Final Report: Microbial Community Microarrays to Assess Chemical and Biological Characteristics of Water QualityEPA Contract Number: EPD07041
Title: Microbial Community Microarrays to Assess Chemical and Biological Characteristics of Water Quality
Investigators: Marshall, Michael
Small Business: Southeast TechInventures, Inc.
EPA Contact: Manager, SBIR Program
Project Period: March 1, 2007 through August 31, 2007
Project Amount: $69,904
RFA: Small Business Innovation Research (SBIR) - Phase I (2007) RFA Text | Recipients Lists
Research Category: SBIR - Water and Wastewater , Small Business Innovation Research (SBIR)
This research focused on demonstrating that a suite of environmental microorganisms can serve as bioindicators for the presence of chemical pollutants, using mercury as a proof of principle. Ultimately, “DNA fingerprints” could be derived from microbial mercury bioindicators and used to develop a DNA microarray, or other probe-based platform, for mercury detection. Specific Phase I objectives included: (1) recovering a collection of microbes that are relatively common in mercury-contaminated sediments, but not in uncontaminated sediments; and (2) developing a mercury detection component for the WaterChip™ technology, a patented system based on unique “DNA fingerprints” derived from microbial bioindicators. To accomplish the first objective, sediment samples were collected from mercury-contaminated and uncontaminated sites in the Great Lakes (GL-DRTC), the Florida Everglades (FL-WCA1), and the Holston River (NFHR 94, 80.8 and 77) in Saltville, Virginia, and analyzed for total mercury and monomethyl mercury levels. Samples were processed by extracting genomic DNA, amplifying SSU rDNA by polymerase chain reaction (PCR), and cloning PCR amplicons to produce plasmid clone libraries. Fifty clones from each library were sequenced, aligned, and grouped into operational taxonomic units (OTUs) so that sample comparisons could be made. OTUs that occurred in at least two mercury-contaminated samples, but not in uncontaminated samples, were selected as candidate mercury bioindicators. Real-time quantitative PCR experiments were used to assess relative abundance patterns of candidates across all samples, including the control sample. This provided a method for screening candidates that produced quantitative differences, 5-fold or greater, between mercury-contaminated and uncontaminated samples. To accomplish the second objective, a 50-mer oligonucleotide probe was derived from the V2 region and checked for uniqueness by aligning it with probes already spotted on the prototype WaterChip™. Microarrays then were printed with candidate probes and validated amplification/labeling protocols were used to test probe specificity by hybridizing reaction products to the array. In a control experiment, a plasmid clone dilution prepared using bioindicator rDNA was amplified in a multiplex PCR and hybridized overnight for 20 hours at 48°C. This process then was repeated using genomic DNA known to contain the bioindicator rDNA.
Based on sequence analyses, 302 OTUs were recovered from four samples: NFHR 77 and 80.8, FL-WCA1, and GL-DRTC. Rank-abundance profiles were generated for each sample and informally represent “species diversity.” After comparing OTUs across samples, 14 were selected as candidate mercury bioindicators based on their presence in mercury-contaminated samples only. However, candidates did not include any OTUs recovered in FL-WCA1 for two reasons: (1) none of the 16S rDNAs in this sample were found in other mercury-contaminated samples; and (2) the total mercury analysis for FL-WCA1 was below reporting limits, suggesting that it was actually uncontaminated. Of the 14 candidates, only two were identified at the species level based on Basic Local Alignment Search Tool (BLAST) alignments and a 97.5 percent sequence identity threshold, and the other 12 represent either reported, but uncultured, environmental isolates or novel microbes. Interestingly, one candidate aligned very closely (99% sequence identity) with an unidentified sequence (Accession No. EF552046) that was isolated in a research study described only as a “Microbial Community Analysis of Two Field-Scale Sulfate-Reducing Bioreactors Treating Mine Drainage” (http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?db=nuccore&id=146575902 Exit ). This suggests that certain microbes, characterized as bioindicators, may provide ecological functions that could be used to treat contaminated environments. Thus, the bioindicator screening methodology used in this project also may reveal candidate bioremediators. Quantitative polymerase chain reaction (q-PCR) results generally confirmed that candidate bioindicators could be detected in all mercury-contaminated samples (except FL-WCA1) and quantitative differences corresponded with their relative abundance, as suggested by sampling clone libraries. The results of the microarray hybridizations confirmed that bioindicator probes are highly specific under stringent conditions, including a multiplex PCR containing 46 sets of primers and hybridization reactions carried out at 48°C for 20 hours.
Several important outcomes resulted from this project. In general, the sampling and experimental methodologies utilized can detect potential bioindicators in the environment based on the presence of unique DNA signatures that are recovered. This process can be used to expand the diagnostic capacity of the WaterChip™ (Patent No. 7214492) to mercury detection, in which the microbial community would act as an indirect detection assay. This would permit widely dispersed water sources to be tested because many of the same bioindicators would be endemic in each environment. Also, mercury levels were measured using approved EPA testing protocols, so the WaterChip™ could be evaluated in the context of current EPA regulatory guidelines. Suites of microbial bioindicators that are diagnostic of other environmental chemicals and pollutants could be developed in the same way as mercury and incorporated into the WaterChip™ in modular fashion. Ultimately, this could lead to a cost-effective and comprehensive tool for evaluating sediment and water quality. Further, this approach has potential as a method for discovering microbial bioremediators.
The approach used to recover microbial bioindicators and derive “DNA fingerprints,” however, is not tied to microarray technology and other testing platforms may prove to be more suitable and economical. PCR-based technologies, in general, offer numerous advantages over current analytical methods because they are relatively inexpensive and more discriminating than other methods, allowing for commercialization opportunities. For example, municipal drinking water authorities conduct heavy metals testing infrequently due to the high cost and the fact that mercury accumulates within given populations gradually over a long period of time. They have indicated, however, that a rapid test for pathogens in water is very desirable because it could prevent unnecessary risk to the general population. In addition, water system cleanup costs could be substantially minimized if pathogen detection occurred early in the water treatment process. Thus, water quality authorities would consider a more economical metal test using PCR-based bioindicator detection much more beneficial if it was packaged together with pathogen detection.