The following 6- to 12-week-old mice were used in this study: C57BL/6 mice (Japan Clea, Tokyo, Japan), B6.Cg-Tg(Itgax-cre)1–1Reiz/J mice (CD11c-Cre mice; The Jackson Laboratory, Bar Harbor, ME, USA) (15), and B6.R26:lacZbpAfloxDTA mice (R-DTA mice) (16). R-DTA mice and CD11c-Cre mice were cross-mated to generate CD11c-Cre:R-DTA mice used as CD11c:DTA mice, and their wild-type (WT) littermates were used as CD11c^hi DC-sufficient control mice (14). B6.CD45.1⁺OT-II T-cell receptor (TCR) Tg mice holding ovalbumin (OVA)-specific CD4⁺ T cells (B6.CD45.1⁺OT-II mice) and B6.CD45.1⁺OT-I TCR Tg mice holding OVA-specific CD8⁺ T cells (B6.CD45.1⁺OT-I mice) were generated as described previously (14, 17, 18). All mice were housed under specific pathogen-free conditions in the animal facility, and all experiments were approved by the Animal Experiment Committee and Gene Recombination Experiment Committee at the University of Miyazaki.

OVA-transfected derivative of murine melanoma cell line (B16-OVA) (19) and murine colon adenocarcinoma cell line MC38 (20) were kindly provided by Dr. Shin-ichiro Fujii (RIKEN Center for Integrative Medical Sciences, Japan). B16-OVA and MC38 cells were cultured in RPMI 1640 (Wako, Osaka, Japan) containing an antibiotic-antimycotic (Wako) and 10% heat-inactivated fetal calf serum (FCS; Gibco, Grand Island, NY, USA). The cell line was cultured at 37°C in a humidified atmosphere of 5% CO₂ and air.

Single-cell suspensions were prepared from the spleen (Spl) and inguinal lymph nodes (LNs) used as TdLNs after tumor inoculation as described previously (14, 17, 18, 21, 22). In brief, tissue samples were digested with collagenase type III (Worthington Biochemical, Lakewood, NJ, USA) at 37°C for 20 min and were ground between glass slides. Cell suspensions of Spl were treated with red blood cell lysis buffer (Sigma-Aldrich, St. Louis, MO, USA). Single-cell suspensions of leukocytes were prepared by filtering through a 100-µm cell strainer (BD Biosciences, San Jose, CA, USA). CD4⁺ T cells, CD45.1⁺Vα2⁺OT-II CD4⁺ T cells, or CD45.1⁺Vα2⁺OT-I CD8⁺ T cells were purified from splenocytes of C57BL/6 mice, B6.CD45.1⁺OT-II mice, or B6.CD45.1⁺OT-I mice with mouse CD4 T lymphocyte Enrichment Set-DM (BD Biosciences) or mouse CD8 T lymphocyte Enrichment Set-DM (BD Biosciences). Tumor tissues were ground between glass slides, and CD45⁺ leukocytes were prepared by AutoMACS with mouse CD45 Microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany).

Cells were stained with fluorescein-conjugated monoclonal antibodies (mAbs) after Fc blocking with mAb to CD16/CD32 (clone 2.4G2, BD Biosciences) listed in Supplementary Table 1 in the Supplementary Material . Intracellular staining was performed using the Fixation and Permeabilization kit (eBioscience, San Diego, CA, USA). Analysis of fluorescence staining was performed using FACSVerse flow cytometer (BD Biosciences), CytoFLEX flow cytometer (Beckman Coulter, Brea, CA, USA), and FlowJo software (version 10.9.0; Tree Star, Ashland, OR, USA) according to the gating strategy to identify the leukocytes in LNs, Spl, and tumor tissues as described in Supplementary Figures 1 – 3 in the Supplementary Material .

Quantitative reverse transcription polymerase chain reaction (RT-qPCR) was performed as described previously (14, 21, 22). Total RNA was extracted from cells by RNeasy plus micro kit (Qiagen, Valencia, CA, USA), and the first-strand cDNA was synthesized with oligo(dT)20 primer using the PrimeScript RT Master Mix (Takara, Shiga, Japan) from total RNA. Transcriptional expression levels were analyzed by using SYBR® Premix Ex Taq II on Thermal Cycler Dice (Takara) with specific primer pairs listed in Supplementary Table 2 in the Supplementary Material after normalization for Gapdh expression.

Mice were treated daily with or without gilteritinib, an inhibitor of Fms-like tyrosine kinase 3 (Flt3), (SelleckChem, Houston, TX, USA; 200 μg/mouse) (23) by oral gavage using 20-gauge, 1.5-inch feeding needles (Natume, Tokyo, Japan) for 20 days. Spl was obtained from the mice on day 21 after the start of administration.

Sera were assayed for IL-1β (R&D Systems, Minneapolis, MN, USA), IL-2 (eBioscience), and IL-15 (R&D Systems) using enzyme-linked immunosorbent assay (ELISA) kits according to the manufacturers’ instructions.

WT mice or CD11c:DTA mice were inoculated subcutaneously (s.c.) with B16-OVA (1 × 10⁵) or MC38 (5 × 10⁵) in the right back. Tumor size was measured every day after the inoculation for 8 to 18 days using a digital caliper (CP-15CP; Mitsutoyo, Kanagawa, Japan), and tumor volumes were estimated using the following ellipsoidal formula: length × width × height ×0.52. Spl, TdLNs, and tumor tissues were obtained from mice on day 18 after tumor inoculation. In some experiments, Spl, TdLNs, and tumor tissues were obtained from tumor-bearing WT mice or tumor-bearing CD11c:DTA mice with delayed tumor growth on days 19–21 after tumor inoculation when the values of the tumor volumes of tumor-bearing WT mice as controls were in the range of their mean ± standard deviation (s.d.). For the depletion of Gr-1⁺ cells (24, 25), B16-OVA-bearing mice received anti-Gr-1 mAb (clone RB6–8C5, Bio X Cell, Lebanon, NH, USA; 200 µg/mouse) or control Ab (rat IgG Wako; 200 µg/mouse) by i.p. injections every 3 days for 15 days starting from the day of tumor inoculation until the end of the experiment.

Gr-1⁺ cells used as myeloid-derived suppressor cells (MDSCs) were collected by biotin-conjugated anti-Gr-1 mAb (clone RB6–8C5, BioLegend) and Streptavidin Particles Plus-DM (BD Biosciences) from Spl of B16-OVA-bearing mice on day 18–21 after tumor inoculation. CD4⁺ T cells obtained from C57BL/6 mice were labeled with eFluor™ 670 (Thermo Fisher Scientific, Waltham, MA, USA; 2.5 µM) for 10 min at 37°C. After washing twice with cold medium, CD4⁺ T cells (2 × 10⁴) were stimulated with or without Dynabeads Mouse T-Activator CD3/CD28 (Thermo Fisher Scientific) in the presence or absence of Gr-1⁺ cells (2 × 10⁴) in a 96-well plate. After 4 days, the gated CD4⁺ T cells were analyzed for eFluor™ 670 dilution to detect the dividing cells by flow cytometry.

Analysis of Ag-specific priming of CD4⁺ T cells or CD8⁺ T cells in vivo was performed as described previously (14, 17, 18). In brief, eFluor™ 670-labeled CD45.1⁺OT-II CD4⁺ T cells or CD45.1⁺OT-I CD8⁺ T cells as described above (each 5 × 10⁶/mouse) were intravenously (i.v.) injected into mice. Subsequently, mice were intraperitoneally (i.p.) injected with OVA protein (A5503, Sigma-Aldrich) at 24 hrs after adoptive transfer. The gated CD45.1⁺OT-II CD4⁺ T cells or CD45.1⁺OT-I CD8⁺ T cells in LNs were analyzed for eFluor™ 670 dilution to detect the dividing cells by flow cytometry at 2 days after immunization. In another experiment, B16-OVA-bearing mice on day 18 after tumor inoculation or untreated naïve mice were i.v. or intratumorally (i.t.) injected with eFluor™ 670-labeled CD45.1⁺OT-II CD4⁺ T cells or CD45.1⁺OT-I CD8⁺ T cells (each 5 × 10⁶/mouse) to detect their accumulation in Spl, TdLNs, or tumor tissues. Analysis of the gated CD45.1⁺OT-II CD4⁺ T cells or CD45.1⁺OT-I CD8⁺ T cells in Spl, TdLNs, and tumor tissues was performed at 2 days after adoptive transfer as described above.

Data are expressed as the mean ± s.d. from three to eight individual samples in a single experiment of at least three independent experiments. Statistical analysis was performed using Prism 9 (GraphPad Software, Boston, MA, USA) using the two-sided unpaired Student’s t-test. A P-value for significance was indicated with <0.05 or 0.01.

To clarify the contribution of CD11c^hi DCs to the protection from the development of tumors, we generated a mouse model with the constitutive deficiency of CD11c^hi DCs by cross-mating CD11c-Cre mice (15) and R-DTA mice harboring the diphtheria toxin (DT) α chain (DTA) under control of a loxP-flanked stop cassette in the ubiquitously expressed ROSA26 locus (16) for producing CD11c-Cre:R-DTA double-Tg mice (referred to as CD11c:DTA mice) (14). CD11c:DTA mice showed almost complete depletion of CD11c^hi DCs in LNs and Spl under homeostatic conditions when compared with WT mice ( Supplementary Figures 4 , 5 in the Supplementary Material ). Similar to the published report (14), the partial reduction of the frequency of pDCs was observed in LNs and Spl of CD11c:DTA mice as compared with WT mice ( Supplementary Figures 4 , 5 in the Supplementary Material ).

We addressed the influence of the constitutive ablation of CD11c^hi DCs on the constituency of leukocytes in lymphoid tissues under steady-state conditions. CD11c:DTA mice showed reduced frequencies of CD4⁺ T cells and CD8⁺ T cells, whereas they exhibited the enhanced proportions of Gr-1⁺CD11b⁺F4/80⁻ polymorphonuclear leukocytes (PMNs) and Gr-1⁺CD11b⁺F4/80⁺ monocytes/macrophages in LNs and Spl when compared with WT mice ( Supplementary Figures 4 , 5 in the Supplementary Material ). In addition, CD11c:DTA mice showed the enhanced frequency of B220⁺ B cells in LNs ( Supplementary Figure 4 in the Supplementary Material ).

Taken together, these results indicate that the constitutive deficiency of CD11c^hi DCs leads to abnormal composition of leukocytes in lymphoid tissues.

Given that Flt3 ligand (Flt3L) is crucial for the development of various types of leukocytes (26–28), we have previously reported the linking of the enhanced generation of myeloid-lineage cells with massive serum production of Flt3L in CD11c:DTA mice (14). To clarify the role of Flt3L on the enhanced development of Gr-1⁺CD11b⁺F4/80⁻ PMNs and Gr-1⁺CD11b⁺F4/80⁺ monocytes/macrophages under the constitutive absence of CD11c^hi DCs, we examined the influence of gilteritinib, an inhibitor of Flt3 (23), on their generation in CD11c:DTA mice. Treatment of CD11c:DTA mice with gilteritinib inhibited the enhanced generation of Gr-1⁺CD11b⁺F4/80⁻ PMNs and Gr-1⁺CD11b⁺F4/80⁺ monocytes/macrophages in LNs and Spl under homeostatic conditions ( Supplementary Figure 6 in the Supplementary Material ).

Collectively, these results indicate that Flt3L is required for the enhanced generation of PMNs and monocytes/macrophages under the constitutive loss of CD11c^hi DCs.

We also examined the expression level of CD11c on leukocytes in LNs and Spl. We showed an expression of CD11c on I-A/I-E^hiCD11c⁺XCR1⁺SIRPα⁻ migratory cDC1, I-A/I-E^hiCD11c⁺XCR1⁻SIRPα⁺ migratory cDC2, I-A/I-E⁺CD11c⁺XCR1⁺SIRPα⁻ resident cDC1, I-A/I-E⁺CD11c⁺XCR1⁻SIRPα⁺ resident cDC2, I-A/I-E^hiCD11c⁺CD207⁺XCR1⁻SIRPα⁺ Langerhans cells (LCs) (29), and I-A/I-E^hiCD11c⁺XCR1⁻SIRPα⁺CD64⁺ monocyte-derived DCs (Mo-DCs)/DC3 (30) in LNs as well as I-A/I-E⁺CD11c⁺XCR1⁺SIRPα⁻ cDC1 and I-A/I-E⁺CD11c⁺XCR1⁻SIRPα⁺ cDC2 in Spl ( Supplementary Figure 7 in the Supplementary Material ). In addition, CD11c⁺B220⁺Siglec-H⁺NK1.1⁻ pDCs showed a higher expression of CD11c than CD11c⁺B220⁺Siglec-H⁻NK1.1⁺ natural killer (NK) cells (31, 32) in LNs and Spl, while their expression levels were lower than those of cDC subsets ( Supplementary Figure 7 in the Supplementary Material ). We did not observe an expression of CD11c on CD3⁻CD19⁻NK1.1⁺NKp46⁺CD49a⁻CD49b⁺ NK cells (30), CD3⁻CD19⁻NK1.1⁺NKp46⁺CD49a⁺CD49b⁻ group 1 innate lymphoid cells (ILC1s) (33), and CD3⁻CD19⁻CD127⁺RORγt⁺ ILC3s (34) in LNs and Spl in WT mice ( Supplementary Figure 7 in the Supplementary Material ).

We further examined the constituencies of the subsets of leukocytes in LNs and Spl in CD11c:DTA mice. CD11c:DTA mice showed the marked reduction of I-A/I-E^hiCD11c⁺XCR1⁺SIRPα⁻ migratory cDC1, I-A/I-E^hiCD11c⁺XCR1⁻SIRPα⁺ migratory cDC2, I-A/I-E⁺CD11c⁺XCR1⁺SIRPα⁻ resident cDC1, I-A/I-E⁺CD11c⁺XCR1⁻SIRPα⁺ resident cDC2, I-A/I-E^hiCD11c⁺CD207⁺XCR1⁻SIRPα⁺ LCs, and I-A/I-E^hiCD11c⁺XCR1⁻SIRPα⁺CD64⁺ Mo-DCs/DC3 in LNs as well as I-A/I-E⁺CD11c⁺XCR1⁺SIRPα⁻ cDC1 and I-A/I-E⁺CD11c⁺XCR1⁻SIRPα⁺ cDC2 in Spl ( Supplementary Figures 8A–L , 9A–C in the Supplementary Material ). In addition, CD11c:DTA mice exhibited a milder reduction of CD11c⁺B220⁺Siglec-H⁺NK1.1⁻ pDCs than cDC subsets in LNs, whereas their similar significant reductions were observed in Spl ( Supplementary Figures 8M–O , 9D–F in the Supplementary Material ). In contrast, CD11c:DTA mice had normal constituencies of CD11c⁺B220⁺Siglec-H⁻NK1.1⁺ NK cells (35), CD3⁻CD19⁻NK1.1⁺NKp46⁺CD49a⁻CD49b⁺ NK cells, and CD3⁻CD19⁻NK1.1⁺NKp46⁺CD49a⁺CD49b⁻ ILC1s in LNs and Spl ( Supplementary Figures 8M–R , 9D–I in the Supplementary Material ). CD11c:DTA mice also showed a reduction of CD3⁻CD19⁻CD127⁺RORγt⁺ ILC3s in LNs, but not Spl, as compared with WT mice ( Supplementary Figures 8S–U , 9J–L in the Supplementary Material ). We also observed a higher production of IL-2 or lower productions of IL-1β and IL-15 in CD11c:DTA mice than in WT mice ( Supplementary Figure 9M in the Supplementary Material ).

Taken together, these results indicate that CD11c^hi DC subsets are mainly ablated in lymphoid tissues in CD11c:DTA mice.

We utilized the experimental model of a murine melanoma tumor cell line with poor immunogenicity that was engineered to express OVA (B16-OVA) (19). When compared with WT mice, CD11c:DTA mice displayed a significant progression of tumor growth for the duration of the experiment ( Figures 1A–D ). We also observed that CD11c:DTA mice exhibited an enhanced development of murine colon adenocarcinoma cell line MC38 known as immunogenic tumor (20) compared with WT mice ( Figures 1E–H ). Therefore, these results indicate that the congenital deficiency of CD11c^hi DCs causes the enhanced progression of tumors.

We also compared the cellularity of tumor-infiltrating leukocytes between WT mice and CD11c:DTA mice under tumor-bearing conditions. When compared with tumor-bearing WT mice, tumor-bearing CD11c:DTA mice exhibited the marked infiltrations of Gr-1⁺CD11b⁺F4/80⁻ PMNs, Gr-1⁺CD11b⁺F4/80⁺ monocytes/macrophages, and B220⁺ B cells, whereas they displayed the significant reduction in the accumulations of CD11c^hi cDCs, CD11c⁺Siglec-H⁺ pDCs, and CD3⁺CD8⁺ T cells ( Figures 1I–K ). Similar differences in the cellular compositions in TdLN and Spl between WT mice and CD11c:DTA mice were observed under tumor-bearing conditions when compared with those under steady-state conditions ( Supplementary Figures 10 , 11 in the Supplementary Material ).

Furthermore, tumor-bearing CD11c:DTA mice exhibited a significant reduction of I-A/I-E^hiCD11c⁺XCR1⁺SIRPα⁻ migratory cDC1 and I-A/I-E^hiCD11c⁺XCR1⁻SIRPα⁺ migratory cDC2, I-A/I-E^hiCD11c⁺CD207⁺XCR1⁻SIRPα⁺ LCs, and I-A/I-E^hiCD11c⁺XCR1⁻SIRPα⁺CD64⁺ Mo-DCs/DC3 in tumor tissues, I-A/I-E^hiCD11c⁺XCR1⁺SIRPα⁻ migratory cDC1 and I-A/I-E^hiCD11c⁺XCR1⁻SIRPα⁺ migratory cDC2, I-A/I-E⁺CD11c⁺XCR1⁺SIRPα⁻ resident cDC1, I-A/I-E⁺CD11c⁺XCR1⁻SIRPα⁺ resident cDC2, I-A/I-E^hiCD11c⁺CD207⁺XCR1⁻SIRPα⁺ LCs, and I-A/I-E^hiCD11c⁺XCR1⁻SIRPα⁺CD64⁺ Mo-DCs/DC3 in TdLNs as well as I-A/I-E⁺CD11c⁺XCR1⁺SIRPα⁻ cDC1 and I-A/I-E⁺CD11c⁺XCR1⁻SIRPα⁺ cDC2 in Spl compared with those in tumor-bearing WT mice ( Supplementary Figures 12 , 13A–O in the Supplementary Material ). Tumor-bearing CD11c:DTA mice also showed a reduction of CD3⁻CD19⁻CD127⁺RORγt⁺ ILC3s in TdLNs as compared with tumor-bearing WT mice ( Supplementary Figures 13P–R in the Supplementary Material ). However, tumor-bearing CD11c:DTA mice exhibited higher accumulation of CD11b⁺F4/80⁺CD80⁺ M1-like macrophages (36–38) in TdLNs as well as CD11b⁺F4/80⁺CD80⁺ M1-like macrophages and CD11b⁺F4/80⁺CD206⁺ M2-like macrophages (36–38) in Spl than tumor-bearing WT mice ( Supplementary Figure 14 in the Supplementary Material ). We also showed that Gr-1⁺ cells composed of Gr-1⁺CD11b⁺F4/80⁻ PMNs and Gr-1⁺CD11b⁺F4/80⁺ monocytes/macrophages in Spl in tumor-bearing WT mice and tumor-bearing CD11c:DTA mice had a similar inhibitory effect on the division of anti-CD3/CD28-stimulated eFluor™ 670-labeled CD4⁺ T cells ( Supplementary Figures 15A, B in the Supplementary Material ), indicating that they act as MDSCs (39).

Taken together, these results indicate that the constitutive loss of CD11c^hi DCs enhances the accumulation of MDSCs in tumor tissues under tumor-bearing conditions.

It has been shown that MDSCs exhibited an immunosuppressive potential to support tumor progression (40–42). To clarify the contribution of MDSCs to the development of tumors, we addressed the effect of the depletion of MDSCs by anti-Gr-1 mAb (24, 25) on the progression of tumor growth in CD11c:DTA mice. The elimination of MDSCs by anti-Gr-1 mAb exhibited a more dramatic reduction of tumor growth in CD11c:DTA mice than tumor-bearing WT mice ( Supplementary Figures 15C–F in the Supplementary Material ), probably due to a more prominent accumulation of MDSCs in tumor tissues ( Figures 1I–K ).

Taken together, these results indicate that MDSCs are involved in the enhanced development of tumors under the constitutive loss of CD11c^hi DCs.

We compared the activation status of T cells in lymphoid tissues between WT mice and CD11c:DTA mice. CD11c:DTA mice had reduced fractions of CD4⁺ T cells and CD8⁺ T cells expressing CD44 as a marker for T-cell activation (43) in LNs and Spl as compared with WT mice in the homeostatic conditions ( Supplementary Figures 16 , 17 in the Supplementary Material ). Furthermore, CD4⁺ T cells and CD8⁺ T cells in LNs and Spl did not express any immune checkpoint molecules including programmed death (PD)-1, T-cell immunoglobulin and mucin-domain containing-3 (TIM-3), and lymphocyte activation gene (LAG)-3 in both WT mice and CD11c:DTA mice under steady-state conditions ( Supplementary Figures 16 , 17 in the Supplementary Material ). However, tumor-infiltrating CD4⁺ tumor-infiltrating lymphocytes (TILs) and CD8⁺ TILs obtained from tumor-bearing WT mice not only had the marked proportions of CD4⁺CD44⁺CD62L⁻ T cell subset and CD8⁺CD44⁺CD62L⁻ T cell subset, which function as effector memory T (T_EM) cells (44), but also expressed PD-1, TIM-3, and LAG-3 ( Figures 2A–D ; Supplementary Figures 18A, B in the Supplementary Material ). In contrast, CD11c:DTA mice had a higher proportion of CD4⁺CD44⁺CD62L⁻ T cells in tumor tissues than WT mice ( Figures 2A, B ; Supplementary Figure 18A in the Supplementary Material ). Interestingly, CD4⁺ TILs, but not CD8⁺ TILs, obtained from CD11c:DTA mice showed lower expressions of PD-1 and TIM-3 than those obtained from WT mice ( Figures 2A–D ; Supplementary Figures 18A, B in the Supplementary Material ).

CD4⁺ T cells and CD8⁺ T cells displayed an enhanced expression of CD44 in TdLNs and Spl in WT mice during tumor progression ( Figures 2E–L ; Supplementary Figure 18C–F in the Supplementary Material ). When compared with tumor-bearing WT mice, CD4⁺ T cells, but not CD8⁺ T cells, obtained from tumor-bearing CD11c:DTA mice exhibited enhanced expressions of PD-1 and TIM-3 in TdLNs ( Figures 2E–H ; Supplementary Figures 18C, D in the Supplementary Material ). Similar to the homeostatic conditions, CD11c:DTA mice showed reduced or enhanced fractions of CD4⁺CD44⁺ T cells and CD8⁺CD44⁺ T cells or CD8⁺CD62L⁻ T cells in Spl as compared with WT mice under tumor-bearing conditions ( Figures 2I–L ; Supplementary Figures 18E, F in the Supplementary Material ). Furthermore, tumor-bearing CD11c:DTA mice showed an enhanced expression of TIM-3 on CD4⁺ T cells and PD-1 and TIM-3 on CD8⁺ T cells in Spl as compared with those obtained from tumor-bearing WT mice ( Figures 2I–L ; Supplementary Figures 18E, F in the Supplementary Material ). Collectively, these results indicate that the constitutive loss of CD11c^hi DCs alters the activation status of T cells under tumor-bearing conditions.

To clarify the role of CD11c^hi DCs in the Ag-specific T-cell priming, WT mice and CD11c:DTA mice were adoptively transferred with eFluor™ 670-labeled OT-II CD4⁺ T cells or OT-I CD8⁺ T cells expressing the OVA-specific TCR (14, 17, 18) before systemic injection of OVA protein to analyze their Ag-specific division in LNs. In contrast to the prominent division of OT-II CD4⁺ T cells or OT-I CD8⁺ T cells in LNs in WT mice after systemic administration of OVA protein, their responses were significantly reduced in CD11c:DTA mice ( Supplementary Figure 19 in the Supplementary Material ).

To determine the influence of the constitutive elimination of CD11c^hi DCs in the initiation of Ag-specific T-cell responses in tumor-bearing mice, eFluor™ 670-labeled OT-II CD4⁺ T cells or OT-I CD8⁺ T cells were adaptively i.t. or i.v. transferred into B16-OVA-bearing mice to determine their Ag-specific division in tumor tissues or TdLNs and Spl, respectively. While tumor-bearing WT mice exhibited Ag-specific division of OT-I CD8⁺ T cells in tumor tissues and TdLNs as well as OT-II CD4⁺ T cells in TdLNs, they did not show the apparent responses of OT-II CD4⁺ T cells in tumor tissues and Spl as well as OT-I CD8⁺ T cells in Spl ( Figures 3A–L ). However, tumor-bearing CD11c:DTA mice displayed the profound reduction of Ag-specific divisions of OT-I CD8⁺ T cells in tumor tissues and TdLNs as well as OT-II CD4⁺ T cells in TdLNs as compared with tumor-bearing WT mice ( Figures 3C, D, G–J ).

To examine the effect of the constitutive elimination of CD11c^hi DCs on the generation of Ag-specific CTLs, we evaluated the occurrence of Ag-specific CD8⁺ T cells based on binding with the MHC I-OVA pentamer in tumor-bearing mice. MHC I-OVA pentamer⁺CD44^highCD8⁺ T cells were detected in tumor tissues, TdLNs, and Spl in tumor-bearing WT mice ( Figures 3M–U ). However, tumor-bearing CD11c:DTA mice showed a marked reduction of MHC I-OVA pentamer⁺CD44^highCD8⁺ T cells in tumor tissues, TdLNs, and Spl as compared with tumor-bearing WT mice ( Figures 3M–U ). Similarly, tumor-bearing CD11c:DTA mice exhibited a lower frequency of MHC I-gp100 tetramer⁺CD44^highCD8⁺ T cells in tumor tissues than tumor-bearing WT mice ( Supplementary Figure 20 in the Supplementary Material ). Taken together, these results indicate that the congenital deficiency of CD11c^hi DCs diminishes Ag-specific responses of T cells under tumor-bearing conditions.

To address the effect of the constitutive absence of CD11c^hi DCs on the composition of CD4⁺Foxp3⁺ T_reg cells, we compared their proportion between WT mice and CD11c:DTA mice. CD11c:DTA mice displayed a reduced frequency of CD4⁺Foxp3⁺ T_reg cells, whereas they exhibited the enhanced frequency of CD4⁺Foxp3⁺ retinoic acid-related orphan receptor γt (RORγt)⁺ T_reg cells (45, 46), in LNs and Spl when compared with WT mice in homeostatic conditions ( Supplementary Figure 21 in the Supplementary Material ). However, tumor-bearing WT mice and CD11c:DTA mice showed comparable accumulations of CD4⁺Foxp3⁺ T_reg cells and CD4⁺Foxp3⁺RORγt⁺ T_reg cells in tumor tissues ( Figures 4A–C ). Similar to the homeostatic conditions, tumor-bearing CD11c:DTA mice showed a reduced proportion of CD4⁺Foxp3⁺ T_reg cells and an enhanced proportion of CD4⁺Foxp3⁺RORγt⁺ T_reg cells in TdLNs and Spl as compared with tumor-bearing WT mice ( Figures 4D–I ). Taken together, these results suggest that the constitutive elimination of CD11c^hi DCs alters the composition of CD4⁺Foxp3⁺ T_reg cells in lymphoid tissues under homeostatic and tumor-bearing conditions.

Tumor cells and tumor-infiltrating immune cells secrete several soluble factors that are thought to generate the immunosuppressive tumor microenvironment (TME), which attenuates the anti-tumor immune responses, leading to progressive tumor growth (47–49). We therefore compared the transcriptional expressions of the immunosuppressive cytokines and metabolite-generating enzymes in tumor tissues between tumor-bearing WT mice and CD11c:DTA mice ( Figure 5 ). Tumor tissues obtained from tumor-bearing CD11c:DTA mice exhibited higher transcriptional expressions of IL-10, transforming growth factor (TGF)-β, indoleamine 2,3-dioxygenase (IDO), vascular endothelial growth factor (VEGF), arginase, inducible nitric oxide synthase (iNOS), cyclooxygenase (COX)-2, and membrane-associated prostaglandin E2 synthase (mPGES) than tumor-bearing WT mice. Collectively, these results indicate that the congenital deficiency of CD11c^hi DCs promotes the production of immunosuppressive mediators in tumor tissues.