Secretion of Rhoptry and Dense Granule Effector Proteins by Nonreplicating Toxoplasma gondii Uracil Auxotrophs Controls the Development of Antitumor Immunity

Abstract

Nonreplicating type I uracil auxotrophic mutants of Toxoplasma gondii possess a potent ability to activate therapeutic immunity to established solid tumors by reversing immune suppression in the tumor microenvironment. Here we engineered targeted deletions of parasite secreted effector proteins using a genetically tractable Δku80 vaccine strain to show that the secretion of specific rhoptry (ROP) and dense granule (GRA) proteins by uracil auxotrophic mutants of T. gondii in conjunction with host cell invasion activates antitumor immunity through host responses involving CD8α+ dendritic cells, the IL-12/interferon-gamma (IFN-γ) TH1 axis, as well as CD4+ and CD8+ T cells. Deletion of parasitophorous vacuole membrane (PVM) associated proteins ROP5, ROP17, ROP18, ROP35 or ROP38, intravacuolar network associated dense granule proteins GRA2 or GRA12, and GRA24 which traffics past the PVM to the host cell nucleus severely abrogated the antitumor response. In contrast, deletion of other secreted effector molecules such as GRA15, GRA16, or ROP16 that manipulate host cell signaling and transcriptional pathways, or deletion of PVM associated ROP21 or GRA3 molecules did not affect the antitumor activity. Association of ROP18 with the PVM was found to be essential for the development of the antitumor responses. Surprisingly, the ROP18 kinase activity required for resistance to IFN-γ activated host innate immunity related GTPases and virulence was not essential for the antitumor response. These data show that PVM functions of parasite secreted effector molecules, including ROP18, manipulate host cell responses through ROP18 kinase virulence independent mechanisms to activate potent antitumor responses. Our results demonstrate that PVM associated rhoptry effector proteins secreted prior to host cell invasion and dense granule effector proteins localized to the intravacuolar network and host nucleus that are secreted after host cell invasion coordinately control the development of host immune responses that provide effective antitumor immunity against established ovarian cancer.

Fig 1. Immunity to T. gondii does not diminish the potency of the antitumor response stimulated by uracil auxotrophs.

(A) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with phosphate buffered saline (PBS) or mice were vaccinated i.p. with tachyzoites (the acute replicative form of T. gondii) of uracil auxotrophs (OMP [13] or CPS [5]) at 8, 20, and 32 d after tumor challenge (the three-dose treatment schedule). (B) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS or were vaccinated once (8 d), twice (8, 20 d), three times (8, 20, 32 d), or five times (8, 20, 32, 44, 56 d) i.p. with tachyzoites of uracil auxotrophs. (C) Groups of C57BL/6 mice were vaccinated (VAC) with uracil auxotrophs to establish protective immunity or mice were treated with PBS (naive). Twelve months later ID8DV tumors were established in vaccinated or age matched naive mice and tumors were treated with PBS or were vaccinated with uracil auxotrophs using the three-dose treatment schedule. Data is representative of at least two independent experiments. ns was not significant, *p<0.05, **p<0.01, ***P<0.001, ****p<0.0001.

Fig 2. IL-12, CD8α⁺ dendritic cells, and IFN-γ are essential for stimulation of the antitumor response elicited by uracil auxotrophs.

ID8DV ovarian tumors were established and groups of mice were treated with PBS or were vaccinated i.p. with tachyzoites of uracil auxotrophs (OMP) using the three-dose treatment schedule. (A) Wild-type or IL-12p35^-/- C57BL/6 mice. (B) Wild-type or MyD88^-/- C57BL/6 mice. (C) Wild-type or Batf3^-/- C57BL/6 mice. (D) Wild-type or IFN-γ^-/- C57BL/6 mice. Data is representative of two independent experiments. ns was not significant, **p<0.01, ***P<0.001, ****p<0.0001.

Fig 3. CD8⁺ and CD4⁺ T cells are required for the antitumor response.

(A) ID8DV ovarian tumors were established in wild-type or CD8^-/- C57BL/6 mice and groups of mice were treated with PBS or were vaccinated i.p. with tachyzoites of uracil auxotrophs (OMP) using the three-dose treatment schedule. (B-C) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS or were vaccinated i.p. with tachyzoites of uracil auxotrophs (OMP) on day 8 and day 20. Groups of mice were treated with isotype control antibody, or (B) αCD8 antibody, or (C) αCD4 antibody on days 7, 8, 11, 19, 20, 23 after tumor challenge. Data is representative of two independent experiments. ns was not significant, **p<0.01, ***P<0.001, ****p<0.0001.

Fig 4. Parasite extracts and heat killed uracil auxotrophs fail to stimulate antitumor responses or IFN-γ production in the tumor microenvironment.

(A) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS, vaccinated i.p. with tachyzoites of uracil auxotrophs (OMP or CPS) or vaccinated i.p. with heat killed (HK) tachyzoites of uracil auxotrophs (OMP or CPS) using the three-dose treatment schedule. (B) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS, vaccinated i.p. with tachyzoites of uracil auxotrophs (CPS) or vaccinated i.p. with excreted/secreted protein (ESA) extracts, tachyzoite lysate antigen (TLA) extracts, or soluble tachyzoite antigen (STAg) extracts using the three-dose treatment schedule. (C-E) ID8DV tumors were established in groups of C57BL/6 mice for 25 days then tumor-bearing mice were treated with PBS, vaccinated with heat killed (HK) uracil auxotrophs or vaccinated with live uracil auxotrophs (experiments used the yellow fluorescent protein expressing CPS strain CPS-YFP [96]) and peritoneal levels of (C) IL-12p40, (D) IL-12p70, (E) IFN-γ, and (F) the absolute number or percentage of YFP⁺CD11c⁺ cells present in the tumor microenvironment were measured 18 and 66 h after treatments. Data is representative of two independent experiments. ns was not significant, *p<0.05, **p<0.01, ****P<0.0001.

Fig 5. Secretion and active invasion is required for the antitumor response.

ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS, or were vaccinated i.p. with 4BPB treated, mycalolide B treated, or untreated tachyzoites of uracil auxotrophs (OMP) using the standard three-dose treatment schedule. Data is representative of three independent experiments. ns was not significant, ****P<0.0001.

Fig 6. Secreted rhoptry proteins ROP35 and ROP38 and dense granule proteins GRA2, GRA12, and GRA24 are required for the antitumor response.

(A) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS or were vaccinated i.p. with tachyzoites of uracil auxotrophs using the three-dose treatment schedule. PBS treated or vaccinated i.p. with OMP or uracil auxotrophs lacking rhoptry proteins ROP21, ROP35, or ROP38, or vaccinated with a ROP35 complemented strain. (B) Validation of apical rhoptry localization of expressed C-terminal HA-tagged ROP35 in the complemented OMPΔrop35::ROP35 strain. DAPI stains the nuclei of both parasites and the host HFF cells they invaded. Localization of the HA tag (revealed by green fluorescence) is associated with the apical rhoptry organelles. Vacuole locations in the host cell are shown by differential interference contrast (DIC) microscopy. (C) PVM localization of ROP35. Parasites were allowed to invaded HFF cells for 14 h and after fixation the cytosolic surface of the PVM was exposed using 0.002% digitonin to expose the cytosolic surface of the PVM and ROP35 was localized. Vacuole locations (PVM) in the host cell are shown by differential interference contrast (DIC) microscopy (red arrowheads). The ROP35 HA tag is associated with the PVM. (D) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS or were vaccinated i.p. with tachyzoites of uracil auxotrophs using the three-dose treatment schedule. PBS treated or vaccinated i.p. with OMP or uracil auxotrophs lacking dense granule proteins GRA2, GRA12, or GRA24. Data is representative of two independent experiments. ns was not significant, **p<0.01, ***P<0.001, ****P<0.0001.

Fig 7. Rhoptry proteins ROP5, ROP17, and ROP18 are required for the antitumor response.

(A) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS, or were vaccinated i.p. with tachyzoites of uracil auxotrophs (OMP), or were vaccinated i.p. with uracil auxotrophs (OMP) lacking rhoptry proteins ROP5, ROP17, or ROP18 using the three-dose treatment schedule. (B) Mouse embryonic fibroblasts were stimulated with IFN-γ and parasite survival (measured as PFU) was determined for uracil auxotrophs mutants lacking specific rhoptry or dense granule proteins. (C) Relative parasite invasion efficiency of uracil auxotrophs was measured in MEFs. The parasite to PFU ratios (invasion efficiency) was measured in at least 4 independent assays and compared to the invasion efficiency of the parental OMP strain. (D) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS, or were vaccinated i.p. with tachyzoites of type I uracil auxotrophs (OMP), or were vaccinated i.p. with type II OMP, or were vaccinated i.p. with type II strains lacking ROP5 or ROP18 using the three-dose treatment schedule. Data is representative of at least two independent experiments. ns was not significant, *p<0.05, **p<0.01, ****P<0.0001.

Fig 8. ROP18 PVM association is required for the antitumor response but the kinase function of ROP18 is dispensable.

(A) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS, or were vaccinated i.p. with tachyzoites of uracil auxotrophs (OMP), OMPΔrop18, or were vaccinated i.p. with OMPΔrop18 uracil auxotrophs that were complemented with wild-type or mutant ROP18 gene alleles using the three-dose treatment schedule. (B) Mouse embryonic fibroblasts were stimulated with IFN-γ and parasite survival (measured as PFU) was determined for type I and type II uracil auxotrophs and type I uracil auxotrophs expressing wild-type or mutant alleles of ROP18. (C) PVM localization of wild-type and mutant ROP18 proteins. Parasites were allowed to invaded HFF cells for 14 h and after fixation the cytosolic surface of the PVM was exposed using 0.002% digitonin and HA tagged ROP18 was localized. Vacuole locations (PVM) in the host cell are shown by differential interference contrast (DIC) microscopy (red arrowheads). Wild-type and the kinase-dead ROP18 mutant localized to the PVM, however, deletion of the RAH2(ATF) domain abolished PVM association. (D) ID8DV ovarian tumors were established in C57BL/6 mice and groups of mice were treated with PBS, or were vaccinated i.p. with tachyzoites of uracil auxotrophs (OMP) or OMPΔrop18, or were vaccinated i.p. with mycalolide B treated tachyzoites of uracil auxotrophs (OMP) or OMPΔrop18 using the three-dose treatment schedule. Data is representative of at least two independent experiments. ns was not significant, **P<0.01, ****P<0.0001.