The Toxoplasma gondii-shuttling function of dendritic cells is linked to the parasite genotype

Abstract

Following intestinal invasion, the processes leading to systemic dissemination of the obligate intracellular protozoan Toxoplasma gondii remain poorly understood. Recently, tachyzoites representative of type I, II and III T. gondii populations were shown to differ with respect to their ability to transmigrate across cellular barriers. In this process of active parasite motility, type I strains exhibit a migratory capacity superior to those of the type II and type III strains. Data also suggest that tachyzoites rely on migrating dendritic cells (DC) as shuttling leukocytes to disseminate in tissue, e.g., the brain, where cysts develop. In this study, T. gondii tachyzoites sampled from the three populations were allowed to infect primary human blood DC, murine intestinal DC, or in vitro-derived DC and were compared for different phenotypic traits. All three archetypical lineages of T. gondii induced a hypermigratory phenotype in DC shortly after infection in vitro. Type II (and III) strains induced higher migratory frequency and intensity in DC than type I strains did. Additionally, adoptive transfer of infected DC favored the dissemination of type II and type III parasites over that of type I parasites in syngeneic mice. Type II parasites exhibited stronger intracellular association with both CD11c(+) DC and other leukocytes in vivo than did type I parasites. Altogether, these findings suggest that infected DC contribute to parasite propagation in a strain type-specific manner and that the parasite genotype (type II) most frequently associated with toxoplasmosis in humans efficiently exploits DC migration for parasite dissemination.

FIG. 1.

Transmigration of DC infected with type I, II, and III parasite strains in vitro. DC were preincubated with freshly egressed tachyzoites for 6 h, and transmigration was measured in a Transwell system as described in Materials and Methods. (A) Transmigration of human monocyte-derived DC. DC were infected (MOI of 5) with tachyzoites from 19 strains and clinical isolates, indicated by abbreviations, followed by incubation for 18 h in Transwell inserts. Infection rates of DC were 60 to 85%. Circles, triangles, and diamonds represent the mean values from two to four experiments performed in triplicate for each strain. Solid bars represent the mean values for each group. Asterisks indicate significant difference (P < 0.001; one-way ANOVA). ns, not significant. (B) Kinetics of transmigration for murine bone marrow-derived DC after incubation with various T. gondii strains (MOI of 5), tachyzoite lysate (10 μg/ml), or lipopolysaccharide (100 ng/ml) in Transwell inserts. Abbreviations indicate parasite strains used. A representative experiment from two is shown.

FIG. 2.

Transmigration of primary SI DC and PP DC after T. gondii infection in vitro. Intestinal DC were isolated from C57BL/6 mice as indicated in Materials and Methods. (A) Flow cytometric contour plots show expression of CD11c and MHC-II (I-A/I-E) or isotype controls, after purification. (B) DC from the SI and PP, positively selected for CD11c, were infected (MOI of 1 to 3) with freshly egressed GFP-transfected tachyzoites (type I, RH-LDM; type II, ME49-PTG) in culture medium (CM) for 4 h and subsequently stained for MHC-II expression. Contour plots show viable (PI^neg) MHC-II and GFP double-positive cells, i.e., infected DC. (C and D) Bar diagrams show the transmigration frequency (mean ± standard deviation) of CD11c⁺ SI DC and PP DC, respectively, after 4 h of incubation (MOI of 1) with type I (RH-LDM) or type II (ME49-PTG) parasites in Transwell inserts. Asterisks indicate significant differences (*, P < 0.05; **, P < 0.01; Student's t test). Data from a representative experiment performed in triplicate are shown.

FIG. 3.

Transmigration of primary human blood DC after T. gondii infection in vitro. Primary human myeloid DC were isolated from healthy blood donors as indicated in Materials and Methods. (A) A contour plot shows purified CD1c⁺ cells, where the gate includes the dominating viable cell population. A histogram shows the proportion of CD11c⁺ cells. (B) Immunofluorescence staining of human blood DC (phalloidin-Alexa Fluor 594 stain for top and middle panels; phalloidin-Alexa Fluor 488 stain for bottom panels) infected with T. gondii (type I GFP, RH-LDM; type II GFP, ME49-PTG; type II RFP, PRU). Overlay with DAPI (blue). Scale bar, 3 μm. (C) Transmigration frequency of primary human myeloid blood DC. DC were infected (MOI of 3) with tachyzoites (type I, RH-LDM; type II, ME49-PTG) for 2 h and incubated for 6 h in Transwell inserts. Transmigrated cells were quantified using a hematocytometer. Symbols represent DC from individual donors. An asterisk indicates significant difference (P < 0.05; paired t test).

FIG. 4.

Adoptive transfers of Toxoplasma-infected DC result in increased parasite loads. (A to D) C57BL/6 mice were inoculated i.p. with 10⁶ CFU of freshly egressed tachyzoites (filled triangles) (A and B) or with 10⁶ CFU of tachyzoite-infected DC (open circles) (C and D). Normalizations of inocula were performed as described in Materials and Methods. Parasite loads were quantified by a plaquing assay 16 h postinoculation. Representative strains were used for indicated genotypes (type I, RH-LDM; type II, ME49-PTG; type III, CTG). Asterisks indicate significant differences (*, P < 0.05; **, P < 0.01; one-way ANOVA). ns, not significant. (E and F) The relative differences in parasite loads in mice infected with type I, II, or III strains were calculated for spleens and MLN. The relation of the magnitude of parasite load for each strain was determined as the ratio of the mean parasite loads after inoculation of tachyzoite-infected DC to the mean parasite loads after inoculation of free tachyzoites. Mean parasite tissue loads (black lines) from individual mice from three to four separate experiments are shown.

FIG. 5.

DC are parasitized early during infection. C57BL/6 mice were inoculated i.p. with 2 × 10⁶ CFU of freshly egressed GFP-transfected type I (RH-LDM) and type II (ME49-PTG) tachyzoites. After 32 h, CD11c⁺ and CD11c⁻ populations were isolated, and infection frequencies were evaluated by flow cytometry and plaquing assay as indicated in Materials and Methods. (A) Density plots show CD11c⁺ cells from infected mice stained for MHC-II or an isotype control. Gates are set for infected (GFP⁺) cells. (B) Bar diagrams (spleen and MLN) show infection frequencies (means ± standard errors of the means) of infected CD11c⁺ (CD3⁻, CD19⁻, Gr1⁻, NK1.1⁻) and CD11c⁻ cells for inoculations with type I and type II parasites, respectively. Asterisks indicate significant differences (**, P < 0.01; paired t test). (C) Diagrams (spleen and MLN) show the relative numbers of infected DC (CD11c⁺, CD3⁻, CD19⁻, Gr1⁻, NK1.1⁻) per 10³ parasites (toxo) for inoculations with type I and type II parasites, respectively. Black lines indicate the mean values for each group of five mice (triangles and squares). Asterisks indicate significant differences (*, P < 0.05; Student's t test). Results of a representative experiment with five mice/group are shown.

FIG. 6.

Association with leukocytes characterizes the dissemination of type II parasites. (A) C57BL/6 mice were coinoculated i.p. with 2.5 × 10⁶ CFU GFP-expressing type I (RH-LDM) tachyzoites and 2.5 × 10⁶ CFU RFP-expressing type II (PRU) tachyzoites. Normalizations of inocula were performed as described in Materials and Methods. After 16 h, spleens were extracted, and cells were analyzed by flow cytometry. The gates in the plots include extracellular (e; FSC^low, PI^neg) and intracellular (i; FSC^high, PI^neg) parasites, respectively. Histograms show a disproportionate distribution of GFP⁺ cells (type I), with a dominance of the extracellular fraction. In contrast, RFP⁺ cells (type II) display an even distribution. FSC, forward scatter. (B) Bone marrow-derived DC were separately infected with type I (RH-LDM) and type II (PRU) tachyzoites in vitro. C57BL/6 mice were coinoculated i.p. with infected DC suspensions containing 2.5 × 10⁶ CFU GFP-expressing type I (RH-LDM) and 2.5 × 10⁶ CFU RFP-expressing type II (PRU) tachyzoites. Normalizations of inocula were performed as described in Materials and Methods. After 16 h, spleens were extracted, and cells were analyzed by flow cytometry. Histograms show a disproportionate distribution of GFP⁺ cells (type I) similar to that in panel A. In contrast, RFP⁺ cells (type II) show greater numbers and a strong dominance of the intracellular parasite fraction compared to the data displayed in panel A. (C and D) Bar diagrams show the mean (± standard deviation) percentage of intracellular parasites in the spleen and MLN, respectively, for the experimental setups described for panel A (Toxo) and panel B (Toxo-DC). Asterisks indicate significant differences (**, P < 0.01; ***, P < 0.001; paired t test and Student's t test for Toxo versus Toxo-DC). ns, not significant. Data from a representative experiment with eight mice/group are shown.

FIG. 7.

Enrichment of migratory Toxoplasma-infected DC in the spleen after i.p. inoculation. DC were labeled with cell tracker (BODIPY B22802) and separately infected with type I (GFP⁺, RH-LDM) or type II (RFP⁺, PRU) parasites in vitro before i.p. coinoculation in C57BL/6 mice. After 16 h, spleens were extracted and disseminated BODIPY-positive (BODIPY⁺) cells were assessed by flow cytometry after exclusion of nonviable cells (PI⁺). (A) Plot shows the distribution of GFP⁺ (type I), RFP⁺ (type II), and GFP⁻/RFP⁻ (uninfected) BODIPY⁺ cells inoculated. (B) Plots show the distribution of GFP⁺ (type I), RFP⁺ (type II), GFP⁺/RFP⁺ (double infected), and GFP⁻/RFP⁻ (uninfected) BODIPY⁺ cells in the spleen. (C) Mean distribution ratios (±standard deviations) of the populations displayed in panels A and B at the time point of i.p. inoculation (i.p. ratios) and spleen extraction (spleen ratios). Bar diagram shows the mean (±standard deviation) relative increase coefficient for each population in the spleen related to the population in the peritoneal cavity. Data from two independent experiments are shown.