Generation of recombinant skin in vitro by adeno-associated virus type 2 vector transduction

DNA VIRUSES: OTHER HSV after intratumoral injection. To obtain systemic delivery of virus, mice bearing brain tumors were injected in the intracarotid artery. BBB disruption (BBBD) was assessed using Evans bluealbumin (68.5 kD, which does not cross the intact BBB) after intracarotid injection of PBS, 25% mannitol or 1.8M arabinose at 0.5ml/s/kg for 20s followed immediately by 2% Evans Blue. As expected, blue staining is restricted to the ipsilateral side of the brain after BBBD with mannitol and arabinosie, but not PBS. Because of toxicity in the arabinose injected mice, we performed the treatment experiments with mannitol BBBD. G47∆ infection and replication was detected by X-gal staining. MDA-MB-435 tumors in the right frontal lobe had extensive X-gal staining if treated with G47D after BBBD, but very few or no staining cells without BBBD. Minimal staining was seen in normal brain tissue and infrequent staining in lung or liver that did not vary with or without BBBD. The mice treated with G47D after BBBD survived significantly longer than Mock (p<0.0001, Logrank test) or G47D without BBBD (p< 0.02, Logrank test). There was no significant difference in survival between G47D after PBS and Mock after BBBD. In mice with tumor contralateral to the injection side, G47∆ with BBBD did not increase survival. The ability to successfully perform BBBD in mice allows to examine intracarotid artery injection in syngeneic tumor models and transgenic or knockout mice.

799. Initial Characterization of a SuperPermissive Cell Line for Adeno-Associated Virus (AAV): Microarray-Intimated CK2 Helps AAV Replication Hong You,1 Yong Liu,1 Jawahar L. Mehta,1 Paul L. Hermonat.1 Internal Medicine, University of Arkansas for Medical Sciences, Little Rock, AR.

1

Adeno-associated virus (AAV) infection is negatively associated with cervical cancer and is being used as a human gene therapy vector. However, AAV is very selective as to the specific conditions under which it replicates. The identification of cellular genes involved in AAV replication will allow for a better understanding of AAV’s molecular biology and AAV’s oncoprotective phenotype, as well as allow for the easier generation of recombinant (r)AAV for gene therapy. Here we identify a primary cervical cancer cell line, PT3, which allowed for 10 fold higher AAV DNA replication levels than other isolates and normal primary keratinocytes. Microarray gene expression comparative analysis of PT3 cells to others identified many differences in gene expression. Casein kinase 2 (CK2) was found to be the most highly expressed protein kinase in PT3 cells. Following this lead CK2 stimulated AAV DNA replication in normal keratinocyte-based organotypic epithelial rafts (skin) both as an expressed cDNA ( subunit) or as transfected protein. CK2 also phosphorylated the AAV encoded Rep78, major replication/ regulatory protein in vitro. These data suggest that CK2 may be a cellular “helper” gene for AAV, possibly through phosphorylating Rep78. However, the number of genes identified by microarray comparative analysis suggested that, in addition to UP-regulated genes such as CK2, the genes which are DOWN-regulated might also be significant for AAV replication as four fold more such downregulated genes were identified.

Molecular Therapy Volume 9, Supplement 1, May 2004 Copyright  The American Society of Gene Therapy

800. Generation of Recombinant Skin In Vitro by Adeno-Associated Virus Type 2 Vector Transduction Nalini Agrawal,1 Yong Liu,1 Jawahar L. Mehta,1 Hong You,1 Paul L. Hermonat.1 1 Internal Medicine, University of Arkansas for Medical Sciences, Little Rock, AR. It has long been recognized that skin may be a particularly good target for pharmacologic gene therapy and as a platform for the secreting of systemically distributed molecules. Adeno-associated virus (AAV type 2) might be a useful vector for such skin gene therapy as skin is the natural host tissue for AAV in which it functions as an autonomous parvovirus. Here it is shown that recombinant (r)AAV vectors containing the granulocyte macrophage-colony stimulating factor (GM-CSF), human papillomavirus E6, or green fluorescent protein (GFP) transgenes could transduce primary human keratinocytes in ex vivo culture and that these transduced cells could then be used to form a transgene-positive recombinant skin using the organotypic epithelial raft culture system. Transduction of keratinocytes was demonstrated by six different techniques: reverse transcriptase-polymerase chain reaction (RT-PCR) for RNA expression, enzyme linked immunosorbant assay (ELISA) for product secretion, intracellular staining for protein expression, vectorchromosomal junction PCR and Southern blot analysis of proviral sequences, and in situ immunohistochemistry and ultraviolet light fluorescence for GFP protein expression. ELISA demonstrated that AAV/GM-CSF/Neo infected keratinocyte/raft skins secreted GMCSF (measured as high as 25 ng/cm2 of skin) and maintained expression out to 60 days post-infection. These data support the utility and efficiency of AAV-based gene delivery to produce genetically altered keratinocytes and r-skin.

801. Novel Targeting Non-AAV Parvoviral Vectors for Gene Therapy Refael Itah,1,2 Igor Etingov,2 Clay Davis,1 Jacov Tal.1,2 1 Developmental Molecular Genetics, Ben-Gurion University, Beer-Sheva, Israel; 2Virology, Ben-Gurion University, BeerSheva, Israel. Parvoviruses infect and are able to kill mid- to late-embryo and cancer cells, but are not pathogenic to fully differentiated, nor to early embryonic cells. While their pathogenicity in vivo is highly species specific, they frequently cross the species barrier when infecting oncogenically transformed cells. Hence, the parvovirus minute virus of mice (MVM) can establish lytic infections in a variety of transformed human cells. The tissue specificity of MVM is largely determined intracellularly by a capsid host-range determinant (hrd), which presumably interacts with an unknown cellular receptor - an interaction that facilitates uncoating of the infecting virus. When forced upon non-permissive cells, MVM was shown to evolve by generating variants with altered host-range properties. We have isolated and sequenced a number of such variants by end-point dilution and plaque purification. The isolates varied from the wild type virus by two types of mutations: a base substitution at nt 1955, in the single Sp1 binding site of the viral P38 promoter (which directs the synthesis of the two capsid proteins), and variable combinations of mutations in the codons for 6 amino acids within a 220 amino acid stretch of the viral capsid. Each of the viruses was examined for its ability to grow and induce cell death in a variety of cell lines. The results demonstrated that MVM isolates with different combinations of surface amino acids possessed characteristic host range properties. The examination of MVM host range mutants in tissue culture cells represents only a fraction of what is likely to occur during S303