63 effects of DNA methylation on AAV integration and transgene expression

63 Effects of DNA Methylation on AAV Integration and Transgene Expression Molecular Therapy Volume 18, Supplement 1, May 2010 Copyright © The American Society of Gene & Cell Therapy S24 AAV VECTORS I[.]

Molecular Therapy Volume 18, Supplement 1, May 2010 Copyright © The American Society of Gene & Cell Therapy S24

Variants of AAV9: Vector Development and

Interrogation of Correlations between Capsid

Structure and Vector Biology

Li Zhong,1 Jun Xie,1 Qing Xie,1 Kim Van Vliet,2 Qiang Wang,3

Mengxin Li,1 Qin Su,1 Ran He,1 Jason Goetzmann,4 Mavis

Mckenna,2 Terry Flotte,1 Guangping Gao.1

1 Gene Therapy Center, University of Massachusetts Medical

School, Worcester, MA; 2 Department of Biochemistry and

Molecular Biology, University of Florida, Gainesville, FL;

3 Vaccine Research Institute, The 3rd Af liated Hospital of Sun

Yat-Sen University, Guangzhou, Guangdong, China; 4 New Iberia

Research Center, University of Louisiana at Lafayette, New Iberia,

LA.

LZ, JX and QX contribute equally to this study

Developments in AAV vectorology resulted in a panel of ef cient

vehicles that were originated from either natural serotypes/variants,

direct evolution or rationale design The vector-based on AAV9 stands

out in transduction of almost all major tissues in rodent models In an

attempt to isolate more natural variants of AAV9, we rescued more

than 50 full length capsid sequences that are closely related to AAV9

from chimpanzee tissues A subset of these novel AAV9 variants that

were retrieved from either cellular DNA by PCR (Csp3 and Csp7)

or RNA by RT-PCR (ClgF1, ClgF4, ClvD8, ClvR7 and ClvR9) were

selected for vector development and evaluation Possible correlations

between capsid structure and vector biology were interrogated

Differences in capsid sequences between AAV9 and these seven novel

variants range from 4 to 6 amino acids Three vectors can be packaged

ef ciently whereas the other four cannot be, suggesting that several

amino acids are critical for AAV9 packaging Those three vectors were

evaluated in C57BL/6 mice for nLacZ and α1-anti-trypsin (AAT)

gene transduction after intravenous (i.v.), intramuscular (i.m.), and

intra-nasal (i.n.) administration A preliminary study in comparison

of nLacZ transduction in liver, skeletal muscle and lung suggested

that Csp3 outperforms AAV9 In the lung, both Csp3 and ClvD8

primarily target alveoli as does AAV9 However, preliminary data

from more quantitative comparisons at early time points using AAT

as a reporter gene demonstrated no signi cant differences in liver

and muscle transduction between AAV9 and Csp3 but a slight lead

by Csp3 in AAT transduction in lung after i.n delivery Additionally,

while ClgF1 vector leads to poor expression of both genes in all target

tissues, the ClvD8 accomplishes decent transduction in muscle and

lung Using newly established crystal structure of AAV9 VP3 as the

model, we compared the capsid structure of Csp3, which differs from

AAV9 in 6 amino acid residues, with that of AAV9 Of these six amino

acids, aa203M is conserved, and although not present in the crystal

structure, should be located inside near the 5 fold pore and may play a

role in AAV transduction Aa259Q is located at the subunit interface of

monomers that form pentamers Amino acid 321Q, which is near the

base of 5-fold pore, is in a highly conserved region of the capsid, as

is aa335A, which are inside the 5-fold channel Aa495Q and aa640M

are located on the surface of 3-fold symmetry axis Further studies are

underway to elucidate the potential correlations between the capsid

structure and viral activity and critical functional domains in the AAV9

capsid that may contribute to its interesting vector biology

Retina Highlights Species-Speci c Differences in AAV-Mediated Photoreceptor Transduction

Claudio Mussolino,1 Francesco Viola,4 Michele Della Corte,5

Settimio Rossi,5 Umberto Di Vicino,1 Simona Neglia,1 Monica Doria,1 Francesco Testa,5 Roberto Giovannoni,6 Manuela Crasta,7

Massimo Giunti,8 Edoardo Villani,4 Marialuisa Lavitrano,6 Maria Laura Bacci,8 Roberto Ratiglia,4 Francesca Simonelli,1,5 Alberto

Auricchio,1,3 Enrico M Surace.1

1 Telethon Institute of Genetics and Medicine (TIGEM), Naples, Italy; 2 S.E.M.M.-European School of Molecular Medicine,

Naples, Italy; 3 Dept of Pediatrics, Federico II University, Naples, Italy; 4 U.O Oculistica, Fondazione IRCCS, Ospedale Maggiore

Policlinico, Mangiagalli e Regina Elena, University of Milan, Milan, Italy; 5 Dept of Ophthalmology, Second University of

Naples, Naples, Italy; 6 Dept of Surgical Sciences, University of Milano-Bicocca, Monza, Italy; 7 Visionvet, Eye Clinic for Animal,

Bologna, Italy; 8 Dept of Veterinary Morphophysiology and Animal Production (DIMORFIPA), University of Bologna, Bologna, Italy.

Recent success in clinical trials supports the use of adeno-associated viral (AAV) vectors for gene therapy of retinal diseases

caused by defects in the retinal pigment epithelium (RPE) In contrast, the ef cacy of AAV-mediated gene transfer to retinal photoreceptors, considered as the major site of inherited retinal diseases, appears less robust In addition, while AAV-mediated RPE transduction appears ef cient independently of the serotype used and species treated,

AAV-mediated photoreceptor gene transfer has not been systematically investigated thus far in large animal models In the present study

we used the porcine retina which has a high cone/rod ratio This feature allows to properly evaluate both cone and rod photoreceptors transduction and compare the transduction characteristics of AAV2/5, 2/7 and 2/8, the three most ef cient AAV vector serotypes for photoreceptor targeting Here we show that each of the three serotypes tested transduces both RPE and photoreceptors In contrast

to previous  ndings in mice, our results suggest that AAV2/5 infects and transduces photoreceptor more ef ciently than AAV2/7 and 2/8 underscoring the relevance of studying AAV transduction properties

in different species The use of the photoreceptor-speci c rhodopsin promoter restricts transgene expression to porcine rods and cones, and results in photoreceptor transduction levels similar to those obtained with the ubiquitous promoters tested Finally, immunological, toxicological and biodistribution studies support the safety of AAV subretinal administration to the large porcine retina The data presented here on AAV-mediated transduction of the cone-enriched porcine retina may impact on the development of gene-based therapies

for rare and common severe photoreceptor diseases

Anti-Microtubule Drugs through RhoA GTPase Activation

Ping-Jie Xiao,1,2 Richard J Samulski.1,3

1 Gene Therapy Center, University of North Carolina at Chapel Hill, Chapel Hill, NC; 2 Cell and Developmental Biology,

University of North Carolina at Chapel Hill, Chapel Hill, NC;

3 Pharmacology, Universityt of North Carolina at Chapel Hill, Chapel Hill, NC.

Adeno-associated virus (AAV) is considered a promising gene therapy vector, since long-term gene expression, low toxicity, and disease correction have been observed in animal models Among all AAV serotypes, AAV2 is the most widely used vector and proven to be safe in numerous Phase I clinical trials However, limited information

is available regarding the molecular bases of host cells that govern the intracellular processing of AAV vectors, including subcellular traf cking, nuclear targeting, uncoating, and episomal persistence

Molecular Therapy Volume 18, Supplement 1, May 2010

Cytoskeletons including microtubules and actin  laments have been shown to facilitate several viruses’ infection, such as adenovirus, HIV and CPV By disrupting microtubules using anti-microtubule (ATM) drugs such as Nocodazole and Vinblastine, we recently observed a several-fold enhancement of AAV2 transduction in both Hela cells and Normal Human Fibroblasts (NHF) Using Aphidicolin (APH), we were able to rule out that this enhancement of AAV2 transduction was cell cycle arrest related The above studies suggested that disruption

of microtubules in the cells can facilitate AAV transduction Use of microtubule stabilizing drug (Taxol) simultaneously further support this hypothesis Previous studies have shown that proteosome inhibitor (PI) such as MG132 enhance AAV transduction Our data showed a synergistic effect on the enhancement of AAV transduction when the cells were simultaneously treated with PI and ATM drugs

Given the fact that disassembly of microtubules activates RhoA, we further tested the effect of RhoA protein on the AAV2 transduction

Our preliminary data showed that increased level of active RhoA proteins enhanced AAV2 transduction and knock-down of RhoA protein reduced AAV2 transduction Based on these results, we hypothesize that activation of RhoA by microtubule disassembly promotes the ef ciency of AAV2 intracellular processing such as subcellular traf cking and nuclear targeting We are starting to explore the intracellular processes of AAV2 and corresponding molecular basis underlying the above  ndings The data collected will advance our knowledge regarding AAV biology, and will provide experimental basis for enhancing AAV2 performance using anti-cancer reagents such as ATM drugs

Vectors Alters Transduction Properties in the Retina

Hilda Petrs-Silva,1 Astra Dinculescu,1 Seok H Min,1 Li Zhong,1

Sergei Zolotukhin,1 Arun Srivastava,1 Alfred Lewin,1 William W

Hauswirth.1

1 Ophthalmology, University of Florida, Gainesville, FL; 2 Pediatric, University of Florida, Gainesville, FL; 3 Molecular genetics and Microbiology, University of Florida, Gainesville, FL.

Purpose: Surface exposed capsid amino acid residues are key

elements in AAV-mediated transduction since they de ne not only cell tropism, but perhaps also intracellular traf cking and hence the onset and intensity of passenger gene expression Recently we showed that single tyrosine to phenylalanine (Y-F) mutations in AAV serotype 2 can greatly improve transduction in retinal ganglion cells Seven tyrosine residues map to the capsid surface in the AAV

Therefore, to more completely analyze the AAV tropism of Y-F mutant vectors, different sets of multiple Y-F mutants were made and tested

in murine retina

Methods: Intravitreal injections of AAV type 2 wild type (WT)

and Y-F capsid mutant vectors with identical small CBA promoters expressing green  uorescent protein (GFP) were performed in adult mice GFP expression was evaluated in retinal whole mounts and in retinal sections by GFP immunohistochemistry 4 weeks after vector treatment

Results: All the AAV2 Y-F mutants have the capacity to transduce

retinal ganglion cells like WT-AAV-2 Triple Y-F mutant (730-500-444) and sextuple Y-F mutant (730-704-700-500-444-272) F both showed enhanced transduction in ganglion cells upon analysis of  at mount whole retinas Analysis of retinal cross sections showed a very distinct patter of transduction through the all cell layers for some mutants For double Y-F mutants (444 -730) and (704-730), pentuple Y-F mutant (730-704-500-444-272), sextuple Y-F mutant (730-704-700-500-444-272) and septuple Y-F mutant (730+704+700+500+444+272+250), GFP expression was concentrated on the ganglion cell layer as seen in WT-AAV The triple Y-F mutant (730-500-444) exhibited a signi cant transduction of both photoreceptors within the inner cell

layer and Muller glia cells The quadruple Y-F mutant (730-500-444-272) exhibited the most diverse range of retinal cell transduction, with signi cant transduction of all retinal cell layers, including the pigmented epithelial cell layer

Conclusions: We have found that introducing multiple Y-F

mutations can alter vector properties in unanticipated but potentially very useful ways Many of the mutant vectors exhibited the ability

to very ef ciently transduce retinal cells and a subset were also able

to penetrate the entire thickness of the retina to transduce retinal cells separated from the site of administration by many cell layers,

a property never previously seen with conventional AAV vectors Particularly the possibility to transduce all retinal cell types in the outer retina without detachment after an intravitreal delivery represents a potentially much safer paradigm for retinal gene therapy, mostly when the genetic defect or degenerative process has left the retina fragile and prone to further damage upon retinal detachment

Studies on Vesicular Traf cking Pathways of AAV Serotypes

Shen Shen,1,2 Xin Ming,3 Aravind Asokan,1,4 Rudy L Juliano.3

1 Gene Therapy Center, University of North Carolina, Chapel Hill, NC; 2 Department of Biochemistry and Biophysics, University

of North Carolina, Chapel Hill, NC; 3 Division of Molecular Pharmaceutics, School of Pharmacy, University of North Carolina, Chapel Hill, NC; 4 Department of Genetics, University of North Carolina, Chapel Hill, NC.

Cytoskeletal proteins and Rab GTPases are involved in a spectrum

of intracellular trafficking events, such as vesicle formation, traf cking, sorting, and recycling Such events are critical regulators

of viral transport within host cells, as demonstrated for hepatitis B and

C, dengue virus, foot-and-mouth disease virus etc Adeno-associated virus type 2 (AAV2) is known to enter cells through clathrin-dependent receptor-mediated endocytosis, following which AAV2 capsids have been shown to travel through late (Rab7) and recycling (Rab11) endosomal compartments in a dose-dependent manner These earlier studies suggest the potential involvement of a wide variety of dynamin and Rab GTPases in the intracellular transport of AAV serotypes In order to test the aforementioned hypotheses, we have studied the effect

of over-expressing wild type/dominant negative dynamin-GFP and different Rab-GFP fusion proteins on the transduction ef ciency and vesicular traf cking of different AAV serotypes Wild type dynamin appears to enhance transduction ef ciency of several AAV serotypes

in vitro Further, dual color image analysis of AAV-mediated RFP gene expression in cells transfected earlier with dominant negative dynamin/Rab-GFP fusion constructs suggests that Rab5/Rab7/Rab11, but not Rab4/Rab9 might play signi cant roles in vesicle transport

of AAV2 Further studies outlining the effect of dynamin and Rab GTPases on transduction ef ciency and vesicular traf cking of other AAV serotypes will also be presented

Integration and Transgene Expression

Diptiman Chanda,1 Jonathan A Hensel,1 Jerome T Higgs,1 Niroop Kaza,1 Rajat Grover,1 Selvarangan Ponnazhagan.1

1 Pathology, University of Alabama, Birmingham, AL.

Site-speci c integration of wild-type adeno-associated virus (AAV)

in the AAVS1 site of human chromosome 19 is well known Induction

of site-speci c integration of recombinant AAV (rAAV) into AAVS1 may result in long-term expression of therapeutic transgene(s) Previous studies have demonstrated that the leading 510 nucleotides

at the 5’ end of AAVS1 sequence are key to AAV integration and containing homology to the Rep binding site (RBS) and the terminal resolution site (TRS) of AAV AAVS1 sequence analysis also indicated

Molecular Therapy Volume 18, Supplement 1, May 2010 Copyright © The American Society of Gene & Cell Therapy S26

the presence of signi cant number of CpG motifs in this region CpG

islands are widely distributed in the mammalian genome and typically

methylated at the cytosine bases Unmethylated CpGs are located

mainly in the 5’ regulatory region of active genes, and addition of

methyl group to the cytosine may lead to silencing of gene expression

Hypermethylation of CpGs in the promoter region of tumor suppressor

genes is common in majority of cancers and activation of such genes

using demethylating agents offer better prognosis Interestingly,

wild-type AAV has been known to have anti-oncogenic effects, in

particular in HPV-related carcinogenesis In the present study we

have examined if accumulation of methyl groups in the AAVS1

sequence has effects on AAV transgene expression and integration

To this end, EBV based shuttle vector p220.2 containing the 8.2 kb

AAVS1 site was in vitro methylated using CpG methylase and methyl

donor S-adenosyl methionine (SAM) and transfected into C18 cells

(derivative of 293 cells expressing EBNA-1) using lipofectamine

reagent Control C18 cells were transfected with unmodi ed

p220.2-AAVS1 plasmid The cells were then infected with wild-type AAV and

low molecular weight DNA was extracted by HIRT method 96 hours

after the infection To determine the effect on integration, a nested

PCR was performed using primers  anking the AAV R-ITR region and

AAVS1 sequence, to compare AAV integration between unmethylated

and methylated AAVS1 Genomic DNA isolated from Detroit-6 cells

(containing integrated AAV sequence) was used as a PCR control

The PCR products were run on 1.5% agarose gel and transferred to

a nylon membrane followed by Southern Hybridization using probes

complementary to AAV and AAVS1 Results indicated a lack of AAV

integration into methylated p220.2-AAVS1, which may have been due

to inhibition of Rep binding to AAVS1-RBS due to CpG methylation

To determine the effect of methylation status on AAV transgene

expression, various human cancer cell lines; prostate cancer, PC3 &

Du145; Colon Cancer, HT-29; breast cancer, MDA-MB-435 were

tested for rAAV transduction in presence of different concentrations

of demethylating agent, 5-Aza-2’-deoxycytidine Results of these

studies indicated a signi cant enhancement of transgene expression

following demethylation of genomic DNA, suggesting the potential

of altering the methylation status of target cells to achieve enhanced

AAV transduction/transgene expression

and Other Organs Following Intravenous

Administration of a Recombinant Porcine AAV

Alexander J Bello,1,2 Allan R Chand,1,2 Monica Doria,3 Kaylie

Tran,1 Mariacarmela Allocca,3 Alberto Auricchio,3 Gary

Kobinger.1,2

1 Special Pathogens, Public Health Agency of Canada, Winnipeg,

MB, Canada; 2 Medical Microbiology, University of Manitoba,

Winnipeg, MB, Canada; 3 Telethon Institute of Genetics and

Medicine, Naples, Italy.

The discovery of Adeno-associated virus (AAV) was first

described as a contaminant in Adenoviral stocks As more AAVs

were discovered from natural sources, each serotype was found

to have differing phenotypic properties such as tissue tropism,

immunogenicity, and transduction ef ciency Ongoing AAV vector

development includes engineering AAV capsids to incorporate

properties from various serotypes which has been described as

‘Directed Evolution’ However, this strategy is limited to the number

of AAVs with similar genetic sequences belonging to similar clades

Divergent AAVs such as AAV5 have been found dif cult to ‘shuf e’

with other existing AAVs One solution to this is to isolate new AAVs

with similar genetic sequences We recently described AAVpo1, a new

AAV serotype isolated from porcine tissues which belongs to the same

clade as AAV5 A recently isolated novel porcine AAV (AAVpo4) was

found to belong to a divergent clade, but presumably with genetic

sequences allowing DNA shuf ing with AAV5 Our primary research

objectives were to isolate new AAV sequences from porcine tissues, produce vectors based on these novel AAVs, and characterise them

in vitro and in vivo We report here the isolation of 2 novel AAVs based on genetic make up and alignments with existing AAVs One

of them, AAVpo4, was used to generate AAV2/po4-LacZ vector by triple transfection which was evaluated in vivo Biodistribution of AAV2/po4-LacZ in intravenous tail-vein injected mice showed highly ef cient transduction of the liver, with increased targeting to various organs in comparison to AAV2/5 DNA shuf ing with AAV5 as well

as assessment of transduction ef ciency in the eye, muscle and lungs following subretinal, intramuscular or intranasal administration will

be presented AAVpo4 can be added as a new candidate for liver gene transfer as well as for other target tissues of interests for different applications including inherited or acquired disorders and vaccine development

for Tailoring the AAV Capsid

Luk H Vandenberghe,1 Ru Xiao,1 Christie L Bell,1 Kim Van Vliet,2 Jing Lu,1 Mavis Agbandje-McKenna,2 James M Wilson.1

1 Gene Therapy Program, Department of Pathology & Laboratory Medicine, University of Pennsylvania, Philadelphia, PA; 2 Center for Structural Biology, Department of Biochemistry and Molecular Biology, University of Florida, Gainesville.

Vectors based on AAV are attractive vehicles for in vivo gene

transfer The many available variants of the AAV capsid, the main determinant of its biology, distinguish themselves in tropism, gene transfer ef ciency, receptor use and interaction with the host immune system These features can either be desired or to be avoided and therefore determine the selection of capsid for a particular application

As the field evolves, more hurdles for safe and efficient gene transfer are elucidated, novel therapeutic targets are de ned, and the requirements of the vector to perform in speci c applications are expanded Here we propose a novel methodology, named Combinatorial Domain Diversi cation (CDD) to engineer AAV particles by uncoupling functional determinants from a single particle structure in order to combine or transfer phenotypes from one AAV capsid to another The generation of engineered capsid variants has been achieved in either one of two ways: directed evolution or rational design Rational design strategies are limited in that they require a full understanding of the structure-function interplay, are high risk and are performed iteratively Biopanning strategies are possibly able

to overcome these concerns yet suffer from signi cant interference

of non-functional diversity within its libraries and limitations in

the available, mostly in vitro, selection systems Our methodology

incorporates the available structural and functional understanding of the AAV particle while advancing the power of purifying selection based on functional transduction, rather than viral replication

Practically, CDD is a PCR-based methodology in which a full length AAV capsid VP1 molecule is assembled from individual building blocks These building blocks are delineated based on functional mapping and structural information of known AAV biology CDD permits certain building blocks, such as conserved domains, to remain untouched, whereas other domains, e.g hypervariable domains, to

be diversi ed Next to this positional control, the diversity itself can

be controlled by varying individual sets of residues within domains and/or modifying their serotype origin with as primary objective

maximizing functional diversity and thereby increasing the ef ciency

of the library

Molecular Therapy Volume 18, Supplement 1, May 2010

Expression and Bioactivity in Fibroblast-Like Synoviocytes from Rheumatoid Arthritis Patients

Caroline J Aalbers,1,2 Scott A Loiler,1,2 Federico Mingozzi,3 Paul-Peter Tak,1,2 Margriet J Vervoordeldonk.1,2

1 Arthrogen B.V., Amsterdam, Netherlands; 2 Division of Clinical immunology and Rheumatology, Academical Medical Center – University of Amsterdam, Amsterdam, Netherlands; 3 Childrens Hospital of Philadelphia, Philadelphia, PA.

Introduction

We are currently developing local intra-articular gene therapy for the treatment of rheumatoid arthritis (RA) using interferon beta (IFNβ)

as therapeutic gene and rAAV5 as vector Therefore, we generated a rAAV5 vector that expresses hIFNβ under control of an in ammation

inducible promoter (rAAV5.NFκB.hIFNβ) Objective

As part of our clinical gene therapy program we are developing an

in vitro transduction protocol to transduce  broblast-like synoviocytes (FLS) obtained from the joints of RA patients or animal species that will potentially be used in the non-clinical pharmaco-tox studies, for the evaluation of transgene expression and bioactivity Bioactivity is measured by analyzing the effect of IFNβ on IL-6, IL-8 and MMP3

To optimize transduction and expression of the transgene in FLS

we tested the addition of doxorubicin and in ammatory stimuli to activate the NFκB promoter at different time points The in uence

of methotrexate (MTX), a widely used antirheumatic drug, on

transduction was investigated Methods

Primary human, monkey, mouse and rat, and non-primary rabbit FLS were transduced with rAAV5.NFκB.hIFNβ (MOI 200.000) To improve transduction ef ciency the proteasome inhibitor doxorubicin (0.4 µM) was added 4 hours post-transduction To activate the NFκB promoter, cells were stimulated with TNFα (1 ng/ml) Two timelines were tested, with stimulation either 4 or 24 hours post-transduction

To evaluate the in uence of MTX, human FLS were incubated in medium with MTX added to the medium pre- and post-transduction

in 3 concentrations (10 nM, 1 µM, 100 µM) For each experiment supernatants were harvested 48 hours after stimulation Human IFNβ and the biological effect on pro-in ammatory cytokine production

and MMP3 were measured by ELISA Results

Human IFNβ production was detectable in all species, with comparable levels in human RA and monkey FLS and the highest production in non-primary rabbit FLS In human FLS the inhibition

of IL-8 and MMP3 secretion (for both P<0.05) by hIFNβ was most pronounced with stimulation 24 hours after transduction The effect on IL-6 production was more variable and did not show a clear reduction

For monkeys FLS a comparable decrease in IL-8 secretion (P<0.01) but not for IL-6 was seen No biological effect was observed after transduction of rabbit and rodent FLS MTX treatment did not show

an effect on hIFNβ production and bioactivity Conclusion

Transduction with rAAV5.NFκB.hIFNβ results in signi cant hIFNβ expression in FLS of all species The transgene is bioactive in human and monkey FLS This supports the use of non-human primates

in the non-clinical pharmaco-tox program In rabbits and rodents no bioactivity was observed of hIFNβ on pro-in ammatory cytokine and MMP3 production MTX did not in uence hIFNβ production

or its bioactivity, indicating that MTX can be used by RA patients included in a clinical trial with rAAV5

the In Vivo Transduction Ef ciency of Adeno-Associated Virus Type 2

Gilles Moulay,1 Sylvie Boutin,1 Carole Masurier,1 Daniel Scherman,2 Antoine Kichler.1,2

1 Genethon, BP60, Evry Cedex, France; 2 CNRS UMR 8151 – U1022 Inserm, Unité de Pharmacologie Chimique et Génétique et d’Imagerie, Université Paris Descartes, Chimie Paristech, Paris Cedex, France.

Background The success of muscular dystrophy gene therapy

requires widespread and stable gene delivery In this context, adeno-associated virus (AAV) has attracted great attention as an optimal vehicle for body-wide gene delivery However, for the successful treatment of a disease such as Duchenne muscular dystrophy large amounts of recombinant vector are essential Injection of very high doses of viral vectors not only raises questions about the technical feasibility of the large scale production but also about the overall safety of the approach One way to overcome both problems would

be to  nd strategies able to increase the transduction ef ciency in vivo In the present work, we investigated whether polymers can act as adjuvants to increase the ef ciency of AAV-2 vector after

systemic injection Methods Our strategy consists in the injection

of cationic or anionic polymers before intravenous administration

of Balb/c mice with AAV-2 encoding the reporter gene mSeAP (a murine secreted alkaline phosphatase) The transduction ef cacy was followed by quantifying the mSeAP in serum at different time points Vector biodistribution was evaluated in different tissues by quanti cation of the mSeAP protein as well as by real time PCR

of the recombinant viral genomes Results The pre-injection of a

cationic polymer resulted in a 4 to 12-fold increase of seric mSeAP levels Histochemical analysis showed the appearance of mSeAP positive muscle  bers with this strategy when no staining was visible

in control AAV-2 injected mice The PCR data con rmed an overall increased tissue transduction by AAV-2 Pre-injection of an anionic polymer resulted in a 2-fold increase of mSeAP expression in serum Interestingly, while this latter increase is moderate this strategy permitted to signi cantly reduce the neutralizing antibody titer

raised against the AAV-2 capsid Conclusion Our results show that

strategies of pre-injection of polymers can be used either to improve the overall transduction of systemically administered AAV-2 or to reduce the humoral response against the capsid proteins

Doses Using AAV-Protein Phosphotase-5 Helper Virus

Brandon K Sack,1 Jayandharan R Giridhara,1 Zhong Li,1 Herzog

W Roland,1 Srivastava Arun.1

1 Pediatrics, Division of Cellular and Molecular Therapy, University of Florida, Gainesville, FL.

One major obstacle for single-stranded vector transduction is the need convert the single-stranded DNA to double-stranded DNA before transcription Our work and the work of others has shown that

a protein, FKBP52, is a principle inhibitor of this rate limiting step Dephosphorylation of this protein by phosphotases such as T cell protein tyrosine phosphatase (TC-PTP) and protein phosphatase-5 (PP5) can augment gene therapy Our previous studies have shown that co-injection of conventional single-stranded adeno-associated virus 2 (ssAAV2) vectors carrying the enhanced green  uorescent protein (EGFP) gene with self-complementary (sc) AAV2-TC-PTP and scAAV2-PP5 vectors resulted in a ∼16- fold increase in EGFP expression in murine hepatocytes in vivo Here, we optimized this strategy to achieve transgene expression at reduced vector/helper virus doses by using hepatocyte-speci c promoters, tyrosine-mutant AAV2 capsids, and more ef cient AAV serotypes For experiments