Microbial diversity and gene mining in Antarctic Dry Valley mineral soils

Abstract

Soil communities are regarded as among the most complex and diverse assemblages of microorganisms with estimated bacterial numbers in the order of 109-1 cells.g. Studies on extreme soils however, have reported lower cell densities, supporting the perception that the so-called extreme environments exhibit low species diversity. To assess the extent of microbial diversity within an extreme environment, the mineral soils of the Dry Valleys, Ross Dependency, Eastern Antarctica were investigated using 16S rDNA analysis. Three mineral soils designated MVG, PENP and BIS were analysed, each differing with respect to altitude, protein, lipid, water and DNA content. The mid-altitude sample, MVG, yielded the highest levels of DNA and the low altitude BIS soil contained the highest levels of protein, lipid and water. 16S clone libraries were constructed and 60 unique clones were identified and sequenced. BLASTn analysis revealed eight phylogenetic groups with Cyanobacteria, Actinobacteria and Acidobacteria representing the majority. The Cyanobacterial phylotypes were unique to the desiccated high-altitude soils of the PENP sample, suggesting a soil-borne Cyanobacterial population. 21% of the phylotypes identified were assigned as ‘uncultured’. DNA isolated from the Antarctic mineral soils was also used to construct a metagenomic clone library consisting of 90700 clones with an average insert size of 3.5 kb, representing an estimated 3.4% of the available metagenome. Activity-based screening of the library for genes conferring lipolytic activity yielded no positive clones. It is suggested that the failure to produce positive clones might be a result of insufficient nucleotide coverage of the metagenomic DNA. The metagenomic DNA extracted from the Dry Valley mineral soils was further analyzed using PCR. Two sets of degenerate primers based on conserved regions within lipolytic genes were used to target lipase and esterase genes. One set of primers was selected from a previous study. A second primer set was designed manually from amino acid alignments of true lipase genes from family I, sub-families I-VI. PCR analysis resulted in nine partial gene fragments varying between 240 bp and 300 bp. Bioinformatic analysis revealed that all nine partial gene fragments harboured α/β-hydrolase motifs, putatively identifying two esterases and three lipases from both bacterial and fungal origin.