Entamoeba lysyl-tRNA synthetase contains a cytokine-like domain with chemokine activity towards human endothelial cells

Abstract

Immunological pressure encountered by protozoan parasites drives the selection of strategies to modulate or avoid the immune responses of their hosts. Here we show that the parasite Entamoeba histolytica has evolved a chemokine that mimics the sequence, structure, and function of the human cytokine HsEMAPII (Homo sapiens endothelial monocyte activating polypeptide II). This Entamoeba EMAPII-like polypeptide (EELP) is translated as a domain attached to two different aminoacyl-tRNA synthetases (aaRS) that are overexpressed when parasites are exposed to inflammatory signals. EELP is dispensable for the tRNA aminoacylation activity of the enzymes that harbor it, and it is cleaved from them by Entamoeba proteases to generate a standalone cytokine. Isolated EELP acts as a chemoattractant for human cells, but its cell specificity is different from that of HsEMAPII. We show that cell specificity differences between HsEMAPII and EELP can be swapped by site directed mutagenesis of only two residues in the cytokines' signal sequence. Thus, Entamoeba has evolved a functional mimic of an aaRS-associated human cytokine with modified cell specificity.

Figure 1. Bioinformatic analysis of EELP.

(A) Protein alignment of EELP sequences from different Entamoeba species. (B) Protein alignment of mature EMAPII from Homo sapiens (HsEMAPII), C-terminal EMAPII-like domain of human tyrosyl-tRNA synthetase (HsCtYRS) and C-terminal domain of Entamoeba lysyl-tRNA synthetase (EhCtKRS). The boxed sequence corresponds to the heptapeptide migration motif. (C) Phylogenetic analyses of EELP protein sequences from different Entamoeba species. Numbers on each branch correspond to bootstrap values in the distance consensus tree.

Figure 2. EhKRS and EhMRS are up-regulated by inflammation signals.

(A) qRT-PCR of EhKRS (closed bars) and EhMRS (open bars) gene expression relative to Entamoeba histolytica threonyl-tRNA synthetase (EhTRS). Trophozoites stimulated with 100 ng/ml LPS or 100 ng/ml TNF-α or co-cultured with endothelial cells (HUVEC), colonic epithelial cells (DLD1), or primary monocytes (Monos) pre-activated (Act) or not with 100 ng/ml LPS for 6 h. Values are relative to non-stimulated amoebae (ctl) and depicted as mean ± SD from at least two independent experiments performed in triplicate. (B) Quantification of EhKRS protein expression in trophozoites stimulated for 6 hours (open bars) and 24 hours (closed bars) with 100 ng/ml LPS or 100 ng/ml TNF-α or co-cultured with monocytes (Monos) or monocytes previously activated with LPS for 6 hours (Act Monos). The fluorescence units were normalized to non-stimulated (ctl) trophozoites and depicted as mean ± SD from three different experiments. The number of cells counted per each condition was n = 40. (*** p<0.0001 vs ctl). Inserts show microphotographs of immunodetection of EhKRS protein (α-NtEhKRS) in trophozoites at 24 hours of stimulation. (C) Immunolocalization of EhKRS in Entamoeba trophozoites using affinity purified antibodies against EhKRS N-terminal (α-NtEhKRS) and C-terminal (α-EELP) domains. Controls of secondary antibodies merged with DAPI are shown in bottom row. (D) EhKRS cellular localization was evaluated by immunoblot analysis of E. histolytica lysates, marked as T (whole cell lysate), C (cytoplasmic fraction), and MB (membrane bound fraction). Specific antibodies of cytoplasmic fraction (α-actin) and membrane fraction (α-LGL) were used. α-EELP antibody recognizes a full length protein (top panel) and a C-terminal cleaved product (bottom panel). α-NtEhKRS (1∶250), α-EELP (1∶50), α-LGL (1∶1000) and α-actin (1∶1000) antibody dilutions were used.

Figure 3. EhKRS protease processing.

Immunoblots analysis with α-EELP antibody of recombinant EhKRS protein digestions using Entamoeba lysates (A) or human leukocyte elastase (B). (A) 1 µg recombinant EhKRS was incubated at 37°C for 30 minutes without E. histolytica crude extract (Eh CE; lane 1) or with 0.5 µg Eh CE (lane 2–4); 1 µg Eh CE (lane 5–7); 5 µg Eh CE (lane 8–10) or 5 µg Eh CE plus protease inhibitors (lane 11–13) for 5, 15 or 30 minutes. (B) Digestion of 1 µg recombinant EhKRS protein with elastase for 30 minutes at 37°C. (C) EhMRS protease processing. Digestion of 1 µg recombinant EhMRS at 5, 15 and 30 minutes with 1 µg of Eh CE (lane 2–4) or 5 µg Eh CE (lane 5–7). Recombinant EhMRS control without Eh CE (lane 1). Digestion products were detected by immunoblot using α-EELP antibody. Arrow shows EELP product resulting from recombinant EhKRS or EhMRS. Asterisk denotes recombinant EELP with 6 His tag plus 36 amino acids at the N-terminal. Boxed sequence corresponds to the N-terminal sequence of EELP after digestion from EhKRS by elastase, as determined by Edman degradation.

Figure 4. Effect of EELP on host cell migration.

(A) Endothelial cell and (B) monocyte chemotaxis towards different EMAPII bearing proteins. VEGF (1 nM) was used as a common chemoattractant for both cells. Data are the mean ± SEM for at least three different experiments. Inserts show a representative photograph of cell migration. Pore membranes for HUVEC and monocyte migration were of 8 µm diameter and 5 µm diameter, respectively. Cells are stained in purple. Migration is plotted as Migration Index (MI; number of cells migrating in each condition/number of cells migrating in basal medium). Ctl, basal medium in the lower well; VEGF, vascular endothelial grothw factor; rhEMAPII, recombinant human EMAPII; HsCtYRS, C-terminal EMAPII-like domain of human tyrosyl-tRNA synthetase; EELP, Entamoeba EMAPII-like polypeptide; EhKRS, lysyl-tRNA synthetase of Entamoeba; EhKRSΔCt, EhKRS depleted of EELP domain; HumEELP, humanized EhCtKRS (see Figure S1 and text). *** p<0.0001 vs ctl.

Figure 5. Calcium mobilisation and cellular internalization of EELP.

(A) Calcium flux was observed by Fluo-4 fluorescence in monocytes stimulated with 10 nM rhEMAPII, HsCtYRS, EhKRS, EhKRSΔCt, EELP, and HumEELP. An increase in fluorescence proves increase in cytoplasmic calcium and was monitored as indicated in Methods. RFU, relative fluorescence units; the rest of abbreviations are the same as listed in Figure 4. Arrow indicates when stimuli were added. (B) EELP and HumEELP are internalized in endothelial cells and monocytes. These cells were incubated for 1 hour with 100 nM of His tagged purified proteins. Cell extracts were analyzed by western blot with α-His antibody. Control of protein loading was evaluated with α-tubulin antibody. DmMp20; recombinant Drosophila melanogaster Muscle protein of 25 kDa (see also Figure S1A).

Figure 6. The effect of conditioned HUVEC medium on monocyte migration.

(A) Number of monocytes that have migrated to 1 nM rhEMAPII or supernatants from HUVEC cultured with EELP for 1, 4 or 14 hours in the lower wells of a 24-transwell Boyden plate or (B) that have migrated to 1 nM rhEMAPII, 1 nM EELP, supernatants from HUVEC cultured for 14 hours with 1 nM EhKRS (SN EhKRS), 1 nM EhKRSΔCt (SN EhKRSΔCt) 1 nM EELP (SN EELP) or 1 nM EhMRS (SN EhMRS). Values are plotted as Mean ± SEM of three independent experiments.