Epigenetic repression of IL-1β by C1QBP restricts Dectin-1-mediated host defense against murine invasive candidiasis

Abstract

Invasive candidiasis, predominantly caused by Candida albicans, threatens immunocompromised individuals with high mortality. Dendritic cell-derived interleukin-1β, orchestrated via Dectin-1-mediated β-glucan recognition, is central to antifungal defense. While the Dectin-1/Syk/CARD9 axis governing Il1b transcription is defined, the epigenetic regulatory mechanisms remain elusive. Here, we identify Complement C1q binding protein (C1QBP) as a critical negative regulator of Dectin-1-mediated interleukin-1β production. C1QBP sequesters phosphorylated protein kinase C δ in the cytoplasm, preventing its nuclear transport. Nuclear phosphorylated-PKCδ activates the RNF20/RNF40 ubiquitin ligase complex, which recruits the methyltransferase DOT1L to catalyze histone H3 lysine 79 trimethylation at the Il1b locus to promote its transcription. DC-specific C1QBP depletion enhances interleukin-1β secretion, amplifies protective T helper 17 cell responses, and improves host resistance to systemic C. albicans infection in mice. Collectively, our findings establish C1QBP as a key checkpoint in a PKCδ-dependent epigenetic pathway that constrains antifungal immunity, highlighting its therapeutic potential for invasive fungal diseases.

Engagement of Dectin-1 on dendritic cells drives IL-1β production and is a critical component of the immune response to invasive candidiasis. Here the authors show epigenetic repression of IL-1β by a complement binding protein can restrain the dectin1 driven immune response in a murine model of invasive candidiasis.

Introduction

Invasive fungal infections pose a severe threat to global public health, with Candida albicans accounting for over 40% of invasive cases and driving mortality rates exceeding 40% in high-risk cohorts¹. The innate immune system serves as the first line of defense against pathogenic fungi, wherein interleukin-1β (IL-1β)-mediated inflammatory responses in dendritic cells (DCs) play a central role^2,3. IL-1β orchestrates the recruitment and activation of both innate and adaptive immune cells, ultimately promoting the polarization of T helper 17 (Th17) cells that secrete IL-17 and IL-22 to recruit neutrophils, which are critical for clearing C. albicans infections⁴. The transcriptional regulation of Il1b, encoding the IL-1β precursor pro-IL-1β, is a critical checkpoint in this immune response, governed by a sophisticated network of signaling pathways.

The transcription of Il1b is tightly regulated by pattern recognition receptors (PRRs). Among these, Dectin-1 (encoded by Clec7a), a C-type lectin receptor (CLR) predominantly expressed on murine DCs and macrophages, plays a non-redundant role in recognizing β-glucans on the C. albicans cell wall and initiating downstream signaling cascades⁵. This non-redundant function is underscored by the heightened susceptibility to C. albicans infections observed in Dectin-1-deficient mice and humans carrying loss-of-function Clec7a variants^5,6. Upon binding β-glucans, Dectin-1 triggers a well-characterized signaling cascade involving spleen tyrosine kinase (Syk), protein kinase C δ (PKCδ), and the CARD9/BCL10/MALT1 (CBM) signalosome, which subsequently activates nuclear factor kappa B (NF-κB), a master transcriptional regulator of Il1b^2,7–9. Beyond this canonical NF-κB pathway, Dectin-1 also engages alternative signaling modules. For instance, in pathological contexts such as fungal keratitis, Dectin-1 activation induces c-Jun N-terminal kinase (JNK) phosphorylation, which contributes significantly to IL-1β production¹⁰. Furthermore, Dectin-1 can induce a non-canonical inflammasome wherein the CBM complex recruit caspase-8 and the adaptor ASC (apoptosis-associated speck-like protein containing a CARD) to facilitate pro-IL-1β processing independently of pathogen internalization and the canonical caspase-1 inflammasome⁹. Moreover, negative feedback mechanisms are crucial for maintaining immune homeostasis. For example, DC-derived thymic stromal lymphopoietin (TSLP) has been identified as an autocrine regulator that dampens Syk phosphorylation and limits Il1b transcription, thereby preventing excessive inflammation¹¹. While the Dectin-1/Syk/CARD9 signaling pathway governing pro-IL-1β synthesis and inflammasome-mediated maturation has been well characterized^2,3,9, molecular mechanisms that epigenetically regulate Il1b gene transcription remain largely elusive.

Recent breakthroughs in immune epigenetics have underscored histone modifications as pivotal modulators of inflammatory gene transcription^12,13. For instance, lipopolysaccharide (LPS)-induced enrichment of trimethylated histone H3 lysine 36 (H3K36me3) at the Il1b locus fine-tunes its transcriptional output during Toll-like receptor 4 (TLR4)-mediated innate immune activation¹⁴. Moreover, methylation of histone H3 lysine 79 (H3K79me1/2/3), a modification catalyzed by the conserved methyltransferase DOT1L¹⁵, modulates the transcription of Il6 and Ifnb1 to govern proinflammatory cytokine production¹⁶. Despite these insights, cell-intrinsic mechanisms that epigenetically constrain Dectin-1-driven Il1b transcription remain incompletely defined, representing a critical knowledge gap in the molecular regulation of antifungal immunity.

Complement C1q binding protein (C1QBP, also known as gC1qR or p32) is a highly conserved, multifunctional protein localized primarily in the mitochondria, but also found in the cytoplasm and on the cell surface^17–19. Initially identified as a binding partner for the complement component C1q, C1QBP has been implicated in a broad spectrum of cellular processes, ranging from mitochondrial metabolism and RNA splicing to the modulation of multiple intracellular signaling cascades²⁰. Notably, a recent study shows that mitochondrial C1QBP drives effector CD8⁺ T cell differentiation via metabolic-epigenetic reprogramming during antiviral and antitumor immune responses²¹. Despite these advances, the functional contribution of C1QBP to Dectin-1-elicited Il1b transcription remains entirely uncharacterized.

In this study, we identified a previously unrecognized role of C1QBP as a critical negative regulator of Dectin-1-mediated IL-1β production in DCs during the murine disseminated candidiasis. We demonstrated that C1QBP constrains antifungal immunity by sequestering phosphorylated PKCδ in the cytoplasm, thereby preventing its nuclear translocation and subsequent epigenetic activation of the Il1b gene transcription. Our findings reveal an intricate spatial and epigenetic regulatory circuit that fine-tunes inflammatory responses to fungal pathogens, positioning C1QBP as a potential therapeutic target for enhancing host defense against invasive fungal infections.

Results

C1QBP constrains Il1b expression in cDC2s during systemic C. albicans infection

As the kidneys represent the primary target organ and major site of immune activation during systemic candidiasis^22,23, we used renal immune cells to delineate the cellular origin and regulatory mechanisms governing Il1b expression during C. albicans infection. We performed single-cell RNA sequencing (scRNA-seq) on 33,827 renal immune cells isolated from mice challenged with a sublethal dose (2 × 10⁵ CFU) of C. albicans strain SC5314 (Fig. 1A and Supplementary Fig. 1A). Analysis of Il1b expression revealed its broad induction across multiple immune lineages at days 2 and 6 post-infections (Fig. 1A and Supplementary Fig. 1B). Notably, dendritic cells (DCs) exhibited a most regulated upregulation of Il1b at day 2, implying them as a principal source of IL-1β (Fig. 1A and Supplementary Fig. 1B). Subclassification of DCs into five subsets, including conventional type 1 dendritic cells (cDC1s), cDC2s, monocyte-derived DCs (moDCs), plasmacytoid DCs (pDCs), and regulatory DCs (RegDCs), revealed that cDC2s displayed the highest Il1b expression at days 2 and 6 (Fig. 1B and Supplementary Fig. 1C). These data collectively establish cDC2s as a predominant source of renal IL-1β during C. albicans infection, underscoring the need to define their cell-intrinsic regulatory program.

Fig. 1. Systematic identification of C1qbp as a transcriptional constraint on Il1b in renal cDC2s during C. albicans infection.

A UMAP visualization of renal CD45⁺ immune cells from naïve (day 0) or C. albicans-infected wild-type (WT) mice (days 2 and 6 post-infection; n = 3 per group), based on scRNA-seq analysis. Dot plot showed Il1b expression across immune cell clusters. B UMAP of renal dendritic cell (DC) subsets and dot plot of Il1b expression across DC subsets at indicated times. C Volcano plot of differentially expressed genes (DEGs) between Il1bʰⁱ and Il1bˡᵒ cDC2s (sorted day 2 post-infection). Key functional modules including C1qbp were highlighted. D Gene Ontology (GO) analysis of DEGs from (C), showing top enriched terms including related genes. E Violin plots showing C1qbp expression across DC subsets at days 0, 2, and 6 post-infection. F GO analysis of DEGs between C1qbpˡᵒ and C1qbp^hi cDC2s (day 6 post-infection). G Gene Set Variation Analysis (GSVA) scores for selected GO pathways in C1qbp^hi vs C1qbpˡᵒ cDC2s across infection time points. H KEGG pathway enrichment analysis of upregulated genes in C1qbpˡᵒ cDC2s (day 2 vs. naïve). I Flow cytometry analysis of the frequency and cell number of IL-1β⁺ and C1QBP⁺ cells within renal DC subsets at indicated times post-infection (n = 5 per group). Data were mean ± SEM; statistical significance was determined by unpaired two-tailed t-test. ns, not significant. Source data are provided as a Source Data file.

To investigate the transcriptional basis for elevated Il1b expression in cDC2s, we performed RNA sequencing comparing Il1b‑high (Il1bʰⁱ) and Il1b‑low (Il1bˡᵒ) cDC2 subsets (Fig. 1C). Unbiased analysis identified a suite of differentially expressed genes (DEGs), which we categorized into functional modules encompassing inflammatory response, epigenetic regulation, intracellular signaling, and mitochondrial regulators (Fig. 1C). Among these, the mitochondrial regulator C1QBP emerged as a strong candidate negative regulator of Il1b, showing significant downregulation in Il1bʰⁱ cDC2s (Fig. 1C). Gene Ontology (GO) analysis further highlighted C1QBP as a multifunctional regulator with transcription factor-binding activity and multi-compartment localization (mitochondrion, cytoplasm, nucleus) (Fig. 1D). Based on its downregulation and functional associations with homeostatic and transcriptional pathways, we nominated C1QBP as a key candidate repressor of Il1b expression.

To determine whether low C1qbp expression is a distinctive feature of high IL-1β-producing cDC2s, interrogation of scRNA-seq data confirmed that cDC2s not only maintained significantly lower C1qbp levels than all other DC subsets at days 2 and 6 post-infection, but also exhibited a marked reduction compared to their un-infected state, a decline that coincided with peak Il1b expression (Fig. 1E). To characterize the functional landscape associated with low C1qbp expression, we performed GO analysis on DEGs from C1qbpˡᵒ versus C1qbp^hi cDC2s. This analysis further uncovered significant enrichment of C1QBP in transcriptional regulatory processes, including “negative regulation of mRNA splicing” and “regulation of transcriptional activity” (Fig. 1F and Supplementary Fig. 1D). These data suggest that diminished C1qbp expression may alleviate a transcriptional brake, thereby facilitating elevated Il1b expression in C1qbpˡᵒ cDC2s.

To dynamically assess the impact of C1QBP in cDC2s during infection, we performed Gene Set Variation Analysis (GSVA) across time points post-infection (Fig. 1G). The activity of the “negative regulation of IL-1β production” pathway was significantly suppressed in C1qbpˡᵒ cDC2s, consistent with their high IL-1β output. Concurrently, we observed enhanced activity in both “activation of protein kinase C activity” and “negative regulation of protein localization to the nucleus” (Fig. 1G). These coordinated changes support a model in which C1QBP acts as an upstream regulator that constrains Il1b expression, potentially by limiting the activation and nuclear translocation of a PKC family kinase.

We next performed KEGG enrichment analysis on genes upregulated in renal C1qbpˡᵒ cDC2s at day 2 post-infection, a stage marked by strong Il1b induction during C. albicans infection (Fig. 1H). These genes were enriched in key antifungal pathways, including oxidative phosphorylation, antigen processing and presentation, and CLR signaling pathway, the latter encompassing the fungal PRR encoding gene Clec7a and its downstream target Il1b (Fig. 1H). This enrichment profile functionally links low C1qbp expression to broad activation of antifungal effector programs in cDC2s.

We further used flow cytometry assay to quantify IL-1β and C1QBP production across DC subsets. This analysis revealed that cDC2s uniquely exhibited a significant increase in the frequency and number of IL-1β-producing cells coupled with a concurrent decrease in C1QBP-producing cells at day 2 post-infection, a signature not observed in other DC subsets (Fig. 1I and Supplementary Fig. 2A,B). These data unequivocally confirm cDC2s as the dominant IL-1β-producing subset and establish low C1qbp expression as a defining feature of pro-inflammatory cDC2s during C. albicans infection.

DC-specific C1QBP deficiency potentiates IL-1β-dependent Th17 polarization and enhances host antifungal host defense

To functionally explore the role of C1QBP in restricting IL-1β production and antifungal immunity, we generated mice with DC-specific deletion of C1qbp (C1qbp^fl/flCD11c^Cre/+) (Supplementary Fig. 3A). Following challenge with a sublethal inoculum of C. albicans, C1qbp^fl/flCD11c^Cre/+ mice exhibited significantly improved survival compared to C1qbp^fl/fl control mice (Fig. 2A), demonstrating that DC-expressed C1QBP constrains host resistance to candidiasis. Concordant with the survival benefit, fungal burdens in the kidneys and brains of C1qbp^fl/flCD11c^Cre/+ mice were substantially lower at days 2, 4, and 6 post-infections, with a transient reduction also observed in the spleen (Fig. 2B). In contrast, fungal loads in the liver and lungs were comparable between genotypes, underscoring the organ-specific nature of the C1QBP deficiency phenotype. Renal histopathological examination further corroborated enhanced fungal control: periodic acid–Schiff (PAS) staining revealed significantly diminished invasive fungal hyphae in the kidneys of C1qbp^fl/flCD11c^Cre/+ mice, whereas hematoxylin and eosin (H&E) staining indicated comparable levels of inflammatory cell infiltration (Fig. 2C). This disparity suggests that the improved outcome is mediated by a specific antifungal mechanism rather than a general enhancement of leukocyte recruitment to the site of infection (Supplementary Fig. 3B and 3C).

Fig. 2. DC-intrinsic C1QBP deficiency enhances IL-1β-dependent Th17 immunity and host defense against C. albicans.

A Survival of C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice after intravenous C. albicans infection (n = 6 per group). B Fungal burdens (CFU) in indicated organs at days 2, 4, and 6 post-infection (n = 5 per group). C Representative images of H&E- and PAS-stained kidney sections at day 4 post-infection (n = 5 mice per group). Scale bars: 50 µm. Arrows indicated fungal hyphae. D Flow cytometry analysis of IL-1β⁺ cells within renal DC subsets in C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice at the indicated time points post-infection (n = 5 per group). E, F ELISA quantification of cytokines in kidney homogenates: (e) innate cytokines at day 2 post-infection, including IL-1β, TNF-α, IL-6, and IL-10, (f) T helper cell-related cytokines at day 6 post-infection, including Th17-associated cytokines (IL-17A, IL-17F, IL-22) and IFN-γ (n = 5 per group). G Fungal burdens in the kidneys and brain of wild-type (WT) mice that received intravenous injection of PBS (PBS → WT) or adoptive transfer of 2 × 10⁶ cDC2s from WT mice (WT cDC2→WT) or C1qbp^fl/flCD11c^Cre/+ mice (C1QBP KO cDC2→WT), followed by intravenous infection with 2 × 10⁵ CFUs of C. albicans for 2 days (n = 5 mice per group). ELISA quantification of cytokines in kidney homogenates from the mice described in (G): (H) innate cytokines at day 2 post-infection, including IL-1β and IL-6, (I) T helper cell-related cytokines at day 6 post-infection, including Th17-associated cytokines (IL-17A, IL-17F, IL-22) and IFN-γ (n = 5 per group). J Survival of mice treated with IL-1β-neutralizing antibody (αIL-1β) or control IgG after infection (n = 10 per group). K Fungal burdens in kidney and brain at day 2 post-infection in αIL-1β or control IgG-treated mice (n = 5 per group). L, M Representative H&E and PAS staining of kidneys from antibody-treated mice at day 4 post-infection (n = 5 mice per group). Scale bars, 20 µm. Arrows indicated fungal hyphae. Data were mean ± SEM; significance determined by unpaired two-tailed Student’s t test (B–I, K-L) or Log-rank (Mantel-Cox) test (A, J). ns, not significant. Source data are provided as a Source Data file.

To determine whether the enhanced host defense resulting from C1QBP deficiency was attributable to specific regulation of IL-1β, we analyzed cytokine production in renal DCs. Flow cytometric analysis confirmed that C1QBP deficiency specifically increased both the frequency and absolute number of IL-1β-producing cDC2s, with minimal effects on other DC subsets (Fig. 2D and Supplementary Fig. 4A and 4B). Consistent with this finding, renal cytokine profiling showed a selective elevation in IL-1β protein levels, while concentrations of TNF-α, IL-6, and IL-10 remained unchanged (Fig. 2E). Given the established pivotal role of IL-1β in promoting Th17 cell responses^24,25, we next assessed the adaptive immune response. Measurement of renal cytokines revealed that C1qbp^fl/flCD11c^Cre/+ mice produced significantly higher levels of the Th17-associated cytokines IL-17A, IL-17F, and IL-22 compared to controls, whereas the Th1 cytokine IFN-γ was unaffected (Fig. 2F). This increase in cytokine levels was supported by a corresponding elevation in the frequency of renal Th17 cells, as confirmed by flow cytometry (Supplementary Fig. 5A and 5B). These data demonstrate that C1QBP deletion selectively enhances IL-1β-dependent Th17 polarization during C. albicans infection.

To directly establish the role of cDC2s in mediating the negative regulation of IL-1β production by C1QBP, we performed adoptive transfer of equal numbers of bone marrow-derived cDC2s from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice into WT recipient mice. Adoptive transfer of C1QBP-deficient cDC2s into WT recipients significantly reduced fungal burdens in the kidneys and brain at day 2 post-C. albicans infection (Fig. 2G and Supplementary Fig. 6A) and markedly increased IL-1β protein levels in the kidneys, while IL-6 expression remained unchanged (Fig. 2H). At day 6 post-infection, the frequency and the number of Th17 cells and the levels of Th17-related cytokines in kidney tissues were also significantly upregulated, whereas IFN-γ expression remained unaltered (Fig. 2I and Supplementary Fig. 6B and 6C). In contrast, myeloid cell-specific C1QBP deletion (C1qbp^fl/flLyz2^Cre/+) did not affect host survival, tissue fungal burdens, or the expression of IL-1β, IL-6, IL-17A, and IFN-γ following C. albicans challenge (Supplementary Fig. 6D to 6H). These data confirm that C1QBP-deficient cDC2s, rather than other myeloid cell populations, mediate the enhanced IL-1β production, Th17 adaptive immune responses, and protective antifungal immunity observed in C1qbp^fl/flCD11c^Cre/+ mice.

To establish the essential role of IL-1β in the enhanced antifungal defense conferred by C1QBP deletion, we administered an IL-1β-neutralizing antibody (αIL-1β) during systemic C. albicans infection. IL-1β neutralization abolished the survival advantage of C1qbp^fl/flCD11c^Cre/+ mice (Fig. 2J) and reversed the reduction in fungal burden observed in the kidney and brain (Fig. 2K). Histopathological analysis confirmed that αIL-1β treatment restored fungal invasion in C1qbp^fl/flCD11c^Cre/+ mice without significantly altering inflammatory infiltration (Fig. 2L and M). Furthermore, IL-1β blockade attenuated both the frequency of renal Th17 cells and their cytokine production in C1qbp^fl/flCD11c^Cre/+ mice at day 6 post-infection (Supplementary Fig. 6I to 6K). Collectively, these data establish a linear mechanistic pathway whereby DC-specific deletion of C1QBP unleashes IL-1β production specifically in cDC2s, which in turn drives a robust protective Th17 response and enhances host antifungal defense against systemic candidiasis.

C1QBP constrains Dectin-1-induced IL-1β production by limiting DOT1L-mediated H3K79me3 deposition across the Il1b gene body

To delineate the molecular mechanism by which C1QBP deletion enhances IL-1β production in DCs, we investigated its potential impact on the Dectin-1 signaling pathway, the primary upstream driver of IL-1β production in response to fungal β-glucans. Bone marrow‑derived DCs (BMDCs) from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice were stimulated with curdlan, a particulate β-glucan and specific Dectin-1 agonist (Supplementary Fig. 7A). Comparative transcriptomic analysis identified a core set of 558 genes that were induced by Dectin-1 activation in control BMDCs and further significantly amplified in C1QBP-deficient BMDCs under identical stimulation conditions (Fig. 3A), indicating that C1QBP deficiency selectively potentiates a discrete Dectin‑1-driven transcriptional program. Protein-protein interaction (PPI) network analysis of the top differentially expressed genes (DEGs) positioned Il1b as a highly connected hub within a dense inflammatory module that also included Tnf, Il1a, and Syk (Fig. 3B), indicating that C1QBP deficiency coordinately upregulates a coherent inflammatory network centered on Il1b. Functional enrichment analysis showed that the amplified transcriptome was highly focused on inflammatory and innate immune signaling pathways, including “inflammatory response”, “IL-1 signaling pathway”, and “positive regulation of cytokine production” (Fig. 3C). To further delineate the functional associations of these amplified genes, we performed gene-term connectivity analysis (Fig. 3D). This analysis confirmed that the enhanced expression of Il1b is functionally coupled to the coordinated upregulation of a broader inflammatory transcriptional program. Notably, Il1b-associated term “positive regulation of transcription by RNA polymerase II” included the histone methyltransferase Dot1l (Fig. 3D), which catalyzes H3K79me3^26,27, thereby linking the amplified transcriptional output to epigenetic regulation. Gene set enrichment analysis (GSEA) further confirmed significant enrichment of the gene set “epigenetic regulation of gene expression” in the Dectin‑1-induced transcriptome amplified by C1QBP deficiency (Fig. 3E), underscoring epigenetic remodeling as a key mechanism by which C1QBP constrains inflammatory output.

Fig. 3. C1QBP constrains Dectin-1-induced IL-1β production by limiting DOT1L-mediated H3K79me3 deposition at the Il1b locus.

A Venn diagram showing overlap of DEGs upregulated by curdlan stimulation in C1qbp^fl/fl (Ctrl) and C1qbp^fl/flCD11c^Cre/+ (KO) BMDCs (558 genes). B Protein-protein interaction (PPI) network of top amplified DEGs from (A) in KO BMDCs. C Functional enrichment analysis of the amplified DEGs from (A), showing top enriched inflammatory and immune pathways. D Gene-term connectivity plot illustrating the association between core amplified DEGs (outer ring, colored by log₂ fold-change) and their significantly enriched GO terms (inner ring). E Gene set enrichment analysis (GSEA) showing significant enrichment in amplified DEGs from (A). F Il1b mRNA expression in BMDCs stimulated with curdlan for 4 h. G ELISA of IL-1β levels in the supernatants of C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ BMDCs, pretreated with epigenetic inhibitors (Bobcat339 10 μM, Pargyline 6 μM, MTA 20 μM), or vehicle control (DMSO), followed by stimulation with curdlan (50 μg/mL) for 18 h and ATP (5 mM) for 1 h. H Immunoblot analysis of curdlan-induced pro-IL-1β expression in C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ BMDCs, with or without MTA pretreatment at specified time points. I ELISA of IL-1β levels in the supernatants of C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ BMDCs, pretreated with the Dot1l inhibitor pinometostat (2 μM), the Ezh2 inhibitor GSK343 (1 μM), or vehicle control (DMSO), followed by stimulation with curdlan (50 μg/mL) for 18 hours and ATP (5 mM) for 1 h. J Immunoblot analysis of H3K79 methylation in C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ BMDCs, with or without pretreatment with pinometostat at designated time points. (K) ChIP-PCR analysis assessing H3K79me3 binding to the Il1b gene in C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ BMDCs, stimulated with curdlan (50 μg/mL) for 3 hours, using control IgG or anti-H3K79me3 antibodies. L ChIP-PCR analysis of H3K79me3 binding to the Il1b gene body in C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ BMDCs, with or without pretreatment with pinometostat, following stimulation with curdlan (50 μg/mL) for 3 hours, utilizing control IgG or anti-H3K79me3 antibodies. Data in (F–J) were presented as mean ± SEM, n = 5(F, G, I, K, L), n = 3(H, J) biologically independent samples. Statistical significance was determined by unpaired two-tailed Student’s t-test. ns, not significant. Source data are provided as a Source Data file.

We next validated that C1QBP deletion significantly enhanced Il1b mRNA expression and its protein secretion in BMDCs specifically stimulated with curdlan or zymosan (Fig. 3F and Supplementary Fig. 7B). A targeted pharmacological screen showed that the hypersecretion of IL‑1β in C1QBP-deficient BMDCs was abolished by the global methyltransferase inhibitor MTA, but not by inhibitors of DNA methylation (Bobcat339) or specific histone demethylases (Pargyline) (Fig. 3G). Consistently, MTA also suppressed elevated pro-IL‑1β protein levels (Fig. 3H) without impairing cell viability (Supplementary Fig. 7C), indicating that the exaggerated IL‑1β response depends on a SAM‑utilizing methyltransferase. Among candidates tested, inhibition of DOT1L with pinometostat, but not EZH2 with GSK343, selectively abolished IL‑1β hypersecretion in C1QBP-deficient BMDCs (Fig. 3I). Corroborating these findings, C1QBP deficiency specifically increased H3K79me3 levels following curdlan stimulation, with no significant changes in H3K79me1/2 or other histone modifications examined (Fig. 3J and Supplementary Fig. 7D). Pinometostat treatment concurrently reduced both elevated H3K79me3 and pro-IL‑1β levels (Fig. 3J). Chromatin immunoprecipitation (ChIP)-qPCR analysis further demonstrated that C1QBP deletion significantly enhanced H3K79me3 occupancy specifically across the Il1b gene body, but not at its promoter, and this increase was abolished by pinometostat treatment (Fig. 3K and L, Supplementary Fig. 7E).

In parallel, bone marrow-derived cDC2s from C1qbp^fl/flCD11c^Cre/+ mice exhibited significantly enhanced Il1b mRNA expression and IL-1β protein secretion following specific stimulation with curdlan (Supplementary Fig. 7F and 7G). Similarly, C1QBP deficiency specifically elevated H3K79me3 and pro-IL-1β levels following curdlan stimulation, with no significant changes in H3K79me1/2 or other histone modifications examined (Supplementary Fig. 7H). Pinometostat treatment concurrently reduced IL-1β hypersecretion in C1QBP-deficient cDC2s (Supplementary Fig. 7I). Collectively, these data identify C1QBP as a critical negative regulator of Dectin-1-induced Il1b transcription, pro-IL-1β production, and IL-1β secretion. Mechanistically, C1QBP exerts this function by limiting DOT1L-mediated H3K79me3 deposition across the Il1b gene body, with comparable regulatory roles observed in both BMDCs and primary cDC2s.

C1QBP suppresses Dectin-1-induced Il1b transactivation by inhibiting RNF20/RNF40-mediated H2B ubiquitination and H3K79me3 deposition

Since monoubiquitination of histone H2B at lysine 120 (H2B-ub) is an established prerequisite for DOT1L-mediated H3K79 methylation^28,29, we explored whether C1QBP regulates this upstream epigenetic event. In C1QBP-deficient BMDCs, curdlan stimulation induced a time-dependent increase in H2Bub1, but not in total H2B or histone H4 levels (Fig. 4A). This effect was specific to the H2Bub1-H3K79me3 axis, as C1QBP deficiency did not alter the transcription of metabolic genes or affect H3K79 methylation via glycolytic inhibition (Supplementary Fig. 8A and 8B). We therefore investigated the H2B ubiquitin ligase complex RNF20/RNF40. Further analysis revealed that C1QBP deficiency in BMDCs significantly enhanced curdlan-induced phosphorylation of RNF20 and RNF40, without affecting their basal expression or protein stability (Fig. 4B). No changes of Rnf20 or Rnf40 mRNA levels were observed in C1QBP-deficient BMDCs (Supplementary Fig. 8C), indicating that C1QBP regulates RNF20/RNF40 activation rather than transcriptional control. To functionally validate this, we depleted Rnf20 or Rnf40 via siRNA (Supplementary Fig. 8D). Knockdown of either subunit significantly reduced the elevated levels of H2Bub1, H3K79me3, and pro-IL-1β protein in C1QBP-deficient BMDCs (Fig. 4C and D). Accordingly, the enhanced H3K79me3 occupancy across the Il1b gene body was abolished by RNF20/RNF40 depletion (Fig. 4E). At the transcriptional level, Rnf20 or Rnf40 knockdown significantly reduced Il1b mRNA expression (Fig. 4F) and abrogated IL-1β hypersecretion in C1QBP-deficient BMDCs following curdlan stimulation (Fig. 4G). Collectively, these data establish that C1QBP constrains Dectin‑1‑induced IL‑1β production by inhibiting RNF20/RNF40 activation, thereby limiting H2Bub1 availability and suppressing DOT1L-dependent H3K79me3 deposition and transcriptional elongation at the Il1b locus.

Fig. 4. C1QBP interacts with RNF20/RNF40 to suppress H2B-ub-dependent H3K79me3 and Il1b transactivation.

A Immunoblotting was performed to assess the levels of H2B monoubiquitylation, total H2B and H4K16 acetylation in BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice following stimulation with plate-coated curdlan (50 μg/mL) at various time points. B Co-immunoprecipitation (Co-IP) of endogenous C1qbp with Rnf20 or Rnf40 in curdlan-stimulated from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice, at the indicated time intervals. Immunoblot analysis of H2Bub1, H3K79me3, and pro-IL-1β in KO BMDCs transfected with non-targeting control (NC), Rnf20 (C), or Rnf40 (D) siRNA after curdlan (50 μg/mL) stimulation at designated time points. E ChIP-qPCR analysis of H3K79me3 occupancy at the Il1b gene body in siRNA-in BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice transfected with Rnf20/40-targeting or NC siRNA after curdlan stimulation (50 μg/mL) for 3 hours, utilizing control IgG or anti-H3K79me3 antibodies. F Il1b mRNA expression levels in C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ BMDCs transfected with siRNAs targeting Rnf20, Rnf40, or NC, after stimulation with curdlan (50 μg/mL) for 6 h. G IL-1β secretion in the supernatants from BMDCs of C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice, transfected with Rnf20/40-targeting or NC siRNA, followed by stimulation with curdlan (50 μg/mL) for 18 h and ATP (5 mM) for 1 h. Data were presented as mean ± SEM, n = 3 (A–D), n = 5 (E–G) biologically independent samples. Statistical significance was determined by unpaired two-tailed Student’s t test. ns, not significant. Source data are provided as a Source Data file.

C1QBP gates RNF20/RNF40-dependent H3K79me3 and Dectin-1-mediated IL-1β production by restraining PKCδ nuclear trafficking

Having established that C1QBP suppresses the RNF20/RNF40 complex, we next investigated the mechanistic basis for this inhibition within Dectin-1 signaling. We performed anti‑RNF20 immunoprecipitation coupled with mass spectrometry in curdlan‑stimulated BMDCs. PPI network analysis of the enriched partners revealed a specific enhancement in the association between RNF20 and kinase PKCδ, as well as its cognate partner RNF40, specifically in C1QBP-deficient BMDCs (Fig. 5A). Subsequent exogenous co-immunoprecipitation assays confirmed the direct interactions between PKCδ and both RNF20 and RNF40 (Supplementary Fig. 9A). The interactions were markedly enhanced in the nuclear fraction of C1QBP-deficient BMDCs following curdlan stimulation (Supplementary Fig. 9B). Furthermore, in vitro kinase assays established that PKCδ directly phosphorylated serine residues on RNF20 and RNF40 (Fig. 5B). Notably, C1QBP deficiency did not affect proximal Dectin-1 pathway activation, as phosphorylation of Syk and IκBα remained comparable to controls (Supplementary Fig. 9C), positioning the role of C1QBP downstream of Dectin-1 signaling. These data identify PKCδ as a critical link connecting proximal Dectin‑1 signaling to the C1QBP-restrained RNF20/40 complex.

Fig. 5. C1QBP restrains PKCδ nuclear translocation to gate RNF20/RNF40-dependent H3K79me3 and IL-1β production.

A PPI network of RNF20-associated proteins identified by mass spectrometry in curdlan-stimulated C1qbp^fl/flCD11c^Cre/+ (KO) BMDCs. B Immunoblot analysis showing serine phosphorylation of HA-tagged Rnf20, HA-tagged R40, or Flag-tagged Card9 after incubation with purified PKCδ (0.5 μg) in the presence or absence of ATP (50 μM). C Co-IP analysis of RNF20/RNF40 complex and serine phosphorylation in KO BMDCs transfected with PKCδ or control siRNA following curdlan stimulation (50 μg/mL) for 3 h. D Immunoblot of phospho- and total PKCδ in nuclear and cytoplasmic fractions of BMDCs after curdlan stimulation. E Immunoblot of nuclear PKCδ in WT BMDCs transfected with Rack1 or control siRNA after curdlan stimulation. F Immunofluorescence staining of RACK1 (red) and PKCδ (green) in BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice after curdlan stimulation (n = 8 per group). Scale bars, 10 µm. G Co-IP of PKCδ with RNF20/RNF40 and (H) Immunoblot of H2Bub1 and H3K79me3 in Rack1-knockdown WT BMDCs after curdlan stimulation. I Immunoblot of H2Bub1 and H3K79me3 in BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice treated with nuclear import inhibitor ivermectin during curdlan stimulation. J Immunoblot analysis showing levels of pro-IL-1β expression, H2B monoubiquitylation, and H3K79 methylation in BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice, transfected with PKCδ-targeting siRNA or NC siRNA, after stimulation with curdlan. KChIP-qPCR analysis of H3K79me3 occupancy at the Il1b gene body in PKCδ-knockdown BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice. Il1b mRNA (L) and IL-1β secretion (M) in PKCδ-knockdown BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice after curdlan stimulation. Data were presented as mean ± SEM, n = 3(B–E, G–J), n = 8(F), n = 5(K–M) biologically independent samples. Statistical significance was determined by unpaired two-tailed Student’s t-test. ns, not significant. Source data are provided as a Source Data file.

To define the functional role of PKCδ, we depleted it using validated siRNA (Supplementary Fig. 9D, E). In C1QBP-deficient BMDCs, PKCδ knockdown significantly reduced curdlan-induced RNF20/RNF40 phosphorylation levels (Fig. 5C). We next examined the subcellular localization underlying this activity. Subcellular fractionation revealed that C1QBP deficiency promoted a time‑dependent increase in the nuclear accumulation of both phosphorylated and total PKCδ upon curdlan stimulation, accompanied by a decrease in the cytoplasmic pool (Fig. 5D), indicating that C1QBP restricts the curdlan‑induced nuclear translocation of PKCδ in BMDCs.

Given that Rack1 was significantly downregulated alongside C1qbp in Il1b^hi cDC2s (Fig. 1C) and its known role in compartmentalizing activated PKC isoforms³⁰, we hypothesized that diminished RACK1 expression contributes to altered PKCδ localization upon C1QBP deficiency. Supporting this, RACK1 knockdown (Supplementary Fig. 9F) in wild‑type BMDCs sufficed to promote PKCδ nuclear accumulation after curdlan stimulation (Fig. 5E), phenocopying the distribution in C1QBP-deficient cells. Immunofluorescence analysis further revealed that in wild‑type BMDCs, curdlan stimulation reduced cytoplasmic RACK1 intensity and increased nuclear PKCδ signal. These changes were amplified in C1QBP-deficient BMDCs, which exhibited a greater decline in cytoplasmic RACK1 and more pronounced nuclear PKCδ accumulation upon stimulation (Fig. 5F). This reciprocal pattern visually supports a model wherein C1QBP deficiency diminishes cytoplasmic RACK1 retention, thereby facilitating PKCδ nuclear entry. We further found that RACK1 knockdown in wild‑type BMDCs significantly enhanced the interaction of PKCδ with both RNF20 and RNF40 after curdlan stimulation (Fig. 5G), indicating that reduced RACK1 levels promote the formation of the nuclear PKCδ–RNF20/RNF40 complex, thereby mechanistically linking altered subcellular localization to augmented epigenetic regulator activity. In wild-type BMDCs, RACK1 knockdown significantly increased H2B monoubiquitination and H3K79me3 following curdlan stimulation, without altering total H2B, H3K79me1/2, or global H3 levels (Fig. 5H). This specific augmentation of the H2Bub1‑H3K79me3 axis upon RACK1 knockdown functionally connects altered PKCδ subcellular dynamics to enhanced RNF20/RNF40‑DOT1L epigenetic pathway activity. To establish that PKCδ nuclear entry is necessary for this amplified epigenetic response, we used the nuclear import inhibitor ivermectin, which blocks Impα/β1‑dependent translocation³¹. Ivermectin treatment abolished the enhancement of β‑glucan‑induced H2Bub1 and H3K79me3 in C1QBP‑deficient BMDCs, reducing both marks to control levels (Fig. 5I), by blocking PKCδ nuclear accumulation (Supplementary Fig. 9G). These data demonstrate that Impα/β1‑mediated nuclear import of PKCδ is required for elevated H2B ubiquitination and H3K79 trimethylation upon C1QBP deficiency.

We further investigated the requirement of PKCδ for elevated epigenetic marks and IL‑1β production in C1QBP-deficient BMDCs. PKCδ knockdown in C1QBP-deficient BMDCs abolished the time‑dependent increases in H2Bub1, H3K79me3, and pro-IL‑1β protein upon curdlan stimulation, without affecting total H2B, H3K79me1/2, or global H3 levels (Fig. 5J). ChIP‑qPCR further showed that C1QBP deficiency enhanced H3K79me3 occupancy across the Il1b gene body after curdlan stimulation, an enrichment completely abolished by PKCδ knockdown (Fig. 5K), demonstrating that PKCδ is necessary for directing DOT1L‑dependent H3K79 trimethylation to the Il1b locus. Functionally, PKCδ depletion in C1QBP-deficient BMDCs significantly reduced the heightened Il1b mRNA expression (Fig. 5L) and IL‑1β secretion (Fig. 5M) to control levels. Collectively, C1QBP acts as a brake by maintaining RACK1 to constrain PKCδ nuclear import. Deficiency in this axis licenses PKCδ to phosphorylate and activate the nuclear RNF20/RNF40 complex, culminating in H3K79me3 deposition at the Il1b locus and enhanced IL-1β secretion.

C1QBP integrates with Dectin-1/Syk signaling to spatially restrain PKCδ nuclear trafficking and suppress H3K79me3-dependent hyper-transcription of Il1b

Having delineated the role of the C1QBP-RACK1 module in restricting PKCδ nuclear translocation to control downstream epigenetic activation, we next investigated how this spatial checkpoint integrates with upstream Dectin-1 signaling. Co-IP assays conducted in 293 T cells showed a direct physical interaction between Myc-tagged PKCδ and Flag-tagged C1QBP (Fig. 6A). Domain-mapping studies showed that the C2 domain of PKCδ was essential for binding C1QBP, while C1QBP required an intact tertiary structure, as deletion of its mitochondrial-targeting motif (MTS), disordered region, or MAM33 domain abrogated the association (Supplementary Fig. 10A and 10B). This interaction was physiologically relevant as evidenced by elevated levels of endogenous PKCδ co-precipitating with C1QBP in BMDCs after curdlan stimulation (Fig. 6B), indicating that Dectin-1 activation actively promotes C1QBP-PKCδ complex formation. Confocal microscopy revealed that upon curdlan stimulation, activated PKCδ translocated to mitochondria, where it extensively co-localized with the mitochondrial pool of C1QBP (Fig. 6C). Quantitative Pearson’s correlation analysis confirmed a significant increase in their mitochondrial co-localization following curdlan stimulation (Fig. 6C), providing spatial evidence that Dectin-1 signaling recruits activated PKCδ to a C1QBP-enriched mitochondrial compartment for sequestration. To characterize the functional properties of this interaction, we inhibited PKCδ activation using Rottlerin during curdlan stimulation. While Rottlerin treatment abolished PKCδ phosphorylation without affecting its total abundance, it significantly reduced the amount of both phosphorylated and total PKCδ co-precipitating with C1QBP (Fig. 6D), demonstrating that C1QBP preferentially binds the activated, phosphorylated form of PKCδ.

Fig. 6. C1QBP integrates with Dectin-1/Syk signaling to sequester PKCδ and suppress downstream epigenetic activation.

A Immunoblot analysis demonstrating the interaction between C1qbp and PKCδ in 293 T cells transfected with various combinations of expression constructs, following immunoprecipitation using an anti-Flag antibody. BCo-IP of endogenous C1qbp and PKCδ in WT BMDCs after curdlan stimulation (50 μg/mL) at specified time points. C Confocal microscopy showing mitochondrial(turquoise) co-localization of C1qbp (red) and PKCδ (green) in BMDCs after curdlan stimulation for 60 minutes. Scale bars, 5 µm. Pearson’s correlation coefficient was quantified. D Immunoblot analysis of C1qbp interactions with phosphorylated PKCδ in BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice stimulated with curdlan (50 μg/mL), with or without rottlerin (20 μM), at indicated time points. E Co-immunoprecipitation and immunoblot analysis of the interactions between PKCδ and Syk, along with PKCδ phosphorylation levels, in BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice following curdlan stimulation (50 μg/mL) at specified time points. F Immunoblot analysis of PKCδ-Syk interactions in 293 T cells overexpressing designated protein combinations, following immunoprecipitation with an anti-Myc antibody. G FRET assays revealing interactions between Cy3-labeled PKCδ (green) and Cy5-labeled Syk (red) in BMDCs from C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice after 1 h of curdlan stimulation (50 μg/mL, n = 8 per group). Scale bars represent 10 μm. H Immunoblot analysis of C1qbp interactions with PKCδ in wild-type, Clec7a^–/–, or Syk^fl/flLyz2^Cre/+ BMDCs stimulated with plate-coated curdlan (50 μg/mL) for 30 min, following immunoprecipitation with IgG or anti-C1QBP antibodies. Immunoblot analyses depicting nuclear translocation and phosphorylation of PKCδ (I), nuclear interactions between RNF20/RNF40 and PKCδ (J), and assessments of H2B monoubiquitylation and H3K79 methylation (K) in wild-type, Clec7a^–/–, or Syk^fl/flLyz2^Cre/+ BMDCs stimulated with curdlan (50 μg/mL) at specified time points. L ChIP-qPCR analysis of H3K79me3 occupancy at the Il1b gene body in BMDCs from wild-type, Clec7a^-/-, or Syk^fl/flLyz2^Cre/+ mice after 3 hours of curdlan stimulation (50 μg/mL), using control IgG or anti-H3K79me3 antibodies. Data were presented as mean ± SEM, n = 3(A-B, D-F, H–K), n = 8(G), n = 5(L) biologically independent samples. Statistical significance was determined by unpaired two-tailed Student’s t test. ns, not significant. Source data are provided as a Source Data file.

We next explored how C1QBP engagement influences the activation of PKCδ within the Dectin-1 pathway. Reconstitution Co-IP assays confirmed that PKCδ can independently bind either its activating kinase Syk or C1QBP, and that all three proteins can be co-precipitated in a complex (Supplementary Fig. 10C). Endogenous Co-IP in BMDCs showed that curdlan stimulation induced a progressive association between PKCδ and Syk, which was markedly enhanced in C1QBP-deficient BMDCs (Fig. 6E). This enhanced physical coupling correlated with a more intense and transient peak of PKCδ phosphorylation following curdlan stimulation (Fig. 6E), indicating that C1QBP intrinsically constrains the signal-induced coupling of PKCδ to Syk, thereby limiting the amplitude of PKCδ activation. To dissect the mechanism, we employed a reconstituted system with titrated expression of HA-C1QBP (Fig. 6F). Increasing HA-C1QBP levels enhanced its co-precipitation with Myc-PKCδ and led to a dose-dependent displacement of Flag-Syk from the PKCδ complex, accompanied by a proportional reduction in PKCδ phosphorylation (Fig. 6F). This demonstrates that C1QBP acts as a competitive scaffold, binding PKCδ to sterically hinder the functional Syk-PKCδ kinase-substrate interaction and quench phosphorylation. Fluorescence resonance energy transfer (FRET) analysis in BMDCs provided nanoscale spatial validation: the curdlan-induced FRET signal between Cy3-PKCδ and Cy5-Syk was significantly amplified in C1QBP-deficient BMDCs (Fig. 6G), confirming that C1QBP restricts the close-range interaction between PKCδ and Syk.

We then interrogated the hierarchical relationship between upstream Dectin-1-Syk signaling and the C1QBP-PKCδ scaffold. Genetic ablation of Dectin-1 or Syk significantly impaired the curdlan-induced association between C1QBP and PKCδ (Fig. 6H), indicating that efficient, signal-enhanced complex formation is licensed by upstream Dectin-1/Syk activity. Subcellular fractionation analysis revealed that the curdlan-induced, time-dependent nuclear accumulation of phosphorylated and total PKCδ was virtually abolished in BMDCs lacking either Dectin-1 or Syk (Fig. 6I). Consequently, the stimulation-dependent interaction of PKCδ with RNF20/RNF40 was essentially abolished in these cells (Fig. 6J), as was the amplification of the H2Bub1-H3K79me3 axis (Fig. 6K). Consistent with these molecular deficits, the curdlan-induced enrichment of H3K79me3 across the transcribed region of the Il1b locus was eliminated in BMDCs lacking either Dectin-1 or Syk (Fig. 6L). Collectively, these data suggest that C1QBP acts a critical spatial checkpoint downstream of Dectin-1/Syk signaling that sequesters PKCδ to disrupt active Syk-signalosome assembly, subsequently inhibiting RNF20/40 activation and H3K79me3-dependent Il1b transcription.

C1QBP constrains host antifungal defense by epigenetically restricting DOT1L-mediated IL-1β production, thereby limiting the protective Th17 response

To determine whether the C1QBP/PKCδ/DOT1L axis governs antifungal immunity in vivo, we assessed the effect of DOT1L inhibition following C. albicans infection. During infection, DC‑specific C1QBP‑deficient mice exhibited significantly improved survival and reduced fungal burdens in the kidneys and brains compared to controls (Fig. 7A, B). Strikingly, treatment with the DOT1L inhibitor pinometostat completely abolished both the survival advantage and the reduction in organ fungal loads in C1QBP-deficient mice, bringing them to levels comparable to pinometostat‑treated controls (Fig. 7A, B). Consistent with the observed role of DOT1L in driving Il1b transcription in vitro, renal IL‑1β levels were elevated in C1QBP-deficient mice, and this increase was abrogated by pinometostat treatment (Fig. 7C). In contrast, IL‑6 levels were unaffected by genotype or treatment (Fig. 7C). Histopathological analysis confirmed that the diminished fungal invasion revealed by PAS staining in C1QBP-deficient mice was reversed upon DOT1L inhibition, whereas inflammatory infiltration shown with H&E staining was comparable across all groups (Fig. 7D). These data directly link the C1QBP/DOT1L axis to the control of IL‑1β production and fungal clearance in target organs.

Fig. 7. C1QBP constrains host antifungal defense by epigenetically restricting DOT1L-mediated IL-1β production and consequent Th17 response.

A Survival of C1qbp^fl/fl and C1qbp^fl/flCD11c^Cre/+ mice treated with DOT1L inhibitor pinometostat (1 mg/mouse) or vehicle after C. albicans infection (n = 6 per group). B Fungal burdens in kidney and brain at day 2 post-infection in treated mice (n = 5 per group). CELISA quantification of IL-1β and IL-6 in kidney homogenates at day 2 post-infection in treated mice (n = 5 per group). D Representative H&E and PAS staining of kidney sections at day 4 post-infection (n = 5 mice per group). Scale bars: 100 µm. Arrows indicated fungal hyphae. E Flow cytometry analysis of renal Th17 cell frequencies (IL-17A⁺, IL-17F⁺, IL-22⁺ CD4⁺ T cells) at day 6 post-infection in treated mice (n = 5 per group). Data were mean ± SEM; significance determined by unpaired two-tailed t test (B–E) or Log-rank (Mantel-Cox) test (A). ns, not significant. Source data are provided as a Source Data file.

Given the pivotal role of IL‑1β in promoting Th17 responses, we analyzed the adaptive immunity of infected mice. Flow cytometric analysis of renal lymphocytes revealed that C1QBP deficiency significantly increased the frequencies of CD4⁺ IL‑17 A⁺, CD4⁺ IL‑17 F⁺, and CD4⁺ IL‑22⁺ T cells (Fig. 7E). Pinometostat treatment reversed this enhancement, reducing the frequency of Th17 cells in C1QBP-deficient mice to control levels, while the frequency of IFN‑γ⁺ T cells remained unchanged (Fig. 7E). Together, these data demonstrate that C1QBP constrains host antifungal defense by serving as an epigenetic gatekeeper of DOT1L-mediated IL-1β production, which in turn restrains a protective Th17 response against C. albicans infection.

Discussion

Dectin-1-dependent IL-1β production is a cornerstone of antifungal immunity, yet the transcriptional and epigenetic mechanisms that govern its magnitude and specificity have remained poorly defined. In this study, we identify C1QBP as a critical transcriptional repressor that constrains Dectin-1-driven IL-1β production in cDC2s during systemic C. albicans infection. We demonstrate that C1QBP functions as a spatial rheostat within the Dectin-1/Syk signaling by sequestering phosphorylated PKCδ in the cytoplasm, thereby preventing its nuclear translocation and subsequent activation of the RNF20/RNF40-DOT1L-H3K79me3 axis at the Il1b locus. DC-specific ablation of C1QBP unleashed this epigenetic pathway, amplifying IL-1β-dependent Th17 responses and enhancing host resistance to disseminated candidiasis. These findings not only elucidate a spatially regulated checkpoint for innate immune gene transcription but also nominate C1QBP as a therapeutic target to potentiate antifungal defense.

Our multi-omics approach unequivocally established cDC2s as the dominant source of renal IL-1β during C. albicans infection. The inverse correlation between C1qbp expression and Il1b induction in cDC2s, coupled with GSVA revealing suppressed “negative regulation of IL-1β production” in C1qbp^lo cells, positioned C1QBP as a key homeostatic regulator. C1QBP was initially characterized as a component of the classical complement pathway, where it binds C1q to facilitate immune complex clearance³². However, emerging evidence has expanded its functional repertoire beyond complement activation³³, including roles in cancer, where it promotes tumor survival and metastasis via modulation of mitochondrial metabolism and chromatin accessibility³⁴. Our study extends this paradigm by defining an unknown role for C1QBP as an epigenetic gatekeeper in DCs during fungal infection. The induction of C1QBP in DCs upon Dectin-1 engagement aligns with its emerging role as a modulator of innate immunity. A recent study indicates C1QBP restrains TLR4-driven inflammation by stabilizing mitochondrial complexes, suggesting a conserved function in mitigating excessive responses across PRR pathways³⁵.

Notably, we identify a previously uncharacterized cytoplasmic role for C1QBP as a molecular brake on aberrant p-PKCδ nuclear translocation. This mechanism is consistent with evidence that C1QBP shuttles dynamically between cellular compartments to regulate processes such as apoptosis and ribosome biogenesis³⁶. The most salient mechanistic insight of our work is the elucidation of how a cytoplasmic-mitochondrial protein can govern nuclear epigenetic events. As a Dectin-1 signal-dependent compartmentalization factor, C1QBP disruption specifically enables p-PKCδ nuclear entry, diverging from plasma membrane-focused signaling. Recent studies have highlighted nuclear functions of p-PKCδ in transcriptional regulation, including Stat1 phosphorylation and NF-κB activation^37,38. Spatially, our data reveal that C1QBP, through maintaining RACK1 expression, anchors activated PKCδ at the mitochondria, thereby limiting its nuclear access. The enhanced co-localization of p-PKCδ with mitochondrial C1QBP upon curdlan stimulation suggests that Dectin-1 signaling actively recruits PKCδ to a C1QBP-rich compartment for sequestration. This model is supported by the competitive displacement of Syk from PKCδ by C1QBP, which dampens PKCδ phosphorylation and subsequent nuclear import. By sequestering p-PKCδ in the cytoplasm, C1QBP prevents these nuclear events, positioning itself as a key node in inflammatory control. Extending this to CLR signaling, we show that C1QBP directly interferes with Syk-PKCδ complex assembly, with competition between C1QBP and Syk for p-PKCδ binding representing a mechanism to dampen signal amplification downstream of Dectin-1. Given the role of PKCδ as an integrator of signals from multiple immune receptors³⁹, C1QBP may serve as a universal modulator of PKCδ-dependent responses beyond antifungal immunity.

Histone modifications, particularly H3K79me3, have emerged as central regulators of immune gene expression. For example, in influenza infection, H3K79me3 enhances interferon-induced antiviral responses by promoting chromatin accessibility¹⁶. In contrast, our data reveal that H3K79me3 deposition at the Il1b locus is tightly constrained by C1QBP to prevent excessive IL-1β production. The requirement for H2B-Ub-dependent H3K79me3 in Il1b transactivation points to a two-tiered epigenetic control of cytokine production: RNF20/RNF40-mediated H2B-Ub facilitates DOT1L activity, driving H3K79me3 deposition and transcriptional elongation of Il1b. This pathway, selectively activated by Dectin-1 signaling, is suppressed by C1QBP’s interception of p-PKCδ nuclear trafficking. Importantly, C1QBP specifically modulates H3K79me3 at Il1b without affecting other Dectin-1-induced cytokines (e.g., IL-6), likely reflecting locus-specific differences in enhancer-promoter architecture or cofactor recruitment, highlighting the precision of epigenetic regulation in innate immunity.

Our in vivo findings reinforce the physiological relevance of this pathway: C1QBP deficiency enhances fungal clearance via amplification of the IL-1β/ Th17 axis. This aligns with established roles for IL-1β in promoting Th17 responses against C. albicans and corroborates human genetic studies linking Dectin-1 polymorphisms (e.g., Y238X) to mucocutaneous candidiasis, driven by impaired IL-17 production⁴⁰. Our work suggests C1QBP expression or function may contribute to this variability: polymorphisms in C1QBP or its interactors (e.g., PKCδ) could alter checkpoint efficacy, influencing clinical susceptibility to fungal infections. Given that CARD9 deficiency causes severe fungal infections due to defective CLR signaling^40,41, our findings position C1QBP as a modifiable component of this critical pathway. Notably, while Th17 responses are indispensable for mucosal antifungal host defense, their contribution to protection against systemic candidiasis remains context-dependent^25,42. Using a sublethal systemic C. albicans infection model, we demonstrate that C1QBP deficiency augments fungal clearance and confers robust renal protection via selective potentiation of the IL-1β/Th17 axis, although the relative contribution of Th17-mediated immunity to this protective phenotype may vary across different experimental paradigms and disease stages.

While we recognize that the CD11c-Cre transgenic line, despite being the most widely used tool for cDC targeting, exhibits low-level recombination in a small subset of non-DC cell types^43,44, multiple independent lines of evidence validate the specificity of cDC2-mediated regulation in our system. Specifically, cDC2-specific adoptive transfer experiments and Lyz2-Cre-mediated myeloid lineage knockout studies collectively confirm that the observed protective phenotype is exclusively driven by C1QBP ablation in cDC2s. Furthermore, our current findings were established in immunocompetent mice to delineate the fundamental cellular and molecular mechanisms underlying C1QBP-dependent immune regulation. Future investigations using clinically relevant immunocompromised models will be critical to fully elucidate the translational potential of targeting this pathway for the treatment of invasive candidiasis.

In summary, we redefine C1QBP as a pivotal epigenetic checkpoint that prevents excessive inflammation during fungal sensing. By sequestering p-PKCδ, C1QBP limits nuclear RNF20/RNF40 activation and H3K79me3 licensing at Il1b, balancing effective pathogen clearance and protection from immunopathology. This mechanism expands the paradigm of innate immune regulation by introducing spatial control of kinase signaling as a gatekeeper of epigenetic reprogramming. Targeting C1QBP may thus offer a strategy to recalibrate immune responses in fungal infections and beyond.

Methods

Mice

C57BL/6 J mice (6-8 weeks old) were purchased from GemPharmatech Co., Ltd. (Nanjing, China). Clec7a^-/- and Card9^-/- mice were kindly provided by Dr. X. Lin (Tsinghua University School of Medicine, Beijing, China), while Syk^fl/fl, Syk^fl/flLyz2^Cre/+ mice were kindly provided by Dr. H. Xiao (Institut Pasteur of Shanghai, China). C1qbp^fl/fl, CD11c^Cre/+, and Lyz2^Cre/+ mice were purchased from GemPharmatech Co., Ltd. All mice were genotyped via polymerase chain reaction (PCR) using genomic DNA extracted from tail tissue and housed under specific pathogen-free (SPF) conditions at Tongji University. The animal housing environment was maintained under specific pathogen-free conditions with a 12-h light and 12-h dark cycle, ambient temperature around 21–22 °C, and humidity between 40 and 70%. All animal experimental procedures were approved by the Animal Ethics Committee of Tongji University School of Medicine (Approval No.: TJAA09021102). All animal studies were performed using sex-and age-matched male or female mice.

Cell lines

HEK293T cells were cultured in 10 cm Petri dishes using Dulbecco’s Modified Eagle Medium (DMEM; Meilunbio, MA0212) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin (Meilunbio, MA0110). Cultures were maintained at 37 °C in a humidified incubator with 5% CO₂.

Generation and stimulation of BMDCs

Bone marrow cells were isolated from the femurs and tibias of mice and cultured in RPMI 1640 medium supplemented with 10% FBS and 20 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF). Half of the medium was replaced on days 3 and 6 of culture. On day 8, non-adherent CD11c⁺ BMDCs were collected. For enzyme-linked immunosorbent assay (ELISA), 24-well plates were coated with 10 μg curdlan (Sigma-Aldrich, C7821) or 10 μg depleted zymosan (Invivogen, tlrl-zyd) dissolved in ethanol; subsequently, 5 × 10⁵ BMDCs were seeded and stimulated for 18 h. For immunoblotting or histone modification assays, 6-well plates were coated with 100 μg curdlan, and 5 × 10⁶ BMDCs were seeded for stimulation at the indicated time points. For co-immunoprecipitation (Co-IP) or chromatin immunoprecipitation (ChIP)-PCR assays, 10 cm dishes were coated with 400 μg curdlan, and 1 × 10⁷ BMDCs were seeded for stimulation as specified.

Fungal strains and culture

The reference strain SC5314 of C. albicans (American Type Culture Collection, ATCC) was cultured in yeast extract-peptone-dextrose (YPD) medium at 30 °C for 16–18 h. Yeast cells were washed three times with phosphate-buffered saline (PBS) and quantified using a hemocytometer. Heat-killed C. albicans was prepared by incubating yeast cells at 95 °C for 1 h. Hyphal induction was performed by culturing yeast cells in RPMI 1640 medium (Meilunbio, MA0215) at 37 °C for 3 h.

Mouse model of systemic Candidiasis

For survival analysis, mice were intravenously injected with 3 × 10⁵ colony-forming units (CFU) of C. albicans via the lateral tail vein. For fungal burden quantification, mice received an intravenous injection of 2 × 10⁵ CFU of C. albicans. Kidneys were harvested for CFU counting, cytokine detection, flow cytometric analysis, and histopathological examination.

Fungal burden analysis

Kidneys were aseptically excised, weighed, and homogenized in 1 mL PBS using a high-speed cryogenic tissue grinder (Servicebio, China). Serial dilutions of the homogenates were plated on Sabouraud dextrose agar plates, which were then incubated at 37 °C for 24 h for CFU enumeration.

Histology assay

Kidneys were fixed in 4% paraformaldehyde, embedded in paraffin, and sent to Servicebio for periodic acid-Schiff (PAS) and hematoxylin-eosin (H&E) staining. Stained sections were scanned and analyzed using CaseViewer software.

Isolation of CD45⁺ immune single-cells from Kidneys

Kidneys were lysed in lysis buffer containing 0.1% NP-40, 10 mM Tris-HCl (pH 7.4), 146 mM NaCl, 1 mM CaCl₂, 21 mM MgCl₂, and 1 U/μL RNase inhibitor (Thermo Scientific) for 7 min. Lysis efficiency was verified by trypan blue staining. Cells were washed with buffer containing 0.01% bovine serum albumin (BSA), filtered through a 40 μm cell strainer, and centrifuged at 4 °C. CD45⁺ cells were purified using magnetic microbeads (Miltenyi Biotec) and pooled from three mice per experimental group. Single-cell RNA sequencing (scRNA-seq) libraries were constructed using the 10× Genomics Chromium Next GEM Single Cell 3’ Reagent Kit v4 and sequenced on the BGI DNBSEQ-T7 platform with paired-end 100 bp reads, targeting ≥30,000 reads per cell. Sequencing and downstream bioinformatics analyses were performed by OE Biotech Co., Ltd. (Shanghai, China).

Single-cell RNA-seq analysis

Raw sequencing FASTQ reads were mapped to the mouse mm10-3.0.0 reference genome assembly utilizing 10×Genomics Cell Ranger v9.0.0 analytical suite. This workflow simultaneously quantified unique molecular identifier (UMI) abundance corresponding to individual cellular barcodes and constructed raw UMI count matrices. All subsequent downstream computational analyses were implemented in the R programming environment with the Seurat v4.0.0 toolkit.

Transcriptomic normalization was executed via the NormalizeData function adopting the “LogNormalize” normalization strategy: per-cell total UMI counts were scaled to a uniform library size of 10000, followed by natural logarithmic transformation. A total of 2000 highly variable genes (HVGs) were screened through the FindVariableGenes algorithm, where FastExpMean and FastLogVMR were designated as the mean expression estimator and dispersion calculation function, respectively. Linear dimensionality reduction was achieved by principal component analysis (PCA) implemented via the RunPCA routine. For cross-sample batch effect mitigation, dataset integration and batch correction were performed using the RunHarmony function from the Harmony R package v1.0.

Graph-based unsupervised cellular clustering was accomplished with the FindClusters module. Nonlinear two-dimensional manifold visualization of single-cell transcriptional profiles was generated via uniform manifold approximation and projection (UMAP) using the RunUMAP function. Cluster-enriched signature marker genes were uncovered by FindAllMarkers with the presto statistical test specified by the parameter test.use=presto. Intergroup differential gene expression profiling was conducted using FindMarkers with the identical presto testing framework; raw P values derived from differential testing were subjected to Bonferroni multiple-testing correction. The GSEABase v1.44.0 R package was deployed to import curated gene set resources, which were retrieved and preprocessed from the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (https://www.kegg.jp/) and Gene Ontology (GO) database (https://geneontology.org/) for subsequent functional enrichment interrogation.

Bulk RNA-seq analysis

Total cellular RNA was isolated via the miRVana RNA Isolation Kit (Invitrogen) following the manufacturer’s standard experimental workflow. RNA integrity evaluation was implemented on an Agilent 2100 Bioanalyzer system (Agilent Technologies). Only specimens possessing an RNA Integrity Number (RIN) no less than 7 were subjected to cDNA library construction employing the TruSeq Stranded mRNA LT Sample Prep Kit (Illumina). Constructed sequencing libraries were subjected to paired-end sequencing on Illumina HiSeq platforms (HiSeq 2500 or HiSeq X Ten), generating 125 bp or 150 bp paired-end sequencing reads; all sequencing services were commissioned to OE Biotech Co., Ltd. (Shanghai, China).

Raw paired-end sequencing reads underwent adapter clipping and quality filtration with Trimmomatic v0.39. Post-quality-control clean reads were subsequently mapped to the designated reference genome utilizing HISAT2 v2.1.0. Global alignment quality metrics were quantified via Qualimap v2.2.1, and gene-level read count matrices were generated by HTSeq v0.11.1.

For C1QBP loss-of-function knockdown assays, bone marrow-derived dendritic cells (BMDCs) with C1QBP depletion were designated as the experimental group, whereas wild-type BMDCs served as matched negative controls. Transcriptomic datasets from C1QBP-knockdown samples were comparatively analyzed against their corresponding wild-type control counterparts.

Differential expression profiling for screening differentially expressed genes (DEGs) was performed using the edgeR R package v3.34.0 based on Fragments Per Kilobase of transcript per Million mapped reads (FPKM) normalized expression values. Transcripts satisfying the dual screening thresholds of |log₂(fold change)|å 1 and nominal P-value <0.05 were defined as statistically significant DEGs.

Protein-protein interaction (PPI) networks for genes within modules of biological interest were retrieved from the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database. The Cytoscape v3.10.1 software with its CytoHubba plug-in was further applied to pinpoint hub genes in PPI networks via the Maximal Clique Centrality (MCC) scoring algorithm.

Immunofluorescence staining

BMDCs (5 × 10⁵) were seeded onto 14 mm glass-bottom confocal dishes and stimulated with 10 μg curdlan for the indicated time points. Cells were fixed in 4% paraformaldehyde for 15 min, permeabilized with 0.25% Triton X-100 for 20 min, and blocked with 2% BSA in PBS for 1 h at room temperature. Samples were incubated with the indicated primary antibodies overnight at 4 °C, washed three times with PBS, and then incubated with appropriate fluorophore-conjugated secondary antibodies for 1 h at room temperature. Mitochondria were counterstained with MitoTracker™ Red CMXRos, and nuclei were counterstained with 4’,6-diamidino-2-phenylindole (DAPI; Beyotime, C1002, 2 μg/mL). Images were acquired using a Zeiss LSM900 confocal laser-scanning microscope. Pearson’s correlation coefficients were calculated using the Coloc2 plugin in ImageJ software (version 1.8.0). The antibodies used in this process were as follows: AF 488-anti-Rabbit IgG (Jackson,711-545-152, 1:500), AF 647-anti-Rabbit IgG (Jackson, 711-605-152, 1:500), AF-647 anti-GC1qR (Abcam, Ab305041, 1:100).

Fluorescence resonance energy transfer (FRET)

14 mm confocal dishes were coated with 10 μg curdlan overnight. BMDCs (5 × 10⁵) were seeded and stimulated for 60 min, then fixed with 4% paraformaldehyde, permeabilized, and blocked as described in the Immunofluorescence Staining section. Cy5-conjugated anti-Syk antibody (1:200, Bioss, bs-0685R-Cy5) and Cy3-conjugated anti-PKCδ antibody (1:200, Bioss, bs-23441R-Cy3) were mixed at a 1:1 donor/acceptor ratio and incubated with the samples for 2 h at room temperature. Cells were washed three times with PBS, and FRET efficiency was determined using the acceptor photobleaching method on a Zeiss LSM900 microscope.

Western blotting

BMDCs isolated from the indicated mice were stimulated with curdlan and lysed in Western & IP Buffer (Beyotime, P0013) for protein extraction. For histone modification assays, histones were purified using the Histone Extraction Kit (Bestbio, BB-3117). For phosphorylation detection, BMDC lysates were clarified by centrifugation (12,000 × g, 10 min, 4 °C), denatured in SDS loading buffer at 95 °C for 10 min, separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE), and transferred to polyvinylidene difluoride (PVDF) membranes. Membranes were blocked in Tris-buffered saline with Tween 20 (TBST) containing 5% non-fat milk for 1 h, incubated with primary antibodies overnight at 4 °C, washed with TBST, and then probed with horseradish peroxidase (HRP)-conjugated secondary antibodies for 1 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence (ECL) substrate (Abclonal, RM00021) on a Tanon 5200S imaging system, and densitometric analysis was performed using ImageJ software. The antibodies used in this process were as follows: H3K79me Rabbit mAb (CST, 12522, 1:50000), H3K79me2 Rabbit mAb (Abclonal, A22086, 1:50000), H3K79me3 Rabbit mAb (CST, 74073S, 1:50000), H3K79pan-me Rabbit mAb (PTM BIO, PTM-7435, 1:5000), H3K4pan-me Rabbit mAb (HUABIO, HA751067, 1:5000), H4K16ac Rabbit mAb (HUABIO, ET7017-89, 1:10000), H3K36me3 Rabbit mAb (Abclonal, A20379, 1:10000), Histone H2B Rabbit mAb (HUABIO, HA750287, 1:10000), IL-1 beta Mouse mAb (HUABIO, HA610004, 1:1000), H2B-Ub Rabbit mAb (CST, 5546 T, 1:10000), H3 Rabbit mAb (CST, 4620, 1:10000), RNF20 Rabbit mAb (CST, 9425, 1:1000), RNF40 Rabbit mAb (Abclonal, A9598, 1:2500), Syk Rabbit mAb (CST, 13198S, 1:1000), p-Syk Rabbit mAb (CST, 2710S, 1:1000), PKCδ Rabbit mAb (CST, 2058S, 1:1000), p-PKCδRabbit mAb (CST, 2055S, 1:1000), IκBα Mouse mAb (CST, 4814S, 1:1000), p-IκBα Mouse mAb (CST, 9246S, 1:1000), p-Ser antibody (Santa Cruz, sc-81514, 1:200), β-Actin Rabbit mAb (Abclonal, AC026, 1:5000), Lamin B1 Rabbit mAb (CST, 12586, 1:1000), GAPDH XP® Rabbit mAb (CST, 4283S, 1:1000), Myc-Tag Mouse Ab (Abmart, M20002, 1:5000), Flag-Tag Rabbit mAb (CST, 42544S, 1:1000), HA-Tag Rabbit mAb (CST, 3724S, 1:1000), C1QBP Rabbit mAb (CST, 6502S, 1:1000), Goat anti-mouse IgG-HRP (Abmart, M21001L, 1:10000), Goat anti-rabbit IgG-HRP (Abmart, M21002L, 1:10000), RACK1 Rabbit mAb (HUABIO, ET7109-04, 1:1000).

Co-immunoprecipitation

For ectopic interaction assays, HEK293T cells expressing C1QBP-HA, C1QBP-Flag, RNF20-HA, RNF40-HA, Syk-Flag, or PKCδ-Myc were lysed in Western & IP Buffer. For endogenous interaction assays, 1 × 10⁷ BMDCs were lysed in the same buffer. Lysates were incubated with antibody-conjugated magnetic beads or primary antibodies followed by Protein A/G beads (Thermo Fisher, 88803). Immunocomplexes were washed, eluted, and analyzed by Western blotting as described above. The IP antibodies used in this process were as follows: Mouse RNF20 antibody (IP 1 μg, CST, 9425), Mouse RNF40 antibody (IP 1 μg, Abclonal, A9598), Mouse PKCδ antibody (IP 2 μg, CST, 2058S), Mouse GC1qR antibody (IP 1 μg, Abcam, ab101267), IgG antibody (IP 1 μg, Abmart, M21002L), Anti-Flag antibody (IP 1 μg, CST, 14793S), Anti-HA antibody (IP 1 μg, CST, 3724S), Anti-Myc antibody (IP 1 μg, Abmart, M20002).

Quantitative real-time PCR

Total RNA was extracted from BMDCs using TRIzol reagent (Takara, 9109) and reverse-transcribed into complementary DNA (cDNA) using RT Master Mix (Abclonal, RK20433). qRT-PCR was performed using Power SYBR Green PCR Master Mix (Abclonal, RK21203) on a LightCycler 96 system (Roche). Primer sequences are listed in SupplementaryTable 1.

Chromatin Immunoprecipitation (ChIP)-qPCR

ChIP assays were performed using the SimpleChIP Enzymatic Chromatin IP Kit (Cell Signaling Technology, 9002/9003) according to the manufacturer’s instructions. BMDCs were stimulated with curdlan for 3 h, crosslinked with 1% formaldehyde, and lysed to isolate nuclei. Chromatin was digested with Micrococcal Nuclease to generate fragments of 150-900 bp and sonicated to disrupt nuclear membranes. Lysates were immunoprecipitated with antibodies specific for H3K79me1 (CST, 12522), H3K79me2 (Abclonal, A22086), H3K79me3 (CST, 74073S), or IgG (as a negative control). Immunoprecipitated DNA was purified via column elution, and enrichment at specific loci was quantified by qPCR using primers listed in Supplementary Table 1.

Enzyme-linked immunosorbent assay (ELISA)

Levels of mouse TNF-α (88-7324-77), IL-6 (88-7064-88), IL-1β (88-7013A-88), IL-10 (88-7105-77), IL-17A (88-7371-88), IL-17F (88-7472-77), IL-22 (88-7422-88), and IFN-γ (88-7314-88) in kidney homogenates or culture supernatants were quantified using ELISA kits (all from Invitrogen). All assays were performed strictly according to the manufacturers’ protocols.

Flow cytometry

Kidneys from C. albicans infected mice were aseptically harvested, digested with collagenase D (1 mg/mL, Roche, 11088866001) at 37 °C for 90 min, and filtered through a 70 μm cell mesh to obtain single-cell suspensions. Cells were stained with the following fluorophore-conjugated antibodies (1:400 dilution, unless otherwise specified; all from BioLegend except where noted): FITC-CD45 (103108), PerCP/Cy5.5-CD11b (101228), PE/Dazzle 594-F4/80 (123110), PE-Ly6C (128008), BV421-Ly6G (127628), APC-CD11c (117310), BV421-I-A/I-E (107639), FITC-CD3 (100203), BV421-CD4 (100438), PE-CD24 (101807), FITC-CD64 (161007), Bv605- I-A/I-E (107639), Bv421-CD11b (101235), FITC-BST2 (127008), Bv650-CD45R/B220 (103241) and eFluor 780 Fixable Viability Dye (Invitrogen, 65-0865). For intracellular cytokine staining, cells were stimulated with Cell Activation Cocktail containing brefeldin A (1:500, BioLegend, 423303) for 6 h, permeabilized using 10×Permeabilization Buffer (Invitrogen, 88-8824-00), and stained with the following fluorophore-conjugated antibodies (1:100 dilution, all from BioLegend except where noted): PE-594-IL-17A (506938), PerCP/Cy5.5-IL-22 (516411), PE-IL-17F (517008), BV650-IFN-γ (505832), V450-IL-1β (eBioscience, 48-7114-82) and Alexa Fluor 647-C1QBP (1:500, Abcam, Ab305041). Samples were analyzed on a FACSCelesta flow cytometer (BD Biosciences) using BD FACSDiva v8.0.1 software, and data were processed using FlowJo v10 software.

siRNA-mediated knockdown

Small interfering RNAs (siRNAs) targeting Rnf20, Rnf40, and PKCδ, as well as non-targeting control (NC) siRNAs, were synthesized by SynBio Co., Ltd. (Shanghai, China). siRNAs and Lipofectamine 3000 (Thermo Fisher Scientific, L3000015) were separately diluted in Opti-MEM medium (Gibco, 31985070), mixed, and incubated for 15 min at room temperature before being added to day 8 BMDC cultures. Knockdown efficiency was verified by Western blotting. Primer sequences are listed in Supplementary Table 1.

In vitro phosphorylation assay

HEK293T cells were transfected with expression plasmids encoding Rnf20, Rnf40, or Card9 using Lipofectamine 3000 according to the manufacturer’s protocol. Twenty-four hours post-transfection, cells were lysed in Western/IP Buffer supplemented with 100 μM phenylmethylsulfonyl fluoride (PMSF; Sangon Biotech) and incubated on ice for 10 min. Lysates were clarified by centrifugation (14,000 × g, 10 min, 4 °C), and supernatants were incubated with antibody-conjugated magnetic beads (MCE) for 1 h at 4 °C with gentle rotation. Beads were washed twice with RIPA buffer and twice with PBS, and the bound complexes were equally divided for in vitro phosphorylation assays using the PKCδ Kinase Enzyme System (Promega, V3401) according to the manufacturer’s instructions. Reactions were carried out in the presence of 50 μM ATP for 1 h and terminated by adding SDS loading buffer, followed by boiling at 95 °C for 10 min.

Isolation and activation of CD4⁺ T cell

Naïve CD4⁺ T cells were isolated from mouse spleens using the MojoSort™ Mouse CD4 Naïve T Cell Isolation Kit (BioLegend, 480040), achieving a purity of >95%. Isolated cells (3 × 10⁵) were activated in 96-well U-bottom plates coated with anti-CD3e antibody (5 μg/mL, BD Pharmingen, 553057) in the presence of soluble anti-CD28 antibody (2 μg/mL, BD Pharmingen, 553294) in complete RPMI medium containing 50 U/mL penicillin and 50 μg/mL streptomycin. Cells were co-cultured with BMDC supernatants (with or without curdlan stimulation) for 18 h and subjected to flow cytometric analysis after 5 days of culture.

Tandem mass spectrometry (MS)

BMDCs from C1qbp^fl/fl (WT), and C1qbp^fl/flCD11c^Cre/+ (KO) mice were stimulated with curdlan and lysed for protein extraction. Cell lysates were centrifuged at 12,000 × g for 10 min, and supernatants were incubated with anti-RNF20 primary antibody (CST, 9425) or anti-IgG antibody overnight, followed by incubation with Protein A/G beads. Protein-conjugated magnetic beads were washed three times with 0.5% TBST buffer. Bound proteins were eluted with 40 μL of 1× SDS loading buffer, separated by SDS-PAGE, and visualized by Coomassie Brilliant Blue staining.

Coomassie-stained gel bands were excised, diced, destained, dehydrated with acetonitrile, and air-dried. Gels were reduced with 10 mM DTT (56 °C, 60 min) and alkylated with 55 mM iodoacetamide (room temperature, dark, 45 min), washed, dried, and digested overnight at 37 °C with trypsin (10 ng μL⁻¹ in 25 mM NH₄HCO₃). Peptides were extracted by sonication, vacuum-dried, reconstituted in 0.1% TFA/0.5% acetonitrile, and desalted via reversed-phase solid-phase extraction before drying. Protein identification was performed by LC-MS using a Q-Exactive mass spectrometer (Thermo Scientific), with raw data processed via Proteome Discoverer. The numbers of matched peptides and identified proteins were 1671/399 (WT-unsti), 1598/359 (WT-sti), 1386/338 (KO-unsti), 1453/343 (KO-sti), and 1536/378 (IgG), respectively. Eluted protein samples from the three wells were pooled into one combined sample to achieve sufficient protein input for robust peptide identification. Number of biological replicates (n = 1). Peptides were matched against the UniProt Mus musculus protein database for identification.

Statistical analysis

Statistical analyses were performed in GraphPad Prism 8. Differences between two groups were assessed by unpaired two-sided Student’s t test; survival curves were compared using the log-rank test. Data are presented as mean ± SEM from at least three independent experiments, unless otherwise indicated. Representative images from immunofluorescence, PAS staining, and immunoblotting are shown.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

Hui Wang, Yebo Gu, and Xin-Ming Jia designed research. Hui Wang performed research; Yi-Heng Yang, Jia-Jie Tang, Xian-Long Wang, Xin-Yi Liu, Han Wang, Han-Yu Wang, Rong-Rong Wang, and Hui-Hui Ma contributed new reagents/analytic tools; Hui Wang, Yebo Gu, and Xin-Ming Jia analyzed data; and Hui Wang, YeBo Gu and Xin-Ming Jia drafted the manuscript.

Peer review

Peer review information

Nature Communications thanks Teresa Zelante, Mariolina Bruno, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Funding

This work was supported by the National Natural Science Foundation of China (82430072 and 82371777 to X.M.J and 82402071 to Y.G), National Key Research and Development Program of China (2023YFF0613602 to Y.G.)

Data availability

The data supporting the findings from this study are available within the manuscript and its supplementary information. The raw data and the processed data of bulk RNA-seq from C1QBP depleted BMDCs (KO) and control BMDCs (WT), have been deposited at the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) repository under accession numbers GSE313539 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE313539). The raw data of scRNA-seq from naïve (day 0) or C. albicans-infected wild-type (WT) mice (days 2 and 6 post-infection) have been deposited in the GEO database (GSE337405, https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE337405). The mass spectrometry raw data of RNF20-binding proteins from C1QBP depleted BMDCs (KO) and control BMDCs (WT) have been deposited to the ProteomeXchange Consortium via the PRIDE repository with the dataset identifier PXD079046. Source data are provided with this paper.  Source data are provided with this paper.

Code availability

This study does not include any original custom code. All single-cell transcriptome analyses were completed using standardized commercial bioinformatic pipelines from OE Biotech. All software packages, versions and full analytical parameters utilized are fully documented within the Methods section of this manuscript.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41467-026-76756-1.