Escherichia coli K-12 has long been known not to produce an 0 antigen. We recently identified two independent mutations in different lineages of K-12 which had led to loss of 0 antigen synthesis (D. Liu and P. R. Reeves, Microbiology 140:49-57, 1994) and constructed a strain with all rjb (0 antigen) genes intact which synthesized a variant of 0 antigen 016, giving cross-reaction with anti-017 antibody. We determined the structure of this 0 antigen to be -with an 0-acetyl group on C-2 of the rhamnose and a side chain c-D-Glcp on C-6 of GlcNAc. 0 antigen synthesis is rfe dependent, and D-GlcpNAc is the first sugar of the biological repeat unit. We sequenced the rjb (0 antigen) gene cluster and found 11 open reading frames. Four rhamnose pathway genes are identified by similarity to those of other strains, the rhamnose transferase gene is identified by assay of its product, and the identities of other genes are predicted with various degrees of confidence. We interpret earlier observations on interaction between the rjb region ofEscherichia coli K-12 and those ofE. colh 04 and E. coi Flexneri. All K-12 rjb genes were of low G+C content for E. coil. The rhamnose pathway genes were similar in sequence to those of (Shigella) Dysenteriae 1 and Flexneri, but the other genes showed distant or no similarity. We suggest that the K-12 gene cluster is a member of a family of rjb gene clusters, including those of Dysenteriae 1 and Flexneri, which evolved outside E. coli and was acquired by lateral gene transfer.Escherichia coli K-12 was isolated in 1922 and used as a standard E. coli strain at Stanford University for many years; the strains which survive all derive from cultures given to E. Tatum and others in the 1940s and early 1950s, when E. coli K-12 was first used for the genetic studies which led to its adoption as the major strain for laboratory study. After 50 years of intensive study, E. coli K-12 is arguably the best understood of all organisms, having been used for studies of many facets of living organisms, outlined in a two-volume book on E. coli and Salmonella enterica (45). Currently, 50% of its genome has been sequenced (56), and completion of this task will increase the focus of attention on E. coli K-12 for studies which integrate genomic information and biochemical processes.There are, however, significant gaps in our knowledge of E. coli K-12. During its first 25 to 30 years in the laboratory, it probably accumulated a range of mutations which improved adaptation to a laboratory environment but destroyed its ability to survive in its natural environment. Among them were mutations in the rjb gene cluster which led to loss of 0 antigen synthesis. The 0 antigen, a repeat unit polysaccharide which is a component of lipopolysaccharide (LPS), is the major surface antigen of many gram-negative bacteria (see reference 53 for a review) and, for E. coli, was present in by far the majority of strains when first isolated. However, it is often lost during culture, presumably because it offers no advantage under suc...

No abstract

Rhizobium tifolii mutants containing Escherichia coli lac gene fusions to specific nodulation (nod) genes were used to characterise phenolic compounds secreted from the roots of white clover (Trifolium repens) plants. These compounds either had stimulatory or inhibitory effects upon the induction of the nod genes. The stimulatory compounds were hydroxylated flavones and the most active compound was 7,4'-dihydroxyflavone. The inhibitory compounds present in white clover root exudates were umbelliferone (a coumarin) and formononetin (an isoflavone). Transcriptional activation of nod gene promoters in response to short exposures (3 h) of 7,4'-dihydroxyflavone was growth phase dependent; cells in early log phase were highly responsive to flavone additions in vitro and nod gene induction could be detected within 20 min of exposure at 5 x 10-7 M. Cells in other growth phases were generally unresponsive. A 10-fold molar excess of umbelliferone to 7,4'-dihydroxyflavone resulted in complete inhibition of nod gene induction. Some commerciallyobtained flavones were found to have weak stimulatory activity but could also inhibit nod gene induction by more effective stimulatory compounds. Strong stimulatory and inhibitory compounds all possessed a 7-hydroxy moiety and showed other structural similarities. This suggested that there was one binding site for these compounds. Because the response to these compounds was rapid, we propose that these phenolics act at the bacterial membrane or that an active uptake system is involved.

Several transposon Tn5-induced mutants of the broad-host-range Rhizobium sp. strain NGR234 produce little or no detectable acidic exopolysaccharide (EPS) and are unable to induce nitrogen-fixing nodules on Leucaena kucocephala var. Peru or siratro plants. The ability of these Exo-mutants to induce functioning nodules on Leucaena plants was restored by coinoculation with a Sym plasmid-cured (Nod-Exo+) derivative of parent strain NGR234, purified EPS from the parent strain, or the oligosaccharide from the EPS. Coinoculation with EPS or related oligosaccharide also resulted in formation of nitrogen-fixing nodules on siratro plants. In addition, an Exo-mutant (ANU437) of Rhizobium trifolii ANU794 was able to form nitrogen-fixing nodules on white clover in the presence of added EPS or related oligosaccharide from R. trifolii ANU843. These results demonstrate that the absence of Rhizobium EPSs can result in failure of effective symbiosis with both temperate and subtropical legumes.A complex multistep interaction between the soil bacterium Rhizobium and specific leguminous plants results in the induction of nitrogen-fixing nodules on legume roots (19 and references therein). The early steps of the interaction are characterized by the distortion or curling of the root hair cells. The cell walls of the root hairs are penetrated after 24 h by a compatible Rhizobium strain, and an infection thread is synthesized by the plant after the nucleus of this cell has migrated to the infection site (4, 17). The bacteria are carried toward the root cortex inside the infection thread, where they actively divide. Shortly before or concurrent with initiation of infection thread synthesis, cortical cell division is thought to be induced by the Rhizobium strain, presumably by diffusible substances released by the bacterium (1, 2).Another feature of the Rhizobium-legume interaction is the host specificity displayed. Fast-growing ("temperate") Rhizobium strains, for example, usually nodulate only one plant species effectively, whereas slow-growing Bradyrhizobium strains typically have a broad host range. In contrast, the fast-growing Rhizobium sp. strain NGR234 (28) possesses an unusually extensive host range, which includes a variety of tropical and temperature legumes as well as the nonlegume tropical tree Parasponia andersonii (29).Since the initial interaction between the symbionts occurs at the surface of the two organisms, cell surface molecules may be important in determining the outcome of the infection. Rhizobia characteristically produce large amounts of exopolysaccharides (EPSs) on various laboratory media, and the colonies formed are mucoid (Muc+) Rhizobium polysaccharides, particularly EPSs and lipopolysaccharides, have been postulated to be involved in the infection and nodulation of legumes (24, 25), including specific adhesion to the root hair surfaces (9) and the determination of host specificity (13). Exo-mutants of Rhizobium meliloti are apparently affected at an early stage of infection (15). In other species, EPS m...