Content uploaded by Ben D Tall
Author content
All content in this area was uploaded by Ben D Tall on Sep 18, 2018
Content may be subject to copyright.
BACTERIAL ADHERENCE
AND
VIABILITY ON
CUTTING BOARD SURFACES
SIMIN
H.
ABRISHAMI', BEN
D.
TALL2, THOMAS
J.
BRUURSEMA',
PAUL
S.
EPSTEIN'
and
DHIREN
B.
SHAH2v3
'Department of Microbiology,
NSF
International
Ann Arbor, Michigan
48113
and
'Division of Microbiological Studies
Center for Food
Safety
and Applied
Nutrition
U.S.
Food
and
Drug Administration
Washington,
DC
20204
Received
for
Publication December
1, 1993
Accepted
for
Publication January
13, 1994
ABSTRACT
The adherence and viability of
Escherichia coli
inoculated onto the surfaces
of plastic cutting boards and new and used wood cutting boards were evaluated.
Most
of
the inoculum was recovered
from
all surfaces after resident drying times
of
5
min and
from
plastic
surfaces
at
24
h.
When
the exposure time was
extended to
2
h,
>
90%
of
the cells placed
on
new and used dry wood surfaces
were not recovered after vigorous rinsing. Scanning electron microscopy
showed that the bacteria resided within the structural xylem fibers and vegetative
elements of the wood. After resident drying times
of
up
to
2
h, almost
75%
of
the adherent bacteria
on
the wood surfaces were viable, as defined
by
a
nalidixic
acid direct viable count procedure. Microcosm studies showed
no
intrinsic
growth-supporting
or
toxic properties
of
the cutting board materials. Bacteria
that adhered to plastic
su
faces were more easily removed by low-temperature
washing than were cells that adhered to wood surfaces. These studies
demonstrated that bacteria adhering to wood surfaces resided within the
structural and vegetative elements
of
the wood's xylem tissues and were viable;
wood was more retentive than plastic; penetration
of
the inoculum liquid
promoted cell adherence to the wood matrix; and conditioning
of
wood with
water before inoculation interfered with bacterial adherence.
3Author
to
whom
requests
for
reprints should be addressed.
Journal of Food Safety
14 (1994) 153-172.
All
Rights Reserved.
Vopyright
1994
by
Food
&
Nutrition Press, Inc.,
Trumbull,
Connecticut.
153
154
S.
ABRISHAMI, B. TALL, T. BRUURSEMA,
P.
EPSTEIN
and
D.
SHAH
INTRODUCTION
Recent outbreaks of enterohemorrhagic
E.
coli
in hamburger meat (CDC
1993),
salmonellosis associated with sliced watermelon, powdered milk, and
infant formula (Louie
et
al.
1993);
and
Cumpylobucter-contaminated
chicken
carcasses (Giesendorf
et
al.
1992;
Oosterom
et al.
1983)
have focused public
awareness on bacterial contamination
of
foods and current sanitary food handling
and processing practices. The sanitation of food preparation surfaces, including
cutting boards and cutting board materials, is important in the prevention of
these outbreaks. It is now well known that bacteria can adhere to many natural
and man-made surfaces (Costerton
et
al.
1978, 1987).
Once attached and under
favorable conditions, they can multiply, form microcolonies, elaborate hydrated
exopolysaccharides, and eventually develop into highly complex, dynamic
microbial ecosystems called biofilms. Many pathogenic and spoilage bacteria
form biofilms on materials commonly used in food processing equipment, such
as stainless steel, glass, rubber, and Teflon gaskets (Costerton
et
al.
1987;
Mosteller and Bishop
1993).
Once established, pathogenic bacteria can detach
and contaminate foods in sufficient numbers to be infectious, or, given optimal
conditions such as those associated with temperature abuse, may multiply
to
numbers considered
to
be infectious (Mosteller and Bishop
1993).
Health
agencies often recommend the use
of
nonporous plastic cutting board surfaces
for food preparation. Wood cutting boards are considered unsafe because their
high porosity and absorbency allow for retention
of
microbial contaminants,
making sanitization difficult.
In addition to viability, easy removal of pathogenic microorganisms from food
preparation surfaces is important in the safety of a particular type of material.
As
yet unpublished results of Cliver and Ak as reported by Raloff
(1993)
suggested that wood cutting boards possess properties that kill bacteria and
therefore might be inherently safer to use than plastic. However, they were
unable to find any factors in wood responsible for the bactericidal effect. These
studies also showed that bacteria were more easily recovered from plastic cutting
board surfaces than from wood surfaces.
Before Raloff's report there had been no accurate studies
of
the adherence of
bacteria to cutting board surfaces. Because of the paradoxical implications of
these results, we undertook the present investigation. Our objectives were to
evaluate bacterial attachment and viability to plastic and wooden cutting board
surfaces, to determine
if
either material exhibited bacterial growth-promoting or
inhibiting properties, and to examine the retention
of
bacteria on surfaces after
a low temperature, non-detergent wash in a modified commercial dishwasher.
BACTERIA
ON
CU?TING
BOARDS
155
MATERIALS
AND
METHODS
Escherichia coli
Strains
Escherichia coli
strains ATCC 11229 and 933
J
were used in this study.
E.
coli
ATCC 11229 was stored on nutrient agar slants (Difco Laboratories,
Detroit,
MI);
strain 933
J
(0157:H7) was maintained at -7OC in Luria-Bertani
(LB) broth with 12% glycerol. We studied
E. coli
ATCC 11229 because of its
broad use in the testing of food equipment, particularly in demonstrating the
efficacy of sanitization of commercial equipment, and because it has been used
extensively to test surface disinfectants (American NSF 1989; AOAC, 1990;
NSF 1992).
E.
coli
strain 933
J
has been studied extensively, and much is
known about its pathogenicity, lysogeny
,
toxigenicity, and adherence properties
(O’Brien and Holmes 1987; Strockbine
er
al.
1986; Karch
er
al.
1987).
Inocula Preparation
E. coli
ATCC 11229 was used in analyses of adherence, viability, microcosm,
and low-temperature cleaning. Cells were harvested from nutrient agar slants
with phosphate-buffered dilution water (PBDW) after a 24-h incubation at 35C.
The inoculum was prepared by suspending
E. coli
ATCC 11229 cells in a
solution of
50%
horse serum (Sigma Chemical Co., St. Louis,
MO)
and
10%
vegetable oil (Kroger Co., Cincinnati, OH) in PBDW to a final cell concentra-
tion of 107-109 colony forming units (CFU)/ml. Replicate pour plate quantitative
cultures using violet red bile (VRB) agar (Difco) were prepared to confirm
viability and inoculum size (APHA 1989). Both the serum and vegetable oil
were examined and found
to
contain no bacteria.
E. coli
strain 933
J
was used for the scanning electron microscopy
(SEM)
studies. LB broth was inoculated from a frozen stock and incubated overnight,
with aeration at 30C. The overnight culture was diluted 1 :20 in fresh LB broth
and re-incubated as described above until the culture reached a Klett value of 80
units. The cells were harvested by centrifugation
(5,000
x
g
for 10 min) and
suspended in phosphate-buffered saline (PBS), pH 7.2, to a Klett value of 200
units (approximately
lo9
CFU/mL). This cell suspension was used as inoculum.
Preparation
of
Cutting Board
The new wood cutting boards used in this study were constructed from hard
maple and were kindly supplied by John
Boos
and Company (Effingham, IL).
The plastic cutting boards were molded from clear acrylic
(U.S.
Acrylic, Inc.,
Northbrook,
IL).
Used wood and plastic cutting boards were supplied by the
NSF
International staff. The cutting board surfaces used in the viability and
low-temperature cleaning experiments were first cut with a band saw into 8.9
156
S.
ABRISHAMI, B. TALL,
T.
BRUURSEMA,
P.
EPSTEIN
and
D.
SHAH
x
8.9 cm2 planchets. The maple planchets used for
SEM
were cut from a 13-
mm
diameter dowel (edge grain cut). The planchets used in the viability and
low-temperature cleaning studies were sterilized either by boiling in deionized
water or steam-sterilized in an autoclave for
5
min. Those used for SEM
analysis were sterilized by irradiation for
5
min per side, using germicidal
ultraviolet light. Sterility of the planchets was confirmed before testing by
placing them in individual sterile 150
x
15
mm
petri dishes and covering them
with a poured layer
of
tempered VRB agar (50C). After cooling, the cultures
were incubated for 24 h at 35C. All sterilized planchets were negative for
bacterial growth.
Because of the absorptive nature
of
wood, the wood planchets were tested
both dry and wet. The dried wood planchets, used in the viability, recovery and
low-temperature cleaning assays, were dried in
an
uncovered 150
x
15
mm
petri dish for 24 h under ambient conditions in a laminar flow hood, which
provided a vertical flow of HEPA-filtered air at approximately 100 ft/min. The
new and used wooden planchets designated as "wet" were conditioned in sterile
deionized water for 30 min under the same conditions.
Viability and Recovery
of
E. coli
from Cutting Board Surfaces
To simulate general use conditions, we inoculated duplicate plastic and dry
and wet wooden cutting board planchets with 1
ml
of the serum-oil-cell
suspension containing 107-109 CFU/ml of
E.
coli
strain 11229. The inoculum
was spread with a sterile bent glass rod and air-dried for
5
min,
2
h, or 24 h in
the laminar flow biological hood. At stated times, cells adhering to the surface
of the planchets were eluted by transferring each planchet to a 15.2
x
15.2 cm
Corning cylindrical jar (Coming Inc., Coming,
NY)
containing 99
ml
of PBDW
(Fig. 1). The planchet in the jar was shaken for 2-3 min and the bacteria eluted
from the cutting board surfaces were enumerated in all experiments by the pour
plate technique, using VRB agar. Finally, the rinsed planchets were cultured as
described earlier. Colonies were enumerated after
24
h of incubation at 35C.
Adherence Assay and Direct Viable Count-Scanning Electron Microscopy
Adherence was assayed by placing 30
p1
of an inoculum
of
E.
coli
strain 933
J
containing approximately
lo9
CFUlml
onto one side of five separate maple
planchets. The inoculum was air-dried for 10 min, after which one
of
the
planchets was processed immediately for evaluation by
SEM.
Two of the
remaining four planchets were placed inoculum-side down in a
50
x
15
mm
petri dish containing
10
ml
of LB medium; the other two were placed in 10
ml
of
LB
medium supplemented with
4
pglml
of nalidixic acid and incubated at
30C. A modification of the direct viable count procedure described by Kogure
(DVC-SEM)
BACTERIA ON
CUTTING
BOARDS
157
ef
al.
(1978) was performed by incubating the inoculated planchets for
5
h. A
second experiment was performed as described above, except that the inoculum
was air-dried for 2 h at 20C. Replicate spread plate quantitative cultures using
MacConkey agar were made of the inoculum and LB cultures to confirm
viability, percentage of cells adherent to the surface, and inoculum size. One
planchet each, from the LB culture and the nalidixic acid supplemented LB
culture, was processed for SEM.
Scanning Electron Microscopy
(SEM)
All of the planchets were vapor-fixed in 2% osmium tetroxide for 2 h; washed
three times in 0.2 M sodium cacodylate buffer, pH 7.2; dehydrated in an
ascending graded ethanol series (30-100%); critical point-dried (purge time, 15
min in liquid C02, using
an
Autosamdri 814; Tousimis Research Corp.,
Rockville, MD); mounted onto aluminum stubs; and sputter coated with
60%
gold-40% palladium, in argon, for 2 min at 15
mA,
using a PS-2
IS1
sputter
coater. The specimens were then viewed
in
an
IS1
super
111
A scanning electron
microscope at
an
accelerating voltage of 15 kV. Cell size was determined on
scanning electron photomicrographs at
7000
diameters and approximately 100
cells were counted for each analyses.
Microcosm Assay
Wood and plastic cutting boards were also tested for their growth-promoting
potential. Duplicate microcosms containing 11 g each of wood cutting board
dust, wood cutting board chips, and plastic cutting board pieces were individual-
ly
placed in 99
ml
of PBDW. Each microcosm was inoculated with 1 mL of a
cell suspension of
E.
coli
strain 11229 (107-108 CFU/ml) contained in PBDW.
Microcosms containing a similar concentration of
E.
coli
strain 11229 in PBDW
served as control. The microcosms were incubated for 24 h at 35C, and cell
viability was determined by the pour plate technique, using VRB as described
earlier.
Low Temperature Cleaning
of
Cutting Boards in a Commercial Dishwasher
Five replicates each of new and used wood and plastic planchets were
inoculated with 1.0
mL
of
lo7
CFU/ml of
E.
coli
strain 11229. Inoculated
planchets were placed and washed without detergent in a
Hobart
UM-4D
dishwashing machine, within 3 min resident time, using dechlorinated tap water
at a temperature of 15.5-21.1C. All operations of the dishwasher were
automatic, including a 30-s fill of 2.3
i-
0.1 gal; a 120-s wash; a 30-s drain; a
30-s rinse of 2.3
f
0.1 gal; and a 30-s final rinse. Total cycle time was 240
s.
To eliminate cross-contamination of materials, the dishwasher rack was
158
S.
ABRISHAMI, B. TALL, T. BRUURSEMA,
P.
EPSTEIN
and
D. SHAH
modified
so
that
for
each cycle, the position
of
a single test sample was
identical. After the wash, the planchets were placed in
99
ml
of
PBDW
in a
15.2
x
15.2 cm Coming cylindrical
jar
and vigorously shaken for
2-3
min.
The bacteria recovered from the surfaces were enumerated by the pour plate
technique, using
VRB
agar as described above.
A
flow
chart of this procedure
is shown in Fig.
1.
E.
coli
in suspending medium
(50%
serum/lO
%
oil
in
PBDW)
1
mL
1
mL
1
mL
on each planchet in
99
mL
diluent on each planchet
(density control)
uniformly spread
washed in dish
washing machine
I
uniformly spread
cells eluted from surface
in
99
mL
diluent
cells eluted from surface
in
99
mL
diluent
assayed in
VRB
FIG.
1.
PROCEDURE FOR REMOVING ADHERENT BACTERIA FROM CUTTING
BOARD SURFACES BEFORE AND AFTER LOW-TEMPERATURE WASHING.
BACTERIA
ON CUTTING
BOARDS
159
RESULTS
Viability and Recovery
of
Bacteria
from
Cutting Board Planchets
The behavior of inocula placed on the different cutting board surfaces at
various times was characterized. After a 5-min resident time, the surface of all
wooden planchets spread with 1
.O
ml
of the serum-oil-PBDW inoculum appeared
wet. When the exposure time was extended to 2 h, all of the liquid was
absorbed by the new and used dry wood, but not by the water-conditioned new
and used wood. Nearly all of the bacteria placed on the wooden surfaces for the
5-min resident time were non-adherent and were recovered in the rinse after
vigorous shaking (Table
1).
Similarly, adherence of bacteria was not significant
after the 2-h resident time on water-conditioned new and used wood. In
contrast, of the cells placed on the new and used dry wood,
>95%
and
91
%,
respectively, were not recovered in the rinse (Table
1).
This result suggests that
penetration of the inoculum liquid promoted cell adherence to the wood matrix.
The inoculum placed
on
plastic planchets was not absorbed or desiccated even
after 24
h
of resident time, and almost all of the cells of the inoculum were
recovered in the rinse (Table 1).
Adherence Assay and DVC-SEM
The inoculum in PBS, applied to dry new wood planchet surfaces, was totally
absorbed into the wood by
10
min.
SEM
showed many bacteria associated with
the cytoplasmic regions of the dried structural and vegetative elements of the
xylem tissue of the maple wood (Fig. 2). Of the inoculum cells (3.7
x
lo7
CFU/mL) placed on these planchets, only
4.5
X
lo6
CFU/mL (12.2%) were
recovered after vigorous agitation and subsequent culture. These results suggest
that
88%
of the cells were not recovered and were adherent to the cutting board
surface. To determine whether the localized bacteria were viable, we incubated
the planchets in LB broth containing nalidixic acid. Under these conditions, as
expected, viable cells formed filaments (Fig. 3). The size distribution of cells
placed on new wood surfaces before and following incubation in LB broth with
or without nalidixic acid is presented in Fig.
6.
The size measurements of
individual cells were made from scanning electron photomicrographs represent-
ing a final magnification of 7000 diameters in
all
experiments. These data
showed that 85% of the cells of the inoculum placed on new wood were less
than
10
mm
in
length. Similarly, for cells placed on new wood and incubated
in
LB broth without nalidixic acid, 95% of the cells were less than
10
mm
in
length and consisted of single cells or doublets (Fig.
4).
In contrast, 75
%
of the
cells placed
on
new wood and incubated in LB broth containing nalidixic acid
were longer than
10
mm.
Nearly 45
%
of these cells were longer than 15
mm.
TABLE
1.
RECOVERY OF
ESCHERICHIA
COLI
STRAIN
11229
FROM NEW AND USED
WOOD
AND PLASTIC PLANCHETS
AFTER
5
MIN,
2
H,
AND
24
H
RESIDENT DRYING TIME
~ ~~~~~
Number
of
cells recovered
(CFU/mL)
Log
Plancheta
5
min reduction 2h
NR ~og,~N,/
~,b
NR
New, Dry
1.9
x
lo7
-0.22
New, Wet
2.3
x
lo7
-0.14
Used, Dry
2.9
x
lo7
-0.04
Used, Wet
1.9
x
lo'
-0.22
Plastic ND ND
6.35
x
lo5
-1.70
1.35
x
107
-0.37
1.3
x
10'
-1.09
5.7
x
10'
-0.44
ND ND
ND~
ND
ND ND
ND ND
ND ND
2.9
x
lo9
0.06
~
a
Sterile planchets were dried and water-conditioned as described in Materials and Methods. Serum-
oil-PBDW was used to suspend cells
of
the inoculum. New and used wood planchets were inoculated
with
3.2
x
lo7
CFU/ml. Used wood
(2
h, only) and plastic planchets were inoculated with
1.6
x
lo9
CFU/ml and 2.5
x
lo9
CFU/ml, respectively.
Log
reduction was calculated by dividing the number of CFU/ml recovered from each planchet
(N,)
by
the inoculum size
(N,).
ND,
not determined.
TABLE
1.
RECOVERY OF
ESCHERICHIA
COLI
STRAIN
11229
FROM NEW AND USED
WOOD
AND PLASTIC PLANCHETS
AFTER
5
MIN,
2
H,
AND
24
H
RESIDENT DRYING TIME
~ ~~~~~
Number
of
cells recovered
(CFU/mL)
Log
Plancheta
5
min reduction 2h
NR ~og,~N,/
~,b
NR
New, Dry
1.9
x
lo7
-0.22
New, Wet
2.3
x
lo7
-0.14
Used, Dry
2.9
x
lo7
-0.04
Used, Wet
1.9
x
lo'
-0.22
Plastic ND ND
6.35
x
lo5
-1.70
1.35
x
107
-0.37
1.3
x
10'
-1.09
5.7
x
10'
-0.44
ND ND
ND~
ND
ND ND
ND ND
ND ND
2.9
x
lo9
0.06
~
a
Sterile planchets were dried and water-conditioned as described in Materials and Methods. Serum-
oil-PBDW was used to suspend cells
of
the inoculum. New and used wood planchets were inoculated
with
3.2
x
lo7
CFU/ml. Used wood
(2
h, only) and plastic planchets were inoculated with
1.6
x
lo9
CFU/ml and 2.5
x
lo9
CFU/ml, respectively.
Log
reduction was calculated by dividing the number of CFU/ml recovered from each planchet
(N,)
by
the inoculum size
(N,).
ND,
not determined.
FIG.
2.
SCANNING ELECTRON PHOTOMICROGRAPH SHOWING
E.
COLI
INOCULUM STRAIN
933
J
ADHERING
TO
WOOD SURFACE AFTER AIR DRYING FOR
10
MIN
c
Note:
Box in
A
delineates area magnified in B.
The
letters
S
and
V
denote
the
structural elements and vegetative
E
elements
of
the
xylem tissue. Bar markers represent
10
pm
and
1
pm,
respectively
162
S.
ABRISHAMI,
B.
TALL,
T.
BRIJURSEMA, P. EPSTEIN and
D.
SHAH
BACTERIA ON CUTTING BOARDS
163
0
im
164
S.
ABRISHAMI,
B.
TALL, T. RRUURSEMA,
P.
EPSTEIN and D. SHAH
If we use a cut-off
of
10
mm
as the average cell length before nalidixic acid
exposure, the data suggest that at least
75%
of the cells were viable after a
resident drying time of
10
min
on
the wood surface. Similar results were
observed for cells subjected to
a
2-h resident drying time with
>75%
cell
viability (Fig.
6).
Unlike new wood, the used wood planchets required
>
90
min for penetration
of the inoculum liquid.
SEM
of uninoculated used wood surfaces showed
multiple layers of organic material covering the structural and vegetative
elements
of
the wood tissues (compare Fig. 2 and
5).
Inoculum dried for
2
h
on
used wood and analyzed in an experiment analogous to the nalidixic acid
procedure described above is shown in Fig.
6.
The results also showed that
74% of cells were still viable.
FIG.
5.
SCANNING ELECTRON PHOTOMICROGRAPH
OF
UNINOCULATED, USED
CUTTING BOARD SHOWING ORGANIC MATERIAL ASSOCIATED WITH THE
SURFACE
Bar marker represents
100
pm.
z
I-
<
-1
3
a
0
n
s?
-1
-1
W
0
s
CELL
SIZE
(mm
at
7000X)
FIG.
6.
SIZE DISTRIBUTION
OF
ENTEROHEMORRHAGIC
E.
COLI
PLACED ON
WOOD
SURFACES
(-El-)
E.
coli
cells
on
new wood dried
for
10
min;
(-W-)
cells placed
on
new wood dried
for
10
min
and incubated in LB broth
for
5
h;
(-O-)
cells
placed
on
new wood dried
for
10
min
and incubated
in LB broth with
4
pglml
of
nalidixic acid
for
5
h;
(-0.)
cells
placed
on
new wood dried
for
2
h
and incubated in LB broth with
4
pg/mL
of
nalidixic acid
for
5
h;
and
(-
-)
cells placed
on
used
wood dried
for
2
h
and incubated in LB broth with
4
pglml
of
nalidixic acid
for
5
h.
Microcosm Assay
To determine whether the properties of the cutting board materials were
intrinsically beneficial or deleterious, we analyzed microcosms containing
E.
coli
and wood dust, wood chips, or plastic pieces. These materials were found to
be innocuous, imparting no beneficial or deleterious effects
on
the viability of
the cells (Table
2).
Effects of Low-Temperature
Cleaning
The effects of low-temperature cleaning were studied under conditions
analogous
to
washing in a sink under a steady stream of cold water. New and
used wood and plastic planchets were washed in a commercial dishwasher
to
maintain identical wash conditions for all the planchets. After cold water
washing, more bacteria were removed from the plastic surfaces than from the
wood surfaces (Table
3),
and after cold water washing, more bacteria were
retained
on
new
than
on
used wood.
1
I
166
S.
ABRISHAMI, B. TALL, T. BRUURSEMA, P. EPSTEIN
and
D.
SHAH
TABLE
2.
MICROCOSM STUDIES: PERCENTAGE
OF
VIABLE
ESCHERICHIA
COLI
STRAIN
11229
CELLS AFTER
5
MIN AND
24
H
INCUBATION
WITH CUTTING BOARD MATERIALS
Number
of
cells
specimen
(CFU/mL)
%
Cells
wood
Inoculum
6.6
x
lo7
100
Wood dust,
24
h
7.2
x
10’
109
Wood
chips,
5
min
8.1
x
lo’
122
Wood dust,
5
min
6.4
x
lo7
96
Wood
chips,
24
h
5.6
x
lo7
84
Plastic
Inoculum
3.3
x
108
100
Plastic,
24
h
3.4
x
lou
103
DISCUSSION
The sanitation of food preparation surfaces, including cutting boards and
cutting board materials, is critical for the control of microbial contamination of
foods and is a significant concern of food preparation and processing industries
and public health agencies. Biofilms are thought
to
serve as reservoirs for
pathogenic and spoilage microorganisms which contaminate foods processed by
unsanitary means (Mosteller and Bishop 1993). Recently, these issues have been
complicated by the hypothesis that bacteria associated with biofilms are
metabolically different from those in broth or in
in
virro
culture and that some
sanitization procedures and disinfectants may, in fact, be ineffective on biofilm
microorganisms (Costerton
et
al.
1987).
For many years the decision to use nonporous plastic rather than wood cutting
board material for
food
preparation, was
based
on
anecdotal or common sense
information (Felix 1993; Raloff 1993).
It
was believed that because wood is
porous
it
would adsorb bacteria from food material more readily than plastic and
could develop into a reservoir
for
foodborne pathogens and spoilage microorgan-
isms. Our study was undertaken because the preference of nonporous over
porous surfaces was recently questioned (Raloff 1993). Specifically, Raloff
TABLE
3.
REMOVAL
OF
BACTERIA
FROM
CUTTING BOARD
SURFACES
AFTER LOW-TEMPERATURE CLEANINWh~'
Number
of cells
recovered Log reduction
Planchetd (CFU/mL) Range
Loglo
Nml NA' Range
W
b
2
F
Plastic, new
3.4
x
lo2
1-7.0
X
lo2
-4.76
-4.52
to
-5.36
Plastic, used
6.0
x
lo2
2-14.0
x
lo2
-4.52
-4.20 to
-5.04
Wood, new
4.4
x
105
2.6-6.2
x
lo5
-1.65
-1.65
to
-1.95
Wood, used
6.5
x
lo4
3-15.0
x
104
-2.49
-2.19
to
-2.88
m
0
%
21
"All
surfaces were inoculated with approximately
2
x
lo7
CFU/mL.
bData is represented as an average of
5
replicates analyzed for each surface.
'All planchets embedded into
VRB
were culture positive.
'Sterile planchets were defined as new and used
as
described in Materials and Methods.
'Log reduction was calculated by dividing the number
of
CFU/mL recovered from each
planchet
(Nm)
by the inoculum size (NA).
TABLE
3.
REMOVAL
OF
BACTERIA
FROM
CUTTING BOARD
SURFACES
AFTER LOW-TEMPERATURE CLEANINWh~'
Number
of cells
recovered Log reduction
Planchetd (CFU/mL) Range
Loglo
Nml NA' Range
W
b
2
F
Plastic, new
3.4
x
lo2
1-7.0
X
lo2
-4.76
-4.52
to
-5.36
Plastic, used
6.0
x
lo2
2-14.0
x
lo2
-4.52
-4.20 to
-5.04
Wood, new
4.4
x
105
2.6-6.2
x
lo5
-1.65
-1.65
to
-1.95
Wood, used
6.5
x
lo4
3-15.0
x
104
-2.49
-2.19
to
-2.88
m
0
%
21
"All
surfaces were inoculated with approximately
2
x
lo7
CFU/mL.
bData is represented as an average of
5
replicates analyzed for each surface.
'All planchets embedded into
VRB
were culture positive.
'Sterile planchets were defined as new and used
as
described in Materials and Methods.
'Log reduction was calculated by dividing the number
of
CFU/mL recovered from each
planchet
(Nm)
by the inoculum size (NA).
168
S.
ABKISHAMI,
B.
TALL,
T.
BRUURSEMA,
P. EPSTEIN
and
D.
SHAH
(1993)
cited
as
yet unpublished results of Cliver and
Ak
which led to the
conclusion that once localized in wood, bacteria die because they are not
recoverable; however, no toxic materials intrinsic
to
wood were identified. Our
goal, therefore, was to evaluate bacterial attachment and viability to plastic and
wooden cutting board surfaces. This study also examined whether either
material contained properties that would promote bacterial growth, and whether
bacteria are retained
on
the surfaces after a low-temperature, non-detergent wash
in a modified commercial dishwasher.
Viability and Recovery
of
Bacteria
from
Cutting Board
Planchets
Our results corroborated the anecdotal concept that bacteria are better retained
on wooden than plastic cutting board surfaces. Although bacterial counts from
both dry and wet wood cutting board surfaces showed no significant reduction
after
5
min of resident drying time, a higher percentage of cells of the inoculum
were retained on the dry wood surface than
on
the water-conditioned wood
surface after
a
2-h resident drying time. These results suggest that saturation
of wood with moisture interfered with the penetration of the inoculum liquid and
cell adherence. In contrast, plastic surfaces retained little inoculum even after
24
h
of
resident drying time. These data are consistent with the concept that
because wood is porous and absorbent it will retain more bacteria than plastic.
The results
also
showed that used wood surfaces are less retentive than new
wood surfaces, suggesting that there is some benefit afforded to wood surfaces
made less porous through use or further refinement.
SEM
of used wood
surfaces showed multiple layers of organic material covering the structural and
vegetative elements of the wood tissues. This would be expected of
a
wood
cutting board surface treated with several applications of mineral oil and used
for food preparation. The oiling of the wooden surface may increase its
hydrophobicity, changing the surface from porous
to
nonporous and preventing
bacteria from adhering.
Because bacteria found in food processing plants are commonly associated
with serum and fat, we investigated the behavior of both the serum-oil-PBDW
and PBS inocula preparations on the different surfaces. The inocula prepared
in PBS were totally absorbed into new wood within
10
min, whereas the serum-
oil inoculum required 2 h to be absorbed into a new dry wood surface.
In
contrast, inocula placed on water-conditioned wood and on plastic surfaces were
not absorbed after 2 or
24
h, respectively, suggesting that the properties of
serum-oil mixture protected the inocula from desiccation on these surfaces and
that the hydrophobic properties of the conditioned new wood and used wood
surfaces were similar to those of plastic surfaces. The miscibility properties of
the serum-oil and water mixture may have acted as a phase partitioning agent,
not allowing the inoculum liquid and cells to penetrate the deeper layers of the
BACTERIA
ON
CUTTING BOARDS
169
water-conditioned wood and plastic surfaces. This would account for the greater
recovery of inoculum cells from these surfaces.
Adherence Assay
and
DVC-SEM
SEM
of inoculated new wooden surfaces showed that bacteria resided within
the structural and vegetative elements of the xylem plant tissues. It may be
speculated that the same water-, food-, and mineral-conducting functions
attributed to xylem plant tissues were a means of redistributing the inoculum
cells throughout the cutting board surface, entrapping them within the xylem
tissues
as
the capillary force of the inoculum liquid dissipated upon drying.
Because of the interpretations reported by Raloff (1993), which suggested that
99.9% of bacteria in
an
inoculum applied to wood were dead, we demonstrated
the viability of cells residing within the xylem tissues in an experiment based
on
the direct viable count procedure described by Kogure
el
af.
(1978). By
modifying the procedure for
SEM
analysis, we took advantage of the inhibitory
effect of nalidixic acid on DNA synthesis. This synthetic antibiotic uncouples
DNA synthesis by inhibiting DNA gyrase, yet allows other synthetic pathways
to continue to operate. Unable to replicate their genome and complete their cell
division cycle, the bacteria become elongated.
In
recent years this technique has
become the cornerstone for defining the viable but nonculturable (VNC)
bacterial state (Colwell
et
al.
1990). Colwell
et
al.
(1990) demonstrated the
existence of
VNC
cells and hypothesized that these VNC cells may play a role
in the epidemiology of human diseases such as cholera, salmonellosis (Roszak
el
al. 1984), and campylobacteriosis (Rollins and Colwell 1986). Our results
demonstrated that most cells residing within the xylem tissues with resident
drying times of up to
2
h were metabolically active and viable.
Microcosm Assay
To
investigate the hypothesis that the cutting board materials had intrinsic
beneficial or deleterious properties, we analyzed microcosms containing
E.
cofi
and wood dust, wood chips, or plastic pieces. We found that these materials
were innocuous, imparting
no
beneficial or deleterious effects
on
the viability
of the cells. These results are consistent with those cited by Raloff (1993).
Effects
of
Low-Temperature
Cleaning
The results of the low temperature cleaning experiment demonstrated that
more bacteria were adherent to wood surfaces than to plastic surfaces and that
cold water washing did not dislodge bacteria from the wooden surfaces as easily
as from the plastic surfaces. These results corroborate the findings of Cliver
170
S.
ABRISHAMI,
9.
TALL,
T.
BRUURSEMA,
P.
EPSTEIN
and
D.
SHAH
and Ak as reported by Raloff (Raloff 1993). Although our experimental design
of low temperature wash may simulate a common household practice of rinsing
cutting boarqs without the use of sanitizers, additional cleaning protocols using
hot water and sanitizers need to be explored to understand the dynamics of
bacterial survival in porous materials, such as wood.
In summary, the results of our study suggest that bacteria adherent
to
wood
surfaces reside in the inner structural and nutrient-conducting elements of xylem
wood tissues; these bacteria were viable after resident times of at least
2
h;
wood had greater retentive properties than plastic even under cold water rinsing
conditions; penetration of the inoculum liquid promoted cell adherence
to
the
wood matrix; and conditioning of wood interfered with bacterial adherence.
ACKNOWLEDGMENTS
We thank John Boos and Company for providing the wood cutting boards, and
the NSF International staff who graciously provided their personal cutting
boards. We acknowledge Gregory C. Earlam, NSF International, for his
technical assistance during the low- temperature cleaning experiments. We
thank Drs. William Obermeyer and Joseph M. Betz, Biological and Organic
Chemistry Branch, Division
of
Natural Products, Center for Food Safety and
Applied Nutrition, Food and Drug Administration, Washington, DC, for their
helpful discussions regarding plant structure; and Walter B. Sisson and James
A. Easterling, FDA, for their help
in
preparing the maple dowel planchets.
REFERENCES
American National Standard/NSF International Standard 6. 1989. Dispensing
freezers. In
Mechanical Sanitizing Test Methods: Heat Treatment Dispensing
Freezers--Microbiological
Performance,
pp. C 1-C2,
D
1 -D2. NSF Intema-
tional, Ann Arbor,
MI.
American Public Health Association (APHA). 1989. Microbiological examina-
tion: heterotrophic plate count.
In
Standard Methods for the Examination
of
Water and Wastewater,
17th ed.
(L.S.
Clesceri, A.E., Greenberg and R.R.
Trussell, eds.) pp.
9-54
-
9-57, APHA, Washington, DC.
Association of Official Analytical Chemists (AOAC). 1990.
Oncia1 Methods
of
Analysis,
15th ed. Secs. 960.09. AOAC, Arlington, VA.
Centers for Disease Control. 1993. Preliminary report: foodbome outbreak of
Escherichia coli
0157:H7 infections from hamburgers-western United
States, 1993. Morbid. Mortal. Weekly Rep.
42,
85-86.
BACTERIA ON CUTTING BOARDS
171
COLWELL, R.R.
et
al.
1990. Environmental aspects of
Vibrio cholerae
in
transmission of cholera. In
Advances
in
Research on Cholera and Related
Diarrheas,
7th ed. (R.B. Sack and Y. Zinnaka, eds.) pp. 327-343, KTK
Scientific Publishers,
Tokyo.
COSTERTON, J.W.
et
al.
1987. Biofilms in nature and disease. AMU. Rev.
Microbiol.
41,
435-464.
COSTERTON, J.W., GEESEY, G.G. and CHENG, K.J. 1978. How bacteria
stick. Sci Am.
238,
86-95.
FELIX, C.W. 1993. Wooden cutting boards see new revival. Food Prot. Rep.
9,
8.
GIESENDORF, B.A., QUINT, W.G.V., HENKENS, M.H.C., STEGEMAN,
H.,
HUF, F.A. and NIESTERS, H.G.M. 1992. Rapid and sensitive detection
of
Camplyobacter
spp. in chicken products by using polymerase chain
reaction. Appl Environ. Microbiol.
58,
3804-3808.
KARCH, H., HEESEMAN, J., LAUFS, R., O’BRIEN, A.D., TACKET, C.O.
and LEVINE,
M.M.
1987. A plasmid of enterohemorrhagic
Escherichia coli
0157:H7 is required for expression
of
a new fimbrial antigen and for
adhesion to epithelial cells. Infect. Immun.
55,
451-461.
KOGURE, K., SIMIDU, U. and TAGA, N. 1978. A tentative direct microscop-
ic method for counting living marine bacteria. Can. J. Microbiol.
25,
415-
420.
LOUIE, K.K., PACCAGNELLA, A.M., OSEI, W.D., LIOR, H., FRANCIS,
B
.
J
,
and OSTERHOLM
,
M .T
.
1993.
Salmonella
serotype
tennessee
in
powdered milk products and infant formula--Canada and United States, 1993.
Morbid. Mortal. Weekly Rep.
42,
516-517.
MOSTELLER, T.M. and BISHOP, J.R. 1993. Sanitizer efficacy against
attached bacteria in a milk biofilm. J. Food Prot.
56,
34-41.
NSF INTERNATIONAL STANDARD. 12. 1992. Automatic ice making
equipment. In
In-Place Cleaning/Sanitizing Test Methods,
pp.
AI-A2. NSF
International, AM Arbor, MI.
O’BRIEN, A.D. and HOLMES,
R.K.
1987. Shiga and shiga-like toxins.
Microbiol. Rev.
52,
206-220.
OOSTEROM, J., NOTERMANS,
S.,
KARMAN, H. and ENGELS, G.B.
1983. Origin and prevalence
of
Campylobacter jejuni
in poultry processing.
J.
Food Prot.
46,
339-344.
RALOFF,
J.
1993. Wood wins: plastic trashed for cutting meat. Science News
ROLLINS, D.M. and COLWELL, R.R. 1986. Viable but nonculturable stage
of
Campylobacter jejuni
and its role in survival in the natural aquatic
environment. Appl. Environ. Microbiol.
52,
53 1-538.
143,
84-85.
172
S.
ABRISHAMI,
B.
TALL,
T.
BRUURSEMA,
P.
EPSTEIN
and
D.
SHAH
ROSZAK, D.B., GRIMES, D.J. and COLWELL, R.R.
1984.
Viable but
nonrecoverable stage of
Salmonella enteritidis
in aquatic systems. Can.
J.
Microbiol.
30,
334-338.
STROCKBINE, N.A., MARQUES, L.R.M., NEWLAND, J.W., SMITH,
H.W., HOLMES,
R.K.
and O’BRIEN, A.D.
1986.
Two toxin-converting
phages from
Escherichia
coli
0157:H7
strain
933
encode antigenically
distinct toxins with similar biologic activities. Infect. Immun.
53,
135-140.
ADDENDUM
While this manuscript was in press, two papers on cutting board microbiolo-
gy by Dr. Cliver and colleagues have appeared in print:
AK,
et al.
1994.
J. Food Prot.
57,
16-22; 23-30,
