Direct evidence for rifampicin-promoted readthrough of the partial terminator tL7 in the rpoBC operon of Escherichia coli

Rifampicin exerts anti-mucoviscous activity against hypervirulent Klebsiella pneumoniae via binding to the RNA polymerase β subunit

OBJECTIVES

In hypervirulent Klebsiella pneumoniae (hvKP), the hypermucoviscous capsule is known to be a major virulence determinant. We previously discovered that rifampicin (RFP), a bactericidal drug that binds to and inhibits the β subunit of RNA polymerase (RpoB), elicits anti-mucoviscous activity against hvKP by suppressing rmpA, a regulator of capsule production. Here, we aimed to determine whether RFP exerts this effect at sub-growth-inhibitory concentrations via its binding to RpoB.

METHODS

Five spontaneous RFP-resistant mutants (R1-R5) were prepared from an hvKP clinical isolate and subjected to whole genome sequencing and mucoviscosity analyses. Subsequently, a two-step allelic exchange procedure was used to create a rpoB mutant R6 and revertants with wild-type rpoB from R1-R5 (named R1'-R5'). Transcription levels of rmpA and the capsular polysaccharide polymerase gene magA and capsule thickness of R1-R5 and R1'-R5' grown without or with RFP were evaluated by quantitative reverse transcription polymerase chain reaction and microscopic observation using India ink staining.

RESULTS

R1-R5 all had non-synonymous point mutations in rpoB and were highly resistant to the bactericidal effects and anti-mucoviscous activity of RFP. While the properties of R6 were similar to those of R1-R5, the responses of R1'-R5' to RFP were identical to those of the wild type. rmpA and magA transcription levels and capsule thickness correlated well with the mucoviscosity levels.

CONCLUSIONS

RFP exerts anti-mucoviscous activity by binding to RpoB. The mechanism of how this causes rmpA suppression remains to be explored.

Heterogeneous Distribution of Proton Motive Force in Nonheritable Antibiotic Resistance

Bacterial infections that are difficult to eradicate are often treated by sequentially exposing the bacteria to different antibiotics. Although effective, this approach can give rise to epigenetic or other phenomena that may help some cells adapt to and tolerate the antibiotics. Characteristics of such adapted cells are dormancy and low energy levels, which promote survival without lending long-term genetic resistance against antibiotics. In this work, we quantified motility in cells of Escherichia coli that adapted and survived sequential exposure to lethal doses of antibiotics. In populations that adapted to transcriptional inhibition by rifampicin, we observed that ~1 of 3 cells continued swimming for several hours in the presence of lethal concentrations of ampicillin. As motility is powered by proton motive force (PMF), our results suggested that many adapted cells retained a high PMF. Single-cell growth assays revealed that the high-PMF cells resuscitated and divided upon the removal of ampicillin, just as the low-PMF cells did, a behavior reminiscent of persister cells. Our results are consistent with the notion that cells in a clonal population may employ multiple different mechanisms to adapt to antibiotic stresses. Variable PMF is likely a feature of a bet-hedging strategy: a fraction of the adapted cell population lies dormant while the other fraction retains high PMF to be able to swim out of the deleterious environment. IMPORTANCE Bacterial cells with low PMF may survive antibiotic stress due to dormancy, which favors nonheritable resistance without genetic mutations or acquisitions. On the other hand, cells with high PMF are less tolerant, as PMF helps in the uptake of certain antibiotics. Here, we quantified flagellar motility as an indirect measure of the PMF in cells of Escherichia coli that had adapted to ampicillin. Despite the disadvantage of maintaining a high PMF in the presence of antibiotics, we observed high PMF in ~30% of the cells, as evidenced by their ability to swim rapidly for several hours. These and other results were consistent with the idea that antibiotic tolerance can arise via different mechanisms in a clonal population.

Linking system-wide impacts of RNA polymerase mutations to the fitness cost of rifampin resistance in Pseudomonas aeruginosa

Fitness costs play a key role in the evolutionary dynamics of antibiotic resistance in bacteria by generating selection against resistance in the absence of antibiotics. Although the genetic basis of antibiotic resistance is well understood, the precise molecular mechanisms linking the genetic basis of resistance to its fitness cost remain poorly characterized. Here, we examine how the system-wide impacts of mutations in the RNA polymerase (RNAP) gene rpoB shape the fitness cost of rifampin resistance in Pseudomonas aeruginosa. Rifampin resistance mutations reduce transcriptional efficiency, and this explains 76% of the variation in fitness among rpoB mutants. The pleiotropic consequence of rpoB mutations is that mutants show altered relative transcript levels of essential genes. We find no evidence that global transcriptional responses have an impact on the fitness cost of rifampin resistance as revealed by transcriptome sequencing (RNA-Seq). Global changes in the transcriptional profiles of rpoB mutants compared to the transcriptional profile of the rifampin-sensitive ancestral strain are subtle, demonstrating that the transcriptional regulatory network of P. aeruginosa is robust to the decreased transcriptional efficiency associated with rpoB mutations. On a smaller scale, we find that rifampin resistance mutations increase the expression of RNAP due to decreased termination at an attenuator upstream from rpoB, and we argue that this helps to minimize the cost of rifampin resistance by buffering against reduced RNAP activity. In summary, our study shows that it is possible to dissect the molecular mechanisms underpinning variation in the cost of rifampin resistance and highlights the importance of genome-wide buffering of relative transcript levels in providing robustness against resistance mutations.

Antibiotic resistance mutations carry fitness costs. Relative to the characteristics of their antibiotic-sensitive ancestors, resistant mutants show reduced growth rates and competitive abilities. Fitness cost plays an important role in the evolution of antibiotic resistance in the absence of antibiotics; however, the molecular mechanisms underlying these fitness costs is not well understood. We applied a systems-level approach to dissect the molecular underpinnings of the fitness costs associated with rifampin resistance in P. aeruginosa and showed that most of the variation in fitness cost can be explained by the direct effect of resistance mutations on the enzymatic activity of the mutated gene. Pleiotropic changes in transcriptional profiles are subtle at a genome-wide scale, suggesting that the gene regulatory network of P. aeruginosa is robust in the face of the direct effects of resistance mutations.

Rifampin-induced initiation of chromosome replication in dnaR-deficient Escherichia coli cells

The dnaR130 mutant of Escherichia coli, which was thermosensitive in initiation of chromosome replication, was capable of thermoresistant DNA synthesis in the presence of rifampin at a low concentration that allowed almost normal RNA synthesis. The DNA synthesis in the presence of the drug depended on protein synthesis at the high temperature. The protein synthesis in the dnaR-deficient cells provided a potential for thermoresistant DNA synthesis to be induced at a high dose of the drug that almost completely prevented RNA synthesis. The induced synthesis was synchronously initiated from oriC and proceeded semiconservatively toward terC. The replication depended on the dnaA function, which was essential for normal initiation of replication from oriC. The capability for drug-induced replication was abolished by certain rifampin resistance mutations in the beta subunit of RNA polymerase. Thus, the drug can induce the dnaA-dependent initiation of replication in the dnaR-deficient cells through its effect on RNA polymerase. This result implies that the dnaR product is involved in the transcription obligatory for the initiation of replication of the bacterial chromosome.

Determination of the growth rate-regulated steps in expression of the Escherichia coli K-12 gnd gene

In Escherichia coli K-12 strain W3110, the amount of 6-phosphogluconate dehydrogenase relative to that of total protein, i.e., the specific enzyme activity, increases about threefold during growth in minimal media over the range of growth rates with acetate and glucose as sole carbon sources. Previous work with gnd-lac operon and protein fusion strains indicated that two steps in the expression of the gnd gene are subject to growth rate-dependent control, with at least one step being posttranscriptional. With both Northern (RNA) and slot blot analyses, we found that the amount of gnd mRNA relative to that of total RNA was 2.5-fold higher in cells growing in glucose minimal medium than in cells grown on acetate. Therefore, since the total mRNA fraction of total RNA is essentially independent of the growth rate, the amount of gnd mRNA relative to that of total mRNA increases about 2.5-fold with increasing growth rate. This indicates that most of the growth rate-dependent increase in 6-phosphogluconate dehydrogenase can be accounted for by the growth rate-dependent increase in gnd mRNA level. We measured the decay of gnd mRNA mass in the two growth conditions after blocking transcription initiation with rifampin and found that the stability of gnd mRNA does not change with growth rate. We also used a gnd-lacZ protein fusion to measure the functional mRNA half-life and found that it too is growth rate independent. Thus, the growth rate-dependent increase in the level of gnd mRNA is due to an increase in gnd transcription, and this increase is sufficient to account for the growth rate regulation of the 6-phosphogluconate dehydrogenase level. The dilemma posed by interpretations of the properties of gnd-lac fusion strains and by direct measurement of gnd mRNA level is discussed.

Transcription frequency modulates the efficiency of an attenuator preceding the rpoBC RNA polymerase genes of Escherichia coli: possible autogenous control

Expression of the rpoBC genes encoding the beta and beta' RNA polymerase subunits of Escherichia coli is autogenously regulated. Although previous studies have demonstrated a post-transcriptional feedback mechanism, complex transcriptional controls of rpoBC expression may also contribute. We show that an attenuator (rpoBa) separating the ribosomal protein (rpl) genes from the rpoBC genes in the rplKAJLrpoBC gene cluster is modulated in its efficiency in response to changes in the frequency of transcription initiated by promoters located upstream. A series of rplJLrpoBalacZ transcriptional fusions was constructed on lambda vectors in which transcription into the rpoBa attenuator was varied by using a variety of promoters with different strengths. beta-galactosidase assays performed on monolysogens of the recombinant phage show that with transcription increasing over a 40-fold range, readthrough of rpoBa decreases from 61% to 19%. In contrast, two other well-characterized terminators show nearly constant efficiencies over a similar range of transcription frequencies. Using a set of phage P22 ant promoter variants with single-nucleotide changes in the promoter consensus sequences also demonstrates that the modulation of rpoBa function appears to be unrelated to the phenomenon of 'factor-independent antitermination' reported by others. The implications for autogenous control of RNA polymerase synthesis are discussed.

In vivo analysis of overlapping transcription units in the rplKAJLrpoBC ribosomal protein-RNA polymerase gene cluster of Escherichia coli

Transcription of the rplKAJLrpoBC ribosomal protein (rpl) RNA polymerase (rpo) gene cluster is governed by a complex set of signals. To dissect the transcription units active in vivo and to quantify the relative contribution of each, an extensive array of rplKAJLrpoB/lacZ gene fusions were constructed on lambda phage derivatives and introduced in single copy into the chromosomes of lac- cells. Measurements of beta-galactosidase production from fusions containing wild-type and/or mutagenized rplrpo DNA fragments permitted the establishment of high-resolution transcription profiles of the gene cluster. The results show that transcription initiated at the upstream rplKp promoter (located just before rplK) does not terminate before the rplJp promoter (located upstream from rplJ), but instead reads through into the distal genes. In addition, rplJp continues to function efficiently in the presence of readthrough transcription from rplKp. As a result the rplJL genes are transcribed at almost twice the frequency of the upstream rplKA genes. However, the transcription of rpoB, which is situated downstream from the previously identified attenuator (rpoBa), is only marginally increased (20%) when both promoters are present. This suggests that although both transcription units overlap, transcriptional termination at rpoBa is modulated in response to the frequency of initiation from both promoters.

Regulation of proline utilization in enteric bacteria: cloning and characterization of the Klebsiella put control region

Enteric bacteria can grow on proline as the sole nitrogen and carbon source. Expression of the proline utilization (put) operon in Klebsiella strains and Escherichia coli is responsive to nitrogen regulation. In contrast, Salmonella typhimurium cannot activate put operon expression when growing in medium with glucose as a carbon source and proline as the sole nitrogen source. To compare nitrogen regulatory sites in the control regions of the put operons in these three closely related genera, we cloned the Klebsiella put operon onto a plasmid. The putA and putP genes were localized on the plasmid by transposon mutagenesis. The DNA sequence of the put control region was determined and compared with those of the put control regions from S. typhimurium and E. coli. The overall size and organization of the put control region were very similar in all three bacteria. However, no obvious ntr regulatory sites were found in this region, and transcription of the put genes started at the same sites during growth with limiting or excess nitrogen. These results strongly suggested that the Klebsiella put operon may not be directly regulated by the ntr system.

Regulation of proline utilization in Salmonella typhimurium: a membrane-associated dehydrogenase binds DNA in vitro

The PutA protein is a membrane-associated enzyme that catalyzes the degradation of proline to glutamate. Genetic evidence suggests that in the absence of proline, the PutA protein also represses transcription of the putA and putP genes. To directly determine whether PutA protein binds to the put control region, we analyzed gel retardation of put control region DNA by purified PutA protein in vitro. The put control region is 420 bp. Purified PutA protein bound specifically to several nonoverlapping fragments of control region DNA, indicating the presence of multiple binding sites in the control region. Electrophoretic abnormalities and behavior of circularly permuted fragments of control region DNA indicate that it contains a region of intrinsically curved DNA. To determine whether the multiple binding sites or the DNA curvature are important in vivo, two types of deletions were constructed: (i) deletions that removed sequences predicted to contribute to DNA curvature as well as potential operator sites and (ii) deletions that removed only potential operator sites. Both types of deletions increased expression of the put genes but were still induced by proline, indicating that multiple cis elements are involved in repression. These data suggest a model for put repression that invokes the formation of a complex between PutA protein molecules bound at different sites in the control region, brought into proximity by a loop of curved DNA.

Isolation and mapping of Escherichia coli mutations conferring resistance to division inhibition protein DicB

Temperature-sensitive dicA mutants of Escherichia coli, dicA1(Ts), are blocked for cell division, owing to derepressed expression of a division inhibition gene, dicB. We isolated mutants which survived a high temperature in the dicA1 background and which survived induced expression of dicB carried by a high-copy-number plasmid. Most of the mutations conferred very slow growth on the cells. Two were mapped to the 90-min cluster of genes involved in translation and transcription, in or very close to gene rpoB. The majority of the other mutations were found to cause variable degrees of minicell formation and to map within or very close to the minB locus. Contrary to these mutations, the canonical min-1 mutation did not confer resistance to DicB.