2000 RESEARCH PROJECTS Program Element 5 Assessment PROJECT: Optimization of Nonlinear Data Analysis Tools for the Assessment of Microbial Communities PRINCIPAL INVESTIGATOR: Craig C. Brandt PROGRAM ELEMENT 5 Assessment A major challenge in the successful implementation of in situ bioremediation is understanding the structure of the indigenous microbial community and how this structure is affected by environmental conditions. Culture-independent approaches that use biomolecular markers have become the key to comparative microbial community analysis. However, the large number and complex relationships among these measurements makes conventional statistical analysis of the data difficult. In our current NABIR research, we are developing artificial neural networks (ANNs) tools for relating changes in microbial biomarkers to the concentration of heavy metals. In this project, we propose to (1) refine ANN methods to handle a small sample size data sets; (2) use ANNs to reduce the dimensionality of measurement-rich data sets; (3) develop techniques for combining multiple data sets to increase the power of an ANN analysis; (4) demonstrate the utility of ANNs with a variety of data relevant to NABIR; and (5) provide ANN data analysis tools to other researchers and guidance in their use. The tools resulting from this research will contribute to the goals of the NABIR Program to provide better means for assessing the subsurface microbial community, thus improving bioremediation efforts for metals and radionuclides. PROJECT: Expanded Rapid, Comprehensive, Lipid Biomarker Analysis for Subsurface, Community Composition and Nutritional/Physiological Status as Monitors of Remediation and Detoxification Effectiveness PRINCIPAL INVESTIGATOR: David. C. White PROGRAM ELEMENT 5 Assessment We will assemble and validate a rapid, comprehensive, cost-effective suite of lipid biomarker measurements to quantify microbial community structure, activity and effectiveness thereby providing defensible community based endpoints for bioavailability and bioremediation success. The system will integrate enhanced rapid extraction of lipid biomarkers from groundwater membrane filter retentates or subsurface samples based on increased pressure/temperature that will deliver neutral lipids and polar lipids in two isolated fractions in less than an hour. The fractions are separated by High Performance Liquid Chromatography (HPLC) without the need for derivatization and volatilization of gas chromatography(thus greatly expanding the lipid components that can be analyzed). Electrospray Ionization (ESI) will efficiently transfer the ions in solution to the gas phase for MS/MS analysis. The system sequentially extracts neutral lipids from the environmental samples by supercritical carbon dioxide with a methanol modifier. Next, the residue is "flash" extracted with polar solvents at high pressure/temperature to recover polar lipids. Each fraction is then analyzed by HPLC/ESI/MS. In tandem MS/MS either collisionally activated dissociation (CAD) reactions or neutral/loss gain reactions performed in the collision cell of the mass spectrometer between the two analytical quadrupoles (MS/CAD/MS) enhances the specificity and sensitivity of the detection so smaller groundwater samples can be utilized. PROJECT: Development and Use of rRNA Gene-Based Microarrays for Assessing Microbial Community Composition and Dynamics PRINCIPAL INVESTIGATOR: Jizhong Zhou PROGRAM ELEMENT 5 Assessment Rapid, parallel, and cost-effective detection tools that can be operated in real time and in field-scale heterogeneous environments are needed for assessing microbial communities that impact the in situ bioremediation of radionuclides and metals. The objectives of this project are: (1) to optimize and validate rRNA gene-based microarrays for assessing microbial community composition and dynamics at radioactive and mixed waste sites; and (2) to create and implement new computer algorithms for designing oligonucleotide probes that are specific for different taxonomic groups of targeted organisms. We will optimize hybridization conditions with small-scale model oligonucleotide microarrays in terms of sensitivity, specificity, and quantitation, and validate larger prototype oligonucleotide arrays using environmental samples from the proposed FRC. We will also devise new bioinformatics programs that facilitate the probe design process for microarray applications. The research proposed here should provide a rapid, quantitative, field-applicable, and cost-effective tool for monitoring environmental microbial communities that, in turn, permits a more effective assessment of bioremediation strategies and endpoints. [Back to Award Recipients Page]