Introduction

The beach water at the mouth of San Juan Creek and the portion of the creek immediately upstream of the beach fails State bacteriological standards regularly. While there are suspected sources of bacterial contamination, no comprehensive study has been carried out to address this problem. In addition, no bacteriologic survey has been carried out for the remainder of the watershed so there is no information available on potential sources of pollution. This study has three purposes:

1) Provide a bacterial survey of the water quality of the San Juan Creek watershed in dry weather conditions and locate any areas with bacteriologic water quality problems. Survey problem areas in detail.

2) Determine the source of the bacteria found in the problem areas using bacterial source tracking.

3) Compare two different techniques of bacterial source tracking, Ribotyping and Antimicrobial Resistance Testing (ARA) to determine the accuracy of these techniques.

This is the majority of the work required by Agreement 9-182-190-0 between the State Water Resources Control Board and Orange County PFRD.

Study Design

Phase 1: Bacteriological Survey of Watershed and Adjacent Beach Recreational Water

Phase 1 is designed to quickly survey the watershed to locate areas with bacterial pollution problems. Sites will be selected to sample tributaries, storm drains and to sample known or suspected problem areas and recreational water at the mouth of the creek. Approximately 30 sites will be sampled weekly for 10 weeks (300 samples).

Bacteriological testing will include total coliform and fecal coliform by MTF or membrane filtration and enterococcus by Enterolert or membrane filtration.

PFRD, environmental health and public health laboratory staff will determine the sampling sites with input from the technical advisory committee. PFRD will provide necessary access information for sampling sites. The Public Health Laboratory will do the sampling, testing and analysis.

Phase 2: Detailed Bacteriological Survey of Identified Problem Areas

Phase 2 is designed to further characterize locations with water quality problems identified in Phase 1. These sites will be tested in detail to determine maximum and minimum bacterial counts as well as temporal and geographic boundaries of problem. Existing data indicates that one known problem area is the very lower end of the creek, which is often prevented from flowing into the ocean by a sand berm. Samples will be taken from at least one known problem area and a second area geographically removed. E. coli and enterococcus isolates from a sub-set of samples will be saved to use in Phase 3.

It is expected that approximately 200 samples will be collected over 10 weeks. From a sub-set of these, 5 isolates of E. coli and enterococcus per sample for each bacterium will be saved for phase 3 analyses.

The public health laboratory will perform sampling and testing. Sampling strategy will be determined by public health laboratory staff and with input from PFRD, environmental health and the technical advisory committee.

Phase 3: Source Identification by ARA and Ribotyping: Library Preparation and Technique Accuracy Determination

Phase 3 is designed to build the necessary databases and determine the accuracy of two methods of bacterial source tracking, ARA (Antimicrobial Resistance Analysis) and Ribotyping. Overall, this will be done by constructing a library of bacterial isolates from known species, performing the source analysis testing to build or add to an existing database and determining the accuracy of the methods utilizing bacterial strains from known sources not included in the database.

ARA Technique:

ARA differentiates bacteria from different species of animals by comparison of their resistance to antibiotics. Fecal samples from known animal species and human sources are taken, E. coli and enterococcus bacteria are isolated and a collection of isolates is constructed. Each bacterial isolate is tested against 4 concentrations of 8 different antibiotics. Resistance patterns by species analyzed by discriminate analysis are used to classify individual strains into the most likely group (species).

Several recent published studies done by several researchers on different watersheds have shown the utility of the ARA technique and have documented the accuracy as being between 75% and 85% (Wiggins, Hagedorn, Parveen, Harwood). In this study, dual analysis of both E. coli and enterococcus isolates is expected to improve the accuracy somewhat. In addition, isolates will be frozen so they can be re-analyzed by other techniques in the future. The technology is fairly simple resulting in lower cost per isolate and an increased probability that the technique can be rapidly imported into the Public Health Laboratory for future studies if it proves to be an accurate technique. This technique produces matches based on the local database of known isolates. This requires a large database be constructed. However, the database can be utilized for future bacterial pollution problems.

The initial part of this technique is to construct a library of strains from known species. Constructing the library will require 500 strains of each bacterium from human sources and from each animal species that may be a major contributor. This will be done by collecting 100 samples per species and picking 5 isolates from each sample for each bacterium. This will result in 2500 E. coli strains and 2500 Enterococcus strains included in the database if humans and 4 species of animals are considered potential sources.

Quality assurance is carried out first by determining the initial accuracy of the technique by holdout analysis, a statistical study of the database. Additional quality assurance is covered under the Quality Assurance section below.

Samples will be collected by public health laboratory staff and any other organizations that can assist (animal control, sewage treatment plant personnel, volunteers, humane societies). Bacteria will be isolated and identified and frozen by the public health laboratory staff. Valerie Harwood, Ph.D., Assistant professor, University of South Florida, and Tampa, FL will subcontract with the public health laboratory to perform the ARA analysis.

Ribotyping Technique:

Ribotyping differentiates bacteria by detecting changes in restriction enzyme cutting sites in the bacterial genome. Restriction fragments are separated by size on an agarose gel and the fragments containing r-RNA (ribosomal RNA) gene sequences are detected using a complimentary probe. The patterns of different sized fragments are compared between isolates. Isolates with the same patterns are considered related.

Similar to ARA, a library of E. coli strains from known animal species and human sources is utilized to identify the source of strains isolated from problem areas. Ribotype patterns, analyzed by discriminant analysis are used to classify strains into the most likely group (species). Unlike ARA, large isolate libraries, maintained by the testing laboratory have been traditionally utilized for matching along with the local library. However, a local isolate library is necessary to identify up to one third of strains. The accuracy of a geographically separated database is not established at this time. In the one published study utilizing ribotyping, an accuracy of 82% was achieved (Parveen 1999).

To construct the local library will require 100 strains of E. coli from human sources and from each animal species that may be a major contributor. This will be done by collecting 100 samples per species and picking 1 isolate from each sample. This will be a subset of the 5 samples/specimen taken for ARA analysis. This will result in 500 strains for analysis if 4 animal species and humans are considered potential sources. This library will be utilized with the existing library by the testing laboratory to determine sources.

The samples that will be utilized are the same as ARA analysis. Bacteria will be isolated, identified and frozen by the public health laboratory staff. The ribotyping technique published by S. Parveen of the University of Florida (Parveen, 1999) will be utilized for analysis. The testing laboratory will be George Lukasik, Ph.D., Biological Consulting Services of North Florida, Gainsville, Florida or another acceptable laboratory.

Quality assurance:

Two types of quality assurance tests will be carried out:

1) Reproducibility. 20 strains of each bacterium will be submitted to the testing laboratory 3 different times blinded. The percentage of reproducibility will be determined for each strain and summed.

2) Accuracy. Final accuracy of each technique is determined by an analysis of 100 isolates from known species (which are not residing in the database) that are submitted to the reference laboratory blinded. These bacterial strains will be isolated from collected along side the data base samples and held back. The analysis of rate of correct classification will be presented in a table similar to tables 1 and 2 of Harwood et. al. 2000. After this analysis adjustments can be made in the analysis technique to maximize accuracy if necessary. The same strains will be utilized for both typing techniques so a direct comparison can be made.

Phase 4: Source Identification by ARA and Ribotyping: Source Identification of Watershed Isolates.

Phase 4 is the analysis of bacterial isolates from problem areas that have been collected in phases 2 and 3 utilizing the two source tracking techniques and a comparison of the results of the techniques. Approximately 200 bacterial isolates (per species) from the problem areas will be provided to the two reference laboratories and will be compared to the library of isolates from known species to determine potential source. For ARA, a comparison is made between the results for the two different bacterial databases (E. coli and enterococcus) to determine a final result. For ribotyping the analysis will be done comparing the unknown E. coli strains to the local database and to the testing laboratory's database. The ARA results will be compared to that of the ribotyping analysis and written report issued. The report is also prepared for publication in a peer-reviewed journal.

Public health laboratory staff will do the sampling and bacterial isolation and identification. Dr. Harwood, Dr. Lukasik and Dr. Moore will write the final report.

Schedule

Phase 1: April 16 - June 22

Phase 2: June 22 - August 10

Phase 3: June 22 - November 16

Phase 4: November 16 - May 15, 2002

Budget

See spreadsheet

References

Hagedorn, C., S. L. Robinson, J. R. Filtz, S. M. Grubbs, T. A. Angier, and R. B. Beneau. 1999. Determining sources of fecal pollution in a rural Virginia watershed with antibiotic resistance patterns in fecal streptococci. Applied and Environmental Microbiology. 65(12):5522 - 5531.

Harwood, V.J., J. Whitlock and V. Washington. 2000. Classification of antibiotic resistance patterns of indicator bacteria by discriminant analysis: use in predicting the source of fecal contamination in subtropical waters. Appl Environ Microbiol. 66:3698-3704.

Parveen, S., R. L. Murphree, L. Edmiston, C. W. Kaspar, K. M. Portier, and M. L. Tamplin. 1997. Association of multiple-antibiotic-resistance profiles with point and non-point sources of Escherichia coli in Apalachicola Bay. Appl. Environ. Microbiol. 63(7):2607-12.

Parveen, S., K. M. Portier, K. Robinson, L. Edmiston and M. L. Tamplin. 1999. Discriminant Analysis of Ribotype Profiles of E. coli for Differentiating Human and Non-Human Sources of Fecal Pollution", Appl Environ Microbiol. 65:3142-3147.

Wiggins, B. A., R. W. Andrews, R. A. Conway, C. L. Corr, E. J. Dobratz, D. P. Dougherty, J. R. Eppard, S. R. Knupp, M. C. Limjoco, J. M. Mettenburg, J. M. Rinehardt, J. Sonsino, R. L. Torrijos, and M. E. Zimmerman. 1999. Use of antibiotic resistance analysis to identify nonpoint sources of fecal pollution. Appl Environ Microbiol. 65:3483-6.