Selecting rep-PCR markers to source track fecal contamination in Laguna Lake, Philippines
| Autor: | Windell L. Rivera, Kevin Labrador, Joseth Jermaine M. Abello, Gicelle T Malajacan, Christopher Rensing, Mae Ashley G Nacario, Luiza H Galarion |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Microbiology (medical)
Veterinary medicine Swine Philippines Biology Polymerase Chain Reaction law.invention Feces Fingerprint law Animals Water Pollutants Waste Management and Disposal Genotyping Polymerase chain reaction Water Science and Technology Significant difference Public Health Environmental and Occupational Health Linear discriminant analysis DNA Fingerprinting Fecal coliform Lakes Infectious Diseases DNA profiling Principal component analysis Cattle Female Chickens Environmental Monitoring |
| Zdroj: | Journal of Water and Health. 18:19-29 |
| ISSN: | 1996-7829 1477-8920 |
| DOI: | 10.2166/wh.2019.042 |
| Popis: | Fecal contamination is one of the factors causing deterioration of Laguna Lake. Although total coliform levels are constantly monitored, no protocol is in place to identify their origin. This can be addressed using the library-dependent microbial source tracking (MST) method, repetitive element sequence-based polymerase chain reaction (rep-PCR) fingerprinting. Serving as a prerequisite in developing the host-origin library, we assessed the discriminatory power of three fingerprinting primers, namely BOX-A1R, (GTG)5, and REP1R-1/2-1. Fingerprint profiles were obtained from 290 thermotolerant Escherichia coli isolated from sewage waters and fecal samples of cows, chickens, and pigs from regions surrounding the lake. Band patterns were converted into binary profiles and were classified using the discriminant analysis of principal components. Results show that: (1) REP1R-1/2-1 has a low genotyping success rate and information content; (2) increasing the library size led to more precise estimates of library accuracy; and (3) combining fingerprint profiles from BOX-A1R and (GTG)5 revealed the best discrimination (average rate of correct classification (ARCC) = 0.82 ± 0.06) in a two-way categorical split; while (4) no significant difference was found between the combined profiles (0.74 ± 0.15) and using solely BOX-A1R (0.76 ± 0.09) in a four-way split. Testing the library by identifying known isolates from a separate dataset has shown that a two-way classification performed better (ARCC = 0.66) than a four-way split (ARCC = 0.29). The library can be developed further by adding more representative isolates per host source. Nevertheless, our results have shown that combining profiles from BOX-A1R and (GTG)5 is recommended in developing the MST library for Laguna Lake. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|