Requirement for Exopolysaccharide in the Mesorhizobium-Lotus Symbiosis

Abstract

The establishment of a successful symbiotic interaction between rhizobia and legumes requires complex molecular communication between the two partners in order to determine compatibility and co-ordinate symbiotic developmental responses. Although a symbiotic association, the invasion of plant nodules by rhizobia bears parallels to plant pathogen interactions with a successful outcome requiring either avoiding recognition by the host plant’s innate immune system or suppressing its activation. Exopolysaccharide (EPS) production is ubiquitous amongst rhizobia and the aim of this work was to determine its function in the Mesorhizobium-Lotus symbiosis. Evidence is presented that wild-type EPS may suppress a defence response and, in its absence, truncated EPS molecules produced by certain EPS mutants may cause the bacterium to be recognised as a foe. Other EPS mutants may remain blind to the defence system. Colony and symbiotic phenotypes of a range of M. loti R7A EPS mutants were examined on Lotus corniculatus and L. japonicus Gifu. Strains disrupted in the early stages of EPS biosynthesis were symbiotically proficient (e.g. exoB and, to a lesser extent, exoA mutants), whilst strains affected in later stages (e.g. exoU mutants) were severely impaired at the stage of infection thread initiation. EPS extracts isolated from wild-type M. loti and mutant strains were chemically characterised and a proposed structure and biosynthetic pathway for R7A EPS was determined. R7A produces an O-acetylated acidic EPS that is an octasaccharide consisting of glucose, galactose, glucuronic acid and riburonic acid residues. R7A EPS mutants produced EPS fractions that contained varying glycosyl linkages, indicating variations in low-molecular-weight EPS production and/or alternative polysaccharides depending on the particular mutant strain. Strains R7AexoB and R7AexoU that exhibit contrasting symbiotic phenotypes were thoroughly examined. The results suggested that R7AexoB was symbiotically proficient due to either complementary signalling by an alternative polysaccharide in the absence of EPS or the absence of surface polysaccharides that normally elicit a plant defence response. R7AexoU was severely impaired in nodulation, most likely due to the production of a truncated EPS molecule that is actively perceived by the plant resulting in the activation of a defence response. Pre-inoculation results suggested that the defence response elicited by R7AexoU could be dampened by wild-type R7A but not by R7AexoB. Co-inoculation studies with a ∆nodA mutant suggested that EPS is required both for infection thread initiation and for release from the infection thread into the nodule primordia. Several symbiotically-proficient exoU suppressor mutants were isolated following transposon mutagenesis. The insertions in these strains were in the exoA or exoL genes involved in the early stages of EPS biosynthesis, or in four genes not previously linked to EPS biosynthesis. Mutagenesis in an R7A background indicated that the four novel genes were not involved in wild-type EPS production; however, the colony phenotypes of the suppressor mutants suggested that their symbiotic proficiency was due to the prevention of truncated EPS production by R7AexoU. Various mutation and complementation approaches were employed to investigate if EPS produced by M. loti NZP2037 is involved in the strain’s unique symbiotic abilities. Results suggested that it was not EPS but perhaps lipopolysaccharide or capsular polysaccharide produced by NZP2037 that was responsible for the observed differing symbiotic ability compared to other M. loti strains. The results obtained in this study support a signalling role for M. loti EPS that acts to suppress host defence responses, allowing for infection thread formation and the development of nitrogen-fixing nodules.