Characterisation of rhizobia and studies on N₂ fixation of common weed legumes in New Zealand : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy

Abstract

Most legume species can fix atmospheric N₂ via symbiotic bacteria (collectively termed rhizobia) in nodules on their roots, thus allowing them to colonise marginal land with low soil N availability. Over the last 150 years, over 100 legume species from different continents have become naturalised in NZ and many of these are now common weeds. The major objective of this study was to genotypically characterise rhizobial isolates which produce N₂-fixing nodules on common weed legumes in NZ soils via phylogenetic analyses of 16S rRNA, recA, nifH, nodA and/or nodC gene sequences to establish their identity, diversity and presumptive origin(s). In addition, detailed studies were carried out on particular legumes or rhizobia linked to their ‘novelty’ outside and/or inside NZ. All legume plants sampled in the field were nodulated. Phylogenetic analyses of housekeeping and symbiosis gene sequences of 70 rhizobial isolates recovered from 19 legume species indicated that there is a diverse range of rhizobia which effectively nodulate weed legumes in NZ soils (Burkholderia, Bradyrhizobium, Ensifer, Mesorhizobium and/or Rhizobium depending on plant species). Almost all legumes were only effectively nodulated by rhizobia in a single genus. Exceptions were: (i) Dipogon lignosus, which had both determinate and indeterminate nodules, and was effectively nodulated by species of Burkholderia, Bradyrhizobium and Rhizobium and, (ii) Medicago sativa which was effectively nodulated by Ensifer sp. and Rhizobium sp.. This is the first report of a beta-rhizobia in NZ soils. Many of the rhizobia are likely to have been introduced into NZ either via plant materials/soil or commercial inoculants used in the field. However, for six Mesorhizobium isolates, there is a possibility of lateral transfer of symbiosis genes from introduced to native strains. There is also evidence of the occurrence of indigenous bradyrhizobia that are not associated with NZ native legumes but which are capable of nodulating the genistoid legumes. Host-specificity work confirmed that both native and exotic weed legumes are nodulated by disparate rhizobial populations in NZ. Bradyrhizobia could effectively nodulate legumes in the tribes Acacieae (Acacia), Genisteae (Chamaecytisus, Cytisus, Lupinus and Ulex), Loteae (Lotus and Ornithopus) and Phaseoleae (Dipogon). The symbiosis genes of the rhizobial isolates recovered here were largely plant species/tribe specific. However, bradyrhizobial cross-nodulation studies on seven legume genera of the Acacieae, Genisteae and Loteae indicated that legume species of the tribe Genisteae formed effective nodules with all Bradyrhizobium spp. tested. Those of the tribes Acacieae and Loteae only formed effective nodules with isolates associated with plants of their respective tribes. These plants generally showed greatest total dry weight when inoculated with strains isolated from the same host plants/related plants of the same genus. Cross-nodulation studies on M. sativa and Melilotus indicus indicated that they could form N₂-fixing nodules with Ensifer and Rhizobium isolates. However, these isolates showed differences in their ability to promote growth of M. sativa. Tolerance patterns of the rhizobial isolates to environmental stresses were largely based on the genus the isolates belonged to. The Burkholderia isolates were generally the most stress-tolerant and also showed phosphate solubilisation and siderophore production. However, the Ensifer isolates were the most salt tolerant. ¹⁵ natural abundance analysis indicated that N₂ fixation contributed substantially to the total N nutrition of Lupinus arboreus and Ulex europaeus sampled at sand dunes and hedges bordering intensive agricultural systems in Canterbury, respectively. Glasshouse studies via ¹⁵N-isotope dilution analysis showed that U. europaeus is a facultative N₂ fixer with an increased reliance on soil N in comparison to N₂ fixation as soil N levels increase. In conclusion, a wide range of rhizobia form effective nodules with specific weed legumes in NZ soils and many are likely to have been transported from abroad via plant material, soil and/or field inoculant but there is a possibility of pre-existing bradyrhizobial population occurring in NZ soils. Generally, their symbiosis genes are plant species/tribe specific but some legumes are able to form effective nodules with a wider range of rhizobia than those recovered from them in the field. The ability to fix N₂ could be an important factor in the establishment of these weed legumes in NZ.