Soybean (Glycine max (L.) Merr.); nodulation mutant characterization: Role of isoflavonoid, nitrate, and abscisic acid in nodulation control

Isoflavonoids (daidzein, genistein, and coumestrol) are known to be inducers of nod genes in Bradyrhizobium japonicum. This study evaluated the effect of inoculation and nitrogen on isoflavonoid concentration in soybean (Glycine max (L.) Merr.) root extracts from selected hypernodulating mutants (NOD1-3, NOD2-4, and NOD3-7), a nonnodulating mutant (NN5), and the Williams parent. There were no significant differences in isoflavonoid concentrations of root extracts among lines when plants were not inoculated, while all mutants had higher isoflavonoid concentrations at 12 days after inoculation than did the Williams control when plants were inoculated. Nitrogen (urea, (NH$\sb4)\sb2$SO$\sb4$, and NO$\sb3\sp-$) application markedly decreased isoflavonoid concentrations in all soybean lines, with NO$\sb3\sp-$ being most inhibitory.Split-root studies with both Williams and NOD1-3 showed that root isoflavonoid concentration and nodule number, weight, and nitrogenase activity were markedly suppressed on the root-half receiving 5 mM NO$\sb3\sp-$, compared with the other root-half receiving 0 mM NO$\sb3\sp-$. Grafting studies between NOD1-3 and Williams showed that NOD1-3 shoots induced hypernodulation on roots of both Williams and NOD1-3 while Williams shoots induced normal nodulation on root stocks of both lines. Shoot control of hypernodulation in grafted plants was possibly due to NOD1-3 shoots inducing markedly higher root isoflavonoid concentrations than Williams shoots. Grafting confirmed that the nonnodulating phenotype of the NN5 mutant was strictly root controlled.Root medium application of ABA resulted in a decrease of nodulation in both Williams and NOD1-3, which may be causally related to the decrease root isoflavonoid concentrations. Isoflavonoid application into root medium enhanced nodulation in both Williams and NOD1-3, but it failed to induce nodulation on the NN5 nonnodulating mutant. This indicates that the nonnodulating phenotype is totally unrelated to isoflavonoids.The results shown here can be interpreted as evidence for involvement of isoflavonoids in nodulation control and support the previous conclusions of (1) greater nodulation capabilities of hypernodulating mutants than the Williams parent, (2) N inhibition of nodulation and N$\sb2$ fixation, (3) more marked inhibitory effect of NO$\sb3\sp-$ than of urea on nodulation, (4) shoot control of hypernodulation phenotype, and (5) ABA inhibition of nodulation. The lack of any difference in nodulation and isoflavonoid concentration of a hypernodulating mutant and Williams in early stages after inoculation indicated that the role of isoflavonoids in differential nodulation, if any, was on advanced stages of nodule ontogeny rather than on initial infection stages.Ope