Insight into soil microbial ecology through the development and application of serial analysis of ribosomal sequence tags (SARST) and a community-specific microarray

In order to overcome technical hurdles hindering progress in soil microbial ecology, this project involved the development and application of a molecular assay for measuring microbial community composition and diversity. Serial analysis of ribosomal sequence tags (SARST) generates libraries of a short and variable portion of the 16S rRNA gene. A series of enzymatic reactions amplifies and ligates ribosomal sequence tags (RSTs) into concatemers that are cloned and sequenced. On average, 5-10 RSTs were obtained from multiple phylotypes with each sequencing reaction. SARST was initially tested on: (1) activated sludge, (2) boreal forest soil, (3) a mixture of pure cultures and (4) duplicate libraries from an arctic tundra soil sample. SARST was also used to study microbial communities within soils sampled from Canada and Spain.;Bacterial diversity of geographically distinct locations was characterized for Canadian arctic tundra and boreal forest soil biomes. Composite samples taken from arctic tundra demonstrated significantly higher estimates of bacterial diversity than boreal forest soils. Also, despite climate differences, the overall phylotype compositions of boreal forest soil libraries did not cluster distinctly from tundra samples. As a result of large RST libraries, individual phylotypes were identified that were associated with all soil libraries and others were unique to specific samples. These sequences represent potentially cosmopolitan and endemic distributions of these organisms.;Several soils in the Basque Country of Spain are contaminated with high concentrations of hexachlorocyclohexane (HCH). However, the impact of HCH on soil bacterial community structure is not well understood. SARST enabled the design of a community-specific microarray for comparing multiple soil samples. The RST array was tested with pure cultures and by comparing soil hybridizations to denaturing gradient gel electrophoresis (DGGE) fingerprints from these same soils. Hybridization and DGGE results suggested that in contaminated soils, microbial communities were less diverse and had a higher proportion of predominant phylotypes. Several probe signals were correlated with HCH contamination (r>0.70), including one corresponding to Sphingomonas , a genus with known HCH degraders. Environmental parameters such as depth, pH and organic matter were also correlated with community structure in the soil samples.*;*This dissertation is a compound document (contains both a paper copy and a CD as part of the dissertation).