Ethyl Pyruvate Modulates Murine Dendritic Cell Activation and Survival Through Their Immunometabolism

Abstract

Attenuating the innate immunity activation could ameliorate inflammation and disease in settings such as transplant rejection or autoimmunity. Recently, a pivotal role for metabolic re-programming in TLR-induced dendritic cell (DC) activation has emerged. Ethyl pyruvate (EP), a pyruvate derivative, possesses anti-inflammatory properties in vitro and in animal models of disease. However, its effects on DCs remain elusive. We found that EP attenuated LPS-induced activation of murine GM-CSF bone marrow-derived dendritic cells (DCs) in vitro, reducing pro-inflammatory cytokine and IL-10 production, costimulatory molecule and MHC expression, the type I Interferon (IFN-I) response, the LPS-induced cell death, and the ability of DCs to stimulate allogeneic T cells. DC activation induced by TLR7 and TLR9 ligands was also suppressed by EP in vitro. Finally, EP decreased TLR-induced activation stimulated in vivo in conventional DCs and inflammatory monocytes. Investigating EP mechanisms, we found that EP decreased glycolysis and mitochondrial respiration, upon and in absence of TLR stimulation, by reducing ERK, AKT, and nitric oxide (NO) activation. These results indicate that EP inhibits most of the DC biological responses to TLR triggering, altering the metabolic reprogramming necessary for DC activation.

Keywords: DC activation; TLR; dendritic cells; ethyl pyruvate; metabolism.

Figure 1

Ethyl pyruvate (EP) dose titration response. Dose titration response to EP in DCs in presence or absence of 100 ng/ml LPS. EP was administered 1 h prior to LPS stimulation (A–D) DC viability. Cells were stained with AnnexinV and 7-AAD to assess viability by flow cytometry. Results are shown as percent Annexin V-7AAD double negative cells (alive) in the CD11c gate. Statistical significance was analyzed using One-way ANOVA followed by the Bonferroni multiple comparisons test and is shown by the symbols + between conditions in absence of LPS and the untreated control, and by the symbol # between conditions in presence of LPS and the LPS control. Statistical significance was also analyzed using Two-way ANOVA followed by the Sidak multiple comparisons test and is shown by the symbol * between conditions with and without LPS in presence of the same EP concentration (E–G) Cytokine production. Culture supernatants were analyzed by ELISA for levels of TNF-a at 8 h as well as IL-12p70 and IL-6 at 24 h post-LPS stimulation. Results are the average of 2 independent experiments with two biological replicates per experiment (n = 2, 4 data numbers). Data are shown as mean ± SEM. Statistical significance was analyzed using One-way ANOVA followed by the Bonferroni multiple comparisons test. (H–O) DCs were treated with 10mM EP either 1 h before (H–K) or 1 h after (L–O) LPS stimulation, and supernatants were analyzed for (H,L) TNF-a, (I,M) IL-12p70, (J,N) IL-6, (K,O) IL-10 by ELISA. Results are shown as mean ± SEM from 4 to 5 independent experiments (n = 4–5) (H–K) and 3 independent experiments (n = 3) (L–O). Data were analyzed using the two-tailed unpaired Student's t-test. P-values P < 0.05 were considered of statistical significance. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 2

EP decreases LPS-induced DC up-regulation of surface costimulatory molecule and MHC expression. (A–C) Pre-treatment with EP. DCs were pre-treated with 10mM EP for 1 h before stimulation with LPS 100 ng/ml, then were harvested 24 h (A–C) and 48 h (C) later and analyzed by flow cytometry for surface costimulatory molecule CD86, CD80, CD40, and MHC class I and Class II expression. Results are expressed as mean ± SEM of at least 6 independent experiments (n = 6). (D,E) Delayed treatment with EP. DCs were stimulated with 100 ng/ml LPS followed by 10mM EP 1 h later, then analyzed after 24 h by flow cytometry. Results are expressed as mean ± SEM of 4 independent experiments (n = 4). Data were analyzed using the two-tailed unpaired Student's t-test. P-values P < 0.05 were considered of statistical significance. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 3

EP decreases the IFN-I response in LPS-stimulated DCs. DCs were pre-treated with 10mM EP for 1 h before LPS 100 ng/ml. Cells were harvested after (A) 1 h for Ifnb level of expression and (B–E) 6 h for the IFN-I-stimulated gene expression analysis by real-time RT-PCR. Values were expressed as fold difference in mRNA from the control (untreated cells). (F) CXCL10 ELISA from 24 h culture supernatants. Results in (A) are the average of biological replicates from 1 representative of 5 independent experiments (n = 5). Results in (B–E) are from 10 independent experiments (n = 10) and in (F) from 5 independent experiments (n = 5). Results are expressed as mean ± SEM. (G) EP reduces the ability of DCs to stimulate an allogeneic T cell response in the MLR assay. We treated DCs with LPS ± EP pre-treatment for 24 h, then we washed and used them as APCs to induce lymphocyte proliferation in an in vitro MLR assay. We used a DC:splenocyte ratio of 1:30. Lymphocyte proliferation was measured by ³H thymidine incorporation. Stimulation index was calculated by dividing the counts per minute (cpm) for each condition by the cpm of lymphocytes exposed to untreated DCs. Results are from 3 to 4 biological replicates of 2 independent experiments (n = 2, 7 data numbers). Data was analyzed using the two-tailed unpaired Student's t-test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 4

EP suppresses CpG B- and R848-induced DC activation by decreasing cytokine production and surface costimulatory molecules. DCs were pre-treated with 10mM EP for 1 h before stimulation with CpG B 10 μg/ml or R848 1 μg/ml. Cells and supernatants were harvested 24 h and 6 or 24 h post-stimulation, respectively, and analyzed by ELISA for cytokine levels and by flow cytometry for surface costimulatory molecule expression. EP decreases CD40 expression (A) and production of TNF-a (B,F), IL-6 (C,G), IL-10 (E), and IL-12p70 (D,H) upon CpG B (A–D) and R848 (A,E–H) stimulation. Results are the average of biological replicates from 2 to 3 independent experiments (n = 2–3) and are expressed as mean ± SEM. Data was analyzed using the two-tailed unpaired Student's t-test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 5

EP does not affect IKB degradation but decreases ERK phosphorylation in LPS-stimulated DCs. DCs were pre-treated with 10mM EP for 1 h before 100 ng/ml LPS. (A,B) Cells were harvested 1 h after LPS stimulation and IKBa (A) and IKBb (B) protein expression was detected by western blotting. Densitometry was analyzed by calculating the ratio of IKBa or IKBb to actin. Results are normalized to the untreated control and shown as mean ± SEM of 5–6 independent experiments (n = 5–6). (C) p-ERK/Total ERK detection. Cells were harvested 15 and 30min after LPS stimulation and ERK1/2 protein expression was detected by western blotting. Densitometry was analyzed by calculating the ratio of phosphorylated p-ERK to total ERK. GAPDH was included as a confirmation control only. Results are normalized to the untreated control and shown as mean ± SEM of 7 independent experiments (n = 7). Data was analyzed using the ratio paired two-tailed Student's t-test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 6

EP alters DC metabolism by targeting AKT and NO. We performed metabolic assays on DCs pre-treated with EP (10mM) 1 h before LPS (100 ng/ml) stimulation. (A) ECAR measurements at the acute time point (30min after LPS addition or at rate 7 of the assay, also equivalent to 30min after the beginning of the seahorse run; top) and 24 h post-stimulation (24 h post-LPS addition is equivalent to rate 7 of the assay or 30min after the beginning of the seahorse run; bottom), before the addition of mitochondrial inhibitors. Results are the average of 4 independent experiments (n = 4). (B) OCR measurements. DCs were stimulated with LPS for 30min (measurements taken at rate 7 of the assay, also equivalent to 30min after the beginning of the seahorse run; top) and 24 h (equivalent to rate 7 of the assay or 30min after the beginning of the seahorse run; bottom), and OCR was measured as a response to mitochondrial inhibitors: 1 μM oligomycin, 1.5 μM fluorocarbonyl cyanide phenylhydrazone (FCCP), and 100 nM rotenone plus 1 μM antimycin A. Representative OCR plots are shown on the left and the average of 4 independent experiments on the right (n = 4). (C) Energetic maps of control and treatments at the acute (30min; top) and the 24 h (bottom) time points. (D) Cells were harvested at 30min and 1 h after LPS stimulation and AKT phosphorylation measured by western blotting. Densitometry was analyzed by calculating the ratio of p-AKT to actin. Results are plotted as change in ratio from the untreated control and are shown as mean ± SEM of 3 independent experiments (n = 3). No change in trend when compared to total AKT. (E) Supernatants were analyzed for nitrite concentration as a proxy for nitric oxide levels using the Griess reagent kit. (F) Total RNA was extracted from DCs to determine Inos level of expression by qRT-PCR 6 h post-LPS using the Ct method. Values were normalized to cyclophillin and expressed as fold difference in mRNA from untreated cells. Results are shown as mean ± SEM and are from 6 independent experiments for nitrite levels (n = 6) and from 3 independent experiments for Inos level of expression (n = 3). Data in (A,B,E) was analyzed using the two-tailed unpaired Student's t-test and in (B) using the ratio paired two-tailed Student's t-test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 7

EP decreases the in vivo TLR-induced cDC activation without affecting their survival. C57BL/6 mice were injected i.p. with 80 mg/kg of EP in PBS (vehicle) 1 h before the injection of 30 μg/mouse of TLR7 ligand R848 in PBS (vehicle). EP was further administered again 4, 8, and 20 h after R848 stimulation. (A,C) Spleens and mesenteric lymph nodes (mLN) were stained for surface lineage markers to recognize cDCs (CD11c^Hi, CD11b^Int) and inflammatory monocytes (iMo, CD11c^Int, CD11b^Hi, and Ly6C^pos) and (B,D) the costimulatory molecule CD86 24 h after the TLR stimulation. Results are expressed as mean and single results of 2 independent experiments (n = 3–6 for spleens and n = 2–3 for mLN). Data was analyzed using the two-way ANOVA and the Sidak correction for multiple comparisons test. P-values P < 0.05 were considered of statistical significance. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.