THE CHARACTERISTICS AND DIVERSITY OF INDIGENOUS RHIZOBIA THAT NODULATE SELECTED INDIGENOUS MULTIPURPOSE LEGUMINOUS TREES AND SHRUBS IN THREE SOILS OF GHANA

ABSTRACT Trees in general and leguminous trees in particular form an integral part of the traditional farming systems in Ghana. Compared to other plants, leguminous trees have the advantage that, they are generally capable of growing better on N-deficient soils due to their ability to convert unavailable atmospheric N2 into plant utilizable N. However, several factors including the abundance and effectiveness of the specific rhizobial partner, the available N and P in soil, are among the important factors that severely affect how much N2 can be fixed in these trees. This study was thus conducted to assess the abundance and characteristics of the rhizobia that nodulate 18 selected indigenous tree legumes grown in three representative soils of Ghana and to ascertain important soil nutrient constraints that affect their nodulation, nitrogen fixation and growth. The three soils belonged to the Hatso, Toje and Alajo local series (equivalent to Haplic lixisol, Rhodic lixisol and Calcic vertisol, respectively). The 200 Rhizobium isolates obtained from nodules of these tree legumes were found to be highly diverse and varied in their abilities to nodulate legumes other than the host plants from which they were isolated. The isolates were further characterized culturally, metabolically, phenotypically and for their effectiveness in fixing atmospheric nitrogen. Of the 10 multi-purpose shrubs and tree species belonging to the subfamily Mimosoideae examined as much as 70% of them formed nodules in the three soils, whiles only 20% and 10% of the tree legumes that formed nodules belong to Papilionoideae and Caesalpinoideae sub-families respectively. As to the Rhizobium isolates from these shrubs and tree species, those obtained from Pithecelobium spp and Melletia thonningi appeared to be highly specific, nodulating only their respective homologous hosts, while those from Acacia mangium, Albizia lebbek and Acacia auricloformis appeared to be slightly promiscuous, and moderately promiscuous for those from Leucaena leucocephela, Crotalaria ochloleuca, Cajanus cajan and Vigna unguiculata;, the NGR 234 strain, included as a standard was the most promiscuous, forming nodules on almost 67% of the eight host legumes on which it was tested. With regards to the occurrence of rhizobia in each soil, the three soils were found to harbour variable populations of indigenous rhizobia, ranging from 22 cells/g of soil for A. mangium and C. ochloleuca in Alajo soil, to 5,200 cells/g for compatible strains of Pithecelobium sp. (in each of the three soils), M. thonningi (in Toje and Hatso soils) and A. lebbek (in only the Hatso soil). Further examination of the characteristics of the Rhizobium isolates revealed that each shrub and tree legume species was nodulated by strains belonging to both the fast-growing (Rhizobium) and slow-growing (Bradyrhizobium) types. In terms of abilities to fix N2, the greatest majority (63%) of the isolates were found to be ineffective on their homologous host plants, with ineffectiveness ranging from 43% of the Acacia auricloformis strains to 88% of isolates from Acacia mangium. The proportion classified as being effective averaged 15% and ranged from 5% in the case of Acacia mangium to 28% of the Albizia lebbek isolates. The polymerase chain reaction (PCR) of the 16S rRNA gene of almost all the 60 selected Rhizobium isolates from the tree legumes gave a single band of size 1.5 kb, corresponding to the expected band size reported for rhizobia, while the combined restrictions of the 16S rRNA genes of the 60 rhizobia isolates with five enzymes (Hae 111, Rsa 1, Hpa 1, Hpa 11 and Alu 1) distinguished clearly 60 different combinations of patterns or fingerprints, representing 60 distinct 16S rRNA genotypes among the isolates. Characterization of the test rhizobia isolates based on simple PCR of the intergenic spacer (ITS) gene was most resolute as it provided several distinct band sizes indicating great variation among the rhizobia isolates’ ITS. However, restriction of the ITS with three restriction enzymes (Hae 111, Hpa 11 and Alu 1) did not give much further distinctions among the test isolates. Experiments conducted in pots in the screenhouse using the Hatso and Toje soils indicated significant increases in both number of nodules formed, N2 fixation and total biomass of A. lebbek, A.mangium and A. auricloformis with P addition to soils. However, the application of phosphorus beyond 90 kg P/ha resulted in significant decrease in all these parameters in Hatso and Toje soils. With nitrogen fertilizer application, even the minimum level (40 kg N/ha) resulted in significant reduction in numbers of nodules formed on the tree legumes, even though it resulted in significant increase in plant biomass. However, phosphorus application mitigated the negative effect of nitrogen addition. Rhizobium inoculation also resulted in significant increase in both number of nodules formed and total biomass of the tree legumes over the uninoculated controls. Thus although the rhizobia that nodulate tree legumes in Ghanaian soils may be widespread, the rhizobia that occur in each soil appear to be heterogeneous or very diverse both phenotypically and genotypically and in their abilities to form effective nodules. To enhance nodulation, nitrogen fixation and growth of the tree legumes, consideration must be given to augmenting these soils with selected, highly effective strains together with phosphorus fertilizer.