Self-inactivating Gammaretroviral Vectors for Gene Therapy of X-linked Severe Combined Immunodeficiency

Gammaretroviral SIN vectors for gene therapy of SCID-X1

SIN gammaretroviral vectors incorporating the human IL2RG complementary DNA (cDNA) transgene were constructed using the previously described SRS11 vector backbone (Figure 1a).¹⁷ These vectors contain a Rous sarcoma virus promoter within the 5′ LTR enabling the production of high-titre gammaretro-viral supernatants. After an initial evaluation of several regulatory elements including the endogenous Wiskott–Aldrich syndrome and human phosphoglycerate kinase gene promoters (data not shown), two were chosen for further study. These were the U3 region of the SFFV LTR (SFFV promoter) and the short form (intron-deleted) of the human elongation factor 1-α promoter (EFS), both of which have been demonstrated to regulate high levels of transgene expression in hematopoietic lineages, including human CD34⁺ cells.^18,19,22 In addition, a safety-modified post-transcriptional regulatory element (devoid of the X-protein open-reading frame) was introduced to enhance titres and gene expression levels. The IL2RG cDNA (equipped with a Kozak consensus sequence) and post-transcriptional regulatory element were verified by sequencing.

Vector characterization in cell lines and wild-type lineage negative cells

First, the performance of the new SIN vectors was compared to MFG γc, the LTR-regulated gammaretroviral vector used in the SCID-X1 clinical trials. Figure 2a shows a Northern blot analysis of transduced SC1 cells (murine fibroblast cell line) indicating the expected RNA processing. MFG γc, which harbors an intron within the leader region, shows spliced and unspliced transcripts while the SIN vectors (SRS11 derivatives on right) show only the transcript initiated by the internal promoter. To assess expression levels, murine fibroblasts (SC1), cytotoxic T cells (CTLL-2, constitutively expressing IL-2 receptors) and murine lin^– cells (C57/Bl6 wild-type background) were transduced. To ensure that the majority of cells contained only single integration events, transduction efficiencies <35% were used (Figure 2b). Whilst the SRS11 SF γc vector clearly mediated the highest expression in SC1 cells, differences in expression strength were less prominent between vectors in transduced CTLL-2 cells. Following transduction of murine lin^– cells, the level of γc expression in MFG γc-transduced cells was slightly superior to the levels observed following transduction with either of the SIN vectors. Taken together, expression differences between vectors were minor and dependent on cell type.

In vitro differentiation of T and B lymphocytes from transduced murine Il2rg^−/− lin^– cells

To assess the ability of the SIN gammaretroviral vectors to restore the lymphoid differentiation potential of γc-deficient hematopoietic progenitors, lin^– cells isolated from the bone marrow of Il2rg^−/− mice were transduced with the SIN vector series, MFG γc or a control vector encoding enhanced green fluorescent protein (eGFP) (SF91GFPpre) and seeded onto OP9-eGFP or OP9-DL1 stromal layers, which provide a convenient system for the in vitro development of B or T lymphocytes, respectively.^23,24 Untransduced C57/Bl6 wild-type lin^– cells were also seeded as controls. The γc-deficient lin^- cells transduced with either of the SIN gammaretroviral vectors or MFG γc were found to have restored lymphoid differentiation potential and to have undergone in vitro B- or T-cell development following 13 days of co-culture on OP9 stromal layers at a similar rate as observed for C57/Bl6 wild-type cells (Figure 3a and b). Furthermore, the transduced γc-deficient lin^– cells were able to develop into mature CD4⁺ and CD8⁺ T lymphocytes following 31 days of co-culture on OP9-DL1 stromal layers (Figure 3c). γc-Deficient lin^– cells transduced with the control SF91GFPpre vector however, failed to undergo lymphoid differentiation under identical culture conditions and were no longer viable at the time of analysis. These results demonstrate that both the SFFV and EFS promoters in the context of gammaretroviral SIN vectors are able to regulate expression of the human γc transgene in murine γc-deficient hematopoietic progenitors to sufficient levels that restore lymphoid development in vitro.

Reconstitution of T, B and natural killer cells following ex vivo gene therapy of the SCID-X1 murine model

The SIN gammaretroviral vectors were subsequently tested in vivo using a murine model of SCID-X1. In contrast to SCID-X1 patients, Il2rg^−/− mice develop a proportion of T lymphocytes and are susceptible to inflammatory bowel disease. A second strain, Il2rg^−/−Rag2^−/−c5^−/− mice, were therefore used as transplant recipients as these animals have a stable, alymphoid phenotype and are a convenient model for reconstitution studies.²⁵ γc-Deficient lin^– cells transduced twice with ecotropic retroviral particles at a defined multiplicity of infection (MOI) were engrafted into sub-lethally irradiated, Il2rg^−/−Rag2^−/−c5^−/− recipient mice. To assess lin^– cell transduction efficiency, a proportion of transduced progenitors were seeded in semi-solid medium supplemented with murine cytokines and cultured for ~2 weeks, after which colonies were picked and assessed for the presence of integrated provirus by polymerase chain reaction (PCR) using primers specific to the IL2RG transgene. High transduction levels were achieved for both the SIN gammaretroviral vectors and the LTR-regulated vector, with between 79 and 91% of colonies found to harbor integrated provirus (Figure 4a).

Lymphoid reconstitution of transplanted Il2rg^−/−Rag2^−/−c5^−/− mice was confirmed by the detection of lymphocytes in tail-bleed samples 15–19 weeks post-engraftment, with similar levels of T and B lymphocytes detected for animals transplanted with SIN gammaretrovirally transduced (n = 14) or MFG γc–transduced (n = 6) γc-deficient lin^– cells (data not shown). Recipient mice that had been injected with Il2rg^−/− lin^– cells transduced with the control SF91GFPpre vector had no detectable circulating B or T lymphocytes (n = 4); eGFP⁺ cells were however detected in the periphery of two out of four animals (data not shown).

Engrafted mice were killed 23 weeks post-transplantation and the spleens, bone marrow, and thymi assessed for lymphoid reconstitution. Analysis by flow cytometry revealed the presence of T, B, and natural killer cell populations in transplanted animals, with comparable levels of lymphoid reconstitution observed for the SIN vectors (SRS11 SF γc and SRS11 EFS γc) and the LTR-regulated vector (MFG γc) (Figure 4b and c and Table 1). By contrast, Il2rg^−/−Rag2^−/−c5^−/− mice transplanted with SF91GFPpre-transduced Il2rg^−/− progenitors failed to reconstitute (Figure 4d). The proviral copy number in the spleens of transplant recipients was determined by quantitative PCR. A comparable mean vector copy number was observed for mice transplanted with SRS11 SF γc–transduced (0.29 per cell), SRS11 EFS γc–transduced (0.47 per cell) or MFG γc–transduced (0.27 per cell) γc-deficient progenitors, indicating that lymphoid reconstitution following ex vivo gene therapy with either of the SIN gammaretroviral vectors is achievable at relatively low vector copy number (Figure 4e).

Restoration of lymphocyte function following ex vivo gene therapy of the SCID-X1 murine model using SIN gammaretroviral vectors

To assess functionality of the restored T lymphocytes in the SIN vector transplanted mice, splenocytes were stimulated in vitro with the mitogen concanavalin A alone or in the presence of the γc-dependent cytokine, IL-2. In reconstituted Il2rg^−/−Rag2^−/−c5^−/− mice, splenocytes were capable of proliferation in response to concanavalin A at levels higher than those observed for splenocytes from Il2rg^−/− donor mice, with significantly higher levels of proliferation observed for the SRS11 EFS γc group (Figure 5a). Furthermore, in all but two of the transplanted mice tested, IL-2 further stimulated the concanavalin A mitogenic response, with significantly increased levels of proliferation observed for animals reconstituted with lin^– cells transduced with either SIN vector. It is worth noting, however, that the basal proliferation levels for Il2rg^−/− mice and transplant recipients were significantly higher than for wild-type C57/Bl6 mice; consequently, even though the absolute increase in proliferation was equivalent to that seen in C57/Bl6 animals for several experimental mice, the relative increase as expressed here is lower.²⁶ These data demonstrate that SIN gammaretroviral human γc gene transfer in this model is able to restore expression of functional IL-2 receptor complexes on murine T cells.

As a measure of restoration of humoral immunity in reconstituted MFG γc and SRS11 EFS γc mice, circulating plasma immunoglobulin (Ig) levels were analyzed by enzyme-linked immunosorbent assay (ELISA). γc-Deficient mice exhibit abnormal serum Ig levels reflecting the defective B-cell development in these animals, with levels of IgG1 and IgG2a in particular diminished in this model.²⁷ Following ex vivo gene therapy however, levels of IgG and IgG1 in the plasma of mice reconstituted with MFG γc or SRS11 EFS γc retrovirally transduced cells had increased as compared to levels in Il2rg^−/− donor mice, with reconstitution of serum IgG2a levels appearing more efficient for the SRS11 EFS γc mice (Figure 5b). IgG1 class-switching is dependent on IL-4 secreted by T_H2 CD4⁺ T-helper cells, hence these data in particular indicate restoration of both T-lymphocyte function and γc-containing cytokine receptors on B lymphocytes in the reconstituted mice.²⁸

Restoration of γc-mediated signaling in a SCID-X1 T-cell line following transduction with GALV-pseudotyped SRS11 EFS γc

The common cytokine receptor γc is shared by the receptors for IL-2, 4, 7, 9, 15, and 21, which signal through association with the JAK family of tyrosine kinases. In particular γc associates with JAK3 following ligand binding, which causes the subsequent phosphorylation of the transcription factor, STAT5. To assess whether the SIN vector containing the internal EFS promoter (SRS11 EFS γc) can express the γc transgene to therapeutic levels in human cells, the γc-deficient T-cell line, ED-7R, was transduced with GALV-pseudotyped gammaretrovirus. The transduced cells were analyzed for the presence of restored functional IL-2 receptor complexes by stimulation with IL-2 and subsequent analysis of downstream signaling by staining for phosphorylated STAT5. Following two rounds of transduction with SRS11 EFS γc at low MOI, ~20% of cells had restored IL-2-mediated signaling as assessed by the presence of phosphorylated STAT5, demonstrating that the SRS11 EFS γc vector is capable of directing γc expression in human cells to levels that enable the rescue of γc-dependent cytokine signaling (Figure 6).

Vector construction

The coding sequence for the human common γc (IL2RG) was amplified via PCR using primers 5′ IL2RG age (5′-CTGAACCGGTC GCCACCATGTTGAAGCCATCATTACCAT-3′; including a Kozak consensus sequence, restriction sites underlined) and 3′ IL2RG γ sal bam (5′-GCTAGGATCCGTCGACTCAGT TATCTAGGTTTCAGG CTTTAGGGTGT-3′), and verified by sequencing. To construct the gammaretroviral vector SRS11 SF γc, the γc cDNA was cloned into pSRS11 SF GFPpre,¹⁷ a SIN vector optimized for high-titre virus production and harboring a safety-modified post-transcriptional regulatory element (devoid of the X-protein) of the woodchuck hepatitis virus.³³ To design SRS11 EFS γc, the EF1-α short (EFS) promoter (242 base pairs)²² was cloned into SRS11 SF γc. As control vectors, we used pSF91GFPpre,^34,35 a retroviral vector harboring the eGFP transgene under control of the SFFV U3 promoter, and MFG γc, the conventionally used Moloney murine leukemia virus LTR-driven retroviral vector used in the clinical trial in Paris and London (kindly provided by Dr. M Cavazzana-Calvo, Hopital Necker-Enfants Malades, Paris, France).

Production of vector supernatants

Retroviral supernatant production was performed on 293T cells using Moloney murine leukemia virus gag/pol (M57-DAW) and GALV or ecotropic Moloney murine leukemia virus envelope helper plasmids and titrated on HT1080 or SC1 cells, respectively, as previously described.^16,36 For titration of γc vectors, transduced cells were stained with an anti-human CD132 antibody (BD Pharmingen, Oxford, UK) prior to analysis by flow cytometry. 293T, HT1080, and SC1 (murine fibroblast) cells were maintained in Dulbecco’s modified Eagle’s medium (Biochrom, Berlin, Germany) supplemented with 10% fetal calf serum (FCS), 100 U/ml penicillin/streptomycin, and 2 mmol/l glutamine.

Transduction of cell lines and primary cells

CTLL-2 (murine lymphoblast cell line, American Type Culture Collection TIB-214) and ED-7R cells (derived from an adult T-cell leukaemia line deficient in IL2RG gene expression³⁷) were cultured in Roswell Park Memorial Institute-1640 medium (Biochrom, Berlin, Germany), SC1 cells as described above. Both media were supplemented with 10% FCS, 100 U/ml penicillin/streptomycin, and 2 mmol/l glutamine. 1 × 10⁵ cells were transduced in 12-well plates in the presence of protamine sulfate using centrifugation at 2,000 r.p.m. for 1 hour. Transduction of CTLL-2 and SC-1 cells aimed for <35% positive cells (MOI of 1.5 and 0.3, respectively) to exclude multiple integrants in the majority of transduced cells. ED-7R cells were transduced using two rounds of transduction at an MOI of 3 over 2 days.

Bone marrow cells were harvested from femurs and tibias of C57/Bl6 mice (Charles River Laboratories, Wilmington, MA) and further purified by magnetic sorting using the Lineage Cell depletion kit (Miltenyi, Bergisch Gladbach, Germany) to obtain lin^– cells. Lin^– cells were cultured in StemSpan (Cell Systems, St. Katharinen, Germany) supplemented with 2 mmol/l glutamine, 200 U/ml penicillin/streptomycin, and cytokines [50 ng/ml murine stem cell factor, 100 ng/ml hFlt3L, 100 ng/ml hIL-11, 20 ng/ml mIL-3; all from PeproTech, London, UK], and pre-stimulated for 2 days. Transduction at an MOI of 3 was performed on RetroNectin-coated (TaKaRa, Otsu, Japan) dishes by spinoculation for 30 minutes at 4 °C. Cells were analyzed by flow cytometry (FACSCalibur, Becton-Dickinson, Heidelberg, Germany) 5 days after transduction.

Northern blot

Standard procedures were performed as previously described.²² Specific probes were directed against the coding sequence of IL2RG and GAPDH.

Co-culture of transduced lin^– cells on OP9 stromal layers

Early passage OP9-eGFP and OP9-DL1 stromal cells were grown to confluency in 10-cm tissue culture dishes in minimum essential media alpha (Gibco BRL, Invitrogen, Paisley, UK) supplemented with 20% FCS and 5% penicillin/streptomycin. 2 × 10⁴ stromal cells per well were seeded in 24-well tissue culture plates and left to adhere overnight. The following day the media was removed from the plates and 6,500–10,000 lin^– cells (transduced or untransduced) were seeded on top of the stromal layers in 2 ml of minimum essential media alpha supplemented with 12% FCS, 1 mmol/l HEPES, 1 mmol/l sodium pyruvate, 50 μg/ml gentamicin, 1% penicillin/streptomycin, 2 mmol/l Glutamax, 0.5 mmol/l 2-mercaptoethanol, and cytokines (0.5 ng/ml mIL-7, 5 ng/ml mFlt3L for OP9-DL1 co-cultures or 5 ng/ml mIL-7, 5 ng/ml mFlt3L for OP9-eGFP co-cultures). Every 4 days 1 ml of media was removed from the top of the plates so as not to disturb the differentiating lin^– cells and 1 ml of fresh co-culture media added. On days 13 and 31 (the final analysis of control C57/Bl6 cells was performed on day 15) the cells were removed from the plates by vigorous pipetting, strained through a 70-μm nylon cell strainer to remove the OP9 stromal layers, stained with antibodies to lymphocyte cell surface markers (BD Pharmingen, Oxford, UK) and analyzed by flow cytometry (CyAn ADP, DakoCytomation, Cambridgeshire, UK) for lymphoid differentiation.

Animals

Il2rg^−/− mice³⁸ were a kind gift from Dr. J DiSanto (Hopital Necker-Enfants Malades, Paris, France). Il2rg^−/−Rag2^−/−c5^−/− mice were generated by crossing Il2rg^−/−Rag2^−/−c5^+/+ mice³⁹ with A/J mice (Il2rg^+/+ Rag^+/+c5^−/−; Harlan UK, Bicester, UK). All experimental procedures were approved by the Institutional Research Ethics Committee (Institute of Child Health, University College London, UK) and performed according to UK Home Office Animal Welfare Legislation.

Ex vivo gene therapy of the SCID-X1 murine model

Bone marrow cells were harvested from the femurs and tibias of Il2rg^−/− or C57/Bl6 mice and the lin^– progenitors isolated using a StemSep mouse progenitor enrichment cocktail and negative selection columns as per the manufacturer’s instructions (StemCell Technologies, London, UK). Lin^– cells were pre-stimulated in StemSpan SFEM serum-free medium (StemCell Technologies) supplemented with 1% penicillin/streptomycin and cytokines (100 ng/ml murine stem cell factor, 100 ng/ml mFlt3L, 100 ng/ml hIL-11, 20 ng/ml mIL-3) for 2 days. Two rounds of transduction at an MOI of 1–2 were performed on RetroNectin-coated (TaKaRa, Otsu, Japan) dishes by spinoculation for 30 minutes at 950g at 4 °C. The transduced cells were harvested on day 5 and between 3 and 10 × 10⁵ transduced lin^– cells, resuspended in 300 μl Roswell Park Memorial Institute medium, were intravenously injected into the tail veins of sub-lethally irradiated Il2rg^−/−Rag2^−/−c5^−/− mice (6 Gy).

Lin^– cell transduction efficiency

To assess lin^– cell transduction efficiency, 1 × 10⁴ transduced lin^– cells were treated with 25 U of Benzonase nuclease to remove any contaminating plasmid DNA and seeded in 3 ml of methocult (StemCell Technologies) into two 10-mm tissue culture dishes and incubated at 37 °C for 14–21 days. Approximately 28 colonies per condition were picked using a P20 Gilson pipette, lyzed immediately in 20 μl DNA lysis buffer (10 mmol/l Tris 1 mmol/l EDTA pH 8.0, 0.5% NP40, 0.5% Tween 20, 1.25 mg/ml Proteinase K) and genomic DNA isolated. To confirm the presence of DNA, PCR was performed using primers to the murine house-keeping gene, Hprt, using primers HPRT-F: 5′-TCCCCAGACTTTTGATTTGC-3′ and HPRT-R: 5′-GGAAAATACAGCCAACACTGC-3′. To assess for the presence of integrated provirus, PCR was performed using primers γc-F (exon 3): 5′-CTGGCTGTCAGTTGCAAAAA-3′ and gammac-R (exon 8): 5′-GAGATAACCACGGCTTCCAA-3′ specific for the human IL2RG cDNA.

Analysis of immune reconstitution

Analysis of immune reconstitution by flow cytometry and T-cell proliferation assay was performed 3–5 months post-transplantation as previously described.²⁵

Ig ELISA

Serum isolated from the peripheral blood of transplanted mice was diluted appropriately and added to NUNC-Immuno ELISA plates (NUNC, Fischer Scientific UK, Loughborough, UK) pre-coated with capture antibody—IgG (AbD Serotec, Oxford, UK), IgG1 (BD Pharmingen), or IgG2a (BD Pharmingen). Plates were incubated for 1–2 hours at room temperature, washed three times with phosphate-buffered saline/0.05% Tween 20 and incubated with a horseradish peroxidase–conjugated detection antibody at room temperature for 1 hour. For the IgG ELISAs a biotinylated detection antibody was used followed by 30 minute incubation with a Strepavidin–horseradish peroxidase tertiary antibody. Plates were washed a further three times in phosphate-buffered saline/0.05% Tween 20 after which substrates A and B from an ELISA Duo-set substrate reagent pack (R&D Systems, Abingdon, UK) were mixed in equal quantities and 100 μl of the mixed solutions added to each well, plates were incubated at room temperature in the dark for ~10 minutes to allow color change, at which time the reaction was stopped by adding 50 μl stop solution/well (2 N H₂SO₄). The absorbance of each plate was then read at 405 nm using a FLUOstar Optima plate reader (BMG LABTECH).

Determination of gammaretroviral copy number by quantitative real-time PCR

To harvest genomic DNA, 1 × 10⁵ splenocytes were lyzed in 20 μl DNA lysis buffer (10 mmol/l Tris, 1 mmol/l EDTA pH 8.0, 0.5% NP40, 0.5% Tween 20, 1.25 mg/ml Proteinase K) and genomic DNA isolated. Quantitative real-time PCR for copy number was performed using primers and probes to the IL2RG cDNA and murine Titin gene as previously described.²⁵

STAT5 phosphorylation assay

This assay was performed as previously described.²⁵ The control ED-7R-γc cells have been described previously³⁷ and were maintained in Roswell Park Memorial Institute-1640 medium supplemented with 10% FCS, 100 U/ml penicillin/streptomycin, and 2 mmol/l glutamine.