Molecular Microbiology

Sitnikov, Dmitry M.; Schineller, Jeffrey B.; Baldwin, Thomas O.

doi: 10.1111/j.1365-2958.1995.mmi_17050801.xpmid: 8596430

The phenomenon of cell‐density‐dependent control of gene expression, called autoinduction, has long been a subject of interest and investigation in bioluminescent marine bacteria. It is now becoming clear that many other bacteria, including animal and plant pathogens, use an autoinduction mechanism to regulate a variety of functions. Cell‐density‐dependent gene expression provides an excellent example of multicellular behaviour in the prokaryotic kingdom where a single cell is able to communicate and sense when a minimal population unit, a ‘quorum’ of bacteria, is achieved in order for certain behaviour of the population to be performed efficiently. Regulation of bacterial bioluminescence has been studied for many years and represents the best model system for understanding the mechanism of cell‐density‐dependent gene expression. This review will focus on transcriptional regulation of the Vibrio fischeri luminescence genes emphasizing the role of the transcriptional activator LuxR and possible autoinduction mechanisms that occur in E. coli Alternative views and opinions regarding the molecular details of the autoinduction mechanism will be discussed.

Mulder, Wietse; Scholten, Inge H.J.M.; Grivell, Leslie A.

doi: 10.1111/j.1365-2958.1995.mmi_17050813.xpmid: 8596431

In Saccharomyces cerevisiae transcription of QCR8, encoding subunit VIII of the mitochondrial ubiquinol cytochrome c oxidoreductase, is subject to glucose repression, whereas in the distantly related yeast Kluyveromyces lactis it is not. The homologous promoter regions lack overall DNA‐sequence identity, but do share binding sites for the transcription factors Abf1p, Cpf1p and Hap2/3/4p. For S. cerevisiae it has been established that these factors are involved in growth and carbon source control of nuclear genes encoding mitochondrial proteins. Here we present experimental evidence that K. lactis counterparts of Abf1p and Cpf1p bind independently to the KIQCR8 promoter. The presence of the KIHap2/3/4p binding site enhances binding of KIAbf1p, strongly suggesting that the KIHap2/3/4p complex stabilizes binding of KIAbf1p. In reciprocal exchanges of the QCR8 genes between S. cerevisiae and K. lactis, overall regulation of transcription was found to be species specific. In contrast to S. cerevisiae, additional elements and factors in K. lactis, distinct from Abf1p, Cpf1p and Hap2/3/4p, are able to activate transcription of the QCR8 gene during both fermentative and non‐fermentative growth, as well as to induce transcription during growth on non‐fermentable carbon sources. In both yeasts, Abf1p is involved in transcriptional activation under both fermentative and non‐fermentative growth conditions. Hap2/3/4p plays a minor activational role during fermentative growth, but is mainly responsible for transcriptional induction under non‐fermentative growth conditions. Under these latter conditions Abf1p and Hap2/3/4p activate transcription synergistically.

Skarstad, Kirsten; Wold, Sture

doi: 10.1111/j.1365-2958.1995.mmi_17050825.xpmid: 8596432

The DnaC protein is required for loading the DnaB helicase at oriC. Thus DnaC promotes the formation of the pre‐replication complex, but must leave the complex in order for the DnaB protein to function as a helicase. In vitro, a slight excess of DnaC inhibits the movement of replication forks by inhibiting DnaB helicase activity (Allen and Kornberg, 1991). Here we show that inhibition of DNA replication by excess DnaC also occurs in vivo. The rate of replication‐fork movement was measured by flow cytometry. Initiation of replication was inhibited with rifampicin and the rate of fork movement monitored during replication run‐out by measuring the increase in the fraction of the cell population with fully replicated chromosomes. The replication rate was inversely related to the amount of excess DnaC protein. Initiation of replication was also inhibited. Co‐overexpression of DnaB protein alleviated the inhibition of replication caused by moderate excess of DnaC. The results show that DnaC interacts with replication forks during elongation in vivo, probably by binding to DnaB and inhibiting its helicase activity. Therefore, the ratio of DnaC to DnaB and the affinity of DnaC for a helicase hexamer at an established replication fork are of great importance for the rate of replication fork movement also in vivo.

Soler, M.; Plovins, A.; Martín, H.; Molina, M.; Nombela, C.

doi: 10.1111/j.1365-2958.1995.mmi_17050833.xpmid: 8596433

MKK1/MKK2 and SLT2 (MPK1) are three Saccharomyces cerevisiae genes, coding for protein kinases, that have been postulated to act sequentially as part of the Pkc1p signalling pathway, a phosphorylation cascade essential for cell integrity. By using the ‘two‐hybrid system’ and co‐purification experiments on glutathione‐agarose beads, we have shown that Slt2p interacts in vivo and in vitro with both Mkk1p and Mkk2p, thus confirming a previous suggestion based on epistasis experiments of the corresponding genes. Plasmid constructs of the SLT2 gene, deleted in the whole C‐terminal non‐kinase region or part of it, and therefore containing all of the conserved kinase subdomains, were still functional in complementation of the slt2 lytic phenotype and in vivo interaction with Mkk1p and Mkk2p. In contrast, the Slt2p C‐terminal domain (162 residues) that carries a glutamine‐rich fragment followed by a 16 polyglutamine tract, was shown to be dispensable for complementation and in vivo association with Mkk1p and Mkk2p. We have also demonstrated that the N‐terminal putative regulatory domain of these two MAP kinase activators is the main region involved in the interaction with Slt2p.

Pedersen, Henrik; Dall, Jens; Dandanell, Gert; Valentin‐Hansen, Poul

doi: 10.1111/j.1365-2958.1995.mmi_17050843.xpmid: 8596434

Repression by CytR depends on the formation of nucleoprotein complexes in which the CytR repressor and the cAMP‐CRP activator complex bind co‐operatively to the DNA. Transcription initiation from CytR‐regulated promoters requires cAMP‐CRP; therefore, the cAMP‐CRP complex functions both as an activator and as a co‐repressor in these promoters. Another interesting aspect of the CytR regulon is that each promoter appears to have individual features. Therefore, structural and functional rules governing the formation of repression and activation complexes in one promoter may not be valid for other promoters of the CytR regulon. Here we show that the Escherichia colinupG gene contains one CytR‐ and four CRP‐binding sites in the control region. Notably, the architecture of the CytR binding site is different from previously described targets. In addition, the CytR repressor triggers a DNA repositioning of a cAMP‐CRP complex in the −35 region upon binding to its operator. Thus, formation of the repression and activation complexes at the nupG promoter involves different subsets of CRP‐binding sites. These findings show that the bacterium uses positive and negative regulatory modules to differentially control the expression of CytR‐ and cAMP‐CRP‐regulated genes.

Marceau, Michael; Beretti, Jean‐Luc; Nassif, Xavier

doi: 10.1111/j.1365-2958.1995.mmi_17050855.xpmid: 8596435

Pili are indispensable in adhesion of encapsulated Neisseria meningitidis (MC) to eukaryotic cells. Intrastrain variability with respect to the degree of adhesion is the result of pilin antigenic variation. We have localized the region responsible for this variability to the 20‐amino‐acid hypervariable domain of pilin. The replacement of an aspartic acid, located in the hypervariable region of a low‐adhesive variant by a lysine restored high adhesiveness. To assess whether hyperadhesiveness confered by some pilin variants was related to the generation of a new pilus‐associated ligand, high‐ and low‐adhesive variants were purified. In a first step, low‐ and high‐adhesive pilins were fused to maltose binding protein (MBP). These hybrid proteins bound epithelial cells with the same affinity. Truncated MBP pilin fusions identified a cell‐binding domain within the 77 residues of the N‐terminal end of mature pilin. This region of the protein is common to low‐ and high‐adhesive derivatives used in this work, thus eliminating the possibility that high adhesiveness confered by some pilin variants was because of the generation of a new pilus‐associated ligand. Electron‐microscopic examination showed that low‐adhesive derivatives expressed long and distinct pili and adhered as single cells. In contrast, pili of derivatives expressing high‐adhesive pilins, either wild type or mutagenized from the low‐adhesive variant, formed large bundles which bound bacteria and caused them to grow as colonies on infected mono‐layers. These data demonstrate that aggregative pili promote high adhesiveness of encapsulated MC.

O'Rourke, Maria; Ison, Catherine A.; Renton, Adrian M.; Spratt, Brian G.

doi: 10.1111/j.1365-2958.1995.mmi_17050865.xpmid: 8596436

A single gonococcus possesses a family of 11 distinct and highly variable opa genes. The extensive variation and rapid evolution of the opa gene repertoire has been exploited to provide a high‐resolution typing method for studies of the short‐term transmission of gonorrhoea. The 11 opa genes are amplified with a single pair of primers by the polymerase chain reaction, digested with frequently‐cutting restriction enzymes, and the fragments are fractionated on polyacrylamide to provide an opa‐type. The method appeared to be highly discriminatory as the opa‐types of gonococci, isolated world‐wide over the last 30 years, were all different. Opa‐typing discriminated between isolates of the same auxotype/serovar class. Similarly, there were 41 opa‐types among 43 consecutive isolates from a sexually transmitted disease (STD) clinic. The two pairs of isolates from this clinic that gave the same opa‐types were identical by other criteria and may have been from unsuspected sexual contacts. With one minor exception, identical opa‐types were obtained from gonococci recovered from known sexual contacts. These results suggest that variation in the family of 11 opa genes evolves so rapidly that the opa‐types of gonococci are distinguishable, unless the isolates are from sexual contacts or a short chain of disease transmission. The identification of gonococci with identical opa‐types is therefore believed to be a good indicator that the individuals from which they were recovered were sexual partners, or part of a short chain of disease transmission.

Langdon, Tim; Sheerins, Alan; Ravagnani, Adriana; Gielkens, Marco; Caddik, Mark X.; Arst, Herbert N.

doi: 10.1111/j.1365-2958.1995.mmi_17050877.xpmid: 8596437

Mutational analysis has enabled identification and localization of an upstream exon of the areA gene of Aspergillus nidulans mediating nitrogen metabolite repression. A mutation in the initiation codon and frameshift mutations, which revert by restoration of the reading frame, established the coding role of the exon and mutations affecting intron splicing in conjunction with DNA sequencing of reverse transcriptase polymerase chain reaction (RT—PCR) products localized the coding region intron. The resulting AREA translation product would have 876 residues. Deletion of the upstream exon such that translation of the remaining areA coding region would yield a protein containing only the 719 C‐terminal residues has only a subtle phenotype, very similar to those resulting from single amino acid replacements in upstream exon‐encoded regions of strong sequence similarity to the Neurospora crassa and Penicillium chrysogenum homologues. A number of areA mRNAs of different sizes are synthesised and appear to be functionally redundant. Synthesis of at least the smallest mRNA(s) is probably subject to autogenous activation. Suppression of frameshift mutations by compensating mutations preventing intron splicing suggests that insertion of a markedly hydrophobic sequence can impair AREA function. Finally, translational initiation for areA can occur within a region of at least 123 nucleotides.

Deretic, V.; Philipp, W.; Dhandayuthapani, S.; Mudd, M.H.; Curcic, R.; Garbe, T.; Heym, B.; Via, L.E.; Cole, S.T.

doi: 10.1111/j.1365-2958.1995.mmi_17050889.xpmid: 8596438

The systems participating in detoxification of reactive oxygen intermediates in Mycobacterium tuberculosis are believed to play a dual role in the biology of this highly adapted human pathogen: (i) they may contribute to the survival of this bacterium in the host; and (ii) alterations in the gene encoding catalase/peroxidase have been linked to this organism's resistance to the front‐line antituberculosis drug isoniazid. These relationships prompted us to extend investigations of the oxidative‐stress‐response systems in M. tuberculosis by analysing the alkyl hydroperoxide reductase gene ahpC and its putative regulator oxyR. Surprisingly, the oxyR gene was found to be inactivated by multiple lesions in M. tuberculosis H37Rv. These alterations were observed in all M. tuberculosis strains tested, and in members of the M. tuberculosis complex: Mycobacterium bovis BCG, Mycobacterium africanum, and Mycobacterium microti. The corresponding region carrying these genes in Mycobacterium leprae, an organism not sensitive to isoniazid, has a complete oxyR gene divergently transcribed from ahpC. An increase in minimal inhibitory concentration for isoniazid was observed upon transformation of M. tuberculosis H37Rv with cosmids carrying the oxyR—ahpC region of M. leprae. In keeping with the observed inactivation of oxyR, transcriptional activity of the corresponding region in M. tuberculosis was an order of magnitude lower than that of the oxyR gene from M. leprae. While the loss of this putative regulator of oxidative‐stress response in M. tuberculosis is paradoxical considering the fact that survival in host macrophages is regarded as a critical feature of this pathogen, it offers a partial explanation for the exquisite sensitivity of M. tuberculosis to isoniazid.

Dhandayuthapani, S.; Via, L.E.; Thomas, C.A.; Horowitz, P.M.; Deretic, D.; Deretic, V.

doi: 10.1111/j.1365-2958.1995.mmi_17050901.xpmid: 8596439

The green fluorescent protein (GFP) of the jellyfish Aequorea victoria offers certain advantages over other bioluminescence systems because no exogenously added substrate or co‐factors are necessary, and fluorescence can be elicited by irradiation with blue light without exposing the cells producing GFP to invasive treatments. A mycobacterial shuttle‐plasmid vector carrying gfp cDNA was constructed and used to generate transcriptional fusions with promoters of interest and to examine their expression in Mycobacterium smegmatis and Mycobacterium bovis BCG grown in macrophages or on laboratory media. The promoters studied were: (i) ahpC from Mycoosis and Mycobacterium leprae, a gene encoding alkyl hydroperoxide reductase which, along with the divergently transcribed regulator oxyR, are homologues of corresponding stress‐response systems in enteric bacteria and play a role in isoniazid sensitivity; (ii) mtrA, an M. tuberculosis response regulator belonging to the superfamily of bacterial two‐component signal‐transduction systems; (iii) hsp60, a previously characterized heat‐shock gene from M. bovis; and (iv) tbprc3, a newly isolated promoter from M. tuberculosis. Expression of these promoters in mycobacteria was analysed using epifluorescence microscopy, laser scanning confocal microscopy, fluorescence spectroscopy, and flow cytometry. These approaches permitted assessment of fluorescence prior to and after macrophage infection, and analyses of promoter expression in individual mycobacteria and its distribution within populations of bacterial cells. Bacteria expressing GFP from a strong promoter could be separated by fluorescence‐activated cell sorting from cells harbouring the vector used to construct the fusion. In addition, the stable expression of mtrA‐gfp fusion in M. bovis BCG facilitated localization and isolation of phagocytic vesicles containing mycobacteria. The experiments presented here suggest that GFP will be a useful tool for analysis of mycobacterial gene expression and a convenient cell biology marker to study mycobacterial interactions with macrophages.