Nodulation of field-grown Pisum sativum and Vicia faba: Competitiveness of inoculant strains of Rhizobium leguminosarum bv. viciae determined by an indirect, competitive ELISA method

Abstract

An indirect, competitive, enzyme-linked immunosorbent assay (ELISA) was developed to indentify strains (SU391, SU303, WSM937, NZP5472) of Rhizobium leguminosarum bv. viciae in the nodules of Pisum sativum and Vicia faba and to study the effect of legume species and indigenous R. l. viciae on the nodulating-competitiveness of these strains, applied as seed inoculants. The study involved field-grown crops at Avondale and Merredin in Western Australia and Wagga Wagga in New South Wales. A multivariate statistical procedure was used to determine an optimal classification rule by which nodule isolates were allocated as an inoculant strain or indigenous strain. The legume significantly affected the proportion of nodules formed by inoculant strains at Merredin. Strains (SU303, WSM937 and NZP5472) with V. faba cv. Fiord were very competitive with indigenous rhizobia, forming 53–92% of the nodules. In comparison, inoculant strains on P. sativum at Merredin were found in < 15% of the nodules. The proportion of nodules formed by inoculant strains varied significantly between the sites, feasibly in response to the number of indigenous R. l. viciae. Thus, soil populations of R. l. viciae at sowing time were <30, 537 and 16,980 g −1 soil at Avondale, Merredin and Wagga Wagga, respectively; and the corresponding proportional representation of inoculant rhizobia in nodules was 63–79, < 15 and < 5% on P. sativum, and 91, 80 and < 5% on V. faba. Therefore, it appears that indigenous R. l. viciae numbers of at least 500 g −1 soil at sowing may severely reduce the nodulating-competitiveness of seed-applied inoculant strains with pea; and numbers much > 500 g −1 soil, the nodulating-competitiveness of inoculant strains applied to faba bean. No inoculant strain was remarkably better than another at competing with indigenous R. l. viciae. We conclude that seed inoculation often may be more successful with V. faba than P. sativum, which is fortuitous because indigenous R. l. viciae are more likely to be less efficient for N 2 fixation with faba bean (cv. Fiord) than with pea. The relatively poor success with inoculation of P. sativum suggests the need to determine its efficiency for N 2 fixation with indigenous R. l. viciae, and to ensure novel cultivars are effectively nodulated by these rhizobia. The poor representation of inoculant strains in nodules on faba bean grown in soils with very large numbers of soil-borne R. l. viciae may severely reduce its productivity. Moreover, we propose that seed inoculation may not achieve a solution to this problem.