Mosaic genes and mosaic chromosomes: intra- and interspecies genomic variation of Streptococcus pneumoniae.

Streptococcus pneumoniae remains a major causative agent of serious human diseases. The worldwide increase of antibiotic resistant strains revealed the importance of horizontal gene transfer in this pathogen, a scenario that results in the modulation of the species-specific gene pool. We investigated genomic variation in 20 S. pneumoniae isolates representing major antibiotic-resistant clones and 10 different capsular serotypes. Variation was scored as decreased hybridization signals visualized on a high-density oligonucleotide array representing 1,968 genes of the type 4 reference strain KNR.7/87. Up to 10% of the genes appeared altered between individual isolates and the reference strain; variability within clones was below 2.1%. Ten gene clusters covering 160 kb account for half of the variable genes. Most of them are associated with transposases and are assumed to be part of a flexible gene pool within the bacterial population; other variable loci include mosaic genes encoding antibiotic resistance determinants and gene clusters related to bacteriocin production. Genomic comparison between S. pneumoniae and commensal Streptococcus mitis and Streptococcus oralis strains indicates distinct antigenic profiles and suggests a smooth transition between these species, supporting the validity of the microarray system as an epidemiological and diagnostic tool.

Microarray-based identification of a novel Streptococcus pneumoniae regulon controlled by an autoinduced peptide.

We have identified in the Streptococcus pneumoniae genome sequence a two-component system (TCS13, Blp [bacteriocin-like peptide]) which is closely related to quorum-sensing systems regulating cell density-dependent phenotypes such as the development of genetic competence or the production of antimicrobial peptides in lactic acid bacteria. In this study we present evidence that TCS13 is a peptide-sensing system that controls a regulon including genes encoding Blps. Downstream of the Blp TCS (BlpH R) we identified open reading frames (blpAB) that have the potential to encode an ABC transporter that is homologous to the ComA/B export system for the competence-stimulating peptide ComC. The putative translation product of blpC, a small gene located downstream of blpAB, has a leader peptide with a Gly-Gly motif. This leader peptide is typical of precursors processed by this family of transporters. Microarray-based expression profiling showed that a synthetic oligopeptide corresponding to the processed form of BlpC (BlpC*) induces a distinct set of 16 genes. The changes in the expression profile elicited by synthetic BlpC* depend on BlpH since insertional inactivation of its corresponding gene abolishes differential gene induction. Comparison of the promoter regions of the blp genes disclosed a conserved sequence element formed by two imperfect direct repeats upstream of extended -10 promoter elements. We propose that BlpH is the sensor for BlpC* and the conserved sequence element is a recognition sequence for the BlpR response regulator.

Visualising gene expression in its metabolic context.

Relative changes in mRNA as well as protein levels induced by sublethal doses of antibiotics on bacteria are measured and results visualised in the context of metabolic pathway diagrams. The mRNA levels present at a given time point after the addition of the antibiotic are measured using microarrays from Affymetrix. Additionally, the relative amount of each protein synthesised during 3 minute intervals sampled at the given times is measured using radio-labelling followed by two-dimensional polyacrylamide gel electrophoresis and the subsequent analysis of the images produced by exposure to a phosphorimager. Metabolic pathway diagrams are both constructed in-house and imported from KEGG (Kyoto Encyclopedia of Genes and Genomes). Both protein and mRNA expression data can be displayed in the pathway diagrams such that the colour of the vectors or enzyme identifiers indicate the relative change in expression level and reproducibility.

Domain organization and molecular characterization of 13 two-component systems identified by genome sequencing of Streptococcus pneumoniae.

In bacteria, adaptive responses to environmental stimuli are often initiated by two-component signal transduction systems (TCS). The prototypical TCS comprises two proteins: a histidine kinase (HK, hk) and a response regulator (RR rr). Recent research has suggested that compounds that inhibit two-component systems might have good antibacterial activity. In order to identify TCS that are crucial for growth or virulence of Streptococcus pneumoniae, we have examined the genomic sequence of a virulent S. pneumoniae strain for genes that are related to known histidine kinases or response regulators. Altogether 13 histidine kinases and 13 response regulators have been identified. The protein sequences encoded by these genes were compared with sequences deposited in public databases. This analysis revealed that two of the 13 pneumococcal TCSs have been described before (ciaRH and comDE) and two are homologous to the yycFG and the phoRP genes of Bacillus subtilis. All the pneumococcal response regulators contain putative DNA binding motifs within the C-terminal output domain, implying that they are involved in transcriptional control. Two of these response regulators are obviously the first representatives of a new subfamily containing an AraC-type DNA-binding effector domain. To assess the regulatory role of these transcription factors, we disrupted each of the 13 response regulator genes by insertional mutagenesis. All the viable mutant strains with disrupted response regulator genes were further characterized with regard to growth in vitro, competence, and experimental virulence. Two response regulator genes could not be inactivated, indicating that they may regulate essential cellular functions. The possibility of using these systems as targets for the development of novel antibacterials will be discussed.

Competence-specific induction of recA is required for full recombination proficiency during transformation in Streptococcus pneumoniae.

Transcriptional activation of the recA gene of Streptococcus pneumoniae was previously shown to occur at competence. A 5.7 kb recA-specific transcript that contained at least two additional genes, cinA and dinF, was identified. We now report the complete characterization of the recA operon and investigation of the role of the competence-specific induction of recA. The 5.7 kb competence-specific recA transcript is shown to include lytA, which encodes the pneumococcal autolysin, a protein previously shown to contribute to virulence of S. pneumoniae. Uncoupling (denoted Ind-) of recA and/or the downstream genes was achieved through the placement of transcription terminators within the operon, either upstream or downstream of recA. Prevention of the competence-specific induction of recA severely affected spontaneous transformation. Transformation efficiencies of recA+ (Ind-) and of wild-type cells were compared under various conditions and with different donor DNA. Chromosomal transformation was reduced 17-(chromosomal donor) to 45-fold (recombinant plasmid donor), depending on the donor DNA, and plasmid establishment was reduced 129-fold. Measurement of uptake of radioactively labelled donor DNA in transformed cells in parallel with scoring for transformants (chromosomal donor) revealed normal uptake, but a 21-fold reduction in recombination in a recA+ (Ind-) strain, indicating that the transformation defect was primarily in recombination. Strikingly enough, a much larger (460-fold) reduction in recombination was observed for the shortest homologous donor fragment used (878 nucleotides long). Possible interpretations of the observation that basal RecA appears unable to promote efficient recombination whatever the number and the length of donor fragments taken up are proposed. The role of recA induction is discussed in view of the potential contribution of transformation to genome plasticity in this pathogen.

Bacterial transcript imaging by hybridization of total RNA to oligonucleotide arrays.

We have used high-density oligonucleotide probe arrays (chips) for bacterial transcript imaging. We designed a chip containing probes representing 106 Hemophilus influenzae genes and 100 Streptococcus pneumoniae genes. The apparent lack of polyadenylated transcripts excludes enrichment of mRNA by affinity purification and we thus used total, chemically biotinylated RNA as hybridization probe. We show that hybridization of Streptococcus RNA to a chip allows simultaneous quantification of the transcript levels. The sensitivity was found to be in the range of one to five transcripts per cell. The quantitative chip results were in good agreement with conventional Northern blot analysis of selected genes. This technology allows simultaneous and quantitative measurement of the transcriptional activity of entire bacterial genomes on a single oligonucleotide probe array.

Translation of trpG in Bacillus subtilis is regulated by the trp RNA-binding attenuation protein (TRAP).

trpG of Bacillus subtilis encodes a glutamine amidotransferase subunit that is involved in the synthesis of both folic acid and L-tryptophan. Expression of trpG is negatively regulated by tryptophan even though this gene is located within a folic acid biosynthetic operon. Examination of both transcriptional and translational gene fusions to lacZ involving trpG and direct measurements of trpG mRNA levels and TrpG polypeptide accumulation demonstrated that translation of trpG is regulated by tryptophan whereas transcription is not. these studies also show that this regulation is mediated by the trp RNA-binding attenuation protein. Deletion and point mutations indicated that regulation is dependent on a series of G/UAG trinucleotide repeats surrounding the putative ribosome-binding site for trpG. Our results are consistent with a model in which the tryptophan-activated trp RNA-binding attenuation protein and ribosomes compete for binding to trpG mRNA.

Enzymic characterization of Bacillus subtilis GTP cyclohydrolase I. Evidence for a chemical dephosphorylation of dihydroneopterin triphosphate.

GTP cyclohydrolase I catalyses the first committing step in the biosynthesis of the pterin moiety of folic acid: conversion of GTP to dihydroneopterin triphosphate. GTP cyclohydrolase I of Bacillus subtilis was purified to homogeneity and shown to have a homo-octameric structure. The enzyme had an apparent Km for GTP of 4 microM and, in the absence of cations, a Vmax. of 80 nmol/min per mg of protein. K+ ions moderately increased its Vmax., whereas UTP and Ca2+ and Mg2+ ions drastically increased its Km for GTP. Dihydrofolate and other products of the folate and tetrahydrobiopterin pathways did not inhibit GTP cyclohydrolase I. In addition to their effect on the enzyme activity, Ca2+ and Mg2+ ions catalysed the chemical dephosphorylation of dihydroneopterin triphosphate to non-cyclic dihydroneopterin monophosphate, the substrate for the phosphomonoesterase reaction in folate biosynthesis. This dephosphorylation was specific and did not require the action of a phosphatase. We suggest a physiological role for Ca2+ ions and UTP in regulation of folate biosynthesis at the levels of GTP cyclohydrolase I and dephosphorylation of dihydroneopterin triphosphate.

Construction and characterisation of a yeast artificial chromosome library containing three haploid maize genome equivalents.

We have constructed a yeast artificial chromosome (YAC) library using high-molecular-weight DNA prepared from agarose-embedded leaf protoplasts of the maize inbred line UE95. This library contains 79,000 clones with an average insert size of 145 kb and should therefore represent approximately three haploid genome equivalents. The library is organised as an ordered array in duplicate microtitre plates. Forty-one pools of DNA from 1920 individual clones have been prepared for rapid screening of the library by the polymerase chain reaction (PCR). Using this approach, together with conventional colony hybridisation, we have been able to identify between one and eight positive clones for every probe used.