Enhanced Glycolysis Is Required for Antileishmanial Functions of Neutrophils Upon Infection With Leishmania donovani

Abstract

Visceral leishmaniasis (VL) is a fatal parasitic disease if untreated. Treatment options of VL diminish due to emerging drug resistance. Although the principal host cells for the multiplication of Leishmania are macrophages, neutrophils are the first cells infected with the parasites rapidly after parasite inoculation. Leishmania can survive in neutrophils despite the potent antimicrobial effector functions of neutrophils that can eliminate the parasites. Recently, the growing field of immunometabolism provided strong evidence for the therapeutic potential in targeting metabolic processes as a means of controlling immune effector functions. Therefore, the understanding of the immunometabolic profile of neutrophils during Leishmania infection could provide new promising targets for host-directed therapies against VL. To our knowledge, this is the first study addressing the bioenergetics profile of L. donovani-infected primary human neutrophils. Transcriptome analysis of L. donovani-infected neutrophils revealed an early significant upregulation of several glycolytic enzymes. Extracellular flux analysis showed that glycolysis and glycolytic capacity were upregulated in L. donovani-infected neutrophils at 6 h post infection. An increased glucose uptake and accumulation of glycolytic end products were further signs for an elevated glycolytic metabolism in L. donovani-infected neutrophils. At the same time point, oxidative phosphorylation provided NADPH for the oxidative burst but did not contribute to ATP production. Inhibition of glycolysis with 2-DG significantly reduced the survival of L. donovani promastigotes in neutrophils and in culture. However, this reduction was due to a direct antileishmanial effect of 2-DG and not a consequence of enhanced antileishmanial activity of neutrophils. To further address the impact of glucose metabolism during the first days of infection in vivo, we treated C57BL/6 mice with 2-DG prior to infection with L. donovani and assessed the parasite load one day and seven days post infection. Our results show, that seven days post-infection the parasite load of 2-DG treated animals was significantly higher than in mock treated animals. This data indicates that glycolysis serves as major energy source for antimicrobial effector functions against L. donovani. Inhibition of glycolysis abrogates important neutrophil effector functions that are necessary the initial control of Leishmania infection.

Keywords: 2-deoxyglucose; glucose metabolism; glycolysis (glycolytic pathway); host-directed therapy (HDT); leishmania; neutrophils; oxidative phosphorylation (OXPHOS); transcriptome (RNA-seq).

Figure 1

L. donovani induces enhanced glycolysis in neutrophils. (A) Heatmap of transcripts of selected glycolytic genes for six samples without and six samples 6 h after L. donovani infection. Displayed expression values have been normalized across samples by sleuth and are provided as log2 transcripts per million (log2 TPM). Shown are transcripts of human glycolytic genes according to KEGG pathway glycolysis if this transcript (i) is significantly differentially expressed according to sleuth likelihood ratio test with multiple testing corrected p-value, i.e., q-value, <0.05 and (ii) is highly expressed (sleuth mean of natural log counts ≥5). Heatmap annotation shows that samples cluster primarily by condition (infected vs. uninfected) and secondarily by donor (A, B, C). A heatmap of the highest expressed transcript of every gene that could be mapped to the KEGG glycolysis pathway (human as well as L. major) is provided in Supplementary Figure 3. (B) Estimated effect sizes according to sleuth Wald test for the most highly expressed, significant transcript of genes mapping to KEGG pathway glycolysis (colored are only genes with at least one transcript having mean of natural log count ≥5). Highly expressed and significantly differentially expressed human genes shown in (A) constitute the major glycolytic axis, and consistently show an increase of expression in infected neutrophils with the exception of BPGM, which has opposite effect by similar order of magnitude.

Figure 2

Glycolysis stress test profile of L. donovani-infected neutrophils. Primary human neutrophils were infected with L. donovani promastigotes (ratio 1:10) for 3 h at 37°C and 5% CO₂. Uninfected cells served as control. The infection rate was determined by Giemsa staining of cytocentrifuged samples. Subsequently, free, non-ingested parasites were removed by washing. After 6 h post infection the glycolysis stress test profile was determined by using the Seahorse extracellular flux analyzer (A). Extracellular acidification rate (ECAR) measurements following the sequential injection of 5 mM glucose, 1 μM oligomycin, and 10 mM 2-DG (dotted lines indicate injection time) were used to calculate key parameters of glycolytic function. Glycolysis (B) was calculated by subtraction of 2-DG-mediated ECAR from glucose-mediated ECAR. Glycolytic capacity (C) was calculated by subtraction of 2-DG-mediated ECAR from oligomycin-mediated ECAR. Bar graphs show mean ± SD (n = 8), ^**p ≤ 0.01.

Figure 3

Mitochondrial stress test profile of L. donovani-infected neutrophils. Primary human neutrophils were infected with L. donovani promastigotes (ratio 1:10) for 3 h at 37°C and 5% CO₂. Uninfected cells served as control. Successful infection was determined by Giemsa staining of cytocentrifuged samples. Subsequently, free, non-ingested parasites were removed by washing. After 6 h post infection the mitochondrial stress test profile was determined by using the Seahorse extracellular flux analyzer (A). The measurement of basal oxygen consumption rate (OCR) was followed by sequential injections of 1 μM oligomycin, 1.5 μM FCCP, and 1 μM rotenone/antimycin A (dotted lines indicate injection time). OCR measurements were used to calculate key parameters of mitochondrial function. Non-mitochondrial respiration (B) was calculated as OCR after rotenone/antimycin A injection. Basal respiration (C) was calculated by subtraction of rotenone/antimycin A-mediated OCR from basal OCR. Maximal respiration (D) was calculated by subtraction of rotenone/antimycin A-mediated OCR from FCCP-mediated OCR. Proton leak (E) was calculated by subtraction of non-mitochondrial respiration from oligomycin-mediated OCR. ATP production (F) was calculated by subtraction of oligomycin-mediated OCR from basal OCR. Bar graphs show mean ± SD (n = 3), ^**p ≤ 0.01, ns, not significant.

Figure 4

2-NBDG uptake of L. donovani-infected neutrophils. Primary human neutrophils were infected with L. donovani promastigotes (ratio 1:10) for 3 h at 37°C and 5 % CO₂. Uninfected cells served as control. The infection rate was determined by Giemsa staining of cytocentrifuged samples. Subsequently, free, non-ingested parasites were removed by washing. After 6 h post infection the fluorescent glucose analog 2-NBDG was added to the cells in glucose-free medium for the last 10 min of incubation time and 2-NBDG uptake was analyzed by flow cytometry. The bar diagram shows the autofluorescence corrected mean fluorescence intensity (MFI) of 2-NBDG uptake ± SD (n = 3), ^***p ≤ 0.001.

Figure 5

Lactate secretion and pyruvate content of L. donovani-infected neutrophils. Primary human neutrophils were infected with L. donovani promastigotes (ratio 1:10) for 3 h at 37°C and 5% CO₂. Uninfected cells served as control. The infection rate was determined by Giemsa staining of cytocentrifuged samples. Subsequently, free, non-ingested parasites were removed by washing. Whole cell lysates were generated after 6 h post infection by boiling cells at 90°C for 5 min. Cell-free supernatants were collected after 6 h post infection. Lactate was detected in culture supernatants by using a lactate assay kit. Pyruvate was detected in whole cell lysates by using a pyruvate assay kit. Bar diagrams show mean concentration of lactate (A) after subtraction of the medium blanks and pyruvate (B) calculated by interpolation from standard curve ± SD (n = 5), ^*p ≤ 0.05, ^****p ≤ 0.0001.

Figure 6

ATP concentration in L. donovani-infected neutrophils as response to metabolic inhibitors. Primary human neutrophils were infected with L. donovani promastigotes (ratio 1:10) for 3 h at 37°C and 5% CO₂. Uninfected cells served as control. The infection rate was determined by Giemsa staining of cytocentrifuged samples. Subsequently, free, non-ingested parasites were removed by washing. After 6 h post infection cells were treated with 10 mM 2-DG for 3 h at 37°C and 5% CO₂. PBS served as solvent control. Whole cell lysates were prepared and the ATP concentration was determined by using the ATP determination kit. Bar diagrams show the mean ATP concentration ± SD (n = 3) in uninfected (A) and L. donovani-infected neutrophils (B), ^*p ≤ 0.05, ^***p ≤ 0.001, ^****p ≤ 0.0001.

Figure 7

Survival of L. donovani promastigotes in 2-DG-treated neutrophils. Primary human neutrophils were infected with L. donovani promastigotes (ratio 1:10) for 3 h at 37°C, 5% CO₂. The infection rate was determined by Giemsa staining of cytocentrifuged samples. After removing the free, non-ingested parasites the infected cells were treated with PBS, 5 mM, 50 mM, or 100 mM 2-DG. Survival of parasites was assessed after 24 h post infection by using the limiting dilution assay. The bar diagram shows the mean survival rates (%) normalized to PBS-treated control cells ± SD (n = 3), ^****p ≤ 0.0001.

Figure 8

ROS production of 2-DG-treated L. donovani-infected neutrophils. Primary human neutrophils were infected with L. donovani promastigotes (ratio 1:10) for 3 h at 37°C, 5% CO₂. The infection rate was determined by Giemsa staining of cytocentrifuged samples. After removing the free, non-ingested parasites the infected and uninfected cells were treated with PBS, 5 mM, 50 mM, or 100 mM 2-DG. After 24 h post infection the MPO-derived ROS production was measured for 1 h at 37°C and 5% CO₂ after the stimulation with 20 nM PMA by using the luminol-based chemiluminescence assay. A representative curve of luminol chemiluminescence is shown in panel (A). The bar diagram (B) shows the mean area under the curve (AUC) values ± SD (n = 3), ^*p ≤ 0.05.

Figure 9

Survival of L. donovani promastigotes in the presence of 2-DG. L. donovani promastigotes were adjusted to 25 x 10⁶ cells/ml in complete medium and treated with 5 mM, 50 mM, or 100 mM 2-DG for 24 h at 37°C and 5% CO₂. PBS treatment served as solvent control. Survival of parasites was assessed after 24 h by using the limiting dilution assay. The bar diagram shows the mean normalized to PBS-treated control cells ± SD (n = 3), ^***p ≤ 0.001, ns, not-significant.

Figure 10

Impairment of glycolytic pathway in vivo increases susceptibility to L. donovani infection. (A) Mice were pre-treated with 2-deoxyglucose (2-DG) on days −3, −2 and −1 before infection. Untreated and 2-DG-treated mice were then infected intraperitoneally with 1 x 10⁸ L. donovani promastigotes and mice were euthanized at day 1 (d1) and day 7 (d7) post-infection. (B) Parasite burden in the liver and the spleen of infected animals at day 1 and day 7 post-infection. (C) The total number of splenocytes were accessed in untreated and 2-DG-treated mice. (D) The percentage and number of neutrophils were quantified in the spleen. (E) Expression of CD62L and CD11b were evaluated in neutrophils of infected animals at both timepoints. Data is expressed as mean ± SD. ^*p < 0.05; ^**p < 0.01; ^***p < 0.001. (F) The percentage and number of monocytes were quantified in the spleen. Expression of MHCII was evaluated in monocytes of infected animals at both time points (d1 and d7). Data is expressed as mean ± SD. ^*p < 0.05; ^**p < 0.01.