TRANSPOSON, GENE TRANSFER SYSTEM AND METHOD OF USING THE SAME

Abstract

The present invention refers to hyperactive variants of a transposase of the transposon system Sleeping Beauty (SB). The invention further refers to corresponding nucleic acids producing these variants, to a gene transfer system for stably introducing nucleic acid(s) into the DNA of a cell by using these hyperactive variants of a transposase of the transposon system Sleeping Beauty (SB) and to transposons used in the inventive gene transfer system, comprising a nucleic acid sequence with flanking repeats (IRs and/or RSDs). Furthermore, applications of these transposase variants, the transpsoson, or the gene transfer system are also disclosed such as gene therapy, insertional mutagenesis, gene discovery (including genome mapping), mobilization of genes, library screening, or functional analysis of genomes in vivo and in vitro. Finally, pharmaceutical compositions and kits are also encompassed.

Claims

1-45. (canceled)

46. A method of inserting a transposon into the nucleic acid of a cell, comprising introducing into said cell a nucleic acid encoding a transposase and a nucleic acid encoding a transposon; wherein said transposase is a sleeping beauty 10 (SB10) polypeptide variant of SEQ ID NO: 1, wherein said variant has transpositional activity at least twice the transpositional activity of SEQ ID NO: 1 wherein the amino acid sequence of said SB10 polypeptide variant differs from SEQ ID NO:1 by 1 to 20 amino acids and wherein one mutation is K14R; and wherein said transposon comprises at least two repeats recognized by the transposase, wherein the repeats are inverted repeats (IRs) and/or inverted repeats/direct repeats (IR/DRs), and the transposase mediates the insertion of the transposon into the nucleic acid of the cell.

47. The method of claim 46, wherein the variant comprises 2 to 20 mutations.

48. The method of claim 46, wherein the SB10 polypeptide variant is a combination of mutations selected from the group consisting of: Variant 1: K14R/R214D/K215A/E216V/N217Q; Variant 3: K14R/K30R/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M24 3H; Variant 7: K14R/T83A/M243Q; Variant 8: K14R/T83A/I100L/M243Q; Variant 9: K14R/T83A/R143L/M243Q; Variant 10: K14R/T83A/R147E/M243Q; Variant 11: K14R/T83A/M243Q/E267D; Variant 12: K14R/T83A/M243Q/T314N; Variant 13: K14R/K30R/I100L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H; Variant 14: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H; Variant 15: K14R/K30R/R147E/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H; Variant 16: K14R/K30R/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/E267D; Variant 17: K14R/K30R/A205K/H207V/K208R/D210E//R214D/K215A/E216V/N217Q/M243H/T314N; Variant 18: K14R/K30R/A205K/H207V/K208R/D210E//R214D/K215A/E216V/N217Q/M243H/G317E; Variant 19: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H; Variant 20: K14R/K30R/R147E/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/T314N; Variant 21: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/E267D; Variant 22: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/T314N; Variant 23: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/G317E; Variant 24: K14R/K33A/R115H/R143L/R214D/K215A/E216V/N217Q/M243H; Variant 25: K14R/K33A/R115H/R147E/R214D/K215A/E216V/N217Q/M243H; Variant 26: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/E267D; Variant 27: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/T314N; Variant 28: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/G317E; and Variant 29: K14R/T83A/M243Q/G317E.

49. The method of claim 46, wherein the variant further comprises mutations R214D/K215A/E216V/N217Q.

50. The method of claim 49, wherein the variant is a combination of mutations selected from the group consisting of: Variant 1: K14R/R214D/K215A/E216V/N217Q; Variant 3: K14R/K30R/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M24 3H; Variant 13: K14R/K30R/I100L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H; Variant 14: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H; Variant 15: K14R/K30R/R147E/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H; Variant 16: K14R/K30R/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/E267D; Variant 17: K14R/K30R/A205K/H207V/K208R/D210E//R214D/K215A/E216V/N217Q/M243H/T314N; Variant 18: K14R/K30R/A205K/H207V/K208R/D210E//R214D/K215A/E216V/N217Q/M243H/G317E; Variant 19: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H; Variant 20: K14R/K30R/R147E/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/T314N; Variant 21: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216 V/N217Q/M243H/E267D; Variant 22: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/T314N; Variant 23: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/G317E; Variant 24: K14R/K33A/R115H/R143L/R214D/K215A/E216V/N217Q/M243H; Variant 25: K14R/K33A/R115H/R147E/R214D/K215A/E216V/N217Q/M243H; Variant 26: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/E267D; Variant 27: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/T314N; and Variant 28: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/G317E.

51. The method of claim 46, wherein the variant comprises at least mutations R214D/K215A/E216V/N217Q and K14R; and 2 to 6 additional mutations or groups of mutations selected from the group consisting of: K30R, K33A, R115H, R143L, R147E, A205K/H207V/K208R/D210E; M243H; E267D; T314N; and G317E.

52. The method of claim 51, wherein the variant comprises a combination of mutations selected from the group consisting of mutations: Variant 3: K14R/K30R/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M24 3H; Variant 14: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H; Variant 15: K14R/K30R/R147E/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H; Variant 16: K14R/K30R/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/E267D; Variant 17: K14R/K30R/A205K/H207V/K208R/D210E//R214D/K215A/E216V/N217Q/M243H/T314N; Variant 18: K14R/K30R/A205K/H207V/K208R/D210E//R214D/K215A/E216V/N217Q/M243H/G317E; Variant 19: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H; Variant 20: K14R/K30R/R147E/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/T314N; Variant 21: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/E267D; Variant 23: K14R/K30R/R143L/A205K/H207V/K208R/D210E/R214D/K215A/E216V/N217Q/M243H/G317E; Variant 24: K14R/K33A/R115H/R143L/R214D/K215A/E216V/N217Q/M243H; Variant 25: K14R/K33A/R115H/R147E/R214D/K215A/E216V/N217Q/M243H; Variant 26: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/E267D; Variant 27: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/T314N; and Variant 28: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/G317E.

53. The method of claim 46, wherein the variant includes at least the following group of mutations: R214D/K215A/E216V/N217Q, and K14R; and 3 to 4 additional mutations selected from the group consisting of: K33A, R115H, R143L, R147E, M243H; E267D; T314N; G317E; and 0 or 1 additional mutation selected from the group consisting of: R143L, R147E, E267D; T314N; and G317E.

54. The method of claim 53, wherein the variant is a combination of mutations selected from the group consisting of: Variant 19: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H; Variant 24: K14R/K33A/R115H/R143L/R214D/K215A/E216V/N217Q/M243H; Variant 25: K14R/K33A/R115H/R147E/R214D/K215A/E216V/N217Q/M243H; Variant 26: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/E267D; Variant 27: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/T314N; and Variant 28: K14R/K33A/R115H/R214D/K215A/E216V/N217Q/M243H/G317E.

55. The method of claim 46, wherein the variant has transpositional activity at least ten times the transpositional activity of SEQ ID NO: 1.

56. The method of claim 46, wherein said cell is a lymphocyte.

57. The method of claim 46, wherein said cell is derived from the hematopoietic system.

58. The method of claim 48, wherein said cell is a T cell.

59. The method of claim 46, wherein the nucleic acid encoding said transposase is a mRNA.

60. The method of claim 46, wherein the nucleic acid encoding said transposase is a DNA.

61. A method for producing a protein encoded by a transposon, comprising: i) providing a gene transfer system comprising: a) a polypeptide variant of sleeping beauty 10 (SB10) transposase or an isolated nucleic acid encoding a variant of SB10 transposase, wherein said variant comprises an amino acid sequence differing from the sequence of SB10 transposase of SEQ ID NO: 1 by 1-20 amino acids and at least the following mutation: K14R; and b) a transposon comprising a nucleic acid sequence positioned between at least two repeats, wherein the repeats are inverted repeats (IRs) and/or inverted repeats/direct repeats (IR/DRs), wherein the repeats can bind to the SB10 polypeptide variant, wherein the protein is encoded by the nucleic acid sequence positioned between the at least two repeats; ii) providing a target mammalian cell; iii) transfecting the target mammalian cell with the gene transfer system to produce a transfected target mammalian cell; and iv) expressing the protein.

Description

DESCRIPTION OF FIGURES

[0422] FIG. 1. Scheme of a Class II cut-and-paste transposable element (TE), the binary transposition system created by dissecting the transposase source from the transposon, and its transposition. ITR, inverted terminal repeat. TEs are moving in the host genome via a “cut-and-paste” mechanism. The mobile DNA elements are simply organized, encoding a transposase protein in their simple genome flanked by the inverted terminal repeats (ITR). The ITRs carry the transposase binding sites necessary for transposition. Their activities can easily be controlled by separating the transposase source from the transposable DNA harboring the ITRs, thereby creating a non-autonomous TE. In such a two-component system, the transposon can only move by transsupplementing the transposase protein. Practically any sequence of interest can be positioned between the ITR elements according to experimental needs. The transposition will result in excision of the element from the vector DNA and subsequent single copy integration into a new sequence environment.

[0423] FIG. 2A-2B. A part of the protein alignment of the Tel transposase sequences along the whole Tel family with no respect of the similarity to SB. FIG. 2(A) Part of the alignment with one picked hyperactive AA substitution as an example. FIG. 2(B) Similarity tree of the alignment.

[0424] FIG. 3A-3B. A part of the protein alignment of the Tel transposase sequences more related to SB. FIG. 3(A) Part of the alignment with two picked hyperactive M substitutions as examples. FIG. 3(B) Similarity tree of the alignment.

[0425] FIG. 4. Hyperactive mutations forming the base for the shuffling, and their grouping to the particular restriction digestions for reducing the wt sequence content.

[0426] FIG. 5. DNasel treated isolated fragment populations (lane 1 and 2) run on a 12% poly-acrylamide gel. M, marker

[0427] FIG. 6A-6B FIG. 6(A) PCR reassembly reaction (lane 1); M, marker. FIG. 6(B) Final PCR step for cloning of the full length CDS on the diluted PCR reassembly reaction template. Lane 1, forward and reverse cloning primers are added; lane 2, forward cloning primer is added; lane 3, reverse cloning primer is added; M, marker.

[0428] FIG. 7. Distribution of clone classes in the unselected libraryl.

[0429] FIG. 8A-8B. FIG. 8(A) Mutational participation of the 7 most hyperactive clones isolated from the shuffling library. FIG. 8(B) Particular statistical features of the selected hyperactive clones of the library.

[0430] FIG. 9. Summary of our strategy for the manual improvement of the hyperactive clones harvested from the shuffling library.

[0431] FIG. 10: FIG. 10 shows the amino acid sequence of SB10 (SEQ. ID. No. 1).

[0432] FIG. 11. construction of the vectors used for e.g. the experiments shown in FIGS. 14 to 26

[0433] FIG. 12. overview of the non-hyperactive transposases; SBl 0 is the wild type transposase, SB 11 and SB DNGP are other non-hyperactive or inactive transposases having mutations over SB 10 as indicated and used herein for comparative reasons. SBl 0 was originally published by Ivies et al. (1997), Cell 91: 501-510 (FIG. 10), while SB 11 was originally published by Geurts et al. (2003), Mol. Therapy 8: 108-117. SB 11 contains the mutations T136R, M243Q, VVA253HVR.

[0434] FIG. 13. overview of the hyperactive mutant transposases SB M3a (containing the mutations K13A, K33A, T83A and R214D/K215NE216V/N217Q over SB10, Variant 30, table II), SB 3O5-Kl 4R (Variant 19, table II), SB/6A5 (Variant 3, table II), SB 100×(Variant 27, table II), all of them derived from SB10. Preferred variants are derived from the sequence of SB10 with the mutations indicated.

[0435] FIG. 14. comparative analysis of hyperactive transposases SB M3a, SB6/AS versus non-hyperactive SB10 and SB11 in erythroid lineage

[0436] FIG. 15. comparative analysis of hyperactive transposases M3a, SB6/AS and SB 3O5-K14R in erythroid lineage

[0437] FIG. 16. comparative analysis of hyperactive transposases SB6/AS and SB100X in erythroid lineage

[0438] FIG. 17. comparative analysis of hyperactive transposases SB6/AS and SB1OOX in megakaryotic lineage

[0439] FIG. 18. comparative analysis of hyperactive transposases SB6/AS and SB100X in granulocyte/macrophage/monocyte lineage

[0440] FIG. 19. relative gene transfer efficiency of transposase SB100X as compared with the hyperactive transposase SB6/AS

[0441] FIG. 20. relative gene transfer activity of the mutant hyperactive transposases SB M3a, SB6/AS, SB3 O5-K14R and SB100X, compared to wi Id type transposase SB10 and the mutant non-hyperactive transposase SB11

[0442] FIG. 21. Stable gene transfer efficiency in human muscle progenitor/stem cells using mutant hyperactive transposase SB M3a

[0443] FIG. 22. Comparative analysis of gene transfer in human muscle progenitor/stem cells using mutant hyperactive transposase SB M3a or SB6/AS, compared with wild type transposase SB10 or mutant non-hyperactive transposase SB11

[0444] FIG. 23. Comparative analysis of hyperactive transposase SB100x, SB 3DS-K14R, SB M3a versus non-hyperactive transposase SB11 in muscle progenitor cells.

[0445] FIG. 24. Levels of Factor IX expression in human muscle progenitor cells following stable transfection using the mutant hyperactive transposase SB M3a

[0446] FIG. 25. Levels of in vivo expression of Factor IX in liver of mice after transfection using the mutant hyperactive SB 100X transposase, in comparison with the mutant non-hyperactive SB11 transposase, or an inactive control.

[0447] FIG. 26. Levels of in vivo expression of Factor IX in liver of mice, after transfection using the mutant hyperactive SB 1 00X transposase, showing a stable expression after partial hepatectomy.

EXAMPLES

[0448] Description of the Experimental Strategy [0449] I.) Collecting hyperactive mutations within the SB transposase coding sequence (CDS) for the shuffling. [0450] a) Mutagenesis through the whole SB CDS. [0451] b) Selection of hyperactives using an activity test-system. [0452] II.) In vitro recombination of the selected mutants by DNA shuffling. [0453] a) Isolation of the point mutations on 100-300 by fragments. [0454] b) DNasel breakage to 30-70 bp fragments. [0455] c) PCR shuffling and cloning of the library. [0456] d) Sequencing the library. [0457] III.) Searching for clones exhibiting high transpositional activity. [0458] a) Large scale purification of shuffling clones. [0459] b) Test of the library clones for transpositional activity in Hela cells. [0460] c) Manual creation of promising new combinations based on the sequencing data of the selected hyperactive clones

[0461] In all tests for activity as a transposase described here SB10 (Ivies, Z., Hackett, P. B., Plasterk, R. H. and Izsvak, Zs. (1997) Molecular reconstruction of Sleeping Beauty, a Tel-like transposon from fish, and its transposition in human cells. Ce//91:501-510) was used as a comparator. [0462] I.a) Mutagenesis through the whole SB coding sequence. [0463] The Tel family of transposons is the biggest transposon family representing a lot of related sequences available for comparison. As a first step a number of single M substitutions were designed. A range of related transposase sequences were aligned to find promising positions to be changed in the SB CDS (coding sequence). Although emphasis was laid on getting the new M from known active sequences, also sources with no information of their activity and some known inactive sequences were used, too. So, the transposase CDSs of other known related Tel transposones (FIG. 2.), most of which are coding for active transposases, were aligned with SB10, followed by a second alignment with a range of other Tel transposase CDSs more closely related to the SB transposase sequence (compare FIG. 2B and FIG. 3B). FIG. 1A and FIG. 2A demonstrate some examples of M substitution design using the first and the second alignments respectively. [0464] I.b) Selection of hyperactives using an activity test-system. [0465] The transpositional activity of all the mutations created was tested using the classical binary transposition assay (Ivies, 1997, see above). This test was the standard test for transposase activity used here. The scheme of the two component system is depicted on Figl. Briefly, Hela cells were cotransfected with the transposon vector carrying the neomycine resistance gene (NeoR) between the SB inverted repeats (pTNeo), and with the Polypeptide (transposase variant) expressing plasmid vector where the expression of the mutant SB transposases was driven by the CMV promoter. Following transfection it was selected for two weeks with G418 administration for the integration events of the NeoR transposon into the Hela cells genome. Finally the G418 resistant colonies were stained and counted. SBlO transposase CDS were used as a control to adjust the threshold level of activity and an inactive version of the SB transposase as a negative control. The tests were performed as duplicates on 12 well tissue culture plate formats. [0466] All the polypeptides (all single mutations) according to the invention (transposase variants) causing at least 200% hyperactivity compared to SB10 in the above assay were selected for further use in the shuffling experiment below. The hyperactivity of these variants was typically between 200-400% compared to SB10. [0467] II.a) Isolation of the point mutations on 100-300 bp fragments. [0468] The PCR shuffling method originally published by Stemmer W. P. C. in 1994 (Stemmer, W. P. C. (1994) DNA shuffling by random fragmentation and reassembly: In vitro recombination for molecular evolution. Proc. Natl. Acad Sci. 91:10747-10751) is a suitable method for mixing related parental. All the hyperactive mutations on smaller parts of the transposase CDS were isolated. The isolated fragments were broken to a 30-70 bp fragment population by DNasel to facilitate high recombination rates. [0469] 41 single hyperactive mutations were collected to combine them in DNA shuffling (FIG. 4). The particular mutations on smaller fragments of the CDS were isolated using restriction endonucleases to reach higher average mutation number/clone. The fragment sizes and the groups of mutations isolated by the same digestions are summarized on FIG. 4. At the 5′ and 3′ ends of the CDS some extra flanking DNA to the fragments included to allow rebuilding the full length CDS in the shuffling. The predicted average number of mutations pro clone (see FIG. 4.) was calculated to be about 4 in the case of this particular library (FIG. 4.). [0470] I1.b) DNasel breakage to 30-70 bp fragments [0471] Next the fragments were broken in a random fashion by DNasel digestions. The similarly sized fragments were treated in groups taking care for their same ratio in the total population. Then the mixtures of broken DNA molecules carrying the mutations were run on 12% acrylamide gels and the 30-70 bp populations of fragments were isolated. An example of the isolated fragment populations is presented on FIGS. [0472] I1.c) PCR shuffling and cloning of the library [0473] The isolated fragment populations were shuffled to reassemble the SB transposase CDS. Approximately the same amount of all individual mutations was used in the PCR reassembly reaction. As non-overlapping restriction fragment populations for narrowing the CDS around the mutations were used the addition of bridging oligos (for sequence see connect1-3 on Table1) was also necessary to connect the neighboring fragment groups to finally get the full length SB transposase CDS. The PCR reassembly reaction was done similarly to Stemmer, 1994, (see above). Briefly, the isolated 30-70 bp fragment populations of all the selected hyperactive mutations were added in the same ratio into the PCR reaction. The final concentration of DNA in the mixture was about 20 ngtμl. Further 2 pmol of each bridging oligos (see Table1.) was added. High-fidelity polymerase was used to minimize the introduction of further mutations created by the PCR reaction itself. The program for the PCR reassembly was the following: 1) 94 C.°-60 sec, 2) 94 C.°-30 sec, 3) 50 C.°-30 sec, 4) 68 C.°-1 min, 5) 68 C.°-5 min, and 40 cycles has been made of the 2-4 steps. The transposase CDS reassembly to the higher molecular weight was nicely visible after 40 cycles (FIG. 6A.). As the next step a second PCR reaction was carried out with the SB cloning primers SBclnfw and SBclnrev (see sequence on Table1.) using the 40× diluted assembly reaction as a template, to amplify the full length transposase CDS. The full length CDS (1023 bp) was amplified using the forward and reverse cloning primers together, in contrast to the situation when theses were added alone (FIG. 6B). The forward primer carried the recognition site of the endonuclease Spel. and a Kozak sequence while the reverse one carried an ApaI. recognition site, besides both of the primers beard 26 bp of the very ends of the transposase CDS. This gave the possibility to efficiently clone the isolated 1023 bp product pool of the second PCR reaction into a suitable vector designed and created for the purposes of the library (data not shown) digested with the same enzymes.

TABLE-US-00001 TABLE I Oligonucleotides used for the creation of the shuffling library. Connect1 5′ gtaccacgttcatctgtacaa acaatagtacgcaagtataa 3′ Connect2 5′ cgacataagaaagccagactac ggtttgcaactgcacatgggg 3′ Connect3 5′ atattgaagcaacatctcaaga catcagtcaggaagttaaagcttgg tcg 3′ SBclnfw 5′ ggtcactagtaccatgggaaaa tcaaaagaaatcagcca 3′ SBclnrev 5′ ggtcgggcccctagtatttggt agcattgcctttaa 3′ [0474] 11.d) Sequencing the library. [0475] As a next step the library of the shuffling clones (see Ile) were characterized. The library was transferred into E. Coli DH5 competent cells, then isolated and 45 reassembled CDS fully sequenced. It was found that all the 45 CDS were full length without insertions or deletions and moreover only a very low incidence of extra mutations were observed. Only 2 specific nucleotide positions in the 1023 bp long CDS were found, where typical point mutations were inserted by the shuffling process itself into some of the clones. None of them caused M change, and they remained silent on the protein level. After aligning the sequences to the S810 transposase CDS the clonal distribution of the 41 mutations taken into the shuffling were identified. The incidence of the mutations was fairly statistical in the unselected library, 31 of the 41 mutations introduced into the shuffling in the 45 sequenced clones were identified (data not shown). However, the average number of mutations/clone was only about 2 mutations in contrast to the prediction (see above; FIG. 4.). The reason for this is possibly the 30-70 bp length of the fragments carrying the individual mutations in the shuffling. The majority of the 41 mutations were separated from their neighboring mutations along the transposase CDS by less then 70 bp long sequences. As a consequence in the shuffling reassembly reaction the 30-70 bp fragments could partially exclude each other from a given chain elongation reaction, thereby decreasing the recombination rate between the neighboring mutations. 2 libraries were created with slight modifications in the shuffling setup (data not shown) and sequenced 23 and 22 clones of library 1 and library 2 respectively. Library 1 had 2.2 mutations/clone while library 2 had only 1.8, so library 1 was used for the further experiments. The clonal distribution of mutations in library 1 is shown on FIG. 7. [0476] III.a) Large scale purification of shuffling clones. [0477] The cell culture system described above was used for the activity tests. A large scale automated purification of plasmid DNA of the shuffling clones was done using a pipetting robot and a plasmid kit. The plasmid preparations were producing fairly similar yields and their quality was tissue culture compatible. All the plasmid samples were run on agarose gel to verify their similar concentrations and quality. Plasmid DNA of about 2000 clones was purified. [0478] 111.b) Test of the library clones for transpositional activity in Hela cells. [0479] The clones were tested in transposition assays in HeLa cells as described in Ib) above with the difference that 96 well formats were used. All the tests were done as duplicates. For reference S816 (Baus, 2005) was used on all the plates. All the clones that showed similar or higher activity compared to S81 6 on the duplicated 96 well test plates were chosen for further operations. Further the activity of the best 20 clones on 12 well formats were verified. 7 (Variants 1 to 7) of the 20 retested clones showed clearly higher activity compared to S816. The best 2 clones (Variants 2 and 3) exhibited about 2 times higher activity than S816 which means about 30 times higher activity compared to S810. [0480] 111.c) Manual creation of promising new combinations based on the sequencing data of the selected hyperactive clones [0481] The best 20 clones retested on 12 well format and also 18 other clones still showing high activity in the range of S816, (thus 38 clones all together), were fully sequenced to collect a data pool. The mutational content of the best 7 clones is shown in FIG. 8A. The combinations 3D5 and 6A5 (variants 2 and 3) proved to be the best showing 30-32 times higher activity compared to SB10. By analyzing the sequencing data pool of all 38 active clones it was observed that (i) the mutation number/clone is growing with the activity and it reaches the 3,6 mutations/clone as average in the group of the most hyperactive 7 clones (FIG. 8B). Moreover, (ii) it was also realized that the incidence of some of the mutations is increasing among clone groups parallel to increased transpositional activity of the groups. The most obvious example for this was the increasing incidence of the 214DAVQ mutation (FIG. 8B). Moreover, this particular mutation appeared as the core mutation of most of the hyperactive combinations reaching or exceeding the activity range of S816. [0482] Among the 38 sequenced hyperactive clones only 8 containing 4 mutations and 5 containing 5 mutations were found. This means 21% and 13% incidence of 4 and 5 mutation carrying clones, respectively, in the selected library. No clones were identified bearing more than 5 mutations. Moreover, among the 7 best variants already 4 carried 4 or 5 mutations (see FIG. 8). In the unselected library the incidence of clones having 4 mutations was less then 10%. Thus, the hyperactivity in the range of S816 really correlates well with bearing 4 or 5 mutations/clone. [0483] After analyzing the sequencing data of the 38 most hyperactive shuffling clones 3 clones were chosen for further mutagenesis: 3D5, 6A5 and 12B1 (variants 2, 3 and 7) (FIG. 9). Also based on the sequencing data 6 “friendly” mutations were identified (FIG. 9) with the hope to successfully combining them to the 3 chosen clones. The resultant combinations and their transpositional activity were measured on 12 well formats are shown on FIG. 9 and listed in Table II. In addition, a particular clone (variant 1) was exceptionally bearing only 2 mutations. Two of these clones were identical (see 2G6 and 6G2 on FIG. 8). This exceptional combination of the K14R and the 214DAVQ mutations was obviously not simply additive in terms of hyperactivity but it was rather a multiplier combination. Based on this observation the Kl 4R mutation was introduced into the best 3D5 combination, by which the resultant clone containing both mutations Kl 4R and 214 DAVQ. The established clone (variant 19) showed highly enhanced activity (FIG. 9).

Overview of Transposase Activity of Tested Variants {Table II)

[0484]

TABLE-US-00002 TABLE II Activity compared to SB10 Variant (with mutation pattern) (factor) Variant 1: Kl 4R/R21 4D//K21 SNE216V/N21 7Q; −20 Variant 2: K33NR1 15H//R214D/K215NE216V/N217Q//M243H; −30 Variant 3: KI 4R/K30R//A205K/H207V/K208R/D21OE// −30 R214D/K215N E216V/N217Q//M243H; Variant 4: KI 3D/K33Nf83N/H207V/K208R/D21OE//M243Q; −20 Variant 5: KI 3NK33N/R214D/K215NE216V/N217Q; −20 Variant 6: K33Nf83N/R214D/K215NE216V/N217Q//G317E; −20 Variant 7: Kl 4R/f83NM243Q; −15 Variant 8: K14R/f83Nl100UM243Q; −5 Variant 9: K14R/f83NR143UM243Q; 20-30 Variant 10: Kl 4R/f83NR147E/M243Q; 20-30 Variant 11: KI 4R/f83NM243Q/E267D; 15-20 Variant 12: K14R/T83NM243Qfr314N; −10 Variant 13: Kl4R/K30R/11 OOU/A205K/H207V/K208R/D210E//R214D/ K215NE216V/N217Q//M243H; Variant 14: Kl4R/K30R/R143U/A205K/H207V/K208R/D21 OE//R214D/ −40 K215NE216V/N217Q//M243H; Variant 15: Kl 4R/K30R/Rl 47E//A205K/H207V/K208R/D210E//R214D/ −30 K215NE216V/N217Q//M243H; Variant 16: Kl 4R/K30R//A205K/H207V/K208R/D21 OE//R214D/K215N −30 E216V/N217Q//M243H/E267D; Variant 17: Kl 4R/K30R//A205K/H207V/K208R/D21OE//R214D/K215N −25 E216V/N217Q//M243Hfr314N; Variant 18: Kl 4R/K30R//A205K/H207V/K208R/D21OE//R214D/K215N −25 E21 6V/N217Q//M243H/G31 7E; Variant 19: K14R/K33NR115H//R2140/K21SNE216V/N217Q//M243H; 70-80 Variant 20: KI 4R/K30R/Rl 47E//A205K/H207V/K208R/D210E//R2l 4D/ −40 K215N E216V/N217Q//M243Hfr314N; Variant 21: Kl4R/K30R/RI 43U/A205K/H207V/K208R/D21OE//R214D/ −50 K215N E216V/N217Q//M243H/E2670; Variant 22: KI 4R/K30R/Rl 43U/A205K/H207V/K208R/D21OE//R214D/ K21 SN E216V/N217Q//M243Hfr314N; Variant 23: KI 4R/K30R/Rl 43U/A205K/H207V/K208R/D21OE//R214D/ −35 K215N E216V/N217Q//M243H/G317E; Variant 24: KI 4R/K33NR115H/R143U/R214D/K215NE216V/N217Q// 70-80 M243H; Variant 25: KI 4R/K33NR115H/R147E//R214D/K215NE216V/N217Q// 70-80 M243H; Variant 26: Kl 4R/K33NR1 1 5H//R214D/K215NE216V/N217Q//M243H/ 70-80 E267D; Variant 27: KI 4R/K33NR1 1 5H//R214D/K215NE216V/N217Q//M243H/ 90-100 T314N; Variant 28: KI 4R/K33NR115H//R2l 4D/K215NE216V/N2l 7Q//M243H/ 80-90 G317E; Variant 29: KI 4Rff83NM243Q/G31 7E; Variant 30: KI 3NK33Aff83N/ R214D/K215NE216V/N217Q −10

[0485] Further examples IV to IX were carried out with the object to determine the activity of various hyperactive transposase mutants as compared to non-hyperactive or inactive mutants (control experiments) in cell lines of various lineages (see FIGS. 14 to 26). The conditions used for these Examples are given in the following.

[0486] Description of the Experimental Strategy

[0487] Materials and Methods for Examples IV to IX

[0488] A) Sleeping Beautytransposon System [0489] The SB transposon system is a binary system composed of (i) the inverted repeat/direct repeats (IR/DR) flanking the gene of interest, and (ii) the expression cassette encoding the transposase. Different transposons containing the gene of interest and different transposases were used in this study (see also above). [0490] 1.) SB transposon-based vectors [0491] (i) pT2-HB-CAG-GFP [0492] The pT2-HB-CAG-GFP transposon is a SB transposon vector in which the GFP reporter gene is transcriptionally regulated by the CAG promotor. The CAG promotor is a chimeric promoter composed of the CMV (human cytomegalovirus) immediate early enhancer in conjunction with the chicken b-actin/rabbit-b-globin hybrid promoter and intron (CAG) http://www.belspo.be/bccm/Imbp.htm; LMBP 2453); (pA: polyadenylation signal) (FIG. 11). [0493] (ii) pT2-HB-CMV-FIX-neo [0494] The pT2-HB-CMV-FIX-neo transposon is a SB transposon vector in which the human coagulation factor IX cDNA (FIX) is driven by the CMV promoter. The vector also contains a Simian Virus 40 (SV40) promoter driving a neomycin resistance gene (NeoR) that confers resistance to G418 (Geneticin) in stably transfected eells (FIG. 11). [0495] (iii) pT2-HB-CMV-GFP-neo [0496] The pT2-HB-CMV-GFP-neo transposon is a SB transposon vector in which the GFP reporter gene is transcriptionally regulated by the CMV promotor. The vector also contains a SV40 promoter driving a neomycin resistance gene (NeoR) that confers resistance to G418 (Geneticin) in stably transfected cells (FIG. 11). [0497] (iv) pT2-HB-Apo/MT-FIX [0498] The pT2-HB-Apo/MT-FIX transposon is a SB transposon vector in which the FIX cDNA was driven from the ApoE HCR/AAT promoter composed of the apolipoprotein E enhancer/al-antitrypsin promoter, the hepatocyte control region (HCR) and the first FIX intron (kindly provided by Dr. Miao, University of Washington) (FIG. 11). [0499] 2.) Transposases [0500] All transposases (active, inactive of hyper-active) are encoded by a CMV expression plasmid and contain different mutations in the DNA binding domain, catalytic domain or both (FIGS. 12 & 13) compared to the originally reconstructed SB10 Sleeping Beauty transposase. The SB-DNGP encodes an inactive SB transposase due to the deletion of the DDE catalytic domain. To generate the SB GFP plasmid, the SB100x transposase (Variant 27) was replaced with GFP.

[0501] B) Cells [0502] Umbilical cord blood (UCB) mononuclear cells were separated from UCB over Ficoll/Hypaque by centrifugation at 2400 rpm for 30 min at 20° C., then washed with PBS containing 2 mM EDTA and centrifuged twice at 1000 rpm for 10 min. The CD34+ cells were further enriched by immunomagnetic separation according to the manufacturers instructions (Miltenyi Biotech Inc. CA, USA) using magnetic beads conjugated to anti-CD34 antibodies. This immunomagnetic cell separation typically yielded >95% CD34+ cells which are enriched for hematopoietic stem/progenitor (HSC) cells. [0503] Primary human skeletal muscle stem/progenitors cells (myoblasts) were obtained by needle biopsy.sup.5 from the vastus lateralis muscle of volunteers. Myoblasts were expanded in SkGM medium, as described by the manufacturer (Cambrex Bio Science, MD USA).

[0504] C) Mice [0505] C57Bl/6 mice were hydrodynamically transfected with 50 micrograms of transposon with 25 μg of transposase plasmid diluted in 2 ml of PBS and injected into the tail vein. Typically, the injection took less than 10 seconds for each mouse and is results in efficient hepatic gene delivery.

[0506] D) Transfection [0507] Nucleofection of CD34+ HSCs was done according to the optimized protocol for human CD34+ cells using the nucleofection kit developed by Aamaxa Biosystems (Amaxa Biosystems, Cologne Germany). The U-01 program was employed using the Amaxa electroporation device (Nucleofector I, Cologne Germany). Enriched CD34+ cells in PBS were centrifuged at 1200 rpm for 10 min and re-suspended in Nucleofector buffer. Typically, 1.5×10.sup.5 cells in 100 microliter of human CD34 cell Nucleofector buffer (Amaxa Biosystems, Cologne Germany) per cuvette were subjected to electroporation with purified plasmids containing the transposon (10 microgram) and transposase (5 microgram) (concentration: 1 microgram/microliter). [0508] Nucleofection of human muscle progenitor/stem cells (myoblasts) was done according to the optimized protocol for human myoblasts using the nucleofection kit developed by Aamaxa Biosystems (Amaxa Biosystems, Cologne Germany). The A-33 program was employed using the Amaxa electroporation device (Nucleofector I, Cologne Germany). Myoblasts in PBS were centrifuged at 1200 rpm for 10 min and resuspended in Nucleofector buffer. Typically, 10.sup.6 cells in 100 microliter of Primary Smooth Muscle Cell Nucleofector buffer (Amaxa Biosystems, Cologne Germany) per cuvette were subjected to electroporation with purified plasmids containing the transposon (3.6 microgram) and transposase (1.4 microgram) (concentration: 1 microgram/microliter). Transfected myoblasts were selected in G418 (400-600 microgram/ml).

[0509] E) Clonogenic Assays [0510] 1.) CFU-Mk (Megakaryocytes/Platelets) [0511] Megakaryocytic clonogenic assays were performed by adding 50 microliter of Stemline medium (Sigma-Aldrich, USA) supplemented with SCF 100 ng/ml, IL-6 20 ng/ml, IL-3 100 ng/ml, Flt3-L 20 ng/ml and TPO 100 ng/ml to the 100 microliter of electroporated CD34+ cell suspension. Fifty microliter of the final cell suspension was then added to 450 microliter of megakaryocyte differentiation medium corresponding to Myelocult H5100 (Stemcell Technologies, Vancouver Canada) supplemented with TPO 25 ng/ml, hSCF 25 ng/ml, hll-6 10 nglml, hillb 10 nglml and seeded over 3 wells in a 24-well plate, hence containing 5×10.sup.4 cells per well. At day 6 post-transfection, medium was changed by centrifuging the plate briefly, discarding the supernatant and adding fresh megakaryocyte differentiation medium. At day 10, colonies were counted. GFP expression was monitored using the Olympus fluorescence inverted microscope and CFU-Mk colonies were scored with this microscope. In addition, the automated Zeiss Inverted Microscope was employed. [0512] 2.) CFU-GM (Granulocyte/monocyte/macrophage) [0513] Granulocyte/monocyte/macrophage clonogenic assays were performed by adding 30 microliter of the final cell suspension to 270 microliter of granulocyte/monocyte/macrophage differentiation medium corresponding to semi-solid Methocult GF H4534 (Stemcell Technologies, Vancouver Canada) composed of 1% methylcellulose (4000 cps), 30% fetal bovine serum, 1% bovine serum albumin, 10-4 M 2-mercaptoethanol, 2 mM L-glutamine, 50 ng/ml rhSCF, 10 nglml rhGM-CSF, 10 ng/ml rhlL-3 in Iscove's MDM. The cell suspension were seeded over 3 wells in a 24-well plate, hence containing 5×10.sup.4 cells per well. At day 14, colonies were counted. GFP expression was monitored using the Olympus fluorescence inverted microscope and CFU-GM colonies were scored with this microscope. In addition, the automated Zeiss Inverted Microscope was employed. [0514] 3.) CFU-E (erythrocytes) [0515] Erythroid clonogenic assays were performed by adding 30 microliter of the final cell suspension to 270 microliter of erythoid differentiation medium corresponding to semi-solid Methocult SF.sup.8.sub.IT H4436 (Stemcell Technologies, Vancouver Canada) composed of methylcellulose, fetal bovine serum, bovine serum albumin, 2-mercaptoethanol, L-glutamine, rhSCF, rhGM-CSF, rhll-3, rhlL-6, rhG-CSF, rh Epo in Iscove's MOM. The cell suspension were seeded over 3 wells in a 24-well plate, hence containing 5×10.sup.4 cells per well. At day 7, colonies were counted which typically contained about 70% glycophorin N cells, a characteristic marker of erythroid cells. GFP expression was monitored using the Olympus fluorescence inverted microscope and CFU-E colonies were scored with this microscope. In addition the automated Zeiss Inverted Microscope was employed.

[0516] F) Detection of FIX [0517] The level of FIX in culture supernatant or in citrated plasma was assayed for FIX antigen by Asserachrome IX sandwich ELISA (Asserachrome/Diagnostica Stago, Parsippany, NJ, USA). Blood was collected by retro-orbital bleeds.

[0518] G) Microscopy [0519] Epifluorescence and bright field images were taken with Zeiss Axiovert 200M microscope, using the Axiovision 4.6 program and AxioCam MR3 camera. If not mentioned otherwise, pictures were taken by automatic exposure time selection and optimal display of the minimum and maximum contained gray or color value (A. Min/Max option). The settings were kept as same throughout a series of imaging and were not reset at each individual image. Confocal microscopy was carried out with Axiovert 100M, LSM510, Zeiss using the AxioPlan 2 LSM 510 version 2.8 software. In all images GFP expression was monitored at 488 nm excitation wavelength.

Examples IV to IX

Example IV

[0520] Transposition in Human CD34+ Hematopoietic Stem/Progenitor Cells

[0521] This Example was intended to provide a comparative analysis of hyperactive transposases SB M3a and SB6/AS, respectively, versus the non-hyperactive transposases SB10 and SB11 in erythroid lineage. Human CD34+ HSC were transfected by nucleofection with the pT2-HB-CAG-GFP and transposase expression vectors encoding SB M3a and SB 6/AS as described in the Materila and Method section for Examples IV to IX. The performance of these novel engineered transposases was compared with that of the originally derived SB10 transposase and SB 11. The total number of CFU-E colonies, the absolute number of GFP+ CFU-E colonies and the % GFP+ CFU-E colonies are shown in FIG. 14.

[0522] The results indicate that transposases SB M3a and SB 6/AS lead to a robust increase in % GFP+ colonies compared to the originally derived SB 10 transposase and SB 11. In contrast, no GFP+ CFU-E colonies were detectable after co-transfection with the inactive transposase SB DNGP in which the catalytic site had been mutated. Hence, the inventive SB M3a and SB 6/AS transposases correspond to the inventive group of “hyper-active” transposases that result in more efficient transposition in human CD34+ HSC compared to non-hyperactve transposase SB 10 and SB1 1. The total number of CFU-E colonies remained unchanged after electroporation with the various constructs, suggesting that there is no overt toxicity associated with over-expression of these hyper-active transposases which underscores the safety of this approach.

Example V

[0523] Transposition in Human CD34/Hematopoietic Stern/Progenitor Cells

[0524] This comparative analysis was designed to determine the results of hyperactive transposases M3a, SB6/A5 and SB 3D5-K14R in erythroid lineage. Human CD34+ HSC were transfected by nucleofection with the pT2-HB-CAG-GFP and transposase expression vector encoding SB M3a, SB 6/AS or SB 3D5-K14R, as described above. The total number of CFU-E colonies, the absolute number of GFP+ CFU-E colonies and the % GFP+ CFU-E colonies are shown in FIG. 15.

[0525] The results indicate that the transposases SB 6/AS and SB 3D5-Kl 4R lead to a significant increase of 2 and 4.8-fold in % GFP+ relative to the hyperactive transposase SB M3a. In contrast, no GFP+ CFU-E colonies were detectable after co-transfection with the inactive transposase SB DNGP in which the catalytic site had been mutated. This indicates that the SB 3D5-K14R results in even more robust transposition than the hyperactive SB M3a and SB 6/AS transposases. Hence, the data shown in FIGS. 14 & 15 indicate that SB M3a, SB 6/AS and SB 3D5-Kl 4R correspond to the group of “hyper-active” transposases that result in more efficient gene transfer in human CD34+ HSC compared to non-hyperactive transposases SB10 and SB11. The total number of CFU-E colonies remained unchanged after electroporation with the various constructs, suggesting that there is no overt toxicity associated with over-expression of these hyper-active transposases which underscores the safety of this approach.

Example VI

[0526] Transposition in Human CD34+ Hemapoietic Stem/Progenitor Cells

[0527] This Example was intended to provide a comparative analysis of hyperactive transposases SB6/A5 (Variant 3) versus SB100X (Variant 27) in erythroid, megakaryocytic and granulocytidmacrophage monocyte/lineage. Human CD34+ HSC were transfected by nucleofection with the pT2-HB-CAG-GFP and transposase expression vector encoding SB 6/AS or SB 1 00x, as described above. The total number of CFU-E colonies, the absolute number of GFP+ CFU-E colonies and the % GFP+ CFU-E colonies are shown in FIG. 16. The total number of CFU-Mk colonies, the absolute number of GFP+ CFU-Mk colonies and the % GFP+ CFU-Mk colonies are shown in FIG. 17. The total number of CFU-GM colonies, the absolute number of GFP+ CFU-GM colonies and the % GFP+ CFU-GM colonies are shown in FIG. 18. The % relative increase in % GFP+ CFU-E, CFU-Mk and CFU-GM colonies following transposition with SB100 vs. SB 6/AS is shown in FIG. 19.

[0528] The results indicate that the transposases SB 1 00x (Variant 27) lead to a robust increase in % GFP+ colonies compared to the hyperactive transposase SB 6/AS in all lineages (CFU-E, CFU-Mk, CFU-GM). The increase in % GFP CFU's, that reflects the concomitant increase in stable gene transfer efficiencies following SB-mediated transposition, was consistent among the different lineages and hereby provides compelling evidence that a genuine hematopoietic stem/progenitor cells had been stable and efficiently transfected using this transposon technology. In contrast, no GFP+ CFU-E, CFU-Mk and CFU-GM colonies were detectable after co-transfection with the inactive transposase SB DNGP in which the catalytic site had been mutated. The total number of CFU-E, CFU-Mk and CFU-GM colonies remained unchanged after electroporation with the various constructs, suggesting that there is no overt toxicity associated with over-expression of hyper-active transposases SB1O0x or SB 6/AS.

[0529] Comparative analysis of the different transposases in the erythroid lineage indicates that all inventive hyperactive transposases (SB M3a, SB 6AS, SB 3D5-K14R and SB 100x) result in more efficient stable gene transfer in CD34+ HSCs and hence a higher % GFP+ colonies compared to when the originally derived transposase SB10 and its derivative SBl 1 were used (FIG. 20). The SBl00x was the most efficient transposase resulting in ˜100-fold increase in GFP expression and stable gene transfer efficiencies compared to SBl 0. This is the first demonstration of such robust stable gene transfer in primary cells, particularly lymphohematopoietic cells, including stem cells, and more in particular hematopoietic stem/progenitor cells using transposon technology. Up to now, no such high stable gene transfer efficiencies have ever been reported using a non-viral gene transfer approach in stem cells, particularly in CD34+ HSCs. These data are consistent with a recent demonstration that only a minor fraction of CD34+ HSCs can be stably transfected when non-hyperactive transposases are used consistent with the low % GFP expression in clonogenic assays (Hollis et al. Exp Hematol. 2006 October; 34(10): 1333-43) which warrants and justifies the development of hyper-active transposases as provided herein.

Example VII

[0530] Transposition in Human Muscle Stem/Progenitor Cells

[0531] This Example was intended to validate inventive hyperactive transposases SBM3a and SB6/A5. Human muscle stem/progenitor cells (myoblasts) were transfected by nucleofection with the pT2-HB-CMV-GFP-Neo (see FIG. 11) and transposase expression vector encoding the hyperactive SB M3a transposase, as described above. Transfected cells were enriched after G418 selection. High and stable levels of GFP expression were obtained and most cells survived the G418 selection (FIG. 21). In contrast, only a limited number of GFP+ cells were detectable after cotransfection with the inactive transposase SB DNGP in which the catalytic site had been mutated. These cells ultimately failed to survive the G418 selection consistent with poor stable gene tranfer efficiencies. Comparison of the hyperactive SB M3a transposase with the orginally derived SBlO and its derivative SB11 confirm the superior transposition efficiency of SB M3a consistent with a robust increase in GFP+ transfected cells. Hence, the superior gene transfer efficiencies that can be obtained with hyperactive transposases is not unique to a given primary cell but can be extended to other cell types, including other stem/progenitor cells such as muscle stem/progenitor cells (myoblasts). This superior gene transfer potential of inventive hyperactive transposases translates into efficient and stable production of therapeutically relevant proteins like human coagulation factor IX (FIG. 22).

Example VIII

[0532] Transposition in Human Muscle Stem/Progenitor Cells

[0533] This Example serves to validate hyperactive transposases SB 100x, SB 3D5-K14R, SB M3a vs. non-hyperactive SB 11. Human muscle stem/progenitor cells (myoblasts) were transfected by nucleofection with the pT2-HB-CMV-GFP-Neo and transposase expression vector encoding the hyperactive SB 100x, SB 3D5-Kl 4R, SB M3a transposase, vs. non-hyperactive SB transposase (SBl 1) as described above. Transfected cells were enriched after G418 selection (7 days selection). High and stable levels of GFP expression were obtained and most cells survived the G418 selection (FIG. 23). In contrast, only a limited number of GFP+ cells were detectable after cotransfection with the inactive transposase SB DNGP or SB (“inactive control”) in which the catalytic site had been mutated. These cells ultimately failed to thrive under G418 selection consistent with poor stable gene tranfer efficiencies. The percentage GFP+ cells was limited when the non-hyperactive SB11 transposase was used.

[0534] Comparison of the hyperactive SB transposases with the SB10-derivative SB11, confirm the superior transposition efficiency of the SB 100x, SB 3D15-K14R and SB M3a, consistent with a robust increase in % GFP+ transfected cells. The SB100× transposase yielded the highest % GFP+ cells. The stable gene transfer efficiency as reflected by the % GFP cells was less when the SB 3D15 transposase was used relative to SB100x. The stable gene transfer efficiency as reflected by the % GFP cells was less with the SB M3a transposase compared to SB 3D5-K14R. Hence, the relative differences in transposition/stable gene transfer obtained with different transposases in human muscle progenitor/stem cells correlated with the relative differences in gene transfer in other primary cell types, particularly CD34 human hematopoietic stem/progenitor cells (FIG. 20). Hence, the superior gene transfer efficiencies that can be obtained with hyperactive transposases is not unique to a given primary cell but can be extended to other cell types, including other stem/progenitor cells such as muscle stem/progenitor cells (myoblasts). This superior gene transfer potential of hyperactive transposases translates into efficient and stable production of therapeutically relevant proteins like human coagulation factor IX following transfection with the SB 3D5-Kl 4R transposase and an SB transposon containing FIX (FIG. 24).

Example IX

[0535] Transposition In Vivo

[0536] This Example was intended to validate inventive hyperactive transposase SBlOOX. To assess whether the hyperactive transposase SB100x also resulted in more robust gene transfer in vivo compared to when non-hyperactive transposases are used, a liver-directed gene transfer experiment was conducted as described above. To achieve this, a plasmid containing a transposon expressing factor IX (FIX) from a potent liver-specific promoter (pT2-HB-Apo/MT-FIX) (see FIG. 11) was hydrodynamically transfected along with the transposase construct (hyperactive SB 100x vs. non-hyperactive SB 11 vs. inactive SB DNGP or SB GFP) by rapid tail vein injection in C57Bl/6 mice. Stable and high therapeutic factor IX levels were obtained when the hyperactive SB 1O0x was used (FIG. 25). In contrast, expression gradually declined when the inactive transposase control was employed (SB GFP). Expression of FIX following co-transfection in vivo of the FIX transposon with the hyper-active SB 100x transposase was also much more robust than when the non-hyperactive SB 11 transposase was used. Indeed, SB11-mediated transposition resulted in FIX expression that gradually declined to levels slightly above that of the control plasmid that encodes a defective transposase (SB DNGP). These results indicate that prolonged expression of FIX following SB 1 00x transfection in vivo could be ascribed to efficient stable transposition and hereby confirm the hyper-active transposition properties of SB 1O0x in vivo.

[0537] To confirm that the FIX transposon had been stably integrated into the hepatocyte genome following in vivo gene transfer, hepatocyte cell cycling was induced following partial hepatectomy (Phx) (FIG. 26). This procedure consists of surgically removing 60% of the liver. In the weeks following Phx, the liver regenerates by de nova proliferation of hepatocytes until the normal liver mass had been re-established. Since Phx did not reduce the FIX levels when the SB 1 00x was used, it provides conclusive evidence that the transgene had integrated into the genome of the in vivo transfected hepatocytes. In contrast, FIX expression declined in the absence of stable genomic integration following hydrodynamic co-transfection of the FIX transposon with expression plasmids that encoded either an inactive transposase (SB-DNGP, inactive control) or no transposase (MV-MCS).