- Volume 145, Issue 9, 1999

The authors have previously demonstrated that adjuvant-mediated differences in early cellular responses to antigens significantly affect subsequent adaptive immune responses. To investigate further the contribution of adjuvant to adaptive immune responses, outer-membrane proteins (OMP) purified from the respiratory pathogen Actinobacillus pleuropneumoniae, given either alone (antigen group) or complexed with SAMA4 (vaccine group), were injected intradermally into groups of mice. Controls were given PBS. Inclusion of adjuvant did not significantly alter the kinetics of antibody responses against OMP in serum or respiratory tract washings (RTW) over 21 weeks. Re-exposure to OMP at 21 weeks also induced identical recall responses in both immunized groups. However, differences between the responses of the vaccine and antigen groups were apparent when sera and RTW were reacted against OMP and OMP-derived polysaccharide antigens (ODPA). Serum and RTW reactivity against protein antigens was stronger in the vaccine group than in the antigen group. Serum and RTW from the vaccine group also reacted against a greater number of proteins than did the antigen group. Although serum reactivity against ODPA was equivalent for both groups, RTW from the vaccine group reacted only faintly against ODPA compared with the antigen group. The results suggested that shifting of antibody reactivity away from polysaccharide antigens toward protein antigens was an adjuvant-mediated effect. The rapid death of controls following intranasal inoculation confirmed that protection was ultimately dependent on the presence of specific antibodies in the serum and respiratory tract. However, since both groups responded equally to intranasal infection with A. pleuropneumoniae, as seen by the rapid clearance of bacteria from the lungs, the biological significance of any differences between the groups was unclear. Knowledge of the effects of adjuvants may provide a rational basis for adjuvant selection and the ability to manipulate immunological outcomes more precisely.

The 16S toxin and one subcomponent of haemagglutinin (HA), designated HA1, were purified from a type D culture of Clostridium botulinum by a newly established procedure, and their HA activities as well as that of purified type C 16S toxin were characterized. SDS-PAGE analysis indicated that the free HA1 forms a polymer with a molecular mass of approximately 200 kDa. Type C and D 16S toxins agglutinated human erythrocytes in the same manner. Their HA titres were dramatically reduced by employing erythrocytes that had been previously treated with neuraminidase, papain or proteinase K, and were inhibited by the addition of N-acetylneuraminic acid to the reaction mixtures. In a direct-binding test to glycolipids such as SPG (NeuAcα2-3Galβ1-4GlcNAcβ1-3Galβ1-4Glcβ1-Cer) and GM3 (NeuAcα2-3Galβ1-4Glcβ1-Cer), and glycoproteins such as glycophorin A and/or B prepared from the erythrocytes, both toxins bound to sialylglycolipids and sialoglycoproteins, but bound to neither neutral glycolipids nor asialoglycoproteins. On the basis of these results, it was concluded that type C and D 16S toxins bind to erythrocytes through N-acetylneuraminic acid. HA1 showed no haemagglutination activity, although it did bind to sialylglycolipids. We therefore speculate that binding to glycoproteins rather than to glycolipids may be important in causing haemagglutination by type C and D 16S toxins.

Glycine betaine is mostly known as an osmoprotectant. It is involved in the osmotic adaptation of eukaryotic and bacterial cells, and accumulates up to 1 M inside cells subjected to an osmotic upshock. Since, like other osmolytes, it can act as a protein stabilizer, its thermoprotectant properties were investigated. In vitro, like protein chaperones such as DnaK, glycine betaine and choline protect citrate synthase against thermodenaturation, and stimulate its renaturation after urea denaturation. In vivo, the internal concentration of glycine betaine is neither increased nor decreased after heat shock (this contrasts with a massive increase after osmotic upshock). However, even in exponential-phase bacteria grown in usual minimal salts media, the internal glycine betaine concentration attains levels (around 50 mM) which can protect proteins against thermodenaturation in vitro. Furthermore, glycine betaine and choline restore the viability of a dnaK deletion mutant at 42 °C, suggesting that glycine betaine not only acts as a thermoprotectant in vitro, but also acts as a thermoprotectant for Escherichia coli cells in vivo.

The wzy/rfc gene, encoding the O-antigen polymerase, of Salmonella enterica serovar Typhimurium has been previously cloned and sequenced. In the present work, the wzy transcriptional startpoint was initially identified by primer extension. Next, wzy promoter strength in Escherichia coli K-12 was measured, and was found to be greater than that of the induced lac promoter. To define the Wzy translational startpoint, DNA including the wzy promoter and the putative first five residues of the Wzy protein was fused to the N-terminus of glutathione-S-transferase, and the fusion protein purified by affinity chromatography. N-terminal amino acid sequencing yielded the Wzy translational startpoint. Next, the Wzy protein was C-terminally tagged with the FLAG peptide, and immunoblotting of an S. typhimurium strain expressing a low-copy wzy–FLAG gene (five copies per cell) localized the intact Wzy protein in the cytoplasmic membrane of S. typhimurium cells. The Wzy protein was not well-expressed from a multi-copy wzy-FLAG + plasmid in S. typhimurium, or in E. coli K-12.

The Bordetella bronchiseptica tonB gene was cloned by detection of a chromosomal restriction fragment hybridizing with each of two degenerate oligonucleotides that corresponded to Pro-Glu and Pro-Lys repeats characteristic of known TonB proteins. The tonB Bb gene was situated upstream of exbB and exbD homologues and downstream of a putative Fur-regulated promoter. Hybridization results indicated that the tonB operon and flanking regions were highly conserved between B. bronchiseptica, Bordetella pertussis and Bordetella parapertussis. Disruption of tonB in B. bronchiseptica resulted in inability to grow in iron-limiting media, and inability to utilize alcaligin, enterobactin, ferrichrome, desferroxamine B, haemin and haemoglobin. Although it was not possible to inactivate tonB in a clinical B. pertussis isolate, tonB was disrupted in a laboratory B. pertussis strain previously selected for the ability to grow on Luria–Bertani medium. This B. pertussis tonB mutant shared a similar iron complex utilization deficient phenotype with the B. bronchiseptica tonB mutant. The B. bronchiseptica tonB operon present on a plasmid did not complement an Escherichia coli tonB mutant, but inefficient reconstitution of enterobactin utilization was observed in one fepA mutant harbouring plasmid copies of the B. pertussis fepA homologue and tonB Bb operon.

To survive and multiply in different environments, Vibrio cholerae has to co-ordinately regulate the expression of genes involved in adaptive responses. In many pathogens, adaptive responses, including pathogenic responses, are regulated by two-component regulator (TCR) systems. It is likely that members of a TCR family play a role in the regulation of processes involved in intestinal colonization, and therefore pathogenesis, in V. cholerae. We have identified and characterized a TCR system of V. cholerae: this system is a homologue of Escherichia coli PhoBR. The presence of a putative Pho box suggests that the V. cholerae phoBR operon is regulated by inorganic phosphate levels. The phoR and phoB genes are organized the same way as in E. coli. Mutation of the V. cholerae phoB gene affected the expression of the putative Pho regulon, including PhoA, but did not affect the production of cholera toxin. V. cholerae phoB mutants are less able to colonize rabbit intestine than wild-type V. cholerae. The addition of inorganic phosphate at a high concentration to the inoculum only partially restored the ability of the mutants to colonize the intestine, suggesting that the V. cholerae Pho regulon in vivo may not be regulated by inorganic phosphate levels alone.

In previous studies the authors cloned and characterized the DNA sequence of the regions at both ends of the O7-specific lipopolysaccharide (LPS) biosynthesis cluster of Escherichia coli VW187 (O7:K1), and identified the biosynthetic genes for dTDP-rhamnose and GDP-mannose, as well as one of the candidate glycosyltransferases. In this work the complete DNA sequence of a 6·9 kb intervening region is presented. Seven new ORFs were identified. All the functions required for the synthesis and transfer of the O7 LPS were assigned on the basis of complementation experiments of transposon insertion mutants, and amino acid sequence homology to proteins involved in LPS synthesis of other bacteria. Of the seven ORFs, two encoded membrane proteins that were homologous to the O-antigen translocase (Wzx) and polymerase (Wxy), two were involved in the biosynthesis of dTDP-N-acetylviosamine, and the remaining three showed homologies to sugar transferases. The O antigen chain length regulator gene wzz was also identified in the vicinity of the O7 polysaccharide cluster. O7-specific DNA primers were designed and tested for serotyping of O7 E. coli strains.

The fibrinogen-binding protein ClfA and the collagen-binding protein Cna are surface-associated adhesins of Staphylococcus aureus. ClfA has a dipeptide repeat region R composed mainly of serine and aspartate residues, more than 40 of which are required along with the 28-residue region W, the LPXTG motif and region M to display the ligand-binding region A on the cell surface in a functional form. Cna has a 61-residue region W and at least one 187-residue region B linking the collagen-binding region A to peptidoglycan. A cna mutant of S. aureus lacking region B was shown to bind collagen at the same level as wild-type Cna+ cells, indicating that region B is not necessary for ligand binding. Furthermore, altering the number of B repeats did not influence the level of collagen binding. In order to study the ability of C-terminal domains of Cna and ClfA to support functional ligand-binding activity of different adhesins, chimeric proteins were constructed and expressed in S. aureus. Surprisingly, the presence of a single Cna B domain and a nonapeptide linker located between ClfA region A and Cna region WM failed to support fibrinogen binding by S. aureus cells, despite the fact that ClfA region A was detected on the bacterial surface by immunoblotting. In contrast, the ClfA region A–Cna region B hybrid expressed as a recombinant protein in Escherichia coli did bind fibrinogen in Western ligand blots and in an ELISA-type assay. It is concluded that Cna region B cannot support functional display of ClfA region A on the bacterial cell surface. However, the ClfA dipeptide repeat region R and region WM did promote functional surface expression of the Cna collagen-binding domain in a hybrid Cna–ClfA protein.

An Escherichia coli–mycobacterial shuttle vector, pJCluc, containing a luciferase reporter gene, was constructed and used to analyse the Mycobacterium tuberculosis katG promoter. A 1·9 kb region immediately upstream of katG promoted expression of the luciferase gene in E. coli and Mycobacterium smegmatis. A smaller promoter fragment (559 bp) promoted expression with equal efficiency, and was used in all further studies. Two transcription start sites were mapped by primer extension analysis to 47 and 56 bp upstream of the GTG initiation codon. Putative promoters associated with these show similarity to previously identified mycobacterial promoters. Deletions in the promoter fragment, introduced with BAL-31 nuclease and restriction endonucleases, revealed that a region between 559 and 448 bp upstream of the translation initiation codon, designated the upstream activator region (UAR), is essential for promoter activity in E. coli, and is required for optimal activity in M. smegmatis. The katG UAR was also able to increase expression from the Mycobacterium paratuberculosis P AN promoter 15-fold in E. coli and 12-fold in M. smegmatis. An alternative promoter is active in deletion constructs in which either the UAR or the katG promoters identified here are absent. Expression from the katG promoter peaks during late exponential phase, and declines during stationary phase. The promoter is induced by ascorbic acid, and is repressed by oxygen limitation and growth at elevated temperatures. The promoter constructs exhibited similar activities in Mycobacterium bovis BCG as they did in M. smegmatis.

Tuberculosis (TB) was diagnosed in 10 seals from three species (Arctocephalus australis, Arctocephalus tropicalis and Otaria flavescens) found in South America. The mycobacteria isolated from these cases belonged to the Mycobacterium tuberculosis complex, as determined by RFLP using an IS6110 probe, spoligotyping, analysis of the 16S rRNA gene sequence and by PCR-restriction analysis of hsp65. Polymorphisms in gyrA, katG, oxyR and pncA were investigated in some of the isolates, as well as the presence of the MPB70 antigen. The insertion sequence IS6110 was present in three to seven copies in the genome of the mycobacteria isolated from seals. Using the IS6110 probe, six patterns (designated A, B, C, D, E and F) were identified from 10 different isolates. Patterns A and B were found for the mycobacteria isolated from two and four seals, respectively, indicating an epidemiological relationship between isolates grouped according to their IS6110 RFLP. The mycobacteria isolated from seals shared the majority of their IS6110 DNA-containing restriction fragments, and nine isolates had an identical spoligotype; only one isolate showed a minor difference in its spoligotype. In addition, none of these spoligotypes were found in other M. tuberculosis complex strains. These results suggest that the isolates from seals constitute a unique group of closely related strains. The mycobacteria isolated from seals showed polymorphisms at gyrA codon 95 and katG codon 463, as do group 1 M. tuberculosis, and M. bovis. Group 1 mycobacteria are associated with cluster cases. The spoligotypes found in the mycobacteria isolated from seals lack spacers 39–43, as does M. bovis, but the MPB70 antigen, which is highly expressed in M. bovis and minimally expressed in M. tuberculosis, was not detected in these mycobacteria. The mycobacteria isolated from seals also showed oxyR and pncA polymorphisms specific to M. tuberculosis. In conclusion, the mycobacteria that cause TB in seals in the South-Western Atlantic are a related group, and based on the combination of genetic characteristics, belong to a unique genotypic group within the M. tuberculosis complex.

Streptomyces chromosomes are linear DNA molecules and yet their genetic maps based on linkage analysis are circular. The only other known examples of this phenomenon are in the bacteriophages T2 and T4, the linear genomic sequences of which are circularly permuted and terminally redundant, and in which replication intermediates include long concatemers. These structural and functional features are not found in Streptomyces. Instead, the circularity of Streptomyces genetic maps appears to be caused by a completely different mechanism postulated by Stahl & Steinberg (1964 R32 , Genetics 50, 531–538) – a strong bias toward even numbers of crossovers during recombination creates misleading genetic linkages between markers on the opposite arms of the chromosome. This was demonstrated by physical inspection of the telomeres in recombinant chromosomes after interspecies conjugation promoted by a linear or circular plasmid. The preference for even numbers of crossovers is probably demanded by the merozygosity of the recombining chromosomes, and by the association between the telomeres mediated by interactions of covalently bound terminal proteins.

The enhanced green fluorescent protein (EGFP) gene is a modified version of the green fluorescent protein gene of the jellyfish Aequorea victoria with a codon usage that corresponds well to that found in many GC-rich streptomycete genes. Here the use of EGFP as a reporter for the analysis of spatially and temporally regulated gene expression in Streptomyces coelicolor A3(2) is demonstrated. The EGFP gene was inserted into plasmids that can replicate in Escherichia coli, greatly facilitating the construction of EGFP gene fusions. The plasmids can be transferred readily to S. coelicolor by conjugation, whereupon two of them (pIJ8630 and pIJ8660) integrate at the chromosomal attachment site for the temperate phage ϕC31. These vectors were used to analyse the spatial and temporal expression of sigF, which encodes a σ factor required for spore maturation, and of redD, a pathway-specific regulatory gene for the production of undecylprodigiosin, one of the four antibiotics made by S. coelicolor. While transcription of sigF appeared to be confined to developing and mature spore chains, transcription of redD occurred only in ageing substrate mycelium. A further plasmid derivative (pIJ8668) was made that lacks the ϕC31 attachment site, allowing the EGFP gene to be fused transcriptionally to genes of interest at their native chromosomal locations.

Cytological analysis of a series of Streptomyces coelicolor A3(2) mutants with disruptions of early sporulation (whi, for white aerial mycelium) genes in an isogenic background has provided new information about the role of whiH, and confirmed and extended previous knowledge about whiG, whiA and whiB. The characteristic straight aerial hyphae of whiG mutants contained normally spaced vegetative-like septa, while mutants in whiA or whiB had abnormally long and coiled aerial hyphae almost devoid of septation. whiG, whiA and whiB were all absolutely required for sporulation septation, and for all visible signs of nucleoid condensation and partitioning and other changes associated with later stages of sporulation. On the other hand, whiH appeared to enhance low basal levels of these processes. Thus, whiH mutant aerial hyphae were divided into loosely coiled fragments of variable sizes by what appeared to be a few sporulation septa. These fragments showed some spore-like characteristics and contained condensed and aberrantly partitioned nucleoids. whiG, whiA and whiB were epistatic to whiH on the criterion that they prevented such fragments from forming in double mutants. These spore-like features and the synthesis of clearly detectable levels of the whiE-directed grey spore pigment were not due to any residual activity of previously studied whiH alleles since they were retained by a constructed whiH null mutant. A model is presented that explains the mutant phenotypes by proposing two early developmental decision points involved in commitment to sporulation septation, one requiring whiG and the other requiring whiA and whiB.