POLYPEPTIDES FOR ENGINEERING INTEGRASE CHIMERIC PROTEINS AND THEIR USE IN GENE THERAPY

Abstract

The present invention relates to polypeptide for engineering integrase chimeric proteins and their use in gene therapy. In particular, the present invention relates to a polypeptide which comprises the amino acid sequence ranging from the amino acid residue at position 617 to the amino acid residue at position 622 in SEQ ID NO: 1 or a function conservative thereof.

Claims

1. A polypeptide which comprises the amino acid sequence ranging from the amino acid residue at position 617 to the amino acid residue at position 622 in SEQ ID NO:1 or a function conservative thereof.

2. The polypeptide of claim 1 wherein the function-conservative variant comprises an amino acid sequence having 50; 51; 52; 53; 54; 55; 56; 57; 58; 59; 60; 61; 62; 63; 64; 65; 66; 67; 68; 69; 70; 71; 72; 73; 74; 75; 76; 77; 78; 79; 80; 81; 82; 83; 84; 85; 86; 87; 88; 89; 90; 91; 92; 93; 94; 95; 96; 97; 98; or 99% identity with the amino acid sequence ranging from the amino acid residue at position 617 to the amino acid residue at position 622 in SEQ ID NO:1.

3. The polypeptide of claim 1 wherein the amino acid at position 618 or 620 is substituted by any amino acid.

4. The polypeptide of claim 1 wherein the amino acid residue at position 618 is substituted by an alanine (A).

5. The polypeptide of claim 1 wherein the amino acid residue at position 620 is substituted by an aspartic acid residue (D).

6. The polypeptide of claim 1 which comprises the amino acid sequence as set forth in SEQ ID NO:3 (KEMDSL).

7. The polypeptide of claim 1 which comprises the amino acid sequence selected from the group consisting of SEQ ID NO:2 to SEQ ID NO:12 and amino acid sequences having at least 50% of identity with SEQ ID NO:2 to SEQ ID NO:12.

8. The polypeptide of claim 1 which is fused to a heterologous polypeptide.

9. A chimeric integrase which comprises the polypeptide of claim 1.

10. The chimeric integrase of claim 9 which comprises a retroviral integrase selected from the group consisting of feline immunodeficiency virus (FIV) integrases, Foamy virus (FV) integrases, murine leukemia virus (MLV) integrases, and lentivirus integrases.

11. The chimeric integrase of claim 10 wherein the retroviral integrase has a sequence having at least 80% of identity with SEQ ID NO: 13 or SEQ ID NO:14.

12. The chimeric integrase of claim 9 wherein the polypeptide is fused at the C-terminal end of the integrase protein or at the N-terminal end of the integrase protein.

13. The chimeric integrase of claim 9 wherein the polypeptide is made part of a retroviral integrase.

14. A nucleic acid molecule which encodes a polypeptide of claim 1 or a chimeric integrase comprising the polypeptide.

15. A retroviral vector which comprises a nucleic acid molecule encoding the chimeric integrase of claim 9.

16. The retroviral vector of claim 15 which is derived from a retrovirus selected from the group consisting of alpharetroviruses, betaretroviruses, gammaretroviruses, deltaretroviruses, epsilonretroviruses, lentiviruses and spumaviruses.

17. The retroviral vector of claim 15 which comprises a transgene.

18. A method for expressing a transgene from the genome of a plurality of cells comprising infecting the plurality of cells with the retroviral vector of claim 17 under conditions which i) result in integration of the retroviral vector into the genome of the plurality of cells and ii) permit expression of the transgene from the integrated retroviral vector.

19. A method of performing gene therapy in a subject in need thereof, comprising providing the subject with a transgene by administering to the subject the retroviral vector of claim 17, under conditions that allow expression of the transgene within the subject.

Description

FIGURES

[0083] FIG. 1A-C. Ty1 IN interacts with AC40 but not with its S. pombe ortholog, AC40sp.

[0084] (A) Two-hybrid interaction between different domains of Ty1-IN and AC40 or AC40sp. Left, Ty1 structure. Two open reading frames, GAG and POL (historically TYA and TYB), are flanked by two LTRs (black triangles). Protease (PR), integrase (IN), reverse transcriptase-RNAse H (RT/RH). The left-right arrow indicates Ty1 sequences recovered in the two-hybrid screen (2-HS) with AC40 as bait (coordinates 2706-4342 in Ty1). Ty1-IN regions fused to Ga14 activating domain (GAD) are depicted. Right, serial 10-fold dilutions of liquid cultures of cells expressing AC40 or AC40sp fused to Gal4 binding domain (GBD) and the various GAD-IN proteins were plated onto non-selective (Control) or selective (Interaction) media to detect interactions. (B) Co-immunoprecipitation of AC40 and IN-Strep. A 6HA epitope was inserted at the end of the endogenous RPC40 gene (AC40-HA) and IN was fused with streptavidine (IN-Strep) and expressed from a galactose inducible promoter. Immunoprecipitation of protein extracts was performed with anti-HA magnetic beads coupled to IgG. Proteins were revealed on western-blots with 12CA5 anti-HA and anti-Strep antibodies. (C) Co-immunoprecipitation of AC40 and IN expressed from a galactose inducible functional Ty1 element. Immunoprecipitation conditions are as in FIG. 3B. Proteins were revealed with 12CA5 anti-HA and 8B11 anti-IN antibodies.

[0085] FIG. 2A-D. The interaction between Ty1-IN and AC40 is important for Ty1 targeting upstream two tRNA genes but not for Ty1 overall mobility.

[0086] (A) Ty1 integration into the URA3-tG(GCC)B locus. Histogram shows the average rate of 5FOA.sup.R colonies obtained from cultures (32 and 55 for AC40 and AC40sp expressing strains, respectively) and induced for Ty1 retrotransposition at 20 C. (grey bar) and of Ty1 insertions in URA3/5FOA.sup.R colony identified by PCR (black bars) with standard errors. (B) PCR analysis of 5FOA.sup.R colonies from independent cultures (30 and 27 for AC40 and AC40sp expressing strains, respectively) to detect Ty1 insertion events in URA3 (primers indicated by leftwards and rightwards black triangles). Representative PCR results from 4 colonies for each strain are shown, a 6-kb band indicates a Ty1 element, a 2,5-kb band no insertion. A negative control (C) is shown: URA3-tG(GCC)B locus of a strain grown at the 30 C. non permissive temperature for Ty1 retrotransposition. (C) Detection of Ty1 de novo insertions upstream of tG(GCC)C in AC40 and AC40sp expressing strains. Top: tG(GCC)C locus and experimental design. DNA was prepared from 5 independent cultures of each strain grown at 20 C. and expressing Ty1 from the galactose inducible GAL1 promoter. PCR reactions were carried out with primers specific for tG(GCC)C and Ty1 (leftwards and rightwards black triangles). Ty1 insertion occurs with equal frequency in both orientations (3, 4). Only insertion events in opposite orientation to tG(GCC)C were analyzed by PCR. N.S. a non specific band corresponding to the amplification of a fragment from adjacent Ty1 and Ty2 elements in the genome. (D) Frequency of Ty1his3A1 retrotransposition in AC40 and AC40sp expressing strains, indicated as the number of His.sup.+ prototrophs divided by the total number of cells. Data from 6 independent cultures are represented as box plots.

[0087] FIG. 3A-C. Association of Ty1 insertions with chromosomal features in WT (AC40) and loss-of-interaction mutant (AC40sp) strains.

[0088] (A) Ty1 insertions upstream of the 275 tRNA genes of S. cerevisiae were aggregated into a single distribution measuring distance to the start of transcription (position 0 on the x-axis). Black points above and below the x-axis denote convergent and divergent integrations, relative to the tRNA gene, respectively. LOESS (Locally weighted scatterplot smoothing) curves in red and blue indicate the general trends. The grey backdrop indicates the level of nucleosome at each position, based on the sequencing of chromosomal DNA after digestion by micrococcal nuclease (30). The y-axis scale is normalized to the number of total integrations in each sample. The nucleosomes were linearly scaled for visibility of the periodicity. Left panel shows Ty1 insertions in WT (AC40) and right panel, Ty1 insertions in mutant (AC40sp). (B) Graphic of a subset of representative features with positive and negative AUC0.5 (AUC, Area Under the Curve values of the Receiver Operator Characteristic). 0 indicates a model of no predictive power, values of magnitude 0.5 indicate perfect prediction. Negative and positive values denote a feature that is associated with the control and with integration, respectively. Nucleosome AUC values were obtained from chromosomal regions for which nucleosomes data are available, the other AUC values were obtained from the entire genome dataset. (C) Top: Distribution of Ty1 insertions on chromosome III. Vertical bars indicate the number of unambiguous integrations at a particular site. Red and blue bars above and below the x-axis indicate insertions in WT (AC40) and loss of interaction mutant (AC40sp), respectively. Vertical lines indicate tRNA genes in forward (blue) or reverse (pink) orientations. Bottom: Zoom of the distribution of Ty1 insertions on the right subtelomere of chromosome III (coordinate 292388 to 316620). Depicted are the HMR locus (green thick line), the tT(AGU)C tRNA gene (light brown triangle), a large subtelomeric region containing non-essential genes (grey thick line), an ARS (blue thick line) and the telomere TELO3R (black thick line). X-axis denotes position along the chromosome at 1000-bp resolution for the whole chromosome and 100-bp for the right subtelomere. Y-axis scale shows percentage of total integrations in the genome per bin.

[0089] FIG. 4: Scheme of the two-hybrid system. Eukaryotic transcription factors consist of two main domains that are folded independently: a DNA binding domain (BD) that binds specific target sequences on a promoter and an activation domain (AD) that recruits the transcription machinery. The basic concept of our yeast two-hybrid screen is based on the reconstitution of an active transcription factor resulting from the interaction between two proteins, the bait protein, here AC40, fused to the BD of GAL4 (GBD), and the other protein, the prey, here the different IN constructs fused to the AD of GAL4. The basis of the two-hybrid assay is that the transcription of the HIS3 reporter gene correlates to the interaction between the bait and prey proteins.

[0090] FIG. 5: Characterization of Ty1 TD by two-hybrid assay. On the top, Ty5 TD is depicted as an example. It was the first targeting domain identified in the integrase of a retrolelement. It corresponds to the minimal region interacting with the Sir4 protein, which targets Ty5 integration into heterochromatin regions. Ty1 TD was predicted on the basis of the alignments of the C-terminal region Ty1, Ty2 and Ty4 integrases because the three retrotransposons have the same preference to integrate upstream of Pol III transcribed genes. The rectangles indicate regions/amino acids that were well conserved between the 3 integrases. Alanine substitution of LE601-602, VS609-610, T616 (together with the deletion of W614), K617, N618, M619, R620, 5621, L622, E623 and SLE621-623 was performed on GAD-IN578-635 and the interaction of the different fusion proteins with GBD-AC40 was tested. Growth indicates an interaction. (+) indicates an interaction between IN and AC40, () no interaction.

[0091] FIG. 6A-B: Characterization of the minimal Ty1 TD by two-hybrid assay. A) The peptide 617KNMRSLEPP625 (SEQ ID NO: 7) of Ty1 IN was fused to GAD and the interaction between the two protein fusions were tested by two-hybrid assay. Positive control: IN.sub.578-635, negative control: IN.sub.578-635 with the R620A substitution, Ty#125: positive control, which corresponds to a sequence obtained in the original two-hybrid screen described in Example 1. Results indicate that the peptide 617KNMRSLEPP625 interacts weakly with AC40. B) The sequences IN.sub.596-630 and IN.sub.617-623 (SEQ ID NO: 4) fused to the streptavidine tag sequence interact with AC40. Positive control: IN.sub.578-635, negative control IN.sub.1-578.

[0092] FIG. 7: Replacement of the sequence 617KNMRSL622 by the sequence 617KEMDSL622 (SEQ ID N: 2) in GAD-IN578-635 maintains the interaction with GBD-AC40. Interaction between GBD-AC40 and the different GAD-IN protein constructs were tested by two-hybrid assay. Negative control (GAD-IN578-635 with the S621A substitution), positive control, GAD-IN578-635 with the E623A substitution).

[0093] FIG. 8A-B: Mutations that abolish IN/AC40 interaction decrease Ty1 integration upstream of a Pol III transcribed genes but not Ty1 overall mobility.

A) Ty1 overall mobility decreases no more than two-fold in Ty1 mutants carrying mutations that abolish IN/AC40 interaction. Relative frequency of Ty1 retrotransposition in yeast strains expressing various mutants of Ty1his3AI, in which either of the K617, M619, R619, S621 or L622 residues where modified in A (alanine). For each strain, the number of His.sup.+ cells was divided by the total number of cells. Values were set at 100% of relative retrotransposition for WT Ty1 and normalized for Ty1 mutant elements.
B) Ty1 integration upstream of a Pol III genes is severely affected in Ty1 mutants carrying mutations that abolish IN/AC40 interaction. Detection of Ty1 insertion events upstream of tG(GCC)C in yeast strains expressing various mutants of Ty1 (in which either of the S621 or L622 residues where modified in A (alanine)). A periodic banding pattern is characteristic of a functional Ty1 element (WT). Loss of this pattern in mutants indicates that they have lost the ability to integrate upstream of Pol III genes. The loss of this pattern was also obtained in mutants of Ty1 in which either of the K617 or M619 residues where modified in A (alanine). Experimental design is given in the legend of FIG. 2C.

EXAMPLES

Example 1: An RNA Polymerase III Subunit Determines Sites of Retrotransposon Integration

[0094] A balance exists between the short-term detrimental effects of transposable elements as mutagens and the long-term, positive role they play as agents of genome plasticity and evolution (1). Important in maintaining this balance is integration site choice (ISC). Integration is typically not random and often occurs in regions of the genome where insertion is benign. For the retrotransposonsobligate genomic parasitesintegration preferentially occurs in regions that are gene poor, likely the consequence of selection for mechanisms that favor non-deleterious insertions. Ty1, the most active and abundant LTR-retrotransposon of S. cerevisiae, preferentially integrates within a 1-kb window upstream of Pol III-transcribed genes (2). Ty1 targeting requires a functional Pol III promoter in the target gene (2-4) and is influenced by the chromatin-remodeling factor Isw2 and the Bdp1 subunit of TFIIIB (5, 6). However, the primary determinant of Ty1 ISC and its contribution in protecting the genome from deleterious insertions is still unknown.

[0095] In an attempt to identify protein interactors of Pol III, we performed systematic yeast two-hybrid screens using the 17 Pol III subunits as baits and a library of yeast DNA as prey. We identified a specific interaction between the AC40 subunit and a region of the Ty1 TyB/POL protein that encompasses the C-terminus of integrase (IN) and the N-terminus of reverse transcriptase (FIG. 1A). The screen also recovered Ty2 and Ty4 overlapping the same protein domains as Ty1, consistent with similar integration preferences for these 3 elements (7). HA-tagged AC40 (AC40-HA) co-immunoprecipitated IN when ectopically expressed either as a streptavidine-fusion (IN-Strep) (FIG. 1B) or from a functional Ty1 element (FIG. 1C).

[0096] Ty1 IN has two phylogenetically conserved regions (an N-terminal zinc-binding domain and a catalytic core with the DDX.sub.35E motif) and a less conserved C-terminus containing a bi-partite nuclear localization signal required for IN nuclear localization and efficient retrotransposition (FIG. 1A) (8, 9). A two-hybrid assay confirmed the interaction between AC40 and full-length IN (IN.sub.1-635) and revealed that the IN N-terminus and catalytic core were dispensable (IN.sub.193-635), whereas the last 57 amino acids of the IN C-terminus were necessary and sufficient (FIG. 1A, IN.sub.1-578 versus IN.sub.578-635). The integrases of several retrotransposons and retroviruses interact with a cellular DNA-bound protein to mediate ISC (10-16). Similar to the yeast retrotransposon Ty5 and murine leukemia virus (MLV), the Ty1 targeting domain is located at the IN C-terminus, suggesting in some cases this region evolved to interact with specific targeting cofactors.

[0097] Since AC40 is an essential protein, we addressed its role in Ty1 ISC by using the Schizosaccharomyces pombe rpc40.sup.+ ortholog of RPC40, which encodes AC40. The S. pombe AC40 (AC40sp) could restore the viability of S. cerevisiae cells lacking AC40 and sustain normal Pol III transcription (17). However, GBD-AC40sp, when expressed at levels similar to GBD-AC40, did not show two-hybrid interaction with Ty1 IN (FIG. 1A and S1D), implying that in its native state, AC40sp and Ty1 IN do not interact or only very poorly. Therefore, AC40sp was used as a loss-of-interaction mutant. We first determined the role of AC40 in Ty1 integration at Pol III-transcribed genes using a quantitative assay that relies on a URA3 gene located upstream of the glycine tRNA gene tG(GCC)B, a hotspot of Ty1 integration. In this assay, Ty1 insertions upstream of tG(GCC)B inactivate URA3, resulting in Ura.sup. cells that grow on 5FOA plates. Endogenous Ty1 retrotransposition was induced in rpc40 strains expressing AC40 or AC40sp at similar levels by switching growth to 20 C., a temperature permissive for Ty1 retrotransposition. A 3.5-fold decrease in the number of 5FOA.sup.R colonies per cell was observed in the presence of AC40sp (FIG. 2A). The fraction of 5FOA.sup.R cells arising from Ty1 integration in cells expressing AC40sp was 30-fold lower than in AC40 expressing cells (FIGS. 2A and 2B). Therefore, AC40/IN interaction is determinant for Ty1 integration at this reporter gene.

[0098] We further demonstrated the role of AC40 in Ty1 ISC by analyzing unselected insertions at the tRNA gene tG(GCC)C, another Ty1 integration hotspot, in strains that express a galactose-inducible Ty1 element (FIG. 2C). In the presence of AC40, we observed a 70-bp periodic banding pattern characteristic of Ty1 insertion in nucleosomal DNA upstream of tRNA genes (3, 4, 18). In cultures expressing AC40sp, the pattern was dramatically altered, showing both a significant decrease in Ty1 integration and an apparent relaxed periodicity between hotspots (FIG. 2C). The same alteration in the pattern of integration at the tG(GCC)C locus was observed in AC40sp-expressing cells for endogenous Ty1 elements. Hence, AC40/IN interaction is critical for Ty1 integration specificity upstream of Pol III-transcribed genes.

[0099] To establish whether Ty1 integration frequency depends on the AC40/IN interaction, we used a chromosomal Ty1 element marked with a his3AI reporter gene, which confers His.sup.+ prototrophy to cells upon retrotransposition. Frequencies of His.sup.+ cells generated in strains expressing AC40 or AC40sp were similar (FIG. 2D). It is conceivable that in the loss-interaction mutant an overrepresentation of Ty1HIS3 insertions could be due to homologous recombination (HR) with endogenous elements (19). We determined the frequency of IN-dependent Ty1HIS3 insertion events in a rad52 mutant, which is deficient in HR (FIG. 2D). His.sup.+ events were slightly higher in rad52 mutant cells, consistent with previous studies showing that although the Rad52 pathway contributes to Ty1 HR, it globally inhibits Ty1 retrotransposition (20). Notably, the frequency of IN-dependent integration is 2-fold lower in AC40sp cells compared to AC40 cells, suggesting that a higher proportion of Ty1 insertions might be due to recombination in the absence of interaction. However, this slight decrease can not explain the dramatic effect observed on Ty1 integration at the URA3-tG(GCC)B reporter gene. Altogether, these results indicate that the AC40/IN interaction is required for the canonical Ty1 targeting pattern upstream of two representative tRNA genes, but the interaction does not influence integration efficiency. Therefore, the interaction with AC40 could function specifically to restrict Ty1 integration upstream of Pol III-transcribed genes to prevent potentially harmful insertions at secondary targets in the genome.

[0100] We compared the genome-wide integration patterns of Ty1 in the WT and AC40sp loss-of-interaction mutant strains. The WT strain exhibited a typical integration pattern at Pol III-transcribed genes (3, 4) (FIG. 3A). On the contrary, the loss-of-interaction mutant displayed a fundamentally different profile with only a slight bias to integrate near tRNA genes. Residual interaction between IN and AC40sp could explain this modest bias. To determine if other genomic features influence Ty1 ISC in the absence of the AC40/IN interaction, we used single (FIG. 3B) and multi-dimensional logistic regression models to associate genomic features with integration hotspots. Ty1 insertions in the WT strain were associated with upstream regions of tRNA genes or features associated with these sites, such as pre-existing retrotransposons (FIG. 3B). Regions with well-positioned nucleosomes were also favored and open reading frames (ORFs) avoided. In the AC40sp loss-of-interaction mutant, tRNA genes were no longer the primary targeting determinant but nucleosomes were still favored. Telomeres and subtelomeric regions were also preferred targets in the WT and even more so in the mutant (FIGS. 3B, and 3C). This propensity for insertions to occur in heterochromatic domains was observed for all chromosomes. Noteworthy, Ty5, a related yeast retrotransposon, also targets heterochromatin by specifically interacting with the heterochromatin protein Sir4 (11). Thus, an interaction between Ty1 IN and heterochromatin proteins could be involved in Ty1's secondary target site preference.

[0101] We identified an interaction between Ty1 IN and the Pol III subunit AC40 that determines Ty1 ISC upstream of Pol III genes. Previous studies indicated that Ty1 integration requires a functional Pol III promoter, and our characterization of AC40 as a cofactor of Ty1 integration explains this requirement. The retroelements Ty3 of S. cerevisiae and TRES-A of Dictyostelium discoideum also integrate at Pol III-transcribed genes, but their targeting involves an interaction with TFIIIB (21, 22). Our study demonstrates a direct role of an RNA polymerase in integration targeting. In Pol III, AC40 is located at the periphery of the complex in close proximity to the upstream DNA region (23). This location might explain why Ty1 integrates mostly upstream of Pol III transcribed-genes and rarely downstream. Ty1 does not integrate near or into genes transcribed by Pol I, which also contains AC40 (3, 4), The AC40/IN interaction could require additional cofactor(s) specific to Pol III transcription. Noteworthy, AC40 in Pol III is in close contact with TFIIIB (24), which contains Bdp1, a protein required for the Isw2-dependent periodic integration of Ty1 (6). This cofactor could participate directly or indirectly to the interaction. Ty1 integrates preferentially in nucleosomes, and the chromatin at Pol III promoters is relatively open structure with well-organized nucleosomes (25, 26). This organization is not found at Pol I genes, which are either actively transcribed or packed into a tight and repressive nucleosomal structure (27)chromatin states that might disfavor Ty1 integration.

[0102] The Pol III transcription complex, which includes AC40, is regulated by nutrient and stress signaling pathways (28). As a result, Ty1 exhibits the potential to act as a mutagenic agent under conditions that affect AC40/IN interaction or Pol III transcription. tabOur work reveals that Ty1 also targets subtelomeres, especially in the absence of AC40/IN interaction. Yeast chromosome ends contain non-essential fast-evolving gene families generally needed to respond to environmental changes (29). Therefore, targeting Ty1 integration to subtelomeres could further protect the yeast genome from Ty1 mobility, while potentially promoting evolutionary adaptation and gene innovation in response to stress.

[0103] Understanding the molecular mechanism of targeted integration of the Ty1 retroelement has been a Holy Grail of mobile genetic element research for decades. With our study, we elucidate this molecular mechanism by indicating that the AC40 subunit of Pol III is the major cellular protein that targets Ty1 integration to Pol III genes, through its interaction with Ty1 integrase. Importantly, the tethering of integration complexes to the cell genome, through an interaction between the integrase and cellular proteins bound at favored insertion sites, is conserved with MLV and HIV retroviruses. The main difference is that MLV and HIV integrases interact with cellular factors that target retroviral integration in the promoter or transcription unit of Pol II transcribed genes, leading to potential gene deregulation or inactivation upon their integration.

[0104] The stable integration of retroelements into the host cell genome constitutes a major advantage for a retrovirus-based gene delivery system, especially for gene therapy, which aims at long-term correction of genetic defects, but also raises concerns about potential mutagenesis and oncogene activation. Gene therapy with MLV-derived retroviral vectors to treat severe immunodeficiency obtained impressive therapeutic success. However, the development of leukemia caused by insertional mutagenesis in some patients uncovered a serious unanticipated side effect due to the transactivation of adjacent oncogenes by the enhancer activity of the viral LTR. A key advance was the creation of more sophisticated self-inactivating (SIN) vectors, designed to avoid such transactivation events upon integration. However, these vectors still integrate near proto-oncogenes, maintaining a Damocles sword on the safety of such vectors. Therefore, a potential improvement for the future is to target gene therapy vectors to safe regions of the genome. Ty1 targeting mechanism upstream of individually non-essential Pol III genes provides safe landing sites that prevent deleterious consequences on cell fitness, and is therefore a promising alternative to already existing retroviral-based vectors designed for human gene therapy. Accordingly, the domain of Ty1 responsible for the interaction with the RNA polymerase III could thus be suitable to engineering chimeric integrases of retrovirus so as to drive the targeted integration of a transgene into safe regions of a eukaryotic genome, in order to reduce potential detrimental effect.

Example 2: Characterization of Ty1 IN Targeting Domain

[0105] We used the two-hybrid approach as depicted in FIG. 4 to further characterize the TY1 domain responsible for the interaction with AC40: if the interaction occurs, then the HIS3 is activated and the growth can occur in the absence of histidine; if no interaction occurs then there is no activation of HIS3 and accordingly the growth cannot occur in the absence of histidine. Briefly, the experimental design was as follows: strain PJ69-4A is transformed with the 2 vectors pAS2 and pACTII, containing either GBD-AC40 or any of the GAD-IN constructs tested in the examples. Cultures were grown overnight at 30 C. Serial 10-fold dilutions of aliquots of 1 DO600, washed in 1 ml of H20, were plated on HC medium lacking leucine and tryptophane (control) or leucine, tryptophane and histidine (interaction). Leucine and tryptophan prototrophies guaranty the presence of the 2 vectors. Plates were incubated 3 days at room temperature. As shown in FIG. 5, the minimal sequence responsible for the interaction of IN578_635 with AC40 is the sequence which ranges from the amino acid residue at position 617 to 623. We also demonstrated that the amino acid residues at position 618 or 623 can be substituted. In FIG. 6 we showed that the sequence KNMRSLEPP can also interact with AC40. In FIG. 7, we replaced in GAD-IN578-635, the sequence 617KNMRSL622 by KEMDSL (the amino acids that were modified are underlined) to demonstrate that the sequence KEMDSL also interacts with AC40.

REFERENCES

[0106] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure. [0107] 1. H. L. Levin, J. V Moran, Nat Rev Genet 12, 615-627 (2011). [0108] 2. S. E. Devine, J. D. Boeke, Genes Dev. 10, 620-633 (1996). [0109] 3. J. A. Bailer, J. Gao, R. Stamenova, M. J. Curcio, D. F. Voytas, Genome Res. 22, 704-13 (2012). [0110] 4. L. Mularoni et al., Genome Res 22, 693-703 (2012). [0111] 5. M. E. Gelbart, N. Bachman, J. Delrow, J. D. Boeke, T. Tsukiyama, Genes Dev 19, 942-954 (2005). [0112] 6. N. Bachman, M. E. Gelbart, T. Tsukiyama, J. D. Boeke, Genes Dev 19, 955-964 (2005). [0113] 7. J. M. Kim, S. Vanguri, J. D. Boeke, A. Gabriel, D. F. Voytas, Genome Res. 8, 464-478 (1998). [0114] 8. M. Kenna, C. Brachmann, S. Devine, J. Boeke, Mol Cell Biol 18, 1115-1124 (1998). [0115] 9. S. P. Moore, L. A. Rinckel, D. J. Garfinkel, Mol Cell Biol 18, 1105-1114 (1998). [0116] 10. J. Kirchner, C. M. Connolly, S. B. Sandmeyer, Science 267, 1488-1491 (1995). [0117] 11. W. Xie et al., Mol Cell Biol 21, 6606-6614 (2001). [0118] 12. Y.-E. E. Leem et al., Mol Cell 30, 98-107 (2008). [0119] 13. A. Engelman, P. Cherepanov, PLoS Pathog 4, e1000046 (2008). [0120] 14. J. De Rijck et al., Cell Rep. 5, 886-894 (2013). [0121] 15. S. S. Gupta et al., J. Vivol. 87, 12721-36 (2013). [0122] 16. A. Sharma et al., Proc. Natl. Acad. Sci. U.S.A 110, 12036-41 (2013). [0123] 17. G. V Shpakovski, E. K. Shematorova, Curr Genet 36, 208-214 (1999). [0124] 18. N. Bachman, Y. Eby, J. D. Boeke, Genome Res 14, 1232-1247 (2004). [0125] 19. G. Sharon, T. J. Burkett, D. J. Garfinkel, Mol Cell Biol 14, 6540-51. (1994). [0126] 20. A. J. Rattray, B. K. Shafer, D. J. Garfinkel, Genetics 154, 543-56. (2000). [0127] 21. L. Yieh, G. Kassavetis, E. P. Geiduschek, S. B. Sandmeyer, J Biol Chem 275, 29800-29807 (2000). [0128] 22. T. Chung, O. Siol, T. Dingermann, T. Winckler, Mol. Cell. Biol. 27, 8492-501 (2007). [0129] 23. C. Fernandez-Tomero et al., Nature 502, 644-9 (2013). [0130] 24. A. Vannini, P. Cramer, Mol. Cell 45, 439-46 (2012). [0131] 25. K. Brogaard, L. Xi, J.-P. Wang, J. Widom, Nature 486, 496-501 (2012). [0132] 26. Y. Kumar, P. Bhargava, BMC Genomics 14, 402 (2013). [0133] 27. K. Merz et al., Genes Dev. 22, 1190-204 (2008). [0134] 28. R. D. Moir, I. M. Willis, Biochint Biophys. Acta 1829, 361-75 (2013). [0135] 29. C. A. Brown, A. W. Murray, K. J. Verstrepen, Curr. Biol. 20, 895-903 (2010). [0136] 30. N. Kaplan et al., Nature 458, 362-366 (2009).