A sialic acid-binding protein in Toxoplasma gondii contains a conserved globular domain in apicomplexan parasites

Abstract

Background: Apicomplexan protozoans employ an intricate invasion mechanism involving dynamic interactions with host cells, characterized by sequential secretion of adhesins and lectins. Our laboratory previously identified TgSABP1, a novel Toxoplasma gondii adhesin, demonstrating specific binding affinity for sialic acid (SA) receptors on host cell surfaces. However, the structural determinants governing SA recognition by this adhesin remain undefined.

Methods: Three-dimensional structural predictions of TgSABP1 and homologous proteins were generated using AlphaFold2. Bio-layer interferometry (BLI) quantified the binding affinities between the recombinant proteins and ligands. Competitive BLI assays evaluated small molecules that potentially inhibit the TgSABP1-sialyllactose interactions. Molecular docking simulations employing AutoDock Vina software elucidated ligand-binding site interactions. In vitro invasion inhibition assays were performed to assess the therapeutic potential of lead compounds targeting TgSABP1 against T. gondii tachyzoites.

Results: AlphaFold2 structural predictions revealed that TgSABP1 and its homologues contain a conserved globular domain (pLDDT > 90) with significant structural homology (with root-mean-square deviation [RMSD] < 4 Å) to a Plasmodium falciparum invasion-related protein PfIMP2 (PDB: 5LG9). BLI quantification demonstrated the micromolar binding affinities of the recombinant proteins for 3'-sialyllactose-polyacrylamide (PAA) and 6'-sialyllactose (6'SL)-PAA. Intriguingly, although recombinant TgSABP1 showed stronger lactose binding (K_D = 0.02 ± 0.01 M) compared to SA (K_D = 2.07 ± 0.45 M), only the latter exhibited an inhibition on the TgSABP1-6'SL-PAA interaction. Virtual screening of Food and Drug Administration (FDA)-approved compounds identified eltrombopag as a high-affinity molecule (ΔG_bind = -8.3 kcal/mol) targeting the SA-binding pocket in TgSABP1. Functional validation demonstrated that eltrombopag effectively blocked the TgSABP1/6'SL-PAA interaction and significantly decreased host cell invasion of T. gondii tachyzoites.

Conclusions: Our study reveals a conserved globular domain of apicomplexan parasites as a novel SA-binding domain. Structural and functional characterization demonstrates its critical role in mediating TgSABP1-host cell interactions. Targeting this SA-binding pocket with eltrombopag effectively decreased T. gondii tachyzoite invasion, suggesting its therapeutic potential as an anti-invasion target. These findings not only elucidate a conserved mechanism underlying host receptor recognition in apicomplexans, but also establish a structural framework for the rational design of broad-spectrum inhibitors targeting invasion-related lectin domains.

Keywords: Toxoplasma gondii; Globular domain; Sialic acid.

Fig. 1

Analysis of three-dimensional (3D) structures of TgSABP1 and its homologues in T. gondii and Plasmodium parasites and their binding affinity to sialylglycopolymers. A–F A credible globular domain was present in the structures of TgSABP1, PfIMP2, PbIMP2, PfIMP1, TgIMP2.1, and TgIMP1. G The affinities of the recombinant proteins to 3′SL-PAA (red) and 6′SL-PAA (blue) were determined by BLI assay. K_D value represents the equilibrium dissociation constant. ns, no significant difference, *P < 0.05 (unpaired Student’s t-test). All data are presented as the means ± SD (n = 3)

Fig. 2

Specific binding of recombinant PfIMP2 to human erythrocytes. A Coomassie-stained SDS-PAGE of purified GST-tagged PfIMP2 (~ 44 kDa). B SDS-PAGE showing the same amounts of proteins were loaded in the gel after RBC-binding. C Western blot showing the binding of GST-PfIMP2 to RBC in 2.5 and 5 μM concentrations with two replicates. D Immunofluorescence micrographs demonstrating specific binding of GST-tagged PfIMP2 to human erythrocyte, scale bar = 20 μm. Images representative of three biological replicates

Fig. 3

Inhibition analysis of SA on the binding of the recombinant TgSABP1 to 6′SL-PAA with lactose as a control. A, B Real-time binding sensograms for SA (A) vs. lactose (B). The K_D value represents the equilibrium dissociation constant. Data are presented as means ± SD (n = 3). The R² value represents the fitting degree of the kinetic curve. C Comparative K_D values of SA and lactose with TgSABP1. **P < 0.01 (unpaired Student’s t-test). D, E Competitive binding profiles with SA (D) and lactose (E) in TgSABP1-6′SL-PAA interactions

Fig. 4

Molecular docking analysis of TgSABP1–SA interactions. A 6′SL (red sticks) forms eight hydrogen bonds (blue lines) with seven residues in TgSABP1. B SA (green sticks) binds to the TgSABP1 with one salt bridge (yellow dashes), one hydrophobic interaction (gray dashes) interaction, and five hydrogen bonds (blue lines). C Lactose (yellow sticks) forms seven hydrogen bonds (blue lines) with five residues of TgSABP1. D Structural overlap showing the relative positions of 6′SL (red), SA (green), and lactose (yellow) on the surface of TgSABP1 (cyan)

Fig. 5

Docking and interaction validation between oseltamivir, eltrombopag, and TgSABP1. A Oseltamivir (blue) forms two hydrogen bonds (blue) and one hydrophobic interaction (gray dashes) with TgSABP1. B Eltrombopag (gray) formed five hydrogen bonds (blue lines) and five hydrophilic interactions (gray dashes) with TgSABP1. C Structural overlap showing the relative positions of 6′SL (red), oseltamivir (blue), and eltrombopag (light blue) on the surface of TgSABP1 (cyan). D, E Competitive binding profiles with oseltamivir phosphate (D) and eltrombopag (E) on TgSABP1-6′SL-PAA interactions. F, G Real-time binding sensograms for oseltamivir phosphate (F) and eltrombopag (G) with recombinant TgSABP1. The K_D value represents the equilibrium dissociation constant. The R² value represents the fitting degree of the kinetic curve. Data are presented as means ± SD (n = 3)

Fig. 6

Cytotoxicity and anti-invasion activity of eltrombopag against T. gondii. A Cytotoxicity of eltrombopag on Vero cells after 2 h and 24 h exposure was detected by CCK-8 assay. B Cell viability curves of eltrombopag on Vero cells exposed to eltrombopag for 2 h and 24 h. C Invasion efficiency quantification of eltrombopag was calculated through PVs/DAPI⁺ cells. D The number of parasites per vacuole was determined in the eltrombopag-treated and DMSO groups. All data are expressed as the mean ± SD (n = 5–7). One-way analysis of variance (ANOVA) test for panels A–C and two-way ANOVA test with Dunnett’s multiple comparisons test for panel D. ns, not significant, *P < 0.05, **P < 0.01, ***P < 0.001