All mice were on C57BL/6 background and maintained under specific pathogen-free conditions at the Animal Facility of University Hospital Essen. Female C57BL/6 mice were purchased from Envigo Laboratories (Envigo CRS GmbH, Rossdorf, Germany).

iDTRxCD11ccre (B6.Gt(Rosa26)Sortm1(HBEGF)Awai/Tg(Itgax-cre)1-1Reiz) mice, kindly provided by Ari Waisman, Mainz, Germany and CD169DTR (Siglec1tm1(HBEGF)Mtka) mice have been described previously (15, 24). SpiC^flox/flox x vav1cre (B6.Cg-Tg(VAV1-cre)1Graf/Mdf/Spictm1Kmm) mice express the cre recombinase in hematopoietic stem cells and contain loxP-flanks at the spi-C transcription factor locus, which controls RPMΦ development and were kindly provided by Prof. Manfred Kopf, ETH Zürich, Switzerland. OTII (Cg-Tg(TcraTcrb)425Cbn/J) mice express a transgenic T cell receptor recognizing the MHC class II (I-Ab)-restricted ovalbumin peptide_323-339 (25) and were kindly provided by Dr. Tetyana Yevsa, Medical School, Hannover, Germany. To ensure unbiased group allocation, age- and sex-matched animals were randomly assigned to treatment groups. All animal experiments were carried out in accordance with the guidelines of the German Animal Protection Law and were approved by the state authority for nature, environment, and customer protection, North Rhine-Westphalia, Germany.

Spleen samples were excised and washed in PBS before being embedded in Tissue-Tek (Sakura Finetek, Tokyo, Japan). The samples were then rapidly frozen in liquid nitrogen and stored at -80°C. Spleen slices were sectioned using a Cryostat microtome (Leica) at a thickness of 8 µm and mounted on HistoBond^® microscope slides (Paul Marienfeld GmbH & Co. KG). The sections were subsequently frozen at -80°C for a minimum of 24 hours before staining. Spleen sections were retrieved from the freezer and allowed to dry at room temperature before use. Samples were then fixed in acetone for 10 minutes, followed by two washes with PBS for 3 minutes each. The areas of interest were outlined with a hydrophobic pen, and the sections were washed again in PBS for 3 minutes. To block nonspecific binding sites, samples were incubated in wash buffer (PBS + 2% FCS) for 45 minutes, then washed twice more in PBS for 3 minutes each. Excess buffer was gently removed, and sections were incubated with FITC-conjugated F4/80, APC-conjugated CD169 and PE-conjugated CD209b (all Miltenyi Biotec, Bergisch Gladbach, Germany) for 30 minutes. After incubation, the sections underwent two additional PBS washes (3 minutes each). Finally, mounting medium and a coverslip were applied.

For infection, the non-lethal Plasmodium strain P. yoelii 17XNL was used. Mice were intravenously injected with 1 x 10⁵ P. yoelii iRBCs in 100 µl PBS. Parasitemia was assessed on days 3, 5 and 7 p.i. with the Hematology analyzer XN-31 (Sysmex, Kobe, Japan) as described (26).

To obtain single-cell suspensions from the spleen, the organ was minced into a homogeneous paste before enzymatic digestion (DMEM supplemented with 1 mg/ml Collagenase D and 100 µg/ml DNase) at 37°C for 30 minutes. Following incubation, the cells were mechanically dissociated by passing them through a 100 μm strainer. Erythrocyte lysis buffer was added, and the samples were incubated for 2 minutes at room temperature before being diluted with MΦ medium (DMEM supplemented with 10% FCS and 5 mM EDTA). The suspension was then centrifuged at 300 × g for 10 minutes at 4°C. Finally, the obtained cells were resuspended in PBS supplemented with 2% FCS and 2 mM EDTA for further processing.

CD4⁺ T cells were obtained from OT-II mice by rinsing spleens with erythrocyte lysis buffer and washing with PBS supplemented with 2 mM EDTA and 2% FCS. CD4⁺ T cells were isolated from splenocytes by using the CD4⁺ T cell isolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s recommendation. DCs were obtained from splenocytes of WT C57BL/6 mice by using CD11c MicroBeads UltraPure (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the manufacturer’s recommendation. Macrophages were isolated as described, and splenic macrophage subpopulations were sorted using an Aria II Cell Sorter (BD Biosciences, Heidelberg, Germany). CD19, NK1.1, Ly-6G, Ly-6C (BD Biosciences, Heidelberg, Germany) CD3 (BioLegend, San Diego, USA) were used as fluorescein isothiocyanate (FITC), CD11b (BioLegend, San Diego, USA) was used as R-Phycoerythrin-Cyanine dye 7 (PE-Cy7), CD169 (BioLegend, San Diego, USA) was used as Allophycocyanine (APC), CD209b (Miltenyi Biotec, Bergisch Gladbach, Germany) was used as R-Phycoerythrin (PE), F4/80 (Invitrogen, Carlsbad, USA) was used as Pacific Blue (PB), and TIM-4 (Invitrogen, Carlsbad, USA) was used as Peridinin chlorophyll protein-Cyanine5.5 (PerCP-Cy5.5). RPMΦ were defined as lineage negative (neg. lin.) (CD3^-, CD19^-, LyG^-, NK1.1^-), CD11b^int and F4/80⁺. MZMΦ were defined as lineage negative (CD3^-, CD19^-, LyG^-, NK1.1^-), MARCO⁺ TIM4⁺ CD11b⁺, and MMMΦ as lineage negative (CD3^-, CD19^-, LyG^-, NK1.1^-), CD169⁺ TIM4⁺ CD11b⁺. Once all cell types had been prepared, CD4⁺ T cells were stained with eFluor 670 (eBioscience, ThermoFisher Scientific, Langenselbold, Germany). For this, the cells were washed twice with PBS, resuspended in 2 mL of PBS at room temperature to a concentration of 1 × 10⁷ cells and incubated with eFlour 670 in darkness at 37°C for 10 minutes. Staining was halted by adding 4 mL of cold FCS, followed by an additional 5 minute incubation on ice. Cells were then washed three times with IMDM (supplemented with 10% FCS, 1% Penicillin/Streptomycin and 0.1% Mercaptoethanol), and the supernatant was removed before final resuspension. Cells were plated at a ratio of 2 × 10⁴ APCs to 2 × 10⁴ T cells with 1 µg/ml OVA_323-339-peptide, and incubated at 37°C. T cell proliferation was assessed after 48 hours incubation. Cells were harvested and analyzed using flow cytometry.

CD11b, CD62L, CD69 (eBioscience, ThermoFisher Scientific, Langenselbold, Germany) CD19, CD3, CD4, CD86, IL-10, IL-6, Ly-6G, NK1.1, TIM-4 (BD Biosciences, Heidelberg, Germany), CD11c, CD169, CD192, CD3017, CX3CR1, F4/80, IFN-γ, IL-12/IL-23, Ly-6C (BioLegend, San Diego, USA), CD209b (Miltenyi Biotec, Bergisch Gladbach, Germany), granzyme B (Invitrogen, Carlsbad, USA), and MARCO (R&D Systems, Bio-Techne, Wiesbaden, Germany) were used as Allophycocyanine (APC), Brilliant Ultra Violet 737 (BUV737), Brilliant Ultra Violet 805 (BUV805), Horizon V450 (V450), Brilliant Violet 510 (BV510), Brilliant Violet 605 (BV605), Brilliant Violet 711 (BV711), Brilliant Violet 786 (BV786), Fluorescein isothiocynat (FITC), Peridinin-Chlorophyll-Protein (Per-CP), Peridinin chlorophyll protein-Cyanine5.5 (PerCP-Cy5.5), R-Phycoerythrin (PE), R-Phycoerythrin-Cyanine dye 5 (PE-Cy5), R-Phycoerythrin-Cyanine dye 7 (PE-Cy7). Dead cells were identified using the fixable viability dye eFluor 780 (eBioscience, ThermoFisher Scientific, Langenselbold, Germany).

Before surface staining, cells were resuspended in TruStain FcX™PLUS (Biolegend, San Diego, USA) and incubated for 15 minutes on ice. Cells were then centrifuged at 300 x g at 4°C and afterwards stained with the respective surface antibody mix. After surface staining, intracellular staining was performed if applicable. For granzyme B, the Foxp3 staining kit (eBioscience, ThermoFisher Scientific, Langenselbold, Germany) was used following the manufacturer’s instructions. Briefly, cells were fixed in fixation/permeabilization buffer and incubated at RT for 1 hour. Then, cells were permeabilized with permeabilization buffer and centrifuged at 300 x g for 5 min before resuspending cells with the respective intracellular antibodies and incubation for 30 min at 4°C. To assess IFNγ expression, cells were stimulated with 10 ng/mL phorbol 12-myristate 13-acetate and 100 μg/mL ionomycin (both from Sigma-Aldrich, Munich, Germany) for 4 hours in the presence of 5 μg/mL Brefeldin A (Sigma-Aldrich, Munich, Germany). Subsequently, cells were treated with 2% paraformaldehyde and 0.1% IGEPAL CA-630 (Sigma-Aldrich, Munich, Germany) before being stained with the respective antibodies for 30 minutes at 4°C. Flow cytometric analyses were performed using FACS Symphony A5, with data acquisition and analysis conducted using DIVA software (BD Biosciences, Heidelberg, Germany).

For the depletion of DCs, iDTRxCD11ccre transgenic (TG) mice were administered diphtheria toxin (DT) (Sigma-Aldrich, Munich, Germany) at a dose of 12 ng/kg body weight in PBS intraperitoneal (i.p.) every other day. As a control group, iDTRxCD11ccre wild-type (WT) littermates received the same DT amount.

CD169⁺ cell depletion was achieved by administration of 30 ng/kg bodyweight DT in PBS i.p. every 5 days. Control groups received PBS i.p. on the same days.

Macrophages were depleted by administration of 200 µl clodronate liposomes (CL) (Liposoma BV, Amsterdam, Netherlands) i.p. every 5 days. Control groups received 200 µl PBS filled liposomes (PBSL) (Liposoma BV, Amsterdam, Netherlands).

Statistical analyses were performed using Graph Pad Prism Software (Graph Pad Software, La Jolla, CA). Gaussian distribution was determined using D’Agostino & Pearson test, Shapiro-Wilk test or Kolmogorov-Smirnov test. If data passed normality test, unpaired Student’s t-test was used, if data were not distributed normally, Mann-Whitney U-test was performed. If two or more groups were to be analyzed, statistical analyses was calculated by one-way ANOVA or Kruskal-Wallis test. Statistical significance was considered as P-values of 0.05 or less (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Since CD4⁺ T cells play a crucial role in controlling acute infection and establishing long-term immunity, their activation was analyzed upon DC depletion by using DT-treated iDTRxCD11ccre (TG) mice and P. yoelii infection. DC depletion was performed one day prior to infection with P. yoelii. The mice were sacrificed seven days post-infection (p.i.) to analyze T cell responses and parasitemia as previously described (15). We first confirmed our previous results regarding CD4⁺ T cell activation. Wild-type (WT) mice infected with P. yoelii showed a significant rise in CD69⁺ CD4⁺ T cells, indicating enhanced activation ( Supplementary Figure 1 ). This increase was also seen in P.yoelii-infected DC-depleted transgenic iDTRxCD11ccre mice, although to a significantly lesser extent than in infected WT mice. Accordingly, infected TG mice displayed a marked decrease in CD69-expressing CD4⁺ T cells compared to infected WT mice, suggesting reduced T cell activation in the absence of DCs ( Supplementary Figure 1A ). In agreement with our previous data (15), IFNγ production by CD4⁺ T cells significantly increased upon infection in both WT and TG mice. Interestingly, despite the lack of DCs, DT-treated TG mice showed even higher levels of T cell activation, as evidenced by increased IFNγ-producing CD4⁺ T cells compared to non-infected DT-treated TG mice ( Supplementary Figure 1B ). Upon P. yoelii infection the expression of additional T cell activation markers, i.e. CD62L and granzyme B, were markedly reduced and increased, respectively ( Supplementary Figures 1C, D ). DT treatment resulted in a slight increase and decrease of CD62L and granzyme B, respectively, in TG mice upon infection, but this effect did not reach significance ( Supplementary Figures 1C, D) . In summary, we verify that DC-depleted mice show a decrease in CD4⁺ T cell responses during P. yoelii infection, but retain substantial activation, which may possibly be due to a compensatory effect of other APCs, such as macrophages.

Upon DC depletion, P. yoelii-infected animals show an increase in the splenic macrophage compartment ( Supplementary Figure 2D ), indicative of a compensatory mechanisms during hematopoiesis. As the splenic macrophage subsets MZMΦ and MMMΦ rely on the integrity of the MZ structure (27, 28), we first investigated possible effects of DC depletion on the splenic architecture in DT-treated iDTRxCD11ccre mice (TG) and control littermates (WT). Immunofluorescence staining revealed no apparent changes in splenic architecture ( Figure 1A ). The distinction between the WP and RP remained intact in DC-depleted mice. The distribution of RPMΦ (green), MZMΦ (yellow), and MMMΦ (red) was similar between the spleens of DT-treated WT and TG mice, with all macrophages maintaining their typical locations within the spleen. After P. yoelii infection, both DT-treated WT and TG mice showed a significant decline in RPMΦ frequency ( Figure 1B , gating strategy Supplementary Figure 3 ). In contrast, MZMΦ analysis revealed an increase in their abundance in non-infected mice lacking DCs, which became significant following P. yoelii infection in DT-treated TG mice ( Figure 1C ). This suggests a significant modulation of MZMΦ population in response to both, DC depletion and infection. Similarly, MMMΦ frequency exhibited a significant rise in DT-treated TG mice compared to WT mice following P. yoelii infection ( Figure 1D ). Next, we analyzed the cytokine profile of splenic macrophage subsets from DC-depleted mice after P. yoelii infection by flow cytometry and detected significant changes. While the frequency of IL-12/IL-23⁺ RPMΦ from DT-treated TG mice significantly decreased ( Figure 1E ), the percentage of IL-12/IL-23-producing MZMΦ significantly increased upon infection ( Figure 1F ). Additionally, cytokine analysis of MMMΦ showed a significant decrease in IL-6-producing MMMΦ ( Figure 1G ).

These findings suggest an enhanced response of MMMΦ and MZMΦ populations to P. yoelii infection in the absence of DCs, possibly indicating altered immune dynamics in this context.

To further investigate the overall impact of splenic macrophages on the outcome of P. yoelii infection, we depleted macrophages using the well-established CL model. CL were administered 3 days before P. yoelii infection and again 2 days p.i. As a control, PBS-filled liposomes (PBSL) were used, and mice were analyzed 7 days p.i. ( Figure 2A ). The administration of CL resulted in an efficient depletion of all splenic macrophage subsets as confirmed by immunohistochemistry ( Figure 2B ) as well as flow cytometric analysis ( Figures 2C-E ). To ensure that no other phagocytic cells were impacted by CL treatment, various immune cell populations were examined for potential changes in frequency by flow cytometry ( Supplementary Figure 4 ). Herein, the neutrophil population was significantly increased, suggesting a possible compensatory effect upon macrophage depletion and cDCs were reduced, but still present. In macrophage-deficient mice infected with P. yoelii, a significant reduction in CD69⁺ CD4⁺ T cells was observed ( Figure 2F ). While granzyme B is classically associated with cytotoxic CD8⁺ T cells, it is also expressed by specific CD4⁺ T cell subsets, called cytotoxic Th1 cells. Granzyme B⁺ CD4⁺ T cells appear to participate in anti-Plasmodium immunity, but the mechanism remains elusive, making them interesting for our analysis (29). Herein, we observed a slight trend to lower frequencies of IFNγ- and granzyme B-expressing CD4⁺ T cells from the CL-treated, infected group, which did not reach significance ( Figures 2G, H ). In contrast, CD62L levels were not altered in the infected groups, but showed a significant decrease upon P. yoelii infection indicating an overall increase in CD4⁺ T cell activation ( Figure 2I ). Together, these results suggest that macrophages contribute, at least in part, to CD4⁺ T cell activation during infection.

We next quantified parasite blood infection dynamics following macrophage depletion ( Figure 3 ). In the first infection experiment, CL administration was initiated three days before P. yoelii infection ( Figure 3A ). Here, a significant increase in parasitemia was observed 5 days p.i. in CL-treated mice compared to PBSL-treated controls. However, by day 7 p.i., parasitemia levels in CL-treated mice decreased and resulted in equalized levels between the groups, suggesting that other phagocytic cells may compensate for the loss of macrophage function. To determine whether this equalizing effect occurs when macrophages are depleted after an infection is established, C57BL/6 mice were first infected with P. yoelii and treated with PBSL or CL starting 2 days p.i. ( Figure 3B ). Interestingly, in this experimental setting a significant increase in parasitemia was detected 24 hours later (3 days p.i.) in CL-treated mice. This exacerbation became even more pronounced by day 5 p.i. and peaked at an excessively high parasitemia of approximately 40% by day 7 p.i. ( Figure 3B ). Together, the altered blood infection dynamics in CL-treated mice corroborate the notion that macrophages are the main cells essential for parasite clearance during P. yoelii infection.

Given that the CL treatment depletes all splenic macrophage subsets, the next step was to identify which macrophage subset might be responsible for the reduced CD4⁺ T cell activation and uncontrolled parasite burden observed during P. yoelii infection in macrophage-depleted mice. To determine whether different splenic macrophage populations can generally induce a T cell response, we first analyzed their T cell activating capacity in vitro. Therefore, OVA_323-339-peptide was loaded on sorted splenic macrophage subpopulations as well as on DCs, which were co-cultured with eFluor670-labeled OVA-specific CD4⁺ T cells. T cell proliferation was then assessed by the dilution of the proliferation dye ( Supplementary Figure 5 ). All splenic macrophage subpopulations were capable of inducing an OVA-specific T cell proliferation, with RPMΦ and MMMΦ demonstrating the strongest potential, followed by MZMΦ. These results suggest that splenic macrophages can elicit T cell responses at least in vitro.

Since a significant increase in MMMΦ was observed during P. yoelii infection in DC-depleted mice ( Figure 1D ), along with their potential to induce CD4⁺ T cell proliferation in vitro ( Supplementary Figure 5 ), this macrophage subset was of particular interest for further analysis. Therefore, CD169DTR mice were used to specifically deplete MMMΦ via diphtheria toxin (DT). DT treatment started at day 3 prior to P. yoelii infection and was repeated every 5 days. Parasitemia was monitored on days 3, 5, and 7 p.i., and mice were sacrificed for analysis 7 days p.i. ( Figure 4A ). While DT administration in CD169DTR mice did not alter splenic architecture ( Figure 4B ), or other immune cells ( Supplementary Figure 6 ), it resulted in a reduction in MMMΦ, and to a lesser extent, MZMΦ, as determined by immunofluorescence analysis ( Figure 4B ) and confirmed by flow cytometry ( Figures 4C-E ). To assess the effect of MMMΦ and MZMΦ depletion on CD4⁺ T cells, we analyzed the T cell response in P. yoelii-infected DT-treated CD169DTR mice. We did not detect differences in CD4⁺ T cell frequencies following DT treatment and P. yoelii infection ( Figure 5A ). Moreover, the expression of activation markers, i.e. CD69, CD62L, and granzyme B, did not differ between MMMΦ-depleted and control mice after infection ( Figures 5B-D ). Of note, MMMΦ and MZMΦ depletion led to a slight, but significant, reduction in IFNγ⁺ CD4⁺ T cells at 7 days p.i. ( Figure 5E ). However, depletion of MMMΦ and MZMΦ did not affect the course of parasitemia until day 7 p.i. ( Figure 5F ), excluding an important role in phagocytosis of parasite-infected erythrocytes.

In conclusion, MMMΦ and MZMΦ contribute to the induction of a T_H1 response, but their contribution to clear Plasmodium-infected RBCs is likely negligible.

Since the general depletion of splenic macrophages in P. yoelii-infected mice by CL treatment resulted in a significant increase in parasitemia ( Figures 3A, B ), but MMMΦ and MZMΦ depletion did not significantly alter parasite burden, we next analyzed the impact of RPMΦ on clearing P. yoelii-infected RBCs. We used SpiC^flox/flox x vav1cre mice, a transgenic mouse model in which RPMΦ development is selectively controlled. KO mice and WT littermates were infected with P. yoelii and analyzed 7 days p.i. ( Figure 6A ). Indeed, we could demonstrate, a significant reduction of the RPMΦ population in SpiC^flox/flox x vav1cre KO mice compared to SpiC^flox/flox x vav1cre WT littermates ( Figures 6B, D ), while pDCs, cDCs, B cells, and neutrophil frequencies are not altered due to the RPMΦ deficiency ( Figure 6C ). We note that MZMΦ and MMMΦ showed a trend towards increased frequencies, which did not reach significance in infected KO mice compared to WT mice ( Figures 7A, B ). The absence of RPMΦ did not impact the CD4⁺ T cell response during P. yoelii infection, as no notable differences in CD4⁺ T cell frequencies and activation status were found between WT and KO mice upon infection ( Figures 7C-E ). This implicates, that the RPMΦ subpopulation does not contribute to CD4⁺ T cell priming. However, RPMΦ deficiency resulted in significant elevated parasitemia at 5 days p.i. ( Figure 7F ). This impairment of parasite clearance is comparable to the findings we obtained after depleting all macrophages by CL. In summary, the data suggest that RPMΦ are not essential for the induction of CD4⁺ T cell responses during P. yoelii infection, but were identified as being important for clearing P. yoelii iRBCs.