Murine precursors to type 1 conventional dendritic cells induce tumor cytotoxicity and exhibit activated PD-1/PD-L1 pathway

Abstract

The immediate precursor to murine type 1 conventional DCs (cDC1s) has recently been established and named "pre-cDC1s". Mature CD8α+ cDC1s are recognized for suppressing graft-versus-host disease (GvHD) while promoting graft-versus-leukemia (GvL), however pre-cDC1s have not previously been investigated in the context of alloreactivity or anti-tumor responses. Characterization of pre-cDC1s, compared to CD8α+ cDC1s, found that a lower percentage of pre-cDC1s express PD-L1, yet express greater PD-L1 by MFI and a greater percent PIR-B, a GvHD-suppressing molecule. Functional assays were performed ex vivo following in vivo depletion of CD8α+ DCs to examine whether pre-cDC1s play a redundant role in alloreactivity. Proliferation assays revealed less allogeneic T-cell proliferation in the absence of CD8α+ cDC1s, with slightly greater CD8+ T-cell proliferation. Further, in the absence of CD8α+ cDC1s, stimulated CD8+ T-cells exhibited significantly less PD-1 expression compared to CD4+ T-cells, and alloreactive T-cell death was significantly lower, driven by reduced CD4+ T-cell death. Tumor-killing assays revealed that T-cells primed with CD8α-depleted DCs ex vivo induce greater killing of A20 B-cell leukemia cells, particularly when antigen (Ag) is limited. Bulk RNA sequencing revealed distinct transcriptional programs of these DCs, with pre-cDC1s exhibiting activated PD-1/PD-L1 signaling compared to CD8α+ cDC1s. These results indicate distinct T-cell-priming capabilities of murine pre-cDC1s compared to CD8α+ cDC1s ex vivo, with potentially clinically relevant implications in suppressing GvHD while promoting GvL responses, highlighting the need for greater investigation of murine pre-cDC1s.

Fig 1. Pre-cDC1s differ from CD8α+ cDC1s in their steady-state expression of co-stimulatory and co-inhibitory molecules.

Splenic pan DCs were isolated from naïve BALB/c mice and analyzed by flow cytometry to characterize pre-cDC1s and CD8α+ cDC1s. (A) Representative flow plot demonstrating gating strategy. Gated on CD11c+B220- (conventional DCs); pre-cDC1s are gated as CD24^highCD8α- and CD8α+ cDC1s are gated as CD8α+. (B) Mean absolute cell number of respective DC subsets shown with SEM. (C-F) Mean percent and MFI with representative histograms shown with SEM for expression of (C) PD-L1, (D) PIR-B, (E) CD70, and (F) ICOSL on CD8α+ cDC1s (black) and pre-cDC1s (cyan). Data is representative of two experiments. (n = 5 per group). Paired t tests were used to determine significance between DC subsets. **P<0.01, ***P<0.001.

Fig 2. Using a CD8α-depleting antibody to determine functional redundancy.

BALB/c mice were dosed with 200 μg of 2.43 mAb antibody intraperitoneally on day -3 and day -1. On day 0, splenic DCs were isolated and analyzed by flow cytometry. Data is representative of three independent experiments. (n = 4 per group) (A) Representative flow plots depicting relative cDC1 proportions in naïve DCs (left) and 2.43 mAb DCs (right). Gated on CD11c+B220- (conventional DCs). (B) Mean percent of CD8α+ cDC1s (black) and pre-cDC1s (cyan) in untreated and 2.43 mAb-treated DCs shown with SEM. Two-way ANOVA and Dunnett’s multiple comparisons used to determine significance. **P<0.01, ****P<0.0001. (C) Mean absolute cell number of pre-cDC1s (CD24^highCD8α-) in untreated (black) and 2.43 mAb-treated (magenta) groups shown with SEM. Unpaired t test used to determine significance. *P<0.05. (D) Schematic depicting experimental design for proposed allogeneic mixed leukocyte reaction (MLR) using 2.43 mAb treatment.

Fig 3. 2.43 mAb-treated DCs induce less proliferation, altered PD-1 expression, and depressed induction of cell death of alloreactive T-cells.

BALB/c mice were dosed with 200 μg of 2.43 mAb antibody intraperitoneally on day -3 and day -1. On day 0, splenic DCs were isolated and plated in an allogeneic MLR. Allogeneic T-cell proliferation was assessed on day 3 and day 4 of co-culture. Data is representative of two independent experiments. (n = 4 per group) (A-B) ModFit Software analysis of CellTrace dye dilutions. (A) Representative histograms depicting proliferation of H2K^b+ T-cells following stimulation with naïve DCs (top) or 2.43 mAb DCs (bottom) on days 3 and 4. Proliferation Index (PI) for representative histograms is boxed. (B) Mean PI for naïve or 2.43 mAb DCs shown with SEM. (C-G) Flow analysis of the proliferative fraction of T-cells was determined by first gating on H2K^b+CellTrace^low T-cells. (C) Mean percent of proliferated T-cells that are CD4+ (left) and CD8+ (right) on day 3 shown with SEM. (D) Mean percent of proliferated CD4+ (left) and CD8+ (right) T-cells expressing PD-1 on day 3 shown with SEM. (E) Mean percent of proliferated T-cells that are CD4+ (left) and CD8+ (right) on day 4 shown with SEM. (F) Mean percent of proliferated CD4+ (left) and CD8+ (right) T-cells expressing PD-1 on day 4 shown with SEM. (G) Mean percent of proliferated T-cells (left) and CD4+ T-cells (right) that are Propidium Iodide+ shown with SEM. Unpaired t tests were used to determine significance between groups. *P<0.05.

Fig 4. 2.43 mAb-treated DCs retain a lower frequency of CD8α+ cDC1s.

BALB/c mice were dosed with 200 μg of 2.43 mAb antibody intraperitoneally on day -3 and day -1. On day 0, splenic DCs were isolated and plated in an allogeneic MLR. DC composition was determined by flow cytometry on day 2 of co-culture. Data is representative of two independent experiments. (n = 4 per group) (A) Representative flow plots depicting percent of H2K^d+CD11c+ DCs in naïve DCs (left) and 2.43 mAb DCs (right) on day 2 of co-culture. (B) Mean percent of H2K^d+CD11c+ DCs for naïve (black) or 2.43 mAb DCs (magenta) shown with SEM. Unpaired t test was used to determine significance between groups. **P<0.01. (C) Representative flow plots depicting percent CD8α+ cDC1s and pre-cDC1s in naïve (left) and 2.43 mAb DCs (right). (D) Mean percent CD8α+ cDC1s (black bars) and pre-cDC1s (teal bars) in naïve and 2.43 mAb DCs shown with SEM. 2way ANOVA and Šidák’s multiple comparison used to determine statistical significance. **P<0.01.

Fig 5. Depletion of CD8α+ cDC1s results in improved tumor-killing.

BALB/c mice were dosed with 200 μg of 2.43 mAb antibody intraperitoneally on day -3 and day -1. On day 0, splenic DCs were isolated and plated with allogeneic T-cells. On day 3 of co-culture A20-Luc tumor cells were added to wells at the indicated concentrations. On day 4, luciferin was added to wells and plates were imaged. BLI was measured for each independent well and averaged among technical replicates. BLI values were used to quantify specific killing as a percent of tumor alone control. Data is pooled from two independent experiments. (n = 6 per group) (A) Schematic depicting the experimental timeline. (B) Representative bioluminescence image depicting plate layout and showing BLI intensity on a scale from low (green) to high (red) indicative tumor burden. (C-D) Mean percent specific killing is shown with SEM for A20-Luc concentrations of (C) 1x10⁵ and (D) 2x10⁵ A20-Luc tumor cells. Unpaired t test used to determine significance between groups. *P<0.05, **P<0.01.

Fig 6. Pre-cDC1 and CD8α+ cDC1 exhibit distinct transcriptional programs.

Splenic DCs were isolated from naïve BALB/c mice and pre-cDC1s and CD8α+ cDC1s were cell sorted for RNA sequencing. (A) Enhanced volcano plot identifying genes that are up or down-regulated in pre-cDC1s compared to CD8α+ cDC1s. (B-C) Pathway results using Qiagen’s IPA software. P = 0.05 corresponds to a 1.3 -log p-value, and anything above this value is considered statistically significant. (B) Pathways predicted to be significantly inhibited (z-score < -2) in pre-cDC1s compared to CD8α+ cDC1s. (C) Pathways predicted to be significantly activated (z-score < 2) in pre-cDC1s compared to CD8α+ cDC1s. (D) Heat map showing relative gene expression levels for the PD-1/PD-L1 cancer immunotherapy pathway. Full documentation, methodology, and raw data for this data set is available online at: https://doi.org/10.25422/azu.data.14241902.