Phenotype and functional evaluation of ex vivo generated antigen-specific immune effector cells with potential for therapeutic applications

Healthy donors' buffy coats were obtained from Civitan Blood Center (Gainesville, FL, USA). PBMC were prepared by gradient density centrifugation in Ficoll-Hypaque (GE Healthcare Bio-Sciences AB, NJ, USA) as previously described. [16] Viability was determined by trypan blue staining. Autologous B lymphoblastoid cell line (BLCL) was generated by transforming peripheral blood B lymphocytes with EBV as described previously. [17] The BLCL were continuously propagated in RPMI 1640 medium supplemented with 2 mM L-glutamine, 100 μg/ml streptomycin, 100 IU/ml penicillin and 10% heat inactivated fetal bovine serum (FBS) at 37°C with 5% CO₂.

The mixtures of 11 amino acid overlapping pentadecapeptides (122 peptides) spanning the entire 497 amino acids (NCBI Accession number P13285) of LMP2A of the EBV (human herpesvirus 4, strain B95-8) and the Wilms' tumor antigen (WT1, 449 amino acids, 110 peptides) were purchased from JPT Peptide Technologies GmbH (Berlin, Germany).

PBMC were plated into 6-well plate at 1 × 10⁷ cells/well and adhered for 2 hours in AIM-V (Gibco-BRL, CA, USA). The non-adherent cells were removed gently and frozen as source of lymphocytes for co-culture use. Adherent monocytes were cultured in AIM-V supplemented with 50 ng/ml of GM-CSF and 25 ng/ml IL-4 (eBiosource International, Inc. Camarillo, CA, USA). For the generation of 2 day DC, the adherent cells were cultured with GM-CSF and IL-4 for 24 h and incubated for another 24 h with TNFα (50 ng/ml), IL-1β (10 ng/ml), IL-6 (10 ng/ml, all from R&D systems, MN, USA) and PGE2 (1 uM, Sigma-Aldrich, MO, USA) to induce maturation. For the generation of 5 day DC, cells were cultured with GM-CSF and IL-4 for 5 days. On day 3, half of the medium was replaced with fresh medium containing GM-CSF and IL-4. On day 5, the immature DC were induced into maturation with TNFα (50 ng/ml), LPS (1 μg/ml, Sigma-Aldrich) and IFN-γ (50 ng/ml, R&D systems).

PBMC were resuspended in AIM-V plus 5% human AB serum at 1 × 10⁷ cells/ml and mixed with pentadecapeptides for EBV-LMP2A (10 ug/ml), mouse anti-human CD28 antibody (Ab, 1 ug/ml, eBioscience, San Diego, USA) and human β2-microglobulin (1 ug/ml, Sigma) to enhance antigen presentation and costimulation. The cells were incubated in a 37°C humidified incubator for 3–13 hours. IFN-γ secreting cells were enriched with the IFN-γ Catch Reagent and CD137 positive cells with PE-conjugated monoclonal anti-CD137 Ab (clone 4B4-1), followed by affinity isolation using anti-PE microbeads according to the manufacturer's instruction (Miltenyi Biotech Inc. Auburn, CA, USA).

DC-activated antigen-specific immune cells were generated as previously described. [16] In brief, mature 2 day DC were loaded with LMP2A peptides (2.5 ug/ml) for 2 h and irradiated (20 Gy, or 2,000 rads). The antigen-pulsed DC were cocultured with autologous non-adherent PBMC at a ratio of 1:20 in AIM-V with 5% human AB serum. On day 3, half of the medium was replaced with fresh medium supplemented with IL-2 (12.5 U/ml), IL-7 (5 ng/ml) and IL-15 (20 ng/ml, all from Gentaur, Aachen, Germany). Half of the medium was replaced with fresh medium with cytokines every other day.

Mature DC were analyzed using a four-color panel of monoclonal Ab including PE-anti-CD14, FITC-anti-HLA-DR, PE-anti-CD86, APC-anti-CD1a, PE-anti-CD 83, APC-anti-CD40 and FITC-anti-HLA-I (BD Biosciences, San Jose, CA, USA), APC-anti-DC-SIGN and PE-cy7-anti-CD11c (eBioscience), and incubated for 30 min at 4°C. Isotype-matched antibodies were used for controls. The Ab-labeled cells were washed twice with PBS containing 1% FBS and analyzed with FACSCaliber or FACSAria using FACSDiva software (BD Biosciences) and Flowjo software (Tree Star, Inc., Ashland, OR, USA). For memory T cell analysis, the cells were stained with PE-anti-IFN-γ or PE-anti-CD137, in combination with APC-anti-CD4 (clone RPA-T4), Pacific blue-anti-CD8 (clone RPA-T8), FITC-anti-CD45RA (clone HI100), PE-cy7-anti-CCR7 (clone 3D12), Percp-cy5.5-anti-CD28 (clone 293, all from BD Biosciences) and Alexa-fluo 750-anti-CD27 (clone O323, eBioscience) at 4°C for 30 min and washed twice with PBS containing 1% FBS. The percentage of different T cell subsets was analyzed using FACSAria with FACSDiva and Flowjo softwares.

These assay were performed as described. [18] Briefly, 2 × 10⁵ LMP2A-specific T cells were stimulated for 5 h in a 96-well plate with irradiated (20 Gy) antigen-loaded autologous DC. Monensin A (Sigma-Aldrich) and FITC-conjugated Abs for CD107a or isotype matched Abs (BD Pharmingen, San Diego, CA, USA) were added 1 hour after stimulation and incubated for 5 hours. Cells were then stained with Abs against CD4 and CD8 and fixed, permeabilized with Cytofix/Cytoperm solution and stained with Ab against IFN-γ (all from BD Pharmingen) at 4°C for 20 min. Unrelated peptide group was included as a negative control for spontaneous CD107a expression and/or cytokine production.

Peptide-major histocompatibility complex (MHC)-pentamer conjugate specific for EBV-LMP2A/TYGPVFMCL (HLA-A*2402 restricted) was purchased from Proimmune (Springfield, VA, USA). T cells were incubated with PE-labeled peptide MHC-pentamer at room temperature for 10 min, washed and stained with APC-anti-CD3 and FITC-anti-CD8 Ab (BD Pharmingen) on ice for 30 min, and analyzed using FACSAria. At least 1 × 10⁵ events were collected for each sample.

The CFSE-based proliferation assay was performed as previously described [19]. Briefly, LMP2A-specific T cells were washed and labeled with 1 uM CFSE (Molecular Probes, Inc., Eugene, OR, USA). The labeled cells were washed and plated into 96-well U-bottom wells at 1 × 10⁵ cells per well. Autologous DC were loaded with peptides (2.5 ug/ml) for 2 hours and irradiated (20 Gy). The irradiated DC were added to the CFSE-labeled T cells at a ratio of 1:20 and cultured in AIM-V with 5% human AB serum. After 4 days, cells were harvested and analyzed with flow cytometry.

The immune cell cytotoxicity assay was based on Jedema et al. with minor modifications. [20] The target cells were washed with PBS, and labeled with 1 uM CFSE (Molecular Probes) at 5 × 10⁶ cells per ml at 37°C for 15 minutes. The reaction was stopped with the addition of 10 volumes of complete RPMI containing 10% FCS, followed with a 30 min incubation at 37°C. After two washes, the CFSE-labeled target cells were resuspended in AIM-V containing 5% human AB serum. The cell concentration was adjusted to 1 × 10⁵ cells/ml before plating into 96-well microtiter plates at 100 ul/per well. The effector cells were then mixed with target cells at a ratio of 1:1. The plates were incubated in a humidified atmosphere of 5% CO₂ and 37°C. The target cells included irradiated (20 Gy) autologous DC loaded with LMP2A peptides or control WT1 peptides, and irradiated (100 Gy) autologous BLCL. The effector cells included LMP2A peptide-stimulated IFN-γ-selected cells, LMP2A pulsed DC-expanded effector cells, LMP2A peptide-stimulated IFN-γ-negative PBMC, and control PBMC. After 6 hour of incubation, the cells were mixed with 10,000 Flow-Count Fluorospheres (BD Pharmingen) and followed by flow cytometry analysis. To stain for dead cells, 7-AAD (10 ug/ml) was added and incubated for 30 min on ice. For each sample, 5,000 microbeads were acquired, facilitating the calculation of absolute numbers of target cells. The percentage of survival was determined as the following:

(R refers to different CTL ratio groups; R = 0 refers to the CTL = 0 control)

The small population of immune effector cells specific for a particular antigen may be isolated based on expression of antigen-specific activation markers. IFN-γ and CD137 have been identified as antigen-specific activation markers. Upon antigen stimulation, the IFN-γ secreting cells can be captured with anti-IFN-γ Ab conjugated with an anti-surface marker Ab. Alternatively, CD137 positive cells can be directly isolated using anti-CD137 Ab. Specific T cell immune response to the peptide library spanning the entire LMP2A sequence has been previously documented. [21] We first established the expression kinetics of IFN-γ and CD137 in PBMC after stimulation with a pool of EBV LMP2A pentadecapeptides, as reports by others, the expression peaked at 6–12 hr for IFN-γ and 24 hr for CD137 (data not shown). [15, 22] The IFN-γ- and CD137-expressing cells were isolated using Ab-conjugated magnetic bead affinity columns (Miltenyi Biotech). After examining a large number of donors, we found that the antigen-specific response of an individual could be either CD8 or CD4 T cell dominant, which can be donor- and/or antigen-dependent. Representative results are illustrated in Fig. 1A (a CD4-dominant donor to the left, and a CD8-dominant donor to the right). The results indicated that the CD4/CD8 ratio of the IFN-γ-selected cells correlated with the donor's T cell dominance phenotype. However, the CD137-enriched cells consistently showed an increased CD4 to CD8 ratio regardless of the donor's phenotype (Fig. 1A, bottom); the latter could be due to the increased proportion of CD137 positive CD4 cells in the total population of T cells before and after stimulation (Fig. 1B). However, quantitative analysis of the expression level of CD137 demonstrated that CD8 T cells expressed higher density of surface CD137 than did CD4 T cells in both CD4 and CD8 dominant individuals (see geometric means in Fig. 1C).

The phenotype of the immune cells of the two affinity isolation methods has not been characterized in the past due to the limited number of the total harvested cells. We subjected the IFN-γ- and CD137 affinity-purified cells to a multi-color flow cytometry analysis for central memory (Tcm, CCR7+, CD45RA-), effector memory (Tem, CCR7-, CD45RA-), terminal effector (Teff, CCR7-, CD45RA+), differentiation (CD27 and CD28) and migration (CCR7) markers of T cells in addition to CD4, CD8 and IFN-γ (or CD137). We found that there was a trend of increased early-differentiated Tcm and Tem cells in the dominant populations (Fig. 2A and 2B, CD4- and CD8-dominant donors, respectively).

To see if the enriched immune effector cells could be further expanded, we cultured them on irradiated autologous feeder PBMC; the cells expanded approximately 600-fold in two weeks (not shown). Before expansion, the CD137 enriched cells contained a higher ratio of CD4 T cells (Fig. 1B, 37.6 vs. 13.1); however, the donor's dominant phenotype was restored after expansion in culture (Fig. 3A). Furthermore, we consistently observed an increased population of CD3^-CD56⁺ cells in the CD137+ cell expansion culture compared with the IFN-γ cell expansion culture (Fig. 3A top, 32.7% and 8.5% versus 2.3% and 2.6%, for a CD4-dominant and a CD8-dominant donors, respectively). Compared with the freshly isolated effector cells that contained more memory effector cells (Fig. 2), the cultured cells differentiated toward Tem and (Teff) cells after expansion (Fig. 3B and 3C). Intracellular staining for effector cytokines showed that the expanded IFN-γ and CD137 cells displayed high antigen-specific activities, up to 59% and 34% for IFN-γ and CD137 selected cells, respectively, when restimulated with DC-pulsed with the specific antigens (LMP2A petadecapeptides) or autologous EBV positive BLCL (Fig. 4A and 4B, control: WT-1 peptide-pulsed DC). To demonstrate antigen-specific cytolytic activity, we incubated these cells with different target cells as illustrated in Fig. 4C. The effector cells killed specific target cells with high specificity including autologous BLCL and LMP2A peptide-pulsed DC, but not control WT-1 peptide-pulsed DC (Fig. 4C, effector to target ratio 1:1).

Antigen-specific immune effector cells can also be generated through DC activation. The latter protocol includes a DC preparation step, followed by antigen exposure and lymphocyte coculture. For the preparation of DC, we compared the 2 day and the 5 day protocols. [23] We analyzed the surface markers for plastic-adherent monocytes and mature DC with flow cytometry (data shown in Additional file 1), and analyzed phenotypes of the 2 day and the 5 day mature DC (Fig. 5). The 2 day DC expressed higher levels of class I/II MHC (HLA-I and HLA-DR), costimulatory molecules (CD86 and CD40) and maturation marker CD83. We also found that the 2 day DC induced primary and secondary immune response against viral or cancer antigens at efficiencies equal to or better than the 5 day DC (manuscript in preparation). Therefore, the 2 day DC protocol was adopted for later experiments.

The 2 day DC were pulsed with the pooled LMP2A pentadecapeptides, irradiated and cocultured with autologous lymphocytes at a ratio of 1:20. The cell number usually decreases around day 5 after coculture, followed by an increase of a few fold around day 17, suggesting a loss of non-specific cells followed by expansion of antigen-specific cells (see representative growth curves in Fig. 6A). Flow cytometry analysis of the DC-activated cells in culture from four different donors at day 0, 12 and 19 indicated that most donors generated CD3+CD56- T cells, but some generated a large proportion of CD3-CD56+ NK cells and CD3+CD56+ cells (e.g. donor 3, Fig. 6B). The relative ratios of CD4+ T cells, CD8+ T cells, NK cells and CD3+CD56+ cells in the coculture appeared to be donor-dependent (Fig. 6B and 6C). Furthermore, phenotype analysis with multi-color flow cytometry demonstrated that the ex vivo expanded cells contained mostly Tem and Teff cells (Fig. 7A and 7B, representative of CD4- and CD8-dominant donors).

We next evaluated the DC-activated immune cells for antigen-specific effector functions based on intracellular staining of IFN-γ, IL-2, and the degranulation marker CD107a after restimulation with peptide-loaded DC (Fig. 8A). For direct comparison, the same donor whose cells were analyzed for the IFN-γ-selected effector functions (Fig. 4B) was chosen. Analyses of the immune effector cells without further stimulation (IE cell alone), or stimulated with non-specific peptides (Non-specific) or with the LMP2A pentadecapeptides (LMP2A-specific) demonstrated LMP2A-specific expression of IFN-γ, IL-2 and CD107a in the CD8 T cells, and to a lesser extent, the CD4 T cells (Fig. 8A top: IL-2 and IFN-γ, and bottom: CD107a and IFN-γ). The antigen-specific effector function was further confirmed with the CFSE-based proliferation assay, as well as MHC-peptide pentamer specific for the LMP2A epitope-specific T cell receptors of the CD8 T cells (data not shown).

To demonstrate antigen-specific killing, we directly compared the DC-activated cells to the IFN-γ captured cells (as shown in Fig. 4C) using an in vitro cytolytic assay (Fig. 8B). At an effector to target cell ratio of 1:1, the immune cells effectively killed autologous BLCL and LMP2A-pulsed DC (DC-LMP2A), but not DC pulsed with control WT1 pentadecapeptides (DC-WT1). Therefore, effector function analyses including IFN-γ release and cytotoxicity assays suggest that the DC-activated cells had lower activities than the expanded IFN-γ effector cells.