Progress 09/01/02 to 08/31/06
Outputs
Microbial source tracking (MST) is a process by which fecal bacteria found in water are identified according to their original host animal. This process permits users to determine which animal sources are contributing the most fecal pollution to a waterway. There are a variety of methods available for MST, but at present library-independent genetic markers hold the most promise for identifying and partitioning fecal pollution. The purpose of this project was to identify novel genetic markers using Enterococci as indicator species. We used a unique mixed-genome microarray to screen approximately 4,300 gene fragments from enterococci collected and cloned from humans, cows, dogs, cervids, and waterfowl. We conducted 384 hybridization experiments (augmented with 38 recent hybridizations from goose feces) and identified approximately 100 putative markers. Subsequent validation using PCR testing from samples collected around the country indicated that 15 of these markers
were host-specific. A paper describing this work has been published in Applied and Environmental Microbiology (2006, vol 72 pages 1843-1851). During the summer of 2005 we field tested these markers in the Colville River drainage of northeastern Washington State. There is no gold standard by which to assess the outcome of these tests, but the cow and human markers generated data consistent with local knowledge of fecal pollution. Thus, we considered the effort successful and a manuscript is in preparation. We have developed a multiplex bead suspension array for these markers and continue to optimize the assay in preparation for publication. We have submitted an additional manuscript that describes an additional marker that appears useful for detecting goose fecal pollution. Finally, this project has raised a number of questions about the type of inferences that can be made using library-independent MST markers. Indeed, it is very possible to derive completely incorrect conclusions from
this type of assay depending on the underlying distribution of markers and we are developing a simulation program to examine this question and to examine the impact of assumptions underlying MST analyses; preliminary results are consistent with potentially problematic interpretations and this work identifies priority areas for future research. Unfortunately, we have been unable to garner additional USDA support for this research effort.
Impacts
Water managers across the country are confronted with biologically compromised waters. Mitigation efforts require knowledge about where the fecal pollution arises and objective source tracking methods are needed to garner support of stakeholders before cooperation can be expected. Our work identified 16 genetic markers for this purpose and our markers target enterococci for which there is a long history of use as a fecal indicator. Additional work in the field and with multiplex assay development adds to our ability to apply these markers in a cost-effective and relatively high throughput manner. Simulation studies indicate that significant analytical questions remain for the application of these markers; under some circumstances it is possible to make incorrect inferences about the primary sources of fecal pollution.
Publications
- Soule, M, EA Kuhn, F Loge, J Gay, and DR Call. 2006. Using DNA microarrays to identify library-independent markers for bacterial source tracking. Applied and Environmental Microbiology 72:1843-1851.
- One additional manuscript is in review (as of July 2006) and three additional manuscripts are in preparation. In addition, multiple presentations have been made to water managers and the public on behalf of this research effort.
|
Progress 07/31/03 to 07/31/04
Outputs
Microbial source tracking (MST) is a process by which fecal bacteria found in water are identified according to their original host animal. This process permits users to determine which sources are contributing the most fecal pollution to a waterway. There are a variety of methods available for MST, but for bacteria most of these methods rely on developing a classification library based on antibiotic resistance profiles or DNA fingerprinting methods. The library is then used to develop a classification function for use with water samples. These methods are generally limited in geographical range and the classification methods are imprecise. The quality of the libraries and validation data, if any, vary widely. An alternative approach is to identify genetic markers that are specific to a host. That is, if there are host-specific strains of bacteria, then there should be host-specific genetic markers that could be used for MST. In theory, these markers should be
independent of the geographic region and they would not require investment in a classification library. The goal of this project is to identify DNA markers for Enterococcus species that are strain specific for humans, cows, dogs, cervids, and waterfowl. We constructed mixed-genomic libraries from known-source isolates of Enterococcus. DNA fragments from these libraries (>4,300) were arrayed onto glass slides and a total of 384 strains of Enterococcus from five different hosts were hybridized to the arrays to identify genetic markers that were exclusive to specific hosts. In the process of this work we demonstrated that the microarrays themselves can be used as a library-dependent classification system. More importantly, we were able to identify approximately 70 markers that appeared to be host-specific. We selected a subset of these markers based on their potential as markers (encoding surface bound molecules or involved in carbohydrate metabolism) and determined the sequences of the
markers and developed PCR primers. PCR was then used to test fecal samples collected from around the United States to confirm specificity and sensitivity of the markers. A validation study is currently underway and preliminary results indicate that we may have identified 13 markers suitable for host-specific identification of enterococci originating from humans, cattle, dogs, and cervids. Sensitivity is <100% (not all sources harbor the markers), but specificity appears to be 100% by PCR. We were unable to identify host-specific markers for waterfowl. Validation tests are continuing and we will proceed with a pilot study of surface water samples and begin developing a more practical test based on these markers.
Impacts
This study is designed to identify genetic markers that are suitable for linking fecal bacteria (Enterococcus) to their original host animals (humans, cattle, dogs, and deer/elk). These markers should be suitable for discriminating between these hosts over broad geographic regions independent of cumbersome classification libraries and proprietary classification functions. If successful these tools will help watershed managers more effectively target mitigation plans for waters that are compromised by fecal pollution.
Publications
- Call, D.R., Soule, M., Loge F., and Gay, J.M. 2004. Identifying host-specific markers of fecal pollution using mixed-genome microarrays. Presentation at the USDA-CSREES National Water Quality Conference, Clearwater, FL, January 2004.
|
Progress 01/01/02 to 12/31/02
Outputs
The goal of this project is to develop a definitive tool for identifying host-specific fecal pollution in waterways. Although indicator bacteria, Enterococcus spp. in particular, will likely remain the mainstay of water quality analysis, classic biochemical, antibiotic resistance, and DNA fingerprinting methods are inadequate for identification of host-specific markers. We proposed to use an entirely novel approach, a mixed-genome microarray, to identify DNA markers for Enterococcus species that are strain specific for origination from humans, cows, dogs and waterfowl. Waterfowl were added to the array in response to reviewer suggestions. To construct the array, genomic DNA from host-specific strains is pooled and used to produce a shotgun library consisting of DNA fragments between 500 and 800 bp in size. Four separate libraries will be constructed and each library will include DNA from 50 strains of Enterococcus. Clones (n = 750) from each library will be used to
generate a 3,000 spot microarray. Genomic DNA from Enterococcus isolates obtained from these and other host organisms will then be hybridized to the array and probes most closely associated with specific hosts will be identified. These probes will further validated using geographically dispersed isolates and through examination of the distribution of markers in part of the Yakima River, Washington. We recruited a postdoctoral fellow for this project and we have isolated the necessary 50 strains of Enterococcus for each of the four libraries. Library construction has commenced.
Impacts
Host-specific markers of fecal pollution identified in this study are readily adaptable to existing water quality assessment technologies (e.g., PCR). These markers will permit partitioning the contributions from different host species to the waste stream, which is critical for improving waste management, health risk analysis, total daily maximum load (TDML) calculations, and for assessing impacts of changing land use and animal husbandry practices.
Publications
- No publications reported this period
|
|