Prothymosin α and a prothymosin α-derived peptide enhance T_H1-type immune responses against defined HER-2/neu epitopes

We have previously shown that proTα and proTα(100–109) efficiently mature human DCs in vitro, as indicated by the induction of surface expression of established DC-markers to levels comparable to those induced by lipopolysacharide (LPS) [23] or tumor necrosis factor (TNF)-α (this report). As shown in Figure 1, LPS-matured DCs significantly upregulated the expression of HLA-DR, CD11b, CD80, CD83, CD86 and CD40, to levels comparable to TNF-α-matured DCs (p < 0.001 compared to iDCs for all values). Similarly, both agents caused a reduction of CD14 expression. In agreement with our previous study [23], iDCs matured with either proTα or its decapeptide presented a similar phenotype to LPS- or TNF-α-matured DCs, upregulating the expression of all co-stimulatory molecules (p < 0.05-0.001, compared to iDCs; Figure 1) and downregulating CD14 (p < 0.01, compared to iDCs).

In conjunction with their phenotype, functionally competent mature DCs secrete pro- and anti-inflammatory cytokines [6]. Therefore, we assessed the production of TNF-α, interleukin (IL)-12 and IL-10 from iDCs and mature DCs and determined the IL-12:IL-10 ratio. We present these data because high TNF-α levels, as well as a balance between IL-12:IL-10 in favor of IL-12 have been proposed to promote T_H1-polarization [24]. As shown in Figure 2, iDCs produced low amounts of all three cytokines, whereas mature DC supernatants collected 48 h after addition of LPS or TNF-α were rich in TNF-α and IL-12. Compared to iDCs, higher TNF-α and IL-12 levels were also found in supernatants of cultures of DCs matured with proTα and proTα(100–109). Although the absolute concentrations of cytokines varied among the differentially matured DCs, their overall cytokine-production patterns were comparable. Most importantly, the mean IL-12:IL-10 ratios were similar (6.61, 6.45, 7.89 and 5.18, for LPS-, TNF-α-, proTα- and proTα(100–109)-matured DCs, respectively). These data suggest that the peptides bias immunoreactivity towards a pro-inflammatory T_H1-type of response.

Finally, in the presence of a blocking antibody against TLR-4 (a-TLR-4; Figure 2), lower amounts of cytokines were secreted by LPS-, proTα- and proTα(100–109)-matured DCs, but not TNF-α-matured DCs. Notably, a-TLR-4 reduced the levels of LPS- and proTα-induced IL-12 production by 55 and 47%, respectively (p < 0.05), implying that IL-12 production by LPS- and proTα-matured DCs is at least partially, TLR-4-dependent [25].

As optimally matured DCs prime antigen-specific CD4+ and CD8+ T cell activation and proliferation of naive T cells [7], we next assessed whether proTα- and proTα(100–109)-matured DCs are functionally competent, i.e., induce in vitro the selective expansion of tumor antigen-specific T cells.

Monocyte-derived DCs matured for 48 h with proTα, proTα(100–109) or TNF-α (used as a conventional DC maturation agent) were pulsed with the HLA-A2 and HLA-DR4-restricted HER-2/neu(369–377) [HER-2(9₃₆₉)] and HER-2/neu(776–790) [HER-2(15₇₇₆)] epitopes, and used to prime autologous naïve T cells isolated from the peripheral blood of HLA-A2+/DR4+ healthy donors. T cells were restimulated twice, at weekly intervals, with similarly matured autologous DCs. Twelve hours after the third stimulation their production of TNF-α, interferon (IFN)-γ, IL-2, IL-4, IL-10 and IL-17 was analysed. Figure 3 shows the percentages of IFN-γ+, IL-2+, IL-4+ and IL-10+ CD4+ T cells from one representative donor of 5 tested with similar results (Additional file 1: Table S1A). In the presence of unpulsed TNF-α-matured DCs, only a low percentage of CD4+ T cells produced IFN-γ (0.02%), which was significantly increased (23.30%) in the presence of the HER-2/neu peptides. An analogous increase in the percentage of IFN-γ-producing cells was also recorded in CD4+ T cells stimulated with proTα- or proTα(100–109)-matured DCs in the presence of the same peptides (21.78% and 22.93%, respectively, compared to 0.01% and 0.02% in the absence of HER-2/neu peptides; Figure 3). The percentages of IL-2-producing CD4+ T cells in all groups were also significantly enhanced upon stimulation with HER-2/neu peptide-pulsed DCs (35.02% for TNF-α-, 31.51% for proTα- and 37.18% for proTα(100–109)-matured DCs, compared to 0.08%, 0.09% and 0.71% of the respective unpulsed groups; Figure 3). A similar enhancement was also seen for TNF-α-producing CD4+ T cells (Additional file 1: Table S1A). In contrast, production of IL-4 and IL-10 was only marginally increased when CD4+ T cells were stimulated with HER-2/neu peptide-pulsed DCs, regardless of the factor used for DC maturation (Figure 3). Specifically, the percentage of CD4+ T cells producing IL-4 was increased from 0.01% (without peptides) to 1.17% among T cells stimulated with peptide-pulsed TNF-α-matured DCs, from 0.01% to 0.61% in the proαα- and from 0.02% to 0.82% in the proTα(100–109)-matured DC groups. A minor enhancement was also recorded for IL-10 production; IL-10+ cells increased from 0.03% to 0.11% in the TNF-α-, from 0.01% to 0.02% in the proαα- and from 0.01% to 0.02% in the proTα(100–109)-matured DC cultures without and with HER-2/neu peptides, respectively. IL-17 production exhibited a similar pattern of marginal increase in CD4+ T cells stimulated in the presence of all matured antigen-pulsed DCs (Additional file 1: Table S1A). These data suggest that proTα- and proTα(100–109)-matured DCs are immunocompetent and in the presence of specific tumor antigenic epitopes, favor the in vitro production of pro-inflammatory (IFN-γ, IL-2, TNF-α), rather than anti-inflammatory cytokines (IL-4, IL-10) and IL-17 by CD4+ T cells, inducing T_H1-type immune responses.

Cell-mediated immunity requires initial collaboration between T_H1 CD4+ and CD8+ T cells [26]. Thus, we next investigated whether proTα- and proTα(100–109)-matured DCs can elicit tumor peptide-specific cytotoxic T cell immune responses.

CD8+ T cells recovered from the same stimulation cultures as aforementioned were assessed for the intracellular production of TNF-α. As shown in Figure 4A, they also exhibited a similar pattern of enhanced cytokine production in the presence of HER-2/neu peptides as did CD4+ T cells. The percentage of TNF-α+ cells was increased from 0.35% (unpulsed) to 47.52% (pulsed) when T cells were stimulated with TNF-α-matured DCs, and from 0.12% to 45.38% for proTα- and from 0.13% to 42.88% for proTα(100–109)-matured DCs (Figure 4A). In addition and in accordance with the results recorded for CD4+ T cells, IL-2- and IFN-γ-producing CD8+ T cells were also increased in the presence of peptide-pulsed DCs in the cultures, whereas differences in the percentages of IL-10-producing CD8+ T cells were only marginal (Additional file 1: Table S1A).

The same cells were assessed for the expression of CD107a, as a surrogate marker for cytotoxicity [27]. In the absence of HER-2(9₃₆₉), a low percentage of CD8+ T cells stimulated with TNF-α-matured DCs expressed CD107a (3.70%; Figure 4A), which increased when cells were stimulated with HER-2(9₃₆₉)-pulsed DCs (54.75%). Similar CD107a upregulation was observed in CD8+ T cells stimulated with proTα- and proTα(100–109)-matured HER-2(9₃₆₉)-pulsed DCs (36.86% and 41.99%, respectively, compared to 2.80% and 2.17% of the unpulsed groups; Figure 4A). Since TNF-α mediates target cell damage and CD107a-expressing CD8+ T cells are cytotoxic [27], our results suggest that proTα- and proTα(100–109)-matured DCs efficiently activate CD8+ cytotoxic T cells, which were able to kill targets presenting the immunogenic epitope versus which they were primed.

Cytotoxic activity was verified by using ⁵¹Cr-labeled HLA-A2+ T2 cells loaded with HER-2(9₃₆₉) or an irrelevant epitope, tyrosinase(369–377) [tyr(9₃₆₉)]. CD8+ T cells thrice stimulated with peptide-pulsed TNF-α-, proTα- or proTα(100–109)-matured DCs were coincubated with these peptide-loaded T2 targets. The results showed that CD8+ T cell mean cytotoxicity against non-peptide loaded T2 targets did not exceed 30% in any group (26.9% for TNF-α, 23.7% for proTα- and 21.4% for proTα(100–109)-matured DCs; Figure 4B), whereas HER-2(9₃₆₉)-loaded T2 targets were lysed twice as efficiently by CD8+ T cells recovered from all stimulation cultures (49.9% for TNF-α-, 46.6% for proTα- and 40.4% for proTα(100–109)-matured DCs; Figure 4B). Cytotoxicity against T2 targets loaded with tyr(9₃₆₉) was low and in no instance exceeded 30%. These cytotoxic responses were significantly decreased by monoclonal antibody (mAb) to MHC class I molecules, suggesting that the CD8+ T cells generated by our stimulation protocol are MHC class I-restricted and HER-2(9₃₆₉)-specific (Figure 4B).

Based on previous studies associating T cell polyfunctionality with high IFN-γ production and the quality of the elicited responses [28, 29], we carried out a functional analysis of the HER-2(9₃₆₉)-specific CD8+ T cells generated in these experiments. Using FlowJo software, we analyzed their ability to produce effector cytokines (IFN-γ, TNF-α and IL-2) and to degranulate (expression of CD107a). Quantifying the fraction of the responsive CD8+ T cells producing any one (1+), any two (2+), any three (3+) or all four (4+) mediators, we observed that approximately a mean ~16% of the responsive CD8+ T cells were 2+ cells, regardless of the agent used to mature the DCs that stimulated them (16.26% for TNF-α; 16.92% for proTα; 15.95% for proTα(100–109); Figure 5). In all experimental groups, 3+ cells were also detected in increased percentages (8.13% for TNF-α; 7.03% for proTα; 4.34% for proTα(100–109)). In contrast, very few 4+ cells were detected under any conditions. Taken together, these data suggest that proTα- or proTα(100–109)-matured DCs were able to induce polyfunctional (2+, 3+) CD8+ peptide-specific T cell responses at least as well as TNF-α-matured DCs.

T cell proliferation was assessed by the incorporation of ³H-thymidine. HER-2(15₇₇₆)-sensitized T cells coincubated with HER-2(15₇₇₆)-, tyrosinase(448–462) [tyr(15₄₄₈)]-pulsed or unpulsed DCs, specifically proliferated in response only to the HER-2/neu epitope (Table 1). ProTα- and proTα(100–109)-matured DCs showed relatively high mean stimulation indices (S.Is) (2.64 and 2.26, respectively), comparable to those recorded for TNF-α-matured DCs (3.09). Addition of mAb to MHC class II molecules reduced mean S.Is in all groups (0.87 for TNF-α; 0.93 for proTα; 0.98 for proTα(100–109)). These results suggest that following our in vitro culture protocol, peptide-reactive T cells are generated, which proliferate only in a HER-2(15₇₇₆)-dependent MHC class II-restricted manner.

We have previously reported that stimulation of human monocytes with proTα upregulated IRAK-4, a protein kinase involved in TLR downstream signaling [19], whereas Mosoian et al. [20] showed that proTα ligates TLR-4 and signals through both TRIF- and MyD88-dependent pathways. To determine whether TLR-4 is triggered by our peptides, we studied the kinetics of TLR-4 surface expression on proTα and proTα(100–109)-stimulated DCs. Immature DCs (iDCs) and DCs matured with LPS (a known TLR-4 ligand; [30]), proTα or proTα(100–109) for 15 min, 30 min, 1 h, 18 h and 36 h were analyzed by flow cytometry. The percentage of surface TLR-4 expression over time is presented in Figure 6. Maturation of DCs with LPS led to an early (15 and 30 min) decrease of TLR-4 expression (by ~15%) due to internalization [31], and a subsequent increase from 1 to 18 h [32]. At 36 h post-LPS addition, TLR-4 expression was lower and comparable to that of iDCs (0 h). ProTα and proTα(100–109), marginally downregulated TLR-4 expression at 30 min and, similarly to LPS, transiently increased it from 1 to 18 h. As with LPS-matured DCs, basal levels of TLR-4 expression were detected at 36 h post-maturation. The similar kinetics of TLR-4 expression in LPS-, proTα- and proTα(100–109)-matured DCs is consistent with the notion that the two peptides interact with TLR-4.

To extend these findings, we next investigated the intracellular expression levels of three adaptor molecules that participate in signaling pathways downstream of TLR-4, namely TRIF, an adaptor molecule common to TLR-3 and -4 signaling; TIRAP, a signaling adaptor common to TLR-2, and -4; and MyD88, a molecule upregulated upon ligation of all TLRs except TLR-3 [33]. We specifically selected these three adaptors because this constellation is unique to TLR-4 activation. Total cell extracts from iDCs and DCs matured with LPS, proTα or proTα(100–109) for 1 h and 18 h were immunoblotted (Figure 7A). Upon densitometric quantification of each protein band detected, expression relative to GAPDH was calculated. As shown in Figure 7B, addition of LPS led to a significant ~2-3 fold increase of the expression of all three adaptors within 1 h (3.05 for TRIF, 2.88 for TIRAP and 1.81 for MyD88) relative to iDCs (1.38 for TRIF, 1.00 for TIRAP and 0.74 for MyD88). At 18 h post-addition of LPS, expression of all adaptors was decreased and again comparable to iDCs (1.35 for TRIF, 1.16 for TIRAP and 1.20 for MyD88). A similar trend of increased expression was also observed 1 h after addition of proTα (2.537 for TRIF, 2.28 for TIRAP and 1.577 for MyD88) or proTα(100–109) (1.62 for TRIF, 1.423 for TIRAP, 1.06 for MyD88), although in the latter case, the detected protein levels were lower. As with LPS, 18 h after proTα or proTα(100–109) DC-stimulation, the expression of TRIF, TIRAP and MyD88 was reduced and was similar to iDCs. These data, in conjunction with the cytokine profile shown in Figure 2, suggest that LPS, proTα, and possibly also proTα(100–109), activate DCs at least partly through one common TLR-4-dependent intracellular signaling pathway.

ProTα(100–109), and the tumor antigen epitopes HER-2(9₃₆₉), tyr(9₃₆₉) (HLA-A2-restricted) [64], HER-2(15₇₇₆) and tyr(15₄₄₈) (HLA-DR4-restricted) [36, 64] were synthesized by the Fmoc (9-fluorenylmethoxycarbonyl)/tBu chemistry utilizing a multiple peptide synthesizer Syro II (MultiSynTech, Witten, Germany). Crude peptides were purified by HPLC on a reverse phase C18 Nucleosil 100-5C column (HPLC Technologies, UK) to a purity of >95%, using a linear gradient of 5.8% acetonitrile in 0.05% trifluoroacetic acid for 45 min. All peptides were characterized by matrix-assisted laser desorption ionization-time of flight mass spectrometry and results were in all cases in agreement with the calculated masses. Human recombinant proTα was purchased from Alexis Biochemicals, CA, USA and passed through an Endotoxin removal column (Pierce Biotechnology). Prior to their use, all peptides and proTα were tested for endotoxin levels using the LAL chromogenic Endotoxin Quantitation kit (Pierce Biotechnology, IL, USA) according to the manufacturer’s instructions. They were endotoxin-free.

Human T2 cells (HLA-A*0201) were cultured in RPMI 1640, supplemented with 10% heat-inactivated fetal bovine serum (FBS), 2 mM L-glutamine, 10 mM Hepes, 5 μg/mL Gentamycin, 10 U/mL Penicillin and 10 U/mL Streptomycin (all from Lonza, Cologne, Germany), at 37°C, in a humidified 5% CO₂ incubator.

Buffy coats were collected from HLA-A2+ and DR4+ healthy blood donors. Prior to blood draw, individuals gave their informed consent according to the regulations approved by the 2nd Peripheral Blood Transfusion Unit and Haemophilia Centre, ‘Laikon’ General Hospital Institutional Review Board, Athens, Greece. PBMCs were isolated by centrifugation over Ficoll-Histopaque (Lonza) density gradient, resuspended in X-VIVO 15 (Lonza) or cryopreserved in FBS-10% DMSO (Sigma-Aldrich Chemical Co., St Louis, MO, USA) for later use.

Highly enriched monocytes (>80% CD14+) were obtained from PBMCs by plastic adherence for 2 h at 37°C [65]. Non-adherent cells were removed and cryopreserved. Monocytes were cultured for 5 days in X-VIVO 15 supplemented with 800 IU/mL recombinant human granulocyte macrophage colony-stimulating factor (GM-CSF) and 500 IU/mL recombinant human IL-4 (both from R&D Systems GmbH, Wiesbaden-Nordenstadt, Germany). On day 5, iDCs were treated with LPS (0.5 μg/mL; Sigma-Aldrich), TNF-α (10 ng/mL; R&D Systems), proTα (160 ng/mL) or proαα(100–109) (25 ng/mL) for 1–48 h, concentrations already reported to induce DC maturation [23]. Mature DCs were recovered at various time points for phenotypic and TLR-4 analysis by flow cytometry and immunoblotting, or were used to stimulate autologous T cells. Supernatants from 48 h matured DCs were also collected and the concentrations of TNF-α, IL-10 and IL-12 were quantified using commercially available ELISA kits (all from Life Technologies Corporation, Carlsbad, USA), according to manufacturer’s instructions. For TLR-4 neutralization experiments, iDCs were pre-incubated in the presence of anti-TLR-4 (a-TLR-4) neutralizing monoclonal antibody (mAb; clone W7C11) or an irrelevant mouse IgG1 mAb (both from InvivoGen, San Diego, USA) at a final concentration of 10 μg/mL for 1 h and further stimulated with LPS, proTα or proTα(100–109) for 48 h. TNF-α, IL-10 and IL-12 were determined in culture supernatants.

For T cell stimulation, 48 h matured DCs (1×10⁶/mL) were pulsed with 50 μg/mL HER-2(9₃₆₉) and HER-2(15₇₇₆) for 6 h at 37°C, in a humidified 5% CO₂ incubator in X-VIVO 15. DCs were washed twice, resuspended in X-VIVO 15 and added to autologous lymphocytes (non-adherent fraction) at a DC:lymphocyte ratio of 1:10. T cells were stimulated thrice at weekly intervals and on days 3 and 5 after each stimulation, 40 IU/mL IL-2 (Proleukin; Novartis Pharmaceuticals Ltd, UK) were added to the cultures. At the third stimulation, Golgi-Plug (1 μL/mL; Becton-Dickinson (BD) Biosciences, Erembodegem, Belgium) was added in the cultures, and 12 h later, T cells were harvested and analyzed for cytokine production by flow cytometry.

For DC phenotype analysis, iDCs and mature DCs were stained for the surface molecules HLA-DR, CD80, CD83, CD86, CD11b, CD40 and CD14. Triple staining was performed using appropriate combinations of FITC-, PE- or PE-Cy5-labelled mouse anti-human IgG1 and IgG2 mAbs (BD Biosciences) at saturating concentrations for 30 min on ice. DCs were also stained with irrelevant anti-human IgG1 and IgG2 mAbs (BD Biosciences), as isotype controls. Samples were measured using a FACSCalibur flow cytometer (BD Biosciences) and data were analyzed using CellQuest software. MFI was evaluated for each marker.

The cytotoxic activity of thrice stimulated T cells was determined by standard ⁵¹Cr- release assay. T2 cells were incubated for 2 h at 37°C with 10 μg/mL HER-2(9₃₆₉) or tyr(9₃₆₉), washed and labeled with sodium chromate, as previously described [21]. Non-loaded T2 were similarly labeled for controls. Effectors (1×10⁶/mL in X-VIVO 15; 100 μL/well) were seeded in 96-well U-bottom plates (Greiner Bio-one, Kirchheim, Germany) and T2 targets were added (5×10⁴/mL; 100 μL/well), at an effector:target (E:T) ratio of 10:1. Where indicated, mAb to MHC class I molecules (W6/32, kindly donated by Prof. S. Stevanovic, University of Tübingen) was added to the cultures at a final concentration of 5 μg/mL for the entire incubation period [66]. After 18 h of coincubation at 37°C, 5% CO₂, 100 μL of supernatant were removed from each well and isotope (counts per minute (cpm)) was counted in a γ-counter (1275 Mini-gamma LKB Wallac, Turku, Finland). To determine maximal and spontaneous isotope release, targets were incubated with 3 N HCl and in plain medium, respectively. All cultures were set in triplicate. Percentage of specific cytotoxicity was calculated according to the formula: [(cpm experimental-cpm spontaneous)/(cpm maximal-cpm spontaneous)] ×100.

Stimulated T cells were seeded in 96-well U-bottom plates (1 × 10⁶/mL; 100 μL). Autologous matured DCs pulsed with 50 μg/mL HER-2(15₇₇₆) or tyr(15₄₄₈) for 6 h, were added (1 × 10⁵/mL; 100 μL/well) and cocultured for 5 days. T cells incubated with unpulsed matured DCs or in the presence of IL-2 (500 IU/mL) were used as controls. Where indicated, mAb to MHC class II molecules (L243, kindly donated by Prof. S. Stevanovic) was added to the cultures at a concentration of 5 μg/mL for the entire culture period [66]. For the last 18 h of culture, 1 μCi ³H-thymidine (Amersham Pharmacia Biotech, Amersham, Bucks, UK) was added per well and cells were harvested in a semi-automatic cell harvester (Skatron Inc., Tranby, Norway). The amount of incorporated radioactivity, proportional to DNA synthesis, was measured in a liquid scintillation counter (Wallac, Turku, Finland) and expressed as cpm. The S.I. of each experimental group was calculated using the formula: (average cpm of sample in the presence of peptide-pulsed DCs)/(average cpm of sample in the presence of unpulsed DCs).

Total cell extracts from 4–5×10⁵ iDCs and DCs matured with LPS, proTα or proTα(100–109) were extracted as described [67]. Briefly, cells were lysed in NP-40 lysis buffer (1% NP-40, 150 mM NaCl, 50 mM Tris pH 8.0) containing protease inhibitors (Protease Inhibitor Cocktail, Sigma-Aldrich) and lysates were cleared by centrifugation for 10 min at 19,000 g (4°C). The protein content of extracts was determined by the Bradford assay, samples were mixed with reducing Laemmli buffer and equal protein amounts (15–25 μg) were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis using 12% (w/v) polyacrylamide gels. Separated proteins were blotted on nitrocellulose membranes and probed with primary antibodies (goat anti-human αRIF/Novus Biologicals, Ltd, Cambridge, UK; rabbit anti-human MyD88 and rabbit anti-human TIRAP/eBioscience; rabbit anti-human GAPDH/Santa Cruz Biotechnology Inc, Santa Cruz, CA, USA) and horseradish peroxidase (HRP)-conjugated secondary antibodies (anti-rabbit-IgG and anti-goat-IgG/Santa Cruz Biotechnology). Immunoblots were developed using an enhanced chemiluminescence reagent kit (Santa Cruz Biotechnology) and quantified by scanning densitometry (Gel Analyzer v.1.0, Biosure, Athens, Greece).