CICEET Progress Report for the period 9/02/07 Through 3/01/08
Project Title: Assessment of microbial contaminants pertinent to swimmingrelated
illnesses at Doheny Beach, Orange County, CA
Principal Investigator(s): Rachel T. Noble
Additional Investigator(s): Jed Fuhrman, Troy Scott, Jill Stewart (separate budget)
Project Start Date: May 1, 2007
Project Objectives for This Reporting Period
Objectives
The objectives are to analyze samples from the Doheny Beach
Epidemiological Study, which was abandoned in July, 2007, and also conduct the same
analyses for the City of Avalon (Catalina Island) Epidemiological Study. The analyses
funded by this grant are specific to the PI.
PI Noble: QPCR E. coli analysis, QPCR Bacteroides thetaiotamicron analysis
Co-PI Scott: ESP Gene analyses, quantitative and qualitative
Co-PI Fuhrman: Hepatitis A Virus analysis
Co-PI Stewart: Real time polymerase chain reaction (PCR) assays for (1) enteroviruses,
and (2) Methanobrevibacter smithii, an archean associated with human-source fecal
contamination.
Tasks to meet objectives
In summary, the tasks are to 1) participate in Performance Based Study at SCCWRP, 2)
conduct analyses of samples from Doheny Beach Epi study, 3) conduct analyses for Avalon
Epi Study, 4) conduct data analysis as individual researchers, 5) conduct data analysis as
multi-disciplinary team, 6) integrate analyses with health outcomes.
Progress on Tasks
Performance Based Study: Researchers at SCCWRP put together a panel of blind
samples and sent them to UNC Chapel Hill Institute of Marine Sciences (Noble), Univ.
Southern California (Fuhrman), Source Molecular, Inc. (Scott) and NOAA Charleston
(Stewart) to evaluate the novel methods proposed for the epidemiology studies. A total
of 36 samples were prepared, including negative controls (PBS), ambient water with no
fecal indicator bacteria (FIB), ambient water with low concentrations of FIB, sewage
spiked offshore water, sewage spiked beach water, water seeded with gull guano from a
rehab center, water seeded with gull guano from the beach, and water from the Tijuana
River. The methods tested at were Bacteroides thetaiotamicron QPCR (Noble), E. coli
QPCR (Noble), Hepatitis A Virus QRTPCR (Fuhrman), ESP Gene Quantitative and
Qualitative analysis (Scott) and single agar layer method for male-specific (F+) and
somatic (F-) coliphages, a duplex Luminex assay that simultaneously detects M. smithii
and Enterococcus, and monoplex qPCR assays for M. smithii, Enterovirus and Norovirus
by NOAA Charleston (Stewart). All tested methods successfully scored PBS samples as
negative (no false positives) and the Tijuana River water as positive (no false negatives).
For NOAA Charleston, methods differed in their sensitivity to detecting sewage spiked in
seawater, with coliphage and Luminex assays most effective and qPCR unable to detect
some of the lower concentrations. The data from the Performance Study was detailed in
the Progress Report of September 1, 2007.
Doheny Beach Epidemiology Study:
At Doheny Beach, 104 total water samples were collected from up to five locations, three
times a day, and one day a week from May 28 - June 23, 2007, and on three separate days
from June 30 - July 4, 2007. Rapid QPCR E. coli and Bacteroides thetaiotamicron
analysis for all of these samples is complete and the data has been submitted to SCCWRP
for integration with the epidemiological analyses. See raw data as presented in Tables 1
and 2. Hepatitis A Virus assays for all of these samples is complete and the data has been
submitted to SCCWRP for inclusion in the epidemiological analyses. ESP gene analysis
for all of these samples is complete and the data has been submitted to SCCWRP for
inclusion in the epidemiological data analyses. Rapid qPCR analysis of Enteroviruses
and Methanobrevibacteri smithii is completed and results were negative for all of the
Doheney Beach samples (data not shown). Results of additional coliphage analysis are
presented in table below.
Avalon, Catalina Island Epidemiology Study: At the City of Avalon, on Catalina Island, 325 total water samples were analyzed from four locations, up to three times a day, over four consecutive days, once a week from July 26 August 26, 2007. Additional samples were collected over five days from August 30 September 3, 2007 and over two consecutive days from September 8-9, 2007. ESP Gene analysis from all of these samples is complete. QPCR E. coli and Bacteroides thetaiotamicron analyses are complete. HAV, Enterovirus, coliphage and M. brevii analyses are also complete. During the Avalon study correlative relationships with traditional methods and the QPCR results were not apparent. However, the traditional methods have been identified as having up to a 40% false positive rate (Moore et al. data in preparation) so that finding is not surprising. Enteroviruses were detected in some of the Avalon samples (Table 4), with the majority of positives occurring from site B. Similar to the Doheny results, M. smithii was not detected by qPCR in any of the Avalon samples. Coliphages were also detected sporadically as detailed in tables below.
QPCR Assay and Interpretation:
For all QPCR assays, B. thetaiotamicron and E. coli concentrations were measured using
QPCR assays. OmniMix beads (including Taq, buffers, and dNTPs at appropriate
concentrations) were used for all analyses. B. thetaiotamicron primers and probe,
designed by Blackwood, used TaqMan technology and targeted a 110 base pair fragment
of 16S rRNA. For the E. coli assays, lyophilized beads with Scorpion primers and probe,
designed in collaboration by Cepheid and the Noble lab to target the uid a gene, were
used rather than wet chemistry. B. thetaiotamicron concentrations were quantified using
a standard curve; E. coli concentrations calculated by the delta-Ct method.
Additionally, a range of controls were used. Negative controls were used during all assays and all DNA extractions to ensure that samples were not cross-contaminated. For the both assays, an external standard curve was run with each individual run. A standard curve was generated using 3-4 data points: 10,000 cell/reaction; 1,000 cell/reaction; 100 cell/reaction; and 10 cell/reaction. This curve was used for the B. thetaiotamicron absolute quantification and to ensure high PCR efficiency for the E. coli runs. Only data from runs with ~90% efficiency or higher were included in this data set. Additionally, all filters were spiked with approximately 100,000 cells of L. lactis prior to extraction. QPCR assays measuring L. lactis were used as a control for extraction efficiency and PCR inhibition. QPCR data measuring less than 85,000 cells of L. lactis were assumed to be inhibited (whether in the extraction or PCR step). B. thetaiotamicron and E. coli assays were repeated for samples demonstrating inhibition using a range of dilutions, from 1:5 to 1:500, in order to dilute out the inhibitors. Less than 8% of samples had fewer than 85,000 cells of L. lactis. In fact, many samples seemed to have naturally occurring L. lactis populations, with much higher numbers than the spikes. In these cases, there may have been inhibition, but we were unable to detect it. For those samples, B. thetaiotamicron and E. coli concentrations calculated from the QPCR represent a conservative estimate.
One hundred percent of samples were analyzed. Samples from several dates following the move from Avalon arrived defrosted, but replacement filters from SCCWRP’s community filter collection were received as prompt replacements.
Thirty-six percent of samples taken from Doheny Beach contained B. thetaiotamicron, suggesting human fecal contamination. B. thetaiotamicron was found several times at A, B, D, and E sites but only once at a C site. The C site consistently had the highest E. coli concentrations though, which was expected since the C site was in a pond at the end of San Juan Creek and prime real estate for waterfowl.
Forty nine percent of samples taken from Avalon contained B. thetaiotamicron. On 7/26/07 to 7/29/07, 8/12/07 to 8/16/07, and 8/30/07 to 9/3/07, B. thetaiotamicron concentrations were high across all sites. For all other study dates, concentrations were extremely low across nearly all sites. E. coli concentrations were generally high across all sites and dates.
ESP Gene Assay and Interpretation:
For the ESP gene detection from the E. faecium specie, the qPCR data generated were
used to establish baseline ranges for marker and reference levels and to establish ratios to
which unknown samples can be compared. The samples were analyzed using
conventional PCR specific to the esp gene target, from enriched samples, and a further
quantitative analysis was conducted using a more recently developed esp gene primer
probe set for quantification and verification. See raw data in Table 3 below. Quantitative
results were reported as ranges with regard to the portion of Enterococci that were
deemed to be of human origin. These ranges were estimated based on internal laboratory
reference Tables comprised of raw sewage influent samples collected from throughout
the US and Canada (Scott et al. accepted for publication). Overall, the Doheny Beach
samples were predominantly negative whereas the Avalon Beach samples showed
intermittent spikes in what appeared to be wastewater plumes. All positive samples
contained some degree of background bacteria that were not determined to be of human
origin.
Hepatitis A virus detection and quantification (Fuhrman laboratory, USC):
Hepatitis A detection was done by a TaqMan-based QRT-PCR, using the SH-Poly
primers and probe developed by Houde et al (2007). The TaqMan real-time PCR assay
was completed in 25µl total reaction volume with 23µl master mix and 2µl of extracted
RNA. The 1-step Brilliant QRT-PCR core reagent kit (Stratagene, La Jolla, CA, USA
#600532) was used for the master mix. Primers were synthesized by Operon
Biotechnologies Inc (Huntsville, AL, USA) and the SH-Poly-Q probe by Biosearch
Technologies (Novato, CA, USA). The reaction contained 5.5mM MgCl2, 300nM
forward primer SH-Poly A, 400nM reverse primer SH-Poly-1, and 200nM TaqMan probe
SH-Poly-Q. The remaining components of the master mix were added per
manufacturer’s instructions: 2.5µl 10x Core RT-PCR Buffer, 1.0µl 20mM dNTP, 0.5µl
diluted StrataScript reverse transcriptase, and 0.25µl SureStart Taq DNA polymerase
with the remaining volume in RNAse-free water. Thermal cycling conditions were as
follows and performed on a Stratagene Mx3000P: 45°C for 30:00; 95°C for 10:00; 50
cycles of 95°C for 0:15 and 60°C for 1:00.
The PCR conditions were updated in November 2007, to reflect the introduction of a new Brilliant II QRT-PCR core reagent kit (600810) and the discontinuation of the 1-step Brilliant QRT-PCR core reagent kit (600532); both kits are from Stratagene. The reaction contained 5.5mM MgCl2, 300nM forward primer SH-Poly A, 400nM reverse primer SH-Poly-1, and 200nM TaqMan probe SH-Poly-Q. The remaining components of the master mix were added per manufacturer’s instructions: 2.5µl 10x Core RT-PCR Buffer, 1.0µl 20mM dNTP, 1.0 µl provided reverse transcriptase, and 0.25µl SureStart Taq DNA polymerase. We also added 0.5µl of Rnase Inhibitor (ABI, 20IU/ml), 0.75µl T4 gene (Ambion, 2mg/ml), with the remaining volume in RNAse-free water. Thermal cycling conditions were as follows and performed on a Stratagene Mx3000P: 50°C for 30:00; 95°C for 10:00; 50 cycles of 95°C for 0:15 and 60°C for 1:00. The amount of RNA amplified was reduced from 2µl to 1µl because of consistent inhibition from the Doheny Beach samples.
All samples were amplified in duplicate along with duplicate positively spiked samples (2µl at 1x105 copies/µl). ). A 123bp synthesized ssDNA oligo was used as a positive control (IDT, Coralville Iowa, USA). This oligo was synthesized for the amplicon region (107bp) with additional bases on either end. The oligo’s sequence was based on the desired region within the complete genome sequence for Hepatitis A (gi|329582|gb|M14707.1|HPA Hepatitis A virus (wild-type) RNA, complete genome). Positive spikes were also used to test for inhibition by observing if a delay in amplification occurred from the positive control alone to the spike with sample. If inhibition was observed, the samples were amplified again within a few hours using only 1µl of extracted RNA to lessen the effect of any inhibitory compounds present. RNA was stored at -4°C between amplifications to prevent any loss from an additional freezethaw cycle. Only 1µl of RNA extract was used to test for HAV in the samples from Avalon.
PCR product from amplification of the oligo alone was cloned to use as a standard curve. Costafreda et al (2006) suggested that a dsDNA plasmid provides a quantification that most closely resembles use of the whole Hepatitis A genome. The standard curve was prepared from 2 x 106 copies to 2 x 100 by a serial tenfold dilution.
Have the results/data gathered during this reporting period changed the project objectives when compared to your original proposal? Please explain.
Yes, the Doheny Beach study was abandoned due to good water quality and study was
moved to City of Avalon. Other than the location from which the samples are being
taken, nothing else has changed.
Dissemination activities during this reporting period (please include the number of participants where applicable).
This period has largely been used to conduct sample
analysis, and conduct data analysis. There is a special Water Research issue that is slated
for publication for late 2008 that will include the results from the April Evaluation Study
and potentially some results from the two epidemiological studies. Although none of the
submissions are complete at this time, we envision at least 3 manuscripts to stem from the
funded work. These will likely be submitted to Applied and Environmental Microbiology
and Water Research.
Difficulties
Switching study locations and problems with shipments of samples have
slowed the analyses completion times but those hurdles have been long overcome.
Data Generated to date
For all the amplifications, all the negative controls were negative and all positive
controls amplified as expected. Summary of results is provided in the attached excel
spreadsheet. Refer to Sheet “HAV QPCR Results.”
ND = No copies of HAV were detected by QPCR
ND* = No copies of HAV detected, but RNA was amplified after one freeze-thaw cycle.
ND** = No copies detected but RNA was amplified with over-diluted RT enzyme and
the improper RT temperature.
Only those samples that met one of the following criteria were tested:
a) Total Enterococcus >104/100ml
b) FecalColifornms >400/100ml
c) TotalColiforms >1000/100ml
A total of 126 samples met one of these criteria. Of these 126, RNA extracts were not provided for 21 samples, the A15, Ad12, C15, Cd12 series. We were unable to locate and test one sample: B15-167.
Only one sample amplified positively, A12-76; however, only one of the two replicates amplified at a copy number of 27.05copies/1ul RNA extract so these results are not conclusive. The sample was re-analyzed (unfortunately after another freeze-thaw cycle), and no viral gene copies were detected.
Project Objectives for Next Reporting Period
Objectives
Our contract with CICEET has been extended to March 2009. In the time
frame of March 2008 through March 2009 we expect to finalize data analysis, conduct
analyses as multi-disciplinary team, and write manuscripts. Furthermore, this time will be
used to conduct Enterococcus speciation (partially supported by SCCWRP).
Work Plan to Meet Objectives
Continue with analyses as described in proposal. PI
meetings will occur for data discussions.
Dissemination Objectives for next reporting period
Analysis of this data and comparison
to human health outcomes will continue in the first two quarters of 2008. Multiple
publications are expected, including one detailing results from the preliminary study,
others detailing the microbiological analysis, and a final one including health outcomes
from the epidemiological study. The lead for this final publication will likely be
epidemiologists at UC Berkeley.
Overall Project Timeline Update
Work appears to be going on schedule.
Expenditures
They are in range.