MAPK/ERK activation in macrophages promotes Leishmania internalization and pathogenesis

Abstract

The obligate intracellular parasite Leishmania binds several receptors to trigger uptake by phagocytic cells, ultimately resulting in visceral or cutaneous leishmaniasis. A series of signaling pathways in host cells, which are critical for establishment and persistence of infection, are activated during Leishmania internalization. Thus, preventing Leishmania uptake by phagocytes could be a novel therapeutic strategy for leishmaniasis. However, the host cellular machinery mediating promastigote and amastigote uptake is not well understood. Here, using small molecule inhibitors of Mitogen-activated protein/Extracellular signal regulated kinases (MAPK/ERK), we demonstrate that ERK1/2 mediates Leishmania amazonensis uptake and (to a lesser extent) phagocytosis of beads by macrophages. We find that inhibiting host MEK1/2 or ERK1/2 leads to inefficient amastigote uptake. Moreover, using inhibitors and primary macrophages lacking spleen tyrosine kinase (SYK) or Abl family kinases, we show that SYK and Abl family kinases mediate Raf, MEK, and ERK1/2 activity and are necessary for uptake. Finally, we demonstrate that trametinib, a MEK1/2 inhibitor used to treat cancer, reduces disease severity and parasite burden in Leishmania-infected mice, even if it is started after lesions develop. Our results show that maximal Leishmania infection requires MAPK/ERK and highlight potential for MAPK/ERK-mediated signaling pathways to be novel therapeutic targets for leishmaniasis.

Keywords: Kinase; Leishmania; MAPK/ERK; Macrophage; Parasite.

Fig. 1.. The MAPK/ERK pathway is required for optimal amastigote uptake.

(A) Trametinib decreases IgG-opsonized amastigote uptake. RAW 264.7 cells were treated with 3.3 μM imatinib, 1 μM trametinib or DMSO for 2 h, and 10 IgG-opsonized amastigotes were incubated per RAW 264.7 cell for 2 h at 37 °C. An immunofluorescence assay distinguished between intracellular (green) and extracellular (red + green = orange) amastigotes. Nuclei are labeled with Hoescht 33342 (blue). Scale bar, 100 μm. (B) Graph showing the mean phagocytic index (PI) ± standard deviation (SD) for RAW 264.7 cells normalized to DMSO for each experiment (100%), quantified from n = 6 biological experiments. Comparisons between imatinib and trametinib by ANOVA were nonsignificant. (C) Imatinib and trametinib do not affect the total (external + internal) number of amastigotes per 100 RAW cells (termed an adhesion index). Shown are percentages of total amastigotes per 100 imatinib-treated and trametinib-treated RAW 264.7 cells relative to DMSO-treated cells for the experiment shown in B. (D) Titration of trametinib demonstrates that the reduction of amastigote internalization is maximal at ~100x the purified MEK1/2 IC₅₀. RAW 264.7 cells were treated with increasing doses of trametinib (0.25, 0.5, 1, or 2.5 μM) or DMSO. Bar graph shows the mean PI ± SD for trametinib-treated samples relative to DMSO-treated samples for each experiment (100%) for 4 biological experiments. (E) The adhesion index for amastigotes from the experiment in D is not affected by increasing doses of trametinib. Experiment performed as in C. (F) The relative decrease in PI for trametinib-treated samples persists even after prolonged incubation. RAW 264.7 cells were treated with DMSO (D) or 1 μM trametinib (T) prior to addition of IgG-opsonized amastigotes for 1 h, 2 h, or 4 h, as indicated. Shown is the mean PI ± SD for each time point for the trametinib-treated condition, relative to the DMSO-treated condition, calculated from 4 biological experiments. (G) Treating samples with trametinib for prolonged periods does not affect the adhesion index for amastigotes. Shown are the total number of amastigotes (internal + external) per 100 RAW 264.7 cells for the experiments performed in F. For all experiments, **P < 0.01, ****P < 0.0001, ns = nonsignificant compared to the relevant DMSO-treated category by ANOVA. (H) Trametinib is not toxic to amastigotes. Shown is one representative experiment of five experiments with two technical replicates following the number of amastigotes growing in untreated, DMSO-treated, or trametinib-treated media over 5 days relative to on day 0. ns, P > 0.05 compared to DMSO category by ANOVA.

Fig. 2.. MAPK/ERK are activated during Leishmania uptake by Mφ.

Amastigotes were added to ≥ 70% confluent RAW 264.7 cells. Equal amounts of protein from whole lysates were employed for each category. Top: Representative blots displaying relative activation of MAPK/ERK. Phosphorylated (A) ERK1/2, (B) MEK1/2, (C) cRaf (Raf-1) and the corresponding total proteins were detected by Western Blot. An unrelated protein, GAPDH, is shown as an additional loading control. Bottom: Quantification of phosphorylated MAPK/ERK levels. Graphs shown in (A) for ERK1/2, (B) for MEK1/2, and (C) for cRaf represent mean ± SD of relative phosphorylated kinase levels, normalized to total kinase levels, for at least 3 biological replicates per condition. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by ANOVA compared to the 0 m time point for each set of kinases.

Fig. 3.. PMA activates ERK1/2 to modulate amastigote uptake.

(A, B) PKC typically is only slightly activated during amastigote uptake. PMA serves as a positive control. Amastigotes were added to monolayers of RAW 264.7 cells and processed as above. Shown are (A) representative blots demonstrating and (B) bar graphs quantifying relative activation of pPKC/PKC. *P < 0.05, ***P < 0.001, ****P < 0.0001 compared with samples indicated by brackets (ANOVA). n = 3 experiments. (C) PMA is not toxic to amastigotes. Shown is one representative experiment of five experiments with two technical repeats following the number of amastigotes growing in DMSO-treated or PMA-treated media over 5 days relative to on day 0. ns, P > 0.05 compared to DMSO category by ANOVA. (D) Representative image demonstrating that PMA rescues trametinib inhibition of Leishmania amastigote uptake. BMDM were treated with DMSO, 1 μM PMA, 1 μM trametinib or a combination of 1 μM trametinib and 1 μM PMA prior to incubation with IgG-opsonized amastigotes for 2 h. (E) PMA increases IgG-opsonized amastigote uptake. BMDM were treated with DMSO or increasing doses of PMA (0.25, 0.5, 1, 2.5 μM) prior to incubation with IgG-opsonized amastigotes for the times indicated. n = 4 experiments. **P < 0.01, ***P < 0.001, ****P < 0.0001 compared to DMSO category by ANOVA. (F) PMA does not affect the adhesion index for amastigotes. ns, not significant by ANOVA. (G) Mean PI ± SD for BMDM treated with trametinib/PMA relative to DMSO-treated samples for each experiment (100%), obtained from 3 biological experiments. ****P < 0.0001, ns = nonsignificant by ANOVA compared to DMSO-treated sample. (H) Trametinib/PMA does not affect amastigote adhesion for the experiment shown in G.

Fig. 4.. SYK and Arg/Abl stimulate ERK1/2 phosphorylation during amastigote uptake.

(A) Phosphorylation of ERK1/2 in RAW 264.7 cells is induced upon amastigote uptake and is decreased by imatinib (Arg/Abl inhibitor), entospletinib (SYK inhibitor), SCH772984 (ERK1/2 inhibitor), and trametinib (MEK1/2 inhibitor). PMA serves as a positive control for pERK stimulation. RAW 264.7 cells were allowed to adhere to plates, and then treated with compounds or DMSO prior to IgG-opsonized amastigote addition, lysis, and immunoblotting. Representative immunoblots of pERK1/2 (top) and total ERK1/2 (bottom) [with (+) or without (−) amastigotes] are shown. (B) Relative pERK levels, normalized to ERK1/2 levels, among categories shown in A. Levels of pERK are normalized to amastigote-exposed DMSO-treated RAW 264.7 cells (100%). 5 biological replicates are quantified. ***P < 0.001, ****P < 0.0001 compared with amastigote-stimulated, DMSO-treated RAW 264.7 cells (ANOVA). (C) Phosphorylation of ERK1/2 induced upon FcγR engagement is decreased in SYK^−/− and Arg^−/−Abl^−/− BMDM compared to WT BMDM. Shown is a representative immunoblot of pERK1/2 (top) and total ERK1/2 (bottom) in WT, SYK^−/− and Arg^−/−Abl^−/− BMDM treated with DMSO (D) or PMA (P), ± incubation with amastigotes. (D) Graph showing relative levels of pERK1/2, normalized to ERK1/2 levels, among WT, SYK^−/−, and Arg^−/−Abl^−/− BMDM treated with DMSO (D) or PMA (P), ± amastigotes (A). **P < 0.01, ***P < 0.001, ****P < 0.0001, ns: not significant by ANOVA (n = 6 experiments), compared to DMSO + amastigote category for each condition. (E–G) Primary cells lacking Arg and Abl exhibit decreases in amastigote uptake, but PMA can partially rescue this amastigote internalization defect. (E) Representative image performed as in Fig. 1. (F) Mean phagocytic index ± SD for PMA-treated and/or Arg^−/− Abl^−/− BMDM relative to WT DMSO-treated BMDM. (G) DMSO and PMA do not affect the adhesion indexes for the experiments shown in F. n for F-G = 3 experiments. **P < 0.01; ***P < 0.001; ****P < 0.0001; ns, not significant compared with DMSO-treated WT BMDM (ANOVA).

Fig. 5.. ERK 1/2’s role in Leishmania uptake is greater than for phagocytosis of beads.

(A) Trametinib mildly decreases bead internalization. RAW 264.7 cells were treated with 3.3 μM imatinib, 1 μM trametinib or DMSO for 2 h prior to addition of 10 IgG-opsonized beads per RAW 264.7 cell for 2 h at 37 °C. Treating RAW 264.7 cells with trametinib decreases uptake of IgG-opsonized beads to a lesser degree than imatinib. Graph shows the mean PI ± SD for each category, normalized to DMSO (100%). ****P < 0.0001, ns: not significant compared to indicated categories by ANOVA (n = 6 biological experiments). (B) Imatinib and trametinib does not affect the adhesion index for beads in the experiments shown in A. (C–H) IgG-coated beads activate the MAPK/ERK pathway, but phosphorylation is not sustained. Beads were added to > 70% confluent RAW 264.7 cells and experiments were processed as in Fig. 2. Top: Phosphorylated forms of (C) ERK, (E) MEK1/2, (G) cRaf and the corresponding total kinase were detected by Western blotting. Bottom: Relative activation was quantified for (D) ERK, (F) MEK1/2, and (H) cRaf, using the methods described in Fig. 2. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 and ns: not significant compared to 0 m timepoint by ANOVA (n ≥ 3 experiments).

Fig. 6.. MAPK/ERK permit efficient infection in a mouse model of cutaneous leishmaniasis.

(A) Mouse footpad model for cutaneous leishmaniasis. C57BL/6 mice diluent and trametinib-treated groups were injected with promastigotes in the hind footpad. Mice were given 2 mg/kg body weight of trametinib or an equal volume of diluent by oral gavage daily, starting 7 days before infection and continuing until the mice were euthanized. Two experiments were performed. Shown in the following panels is one representative experiment containing 10 female mice per group. (B) Trametinib given prior to infection reduces lesion size in mice. Depth of foot swelling was monitored with calipers by a blinded investigator. Results represent the mean ± SD foot size increase over the uninfected foot (normalized to 1). The control (diluent) and drug-treated groups were compared to swelling at time 0 by ANOVA; *P < 0.05, ****P < 0.0001, ns, not significant. (C) Trametinib given prior to infection significantly reduces parasite burden. Parasite burdens in infected footpads were determined by limiting dilution as described in Materials and Methods. Plotted is the mean number of parasites (i.e., parasite burden) in millions at the end of the experiment shown in B ± SD. In this example, there were 1.41 × 10⁶ parasites in the control mice versus 2.03 × 10⁵ parasites in the trametinib-treated mice. *P < 0.05 (two-tailed t-test). (D) Cytokine/chemokine analysis of draining lymph nodes from above mice, harvested at experiment termination, and treated with Con A (positive control), 1 × 10⁶ Leishmania promastigotes/sample (high stimulation) or 1 × 10⁵ promastigotes/sample (low stimulation) X 72 h. Cultured supernatants were harvested for analysis. *P < 0.05 (ANOVA). (E) C57BL/6 mice diluent and trametinib-treated groups were injected with L. amazonensis promastigotes as above. Starting 6 weeks after inoculation (arrow), mice were given 2 mg/kg body weight of trametinib or vehicle by oral gavage daily and continuing until the experiment was terminated. (F) Trametinib given once lesions have begun reduces lesion size in trametinib-treated mice compared to diluent-treated mice. Arrow shows start time of drug treatment. Shown is one representative experiment containing 10 female mice per group. Statistics calculated as in B. (G) Lesions in mice treated with trametinib 6 weeks after infection contain fewer Leishmania parasites than control mice. Plotted is the mean ± SD lesion parasite burden in millions at the end of the experiment shown in E; in this example, there were 1.01 × 10⁶ parasites in the control mice versus 8.29 × 10⁴ parasites in the trametinib-treated mice (>10-fold reduction). *P < 0.05 (two-tailed t-test). (H) MAPK/ERK signal transduction in Mφ facilitates Leishmania uptake and pathogenesis. Upon FcγR ligation by amastigotes, SFK, Arg, and SYK are activated and relay signals to the MAPK/ERK pathway, which allows the internalization of Leishmania. Diagrams created with Biorender.