Effects of rifampicin on synthesis and functional activity of DNA-dependent RNA polymerase in Escherichia coli

Genome-wide screen reveals cellular functions that counteract rifampicin lethality in Escherichia coli

IMPORTANCE

Rifamycins are a group of antibiotics with a wide antibacterial spectrum. Although the binding target of rifamycin has been well characterized, the mechanisms underlying the discrepant killing efficacy between gram-negative and gram-positive bacteria remain poorly understood. Using a high-throughput screen combined with targeted gene knockouts in the gram-negative model organism Escherichia coli, we established that rifampicin efficacy is strongly dependent on several cellular pathways, including iron acquisition, DNA repair, aerobic respiration, and carbon metabolism. In addition, we provide evidence that these pathways modulate rifampicin efficacy in a manner distinct from redox-related killing. Our findings provide insights into the mechanism of rifamycin efficacy and may aid in the development of new antimicrobial adjuvants.

Proximate and ultimate causes of the bactericidal action of antibiotics

During the past 85 years of antibiotic use, we have learned a great deal about how these 'miracle' drugs work. We know the molecular structures and interactions of these drugs and their targets and the effects on the structure, physiology and replication of bacteria. Collectively, we know a great deal about these proximate mechanisms of action for virtually all antibiotics in current use. What we do not know is the ultimate mechanism of action; that is, how these drugs irreversibly terminate the 'individuality' of bacterial cells by removing barriers to the external world (cell envelopes) or by destroying their genetic identity (DNA). Antibiotics have many different 'mechanisms of action' that converge to irreversible lethal effects. In this Perspective, we consider what our knowledge of the proximate mechanisms of action of antibiotics and the pharmacodynamics of their interaction with bacteria tell us about the ultimate mechanisms by which these antibiotics kill bacteria.

Elongation by RNA polymerase II: structure-function relationship

RNA polymerase II is the eukaryotic enzyme that transcribes all the mRNA in the cell. Complex mechanisms of transcription and its regulation underlie basic functions including differentiation and morphogenesis. Recent evidence indicates the process of RNA chain elongation as a key step in transcription control. Elongation was therefore expected and found to be linked to human diseases. For these reasons, major efforts in determining the structures of RNA polymerases from yeast and bacteria, at rest and as active enzymes, were undertaken. These studies have revealed much information regarding the processes involved in transcription. Eukaryotic RNA polymerases and their homologous bacterial counterparts are flexible enzymes with domains that separate DNA and RNA, prevent the escape of nucleic acids during transcription, allow for extended pausing or "arrest" during elongation, allow for translocation of the DNA and more. Structural studies of RNA polymerases are described below within the context of the process of transcription elongation, its regulation and function.

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

The RNA polymerase subunits beta and beta' of Escherichia coli, encoded by the genes rpoB and rpoC, are co-transcribed with four 50 S ribosomal protein genes, rplKAJL. After treatment with the antibiotic rifampicin a partial uncoupling of rpoBC from rplKAJL transcription occurs. We have been investigating the role played in uncoupling by tL7, an 80% efficient terminator of transcription present in the 319 bp intercistronic space between rplL and rpoB, using S1 nuclease mapping of transcripts produced in vivo in normal (rpoBC haploid) strains. Our results show directly that rifampicin stimulates readthrough of tL7 on the chromosome by approximately twofold, an effect sufficient to explain the observed increase in beta beta' protein synthesis. We also provide preliminary evidence for the map position of PL7, and show that both this and P beta, two very weak promoters which might in principle be activated by rifampicin, are not in fact stimulated by the drug.

Regulation of alpha operon gene expression in Escherichia coli. A novel form of translational coupling

The alpha operon of Escherichia coli contains the genes for ribosomal proteins S13, S11, S4, RNA polymerase subunit alpha, and r-protein L17, in this order. Previous studies have shown that translation of all four ribosomal proteins is regulated by S4, and that binding of S4 to the mRNA at the start site for S13 translation is probably responsible for the regulation of translation of S13, S11 and S4. The alpha gene is "unique" in that it is located between the genes for two ribosomal proteins (S4 and L17) and yet appears to be regulated independently of them. In the present studies, we have measured the synthesis rates of all the alpha operon proteins under a variety of physiological conditions. Our results confirm that alpha gene expression is regulated independently of the co-transcribed ribosomal protein genes and is relatively insensitive to translational feedback repression by S4. S1 nuclease analysis of alpha operon mRNA failed to reveal the presence of any unique transcription start or mRNA cleavage that leads to separation of the alpha cistron from preceding ribosomal protein cistrons. Therefore, it appears that differential regulation of alpha synthesis takes place at the level of mRNA translation. We have also carried out a deletion analysis of the alpha operon leader and identified a region of the alpha operon leader mRNA that is required for regulation by S4. Furthermore, deletion of this region results in increased synthesis of L17 together with S13, S11 and S4, whereas alpha synthesis did not increase significantly. Therefore, we conclude that interaction of S4 with this single target site results in translational repression of not only the proximal three cistrons for S13, S11 and S4 but also that of the last cistron, L17, without affecting the intervening alpha cistron.

Overexpression of the dnaA gene in Escherichia coli B/r: chromosome and minichromosome replication in the presence of rifampin

The replication of chromosomes and minichromosomes in Escherichia coli B/r was examined under conditions in which the dnaA gene product was overproduced. Increased levels of the DnaA protein were achieved by thermoinduction of the dnaA gene, under the control of the lambda pL promoter, or by cellular maintenance of multicopy plasmids carrying the dnaA gene under the control of its own promoters. Previous work has shown that overproduction of DnaA protein stimulates replication of the chromosomal origin, oriC, but that the newly initiated forks do not progress along the length of the chromosome (T. Atlung, K. V. Rasmussen, E. Clausen, and F. G. Hansen, p. 282-297, in M. Schaechter, F. C. Neidhardt, J. L. Ingraham, and N. O. Kjeldgaard, ed., The Molecular Biology of Bacterial Growth, 1985). In the present study, it was found that overproduction of DnaA protein caused both a two- to threefold increase in the amount of residual chromosome replication and an extended synthesis of minichromosome DNA in the presence of rifampin. The amount of residual chromosome replication was consistent with the appearance of functional replication forks on the majority of the chromosomes. Since the rate of DNA accumulation and the cellular DNA/mass ratios were not increased significantly by overexpression of the dnaA gene, we concluded that the addition of rifampin either enabled stalled replication forks to proceed beyond oriC or enabled new forks to initiate on both chromosomes and minichromosomes, or both.

Sensitivity of DNA-repair-deficient strains of Escherichia coli to rifampicin killing

We have analyzed the role of RNA polymerase in DNA repair using the antibiotic rifampicin which binds specifically to the beta subunit of the enzyme. Several DNA-repair-deficient strains such as recA, uvr, and polA, and their isogenic parents were used for this study. All repair-deficient strains were found to be hypersensitive to rifampicin killing. Compared to the isogenic parent strains, recA strains are about 50 times more sensitive and the polA strain is about 100 times more sensitive to rifampicin killing. UvrA and uvrB strains are slightly more sensitive to rifampicin than the wild-type strains. The hypersensitivity of repair-deficient strains to rifampicin killing is totally abolished by the introduction of rifampicin-resistant mutations into these strains. We have examined the effect of rifampicin on RNA and protein synthesis in repair-deficient and -proficient strains. RNA and protein synthesis were found to be inhibited by rifampicin to the same extent among all the strains tested. The results also show that the resumption of DNA synthesis was significantly disrupted in DNA-repair-deficient strains following drug removal. Taken together these results suggest that RNA polymerase plays an essential role in DNA metabolism and such function may be replaced by polA and recA gene products and to a lesser extend by uvrA and uvrB gene products.

Expression of srnB gene of F plasmid by altered RNA polymerase in Escherichia coli

Degradation of otherwise stable rRNA and tRNA takes place in the presence of rifampin, dependent on the F plasmid srnB gene. We have reported that a protein newly synthesized in the presence of rifampin might be a product of the srnB gene required for stable RNA degradation (Ito, R. and Ohnishi, Y. (1983) Biochim. Biophys. Acta 739, 27-34). Here we have further studied the mechanism of srnB expression. Among eighteen mutants with altered RNA polymerase, two (TJ2470 (rpoC4) and TJ302 (rpoC56)) showed RNA degradation at high temperature (42 degrees C) when the srnB gene was present. Labeling proteins at 42 degrees C in strain TJ2470 indicated that a protein of molecular weight 12 000 was a product of the srnB gene, and that expression of the srnB gene provoked RNA degradation. Using plasmid pTK4, in which the srnB gene is inserted downstream of the promoter of lacZ, lac promoter-dependent expression of the srnB gene, with production of the putative protein product, also induced RNA degradation at 42 degrees C, with no requirement for added rifampin or altered RNA polymerase. RNA degradation in these conditions was quite similar to that in the case of the addition of rifampin; e.g., it showed some responses to Mg2+, temperature and RNAase I content of the cells. Expression of the srnB gene dependent on lac promoter was also observed in minicells. Thus, it is inferred that the srnB gene is probably repressed under normal conditions with its own promoter; its expression initiates RNA turnover.

Rifampicin-induced chromatid exchanges in Sarcoma-180 mouse ascites cells

Sarcoma-180 ascites tumour-bearing male mice were injected i.p. with single dose of 0.5, 2.5, 5 mg/kg body wt. of rifampicin. Cells were sampled for mitotic chromosome analysis 4, 16 or 24 h after treatment. The maximal yield of chromatid-type aberrations induced was found 24 h after treatment with 5 mg/kg of rifampicin. More than 60% of the cells carried at least one chromatid exchange. The majority of these were exchanges derived from breaks in the centromeric heterochromatin.

Translational control of the expression of the beta subunit gene of E. coli RNA polymerase

Using the plasmid pNF1337 as template, a mRNA preparation has been obtained that directs the in vitro synthesis of fMet-Val, the N-terminal dipeptide of the beta subunit of RNA polymerase. RNA polymerase holoenzyme specifically inhibits the mRNA-directed synthesis of fMet-Val showing that the autoregulation by RNA polymerase of beta, beta 1 synthesis is at the level of translation. L factor (nusA gene product) stimulates the synthesis of fMet-Val from a DNA template but not from mRNA. Rifampicin has no effect on the mRNA-directed synthesis of fMet-Val or the ability of RNA polymerase to inhibit fMet-Val synthesis.

Evidence that rifampicin can stimulate readthrough of transcriptional terminators in Escherichia coli, including the attenuator of the rpoBC operon

The genes encoding the beta and beta' subunits of RNA polymerase in E.coli, rpoB and rpoC, lie downstream of at least two ribosomal protein genes, rplJ (encoding L10) and rplL (L7/12), in a common operon. All four genes are served by promoter PL10, and an attenuator (partial terminator) of transcription, t1, lies between rplJL and rpoBC. Treatment of E.coli with rifampicin, under conditions producing partial inhibition of general RNA synthesis, can stimulate transcription of rpoBC. We have investigated the locus of this effect by fusing PL10 and t1 separately to galK, in suitable plasmids. Our studies of these fusions, and similar fusions involving transcriptional terminators derived from coliphage T7, indicate that low concentrations of rifampicin cause increased readthrough of several different transcriptional terminators in E.coli in vivo, including rpo t1. We discuss whether or not this unspecific mechanism is solely responsible for the observed stimulatory effects of the drug on rpoBC transcription.

Effects of olivacine on the metabolism of proteins and nucleic acids in Escherichia coli

Olivacine, a semi-synthetic isomer of ellipticine, interferes with the growth of Escherichia coli. Investigation of the effects of graded concentration shows that, in vivo, the synthesis of overall proteins is markedly more sensitive to low concentrations than is the synthesis of RNA and DNA, respectively. Furthermore, at moderate concentrations, olivacine is much more inhibitory to the synthesis of cytoplasmic proteins than to that of envelope proteins. At lethal concentration (10-4 M) all macromolecular syntheses are inhibited, DNA synthesis, however, to a lower extent. Olivacine blocks the synthesis of polyribonucleotide and polypeptide chains at the elongation step. The resulting nascent products of both transcription and translation are unstable, whereas completed RNA molecules are partially stabilized in the presence of the drug.

Negative regulation of beta and beta' synthesis by RNA polymerase

The genes for the beta and beta' subunits of RNA polymerase, rpoB and rpoC, and the genes for the two ribosomal proteins, rplL and rplJ, are part of the beta operon. Although this operon and contains a single strong promoter, the genes of the operon are not always coordinately expressed in vivo. This has now been confirmed in vitro where the lack of coordinate expression has been shown to be correlated with the selective inhibition of rpoB and rpoC gene expression by RNa polymerase. Rifampicin, which stops the initiation of transcription, also relieves this autogenous inhibition of beta and beta' (beta beta') synthesis. The inhibitory action of RNA polymerase and its reversal by rifampicin most likely occurs at a posttranscriptional or translation level.

Comparison of antibacterial and antiimmune effects of certain rifamycins

Comparison of the in vivo and in vitro immunosuppressive activities of the five rifamycins with their in vitro antibacterial and anti-ribonucleic acid polymerase activities indicated that correlation was poor. Examination of their activities on mitogen-induced blastogenesis in human peripheral blood leukocytes and inhibition of delayed-type hypersensitivity to partially purified protein derivative in immunized mice demonstrated that correlation was usually good. Antibacterial activity in cultures and the activities of the rifamycins inhibiting deoxyribonucleic acid-dependent ribonucleic acid polymerase appeared to correlate well. However, when these two types of activity, antiimmune and antibacterial, were compared, correlation was poor on occasion and indicated that the antiimmune activities and antibacterial activities of the rifamycins are probably not related.

Autogenous and post-transcriptional regulation of RNA polymerase synthesis

The regulation of gene expression was studied, for the Escherichia coli rpoBC operon, which includes the genes, rpoB and rpoC, for the beta and beta subunits of RNA polymerase, and rplJ and rplL, for the two proteins, L10 and L7/12, of the 50S ribosome. The gene organization agrees well with the accumulated observations indicating the coordinate synthesis of RNA polymerase and ribosomes under various growth conditions for wild-type E. coli cells. On the other hand, the differential regulation of the two essential components observed under restrictive growth conditions, after addition of various drugs or with certain mutants, in particular those carrying mutations in the RNA polymerase genes, was found to take place through two novel regulation systems: The transcriptional termination at an internal attenuation site and the two autogenous and posttranscriptional controls, being specific for the two ribosomal protein genes and the two RNA polymerase subunit genes, respectively. The majority of the transcription initiated from the promoter rpoP beta terminates at an attenuator site between the promoter-proximal rplJL and the promoter-distal rpoBC genes. The frequency of the attenuation seems to control the relative level of RNA polymerase synthesis to that of ribosomes. The expression of rpoBC genes is subject to an autogenous regulation, in which both RNA polymerase holoenzyme and alpha 2 beta complex function as regulatory molecules with repressor activity. The autogenous regulation was found to operate at post-transcriptional step(s), probably at the level of translation. During the study on the regulation of RNA polymerase synthesis, we noticed that the rpoBC operon contained another autogenous regulation circuit, in which the synthesis of L10 and L7/12 was specifically repressed by the L10-L7/12 complex. Molecular mechanisms and physiological meanings of the novel regulations are discussed.

Cloning of DNA of the rpoBC operon from the chromosome of Escherichia coli K12

We provide evidence that, in terms of transcriptional organisation, the rpoBC operon carried by lambdarifd 18 accurately represents the corresponding region of the E. coli K12 chromosome. A restriction fragment of E. coli K12 chromosomal DNA carrying the genes rpoBC (encoding the beta and beta' subunits of RNA polymerase) and rplL (coding for ribosomal proteins L7/L12) was cloned in a lambda vector, and the resulting phage tested for gene expression. In common with the corresponding fragment of lambdarifd 18 DNA, the chromosomal fragment has no strong promoter for rplL or rpoBC transcription. Another new phage was constructed by adding, to the restriction fragment carrying the rplL rpoBC structural genes from lambdarifd 18, a sequence from the E. coli K12 chromosome which includes a promoter for these genes. As in lambdarifd 18 itself, this promoter is shared with rplJ but not with rplKA. The properties of the latter phage also show that the dominant rifampicin-resistance characteristic of lambdarifd 18 results from more than one mutation.

The effect of rifampicin upon the transcription of RNA polymerase beta-gene in Escherichia coli

We studied the rate of synthesis of beta-and beta'-subunits of DNA-dependent RNA polymerase and the rate of beta-polypeptide mRNA synthesis in rifampicin-treated bacteria. The chosen antibiotic doses did not significantly inhibit the total RNA and protein synthesis in rifampicin-sensitive bacteria. For RNA-DNA hybridization experiments a pOD162 plasmid was constructed carrying a fragment of the rpoB gene and no other chromosome DNA regions. It was found that low doses of rifampicin cause an absolute and a relative increase in the rate of synthesis of the specific mRNA for the beta-subunit, suggesting a stimulation of the corresponding gene transcription and excluding the possibility of a less pronounced inhibition of the rpoB gene expression compared to that of most other genes. However the relative acceleration of transcription is substantially higher than the absolute one. The stimulating effect of rifampicin on the beta-polypeptide synthesis is also demonstrated in a coupled system of transcription and translation directed by lambda rifd47 DNA. The possible mechanisms of the rifampicin action are discussed.

Gene expression in Escherichia coli B/r during partial rifampicin-mediated restrictions of transcription initiation

The antibiotic rifampicin inhibits transcription initiation, but not the elongation and completion of nascent RNA transcripts. Addition of low concentrations of rifampicin only partially blocks initiation but at the same time specifically alters the general pattern of transcription in the culture. The transcription of genes specifying the beta and beta' subunits of RNA polymerase, and to a lesser extent of the genes specifying the RNA and protein components of the ribosome, was specifically stimulated relative to total transcription. In contrast, the transcription of the lactose operon was selectively reduced. These results are consistent with the ideas that the level of expression of the genes specifying the beta and beta' subunits is sensitive to the general rate of RNA synthesis in the culture, and that the expression of the beta and beta' RNA polymerase genes is related to the expression of ribosome component genes.

Rate of major protein synthesis during the cell cycle of Caulobacter crescentus

The rate of major protein synthesis was examined during the synchronous differentiation of Caulobacter crescentus. Total cell proteins were pulse-labeled with [35S]methionine at different times in the swarmer cell cycle and analyzed by sodium dodecyl sulfate- polyacrylamide gel electrophoresis. The rates of synthesis of total cell proteins and of about one-half of the individual major proteins examined increased through G1 and S periods but remained nearly constant during G2 period. The rates of synthesis of the other half of the individual major proteins either increased continuously throughout the swarmer cell cycle or doubled during S period. One stage-specific protein was also detected in late S period. For most of the major proteins examined, the rate of synthesis in the swarmer cell was less than that in the stalked cell. It seemed that, before the onset of G2 period, the Caulobacter cell was already able to synthesize each major protein at the additive rate of the two progeny cells. Compared to the stability of cellular proteins, the functional degradation rate of mRNA coding for individual major proteins was rapid, with half-lives of 0.4 to 5.8 min. It thus seems that the rate of major protein synthesis mainly reflects the transcriptional control of gene expression.

Escape synthesis of RNA polymerase subunits and termination factor rho following induction of prophage lambda in Escherichia coli

Synthesis of RNA polymerase subunits and of transcription termination factor p was studied after thermoinduction of prophage lambdac1857 located at several unusual sites on the chromosome of Escherichia coli. When a lysogen carrying the prophage at the bfe gene was induced at 42 degrees C, the rate of synthesis of core polymerase subunits (alpha, beta and beta') rapidly decreased, followed by a marked increase after about 10 min. The latter increase was observed specifically in the "bfe lysogen" and not in any of the other lysogens tested. Similarly, the rate of synthesis of p factor increased appreciably in the induced ilv lysogen carrying the prophage at the ilv gene, and possibly in the bfe lysogen as well, but not in other lysogens examined. Taken together with other evidence, these results suggest that the enhanced syntheses of beta and beta' subunits of RNA polymerase and of p factor observerd represent "escape synthesis", resulting from the close linkage of the prophage genome to the respective structural genes. In contrast, omega factor synthesis was stimulated upon induction of any of the lysogens used without respect to the site of prophage location, suggesting the involvement of an entirely different mechanism.

Induction of sigma factor synthesis in Escherichia coli by the N gene product of bacteriophage lambda

Thermoinduction of cells of E. coli carrying prophage lambdacI857 within the bfe gene brings about not only "escape synthesis" of core subunits of the DNA-dependent RNA polymerase (RNA nucleotidyltransferase, nucleosidetriphosphate:RNA nucleotidyltransferase, EC 2-7-7-6), but also a striking stimulation of sigma factor synthesis. The latter phenomenon, termed sigma induction, is generally observed after lambda phage infection or prophage induction. A series of experiments with various bacterial and phage strains led us to conclude that the N gene product of lambda is directly involved to the sigma induction. These and other results obtained with mutants defective in transcription termination factor rho suggest the involvement of a rho-sensitive site in the control of sigma gene expression in E. coli.