H U M A N G E N E T H E R A P Y 8:2103-2115 (November 20, 1997) Mary Ann Liebert, Inc. A Recombinant El-Deleted C a n i n e A d e n o v i r a l V e c t o r C a p a b l e o f T r a n s d u c t i o n a n d E x p r e s s i o n o f a T r a n s g e n e i n H u m a n - D e r i v e d C e l l s a n d I n V i v o BERNARD KLONIKOWSKI,i'5 PASCALE GILARDI-HEBENSTREIT,^ lULIETTE HADCHOUEL,^ VOAHANGY RANDRIANARISON,^ SYLVIE BOUTIN,^ PATRICE YEH,' MICHEL PERRICAUDET,' and ERIC J. KREMER* ABSTRACT Human adenovirus (HAV) serotypes 2 and 5 are commonly used as vector backbones for adenovirus-medi- ated gene transfer. However, H A V s were chosen as a backbone for the vectors for historical reasons and have a n u m b e r of significant disadvantages w h e n used as a shuttle for gene transfer in h u m a n s . A s an initial trial to circumvent some of the shortcomings of HAV vectors, we have produced an El-deleted canine adenovirus type 2 ( C A V - 2 ) vector for gene transfer. Initially, we demonstrated that C A V - 2 undergoes an abortive viral cycle in a wide range of human-derived cell lines. Second, we assayed h u m a n sera containing H A V - 5 neu- tralizing antibodies for their ability to inhibit CAV-2-induced plaques on permissive cells. In the cohort tested, our data demonstrate that the humoral response directed against H A V - 5 does not inhibit C A V - 2 plaque for- mation in the majority of cases. Canine cell lines expressing the El region of CAV-2 were generated and char- acterized. A recombinant CAV vector (CAVRSV/3gal) deleted in the El region and harboring lacZ was con- structed. We s h o w that CAVRSV/3gal is able to transduce and direct expression of the transgene in vitro in a variety of m ammalian cells, most notably primary human-derived cells. In addition, gene transfer is demon- strated in vivo using chick embryos. OVERVIEW SUMMARY human cells in vitro. In addition, —10'* transduction units of CAVRSVpgal were able to transduce a wide variety of There are over 100 members of the adenovirus group iden- cell types in 2-day-old chick embryos. tified, with 49 belonging to the h u m a n type. H u m a n aden- oviruses have been the most extensively characterized and were unfortunately chosen as backbones for gene transfer INTRODUCTION vectors for precisely this reason. For gene transfer in hu- mans, non-human adenoviral vectors eliminate several in- CJince 1985, recombinant human adenovirus ( H A V ) vec- nate disadvantages of human-derived vectors while retain- kJioRS have been used for gene transfer in vivo (Ballay et al, ing most, if not all, of the well-characterized advantages of 1985). Although recombinant HAV were initially used in vivo adenoviral vectors. Our initial efforts with C A V - 2 have as vaccines (for review, see Randrianarison-Jewtokoff and Per- shown that this non-human vector can infect, and is repli- ricaudet, 1995), it became apparent that these vectors held con- cation defective in, human-derived cell. An El-deleted ca- siderable potential for gene therapy. Fortiinately, preclinical nine adenovirus type 2 vector containing lacZ was able to testing and phase Ittialshave exposed some of the limitations direct expression of the transgene in h u m a n cells and non- of current HAV vectors (Crystal et al, 1994; Yang et al, 1994). 'CNRS U RA 1301/Rh6ne-Poulenc Rorer Gencell, Laboratoire de Genetique des Virus Oncogenes, InstiUit Gustave Roussy, Villejuif, France 94805. ^Unite INSERM 368, Ecole Nomiale Superieure, Paris, France, 75005. ^Unite de Genetique Mol&ulaire Developpement, histittit Pasteur, Paris, France, 75015. "Genetiion II Programme Therapie G6nique, 91002 Evry, France. ^Present address: INRA U R A de Gen. Mol. & Cell., ENVA, Maison-Alfort, France 94704. 2103