means of BamHI digestion. Adequate primers were then constructed for the sequencing of the mutated TGF-ß1 cDNA by the Protein Chemistry and Nucleic Acid Laboratory at Washington University. Purity of the plasmid after mega preps (Endo-free Mega Kit; Quiagen Inc, Valencia, Calif) was tested at Charles River Laboratories (Charleston, SC), and the endotoxin units were approximately 10 EU/mL (EU > 10).
Study design
Nontransplant setting
Thirty-six rats divided into 4 groups (n = 3 each for 3 time points) received 1000 µg of vIL-10, mutated TGF-ß1, or both by means of separate injections of 1 mL in each gastrocnemius muscle. Plasmids were suspended in 0.9% saline solution, and control groups were injected with 1 mL of saline solution.
All groups were killed 24, 48, or 168 hours after treatment. Muscles were immediately frozen in liquid nitrogen at -70°C. Blood was collected and centrifuged at 1000g for 30 minutes at 4°C. The plasma obtained was divided into 2 vials: one was stored at -70°C for enzyme-linked immunosorbent assay (ELISA) analysis, and the other underwent a second cycle of centrifugation (15,000g for 15 minutes) for TGF-ß1 expression measurement. 20
Transplant setting
Thirty-six rats divided into 6 groups (n = 6 each) received intramuscular injection in the gastrocnemius muscle of 1000 µg of plasmid diluted in 1 mL of saline solution with chloramphenicol acetyl transferase (CAT); CAT and ß-galactosidase (ß-Gal); or vIL-10, TGF-ß1, or vIL-10 and TGF-ß1. CAT and ß-Gal are nonfunctional reporter genes used as plasmid controls. One more control group was not treated with any vector. Cotransfection groups used separate intramuscular injections in contralateral muscles.
Eighteen hours before transplantation, donors were harvested as previously described. 14-16 In brief, donors were anesthetized with intraperitoneal pentobarbital of 0.3 mL (30 mg/kg), intubated with a 14-gauge catheter by means of tracheotomy, and mechanically ventilated (model 683; Harvard Apparatus Co, South Natick, Mass) with room air. Lungs were flushed with 20 mL of cold low potassium dextran 1% glucose solution through the main pulmonary artery. The heart-lung block was removed, and the pulmonary artery, pulmonary vein, and left main bronchus were prepared with polyethylene cuffs and preserved at 4°C for 18 hours.
Forty-eight hours after intramuscular treatment, the recipients underwent transplantation. After anesthesia with subcutaneous injection of ketamine (25 mg/kg), atropine (30 mg/kg), and halothane inhalation (0.5%-1%) from orotracheal intubation, a left thoracotomy was performed. The left pulmonary vessels and main bronchus were anastomosed by the previously described cuff technique. 21 Ventilation and perfusion were restored, and a temporary drainage chest tube was placed, which was subsequently removed in the incubator (Intensive Care System; Thermocare Inc, Incline Village, Nev) after recovery from anesthesia.
Twenty-four hours after transplantation, the recipients were killed. Animals were reanesthetized by means of intraperitoneal pentobarbital injection and their lungs were ventilated with 100% oxygen. Median laparosternotomy was performed, and the right hilum was clamped. The animals' lungs were ventilated for 5 minutes by means of a tidal volume of 1.5 mL, peak end-expiratory pressure of 1 cm H2O, and rate of 100 breaths/min. Blood samples were then collected from the ascending aorta. The left lung was frozen in liquid nitrogen and stored in 3 separate sections: the upper section was used for ELISA; the middle section was used for myeloperoxidase (MPO) assay; and the lower section was weighed, dried at 70°C for 48 hours, and then reweighed for calculation of the wet/dry ratio (W/D). A portion of the native right lung of the donor was taken for comparison.
Six additional recipient rats were intramuscularly cotransfected with vIL-10 plus TGF-ß1 (n = 3) or not transfected (n = 3) and underwent the same transplant procedure as above for immunohistochemistry assessment. Their left lungs were flushed through the main pulmonary artery with saline solution at a pressure of 20 cm H2O and then fixed with Histochoice (AMRESCO, Solon, Ohio). No functional or efficacy data were obtained from these animals.
Assessment
MPO
Quantitative lung homogenate MPO activity as a measure of neutrophil sequestration was determined as previously described, 22 with some modification. 14 Optical density was measured at 460 nm with a spectrophotometer (Genesis 5; Spectronic Instruments, Inc, Rochester, NY). For each sample, the 5-minute reading was subtracted from the 20-minute reading and standardized to the total protein present in that sample (BCA Protein Assay Kit; Pierce, Rockford, Ill). MPO enzyme activity is expressed in units on the basis of the amount of protein required to effect a 
change of 1 optical density unit per minute at room temperature.
ELISA
All muscle and lung tissue studied was homogenized in lysis solution containing 100 mmol/L potassium phosphate (pH 7.8), 0.2% Triton X-100 with pepstatin A (5 µg/mL; Roche Molecular Biochemicals, Mannheim, Germany), and protease inhibitor cocktail (Complete mini, Roche Molecular Biochemicals). The homogenate was incubated for 15 minutes at room temperature and then centrifuged at 15,000 rpm for 15 minutes. The supernatant was collected and analyzed for proinflammatory cytokines. 18
TGF-ß1 protein activity in muscle and plasma was detected by means of ELISA according to the protocol of R&D Systems (Minneapolis, Minn) for human TGF-ß1, which is also cross-reactive to rat and mouse. The procedure to activate the latent form of TGF-ß1 in the ELISA kit was not used for the detection of active TGF-ß1.
The presence of vIL-10 was detected according to the BD Pharmingen (San Diego, Calif) cytokine and chemokine ELISA protocol, with few modifications. In brief, the purified rat monoclonal antibody for vIL-10 (BD Pharmingen) was diluted to 2 µg/mL in a ligand solution (0.1 mol/L Na2HPO4 at pH 9.0). After overnight incubation at 4°C in a 96-well plate (MaxiSorp Surface; Nalge Nunc Int, Rochester, NY), the wells were incubated with blocking buffer (1% bovine serum in phosphate-buffered saline solution [PBS]) for 2 hours to prevent nonspecific binding. After washing with PBS-Tween, all recombinant vIL-10 standards (BD Pharmingen) and the samples were added for 3 hours. After washing with PBS-Tween-20, the wells were incubated with biotinylated anti-vIL-10 antibody (BD Pharmingen) and streptavidin-horseradish peroxidase. The colorimetric reaction was activated by the substrate solution (2,2'-azino-bis-3-etilenbenzotazolin-6-sulfonic acid) and read after 40 minutes (optical density, 405 nm).
The quantitative expression of proinflammatory cytokines (IL-1ß, IL-2, TNF-, and IFN-) was performed according to the protocols of R&D Systems kits.
All samples were standardized to the total protein with the BCA kit (Pierce).
Immunohistochemistry
Immunohistochemistry, as previously described, 23 was performed according to the protocol of the Tyramide Signal Amplification kit (NEN Life Science Products, Boston, Mass). In brief, slides were incubated for 4 hours at room temperature with biotinylated rat anti-mouse/anti-human TGF-ß1 antibody (BD Pharmingen) at 1:40 dilution in TNB buffer (Blocking Reagent in 100 mmol/L Tris and 500 mmol/L NaCl, pH 7.4) and 0.1% saponin. Slides were washed in TBS (100 mmol/L Tris and 500 mmol/L NaCl/0.2% Triton X, pH 7.4) and 0.1% saponin. Slides were then incubated with streptavidin-horseradish peroxidase for 30 minutes, followed by biotinyl tyramide for 10 minutes. Finally, slides were incubated for 30 minutes with streptavidin-alkaline phosphatase. Chromogenic detection for TGF-ß1 was performed with Vector Blue Alkaline Phosphatase Substrate (Vector Laboratories, Burlingame, Calif) in 100 mmol/L Tris, pH 8.2, including 5 mmol/L levamisole and 50 µL of Tween-20. For subsequent double staining to detect vIL-10, slides were washed with deionized water. The blocking step in Super Blocking Buffer was performed. Slides were incubated overnight at room temperature with biotinylated rat antiviral IL-10 antibody (BD Pharmingen) diluted 1:20 in TNB Buffer/0.1% saponin. The detection method was performed as stated above, except that streptavidin-horseradish peroxidase was substituted for streptavidin-alkaline phosphatase in the final step. Slides were developed for 5 minutes with Vector NovaRed Substrate (Vector Laboratories) and transferred to TE buffer (tris-ethylenediamine tetraacetic acid) at pH 8.0 to stop the reaction. Slides were counterstained with methyl green nuclear (Vector Laboratories) for 5 minutes and permanently mounted with Cytoseal 60 (Stephens Scientific, Kalamazoo, Mich).
The pattern of protein expression and histology using hematoxylin and eosin staining was assessed by a blinded observer (J.H.R.).
Statistics
Parametric data were analyzed by means of 1-way analysis of variance and the Fisher post hoc multiple comparison test. Values are reported as means ± SD. Data not normally distributed were converted to a square root transformation before the analysis of variance was performed.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
, B). On the basis of these results, 48 hours before transplant was chosen as the time point for intramuscular injection of vIL-10 naked plasmid in the I/R injury model.
|
, C) after 24 and 48 hours was significantly increased in the cotransfected groups compared with that seen in control animals (P < .0001) or that seen in groups transfected with TGF-ß1 alone (P = .0086 to 24 hours and P = .0027 to 48 hours). Similarly, in plasma (Fig 1, D) significant differences were observed between cotransfected groups and control animals (24 and 48 hours, P < .0001) or with groups transfected with TGF-ß1 alone (24 and 48 hours, P < .0001). As with vIL-10, these results support selecting 48 hours before transplant as the optimum intramuscular injection time for TGF-ß1 in the I/R injury model. Interestingly, there were no differences in the TGF-ß1 levels between the 24- or 48-hour cotransfected groups in both muscle and plasma.
Transplant setting
PaO2
Lungs from the cotransfected group had superior PaO2 levels (Table 1) compared with those of the nontreated control group (P = .0002), the CAT group (P < .0001), or the CAT plus ß-Gal group (P < .0001) and those of groups treated with TGF-ß1 alone (P = .0005) or vIL-10 alone (P = .011). Cotransfection provides a synergistic benefit versus either cytokine alone in graft function as measured by PaO2.
|
), was significantly lower among the TGF-ß1, vIL-10, vIL-10 plus TGF-ß1, and control groups (P = .03, P = .015, and P = .028, respectively). Furthermore, there was no difference between the above study groups and normal flushed nontransplanted lungs (P = .8). This indicates that the flushing process does not affect the W/D ratio, and cotransfection provides a benefit in reducing lung edema. However, cotransfection was not superior to either cytokine alone in reducing the edema.
MPO
MPO activity, which reflects neutrophil sequestration (Table 1), was decreased significantly in the cotransfected group versus that seen in the no treatment, CAT, or CAT plus ß-Gal groups (P = .0028, P = .0023, and P = .0017, respectively). Also, vIL-10 had significantly decreased MPO versus that seen in the no treatment, CAT, or CAT plus ß-Gal groups (P = .03, P = .034, and P = .026, respectively). These results indicate that vIL-10 and vIL-10 plus TGF-ß1 reduce neutrophil sequestration but not TGF-ß1 alone.
Proinflammatory cytokines
For this analysis, we added 2 groups: flushed normal nontransplanted lungs and flushed transplanted lungs after 1 hour of ischemia at 4°C (1 hour, no transfection).
IL-1ß levels (Table 2 and Fig 2, A) in the cotransfected group were significantly lower compared with those in all other control and study groups (1 hour, no transfection, P = .0039; 18 hours, no transfection, P = .0063; CAT, P = .0073; CAT plus ß-Gal, P = .0021; TGF-ß1, P = .0021; and vIL-10, P = .022). However, transfection with either cytokine alone did not significantly reduce the IL-1ß levels. Also, no difference was noted between cotransfected and normal nontransplanted lungs (P = .688).
|
|
(Table 2 and Fig 2, B) was undetectable in normal nontransplanted lungs. vIL-10 significantly reduced TNF- levels versus those in the 18 hours with no transfection, CAT, and CAT plus ß-Gal groups (P = .024, P = .012, and P = .011, respectively), whereas TGF-ß1 significantly reduced TNF- levels versus those seen in the CAT and CAT plus ß-Gal groups (P = .039 and P = .034, respectively). Each cytokine was able to reduce the TNF- expression, but cotransfection synergistically lowered the TNF- expression compared with that seen in all control and study groups (P < .0001 for all groups).
IFN- (Table 2 and Fig 2, C) was also undetectable in normal nontransplanted lungs and was not significantly different from levels in the cotransfected group (P = .2). Like IL-1ß, cotransfection leads to a normal cytokine balance with respect to IFN-. In addition, the cotransfected group had a significantly lower IFN- level compared with all control and study groups (P < .0001 for all groups). TGF-ß1 was only significantly decreased compared with that in the CAT group (P < .0001), and vIL-10 was significantly decreased when compared with the TGF-ß1 group (P = .036), as well as the flushed transplanted lungs, CAT, and CAT plus ß-Gal control groups (P = .008, P < .0001, and P < .0001, respectively). This shows that IL-10 is effective in reducing IFN-, whereas TGF-ß1 is not. Yet the cotransfection results indicate that TGF-ß1, by possibly augmenting IL-10 levels, allows for a greater reduction in IFN- levels.
Last, IL-2 expression (Table 2 and Fig 2, D) was very low in the 3 study groups of TGF-ß1, vIL-10, and vIL-10 plus TGF-ß1 and was significantly reduced compared with that in the following control groups: 1 hour, no transfection; 18 hours, no transfection; CAT; and CAT + ß-Gal, respectively (P = .028, P = .0048, P = .0022, and P = .001 for TGF-ß1 and vIL-10 alone; P = .01, P = .001, P = .0007, and P = .0002 for vIL-10 plus TGF-ß1). There does not appear to be any synergism exhibited with cotransfection with respect to IL-2 levels, yet either cytokine alone is extremely effective in reducing IL-2.
Histology
Hematoxylin and eosin staining in cotransfected lungs showed moderate edema and mild interstitial neutrophil infiltration when compared with that seen in normal lungs (Fig 3, A). Severe edema and moderate interstitial infiltration of neutrophils with rare microabscesses and hyaline membranes appears in nontreated lungs (Fig 3, B).
|
, B) and some localized areas of type II cells (Fig 4, B). Also, most of the type I and II alveolar epithelial cells were stained positive for TGF-ß1 (Fig 4, A, for both and Fig 4, B, for type I).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
5 During reperfusion of ischemic tissue, the classical complement pathway is activated, and the potent chemotactic factors C5a and C1q are released. These binding proteins attract macrophages. Macrophage migration is also partially mediated by activated TGF-ß1 and macrophage chemoattractant protein 1 released from injured cells. 5 Macrophages then release the preformed cytokines TNF-, IL-1, and IL-6, which increase the expression of neutrophil adhesion molecules, such as intercellular adhesion molecule 1, on endothelial cells. 5,6 The chemokines IL-8 and macrophage inflammatory protein 1 also mediate neutrophil chemotaxis. The late phase of I/R injury occurs several hours after this inflammation begins. 24 Normally, neutrophils are more concentrated within the pulmonary capillaries than in systemic blood. 24,25 The inflammatory mediators, such as C5a, TNF-, and IL-8, cause neutrophils to become less deformable and foster pulmonary capillary sequestration. 25 The sequestered neutrophils adhere to the capillary endothelium, migrate into the lung parenchyma and air spaces, and contribute to I/R injury by generating oxygen-derived free radicals, releasing proteases, and secreting arachidonic acid metabolites and other proinflammatory mediators. 7,25
In this study we evaluated the effects of IL-10 and TGF-ß1 and their anti-inflammatory role in lung transplant I/R injury. Cellular IL-10 is regularly produced during inflammatory processes by monocytes, macrophages, and T lymphocytes. 8 It inhibits the production of proinflammatory cytokines, such as IL-1, IL-2, IL-6, IL-12, and TNF-, and inhibits natural killer cell production of IFN-. Conversely, IL-10 might have an important role in the differentiation of mastocytes. 8,26 Human IL-10, similar to cellular IL-10, has anti-inflammatory effects. 26,27 Viral IL-10 is structurally similar to cellular IL-10 in the sequence encoding the mature protein. It also inhibits the activity of the cytokines and downregulates the expression of major histocompatibility complex II on the surface of monocytes. Because vIL-10 does not possess the costimulatory effect of cellular IL-10 on mastocytes, it is an effective immunosuppressant agent. 12,26,28,29
Similar to IL-10, TGF-ß1 plays an important role in the modulation of the inflammatory process and in several other processes, such as embryonic development, tumorigenesis, wound healing, and fibrosis. 30 Its immunosuppressive activity is related to antagonizing inflammatory cytokines, such as IL-1, TNF-, and IFN-. 12,29 Gene transfer strategies of IL-10 and TGF-ß1 are important because of their short half-life when administered in vivo. Human studies have demonstrated a half-life of approximately 2 to 5 hours when IL-10 is given intravenously or subcutaneously. 31 Animal studies show a half-life of 100 minutes for the latent form of TGF-ß1 compared with 2 to 3 minutes for the active form. 19 Conversely, a single application of IL-10 or TGF-ß1 gene transfer might provide significant protein expression to reduce I/R injury. Furthermore, unlike other types of injury, gene transfer in experimental and clinical lung transplantation can provide cytokine expression before the onset of I/R injury.
The naked plasmid vector was chosen for gene delivery in our model for several reasons. We previously reported that muscle expression of gene transfer products using adenoviral vectors might enter the bloodstream and have beneficial effects in acute lung rejection. 32 The viral vector has a higher capability to transfect cells compared with that of naked plasmid but might affect the pretransplant and posttransplant recovery secondary to the possible inflammatory side effects and the potential immunologic response of the recipient to the vector. Despite the shortcomings of lower expression, lower transfection efficiency, and high concentration necessary to have a beneficial response, the advantages of using plasmids are (1) no systemic toxicity and consequently no side effects, 33-35 (2) the possibility of enhancing gene transfer efficiency and transgene expression with several injections, 29 (3) relatively lower cost, and (4) easy applicability in the clinical setting by means of simple intramuscular administration. We have shown (unpublished data), using intramuscular injection of CAT naked plasmid, that the gastrocnemius muscle was markedly transfected compared with the heart, lung, kidney, spleen, or liver. This supports that naked plasmid use produces no systemic toxicity or side effects. Also, we found elevated CAT protein expression in muscle for 7 days. This supports the advantage of a single intramuscular injection of naked plasmid over intravenous or pharmacologic delivery of cytokines.
In the expression phase of the study, the data show that at least 48 hours after injection are required for both anti-inflammatory cytokines to be sufficiently present in the rodent system. We also found a marked increase in TGF-ß1 expression in the cotransfected groups compared with in the groups treated with either cytokine alone. This suggests that vIL-10 augments TGF-ß1 muscle and plasma expression. We also found that TGF-ß1 expression was significantly increased after both 24 and 48 hours after injection. This implies that one could perform intramuscular injection of TGF-ß1 and vIL-10 24 hours before transplantation and possibly obtain similar results with respect to reducing I/R injury as those found when injections are performed 48 hours before transplantation. Also, this suggests that cotransfection allows for earlier recipient plasmid administration because the delivery of 2 agents will augment and expedite each other's expression compared with the delivery of a single agent.
In contrast, IL-10 expression 24 hours after injection was not significantly increased compared with that 48 hours after injection. The main factor that might decrease exogenous IL-10 expression 24 hours after injection is the low transfection efficiency in muscle fibers. Another factor is related to the short systemic half-life of IL-10. In contrast, TGF-ß1 has a very short life, 32 but its plasma levels correlate with muscle levels, leading us to believe that what is being produced in muscle is remaining intact in plasma. This could be due to the large volume of plasmid used that creates damage to the muscle fibers and allows for greater transfection efficiency. 27,36
Immunohistochemistry results show type I and II pneumocytes are seen in transplanted lungs stained for TGF-ß1 after only intramuscular injection of plasmid. Bellocq and colleagues 37 found that reactive oxygen intermediates might increase TGF-ß1 release from human epithelial alveolar cells in vitro. Therefore, the reperfusion period after transplantation might provide the stimulus for such oxygen intermediates to activate alveolar cells to release and stain for TGF-ß1 in our model. One hypothesis to explain this finding is that TGF-ß1 released systemically from muscle fibers might be taken up by type I or II pneumocytes. Another possibility is that inflammatory cells, such as macrophages, infiltrate the lung graft to produce either endogenous TGF-ß1 or to stimulate alveolar cells to produce TGF-ß1. 27 Wolff and colleagues 34 and Davis and coworkers 33 have demonstrated the efficacy of the naked plasmid for direct gene transfer into rodent skeletal muscle. However, the mechanisms for how plasmids enter muscle fibers 38 or how a cytokine might augment the activity of another cytokine remain unsolved. 35,36
Attenuation of lung I/R injury was noted after the intramuscular cotransfection treatment of these 2 potent cytokines. When the PaO2, MPO, and inflammatory cytokine results are examined, a synergistic benefit is seen with cotransfection. This probably reflects the fact that IL-10 and TGF-ß1 work at different points in the inflammatory pathway. 12,29 MPO was significantly reduced in the cotransfected and vIL-10 groups compared with the reduction in the control groups. Yet the reduction in MPO is more marked with cotransfection compared with use of vIL-10 alone. This leads one to believe that TGF-ß1 augments IL-10 expression, leading to increased IL-10 levels and accounting for the observed difference in cotransfected animals. This also implies that TGF-ß1 indirectly reduces MPO activity in the cotransfection group through its effect on IL-10 expression. Arterial oxygenation is also superior with cotransfection versus use of either cytokine alone. With IL-1ß, either cytokine alone was unable to significantly reduce its expression, but cotransfection was successful in returning the level to that of normal control lungs. Similarly for IFN-, vIL-10 and not TGF-ß1 reduced its levels, but cotransfection markedly reduced expression compare with that seen in normal lungs. This also supports the idea that TGF-ß1 augments IL-10 expression, accounting for the greater reduction in IFN- levels in cotransfected animals. The IL-2 level was significantly reduced with IL-10, TGF-ß1, and cotransfection. Interestingly, IL-2 expression was higher in normal flushed lungs compared with in the 3 study groups. This means that either the flushing process increased IL-2 or that IL-10, TGF-ß1, and cotransfection all effectively reduced IL-2 levels to below baseline.
In conclusion, recipient intramuscular gene transfer of IL-10 and TGF-ß1 attenuates I/R injury. The plasmid-mediated gene cotransfection achieved not only gene expression but also beneficial immunologic effects by providing a synergistic effect that significantly improved lung function after prolonged cold ischemia. Multiple gene transfer of anti-inflammatory cytokines might not only reduce I/R injury but also might have an effect on preventing acute and chronic rejection in the transplant setting. These possibilities will be examined in future studies.
|
Appendix: Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
I have a couple of questions. First, for clinical applications, how are you going to apply this to patients who only know 6 to 8 hours before transplant that they are going to get a graft? Second, have you considered putting IL-10 or TGF-ß into an adeno-associated virus, so that you could actually dose an experimental animal or an individual weeks or months ahead of time and still have the salutary effect when a donor organ was identified? Third, following up on that question, if a longer-term expression is desired or possible, are there any toxic effects of long-term overexpression?
Dr Daddi. Thank you Dr Smythe for your comments. In response to the first question, we agree that 48 hours before transplant for vector administration is not entirely useful for clinical application at this time. Two immediate uses for this approach are in living lobar lung transplantation and acute rejection. We are testing in our laboratory the effect of adenoviral vectors encoding TGF-ß and human IL-10 in the acute rejection setting. Preliminary data confirm the anti-inflammatory effect on the transplanted graft. Ideally, this will lead us to further investigations on the molecular mechanisms involved in lung transplantation.
For the second question, we have not used or considered using adeno-associated virus for long-term use in our model. It is a very good idea worth examining further.
For the third question regarding long-term problems with overexpression of these cytokines, we have not detected any to date. We confirm that compared with the first-generation adenoviral vectors, naked plasmid elicits lower plasma expression, yet in comparison, it is nontoxic to the recipient. We also reasonably think that because of the nontoxicity of this vector, it should be applicable in clinical trials involving organ transplantation. Giving several injections of naked plasmid alone or in combination with adenoviral vector administration might be a useful therapeutic instrument. Using the plasmids 1 to 2 days or perhaps 5 days after transplantation might benefit the lung graft, and we are studying some protocols about performing this. On the other hand, you asked me whether there are any side effects of the naked plasmid on animals. We saw that there was no side effect of the naked plasmid in the postoperative period in the study groups, and we noted no postoperative complications. No preoperative or postoperative mortality occurred. Morbidity was related to the control groups. Unfortunately, our I/R injury model does not give us the possibility to investigate the long-term expression. Data from similar expression work that we conducted on day 7 with intramuscular injection of naked plasmid encoding CAT shows systemically a very mild level of protein expression and no side effects. Your comment is an interesting one and should be investigated further.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendix: DiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. S. Wilkes A Breath of Fresh Air for Lung Transplant Recipients Science Translational Medicine, October 28, 2009; 1(4): 4ps5 - 4ps5. [Full Text] [PDF]
|
|
|
|
 |
|
 P. L. Hermonat, D. Li, B. Yang, and J. L. Mehta Mechanism of action and delivery possibilities for TGF{beta}1 in the treatment of myocardial ischemia Cardiovasc Res, May 1, 2007; 74(2): 235 - 243. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Suda, N. Daddi, T. Tagawa, S. A. Kanaan, B. D. Kozower, J. H. Ritter, and G. A. Patterson Recipient intramuscular cotransfection of transforming growth factor {beta}1 and interleukin 10 ameliorates acute lung graft rejection J. Thorac. Cardiovasc. Surg., April 1, 2005; 129(4): 926 - 931. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Tagawa, B. D. Kozower, S. A. Kanaan, N. Daddi, M. Muraoka, T. Oka, J. H. Ritter, and G. A. Patterson Gene transfer of tumor necrosis factor inhibitor improves the function of lung allografts J. Thorac. Cardiovasc. Surg., June 1, 2004; 127(6): 1558 - 1563. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Waldow, K. Alexiou, W. Witt, F. M. Wagner, V. Gulielmos, K. Matschke, and M. Knaut Attenuation of Reperfusion-Induced Systemic Inflammation by Preconditioning With Nitric Oxide in an In Situ Porcine Model of Normothermic Lung Ischemia Chest, June 1, 2004; 125(6): 2253 - 2259. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
