A. Willems, J. Schelkens, M. Gillis
Laboratory of Microbiology, Dept of Biochemistry, Physiology & Microbiology, University of Gent, K.L. Ledeganckstraat 35, B-9000 Gent, Belgium
Bradyrhizobium contains three named species (B. japonicum, B. elkanii, B. liaoningense) and a number of unnamed groups from various host plants. Genospecies I-XI (Willems et al. 2001a, 2001b) were revealed by DNA:DNA hybridization and genospecies I-VII confirmed by sequence analysis of the internal transcribed spacer (ITS) between the 16S and 23S rRNA genes. We have extended the ITS sequence analysis to 11 genospecies to show that it is a good indicator of close genetic relationship, gives valuable information on the genospecies grouping, and predicts where DNA:DNA hybridization will be useful. Such data permit easy comparison between laboratories. Three of the 11 genospecies represent the currently recognized species. To aid in naming the other eight genospecies, we used the following techniques on a subset of strains to find differentiating features: API ZYM and API 50CH tests, antibiotic sensitivity tests and fatty acid analysis.
API ZYM tests were not useful as just 3 of 19 tests were positive and only alkaline phosphatase could be used to differentiate genospecies VI and VIII from the others. API 50CH gave 27/49 positive results. Because of the slow growth, the procedure was modified and growth scored on days 2, 4, 7 and 9. Eleven strains were tested but more are needed to assess the system's potential.
Antibiotic susceptibility testing used the disc diffusion method on modified YMA (containing 5 g/L mannitol and MgCl2.6H20 instead of MgSO^FhO). Plates were inoculated with a cell suspension (density McFarland 0.5) and incubated with antibiotic disks for 4 days at 28°C before inhibition zone diameters were read using a digital caliper. The following antibiotics were useful to differentiate between some of the genospecies: kanamycin (K30), carbenicilin (CAR100), tetracycline (TE30), cefuroxime (CXM30), minocycline (MH30), rifampicin (RD5/30) and novobiocin (NV30). Overall, the genospecies VI and VIII were most susceptible and the genospecies II (B. elkanii), X and XI were resistant to many of the antibiotics tested.
For fatty acid analysis, each strain was grown on a plate of modified YMA medium as described above and from these cultures a cell suspension was prepared (density McFarland 5) to uniformly inoculate 1 to 6 plates of the same medium. The number of plates to be used was assessed from the growth of the first culture. Plates were incubated for 4 days at 28°C. Fatty acid methyl esters were extracted and analyzed using the MIDI system (Microbial ID Inc.). The major fatty acids found were 18:1 w7c and 16:0, in line with literature data. Some of the smaller constituents allowed differentiation of the genospecies, but not all genospecies could be differentiated.
So far, we can conclude that phenotypic data for the differentiation of the different Bradyrhizobium genospecies are difficult to obtain. Most procedures have to be modified because of the slow growth of bradyrhizobia. Some groups, such as the genospecies containing photosynthetic isolates (VI and VIII), may be recognized using fatty acid composition and antibiotic susceptibilities, but more strains from the different groups need to be included to confirm this. The API 50CH system may provide further differentiation, but also here, more strains need to be included.
Was this article helpful?