Quorum-Sensing Signaling Molecule 2-Aminoacetophenone Mediates the Persistence of Pseudomonas aeruginosa in Macrophages by Interference with Autophagy through Epigenetic Regulation of Lipid Biosynthesis

Abstract

Macrophages are crucial components of the host's defense against pathogens. Recent studies indicate that macrophage functions are influenced by lipid metabolism. However, knowledge of how bacterial pathogens exploit macrophage lipid metabolism for their benefit remains rudimentary. We have shown that the Pseudomonas aeruginosa MvfR-regulated quorum-sensing (QS) signaling molecule 2-aminoacetophenone (2-AA) mediates epigenetic and metabolic changes associated with this pathogen's persistence in vivo. We provide evidence that 2-AA counteracts the ability of macrophages to clear the intracellular P. aeruginosa, leading to persistence. The intracellular action of 2-AA in macrophages is linked to reduced autophagic functions and the impaired expression of a central lipogenic gene, stearoyl-CoA desaturase 1 (Scd1), which catalyzes the biosynthesis of monounsaturated fatty acids. 2-AA also reduces the expression of the autophagic genes Unc-51-like autophagy activating kinase 1 (ULK1) and Beclin1 and the levels of the autophagosomal membrane protein microtubule-associated protein 1, light chain 3 isoform B (LC3B) and p62. Reduced autophagy is accompanied by the reduced expression of the lipogenic gene Scd1, preventing bacterial clearance. Adding the SCD1 substrates palmitoyl-CoA and stearoyl-CoA increases P. aeruginosa clearance by macrophages. The impact of 2-AA on lipogenic gene expression and autophagic machinery is histone deacetylase 1 (HDAC1) mediated, implicating the HDAC1 epigenetic marks at the promoter sites of Scd1 and Beclin1 genes. This work provides novel insights into the complex metabolic alterations and epigenetic regulation promoted by QS and uncovers additional 2-AA actions supporting P. aeruginosa sustainment in macrophages. These findings may aid in designing host-directed therapeutics and protective interventions against P. aeruginosa persistence. IMPORTANCE This work sheds new light on how P. aeruginosa limits bacterial clearance in macrophages through 2-aminoacetophenone (2-AA), a secreted signaling molecule by this pathogen that is regulated by the quorum-sensing transcription factor MvfR. The action of 2-AA on the lipid biosynthesis gene Scd1 and the autophagic genes ULK1 and Beclin1 appears to secure the reduced intracellular clearance of P. aeruginosa by macrophages. In support of the 2-AA effect on lipid biosynthesis, the ability of macrophages to reduce the intracellular P. aeruginosa burden is reinstated following the supplementation of palmitoyl-CoA and stearoyl-CoA. The 2-AA-mediated reduction of Scd1 and Beclin1 expression is linked to chromatin modifications, implicating the enzyme histone deacetylase 1 (HDAC1), thus opening new avenues for future strategies against this pathogen's persistence. Overall, the knowledge obtained from this work provides for developing new therapeutics against P. aeruginosa.

Keywords: 2-aminoacetophenone; MvfR; PqsR; Pseudomonas aeruginosa; Scd1; autophagy; epigenetic reprogramming; fatty acids; histone deacetylation; immunometabolism; macrophages; persistence; quorum sensing.

FIG 1

The MvfR-regulated quorum sensing signaling molecule 2-AA sustains P. aeruginosa burden in macrophages. (A) Enumeration of the intracellular P. aeruginosa, as CFU/10⁶ RAW 264.7 macrophages (mΦ) at 1, 2, and 3 h postinfection with WT PA14 (green) or the isogenic mutants nonproducing 2-AA mvfR (red) and pqsA (blue) and the 2-AA producing isogenic mutant pqsBC (gray). Bacterial loads were significantly reduced in mvfR and pqsA mutants relative to PA14 and pqsBC. (B) Enumeration of the P. aeruginosa intracellular CFU/10⁶ RAW 264.7 mΦ in the absence or presence of 200 μM 2-AA (dotted) at 1 h and 3 h postinfection with PA14 or mutants mvfR, pqsA, or pqsBC. The addition of 2-AA significantly increases the intracellular bacterial load in mvfR and pqsA-infected macrophages relative to those infected with these nonproducing 2-AA mutants at 3 h postinfection. The error bars denote ± SD. One-way ANOVA followed by Tukey posttest was applied. ***, P < 0.001; ns indicates no significant difference. Data represent n ≥ 3 independent replicates. Each circle represents data from one replicate.

FIG 2

Impairment of autophagy contributes to the P. aeruginosa intracellular bacterial load in infected macrophages. (A) Enumeration of the intracellular P. aeruginosa CFU/10⁶ RAW264.7 mΦ at 3 h postinfection with PA14 (green) strain or the isogenic mutants nonproducing 2-AA mvfR (red) and pqsA (blue) in the absence or in the presence of exogenously added 200 μM 2-AA (dotted), or 10 mM rapamycin (striped). The addition of the autophagic inducer rapamycin significantly decreases the intracellular CFU in PA14-infected macrophages (green striped) and isogenic mutants mvfR (red striped) compared to infected macrophages that did not receive rapamycin (solid color). This decrease is abolished in all macrophage-infected groups treated with 2-AA (dotted colors), despite the presence of rapamycin (striped colors). (B) Representative confocal images depict LC₃B puncta (red) in BMDM cells infected with PA14, mvfR, or pqsA (top). PBS was used as a control. PA14-infected macrophages show fewer LC₃B puncta than macrophages infected with mvfR and pqsA. However, supplementation with 2-AA (200 μM) in macrophages infected with mvfR and pqsA (bottom) also showed lower LC₃B punctation than their nonsupplemented counterparts. The experiment was repeated independently three times with similar results. (C and D) RT-qPCR analysis of PA14 (green)-infected RAW264.7 macrophages reveal significantly lower ULK1 and Beclin1 expression levels relative to mvfR (red)- or pqsA (blue)-infected macrophages, while 2-AA (200 μM) addition in both mvfR (red dotted)- and pqsA (blue dotted)-infected cells decreases the expression of these genes. The error bars denote ± SD. One-way ANOVA followed by Tukey posttest was applied. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns indicates no significant difference. Data represent n = 6 independent replicates. Each dot represents data from one replicate. (E) Representative immunoblot of p62, LC₃B protein levels in PA14-, mvfR-, or pqsA-infected macrophages ± 2-AA (200 μM). β-Actin was used as a control. The blots are representative of three independent experiments.

FIG 3

2-AA affects membrane lipids. Representative confocal images of BMDM cells depicting membrane ganglioside staining with fluorescent-labeled cholera toxin B (red) (A) untreated (top) or treated (bottom) with 2-AA (200 μM) for 1 h, 3 h, 6 h, 9 h, and 24 h and infected (B) with PA14, mvfR, pqsA, and pqsBC (top), or treated with 2-AA (bottom) for 3 h. (C and D) Corresponding graphical intensity plots of panels A and B, respectively. Arbitrary fluorescent units depict the intensity of membrane staining (red) versus the intensity of DAPI (blue). Significant reduction in membrane staining relative to the PBS-treated cells is observed within 3 h post-2-AA treatment. The error bars denote ± SD. One-way ANOVA followed by Tukey posttest was applied. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns indicates no significant difference. The images are representative of three independent experiments.

FIG 4

2-AA affects lipogenic gene expression in macrophages. (A) Representative confocal images of SCD1 immunocytochemical staining (red) in BMDM treated (bottom)/untreated (top) with 2-AA. The macrophages were counterstained with DAPI (Blue). (B) Representative confocal images showing SCD1 expression (red) in BMDM infected with PA14, mvfR, pqsA, and pqsBC (top). Supplementation of 2-AA (200 μM) in macrophages infected with mvfR or pqsA (bottom) showed lower SCD1 staining than their nonsupplemented counterparts. (C) RT-qPCR analysis of Scd1 gene expression in RAW264.7 mΦ infected with PA14 (green), mvfR (red), or pqsA (blue). Supplementation of 2-AA in infected cells with mvfR (red dotted), pqsA (blue dotted), or uninfected (white dotted) cells decreases Scd1 expression levels. (D) Bacterial load represented as CFU/10⁶ mΦ in the absence and presence of 10 μM SCD1 inhibitor (SCD1i) or 200 μM 2-AA. The bacterial load increases when infected RAW 264.7 cells are treated with either SCD1i or 2-AA. (E) Enumeration of PA14 and mvfR intracellular CFU/10⁶ RAW 264.7 mΦ cells in the absence and presence of palmitoyl-CoA (50 μM) and stearoyl-CoA (50 μM). The addition of these compounds decreases the intracellular bacterial load in the presence of 2-AA. Addition of ATP (10 μM) was used as a control in PA14 cells. Data represent n ≥ 5 independent replicates. Each point represents data from one replicate. The error bars denote ± SD. One-way ANOVA followed by Tukey’s honestly significant difference (HSD) post hoc test was applied; ****, P < 0.0001; ns indicates no significant difference.

FIG 5

Impairment of lipid homeostasis and autophagy by 2-AA is HDAC1-dependent. (A and B) Representative confocal images and graphical intensity plots (Ai and Bi) depicting enhanced membrane lipid staining with fluorescent-labeled cholera toxin B (red) (A) or LC₃B puncta (Red) (B) in RAW 264.7 and hdac1 KD isogenic cells uninfected and infected with PA14 (top) and/or treated with 2-AA (bottom). (C) Enumeration of PA14 intracellular CFU/10⁶ in RAW 264.7 or hdac1 KD mΦ cells. Loss of HDAC1 clears PA14 more efficiently than the wild-type macrophages. Data represent n = 6 independent replicates. Each dot represents data from one replicate. The error bars denote ± SD. One-way ANOVA followed by Tukey posttest was applied. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; and ns indicates no significant difference.

FIG 6

HDAC1-mediated deacetylation of H3K18 at the promoter of Beclin1 and Scd1 decreases their gene expression. Analysis of Scd1 (A) and Beclin1 (B) transcript levels by RT-qPCR. Shown is the relative gene expression in PA14 (green)-, mvfR (red)-, or pqsA (blue)-infected RAW 264.7 and hdac1 KD cells (dotted bars) compared to uninfected cells. The gene expression levels of Scd1 and Beclin1 in hdac1 KD cells (dotted) are significantly higher than in RAW 264.7-infected cells (plain). (C) Occurrence of H3K18ac marks on the Beclin1 and Scd1 promoters. ChIP-qPCR was used to show the enrichment of H3K18ac marks on the Beclin1 (orange) promoter and Scd1 promoter (yellow) in RAW 264.7 cells treated with 2-AA (400 mM) compared to untreated macrophages or HDAC1-knockdown treated macrophages. Each circle represents data from one replicate. The error bars denote ± SD. One-way ANOVA followed by Tukey’s HSD posttest was applied. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns indicates no significant difference. (D) Schematic representation of 2-AA effects on macrophages. Once P. aeruginosa is present in macrophages, 2-AA secures its intracellular sustenance through changes in the host autophagy by affecting autophagic nucleation and lipid biosynthesis. Alteration in macrophage lipid homeostasis results from changing the expression of the key lipogenic gene Scd1, which also affects the macrophage membrane, possibly through reduced monounsaturated fatty acid (MUFA) biosynthesis. The epigenetic eraser HDAC1 mediates inhibition of MUFAs biosynthesis and autophagy.