Uncovering a hidden functional role of the XRE-cupin protein PsdR as a novel quorum-sensing regulator in Pseudomonas aeruginosa

Abstract

XRE-cupin family proteins containing an DNA-binding domain and a cupin signal-sensing domain are widely distributed in bacteria. In Pseudomonas aeruginosa, XRE-cupin transcription factors have long been recognized as regulators exclusively controlling cellular metabolism pathways. However, their potential functional roles beyond metabolism regulation remain unknown. PsdR, a typical XRE-cupin transcriptional regulator, was previously characterized as a local repressor involved solely in dipeptide metabolism. Here, by measuring quorum-sensing (QS) activities and QS-controlled metabolites, we uncover that PsdR is a new QS regulator in P. aeruginosa. Our RNA-seq analysis showed that rather than a local regulator, PsdR controls a large regulon, including genes associated with both the QS circuit and non-QS pathways. To unveil the underlying mechanism of PsdR in modulating QS, we developed a comparative transcriptome approach named "transcriptome profile similarity analysis" (TPSA). Using this TPSA method, we revealed that PsdR expression causes a QS-null-like transcriptome profile, resulting in QS-inactive phenotypes. Based on the results of TPSA, we further demonstrate that PsdR directly binds to the promoter for the gene encoding the QS master transcription factor LasR, thereby negatively regulating its expression and influencing QS activation. Moreover, our results showed that PsdR functions as a negative virulence regulator, as inactivation of PsdR enhanced bacterial cytotoxicity on host cells. In conclusion, we report on a new QS regulation role for PsdR, providing insights into its role in manipulating QS-controlled virulence. Most importantly, our findings open the door for a further discovery of untapped functions for other XRE-Cupin family proteins.

Fig 1. The expression of PsdR influences QS activity.

(A-C) Las, Rhl and PQS activity. The activities of the Las, Rhl and PQS QS systems are reflected by the fluorescence levels of the expressed reporters PlasR-GFP (A), PrhlA-GFP (B) and PpqsA-GFP (C), respectively. Fluorescence values were obtained from bacteria cultured in casein broth for 12 h using a flow cytometer (FITC channel). (D-F) The relative concentrations of QS-controlled metabolites, C4-HSL (D), pyocyanin (E) and hydrogen cyanide (F) in shown strains. Production of C4-HSL in WT was set to 100%. WT, wild-type strain PAO1; ΔpsdR, PsdR-null mutant; ΔpsdR::PrrnB::psdR, PsdR-null mutant carrying a single copy of psdR driven by the rrnB promoter; ΔpsdR::miniTn7, PsdR-null mutant carrying a single copy of empty miniTn7 vector. A one-way ANOVA with Bonferroni posttest was used for statistical analysis. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig 2. Transcriptome analysis of the PsdR-expressing cells.

(A) Volcano plots showing the magnitude of differential gene expression induced by the PsdR-expressing strain (WT::PrrnB::psdR) compared to the wild-type strain PAO1. Each dot represents one annotated sequence with detectable expression. Red dots, up-regulated genes; blue dots, down-regulated genes; green dots, genes that are not differentially expressed. Thresholds for defining a significance in differential expression (log₂ (fold change) ≧ |1.0|, P value ≦ 0.05) are shown as dashed lines. (B) KEGG pathway analysis of differentially expressed genes in the WT::PrrnB::psdR strain compared with wild-type strain. Gene count number and a negative log₁₀ P value shows the enrichment of differentially expressed genes in the corresponding pathway. (C-E) A negative log₂ (fold change) value indicates the down-regulation of genes encoding QS system components (C) involved in QS-controlled products (D) and associated with the Las core regulon [26] (E) in the WT::PrrnB::psdR strain compared with wild-type strain. WT::PrrnB::psdR, wild-type PAO1 strain carrying a single copy of psdR driven by the rrnB promoter. Data are presented as mean ± SD.

Fig 3. Gene expression profile in regulator transcriptomes.

(A-B) Expression shift of down-regulated (A) or up-regulated (B) genes in each regulator transcriptome. Transcriptome similarity was estimated using one-sided KS test. The graph shows the corresponding negative log₁₀-transformed P value for each regulator transcriptome. (C) Empirical cumulative distribution functions (ECDF) analysis for gene expression changes in PsdR-expressing strain compared with LasR-null, PhoB-null, PchR-null, RhlR-null and PqsR-null. Genes with a two-fold expression level change in the PsdR-expressing strain were applied to produce the plot. WT::PrrnB::psdR, wild-type PAO1 strain carrying a single copy of psdR driven by a rrnB promoter.

Fig 4. PsdR directly binds to the promoter region of lasR.

(A) Schematic diagram of the promoter region of the lasR gene. Cis-acting elements are colored. The arrow indicates the translation start site. Putative PsdR-binding sites are highlighted in red. (B) Sequence alignment of the promoter regions of mdpA and lasR. (C) SDS-PAGE of purified His-tagged PsdR protein. (D) EMSA results showing that the his-tagged PsdR protein and the probe DNA containing putative PsdR-binding site formed a protein-DNA complex. This protein-DNA complex was disrupted when the putative PsdR-binding site was deleted in the probe DNA. PlasR-WT, promoter DNA of the lasR gene containing the putative PsdR-binding site; PlasR-mut, lasR probe DNA deleting the putative PsdR-binding site. Probe DNA was either 3’end biotin-labeled or unlabeled. The protein-probe complex and unbound DNA probe and are indicated by arrows. The experiment was independently performed more than three times and obtained data were similar.

Fig 5. PsdR modulates QS activity in a LasR-dependent manner.

(A-B) Evaluation of respective Rhl and PQS QS activity in shown strains by using PrhlA-GFP (A) and PpqsA-GFP (B) reporter plasmids. Strains were grown in casamino acids medium. Fluorescence values were determined from the following strains: WT, wild-type PAO1 strain; ΔpsdR, PsdR-null mutant; ΔlasR, LasR-null mutant; ΔpsdRΔlasR, double deletion mutant. Data are means ± SD from a representative experiment (n ≧ 4). In some cases, the error bars are too small to be seen.

Fig 6. PsdR expression affects QteE-mediated LasR regulation.

The expression levels of lasR reflected by the PlasR-GFP report plasmid in shown strains. These strains were grown in casamino acids medium. WT, wild-type PAO1 strain; ΔpsdR, PsdR-null mutant; ΔqteE, QteE-null mutant; ΔqteEΔpsdR, qteE and psdR double deletion mutant; ΔqteEΔpsdR::PrrnB::psdR, qteE and psdR double deletion mutant containing a single copy of psdR driven by a rrnB promoter. Data are means ± SD from a representative experiment (n ≧ 4).

Fig 7. PsdR inactivation enhances host cell cytotoxicity.

The killing assay with human lung cancer Chinese hamster ovary (CHO) cells was performed with equal amounts of the indicated strains. After incubation for 6 h, the release of cytosolic lactate dehydrogenase (LDH) from infected cells was quantified. The released amount of LDH inoculated with the wild-type strain (WT) was set to 100%. A one-way ANOVA with Bonferroni posttest was used for statistical analysis. *P < 0.05, ** P < 0.01, *** P < 0.001.

Fig 8. Diagrams summarizing regulatory mechanisms and functions of PsdR.

(A) Diagrams illustrating that the lasR expression levels are regulated by PsdR and the effects on downstream gene expression. (B) The identified direct target genes of PsdR and their associated biological functions.