doi: 10.1038/s41467-018-04472-6.
Activity of acetyltransferase toxins involved in Salmonella persister formation during macrophage infection
Affiliations
- PMID: 29777131
- PMCID: PMC5959882
- DOI: 10.1038/s41467-018-04472-6
Item in Clipboard
Activity of acetyltransferase toxins involved in Salmonella persister formation during macrophage infection
Julian A Rycroft et al.
Nat Commun.
.
Display options
Format
Display options
Format
Abstract
Non-typhoidal Salmonella strains are responsible for invasive infections associated with high mortality and recurrence in sub-Saharan Africa, and there is strong evidence for clonal relapse following antibiotic treatment. Persisters are non-growing bacteria that are thought to be responsible for the recalcitrance of many infections to antibiotics. Toxin-antitoxin systems are stress-responsive elements that are important for Salmonella persister formation, specifically during infection. Here, we report the analysis of persister formation of clinical invasive strains of Salmonella Typhimurium and Enteritidis in human primary macrophages. We show that all the invasive clinical isolates of both serovars that we tested produce high levels of persisters following internalization by human macrophages. Our genome comparison reveals that S. Enteritidis and S. Typhimurium strains contain three acetyltransferase toxins that we characterize structurally and functionally. We show that all induce the persister state by inhibiting translation through acetylation of aminoacyl-tRNAs. However, they differ in their potency and target partially different subsets of aminoacyl-tRNAs, potentially accounting for their non-redundant effect.
Conflict of interest statement
The authors declare no competing interests.
Figures
Non-typhoidal Salmonella strains induce persister formation following internalization by human macrophages. a Fold increase in persisters caused by 30 min internalization in primary human macrophages relative to levels in inocula in clinical isolates of S. Typhimurium (white and light grey bars) and S. Enteritidis (dark grey bars), measured after exposure to gentamicin. Data represent the mean ± SEM (n ≥ 5) and were analysed using one-way ANOVA. b Comparison of the repertoire of class II TA modules between S. Typhimurium and S. Enteritidis serovars. Boxes of same colour—fully conserved; white boxes—absence; different colour boxes—polymorphism. c Proportion of bacteria surviving 4 h exposure to bactericidal concentrations of cefotaxime in cultures of S. Typhimurium 12023 wild-type, ΔtacAT2, ΔtacAT2 pBAD33::tacT2STm, ΔtacAT2 pBAD33::tacT2SEn, ΔtacAT2 pBAD33::tacT2SEn and pCA24N::tacA2, or ΔtacAT3, ΔtacAT3 pBAD33::tacT3, ΔtacAT3 pBAD33::tacT3 and pCA24N::tacA3. Arabinose and IPTG were added to all cultures in the fresh medium during the lag phase, then antibiotic treatment started 1 h later. Data represent the mean ± SEM (n ≥ 3) and were analysed using a Student’s t test, compared to WT (ns, non-significant; ***p < 0.005)
TacT2SEn and TacT3 are functional acetyltransferase toxins. a Left panel: Growth curves of S. Typhimurium 12023 ΔtacTA2 carrying pBAD33 (−TacT2), pBAD33::tacT2STm (+TacT2STm), pBAD33::tacT2SEn (+TacT2SEn), pCA24N::TacA2 (+TacA2) or pBAD33::tacT2SEn and pCA24N::tacA2 (+TacT2SEn + TacA2). Right panel: Growth curves of S. Typhimurium 12023 ΔtacTA3 carrying pBAD33 (−TacT3), pBAD33::tacT3 (+TacT3), pCA24N::TacA3 (+TacA3) or pBAD33::tacT3 and pCA24N::tacA3 (+TacT3 + TacA3). All cultures were supplemented with arabinose and IPTG in fresh rich medium during the lag phase and growth was monitored by optical density. b Left panel: Growth curves of S. Typhimurium 12023 ΔtacTA2 expressing from pBAD33, the wild-type S. Enteritidis toxin (+TacT2SEn) or point mutant toxins (+TacT2Y137FSEn) or (+TacT2R88GSEn), or carrying the empty vector (−TacT2). Right panel: Growth curves of S. Typhimurium 12023 ΔtacTA3 expressing from pBAD33, the wild-type toxin (+TacT3), or point mutant toxins (+TacT3Y143F) or (+TacT3R94E), or carrying the empty vector (−TacT3). All cultures were supplemented with arabinose in fresh rich medium during the lag phase. c Left panel: Cartoon representation of the dimeric structure of TacT3Y143F. Chain A is coloured from blue at the N terminus to red at the C terminus, and Chain B is coloured light grey. Right panel: Cartoon superimposition of TacT3Y143F (coloured) and TacTY140F (grey). TacT3- and TacT-bound Ac-CoA molecules are shown as black and grey sticks, respectively. (PDB: 6G96). d Electrostatic potential of the surface of the modelled TacT2SEn (left panel) or TacT3 (right panel) dimer showing positive potential in blue and negative in red. Residues mentioned in the text are labelled. a, b Data represent the mean ± SEM (n ≥ 3) and were analysed using a Student’s t test (ns, non-significant)
Tac toxins block translation through acetylation of aminoacyl-tRNAs. a Cell-free expression assays leading to the production of the control protein DHFR (red asterisk) from template DNA without toxins or with purified TacT, TacT2STm, TacT2SEn or TacT3, added from the onset of the assay. [14C]Ac-CoA was added to all the samples. All samples were analysed by SDS-PAGE, and the production of DHFR was revealed by Coomassie staining (CM). tRNAs extracted from the samples were analysed by acid-urea PAGE and revealed by methylene blue staining (MB) and acetylation tracked by autoradiography (AR). Equivalent amount of toxins added to the sample was tested by western blotting against His tag of the toxins (WB). Quantification of inhibition of the production of DHFR (spotted bars) and acetylation of tRNAs (dashed bars) in all conditions is reported in the bar charts, where data represent the mean ± SEM (n ≥ 3). b Surface of the Ac-CoA-binding site of TacT3 with Ac-CoA represented as solid sticks for the orientation in TacT3 and shadowed for the orientation of Ac-CoA in TacT. Trp142 is in yellow and Gly145 is in red. c Exposure of tRNA molecules acetylated by TacTs to Pth treatment in vitro. tRNA molecules acetylated by different TacTs in vitro were subsequently incubated with purified Pth, acetylation was assessed by autoradiography, before or after samples were treated with purified Pth. All the samples were supplemented with [14C]Ac-CoA. Treated tRNA molecules were separated on acid-urea polyacrylamide gel and revealed by methylene blue staining (MB) (top panel). Acetylation was tracked by autoradiography (AR) (lower panel). d Model of activity of Tac toxins
TacT2SEn acetylates aminoacyl-tRNAs more efficiently than TacT2STm. a TacT2 polymorphism. Electrostatic potential of the surface of the modelled TacT2SEn (top panels) or TacT2STm (lower panel) dimer showing positive potential in blue and negative in red. Arrow points toward the polymorphic amino acid. A 45° rotation about the y-axis from the first view of TacT2SEn showing the negative electrostatic potential of E29 in TacT2STm. b Extracted tRNA molecules from S. Typhimurium 12023 ΔtacATtacAT2tacAT3 were treated with equivalent amounts of TacT2STm or TacT2SEn as assessed by western blotting. [14C]Ac-CoA was added to all samples from the onset of the assay and the pH of the reactions was set over a 5.9–7.5 range. Samples were analysed by acid-urea PAGE, tRNAs revealed by methylene blue staining (MB) (top panel) and acetylation tracked by autoradiography (AR) (lower panel). Quantification of acetylation of tRNAs in all conditions is reported in the bar chart, where data represent the mean ± SEM (n = 3). c TacT2SEn (red) K29 and TacT (grey) K31 form two hydrogen bonds between the epsilon amine group of the lysine and carbonyl oxygens from peptide bonds on a neighbouring chain. In a model of TacT2STm (blue), no hydrogen bond can be formed with E29. In TacT3 (orange), a stabilizing hydrogen bond is formed between Q32 and the side chain of the neighbouring R17. Growth curves of S. Typhimurium 12023 ΔtacAT expressing from pBAD33, tacT (+TacT), point mutant toxin (+TacTK31E) or carrying the empty vector (−TacT), or ΔtacAT2 expressing from pBAD33 the Enteritidis wild-type toxin (+TacT2SEn) or carrying the empty vector (−TacT2). All cultures were supplemented with arabinose in fresh rich medium during lag phase. Data represent the mean ± SEM (n ≥ 3). d Binding of the two isoforms of TacT2 to tRNA molecules. Equivalent amounts of purified His-tagged inactive toxins TacT2STmY137F and TacT2SEnY137F were incubated with tRNA molecules extracted from S. Typhimurium 12023 ΔtacATtacAT2tacAT3 and subsequently pulled down with anti-His antibody. The input and output tRNA molecules were analysed by acid-urea PAGE and revealed by methylene blue staining (MB—top panels). Equivalent pulldown of the two toxin isoforms was tested by SDS-PAGE and Coomassie staining (CM—lower panel). Quantification of the amount of tRNA molecules pulled down by the toxins is reported in the bar chart where data represent the mean ± SEM (n ≥ 3)
The three Tac toxins promote acetylation of specific subsets of aa-tRNAs identification of amino acids acetylated by the different toxins. a Toxin-corrupted aminoacyl-tRNAs were extracted from cell-free expression assays, and treated with purified Pth to release the acetylated amino acids from tRNAs and subsequently analysed by LC-MS. b Specificity of TacT, TacT2SEn, TacT2STm and TacT3. Size of pie represents the relative amounts of acetylated amino acids recovered
References
Publication types
MeSH terms
Substances
Grants and funding
LinkOut - more resources
Full Text Sources
Other Literature Sources
Medical