Following infection or exposure to microbial products, innate immune cells may enter a state of tolerance several hours later, as depicted by their diminished responsiveness or unresponsiveness to microbial re-exposure (Foster et al., 2007; Bagchi et al., 2007; Masyutina et al., 2023; Divangahi et al., 2021). Given that 2-AA-tolerized cells are non-responsive to 2-AA re-exposure, we investigated the impact of 2-AA on energy homeostasis and metabolism in non-tolerized and tolerized BMDM cells (Figure 1). We quantified the levels of ATP and acetyl-CoA, which is a fuel for ATP production, in a series of in vitro experiments in which murine BMDM cells were exposed to 2-AA (first 2-AA exposure) or re-exposed (second 2-AA exposure) to respectively determine their activation, and tolerance as well as memory to this QS molecule (Figure 1A). Initially, following the first 2-AA exposure (black bars), a significant increase in the concentrations of intracellular ATP (Figure 1B) and acetyl-CoA (Figure 1C) was observed in BMDM cells exposed to 2-AA for 1 and 6 hr compared to their corresponding naïve control cells (gray bars). However, cells exposed to 2-AA for 48 hr (black bar) appeared to have entered a state of tolerance as ATP and acetyl-CoA levels were decreased compared to 2-AA exposures for a shorter time (Figure 1B and C). The tolerance of these cells as well as their memory to 2-AA was determined by their responsiveness to a second 2-AA exposure (Figure 1A). To achieve this, these cells were washed to remove 2-AA and allowed to rest for 24 or 106 hr in the absence of 2-AA before receiving the second 2-AA exposure for 1 or 6 hr (Figure 1A). Figure 1B and C (red bars) shows the unresponsiveness of BMDM cells to 2-AA re-exposure as evidenced by the similar levels of ATP and acetyl-CoA compared to cells exposed to 2-AA for 48 hr (first exposure black bars). The sustained unresponsiveness of the cells to re-exposure indicates that the cells exhibited innate immunological memory, which could not be reverted by 2-AA re-exposure. Similar responsiveness patterns of ATP and acetyl-CoA as with BMDM cells were observed in murine macrophage RAW 264.7 cells (Figure 1—figure supplement 1A and Figure 1—figure supplement 1B) or human monocyte THP-1 cells (Figure 1—figure supplement 1C and Figure 1—figure supplement 1D) following exposure or re-exposure to 2-AA. These findings further support our observations with BMDM cells and confirm that either cell type can be used to study responses to 2-AA. These results reinforce the notion that 2-AA impacts crucial energy metabolites in macrophages.

Moreover, we sought to determine whether 2-AA-tolerized BMDM cells would respond to heterologous stimulus. To test this, we re-exposed the cells to LPS, an outer membrane bacterial component that strongly induces macrophages. LPS re-exposure of the 2-AA-tolerized BMDM cells did not significantly (p=0.9) increase ATP levels (Figure 1—figure supplement 2), indicating that 2-AA cross-tolerized the cells to this heterologous bacterial immunostimulant. On the other hand, as expected, we observed increased ATP levels following LPS stimulation of the non-tolerized BMDM cells (Figure 1—figure supplement 2). The broad spectrum of 2-AA cross-tolerization strongly suggests an interplay of epigenetic mechanism rather than ligand-receptor-mediated signaling-based mechanism in bringing out the observed 2-AA-mediated tolerance.

To further investigate the 2-AA’s impact on the bioenergetics of macrophages, we assessed by Seahorse assay the oxygen consumption rate (OCR) as an index of OXPHOS in 2-AA exposed (for 1 and 24 hr) and re-exposed (second exposure for 1 hr) BMDM cells (Figure 1D). Values of basal (Figure 1E) and maximal (Figure 1F) mitochondrial respiration and spare respiratory capacity (Figure 1—figure supplement 3) interpreted from these measurements revealed that 1 hr 2-AA exposure of naïve cells significantly increased OCR compared to naïve control cells. At 24 hr post-2-AA exposure, cells had a significantly reduced basal OCR level than the control BMDM cells, indicating decreased OXPHOS and an overall quiescent phenotype of tolerized macrophages. Re-exposure of tolerized macrophages with 2-AA also did not augment maximal and basal OCR (Figure 1D–F), indicating the unresponsiveness of tolerized BMDM cells. Ultramicroscopic examination of RAW 264.7 cells exposed to 2-AA for 48 hr although shows the same number of mitochondria as in control cells, their mitochondrial morphology is altered, appearing smaller and round and having reduced cristae indicating dysfunctional mitochondria (Figure 1—figure supplement 4).

These OCR data are consistent with the previous direct measurements of ATP (Figure 1B) and acetyl-CoA (Figure 1C) and support the notion that 2-AA tolerization induces a quiescent phenotype in these cells characterized by defective OXPHOS. These findings confirm that 2-AA mediates energy homeostatic and metabolic alterations in the immune cells and that 2-AA tolerization arrests these cells in a sustained 2-AA-unresponsive state.

Since acetyl-CoA and ATP levels were reduced due to 2-AA tolerization, we investigated whether the mitochondrial dysfunction observed is related to pyruvate metabolism. Pyruvate links glycolysis with mitochondrial production of acetyl-CoA and ATP via the tricarboxylic acid (TCA) cycle and OXPHOS (Galli and Saleh, 2020). RAW 264.7 cells initially exposed to 2-AA for 1 or 3 hr (black bars) exhibited significantly higher concentrations of cytosolic and mitochondrial pyruvate compared to the corresponding naïve control cells (Figure 2A). Pyruvate significantly decreased over time in cytosolic and mitochondrial fractions as cells were entering into the tolerized state (Figure 2A). Tolerized cells remain unresponsive to 2-AA re-exposure (red bars), exhibiting no significant change in the concentration of pyruvate in the cytosolic or mitochondrial fraction (Figure 2A). A similar result was observed in THP-1 cells (Figure 2—figure supplement 1A and Figure 2—figure supplement 1B).

To gain a better understanding of the 2-AA-mediated reductions in mitochondrial pyruvate, ATP production, and acetyl-CoA levels, we sought to determine the involvement of the Mpc1 (Koh et al., 2018) that is crucial in transporting pyruvate from the cytosol into the mitochondria (Xue et al., 2021). Western blot studies of whole-cell lysates of RAW 246.7 cells exposed to 2-AA showed a significant decrease over time in the abundance of Mpc1 protein, and tolerized cells remain unresponsive to 2-AA re-exposure (Figure 2B). Together, these findings suggest that 2-AA tolerization dysregulates Mpc1-mediated transport of the pyruvate into mitochondria.

Given that the Ppargc1a/Esrra axis is the transcriptional regulatory node that activates the expression of Mpc1 (Schreiber et al., 2003), we determined the protein levels of Esrra. Western blot studies of whole-cell lysates showed a significant decrease over time in the abundance of Esrra protein following first exposure of RAW 246.7 cells to 2-AA, while tolerized cells remain unresponsive to 2-AA re-exposure (Figure 3A). By assessing the cytosolic and nuclear abundance of Esrra in tolerized cells, we found that both fractions had lower levels of Esrra protein than their corresponding control cells (Figure 3B).

Furthermore, because an Esrra-binding putative DNA motif (TNAAGGTCA) has been identified upstream of the Mpc1 promoter in humans (Koh et al., 2018), and we found the same motif to be present 1.5 kB upstream of the transcription start site of mouse Mpc1 promoter, we analyzed if tolerization led to reduced Esrra binding to the Mpc1 promoter motif TNAAGGTCA. Chromatin immunoprecipitation (ChIP) assay followed by qPCR analysis revealed that in tolerized cells, Esrra binds approximately fourfold less efficiently to the putative binding site on the Mpc1 than in corresponding control cells (Figure 3C).

Since the transcriptional activity of Esrra is regulated through a protein-protein interaction with the transcriptional coactivator Ppargc1a (Schreiber et al., 2003), we assessed whether the formation of the Ppargc1a/Esrra complex is affected in 2-AA-tolerized cells. Using paraformaldehyde fixed nuclear lysates of RAW 264.7 tolerized and naïve cells, we performed a co-immunoprecipitation (co-IP) of Ppargc1a and reversed co-IP with Esrra followed by immunoblotting (Figure 3D). As shown in Figure 3D less Esrra was detected when Ppargc1a was pulled down in the presence of 2-AA than in naïve control cells. Moreover, reverse co-IP with Esrra protein showed lower levels of Ppargc1a in the presence of 2-AA than in naïve control cells (Figure 3D). Control IgG showed no detectable levels of Esrra following nuclear fraction pull-down. These results indicate that in the tolerized macrophages, Ppargc1a/Esrra interaction is impaired (Figure 3D).

Moreover, since Ppargc1a enhances the transcriptional activity of Esrra through their interaction, and Esrra transcription is regulated via an autoregulatory loop (Laganière et al., 2004), we examined the effect of tolerization on the transcription of Ppargc1a, Esrra, and its target gene Mpc1. Using RAW 264.7 cells infected with the wild-type PA strain PA14 or isogenic ΔmvfR mutant, which does not produce 2-AA (Kesarwani et al., 2011) we observed lower RNA transcript levels of Mpc1 (Figure 4A), Esrra (Figure 4B), and Ppargc1a (Figure 4C) in PA14 compared to ΔmvfR. The addition of 2-AA to ΔmvfR decreased the levels of these gene transcripts to levels similar to the PA14 infection condition, supporting the role of 2-AA in effect. To test if the observed effects are due to Mpc1-dependent reduction of mitochondrial ATP generation, we supplemented the macrophages with ATP. The findings indicate that the addition of ATP in PA14-infected cells elevated the transcript levels of Esrra, Mpc1, and Ppargc1a, reaching levels similar to those observed in ΔmvfR infected macrophages (Figure 4A, B, and C). These results indicate that reduced Mpc1 function is due to 2-AA tolerization on the transcriptional activation of Mpc1 through the Ppargc1a/Esrra axis.

2-AA mediates persistence of PA in vivo, dampens pro-inflammatory responses, and increases the intracellular burden of this pathogen in macrophages via epigenetic modifications (Bandyopadhaya et al., 2012; Bandyopadhaya et al., 2016b; Chakraborty et al., 2023). Here, we used RAW 264.7 cells infected with the PA14 or the 2-AA-deficient ΔmvfR mutant to assess the clearance of PA by macrophages. Macrophages infected with PA14 showed increased bacterial burden than the cells infected with ΔmvfR, and 2-AA addition to ΔmvfR led to increased bacterial burden (Figure 4D). Furthermore, we used UK5099 and ATP to interrogate whether Mpc1-mediated mitochondrial pyruvate import and bioenergetics are linked to the clearance of PA intracellular burden. The addition of the UK5099 inhibitor strongly enhanced the bacterial intracellular burden in ΔmvfR infected macrophages compared to the non-inhibited ΔmvfR infected cells, reaching a similar burden to those infected with PA14 (Figure 4D). Conversely, exogenously added ATP to macrophages infected with PA14 strongly reduced the PA intracellular burden (Figure 4D).

Given that 2-AA tolerization decreases the expression of pro-inflammatory cytokine TNF-α by hypo-acetylating the core histone 3 lysine 18 acetylation (H3K18ac) mark at TNF-α promoter (Bandyopadhaya et al., 2016b), we investigated the link between bioenergetics and TNF-α expression in infected macrophages. UK5099 or ATP was added exogenously to infected macrophages with PA14 or ΔmvfR (Figure 4E). As shown in Figure 4E, PA14-infected cells showed lower TNF-α transcript levels compared to the ΔmvfR infected cells. Supplementation of 2-AA to ΔmvfR infected cells led to a decrease in the TNF-α transcript levels (Figure 4E). The addition of the Mpc1 inhibitor, UK5099, in ΔmvfR infected cells counteracted the increase in TNF-α transcript levels observed in ΔmvfR infected cells in the absence of UK5099 (Figure 4E). Conversely, exogeneous addition of ATP to PA14-infected cells enhanced the transcription of TNF-α compared to the untreated PA14-infected cells, while no difference in TNF-α expression levels were observed in ΔmvfR infected macrophages in presence or absence of ATP (Figure 4E). These findings strongly suggest that 2-AA tolerization severely alters macrophages’ ability to facilitate the clearance of PA intracellular burden, via the reduction in Mpc1-mediated pyruvate import into mitochondria, ATP levels, and TNF-α transcription.

To determine whether the decrease in the key metabolites mediated by 2-AA is also observed in vivo during infection, we quantified the levels of ATP and acetyl-CoA in murine spleen tissues at 1, 5, and 10 days (Figure 5). This organ was selected for our immunometabolic studies due to its role in regulating not only local but also systemic (whole-body) immune responses, facilitated by various immune cells including macrophages (Bronte and Pittet, 2013). Mice were infected with PA strain PA14, or isogenic mutant ΔmvfR, which does not produce 2-AA (Kesarwani et al., 2011). It is important to note that it is not possible to generate or utilize a bacterial mutant that is defective in 2-AA only because 2-AA is formed by spontaneous decarboxylation rather than by an enzyme-catalyzed reaction (Starkey et al., 2014; Dulcey et al., 2013; Fetzner and Drees, 2013). Therefore, animals that were either infected with ΔmvfR and received 2-AA (ΔmvfR + 2-AA) at the time of infection or uninfected mice injected with a single dose of 2-AA served as direct controls (Figure 5). Naïve and sham mice groups served as additional basal controls (Figure 5).

Infection with PA14 or uninfected mice injected with 2-AA led to a significant decrease in both ATP (Figure 5A) and acetyl-CoA (Figure 5B) concentrations in spleen tissues compared to ΔmvfR infected mice that sustained higher ATP and acetyl-CoA levels similar to naïve and sham control groups across the time points tested. However, in mice infected with ΔmvfR and 2-AA injected (ΔmvfR + 2-AA) at the time of infection, ATP and acetyl-CoA concentrations in the spleen decreased to levels comparable to those of PA14-infected animals, strongly indicating the biological function of 2-AA in decreasing these key metabolites. These in vivo findings further support the adverse action of 2-AA on host energy homeostasis and metabolism observed in in vitro studies.

The metabolic alterations observed are associated with the host tolerance to PA persistence (Figure 5C). Using Pseudomonas isolation agar plates, we evaluated the bacterial load at the infection site over the course of 10 days, obtaining samples at 1, 5, and 10 days. At 1 day post-infection, mice infected with PA14, ΔmvfR, or ΔmvfR + 2-AA exhibited no difference in bacterial burden (Figure 5C), verifying the ability of all strains to establish infection. At 5 and 10 days post-infection, mice infected with PA14 or ΔmvfR + 2-AA sustained the bacterial burden at a significantly higher bacterial burden over time compared to those infected with ΔmvfR (Figure 5C).

To strengthen the relevance of our in vivo data, we performed additional in vivo experiments. In this set of in vivo studies, mice received the first exposure to 2-AA by injecting 2-AA only and the second exposure through infection with PA14 or ΔmvfR 4 days post-2-AA injection. As shown in Figure 5—figure supplement 1, the levels of ATP and acetyl-CoA in the spleen of infected animals and the enumeration of the bacterial counts were similar between PA14 and ΔmvfR receiving the first 2-AA exposure and agree with the ‘one-shot infection’ findings presented in Figure 5 with the PA14 or ΔmvfR + 2-AA infected mice or those receiving 2-AA only. These results are consistent with our previous findings, showing that 2-AA impedes the clearance of PA14 (Bandyopadhaya et al., 2012; Bandyopadhaya et al., 2016b) and provide compelling evidence that the metabolic alterations identified may favor PA persistence in infected tissues.

RAW 264.7 and THP1 cells were obtained from ATCC. Short tandem repeat analysis (DNA fingerprinting) was performed to determine the identity and uniqueness of a human line through Harvard Catalyst core facility. Periodic assays were performed to detect mycoplasma using PlasmoTest kit (InvivoGen). In addition, we used fluorescent Hoechst staining, Hoechst 33258 fluorescent dye, that binds specifically to DNA to reveal possible mycoplasma infections through their characteristic patterns of extracellular particulate or filamentous at ×500 magnification. BMDM cells were grown in Roswell Park Memorial Institute (RPMI) 1640 medium, while RAW 264.7 and THP1 cells were grown in Dulbecco’s Modified Eagle Medium (DMEM).

Mouse BMDM cells were isolated from the femur of CD1 6-week-old male and female mice and used to estimate the levels of various metabolites (Figure 2A) following exposure and re-exposure to 2-AA (Figure 1A). The first exposure of naïve BMDMs (10⁶/mL in six-well plates) was achieved with 400 µM 2-AA for 1, 6, or 48 hr. Cells exposed for 48 hr were used for the second 2-AA exposure. These cells were washed to remove residual 2-AA, left in the RPMI 1640 medium in the absence of 2-AA for either 24 or 106 hr, and re-exposed to 2-AA (200 µM) for 1 or 6 hr, respectively. For RAW 264.7 (10⁶/mL in six-well plates) and THP-1 cells (10⁶/mL in six-well plates) DMEM was used, the 2-AA concentration and conditions used were the same as those used for BMDM cells. The hours of the first round and second round of exposure to 2-AA are indicated in Figure 1.

For the LPS stimulation studies, cells were stimulated with 400 µM 2-AA or 100 ng/mL LPS for 6 hr and 800 µM of 2-AA for 48 hr. Cells receiving first exposure for 48 hr were washed and re-exposed to 400 µM of 2-AA or 100 ng/mL LPS for 6 hr.

The levels of ATP acetyl-CoA were assessed in BMDM and RAW 264.7 macrophage cells following exposure to 2-AA at the times indicated and by utilizing the ATP Assay kit (cholorimetric/fluorometric) (#ab83355, Abcam) and the PicoProbe AcCoA assay kit (ab87546, Abcam), respectively, according to the manufacturer’s instruction. Quantifications of ATP and acetyl-CoA were performed in triplicates.

Seahorse analysis was performed according to the previously published protocols (Van den Bossche et al., 2015). Briefly, freshly prepared BMDM cells were reseeded in complete RPMI-1640 cells using Seahorse plates (Agilent cat. no. 103729100) at a density of 5×10⁴ cells per well. Cells were exposed to 400 µM 2-AA for 1 or 24 hr (first exposure). Cells receiving first exposure for 24 hr were washed and re-exposed to 200 µM of 2-AA for 1 hr. Naïve cells were used as control. Prior to initiating Seahorse measurements, cells were washed, and Seahorse XF DMEM supplemented with 2 mM glutamine (Gibco cat. no. 25030-081) was added to each well. Cells were allowed to stabilize in a 37°C incubator without CO₂ for 1 hr. The Seahorse cartridge was hydrated and calibrated as per the manufacturer’s instructions. The Mitochondrial Stress Test Kit (Agilent cat. no. 10395-100) was used (oligomycin at 1 μM, FCCP [1.5 µM], rotenone [0.5 μM], and antimycin A [0.5 μM]) with slight modifications according to the published protocol (Van den Bossche et al., 2015) that included injection of 25 mM glucose and sodium pyruvate (1 μM). All samples N=4 were run in a Seahorse XFe96 Analyzer, and data were analyzed using Wave and plotted using GraphPad Prism software.

Pyruvate is produced in the cytosol and is transported into the mitochondria. RAW 264.7 cells were plated at 1×10⁶ to incubate overnight at 37°C in a CO₂ incubator. Cells were exposed to 2-AA for 1, 3, 24, or 48 hr or re-exposed for 1 hr as described above. Mitochondria and cytosolic fractions of each group were isolated utilizing the Mitochondria Isolation Kit for cultured cells (#ab110170, Abcam) following the manufacturer’s protocol. Briefly, cells were collected with a cell lifter and pelleted by centrifugation at 1000×g, frozen, and then thawed to weaken the cell membranes. The cells were resuspended in Reagent A and transferred into a pre-cooled Dounce Homogenizer. The homogenates were centrifuged at 1000×g for 10 min at 4°C and saved as supernatant #1 for the cytosolic fractions. The pellet was resuspended in Reagent B, followed by repeat rupturing and centrifugation. The pellet was collected and resuspended in 500 μL of Reagent C supplemented with Protease Inhibitor cocktails (P8340, Sigma-Aldrich). Following separation, a Bradford assay was conducted to determine the protein concentration in each fraction. Pyruvate levels were determined in cellular fractions and mitochondrial fractions by using the Pyruvate Assay Kit (#ab65342, Abcam) according to the manufacturer’s instructions. Briefly, after deproteinization using perchloric acid, the samples were neutralized in ice-cold 2 M KOH. 10 µL samples were incubated with reaction mix and kept on the plate at room temperature for 30 min in the dark. The absorbance was measured in a microplate reader (Tecan Group Ltd, Männedorf, Switzerland) at 570 nm, and the results were shown in three independent experiments.

The PA strain known as Rif^R human clinical isolate UCBPP-PA14 (also known as PA14) was used (Rahme et al., 1995). The bacteria were grown at 37°C in lysogeny broth (LB) under shaking and aeration or on LB agar plates containing appropriate antibiotics. PA14 and isogenic mutant ΔmvfR (Cao et al., 2001) cultures were grown in LB from a single colony to an optical density of 600 nm (OD₆₀₀) of 1.5, diluted 1:50,000,000 in fresh LB media, and grown overnight to an OD₆₀₀ of 3.0.

RAW 264.7 macrophage cells were plated on six-well cell culture-treated plates overnight in pyruvate-free DMEM. After 3 hr of incubation with 10 µM UK5099, 20 µM ATP or 400 µM 2-AA, cells were infected with PA14 and isogenic mutant ΔmvfR at 5 MOI for 30 min at 37°C in 5% CO₂. Unbound bacteria were removed by washing once with cold DMEM. Afterward, the cells were incubated with 100 µg/mL of gentamicin for 30 min to eliminate residual extracellular bacteria. The cells were washed with DMEM, transferred to free medium without gentamicin and kept for 3 hr at 37°C in 5% CO₂. The infected cells were scraped after 3 hr, centrifuged at 500×g, and lysed in distilled water. The lysed cells were immediately diluted in PBS and plated on LB agar plates to assess bacterial presence. Bacterial colony-forming units (CFUs) were counted after incubating the plates overnight at 37°C. Untreated cells infected with PA14 were set as 100% and the reduction of the bacterial load was expressed as %CFU.

For the OXPHOS inhibition assay, RAW 264.7 macrophage cells were treated with UK5099 (10 µM, Sigma-Aldrich, dissolved in DMSO) 3 hr prior to first or second rounds of 2-AA exposure. RAW 264.7 macrophages supplemented with ATP (20 µM, Sigma-Aldrich, dissolved in PBS) received ATP at the time of first and second 2-AA exposure.

For all immunoprecipitation assays, protein A/G agarose Magnetic beads (Pierce) were used after washing in 1× IP buffer. Nuclear lysates from RAW 264.7 tolerized cells exposed to 2-AA for 24 hr or non-exposed cells were diluted in 1× IP buffer, and 500 µg of protein was taken for each experiment. The lysates were precleared by using unbound 50 µL protein A/G magnetic beads for 2 hr at room temperature on a rotator. Precleared lysates were either incubated with Esrra, Ppargc1a, or rabbit IgG antibody overnight at 4°C. 100 µL protein A/G magnetic beads were used to pull down the antibody-protein complex and washed twice with IP buffer to remove unbound proteins. Magnetic beads were then eluted in 2× Laemmli SDS-PAGE loading buffer. Immunoprecipitated Esrra and Ppargc1a was detected by western blot analyses, using conformational-specific Anti Rabbit antibody (TruBlot).

Cells were plated at 6×10⁵ cells per well in six-well plates. Cells were washed with PBS and subsequently lysed using RIPA lysis buffer containing 1 mM phenylmethylsulfonyl fluoride. 20 μg of proteins were separated by electrophoresis on any KD (Kilo Dalton) (Bio-Rad, cat no. 4569033) SDS-polyacrylamide gel. Proteins were transferred to a 0.2 μm polyvinylidene fluoride membrane (Millipore, Billerica, MA, USA) using a Bio-Rad semi-dry instrument. After blocking with 5% BSA in TBS containing 0.1% Tween-20 for 1 hr at room temperature, the membranes were incubated with a primary antibody Esrra (Abcam, #ab76228), Mpc1 (D2L91, Cell Signaling), and anti-β-actin (cat no. sc-47778) (Santa Cruz Biotechnology) overnight at 4°C. Following washing, the membranes were incubated with an anti-rabbit secondary antibody, and the bands were detected by SuperSignal West Pico Chemiluminescent Substrate (Thermo Scientific) reaction, according to the manufacturer’s instructions. The blots were visualized in the ChemiDOC Imaging system (Bio-Rad Laboratories, Inc, Hercules, CA, USA). The bands were analyzed densitometrically using QuantityOne software (Bio-Rad).

For ChIP studies, RAW 264.7 macrophage cells exposed to 2-AA for 24 hr were cross-linked in 1% (vol/vol) methanol-free formaldehyde for 10 min and then placed in 0.125 M glycine for 5 min at room temperature. Using the truChIP High Cell Chromatin Shearing kit (Covaris, USA), cells were prepared for sonication according to the Covaris protocol. Approximately 1 × 10⁷ cells were plated in a 12 mm × 12 mm tube and subjected to shearing with the Covaris S220 sonicator for 8 min (140 peak power, 5 duty factor, 200 cycles/burst). The Magna ChIP A/G kit (Millipore, USA) was used for the subsequent immunoprecipitations according to the manufacturer’s protocol. Briefly, chromatin from approximately 10⁶ cells was incubated overnight at 4°C with 2 μg of anti-Esrra or anti-Ppargc1a (Abcam, USA) ChIP-grade antibody and 20 μL of A/G magnetic beads. The beads were washed serially (5 min each) with low-salt wash buffer, high-salt wash buffer, LiCl wash buffer, and TE buffer from the kit at 4°C. Chromatin was eluted with elution buffer containing Proteinase K at 62°C for 4 hr, then incubated at 95°C for 10 min. DNA was isolated by column purification (QIAquick PCR purification kit).

Real-time ChIP-qPCR was performed with the Brilliant II SYBR green super mix (Agilent, USA). Forward (AGTGGTGACCTTGAACTTCCC) and reverse (CTGAAGACGACCTTCCCCTT) primers were chosen to amplify a genomic locus of Mpc1 promoter, which had a putative ERR-binding site (TNAAGGTCA) at 1582 bp upstream of the start site. Normalized values were calculated using the percent-input method relative to the IgG. The assay was performed three times.

Total RNA from all the groups mentioned above was isolated from approximately 2 × 10⁶ cells with the RNeasy minikit (QIAGEN, USA), and cDNA was prepared with the iScript Reverse transcription kit (Bio-Rad, USA), as per the manufacturer’s instruction. Real-time PCR was conducted using the PowerUP SYBR Green Master mix (Applied Biosystems, USA) and primer sets for mouse Esrra (forward: ACTACGGTGTGGCATCCTGTGA; reverse: GGTGATCTCACACTCATTGGAGG), Ppargc1a (forward: GAATCAAGCCACTACAGACACCG; reverse: CATCCCTCTTGAGCCTTTCGTG), MPC-1 (forward: CTCCAGAGATTATCAGTGGGCG; reverse: GAGCTACTTCGTTTGTTACATGGC), TNF-α (forward: GGTGCCTATGTCTCAGCCTCTT; reverse: GCCATAGAACTGATGAGAGGGAG) and mouse 18S rRNA (forward: GTTCCGACCATAAACGATGCC; reverse: TGGTGGTGCCCTTCCGTCAAT). The transcript levels of all the genes were normalized to 18S rRNA with the ΔΔCT method. The relative expression was calculated by normalizing transcript levels to those of PA14-infected cells. The assay was conducted in triplicate; means and standard deviations were calculated for each group.

The full-thickness thermal burn injury and infection (BI) model (Bandyopadhaya et al., 2012; Bandyopadhaya et al., 2016b; Maura et al., 2018) was used to assess the effect of 2-AA on metabolic alterations in the spleens of 6-week-old CD1 male mice and bacterial burden (Charles River Labs, USA). A full-thickness thermal burn injury involving 5–8% of the total body surface area was produced on the shaved mouse abdomen dermis, and an inoculum of ~1 × 10⁴ PA14 and ΔmvfR (Cao et al., 2001) cells in 100 μL of MgSO₄ (10 mM) was injected intradermally into the burn eschar. For the groups that received 2-AA, mice were injected intraperitoneally with 100 μL of 2-AA (6.75 mg/kg). The entire procedure was done under the influence of anesthesia. One of the groups of mice infected with the PA14 isogenic mutant ΔmvfR (Cao et al., 2001) also received 100 μL of 2-AA in PBS (6.75 mg/kg) at the time of infection (ΔmvfR + 2-AA) and served as an additional control. CFU counts from rectus abdominus muscle (underlying the burn-infected tissue) were assessed in groups of four mice each at 1, 5, and 10 days post-BI by plating diluted muscle homogenate on Pseudomonas Isolation Agar (Sigma-Aldrich) plates containing rifampicin (50 mg/L). Spleen samples from mice were collected for metabolite analyses from all mice groups at 1, 5, and 10 days post-BI to assess ATP (#ab83355 Abcam) and acetyl-CoA (#ab87546, Abcam) as described above. Tissues were homogenized, 100 mg of homogenate was centrifuged at 4°C for 15 min at 10,000×g, and supernatants were collected. The supernatant was mixed with 400 µL of 1 M perchloric acid. The deproteinized supernatant was neutralized by 3 M KHCO₃. The ATP and acetyl-CoA assays were performed as described in the in vitro section.

RAW 264.7 macrophages exposed to 2-AA and corresponding controls were fixed with 2% glutaraldehyde in 0.1 M cacodylate buffer and post-fixed in 1% OsO₄ in 0.1 M cacodylate buffer for 1 hr on ice. The cells were stained all at once with 2% uranyl acetate for 1 hr on ice, after which they were dehydrated in a graded series of ethanol (50–100%) while remaining on ice. Ultrathin (70 nm) sections were cut using a Leica EMUC7 ultramicrotome and collected onto formvar-coated grids (EMS, Hatfield, PA, USA). Sections were contrast-stained using 2.0% aqueous uranyl acetate. Grids were examined at 80 kV in a JEOL 1011 transmission electron microscope (Peabody, MA, USA) equipped with an AMT digital camera and proprietary image capture software (Advanced Microscopy Techniques, Danvers, MA, USA).