As the history and the diversity of rhizobia in the soil are quite different for both crops, the approaches of the strain selection programs have to be different. For the soybean crop, a continuous selection program is mandatory, since bacteria have to meet the increased N demand of more productive cultivars. The main approach consists of re-isolating adapted strains from areas, which have been previously inoculated, looking for variant genotypes with higher competitiveness and BNF capacity (Santos et al. 1999; Hungria, Vargas 2000). New strains are annually tested in national field trials, and recommendation for commercial inoculants can be changed if a more efficient and competitive strain is identified. Currently reinoculation guarantees a mean increase of 4.5% in grain yield, but can reach values as high as 25%. Therefore inoculation is practiced by 58% of the farmers, with 12.5 million doses (55% peat based and 45%> liquid) of inoculants sold in the country last year, representing 99% of the Brazilian market.
A different approach is used for the common bean crop and the strain selection is based on the search for efficient and competitive strains within the diverse indigenous population. A first premise of the Brazilian program is that strains have to belong to the species Rhizobium tropici, which shows higher genetic stability. A successful example was the strain PRF 81 (=SEMIA 4080), isolated from a soil of Paraná State that consistently increased yield by up to 900 kg ha"1, reaching yields of 3000 to 4000 kg ha"1 (Hungría et al. 2000). The sequencing of 16S rRNA genes has shown a high similarity of PRF 81 with Rhizobium genomic species Q strain BDV5102 isolated from Daviesa leptophylla in Australia (Lafay, Burdon 1998).
82^ 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99
HAMBI 540 S02
LMG 383 USDA 205
R. %uautlense A. radiobacter S.fredii uota ¿u
Sinorhizobium sp. C4 R. giardinii bv. giardinii H 152 244 Rizobium sp.
Rhizobium sp. USDA 1920 068 R. etli 096 R. etli 116 R. etli 190 R. etli
R. etli Olivia 4
R. mongolense USDA 1877T 079 R. etli
R. etli „„„ «. etli 029 R. etli 230 R. etli 061 R. etli 076 R. etli 082 R. etli 206 R. etli 164 R. etli 031 R. etli 039 R. etli 069 R. etli 160 R. etli 074 R. etli 133 R. etli 178 R. etli 072 R. etli 126 R. etli 246 R. etli 129 Rizobium sp. 215 Rizobium sp.
Rhizobium sp. OR 191
R. mongolense USDA 1844
R onlhrttm hv oalliriim R f\
K. mongolense USUA 1844' R. gallicum bv. gallicum R 602sp Rizobium sp. WSM1583 R. leg. bv. viceae ATC10Q04t „ R. leg. bv. phaseoli ATCC 8002 100 R. leguminosarum
Rhizobium genomic sp. Q BDV 5102 A. tumefasriens
PRF 81 LMG 9518
A. tumefasciens 233 R. tropici R. tropici IIB R. tropici IIA 034 R. tropici 093 Rizobium sp. T
R. tropici IIB CIAT 899T 077 R. tropici IIB
R. tropici IIB LMG 9517 086 Mesorhizobium sp.
M. plurifarium LMG 10056 062 Mesorhizobium sp. 084 Rizobium sp. 241 Rizobium sp.
Figure 2. Dendrogram built with the UPGMA algorithm with the aligned 16S rRNA partial sequences of Brazilian common bean isolates and of several type and reference strains. After Grange (2001).
Response to inoculation also depends on soil/crop management practices. In soybean, the zero-tillage management system reduces soil temperature and increases soil moisture resulting in a higher number of bradyrhizobial cells in the soil, higher rates of BNF and higher grain yield (Hungría, Stacey 1997; Ferreira el al. 2000; Hungría, Vargas 2000). Rhizobial number of diversity is also affected by soil management, and when samples from eight different sites in Brazil were analyzed, using soybean promiscuous primitive cultivars as trap plants, the following rhizobia were identified: R. tropici, R. huautlense, Rhizobium genomic species Q, rhizobia resembling agrobacteria, Rhizobium OR191, B. japonicum and B. elkanii. In soils under no-tillage, species present included B. japonicum, B. elkanii and R. tropici, while under conventional tillage the only species detected was R. tropici (unpublished data). These results indicate that even when management techniques considered appropriate for the tropics are used, rhizobial diversity can be drastically reduced.
For the bean crop, PCR-RFLP analysis of the 16S-23S rRNA intergenic spacer (IGS) and the 16S rRNA gene, indicated that the rhizobial populations from bean nodules cultivated in an unlimed acidic oxisol were less diverse than those from the limed soil (Andrade 1999). The 16S rRNA gene nucleotide sequences of isolates presented similarity values ranging from 97 to 100% with R. etli, R. gallicum, R. tropici, R. mongolense, R. leguminosarum, Sinorhizobium meliloti, Agrobacterium rhizogenes and A. tumefaciens. However, isolates affiliated to R. tropici IIB and to R. leguminosarum bv. phaseoli were predominant independent of lime application. The richness index (number of IGS groups) increased from 2.2 to 5.7 along the soil-liming gradient and, based on species, showed an increase from 0.5 to 1.4. The Shannon index (species diversity) ranged from 0.9 in unlimed soil to 1.4 in limed soil, and based on the number of IGS groups, the diversity increased from 1.8 to 2.8 (Andrade 1999).
Was this article helpful?