Nipah virus envelope glycoprotein pseudotyped lentivirus

Abstract

The present invention relates to pseudotyped retrovirus-like particles or retroviral vectors comprising both engineered envelope glycoproteins derived from a virus of the Paramyxoviridae family fused to a cell targeting domain and fused to a functional domain. The present invention also relates to the use of said pseudotyped retrovirus-like particles or retroviral vectors to selectively modulate the activity of specific subsets of cells, in particular of specific immune cells. These pseudotyped retrovirus-like particles or retroviral vectors are particularly useful for gene therapy, immune therapy and/or vaccination.

Claims

1. A pseudotyped lentiviral vector comprising: a) at least one cell targeting fusion protein comprising (i) an envelope glycoprotein G of a Nipah virus that lacks amino acids 2-34 of SEQ ID NO: 9 (GcΔ34) and comprises the point mutations E501A, W504A, Q530A, and E533A by comparison to the sequence of SEQ ID NO:9 and (ii) at least one cell targeting domain, b) at least one envelope glycoprotein F of a Nipah virus that lacks amino acids 525-546 of SEQ ID NO:11 (FΔ22), and c) a nucleic acid encoding a chimeric antigen receptor.

2. A pseudotyped lentiviral vector comprising: a) at least one cell targeting fusion protein comprising an envelope glycoprotein G of a Nipah virus that lacks amino acids 2-34 of SEQ ID NO: 9 (GcΔ34) and comprises the point mutations E501A, W504A, Q530A, and E533A by comparison to the sequence of SEQ ID NO:9, and b) at least one glycoprotein that is an envelope glycoprotein F of a Nipah virus that lacks amino acids 525-546 of SEQ ID NO:11 (FcΔ22).

3. The pseudotyped lentiviral vector of claim 2, wherein the at least one cell targeting fusion protein further comprises at least one cell targeting domain.

4. The pseudotyped lentiviral vector of claim 1, wherein the at least one cell targeting domain is specific for CD3, CD8, or CD4.

5. The pseudotyped lentiviral vector of claim 1, wherein the at least one cell targeting domain is specific for CD8 or CD4.

6. The pseudotyped lentiviral vector of claim 1, wherein the at least one cell targeting domain is a Darpin or an scFv.

7. The pseudotyped lentiviral vector of claim 3, wherein the at least one cell targeting domain is a Darpin or an scFv.

8. The pseudotyped lentiviral vector of claim 1, wherein the at least one cell targeting domain is an scFv specific for CD8.

9. The pseudotyped lentiviral vector of claim 1, wherein the encoded chimeric antigen receptor is specific for CD19.

10. A pseudotyped lentiviral vector comprising: a) a cell targeting fusion protein comprising (i) a truncated glycoprotein G of a Nipah virus that lacks amino acids 2-34 of SEQ ID NO:9 (GcΔ34) comprises the point mutations E501A, W504A, Q530A, and E533A by comparison to the sequence of SEQ ID NO:9, and (ii) an scFv that is specific for CD8, b) a truncated glycoprotein F of a Nipah virus that lacks amino acids 525-546 of SEQ ID NO: 11 (FcΔ22), and c) a nucleic acid encoding a chimeric antigen receptor that is specific for CD19.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1: Efficient and stable ErbB2-CAR gene transfer into resting CD4.sup.+ T cells by 4.sup.H/IL7.sup.H-LV. To assess if 4.sup.H/IL7.sup.H-LV can efficiently transfer therapeutic genes (ErbB2 specific CAR, ErbB2-CAR) into resting T cells, freshly isolated CD3.sup.+ T cells were transduced with 4.sup.H-LV or 4.sup.H/IL7.sup.H-LV. Cells transduced with VSV-LV or left untransduced (ut) were used as negative controls. Three days later, half of the transduced cells were analyzed by FACS to determine the percentage of ErbB2-CAR.sup.+ cells in the CD4- or CD8-gated cells. To confirm stable ErbB2-CAR expression, the other half of the transduced cells was further cultivated in the presence of anti-CD3/anti-CD28/IL-2 stimulus, after a washing step, for additional three days. Then, the percentages of CD4.sup.+/ErbB2.sup.+ cells and CD8.sup.+/ErbB2.sup.+ cells were determined by FACS.

(2) FIG. 2: Mutation of the NiV envelope glycoprotein G to ablate natural receptor recognition. Binding of Ephrin-B2 (A) and B3 (B) to NiV-G mutants is shown. 293T cells were transfected either mock, plasmids encoding for GcΔ34.sup.His, or with different GcΔ34.sup.EpCAM mutants: unmutated (GcΔ34.sup.EpCAM), E533A (GcΔ34.sup.EpCAMmut1), Q530A+E533A (GcΔ34.sup.EpCAMmut2.1), E501A+W504A (GcΔ34.sup.EpCAMmut2.2) or E501A+W504A+Q530A+E533A (GcΔ34.sup.EpCAMmut4) and incubated with 1 μg/ml recombinant Fc-EphrinB2 or B3. The amounts of binding of the receptor by the different mutants are shown as MFI values. Statistics are relating to unmutated GΔ34-DARPin-Ac1 (n=3; mean±standard deviations (SD) are shown; *, P<0,1**, P<0,01;***, P<0,001 by unpaired t test).

(3) FIG. 3: Selective activation of CD4.sup.+ T cells by 4.sup.H/IL7.sup.H-VLP. Freshly isolated CD3.sup.+ T cells were left untransduced (ut) or transduced with different types of VLPs pseudotyped with CD4-DARPin-Hmut (4.sup.H-VLP), CD4-DARPin/IL7-Hmut (4.sup.H/IL7.sup.H-VLP) or IL7-Hnse (Hnse/IL7.sup.H-VLP). Three days later the cells were stained by CD8, CD4, and CD71 antibodies. The CD71 expression in the CD4- or CD8-gated T cells is shown.

(4) FIG. 4: Functionally displayed IL7 on the targeted-VLPs promote T cell survival. The forward/side scatter profiles of adult resting CD3.sup.+ T cells that were incubated for 6 days with virus-like particles (VLP) pseudotyped with CD4-targeted MV-H (4.sup.H-VLP), CD8-targeted MV-H (8.sup.H-VLP), IL7-displaying CD4-targeted MV-H (4.sup.H/IL7.sup.H-VLP), or IL7-displaying CD8-targeted MV-H (8.sup.H/IL7.sup.H-VLP). Cells cultivated in the medium alone (untransduced, ut) or in the presence of 15 ng/ml IL7 were used as controls. The percentage of viable cells is indicated in each dot blot.

(5) FIG. 5: Functionally displayed IL7 on the CD4-targeted LVs promote T cell survival. The forward/side scatter profiles of adult resting CD3.sup.+ T cells that were incubated for 6 days with lentiviral vectors (LV) pseudotyped with VSVG (VSV-LV), MV-Hmut (Hnse-LV), CD4-targeted MV-H (4.sup.H-LV), or IL7-displaying CD4-targeted MV-H (4.sup.H/IL7.sup.H-LV). Cells cultivated in the medium alone (untransduced, ut) or in the presence of 15 ng/ml IL7 were used as controls. The percentage of viable cells is indicated in each dot blot.

(6) FIG. 6: Selective activation of CD4.sup.+ T cells by 4.sup.H/IL7.sup.H-LV. Freshly isolated CD3.sup.+ T cells were left untransduced (ut) or transduced with different types of LVs pseudotyped with CD4-DARPin-Hmut (4.sup.HΔ18-LV), Hnse (Hnse-LV), or VSVG (VSVG-LV). Here IL7 was displayed on two different versions of Hmut, either with 20 amino acids C-truncated (HΔ20) or with 15 amino acids C-truncated (H≢15). Accordingly, two types of 4.sup.H/IL7.sup.H-LV were produced and transduced the resting T cells. Three days later the cells were stained by CD3, CD8, CD69, and CD71 antibodies. The activation markers CD69 and CD71 expression in CD8.sup.+ and CD8.sup.− (CD4.sup.+) cells is shown.

(7) FIG. 7: Schematic representation of pseudotyped lentiviral vectors (LV). 4.sup.H-LV is a CD4-targeted LV based on MeV glycoprotein pseudotypes. 4.sup.H/IL7.sup.H-LV is a IL7-displaying CD4-targeted LV based on MV glycoprotein pseudotypes. The envelope glycoprotein H (MV-H) is truncated at its cytoplasmic tail, for example by 15, 18, or 20 amino acids and at least two out of four point mutations (Y481A, R533A, S548L, and F549S) are introduced to blind its natural receptor recognition of SLAM and CD46, thereby resulting in the envelope glycoprotein Hmut. Afterwards, a targeting domain (for example CD4-DARPin) and/or a cytokine (for example IL7) are fused to the mutant H (Hmut), with or without a linker. The MV envelope glycoprotein F is truncated at its cytoplasmic tail by 30 amino acids (MV-F.sub.Δ30).

(8) FIG. 8: Schematic representation of pseudotyped virus-like particles (VLP). 8.sup.G-VLP is a CD8-targeted VLP based on NiV glycoprotein pseudotypes. 8.sup.G/IL7.sup.G-VLP is a IL7-displaying CD8-targeted VLP based on NiV glycoprotein pseudotypes. The NiV-G envelope glycoprotein is truncated at its cytoplasmic tail by 34 amino acids (Δ34) and four point mutations (E501A, W504A, Q530A, and E533A) are introduced to blind its natural receptor recognition of Ephrin-B2/B3, thereby resulting in the envelope glycoprotein NiV-G.sub.Δ34. Afterwards, a targeting domain (CD8-scFv) and/or a cytokine (IL7) are fused to the mutant G protein (NiV-G.sub.Δ34) with or without a linker. The NiV envelope glycoprotein F is truncated at its cytoplasmic tail by 22 amino acids (NiV-F.sub.Δ22).

(9) FIG. 9: Efficient and selective transduction of resting CD4.sup.+ T cells by 4.sup.H/IL7.sup.H-LV. Freshly isolated CD3.sup.+ T cells were left untransduced (ut) or transduced with different types of LVs pseudotyped with CD4-DARPin-Hmut (4.sup.HΔ18-LV), Hnse (Hnse-LV), or VSVG (VSVG-LV). Here IL7 was displayed on two different versions of Hmut, either with 20 amino acids C-truncated (HΔ20) or with 15 amino acids C-truncated (HΔ15). Accordingly, two types of 4.sup.H/IL7.sup.H-LV (4.sup.HΔ18/IL7.sup.HΔ15-LV and 4.sup.HΔ18/IL7.sup.HΔ20-LV) were produced and transduced the resting T cells. Three days later the cells were stained by CD3, CD8, and CD4 antibodies. The GFP expression in CD4- or CD8-gated T cells is shown.

(10) FIG. 10: Neutralization of NiV glycoprotein pseudotyped LVs. CHO-EpCAM or CHO-Ephrin-B2 cells were transduced with NiVmutEpCAM-LV (circle), MVEpCAM-LV (square), NiVwt-LV (triangle), or VSVG-LV (diamond) at an MOI of 0.4 after incubation with serial dilutions of pooled human serum (IVIG) for 2 h at 37° C. After 72 h, GFP+ cells were determined by flow cytometry and the number of GFP+ cells relative to untreated control is shown (n=3).

(11) FIG. 11: Selective activation of CD8.sup.+ T cells by 8.sup.G/IL7.sup.G-LV. Freshly isolated CD3.sup.+ T cells were left untransduced (ut) or transduced with different types of LVs pseudotyped with CD8-scFv-Gmut (8.sup.G-LV) or CD8-scFv-Gmut/IL7-Gmut (8.sup.G/IL7.sup.G-LV). Three days later the cells were stained by CD8, CD4, and CD71 antibodies. The CD71 expression in the CD4- or CD8-gated T cells is shown.

(12) FIG. 12: Efficient and selective transduction of resting CD8.sup.+ T cells by 8.sup.G/IL7.sup.G-LV. Freshly isolated CD3.sup.+ T cells were left untransduced (ut) or transduced with different types of LVs pseudotyped with CD8-scFv-Gmut (8.sup.G-LV) or CD8-scFv-Gmut /IL7-Gmut (8.sup.G/IL7.sup.G-LV). Three days later the cells were stained by CD3, CD8, and CD4 antibodies. The GFP expression in CD4- or CD8-gated T cells is shown.

(13) FIG. 13: Efficient CAR19 gene transfer into resting CD8.sup.+ T cells by 8.sup.G/IL7.sup.G-LV. Freshly isolated CD3.sup.+ T cells were left untransduced (ut) or transduced with different types of LVs pseudotyped with CD8-scFv-Gmut (8.sup.G-LV), CD8-scFv-Gmut /IL7-Gmut (8.sup.G/IL7.sup.G-LV) or VSVG (VSV-LV). The therapeutic gene encoding a CD19-specfic CAR (CAR19) was delivered by the LVs. Three days after transduction, the CAR19 expression was detected by anti-cmyc antibody. The percentage of CAR19 expressing cells in CD4.sup.+ or CD8.sup.+ T cells is shown.

(14) FIG. 14: 4.sup.H/IL7.sup.H-LVs allow targeted in vivo activation of CD4+ T cells in humanized mice.

(15) NOD/SCID gc−/− mouse were injected with cord blood T cells and 2 months upon engraftment (20% T cell reconstitution), 100 microliters of 4.sup.H/IL7.sup.H-LV (1E6 IU) or 4.sup.H/IL7.sup.H-LV were injected by IV. Two weeks after injection of the vectors, the mice were sacrificed and the splenocytes were evaluated for % of CD71+CD4+ T cells (activation marker).

EXAMPLES

(16) Material and methods

(17) Generation of the Constructs

(18) The plasmid pHL3-Ac1 coding for truncated and mutated MV HcΔ18mut protein and for a (G.sub.4S).sub.3 linker (L3) between H and the His-tagged DARPin Ac1 was generated by inserting the PCR amplified coding sequence of the EpCAM specific DARPin Ac1 (Stefan et al. 2011) from pQE30ss_Ac1_corr into the backbone of plasmid pHL3-HRS3opt2#2(Friedel et al. 2015) via Sfil/Notl.

(19) All plasmids encoding Nipah virus G protein variants were derived from plasmid pCAGGS-NiV-codonop-Gn. The coding sequence for the Ac1 targeting domain was fused to the C-terminus of the G protein reading frame by PCR amplification of each fragment and simultaneously introducing a common Agel restriction site, which was used for ligation resulting in plasmid pCAGGS-NiV-G-DARPin-Ac1. All other targeting domains were exchanged via Agel/Notl. Truncations of the G protein cytoplasmic tail were introduced by PCR amplification of the G protein reading frame and insertion of the PCR fragments into pCAGGS-NiV-G-DARPin-Ac1 resulting in plasmids pCAGGS-NiV-GcΔ33-DARPin-Ac1 and pCAGGS-NiV-GcΔ34-DARPin-Ac1.The His-tagged G and GcΔ34 proteins were generated by PCR amplification from pCAGGS-NiV-codonop-Gn. The fragments were cloned via Pacl/Notl restriction into the plasmid backbone of pCAGGS-NiV-G-DARPin-Ac1 resulting in pCAGGS-NiV-G-His and pCAGGS-NiV-GcΔ34-His, respectively. Mutations interfering with natural receptor recognition were introduced into the NiV-GcΔ34-DARPin-Ac1 protein coding sequence by site-directed mutagenesis. Each mutation was generated by amplification of two fragments carrying the designated mutation with homologous regions at the mutation site. These fragments were fused and amplified by a flanking primer pair. Resulting fragments were cloned into pCAGGS-NiV-GcΔ34-DARPin-Ac1 via Rsrll/Agel, generating the plasmids pCAGGS-NiV-GcΔ34EpCAMmut.

(20) For the generation of the NiV-F variants, the coding sequences for FcΔ22 and FcΔ25 were amplified from pCAGGS-NiV-F and cloned via Pacl/Sacl restriction into the plasmid backbone of pCAGGS-NiV-codonop-Gn resulting in the plasmids pCAGGS-NiV-FcΔ22 and pCAGGS-NiV-FcΔ25. AU1 tagged NiV-F variants used for Western blot analysis of vector particles were generated by amplifying the NiV-F variants from pCAGGS-NiV-F and simultaneously adding the AU1 tag N-terminally. The resulting PCR fragments were cloned via Pacl/Sacl restriction digest into the backbone of pCAGGS-NiV-codonop-Gn, resulting in the plasmids pCAGGS-AU1-NiV-F, pCAGGS-AU1-NiV-FcΔ22, and pCAGGS-AU1-NiV-FcΔ25.

(21) Vector Production

(22) Vector particles were generated by transient transfection of HEK-293T cells using polyethylenimine (PEI). Twenty-four hours before transfection, 2.5×107 cells were seeded into a T175 flask. On the day of transfection, the cell culture medium was replaced by 10 ml DMEM with 15% FCS and 3 mM L-glutamine. The DNA mix was prepared by mixing 35 μg of total DNA with 2.3 ml of DMEM without additives.

(23) For example, following optimization of G to F ratios, 0.9 μg of plasmid encoding for GcΔ34DARPin/scFv variants was mixed with 4.49 μg plasmid encoding for F variants. 14.4 μg of HIV-1 packaging plasmid pCMVΔR8.9 and 15.1 μg of transfer-LV plasmids were used. The transfection reagent mix was prepared by mixing 140 μl of 18 mM PEI solution in H2O with 2.2 ml DMEM without additives. This solution was mixed with the DNA mix, vortexed, incubated for 20 minutes at room temperature and added to the HEK-293T cells, resulting in DMEM with 10% FCS, 2 mM L-glutatmine in total. 24 h later, the medium was exchanged with DMEM with 10% FCS, 2 mM L-glutatmine in order to remove remaining PEI/DNA complexes. At day two post transfection, the cell supernatant containing the lentiviral vectors was filtered through a 0.45 μm filter. If needed, vector particles were purified by centrifugation at 450×g for 24 h over a 20% sucrose cushion. The pellet was resuspended in phosphate-buffered saline (PBS). For transduction, 8×103 of CHO-EpCAM and SK-OV-3 cells or 2×104 Molt4.8 and Raji cells were seeded into a single well of a 96-well-plate and transduced on the next day. When needed, medium of cells was replaced with medium containing different concentrations of Bafilomycin A1 (Santa Cruz Biotechnology, Inc, Dallas, USA) and cells were pre-incubated 30 min at 37° C. before vector was added. For titration, at least four serial dilutions of vector particles were used. After 72 h, the percentage of green fluorescentprotein (GFP)-positive cells was determined by flow cytometry. Transducing units/ml (t.u./ml) was calculated by selecting the dilutions showing linear correlation betweendilution factor and number of GFP-positive cells (number of transduced cells/volume of vector in μl/0.001).

(24) Schematic presentations of the pseudotyped retrovirus-like particle or retroviral vector according to the invention are depicted in FIGS. 7 and 8.

Example 1

Selective Activation of CD4+ T Cells

(25) In order to activate CD4.sup.+ but not CD8.sup.+ lymphocytes, VLPs were generated that display a CD4-specific DARPin as targeting ligand and IL-7 as activation domain on the MV H protein along with the fusion protein (F) resulting in the 4.sup.H/IL7.sup.H-VLP (comprising proteins of sequence SEQ ID NO: 2, 14 and 15). To generate the particles, a transfection protocol was established. Briefly, 0.45 μg of pCG-Hmut-CD4-DARPin (Zhou et al., J Immunol, 2015), 0.45 μg of pCG-HΔ15-IL7 (provided by Els Verhoeyen), 4.7 μg of pCG-FcΔ30 (Funke et al., Molecular therapy, 2008), 14.4 μg of HIV-1 packaging plasmid pCMVΔR8.9 (Funke et al., Molecular therapy, 2008), and 15.1 μg of pCG-1 were used for transfection of HEK293T cells in one T175 flask. After harvesting, the vector particles were further concentrated via ultracentrifugation. Titers of ˜10.sup.7 to/ml were obtained for LVs and titers of ˜1.5 ug p24/ml were obtained for VLPs.

(26) The function of 4.sup.H/IL7.sup.H-VLP was demonstrated on freshly isolated resting T cells. Briefly, CD3.sup.+ T cells were isolated by negative selection using Pan T cell isolation kit (Miltenyi Biotech) from adult peripheral blood. Subsequently, the cells were incubated with 4.sup.H/IL7.sup.H-VLP in the absence of any stimulation. Three days later, the CD4.sup.+ T cells but not the CD4.sup.− T cells in the cell culture were activated, as confirmed by FACS analysis for the CD71 activation maker (FIG. 3). Non-targeted particles displaying IL-7 activated both, CD4.sup.+ T cell and CD4.sup.− T cells. IL-7 is known to be a T cell survival cytokine. Therefore, it was demonstrated that all IL7-displaying VLPs were as effective as recombinant human IL-7 at preventing the death of primary T cells, while most of the resting T cells placed in culture were dead after six days in the presence of conventional IL7-deficient VLP particles (FIG. 4).

(27) The capacity of promoting T cell survival and selectively activating CD4.sup.+ T cells was also acquired by IL7-displaying CD4-targeted LVs (4.sup.H/IL7.sup.H-LV). Viral particles were produced as described above, except that pCG-1 was substituted by the transfer plasmid encoding GFP. In addition, in order to test the flexibility of the system and if the function of particles could be enhanced by using other version of Hmut, two different plasmids encoding Hmut-IL7 were used for vector production. pCG-HΔ15-IL7 is with 15 aa cytoplasmic truncation of Hmut-IL7 (for 4.sup.H/IL7.sup.HΔ5-LV production) and pCG-HΔ20-IL7 is with 20 aa cytoplasmic truncation of Hmut-IL7 (for 4.sup.H/IL7.sup.HΔ20-LV production). Without specific indication, IL7-displaying VLP and LV were produced with pCG-HΔ15-IL7. Similar to what was achieved by 4.sup.H/IL7.sup.H-VLP, when freshly isolated human CD3.sup.+ T cells were incubated with the indicated LVs in the absence of stimulation for 6 days, 4.sup.H/IL7.sup.H-LV significantly promoted T cell survival compared to the parental 4.sup.H-LV and non-targeted Hnse-LV (FIG. 5). As expected, the highest percentage of viable cells was observed in rhIL-7 treated positive group and the least viable cells were in the VSV-LV transduce or untransduced (ut) groups (FIG. 5). Moreover, 4.sup.H/IL7.sup.H-LV was as potent as 4.sup.H/IL7.sup.H-VLP in selectively stimulating its target cell population in the mixed cell culture. As shown in FIG. 6, 4.sup.H/IL7.sup.H-LV selectively upregulate activation maker CD69 and CD71 expression CD4.sup.+ cells but not CD4.sup.− cells. There were no differences between 4.sup.H/IL7.sup.HΔ15-LV and 4.sup.H/IL7.sup.HΔ20-LV in CD4.sup.+ T cell stimulation. Therefore, functional IL7-displaying targeted LVs are successfully generated.

Example 2

Delivery of Tumor-Specific Chimeric Antigen Receptors into Resting T Cells

(28) Since IL7-displaying vectors triggered the resting T cells activation, it is expected that these cells are permissive for lentiviral transduction. In order to prove it, two versions of 4.sup.H/IL7.sup.H-LV delivering GFP transgene were generated. As shown in FIG. 9, both 4.sup.H/IL7.sup.HΔ15-LV and 4.sup.H/IL7.sup.HΔ20-LV can efficiently and selectively transduce resting CD4.sup.+ T cells in the cell mixture.

(29) Chimeric antigen receptors (CARs) are a powerful tool for cancer therapy. So far, T cell subsets must be purified and activated for genetic delivery of CARs. Here, it is demonstrated that CARs can be delivered selectively into resting CD4.sup.+ cells. The particles were generated as described in example 1 except that pCG-1 was substituted by the CAR encoding transfer plasmid. When resting T cells were transduced with 4.sup.H/IL7.sup.H-LV delivering ErbB2-specific chimeric antigen receptors (ErbB2-CAR), ErbB2-CAR expression was only observed in the CD4.sup.+ T cell population. Compared to the IL7-deficienct CD4-targeted LV (4.sup.H-LV), the 4.sup.H/IL7.sup.H-LV was more efficient in delivering the CAR gene while retaining the selectivity for the CD4.sup.+ T cells (FIG. 1).

Example 3

The MV Glycoproteins can be Replaced by those of NiV to Selectively Activate and Transduce Resting Human CD8+T Cells

(30) In order to efficiently pseudotype VLPs or LVs with NiV glycoproteins, the NiV-G was truncated at its cytoplasmic tail by 34 amino acids (GcΔ34) and the NiV-F was truncated at its cytoplasmic tail by 22 amino acids (FΔ22). Next, the natural receptor recognition of GcΔ34 for Ephrin-B2/B3 was destroyed by introducing four point mutations (E501A, W504A, Q530A, E533A) into GcΔ34. LVs pseudotyped with this engineered G protein completely lost binding to the natural Ephrin-B2 and -B3 NiV receptors (FIG. 2) and did not enter into cells expressing these receptors.

(31) Moreover, NiV-pseudotyped LVs have some attractive features, like high production yield and resistance to intravenous immunoglobulins. Since there is no vaccination against NiV and the few outbreaks are limited to a few cases in Malaysia, Bangladesh and India (SEARO|WHO South-East Asia Region), there should be no neutralizing antibodies present in humans. To demonstrate this, intravenous immunoglobulin (IVIG; Intratect®) covered the widest range of human serum donors, was incubated with NiVwt-LV, NiVmut.sup.EpCAM-LV, MV.sup.EpCAM-LV and VSV-LV at increasing concentrations prior to the transduction of target cells. GFP expression was then determined by flow cytometry three days post transduction. As expected, transduction mediated by VSVG-LV and NiVwt-LV was unaffected by the treatment with IVIG. MV.sup.EpCAM-LV, on the other hand, showed a dose-dependent decrease in transduction rates with a complete neutralization at 100 μg/ml of IVIG. In contrast, NiVmut.sup.EpCAM-LV was resistant against IVIG at all concentrations used and must thus be at least 10000-fold less sensitive against human immunoglobulin than the corresponding MV-based vector (FIG. 10). These results show that receptor-targeted vectors based on NiV glycoproteins will not be neutralized when injected into humans.

(32) Therefore, due to the above features of the NiV-G, the NiV-pseudotypes were further used to generate cytokine-displaying CD8-targeted particles. Here, a CD8 specific scFv derived from OKT8 was displayed on the NiV-GΔ34mut4 to generate envelope plasmid pCG-Gmut-CD8scFv and human IL7 was displayed on the NiV-GΔ34mut4 to generate envelope plasmid pCG-Gmut-IL7. In order to produce 8.sup.G/IL7.sup.G-LV, 0.45 μg of pCG-Gmut-CD8scFv, 0.45 μg of pCG-Gmut-IL7, 4.7 μg of pCG-FcΔ22, 14.4 μg of HIV-1 packaging plasmid pCMVΔR8.9, and 15.1 μg of transfer-LV plasmids were used for transfection of HEK293T cells in one T175 flask. Titers of ˜10.sup.7 to/ml were obtained for LVs and titers of ˜1.5 ug p24/ml were obtained for VLPs.

(33) Similar to what was achieved by 4.sup.H/IL7.sup.H-LV, 8.sup.G/IL7.sup.G-LV (comprising proteins of sequence SEQ ID NO: 4, 6 and 16) was able to selectively stimulate and transduce CD8.sup.+ T cells in the peripheral blood mixture. As shown in FIG. 11, when freshly isolated human CD3.sup.+ T cells were incubated with 8.sup.G/IL7.sup.G-LV for 3 days, not only the CD8.sup.+ T cell subsets in the culture upregulated the CD71 expression but also only this cell population efficiently expressed the reporter transgene GFP (FIG. 12) or therapeutic transgene CAR19 (FIG. 13). In addition, compared to the parental IL7-deficient CD8-targeted LV (8.sup.G-LV), the 8.sup.G/IL7.sup.G-LV was more efficient in activating the target cells and delivering transgenes without compromise on specificity for transduction of the CD8.sup.+ T cells.

Example 4

Identification of an Optimal Ratio of CD8-Targeting NiV-G to IL7-Displaying NiV-G

(34) In order to hit more target cells, improve the titer and enhance the specificity of the particles for selective activation and transduction of a distinct cell population, while maximally reduce the off-target effect, the transfection protocol is further optimized for vector production and carefully determine the amounts of NiV-G.sup.targeting domain (or MV-H.sup.targeting domain) and NiV-G.sup.functional domain (MV-H.sup.functional domain) encoding plasmids maintained in the packaging cells.

(35) For the production of G.sup.8/G.sup.IL7-LV, as described in Example 3, total amounts of 0.9 μg of pCG-Gmut plasmids together with 4.7 μg of pCG-FcΔ22, 14.4 μg of HIV-1 packaging plasmid pCMVΔR8.9, and 15.1 μg of pSEW were used for transfection of HEK293T cells in one T175 flask. Meanwhile, pCG-Gmut-CD8scFv and pCG-Gmut-IL7 plasmids are mixed at different ratios at 20:1, 10:1, 5:1, 1:1, or 1:5. The respective cell supernatants containing 8.sup.G/IL7.sup.G-LVs are used for transduction of CD8.sup.+ Molt or A301 cells and, 48 h after transduction, the percentage of GFP cells are measured by FACS to calculate the titers. In addition, 8.sup.G/IL7.sup.G -LVs are used for transduction of freshly isolated CD3.sup.+ T cells and then the CD71 and GFP expression in CD8.sup.+ and CD8.sup.− cells are determined. The best ratio is identified as giving the highest titer, highest expression of CD71 and GFP in target population and lowest expression in non-target population.

Example 5

Characterization of the Transduced Resting T Cells by IL7-Displaying Targeted VLPs and LVs

(36) T cell compartments are highly heterogeneous and T cell subsets are phenotypically and functionally distinct. As far as T cell therapy for cancer is concerned, less differentiated cells are usually correlated with better antitumor effect in vivo. Therefore, it's of importance to identify the T cell subsets that activated or modified by the vector particles according to the invention. Take 8.sup.G/IL7.sup.G-LV-CAR for example, the phenotypes of the transduced resting T cells should be characterized in comparison to transduced pre-stimulated T cells (with most widely used protocol of CD3/CD28 stimulation). Briefly, 3 days after transduction of resting or pre-stimulated CD3.sup.+ T cells, the expression of multiple T cell markers is monitored, including CD11a, CD11b, CD25, CD27, CD28, CD45RA, CD45RO, CD62L, CD69, CD71, CD95, CD127, and CCR7. These molecules are chosen as they have been used in the past to differentiate CD45RA+CD45RO.sup.−CD62L.sup.highCD95.sup.−CD27.sup.highCCR7.sup.high naïve (T.sub.N), CD45RA.sup.+CD45RO.sup.−CD62L.sup.highCD95.sup.+CD27.sup.highCCR7.sup.high stem cell memory (T.sub.SCM), CD45RA.sup.−CD45RO.sup.highCD62L+CD95.sup.+CD27.sup.+central memory (T.sub.CM), CD45RA.sup.−/+CD45RO.sup.highCD62L.sup.−CD95+CD27.sup.−/+CCR7.sup.− effector memory (T.sub.EM), and CD45RA.sup.−/+CD45RO.sup.+CD62L.sup.−CD95.sup.highCD27.sup.−CD28.sup.−CCR7.sup.− effector (T.sub.E) T cells. The phenotypes of CAR.sup.+ T cells are analyzed by flow cytometry.

(37) In addition, pre-sorted CD4+ T cell subsets (T.sub.SCM, T.sub.CM, T.sub.EM, and T.sub.E) are incubated with 4.sup.H/IL7.sup.H-VLP for example, to demonstrate the advantages of current invention for T cell activation over the conversional TCR stimulation. Briefly, 3 days after incubation with 4.sup.H/IL7.sup.H-VLP or stimulation by CD3/28 antibodies, above indicated T cell subsets are analyzed for phenotype skewness and activation levels. It is expected that IL7-displaying T cell targeted vector particles activate and transduce less differentiated T cells.

Example 6

Enhance the Function of Ex Vivo Produced CART Cells and Simplify the CART Cell Production Procedure

(38) Adoptive T cell immunotherapy has been demonstrated to be an effective regime in clinic for treatment of various types of diseases. However, its broader application has been limited by several issues. On one hand, the production of CAR-engineered or TCR-engineered T cells is very expansive and labor-intensive due to the in vitro stimulation, long-term expansion, and cell isolation and purification. On the other hand, the effector T cells that have been expanded in vitro often persist poorly in vivo, and fail to exhibit a sustained antitumor effect. The current invention provides the potential solutions for these challenges by combining the cell stimulation and gene transfer in a single step and enabling resting T cell transduction. Using this invention, the T cell production process can be significantly simplified therefore reducing the cost of T cell manufacture. Moreover, compared to the conventionally produced cells, CAR/TCR-engineered T cells produced by current invention can be more potent in vivo, due to the potential of efficiently engineering less differentiated cells as demonstrated in Example 5.

(39) To demonstrate these features, freshly isolated human PBMC or CD8.sup.+ T cells are transduced with for example 8.sup.G/IL7.sup.G-LV-CAR or VSV-LV-CAR as control. Meanwhile, conventional CAR T cells are prepared in parallel according to the most widely uses protocol. Briefly, isolated human PBMC or CD8.sup.+ T cells from the same donor are simulated with CD3/CD28 antibodies and IL2. Then cells are transduced with 8.sup.G/IL7.sup.G-LV-CAR or VSV-LV-CAR followed by 10 days expansion in the presence of IL2. 2-3 days after vector transduction of resting T cells or 12 days after transduction of stimulated T cells, these cells are co-cultivated with target tumor cells in vitro or infused in the tumor bearing humanized mouse. Afterwards, in vitro tumor cells lysis and in vivo tumor remission, animal survival, tumor-reactive T cell persistence and proliferation are assessed.

Example 7

Selective Activation and Transduction by IL7-Displaying Targeted Particles in In Vivo like Conditions

(40) In a next step, the feasibility of using the viral particles to target and functionally modify the distinct cells in in vivo settings is investigated. Thus, transduction of fresh human peripheral blood with the viral particles is performed, for example with G.sup.8/G.sup.IL7-LV. This allows evaluation of targeted activation and gene transfer in CD8.sup.+ T cell population in the presence of an active human complement system, an obstacle encountered by viral particles in vivo.

(41) Briefly, fresh peripheral blood is incubated with 8.sup.G/IL7.sup.G-LV-GFP or other control vectors (VSV-LV, 8.sup.H/IL7.sup.H -LV, 8.sup.G-LV, or VSV/IL7.sup.G-LV) for 6-8 h. Afterwards, total PBMC are isolated from the blood and further cultured in the T cell culture medium without adding any stimulatory reagents. After 3-4 days, CD8.sup.+ and CD8.sup.− cells in the culture are evaluated for CD71 and GFP expression. To confirm the stable gene transfer, a small fraction of cells are transferred into the culture medium supplemented with CD3/CD28 antibodies and IL2 for additional 3 days before the percentages of GFP cells are analyzed.

Example 8

IL7-Displaying Targeted Particles Selectively Activate and Transduce Circulating Resting T Cells in Humanized Mouse Models

(42) In order to demonstrate the particles according to the invention are able to be locally or systemically applied in vivo to allow the cell type specific activation or modification, they are applied in mice engrafted with human CD34.sup.+ hematopoietic stem cells (HSC) from healthy donors. In this mouse model, multilineage human hematopoietic cells can be generated and these cells are tolerated by the mouse host.

(43) Here, the application of G.sup.8/G.sup.IL7-pseudotpyed particles for human CD8.sup.+ T cell activation and modification is taken for example. Briefly, after having confirmed that the human immune system was successfully established by detecting around 10% of human CD3.sup.+ T cells, G.sup.8/G.sup.IL7-pseudotpyed particles are applied systemically via intravenous injection (iv) or locally via intrasplenic or intrathymic injection. G.sup.8/G.sup.IL7-VLP is injected to assess the feasibility of selective activation of human CD8.sup.+ T cells in vivo. In this experiment setting, the activation level/duration and CD8.sup.+ T cells proliferation are analyzed at the different time points after VLP injection. G.sup.8/G.sup.IL7-LV-GFP is injected to assess the function of in vivo gene delivery. In this experiment setting, the GFP expression, the phenotypes of transduced cells, and their proliferation and persistence are analyzed at the different time points after LV injection. G.sup.8/G.sup.IL7-LV-CAR is injected to assess the feasibility of therapeutic gene delivery and the in vivo generation of CAR T cells. In this experiment setting, the CAR expression, the phenotypes of transduced cells, the proliferation and persistence of transduced cells in the presence/absence of target tumor cells/antigens, the elimination of local/systemic tumor are analyzed at the different time points after LV injection.

Example 9

Selective Activation of Rhesus Macaque CD4.SUP.+ T Cells

(44) T cells of non-human primate (NHP) are phenotypically similar to human T cells, making it the best animal model for human immunity and immunotherapies. As far as T cell therapy is concerned, the NHP model could be more reliable in predicting the effect, dosage, application routes, and safety of potential medicaments. Therefore, a tool to specifically modify or inducing a functional change in a distinct cell type in NHP is highly desirable.

(45) Here, the 4.sup.H/IL7.sup.H-SIV vector and its application for rhesus macaque are described in detail as an example. First, a rhesus macaque CD4-specific DARPin57.2 (CD4.sup.D572) is used as targeting domain displayed on the Hmut protein (H-CD4.sup.D572) and the rhesus macaque reactive IL7 is used as functional domain displayed on the Hmut (H-rmIL7). Then the above two H proteins, for example with the ratios determined in Example 4, are used to pseudotype simian immunodeficiency virus mac 251 (SIVmac251)-derived LV. After determining the vector yield and titers by P27 ELISA/Nano-sight and transduction of CD4.sup.+ simian cells, freshly isolated simian CD3.sup.+ T cells are transduced with 4.sup.H/IL7.sup.H-SIV delivering GFP or CARs. Then, the activation makers and transgene expression in CD4.sup.+ vs CD4.sup.− T cells was determined as described above. Next, the possibility of in vivo application is assessed in NHP models. For this purpose, a single dose of vectors is injected slowly into two axillary lymph nodes of each of the enrolled animals (eg. rhesus macaque). One week post-injection, blood samples are collected and one of the two injected lymph nodes is removed by surgery and dissociated into single cell suspensions. The detailed analysis is performed to analyse the phenotypes, specific activation, transgene expression levels and/or function (when therapeutic transgene is delivered) of in vivo transduced T cells. Moreover, the second lymph node receives a second dose of the same vector and the enrolled animals are monitored for at six months to assess the proliferation, persistence, and/or function (when therapeutic transgene is delivered) of in vivo transduced T cells.

Example 10

Selectively Active and Expand Antigen-Specific T Cells in Immunocompetent Mouse Model

(46) The potential of targeting cytokine to function at the disease site while avoiding the systemic toxicity can be well evaluated in an immunocompetent mouse melanoma tumor model. IL12 displaying murine CD8-targeted VLP (m8.sup.G/mIL12.sup.G-VLP) is taken as an example. Previous studies have shown that IL12 application can enhance the transferred tumor-reactive T cells tolerance and antitumor efficacy, but it is very toxic systemically. To overcome this drawback, in our study, m8.sup.G/mIL12.sup.G-VLP is used to control the function of IL12 to take place at the tumor sites. Wide type 057/Bl6 mouse is transplanted with B16/OVA melanoma cells, after the establishment of tumor, the mouse is treated by irradiation and transferred with T cells isolated from C57/Bl6 OT-I transgenic mouse. After OT-I T cell transplantation, the mice are treated with m8.sup.G/mIL12.sup.G-VLP, or with murine IL12, or with PBS as controls via i.v. injection. The tumor volume, in vivo persistence and function of OT-I T cells, and mouse survival are analyzed.

Example 11

In Vivo Targeted Targeted Activation of Human CD4+ T Cells.

(47) In order to evaluate the in vivo performance of the MV-4.sup.H/IL7.sup.H-LVs for activating human CD4.sup.+ T cells, NOD/SCID gc−/− (NSG) mice were engrafted with CD3.sup.+ cord blood T cells. Human T cell reconstitution was weekly determined in the blood. When 20% of hT cells was detected (% hCD3.sup.+ cells/total hCD45.sup.++mCD45.sup.+ cells), the mice were injected IV with 1E6 IU MV-4.sup.H/IL7.sup.H-LVs or MV-4.sup.H-LVs. Two weeks post-injection of the vectors, the mice were sacrificed and the splenocytes and peripheral blood mononuclear cells were isolated. Remarkably, it was found a stronger specific IL-7 activation revealed by a CD71 late activation marker of the human CD4+ T cells in case of the MV-4.sup.H/IL7.sup.H-LVs (25% CD4+ CD71+ cells) as compared to MV-4.sup.H-LVs (8% CD4+ CD71+ cells) (FIG. 14). Moreover, the CD4-T cells, which represent the CD8+ T cells, show identical expression of CD71 for both targeting vectors. The latter emphasizes that the IL-7 survival signaling was only activating the targeted CD4+ T cells and not the CD8+ cells. In conclusion, next to the in vitro specific activation of the target cells achieved with MV-4.sup.H/IL7.sup.H-LVs, the specific activation of CD4+ T cells by MV-4.sup.H/IL7.sup.H-LVs was confirmed in vivo in human blood system mouse model.