ube number i.e. NB-AM080911-01. One tube containing a morning sample filter and one containing an afternoon sample filter, along with their respective controls, underwent DNA extraction using crude bead-beating as described in Lavender and Kinzelman (2009). Briefly, DNA extraction buffer was prepared by dissolving salmon testes DNA (#D-1626, Sigma, St. Louis, MO) in AE Buffer (#19077, Qiagen, Hilden, Germany) to a working concentration of approximately 10 µm/mL confirmed by reading at A260. This working solution was diluted to make 0.2 µm/mL salmon DNA/extraction buffer with AE buffer. The introduction of non-target DNA (salmon testes DNA, aka specimen processing control or SPC) to both the calibrator and experimental samples normalizes the relative recovery of DNA through comparison of recovered reference DNA which was equally added to both samples. After preparation of the extraction buffer, 590 µm of 0.2 µm/mL salmon DNA extraction buffer was added to the micro-centrifuge tube containing the filter, bead beaten for two minutes at the maximum speed (BioSpec Mini-BeadBeater-8, Bartlesville, OK), and then centrifuged at 12,000 rpm for one minute. Once completed, 250 µL of supernatant was transferred to a 1.7 mL low-retention micro-centrifuge tube (#C-3228-1, GeneMate, Kaysville, UT) and centrifuged again for 5 minutes. The final supernatant was placed into a new micro-centrifuge tube, labeled as 1x DNA extract and stored at 4˚C until the onset of qPCR analysis. Under certain environmental conditions, such as heavy rainfall or turbidity greater than 20 NTU, the 1x DNA extract was diluted 1:5 with AE buffer prior to analysis.

Quantification of E. coli using qPCR. Replicate aliquots of 5 μL of extracted sample/calibrator/control DNA was added to 20 μL of PCR mixture containing lyophilized reagents: Omnimix HS beads (2 beads/4 reactions) (Cepheid; Sunnyvale, CA), and E. coli (target) or Salmon testes DNA (SPC control) Smartbeads™ (1 bead/4 reactions) (BioGx; Birmingham, AL). Each run also included quality control samples: 1) a no template control (NTC); 2) a negative extraction control (NEC); and 3) a calibrator (CAL). These single samples were prepared by combining 20 µL PCR master mix with 1) 5 µL AE buffer; 2) 5 µL negative extraction control DNA extract (sterile PBS carried through the DNA extraction phase); and 3) 5 µL laboratory-prepared whole cell calibrator DNA extract (10 µL of E. coli at a concentration of 3.0 × 109 cells/mL spotted onto a blank filter and extracted) respectively. Quantification was performed on the SmartCycler II (Sunnyvale, CA) using the following cycling parameters: hold at 95˚C for 120 seconds (optics off) (Stage 1), followed by 45 cycles of 95˚C for 5 seconds (optics off) (Stage 2), and 62˚C for 43 seconds (optics on).

Statistical analysis. The SmartCycler software automatically calculates cycle threshold (CT) values for each sample by the second derivative method. Replicate DNA extracts not agreeing within one CT were reanalyzed prior to performing further calculations. Once agreement was achieved, replicate CT values were averaged and calibrator cell equivalent (CCE) values determined using the ddCT method [8]. If the difference between the sample and SPC CT was greater than 3.0 the sample was considered inhibited and was excluded from the dataset.

Statistical analysis of data (ANOVA and correlation) was performed using WINKS SDA 6.0 Software (Texasoft, Cedar Hill, TX). Values for viable cell counts (CFU or MPN) and CCE were log-normalized prior to analysis. A test for equality of variance indicated the necessity of using a two-sample t-test. Statistical decisions were made at p = 0.05 unless otherwise stated.

3. Results

A survey of morning (0700) and afternoon (1200) surface water samples collected from North Beach (Racine, WI) was conducted in order to determine if diurnal variation existed prior to implementation of qPCR assays.

Culture-based assays. Two culture-based assays were used, Colilert/Colilert-18® and US EPA Method 1603 (modified m-TEC agar, membrane filtration). There was no significant difference in mean E. coli values between the two culture-based methods of enumeration; US EPA Method 1603 and IDEXX Colilert/Colilert-18®, for either the AM (0700) or PM (1200) sampling period (pAM = 0.539, pPM = 0.245) (Tables 1 and 2). The mean values of 0700 and 1200 E. coli values as determined by the IDEXX methods did not differ significantly [ANOVA, t (56) = 0.7, p = 0.486] suggesting diurnal variation, on average, was not present at the study site (Figure 1(a)). Samples enumerated using modified m-TEC agar (US EPA Method 1603) also demonstrated no significant difference in E. coli concentrations in morning versus afternoon samples (p = 0.959). However, there were four instances out of 22 sample collection dates (18%) where the cfu/100 mL differed between AM and PM samples; three occasions in which the AM sample had higher counts than the afternoon and one where the PM sample had the higher count.

Table 1. ANOVA: Single factor, AM.

Table 2. ANOVA: Single factor, PM.


Figure 1. Comparison of log converted E. coli densities from early morning (0700) vs. afternoon (1200) sample collection by Colilert/Colilert-18® (a) and qPCR (b).

qPCR assay. A test for equality of variance indicated that the variances of the two groups (0700 and 1200 sample collection times) were significantly different (Table 3). Because the variances of the two groups were significantly different (1.469 AM vs. 4.343 PM), a twosample t-test was performed. This analysis indicated that the means of the two groups were not significantly different (2.418 AM and 3.171 PM) and no significant diurnal variation was present (unequal variances t-test, t (35.9) = 1.72; p = 0.094) (Figure 1(b), Table 3).

Regulatory agreement. During the course of this study there were only two out of 22 instances where there was regulatory decision disagreement between culture-based assays (Colilert/Colilert-18® or US EPA Method 1603) and qPCR (data not shown). Although significant diurnal variation was not observed, differences in daily regulatory action, based on the time of sample collection (0700 or 1200) were noted on several occasions (Figure 2). There were five of 22 (22%) instances where 1200 (PM) CCE/100 mL values exceeded their 0700 counterparts when qPCR was employed as the analytical assay. Of these, there were two instances (10%) when the culturebased assays did not reflect the same relationship. Sanitary survey data (ambient environmental conditions, bather density, presence of wildlife, etc.) indicated that changes in weather frequently occurred between AM and PM sample collection events, e.g. measurable precipitation and/or increases in wave height leading to increased

Table 3. ANOVA: Single factor, qPCR.

Figure 2. E. coli CCE/100 mL, AM (dark gray) vs. PM (light gray), North Beach (2011).

turbidity. These differences could explain the variability in E. coli densities as enumerated by the culture-based and qPCR assays. Culture-based assays detect only cells possessing the ability to grow on selective media under optimal conditions, whereas qPCR detects these cells in addition to: viable but not culturable (VBNC), DNA contributed from dead cells, and free environmental DNA.

4. Discussion

The determination of diurnal variation in surface waters is essential for both beach management and assurance of public safety in the context of recreational water exposure. The most efficient method vetted to date for the determination of same day water quality management decisions is qPCR. Development of qPCR assays as effective and reliable methods for the determination of FIB, with demonstrable relationships to human health effects, has been presented [8,24]. These advancements will allow for the replacement of slower, culture-based methods. Due to the rapid nature of qPCR assays, the time from sample collection to public notification can be reduced from 18 to 24 hours down to less than two hours [25]. In order to facilitate the generation of results prior to the opening of public bathing beaches, sample collection would need to occur in the early morning hours. Previous studies, employing culture-based assays, have demonstrated significant diurnal variation at Great Lakes beaches, including the study site, where samples collected in the early morning would have resulted in more frequent water quality advisories [1,2]. In these instances, UV light may play a role in the decrease of FIB concentrations on sunny days as night gives way to day. In contrast, the US EPA NEEAR epidemiological studies, performed at sewageimpacted beaches, indicated that the qPCR signal remained flat through the day [26,27]. Therefore, when considering the implementation of qPCR on non-sewage impacted beaches, it is important to determine if time of day would introduce bias with respect to the posting of water quality advisories. While the protection of public should be of primary concern, public officials would be reluctant to implement qPCR if the necessity of early morning sample collection resulted in an increase in Type I errors, i.e. posting additional advisories in the absence of credible human health risk elevations.

This study indicated that significant diurnal variation was not detected at North Beach during the summer of 2011. However, different management decisions would have been made based on the time of sample collection on five instances; two of which when there existed a disparity between culture-based assays and qPCR. Notwithstanding, there were only two instances out of 22 paired sampling events (10%) in which the use of qPCR would have resulted in a different beach management decision.

Changes in ambient conditions were one possible explanation for these differences and the use of sanitary surveys may aid those considering the implementation of rapid molecular methods for the assessment of recreational surface waters (by predicting when changes to the aquatic environment occur). When changes in ambient water quality are likely to result in an increased risk to public health, the use of qPCR assays, by virtue of their rapid turnaround time (approximately three hours), will allow beach managers to re-assess the beach environment within a single day. Combined with other non-analytical rapid estimations of FIB, such as predictive models, future decisions concerning management and public safety can be made more quickly and economically, i.e. model derived estimations of FIB elevations may be confirmed in near real-time via qPCR [28,29].

In areas where significant diurnal variation is noted in conjunction with consistently low bather densities, the use of early AM E. coli analyses as determined by qPCR, may not be the best beach management choice. In these instances, sampling during periods of peak bather density (typically during the afternoon hours) and employing culture-based assays such as Colilert® or predictive models may be the best management decision for the protection of public health. Future work should continue at this, and other sewage and non-sewage impacted coastal beaches, to determine whether or not diurnal variation impacts the implementation and effectiveness of rapid molecular analytical methods such as qPCR for the determination of recreational water quality.

5. Conclusions

An assessment of diurnal variation was conducted during the summer of 2011 at North Beach (Racine, WI) to determine implications, if any, surrounding the necessity of early morning sample collection as part of qPCR implementation. Through the analysis of study results it was demonstrated that:

• No significant diurnal variation was present at North Beach in Racine, WI by either culture-based or qPCR assays;

• Early morning samples were representative of daily water quality in the absence of significant changes in ambient environmental conditions as determined by the routine on-site sanitary survey data;

• Employing a single, rapid analytical method would contain costs as well as ensure public safety, however the use of models in conjunction with rapid lab-based methods may provide an additional line of evidence when making beach management decisions.

6. Acknowledgements

Research and monitoring was funded by US EPA contract #EP115000072, Federal BEACH Act dollars administered by the WI Department of Natural Resources, and the city of Racine, WI. The authors wish to thank Joseph Granite and Jennifer Creekmur for their assistance with sample collection and processing. Finally this opportunity would not have been possible without the assistance of the McNair Scholars and TRiO Programs at Marian University (Fond du Lac, WI).


  1. R. Whitman and M. Nevers, “Escherichia coli Sampling Reliability at a Frequently Closed Chicago Beach: Monitoring and Management Implications,” Environmental Science & Technology, Vol. 38, No. 16, 2004, pp. 4241- 4246. doi:10.1021/es034978i
  2. J. Kinzelman, “Investigating Bathing Water Quality Failures and Initiating Remediation for the Protection of Public Health,” Ph.D. Dissertation, University of Surrey, Guildford, 2005.
  3. National Technical Advisory Committee (NTAC), “Water Quality Criteria, Washington, DC,” Federal Water Pollution Control Administration, 1968.
  4. United States Environmental Protection Agency (US EPA), “Clean Water Act,” United States Environmental Protection Agency, Washington DC, 1977.
  5. US EPA, “Ambient Water Quality Criteria for Bacteria, Office of Water Regulation and Standards, Criteria and Standards Division,” United States Environmental Protection Agency, Washington DC, 1986.
  6. US EPA, “Beaches Environmental Assessment and Coastal Health (BEACH) Act of 2000,” United States Environmental Protection Agency, Washington DC, 2000.
  7. National Resources Defense Council (NRDC), “Testing the Waters: A Guide to Water Quality at Vacation Beaches,” 2011. http://www.nrdc.org/water/oceans/ttw/titinx.asp
  8. R. Haugland, S. Siefring, L. Wymer, K. Brenner and A. Dufour, “Comparison of Enterococcus Measurements in Freshwater at Two Recreational Beaches by Quantitative Polymerase Chain Reaction and Membrane Culture Analysis,” Water Research, Vol. 39, 2005, pp. 559-568. doi:10.1016/j.watres.2004.11.011
  9. United States House of Representatives, “H.R. 2537, Beach Protection Act of 2007,” 110th United States Congress, Washington DC, 2007, pp. 1-20. http://www.gpo.gov/fdsys/pkg/CRPT-110hrpt491/pdf/CRPT-110hrpt491.pdf
  10. United States House of Representatives, “H.R. 2093, Clean Coastal Environment and Public Health Act of 2009,” 111th United States Congress, Washington DC, 2009, pp. 1-16. http://www.gpo.gov/fdsys/pkg/BILLS-111hr2093rh/pdf/BILLS-111hr2093rh.pdf
  11. J. Griffith, D. Moore, C. McGee and S. Weisberg, “Technical Report 506,” Southern California Coastal Water Research Project, Costa Mesa, 2007. ftp://ftp.sccwrp.org/pub/download/DOCUMENTS/TechnicalReports/506_beta_testing.pdf
  12. S. Siefring, M. Varma, E. Atikovic, L. Wymer and R. Haugland, “Improved Real-Time PCR Assays for the Detection of Fecal Indicator Bacteria in Surface Waters with Different Instrument and Reagent Systems,” Journal of Water and Health, Vol. 6, 2008, pp. 225-237. doi:10.2166/wh.2008.022
  13. R. Bushon, C. Likirdopulos and A. Brady, “Comparison of Immunomagnetic Separation/Adenosine Triphosphate Rapid Method to Traditional Culture-Based Method for E. coli and Enterococci Enumeration in Wastewater,” Water Research, Vol. 43, No. 19, 2009, pp. 4940-4946. doi:10.1016/j.watres.2009.06.047
  14. J. Lavender and J. Kinzelman, “A Cross Comparison of qPCR to Agar-based or Defined Substrate Test Methods for the Determination of Escherichia coli and Enterococci in Municipal Water Quality Monitoring Programs,” Water Research, Vol. 43, No. 19, 2009, pp. 4967-4979. doi:10.1016/j.watres.2009.08.010
  15. J. Kinzelman, R. Bushon, S. Dorevitch and R. T. Noble, “Comparative Evaluation of Molecular and Culture Methods for Fecal Indicator Bacteria for Use in Inland Recreational Waters,” Water Environment Research Foundation, IWA Publishing, London, 2011, 360 p.
  16. US EPA, “Recreational Water Quality,” United States Environmental Protection Agency, Washington DC, 2011.
  17. J. Kinzelman, C. Ng, E. Jackson, S. Gradus and R. Bagley, “Enterococci as Indicators of Lake Michigan Recreational Water Quality: Comparison of Two Methodologies and Their Impacts on Public Health Regulatory Events,” Applied and Environmental Microbiology, Vol. 69, No. 1, 2003, pp. 92-96. doi:10.1128/AEM.69.1.92-96.2003
  18. R. Converse, J. Griffith, R. T. Noble, R. Haugland, K. Schiff and S. Weisberg, “Correlation between Quantitative PCR and Culture-Based Methods for Measuring Enterococcus spp. over Various Temporal Scales at Three California Marine Beaches,” Applied and Environmental Microbiology, Vol. 78, No. 4, 2012, pp. 1237-1242. doi:10.1128/AEM.07136-11
  19. M. Gregory and E. Frick, “Indicator Bacteria Concentrations in Streams of the Chattahoochee River National Recreation Area, March 1999-April 2000,” Proceedings 2001 GWRC Conference, Athens, 26-27 March 2001, pp. 510-513.
  20. J. Kinzelman and S. McLellan, “Success of ScienceBased Best Management Practices in Reducing Swimming Bans—A Case Study from Racine, Wisconsin, USA,” Aquatic Ecosystem Health Manage, Vol. 12, No. 2, 2009, pp. 187-196. doi:10.1080/14634980902907466
  21. Wisconsin Department of Natural Resources (WI DNR), “Beach Monitoring Program Requirements,” 2011. http://dnr.wi.gov/org/water/wm/wqs/beaches/BeachMonitoringRequirements.pdf
  22. J. Kinzelman, A. Dufour, L. Wymer, G. Rees, K. Pond and R. Bagley, “Comparison of Multiple Point and Composite Sampling for Monitoring Bathing Water Quality,” Lake and Reservoir Management, Vol. 22, No. 2, 2006, pp. 95- 102. doi:10.1080/07438140609353887
  23. US EPA, “Method 1603: Escherichia coli (E. coli) in Water by Membrane Filtration Using Modified Membrane-Thermotolerant Escherichia coli Agar (Modified mTEC),” US EPA Office of Water, Washington DC, 2002.
  24. A. Blackwood, S. Yu, J. Gregory and R. T. Noble, “Rapid qPCR Assays for Escherichia coli and Enterococcus in Recreational Waters: Equivalent to Existing Methods?” Proceedings ASM National Meeting, Orlando, 21-26 May 2006. http://ieg.ou.edu/ASM2006/data/papers/Q_494.htm
  25. R. T. Noble, A. Blackwood, J. Griffith, C. McGee and S. Weisberg, “Comparison of Rapid Quantitative PCR-Based and Conventional Culture-Based Methods for Enumeration of Enterococcus spp. and Escherichia coli in Recreational Waters,” Applied and Environmental Microbiology, Vol. 76, No. 22, 2010, pp. 7437-7443. doi:10.1128/AEM.00651-10
  26. T. Wade, R. Calderon, E. Sams, M. Beach, K. Brenner, A. Williams and A. Dufour, “Rapidly Measured Indicators of Recreational Water Quality are Predictive of Swimming-Associated Gastrointestinal Illness,” Environmental Health Perspectives, Vol. 114, No. 1, 2006, pp. 24-28. doi:10.1289/ehp.8273
  27. T. Wade, et al., “Rapidly Measured Indicators of Recreational Water Quality and Swimming-Associated Illness at Marine Beaches: A Prospective Cohort Study,” Environmental Health, Vol. 9, No. 66, 2010.
  28. A. Mednick and D. Watermolen, “Beach Pathogen Forecasting Tools: Pilot Testing, Outreach, and Technical Assistance,” Wisconsin Department of Natural Resources, Bureau of Science Services, Miscellaneous Publication, 2009.
  29. J. Telech, K. Brenner, R. Haugland, E. Sams, A. Dufour, L. Wymer and T. J. Wade, “Modeling Enterococcus Densities Measured by Quantitative Polymerase Chain Reaction and Membrane Filtration Using Environmental Conditions at Four Great Lakes Beaches,” Water Research, Vol. 43, No. 19, 2009, pp. 4947-4955. doi:10.1016/j.watres.2009.07.002

Journal Menu >>