A Pseudomonas aeruginosa small RNA regulates chronic and acute infection

Abstract

The ability to switch between different lifestyles allows bacterial pathogens to thrive in diverse ecological niches^1,2. However, a molecular understanding of their lifestyle changes within the human host is lacking. Here, by directly examining bacterial gene expression in human-derived samples, we discover a gene that orchestrates the transition between chronic and acute infection in the opportunistic pathogen Pseudomonas aeruginosa. The expression level of this gene, here named sicX, is the highest of the P. aeruginosa genes expressed in human chronic wound and cystic fibrosis infections, but it is expressed at extremely low levels during standard laboratory growth. We show that sicX encodes a small RNA that is strongly induced by low-oxygen conditions and post-transcriptionally regulates anaerobic ubiquinone biosynthesis. Deletion of sicX causes P. aeruginosa to switch from a chronic to an acute lifestyle in multiple mammalian models of infection. Notably, sicX is also a biomarker for this chronic-to-acute transition, as it is the most downregulated gene when a chronic infection is dispersed to cause acute septicaemia. This work solves a decades-old question regarding the molecular basis underlying the chronic-to-acute switch in P. aeruginosa and suggests oxygen as a primary environmental driver of acute lethality.

Fig. 1. The expression level of PA1414 is the highest of the P. aeruginosa genes expressed during human chronic infections.

a, Differential gene expression of P. aeruginosa between human infections and common in vitro conditions. For each gene, the log₂[fold change] is plotted against the Wald test P value. Bubbles with black borders highlight the 30 signature genes of P. aeruginosa human infection. Bubble size indicates mRNA read abundance during human infections. Blue bars on the y axis indicate genes (with negative log₂[fold change] values) outside the scale range. Dashed grey lines indicate the cutoffs (−log₁₀[adjusted P value] > 2, |log₂[fold change]| > 1) for identifying differentially expressed genes. The upregulated (green) and downregulated (blue) genes in humans are colour-coded differently. Grey bubbles indicate genes that are not differentially expressed. b, Relative transcript abundance (TPM) of PA1414 compared to those of all other protein-coding sequences (CDSs) and non-coding sRNAs in 54 transcriptomes examined in a, sorted from the highest to the lowest PA1414 TPM. Average PA1414 TPM is indicated with dashed lines. c, PA1414 orthologues are found only in P. aeruginosa. Green and black bars indicate the presence and absence of orthologues, respectively. d, DNA sequence conservation of PA1414 and its neighbouring genes among 258 PA1414-containing P. aeruginosa isolates. Below, the percentage of orthologues identical to the PA1414 allele from PA14 at each nucleotide is shown. Source Data

Fig. 2. PA1414 encodes an oxygen-responsive sRNA that governs anaerobic ubiquinone biosynthesis.

a, PA1414 RNA abundance (TPM) and the corresponding ranking (out of all protein-coding genes and sRNAs) in 202 P. aeruginosa transcriptomes. Dashed lines indicate the cutoffs (PA1414 TPM > 10³, PA1414 ranking < 10²) for identifying transcriptomes with high PA1414 expression. b, β-galactosidase assay examining the induction of PA1414 under anaerobic and static conditions. P_PA1414-lacZ, lacZ transcriptionally fused to PA1414 promoter; P_PA1414*, PA1414 promoter containing mutations in the Anr-binding motif; MrT7, MAR2xT7 transposon. n ≥ 4 independent experiments. c, Northern blot analysis of PA1414 expression. Estimated size (nucleotides, nt) are indicated on the left. 5S rRNA served as a loading control. For gel source data, see Supplementary Fig. 1. d, Colony biofilm growth of ΔPA1414 and WT in different P. aeruginosa strain backgrounds. The presence and absence of oxygen are indicated. The CFU of ΔPA1414 was normalized against the CFU of the WT in each experiment (n = 4). A dashed line highlights the point where the CFU ratio = 1. e, Percentage of tetracycline-resistant (Tet^R) cells before and after daily passages under anaerobic conditions (n = 3). Tet^S, tetracycline sensitive; Δ, ΔPA1414. f, RNA-seq reads aligned to the PA1414 locus during standard in vitro growth and human infections. Blue shade indicates Rho-independent terminator in PA1414. g, Colony biofilm growth of ΔsicX harbouring different sicX mutations under anaerobic conditions. n = 4 independent experiments. h, Comparative proteomic study identifies the targets of SicX under both static and anaerobic conditions. i, β-galactosidase assay evaluating the translational control of SicX on ubiUVT under anaerobic conditions (n ≥ 4). j, β-galactosidase assay evaluating the transcription of ubiUVT in the absence of SicX or Anr under static conditions (n ≥ 8). k, Quantification of anaerobic UQ9 synthesis in different strains (n = 4). l, A model of dual regulation of ubiUVT expression by Anr and SicX. Error bars represent standard deviation from the mean. Significant differences (compared to the WT) are indicated with asterisks (*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; two-tailed Mann–Whitney test). Source Data

Fig. 3. SicX governs chronic–acute lifestyle transition in the mammalian host.

a, Schematic of the mouse chronic wound model (10-day infection). b, P. aeruginosa burden in wounds. Lines indicate means. c, P. aeruginosa burden in the spleen (y axis). Corresponding wound burden is indicated on the x axis. d, Survival curves of mice infected with WT (n = 24) and ΔsicX (n = 24). e, Restoration of ubiUVT translation in ΔsicX prevented dissemination, and deletion of ubiUVT in WT caused dissemination. f, P. aeruginosa spatial organization in wounds. C, core; E, edge. g, Schematic of the mouse pneumonia model (2-day infection). h, P. aeruginosa burden in lungs. Lines indicate means. i, P. aeruginosa burden in the spleen. Lines indicate means. Two-tailed Mann–Whitney test was used to compare the bacterial burden differences in b,c,h,i. Log-rank test and two-tailed Wilcoxon matched-pairs signed-rank test were used for d,f, respectively. Summary of n = 6 independent experiments for b–d; n = 3 independent experiments for e,f; n = 2 independent experiments for h,i. NS, not significant. Source Data

Fig. 4. SicX is a biomarker for chronic–acute lifestyle transition.

a, Schematics of in vivo dispersal of mature biofilms in mouse chronic wounds. b, Venn diagram of genes differentially expressed before (n = 3 animals) and after GH (n = 2) or cis-DA (n = 2) treatment. c, sicX (orange) is the top downregulated gene among the 132 differentially expressed genes identified under both GH and cis-DA treatments. d, Heat map of differentially expressed (DE) genes involved in alginate synthesis and secretion, anaerobic respiration and aerobic respiration. Aerobic terminal oxidases with low or high oxygen affinities are indicated. Blank cells indicate genes not identified as differentially expressed. e, A working model of how SicX allows P. aeruginosa to establish chronic local infection in a host environment with low oxygen. No or low SicX promotes systemic infection. Source Data

Extended Data Fig. 1. PA1414 orthologs are restricted in P. aeruginosa.

The maximum-likelihood tree is built with 16s rDNA sequences from representative strains across the Pseudomonas genus. Clusters of abundant species are indicated with colored boxes. Strains that were reported to have been isolated from human or hospital environments are highlighted.

Extended Data Fig. 2. The expression of PA1414 under low oxygen conditions.

a. Schematic of mutagenesis of the Anr binding motif. b. Figure 2a with the addition of five in vitro static growth transcriptomes (blue circles; the RNA-seq libraries were prepared following the same method used in our study of P. aeruginosa transcriptomes in humans). c. Planktonic growth (OD₆₀₀) of WT and ΔPA1414 in the presence or absence of oxygen. Mean values from 3 independent experiments are shown. d. RNA-seq reads aligned to the PA1414 locus (sicX) during in vitro aerobic and anaerobic growth. e. Colony biofilm growth of ΔsicX harboring different sicX mutations (as illustrated in Fig. 2g) under aerobic conditions. Error bars represent standard deviation from the mean. n = 4 independent experiments. Two-tailed Mann-Whitney test comparing WT and each other strain was performed (n.s., not significant). f. Northern blot analyses on different sicX alleles as depicted in Fig. 2g. The sRNA abundance was estimated by first normalizing against the 5S rRNA (loading control) and next comparing to the WT level. Anaerobic growth fitness was indicated with + (no defect) and − (defect). For gel source data, see Supplementary Fig. 1.

Extended Data Fig. 3. A proposed base-pairing interaction between SicX and the 5’ UTR of ubiUVT.

a. Predicted base-pairing between SicX and ubiUVT mRNA. Nucleotides highlighted in bold were subject to point mutations (A → U, U → A, G → C, C → G), which were assessed via the ubiUVT-lacZ translational reporter. Colony biofilms grown under anaerobic conditions were assessed for β-galactosidase activities as described. Heatmap displays the mean values. Nucleotides subject to compensatory base changes are highlighted with magenta borders. Predicted base-pairing regions are highlighted with dashed borders. b. β-galactosidase activities (individual values) of the experiment shown in (A) quantified as described. Potential compensatory mutations are highlighted in blue. Green dashed lines indicate the mean β-galactosidase activities in WT and ΔsicX. A. U., arbitrary unit. n = 5 independent experiments. c. β-galactosidase assays evaluating whether 5’ UTR mutations can alter ubiUVT translation independently of SicX during static growth. WT and ΔsicX harboring the ubiUVT-lacZ reporter (grown as anaerobic colony biofilms) were also assessed using this assay. n ≥ 3 independent experiments. d. β-galactosidase assays indicating that the presence or absence of Hfq does not affect the induction of ubiUVT translation by SicX. n ≥ 5 independent experiments. Error bars represent standard deviation from the mean. Two-tailed Mann-Whitney test was performed for b–c (n.s., not significant).

Extended Data Fig. 4. DNA sequence conservation of the ubiUVT operon.

a. The percent identity of 258 ubiUVT orthologs (compared to the ubiUVT allele from PA14) at the single-nucleotide scale. The sicX binding site at the 5’ UTR is highlighted. b. DNA sequence alignment of the upstream intergenic region of 258 ubiUVT orthologs. c. DNA sequence alignment of 258 sicX orthologs (only the sRNA region is shown). The seed region in SicX and the target region in the 5’ UTR of ubiUVT are indicated.

Extended Data Fig. 5. SicX expression in mouse chronic wounds.

a. RNA-seq reads aligned to the PA1414 locus (sicX) in mouse chronic wound infection. b. sicX RNA level in the mouse chronic wound is comparable to that in humans. Lines indicate means. Two-tailed Mann-Whitney test was used to compare the bacterial burdens (n.s., not significant). c. Bacterial burdens in the wound (dissemination information shown in Fig. 3e). Lines indicate means. n = 9 animals (for each group) over three independent experiments. Two-tailed Mann-Whitney test was used to compare the bacterial burdens of WT and each other mutant (n.s., not significant). Source Data

Extended Data Fig. 6. WT and ΔsicX showed similar spleen colonization in a systemic infection model.

P. aeruginosa cells (~10⁷ CFU) were subcutaneously injected into the mouse inner thigh to induce systemic infection. Bacterial burdens in spleens were assessed at 16 h post-infection. WT (n = 7) and ΔsicX (n = 8). Lines indicate means. Two-tailed Mann-Whitney test was used. Source Data

Extended Data Fig. 7. Decoupling the translation of ubiUVT from SicX.

a. Schematic of mutagenesis in the 5’ UTR of ubiUVT: SicX binding site, ribosome binding site, and start codon are colored differently; the stem-loop structure is highlighted; mutagenesis (ubi^C) is indicated with an arrow. b. Restoring ubiUVT translation in ΔsicX. β-galactosidase assay was performed to assess the translational activity of ubiUVT with and without 5’ UTR mutation as described in A (n = 6). c–d. Growth yield of colony biofilms in the presence or absence of oxygen (n = 4). Error bars represent standard deviation from the mean. Significant differences (compared to the WT) are indicated with asterisks (*, P < 0.05; **, P < 0.01; Two-tailed Mann-Whitney test).

Extended Data Fig. 8. SicX (in PAO1 strain background) governs chronic-to-acute transition.

a. Schematic of the mouse pneumonia infection experiment. b. P. aeruginosa burden in the lung. Lines indicate means. c. P. aeruginosa burden in the spleen. Lines indicate means. Two-tailed Mann-Whitney test was used to compare bacterial burdens. Summary of two biological replicates for b and c. Source Data

Extended Data Fig. 9. Functional enrichment of differentially-expressed genes in P. aeruginosa after GH or cis-DA treatment.

Gene Ontology (GO) terms are labeled on the left. One-tailed Fisher’s exact test was performed. Down, downregulated; up, upregulated.

Extended Data Fig. 10. Relatively low expression of Rsm sRNAs during infections.

a. Schematics of the Gac-Rsm cascade in P. aeruginosa. Components that promote biofilm (green) and planktonic (blue) lifestyles are color-coded. b. Relative transcript abundance (TPM, transcripts per million) of sicX compared to all protein-coding sequences (CDS), rsmYZ sRNAs, and the remaining non-coding sRNAs in 202 transcriptomes examined in Fig. 2a. Transcriptomes are sorted from the highest to the lowest sicX TPM. c. A heatmap showing the mean TPM of CDS, sRNAs, rsmYZ, and sicX in P. aeruginosa transcriptomes. d. β-galactosidase assays demonstrating the expression of sicX is independent from GacS-GacA. P_sicX-lacZ, lacZ transcriptionally fused to sicX promoter. MrT7, MAR2xT7 transposon. n = 4 in all experiments. Error bars represent standard deviation from the mean. Significant difference (compared to WT) was analyzed with a Two-tailed Mann-Whitney test (n.s., not significant).