T-bet-dependent ILC1- and NK cell-derived IFN-γ mediates cDC1-dependent host resistance against Toxoplasma gondii.

Host resistance against intracellular pathogens requires a rapid IFN-γ mediated immune response. We reveal that T-bet-dependent production of IFN-γ is essential for the maintenance of inflammatory DCs at the site of infection with a common protozoan parasite, Toxoplasma gondii. A detailed analysis of the cellular sources for T-bet-dependent IFN-γ identified that ILC1s and to a lesser degree NK, but not TH1 cells, were involved in the regulation of inflammatory DCs via IFN-γ. Mechanistically, we established that T-bet dependent innate IFN-γ is critical for the induction of IRF8, an essential transcription factor for cDC1s. Failure to upregulate IRF8 in DCs resulted in acute susceptibility to T. gondii infection. Our data identifies that T-bet dependent production of IFN-γ by ILC1 and NK cells is indispensable for host resistance against intracellular infection via maintaining IRF8+ inflammatory DCs at the site of infection.


Introduction
Host defense against intracellular pathogens requires a quick and effective type I immune response. A coordinated response of innate myeloid and lymphoid cells is critical for both rapid pathogen restriction and activation of the adaptive immune response. The transcription factor T-bet, encoded by Tbx21, has been shown to play a critical role for the effector function of innate and adaptive lymphocytes in response to intracellular pathogens via regulation of IFN-γ production [1][2][3][4][5][6][7][8]. The cytokine IFN-γ is indispensable for host defense as it is essential for the induction of anti-microbial IFN-γ-inducible genes, which results in intracellular microbial clearance.
The obligate intracellular protozoan parasite Toxoplasma gondii is a potent inducer of IFNγ and has been exploited to characterize the host's innate and adaptive type I immune responses against intracellular pathogens [9]. Immunity against T. gondii requires a type I CD4+ T helper cell (T H 1)-derived IFN-γ response, and in the absence of either CD4+ T cells or IFN-γ, the host rapidly succumbs to infection [10][11][12]. Therefore, it was anticipated that the rapid susceptibility observed in T. gondii infected T-bet-deficient (Tbx21 -/-) mice was due to the absence of CD4+ T cell-derived IFN-γ. However, our group and others have recently observed that Tbx21 -/mice maintained IFN-γ producing CD4+ T cells during parasite infection, while remaining highly susceptible to infection [4,5]. These data suggest an innate T-betdependent mechanism that is critical for a protective type I immune response against T. gondii infection.
Natural killer (NK) cells and group 1 innate lymphoid cells (ILC1s) are critical sources of innate IFN-γ during intracellular infection [13,14]. It has been established that the transcription factors, Eomesodermin (Eomes) and T-bet play a role in NK cell development [15,16]. However, while Eomes is indispensable for NK cell maturation and effector function, T-bet plays a more limited role in the development, migration, and cytokine production of NK cells [4, [16][17][18]. During T. gondii infection, NK cell-derived IFN-γ stimulates the effector function of inflammatory myeloid cells, and in the absence of NK cells the host immunity to T. gondii is compromised [13,19].
Similar to NK cells and T H 1s, tissue resident ILC1s have been identified as a critical source of IFN-γ necessary for pathogen restriction [14,[20][21][22]. Unlike NK cells, the maturation and cytokine production of ILC1s is T-bet-dependent and Eomes-independent [14]. Tissue resident ILC1s rapidly respond to type I conventional DC (cDC1)-derived IL-12 in an antigen independent manner, leading to IFN-γ production [22,23]. In addition to providing an early source of IFN-γ, ILC1s can also augment the recruitment of innate inflammatory myeloid cells during T. gondii infection [14], demonstrating that crosstalk between innate myeloid cells and ILC1s is a critical component of an effective innate type I immune response against intracellular pathogens.
Because multiple cellular sources of IFN-γ play a complex and non-redundant role for Tbet-dependent host defense against the common protozoan parasite T. gondii, we sought to determine the role T-bet plays in coordinating innate myeloid cells and lymphocytes to work in concert with one another for host resistance. Our experiments revealed that in the absence of T-bet, inflammatory DCs were significantly compromised at the site of infection. We identified that parasite-mediated T-bet-dependent ILC1-derived IFN-γ is crucial for maintaining inflammatory DCs during infection. In addition, NK cell-derived IFN-γ played a similar, although less profound, role in the regulation of DCs during T. gondii infection. Importantly, the absence of DCs could be rescued by exogenous administration of IFN-γ during T. gondii infection, indicating a critical role for early T-bet-dependent IFN-γ in regulating inflammatory DCs. Mechanistically, we uncovered that innate IFN-γ was required for induction of the transcription factor interferon regulatory factor-8 (IRF8) in cDC1s, which are required for immunity against T. gondii. Our results establish that during T. gondii infection, T-bet-dependent IFN-γ is indispensable for regulating inflammatory IRF8+ cDC1s, leading to parasite clearance and host survival.

T-bet is critical to maintain inflammatory DCs during T. gondii infection
To establish T-bet's protective role in host defense against intracellular pathogens, we implemented the well-established intraperitoneal (i.p.) model of infection with T. gondii that triggers and depends on a robust CD4+ T H 1-derived IFN-γ response [10,24]. Considering the importance of T-bet in the regulation of T H 1 cells [7], we anticipated that T-bet-deficient mice would demonstrate enhanced susceptibility to T. gondii similar to T-cell deficient mice due to the lack of T-bet-mediated T H 1-derived IFN-γ production. Strikingly, mice lacking T-bet were more susceptible to T. gondii compared to Rag2 -/animals and were unable to survive past day 10 of the acute stage of infection ( Fig 1A). These data suggested that the transcription factor Tbet is involved in innate host defense and is essential for host survival through the acute stage of parasitic infection independently of the regulation of T H 1 immunity to the parasite.
It has been well-established that T. gondii-mediated type I immunity requires DCs for host defense [25][26][27][28][29]. Work from our lab and others have demonstrated that DC-deficiency results in acute host susceptibility to T. gondii [26,30]. Therefore, we examined if the absence of T-bet compromised inflammatory DC-mediated immunity. We observed that in naïve mice, lacking T-bet had no discernable effect on the presence of DCs in the peritoneal cavity (Fig 1B), defined as lineage-negative (CD3 -CD19 -NK1.1 -; Lin -) CD11c + MHCII + cells (S6A Fig). In striking contrast, Tbx21 -/mice had a significantly reduced frequency of inflammatory DCs compared to both WT and Rag2 -/animals when infected with T. gondii (Figs 1C, 1D and S1).
To determine if T. gondii-mediated DCs required intrinsic T-bet expression, we employed a mouse model of DC-restricted T-bet deficiency using the CD11c-Cre system (CD11c-Cre x Tbx21 flox/flox mice, DC-Tbx21 -/-). Similar to WT controls, DC-Tbx21 -/retained Lin -CD11c + MHCII + DCs during parasite infection (S2A Fig). A detailed analysis of inflammatory DCs present at the site of infection revealed that as early as day 3 post-infection, there was a noticeable reduction in DCs analyzed in Tbx21 -/mice when compared to WT mice (S1A Fig). Furthermore, by day 5 post-infection, inflammatory DCs were practically undetectable in the absence of T-bet, a trend that was also observed by day 8 following infection ( Fig 1C and  1D). Overall, our experiments established a critical innate function of T-bet for regulating the presence of inflammatory DCs during acute parasitic infection.

Early T-bet-dependent IFN-γ is critical for inflammatory DCs via regulation of IRF8
The IFN-γ inducible transcription factor IRF8, originally described as interferon consensus sequence-binding protein (ICSBP), is known to play a key role in the development and survival of cDC1s, also known as CD8 + BATF3 + DCs [31][32][33][34][35]. Therefore, we investigated if T-bet regulates the presence of inflammatory DCs via induction of IRF8. We observed that in naïve mice, T-bet was dispensable for the presence of a small population of peritoneal IRF8 + DCs, similar to WT and Rag2 -/animals (Figs 2A, 2B and S1D). Strikingly, T. gondii infection resulted in the rapid accumulation of inflammatory IRF8 + DCs in WT mice and this population was virtually absent in Tbx21 -/mice by day 8 post-infection (Figs 2C-2F and S1E). Additionally, the absence of IRF8 + DCs in Tbx21 -/resulted in an overall reduction of IL-12 To identify if the lack of T-bet resulted in impaired early IFN-γ production during T. gondii infection, we performed a detailed analysis of T. gondii-mediated IFN-γ production. Our mice were i.p. infected and PECs were harvested on days 0, 5, and 8 (B, C, D) and Lin -CD11c + MHCII + DCs were analyzed by flow cytometry. Average frequencies of Lin -CD11c + MHCII + DCs (B, C, D) in the PECs were analyzed on days 0, 5, and 8 following infection. Results are representative of three-independent experiments involving at least 3 mice per group. Statistical analyses were done using Log-rank (Mantel Cox) test or unpaired t-test analysis of individual groups, �� P<0.01, ��� P<0.001, ���� P<0.0001. Error bars, standard error mean. https://doi.org/10.1371/journal.ppat.1008299.g001

PLOS PATHOGENS
ILC1s maintain dendritic cells via IFN-gamma results identified that T-bet was critical for early IFN-γ production on days 3 and 5 of parasite infection ( Fig 3A). Therefore, we hypothesized that early IFN-γ is required to maintain DCs during T. gondii infection. To test our hypothesis, IFN-γ was administered to infected Tbx21 -/mice, which not only significantly augmented inflammatory DCs at the site of infection ( Fig  3B and 3C), but also rescued IRF8 expression in DCs (Fig 3D-3F). These data demonstrate Tbet-dependent early IFN-γ production is essential for maintaining inflammatory DCs via regulation of IRF8.

T-bet is essential for ILC1-derived IFN-γ and contributes to NK cellderived IFN-γ
To thoroughly define the function of IFN-γ in regulating inflammatory DCs, we examined if neutralizing IFN-γ eliminates their presence at the site of infection. We observed that blocking IFN-γ during T. gondii infection resulted in a striking reduction of inflammatory DCs in both WT and T cell-deficient mice (Fig 4A-4C). These results revealed that IFN-γ from innate immune cells was sufficient to maintain inflammatory DCs in the absence of T cell-derived IFN-γ. Among innate immune functions, T-bet has been described to play an important role for NK cell function and ILC1 development, and both of these cell types primarily control acute T. gondii infection via their production of IFN-γ [14][15][16]. Therefore, we examined IFN-γ production by NKs, ILC1s, neutrophils, T, and B cells on days 3 and 5 of infection and observed that NKs and ILC1s are the primary sources of the effector cytokine during the acute stage of infection (S3A Fig). To characterize the importance of innate IFN-γ for the maintenance of inflammatory DCs, we treated WT and Rag2 -/mice with anti-NK1.1 antibodies resulting in a profound loss of inflammatory DCs (Fig 4A-4C), further identifying that innate lymphoid cells are critically involved in sustaining parasite-triggered inflammatory DCs. To further explore if ILC1s were sufficient and necessary for maintaining DCs during infection, we next tested the effects of selective depletion of Thy1+ cells in WT and Rag2 -/mice. Depletion of Thy1 expressing cells resulted in a significant decrease of DCs in WT and Rag2 -/mice (Fig 4A-4C). Moreover, Thy1-mediated depletion of ILC1s in Rag2 -/mice resulted in increased pathogen burdens comparable to levels seen in T-bet-deficient animals (S3B Fig). Our data implicates that either NK cell-or ILC1-derived IFN-γ, in a T-bet-dependent manner, are required to maintain inflammatory DCs during T. gondii infection.
Our experiments with WT and T-bet-deficient mice revealed that in response to T. gondii infection, NK cells produced large amounts of IFN-γ in a T-bet-independent manner. This was evident from comparative analysis of IFN-γ production by NK cells in WT and Tbx21 -/mice on days 3 and 5 post-infection (Figs 4D, 4F and S4A). Nevertheless, an adoptive transfer of NK cells into T-bet-deficient mice partially rescued the presence of inflammatory DCs ( Fig  4H). These results revealed that while T-bet is not absolutely required for IFN-γ production by NK cells during T. gondii infection, T-bet-regulated production of IFN-γ by NK cells plays a role in the regulation of inflammatory DCs.
Most importantly, we observed robust T-bet expressing ILC1-derived IFN-γ responses during infection in WT mice (Fig 4E and 4G). We identified ILC1s from the peritoneal cavity as CD3 -CD19 -Ly6G -NKp46 + CD127 + (S6B Fig) and, upon further evaluation, confirmed this population was also NKp46 + GranzymeB -CD49a + CD49b -CD127 + CD200R + (S5 Fig), and as anticipated were absent in Tbx21 -/mice at all examined time points following T. gondii infection (Fig 4E and 4G). Therefore, T-bet deficiency resulted in two major defects in innate immunity: a complete lack of ILC1s and inflammatory DCs at the site of infection. We further examined a role for ILC1s in maintenance of inflammatory DCs during T. gondii infection by analyzing parasite infected lymphocyte-deficient (Rag2 -/γc -/-) mice, which also lack ILC1s. We

PLOS PATHOGENS
ILC1s maintain dendritic cells via IFN-gamma observed that similar to T-bet deficiency, Rag2 -/γc -/mice had very few inflammatory DCs (S4B Fig), further demonstrating that ILC1s are essential for the presence of inflammatory DCs during parasite infection. Finally, an adoptive transfer of ILC1 into Tbx21 -/mice restored inflammatory DCs at the site of infection (Figs 4H and S6C). Taken together, these result suggest that ILC1s play a major role in the regulation of inflammatory DCs.

IRF8+ DCs are critical for pathogen clearance against intracellular infection
Our data has identified that early ILC1-derived IFN-γ is critical for maintaining IRF8 + DCs during infection and limiting parasite replication throughout the host. Therefore, we hypothesized that conditionally deleting IRF8 expression by DCs would result in uncontrolled parasite replication during infection and rapid host mortality. To test our hypothesis, we infected mice with a DC-restricted deficiency of IRF8 by using the CD11c-Cre system (CD11c-Cre x Irf8 flox/ flox mice, DC-Irf8 -/-). Infected DC-Irf8 -/mice revealed an overall reduction of total MHCII + CD11c + DCs by day 5 post-infection ( Fig 5A). Moreover, the absence of IRF8 expression in MHCII + CD11c + DCs resulted in dramatically elevated parasite burden both at the site of infection and in the spleen on day 5 post infection (Fig 5B and 5C). We then examined if IRF8-deficient DCs were required for host survival against T. gondii. Similarly to T-bet-deficient mice, DC-Irf8 -/animals succumb to parasite infection rapidly (Fig 5D). Our data establishes that early and rapid T-bet-dependent innate IFN-γ is indispensable for maintaining IRF8 + cDC1s, which are required for pathogen clearance and host survival.

Discussion
The cytokine IFN-γ is critical for triggering cellular anti-parasitic defense mechanisms that are essential for the destruction of the intracellular parasitophorous vacuole and T. gondii clearance [9,12, [36][37][38][39][40][41][42][43]. Classically, the transcription factor T-bet is considered the master regulator for determining the CD4+ T H 1 lineage and IFN-γ production [7]. However, recent studies revealed that that T-bet is largely dispensable for CD4+ T cell-derived IFN-γ responses [4,5]. Herein we demonstrate that Tbx21 -/mice succumb to infection significantly quicker than mice lacking T cells, suggesting an innate T-bet-dependent mechanism of host immunity, critical for survival of the acute stage of infection.
Innate lymphoid cells are critical early responders to intracellular pathogens. During T. gondii infection, two sets of ILCs are known to have essential roles for host resistance: NK cells and ILC1s. T-bet is associated with the maturation, egress, and effector function of NK cells [15,17,18]. Yet, NK cell development is not impaired in the absence of T-bet in this model [16]. Moreover, T. gondii-triggered NK cell-derived IFN-γ production plays a critical role for initiating the effector function of myeloid cells that are required for host resistance [13, 19,44]. Our results demonstrate that the absence of T-bet does not impede the recruitment of NK cells

ILC1s maintain dendritic cells via IFN-gamma
to the site of infection, and only partially compromised their capability to produce IFN-γ. These data establish that during T. gondii infection, T-bet plays a limited role for the migration of NK cells to the peritoneum and their IFN-γ production.

ILC1s maintain dendritic cells via IFN-gamma
Along with NK cells, ILC1s have been shown to be a critical early source of IFN-γ during microbial invasion. Thus, it is imperative to perform clear phenotyping of NK and ILC1 populations. Based on previous studies investigating peritoneal NKs and ILC1s, we defined NKs as NKp46 + GranzymeB + CD49a -CD49b + CD127 -CD200Rand ILC1s as NKp46 + GranzymeB -C-D49a + CD49b -CD127 + CD200R + [22,23]. Early ILC1-derived IFN-γ is critical for host resistance against Mouse Cytomegalovirus (MCMV), Clostridium, Salmonella, and T. gondii [14,[20][21][22]. T-bet is required for the maturation and cytokine production of ILC1s [14]. Previous groups have observed that T. gondii infection mediates ILC1-derived IFN-γ and plays an important role in parasite clearance [14]. By examining T. gondii-infected Tbx21 -/mice, we were able to define a host defense function of T-bet-dependent ILC1s during T. gondii infection. T-bet-deficiency resulted in the complete loss of ILC1s in comparison to a minor reduction in IFN-γ production by NK cells or CD4+ T cells [5,14]. This suggested that the significant reduction of inflammatory DCs observed in Tbx21 -/mice was caused by the absence of T-bet-dependent ILC1-derived IFN-γ. By antibody depletion of ILC1s or IFN-γ neutralization, we revealed that ILC1-derived IFN-γ plays a major role in maintaining inflammatory DCs and restricting parasite growth both locally and in peripheral tissues. Nevertheless, it is important to note that T-bet regulated NK cells can also augment inflammatory DCs at the site of the infection. We also observed that absence of T-bet resulted in the reduction of inflammatory monocytes during the acute stage of the infection (S2B Fig). Thus, the critical function of T-bet-dependent ILC1-and NK cell-derived IFN-γ is to maintain inflammatory DCs at the site of infection.
We and others have previously established that the transcription factor IRF8 is essential for host resistance to T. gondii [34,45]. IRF8-deficiency resulted in acute susceptibility to T. gondii due to the absence of cDC1s [46]. Our results revealed that early T-bet-dependent IFN-γ was critical for maintaining inflammatory DCs via regulation of IRF8 expression and DC-specific IRF8 is required for host resistance to the parasite.
This study defines the critical function of early T-bet-dependent IFN-γ in sustaining inflammatory IRF8 + DCs during intracellular pathogen infection. Our results demonstrate continuous crosstalk between DCs and ILC1s mediated by IL-12 and IFN-γ, where DCderived IL-12 triggers an early IFN-γ response from ILC1s and NK cells and IFN-γ produced by innate lymphoid cells is essential for maintaining inflammatory DCs during infection via regulation of IRF8.

Ethics statement
All mice were maintained at in the pathogen-free American Association of Laboratory Animal Care-accredited animal facility at the University of Rochester Medical Center, Rochester, NY.
All animal experimentation (animal protocol #102122) has been reviewed and approved by the University Committee on Animal Resources (UCAR), the Institutional Animal Care and Use Committee (IACUC).

Animals
C57BL/6, Rag2 -/-, Tbx21 -/-, CD11c-Cre, Irf8 flox/flox , and Tbx21 flox/flox mice were obtained from Jackson Laboratory (Bar Harbor, ME) and Rag2 -/γc -/mice were obtained from Taconic (Rensselaer, NY). All control and experimental mice were age-and sex-matched within all individual experiments. This study included both male and female mice, and the data derived from male and female mice identified no sex-specific differences in the performed experiments.

ILC1s maintain dendritic cells via IFN-gamma
Toxoplasma gondii infection and qPCR. All mice were i.p. infected with an average of 20 T. gondii cysts of the ME49 strain. At days 0, 3, 5, and 8 post-infection, the animals were necropsied. In some experiments, mice were injected i.p. with 50 ng of IFN-γ (R&D Systems) on days 0, 1, 2, and 3. In some experiments, mice were injected i.p. with 200 μg of anti-IFN-γ (BioxCell) 500 μg of anti-NK1.1 (BioxCell), or 200 μg of anti-Thy1.2 (BioxCell) on days 0, 3, 5. To determine T. gondii pathogen loads, total genomic DNA from animal tissue was isolated by using the DNeasy Blood and Tissue Kit (Qiagen) according to manufacturer's instructions. PCR were performed by using SSOFast Eva Green Supermix (BioRad). Samples were measured by qPCR using a MyiQ Real-Time PCR Detection System (BioRad), and data from genomic DNA was compared with a defined copy number standard of the T. gondii gene B1.
ELISA analysis. The IFN-γ and IL-12/23p40 concentration in the sera or peritoneal exudate fluid (PEF) was analyzed by standard sandwich ELISA kit according to manufacturer's instructions (ThermoFisher).

Measurements of immune cells responses.
To assay the responses of mice infected with T. gondii, the PECs were harvested from C57BL/6, Rag2 -/-, Tbx21 -/-, DC-Irf8 -/-, DC-Tbx21 -/-, and Rag2 -/γc -/mice on days 0, 3, 5, or 8 post-infection. To examine neutrophils, T cells, B cells, NK cells and ILC1s responses, single-cell suspension of PECs were restimulated with PMA (20 ng/mL) and ionomycin (1 μg/mL) (Sigma-Aldrich) for 4 hours in the presence of GolgiPlug (Brefeldin A, BD Biosciences). After isolation and in vitro restimulation, the cells were washed once in phosphate-buffered saline and stained with Zombie Yellow (BioLegend) to assess live vs. dead status of cells. Cells were then washed with phosphate-buffered saline + 1% fetal bovine serum and stained with fluorochrome-conjugated antibodies. For intracellular staining and subsequent washing, cells were permeabilized overnight at 4˚C with the Foxp3/Transcription Factor Staining Buffer Set according to the manufacturer's instructions (ThermoFisher).
Cells were sorted using a FACSAria II cell sorter (BD Biosciences). Cell fluorescence was measured using an LSRII flow cytometer (BD Biosciences), and data were analyzed using FlowJo Software (Tree Star, Ashland, OR).

Statistical analysis
All data were analyzed with Prism (Version 8; GraphPad, La Jolla, CA). These data were considered statistically significant when P-values were <0.05.

PLOS PATHOGENS
ILC1s maintain dendritic cells via IFN-gamma