• rss feed
• contact

Terence L. Marsh, Ph.D.

Research

Microbial Community Analysis

We have developed a new approach for generating profiles of microbial communities referred to as terminal restriction fragment length polymorphism (T-RFLP). This is a culture-independent approach that combines the resolution powers and high throughput features of contemporary automated sequencing gel technology with the phylogenetic richness of the rRNA database. Briefly, community DNA is extracted directly from a substratum and used as a template for PCR amplification of 16S rRNA genes. One or both of the primers employed in the amplification are derivitized with a fluorescent ligand at the 5' terminus. End-labeling the primer permits the identification of only the terminal fragment in a restriction digest of the PCR products. In the case of a complex community this reduces the total number of restriction fragments from a standard RFLP analysis and allows one to conclude that each fragment represents one ribotype or operational taxonomic unit. Hence the approach offers a rapid and sensitive method to track many populations within a community as well as derive comparative estimates of diversity. The advantage of this approach is realized most effectively in comparative community analysis when unique ribotypes in an altered community can be compared to a controlled baseline community profile. This allows the detection of unique ribotypes against even a complex background. We have successfully applied this approach to a number of communities including activated sludge, soil, termite hindgut, and aquifer. We plan to extend this approach to other phylogenetically relevant markers in order to increase the number of populations in a community that can be tracked. The general approach has great promise in that communities can be dissected both phylogenetically and physiologically through a rational and systematic design of primers. Our laboratory is currently involved in the following projects that employ T-RFLP.

• Deep Subsurface We are characterizing the oligotrophic communities of the deep subsurface. This is an NSF funded project that is a spin off of a DOE deep drilling project in New Mexico. Results from the initial funding identified populations of oligotrophs with unusual phylogenetic signatures. We are characterizing these isolates as well as the community from which they were derived. Additional enrichments have been set up to confirm the initial observation and phylogenetic markers other than 16S rRNA will be used to confirm (or confound) the original observation. We view this line of experimentation as potentially very intriguing. In a recent publication Whitman et. al. (6) identify the subsurface as one of the major repositories of bacteria. Characterizing this large community should be a primary direction if we are to understand the role of the microbe in geochemical cycles and events.
• Soil Communities We have applied this technology to soil communities of contaminated and uncontaminated soils. We are currently funded to characterize the microbial communities of a super fund site in the upper peninsula of MI that has been contaminated with heavy metals and organic solvents. My collaborator and Co-P.I. for this project is David T. Long -- Geology-MSU. These studies have been a rich proving ground for the terminal restriction fragment approach. In our initial analyses of chromium contaminated soils from this site we identified terminal fragments that appeared to be unique to sites with high levels of chromium. Based on the results from three separate restriction digests, we tentatively identified the fragments as derived from the Cytophaga-Flexibacter phylogenetic group. This was confirmed with Cytophaga specific primers applied to T-RFLP as well as results from direct sequencing of 100 unique isolates from an rRNA gene library (in collaboration with Mike Dojka and Norman Pace, manuscript in preparation).
• Aquifer Communities We are just beginning an analysis of the microbial communities of aquifers. This is a logical extension of our work with soils and the deep subsurface. The microbial communities of aquifers have only recently been the focus of intensive investigations. It is clear that the aquifer systems of the continents are of vital importance to a healthy functioning ecosystem. Knowledge of the rate at which water is cycled through the aquifer system and the extent and rate of remediation that occurs naturally within these systems is of critical importance to understanding the level of contaminants that can be loaded onto this fragile system. The microbial community plays a critical role in these geochemical processing events and knowledge of the microbial community structure of aquifers can provide insight into the capacity of natural remediation within an aquifer as well as provide a basis for predictions regarding the reclamation of severely contaminated aquifers. We have begun characterizing the aquifer communities at three sites including the Bachman site (collaboration with Jim Tiedje and Frank Loefler), the Dover Air Force Base aquifer (collaboration with John Davis @ DOW, funded through RTDF), and a pristine aquifer on the Delmarva Peninsula (DOE-NABIR funded).
• Preparation of a T-RFLP web site We are currently preparing a web site to facilitate community analysis with T-RFLP. Niels Larsen has constructed a web site for the comparative analysis of microbial communities using the output from a T-RFLP run on the ABI automated sequencer (http://rdp.cme.msu.edu) We are also developing a web site for determining the terminal restriction fragment sizes for all complete rRNA sequences and placing the information in a phylogenetic context. This is in collaboration with Jim Cole, Paul Saxman, and Jim Tiedje at the Center for Microbial Ecology at MSU.

Evolution of microbial populations and genome plasticity

• Long-term Evolution of a Soil Isolate. We have continued the analysis of eighteen evolved populations of a soil isolate, TFD41, (Ralstonia sp.) that were maintained under laboratory conditions for 1000 generations (3). This is a long-term experiment in evolution that parallels the long-term E.coli experiment of Lenski and collaborators (2). The focus of our experiments can be divided into a more detailed assessment of the major phenotypic changes that have occurred over the 1000 generations and a characterization of the genotypic change(s) common to all evolved lines. From the outset, a long term objective has been to catalog and understand the genotypic and resulting phenotypic changes that have led to the increase in relative fitness of the evolved populations in comparison to the ancestral stock. We have detected large changes in cell morphology that include significant alteration to the outer envelope (see SEM portraits below). Nakatsu et al (4) have identified a deletion that is common to all of the evolved populations. We have sequenced part of this deletion and are currently in the process of walking the chromosome to the deletion junctions. The region that is deleted will be reintroduced into several of the evolved populations to test for the effects of this region on relative fitness.
• Genome Plasticity. During our investigations of the genotypic changes in the long-term evolution experiment described above, we noticed significant instability in the genome of this organism. We have identified regions on the chromosome that are deleted at an unusually high rate. There is precedence for this in the literature. A number of microbial systems have been reported to have unstable regions of the chromosome. Moreover Max Mergeay and colleagues (5) have reported on a process referred to as Temperature Induced Mutation and Mortality (TIMM) in various Ralstonia sp. The genotypic consequences of this phenomenon have been identified as deletions and rearrangements of the genome (both chromosome and plasmid) but the mechanics of the process are unknown. Our natural isolate (TFD41) displays a stress-induced phenotype that appears similar to that described by Mergeay. We have identified a region of the chromosome that is selectively deleted in populations that have survived the stress. Currently we are attempting to identify the deletion junctions as a beginning in an analysis of this process. We will also determine if TIMM has contributed to the evolution of TFD41 in the 1000 generation experiment described above.