PORCINE ADENQ-ASSOCIATED VIRUSES

PRIOR APPLICATION INFORMATION

The instant application claims the benefit of US Provisional Patent Application 60/969,733, filed September 4, 2007. BACKGROUND OF THE INVENTION

Adeno-associated viruses (AAV) are non-enveloped, non-pathogenic viruses with a single-stranded genome approximately 4.7 kb in length of either positive or negative polarity. The replication of AAV necessitates a coinfection with helper virus such as Adenovirus or Herpes Simplex virus. In the absence of helper virus, AAV integrates into the host chromosome to establish a latent infection. AAV has been isolated from many organisms including human, nonhuman primates, ovine, avian, snake, bovine, murine, and caprine (1-13). It has been reported that various AAV serotypes have differing tropisms and can transduce specific organs more efficiently than others, such as the high transduction efficiency of AAV5 in airway epithelial, muscle, and retina cells as compared to AA V2 (14, 15).

Gene transfer vectors derived from AAV can deliver genes in a variety of tissues in vivo. Recent observations have also promoted the evaluation of AAV as potential genetic vaccine vectors due to their long-term expression profile which can stimulate robust antibody responses (16, 17). Currently, clinical trials are taking place using AAV as a vector for the treatment of ailments such as Parkinson's disease, Cystic fibrosis, Leber's congenital amaurosis, HIV infection, and various other genetic disorders (18, 19, 20, 21 ). These recombinant AAV (rAAV) vectors are based on the well characterized human AAV serotype 2 which is seroprevalent in up to 80% of the human population with neutralising antibodies found in 35% of them (22).

The AAV capsid proteins encoded by the cap gene were shown to be the main determinant for tissue tropism and constitute an important target for the immune response. The cap gene encodes three proteins expressed from two different alternatively spliced transcripts; VP1 , VP2, and VP3. Some serotypes of AAV were found to be less immunogenic than others which may translate into better gene transfer vehicle and less efficient vaccine vector or vice versa (23, 24, 25). In the past, AAV has conventionally been isolated from contaminated

adenoviral stocks or young, sick animals or children (1 , 2, 3, 4, 5, 6, 7, 9, 12). Recently, isolation of new AAV isolates was extensively performed using PCR amplification of AAV sequences from genomic DNA of different animal species (8, 10, 11 , 13). In this study, novel porcine AAV sequences were identified by PCR using genome walking strategies. We describe the isolation of novel AAVpol and the characterisation of the serological profile and tissue tropism in vitro and in vivo.

Adeno-associated virus (AAV) is a non-pathogenic parvovirus with a single- stranded DNA genome. The virus typically has a relatively small genome size.

While other parvoviruses replicate autonomously, wild type AAV requires co-infection with a helper virus for lytic phase reproduction. In the absence of a helper virus, wild-type AAV establishes a latent, non-productive infection with long- term persistence.

SUMMARY OF THE INVENTION According to a first aspect of the invention, there is provided a method for the preparation of a recombinant porcine AAV particle comprising: providing a cell engineered to express: a peptide having at least 80% identity to a porcine AAV capsid protein selected from the group consisting of VP1 , VP2 and VP3; a functional rep gene; a minigene comprising two AAV inverted terminal repeats and a transgene inserted between the two inverted terminal repeats; and sufficient helper functions to permit packaging of the minigene into a porcine AAV capsid particle. According to a second aspect of the invention, there is provided a single dose, injectable pharmaceutical composition comprising 1X10 ⁹ -1X10 ¹³ purified recombinant AAV particles prepared according to the method of claim 1 and a suitable excipient.

BRIEF DESCRIPTION OF THE DRAWINGS AND TABLES

Figure 1. Phylogenetic tree representing alignments of known AAV serotypes with novel porcine AAVs. Alignments were performed based on the

truncated regions of known AAV serotypes corresponding to those regions of the AAV genomes isolated from porcine tissues. The 1.7 kb rep-cap sequence was isolated for AAVpol . The 1.4kb sequence was found for AAVpol , AAVpo2, and AAVpo3. A. Left phylogenetic tree of 1.7 kb truncated regions of known AAV serotypes aligned with the isolated 1.7 kb region for AAVpol and right phylogenetic tree of truncated 1.4 kb "end-of-cap" region for published AAV serotypes aligned with the novel AAVpol , -po2, and -po3 1.4 kb regions. B. Figure depicting the full-length AAV genome. The AAV genome is composed of a rep gene and a cap gene flanked by ITRs on either side. The lines for the 1.7 kb rep- cap region and 1.4 end-of-cap region represent the regions of the AAV genome isolated for AAVpol , -po2, and -po3, as well as the regions of published AAV serotypes used for alignment.

Figure 2. Western blot of A. AAV2/5 capsid proteins and B. AAV2/po1 VP1 ,

VP2, and VP3 using AAV VP1 , VP2, VP3 monoclonal antibodies. AAV2/5 VP1 has a predicted molecular weight of 80.42 kDa while AAV2/po1 VP1 has a predicted weight of 80.18 kDa using DNASTAR Lasergene 7 Protean software. Proteins were isolated from transfected HEK 293T cells and separated by a 10% SDS

PAGE, and subsequently transferred to nitrocellulose paper. Mouse monoclonal anti-VP1 , -VP2, and -VP3 were used as the primary antibody and goat anti-mouse horseradish peroxidase (HRP) conjugated antibody as secondary antibody.

Figure 3. Electron microscopy of chimeric AAV2/5 and AAV2/po1 CsCI- purified particles containing the LacZ transgene at 100,000 X magnification. Particles were produced by transfection of subconfluent HEK 293T cells and purified 48 hours later. Figure 4. Muscle susceptibility to AAV2/5 or AAV2/po1 following intramuscular injection in BALB/c mice. The mice were sacrificed and muscle was harvested 30 days p.i., frozen with O. C. T., and sections were cut using a cryostat.

Figure 5. AAV2/po1 or AAV2/5 transduction following subretinal injection in C57BL/6 mice. EGFP was used as a reporter gene. The retinal cell layers are of the eyecup depicting the RPE (retinal pigment epithelium), ONL (photoreceptor outer nuclear layer), and the GCL (ganglion cell layer).

Figure 6. Nucleotide sequence of AAVpoia Rep/Cap 1643 bp (SEQ ID No. 18).

Figure 7. Nucleotide sequence of AAVpoi b End of Cap 1398 bp (SEQ ID No. 19).

Figure 8. Nucleotide sequence of AAVpoi c Full Cap 2151 bp (SEQ ID No. 20). Figure 9. Nucleotide sequences of AAVpo2 End of Cap 1385 bp (SEQ ID

No. 21) and AAVpo3a Signature Sequence 252 bp (SEQ ID No. 22).

Figure 10. Nucleotide sequence of AAVpo3b End of Cap 1549 bp (SEQ ID No. 23).

Figure 1 1. Amino acid sequences of AAVpol VP1 Peptide (SEQ ID No. 24), AAVpol VP2 Peptide (SEQ ID No. 25) and AAVpol VP3 Peptide (SEQ ID No. 26).

Table 1. Primer names and primer sequences used to isolate novel AAV sequences from various porcine tissues.

Table 2: Nucleotide and amino acid identity of AAVpol VP1 capsid with other known AAV serotypes. Table 3: Tropism of AAV2/po1 or AAV2/5 in different cell lines.

Table 4: Biodistribution of AAV2/5 and AAV2/po1 after IV Tail Vein injection in mice.

Table 5: Neutralisation profile of homologous or heterologous AAV2/po1 or AAV2/5 anti-serum. Table 6: Neutralisation of AAV2/po1 or AAV2/5 by pooled Human Ig

(Immune Globulin Intravenous (Human) Carimune® NF).

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are now described. All publications mentioned hereunder are incorporated herein by reference. Described herein is the discovery of a number of novel porcine adeno- associated viruses and nucleotide and amino acid sequences associated therewith. As will be appreciated by one of skill in the art and as discussed below, these sequences can be used advantageously in the generation of delivery

constructs, for example, for the delivery of an antisense agent, a gene therapy agent or a vaccine agent.

It is noted that high antigenic compatibility and low toxicity is associated with xenograft transplantation of porcine tissues in immunodeficient human recipients. We hypothesized that AAVs of porcine origin, if existing, could be highly compatible to human tissues and of high efficiency for in vivo gene transfer. Porcine tissues were screened for the presence of AAV using universal primers designed from an alignment of published AAV sequences. In total, AAV sequences were detected in 9 out of 21 farmed pigs. Several AAV sequences were isolated from various porcine tissues, and BLAST analysis confirmed high to low homology with known AAV sequences of different origin. Sequence analysis confirmed the isolation of at least three novel porcine AAV isolates which we named AAVpol , - po2, and -po3 (see Figures 6-1 1 ). The AAVpol capsid protein was closely related to AAV5 capsid with 87.1 % amino acid identity. AAVpo2 was found to be related to human AAV2 whereas the isolated sequence of AAVpo3 was highly divergent from all AAV isolates previously described. Hybrid vector particles based on AAV2 rep and AAVpol cap with a packaged LacZ transgene were successfully produced. The novel AAV2/po1 could efficiently transduce muscle fibers or the retinal pigment epithelium following intramuscular or subretinal injection in mice respectively. Serological analysis indicates that AAVpol is a unique serotype and that pre-existing immunity to AAVpol could not be detected in the human sera evaluated. Novel AAVs derived from porcine tissues may significantly contribute to the generation of new preventive or curative clinical modalities acceptable for human use. As will be appreciated by one of skill in the art, such viruses and the nucleotide sequences thereof can be used in the construction of useful products and reagents, for example, for the production of gene replacement or gene expression vectors.

For example, therapeutic or prophylactic therapies in which the constructs described herein have or are likely to have utility include but are by no means limited to blood disorders, lung disorders, neurological disorders and muscle disorders.

As will be appreciated by one of skill in the art, the genetic factors required

for several disease states are known and accordingly constructs may be prepared which either reduce expression, reduce translation or replace such genetic factors Examples include but are by no means limited to hormones and growth and differentiation factors, including but by no means limited to insulin, glucagons, growth hormone, parathyroid hormone, growth hormone releasing factor, follicle stimulating hormone, luteinizing hormone, cumin chorionic gonadotropin, vascular endothelial growth factor, angiopoietins, angiostatin, granulocyte colony stimulating factor, erythropoietin, connective tissue growth factor, basic fibroblast growth factor, acidic fibroblast growth factor, epidermal growth factor, platelet-derived growth factor, insulin growth factors I and II, any one of the transforming growth factor superfamily, including TGFα, activins, inhibins or any of the bone morphogenic proteins (BMP) BMPs 1 -15, any one of the heregluιn/neuregulιn/ARIA/neu differentiation factor family of growth factors, nerve growth factor, brain-derived neurotrophic factor, neurotrophns NT-3 and NT-4/5, ciliary neurotrophic factor, glial cell line derived neurotrophic factor, neurturin, agrin, any one of the family of semaphorins/collapsins, netrιn-1 and netrιn-2, hepatocyte growth factor, ephπns, noggin, sonic hedgehog, tyrosine hydroxylase and cystic fibrosis transmembrane conductance regulator Further examples as well as additional information on the construction and use of AAV particles may be found at least in published US Patent Application 2007/0036760 which is incorporated herein by reference, particularly for the sections describing preparation of constructions and use of said constructs

For example, as can be seen in Figure 1 , AAVpo2 is phylogenetically related to AAV2 and accordingly it is expected that AAVpo2 may have similar properties, that is, expression in skeletal muscles, neurons, vascular smooth muscle cells and hepatocytes

Similarly, AAVpol is related to AAV5 and accordingly is expected to have similar properties However, AAVpol is functionally different than AAV5 and transduces less cell types, as discussed herein As shown in Figure 1 , AAVpo3 is more distantly related to a number of AAV isolates

As will be appreciated by one of skill in the art, in some embodiments, the ϊelatedness' of different clades can be used advantageously when designing or

developing a treatment program. For example, as discussed below, ITRs from one clade are likely to be compatible with capsid proteins from other closely related clades. In other embodiments, following an initial treatment with a given clade, a second treatment may be administered using a distantly related clade, thereby reducing the likelihood of cross-reactivity, that is, that the second delivery construct will be recognized by the host's immune system. It is of note that the first and second delivery constructs may carry the same or a similar insert (for example, an antigenic agent for vaccination) or may contain different inserts.

AAV is found in many tissues, including the heart of a high percentage of human and nonhuman primates. Transplantation of pig tissues (eg heart valves) into human has been well documented. During numerous xenotransplantations in immunocompromised individuals, no side effects due to sudden viral replication originating from the transplanted tissue were noted. On the other hand, viruses transmitted by nonhuman primates are often dangerous pathogens (Ebola etc) and can adapt and become human pathogens (SIV to HIV). Historically this is also a concern (transmission of SV40 from the polio vaccine that was produced in monkey cells.

As will be apparent to one of skill in the art, the capsid is responsible for the tropism and vector entry and thus gene transfer efficacy. Accordingly, in one embodiment of the invention, there is provided a method for the preparation of a recombinant porcine AAV particle comprising: providing a cell engineered to express: a peptide having at least 80% identity to a porcine AAV capsid protein selected from the group consisting of vp1 , vp2 and vp3; a minigene comprising two AAV inverted terminal repeats and a transgene inserted between the two inverted terminal repeats; a functional rep gene compatible with said AAV inverted terminal repeats; and sufficient helper functions to permit packaging of the minigene into a porcine AAV capsid particle, and recovering said porcine AAV capsid particle from said cell. As will be appreciated by one of skill in the art, the cell as described above may be described as a packaging cell. Specifically, the cell may be engineered by

means known in the art to express, either constitutively or inducibly, the elements listed above. It is further of note that these expression constructs may be supplied by transient expression constructs or by expression constructs integrated into the genome of the packaging cell. It is further noted that the amino acid sequences of AAVpol VP1 (SEQ ID No. 24), VP2 (SEQ ID No. 25) and VP3 (SEQ ID No. 26) are provided in Figure 11.

As will be appreciated by one of skill in the art, 'porcine AAV capsid protein' refers to porcine-derived AAV virus sequences which encode all or part of at least one of VP1 , VP2 or VP3. Attention is directed to figures 6-1 1 for examples of such sequences and/or sequences from which said VP1 , VP2 and VP3 amino acid sequences may be derived (for example, by searching an appropriate nucleotide sequence for a homologous open reading frame).

In some embodiments of the invention, the peptide has at least 80%, at least 81 %, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91 %, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% identical to a procine adeno-associated virus capsid protein selected from the group consisting of VP1 , VP2 and VP3. In some embodiments, VP1 , VP2 and VP3 are the amino acid sequences of AAVpol VP1 (SEQ ID No. 24), VP2 (SEQ ID No. 25) and VP3 (SEQ ID No. 26) respectively.

A peptide having at least 80% identity to a porcine AAV capsid protein may be a porcine AAV capsid protein related to a sequence shown in Figure 11 , or may be a humanized version thereof or a chimeric version of two or more AAV capsid proteins. As will be appreciated by one of skill in the art, such chimeric hybrids may easily be assembled by comparing two or more AAV capsid peptides for regions of high similarity (conserved regions) and regions of lesser similarity (variable regions) and then preparing chimeric capsids by substitution or combination of regions or domains of interest.

It is also important to note that on comparison of VP1 , VP2 and VP3 sequences from a variety of adeno-associated viruses, it is possible to determine highly conserved regions of the capsid protein in which alteration, insertion and/or deletion is unlikely to be tolerated. Similarly, variable or non-conserved or less highly conserved regions can be determined in a similar manner which one of skill

in the art will understand are more likely to tolerate alteration

As will be appreciated by one of skill in the art, the transgene is inserted between the two inverted terminal repeats It is of note that the ITRs and the AAV REP protein must be compatible, however, it is noted that determining such compatibility is well within the skill of one of knowledgeable in the art It is also important to note that the AAV ITR does not necessarily need to be derived from porcine AAV

As discussed above, the transgene may be used to correct or ameliorate gene deficiencies or defects which may include deficiencies in which normal genes are expressed at less than normal levels or deficiencies in which the functional gene product in not expressed at all In other embodiments, the transgene may encode an antisense construct arranged to reduce expression of a specific target or the transgene may encode an antigen or immunogenic domain for use as a vaccine While the currently accepted insert or transgene size limit is ~4kb it is believed that up to 10kb can be incorporated and transferred although efficacy is lower (about 80%) Above 10 kb, efficiency of transfer starts dropping significantly

As will be appreciated by one of skill in the art, 'helper functions' include but are by no means limited to sequences capable of inducing AAV expression It is of note that such sequences are well known to one of skill in the art In one embodiment, the host cell is provided with and/or contains an E1 a gene product, an E1b gene product, an E2a gene product and/or an E4 ORF6 gene product As will be known to one of skill in the art, the helper virus is primarily required by AAV to drive the cell into "S phase" in order to drive viral DNA replication While both Adenovirus and HSV have the ability to do this, other means known in the art may also be used within the invention

It is of note that the host cell or packaging cell may be selected from any suitable biological organism, including prokaryotic and eukaryotic cells, including but by no means limited to insect cells, yeast cells and mammalian cells In a preferred embodiment, the cells are HEK 293T cells

As will be appreciated by one of skill in the art, the assembled particles may be recovered from the packaging cell and purified by any suitable means known in the art, for example, by cesium chloride gradient It is important to note that as

used in this context, 'purity' does not require absolute purity but only that the particles have been substantially enriched in concentration or isolated or purified away from cellular components.

In some embodiments, there is provided a single dose, injectable pharmaceutical composition comprising 1X10 ⁹ -1X10 ¹³ purified recombinant AAV particles and a suitable excipient.

The cap gene can trans-complement AAV chimeric particles with the AAV2 rep gene and produce functional vector capable of achieving gene transfer in different susceptible tissues in vitro and in vivo. The present study reports the isolation of sequences defining three novel AAV isolates isolated from porcine tissues. The complete sequence of the cap gene of the isolate AAVpol was used to generate functional particles that showed strong tropism mainly for muscle fibers and the retinal pigment epithelium. Good transduction of muscle fibers combined to poor transduction of other tissues could make AAV2/po1 an attractive gene transfer vehicle for systemic delivery and possibly help the treatment of myopathies such as Duchenne muscular dystrophy. Muscles susceptibility will need to be evaluated in other animal models to further investigate this hypothesis since AAV2/po1 , which has also been isolated from porcine liver and lung, may demonstrate a different preferential tropism in other species. A high histocompatibility has been documented of porcine tissues into human recipients following xenotransplantation. We hypothesized that this compatibility would translate in porcine AAVs with good transduction efficiency and high safety profile for therapeutic applications in humans. An interesting observation is the isolation of AAV2 sequences identical or closely resembling the original sequence of the virus isolated from human tissue. Contamination of samples with exogenous AAVs is always a concern when isolating sequences with a methodology as sensitive as PCR. The complete absence of AA V2 vector in our research institute before the isolation of sequences from genomic pig DNA strongly suggest that the isolation of AAV2 sequences was not the result of contamination but rather present in the porcine tissues extract tested. This may indicate zoonotic transmission and/or co-evolution of certain AAVs in pigs and some humans supporting the concept of high compatibility between the two species. If accurate, this phenomenon may however be of low incidence since AAV2/po1 was not

neutralised by the pooled human Ig tested.

The absence of neutralising antibody to AAV2/po1 in pooled human sera adds a desired characteristic for in vivo administration of the vector in humans. The long-term safety profile following porcine-derived AAVs in vivo remains to be investigated. Finally, absence of cross-neutralisation with anti-serum against AAV2/po1 and anti-serum from all other serotypes evaluated strongly suggest that AAVpol is a unique serotype and structurally different from the other AAVs evaluated.

The invention will now be further described by way of example; however, the examples do not necessarily limit the invention.

Use of PCR to Determine AAVpol , -po2, and -po3 Sequences from Porcine Tissues.

Since AAV has the ability to integrate its genome into the host chromosome during a latent infection, we employed the use of PCR techniques to isolate novel AAV sequences from genomic porcine DNA similarly to previous work reported by Gao et al. (8) and Mori et al. (1 1 ). Various pig tissues including heart, lung, spleen, liver, and gut were harvested from rural private farms and industrial slaughterhouses, and DNA was isolated and screened for the presence of novel AAV sequences by PCR. The use of primers specific for conserved regions of the AAV genome generated at least three distinct AAV sequences from porcine tissues we called AAVpol , AAVpo2 and AAVpo3. A 258bp sequence corresponding to the novel AAVpol was found in various organs from nine pigs obtained from three different locations. In contrast, the highly divergent 252bp and 1.4kb AAVpo3 sequence was only found in the intestine of one animal. Along with the unique porcine AAV sequences, other sequences highly similar (with 1 to 20 nucleotide changes throughout the cap gene) or identical to AAV2 cap were also isolated from approximately 12.5% of the porcine tissues screened. PCR amplification was also performed to isolate a 1.64kb sequence which spans the rep and cap genes and a 1.4 kb sequence which overlaps with the 1.64 kb fragment and covers the 3' end of the cap gene (Figure 1 ). The entire cap gene of the AAVpol could be isolated from 63% of the tissues obtained from privately owned farmed pigs with the spleen having the highest success with 75% of AAV

sequence recovery. Porcine lung, heart and liver were second for successful isolation of AAV sequences. Only one farmed-animal proved negative for AAV sequences, while tissue samples originating from the slaughterhouse showed only 4% of AAV sequence recovery from 51 samples. Alignment of the nucleotide sequence of AAVpol , -po2, or -po3 with available AAV sequences revealed that AAVpol was more closely related to AAVgo.1 , AAVpo2 to AAV2, and that AAVpo3 was uniquely divergent from the others (Figure 1). AAVpol cap showed 89.2% amino acid identity and 86.3% nucleotide identity with AAVgo.1 and 87.1 % amino acid identity and 84.7% nucleotide identity with AA V5 (Table 2). The predicted amino acid sequences for the capsid proteins of known AAV serotypes when compared to AAVpol show that region C, G, and H are the most divergent regions.

Production of AAV2/po1 hybrid vector. Several studies have demonstrated that AAV2 rep can trans complement AAV cap gene resulting in successful production of functional AAV particles with the capsid protein of interest. Triple transfection of the trans packaging plasmid encoding for AA V2 rep and -po1 cap genes, a LacZ expressing c/s-plasmid and the helper plasmid containing the adenoviral genes necessary to drive production of AAV in HEK 239T cells resulted in the successful generation of chimeric AAV2/po1 particles (Figure 2 and 3). Purification of AAV2/po1 by cesium chloride gradient produced particles showing the distinct pattern of VP1 , VP2 and VP3 on a denaturing acrylimide gel visualized by Western blot using anti-VP1 , -VP2, and -VP3 antibodies (Figure 2). Protean software of DNASTAR Lasergene 7 predicted the VP1 protein of AAVpol to have 716 amino acids with a molecular weight of 80.18 kDa, and the VP1 protein of AAV5, used as a positive control, to contain 724 amino acids with a predicted molecular weight of 80.42 kDa. Electron microscopy revealed that AAVpol particles were comparable in shape and size to AAV2/5 with a diameter of 20-25 nm (Figure 3). Viral titres of 0.4 to 4 x 10 ¹³ total GC of AAVpol per preparation of 3 to 4 ml could routinely be obtained as assessed by Taqman PCR.

Transduction efficiency of AAV2/po1 in cultured cells. Transduction efficiency was evaluated for AAV2/po1 on different cell types in vitro. The closely related

AAV2/5 vector was also evaluated in parallel and used as a benchmark. Subconfluent cell lines from different species including mouse (NIH 3T3), dog (MDCK), pig (VirBlei), monkey (VeroE6) and human (HEK 293T and HeLa) were infected with 1 x 10 ⁹ GC of each AAV expressing LacZ and the number of positive blue cells recorded after 48 hours. Results, summarized in Table 3, show that AAV2/po1 transduced all the cultured cell lines tested less efficiently than AAV2/5, with the exception of Vero E6 and VirBlei cells. VirBlei is a porcine retinal cell line which showed substantially higher susceptibility to AAV2/po1 than AAV2/5.

Tropism of AAV2/po1 in mice. The tropism and transduction efficiency of AAV2/po1 was studied in mice in parallel to AAV2/5 that was used as a control. 1 x 10 ¹¹ GC was administered per mouse in the muscle, liver or lung by intramuscular, tail vein or intranasal injection and were harvested and analysed 30 (liver and muscle) or 60 (lung) days after administration. LacZ positive cells could not be detected in AAV2/po1 -treated lungs and only rare scattered positive cells were observed in the liver (only anecdotal cells could be detected per 20 liver sections). In contrast, high transduction efficiency was observed in the muscle following I. M. injection of AAV2/po1 which was comparable to that observed with AAV2/5 in the same conditions (Figure 4). Subretinal injections of AAV2/po1 also resulted in high transduction efficiency of the retinal pigment epithelium and photoreceptor layers again comparable to the efficiency recorded for AAV2/5 (Figure 5). Cells in the inner nuclear layer and some cells morphologically similar to Muller Cells were also transduced following AAV2/po1 administration.

Biodistribution of AAV2/po1 was also evaluated in mice from various organs 30 days after systemic administration by tail vein injection (Table 4). Results showed that AAV GC numbers, established from the copy number of the LacZ gene per cell, were significantly lower for AAV2/po1 than for AAV2/5 in all organs studied with the exception of the muscle.

Serological profile of AAV2/po1. Neutralisation assays were performed with the antisera collected from AAV2/po1 and AAV2/5 intramuscularly injected mice in order to evaluate cross-neutralisation between the two related isolates. Antiserum from AAV2/po1 or AAV2/5 treated mice had neutralising antibody titer of 1 :1280 or

1 :5120 against themselves respectively (Table 5). In contrast, AAV2/po1 antiserum had undetectable neutralising antibody against AAV2/5 and vice versa. Overall, AAV2/po1 was not neutralised by serum collected from mice exposed to AAV2/2, - 2/3, -2/4, -2/5, -2/6, -2/7 or -2/8 vector. The AAV2/po1 and AAV2/5 expressing LacZ were also incubated with pooled Human Ig at various dilutions to address the serosusceptibility of the vectors to antibody found in humans. At least fifty percent of AAV2/5 particles were neutralised with 7.5 x 10 ^"5 mg/mL of pooled human Ig while neutralisation was not detected for AAV2/po1 with undiluted human Ig at 12 mg/ml (Table 6).

Isolation of genomic DNA from porcine tissues and amplification of AAV sequences. Various porcine tissues including the lung, heart, spleen, gut, and liver were screened for the presence of AAV sequences. Porcine tissue samples were collected from various locations such as industrial slaughterhouses and private rural farms. Genomic DNA was isolated from approximately 25 mg of each tissue using QIAamp® DNA Mini Kit (QIAGEN). Primers SIG+ and SIG- (5 ¹ - GGTAATTCCTCGGGAAATTGGCATT-3' (SEQ ID No. 3) and 5'- GAATCCCCAGTTGTTGTTGATGAGTC-3' (SEQ ID No. 2) respectively) were used to PCR amplify short fragments referred as the "signature region" as previously described by Gao et al. (8). These fragments of 252 and 258 bp corresponded to two novel AAV sequences from porcine tissues generating AAVpol and -po3 fragments. BLAST analysis showed that AAVpol and -po3 sequences were significantly different from previously described AAV sequences available on GenBank. The 5' sequence of cap was isolated with the primer RC+ described by Mori et al. (1 1); δ'-GGTGCGTAAACTGGACCAATGAGAAC-S' (SEQ ID No. 1 ) and the primer SIG- generating a 1.64 kb fragment encompassing the end of the rep gene and beginning of cap region. . To isolate the 3' end of the cap gene, three successive primers were generated based on conserved regions of the AAV genome along with a reverse degenerate primer in a three step nested thermal-asymmetric interlaced (TAIL) PCR. The three successive primers are: PAAVSP1 + (δ'-GGARATTGGCATTGCGATTCC-S' (SEQ ID NO. 8)), PAAVSP2+ (5'-GACTTCAACCGCTTCCACAGCCAC-S' (SEQ ID NO. 9)), and PAAVSP3+ (5'- GACTCATCAACAACCWACTGGGG-3' (SEQ ID No. 4)). The sequence of the

degenerate primer CED- is: 5'-ACTGAMACGAAT(H/-)AMMCGGTTTATTGA-3' (SEQ ID No. 5). A 1.4kb fragment which overlapped with the 1.64kb 5' end sequence was isolated using these primers. Specific primers were also designed to isolate the full length cap gene from porcine tissues in one PCR with the following sequences: PO1 CAPBEGIN+ (δ'-ATGTAGTGGATCTTGACGATG-S' (SEQ ID No. 6)) and PO1 CAPEND- (5'- CATGTGACAGGATAGGGTTA-3' (SEQ ID No. 7)). All PCR amplified fragments were gel purified using QIAguick® Gel Extraction Kit (QIAGEN) and subsequently cloned into pCR®2.1-TOPO® Vector (Invitrogen). DNA sequencing was performed by the DNA Core Facility at the National Microbiology Laboratory.

Production of hybrid AAV2/po1 particles. Chimeric AAV particles were produced with AAV2 rep and AAVpol cap expressing plasmids. AAVpol cap was cloned in place of the AAV cap gene in the packaging plasmid p600 trans using Swa\ and Not\ restriction sites, generating pACK 2/po1. The AAV2CMVLacZ or EGFP plasmids (26) expressing the LacZ or EGFP reporter genes from a CMV promoter were used as the c/s-plasmid containing the recombinant AAV genome. Triple transfection of Human Embryonic Kidney 293T (HEK 293T) cells with AAV2CMVLacZ or EGFP, the packaging plasmids pACK2/po1 or pACK2/5 (29) and the helper plasmid pDELTA F6 containing adenoviral genes necessary to drive AAV replication was performed with CaPO ₄ as described previously (27, 28). Recombinant AAV vectors were all purified by CsC^ gradients as previously described (28). Titres of AAV vector preparations were determined by TaqMan® PCR for genome copy (GC) with primers LACZ+ (5'- TTAACCCGCCATGCTACTTATCTA-S' (SEQ ID NO. 15)), LACZ- (5'- TGAACTAATGACCCCGTAATTGATT-3' (SEQ ID No. 16)), and probe LACZP (5'- CTCTAGGAAGATCGGAATTCGCCCTTAAGCT AG-3' (SEQ ID No. 17)) or with primers for the bGH poly A as previously described (26). The number of transducing particles per GC was also determined for AAV2/po1 and AAV2/5 on HEK 293T cells (on average, for every 1 x 10 ⁹ GC, AAV2/po1 and AAV2/5 had 2 x 10 ⁵ and 4 x 10 ⁵ transducing units (TU) respectively). The AAV2/5-CMVEGFP and AAVpol -CMVEGFP were produced by the Telethon Institute of Genetics and

Medicine (TIGEM) AAV vector core. AAV specimens were prepared for electron microscopy by adsorption to glow discharged carbon coated formvar films on 400 mesh copper grids for one minute, and negatively contrasted with 2% methylamine tungstate (Nanoprobes, Yaphank, New York). Specimens were imaged in a FEI Tecnai 20 transmission electron microscope operating at 200 kV, at a nominal instrument magnification of 20000Ox. Digital images of the specimens were acquired by an AMT Advantage XR 12 CCD camera (AMT, Danvers, MA).

Serology and neutralisation assay. Recombinant AAV2/5 and AAV2/po1 vectors containing the LacZ transgene were injected intramuscularly into BALB/c mice (1 x 10 ¹¹ GC per injection) in the left tibialis anterior. Serum samples were collected 28 days post infection for serology analysis. The sera were inactivated at 56°C for 45 minutes. Serial dilutions of each sample (1 :10, 1 :20, 1 :40, etc, in 50 μl of DMEM) was mixed with equal volume of the appropriate AAV encoding the LacZ reporter gene (80-100 transducing units/well) and incubated at 37 ⁰ C for 60 minutes. The mixture was then transferred onto subconfluent HEK 293T cells in 96-well flat- bottomed plates and incubated for 90 minutes at 37 ⁰ C in 5% CO ₂ . Control wells were infected with equal amount of AAV vector without the addition of serum or with the addition of non-immune serum. 100 μl of DMEM supplemented with 20% FBS was then added to each well and the plates were incubated at 37°C in 5% CO ₂ for 48 hours. Cells were subsequently stained with 5-bromo-4-chloro-3-indolyl β-D-galactoside (X-gal) and examined under a microscope. The same type of neutralisation assay was also performed with Immune Globulin Intravenous Human Carimune® NF (CSL Behring) in place of the sera. Sample dilutions which showed >50% reduction in the number of LacZ positive cells compared to controls scored positive for neutralising antibody.

Histology and histochemical staining. For muscle gene delivery, the tibialis anterior was harvested 30 days after intramuscular injection of mice. The muscle was embedded in O. C. T. medium and snap frozen in liquid nitrogen-cooled isopentane. For lung gene delivery, AAV2/5 or AAV2/po1 vector expressing LacZ was administered intranasally in BALB/C mice at 1 x 10 ¹¹ GC/mouse in 50 μl. The lungs and trachea were harvested and inflated with a 1 :1 solution of PBS and

O.C.T. medium 64 days post-administration and embedded in O. CT. medium. For liver gene transfer, each mouse was injected in the tail vein with 1 x 10 ¹¹ GC in 100 μl and the liver was harvested 30 days later and embedded in O.C.T. medium. Lung, trachea and liver samples were frozen on ethanol-cooled dry ice. For all tissues, 10μm thick sections were cut with a cryostat and fixed with 1.6 % glutaraldehyde. The sections were then incubated overnight with X-gal at 37°C, lightly stained with eosin and dehydrated in 70-100% ethanol solutions. Sections were analysed on a light microscope.

Subretinal Injections and retinal transduction assessment. All procedures on animals were performed in accordance with institutional guidelines of the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Four-week- old male C57BL/6 mice (Harlan) were used for injections of rAAV vector Before vector administration, mice were anesthetized with an intraperitoneal injection of avertin (1.25% [wt/vol] 2,2,2-tribromoethanol and 2.5% [vol/vol] 2-methyl-2- butanol [Sigma-Aldrich]) at 2 ml/100 g of body weight. Subretinal vector administrations were performed as described previously using 1 ul of a vector solution containing 1 x 10 ¹² GC/ml. A month after vector administration, mice were sacrificed and their eyeballs were harvested and fixed overnight by immersion in 4% paraformaldehyde. The eyeballs were cut so that the lens and vitreous could be removed, leaving the eyecup. Mice eyecups were infiltrated with 30% sucrose for cryopreservation and were embedded in tissue-freezing medium (O.C.T. matrix; Kaltek). For each eye, 150 to 200 serial sections (10 μm thick) were cut along the horizontal meridian, and the sections were progressively distributed on 10 slides so that each slide contained 15 to 20 sections representative of the whole eye at different levels. The sections were mounted with Vectashield (Vector Laboratories, Inc.), and retinal histology images were obtained with an Axiocam (Carl Zeiss) at 2OX magnification.

Biodistribution in mice. AAV2/5 or AAV2/po1 expressing LacZ was administered at 1 x 10 ¹¹ GC in 100μl PBS per C57BL/6 or BALB/c mouse by IV tail vein injection. Various organs including the heart, liver, spleen, kidney, lungs, small intestine, large intestine, muscle, and pancreas were collected 28 days post-

injection and analysed by TaqMan® analysis for the amount of rAAV GC per cell. Briefly, DNA was extracted from approximately 25mg of each tissue sample with QIAamp® DNA Mini Kit and TaqMan® amplification was performed as described previously (30) with the primers LACZ+, LACZ- and probe LACZP described above. The signal was considered positive when the number of GC was superior to 3 times of the number obtained from PBS-injected control tissues. The number of transduced rAAV genomes in each organ is reported as number of GC/cell (500ng DNA contained within 1.5 x 10 ⁴ cells).

While the preferred embodiments of the invention have been described above, it will be recognized and understood that various modifications may be made therein, and the appended claims are intended to cover all such modifications which may fall within the spirit and scope of the invention.

REFERENCES

1. Atchison, R. W., Castro, B. C. & Hamman W. M. (1965) Science 149, 754-756.

2. Yates, V. J., el-Mishad, A. M., McCormick, K. J. & Trentin, J. J. (1973) Infect. Immun. 7, 973-980. 3. Myrup, A. C ₁ Mohanty, S. B. & Hetrick, F. M. (1976) Am. J. Vet. Res. 37, 907- 910.

4. Coria, M. F. & Lehmkuhl, H. D. (1978) Am. J. Vet. Res. 39, 1904-1906.

5. Clarke, J. K., McFerran, J. B., McKillop, E. R. & Curran, W. L. (1979) Arch. Virol. 60, 171-176. 6. Bantel-Schaal, U. & zur Hausen, H. (1984) Virology 134, 52-63.

7. Rutledge, E. A., Halbert, C. L. & Russell, D. W. (1998) J. Virol. 72, 309-319.

8. Gao, G. P., Alvira, M. R., Wang, L., Calcedo, R., Johnston, J. & Wilson, J. M. (2002) Proc. Natl. Acad. Sci. USA 99, 1 1854-1 1859.

9. Farkus, S. L., Zadori, Z., Benko, M., Essbauer, S., Harrach, B. & Tijssen, P. (2004) J. Gen. Virol. 85, 555-561.

10. Gao, G., Vandenberghe, L. H., Alvira, M. R., Lu, Y., Calcedo, R., Zhou, X. SWilson, J. M. (2004) J. Virol. 78, 6381-6388.

11. Mori, S., Wang, L., Takeuchi, T. & Kanda, T. (2004) Virology 330, 375-383.

12. Olsen, E. J., Haskell, S. R., Frank, R. K., Lehmkuhl, H. D., Hobbs, L. A., Warg, J. V., Landgrat, J. G. & Wunschmann, A. (2004) J. Vet. Diag. Invest. 16, 461-464.

13. Lochrie, M. A., Tatsuno, G. P., Arbetman, A. E., Jones, K., Pater, C ₁ Smith, P. H., McDonnell, J. W., Zhou, S. Z., Kachi, S., Kachi, M., Campochiaro, P. A., Pierce, G. F. & Colosi, P. (2006) Virology 353, 68-82.

14. Davidson, B. L., Stein, C. S., Heth, J. A., Martins, I., Kotin, R. M., Derksen, T. A., Zabner, J., Ghodsi, A. & Chiorini, J. A. (2000) Proc. Natl. Acad. Sci. USA 97,

3428-3432.

15. Zabner, J., Seiler, M., Walters, R., Kotin, R. M., Fulgeras, W., Davidson, B. L. & Chiorini, J. A. (2000) J. Virol. 74, 3852-3858.

16. Lai, C. M., Shen, W. Y., Brankov M., Lai' Y. K. Y., Barnett N. L., Lee- S. Y., YeO ¹ I. W. S., Mathur R., Ho J. E. S., Pineda' P., Barathi A., Ang C. L., Constable !.

J. & Rakoczy, E. P. (2005) MoI. Ther. 12, 659-668.

17. Rivera, V. M., Gao, G. P., Grant, R. L., Schnell, M. A., Zoltick, P. W., Rozamus, L. W., Clackson, T. and Wilson, J. M. (2005) Blood 105, 1424-1430.

18. Carter, B. J. (2005) Human Gene Therapy 16, 541-550.

19. i ■ . , ^* . \ , ^τ J , , - ^J . , . , ,» ⁽ . - , . , i., '-• ,., ^• . - ., ^• \ and , ■ .. (2007) The

Lancet 369, 2097-2105. 20. Moss, R. B., MiIIa, C, Colombo, J., Accurso, F., Zeitlin, P. L., Clancy, J. P.,

Spencer, L. T., Pilewski, J., Waltz, D. A., Dorkin, H. L., Ferkol, T., Pian, M.,

Ramsey, B., Carter, B. J., Martin, D. B. & Heald, A. E. (2007) Human Gene

Therapy 18, 726-732.

21. Buch, P. K. & Bainbridge, J. W. (2008) Gene Therapy 15, 849-857. 22. Chirmule, N., Propert, K. J., Magosin, S. A., Qian, Y., Qian, R. And Wilson, J.

M. (1999) Gene Therapy 6, 1574-1583.

23. Wang, Z., Blankinship, M. J., Gregorevic, P., Little, M., T., Storb, R. J., Allen, J.

M , Tapscott, S. J., Chamberlain, J. S. & Kuhr, C. S. (2005) Molecular Therapy U,

S307. 24. Manno, C. S., Pierce, G. F., Arruda, V. R., Glader, B., Ragni, M., Rasko, J. J.

E., Ozelo, M. C, Hoots, K., Blatt, P., Konkle, B., Dake, M., Kaye, R , Razavi, M ,

Zajko, A., Zehnder, J., Rustagi, P., Nakai, H., Chew, A., Leonard, D., Wright, J.

F., Lessard, R. R., Sommer, J. M., Tigges, M., Sabatino, D., Luk, A., Jiang, H.,

Mingozzi, F., Couto, L., Ertl, H. C, High, K. A. & Kay, M. A. (2006) Nature Medicine 12, 342-347.

25. Vandenberghe, L. H., Wang, L., Somanathan, S., Zhi, Y., Figueredo, J , Calcedo, R., Sanmiguel, J , Desai, R. A., Chen, C. S., Johnston, J., Grant, R. L., Gao, G. & Wilson, J. M. (2006) Nature Medicine 12, 967-971.

26. Auricchio, A., Hildinger, M , O'Connor, E., Gao, G. & Wilson, J. M. (2001) Hum. Gene Ther. 12, 71-76.

27. Graham, F. L. & van der Eb, A. J. (1973) Virology 52, 456-467

28. Xiao, X., Li, J. & Samulski, R. J. (1998) J. Virol. 72, 2224-2232.

29. Hildinger, M., Auricchio, A., Gao, G., Wang, L., Chirmule, N. & Wilson, J. M. (2001) J. Virol. 75, 6199-6203. 30. Holland, P. M., Abramson, R. D., Watson, R. & Gelfand, D. H. (1991 ) Proc. Natl. Acad. Sci. U.S.A. 88, 7276-7280.

Table 1. Primer names and primer sequences used to isolate novel AAV sequences from various porcine tissues.

Primer Name Primer Sequence 1. CP3 5'-GGTGCGTAAACTGGACCAATGAGAAC-S' (SEQ ID NO.

1)

2. Sig3072+ δ'-GAATCCCCAGTTGTTGTTGATGAGTC-S' (SEQ ID NO. 2)

3. Sig2839- δ'-GGTAATTCCTCGGGAAATTGGCATT-S' (SEQ ID NO. 3)

4. Cap3158+ δ'-GACTCATCAACAACWACTGGGG-S' (SEQ ID NO. 4) 5. CapEndDeg- δ'-ACTGAMACGAATHAMMCGGTTTATTGA-S' (SEQ ID NO.

5)

6. P5LCapBeg+ δ'-ATGTAGTGGATCTTGACGATG-S' (SEQ ID NO. 6)

7. PδLCapEnd- δ'-CATGTGACAGGATAGGGTTA-S' (SEQ ID NO. 7)

8. AlexGPí δ'-GGARATTGGCATTGCGATTCC-S' (SEQ ID NO. 8) 9. PigSpforí δ'-GACTTCAACCGCTTCCACAGCCAC-S' (SEQ ID NO. 9)

10. ITR5+ δ'-TTGGCCACTCCCTCTMGCGC-S' (SEQ ID NO. 10)

11. RevRep1397- δ'-ACGCASCCGTAAAAGGGCAC-S' (SEQ ID NO. 1 1)

12. ITR3- δ'-TTTGSCCACTCCCTCTMTGCGC-S' (SEQ ID NO. 12)

13. EndCapSeq+ δ'-CGTGGACTTTGCTCCAGAC-S' (SEQ ID NO. 13) 14. AAVpo3 p δ'-CAGTCTCGGGGTGAGAAATG-S' (SEQ ID NO. 14)

Table 2: Nucleotide and amino acid identity of AAVpol VPl capsid with other known AAV serotypes.

AAVpol

% nucleotide identity ¹ % amino acid identity

AAVl 64/7 59^0

AAV2 64.2 59.5

AAV3 63.7 59.6

AAV3b 63.7 59.7

AA V4 61.5 53.6

AAV5 84.7 87.1

AAV6 64.4 58.7

AAV7 64.6 59.2

AAV8 66.1 59.0

AAV9 64.4 56.8

AAVlO 65.2 58.8

AAVI l 61.0 54.6

AAV12 60.3 53.7

AAVgo.l 86.3 89.2

BAAV 62.1 57.1

Alignments were performed using the CLUSTAL W method of MegAlign software (DNASTAR Lasergene 7).

Table 3: Tropism of AAV2/pol or AAV2/5 in different cell lines.

Table 4: Biodistribution of AAV2/5 and AAV2/pol after IV Tail Vein injection in mice.

AAV2/pol AAV2/5

Heart 0.004±0.002 0.042±0.052

Liver 0.003±0.003 0.186±0.265

Large Intestine 0.010±0.008 0.158±0.133

Small Intestine 0.004±0.006 0.019±0.012

Lung 0.035±0.004 0.955±1.406

Kidney 0.009±0.009 0.171±0.284

Spleen 0.208±0.339 1.774±1.592

Pancreas 0.004±0.007 0.030±0.029

Muscle 0.080±0.138 0.041±0.027 rAAV genomes LacZ were targeted for TaqMan PCR. Values present the number of genome copies present per cell.

Table 5: Neutralisation profile of homologous or heterologous AAV2/pol or AAV2/5 antiserum.

Values reported represent the dilution of anti-sera required to neutralise >50% of AAV- LacZ particles.

Table 6: Neutralisation of AAV2/pol or AAV2/5 by pooled Human Ig (Immune Globulin Intravenous (Human) Carimune® NF).

Values represent the concentration of the pooled Human Ig required to neutralise >50% of AAV-LacZ particles.