Glucose metabolism is upregulated in the mononuclear cell proteome during sepsis and supports endotoxin-tolerant cell function

Abstract

Metabolic adaptations shape immune cell function. In the acute response, a metabolic switch towards glycolysis is necessary for mounting a proinflammatory response. During the clinical course of sepsis, both suppression and activation of immune responses take place simultaneously. Leukocytes from septic patients present inhibition of cytokine production while other functions such as phagocytosis and production of reactive oxygen species (ROS) are preserved, similarly to the in vitro endotoxin tolerance model, where a first stimulation with lipopolysaccharide (LPS) affects the response to a second stimulus. Here, we sought to investigate how cellular metabolism is related to the modulation of immune responses in sepsis and endotoxin tolerance. Proteomic analysis in peripheral blood mononuclear cells (PBMCs) from septic patients obtained at intensive care unit admission showed an upregulation of proteins related to glycolysis, the pentose phosphate pathway (PPP), production of ROS and nitric oxide, and downregulation of proteins in the tricarboxylic acid cycle and oxidative phosphorylation compared to healthy volunteers. Using the endotoxin-tolerance model in PBMCs from healthy subjects, we observed increased lactate production in control cells upon LPS stimulation, while endotoxin-tolerant cells presented inhibited tumor necrosis factor-α and lactate production along with preserved phagocytic capacity. Inhibition of glycolysis and PPP led to impairment of phagocytosis and cytokine production both in control and in endotoxin-tolerant cells. These data indicate that glucose metabolism supports leukocyte functions even in a condition of endotoxin tolerance.

Keywords: LPS; PBMCs; endotoxin-tolerance; glycolysis; immunometabolism; pentose phosphate pathway; sepsis.

Figure 1

Altered pathways in septic patient’s proteome ad D0 and D7. (A): Gene ontology biological processes significantly enriched at D0 and D7. Color scale represents -Log₁₀ (FDR). (B): metabolism-targeted IPA analysis. All selected pathways presented B–H p-value <0.05. Dots size represents the z score, which predicts the activation (red dots) or inhibition (blue dots) of a given pathway. Proteomic data from patients at D0 were obtained from (16).

Figure 2

Schematic representation of glycolysis, pentose phosphate pathway, and TCA cycle in septic patients’ PBMCs at D0 (A) and D7 (B). The arrows indicate enzymes catalyzing each reaction. Red arrows indicate an average upregulation and blue arrows indicate an average downregulation based on the Log₂ (fold-change) compared to healthy volunteers. Metabolic maps were generated using Escher.

Figure 3

TNF-α production and metabolic profile of endotoxin tolerant PBMCs. (A): PBMCs were incubated for 48h (1st stimulus) with medium (-) or different concentrations of LPS and challenged with 100 ng/mL LPS (+) or medium (-) for 24h. (B): Percentage of CD14⁺ monocytes producing TNF-α, measured by flow cytometry. (C): Lactate concentration in culture supernatants. (D): Intracellular ATP production. Graphs show mean and SD and each dot represents one individual donor evaluated in all experimental conditions (n = 5 donors); p-values indicate differences between each condition as measured by repeated measures one-way ANOVA followed by Bonferroni posttest.

Figure 4

Phagocytic capacity in control and endotoxin tolerant PBMCs is affected by the modulation of cell metabolism. (A): Schematic figure of the inhibition of cell metabolism using 2-DG, 6-AN, and Oligomycin and its specific targets. (B): PBMCs were incubated for 48h (1^st stimulus) with medium or 100 ng/mL of LPS, washed, and incubated for 22 h with the metabolic modulators before exposition to pHrodo green E. coli particles for 2.5 h. (C): Phagocytic capacity of control (medium) and endotoxin tolerant (100 ng/mL LPS) CD14⁺ monocytes measured as the median fluorescence intensity (MFI) of pHrodo green by flow cytometry. (D): Phagocytic capacity of control and endotoxin tolerant monocytes incubated with metabolic inhibitors. Graphs show mean and SD and each dot represents one individual donor evaluated in all experimental conditions (n = 5 donors); p-values indicate differences between each condition as measured by paired Student’s t-test (C) or repeated measures one-way ANOVA followed by Bonferroni posttest (D). ns, not significant.

Figure 5

Glucose metabolism supports cytokine production in control and endotoxin tolerant PBMCs. (A): PBMCs were incubated for 48h (1^st stimulus) with medium or 100 ng/mL of LPS, washed, and incubated for 1 h with 2-DG, 6-AN, and Oligomycin before challenge with 100 ng/mL LPS. TNF-α (B) and IL-6 (D) concentration in the culture supernatant of control (-) and endotoxin-tolerant PBMCs challenged with LPS. TNF-α (C) and IL-6 production (E) by control (-) and endotoxin tolerant PBMCs incubated with metabolic inhibitors along with LPS challenge. Graphs show mean and SD and each dot represents one individual donor evaluated in all experimental conditions (n = 6 donors); p-values indicate differences between each condition as measured by paired Student’s t-test (B, D) or repeated measures one-way ANOVA followed by Bonferroni posttest (C, E). ns, not significant.