Enterococcal PrgU Provides Additional Regulation of Pheromone-Inducible Conjugative Plasmids

Abstract

Efficient horizontal gene transfer of the conjugative plasmid pCF10 from Enterococcus faecalis depends on the expression of its type 4 secretion system (T4SS) genes, controlled by the P_Q promoter. Transcription from the P_Q promoter is tightly regulated, partially to limit cell toxicity caused by overproduction of PrgB, a T4SS adhesin. PrgU plays an important role in regulating this toxicity by decreasing PrgB levels. PrgU has an RNA-binding fold, prompting us to test whether PrgU exerts its regulatory control through binding of prgQ transcripts. We used a combination of in vivo methods to quantify PrgU effects on prgQ transcripts at both single-cell and population levels. PrgU function requires a specific RNA sequence within an intergenic region (IGR) about 400 bp downstream of P_Q. PrgU interaction with the IGR reduces levels of downstream transcripts. Single-cell expression analysis showed that cells expressing prgU decreased transcript levels more rapidly than isogenic prgU-minus cells. PrgU bound RNA in vitro without sequence specificity, suggesting that PrgU requires a specific RNA structure or one or more host factors for selective binding in vivo. PrgU binding to its IGR target might recruit RNase(s) for targeted degradation of downstream transcripts or reduce elongation of nascent transcripts beyond the IGR. IMPORTANCE Bacteria utilize type 4 secretion systems (T4SS) to efficiently transfer DNA between donor and recipient cells, thereby spreading genes encoding antibiotic resistance as well as various virulence factors. Regulation of expression of the T4SS proteins and surface adhesins in Gram-positive bacteria is crucial, as some of these are highly toxic to the cell. The significance of our research lies in identifying the novel mechanism by which PrgU performs its delicate fine-tuning of the expression levels. As prgU orthologs are present in various conjugative plasmids and transposons, our results are likely relevant to understanding of diverse clinically important transfer systems.

Keywords: conjugation; regulation; type 4 secretion systems.

FIG 1

Schematic overview of the prgQ operon and its regulation. (A) Schematic illustration of the genetic organization of the entire prgQ operon (not to scale). Note that prgU is situated within the cassette containing the surface adhesin, even though PrgU belongs to the regulatory proteins. (B) The inhibitory peptide I is transcribed from the pCF10 plasmid (#1), whereas the inducing C peptide is transcribed from the genome of all E. faecalis cells (#2). Both I and C peptides are secreted to the outside of the cell, where the C peptide from donor cells gets partially degraded by the PrgY protease (#2) (38, 39). Both I and C peptides are subsequently bound by PrgZ and imported via a permease (23). The transcriptional regulator PrgX is a dimer in its apo state but tetramerizes upon binding either I or C peptide. Depending on which pheromone is bound, it either represses transcription of the prgQ operon (PrgX/I) or induces it (PrgX/C) (#3) (22, 38, 40). Induction of the prgQ operon produces 3 transcripts, Q_S, Q_L and Full (#4) (40). The prgX operon also produces two transcripts, one of which is the anti-Q RNA that aids the formation of a terminator structure of the prgQ operon at the inverted repeat sequence 1 (IRS1) (#5) (7). In uninduced cells, anti-Q levels are sufficient to interact with all nascent prgQ transcripts, favoring formation of the IRS1 structure and transcript termination. Pheromone induction leads to the production of many more prgQ transcripts, which then overwhelm the pool of anti-Q. In the unpaired prgQ transcripts, IRS1 does not form, and transcription extends through the rest of the operon. The findings that we present in this article show that PrgU interacts with a region in the IGR, in between IRS1 and IRS2, thereby decreasing the expression of proteins transcribed from downstream genes (#6).

FIG 2

PrgU inhibition of gene expression depends on the intergenic region (IGR) and is independent of PrgX. (A) Schematic overview of the plasmids used in this study. The pMC plasmids carry the P_Q promoter with its regulatory region and the transcriptional lacZ reporter. This reporter gene is positioned at the start site of prgR in pMC2 and pMC9 (Fig. 1B) and at the 3′ end of prgQ in pMC3. In pMC9 the prgX transcriptional repressor upstream from the P_Q promoter is deleted. (B) β-Galactosidase activities originating from expression of the lacZ reporter gene on the pMC plasmids listed in panel A (coded in grayscale) in OG1RF cells without a pCF10 plasmid (−) or with wild-type pCF10 or pCF10ΔprgU. These cells also contain the pDL278p23 vector (P₂₃) or this vector constitutively expressing prgU (P₂₃::U). n = 3 independent biological replicates, and the error is the standard deviation.

FIG 3

Effects of PrgU on pCF10 transcription at the population and single-cell levels. (A and B) HCR analysis of prgB and prgG transcripts in single cells as a quantification of the mean fluorescence intensity per cell of 500 random cells, using hybridization probes against prgB (A) and pcfG (B) transcripts. Thresholds (thin dotted lines) used are 30 AU and 50 AU, respectively. Transcription was induced with (+) 10 ng/ml peptide C (cCF10 pheromone) for 30 or 60 min. Uninduced cells (−) at 0 min and pCF10ΔprgU^Res-containing cells (which cannot be induced) were used as negative controls. pCF10ΔprgABUC-containing cells are a negative control for prgB. (C) mRNA reads in the prgA to -C region in induced pCF10 wild-type and pCF10ΔprgU cells. The height of the blue lines indicates the transcription level in each strain. The complete RNA-seq data for this experiment are available in Data Sets S1 and S2.

FIG 4

PrgU expression reduces the formation of full-length [Full (pMC2)] transcript and leads to formation of Q_L and, at high concentrations, an increase in Q_S transcripts. Northern blot analyses of total RNA extracts probed with an oligonucleotide specific for the 5′ end of the IGR or 5S RNA as a loading control. All strains analyzed were OG1RF cells containing the pMC2 plasmid, induced by addition of exogenous C peptide. In one strain, no additional plasmid was present (−). The other strains contained either the pMB11 vector that constitutively expresses prgU (P₂₃::prgU), wild-type pCF10, or pCF10ΔprgU. Note that Q_S and Q_L are also formed from the pCF10 plasmids, while the full-length transcript produced from pCF10 is much larger and not visible here.

FIG 5

PrgU-FLAG binds to the Q_L and part of the Q_S transcripts. Cell lysates from C peptide-induced OG1RF(pCF10) strains carrying either pMB11 (expresses PrgU from the P₂₃ promoter, P₂₃::prgU) or pMC10 (expresses FLAG-tagged PrgU from the P₂₃ promoter, P₂₃::prgU-FLAG) were mixed with magnetic beads coated with FLAG antibodies. (A) Samples from the total cell lysates (input) and the elution fractions from the washed beads (pulldown) were subjected to Northern blot analysis using an oligonucleotide protein probe specific for the 5′ end of IGR RNA. Transcripts corresponding to the size of Q_S and Q_L bound to the magnetic beads in the presence of FLAG-tagged PrgU but not with the nontagged control. Note that full-length transcripts of the prgQ operon from pCF10 are much larger and not visible here. MW, molecular weight. (B) RNA-seq analysis of the RNA from PrgU pulldown fractions from lysates of either C peptide-induced (+) or noninduced (−) cells. PrgU-FLAG1 and -2 represent two biologically separate pulldowns. The top panel graphically depicts the relative read counts in the prgQ region, whereas the lower panel shows an enlargement with the RNA sequence. The sequence corresponding to IRS1 is highlighted by a black box, and the putative 3′ end of the prominent pulldown product is indicated by an arrow.

FIG 6

Oligomeric state of PrgU. (A) GEMMA of 0.01 mg/ml PrgU. The determined molecular masses are shown above the peaks. (B) SEC-MALS profile of PrgU, loaded at a protein concentration of 1 mg/ml, showing a molecular mass of 29 kDa over the range of the peak.

FIG 7

PrgU binding to IGR. (A) EMSA with dsDNA of the IGR or an equally long control dsDNA with increasing concentrations of PrgU-His (0 to 13 μM). In all cases, 100 nM dsDNA was used along with IGR (lanes 1 to 14) or control (lanes 16 to 29). Lanes 1 and 16 contain only 100 nM DNA, ds-IGR or ds-control, respectively, and lane 15 contains only the highest concentration of protein used in this assay (13 μM PrgU). (B) Data from MST experiment with a constant concentration of His-labeled PrgU (50 nM) and a varied concentration (between 0 and 40 μM) of the nonlabeled binding partner (RNA IGR and control RNA). n = 3 independent measurements; error bars represent the standard deviation.