Modulation of dendritic cell metabolism by an MPLA-adjuvanted allergen product for specific immunotherapy

Abstract

Background: Recently, bacterial components were shown to enhance immune responses by shifting immune cell metabolism towards glycolysis and lactic acid production, also known as the Warburg Effect. Currently, the effect of allergen products for immunotherapy (AIT) and commercial vaccines on immune cell metabolism is mostly unknown.

Objective: To investigate the effect of AIT products (adjuvanted with either MPLA or Alum) on myeloid dendritic cell (mDC) metabolism and activation.

Methods: Bone marrow-derived mDCs were stimulated with five allergoid-based AIT products (one adjuvanted with MPLA, four adjuvanted with Alum) and two MPLA-adjuvanted vaccines and analyzed for their metabolic activation, expression of cell surface markers, and cytokine secretion by ELISA. mDCs were pre-incubated with either immunological or metabolic inhibitors or cultured in glucose- or glutamine-free culture media and subsequently stimulated with the MPLA-containing AIT product (AIT product 1). mDCs were co-cultured with allergen-specific CD4+ T cells to investigate the contribution of metabolic pathways to the T cell priming capacity of mDCs stimulated with AIT product 1.

Results: Both the MPLA-containing AIT product 1 and commercial vaccines, but not the Alum-adjuvanted AIT products, activated Warburg metabolism and TNF-α secretion in mDCs. Further experiments focused on AIT product 1. Metabolic analysis showed that AIT product 1 increased glycolytic activity while also inducing the secretion of IL-1β, IL-10, IL-12, and TNF-α. Both rapamycin (mTOR-inhibitor) and SP600125 (SAP/JNK MAPK-inhibitor) dose-dependently suppressed the AIT product 1-induced Warburg Effect, glucose consumption, IL-10-, and TNF-α secretion. Moreover, both glucose- and glutamine deficiency suppressed secretion of all investigated cytokines (IL-1β, IL-10, and TNF-α). Glucose metabolism in mDCs was also critical for the (Th1-biased) T cell priming capacity of AIT product 1-stimulated mDCs, as inhibition of mTOR signaling abrogated their ability to induce Th1-responses.

Conclusion: The AIT product and commercial vaccines containing the adjuvant MPLA were shown to modulate the induction of immune responses by changing the metabolic state of mDCs. Better understanding the mechanisms underlying the interactions between cell metabolism and immune responses will allow us to further improve vaccine development and AIT.

Keywords: MPLA: monophosphoryl lipid A; Warburg Effect; allergen specific immunotherapy; immune metabolism; vaccine.

Figure 1

An MPLA-containing AIT product can activate the metabolism of dendritic cells. C57BL/6 bone marrow-derived mDCs were stimulated with either 0.025, 0.25, 2.5, 5, or 50 (only for AIT products 2 to 5) PNU/ml of the different AIT products (see Table 1 for coding and composition of the products and Table 2 for protein concentrations) or 10 µg/ml LPS as a positive control for 72 h and analyzed for the activation of mDC metabolism and cytokine secretion. The Warburg Effect, glucose consumption from the culture medium, and TNF-α secretion were determined 72 h post-stimulation. Data are mean results of three independent experiments ± SD. Data displayed no gaussian normal distribution. For statistical analysis a Kruskal-Wallis test with correction for multiple comparisons according to Dunn was applied. Statistical significance was achieved at **p<0.01, ***p<0.001, respectively with “n.s.” representing non-significant results.

Figure 2

MPLA-containing vaccine preparations also activate the metabolism of dendritic cells. C57BL/6 bone marrow-derived mDCs were stimulated with either AIT product 1, two commercial, MPLA-containing vaccines (see Table 1 for coding and composition of the products), or 10 µg/ml LPS as a positive control for 72 h and analyzed for the activation of mDC metabolism and cytokine secretion (A). Cells were analyzed for the induced Warburg Effect and TNF-α secretion 72 h post-stimulation and extracellular acidification rates (ECAR) and oxygen consumption rates (OCR) in cycle 7 (42 min) post-stimulation using Seahorse technology (B). Stimulation concentrations for all three tested products were normalized to contain 0.76 µg/ml of MPLA (corresponding to a total protein amount of 5 PNU/ml of AIT product 1). For Extracellular Flux Assays, mDCs adhered to the plastic plate were stimulated with either 100 ng/ml of LPS or the indicated stimulation concentrations of AIT product 1 or vaccine 1 and 2 (with stimulation concentrations normalized to their MPLA content) for the indicated durations (B,C). 14 cycles (84 min) post-stimulation, the ATP synthase, the electron transfer chain, and glycolysis were inhibited, respectively, by sequential injection of oligomycin, rotenone/antimycin A (Rot/AA), and 2-deoxy-glucose (2-DG) for 8 cycles (48 min) each. Data are either representative (C) or mean results (B) of three to five independent experiments ± SD Data displayed a gaussian normal distribution. For statistical analysis a ONE-way ANOVA with correction for multiple comparisons according to Tukey was applied. Statistical significance was achieved at *p<0.05, **p<0.01, ***p<0.001, respectively with “n.s.” representing non-significant results.

Figure 3

AIT products and MPLA-adjuvanted vaccines upregulate surface marker expression of bone marrow-derived mDCs. C57BL/6 mDCs were stimulated for 3 days with either the indicated AIT products or MPLA-containing vaccines (all corresponding to a total protein amount of 5 PNU/ml of AIT product 1), cells were harvested, and stained for surface expression of CD40, CD69, CD80, or CD86 (A). Expression levels of the indicated maturation and activation markers on CD11b+CD11c+B220- mDCs after stimulation with either AIT products (B) or MPLA-containing vaccines (C) were determined by flow cytometry. Dashed lines: isotype control-stained cells, light grey filled curves: unstimulated cells, colored curves: stimulated as indicated. Data are representative results taken from one out of two experiments.

Figure 4

Stimulation of mDCs with MPLA results in similar activation of glucose metabolism and cytokine secretion to AIT product 1. C57BL/6 bone marrow-derived mDCs were stimulated with increasing concentrations of either MPLA, L-Tyrosine, the mixture of MPLA and L-Tyrosine, or AIT product 1 (all equivalent to either 0.025, 0.25, 1.25, 2.5, or 5 PNU/ml of AIT product 1) or 10 µg/ml LPS as a positive control for 72 h and analyzed for the activation of mDC metabolism and cytokine secretion (A). The induced Warburg Effect (B) and the secretion of the indicated cytokines (C–E) were determined 72 h post-stimulation. Data are mean results of three independent experiments ± SD. Data displayed a gaussian normal distribution. For statistical analysis a ONE-way ANOVA with correction for multiple comparisons according to Tukey was applied. Statistical significance was achieved at *p<0.05, **p<0.01, ***p<0.001, respectively with “n.s.” representing non-significant results.

Figure 5

AIT product 1-mediated mDC activation is mainly driven by mTOR-dependent glycolysis. C57BL/6 bone marrow-derived mDCs were pre-treated with the indicated amounts of either the mTOR inhibitor rapamycin (0.1, 1, or 10 nM), the hexokinase 2 inhibitor 2-deoxyglucose (2-DG) (0.01 or 0.1 mM), the amino acid metabolism inhibitor BPTES (0.02 or 0.2 µM), or the inhibitors of fatty acid synthase cerulenin (0.02 or 0.2 µg/ml) and fatty acid oxidation etomoxir (0.05 or 0.5 µM) for 90 minutes and subsequently stimulated with 2.5 PNU/ml AIT product 1 (containing 0.38 µg/ml MPLA) for additional 72 h. 10 µg/ml LPS served as a positive control. mDCs were subsequently analyzed for the activation of mDC metabolism and cytokine secretion (A). The Warburg Effect (B), glucose concentration in the medium and metabolic rates (C), and cytokine secretion (D) were determined 72 h post-stimulation. Data are mean results of three independent experiments ± SD. Data displayed a gaussian normal distribution. For statistical analysis a ONE-way ANOVA with correction for multiple comparisons according to Tukey was applied. Statistical significance was achieved at *p<0.05, **p<0.01, ***p<0.001, respectively with “n.s.” representing non-significant results.

Figure 6

Activation of mDC metabolism, IL-10-, and TNF-α secretion by AIT product 1 depend on SAP/JNK MAPK-signaling. C57BL/6 bone marrow-derived mDCs were pre-treated with the indicated amounts of the SAP/JNK MAPK inhibitor SP600125 for 90 minutes and subsequently stimulated with 5 PNU/ml of AIT product 1 (containing 0.76 µg/ml MPLA) for additional 72 h (A). 10 µg/ml LPS served as a positive control. mDCs were analyzed for the activation of mDC metabolism and cytokine secretion. The Warburg Effect, glucose concentration in the medium, metabolic rates, and cytokine secretion were determined 72 h post-stimulation (B). Data are mean results of three independent experiments ± SD. Data displayed a gaussian normal distribution. For statistical analysis a ONE-way ANOVA with correction for multiple comparisons according to Tukey was applied. Statistical significance was achieved at *p<0.05, **p<0.01, ***p<0.001, respectively with “n.s.” representing non-significant results.

Figure 7

The stimulation of mDCs with AIT product 1 in different culture media demonstrates a glucose- and glutamine-dependency for the activation of mDC metabolism and cytokine secretion. C57BL/6 bone marrow-derived mDCs were stimulated with 2.5 PNU/ml of AIT product 1 (containing 0.38 µg/ml MPLA) for 72 h in either complete culture medium, glucose-free, or glutamine-free medium. 10 µg/ml LPS served as a positive control. mDCs were subsequently analyzed for the activation of mDC metabolism and cytokine secretion (A). The Warburg Effect (B), glucose concentration in the medium and metabolic rates (C), and cytokine secretion (D) were determined 72 h post-stimulation. n.d. = not detectable due to lack of glucose in medium. Data are mean results of three independent experiments ± SD. Data displayed a gaussian normal distribution. For statistical analysis a ONE-way ANOVA with correction for multiple comparisons according to Tukey was applied. Statistical significance was achieved at *p<0.05, **p<0.01, ***p<0.001, respectively with “n.s.” representing non-significant results.

Figure 8

The stimulation of mDC:T cell co-cultures with AIT product 1 leads to increased secretion of Th1 cytokines. BALB/C mDCs were co-cultured with CD4+ T cells isolated from spleens of BALB/C mice that were previously immunized with the major birch pollen allergen Bet v 1 and Alum. Prior to co-cultures, mDCs were pre-treated for 16 hours with metabolic inhibitors (5 nM rapamycin, 0.5 mM 2-DG, or 1 µM BPTES), then the medium was changed, T cells were added, and co-cultures were re-stimulated with either 2 µg/ml Bet v 1 alone or together with 2.5 PNU/ml of AIT product 1 (containing 0.38 µg/ml MPLA) for additional 70 hours (A). The Warburg Effect was determined in mDCs (bars without pattern) and the co-cultures (bars with pattern) (B). Cytokine secretion was determined for IL-2 24 h post-stimulation and for IL-5, IL-13, IFN-γ, IL-12p70, IL-10, IL-6, TNF-α, and IL-1β 72 h post-stimulation (C). Data are mean results of three independent experiments ± SD. Data displayed a gaussian normal distribution. For statistical analysis a ONE-way ANOVA with correction for multiple comparisons according to Tukey was applied. Statistical significance was achieved at *p<0.05, **p<0.01, ***p<0.001, respectively with “n.s.” representing non-significant results.