A Family of Toxoplasma gondii Genes Related to GRA12 Regulate Cyst Burdens and Cyst Reactivation

Abstract

Toxoplasma gondii causes a chronic infection that renders the immunocompromised human host susceptible to toxoplasmic encephalitis triggered by cyst reactivation in the central nervous system. The dense granule protein GRA12 is a major parasite virulence factor required for parasite survival during acute infection. Here, we characterized the role of four GRA12-related genes in acute and chronic stages of infection. While GRA12A, GRA12B, and GRA12D were highly expressed in asexual stage tachyzoites and bradyzoites, expression of GRA12C appeared to be restricted to the sexual stages. In contrast to deletion of GRA12 (Δgra12), no major defects in acute virulence were observed in Δgra12A, Δgra12B, or Δgra12D parasites, though Δgra12B parasites exhibited an increased tachyzoite replication rate. Bradyzoites secreted GRA12A, GRA12B, and GRA12D and incorporated these molecules into the developing cyst wall, as well as the cyst matrix in distinct patterns. Similar to GRA12, GRA12A, GRA12B, and GRA12D colocalized with the dense granules in extracellular tachyzoites, with GRA2 and the intravacuolar network in the tachyzoite stage parasitophorous vacuole and with GRA2 in the cyst matrix and cyst wall. Chronic stage cyst burdens were decreased in mice infected with Δgra12A parasites and were increased in mice infected with Δgra12B parasites. However, Δgra12B cysts were not efficiently maintained in vivo Δgra12A, Δgra12B, and Δgra12D in vitro cysts displayed a reduced reactivation efficiency, and reactivation of Δgra12A cysts was delayed. Collectively, our results suggest that a family of genes related to GRA12 play significant roles in the formation, maintenance, and reactivation of chronic stage cysts.IMPORTANCE If host immunity weakens, Toxoplasma gondii cysts recrudesce in the central nervous system and cause a severe toxoplasmic encephalitis. Current therapies target acute stage infection but do not eliminate chronic cysts. Parasite molecules that mediate the development and persistence of chronic infection are poorly characterized. Dense granule (GRA) proteins such as GRA12 are key virulence factors during acute infection. Here, we investigated four GRA12-related genes. GRA12-related genes were not major virulence factors during acute infection. Instead, GRA12-related proteins localized at the cyst wall and cyst matrix and played significant roles in cyst development, persistence, and reactivation during chronic infection. Similar to GRA12, the GRA12-related proteins selectively associated with the intravacuolar network of membranes inside the vacuole. Collectively, our results support the hypothesis that GRA12 proteins associated with the intravacuolar membrane system support parasite virulence during acute infection and cyst development, persistence, and reactivation during chronic infection.

Keywords: GRA12; Toxoplasma gondii; bradyzoite differentiation; chronic infection; cyst development; cyst matrix; cyst persistence; cyst reactivation; cyst wall; cysts; dense granule; dense granules; intravacuolar network; virulence.

FIG 1

GRA12-related proteins localize to tachyzoite dense granules, are secreted into the PV, and behave as a soluble and transmembrane PV proteins. (A) Extracellular tachyzoites were stained with anti-HA (α-HA) and anti-GRA5 (α-GRA5) antibody. Tachyzoite nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). Tachyzoites were located using differential interference contrast (DIC) microscopy and imaged by confocal microscopy. Number (n) of tachyzoites analyzed (n = 46 for GRA12A, n = 39 for GRA12B, n = 9 for GRA12C, n = 19 for GRA12D). Bars = 5 μm. (B to D) Human foreskin fibroblast (HFF) cells on coverslips were infected with 0.5 MOI tachyzoites of each strain. Vacuoles were located using DIC microscopy, and parasite and host nuclei were stained with DAPI and imaged by confocal microscopy. Vacuoles were stained with anti-HA and anti-GRA5 (B), anti-GRA2 (C), or anti-GRA1 (D). Number (n) of vacuoles analyzed, 6 to 42. Bars = 5 μm. (E) PV proteins were fractionated by ultracentrifugation into high-speed pellet (HSP) and high-speed supernatant (HSS) fractions. Proteins were resolved by electrophoresis, transferred to membranes, and probed with anti-HA rabbit antibody specific for HA-tagged GRA12A, GRA12B, or GRA12D, labeled with goat anti-rabbit IgG coupled to peroxidase, and visualized by chemiluminescence.

FIG 2

GRA12-related proteins localized to the cyst periphery and cyst matrix in immature cysts. (A, C, and E) HFF cells were infected with complemented GRA12-related HA-tagged protein strains, and in vitro cysts were differentiated for 1 day (1d) (A), 2 days (2d) (C), or 3 days (3d) (E). Cysts were located using DIC microscopy and imaged by confocal microscopy. The presence of bradyzoites was verified by locating parasite nuclei using DAPI stain and verifying that each parasite nucleus was surrounded by expression of cytosolic GFP (GFP⁺ bradyzoites). DAPI stains the host and parasite nuclei. Cysts were stained with DBA and anti-HA antibody. Panels show DAPI and GFP, DBA and GFP, DBA, anti-HA, DBA and anti-HA, and DIC (cyst periphery indicated by white arrowhead). Number (n) of cysts analyzed, 10. Bars = 5 μm. (B, D, and F) Fluorescence intensities of DBA, GRA12A, GRA12B, GRA12D, and GFP were measured at the cyst periphery and within the cyst (cyst interior) (n = 10 cysts) at 1 day (B), 2 days (D), or 3 days (F). Data plotted as the ratio mean of fluorescence intensity at the cyst periphery to the cyst interior ± standard error of the mean (SEM) (error bar). The numerical ratio for the mean fluorescence intensity is labeled. A one-way ANOVA test was used to reveal significance in mean fluorescence intensity [F(4, 36) = 68.30 at 1 day, 55.35 at 2 days, 115.5 at 3 days, P < 0.0001], and then a Tukey test was used to calculate P values in pairwise comparison to DBA and shown as follows: *, P < 0.05; ***, P < 0.001; ****, P < 0.0001; NS, not significant.

FIG 3

GRA12-related proteins localize to the cyst wall and cyst matrix in mature cysts. (A and B) HFF cells were infected with complemented GRA12-related HA-tagged protein strains, and in vitro cysts were differentiated for 7 days (A) or 10 days (B). Cysts containing GFP⁺ bradyzoites were located using DIC microscopy and imaged by confocal microscopy. Cysts were stained with DBA and anti-HA antibody (α-HA). Panels show DBA and GFP, anti-HA, DBA and anti-HA, and DIC (cyst periphery indicated by white arrowhead). Number (n) of cysts analyzed, 8 to 12. Bars = 5 μm. (C to H) Fluorescence intensity profiles of representative 7- and 10-day-old cysts, shown in panels A and B, were generated to quantify the location of DBA relative to GRA12A (C and D), GRA12B (E and F), or GRA12D (G and H) at the cyst wall. Dotted black lines define the cyst wall region. The dotted purple line indicates the middle of the cyst wall, which corresponds to the peak DBA fluorescence intensity. (I and J) Fluorescence intensities of DBA, GRA12A, GRA12B, GRA12D, and GFP were measured at the cyst periphery and within the cyst (cyst interior) at 7 days (n = 8 cysts) (I) or 10 days (n = 7 to 10 cysts) (J). Data plotted as the ratio mean of fluorescence intensity at the cyst periphery to the cyst interior ± SEM. The numerical ratio for the mean fluorescence intensity is labeled. A one-way ANOVA test was used to reveal significance in mean fluorescence intensity [F(4, 28) = 223.4 at 7 days, F(4,42) = 134.4 at 10 days, P < 0.0001], and then a Tukey test was used to calculate P values in pairwise comparison to DBA as follows: ****, P < 0.0001.

FIG 4

GRA12-related proteins colocalize with GRA2 in the cyst matrix and at the cyst wall. HFF cells were infected with complemented GRA12-related HA-tagged protein strains, and in vitro cysts were differentiated for 6 h (A), 1 day (B), 2 days (C), 3 days (D), 7 days (E), or 10 days (F). Cysts containing GFP⁺ bradyzoites were located using DIC microscopy and imaged by confocal microscopy. DAPI stained host and parasite nuclei. Cysts were stained with anti-GRA2 and anti-HA antibody. Panels show DAPI and GFP, anti-GRA2, anti-HA, anti-GRA2 and anti-HA, and DIC (cyst periphery/wall indicated by white arrowhead). GRA2 colocalized with GRA12-related proteins in 100% of cysts evaluated at 6 h (n = 11 to 13), with GRA12A and GRA12B in 100% of cysts evaluated at 1 day (n = 13 to 16) and 2 days (n = 6 to 12), and with GRA12D in 100% of cysts evaluated at 3 days (n = 17). Occurrence of GRA2 colocalization phenotype was 100% in all cysts analyzed, 1 day (n = 9 to 16), 2 days (n = 7 to 12), 3 days (n = 17 to 22), 7 days (n = 10 to 24), and 10 days (n = 12 to 15). Bars = 5 μm.

FIG 5

GRA12-related gene knockout mutants differentiate normally in vitro and exhibit defects in cyst burdens in vivo. (A) HFFs were infected with the indicated strains, and cysts differentiated in vitro cysts for 7 days. Cysts containing GFP⁺ bradyzoites were located using DIC microscopy and imaged by confocal microscopy. Cysts were stained with DBA. Number (n) of cysts analyzed, 6 to 8. Bars = 5 μm. (B) Cysts from each strain were analyzed to determine the ratio of DBA staining intensity at the cyst periphery relative to the cyst interior. Data were plotted as the average ratio mean of fluorescence intensity at the cyst periphery to the cyst interior ± SEM for each strain (n = 6 to 8 cysts). A one-way ANOVA analysis showed no significance in the mean fluorescence intensity ratio [F(3, 27) = 2.91, P = 0.0527]. (C) C57BL/6 mice were infected with 2 × 10² tachyzoites by intraperitoneal injection, and brain cysts were measured 3 weeks after infection. Data shown are cumulative results from two to six independent experiments for each strain tested and are shown as means ± SEM. PruΔku80 (three experiments; n = 12 mice), Δgra12A (two experiments; n = 8 mice), Δgra12A::GRA12A (two experiments; n = 6 mice), Δgra12B (two experiments; n = 8 mice), Δgra12B::GRA12B (four experiments; n = 13 mice), Δgra12D (six experiments; n = 21 mice), Δgra12D::GRA12D (three experiments; n = 10 mice). At least 10% of each brain was scanned for GFP⁺ cysts. A one-way ANOVA test revealed significance in the number of cysts per brain [F(6, 71) = 7.5, P < 0.0001], and a Tukey test was used to calculate P values in comparison to parental control PruΔku80 and the respective complemented strain as follows: **, P < 0.01; NS, not significant.

FIG 6

GRA12-related gene knockout mutants exhibit reduced cyst maintenance and cyst reactivation. (A) Reactivation assays were performed using in vitro cysts differentiated for 3 days after HFF cell cultures were infected with 2 × 10² tachyzoites of parental PruΔku80, Δgra12A::GRA12A or Δgra12A, Δgra12B::GRA12B or Δgra12B, and Δgra12D::GRA12D or Δgra12D. Percent reactivation (mean ± SEM) from three independent experiments is shown. A one-way ANOVA test revealed significance in reactivation [F(6, 20) = 8.642, P = 0.0001], and a Tukey test was used to calculate P values in comparison to parental control PruΔku80 and the respective complemented strain as follows: *, P < 0.05; **, P < 0.01. (B) C57BL/6 female mice (n = 6 to 13) were infected i.p. with 2 × 10² tachyzoites of PruΔku80, Δgra12B, or Δgra12B::GRA12B strain, and brain cyst burdens were measured 3 and 6 weeks postinfection (mean ± SEM). Data shown are cumulative results from two to four independent experiments for each strain. At 3 weeks, PruΔku80 (two experiments, n = 8), Δgra12B (two experiments, n = 8), and GRA12B (four experiments, n = 13) strains are shown. At 6 weeks, PruΔku80 (two experiments, n = 6), Δgra12B (two experiments, n = 6), and GRA12B (three experiments, n = 10) strains are shown. A one-way ANOVA test revealed significance in reactivation [F(5, 45) = 9.909, P < 0.0001], and a Tukey test was used to calculate P values in comparison to parental control PruΔku80 and the respective complemented strain as follows: **, P < 0.01; NS, not significant.