Transgene IL-6 Enhances DC-Stimulated CTL Responses by Counteracting CD4+25+Foxp3+ Regulatory T Cell Suppression via IL-6-Induced Foxp3 Downregulation

Dendritic cells (DCs), the most potent antigen-presenting cells have been extensively applied in clinical trials for evaluation of antitumor immunity. However, the efficacy of DC-mediated cancer vaccines is still limited as they are unable to sufficiently break the immune tolerance. In this study, we constructed a recombinant adenoviral vector (AdVIL-6) expressing IL-6, and generated IL-6 transgene-engineered DC vaccine (DCOVA/IL-6) by transfection of murine bone marrow-derived ovalbumin (OVA)-pulsed DCs (DCOVA) with AdVIL-6. We then assessed DCOVA/IL-6-stimulated cytotoxic T-lymphocyte (CTL) responses and antitumor immunity in OVA-specific animal tumor model. We demonstrate that DCOVA/IL-6 vaccine up-regulates expression of DC maturation markers, secretes transgene-encoded IL-6, and more efficiently stimulates OVA-specific CTL responses and therapeutic immunity against OVA-expressing B16 melanoma BL6-10OVA in vivo than the control DCOVA/Null vaccine. Moreover, DCOVA/IL-6-stimulated CTL responses were relatively maintained in mice with transfer of CD4+25+Foxp3+ Tr-cells, but significantly reduced when treated with anti-IL-6 antibody. In addition, we demonstrate that IL-6 down-regulates Foxp3-expression of CD4+25+Foxp3+ Tr-cells in vitro. Taken together, our results demonstrate that AdV-mediated IL-6 transgene-engineered DC vaccine stimulates potent CTL responses and antitumor immunity by counteracting CD4+25+ Tr immunosuppression via IL-6-induced Foxp3 down-regulation. Thus, IL-6 may be a good candidate for engineering DCs for cancer immunotherapy.


Introduction
Immune surveillance by CD8 + cytotoxic T lymphocytes (CTLs) represents a major mechanism for the detection and elimination of pathogen-infected cells. CTLs are also essential for effective immunity against tumors [1]. Dendritic cells (DCs) are the most potent professional antigen-presenting cells of the immune system, uniquely capable of stimulating tumor-specific CD4 + and CD8 + T cell immune responses leading to CTL tumor infiltration and tumor regression [2,3]. DC vaccines have been extensively applied in experimental animal models and clinical trials for evaluation of antitumor immunity [4,5]. While only a proportion of the tumor immunotherapy clinical trials carried out so far have yielded positive results, those using DCs as carrier of tumor antigens have obtained the highest rates of success amongst others [6]. However, in general, the efficacy of DC-mediated cancer vaccine is still limited, mostly because DC vaccines are unable to sufficiently break the suppressive tumor microenvironment and immune tolerance in cancer patients [4].
Inflammatory cytokines such as IL-2, IL-6, IL-12, IL-15 and TNF-α play an important role in inflammation, innate and adaptive immunity [7]. To improve the efficacy of DC vaccine, DCs were genetically modified to produce IL-2 or IL-12 [8,9]. These engineered DC vaccines induced potent antitumor immunity via activation of strong CTL responses. It was also demonstrated that IL-15 transgene expression of engineered DCs increased their functional effect and survival, and became resistant to tumor-induced DC apoptosis via up-regulation of DC markers and Bcl-2, respectively [10]. We previously demonstrated that inflammatory cytokine TNF-α transgene-expressing DCs underwent augmented cellular maturation and induced more robust CTL responses and antitumor immunity [11]. However, the impact of genetically modified-DCs with IL-6 transgene in antitumor vaccine has not been studied.
In this study, we cloned murine inflammatory cytokine IL-6 gene from ConA-stimulated T cells by reverse transcription-polymerase chain reaction (RT-PCR) and constructed a recombinant adenoviral vector AdV IL-6 using the cloned IL-6 cDNA. We then generated IL-6 transgene-engineered DC (DC IL-6 ) vaccine by transfection of murine bone marrow (BM)-derived ovalbumin (OVA)-pulsed DCs (DC OVA ) with AdV IL-6 and further assessed DC OVA/IL-6 -stimulated CTL responses and antitumor immunity in an OVA-specific animal tumor model.

AdV IL-6 -Transfected DCs Upregulate Expression of I a b , CD54, CD80 and IL-6
To assess the impact of genetically modified-DCs with IL-6 transgene in antitumor vaccine, we first constructed a recombinant adenoviral vector AdV IL-6 expressing transgene IL-6 under the regulation of the cytomegalovirus (CMV) early/immediate promoter/enhancer ( Figure 1A). To assess the transcriptional IL-6 expression, RNA extracted from AdV IL-6 was subjected to reverse transcription-polymerase chain reaction (RT-PCR) analysis using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as the loading control. As shown in Figure 1B, a significant amount of IL-6 expression was found in recombinant adenovirus AdV IL-6 , but not in the control adenovirus AdV Null without any transgene insertion. We demonstrated that DC OVA expressed DC marker CD11c, adhesion molecule CD54 and DC maturation markers CD40, CD80 and I a b ( Figure 1C), and secreted little amount of IL-6 (0.03 ng/mL). We also demonstrated that both AdV IL-6 -and AdV Null -transfected DC OVA/IL-6 and DC OVA/Null up-regulated CD40, CD54, CD80, and I a b ( Figure 1C), indicating that AdV-mediated transfection enhances DC maturation. In addition, we also found that AdV Null -transfected DC OVA/Null secreted some IL-6 (0.40 ng/mL) whereas AdV IL-6 -transfected DC OVA/IL-6 secreted much more IL-6 (1.90 ng/mL), indicating that AdV transfection induces DCs to express the inflammatory cytokine IL-6.

AdV IL-6 -Transfected DCs Stimulate Potent CTL Responses
To assess DC OVA/IL-6 vaccine-stimulated CTL responses, we intravenously (i.v.) immunized C57BL/6 mice with DC OVA/IL-6 . Six days later, the amount of OVA-specific CD8 + T cells in the peripheral blood was measured using PE-labeled H-2K b /OVA 257-264 tetramer and FITC-anti-CD8 + antibody staining by flow cytometry. As illustrated in Figure 2A, the percentage of double positive (PE-tetramer + and FITC-CD8 + ) cells in the total CD8 + population is significantly higher in the DC OVA/IL-6 -immunized mice (2.79%) compared to the control DC OVA/Null -immunized mice (0.63%) (p < 0.05), with both immunized groups showing a significant difference compared to the control PBS-immunized mice (p < 0.05), indicating that DC OVA/IL-6 immunization stimulates potent OVA-specific CD8 + T cell responses.

IL-6 Induces Foxp3 down-Regulation of CD4 + 25 + Foxp3 + Tr Cells
IL-6 has been reported to inhibit the generation and counteract the immunosuppression of CD4 + 25 + Foxp3 + Tr cells [12,13]. To assess the mechanism for IL-6-induced counteraction, we cultured CD4 + 25 + Foxp3 + Tr cells in the presence or absence of IL-6. We found that IL-6-treated CD4 + 25 + Foxp3 + Tr cells down-regulated Foxp3 expression ( Figure 2B), indicating that IL-6-induced counteraction of CD4 + 25 + Foxp3 + Tr immunosuppression may be via Foxp3 down-regulation. Figure 2. DC OVA/IL6 stimulates potent CTL responses. (A) C57BL/6 mice were intravenously (i.v.) immunized with PBS, DC OVA/IL-6 and DC OVA/Null . On day six after immunization, mouse tail blood samples were stained with PE-labeled H-2Kb/OVA 257-264 tetramer (Beckman-Coulter, Mississauga, ON, Canada) and FITC-labeled anti-CD8 + antibody, followed by flow cytometric analysis. One day after CD4 + 25 + Foxp3 + Tr cells transfer, C57BL/6 mice were i.v. immunized with DC OVA/IL-6 , DC OVA/Null and the CTL responses were analyzed by flow cytometry with or without i.v. treatment of anti-IL-6 antibody (0.5 mg/mL). The value in each panel represents the percentage of OVA-specific (tetramer-positive) CD8 + T cells vs. the total CD8 + T cell population. The value in parenthesis represents the standard deviation (SD). * p < 0.05 vs. cohorts of the DC OVA/Null group and ** p < 0.05 vs. cohorts of DC OVA/IL-6 + Tr group (student t test); (B) CD4 + 25 + Foxp3 + Tr cells were incubated with IL-2 with or without IL-6 overnight. After fixation, the cell membranes were permeabilized and then stained with PE-Cy5-conjugated anti-Foxp3 + antibody followed by flow cytometric analysis; (C) In vivo cytotoxicity assay. Six days after immunization, the immunized mice were i.v. injected with a mixture of CFSE high and CFSE low -labeled splenocytes (at 1:1 ratio) that had been pulsed with OVAI and the control Mut1 peptide, respectively. After sixteen hours, spleens of immunized mice were removed and the percentages of the residual CFSE high (H) and CFSE low (L) target cells remaining in the recipients' spleens were analyzed by flow cytometry. The value in each panel represents the percentage of CFSE high vs. CFSE low target cells remaining in spleen. The value in parenthesis represents the standard deviation (SD). * p < 0.05 vs. cohorts of the DC OVA/Null group (student t test). One representative experiment of two is shown.

AdV IL-6 -Transfected DC-Stimulated CD8 + T Cells Are Effector CTLs
To analyze the differentiation of DC OVA/IL-6 -stimulated CD8 + T cells into effector CTLs, an in vivo cytotoxicity assay was performed. We adoptively i.v. transferred OVAI peptide-pulsed and strongly carboxyfluorescein diacetate succinimidyl ester (CFSE)-labeled splenocytes (CFSE high ) as the OVA-specific target cells, as well as the control peptide-pulsed and weakly CFSE-labeled splenocytes (CFSE low ) as the control non-specific target cells into recipient mice six days after immunization with DC OVA/IL-6 and DC OVA/Null . Flow cytometric analysis was performed to examine the ability of activated T cells to induce specific killing of the above target cells sixteen hours after target cell transfer. In Figure 2C, the cell killing was specifically targeted towards OVAI-pulsed CFSE high target cells, and the levels of CFSE low cells remain unaffected. Mice immunized with DC OVA/Null had a decrease of 46% OVAI-pulsed CFSE high target cells, whereas mice immunized with DC OVA/IL-6 had a significantly greater degree of loss of OVAI-pulsed CFSE high target cells (91.3%) (p < 0.05), indicating that DC OVA/IL-6 -stimulated CD8 + T cells are effector CTLs with more efficient killing activity for OVA-specific target cells.

AdV IL-6 -Transfected DCs Induce Potent Antitumor Immunity
To study whether DC OVA/IL-6 is capable of inducing therapeutic immunity against six-day-established tumor, we i.v. injected mice with the highly metastatic OVA-expressing B16 melanoma cells BL6-10 OVA (1 × 10 6 cells). Six days later, mice were i.v. immunized with DC OVA/IL-6 . Three weeks after tumor cell challenge, mice were sacrificed and numbers of lung metastatic tumor colonies were counted. As shown in Table 1, DC OVA/Null immunization only cured 50% of the mice (4/8). However, the median number of lung tumor colonies in DC OVA/Null -immunized group was 49, which is much less than that (>300) in PBS control group (p < 0.05). In contrast, DC OVA/IL-6 immunization was able to protect 100% of mice (8/8) from tumor growth, indicating that DC OVA/IL-6 vaccine stimulates potent therapeutic immunity against six-day-established B16 melanoma.

Discussion
Dendritic cells (DCs) are a subset of white blood cells that are critical to most aspects of adaptive immunity because of their central role in initiation of T-cell responses [14][15][16]. As dendritic cells (DCs) are the most potent antigen-presenting cells (APCs) [14][15][16], engineering DCs is likely to yield improved therapeutic vaccines [17] by inducing or promoting efficient antitumor immune responses in cancer patients [18]. Previous reports indicate that intratumoral injection of DCs, engineered to express a combination of different cytokines, such as IL-12, IL-21, or IFN-α, showed potent therapeutic effect against established tumors [19]. Vogt et al. [20] have reported that intratumoral injection of adenoviral vector-transfected DCs with IL-12 over-expression was crucial for effective tumor regression. Qu et al. [21] have demonstrated the therapeutic effectiveness of intratumorally delivered DCs engineered to express the pro-inflammatory cytokine IL-32.
The cytokine IL-6 secreted by many different cells, including the monocyte/macrophages, fibroblasts, endothelial cells, keratinocytes, mast cells, T cells, and DCs acts as a central regulator of inflammatory processes [22]. It plays a key role in progression from the initial innate immune responses to infection to adaptive immune responses [23]. IL-6 is involved in the maturation of B cells and development of a major proinflammatory T cell population, the pathogenic CD4 + Th17 cells [24,25]. IL-6 has been reported to inhibit the generation of and counteract the immunosuppression of CD4 + 25 + Tr cells [12,13]. We have shown that IL-6 counteracts CD4 + Th2 cell's IL-10-mediated immunosuppression [26]. These unique characteristics of IL-6 suggest that it may be a good candidate transgene to engineer DCs for the development of new DC-based vaccines capable of overcoming immunosuppression leading to potent CTL responses and antitumor immunity.
To assess the impact of genetically modified-DCs with IL-6 transgene in antitumor vaccine, we constructed the recombinant adenoviral vector AdV IL-6 expressing transgene IL-6 and the control adenoviral vector AdV Null without any transgene insertion. We have previously shown that AdV transfected DCs expressed inflammatory cytokines such as IL-1β and IL-12 [27]. In this study, we found that AdV Null -transfected DC OVA/Null secreted IL-6 (0.4 ng/mL), indicating that AdV transfection also induces DCs to express inflammatory cytokine IL-6. Furthermore, AdV IL-6 -transfected DC OVA/IL-6 secreted much more IL-6 (1.95 ng/mL), indicating that DC OVA/IL-6 cells also secrete transgene-encoded IL-6. In addition, we also demonstrated that both AdV IL-6 -and AdV Null -transfected DC OVA/IL-6 and DC OVA/Null up-regulated CD54, CD80 and I a b , indicating that AdV-mediated transfection enhances DC maturation, which is consistent with several previous reports [27][28][29]. The AdV-induced DC maturation has been shown to be linked to NF-κB-dependent [30] and PI3 kinase-mediated TNF-α induction pathway [31].
In this study, we demonstrated that DC OVA/IL-6 vaccine stimulates potent effector CTL responses and immunity against OVA-expressing B16 melanoma. The polyclonal naïve CD4 + 25 + Foxp3 + Tr cells develop in the thymus and then enter peripheral tissues where they suppress the activation of other self-reactive T cells [32]. The transcription factor Foxp3 controls regulatory T cell development [33,34]. The activation and transgene expression of Foxp3 have been reported to induce immune suppressive effects of T cells, DCs and macrophages [35,36]. It has also been shown that an elevated number of Tr cells was detected in tumors [37,38], which suppressed the antitumor immune responses by inhibition of T cell proliferation and effector function [39][40][41] as well as DC maturation [42]. Therefore, the question of how to combat immune tolerance becomes a critical challenge in cancer vaccine development [43]. In this study, we demonstrate that DC OVA/IL-6 vaccine counteracts CD4 + 25 + Foxp3 + Tr-mediated immunosuppression in mice with transfer of purified naïve CD4 + 25 + Foxp3 + Tr cells. In addition, for the first time, we demonstrate that the counteraction of CD4 + 25 + Foxp3 + Tr suppression by DC OVA/IL-6 vaccine is derived from transgene-encoded IL-6 signaling and possible via IL-6-induced Foxp3 down-regulation. OVA protein is a well-established model antigen to study anti-tumor immunity [44]. Many previous studies also have used OVA as a model antigen for tumor immunotherapeutic studies [45][46][47]. However, future study using less immunogenic tumor antigen will be interesting. Overall, our results suggest that IL-6 may overcome Tr-mediated suppression of antigen-specific T cell responses in tumor microenvironments. It has also been demonstrated that IL-6 activates in vivo T cell responses [12] and exerts anti-apoptotic activity on a wild variety of cells, including the naïve and activated T cells [48][49][50]. Therefore, the potent CTL responses and antitumor immunity induced by DC OVA/IL-6 vaccine may be derived from a combination of the above transgene-encoded IL-6-mediated stimulatory effects.

Reagents, Cell Lines and Animals
The biotin-labeled antibodies (Abs) specific for CD11c, CD40, CD54, CD80, I a b , and fluorescein isothiocyanate (FITC)-or phycoerythrin (PE)-labeled Abs specific for CD4, CD8, and CD44 were obtained from PharMingen Canada Inc. (Mississauga, ON, Canada). The anti-ovalbumin (OVA) Ab was obtained from Sigma (Oakville, ON, Canada). The PE-labeled H-2K b /OVA 257-264 tetramer was obtained from Beckman Coulter (San Diego, CA, USA). The PE-Cy5-conjugated anti mouse Foxp3 antibody was obtained from eBioscience Inc. (San Diego, CA, USA). The highly lung metastatic OVA-expressing B16 melanoma cell line BL6-10 OVA was generated in our laboratory [51]. Naïve C57BL/6 mice were obtained from Jackson Laboratory (Bar Harbor, ME, USA). All animal experiments were carried out in accordance with the Canadian Council for Animal Care guidelines.

Recombinant Adenovirus Construction
The construction of recombinant adenovirus (AdV) expressing IL-6 (AdV IL-6 ) was performed by insertion of mouse IL-6 gene cloned from ConA-stimulated T cells into pShuttle vector (Stratagene Inc., La Jolla, CA, USA) using the cloned IL-6 cDNA to form pLpA IL-6 expressing IL-6 gene [11]. The PmeI-digested shuttle vector was then co-transformed into BJ5183 E. coli cells already containing the backbone vector for homologous recombination to form the recombinant vector AdV IL-6 as described previously ( Figure 1A) [52,53]. The control AdV Null without any transgene insert was previously constructed in our laboratory ( Figure 1A) [52,53]. The recombinant AdV IL-6 vector was then linearized by PacI digestion, and then transfected into 293 cells using lipofectamine (Gibco/BRL, Burlington, ON, Canada) to generate recombinant adenovirus AdV IL-6 as described previously [52,53]. All recombinant AdVs were amplified in 293 cells and purified by cesium chloride ultracentrifugation gradients [52,53]. To assess transgene IL-6 expression, we performed RT-PCR using RNA purified from AdV IL-6 -transfected 293 cells as described previously [52,53].

Preparation of Dendritic Cells
C57BL/6 mouse bone marrow (BM)-derived dendritic cells (DCs) were prepared as described previously [1]. Briefly, BM cells from femora and tibia of naïve C57BL/6 mice were depleted of red blood cells with 0.84% Tris-ammonium chloride, and plated in DC culture medium (Dulbecco's Modified Eagle Medium (DMEM) plus 10% fetal calf serum (FCS), granulocyte macrophage colony-stimulation factor (GM-CSF) (20 ng/mL) and IL-4 (20 ng/mL)). On day three, the non-adherent granulocytes, T and B cells were gently removed and fresh media was added. After two days, the loosely adherent proliferating DC aggregates were dislodged and re-plated. On day six, DCs displaying typical morphologic features (that is, numerous dendritic processes) were further pulsed with ovalbumin (OVA) (0.5 mg/mL) for overnight and termed DC OVA .

Adenovirus (AdV) Transfection of DCs
DC OVA were transfected with AdV IL-6 expressing the transgene IL-6 and control AdV Null without any transgene insert at a multiplicity of infection (MOI) of 150 to form DC OVA/IL-6 and DC OVA/Null vaccines as previously described [52,53]. The transfected cells were then harvested for phenotypic analysis by flow cytometry. Moreover, the supernatants of DC OVA/IL-6 and DC OVA/Null were assessed for the secretion of IL-6 using the IL-6 enzyme-linked immunosorbent assay (ELISA) kit (BD Bioscience, Mississauga, ON, Canada).

Flow Cytometric Analysis
For phenotypic analysis, DC OVA , DC OVA/IL-6 , and DC OVA/Null were stained with biotin-conjugated anti-mouse antibodies (2 mg/mL) specific for major histocompatibility complex (MHC) class II (I a b ), CD40, CD54 or CD80 and the cells were analyzed by flow cytometry. For tetramer analysis, peripheral blood of immunized C57BL/6 mice [DC OVA , DC OVA/IL-6 , and DC OVA/Null (1 × 10 6 cells/mouse)] were stained with PE-labeled H-2K b /OVA 257-264 tetramer and FITC-labeled anti-CD8 antibody on day six after immunization followed by flow cytometric analysis. In another set of experiments, mice were first i.v. transferred with CD4 + 25 + Foxp3 + Tr cells (1 × 10 6 cells/mouse). One day after Tr cell transfer, mice were i.v. immunized with DC OVA/IL-6 , DC OVA/Null with or without i.v. treatment of anti-IL-6 antibody (0.5 mg/mL), and the CTL responses were analyzed six days after immunization by flow cytometry. To assess Foxp3 expression, CD4 + 25 + Foxp3 + Tr cells were incubated in culture medium with IL-2 (40 units/mL) in the presence or absence of IL-6 (40 ng/mL) for overnight. The cells were permeabilized with cytofix/cytoperm solution (BD Biosciences, San Diego, CA, USA) and then stained with PE-Cy5-conjugated anti-Foxp3 antibody followed by flow cytometric analysis.

In Vivo Cytotoxicity Assay
The in vivo cytotoxicity assay was performed as described previously [51]. Briefly splenocytes derived from naïve C57BL/6 mice were incubated with high (3.0 μM, CFSE high ) or low (0.6 μM, CFSE low ) concentrations of CFSE. CFSE high cells were further pulsed with OVAI (OVA 257-264 ) peptide (SIINFEKL), and washed extensively to remove free peptide. However, the CFSE low cells were pulsed with the control Mut peptide (FEQNTAQP) to become the internal controls. CFSE high and CFSE low target cells were co-injected i.v. at a ratio of 1:1 into the above immunized mice six days after immunization. Sixteen hours after injection, spleens were removed from the immunized mice to analyze the residual OVA-specific CFSE high and irrelevant control CFSE low target cells remaining in recipients' spleens by flow cytometry.

Animal Studies
For evaluation of therapeutic antitumor immunity, C57BL/6 mice were first challenged by i.v. injection with 1 × 10 6 OVA-expressing BL6-10 OVA tumor cells. Six days after tumor cell injection, mice were vaccinated i.v. with 1 × 10 6 engineered DC OVA/IL-6 , DC OVA/Null , and PBS, respectively. Three weeks after tumor cell challenge, mice were sacrificed, and numbers of lung metastatic tumor colonies were counted. The metastasis on freshly isolated lungs appeared as discrete black pigmented foci that can be easily distinguishable from normal lung tissues and further confirmed by histopathological examination. Tumor metastatic foci too numerous to count were assigned an arbitrary value of >300.

Statistical Analyses
Statistical analyses were performed using Student's t-test or Mann-Whitney U test to compare variables from different groups [54]. A value of p < 0.05 is considered significant.

Conclusions
The results of our study demonstrate that AdV-mediated IL-6 transgene-engineered DC vaccine stimulates potent CTL responses and antitumor immunity by counteracting CD4 + 25 + Tr immunosuppression via IL-6-induced Foxp3 down-regulation. Thus, IL-6 may be a good candidate for engineering DCs for effective cancer immunotherapy.