Mycobacterium tuberculosis H37Rv strain was grown at 37°C in Middlebrook 7H9 medium (Difco) supplemented with 10% albumin-dextrose-catalase (Difco) and 0.05% Tween-80 (Sigma-Aldrich). The Mtb γ-irradiated H37Rv strain (NR-49098) and its total lipids’ preparation (NR-14837) were obtained from BEI Resource, USA.

Buffy coats from healthy donors were prepared at Centro Regional de Hemoterapia Garrahan (Buenos Aires, Argentina) according to institutional guidelines (resolution number CEIANM-664/07). Informed consent was obtained from each donor before blood collection. Peripheral blood mononuclear cells were obtained by Ficoll gradient separation on Ficoll-Paque (GE Healthcare). Then, monocytes were purified by centrifugation on a discontinuous Percoll gradient (Amersham) as previously described (18). After that, monocytes were allowed to adhere to 24-well plates (Costar) at 5 × 10⁵ cells/well for 1 h at 37°C in warm RPMI-1640 medium (GIBCO). The cells were then washed with warm PBS twice. The final purity was checked by fluorescence-activated cell sorting analysis using an anti-CD14 monoclonal antibody (mAb) and was found to be >90%. The medium was then supplemented to a final concentration of 10% fetal bovine serum (FBS, Sigma-Aldrich) and human recombinant Macrophage Colony-Stimulating Factor (M-CSF, Peprotech) at 10 ng/ml. Cells were allowed to differentiate for 5–7 days.

Pleural effusions were obtained by therapeutic thoracentesis by physicians at the Hospital F. J Muñiz (Buenos Aires, Argentina). The research was carried out in accordance with the Declaration of Helsinki (2013) of the World Medical Association, and was approved by the Ethics Committees of the Hospital F. J Muñiz and the Academia Nacional de Medicina de Buenos Aires (protocol number: NIN-1671-12). Written informed consent was obtained before sample collection. We collected a total of 43 PE samples which were classified according to their etiology, being 38 of them associated to TB (TB-PE) and 5 to heart failure (HF-PE). Individual samples of confirmed TB patients were used for correlation analysis. A group of samples (n = 10) were pooled and used for in vitro assays to treat macrophages. The selection of these samples was based merely on practical reasons, involving those samples that have been collected earlier through the course of this study. The diagnosis of TB pleurisy was based on a positive Ziehl–Nielsen stain or Lowestein–Jensen culture from PE and/or histopathology of pleural biopsy, and was further confirmed by an Mtb-induced IFN-γ response and an ADA-positive test (19). Exclusion criteria included a positive HIV test, and the presence of concurrent infectious diseases or non-infectious conditions (cancer, diabetes, or steroid therapy). None of the patients had multidrug-resistant TB. Those PE samples derived from patients with pleural transudates secondary to heart failure (HF-PE, n = 5) were employed to prepare a second pool of PE, used as control of non-infectious inflammatory PE. The PE were collected in heparin tubes and centrifuged at 300 g for 10 min at room temperature without brake. The cell-free supernatant was transferred into new plastic tubes, further centrifuged at 12,000 g for 10 min and aliquots were stored at −80°C. After having the diagnosis of the PE, pools were prepared by mixing same amounts of individual PE associated to a specific etiology. The pools were decomplemented at 56°C for 30 min, and filtered by 0.22 µm in order to remove any remaining debris or residual bacteria.

Macrophages were plated on glass coverslips within a 24-well tissue culture plate (Costar) at a density of 5 × 10⁵ cells/ml per well with or without 20% v/v of PE, 10 µg/ml of Mtb lipids (BEI resources) or infected with Mtb (MOI 2:1) for 24 h. When indicated, cells were pre-incubated with either Cucurbitacin I (50–100 nM, Sigma-Aldrich), or the STAT3 inhibitor Stattic (1–20 µM, Sigma-Aldrich), or the ACAT inhibitor Sandoz 58-035 (5–50 µg/ml, Sigma-Aldrich) for 2 h prior TB-PE addition and for further 24 h during TB-PE incubation. DMSO alone was used as control. Alternatively, cells were treated with recombinant human IL-10 (Peprotech) at the indicated doses. Foam cell formation was followed by Oil Red O (ORO) staining (Sigma-Aldrich) as previously described (20) at 37°C for 1–5 min, and washed with water three times. For the visualization of the lipid bodies, slides were prepared using the aqueous mounting medium Poly-Mount (Polysciences), observed via light microscope (Leica) and finally photographed using the Leica Application Suite software. Alternatively, after fixation, cells were labeled with 1 µg/ml of BODIPY 493/503 (Life technologies) for 15 min in order to visualize the lipid bodies by green fluorescence emission using a confocal microscope (Olympus BX51).

Infections were performed in the biosafety level 3 laboratory at the Unidad Operativa Centro de Contención Biológica, ANLIS-MALBRAN (Buenos Aires), according to the biosafety institutional guidelines. Macrophages seeded on glass coverslips within a 24-well tissue culture plate (Costar) at a density of 5 × 10⁵ cells/ml were infected with Mtb H37Rv strain at a MOI of 2:1 during 1 h at 37°C. Then, extracellular bacteria were removed gently by washing with pre-warmed PBS, and cells were cultured in RPMI-1640 medium supplemented with 10% FBS and gentamicin (50 µg/ml) for 24 h. The glass coverslips were fixed with PFA 4% and stained with ORO, as was previously described.

Total cholesterol was determined in TB-PE or cell lysates using the Colestat Enzimatico kit according to manufacturer instructions (Wiener Lab, Argentina). This assay is based in Trinder reaction in which cholesterol in the sample is quantified by enzymatic hydrolysis of cholesterol esters (21).

Total lipids were extracted from the same number of macrophages with methanol/chloroform (2:1 v/v) as described by Bligh and Dyer (22). After extraction, lipids were dried and analyzed by thin layer chromatography on silica gel 60 F254 plates (Merck), using hexane/diethyl ether/acetic acid (75:25:1, v/v/v) as the developing solvent. Chemical staining with Cu-phosphoric was used for detection.

Macrophages were centrifuged for 7 min at 1,200 rpm and then stained for 40 min at 4°C with fluorophore-conjugated antibodies FITC-anti-CD36 (clone 5-271), PerCP.Cy5.5-anti-CD14 (clone HCD14), PE-anti-CD163 (clone GHI/61), FITC-anti-CD206 (clone C068C2) (Biolegend), FITC-anti-HLA-DR (clone G46-6), or PE-anti-CD274 (clone MIH1) (BD Biosciences), and in parallel, with the corresponding isotype control antibody. After staining, the cells were washed with PBS 1×, centrifuged and analyzed by flow cytometry using FACSCalibur cytometer (BD Biosciences). The median fluorescence intensity were analyzed using FCS Express V3 software (De Novo Software, Los Angeles, CA, USA). The mean of the median fluorescence intensities (MFI) of the independent assays were used for comparisons between experimental conditions.

The amounts of human TNF-α, IL-1β, IL-10, IL-6, IFN-γ, and IL-4 were measured by ELISA, according to manufacturers’ instructions kits (TNF-α, IL-4, and IFN-γ Ready-SET-Go!™ Kits from eBioscience; IL-10, IL-1β, and IL-6 ELISA MAX™ Deluxe Kits from Biolegend). The detection limit was 3 pg/ml for TNF-α; 8 pg/ml for IL-10 and IL-6; 6.25 pg/ml for IFN-γ and IL-4; and 15.6 pg/ml for IL-1β. Murine TNF-α and IL-10 were measured by ELISA, according to manufacturer’s instructions (OptEIA™ Set kits from BD Bioscience) with a detection limit at 30 pg/ml.

Tuberculosis-PE were incubated with 10 µg/ml of the following neutralizing antibodies for 1 h at 4°C for the specific depletion of IL-10 (αIL-10, clone 19F1; Biolegend), IL-6 (αIL-6, clone MQ2-13A5; BD Bioscience), IL-1β (αIL-1β, clone 8516.311; SIGMA), or TNF-α (αTNF-α, clone MAb1; Biolegend). Then, 100 µl/ml of Protein G Sepharose beads (Amersham) were added and incubated for 1 h at 4°C. Finally, TB-PE were centrifuged at 12,000 g to remove antibody-bead complexes and then filtered (0.22-µm pores) before use. IFN-γ depletion was performed by incubating TB-PE for 2 h in sterile 96-well plates that had been coated with the capture antibody provided by the Human IFN-γ ELISA Kit (BD Bioscience). In all cases, depletions were controlled by ELISA.

Macrophages exposed (or not) to TB-PE, and depleted (or not) for IL-10, were prepared for transmission electron microscopy study. For this purpose, cells were centrifuged at 6,000 rpm for 1 min, fixed in 1% glutaraldehyde dissolved in 0.1 M cacodylate buffer (pH 7); post-fixed in 2% osmium tetroxide; dehydrated with increasing concentrations of ethanol and gradually infiltrated with Epon resin (Pelco). Thin sections were contrasted with uranyl acetate and lead citrate (Electron Microscopy Sciences, Fort Washington, PA, USA) and examined with a FEI Tecnai transmission electron microscope (Hillsboro, OR, USA). For morphometry, 30 cells from each condition were randomly selected and digitalized at 40,000× magnification. Then the area of lipid electron dense vacuoles per cell were measured and compared in each experimental group by automated morphometry.

We purified and maintained CD4 T cells from blood derived from healthy PPD⁺ donors as previously demonstrated (23). In parallel, we cultured autologous monocytes to generate macrophages and exposed them or not to TB-PE 20% v/v for 24 h. The next day, the medium was replaced, and macrophages were loaded with mycobacterial antigens by adding γ-irradiated Mtb at 2 bacteria to 1 macrophage ratio for further 24 h. Thereafter, the medium was replaced and autologous carboxyfluorescein succinimidyl ester (CFSE)-labeled CD4 T cells (Invitrogen) were added at a ratio of 10 lymphocytes to 1 macrophage for 5 days, as detailed previously (24). To determine IFN-γ production among proliferating CD4 T cells (CFSE^low), brefeldin A (5 µg/ml; Sigma Chemical Co.) was added 4 h prior the end of the coculture to block cytokine secretion. Thereafter, CFSE-labeled lymphocytes were first stained with anti-CD4-PerCP-Cy5.5 mAbs (eBioscience), then fixed with 0.5% PFA for 15 min. Second, cells were permeabilized with 500 µl Perm2 (Becton Dickinson, Cockysville, MD, USA) for 10 min and incubated with anti-IFN-γ-PE mAb (Invitrogen, CA, USA). Cells were gated according to its FSC and side scatter (SSC) properties analyzed on FACScan (Becton Dickinson). In order to gate out dead lymphocytes, the gate of CD4 T cells with increased SSC and low FSC was excluded (25). Isotype matched controls were used to determine auto-fluorescence and non-specific staining. Analysis was performed using the FCS Express (De Novo Software) and results were expressed as percentages of IFN-γ^pos/CFSE^low/CD4^pos T cells induced by the different macrophage populations. To complement these results, the amounts of IFN-γ released in the supernatant throughout the coculture were determined by ELISA according to the manufacturer’s kit indications (BD Bioscience).

Macrophages exposed (or not) to TB-PE, were infected with H37Rv Mtb strain at a MOI of 1 bacteria/cell in triplicates. After 2 h, extracellular bacteria were removed by gently washing four times with pre-warmed PBS. At 4 h and days 3, 6, and 10, cells were lysed in 0.1% SDS and neutralized with 20% Bovine Serum Albumine in Middlebrook 7H9 broth. Serial dilutions of the lysates were plated in triplicate, onto 7H11-Oleic Albumin Dextrose Catalase (OADC, Difco) agar medium for CFU scoring at 21 days later.

Macrophages were treated (or not) with TB-PE. Following the different experimental treatments, cells were lysed in ice-cold buffer consisting of 150 mM NaCl, 10 mM Tris, 5 mM EDTA, 1% SDS, 1% Triton X-100, 1% sodium deoxycholate, gentamicin/streptomycin, 0.2% azide plus a cocktail of protease inhibitors (Sigma-Aldrich). Lysates were incubated on ice for 3 h and cleared by centrifugation for 15 min at 14,000 rpm at 4°C. Protein concentrations were determined using the BCA protein assay (Pierce). Equal amounts of protein (40 µg) were then resolved on a 10% SDS-PAGE. Proteins were then transferred to Hybond-ECL nitrocellulose membranes (Amersham) for 2 h at 100 V and blocked with 1% BSA-0.05% Tween-20 for 1 h at room temperature. Membranes were then probed with primary anti-human ACAT (1:200 dilution, SOAT; Santa Cruz) or anti-human pY705-STAT3 (1:1,000 dilution, Cell Signaling Technology, clone D3A7) overnight at 4°C. After extensive washing, blots were incubated with a HRP-conjugated goat anti-rabbit IgG Ab (1:5,000 dilution; Santa Cruz Biotechnology) or HRP-conjugated goat anti-mouse IgG Ab (1:2,000 dilution; Santa Cruz Biotechnology) for 1 h at room temperature. Immunoreactivity was detected using ECL Western Blotting Substrate (Pierce). Protein bands were visualized using Kodak Medical X-Ray General Purpose Film. For internal loading controls, membranes were stripped by incubating in buffer consisting of 1.5% Glycine, 0.1% SDS, 1% Tween-20, pH 2.2 for 10 min twice, extensively washed and then reprobed with anti-β-actin (1:2,000 dilution; ThermoFisher, clone AC-15) or anti-STAT3 Ab (1:1,000 dilution; Cell Signaling Technology, clone D1A5). Results from Western blot were analyzed by densitometric analysis (Image J software).

Macrophages were plated on glass coverslips and were treated with or without 20% v/v of TB-PE for 24 h. Cells were fixed with PFA 4% for 20 min at room temperature and then PFA was quenched with 50 mM NH₄Cl for 2 min. Cells were rinsed in PBS once and then were labeled with 1 µg/ml of BODIPY 493/503 (Life technologies) for 15 min before permeabilization with PBS-Triton X-100 0.1% for 10 min. Cells were then incubated with PBS-BSA 3% w/v for 30 min prior to overnight incubation at 4°C with primary anti-human pY705-STAT3 (dilution 1/100, Cell Signaling Technology, Clone D3A7). Cells were then washed and incubated with Goat anti-Mouse IgG, AlexaFluor 555 (dilution 1/1,000, Cell Signaling Technology) for 1 h at room temperature. Cells were extensively washed and then incubated for 10 min with DAPI in PBS-BSA 1% (500 ng/ml, Sigma-Aldrich). Finally, slides were mounted and visualized with a Leica DM-RB fluorescence microscope.

Wild-type (WT) BALB/c and IL-10 knockout (IL-10 KO) (C.129P2(B6)-IL-10 tm1Cgn/J) BALB/c male mice (8–12 weeks old), were obtained from the Leloir Institute Foundation. Animals were bred and housed in accordance with the guidelines established by the Institutional Animal Care and Use Committee of Institute of Experimental Medicine (IMEX)-CONICET-ANM. All animal procedures were shaped to the principles set forth in the Guide for the Care and Use of Laboratory Animals (26).

Femurs and tibia from WT and IL-10 KO mice were removed after euthanasia and the bones were flushed with RPMI-1640 medium by using syringes and 25-Gauge needles. The cellular suspension was centrifuged and the red blood cells were removed. The BMDMs were obtained by culturing the cells with RPMI-1640 medium containing l-glutamine, pyruvate, β-mercaptoethanol (all from Sigma-Aldrich), 10% FCS and 20 ng/ml of murine recombinant M-CSF (Biolegend) at 37°C in a humidified incubator for 7–8 days. Differentiated BMDMs were re-plated into 24-well tissue culture plates in complete medium prior to cell stimulation with Mtb lipids for 24 h. Alternatively, recombinant murine IL-10 (Peprotech) at 10 ng/ml was added to the cultures for 24 h.

All values are presented as mean and SEM of a number of independent experiments. Independent experiments are defined as those performed with macrophages derived from monocytes isolated independently from different donors. As most of our datasets did not pass the normality tests, non-parametric tests were applied. Comparisons between more than two paired data sets were made using the Friedman test followed by Dunn’s Multiple Comparison Test. Comparisons between two paired experimental conditions were made using the two-tailed Wilcoxon Signed Rank. Correlation analyses were determined using the Spearman’s rank test. For all statistical comparisons, a p value < 0.05 was considered significant.

It has been demonstrated that Mtb induces a lipid-rich foam cell phenotype in host macrophages (6–8, 27), possibly via TLR2 and TLR6 activation (28). Moreover, merely isolated lipids from Mtb were able to induce the foamy phenotype (7, 27). Based on this knowledge, we aimed to determine whether soluble factors found in physiological samples derived from active TB patients could promote the generation of FM. To this end, we treated human M-CSF-driven macrophages with cell-free preparations of PE derived from patients with tuberculosis (TB-PE) in order to mimic a genuine microenvironment derived during Mtb infection. According to the pattern of staining of neutral lipids with ORO, we observed that TB-PE treatment induced lipid bodies accumulation in macrophages to the same extent as Mtb infection or exposure to Mtb-derived lipids (Figure 1A). We also demonstrated that the formation of lipid bodies was specific for TB-PE treatment in comparison to PE from patients with heart failure (HF-PE) (Figure 1B). Moreover, biochemical analysis of the lipids recovered from macrophages identified higher abundance of cholesteryl esters and triacylglycerols in TB-PE-treated macrophages in comparison to control or HF-PE-treated macrophages (Figure 1C). Additionally, unlike HF-PE, the treatment with TB-PE resulted in an increase of the total intracellular cholesterol content (Figure 1D) and a twofold increase of CD36 expression (Figure 1E). Therefore, soluble factors released during pleural Mtb infection induced lipid body accumulation in human macrophages that typically confirm the FM phenotype.

In order to assess whether the differentiation into FM induced by TB-PE may have a negative impact on the development of the antimycobacterial response, we determined phenotype and functions of TB-PE-induced FM. As observed in Figure 2A, macrophages treated with TB-PE displayed a higher expression of anti-inflammatory markers such as CD163, mannose receptor or CD206 (MRC1), and PD-L1, and a lower expression of the MHC class II cell-surface receptor, HLA-DR. In accordance with the acquisition of this anti-inflammatory profile, TB-PE-induced FM secreted higher levels of IL-10 and lower levels of TNF-α upon stimulation with irradiated Mycobacterium tuberculosis (iMtb) than untreated cells (Figure 2B). In order to assess the effect of TB-PE treatment on the presentation of mycobacterial antigens, macrophages were treated with or without TB-PE for 24 h, washed and stimulated with iMtb for further 24 h. Thereafter, cells were washed and cocultured with autologous CFSE-labeled CD4 T cells derived from healthy PPD positive subjects for 5 days. Based on the CFSE labeling of CD4 T cells, while TB-PE-treated macrophages did not induce differential levels of proliferation of antimycobacterial CD4 T cells (Figure 2C), they promoted differential distribution of the IFN-γ producing clones (Figure 2D). In particular, those macrophages treated with TB-PE induced lower percentages of IFN-γ producing clones (Figure 2D) and a lower release of IFN-γ (Figure 2E), as compared to untreated macrophages in response to iMtb stimulation. Of note, IL-4 and IL-17 contents were undetectable in these assays (data not shown). Therefore, the activation of Ag-specific IFN-γ-producing CD4 T cells was impaired when macrophages were exposed to TB-PE, suggesting that the accumulation of lipid bodies is accompanied by a reduced capacity to activate antimycobacterial Th1 cells.

Next, we assessed whether the treatment with TB-PE had an impact on the control of the bacillary load. To accomplish this, macrophages were treated with TB-PE for 24 h, washed and infected with Mtb. Although no differences were observed in the uptake of the mycobacteria as judged by the scoring of the CFU at 4 h post-infection (Figure 2F), a significant increase in the bacillary load was observed in TB-PE-treated macrophages at later time points (Figure 2G). Therefore, FM induced by TB-PE are more susceptible to Mtb intracellular growth.

In order to evaluate the host factors involved in promoting the accumulation of lipid bodies by TB-PE, we depleted different cytokines known to be highly present in this fluids (29, 30), including IL-10, TNF-α, IL-1β, IL-6, and IFN-γ. We then evaluated the ability of these depleted-PE to induce the foamy phenotype in macrophages. The depletion of each individual cytokine was confirmed by ELISA (Figure S1A in Supplementary Material). Interestingly, only the depletion of IL-10 from TB-PE was capable of preventing lipid bodies accumulation (Figure 3A). This result was also confirmed by labeling the cells with Bodipy staining (Figure S1B in Supplementary Material). In addition, macrophages exposed to IL-10-depleted TB-PE showed a reduction of intracellular cholesterol content and CD36 cell-surface expression in comparison to those cells treated with non-depleted or depleted of any other cytokines (Figures 3B,C). In line with these results, the addition of recombinant IL-10 to the IL-10-depleted TB-PE restored the acquisition of the foamy phenotype in a dose-dependent manner (Figure 3D). Of note, the addition of IL-10 in the absence of TB-PE did not induce the foamy phenotype (Figure 3D). Thereafter, we determined the presence of the lipidic vacuoles by electron microscopy in macrophages treated with TB-PE depleted (or not) of IL-10. As it is shown in Figure 3E, while untreated macrophages showed numerous pseudopodia and empty cytoplasmic vacuoles, TB-PE-treated macrophages displayed electron dense lipid osmiophilic vacuoles, which correspond to lipid bodies. Interestingly, macrophages treated with IL-10-depleted TB-PE showed smaller-sized lipidic vacuoles than those exposed to non-depleted TB-PE. These results indicate that the IL-10 present in TB-PE is a key host factor promoting the lipid-laden phenotype in the presence of exogenous lipids.

In order to evaluate whether IL-10 level is associated to the acquisition of the foamy phenotype in the course of a Mtb natural infection, we assessed the levels of IL-10 and total cholesterol in individual preparations of TB-PE, and the numbers of FM found within the pleural fluids mononuclear cells (PFMC). We found that levels of IL-10 and total cholesterol were positively correlated (Figure 4A), unlike other cytokines such as IFN-γ, IL-6, TNF-α, and IL-1β (Figure S2 in Supplementary Material). Noticeably, there was also a positive correlation between the level of IL-10 and the number of FM among the PFMC (Figure 4B). Moreover, lipid-laden CD14⁺ cells were found in the pleural compartment but not in paired peripheral blood (Figure 4C). These results support the idea that IL-10 potentiates the acquisition of the foamy program in human macrophages in the context of a natural infection.

To confirm the role of IL-10 in the differentiation of FM, we evaluated whether BMDM derived from IL-10-knock out (KO) mice could indeed become FM after the treatment with lipids from Mtb. First, we determined IL-10 production in M-CSF-derived BMDM from WT mice stimulated with mycobacterial lipids for 24 h. As shown in Figure 5A, unlike the undetectable level of TNF-α (data not shown), BMDM secreted IL-10 in response to mycobacterial lipid-stimulation. Second, we compared whether BMDM derived from WT or IL-10-KO mice differed in their propensity to accumulate lipid bodies in response to Mtb-derived lipids. As illustrated in Figures 5B,C, IL-10-deficiency prevented the foamy phenotype induced by Mtb lipids, which in turn could be partially reverted by the addition of exogenous IL-10. In agreement with our previous results, exogenous IL-10 did not induce the foamy phenotype in the absence of the source of lipids in BMDM. These results obtained in murine macrophages confirm those obtained in human TB-PE-treated macrophages, demonstrating the key role of IL-10 in favor of the differentiation program toward foamy cells.

Considering the role of IL-10 in promoting the differentiation of macrophages into FM, and that STAT3 is a pivotal transcriptional factor induced by IL-10 (31, 32), we studied the contribution of STAT3 activation in the induction of the foamy phenotype by measuring its phosphorylated form (pSTAT3). First, we determined that STAT3 was activated in TB-PE-treated macrophages detecting its phosphorylated form by western blot and immunofluorescence microscopy (Figures 6A,B). As depicted in Figure 6B FM induced by TB-PE showed nuclear localization of STAT3 phosphorylated on tyrosine 705, reflecting STAT3 activation. Importantly, pharmacological inhibition of STAT3 with Stattic (or with cucurbitacin I) prevented the accumulation of lipid bodies in macrophages in a dose-dependent manner (Figure 6C; Figures S3A,B in Supplementary Material). Therefore, the enhancement of FM differentiation driven by IL-10 is mediated by STAT3.

In order to elucidate the mechanism by which IL-10 promotes FM differentiation, we assessed the expression of ACAT, which is central for the biogenesis of lipid bodies by converting free- into esterified-cholesterol that are eventually packed inside the lipid droplets (33). We first showed that the foamy phenotype induced by TB-PE was dependent on ACAT activity, as judged both by the increase of ACAT expression after TB-PE treatment (Figure 7A), and by the blocking of the accumulation of lipid bodies in the presence of Sandoz, a specific inhibitor of ACAT (Figure 7B). Noticeably, the inhibition of ACAT lead to reduced amounts of both cholesteryl esters and triacylglycerols in TB-PE-treated macrophages (Figure 7C), confirming the involvement of ACAT in FM formation upon TB-PE treatment. We then assessed ACAT expression in IL-10-depleted TB-PE and we found that its expression was reduced in the absence of IL-10 (Figure 7A). In agreement with this result, ACAT induction by TB-PE was abolished when STAT3 activity was inhibited (Figure 7D). Therefore, the IL-10/STAT3 axis activated by TB-PE enhances FM differentiation through the upregulation of ACAT expression, leading to an increased biogenesis and accumulation of lipid bodies.

Based on our findings, we propose a model for the modulation of FM in the context of a physiologically relevant microenvironment promoted by Mtb infection for which the axis IL-10/STAT3 induces the accumulation of lipid bodies via ACAT upregulation. This in turn is accompanied by an increase of CD36 and the acquisition of immunosuppressive properties, such as a reduced induction of antimycobacterial Th1 clones, an enhanced production of IL-10 and a more permissive phenotype for bacillary growth (Figure 8).