Susan Libes

Erin Burge

Mike Trapp

Amanda Sturgeon

2014 Aquatic Microbiology

Conference 5/6/14

Building local capacity for microbial source tracking in the Myrtle Beach Urbanized Area

Shout out to Charles

• Learned the promise of multi-tracer approach in a joint effort (2009-2010) with Thomas & Hutton Engineering and Virginia Tech

• Provided impetus to grow our own local capacity for microbial source tracking

• Regulatory Drivers – 80 sites 303(d) listed for bacteria; 2 TMDLs

• Recreational impairments – Freshwater: E. coli – Saltwater: Enterococcus

• Shellfish impairments – National Shellfish Sanitation Program: Fecal Coliform

– NPDES Phase II stormwater program requirement • First permit cycle started in 2007

3

• Horry County

– Myrtle Beach (27,109)

– N. Myrtle Beach (13,754)

– Surfside Beach (3,837)

– Conway (17,103)

– Briarcliffe Acres (457)

– Atlantic Beach (334)

– Unincorporated (68,302*)

• Georgetown Co. (5,233*)

US Census of 2000*

Myrtle Beach Urbanized Area Small Separate Storm Sewer Systems (SMS-4’s)

Strategic Approach

Weight

of Evidence

Microbial Source Tracking

Intensive sampling using standard FIB measurements;

Infrared thermography

Sanitary Survey

Identification of impaired areas based on long-term monitoring

MST Protocol • Use existing best practices

• Watershed approach – Unified Subwatershed and Site Reconnaissance – SMS4’s stormwater infrastructure maps – Wet and dry sampling

• Fecal indicator bacteria – Include sediments

• qPCR assays • Chemical tracers

– Caffeine – Optical Brighteners – (Eutrophication/Hypoxia)

• Weight of evidence – Percentile-based rankings

Water level logger

Why Local Capacity for MST?

• Local knowledge of landscape – Help with site selection

• Historical memory of projects across landscape – Have better context for

data interpretation

• Logistics – Facilitates stormwater

sampling – Available for collaborative

meetings

• Engage students – Leverage research

qPCR approach

• Adapt from literature

• Local validation

– Matrix spiking into natural waters

– False positives and negatives

• Validated assays – GenBac3 (EPA 2010) – BacHum-UCD (Kildare et al. 2007) – GFD-Bird (Green et al. 2012) – BacCan-UCD (Kildare et al. 2007)

qPCR analyst: Amanda Sturgeon

Case Study:

Withers Swash

Largest drainage basin in Myrtle Beach

Drains 1/3 of CMB

Withers Swash: Enterococcus and E. coli concentrations

BacHum concentrations (genome copies/100 mL) in Withers Basin sub-watersheds. • Contamination rare

Homeless camps Sewer line break

• Downstream sewer- line break

5 of 72 samples had 77-760 copies/100 mL

quartiles

BacCan concentrations (genome copies/100 mL)

in Withers Basin sub-watersheds.

• Widespread

• Reinforced findings from sanitary surveys

• Coyotes

Weight-of-evidence approach • Percentile ranking by parameter

• Rankings assigned an index value and averaged by site

• Average values translated into a qualitative descriptor

– Minor; Significant; Strong; Very Strong

– Color-coded matrixes

Site qualitative descriptors: regulatory FIB and qPCR markers.

Very PC

Sites with “strong” and “very strong” levels of BacHum

Sites with “very strong” levels of BacCan

Percentages determined by dividing the average rank order by the maximum rank order for each tracer. Percentages greater than 50% are shaded red.

Corroboration with other tracers

Sites with strong evidence for human or canine sourced Bacteroides.

Dog Waste Outreach Campaign

Comparison of Common Non-Point Source Fecal Pollution to Recreational Waters: Regulatory Fecal Indicator Bacteria vs. Host Specific Genetic

qPCR markers in Avian and Canine Fecal Matter Aleksandar Dimkovikj1 and J. Michael Trapp2

1Department of Marine Science, Coastal Carolina University, P.O. Box 261954, Conway, South Carolina 29528-6054 2Burroughs & Chapin Center for Marine and Wetland Studies, Coastal Carolina University, P.O. Box 261954, Conway, South Carolina 29528-6054

Results

Summary

Abstract

Methodology

Introduction

Fecal indicator bacteria (FIB) are an important form of pollution in inland and coastal recreational waters and are of particular concern to human health. In this study, we conduct a comparison of regulatory FIB (E. coli) to host-specific qPCR assays on direct canine and sea bird fecal grab samples from the Grand Strand of South Carolina. Results suggest that inter-specimen variability makes interpretation of qPCR results difficult to attribute a percentage of the FIB load to a particular host. Temporal variability of the addition of waste to the system further complicates interpretation.

Water quality impairments are commonly associated with elevated concentrations of fecal indicator bacteria (FIB). Microbial source tracking (MST) aims to identify the sources of FIB pollution so targeted remediation strategies can be used to improve water quality. Bacterial culture-based methods dependant on active substrate metabolism are typically used to quantify FIB and can offer some geographic source information. Results from these methods, however, do not provide information on the source of the pollution. Molecular techniques, such as polymerase chain reaction (PCR), offer a quick and sensitive approach for quantifying FIB concentrations and host specific quantification by targeting genetic markers in the bacteria unique to the host organism. In this study, dog and bird fecal grab samples were examined for regulatory E. coli (Colilert®-18), a general FIB qPCR assay (GenBac) and host specific qPCR assays for dog and avian sources (BacCan and Bird GFD) assays. In a second experiment dog and bird fecal matter was aged and sampled over time to compare how the ratio between E. coli and genetic markers change temporally.

Fecal samples collected from selected canines were diluted to 10-5 mL final dilution factor, while selected sea bird (Laridae) samples were combined and diluted to 10-6 mL final dilution factor. Collected samples from an Italian Greyhound and bird mass-fecal sample were incubated (22◦C for ~ 14 days) and dilution and filtering protocols were repeated. Bacterial DNA was extracted from filters by mechanical and chemical cell lysis, and quantitative PCR was used to quantify the presence of FIB with the GenBac Assay, the BacCan Assay and the Bird GFD Assay. Additionally culturable E. Coli was measured by Colilert®-18.

Figure 1. A: Homogenized canine (n = 12) and sea bird (n = 25) fecal samples were diluted (1X PBS) and vacuum filtered in clean environment. B: Subsamples were analyzed with Colilert®-18 for E. coli. C: DNA was extracted from filters using the UltraClean® Soil DNA Isolation Kit Method. D: Quantitative Polymerase Chain Reaction (qPCR) was conducted by TacMan with Strategene®

Figure 2. Flow chart indicating how canine (top) and sea bird (bottom) fecal samples

were diluted with Phosphate Buffered Saline (1X PBS) and analyzed for qPCR and FIB

concentrations.

Concentrations of FIB and species specific genetic markers varied greatly between individuals. Surprisingly some individuals showed no E. coli in fresh feces but had relatively strong genetic marker results.

A weak correlation existed across individually sampled canines between E. coli and BacCan (R2 = 0.0366; p = 0.0187), and E. coli and GenBac (R2 = 0.0366; p = 1.97 X 10-6) determined FIB concentrations.

Genetic marker decay rates for aged canine fecal sample was rapid and logarithmic in slope. Corresponding E. coli concentrations progressively increased with time to maximum quantifiable level.

Decay rate for aged sea bird fecal sample was very rapid while E. coli persisted longer with high concentrations.

There is a great deal of variability in the concentration of FIBs and genetic markers between individuals.

The source specific signal seems to disappear quickly while FIB concentrations appear to continue to rise after leaving the organism. Thus any detection of a source specific signal should be considered significant .

This variability between bacterial concentrations in fecal samples limits interpretation of qPCR and Colilert results which complicates the assignment of FIB percent load to a particular host.

These results suggest that a multiple tracer weight of evidence approach including traditional WQ measurements and qPCR methods are necessary for meaningful data interpretation.

Donsel, D., J., V., Geldreich, E., E., Clarke, N., A. 1967. Seasonal Variations in Survival of Indicator Bacteria in Soil and Their Contribution to Storm-water Pollution. Appl. Environ. Microbiol. 15, 1362-1370.

Krometis, A., L., Noble, R., T., Characklis, G., W., Blackwood, A., D., Sobsey M., D. 2013. Assessment of E. coli partitioning behavior via both culture-based and qPCR methods. Water Sci. & Technol. 68.6, 1359-1369.

Oliver, J. D. 2000 The public health significance of viable but nonculturable bacteria. In: Nonculturable Microorganisms in the Environment (R. R. Colwell & D. J. Grimes, eds). ASM Press, Washington, DC, pp. 277–300.

USEPA 2012 National Section 303(d) List Fact Sheet.http://oaspub.epa.gov/waters/national_rept.control (accessed December 2013)

Bibliography

Figure 4. Individual canine samples exhibit high inter-species and applied method variability

for FIB.

Figure 5.

Weak

correlation

between

culture based

and qPCR

determined

bacterial

concentration

s for canine

fecal

samples.

Figure 7.

Genetic

marker

decay rates

(qPCR) for

sea bird

fecal matter

were highly

accelerated

as compared

to E. Coli

which

persisted

longer with

high

concentratio

ns.

Figure 6.

Genetic

marker

decay rates

determined

with qPCR

were

accelerated

and

exponential

in slope (R2

= 0.991) as

compared to

E. Coli for

canine fecal

matter.

Contact information: Michael Trapp (jtrapp@coastal.edu); Aleksandar Dimkovikj (adimkov@coastal.edu)

A

C

D

B

Wood, J., J.M. Trapp, S.M. Libes, and E.J. Burge. 2013. Watershed Assessment Report: Stormwater Management Planning: Development of a Pilot Investigative Approach to Remediate Bacterial Source Impairments along the Grand Strand. Final Report, prepared under the authority of Section 22 of the Water Resources Development Act of 1974 for the US Army Corps of Engineers, Charleston District; Horry County, SC; Georgetown County, SC; City of Myrtle Beach, SC; and City of North Myrtle Beach, SC. 100 pgs + Appendices. http://www.coastal.edu/envsci/projects/pollution/documents.html

FLOW CHART FOR MST TRACKING IN SWASHES OF THE GRAND STRAND Phase I: Identification of sites with FIB exceeding state WQS

FLOW CHART FOR MST TRACKING IN SWASHES OF THE GRAND STRAND Phase II: MST to identify geographic locations of sources and major host animal sources

qPCR challenges

• Need more assays – Urbanized (nuisance) wildlife

– Cats

• Access to calibrating standards

• Couldn’t quantify each assay’s % contribution to Bacteroides population

• Fate of tracer in the wild – Differential persistence

– Inoculation of soils and sediments

Questions?