Food Protection Trends - November/December 2024 - 422

ml of phosphate-buffered saline (VWR Chemicals, Solon,
OH). Salmonella strains were equally combined to form a
3-ml multistrain Salmonella cocktail. The initial population
was verified by plating on TSA.
Sample preparation and inoculation
Banana leaves were collected from banana trees locally
grown in an outdoor farm in Hutchinson, KS and transported
at ca. 20°C (car air-conditioning system) in polyethylene
bags to Kansas State University. Leaves were stored at −80°C
until processed. Bamboo baskets were purchased from a local
vendor and stored at room temperature in the laboratory.
Experiments were performed on 3- by 3-cm2
banana leaves and 1.2- by 3-cm2
samples of
samples of bamboo baskets.
Samples of both surfaces were prepared with a sterile
template under aseptic conditions. Samples were placed
into a sterile petri dish and exposed to UV-C light for 15
min to eliminate background microflora. Surfaces used for
wiping and scrubbing were also UV sterilized. Samples were
inoculated by applying ten 10-μl drops of the Salmonella
cocktail onto each sample. Banana leaves were inoculated on
the underside of the leaf. Bacteria were allowed to attach for
60 min following inoculation.
Treatment description and allocation
Samples obtained from a single banana leaf were randomly
assigned to treatments, such that a leaf served as a blocking
structure in the experimental design. Similarly, samples
obtained from an individual bamboo basket were randomly
assigned to treatments, such that a basket served as a
blocking structure. Six banana leaves and six bamboo baskets
were used in this study.
Each treatment was defined by the combination of
cleaning and sanitizing methods. Three cleaning methods
were evaluated: (i) wiping with a dry paper towel vertically
and horizontally five times in each direction, (ii) scrubbing
with a dry kitchen sponge vertically and horizontally five
times in each direction, and (iii) no cleaning. At each change
of direction a different part of the dry paper towel and dry
kitchen sponge was used to avoid cross-contamination.
After cleaning, five sanitizing methods were used: (i) 200
ppm of disinfecting bleach (Clorox Company, Oakland,
CA) applied for 3 min, (ii) 200 ppm of disinfecting bleach
applied for 5 min; (iii) 400 ppm of peracetic acid solution
(PAA; Shield-Brite 15.0, Pace International, Wapato, WA)
applied for 3 min; (iv) 400 ppm of PAA applied for 5 min;
and (v) no sanitizing. The sanitizing method was performed
by immersing the sample into a 50-ml test tube containing 20
ml of PAA or bleach solution. Sanitizer solutions were freshly
prepared from bulk ingredients before each application,
and the concentration of each sanitizer was measured with
commercial test kits (Peracetic Acid Test Kit and Chlorine
Test Kit, ChemWorld, Kennesaw, GA).
Salmonella attachment determined by scanning electron
microscopy
Samples (1 by 1 cm2
) of banana leaves and bamboo
baskets were prepared and inoculated as described above for
microscopic examination. Inoculated samples were stained
with SYTO9 and SYTOX red (Thermo Fisher Scientific,
Eugene, OR) and fixed with 10% formalin as previously
reported by Mendez et al. (18). Samples were analyzed by
scanning electron microscopy (SEM; Hitachi S-3500N,
Tokyo, Japan) in the Microscopy Facility, Division of Biology,
at Kansas State University.
Statistical analysis
A general linear mixed model was used to evaluate the
effect of cleaning and sanitizing methods on Salmonella
counts, which were averaged over duplicate count plates
for each sample and expressed as log CFU/cm2
. A separate
analysis was conducted on data obtained from banana leaves
and from bamboo baskets. In both cases, the linear predictor
included the fixed effects of cleaning, sanitizing methods, and
their two-way interaction. For banana leaves, the statistical
analysis excluded data corresponding to sanitizing with
PAA because Salmonella counts were below the limit of
quantification (LOQ; <1.44 log CFU/cm2
) for all samples.
The random effect of leaf or basket was incorporated into
the model as an overarching blocking structure. Restricted
maximum likelihood was used to estimate variance
components, and the Kenward-Roger's approach was
used for estimation of degrees of freedom and standard
error correction. Model assumptions were checked and
Food Protection Trends - November/December 2024

Table of Contents for the Digital Edition of Food Protection Trends - November/December 2024
