Genetic diversity of bacterial β-glucosidase-encoding genes as a function of soil management

Abstract

This study is the first approach to evaluate the diversity of bacterial β-glucosidase-encoding gene sequences, aiming to identify the main environmental factors structuring bacterial β-glucosidase genetic diversity in semiarid soils. Two agricultural management systems, soils under spontaneous cover vegetation vs. noncovered herbicide-treated soils, were tested. The weed biomass generated in the former was estimated around 2,600 kg ha−1 year−1, whereas leaves and root exudates from olive trees were the only input of C biomass in the latter. Dendrograms generated from polymerase chain reaction–denaturing gradient gel electrophoresis profiles of bacterial β-glucosidase-encoding genes revealed two clusters determined by soil treatment and sharing <20 % similarity. The sequences of a total of 59 DNA fragments, representing 39 operational taxonomic units, were successfully determined. The Proteobacteria phylum clearly dominated all the soil samples, but representatives of Chloroflexi, Deinococci, Actinobacteria, Thermotogae, and Firmicutes class were also detected. Management strategies favoring the presence of spontaneous vegetation determined a higher genetic diversity of β-glucosidase-encoding genes of soil bacteria. However, since there is little information of β-glucosidase gene sequences available in databases, it is difficult to establish particular relationships between bacterial networks for C degradation and land use. Results from canonical correspondence analysis indicated that bacterial metabolic networks for oligomeric C substrates utilization were affected by the physicochemical properties of the soil; the uppermost 10 cm of covered soil clustered together and were positively correlated with some chemical properties related to soil fertility, whereas less influence of soil texture was observed for the deeper layers of bare soils.