Introduction

Treatment of graft-versus-host disease in the setting of allogeneic hematopoietic stem cell may be achieved by the ex vivo transfer of the herpes-simplex thymidine kinase (HSV-tk) gene in donor T cells, allowing these gene-modified cells (GMC) to be sensitive to ganciclovir.^{1, 2, 3}

In our first clinical trial, GMC production required a 12-day ex vivo culture period including anti-CD3 T-cell polyclonal activation, retroviral-mediated transfer of the HSV-tk and neomycin phosphotransferase II (Neo-R) genes and a selection of GMC based upon their resistance to G418, a neomycin analog.⁴ This protocol resulted in the transfusion of a population of T cells with a reduced anti-EBV response⁵ but harboring a normal, or even enhanced, anti-CMV response.⁶ This difference appears to be related to a differential susceptibility of these two populations to activation-induced cell death as well as a differential proliferation rate following CD3 stimulation. We further demonstrated that anti-EBV reactivity can be improved by activation of T cells via CD3/CD28 costimulation rather than by CD3 ligation alone, and by selecting GMC by virtue of a cell-surface marker encoded by the retroviral vector, such as a truncated low-affinity nerve growth factor receptor (LNGFR), rather than by G418 selection.⁵

Moreover, we previously demonstrated that the stimulation of T cells with alloantigen prior to transduction results in a preferential transduction of alloreactive T cells.⁷ As division is required for oncoretrovirus-mediated gene transfer, this bias in transduction is likely due to the fact that there is increased proliferation of alloreactive cells.⁸ Such an observation was subsequently extended by Koehne et al⁹ to other antigens such as EBV. In the present study, we demonstrate that antigen-specific stimulation might not always be the unique possibility to achieve preferential transduction. Indeed, after a polyclonal activation stimulus, anti-EBV (EBV-T) and anti-CMV (CMV-T) CD8+ T cells are more susceptible to retroviral-mediated transduction than bulk CD8+ cells.

Results and discussion

Using LNGFR-coding vectors, we previously demonstrated that the frequency of EBV-T cells, as detected by reactivity against BMLF1 viral protein, was higher in the LNGFR+ CD8+ fraction than in the LNGFR-negative CD8+ fraction or even in cultured nontransduced, nonselected control cells.⁵ These data strongly suggested that, although the cells were activated by a polyclonal stimulus prior to transduction, BMLF1-specific cells were preferentially transduced as compared to the bulk CD8+ T-cell population. Similar observations were made for CMV-T cells, as monitored using pp65 tetramers.⁶ To exclude an artefact due to immunomagnetic-based cell sorting of the transduced cells, we evaluated transduction efficiency of tetramer-positive and tetramer-negative cells prior to sorting by simultaneous staining for tetramer, LNGFR and CD8 expression. To remain relevant with regard to our previous clinical trial,³ gene transfer was performed after soluble CD3 monoclonal antibody (mAb)+interleukin-2 (IL-2)-mediated T-cell activation (CD3/IL-2). Transduction efficiency was always higher in the EBV-T and CMV-T cells than in bulk CD8+ T cells (Figure 1). This increased transduction efficiency was observed irrespective of the vector used (amphotropic or GALV-pseudotyped vector) or the transduction method (supernatant incubation or spin transduction, data not shown). When T cells were alternatively activated by CD3/CD28 beads (Xcyte Therapies Inc., Seattle, WA, USA)+IL-2 (CD3/CD28/IL-2), transduction efficiencies of bulk CD8+ cells remained similar⁵ while higher transduction efficiencies were also observed within EBV-T and CMV-T cells (data not shown).

Figure 1.

Transduction efficiencies of antiviral CD8+ T cells versus bulk CD8+ T cells. Peripheral blood mononuclear cells from HLA-A*0201-positive, EBV and/or CMV seropositive healthy volunteer adult donors were activated by 10 ng/ml soluble CD3 mAb (OKT3, Jansen-Cilag, Levallois Perret, France) and 500 UI/ml recombinant human IL-2 (Cetus, Rueil Malmaison, France)^{3, 4} and transduced at day 3 by spin transduction or incubation with retroviral vector-containing supernatants harboring either the LNGFR-coding amphotropic SFCMM3 vector¹⁰ (dashed lines, black symbols) or the gibbon ape leukemia virus (GALV)-pseudotyped LNGFR-coding LNSN vector^{5, 11} (full lines, white symbols). Cells were stained at day 7, as previously described⁵ with an anti-human LNGFR mAb for 20 min at 4°C, then washed and incubated with goat-anti-mouse IgG-fluoroisothiocyanate (FITC) (Clinisciences, Montrouge, France) for 20 min at 4°C. Cells were then stained with phycoerythrin (PE)-labeled HLA-A*0201 tetramers complexed with EBV-derived peptides (lytic protein BMLF1: GLCTLVAML; latent protein EBNA-3C: VLEETSVML) or CMV-derived peptides (IE1 early antigen: LLDFVRFMGV; pp65 late antigen: NLVPMVATV) and CD8-phycocyanin-5 (PC5) mAb (Beckman-Coulter Immunotech, Marseille, France) for 30 min at room temperature.^{12, 13} After washing, cells were analyzed on a FACSCalibur flow cytometer using CellQuest software (Becton Dickinson). Analyses were performed after gating on the CD8⁺ lymphocyte population. For each individual staining, the percentage of LNGFR-positive cells is represented on the Y-axis for the bulk CD8+ cells (CD8+) and the tetramer-specific CD8+ cells (EBV: BMLF1+ and EBNA3C+; CMV: pp65+ and IE1+). Each symbol represents one independent experiment. CD3- CD56+ NK cells represented <5% of GMC^{6, 14} and were excluded from analysis by their low level of CD8 expression.

Full figure and legend (26K)

The preferential transduction of EBV-T or CMV-T cells could be related to two mechanisms. One mechanism relates to a potentially higher surface expression of the receptor for the envelope used to pseudotype the retroviral vector (ie PiT-2 for amphotropic Env¹¹) on tetramer-positive cells. At day 0, Pit-2+ cell frequency was significantly higher in tetramer-positive cells (BMLF1: 3.771.48%; pp65: 0.970.48%; n=8) than in the bulk CD8+ cells (0.080.04%; n=8; BMLF1 or pp65 versus CD8+: P<0.05, Wilcoxon's signed rank test). Upon CD3/IL-2 activation, the percentage of cells expressing PiT-2 was strongly increased in both tetramer-positive and tetramer-negative CD8+ cells (Figure 2a), suggesting an induction of Pit-2 expression rather than a preferential expansion of those cells expressing a high level of Pit-2. Pit-2 expression remained, at the time of retroviral-mediated transduction (day 3), significantly higher in tetramer-positive cells than in the bulk CD8+ T cells (Table 1). Use of CD3/CD28 beads induced similar levels of Pit-2 expression on bulk CD8+ cells or tetramer-positive cells to those observed when activation was performed with CD3 mAbs (data not shown). The apparent relationship between expression of Pit -2, a type III sodium phosphate cotransporter, and activation state in T cells might be related to an increased requirement for exogenous phosphate uptake in dividing cells.^{11, 17} Although the tools necessary to assess expression of the GALV receptor (Pit-1, another type III sodium phosphate cotransporter¹¹) are not available, it is possible that this receptor is also expressed at higher levels on EBV-T and CMV-T cells since preferential transduction was also observed with a GALV-pseudotyped vector.

Figure 2.

(a) Expression of the amphotropic receptor Pit-2 by pp65-specific CD8+ cells (shaded area) compared to bulk CD8+ (black line) 3 days after CD3/IL2 activation. Cells were stained with (or without, as a negative control) a fusion protein comprising the N-terminal 379-amino-acid receptor binding domain of the amphotropic murine leukemia virus Env SU region fused to a C-terminal rabbit immunoglobulin Fc tag (A_RBD),¹⁵ then FITC-conjugated goat-anti-rabbit IgG (Sigma, St Louis, MO, USA), then with PE-labeled pp65/HLA-A*0201 tetramer and CD8-PC5 staining. The gate for A_RDB+ (ie Pit-2+) cells (M1) was set with, as the negative control, CD8+ cells stained with FITC-conjugated rat-anti-rabbit IgG alone. Similar results were obtained with BMLF1/HLA-A*0201 tetramer-specific staining (not shown). (b) Cell division analysis of pp65-specific (open circles) and BMLF1-specific (open squares) CD8+ cells, compared to bulk CD8+ (black triangles), after CD3/IL-2 activation. Peripheral blood mononuclear cells were incubated (5 10⁶ cells/ml) for 10 min at room temperature in phosphate buffer saline (BioWhittaker, Emerainville, France) containing 5 M 5-carboxyfluorescein diacetate succinimidyl ester (CFSE, Molecular Probes, Leiden, Netherlands). Cells were then washed once with human serum (EFS Bourgogne/Franche-Comté, Besançon, France), then twice with phosphate buffer saline before CD3/IL-2 T-cell activation. At day 3, cells were stained with HLA-A*0201 tetramers and PC5-CD8 mAb. The graph represents the initial frequency of cells at day 0 that have subsequently undergone the indicated number (N) of cell divisions, calculated according to Wells et al.¹⁶ Briefly, if a single T cell divides n times between day 0 and day 3, it will generate 2ⁿ daughter cells. Thus, the absolute number of original, or precursor, T cells required to have generated these daughters is extrapolated by dividing the number of daughters under each division peak N by 2ⁿ. The day-3 CFSE staining profiles of the corresponding experiment are shown for the pp65-specific cells (shaded area) and bulk CD8+ cells (black line) in the upper right corner of the graph.

Full figure and legend (98K)

Table 1 - Percent of Pit-2+ cells within bulk CD8+ or tetramer-specific cells 3 days after CD3/IL2 activation.

Full table

The second mechanism that may have contributed to the preferential transduction of EBV-T and CMV-T cells could be related to their proliferative potential. Using CFSE in order to study precisely proliferation and generation of daughter cells, we found that there was increased proliferation of EBV-T and CMV-T cells as compared with bulk CD8+ cells following CD3/IL-2 stimulation. A CFSE staining was performed at day 0, before CD3/IL-2 activation, followed by a tetramer costaining performed at day 3. Our data demonstrate (Table 2) that the fraction of cells having undergone at least two cell divisions was significantly higher in BMLF1- and pp65-specific cells than in bulk CD8+ cells. One representative experiment is shown in Figure 2b, together with its CFSE profiles of bulk CD8+ cells and CMV-T cells. Moreover, the maximum number of cell divisions at this time point was also significantly higher in BMLF1- and pp65-specific cells than in bulk CD8+ cells (Table 2), resulting in an increased proportion of cells that are sensitive to oncoretroviral-mediated transduction. Similar results were observed when using CD3/CD28 beads instead of CD3 mAbs (data not shown).

Table 2 - Day-0 frequencies of CD8+ cell precursors that subsequently divide at least twice upon CD3 mAb/IL2 activation.

Full table

In conclusion, we show that polyclonal stimulation results in a biased retroviral-mediated transduction of certain T cells. Knowledge of such skewing in antigen specificity after retrovirus-mediated transduction will be important for the design of cellular immunotherapy approaches and benefit the targeting of T cells specific for CMV and EBV antigens. It remains to be determined whether such a preferential transduction is also observed in T cells specific for other antigens and can be extended to specific T-cell subsets.

References

1. Tiberghien P et al. Ganciclovir treatment of herpes simplex thymidine kinase-transduced primary T lymphocytes: an approach for specific in vivo donor T-cell depletion after bone marrow transplantation? Blood 1994; 84: 1333–1341. | PubMed | ISI | ChemPort |
2. Bonini C et al. HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft-versus-leukemia. Science 1997; 276: 1719–1724. | Article | PubMed | ISI | ChemPort |
3. Tiberghien P et al. Administration of herpes simplex-thymidine kinase-expressing donor T cells with a T-cell-depleted allogeneic marrow graft. Blood 2001; 97: 63–72. | Article | PubMed | ISI | ChemPort |
4. Robinet E et al. A closed culture system for the ex vivo transduction and expansion of human T lymphocytes. J Hematother 1998; 7: 205–215. | PubMed |
5. Sauce D et al. Retrovirus-mediated gene transfer in primary T lymphocytes impairs their anti-Epstein–Barr virus potential through both culture-dependent and selection process-dependent mechanisms. Blood 2002; 99: 1165–1173. | Article | PubMed | ChemPort |
6. Sauce D et al. Retrovirus-mediated gene transfer in polyclonal T cells results in lower apoptosis and enhanced ex vivo cell expansion of CMV-reactive CD8T cells as compared to EBV-reactive CD8T cells. Blood 2003; 102: 1241–1248. | Article | PubMed |
7. Contassot E et al. Retrovirus-mediated transfer of the herpes simplex type I thymidine kinase gene in alloreactive T lymphocytes. Hum Gene Ther 1998; 9: 73–80. | PubMed | ChemPort |
8. Miller DG, Adam MA, Miller AD. Gene transfer by retrovirus vectors occurs only in cells that are actively replicating at the time of infection. Mol Cell Biol 1990; 10: 4239–4242. | PubMed | ISI | ChemPort |
9. Koehne G, Gallardo HF, Sadelain M, O'Reilly RJ. Rapid selection of antigen-specific T lymphocytes by retroviral transduction. Blood 2000; 96: 109–117. | PubMed | ISI | ChemPort |
10. Garin MI et al. Molecular mechanism for ganciclovir resistance in human T lymphocytes transduced with retroviral vectors carrying the herpes simplex virus thymidine kinase gene. Blood 2001; 97: 122–129. | Article | PubMed | ISI | ChemPort |
11. Kavanaugh MP et al. Cell-surface receptors for gibbon ape leukemia virus and amphotropic murine retrovirus are inducible sodium-dependent phosphate symporters. Proc Natl Acad Sci USA 1994; 91: 7071–7075. | Article | PubMed | ChemPort |
12. Bodinier M et al. Efficient detection and immunomagnetic sorting of specific T cells using multimers of MHC class I and peptide with reduced CD8 binding. Nat Med 2000; 6: 707–710. | Article | PubMed | ISI | ChemPort |
13. Romero P et al. Ex vivo staining of metastatic lymph nodes by class I major histocompatibility complex tetramers reveals high numbers of antigen-experienced tumor-specific cytolytic T lymphocytes. J Exp Med 1998; 188: 1641–1650. | Article | PubMed | ISI | ChemPort |
14. Sauce D, el al. Influence of ex vivo expansion and retrovirus-mediated gene transfer on primary T lymphocyte phenotype and functions. J Hematother 2002; 11: 929–940. | Article |
15. Manel N et al. The HTLV receptor is an early T-cell activation marker whose expression requires de novo protein synthesis. Blood 2003; 101: 1913–1918. | Article | PubMed | ISI | ChemPort |
16. Wells AD et al. Following the fate of individual T cells throughout activation and clonal expansion. J Clin Invest 1997; 100: 3173–3183. | PubMed | ISI | ChemPort |
17. Chien ML, Foster JL, Douglas JL, Garcia JV. The amphotropic murine leukemia virus receptor gene encodes a 71-kilodalton protein that is induced by phosphate depletion. J Virol 1997; 71: 4564–4570. | PubMed | ChemPort |

Acknowledgements

This work was supported by the Programme Hospitalier de Recherche Clinique (PHRC 2000 #RC39), the Association Française contre la Myopathie and the Ministère de l'Enseignement Supérieur et de la Recherche (Centre Réseau de Développement des Thérapies Géniques (CRTG)), the Association pour la Recherche sur le Cancer (Grants # 4350 & # 7665), the Ligue contre le Cancer, Comité du Doubs and the European Community (Biomed contract # CT97-2074 and QLK3-CT-2001-01265). Patricia Mercier has benefited from a grant from the Ligue Contre le Cancer, Comité du Doubs. We also thank Dr Mark Bonyhadi (Xcyte Therapies Inc., Seattle, WA, USA) for providing us with CD3/CD28 beads.