Gene-Based Treatment of Motor Neuron Diseases

Home , Lower motor neuron, Motor neuron, Motor neuron disease

INVITED REVIEW ABSTRACT: Motor neuron diseases (MND), such as amyotrophic lateral sclerosis (ALS) and spinal muscular atrophy (SMA), are progressive neurodegenerative diseases that share the common characteristic of upper and/or lower motor neuron degeneration. Therapeutic strategies for MND are designed to confer neuroprotection, using trophic factors, anti-apoptotic proteins, as well as antioxidants and anti-excitotoxicity agents. Although a large number of therapeutic clinical trials have been attempted, none has been shown satisfactory for MND at this time. A variety of strategies have emerged for motor neuron gene transfer. Application of these approaches has yielded therapeutic results in cell culture and animal models, including the SOD1 models of ALS. In this study we describe the gene-based treatment of MND in general, examining the potential viral vector candidates, gene delivery strategies, and main therapeutic approaches currently attempted. Finally, we discuss future directions and potential strategies for more effective motor neuron gene delivery and clinical translation. Muscle Nerve 33: 302–323, 2006

GENE-BASED TREATMENT OF MOTOR NEURON DISEASES

THAIS FEDERICI, PhD,1 and NICHOLAS M. BOULIS, MD1,2

1 Department of Neuroscience, Cleveland Clinic Foundation, NB2-126A, 9500 Euclid Avenue, Cleveland, Ohio 44195, USA 2 Center for Neurological Restoration, Cleveland Clinic Foundation, Cleveland, Ohio, USA

Accepted 5 August 2005

Motor neuron diseases (MND) are progressive neu- ginally effective. A detailed understanding of the rodegenerative disorders with different etiologies pathogenesis of MND remains elusive, and this gap and clinical variability, but a common ﬁnal event: continues to impede the elaboration of speciﬁc treat- loss of upper and/or lower motor neurons. Amyo- ment strategies. Nonetheless, based on our limited trophic lateral sclerosis (ALS) and spinal muscular insight into the pathogenesis of the MNDs, a variety atrophy (SMA) are the most frequent forms of MND of approaches have emerged. Although these are not and therefore the most studied.100,144,184,185 Al- likely to be curative, they may provide life-prolong- though a variety of therapeutic trials in ALS and ing treatments. Gene-based treatment has several SMA are ongoing, available treatments remain mar- advantages over alternative paradigms that may deliver these therapies for MND.5,29,167

Available for Category 1 CME credit through the AANEM at www. aanem.org. AMYOTROPHIC LATERAL SCLEROSIS Abbreviations: AAV, adeno-associated virus; ALS, amyotrophic lateral sclerosis; BDNF, brain-derived neurotrophic factor; BHK, baby hamster kidney; ALS, also known as Lou Gehrig’s disease or adult CNS, central nervous system; CNTF, ciliary neurotrophic factor; CT-1, cardiotrophin-1; EAAT2, excitatory amino acid transporter 2; EIAV, equine infec- motor neuron disease, was first described by Charcot tious anemia virus; FALS, familial amyotrophic lateral sclerosis; GDNF, glial in 1869 and is the most common form of MND. ALS cell line–derived neurotrophic factor; GFAP, glial fibrillary acidic protein; GFP, green fluorescent protein; HIV-1, human immunodeficiency virus type 1; HSV, is characterized by a progressive degeneration of herpes simplex virus; IAPs, inhibitors of apoptosis proteins; IGF-1, insulin-like lower and upper motor neurons in the cerebral growth factor 1; MND, motor neuron diseases; NAIP, neuronal apoptosis inhibitory protein; NT-3, neurotrophin-3; p75NTR, p75 neurotrophin receptor; cortex, brainstem, and spinal cord. In turn, this neu- pmn, progressive motor neuropathy; PNA, peptide nucleic acid; RabG, rabies ronal loss causes weakness, spasticity, and muscular G protein; RNAi, RNA interference; SALS, sporadic amyotrophic lateral sclerosis; siRNA, small interfering RNA; SMA, spinal muscular atrophy; SMN1 and atrophy that may evolve to paralysis. The disease SMN2, survival motor neuron gene; SOD1, copper–zinc superoxide dismutase; trk, receptor tyrosine kinases; VEGF, vascular endothelial growth culminates in death within 2–5 years of onset, gen- factor; VSV-G, vesicular stomatitis virus G glycoprotein; XIAP, X-linked inhib- erally due to respiratory failure. To date, ALS re- itors of apoptosis protein Key words: amyotrophic lateral sclerosis; gene therapy; retrograde axonal mains untreatable and all therapeutic trials are transport; spinal muscular atrophy; viral vectors merely palliative.47,100,184,185 N. M. Boulis; e-mail: [email protected] Correspondence to: ALS is a multifactorial disease, with etiological © 2005 Wiley Periodicals, Inc. Published online 14 October 2005 in Wiley InterScience (www.interscience. heterogeneity and a high variability of clinical pre- wiley.com). DOI 10.1002/mus.20439 sentation. The sporadic form (SALS) is widely pre-

302 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 dominant, including approximately 90% of all cases, androgen receptor gene.66,90,100,170 As with ALS, clin- whereas the familial variant (FALS) affects less than ical trials have made no significant impact on SMA 10% of these patients and is usually autosomal-dom- survival. inant.61,184,185 Close to 20% of familial cases have been found to RESEARCH have mutations in the copper–zinc (Cu/Zn) superoxide dismutase-1 (SOD1) gene on chromosome Research using animal models that reproduce the 21q.91,163 Over 100 different mutations of the SOD1 phenotype of MND has provided substantial insight 21,58 gene have already been described and several trans- into the human disease over the past decade. genic animal models for the SOD1 gene mutation There are many different mouse models for ALS, have been created.77,161 including those with spontaneous mutations, such as No satisfactory treatment is available for ALS at the motor neuron degenerative (mnd) mouse, the this time. Patient care focuses on symptomatic treat- wasted mouse, the wobbler mouse, and the progres- ment and physical therapy. Assisted ventilation and sive motor neuronopathy (pmn) mouse; and geneti- nutrition can transiently overcome the loss of upper cally engineered mice, such as the SOD1 transgenic airway and respiratory muscular control. A large mouse and the transgenic mouse that overexpresses number of therapeutic trials have been attempted. mutant neurofilaments. Although all these models Nonetheless, it was not until 1996 that the first drug are complementary and very helpful, the complex approved by the FDA for the treatment of patients etiology of ALS remains unclear and more studies with ALS reached the market. Riluzole is an anti- are required to provide therapies capable of prevent- glutamatergic substance that blocks the presynaptic ing or slowing the disease course.197 release of glutamate.103,104 The efficacy of riluzole is In the case of SMA, as SMN is a highly expressed questionable, however, with minimal therapeutic protein essential for survival, the development of benefits.177 animal models is extremely difficult. For example, the attempt to create a complete SMN knockout in 172 SPINAL MUSCULAR ATROPHY mice was hampered by its embryonic lethality. In fact, the age at onset of disease is a challenge among SMA is the generic name for a heterogeneous group all the different mice models of MND. More re- of diseases characterized by the degeneration of mo- cently, the creation of several mouse models, includ- tor neurons in the anterior horn of the spinal cord ing the novel SMN knockout mouse,66,144 has facili- and the brainstem (lower motor neurons), causing tated improved understanding of SMA and the progressive muscle weakness and atrophy. Like ALS, development of therapeutic strategies. SMA is invariably fatal. SMA remains one of the most Currently gene therapy is among the most prom- frequent genetic causes of death in childhood.43,144 ising treatments for ALS5,29,59,133 and SMA.13,55,110 The classification of SMA is controversial and This review focuses on the gene-based treatment of includes several forms, separated according to age of motor neuron diseases in general, surveying the po- onset, severity of symptoms, and evolution. Pediatric tential viral vector candidates, delivery strategies, SMA can be roughly divided into three forms: type 1, and main therapeutic approaches that have severe, infantile, also known as Werdnig–Hoffmann emerged. Finally, recent progress and potential fu- disease; type 2, the intermediate form; and type 3, ture strategies toward an effective motor neuron mild, juvenile, also called Kugelberg–Welander dis- gene therapy are also discussed, with emphasis on ease. Adult SMA includes distinct complex syn- ALS and SMA. dromes. One example is Kennedy’s syndrome, also known as progressive spinal and bulbar muscular PATHOGENESIS atrophy.43,46,90,100 The childhood forms, although phenotypically Several mechanisms have been postulated to explain different, are recessive autosomal disorders. The ma- motor neuron death in ALS. However, none of these jority are caused by homozygous deletion or muta- is completely satisfactory to elucidate the entire pro- tions in the telomeric copy of the survival motor cess.58,177,184 Among the proposed mechanisms, glu- neuron gene (SMN1) on chromosome 5q, which tamatergic oxidation and excitotoxicity have been codes the functional copy of the SMN protein. explored in the greatest depth. This mechanism pos- Kennedy’s disease, by contrast, is an X-linked reces- tulates that an excess of glutamate outside of the sive disorder and its molecular basis resides in the motor neuron, probably caused by inefficient clear- expansion of a trinucleotide (CAG) repeat in the ance (e.g., altered metabolism or defective function

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 303 of glutamate transporters), leads to an increase in SOD1, and findings of abnormal mitochondrial mor- levels of free intracellular calcium, resulting in a phology are evidence corroborating the role of mo- cascade of damaging processes by the generation of tor neuron apoptosis in ALS.75,112,123,149,164,203 In the free radicals.47,177 Conversely, the oxidative stress SOD1 mouse model of ALS, caspase-1 and caspase-3 that occurs due to mutations in the SOD1 gene can are sequentially activated.199 Caspase-9 activation, cy- result in more accumulation of glutamate in the toplasmic cytochrome c, and mitochondrial Bax neural tissue and more excitotoxicity. In addition, translocation have also been documented in the the oxidative lesion damages the membranes of the SOD1 mouse spinal cord.75 In addition, expression of neurons and, consequently, blocks cellular func- a mutant caspase-1 protein65 as well as the intraven- tions. The role of glia in this mechanism remains tricular injection of caspase inhibitors,112 both capa- controversial. Reactive oxygen and reactive nitrogen ble of inhibiting MN apoptosis, improve the survival species (ROS and RNS) may derive from activated of SOD1 mutant animals. glia. However, the presence of oxidants can also lead Finally, autoimmune disease9 and viral infec- to inflammatory reactions and activate glial tion23 have also been implicated as possible mecha- cells.7,17,82,128 nisms for motor neuron degeneration in ALS. In Neurofilament aggregation within the cell body fact, at present it remains undetermined which pro- and proximal axon of motor neurons has also been cess is dominant and whether one of these mecha- proposed as a possible mechanism. As with glial nisms gives rise to the others. The cascade of events activation, whether these aggregates are a cause or that leads motor neurons to death is complex and effect of the underlying pathophysiological process these events are all linked. For this reason, gene is unclear. These deposits can also lead to axonal therapy approaches to ALS have, for the most part, transport defects that culminate in neuronal death. focused on inhibiting the final common pathway of Conversely, they may result from a perturbation of motor neuron death, apoptosis. The therapeutic axonal transport.107,164,177,204 transgenes employed to achieve this end have in- Whether the problem resides in neurons or as- cluded intracellular inhibitors of pro-apoptotic cas- trocytes remains an open question. Some evidence cades and neurotrophic factors. As will be discussed, supports the idea that reactive astrocytes cause mo- attempts have also been made to inhibit motor neu- tor neuron death by a nerve growth factor (NGF)– ron excitotoxicity and downregulate the expression mediated mechanism involving nitric oxide and per- of mutant genes in familial forms of ALS. oxynitrite formation.45 In fact, astrogliosis is a In contrast to idiopathic ALS, the mechanisms of common finding in ALS patients and ALS transgenic SMA appear to have been better elucidated. Like mice. However, it remains undetermined whether ALS, a preponderance of data suggest that a pro- this is a primary finding or a consequent event upon grammed neuronal death (apoptosis) plays the ma- the oxidative stress in motor neurons.169 A neuroin- jor role in this disorder. Indeed, the specific gene flammatory process with the accumulation of reac- that causes SMA has been identified. Deletion or tive microglial cells is also observed in the degener- mutations in the SMN1 gene leads to motor neuron ating areas and may contribute to motor neuron apoptosis, forming the pathogenesis of SMA. It was degeneration.86,218 Finally, a significant body of liter- demonstrated that SMN delays the onset of apopto- ature supports the downregulation of the excitatory sis and that the C terminal of hSMN is important for amino acid transporter 2 (EAAT2), expressed its cell-death inhibitory function.200 However, the mainly in astrocytes, as a cause of motor neuron reasons that motor neuron degeneration occurs se- excitotoxicity.17,40,73 lectively in the spinal cord are not yet under- Apoptosis, triggered by a mitochondrial-depen- stood.46,66,90,144,170 More importantly, the develop- dent pathway or induced by glutamatergic oxidation, mental time window for the critical insult to the leads to motor neuron death in model systems and motor neuron is not clearly understood. Although it plays an important role in the pathogenesis of is tempting to believe that SMN1 replacement in ALS.153 Apoptosis involves the activation of a specific motor neurons will prevent motor neuron degener- intracellular cascade that culminates in DNA frag- ation, this approach is doomed to failure if the ab- mentation and chromatin clumping. Proteases sence of SMN initiates a delayed deterioration in known as caspases are a critical precursor to the utero.129 Similarly, it is not entirely clear that SMN cascade that leads to DNA fragmentation. The re- deficiency in motor neurons is the cause of their lease of mitochondrial contents can activate this death. Despite these theoretical concerns, SMA gene caspase cascade. The overexpression and functional- therapy focuses on SMN1 replacement in motor neu- ity of different caspases reported in the mutant rons as a first-pass approach.

304 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 Thus, whereas the molecular mechanisms under- the immunoprivileged compartment of the central lying MND are not entirely clear, all forms of ALS nervous system (CNS). This mechanism can be and SMA have a common outcome: motor neuron equally employed to deliver therapeutic genes to injury and death. We believe that the most effective threatened motor neurons. Vectors derived from treatments will derive from a better understanding adenovirus, herpes simplex virus (HSV), adeno-asso- of the possible genetic, toxic, inflammatory, devel- ciated virus (AAV), and lentivirus are potential can- opmental, and infectious etiologies of this process. didates for the treatment of neurodegenerative dis- However, given the tremendous human cost associ- orders, because they can transduce neurons in vitro ated with motor neuron diseases, attempts at em- and in vivo.4,52,69,72,142 ploying less specific strategies are not only warranted Delivery of genes, trophic factors, anti-apoptotics, but imperative. antioxidants, and other drugs has been used extensively in animal models to prevent or restore neuron GENE THERAPY damage. We will discuss the therapeutic strategies, delivery methods, and candidate genes that are Human gene therapy using viral vectors has gained emerging for a gene-based treatment of MND.154 considerable attention over the past two decades. Adenoviral Vectors. Adenoviral vectors are non- Gene therapy is considered one of the most promis- enveloped, double-stranded DNA viral vectors. They ing approaches for disorders that currently lack ef- have been one of the most widely used vehicles for fective treatment, including inherited and acquired gene delivery. Many studies conducted in the early diseases. Although its original concept was focused 1990s characterized the neural tropism of adenoviral on the treatment of inherited genetic diseases, the vectors in vitro and in vivo.4,16,42,51,108 Although ade- current status of gene therapy shows cancer as the novirus is not known to infect the CNS, work with major indication in clinical trials (Journal of Gene adenoviral vectors has demonstrated their capacity Medicine clinical trial site: http://www.wiley.co.uk/ to undergo retrograde axonal transport.36,37,39,131,137 genetherapy/clinical). Gene therapy presents an ex- In a therapeutic context, several investigators de- citing prospect for treatment of neurodegenerative scribed the use of adenoviral vectors encoding dif- disorders, including MND, that also urgently need ferent neurotrophic factors, by intramuscular injec- an effective treatment. tion of these vectors in newborn normal rats, pmn mutant mice, or SMA mutant mice.19,32,70,78,79,110 In Viral Vectors. In order to achieve efficient gene de- most reports, investigators used the same recombi- livery, viral vector genomes encode a functional gene nant adenoviral vector construction, but different expression cassette. The expression cassette is the neurotrophic factor genes driven by the Rous sar- coding region (DNA sequence that contains the coma virus (RSV) long-terminal repeat (LTR) pro- transgene), flanked by the regulatory elements nec- moter showing therapeutic effects such as neuropro- essary for its transcription (promoter and polyade- tection and improved lifespan. nylation signal). Promoters drive and can even re- Vectors based on human adenovirus serotypes 2 strict gene expression to certain cell types, whereas and 5 continue to show increasing promise as gene the polyadenylation (polyA) sequence is involved in therapy delivery vehicles. However, the host immune the end of the transcription process. Expression cas- response remains a major safety issue in the context settes may be simple, double, or multigenic, for the of adenoviral-mediated human gene therapy, mak- expression of one or more genes of interest, according clinical application of these vectors in the ner- ing to the required therapy and the cloning capacity vous system questionable.117,118 The immune re- of each vector. sponse to adenoviral vectors also plays a critical role In theory, a viral vector candidate for the treat- in turning off CNS gene expression mediated by ment of MND should be able to deliver therapeutic these vectors.192 Thus, different investigators have genes to motor neurons, while preserving motor reported varying durations of gene expression by function. It is also desirable for the vector to be these vectors in the CNS. capable of retrograde transport and to mediate long- The recent development of new generations of term gene expression.167 Retrograde axonal trans- adenoviral vectors might help to overcome this prob- port provides a direct conduit through which a virus lem. Helper-dependent adenoviral vectors have can move from the periphery into the central ner- demonstrated reduced toxicity, as well as increased vous system, bypassing the blood–brain barrier. Neu- cloning capacity and prolonged transgene expres- rotropic viruses have employed this evolutionary sion, making them viable for use in the future.181 strategy to avoid the immune system, passing into Nonetheless, at present, this vector system has

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 305 proven somewhat cumbersome. These difficulties ruses require a helper virus to infect hosts, which can are justified if an increased cloning capacity is re- be either adenovirus, HSV, or vaccinia virus. Recombi- quired by the need for larger therapeutic genes. At nant AAV (rAAV) vectors are capable of transducing present, most of the therapeutic expression cassettes numerous dividing and nondividing cell types. They that have been developed for application to MND promote prolonged in vivo gene expression and are, remain small enough to accommodate in other vec- for the most part, nonpathogenic in the wild-type, tor systems. The need for a large cloning capacity for giving them an attractive safety profile. Despite their the delivery of the dystrophin gene in neuromuscular low immunogenicity, some immune response has been gene therapy of Duchenne’s muscular dystrophy has reported against the vector capsid and the transgene. driven the continued development of this class of An intensive effort is currently being made to improve vectors. AAV vector systems in safety, stability, and regulation of HSV Vectors. Vectors derived from herpes sim- gene expression.217 plex virus type 1 (HSV-1), a natural neurotropic, en- To date, eight different serotypes have been developed, double-stranded DNA virus, are promising for scribed, which display different tissue tropisms and gene therapy applications. This virus has the advan- patterns of transduction.50,189 For this reason, the vast tages of a large cloning capacity and the potential to majority of clinical trials ongoing and planned for ap- remain in a latent state within neurons. Further, be- plication to neurodegenerative diseases are based on cause HSV-1 naturally infects nerve terminals and the AAV vectors. The clinical protocol for Canavan’s dis- sensory neurons of the dorsal root ganglia, it has ease, using AAV-aspartoacylase (ASPA), was the pio- evolved effective mechanisms for retrograde transport. neering approved clinical use of AAV vectors in the Vectors derived from HSV-1 have been generated and human brain.92 Other examples of approved protocols demonstrated to infect neurons in vitro and in vivo, include the ongoing phase I clinical trials for Parkin- and undergo retrograde transport.62,69,113,121,151 The son’s disease, using AAV–hAADC (aromatic-l-amino tendency to infect sensory neurons lends HSV to ap- acid decarboxylase) and AAV–GAD (glutamic acid de- plication in the treatment of pain and sensory neurop- carboxylase), as well as the use of AAV–NGF for Alz- athies. Nonetheless, these vectors are capable of motor heimer’s disease (Journal of Gene Medicine clinical trial neuron gene delivery. site: http://www.wiley.co.uk/genetherapy/clinical). It Together, these characteristics, including the is hoped that all these clinical trials will be able to possibility of insertion of multiple genes, are advan- confirm the safety of AAV vectors in the human CNS. tageous for the development of delivery systems for Serotype 2 (rAAV2) is the most studied and, the treatment of MND.148 Progress has been made to therefore the most used, in in vivo studies and clin- overcome immunity and transgene expression is- ical trials. Experiments comparing the transduction sues, supporting the use of these vectors as therapeu- efficiency of distinct serotypes in the neonatal and tic candidates for the treatment of MND.143,151 The adult CNS of rats and mice have shown distinct ability of the herpes latency promoter to drive gene patterns of gene expression, depending on the sero- expression while the virus remains in a quiescent type, the promoter, and the injected area of brain. state in the neuron has created hope that sustained After intrastriatal injection, for example, whereas gene expression can be achieved in the nervous rAAV2 transduces predominantly neurons,18,127 sero- system through HSV vectors.155,168 However, at types 1, 4, and 5 show a broader tropism, also trans- present, sustained expression has remained difficult ducing astrocytes and ependymal cells.52,201 to achieve. Furthermore, the ability of these vectors to be It is also critical to point out that application of transported retrogradely both in the CNS95 and pe- viral vectors to large-scale human therapy will re- ripheral nervous system38,39,119 makes them even quire the production of large quantities of vector. more appropriate and attractive candidates for gene This so-called “scaleability” problem may undermine therapy for MND. Nonetheless, in our experience, therapies based on elegant vector systems that are rAAV remote delivery has proven less reliable than hard to “scale up.” Proponents of HSV vectors point HSV and adenoviral vectors. out that these vectors have been widely employed in Lentiviral Vectors. Lentiviral vectors are among a variety of clinical trials and have proven relatively the most promising viral vectors for clinical applica- easy to produce in large quantities. tions for neurodegenerative diseases. They combine AAV Vectors. Adeno-associated viral vectors (AAV) advantages of midrange cloning capacity with stable are based on the adeno-associated virus, a single- gene expression and an attractive safety profile due stranded DNA, nonpathogenic, defective member of to a minimal inflammatory response to the vectors. the Parvoviridae family, genus dependovirus. These vi- In addition, as discussed in what follows, the mem-

306 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 brane capsule of lentiviral vectors renders alterations in tropism relatively easy. Lentiviral vectors are based on the single-stranded RNA (ssRNA) lentiviruses, which are a subclass of retrovirus. They are able to transduce both dividing and nondividing cells, including neurons, to integrate transgenes of interest into their target cells, promoting long-term gene expression. They can also be retrogradely transported depending on the specific glycoprotein expressed on the envelope.12,126,141,142 Currently, vector systems derived from different lentiviruses are under development, and a series of improvements have been made in these vectors, making them both safer and more efficacious. The most widely used lentiviral vectors for gene transfer to the CNS are based on human immunodeficiency virus type 1 (HIV-1). Besides HIV-1, vectors derived from other lentiviruses, such as the non- primate equine infectious anemia virus (EIAV) and feline immunodeficiency virus (FIV), represent other options of gene transfer vector systems to the CNS.14,152 FIGURE 1. Schematic diagram showing methods for gene-based motor neuron therapies. Therapeutic transgenes can be deliv- Lentiviral vectors are usually pseudotyped with ered to upper and lower motor neurons in the brain and spinal the vesicular stomatitis virus G glycoprotein (VSV-G) cord by direct injection; remote gene transfer (using a vector that envelope, but the use of alternative envelopes is of can be retrogradely transported from muscles and peripheral interest and has been explored in the engineering of nerves), or by ex vivo gene delivery of cells engineered in vitro. new vectors. The lentipseudotyping strategy is discussed in greater detail in a later section focused on targeted vectors. effective for motor neuron gene delivery, but not The large-scale production of AAV and lentiviral glial gene delivery. If the core defect of MND proves vectors remains to be established. The production of to be glial, remote delivery will be insufficient. The these vectors requires stable packaging cell lines. inherent lack of specificity created by direct injec- Moreover, in the case of AAV vectors, the low vector tion can be overcome through the modification of solubility and aggregation of concentrated vector are vector tropism or the development of cell-type–spe- other challenges in production. cific expression cassettes through specific promoters. Furthermore, a variety of data suggest that prac- MOTOR NEURON GENE DELIVERY tical remote delivery may not be possible in human Theoretically, gene transfer to upper and lower mo- MND patients.87,90,107,139,204 First, axonal transport tor neurons can be achieved either by direct injec- may be impaired in MND, preventing adequate de- tion into the brain and spinal cord, remote delivery livery in symptomatic patients. Second, there are using a vector that can be retrogradely transported, inherent disadvantages to the remote delivery, as or by ex vivo gene transfer (Fig. 1). These methods discussed in greater detail in the next section. of delivery overcome the challenge that the blood– We have demonstrated effective gene transfer brain barrier represents to the passage of therapeu- after direct spinal cord injection with three different tic macromolecules into the CNS. However, the fea- vectors, including adenoviral NGF in normal rats,35 sibility of these techniques of administration is AAV2 vectors, and a rabies G (RabG) protein– hampered by other difficulties. pseudotyped lentiviral vector in SOD1 transgenic mice188 (Fig. 2). Nonetheless, direct injection of the Direct Injection. All the vectors just described are human spinal cord remains a daunting challenge capable of neuronal gene delivery in vivo. Direct with serious potential morbidity. However, taking injection can reliably deliver high titers of vector into into account the grave prognosis of MND patients the targeted region. Further, because remote deliv- and the imperative for practical clinical translation ery paradigms depend on axonal transport, they are in the near future, we have begun work to develop

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 307 more feasible for human application. Intramuscular Delivery. The feasibility of this method of gene delivery using viral vectors has previously been described using normal animals. For example, Kahn and colleagues94 constructed adenoviral vectors encoding genes for various neurotrophic factors and demonstrated axonal transport in spinal motor neurons after intramuscular injection. Yamamura and collaborators211 also reported gene expression in the anterior-horn motor neurons after intramuscular inoculation with an HSV vector. Recently, the many successful attempts using animal models of MND have encouraged the use of this route of delivery. Among others, Wang and col- 202 FIGURE 2. Direct gene delivery of an advanced generation vec- leagues showed gene expression and neuroprotec- tor into the cervical spinal cord of a SOD1 transgenic mouse. tive effects in muscle and anterior-horn motor neu- Effective motor neuron gene expression can be achieved in the rons after intramuscular injection of an AAV–GDNF cervical spinal cord of mutant mice through intraparenchymal (glial cell line–derived neurotrophic factor) vector injection without deleterious behavioral effects, reduced survival, in a transgenic animal model of ALS. Lesbordes et or substantial inflammation, using an EIAV vector. X-gal staining 110 showing the transduced area by the EIAV–LacZ vector. al. demonstrated that intramuscular injection of adenoviral vector–cardiotrophin-1 (CT-1) improved therapeutic effects in a mouse model of SMA. Kaspar the tools and techniques to make restorative spinal and colleagues96 also achieved retrograde viral deliv- cord surgery possible. ery and neuroprotective effects by intramuscular injection of AAV vectors encoding insulin-like growth Remote Delivery. As just discussed, several types of factor 1 (IGF-1) in a mouse model of ALS. Finally, viral vectors are capable of migrating by retrograde Mazarakis and Azzouz used pseudotyped EIAV to axonal transport to the cell body of projection neu- achieve gene expression in the spinal cord, after rons, after internalization at the axon terminal36,38 (Fig. 3). This noninvasive remote gene delivery, that is, from sites outside the CNS (muscle or peripheral nerves), might be applicable to the future clinical treatments of a variety of neurodegenerative diseases. In addition, direct intraparenchymal spinal injection of vectors capable of retrograde axonal movement provides a strategy that is potentially capable of gene delivery to both upper and lower motor neurons. Despite the fact that some investigators have described abnormal axonal transport in the SOD1 mutant mouse,139 other groups have achieved successful delivery using this method.131,157,212 Both routes of administration, intramuscular and intraneural, have been used extensively in animal models. Both must overcome real challenges for human gene therapy, due to larger muscle volume and the length of human motor neuron axons. Furthermore, the high binding affinity of some vectors in muscle and the FIGURE 3. Schematic representation of peripheral viral vector limited binding affinity at synaptic terminals reduce uptake. Adenoviral vectors, adeno-associated viral vectors, and neuronal uptake and retrograde axonal transport, RabG-pseudotyped lentiviral vectors can be retrogradely trans- especially of AAV vectors.50,83,124 ported to the cell body of motor neurons (MN) and deliver therapeutic transgenes. In ALS, axonal defects may limit remote Progress in vector design, by restricting the bind- gene delivery. Enhanced uptake and retrograde transport ing affinity and enhancing the delivery to target achieved through vector pseudotyping may be required to over- cells, might optimize these techniques to make them come these defects.

308 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 with intramuscular or subcutaneous injection.36 More recently, we confirmed the role of axonal transport in remote delivery of adenoviral and AAV vectors, through colchicine inhibition, after intraneural injection.39 Subsequently, we also demonstrated remote delivery of AAV genes to the spinal cord following peripheral nerve injection38 (Fig. 4). Finally, we reported EIAV gene expression in cervical spinal cord motor neurons of SOD1 mice after brachial plexus injection187 (Fig. 5). Together, these distinct approaches point to the potential of this methodology for minimally invasive spinal cord mo- FIGURE 4. Adeno-associated viral vector gene expression in the tor neuron gene therapy. spinal cord after remote delivery by sciatic nerve injection. Anti– green fluorescent protein (anti-GFP) immunofluorescence re- Ex Vivo Gene Therapy. In ex vivo gene therapy, cells veals GFP-positive ventral horn motor neurons, after intraneural injection of rAAV.GFP. are genetically modified and then transplanted into the region of injury. The use of this strategy has been explored for the treatment of different neurodegen- multiple or single intramuscular injections of lenti- erative diseases, including MND, by grafting cells viral vectors, in different approaches.13,15,157 At- that can secrete neurotrophic factors and thus con- tempts have been made in some of these experiments to demonstrate the importance of retrograde vector delivery for therapy. However, some of the observed therapeutic effects may result from retrograde transport of the transgene product rather than the vector itself. Nonetheless, this potential dual mechanism for therapy represents an advantage for intramuscular delivery strategies. As previously alluded to, whereas the intramuscular approach has shown efficacy with different vector systems in mouse models, it faces a difficult challenge in translation to humans. The nerves of humans, particularly those that innervate the distal extremities and diaphragm, are extremely long. More importantly, the muscle mass of humans vastly exceeds that of the mouse, requiring vector diffusion across long distances to achieve uptake at the neuromuscular junction. To date, there are no data to support the feasibility of this approach in a large- animal model. Intraneural Delivery. In order to avoid the problem of muscle binding and uptake, but more importantly to concentrate vector in a region with many motor axons, we have experimented with injections directly into the peripheral nervous system. Al- though less invasive than spinal cord surgery, this approach still carries potential morbidity. Further- more, the majority of viral uptake is likely to occur at the synaptic terminal rather than the axon itself. Thus, the precise mechanism whereby vectors enter FIGURE 5. Remote delivery of a RabG-pseudotyped EIAV vector axons is unclear. in a SOD1 transgenic mouse. (A) Photomicrograph showing gene expression of RabG.EIAV.LacZ vector. ␤-Galactosidase We previously demonstrated that sciatic nerve expression in cervical spinal cord motor neurons (arrows). (B) injection resulted in significantly higher adenoviral Macroscopic view of the injection site (brachial plexus) and the gene expression in spinal cord neurons as compared remote delivery site (cervical spinal cord).

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 309 fer neuroprotection. This approach can also over- scriptional elements into viral genomes to restrict come the difficulties that arise from the systemic the expression of therapeutic genes to certain target delivery of these molecules.193 cells. The use of cell-specific promoters,22,102 as well Encapsulated baby hamster kidney (BHK) cells as enhancers of genes for motor neuron–specific genetically engineered to release hCNTF were im- expression10,196 are improvements that might also planted in pmn mutant mice, resulting in therapeutic optimize the targeted gene expression strategy. effects, including an increase in mean survival of the animals and reduction of motor neuron loss.166 A Lentipseudotyping. Because of their lipid mem- phase I study involving six ALS patients also demon- brane capsule, lentiviral vectors can be easily tar- strated the feasibility of the ex vivo gene delivery geted through the insertion of envelope glycopro- technique. Patients were intrathecally implanted teins. These vectors are usually pseudotyped with the with polymer capsules containing engineered BHK vesicular stomatitis virus G glycoprotein (VSV-G). cells. These capsules were demonstrated to release VSV-G–pseudotyped lentivectors do not undergo re- CNTF. Consequently, measurable CNTF levels were liable retrograde transport.96 However, the use of detected in the cerebrospinal fluid for at least 17 alternative envelopes may allow the targeting of spe- weeks posttransplantation.3 In another approach, cific cell types. For example, when injected into the myoblasts retrovirally transduced with GDNF were muscle, RabG-pseudotyped lentiviral vectors are ret- implanted into hindlimb muscles of SOD1 mutant rogradely transported and transduce motor neurons mice; GDNF gene delivery prevented motoneuron of lumbar spinal cord, in contrast to VSV-G vectors loss and disease progression.138 that transduce exclusively muscle at the site of injec- Ex vivo gene transfer has also been applied tion.126 Another work from the same investigators through spinal cord transplant. Recently, efforts described additional lentipseudotyped vectors with have been made to integrate the application of stem distinct transduction patterns and retrograde trans- cell derivatives and gene transfer through ex vivo port ability, when injected into the brain or spinal gene transfer. Klein and collaborators98 demon- cord.207 strated that astrocytes derived from human neural Recently, we reported the neurotropism of novel progenitors could be induced to secrete GDNF after HIV-1–based lentiviral vectors pseudotyped with gly- ex vivo lentiviral vector–mediated GDNF gene deliv- coproteins from different members of the Rhab- ery. These cells were shown to survive for up to 11 doviridae family, Lyssavirus genus (rabies virus, Euro- weeks and integrate into the gray and white matter pean bat Lyssavirus, Lagos bat Lyssavirus, and of the spinal cord in SOD1 mutant rats. GDNF- Duvenhage). We demonstrated their neuronal gene producing astrocytes resulted in upregulation of transfer in vitro191 and their retrograde transport in choline acetyltransferase staining, suggesting physi- vivo, after sciatic nerve injection (Fig. 6). ological activity of the trophic factor. However, as with direct injection for in vivo gene transfer, tech- AAV Pseudotyping. Because their genomes are al- niques for direct spinal cord injection must exist to most identical, the serotype-specific differences of render this approach feasible for human applica- transduction observed with AAV vectors are due to tion. differences in their capsids and receptors required for attachment and internalization of these vec- 165 TARGETED VECTOR tors. The use of capsids of other serotypes rather than AAV2 is an alternative that may provide advan- Targeted gene therapy is a strategy that consists in tages to this vector system.41,156,174 creating vectors that can achieve restricted cell pop- Burger and collaborators41 designed the hybrid ulations. It aims at increasing the specificity and serotypes rAAV2/1, rAAV2/2, and rAAV2/5, and safety of gene therapy, avoiding vector diffusion to reported an almost exclusive neurotropism for all unsuitable structures.22 Vectors can be genetically of them after injection into distinct regions of the engineered to ablate their natural tropism and con- CNS, with differences just in distribution. More fer new binding specificity. In the context of remote recently, Shevtsova and colleagues174 described delivery for MND gene therapy, the goal is to en- the better performance and increased spread of hance vector uptake at the neuromuscular junction serotype 5 capsid in the brain compared to the through increased motor neuron affinity and limit- serotype 2 capsid. ing muscle affinity. Currently, targeting strategies include modifica- Neurotropic Peptides. Targeted gene delivery can tions of the vector’s capsid or incorporation of tran- also be achieved using peptides with selective affinity

310 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 FIGURE 6. LacZ gene expression following intraneural injection of a Lagos-pseudotyped HIV-based lentiviral vector. Dorsal root ganglia (DRGs) ␤-galactosidase expression revealed by X-gal staining, after remote delivery into the sciatic nerve of a normal adult rat.

for motor neurons, such as the nontoxic tetanus genetics of SMA have motivated SMN1 gene replace- toxin C-fragment that avidly binds to peripheral neu- ment strategies that have shown promise.13 Similarly, rons.63 Some studies have demonstrated the feasibil- an understanding of the molecular genetics of the ity and the advantages of this strategy. Nonviral SOD1 mutant variant of familial ALS has motivated a neuronal gene delivery mediated by the nontoxic selective vector-mediated gene knockdown.134,157,158 tetanus toxin C-fragment has been demonstrated.99 Neuronal targeting of cardiotrophin-1 has been Trophic Factors. Neurotrophins and other trophic achieved through C-fragment coupling.33 Finally, an factors are molecules that promote several effects in SMN:tetanus toxin C-fragment fusion protein was neurons such as differentiation, maintenance of developed for the targeted delivery of the SMN pro- function, synaptic plasticity, survival, and regenera- tein to motor neurons.64 The selective retargeting to tion. Their receptors are the high-affinity members neurons using targeted adenoviral vectors with the of the family of receptor tyrosine kinases (trks; trk A, tetanus toxin C-fragment was also achieved after in- B, and C) and a low-affinity receptor, the p75 neu- tramuscular injection.171 rotrophin receptor (p75NTR). These receptors are In addition to chemically modifying the coats of expressed almost exclusively in the nervous system. viral vectors to create novel motor neuron targeting Some neurotrophic factors are able to bind to mul- strategies, protein capsids can be altered through tiple distinct receptors.54,71,193 alterations in the genes encoding the capsid pro- Several studies have demonstrated alterations in teins. Alteration in AAV tropism has been achieved neurotrophins and their receptors in ALS57,140 that by inserting oligonucleotides into the cap gene of may be relevant in the development of therapeutic AAV2 vectors. This modification can be done by strategies. However, the up- or downregulation of 208 175 peptide insertion into the VP2 and VP3 protein. neurotrophic factors does not appear to be a major We have recently employed phage display to identify cause, but rather a consequence of ALS, contribut- short peptides with selective motor neuron binding ing to neuronal death. Nonetheless, therapy with properties60,116 and inserted it into the VP3 capsid neurotrophins has been extensively explored in protein of AAV2.53 MND models due to their neuroprotective effects. The use of viral vectors encoding neurotrophic THERAPEUTIC TRANSGENES factors to protect motor neurons from acute death Because the molecular mechanisms of motor neu- has been applied exhaustively in a variety of models. ron diseases remain incompletely understood, the However, different studies using viral vectors to de- goals of gene therapy strategies have been focused liver these neurotrophins in vivo have identified lim- on neuroprotection through trophic factors, anti- itations in their effectiveness, such as short half-life apoptotic proteins, and antioxidants and anti-excito- of the neurotrophic factors, poor access due the toxicity genes.5,59 The better understood molecular presence of the blood–brain barrier, insufficient or

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 311 nonspecific delivery, and adverse effects.20,31,193 Even this study, encapsulated BHK cells were modified to so, a number of neurotrophic factors have been secrete hCNTF and implanted into the lumbar in- demonstrated to promote therapeutic benefits, trathecal space of six ALS patients.3 prompting the development of clinical trials. Vector systems to deliver CNTF, such as adenovi- Brain-Derived Neurotrophic Factor. Brain-derived ral and lentiviral vectors, have been shown to be neurotrophic factor (BDNF), among other neuro- successful in achieving neuroprotection. In the ad- protective properties, can prevent glutamate- enoviral study, the investigators compared different induced neurotoxicity, an important mechanism in- routes of administration of adenoviral vector–CNTF volved in the pathogenesis of ALS.114,176 A phase III in pmn mutant mice. While intramuscular and intra- study involving subcutaneous delivery of BDNF venous injections resulted in transduction of skeletal failed to demonstrate clinical benefit in patients with muscle fibers and hepatocytes, increased lifespan, ALS.8 Likewise, intrathecal administration of BDNF and reduction in motor axonal degeneration, the also failed to show treatment efficacy in ALS pa- intracerebroventricular injection resulted in trans- tients, as reported in a phase I/II trial.145 Defects in gene expression restricted to ependymal cells and delivery have been proposed to underlie the gulf no neuroprotective effects.80 In the other study, tet- between activity in models and the clinical setting. racycline-regulated lentiviral vectors were used to Viral gene transfer has been applied in an attempt to investigate the dose-dependent neuroprotective ef- enhance delivery. Pretreatment with adenoviral vec- fect of CNTF. Using a quinolinic acid (QA) rat tor–BDNF prevented the death of axotomized facial model of Huntington’s disease, the researchers motoneurons in newborn rats.70 BDNF has not been showed that, in QA-lesioned rats, gene-controlled applied using more practical, advanced generation delivery of CNTF was associated with reduction of vector systems. the motor deficit.160 Cardiotrophin-1. Cardiotrophin-1 (CT-1) is a Recently, conflicting data have emerged from muscle-derived cytokine that has been shown to pre- studies on endogenous CNTF in clinical MND. Some vent or delay motor axonal degeneration in different postmortem investigations and CNTF measurements mouse models of MND, including pmn,32 wobbler,136 in the cervical enlargement of the spinal cord have SOD1,34 and SMA mutant mice (carrying a deletion attributed motor neuron loss to decreased CNTF of Smn exon 7).110 Most of these studies explored expression in the spinal cord of ALS patients.6,109,146 the use of adenoviral vectors encoding the CT-1 By contrast, it has been demonstrated that CNTF gene, after intramuscular injection, and demon- serum levels are increased in patients with ALS.89 strated neuroprotection. Indeed, in the SMA ap- Moreover, gene expression of CNTF is upregulated proach, the investigators also pointed out the neu- in degenerating spinal motor neurons from autop- roprotective role of CT-1 during the postnatal sied patients with SALS.93 Therefore, although period, in addition to its previously reported role CNTF might have potential therapeutic utility in during the embryonic period.147 As discussed earlier, ALS and other neurodegenerative diseases, its role because neurotrophic factors can themselves be in the pathogenesis of ALS remains unclear. taken up and delivered through retrograde axonal Glial Cell Line–Derived Neurotrophic Factor. Glial transport, intramuscular vector injection may result cell line–derived neurotrophic factor (GDNF) was in gene- or gene product–mediated effects on motor discovered over a decade ago and is currently con- neurons. sidered to be one of the most potent neuroprotec- Ciliary Neurotrophic Factor. Ciliary neurotrophic tive molecules.26,28,85,219 Consequently, viral vector– factor (CNTF), a member of the same family of mediated delivery of GDNF has been applied to the CT-1, promotes motor neuron survival in vitro and CNS in a variety of animal models of neurodegen- in vivo like other neurotrophic factors.97,120,166 CNTF erative diseases, including ALS. has also been shown to effectively reduce motor Many different groups have reported the efficacy of neuron degeneration in animal models such as pmn GDNF gene delivery using the SOD1(G93A) transgenic mice.173 mouse model of ALS. Intramuscular injections of ad- The systemic delivery of recombinant human enoviral vector–GDNF1,122 and AAV–GDNF96,119,202 at CNTF has failed to demonstrate efficacy. A placebo- different ages before the disease onset, conferred neu- controlled trial in 570 patients with ALS showed no roprotective effects, delayed disease onset, and pro- beneficial effects, but rather side effects at all tested longed survival of animals (Fig. 7). Also, as reported doses.132 However, a phase I study using an ex vivo previously, the ex vivo gene delivery of GDNF using gene delivery approach demonstrated measurable myoblasts retrovirally transduced and implanted into levels of CNTF without deleterious side effects. In hindlimb muscles, prevented motor neuron degener-

312 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 FIGURE 7. Survival analysis of G93A SOD1 transgenic mice after different treatments. Kaplan–Meier graph demonstrates increase in lifespan of SOD1 mice, after intramuscular injection of Ad–GDNF (red), AAV–IGF-1 (green), EIAV–VEGF (blue), or EIAV-mediating expression of RNAi targeted to the human SOD1 gene (orange). Untreated SOD1 animals survive an average of 135 days (black). 1: Acsadi et al.1; 2: Kaspar et al.96; 3: Azzouz et al.15; 4: Ralph et al.157 ation and delayed the progression of disease in ALS affects MND patients, this dual effect has generated mice.138 However, although direct injection of lentivi- significant enthusiasm for the application of IGF-1. ral GDNF into the spinal cord promoted long-term However, to date, studies addressing the efficacy of expression in motor neurons, it did not prevent motor this trophic factor in the treatment of MND have had neuron loss and muscle denervation in transgenic conflicting results.24,133,135 mice.76 Recent placebo-controlled studies showed re- Zhao and collaborators219 used the glial fibrilla- combinant human IGF-1 to be a modestly effective tory acidic protein (GFAP)–specific promoter to ex- drug treatment for ALS. Although one investigation, press GDNF in astrocytes of transgenic mice, dem- a European trial, found no satisfactory clinical ef- onstrating its effects on motor neuron survival. They fects,30 a North American trial showed slowing of achieved restricted gene expression predominantly disease progression in patients treated with recom- in astrocytes throughout the brain and spinal cord. binant IGF-1.105 Other clinical trials are currently in The overexpression of GDNF prevented pro- progress. grammed cell death during development and pro- Kaspar and colleagues96 reported efficient re- tected facial motoneurons following axotomy. mote AAV-mediated IGF-1 gene delivery followed For a detailed description of GDNF experimental intramuscular injections, promoting prolonged sur- models the reader is referred to recent articles by vival and delaying disease progression in SOD1 mu- Bohn and colleagues.27,28 tant animals (Fig. 7). In an in vitro model of ALS, Insulin-Like Growth Factor 1. The neuroprotec- experiments suggested that motor neurons express- tive role of insulin-like growth factor 1 (IGF-1) was ing the IGF-1 gene may be capable of protecting reported over a decade ago.111 IGF-1 has the distinct bystander neurons through secretion of the fac- advantage of direct trophic effects on both muscle tor.198 We have also constructed RabG.EIAV vectors and motor neurons. Because diffuse muscle atrophy for the delivery of the IGF-1 gene. This vector has

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 313 proven to induce motor neuron axon elongation Two different study approaches pointed out its and neuroprotection in vitro.190 In addition, this neuroprotective effect and its application in ALS vector sporadically prolonged survival in the SOD1/ treatment. The first study demonstrated that direct G93A model after intramuscular remote injections at intracerebroventricular delivery of recombinant relatively low titers. VEGF was able to delay disease onset, improve motor Nerve Growth Factor. Nerve growth factor performance, and prolong survival in two rat models (NGF) was one of the first members of the neuro- of ALS. Moreover, the investigators showed that trophic factor family to be described.54,193 Although rVEGF was anterogradely transported in peripheral NGF levels were decreased postmortem in cerebral axons, protecting cervical motoneurons and preserv- motor cortex and the dorsal spinal cord of ALS ing neuromuscular junctions.183 More recently, Az- patients, levels in the lateral column of spinal cord zouz and colleagues15 pointed out its neuroprotec- were increased.6 This paradoxical observation has tive effect and its application in ALS treatment. been attributed to deleterious reactive astrocyte pro- Intramuscular injection of RabG-pseudotyped EIAV– duction.150 VEGF resulted in retrograde transport and motor In early experiments, we demonstrated the feasi- neuron delivery of VEGF, promoting prolonged sur- bility of adenoviral-mediated NGF gene delivery and vival in an ALS mouse model (Fig. 7). expression in the spinal cord for application to spi- Summary. Although therapeutic application of nal cord injury.35 Despite NGF’s neuroprotective ac- protein trophic factors has yielded disappointing re- tivity, it has not been successfully applied for ALS sults in clinical trials, gene delivery may overcome treatment. In fact, the suggestion that motor neuron the barrier of sustained and targeted delivery. The death is exacerbated by reactive astrocyte NGF secre- combination of two or more factors may also im- tion has prompted the proposal that NGF gene si- prove the chances of success with this strategy.24,78,97 lencing, rather than delivery, should be applied in ALS.194 Anti-Apoptotic Proteins. Apoptosis plays an impor- Neurotrophin-3. Like NGF and BDNF, neurotro- tant role in neuronal cell death and is considered to phin-3 (NT-3) is a member of the neurotrophin be one of the main possible causes of MND.153,170,200 family. The intramuscular administration of an ad- The apoptotic mechanism is driven by a family of enoviral vector coding for NT-3 in pmn mutant mice proteases (caspases) and controlled by a family of resulted in beneficial effects, including increased proteins (Bcl-2), including Bcl-xL, which regulates survival, reduction of motor axon loss, improvement the cell death repressor activity in the apoptotic in neuromuscular function, and efficient reinnerva- pathway. The downregulation of Bcl-2 has been dem- tion of muscle fibers.79 However, since these initial onstrated in ALS123 and SMA.180 Because anti-apop- experiments, NT-3 has not been explored exten- totic proteins function in the intracellular space, sively for motor neuron protection. It is important to gene transfer provides the advantage of circumvent- consider that NT-3 utilizes the trk C receptor, which ing the need for uptake across the neuronal mem- is not abundantly expressed in either normal or ALS brane. spinal cord motor neurons.57 Bcl-2 and Bcl-xL. The Bcl-2 family includes both Vascular Endothelial Growth Factor. As implied anti-apoptotic and pro-apoptotic members. Bcl-2 by its name, vascular endothelial growth factor family proteins regulate apoptosis by the mitochon- (VEGF) was originally described in the context of drial pathway. Bcl-xL is a member of the Bcl-2 family angiogenesis. However, recent observations have of proteins and is the dominant inhibitor of apopto- demonstrated that this trophic factor has potent di- sis in cell death repressor activity.2,159 The anti-apop- rect neuroprotective properties.15,48,74,106,182,183 totic activity of both molecules, Bcl-2 and Bcl-xL, has Because of its mixed vascular and neurotrophic been achieved in vitro and in vivo, making their effects, VEGF can exercise direct and indirect pro- therapeutic use an attractive tool to confer neuro- tective effects. VEGF can act directly on motor neu- protection. rons as a neurotrophic factor and also regulate Adenoviral vector–Bcl-xL125 and AAV–Bcl-xL67 blood supply to these cells.182 Some studies revealed gene delivery have been shown to prevent apoptosis that reduced levels of VEGF can predispose mice in vitro. These results were obtained in normal pri- and humans to ALS, suggesting a role for VEGF in mary rat neuronal cultures,125 as well as in cultured the pathogenesis of ALS.48 This hypothesis was cor- motor neurons (Fig. 8) and in a glutamate-induced ␦ ␦ roborated by the observation that VEGF / –knock-in apoptosis neuroblastoma cell model.67 mice developed degeneration of adult motor neu- Adenoviral vector–Bcl-2 and AAV–Bcl-2 have also rons and symptoms similar to those of ALS.106 been applied to achieve neuroprotective effects in

314 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 FIGURE 8. Cultured motor neurons transduced by an AAV–Bcl-xL.GFP vector. (A) E15 primary rat motor neurons in vitro (Phase). (B) GFP expression, after transduction with an AAV–Bcl-xL.GFP vector.

animal models of ALS. Yamashita and co-workers212 IAPs are considered stronger and, thus, are po- used intramuscular injection of adenoviral vector– tential candidates for anti-apoptotic treatment strat- Bcl-2 into the tongue of mutant SOD1 transgenic egies, as we showed in an in vitro model of ALS. mice to induce Bcl-2 expression in motor neurons of Using a glutamate-induced model of excitotoxicity, the hypoglossal nuclei by retrograde transport of the we demonstrated that the reduction in cell death was vector. They reported that Bcl-2 expression was able associated with the neuroprotective effects of the to promote motor neuron survival213 by regulating adenoviral vector–XIAP.GFP vector present in the caspase-1 activation and retarding the release of the primary cultured motor neurons.68 pro-apoptotic protein cytochrome c into the cy- tosol.214 Using the same SOD1 transgenic mice, Az- Anti-Excitotoxicity. Excitatory Amino Acid Trans- zouz and collaborators11 also demonstrated Bcl-2 ex- porter 2. Excitotoxicity and glutamate-induced apo- pression in motor neurons, as well as increased ptosis has been considered a major mechanism in the motor neuron survival, after intraspinal injection of pathogenesis of ALS.154,177,185 Increased concentra- an rAAV–Bcl-2 vector. Furthermore, the investigations of glutamate in the cerebrospinal fluid of ALS tors reported a delay in the onset of disease, without patients, decreased levels in the glial glutamate trans- increase in lifespan. porter EAAT2 in the spinal cord and motor cortex of X-Linked Inhibitor of Apoptosis Protein/Neuronal Ap- ALS patients, as well as reports of loss of function of optosis Inhibitory Protein. The inhibitors of apopto- EAAT2 in transgenic SOD1 models, provide substantial sis proteins (IAPs) regulate both the upstream “ini- support for the involvement of this mechanism.84,88 tiator” caspase (caspase-9) as well as downstream Engineered cells overexpressing EAAT2 have “effector” caspases (caspase-3 and -7). Seven mem- been used to deliver EAAT2 gene in vitro and thus bers of the IAP family have been identified. They are confer neuroprotection of motor neurons against the neuronal apoptosis inhibitory protein (NAIP), glutamate toxicity.206 More recently, AAV and lenti- X-linked inhibitor of apoptosis protein (XIAP), viral vector systems have been developed to deliver c-IAP1 (HIAP-1), c-IAP2 (HIAP-2), survivin, livin, EAAT2 gene in vitro and in vivo.81,205 Both studies and Ts-IAP.115 Transfer of the NAIP and XIAP genes showed that EAAT2 can be delivered to motor neu- has proven to have a therapeutic effect in a variety of rons, but have not yet demonstrated a neuroprotec- neurodegenerative disease animal models. In the tive effect in ALS mouse models. four-vessel occlusion model, adenoviral vector– In fact, because EAAT2 is naturally expressed in mediated delivery of IAPs has been shown to prevent astrocytes, targeted delivery is an approach that may CA-1 neuronal death209 and associated behavioral be considered in this case. A gene transfer protocol deficit.210 Vector-mediated XIAP delivery was able utilizing rAAV vector to drive expression of EAAT2 to protect axotomy-induced apoptosis in retinal gan- by the GFAP-specific promoter has been proposed. glion.101 It was also reported that NAIP gene delivery The GFAP promoter is particularly active in the re- reduced 6-hydroxydopamine–induced apoptosis in active astrocytes present in the spinal cord of ALS the rat substantia nigra.49 patients (Paola Leone, personal communication).

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 315 OTHER STRATEGIES have suggested that stem cell therapy, in which Gene Silencing. RNA Interference. Traditionally, degenerated neurons are replaced in order to pro- gene transfer has provided a means for correction of mote recovery in the areas of motor neurons loss, genetic mutations and the restoration of genes or is also a good strategy for the treatment of neuro- proteins that they encode. However, RNA interfer- degenerative diseases, such as Parkinson’s and ence (RNAi) is emerging as an exciting therapeutic Huntington’s diseases and, more recently, ALS approach for the treatment of several genetic disor- and SMA. A recent study has demonstrated the functionality of motor neurons derived from em- ders, selectively silencing the expression of mutant, bryonic stem cells, supporting the idea of cell but not wild-type, genes.44 RNAi is an evolutionary replacement.130 However, many challenges limit conserved process by which double-stranded RNA the success of the stem cell therapy. Besides clin- degrades messenger RNA where there is homology, ical safety and ethical issues concerning the use of silencing posttranscriptional gene expression.44,216 embryonic stem cells, the ability of these cells to The potential of RNAi as a therapy for ALS was migrate, survive, and become functional in the demonstrated by selective inhibition of the mutant environment where they will be grafted has to be SOD1 form, using small interfering RNA (siRNA) considered.178,179 In addition, stem cell–derived sequences with this specificity.56,215 neurons must integrate into a functioning network Improvements on the “vectorization” of siRNA forming afferent and efferent synapses. In order to and the targeted delivery of viral vectors will make reproduce the neuromuscular junction, the axons this technology more practical for application to of these transplanted cells must cross the white neurodegenerative diseases. Two very recent ap- matter of the ventral spinal cord. This white mat- proaches using lentiviral vectors, which can integrate ter contains molecules that suppress axonal the host genome and thus provide long-term gene growth. Further, these axons can be misdirected expression, showed success in silencing the SOD1 into the white matter tracks. Thus, strategies are gene in SOD1 models of ALS.157,158 Furthermore, the required that can guide these axons into the ven- investigators demonstrated a significant increase in tral horn. Viral gene transfer has been used to the lifespan of SOD1 mutant animals157 (Fig. 7). A guide axons in the corpus callosum and spinal third study also showed the benefit of siRNA therapy cord.162 An alternative approach is grafting of for ALS, after retrograde transport of an AAV vector adult stem cells, in order to help “sick” neurons to to the spinal cord following intramuscular injec- recover.186 tion.134 A distinct alternative for silencing gene expres- TOWARD THE CLINIC sion is the antisense peptide nucleic acid (PNA). Systemic intraperitoneal administration of this anti- Neurodegenerative disorders are progressive debili- sense PNA to SOD1 mutant mice conferred signifi- tating diseases that usually have a dire outcome, due cant reduction in p75 neurotrophin receptor gene to the loss of neural tissue. There remain no effective expression.17,194 Because this approach would be ex- treatments for this group of disorders. Gene therapy pected to damp neurotrophic activity, it further sup- employing neuroprotective approaches is an excit- ports the idea that NGF may contribute to motor ing strategy that may prevent neuronal death. Ex neuron degeneration. vivo growth factor gene therapies involving CNTF gene delivery for Huntington’s disease,25 as well as Gene Replacement. The direct replacement of de- NGF for Alzheimer’s disease,195 are examples of fective genes is ideal to treat SMA by increasing the growth factor phase I clinical trials that have been expression of functional SMN protein from the recently performed for neurodegenerative disor- SMN2 gene. This approach has been attempted re- ders. cently using a lentiviral vector expressing human The present review has emphasized different SMN. The ability of this system to restore SMN pro- therapeutic strategies focusing on neuroprotection tein levels in both SMA type 1 fibroblasts and in SMA for MND. A preponderance of evidence supports the mice, has been demonstrated after multiple intra- feasibility of this approach in animal models, espe- muscular injections of a RabG–EIAV vector.13 cially in SOD1 models of ALS. However, these models Stem Cell Therapy. Ex vivo gene therapy, em- reproduce only the FALS, which affects 10% or less ploying stem cell derivatives genetically engi- of ALS patients and, among these, only about 20% neered to enhance neurotrophic factors, has al- have the mutation in the SOD1 gene. Moreover, ready been discussed. In addition, several reports none of these strategies has been tested in humans.

316 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 Practical barriers to gene-based treatment in hu- 6. Anand P, Parrett A, Martin J, Zeman S, Foley P, Swash M, et al. Regional changes of ciliary neurotrophic factor and nerve mans include the duration of gene expression, the growth factor levels in post mortem spinal cord and cerebral distribution of gene expression (targeting of gene cortex from patients with motor disease. Nat Med 1995;1: delivery and broad distribution), and the safety of 168–172. the system (random insertion into human genome, 7. Anneser JM, Chahli C, Ince PG, Borasio GD, Shaw PJ. Glial proliferation and metabotropic glutamate receptor expres- insertion into germ cell lines and inflammatory in- sion in amyotrophic lateral sclerosis. J Neuropathol Exp jury to tissue). The appropriate routes for delivery Neurol 2004;63:831–840. and control of gene expression remain as significant 8. Anonymous. A controlled trial of recombinant methionyl human BDNF in ALS: the BDNF Study Group (Phase III). obstacles. We have begun delivery experiments in Neurology 1999;52:1427–1433. large-animal models to investigate the limits of direct 9. Appel SH, Smith RG, Engelhardt JI, Stefani E. Evidence for and remote injections. The availability of sufficient autoimmunity in amyotrophic lateral sclerosis. J Neurol Sci 1993;118:169–174. resources to “scale-up” vectors for human applica- 10. Arber S, Han B, Mendelsohn M, Smith M, Jessell TM, Soc- tion on a large scale will similarly limit application to kanathan S. Requirement for the homeobox gene Hb9 in MND. the consolidation of motor neuron identity. Neuron 1999; 23:659–674. 11. Azzouz M, Hottinger A, Paterna JC, Zurn AD, Aebischer P, FUTURE DIRECTIONS Bueler H. Increased motoneuron survival and improved neuromuscular function in transgenic ALS mice after in- The development of practical gene-based therapy traspinal injection of an adeno-associated virus encoding for MND requires a better understanding of the Bcl-2. Hum Mol Genet 2000; 9:803–811. 12. Azzouz M, Kingsman SM, Mazarakis ND. Lentiviral vectors pathophysiology of these diseases. For example, the for treating and modeling human CNS disorders. J Gene reasons for this selective motor neuron vulnerability Med 2004;6:951–962. may hold important keys to novel neuroprotective 13. Azzouz M, Le T, Ralph GS, Walmsley L, Monani UR, Lee DC, et al. Lentivector-mediated SMN replacement in a mouse strategies. Targeting the early steps of these diseases model of spinal muscular atrophy. J Clin Invest 2004;114: may be the key to preventing motor neuron damage, 1726–1731. instead of correcting damage that has already oc- 14. Azzouz M, Mazarakis N. Non-primate EIAV-based lentiviral vectors as gene delivery system for motor neuron diseases. curred. Toward this end, better diagnostic aids may Curr Gene Ther 2004;4:277–286. be required. Microarray technology may provide di- 15. Azzouz M, Ralph GS, Storkebaum E, Walmsley LE, Mitro- rect information on the genes leading to neurode- phanous KA, Kingsman SM, et al. VEGF delivery with retro- 93 gradely transported lentivector prolongs survival in a mouse generation. Unlocking the specific mechanisms of ALS model. Nature 2004;429:413–417. MND may provide novel therapeutic target genes. 16. Bajocchi G, Feldman SH, Crystal RG, Mastrangeli A. Direct Finally, the co-delivery of therapeutic agents may in vivo gene transfer to ependymal cells in the central ner- 24,59,78,97 vous system using recombinant adenovirus vectors. Nat provide synergistic or additive effects. Re- Genet 1993;3:229–234. gardless of these improvements, variations of the 17. Barbeito LH, Pehar M, Cassina P, Vargas MR, Peluffo H, strategies discussed are likely to be subjected to clin- Viera L, et al. A role for astrocytes in motor neuron loss in ical trials over the next decade. amyotrophic lateral sclerosis. Brain Res Brain Res Rev 2004; 47:263–274. The authors thank the Amyotrophic Lateral Sclerosis Association 18. Bartlett JS, Samulski RJ, McCown TJ. Selective and rapid (ALSA) and the National Institute of Neurological Diseases and uptake of adeno-associated virus type 2 in brain. Hum Gene Stroke (KO8 NS43305) for support. Ther 1998;9:1181–1186. 19. Baumgartner BJ, Shine HD. Targeted transduction of CNS neurons with adenoviral vectors carrying neurotrophic factor genes confers neuroprotection that exceeds the transduced population. J Neurosci 1997;17:6504–6511. REFERENCES 20. Begley DJ. Delivery of therapeutic agents to the central 1. Acsadi G, Anguelov RA, Yang H, Toth G, Thomas R, Jani A, nervous system: the problems and the possibilities. Pharma- et al. Increased survival and function of SOD1 mice after col Ther 2004;104:29–45. glial cell–derived neurotrophic factor gene therapy. Hum 21. Bendotti C, Carri MT. Lessons from models of SOD1-linked Gene Ther 2002;13:1047–1059. familial ALS. Trends Mol Med 2004;10:393–400. 2. Adams J, Cory S. The Bcl-2 protein family: arbiters of cell 22. Benitez JA, Segovia J. Gene therapy targeting in the central survival. Science 1998;281:1322–1326. nervous system. Curr Gene Ther 2003;3:127–145. 3. Aebischer P, Schluep M, Deglon N, Joseph JM, Hirt L, Heyd 23. Berger MM, Kopp N, Vital C, Redl B, Aymard M, Lina B. B, et al. Intrathecal delivery of CNTF using encapsulated Detection and cellular localization of enterovirus RNA se- genetically modified xenogeneic cells in amyotrophic lateral quences in spinal cord of patients with ALS. Neurology sclerosis patients. Nat Med 1996;2:696–699. 2000;54:20–25. 4. Akli S, Caillaud C, Vigne E, Stratford-Perricaudet LD, Poe- 24. Bilak MM, Corse AM, Kuncl RW. Additivity and potentiation naru L, Perricaudet M, et al. Transfer of a foreign gene into of IGF-I and GDNF in the complete rescue of postnatal the brain using adenovirus vectors. Nat Genet 1993;3:224– motor neurons. Amyotroph Lateral Scler Other Motor Neu- 228. ron Disord 2001;2:83–91. 5. Alisky JM, Davidson BL. Gene therapy for amyotrophic lat- 25. Bloch J, Bachoud-Levi AC, Deglon N, Lefaucheur JP, Winkel eral sclerosis and other motor neuron diseases. Hum Gene L, Palfi S, et al. Neuroprotective gene therapy for Hunting- Ther 2000;11:2315–2329. ton’s disease, using polymer-encapsulated cells engineered

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 317 to secrete human ciliary neurotrophic factor: results of a 45. Cassina P, Peluffo H, Pehar M, Martinez-Palma L, Ressia A, phase I study. Hum Gene Ther 2004;15:968–975. Beckman JS, et al. Peroxynitrite triggers a phenotypic trans- 26. Bohn MC. A commentary on glial cell line–derived neuro- formation in spinal cord astrocytes that induces motor neu- trophic factor (GDNF). From a glial secreted molecule to ron apoptosis. J Neurosci Res 2002;67:21–29. gene therapy. Biochem Pharmacol 1999;57:135–142. 46. Cifuentes-Diaz C, Frugier T, Melki J. Spinal muscular atro- 27. Bohn MC, Connor B, Kozlowski DA, Mohajeri MH. Gene phy. Semin Pediatr Neurol 2002;9:145–150. transfer for neuroprotection in animal models of Parkin- 47. Cleveland DW, Rothstein JD. From Charcot to Lou Gehrig: son’s disease and amyotrophic lateral sclerosis. Novartis deciphering selective motor neuron death in ALS. Nat Rev Found Symp 2000;231:70–89. Neurosci 2001;2:806–819. 28. Bohn MC. Motoneurons crave glial cell line–derived neuro- 48. Cleveland JL. A new piece of the ALS puzzle. Nat Genet trophic factor. Exp Neurol 2004;190:263–275. 2003;34:357–358. 29. Boillee S, Cleveland DW. Gene therapy for ALS delivers. 49. Crocker S, Wigle N, Liston P, Thompson C, Lee C, Xu D, et Trends Neurosci 2004;27:235–238. al. NAIP protects the nigrostriatal dopamine pathway in an 30. Borasio GD, Robberecht W, Leigh PN, Emile J, Guiloff RJ, intrastriatal 6-OHDA rat model of Parkinson’s disease. Eur Jerusalem F, et al. A placebo-controlled trial of insulin-like J Neurosci 2001;14:391–400. growth factor-I in amyotrophic lateral sclerosis. European 50. Daly TM. Overview of adeno-associated viral vectors. Meth- ALS/IGF-I Study Group. Neurology 1998;51:583–586. ods Mol Biol 2004;246:157–165. 31. Borasio GD, Sloan R, Pongratz DE. Breaking the news in 51. Davidson B, Allen E, Kozarsky K, Wilson J, Roessler B. A amyotrophic lateral sclerosis. J Neurol Sci 1998;160(suppl model system for in vivo gene transfer into the CNS using an 1):S127–133. adenoviral vector. Nat Genet 1993;3:219–223. 32. Bordet T, Schmalbruch H, Pettmann B, Hagege A, Castel- 52. Davidson BL, Stein CS, Heth JA, Martins I, Kotin RM, Derk- nau-Ptakhine L, Kahn A, et al. Adenoviral cardiotrophin-1 sen TA, et al. Recombinant adeno-associated virus type 2, 4, gene transfer protects pmn mice from progressive motor and 5 vectors: transduction of variant cell types and regions neuronopathy. J Clin Invest 1999;104:1077–1085. in the mammalian central nervous system. Proc Natl Acad 33. Bordet T, Castelnau-Ptakhine L, Fauchereau F, Friocourt G, Sci USA 2000;97:3428–3432. Kahn A, Haase G. Neuronal targeting of cardiotrophin-1 by 53. Davis A, Stachler M, Shi W, Liu J, Boulis N, Bartlett J. coupling with tetanus toxin C fragment. Mol Cell Neurosci Manipulating AAV2 tropism for enhanced delivery of AAV 2001;17:842–854. vectors to the spinal cord. Mol Ther 2005;11(suppl):S46. 34. Bordet T, Lesbordes JC, Rouhani S, Castelnau-Ptakhine L, 54. Dawbarn D, Allen SJ. Neurotrophins and neurodegenera- Schmalbruch H, Haase G, et al. Protective effects of car- tion. Neuropathol Appl Neurobiol 2003;29:211–230. diotrophin-1 adenoviral gene transfer on neuromuscular 55. DiDonato CJ, Parks RJ, Kothary R. Development of a gene degeneration in transgenic ALS mice. Hum Mol Genet 2001; therapy strategy for the restoration of survival motor neuron 10:1925–1933. protein expression: implications for spinal muscular atrophy 35. Boulis NM, Bhatia V, Brindle TI, Holman HT, Krauss DJ, therapy. Hum Gene Ther 2003;14:179–188. Blaivas M, et al. Adenoviral nerve growth factor and beta- 56. Ding H, Schwarz DS, Keene A, el Affar B, Fenton L, Xia X, galactosidase transfer to spinal cord: a behavioral and histo- et al. Selective silencing by RNAi of a dominant allele that logical analysis. J Neurosurg Spine 1999;90:99–108. causes amyotrophic lateral sclerosis. Aging Cell 2003;2:209– 36. Boulis NM, Turner DE, Dice JA, Bhatia V, Feldman EL. 217. Characterization of adenoviral gene expression in spinal 57. Duberley RM, Johnson IP, Anand P, Leigh PN, Cairns cord after remote vector delivery. Neurosurgery 1999;45: NJ. Neurotrophin-3-like immunoreactivity and Trk C expres- 131–137. sion in human spinal motoneurones in amyotrophic lateral 37. Boulis NM, Turner DE, Imperiale MJ, Feldman EL. Neuro- sclerosis. J Neurol Sci 1997;148:33–40. nal survival following remote adenovirus gene delivery. 58. Dupuis L, Muller A, Meininger V, Loefﬂer JP. Molecular J Neurosurg Spine 2002;96:212–219. mechanisms of amyotrophic lateral sclerosis: recent contri- 38. Boulis NM, Noordmans AJ, Song DK, Imperiale MJ, Rubin A, butions from studies in animal models. Rev Neurol (Paris) Leone P, et al. Adeno-associated viral vector gene expression 2004;160:35–43. in the adult rat spinal cord following remote vector delivery. 59. Eisen A, Weber M. Treatment of amyotrophic lateral sclero- Neurobiol Dis 2003;14:535–541. sis. Drugs Aging 1999;14:173–196. 39. Boulis NM, Willmarth NE, Song DK, Feldman EL, Imperiale 60. Federici T, Liu J, Teng Q, Garrity-Moses M, Yang J, Boulis N. MJ. Intraneural colchicine inhibition of adenoviral and ad- In vivo spinal cord uptake and neuronal binding properties eno-associated viral vector remote spinal cord gene delivery. of Tet1: a potential means for enhanced spinal cord AAV Neurosurgery 2003;52:381–387. delivery. Mol Ther 2005;11(suppl):S332. 40. Bruijn LI, Becher MW, Lee MK, Anderson KL, Jenkins NA, 61. Figlewicz DA, Orrell RW. The genetics of motor neuron Copeland NG, et al. ALS-linked SOD1 mutant G85R medi- diseases. Amyotroph Lateral Scler Other Motor Neuron Dis- ates damage to astrocytes and promotes rapidly progressive ord 2003;4:225–231. disease with SOD1-containing inclusions. Neuron 1997;18: 62. Fink DJ, Sternberg LR, Weber PC, Mata M, Goins WF, Glo- 327–338. rioso JC. In vivo expression of beta-galactosidase in hip- 41. Burger C, Gorbatyuk OS, Velardo MJ, Peden CS, Williams P, pocampal neurons by HSV-mediated gene transfer. Hum Zolotukhin S, et al. Recombinant AAV viral vectors Gene Ther 1992;3:11–19. pseudotyped with viral capsids from serotypes 1, 2, and 5 63. Francis JW, Bastia E, Matthews CC, Parks DA, Schwarzschild display differential efﬁciency and cell tropism after delivery MA, Brown RH Jr, et al. Tetanus toxin fragment C as a vector to different regions of the central nervous system. Mol Ther to enhance delivery of proteins to the CNS. Brain Res 2004; 2004;10:302–317. 1011:7–13. 42. Caillaud C, Akli S, Vigne E, Koulakoff A, Perricaudet M, 64. Francis JW, Figueiredo D, van der Spek JC, Ayala LM, Kim Poenaru L, et al. Adenoviral vector as a gene delivery system YS, Remington MP, et al. A survival motor neuron:tetanus into cultured rat neuronal and glial cells. Eur J Neurosci toxin fragment C fusion protein for the targeted delivery of 1993;5:1287–1291. SMN protein to neurons. Brain Res 2004;995:84–96. 43. Camu W. Spinal muscular atrophies in the adult. Rev Neurol 65. Friedlander R, Brown R Jr, Gagliardini V, Wang J, Yuan (Paris) 2004;160:225–228. J. Inhibition of ICE slows ALS in mice. Nature 1997;388:31. 44. Caplen NJ. Gene therapy progress and prospects. Downregu- 66. Frugier T, Nicole S, Cifuentes-Diaz C, Melki J. The molecular lating gene expression: the impact of RNA interference. bases of spinal muscular atrophy. Curr Opin Genet Dev Gene Ther 2004;11:1241–1248. 2002;12:294–298.

318 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 67. Garrity-Moses M, Teng Q, Tanase D, Liu J, Bhat M, Boulis N. microglia/macrophages in amyotrophic lateral sclerosis spi- AAVCMV.Bcl-xLGFP gene delivery to SHSY-5Y neuroblas- nal cord tissue. Ann Neurol 2004;55:221–235. toma cells as a neuroprotective agent in glutamate-induced 87. Holzbaur EL. Motor neurons rely on motor proteins. Trends apoptosis. Mol Ther 2003;7(suppl):S96. Cell Biol 2004;14:233–240. 68. Garrity-Moses M, Teng Q, Liu J, Tanase D, Rezai A, Boulis N. 88. Howland DS, Liu J, She Y, Goad B, Maragakis NJ, Kim B, et Adenoviral XIAP expression protects motor neurons in an in al. Focal loss of the glutamate transporter EAAT2 in a trans- vivo model of ALS. Soc Neurosci Abstr 2004;312–310. genic rat model of SOD1 mutant-mediated amyotrophic 69. Geller AI, Freese A. Infection of cultured central nervous lateral sclerosis (ALS). Proc Natl Acad Sci USA 2002;99: system neurons with a defective herpes simplex virus 1 vector 1604–1609. results in stable expression of Escherichia coli beta-galactosi- 89. Ilzecka J. Increased serum CNTF level in patients with amyo- dase. Proc Natl Acad Sci USA 1990;87:1149–1153. trophic lateral sclerosis. Eur Cytokine Netw 2003;14:192– 70. Gimenez y Ribotta M, Revah F, Pradier L, Loquet I, Mallet J, 194. Privat A. Prevention of motoneuron death by adenovirus- 90. Jablonka S, Sendtner M. Molecular and cellular basis of mediated neurotrophic factors. J Neurosci Res 1997;48:281– spinal muscular atrophy. Amyotroph Lateral Scler Other 285. Motor Neuron Disord 2003;4:144–149. 71. Glass DJ, Yancopoulos GD. The neurotrophins and their 91. Jafari-Schluep HF, Khoris J, Mayeux-Portas V, Hand C, Rou- receptors. Trends Cell Biol 1993;3:262–268. leau G, Camu W. Superoxide dismutase 1 gene abnormali- 72. Glorioso JC, Mata M, Fink DJ. Therapeutic gene transfer to ties in familial amyotrophic lateral sclerosis: phenotype/ the nervous system using viral vectors. J Neurovirol 2003;9: genotype correlations. The French experience and review of 165–172. the literature. Rev Neurol (Paris) 2004;160:44–50. 73. Gong YH, Parsadanian AS, Andreeva A, Snider WD, Elliott 92. Janson C, McPhee S, Bilaniuk L, Haselgrove J, Testaiuti M, JL. Restricted expression of G86R Cu/Zn superoxide dis- Freese A, et al. Clinical protocol. Gene therapy of Canavan mutase in astrocytes results in astrocytosis but does not cause disease: AAV-2 vector for neurosurgical delivery of asparto- motoneuron degeneration. J Neurosci 2000;20:660–665. acylase gene (ASPA) to the human brain. Hum Gene Ther 74. Greenberg DA, Jin K. VEGF and ALS: the luckiest growth 2002;13:1391–1412. factor? Trends Mol Med 2004;10:1–3. 93. Jiang YM, Yamamoto M, Kobayashi Y, Yoshihara T, Liang Y, 75. Guegan C, Vila M, Rosoklija G, Hays AP, Przedborski S. Terao S, et al. Gene expression profile of spinal motor Recruitment of the mitochondrial-dependent apoptotic neurons in sporadic amyotrophic lateral sclerosis. Ann Neu- pathway in amyotrophic lateral sclerosis. J Neurosci 2001;21: rol 2005;57:236–251. 6569–6576. 94. Kahn A, Haase G, Akli S, Guidotti JE. Gene therapy of 76. Guillot S, Azzouz M, Deglon N, Zurn A, Aebischer P. Local neurological diseases. C R Seances Soc Biol Fil 1996;190:9– GDNF expression mediated by lentiviral vector protects fa- 11. cial nerve motoneurons but not spinal motoneurons in 95. Kaspar BK, Erickson D, Schaffer D, Hinh L, Gage FH, Peter- SOD1(G93A) transgenic mice. Neurobiol Dis 2004;16:139– son DA. Targeted retrograde gene delivery for neuronal 149. protection. Mol Ther 2002;5:50–56. 77. Gurney ME, Pu H, Chiu AY, Dal Canto MC, Polchow CY, 96. Kaspar BK, Llado J, Sherkat N, Rothstein JD, Gage FH. Alexander DD, et al. Motor neuron degeneration in mice Retrograde viral delivery of IGF-1 prolongs survival in a that express a human Cu,Zn superoxide dismutase muta- mouse ALS model. Science 2003;301:839–842. tion. Science 1994;264:1772–1775. 78. Haase G, Kennel P, Pettmann B, Vigne E, Akli S, Revah F, et 97. Kato AC, Lindsay RM. Overlapping and additive effects of al. Gene therapy of murine motor neuron disease using neurotrophins and CNTF on cultured human spinal cord adenoviral vectors for neurotrophic factors. Nat Med 1997; neurons. Exp Neurol 1994;130:196–201. 3:429–436. 98. Klein SM, Behrstock S, McHugh J, Hoffmann K, Wallace K, 79. Haase G, Pettmann B, Vigne E, Castelnau-Ptakhine L, Suzuki M, et al. GDNF delivery using human neural progen- Schmalbruch H, Kahn A. Adenovirus-mediated transfer of itor cells in a rat model of ALS. Hum Gene Ther 2005;16: the neurotrophin-3 gene into skeletal muscle of pmn mice: 509–521. therapeutic effects and mechanisms of action. J Neurol Sci 99. Knight A, Carvajal J, Schneider H, Coutelle C, Chamberlain 1998;160(suppl 1):S97–105. S, Fairweather N. Non-viral neuronal gene delivery mediated 80. Haase G, Pettmann B, Bordet T, Villa P, Vigne E, Schmal- by the HC fragment of tetanus toxin. Eur J Biochem 1999; bruch H, et al. Therapeutic benefit of ciliary neurotrophic 259:762–769. factor in progressive motor neuronopathy depends on the 100. Krivickas LS. Amyotrophic lateral sclerosis and other motor route of delivery. Ann Neurol 1999;45:296–304. neuron diseases. Phys Med Rehabil Clin N Am 2003;14:327– 81. Haenggeli C, Kaspar B, Lee H, Gage F, Rothstein J. Thera- 345. peutic viral vector delivery of excitatory amino acid trans- 101. Kugler S, Straten G, Kreppel F, Isenmann S, Liston P, Bahr porters in a mouse model of ALS. Amyotroph Lateral Scler M. The X-linked inhibitor of apoptosis (XIAP) prevents cell Other Motor Neuron Disord 2004;5:55. death in axotomized CNS neurons in vivo. Cell Death Differ 82. Hall ED, Oostveen JA, Gurney ME. Relationship of micro- 2000;7:815–824. glial and astrocytic activation to disease onset and progres- 102. Kugler S, Lingor P, Scholl U, Zolotukhin S, Bahr M. Differ- sion in a transgenic model of familial ALS. Glia 1998;23: ential transgene expression in brain cells in vivo and in vitro 249–256. from AAV-2 vectors with small transcriptional control units. 83. Hauck B, Xiao W. Characterization of tissue tropism deter- Virology 2003;311:89–95. minants of adeno-associated virus type 1. J Virol 2003;77: 103. Lacomblez L, Bensimon G, Leigh PN, Guillet P, Meininger 2768–2774. V. Dose-ranging study of riluzole in amyotrophic lateral 84. Heath PR, Shaw PJ. Update on the glutamatergic neuro- sclerosis. Amyotrophic Lateral Sclerosis/Riluzole Study transmitter system and the role of excitotoxicity in amyotro- Group II. Lancet 1996;347:1425–1431. phic lateral sclerosis. Muscle Nerve 2002;26:438–458. 104. Lacomblez L, Bensimon G, Leigh PN, Guillet P, Powe L, 85. Henderson CE, Phillips HS, Pollock RA, Davies AM, Le- Durrleman S, et al. A confirmatory dose-ranging study of meulle C, Armanini M, et al. GDNF: a potent survival factor riluzole in ALS. ALS/Riluzole Study Group II. Neurology for motoneurons present in peripheral nerve and muscle. 1996;47(suppl):S242–250. Science 1994;266:1062–1064. 105. Lai EC, Felice KJ, Festoff BW, Gawel MJ, Gelinas DF, Kratz R, 86. Henkel JS, Engelhardt JI, Siklos L, Simpson EP, Kim SH, Pan et al. Effect of recombinant human insulin-like growth fac- T, et al. Presence of dendritic cells, MCP-1, and activated tor-I on progression of ALS. A placebo-controlled study. The

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 319 North America ALS/IGF-I Study Group. Neurology 1997;49: 123. Martin LJ. Neuronal death in amyotrophic lateral sclerosis is 1621–1630. apoptosis: possible contribution of a programmed cell death 106. Lambrechts D, Storkebaum E, Carmeliet P. VEGF: necessary mechanism. J Neuropathol Exp Neurol 1999;58:459–471. to prevent motoneuron degeneration, sufficient to treat 124. Martinov VN, Sefland I, Walaas SI, Lomo T, Nja A, Hoover F. ALS? Trends Mol Med 2004;10:275–282. Targeting functional subtypes of spinal motoneurons and 107. LaMonte BH, Wallace KE, Holloway BA, Shelly SS, Ascano J, skeletal muscle fibers in vivo by intramuscular injection of Tokito M, et al. Disruption of dynein/dynactin inhibits ax- adenoviral and adeno-associated viral vectors. Anat Embryol onal transport in motor neurons causing late-onset progres- (Berl) 2002;205:215–221. sive degeneration. Neuron 2002;34:715–727. 125. Matsuoka N, Yukawa H, Ishii K, Hamada H, Akimoto M, 108. Le Gal La Salle G, Robert JJ, Berrard S, Ridoux V, Stratford- Hashimoto N, et al. Adenovirus-mediated gene transfer of Perricaudet LD, Perricaudet M, et al. An adenovirus vector Bcl-xL prevents cell death in primary neuronal culture of the for gene transfer into neurons and glia in the brain. Science rat. Neurosci Lett 1999;270:177–180. 1993;259:988–990. 126. Mazarakis ND, Azzouz M, Rohll JB, Ellard FM, Wilkes FJ, 109. Lee DA, Gross L, Wittrock DA, Windebank AJ. Localization Olsen AL, et al. Rabies virus glycoprotein pseudotyping of and expression of ciliary neurotrophic factor (CNTF) in lentiviral vectors enables retrograde axonal transport and postmortem sciatic nerve from patients with motor neuron access to the nervous system after peripheral delivery. Hum disease and diabetic neuropathy. J Neuropathol Exp Neurol Mol Genet 2001;10:2109–2121. 1996;55:915–923. 127. McCown TJ, Xiao X, Li J, Breese GR, Samulski RJ. Differen- 110. Lesbordes JC, Cifuentes-Diaz C, Miroglio A, Joshi V, Bordet tial and persistent expression patterns of CNS gene transfer T, Kahn A, et al. Therapeutic benefits of cardiotrophin-1 by an adeno-associated virus (AAV) vector. Brain Res 1996; gene transfer in a mouse model of spinal muscular atrophy. 713:99–107. Hum Mol Genet 2003;12:1233–1239. 128. McGeer PL, McGeer EG. Glial cell reactions in neurodegen- 111. Lewis ME, Neff NT, Contreras PC, Stong DB, Oppenheim erative diseases: pathophysiology and therapeutic interven- RW, Grebow PE, et al. Insulin-like growth factor-I: potential tions. Alzheimer Dis Assoc Disord 1998;12(suppl 2):S1–6. for treatment of motor neuronal disorders. Exp Neurol 129. McWhorter ML, Monani UR, Burghes AH, Beattie CE. 1993;124:73–88. Knockdown of the survival motor neuron (Smn) protein in 112. Li M, Ona VO, Guegan C, Chen M, Jackson-Lewis V, An- zebrafish causes defects in motor axon outgrowth and path- drews LJ, et al. Functional role of caspase-1 and caspase-3 in finding. J Cell Biol 2003;162:919–931. an ALS transgenic mouse model. Science 2000;288:335–339. 130. Miles GB, Yohn DC, Wichterle H, Jessell TM, Rafuse VF, 113. Lilley CE, Groutsi F, Han Z, Palmer JA, Anderson PN, Latch- Brownstone RM. Functional properties of motoneurons de- man DS, et al. Multiple immediate-early gene-deficient her- rived from mouse embryonic stem cells. J Neurosci 2004;24: pes simplex virus vectors allowing efficient gene delivery to 7848–7858. neurons in culture and widespread gene delivery to the 131. Millecamps S, Mallet J, Barkats M. Adenoviral retrograde central nervous system in vivo. J Virol 2001;75:4343–4356. gene transfer in motoneurons is greatly enhanced by prior 114. Lindholm D, Dechant G, Heisenberg CP, Thoenen H. Brain- intramuscular inoculation with botulinum toxin. Human derived neurotrophic factor is a survival factor for cultured Gene Therapy 2002;13:225–232. rat cerebellar granule neurons and protects them against 132. Miller RG, Petajan JH, Bryan WW, Armon C, Barohn RJ, glutamate-induced neurotoxicity. Eur J Neurosci 1993;5: Goodpasture JC, et al. A placebo-controlled trial of recom- 1455–1464. binant human ciliary neurotrophic (rhCNTF) factor in 115. Liston P, Fong W, Korneluk R. The inhibitors of apoptosis: amyotrophic lateral sclerosis. rhCNTF ALS Study Group. there is more to life than Bcl2. Oncogene 2003;22:8568– Ann Neurol 1996;39:256–260. 8580. 133. Miller TM, Cleveland DW. Has gene therapy for ALS ar- 116. Liu JK, Teng Q, Garrity-Moses M, Federici T, Tanase D, rived? Nat Med 2003;9:1256–1257. Imperiale MJ, et al. A novel peptide defined through phage 134. Miller TM, Kaspar BK, Kops GJ, Yamanaka K, Christian LJ, display for therapeutic protein and vector neuronal target- Gage FH, et al. Virus-delivered small RNA silencing sustains ing. Neurobiol Dis 2005;19:407–418. strength in amyotrophic lateral sclerosis. Ann Neurol 2005; 117. Lowenstein PR. Immunology of viral-vector-mediated gene 57:773–776. transfer into the brain: an evolutionary and developmental 135. Mitchell JD, Wokke JH, Borasio GD. Recombinant human perspective. Trends Immunol 2002;23:23–30. insulin-like growth factor I (rhIGF-I) for amyotrophic lateral 118. Lowenstein PR, Castro MG. Inflammation and adaptive im- sclerosis/motor neuron disease. Cochrane Database Syst Rev mune responses to adenoviral vectors injected into the 2002:CD002064. brain: peculiarities, mechanisms, and consequences. Gene 136. Mitsumoto H, Klinkosz B, Pioro EP, Tsuzaka K, Ishiyama T, Ther 2003;10:946–954. O’Leary RM, et al. Effects of cardiotrophin-1 (CT-1) in a 119. Lu YY, Wang LJ, Muramatsu S, Ikeguchi K, Fujimoto K, mouse motor neuron disease. Muscle Nerve 2001;24:769– Okada T, et al. Intramuscular injection of AAV-GDNF results 777. in sustained expression of transgenic GDNF, and its delivery 137. Miwa H, Shibata M, Okado H, Hirano S. Tracing axons in to spinal motoneurons by retrograde transport. Neurosci the peripheral nerve using lacZ gene recombinant adenovi- Res 2003;45:33–40. rus and its application to regeneration of the peripheral 120. Magal E, Burnham P, Varon S. Effects of ciliary neuronotro- nerve. J Neuropathol Exp Neurol 2001;60:671–675. phic factor on rat spinal cord neurons in vitro: survival and 138. Mohajeri M, Figlewicz D, Bohn M. Intramuscular grafts of expression of choline acetyltransferase and low-affinity nerve myoblasts genetically modified to secrete glial cell line-de- growth factor receptors. Brain Res Dev Brain Res 1991;63: rived neurotrophic factor prevent motoneuron loss and dis- 141–150. ease progression in a mouse model of familial amyotrophic 121. Maguire-Zeiss KA, Bowers WJ, Federoff HJ. HSV vector– lateral sclerosis. Hum Gene Ther 1999;10:1853–1866. mediated gene delivery to the central nervous system. Curr 139. Murakami T, Nagano I, Hayashi T, Manabe Y, Shoji M, Opin Mol Ther 2001;3:482–490. Setoguchi Y, et al. Impaired retrograde axonal transport of 122. Manabe Y, Nagano I, Gazi MS, Murakami T, Shiote M, Shoji adenovirus-mediated E. coli LacZ gene in the mice carrying M, et al. Adenovirus-mediated gene transfer of glial cell mutant SOD1 gene. Neurosci Lett 2001;308:149–152. line–derived neurotrophic factor prevents motor neuron 140. Mutoh T, Sobue G, Hamano T, Kuriyama M, Hirayama M, loss of transgenic model mice for amyotrophic lateral scle- Yamamoto M, et al. Decreased phosphorylation levels of rosis. Apoptosis 2002;7:329–334. TrkB neurotrophin receptor in the spinal cords from pa-

320 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 tients with amyotrophic lateral sclerosis. Neurochem Res RNA interference retards disease onset and progression in a 2000;25:239–245. mouse model of ALS. Nat Med 2005;11:423–428. 141. Naldini L, Blomer U, Gage FH, Trono D, Verma IM. Effi- 159. Reed J. Bcl-2 family proteins. Oncogene 1998;17:3225–3236. cient transfer, integration, and sustained long-term expres- 160. Regulier E, Pereira de Almeida L, Sommer B, Aebischer P, sion of the transgene in adult rat brains injected with a Deglon N. Dose-dependent neuroprotective effect of ciliary lentiviral vector. Proc Natl Acad Sci USA 1996;93:11382– neurotrophic factor delivered via tetracycline-regulated len- 11388. tiviral vectors in the quinolinic acid rat model of Hunting- 142. Naldini L, Blomer U, Gallay P, Ory D, Mulligan R, Gage FH, ton’s disease. Hum Gene Ther 2002;13:1981–1990. et al. In vivo gene delivery and stable transduction of non- 161. Ripps ME, Huntley GW, Hof PR, Morrison JH, Gordon JW. dividing cells by a lentiviral vector. Science 1996;272:263– Transgenic mice expressing an altered murine superoxide 267. dismutase gene provide an animal model of amyotrophic 143. Natsume A, Mata M, Wolfe D, Oligino T, Goss J, Huang S, et lateral sclerosis. Proc Natl Acad Sci USA 1995;92:689–693. al. Bcl-2 and GDNF delivered by HSV-mediated gene trans- 162. Romero M, Rangappa N, Garry M, Smith G. Functional fer after spinal root avulsion provide a synergistic effect. regeneration of chronically injured sensory afferents into J Neurotrauma 2002;19:61–68. adult spinal cord after neurotrophin gene therapy. J Neuro- 144. Nicole S, Diaz CC, Frugier T, Melki J. Spinal muscular sci 2001;21:8408–8416. atrophy: recent advances and future prospects. Muscle 163. Rosen DR, Siddique T, Patterson D, Figlewicz DA, Sapp P, Nerve 2002;26:4–13. Hentati A, et al. Mutations in Cu/Zn superoxide dismutase 145. Ochs G, Penn RD, York M, Giess R, Beck M, Tonn J, et al. A gene are associated with familial amyotrophic lateral sclero- phase I/II trial of recombinant methionyl human brain sis. Nature 1993;362:59–62. derived neurotrophic factor administered by intrathecal in- 164. Rowland LP. Six important themes in amyotrophic lateral fusion to patients with amyotrophic lateral sclerosis. sclerosis (ALS) research, 1999. J Neurol Sci 2000;180:2–6. Amyotroph Lateral Scler Other Motor Neuron Disord 2000; 165. Rutledge EA, Halbert CL, Russell DW. Infectious clones and 1:201–206. vectors derived from adeno-associated virus (AAV) serotypes 146. Ono S, Imai T, Igarashi A, Shimizu N, Nakagawa H, Hu J. other than AAV type 2. J Virol 1998;72:309–319. Decrease in the ciliary neurotrophic factor of the spinal cord 166. Sagot Y, Tan SA, Baetge E, Schmalbruch H, Kato AC, Aebi- in amyotrophic lateral sclerosis. Eur Neurol 1999;42:163– scher P. Polymer encapsulated cell lines genetically engi- 168. neered to release ciliary neurotrophic factor can slow down 147. Oppenheim RW, Wiese S, Prevette D, Armanini M, Wang S, progressive motor neuronopathy in the mouse. Eur J Neu- Houenou LJ, et al. Cardiotrophin-1, a muscle-derived cyto- rosci 1995;7:1313–1322. kine, is required for the survival of subpopulations of devel- 167. Sahenk Z, Seharaseyon J, Mendell JR, Burghes AH. Gene oping motoneurons. J Neurosci 2001;21:1283–1291. delivery to spinal motor neurons. Brain Res 1993;606:126– 148. Palmer JA, Branston RH, Lilley CE, Robinson MJ, Groutsi F, 129. Smith J, et al. Development and optimization of herpes 168. Scarpini CG, May J, Lachmann RH, Preston CM, Dunnett simplex virus vectors for multiple long-term gene delivery to SB, Torres EM, et al. Latency associated promoter transgene the peripheral nervous system. J Virol 2000;74:5604–5618. expression in the central nervous system after stereotaxic 149. Pasinelli P, Borchelt DR, Houseweart MK, Cleveland DW, delivery of replication-defective HSV-1–based vectors. Gene Brown RH Jr. Caspase-1 is activated in neural cells and tissue Ther 2001;8:1057–1071. with amyotrophic lateral sclerosis-associated mutations in 169. Schiffer D, Fiano V. Astrogliosis in ALS: possible interpreta- copper–zinc superoxide dismutase. Proc Natl Acad Sci USA tions according to pathogenetic hypotheses. Amyotroph Lat- 1998;95:15763–15768. eral Scler Other Motor Neuron Disord 2004;5:22–25. 150. Pehar M, Cassina P, Vargas MR, Castellanos R, Viera L, 170. Schmalbruch H, Haase G. Spinal muscular atrophy: present Beckman JS, et al. Astrocytic production of nerve growth state. Brain Pathol 2001;11:231–247. factor in motor neuron apoptosis: implications for amyotro- 171. Schneider H, Groves M, Muhle C, Reynolds PN, Knight A, phic lateral sclerosis. J Neurochem 2004;89:464–473. Themis M, et al. Retargeting of adenoviral vectors to neu- 151. Perez MC, Hunt SP, Coffin RS, Palmer JA. Comparative rons using the Hc fragment of tetanus toxin. Gene Ther analysis of genomic HSV vectors for gene delivery to motor 2000;7:1584–1592. neurons following peripheral inoculation in vivo. Gene Ther 172. Schrank B, Gotz R, Gunnersen JM, Ure JM, Toyka KV, Smith 2004;11:1023–1032. AG, et al. Inactivation of the survival motor neuron gene, a 152. Poeschla EM. Non-primate lentiviral vectors. Curr Opin Mol candidate gene for human spinal muscular atrophy, leads to Ther 2003;5:529–540. massive cell death in early mouse embryos. Proc Natl Acad 153. Przedborski S. Programmed cell death in amyotrophic lat- Sci USA 1997;94:9920–9925. eral sclerosis: a mechanism of pathogenic and therapeutic 173. Sendtner M, Schmalbruch H, Stockli KA, Carroll P, importance. Neurologist 2004;10:1–7. Kreutzberg GW, Thoenen H. Ciliary neurotrophic factor 154. Przedborski S. Molecular targets for neuroprotection. prevents degeneration of motor neurons in mouse mutant Amyotroph Lateral Scler Other Motor Neuron Disord 2004; progressive motor neuronopathy. Nature 1992;358:502–504. 5(suppl 1):14–18. 174. Shevtsova Z, Malik JM, Michel U, Bahr M, Kugler S. Promot- 155. Puskovic V, Wolfe D, Goss J, Huang S, Mata M, Glorioso JC, ers and serotypes: targeting of adeno-associated virus vectors et al. Prolonged biologically active transgene expression for gene transfer in the rat central nervous system in vitro driven by HSV LAP2 in brain in vivo. Mol Ther 2004;10:67– and in vivo. Exp Physiol 2005;90:53–59. 75. 175. Shi W, Bartlett J. RGD inclusion in VP3 provides adeno- 156. Rabinowitz JE, Rolling F, Li C, Conrath H, Xiao W, Xiao X, associated virus type 2 (AAV2)-based vectors with a heparan et al. Cross-packaging of a single adeno-associated virus sulfate–independent cell entry mechanism. Mol Ther 2003; (AAV) type 2 vector genome into multiple AAV serotypes 7:515–525. enables transduction with broad specificity. J Virol 2002;76: 176. Shimohama S, Tamura Y, Akaike A, Tsukahara T, Ohara O, 791–801. Watanabe S, et al. Brain-derived neurotrophic factor pre- 157. Ralph GS, Radcliffe PA, Day DM, Carthy JM, Leroux MA, treatment exerts a partially protective effect against gluta- Lee DC, et al. Silencing mutant SOD1 using RNAi protects mate-induced neurotoxicity in cultured rat cortical neurons. against neurodegeneration and extends survival in an ALS Neurosci Lett 1993;164:55–58. model. Nat Med 2005;11:429–433. 177. Silani V, Braga M, Cardin V, Scarlato G. The pathogenesis of 158. Raoul C, Abbas-Terki T, Bensadoun JC, Guillot S, Haase G, ALS: implications for treatment strategies. Neurol Neuro- Szulc J, et al. Lentiviral-mediated silencing of SOD1 through chir Pol 2001;35:25–39.

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 321 178. Silani V, Leigh N. Stem therapy for ALS: hope and reality. 197. Veldink JH, Van den Berg LH, Wokke JH. The future of Amyotroph Lateral Scler Other Motor Neuron Disord 2003; motor neuron disease: the challenge is in the genes. J Neu- 4:8–10. rol 2004;251:491–500. 179. Silani V, Cova L, Corbo M, Ciammola A, Polli E. Stem-cell 198. Vincent AM, Mobley BC, Hiller A, Feldman EL. IGF-I pre- therapy for amyotrophic lateral sclerosis. Lancet 2004;364: vents glutamate-induced motor neuron programmed cell 200–202. death. Neurobiol Dis 2004;16:407–416. 180. Soler-Botija C, Ferrer I, Alvarez JL, Baiget M, Tizzano EF. 199. Vukosavic S, Stefanis L, Jackson-Lewis V, Guegan C, Romero Downregulation of Bcl-2 proteins in type I spinal muscular N, Chen C, et al. Delaying caspase activation by Bcl-2: a clue atrophy motor neurons during fetal development. J Neuro- to disease retardation in a transgenic mouse model of amyo- pathol Exp Neurol 2003;62:420–426. trophic lateral sclerosis. J Neurosci 2000;20:9119–9125. 181. St George JA. Gene therapy progress and prospects: adeno- 200. Vyas S, Bechade C, Riveau B, Downward J, Triller A. Involve- viral vectors. Gene Ther 2003;10:1135–1141. ment of survival motor neuron (SMN) protein in cell death. Hum Mol Genet 2002;11:2751–2764. 182. Storkebaum E, Lambrechts D, Carmeliet P. VEGF: once 201. Wang C, Wang CM, Clark KR, Sferra TJ. Recombinant AAV regarded as a specific angiogenic factor, now implicated in serotype 1 transduction efficiency and tropism in the murine neuroprotection. Bioessays 2004;26:943–954. brain. Gene Ther 2003;10:1528–1534. 183. Storkebaum E, Lambrechts D, Dewerchin M, Moreno-Mur- 202. Wang LJ, Lu YY, Muramatsu S, Ikeguchi K, Fujimoto K, ciano MP, Appelmans S, Oh H, et al. Treatment of motoneu- Okada T, et al. Neuroprotective effects of glial cell line– ron degeneration by intracerebroventricular delivery of derived neurotrophic factor mediated by an adeno-associ- VEGF in a rat model of ALS. Nat Neurosci 2005;8:85–92. ated virus vector in a transgenic animal model of amyotro- 184. Strong M, Rosenfeld J. Amyotrophic lateral sclerosis: a re- phic lateral sclerosis. J Neurosci 2002;22:6920–6928. view of current concepts. Amyotroph Lateral Scler Other 203. Wengenack TM, Holasek SS, Montano CM, Gregor D, Cur- Motor Neuron Disord 2003;4:136–143. ran GL, Poduslo JF. Activation of programmed cell death 185. Strong MJ. The basic aspects of therapeutics in amyotrophic markers in ventral horn motor neurons during early pr- lateral sclerosis. Pharmacol Ther 2003;98:379–414. esymptomatic stages of amyotrophic lateral sclerosis in a 186. Svendsen CN, Langston JW. Stem cells for Parkinson disease transgenic mouse model. Brain Res 2004;1027:73–86. and ALS: replacement or protection? Nat Med 2004;10:224– 204. Williamson TL, Cleveland DW. Slowing of axonal transport 225. is a very early event in the toxicity of ALS-linked SOD1 187. Tanase D, Garrity-Moses M, Liu J, Teng Q, Mazarakis N, mutants to motor neurons. Nat Neurosci 1999;2:50–56. Boulis N. Techniques for remote cervical spinal cord deliv- 205. Wisman L, Coenen S, Van Wingerden W, Hol E, Verhaagen ery of neurotherapeutic RabG-EIAV encoded genes. Mol J, Van Muiswinkel F, et al. Lentiviral vectors coding for Ther 2003;7(suppl):S97. EAAT2 to protect motoneurons from glutamate toxicity. Soc 188. Tanase K, Teng Q, Krishnaney AA, Liu JK, Garrity-Moses Neurosci Abstr 2003;96.96. ME, Boulis NM. Cervical spinal cord delivery of a rabies G 206. Wisman LA, van Muiswinkel FL, de Graan PN, Hol EM, Bar protein pseudotyped lentiviral vector in the SOD-1 trans- PR. Cells over-expressing EAAT2 protect motoneurons from genic mouse. J Neurosurg Spine 2004;1:128–136. excitotoxic death in vitro. Neuroreport 2003;14:1967–1970. 189. Tenenbaum L, Chtarto A, Lehtonen E, Velu T, Brotchi J, 207. Wong LF, Azzouz M, Walmsley LE, Askham Z, Wilkes FJ, Levivier M. Recombinant AAV-mediated gene delivery to the Mitrophanous KA, et al. Transduction patterns of central nervous system. J Gene Med 2004;6(suppl 1):S212– pseudotyped lentiviral vectors in the nervous system. Mol 222. Ther 2004;9:101–111. 190. Teng Q, Garrity-Moses M, Liu J, Tanase D, Mazarakis N, 208. Wu P, Xiao W, Conlon T, Hughes J, Agbandje-McKenna M, Walmsely L, et al. EIAV-IGF-I gene transfer to motor Ferkol T, et al. Mutational analysis of the adeno-associated neurons enhances axonal length in vitro. Mol Ther 2004; virus type 2 (AAV2) capsid gene and construction of AAV2 9(suppl):S19. vectors with altered tropism. J Virol 2000;74:8635–8647. 209. Xu D, Crocker S, Doucet J, St-Jean M, Tamai K, Hakim A, et 191. Teng Q, Garrity-Moses M, Federici T, Yang J, Kutner R, al. Elevation of neuronal expression of NAIP reduces isch- Tordo N, et al. Neuronal gene transfer with HIV-1–based emic damage in the rat hippocampus. Nat Med 1997;3:997– lentiviral vectors pseudotyped with Lyssavirus glycoproteins. 1004. Mol Ther 2005;11(suppl):S187. 210. Xu D, Bureau Y, McIntyre DC, Nicholson DW, Liston P, Zhu 192. Thomas CE, Schiedner G, Kochanek S, Castro MG, Lowen- Y, et al. Attenuation of ischemia-induced cellular and behav- stein PR. Peripheral infection with adenovirus causes unex- ioral deficits by X chromosome-linked inhibitor of apoptosis pected long-term brain inflammation in animals injected protein overexpression in the rat hippocampus. J Neurosci intracranially with first-generation, but not with high-capac- 1999;19:5026–5033. ity, adenovirus vectors: toward realistic long-term neurolog- 211. Yamamura J, Kageyama S, Uwano T, Kurokawa M, Imakita ical gene therapy for chronic diseases. Proc Natl Acad Sci M, Shiraki K. Long-term gene expression in the anterior USA 2000;97:7482–7487. horn motor neurons after intramuscular inoculation of a live 193. Thorne RG, Frey WH II. Delivery of neurotrophic factors to herpes simplex virus vector. Gene Ther 2000;7:934–941. the central nervous system: pharmacokinetic considerations. 212. Yamashita S, Mita S, Arima T, Maeda Y, Kimura E, Nishida Y, Clin Pharmacokinet 2001;40:907–946. et al. Bcl-2 expression by retrograde transport of adenoviral 194. Turner BJ, Cheah IK, Macfarlane KJ, Lopes EC, Petratos S, vectors with Cre–loxP recombination system in motor neu- Langford SJ, et al. Antisense peptide nucleic acid–mediated rons of mutant SOD1 transgenic mice. Gene Ther 2001;8: knockdown of the p75 neurotrophin receptor delays motor 977–986. neuron disease in mutant SOD1 transgenic mice. J Neuro- 213. Yamashita S, Mita S, Kato S, Okado H, Ohama E, Uchino M. chem 2003;87:752–763. Effect on motor neuron survival in mutant SOD1 (G93A) 195. Tuszynski MH, Thal L, Pay M, Salmon DP, Bakay R, et al. A transgenic mice by Bcl-2 expression using retrograde axonal phase 1 clinical trial of nerve growth factor gene therapy for transport of adenoviral vectors. Neurosci Lett 2002;328:289– Alzheimer disease. Nat Med 2005;11:551–555. 293. 196. Uemura O, Okada Y, Ando H, Guedj M, Higashijima S, 214. Yamashita S, Mita S, Kato S, Okado H, Ohama E, Uchino M. Shimazaki T, et al. Comparative functional genomics re- Bcl-2 expression using retrograde transport of adenoviral vealed conservation and diversification of three enhancers vectors inhibits cytochrome c release and caspase-1 activa- of the isl1 gene for motor and sensory neuron-specific ex- tion in motor neurons of mutant superoxide dismutase 1 pression. Dev Biol 2005;278:587–606. (G93A) transgenic mice. Neurosci Lett 2003;350:17–20.

322 Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 215. Yokota T, Miyagishi M, Hino T, Matsumura R, Tasinato A, 218. Zhao W, Xie W, Le W, Beers DR, He Y, Henkel JS, et al. Urushitani M, et al. siRNA-based inhibition speciﬁc for mu- Activated microglia initiate motor neuron injury by a nitric tant SOD1 with single nucleotide alternation in familial ALS, oxide and glutamate-mediated mechanism. J Neuropathol compared with ribozyme and DNA enzyme. Biochem Bio- Exp Neurol 2004;63:964–977. phys Res Commun 2004;314:283–291. 219. Zhao Z, Alam S, Oppenheim RW, Prevette DM, Evenson 216. Yu JY, DeRuiter SL, Turner DL. RNA interference by expres- A, Parsadanian A. Overexpression of glial cell line–d- sion of short-interfering RNAs and hairpin RNAs in mam- erived neurotrophic factor in the CNS rescues motoneu- malian cells. Proc Natl Acad Sci USA 2002;99:6047–6052. rons from programmed cell death and promotes their 217. Zaiss AK, Muruve DA. Immune responses to adeno-associ- long-term survival following axotomy. Exp Neurol 2004; ated virus vectors. Curr Gene Ther 2005;5:323–331. 190:356–372.

Motor Neuron Diseases and Gene Therapy MUSCLE & NERVE March 2006 323