VIRAL VECTOR FOR THE TARGETED TRANSFER OF GENES IN THE BRAIN AND SPINAL CORD

Abstract

The invention relates to novel peptides, polypeptides or proteins which specifically bind to cells of the brain and/or the spinal cord. The peptides, polypeptides or proteins can be part of a viral capsid, and they can be used for guiding a recombinant viral vector selectively to the brain and/or spinal cord after systemic administration to a subject, where it provides for a tissue-specific expression of one or more transgenes. The invention therefore also relates to a recombinant viral vector, preferably an AAV vector, comprising a capsid containing at least one of the peptides, polypeptides or proteins of the invention and at least one transgene which is packaged within the capsid. The viral vector is particularly suitable for the therapeutic treatment of a disease or functional disorder of the brain and/or the spinal cord. The invention further relates to cells and pharmaceutical compositions comprising the viral vector of the invention.

Claims

1. A peptide, polypeptide, or protein that specifically binds to cells of the brain and/or spinal cord, characterized in that it comprises the amino acid sequence of SEQ ID NO:6.

2. A peptide, polypeptide, or protein according to claim 1, characterized in that it comprises the following: (a) one of the amino acid sequences of SEQ ID NO:2-5, or (b) an amino acid sequence which differs from the amino acid sequence of SEQ ID NO:1 by modification of one amino acid.

3. The protein according to claim 1, which is a capsid protein of a viral vector, preferably a capsid protein of an adeno-associated virus (AAV).

4. The protein according to claim 3, which is a capsid protein of an AAV of a serotype selected from the group consisting of serotypes 2, 4, 6, 8, and 9.

5. The protein according to claim 4, which is a capsid protein of an AAV of serotype 2.

6. (canceled)

7. The protein according to claim 1, wherein the peptide is present in the region of amino acids 550-600 of the capsid.

8. The protein according to claim 1, comprising the following: (a) an amino acid sequence having at least 80% identity to the amino acid sequence of SEQ ID NO:10 and further comprises the sequence of SEQ ID NO:1 or an amino acid sequence which differs from the amino acid sequence of SEQ ID NO:1 by modification of one amino acid; or (b) a fragment of the amino acid sequences defined in (a).

9. A viral capsid which comprises a peptide, polypeptide, or protein according to claim 1.

10. A nucleic acid which encodes a peptide, polypeptide, or protein comprising the amino acid sequence of SEQ ID NO:6 that specifically binds to cells of the brain and/or spinal cord.

11. A plasmid which comprises a nucleic acid according to claim 10.

12. A recombinant viral vector which comprises a capsid and a transgene packaged therein, wherein the capsid comprises at least one capsid protein comprising a peptide, polypeptide, or protein comprising the amino acid sequence of SEQ ID NO:6 that specifically binds to cells of the brain and/or spinal cord.

13. The recombinant viral vector according to claim 12, which is a recombinant AAV vector.

14. The recombinant viral vector according to claim 13, which is an AAV vector of a serotype selected from the group consisting of serotypes 2, 4, 6, 8, and 9.

15. (canceled)

16. The recombinant viral vector according to claim 12, wherein the transgene encodes one of the following proteins: a membrane or tight junction protein, a neuraminidase, glucuronidase, a chemokine antagonist, neurotrophic factor of glia cells (GDNF), neprilysine, cholesterol 24 hydroxylase, aromatic L-amino acid decarboxylase, a tyrosine hydroxylase, GTP cyclohydrolase I, and survival of motor neuron (SMN) protein.

17. The recombinant viral vector according to claim 16, wherein the transgene encodes a neuraminidase.

18. The recombinant viral vector according to claim 12, wherein the transgene is in the form of an ssDNA or a dsDNA.

19. A method for the targeted delivery of viral vectors to the brain and/or spinal cord in a subject, comprising administering a recombinant viral vector according to claim 12 to the subject.

20. The method of claim 19, wherein the vector is formulated for intravenous administration.

21. A cell which comprises a peptide, polypeptide, or protein according to claim 1.

22. A pharmaceutical composition which comprises a peptide, polypeptide, or protein according to claim 1.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0065] FIG. 1 shows the in vivo selection method of the AAV-peptide library used according to the invention.

[0066] 1.: Randomized AAV peptide library with approx. 1×10.sup.8 different capsid variants;

[0067] 2.: Withdrawal of the target organ, 8 days after injection;

[0068] 3.: DNA isolation and amplification of viral DNA fragments via real time PCR;

[0069] 4.: Cloning into peptide library plasmids for sequencing and production of a secondary peptide library;

[0070] 5.: Co-transfection of HEK293T cells; production of a secondary peptide library;

[0071] 6.: secondary AAV peptide library for further rounds of selection; contains pre-selected capsid variants;

[0072] 7.: intravenous injection of the peptide library into the mouse.

[0073] FIG. 2 shows the sequences of the brain-specific peptides identified by means of the selection method. The peptide with the sequence shown in SEQ ID NO:1 was identified in the fourth round of selection.

[0074] FIG. 3 shows the gene expression after intravenous injection of recombinant AAV vectors into the living mouse. The luciferase expression was measured 14 days after systemic injection of 5×10.sup.10 vector genomes using the IVIS® 200 Imaging System. A: vectors on the basis of the unmodified AAV2 wild-type capsid show gene expression predominantly in the liver. There is no measurable gene expression in the brain. B: In contrast, the recombinant AAV vector rAAV2-NRGTEWD (SEQ ID NO:1) induces a strong and specific gene expression in the brain.

[0075] FIG. 4 shows a long-term expression analysis in mouse after systemic administration of recombinant rAAV2-NRGTEWD (SEQ ID NO:1) vector. Repeated measurements using the IVIS® 200 Imaging System exhibit stable gene expression in the brain over a period of 168 days (n=2).

[0076] FIG. 5 shows the results of the determination of luminescence and vector 14 days after intravenous injection of 5×1010 vector genomes of the recombinant AAV vector rAAV2-NRGTEWD (SEQ ID NO:1). A: Measuring gene expression as bioluminescence in the lysates of different organs of the mouse showed a high specificity of the vector for the brain. B: Measuring the distribution of vector genomes in the lysates showed a significant accumulation of the vector in the brain. Mean values+standard deviation. Statistics were calculated by one-way ANOVA. p<0.05=*; p<0.01=**; p<0.001=*** for n=3.

EXAMPLES

[0077] All data was determined as mean values+standard deviation (SD). The statistical analysis was performed using the GraphPad Prism 3.0 program (GraphPad Software, San Diego, USA). Data was analyzed by one-way ANOVA followed by multiple comparison tests as per Bonferroni. P values >0.05 were considered significant.

Example 1: Selection of AAV2 Peptide Libraries

[0078] For the selection of tissue-specific AAV2 capsids, a random peptide library was prepared and selected in five rounds. A random X.sub.7-AAV peptide library with a theoretical diversity of 1×10.sup.8 individually occurring clones was prepared using a two-stage protocol as previously described [26-27]. A degenerate oligonucleotide was first produced which codes for seven randomized amino acids at nucleotide position 3967 in the AAV genome, which corresponds to the amino acid position R588 in VP1. The oligonucleotide had the sequence: 5′-CAGTCGGCCAGAGAGGC(NNK).sub.7GCCCAGGCGGCTGACGAG-3′ (SEQ ID NO:12). The second strand was produced using a Sequenase (Amersham, Freiburg, Germany) and the primer with the sequence 5′-CTCGTCAGCCGCCTGG-3′ (SEQ ID NO:13). The double-stranded insert was cut with BglI, purified with the QIAquick Nucleotide Removal Kit (Qiagen, Hilden, Germany) and ligated into library plasmid pMT187-0-3 that had been digested with SfiI [26]. The diversity of the plasmid library was determined by the number of clones grown from a representative aliquot of transformed, electrocompetent DH5a bacteria on agar containing 150 mg/ml ampicillin. Library plasmids were harvested and purified by using the Plasmid Preparation Kit from Qiagen. The AAV library genomes were packaged into chimeric wild-type and library AAV capsids (AAV transfer shuttle) by transfecting 2×10.sup.8 293 T cells in 10 cell culture dishes (15 cm) with the plasmid pVP3 cm (containing the wild-type cap genes with modified codon usage without the inverted terminal repeats) [27], the library plasmids and the pXX6 helper plasmid [28], wherein the ratio between the plasmids was 1:1:2. The resulting AAV library transfer shuttles were used to infect 2×10.sup.8 293 T cells in cell culture dishes (15 cm) with an MOI of 0.5 replication units per cell. Cells were superinfected with Ad5 (provided by the Laboratoire de Therapie Génique, France), with an MOI of 5 plaque-forming units (pfu/cell). The final AAV display library was harvested from the supernatants after 48 hours. The supernatants were concentrated using VivaSpin columns (Viva Science, Hannover, Germany) and purified by iodixanol density gradient ultracentrifugation as previously described [29], and titrated by real-time PCR using the cap-specific primers 5′-GCAGTATGGTTCTGTATCTACCAACC-3′ (SEQ ID NO:14) and 5′-GCCTGGAAGAACGCCTTGTGTG-3′ (SEQ ID NO:15) with the LightCycler system (Roche Diagnostics, Mannheim, Germany).

[0079] For the in vivo selection 1×10.sup.11 particles of the genomic library were injected into the tail vein of FVB/N mice. The particles were given 8 days for the distribution and the infection of the target cells. After 8 days, the mice were killed and the brains were removed. The total DNA of the tissue was extracted using the DNeasy Tissue Kit (Qiagen). The random oligonucleotides that were included in AAV particles of the library and had accumulated in the tissue of interest were amplified by nested PCR using the primers 5′-ATGGCAAGCCACAAGGACGATG-3′ (SEQ ID NO:16) and 5′-CGTGGAGTACTGTGTGATGAAG-3′ (SEQ ID NO:17) for the first PCR and the primers 5′-GGTTCTCATCTTTGGGAAGCAAG-3′ (SEQ ID NO:18) and 5-TGATGAGAATCTGTGGAGGAG-3′ (SEQ ID NO:19) for the second PCR. The PCR-amplified oligonucleotides were used to prepare secondary libraries for three additional rounds of selection. The secondary libraries were generated like the primary libraries (see above), but without the additional step of producing transfer shuttles. The secondary plasmid library was used to transfect 2×10.sup.8 293 T cells in cell culture dishes (15 cm) at a ratio of 25 library plasmids per cell, wherein the transfection reagent Polyfect (Qiagen) was used. After each round of selection, several clones were sequenced. The applied selection method is shown in FIG. 1.

[0080] Results: After five rounds of selection, several brain-binding capsids were selected. Capsids comprising the peptide sequences NRGTEWD (SEQ ID NO:1) were found to bind particularly strong to cells of the brain and spinal cord (see below). A further group of peptides, which also showed specificity for the brain and spinal cord comprised the peptide ADGVQWT (SEQ ID NO:2), DDGVSWK (SEQ ID NO:3), SDGLTWS (SEQ ID NO:4) and SDGLAWV (SEQ ID NO:5). These peptides comprised the general motif XDGXXWX (SEQ ID NO:6). The peptides obtained in the various rounds of selection are shown in FIG. 2.

Example 2: Preparation and Quantification of Recombinant AAV Vectors in HEK293T Cells

[0081] The clones enriched in Example 1 were produced as recombinant AAV vectors and tested for their transduction profile. Recombinant AAV vectors were produced by triple transfection of HEK293T cells. The cells were incubated at 37° C., 5% CO.sub.2 in Dulbecco's modified Eagle Medium (Invitrogen, Carlsbad, USA), supplemented with 1% penicillin/streptomycin and 10% fetal calf serum. Plasmid DNA was transfected into 293T cells with the transfection agent Polyfect (Qiagen, Hilden, Germany). Four days after transfection, the cells were harvested and lysed, and the vectors were purified by means of iodixanol density gradient ultracentrifugation as previously described [29]. For the transfections, pXX6 was used as adenoviral helper plasmid [28], which encodes the luciferase gene pUF2-CMV-luc [27] or the GFP gene pTR-CMV-GFP [30], as was a plasmid encoding the AAV capsid of interest. The plasmids encoding the AAV capsid mutants which had been previously selected from the AAV library, and wild-type controls, were modified pXX2-187 [31] or pXX2 [28]. The inserts were processed as described into library inserts (see above). To quantify the recombinant vectors, the genomic titer was determined by the LightCycler system, as previously described [32], by real-time PCR using the CMV-specific primers 5′-GGCGGAGTTGTTACGACAT-3′ (SEQ ID NO: 20) and 5′-GGGACTTTCCCTACTTGGCA-3′ (SEQ ID NO:21).

[0082] Results: It was found that the yield with respect to virus titer for recombinant viruses with luciferase reporter gene was comparable to that of vectors which comprised a wild type AAV2 capsid which indicates that the accumulated peptides do not affect the assembly of the capsid or the packaging of the gene.

Example 3: Examination of the Tropism of the Recombinant AAV Vectors In Vivo

[0083] To be able to examine the tropism of the enriched peptides in vivo, the peptides were introduced into the capsid of a recombinant vector comprising a luciferase reporter gene. Vectors with mutated capsids were injected into mice along with control vectors. The AAV vectors were administered intravenously at a dose of 5×10.sup.10 vector genomes (vg)/mouse (n=3 animals per injected AAV clone). On day 14, the animals were anesthetized with isoflurane. The luciferase expression was analyzed using a Xenogen IVIS200 Imaging System (Caliper Lifescience, Hopkinton, USA) with the Living Image 4.0 (Caliper) software, following intraperitoneal injection of 200 μl of luciferin substrate (150 mg/kg, Xenogen) per mouse. Representative, in vivo bioluminescence images of the expression of the transgene at different positions (ventral, dorsal, lateral) were taken when the luminescence in relative light units (photons/sec/cm.sup.2) reached the highest intensity.

[0084] Then the animals were sacrificed, the organs of interest were removed quickly, and images of the expression of the transgene in individual organs were immediately taken. The organs were then frozen in liquid nitrogen and stored at −80° C. To quantify the luciferase expression, the organs were homogenized in reporter lysis buffer (RLB, Promega, Madison, USA). The determination of the luciferase reporter gene activity was carried out in a luminometer (Mithras LB9 40, Berthold Technologies, Bad Wildbad, Germany) at 10-second intervals after the addition of 100 μL luciferase assay reagent (LAR, Promega), with a 2-second delay between each of the measurements. The values were normalized in each sample with respect to the total amount of protein using the Roti NanoQuant protein assay (Roth, Karlsruhe, Germany).

[0085] Results: The in vivo measurement of bioluminescence after 14 days showed that the peptide NRGTEWD (SEQ ID NO:1) led to an expression of the transgene in the brain (10.sup.4 p/sec/cm.sup.2/r). These results were confirmed by the control experiments carried out ex vivo with explanted organs. A randomly selected control clone of the non-selected library (CVGSPCG) (SEQ ID NO:43) led to a weak gene expression that occurred primarily in the heart and in some parts of the abdomen, but not in the brain (not shown). Wild-type AAV2 caused a weak gene expression in the heart, liver and skeletal muscle, but not in the brain (FIG. 3A).

[0086] The examination of the luciferase activity of tissue lysates from representative organs revealed that the vectors which comprised the brain-specific NRGTEWD (SEQ ID NO:1) capsid led to a strong and specific gene expression in the brain (1.1×10.sup.7 RLU/mg protein, see FIG. 5A). In other organ tissues, the NRGTEWD (SEQ ID NO:1) luciferase vector hardly showed expression.

[0087] The results further showed that the brain-specific expression of the transgene mediated by the NRGTEWD (SEQ ID NO:1) vector remained organ-specific over a long period. After intravenous administration of the AAV2 NRGTEWD (SEQ ID NO:1) luciferase vectors, the expression of the transgene was measured over a period of 168 days. The radiation emitted in the brain region was determined quantitatively. Over the entire period of time, the expression of the transgene was stable at a high level, and was limited to the brain (FIG. 4).

Example 4: Analysis of the Vector Distribution

[0088] In order to check whether the brain-specific expression of the transgene of intravenously injected NRGTEWD (SEQ ID NO:1) vectors is based on a specific homing, first the distribution of vectors was investigated 14 days after intravenous administration of 5×10.sup.10 gp/mouse. The quantification of the vector genomes was performed by real-time PCR. First, the total DNA was extracted from the organ concerned at various time points after intravenous administration of 5×10.sup.10 vg/mouse using a tissue homogenizer (Precellys 24, Peqlab, Erlangen, Germany) and the DNeasy Tissue Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The DNA was quantified using a spectrophotometer (NanoDrop ND-2000C, Peqlab). The analysis of the AAV vector DNA in the tissues was performed by quantitative real-time PCR using the above-described CMV-specific primer, wherein 40 ng of template were used, normalized with respect to the total DNA.

[0089] Results: The quantification of the vector genomes by real-time PCR showed a brain-specific homing of NRGTEWD (SEQ ID NO:1). The amount of vector genomes which could be detected in the brain (1.6×10.sup.4±7.1×10.sup.3 vg/100 ng total DNA) was significantly higher than the amount of vector genomes which was detected in another organ (FIG. 5B). To determine the direct correlation between vector homing and expression of the transgene, the vector distribution of wild-type AAV2, the control peptide CVGSPCG and the brain-specific peptide NRGTEWD (SEQ ID NO:1) was measured 14 days after intravenous administration of 5×10.sup.10 gp/mouse, i.e., at the time when the expression of the transgene was determined (see above). The genomes provided by wild-type AAV2 vectors were mainly recovered from the liver and spleen, and the genomes of vectors which had the control peptide were obtained largely from the spleen (data not shown). In total, the amount of vector genomes which were detected in the spleen were relatively equal in all examined capsid variants, suggesting a non-specific capture mechanism for the particles in the reticulo-endothelial system which is independent of the provision and the expression of the transgene. In contrast, the distribution data of genomes which were provided by vectors which had the brain-specific peptide NRGTEWD (SEQ ID NO:1) was highly similar to the expression data of the transgene, with a highly specific accumulation observed in the brain. The amount of vectors detected in the brain which showed the peptide NRGTEWD (SEQ ID NO:1) was about 168-fold higher than in brains, which were injected with a wild-type vector or a control capsid vector. Overall, this data indicates that a brain-specific expression of the transgene, mediated by NRGTEWD (SEQ ID NO:1) vectors, is achieved by a tissue-specific homing of circulating particles.

Example 5: Immunohistochemistry and Histology

[0090] Immunohistochemistry was used to visualize the expression of the transgene at the cellular level in the brain, as well as in a control organ, 14 days after the intravenous administration of the rAAV-GFP vector having the peptide NRGTEWD (SEQ ID NO:1) and/or the wild-type AAV capsid as control. The brains of the animals were fixed with 4% (w/v) paraformaldehyde. The tissues were embedded in paraffin. Sections with a thickness of 2 μm were removed from wax, rehydrated and used for immunohistochemistry. An immunohistochemical procedure was performed using polyclonal antibodies for GFP (A-11122, Invitrogen) or CD31 (AB28364, Abcam, Cambridge, USA). The activity of the endogenous peroxidase was inactivated with 1% H.sub.2O.sub.2 in methanol for 30 minutes. Prior to staining with CD31, the sections were heated in citrate buffer (pH 6.0) for 20 minutes at 100° C. After washing in PBS, the sections were incubated for 30 minutes with PBS, 10% goat serum (Vector Lab, Burlingame, USA) and 2% milk powder (Roth). Primary antibodies were allowed to bind for 1 hour at 37° C. After washing in PBS, the sections were incubated for 30 minutes with a secondary, biotinylated goat anti-rabbit antibody (Vector Lab). Bound antibodies were visualized by using the VECTASTAIN-Elite ABC kit (Vector Lab) and 3,3′-diaminobenzidene (DAB, Sigma-Aldrich, St. Louis, USA). Selected sections were counterstained with Hemalum.

[0091] Results: In the brains of mice injected with rAAV-NRGTEWD (SEQ ID NO:1), a microscopic examination showed intensive staining of the endothelial cells over the entire microvasculature and to a slightly lesser extent in the large vessels (data not shown). In contrast, brain tissue of mice which was injected with wild-type AAV2 vector showed no staining. To confirm the tissue specificity, the liver was analyzed as a control organ (a tissue which is known to frequently demonstrate high expression of a transgene after injection of wild-type AAV2 vector). In the liver, hepatocyte staining was observed after administration of wild-type rAAV2 vector; but no staining was observed after administration of rAAV2-NRGTEWD (SEQ ID NO:1) vector. The endothelial lineage of cells transduced with the vectors was confirmed by CD31 staining, wherein the pattern obtained by the GFP staining was confirmed in serial sections of the brains of mice injected with rAAV2-NRGTEWD (SEQ ID NO:1) (data not shown).

[0092] The examination of the spinal cord of mice which have been injected with rAAV-NRGTEWD (SEQ ID NO:1) also revealed an intensive staining of the endothelial cells of the microvasculature which shows that not only the endothelial cells of the brain but rather the endothelial cells of the complete central nervous system are transducable with the peptides of the invention (data not shown).

Example 6: Production and Quantification of Recombinant AAV Vectors Using the Baculovirus Expression System in Sf9 Insect Cells

[0093] For the production of recombinant AAV vectors in Sf9 insect cells [33-35] the modified AAV2 genome having the oligonucleotide insert in the cap gene which encodes the peptide insertion (see above) was cloned into the donor plasmid pFASTBAC Dual (Life Technologies, Darmstadt, Germany). In addition, an artificial intron was inserted into the donor plasmid which included the polh promoter, thereby giving plasmid pFBD-Rep.sub.in/Cap.sub.in [35]. For establishing the donor plasmid pFB-CAG-eGFP, the CAG promoter and the eGFP gene were cloned together with the SV40 polyadenylation signal and the AAV2 ITRs into plasmid pFASTBAC1 (Life Technologies). The donor plasmids were used for transforming DH10Bac E. coli cells which were subsequently used for isolating recombinant bacmids that comprised the recombinant AAV genome or the eGFP transgene cassette, respectively. The bacmids (9 μg) were used for transfection of 1×10.sup.6 Sf9 cells using the Fectofly-Transfektionreagenz (Polyplus Transfection/VWR International GmbH, Darmstadt, Germany) in a 6-well format. After 3 days of incubation of the transfected Sf9 insect cells at 27° C. in insect X-Press Medium (Lonza, Cologne, Germany) with 1% Gentamycin (Lonza), 500 μl of the recombinant baculoviruses present in cell culture supernatants were used for the amplification of 2.5×10.sup.7 fresh Sf9 cells in T175 cell culture flasks for additional 3 days at 27° C. in Insect X-Press Medium (Lonza) with 1% Gentamycin. The baculoviruses amplified in this way were used for infecting fresh Sf9 cells for producing recombinant AAV vectors. For this purpose, recombinant baculovirus with inserted AAV genome and recombinant baculovirus with inserted eGFP transgene cassette were mixed and used together in 400 ml Insect X-Press Medium with 1% Gentamycin in a 1 L Erlenmeyer flask for infecting 6×10.sup.8 insect cells. The cells were subsequently incubated at 27° C. under agitation (110 rpm). 4 days after infection, the cells were harvested, lysed, and the AAV vectors were purified via iodixanol gradient ultra centrifugation as described before [29]. For quantification of the recombinant vectors, the genomic titer was determined by quantitative real time PCR using the CMV specific primer of SEQ ID NO:20 and SEQ ID NO:21 in the LightCycler system as described before [32].

[0094] Results: By using the baculovirus expression system, higher titers of recombinant AAV vectors were achieved in Sf9 insect cells compared to the production in HEK293T cells after triple transfection. While the yield of virus production in HEK293T cells had a maximum of 1.9×10.sup.4 genomic particles per cell, a yield of up to 7.9×10.sup.4 genomic particles per cell were observed in Sf9 insect cells. It could further be observed that rAAV2-NRGTEWD (SEQ ID NO:1) vectors produced in Sf9 insect cells had a higher affinity for neurons than comparable recombinant vectors, which have been produced in HEK293T cells (data not shown). Thus, the choice of the specific production process for the recombinant vectors provides the possibility of increasing the specificity of the vectors for the neuron or endothelial cells, respectively.

READINGS

[0095] [1] Tenenbaum et al., 2003, Curr Gene Ther, 3:545-565 [0096] [2] Work et al., 2006, Mol Ther, 4:683-693 [0097] [3] Shi et al., 2006, Hum Gene Ther, 17:353-361 [0098] [4] US 2007/0172460 A1 [0099] [5] Michelfelder et al., 2009, PLOS One, 4(4): e5122 [0100] [6] Kaplitt et al., 2007, Lancet, 369(9579): p. 2097-105 [0101] [7] Gray et al., 2010, Mol Ther, 18(3): p. 570-8 [0102] [8] Shevtsova et al., 2005, Exp Physiol. 90(1): p. 53-9 [0103] [9] Gray et al., 2011, Hum Gene Ther, 22(9): p. 1143-53 [0104] [10] Aebischer et al., 1996, Nat Med, 2(6): p. 696-9 [0105] [11] McCown et al., 1996, Brain Res, 713(1-2): p. 99-107 [0106] [12] During et al., 1998, Gene Ther. 5(6): p. 820-7 [0107] [13] Nagabhushan Kalburgi et al., 2013, Discov Med, 15(81): p. 111-9 [0108] [14] Alonso et al., 2013, Mol Ther Nucleic Acids, 2: p. e73 [0109] [15] Shi and Bartlett, 2003, Mol Ther, 7:515-525 [0110] [16] Loiler et al., 2003, Gene Ther, 10:1551-1558 [0111] [17] Rabinowitz et al., 1999, Virology, 265:274-285 [0112] [18] Wu et al., 2000, J Virol, 74:8635-8647 [0113] [19] Shi et al., 2001, Hum Gene Ther, 12:1697-1711 [0114] [20] Warrington et al., 2004, J Virol, 78:6595-6609 [0115] [21] Girod et al., 1999, Nat Med, 5:1052-1056 [0116] [22] Grifman et al., 2001, Mol Ther, 3:964-975 [0117] [23] Opie et al., 2003, J Virol, 77:6995-7006 [0118] [24] Kern et al., 2003, J Virol, 77:11072-11081 [0119] [25] Russell et al., 1998, J Virol, 72:309-319) [0120] [26] Müller et al., 2003, Nat Biotechnol 21, 1040-1046 [0121] [27] Waterkamp, et al., 2006, J Gene Med 8, 1307-1319 [0122] [28] Xiao et al., 1998, Journal of Virology 72, 2224-2232 [0123] [29] Zolotukhin, et al., 1999, Gene Ther 6, 973-985 [0124] [30] McCarty et al., 2001, Gene Ther 8, 1248-1254 [0125] [31] Michelfelder, et al., 2007, Exp Hematol 35, 1766-1776 [0126] [32] Rohr et al., 2005, J Virol Methods 127, 40-45 [0127] [33] Urabe et al., 2002, Hum Gene Ther 13, 1935-1943 [0128] [34] Kohlbrenner et al., 2005, Mol Ther 12, 1217-1225 [0129] [35] Chen, 2008, Mol Ther 16, 924-930 (2008)