Background Gene transfer to peripheral arteries has been accomplished with catheter-based approaches. Recently, gene transfer to the carotid artery and intracranial vessels has been achieved both in vitro and in vivo. Although gene therapy for cerebral vascular disease may not be accomplished for years, currently available methods probably will allow novel approaches to the study of vascular biology.
Purpose This mini-review summarizes current methodology and describes some potential goals of gene therapy. Transfection of vessels might be used to prevent vasospasm after subarachnoid hemorrhage, stimulate growth of collateral blood vessels, stabilize atherosclerotic plaques, and prevent restenosis after angioplasty. Gene transfer approaches also may be useful in treating ischemia by inhibition or overexpression of cytokines and by effects on neurons. Some formidable barriers to gene therapy are the current lack of safe and effective vectors for gene transfer, the difficulty in delivering vectors to intracranial vessels, and the transience of transfection.
Conclusions At present, gene transfer is a promising tool for the study of vascular biology. Obstacles to gene therapy for cerebral vascular disease seem sufficiently large that new approaches, rather than refinement of current approaches, may be needed. Progress toward gene therapy probably will be made in steps rather than leaps.
After several years of unfettered excitement and hype, it is now clear that gene therapy is at a very early stage of development. Nevertheless, despite great obstacles, this area of research has considerable promise. In this mini-review, we summarize the promise and the problems associated with gene therapy. The present state of the art will be summarized in relation to gene therapy for cerebrovascular disease. We then speculate briefly about the future.
The Promise of Gene Therapy
The underlying goal of gene therapy for cerebrovascular disease is to introduce cDNA into a vessel or perivascular tissue and produce a substance that favorably modulates vascular growth or function. Theoretically, this could be performed via introduction of naked DNA, but because this approach provides only minimal gene transfer and expression, it does not look promising. Thus, a variety of studies have focused on methods to deliver and express genes in blood vessels. The most promising currently available vectors for delivery of genes to blood vessels are viral vectors (adenovirus and retrovirus), DNA-cationic liposome complexes, and viral conjugate vectors (liposomes with a viral coat).
Intracranial Vascular Disease
Gene transfer to intracranial blood vessels is particularly attractive because surgical intervention for these vessels is difficult, and some major clinical problems remain resistant to nonsurgical treatment. We have speculated about possible applications of gene therapy to several clinical problems (Fig 1⇓).
First, gene therapy to prevent vasospasm after subarachnoid hemorrhage seems to be potentially feasible. For example, a neurosurgeon might clip an aneurysm and administer a vector (perhaps in CSF)1 that contains a gene which encodes a vasoactive protein that inhibits vasospasm. An alternative approach would be appropriate if it is demonstrated that vasospasm is caused by a specific substance, such as endothelin-1. It might be possible to inhibit vasospasm by introducing a gene that inhibits the induction or action of the spasm-producing substance.
It often takes a few days to achieve maximal gene expression and synthesis of a protein from viral vectors. Because vasospasm typically does not occur until several days after subarachnoid hemorrhage, the timing for gene therapy seems feasible. Current vectors allow only transient expression of transgenes, but fortuitously the risk of vasospasm is also transient. Thus, even with currently available vectors, which provide transient transfection, gene therapy to prevent vasospasm after subarachnoid hemorrhage may be achievable.
Second, gene therapy might be useful for stimulation of growth of collateral blood vessels. Some growth factors (for example, VEGF) stimulate vascular proliferation in the presence of ischemia. Thus, the concept that VEGF might be useful for stimulation of angiogenesis in the peripheral circulation and coronary vessels has received a lot of attention.2 A recent study suggests that intracoronary adenoviral-mediated gene transfer of FGF stimulates angiogenesis and ameliorates stress-induced myocardial ischemia.3 The goal of this approach is to stimulate growth of collateral vessels to vulnerable regions. In the presence of cerebrovascular disease, it might be possible to instill a vector with the cDNA for VEGF or FGF. In the presence of intermittent ischemia, gene transfer of VEGF or FGF might stimulate growth of collateral blood vessels in the area at risk.
Third, gene therapy might be useful for treatment of brain tumors.4 Most efforts are directed at expression of a protein that kills tumor cells, but a theoretical approach to treatment of brain tumors is to inhibit vascular proliferation and thus produce ischemia in the tumor.
Extracranial Vascular Disease
Gene therapy might be useful for treatment of disease in the carotid artery. Treatment of vertebral artery lesions might also be useful, especially because the vessels are relatively inaccessible to surgery. For example, a goal of gene therapy might be to transfect an atherosclerotic lesion with a gene that inhibits thrombosis or proliferation of the lesion. Gene transfer approaches have been used to interrupt the cell cycle and thereby inhibit proliferation of vascular muscle.5 6 7 Studies have focused on inhibition of restenosis, and enormous problems related to transient expression of transgenes would need to be solved before similar approaches might be useful for inhibition of the chronic proliferation that is characteristic of atherosclerosis.
A more realistic but nevertheless formidable goal would be to use gene therapy to stabilize the atherosclerotic plaque. Studies in coronary arteries suggest that a thick fibrous cap may protect atherosclerotic lesions against rupture and that macrophage-rich regions of atherosclerotic lesions may be especially prone to rupture. One might theoretically target these regions to stabilize atherosclerotic plaques.8
Restenosis has been an enormous problem in coronary arteries after balloon angioplasty. Restenosis has been remarkably resistant to a variety of therapeutic approaches. Nevertheless, there are some novel and promising gene therapy approaches to restenosis.5 6 If balloon angioplasty proves to be useful for cranial vessels, restenosis may emerge as a major problem. Thus, gene therapy for restenosis might be applicable to cranial as well as coronary and other blood vessels.
The Problems of Gene Therapy
Gene therapy will require safe and effective vectors for gene transfer. An optimal vector is not available, and although development of vectors is an extremely active area of focus, it seems likely that vectors will be the major hurdle for gene therapy. Some vectors are summarized very briefly (Table⇓). There have been several reviews of vectors in relation to gene transfer to other vascular beds.9 10
Cationic liposomes appear to be a relatively safe but inefficient method for gene transfer to blood vessels and other target tissues. Liposomes offer many advantages: large cDNA sequences can be incorporated into the liposome, the vector is relatively easy to prepare for clinical use, and cell division is not required for transfection of cells by liposomes. A major problem is that liposomes are far less effective than viral vectors in transferring genes into cells.
Although viral vectors offer the strong advantage that they are more effective than liposomes, vectors may be less safe, and the size of cDNA inserts is limited. Retroviral vectors were used in initial studies of gene transfer to vessels,11 in part because they had been effective in gene transfer to bone marrow and other organs. Retroviral vectors integrate into chromosomal DNA of the target tissue, which potentially could result in stable gene expression but also could produce mutagenesis. A second major problem with retroviruses is that replication of target cells is necessary for gene transfer. For these reasons, it seems unlikely that retrovirus will be useful for gene transfer to cerebral blood vessels.
Because herpesvirus is neurotrophic, it is possible that the virus will be useful for gene transfer that is targeted for neurons,12 perhaps including perivascular nerves. In the attempt to target cerebral blood vessels, however, a different virus probably will be more useful.
Recombinant adenovirus is a promising vector for gene transfer to neurons13 and blood vessels.9 Recombinant adenoviral vectors have been engineered to be replication-deficient by deleting a portion of the viral genome (the E-1 region) that is required for replication. A foreign cDNA then can be inserted into the region of the genome that has been removed. A great advantage of recombinant adenoviral vectors is that they can transfect many different cell types, independent of cell division. Thus, adenoviral vectors can infect blood vessels despite the low rate of replication of cells in the vessel wall.
One major limitation of adenoviral vectors is that gene expression is transient, usually lasting for only a few days or weeks, depending on the tissues that are transfected and the immunocompetency of the host. A second major limitation is that adenoviral vectors produce inflammation.14 15 Although it is likely that both problems (transient duration of transgene expression and induction of inflammation) can be addressed by construction of appropriate viral vectors or by coexpression of cytokines such as IL-1016 that inhibit inflammation, these limitations currently are major problems.
A promising approach is to combine liposomes with inactivated viral particles. The viral coat facilitates entry of liposomes into cells. This approach improves efficiency of gene transfer and minimizes the toxic effects of viruses. Vascular function can be altered using a plasmid encapsulated in a viral-liposome conjugate.5 Adenoviral vectors appear to be more efficient, however, than viral-liposome conjugates. Nevertheless, viral-liposome conjugates are promising and likely to be useful in the development of novel vectors.
An attractive concept is that gene transfer can be “targeted” to appropriate tissues. Targeting can be accomplished by delivery (ie, a tissue-specific injection of a vector), by tissue-specific promoters to drive expression of a transgene, and by tissue specificity of vectors (eg, neurotropic herpesvirus). After intravascular injection of a vector, adherence of the vector to specific sites in endothelium might be very useful. One might attempt to identify and target unique binding sites on cerebral endothelial cells, to achieve preferential gene transfer to intracranial blood vessels. This approach has enormous barriers because the vector would need to be transported along the periphery of flowing blood so that it would come in contact with endothelial cells, and there would need to be extremely rapid binding during blood flow past the sites. Nevertheless, there is considerable work under way to achieve endothelium-specific targeted gene transfer.
A second approach would be to attempt to achieve tissue-specific targeting based on the use of specific promoters. For example, an adenoviral vector might be used to infect all cells with which it comes in contact. Expression of a transgene, however, depends (in part) on the efficiency of the viral promoter. Thus, for example, in regions of an atherosclerotic vessel that are sites of pronounced inflammation and activation of nuclear factor-κB,17 appropriate viral promoters may be activated. We have used this approach recently to achieve greater transgene expression in atherosclerotic than in normal arteries18 through activation of the cytomegalovirus promoter that responds to a variety of inflammatory stimuli.19
We have used a novel approach to target different regions of the brain.1 Adenoviral vectors were injected into CSF in 20% sucrose, which is dense and thus accumulates in the most dependent region of the intracranial cavity. By tilting the head in various positions, we achieved preferential expression of a transgene in perivascular tissues of either hemisphere, the circle of Willis, or the ventral brain stem.
Delivery of Vectors
Gene transfer to blood vessels has generally been accomplished with intravascular administration of the vectors.5 6 7 Unfortunately, there are major limitations in transduction of genes to cerebral blood vessels with intravascular approaches. One limitation is that efficient gene transfer requires either interruption of blood flow for several minutes or the use of a double balloon catheter that produces transfection in only a limited region of a large artery. A second limitation is that the blood-brain barrier may attenuate infection of cerebral vessels beyond the endothelium after intravascular injection.
A novel approach to gene delivery to the brain involves osmotic disruption of the blood-brain barrier before injection of adenovirus. This approach allows access of adenoviral vectors to glia20 but, to our knowledge, has not resulted in transgene expression in cerebral blood vessels.
To circumvent these obstacles, we have developed two novel approaches for gene transfer to cerebral blood vessels. One approach is to inject recombinant adenovirus into CSF. After injection of adenovirus into CSF, there is prolonged contact of adenovirus with the target tissue. This approach allows efficient transfection of the meninges over cerebral vessels (Fig 2⇓),1 with limited expression of transgene in the vessels. Although transgene expression is limited largely to perivascular meninges, we anticipate that injection of vectors into CSF may be useful. For example, expression of one of the genes for NOS in perivascular meninges may result in release of NO, a highly diffusible substance, and thus might produce relaxation of the nearby cerebral blood vessels. This mechanism may be analogous to NO-containing perivascular nerves, which are located in adventitia. We speculate, for example, that this approach could be useful in prevention of vasospasm after subarachnoid hemorrhage. A major obstacle, however, is that injection of adenoviral vectors produces an inflammatory response.15 This approach will not be therapeutically useful until a new generation of viral vectors, which will not produce inflammation, is developed or until other approaches (such as coadministration of IL-10) are developed to suppress the immune response.
Transfection after perivascular injection of virus has also been accomplished in peripheral blood vessels.21 Injection of an adenoviral vector in the periarterial sheath of monkeys and rabbits produces substantial expression of transgene in the adventitia of the femoral and carotid arteries. No gene transfer was observed to cells in the intima or media. Nevertheless, we speculate that a highly diffusible substance, such as NO, may be produced by transgenes in the adventitia and affect underlying vascular muscle.
When adenovirus is injected in the carotid sheath, it transfects the outer layers of adventitia but not the inner layers. Thus, after transfection, a diffusible gene product must diffuse through the inner layers of adventitia before it reaches smooth muscle of the carotid artery. This potential problem is likely to be less important in intracranial arteries, with thin adventitia, than in extracranial vessels, in which the adventitia is thicker. Thus, gene transfer to perivascular tissues and the adventitia may be more useful in intracranial vessels than in extracranial arteries.
For the reasons summarized above, catheter-based gene transfer may be useful for a variety of blood vessels, but it seems unlikely that this approach will be especially useful for gene transfer to intracranial vessels. Thus, the concept that vectors may be delivered to the CSF and perivascular tissues may be useful for gene transfer to intracranial vessels.
Functional consequences of gene transfer depend not only on the efficiency of gene transfer, in relation to the number of cells transfected, but also on the gene product. Thus, when a transgene produces a product that remains intracellular, it may be necessary to transfect virtually all of the targeted cells. For example, to inhibit the synthesis of a product that remains within the cells and is toxic to cells, if 50% of targeted cells are infected, the other 50% may remain vulnerable. In that case, it might be difficult to detect a protective effect in an organ despite successful transfection of 50% of cells.
In contrast, when the transgene product is secreted or diffuses into the extracellular space, even if it is possible to transfect only a few cells, the product may be released into the extracellular space and perhaps achieve a high concentration. Thus, if one were able to transfect 50% of cells, and they released a product into the extracellular space that normally is not present, one might be able to achieve effective concentrations of a product.
Examples of some products that are released into the extracellular space that might be efficacious, even if only a small number of cells are transfected, include genes expressing NOS, angiogenesis factors, and the extracellular isoform of SOD (EcSOD). Examples of some enzymes that remain intracellularly, and thus may require transfection of genes to virtually all targeted cells, include CuZnSOD and MnSOD. Gene transfer to choroid plexus, either by intraventricular injection or nasal instillation,22 may allow secretion of gene products into the CSF, where they may then be distributed to larger portions of brain.
Present Status of Gene Transfer and Gene Therapy
Studies of Vascular Biology
Gene transfer approaches are a novel, potentially valuable method to study vascular biology. An important question, however, is why not simply administer the gene product instead of the gene? For example, one might administer a nitrovasodilator such as nitroprusside instead of transfecting the gene for NOS. Or, one might administer IL-1ra23 instead of transfecting the gene for IL-1ra. Gene transfer approaches, however, offer some unique advantages over administration of the gene product.
First, it is possible with gene transfer approaches to study the role of different isoforms of an enzyme. For example, although one can administer nitroprusside and thus study the effects of NO, it is not possible to distinguish between the role of different isoforms of NOS (neuronal, inducible, and endothelial), all of which generate NO. The role of different isoforms of NOS is a very active area of research. For example, endothelial NOS may be protective during cerebral ischemia, but neuronal NOS24 and inducible NOS25 may be harmful. Protective versus harmful effects of different isoforms of NOS probably relate to enzyme localization and the quantity of NO produced. The role of different isoforms of NOS may be distinguished by transfection with the genes for the three isoforms of NOS. This approach is analogous to studies of transgenic mice, except that overexpression is local.
Similarly, SOD probably plays a critical role in protection against oxidative stress and perhaps in gene regulation. But the relative importance of CuZnSOD, MnSOD, and EcSOD is not known and cannot be addressed by administration of exogenous SOD, unless recombinant SOD is used. An approach to distinguish between the various isoforms of SOD is to transfect blood vessels with the genes for the different isoforms.
Second, targeted transfection might distinguish the location of an effect of a substance. For example, it might be possible to distinguish between intracellular effects (eg, MnSOD or CuZnSOD) versus extracellular effects (eg, EcSOD). In addition, it might be possible to distinguish between products released from endothelium versus adventitia. It is feasible to “target” endothelium or adventitia with a viral vector in a way that is extremely difficult with a diffusible transgene product, such as NO.
Third, it may be possible to replace genes after targeted disruption (or “knockout”) in mice. The number of knockout mice available is increasing rapidly, and correction of the abnormality with systemic (usually hepatic) transfection of a gene, or localized expression, may be useful in characterizing the gene function.
Addressing the Problems
Current approaches to gene transfer to cerebral blood vessels are on the verge of becoming a useful tool to study physiological and pathophysiological mechanisms. At the same time, it is likely that approaches currently being developed will ultimately be useful for gene therapy. As outlined in “Problems of Gene Therapy,” there are active efforts in many laboratories to develop better vectors, to develop novel approaches to delivery of vectors, and to advance gene targeting.
In gene therapy for cerebral vascular disease, an important goal is to target vessels, not neurons or perhaps glia. An alternative approach is to express a transgene that bathes the brain in a protective protein. For example, a recombinant virus with the gene for IL-1ra has been reported to increase concentrations of IL-1ra in brain and CSF 5-fold to 50-fold and to attenuate infarct volume after occlusion of the middle cerebral artery.22
Some viral promoters, such as the cytomegalovirus major immediate early promoter/enhancer, drive constitutive expression, and response elements also allow inducible expression. Preliminary studies suggest that transgene expression can be enhanced after gene transfer to the carotid artery ex vivo.19 After gene transfer, it would be potentially useful to regulate expression of the transgene. A variety of approaches may be used to regulate transgene expression, including design of recombinant adenovirus with inducible expression. Another attractive approach would be to use a promoter that responds to the gene product and thus is inhibited by excessive concentrations of the product.
The Future of Gene Therapy: Steps Versus Leaps
What is the future of gene therapy for cerebral vascular disease? Gene transfer to blood vessels has been possible for only a few years, and gene transfer to intracranial vessels has been achieved only in the past year. After the achievement of successful transfection of vessels with a reporter gene, the next step will be to alter vascular function. It is possible to alter function of the carotid artery5 and intracranial blood vessels, at least in organ culture.26 Alteration of function of cerebral vessels in vivo has not been accomplished, to our knowledge, and will be a critical next step.
Although there has been steady progress in achieving gene transfer to cerebral blood vessels, this area of research is extremely young and in many ways primitive. There are many formidable hurdles that must be cleared before gene therapy for cerebral vascular disease is a realistic possibility. It will first be necessary to develop safe and effective vectors, clinically applicable approaches to deliver vectors to extracranial and intracranial blood vessels, and, for some applications, the ability to achieve prolonged transfection.
Barriers to gene therapy are sufficiently large that it seems likely that new approaches may be needed to address the major problems, rather than simple refinement of current approaches. It is possible that gene therapy for cerebral vascular disease will burst on the scene in the next few years. It seems more likely, however, that gene therapy will be achieved in steps rather than in a great leap. Nevertheless, the opportunity to develop new methods to study cerebral vascular biology, with potential for unique therapeutic approaches, is extremely attractive.
Selected Abbreviations and Acronyms
|FGF||=||fibroblast growth factor|
|IL(-1ra)||=||interleukin(-1 receptor antagonist protein)|
|NO(S)||=||nitric oxide (synthase)|
|VEGF||=||vascular endothelial growth factor|
Original studies by the authors were supported by National Institutes of Health grants NS-24621, HL-16066, HL-14388, and AG-10269; research funds from the Veterans Administration; and funds from the Carver Trust of the University of Iowa. Frank M. Faraci is an Established Investigator of the American Heart Association. We thank Drs Beverly Davidson, Hiroaki Ooboshi, and David Rios for critical review of the manuscript and Sydney Harned for typing the manuscript.
- Received May 28, 1996.
- Accepted June 5, 1996.
- Copyright © 1996 by American Heart Association
Ooboshi H, Welsh MJ, Rios CD, Davidson BL, Heistad DD. Adenovirus-mediated gene transfer in vivo to cerebral blood vessels and perivascular tissue. Circ Res. 1995;77:7-13.
Bauters C, Asahara T, Zheng LP, Takeshita S, Bunting S, Ferrara N, Symes JF, Isner JM. Physiological assessment of augmented vascularity induced by VEGF in ischemic rabbit hindlimb. Am J Physiol. 1994;267(Heart Circ Physiol 36):H1263-H1271.
Giordano FJ, Ping P, McKirnana MD, Nozaki S, DeMaria AN, Dillmann WH, Mathieu-Costello O, Hammond HK. Intracoronary gene transfer of fibroblast growth factor-5 increases blood flow and contractile function in an ischemic region of the heart. Nature Med. 1996;2:534-539.
Nilaver G, Muldoon LL, Kross RA, Pagel MA, Breakefield XO, Davidson BL, Neuwelt EA. Delivery of herpesvirus and adenovirus to nude rat intracerebral tumors after osmotic blood-brain barrier disruption. Proc Natl Acad Sci U S A. 1995;92:9829-9833.
Von der Leyen HE, Gibbons GH, Morishita R, Lewis NP, Zhang L, Nakajima M, Kaneda Y, Cooke JP, Dzau VJ. Gene therapy inhibiting neointimal vascular lesion: in vivo transfer of endothelial cell nitric oxide synthase gene. Proc Natl Acad Sci U S A. 1995;92:1137-1141.
Ohno T, Gordon D, San H, Pompili VJ, Imperiale MJ, Nabel GJ, Nabel EG. Gene therapy for vascular smooth muscle cell proliferation after arterial injury. Science. 1994;265:781-784.
Chang MW, Barr E, Seltzer J, Jiang Y-Q, Nabel GJ, Nabel EG, Parmacek MS, Leiden JF. Cytostatic gene therapy for vascular proliferative disorders with a constitutively active form of the retinoblastoma gene product. Science. 1995;267:518-522.
Nabel EG. Gene therapy for cardiovascular disease. Circulation. 1995;91:541-548.
Finkel T, Epstein SE. Gene therapy for vascular disease. FASEB J. 1995;9:843-851.
Nabel EG, Plautz G, Boyce FM, Stanley JC, Nabel GJ. Recombinant gene expression in vivo within endothelial cells of the arterial wall. Science. 1989;244:1342-1344.
Glorioso JC, Bender MA, Goins WF, DeLuca N, Fink DJ. Herpes simplex virus as a gene-delivery vector for the central nervous system. In: Kaplitt MG, Looewy AD, eds. Viral Vectors: Gene Therapy and Neuroscience Applications. New York, NY: Academic Press Inc; 1995;1-23.
Newman KD, Dunn PF, Owens JW, Schulick AH, Virmani R, Sukhova G, Libby P, Dichek DA. Adenovirus-mediated gene transfer into normal rabbit arteries results in prolonged vascular cell activation, inflammation, and neointima hyperplasia. J Clin Invest. 1995;96:2955-2965.
Qin L, Chavin KD, Ding H, Tahara H, Favaro JP, Woodward JE, Suzuki T, Robbins PD, Lotze MT, Bromberg J. Retrovirus-mediated transfer of viral IL-10 gene prolongs murine cardiac allograft survival. J Immunol. 1996;156:2316-2323.
Ooboshi H, Rios CD, Chu Y, Christenson SD, Faraci FM, Davidson BL, Heistad DD. Augmented adenovirus-mediated gene transfer to atherosclerotic aorta. J Invest Med. 1996;44:282A. Abstract.
Christenson SD, Ooboshi H, Faraci F, Davidson BL, Heistad DD. Approaches to enhance expression after adenovirus-mediated gene transfer to carotid artery. FASEB J. 1996;10:A276. Abstract.
Rios CD, Ooboshi H, Piegors DJ, Davidson BL, Heistad DD. Adenovirus-mediated gene transfer to atherosclerotic vessels: a novel approach. Arterioscler Thromb Vasc Biol. 1995;15:2241-2245.
Huang Z, Huang PL, Panahian N, Dalkara T, Fishman MC, Moskowitz MA. Effects of cerebral ischemia in mice deficient in neuronal nitric oxide synthase. Science. 1994;265:1883-1885.
Iadecola C, Zhang F, Xu X. Inhibition of inducible nitric oxide synthase ameliorates cerebral ischemic damage. Am J Physiol. 1995;268:R286-R292.
Chen AFY, Kinoshita H, Tsutsui M, O'Brien T, Pompili VJ, Crotty TB, Katusic ZS. Effect of recombinant endothelial nitric oxide synthase gene expression on reactivity of isolated canine basilar artery. FASEB J. 1966;10:A303. Abstract.