cDC1 coordinate innate and adaptive responses in the omentum required for T cell priming and memory

Abstract

In the peritoneal cavity the omentum contains fat associated lymphoid clusters (FALCs) whose role in response to infection is poorly understood. After intraperitoneal immunization with Toxoplasma gondii, conventional type 1 dendritic cells (cDC1) were critical to induce innate sources of IFN-γ and cellular changes in the FALCs. Unexpectedly, infected peritoneal macrophages that migrated into the FALCs primed CD8⁺ T cells. While T cell priming was cDC1-independent, these DC were required for maximal CD8⁺ T cell expansion. An agent-based computational model and experimental data highlighted that cDC1 impacted the magnitude of the proliferative burst and promoted CD8⁺ T cell expression of nutrient uptake receptors and cell survival. Thus, although FALCs lack the organization of secondary lymphoid organs, cDC1 resident in this tissue coordinate innate responses to microbial challenge and provide secondary signals required for T cell expansion and memory formation.

ONE-SENTENCE SUMMARY:

cDC1 are not required for T cell priming but coordinate immune responses in the omentum to generate protective T cell responses.

INTRODUCTION

Clusters of immune cells in white adipose tissue (WAT) form fat-associated lymphoid clusters (FALCs) (1, 2) that provide protective immunity following a breach in local lymphatics or mucosal barrier sites (3–6). In response to inflammation within serous cavities, FALCs increase in size as leukocytes from the blood or cells from the cavities access these structures (1, 5, 7). In the peritoneum, the omentum is composed of mesothelial sheets that enclose adipose tissue densely populated with FALCs (termed milky spots, MS) that are enriched for B cells but also other immune cells (8). These vascularized MS associate with stomata in the mesothelial layer of the omentum and allow entry of inflammatory cells into the peritoneum (5, 6, 9, 10), but are also a site of cellular egress and antigen drainage from the peritoneum (6, 7, 9–11). While soluble antigen from the peritoneum can drain to the mediastinal and parathymic lymph nodes (LNs) (12), cells and particulates drain almost entirely to the omentum (9, 11). In higher vertebrates, the MS in the omentum contribute to humoral responses (7), but are less organized compared to conventional secondary lymphoid organs (SLOs) (8). The reduced complexity of FALCs and differences in the stromal cell network have contributed to the idea that these sites have a secondary role in the induction of immune responses.

Within LNs the positioning of cDC is critical for the efficient capture and presentation of antigen to generate T cell responses (13–15). This localization is of particular relevance to cDC1 that cross-present exogenous antigen to CD8⁺ T cells (16, 17). However, studies with vaccinia virus (VV) concluded that in LN cDC1 were not required for priming, but rather that uninfected cDC1 allow communication between CD4⁺ and CD8⁺ T cells that shape CD8⁺ T cell differentiation (13). This role for cDC1 as a platform to help CD8⁺ T cell responses is apparent in other systems (15, 17), but less is known about the localization and function of cDC in the FALCs. Intraperitoneal (i.p.) infection with a live, replication-deficient strain of T. gondii results in DC-dependent parasite-specific CD8⁺ and CD4⁺ T cell responses (18, 19). In the present study, a replication-deficient strain of T. gondii that expresses ovalbumin (OVA) and Cre recombinase was combined with a Cre-sensitive reporter mouse to define the events involved in CD8⁺ T cell priming in the omentum. These data sets formed the basis for an agent-based model to predict how perturbations in priming would impact CD8⁺ T cell responses. Together, these studies demonstrate that cDC1 mediate inflammatory changes in the FALCs, and – although infected macrophages prime CD8⁺ T cells – cDC1 in the FALCs provide secondary signals required for T cell expansion and memory formation.

RESULTS

Immunization alters organization of the omental MS

Injection of the CPS strain of T. gondii i.p. results in rapid infection of the resident cells. To track host-parasite interactions, CPS that express mCherry and inject Cre recombinase into host cells (CPS-Cre-mCherry) were combined with Cre-sensitive Ai6 reporter mice to identify cells that had phagocytosed parasites (Toxo⁺ZsGreen⁻), had been infected (Toxo⁺ZsGreen1⁺), or been injected with Cre (Toxo⁻ZsGreen⁺) (20, 21). In the first 2 days post-immunization (dpi), infected cells had moved from the peritoneum to the omentum but were rare in the draining parathymic and mediastinal lymph nodes (DLN) (Fig. 1A). By 2 dpi in the omentum there was an influx of large peritoneal macrophages (LPM), MHCII⁺ monocytes, and macrophages (Mϕ) (Fig. 1B, C), and CD64⁺ LPM and Mϕ composed the majority of infected or injected cells (Fig. 1D, E). Imaging revealed the MS contained distinct regions enriched for B cells and a central T cell zone (Fig. 1F–G). Reticular cells and reticular fibers formed an extensive network, and infected cells were present in a mantle of CD11b⁺ myeloid cells (Fig. 1H). The use of Lyz2-Cre mice interbred with Ai6 reporters (Lyz2^Ai6), indicated that this mantle was largely composed of monocytes (fig. S1A–F). Infected ZsGreen⁺ macrophages were observed in regions enriched with T and B cells (Fig. 1I–J) but not outside the MS (fig. S1G). Thus, after CPS immunization infected macrophages migrate from the peritoneum into the MS accompanied by monocyte recruitment.

Fig. 1. CPS immunization results in antigen transport to MS.

(A) Intraperitoneal immunization of Ai6 mice with CPS-Cre-mCherry parasites. Quantification of cells phagocytosing the parasite (Toxo⁺ZsGreen⁻), infected cells (Toxo⁺ZsGreen⁺), and injected cells (Toxo⁻ZsGreen⁺) from the peritoneum to the omentum and DLN at 1 and 2 dpi. Data represent summation of the means.

(B to E) UMAP projection of cellularity of the omentum of naïve and CPS-Cre-mCherry-immunized Ai6 mice at 2 dpi. (B) Color-coded for clusters of non-hematopoietic cells (CD45⁻), B cells (CD45⁺CD64⁻FcER1α⁻cKit⁻SiglecH⁻Ly6G⁻CD19⁺MHCII⁺B220⁺), T cells (CD45⁺CD64⁻FcER1α⁻cKit⁻SiglecH⁻Ly6G⁻TCRβ⁺CD5⁺MHCII⁻CD3⁺), CD64⁺ (CD45⁺CD64⁺CD11b⁺), Neutrophils (CD45⁺CD64⁻FcER1α⁻cKit⁻SiglecH⁻Ly6G⁺), Eosinophils (CD45⁺CD64⁻FcER1α⁻cKit⁻SiglecH⁻Ly6G⁻TCRβ⁻CD5⁻CD19⁻NK1.1⁻CD3⁻CD11b⁺SiglecF⁺), cDC1 (CD45⁺CD64⁻FcER1α⁻cKit⁻SiglecH⁻CD11c⁺MHCII⁺XCR1⁺CD11b⁻), cDC2 (CD45⁺CD64⁻FcER1α⁻cKit⁻SiglecH⁻CD11c⁺MHCII⁺XCR1⁻CD11b⁺).

(C) UMAP projection of changes in the CD64⁺ population between naïve (top) and immunized (bottom) Ai6 mice. Color-coded for indicated cell-type, LPM (CD45⁺CD64⁺CD11b⁺CD102⁺), MHCII⁺ Mo (CD45⁺CD64⁺CD11b⁺CD102⁻MHCII⁺Ly6C⁺), Mϕ (CD45⁺CD64⁺CD11b⁺CD102⁺MHCII⁺Ly6C⁻).

(D) Cells phagocytosing parasites (red), infected cells (yellow), and injected cells (green) in the omentum color-coded on UMAP projection from (B).

(E) Quantification of cellular distribution of infected and injected cells from immunized omenta. DLN, Draining LN; UMAP, Uniform Manifold Approximation and Projection; Neut, Neutrophil; Eos, eosinophils; cDC, conventional dendritic cell; LPM, large peritoneal macrophage; Mo, monocyte; Mϕ, macrophage.

(F and G) Omentum MS from WT mice injected with PBS or 10⁵ CPS parasites at 2 dpi showing regions enriched in B cells and T cells. CD4 (green), CD8β (red), B220 (blue).

(H to J) MS from Ai6 mice immunized with 10⁵ CPS-Cre-mCherry parasites at 2 dpi. (H and I) ZsGreen (green), CPS-Cre-mCherry (red), ER-TR7 (white), CD11b (yellow). (J) ZsGreen (green), CPS-Cre-mCherry (red), CD3 (white), B220 (blue). Data are representative plots from 2–3 experiments using 4–5 mice.

cDC1 coordinate innate changes in the MS

cDC1 are the source of IL-12 that stimulates group 1 ILCs to produce IFN-γ required for resistance to T. gondii (22). To determine the impact of cDC1 and IL-12 on the immune response to CPS, Batf3^−/− mice that lack cDC1 (fig. S2A–B) and Il12b^−/− mice were compared to WT mice. Batf3^−/− and Il12b^−/− mice had a major defect in recruitment of Ly6C^HI monocytes into the omentum (Fig. 2A–C, fig. S2C) and peritoneum (fig. S2D–F), and reduced NK and ILC1 production of IFN-γ (Fig. 2D–F, fig. S2G–I). Comparison of immunized Rag2^−/− and Rag2^−/−γc^−/− mice (lacking ILCs) revealed that ILCs are required to recruit monocytes to the peritoneum (fig. S2J–K). Moreover, CPS immunization induces a loss of LPM in the peritoneum associated with their migration to the omentum that was dependent on cDC1 and IL-12 (fig. S2L–M). Nevertheless, infected and injected cells accumulated in the MS of Ai6 and Ai6/Batf3^−/− mice (Fig. 2G–I), indicating that infected LPM migrated into the MS in the absence of cDC1 (Fig. 2J). Thus, cDC1 coordinate the recruitment of monocytes into the omentum and peritoneum and resident macrophages from the peritoneum into the MS, but migration of infected cells into the omentum is cDC1-independent.

Fig. 2. Recruitment of innate immune cells to MS requires cDC1 and IL-12.

(A to C) Quantification of monocytes in the omenta of WT, Batf3^−/− and Il12b^−/− mice after immunization with CPS parasites (2 dpi). (A) Flow cytometry analysis of monocytes (CD64⁺CD11b⁺CD102⁻Ly6C^HI).

(B and C) Quantification of monocytes (B) as a fraction of live cells in the omentum and (C) the number of monocytes in the omentum. Data represent 1 experiment from 2–3 independent experiments with 4–5 mice.

(D to F) Quantification of IFN-γ production by ILCs in the omenta of WT, Batf3^−/−, and Il12b^−/− mice after immunization with CPS parasites (1 dpi). (D) Flow cytometry analysis of NK cells (Lin⁻NK1.1⁺NKp46⁺EOMES⁺CD200R⁻) and ILC1s (Lin⁻NK1.1⁺NKp46⁺EOMES⁻CD200R⁺) producing IFN-γ.

(E and F) Quantification of total number of (E) NK cells and (F) ILC1s producing IFN-γ. Lin (CD3, CD5, CD19, B220, F4/80).

(G) Quantification of numbers of infected and injected Live Cells in the omentum of Ai6 and Ai6/Batf3^−/− mice after immunization with CPS-Cre-mCherry parasites (2 dpi).

(H) 3D rendering of infected cells in MS of Ai6/Batf3^−/− mice immunized with CPS-Cre-mCherry parasites (2 dpi). ER-TR7 (gray), ZsGreen (green).

(I) Cellular distribution of infected and injected cells in the omenta of Ai6 and Ai6/Batf3^−/− mice with same gating strategy as (Fig. 1B).

(J) Z slice showing infected cells in T cell enriched region of MS of Ai6/Batf3^−/− mice immunized with CPS-Cre-mCherry parasites (2 dpi). ZsGreen (green), CD3 (red), B220 (blue).

Data are representative plots from 2–3 independent experiments all using 4–5 mice. (B, C, E, and F) Bar graphs indicate mean ± SD. (G) Bar graph indicates summation of means. Data analyzed by (B and C) 2-way ANOVA or (E, F, and G) unpaired T test. ns, not significant; *p < 0.05; **p < 0.01; ****p < 0.0001.

CD8⁺ T cell priming occurs in the omentum

The presence of infected macrophages in an area rich in T cells in the MS suggested that this may be a site of T cell priming. Therefore, CPS parasites that expressed OVA (CPS-OVA) were combined with OT-I T cells that transiently expressed Nur77^GFP after TCR engagement (OT-I/Nur77^GFP). CellTrace Violet (CTV) labeled OT-I/Nur77^GFP T cells were transferred intravenously (i.v.) into congenically distinct hosts that were then immunized with CPS-OVA or heat-killed (HK) CPS-OVA. At 3 dpi, only CPS-OVA induced marked Nur77^GFP expression and proliferation in the omentum (Fig. 3A–B) with low levels of OT-I T cell activation observed in the DLN (fig. S3A). Omentectomy at 2 dpi resulted in a 70% decrease in the number of parasite-specific CD8⁺ and CD4⁺ T cells in the peritoneum at 10 dpi (fig. S3B). While this surgical intervention would result in a reduced parasite antigen load, this result is consistent with the omentum as a site of antigen drainage and T cell priming. Furthermore, i.p. transfer of OT-I/Nur77^GFP T cells followed by immunization with CPS or CPS-OVA revealed that by 18 hours post-immunization only the CPS-OVA resulted in upregulation of CD69 and Nur77^GFP (Fig. 3C–D).

Fig. 3. CD8⁺ T cell activation in the omentum.

(A and B) Intravenous transfer of 10⁶ CTV-labeled OT-I/Nur77^GFP T cells into WT mice that were immunized i.p. with either CPS-OVA or heat-killed (HK) CPS-OVA parasites 24 hr later. (A) Flow cytometric analysis of transferred CTV⁺ OT-I/Nur77^GFP T cells in the omentum of recipient WT mice at 3 dpi.

(B) Quantification of the activated (Nur77^GFP+ and divided) OT-I/Nur77^GFP T cells.

(C and D) Intraperitoneal transfer of 10⁶ CTV-labeled OT-I/Nur77^GFP T cells into WT mice that were immunized with CPS or CPS-OVA parasites 2 hr later. (C) Flow cytometric analysis of transferred CTV⁺ OT-I/Nur77^GFP T cells in the omentum of recipient WT mice at 18 hpi measured.

(D) Quantification of the fraction of activated (CD69⁺Nur77^GFP+) OT-I/Nur77^GFP T cells in the omentum.

(E and F) Intravital imaging of OT-I/Nur77^GFP T cells and P14 T cells in the MS of Ai6 mice immunized with CPS-Cre-OVA-mCherry at 20 hpi. At 18 hpi, 2 × 10⁵ CTV-labeled OT-I/Nur77^GFP and 2 × 10⁵ CTY-labeled P14 T cells were transferred i.p. into Ai6 mice, and 2 hours later the omentum was imaged. (E) Analysis of the Mean Track Speed and Arrest Coefficient for OT-I/Nur77^GFP and P14 T cells.

(F) The association of CTV-labeled OT-I/Nur77^GFP and CTY-labeled P14 with infected ZsGreen⁺ cells in the MS. CTV (cyan), ZsGreen (green), CTY (red).

(G and H) Naïve OT-I/Nur77^GFP T cell priming in vitro for 72 hr by naïve, uninfected (ZsG⁻), or infected (ZsG⁺) LPM sorted from naive or immunized (5 × 10⁵ CPS-Cre-OVA-mCherry parasites i.p. for 18 h) Ai6 mice. (G) Flow cytometric analysis of CTV⁺ OT-I/Nur77^GFP T cells by each group of LPM.

(H) Quantification of the fraction of activated (Nur77^GFP+, divided) OT-I/Nur77^GFP T cells in each group.

(I and J) Intraperitoneal transfer of 10⁶ OT-I/Nur77^GFP T cells into β2M^fl/fl or Lyz2-β2M mice that were immunized with CPS-OVA parasites 2 hr later. (I) Flow cytometric analysis of transferred OT-I/Nur77^GFP T cells in the omentum of recipient mice at 18 hpi.

(J) Quantification of the fraction of activated (CD69⁺Nur77^GFP+) OT-I/Nur77^GFP T cells in the omentum.

(K to M) Imaging and analysis of 5 × 10⁵ i.p.-transferred CTV-labeled OT-I/Nur77^GFP T cell clustering in the MS of WT mice immunized with CPS-OVA at 18 hpi. (K) OT-I T cell clustering in the T cell enriched area of the MS. CTV (red), CD3 (gray), B220 (blue).

(L) CTV⁺ OT-I/Nur77^GFP+ T cell clustering in MS. CTV (red), Nur77^GFP (green), ER-TR7 (white).

(M) K-function statistical analysis of Nur77^GFP+ OT-I T cell clustering in 9 individual MS (black lines) from 2 individual mice was averaged (red line). Dashed red lines indicate significant P values of 0.05 and 0.95.

(N to Q) Imaging and analysis of 5 × 10⁵ i.p.-transferred CTV-labeled OT-I/Nur77^GFP T cells clustering in the MS of Ai6 or Snx22^GFP/+ mice that were immunized with CPS-Cre-OVA-mCherry parasites at 18 hpi. (N) Imaging of infected cells and CTV⁺ OT-I/Nur77^GFP+ T cell clustering in MS of Ai6 mice. CTV (red), Nur77^GFP (green), ZsGreen (magenta), CPS-Cre-OVA-mCherry (white), B220 (blue). Nur77^GFP and ZsGreen were distinguished and pseudocolored using Imaris imaging software.

(O) Immunofluorescence imaging of cDC1 and CTV⁺ OT-I/Nur77^GFP+ T cell clustering in the MS of Snx22^GFP/+ mice. CTV (red), Nur77^GFP (green), Snx22^GFP (gray), B220 (blue). Nur77^GFP and Snx22^GFP were distinguished and pseudocolored using Imaris imaging software.

(P and Q) Cross K-function statistical analysis of Nur77^GFP+ OT-I T cell clusters compared to the ZsGreen⁺ cells in Ai6 mice (P) or cDC1 in Snx22^GFP/+ mice (Q). Analysis of clustering in (P) 9 individual MS (black lines) from 2 individual mice or (Q) 6 individual MS (black lines) from 2 individual mice. Average values indicated by solid red lines, dashed red lines indicate significant P values of 0.05 and 0.95.

CTV, CellTrace Violet; CTY, CellTrace Yellow. Data are representative plots from (B and H) 2, (D, J, P, and Q) 3, or (M) 7 independent experiments all using 2–6 mice. (B, D, E, H, and J) Bar graphs indicate mean±SD. Data analyzed by (H) One-way ANOVA or (B, D, E, and G) unpaired T test. ns, not significant; ***p < 0.001; ****p < 0.0001.

Next, Ai6 mice were immunized with CPS-Cre-OVA-mCherry parasites and 18 hours post-immunization (hpi) CTV-labeled OT-I/Nur77^GFP T cells along with CellTrace Yellow-labeled P14 T cells (an internal control specific for the gp33 peptide of LCMV) were transferred i.p. into immunized Ai6 mice. Intravital imaging of the omentum showed that OT-I/Nur77^GFP T cells had a significantly slower migration speed and a larger proportion exhibiting an arrested behavior around infected ZsGreen⁺ cells than the P14 T cells (Fig. 3E). Occasional stalling (a hallmark of APC:T cell interactions) of OT-I/Nur77^GFP T cells in direct contact with an infected ZsGreen⁺ cell was observed (Fig. 3F). To assess whether LPM could prime CD8⁺ T cells, uninfected (ZsGreen⁻) and infected (ZsGreen⁺) LPM from CPS-Cre-OVA-mCherry immunized Ai6 mice were purified and co-cultured with CTV-labeled OT-I/Nur77^GFP T cells. While LPM from naïve controls or uninfected LPM from vaccinated mice did not activate the OT-I/Nur77^GFP T cells, infected LPM induced OT-I T cells to express Nur77^GFP and undergo division (Fig. 3G–H). Likewise, when mice lacking β2-microglobulin (β2M) in macrophages (Lyz2-β2M mice) (fig. S3C) were vaccinated, the fraction of CD69⁺Nur77^GFP+ OT-I T cells was significantly decreased (Fig. 3I–J).

The ability of activated T cells to cluster around APCs is a hallmark of T cell priming. Imaging of the MS at 18 hpi with CPS-Cre-OVA-mCherry revealed clusters of OT-I T cells in the T cell zone (Fig. 3K) that contained CTV⁺Nur77^GFP+ OT-I T cells while CTV⁺Nur77^GFP- OT-I T cells appeared more dispersed (Fig. 3L). Random permutation tests of clustering were employed using K-function statistics to determine that Nur77^GFP+ OT-I T cells exhibited significant clustering compared to the total OT-I T cell population at a radius of k = 4 μm (Fig. 3M). This analysis also revealed that Nur77^GFP+ OT-I T cells were dispersed away from ZsGreen⁺ macrophages in Ai6 mice (Fig. 3N, P), and instead clustered around cDC1 (Snx22^GFP+ (23), Fig. 3O) at radii larger than 5 μm (Fig. 3Q, Movie S1). Thus, infected LPM present antigen via MHC Class I to naive CD8⁺ T cells and these activated CD8⁺ T cells then form secondary clusters around cDC1.

cDC1 are required for CPS-induced T cell responses

To assess the role of cDC1 in the generation of CPS-induced T cell responses, WT and Batf3^−/− mice were vaccinated and assessed at 8–11 dpi or >30 dpi. At acute time points, Batf3^−/− mice had a reduced population of parasite-specific CD4⁺ and CD8⁺ T cells in the spleen (Fig. 4A–D). Moreover, the ability of CPS-OVA to expand OT-I T cells was significantly decreased in Batf3^−/− mice (Fig. 4E). Similar results were observed with Irf8 +32^−/− mice, that lack cDC1 but are Batf3 sufficient (24) (Fig. 4F). The lack of cDC1 also resulted in a significant decrease in the number (Fig. 4G–H) and altered expression of CXCR3 and KLRG1 by parasite-specific CD8⁺ T cells in the peritoneum (Fig. 4I–J). This phenotypic difference was T cell-extrinsic as it was also observed in OT-I T cells primed in Batf3^−/− mice (Fig. 4K–L). By 30 dpi the endogenous parasite-specific CD8⁺ T cell population was still reduced in Batf3^−/− and Irf8 +32^−/− mice (Fig. 4M–N). When CPS-immunized mice were challenged with the virulent RH strain of T. gondii, WT mice were completely protected while immunized Batf3^−/− mice (Fig. 4O) or Irf8 +32^−/− mice succumbed to this challenge (fig. S3D).

Fig. 4. T cell responses to CPS require cDC1.

(A to D) WT and Batf3^−/− mice were immunized with CPS parasites. T cell responses in the spleen were assessed at 9 dpi. (A) Flow cytometric analysis of T. gondii-specific (tgd057:K^b+) CD8⁺ T cells.

(B) Quantification of tgd057:K^b+ CD8⁺ T cells.

(C) Flow cytometric analysis of T. gondii-specific (AS15:I-A^b+) CD4⁺ T cells.

(D) Quantification of AS15:I-A^b+ CD4⁺ T cells.

(E) Intravenous transfer of 5 × 10³ congenically distinct OT-I T cells into WT and Batf3^−/− mice that were immunized with CPS-OVA parasites 24 hr later. Quantification of OT-I T cells in the spleen at 8 dpi

(F) Quantification of the number of tgd057:K^b+ CD8⁺ T cells in the spleens of WT and Irf8 +32^−/− mice immunized i.p. with CPS parasites at 11 dpi.

(G to J) WT and Batf3^−/− mice were immunized i.p. with CPS parasites. T cell responses in the peritoneum were assessed at 9 dpi. (G) Flow cytometric analysis of tgd057:K^b+ CD8⁺ T cells.

(H) Quantification of tgd057:K^b+ CD8⁺ T cells.

(I) Flow cytometric analysis of the T_mem (CXCR3⁺KLRG1⁻) and T_eff (CXCR3⁻KLRG1⁺) phenotypes of tgd057:K^b+ CD8⁺ T cells.

(J) Quantification of the fraction of T_mem and T_eff phenotypes.

(K and L) Intravenous transfer of 5 × 10³ congenically distinct OT-I T cells into WT and Batf3^−/− mice that were immunized i.p. with CPS-OVA parasites 24 hr later. T cell responses in the peritoneum were assessed at 8 dpi. (K) Flow cytometric analysis of the T_mem and T_eff phenotypes of OT-I T cells.

(L) Quantification of the fraction of the T_mem and T_eff phenotypes of OT-I T cells.

(M and N) WT and Batf3^−/− (M) or Irf8 +32^−/− (N) mice were immunized with 10⁵ CPS parasites. Quantification of the number of tgd057:K^b+ CD8⁺ T cells in the spleen at 30 dpi.

(O) WT and Batf3^−/− mice were immunized with CPS parasites. At 30 dpi, immunized WT and Batf3^−/− mice and naïve WT mice were challenged with the lethal RH strain of T. gondii and survival assessed.

(P and Q) (P) Flow cytometric analysis of donor OT-I T cell populations sorted from CPS-OVA immunized WT (CD45.1.2⁺) and Batf3^−/− (CD45.1⁺) mice in the peritoneum of WT recipients infected with Pru-OVA i.p. at 12 dpi.

(Q) Quantification of the fraction of donor OT-I T cells from WT and Batf3^−/− mice in the peritoneum, spleen, and lungs of individual infected WT recipients.

Data are representative plots from (E, F, L, M, N, O, and Q) 2 or (B, D, H, and J) 3 independent experiments all using 4–5 mice. (B, D, E, F, H, J, L, M, and N) Bar graphs indicate mean±SD. (Q) Plots of paired donor cell fractions for each mouse. Data analyzed by (B, D, E, F, H, J, L, M, and N) unpaired T test or (Q) paired T test. *p < 0.05; **p < 0.01, ***p < 0.0001.

To compare the fitness of memory CD8⁺ T cells generated in the absence of cDC1, congenically distinct OT-I T cells were transferred into WT and Batf3^−/− mice that were then immunized with CPS-OVA. At 42 dpi, memory OT-I T cells were isolated from WT and Batf3^−/− hosts, mixed 1:1, and transferred into WT mice that were then infected with an OVA-expressing avirulent strain of T. gondii (Pru-OVA). Prior to transfer the OT-I T cells from Batf3^−/− hosts had a reduced central memory CD8⁺ T cell population (fig. S3E). Following re-challenge, CD45.1⁺ OT-I T cells from the Batf3^−/− hosts had a significant defect in expansion compared to CD45.1.2⁺ OT-I T cells from WT mice (Fig. 4P–Q, fig. S3F). Thus, the absence of cDC1 results in a defect in number and phenotype of parasite-specific CD4⁺ and CD8⁺ T cells at acute and memory time points and a defect in resistance to rechallenge.

Role of IL-12 and cDC1 in CD4⁺ and CD8⁺ T cell expansion

Because cDC1 are an important source of IL-12 that promotes T cell responses during toxoplasmosis (22), experiments were performed to test if treatment of Batf3^−/− mice with IL-12 would restore the parasite-specific T cell response. In WT mice, treatment with rIL-12 did not affect the AS15:I-A^b+ CD4⁺ T cell response, but restored the frequency and enhanced numbers of these CD4⁺ T cells in Batf3^−/− mice (Fig. 5A–C). In contrast, treatment of Batf3^−/− mice with rIL-12 had a small but insignificant impact on the frequency, number, and differentiation of tgd057:K^b+ CD8⁺ T cells (Fig. 5D–I). Thus, cDC1 have an IL-12-independent role in the induction of a parasite-specific CD8⁺ T cell response.

Fig. 5. IL-12-independent role for cDC1s in the CD8⁺ T cell response to CPS.

(A to I) WT and Batf3^−/− mice were immunized with CPS parasites followed by i.p. administration of either PBS or 200 ng rIL-12 at 0, 24, and 48 hpi. T cell responses in the peritoneum were assessed via flow cytometry at 11 dpi. (A) Representative plots of AS15:I-A^b+ CD4⁺ T cells.

(B and C) The fraction and number of AS15:I-A^b+ CD4⁺ T cells.

(D) Representative plots of tgd057:K^b+ CD8⁺ T cells.

(E and F) The fraction and number of tgd057:K^b+ CD8⁺ T cells. Data (n = 6–8) represent data from 2 independent experiments.

(G) Representative plots of the T_eff and T_mem phenotypes of tgd057:K^b+ CD8⁺ T cells.

(H and I) The fraction and number of T_mem or T_eff phenotypes of tgd057:K^b+ CD8⁺ T cells. Data in (B, C, E, and F) are pooled from 2 independent experiments with 3–5 mice, and (H and I) are representative of 2 independent experiments with 3–5 mice. Bar graphs indicate mean±SD. All statistical comparisons were unpaired T test. ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

To investigate how the absence of cDC1 impacted the T cell response, OT-I/Nur77^GFP T cells were transferred into WT and Batf3^−/− mice that were then immunized with CPS-OVA. At 18 hpi, OT-I/Nur77^GFP T cells showed similar levels of activation (CD69⁺Nur77^GFP+) in the omentum of WT and Batf3^−/− mice (Fig. 6A–B), but Nur77^GFP+ OT-I T cells did not cluster in Batf3^−/− mice (Fig. 6C–D). Next, Xcr1-β2M mice that lack β2M on cDC1 (Fig. 6E–F) showed a decreased acute OT-I T cell response after CPS-OVA vaccination (Fig. 6G–H). Analysis of the kinetics of CTV-labelled OT-I T cell response in multiple tissues of WT and Batf3^−/− mice revealed that OT-I T cells were not detected in large numbers in the DLN or spleen until they had undergone several divisions (fig. S3G). In the omentum at 2 dpi the activation and proliferative response of OT-I T cells in the Batf3^−/− and WT mice were similar. However, by 3 dpi a significantly smaller fraction of OT-I T cells had divided in the Batf3^−/− mice (Fig. 6I–J). Thus, in the absence of cDC1 there is no defect in initial activation of OT-I T cells, but cDC1 are required to present antigen to generate a maximal CD8⁺ T cell response.

Fig. 6. cDC1 are not required for CD8⁺ T cell activation.

(A and B) Intraperitoneal transfer of 10⁶ congenically distinct OT-I/Nur77^GFP T cells into WT and Batf3^−/− mice that were immunized with CPS or CPS-OVA parasites 2 hr later. (A) Flow cytometric analysis of OT-I/Nur77^GFP T cells.

(B) Quantification of the fraction of activated (CD69⁺Nur77^GFP+) OT-I/Nur77^GFP T cells in the omentum.

(C and D) Imaging of 5 × 10⁵ i.p. transferred CTV-labeled OT-I/Nur77^GFP T cells clusters in the MS of Ai6/Batf3^−/− mice that were immunized with CPS-Cre-OVA-mCherry parasites at 18 hpi. (C) Images of CTV⁺ OT-I/Nur77^GFP T cell clustering. CTV (red), ZsGreen, Nur77^GFP (green), CD3 (gray), CPS-Cre-OVA-mCherry (magenta), B220 (blue).

(D) K-function statistical analysis of Nur77^GFP+ OT-I T cell clustering in MS from WT mice (cyan lines, taken from Fig. 3M) and Batf3^−/− mice (black lines). Analysis in Batf3^−/− mice was done for 7 individual MS from 2 individual mice.

(E to H) Intravenous transfer of 5 × 10³ OT-I/Nur77^GFP T cells into β2M^fl/fl and Xcr1-β2M mice immunized with CPS-OVA parasites 24 hr later. (E) Flow cytometric analysis of the efficiency of β2M deletion in cDC1 in the spleen of β2M^fl/fl and Xcr1-β2M mice at 7 dpi.

(F) Quantification of the fraction of β2M⁺ cDC1 in the spleen at 7 dpi.

(G) Flow cytometric analysis of the OT-I T cell population in β2M^fl/fl and Xcr1-β2M mice in the spleen at 7 dpi.

(H) Quantification of OT-I T cells in the spleen at 7 dpi.

(I and J) Intraperitoneal transfer of 5 × 10⁵ CTV-labeled OT-I/Nur77^GFP T cells into WT and Batf3^−/− mice immunized with CPS-OVA parasites 2 hr later. (I) Flow cytometric analysis of the activation (Nur77^GFP+) and division (CTV dilution) of OT-I T cells in the omentum at 2 and 3 dpi.

(J) Quantification of the fraction of OT-I T cells that have undergone division.

Data are representative plots from (B, F, H, and J) 2 or (D) 4 independent experiments all using 3–6 mice. Bar graphs indicate mean±SD. Data analyzed by unpaired T test. ns = not significant, *p < 0.05; **p < 0.01, ***p < 0.001, ****p < 0.0001.

Agent-based modeling of T cell priming and expansion

To understand how absence of cDC1 affects the CD8⁺ T cell response, an agent-based stochastic model – the STochastic Omentum REsponse (STORE) model – was developed to identify how perturbations to the system would affect CD8⁺ T cell responses. The different steps of the T cell response are mostly described by probability distribution functions based on literature values (Fig. 7A, Supplementary Methods). Six model parameters were derived from our experiments: 1) the number of transferred OT-I, 2) the ratio of transferred OT-I to the number of APCs generated by CPS-OVA (r_AP), the probabilities of 3) an OT-I and APC binding (p_AP,bind), 4) an APC leaving the system (p_BP,leave), 5) cells in divisions 3–6 leaving (p_EH,leave), and 6) cells in divisions 7–8+ leaving (p_IJ,leave). After CPS-OVA immunization, the use of this model in a dynamic simulation tracked the fate of individual OT-I T cells in the omentum of each mouse for 8 days. The stochasticity and fidelity of the STORE model are illustrated in Fig. 7B–E where the experimental average cell counts of OT-I T cells at 2, 3, 4, 6, and 8 dpi (table S1) are plotted with a simulation run (table S4). In this example, the APC population is rate limiting and allows naïve OT-I T cells (A) to accumulate rapidly over the first 30 hours and then decay exponentially for the remainder of the 8-day period (Fig. 7B). The number of free APCs (P) remains low throughout the simulation as any available APC almost immediately pairs with a naïve OT-I T cell (AP) (Fig. 7C) and converts to a Nur77^GFP+ OT-I:APC pair (BP). The duration of the BP population is roughly 24 hours to approximate the second phase of DC:T cell interactions during priming (25). The BP trajectory follows the linear decrease of the APC population over time as approximated by the model (p_BP,leave). Simulation of divisions 1–6 (Fig. 7D–E) reflect the 24-hour T cell priming cycle of BP and the 5.3 hour division cycle (26). The effect of the linear reduction of BP over time is evidenced by the decreasing cell counts in divisions 1–6 (Fig. 7D–E). Division 8+ (J) has a different profile (Fig. 7E), because these cells are all CTV⁻ and continue dividing (27) making J the sum of all divisions ≥8.

Fig. 7. Stochastic modeling implicates defect in OT-I expansion in Batf3^−/− mice.

(A) Schematic of STORE model of OT-I T cell activation depicting the set (red text) and variable (purple text) variables.

(B to P) Intraperitoneal transfer of 5 × 10⁵ OT-I/Nur77^GFP T cells into WT and Batf3^−/− mice immunized with CPS-OVA parasites 2 hr later.

(B-E) Representative plots from a simulation of the STORE model where solid lines indicate simulation data and filled circles indicate number of OT-I T cells in each group measured by flow cytometry.

(B) The number of naïve OT-I T cells over time.

(C) The numbers of unbound APCs and APC binding pairs with Nur77^GFP- (AP) and Nur77^GFP+ (BP) OT-I T cells over time.

(D) The number OT-I T cells in divisions 1–4 versus time.

(E) The number of OT-I T cells in divisions 5–8+ versus time.

(F) The number of OT-I T cells at different time points from simulations and experimental data (average) in WT and Batf3^−/− recipient mice.

(G) Flow cytometric analysis of OT-I T cell death (Annexin V⁺ PI⁺) in divisions 3 and 4 in WT and Batf3^−/− mice.

(H) The fraction of dead OT-I T cells in divisions 3 and 4 in WT and Batf3^−/− mice.

(I) UMAP projection of OT-I T cells from WT and Batf3^−/− mice.

(J) Heat map overlay of CTV and expression levels of CD69, CD11a, CD71, CD98, and CD25 by OT-I T cells from WT and Batf3^−/− mice on UMAP projection from (I).

(K-L) Flow cytometric analysis of the expression of CD71, and MFI of CD71 expression by divisions 3–6 of OT-I T cells from WT and Batf3^−/− mice.

(M-N) Flow cytometric analysis of the expression of CD98, and MFI by divisions 3–6 of OT-I T cells from WT and Batf3^−/− mice.

(O-P) Flow cytometric analysis of the expression of CD25, and quantification of the fraction of CD25⁺ OT-I T cells in divisions 3–6 from WT and Batf3^−/− mice.

STORE, STochastic Omentum REsponse; PI, propidium iodide. Data are representative plots from (F, H) 2 or (L, N, P) 6 independent experiments all using 4–5 mice. Bar graphs indicate mean±SD. Data analyzed by unpaired T test. *p < 0.05; **p < 0.01, ***p < 0.001, ****p < 0.0001.

The comparison of simulations of the kinetics of the OT-I T cell response in WT and Batf3^−/− mice using the STORE model with actual experimental data showed that the number of each T cell division at each time point matched well (Fig. 7F). This evaluation highlighted that the major difference between WT and Batf3^−/− mice is the probability of OT-I T cells in divisions 3–6 leaving the system by 68 hpi. When OT-I T cells in the omentum were examined for differing rates of cell death at 3 dpi, cells in divisions 3 and 4 showed increased cell death in Batf3^−/− compared to WT mice (Fig. 7G–H). The blastogenic T cell response requires increased metabolic activity driven by aerobic glycolysis and mitochondrial activity, enhanced responsiveness to IL-2, and nutrient uptake. In OT-I T cells that had become activated and begun to proliferate in WT or Batf3^−/− mice, there was no difference in mitochondrial mass or surface expression of the glucose uptake receptor Glut1 (fig. S3H–I). UMAP analysis of expression of a panel of T cell activation-dependent receptors and CTV dilution at 3 dpi differentiated between OT-I T cells from WT versus Batf3^−/− recipients (Fig. 7I). While expression levels of CD69 and CD11a showed little difference between these populations, the nutrient uptake receptors CD71 and CD98, as well as the cytokine receptor IL-2Rα (CD25), showed increased expression in OT-I cells from WT recipients (Fig. 7J). The expression level of CD71, CD98, and CD25 was significantly higher in divisions 3–6 of OT-I cells in WT mice compared to those from Batf3^−/− mice (Fig. 7K–P). Together, the agent-based model and experimental data sets indicate that after T cell priming cDC1 promote entry of CD8⁺ T cells into a pro-survival expansion phase, and that the loss of cDC1 resulted in a defect in the expression of nutrient and IL-2 receptors that support the T cell proliferative burst (28, 29).

DISCUSSION

The importance of WAT and the FALCs located therein to the immune response is appreciated, but reports that FALCs lack organized T and B cell regions, follicular dendritic cells (FDCs), and a central reticular network suggest FALCs are not capable of the same functions as SLOs (7). However, adaptive immune responses thought to occur exclusively in SLOs can be initiated in the omentum (7) and fat-resident memory CD8⁺ T cells benefit from the local metabolic resources (3). Here, we demonstrate cDC1 in FALCs have the same capacity as their counterparts in SLOs to coordinate changes in cellularity associated with ILC activation and monocyte mobilization along with CD8⁺ T cell priming and expansion. These findings illustrate the dynamic nature of the FALCs and highlight them as a site of interactions between the innate and adaptive responses. Given the differences in organization between the SLO and FALCs, what are the features of these lymphoid aggregates that allows them to accomplish the same functions as SLO? First, the placement of FALCs at a serosal surface facilitates the collection of antigens leaving the peritoneum via convective flow or trafficking by immune cells through the stomata in the mesothelium into the FALCs (9, 12). Second, stromal cell chemokine production recruits ILCs to the FALCs (6, 7, 9–11). Third, while tissue-resident macrophages are considered nonmigratory, the macrophages resident in body cavities do not adhere to this principle and can access other tissues (30). This migratory capacity allows these macrophages to deliver antigen to the FALCs. Perhaps, FALCs should be considered as evolutionarily conserved, mesoderm-derived structures that are specialized for surveillance of cavities and are a relevant site for initiation of T- and B-cell-mediated immunity.

The ability of antigen-laden cDC to migrate from tissues to different regions of SLOs is considered critical to initiate and coordinate T cell activation (14). Utilization of Cre-secreting CPS parasites revealed infected peritoneal macrophages that migrate into the FALCs were sufficient for initial priming of the CD8⁺ T cells. This migration of infected macrophages is likely enhanced by parasite effector proteins that promote leukocyte migration (31). This unanticipated conclusion has parallels to studies that examined the events associated with T cell priming following subcutaneous introduction of VV. In that setting, early activation of CD8⁺ T cells was associated with infected macrophages in the subcapsular sinus (SCS) of the draining LN before T cell cluster formation around cDC adjacent to the macrophage-rich zone (32). A subsequent study utilized VV to show that infected cDC populations mediate CD8⁺ T cell priming, and was the only prior study to show that cDC1 in the spleen and LN were not required for this priming (13). The topic of which cell type is involved in the initial presentation of cognate peptide to T cells is complicated. For example, strategies that target antigen to cDC subsets (33) reveal the function of cDC but do not reflect the biology of different infections. Because many pathogens vary in their cellular tropism and microbial evasion mechanisms that target antigen presentation, it seems likely that multiple mechanisms to allow T cell priming would exist.

A common feature of studies that imaged early cDC:T cell interactions during infection is the observation that “secondary” cluster formation occurs around cDC that were not infected, but it was unclear if cDC that nucleated the clusters had been previously infected or whether they were presenting antigen. The observation that cDC1 enriched in the T cell zone of the MS were associated with clusters of activated OT-I T cells illustrates a process common to FALCs and SLO. The use of the Cre-expressing parasites established that these cDC1 had neither been infected nor injected by the parasites, but the loss of CD8⁺ T cell responses in the Xcr1-β2M mice indicated that these cDC1 present antigen and do not utilize MHC Class I dressing reported in other systems (34, 35). Given the ability of cDC1 to cross-present antigen, it seems likely that uninfected cDC1 acquire antigen secreted from the parasite or released upon death of infected cells. Regardless, these cDC provide a platform to facilitate T cell help (13, 15, 17) and crosstalk between CD8⁺ T cells (36). How absence of this activity affects T cell proliferation and differentiation is not well understood. The use of differential equations has helped to understand T cell activation, expansion and differentiation pathways (27, 37, 38). However, these deterministic models do not account for stochasticity in biological samples and are poorly suited for the analysis of small numbers of cells or to capture rare events such as initial APC:T cell interactions (39). To better understand how loss of cDC1 impacts T cell expansion, the agent-based STORE model was developed. Importantly, this stochastic model required only a small number of parameters to recapitulate the early in vivo response of OT-I T cells and predicted that a loss of cDC1 would lead to an increased rate of cell death of OT-I T cells in their early divisions. This result led to the recognition that early cDC1:CD8⁺ T cell interactions promote expression of nutrient and cytokine receptors that support T cell proliferation required for the generation of memory CD8⁺ T cell responses. The development of an accurate stochastic model of early T cell responses provides an opportunity to test how changes in T cell priming impact memory formation and should allow for the directed design of vaccination strategies to elicit a protective cellular immune response.

MATERIALS AND METHODS

Study design

This study was designed to understand the role of cDC1 in orchestrating the generation of a protective memory CD8⁺ T cell population in the FALCs of the omentum in response to vaccination with a non-replicating strain of T. gondii. To do so, we primarily compared the immune response to vaccination in WT mice and mice lacking cDC1 (Batf3^−/− and Irf8 +32^−/− mice). Measurements of CD8⁺ T cells at acute and memory time points in these different mouse models demonstrated the importance of cDC1 in generating these populations. The early priming events of CD8⁺ T cells in the omentum were thus examined and an agent-based model was developed to understand the kinetics and mechanism of these differences. Endpoint analyses included flow cytometry, survival curves, fluorescence imaging, cluster analysis using K-function statistics, and agent-based modeling. All experiments used randomly assigned mice without investigator blinding. All data points and n values reflect biological replicates, and no data was excluded. The specific numbers and genotypes of the mice, the number of replicates, and the statistics performed are included in the legend of each figure.

Mice

All procedures involving mice were reviewed and approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania (Animal Welfare Assurance Reference Number #A3079-01) and were in accordance with the guidelines set forth in the Guide for the Care and Use of Laboratory Animals of the National Institute of Health.

Batf3^−/− (Stock #013755, RRID:IMSR_JAX:013755), Il12b^−/− (Stock #002693, RRID:IMSR_JAX:002693), Ai6 (Stock #007906, RRID:IMSR_JAX:007906), Nur77^GFP (Stock #016617, RRID:IMSR_JAX:016617), OT-I (Stock #003831, RRID:IMSR_JAX:003831), Irf8 +32^−/− (Stock #032744, RRID:IMSR_JAX:032744), Rag1^−/− (Stock #002216, RRID:IMSR_JAX:002216), and Lyz2-Cre (Stock #004781, RRID:IMSR_JAX:004781) mice were obtained from Jackson Laboratories. C57BL/6 (Stock #B6-F/M) and Rag2^−/−γc^−/− (Stock #4111) mice were obtained from Taconic Biosciences (RRID:SCR_016410). CD45.1 (Stock #564) mice were obtained from Charles River Laboratories. Snx22^GFP/+ mice (on a 129S1 background) were produced by standard homologous recombination in ES cells (54), and were a kind gift from Dr. Kenneth M. Murphy at Washington University in St. Louis. Snx22^GFP/+ mice used here were an F1 cross with C57BL/6 mice. Xcr1-Cre mice were generated previously (30) and were a kind gift from Dr. Kenneth M. Murphy at Washington University in St. Louis. Floxed β2-microglobulin mice were obtained from Dr. Dorian B. McGavern at the NIH in Bethesda. All mice are on a C57BL/6 background unless otherwise noted. For IL-12p70 add back experiments, C57BL/6 and Batf3^−/− mice were treated with an intraperitoneal (i.p.) injection of either PBS (Corning: 21-0311-CM) or 200 ng of recombinant murine IL-12p70 (PeproTech: 210-12) immediately after immunization and once per day for the next two days after immunization. All mice were kept in specific-pathogen-free conditions at the School of Veterinary Medicine at the University of Pennsylvania.

Immunizations and infections

All Toxoplasma gondii immunizations were performed using 1–2 × 10⁵ cpsII parasites, cpsII-OVA parasites, or cpsII-Cre-OVA-mCherry parasites. Infectious challenge experiments were performed using 10⁴ Pru-OVA parasites or RH parasites. CpsII-OVA parasites have been previously described (36) and were derived from the RHΔcpsII clone, which was provided as a generous gift by Dr. David Bzik. CpsII-Cre-OVA-mCherry parasites were derived from the cpsII-OVA clone using the previously described methods (36, 43). Pru-OVA parasites have also been described previously (36). Parasites were cultured and maintained by serial passage on human foreskin fibroblast cells in the presence of parasite culture media [71.7% DMEM (Corning: 10-017-CM), 17.9% Medium 199 (Gibco: 11150-059), 9.9% Fetal Bovine Serum (FBS)(Atlanta Biologics: S11150H), 0.45% Penicillin and Streptomycin (Gibco: 15140-122)(final concentration of 0.05 units/ml penicillin and 50 μg/ml streptomycin), 0.04% gentamicin (Gibco: 15750-060)(final concentration of 0.02 mg/ml gentamicin)], which was supplemented with uracil (Sigma-Aldrich: U1128) (final concentration of 0.2 mM uracil). For infections, parasites were harvested and serially passed through 18 and 26 gauge needles (BD: 305196, 305115) before filtration with a 5 μM filter (PALL Acrodisc: 4650). Parasites were washed extensively with PBS and mice were injected i.p. with parasites suspended in PBS.

T cell transfers and tissue harvesting

For T cell transfers OT-I mice were interbred with CD45.1/Nur77^GFP mice or CD45.1 mice. To isolate OT-I or P14 CD8⁺ T cells, lymph nodes and spleen were harvested and leukocytes from the spleen and draining lymph nodes were obtained by processing spleens and lymph nodes over a 70 μm filter (Fisher Scientific: 22-363-548) and washing them in complete RPMI [90% RMPI 1640 (Corning: 10-040-CM), 10% FBS, 1% penicillin-streptomycin, 1 mM sodium pyruvate (Corning: 25-000-Cl), 1% nonessential amino acids (Gibco: 11140-050), and 0.1% β-mercaptoethanol (Gibco: 21985-023)]. Red blood cells were then lysed by incubating for 5 minutes at room temperature in 5 ml of lysis buffer [0.864% ammonium chloride (Sigma-Aldrich: A0171) diluted in sterile de-ionized H₂O)], followed by washing with complete RPMI. For transfer of low numbers of OT-I CD8⁺ T cells, splenocytes were rinsed with PBS and total splenocytes containing 5 × 10³ or 5 × 10⁴ OT-I T cells were transferred by i.v. injection into recipient mice. For the transfer of high numbers of OT-I CD8⁺ T cells, cells were then purified by magnetic activated cell sorting (MACS) using the CD8a+ T Cell Isolation Kit (Miltenyi Biotec: 130-104-075). For cell division or microscopy studies, purified OT-I or P14 T cells were then fluorescently labeled using the CellTrace Violet (CTV, ThermoFisher Scientific: C34557) or CellTrace Yellow (CTY, ThermoFisher Scientific: C34567) labeling kits. OT-I T cells were then transferred by either i.v. or i.p. injection into recipient mice.

For the transfer of memory OT-I T cells, total splenocytes containing 5 × 10⁴ WT CD45.1⁺ OT-I T cells were i.v. transferred into Batf3^−/− mice and total splenocytes containing 5 × 10⁴ WT CD45.1.2⁺ OT-I T cells were i.v. transferred into WT mice 1 day prior to immunization with 2 × 10⁵ CPS-OVA parasites. At 42 dpi, spleens from WT and Batf3^−/− hosts were harvested and processed as above. Single cell suspensions from mice of each host genotype were then pooled and resuspended in sterile FACS buffer. OT-I T cells from WT and Batf3^−/− hosts were then FACS purified and mixed 1:1 for the i.v. transfer of 2 × 10⁴ total OT-I T cells into naïve WT mice. These recipient WT mice were then infected i.p. 24 hours later with 10⁴ Pru-OVA tachyzoites.

Peritoneal exudate cells were obtained by peritoneal lavage with 8 mL of ice cold PBS. Omentum was isolated, incubated in 0.4 U/mL of LiberaseTL (Roche: 5401020001) for one hour at 37°C, passed through an 18G needle, and processed over a 70 μm filter. Leukocytes from the spleen and draining lymph nodes were obtained by processing spleens and lymph nodes, washing them in complete media, and lysing red blood cells (see above). Cells were then resuspended in complete RPMI.

T cell priming in vitro

For in vitro assays measuring priming of naïve CD8⁺ T cells by infected macrophages, Ai6 mice were immunized i.p. with 5 × 10⁵ CPS-Cre-OVA-mCherry parasites. At 18 hpi, peritoneal exudate cells were harvested from naïve or immunized Ai6 mice. Large peritoneal macrophages (LPM) were then FACS purified after double sorting to isolate naïve LPM from naïve Ai6 mice or uninfected (ZsGreen⁻) and infected (ZsGreen⁺) LPM from 5 different immunized Ai6 mice. These purified populations were then plated in a 96-well plate at 5 × 10³ cells per well. Naïve OT-I/Nur77^GFP T cells that had been MACS purified and labeled with CTV were then added at 4 × 10⁴ cells per well to each well containing LPM. OT-I/Nur77^GFP T cell activation was then measured 72 hours later.

Omentectomy surgeries

The surgical removal of the omentum was performed using standard aseptic technique. Prior to surgery, hair was removed from the ventral side of mice and a single dose of 0.1 mg/kg of buprenorphine was administered by i.p. injection. Mice were then sedated using isoflurane in an induction box, and anesthesia was maintained with a face mask with a constant flow of 1–2% isoflurane. Mice were then placed on a sterile surgical cloth covering a heating pad to provide heat support. The surgical site was then cleaned using a chlorhexidine wipe. A suprapubic midline laparotomy was performed to visualize and expose the omentum. The omentum was then removed using a cauterization pen to sever the connective tissue and blood vessel at each end of the omentum. For sham surgeries, an abdominal incision was made and the omentum was exposed but not removed. The peritoneal wall was closed using 5-0 synthetic absorbable sutures, and the overlaying skin was closed using the same sutures as well as surgical glue. The mice were observed and allowed to recover in a clean cage with a heating pad beneath for heat support until resuming normal activity. Buprenorphine was given to all mice postoperatively at 0.1 mg/kg every 4 hours via subcutaneous injection for the first 48 hours after surgery. Animals were monitored to assess for signs of pain or discomfort and wounds were examined for signs of infection every day for the remainder of the experiment.

Intravital imaging

Ai6 mice were immunized i.p. with 10⁵ CPS-Cre-OVA-mCherry parasites, and 18 hours later 2 × 10⁵ naïve, CTV-labeled OT-I/Nur77^GFP T cells and 2 × 10⁵ naïve, CTY-labeled P14 T cells were transferred via i.p. injection. Starting 2 hours later, recipient Ai6 mice were anesthetized with isoflurane and maintained at a core temperature of 37°C. The peritoneum was surgically opened to access the omentum, and a small portion of visceral adipose tissue was resected to assist in immobilizing the omentum via a soft-tissue vacuum apparatus (VueBio). No greater than 40 kPa of pressure was applied to steady the omentum for imaging. Image acquisition was performed on a Leica SP8 multiphoton confocal with a 20X water immersion objective (1.0 NA) with resonant scanner (8000 kHz) and 4 external HyD detectors. The excitation wavelength of Chameleon Vision II Ti:Sapphire laser was tuned for optimal detection of the labeled OT-I and P14 T cells in each experiment, typically 880 nm. Images collected x-, y-, and z-plane data over time with a step size of 2 μm and a z-thickness that allowed for a complete z-series every 22s. This was carried out for approximately 30 minutes for each region imaged. Resulting images were segmented in Imaris (v9.7.2, Bitplane) with spot-specific position data exported and analyzed in R with the CellTrackR package.

Flow cytometry

Cells were washed with FACS buffer [1× PBS, 0.2% bovine serum antigen (Gemini: 700-100P), 1 mM EDTA (Gibco: 15575-038)] and incubated in Fc block [99.5% FACS Buffer, 0.5% normal rat IgG (Invitrogen: 10700), 1 μg/ml 2.4G2 (BioXCell: BE0307)] at 4°C for 10 min prior to staining. If cells were stained for cell death using LIVE/DEAD staining, LIVE/DEAD Fixable Aqua Dead Cell marker (Invitrogen: L34957), GhostDye Violet 510 Viability Dye (TONBO Biosciences: 13-0870-T100), or GhostDye Red 780 Viability Dye (TONBO Biosciences: 13-0865-T100) was included during incubation with Fc block. Tetramer-specific CD8⁺ and CD4⁺ T cells were measured by staining in 50 μL FACS buffer containing MHCI and MHCII tetramers specific for endogenous T. gondii antigens for 1 hour at room temperature. Cells were surface stained in 50 μL at 4°C for 15–20 min and washed in FACS buffer prior to acquisition. For intracellular cytokine staining, 10⁶ cells from the peritoneal exudate or omentum were plated in a 96-well plate and incubated with 1X Brefeldin A (Sigma-Aldrich: B7651) at 37°C for 3 hr. For intracellular cytokine and transcription factor staining, cells were rinsed with FACS buffer and surface stained as described above, fixed using the eBioscience Foxp3 Transcription Factor Fixation/Permeabilization Concentrate and Diluent (ThermoFisher Scientific: 00-8222) for 30 min at 4°C, and then washed with FACS buffer. Cells were then stained for intracellular cytokines and transcription factors in 50 μL 1X eBioscience Permeabilization Buffer (ThermoFisher Scientific: 00-8333-56) at 4°C for at least 1 hr. Cells were then washed in FACS buffer prior to acquisition. The APC Annexin V Kit (BioLegend: 640920) was used to measure cell death via Annexin V and propidium iodide (PI). Cells were surface stained and washed with FACS buffer then resuspended in 100 μL Annexin V binding buffer (BioLegend: 79998). Cells were then incubated with 5 μL APC Annexin V (BioLegend: 640920) and 10 μL PI solution (BioLegend: 79997) for 15 min at room temperature in the dark. Prior to acquisition, 400 μL of Annexin V binding buffer was added to each sample. For mitochondrial staining 2 × 10⁶ cells were suspended in 100 μL HBSS (Gibco: 14175-079) supplemented with 0.5 mM MgCl₂ (Sigma-Aldrich: M4880). Mitochondrial stains were then added at the following concentrations: 50 nM MitoTracker Green FM (Invitrogen: M7514), 25 nM MitoTracker Deep Red FM (Invitrogen: M22426) and 5 mM MitoSOX Red (Invitrogen: M36008). Samples were incubated at 37°C for 30 min, rinsed with FACS buffer, and surface stained as above prior to acquisition.

The following antibodies were used for staining: B220: BV510, BioLegend: 103248, clone: RA3-6B2, RRID:AB_2650679; B220: ef450, eBioscience: 48-0452-82, clone: RA3-6B2, RRID:AB_1548761; B220: PerCp-Cy5.5, BioLegend: 103236, clone: RA3-6B2, RRID:AB_893356; CCR2: AF647, BioLegend: 150604, clone: SA203G11, RRID:AB_2566140; CD102: BUV395, BD Biosciences: 740227, clone: 3C4 (mIC2/4), RRID:AB_2739975; CD102: ef450, eBioscience: 48-1021-82, clone: 3C4 (mIC2/4), RRID:AB_1548749; CD102: FITC, BioLegend: 105606, clone: 3C4 (mIC2/4), RRID:AB_313199; CD115: BV605, BioLegend: 135517, clone: AF398, RRID:AB_2562760; CD11a: PerCp-Cy5.5, BioLegend: 101124, clone: M17/4, RRID:AB_2562932; CD11b: APC-ef780, eBioscience: 47-0112-82, clone: M1/70, RRID:AB_1603193; CD11b: BV605, BioLegend: 101257, clone: M1/70, RRID:AB_2565431; CD11b: BV650, BioLegend: 101259, clone: M1/70, RRID:AB_2566568; CD11c: APC-Fire750, BioLegend: 117352, clone: N418, RRID:AB_2572124; CD11c: APC-R700, BD Biosciences: 565872, clone: N418, RRID:AB_2744277; CD11c: BUV737, BD Biosciences: 612797, clone: HL3, RRID:AB_2870124; CD172a: APC, BioLegend: 144014, clone: P84, RRID:AB_2564061; CD172a: APC-Cy7, BioLegend: 144018, clone: P84, RRID:AB_2629558; CD19: BV605, BioLegend: 115539, clone: 6D5, RRID:AB_11203538; CD19: ef450, eBioscience: 48-0193-82, clone: eBio1D3, RRID:AB_2734905; CD19: PerCp-Cy5.5, BioLegend: 152406, clone: ID3/CD19, RRID:AB_2629815; CD200R: APC, eBioscience: 17-5201-82, clone: OX110, RRID:AB_10717289; CD25: APC, eBioscience: 17-0251-82, clone: PC61.5, RRID:AB_469366; CD25: PE-Cy7, eBioscience: 25-0251-82, clone: PC61.5, RRID:AB_469608; CD3: APC-ef780, Invitrogen: 47-0032-82, clone: 17A2, RRID:AB_1272181; CD3: BV605, BioLegend: 100237, clone: 17A2, RRID:AB_2562039; CD3: BV750, BioLegend: 100249, clone: 17A2, RRID:AB_2734148; CD3: BV785, BioLegend: 100232, clone: 17A2, RRID:AB_2562554; CD3: PacBlue, BioLegend: 100214, clone: 17A2, RRID:AB_493645; CD3e: PE-cf594, BD Biosciences: 562286, clone: 145-2C11, RRID:AB_11153307; CD3e: PerCp-Cy5.5, Invitrogen: 45-0031-82, clone: 145-2C11, RRID:AB_1107000; CD335: BV650, BioLegend: 137635, clone: 29A1.4, RRID:AB_2734200; CD335: PE-Dazzle594, BioLegend: 137630, clone: 29A1.4, RRID:AB_2616666; CD4: AF700, BioLegend: 100536, clone: RM4-5, RRID:AB_493701; CD4: BV650, BioLegend: 100555, clone: RM4-5, RRID:AB_2562529; CD4: BV711, BioLegend: 100447, clone: GK1.5, RRID:AB_2564586; CD4: BV785, BioLegend: 100552, clone: RM4-5, RRID:AB_2563053; CD44: BV605, BD Biosciences: 563058, clone: IM7, RRID:AB_2737979; CD44: BV785, BioLegend: 103059, clone: IM7, RRID:AB_2571953; CD45: BUV805, BD Biosciences: 748370, clone: 30-F11, RRID:AB_2872789; CD45.1: APC-ef780, Invitrogen: 47-0453-82, clone: A20, RRID:AB_1582228; CD45.1: BV711, BioLegend: 110739, clone: A20, RRID:AB_2562605; CD45.1: ef450, Invitrogen: 48-0453-82, clone: A20, RRID:AB_1272189; CD45.1: PE, BD Biosciences: 553776, clone: A20, RRID:AB_395044; CD45.1: PE-cf594, BD Biosciences: 562452, clone: A20, RRID:AB_11152958; CD45.1: PerCp-Cy5.5, Invitrogen: 45-0453-82, clone: A20, RRID:AB_1107003; CD45.2: APC-ef780, Invitrogen: 47-0454-82, clone: 104, RRID:AB_1272175; CD45.2: BV711, BioLegend: 109847, clone: 104, RRID:AB_2616859; CD49b: FITC, BD Biosciences: 553857, clone: DX5, RRID:AB_395093; CD5: ef450, eBioscience: 48-0051-82, clone: 53-7.3, RRID:AB_1603250; CD5: PerCp-Cy5.5, eBioscience: 45-0051-82, clone: 53-7.3, RRID:AB_914334; CD62L: APC-ef780, eBioscience: 47-0621-82, clone: MEL-14, RRID:AB_1603256; CD62L: BV711, BioLegend: 104445, clone: MEL-14, RRID:AB_2564215; CD64: PE-Cy7, BioLegend: 139314, clone: X54-5/7.1, RRID:AB_2563904; CD69: BV711, BioLegend: 104537, clone: H1.2F3, RRID:AB_2566120; CD69: ef450, Invitrogen: 48-0691-82, clone: H1.2F3, RRID:AB_10719430; CD69: PE-Cy7, BD Biosciences: 552879, clone: H1.2F3, RRID:AB_394508; CD69: PerCp-Cy5.5, eBioscience: 45-0691-82, clone: H1.2F3, RRID:AB_1210703; CD71: PE, BioLegend: 113807, clone: R17217, RRID:AB_313568; CD8a: BV605, BioLegend: 100744, clone: 53-6.7, RRID:AB_2562609; CD8a: BV650, BioLegend: 100742, clone: 53-6.7, RRID:AB_2563056; CD8a: BV711, BioLegend: 100747, clone: 53-6.7, RRID:AB_11219594; CD8b: APC-ef780, Invitrogen: 47-0083-82, clone: ebioH35-17.2, RRID:AB_2573943; CD8b: FITC, BioLegend: 126606, clone: YTS156.7.7, RRID:AB_961295; CD90.2: AF700, BioLegend: 105320, clone: 30-H12, RRID:AB_493725; CD98: PE-Cy7, BioLegend: 128214, clone: RL388, RRID:AB_2750547; CD117/c-kit: APC-ef780, Invitrogen: 47-1172-80, clone: ACK2, RRID:AB_1582227; CXCR3: BV650, BioLegend: 126531, clone: CXCR3-173, RRID:AB_2563160; CXCR3: PE-Cy7, BioLegend: 126516, clone: CXCR3-173, RRID:AB_2245493; EOMES: PE, Invitrogen: 12-4875-80, clone: Dan11mag, RRID:AB_1603278; F4/80: ef450, Invitrogen: 48-4801-80, clone: BM8, RRID:AB_1548756; FcεR1a: APC, Invitrogen: 17-5898-82, clone: 1-Mar, RRID:AB_10718824; Glut1: AF647, AbCam: ab195020, clone: EPR3915, RRID:AB_2783877; I-A/I-E: AF700, BioLegend: 107622, clone: M5/144.15.2, RRID:AB_493727; I-A/I-E: BV711, BioLegend: 107643, clone: M5/144.15.2, RRID:AB_2565976; IFN-γ: PE-Cy7, Invitrogen: 25-7311-82, clone: XMG1.2, RRID:AB_469680; KLRG1: BV711, BioLegend: 138427, clone: 2F1/KLRG1, RRID:AB_2629721; KLRG1: FITC, eBioscience: 11-5893-82, clone: 2F1, RRID:AB_1311265; LFA1: PerCp-Cy5.5, BioLegend: 141008, clone: H155-78, RRID:AB_1311265; Ly6C: BV570, BioLegend: 128030, clone: HK1.4, RRID:AB_2562617; Ly6C: BV785, BioLegend: 128041, clone: HK1.4, RRID:AB_2565852; Ly6G: BUV563, BD Biosciences: 612921, clone: 1A8, RRID:AB_2870206; Ly6G: PE, BioLegend: 127608, clone: 1A8, RRID:AB_1186099; NK1.1: BV711, BioLegend: 108745, clone: PK136, RRID:AB_2563286; NK1.1: PerCp-Cy5.5, BioLegend: 108728, clone: PK136, RRID:AB_2132705; SiglecF: PE, BD Biosciences: 552126, clone: E50-2440, RRID:AB_394341; SiglecH: BV421, BD Biosciences: 566581, clone: 440C, RRID:AB_2739747; TCR Vα2: PE, BioLegend: 127808, clone: B20.1, RRID:AB_1134183; TCR Vα2: Superbright780, Invitrogen: 78-5812-82, clone: B20.1, RRID:AB_2735086; TCR Vβ5.1,5.2: APC, BioLegend: 139506, clone: MR9-4, RRID:AB_10933250; TCRβ: PerCp-Cy5.5, BioLegend: 109228, clone: H57-597, RRID:AB_1575173; Tetramer MHCI: PE, NIH Tetramer Core, peptide: SVLAFRRL; Tetramer MHCII: PE, NIH Tetramer Core, peptide: AVEIHRPVPGTAPPS; Tetramer MHCI: APC, NIH Tetramer Core, peptide: SVLAFRRL; Tetramer MHCII: APC, NIH Tetramer Core, peptide: AVEIHRPVPGTAPPS; XCR1: BV650, BioLegend: 148220, clone: ZET, RRID:AB_2566410; XCR1: BV785, BioLegend: 148225, clone: ZET, RRID:AB_2783119; TCF1: AF488, Cell Signaling: 6444S, clone: C63D9, RRID:AB_2797627; Tbet: AF647, BioLegend: 644804, clone: 4B10, RRID:AB_1595466; CD86: AF700, BioLegend: 105024, clone: GL-1, RRID:AB_493721; Ki67: AF700, BD Biosciences: 561277, clone: B56, RRID:AB_10611571; CD45.1: APC, eBioscience: 17-0453-82, clone: A20, RRID:AB_469398; CD64: APC, BioLegend: 139306, clone: X54-5/7.1, RRID:AB_11219391; CD8β: APC, BioLegend: 126614, clone: YTS156.7.7, RRID:AB_2562775; CD44: APC-Cy7, BioLegend: 103028, clone: IM7, RRID:AB_830785; F4/80: APC-ef780, eBioscience: 47-4801-82, clone: BM8, RRID:AB_2735036; Ly6C: APC-ef780, eBioscience: 47-5932-82, clone: HK1.4, RRID:AB_2573992; KLRG1: BUV395, BD Biosciences: 740279, clone: 2F1, RRID:AB_2740018; CD4: BUV496, BD Biosciences: 612952, clone: GK1.5, RRID:AB_2813886; CD8α: BUV563, BD Biosciences: 748535, clone: 53-6.7, RRID:AB_2872946; CD11b: BUV661, BD Biosciences: 565080, clone: M1/70, RRID:AB_2722548; CD122: BUV661, BD Biosciences: 741493, clone: TM-β1, RRID:AB_2870951; CD19: BUV661, BD Biosciences: 612971, clone: 1D3, RRID:AB_2870243; CD11a: BUV805, BD Biosciences: 741919, clone: 2D7, RRID:AB_2871232; CD8α: BUV805, BD Biosciences: 612898, clone: 53-6.7, RRID:AB_2870186; CD127: BV421, BioLegend: 135027, clone: A7R34, RRID:AB_2563103; CD19: BV421, BioLegend: 115538, clone: 6D5, RRID:AB_11203527; CXCR3: BV421, BioLegend: 126529, clone: CXCR3-173, RRID:AB_2563100; CD43: BV605, BD Biosciences: 747726, clone: S7, RRID:AB_2872201; CD62L: BV605, BioLegend: 104437, clone: MEL-14, RRID:AB_11125577; CD19: BV650, BioLegend: 115541, clone: 6D5, RRID:AB_11204087; CD27: BV650, BioLegend: 124233, clone: LG.3A10, RRID:AB_2687192; CX3CR1: BV785, BioLegend: 149029, clone: SA011F11, RRID:AB_2565938; CD122: FITC, BioLegend: 123207, clone TM-β1, RRID:AB_940611; CD80: FITC, eBioscience: 11-0801-85, clone: 16-10A1, RRID:AB_465134; β2-microglobulin: PE, BioLegend: 154504, clone: A16041A, RRID:AB_2721340; CD212: PE, BD Biosciences: 551974, clone: 114, RRID:AB_394310; CD127: PE-Cy7, BioLegend: 135014, clone: A7R34, RRID:AB_1937265; CD71: PerCp-Cy5.5, BioLegend: 113816, clone: R17217, RRID:AB_2565482; CX3CR1: PerCp-Cy5.5, BioLegend: 149009, clone: SA011F11, RRID:AB_2564493; CD8β: PerCp-ef710, Invitrogen: 46-0083-82, clone: ebioH35-17.2, RRID:AB_10669709.

Samples were run on a LSR Fortessa (BD), LSRII (BD), FACSymphony A3 (BD), or FACSymphony A5 (BD) and analyzed using FlowJo Software (TreeStar).

Immunofluorescence imaging

Whole omenta were harvested from mice and fixed in 1% PFA overnight at 4°C. After rinsing, tissue was blocked using 10% BSA, 0.5% normal rat serum (Invitrogen), and 1 μg/ml 2.4G2 (BD) in PBS for 1 hr at room temperature. For immunofluorescence staining, omenta were incubated in PBS containing primary antibodies at 4°C for 3 days and subsequently rinsed with PBS overnight. Antibodies included: B220: AF647, BD Biosciences: 557683, clone: RA3-6B2; B220: BV510, BioLegend: 103248, clone: RA3-6B2; CD11b: AF700, eBioscience: 56-0112-82, clone: M1/70; CD3: AF532, Invitrogen: 58-0032-82, clone: 17A2; CD4: AF532, Invitrogen: 58-0042-82, clone: RM4-5; CD8b: ef450, Invitrogen: 48-0082-82, clone: eBioH35-17.2; ER-TR7: AF647, Santa Cruz Biotechnology: sc-73355 AF647, clone: ER-TR7. Tissues were then mounted on slides using Prolong Diamond (Invitrogen: P36970) or Prolong Glass mounting media (Invitrogen: P36984) and imaged using a Leica TCS SP8 STED microscope and acquired using LAS X (Leica Microsystems). Images were processed and analyzed using LAS X and Imaris (Oxford Instruments).

Cluster analysis

Clustering was analyzed using K function statistics as described in full in Appendix C. Briefly, to determine clustering of Nur77^GFP+ OT-I T cells, the average number of Nur77^GFP+ cells within a specified distance, k, of each Nur77^GFP+ cell was compared with those averages calculated for 999 randomly permuted patterns of the full population of both Nur77^GFP+ and Nur77^GFP- cells. The subpopulation of Nur77^GFP+ cells was then deemed to be significantly clustered at scale k if the fraction of patterns with averages as large as the observed pattern was not greater than 0.05. To determine whether Nur77^GFP+ cells were clustered around a specific APC population, the target population is deemed to be significantly clustered around the reference population at scale k if the fraction of randomly permuted populations with an average number of target cells with k as large as the observed population is not larger than 0.05. Similarly, the target population is deemed to be significantly dispersed away from the reference population at scale k if the fraction of populations with averages as small as the observed population is not larger than 0.05.

STochastic Omentum REsponse (STORE) model

The STORE model is a dynamic stochastic mathematical model used to simulate the early OT-I T cell response to immunization in the omentum and is described in full in the Supplementary Methods. Briefly, the model is based off experimental data measuring the kinetics of the OT-I T cell response in the omentum through the first 8 days of immunization. Simulations of the model track the interactions of transferred OT-I T cells with APC populations and the subsequent T cell divisions that drive the expansion of the OT-I T cell population. The behavior of every cell within the system is tracked individually throughout the simulation and is determined by probability distributions estimated from the literature or taken as a separate parameter. The goal of the model is to recapitulate the experimental data for OT-I T cells with the tabulated parameters defining the differences between the responses of OT-I T cells in the WT and Batf3^−/− mice.

Statistical analysis

Statistical analysis was performed using GraphPad PRISM 9.0 software. The statistical analysis and significance values for each comparison are indicated in each figure caption.

Supplementary Material

movie S1

Movie S1. cDC1 localize with OT-I T cells in the T cell enriched region of the MS

data file S1

Data file S1: Raw data file (Excel spreadsheet)

reproducibility checklist

MDAR Reproducibility Checklist

main supplementary materials

Supplementary Methods

Fig. S1. Visualizing recruitment of Lyz2⁺ myeloid cells and infected cells to the MS

Fig. S2. Immunization-induced recruitment of monocytes to the peritoneum requires cDC1 and IL-12p40

Fig. S3. Characterization of acute and memory T cell responses during CPS immunization

Fig. S4. Overview of STORE model system, probability distributions, and algorithm

Fig. S5. Results of STORE model simulations

Fig. S6. Results of random permutation tests of clustering using K-function statistics

Table S1. Simulation results of OT-I T cell kinetics Experiment 1

Table S2. Simulation results of OT-I T cell kinetics Experiment 2

Table S3. STORE model optimization results

Table S4. Average of STORE model optimization results

Table S5. Training and testing results for STORE model

DATA AND MATERIALS AVAILABILITY

All data needed to evaluate the conclusions in the paper are present in the paper or the Supplementary Materials. Requests for materials should be directed to the corresponding author.