Crystal structure of calcium binding protein-5 from Entamoeba histolytica and its involvement in initiation of phagocytosis of human erythrocytes

Abstract

Entamoeba histolytica is the etiological agent of human amoebic colitis and liver abscess, and causes a high level of morbidity and mortality worldwide, particularly in developing countries. There are a number of studies that have shown a crucial role for Ca2+ and its binding protein in amoebic biology. EhCaBP5 is one of the EF hand calcium-binding proteins of E. histolytica. We have determined the crystal structure of EhCaBP5 at 1.9 Å resolution in the Ca2+-bound state, which shows an unconventional mode of Ca2+ binding involving coordination to a closed yet canonical EF-hand motif. Structurally, EhCaBP5 is more similar to the essential light chain of myosin than to Calmodulin despite its somewhat greater sequence identity with Calmodulin. This structure-based analysis suggests that EhCaBP5 could be a light chain of myosin. Surface plasmon resonance studies confirmed this hypothesis, and in particular showed that EhCaBP5 interacts with the IQ motif of myosin 1B in calcium independent manner. It also appears from modelling of the EhCaBP5-IQ motif complex that EhCaBP5 undergoes a structural change in order to bind the IQ motif of myosin. This specific interaction was further confirmed by the observation that EhCaBP5 and myosin 1B are colocalized in E. histolytica during phagocytic cup formation. Immunoprecipitation of EhCaBP5 from total E. histolytica cellular extract also pulls out myosin 1B and this interaction was confirmed to be Ca2+ independent. Confocal imaging of E. histolytica showed that EhCaBP5 and myosin 1B are part of phagosomes. Overexpression of EhCaBP5 increases slight rate (∼20%) of phagosome formation, while suppression reduces the rate drastically (∼55%). Taken together, these experiments indicate that EhCaBP5 is likely to be the light chain of myosin 1B. Interestingly, EhCaBP5 is not present in the phagosome after its formation suggesting EhCaBP5 may be playing a regulatory role.

Figure 1. Structure of EhCaBP5.

(A) Overall structure of Entamoeba histolytica calcium binding protein-5 (EhCaBP5) at 1.9 Å resolution, shown as a ribbon diagram. All of the alpha helices (α1 to α8) are in blue, the small antiparallel β-sheet (β1 and β2) is represented in pink, and the bound Ca²⁺ is shown as a red sphere. (B) Ca²⁺ coordination of EhCaBP5: The unconventional mode of Ca²⁺binding by the EF hand motif of EhCaBP5 (see text for details).

Figure 2. Structural comparison.

A) Alignment of the Ca²⁺ bound EF-hand motif of EhCaBP5 (blue) with that of Potato CaM (pink, pdb code 1RFJ). The r.m.s.d. of alignment is 2.38 Å. The interhelical angles of the EF-hand motifs of EhCaBP5 and CaM are 64.3 Å and 89.3 Å, respectively, indicative in turn of the closed and open conformations. (B) Alignment of the Ca²⁺ bound EF-hand motif of EhCaBP5 (blue) with that of squid ELC (yellow, pdb code 1QVI), physarum ELC (green, pdb code 2BLO), and a mutant CaM (pink, pdb code 1Y6W). The rmsd's of the alignments are1.6 Å, 1.3 Å and 1.08 Å, respectively.

Figure 3. Binding of EhCaBP5 with IQ motif.

SPR Sensograms representing interaction between myosin 1B IQ motif peptide (IQKAWKGYRNRKR) and EhCaBP5 in presence of Ca²⁺ (3A) and absence of Ca²⁺ (presence of 5 mM EGTA) (3B). Different concentrations of CaBP5 were injected onto the sensor chip to calculate the dissociation constant (Kd). BSA was also injected at concentration 750 nM (Fig. 3B, grey colour).

Figure 4. Interaction of EhCaBP5 with myosin 1B.

Total (800 µg) E. histolytica lysate was incubated with Sepharose-anti-CaBP5 antibody conjugate for 6 h at 4°C with shaking. The beads were then washed and the bound material was then eluted and analysed by western blotting followed by immunostained with anti-myosin 1B antibody raised in rabbit. The blot was reprobed with anti-CaBP5 antibody raised in mice. The total input lysate was also probed for the presence of EhCaBP5 and myosin 1B by their respective antibodies.

Figure 5. Immunolocalization of EhCaBP5 in E. histolytica.

Trophozoites grown for 48 h were transferred to pre-warmed coverslips for 10 min at 37°C. The cells were then fixed with 3.7% paraformaldehde/PBS, permeabilized with 0.1% triton X-100/PBS and then stained with anti-EhCaBP5 antibody followed by secondary antibody Alexa 488. Hoechst was used to stain the nuclei. Bar represents 10 µm. Magnification 60X.

Figure 6. Distribution of EhCaBP5, myosin 1B and actin in E. histolytica during erythrophagocytosis.

(A) Co- localization of EhCaBP5 and myosin 1B in E. histolytica cells during erythrophagocytosis. Cells were grown for 48 h and incubated with RBCs for 10 min at 37°C. The cells were then fixed and immunostained with anti- EhCaBP5 and anti-myosin 1B antibodies followed by Alexa-488 (green) and Alexa-555 (red) secondary antibodies. Arrow head depicts the co- localization of CaBP5 and myosin 1B in the phagocytic cup (upper panel) and an asterisk mark shows the absence of EhCaBP5 in the phagosome. (B) Co-localization of EhCaBP5 with F-actin. Trophozoites were stained with anti-CaBP5 antibody and TRITC-phalloidin (red) was used to stain the F-actin. The secondary antibody used for EhCaBP5 was Alexa-488 (green). Bar represents 10 µm. (DIC, differential interference contrast).

Figure 7. Down regulation of EhCaBP5 reduces the rate of phagocytosis.

(A) Schematic representation of sense and anti-sense construct. EhCaBP5 was cloned in sense and anti-sense orientation in BamH1 and Kpn1 site of pEhHyg-TetR-O-CAT vector. (B, C) Western blot analysis of anti-sense (B) and Sense EhCaBP5 (C). Thirty microgram of lysate from sense and forty microgram of anti-sense cells in presence and absence of tetracycline were loaded to each lane. HM1 was used as a positive control and coactosin as a loading control. (D) RBC uptake assay performed with cells expressing sense and anti-sense constructs in the presence and absence of tetracycline. The experiment was repeated three times independently in triplicates. Comparisons were made with respect to cells vector alone or cells without addition of tetracycline. Statistical significance was determined by paired- t test. P-values for ** is P<0.001, #P>0.05.

Figure 8. Erythrophagocytosis of cells overexpressing EhCaBP5 antisense and sense RNA constructs in the presence and the absence of tetracycline.

Amoebic cells overexpressing EhCaBP5S and EhCaBP5AS with and without tetracycline were incubated with RBC for 10 min. The cells were then stained with Alexa 488(EhCaBP5) and myosin 1B (red). Solid arrow depicts the phagocytic cup and asterisk shows the phagosome. (scale bar, 10 µm; DIC, differential interference contrast).

Figure 9. Erythrophagocytosis of amoebic cells overexpressing EhCaBP5 antisense RNA in presence of tetracycline.

Amoebic cells were incubated with RBC for 15 min. The cells were then stained with Alexa 488 (EhCaBP5) and myosin 1B (red). Solid arrow heads represent the closure of phagocytic cup. (scale bar, 10 µm; DIC, differential interference contrast).

Figure 10. Protein-peptide Interaction.

A) Rosetta-docked lowest energy model of EhCaBP5 (grey ribbon) bound with an IQ motif (yellow ribbon for backbone and sticks for side chains). B) Schematic representation of the contacts between EhCaBP5 and the peptide. The interaction shown by dotted orange lines are the non-bonded interactions, in which the width of the striped line is proportional to the number of atomic contacts. A blue line between any two residues indicates hydrogen bond interactions. C) Superimposition of the native EhCaBP5 crystal structure on the model of peptide-bound EhCaBP5 shows a predicted 15° rotation of the C-lobe relative the N-lobe upon binding of the peptide.

Figure 11. Model Validation.

SPR Sensogram represents the binding of Native IQ motif peptide (green) and Mutant IQ-M2 peptide (red) to EhCaBP5 immobilized surface.