ORIGINAL RESEARCH published: 07 February 2019 doi: 10.3389/fimmu.2019.00157 Ethyl Pyruvate Induces Tolerogenic Dendritic Cells Neda Djedovic 1 , María José Mansilla 2,3 , Bojan Jevtić 1 , Juan Navarro-Barriuso 2,3 , Tamara Saksida 1 , Eva M. Martínez-Cáceres 2,3 and D – orde – Miljković 1* 1 Department of Immunology, Institute for Biological Research “Siniša Stanković,” University of Belgrade, Belgrade, Serbia, 2 Immunology Division, Germans Trias i Pujol University Hospital and Research Institute, Badalona, Spain, 3 Department of Cellular Biology, Physiology, and Immunology, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, Spain Dendritic cells (DC) are professional antigen presenting cells that have a key role in shaping the immune response. Tolerogenic DC (tolDC) have immuno-regulatory properties and they are a promising prospective therapy for multiple sclerosis and other autoimmune diseases. Ethyl pyruvate (EP) is a redox analog of dimethyl fumarate (Tecfidera), a drug for multiple sclerosis treatment. We have recently shown that EP ameliorates experimental autoimmune encephalomyelitis, a multiple sclerosis murine model. Here, we expanded our study to its tolerogenic effects on DC. Phenotypic analysis has shown that DC obtained from mice or humans reduce expression of molecules required for T cell activation such as CD86, CD83, and HLA-DR under the influence of EP, while CD11c expression and viability of DC are not affected. Furthermore, EP-treated DC Edited by: Herman Waldmann, restrain proliferation and modulate cytokine production of allogeneic lymphocytes. These University of Oxford, United Kingdom results demonstrate that EP has the ability to direct DC toward tolDC. Reviewed by: Keywords: dendritic cells, ethyl pyruvate, autoimmunity, tolerogenicity, immune-regulation Tatjana Nikolic, Leiden University, Netherlands Ursula Grohmann, University of Perugia, Italy INTRODUCTION *Correspondence: Tolerogenic dendritic cells (tolDC) are promising candidates for the cell-based immunotherapy – orde D – Miljković georgije_zw@yahoo.com of autoimmune disorders, including multiple sclerosis (1). They are effective in restraining antigen-specific and allogeneic T cell responses in vitro and their beneficial effects have been Specialty section: demonstrated in the treatment of animal models of various autoimmune disorders (2). Moreover, This article was submitted to their administration to humans has been shown safe and efficient by increasing the proportion Immunological Tolerance and of regulatory T cells in circulation (1). Vitamin D3 and dexamethasone are commonly used for Regulation, induction of tolDC (1, 2), while a number of agents has been shown effective for the induction of a section of the journal tolerogenic properties over dendritic cells (DC). Ethyl pyruvate (EP) is a redox active compound Frontiers in Immunology that has been shown potent as an anti-inflammatory agent (3). It is a safe and simple chemical Received: 27 September 2018 that has already been tested in humans (4). Importantly, it is a redox analog of dimethyl fumarate Accepted: 17 January 2019 (Tecfidera), a drug that is approved for multiple sclerosis treatment (5). Our group has recently Published: 07 February 2019 reported that EP ameliorates experimental autoimmune encephalomyelitis (EAE), an animal model Citation: of multiple sclerosis (6). The major pathogenic T helper (Th) cells in the central nervous system Djedovic N, Mansilla MJ, Jevtić B, autoimmunity are interferon (IFN)-γ-producing Th1 cells and interleukin (IL)-17-producing Th17 Navarro-Barriuso J, Saksida T, cells (7). The beneficial effects of EP in EAE were paralleled with down-regulation of Th1/Th17 Martínez-Cáceres EM and Miljković D– (2019) Ethyl Pyruvate Induces activity (6). Moreover, release/production of IL-6, tumor necrosis factor (TNF) and reactive Tolerogenic Dendritic Cells. nitrogen and oxygen species by macrophages were also inhibited by EP. Noteworthy, IL-6 is known Front. Immunol. 10:157. to potentiate the resistance of effector T cells to regulatory T cells in multiple sclerosis (8), while doi: 10.3389/fimmu.2019.00157 TNF actively contributes to demyelination and axonal degeneration in neuroinflammation (9). Frontiers in Immunology | www.frontiersin.org 1 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate Correspondingly, reactive oxygen and nitrogen species the presence of granulocyte-macrophage colony-stimulating participate in the loss of oligodendrocytes, blood-brain barrier factor (GM-CSF: 1,000 U/ml; Peprotech, Freiburg, Germany) dysfunction, T cell infiltration, and neurodegeneration (10). and interleukin 4 (IL-4: 1,000 U/ml; Peprotech). Cells were Effects of EP were also observed within the CNS, where reactivity replenished on day 4 with fresh medium and cytokines. To of microglia and astrocytes was reduced (11). induce mature DC (mDC), DC were treated with a cocktail of We were also able to demonstrate that EP down-regulated the TNF (1,000 U/ mL), IL-1β (10 ng/mL) (both from Peprotech); expression of antigen presenting molecules on macrophages (6) and PGE2 (1 µM) (Pfizer, New York, NY, United States) on which led us to the investigation of the effects of EP on DC as the day 4. Treatment with Vitamin D3 (vitDC, 1 nM, Calcijex, Kern major professional antigen-presenting cells. Here, we present that Pharma, Terrassa, Spain) was performed on days 0 and 4, while EP exerts potent tolerogenic effect on murine and human DC. It treatment with EP (EPDC) was performed on days 2 and 4. down-regulates the expression of antigen-presenting molecules These cells were also stimulated as mature DCs at day 4 with the on DC, restricts the production of pro-inflammatory cytokines cytokine cocktail. Immature DC (iDC) were not treated with the in DC and diminishes their T cell-activating function. maturation cocktail on day 4. On day 6, DC were harvested and washed extensively twice, before phenotype was determined and functional assays were performed. MATERIALS AND METHODS Monocyte-Derived Human DC Bone Marrow-Derived Murine DC Buffy coats, provided by the Banc de Sang i Teixits (Barcelona, Murine DC were obtained from progenitor bone marrow cells Spain), were obtained from randomized healthy blood donors, that were flushed from femur of NOD and C57BL/6 mice following the institutional Standard Operating Procedures (experiments were approved by the local ethics committee for blood donation and processing. Peripheral blood was of the Institute for Biological Research “Sinisa Stankovic,” in ◦ obtained from untreated relapsing-remitting multiple sclerosis accordance with Directive 2010/63/EU, N 02-09/16). These patients. The Ethical Committee of Germans Trias i Pujol cells were cultured in RPMI 1640 (Biological Industries, Hospital approved the study, and all subjects gave their Kibbutz Beit-Haemek, Israel) supplemented with 10% FCS (PAA informed consent according to the Declaration of Helsinki Laboratories), 2 mM glutamine and 1 mM sodium pyruvate (both (BMJ 1991; 302: 1994). Peripheral Blood Mononuclear Cells from Sigma-Aldrich) (1 × 106 /mL/well in 24-well plate). Bone (PBMC) were isolated by Ficoll-Paque (Lymphoprep, Axis marrow derived dendritic cells (BMDC) were cultivated for Shield, Oslo, Norway) density gradient centrifugation at 400 8 days in the presence of 20 ng/mL of GM-CSF (Peprotech × g for 30 min. Recovered cells were washed twice in PBS or Novus, Littleton, CO), with 100 ng/mL lipopolysaccharide and counted using Perfect Count microspheres (Cytognos SL, (LPS, Sigma-Aldrich) added for the last 24 h of cultivation for Salamanca, Spain) following the manufacturer’s instructions. maturation. Treatment with EP was performed on days 3 and 6 Establishing Monocyte-derived DCs, PBMCs were first depleted (diffEP) or simultaneously with LPS (matEP). Number of viable of CD3+ T cells using the RosetteSepTM Human CD3 Depletion cells was determined by trypan blue exclusion test on a LUNA- Cocktail (StemCell Technologies, Seattle, WA, United States). IITM Automated Cell Counter from Logos Biosystems (Gyeonggi- Afterwards, monocytes were obtained by positive selection using do, South Korea). the EasySep R Human CD14 Positive Selection Kit (StemCell Technologies). For all samples, the purity and viability of the In vivo DC Application monocyte populations were >95 and 90%, respectively, as Experiments were approved by the local ethics committee assessed by the expression of specific markers and Annexin V (Institute for Biological Research “Sinisa Stankovic,” in ◦ and 7-Amino-actinomycin D (7AAD) labeling (BD Biosciences). accordance with Directive 2010/63/EU, N 03-01/17). Murine Monocytes were cultured at 1 × 106 /ml for 6 days in X- DC were prepared as described above. mDC or EPDC were VIVO 15 culture medium (BioWhittaker R , Lonza, Belgium) injected subcutaneously into the hind paw of female 2–3 months supplemented with 2% (vol/vol) heat inactivated AB human old C57BL/6 mice (1 × 106 /80 µl/mouse). Complete Freund’s serum (BioWhittaker R , Lonza, Belgium), 2 mM L-glutamine adjuvant (CFA) was made from incomplete Freund’s adjuvant (Sigma-Aldrich Company LTD, Saint Louis, MO, United States), (Difco, Detroit, MI) supplemented with M. tuberculosis (to 5 100 U/mL penicillin (Cepa S.L, Madrid, Spain), and 100 µg/mL mg/ml, Difco). Each mouse was injected subcutaneously in streptomycin (Laboratorios Normon S.A, Madrid, Spain) in the hind paw with 50 µl of emulsion made from CFA mixed with equal volume of phosphate buffer saline on the following day. Popliteal lymph nodes were isolated from the mice 3 days Abbreviations: 7AAD, 7-Amino-actinomycin D; BMDC, bone marrow derived dendritic; CD, cluster of differentiation; CFA, complete Freund’s adjuvant; CNS, later and subjected to collagenase V (Sigma-Aldrich) digestion. central nervous system; DC, dendritic cells; diffEP, DC trated with EP during Lymph nodes were gently minced with scissors and incubated differentiation/propagation; EAE, experimental autoimmune encephalomyelitis; with 1 mg/mL of Collagenase V solution in RPMI 1640 at 37◦ C EP, ethyl pyruvate; FCS, fetal calf serum; GM-CSF, granulocyte-macrophage with gentle shaking for 20 min. Subsequently, the digestion was colony-stimulating factor; HLA, human leukocyte antigen; iDC, immature DC; stopped with FCS and the cells were pelleted by centrifugation. IFN, interferon; IL, interleukin; LPS, lipopolysaccharide; matEP, DC treated with EP during maturation; mDC, mature DC; PBMC, peripheral blood mononuclear Number of viable cells was determined by trypan blue exclusion cells; PBS, phosphate buffer saline; SD, standard deviation; Th, helper T cells; TNF, test on a LUNA-IITM Automated Cell Counter. For cell tracking, tumor necrosis factor; tolDC, tolerogenic DC; Treg, regulatory T cells. DC were stained with CFSE (2 µM, Invitrogen, Carlsbad, CA, Frontiers in Immunology | www.frontiersin.org 2 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate United States) prior to injection. Determination of CFSE+ cells cytokines with the following antibodies: anti-mouse antibodies among popliteal lymph node cells was performed on a CyFlow against IL-17 or IFN-γ or IL-10 coupled with FITC or PE (all Space flow cytometer (Partec, Munster, Germany). from eBioscience). After washing, acquisition was performed with a CyFlow Space flow cytometer. Samples were gated using CD4+ T Cell Isolation forward (FSC) and side (SSC) scatter to exclude dead cells and CD4+ T cells were purified from cervical lymph nodes obtained debris, as well as using FSC and FSC-W to exclude cell doublets. from BALB/c mice. For purification of CD4+ T cells from the Isotype-matched controls were included in all experiments lymph nodes, a biotin-conjugated antibody specific for CD4 (eBioscience). (Invitrogen) and IMagSAv Particles Plus (BD Biosciences, San Diego, CA) were used in accordance with the manufacturers’ ELISA instructions. CD4+ T cells were stained with CFSE (2 µM) prior Cell culture supernatants were obtained and centrifuged to spin to cultivation with DC. down the cells. Cell-free supernatants were used in sandwich ELISAs as instructed by the producers of the antibody pairs used. Allostimulatory Assays Samples were analyzed in duplicates for murine TNF, murine IL- Human DC were co-cultured with allogeneic human PBMC 6 (R&D Systems), murine IL-1β, murine IL-12 (eBiosciences), (105 cells/ well) in 96-well round-bottom plates. PBMC murine IL-17 (eBioscience), human IFN-γ and human IL-10 were co-cultured for 4 days (96 h) with MDDCs at a 1:20 (R&D Systems). Lower limit of detection was 30 pg/ml, whereas ratio (DC: PBMC). Cell proliferation was determined by upper limit of detection was 10 ng/ml for all of the ELISA tests incorporation of 1 µCi [3 H]-thymidine (PerkinElmer, Waltham, performed. Samples that showed values over the upper limit MA, United States) for 18 h on each well. Murine DC obtained of detection were adequately diluted for the measurement. The from C57BL/6 mice were co-cultured with CFSE-labeled CD4+ results were calculated using standard curves made on the basis of T cells obtained from BALB/c mice (105 cells/well) in 96-well known concentrations of the appropriate recombinant cytokines. round-bottom plates. CD4+ T cells were co-cultured for 5 days (120 h) with MDDCs at a 1:20 ratio (DC: CD4+ T cells). Cell Statistics proliferation was determined by the sequential loss of CFSE One-way ANOVA followed by Tukey’s multiple comparison fluorescence, as detected by flow cytometry. test or Student’s t-test (two-tailed) were used as appropriate for statistical analysis. A p < 0.05 was considered statistically Immunostaining and Flow Cytometry significant. Human DC were washed, re-suspended in 50 µl of PBS and incubated with mAbs for 20 min protected from light at room RESULTS temperature (RT). After washing, acquisition was performed on a FACSCanto II flow cytometer using FACSDiva software EP Exerts Tolerogenic Effects on Mouse (BD Biosciences, CA, United States). Subsequent analyses were BMDC done using FlowJo software (Tree Star, Inc., OR, United States). In order to determine if EP has tolerizing effects on NOD mice Samples were gated using forward (FSC) and side (SSC) scatter BMDC in the course of their maturation, the agent was applied to exclude dead cells and debris. The following human mAbs during the last 24 h of their cultivation, simultaneously with were used: FITC-labeled mAbs: CD86 (BD Biosciences); PE- the maturing agent, i.e., LPS. Alternatively, to assess EP effect labeled mAbs: CD14 (ImmunoTools GmbH, Germany), CD40 on differentiation/propagation of DC from their bone marrow (BD Biosciences); PE-Cyanine dye 7-labeled mAb: CD14 and precursors, the agent was applied to BMDC cultures on days CD11c (BD Biosciences); Allophycocyanin (APC)-labeled mAbs: 3 and 6 of their cultivation. EP applied during the maturation CD83, APC-H7-labeled mAb: HLA-DR (BD Biosciences). had minimal effects on BMDC viability (Figure 1A), while no Murine cells were washed, re-suspended in 100 µL of PBS effects on the cell viability were observed with the alternative supplemented with 2% of mouse serum and incubated with mAbs application of the agent (Figure 1B). While EP applied during the for 30 min at 4◦ C. After washing, acquisition was performed differentiation reduced IL-1β production in BMDC, EP applied with a CyFlow Space flow cytometer. Samples were gated using during maturation had the opposite effect (Figure 1C). Both ways forward (FSC) and side (SSC) scatter to exclude dead cells of EP application led to the reduction of generation of IL-6, IL- and debris, as well as using FSC and FSC-W to exclude cell 12, and TNF in BMDC (Figures 1D–F). Also, CD11c expression doublets. The following murine mAbs were used: PE-Cyanine5 was not affected, while both CD86 and CD40 expression were labeled CD11c and CD86 (eBioscience), FITC-labeled CD40 decreased in EP-treated BMDC, irrespectively of the way of its (eBioscience). application (Figures 1G–J). Prior to intracellular cytokine staining, cells were stimulated with eBioscienceTM Cell Stimulation Cocktail (plus protein EP-Treated Mouse BMDC Are Inefficient transport inhibitors), containing phorbol myristate acetate, Allogeneic Stimulators in vitro ionomycin and brefeldin A and monensin for 4 h, stained To test if the tolerogenic effects of EP can also be observed with anti-CD4 PerCP-Cy5.5 antibody (eBioscience), fixed and in C57BL/6 BMDC, the same application protocols were used permeabilized with eBioscienceTM Intracellular Fixation and as with NOD mice BMDC. EP did not affect cell viability and Permeabilization Buffer Set and then stained for the intracellular CD11c expression, while it reduced generation of the examined Frontiers in Immunology | www.frontiersin.org 3 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate pro-inflammatory cytokines and expression of CD40 and CD86, irrespectively of the way of its application (Figures 2A–J). To examine if these effects of EP affected BMDC ability to activate T cells, EP-treated BMDC were co-cultured with allogeneic CD4+ T cells from BALB/c mice. Proliferation of CD4+ T cells (Figures 3A,B) and their capacity to produce IL-17 (Figure 3C) were reduced in co-cultures with EP-treated BMDC in comparison to those with the untreated BMDC. EP Exerts Tolerogenic Effects on Human MDDC EP was added to MDDC cultures in the process of their differentiation/propagation (EPDC) (Figure 4A). No effect of EP on cell viability was observed on human MDDC that were obtained from healthy subjects (Figure 4B) or individuals suffering from multiple sclerosis (Figure 4C). EP-treated DC were compared to vitamin D-treated DC (vitDC), non-treated mature DC (mDC) and immature DC (iDC) in the following experiments. There was no difference in the cell viability among the examined DC populations (Figures 4D,E). Also, no difference was observed in the level of CD11c expression (Figures 4F–H). However, both vitDC and EPDC reduced expression of HLA-DR, CD86, and CD83 in comparison to mDC, both in healthy and multiple sclerosis individuals (Figures 4F–H). EP-Treated Human MDDC Are Inefficient Allogeneic Stimulators in vitro Tolerogenic properties of EPDC were examined in allogeneic co- culture with PBMC. EPDC inhibited allogeneic cell proliferation in comparison to mDC to the similar extent as vitDC (Figures 5A,B). Effects of EPDC obtained with the highest EP dose applied (12.5 mM) were superior to those of vitDC (Figures 5A,B). There was no difference in the inhibitory activity of EPDC or vitDC obtained from healthy subjects or individuals affected with multiple sclerosis. Levels of IFN-γ and IL-10 in supernatants of the co-cultures were also determined. IFN-γ levels were reduced in the samples obtained from co- cultures performed with EPDC or vitDC in comparison to mDC (Figures 5C,D), irrespectively if MDDC were obtained from healthy individuals or the patients. However, IL-10 levels were higher with EPDC or vitDC in comparison to mDC, only if MDDC were obtained from healthy subjects (Figures 5C,D). Allogeneic cell proliferation and the cytokine levels determined with DC obtained from individual persons are presented, as well (Figures 5E,F) in order to enable discrimination of responses from different donors. FIGURE 1 | Effects of EP on NOD mice DC. BMDC were obtained from NOD mice. They were treated with EP for 24 h simultaneously with LPS (A and EP-Treated Mouse BMDC Show “maturation”) or at days 3 and 6 of the propagation (B and “differentiation”). 3.1 mM EP was used in (C–I). Viability (A,B) was determined by trypan blue Immunomodulatory Function in vivo exclusion assay. Cytokines were determined in cell culture supernatants by EPDC were applied in vivo to mice immunized with CFA as ELISA (C–F). Proportion of cells expressing the listed antigens was determined depicted in Figure 6A. Their effects were compared to those of by cytofluorimetry (G–J). Data are presented as mean + SD from at least three mDC-treated mice, as well as to mice injected with vehicle only independent experiments. Data in (J) are representative plots. *p < 0.05, EP vs. 0, one way ANOVA followed by Tukey’s multiple comparisons test (A,B), (control). In order to track DC in vivo, the cells were labeled two-tailed t-test (C–I). with CFSE. CFSE-labeled EPDC and mDC were detectable in popliteal lymph nodes 4 days after the injection (Figure 6B). Frontiers in Immunology | www.frontiersin.org 4 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate FIGURE 2 | Effects of EP on C57BL/6 mice DC. BMDC were obtained from C57BL/6 mice. They were treated with EP (3.1 mM) for 24 h simultaneously with LPS (“matEP”) or at days 3 and 6 of the propagation (“difEP”) or they were untreated with EP (mDC), as schematized (A). Viability (B) was determined by trypan blue exclusion assay. Cytokines were determined in cell culture supernatants by ELISA (C–F). Proportion of cells expressing the listed antigens was determined by cytofluorimetry (G–J). Data are presented as mean + SD from at least three independent experiments. Data in (J) are representative plots. *p < 0.05, EP vs. mDC, one way ANOVA followed by Tukey’s multiple comparisons test. Popliteal lymph node cells were isolated for phenotypic analyses, nodes (Figure 6F). Both, IL-10+ CD4+ cells and IFN-γ+ CD4+ at the same time. There were no differences in cellularity and cells were also more numerous in mDC-treated than in control proportion of CD4+ cells of popliteal lymph nodes isolated mice popliteal lymph nodes (Figures 6E,F). There were no from vehicle-, mDC-, and EPDC-treated mice (Figures 6C,D). differences in proportion of IL-17+ CD4+ cells among the groups However, there were more IL-10+ CD4+ cells in EPDC-treated (Figure 6G). Importantly, ratio of IL-10+ CD4+ cells to IFN- than in control mice (Figure 6E) and less IFN-γ+ CD4+ cells γ+ CD4+ cells, as well as to IL-17+ CD4+ cells was higher in in EPDC-treated than in mDC-treated mice popliteal lymph popliteal lymph nodes of EPDC-treated mice in comparison to Frontiers in Immunology | www.frontiersin.org 5 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate FIGURE 3 | Effects of murine EP-treated DC on allogeneic reactivity. BMDC were obtained from C57BL/6 mice. Their maturation was induced by LPS in the absence (mDC) or presence of 3.1 mM EP (diffEP). They were co-cultured with CD4+ T cells purified from BALB/C mice lymph nodes. Cell proliferation was determined by CFSE assay (A,B). IL-17 levels were determined in cell culture supernatants by ELISA (C). Data are presented as mean + SD from at least three independent experiments. Data in (B) are representative plots. *p < 0.05, diffEP vs. mDC, one way ANOVA followed by Tukey’s multiple comparisons test. controls (Figures 6H,I). Moreover, ratio of IL-10+ CD4+ cells with the cells obtained from healthy individuals, as well as with to IFN-γ+ CD4+ cells was higher in popliteal lymph nodes of those obtained from individuals affected by multiple sclerosis. EPDC-treated mice in comparison to mDC-treated mice. This implies that EP-treated DC are good candidates for a tolerogenic cell-based therapy. Along this conclusion, EP-treated DISCUSSION DC were comparable to vitamin D3-treated DC, regarding their immunomodulating properties. The highest dose of EP applied Ethyl pyruvate modulates expression of antigen presentation- in our study (12.5 mM) was even superior to standard vitamin related molecules on DC. It also reduces release of pro- D3 dose used (1 nM) when T cell proliferation was determined. inflammatory cytokines from these cells. Moreover, it makes Having in mind that vitamin D3-induced tolDC have been DC inefficient in T cell activation. Importantly, the effects investigated in details in vitro and in vivo in EAE and that they are are observed when EP is applied to DC during their currently tested in clinical trials in multiple sclerosis, the ability differentiation/propagation from precursor cells, as well as when of EP to act on DC in similar fashion is an additional impetus for it is applied simultaneously with the maturation stimulus. EP- further studies toward clinical application of EP-treated DC as a treated DC modulate immune response initiated by CFA in vivo. therapy for autoimmunity. Thus, EP efficiently modulates immune activity of DC. Interestingly, there was a difference in IL-10 production in Our experiments on mice were performed with two mice the co-cultures of tolDC with PBMC, depending on the source strains, i.e., NOD and C57BL/6 mice. The reason to use both of DC, i.e., whether the cells were obtained from healthy or mice strains are our plans to expand the research to in vivo multiple sclerosis individuals. While IL-10 levels were increased systems, where EAE and type 1 diabetes in mice will be analyzed. in the co-cultures of tolDC from healthy individuals, they were Both mouse strains are susceptible to EAE (12) induction, NOD not increased in the co-cultures of tolDC from multiple sclerosis mice develop diabetes spontaneously (13), while C57BL/6 mice individuals. This discrepancy was observed irrespectively of are susceptible to multiple low dose streptozotocin-induced the tolerizing agent used (EP or vitamin D3). It is proposed diabetes (14). Also, in this way we were able to exclude strain- that IL-10 plays the crucial role in immunomodulatory activity specific effects of EP. The absence of genetic predisposition of tolDC, as it induces differentiation of regulatory T cells toward susceptibility of DC to EP were even more convincingly (15). Thus, the observed phenomenon is worthy of further shown with human cells. There, the tolerizing effects of EP on investigation, as it could be potentially disadvantageous to DC were observed with the cells obtained from more than 10 therapeutic potential of EP- and vitamin D3-induced tolDC. different individuals. What is more, the effects were consistent Also, the possibility of inherent reduced capacity of DC and Frontiers in Immunology | www.frontiersin.org 6 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate FIGURE 4 | Effects of EP on human DC. MDDC were propagated from peripheral blood monocytes and matured in the presence of TNF+IL-1β+PGE2 (mDC, vitDC, tEPDC) or were left immature without the treatment (iDC) as depicted (A). MDDC were obtained from healthy subjects (B,D,F,H) or from individuals affected by multiple sclerosis (C,E,G). EP was applied in various concentrations (B,C) or in concentration of 6.2 mM (D–H, EPDC). Vitamin D3 (vitDC) was applied in concentration of 1 nM. Viability (B–E) was determined by 7AAD test. Proportion of cells expressing the listed antigens was determined by cytofluorimetry (F–H). Data are presented as mean + SD from at least five individuals. Data in H are representative plots. *p < 0.05 vs. mDC, two-tailed t-test. Frontiers in Immunology | www.frontiersin.org 7 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate FIGURE 5 | Effects of human EP-treated DC on allogeneic reactivity. MDDC were obtained from healthy subjects (A,C,E) or from individuals affected by multiple sclerosis (B,D,F). The cells matured in the presence of TNF+IL-1β+PGE2 (mDC, vitDC, EPDC) or were left immature without the treatment (iDC). EP was applied in various concentrations (A,B) or in concentration of 3.1 mM (C–F). Vitamin D3 (vitDC) was applied in concentration of 1 nM. MDDC were co-cultured with allogeneic PBMC. Cell proliferation was determined by 3 H incorporation test (A,B,E,F). Cytokines were determined in cell culture supernatants by ELISA (C–F). Data in (A,B) are presented as mean ± SD from a number of individuals depicted as the open circles. Data in (C,D) are presented as mean + SD from four individuals. Data from four individuals are presented in (E,F), where the same person has a unique symbol (circle, square, triangle, or trapezoid). *p < 0.05, vs. mDC in (C,D), one way ANOVA followed by Tukey’s multiple comparisons test. other immune cells of multiple sclerosis patients to produce as between these two groups. Therefore, further studies enrolling IL-10 warrants additional studies as it might be important higher number of individual samples for the exploration of the for understanding the predisposition of humans toward the effects of EP on human DC, including their ability to generate development of multiple sclerosis. Indeed, reduction in the IL-10 are needed. However, the fact that the samples were not proportion of IL-10-producing PBMC in multiple sclerosis adjusted between multiple sclerosis and healthy subjects did not patients (16), as well as of IL-10 mRNA expression in these cells jeopardize the general conclusion about tolerizing potency of EP- (17, 18) were previously reported. However, multiple sclerosis treated DC, i.e., EP-treated DC had the same, if not stronger patients and healthy subjects from whom blood samples were tolerizing effects in comparison to classical tolDC induced by obtained were not adjusted to age, sex, and other relevant vitamin D3, irrespectively if they were obtained from the healthy parameters that could bias our observation on the differential IL- subjects or the patients. 10 response. Also, higher number of samples would be needed EP-treated DC applied to mice immunized with CFA shifted for the determination of the statistical correlation between the draining lymph node cytokine milieu from IFN-γ/IL-17- toward levels of cell proliferation and IL-10 or IFN-γ production within IL-10-dominated. EP-treated DC injected at the site of injection groups of healthy subjects and multiple sclerosis patients, as well efficiently migrated toward regional lymph nodes and imposed Frontiers in Immunology | www.frontiersin.org 8 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate FIGURE 6 | Effects of murine EP-treated DC on CFA-induced immune response in vivo. Schematic representation of C57BL/6 mice treatment (A). Number of CFSE+ cells (B), total number of cells (C), and proportion of CD4+ cells (D) was determined in popliteal lymph nodes by flow cytometry (B,D) or cell counting (C). Proportion of IL-10+ (E), IFN-γ+ (F), and IL-17+ (G) cells among CD4+ cells was determined by flow cytometry. Ratio of the cytokines expressing CD4+ cells is presented, as well (H,I). Data from three experiments are presented as mean ± SD from a number of mice depicted as the circles, squares, or triangles. *p < 0.05, one way ANOVA followed by Tukey’s multiple comparisons test. Frontiers in Immunology | www.frontiersin.org 9 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate the immunomodulatory effect. The observed modulation of the for clinical trials of multiple sclerosis and other autoimmune cytokine-generating T cell balance is particularly fascinating, disorders. as it was observed in strong inflammatory setting induced by CFA. Thus, immunomodulatory effects of EP-treated DC are AUTHOR CONTRIBUTIONS rather potent and the one can expect that the effects would persist in antigen-specific setting, as well. Of course, in the ND performed the main work, analyzed, interpreted, and context of a certain autoimmune disease it would be important critically revised the data. MM performed the work related to to determine if EP-treated DC loaded with relevant antigens MDDC. BJ organ isolation, cell preparation, cytofluorimetry. would acquire specificity of immunomodulatory actions. Thus, JN-B performed the work related to MDDC. TS organ our results imply that EP-treated DC dampen initiation and isolation, co-culture, CD4+ T cell purification. EM-C conception propagation of an immune response in lymphoid organs, an and design, supervised the work, and critically revised the effect that should be useful in restricting the immunopathology of manuscript. ÐM conception and design, supervised the work, autoimmune and chronic inflammatory diseases. This possibility analyzed and interpreted the data, drafted the work, and critically will be explored in details in specific models of autoimmunity, revised the manuscript. All authors finally approved the version such as type 1 diabetes in NOD mice and EAE in our future to be published and agreed to be accountable for all aspects of studies. the work in ensuring that questions related to the accuracy or We have previously shown that EP exerts immunomodulatory integrity of any part of the work are appropriately investigated properties in animal models of multiple sclerosis and type and resolved. 1 diabetes (6, 19). Here we extend our observation and we introduce a novel biological effect of EP, i.e., its ability to induce FUNDING tolerogenic DC in vitro. Moreover, EP-treated DC are efficient in modulating an immune response in vivo. EP applied to DC This work was supported by the Ministry of Education, during the process of differentiation inhibits their response to Science and Technological Development of the Republic the maturation stimulus, both regarding expression of antigen- of Serbia (173035 and 173013) and by project PI14/01175 presenting and co-stimulatory molecules and production of and PI17/01521, integrated in the Plan Nacional de I+D+I relevant pro-inflammatory cytokines. Also, EP-treated DC are and co-supported by the ISCIII-Subdirección General de impotent allogeneic stimulators of CD4+ T cells. Importantly, Evaluación and the Fondo Europeo de Desarrollo Regional the effects are observed in human cells and not only murine (FEDER). This work has been supported by funding of cells. Further studies are needed for determination of molecular Short Term Scientific Missions and by positive discussion mechanisms behind the effects of EP. Also, in vivo studies on through A FACTT network (Cost Action BM1305: www.afactt. animal models of autoimmunity are necessary. These studies eu). COST is supported by the EU Framework Programme should help to elucidate if EP-treated DC are good candidates Horizon 2020. REFERENCES 8. Schneider A, Long SA, Cerosaletti K, Ni CT, Samuels P, Kita M, et al. In active relapsing-remitting multiple sclerosis, effector T cell resistance to adaptive 1. Phillips BE, Garciafigueroa Y, Trucco M, Giannoukakis N. Clinical T(regs) involves IL-6-mediated signaling. Sci Transl Med. (2013) 5:170ra15. tolerogenic dendritic cells: exploring therapeutic impact on human doi: 10.1126/scitranslmed.3004970. autoimmune disease. Front Immunol. (2017) 8:1279. doi: 10.3389/fimmu.2017 9. Brambilla R, Ashbaugh JJ, Magliozzi R, Dellarole A, Karmally S, .01279 Szymkowski DE, et al. Inhibition of soluble tumour necrosis factor is 2. Stojanovic I, Dimitrijevic M, Vives-Pi M, Mansilla MJ, Pujol-Autonell therapeutic in experimental autoimmune encephalomyelitis and promotes I, Rodríguez-Fernandez S, et al. Cell-based tolerogenic therapy axon preservation and remyelination. Brain. (2011) 134(Pt 9):2736–54. experience from animal models of multiple sclerosis type 1 diabetes doi: 10.1093/brain/awr199 and rheumatoid arthritis. Curr Pharm Des. (2017) 23:2623–43. 10. Miljković D, Spasojević I. Multiple sclerosis: molecular mechanisms and doi: 10.2174/1381612823666170214120708 therapeutic opportunities. Antioxid Redox Signal. (2013) 19:2286–334. 3. Kao KK, Fink MP. The biochemical basis for the anti-inflammatory and doi: 10.1089/ars.2012.5068 cytoprotective actions of ethyl pyruvate and related compounds. Biochem 11. Djedović N, Stanisavljevic S, Jevtić B, Momčilović M, Lavrnja I, Miljković Pharmacol. (2010) 80:151–9. doi: 10.1016/j.bcp.2010.03.007 D. Anti-encephalitogenic effects of ethyl pyruvate are reflected in the 4. Fink MP. Ethyl pyruvate. Curr Opin Anaesthesiol. (2008) 21:160–7. central nervous system and the gut. Biomed Pharmacother. (2017) 96:78–85. doi: 10.1097/ACO.0b013e3282f63c2e doi: 10.1016/j.biopha.2017.09.110 5. Xu Z, Zhang F, Sun F, Gu K, Dong S, He D. Dimethyl fumarate for 12. Bernard CC, Johns TG, Slavin A, Ichikawa M, Ewing C, Liu J, et al. Myelin multiple sclerosis. Cochrane Database Syst Rev. (2015) CD011076. oligodendrocyte glycoprotein: a novel candidate autoantigen in multiple doi: 10.1002/14651858.CD011076.pub2 sclerosis. J Mol Med. (1997) 75:77–88. doi: 10.1007/s001090050092 6. Miljković D, BlaŽevski J, Petković F, Djedović N, Momčilović M, Stanisavljević 13. Tochino Y. The NOD mouse as a model of type I, and diabetes. Crit Rev S, et al. A comparative analysis of multiple sclerosis-relevant anti- Immunol. (1987) 8:49–81. inflammatory properties of ethyl pyruvate and dimethyl fumarate. J Immunol. 14. Like AA, Rossini AA. Streptozotocin-induced pancreatic insulitis: new model (2015) 194:2493–503. doi: 10.4049/jimmunol.1402302 of diabetes mellitus. Science (1976) 193:415–7. doi: 10.1126/science.180605 7. Petermann F, Korn T. Cytokines and effector T cell subsets 15. Horton C, Shanmugarajah K, Fairchild PJ. Harnessing the properties of causing autoimmune CNS disease. FEBS Lett. (2011) 585:3747–57. dendritic cells in the pursuit of immunological tolerance. Biomed J. (2017) doi: 10.1016/j.febslet.2011.03.064 40:80–93. doi: 10.1016/j.bj.2017.01.002 Frontiers in Immunology | www.frontiersin.org 10 February 2019 | Volume 10 | Article 157 Djedovic et al. Tolerogenic Effects of Ethyl Pyruvate 16. Ozenci V, Kouwenhoven M, Huang YM, Kivisäkk P, Link H. Multiple 19. Koprivica I, Vujicic M, Gajic D, Saksida T, Stojanovic I. Ethyl pyruvate sclerosis is associated with an imbalance between tumour necrosis factor- stimulates regulatory, T cells and ameliorates type 1 diabetes development in alpha (TNF-α)- and IL-10-secreting blood cells that is corrected by mice. Front Immunol. (2019) 9:3130. doi: 10.3389/fimmu.2018.03130 interferon-beta (IFN-β) treatment. Clin Exp Immunol. (2000) 120:147–53. doi: 10.1046/j.1365-2249.2000.01175.x Conflict of Interest Statement: The authors declare that the research was 17. Glasnović A, Cvija H, Stojić M, Tudorić-Ðeno I, Ivčević conducted in the absence of any commercial or financial relationships that could S, Romić D, et al. Decreased level of sRAGE, in the be construed as a potential conflict of interest. cerebrospinal fluid of multiple sclerosis patients at clinical onset. Neuroimmunomodulation (2014) 21:226–33. doi: 10.1159/0003 Copyright © 2019 Djedovic, Mansilla, Jevtić, Navarro-Barriuso, Saksida, Martínez- 57002 Cáceres and Miljković. This is an open-access article distributed under the terms 18. Belghith M, Bahrini K, Kchaou M, Maghrebi O, Belal S, of the Creative Commons Attribution License (CC BY). The use, distribution or Barbouche MR. Cerebrospinal fluid IL-10 as an early stage reproduction in other forums is permitted, provided the original author(s) and the discriminative marker between multiple sclerosis and neuro- copyright owner(s) are credited and that the original publication in this journal Behçet disease. Cytokine (2018) 108:160–7. doi: 10.1016/j.cyto.2018. is cited, in accordance with accepted academic practice. No use, distribution or 03.039 reproduction is permitted which does not comply with these terms. Frontiers in Immunology | www.frontiersin.org 11 February 2019 | Volume 10 | Article 157