junctival epithelium are also target tissues into which therapeutic genes can be transferred with recombinant adenovirus. In an RGC model of degen- eration, approximately 80% of RGCs could be induced to undergo apop- tosis and degenerate following intraorbital transection of the optic nerve. A single dose of adenovirus encoding brain-derived neurotrophic factor (Ad- BDNF) coinjected with a free radical scavenger, N-tert-butyl-(2-sulfoph- enyl)-nitrone (S-PBN), resulted in the survival of 63% axotomized RGCs, indicating the clinical usefulness of the approach for treating RGCs follow- ing optic nerve transection (115,116). Similar strategies were tested in the management of corneal and conjunctival abnormalities. Adenovirus type 5 (Ad 5) vector is reported to successfully deliver the reporter lacZ gene to these tissues in humans and rats. Maximum lacZ expression occurred 2–7 days after inoculation. Moreover, the nonspecific upregulation of the inflammatory cytokines IL-6, IL-8, and ICAM-1 in the tissues, induced by Ad5 infection, was suppressed with betamethasone, thereby allowing longer- term transgene expression (117). Some groups have found that other com- binations, such as coinjection of E1-deleted AV vectors carrying the lacZ reporter gene with a modified adenovirus encoding a secreted immunomo- dulatory molecule (CTLA4-Ig), could significantly reduce the immunologi- cal consequences of gene transfer with adenoviruses and, thus, promote prolonged transgene expression (118,119). Corneal opacity, a condition associated with the expression of TGF-b, is another serious cause of visual loss. The accessibility of the cornea has encouraged many in the field to turn to gene therapy alternatives to reduce this condition. An example of the potential usefulness of gene therapy approach for treating corneal opacity is shown in a study that used an adenoviral vector encoding a fusion gene containing the human type II TGF-b receptor and the Fc fragment of human IgG (AdTbeta-ExR). Transfection with the recombinant vector has proven successful in the expression of a soluble TGF-b receptor in Balb/c mice (120). High levels of the soluble receptor were found in serum and ocular fluids for at least 10 days after AdTbeta-ExR injection into the femoral muscle of the animals. Furthermore, the overexpression of soluble TGF-b receptor inhibited TGF- b signaling and may have resulted in the reduced corneal opacity observed in mice subjected to silver nitrate-induced corneal injury. Angiogensis and edema were also reduced in the injured corneas Corneal opacity, a condition associated with the expression of TGF-, is another serious cause of visual loss. the accessibility of the cornea has encouraged many in the field to turn to gene therapy alternatives to reduce this condition. An example of the potential usefulness of gene therapy approach for treating corneal opacity is sown in a study that used an ade- noviral vector encoding a fusion gene containing the human type II TGF- 574 Tombran-Tink Copyright © 2003 Marcel Dekker, Inc. receptorandtheFcfragmentofhumanIgG(AdTbeta-ExR).Transfection withtherecombinantvectorhasprovensuccessfulintheexpressionofa solubleTGF-receptorinBalb/cmice(120).Highlevelsofthesoluble receptorwerefondinserumandocularfluidsforatleast10daysafter AdTbeta-EXRinjectionintothefemoralmuscleoftheanimals. Furthermore,theoverexpressionofsolubleTGF-receptorinhibited TGF-signallingandmayhaveresultedinthereducedcornealopacity observedimicesubjectedtosilvernitrate–inducedcornealinjury. Angiogenesisandedemawerealsoreducedintheinjuredcorneaswiththe overexpressionofTGF-b.Takentogether,theresultsfromtheseexperi- mentssuggestthatadenoviral-mediateddeliveryoftherapeuticgenesisa usefulapproachtoattenuatevisualloss. c.Adeno-AssociatedViruses.Adeno-associatedviruses(AAVs)have noknownpathogeniceffectsinhumans.Itisestimatedthat80%ofthe populationdevelopantibodiestotheseviruses.AAVshaveawidehost range,ahightransductionfrequency,andtheypreferentiallyintegrate intothehumangenomeonchromosome19q13.4.Theydonotrequire hostcellreplicationforintegration.TheAAVrepandcapgenecassettes aredeletedbeforepackagingapassengergene.However,thetargetgene capacityofAAVsislimitedtoapproximately4.8kb(Table1).AAVsin- duce less host immune response than adenoviruses because a large percen- tage of their genome is deleted to accommodate the transgene. Thus, few viral proteins are expressed in vivo. The viruses are naturally replication incompetent and usually require the gene function of a coinfected adeno- virus of herpes simplex, as well as trans-complementation of the deleted rep and cap genes to generate viral progeny. Production of high AAV ti- ters is difficult to obtain and toxicity to the host is increased because of the contaminating helper virus (121,122). While this is a potentially powerful gene transfer vehicle, the transgene capacity is a limiting feature in the widespread utility of AAVs in gene transfer approaches. Until recently, transduction of normal, mature photoreceptor cells has been inefficient and limited by toxicity and host immune response directed against viral proteins. AAV transduction appears to obviate some of these problems, but their use is constrained because of passenger gene size limita- tion and low titer. Efficient transduction using recombinant AAVs, however, has been achieved in all retinal layers, the pigment epithelium, and the optic nerve. Stable transgene expression, lower cytopathology, and reduced immunogenicity have been reported after transduction with the recombinant virus in some retinal studies (123–129). In one protocol, the AAV has shown considerable potential as an effective system for delivering a functional active therapeutic gene in the Experimental Approaches to Retinal Diseases 575 Copyright © 2003 Marcel Dekker, Inc. treatment of Leber’s congenital amaurosis (LCA). LCA is a clinically severe retinal degeneration causing near total blindness in children. It is associated with a mutation in the RPE65 gene. In a breakthrough study, a recombinant AAV vector encoding the RPE65 gene was used to test the effectiveness of wide-type RPE65 in a spontaneously occurring RPE65 canine model. The dogs suffer from an early onset of visual dysfunction similar to that seen in humans affected with LCA. Intraocular innoculations with the recombinant AAV-RPE65 construct resulted in effective transduction and expression of the wild-type RPE65 gene product. Gene expression of the wild-type protein correlated with a significant improvement in visual acuity in the transfected animals (130). AAV transfer of another gene, CNTF, was shown to improve photoreceptor function in a rhodopsin knockout mouse model for retinitis pigmentosa. After transfection into the subretinal space, long-term expres- sion of the biologically active, secreted CNTF resulted in prolonged survival of photoreceptors in the rhodopsin knockout (131). In the retinal degenera- tion slow (rds) mice, another mouse model for RP, the wild-type peripherin- 2 gene (Prph2), transferred by an AAV vector, promoted ultrastructure stabilization of the photoreceptor layer in the retina. Prph2 is a photore- ceptor-specific membrane glycoprotein found in the rims of the cell’s outer segment discs. These discs contain photopigments essential for photon cap- ture during visual transduction. Prph forms a complex with the rom-1 gene product in the outersegments. The complex is essential to induce stable generation of outer segments and formation of new stacks of discs. During the study, it was noted that Prph2 transgene overexpression was associated with the reestablishment of complex ultrastructure in the photo- receptor layer. This subsequently led to an electrophysiological correction of the outer nuclear layer of the Prph2 transfected retinas (132). Efficient transduction and long-term gene expression of the wild-type bPDE were also reported to preserve photoreceptor cells after AAV transduction in the retina (133). The results, although preliminary, suggest that AAV is another potentially useful vector for transferring therapeutic genes to the human retina. d. Herpes Simplex Virus. Herpes simplex viruses (HSVs) are large DNA pathogens that infect approximately 60–90% of the world’s popula- tion. The co-evolution of this virus with humans is due, in part, to its ability to evade host immune surveillance. It establishes a latent infection in its host, a condition that allows the virus to remain unnoticed. These viruses are predominantly neurotrophic pathogens that can infect both quiescent and proliferating cells, an important feature in gene therapy protocols for central nervous system (CNS) disorders. HSVs have the po- tential to establish a lifetime latent infection in cells of the nervous system. 576 Tombran-Tink Copyright © 2003 Marcel Dekker, Inc. AcuteinfectionorreactivationoflatentHSVelicitsastrongimmunere- sponsefromthehost,oftenleadingtocornealblindnessorfatalsporadic encephalitis.ChronicepisodesofreactivationoflatentHSVresultinstro- malkeratitisandscarringcornealblindness.Theviralgenomeencodes81 knowngenes,38ofwhichareimportanttoinvitroviralreplication.Dur- ingtheconstructionofarecombinantvector,severaloftheimmediate earlygenesaredeletedtoaccommodateapassengergeneof>150kb (134).Tropism,latentinfectiveactivity,largetransgenecapacity,andthe abilitytoevadethehostimmunesystemaresomedesirablefeaturesinthe useofthisvectorforgenetherapy(Table1). Gene transfer experiments into the cornea, subconjunctiva and ante- rior chamber of the mouse eye with HSV-1 indicate that the virus is an effective gene delivery vehicle in the eye (135,136). The efficiency of HSV- mediated transfer of the lacZ gene was also tested in monkey eyes, human trabecular meshwork, and human ciliary muscle cells. Gene transfer was reported to be successful after determination of the b-galactosidase activity in the infected tissues. However, significant inflammation, mild vitritis, and retinitis were observed in the eye after infection. Transgene delivery and expression in RPE cells, optic nerve, retinal ganglion cells, and the iris epithelium were also reported with HSV (137). The possibilities for HSV as a gene delivery vehicle in retinal degenerative diseases are enormous because the virus has a large gene transfer capacity and can infect a wide range of retinal cells. However, its potential will only be fully realized with modifications that will decrease the vector’s immunogenicity and reduce packaging instability of the target gene. The above-described four viruses are currently the most advanced gene delivery system in clinical protocols, but others, including the lentivirus and human immunodeficiency viruses (HIVs), are being approached as possibi- lities for gene transfer therapy (138–143). They are endowed with features that could be advantageous to the gene therapy approach. Engineering second-generation viruses with predictable biological properties and reduced immunogenicity or developing chimeric vectors that combine the advantageous properties of several delivery systems will enhance the use of gene therapy and provide flexibility in the treatment of many diseases. 4. Non Viral Vectors One of the greatest concerns that researchers are faced with in gene therapy is the safety of viral vectors as gene delivery vehicles. Consequently, con- siderable effort has been devoted to evaluating and designing alternative strategies for gene delivery. Nonviral approaches that are currently in devel- Experimental Approaches to Retinal Diseases 577 Copyright © 2003 Marcel Dekker, Inc. opment take into consideration the size of the therapeutic gene to be deliv- ered, targeting specificity, immunogenicity, and toxicity. a. Naked DNA. Perhaps the simplest nonviral gene delivery system in use today is the transfer of naked DNA directly into cells. The overall efficiency of this method, however, is very poor when compared to viral gene transfer. Without mechanical or chemical help, naked DNA will not enter cells rapidly, and once inside, the nucleic acid is exposed and suscep- tible to enzymatic degradation. In addition, plasmid DNA carrying thera- peutic gene does not usually integrate into the host genome, and gene expression is transient in those cells that are successfully transfected. In spite of these limitations, surprisingly high levels of gene expression have been obtained in a few accessible tissues, such as skin and muscle, using plasmid DNA. In such cases, treatment is carried out by directly injecting the plasmid DNA into the tissue because the DNA is vulnerable to degra- dation in body fluids. So far, the method is safe and nontoxic, but it lacks the ability to transduce a large number of cells and requires surgical pro- cedures to access internal tissues. An improved strategy for delivery nucleic acid directly into cells is by high velocity bombardment of the cells with DNA attached to gold particles using a ‘‘gene gun’’ approach. Microparticle bombardment has shown some impressive results in focal delivery of naked DNA to corneal cells with little damage or irritation to the tissue (144). It is a method that is being devel- oped for more widespread use and may be a solution to some of the pro- blems encountered with viral vectors. Most gene transfer studies in the eye are carried out using viral vector, but plasmid delivery of a few therapeutic genes, such as tissue plasminogen activator and IFN-, has been tested and shows potential benefits in treating corneal-related pathologies (145–157). A few studies have also reported successful gene expression in the retina using plasmid DNA. In one case it was demonstrated that condensed plasmids containing the human fibroblast growth factor genes were able to transduce a small population of choroidal and RPE cells after subretinal injections into RCS rat eyes. FGF gene expression in those tissues consequently resulted in a delay in photoreceptor degeneration (148). While the current methods of delivering naked DNA are still very inefficient, the eye is in a prime location to benefit from improvements in mechanical delivery strate- gies that increase the therapeutic index of this approach. b. Liposomes. Liposomes are probably the most widely known non- viral vectors used to transfer DNA into cells. The strategy involves encap- sulation of plasmid DNA in lipid complexes that are capable of fusing with the cell membrane and delivering the therapeutic genes intracellu- larly. Initially, this approach has encountered difficulties because classical 578 Tombran-Tink Copyright © 2003 Marcel Dekker, Inc. liposomes are negatively charged lipids that do not interact spontaneously with DNA. Charge limitations and the need to separate DNA-liposome complexes after delivery have led to the development of positively charged cationic lipids. These interact with DNA more readily and have proven to be valuable tools that can compact and deliver DNA across the cell mem- brane with greater efficacy (149–170). Cationic liposomes are typically for- mulated using a positively charged lipid and a co-lipid that will stabilize the DNA complex. A commonly used formulation is a mixture of a cyto- fectin with a neutral lipid component such as DOPE. This combination can be formulated into unilammellar vesicles by several methods, includ- ing reverse phase evaporation and microfluidization. Stable complexes are formed when DNA is combined with the vesicles. The DNA is subse- quently condensed in the vesicles, forming nanometric particles that are referred to as lipoplexes. These complexes protect the DNA, interact with cell surface proteoglycans, and enter the cell by endocytosis (170) (Fig. 7). Experimental Approaches to Retinal Diseases 579 Figure 7 Liposome-mediated transfer of nucleic acid. The nucleic acid is con- densed in the liposome to form lipoplexes. These enter the cell by endocytosis. The majority of lipoplexes are trapped in late endosome. A small percentage can either be released into the cytoplasm (mRNA), where they are functional, or traffic non-specifically to the nucleus, where they may form episomes. Copyright © 2003 Marcel Dekker, Inc. Currently, no more than 30 genes transfer–competent cationic liposomes have been developed and are commercially available. Perhaps the most widely used formulations are DOTAP and DOTMA, the latter of which is sold as Lipofectin, an in vitro transfecting agent. A disadvantage of current methods of liposome-mediated gene trans- fer is due to the large percentage of the DNA-bound complexes trapped in late endosomes, where they undergo enzymatic degradation and are no longer therapeutically useful. Only a small percentage of the bound nucleic acid escapes systemic inactivation and endosome entrapment. Those that manage to escape face yet another hurdle of getting to the nucleus and maintaining their functional integrity. As with viral delivery, in vitro effec- tiveness of liposome-mediated gene transfer is often misleading and is a poor guide for clinical efficacy. Conventional liposome formulations lack cell specificity and can take hours for uptake into the cell. They are highly susceptible to inactivation by a number of serum proteins that bind and cause membrane destabilization, a major obstacle for systemic administra- tion of liposomes. A current research focus in the pharmaceutical industry is the development of sterically stable liposome formulations that are resistant to serum disruption and that will not aggregate prior to delivery. One mod- ification currently in development uses conventional liposome lipid mem- branes to covalently attach polymers such as polyethylene glycol (PEG- lipid) to create stealth liposomes (152,155,156,162,169). Properly formulated polymer-grafted liposomes are shown to be sterically stabilized compounds that have long residence times in circulation, increased biodistribution, and reduction in uptake by cells of the reticuloendothelial system. Other clini- cally advantageous features of pegylated liposome pharmacokinetics include dose independence and increased efficacy as a slow release system for ther- apeutically active drugs. This is a fascinating technology that has the poten- tial of being a tailor-made delivery system that will improve the therapeutic index of a number of drugs. Another modification strategy under active investigation is the manu- facture of ligand-targeted liposomal drugs using combinatorial approaches. Such molecular conjugates could potentially be more versatile than the conventional systems (159–164). In a recent study, transfection was observed to be increased in hepatoma cells after the administration of mod- ified lipoplexes containing triantennary galactosyl residues that specifically target hepatoma cells (172). Targeted delivery of doxorubicin to human umbilical vein endothelial cells and subsequent decrease in the survival of the cells were also achieved with immunoliposomes that were conjugated to a monoclonal antibody against E-selectin, a surface marker of HUVECs (173). While targeting will increase transfection efficiency to specific tissues, it does not address problems of DNA release encountered in the endosomes. 580 Tombran-Tink Copyright © 2003 Marcel Dekker, Inc. Some researchers have shown that the association of amphiphilic peptides, such as GALA, a pH-sensitive peptide, with cationic liposomes can induce fusion and permeabilization at acidic pH values and improve release of the DNA from endosomes. The peptides induce osmotic swelling and subse- quent rupture of the endosomal membranes so that the DNA can escape easily (174). These new modifications, however, are not without problems. Competition between ligand-mediated processes and nonspecific interac- tions with the cell membrane can hinder the efficacy of gene delivery and must be resolved before ligand-modified liposomes are of clinical relevance. In addition to engineering modifications that result in specific cell targeting and more efficient DNA release, mechanisms that will increase nucleic acid condensation and promote nuclear targeting are promising areas of research that will improve liposome-mediated gene delivery. Modifications using DOTAP liposome-protamine sulfate-DNA (LPD) for- mulations are shown to produce denser particles when bound to DNA and result in consistently higher gene expression levels (175,176). Complexes formed with polycations are also observed to be much smaller than those formed with liposomes alone and have increased resistance to nuclease degradation. Smaller-size complexes may allow for higher levels of gene expression because of increased cellular internalization. An interesting var- iation to this hybrid concept is the use of UV-irradiated Japanese Sendai virus (HVJ)-cationic liposome to facilitate nuclear targeting. The binding of high mobility group 1 protein (HMG-1) increase the potency of the complex and enhances nuclear targeting and stability of the DNA after delivery into the nuclear envelope. The success of HVJ-liposome complexes in cancer applications is thought to result from the ability of the complex to bypass the endocytosis process, thereby minimizing the difficulties encountered when the DNA is released from the endosomes (171). The development of ‘‘synthetic chemical viruses’’ that are capable of (a) extended blood circula- tion, (b) increased DNA microparticle condensation, (c) improved cellular uptake, (d) flexible tropism, (e) escaping enzymatic degradation, and (f) nuclear targeting is an attractive challenge in the area of biopharmaceutics. If realized, such compounds have enormous potential in gene therapy pro- tocols and may surpass the clinical usefulness of viral vectors. Liposome-based techniques have been optimized to successfully trans- fer functional genes into human primary RPE cells. In one study, differences in the efficiency of transfection were observed between the types of lipo- somes used in the assay. Nontoxic transfer was achieved after each liposome treatment, but the Tfx-50 formulation showed the most significant results when compared to transfection of the RPE cells with other liposome varia- tions, including lipofectin, lipofectamine, Cellfectin, and DMRIE-C (177). A fascinating variation of gene transfer by liposomes was achieved by a Experimental Approaches to Retinal Diseases 581 Copyright © 2003 Marcel Dekker, Inc. group of researchers who used liposome eye drops to transfer rat retinal ganglion cells. Transfection was reported to be efficient and nontoxic to ocular tissues. This approach represents an interesting development in non- surgical gene delivery for retinal diseases (178). The use of another liposome method, hemagglutinating virus of Japan liposomes, was tested for efficacy in delivering tissue inhibitor of metalloproteinase-3 gene into rat RPE cells. Not only was the transfection successful, but expression of the introduced gene inhibited the development of experimental choroidal neovasculariza- tion induced by laser photocoagulation after transfection of the tissue (179). These are only a few examples showing the feasibility of using, nonviral, nontoxic synthetic DNA-complexing derivatives to transfer therapeutic genes to the retina. The development of innovative nonviral delivery system is still in its infancy, but many advantages are associated with their use in gene transfer applications: (a) they can package and deliver a transgene of any size; (b) packaging cell lines are not required to generate high titers; (c) they are non- pathogenic and cannot replicate; (d) immunogenicity, toxicity, and inflam- mation are minimized with their use; and (e) they can become completely synthetic. While these are safe gene delivery systems, the disadvantages currently lie in their overall inefficiency of transfection and their inability to achieve cell-specific and nuclear targeting. Modifications that improve these features will allow synthetic polymer-based gene vectors to be the candidates of choice for pharmacological intervention in many diseases. 5. Gene Knock down Therapy a. Antisense Drugs. Antisense technology is a novel gene delivery method that is increasingly applied to knock down the expression of a specific target gene for therapeutic purposes or to study the function of that gene. The fundamental principle of the antisense approach is to si- lence a gene using a short synthetic DNA or RNA sequence that is homo- logous to that contained within the target gene. Antisense oligodeoxynucleotides (ODNs) are synthesized in the opposite direction of the known complementary DNA sequence and are designed to hybridize specifically with their target sequences to interrupt the production of the corresponding protein. Almost all human diseases are associated with a dysfunctional protein. While most conventional drugs are designed to in- hibit the disease-causing activity of a dysfunctional protein, antisense mo- lecules are designed to inhibit the production of the protein. In principle, gene silencing may be accomplished at the genetic level by inhibiting bio- logical events, such as transcription, translation, or gene splicing (180– 183). During the inhibition of transcription events, ODNs bind to double- 582 Tombran-Tink Copyright © 2003 Marcel Dekker, Inc. stranded DNA to induce the formation of a short triple-helical structure. This structure is mediated by Hoogsteen hydrogen bonds and sterically hinders the transcription of a specific mRNA. In addition, translation of RNA species can be interrupted by binding of ODNs to tRNA or pre- RNA to prevent their transport from the nucleus or by directly interacting with target mRNA molecule after transcription. In cases where inhibition occurs after the transcript is matured, antisense binding to RNA is in- tended to block ribosomal assembly or ribosomal sliding along the mRNA during translation of the protein. ODNs can also be of therapeu- tic value if they are designed to target the intron-exon junctions of prema- ture RNA. In this regard, they prevent splicing events that are essential for maturation of the RNA transcript. Three regions of the RNA that are considered the best targets when designing ODNs are the 5 0 Cap region, the AUG translational initiation codon, and the 3 0 untranslated region. The concept of disabling the function of a mRNA by hybridization of antisense reagents is a simple one, but, like other gene-based therapies, the technology has encountered difficulties in the past. The technical problems experienced in the early pioneering stages of antisense technology are only now being elucidated and are the focus of active study. From these analyses, several features are apparent in the design of effective antisense molecules: determining the length of sequence with the greatest activity and specificity; cellular uptake; specific targeting of the ODN; antisense stability; and toxi- city. Other factors that have influenced the effectiveness of antisense mole- cules are frequency of protein turnover, the intracellular environment of the cell, and the extent of longevity of ODNs after administration. Gene knockdown practices are still under development and will require significant modifications before being clinically acceptable as a ther- apeutic modality. Introducing variations in antisense chemistry by subtle changes in the phosphate or sugar moieties of the nucleic acid backbone is one method that shows success in minimizing nuclease degradation of the molecules. Replacing a nonbridging oxygen with a sulfur atom in the phos- phodiester bond between nucleotides on the phosphate backbone generates a phosphorothioate linkage, which is reported to be one of the most success- ful modification of antisense oligonucleotides to date (184,195). Phosphorothioate compounds have shown efficacy in delivery and are less vulnerable to intracellular nuclease degradation. A disadvantage of their use, however, is that the constructs are chiral and form a racemix mixture of ODN species that exhibit both desirable and undesirable properties in vivo (186). Some ODNs are reported to be toxic, while others show non- specific affinity for proteins (1987). The technical progress in chemical mod- ification of antisense has recently shifted from the first-generation phosphodiester oligonucleotides, which are still nuclease sensitive, to the Experimental Approaches to Retinal Diseases 583 Copyright © 2003 Marcel Dekker, Inc. [...]... ( 199 7) Efficient photoreceptor-targeted gene expression in vivo by recombinant adeno-associated virus Proc Natl Acad Sci USA, 94 (13): 691 6– 692 1 125 Grant, C A., Ponnazhagan, S., Wang, X S., Srivastava, A., and Li, T ( 199 7) Evaluation of recombinant adeno-associated virus as a gene transfer vector for the retina Curr Eye Res., 16 (9) :94 9 95 6 126 Lai, Y K., Rakoczy, P., Constable, I., and Rolling, F ( 199 8)... fibrous protein Nature 401 :93 5 93 8 98 Shenk, T ( 199 6) Adenoviridae The viruses and their replication In: Fields Virology (B N Fields, D M Knippe, and P Howley, eds.) Lippincot-Raven, Philadelphia, pp 2111–2148 99 Stratford-Perricaudet, L D., Makeh, I., Perricaudet, M., and Briand, P ( 199 2) Wide-spread long-term gene transfer to mouse skeletal muscles and heart J Clin Invest., 90 :626–630 100 Bett, A J.,... intravenous administration of cationic lipid-protamine-DNA (LPD) complexes Gene Ther., 4 (9) : 891 90 0 176 Li, S., Rizzo, M A., Bhattachrya, S., and Huang, L ( 199 8) Characterization of cationic lipid-protamine-DNA (LPD) complexes for intravenous gene delivery Gene Ther., 5(7) :93 0 93 7 177 Abul-Hassan, K., Walmsley, R., and Boulton, M (2000) Optimization of non-viral gene transfer to human primary retinal... Hansen, C., and Chin, Y C ( 199 2) Stealth liposomes: an improved sustained release system for 1-beta-D-arabinofuranosylcytosine Cancer Res., 52 (9) :2431–24 39 157 Gabizon, A., and Martin, F ( 199 7) Polyethylene glycol-coated (pegylated) liposomal doxorubicin Rationale for use in solid tumors Suppl 4:15–21 158 Newman, M S., Colbern, G T., Working, P K., Engbers, C., and Amantea, M A ( 199 9) Comparative pharmacokinetics,... and Wyatt, J ( 199 7) Prospects for a visual prosthesis Neuroscientists, 3:251–262 67 Eckmiller, R ( 199 7) Learning retina transplants with epiretinal contacts Ophthalmol Res., 9: 281–2 89 68 Peyman, G., Chow, A Y., Liang, C., Chow, V Y., Perlman, J I., and Peachey, N S ( 199 8) Subretinal semiconductor microphotodiode array Ophthalmic Surg Lasers, 29( 3):234–241 69 Chow, A Y., and Chow V Y ( 199 7) Subretinal... to RNA Curr Opin Mol Ther., 2(3):264–271 190 Srinivasan, S K., and Iversen, P ( 199 5) Review of in vivo pharmacokinetics and toxicology of phosphorothioate oligonucleotides J Clin Lab Anal., 9( 2):1 29 137 191 Lambert, G., Fattal, E., and Couvreur, P (2001) Nanoparticulate systems for the delivery of antisense oligonucleotides Adv Drug Deliv Rev., 47(1) :99 –112 192 Lebedeva, I., and Stein, C A (2001) Antisense... 41:403–4 19 193 de Smet, M D., Meenken, C J., and van den Horn, G J ( 199 9) Fomivirsen—a phosphorothioate oligonucleotide for the treatment of CMV retinitis Ocul Immunol Inflamm., 7(3–4):1 89 198 194 Flores-Aguilar, M., Besen, G., Vuong, C., Tatebayashi, M., Munguia, D., Gangan, P., Wiley, C A., and Freeman, W R ( 199 7) Evaluation of retinal toxicity and efficacy of anti-cytomegalovirus and anti-herpes simplex... Acad Sci USA, 94 ( 19) :103 19 10323 143 Sikorski, R., and Peters, R ( 199 8) Gene therapy Treating with HIV Science, 282(5 393 ):1438 144 Tanelian, D L., Barry, M A., Johnston, S A., Le, T., and Smith, G ( 199 7) Controlled gene gun delivery and expression of DNA within the cornea Biotechniques, 23(3):484–488 145 Sakamota, T., Oshima, Y., Nakagawa, K., Ishibashi, T., Inomata, H., and Sueishi, K ( 199 9) Target gene... Ophthalmol., 48(4):263–278 8 Ho, Ac ( 199 9) Laser treatment in eyes with drusen Curr Opin Ophthalmol., 10(3):204–208 9 Hageman, G S., and Mullins, R F ( 199 9) Molecular composition of drusen as related to substructural phenotype Mol Vis., 5:28 10 Blodi, C F., and Stone, E M ( 199 0) Best’s vitelliform dystrophy Ophthal Paediatr Genet., 11(1): 49 59 11 Bither, P P., and Berns, L A ( 198 8) Stargardt’s disease: a review... site-selective delivery in the cardiovascular system Proc Natl Acad Sci USA, 94 (16):8 795 –8800 Copyright © 2003 Marcel Dekker, Inc 604 Tombran-Tink 174 Simoes, S., Slepushkin, V., Gaspar, R., de Lima, M C., and Duzgunes, N ( 199 8) Gene delivery by negatively charges ternary complexes of DNA, cationic liposomes and transferrin of fusigenic peptides Gene Ther 5(7) :95 5 96 4 175 Li, S., and Huang, L ( 199 7) . adenovirus encoding brain-derived neurotrophic factor (Ad- BDNF) coinjected with a free radical scavenger, N-tert-butyl-(2-sulfoph- enyl)-nitrone (S-PBN), resulted in the survival of 63% axotomized. B. (2000). Survival of long-term retinal cell transplants. Acta. Ophthalmol. Scand., 78: 396 –402. 29. Aramant, R. B., Seiler, M. J., and Ball, S. L. ( 199 9). Successful cotransplanta- tion of intact sheets. Arch. Clin. Exp. Ophthalmol., 238(12) :97 9 98 4. 32. Schraermeyer, U., Kociok, N., and Heimann, K. ( 199 9). Rescue effect of IPE transplants in RCS rats: short-term results. Invest. Ophthalmol. Vis.