Transcriptional regulation and signature patterns revealed by microarray analyses of Streptococcus pneumoniae R6 challenged with sublethal concentrations of translation inhibitors

Abstract

The effects of sublethal concentrations of four different classes of translation inhibitors (puromycin, tetracycline, chloramphenicol, and erythromycin) on global transcription patterns of Streptococcus pneumoniae R6 were determined by microarray analyses. Consistent with the general mode of action of these inhibitors, relative transcript levels of genes that encode ribosomal proteins and translation factors or that mediate tRNA charging and amino acid biosynthesis increased or decreased, respectively. Transcription of the heat shock regulon was induced only by puromycin or streptomycin treatment, which lead to truncation or mistranslation, respectively, but not by other antibiotics that block translation, transcription, or amino acid charging of tRNA. In contrast, relative transcript amounts of certain genes involved in transport, cellular processes, energy metabolism, and purine nucleotide (pur) biosynthesis were changed by different translation inhibitors. In particular, transcript amounts from a pur gene cluster and from purine uptake and salvage genes were significantly elevated by several translation inhibitors, but not by antibiotics that target other cellular processes. Northern blotting confirmed increased transcript amounts from part of the pur gene cluster in cells challenged by translation inhibitors and revealed the presence of a 10-kb transcript. Purine metabolism genes were negatively regulated by a homologue of the PurR regulatory protein, and full derepression in a DeltapurR mutant depended on optimal translation. Unexpectedly, hierarchical clustering of the microarray data distinguished among the global transcription patterns caused by antibiotics that inhibit different steps in the translation cycle. Together, these results show that there is extensive control of transcript amounts by translation in S. pneumoniae, especially for de novo purine nucleotide biosynthesis. In addition, these global transcription patterns form a signature that can be used to classify the mode of action and potential mechanism of new translation inhibitors.

FIG. 1.

Distribution of functions of genes whose relative transcript amounts were affected by treatment of cells with translation inhibitors. The relative abundance is expressed as a percentage of the total number of genes whose relative transcript amount was affected by each inhibitor (Table 1) and includes increased and decreased transcript amounts.

FIG. 2.

Effect of translation inhibitor dosage on global transcription regulation. In two independent experiments, the relative transcript amount of 22 genes changed ≥2-fold in cells treated with 30 ng of erythromycin per ml for 10 min. The corresponding relative fold changes of these genes in cells treated with 120 ng of erythromycin per ml are included for comparison.

FIG. 3.

Northern blot analysis of clpL and purCLFMN-vanZ-purH transcripts from S. pneumoniae R6 cells treated with various antibiotics. Growth of cultures, treatment with compounds, and Northern blotting were performed as described in Materials and Methods. (A) Expression of the 2.1-kb monocistronic clpL transcript hybridized to a probe internal to the clpL ORF. (B) Expression of a ≈10-kb purCLFMN-vanZ-purH operon transcript hybridized to a probe that extends from the end of purC to the beginning of purL. (C) Hybridization signals from both blots were analyzed using ImageQuant software (Molecular Dynamics). The relative fold changes in transcript amounts for each treatment were compared to that of the control (DMSO). DMSO, solvent control; Erm, erythromycin; Cm, chloramphenicol; Tet, tetracycline; Pur, puromycin; Rif, rifampin; Str, streptomycin.

FIG. 4.

Organization of the S. pneumoniae R6 pur gene cluster and relative transcript amounts from its genes in cells subjected to translation inhibition. (A) Organization of the pur gene cluster that encodes the biosynthetic enzymes that convert PRPP to IMP (see Fig. 5). Each arrow represents an ORF. Flanking ORFs comB and strH are included for reference. (B) The relative fold change in transcript amounts of each gene in the pur cluster determined by independent microarray analyses of RNA from cells treated with sublethal concentrations of translation inhibitors (see Materials and Methods and Results; Table 1). Also shown are the relative fold changes in transcript amounts from the spr0264, xpt, and pbuX genes that mediate purine salvage and uptake but that are not genetically linked to the pur gene cluster (see Results and Discussion). Erm, erythromycin; Cm, chloramphenicol; Pur, puromycin; Tet, tetracycline.

FIG. 5.

Purine nucleotide biosynthesis and transport in S. pneumoniae. Proposed pathways for purine transport and the de novo biosynthetic pathway for purine nucleotides are shown. The function of each gene product was assigned based on sequence homology (21). PbuX and Spr0264 are proposed to be transporters. A homologue of gde that mediates the conversion of xanthine to guanine was not identified in the S. pneumoniae R6 genome (21). Enzymes shown in boxes are those whose relative transcript amounts increased significantly in response to puromycin, tetracycline, chloramphenicol, or erythromycin treatment (Fig. 4).

FIG. 6.

Hybridization intensities to the purL oligonucleotides on an Affymetrix chip of RNA isolated from bacteria subjected for 10 min to the indicated treatments. Growth and treatment of bacteria and microarray analyses are described in Materials and Methods. Hybridization intensities are in arbitrary fluorescent units. The hybridization intensities of genes not affected by treatment with translation inhibitors remained essentially constant for the different strains and treatments (data not shown). purR refers to the ΔpurR mutant and R6 refers to its isogenic purR⁺ parent (see Materials and Methods). Additions: no indication, nothing added; +DMSO, DMSO added lacking drug; +Erm30 and +Erm120, addition of 30 or 120 ng of erythromycin per ml.

FIG. 7.

Hierarchical clustering of global transcription patterns determined by microarray analyses. Hierarchical clustering was performed as described in Materials and Methods and the Discussion. The dendrogram shows the relationship between each translation inhibitor and the antibacterial agent triclosan. See Table 4 for the matrix of correlation coefficients used to generate the dendrogram.