Methods of delivering factor VIII encoding nucleic acid sequences

Abstract

There is provided a nucleic acid molecule comprising a nucleotide sequence encoding a factor VIII protein, wherein a B domain portion of the factor VIII protein is encoded by a nucleotides sequence between 90 and 111 nucleotides in length and has an amino acid sequence that is at least 85% identical to SEQ ID NO: 4 which comprises six asparagine residues. Also provided is a factor VIII protein, a vector comprising the above nucleic acid molecule, a host cell, a transgenic animal, a method of treating Haemophilia A, and a method for the preparation of a parvoviral gene delivery vector.

Claims

1. A method for administering to a subject a nucleic acid molecule encoding a Factor VIII protein, comprising: intravenously administering to the subject a nucleic acid molecule encoding a Factor VIII protein, wherein the Factor VIII protein B domain comprises a spacer that is replaced and wherein the B domain comprises the amino acid sequence set forth in SEQ ID NO:4.

2. The method of claim 1, wherein the B domain of said Factor VIII protein is between 30 and 37 amino acids in length.

3. The method of claim 1, wherein the B domain of said Factor VIII protein is 31 amino acids in length.

4. The method of claim 1, wherein the B domain of said Factor VIII protein comprises the amino acid sequence of SEQ ID NO:5.

5. The method of claim 1, wherein the B domain of said Factor VIII protein consists of the amino acid sequence of SEQ ID NO:5.

6. The method of claim 1, wherein the Factor VIII protein encoded by the nucleic acid molecule comprises domains A1, A2, A3, C1 and C2, and wherein the nucleic acid molecule is codon optimised compared to a corresponding wild type sequence.

7. The method of claim 1, wherein said Factor VIII protein comprises the amino acid sequence of SEQ ID NO:6.

8. The method of claim 1, wherein said nucleic acid molecule encoding said Factor VIII protein comprises the nucleic acid sequence of SEQ ID NO:3.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 is a schematic of rAAV plasmids encoding codon optimised hFVIII-N6 (top panel), codon optimised B domain deleted form (right central panel) and hFVIII variants (bottom panel) containing the 6 asparagine moieties (in bold) that are thought to be required for glycosylation. The B domain is in the middle of the constructs shown in lighter grey. In addition, to the FVIII cDNA, the expression cassette also contains a smaller HLP promoter and a synthetic polyadenylation (Synth pA) signal. The size of the FVIII cDNA as well as the whole rAAV expression cassette is also shown. The full sequence of the B domain of the hFVIII variants (variants 1 and 3) also has a 14 amino acid sequence which flanks the sequences shown for each variant. In addition, the B domain deleted form has the same 14 amino acid sequence which acts as a linker between the domains on either side (the A2 and A3 domains).

(2) FIG. 2 shows the mean hFVIII levelsSEM in murine plasma after a single tail vein administration of rAAV-hFVIII constructs pseudotyped with serotype 8 capsid (dose=110.sup.11 vg/mouse, N=6/group).

(3) FIG. 3 shows FVIII activity level in F8/ mice following a single tail vein administration of high dose of rAAV-HLP-codop-hFVIII vectors (dose=110.sup.12 vg/mouse, N=5/6 animals/group).

(4) FIG. 4 shows alkaline gel analysis of the rAAV-HLP-codop-hFVIII viral genome derived from: Codop-BDD-hFVIII, group 3=low dose (3e13/ml), group 5=high dose (9e13/ml); Codop-hfFVIII-N6, group 4=low dose (3e13/ml), group 6=high dose (9e13/ml); and Codop-FVIII-V3=group 7 (high dose, 9e13/ml). High Mass DNA Ladder is shown by group 1 and Quantification standard by group 2. A discrete band at 5 kb is observed with genome extracted from rAAV-Codop-BDD-hFVIII, and rAAV-Codop-FVIII-V3. However the genome in rAAV-Codop-hfFVIII-N6 appears more heterogeneous.

(5) FIG. 5. A: Yield of AAV-HLP-codop-hFVIII variants pseudotyped with serotype 8 capsid B: Alkaline gel analysis of the AAV-HLP-codop-hFVIII viral genome derived from: codop-BDD-hFVIII, (BDD, group 1); codop-N6-hFVIII, (N6, group 2); and codop-FVIII-V3 (V3, group 3). High Mass DNA ladder is shown by group 1 and Quantification standard by group 2. A discrete band at 5 kb is observed with genome extracted from AAV-codop-BDD-hFVIII, and AAV-codop-FVIII-V3. However the genome in AAV-codop-N6-hFVIII appears more heterogeneous.

(6) FIG. 6. A: Mean hFVIII levelsSEM in murine plasma after a single tail vein administration of AAV-codop-hFVIII constructs pseudotyped with serotype 8 capsid (dose=410.sup.12 vg/kg. N=6/group). B: hFVIII expression levels in mice transduced with 210.sup.13 vg/kg corrected for transgene copy number in the liver at 9 weeks after gene transfer.

(7) FIG. 7. FVIII activity level in F8/ mice following a single tail vein administration of low (210.sup.13 vg/kg, Panel A) or high dose (210.sup.14 vg/kg, Panel B) of AAV-HLP-codop-hFVIII vectors.

(8) FIG. 8. Blood loss in F8/ mice following a single tail vein administration of AAV-HLP-codop-BDD-hFVIII (BDD), AAV-HLP-codop-N6-hFVIII (FVIII N6) and AAV-HLP-codop-hFVIII-V3 (FVIII V3) compared to knockout mice treated with vehicle (V) alone or recombinant human FVIII (rFVIII).

(9) FIG. 9. Anti-hFVIII IgG antibody response. A and B: Anti-hFVIII IgG antibody level following gene transfer with low and high doses of AAV-HLP-codop-BDD-hFVIII (circles). AAV-HLP-codop-N6-hFVIII (squares) and AAV-HLP-codop-hFVIII-V3 (triangles) respectively. C. For comparison anti-hFVIII IgG antibody response following administration of recombinant hFVIII protein is shown.

(10) FIG. 10. Biodistribution of vector following peripheral vein administration of 410.sup.13 AAV8-HLP-codop-hFVIII-V3. Results of qPCR analysis of genomic DNA, isolated from the indicated organs at 9 weeks after tail vein administration of 410.sup.13 vg/kg of AAV8 vector using primers unique to codop-hFVIII. Shown is transgene copy number per diploid genomeSE corrected for variation in loading and amplification efficiency using GAPDH primers.

DETAILED DESCRIPTION OF THE INVENTION

(11) In order to develop a safe and efficient gene transfer strategy for the treatment of haemophilia A (HA), the most common inherited bleeding disorder, the inventors have developed a new FVIII variant called codop-hFVIII-V3 (FIG. 1). This variant builds on a previous variant, a 5013 bp codon-optimised FVIII called codop-hFVIII-N6. The inventors have further modified codop-hFVIII-N6 to improve the efficiency with which it is packaged into rAAV without compromising its potency in vivo.

(12) The cDNA in codop-hFVIII-V3 has been modified to reduce its size to 4424 bp (FIG. 1) through the replacement of the 678 bp B domain spacer sequence with a 93 bp linker that codes for 31 amino acids of which 17 amino acids are unique, including the 6 asparagine moieties believed to be required for efficient cellular processing of FVIII.

(13) The context in which these 6 asparagine moieties are brought together is important, rAAV vectors encoding codop-hFVIII-V1 mediated FVIII expression that was 16 and 10 fold lower than vectors encoding codop-hFVIII-V3 and codop-hFVIII-N6, respectively, in cohorts of mice after a single tail vein injection of 110.sup.11 vector genomes (vg)/mouse (FIG. 2). This difference was highly significant (p=0.0015). Importantly, both codop-hFVIII-V3 and codop-hFVIII-N6 mediated significantly higher level of expression than codop-BDD-hFVIII (FIG. 3).

(14) The inventors' data show that a rAAV expression cassette encoding the 5.2 kb codop-hFVIII-V3 is packaged uniformly as a full length provirus as shown in FIG. 4. In contrast, the packaging of codop-hFVIII-N6 expression cassette is heterogenous. This is due to the larger size of the codop-hFVIII-N6 expression cassette, which at 5.7 kb significantly exceeds the packaging capacity of AAV. Packaging of heterogenous proviral DNA raises safety concerns because of the potential to synthesis and express truncated forms of FVIII, which could provoke an immunological response.

(15) By shortening the B domain of the codop-hFVIII-N6 variant but retaining essential features of the B domain sequence, in particular the N-linked glycosylation consensus sequences, the inventors have been able to enable more efficient packaging of the transgene into AAV. In the course of creating novel sequences for this purpose, one particular sequence N6V3 proved to be associated with highly efficient packaging into AAV. This sequence also showed a remarkable and unpredicted further improvement of transgene expression in animal gene transfer studies.

(16) Based on rational analysis of the structure of factor VIII and on its known secretion pathway, requiring interaction with the chaperon protein LMANN-1, the inventors have deduced that the expression improvement may be due to the following reasons.

(17) The interaction of factor VIII B domain with the lectin LMANN-1 requires multiple N-linked carbohydrate side chains to be present and for them to adopt a specific conformation for binding between the nascent glycopeptide and the lectin.

(18) The wild type B domain is nearly 1000 amino acids long with no likely secondary structure. Therefore, this lengthy peptide requires a considerable time for synthesis into the Golgi and further time for the random coil to adopt a suitable structure stochastically to bring together the widely separated carbohydrate side chains into a conformation that would enable binding to the lectin (LMANN-1).

(19) By shortening the sequence to the minimum length possible that still retains 6 potential N-glycosylation sites (17 mer), the time required for synthesis is drastically reduced.

(20) Furthermore only a very small number of conformations or possibly just one can occur in the glycosylated peptide amongst which is the required tertiary structure for binding the lectin. The inventors have calculated that the length of this peptide is only just long enough to span the distance between the C-terminal of the A2 domain and the N-terminal of the A3 domain in the crystal structure of B domain deleted factor VIII at 53 Angstroms. Therefore, the N6V3 peptide is further constrained to an almost linear structure that would limit the number of sterically possible conformations and, provided the carbohydrate side chains are added in appropriate places, enable the chaperon to bind virtually co-translationally, thus optimizing to the maximum degree possible this essential step in the factor VIII specific secretion pathway.

(21) The unique specificity of the novel N6V3 sequence is further supported by the fact that very minor deviation from this sequence, such as retaining a single extra amino acid between each N-glycosylation consensus sequence trimer, greatly reduces the synthesis and secretion efficiency of factor VIII compared to that obtained with other versions of the truncated B domain.

(22) A New Shorter Codon Optimised FVIII Variant: Codop-hFVII-V3

(23) The inventors have modified codop-N6-hFVIII, to improve the efficiency with which it is packaged into AAV virions as full length viral genome without compromising its potency in vivo. This involved the replacement of the 226 amino acid B domain spacer with a 31 amino acid (93 bp) peptide, containing a 14 amino acid linker sequence as in B domain deleted FVIII and 17 amino acids which are unique. This peptide contained the 6 asparagine residues present in codop-N6-hFVIII that are required for efficient intra-cellular processing. The peptide brings these residues in closer proximity. Consequently, this new hFVIII variant (AAV-HLP-codop-hFVIII-V3) is 5.1 Kb in size, 600 bp smaller than AAV-HLP-codop-N6-hFVIII (5.7 kb), and closer to the packaging capacity of AAV of approximately 5.0 kb (FIG. 1). AAV-HLP-codop-hFVIII-V1 contains a 44 amino acid peptide that includes the same 6 asparagine residues instead of the 226 amino acid spacer in codop-N6-hFVIII. For comparison another AAV vector (AAV-HLP-codop-BDD-hFVIII, 5.0 kb in size) was made which contains a codon optimised hFVIII cDNA from which the B domain has been deleted, retaining a small linker sequence of 14 amino acids.

(24) The yield of AAV8-HLP-codop-hFVIII-V3 vector using the standard HEK293 transient transfection method was comparable (FIG. 5) to that of AAV-HLP-codop-N6-FVIII and AAV8-HLP-codop-BDD-hFVIII. Analysis of viral DNA extracted from 2.510.sup.10 particles of each vector preparation following separation on an alkaline agarose gel showed bands of 5 kb, the expected size for the HLP-codop-BDD-hFVIII (Lane 1, FIG. 5B) and HLP-codop-hFVIII-V3 (Lane 3). In comparison, a rather diffuse signal was observed for the genomes extracted from AAV8-HLP-codop-N6-hFVIII suggesting the packaging of a more heterogeneous proviral species (FIG. 5B, Lane 2).

(25) AAV-HLP-Codop-hFVIII-V3 is More Potent than AAV-HLP-Codop-BDD-hFVIII

(26) AAV vectors containing the different codon optimised FVIII variants, pseudotyped with serotype 8 capsid, were injected as a bolus into the tail vein of 4-6 week old males C57B1/6 mice (N=6) at a dose of 410.sup.13 vg/kg to compare their potency in vivo. The highest level of hFVIII expression was observed with AAV-HLP-codop-FVIII-V3 at 1.520.15 IU/ml (15215% of normal. FIG. 6) 4 weeks after gene transfer. In contrast, AAV8-HLP-codop-N6-hFVIII and AAV8-HLP-codop-BDD-hFVIII mediated hFVIII expression at 0.860.11 and 0.670.12 IU/ml respectively. The difference in plasma FVIII levels between the AAV8-HLP-codop-BDD-hFVIII and AAV-HLP-codop-hFVIII-V3 cohorts of mice was highly significant (p=0.0015, student T test). The lowest level of hFVII expression was observed with AAV-HLP-codop-hFVIII-V1 (0.100.01 U/ml). This is significantly (p<0.0001) lower than FVIII expression in the AAV-HLP-codop-hFVIII-V3 cohort of mice, which suggests that the context in which the 6 asparagine residues are brought together in the synthetic B domain amino acid peptide is important. Tail vein administration of a higher dose of vector (210.sup.13 vg/kg) resulted in between 4-30 fold higher level of plasma hFVIII in the cohort transduced with AAV-HLP-codop-hFVIII-V3 when compared to levels achieved in cohorts of mice transduced with AAV-HLP-codop-N6-hFVIII and AAV-HLP-BDD-hFVIII following correction for transgene copy number (FIG. 6B) in the liver at 9 weeks. The difference in hFVIII levels between AAV-HLP-codop-hFVIII-V3 and AAV-HLP-BDD-hFVIII was highly significant (p=0.0062, student T Test).

(27) Biologic Potency of AAV-HLP-Codop-hFVIII-V3 in F8/ Mice

(28) A direct comparison of the biologic potency of codop-N6-FVIII, codop-BDD-FVIII and codop-FVIII-V3 was performed in F8/ mice. Vector encoding each of these FVIII variants, pseudotyped with serotype 8 capsid was administered into the tail vein of male F8/ mice at a dose of 410.sup.12 (low-dose cohort, n=5/6) or 410.sup.13 (high-dose cohort, n=5/6) vg/kg. For all three constructs the kinetics of expression was broadly similar with plasma hFVIII levels reaching peak levels between 2-6 weeks after gene transfer. For a given construct, hFVIII levels were roughly two fold higher in animals transduced with the high dose of vector when compared to the low dose (FIG. 7). Irrespective of the vector dose, peak hFVIII expression in the cohorts of mice transduced with codop-BDD-hFVIII was approximately two fold lower than observed in animals transduced with codop-N6-hFVIII or codop-hFVIII-V3. At the high dose level the difference in hFVIII expression between the codop-BDD-hFVIII cohort and codop-hFVIII-V3 between weeks 4-8 post gene transfer was highly significant (p<0.001 2 way ANOVA). The average ratio of hFVIII coagulation activity (hFVIII:C) to hFVIII antigen was slightly above 1.0, suggesting the transgenic hFVIII molecules were biologically active.

(29) To establish if the FVIII activity correlated with phenotypic correction in AAV-treated mice, blood loss was analysed by tail clip assay at 8 week after gene transfer (FIG. 8). The amount of blood loss in the AAV-codop-hFVIII-injected mice was almost similar for the 3 codop-hFVIII variants and the two dose levels but substantially lower than observed in FVIII/ mice treated with vehicle instead of AAV. This difference between AAV and vehicle treated F8/ mice was highly significant (p<0.001 one-way ANOVA test). The amount of blood loss in the AAV treated animals was comparable to that observed in F8/ mice treated with recombinant human FVIII (rFVIII) suggesting that rAAV-mediated expression of FVIII restores haemostasis to levels observed with recombinant FVIII. Anti-hFVIII antibodies were detected over time in all AAV transduced animals with the highest levels being observed in the high dose AAV-HLP-codop-hFVIII-V3 cohort. When compared to the response observed after administration of recombinant hFVIII protein (2 U FVIII per week for 6 weeks) the response in the AAV-codop-hFVIII transduced animals was at least 400 fold lower and insufficient to completely neutralise FVIII activity as illustrated by the tail clip assay (FIG. 9). Consistent with this inhibition of coagulation was not observed when two murine samples with the highest anti-FVIII IgG level were assessed in a Bethesda assay, suggesting that these antibodies do not have neutralising activity.

(30) Biodistribution studies (FIG. 10) using a sensitive qPCR based assay demonstrated that the AAV8-HLP-codop-hFVIII-V3 proviral DNA was found predominantly in liver with a mean of 5615 proviral copies/cell in the 410.sup.13 vg/kg cohort of mice at 8 weeks after gene transfer, followed by 2.11 copies/cell in the heart, 0.50.2 copies/cell in the spleen and kidney and 0.20.0.05 in the lungs. The detection limit of QPCR is 0.0003 copy/diploid genome.

(31) Materials and Methods

(32) AAV-hFVIII vector production and purification: The BDD deleted and N6 (kindly provided by Professor Steven Pipe (Miao et al, 2004))-human FVIII variants containing the wild type DNA sequences were cloned downstream of the previously described liver specific LP1 promoter (Nathwani et al, 2006). A 5012 bp codon optimized human N6 FVIII (codop-N6-hFVIII) was generated using codons most frequently found in highly expressed eukaryotic genes, (Haas et al, 1996) synthesized and also cloned downstream of the LP1 promoter. The smaller HLP enhancer/promoter was constructed by synthesizing a 251 bp fragment containing a 34 bp core enhancer from the human apolipoprotein hepatic control region (HCR) upstream of a modified 217 bp alpha-1-antitrypsin (hAAT) gene promoter consisting only of the distal X and the proximal A+B regulatory domains. AAV-HLP-codop-N6-hFVIII was generated by cloning the codop-N6-hFVIII cDNA downstream of the HLP promoter but upstream of a 60 bp synthetic polyadenylation signal. The AAV-HLP-codop-FVIII variants 1 and 3 were made by synthesis of a 1485 and a 1446 bp fragment, respectively. HLP-codop-N6-FVIII was cut with KpnI and the 2028 bp fragment was replaced with the synthesised fragments cut with KpnI. AAV vectors were made by the adenovirus free transient transfection method described before (Davidoff et at, 2004). AAV5 pseudotyped vector particles were generated using a chimeric AAV2 Rep-5Cap packaging plasmid called pLT-RCO3 which is based on XX2 (Xiao et al, 1998) and pAAV5-2 (Chiorini et al, 1999) and similar in configuration to that described before (Rabinowitz et al, 2002). AAV8 pseudotyped vectors were also made using the packaging plasmid pAAV8-2 (Gao et al, 2002). AAV2/5 and 2/8 vectors were purified by the previously described ion exchange chromatography method (Davidoff et al, 2004). Vector genome (vg) titers were determined by previously described quantitative PCR and gel based methods (Nathwani et al, 2001), (Fagone et al, 2012). To determine the size of the packaged genome, vector stocks were run on an alkaline gel as previously described in Fagone et al, 2012.

(33) Animal studies: All procedures were performed in accordance with institutional guidelines under protocols approved by the Institutional and/or National Committees for the care and use of animals in the United States and Europe. FVIII-deficient mice (mixed C57B16/J-129 Sv background with a deletion in exon 16) were bred in-house and used for experiments between 8 and 10 weeks of age. Tail vein administration of rAAV vector particles was performed in 7-10 week old male mice as described before (Nathwani et al. 2001).

(34) Determination of Transduction Efficiency and Vector Biodistribution:

(35) Human FVIII ELISA: Human FVIII antigen levels in murine samples were determined by ELISA using a paired FVIII ELISA kit (Affinity Biologicals, Quadratech, Dorking, UK). Flat-bottomed 96-well plates (NUNCMAXISORP, Fisher Scientific, Loughborough, UK) were coated with a combination of two mouse monocloncal antibodies (ESH2 (Sekisui Diagnostica, Axis-Shield, Dundee, UK), and N77110M (Biodesign international, AMS biotechnology, Abingdon, UK)) 50 l of a 100 g/mL in 50 mM carbonate buffer pH9.6 at 4 C. overnight, washed with PBS containing 0.05% TWEEN 20(=PBST), and blocked with 200 L/well of 6% bovine serum albumin (BSA, Sigma, Pool, UK) in PBST during a 1 hour incubation at 37 C. Standards were made by serial dilutions of murine plasma spiked with recombinant human FVIII, starting concentration 41 U/mL (11.sup.th BS 95/608 6.9 IU/mL, NIBSC, South Mimms). Murine samples and standards were diluted 1:10 in kit buffer with 50 l in duplicates. Following a 2 hour incubation at 37 C., the plates were washed and incubated for a further hour with 100 l of horseradish peroxidase conjugated goat anti-human FVIII polyclonal secondary antibody. After a final wash step, plates were developed with o-phenylenediamine dihydrochloride peroxidase substrate (Sigma) and the optical density was assessed spectrophotometrically at 492 nm. Probability of statistical difference between experimental groups was determined by one-way ANOVA and paired student t test using GRAPHPAD PRISMversion 4.0 software (GraphPad, San Diego, Calif.). FVIII activity was measured in a two-stage coagulation assay, using human plasma as a standard.

(36) Blood loss assay: Mice were anaesthetized with tribromoethanol (0.15 mL/10 g bodyweight) and 3 mm of the distal tail was cut with a scalpel. The tail was immersed immediately in 50 ml saline buffer at 37 C. and blood was collected for 30 min. Two parameters were monitored: First, time to arrest of bleeding was measured from the moment of transection. Second, collected erythrocytes were pelleted at 1500 g and lysed in H.sub.2O. The amount of released haemoglobin was determined by measuring the optical density at 416 nm and using a standard curve prepared upon lysis of 20-100 microliter of mouse blood.

(37) Quantification of vector copy number: Genomic DNA was extracted from murine tissues using the DNEASYBlood and tissue kit (Qiagen, Crawley, UK), 37 ng of genomic DNA extracted from various murine tissues was subjected to quantitativereal-time PCR using primers which amplified a 299 bp region of codop-hFVIII (5 primer: 5 AAGGACTTCCCCATCCTGCCTGG 3 and 3 primer: 5 GGGTTGGGCAGGAACCTCTGG 3) as described previously (Nathwani et al, 2011).

(38) Detection of anti-human FVIII antibodies: Plasma samples from mice were screened for the presence of antibodies against hFVIII using an ELISA. A 96 well MAXISORPplate (Nunc) was coated with 50 L of 2 IU/mL recombinant FVIII in 50 m carbonate buffer pH 9.6 at 4 C. overnight. Plates were washed with PBS-T and blocked with 3% BSA/TBS-T (25 mM Tris, 150 mM NaCI, 5 mM CaCl.sub.2, 0.01% TWEEN, p117.5). 50 L of serial dilutions of the plasma samples were prepared in 3% BSA/TBS-T. Following a 2 hour incubation at 37 C., the plates were washed and incubated for a further hour with 100 l of horseradish peroxidase conjugated goat anti-mouse IgG secondary antibody (A8924, Sigma). After a final wash step, plates were developed with o-phenylenediamine dihydrochloride peroxidase substrate (Sigma) and the optical density was assessed spectrophotometrically at 492 nm. Results were expressed as the end-point titer, defined as the reciprocal of the interpolated dilution with an absorbance value equal to five times the mean absorbance background value.

REFERENCES

(39) Altschul S F. Gish W, Miller W, Myers E W, Lipman D J. (1990) J Mol Biol, 215, 403-10. Altschul S F, Madden T L, Schffer A A, Zhang J, Zhang Z, Miller W, Lipman D J. (1997) Nucleic Acids Res, 25, 3389-402 Bantel-Schaal U, Delius H, Schmidt R, zur Hausen H. (1999) J Virol, 73, 939-47 Cerullo V, Seiler M P, Mane V, Cela R, Clarke C, Kaufman R J, Pipe S W, Lee B. (2007) Mol. Ther, 15, 2080-2087. Chao H, Sun L, Bruce A, Xiao X, Walsh C E. (2002) Mol. Ther, 5, 716-722. Chiorini J A, Yang L, Liu Y, Safer B, Kotin R M. (1997) J Viral, 71, 6823-33 Chiorini J A, Kim F, Yang L, Kotin R M. (1999) J Virol, 73, 1309-19 Chen L, Lu H, Wang J, Sarkar R, Yang X, Wang H, High K A, Xiao W. (2009) Mol. Ther, 17, 417-424. Davidoff A M, Ng C Y, Sleep S, Gray J, Azam S, Zhao Y, McIntosh J H, Karimipoor M, Nathwani A C. (2004) J Virol Methods, 12, 209-15. Fagone P, Wright J F, Nathwani A C, Nienhuis A W, Davidoff A M, Gray J T. (2012) Hum Gene Ther Methods, 23, 1-7. Gao G P, Alvira M R, Wang L, Calcedo R, Johnston J, Wilson J M. (2002) Proc Natl Acad Sci USA, 99, 11854-9 Haas J, Park E C, Seed B. (1996) Curr Biol, 6, 315-24. Jiang H, Pierce G F, Ozelo M C, de Paula E V, Vargas J A, Smith P, Sommer J, Luk A, Manno C S, High K A, Arruda V R. (2006) Mol Ther, 14, 452-455. Kaufman R J, Pipe S W, Tagliavacca L, Swaroop M, Moussalli M. (1997) Blood Coagul. Fibrinolysis, 8 Suppl 2, S3-14. Malhotra, J. D., Miao, H., Zhang, K., Wolfson, A., Pennathur. S., Pipe, S. W., & Kaufman, R. J. (2008) Proc. Natl. Acad. Sci. U.S.A, 105, 18525-18530. Miao C H, Thompson A R, Loeb K, Ye X. (2000) Mol Ther, 3, 947-57. Miao H Z, Sirachainan N, Palmer L, Kucab P, Cunningham M A, Kaufman R J, Pipe S W, (2004) Blood, 103, 3412-3419. Muyldermans S. (2001) J Biotechnol, 74, 277-302 Nathwani A C. Davidoff A, Hanawa H, Zhou J F, Vanin E F, Nienhuis A W. (2001) Blood, 97, 1258-65. Nathwani A C, Gray J T, Ng C Y, Zhou J, Spence Y, Waddington S N, Tuddenham E G, Kemball-Cook G, McIntosh J, Boon-Spijker M, Mertens K, Davidoff A M. (2006) Blood, 107, 2653-61. Nathwani A C, Rosales C, McIntosh J, Rastegarlari G, Nathwani D, Raj D, Nawathe S, Waddington S N, Bronson R, Jackson S, Donahue R E, High K A, Mingozzi F, Ng C Y, Zhou J, Spence Y, McCarville M B, Valentine M, Allay J, Coleman J, Sleep S, Gray J T, Nienhuis A W, Davidoff A M. (2011) Mol Ther, 19, 876-85. Needleman S B, Wunsch C D. (1970) J Mol Biol, 48, 443-53. Okuyama T, Huber R M, Bowling W, Pearline R, Kennedy S C, Flye M W, Ponder K P. (1996) Hum Gene Ther, 7, 637-45. Rabinowitz J E, Rolling F, Li C, Conrath H, Xiao W, Xiao X, Samulski R J. (2002) J Virol, 76, 791-801. Rutledge E A, Halbert C L, Russell D W. (1998) J Virol, 72, 309-19. Srivastava A, Lusby E W, Berns K I. (1983) J Virol, 45, 555-64. Wang L, Takabe K, Bidlingmaier S M, Ill C R, Verma I M. (1999) Proc Natl Acad Sci USA, 96, 3906-10. Wu P, Xiao W, Conlon T, Hughes J, Agbandje-McKenna M, Ferkol T, Flotte T, Muzyczka N. (2000) J Virol, 74, 8635-47. Xiao X, Li J, Samulski R J. (1998) J Virol, 72, 2224-32.