| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1999;68:1924-1928
© 1999 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, University of Washington, Seattle, Washington, USA
Address reprint requests to Dr Allen, Division of Cardiothoracic Surgery, University of Washington, Mailstop 356310, 1959 NE Pacific St, Seattle, WA 98195
e-mail: allenm{at}ctd.surgery.washington.edu
Presented at the International Symposium on Myocardial Protection From Surgical Ischemic-Reperfusion Injury, Asheville, NC, Sep 21–24, 1997.
Abstract
Modification of gene expression within the heart could have a dramatic impact on both cardiac transplantation and routine cardiac surgery within the next decade. The advantage of gene therapy is that it would allow organ-selective local delivery of higher levels of cytokines, growth factors, vasodilators, or immunosuppressive drugs than could be safely achieved by systemic administration. Direct transfection or transduction of myocytes, endothelium, and/or vascular smooth muscle cells could increase the density of beta adrenergic receptors, inhibit endothelial adhesion molecule expression, or prevent neointimal formation in coronary bypass grafts. Cell transfer of neonatal or engineered adult myocytes might allow repopulation of infarct areas. The current limitations to effective clinical gene therapy are the variable transfection efficiencies of gene delivery systems, limited duration of gene expression, immune responses to viral vectors, and safety concerns. Ischemia-reperfusion injury will be one of the earliest applications for gene therapy since the short time course of injury and recovery would be amenable to therapeutic approaches with limited durations of action, achievable by currently available delivery vectors.
The question of whether there is a role for gene therapy during cardiac surgery may be the key to the next decade. In spite of the vast and rapid increase in our fund of scientific knowledge about human genes and cell signaling mechanisms, applications in clinical practice are only now beginning. The potential for medical miracles from genetic modulation is something we still read about in newspapers rather than medical journals. To date, only a few gene therapy strategies have been translated into concrete clinical applications [1]. However, with the first use of gene therapy during cardiac surgery reported only just recently, we realize that we are on the verge of what could become a revolution.
On a more basic level, we can all imagine how genetic engineering could aid arterial remodeling or prevent coronary restenosis after angioplasty, making angioplasty a more permanent solution to coronary occlusive disease. Perhaps a more cogent question then is: Will gene therapy facilitate surgical revascularization or replace surgeons?
Let us first consider why it would be advantageous even to develop this new technology for cardiac surgery. First, gene therapy can produce a local effect in an individual organ or vessel. One example of the benefits of such local effects are, as we will discuss later, anti-adhesion strategies. If adhesion molecules were blocked systemically by drugs or antibodies, the patient would be subject to infectious complications. If the blockade occurs only in a single organ or vessel, that risk is minimal. Similarly, coronary vasodilatation might be accomplished by local intracardiac therapy whereas producing the same degree of vasodilatation by systemic drug therapy would risk hypotension.
Secondly, gene therapy may, in the future, be able to achieve a more prolonged effect than could be accomplished otherwise. If the goal is to provide an appropriate cellular environment to facilitate vascular remodeling or infarct revascularization, this environment should ideally be maintained as constant and persistent for months, not days.
There are currently two methods for gene delivery. The first method is direct transfection or transduction of the heart. This can be accomplished by perfusion of genes and vectors through the coronary circulation [2–4] or by direct injection into myocardial sites [5–8]. Such an intraoperative transfection system could be used to produce growth factors for angiogenesis [1], relevant cytokines for immunosuppression [9–11], vasodilators and anti-thrombotic agents for ischemic coronary beds, or to increase cardiac myocyte beta-adrenergic receptor densities [12, 13] and contractility to hypokinetic myocardium. Another intraoperative use of gene therapy would be the pretreatment of saphenous vein grafts before implantation to prevent graft atherosclerosis [14, 15].
A second method is indirect—transducing cells in vitro which are then selected for activity and seeded into specific sites in vivo. Examples of this latter methodology would include the reseeding of the denuded endothelium with engineered endothelial or vascular smooth muscle cells. Another example is the work being done by several investigators to examine the potential for repopulation of infarcts with transplanted myocytes—derived either from neonatal myocytes [16, 17] or from Myo-D transformation of the patient’s own fibroblasts in the healing infarct area into muscle cells [18]. Neonatal skeletal myoblasts implanted into injured areas of rat myocardium have been shown to harbor the capacity to differentiate into the slow twitch fibers characteristic of cardiac myocytes and to contract on stimulation [19]. Further transduction of such transformed cells with increased numbers of beta receptors could enhance the potential of creating a contractile wall of regenerate myocardium [12, 13]. To date, however, achieving reliable survival of transduced cells in infarcted areas remains a challenge [20, 21].
What vectors are available? This continues to be a rapidly accelerating field. For cardiac surgery, especially non-transplant surgery, it seems almost advantageous to design a delivery system that would ensure that gene expression would occur only over a limited period of time, thereby obviating unexpected late results of permanent expression of an introduced gene. Among currently available viral vectors, the adenoviruses offer excellent transduction efficiency, even in non-dividing cells, but short duration of action [8]. However, adenoviruses also incite an inflammatory response [22–24] and can themselves produce intimal lesions [25], a potentially significant disadvantage for patients with coronary disease. Second and third generation vectors are being designed to avoid this problem [26]. At the other end of the spectrum, retroviral vectors can integrate introduced genes into the cell’s genome and provide a long duration of expression, but only transduce actively dividing cells with appreciable frequency [27, 28]. The newer adeno-associated viruses fall somewhere in between [29–31].
For the past 5 years, my laboratory has been utilizing a technique that we developed for intraoperative transfection of donor hearts, a technique which would be equally applicable to transplantation as to routine cardiac surgery, and which could be added to our intraoperative armamentarium without requiring modification of current operative techniques [2].
For our initial work, we have concentrated on liposome vectors. For early Food and Drug Administration approval, especially in elective cardiac surgery or in transplantation, where one is dealing with immunocompromised hosts, the absolute freedom that liposomes offer from infectious risk or wild type recombination, compared to viral vectors, is an advantage. In addition, liposomes can transfect nondividing cells, such as vascular endothelial cells and myocytes. They are taken up rapidly, making them a practical vector for intraoperative methodology. In our work, an exposure time of 10 minutes was all that was necessary for liposome adsorption to endothelial cells. Finally, because they do not incite a host immune response, retreatment is possible.
We first examined the feasibility of liposome transfection in a transplant setting in a rabbit model. During the ischemic time, after isolation of the donor heart and delivery of cardioplegia, cationic cholesterol-based liposomes (Valentis, Burlingame, CA) complexed to a reporter gene were infused through the aortic root into the coronary arteries. The donor heart was then transplanted and gene expression assessed in the transplanted heart in the recipient. Using this methodology, we found excellent donor heart reporter gene expression at 24 hours, [2] and, more recently, out to at least 7 days in hearts and 21 days in carotids. Of interest, this delivery method resulted in equivalent expression in all distributions of the heart, excellent coronary arterial gene expression, and even coronary sinus expression, suggesting that at least some of the liposomes had crossed the capillary bed to the venous side.
Next, we wanted to investigate what percentage of cells were actually transfected. Interestingly, by tagging liposomes with a lipid tracer, we have new evidence that liposomes may be adsorbed to the vast majority of coronary arterial cells—contrary to their reputation for poor transfection efficiency and similar to what has recently been reported in lungs [32]. We are now focusing on improving gene product expression by studying the intracellular trafficking of the liposome-complexed DNA.
Having this technology now available, what should be targeted with gene therapy to move this technology toward clinical applications? My vision for the future of transplantation is that donor organs will undergo genetic modification after harvesting and before implantation such that they will be "immune" to attack by the host: organ-specific immunosuppression. Optimally, the recipient would then not require immunosuppression. How to accomplish this?
In a transplanted organ, the donor heart endothelium serves as the interface between the donor cells and recipient blood. Theoretically, the initial adhesive interactions between recipient leukocytes and donor endothelium should be the first steps in transplant rejection. We can demonstrate these processes in vitro but would they be found in human transplantation? Over the years, we have analyzed adhesion molecule expression in over 400 human posttransplant myocardial biopsies and found that, indeed, expression of the adhesion molecules E-selectin [33] and VCAM-1 [34], are both highly correlated with the onset of rejection. In these myocardial biopsy specimens, lymphocytic infiltrates collect around VCAM-1-expressing venules, whereas areas free of infiltrates are free of expression. Furthermore, treatment of rejection is associated with downregulation of expression [35]. However, this correlation still does not prove that adhesion molecule expression is necessary and sufficient for rejection.
To test this, we used antibodies to adhesion receptors to block rejection. In a heterotopic heart transplant model between outbred strains of rabbits, rejection usually occurs within 7 to 9 days without immunosuppression. Seven days of antibody to CD18, the pan-ICAM receptor, significantly reduced early neutrophilic infiltrates and reduced later lymphocytic infiltration (cellular rejection) of the graft [36]. In this rabbit transplant model, vascular rejection is produced simultaneously with cellular rejection. Antibody to CD18 also eliminated vascular rejection at 7 days [37]. Blocking VLA-4, the integrin receptor for both VCAM-1 on endothelium and for fibrnnectin in the extracellular matrix, also abrogated vascular rejection in these transplanted hearts [37]. Since we have seen similar adhesion molecule expression on both lumenal endothelium and vascular smooth muscle cells in human posttransplant and nontransplant atherosclerosis [38, 39], anti-adhesion strategies may be effective in blocking not only cellular rejection but also later transplant arteriopathy. And, even though the expression of transfected genes may be of limited duration with today’s vectors, transient blockade of costimulatory molecules over the initial weeks to months posttransplant may be all that is required to induce a state of relative tolerance to an allograft [40].
Of equal importance for gene therapy applications in routine cardiac surgery, we found that, during ischemia-reperfusion injury, VCAM-1 was not expressed until 6 hours posttransplant, and did not become maximal until 24 hours posttransplant [36]. Thus, there appears to be a window of time for transcription and translation of transfected genes to take place and still affect this process.
What strategies could be used to inhibit adhesion molecule expression? New adhesion molecules and receptors are described almost monthly. My laboratory has taken the position that, because of this great redundancy of adhesion molecules, gene therapy targeting a factor central to the production of many adhesion molecules would be more likely to be effective. In my estimation, even with imperfect transfection efficiency, producing an incomplete reduction in the expression of a number of adhesion molecules may be equally, or more, effective than producing complete blockade of only one adhesion molecule, as would be accomplished with antibody therapy or antisense therapy. In my laboratory, we have chosen to target NF
B, a transcriptional activator for E-selectin, ICAM-1, and VCAM-1. With inefficient transfection, as is accomplished with currently available liposomes, we chose a secreted gene product, such that secreted products from the few transfected cells can affect their neighbors in paracrine fashion. For this application, we are overexpressing nitric oxide synthase which, in turn, inhibits NFB activation [41] as well as serving as a coronary vasodilator. Indeed, we found that a single intraoperative dose of liposomes complexed to DNA encoding nitric oxide synthase could inhibit leukocyte infiltration and ischemia-reperfusion injury by reducing NFB activation. Adenoviral transduction of inducible nitric oxide synthase has been effective in inhibiting experimental aortic allograft arteriopathy [42], whereas liposome transfection of endothelial nitric oxide synthase has been used to inhibit neointima formation in balloon injury models [43].
With efficient transduction, as can be accomplished with adenoviral vectors, we are using a dominant negative mutant of IB
[44], which is resistant to phosphorylation, and thus prevents the p65 subunit of NFB from translocating to the nucleus with subsequent transcriptional activation of inflammatory mediators. Recently, Sawa and colleagues reported the successful use of a cis element decoy against the NFB binding site, targeting this same pathway, to pretreat rat hearts that were subjected to ischemia-reperfusion injury 3 days later [45]. Pretreatment resulted in decreased neutrophil adherance and reduced IL-8 production compared with controls, evidence that ischemia-reperfusion injury could be prevented by blocking NFB activation.
What evidence do we have that this strategy will be effective? We have recently perfected a method to visualize the activated p65 epitope in the nucleus in fixed tissue by immunocytochemistry. In transplanted hearts, we have preliminary evidence that NFB activation parallels adhesion molecule expression in that it is not detected to any significant degree before 4 hours into reperfusion but, thereafter, increases steadily over the first 24 hours. Again, this would appear to provide a window of time during which transfected gene products could still abrogate ischemia-reperfusion injury.
So, in summary, we have shown that gene transfection can produce localized genetic alteration of transduced organs, a technology that can be applied to ameliorate ischemia-reperfusion injury or transplant rejection. Specifically, liposome-mediated gene transfection may provide a practical alternative to viral vectors, an alternative which could be particularly appealing in immunosuppressed transplant recipients or in patients undergoing elective surgery. And, this same methodology would have equal applicability in routine cardiac surgery for the amelioration of ischemia-reperfusion injury or in transplantation of abdominal organs.
What should we expect in the near future? New adenoviral vectors with less immunogenicity are under current development [46, 47]. And, at the same time, a multitude of different viruses, including disabled pathogens such as herpes viruses [48] and even human immunodeficiency viruses [49], are being explored for their potential use as gene delivery vectors. As clinicians, we hope to see methods of producing target specificity for in vivo transfection such that one could limit transduction to certain cell types. This could become a reality with cardiac myocyte-specific promoters [50] and by complexing liposome vectors with antibodies specific for adhesion ligands expressed on activated endothelial cells [51]. And, importantly, several "on-off" switches have already been developed [52, 53] which, in the future, might allow the physician to regulate the onset and duration of gene product expression in previously transfected genes by simply having the patient take a pill. Concerns about "drug interactions" could take on a new meaning!
The new millenium is upon us. Will cardiac surgery in the next century be different? I would like to leave you with a quote from Adam Kay, a scientist and a pioneer in computer technology, who said, "The surest way to predict the future is to invent it." And, as surgeons, we should—collectively.
Acknowledgments
I would like to acknowledge the researchers in my laboratory who are responsible for the success of our gene therapy work: Drs Akiko Iwata, Sadahiro Sai, Yoshio Nitta, Owen Lawrence, and Ricarda de Fries; medical students Fieka Wijffels, Herbert Chen, and Nicole Campbell; research scientist Robert Thomas; and research technician Christine Rothnie. A portion of the original work referred to in this article was supported by grants from the National American Heart Association.
References
B binding site. Circulation 1997;96(Suppl 9):S280-S284.
This article has been cited by other articles:
|
|
 |
|
  J. Jayakumar, K. Suzuki, I. A. Sammut, R. T. Smolenski, M. Khan, N. Latif, H. Abunasra, B. Murtuza, M. Amrani, and M. H. Yacoub Heat Shock Protein 70 Gene Transfection Protects Mitochondrial and Ventricular Function Against Ischemia-Reperfusion Injury Circulation, September 18, 2001; 104 (2009): I-303 - I-307. [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 |
