Transepithelial migration of Toxoplasma gondii is linked to parasite motility and virulence

Abstract

After oral ingestion, Toxoplasma gondii crosses the intestinal epithelium, disseminates into the deep tissues, and traverses biological barriers such as the placenta and the blood-brain barrier to reach sites where it causes severe pathology. To examine the cellular basis of these processes, migration of T. gondii was studied in vitro using polarized host cell monolayers and extracellular matrix. Transmigration required active parasite motility and the highly virulent type I strains consistently exhibited a superior migratory capacity than the nonvirulent type II and type III strains. Type I strain parasites also demonstrated a greater capacity for transmigration across mouse intestine ex vivo, and directly penetrated into the lamina propria and vascular endothelium. A subpopulation of virulent type I parasites exhibited a long distance migration (LDM) phenotype in vitro, that was not expressed by nonvirulent type II and type III strains. Cloning of parasites expressing the LDM phenotype resulted in substantial increase of migratory capacity in vitro and in vivo. The potential to up-regulate migratory capacity in T. gondii likely plays an important role in establishing new infections and in dissemination upon reactivation of chronic infections.

Figure 1.

Characteristic migration patterns of the type I (RH) and the type II (PTG) strains and the LDM1 clonal line. Parasites were cultured on HFF monolayers with an overlay of 0.75% agarose in culture medium as indicated in Materials and Methods. Migration was assessed 24 h after egress from the original infected cell. Characteristic dispersion patterns of PTG strain (A) and RH strain (B, arrows indicate parasites migrating >110 μm) and the clone LDM1 (C), respectively. Scale bar = 20 μm. Type I (RH) parasites migrating >110 μm were termed LDM and surpassed by >50% the longest migrated distances by type II (PTG) parasites. Graphics show the relative distribution of migrated distances (μm) by different parasite populations calculated using image analysis software as described in Materials and Methods (PTG [A′], n = 306; RH [B′], n = 296; LDM1 [C′], n = 303). The y-axis indicates the relative portion (%) of each category related to the total of the population examined. The distribution of distances migrated by individual parasites was analyzed with the Shapiro-Wilk test of normality, showing that the PTG population approached a normal distribution (W = 0.94; P ≤ 0.0001), whereas RH (W = 0.81; P ≤ 0.0001), and LDM1 (W = 0.76; P ≤ 0.0001) were not normally distributed.

Figure 2.

Transmigration across polarized cell monolayers and extracellular matrix by the type I strain (RH) and the type II strain (PTG). Transmigration assays across polarized MDCK monolayers (A and B) or Matrigel^® (C and D) were performed as indicated in Materials and Methods. Quantification was done by counting CFU (intra-cellular vacuoles containing parasites) in an underlying HFF monolayer (A and C) or by colorimetric detection of β-galactosidase (β-gal) activity immediately after transmigration using the parasite strain RH-lacZ (B and D). Results are means from one representative experiment done in triplicate. (A) Kinetics of transmigration across polarized MDCK cell monolayers revealed differences between type I (RH) and type II (PTG) strains. (B) Heat inactivation (Dead) of type I (RH-lacZ) strain resulted in inhibition of transmigration across polarized MDCK cell monolayers. Parasite culture supernatant was used as control. (C) Kinetics of transmigration across matrigel® showed differences between type I (RH) and type II (PTG) strains. Transmigration was totally inhibited in the presence of 1 μM cytochalasin D (*). (D) Inhibition of migration across Matrigel^® by 1 μM cytochalasin D (CYT D). Parasite culture supernatant was used as control.

Figure 3.

Invasion, motility, and migratory characteristics of the RH (type I) and PTG (type II) strains and the RH-derived clone LDM1. (A) Enhanced long distance migration was observed in the LDM1 clone vs. RH, but was absent in the PTG strain. Frequency of parasites migrating >110 μm (LDM) on an HFF monolayer under agarose as indicated in Materials and Methods. Results are mean (± SD) from three independent experiments. (B) Significantly enhanced transmigration capacity was observed in the clone LDM1 vs. RH (P ≤ 0.005, Student's t test). Frequency of parasites (± SD) transmigrating across polarized MDCK cell monolayers was assessed as indicated in Materials and Methods. (C) Plaquing assays on HFF monolayers showed very similar viability for RH and LDM1, and a slightly reduced viability for PTG. Plaquing assays were evaluated at 5–7 d after inoculation and the number of plaques generated was expressed as a percentage of parasites initially added. Average results are shown from three independent experiments (± SD). (D) Gliding of the RH strain was similar to that of the LDM1, but significantly shorter trail lengths were observed for PTG vs. RH (P ≤ 0.0005). Gliding trail lengths were measured as relative parasite body lengths (5–7 μm) and the average trail length from three independent experiments (± SD) is shown.

Figure 4.

Ex vivo transmigration of T. gondii (RH) across intestinal mucosa. Transmigration assays were performed on mouse ileum in a modified Ussing 2-chamber system as described in Materials and Methods. (A) Section of distal villus from mouse ileum showing tachyzoites (red) invading epithelial cells. Immunohistochemistry was performed with polyclonal rabbit anti-RH serum and goat anti–rabbit antibody conjugated to Alexa 594. Epithelial cells were counterstained with fluorescein-conjugated wheat germ agglutinin and nuclear staining was with DAPI. Scale bar = 5 μm. (B) Close up of epithelial cell invasion by RH tachyzoites (red). In the top part of the micrograph is the apical side of the epithelial cell layer and the lower part shows the limit with the basal lamina (lamina propria). IF was performed as in A. Scale bar = 5 μm. (C and C′) Penetration of tachyzoites to subepithelial cell layers in mouse ileum. (C) Arrows indicate parasite vacuoles in the lamina propria. IF was performed as in A. Scale bar = 5 μm. (C′) Intracellular parasite vacuoles were stained with diaminobenzidine tetrahydrochloride as indicated in Materials and Methods. Epithelial cells were counterstained with hematoxylin. Scale bar = 10 μm. (D) Penetration of RH tachyzoites (green) to the submucosa (white arrows) and invasion of endothelial cells in the vasculature (red arrow). IF was performed with polyclonal rabbit anti-RH serum and goat anti-rabbit antibody conjugated to Alexa 488. Nuclear staining was with DAPI. Scale bar = 20 μm.

Figure 5.

Dissemination of RH (type I) and LDM1(RH clone) parasites in vivo. The number of viable parasites retrieved in the spleens of infected mice is plotted separately and the means are indicated by the horizontal bars. Parasite tissue burden was quantified daily after a challenge of 10⁴ parasites intraperitoneally as indicated in Materials and Methods. Significantly enhanced dissemination was observed in the clone LDM1 vs. RH at day 2 after infection (P ≤ 0.005, Student's t test) and at day 3 (P ≤ 0.001), whereas, by day 4, differences in parasite load were nonsignificant (P ≥ 0.05). The results shown are the mean of two independent experiments consisting of three to four mice per time point.

Figure 6.

Differences in transmigration capacity between type I, type II, and type III strains and expression of the LDM-phenotype. Virulent and nonvirulent parasite strains and clinical isolates were classified by genotype as described (references and 6). Assessment for the LDM-phenotype (parasite migration >110 μm) was performed on HFF cell monolayers under agarose as indicated in Materials and Methods. Graphics show expression (+) of the LDM-phenotype in all type I parasite lines tested and absence (−) of the LDM-phenotype in type II and type III lines. Transmigration of parasites across polarized MDCK cell monolayers was assessed by plaque formation (CFU) as described in Materials and Methods. The transmigration frequency for each strain was calculated as the number of transmigrated parasites (CFU)/1,000 parasites added and represents the mean from three independent experiments (± SD). Significant differences in frequency of transmigration were observed between strains (P ≤ 0.001 for type I vs. type II; P ≤ 0.005 for type I vs. type III, Mann-Whitney U-test) with transmigration frequency differences superior to three orders of magnitude (10³).