Whole β-glucan particles modulate BCG-induced macrophage and CD4⁺ T-cell responses via the Dectin-1–JAK1/STAT1 pathway

Abstract

Tuberculosis (TB) remains a major global health burden, and Bacille Calmette–Guérin (BCG) provides inconsistent protection against adult pulmonary disease. Adjuvant strategies to enhance BCG efficacy are urgently needed. Here we show that whole β-glucan particles (WGP) enhance BCG-induced immune responses and improve protection against mycobacterial challenge. In vitro, WGP enhances macrophage functions associated with antigen processing during BCG exposure, including phagocytosis, lysosomal acidification, intracellular degradation, and the upregulation of MHC-II/CD80/CD86. WGP co-exposure partially attenuates the early IL-10 response induced by BCG. Mechanistically, whole β-glucan particles activate the Dectin-1–JAK1–STAT1 pathway, and disruption of this axis reduces MHC-II upregulation and impairs macrophage-supported OT-II CD4⁺ T-cell proliferation and IFN-γ production. In vivo, using female wild-type and Dectin-1 knockout C57BL/6 J mice, we find that split-site administration of WGP and BCG is better tolerated than same-site mixing and is associated with a less inflammatory early pulmonary myeloid profile. Under this optimized regimen, WGP plus BCG enhances CD4⁺ memory-associated responses and improves bacterial control while reducing lung inflammation in a Mycobacterium tuberculosis H37Ra challenge model. These findings support further evaluation of WGP as an adjunct candidate for BCG-based vaccination strategies and may inform the development of improved tuberculosis vaccines.

Cao et al. report that whole β-glucan particles (WGP) enhance BCG-conditioned macrophage antigen handling and presentation. WGP plus BCG strengthens CD4⁺ memory-associated responses and improves bacterial control while reducing lung inflammation

Introduction

Tuberculosis (TB) remains one of the leading infectious causes of death worldwide, highlighting the need for more effective vaccines^1,2. The Bacille Calmette–Guérin (BCG) vaccine confers protection against severe TB in children but exhibits variable efficacy against adult pulmonary disease^3,4. Accumulating evidence indicates that this limitation may arise, at least in part, from suboptimal early myeloid-cell responses, which shape antigen processing, T-cell priming, and the quality of subsequent adaptive immunity. Macrophage responses to BCG are heterogeneous and context dependent, potentially affecting antigen processing and T-cell priming^3,5. Following BCG revaccination, IL-10 levels increase early in the lung, lymph nodes, and spleen; macrophages and dendritic cells are the primary sources of IL-10, and this response attenuates CD4⁺ Th1-mediated protection^6–8. These findings suggest that early innate programming may be an important determinant of whether BCG exposure is translated into durable protective immunity.

One reason for this limitation is that BCG and related mycobacteria can impose early constraints on macrophage microbicidal and antigen-presenting functions. They delay phagosome–lysosome fusion and prevent phagosomal acidification, interfere with autophagy, and dampen antigen processing^9–11. In some contexts, they also induce IL-10, which suppresses Th1 priming⁶. Collectively, these effects may establish an early ceiling on BCG-elicited myeloid activation, thereby limiting the generation of robust T-cell responses and long-term memory. Because these events occur upstream of adaptive immune induction, strategies that reinforce macrophage activation and antigen presentation may provide a rational means to improve BCG efficacy.

β-Glucans, including whole β-glucan particles (WGP), can enhance antigen uptake and presentation and modulate T helper-cell responses^12–17. In tumor models, oral WGP-encapsulated OVA vaccines increase antigen-specific IgG and CD8⁺ T-cell cytotoxicity¹⁸; a β-glucan-particle/alum (GP-Al) system enhances responses to MUC1¹⁹; and glucan–chitin particle delivery of Coccidioides antigens reduces fungal burden via Dectin-1/2–CARD9-dependent Th1/Th17 immunity²⁰. β-Glucan–induced trained immunity also enhances protection against Trichinella spiralis²¹. Although β-glucans have been extensively studied in trained immunity and anti-mycobacterial host defense, WGP has not been systematically evaluated as an adjuvant in the specific BCG-based TB vaccination framework examined here. Given its particulate nature and immunomodulatory properties, WGP may be particularly suited to enhance macrophage phagocytic activity, intracellular processing, and antigen-presenting capacity during BCG exposure.

Multiple receptors and signaling pathways mediate β-glucan sensing in myeloid cells. Dectin-1 couples to Syk–CARD9 and drives NF-κB and MAPK signaling programs²²; it can also signal via Raf-1 to modulate non-canonical NF-κB activation and T-helper polarization²³. Dectin-1 cooperates with TLR2 to amplify cytokine production in response to particulate β-glucans²⁴. Complement receptor 3 (CR3) can bind β-glucans, particularly soluble forms, thereby modulating phagocyte activation²⁵. Reviews also highlight roles for scavenger receptors and lactosylceramide in β-glucan uptake^26,27. β-Glucans can engage and/or modulate the NLRP3 inflammasome, depending on context and timing²⁸. Among these pathways, Dectin-1 is regarded as a principal receptor for particulate β-glucans in myeloid cells, yet it remains unclear how Dectin-1-linked signaling remodels macrophage responses during BCG exposure and whether this process can improve downstream T-cell priming in vivo.

Here, we investigated whether engagement of a Dectin-1–JAK1–STAT1 signaling axis enhances macrophage phagocytosis, lysosomal acidification, and MHC-II expression, and whether these changes are associated with improved CD4⁺ T-cell proliferation, cytokine production, and memory formation, together with changes in selected CD8⁺ T-cell-associated responses. We further evaluated whether WGP augments BCG-induced immune responses under different in vivo administration strategies, with a particular focus on early macrophage infiltration and subsequent memory T-cell generation. Finally, using a Mycobacterium tuberculosis H37Ra challenge model, we assessed whether WGP can function as an effective adjunct to BCG and enhance protective immunity.

Results

WGP reshapes early macrophage activation programs during BCG exposure and partially counteracts BCG-induced IL-10

To examine how WGP affects macrophage responses during BCG exposure, we analyzed RAW264.7 cells and BMDMs under four conditions: PBS, WGP, BCG, and BCG + WGP. Here, BCG + WGP refers to co-exposure to BCG and WGP for 4 h, followed by removal of BCG and continued culture in the presence of WGP for the indicated times. At 6 h, WGP significantly increased the mRNA expression of iNOS, IL-1β, IL-6, IL-12a, MCP-1, CD80, and CD86 in both RAW264.7 cells and BMDMs compared with PBS (Fig. 1A, B). The BCG + WGP group also showed higher expression of most of these markers than the BCG group, although in many cases the levels did not exceed those induced by WGP alone. In contrast, IL-10 displayed a distinct pattern, with the highest level in the BCG group, the lowest level in the WGP group, and an intermediate level in the BCG + WGP group, suggesting that WGP partially attenuated the early IL-10 response induced by BCG (Fig. 1A, B). At 48 h, IFN-γ and TNF-α levels were higher in the WGP and BCG + WGP groups than in the PBS and BCG groups, whereas no significant difference was observed between the WGP and BCG + WGP groups (Fig. 1C). IL-10 protein levels showed a pattern similar to that of the transcript data, with the highest level in the BCG group and a marked reduction in the BCG + WGP group (Fig. 1C). Flow cytometry further showed that CD80 and CD86 expression on F4/80⁺ macrophages was increased in the WGP and BCG + WGP groups compared with PBS and BCG alone (Fig. 1D). CCK-8 analysis showed that WGP did not impair cell viability. Instead, WGP and BCG + WGP were associated with higher viability-related readouts than PBS or BCG alone at later time points (Fig. 1E, F).

Fig. 1. WGP reshapes macrophage activation under BCG exposure.

A, B RAW264.7 cells (A) and bone marrow-derived macrophages (BMDMs) (B) were treated with PBS (control), WGP alone, BCG alone, or BCG + WGP. Relative mRNA expression of cytokines and activation-associated molecules, including iNOS, IL-1β, IL-6, IL-12a, MCP-1, IL-10, CD80, and CD86, was quantified by RT-qPCR. Data are presented as fold change relative to the PBS group (n = 3 biologically independent samples per group). C IFN-γ, TNF-α, and IL-10 levels in BMDM supernatants were measured by ELISA (n = 3 biologically independent samples per group). D CD80 and CD86 expression on BMDMs was analyzed by flow cytometry; representative plots and MFI values are shown (n = 6 biologically independent samples per group). E, F BMDM proliferation/viability was assessed by CCK-8 assay at 0, 12, 24, and 36 h (n = 3 biologically independent samples per group). Data are shown as mean ± SD from three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns: no significance. One-way ANOVA with Tukey’s multiple-comparison test used for (A–E). Two-way ANOVA with Tukey’s multiple-comparison test used for (F).

Together, these findings indicate that WGP promotes an inflammatory and co-stimulatory macrophage phenotype during BCG exposure while partially attenuating the early IL-10 response induced by BCG.

WGP modulates macrophage uptake, intracellular processing, and antigen-presentation programs during BCG exposure

To assess macrophage function in a reductionist setting, we quantified phagocytosis, intracellular acidification, and MHC-II under four conditions: PBS, WGP, BCG, WGP + BCG. At 4 h, WGP increased the proportion of rhodamine-BCG⁺ BMDMs relative to PBS, and by 24 h the rhodamine signal had declined more prominently in the WGP group than in the PBS group, suggesting enhanced uptake followed by more efficient intracellular processing of internalized bacilli (Fig. 2A). Confocal microscopy was used to assess lysosomal acidification, a critical step in antigen processing and presentation. Lysosomal acidification showed a graded pattern (Fig. 2B): WGP > PBS (p < 0.0001), BCG + WGP > BCG (p < 0.0001), and BCG + WGP > WGP (p < 0.01), whereas BCG vs PBS was not significant (ns). These data indicate that WGP alone promotes lysosomal acidification and that co-treatment with BCG and WGP further enhances this response beyond the level induced by either treatment alone. For MHC-II transcripts, WGP exceeded both PBS and BCG in RAW264.7 cells and BMDMs (p < 0.0001), whereas BCG + WGP was also significantly higher than BCG (p < 0.0001) (Fig. 2C). Immunoblotting confirmed increased MHC-II protein expression in the WGP and BCG + WGP groups relative to the PBS and BCG groups (Fig. 2D). Flow-cytometric analysis of surface MHC-II on F4/80⁺ cells (Fig. 2E) similarly showed WGP > PBS (p < 0.0001), BCG + WGP > BCG (p < 0.0001), and BCG + WGP > WGP (p < 0.0001). Together, these data show that WGP enhances macrophage antigen-handling features, including uptake, lysosomal acidification, and MHC-II expression, during BCG exposure. Although BCG + WGP consistently improved these parameters relative to BCG alone, it did not surpass WGP alone with respect to MHC-II expression, supporting a non-additive pattern consistent with convergence on shared downstream pathways rather than broad in vitro synergy.

Fig. 2. WGP enhances macrophage phagocytosis-associated uptake, lysosomal acidification, and MHC-II expression during BCG exposure.

A Uptake and processing of rhodamine-labeled BCG by BMDMs at 4 h and 24 h (n = 3 biologically independent samples per group). B Lysosomal acidification in BMDMs measured using LysoSensor Green. Values represent means for individual biological replicates (n = 3 biologically independent samples per group). C RT-qPCR analysis of H2-Ab1 (MHC-II) expression in RAW264.7 cells and BMDMs (n = 3 biologically independent samples per group). D Immunoblot analysis of MHC-II protein expression, with GAPDH as a loading control. E Surface MHC-II expression on F4/80⁺ macrophages, shown as MFI (n = 6 biologically independent samples per group). Data are presented as mean ± SD from at least three independent experiments. **p < 0.01; ****p < 0.0001; ns: no significance. Statistical analysis was performed using two-way ANOVA with Tukey’s multiple-comparison test at each time point for A. One-way ANOVA with Tukey’s multiple-comparison test was used for (B), (C), and (E); pre-specified comparisons included WGP vs. BCG + WGP and BCG vs. BCG + WGP.

Having found that WGP enhances macrophage uptake, intracellular processing, and MHC-II expression during BCG exposure, we next asked whether these macrophage changes were associated with stronger T-cell responses in co-culture. To distinguish macrophage-dependent effects from any potential direct action of WGP on T cells, we first examined Dectin-1 expression in splenic T-cell subsets and then tested the effect of WGP on purified T cells in the absence of macrophages. Flow-cytometric analysis showed that Dectin-1 expression was virtually undetectable in splenic CD4⁺ T cells and detectable only in a minor fraction of CD8⁺ T cells, whereas markedly stronger staining was observed in CD11b⁺ myeloid cells (Supplementary Fig. 1A, B). In parallel, purified CD4⁺ and CD8⁺ T cells exposed to WGP alone did not show detectable increases in CFSE dilution or IFN-γ production compared with the PBS control. By contrast, co-culture with BMDMs increased T-cell responses, and these effects were further enhanced when the BMDMs had been treated with WGP (Supplementary Fig. 1C, D). These data indicate that the augmented T-cell responses observed in our system are not explained by a major direct stimulatory effect of WGP on T cells, but instead depend predominantly on APC-mediated activity.

We then asked whether the macrophage antigen-presentation program enhanced by WGP extended beyond the MHC-II axis. In BMDMs, WGP increased MHC-I mRNA and protein expression, and flow-cytometric analysis confirmed higher surface MHC-I levels, although the magnitude of this increase was less pronounced than that observed for MHC-II (Supplementary Fig. 2A–C). Consistent with this pattern, WGP-treated BMDMs promoted not only CD4⁺ T-cell proliferation and IFN-γ production but also a measurable increase in CD8⁺ T-cell activation, while BCG-treated BMDMs showed weaker stimulatory activity than either the WGP or WGP + BCG groups. These findings indicate that WGP broadens macrophage antigen-presenting capacity while preserving macrophage-associated MHC-II regulation and CD4⁺ T-cell activation as the principal mechanistic axis of the study (Supplementary Fig. 2D–F).

Although macrophages remained the central focus of this work, we also examined dendritic cells because both macrophages and dendritic cells function as professional antigen-presenting cells. We therefore asked whether the immunostimulatory effects of WGP extended beyond macrophages to another major APC population. In BMDCs, WGP increased the expression of inflammatory and antigen-presentation-associated genes (Supplementary Fig. 3A), enhanced IFN-γ and TNF-α production (Supplementary Fig. 3B, C), and upregulated CD80, CD86, MHC-I, and MHC-II surface expression on CD11c⁺ cells (Supplementary Fig. 3D). WGP-treated BMDCs also promoted stronger APC-dependent T-cell responses in co-culture assays (Supplementary Fig. 3E, F). These findings indicate that the effects of WGP are not restricted to macrophages, although the principal mechanistic framework of the present study remains focused on macrophage-associated antigen-presentation programs during BCG exposure.

Taken together, these findings show that WGP enhances macrophage uptake, intracellular processing, and antigen-presentation programs during BCG exposure. The supplementary analyses further indicate that WGP alone does not directly induce detectable T-cell activation in the absence of macrophages, that WGP-treated BMDMs support both CD4⁺ and, to a lesser extent, CD8⁺ T-cell responses, and that BMDCs display a similar but complementary pattern of activation. Collectively, these in vitro data define early antigen-processing and antigen-presentation events induced by WGP during BCG exposure, rather than overall vaccine efficacy.

WGP activates Dectin-1–JAK1–STAT1 signaling to enhance macrophage MHC-II expression and macrophage-dependent CD4⁺ T-cell activation

We next investigated the upstream receptor and signaling pathway that mediate the effects of WGP in a reductionist setting. To define receptor usage and downstream signaling, macrophages were stimulated with WGP alone. In both RAW264.7 cells and primary BMDMs, immunoblot analysis showed that WGP rapidly induced phosphorylation of JAK1 and STAT1 at 1 h and increased MHC-II expression at 24 h (Fig. 3A, B). Subcellular fractionation further showed that WGP promoted STAT1 nuclear translocation, whereas JAK1 remained in the cytosolic fraction (Fig. 3C). To determine whether this pathway depends on Dectin-1, we examined BMDMs from Dectin-1-KO mice. Compared with WT cells, Dectin-1-deficient BMDMs showed markedly reduced JAK1 and STAT1 phosphorylation and blunted MHC-II induction after WGP stimulation (Fig. 3D), indicating that Dectin-1 acts upstream of this signaling axis. Consistently, pharmacologic inhibition of JAK1 with abrocitinib or of STAT1 with fludarabine reduced WGP-induced JAK1/STAT1 activation at 1 h and diminished MHC-II expression at 24 h (Fig. 3E, F).

Fig. 3. WGP upregulates macrophage MHC-II expression through the Dectin-1–JAK1–STAT1 pathway and promotes CD4⁺ T-cell activation.

A, B Immunoblot analysis of RAW264.7 cells (A) and primary BMDMs (B) stimulated with WGP for 1 h or 24 h. p-JAK1 and p-STAT1 were assessed at 1 h, and MHC-II at 24 h. Densitometric values above bands are normalized to GAPDH and expressed relative to PBS. C Cytoplasmic and nuclear fractionation after WGP stimulation showing STAT1 nuclear accumulation, with JAK1 remaining predominantly cytoplasmic. GAPDH and Lamin A/C were used as fraction controls. D WT and Dectin-1-KO BMDMs stimulated with WGP, showing loss of p-JAK1/p-STAT1 induction and reduced MHC-II expression in Dectin-1-deficient cells. E, F BMDMs were pretreated with abrocitinib (E) or fludarabine (F) for 2 h, washed, and stimulated with WGP. Inhibitor treatment reduced p-JAK1/p-STAT1 at 1 h and MHC-II expression at 24 h compared with DMSO controls. G OT-II CD4⁺ T cells were co-cultured with WGP-conditioned BMDMs with or without STAT1 inhibition. T-cell proliferation and IFN-γ production were assessed by CFSE dilution and intracellular staining, respectively (n = 3 biologically independent samples per group). Data are presented as mean ± SD from at least three independent experiments. ****p < 0.0001. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple-comparison test at each time point for (G).

We next asked whether this WGP-induced macrophage signaling pathway was functionally linked to macrophage-supported CD4⁺ T-cell activation. To directly assess antigen-presentation capacity under controlled conditions, we used OT-II CD4⁺ T cells, which recognize OVA323–339, thereby allowing a standardized readout of macrophage-dependent T-cell activation. In this reductionist co-culture system, WGP-treated macrophages significantly increased CD4⁺ T-cell proliferation and IFN-γ production (Fig. 3G). Importantly, this effect was markedly attenuated when STAT1 was inhibited in macrophages, indicating that the enhanced T-cell stimulatory activity induced by WGP depends on STAT1 signaling (Fig. 3G).

Taken together, these data show that WGP activates a Dectin-1–JAK1–STAT1 signaling cascade in macrophages, leading to increased MHC-II expression and enhanced macrophage-dependent CD4⁺ T-cell activation in vitro.

Split-site administration of WGP with BCG optimizes early pulmonary myeloid responses and is associated with enhanced CD4⁺ memory T-cell immunity

We then asked whether the macrophage-conditioning effects of WGP observed in vitro could translate into a more favorable in vivo immunization profile when combined with BCG. Because the route and spatial configuration of administration could influence both tolerability and early myeloid remodeling, we first performed an exploratory comparison of different WGP-plus-BCG delivery strategies (Supplementary Fig. 4). In this pilot setting, same-site co-administration of BCG and WGP elicited a stronger early inflammatory reaction, whereas split-site administration was better tolerated. Flow-cytometric analysis of the lung showed that same-site mixed administration was associated with prominent early accumulation of CD45⁺CD11b⁺Ly6G⁺ neutrophils, particularly on day 1, consistent with acute inflammatory recruitment (Supplementary Fig. 4B). By contrast, split-site administration preserved or increased the proportion of CD45⁺Ly6C⁻CD11b⁺F4/80⁺CD64⁺ cells, especially by day 3, suggesting a more favorable pulmonary myeloid profile when the two stimuli were delivered separately (Supplementary Fig. 4C). Histological examination further showed that same-site mixed administration caused more pronounced early inflammatory changes in the lung, including inflammatory-cell infiltration and hemorrhagic changes on day 1, whereas these differences became less evident by day 3 (Supplementary Fig. 4D). Together, these exploratory findings identified split-site administration as the more suitable regimen for subsequent in vivo analyses.

Based on these results, we evaluated the immunological consequences of the split-site regimen in the formal experiment (Fig. 4A). On day 10, flow-cytometric analysis of the lung showed that BCG alone modestly increased Ly6G⁺ cells relative to PBS, whereas the same-site BCG + WGP regimen induced a further increase in this population (Fig. 4B). At the same time, analysis of CD45⁺Ly6C⁻CD11b⁺F4/80⁺CD64⁺ cells demonstrated that WGP or BCG alone each increased this macrophage-enriched population relative to PBS, whereas the split-site BCG + WGP group showed the highest level (Fig. 4C). These data indicate that under the selected regimen, WGP does not simply amplify nonspecific inflammation, but is instead associated with a pulmonary myeloid composition characterized by enhanced accumulation of macrophage-like cells after BCG immunization.

Fig. 4. Split-site administration of WGP and BCG promotes early pulmonary myeloid remodeling and is associated with enhanced CD4⁺ memory T-cell and cytokine responses.

A Experimental scheme. C57BL/6 mice received PBS, WGP alone, BCG alone, a same-site mixture of BCG + WGP, or split-site administration of BCG and WGP, followed by intranasal challenge with M.tb H37Ra. Lung and spleen tissues were collected 28 days later for CFU analysis, and lungs were further processed for acid-fast bacillus staining and histopathological evaluation. Created in https://BioRender.com. B Representative flow-cytometry plots and quantification of lung CD45⁺ CD11b⁺Ly6G⁺ cells on day 10 (n = 6 biologically independent samples per group). C Representative flow-cytometry plots and quantification of lung CD45⁺Ly6C⁻CD11b⁺F4/80⁺CD64⁺ cells on day 10 (n = 6 biologically independent samples per group). D Representative flow-cytometry plots and quantification of central-memory and effector-memory CD4⁺ T cells in the lung and spleen on day 30 (n = 6 biologically independent samples per group). E Representative intracellular cytokine staining plots and quantification of IL-2⁺ and IFN-γ⁺ CD4⁺ T cells in the lung and spleen after ex vivo restimulation on day 30 (n = 6 biologically independent samples per group). Data are presented as mean ± SD from at least three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple-comparison test at each time point for (B), (C), and (E). Two-way ANOVA with Tukey’s multiple-comparison test for (D).

On day 30, analysis of CD4⁺ T cells showed that the split-site WGP + BCG regimen increased the proportions of central memory and effector memory CD4⁺ T cells in both lung and spleen relative to PBS and to each single-treatment group (Fig. 4D). WGP alone yielded intermediate values in several comparisons, whereas the combined regimen consistently showed the strongest enhancement, particularly in the effector-memory compartment. After ex vivo restimulation, the frequencies of IL-2⁺ and IFN-γ⁺ CD4⁺ T cells were also higher in the split-site WGP + BCG group than in the BCG-alone or WGP-alone groups in both tissues (Fig. 4E). Thus, the regimen selected on the basis of early myeloid remodeling was also associated with stronger downstream CD4⁺ memory formation and functional recall responses.

Because this regimen appeared to establish a more favorable antigen-presenting environment in vivo and was associated with enhanced CD4⁺ memory formation, we next examined whether this benefit was also reflected in the CD8⁺ memory compartment. The gating strategy used to identify lung and splenic CD8⁺ memory T cells is shown in Supplementary Fig. 5A, C. In the lung, flow-cytometric analysis showed that the split-site WGP + BCG regimen significantly increased effector memory CD8⁺ T cells relative to PBS and BCG alone, whereas no significant difference was observed in central memory CD8⁺ T cells (Supplementary Fig. 5B). A similar pattern was observed in the spleen, where the split-site regimen also increased the proportions of central memory and effector memory CD8⁺ T cells compared with the single-treatment groups, again with a more prominent enhancement in the effector-memory compartment (Supplementary Fig. 5D). These findings indicate that the immunological benefit associated with the split-site regimen was not restricted to the CD4⁺ compartment, but also extended to CD8⁺ memory T-cell formation in both lung and spleen.

Taken together, these data show that the spatial configuration of WGP and BCG administration critically influences the early pulmonary response, and that split-site administration avoids the exaggerated early inflammatory pattern observed with same-site mixing while favoring a myeloid environment associated with stronger memory T-cell formation and cytokine production. These in vivo findings are therefore consistent with the in vitro observation that WGP conditions antigen-presenting cells toward a more immunostimulatory state, while also indicating that the mode of combination with BCG is an important determinant of the final immune outcome.

Macrophage Dectin-1 is required for CD4⁺ memory T-cell function in a depletion–reconstitution model under the WGP + BCG regimen

Having established that the split-site WGP + BCG regimen enhances CD4⁺ memory T-cell responses in vivo, we next asked whether this effect depends on Dectin-1 signaling, and more specifically whether macrophage-expressed Dectin-1 is required for this enhancement. To define the role of Dectin-1 in WGP-mediated enhancement of CD4⁺ memory T-cell responses, WT and Dectin-1 knockout (Dectin-1 KO) mice were immunized with BCG plus WGP. Flow cytometric analysis on day 30 post-immunization revealed significantly higher proportions of central memory (T_CM) and effector memory (T_EM) CD4⁺ cells in the lungs and spleens of WT mice than in Dectin-1 KO mice (Supplementary Fig. 6A, B). Moreover, cytokine production by CD4⁺ T cells, including IL-2 and IFN-γ, was markedly reduced in Dectin-1 KO mice (Supplementary Fig. 6C, D), supporting a requirement for Dectin-1 in the enhancement of CD4⁺ memory-associated responses under the WGP + BCG regimen.

To determine whether macrophage-expressed Dectin-1 is specifically required for the enhancement of BCG-induced CD4⁺ memory T-cell responses in vivo, we performed an in vivo macrophage depletion assay. WT mice received intravenous (i.v.) injections of either control liposomes or clodronate liposomes to deplete macrophages. Flow-cytometric analysis 48 h after injection confirmed efficient depletion, as the proportion of CD11b⁺F4/80⁺ macrophages decreased from 11.9% to 3.42% in the lung and from 3.57% to 0.45% in the spleen of clodronate-treated mice relative to controls (Supplementary Fig. 7A, B). After confirming depletion efficiency, WT mice were reconstituted on day −1 with either WT or Dectin-1 KO BMDMs following clodronate-liposome pretreatment on day −2. The mice were then immunized with subcutaneous BCG plus intraperitoneal WGP under the established split-site regimen (Fig. 5A; WGP i.p., BCG s.c.).

Fig. 5. Macrophage-expressed Dectin-1 contributes critically to WGP-associated CD4⁺ memory T-cell responses in a depletion–reconstitution model.

A Schematic of the macrophage depletion–reconstitution experiment. WT mice received intravenous clodronate liposomes to deplete macrophages. Forty-eight hours later, mice were reconstituted intravenously with WT or Dectin-1-KO BMDMs, followed by split-site immunization with subcutaneous BCG and intraperitoneal WGP. Created in https://BioRender.com. B, C Thirty days after immunization, flow cytometry was used to assess the proportions of central-memory T cells (T_CM, CD44^hiCD62L^hi) and effector-memory T cells (T_EM, CD44^hiCD62L^low) within the CD4⁺ T-cell population in the lung (B) and spleen (C). Representative plots and summary statistics are shown (n = 3 biologically independent samples per group). D, E Representative flow plots and quantification of IL-2⁺ and IFN-γ⁺ cells within the CD4⁺ T-cell population in the lung (D) and spleen (E). Lung cells were stimulated ex vivo with heat-inactivated BCG for 4–6 h, whereas spleen cells were stimulated for 18–24 h (n = 3 biologically independent samples per group). Data are presented as mean ± SD from at least three independent experiments. ***p < 0.001; ****p < 0.0001. Statistical analysis was performed using two-way ANOVA with Tukey’s multiple-comparison test at each time point for (B) and (C). One-way ANOVA with Tukey’s multiple-comparison test for (D) and (E).

Flow cytometric analysis at day 30 demonstrated significantly higher proportions of CD4⁺ T_CM and T_EM cells in both lungs and spleens of mice reconstituted with WT BMDMs compared to those receiving Dectin-1 KO BMDMs. Specifically, in the lung, T_CM decreased from 14.2% to 4.89% and T_EM from 70.5% to 55.4%, whereas in the spleen, T_CM declined from 7.61% to 4.40% and T_EM from 40.9% to 27.2% (Fig. 5B, C). Concordantly, the frequencies of IL-2⁺ and IFN-γ⁺ CD4⁺ T cells were markedly reduced in both organs after reconstitution with Dectin-1 KO BMDMs (Fig. 5D, E). The gating strategies used to identify lung and splenic T_CM/T_EM and IL-2⁺ and IFN-γ⁺ CD4⁺ T cells are shown in Supplementary Fig. 7C, D.

Together, these data support a macrophage-intrinsic requirement for Dectin-1 in the enhancement of CD4⁺ memory T-cell responses under the WGP + BCG regimen, while not excluding additional contributions from other Dectin-1-expressing myeloid cells.

WGP enhances the protective efficacy of BCG in an M.tb H37Ra challenge model

To determine whether the immunological changes induced by WGP and BCG translated into improved protection against mycobacterial challenge, we next used an M.tb H37Ra intranasal challenge model within the same immunization framework established above. Because the preceding experiments showed that the spatial configuration of WGP and BCG administration critically influenced early myeloid remodeling and subsequent memory T-cell responses, the challenge experiment included not only PBS, WGP alone, and BCG alone, but also both same-site BCG + WGP mixture and split-site BCG + WGP administration groups (Fig. 6A).

Fig. 6. Split-site administration of WGP with BCG enhances protection against M.tb H37Ra challenge and reduces pulmonary inflammation.

A Experimental scheme. C57BL/6 mice received PBS, WGP alone, BCG alone, a same-site mixture of BCG + WGP, or split-site administration of BCG and WGP, followed by intranasal challenge with M.tb H37Ra. Lungs were collected 28 days later for colony-forming unit (CFU) analysis, acid-fast bacillus staining, and histopathological evaluation. Created in https://BioRender.com. B, C Bacterial burden in the lung (B) and spleen (C), expressed as CFU per gram of tissue, at 28 days after M.tb H37Ra challenge (n = 6 biologically independent samples per group). D Representative acid-fast bacillus staining of lung sections from the indicated groups. E Representative hematoxylin and eosin (H&E) staining of lung sections from the indicated groups; boxed regions are shown at higher magnification below. F Quantification of lung inflammation scores in the indicated groups (n = 3 biologically independent samples per group). Data are shown as mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant. Statistical significance was determined by one-way ANOVA with appropriate multiple-comparison testing for (B), (C), and (F).

Twenty-eight days after challenge, bacterial burden analysis showed that the split-site BCG + WGP group exhibited the lowest CFU counts in both the lung and spleen, whereas PBS controls showed the highest burdens (Fig. 6B, C). BCG alone reduced bacterial load relative to PBS, whereas the same-site BCG + WGP mixture showed only partial improvement and remained less effective than the split-site regimen. WGP alone did not provide protection comparable to that of the BCG-containing regimens, supporting its role as an immunomodulatory adjunct rather than a standalone protective intervention in this model (Fig. 6B, C).

Acid-fast bacillus staining further supported these findings, with fewer detectable bacilli in the lungs of mice receiving split-site BCG + WGP than in those receiving BCG alone or the same-site mixture (Fig. 6D). Histopathological examination of lung sections by H&E staining showed that PBS-treated mice displayed the most pronounced inflammatory-cell infiltration and tissue disruption, whereas the split-site BCG + WGP group exhibited better-preserved alveolar architecture and less dense inflammatory infiltration than the BCG-alone and same-site mixture groups (Fig. 6E). Consistent with these observations, quantitative inflammation scores were lowest in the split-site BCG + WGP group among the BCG-containing regimens (Fig. 6F).

Taken together, these data indicate that the optimized split-site WGP + BCG regimen not only strengthens immunological memory but also translates this effect into improved protection against M.tb H37Ra challenge, as reflected by lower bacterial burden and reduced pulmonary inflammation.

Discussion

Our study shows that WGP reshape early macrophage responses during BCG exposure and, under an optimized in vivo administration strategy, are associated with enhanced downstream T-cell immunity. In vitro, WGP enhanced macrophage features related to antigen handling and presentation, including uptake-associated readouts, lysosomal acidification, co-stimulatory molecule expression, and MHC-II upregulation, while partially attenuating the early IL-10 response induced by BCG. Mechanistically, these effects were linked to a Dectin-1–JAK1–STAT1 pathway: WGP rapidly induced JAK1 and STAT1 phosphorylation, promoted STAT1 nuclear translocation, and was followed by increased MHC-II expression. Genetic deletion of Dectin-1, as well as pharmacologic inhibition of JAK1 or STAT1, blunted these responses and reduced macrophage-supported CD4⁺ T-cell proliferation and IFN-γ production in co-culture. Together, these findings place Dectin-1–JAK1–STAT1 upstream of the macrophage functional changes associated with the adjuvant-like activity of WGP in the BCG context.

An important related question is whether the enhanced T-cell responses observed here might reflect a direct effect of WGP on T cells rather than indirect regulation through antigen-presenting cells. Previous studies have suggested that Dectin-1 may be detectable in certain T-cell subsets under specific conditions, whereas direct activation of purified T cells by β-glucans in vitro remains uncertain^29,30. In our system, Dectin-1 expression was minimal in splenic T cells, and WGP alone did not induce detectable T-cell proliferation or IFN-γ production in the absence of macrophages. These findings support the interpretation that the enhanced T-cell responses observed here are predominantly macrophage dependent under our experimental conditions. This APC-conditioning effect was not restricted to macrophages. WGP increased MHC-I expression in BMDMs, modestly extended macrophage-supported responses toward the CD8 compartment, and also enhanced activation- and antigen-presentation-associated features in BMDCs, including increased CD80, CD86, MHC-I, and MHC-II expression together with improved T-cell stimulatory activity. These data suggest that WGP can act on more than one professional myeloid APC population. Even so, the present study remains centered on macrophages because the core mechanistic evidence, including phagocytic processing, lysosomal acidification, Dectin-1-dependent JAK1–STAT1 activation, and the depletion–reconstitution experiments, directly defines a macrophage-intrinsic pathway through which WGP reshapes antigen-presentation programs during BCG exposure.

The in vivo findings further indicate that the manner in which WGP is combined with BCG is a major determinant of the final immune outcome. Same-site mixed administration produced a stronger early pulmonary inflammatory pattern, characterized by increased Ly6G⁺ cell accumulation and more evident inflammatory changes in the lung, whereas split-site administration was associated with a more favorable pulmonary myeloid profile, including higher proportions of CD45⁺Ly6C⁻CD11b⁺F4/80⁺CD64⁺ cells and increased expression of CD80, CD86, MHC-I, and MHC-II. These observations suggest that WGP does not simply amplify inflammation when combined with BCG. Rather, under the current dose and schedule, administration geometry appears to determine whether early innate activation is translated into disruptive inflammatory perturbation or into a myeloid environment more permissive for antigen presentation and downstream adaptive priming. Although split-site delivery was superior in the present study, this does not necessarily mean that same-site administration is intrinsically ineffective, because optimization of dose, BCG-to-WGP ratio, timing, or formulation characteristics may still improve the local balance between beneficial innate activation and excessive inflammation.

Under the optimized split-site regimen, WGP plus BCG was associated with stronger CD4⁺ central-memory and effector-memory responses in both lung and spleen, together with increased IL-2 and IFN-γ production after ex vivo restimulation. We also observed increases in CD8⁺ memory-associated populations. Although these CD8⁺ findings were not mechanistically dissected in the present study, they raise the possibility that WGP-induced APC remodeling may also influence MHC-I-associated antigen presentation, potentially including cross-presentation-related processes. However, the dominant functional pattern observed here remained centered on enhancement of the MHC-II–CD4 axis. WGP alone showed limited activity relative to BCG-containing regimens, supporting the view that in this setting WGP acts primarily as an adjunct that modifies the quality of the immune response to BCG rather than as a standalone protective intervention.

A trained-immunity framework may also help explain part of the present findings. β-glucans are well-established inducers of trained innate responses, and previous studies have shown that they can reprogram monocytes, macrophages, and hematopoietic progenitors toward heightened secondary responsiveness³¹. Although we did not directly test classical trained-immunity endpoints such as long-term epigenetic remodeling, metabolic rewiring, or hematopoietic reprogramming, the enhanced myeloid responsiveness and the route-dependent improvement in downstream T-cell responses observed here are compatible with the possibility that trained-immunity-like conditioning contributes to the in vivo activity of WGP. This interpretation remains suggestive and will require direct testing in future work.

Our data also fit within the broader literature supporting β-glucan particles as immunomodulatory platforms. In tumor and infectious disease models, β-glucan-based systems can enhance antigen uptake, co-stimulatory signaling, and adaptive immune activation^12,32–34. The present study extends this concept to the BCG setting and suggests that the principal value of WGP lies not in indiscriminate amplification of all immune outputs, but in selective improvement of APC conditioning and the downstream T-cell environment. This view is also consistent with our in vitro observation that WGP plus BCG outperformed BCG alone, yet generally did not exceed WGP alone across several macrophage readouts. Rather than indicating broad synergy across all outputs, this pattern is more consistent with convergence on shared downstream processes, including lysosomal acidification, MHC-II expression, and early regulatory control involving IL-10.

The depletion–reconstitution experiment provides an important bridge between the in vitro signaling data and the in vivo immune phenotype. Replacement of host macrophages with Dectin-1-deficient BMDMs reduced day-30 CD4⁺ memory-associated populations and decreased IL-2 and IFN-γ responses under the WGP plus BCG regimen. These data support a macrophage-intrinsic requirement for Dectin-1 in the enhancement of memory-associated immunity and strengthen the interpretation that macrophages are mechanistically important participants rather than simple correlates. At the same time, because BMDCs also responded to WGP in a parallel direction, these findings are best interpreted as defining macrophages as a necessary mechanistic component in the present model, rather than excluding additional contributions from dendritic cells or other Dectin-1-expressing myeloid populations in vivo.

The M.tb H37Ra challenge experiment provided functional support for these immunological findings. Under the optimized split-site regimen, BCG plus WGP yielded the lowest bacterial burdens in both lung and spleen among the BCG-containing groups, whereas same-site mixed administration did not reproduce this degree of benefit. Acid-fast bacillus staining and H&E analysis likewise showed fewer bacilli and reduced pulmonary inflammation in the split-site BCG plus WGP group. These findings support the view that the immunological advantages associated with WGP-mediated APC conditioning and improved administration geometry can translate into improved bacterial control in vivo. At the same time, the H37Ra model differs from virulent M. tuberculosis in both virulence and immunopathology, and the present results should therefore be interpreted as evidence of improved control in a surrogate challenge system rather than as definitive proof of protection against fully virulent tuberculosis.

Several limitations should be acknowledged. Several limitations should be acknowledged. First, although the revised study now includes WGP-only and same-site mixture groups in the relevant in vivo experiments, the overall design still emphasizes WGP as an adjunct to BCG rather than as a complete standalone immunization strategy. Second, antigen specificity was inferred mainly from recall stimulation and functional readouts rather than directly defined by peptide–MHC tetramer approaches or activation-induced marker assays. Third, although a trained-immunity framework is biologically plausible, we did not directly assess epigenetic, metabolic, or hematopoietic reprogramming. Fourth, although the revised study includes BMDC experiments showing that WGP can also enhance dendritic-cell-associated antigen-presentation features, the relative in vivo contribution of dendritic cells versus macrophages to the improved adaptive response was not directly resolved and will require future lineage-specific approaches. Fifth, the challenge experiment used H37Ra at an intermediate time point after immunization and therefore does not substitute for later challenge with virulent M. tuberculosis. Future work should therefore focus on defining antigen specificity more precisely, testing trained-immunity-associated changes in myeloid cells, further resolving the in vivo contributions of different APC populations, optimizing dose and route parameters, and evaluating protective efficacy in more stringent virulent M. tuberculosis models.

In summary, WGP engages Dectin-1–JAK1–STAT1 signaling, enhances macrophage markers of antigen handling and antigen presentation, and partially dampens early BCG-induced IL-10 in vitro. When combined with BCG using an optimized split-site regimen, these mechanistic changes are associated with favorable early myeloid remodeling, enhanced CD4⁺ memory-associated responses, supportive increases in CD8⁺ memory responses, and improved control of M.tb H37Ra with reduced lung inflammation. These findings support continued evaluation of WGP as an adjunct candidate for BCG-based vaccination strategies and further suggest that both innate-cell conditioning and administration geometry are likely to shape its ultimate benefit.

Methods

Bacterial strains and reagents

Mycobacterium bovis BCG Pasteur (ATCC 35734) and Mycobacterium tuberculosis H37Ra (ATCC 25177) were maintained on Middlebrook 7H10 agar (supplemented with OADC) and in Middlebrook 7H9 broth (OADC, 0.05% Tween-80, 0.2% glycerol) at 37 °C with shaking (150 rpm). Mid-logarithmic cultures were pelleted (3000 × g, 10 min), washed 3× with sterile PBS, and resuspended in PBS. Bacterial density was adjusted to the desired inoculum by OD600 and confirmed by back-titration on 7H10 agar. Whole β-glucan particles (WGP) were purchased from Invivogen (France, Cat. No. tlrl-wgp).

Cell lines and animals

Mouse macrophage cell line RAW264.7 was cultured in DMEM medium (Gibco, NY, USA) containing 10% fetal bovine serum (Lonser, Shanghai, China), 100 U/mL penicillin, and 100 μg/mL streptomycin (Beyotime, Shanghai, China). Primary bone marrow-derived macrophages (BMDMs) were isolated from femurs and tibias of 6–8-week-old female wild-type (WT) and Dectin-1-knockout C57BL/6 J mice (GemPharmatech, Shanghai, China). OT-I and OT-II transgenic mice were purchased from Cyagen Biosciences (Suzhou, China).

All animal experiments were approved by the Ethics Committee of Center for Animal Experiments at Wuhan University (Approval No. WP20240336). We have complied with all relevant ethical regulations for animal use.

Mice were maintained under specific-pathogen-free conditions in the Animal Facility of the Center for Animal Experiments at Wuhan University, with a 12-h light-dark cycle. For animal experiments, mice were briefly anesthetized with isoflurane (3% for 1–2 min) to facilitate all injection. Mice recovered within 2–3 min and were returned to their home cage. At experimental endpoints, mice were euthanized by CO₂ inhalation followed by cervical dislocation as a secondary method to ensure death. CO₂ was delivered at a flow rate of 2–3 L/min (30–70% chamber volume per minute) using compressed CO₂ from a cylinder with a pressure regulator. Mice were placed in a clear chamber pre-filled with room air, and CO₂ was gradually introduced. Mice were exposed to CO₂ for at least 5 min after cessation of breathing. Death was confirmed by absence of heartbeat, lack of respiratory effort, and loss of corneal reflex. Cervical dislocation was then performed to ensure death before tissue collection. There is no criteria used for including and excluding animals and no exclusions.

Isolation and culture of BMDMs

Bone marrow cells were isolated from the femurs and tibias of mice by flushing with sterile medium. After erythrocyte removal with ACK lysis buffer, the cells were used for the generation of either bone marrow-derived macrophages or bone marrow-derived dendritic cells.

For BMDMs generation, bone marrow cells were cultured in DMEM supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin, and 25 ng/mL recombinant murine macrophage colony-stimulating factor (M-CSF; ABclonal, Wuhan, China) for 6 days. Fresh medium containing M-CSF was added on day 3.

Macrophage stimulation assays

RAW264.7 and BMDMs were stimulated with PBS, WGP, BCG or their combination (BCG + WGP). To ensure consistent terminology, the in vitro stimulation conditions were defined as follows. (i) WGP: WGP was added to the medium at 50 µg/mL and maintained continuously for up to 48 h. (ii) BCG: BMDMs or RAW264.7 cells were exposed to BCG at a multiplicity of infection (MOI) of 10 for 4 h, the supernatant was discarded, cells were washed three times with PBS, fresh medium was added, and cultures were continued to 48 h. (iii) BCG + WGP: during the first 0–4 h, cells were co-exposed to BCG at MOI 10 and WGP at 50 µg/mL; at 4 h, BCG was removed by three PBS washes, and WGP (50 µg/mL) was re-added and maintained for the indicated time points (6, 12, 24, or 48 h).

CCK-8 assays

BMDM viability was assessed using a Cell Counting Kit-8 (CCK-8, Beyotime, Shanghai, China) according to the manufacturer’s instructions. In brief, cells were seeded in a 96-well plate and treated with WGP, BCG, or their combination. After incubation for 12, 24, or 36 h, CCK-8 reagent was added to each well and incubated for an additional 2–4 h. The absorbance at 450 nm was then measured using a microplate reader (SpectraMax i3, USA) to determine the number of viable cells. PBS-treated cells served as the control.

RT-qPCR

Total RNA was isolated from macrophages using Trizol reagent (Huiyu Bio, Wuhan, China) following the manufacturer’s protocol. RNA purity and concentration were measured using a NanoDrop spectrophotometer (Thermo Fisher Scientific, MA, USA). cDNA synthesis was performed using a reverse transcription kit (Servicebio, Wuhan, China). Quantitative PCR was performed using SYBR Green Master Mix (TOLOBIO, Shanghai, China) on an ABI 7500 Fast system (Applied Biosystems, Singapore). Relative gene expression levels were calculated using the 2^−ΔΔCt method, normalized to GAPDH.

ELISA

Supernatants from stimulated macrophage cultures were collected at 48 h, and cytokine levels of IL-10, IFN-γ, and TNF-α were quantified using ELISA kits (CUSABIO, Wuhan, China) according to the manufacturer’s instructions.

Flow cytometry analysis

For flow cytometry analysis, macrophages were harvested and stained with viability dye (BV510-live) and surface antibodies, including anti-F4/80 (APC), anti-CD80 (PE-Cy7), anti-CD86 (BV650), anti-MHC I (AF700), anti-MHC II (BV711), anti-CD11b (PerCP-Cy5.5), anti-CD45 (APC-Cy7), anti-Ly6c (BV650), anti-CD3 (FITC), anti-CD4 (APC), anti-CD8 (BV480), anti-CD44 (PerCP-Cy5.5), and anti-CD62L (APC-Cy7) (BioLegend, CA, USA), anti-Dectin-1 (BV605). Intracellular cytokines (IL-2, IFN-γ) were detected following fixation and permeabilization. Samples were acquired on a Cytek Aurora flow cytometer (Cytek Biosciences, CA, USA) and analyzed using FlowJo software (TreeStar, v10.6.2).

Phagocytosis and degradation assay

BCG bacteria were labeled with rhodamine (0.7%) for 30 min at 37 °C, washed extensively, and adjusted to 1 × 10⁹ CFU/mL. Macrophages were incubated with labeled BCG (MOI = 10) for 4 h or 24 h in the presence or absence of WGP. Macrophage phagocytosis and intracellular degradation of BCG were quantified by flow cytometry using rhodamine fluorescence intensity.

Lysosomal acidification assessment

BMDMs cells were seeded onto confocal dishes and stimulated as described previously. Lysosomal acidification was assessed by incubating cells with LysoSensor Green DND-189 probe (1 μM, Invitrogen, CA, USA) for 1 h, and nuclei were counterstained with Hoechst33342 for 30 min. Images were captured by a Leica Stellaris 5 confocal microscope (Leica Microsystems, Wetzlar, Germany). LysoSensor Green images were analyzed blinded. Mean fluorescence per cell (≥100 cells/condition/experiment, n ≥ 3 experiments) was extracted in CellProfiler and summarized as mean ± SD.

Western blot

Total cellular proteins or nuclear/cytoplasmic fractions were isolated and quantified using BCA assay (Beyotime, Shanghai, China). Protein samples were separated by SDS-PAGE, transferred to PVDF membranes (Cytiva, NY, USA), and incubated overnight at 4 °C with primary antibodies against MHC-I (Arigo, Hangzhou, China), MHC-II (Affinity Biosciences, OH, USA), JAK1, phospho-JAK1, STAT1, phospho-STAT1, and Lamin A/C (HuaBio, Wuhan, China). Blots were visualized with HRP-conjugated secondary antibodies and ECL detection reagent (Yamei Biotech, Beijing, China).

RAW264.7 cells or BMDMs were stimulated with PBS, WGP, BCG, or BCG + WGP. Cells were harvested at 1 h to detect phosphorylated JAK1 and STAT1, and at 24 h to detect MHC-II expression. For pathway inhibition, BMDMs were pretreated for 2 h with abrocitinib (JAK1 inhibitor) or fludarabine (STAT1 inhibitor)^35,36 in complete medium (vehicle: DMSO), then washed twice with warm PBS to remove inhibitors. Cells were subsequently stimulated with WGP under inhibitor-free conditions. Samples were collected at 1 h for analysis of p-JAK1 and p-STAT1 and at 24 h for MHC-II expression.

CD4⁺ and CD8⁺ T-cell co-culture assays with BMDMs

BMDMs were pretreated for 2 h with fludarabine (STAT1 inhibitor, 10 µM) or vehicle (0.1% DMSO), then washed twice with warm medium. Cells were stimulated with WGP (50 µg/mL, 4 h, 37 °C) and washed twice prior to T-cell co-culture. After WGP stimulation, BMDMs used for CD4⁺ T-cell co-culture were pulsed with OVA323–339 peptide (ISQAVHAAHAEINEAGR, 100 nM, 30 min, 37 °C), then washed twice with warm medium to remove excess peptide. For CD8⁺ T-cell co-culture, BMDMs were pulsed with OVA257–264 peptide (SIINFEKL, 100 nM, 30 min, 37 °C) under the same conditions. Spleens from OT-II and OT-I mice were processed through 70-µm strainers. CD4⁺ T cells from OT-II mice and CD8⁺ T cells from OT-I mice were purified by negative selection to >90% purity and labeled with CFSE (2.5 µM, 10 min, 37 °C). For the mechanistic assay shown in Fig. 3G, CFSE-labeled OT-II CD4⁺ T cells were added to peptide-pulsed BMDMs at a macrophage: T-cell ratio of 1:5 in complete RPMI. Conditions included medium only, BMDMs only, WGP-treated BMDMs, vehicle control, and STAT1 inhibitor. For the supplementary BMDMs comparison experiments, macrophages were stimulated using the conditions described above before co-culture with purified OT-II CD4⁺ or OT-I CD8⁺ T cells. T-cell proliferation was assessed by CFSE dilution, and IFN-γ production was analyzed by intracellular staining and flow cytometry.

BMDC generation, stimulation, and T-cell co-culture assays

For BMDC generation, bone marrow cells were cultured in RPMI 1640 supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin, 20 ng/mL recombinant murine granulocyte-macrophage colony-stimulating factor and 5 ng/mL recombinant murine interleukin-4 for 6–7 days. Fresh medium containing GM-CSF and IL-4 (ABclonal, Wuhan, China) was added on days 3 and 5. Non-adherent and loosely adherent cells were harvested on day 6 or day 7 and used for subsequent stimulation assays. BMDCs were stimulated with PBS, WGP, BCG, or BCG + WGP using the same concentrations and exposure scheme described above for macrophage stimulation assays. For flow-cytometric analysis, BMDCs were stained for CD80, CD86, MHC-II, and MHC-I. For BMDC–T-cell co-culture assays, BMDCs were pulsed with OVA323–339 peptide for OT-II CD4⁺ T-cell assays or with OVA257–264 peptide for OT-I CD8⁺ T-cell assays, washed twice, and then co-cultured with CFSE-labeled purified T cells in complete RPMI. T-cell proliferation was assessed by CFSE dilution, and IFN-γ production was analyzed by intracellular staining and flow cytometry.

Direct WGP stimulation of purified T cells in the absence of APCs

To determine whether WGP directly stimulates T cells independently of antigen-presenting cells, purified OT-II CD4⁺ T cells and OT-I CD8⁺ T cells were cultured without BMDMs or BMDCs under the indicated conditions, including PBS only and WGP only. In parallel, BMDM + PBS and BMDM + WGP co-culture conditions were included as APC-dependent controls. Purified T cells were labeled with CFSE before culture, and proliferation was assessed by CFSE dilution. IFN-γ production was analyzed by intracellular staining and flow cytometry.

Animal immunization and tissue analysis

For in vivo immunization, the cervical dorsal subcutaneous region was chosen as a technically reproducible injection site that permitted spatially controlled delivery of BCG and WGP in the same-site and split-site regimens. For the exploratory comparison of administration strategies, 6- to 8-week-old female C57BL/6J mice were randomly assigned to five groups: PBS, WGP, BCG, BCG + WGP same-site mixture, or BCG + WGP split-site administration. On day 0, mice received: (i) PBS group, PBS control; (ii) WGP group, WGP i.p.; (iii) BCG group, BCG s.c. at the dorsal neck; (iv) BCG + WGP same-site group, a same-site subcutaneous mixture of BCG and WGP at the dorsal neck; or (v) BCG + WGP split-site group, WGP i.p. together with BCG s.c. at the dorsal neck. This exploratory experiment was used to compare tolerability and early pulmonary myeloid responses between same-site and split-site administration and is presented in Supplementary Fig. 4.

For the formal immunization experiments, six- to eight-week-old female C57BL/6 J mice were randomly assigned to four groups: PBS, WGP, BCG, or split-site WGP + BCG. On day 0, mice received: (i) PBS group, 200 µL PBS i.p.; (ii) WGP group, WGP 50 µg per mouse i.p. in 100 µL PBS; (iii) BCG group, 1 × 10⁶ CFU BCG s.c. in 100 µL PBS at the dorsal neck; or (iv) WGP + BCG group, both WGP i.p. and BCG s.c. on the same day at different sites. BCG was administered subcutaneously at the dorsal neck to provide a consistent injection site and to permit clear anatomical separation from intraperitoneal WGP administration in the split-site regimen. These formal experiments were used for subsequent analyses of day-10 myeloid responses and day-30 memory T-cell responses.

For the exploratory comparison of early and later myeloid responses, separate cohorts of mice were euthanized on Days 1, 3, and 10 after treatment, and lungs and spleens were collected at each time point. Lungs were minced and digested in RPMI containing 1 mg/mL collagenase D and 100 µg/mL DNase I for 30 min at 37 °C, then passed through 70-µm cell strainers and treated with ACK buffer for erythrocyte lysis. Spleens were mechanically dissociated and subjected to ACK lysis. Single-cell suspensions were stained with viability dye and antibodies against CD45, Ly6G, Ly6C, CD11b, F4/80, CD64, CD80, CD86, MHC-I, and MHC-II, and analyzed by flow cytometry. Ly6G⁺ cells were quantified separately, and the macrophage-enriched myeloid population was defined as CD45⁺Ly6C⁻CD11b⁺F4/80⁺CD64⁺ cells.

Day 1 and Day 3 samples were used to analyze Ly6G⁺ cells, macrophage-enriched myeloid cells, and the surface expression of CD80, CD86, MHC-I, and MHC-II in both lung and spleen, and these data are shown in Supplementary Fig. 4. Lungs collected on Day 1 and 3 were also processed for hematoxylin and eosin staining. Day 10 samples were used to analyze Ly6G⁺ cells and macrophage-enriched myeloid cells in lung and spleen, and these data are shown in Fig. 4. Surface-marker analyses of CD80, CD86, MHC-I, and MHC-II were not performed in the Day 10 cohort. Fluorescence minus one control was included in the staining panel and are shown in Supplementary Fig. 4. The same FMO controls were used to guide gating for the Day 10 analyses shown in Fig. 4B, C.

For memory CD4⁺ and CD8⁺ T-cell detection, animals were sacrificed on day 30. Lung and spleen cells were prepared as above and either stained directly for memory-phenotype markers (CD3, CD4 or CD8, CD44, CD62L) or restimulated ex vivo with heat-killed BCG (10 µg/mL) in the presence of brefeldin A for 6 h for lung cells or 18 h for splenocytes at 37 °C. After stimulation, cells were surface stained, fixed, permeabilized, and intracellularly stained for IL-2 and IFN-γ in CD4⁺ T cells before acquisition on a Cytek Aurora. Memory subsets were defined as central memory (TCM, CD44^hi CD62L^hi) and effector memory (TEM, CD44^hi CD62L^lo) within live CD4⁺ or CD8⁺ T cells, and cytokine frequencies were expressed as a percentage of total CD4⁺ T cells.

Macrophage depletion and reconstitution experiments

WT mice received intravenous injections of clodronate liposomes (FormuMax, CA, USA) to deplete macrophages. After confirming macrophage depletion, WT or Dectin-1 KO BMDMs were adoptively transferred intravenously at day −1. Subsequently, mice were immunized with WGP i.p. plus BCG s.c. at the dorsal neck. At day 30, lungs and spleens were harvested for CD4⁺ memory T cell and cytokine analysis by flow cytometry, following ex vivo stimulation with heat-killed BCG.

Mycobacterium tuberculosis H37Ra (M.tb H37Ra) infection model

Female C57BL/6J mice (6–8 weeks) were randomized (n = 6/group) into five groups: PBS, WGP, BCG, BCG + WGP same-site mixture, or BCG + WGP split-site administration. The challenge experiment was designed to compare the protective effects of different WGP and BCG administration strategies identified in the preceding immunization experiments. All procedures were approved by the Institutional Animal Care and Use Committee and followed relevant guidelines (BSL-2 for H37Ra).

Mice were immunized on day −28 as follows: the PBS group received PBS control treatment; the WGP group received WGP 50 µg per mouse i.p.; the BCG group received BCG s.c. at the dorsal neck; the BCG + WGP same-site group received a same-site subcutaneous mixture of BCG and WGP; and the BCG + WGP split-site group received WGP i.p. together with BCG s.c. at the dorsal neck. On Day 0 (28 days after immunization), the mice were lightly anesthetized with isoflurane and challenged intranasally with 1 × 10⁷ CFU M.tb H37Ra in 50 µL PBS (25 µL per nostril). Animals were monitored daily, with predefined humane endpoints. On Day 28 post-infection, mice were euthanized; lungs and spleens were aseptically harvested, weighed, and homogenized in 2 mL sterile PBS. Serial 10-fold dilutions (10⁰–10⁻⁵) were plated on Middlebrook 7H10 agar (OADC) and incubated at 37 °C for 14–21 days. Colony-forming units (CFU) were enumerated and expressed as CFU per gram of tissue. Bacterial burden (log¹⁰ CFU) was the primary outcome for dose-response and mechanism studies.

Lung tissues were fixed in 10% neutral-buffered formalin, paraffin-embedded, and sectioned. Acid-fast bacilli were detected by acid-fast bacillus staining according to the manufacturer’s instructions. H&E staining was performed on lung sections to assess inflammatory changes, including peribronchial/perivascular cuffing, alveolar infiltrates, and consolidation.

Randomization and blinding

Randomization method: Mice were randomly assigned to experimental groups using a random number generator (GraphPad QuickCalcs, https://www.graphpad.com/quickcalcs/randomize1/) after genotyping and before any experimental procedures. Confounders: Mice from different groups were housed in separate cages but within the same rack to minimize environmental variation. Treatment order was randomized daily. Cage location was rotated weekly to control for position effects within the animal facility. Investigators were blinded to group allocation during survival monitoring, histological scoring, and behavioral assessments. The investigator performing infections and tissue collection was not blinded due to visible differences in treatments (e.g., injection sites, distinct phenotypes of knockout mice). Data analysis was performed by an investigator blinded to group allocation until final statistical analysis.

Statistics and reproducibility

Statistical analyses were performed using Prism 9 (GraphPad Software). Each group of data was subjected to Shapiro–Wilk test for normal distribution check. Two-tailed unpaired Student’s t-test was used to determine significance between two groups of normally distributed data. Mann–Whitney test was used to determine significance between data without a normal distribution. For comparisons between multiple groups, an ordinary one-way or two-way analysis of variance followed by Tukey post hoc test was used. P-values less than 0.05 were considered statistically significant (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001). ns represents no significance. Sample sizes were determined based on prior studies and pilot experiments to ensure adequate statistical power (≥0.80) to detect biologically meaningful differences. For in vitro experiments, n represents the number of biologically independent experiments performed on different days with independently prepared cells or samples. For in vivo experiments, n represents the number of biologically independent animals. Each biological replicate was measured in technical duplicate or triplicate as specified in figure legends, and the mean of technical replicates was used for statistical analysis.

Reporting summary

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

Author contributions

Y.G.C., H.R.Q., and Q.T. contributed equally to this work and are co-first authors. Q.P., F.L.L., and M.L. jointly supervised this work and are co-corresponding authors. Additional contributions: Conceptualization: F.L.L. and M.L. Methodology: Y.G.C., H.R.Q., Q.T., S.J.S., W.W.Z., X.S.D., F.D., P.R.C., and C.Y.Z. Investigation: F.F.L., M.L., Q.P., and X.L.Z. Visualization: H.R.Q. and Q.T. Funding acquisition: F.L.L. and M.L. Project administration: F.L.L. and M.L. Supervision: X.L.Z., Q.P., F.L.L., and M.L. Writing—original draft: H.R.Q., M.L., and F.L.L. Writing—review & editing: H.R.Q., M.L., and F.L.L.

Peer review

Peer review information

Communications Biology thanks the anonymous reviewers for their contribution to the peer review of this work. Primary Handling Editors: Dr Weishan Huang and Dr Ophelia Bu.

Funding

This work was supported by: National Natural Science Foundation of China grants 82272978 (M.L.), 81971904 (F.L.L.), 82230078 (X.L.Z.); National Key R&D Program of China 2022YFA1303500 (X.L.Z., Q.P.). The Fundamental Research Funds for the Central Universities grants 2042022dx0003, 2042023kf1011 (X.L.Z.).

Data availability

All data supporting the findings of this study are contained within the article and its supplementary content. Uncropped and unprocessed blot and gel images are provided in Supplementary Fig. 8. Numerical values underlying bar graphs, line plots, and statistical analyses for Figs. 1–6 and Supplementary Figs. 1–7 are provided in Supplementary Data 1. All other data are available from the corresponding author (or other sources, as applicable) on reasonable request.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s42003-026-10492-6.