Retrovirus-mediated transduction of primary ZAP-70- deficient ... - Nature

Gene Therapy (2000) 7, 1392–1400  2000 Macmillan Publishers Ltd All rights reserved 0969-7128/00 $15.00 www.nature.com/gt

VIRAL TRANSFER TECHNOLOGY

RESEARCH ARTICLE

Retrovirus-mediated transduction of primary ZAP-70deficient human T cells results in the selective growth advantage of gene-corrected cells: implications for gene therapy M Steinberg1, L Swainson1, K Schwarz2, M Boyer1, W Friedrich3, H Yssel4, N Taylor1 and N Noraz1 1

Institut de Geńe´tique Molećulaire de Montpellier, CNRS UMR 5535 (IFR 24), Montpellier, France; Departments of 2Transfusion Medicine and 3Pediatrics, University of Ulm, Ulm, Germany; and 4INSERM U454, Montpellier, France

Humans lacking the ZAP-70 protein tyrosine kinase present with an absence of CD8+ T cells and defective CD4+ T cells in the periphery. This severe combined immunodeficiency is fatal unless treated by allogeneic bone marrow transplantation. However, in the absence of suitable marrow donors, the development of alternative forms of therapy is desirable. Because lymphocytes are long-lived, it is possible that introduction of the wild-type ZAP-70 gene into CD4+ ZAP-70deficient T cells will restore their immune function in vivo. Initial investigations evaluating the feasibility of gene therapy for ZAP-70 deficiency were performed using HTLV-I-transformed lymphocytes. Although transformation was useful in

circumventing problems associated with the maintenance of ZAP-70-deficient T cells and low gene transfer levels, the presence of HTLV-I precluded any biological studies. Here, we investigated a retrovirus-mediated approach for the correction of primary T cells derived from two ZAP-70-deficient patients. Upon introduction of the wild-type ZAP-70 gene, TCR-induced MAPK activation, IL-2 secretion and proliferation were restored to approximately normal levels. Importantly, this gain-of-function was associated with a selective growth advantage of gene-corrected cells, thereby indicating the feasibility of a gene therapy-based strategy. Gene Therapy (2000) 7, 1392–1400.

Keywords: ZAP-70 deficiency; retroviral transduction; TCR activation; gene therapy

Introduction Severe combined immunodeficiency syndrome (SCID) comprises a heterogeneous group of genetic disorders characterized by defective T and B lymphocyte function. One autosomal recessive variant of SCID is caused by an absence of ZAP-70, a protein tyrosine kinase (PTK) of the Syk family which is recruited to the T cell receptor (TCR) following its stimulation.1 ZAP-70 has a molecular weight of 70 kDa and is expressed at approximately equivalent levels in thymocytes, mature T cells and NK cells.2 ZAP-70-deficient patients are characterized by a selective inability to produce CD8+ T cells. Although mature CD4+ T cells are present at normal numbers in the periphery, they are unable to respond to TCR-mediated stimulation.3–5 Thus, ZAP-70 appears to play a critical role in human T cell ontogeny, as well as in T cell activation. The importance of ZAP-70 in TCR signaling events is underscored by its structure which includes two tandemly arranged Src homology 2 (SH2) domains and a carboxy-terminal kinase domain.1 ZAP-70 can associate with the phosphorylated CD3 and ␨ chains of the TCR

Correspondence: N Noraz, Institut de Geńe´tique Molećulaire de Montpellier, 1919 Route de Mende, 34293 Montpellier, Cedex 5, France Received 28 March 2000; accepted 3 May 2000

complex through its SH2 domain.1,6,7 Following this interaction, ZAP-70 is itself phosphorylated by Src family kinases. The activated Src and Syk/ZAP-70 kinases are then available to phosphorylate cytosolic substrates such as Vav, Lat, PLC␥, Cbl and SLP-76 (reviewed by Chu et al).8 These subsequent steps in the TCR signaling pathway are implicated in the differentiation, proliferation and apoptosis of thymocytes and mature T cells. Like patients with other forms of SCID, ZAP-70deficient patients present with opportunistic infections and failure to thrive. The disease is almost universally fatal in infancy unless treated by allogeneic bone marrow transplantation (BMT). Importantly, histocompatible BMT can generally be successfully performed in these patients without prior cytoablative chemo- or radiotherapy because of the absence of functional immunity. However, for the majority of patients, histocompatible donors are not available and they are therefore transplanted with T cell-depleted BM from haploidentical donors or unrelated donor BM. Unfortunately, transplantation with non-histocompatible BM is associated with a high rate of serious immunologic complications such as graft-versus-host disease, delayed immune reconstitution and abnormal B cell differentiation.9–11 Thus, development of alternative strategies such as gene-based therapies, including the reconstitution of CD4+ T cells or hematopoietic progenitor cells, could be beneficial for patients who do not have histocompatible BM donors.

Gene transfer in primary ZAP-70-deficient T cells M Steinberg et al

Before any clinical gene therapy trials can be contemplated for ZAP-70-deficient patients, it is of utmost importance to ensure that the function of peripheral T cells from these patients can be restored by introduction of the WT ZAP-70 gene. Previously, we performed retrovirus-mediated gene transfer investigations for ZAP-70 deficiency using HTLV-I-transformed ZAP-70-deficient T cells.12 HTLV-I transformation was necessary because, unlike normal or ADA-deficient T cells which can easily be expanded in vitro upon TCR engagement, ZAP-70deficient T cells do not respond to TCR stimulation.3–5 Although we were able to introduce the wild type ZAP70 gene into these HTLV-I-transformed lymphocytes, the presence of HTLV-I confounded any possible biological studies.12 Specifically, because HTLV-I transformation is associated with cell immortalization, as well as the altered expression and function of several protein kinases and phosphatases which play crucial roles in TCRinduced signaling,13–16 these cells likely did not reflect primary ZAP-70-deficient T cells. Thus, while HTLV-I immortalization clearly facilitated our ability to manipulate ZAP-70-deficient T cells in vitro, this approach had severe limitations. In the present study, we have used primary T cells isolated from two ZAP-70-deficient siblings.17 Following establishment of optimal conditions for the ex vivo activation of these T cells, we investigated whether gene transfer could be achieved using a murine leukemia virus-based retroviral vector. Previously, we determined the parameters necessary for reliable high gene transfer efficiencies into normal T cells.18 Here, we adapted this protocol for the transduction of ZAP-70-deficient primary T cells. Upon gene correction of ZAP-70-deficient T cells, we observed a restoration of the TCR signaling cascade; the intracellular Ras-mitogen-activated protein kinase (MAPK) pathway was activated and IL-2 secretion as well as proliferation were normalized. Importantly, this gain-of-function was associated with a selective growth advantage of transduced cells. Thus, these data reinforce the potential application of gene transfer as an alternative therapy for ZAP-70 deficiency.

substitution at aa 507 of the kinase domain and the ZAP70 protein is not expressed17 (Figure 1). Interestingly, this mutation lies within a 13 bp region which is deleted in another ZAP-70-deficient patient.5 Since previous ZAP-70 gene transfer studies were severely limited by the presence of HTLV-I- in the cells,12 we were interested in assessing the effects of introduced ZAP-70 in primary T cells, a physiological context. We have recently determined that CD4+ ZAP-70-deficient T cells can be expanded in vitro with irradiated allogeneic feeder cells, PHA and exogenous IL-2 in the presence of human serum.17 Importantly, we find that a polyclonal population of ZAP-70−/− ␣␤ CD4+ T cells harboring a diverse TCR V␤ repertoire proliferated (data not shown), allowing these cells to be used for the ZAP-70 retrovirusmediated gene transfer experiments described below.

1363

Primary ZAP-70 deficient T cells transduced with a ZAP-70-expressing retroviral vector have a growth advantage over their non-transduced counterparts The human wild-type ZAP-70 cDNA bearing a C-terminal VSV-G tag was cloned into the murine leukemia virus-based LZRS retroviral vector19,20 containing the EGFP coding region downstream of an internal ribosomal entry site (IRES) (Figure 2a). The resulting LZRSZAP-70/EGFP vector was packaged into GALV envelope-expressing retroviral virions, using the PG13 packaging cell line, since this envelope protein results in a high transduction of lymphoid cells.21,22 Since it has been

Results Characterization and in vitro expansion of ZAP-70deficient T cells Two infant siblings (Pt 1 and Pt 2), children of a consanguineous relationship, presented with histories of multiple infections and chronic diarrhea. Examination of peripheral blood T cells revealed a phenotype consistent with a deficiency of the ZAP-70 kinase;3–5 normal numbers of CD4+ lymphocytes (59% and 63% in the two patients, respectively) and a marked paucity of CD8+ cells (3% and 6%, respectively). The phenotype of peripheral CD8+ cells from Pt 2 was further studied and it was demonstrated that the majority of these cells were CD16+ natural killer (NK) cells (72%) whereas in normal individuals, only 6–13% of CD8+ cells are of NK origin. Similar to observations in other ZAP-70-deficient patients, the patients described here demonstrated a profound defect in T cell proliferation upon TCR and mitogen stimulation (data not shown). We previously showed that both patients harbor a homozygous C to T nucleotide transition at position 1729, resulting in an alanine to valine

Figure 1 ZAP-70 deficiency resulting from a mutation at alanine 507 of the kinase domain. CD4+ T cell lysates (1 × 106 cell equivalents) from two SCID patients (Pt 1 and Pt 2) as well as a control donor (CTRL) were fractionated on an SDS-polyacrylamide gel. Membranes were immunoblotted with an ␣-ZAP-70 mAb and reprobed with an ␣-Erk2 mAb to verify protein levels. Immunolabeled proteins were visualized by ECL. The nucleotide sequence corresponding to residues 1713–1736 (aa 502–509) of the kinase domain are shown. As indicated, the two patients harbor a homozygous cytidine to thymidine transition at nucleotide position 1729 resulting in the mutation of an alanine to a valine at aa position 507. Gene Therapy


1364

Figure 2 Generation of a high titer PG13-ZAP-70/EGFP packaging cell line. (a) Schematic representation of the LZRS-ZAP-70/EGFP vector harboring the wild-type ZAP-70 cDNA. The introduced ZAP-70 gene bears a C-terminal vesicular stomatitis virus-protein G epitope tag48 allowing endogenous and ectopic ZAP-70 proteins to be distinguished on the basis of their molecular weights. The positions of the 5⬘ long terminal repeat (5⬘ LTR), viral packaging signal (␺+), multiple cloning site (MCS), internal ribosome entry site (IRES), enhanced GFP coding sequence (EGFP), and the 3⬘ long terminal repeat (3⬘ LTR) are indicated. (b) Flow cytometric analysis of EGFP expression in the PG13 packaging cell line transduced with LZRS-ZAP-70/EGFP retroviral supernatant. PG13 cells expressing EGFP were sorted on a FACS Vantage cytometer. The mean fluorescence intensity (MFI) of non-transduced PG13 cells, the bulksorted PG13-ZAP-70/EGFP pool as well as the PG13-ZAP-70/EGFP clone with the highest capacity to transduce T lymphocytes (clone 10) are shown.

shown that retroviral titer values measured on widely used indicator cell lines (such as NIH 3T3 or HeLa cells) do not reflect the ability of retroviral supernatants to transduce human T cells,23 we tested the capacity of cellfree retroviral supernatants from 20 individual PG13/LZRS-ZAP-70/EGFP clones to transduce the Jurkat leukemia T cell line. The mean transduction of Jurkat T cells after a single exposure to virus from the different clones ranged widely from 5% to 45%. As we previously observed with PG13 clones harboring the empty LZRSEGFP vector,18 EGFP expression in the PG13/LZRS-ZAP70/EGFP clones correlated with their ability to produce virions capable of transducing Jurkat cells. Indeed, virus produced by PG13/LZRS-ZAP-70/EGFP-clone 10, which had the highest level of EGFP expression (MFI of 2109), was most efficient in transducing Jurkat T cells and inversely, clone 6, which expressed the lowest level of EGFP (MFI of 115) produced the lowest level of infectious virus (Figure 2b and data not shown). The original bulksorted PG13/LZRS-ZAP-70/EGFP pool had an MFI of 704 which was intermediate between these two extremes (Figure 2b and data not shown). The expression of EGFP in PG13/LZRS-ZAP-70/EGFP-clone 10 remained stable following continuous culture for over 6 months and was used for all further experiments. It should be noted that retroviral supernatants from PG13/LZRS-ZAP-70/EGFPGene Therapy

clone 10 nevertheless had a lower capacity to infect Jurkat cells than the control packaging clone, PG13/LZRSEGFP-clone 7, with mean transduction efficiencies of 35 ± 8% and 55 ± 8%, respectively. Primary CD4+ T lymphocytes derived from ZAP-70deficient patients were transduced as described, using a fibronectin-based protocol.18 Cells were activated for 2 days with irradiated allogeneic feeder cells, PHA and exogenous IL-2 and exposed to either PG13/LZRS-ZAP70/EGFP clone 10 retroviral supernatant or control PG13/LZRS-EGFP clone 7 supernatant for a 6-h period on 2 consecutive days. Cells were then expanded and assessed for EGFP expression at days 3, 40, 60 and 120 after transduction. As shown in Figure 3a, gene transfer levels were significantly higher following transduction with control LZRS-EGFP retrovirus (46% and 57% in Pt 1 and Pt 2, respectively) in comparison with the LZRSZAP-70/EGFP retrovirus (31 and 36% for Pt 1 and Pt 2, respectively). These values correlate with those obtained following infection of Jurkat T cells. Retroviral gene transfer efficiencies did vary somewhat in individual experiments, but cells from the two patients were equally susceptible to infection with the same retroviral supernatant. Importantly, we always observed a marked augmentation in the percentage of ZAP-70/EGFP-transduced patient T cells upon in vitro expansion (Figures 3a and 3b). In the gene transfer experiment shown here, the percentages of Pt 1 and Pt 2 T cells transduced with ZAP70/EGFP increased from 31 to 67% and 36 to 74%, respectively, following 2 months of culture. Moreover, following 120 days in culture, the percentage of Pt 2 T cells expressing ZAP-70/EGFP rose to greater than 90%. In contrast, following transduction with the control LZRS-EGFP vector, the percentage of patient T cells expressing EGFP remained essentially unchanged. Collectively, these results indicate that ZAP-70-transduced patient T cells have a selective growth advantage over their non-transduced counterparts.

ZAP-70 is highly expressed and appropriately phosphorylated in primary patient T cells transduced with the ZAP-70/EGFP retroviral vector All experiments described herein were performed on T cells which had been expanded in vitro for 2–3 months after transduction. At this time point, approximately 50% and 70% of patient T cells transduced with the empty vector and the ZAP-70/EGFP vector were EGFP+, respectively. The introduced tagged ZAP-70 protein (ZAP-70T) was expressed at levels ranging from 1.5 to three-fold higher than that of endogenous ZAP-70 (ZAP70E) (Figure 4a and data not shown). To determine whether the introduced ZAP-70T would be appropriately tyrosine phosphorylated in the transduced patient T cells, ZAP-70 was immunoprecipitated from either unstimulated or CD3-stimulated control T cells, patient T cells transduced with the empty EGFP vector and patient T cells transduced with the ZAP-70/EGFP vector (Figure 4b). In the patient T cells reconstituted with ZAP-70, ZAP-70 was tyrosine phosphorylated to a level equivalent to that detected in control T cells following TCR aggregation. As expected, no tyrosine phosphorylation of ZAP-70 was observed in the Pt/EGFP cells which do not express this kinase. These results indicate that high levels of ZAP-70 expression can be obtained after retroviral transduction of ZAP-70-deficient primary T cells and


1365

Figure 4 ZAP-70 is highly expressed and appropriately phosphorylated in primary patient T cells transduced with the ZAP-70/EGFP retroviral vector. (a) The level of introduced ZAP-70 (ZAP-70T) in patient T cells transduced with the ZAP-70/EGFP retroviral vector, relative to endogenous ZAP-70 expression (ZAP-70E) in control T cells, was monitored by immunoblot analysis. Patient T cells transduced with a retroviral vector expressing EGFP alone were used as a control. Lysates from 5 × 105 cells were separated by SDS-PAGE and the membrane was immunoblotted with an ␣-ZAP-70 mAb. (b) Control T cells, patient T cells transduced with the EGFP vector and patient T cells transduced with the ZAP70/EGFP vector were either unstimulated (−) or stimulated (+) with a cross-linked ␣-CD3 mAb. ZAP-70 was immunoprecipitated from 2 × 106 cell equivalents with a polyclonal rabbit antibody and immunoprecipitates were fractionated on a polyacrylamide gel. The phosphorylation status of ZAP-70 was monitored by immunoblotting with the 4G10 antiphosphotyrosine mAb (␣-PTyr) (top panel). The positions of endogenous ZAP-70 (ZAP-70E), tagged ZAP-70 (ZAP-70T), and the immunoglobulin heavy chain (IgH) are indicated. The blot was then stripped and reprobed with a ZAP-70-specific mAb to assess the level of immunoprecipitated ZAP70 (bottom panel).

Figure 3 Patient T cells transduced with the LZRS-ZAP-70/EGFP retroviral vector have a selective growth advantage. ZAP-70-deficient patient T cells (Pt 1 and Pt 2) were transduced on fibronectin-coated plates with cell-free GALV-pseudotyped retrovirus harboring either the ZAP70/EGFP or empty EGFP vector. Cells were expanded and EGFP expression was monitored on a FACScan at the indicated time points. (a) The percentage of patient T cells expressing EGFP as a function of time after transduction is shown. (b) Representative FACS profiles of patient T cells at day 3, 60 and 120 after transduction with ZAP-70/EGFP retroviral supernatant. The percentage of EGFP+ cells is indicated in each histogram.

moreover, that introduced ZAP-70 is activated following CD3 crosslinking.

TCR-induced MAPK activation is restored in primary patient T cells transduced with the ZAP-70/EGFP retroviral vector A growing body of evidence indicates that the Ras-Erk pathway is critical for the proliferation of distinct cell types.24 Activation of the MAPKs Erk1 and Erk2 is severely diminished in in vitro expanded ZAP-70-deficient T cells,17 and it was therefore important to first establish whether this signaling pathway was reconstituted following introduction of the wild-type ZAP-70 gene. Control T cells, patient T cells and transduced patient T cells were Gene Therapy


1366

activated for 3 min with crosslinked ␣CD3 mAb and the level of phosphorylated Erk proteins was assessed on immunoblots using a polyclonal Ab which specifically recognizes the dually phosphorylated form of Erk1 and Erk2. As expected, phosphorylation of these proteins was severely diminished in ZAP-70-deficient T cells (Figure 5). In contrast, Erk was phosphorylated to significantly augmented levels in gene-corrected patient T cells. Additionally, a low level of basal Erk phosphorylation was observed in control T cells and ZAP-70-transduced patient T cells, but not in non-transduced patient T cells. CD3-induced Erk/MAPK phosphorylation was slightly lower in gene-corrected patient T cells than in control T cells, likely due to the presence of a small percentage of cells in the former population which did not express ZAP-70 (EGFP−) (30%). Thus, our data strongly suggest that the Ras-Erk cascade is activated at physiological levels in patient T cells expressing the wild-type ZAP-70 gene.

Restoration of ZAP-70 function in primary patient T cells transduced with the ZAP-70/EGFP vector results in an appropriate TCR-induced IL-2 secretion and proliferation As the gene transfer experiments described here were performed on primary ZAP-70-deficient T cells, we were able to monitor the biological consequences of transgene expression. Specifically, TCR-mediated IL-2 secretion and proliferation which are crucial for the development of a normal T cell response and cannot be accurately assessed in the background of a transformed cell. In the primary T cells used here, we measured the level of IL-2 present in the culture media 24 h following stimulation with an immobilized anti-CD3 mAb. Neither control nor patient T cells secreted detectable levels of IL-2 before TCR aggregation (⬍5 pg/ml). Patient T cells had a severe

Figure 5 MAPK activity in ZAP-70-deficient T cells following introduction of the wild-type ZAP-70 gene. Lysates from unstimulated (−) and CD3-crosslinked (+) control T cells, patient T cells transduced with the EGFP vector and patient T cells transduced with the ZAP-70/EGFP vector were immunoblotted with a polyclonal Ab which recognizes the dually phosphorylated forms of Erk1 and Erk2. The blot was stripped and reprobed with an ␣Erk2 mAb. Gene Therapy

deficit in CD3-mediated IL-2 secretion compared with identically activated control T cells (P = 0.01) (Figure 6a). Importantly, the levels of IL-2 secretion in CD3-stimulated patient T cells transduced with the wild-type ZAP70 gene were significantly augmented, to levels approximately equivalent to those observed in control T cells (P ⬎ 0.05). Thus, the marked defect in CD3-induced IL-2 secretion from ZAP-70-deficient T cells can be overcome by the retrovirus-mediated introduction of the wild-type ZAP-70 gene. Finally, we compared the ability of ZAP-70-deficient T cells and their gene-corrected counterparts to proliferate following CD3 engagement. It should be noted that the level of proliferation in normal primary lymphocytes vary extensively from one experiment to another. In order to alleviate any possible bias from a single experiment, the results presented here represent the mean ± standard deviation of samples from four separate experiments, each performed in triplicate. Control T cells, patient T cells transduced with the empty vector, as well as patient T cells transduced with the ZAP-70/EGFP retroviral vector were activated with immobilized antiCD3 mAb and proliferation was measured 3 days later.

Figure 6 CD3-induced IL-2 secretion and proliferation are normalized in ZAP-70-deficient T cells following transduction with the ZAP-70/EGFP retroviral vector. (a) Control T cells, patient T cells transduced with EGFP alone, as well as patient T cells transduced with the ZAP-70/EGFP retroviral vector were either left unstimulated (media) or stimulated with an immobilized anti-CD3 mAb (␣-CD3). IL-2 was measured 24 h later using a commercially available ELISA kit. Values are the mean ± s.d. of triplicate samples from at least three independent experiments. (b) Proliferation was measured following 3 days of culture. Cells were pulsed with 1 ␮Ci of 3 H-thymidine and harvested 18 h later. Values are the mean ± s.d. of four independent experiments, each performed in triplicate. Statistical analyses were performed using a Student’s t test.


As compared with control T cells, CD3-induced proliferation in patient T cells was severely diminished (P = 0.03). However, following transfer of the wild-type ZAP-70 gene, a significantly augmented proliferation was observed (P = 0.01) (Figure 6b). Importantly, gene-corrected and control T cells proliferated to similar levels upon CD3 engagement (P ⬎ 0.05). Taken together, these results demonstrate that the introduction of ZAP-70 allowed patient T cells to respond to TCR stimulation with appropriate levels of IL-2 secretion and proliferation.

Discussion Murine leukemia virus-based retroviral vectors have been investigated extensively during the past decade as vehicles for gene transfer. Although clinical gene therapy trials targeting T lymphocytes and hematopoietic stem cells have demonstrated the feasibility of gene transfer in humans, this approach has been hampered by low levels of retroviral transduction (⬍10%).25–28 Because of concerns regarding the efficiency, safety and reproducibility of previously described retroviral transduction methods, the development of simple protocols which provide reliable high levels of gene transfer is desirable. We as well as others have assessed various parameters important for cell-free retrovirus-mediated gene transfer in T cells including the nature of the pseudotyping envelope, the activating stimulus, and the kinetics of transduction.18,21,22,29 Moreover, the use of fibronectin, which promotes the colocalization of target cells and retrovirus, results in significantly augmented transduction efficiencies of T cells.18,30,31 These advances in gene transfer technology were paramount in allowing us to transduce primary T cells derived from ZAP-70-deficient patients successfully. The ability to generate high-titer packaging lines easily is important for gene transfer applications. Previously, we reported that in the PG13 packaging line, EGFP expression from the LZRS retroviral vector correlated with the level of infectious virus.18 Indeed, the level of T cell transduction following exposure to retroviral supernatants produced by individual PG13/LZRS-EGFP clones varied widely and the highest transduction efficiency was obtained using retrovirus issued from a clone with the highest mean fluorescence intensity (a measure of EGFP expression). This correlation was also observed in the present study, following introduction of the ZAP-70 cDNA upstream of the IRES sequence in the LZRS-EGFP vector. We observed an approximately twoto three-fold increase in T cell transduction using retroviral supernatant produced by PG13/ZAP-70-EGFP clone 10, which had an MFI which was three-fold higher than the initial PG13/ZAP-70-EGFP pool. Collectively, these data demonstrate that high titer PG13 clones, expressing a transgene of interest, may be easily generated by limiting screening to packaging cells with high EGFP expression. It has recently been shown that in the context of a retroviral vector, EGFP expression was significantly reduced in resting T cells when driven from an internal CMV promoter (in the LNCX MuLV-based vector).32 Nevertheless, when transcription was driven from the 5⬘ LTR and translation was dependent on an IRES, EGFP expression did not appear to vary with T cell activation (this report).

The reasons accounting for this discrepancy remain to be determined. Although EGFP expression did not vary within a transduced T cell population, it should be noted that we observed a lower level of EGFP fluorescence in cells transduced with the vector in which ZAP-70 was inserted upstream of the IRES. Thus, it appears that the introduction of a coding sequence between the 5⬘ LTR and the IRES is associated with a lower level of EGFP expression in this vector. As stable expression of the gene of interest is a prerequisite for many gene therapy-based protocols, it appears to be preferable to drive expression from the 5⬘ LTR rather than from an internal CMV promoter. Because of previous difficulties in transducing primary lymphocytes, especially those derived from immunodeficient individuals, most preclinical studies assessing the feasibility of a gene therapy approach for these patients have been performed using transformed cells. We and others assessed gene transfer in ␥c- and JAK3deficient B cells using EBV-transformed lines,33–36 while Jurkat leukemia cells and HTLV-I-transformed T cells have been employed as models for CD3␥- and ZAP-70 deficiencies, respectively.12,37 Here, the effects of introduced ZAP-70 on biological function were analyzed in a primary T cell model. We determined that primary ZAP70-deficient T cells can be expanded in Yssel’s media, which is based on Iscove’s modification of Dulbecco’s modified minimal essential medium and contains transferrin, insulin, ethanolamine, as well as a mixture of saturated and unsaturated fatty acids.38 Since there has been significant concern regarding the safety of reinfusing human cells which have been cultured ex vivo in media containing fetal calf serum (FCS), it is notable that Yssel media is supplemented with human sera and not FCS. Importantly, the ZAP-70-deficient T cells expanded in this manner are polyclonal with a representation of diverse TCR V␤ families (C Serran and NN, unpublished observations), a crucial prerequisite for ZAP-70 gene therapy. The levels of exogenous ZAP-70 in transduced patient T cells were approximately two- to three-fold higher than endogenous ZAP-70 levels in control T cells. In ZAP-70transduced T cells, CD3-induced MAPK activation was significantly augmented, but still decreased as compared with control T cells. Nevertheless, it should be noted that at the time point at which these experiments were performed, approximately 30% of the patient T cells were EGFP− (and by inference, ZAP-70−). These data strongly suggest that ZAP-70-transduced and control T cell populations responded similarly to TCR engagement. This is in striking contrast to non-transduced patient T cells where MAPK activation was minimal. We have previously found that in vitro expanded ZAP-70-deficient T cells do proliferate at low levels upon CD3 activation, likely due to increased expression of the ZAP-70-related Syk kinase.17 However, these cells are clearly defective.17 Here, we show that CD3-induced IL-2 secretion and proliferation are attained only after introduction of the wildtype ZAP-70 gene. It is probable that reconstitution of these TCR-induced responses resulted in the observed in vitro growth advantage of ZAP-70-transduced patient T cells. The selective advantage for ZAP-70-transduced cells was manifested by a significant increase in the number of cells harboring the co-expressed EGFP marker following in vitro expansion (30% to ⬎90%). A similar selec-

1367

Gene Therapy


1368

tive advantage was observed in vitro following transduction of ␥c-deficient EBV-transformed B cell lines.34 Importantly, Cavazzana-Calvo and colleagues have now extended this observation in vivo: following gene transfer into a small fraction of ␥c-deficient CD34+ progenitor cells, all peripheral T cells in the patients express the ␥c transgene.39 There are several advantages and disadvantages of a CD4+ T cell-based gene transfer strategy for the treatment of ZAP-70 deficiency as compared with a lymphoid progenitor-based protocol. The latter approach may be optimal under certain conditions as it will theoretically allow the development of functional CD4+ as well as CD8+ T cells in the periphery. Nevertheless, introduction of ZAP70 into hematopoietic progenitor cells raises several important issues. One potential obstacle concerns the tissue specific expression of ZAP-70. ZAP-70 is normally expressed only in T lineage cells2 and its introduction into stem cells would result in the expression of ZAP-70 in other hematopoietic lineages. This problem could be obviated by the use of vectors containing T cell specific promoters but to date, T cell specific expression has not been achieved using retroviral vectors. Alternatively, it is possible that expression of ZAP-70 in other lineages may not raise any serious predicaments for at least two reasons. (1) In the case of stem cell gene therapy for ADA deficiency, the percentage of transduced T cells present in the periphery was 2–3 logs higher than the percentage of transduced granulocytes.40 This huge discrepancy was explained by the observation that ADA is required for T cell survival, but not for granulocyte survival. A similar phenomenon would probably be observed in ZAP-70 deficiency since we found that reconstituted CD4+ lymphocytes had a selective advantage over their non-reconstituted counterparts. This hypothesis is reinforced by our observation that following a non-conditioned bone marrow transplantation for ZAP-70 deficiency, donorderived cells are exclusively found in the T cell compartment.41 (2) We previously showed that high levels of ectopic ZAP-70 did not appear to have any adverse affects in myeloid cells, likely because it was not activated in these cells following receptor engagement.42 Nevertheless, an exhaustive analysis is still necessary before concluding that ectopic ZAP-70 does not have adverse affects in non-T cell hematopoietic lineages. We are currently assessing the effects of ZAP-70 in hematopoietic progenitor cells using a humanized SCID mouse model. Importantly though, a CD4+ T cell gene transfer approach has distinct advantages for ZAP-70-deficient patients in several clinical situations. Specifically, our results suggest that introduction of ZAP-70 into mature CD4+ peripheral T cells should result in an almost immediate reconstitution of this arm of the immune response. This is crucial for patients who already have opportunistic infections at the time of presentation. Even if these patients could undergo bone marrow transplantation or stem cell-mediated gene therapy, they would likely die of their infection before mature T cells develop (a lag time of at least several weeks). Additionally, as ZAP-70-deficient T cells do not respond to CD3 stimulation alone, it is significant to note that we have recently determined that resting T cells can be transduced to significant levels (approximately 20%) following IL-7 cyto-

Gene Therapy

kine stimulation.43 Notably, the IL-7 signaling pathway is likely to be functional in ZAP-70-deficient T cells. Preliminary results of T cell-based gene therapy for the adenosine deaminase (ADA)-deficient form of SCID have demonstrated that T cells are relatively long-lived after reconstitution.27,44 Moreover, in the presence of normal CD4+ T cells, the absence of functional CD8+ T cells is compatible with life, both in humans and in mice.45,46 Indeed, the first reported patient with this phenotype, due to a mutation in the transporter associated with antigen processing (TAP) gene, was a teenager.45 This is in striking contrast with ZAP-70 deficiency where the disease is universally fatal in infancy. Thus, it is possible that infusion of ZAP-70-reconstituted T cells into patients with severe infections will allow the patient to mount a rapid and protective immune response. The patient could then undergo a definitive treatment such as bone marrow transplantation or a progenitor-based gene therapy at a later time point. Clearly, if the patient is free of infection, these latter therapies would have a much higher chance of success. While ZAP-70 gene therapy is not yet practical, this report demonstrates the potential of gene transfer as an alternative therapeutic strategy for ZAP-70-deficient patients and for patients with other immunological defects.

Materials and methods Cells Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll–Hypaque density gradient centrifugation from two patients, children of a consanguineous relationship, and normal donors following institutional review board approval and informed consent. CD4+ T cells were purified using ␣CD4 coated magnetic beads as per the manufacturer’s instructions (Dynal, Great Neck, NY, USA). Cells were cultured in Yssel’s medium38 supplemented with 1% human AB+ serum and recombinant human IL-2 at 100 U/ml (Chiron, Emeryville, CA, USA). Cells were stimulated weekly or every other week with PHA (0.5 ␮g/ml) (Murex, Dartford, UK) and irradiated feeder cells consisting of PBMC and the EBV-transformed B cell line JY as previously described.47 Retroviral construction and generation of packaging lines The human wild-type ZAP-70 cDNA bearing a C-terminal vesicular stomatitis virus-protein G epitope tag (T) was kindly provided by V Di Bartolo and O Acuto (Institut Pasteur, Paris, France).48 This cDNA was cloned into the EcoRI and NotI sites of the polylinker present in the murine leukemia virus-based LZRS retroviral vector.19,20 Stop codons were introduced in all three open reading frames just upstream of the polylinker in order to avoid translation of any potential gag fusion proteins. In this vector, the EGFP coding region is present downstream of an encephalomyocarditis-derived internal ribosomal entry site (IRES) (Figure 2a). Previously, we described the generation of a high titer PG13 packaging line which produces gibbon ape leukemia virus envelope (GALV)-pseudotyped retrovirus harboring the control LZRS-EGFP retroviral vector (clone 7).18 GALV-pseudotyped retrovirus harboring the LZRS-ZAP-70/EGFP vec-


tor was generated in the same manner. Briefly, PG13 cells49 transduced with LZRS-ZAP-70/EGFP were identified by their autofluorescence and analyzed/sorted on a FACScan or FACS Vantage flow cytometer (Becton Dickinson, San Jose, CA, USA). The sorted PG13 cells were cloned by limiting dilution in 96-well plates and the capacity of cell-free supernatant from each clone to infect Jurkat T cells was monitored. All vector-containing retroviral supernatants were harvested after a 24-h incubation of near confluent cells in a humidified incubator at 32°C. The collected culture medium was filtered through 0.45 ␮m filters and stored at −80°C for further use.

Lymphocyte transductions ZAP-70-deficient CD4+ T cells were activated as described above and 48–72 h later, cells were transduced on fibronectin-coated plates essentially as reported.18 The recombinant fibronectin fragment (CH-296) which contains the connecting segment, cell binding domain and heparin binding domain50 was kindly provided by Takara Shuzo, Otsu Shiga, Japan. The recombinant CH-296 fibronectin fragment was diluted in PBS and used at a concentration of 8 ␮g/cm2 to coat 24-well Falcon dishes for 2 h at room temperature. The unbound fibronectin was then removed and 1 × 106 T cells were added in 0.5 ml of Yssel’s medium and 0.5 ml of retroviral supernatant. After a 6-h exposure to retrovirus at 37°C, T cells were incubated in fresh Yssel’s medium with recombinant IL-2 overnight before the transduction procedure was repeated. EGFP expression was assessed in the transduced patient T cells at the indicated times. Cell stimulations and immunoblots All TCR stimulations were performed at 2–3 months after transduction on cells in ‘resting phase’ which had not been stimulated with irradiated feeder cells for at least 10 days before use. Furthermore, before activation, cells were cultured overnight in Yssel’s medium without rIL2. Cells were then washed in serum-free RPMI, resuspended at 2 × 107 cells/ml, and stimulated with the OKT3 ␣CD3 mAb (10 ␮g/ml) followed by cross-linking with an ␣-mouse F(ab⬘)2 fragment (20 ␮g/ml) (Immunotech, Marseille, France) for 3 min at 37°C. Cells were lysed in a 1% NP40 lysis buffer and postnuclear supernatants were immunoprecipitated for 1 h at 4°C with a rabbit polyclonal antibody specific for ZAP70 (kindly provided by A Weiss, UCSF, CA, USA) followed by collection on protein A sepharose beads (Amersham Pharmacia Biotech, Orsay, France).35 Immunoprecipitates were boiled, resolved on SDS-PAGE gels and transferred electrophoretically to nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany). Membranes were blocked for 30 min in TBS (150 mm NaCl, 20 mm Tris, pH 7.5) containing 5% BSA and 0.1% Tween 20 and incubated with either the 4G10 anti-phosphotyrosine antibody (Upstate Biotechnology, Lake Placid, NY, USA), a monoclonal anti-ZAP-70 antibody (generously provided by A Weiss), a pAb recognizing the dually phosphorylated T183Y185 form of Erk1/Erk2 (antiACTIVE-MAPK) (Promega, Charbonnie`re, France) or an ␣Erk2 mAb (Transduction Laboratories, Lexington, KY, USA) for 1 h at room temperature. Blots were then incubated with horseradish peroxidase-conjugated goat ␣mouse or ␣-rabbit secondary Abs (Amersham, Arlington Heights, IL, USA) and immunoreactive proteins were vis-

ualized using the enhanced chemiluminescence (ECL) detection system (Amersham). For reblotting, filters were stripped as reported.35

1369

Lymphocyte proliferation and cytokine measurements Cells (1 × 105 in a total volume of 100 ␮l) were cultured in triplicate in flat-bottomed 96-well plates in the presence or absence of immobilized OKT3 mAb (1 ␮g/ml). After 3 days in culture, 3H-thymidine (1 ␮Ci/well) (CEA, Saclay, France) was added and cells were cultured for an additional 18 h, harvested on to glass fiber filters and then counted in a scintillation counter (Beckman LS 6000SC; Beckman Coulter, Villepinte, France). For measurement of secreted IL-2, cells were activated as described above and cell-free supernatants were collected 24 h following activation. A commercially available ELISA kit for IL-2 (Immunotech, Marseille, France) was used according to the manufacturer’s instructions.

Acknowledgements We are grateful to V Dardalhon, C Rebouissou, A Weiss, and P Jourdan for their assistance and J Clot for important insight. NN was supported by a fellowship from the AFM and MS is supported by a fellowship from the Fundacion YPF. We thank Christophe Duperray for his expertise and assistance with FACS sorting. Dr Ikunoshin Kato and Setsuko Yoshimura of Takara Shuzo Co. are generously acknowledged for providing the recombinant fibronectin fragment and for their continuing assistance. Supported by grants from the March of Dimes grant #6-FY99–406, the AFM, ARC, INSERM and CNRS (to NT), and JZKF.Ulm.CO.5 (to KS).

References 1 Chan AC et al. ZAP-70: a 70 kd protein-tyrosine kinase that associates with the TCR zeta chain. Cell 1992; 71: 649–662. 2 Chan AC et al. Differential expression of ZAP-70 and Syk protein tyrosine kinases, and the role of this family of protein tyrosine kinases in TCR signaling. J Immunol 1994; 152: 4758–4766. 3 Arpaia E et al. Defective T cell receptor signaling and CD8+ thymic selection in humans lacking zap-70 kinase. Cell 1994; 76: 947–958. 4 Chan AC et al. ZAP-70 deficiency in an autosomal recessive form of severe combined immunodeficiency. Science 1994; 264: 1599–1601. 5 Elder ME et al. Human severe combined immunodeficiency due to a defect in ZAP-70, a T cell tyrosine kinase. Science 1994; 264: 1596–1599. 6 Iwashima M et al. Sequential interactions of the TCR with two distinct cytoplasmic tyrosine kinases. Science 1994; 263: 1136– 1139. 7 Wange RL, Kong AN, Samelson LE. A tyrosine-phosphorylated 70-kDa protein binds a photoaffinity analogue of ATP and associates with both the zeta chain and CD3 components of the activated T cell antigen receptor. J Biol Chem 1992; 267: 11685– 11688. 8 Chu DH, Morita CT, Weiss A. The Syk family of protein tyrosine kinases in T-cell activation and development. Immunol Rev 1998; 165: 167–180. 9 Friedrich W et al. Bone marrow transplantation in severe combined immunodeficiency: potential and current limitations. Immunodeficiency 1993; 4: 315–322. 10 Fischer A et al. European experience of bone marrow transplantation for severe combined immunodeficiency. Lancet 1990; 336: 850–854. Gene Therapy


1370

11 O’Reilly RJ et al. The use of HLA-non-identical T-cell-depleted marrow transplants for correction of severe combined immunodeficiency disease. Immunodefic Rev 1989; 1: 273–309. 12 Taylor N et al. Reconstitution of T cell receptor signaling in ZAP70-deficient cells by retroviral transduction of the ZAP-70 gene. J Exp Med 1996; 184: 2031–2036. 13 Migone TS et al. Recruitment of SH2-containing protein tyrosine phosphatase SHP-1 to the interleukin 2 receptor; loss of SHP-1 expression in human T-lymphotropic virus type I-transformed T cells. Proc Natl Acad Sci USA 1998; 95: 3845–3850. 14 Weil R et al. Altered expression of tyrosine kinases of the Src and Syk families in human T-cell leukemia virus type 1-infected T-cell lines. J Virol 1999; 73: 3709–3717. 15 Koga Y et al. Absence of transcription of lck (lymphocyte specific protein tyrosine kinase) message in IL-2-independent, HTLV-Itransformed T cell lines. J Immunol 1989; 142: 4493–4499. 16 Watanabe T. HTLV-1-associated diseases. Int J Hematol 1997; 66: 257–278. 17 Noraz N et al. Alternative TCR signaling in T cells derived from ZAP-70-deficient patients expressing high levels of Syk. J Biol Chem 2000; 275: 15832–15838. 18 Dardalhon V et al. Green fluorescent protein as a selectable marker of fibronectin- facilitated retroviral gene transfer in primary human T lymphocytes. Hum Gene Ther 1999; 10: 5–14. 19 Heemskerk MH et al. Inhibition of T cell and promotion of natural killer cell development by the dominant negative helix–loop– helix factor Id3. J Exp Med 1997; 186: 1597–1602. 20 Kinsella TM, Nolan GP. Episomal vectors rapidly and stably produce high-titer recombinant retrovirus. Hum Gene Ther 1996; 7: 1405–1413. 21 Lam JS et al. Improved gene transfer into human lymphocytes using retroviruses with the gibbon ape leukemia virus envelope. Hum Gene Ther 1996; 7: 1415–1422. 22 Bunnell BA et al. High-efficiency retroviral-mediated gene transfer into human and nonhuman primate peripheral blood lymphocytes. Proc Natl Acad Sci USA 1995; 92: 7739–7743. 23 Ramiro AR et al. Enhanced green fluorescent protein as an efficient reporter gene for retroviral transduction of human multipotent lymphoid precursors. Hum Gene Ther 1998; 9: 1103–1109. 24 Cantrell D. T cell antigen receptor signal transduction pathways. Annu Rev Immunol 1996; 14: 259–274. 25 Dunbar CE et al. Retrovirally marked CD34-enriched peripheral blood and bone marrow cells contribute to long-term engraftment after autologous transplantation. Blood 1995; 85: 3048–3057. 26 Brenner MK et al. Gene-marking to trace origin of relapse after autologous bone marrow transplantation. Lancet 1993; 341: 85– 86. 27 Blaese RM et al. T lymphocyte-directed gene therapy for ADA− SCID: initial trial results after 4 years. Science 1995; 270: 475–480. 28 Kohn DB et al. Engraftment of gene-modified umbilical cord blood cells in neonates with adenosine deaminase deficiency. Nature Med 1995; 1: 1017–1023. 29 Wu AG et al. Improvement of gene transduction efficiency in T lymphocytes using retroviral vectors. Hum Gene Ther 1999; 10: 977–982. 30 Fehse B et al. Highly-efficient gene transfer with retroviral vectors into human T lymphocytes on fibronectin. Br J Haematol 1998; 102: 566–574. 31 Pollok KE et al. High-efficiency gene transfer into normal and adenosine deaminase-deficient T lymphocytes is mediated by transduction on recombinant fibronectin fragments. J Virol 1998; 72: 4882–4892.

Gene Therapy

32 Pollok KE et al. Costimulation of transduced T lymphocytes via T cell receptor–CD3 complex and CD28 leads to increased transcription of integrated retrovirus. Hum Gene Ther 1999; 10: 2221–2236. 33 Candotti F et al. Retroviral-mediated gene correction for Xlinked severe combined immunodeficiency. Blood 1996; 87: 3097–3102. 34 Hacein-Bey H et al. Gamma-c gene transfer into SCID X1 patients’ B-cell lines restores normal high-affinity interleukin-2 receptor expression and function. Blood 1996; 87: 3108–3116. 35 Taylor N et al. Correction of interleukin-2 receptor function in X-SCID lymphoblastoid cells by retrovirally mediated transfer of the gamma-c gene. Blood 1996; 87: 3103–3107. 36 Candotti F et al. In vitro correction of JAK3-deficient severe combined immunodeficiency by retroviral-mediated gene transduction. J Exp Med 1996; 183: 2687–2692. 37 Sun JY et al. Construction of retroviral vectors carrying human CD3 gamma cDNA and reconstitution of CD3 gamma expression and T cell receptor surface expression and function in a CD3 gamma-deficient mutant T cell line. Hum Gene Ther 1997; 8: 1041–1048. 38 Yssel H et al. Serum-free medium for generation and propagation of functional human cytotoxic and helper T cell clones. J Immunol Meth 1984; 72: 219–227. 39 Cavazzana-Calvo M et al. Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease. Science 2000; 288: 669–672. 40 Kohn DB et al. T lymphocytes with a normal ADA gene accumulate after transplantation of transduced autologous umbilical cord blood CD34+ cells in ADA-deficient SCID neonates. Nature Med 1998; 4: 775–780. 41 Knobloch C et al. Coexistence of donor and host T lymphocytes following HLA-different bone marrow transplantation into a patient with cellular immunodeficiency and nonfunctional CD4+ T cells. Transplantation 1991; 52: 491–496. 42 Taylor N et al. Differential activation of the tyrosine kinases ZAP-70 and Syk after Fc gamma RI stimulation. Blood 1997; 89: 388–396. 43 Dardalhon V et al. Highly efficient gene transfer in naive human T cells with a murine leukemia virus-based vector. Blood 2000 (in press). 44 Bordignon C et al. Gene therapy in peripheral blood lymphocytes and bone marrow for ADA-immunodeficient patients. Science 1995; 270: 470–475. 45 de la Salle H et al. Homozygous human TAP peptide transporter mutation in HLA class I deficiency (published erratum appears in Science 1994; 266: 1464). Science 1994; 265: 237–241. 46 Raulet DH. MHC class I-deficient mice. Adv Immunol 1994; 55: 381–421. 47 Spits H et al. Establishment of human T lymphocyte clones highly cytotoxic for an EBV-transformed B cell line in serumfree medium: isolation of clones that differ in phenotype and specificity. J Immunol 1982; 128: 95–99. 48 Mege D et al. Mutation of tyrosines 492/493 in the kinase domain of ZAP-70 affects multiple T-cell receptor signaling pathways. J Biol Chem 1996; 271: 32644–32652. 49 Miller AD et al. Construction and properties of retrovirus packaging cells based on gibbon ape leukemia virus. J Virol 1991; 65: 2220–2224. 50 Kimizuka F et al. Production and characterization of functional domains of human fibronectin expressed in Escherichia coli. J Biochem (Tokyo) 1991; 110: 284–291.