When does a clone deserve a name? A perspective on bacterial species based on population genetics.

Molecular population-genetic analysis has revealed that for several human diseases, including tuberculosis, plague and shigellosis, the generally accepted taxonomic status of the organisms involved does not fit the usually accepted genus or species criteria. This raises the question of what species concept to apply to bacteria. We suggest that the species definition in bacteria should be based on analysis of sequence variation in housekeeping genes, and also that the "clone" be given official status in bacterial nomenclature. This will allow demotion of the species or genus status of several traditionally recognized human pathogens, but retention of current names of anomalous species and genera as clone names.

Intraspecies variation in bacterial genomes: the need for a species genome concept.

Bacterial populations are clonal. Their evolution involves not only divergence between orthologous genes but also gain of genes from other clones or species, which has only recently been widely appreciated through macrorestriction mapping, genomic subtraction and complete genome sequencing. Genes can also be lost in response to selection or by random mutation after becoming redundant. The bacterial genome is a dynamic structure and intraspecies variation needs to be included in genome analysis if we are to gain insight into the full species genome.

Bacterial polysaccharide synthesis and gene nomenclature.

Gene nomenclature for bacterial surface polysaccharides is complicated by the large number of structures and genes. We propose a scheme applicable to all species that distinguishes different classes of genes, provides a single name for all genes of a given function and greatly facilitates comparative studies.

Genetic organisation and evolution of Yersinia pseudotuberculosis 3,6-dideoxyhexose biosynthetic genes.

3,6-dideoxyhexose (DDH) sugars occur in some of the O antigens of Salmonella enterica and Yersinia pseudotuberculosis, but are otherwise rarely found in nature. Y. pseudotuberculosis DDH biosynthetic genes rfbS (encoding CDP-paratose synthetase) and rfbE (encoding CDP-tyvelose epimerase) were amplified and cloned, and their sequences determined. Comparisons with the equivalent genes of S. enterica show that the genetic arrangement of DDH genes is very similar; however, in Y. pseudotuberculosis there is no suggestion that paratose producing strains are derived from tyvelose-producing strains by inactivation of rfbE, which is the case in S. enterica. The previously determined DNA sequence of the rfb region of an abequose-producing strain was re-examined. It contains the remnants of an insertion sequence (IS) adjacent to a truncated and non-functional rfbE gene. This suggests that the IS was involved in recombination events contributing to O-antigen antigenic diversity in Y. pseudotuberculosis.

Domain organisation in phosphomannose isomerases (types I and II).

Phosphomannose isomerase (PMI) types I and II were found to possess a conserved protein motif. This motif coincides with the catalytic site of the Candida albicans type I PMI, indicating a common catalytic process for both PMI types. The type II PMI are bifunctional enzymes possessing PMI and guanosine diphospho-D-mannose pyrophosphorylase (GMP) activity in separate catalytic domains, which in some species may function as separate proteins.

Presence of different O antigen forms in three isolates of one clone of Escherichia coli.

Escherichia coli strains ECOR2, ECOR3 and K-12 are very closely related in genotype as indicated by multilocus enzyme electrophoresis. We show that they have very different rfb regions indicating that recombination has occurred in this region, and we suggest that it may be associated with niche adaptation.

Molecular analysis of a Salmonella enterica group E1 rfb gene cluster: O antigen and the genetic basis of the major polymorphism.

Salmonella enterica is highly polymorphic for the O antigen, a surface polysaccharide that is subject to intense selection by the host immune system. This polymorphism is used for serotyping Salmonella isolates. The genes encoding O antigen biosynthesis are located in the rfb gene cluster. We report here the cloning and sequence of the 19-kb rfb region from strain M32 (serovar anatum, group E1) and compare it with that of strain LT2 (serovar typhimurium, group B). Genes for biosynthetic pathways common to both strains are conserved and have very similar sequences. In contrast, the five genes for CDP-abequose synthesis, present in strain LT2, are absent in strain M32; three open reading frames (ORFs) of strain LT2, thought to include genes for transferases, are not present in strain M32 but are replaced by three different ORFs with little or low level of similarity. Both rfb gene clusters are low in G + C content, indicating that they were transferred from a common ancestral species with low G + C content to S. enterica relatively recently (in the evolutionary sense). We discuss the recombination and lateral transfer events which may have been involved in the evolution of the polymorphism.

Sequence and analysis of the O antigen gene (rfb) cluster of Escherichia coli O111.

The O antigens found in Salmonella enterica (Se) and Escherichia coli (Ec) show a great deal of diversity, and only three structures are known to be common to both genera. Two of them contain the 3,6-dideoxyheoxse colitose, not found in other serogroups of the two species. The first of these is common to Ec O111 and Se O:35 (sv Adelaide); the other is found in both Ec O55 and Se O:50 (sv Greenside). The genes specific for the synthesis of O antigen are generally located in the rfb gene cluster at map position 45 min in Ec and 42 min in Se. The rfb (O antigen) gene cluster of an Ec O111 strain M92 had been cloned earlier and hybridisation analysis suggested that the rfb clusters of Ec M92 and a Se sv Adelaide strain had been acquired separately by the two species since their divergence. We have now sequenced part of the rfb cluster from Ec M92. We identify two genes of the GDP-colitose pathway, rfbM and rfbK, and show that several other ORFs have similarity to the rfb and cps (capsular polysaccharide) genes. Downstream of this block of genes is an ORF which encodes a protein with predicted transmembrane segments which is presumed to correspond to the rfbX gene. The % G+C values of the Ec M92 rfb sequence are extremely low, indicating that the rfb evolved in a low % G+C species of bacteria before transfer into Ec.

Sequence of the E. coli O104 antigen gene cluster and identification of O104 specific genes.

The Escherichia coli O104 polysaccharide is an important antigen, which contains sialic acid and is often associated with EHEC clones. Sialic acid is a component of many animal tissues, and its presence in bacterial polysaccharides may contribute to bacterial pathogenicity. We sequenced the genes responsible for O104 antigen synthesis and have found genes which from their sequences are identified as an O antigen polymerase gene, an O antigen flippase gene, three CMP-sialic acid synthesis genes, and three potential glycosyl transferase genes. The E. coli K9 group IB capsular antigen has the same structure as the O104 O antigen, and we find using gene by gene PCR that the K9 gene cluster is essentially the same as that for O104. It appears that the distinction between presence as group IB capsule or O antigen for this structure does not involve any difference in genes present in the O antigen gene cluster. By PCR testing against representative strains for the 166 E. coli O antigens and some randomly selected Gram-negative bacteria, we identified three O antigen genes which are highly specific to O104/K9. This work provides the basis for a sensitive test for rapid detection of O104 E. coli. This is important both for decisions on patient care as early treatment may reduce the risk of life-threatening complications and for a faster response in control of food borne outbreaks.

Periplasmic expression of part of the major rotavirus capsid protein VP7 containing all the three antigenic regions in Escherichia coli.

Part of the porcine rotavirus outer capsid protein VP7 containing all the three antigenic regions was expressed as a chimeric protein with bacterial alkaline phosphatase (AP) in E. coli. The construct contains an ompF promoter, the DNA encoding the signal sequence and the first 12 amino acids of mature OmpF, part of vp7 and the DNA encoding mature AP. The chimeric protein is stable, retains the biological property of AP and ability to react with polyclonal antiserum against the virus, and can be exported through the bacterial inner membrane into the periplasm.

Synthesis in Escherichia coli of the major glycoprotein of human rotavirus: analysis of the antigenic regions.

Various regions of the gene encoding the major neutralization antigen, VP7, of human rotavirus have been expressed in Escherichia coli, as N-terminal fusions to beta-galactosidase under the control of the lac promoter. We have determined that the fusion products of two clones containing regions AB (aa 69-158) and ABC (aa 69-319) were antigenic, reacting with antibodies raised against whole virus. When guinea pigs were immunized with fusion protein purified by monoclonal antibody affinity columns, no neutralizing or virus-binding antibodies were detected, but antibodies binding to denatured VP7 were present.

Expression of rotavirus VP7 antigens in fusions with bacterial proteins.

We have cloned antigenic regions of the VP7 gene from rotavirus RV-5 into the lamB gene. The insertions discussed in this paper comprise 250bp of rotavirus DNA and cover the A and B antigenic regions of the protein. The fusion proteins are expressed and are present in the outer membrane: they react with anti-rotavirus antibody. However, as yet we have not been able to demonstrate that the fusion has LamB protein functions with regard to maltodextrins or L phage, and the fusion protein appears to exist in the outer membrane as a monomer rather than as a trimer.

Variation in O-antigens, niche-specific selection and bacterial populations.

Bacterial populations usually consist of distinct clones, often apparently adapted to specific niches. A formal model is developed whereby niche-specific selection maintains the polymorphisms involved in clonal adaptation. Infrequent transfer of non-adaptive alleles to a clone is balanced by the selection for the resident adaptive allele. The model can account for the extensive polymorphism in surface antigens observed in bacteria, and also for the existence of sympatric clones of pathogenic species which differ in host range and/or mode of pathogenesis. Niche-specific selection combined with low levels of genetic transfer can also account for the high level of neutral variation in bacteria, and indirectly for their ability to respond rapidly to environmental changes.

The colanic acid gene cluster of Salmonella enterica has a complex history.

The colanic acid gene cluster of Salmonella enterica LT2 was sequenced and compared with that of Escherichia coli K-12. The two clusters are similar with divergence slightly higher than average for genes of the two species. The cluster was divided into four blocks by GC content and seems likely to have transferred from a higher GC content species to the ancestor of E. coli and S. enterica. All 19 genes of K-12 and 13 genes of LT2 appear to have undergone random genetic drift with amelioration of the GC content. However, in the case of S. enterica, we believe that the six genes of the GDP-fucose pathway group were replaced relatively recently by genes closely related to those of the original donor species. Two repetitive elements were observed: a bacterial interspersed mosaic element in the intergenic region between wzx and wcaK in K-12 only and a RSA (repetitive sequence element) sequence between wcaJ and wzx in LT2 only.

A plasmid-borne O-antigen chain length determinant and its relationship to other chain length determinants.

We identify a function-controlling O antigen chain length for a plasmid-borne gene, cldpHS-2, harboured by Flexneri strains of Escherichia coli known to cause reactive arthritis. The predicted amino acid sequence of the gene product is very similar to those of other cld genes and that of fepE, thought to be part of the enterobactin iron uptake system of E. coli. The predicted proteins are compared with rfb-associated chain length determinants as a family of related genes.

Expression of the O antigen gene cluster is regulated by RfaH through the JUMPstart sequence.

O antigen genes are clustered, with a JUMPstart sequence located upstream. JUMPstart is a 39-bp sequence, present upstream of many polysaccharide gene clusters and also upstream of haemolysin and F factor gene clusters. RfaH is known to regulate the expression of E. coli group II capsule, haemolysin, F factor and the outer core of lipopolysaccharide all of which have the JUMPstart sequence, and has been shown to function as a transcriptional antiterminator in some cases. Using lacZ fusions to genes in the O antigen gene cluster of Salmonella enterica serovar Typhimurium, we found that RfaH also regulates the expression of O antigen. We showed that RfaH enhances expression of the 18-kb O antigen gene cluster, with promoter-distal genes affected more dramatically. We also showed that the JUMPstart sequence was required for RfaH function.

Molecular cloning and expression in Escherichia coli K-12 of chromosomal genes determining the O antigen of an E. coli O2: K1 strain.

The rfb gene cluster which determines the biosynthesis of the O2 O antigen has been cloned from an Escherichia coli O2: K1 strain isolated from a case of septicaemia in chickens. The region required for expression of the O antigen in E. coli K-12 was localised to a 10.7 to 14.15-kb segment which was shown to be chromosomal in origin with a close linkage to the gnd and his genetic loci.