Development of a Rapid Detection Method for Waterborne E. coli O157:H7 M. J. Collins, R. M. Hoffman, B. T. Argue, J. H. Standridge WI State Laboratory of Hygiene, Madison, WI

1
Development of a Rapid Detection Method for
Waterborne E. coli O157H7 M. J. Collins, R. M.
Hoffman, B. T. Argue, J. H. Standridge WI State
Laboratory of Hygiene, Madison, WI
Marty Collins Wisconsin State Lab of Hygiene 2601
Agriculture Drive Madison, WI 53718 (608)
224-6260 Phone [phone]
Fax collinmj_at_mail.slh.wisc.edu
Poster Q-496
Abstract Background E. coli O157H7 is a highly
virulent pathogen implicated in food and
waterborne outbreaks. Documented outbreaks
include over 10,000 cases with several hundred
deaths. Waterborne outbreaks represent 10 of the
reported epidemics. Existing methods for
detection of E. coli O157H7 in water are
non-standardized, have poor sensitivity, or
require enrichment steps resulting in assay times
exceeding 72 hours. These factors significantly
hamper outbreak investigations. The intent of
this study was to develop a rapid and sensitive
method to detect E. coli O157H7 in water.
Methods Various concentration techniques
including diatomaceous earth, and 0.45 micron
membrane filtration were evaluated prior to
choosing filtration through 142 mm diameter, 0.4
micron polycarbonate track-etched membrane
filters. Filters were placed in 25 mL modified
buffered peptone water and incubated at 37?C for
six hours. The organisms were concentrated from
enrichment broth using immunomagnetic separation
with anti-E.coli O157H7 Dynabeads. Bead-cell
complexes were stained with a FITC-conjugated
anti-E. coli O157H7 and screened using flow
cytometry. Samples exhibiting fluorescence were
considered presumptive positive, plated onto
cefixime-tellurite-sorbitol-MacConkey agar and
CHROMagar? O157 plates, incubated for 18 hours,
and typical colonies confirmed with O antisera
agglutination. Results Multiple spikes into
complex lake water samples confirmed the
sensitivity of the assay down to 10 organisms/500
mL. The assay was field tested with 129 samples
collected and analyzed during a 10 week swimming
season. None of the samples were positive for E.
coli O157H7. Conclusions The assay described
above provides a much improved time frame for
detecting E. coli O157H7 in water samples. In a
public health laboratory with flow cytometry
capabilities, presumptive positives can be
attained in less than 8 hours, with confirmed
results available even in laboratories without
flow cytometers, within 24 hours. The assay also
improves sensitivity down to 10 organisms/ 500
mL, an acceptable level for public health
investigations.

• Methods
• Organisms Either E. coli O157H7 ATCC isolate
  700927 or a clinical isolate were used for all
  experiments. Cells were grown to log phase in
  modified buffered peptone water at 37ºC. Stock
  concentrations were determined by flow cytometry
  using Flow Count Beads (Beckman-Coulter,
  Fullerton, CA) with methods previously validated
  in our laboratory. Stock cultures were serially
  diluted in sterile saline to achieve
  concentrations ranging from 10 10,000
  organisms.
• Field testing 129 recreational water samples
  from 3 area lakes were processed using the
  methods listed below. Positive controls
  consisting of natural samples spiked with 103 E.
  coli O157H7 and negative reagent water controls
  were assayed alongside each set of samples.

• Sample concentration
• Three filtration techniques were evaluated by
  filtering 500 mL of natural water collected from
  three Madison, WI area lakes and from a local
  pond. Techniques included
• Diatomaceous earth
• Filta-Max concentrator tube (adapted from the
  FiltaMax system, Idexx Laboratories, Portland,
  ME)
• 142 mm diameter, 0.4 micron polycarbonate,
  track-etched (PCTE) Millipore Isopore membrane
  filters (Millipore, Bedford, MA).
• Flow rate and volume limitations determined.
  PCTE filters were chosen due to ease of use and
  cost.

Introduction E. coli 0157H7 has been implicated
in waterborne disease outbreaks affecting
thousands of individuals throughout North
America, Europe and the Far East. Fatalities
from complications of infection have occurred in
almost all of these outbreaks in part due to the
absence of effective clinical therapies. A
number of these epidemics have been associated
with failures of disinfection systems or
compromises in the integrity of the water
distribution systems serving the community and
the incidence of disease is expected to increase
over the next twenty years based on the rising
number of epidemics observed in recent years.
Current E. coli analytic methods are incapable of
detecting this highly virulent strain and hence
may be ineffective for routine monitoring
purposes, particularly in communities relying
upon water sources on or near concentrated
farming operations. The intent of this study was
to develop a rapid and sensitive method to detect
E. coli O157H7 in water.
2
Flow cytometric histograms depicting peak
FITC-fluorescence (X-axis) vs. log forward angle
light scatter (Y-axis). The presence of
fluorescent events indicates the presence of E.
coli O157H7. Panels A-C show results from
distilled water seeded with 0, 10, and 100 E.coli
O157H7 in 500 ml. Panels B-D show results from
lake water seeded with 0, 10, and 100 E.coli
O157H7 in 500 ml.
Negative control
10 organisms/500 ml
100 organisms/500 ml
A
C
B
Distilled water
F
E
D
Lake water
Conclusions We have developed an E. coli 0157H7
method utilizing IMS with flow cytometric
detection following concentration and a 6 hour
enrichment. This method is capable of detecting
10 organisms in 500 mL natural water samples a
significant sensitivity improvement over
currently accepted methods which may take up to
72 hours for confirmation. In a public health
laboratory with flow cytometry capabilities,
presumptive positives can be attained in less
than 8 hours, with confirmed results available
even in laboratories without flow cytometers,
within 24 hours.
Results 1. Both the FiltaMax concentrator tube
filtration system and the PCTE membrane were
capable of filtering 500 mL natural water
samples. Diatomaceous earth filters were not
capable of filtering the entire sample volume.
PCTE membranes were chosen due to ease of use and
cost. 2. Both reagent water and natural water
samples spiked with 10, 100, and 1000 organisms
were positive using the IMS/flow cytometric
method described above. 3. The assay was field
tested with 129 samples collected and analyzed
during a 10 week swimming season. None of the
samples were positive for E. coli O157H7.
Positive and negative control samples were
acceptable for all samples analyzed.

• Future Research
• Delayed exposure of various antibiotics and use
  of media additives (e.g., antioxidants) to
  improve culturability of viable but
  non-culturable organisms
• PCR confirmation
• Determining optimal enrichment temperatures for
  culturing viable but non-culturable organisms
• Further comparison of CT-SMAC, CHROMagar O157,
  Rainbow Agar O157

Acknowledgements This project is being conducted
by the Wisconsin State Laboratory of Hygiene,
Madison Department of Public Health, and the
United States Geological Survey. This project
is being funded by the U.S. EPA, Project
R82933901-0.