75 targeting of angiogenic endothelial cells using a new AAV 2 insertion site

75 Targeting of Angiogenic Endothelial Cells Using a New AAV 2 Insertion Site liver targeted approach,bloodcellscontainaround0 0I copiesofRV per cell at 5 years after transduction This low level hemat[.]

liver-targeted approach,bloodcells containaround 0.0I copiesofRV per cell at years after transduction This low-level hematopoietic marking and expression may prove to be an important component ofthe therapy However, hematopoietic cell transduction also presents an additional risk for insertional mutagenesis in a population of rapidly dividing cells As part of a long-term safety assessment, blood from RV-treated MPS VII dogs that received direct injections ofRV to 5.5 years previously was analyzed for RV insertion sites using ligation amplification-mediated (LAM) PCR There were relatively few insertionsites detected for each dog A total ofeleven non-artifactual genomic sequences were cloned from five dogs The results of a BLAT search of the UCSC May Z005 Dog Assembly (http://genome.ucsc.edu/cgi-bin/hgBlat)allowed assignmentofnine of these sequences to specific locations in the dog genome Of the genes closest to the nine identifiedsites, the mouse homolog of five of'these were presentin the Retroviral TaggedCancerGene Database (RTCGD(mm8) http://rtcgd.abcc.ncifcr[gov/) Assuming similar gene structures in dog as in human, the locations with respect to genes could be determined, Twowere located within kb upstream ofthe first known exon, three were within intron I, two were within other introns, one was 14 kb downstream of the nearest gene, and the last was at least 60 kb from the nearest annotated gene Three of these dogs had also been analyzed three years previously In that analysis, six unique sequences were identified Two of these corresponded to genes in the RTCGD Only one site, within the last exon of the HMGA2 gene, was detected in both analyses of one of the dogs 73 Identification of Genetic Insulator Elements for Safe Gene Therapy Viral Integrating Vectors Cecile Bauche,' Armelle Gaussin,' Julien De Royer,' Jean Francois Mouscadet,' Christian Auclair; Nicholas Merrnod,' Odile Y Cohcn-Hagucnaucr,'-' I Laboratoire de Biotechnologie et de Pharmacologie Genetique Appliquees, Ecole Normale Superieure, Cachan, France ; 21nstitute ofBiotechnology, University ofLausanne, Lausanne, Switzerland; JDep artment ofMedical Oncology, Hopital SaintLouis, Paris, France Insertional mutagenesis has been demonstrated following the integrationofgene transfervectors that includestrong enhancer/promoters Conversely, the insulator approach can also be investigated as a way to protectthe endogenousgenomic sequences/environment from the risk ofdysregulation resulting from integrating vectors In addition, recent reports have shown that genetic insulatorscurrently under usc, like the HS4 1,2 kp fragment or the 250 bp core clements cannot wholly insulate the genetic environment from potent viral regulatoryelements In this event, there will be no garantec that sideeffects will not reproduce followingclonal selection and expansion, In order to identify short genetic elements capable of insulating the strongest viral regulatory elements from the Fr-MuLV FB29 strain enhancer which are capable of both preventing insertional mutagenesis and the exctinction of transgene expression we have established a standard screening procedure This assay consists of a series of plasmids containing two reporter genes: one mimicking a therapeutic gene under thc control of strong viral enhancer/promoter, and the other one standing for an endogenous gene close to the chromosomalvector integration site, Potentialinsulatorelements interposed between the viral enhancer and the "chromosomal" gene promoter are expected to shield the latter gene from the influenceof the neighboringvector elements The setting up and assay validation make use of the chicken beta-globin 5'HS4 elements, whereas repeats of short genetic elements having more potent insulatoractivity are being screened and retained New insulators are challenged with both the full FOCHA-LrR and the Fr-MuLVenhancer alone SelfS30 inactivating gammaretrovirus-based vectors have been designed by other groups; the virus titers we have obtained from SIN Fr-MuLV FB29 derived, FOCHAvector, i.e., 106pfu/ml (Cohen-Haguenauer, unpublisheddata) are not standard.Aseries ofinsulated gamma- and lenti- SIN-retrovirus constructs have thus been engineered using two combinations of the shortest active stretches, respectively of 268 bp and 157 bp long in total, substituting the U3 LTRenhancer in full The lenti backbone is derived from Luigi Naldini's pSIN18 (with kind permission) Data form these vectors and related comparisons will be presented In adddition, we aim at combining the usc of insulated lentivectors and targeting their integration to heterochromatin Acknowledgement: EC FP6-NoE CLiNIGENE LSHB-CT-Z006-0 18933 (the European network of excellence for the advancement of clinical gene transfer and therapy) 74 Making Retroviral Gene Therapy Safer: Prevention of Retrovirus-Mediated Activation of Cellular Genes near the Integration Site by Engineering of the LTR Youngtae Hong, I Nam-Kyung Yoon, I Sujeong Kim, ' Sunyoung Kim.! Joong Gon Kim,' Jung Woo Rhim,' Hyoung Jin Kang,' Karim Lee,' Jiwon Jang.? I Department ofResearch and Development Viro1l4ed, Seoul, Republic ofKorea; 2Department ofBiological Sciences, Seoul National University, Seoul, Republic ofKorea; 'Depanment of Pediatrics Seoul National University Hospital, Seoul, Republic ofKorea The usc of retroviraI vectors has recently demonstrated its actual clinical benefit in a few inherited diseases However, the leukemia cases foundafter the x-scm gene therapy trial has raised the safety concern of the insertional mutagenesis inherent to the biology of the retrovirus.Although the retrovirus has long been known to integrate into the host chromosome, and thus have the potential to activate the nearby gene, there has been no convenient method of studying or assaying such a cis-activation phenomenon Here we report an in vitro assay system in which the effect ofretroviraI integration on the expression of the neighboring gene can be studied Using this assay, we found that the full-length LTR could indeed activate the neighboring gene expression from a distance and the magnitude of its activation was highly increased when this LTR was placed in the vicinity of the transcription start site of the gene, while the truncated LTR exerted little influence This system might provide a useful tool for selecting the appropriate vector structure as well as studying the molecular mechanism underlying the cis-activation by the viral LTR AAV PRODUCTION AND MODIFICATION 75 Targeting of Angiogenic Endothelial Cells Using a New AAV-2 Insertion Site Jorge Boucas,' Kerstin Lux; Anke Huber,' Sibille Hummc.P Michael Hallek.t-' Luca Perabo,' Hildegard Buning.l -' 'Clinic Ifor Internal Medicine University ofCologne Cologne Germany; 2Centerfor Molecular Medicine Cologne University of Cologne, Cologne , Germany The insertion of small peptides into the amino acid position 587 of the adcno-associatcd virus (AAV-2) capsid opened the field of AAV targeting Since then, recombinant AAV (rAAV) targeting vectors with 587 insertions have been proven to yield increased transduction efficiency in vitro and in vivo and to transduce their target cell via the new insertion receptor interaction Furthermore, combinatorial approaches using AAV display libraries and high throughput-screenings simplified the creation of targeting vectors with desired transduction abilities In this report we demonstrate Molecular Therapy Volume15.Supplement Iã \by 2IlQ7 Cop)Tight â The Amcricm Society o r Gene Therapy how, by analysis of the now availableAAV-2atomic structure, anew position was found and improved to become a valid alternative to 587 By inserting a known peptide - RGD4C - into this new site we were able to show a 3-fold increase in receptor binding capability when compared to the same insertion at position 587 Increased transduction was observed for target cells as e.g, primary angiogenic human umbilical vein endothelial cells (HUVEC) Decreased transduction efficiency for non-target cells was observed when combining insertions with mutations of residues responsible for primary receptor binding Specificity of insertion.receptor interaction was also proven by competition studies using soluble peptides Furthermore, a detailed analysis on different combination of insertion sites, peptides, further capsid modificationsand target cells, allowed us to define optimal transduction conditions and efficiencies According to the analysis of the 3D-structure, peptides inserted at the new site arc less prone to interfere with other viral functions required for efficientcell-transduction By leaving R585 and R588 (main residues responsible for AAV2's primary receptor binding) untouched the new insertion site can be used to broaden AAV2's tropism On the other hand, it can be used to redirect AAV2's tropism by combining peptide insertionwith R585Aand R588Amutations ablatingprimary receptorbinding (heparansulfate proteoglyean, HSPG) Maintaining HSPG binding still allows purification and concentration of such AAV targeting vectors by heparin affinity chromatography, and is especially useful for ex vivo applications Abolishing this binding is of particular use for systemic applications since HSPG dependent retention in liver and spleen can be avoided, thus increasing the in vivo targeting ability of the respective vectors In summary, these results expand the spectrum of targeting technologies for AAV-2 based vectors and offer novel alternatives to address human gene transfer related issues 76 In Vivo Selection of Targeted Gene Transfer Vectors from AAV Random Peptide Display Libraries Ying Ying,I Mueller Oliver,l Waterkamp Daniel,' Kleinschmidt A Juergen.' I Tumorvirologie, Deutsches Krebsforschungszentrum, Heidelberg, Germany; llnnere Medizin, Universitaet Heidelberg, Heidelberg, Germany; 3Haematologie und Onkologie, Universitaetsklinikum, Frelburg, Germany Targeting of viral vectors to cells or tissues of interest is a promising approach to increase the efficiency and safety of gene transfer Our lab has successfully introduced a three-step protocol to produce AAV-2 random peptide display libraries that ensure the encoding of displayed pcptidcs by the packaged AAV genome (Mueller et a! Nat, Biotechnol 2003; 21: 1040-6) Selection of vector targeting sequences on a panel of different cell types demonstrated that the selected targeting sequences not only improved the efficiency but also the specificity ofgene transduction Our current study is aimed at selecting targeting vectors which overcome the physical barriers to the target tissue by in vivo application of AAV peptide display libraries WeselectedAAV2 vectors for mouse cardiac gene transfer by in vitro application on primary neonatal rats cardiomyocytes and furthermore by in vivo homing to mouse heart after intravenous injection ofthe displayed libraries Amplification oflibrary viruses internalized within mice hearts was achieved by organo-typie culture of heart tissue slices super infected with Ad5 Specific peptide sequences could be enriched after rounds of in vitro screening on primary cardiomyocytes or rounds of in vivo biopanning in mice hearts Selected mutants were then evaluated in vivo via tail-vein injection into mice Analysis of gene transfer specificity and transduction efficiency using selected clones will be presented 77 Genetic Modification of AAV1 Capsid for MN Targeted Gene Delivery Thais Federici, I Adam Davis,l Qingshan Tcng,I Jonathan Riley, I Jeffrey S Bartlett.' Nicholas M Boulis.' 'Neurosciences, Cleveland Clinic, Cleveland, OH; lGene Therapy Center; Children s Research Institute and Department 0/ Pediatrics, Ohio State University, Columbus, OH Background: Gene delivery to motor neurons (MN) using adeno-associated virus (AAV) vectors is limited by inefficient vector binding to axons and the high affinity of AAV for muscle We proposed to genetically modify AAV to increase the transduction ofMNs by mimicking the uptake and retrograde delivery of tetanus toxin The binding domain of tetanus toxin (TIC), which binds trisialogangliosides (GTI b), was used to isolate a 12-mer linear peptide, Tet l, by eluting a phage library from GTlb with TIC We have previously shown that Tetl binds selectively to differentiated PCI2 cells, primary MNs, and dorsal root ganglia (ORG) in vitro, and have performed competition ELISAexperiments to demonstrate TetI-specific binding to the targeted GT! b receptor We have also shown Tetl uptake at MN axon termini and retrograde delivery in vitro using Campenot chambers, as well as neuronal binding and spinal cord uptake ofTetl after peripheral administration in vivo I-Iere we provide data suggestingthat insertionof this peptideepitope into the AAVI capsid may improve MV-mediated gene delivery to MNs Methods: Our plan was to insert the Tetl peptide into the AAVVI' I, VP2,and VP3 capsid proteins by PCR-basedsite-directed mutagenesis of the AAVI Cap gene A site was chosen following VI'I amino acid 590 based on our previous studies showing that this region of the capsid protein can tolerate small peptide insertions Packaging plasmids encoding modified AAVl capsid genes were used to produce AAVI.Tet I vectors that were compared to unmodifiedAAVI vectors for their ability to transduce SH-SY5Y cells and for uptake at axon termini The molecular structure ofthe predicted Tetl-modified AAVI capsid was also modeled in silico Results: ModifiedAAVI vectors were produced and shown to increase transduction ofSH·SY5Y cells relative to unmodified particles Uptake of vectors with modified capsid at axon termini was also markedly enhanced compared to the wild-type vector.Current experiments are underway to confirm the presence of the Tetl epitope in the capsid of the modified vectors Although all evidence suggested that the TetI peptide was present and exposed on the surface of modified vector particles, structural modeling predicted suboptimal display of the epitope This finding opens the possibility that efficiency and specificity might be further improved by additional modifications to promote epitope display away from the AAV capsid Conclusions: These findings suggest that modifiedAAVvectors displaying neurotropic peptides might mediate enhanced gene delivery to the spinal cord Moreover, these experiments demonstrate our ability to construct AAV vectors for MN-specific gene delivery through genetic modification of the AAV Cap gene Furthermore, our molecular model suggests that epitope display could be improved by subsequent modification and that even greater enhancement MNspecific gene transfer might be possible 78 Site Specific Modification of AAV Vector Particles with Biophysical Probes and Targeting Ligands Using Biotin Ligase Matthew D Staebler;' Irwin Chen,' Alice Y Ting," Jeffrey S Bartlett.1,3 'Gene Therapy Center; Children s Research Institute Columbus, 011; "Department of Chemistry; Massachusetts Institute ofTechnology, Cambridge MA; JDepartment ofPediatrics, The Ohio State University Columbus Off Adeno-associated virus (AAV) has garnered considerable interest as a gene therapy vector due to its ability to transduce a wide Molecular Therapy Volume 15.Supplement I ~bl' C;oppight © T be American Soci ety of Gene Therapy 2007 S31 ... target tissue by in vivo application of AAV peptide display libraries WeselectedAAV2 vectors for mouse cardiac gene transfer by in vitro application on primary neonatal rats cardiomyocytes and... broaden AAV2 ''s tropism On the other hand, it can be used to redirect AAV2 ''s tropism by combining peptide insertionwith R585Aand R588Amutations ablatingprimary receptorbinding (heparansulfate...how, by analysis of the now availableAAV-2atomic structure, anew position was found and improved to become a valid alternative to 587 By inserting a known peptide - RGD4C - into this new site we