Oil Red O (ORO/O0625), adenosine (ADO/A9251), adenosine 5′-(α,β-methylene) diphosphate (AMP-CP/M3763), adenosine 5′-monophosphate disodium salt (5′ AMP/01930), 8-[4-[((4-cyanophenyl)carbamoylmethyl)oxy]phenyl]-1,3-di (n-propyl)xanthine hydrate (MRS 1754/M6316), CGS21680 (ab120453), MRS1754 (264622-58-4), ZM 241385, Tris (hidroximetil)-aminometano (Tris-HCL/93363), NP-40 (492016), sodium deoxycholate (SDS/D6750) and [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] (MTT/M6494), TritonX-100 (HFH10), bovine serum albumin (BSA/Sigma A2153), trichloroacetic acid (TCA/ThermoFisher), and potassium phosphate (KH2PO4/P9791) were purchased from Sigma-Aldrich, USA. It was also purchased: cOmplete™ EDTA-free Protease Inhibitor Cocktail (04693132001) from Roche, United States; anti-ADA (sc-28346) and anti-GAPDH (sc-32233) from Santa-Cruz, USA; LIVE/DEAD™ Bac Light™ Bacterial Viability kit, Bicincichonic acid kit, 4′,6-Diamidino-2-fenilindol,2-(4-Amidinofenil)-6-indolecarbamidina (DAPI/D1306), Prolong Gold Antifade Mountant (36980), RPMI 1640 medium, (2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), L-Glutamine, MemCode™ Reversible Protein Stain Kit (24580), TRIzol^®, SYBR Green^® PCR master mix, and a DNA-free turbo kit from ThermoFisher Scientific, United States; fetal bovine serum (FBS 0010S/Cripion Biotecnology), anti-mouse CD73 (Ab-54217), anti-mouse Alexa Fluor®-488 (ab150117), and anti-rabbit Alexa Fluor®-594 (ab150080) from Abcam, United Kingdom; anti-rabbit A_2A receptor (AAR002), anti-rabbit A_2B (AAR-004), anti-rabbit A₁ (AAR-006), and anti-rabbit A₃ (AAR-004) from Alomone, anti-p-CREB (ab32096); anti-CREB (ab31387) from Abcam, United States; GoScript Master Mix from Promega.

The live M. leprae Thai-53 strain used throughout the study was isolated from the hind footpads of athymic nu/nu mice kindly provided by Dr. Patrícia Sammarco Rosa and Ms. Daniele Bertolucci of Instituto Lauro de Souza Lima, Bauru, SP, Brazil. Briefly, about 9 months after inoculation of the bacilli (with a bacillary load of about 10⁹–10¹⁰/g of tissue), the mice were euthanized, and M. leprae was purified from their footpads (Trombone et al., 2014). Viability was checked by microscopy using LIVE/DEAD™ Bac Light™ Bacterial Viability Kit following the manufacturer’s guidelines, and the bacilli were quantified before in vitro experiments (Trombone et al., 2014). All procedures were conducted under the Laboratory Animals Welfare Act, CONCEA Guidelines, and Policies for Rodent experiments provided by Ministry of Science and Technology of Brazil and the Institutional Animal Care and Use Committee (Approval CEUA-ILSL 00).

Human Schwann cells (SCs) from the ST88-14 lineage were used in this study. This SC line was derived from patients with neurofibromatosis type 1 and was generously provided by Dr. Jonathan A. Fletcher from the Department of Pathology at Brigham and Women’s Hospital, Harvard Medical School, Boston, USA. ST88-14 cells were cultured in RPMI 1640 medium supplemented with 10% FBS, 10 mM HEPES, and 1 mM L-Glutamine, pH 7.4. Cultures were maintained at 37 °C with 5% CO₂ in 150 cm² flasks until they reached approximately 80% confluence.

SCs were suspended in RPMI 1640 culture medium in 2% FBS without antibiotics and seeded at a density of 5 × 10³ cell/well on 96-well plates for cell viability assays, 3 × 10⁵ cells/well on 6-well plates for RT-qPCR assays, and at 5 × 10⁴ cells/well on 24-well plates for enzymatic activity assays, fluorescence microscopy experiments, and Western blotting analysis. To investigate the desired outcomes, the cells were stimulated with different compounds 30 min prior to infection with live M. leprae. The compounds used were: 5′AMP (100 µM), a substrate of CD73 enzyme; AMP-CP (5 µM), a selective CD73 inhibitor; adenosine (100 µM) and CGS21680 (100 µM), non-selective and selective agonists of the A_2AR, respectively; ZM132485 (1 µM) and MRS1754 (1 µM), non-selective and selective antagonists of A_2AR and A_2BR, respectively. Cells were subsequently infected with M. leprae at a multiplicity of infection (MOI) of 5:1 or 50:1 and incubated at 33 °C with 5% CO₂. The biological responses were assessed at 12, 24, or 48 h post-infection according to the assay to be performed.

To evaluate drug toxicity in SCs in response to compounds used such as 5′AMP (100 µM), AMP-CP (5 µM), CGS21680 (100 µM), ZM132485 (1 µM), and MRS1754 (1 µM), the MTT assay was used. SCs were plated into 96-well plates (10³ cells/well). After 48 h, MTT was added, and the formation of formazan was measured (Mosmann, 1983).

For immunofluorescence analysis, 4 × 10⁴ SCs were plated on glass coverslips in RPMI supplemented with 10% FBS and infected or not with live M. leprae (MOI 5:1) for 24 or 48 h. The cells then were fixed with 4% paraformaldehyde (PFA). They were subsequently incubated with a permeabilization solution composed of phosphate buffer saline (PBS) 1X, 10% FBS, and 0.001% Triton X-100 and were incubated with blocking solution containing 1% BSA diluted in PBS 1X with 0,001% Triton. The expression of CD73, ADA, and A_2AR was evaluated using mouse monoclonal anti-CD73, rabbit polyclonal anti-ADA, and rabbit polyclonal anti-A_2AR, all at a dilution of 1:250 and incubated overnight at 4 °C. The slides were washed thrice with PBS and incubated with Alexa Fluor^® conjugated secondary antibodies (anti-mouse-Alexa Fluor^®-488 or anti-rabbit-Alexa Fluor^®-594) for 1 h at a dilution of 1:1000. Nuclear staining was done using DAPI labeling (1:100). Coverslips were mounted using prolong gold antifade mounting. Images were captured with a Zeiss Axio Observer inverted microscope using the Colibri illuminating system with Plan-Neofluar ×40 objectives (Zeiss, Oberkochen, Germany) equipped with AxioVision Rel. 4.8 software (Zeiss, Oberkochen, Germany). Quantification was performed using ImageJ software (Schneider et al., 2012) measuring the fluorescence area per cell number in each image.

To investigate LDs accumulation, SCs were plated in RPMI supplemented with 2% FBS. The cells were treated with ADO, CGS21680, ZM132485, or MRS1754 as described for the SC infection assays and infected or not with M. leprae (MOI 5:1) for 48 h. Cells were stained with Oil Red O (ORO) (Díaz Acosta et al., 2018). The coverslips were washed with distilled water, and the cells’ nuclei were labeled with DAPI. Coverslips were mounted using prolong gold antifade mounting. The images were observed under an Axio Observer Z1 fluorescence microscope equipped with the AxioVision Rel. 4.8 software. The images were made with a 40 × objective lens and at least ten fields per slide; a total of 200 cells per condition were photographed. The quantification of LDs was performed using ImageJ software, with the Macro Umbral plugin as previously described Díaz Acosta et al. (2018). Macro Umbral calculated the lipid droplet area which was divided by the number of cells present in the analyzed field to determine the LD quantification (LD area/cell). Graphs demonstrate the mean value of LD/cell per condition.

CD73 activity was measured by colorimetric method with malachite green by the rate of released inorganic phosphate (Pi) (Zanin et al., 2012; Allard et al., 2019). Specifically, 5 × 10⁴ SCs were plated on 24-well plates and infected or not with M. leprae (MOI 5:1 or 50:1) for 24 h at 33 °C with 5% CO₂, in the presence or not of the CD73 inhibitor AMPCP (5 µM). The supernatant was discarded, and the wells were thrice washed with reaction medium (phosphate free). The reaction was started by the addition of 500 µL of reaction medium containing 2 mM of CaCl₂, 120 mM NaCl, 5mM KCl, 10 mM glucose, 20 mM Hepes, pH 7.4 pH, and 5 mM 5′AMP, and they were incubated for 1 h at 25 °C. The cell incubation medium (500 μL) was removed and centrifuged at 5,000 × g for 5 min. The supernatant was transferred to a tube containing 500 µL of 5% (w/v) trichloroacetic acid (TCA) on ice. Inorganic phosphate (Pi) production was measured by absorbance of the phosphomolybdate-malachite at 630 nm. The standard curve of Pi was made with a solution of KH₂PO₄, and CD73 activity was calculated by subtracting the nonspecific 5′-AMP hydrolysis measured in the absence of cells. The concentration of Pi released in the reaction was determined by comparison with a standard curve. All samples were taken in biological triplicates, and the CD73 activity was expressed as pmol Pi × min^-1× 10⁻⁴ cells (Zanin et al., 2012; Allard et al., 2019).

To evaluate protein expression, SCs were plated as described for the SC infection assays, treated or not with CGS21680 and infected or not with M. leprae (MOI 50:1) for 48 h. The cells were lysed using a RIPA buffer (50 mM TRIS pH 7.5, 1% NP-40, 0.25% sodium deoxycholate, 0.1% SDS) supplemented with protease inhibitors. Protein concentrations were determined using the BCA Kit. For SDS-PAGE electrophoresis, 20 µg of protein was loaded onto a 12% polyacrylamide gel, and the efficiency of the transfer was confirmed by staining the nitrocellulose membrane with MemCode™ Reversible Protein Stain Kit. The primary antibodies used were anti-A_2AR, (dilution 1:500) and anti-p-CREB (1:500 dilution). Anti-GAPDH (dilution 1:1000) was used as the loading control. Primary antibodies were incubated overnight at 4 °C. After three sequential washes with TBS-Tween and TBS 5 min each, the HRP-conjugated secondary antibodies were incubated 1 h. Chemiluminescence detection was performed using the iBright™ CL1500 Imaging System (A44114), and the relative protein levels were determined using ImageJ software.

SCs were infected with M. leprae (MOI 5:1) overnight at 33 °C in the presence or absence of CGS21680 (100 µM); the mRNA was isolated using TRIzol^® according to the manufacturer’s instructions. Briefly, the nucleic acid content was separated, followed by precipitation of RNA and subsequent production of cDNA. To assess the integrity of the RNA samples, a 1.2% agarose gel electrophoresis technique was employed. Samples showing two corresponding distinct bands without any signs of degradation were considered suited for subsequent analysis. The RT-qPCR was performed to analyze differential gene expression. The sequences of primers are available in Supplementary Table S1. Reverse transcription was performed using GoScript Master Mix. The RT-qPCR reactions were performed in duplicates on a ViiA7 Real-Time PCR System (Applied Biosystems, USA) using Power SYBR Green Master Mix. For this, the cDNA samples were diluted to a concentration of 5 ng/μL, and PCR mix was prepared for application in a 384-well plate. The relative gene expression analysis was performed using adjusted efficiency. Briefly, raw data were exported from ViiA7 software and imported into LinRegPCR v.2022 (Ramakers et al., 2003). Data analysis was carried out using the comparative method ΔΔCT (Livak and Schmittgen, 2001) after normalization with the endogenous housekeeping gene RPL13. Normalized expression values were presented as Fold Change (FC). To take into account differences between biological replicates of NI samples, the mean of all biological replicates was determined along with the ratio between each biological replicate and this mean. Therefore, the NI condition is represented as a mean ± SD.

M. leprae viability was investigated by RT-qPCR as described by Martinez et al. (2009). Briefly, SCs pre-treated or not with CGS21680 (100 µM) were infected with M. leprae for 48 h. RNA was extracted as described in Section 2.10, and DNA was extracted as described in Ramakers et al. (2003). DNA was removed from the RNA samples using DNA-free turbo kit, and the RNA reverse transcription was performed using GoScript Master Mix according to the manufacturer’s instructions. The levels of 16S rRNA were calculated against 16S DNA. LinRegPCR software was used to calculate the PCR efficiency for each biological experiment (Martinez et al., 2009), and the normalization was made considering the efficiency corrections. The primers for the rRNA 16S were used for both cDNA and DNA: sense 5′-GAA ACT GCG AAT GGC TCA TTA AAT CA-3^’ and antisense 5^’- CCC GTC GGC ATG TAT TAG CTC T-3^’. Both cDNA and DNA were measured by TaqMan real-time PCR assay (TaqMan probe 16S 5^’-TGG TTC CTT TGG TCG CTC GCT CC-3^’). M. leprae viability was determined by the comparative Ct method described by Martinez et al. (2009), and viability was calculated as described by Díaz Acosta et al. (2018). Cycling and plate reading were conducted using the Viia7 machine with Real-Time PCR System software.

SCs cultures were infected by M. leprae (MOI 5:1), and the supernatants were collected after 48 h infection for ELISA analysis. The levels of interleukin 6 and 8 (IL-6 and IL-8) were assessed using ELISA with respective Duo-Set human ELISA kits (R&D Systems) following the manufacturer’s guidelines. Optical densities were measured by an Absorbance microplate reader (EON BioTek Instruments), and the data were analyzed using SoftMax R Pro Software version 5.3 (ThermoFisher).

The protein interactome network was generated used STRING (https://string-db.og) version 12.0. The search was done by multiple proteins using Homo sapiens databank. A full STRING network was selected as the network type, with a required score of medium confidence (0.400), and medium false discovery ratio (FDR) stringency corresponding to 5%. Markov clustering (MCL) was applied—a clustering algorithm for graphs that assigns proteins into families based on sequence similarity information with 3.0 of inflation parameters.

Data statistical analysis was conducted using Graph Prism software version 9.0.0, and statistical significance was determined by a two-tailed unpaired t-test for single comparisons or one-way ANOVA tests for multiple comparisons, followed by Bonferroni corrections. The statistical differences between the groups were inferred to be significant when *p < 0.05, **p < 0.01, and ***p < 0.001.

Extracellular ADO (eADO) signaling is regulated through different mechanisms, including CD73 and ecto-ADA, that convert extracellular 5′AMP to eADO and eINO, respectively (Huang et al., 2021). To investigate the impact of M. leprae infection on AdoS, 5′ AMP hydrolysis, the levels of Pi were measured via colorimetric assay (Zanin et al., 2012; Allard et al., 2019). Initially three multiplicity-of-infection (MOI) ratios (5:1, 10:1, and 50:1) were assayed. The results revealed a significant increase in extracellular 5′AMP hydrolysis in infected SCs (Figure 1A) regardless of MOI. M. leprae-infected SCs were 52.21 ± 8.553 pmol Pi × min^-1×10⁻⁴cells (MOI 5:1/p = 0.0004), 49.09 ± 25.27 pmol Pi × min^-1×10⁻⁴cells (MOI 10:1/p = 0.0422), and 60.02 ± 23.27 pmol Pi × min^-1×10⁻⁴cells (MOI 50:1/p = 0.0004) —about three to four times more than the not infected (NI) SCs (Figure 1A). Since no statistical difference was observed in the extracellular hydrolysis of 5′AMP between the different MOIs, the ratio 5:1 was adopted in the remaining experiments. To confirm that hydrolysis of extracellular 5′ AMP resulted from CD73 activity, a non-hydrolysable 5′AMP analogue, adenosine-5'-(α, β-methylene) diphosphate (AMP-CP) —a selective CD73 inhibitor—was used. When applied to the infected cells, it reduced Pi production by 46.8% (p = 0.019) (Figure 1B). AMP-CP’s cytotoxicity was evaluated by MTT assay and did not affect the viability of SCs (data not shown). In addition to CD73 activity, we also analyzed the impact of M. leprae infection on NT5E gene expression which encodes CD73, and we observed a 1.18 log₂ fold change compared to the NI condition (FC = 1.000 ± 0.3681 and M. leprae FC = 1.532 ± 0.3677) (p = 0.0074) (Figure 1C). Immunofluorescence showed diffuse immunostaining of CD73 on M. leprae-infected cells (MIF = 821.3 ± 250.9) with a 1.470-fold change compared to the NI condition (MIF = 555.8 ± 478.7) (p = 0.0170) (Figures 1D, E).

We also analyzed the impact of M. leprae infection on ADA gene expression. Figure 2A shows that infected SCs exhibited an increase of 20.9% in ADA gene expression (FC 1.714 ± 0.5254) in comparison to NI SCs (FC = 1.140 ± 0.3081) (p = 0.0436). ADA protein was analyzed by immunofluorescence 24 and 48 h post-infection (Figures 2B–F). No statistical difference at 24 h infection was found due to the variability between biological replicates. However, the estimation plot (Figure 2D) revealed that all M. leprae-infected biological replicates exhibited higher ADA levels than NI SCs (Figures 2B–D). At 48 h post cellular infection, ADA levels increased with a fold change of 3.37 (MIF = 1001 ± 381.1) compared to the NI condition (MIF = 296.3 ± 206.1) (p = 0.0480) (Figures 2E–G). Taken together, these data indicate that M. leprae infection upregulates both CD73 and ADA, responsible for metabolizing extracellular eAMP to eINO, potentially contributing to the reduction of eADO levels in infected SCs (Supplementary Tables S2–S4 compiles the data presented in this section).

Extracellular ADO activates a family of P1 receptors, such as A₁R, A_2AR, A_2BR, and A₃R, which have different affinity to eADO (Patritti-Cram et al., 2021). To understand the role of eADO during M. leprae infection, we compared the expression profile of these receptors in infected versus NI SCs. The experimental data showed that the ST88-14 SC lineage expresses the genes ADORA1, ADORA2A, and ADORA2B. However, there was no significant modulation of the genes ADORA1 and ADORA2B with M. leprae infection (Figures 3A, C). In contrast, ADORA2A gene expression was downregulated (Figure 3B). The difference between the means of NI versus M. leprae-infected SC was −0.3448 ± 0.1464, which represents a reduction of 34.5% (p = 0.0336) in ADORA2A gene expression in M. leprae-infected SCs (Figure 3C) (Supplementary Table S3). The gene ADORA3 was not detected in this cell line; to confirm this data, we used two primer pairs indicated in Supplementary Table S1. To confirm that M. leprae infection reduces A_2AR levels, immunofluorescence was performed 24 and 48 h post-infection. Consistent with the PCR assay, infection decreased cellular A_2AR by 55.19% (p = 0.0056) and 77.32% (p = 0.0098) at 24 and 48 h, respectively (Figures 4A–D). Western blotting analysis confirmed the downregulation by 63.98% of A_2AR expression 48 h post-infection in SCs, corroborating the PCR and immunofluorescence results (Figures 4E, F).

When activated A_2AR triggers cAMP-PKA-CREB signaling, it results in the phosphorylation of CREB (p-CREB). To investigate whether A_2AR could be active despite its downregulation by M. leprae infection, we used a specific A_2AR agonist, CGS21680, to elevate the level of p-CREB. Figures 4G and H show the results of two biological replicates 48 h post-infection; the expression of p-CREB decreased by 56.77% compared to NI (p = 0.048) (Supplementary Table S5). Nevertheless, when it was preincubated with CGS21680, that effect was reversed to the level of the NI condition.

A_2AR can influence cellular cholesterol homeostasis, decreasing LD formation and exhibiting an antiatherosclerosis role (Reiss et al., 2004; Bingham et al., 2010). We thus investigated the possibility that decreased A_2AR expression induced by M. leprae infection was related to the foamy phenotype observed in infected cells. In this context, the impact of A_2AR on the capacity of M. leprae to induce LD formation in SCs was measured. A_2AR was activated with eADO, and LD formation was measured 48 h post-infection. As expected, M. leprae increased LD accumulation with a 3.96-fold change (p = 0.0153) versus NI SCs. However, when the SCs were treated with eADO, M. leprae-induced LD accumulation was significantly inhibited by 73.9% (p = 0.0257) (Figures 5A, B). To certify that eADO was activating A_2AR, a synthetic A_2AR agonist, CGS21680, was used instead, and this inhibited M. leprae-induced LD accumulation at a similar level as eADO (61.40%, p = 0.0428). In contrast, a synthetic A_2AR antagonist, ZM241385, increased the levels of LDs induced by M. leprae infection by 78.62% (p = 0.0044) compared with M. leprae infected and untreated SC cultures. Furthermore, ZM241385 reversed the inhibitory effect of CGS21680 in M. leprae-infected cells, (Figures 6A, B). Considering that the EC50 of the compound CGS21680, determined in different studies, is in the nM range (Kull et al., 2000; Gao et al., 2020), we also evaluated the effect of CGS21680 at 50 nM on LDs accumulation induced by M. leprae. The result showed that 50 nM CGS21680 continues to inhibit the formation of LDs induced by M. leprae in infected SC and that ZM241385 also reversed this phenotype (Supplementary Figures S1A, B). To rule out a possible interference caused by a high concentration of CGS21680 on the A_2B receptor, we analyzed the effects of a specific A_2BR antagonist, MRS1754, on LD formation induced by M. leprae. However, we firstly checked the expression of A_2BR in M. leprae-infected SCs and confirmed that infection does not change A_2BR expression (Supplementary Figures S2A, B). Nonetheless, despite A_2BR not having its expression modulated by infection, RT-qPCR and immunofluorescence analysis revealed that this receptor is much more abundant in SCs infected by M. leprae than A_2AR, (Figures 3B, C, 4C, D; Supplementary Figures S3A, B). Despite its abundance, we found that MRS1754, unlike ZM241385, did not increase LD formation and was unable to reverse the inhibitory effect of CGS21680 on LD accumulation in M. leprae-infected SCs (Supplementary Figures S2A, B). Overall, this analysis provides valuable insights into the participation of A_2AR as a negative modulator of LD formation in M. leprae-infected SCs (Supplementary Table S6 compiles the data presented in this section).

Our research investigated the crucial role of the A_2AR receptor in the molecular mechanisms of M. leprae infection. This receptor, when activated, reverses the formation of LDs induced by M. leprae in infected SCs. To explore this, we focused on lipid metabolism genes upregulated by M. leprae infection and genes involved in cholesterol efflux upregulated by A_2AR. Before solidifying our hypothesis, we constructed an interaction network using STRING to assess the interconnection between these genes. We included the cytokines IL-6 and IL-8, modulated by M. leprae infection and also involved in lipid homeostasis (Supplementary Figures S4A–C).

The A_2AR receptor, as we anticipated, interacts with CD73 and ADA enzymes, both of which also interact with IL-6. This cytokine, in turn, interconnects with several targets of lipid metabolism, illustrating the intricate web of molecular interactions. We also highlight the direct connection between A_2AR and CREB1, which interacts with transcription factors that modulate lipid metabolism, PPARγ, SREBF1, and SREBF2. Furthermore, CREB1 interacts with the cytokines IL-6 and IL-8, which interact with IL-6, PPARγ, and CD36, a gene regulated by PPARγ (Supplementary Figure S4A). Our network’s green and red nodes represent clustered proteins, each with their cellular processes, as shown in color and indicated in Supplementary Figures S4B, C. The green nodes highlight the processes involved in the negative regulation of the inflammatory response and the metabolic process of adenosine, further emphasizing the complexity of the molecular mechanisms at play. The cellular processes selected to red nodes accentuate the regulation of lipid localization and cholesterol storage, lipid metabolic processes, and the cell-surface receptor signaling pathway. Additional information is compiled in Supplementary Table S7.

This interaction network analysis has shed light on potential molecular mechanisms up- or downregulated by A_2AR that could contribute to decreasing LD accumulation. Importantly, our hypothesis was not just a theoretical construct but was confirmed by rigorous RT-qPCR analysis, adding a layer of confidence to our findings.

Our data showed that M. leprae infection upregulated the expression of pro-lipogenic genes, confirming previous results (Mattos et al., 2011a; Mattos et al., 2011b; Mattos et al., 2012; Mattos et al., 2014), while the activation of A_2AR with CGS21680 reversed this effect (Figure 7). The genes selected for analysis were PLIN3, an LD marker; PPARγ (peroxisome proliferator-activated receptor gamma), a transcription factor that regulates both adipogenic and lipogenic genes; CD36, a cholesterol scavenger receptor, also upregulated by M. leprae infection (Figures 7A–C). Moreover, SREBF1, which encodes the sterol regulatory-element binding proteins 1, a transcription factor that regulates genes of de novo fatty acid synthesis, was also downregulated by CGS21680 (Figure 7D). We did not observe any modulation on the SREBF2 (Figure 7E); however, HMGCR, which encodes 3-hydroxy-3-methylglutaryl CoA reductase, a rate-limiting step in the cholesterol biosynthesis pathway, was also downregulated by CGS21680 (Figure 7F). As well as pro-lipogenic genes, we analyzed the ABCA1 gene, which is regulated by LXR (liver X receptor) and encodes the ATP-binding cassette protein involved in the elimination of excess cholesterol from the cells. The CYP27A1 gene, which encodes a sterol 27-hydroxylase that catalyzes the formation of 27-hydroxycholesterol—an oxysterol activator of LXR—was also investigated. The results presented in Figures 7G and H show that M. leprae increased the expression of ABCA1 and CYP27A1 in infected SCs and that CGS21680 induced a significant increase in the expression of the ABCA1 gene. Regarding CYP27A1, we did not observe any modification when comparing M. leprae to M. leprae + CGS21680. Supplementary Table S9 summarizes the statistical data and the fold change effects of CGS21680. These findings propose a potential molecular mechanism by which the activation of A_2AR can reverse the LD accumulation induced by M. leprae infection.

P1 receptors have been associated both with a pro-inflammatory response (Csóka et al., 2007) and pro-resolving and anti-inflammatory effects (Alam et al., 2009). IL-8 is a cytokine that is positively modulated by M. leprae infection, both in macrophages (Mattos et al., 2009) and Schwann cells (Souza et al., 2022), while IL-6 is downregulated in M. leprae-infected macrophages (Polycarpou et al., 2016). The literature has shown that A_2AR regulates the production of both IL-6 and IL-8 in different study models (Köröskényi et al., 2016; Colombini et al., 2020). We wondered whether A_2AR could affect the production of these cytokines modulated by M. leprae infection. Thus, SCs were infected with M. leprae and treated with CGS21680. The results revealed that M. leprae slightly decreased, with statistically significance, the extracellular levels of IL-6 and increased the production of IL-8—the latter already described in the literature (Díaz Acosta et al., 2018)—compared with the NI SCs (Figures 8A, B). Interestingly, the addition of CGS21680 increased the extracellular levels of IL-6 but inhibited the levels of IL-8 both in NI and M. leprae-infected SCs (Figures 8A, B) suggesting that the downregulation of A_2AR in infected SCs contributes to the maintenance of the extracellular levels of both IL-6 and IL-8 produced by M. leprae-infected SCs. These statistical data are summarized in Supplementary Table S7. Finally, we assessed through RT-qPCR whether the activation of A_2AR could impact M. leprae intracellular viability. We observed that the addition of CGS21680 significantly reduced M. leprae viability in comparison with untreated SCs (Figure 8C). Overall, the observed data suggests that M. leprae, by downmodulating the expression of A_2AR, leads to decreased downstream signaling from this receptor, which contributes to the maintenance of the LD accumulation and intracellular viability of the bacilli. These results are schematically summarized in Figure 9.