Recognition of Brucella abortus drives M2 like polarization and impaired antigen presentation in monocyte derived macrophages

Abstract

Macrophages play essential roles in host defense and the resolution of tissue damage. Under inflammatory conditions, circulating monocytes migrate from the bloodstream to affected tissues, where they differentiate into macrophages whose characteristics are shaped by the local microenvironment. Brucellosis, a zoonotic disease caused by Brucella spp., is characterized by the ability of these pathogens to invade and persist primarily within monocytes and macrophages. Although Brucella spp. can activate these cells, it remains unclear whether Brucella-host interactions modulate monocyte-to-macrophage differentiation. The aim of this work was to examine how Brucella abortus recognition influences macrophage differentiation. Monocytes were differentiated into macrophages with M-CSF in the presence of B. abortus, followed by phenotypic and functional characterization. The macrophages differentiated in the presence of B. abortus presented reduced expression of antigen-presenting markers (HLA-DR, HLA-ABC and CD86), increased M2/pro-regenerative features, and increased basal secretion of cytokines and phagocytic activity. However, they fail to respond to further stimulation (e.g., E. coli LPS) and exhibit poor T cell priming capacity. Macrophages differentiated in the presence of B. abortus exhibit a mixed profile characterized by a reduction in antigen-presenting molecules, which impairs the adaptive immune response, while adopting an M2-like/pro-regenerative phenotype contributes to chronic infection. Moreover, macrophages differentiated in the presence of B. abortus secrete cytokines and chemokines, perpetuating inflammation at the infection site, a hallmark of brucellosis.

Introduction

Macrophages are cells of the innate immune system with a highly adaptable phenotype and function and play essential roles in different contexts, such as tissue development and homeostasis, defense against pathogens, and resolution of tissue damage¹. During embryonic development, tissue-resident macrophages arise from the yolk sac or liver progenitors. However, in inflammatory circumstances, circulating monocytes in the bloodstream, which originate in the bone marrow, are recruited in large numbers to inflamed tissues, where they can differentiate into macrophages^1,2. The phenotype and functions of differentiated macrophages are strongly influenced by the microenvironment, which reflects the health/damage status of the tissue, which is shaped by metabolic conditions, the presence of immune mediators and cells, and pathogen exposure or pathogen-derived molecules³. The molecular motifs of different pathogens can be recognized by several receptors present on macrophages, known as pattern recognition receptors. The activation of different sets of receptors on innate immune cells triggers immune responses and modulates the functional profile of these cells, ultimately shaping the overall inflammatory response.

Brucellosis is a zoonotic disease caused by members of the genus Brucella, which are Gram-negative bacteria that infect domestic and wild animals, including cattle, sheep, goats, pigs and dogs. Several species (e.g., B. melitensis, B. abortus, B. canis and B. suis) can infect humans as incidental hosts^4,5. Brucellosis is recognized as one of the most prevalent bacterial zoonotic diseases, with an estimated incidence exceeding 500,000 new human cases annually and a prevalence of 10 cases per 10,000 inhabitants in endemic regions⁶. Brucella spp. are intracellular pathogens that preferentially infect and replicate in macrophages^4,7 and modulate several immune functions, particularly in monocytes/macrophages, establishing a chronic infection. To accomplish this task, B. abortus is armed with a range of survival strategies that allow bacteria to tilt this warfare in favor of it⁸. Brucella establishes persistent infection inside macrophages, forming a Brucella-containing vacuole, where it resides, becomes resistant to further attacks, and multiplies⁸.

As shown in our previous work and that of others, several innate immune cells are capable of sensing Brucella^9–12, triggering the secretion of proinflammatory cytokines and chemokines¹¹ and the production of reactive oxygen species¹⁰ and nitric oxide^12,13. Additionally, Brucella has different mechanisms to modulate innate and adaptive immune responses to evade them and survive within cells and the organism¹⁴. We have described several of these mechanisms, such as the inhibition of the expression of antigen-presenting molecules increased by interferon-gamma (IFN-γ) exposure^15–20, as well as the downmodulation of the receptor type I of the immunoglobulin G (IgG) constant fraction²¹.

Different reports have shown how Brucella spp. sensing modulates the response generated by macrophages^9,14,22,23, but how the recognition of Brucella spp. modulates the differentiation process from monocytes to macrophages, conditioning the subsequent innate and adaptive response, remains poorly understood.

Thus, the aim of this study was to elucidate whether B. abortus affects the cellular phenotype and functions of macrophages during their differentiation from monocytes. Here, we show that B. abortus modulates the phenotype and function of macrophages, leading to reduced expression of antigen-presenting molecules, increased expression of M2-associated markers, and a decreased ability to activate T cells. On the other hand, these macrophages secrete relatively high levels of cytokines and exhibit increased phagocytic activity.

Results

B. abortus recognition modulates the phenotype of macrophages during their differentiation from monocytes

We first studied whether the recognition of B. abortus by monocytes might modulate the phenotype of macrophages during differentiation events. To this end, monocytes were cultured in the presence of heat-killed B. abortus (HKBa) for 5 days, with the addition of M-CSF to allow for macrophage differentiation. Monocytes cultured with only M-CSF served as a control. Macrophage shape and cell viability were not affected by B. abortus treatment (Fig. 1A, B). To characterize the macrophage phenotype, the expression of HLA-DR, HLA-ABC, CD86, CD80, CD14, CD40 and CD54 was evaluated by flow cytometry. Macrophages differentiated in the presence of B. abortus presented lower expression of the antigen-presenting molecules HLA-DR and HLA-ABC (p < 0.005) and the costimulatory molecule CD86 (p < 0.05). CD54 and CD40 were also downregulated (p < 0.05) (Fig. 1C, D). On the other hand, the expression of CD14 and CD80 was not modified (p > 0.05). These results indicate that the recognition of B. abortus modulates the differentiation profile of macrophages, diminishing the expression of molecules related to antigen presentation and costimulatory functions.

Fig. 1.

B. abortus recognition modulates the phenotype of macrophages. Human monocytes were differentiated into macrophages with M-CSF for 5 days in the presence or absence of B. abortus (B.a.) (HKBa, 1 × 10⁸ bacteria/ml). (A) Cells were fixed, stained with anti-CD14 antibodies (red) and DAPI (blue) and analyzed via fluorescence microscopy. Representative images are shown. Scale bar: 50 μm. (B) Macrophages were stained with 7AAD, and cell viability was evaluated. A representative histogram and the average of 3 experiments are plotted. (C) The expression of the indicated surface markers on macrophages was evaluated by flow cytometry. A representative histogram for each marker is depicted. (D) Mean fluorescent intensities (MFIs) are plotted for each marker. Each point represents a different blood donor (n = 3–12, *p < 0.05, **p < 0.005).

When monocytes are recruited to the site of infection, the microenvironment modulates their functional polarization into phenotypically and functionally distinct macrophages, such as M1 and M2¹. Although this dichotomy is a simplification of macrophage plasticity, it provides a useful framework for exploring functional polarization. Thus, to investigate whether the recognition of B. abortus could modulate M1/M2 polarization during macrophage differentiation, we measured the expression of M1 and M2 markers by flow cytometry and qRT‒PCR. Along with the reduced expression of HLA-DR, HLA-ABC and CD86, we also observed lower expression of the M1 surface marker CD64 (p < 0.05). Moreover, we measured the relative mRNA expression of the inducible nitric oxide synthase (iNOS) enzyme associated with the M1 phenotype and observed lower expression than in control macrophages (p < 0.01) (Fig. 2A). When M2 markers were analyzed, we detected increased expression of CD209 (p < 0.0001) and CD206 (p < 0.0001) in B. abortus-treated cells (Fig. 2A), whereas CD163 and arginase (Arg) were not modulated, suggesting an M2 polarization profile.

Fig. 2.

B. abortus recognition modulates the macrophage polarization profile. Human monocytes were differentiated into macrophages with M-CSF for 5 days in the presence or absence of B. abortus (B.a.) (HKBa, 1 × 10⁸ bacteria/ml). (A) The expression of M1 and M2 surface markers was evaluated via flow cytometry. The mean fluorescence intensity (MFI) is plotted for each marker. (B) Semiautomated analysis of flow cytometry data by tSNE (i). The histogram overlays depict the expression of each analyzed marker (ii). (C) Visualization by tSNE reveals distinct macrophage populations, with the FlowSOM algorithm identifying 4 cell clusters. Each cluster is visualized in a different color (i-ii). The ClusterExplorer algorithm quantifies cell counts within these FlowSOM clusters and facilitates the generation of a heatmap to examine protein expression patterns across each identified cluster (iii). The percentage of each cluster per treatment is shown (iv). (D) Gene expression of tissue remodeling markers was evaluated via RT‒qPCR. Fold enrichments are plotted for each marker. (n = 3–8, *p < 0.05, **p < 0.005, ***p < 0.001).

To characterize the heterogeneity of macrophage populations generated in the presence or absence of B. abortus, we performed high-dimensional flow cytometry analysis using M1/M2 markers. Through t-distributed stochastic neighbor embedding analysis (t-SNE), we observed that both conditions share a similar overall phenotypic structure (Fig. 2Bi), with the main differences residing in the expression of specific M1/M2 markers. Consistent with our previous analysis, histograms revealed a rightward shift in CD209 and CD206 and, to a lesser extent, CD163 in macrophages differentiated in the presence of B. abortus compared with the control, whereas CD64 remained largely unchanged (Fig. 2Bii). This pattern suggests a bias toward alternative activation programs without loss of overall macrophage identity. Both the t-SNE and FlowSOM algorithms enabled the identification of four clusters in macrophages differentiated in the presence or absence of B. abortus (Fig. 2Ci and ii). Cluster 2, corresponding to an M2-like phenotype (CD209^hi/CD206^hi/CD163^int/CD64^low), predominated under both conditions, and its frequency increased in macrophages exposed to B. abortus compared with the control (mean control 50% vs. B.a. 68%, p < 0.001) (Fig. 2Civ). Conversely, Cluster 1 (CD209^int/CD206^low/CD163^low/CD64^low) was markedly decreased (mean control 49% vs. B.a. 29%, p < 0.001), accompanied by a slight increase in Cluster 3 (CD209^low/CD206^hi/CD163^low/CD64^low) in the presence of B. abortus (mean control 1.5% vs. B.a. 2.7%, p = 0.06) (Fig. 2Civ). This pattern suggests that Cluster 3 might represent a transitional phenotype toward alternative macrophages, characterized by early induction of CD206, yet without full acquisition of the regulatory functions associated with CD209 and CD163. In contrast, Cluster 1 remains a population low in alternative markers, potentially linked to more proinflammatory functions. Finally, Cluster 4 (CD209^hi/CD206^hi/CD163^hi/CD64^low), although minor, was slightly increased in macrophages differentiated in the presence of B. abortus (mean control 0.14% vs. B.a. 0.30%, p = 0.058).

Together, these results indicate that B. abortus reshapes macrophage differentiation, favoring M2-like phenotypes while maintaining intermediate states that may contribute to a functionally mixed response.

To complete the characterization of the macrophage population, we analyzed markers related to cell growth and tissue remodeling, such as CD93 (C1qR1), CD300E, RGS2 (Regulator of G protein signaling 2) and VEGF (Vascular Endothelial Growth Factor)^24,25, via RT‒qPCR and detected increased expression of CD300E (p < 0.05) and slight upregulation of VEGF expression (p = 0.06) but no modulation of CD93 or RGS-2 (p > 0.05) when macrophages were differentiated in the presence of B. abortus (Fig. 2D). Overall, these results indicate that the recognition of B. abortus by monocytes modulates their differentiation toward macrophages with an M2-like/pro-regenerative profile.

B. abortus recognition during macrophage differentiation modulates cytokine secretion and phagocytic activity

We evaluated the ability of macrophages differentiated in the presence of B. abortus to secrete the proinflammatory cytokines IL-6, IL-1β and TNF-α; the chemokine CCL2; and the anti-inflammatory cytokines IL-10 and TGF-β. We found an increment in the secretion of IL-6 (mean control 0.15 vs. B.a. 294 ng/ml, p < 0.05), IL-1β (mean control 29 vs. B.a. 7050 pg/ml, p < 0.005) and TNF-α secretion (mean control 11 vs. B.a. 2255 pg/ml, p < 0.05); as well as in the chemokine CCL2 secretion (mean control 175 vs. B.a. 1295 pg/ml, p < 0.05) when macrophages were differentiated in the presence of B. abortus antigens compared to the control. The release of anti-inflammatory cytokines such IL-10 (mean control 72 vs. B.a. 1350 pg/ml, p < 0.005) and TGF-β (mean control 0.3 vs. B.a. 0.52 OD 650nM, p < 0.05) was also increased in B. abortus-differentiated macrophages than in controls (Fig. 3A).

Fig. 3.

B. abortus promotes phagocytosis and cytokine secretion. Human monocytes were differentiated into macrophages with M-CSF for 5 days in the presence or absence of B. abortus (B.a.) antigens (HKBa, 1 × 10⁸ bacteria/ml). (A) The secretion of IL-6, IL-1β, TNF-α, IL-10, TGF-β and CCL-2 in cultured supernatants was measured via ELISA. (B) Engulfment of CFSE + C. albicans (C.a.) was evaluated by flow cytometry (represented as a percentage of CFSE+ macrophages). (C) Engulfment of CFSE + C. a.  was evaluated by fluorescence microscopy. A picture of a representative donor is shown. The number of C.a. per macrophage and the percentage (%) of macrophages that phagocytosed C.a. are plotted. Each point represents an independent monocyte donor. (n = 4–11, *p < 0.05, **p < 0.005).

Monocyte differentiation into macrophages is characterized by an increase in phagocytic activity. Thus, we evaluated this modulation using fluorescence microscopy and flow cytometry. We focused on macrophages that were differentiated in the presence of B. abortus by culturing them in the presence of CFSE-labeled Candida albicans. We observed an increase in both the percentage of macrophages that were able to phagocytose C. albicans and the number of C. albicans engulfed per macrophage in B. abortus-treated cells (Fig. 3B-C). These results indicate that, compared with control macrophages, macrophages differentiated in the presence of B. abortus secrete increased levels of proinflammatory and anti-inflammatory cytokines and chemokines and have increased phagocytic ability.

Macrophages differentiated in the presence of B. abortus display an impaired ability to prime T cells

Next, we evaluated whether heat-killed B. abortus-induced modulation of macrophage differentiation affects CD4⁺ and CD8⁺ T cell activation. To this end, macrophages were cocultured with allogenic T cells, and T cell activation was evaluated by measuring T cell proliferation by the CFSE dilution assay and performing intracellular IFN-γ staining of CD4⁺ and CD8⁺ T cell populations via flow cytometry (Fig. 4A). We observed reduced proliferation when T cells were cocultured with B. abortus-treated cells (Fig. 4B). Moreover, both the CD4⁺ and CD8⁺ populations produced lower levels of IFN-γ when stimulated with B. abortus-differentiated macrophages (Fig. 4C). Next, we evaluated the ability of B. abortus-treated macrophages to activate naïve CD4⁺ T cells. Confirming the ability of B. abortus-treated macrophages to activate T cells, we observed a reduced capacity of B. abortus-treated macrophages to activate naïve CD4⁺ T cells compared with that of control macrophages (Fig. 4D). Taken together, these results indicate that the presence of B. abortus during macrophage differentiation leads to diminished antigen presentation to T cells, likely in part due to the reduced expression of antigen presentation and costimulatory molecules.

Fig. 4.

Macrophages differentiated in the presence of B. abortus fail to activate T lymphocytes. Human monocytes were differentiated into macrophages with M-CSF for 5 days in the presence or absence of B. abortus (B.a.) antigens (HKBa, 1 × 10⁸ bacteria/ml). On day 5, the cells were cocultured with heterologous PBMCs (B) or naïve CD4⁺ T cells (C). The gating strategy is shown (A). T cell proliferation was measured 3 days later with CFSE dye by flow cytometry (B and D). Intracellular IFN-γ was measured by flow cytometry 24 h later (C). Each point represents an independent monocyte/lymphocyte donor. (n = 3–5, *p < 0.05, **p < 0.005, ***p < 0.0001).

Impaired activation of macrophages differentiated in the presence of B. abortus

Next, we investigated whether macrophages differentiated in the presence of heat-killed B. abortus retained the ability to respond to activation stimuli. Therefore, monocytes were differentiated in the presence or absence (control) of B. abortus for 5 days and subsequently stimulated with Escherichia coli lipopolysaccharide (EC-LPS) for an additional 24 h. As shown in Fig. 5A, the recognition of B. abortus during macrophage differentiation impaired the upregulation of activation markers induced by EC-LPS, such as HLA-ABC and CD54 (p < 0.05 and 0.005, respectively), as well as any modulation of HLA-DR and CD86 expression upon EC-LPS stimulation. As previously observed (Fig. 3), B. abortus-treated cells secreted higher levels of cytokines than control untreated cells did. However, when these macrophages were stimulated with EC-LPS, there was no additional increase in cytokine secretion (Fig. 5B). Notably, IL-6, IL-1β and CCL2 production was greater in macrophages differentiated in the presence of B. abortus (with or without additional stimulation with EC-LPS) than in control EC-LPS-stimulated macrophages (Fig. 5B). Alternatively, control macrophages stimulated with EC-LPS secreted greater amounts of TNF-α than B. abortus differentiated cells did. No differences in IL-10 secretion were detected between EC-LPS-activated and B. abortus-differentiated macrophages. Moreover, as shown above, B. abortus-treated macrophages have a greater phagocytic capacity than control macrophages do, which was not modulated by EC-LPS stimulation (Fig. 5C). Together, these findings indicate that B. abortus not only modulates macrophage differentiation but also imposes functional reprogramming that limits their subsequent responsiveness to classical inflammatory stimuli.

Fig. 5.

Macrophages differentiated in the presence of B. abortus exhibit impaired activation. Human monocytes were differentiated into macrophages with M-CSF for 5 days in the presence or absence of B. abortus (B.a.) (HKBa, 1 × 10⁸ bacteria/ml). On day 4, the cells were activated with E. coli LPS. Nonstimulated macrophages served as a negative control (NS). (A) The expression of surface markers on macrophages was evaluated via flow cytometry. Normalized mean fluorescent intensities (MFIs) are plotted for each marker. (B) The secretion of IL-6, IL-1β, TNF-α, TGF-β, IL-10 and MCP-1 was measured in cultured supernatants. (C) Phagocytosis activity was evaluated by the engulfment of CFSE-stained C. albicans (represented as a percentage of CFSE+ macrophages) or fluorescent beads (represented as a percentage of control MIF). Each point represents an independent monocyte donor. (n = 4–10, *p < 0.05, **p < 0.01, **p < 0.005).

Lipoproteins and DNA are key B. abortus PAMPs that modulate the macrophage phenotype

L-Omp19 is an outer membrane lipoprotein of B. abortus that plays a crucial role in bacterial interactions with the host immune system. We previously reported that Brucella lipoproteins have potent stimulatory effects, stimulating TLR2 on myeloid cells^{11,12,21,26,27}. Thus, we assessed the role of L-Omp19 in the modulation of macrophage phenotypes. For this purpose, monocytes were cultured for 5 days with L-Omp19 (500 ng/ml) or U-Omp19 (unlipidated protein form, 500 ng/ml) plus M-CSF to allow macrophage differentiation. The expression levels of HLA-DR, HLA-ABC, CD86, and CD54 were evaluated by flow cytometry (Fig. 6A). Like heat-killed bacteria, macrophages differentiated in the presence of L-Omp19 presented lower expression of HLA-DR (p < 0.01) and CD54 (p < 0.05) than control macrophages did. However, we observed no differences in HLA-ABC or CD86 expression between the treatment groups. U-Omp19 had no effect on the macrophage phenotype for any evaluated marker, indicating that the lipidated moiety is essential for inducing its effect, as previously reported¹¹.

Fig. 6.

L-Omp19 modulates monocyte differentiation. Human monocytes were differentiated into macrophages with M-CSF for 5 days in the presence or absence of different PAMPs present in B. abortus: (A) L-Omp19 or U-Omp-19 (500 ng/ml)), (B) RNA (5 ng/ml), DNA (5 ng/ml) or LPS (1 µg/ml). The expression of surface markers in macrophages was evaluated via flow cytometry. Normalized mean fluorescent intensities (MFIs) are plotted for each marker. (n = 3–8, *p < 0.05, **p < 0.01, ***p < 0.001).

Other relevant B. abortus pathogen-associated molecular patterns (PAMPs) include bacterial DNA and RNA^17,28,29, as well as B. abortus lipopolysaccharide (BA-LPS), which is known for its low immunogenicity¹¹. Therefore, we assessed the contributions of B. abortus DNA, RNA, and BA-LPS to the modulation of the macrophage phenotype. We observed significant downregulation of HLA-DR (p < 0.05) and HLA-ABC (p < 0.05) when monocytes were differentiated into macrophages in the presence of B. abortus DNA, whereas the expression of CD86 and CD54 remained unchanged (Fig. 6B). In contrast, neither B. abortus RNA nor BA-LPS affected the expression of these surface markers.

These findings indicate that among the PAMPs tested, L-Omp19 and bacterial DNA are the main contributors to the modulation of the macrophage phenotype.

Discussion

Monocytes originate from hematopoietic precursors in the bone marrow. After entering the bloodstream, these cells migrate to inflamed peripheral tissues, where they become the predominant mononuclear cell population^1,30. During injury or inflammatory responses, circulating monocytes exit the vasculature and differentiate into macrophage or dendritic cell populations within tissues^{3,23,25,31,32}. Indeed, the detection of microorganisms through distinct sets of pattern recognition receptors by monocytes strongly influences their differentiation process. In the present study, we examined how the recognition of B. abortus by human monocytes influences their differentiation into macrophages. Our results show that this interaction alters both phenotype and function, leading to M2-like macrophages, which have a reduced capacity to activate T cells but display increased phagocytic activity and elevated cytokine production. Similarly, Mycobacterium tuberculosis has been shown to alter monocyte-to-macrophage differentiation by reducing the expression of HLA-DR and CD86, among other markers³³. In contrast, during monocyte-to-macrophage differentiation in the presence of EC-LPS, macrophages exhibit increased expression of major histocompatibility complex (MHC) I, MHC II and CD40³¹. Similarly, we previously reported that B. abortus downregulates HLA-ABC and HLA-DR on monocytes/macrophages activated with IFN-γ^15,16. Here, we describe an altered macrophage phenotype that emerges early during differentiation, prior to activation. This finding suggests that the phenotype acquired by macrophages derived from infiltrating monocytes in response to inflammation may vary depending on the pathogen, shaped by the specific combination of stimuli present during each encounter. Notably, our study demonstrated that macrophages differentiated in the presence of B. abortus fail to be activated by LPS.

Macrophages have high plasticity, and their polarization state significantly influences both innate and adaptive immune responses. Traditionally, they are classified as M1 (proinflammatory) or M2 (anti-inflammatory/regulatory) macrophages; however, additional subsets, such as M3 and M4, have been described³. M1 macrophages are associated with antimicrobial activity and tissue damage, whereas M2 macrophages modulate immune responses and promote tissue repair by secreting cytokines such as IL-10, TGFβ, and VEGF³. Previous studies have shown that M0 macrophages treated with RNA from B. abortus cannot be polarized to either an M1 or an M2 profile³⁴. Additionally, monocytes purified from Brucella-infected patients exhibit neither M1 nor M2-like profiles²³. In contrast, our data revealed increased expression of M2-associated markers, including CD209 (DC-SIGN) and CD206 (mannose receptor), along with reduced expression of CD64 (Fc-γ receptor) and iNOS, in macrophages differentiated in the presence of B. abortus.

Biomarkers for regulatory macrophages, such as CD300E, VEGF, CD93 and RGS-2 (among others), have been identified and can be used to classify proangiogenic and tissue remodeling markers on macrophages^24,25. Here, we observed robust upregulation of CD300E and slight upregulation of VEGF, revealing macrophages with a growth-promoting phenotype. Moreover, we detected elevated secretion of both anti-inflammatory cytokines (IL-10 and TGF-β) and proinflammatory mediators. Similarly, macrophages differentiated in the presence of M. tuberculosis antigens exhibit a reduced capacity to secrete both pro- and anti-inflammatory cytokines³³. In contrast, our findings demonstrate that macrophages differentiated in the presence of B. abortus antigens produce elevated levels of proinflammatory cytokines and chemokines, as well as anti-inflammatory cytokines. This ability of Brucella to induce high levels of proinflammatory cytokines among different cellular models is well established^{11,12,28,35–37}. We observed that phagocytic activity was greater in B. abortus-differentiated macrophages than in control macrophages, reflecting a proinflammatory functional profile despite an M2-like phenotype. Interestingly, when macrophages were stimulated with EC-LPS, they were incapable of secreting higher levels of cytokines, as previously described, whereas monocytes from Brucella-infected patients presented lower levels of proinflammatory cytokines in response to EC-LPS than did those from healthy controls²³.

In line with the low expression of HLA-ABC and HLA-DR, monocyte exposure to B. abortus during macrophage differentiation impairs their ability to activate CD4 + and CD8 + cells. As previously reported, B. abortus decreased antigen presentation to CD4 + and CD8 + T cells through HLA-DR and HLA-ABC downmodulation in macrophages stimulated with IFN-γ^14–16. Here, the observed downregulation of these molecules correlated with reduced T cell activation, possibly facilitating chronic infection. A similar phenomenon was described for M. tuberculosis, where monocyte-derived macrophages cocultured with CD3 + T cells induced less proliferation, likely due to reduced CD86 and HLA-DR expression³³. Similarly, infection with Toxoplasma gondii led to impaired antigen presentation in murine macrophages, which was attributed to decreased MHC-II expression³⁸.

Finally, we found that L-Omp19, a major Brucella PAMP, together with DNA recapitulates the phenotype observed with whole B. abortus antigens, as revealed by the downregulation of HLA-DR, HLA-ABC, and CD54. These results reinforce our previous finding that L-Omp19 is a key PAMP responsible for immune modulation^11,12,15,21. Brucella lipoproteins have been shown to inhibit HLA-DR surface expression (induced by IFN-γ) at the transcriptional level^15,18.

Our study demonstrated that the recognition of B. abortus induces a distinctive phenotype in human monocyte-derived macrophages, characterized by reduced expression of antigen-presenting functions and adhesion molecules. This results in impaired macrophage activation and T cell proliferation, with an M2-like profile, which favors chronic infection. However, these macrophages also produce elevated levels of proinflammatory mediators, which might promote tissue damage and additional monocyte recruitment. This phenotype likely facilitates immune evasion and chronic infection while contributing to the persistent inflammation characteristic of Brucella pathogenesis.

Materials and methods

B. abortus antigens

The B. abortus strain S2308 was grown for 3 to 5 days in tryptic soy agar medium at 37 °C (#BK046HA, Biokar Diagnostics, France). The bacterial stock was prepared in phosphate-buffered saline (PBS), and the bacterial concentration was estimated by the optical density of the culture at 600 nm via spectrophotometry. To obtain heat-killed B. abortus (HKBa), bacteria were washed five times for 10 min each in sterile PBS, heat-killed at 70 °C for 20 min, aliquoted, and stored at − 80 °C until use. B. abortus lipidated outer membrane protein 19 (L-Omp19) and unlipidated Omp19 (U-Omp19) were obtained as previously described¹¹. B. abortus deoxyribonucleic acid (DNA) was purified via extraction with phenol: chloroform: isoamyl alcohol and ethanol precipitation³⁹. To eliminate lipopolysaccharide (LPS) contamination, the DNA was adsorbed with Sepharose-polymyxin B (#P1411, Sigma‒Aldrich, USA). B. abortus ribonucleic acid (RNA) was purified with a Quick-RNA MiniPrep (#R1055, Zymo Research, USA) according to the manufacturer’s instructions and as previously described³⁹. Both recombinant proteins, DNA and RNA, contained less than 0.25 endotoxin U/µg of protein, as assessed by Limulus Amebocyte Lysates (Associates of Cape Cod, Inc., USA). B. abortus S2308 LPS was provided by I. Moriyon (University of Navarra, Pamplona, Spain).

Monocyte purification and macrophage differentiation

Blood samples from healthy donors were obtained from the Blood Bank of the Hospital de Clínicas “José de San Martín”, University of Buenos Aires (Buenos Aires, Argentina). The study protocol was approved by the Institutional Ethics Committee of the School of Medicine, University of Buenos Aires (approval ID: 05904894). All donors provided written informed consent in accordance with the principles of the Declaration of Helsinki and relevant local regulations. Peripheral blood mononuclear cells (PBMCs) were isolated from the buffy coats of healthy anonymous blood donors via density gradient centrifugation using Ficoll‒Hypaque (#171440-03, GE Healthcare, USA). Monocytes were isolated from PBMCs via CD14 magnetic beads (#130-097-052, Miltenyi Biotec, USA). Only monocytes with a purity of up to 85% were used for the experiments, as determined by flow cytometry with an anti-human CD14-FITC antibody (#555397, BD Pharmingen, USA)⁴⁰. To obtain macrophages, 250.000 monocytes were cultured in RPMI 1640 medium (#23400-021, Gibco, USA) for 2 h in 24-well plates. After this, nonadherent cells were removed by three washes with 37 °C pre-warmed PBS, and the medium was replaced with complete culture medium (RPMI 1640 medium containing 10% fetal bovine serum (Natocor, Argentina), 2 mM L-glutamine (#35050061, Gibco), 100 U/ml penicillin and 0.1 mg/ml streptomycin (#15240-062, Gibco), 10 ng/ml macrophage colony-stimulating factor (M-CSF) (#130-096-489, Miltenyi Biotec) alone or plus HKBa (1 × 10⁸ bacteria/ml), or its antigens: L-Omp19 [500 ng/ml], U-Omp19 [500 ng/ml], BA-LPS [1000 ng/ml], RNA [5 µg/ml], and DNA [5 µg/ml], all from B. abortus, which were cultured for 5 days at 37 °C and 5% CO₂. The concentrations of B. abortus antigens used in this study were selected based on previously published studies demonstrating their biological activity^{11,12,15,17,41}.

On the fifth day, the macrophages were washed with phosphate-buffered saline (PBS) and harvested with 5% PBS-EDTA for flow cytometry or RNA isolation, and the supernatants were stored at -80 °C until use. For immunofluorescence microscopy, macrophages were fixed with 4% PFA for 10 min at room temperature and incubated with an anti-human CD14 primary antibody (#55395, BD Pharmingen) for 60 min at 4 °C. After washing, the cells were incubated with an Alexa Fluor™ 546 secondary antibody (#A21133, Life Technologies, USA) for 30 min at 4 °C. Coverslips were mounted with Fluoroshield-DAPI (#F6057, Sigma‒Aldrich). The cells were visualized with an ECLIPSE, Ti-E PFS, Nikon fluorescence microscope, and the images were processed with ImageJ software.

For macrophage activation, on the fourth day of culture, the cells were stimulated with E. coli LPS (EC-LPS 100 ng/ml) (#L2630, Sigma‒Aldrich) or HKBa (1 × 10⁸ bacteria/ml). Unstimulated cells were used as a negative activation control.

Flow cytometry

For surface staining, the following antibodies were used for 30 min at 4 °C: FITC-conjugated anti-human leukocyte antigen (HLA)-DR (#555811), FITC-conjugated anti-human CD86 (#555658), PE-conjugated anti-human HLA-ABC (#555553), PE-conjugated anti-human CD54 (#555511), FITC-conjugated anti-human CD80 (#557227), FITC-conjugated anti-human CD40 (#555588), PerCP-conjugated anti-human CD14 (#340585), FITC-conjugated anti-human CD209 (#561764), FITC-conjugated anti-human CD64 (#560970) (all from BD Pharmingen), PerCP-conjugated anti-human CD163 (#333625) and PE-conjugated anti-human CD206 (#321105) (both from Biolegend, USA). In all cases, isotype-matched control antibodies were used. Samples were acquired via BD FACSCalibur^® and BD FACSAria II^® cytometers. The flow cytometer data were analyzed with FlowJo^® software (version 10.10.0). t-SNE and FlowSOM, which use self-organizing maps (SOM) based on marker expression phenotypes, were employed to assign each individual cell to clusters and metaclusters. FlowSOM was applied with the default settings for 7 samples from both the control and B. abortus treatments.

Viability assay

For the viability assay, macrophages were obtained on the fifth day of culture and washed, followed by incubation with 7-AAD (#559925, BD Pharmingen) for 10 min. After staining, analysis was performed using a BD FACSCalibur^® cytometer and FlowJo^® software. Heat-killed cells were included as a positive control for the technique.

Measurement of cytokines/chemokines

Human interleukin (IL)-6 (#555220), IL-1β (#557953), tumor necrosis factor (TNF) α (#555212), IL-10 (#555157) and chemokine (C-C motif) ligand 2 (CCL-2) (#555179) were measured in culture supernatants from different experiments via sandwich ELISA with paired cytokine-specific antibodies according to the manufacturer’s instructions (all from BD Biosciences, USA). Transforming growth factor (TGF) β was measured in HEK-Blue TGF-β reporter cells (TGF-β responsive reporter cells derived from HEK293 cells) following the manufacturer’s instructions (#hkb-tgfb, InvivoGen, USA). Three technical replicates were performed for each sample.

Phagocytosis assay

Candida albicans was grown to the stationary phase in YPD medium (#Y1375, Sigma‒Aldrich) at 30 °C with orbital shaking at 160 rpm. The labeling of C. albicans with carboxyfluorescein succinimidyl ester (CFSE) (#C34554, Invitrogen, USA) was performed by incubating 1 × 10⁸ yeasts with CFSE (2 µM) for 1 h at 37 °C. Yeasts were then washed twice in PBS and suspended in culture medium. The phagocytosis assay was performed by incubating macrophages with CFSE-labeled C. albicans at a macrophage/yeast ratio of 1:2 for 1 h at 37 °C and 5% CO₂ in complete culture medium⁴². The cells were washed and fixed with 4% PFA for 10 min at room temperature and then labeled with an anti-human CD14 primary antibody for 60 min at 4 °C. After washing, the cells were incubated with an Alexa Fluor™ 546 secondary antibody for 30 min at 4 °C. Coverslips were mounted with Fluoroshield-DAPI. The cells were visualized with an ECLIPSE, Ti-E PFS, Nikon fluorescence microscope, and the images were processed with ImageJ software. The percentage of macrophages phagocytosing C. albicans was evaluated via flow cytometry. For the phagocytosis assay with fluorescent beads, the macrophages were incubated for 2 h with 0.004% w/v 0.1 μm FluoSpheres™ Carboxylate-Modified Fluorescence Microspheres (#F8801, Invitrogen) for 2 h at 37 °C and 5% CO₂^12,13. In both cases, phagocytosis was then evaluated by flow cytometry and fluorescence microscopy.

RNA isolation and qPCR

Total cellular RNA was extracted via a Quick-RNA MiniPrep Kit (Zymo Research), and 1 µg of RNA was used to perform reverse transcription via Improm-II Reverse Transcriptase (#M314A, Promega, USA). Real-time PCR was conducted with FastStart Universal SYBR Green Master (ROX) (#48869500, Sigma-Aldrich-USA) as a DNA-binding fluorescent dye via a StepOne Real-Time PCR System (Applied Biosystems, USA). The primer sets used for amplification were as follows: GAPDH (F: 5’-TTACTCCTTGGAGGCCATGT-3’, R: 5’-CGACCACTTTGTCAAGCTCA-3’), annealing temperature (AT°) 56 °C; CD300e (F: 5’-GTTTCCCCAGCAATTACAACCC-3’, R: 5’-CAGAAGACAGCACCCAGCAT-3’) AT° 60 °C; VEGFA (F: 5’-CATGCCAAGTGGTCCCAGG-3’. R: 5’-GCTGGCTTTGGTGAGGTTTG-3’) AT° 60 °C; RGS2 (F: 5’-AAGAGCGAGGAGAAGCGAG-3’, R: 5’-GCAAGACCATATTTGCTGGCT-3’) AT° 60 °C; CD93 (F: 5’-TGGAGAACCAGTACAGTCCG-3’, R: 5’-GAGTCACGAAATCCCCACCG-3’) AT° 60 °C; iNOS (F: 5’-TCATCCGCTATGCTGGCTAC-3’, R: 5’-CCCGAAACCACTCGTATTTGG-3’) AT° 58 °C; Arg-1 (F: 5’-ACGGAAGAATCAGCCTGGTG-3’, R: 5’-ATCAGTGTGAGCATCCACCC-3’) AT° 56 °C. Cycle thresholds (Ct) were normalized to the Ct of GAPDH, and fold enrichments were calculated compared with the values from unstimulated control cells. Three technical replicates were performed for each sample.

T lymphocyte proliferation assay

Freshly isolated PBMCs were obtained as described previously. Naïve CD4⁺ T cells were enriched from freshly isolated PBMCs via the Naïve CD4⁺ T Cell Isolation Kit (#130-094-131, Miltenyi Biotec) via MACS negative selection following the manufacturer’s instructions. The purity of the isolated cells exceeded 90%. PBMCs or naïve CD4⁺ T cells were labeled with CFSE (2 µM, Invitrogen) following the manufacturer’s instructions. Appropriate staining controls were performed for each donor. Specifically, to confirm the absence of a CFSE-negative population, a sample of CFSE-labeled lymphocytes was analyzed by flow cytometry at Day 0. Then, cells were cocultured with allogeneic macrophages differentiated in the presence of B. abortus antigens (or not, as a control) in complete medium. Under all conditions, the medium was supplemented with recombinant IL-2 (10 U/ml, #554603, BD Pharmingen). Twenty-four hours later, IFN-γ was measured in culture supernatants from different experiments via sandwich ELISA with paired cytokine-specific antibodies according to the manufacturer’s instructions (#555142, BD Pharmingen). Three days later, the cells were labeled via surface staining with the following antibodies for 30 min at 4 °C: Pacific Blue anti-human CD4 (#558116, BD Pharmingen) and APC anti-human CD3 (#300412, BioLegend, USA). The proliferation of CD3⁺CD4⁺ T cells and CD3⁺CD4⁻, and thus CD8⁺ T cells, was assessed via flow cytometry, and the percentage of cells that lost the CFSE label due to cell division was determined.

Statistical analysis

The experiments were performed at least 3 times with different cell donors. Data analysis was performed with GraphPad Prism 8.0 (GraphPad Software, USA). Statistical differences were assessed by paired analysis of variance (ANOVA) followed by the Wilcoxon matched-pairs test or t test. Differences with a p value < 0.05 were considered significant. The data are presented as the means ±SEMs.

Author contributions

ADG, JS and AMR conceived and designed the experiments. ADG, MA, JR and AJBB performed the experiments. ADG, AJBB, JS and AMR analyzed the data and wrote sections of the manuscript. JAS, VAD and GHG supported the work with key suggestions and helped with data interpretation and funding acquisition. AMR supervised the experiments, interpreted the data, acquired the funding and wrote the manuscript. All the authors read, reviewed and approved the final manuscript.

Funding

This research was funded by grants from the Agencia Nacional de Promoción Científica y Tecnológica (ANPCYT-Argentina) (PICT 2017 − 1905, 2017 − 1393, 2019 − 1178, 2021 − 00218), UBACYT from the University of Buenos Aires (20020170100320BA) (Argentina), and the National Science Foundation (NSF)-HBCU-UP (HRD-1911660) grant to VAD. ADG and AJBB are recipients of a fellowship from ANPCYT (Argentina). JR is a recipient of a fellowship from CONICET (Argentina). JS, GHG, and AMR are members of the Research Career of CONICET (Argentina).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Competing interests

The authors declare no competing interests.