287 fusion of either gap 43 or tau to the clostridial tetanus toxin light chain improves synaptic inhibition in the rat spinal cord following adenoviral vector gene delivery

287 Fusion of Either Gap 43 or Tau to the Clostridial Tetanus Toxin Light Chain Improves Synaptic Inhibition in the Rat Spinal Cord Following Adenoviral Vector Gene Delivery Molecular Therapy Volume 2[.]

Molecular Therapy Volume 21, Supplement 1, May 2013

NEUROLOGICAL THERAPY: NEURODEGENERATION AND IMAGING

are assessing HTT mRNA expression in other regions of the brain

and in peripheral tissues Vascular AAV2/9-mi2.1-GFP treatment

signifi cantly prevented weight loss in N171-82Q mice compared

to controls (p<0.05) Motor assays indicated no difference between

groups (p>0.05) Together, these results demonstrate that AAV2/9 is

an exciting gene therapy tool that can be used to deliver HD-specifi c

RNAi constructs to many regions of the brain and peripheral tissues

following a single injection to the vasculature

285 Transcriptional Targeting of HSV

Vector-Mediated Transgene Expression to Specifi c

Sensory Neuron Subtypes

Mark F Doyal,1 Mingdi Zhang,1 George Huang,1 Justus Cohen,1

Nicole Scheff,2 Michael Gold,2 William F Goins,1 Joseph C

Glorioso.1

1 Department of Microbiology and Molecular Genetics, University

of Pittsburgh School of Medicine, Pittsburgh, PA; 2 Department

of Anesthesiology, University of Pittsburgh School of Medicine,

Pittsburgh, PA.

Peripheral injection of replication defective HSV-1 vectors results

in vector uptake at sensory nerve terminals and retrograde delivery

of the genome to the neuronal nuclei within dorsal root (DRG)

and trigeminal ganglia (TG) These genomes enter a latency-like

state and can serve as a persistent platform for expression of

anti-nociceptive products that reduce pain signaling in animal models

of acute and chronic pain We have used the transiently active

HCMV and persistently active viral latency (LAT) promoter to drive

transgene expression; however neither promoter has specifi city for

subpopulations of sensory neurons which mediate the development

of chronic pain Furthermore, in the case of the CMV promoter,

nonspecifi c expression may result in unwanted off-target effects

Here we describe the use of neuronal promoters to target transgene

expression to specifi c sensory neuron subpopulations Using these

transcriptionally targeted vectors we hope to identify sensory neuron

subtypes that contribute to chronic pain signaling and identify those

populations of cells where disruption of the pain response will be most

effective In a proof of principle experiment we tested the promoters

of several neuronal subtype marker genes to target the expression of

the fl uorescent marker mCherry to the neuronal population in which

the promoter is endogenously active Primary rat sensory neurons

were transduced with HSV vectors expressing mCherry under the

control of the promoters for the heavy neurofi lament (highly expressed

in A fi bers), calcitonin gene related peptide (CGRP, expressed

in a population of nociceptive, peptidergic C fibers), transient

receptor potential vanilloid 1 (TRPV1, heat sensitive nociceptors),

neuropeptide Y (upregulated in sensory neurons after nerve injury), or

CMV and HSV LAT promoter controls Promoter targeting effi ciency

was determined by immunofl uorescence using antibodies to the

targeted neuronal marker within a larger population of pan-neuronal

antibody NeuN reactive neurons In vitro experiments in cultured

rat DRG and TG neurons have shown; (i) the CMV promoter drives

transgene expression in both neuronal and non-neuronal cells, whereas

each of the neuronal promoters tested as well as the LAT promoter

restrict expression to neuronal cells only; and (ii) expression from

neuronal promoters is targeted to the appropriate neuronal population

with a high degree of specifi city (70-80%) and sensitivity (80-90%) In

vivo studies to evaluate vector transgene performance by injection of

vectors into the hindpaw of rats corroborated our in vitro model with

neuronal subtype-specifi c expression from neuronal promoters and

nonspecifi c expression from the CMV promoter This will ultimately

be used to target analgesic transgene products in order to precisely

determine the sensory populations contributing to a pain phenotype,

and generate therapeutics with minimal off-target effects (e.g on

non-nociceptive neurons)

Schwannoma Regression

Mehran Taherian,1 Shilpa Prabhakar,2,5 Giulia Fulci,3 Miguel Sena-Esteves,4 Xandra O Breakefi eld,2,5 Gary J Brenner.1

1 Anesthesiology, Massachusetts General Hospital, Boston, MA;

2 Neurology and Radiology, Massachusetts General Hospital, Boston, MA; 3 Neurosurgery, Massachusetts General Hospital, Boston, MA; 4 Neurology, U Mass Medical School, Worcester, MA;

5 Program in Neuroscience, Harvard Medical School, Boston, MA.

Schwannoma tumors are composed of Schwann-lineage cells and form along peripheral, spinal and cranial nerves These tumors can cause pain, sensory/motor dysfunction, and death through compression of peripheral nerves, the spinal cord, and/or the brain stem We have previously shown gene therapy for these tumors using a xenograft model in which immortalized human schwannoma cells (HEI-193) expressing a fl uorescent protein and luciferase are implanted in the sciatic nerve of nude mice Our strategy has been

to express the apoptotic protein, caspase-1 (ICE, IL1-converting enzyme) under the P0 promoter, selectively active in Schwann cells during development, using an adeno-associated vector 1 (AAV1) injected directly into the tumors Caspase-3 is the most common pro-apoptotic target for therapeutic modulation However, unlike caspase-3, caspase-1 has a strong pro-infl ammatory component in addition to induction of apoptosis Caspase-1 activates pro-IL-1b and pro-IL-18, which in turn trigger immune responses mediated by neutrophils and monocytes, or NK cells; it has been associated with both innate and adaptive immunity In addition, caspase-1 is activated

by several chemotherapeutic drugs and can sensitize tumor cells to chemotherapy and radiation Our recent data indicate that AAV1-P0-ICE not only induces apoptosis of infected caspase-1-expressing HEI-193 schwannoma cells, but also kills non-infected tumor cells through a bystander effect mediated by both tumor- and host-specifi c mechanisms Supernatants derived from HEI-193 cells transfected with AAV1-P0-ICE DNA kill non-transfected tumor cells in vitro Similarly, peripheral nerve implantation of a 1:1 mixture of AAV-P0-ICE DNA-transfected and non-transfected HEI-193 cells prevented tumor growth This bystander killing effect induced by AAV1-P0-ICE was also observed when established intra-sciatic HEI-193 tumors were injected with AAV1-P0-ICE transfected cells Interestingly, intra-sciatic implantation of HEI-193 cells 2 weeks after injection at the same location of AAV1-P0-ICE DNA-transfected cells, i.e long after all of the originally implanted caspase-1expressing HEI-193 cells should be dead, also prevents formation of a tumor These data suggest that caspase-1-induced death of HEI-193 cells in vivo leads to local changes in the region of the tumor capable of preventing formation

of a subsequent tumor This investigation of AAV1-P0-ICE induced bystander killing of tumor cells has the potential to generate new therapeutics for the neurofi bromatoses, as well as, for other benign neoplasms Not only can caspase-1 induce apoptosis, but the resulting tumor killing and associated activation of host immune responses may generate a vaccination effect that could control the development and growth of subsequent schwannomas

287 Fusion of Either Gap-43 or Tau to the Clostridial Tetanus Toxin Light Chain Improves Synaptic Inhibition in the Rat Spinal Cord Following Adenoviral Vector Gene Delivery

Eleanor M Donnelly,1 Chalonda R Handy,2 Marie Kim,1 Jeremiah Huang,1 Thais Federici,1 Nicholas M Boulis.1

1 Neurosurgery, Emory University, Atlanta, GA; 2 Nationwide Children’s Hospital, Columbus, OH.

Clostridial bacterial toxins, are assuming increased importance as therapeutic agents for modulating neural transmission in refractory conditions The similarity between the functional components of

Molecular Therapy Volume 21, Supplement 1, May 2013 Copyright © The American Society of Gene & Cell Therapy

S110

NEUROLOGICAL THERAPY: NEURODEGENERATION AND IMAGING

the Clostridial toxins presents an opportunity for the development

of therapies involving the use of the tetanus light chain (LC) As

an alternative to the commonly used botulinum treatment approach

which requires repeated protein injections, viral vector delivery of

LC can give stable LC expression Indeed, in earlier studies we have

demonstrated effective synaptic inhibition using adenoviral

directed-LC expression directed-LC proteases targets proteins in the SNARE complex

responsible for synaptic vesicle docking at the synaptic terminal In

order to being about a more potent inhibition we are investigating

methods of transporting the LC protein from the cell body to synaptic

terminal via fusing the LC protein with sequences of proteins which

are natively transported to the synaptic terminal We have evaluated

the benefi ts of fusing the LC with either the 3’ untranslated region of

the microtubule-associated protein tau or the N-terminus region of

the cytoplasmic protein neuromodulin (Gap-43) for axonal targeting

For our research purposes, four recombinant vectors were produced:

Ad.LC, Ad.LC.Gap-43, Ad.LC.tau and Ad-Green Fluorescent

Protein (GFP) Each vector contained the cytomegalovirus (CMV)

promoter and was constructed to express GFP In all studies, adult

rats received ipsilateral spinal cord injections into the lumbar region

and motor function was assessed by measures of gait and strength We

found that all LC expressing vectors resulted in inhibition of motor

activity without pathological effects as assessed by cresyl violent

staining Importantly, rats treated with Ad.GFP did not demonstrate

changes in locomotor activity at any time during the course of the

study Whereas Ad.LC treated animals showed impairments in

gait and ipsilateral hindlimb strength approximately 5 days after

vector delivery; interestingly, animals that received Ad.LC.Gap-43

demonstrated declines in ipsilateral motor function 1-2 days post

vector administration Moreover, Ad.LC.tau vector delivery resulted

in ipsilateral locomotor inhibition occurring 3-4 days post vector

administration This data demonstrates that gene transfer of LC can be

achieved without cell loss and axonal targeting can result in enhanced

transgene potency and activity We are currently determining the

relative level of GFP expression in each animal group using protein

assays such as immunohistochemisty and western blotting techniques

into the Midbrain of Macaques Results in a Safe

and Effi cient Axonal Transport of AADC into the

Striatum

Krystof S Bankiewicz,1 Adrian P Kells,1 Lluis Samaranch,1 Waldy

San Sebastian,1 Nitasha Sharma,1 John Bringas,1 Phillip Pivirotto,1

John Forsayeth.1

1 Department of Neurological Surgery, University of California,

San Francisco, CA.

Aromatic L-amino acid decarboxylase (AADC) defi ciency is a

rare and debilitating recessive genetic disorder in which mutations

in the AADC gene result in a reduced activity of the AADC enzyme

and lead to an inability to synthesize catecholamines [norepinephrine

(NE), epinephrine (EP), dopamine (DA)] and serotonin As a result,

affl icted children suffer severe movement and affective disorders with

a consequently curtailed and diffi cult life We plan on using

AAV2-hAADC gene transfer to restore AADC levels in the brains of these

affected children An IND-enabling safety study was performed in 18

non-human prmates that evaluated 2 dose levels of AAV2-hAADC

and a saline (control) group Infusate was delivered bilaterally into

the midbrain under MR-guidance using our clinical brain delivery

platform The MRI tracer, gadolinium, was used to visualize AAV

vector administration in real time into the substantia nigra and ventral

tegmental area In all cases, the precise, complete and refl ux-free

delivery of AAV was observed with no clinical adverse effects In

animals that were euthanized at 4 weeks, histological analysis showed

effi cient anterograde axonal transport of AADC to the caudate and

putamen Animals scheduled for euthanasia at 3 and 9 months will

be processed and the results will be presented This data supports the safety and effi cacy of MR-guided real time delivery of AAV2-hAADC into the midbrain of juvenile monkeys and justifi es the use of axonal transport to achieve widespread delivery of AADC in the striatum of AADC-defi cient children

AAV-Mediated D2R80A Reporter Gene Expression

in Live Animal Brains

Sea Young Yoon,1 Datta Ponde,2 Harish Poptani,2 Jessica Bagel,3

Patricia A O’Donnell,3 Charles H Vite,3 John H Wolfe.1,3

1 Research Institute of the Children’s Hospital of Philadelphia, Philadelphia; 2 Department of Radiology, Perelman School

of Medicine, University of Pennsylvania, Philadelphia; 3 W.F Goodman Center for Comparative Medical Genetics, School of Veterinary Medicine, University of Pennsylvania, Philadelphia.

In vivo imaging of vector gene expression is crucial for evaluating the efficacy and response of gene therapy, particularly when traditional methods for evaluating gene expression that require invasive tissue sampling would not be ideal or provide only limited information A positron emission tomography (PET)-based reporter gene imaging system enables noninvasive, repetitive monitoring of transgene expression levels and spatial distribution in the living brain following vector gene delivery Several PET reporter genes have been developed for monitoring gene activity D2R80A is a dopamine-2 receptor mutant that binds radioactive analogs of dopamine but has been inactivated for intracellular signaling (Liang et al., Gene Ther 2001); and has been used to monitor peripheral organ transduction Since radioactive dopamine analogs cross the blood brain barrier, D2R80A can be used to monitor the distribution, magnitude and duration of AAV-mediated expression in the mouse brain The dopamine analogs bind to cells in the striatum where D2R is normally expressed, providing a well defi ned internal control The PET-imaging signals correlated well with the reporter gene expression levels and distribution as assessed post-mortem by immunofl uorescence, in situ hybridization, and real time PCR In the current study, two separate AAV1 vectors, one to express D2R80A and the other to express a therapeutic gene, feline -mannosidase were constructed under the control of the human GUSB gene promoter Both vectors were co-injected in the signifi cantly larger cat brain into two injection tracks unilaterally into the cortex, hippocampus and thalamus, and at 9,

16, 24 weeks post-injection, in vivo PET imaging was performed for longitudinal monitoring following intravenous administration of [18F]-fallypride Reporter gene activity (bound to [18F]-fallypride) was quantifi ed in the injected brain regions relative to the striatal binding No signifi cant decrease in signal was observed over time indicating continued D2R80A production and no adverse immune response against the reporter gene Post-mortem brain analyses are underway to correlate the PET signal in the cats with the expression of both vectors in tissues by immunofl uorescence, in situ hybridization, and real time PCR to assess D2R80A as a surrogate marker for the therapeutic gene [18F]-fallypride PET imaging of D2R80A provides

a noninvasive, quantitative and sensitive method to monitor the vector gene expression, which could facilitate the development of enhanced vector delivery and potentially be used clinically in gene therapy of neurogenetic diseases Funding source: NIH R01-DK063973 and R01-NS038690