Gene Therapy for Retinal Ganglion Cell Neuroprotection in Glaucoma

REVIEW Gene therapy for retinal ganglion cell neuroprotection in glaucoma

AM Wilson1,2 and A Di Polo1,2,3

Glaucoma is the leading cause of irreversible blindness worldwide. The primary cause of glaucoma is not known, but several risk factors have been identiﬁed, including elevated intraocular pressure and age. Loss of vision in glaucoma is caused by the death of retinal ganglion cells (RGCs), the neurons that convey visual information from the retina to the brain. Therapeutic strategies aimed at delaying or halting RGC loss, known as neuroprotection, would be valuable to save vision in glaucoma. In this review, we discuss the signiﬁcant progress that has been made in the use of gene therapy to understand mechanisms underlying RGC degeneration and to promote the survival of these neurons in experimental models of optic nerve injury. Gene Therapy (2012) 19, 127–136; doi:10.1038/gt.2011.142; published online 6 October 2011

Keywords: retinal ganglion cell; neuroprotection; glaucoma

NEUROPROTECTION IN GLAUCOMA: RATIONALE AND 12 causative genes have been identified by linkage studies.5 Never- CURRENT LIMITATIONS theless, further work is required to clarify the relationship between Adult-onset glaucoma is an optic neuropathy that affects more than 50 gene defects and RGC pathophysiology in glaucoma before gene million people around the world, with 47 million having bilateral supplementation therapies can be developed. In addition, multiple (both eyes) blindness due to this disease. As more effective diagnostic genes, individual risk factors and environmental factors are likely to tools become available, the number of cases is expected to rise, and a contribute to glaucoma onset. For these reasons, most gene therapy recent study predicts that there will be 70 million people with strategies for RGC neuroprotection, to date, are based on enhancing glaucoma in 2020.1 Glaucoma is considered to be a group of diseases neuronal survival rather than correcting the primary genetic defect or characterized by progressive optic nerve degeneration that results in injury source. As RGC death in human glaucoma typically occurs over visual field loss and irreversible blindness. The cause of glaucoma is several decades, strategies that do not address the initial insult, but unknown, but several risk factors have been identified, including high might delay RGC death and support vision, are likely to be beneficial. intraocular pressure, age and genetic predisposition. Open angle and Effective neuroprotective strategies should take into account both angle-closure glaucoma, the most common forms of the disease, are the mechanisms of RGC death and the tools to halt or slow down this often associated with high intraocular pressure. A crucial element in process. Gene therapy has gained considerable ground as a strategy to the pathophysiology of all forms of glaucoma is the death of retinal promote RGC neuroprotection after injury. With the advent of viral ganglion cells (RGCs; Figure 1) and, as these are central nervous and nonviral agents suitable for in vivo gene delivery, a number of system neurons, their loss is irreversible. At present, there is no cure studies have now established proof of concept for gene therapy for glaucoma, and the standard treatment is to lower intraocular approaches to enhance the survival capacity of injured RGCs. This pressure by medication or surgery. However, a significant proportion review focuses on the progress made to date on gene therapy strategies of patients continue to experience visual loss even when pharmaco- to promote RGC neuroprotection in experimental models of acute logical treatments effectively reduce eye pressure. Therefore, novel and chronic optic nerve damage. We also highlight the areas where therapeutic approaches are needed for its treatment and management. there has been paucity in the use of gene therapy to manipulate Early-onset glaucoma, comprising primary congenital glaucoma pathways involved in RGC death, and discuss the challenges that lie and juvenile open-angle glaucoma, has a clear genetic basis. Between ahead to move gene therapy toward clinical trials for RGC neuropro- 10 and 30% of individuals with juvenile open-angle glaucoma have tection in glaucoma. The regulation of intraocular pressure through mutations in the gene encoding myocilin,2,3 and, on average, B40% gene therapy targeted to the anterior structures of the eye has been of people with primary congenital glaucoma have mutations in the reviewed recently elsewhere6 and will not be addressed here. gene encoding cytochrome P450 protein.4 Both genes are expressed in the trabecular meshwork and ciliary body, and defects in these genes TOOLS FOR GENE DELIVERY TO INJURED RGCS may cause ocular hypertension. The elucidation of the genetic basis Given the wide range of tools currently available for retinal gene underlying adult-onset glaucoma is difficult because the late onset of transfer, we will focus here only on those that have had a significant disease limits the number of individuals available for linkage analysis. impact on studies of RGC neuroprotection and discuss the advantages However, at least 29 genetic loci for various forms of glaucoma and and disadvantages of each system.

1Department of Pathology and Cell Biology, Universite´ de Montre´ al, Montreal, Quebec, Canada; 2Groupe de Recherche sur le Syste`me Nerveux Central, Universite´ de Montreál, Montreal, Quebec, Canada and 3Department of Ophthalmology, Universite´ de Montreál, Montreal, Quebec, Canada Correspondence: Dr A Di Polo, Department of Pathology and Cell Biology, Universite´ de Montreál, 2900, Boulevard Edouard-Montpetit, Pavillon Roger-Gaudry, Room N-535, Montreal, Quebec H3T 1J4, Canada. E-mail: [email protected] Received 24 August 2011; accepted 30 August 2011; published online 6 October 2011 Gene therapy for retinal ganglion cell AM Wilson and A Di Polo 128

Figure 1 RGCs die in glaucoma. High intraocular pressure is a major risk factor for developing glaucoma. Ocular hypertension damage leads to the death of RGCs, the neurons that convey visual information from the retina to the brain. Loss of RGCs results in progressively worsening vision and blindness.

Going viral: the use of virus vector-based approaches to target been the size constraint for packaging genes larger than 4.7 kb. RGCs or surrounding cells Although methods have been developed to increase the size of Adeno-associated viral vectors. Among all currently available viral delivered transgenes by trans-splicing two independent vectors coad- vector systems, the adeno-associated virus (AAV) vector has emerged ministered to the same tissue,16 this remains a limitation of the AAV as a favored tool for targeting adult RGCs. The ability of AAV vectors system. The laborious work needed to produce AAV vectors has often to transduce distinct retinal cell types depends on the virus serotype, been regarded as a disadvantage. However, recent improvements in the the route of vector administration and the age of the host animal. protocols have facilitated the preparation of high-titer and pure AAV Early studies in adult rat and mouse retinas identified AAV as an stocks following good manufacturing practice, which is a requirement effective tool to transduce a large number of RGCs following intra- for use in humans. vitreal injection of the vector,7–9 whereas subretinal AAV administration leads primarily to transduction of photoreceptors and retinal Adenoviral and lentiviral vectors. The most widely used adenoviral pigment epithelium (RPE).10,11 A variety of different AAV serotypes vectors for ocular gene transfer are based on the adenovirus (Ad)-5 have been tested for retinal gene transfer, but the general consensus is backbone. These vectors were initially generated with deletions of the that AAV-2 vectors provide the most efficient transduction of RGCs early region 1 (DE1) that contains genes required for virus replication, after intravitreal injection.12 Although RGC-specific promoters have rendering vectors replication defective and more suitable for gene not yet been used to drive transgene expression following AAV transfer into mammalian cells. A major disadvantage of these early Ad transduction, constitutive viral promoters, such as the cytomegalo- vectors is the strong cytotoxic and immune reaction elicited upon virus promoter, with a chicken b-actin enhancer have been shown to transduction of host cells.17 Recent versions of Ad vectors have been be effective. The combination of a chicken b-actin-containing pro- produced in which the entire viral genome, except for the terminal moter with the woodchuck hepatitis posttranscriptional regulatory repeat regions required for viral assembly, has been replaced by element allowed transduction of B85% of rat RGCs within 2 weeks exogenous gene sequences. These so-called ‘gutless’ vectors, also of intravitreal virus injection.13 Tetracycline-regulatable promoter referred to as helper-dependent Ad, exhibit considerably reduced systems are also being investigated as a means to regulate AAV- immune response, but can only be produced in the presence of a mediated transgene expression in RGCs.14 helper virus that provides all the proteins required for viral replica- Other advantages of the AAV vector system for in vivo gene delivery tion.18 These new vectors are less immunogenic and mediate long- include that it is not pathogenic and has not been implicated in the term transgene expression following subretinal or intravitreal injec- etiology of any known human disease, it mediates long-term transgene tion.19–22 For example, intravitreal administration of helper-depen- expression that can last for several years in the retina15 and it has low dent Ad promoted transgene expression for up to 1 year in transduced immunogenicity. In the absence of helper virus, wild-type AAV can Mu¨ller glia.22 integrate at a specific site on the 19q to establish latent infection. A cardinal feature of Ad-5 vectors in the adult retina is that following However, the lack of Rep proteins has been shown to compromise intraocular injection, they efficiently transduce nonneuronal cell types— integration specificity, leading to random insertion of recombinant epithelial cells or glia—whereas their ability to infect neurons is poor. AAV. Although viral integration into the genome may contribute to Intravitreal injection of Ad-5 results in preferential transduction of the stability of AAV-mediated transgene expression, a careful evalua- Mu¨ller cells,23 an approach that has been useful to deliver genes encoding tion of the risks associated with insertional mutagenesis is required diffusible neurotrophic factors to promote RGC neuroprotection.23,24 before implementing AAV-based therapies. A disadvantage of AAV has Ad-5 vectors have also been shown to effectively transduce the RPE

Gene Therapy Gene therapy for retinal ganglion cell AM Wilson and A Di Polo 129 following subretinal injections,25,26 and transduced RPE cells have the expression using electroporation can be detected as early as 2 or 3 days potential to produce diffusible factors that enhance RGC survival. after gene delivery in vivo and can last up to 3 weeks. This approach Further studies demonstrated that the arginyl-glycyl-aspartic acid has been used successfully to deliver neurotrophic factors and TXN domain, an integrin-binding motif, was responsible for the RPE thioredoxin genes to promote RGC survival in several models of optic tropism of Ad-5 vectors, and its deletion significantly increased transgene nerve injury.33–35 expression in the neural retina.27 Under some experimental conditions, Small interference RNA (siRNA) has been successfully delivered to limited transduction of RGCs by Ad-5 has been observed. For example, RGCs via injection into the superior colliculus;36,37 however, the introduction of Ad-5 to the superior colliculus or to the transected highly invasive nature of this approach limits its clinical application. optic nerve stump results in retrograde transport of viral particles and Alternatively, siRNA can be injected intravitreally, leading to effective subsequent gene expression in some RGCs.28 delivery to RGCs soon after administration (Figure 2). Although other Subretinal injection of lentiviral vectors results in transgene retinal cells are likely to uptake siRNA delivered to the vitreous expression primarily by RPE cells, although other cells, including chamber, this strategy might be suitable for silencing genes that are some photoreceptors, bipolar cells and Mu¨ller glia, may occasionally specific or highly enriched in RGCs. be transduced (see review by Balaggan et al.29). Some studies have observed limited transduction of the inner retina when HIV-1 CRACKING THE CASE: GENE THERAPY STRATEGIES FOR RGC vectors are delivered intravitreally,8 whereas others have reported NEUROPROTECTION some transduction of RGCs following intravitreal injection of There is little doubt that RGCs are the ultimate victims in glaucoma, HIV-2 vectors.30 but the primary culprits responsible for their demise are still at large. Glaucoma is an age-related, multifactorial disease and it is likely that DNA- and RNA-based technologies to modify RGC gene expression several pathways converge to induce RGC loss. Although the signals Nonviral gene transfer strategies have the advantage of circumventing that contribute to RGC degeneration in glaucoma are not well the safety concerns that stem from the potential immunogenic understood, a number of mechanisms have been proposed, including response and risk of chromosomal integration associated with viral neurotrophic factor deprivation, activation of pro-apoptotic pathways, vectors. The downside of nonviral approaches is that they yield low excitotoxic damage, oxidative stress, growth-inhibitory signals and transduction rates, resulting in limited and often short-lived gene reactive gliosis. The next sections discuss the use of gene delivery expression in vivo. These gene transfer strategies might be useful when systems to increase our knowledge of the mechanisms that lead to transgene expression is required only during a critical time window to RGC death after optic nerve damage and to manipulate molecular boost RGC survival. Sustained transgene expression using nonviral pathways to promote RGC survival in vivo. vectors might be attainable through multiple intraocular injections, topical (corneal) applications or intranasal delivery. A bad case of neuronal neglect: neurotrophic factor deprivation DNA plasmids or oligonucleotides are easy to prepare and can be During development of the nervous system, young neurons require readily injected into the eye, but they are not easily taken up by cells. neurotrophic factors for their survival, differentiation and the estab- Antisense oligonucleotides against pro-apoptotic molecules have been lishment of synaptic connections. Neurotrophic factors are produced delivered to RGCs by intravitreal injection or by retrograde transport in limited amounts; therefore, only neurons that are exposed to via injection in the superior colliculus, but these approaches resulted optimal levels of these factors survive, whereas the rest die by in only modest protection of axotomized RGCs due to limited apoptosis. There is substantial evidence that deprivation of neurotro- transfection efficiency.31,32 Although liposomes have been shown to phins, mainly brain-derived neurotrophic factor (BDNF) and nerve facilitate intravitreal or topical delivery of DNA to the inner retina, the growth factor secreted by targets in the brain, leads to apoptotic death usefulness of this approach in RGC neuroprotection remains largely of RGCs during visual system development.38 It has been proposed unexplored. Among the physical methods to deliver DNA plasmids to that blockade of axonal transport during eye pressure elevation in RGCs, electroporation has received particular attention because it is glaucoma leads to scarcity of target-derived neurotrophic factors and safe, while providing effective gene transfer to these neurons in vivo. neuronal death. Indirect support for this hypothesis is provided by the Electric field strength, pulse duration and stimulation pattern can be observation that both anterograde and retrograde axonal transport in controlled to optimize transduction efficiency of RGCs. Transgene the optic nerve are impaired in animal models of glaucoma.39,40

Figure 2 RGC uptake siRNA injected into the vitreous chamber. Intravitreal injection of a Cy3-labeled siRNA results in robust Cy3 labeling of RGCs, visualized with Fluorogold (FG) (Fluorochrome, Denver, CO, USA). Scale bars¼40 mm. GCL, ganglion cell layer; INL, inner nuclear layer; IPL, inner plexiform layer; ONL, outer nuclear layer; OPL, outer plexiform layer; PS, photoreceptor segment.

Gene Therapy Gene therapy for retinal ganglion cell AM Wilson and A Di Polo 130

Local sources of neurotrophic factors within the retina might also hand, promotes substantial axon regrowth in the order of 100-fold have a prominent role in adult RGC survival. Evidence for this idea increase in the number of RGC axons that regenerate beyond the crush stems from studies where RGC target tissue was ablated in adult site,58,59 an effect that has been partly attributed to oncomodulin, a animals, but neuronal loss was not detected until several months post novel macrophage-derived factor that promotes cAMP-dependent lesion.41–43 These findings suggest that retina-derived neurotrophic axonal regeneration.60 factors may temporarily support the survival of RGCs that have been Long-term studies have demonstrated that the pro-survival effect of disconnected from their targets. Collectively, these studies suggest that BDNF on RGCs is temporary: it delays, but does not prevent, the lack of neurotrophic factors, derived either from the target or the onset of RGC death.23,49 Administration of BDNF protein by repeated retina, may have a role in RGC death in glaucoma. Consequently, the intravitreal injections or osmotic mini pumps failed to extend the time supplementation of neurotrophic factors, either as exogenous proteins course of RGC neuroprotection.49,61 Adenoviral vector-mediated gene or by gene therapy, has been extensively explored as a neuroprotective delivery of BDNF to Mu¨ller cells provided a sustained source of strategy for injured RGCs. Because neurotrophic factors are secreted neurotrophin, leading to a 4.5-fold increase in the number of surviv- by the target cells, it is not a requirement for the transgenes to be ing RGCs at 16 days after axotomy, but this neuroprotective effect was expressed by RGCs. In fact, both autocrine or paracrine expression of transient and declined at 28 days post injury.23 BDNF exerts its neurotrophic factors by surrounding cells, which then influence the biological effect upon binding to its cognate receptor TrkB, therefore response of injured RGCs, have been an attractive gene therapy the transient survival effect of BDNF was attributed to a reduction in strategy23,24,44,45 (Figure 3). RGC cell surface receptor levels.62 Consistent with this, AAV-mediated Among neurotrophic factors, the neurotrophin BDNF stands out gene transfer of TrkB to RGCs markedly increased the duration and for its potent protective effect on injured RGCs. Intraocular injection level of BDNF-induced neuroprotection of these neurons.7 Specifi- of exogenous BDNF protein or viral-mediated BDNF gene transfer cally, AAV.TrkB gene therapy, combined with exogenous BDNF using Ad (Ad.BDNF) or AAV (AAV.BDNF) promotes robust RGC administration, increased the survival of RGCs by 76% at 2 weeks survival after optic nerve transection.23,46–51 The combination of after optic nerve lesion, a time when o10% of these remain in non- BDNF gene therapy with additional therapies, including free radical treated retinas, hence substantially extending RGC survival well scavengers and cell-permeable cyclic AMP (cAMP), further increases beyond the shorter effect of BDNF protein alone. RGC neuroprotection.24,52 In experimental glaucoma induced by Another prominent neurotrophic factor that effectively promotes chronic eye pressure elevation, intraocular delivery of BDNF protein53 RGC neuroprotection in addition to axon regrowth is the ciliary or BDNF gene transfer using an AAV vector led to marked RGC neurotrophic factor (CNTF). Exogenous or endogenous CNTF pro- neuroprotection.54 For example, at 4 weeks of ocular hypertension, tein has been shown to stimulate RGC survival after optic nerve 32% of RGC axon loss was reported in eyes treated with AAV.BDNF axotomy or crush63,64 in experimental glaucoma65,66 and acute auto- compared with 52% in control eyes.54 In spite of its robust pro- immune optic neuritis.67 Four weeks after laser-induced ocular survival effect, BDNF does not stimulate RGC axon regeneration after hypertension, there was a 15% increase in the number of RGC optic nerve injury. Administration of BDNF protein or AAV.BDNF did axons in eyes treated with AAV.CNTF compared with eyes treated not improve RGC axonal regrowth within the optic nerve or into with a control vector.66 Adenoviral-mediated CNTF gene transfer was peripheral nerve autologous grafts.55,56 Intriguingly, although the reported to increase retinal CNTF and CNTF receptor-a levels, which combination of BDNF and a preconditioning lens injury substantially correlated with a 7% increase in the survival of axotomized RGCs.45 enhanced RGC soma survival at 2 weeks post axotomy (71% survival Lentiviral-mediated CNTF gene transfer was also reported to enhance compared with 44% in control eyes), this approach resulted in axonal survival of axotomized RGCs.68 More recently, AAV-mediated expres- dystrophy and regenerative failure.57 Lens injury alone, on the other sion of a secretable form of CNTF was shown to stimulate RGC

Figure 3 Autocrine or paracrine production of neurotrophic factors by transduced cells. Neurotrophic factors can be secreted by cells targeted with viral vectors, such as the RPE (green) or Mu¨ller glia (yellow), to promote RGC survival in a paracrine manner. Alternatively, RGCs can produce and secrete neurotrophic factors that promote their own survival in an autocrine manner (red/purple). GCL, ganglion cell layer; INL, inner nuclear layer; IPL, inner plexiform layer; ONL, outer nuclear layer; OPL, outer plexiform layer.

Gene Therapy Gene therapy for retinal ganglion cell AM Wilson and A Di Polo 131 survival as well as axon regeneration after optic nerve crush or RGCs compared with BDNF or GDNF alone, respectively.73 GDNF peripheral nerve graft transplantation, an effect that was substantially gene transfer using an adenoviral vector or electroporation conferred enhanced in the presence of a cell-permeable cAMP analog.48,63 Three protection of RGCs after optic nerve transaction.44,74 Collectively, weeks after graft transplantation, there was a sixfold increase in the these studies highlight the potential of gene therapy to upregulate number of surviving RGCs that regenerated an axon when CNTF gene retinal neurotrophic factors that stimulate survival and/or regenera- transfer was combined with increased cAMP levels compared with tion of injured RGCs. control optic nerves.63 Other key neurotrophic factors that have been explored using gene Inhibition of apoptotic pathways and manipulation of therapy approaches regarding their ability to regulate RGC survival mitochondrial function and regeneration are fibroblast growth factor-2 (FGF-2) and glial cell Apoptosis or programmed cell death is a common mechanism of line-derived neurotrophic factor (GDNF). FGF-2 is upregulated after neuronal loss in the injured or degenerating visual system. The injury in cells of the inner nuclear layer and ganglion cell layer of the hallmark structural features of apoptosis are cellular round up, adult retina, as well as in the optic nerve and in the optic tract.69 retraction of pseudopodia, reduction of cellular volume, nuclear AAV-mediated FGF-2 gene transfer resulted in a 10-fold increase in fragmentation, modification of cytoplasmic organelles, plasma the number of injured RGC axons that regenerated past the lesion membrane blebbing and engulfment by resident phagocytes. RGCs site at 2 weeks after optic nerve micro-crush lesion.70 The regenerative have been shown to die by apoptosis in optic nerve acute lesion response to FGF-2 upregulation was supported by the finding that models (axotomy and crush),75 experimental glaucoma76 and human FGF receptor-1 and heparan sulfate, known to be essential for glaucoma.77 Apoptotic RGC death was recently confirmed by in vivo FGF-2 signaling, are expressed by adult rat RGCs. In contrast to real-time visualization in ocular hypertensive rat eyes.78 The apoptotic BDNF, FGF-2 transgene expression stimulated RGC axon regrowth, process can be triggered by various stimuli and involves extrinsic and but led to only modest survival of injured RGCs.70,71 Intraocular intrinsic pathways (Figure 4). Extrinsic signals include an array of injection of GDNF protects axotomized RGCs,72 albeit with less death-receptor ligands: Fas-L, tumor necrosis factor-a (TNF-a)and efficacy than BDNF. Combined treatment of BDNF and GDNF TNF-related apoptosis-inducing ligands, which bind to their respec- resulted in 65% and 110% increase in the survival of axotomized tive receptors (Fas, TNF receptor and TNF-related apoptosis-inducing

Figure 4 Extrinsic and intrinsic apoptotic pathways regulating RGC death. The apoptotic death of RGCs can be triggered by various stimuli and involves extrinsic and intrinsic pathways. Extrinsic signals include the death-receptor ligands Fas-L, TNF-a and TNF-related apoptosis-inducing ligand, and their respective receptors, which induce RGC death. Lack of neurotrophic factors may result in deﬁcits of pro-survival pathways, including extracellular signal- regulated kinase 1/2 and phosphatidylinositol-3 kinase. The intrinsic pathway converges on the Bcl-2 family members: pro-apoptotic (Bax, Bad and Bid) or antiapoptotic (Bcl-2 and Bcl-XL), which regulate the mitochondrial outer membrane permeabilization. These proteins control the release of cytochrome c into the cytosol, which can activate the caspases, executioners of apoptosis. The target molecules that have been used in gene therapy studies of RGC neuroprotection are shaded.

Gene Therapy Gene therapy for retinal ganglion cell AM Wilson and A Di Polo 132

ligand receptor) to induce cell death. Evidence of TNF-a and Fas promote neuronal survival.37 In contrast, the suppression of apoptosis ligand upregulation in glaucoma has been observed.79 Death receptor using caspase inhibitors has been investigated with modest success.93 activation results in the recruitment of intracellular adaptor Fas- An exception to this are studies using siRNA to knockdown caspase-2 associated death domain, which then recruits caspase-8 leading to expression in RGCs. A recent study demonstrated that caspase-2 is caspase-3 activation and subsequent cell death. Although these mole- expressed and activated primarily in RGCs following optic nerve cular pathways represent important targets for RGC neuroprotection, injury.94 Moreover, inhibition of caspase-2 expression using a chemi- they have been largely unexplored using gene therapy tools. cally modified siRNA delivered intravitreally led to robust RGC In contrast, the intrinsic apoptotic pathway has been the target of survival after optic nerve crush or cut.94 This strategy is currently many studies using gene therapy, with the common goal of rescuing in phase 1 clinical trials for non-arteritic anterior ischemic RGCs from apoptotic death. There is substantial evidence that optic neuropathy (http://clinicaltrials.gov/ct2/show/NCT01064505?term¼ neurotrophic factors suppress the intrinsic apoptotic cascade by Quark&rank¼3). Another effective approach has been to target baculo- activating intracellular survival signals. Upon binding to their cognate viral IAP repeat-containing 4 (BIRC4), also known as X-linked inhibitor tyrosine kinase receptors, most neurotrophic factors stimulate two of apoptosis protein, a member of the inhibitors of apoptosis family primary pro-survival signaling pathways: the extracellular signal- and a direct inhibitor of several effectors including caspase-3, -7 regulated kinase 1/2 and the phosphatidylinositol-3 kinase pathways.80 and -9.95 AAV-mediated gene transfer of BIRC4/X-linked inhibitor Some factors, such as CNTF, can also activate the Janus kinase/signal of apoptosis protein successfully promoted RGC protection in chronic transducer and activator of transcription 3.81 AAV-mediated gene transfer and acute ocular hypertension models.96,97 For example, at 12 weeks of constitutively active mitogen activated protein kinases/extracellular of ocular hypertension there was a 2.4-fold increase in the number of signal-regulated kinase kinase 1, the obligate upstream activator of extra- surviving RGC axons in AAV.BIRC4-treated eyes compared with control cellular signal-regulated kinase 1/2, resulted in robust survival of RGCs eyes.96 Of interest, combined administration of adenoviral vectors at 2 weeks after axotomy (27% increase over controls) and at 5 weeks of encoding X-linked inhibitor of apoptosis protein or GDNF had a ocular hypertension (38% increase over controls) in rats.82,83 Consistent synergistic effect on the survival of axotomized RGCs, which was greater with the lack of RGC axon regeneration observed with BDNF, gene than upregulation of each individual pathway.44 Therefore, the use of gene transfer of constitutively active mitogen activated protein kinases/ therapy strategies that target multiple antiapoptotic pathways appears to extracellular signal-regulated kinase kinase 1 did not promote axon be a promising avenue to increase RGC survival after optic nerve damage. regrowth in the damaged optic nerve.82 Gene transfer of signaling An alternative neuroprotective strategy that is being developed molecules downstream of CNTF, such as Janus kinase/signal transducer for the treatment of optic neuropathies, and might be relevant to and activator of transcription 3, has not yet been reported, but would glaucoma, is the regulation of genes encoding integral mitochondrial help clarify the role of these intermediaries in RGC survival and regen- proteins. The gene encoding OPA-1, a protein embedded in the inner eration. In contrast to protein tyrosine kinases that promote neuronal mitochondrial membrane, is mutated in the majority of patients with survival, there are several pro-apoptotic protein kinases that contribute autosomal-dominant optic atrophy, a condition characterized by RGC to RGC death, such as the c-Jun N-terminal kinase.84 c-Jun, activated degeneration and childhood blindness. OPA-1 is a GTPase dynamin- by c-Jun N-terminal kinase phosphorylation, mediates transcription like protein that mediates mitochondrial fusion, thus loss of of pro-apoptotic genes and has been shown to be upregulated in rat OPA-1 function leads to mitochondrial fragmentation, cytochrome and monkey models of glaucoma.85,86 siRNA-mediated gene expression c release, mitochondrial DNA damage and increased reactive oxygen knockdown of c-Jun in RGCs resulted in a threefold increase in RGC species.98 Gene therapy for mitochondrial disorders is challenging survival at 2 weeks after optic nerve lesion.37 These studies support because the mitochondrial inner membrane is a substantial barrier for the notion that gene therapy strategies that modulate the activity of gene delivery. A promising technology involves allotopic expression signaling intermediaries may be beneficial for RGC neuroprotection where the therapeutic gene is introduced into the nuclear genome, in glaucoma. and the gene product is then imported into mitochondria using a The battleground of apoptosis is the mitochondria, the organelle targeting sequence. Overexpression of OPA-1 by AAV-mediated gene where cell fate is decided. The Bcl-2 family members were the first transfer increased RGC survival, while decreasing the number molecular targets linked to mitochondrial function to be studied in of reactive astroglia and microglia in DBA/2J mice.99 Leber’s heredi- the context of RGC death.87 Transgenic overexpression of the anti- tary optic neuropathy, a rare condition characterized by RGC death, is apoptotic Bcl-2 protein resulted in sustained RGC survival after also caused by mutations in mitochondrial DNA. Most cases are due axotomy.88 Furthermore, AAV-mediated gene transfer of Bcl-XL or to point mutations in genes encoding several subunits of the BAG1, a Bcl-2-associated protein, promoted RGC survival in optic nicotinamide adenine dinucleotide-ubiquinone oxidoreductase, nerve cut or crush models.89,90 At 8 weeks after axotomy, 46% of which is essential for complex I activity in the electron transport Bcl-XL overexpressing RGCs survived compared with only 6% in chain. RGCs with mutated mitochondrial DNA undergo apoptosis control groups.89 Deletion of the pro-apoptotic Bax gene was shown through calcium-dependent and caspase-independent pathways, a to be neuroprotective for RGC soma, but failed to prevent axon process also associated with the release of toxic mitochondrial factors degeneration in DBA/2J mice, a mouse strain that spontaneously to the cytosol (for example, cytochrome c, apoptosis-inducing factor develops chronic age-related glaucoma.91 Intriguingly, BAX knock- and endonuclease G). Allotopic delivery by electroporation of the down through injection of siRNA into the optic nerve did not result in wild-type human ND4 subunit gene, which carries prevalent muta- significant protection of axotomized RGCs.92 In the final steps of the tions in Leber’s hereditary optic neuropathy patients, promoted RGC intrinsic apoptotic pathway, BAX molecules translocate to the mito- neuroprotection in a rat model of Leber’s hereditary optic neuro- chondrial outer membrane and form pore structures that facilitate pathy.100 Specifically, B38% decrease in RGC counts was observed cytochrome c release into the cytoplasm. Cytochrome c binds to in animals that expressed the mutant ND4 compared with wild- Apaf-1 to form the apoptosome, a complex that recruits and activates type ND4 at 48 days after electroporation.100 More recently, pro-caspase-9, which in turn activates downstream effector caspases. AAV-mediated allotopic delivery of ND4 was shown to be safe in a Apaf-1 gene knockdown in RGCs using siRNA has been shown to phase I clinical trial.101

Gene Therapy Gene therapy for retinal ganglion cell AM Wilson and A Di Polo 133

Toxic enablers: channels and receptors involved in RGC damage of RGCs in glaucoma undergo axonal damage and/or retraction Excitotoxicity, ionic imbalance and oxidative stress have been proof synaptic terminals that interrupt neurotransmission. For those posed to contribute to RGC loss in glaucoma. These pathways of RGC neurons, in addition to neuroprotection, regenerative strategies that injury, activated by numerous receptors and channels, can activate promote axonal regrowth and enhance synaptic connectivity might be downstream mediators of cell death through multiple points of required. Similar to other central nervous system neurons in adult convergence. The central dogma of excitotoxicity is that excess mammals, RGCs have a limited capacity to regenerate an axon within glutamate binds to cell surface glutamate receptors on neurons, the optic nerve. If these neurons are provided with a permissive 2+ primarily N-methyl-D-aspartate receptors, triggering massive Ca environment, such as a peripheral nerve graft, they have the ability to influx and apoptosis. However, N-methyl-D-aspartate receptor antago- regrow axons for long distances and form new synaptic connec- nists have consistently failed in clinical trials, including a recent trial tions.108 The limited ability of RGCs to regenerate an axon is partly testing the efficacy of memantine in glaucoma,102 which suggests that due to the presence of myelin-associated growth inhibitory molecules. other mechanisms contribute to excitotoxic damage in vivo.Alter- Three inhibitors have been identified in myelin: Nogo-A, myelin- native models include a role for neighboring cells, primarily glia, in associated glycoprotein and oligodendrocyte-myelin glycoprotein.109 excitotoxic RGC demise. For example, a recent study demonstrated The Nogo receptor (NgR) serves as a high-affinity receptor for Nogo- that Mu¨ller cells are extremely sensitive to an acute excitotoxic insult A, myelin-associated glycoprotein and oligodendrocyte-myelin glyco- and respond by upregulating TNF-a, which then mediates RGC protein, and has two partners responsible for signal transduction: loss.103 In the case of chronic, neurodegenerative diseases where an p75NTR and the NgR-interacting protein (LINGO-1).109 Adult RGCs excitotoxic component is thought to be involved, such as in glaucoma, do not express p75NTR,110 and a newly identified co-receptor, named blockade of glutamate receptors may not ameliorate disease progres- TROY, has been shown to interact with NgR and LINGO-1 to sion unless other damaging molecules (for example, glia-derived promote growth inhibition in these neurons.111 The strategy to TNF-a) are also inhibited. The molecular pathways involved in RGC block NgR has received attention because it offers a single target to excitotoxicity have been targeted using drugs and pharmacological neutralize signaling via Nogo, myelin-associated glycoprotein or agents, but, so far, gene therapy has not been explored. Given the oligodendrocyte-myelin glycoprotein. Expression of a dominant-nega- complexity of glaucoma, it is likely that more than one signaling tive form of NgR in RGCs using recombinant AAV led to appreciable pathway will need to be targeted to achieve neuroprotection. Gene axonal regeneration, but only when combined with a preconditioning therapy might provide the versatility required to selectively target lens injury that elicited ocular inflammation and axon regrowth.112 different molecular pathways, possibly in different cell types, to Therefore, overcoming myelin-derived inhibitory signals is not suffi- optimize RGC viability. cient to promote RGC axon regeneration unless this strategy is Other pathways converging onto the apoptotic machinery are combined with treatments that stimulate axonal growth. mediated by ionic imbalance, leading to cell shrinkage and death. Damage to the optic nerve triggers a series of events that lead to the Indeed, cell shrinkage precedes mitochondrial dysfunction and apop- formation of a glial scar. Although the glial scar is part of a self-repair tosome formation, and has been shown to occur in RGCs subjected to process that protects the fragile neural tissue and helps repair the high pressure.104 Importantly, cell shrinkage and apoptosis are accom- blood–brain barrier, it is also a major physical and chemical obstacle panied by increased K+ currents and depletion of intracellular K+; that prevents axonal regeneration. The glial scar in central nervous therefore, K+ channels are being investigated in the context of RGC system tissue contains a large number of reactive astrocytes that death. A recent study demonstrated that siRNA-mediated knockdown secrete potent inhibitors of axon growth such as chondroitin sulfate of K+ channels of the Shaker (Kv1) family effectively supports the proteoglycans. In glaucoma, astrocytes in the optic nerve head survival of axotomized RGCs.36 This approach is substantially more undergo cell hypertrophy, leading to major alterations in the bio- specific than common, nonselective channel blockers. chemical composition of the extracellular matrix and biomechanical Oxidative stress, caused by the imbalance between the production properties of the tissue surrounding RGC axons,113 which might of reactive oxygen species and their breakdown by antioxidants, has contribute to regenerative failure. Gene therapy represents a potential been recognized as another central contributor to RGC injury and tool to modulate gene expression in astrocytes to selectively regulate death. Reactive oxygen species superoxide increases in RGCs following expression of inhibitory factors and enhance RGC axon regeneration, optic nerve axotomy, and RGC death is delayed in vivo when while preserving the astrocytic functions that facilitate the nerve repair intracellular superoxide levels are decreased.105 Administration of processes. pegylated superoxide dismutase-1, which catalyzes the dismutation The signaling molecules that contribute to RGC axon re- of superoxide into oxygen and hydrogen peroxide, functions as an generation are being elucidated and are promising avenues to antioxidant and attenuates RGC death.105 Therefore, blockade of tackle the complex problem of optic nerve repair through gene superoxide generation might be beneficial for injured RGCs in therapy strategies. Several growth inhibitory signals converge on the glaucoma. Another strategy involves heme oxygenase, which alleviates Rho GTPase pathway to halt axonal growth. Blockade of RhoA activity tissue injury not only by acting as an antioxidant but also by using cell-permeable C3 ribosyltransferase, directly applied to the modulating the inflammatory response and activating innate anti- transected optic nerve or delivered to RGCs by an AAV vector, resulted apoptotic pathways. Consistent with this, Ad-mediated heme oxyge- in twofold increase in axonal regrowth at 2 weeks after crush nase gene transfer protected RGCs against ischemia/reperfusion injury.114,115 Gene knockdown of the mammalian sterile 20-like injury106 and pressure-induced ischemia.107 Thus, the modulation of kinase-3b by short hairpin RNA delivered to RGCs using AAV antioxidants via gene therapy might be a promising approach to prevented axon growth, demonstrating a role for this kinase in optic protect RGCs from glaucomatous damage. nerve regeneration.116 AAV-mediated deletion of phosphatase and tensin homolog, a negative regulator of the mammalian target of Overcoming axon growth inhibition rapamycin, promoted robust RGC axon regeneration,117 and this Neuroprotection in glaucoma aims at preserving the structure and effect was markedly enhanced by cAMP and the pro-inflammatory function of remaining RGCs. It is likely, however, that a number agent zymosan.118

Gene Therapy Gene therapy for retinal ganglion cell AM Wilson and A Di Polo 134

SUMMARY AND FUTURE PERSPECTIVES CONFLICT OF INTEREST The medical need for neuroprotective therapies for glaucoma is The authors declare no conflict of interest. undeniable. In the last 10 years, there has been tremendous progress in the use of gene therapy strategies not only to understand the ACKNOWLEDGEMENTS molecular basis of RGC death, but also to stimulate the survival and We thank Dr Timothy Kennedy (McGill University) for helpful comments on regeneration of these neurons in a variety of preclinical models of the manuscript, Dr Elena Feinstein (Quark Pharmaceuticals Inc., Ness Ziona, optic nerve injury. Among available recombinant viral vectors, AAV-2 Israel) for providing the reporter siRNA and James Correia for assistance with has received particular attention because of its ability to mediate the figures. This work was supported by the Natural Sciences and Engineering infection of a large number of adult RGCs in vivo with relatively high Research Council of Canada and the Canadian Institutes of Health Research. ADP is a Fonds de recherche en sante´ du Que´bec Chercheur Senior. specificity. Novel nonviral approaches to modulate RGC gene expression, including electroporation and siRNA delivery, are also being investigated with promising results. Several key molecular pathways that dictate the fate of RGCs after optic nerve injury have been 1 Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and explored. The modulation of neurotrophic factor expression and 2020. Br J Ophthalmol 2006; 90: 262–267. 2 Wiggs JL, Allingham RR, Vollrath D, Jones KH, De La Paz M, Kern J et al. Prevalence stimulation of pro-survival signaling pathways have been extensively of mutations in TIGR/myocilin in patients with adult and juvenile primary open-angle studied, and several candidate molecules have emerged as promising glaucoma. Am J Hum Genet 1998; 63: 1549–1552. 3 Shimizu S, Lichter PR, Johnson AT, Zhou Z, Higashi M, Gottfredsdottir M et al. Age- therapeutic targets. Strategies to block the intrinsic apoptotic pathway, dependent prevalence of mutations at the GLC1A locus in primary open-angle toxic mediators and axon growth inhibitors have also been explored glaucoma. Am J Ophthalmol 2000; 130: 165–177. with success, and allotopic gene transfer to treat mitochondrial 4 Campos Mollo E, Lpez-Garrido M-P, Blanco Marchite C, Garcia-Feijoo J, Peralta J, Belmonte-Martnez J et al. CYP1B1 mutations in Spanish patients with primary congenital dysfunction is a promising avenue for several optic neuropathies glaucoma: phenotypic and functional variability. Mol Vis 2009; 15: 417–431. including glaucoma. 5 Fan BJ, Wiggs JL. Glaucoma: genes, phenotypes, and new directions for therapy. JClin Gene therapy has moved rapidly from promising data obtained in Invest 2010; 120: 3064–3072. 6 Liu X, Rasmussen CA, Gabelt BT, Brandt CR, Kaufman PL. Gene therapy targeting the laboratory to ongoing clinical trials for blinding diseases such as glaucoma: where are we? Surv Ophthalmol 2009; 54:472–486. Leber’s congenital amaurosis. In the case of neuroprotective gene 7 Cheng L, Sapieha P, Kittlerova´ P, Hauswirth WW, Di Polo A. TrkB gene transfer protects retinal ganglion cells from axotomy-induced death in vivo. JNeurosci2002; 22: therapy for adult-onset glaucoma, substantial hurdles still need to be 3977–3986. overcome to translate successful preclinical findings to clinical appli- 8 Harvey AR, Kamphuis W, Eggers R, Symons NA, Blits B, Niclou S et al. Intravitreal cations. For example, what molecular pathway should be targeted for injection of adeno-associated viral vectors results in the transduction of different types of retinal neurons in neonatal and adult rats: a comparison with lentiviral vectors. Mol RGC neuroprotection? Unlike many inherited retinal degenerations in Cell Neurosci 2002; 21: 141–157. which supplementation of a healthy copy of the affected gene is a 9 Ali RR, Reichel MB, De Alwis M, Kanuga N, Kinnon C, Levinsky RJ et al. Adeno- sensible approach, the heterogeneity of factors contributing to RGC associated virus gene transfer to mouse retina. Hum Gene Ther 1998; 9:81–86. 10 Ali RR, Reichel MB, Thrasher AJ, Levinsky RJ, Kinnon C, Kanuga N et al. Gene death in glaucoma makes it difficult to settle on any one pathway. transfer into the mouse retina by an adeno-associated virus vector. Hum Mol Genet There might be patients who experience compromised retrograde 1996; 5: 591–594. 11 Bennett J, Maguire AM, Cideciyan AV, Schnell M, Glover E, Anand V et al. Stable axonal transport along the optic nerve and target-derived neurotro- transgene expression in rod photoreceptors after recombinant adeno-associated virus- phin deprivation, whereas others might suffer from increased oxida- mediated gene transfer to monkey retina. Proc Natl Acad Sci USA 1999; 96: tive stress and reduced antioxidant defense mechanisms. Moreover, 9920–9925. 12 Hellstrom M, Ruitenberg MJ, Pollett MA, Ehlert EME, Twisk J, Verhaagen J et al. some individuals may suffer from alterations in multiple pathways Cellular tropism and transduction properties of seven adeno-associated viral vector that converge to trigger or exacerbate RGC death. The availability of serotypes in adult retina after intravitreal injection. Gene Therapy 2009; 16:521–532. novel biomarkers for glaucoma may serve to identify molecular 13 Martin KR, Klein RL, Quigley HA. Gene delivery to the eye using adeno-associated viral vectors. Methods 2002; 28: 267–275. deficits in individual patients and, eventually, to personalize gene 14 Stieger K, Belbellaa B, Le Guiner C, Moullier P, Rolling F. In vivo gene regulation using therapy interventions. Unlike pharmacological treatments, gene tetracycline-regulatable systems. Adv Drug Deliv Rev 2009; 61: 527–541. 15 Stieger K, Schroeder J, Provost N, Mendes-Madeira A, Belbellaa B, Meur GL et al. therapy will allow the simultaneous manipulation of multiple path- Detection of intact rAAV particles up to 6 years after successful gene transfer in the ways with relatively high specificity. retina of dogs and primates. Mol Ther 2008; 17:516–523. Another important issue relates to the effect of long-term mod- 16 Lai Y, Yue Y, Liu M, Ghosh A, Engelhardt JF, Chamberlain JS et al. Efficient in vivo gene expression by trans-splicing adeno-associated viral vectors. Nat Biotech 2005; ification of gene expression, such as that provided by AAV, in 23: 1435–1439. glaucoma. For example, neurotrophic factor supplementation by 17 Reichel MB, Ali RR, Thrasher AJ, Hunt DM, Bhattacharya SS, Baker D. Immune gene therapy is a promising strategy to promote RGC survival in responses limit adenovirally mediated gene expression in the adult mouse eye. Gene Therapy 1998; 5: 1038–1046. several models of optic nerve damage, but the long-term effect and 18 Kumar-Singh R. Barriers for retinal gene therapy: separating fact from fiction. Vis Res safety of increased diffusible neurotrophin levels in the retina are 2008; 48: 1671–1680. 19 Oshima Y, Oshima S, Nambu H, Kachi S, Hackett S, Melia M et al. Increased currently unknown. Sustained AAV-mediated expression of CNTF or expression of VEGF in retinal pigmented epithelial cells is not sufficient to cause BDNF leads to changes in the dendritic structure of transduced choroidal neovascularization. JCellPhysiol2004; 201:393–400. RGCs.56 It will be important to establish whether these morphological 20 Oshima Y, Takahashi K, Oshima S, Saishin Y, Saishin Y, Silva RL et al. Intraocular gutless adenoviral-vectored VEGF stimulates anterior segment but not retinal neovas- changes have a tangible effect on RGC electrophysiological function. cularization. JCellPhysiol2004; 199: 399–411. The use of viral vectors containing genes driven by regulatable 21 Takahashi K, Saishin Y, Saishin Y, Silva R, Oshima Y, Oshima S et al. Intraocular promoters, or the use of DNA or siRNA technologies that afford expression of endostatin reduces VEGF-induced retinal vascular permeability, neovascularization, and retinal detachment. FASEB J 2003; 17:896–898. transient gene expression may resolve some of these issues. In 22 Lamartina S, Cimino M, Roscilli G, Dammassa E, Lazzaro D, Rota R et al. Helper- conclusion, a number of studies have demonstrated proof of principle dependent adenovirus for the gene therapy of proliferative retinopathies: stable gene transfer, regulated gene expression and therapeutic efficacy. JGeneMed2007; 9: for the use of gene therapy for neuroprotection of injured RGCs, 862–874. and several promising molecular pathways have been elegantly 23 Di Polo A, Aigner LJ, Dunn RJ, Bray GM, Aguayo AJ. Prolonged delivery of brain- delineated. The challenge now is to assess how this wealth of derived neurotrophic factor by adenovirus-infected Mu¨ller cells temporarily rescues injured retinal ganglion cells. Proc Natl Acad Sci USA 1998; 95: 3978–3983. knowledge can be translated into potential gene therapies for neuro- 24 Isenmann S, Klo¨cker N, Gravel C, Ba¨hr M. Protection of axotomized retinal ganglion protection in glaucoma. cells by adenovirally delivered BDNF in vivo. Eur J Neurosci 1998; 10: 2751–2756.

Gene Therapy Gene therapy for retinal ganglion cell AM Wilson and A Di Polo 135

25 Bennett J, Wilson J, Sun D, Forbes B, Maguire A. Adenovirus vector-mediated 52 Hellstro¨m M, Harvey AR. Retinal ganglion cell gene therapy and visual system repair. in vivo gene transfer into adult murine retina. Invest Ophthalmol Vis Sci 1994; 35: Curr Gene Ther 2011; 11:116–131. 2535–2542. 53 Ko ML, Hu DN, Ritch R, Sharma SC, Chen CF. Patterns of retinal ganglion cell survival 26 Li T, Adamian M, Roof DJ, Berson EL, Dryja TP, Roessler BJ et al. In vivo transfer of a after brain-derived neurotrophic factor administration in hypertensive eyes of rats. reporter gene to the retina mediated by an adenoviral vector. Invest Ophthalmol Vis Sci Neurosci Lett 2001; 305:139–142. 1994; 35: 2543–2549. 54 Martin KR, Quigley HA, Zack DJ, Levkovitch-Verbin H, Kielczewski J, Valenta D et al. 27 Cashman S, McCullough L, Kumar Singh R. Improved retinal transduction in vivo and Gene therapy with brain-derived neurotrophic factor as a protection: retinal ganglion photoreceptor-specific transgene expression using adenovirus vectors with modified cells in a rat glaucoma model. Invest Ophthalmol Vis Sci 2003; 44: 4357–4365. penton base. Mol Ther 2007; 15: 1640–1646. 55 Fournier AE, McKerracher L. Expression of specific tubulin isotypes increases during 28 Kugler S, Straten G, Kreppel F, Isenmann S, Liston P, Bahr M. The X-linked inhibitor of regeneration of injured CNS neurons, but not after the application of brain-derived apoptosis (XIAP) prevents cell death in axotomized CNS neurons in vivo. Cell Death neurotrophic factor (BDNF). JNeurosci1997; 17: 4623–4632. Differ 2000; 7: 815–824. 56 Harvey AR, Hellstro¨m M, Rodger J. Gene therapy and transplantation in the retinofugal 29 Balaggan KS, Ali RR. Ocular gene delivery using lentiviral vectors. Gene Therapy pathway. In: Verhaagen J, Hol E, Huitinga I, Wijnholds J, Bergen AA, Boer GJ, Swabb DF 2011; (in press). (eds). Progress in Brain Research, vol. 175. Elsevier: Amsterdam, 2009, pp 151–161. 30 Cheng L, Chaidhawangul S, Wong-Staal F, Gilbert J, Poeschla E, Toyoguchi M et al. 57 Pernet V, Di Polo A. Synergistic action of brain-derived neurotrophic factor and lens Human immunodeficiency virus type 2 (HIV-2) vector-mediated in vivo gene transfer injury promotes retinal ganglion cell survival, but leads to optic nerve dystrophy in into adult rabbit retina. Curr Eye Res 2002; 24:196–201. vivo. Brain 2006; 129: 1014–1026. 31 Isenmann S, Gillardon F, Ba¨hr M. Bax antisense oligonucleotides reduce axotomy- 58 Fischer D, Heiduschka P, Thanos S. Lens-injury-stimulated axonal regeneration induced retinal ganglion cell death in vivo by reduction of Bax protein expression. Cell throughout the optic pathway of adult rats. Exp Neurol 2001; 172:257–272. Death Differ 1999; 6: 673–682. 59 Leon SY, Nguyen J, Irwin N, Benowitz LI. Lens injury stimulates axon regeneration in 32 Thaler S, Dietrich K, Ladewig T, Okuno E, Kocki T, Turski WA et al. Aselectivemethod the mature rat optic nerve. J Neurosci 2000; 20: 4615–4626. for transfection of retinal ganglion cells by retrograde transfer of antisense oligonu- 60 YinY,HenzlMT,LorberB,NakazawaT,ThomasTT,JiangFet al. Oncomodulin is a cleotides against kynurenine aminotransferase II. Mol Vis 2006; 12: 100–107. macrophage-derived signal for axon regeneration in retinal ganglion cells. Nat 33 Caprioli J, Munemasa Y, Kwong JM, Piri N. Overexpression of thioredoxins 1 and 2 Neurosci 2006; 9: 843–852. increases retinal ganglion cell survival after pharmacologically induced oxidative 61 Clarke DB, Bray GM, Aguayo AJ. Prolonged administration of NT-4/5 fails to rescue stress, optic nerve transection, and in experimental glaucoma. Trans Am Ophthalmol most axotomized retinal ganglion cells in adult rats. Vis Res 1998; 38: 1517–1524. Soc 2009; 107: 161–165. 62 Meyer-Franke A, Wilkinson GA, Kruttgen A, Hu M, Munro E, Hanson Jr MG et al. 34 Ishikawa H, Takano M, Matsumoto N, Sawada H, Ide C, Mimura O et al. Effect of Depolarization and cAMP elevation rapidly recruit TrkB to the plasma membrane of GDNF gene transfer into axotomized retinal ganglion cells using in vivo electroporation CNS neurons. Neuron 1998; 21: 681–693. with a contact lens-type electrode. Gene Therapy 2004; 12: 289–298. 63 Cui Q, Yip HK, Zhao RC, So KF, Harvey AR. Intraocular elevation of cyclic AMP 35 Mo X, Oshitari T, Negishi H, Dezawa M, Mizota A, Adachi-Usami E. Rescue of potentiates ciliary neurotrophic factor-induced regeneration of adult rat retinal gang- axotomized retinal ganglion cells by BDNF gene electroporation in adult rats. Invest lion cell axons. Mol Cell Neurosci 2003; 22:49–61. Ophthalmol Vis Sci 2002; 43: 2401–2405. 64 Muller A, Hauk TG, Fischer D. Astrocyte-derived CNTF switches mature RGCs to a 36 Koeberle PD, Wang Y, Schlichter LC. Kv1.1 and Kv1.3 channels contribute to the regenerative state following inflammatory stimulation. Brain 2007; 130: 3308–3320. degeneration of retinal ganglion cells after optic nerve transection in vivo. Cell Death 65 Ji JZ, Elyaman W, Yip HK, Lee VW, Yick LW, Hugon J et al. CNTF promotes survival of Differ 2009; 17: 134–144. retinal ganglion cells after induction of ocular hypertension in rats: the possible 37 Lingor P, Koeberle P, Kugler S, Bahr M. Down-regulation of apoptosis mediators by involvement of STAT3 pathway. Eur J Neurosci 2004; 19:265–272. RNAi inhibits axotomy-induced retinal ganglion cell death in vivo. Brain 2005; 128: 66 Pease ME, Berlinicke C, Bloom K, Cone F, Wang Y, Klein RL et al. Effect of CNTF on 550–558. retinal ganglion cell survival in experimental glaucoma. Invest Ophthalmol Vis Sci 38 Thoenen H, Bandtlow C, Heumann R. The physiological function of nerve growth 2009; 50: 2194–2200. factor in the central nervous system: comparison with the periphery. Rev Physiol 67 Maier K, Rau CR, Storch MK, Sattler MB, Demmer I, Weissert R et al. Ciliary Biochem Pharmacol 1987; 109:145–178. neurotrophic factor protects retinal ganglion cells from secondary cell death during 39 Crish SD, Sappington RM, Inman DM, Horner PJ, Calkins DJ. Distal axonopathy with acute autoimmune optic neuritis in rats. Brain Pathol 2004; 14: 378–387. structural persistence in glaucomatous neurodegeneration. Proc Nat Acad Sci USA 68 van Adel BA, Kostic C, De´glon N, Ball AK, Arsenijevic Y. Delivery of ciliary neuro- 2010; 107: 5196–5201. trophic factor via lentiviral-mediated transfer protects axotomized retinal ganglion 40 Pease ME, McKinnon SJ, Quigley HA, Kerrigan-Baumrind LA, Zack DJ. Obstructed cells for an extended period of time. Hum Gene Ther 2003; 14: 103–115. axonal transport of BDNF and its receptor TrkB in experimental glaucoma. Invest 69 Wen R, Song Y, Cheng T, Matthes MT, Yasumura D, LaVail MM et al. Injury-induced upregu- Ophthalmol Vis Sci 2000; 41:764–774. lation of bFGF and CNTF mRNAS in the rat retina. JNeurosci1995; 15: 7377–7385. 41 Carpenter P, Sefton AJ, Dreher B, Lim W-L. Role of target tissue in regulating the 70 Sapieha M, Peltier M, Rendahl KG, Manning WC, Di Polo A. Fibroblast growth factor-2 development of retinal ganglion cells in the albino rat: effects of kainate lesions in the gene delivery stimulates axon growth by adult retinal ganglion cells after acute optic superior colliculus. JCompNeurol1986; 251: 240–259. nerve injury. Mol Cell Neurosci 2003; 24: 656–672. 42 Pearson HE, Stoffler DJ. Retinal ganglion cell degeneration following loss of post- 71 Sapieha M, Hauswirth WW, Di Polo A. Extracellular signal-regulated kinases 1/2 are synaptic target neurons in the dorsal lateral geniculate nucleus of the adult cat. Exp required for adult retinal ganglion cell axon regeneration induced by fibroblast growth Neurol 1992; 116: 163–171. factor-2. J Neurosci Res 2006; 83: 985–995. 43 Pearson HE, Thompson TP. Atrophy and degeneration of ganglion cells in central 72 Klocker N, Braunling F, Isenmann S, Bahr M. In vivo neurotrophic effects of GDNF on retina following loss of postsynaptic target neurons in the dorsal lateral geniculate axotomized retinal ganglion cells. Neuroreport 1997; 10: 3439–3442. nucleus of the adult cat. Exp Neurol 1993; 119: 113–119. 73 Koeberle PD, Ball AK. Neurturin enhances the survival of axotomized retinal ganglion 44 Straten G, Schmeer C, Kretz A, Gerhardt E, Kugler S, Schulz JB et al. Potential cells in vivo: combined effects with glial cell line-derived neurotrophic factor and synergistic protection of retinal ganglion cells from axotomy-induced apoptosis by brain-derived neurotrophic factor. Neurosci 2002; 110: 555–567. adenoviral administration of glial cell line-derived neurotrophic factor and X-chromo- 74 Ishikawa H, Takano M, Matsumoto N, Sawada H, Ide C, Mimura O et al. Effect of some-linked inhibitor of apoptosis. Neurobiol Dis 2002; 11: 123–133. GDNF gene transfer into axotomized retinal ganglion cells using in vivo electroporation 45 van Adel BA, Arnold JM, Phipps J, Doering LC, Ball AK. Ciliary neurotrophic factor with a contact lens-type electrode. Gene Therapy 2005; 12: 289–298. protects retinal ganglion cells from axotomy-induced apoptosis via modulation of 75 Berkelaar M, Clarke DB, Wang Y-C, Bray GM, Aguayo AJ. Axotomy results in delayed death retinal glia in vivo. J Neurobiol 2005; 63: 215–234. and apoptosis of retinal ganglion cells in adult rats. JNeurosci1994; 14: 4368–4374. 46 Chen H, Weber AJ. BDNF enhances retinal ganglion cell survival in cats with optic 76 Quigley HA, Nickells RW, Kerrigan LA, Pease ME, Thibault DJ, Zack DJ. Retinal nerve damage. Invest Ophthalmol Vis Sci 2001; 42: 966–974. ganglion cell death in experimental glaucoma and after axotomy occurs by apoptosis. 47 Klo¨cker N, Kermer P, Weishaupt JH, Labes M, Ankerhold R, Ba¨hr M. Brain-derived Invest Ophthalmol Vis Sci 1995; 36:774–786. neurotrophic factor-mediated neuroprotection of adult rat retinal ganglion cells in vivo 77 Kerrigan LA, Quigley HA, Smith SD, Pease ME. TUNEL-positive ganglion cells in does not exclusively depend on phosphatidyl-inositol-3¢-kinase/protein kinase B human primary open-angle glaucoma. Arch Ophthalmol 1997; 115: 1031–1035. signaling. J Neurosci 2000; 20: 6962–6967. 78 Cordeiro MF, Guo L, Luong V, Harding G, Wang W, Jones HE et al. Real-time imaging of 48 Leaver SG, Cui Q, Plant GW, Arulpragasam A, Hisheh S, Verhaagen J et al. AAV- single nerve cell apoptosis in retinal neurodegeneration. P Natl Acad Sci USA 2004; mediated expression of CNTF promotes long-term survival and regeneration of adult 101: 13352–13356. rat retinal ganglion cells. Gene Therapy 2006; 13: 1328–1341. 79 Tezel G, Carlo Nucci LC, Giacinto B. TNF-a signaling in glaucomatous neuro- 49 Mansour-Robaey S, Clarke DB, Wang Y-C, Bray GM, Aguayo AJ. Effects of ocular injury degeneration. In: Nucci C, Cerulli L, Osborne NN, Bagetta G (eds). Progress in and administration of brain-derived neurotrophic factor on survival and regrowth of Brain Research, vol. 173. Elsevier: Amsterdam, 2008, pp 409–421. axotomized retinal ganglion cells. Proc Natl Acad Sci USA 1994; 91: 1632–1636. 80 Kaplan D, Miller FD. Neurotrophin signal transduction in the nervous system. 50 Mey J, Thanos S. Intravitreal injections of neurotrophic factors support the survival Curr Opin Neurobiol 2000; 10:381–391. of axotomized retinal ganglion cells in adult rats in vivo. Brain Res 1993; 602: 81 Peterson WM, Wang Q, Tzekova R, Wiegand SJ. Ciliary neurotrophic factor and stress 304–317. stimuli activate the Jak-STAT pathway in retinal neurons and glia. J Neurosci 2000; 51 Peinado-Ramon P, Salvador M, Villegas-Perez MP, Vidal-Sanz M. Effects of axotomy 20: 4081–4090. and intraocular administration of NT-4, NT-3, and brain-derived neurotrophic factor on 82 Pernet V, Hauswirth WW, Di Polo A. Extracellular signal-regulated kinase 1/2 mediates the survival of adult rat retinal ganglion cells. A quantitative in vivo study. Invest survival, but not axon regeneration, of adult injured central nervous system neurons in Ophthalmol Vis Sci 1996; 37:489–500. vivo. JNeurochem2005; 93:72–83.

Gene Therapy Gene therapy for retinal ganglion cell AM Wilson and A Di Polo 136

83 Zhou Y, Pernet V, Hauswirth WW, Di Polo A. Activation of the extracellular signal- 101 Lam BL, Feuer WJ, Abukhalil F, Porciatti V, Hauswirth WW, Guy J. Leber hereditary regulated kinase 1/2 pathway by AAV gene transfer protects retinal ganglion cells in optic neuropathy gene therapy clinical trial recruitment: year 1. Arch Ophthalmol glaucoma. Mol Ther 2005; 12:402–412. 2010; 128: 1129–1135. 84 Bessero AC, Chiodini F, Rungger-Bra¨ndle E, Bonny C, Clarke PG. Role of the c-Jun 102 Osborne NN. Recent clinical ﬁndings with memantine should not mean that the N-terminal kinase pathway in retinal excitotoxicity, and neuroprotection by its idea of neuroprotection in glaucoma is abandoned. Acta Ophthalmol 2008; 87: inhibition. JNeurochem2010; 113: 1307–1318. 450–454. 85 Hashimoto K, Parker A, Malone P, Gabelt BT, Rasmussen C, Kaufman PS et al. Long- 103 Lebrun-Julien F, Duplan L, Pernet V, Osswald IK, Sapieha P, Bourgeois P et al. term activation of c-Fos and c-Jun in optic nerve head astrocytes in experimental Excitotoxic death of retinal neurons in vivo occurs via a non-cell-autonomous ocular hypertension in monkeys and after exposure to elevated pressure in vitro. Brain mechanism. JNeurosci2009; 29: 5536–5545. Res 2005; 1054:103–115. 104 Lee JK, Lu S, Madhukar A. Real-time dynamics of calcium, caspase-3/7, and 86 Levkovitch-Verbin H, Quigley HA, Martin KR, Harizman N, Valenta DF, Pease ME et al. morphological changes in retinal ganglion cell apoptosis under elevated pressure. The transcription factor c-jun is activated in retinal ganglion cells in experimental rat PLoS ONE 2010; 5: e13437. glaucoma. Exp Eye Res 2005; 80:663–670. 105 Kanamori A, Catrinescu MM, Kanamori N, Mears KA, Beaubien R, Levin LA. 87 Nickells RW, Semaan SJ, Schlamp CL, Carlo Nucci LCNNO, Giacinto B. Involvement Superoxide is an associated signal for apoptosis in axonal injury. Brain 2010; 133: of the Bcl2 gene family in the signaling and control of retinal ganglion cell death. 2612–2625. In: Nucci C, Cerulli L, Osborne NN, Bagetta G (eds). Progress in Brain Research, 106 Peng PH, Ko ML, Chen CF, Juan SH. Haem oxygenase-1 gene transfer protects retinal vol. 173. Elsevier: Amsterdam, 2008, pp 423–435. ganglion cells from ischaemia/reperfusion injury. Clin Sci 2008; 115: 335–342. 88 Bonfanti L, Strettoi E, Chierzi S, Cenni MC, Liu X-H, Martinou J-C et al. Protection of 107 Hegazy KA, Sharma SC. Functional human heme oxygenase has a neuroprotective retinal ganglion cells from natural and axotomy-induced cell death in neonatal effect on adult rat ganglion cells after pressure-induced ischemia. Neuroreport 2000; transgenic mice overexpressing bcl-2. JNeurosci1996; 16: 4186–4194. 11: 1185–1189. 89 Malik JMI, Shevtsova Z, Bahr M, Kugler S. Long-term in vivo inhibition of CNS 108 Aguayo AJ, Bray GM, Carter DA, Villegas-Perez MP, Vidal-Sanz M, Rasminsky M. neurodegeneration by Bcl-XL gene transfer. Mol Ther 2005; 11: 373–381. Regrowth and connectivity of injured central nervous system axons in adult rodents. 90 Planchamp V, Bermel C, To¨nges L, Ostendorf T, Ku¨gler S, Reed JC et al. BAG1 Acta Neurobiol Exp 1990; 50:381–389. promotes axonal outgrowth and regeneration in vivo via Raf-1 and reduction of ROCK 109 Sandvig A, Berry M, Barrett LB, Butt A, Logan A. Myelin-, reactive glia-, and scar- activity. Brain 2008; 131: 2606–2619. derived CNS axon growth inhibitors: expression, receptor signaling, and correlation 91 Libby RT, Li Y, Savinova OV, Barter J, Smith RS, Nickells RW et al. Susceptibility with axon regeneration. Glia 2004; 46: 225–251. to neurodegeneration in a glaucoma is modiﬁed by Bax gene dosage. PLoS Genet 110 Lebrun-Julien F, Morquette B, Douillette A, Saragovi HU, Di Polo A. Inhibition of 2005; 1:e4. p75NTR in glia potentiates TrkA-mediated survival of injured retinal ganglion cells. 92 Lingor P, Koeberle P, Kugler S, Bahr M. Down-regulation of apoptosis mediators by Mol Cell Neurosci 2009; 40: 410–420. RNAi inhibits axotomy-induced retinal ganglion cell death in vivo. Brain 2005; 128: 111 Shao Z, Browning JL, Lee X, Scott ML, Shulga-Morskaya S, Allaire N et al. TAJ/TROY, 550–558. an orphan TNF receptor family member, binds Nogo-66 receptor 1 and regulates 93 Bilsland J. Caspases and neuroprotection. Curr Opin Investig Drugs 2002; 3: axonal regeneration. Neuron 2005; 45:353–359. 1745–1752. 112 Fischer D, He Z, Benowitz LI. Counteracting the Nogo receptor enhances optic nerve 94 Ahmed Z, Kalinski H, Berry M, Almasieh M, Ashush H, Slager N et al. Ocular regeneration if retinal ganglion cells are in an active growth state. JNeurosci2004; neuroprotection by siRNA targeting caspase-2. Cell Death Dis 2011; 2:e173. 24: 1646–1651. 95 Deveraux QL, Takahashi R, Salvesen GS, Reed JC. X-linked IAP is a direct inhibitor of 113 Johnson EC, Morrison JC. Friend or foe? Resolving the impact of glial responses in cell-death proteases. Nature 1997; 388: 300–304. glaucoma. JGlaucoma2009; 18: 341–353. 96 McKinnon SJ, Lehman DM, Tahzib NG, Ransom NL, Reitsamer HA, Liston P et al. 114 Fischer D, Petkova V, Thanos S, Benowitz LI. Switching mature retinal ganglion cells Baculoviral IAP repeat-containing-4 protects optic nerve axons in a rat glaucoma to a robust growth state in vivo: gene expression and synergy with rhoA inactivation. model. Mol Ther 2002; 5: 780–787. JNeurosci2004; 24: 8726–8740. 97 Renwick J, Narang MA, Coupland SG, Xuan JY, Baker AN, Brousseau J et al. 115 Lehmann M, Fournier A, Selles-Navarro I, Dergham P, Sebok A, Leclerc N et al. XIAP-mediated neuroprotection in retinal ischemia. Gene Therapy 2005; 13: Inactivation of Rho signaling pathway promotes CNS axon regeneration. JNeurosci 339–347. 1999; 19: 7537–7547. 98 Cipolat S, Rudka T, Hartmann D, Costa V, Serneels L, Craessaerts K et al. Mitochon- 116 Lorber B, Howe ML, Benowitz LI, Irwin N. Mst3b, an Ste20-like kinase, regulates axon drial rhomboid PARL regulates cytochrome c release during apoptosis via OPA1- regeneration in mature CNS and PNS pathways. Nat Neurosci 2009; 12: dependent cristae remodeling. Cell 2006; 126: 163–175. 1407–1414. 99 Ju WK, Duong-Polk KX, Lindsey JD, Ellisman MH, Weinreb RN. Increased optic 117 Park KL, Hu Y, Smith PD, Wang C, Cai B, Xu B et al. Promoting axon regeneration atrophy type 1 expression protects retinal ganglion cells in a mouse model of in the adult CNS by modulation of the PTEN/mTOR pathway. Science 2008; 322: glaucoma. Mol Vis 2010; 16: 1331–1342. 963–966. 100 Ellouze S, Augustin S, Bouaita A, Bonnet C, Simonutti M, Forster V et al. Optimized 118 Kurimoto T, Yin Y, Omura K, Gilbert H-y, Kim D, Cen L-P et al. Long-distance axon allotopic expression of the human mitochondrial ND4 prevents blindness in a rat regeneration in the mature optic nerve: contributions of oncomodulin, cAMP, and pten model of mitochondrial dysfunction. Am J Hum Genet 2008; 83: 373–387. gene deletion. J Neurosci 2010; 30: 15654–15663.

Gene Therapy