Co-expression of cytokine and suicide genes to enhance the activity and safety of tumor-specific cytotoxic T lymphocytes

Abstract

The antitumor effect of adoptively transferred tumor-specific cytotoxic T lymphocytes (CTLs) is impaired by the limited capacity of these cells to expand within the tumor microenvironment. Administration of interleukin 2 (IL-2) has been used to overcome this limitation, but the systemic toxicity and the expansion of unwanted cells, including regulatory T cells, limit the clinical value of this strategy. To discover whether transgenic expression of lymphokines by the CTLs themselves might overcome these limitations, we evaluated the effects of transgenic expression of IL-2 and IL-15 in our model of Epstein Barr Virus-specific CTLs (EBV-CTLs). We found that transgenic expression of IL-2 or IL-15 increased the expansion of EBV-CTLs both in vitro and in vivo in a severe combined immunodeficiency disease (SCID) mouse model and enhanced antitumor activity. Although the proliferation of these cytokine genes transduced CTLs remained strictly antigen dependent, clinical application of this approach likely requires the inclusion of a suicide gene to deal with the potential development of T-cell mutants with autonomous growth. We found that the incorporation of an inducible caspase-9 suicide gene allowed efficient elimination of transgenic CTLs after exposure to a chemical inducer of dimerization, thereby increasing the safety and feasibility of the approach.

Figure 1

Construction and functionality of the retroviral vectors. Panel A is the schema of the retroviral vectors used to transduce EBV-CTLs. Panel B is a Western blot analysis showing the expression of ΔCD34 (top panel) and caspase-9 (middle panel) in COS-7 cells transduced with either iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v or ΔCD34v vectors. The lower gel shows the membrane reprobed with anti-GAPDH antibody. Panel C shows the transduction efficiency of EBV-CTLs measured as expression of a truncated form of CD34 (ΔCD34) on the cell surface by FACS analysis. Plots from a representative experiment are shown. Panel D illustrates the kinetics of cytokine release by EBV-CTLs transduced with either iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v or ΔCD34v vectors and stimulated with EBV-LCLs. Cytokines were detected in the culture supernatant at the indicated time after EBV-LCL stimulation and measured by specific ELISAs.

Figure 2

IL-2 and IL-15 transgenic CTLs expand in response to antigen stimulation. EBV-CTLs transduced with either iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v or ΔCD34v vectors were maintained in culture and stimulated once a week with autologous LCLs (E/T ratio 1:1) without addition of exogenous cytokines. Control EBV-CTLs transduced with the ΔCD34v vector were maintained in culture by adding rhIL-2 (50 U/mL). Data represent the mean plus or minus SD of cell expansion of 5 donors (panel A). IL-2 or IL-15 transgenic EBV-CTLs did not significantly expand when they were maintained in culture without specific antigen stimulation. Data represent the mean plus or minus SD of cell expansion of 3 donors (panel B).

Figure 3

IL-2 and IL-15 transgenic CTLs maintain their antigen specificity. EBV-CTLs transduced with either ΔCD34v or iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v vectors were analyzed for their phenotype and antigen specificity. Panel A illustrates the phenotypic analysis of transduced EBV-CTLs. Data represent the mean (± SD) of 5 donors. Panel B illustrates the result of a standard ⁵¹Cr release assay in which killing by CTLs of autologous LCLs, allogeneic LCLs, HSB-2 and K562 cell lines was tested at an E/T ratio of 20:1. Data represent the mean (± SD) of 5 donors. Panel C illustrates a representative staining of EBV-CTLs using tetramers targeting the HLA-B8 BZLF1 peptide RAKFKQLL and the HLA-A2 LMP-2 peptide CLGGLLTMV. Panel D illustrates the IFN-γ ELIspot assay of EBV-CTLs tested against the EBV-peptides HLA-A2 LMP-2 CLGGLLTMV, HLA-B8 EBNA3A FLRGRAYGL and HLA-B8 BZLF1 QAKWRLQTL. Data represent the mean (± SD) of 3 donors.

Figure 4

In vivo expansion of IL-2 and IL-15 transgenic CTLs. SCID mice engrafted with LCLs were injected with either EBV-CTLs control (ΔCD34v) or EBV-CTLs transgenic for IL-2 (iC.ΔCD34/IL-2v) or IL-15 (iC.ΔCD34/IL-15v) (10⁷ cells). To track their homing and in vivo expansion, CTLs were transduced with the vector encoding eGFP-FFLuc. CTL localization and expansion were monitored using an in vivo imaging system (Xenogen-IVIS Imaging System). Mice did not receive exogenous cytokines after CTL transfer. Panel A shows images of representative mice. The signal intensity measured as photon/sec/cm²/sr (p/s/cm²/sr) was increased in mice receiving CTLs transgenic for IL-2 or IL-15 compared with control cells (ΔCD34v). Panel B illustrates the maximum increase in bioluminescence obtained in 20 mice per group. The expansion of iC.ΔCD34/IL-15v CTLs and iC.ΔCD34/IL-2v CTLs was statistically significant compared with control ΔCD34v (P < .001 and P < .002, respectively). IL-2 and IL-15 transgenic CTLs did not significantly expand in response to allogeneic EBV-LCLs. (C) To evaluate whether the increase in bioluminescence signal corresponded to an increased number of CTLs infiltrating the tumor, mice were euthanized and T-cell infiltration in the bioptic samples was measured using antihuman CD3 staining and FACS analysis (ΔCD34v = 3.8 × 10⁵ p/s/cm²/sr; iC.ΔCD34/IL-2v = 3 × 10⁶ p/s/cm²/sr; iC.ΔCD34/IL-15v = 3.7 × 10⁶ p/s/cm²/sr).

Figure 5

Activation of the iCasp-9 suicide gene significantly eliminates IL-2 and IL-15 transgenic CTLs. EBV-CTLs transduced with either ΔCD34v or iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v vectors were plated at 10⁶ cells/well and incubated with or without CID AP20187 at 50 nM. Twenty-four hours later cells were collected, stained with CD34-PE antibody, and the percentage of residual transgenic CTLs was evaluated by FACS analysis. Significant reduction of CD34⁺ cells after incubation with CID was obtained only for EBV-CTLs transduced with either iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15 vectors incorporating the iCasp-9 gene. Panel A illustrates ΔCD34 expression in a representative experiment. Panel B summarizes the effects of the CID on 4 different EBV-CTL lines transduced with either iC.ΔCD34/IL-2v, iC.ΔCD34/IL-15v, or ΔCD34v vectors. The y axis represents the mean (± SD) for CD34⁺ CTLs in the CTL lines before or after incubation with CID. The percentage of CD34⁺ cells remained unchanged in control CTLs transduced with the ΔCD34v vector lacking the suicide gene. In contrast, a significant reduction in the percentage of CD34⁺ CTLs was observed for CTLs transduced with either iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v vectors. Panel C shows that the elimination of transgenic CTLs was mediated by induction of apoptosis as assessed by Annexin-V/7-AAD staining. Panel D shows that the induction of apoptosis/necrosis by CID was dose dependent. CTLs transgenic for IL-2 or IL-15 and selected using CD34 magnetic beads were incubated with different doses of CID, and then 24 hours later the induction of apoptosis/necrosis was evaluated by staining with Annexin-V/7-ADD and FACS analysis.

Figure 6

Activation of the iCasp-9 suicide gene abrogates cytokine production and long-term expansion of IL-2 and IL-15 transgenic CTLs. EBV-CTLs transduced with either iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v vectors were plated at 10⁶ cells/well and incubated with or without the CID AP20187 (50 nM). Panel A illustrates the production of transgenic cytokines IL-2 or IL-15 by iC.ΔCD34/IL-2v and iC.ΔCD34/IL-15v EBV-CTLs before and after exposure to the CID. Neither IL-2 nor IL-15 cytokines could be detected in the supernatants from CTLs treated with the CID. Data represent the mean (± SD) of 4 experiments. Panel B illustrates the long-term expansion of EBV-CTLs transduced with either iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v vectors and stimulated once a week with autologous LCLs. Viable cells were counted by trypan blue exclusion once a week before EBV-LCL restimulation. Exposure to a single dose of CID (50 nM) ablated CTL expansion, while nonexposed CTLs continued to expand. Data represent mean (± SD) of 4 experiments. Panel C illustrates a representative experiment in which IL-2 and IL-15 transgenic CTLs obtained 24 hours after CID exposure were sorted based on the expression of Annexin-V. These CTLs were then cultured without any further addition of CID and stimulated with EBV-LCLs. Annexin-V/7ADD staining showed that these cells progressed to a late stage of apoptosis/necrosis (Annexin-V⁺/7ADD⁺) by days 7 to 9. Panel D shows that CID induced elimination of CD34⁺ cells for IL-2 or IL-15 transgenic CTLs even in their resting phase, 7 days after the last antigen stimulation.

Figure 7

Activation of the iCasp-9 suicide gene eliminates the IL-2 and IL-15 transgenic CTLs in vivo. SCID mice engrafted subcutaneously with LCLs were injected intravenously with EBV-CTLs transduced with either iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v vector, sorted for ΔCD34 expression, and transduced with eGFP-FFLuc vector. When the CTLs were expanding, mice were treated with 2 to 3 doses of the CID AP20187 (50 μg) intraperitoneally 2 days apart. The persistence of the transgenic cells was monitored in vivo using the bioluminescence system. Panel A illustrates in a representative experiment the reduction of the bioluminescence after CID administration. Bioluminescence was significantly reduced in mice receiving IL-2 or IL-15 transgenic CTLs after treatment with CID. In contrast, the signal was not diminished in mice receiving control ΔCD34⁺ CTLs lacking the expression of the suicide gene. The bioluminescence continued to increase in mice receiving IL-15 transgenic CTLs nontreated with CID. Panel B shows the kinetics of bioluminescence in 7 mice (closed symbols) before and after treatment with CID. In mice receiving EBV-CTLs expressing either iC.ΔCD34/IL-2v or iC.ΔCD34/IL-15v followed by the CID greater than 1 log reduction in bioluminescence was observed. In contrast, bioluminescence continued to increase in mice nontreated with CID (4 representative mice, open symbols). (C) Mice showing > 10⁶ photons were euthanized to evaluate the infiltrate of CTLs within the tumor by FACS analysis after staining with antihuman CD3 antibody (top left panel). Mice with similar level of bioluminescence signal were treated with CID and 24 to 72 hours later euthanized to evaluate the effective reduction of CTL infiltration by FACS analysis after staining with antihuman CD3 antibody (top right panel). CID did not reduce the number of CD3⁺ cells in mice receiving control CTLs transduced with ΔCD34v vector (bottom left and right).