| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1998;116:386-390
© 1998 Mosby, Inc.
Cardiothoracic Transplantation |
From the Departments of Cardiothoracic Surgerya andPathology,b Stanford University School of Medicine, FalkCardiovascular Research Building, Stanford, Calif.
Presented at the Thirty-third Annual Meeting of The American Societyof Transplant Physicians, Chicago, Ill., May 2-5, 1997.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 and results in a diffuse coronaryvascular narrowing termed chronic graft vascular disease(CGVD). Current immunosuppressants, although effective at controlling acutealloimmune responses, have not altered the devastating impact of CGVD onlong-term graft survival. Ex vivo somatic gene therapy has been proposed as anovel, highly targeted strategy of genetically manipulating donor allografts toprevent CGVD. Unlike genetic engineering applications such as cloning, somaticgene therapy has yet to prove clinically useful, primarily because ofconsistently poor transfection efficiency.
We 2,3 have used hydrostatic pressure as anovel, technically simple, nontoxic vector for nuclear delivery ofphosphorothioate oligodeoxynucleotides (ODN) to a variety of tissues. Theconsistently high transfection efficiency of this method in cardiac tissuesraises the intriguing possibility of using antisense technology for cardiacallograft immunosuppression. In the present studies, phosphorothioate antisenseoligonucleotides (AS ODN) that bind specifically to the target messengerribonucleic acid (mRNA) of intercellular adhesion molecule–1 (ICAM-1) weredelivered to cardiac allografts ex vivo by means of this hyperbaric vector.ICAM-1 was chosen as the antisense target because its expression is solelytranscriptionally regulated 4and low at baseline in myocardium. 5Furthermore, the central role that ICAM-1 maintains in allograft pathophysiologymakes it an ideal immunosuppressive target.
Specific blockade of this molecule interrupts the positive feedbackcascade of ICAM-1 with its ligand, leukocyte function associated antigen–1(LFA-1), and results in reduced neutrophil-mediated reperfusion injury 6 and donor-specific allografttolerance. 7,8 It has been shown that earlynonspecific inflammation from reperfusion injury up-regulates graftimmunogenicity and therefore increases the rates of acute 9 and chronic rejection. 10 In addition to indirectly promotingCGVD via reperfusion injury, clinical 11and experimental 12 models haverevealed a second, later phase of ICAM-1 up-regulation that coincides with thehistologic onset of vasculopathy, suggesting a more direct pathogenic role ofICAM-1 in development of CGVD.
Regardless of the exact mechanism of effect, we hypothesized that the exvivo blockade of ICAM-1 in donor grafts would reduce the severity ofvasculopathy in a rodent cardiac allograft model of CGVD.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
8 mouse AS ICAM-1 ODN using informationobtained from GenBank regarding the differences of ratR386013"> 13 and mouseR386014"> 14 ICAM-1 mRNA at the homologous 3'untranslated target region:(bold type represents bases that differ betweenmouse and rat ICAM-1 mRNA and AS at this region).
For studies to address the efficiency of transfection, ODN with ascrambled sequence was synthesized and labeled on the 5' and 3' endswith fluorescein isothiocyanate (FITC). For investigation of the true antisensemechanism of AS ODN, control reverse antisense (RAS) sequences were synthesized.LFA-1 and ICAM-1 monoclonal antibodies (MAb) were purified from the ascites ofmice injected with hybridomas (WT.1 and 1A29, gift of M. Miyasaka, Osaka, Japan)and administered via the dorsal penile vein at 1.5 mg/kg per day frompostoperative day 0 to day 6. This dose was confirmed to promote full saturationof their respective ligands on peripheral blood leukocytes 24 hours afteradministration by flow cytometric analysis. Cyclosporine A (INN: ciclosporin)was dissolved in olive oil (10 mg/ml) and administered via gavage at a dose of 5mg/kg for 10 days after the operation to prevent acute rejection and allow forthe development of CGVD in all groups.
Animals
Adult male (8 to 10 weeks old, 230 to 270 gm) PVG (RT1c) and ACI (RT1a)rats were obtained from Harlan Sprague Dawley, Indianapolis, Indiana. Allanimals were maintained in the animal care facilities of the Department ofCardiothoracic Surgery, Stanford University Medical Center, Stanford,California. Their environment was maintained at 21° ± 2° C with atime-regulated light period from 7:00 am to 7:00pm. Rats were provided water and dry food adlibitum. Periodic serologic analysis of room sentinel animals showed that allrats were free of acute viral infection. All animals received humane care incompliance with the "Principles of Laboratory Animal Care"formulated by the National Society for Medical Research and the "Guide forthe Care and Use of Laboratory Animals" prepared by the National Academyof Sciences and published by the National Institutes of Health (NIH PublicationNo. 86-23, revised 1985).
Heart transplantation
Both donor and recipient rats were anesthetized with methoxyflurane(inhalational) and sodium pentobarbital (50 mg/kg intraperitoneally). Heartswere procured and either underwent transfection or mock transfection asdescribed below. After incubation, hearts were grafted heterotopically into theabdomen of allogenic recipients by means of a modification of the methodsdescribed by Ono and Lindsay. 14aImmediately after reperfusion and daily thereafter, grafts were assessed forsuccessful return of rhythmic cardiac contractions on a scale ranging from 0 (nocontractions) to 4 (vigorous contractions). Primary failure was defined ingrafts that did not achieve an immediate palpation score greater than 2. Afteran initial period of successful function, hearts were considered acutelyrejected when palpation scores were less than 1. Grafts with either of thesediagnoses were excluded from further analysis.
In vivo transfection efficiency
After donor rat thoracotomy, ice cold Stanford cardioplegic solution wasapplied to the heart topically and infused into the aortic root proximal to anaortic crossclamp for coronary perfusion. After complete cardioplegic arrest,the coronary arteries were perfused with FITC-tagged scrambled ODN solution (0.5ml of 80 µmol/L ODN in phosphate-buffered saline solution at l ml/min).The heart was then explanted, placed into a well of ODN solution in an ice bathat 4° C, and pressurized to 3 atm in a custom-designed pressure vessel for45 minutes. Donor hearts treated with the above protocol without the use ofpressure served as the control group. After 24 hours' reperfusion in therecipient, grafts were procured, flushed with ice-cold phosphate-buffered salinesolution, and immediately snap frozen in OCT embedding compound in liquidnitrogen. After equilibration to –20° C, 6 µm sections weretaken perpendicular to the long axis of the heart from the superior, midportion,and inferior portions of the grafts. Hoechst dye was used to facilitateidentification of total nuclei under fluorescent microscopy, and transfectionefficiency was assessed by determining the percentage of myocardial nuclearlocalization of fluorescence in 5 random high-power (x100) fields chosenfrom each of three graft regions (superior, midportion, and inferior) for atotal of 15 high-power fields per graft.
Verification of transfection efficacy
After confirmation of transfection efficiency, we then verified that thedelivery of AS ODN by means of the hyperbaric method would successfully preventICAM-1 up-regulation in the PVG to ACI model of acute rejection. On the basis ofpilot studies of ICAM-1 protein expression analyzed in control PVG allografts byWestern blot and immunohistochemistry, postoperative day 3 was chosen as thetime point most appropriate for comparison with AS ODN transfected grafts. PVGgrafts were transfected with AS ICAM-1 ODN or RAS ODN and heterotopicallytransplanted into the abdomen of untreated ACI recipients. Hearts were procuredon postoperative day 3 for immunohistochemical assessment of ICAM-1, CD4, CD8,macrophage, and major histocompatibility complex (MHC) classes I and II usingthe method outlined in the Histostain SP kit (Zymed Laboratories, South SanFrancisco, Calif.). In brief, 6 µm sections were air dried at roomtemperature and fixed in acetone at 20° C. Sections were rehydrated in 1%bovine albumin–phosphate buffered saline solution and then incubated withthe one of the following primary antibodies (from Serotec, Westbury, N.Y.): 1A29(ICAM-1), W3/25 (CD4), MRC OX-8 (CD8), ED1 and ED2 (macrophage), 156 and 280(MHC class I), and 46 (MHC class II). This was followed by incubation with abiotinylated goat antimouse immunoglobulin G (Zymed Laboratories, South SanFrancisco, Calif.). The avidin-biotin complex was applied and diaminobenzidinetetrahydrochloride was used as the chromogen. The substitution of 1%bovine albumin–phosphate buffered saline solution for the primary antibodyserved as the negative (reagent) control. Rat cervical lymph nodes served as theICAM-1 positive control. Sections from transfected and control grafts weresemiquantitatively scored (0 to 3+) for antigen staining intensity by apathologist blinded as to experimental group. Computer-assisted image analysis(C-Imaging Systems, Cranberry Twp., N.J.) was used to measure the total graftcross-sectional area positive for ICAM-1 to provide a quantitative assessment ofthe AS effect.
Assessment of biologic effects of ex vivo AS ICAM-1 ODN
The biologic effects of ex vivo AS ODN transfection (experimental group,n = 25) were compared with five additionalcontrol groups as shown in Table I: an ODN control (n =6), a vector control (n = 6), a negativecontrol (n = 10), and two positive controlgroups. Rats in the positive control groups receiveduntransfected grafts and were treated with a single injection of ICAM-1 MAb(positive control 1, n = 25) or LFA-1 MAb(positive control 2, n = 25) 1.5 mg/kgintravenously just before graft reperfusion. After successful return of heartcontractions, grafts were either procured at 6, 12, and 24 hours to assessreperfusion injury or the recipients were treated with a 10-day course ofcyclosporine A 5 mg/kg per day via gastric gavage to promote long-term graftsurvival and allow for the development of CGVD. Studies of reperfusion injuryanalyzed three parameters of myocardial injury: (1) cardiac edema using percentwet weight assay (%wt/wt)—heart weighed before and after drying inan oven at 100° C for 24 hours; (2) neutrophil infiltration assessed by MPOenzyme activity—total protein isolated from frozen tissue and assayed forchange in absorbance at 470 nm at 1 minute after addition of guaiacol andperoxide, and (3) histologic injury as determined by the presence of contractionband necrosis (%CBN)—after initial review with a pathologist, thepercentage cross-sectional area of a trichrome-stained section involved withcontraction band necrosis was quantitatively determined by means ofcomputer-assisted image analysis (C-imaging systems, Cranberry Twp., N.J.).
|
|
Statistical analysis
Transfection efficiency data (percent nuclear fluorescence) was comparedbetween groups using the Mann-Whitney U test. The parameters of reperfusioninjury, MPO and %wt/wt, were compared by means of analysis of variancewith a post hoc t test. Given the consistentlyunequal standard deviations noted in the mean areas of percent contraction bandnecrosis and percent ICAM-1 staining between groups, the Welcht test was required for comparison of thesedata. CGVD scores were compared between groups by means of the Mann-Whitney Utest. An association between positive perivascular ICAM-1 staining of any givenartery on immunohistochemical analysis (defined as a score > 1) andunderlying vasculopathy of that vessel (defined as CGVD score > 2) wasinvestigated by means of the Fisher exact test.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
|
). Trends toward asignificant reduction in contraction band necrosis after MAb treatment werenoted at 12 hours (10.1 ± 1.5 vs 12.0 ± 1.4,n = 5, p =0.07) and 24 hours (10.8 ± 1.4 vs 12.8 ± 1.9,n = 6, p =0.06) (Fig. 4, C).
|
Compared with the positive control groups, ex vivo AS ODN transfectionwas relatively less effective at preventing early reperfusion injury, displayinga reduction in parameters of reperfusion injury only at the later time points.No difference from control was noted at 6 hours and only trends towardsignificance were seen at 12 hours (MPO: 0.07 ± 0.02 vs 0.11 ±0.04 U/mg protein, n = 5,p = 0.08; %wt/wt: 76.9%± 2.19% vs 79.24% ± 0.71%,n = 5, p =0.06; %CBN: 9.9 ± 2.23 vs 12.0 ± 2.62,n = 5, p =0.20). However, statistical significance was achieved in these parameters at 24hours (MPO: 0.06 ± 0.02 vs 0.10 ± 0.04 U/mg protein,n = 6, p =0.05; %wt/wt: 76.7 ± 1.25 vs 78.6 ± 1.46,n = 6, p <0.05; %CBN: 9.4% ± 1.03% vs 12.8% ±1.9%, n = 6, p <0.02). There was no statistically significant difference between the parametersof reperfusion injury compared with perioperative ICAM-1 MAb at any time point.
Effect of AS ODN on CGVD
An initial 10-day course of cyclosporine A (5 mg/kg per gavage tube)promoted long-term PVG allograft survival (i.e., >60 days) in 53 of 75 ACIrecipients (70.7%). Despite prolonged survival of the vigorously beatingallograft, CGVD reproducibly develops in control PVG grafts with a mean score of1.84 ± 0.75 (n = 17) (TableII). Contrary to impressive effects on reperfusion injury,perioperative MAb administration had minimal benefit against chronic allograftvasculopathy. After a single perioperative dose of either MAb, CGVD scores onpostoperative day 90 were reduced to 1.25 ± 1.01,n = 10 (anti-ICAM-1 MAb) and 1.58 ±1.13, n = 10 (anti-LFA MAb), differenceswhich did not reach statistical significance compared with untreated controls.On the other hand, ex vivo AS ICAM-1 ODN transfection significantly reduced CGVDto 0.98 ± 0.48, n = 10 (p < 0.05, Mann-Whitney U test). This protectiveeffect on graft vasculature was eliminated by using an otherwise identicaltransfection protocol with the substitution of a control RAS ODN sequence (1.73 ±0.98, n = 6) or by merely perfusing graftcoronary arteries with AS ODN without the application of pressure (1.65 ±0.78, n = 6).
|
After a short course of cyclosporine A, control grafts showed a dramaticincrease in the expression of all antigens on postoperative day 30 (Fig. 5). By day 90, CD8 and MHC class I and IIexpression returned to baseline with decreased but persistent CD4 and macrophageaccumulation primarily around those arteries and arterioles expressing ICAM-1(Fig. 6). The presence of ICAM-1 around any givenartery, as seen around 45% of allograft arteries (33 of 76 total vessels,n = 17), was significantly associated witha CGVD score > 2 (p = 0.02, Fisherexact test) and provided for a relative risk of vasculopathy of 2.04 (95%confidence interval: 1.1 to 3.9).
|
|
). Those vessels staining positive for perivascular ICAM-1despite ex vivo AS ODN transfection were at an increased relative risk of CGVD(2.63, 95% confidence interval: 1.1 to 6.3). |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
4Ex vivo AS ODN blockade of ICAM-1 provides for a more graft-specific, yetequally potent antiinflammatory and immunosuppressive strategy that avoids theinfectious risk inherent in recipient ICAM-1 blockade using systemic MAb. Inthis rodent heterotopic transplant model, AS ODN transfection of donor cardiacgrafts resulted in a reduction of postoperative ICAM-1 up-regulation and asignificant blockade of CGVD compared with untransfected control grafts. To ourknowledge, no study to date has demonstrated a biologic effect of gene therapyon the development of CGVD.
Clinical and experimental studies clearly point to an important role forICAM-1 on the pathogenesis of CGVD. By promoting increased reperfusion injury,early (<24 hours) ICAM-1 up-regulation indirectly plays a role in theinitiation of the vascular "response to injury" phenomenon thoughtresponsible for neointimal hyperplasia of CGVD. In addition, the correlation oflate (>30 days) up-regulation of perivascular ICAM-1 expression withdevelopment of CGVD in this and other models 11,12,15 suggests a more direct role forthis molecule in these vascular lesions. MAb provides a positive control groupfor ICAM-1 blockade in the immediate postoperative period, but because of itsrapid clearance, a single perioperative dose would not be expected to influenceevents resulting from a late up-regulation in ICAM-1. A single ex vivo treatmentwith AS ODN, on the other hand, was found to be increasingly effective over thetime period analyzed. For example, when the efficacy of ICAM-1 blockade by ASODN was assessed in our studies, a difference between transfected and controlgroups was more clearly demonstrated in allografts at day 3 than day 1.
This difference in pharmacokinetics of AS and MAb was further illustratedby their divergent biologic effects. Because of the central role that ICAM-1plays on neutrophil diapedesis and toxicity during reperfusion, optimizedblockade of early, preexisting ICAM-1 by means of MAb prevented reperfusioninjury more effectively than AS ODN. More surprising were the results of theCGVD studies. Using the same PVG to ACI model, previous studies have shown thatthe up-regulation of ICAM-1–mediated reperfusion injury resulted inworsened eventual CGVD. 16However, despite its optimal effectiveness at reversing this initial nonspecificinjury, ICAM-1 MAb was much less effective than AS ODN at reducing CGVD. TheICAM-1 MAb used in these studies (1A29), because it is an Ig1 isotype MAb,potentially triggers recipient effector mechanisms after interaction with itsendothelial target (i.e., ICAM-1). It has been proposed that initiation of theseeffector mechanisms may inadvertently "activate" the endothelium andpromote early 17 or late 18 vascular injury. By targetingLFA-1, a ligand of ICAM-1 expressed on leukocytes and not endothelium, the WT.1MAb provides a positive control for the blockade of ICAM-1/LFA-1–mediatedreperfusion injury without these theoretical concerns of vascular injury.However, the reduction of reperfusion injury with LFA-1 MAb also failed toprevent CGVD. These results suggest that the anti-CGVD mechanism of donor graftAS ICAM-1 ODN treatment is not fully explained by an initial reduction inreperfusion injury–mediated immunogenicity and vascular injury. Furtherstudies are needed to determine whether transfection leads to long-termreduction of perivascular ICAM-1 by the persistence of AS ODN or whether somechange in allograft biology is triggered that provides a resistance to CGVD.
Nonsequence-specific effects of ODN transfection on gene expression havebeen well documented, adding complexity to hasty conclusions regarding biologiceffects. The aptamer effect 19and direct antiproliferative and cytokine-inducing actions of certain ODNsequences 20 produceunpredictable, non-AS-related influences on biologic end points. In our studies,a true AS mechanism was supported by the lack of effect of (1) control RAS ODNon either target ICAM-1 protein expression or CGVD, (2) AS ICAM-1 ODN on theexpression of other gene products potentially important for CGVD (e.g., MHCclass I and II), or (3) AS ODN treatment without pressure (vector control group)on CGVD. Indeed, ex vivo hyperbaric delivery of AS ODN to donor cardiac graftsresulted in a 48% transfection rate. Standard viral and lipid vectorshave consistently shown transfection efficiency of less than 10% of totalcells. 21,22Prolonged allograft survival has been accomplished with the use of these moretechnically difficult vectors by local overexpression of factors such as T-cellgrowth factor–ß or interleukin-10. 21,23However, AS ODN blocks target mRNA translation only in successfully transfectedcells. Immunosuppression by means of this strategy obviously depends ontransfection efficiency at levels much closer to that described by thehyperbaric vector.
In conclusion, these studies document the sole use of pressure as ahighly efficient vector for ODN delivery with documented efficacy of ICAM-1blockade. Compared with ICAM-1 MAb, AS ODN was less effective at preventing themanifestations of reperfusion injury, especially at early time points, acomparison which was consistent the pharmacokinetics of these agents. Less wellexplained was their inverse effects on CGVD with ICAM-1 blockade with AS ODN butnot perioperative MAb providing a significant preventive effect on thedevelopment of vascular lesions. Although further investigation regarding themechanism of AS ODN effect in this model is warranted, these studies supportongoing investigations in this laboratory using the more clinically relevantnonhuman primate transplantation model. We think that this highly targetedalteration of donor organ immunogenicity may have an important future role inimmunosuppressive strategies in clinical transplantation.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  X. Zheng, D. Lian, A. Wong, M. Bygrave, T. E. Ichim, M. Khoshniat, X. Zhang, H. Sun, T. De Zordo, J. C. Lacefield, et al. Novel Small Interfering RNA-Containing Solution Protecting Donor Organs in Heart Transplantation Circulation, September 22, 2009; 120(12): 1099 - 1107. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-L. C. Urbain Sense, Antisense, and Common Sense J. Nucl. Med., November 1, 2001; 42(11): 1670 - 1672. [Full Text] [PDF]
|
|
|
|
|
|
C. A. Labarrere, D. R. Nelson, S. J. Miller, J. M. Nieto, J. A. Conner, D. E. Pitts, P. C. Kirlin, and H. G. Halbrook Value of Serum-Soluble Intercellular Adhesion Molecule-1 for the Noninvasive Risk Assessment of Transplant Coronary Artery Disease, Posttransplant Ischemic Events, and Cardiac Graft Failure Circulation, September 26, 2000; 102(13): 1549 - 1555. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ehsan and M. J Mann Antisense and gene therapy to prevent restenosis Vascular Medicine, May 1, 2000; 5(2): 103 - 114. [Abstract] [PDF]
|
|
|
|
 |
|
 B. N. Mora, C. H. R. Boasquevisque, M. Boglione, J. M. Ritter, R. K. Scheule, N. S. Yew, L. Debruyne, L. Qin, J. S. Bromberg, and G. A. Patterson Transforming growth factor-{beta}1 gene transfer ameliorates acute lung allograft rejection J. Thorac. Cardiovasc. Surg., May 1, 2000; 119(5): 913 - 920. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. T. Feeley, A. K. Park, S. Alexopoulos, E. G. Hoyt, M. P. Ennen, R. S. Poston Jr, and R. C. Robbins Pressure delivery of AS-ICAM-1 ODN with LFA-1 mAb reduces reperfusion injury in cardiac allografts Ann. Thorac. Surg., July 1, 1999; 68(1): 119 - 124. [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 |
