- Volume 152, Issue 1, 2006

Production of the secondary metabolite phenazine-1-carboxamide (PCN) by Pseudomonas chlororaphis PCL1391 is crucial for biocontrol activity against the phytopathogen Fusarium oxysporum f. sp. radicis lycopersici on tomato. Regulation of PCN production involves the two-component signalling system GacS/GacA, the quorum-sensing system PhzI/PhzR and the regulator PsrA. This paper reports that a functional rpoS is required for optimal PCN and N-hexanoyl-l-homoserine lactone (C6-HSL) production. Constitutive expression of rpoS is able to complement partially the defect of a psrA mutant for PCN and N-acylhomoserine lactone production. Western blotting shows that rpoS is regulated by gacS. Altogether, these results suggest the existence of a cascade consisting of gacS/gacA upstream of psrA and rpoS, which influence expression of phzI/phzR. Overproduction of phzR complements the effects on PCN and C6-HSL production of all mutations tested in the regulatory cascade, which shows that a functional quorum-sensing system is essential and sufficient for PCN synthesis. In addition, the relative amounts of PCN, phenazine-1-carboxylic acid and C6-HSL produced by rpoS and psrA mutants harbouring a constitutively expressed phzR indicate an even more complex network of interactions, probably involving other genes. Preliminary microarray analyses of the transcriptomics of the rpoS and psrA mutants support the model of regulation described in this study and allow identification of new genes that might be involved in secondary metabolism.

Lactobacillus casei transports glucose preferentially by a mannose-class phosphoenolpyruvate : sugar phosphotransferase system (PTS). The genomic analysis of L. casei allowed the authors to find a gene cluster (manLMNO) encoding the IIAB (manL), IIC (manM) and IID (manN) proteins of a mannose-class PTS, and a putative 121 aa protein of unknown function (encoded by manO), homologues of which are also present in man clusters that encode glucose/mannose transporters in other Gram-positive bacteria. The L. casei man operon is constitutively expressed into a manLMNO messenger, but an additional manO transcript was also detected. Upstream of the man operon, two genes (upsR and upsA) were found which encode proteins resembling a transcriptional regulator and a membrane protein, respectively. Disruption of either upsR or upsA did not affect manLMNO transcription, and had no effect on glucose uptake. Cells carrying a manO deletion transported glucose at a rate similar to that of the wild-type strain. By contrast, a manM disruption resulted in cells unable to transport glucose by the PTS, thus confirming the functional role of the man genes. In addition, the manM mutant exhibited neither inducer exclusion of maltose nor glucose repression. This result confirms the need for glucose transport through the PTS to trigger these regulatory processes in L. casei.

The use of a lysine-overproducing strain of Lactobacillus plantarum in food or feed fermentations may lead to the production of lysine-rich products. The availability of functional genes and information on the regulation of lysine biosynthesis are required to develop a lysine-overproducing strain. The genome sequence of L. plantarum revealed putative lysine biosynthetic genes, some of which may produce isozymes. This study examined the functionality of the genes and the regulation of the first four enzymes of lysine biosynthesis, together with homoserine dehydrogenase, in L. plantarum. The genes were expressed in Escherichia coli, and the regulation of the enzymes was studied in cell extracts of both recombinant E. coli and L. plantarum. Among seven lysine biosynthetic genes studied (aspartokinase genes, thrA1 and thrA2; aspartate semialdehyde dehydrogenase genes, asd1 and asd2; dihydrodipicolinate synthase genes, dapA1 and dapA2; and the dihydrodipicolinate reductase gene, dapB) plus two homoserine dehydrogenase genes (hom1 and hom2), the products of six genes, i.e. thrA2, asd2, dapA1, dapB, hom1 and hom2, showed obvious enzyme activities in vitro. The product of one of the homoserine dehydrogenase genes, hom1, exhibited both homoserine dehydrogenase and aspartokinase activities. However, the aspartokinase activity was mainly due to ThrA2 and was inhibited by l-lysine and repressed by l-threonine, and the homoserine dehydrogenase activity was mainly due to Hom2 and was inhibited by l-threonine. The aspartate semialdehyde dehydrogenase, dihydrodipicolinate synthase and dihydrodipicolinate reductase were not regulated by the end-products of the pathway.

Lipooligosaccharide (LOS) is a major virulence factor of the pathogenic Neisseria. Three galactosyltransferase genes, lgtB, lgtE and lgtH, responsible for the biosynthesis of LOS oligosaccharide chains, were analysed in five Neisseria species. The function of lgtH in Neisseria meningitidis 6275 was determined by mutagenesis and chemical characterization of the parent and mutant LOS chains. The chemical characterization included SDS-PAGE, immunoblot, hexose and mass spectrometry analyses. Compared with the parent LOS, the mutant LOS lacked galactose, and its oligosaccharide decreased by three or four sugar units in matrix-assisted laser desorption ionization (MALDI)-MS analysis. The results show that lgtH encodes a β-1,4-galactosyltransferase, and that the glucose moiety linked to heptose (Hep) in the α chain is the acceptor site in the biosynthesis of Neisseria LOS. To understand the sequence diversity and relationships of lgtB, lgtE and lgtH, the entire lgt-1 locus was further sequenced in three N. meningitidis strains and three commensal Neisseria strains, and compared with the previously reported lgt genes from Neisseria species. Comparison of the protein sequences of the three enzymes LgtB, LgtE and LgtH showed a conserved N-terminal region, and a highly variable C-terminal region, suggesting functional constraint for substrate and acceptor specificity, respectively. The analyses of allelic variation and evolution of 23 lgtB, 12 lgtE and 14 lgtH sequences revealed a distinct evolutionary history of these genes in Neisseria. For example, the splits graph of lgtE displayed a network evolution, indicating frequent DNA recombination, whereas splits graphs of lgtB and lgtH displayed star-tree-like evolution, indicating the accumulation of point mutations. The data presented here represent examples of the evolution and variation of prokaryotic glycosyltransferase gene families. These imply the existence of multiple enzyme isoforms for biosynthesis of a great diversity of oligosaccharides in nature.

The complete nucleotide sequence of the 3475 bp plasmid pCD3.4 from Carnobacterium divergens LV13, which encodes the bacteriocin divergicin A, was determined. Nucleotide sequence, deletion and complementation analyses revealed the presence of a trans-acting replication protein, RepA, and DNA sequences involved in plasmid replication and copy-number control. The DNA region preceding the repA gene probably contains the origin of replication. This sequence includes four and a half direct repeats (iterons) of 22 bp, to which RepA is thought to bind, and an AT-rich region containing a 12 bp repeat, at which initiation of DNA might occur. Further upstream of this sequence resides a fifth iteron required for optimal plasmid replication. The RepA protein shows homology to replication proteins of the pUCL287 subfamily of theta-type replicons. Two ORFs were found downstream of the repA gene that could be deleted without affecting replication and stability of the plasmid. pCD3.4 has a narrow host range, and could only be maintained in Carnobacterium spp.; however, a mutant of the plasmid was obtained that enabled the pCD3.4 replicon to replicate in Enterococcus faecium, but not in Carnobacterium spp. The mutation was located in the C-terminal region of the RepA protein, changing a proline into a serine. This is believed to be the first example of such plasmid-host-range modulation in Gram-positive bacteria.

Bacterial butyryl-CoA CoA-transferase activity plays a key role in butyrate formation in the human colon, but the enzyme and corresponding gene responsible for this activity have not previously been identified. A novel CoA-transferase gene is described from the colonic bacterium Roseburia sp. A2-183, with similarity to acetyl-CoA hydrolase as well as 4-hydroxybutyrate CoA-transferase sequences. The gene product, overexpressed in an Escherichia coli lysate, showed activity with butyryl-CoA and to a lesser degree propionyl-CoA in the presence of acetate. Butyrate, propionate, isobutyrate and valerate competed with acetate as the co-substrate. Despite the sequence similarity to 4-hydroxybutyrate CoA-transferases, 4-hydroxybutyrate did not compete with acetate as the co-substrate. Thus the CoA-transferase preferentially uses butyryl-CoA as substrate. Similar genes were identified in other butyrate-producing human gut bacteria from clostridial clusters IV and XIVa, while other candidate CoA-transferases for butyrate formation could not be detected in Roseburia sp. A2-183. This suggests strongly that the newly identified group of CoA-transferases described here plays a key role in butyrate formation in the human colon.

Tripartite ATP-independent periplasmic (TRAP) transporters are relatively common prokaryotic secondary transporters which comprise an extracytoplasmic solute receptor (ESR) protein and two dissimilar membrane proteins or domains, yet the substrates and physiological functions of only a few of these systems are so far known. In this study, a biophysical approach was used to identify the ligands for the purified Rhodobacter capsulatus RRC01191 and Escherichia coli YiaO proteins, which are members of two phylogenetically distinct families of TRAP-ESRs found in diverse bacteria. In contrast to previous indirect evidence pointing to RRC01191 orthologues being involved in polyol uptake, it was shown that RRC01191 binds pyruvate, 2-oxobutyrate and a broad range of aliphatic monocarboxylic 2-oxoacid anions with varying affinities (K d values 0·08–3 μM), consistent with a predicted role in monocarboxylate transport related to branched-chain amino-acid biosynthesis. The E. coli YiaMNO TRAP transporter has previously been proposed to be an l-xylulose uptake system [ Plantinga et al. (2004) Mol Membr Biol 21, 51–57], but purified YiaO did not bind l- or d-xylulose as judged by fluorescence spectroscopy, circular dichroism or mass spectrometry. Instead, these techniques showed that a breakdown product of l-ascorbate, 2,3-diketo-l-gulonate (2,3-DKG), binds by a simple one-step mechanism with sub-micromolar affinity. The data provide the first evidence for the existence of ESR-dependent transporters for 2-oxoacids and 2,3-DKG, homologues of which appear to be widespread amongst prokaryotes. The results also underline the utility of direct ESR ligand-binding studies for TRAP transporter characterization

Copper uptake in the fission yeast Schizosaccharomyces pombe is carried out by a heteromeric complex formed by two proteins, Ctr4 and Ctr5. In this study, a stable expression system using integrative plasmids was developed to investigate the respective roles of Ctr4 and Ctr5 in copper transport. It was shown that expression of full-length Ctr4 or truncated Ctr4 containing residues 106–289 was required for localization of Ctr5 to the plasma membrane. Likewise, when the full-length Ctr5 or truncated Ctr5 from residues 44–173 was co-expressed with Ctr4, this protein was visualized at the periphery of the cell. To determine the importance of the Mets motifs (consisting of five methionines arranged as Met-X2-Met-X-Met, where X is any amino acid) of Ctr4 and Ctr5 in the heteroprotein complex, we co-expressed Ctr5 lacking the Mets motif and Cys-X-Met-X-Met sequence with wild-type Ctr4 or its mutant derivatives. Conversely, Ctr4 lacking the Mets motif and Met122 was expressed with wild-type Ctr5 or its mutant derivatives. These experiments revealed that the five Mets motifs of Ctr4 and the Ctr4 residue Met122 have equally important roles in copper assimilation. Furthermore, the two partially overlapping Mets motifs and the Cys-X-Met-X-Met sequence in Ctr5 have redundant functions in copper transport, with the latter sequence making a greater contribution than the former. Together, the data reveal that co-expression of both Ctr4 and Ctr5 is necessary for the proper function and localization of the heteroprotein complex to the plasma membrane. Once on the cell surface, the N-terminal regions of Ctr4 and Ctr5 can function independently to transport copper; however, the greatest efficiency is achieved when both N termini are present.

The cpmA gene mediates an output signal in the cyanobacterial circadian system. This gene and its homologues are evolutionarily old, and occur in some non-photosynthetic bacteria and archaea as well as in cyanobacteria. The gene has two functional domains that differ drastically in their level of polymorphism: the N-terminal domain is much more variable than the PurE homologous C-terminal domain. The phylogenetic tree of the cpmA homologues features four main clades (C1–C4), two of which (C1 and C3) belong to cyanobacteria. These cyanobacterial clades match respective ones in the previously reported phylogenetic trees of the other genes involved in the circadian system. The phylogenetic analysis suggested that the C3 subfamily, which comprises the genes from the cyanobacteria with the kaiBC-based circadian system, experienced a lateral transfer, probably from evolutionarily old proteobacteria about 1000 million years ago. The genes of this subfamily have a significantly higher nonsynonymous substitution rate than those of C1 (2·13×10−10 and 1·53×10−10 substitutions per nonsynonymous site per year, respectively). It appears that the functional and selective constraints of the kaiABC-based system have slowed down the rate of sequence evolution compared to the cpmA homologues of the kaiBC-based system. On the other hand, the differences in the mutation rates between the two cyanobacterial clades point to the different functional constraints of the systems with or without kaiA.

Lactobacillus plantarum is a species of considerable industrial and medical interest. To date, the lack of reliable molecular methods for definite identification at strain level has hindered studies of the population biology of this organism. Here, a multilocus sequence typing (MLST) system for this organism is described, which exploits the genetic variation present in six housekeeping loci to determine the genetic relationship among isolates. The MLST system was established using 16 L. plantarum strains that were also characterized by ribotyping and restriction fragment length polymorphism (RFLP) analysis of the PCR-amplified 16S–23S rDNA intergenic spacer region (ISR). Ribotyping grouped the strains into four groups; however, RFLP analysis of the ISRs showed no differences in the strains analysed. In contrast, MLST had a good discriminatory ability. The sequence analysis of the six genes showed 14 different allelic combinations, with 12 of them represented by only one strain. By using this MLST approach we were able to confirm the identity of two strains deposited in the Spanish Type Culture Collection as different strains. Phylogenetic analysis indicated a panmictic population structure of L. plantarum and split decomposition analysis indicated that recombination plays a role in creating genetic heterogeneity in L. plantarum. As MLST allows precise identification, and easy comparison and exchange of results obtained in different laboratories, the future application of this new molecular method could be useful for the identification of valuable L. plantarum strains.

Campylobacter coli is a food-borne pathogen associated increasingly with human gastroenteritis. C. coli has a high prevalence in swine, but is isolated also from cattle and poultry. Multilocus sequence typing (MLST) systems have been developed to differentiate C. coli strains. Although substantial allelic diversity was identified across all seven C. coli MLST loci, no correlations were made in two previous studies between allele or sequence type (ST) and the source of the organism. However, this may be due to either the relatively small number or the low diversity of C. coli strains used to validate both MLST studies. This study describes the typing of 488 C. coli strains from 4 different food animal sources (cattle, chickens, swine and turkeys), collected at different times over a 6 year period from different USA geographical locations. A total of 149 STs were identified. The 185 swine strains were the most diverse, possessing 82 STs. The cattle strains were the most clonal; 52/63 (83 %) strains possessed a single ST (ST-1068). A subpopulation of C. coli strains, collected primarily from turkeys, was identified, containing both C. coli- and Campylobacter jejuni-associated MLST alleles, specifically the C. jejuni allele aspA103. The majority of STs and alleles were host associated, i.e. found primarily in strains from a single food-animal source. Only 12/149 (8 %) STs were found in multiple sources. Additionally, the majority (34/46, 74 %) of major (n>5) alleles were more prevalent in certain hosts (swine, poultry). The presence of host-associated C. coli MLST alleles could lead potentially to more efficient source tracking in this species, especially in the trace-back of both sporadic and outbreak human clinical C. coli strains to animal sources.

Aflatoxins are polyketide-derived secondary metabolites produced by Aspergillus parasiticus, Aspergillus flavus, Aspergillus nomius and a few other species. The toxic effects of aflatoxins have adverse consequences for human health and agricultural economics. The aflR gene, a regulatory gene for aflatoxin biosynthesis, encodes a protein containing a zinc-finger DNA-binding motif. Although Aspergillus oryzae and Aspergillus sojae, which are used in fermented foods and in ingredient manufacture, have no record of producing aflatoxin, they have been shown to possess an aflR gene. This study examined 34 strains of Aspergillus section Flavi. The aflR gene of 23 of these strains was successfully amplified and sequenced. No aflR PCR products were found in five A. sojae strains or six strains of A. oryzae. These PCR results suggested that the aflR gene is absent or significantly different in some A. sojae and A. oryzae strains. The sequenced aflR genes from the 23 positive strains had greater than 96·6 % similarity, which was particularly conserved in the zinc-finger DNA-binding domain. The aflR gene of A. sojae has two obvious characteristics: an extra CTCATG sequence fragment and a C to T transition that causes premature termination of AFLR protein synthesis. Differences between A. parasiticus/A. sojae and A. flavus/A. oryzae aflR genes were also identified. Some strains of A. flavus as well as A. flavus var. viridis, A. oryzae var. viridis and A. oryzae var. effuses have an A. oryzae-type aflR gene. For all strains with the A. oryzae-type aflR gene, there was no evidence of aflatoxin production. It is suggested that for safety reasons, the aflR gene could be examined to assess possible aflatoxin production by Aspergillus section Flavi strains.

The value of the multivariate data analysis tools principal component analysis (PCA) and principal component discriminant analysis (PCDA) for prioritizing leads generated by microarrays was evaluated. To this end, Pseudomonas putida S12 was grown in independent triplicate fermentations on four different carbon sources, i.e. fructose, glucose, gluconate and succinate. RNA isolated from these samples was analysed in duplicate on an anonymous clone-based array to avoid bias during data analysis. The relevant transcripts were identified by analysing the loadings of the principal components (PC) and discriminants (D) in PCA and PCDA, respectively. Even more specifically, the relevant transcripts for a specific phenotype could also be ranked from the loadings under an angle (biplot) obtained after PCDA analysis. The leads identified in this way were compared with those identified using the commonly applied fold-difference and hierarchical clustering approaches. The different data analysis methods gave different results. The methods used were complementary and together resulted in a comprehensive picture of the processes important for the different carbon sources studied. For the more subtle, regulatory processes in a cell, the PCDA approach seemed to be the most effective. Except for glucose and gluconate dehydrogenase, all genes involved in the degradation of glucose, gluconate and fructose were identified. Moreover, the transcriptomics approach resulted in potential new insights into the physiology of the degradation of these carbon sources. Indications of iron limitation were observed with cells grown on glucose, gluconate or succinate but not with fructose-grown cells. Moreover, several cytochrome- or quinone-associated genes seemed to be specifically up- or downregulated, indicating that the composition of the electron-transport chain in P. putida S12 might change significantly in fructose-grown cells compared to glucose-, gluconate- or succinate-grown cells.

Lactic acid bacteria (LAB) are frequently encountered inhabitants of the human intestinal tract. A protective layer of mucus covers the epithelial cells of the intestine, offering an attachment site for these bacteria. In this study bioinformatics tools were used to identify and characterize proteins containing one type of mucus-binding domain, called MUB, that is postulated to play an important role in the adherence of LAB to this mucus layer. By searching in all protein databases 48 proteins containing at least one of these MUB domains in nine LAB species were identified. These MUB domains varied in size, ranging from approximately 100 to more than 200 residues per domain. Complete MUB domains were found exclusively in LAB. The number of MUB domains present in a single protein varied from 1 to 15. In some cases, orthologous proteins in closely related species contained a different number of domains, indicating that repeats of the domain undergo rapid duplication and deletion. Proteins containing the MUB domain were often encoded by gene clusters that encode multiple extracellular proteins. In addition to one or more copies of the MUB domain, many of these proteins contained other domains that are predicted to be involved in binding to and degradation of extracellular components. These findings strongly suggest that the MUB domain is an LAB-specific functional unit that performs its task in various domain contexts and could fulfil an important role in host–microbe interactions in the gastrointestinal tract.