Polyphasic analysis reveals correlation between phenotypic and genotypic analysis in soybean bradyrhizobia (Bradyrhizobium spp.)

Abstract

Soybean bradyrhizobia (Bradyrhizobium spp.) are bacteria that fix atmospheric nitrogen within the root nodules of soybean, a crop critical for meeting global nutritional protein demand. Members of this group differ in symbiotic effectiveness, and historically both phenotypic and genotypic approaches have been used to assess bradyrhizobial diversity. However, agreement between various approaches of assessment is poorly known. A collection (n=382) of soybean bradyrhizobia (Bradyrhizobium japonicum, B. diazoefficiens, and B. elkanii) were characterized by Internal Transcribed Spacer – Restriction Fragment Length Polymorphism (ITS-RFLP), cellular fatty acid composition (fatty acid methyl esters, FAME), and serological reactions to assess agreement between phenotypic and genotypic methods. Overall, 76% of the accessions demonstrated identical clustering with each of these techniques. FAME was able to identify all 382 accessions, whereas 14% were non-reactive serologically. One ITS-RFLP group, containing 36 Delaware isolates, produced multiple ITS amplicons indicating they possess multiple ribosomal RNA (rrn) operons. Cloning and sequencing revealed that these strains contained as many as three heterogenous rrn operons, a trait previously unknown in bradyrhizobia. A representative subset of 96 isolates was further characterized using 16S rRNA and Internal Transcribed Spacer (ITS) amplicon sequencing. ITS sequences showed better inter- and intra-species discrimination (65–99% identity) than 16S sequences (96–99% identity). This study shows that phenotypic and genotypic approaches are strongly correlated at the species level but should be approached with caution. We also suggest using combined 16S and ITS genotyping data to obtain better inter- and intra-species resolution in bradyrhizobia classification.