PixR, a Novel Activator of Conjugative Transfer of IncX4 Resistance Plasmids, Mitigates the Fitness Cost of mcr-1 Carriage in Escherichia coli

Abstract

The emergence of the plasmid-borne colistin resistance gene mcr-1 threatens public health. IncX4-type plasmids are one of the most epidemiologically successful vehicles for spreading mcr-1 worldwide. Since MCR-1 is known for imposing a fitness cost to its host bacterium, the successful spread of mcr-1-bearing plasmids might be linked to high conjugation frequency, which would enhance the maintenance of the plasmid in the host without antibiotic selection. However, the mechanism of IncX4 plasmid conjugation remains unclear. In this study, we used high-density transposon mutagenesis to identify factors required for IncX4 plasmid transfer. Eighteen essential transfer genes were identified, including five with annotations unrelated to conjugation. Cappable-seq, transcriptome sequencing (RNA-seq), electrophoretic mobility shift assay, and β-galactosidase assay confirmed that a novel transcriptional regulator gene, pixR, directly regulates the transfer of IncX4 plasmids by binding the promoter of 13 essential transfer genes to increase their transcription. PixR is not active under nonmating conditions, while the expression of the pixR, pilX3-4, and pilX11 genes increased 3- to 6-fold upon contact with recipient Escherichia coli C600. Plasmid invasion and coculture competition assays revealed the essentiality of pixR for spreading and persistence of mcr-1-bearing IncX4 plasmids in bacterial populations. Effective conjugation is crucial for alleviating the fitness cost exerted by mcr-1 carriage. The existence of the IncX4-specific pixR gene increases plasmid transmissibility while promoting the invasion and persistence of mcr-1-bearing plasmids in bacterial populations, which helps explain their global prevalence. IMPORTANCE The spread of clinically relevant antibiotic resistance genes is often linked to the dissemination of epidemic plasmids. However, the underlying molecular mechanisms contributing to the successful spread of epidemic plasmids remain unclear. In this report, we shine a light on the transfer activation of IncX4 plasmids. We show how conjugation promotes the invasion and persistence of IncX4 plasmids within a bacterial population. The dissection of the regulatory network of conjugation helps explain the rapid spread of epidemic plasmids in nature. It also reveals potential targets for the development of conjugation inhibitors.

Keywords: IncX4; conjugation; fitness; mcr-1; plasmids.

FIG 1

In-depth analysis of the PixR regulon. The first two tracks represent the Tn5 insertions sites in the mutant libraries before and after two consecutive mattings, respectively. Essential genes appear in black in the “essential genes” track. The region highlighted in blue encompasses the ones belonging to the transfer region. The following four tracks represent the RNA-seq read densities of E. coli BW25113/pHNSHP23 and BW25113/pHNSHP23ΔpixR carrying pBAD or pBAD-pixR with 0.2% arabinose induction. For each of these tracks, densities with a positive value correspond to the positive DNA strand, whereas densities with a negative value correspond to the negative DNA strand. The region highlighted in red encompasses all genes in the transfer region showing a drastic change in expression. The eighth track is the genetic map of pHNSHP23. Genes are color coded according to their predicted function as indicated in the key. The last three tracks represent the Cappable-seq results. Strains are detailed on the left. Blue and gray bars indicate the Cappable-seq density on the positive and negative DNA strands, respectively. The orange bar indicates the TSS of the pilX operon. For all tracks, pink dots at the summit of peaks indicate that the signal is beyond the fixed y axis maximal or minimal value.

FIG 2

Identification of genes contributing to IncX4 conjugation. (A) F index for every gene of pHNSHP23, based on HDTM data. F indices were calculated as the insertion index ratio of the normalized transposon insertion read counts in the transconjugant mutant library divided by that in the initial mutant library. The genes with an F index of >100 were considered essential for conjugation and are in blue. The F indices of the pilX, taxAC, and taxB genes were used to set the threshold. (B) Conjugation frequencies of pHNSHP23Δcds4, pHNSHP23Δcds9, pHNSHP23Δcds10, pHNSHP23Δcds16, and complemented strains in E. coli BW25113 or E. coli GDE8P261. The bars each represent the means and standard deviations from biological triplicates. A one-way analysis of variance (ANOVA) with Tukey’s multiple-comparison test was performed. (C) Multiple-sequence alignment of PixR with three members of the LuxR family of transcriptional regulators. TraR (NCBI:protein accession no. WP_001278699), TrlR (GenBank accession no. CUX06573), and QscR (NCBI:protein accession no. WP_088170053) are transcriptional regulators involved in quorum sensing from A. tumefaciens and Pseudomonas aeruginosa. (D) Effect of PixR on IncX4 plasmids conjugation under different conditions. Conjugation assays conducted on solid plates are shown on the left, whereas those performed overnight in LB broth are shown on the right. The bars represent the means and standard deviations from biological triplicates. Bars were compared using two distinct one-way ANOVA with Tukey’s multiple-comparison test. ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05. ns, not significant.

FIG 3

PixR regulon. (A and B) Differential expression analyses of pHNSHP23 genes upon pixR overexpression or deletion. (C) Differential expression analysis of BW25113 genes upon pixR overexpression. The x and y axes represent the log₂ fold change and the adjusted P value, respectively. The vertical lines represent a 2-fold change in expression. Genes were considered differentially expressed (and therefore were marked with blue dots) when they displayed a fold change of more than 2 as well as an adjusted P value of more than 0.05, indicated by the gray horizontal line. (D) Effect of pixR deletion on the mRNA level of pilX3-4, pilX11, taxB, and trbM. The bars represent the means and standard deviations from biological triplicates. Four distinct one-way ANOVAs with Tukey’s multiple-comparison test were performed to compare the relative mRNA levels of a given gene or operon under different conditions. Statistical significance is indicated as follows: ****, P < 0.0001. (E) mRNA levels of pixR, pilX3-4, and pilX11 with and without C600 contact. The bars represent the means and standard deviations from biological triplicates. Bars were compared using unpaired t test.

FIG 4

PixR directly activates the promoter of the pilX operon. (A) Schematic representation of the promoter of the pilX operon. The transcription start site (TSS) was identified by Cappable-seq. The −10 and −35 boxes were predicted by Softberry. (B) A 1% agarose gel loaded with the amplified fragments of cds9-pilX1 and trbM-pill11. A primer located at the 3′ end of trbM was used for reverse transcription. Genomic DNA (gDNA) and RNA were used as positive and negative controls, respectively. (C) β-Galactosidase activities of P_pilX. The activity of P_pilX was monitored from a transcriptional lacZ fusion (P_pilX-lacZ) in E. coli. Cultures were grown in LB broth with 0.2% arabinose at 37°C for 2 h to induce the expression of pixR. The bars represent the means and standard deviations from biological triplicates. They were compared using an unpaired t test (***, P < 0.001). (D) EMSA performed with the P_pilX fragment and PixR. The 6×His-tagged PixR protein was purified by nickel-nitrilotriacetic acid (Ni-NTA) affinity chromatography. Increasing concentrations of PixR protein were incubated with the P_pilX fragment for 1 h.

FIG 5

Ecological role of pixR for plasmid pHNSHP23 invasion and persistence. E. coli BW25113/pHNSHP23ΔpixR::cat (A), BW25113/pHNSHP23 (B), BW25113/pHNSHP23ΔpixRΔmcr-1::cat (E), and BW25113/pHNSHP45 (G) were mixed with a 1,000-fold excess of BW25113 without plasmid initially; BW25113/pHNSHP23ΔpixR::cat (C), BW25113/pHNSHP23ΔpixRΔmcr-1::cat (F), and BW25113/pHNSHP45 (H) were cocultured with BW25113/pHNSHP23 and a 1,000-fold excess of BW25113 without plasmid initially. (D) E. coli BW25113/pHNHSP23, BW25113/pHNSHP23ΔpixR::cat, and BW25113/pHNSHP23ΔpixRΔmcr-1::cat were competed with the reference strain BW25113 in vitro, separately. BW25113/pHNSHP23Δmcr-1 was competed with BW25113/pHNSHP23 in vitro. BW25113/pHNSHP23ΔpixRΔmcr-1::cat was competed with BW25113/pHNSHP23Δmcr-1::cat in vitro. All competitions assays were performed with three biological replicates.

FIG 6

Linear comparison of IncX plasmid transfer region. Genes are labeled in different colors based on their annotation. Shades of gray indicate percent sequence identity.

FIG 7

PixR’s key role in the persistence and invasion of mcr-1-bearing IncX4 plasmids. PixR enhances the conjugation efficiency of IncX4 plasmids by binding the promoter of a set of essential transfer genes to increase their expression. PixR has a strong effect on the persistence of mcr-1-bearing IncX4 plasmids in bacterial populations.