Food Protection Trends - September/October 2024 - 361

Water Source
Creek or Rain
Temperature
Treatment
Enumeration
Petrifilm
(3 reps)
W
Colilert
(2 reps)
Petrifilm
(3 reps)
12oC
32oC
Cl
PAA
Colilert
(2 reps)
Petrifilm
(3 reps)
Colilert
(2 reps)
W
Petrifilm
(3 reps)
Colilert
(2 reps)
Cl
Petrifilm
(3 reps)
Colilert
(2 reps)
PAA
Petrifilm
(3 reps)
Colilert
(2 reps)
FIGURE 1. Overview of experimental design, including water source, temperature, treatment, enumeration, and number of replications completed.
isolated E. coli colony from the NA plates was then used to
make the working inoculum, as described in detail later. The
appearance of each strain as pure culture was documented
before the beginning of the study by plating on E. coli/
coliform (EC) Petrifilm, where the colonies grew blue to
purplish with gas bubbles.
Inoculum preparation
An isolated colony of each ATCC strain was added to a
separate 10-ml brain heart infusion (BHI) broth (Thermo
Scientific Oxoid, Hants, UK) tube. The three tubes were
incubated at 37°C for 24 ± 2 h. A pellet of pure 108
-109
CFU/ml E. coli was then separated from the broth by
separately centrifuging the tubes at 4,300 × g for 15 min
at 4°C. The supernatant broth was discarded and 10 ml of
phosphate-buffered dilution water (PBDW; EMD Millipore,
Billerica, MA) was used to rehydrate each pellet. Preliminary
studies confirmed that these methods result in consistent
concentrations for each strain. The concentration of each
strain was enumerated, which resulted in 9 ml remaining
for each strain following enumeration. The remaining equal
volumes (9 ml each) of the three strains were then mixed
to form a cocktail solution that was used as the working
inoculum (~27 ml total). The working inoculum was
enumerated at the beginning and the end of the trial to
ensure that the E. coli population was consistent throughout
the inoculation trial. To enumerate the concentrations of
the strains individually and the used cocktail, subsequent
dilutions were made in PBDW and plated in duplicate on
EC Petrifilms; the EC Petrifilms were incubated at 37°C for
48 ± 4 h. Individual strain colonies (blue and purplish with
gas bubbles) were counted on the EC Petrifilms, and they
were used as a reference for proper enumeration of generic
E. coli (i.e., appearance of each inoculum strain on an EC
Petrifilm) in the inoculated study. The rain barrel and filtered
creek water native E. coli populations were also enumerated
preinoculation using Colilert Quanti-Tray/2000 (IDEXX
Laboratories, Westbrook, ME).
Water sampling, preparation, and inoculation
A 6.5-liter batch of rain barrel water was collected from
a rain barrel owned by a local produce grower in northeast
Kansas. The rain barrel was a 50-gal covered barrel stored
outside, collecting rainwater from the gutters of a garage
with a metal roof. A sterilized carboy with a 9-liter capacity
was placed directly under the spigot of a rain barrel
(covered with a lid) and used to collect the rainwater
directly. Before collection, the external surface of the rain
barrel spigot was sterilized, and the water was run for 1
min in the carboy before collection. From a creek (mostly
rainfed) used for produce irrigation at the Olathe Kansas
Horticulture Research and Extension Center, 6.5 liters of
creek water was pumped through a sand filter before it was
collected directly into a 10-liter sterile carboy. Following
each collection, a portion (three subsamples) of each water
type was enumerated for naturally occurring coliform and
E. coli populations using Colilert (Table 1). Water samples
were collected on three separate occasions in April 2022.
Both water types were also analyzed for turbidity, pH,
electroconductivity, and total dissolved solids (TDS) (Table
2). Then, each batch of source water was separated into two
water samples, each with a volume of 3 liters. One 3-liter
bottle of each water type was stored overnight (8-10 h) at
32 and 12°C, as described in the FDA/EPA protocol (34), to
simulate cool season and warm season scenarios.
Before inoculation the following day, each water sample
was plated on EC Petrifilm, as described later, to ensure that
naturally occurring populations did not change during the
overnight temperature equilibration. Each 3-liter sample
bottle per water source and temperature (e.g., rain at 32°C
and creek at 12°C) was inoculated with 1 ml of the working
inoculum to achieve a target concentration of ca. 5 log CFU/
ml. The E. coli concentration of each 3-liter bottle of water
was confirmed after inoculation by plating on EC Petrifilm,
as described later. Each inoculated bottle of water was then
equally distributed in three subsamples of 990 ml and placed
back at their respective temperature for approximately 30
min for temperature equilibration.
September/October Food Protection Trends 361

Food Protection Trends - September/October 2024

Table of Contents for the Digital Edition of Food Protection Trends - September/October 2024