- Email: [email protected]
Isolation of indigenous bacteria from a cafeteria kitchen and their biofilm formation and disinfectant susceptibility
Accepted Manuscript Isolation of indigenous bacteria from a cafeteria kitchen and their biofilm formation and disinfectant susceptibility Eun Seob Lim, Jang Eun Lee, Joo-Sung Kim, Ok Kyung Koo PII:
S0023-6438(16)30729-0
DOI:
10.1016/j.lwt.2016.11.060
Reference:
YFSTL 5869
To appear in:
LWT - Food Science and Technology
Received Date: 21 June 2016 Revised Date:
14 October 2016
Accepted Date: 20 November 2016
Please cite this article as: Lim, E.S., Lee, J.E., Kim, J.-S., Koo, O.K., Isolation of indigenous bacteria from a cafeteria kitchen and their biofilm formation and disinfectant susceptibility, LWT - Food Science and Technology (2016), doi: 10.1016/j.lwt.2016.11.060. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Running head: Biofilm Formation and Disinfectant Susceptibility
2
Isolation of indigenous bacteria from a cafeteria kitchen and their biofilm formation
3
and disinfectant susceptibility
4
Eun Seob Lim1, Jang Eun Lee1,2, Joo-Sung Kim1,3, Ok Kyung Koo4,5*
5
1
6
Republic of Korea
7
2
8
si, Gyeonggi-do, Republic of Korea
9
3
RI PT
1
Department of Food Biotechnology, Korea University of Science & Technology, Daejeon,
SC
Traditional Alcoholic Beverages Research Team, Korea Food Research Institute, Seongnam-
10
Republic of Korea
11
4
12
Gyeongsangnam-do, Republic of Korea
13
5
14
Gyeongsangnam-do, Republic of Korea
M AN U
Food Safety Research Team, Korea Food Research Institute, Seongnam-si, Gyeonggi-do,
Department of Food and Nutrition, Gyeongsang National University, Jinju,
TE D
Institute of Agriculture & Life Science, Gyeongsang National University, Jinju,
15
EP
16
*Corresponding author:
18
Ok-Kyung Koo, Ph.D.
19
Department of Food and Nutrition
20
Gyeongsang National University
21
501 Jinjudae-ro,
22
Jinju, Gyeongsangnam-do, 52828
23
Tel: +82-55-772-1441
24
Fax: +82-55-772-1439
25
Email: [email protected]
AC C
17
-1-
ACCEPTED MANUSCRIPT Abstract
27
Bacterial biofilm formation in foodservice facilities is a continuous cross-contamination risk
28
through survival and persistence despite disinfectant treatments. In this study, we evaluated
29
biofilm formation and disinfectant susceptibility of 178 strains obtained from a cafeteria
30
kitchen and 70 foodborne pathogens and analyzed results by multivariate data analyses. A
31
total of 23 areas in a cafeteria kitchen were selected for bacterial isolation and identification.
32
The capacity for biofilm formation was tested using a crystal violet assay, and disinfectant
33
susceptibility was examined using an agar well diffusion assay and resazurin reduction assay.
34
The most frequently isolated genera were Bacillus (33%), Acinetobacter (17%), Kocuria
35
(12%) and Staphylococcus (5%). The genus Bacillus showed the strongest capacity of the
36
biofilm formation. The foodborne bacteria exhibited a wide range of susceptibility to
37
disinfectants, such as sodium hypochlorite, hydrogen peroxide, benzalkonium chloride, lactic
38
acid and citric acid. However, the susceptibilities changed after biofilm formation in a strain-
39
dependent manner, and the relative resistance levels changed among the isolates. Overall, this
40
study will be a great resource for selecting and using disinfectants in foodservice facility
41
hygienic practices.
EP
TE D
M AN U
SC
RI PT
26
42
Keywords: Cafeteria kitchen, Biofilm, Crystal Violet Assay, Disinfectant, Resazurin Reduction
44
Assay
45
AC C
43
-2-
ACCEPTED MANUSCRIPT 46
1. Introduction Food hygiene control and safety management is essential and critical for protecting
48
the public from foodborne illnesses. The HACCP program and other adequate programs have
49
been globally accepted for systemically preventing and controlling food safety hazards
50
(Grönholm, Wirtanen, Ahlgren, & Sjöberg, 1999; Notermans, Gallhoff, Zwietering, & Mead,
51
1995; Notermans & Mead, 1996). Despite efforts to minimize contamination, massive
52
foodborne outbreaks occur throughout the world (Much, Pichler, Kasper, & Allerberger, 2009;
53
Naimi et al., 2003). Such outbreaks can be due to inadequate use of cleaning and sanitizing
54
practices as well as naturally occurring biofilms (Yang, Kendall, Medeiros, & Sofos, 2009).
M AN U
SC
RI PT
47
Biofilm is a bacterial community that adheres to biotic or abiotic surfaces and
56
produces exopolymeric substances to protect from the environmental stress including
57
antibiotic treatment or disinfection (Rayner, Veeh, & Flood, 2004). Previous studies have
58
confirmed the ubiquity of biofilm in household surfaces and suggest the importance of
59
biofilm in household hygiene (Rayner et al., 2004), in various food processing plants (Srey,
60
Jahid, & Ha, 2013) and even in fresh produce and meat (Jessen & Lammert, 2003). In
61
phyllosphere microbiology, 30 to 80% of the total bacterial population in a plant is related to
62
biofilm formation (Lindow & Brandl, 2003). Costerton et al. (1987) claimed that
63
approximately 99% of all bacteria in nature exist as a biofilm and that most microorganisms
64
can survive and contaminate biotic or abiotic surfaces. Further, 80% of bacterial infections in
65
the USA are associated with biofilm (Costerton et al., 1987; Srey et al., 2013).
EP
AC C
66
TE D
55
Bacterial biofilms in food ingredients or food handlers transfer bacteria to food
67
contact surfaces, to other food handlers or the food processing environment, and contaminate
68
the final food product (Pérez-Rodríguez, Valero, Carrasco, García, & Zurera, 2008). Because
-3-
ACCEPTED MANUSCRIPT sessile bacteria can survive and persist in the environment, adequate and appropriate
70
disinfectant use is necessary to minimize the risk of foodborne pathogen contamination (Gil,
71
Selma, López-Gálvez, & Allende, 2009). Naturally acquired bacterial biofilms have been the
72
main target of intervention technologies because they protect pathogenic bacteria and the
73
sessile bacterial cells that were embedded in the biofilm were 10-1000 times more resistant to
74
antibiotic treatment (Davies, 2003; Sanchez-Vizuete, Orgaz, Aymerich, Le Coq & Briandet,
75
2015). Therefore, an increasing resistance to disinfectants is a critical issue in food safety
76
(Meira, de Medeiros Barbosa, Alves Aguiar Athayde, de Siqueira-Júnior, & de Souza, 2012).
SC
RI PT
69
Until now, disinfectant susceptibility in bacterial biofilms was individually evaluated
78
using select disinfectants. We isolated bacteria from a foodservice facility and evaluated their
79
capacity for biofilm formation and disinfectant resistance/susceptibility compared with
80
foodborne pathogens from our culture collection. We also compared the relationship between
81
sessile and planktonic bacteria in response to the same disinfectants and compared the
82
correlations among the bacterial strains and different disinfectants. This evaluation will
83
provide a valid result through comparing bacteria under different antibiotic and surface-
84
attached conditions.
87 88
TE D
EP
86
2. Materials and Methods
AC C
85
M AN U
77
2.1 Bacterial strains and growth conditions Seventy foodborne pathogens including E. coli O157:H7, Listeria monocytogenes,
89
Salmonella and Staphylococcus aureus were obtained from the culture collection of the Food
90
Safety Research Team in the Korea Food Research Institute (Table S1). Bacteria was
91
inoculated in Tryptic Soy Broth (Becton Dickinson Co, Franklin Lakes, New Jersey, USA)
-4-
ACCEPTED MANUSCRIPT 92
and incubated at 37°C for 16 to 18 h in shaking incubator. The cultures were maintained in
93
15% glycerol at -80°C until use.
95
2.2 Isolating microorganisms on food-contact surfaces
RI PT
94
A cafeteria kitchen in a foodservice facility with an average daily attendance (ADA)
97
greater than 250 people was used to isolate bacteria. Twenty-three different food-contact
98
surfaces, including cooking utensils, kitchen appliances and the cooking area, were used to
99
isolate background microflora, and the sampling was performed after daily cooking and
100
cleaning (Fig. S1). Each surface (a total of 100 cm2 area using one 10 × 10 cm or four 5 × 5
101
cm stainless steel frames depending on surface conditions) was swabbed 10 times in vertical,
102
horizontal and diagonal directions, respectively, using Pipette swab plus (3M, Minnesota,
103
USA) in 10 mL of buffered peptone water (BPW). The researchers wore sterile latex gloves
104
during collection to minimize cross-contamination from hands. Each sample was vortexed for
105
1 min to release the bacteria attached to the swab and plated onto plate count agar (PCA, BD)
106
with the appropriate dilution and incubated at 37°C for up to 48 h. The collected samples
107
were further processed within two hours for quantification and isolation.
109 110
M AN U
TE D
EP
AC C
108
SC
96
2.3 Isolation and Identification Bacteria were isolated from the colonies grown on PCA. Colony selection was
111
performed based on the morphology of each colony and the food-contact surface area and
112
178 strains were isolated. Each colony was re-streaked on PCA and inoculated in tryptic soy
113
broth (TSB, Merck & Co., Kenilworth, New Jersey, USA) at 37°C for overnight. The
114
overnight culture was mixed with 15% glycerol and kept at -70°C until next use. In order to -5-
ACCEPTED MANUSCRIPT identify the isolates, the overnight culture was centrifuged at 9,400 ×g for 5 min and the
116
pellet was used for the genomic DNA extraction using DNeasy Blood & Tissue kit (Qiagen,
117
Hilden, Germany) and the DNA concentration was measured using NanoVue (GE Healthcare,
118
Buckinghamshire, UK). The 16S rRNA gene of each DNA was amplified using 27F (5'-
119
GAGTTTGATCMTGGCTCAG-3'),
120
primers (Macrogen, Seoul, Korea). DNA templates, 10 pmol/µL of primers and PCR-grade
121
water were prepared with Takara Ex Taq version 2.0 (Takara, Kusatsu, Japan) to a total of 25
122
µL for each reaction. The amplification was performed by following program; initial
123
denaturation at 95°C for 5 min, then 30 cycles of 1) denaturation at 94°C for 1 min, 2)
124
annealing at 55°C for 30 s and 3) extension at 72°C for 1 min, and the final elongation step at
125
72°C for 7 min. The amplified PCR product was loaded on 0.8% agarose gel in TAE (40 mM
126
Tris·HCl, 40 mM acetate, 1.0 mM EDTA) buffer stained with Staining STAR (DyneBio,
127
Seongnam, Korea) and the amplification was confirmed with ultraviolet transilluminator
128
using Gel DocTM EZ Imager (Bio-Rad, Richmond, CA, USA). The amplicon was further
129
purified by QIAquick PCR purification kit (Qiagen). The sequencing was performed by
130
Macrogen with 512F (5'-CCAGCAGCCGCGGTAAT-3') for the downstream of 16S rRNA
131
gene and 512R (5'-ATTACCGCGGCTGCTGG-3') primers for the upstream of 16S rRNA
132
gene. The sequencing results were analyzed with the EzTaxon (Kim et al., 2012) on the basis
133
of 16S rRNA sequence data (Table S2).
135
(5'-GGTTACCTTGTTACGACTT-3')
EP
TE D
M AN U
SC
1492R
AC C
134
and
RI PT
115
2.4 Biofilm formation on a microtiter plate-crystal violet assay
136
All stains from culture collection and kitchen isolates were inoculated in TSB and
137
incubated for 16-18 h at 37°C. The overnight culture was diluted to approximately 107
-6-
ACCEPTED MANUSCRIPT CFU/mL in TSB, and 200 µL of the culture was inoculated in a 96-well plate at 37°C for 48 h
139
to facilitate biofilm formation on the microtiter plate. After incubation, the culture medium
140
was carefully removed and washed with PBS (phosphate-buffered saline, pH 7.0) once. After
141
washing the microtiter plate with biofilm-forming cells, 200 µL of a 1% crystal violet (CV)
142
solution (bioWORLD, Dublin, Ohio, USA) was added, and the microtiter plate was incubated
143
for 30 min at room temperature. After washing with PBS 3 times, absolute ethanol 200 µL
144
was added and incubated for 15 min at RT to destain the CV. From the destaining solution,
145
100 µL was transferred to a new 96-well plate, and the absorbance was measured at 595 nm
146
using Infinite® 200 PRO NanoQuant (Tecan Group Ltd., Männedorf, Switzerland).
M AN U
SC
RI PT
138
147 148
2.5 Disinfectant assay using non-biofilm bacteria
Twenty-eight strains of kitchen isolates and 18 strains from culture collection were
150
selected for the following experiment based on the CV assay that resulted in strong biofilm
151
forming ability (over the absorbance of 0.7 for the isolates from this study and 0.3 from the
152
culture collection). The bactericidal activity of disinfectants was determined using the agar
153
well diffusion method with Mueller Hinton II agar (BD). An overnight culture of the isolate
154
was streaked using a cotton swab to prepare a lawn plate. A well was generated in the middle
155
of each plate using a sterilized 8 mm cork borer (K-ACE, Jongro-gu, Seoul, Korea) and
156
inoculated with 100 µl of a select disinfectant. The disinfectant concentrations were as
157
follows: 2,000 mg/L NaClO (Junsei, Tokyo, Japan), 500 mg/L benzalkonium chloride (BAC,
158
Kukbo Science, Cheongju, Korea), 2,000 mg/L H2O2 (Daejung, Siheung, Korea), 10% lactic
159
acid (LA, Kanto, Tokyo, Japan), and 10% citric acid (CA, Junsei). The plates were then
AC C
EP
TE D
149
-7-
ACCEPTED MANUSCRIPT 160
incubated at 37°C for 18-24 h. The inhibition activity was measured based on the cleared
161
zone.
162
2.6 Disinfectant assay using biofilm-forming bacteria
RI PT
163
The bactericidal activity to disinfectants after biofilm formation was determined
165
using resazurin reduction assay. Fifty-six strains were inoculated in TSB and incubated for
166
16-18 h at 37°C. The initial part of the experiment was performed the same as CV assay until
167
washing with PBS after incubation of the culture. After washing the microtiter plate with
168
biofilm-forming cells, the PBS wash solution was removed, 200 µL of disinfectant was added,
169
and the cells were incubated for 10 min at room temperature. The disinfectant concentrations
170
were as follows: 50 mg/L NaClO (Junsei), 100 mg/L BAC (Kukbo Science), 2,000 mg/L
171
H2O2 (Daejung), 0.25% LA (Kanto), and 0.25% CA (Junsei), and PBS was used as the
172
negative control. Next, the disinfectant was removed, and 200 µL of Dey-Engley Neutralizing
173
Broth (Sigma-Aldrich, Saint Louis, MO, USA) was added, and the cells were incubated for 5
174
min at room temperature. The neutralizing broth was removed, and the 96-well plate was
175
washed with PBS to remove the neutralizing agent. Thereafter, 200 µL of 0.001% (wt/vol)
176
resazurin (Sigma-Aldrich) was added to each well, and the cells were incubated for 60 min at
177
37°C under dark conditions. After incubation, the fluorescence (λex 570 nm, λem 590 nm) was
178
measured using SpectraMax® M2 (Molecular Devices®, Sunnyvale, CA, USA).The
179
reduction level (%) was calculated using the formula below.
180
Reduction % =
AC C
EP
TE D
M AN U
SC
164
′
′
x 100 .
181
where A = RFU of test wells, A’ = RFU of negative control wells, and res. = RFU of
182
resazurin blue oxidized form. -8-
ACCEPTED MANUSCRIPT 183 184
2.5 Multivariate data analysis The resulting data were analyzed using SIMCA-P version 12.0 (Umetrics, Umeå,
186
Sweden), and a mean-centered scaling method was applied for multivariate statistical
187
analyses. A principal component analysis (PCA), an unsupervised pattern recognition method,
188
was performed to examine the intrinsic variation in the dataset.
190
SC
189
RI PT
185
2.6 Statistical analysis
The result was evaluated with one-way analysis of variance (ANOVA), followed by
192
Tukey Honestly Significantly Difference (HSD) tests with a significance level of 0.05
193
(p<0.05). All statistical analyses were conducted using Minitab 17 (Minitab Inc.,
194
Pennsylvania, USA).
M AN U
191
196 197
3. Results
TE D
195
3.1 Bacteria isolation and identification from food-contact surfaces Twenty-three different food-contact surfaces in a cafeteria kitchen (Fig. S1) were
199
selected, and bacteria were isolated from the surfaces to evaluate the capacity for biofilm
200
formation and disinfectant susceptibility. The total aerobic count varied from 1.60 log
201
CFU/cm2 in stainless steel trays to 6.85 log CFU/cm2 in aprons, and 178 bacterial strains were
202
isolated and identified. The floor contained the most diverse bacteria with 9 genera and 15
203
species, followed by the pretreatment countertops, the cold room, the countertop for the
204
completed menu, and the cutting board for vegetables (Table 1). Five surfaces contained
205
Bacillus, including frying pots, dishwashers, sinks for dishwashing, stainless steel trays and
AC C
EP
198
-9-
ACCEPTED MANUSCRIPT fans. Bacillus spp. was the most frequently isolated bacteria from 19 surfaces out of 23
207
surfaces (Table 1), and 61 strains (33%) were identified as Bacillus spp. Kocuria and
208
Acinetobacter spp. were isolated on 12 and 9 surfaces with 23 strains (12%) and 31 strains
209
(17%), respectively. Staphylococcus spp. was isolated at 5% of the total isolates from the
210
pretreatment countertops, food waste container and completed menu cutting board followed
211
by Pantoea at 4% and Sphingomonas at 3%. Pseudomonas spp. was isolated from the
212
pretreatment countertops and floor and accounted for 2% of total isolated strains. In addition,
213
Enterobacter, Cloacibacterium and Chryseobacterium spp. were isolated and accounted for
214
2%, respectively.
215 216
3.2 Biofilm formation
217
M AN U
SC
RI PT
206
A total of 248 bacterial strains, including bacteria isolated from a cafeteria kitchen
218
and
219
Salmonella and Staphylococcus aureus, were used to evaluate the biofilm-forming capacity
220
using a crystal violet assay (Fig. 1). As a result, each strain exhibited a distinctively different
221
profile for biofilm formation with an absorbance range between zeros to 4.0. The results were
222
classified into three categories based on absorbance: weakly adherent <1.0, 1.0 < moderately
223
adherent <3.0, and 3.0< strongly adherent. Over 50% of the strains were in the genera
224
Acinetobacer, Kocuria, Staphylococcus and Bacillus, which strongly or moderately adhered
225
with the average absorbance 1.46, 1.97 and 1.41, respectively. In particular, over 90% of
226
the Bacillus strains adhered strongly with an average absorbance of 2.88. On the other hand,
227
approximately 85% of the foodborne pathogens from culture collection weakly adhered with
culture
collection
including
E.
coli O157:H7,
Listeria monocytogenes,
AC C
EP
TE D
from
- 10 -
ACCEPTED MANUSCRIPT 228
an average absorbance of 0.39 – 0.63, and a few strains were exceptionally strong at forming
229
biofilm.
230
3.3 Disinfectant susceptibility tests
RI PT
231
The agar well diffusion assay (AWD) was performed to evaluate the
233
susceptibility/resistance of the bacteria to five disinfectants (Fig. 2). This method was
234
selected in order to further perform multivariate data analysis and the treated concentration
235
was fixed to limit the comparison factor. The range of the zone of inhibition for each
236
disinfectant was 13 to 33.5 mm, 8.6 to 26.8 mm, 9.7 to 28.6 mm, 17.2 to 35.1 mm, 12.4 to 35
237
mm on sodium hypochloride, benzalkonium chloride, hydrogen peroxide, lactic acid and
238
acetic acid, respectively (Table S3). We observed disinfectant susceptibility with a relatively
239
lower deviation for each tested genus than resazurin reduction assay. Among the disinfectants,
240
LA and CA were most effective against the isolates with a more than 20 mm average
241
inhibition zone. Acinetobacter exhibited the greatest disinfectant sensitivity to NaClO and
242
H2O2, and Kocuria was significantly more resistant to the two disinfectants; however, the
243
reverse was observed for BAC, LA and CA. The disinfectant susceptibility of L.
244
monocytogenes was similar to Kocuria. Bacterial susceptibility to disinfectants was
245
consistent with the non-biofilm-forming bacteria, and the results were genus or species
246
specific rather than strain specific. For BAC, LA and CA, the antimicrobial activity was
247
divided into two groups: Gram-positive and Gram-negative. Kocuria, L. monocytogenes, and
248
S. aureus were relatively sensitive to the three disinfectants, while Acinetobacter, E. coli
249
O157:H7 and Salmonella were significantly resistant.
250
AC C
EP
TE D
M AN U
SC
232
The resazurin reduction assay (RES) was used to evaluate the viability of biofilm-
- 11 -
ACCEPTED MANUSCRIPT forming cells after a disinfectant treatment (Fig. 2). The range of reduction rate by RES was
252
relatively higher than AWD result and mostly strain dependent. The disinfectant resistance
253
also changed after the biofilm had formed. For example, most Kocuria strains were the most
254
resistant to all disinfectants after the biofilm had formed, while most Bacillus isolates became
255
the most sensitive to BAC among the biofilm-forming isolates. Biofilm-forming L.
256
monocytogenes and E. coli O157:H7 became the most sensitive to NaClO and H2O2,
257
respectively. However, Salmonella and S. aureus became the most resistant bacteria to H2O2,
258
and LA and CA, respectively, after the biofilm had formed.
SC
RI PT
251
To interpret the resistance variations using the five disinfectants and different
260
bacteria strains, a principal components analysis (PCA) of pattern recognition was initially
261
applied to the dataset (Fig. 3). The resistances were based on the AWD assay for each
262
bacterium strain and were partially differentiated using PCA score plots (Fig. 3A, B); no
263
significant differentiations were observed in the RES assay (Fig. 3C, D). We observed a
264
65.84% difference in the first coordinate for AWD, and the data were clearly separated into
265
Gram-positive or Gram-negative groups. L. monocytogenes and Kocuria presented a
266
particularly distinctive skew for relative sensitivity to BAC, CA and LA. The second
267
coordinate explained 17.67% of the difference in the relative sensitivities of S. aureus to
268
H2O2. Unlike AWD, the RES results were not clustered based on bacteria under genus or
269
species level. While the Salmonella and Acinetobacter bacterial strains showed similar
270
susceptibilities using AWD, the bacteria were separated throughout the PCA analysis using
271
the RES data. For the RES data, the first principle component 57.92% of variance distribution
272
for Kocuria showed a relative resistance to BAC, LA and CA, which was the opposite result
273
from the AWD. Bacillus also changed from exhibiting a relatively moderate activity to a
AC C
EP
TE D
M AN U
259
- 12 -
ACCEPTED MANUSCRIPT 274
relative sensitivity to RES.
275
4. Discussion
277
Foodservice facilities are required to ensure that the facilities follow food safety and
RI PT
276
278
hygiene
279
sanitation/disinfection processes because foodservice facilities are the perfect niche for
280
bacteria to thrive due to temperature, humidity and nutrient conditions (Staskel et al., 2007;
281
Stoodley et al., 2002). Bacteria that are “native” to food-contact surfaces have been evaluated
282
in previous studies, such as in the kitchen sink drain (Furuhata, Ishizaki, & Fukuyama, 2010)
283
and cutting board (Abdul-Mutalib et al., 2015), and on household surfaces, including the
284
kitchen counter and sink, bathroom floor, and toilet seat (Gajanan & Singh, 2013). Liu and
285
others (2013) isolated 23 genera out of 117 bacterial strains from fresh-cut produce
286
processing plants (Liu et al., 2013). In this study, we evaluated biofilm formation in bacterial
287
isolates from a foodservice facility in 23 different food-contact and nonfood-contact surfaces
288
using a culturing method. Because the experiment was performed after the regular cleaning
289
and sanitation process, the isolated bacteria should be resistant to sanitation or protected from
290
the sanitation process through biofilm formation.
through
regular
and
appropriate
decontamination
and
EP
TE D
M AN U
SC
practices
Aprons were the most contaminated food-contact surface, and Bacillus, Rhizobium,
292
Kocuria and Pseudoxanthomonas were isolated (Table 1). Aprons are required for workers’
293
personal hygiene, but apron sanitation may have been under-estimated due to indirect food-
294
contact activities. However, aprons can be a great mediator for microorganism cross-
295
contamination or transfer due to frequent direct or indirect food contact. Therefore,
AC C
291
- 13 -
ACCEPTED MANUSCRIPT 296
appropriate sanitation practices for aprons are necessary to minimize the cross-contamination
297
risk. The most diverse bacteria isolation was observed in the floor and included
299
Aeromonas, Kocuria, Bacillus, Pseudomonas and Cloacibacterium (Table 1). The floor does
300
not directly interact with food; however, the floor can collect residue from food ingredients,
301
worker hygienic practices and environmental contaminants to create a diverse microbial
302
community. Liu and others (2013) isolated Bacillus, Enterobacter, Rhnella, Ralstonia and
303
Pseudomonas from nonfood-contact surfaces that are known potential soil, plant, animal and
304
water sources. The sampling method also affected diversity through swabbing various
305
locations with a 5x5 cm surface area.
M AN U
SC
RI PT
298
Bacillus spp. were the most dominant isolate at 33% and were detected on frying
307
pots, dishwashers, fans and other surfaces. Bacillus is ubiquitous and can survive stringent
308
conditions through spore formation (Turnbull, 1996). Therefore, whether Bacillus was
309
transferred from ingredients or food handlers, the bacteria survive under at a high temperature,
310
at low moisture and under sanitizing conditions; thus, it was the most frequently isolated
311
bacteria in the foodservice facilities. Opportunistic pathogens, such as Acinetobacter
312
baumannii, Staphylococcus epidermidis, and Kocuria varians, were isolated; thus, proper
313
handling practice is necessary to minimize growth and cross-contamination (Ryan, 2004).
314
The Enterobacteriaceae family is a hygienic indicator and includes Pantoea, Enterobacter,
315
Raoultella, Escherichia, and Leclercia, which were isolated from 8 surfaces, including cold
316
rooms, countertops, sinks, spice racks, plastic wicker trays, and cutting boards (Table 1). The
317
corresponding surfaces are frequently associated with water and raw ingredients in the pre-
318
treatment area. However, this type of contamination can be controlled during the cooking
AC C
EP
TE D
306
- 14 -
ACCEPTED MANUSCRIPT 319
process. The countertop for the completed menu also contained Enterobacter, and salads with
320
fresh produce are at risk for contamination. Crystal violet (CV) is a dye that binds negatively charged cell surface molecules and
322
exopolysaccharide (EPS) and can efficiently detect bacterial presence and quantify the
323
biomass of biofilm-forming cells (Peeters, Nelis, & Coenye, 2008). In this study, most
324
isolates exhibited a moderate or strong biofilm-forming capacity. The result could be due to
325
the isolation process after the daily cleaning and sanitation practices and the bacteria should
326
have survived the antibiotic treatment. The bacteria were recovered by strongly binding the
327
surfaces or through the protection from food debris or the biofilm EPS. While the tested
328
foodborne pathogens adhere less than the isolates, the shield created by the biofilm
329
community or food debris can increase the chances of survival on food-contact surfaces and
330
eventually cause foodborne illnesses. The CV assay is a high-throughput screening method
331
for biofilm; however, the staining method cannot differentiate viable and non-viable bacteria
332
because it detects the entire biomass. Therefore, the resazurin assay was performed to
333
measure cell viability. The correlation between the CV and resazurin assays was moderate
334
with an r value of 0.57, which validates the reasonable relationship between biomass content
335
and viability (data not shown).
EP
TE D
M AN U
SC
RI PT
321
Depending on the disinfectant, the antimicrobial activity varied for the target bacteria.
337
The surfactant disinfectant BAC primarily kills bacteria through membrane damage where
338
the hydrophilic cationic region destabilizes the bacterial surface through electrostatic
339
interactions, and the hydrophobic region penetrates the cell membrane and causes cell
340
leakage. The cell wall of Gram-positive bacteria is composed of peptidoglycans and teichoic
341
acids that are negatively charged; thus, BAC is more effective against Gram-positive bacteria
AC C
336
- 15 -
ACCEPTED MANUSCRIPT (Fazlara & Ekhtelat, 2012; Fraise, Maillard, & Sattar, 2008). Organic acids, such as lactic
343
acid and citric acid, are active permeabilizers and are effective against Gram–positive, while
344
acetic acid exhibits the opposite result (Lee, Cesario, Owens, Shanbrom, & Thrupp, 2002).
345
Whilst lactic acid can weaken a cell wall by binding the phosphate and carbonyl group
346
protons, it is active against Gram-negative at a low pH. Our results confirm previously
347
accepted antibacterial activities before biofilm forms. However, when bacterial biofilms
348
formed, the susceptibility was transformed from clustered activity to strain-dependent activity.
349
Additionally, the relative susceptibility amongst the bacterial group shifted to relatively
350
resistant or sensitive. When bacterial cells are attached to the abiotic surface, the physiology
351
of the bacteria changes by the stressed condition such as limited nutrients on the surface and
352
the bacteria can cross-protect them from other stresses including disinfectant (Chen & Jiang,
353
2014). For instance, the reduction in bacterial growth rate and changes to its cell membrane
354
composition cause a subsequent decrease in metabolism of antimicrobial agents, stress
355
response modulation, quorum sensing and EPS production, which can limit disinfectant
356
penetration of the target bacteria (Malik & Grohmann, 2012; Otter et al., 2015). In this study,
357
we did not compare the reduced susceptibility using a dose-response or quantify viability
358
using culture method, but we compared the relative responses among bacteria to each
359
disinfectant. Therefore, the absolute difference for each bacterium cannot be compared.
360
However, the bacterial responses to disinfectants under the same concentration were
361
comparable, and we can understand the transformation of relative resistance in each
362
bacterium to determine the pertinent bacterial target for each disinfectant, which can be
363
important information for best hygienic practices.
AC C
EP
TE D
M AN U
SC
RI PT
342
364
- 16 -
ACCEPTED MANUSCRIPT 365
5. Conclusion Consumers’ demand on better quality with safety assured food product has been
367
increased, which can be satisfied using an effective safety program and intervention
368
technology. However, current control steps cannot efficiently kill microorganisms, and the
369
inadequate disinfectant process can limit microbial quality control and promote potential
370
contamination by foodborne pathogens through biofilm formation. This comprehensive study
371
suggests that the relative disinfectant susceptibility of bacteria changes upon forming a
372
biofilm, we must consider re-evaluating the reference bacteria for testing disinfectant
373
activities against sessile bacteria.
M AN U
SC
RI PT
366
374
Acknowledgment
376
This research was funded by a research grant from the Korea Food Research Institute under
377
Grant E0142102-02 and from Gyeongsang National University under Grant 2016-0194.
378
380
References
Abdul-Mutalib, Nordin, N. A., S. A., Osman, M., Ishida, N., Tashiro, K., Sakai, K.,
EP
379
TE D
375
Tashiro, Y., Maeda, T., & Shirai, Y. (2015). Pyrosequencing analysis of microbial
382
community and food-borne bacteria on restaurant cutting boards collected in Seri
383
Kembangan, Malaysia, and their correlation with grades of food premises. International
384
Journal of Food Microbiology, 200, 57–65.
385 386 387
AC C
381
Chen, Z., & Jiang, X. (2014). Microbiological safety of chicken litter or chicken litter-based organic fertilizers: a review. Agriculture, 4, 1-29. Costerton, J. W., Cheng, K. J., Geesey, G. G., Ladd, T. I., Nickel, J. C., Dasgupta, M., &
- 17 -
ACCEPTED MANUSCRIPT 388
Marrie, T. J. (1987). Bacterial biofilms in nature and disease. Annual Review of
389
Microbiology, 41, 435–464.
391
Davies, D. (2003). Understanding biofilm resistance to antibacterial agents. Nature Reviews. Drug Discovery, 2, 114–122.
RI PT
390
Fazlara, A., & Ekhtelat, M. (2012). The Disinfectant Effects of Benzalkonium Chloride on
393
Some Important Foodborne Pathogens. American-Eurasian Journal of Agricultural &
394
Environmental Sciences, 12, 23–29.
395
SC
392
Fraise A. P., Lambert, P. A., & Maillard, J. Y. (2008). Russell, Hugo & Ayliffe’s Principles and Practice of Disinfection, Preservation & Sterilization (4th ed.). England: Wiley-
397
Blackwell.
399 400 401 402
Furuhata, K., Ishizaki, N., & Fukuyama, M. (2010). Characterization of heterotrophic bacteria isolated from the biofilm of a kitchen sink. Biocontrol Science, 15, 21–25. Gajanan, M. V, & Singh, O. V. (2013). Isolation of microbes from common household
TE D
398
M AN U
396
surfaces. Journal of Emerging Investigators, 1, 1–7. Gil, M. I., Selma, M. V., López-Gálvez, F., & Allende, A. (2009). Fresh-cut product sanitation and wash water disinfection: Problems and solutions. International Journal of
404
Food Microbiology, 134, 37–45. Grönholm, L., Wirtanen, G., Ahlgren, K., & Sjöberg, K. N. A. (1999). Screening of
AC C
405
EP
403
406
antimicrobial activities of disinfectants and cleaning agents against foodborne spoilage
407
microbes. Z Lebensm Unters Forsch A, 208, 289–298.
408 409 410
Jessen, B., & Lammert, L. (2003). Biofilm and disinfection in meat processing plants. International Biodeterioration & Biodegradation, 51, 265–269. Kim, O. S., Cho, Y. J., Lee, K. H., Yoon, S. H., Kim, M. C., Na, H. S., Park, S. C., Jeon, Y.
- 18 -
ACCEPTED MANUSCRIPT S., Lee, J. H., Yi, H. N., Won, S. H., & Chun, J. S. (2012). Introducing EzTaxon-e : a
412
prokaryotic 16S rRNA gene sequence database with phylotypes that represent
413
uncultured species. International Journal of Systematic and Evolutionary
414
Microbiology, 62, 716–721.
417 418 419
Activity of Citrate and Acetate. Nutrition, 18, 665–666.
Lindow, S. E., & Brandl, M. T. (2003). Microbiology of the Phyllosphere. Applied and Environmental Microbiology, 69, 1875–1883.
SC
416
Lee, Y. L., Cesario, T., Owens, J., Shanbrom, E., & Thrupp, L. D. (2002). Antibacterial
Liu, N. T., Lefcourt, A. M., Nou, X., Shelton, D. R., Zhang, G., & Lo, Y. M. (2013). Native
M AN U
415
RI PT
411
420
microflora in fresh-cut produce processing plants and their potentials for biofilm
421
formation. Journal of Food Protection, 76, 827–832.
422
Malik, A., & Grohmann, E. (2012). Biofilm Formation by Environmental Bacteria. In M. I. Ansari, K. Schiwon, A. Malik, & E. Grohmann (Eds.), Environmental protection
424
Strategies for Sustainability (pp. 344-345). New York: Springer.
425
TE D
423
Meira, Q. G. D. S., Barbosa, I. D. M., Athayde, A. J. A. A., Siqueira-Júnior, J. P. D., & Souza, E. L. D. (2012). Influence of temperature and surface kind on biofilm formation by
427
Staphylococcus aureus from food-contact surfaces and sensitivity to sanitizers. Food
428
Control, 25, 469–475.
430 431
AC C
429
EP
426
Much, P., Pichler, J., Kasper, S. S., & Allerberger, F. (2009). Foodborne outbreaks, Austria 2007. Wiener Klinische Wochenschrift, 121, 77–85. Naimi, T. S., Wicklund, J. H., Olsen, S. J., Krause, G., Wells, J. G., Bartkus, J. M., Boxrud,
432
D. J., Sullivan, M., Kassenborg, H., Besser, J. M., Mintz, E. D., Osterholm, M. T., &
433
Hedberg, C. W. (2003). Concurrent outbreaks of Shigella sonnei and enterotoxigenic
- 19 -
ACCEPTED MANUSCRIPT 434
Escherichia coli infections associated with parsley: implications for surveillance and
435
control of foodborne illness. Journal of Food Protection, 66, 535–541.
436
Notermans, S., Gallhoff, G., Zwietering, M. H., & Mead, G. C. (1995). The HACCP concept : specification of criteria using quantitative risk assessment. Food Microbiology, 12, 81–
438
90.
440 441
Notermans, S., & Mead, G. C. (1996). Incorporation of elements of quantitative analysis in the HACCP system. International Journal of Food Microbiology, 30, 157–173.
SC
439
RI PT
437
Otter, J. A., Vickery, K., Walker, J. T., Pulcini, E. D., Stoodley, P., Goldenberg, S. D., Salkeld, J. A. G., Chewins, J., Yezli, S., & Edgeworth, J. D. (2015). Surface-attached
443
cells, biofilms and biocide susceptibility: implications for hospital cleaning
444
and disinfection. Journal of Hospital Infection, 89, 16–27.
M AN U
442
Peeters, E., Nelis, H. J., & Coenye, T. (2008). Comparison of multiple methods for
446
quantification of microbial biofilms grown in microtiter plates. Journal of
447
Microbiological Methods, 72, 157–165.
448
TE D
445
Pérez-Rodríguez, F., Valero, A., Carrasco, E., García, R. M., & Zurera, G. (2008). Understanding and modelling bacterial transfer to foods: a review. Trends in Food
450
Science & Technology, 19, 131–144.
452 453 454
Rayner, J., Veeh, R., & Flood, J. (2004). Prevalence of microbial biofilms on selected fresh
AC C
451
EP
449
produce and household surfaces. International Journal of Food Microbiology, 95, 29–39. Ryan, K. J.,& Ray, C. G. (2004). Enterobacteriaceae. In K. J. Ryan (Ed.), Sherris Medical Microbiology (pp. 343-372). New York: McGraw-Hill.
455
Sanchez-Vizuete, P., Orgaz, B., Aymerich, S., Le Coq, D. & Briandet, R. (2015). Pathogens
456
protection against the action of disinfectants in multispecies biofilms. frontiers in
- 20 -
ACCEPTED MANUSCRIPT 457 458 459
Microbiology, 6, 705. Srey, S., Jahid, I. K., & Ha, S. D. (2013). Biofilm formation in food industries: A food safety concern. Food Control, 31, 572–585. Staskel, D. M., Briley, M. E., Field, L. H., & Barth, S. S. (2007). Microbial Evaluation of
461
Foodservice Surfaces in Texas Child-Care Centers. Journal of the American Dietetic
462
Association, 107, 854–859.
Stoodley, P., Sauer, K., Davies, D. G., & Costerton, J. W. (2002). Biofilms as complex
SC
463
RI PT
460
differentiated communities. Annual Review of Microbiology, 56, 187–209.
465
Turnbull, P. C. B. (1996). Bacillus. In S. Baron (Ed.), Medical Microbiology. Texas:
466 467
M AN U
464
Galveston.
Yang, H., Kendall, P. A., Medeiros, L. C., & Sofos, J. N. (2009). Efficacy of Sanitizing Agents against Listeria monocytogenes Biofilms on High-Density Polyethylene Cutting
469
Board Surfaces. Journal of Food Protection, 72, 990–998.
EP
471
AC C
470
TE D
468
- 21 -
ACCEPTED MANUSCRIPT
Table 1. Bacteria identified from different surface area in a cafeteria kitchen. Surface area
Total aerobic count 2 b (log CFU/100cm )
Bacteria
c
RI PT
Surface a No. Cold room
5.90
Brachybacterium, Brevundimonas, Kocuria, Pantoea, Paracoccus, Roseomonas, Sphingobacterium
2
Pretreatment countertops
3.83
Bacillus, Curtobacterium, Gordonia, Kocuria, Pantoea, Pseudomonas, Sphingomonas, Staphylococcus
3
Sink
4.14
Bacillus, Enterobacter, Raoultella
4
Faucet in the sink
5.35
Bacillus, Enhydrobacter, Escherichia, Kocuria
5
Roasting/frying countertop area
3.63
Bacillus, Kocuria, Lysinibacillus
6
Spice rack
4.84
Bacillus, Microbacterium, Pantoea
7
Soup pot
3.59
Acinetobacter, Bacillus
8
Frying pot
2.16
Bacillus
M AN U
SC
1
Countertop for completed menu
3.16
Acidovorax, Acinetobacter, Bacillus, Chryseobacterium, Deinococcus, Enterobacter, Sphingomonas
Rice cooker
2.73
Acinetobacter, Bacillus, Chryseobacterium, Kocuria, Sphingomonas
11
Food waste container
3.21
Bacillus, Staphylococcus
12
Dishwasher
2.53
Bacillus
13
Sink for dishwashing
2.86
Bacillus
14
Stainless steel trays
1.60
Bacillus
15
Plastic wicker tray
4.07
Acinetobacter, Chryseobacterium, Dermacoccus, Exiguobacterium, Kocuria, Pantoea
16
Iron roasting pan
2.00
Acinetobacter, Kocuria
17
Knives
2.00
Acinetobacter, Bacillus
18
Cutting board (Completed menu)
2.45
19
Cutting board (vegetables)
2.97
20
Gloves
21
Apron
22
Fan
23
Floor
EP
TE D
9 10
Arthrobacter, Bacillus, Curtobacterium, Kocuria, Microbacterium, Staphylococcus Acinetobacter, Dermacoccus, Enterobacter, Kocuria, Leclercia, Methylobacterium, Roseomonas
AC C
472
3.97
Acinetobacter, Bacillus, Kocuria
6.85
Bacillus, Kocuria, Pseudoxanthomonas, Rhizobium
4.15
Bacillus Achromobacter, Aeromonas, Bacillus, Cellulosimicrobium, Chryseobacterium, Cloacibacterium, Diaphorobacter, Kocuria, Pseudomonas
4.65
- 22 -
ACCEPTED MANUSCRIPT
a
Please refer to the images in Fig. S1 for the sample number and the surface information.
474
b
Total aerobic count was measured based on the surface swab of 100 cm2 surface area using sterile steel frame and quantified on plate count
475
agar.
476
c
RI PT
473
Bacteria identified on each surface at genus level.
SC
477
M AN U
478 479
AC C
EP
TE D
480
- 23 -
ACCEPTED MANUSCRIPT Figure legends
482
Figure 1. Scatter plot of the crystal violet assay for foodborne bacterial isolates. Each data
483
point represents an average of three replicates. The horizontal lines represent the mean value
484
for each bacterium. The same lowercase letters indicate no significant difference at p<0.05
485
using Tukey’s HSD.
RI PT
481
486
Figure 2. A box plot of the inhibition zone based on the agar well diffusion assay (A-E) and
488
the absorbance of the resazurin reduction assay (F-J) of foodborne bacterial isolates with
489
NaClO (A, F), benzalkonium chloride (B, G), H2O2 (C, H), lactic acid (D, I) and citric acid (E,
490
J). Each data point represents the average of three replicates. The ‘+’ represent the mean
491
value for each bacteria. The same lowercase letters indicate no significant difference at
492
p<0.05 using Tukey’s HSD. The reduction % was calculated as follows. (RFU of treated well
493
with PBS – RFU of treated well with disinfectant) / (RFU of treated well with PBS - RFU of
494
resazurin) x 100
TE D
M AN U
SC
487
495
Figure 3. A principal component analysis score plot (A, C) and loading plot (B, D) based on
497
the agar well diffusion assay (A, B) and resazurin reduction assay (C, D) for foodborne
498
pathogens and bacteria isolates from a cafeteria kitchen.
AC C
499
EP
496
500
Supporting information
501
Figure S1. Images of a cafeteria kitchen. The number indicates the swabbed surface area,
502
which is provided in Table 1.
503 - 24 -
ACCEPTED MANUSCRIPT 504
Table S1. Crystal violet assay for biofilm formation of bacteria from culture collection.
505 506
Table S2. Crystal violet assay for biofilm formation of bacteria isolated in this study.
508
RI PT
507
Table S3. Disinfectant susceptibility of sessile and planktonic bacteria.
AC C
EP
TE D
M AN U
SC
509
- 25 -
SC
RI PT
ACCEPTED MANUSCRIPT
511
M AN U
510
Figure 1.
AC C
EP
TE D
512
- 26 -
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
513
- 27 -
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
Figure 2.
AC C
514
- 28 -
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
515
TE D
518
EP
517
Figure 3.
AC C
516
- 29 -
ACCEPTED MANUSCRIPT Highlights Bacteria isolated from a cafeteria kitchen showed strong biofilm formation capacity. Disinfectant susceptibility changed after biofilm formation in a strain-dependent manner.
RI PT
After biofilm formation, Kocuria increased the resistance to all disinfectants tested. Multivariate analysis revealed a significant increase of relative resistance of
AC C
EP
TE D
M AN U
SC
Kocuria spp. after biofilm.
S0023-6438(16)30729-0
DOI:
10.1016/j.lwt.2016.11.060
Reference:
YFSTL 5869
To appear in:
LWT - Food Science and Technology
Received Date: 21 June 2016 Revised Date:
14 October 2016
Accepted Date: 20 November 2016
Please cite this article as: Lim, E.S., Lee, J.E., Kim, J.-S., Koo, O.K., Isolation of indigenous bacteria from a cafeteria kitchen and their biofilm formation and disinfectant susceptibility, LWT - Food Science and Technology (2016), doi: 10.1016/j.lwt.2016.11.060. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Running head: Biofilm Formation and Disinfectant Susceptibility
2
Isolation of indigenous bacteria from a cafeteria kitchen and their biofilm formation
3
and disinfectant susceptibility
4
Eun Seob Lim1, Jang Eun Lee1,2, Joo-Sung Kim1,3, Ok Kyung Koo4,5*
5
1
6
Republic of Korea
7
2
8
si, Gyeonggi-do, Republic of Korea
9
3
RI PT
1
Department of Food Biotechnology, Korea University of Science & Technology, Daejeon,
SC
Traditional Alcoholic Beverages Research Team, Korea Food Research Institute, Seongnam-
10
Republic of Korea
11
4
12
Gyeongsangnam-do, Republic of Korea
13
5
14
Gyeongsangnam-do, Republic of Korea
M AN U
Food Safety Research Team, Korea Food Research Institute, Seongnam-si, Gyeonggi-do,
Department of Food and Nutrition, Gyeongsang National University, Jinju,
TE D
Institute of Agriculture & Life Science, Gyeongsang National University, Jinju,
15
EP
16
*Corresponding author:
18
Ok-Kyung Koo, Ph.D.
19
Department of Food and Nutrition
20
Gyeongsang National University
21
501 Jinjudae-ro,
22
Jinju, Gyeongsangnam-do, 52828
23
Tel: +82-55-772-1441
24
Fax: +82-55-772-1439
25
Email: [email protected]
AC C
17
-1-
ACCEPTED MANUSCRIPT Abstract
27
Bacterial biofilm formation in foodservice facilities is a continuous cross-contamination risk
28
through survival and persistence despite disinfectant treatments. In this study, we evaluated
29
biofilm formation and disinfectant susceptibility of 178 strains obtained from a cafeteria
30
kitchen and 70 foodborne pathogens and analyzed results by multivariate data analyses. A
31
total of 23 areas in a cafeteria kitchen were selected for bacterial isolation and identification.
32
The capacity for biofilm formation was tested using a crystal violet assay, and disinfectant
33
susceptibility was examined using an agar well diffusion assay and resazurin reduction assay.
34
The most frequently isolated genera were Bacillus (33%), Acinetobacter (17%), Kocuria
35
(12%) and Staphylococcus (5%). The genus Bacillus showed the strongest capacity of the
36
biofilm formation. The foodborne bacteria exhibited a wide range of susceptibility to
37
disinfectants, such as sodium hypochlorite, hydrogen peroxide, benzalkonium chloride, lactic
38
acid and citric acid. However, the susceptibilities changed after biofilm formation in a strain-
39
dependent manner, and the relative resistance levels changed among the isolates. Overall, this
40
study will be a great resource for selecting and using disinfectants in foodservice facility
41
hygienic practices.
EP
TE D
M AN U
SC
RI PT
26
42
Keywords: Cafeteria kitchen, Biofilm, Crystal Violet Assay, Disinfectant, Resazurin Reduction
44
Assay
45
AC C
43
-2-
ACCEPTED MANUSCRIPT 46
1. Introduction Food hygiene control and safety management is essential and critical for protecting
48
the public from foodborne illnesses. The HACCP program and other adequate programs have
49
been globally accepted for systemically preventing and controlling food safety hazards
50
(Grönholm, Wirtanen, Ahlgren, & Sjöberg, 1999; Notermans, Gallhoff, Zwietering, & Mead,
51
1995; Notermans & Mead, 1996). Despite efforts to minimize contamination, massive
52
foodborne outbreaks occur throughout the world (Much, Pichler, Kasper, & Allerberger, 2009;
53
Naimi et al., 2003). Such outbreaks can be due to inadequate use of cleaning and sanitizing
54
practices as well as naturally occurring biofilms (Yang, Kendall, Medeiros, & Sofos, 2009).
M AN U
SC
RI PT
47
Biofilm is a bacterial community that adheres to biotic or abiotic surfaces and
56
produces exopolymeric substances to protect from the environmental stress including
57
antibiotic treatment or disinfection (Rayner, Veeh, & Flood, 2004). Previous studies have
58
confirmed the ubiquity of biofilm in household surfaces and suggest the importance of
59
biofilm in household hygiene (Rayner et al., 2004), in various food processing plants (Srey,
60
Jahid, & Ha, 2013) and even in fresh produce and meat (Jessen & Lammert, 2003). In
61
phyllosphere microbiology, 30 to 80% of the total bacterial population in a plant is related to
62
biofilm formation (Lindow & Brandl, 2003). Costerton et al. (1987) claimed that
63
approximately 99% of all bacteria in nature exist as a biofilm and that most microorganisms
64
can survive and contaminate biotic or abiotic surfaces. Further, 80% of bacterial infections in
65
the USA are associated with biofilm (Costerton et al., 1987; Srey et al., 2013).
EP
AC C
66
TE D
55
Bacterial biofilms in food ingredients or food handlers transfer bacteria to food
67
contact surfaces, to other food handlers or the food processing environment, and contaminate
68
the final food product (Pérez-Rodríguez, Valero, Carrasco, García, & Zurera, 2008). Because
-3-
ACCEPTED MANUSCRIPT sessile bacteria can survive and persist in the environment, adequate and appropriate
70
disinfectant use is necessary to minimize the risk of foodborne pathogen contamination (Gil,
71
Selma, López-Gálvez, & Allende, 2009). Naturally acquired bacterial biofilms have been the
72
main target of intervention technologies because they protect pathogenic bacteria and the
73
sessile bacterial cells that were embedded in the biofilm were 10-1000 times more resistant to
74
antibiotic treatment (Davies, 2003; Sanchez-Vizuete, Orgaz, Aymerich, Le Coq & Briandet,
75
2015). Therefore, an increasing resistance to disinfectants is a critical issue in food safety
76
(Meira, de Medeiros Barbosa, Alves Aguiar Athayde, de Siqueira-Júnior, & de Souza, 2012).
SC
RI PT
69
Until now, disinfectant susceptibility in bacterial biofilms was individually evaluated
78
using select disinfectants. We isolated bacteria from a foodservice facility and evaluated their
79
capacity for biofilm formation and disinfectant resistance/susceptibility compared with
80
foodborne pathogens from our culture collection. We also compared the relationship between
81
sessile and planktonic bacteria in response to the same disinfectants and compared the
82
correlations among the bacterial strains and different disinfectants. This evaluation will
83
provide a valid result through comparing bacteria under different antibiotic and surface-
84
attached conditions.
87 88
TE D
EP
86
2. Materials and Methods
AC C
85
M AN U
77
2.1 Bacterial strains and growth conditions Seventy foodborne pathogens including E. coli O157:H7, Listeria monocytogenes,
89
Salmonella and Staphylococcus aureus were obtained from the culture collection of the Food
90
Safety Research Team in the Korea Food Research Institute (Table S1). Bacteria was
91
inoculated in Tryptic Soy Broth (Becton Dickinson Co, Franklin Lakes, New Jersey, USA)
-4-
ACCEPTED MANUSCRIPT 92
and incubated at 37°C for 16 to 18 h in shaking incubator. The cultures were maintained in
93
15% glycerol at -80°C until use.
95
2.2 Isolating microorganisms on food-contact surfaces
RI PT
94
A cafeteria kitchen in a foodservice facility with an average daily attendance (ADA)
97
greater than 250 people was used to isolate bacteria. Twenty-three different food-contact
98
surfaces, including cooking utensils, kitchen appliances and the cooking area, were used to
99
isolate background microflora, and the sampling was performed after daily cooking and
100
cleaning (Fig. S1). Each surface (a total of 100 cm2 area using one 10 × 10 cm or four 5 × 5
101
cm stainless steel frames depending on surface conditions) was swabbed 10 times in vertical,
102
horizontal and diagonal directions, respectively, using Pipette swab plus (3M, Minnesota,
103
USA) in 10 mL of buffered peptone water (BPW). The researchers wore sterile latex gloves
104
during collection to minimize cross-contamination from hands. Each sample was vortexed for
105
1 min to release the bacteria attached to the swab and plated onto plate count agar (PCA, BD)
106
with the appropriate dilution and incubated at 37°C for up to 48 h. The collected samples
107
were further processed within two hours for quantification and isolation.
109 110
M AN U
TE D
EP
AC C
108
SC
96
2.3 Isolation and Identification Bacteria were isolated from the colonies grown on PCA. Colony selection was
111
performed based on the morphology of each colony and the food-contact surface area and
112
178 strains were isolated. Each colony was re-streaked on PCA and inoculated in tryptic soy
113
broth (TSB, Merck & Co., Kenilworth, New Jersey, USA) at 37°C for overnight. The
114
overnight culture was mixed with 15% glycerol and kept at -70°C until next use. In order to -5-
ACCEPTED MANUSCRIPT identify the isolates, the overnight culture was centrifuged at 9,400 ×g for 5 min and the
116
pellet was used for the genomic DNA extraction using DNeasy Blood & Tissue kit (Qiagen,
117
Hilden, Germany) and the DNA concentration was measured using NanoVue (GE Healthcare,
118
Buckinghamshire, UK). The 16S rRNA gene of each DNA was amplified using 27F (5'-
119
GAGTTTGATCMTGGCTCAG-3'),
120
primers (Macrogen, Seoul, Korea). DNA templates, 10 pmol/µL of primers and PCR-grade
121
water were prepared with Takara Ex Taq version 2.0 (Takara, Kusatsu, Japan) to a total of 25
122
µL for each reaction. The amplification was performed by following program; initial
123
denaturation at 95°C for 5 min, then 30 cycles of 1) denaturation at 94°C for 1 min, 2)
124
annealing at 55°C for 30 s and 3) extension at 72°C for 1 min, and the final elongation step at
125
72°C for 7 min. The amplified PCR product was loaded on 0.8% agarose gel in TAE (40 mM
126
Tris·HCl, 40 mM acetate, 1.0 mM EDTA) buffer stained with Staining STAR (DyneBio,
127
Seongnam, Korea) and the amplification was confirmed with ultraviolet transilluminator
128
using Gel DocTM EZ Imager (Bio-Rad, Richmond, CA, USA). The amplicon was further
129
purified by QIAquick PCR purification kit (Qiagen). The sequencing was performed by
130
Macrogen with 512F (5'-CCAGCAGCCGCGGTAAT-3') for the downstream of 16S rRNA
131
gene and 512R (5'-ATTACCGCGGCTGCTGG-3') primers for the upstream of 16S rRNA
132
gene. The sequencing results were analyzed with the EzTaxon (Kim et al., 2012) on the basis
133
of 16S rRNA sequence data (Table S2).
135
(5'-GGTTACCTTGTTACGACTT-3')
EP
TE D
M AN U
SC
1492R
AC C
134
and
RI PT
115
2.4 Biofilm formation on a microtiter plate-crystal violet assay
136
All stains from culture collection and kitchen isolates were inoculated in TSB and
137
incubated for 16-18 h at 37°C. The overnight culture was diluted to approximately 107
-6-
ACCEPTED MANUSCRIPT CFU/mL in TSB, and 200 µL of the culture was inoculated in a 96-well plate at 37°C for 48 h
139
to facilitate biofilm formation on the microtiter plate. After incubation, the culture medium
140
was carefully removed and washed with PBS (phosphate-buffered saline, pH 7.0) once. After
141
washing the microtiter plate with biofilm-forming cells, 200 µL of a 1% crystal violet (CV)
142
solution (bioWORLD, Dublin, Ohio, USA) was added, and the microtiter plate was incubated
143
for 30 min at room temperature. After washing with PBS 3 times, absolute ethanol 200 µL
144
was added and incubated for 15 min at RT to destain the CV. From the destaining solution,
145
100 µL was transferred to a new 96-well plate, and the absorbance was measured at 595 nm
146
using Infinite® 200 PRO NanoQuant (Tecan Group Ltd., Männedorf, Switzerland).
M AN U
SC
RI PT
138
147 148
2.5 Disinfectant assay using non-biofilm bacteria
Twenty-eight strains of kitchen isolates and 18 strains from culture collection were
150
selected for the following experiment based on the CV assay that resulted in strong biofilm
151
forming ability (over the absorbance of 0.7 for the isolates from this study and 0.3 from the
152
culture collection). The bactericidal activity of disinfectants was determined using the agar
153
well diffusion method with Mueller Hinton II agar (BD). An overnight culture of the isolate
154
was streaked using a cotton swab to prepare a lawn plate. A well was generated in the middle
155
of each plate using a sterilized 8 mm cork borer (K-ACE, Jongro-gu, Seoul, Korea) and
156
inoculated with 100 µl of a select disinfectant. The disinfectant concentrations were as
157
follows: 2,000 mg/L NaClO (Junsei, Tokyo, Japan), 500 mg/L benzalkonium chloride (BAC,
158
Kukbo Science, Cheongju, Korea), 2,000 mg/L H2O2 (Daejung, Siheung, Korea), 10% lactic
159
acid (LA, Kanto, Tokyo, Japan), and 10% citric acid (CA, Junsei). The plates were then
AC C
EP
TE D
149
-7-
ACCEPTED MANUSCRIPT 160
incubated at 37°C for 18-24 h. The inhibition activity was measured based on the cleared
161
zone.
162
2.6 Disinfectant assay using biofilm-forming bacteria
RI PT
163
The bactericidal activity to disinfectants after biofilm formation was determined
165
using resazurin reduction assay. Fifty-six strains were inoculated in TSB and incubated for
166
16-18 h at 37°C. The initial part of the experiment was performed the same as CV assay until
167
washing with PBS after incubation of the culture. After washing the microtiter plate with
168
biofilm-forming cells, the PBS wash solution was removed, 200 µL of disinfectant was added,
169
and the cells were incubated for 10 min at room temperature. The disinfectant concentrations
170
were as follows: 50 mg/L NaClO (Junsei), 100 mg/L BAC (Kukbo Science), 2,000 mg/L
171
H2O2 (Daejung), 0.25% LA (Kanto), and 0.25% CA (Junsei), and PBS was used as the
172
negative control. Next, the disinfectant was removed, and 200 µL of Dey-Engley Neutralizing
173
Broth (Sigma-Aldrich, Saint Louis, MO, USA) was added, and the cells were incubated for 5
174
min at room temperature. The neutralizing broth was removed, and the 96-well plate was
175
washed with PBS to remove the neutralizing agent. Thereafter, 200 µL of 0.001% (wt/vol)
176
resazurin (Sigma-Aldrich) was added to each well, and the cells were incubated for 60 min at
177
37°C under dark conditions. After incubation, the fluorescence (λex 570 nm, λem 590 nm) was
178
measured using SpectraMax® M2 (Molecular Devices®, Sunnyvale, CA, USA).The
179
reduction level (%) was calculated using the formula below.
180
Reduction % =
AC C
EP
TE D
M AN U
SC
164
′
′
x 100 .
181
where A = RFU of test wells, A’ = RFU of negative control wells, and res. = RFU of
182
resazurin blue oxidized form. -8-
ACCEPTED MANUSCRIPT 183 184
2.5 Multivariate data analysis The resulting data were analyzed using SIMCA-P version 12.0 (Umetrics, Umeå,
186
Sweden), and a mean-centered scaling method was applied for multivariate statistical
187
analyses. A principal component analysis (PCA), an unsupervised pattern recognition method,
188
was performed to examine the intrinsic variation in the dataset.
190
SC
189
RI PT
185
2.6 Statistical analysis
The result was evaluated with one-way analysis of variance (ANOVA), followed by
192
Tukey Honestly Significantly Difference (HSD) tests with a significance level of 0.05
193
(p<0.05). All statistical analyses were conducted using Minitab 17 (Minitab Inc.,
194
Pennsylvania, USA).
M AN U
191
196 197
3. Results
TE D
195
3.1 Bacteria isolation and identification from food-contact surfaces Twenty-three different food-contact surfaces in a cafeteria kitchen (Fig. S1) were
199
selected, and bacteria were isolated from the surfaces to evaluate the capacity for biofilm
200
formation and disinfectant susceptibility. The total aerobic count varied from 1.60 log
201
CFU/cm2 in stainless steel trays to 6.85 log CFU/cm2 in aprons, and 178 bacterial strains were
202
isolated and identified. The floor contained the most diverse bacteria with 9 genera and 15
203
species, followed by the pretreatment countertops, the cold room, the countertop for the
204
completed menu, and the cutting board for vegetables (Table 1). Five surfaces contained
205
Bacillus, including frying pots, dishwashers, sinks for dishwashing, stainless steel trays and
AC C
EP
198
-9-
ACCEPTED MANUSCRIPT fans. Bacillus spp. was the most frequently isolated bacteria from 19 surfaces out of 23
207
surfaces (Table 1), and 61 strains (33%) were identified as Bacillus spp. Kocuria and
208
Acinetobacter spp. were isolated on 12 and 9 surfaces with 23 strains (12%) and 31 strains
209
(17%), respectively. Staphylococcus spp. was isolated at 5% of the total isolates from the
210
pretreatment countertops, food waste container and completed menu cutting board followed
211
by Pantoea at 4% and Sphingomonas at 3%. Pseudomonas spp. was isolated from the
212
pretreatment countertops and floor and accounted for 2% of total isolated strains. In addition,
213
Enterobacter, Cloacibacterium and Chryseobacterium spp. were isolated and accounted for
214
2%, respectively.
215 216
3.2 Biofilm formation
217
M AN U
SC
RI PT
206
A total of 248 bacterial strains, including bacteria isolated from a cafeteria kitchen
218
and
219
Salmonella and Staphylococcus aureus, were used to evaluate the biofilm-forming capacity
220
using a crystal violet assay (Fig. 1). As a result, each strain exhibited a distinctively different
221
profile for biofilm formation with an absorbance range between zeros to 4.0. The results were
222
classified into three categories based on absorbance: weakly adherent <1.0, 1.0 < moderately
223
adherent <3.0, and 3.0< strongly adherent. Over 50% of the strains were in the genera
224
Acinetobacer, Kocuria, Staphylococcus and Bacillus, which strongly or moderately adhered
225
with the average absorbance 1.46, 1.97 and 1.41, respectively. In particular, over 90% of
226
the Bacillus strains adhered strongly with an average absorbance of 2.88. On the other hand,
227
approximately 85% of the foodborne pathogens from culture collection weakly adhered with
culture
collection
including
E.
coli O157:H7,
Listeria monocytogenes,
AC C
EP
TE D
from
- 10 -
ACCEPTED MANUSCRIPT 228
an average absorbance of 0.39 – 0.63, and a few strains were exceptionally strong at forming
229
biofilm.
230
3.3 Disinfectant susceptibility tests
RI PT
231
The agar well diffusion assay (AWD) was performed to evaluate the
233
susceptibility/resistance of the bacteria to five disinfectants (Fig. 2). This method was
234
selected in order to further perform multivariate data analysis and the treated concentration
235
was fixed to limit the comparison factor. The range of the zone of inhibition for each
236
disinfectant was 13 to 33.5 mm, 8.6 to 26.8 mm, 9.7 to 28.6 mm, 17.2 to 35.1 mm, 12.4 to 35
237
mm on sodium hypochloride, benzalkonium chloride, hydrogen peroxide, lactic acid and
238
acetic acid, respectively (Table S3). We observed disinfectant susceptibility with a relatively
239
lower deviation for each tested genus than resazurin reduction assay. Among the disinfectants,
240
LA and CA were most effective against the isolates with a more than 20 mm average
241
inhibition zone. Acinetobacter exhibited the greatest disinfectant sensitivity to NaClO and
242
H2O2, and Kocuria was significantly more resistant to the two disinfectants; however, the
243
reverse was observed for BAC, LA and CA. The disinfectant susceptibility of L.
244
monocytogenes was similar to Kocuria. Bacterial susceptibility to disinfectants was
245
consistent with the non-biofilm-forming bacteria, and the results were genus or species
246
specific rather than strain specific. For BAC, LA and CA, the antimicrobial activity was
247
divided into two groups: Gram-positive and Gram-negative. Kocuria, L. monocytogenes, and
248
S. aureus were relatively sensitive to the three disinfectants, while Acinetobacter, E. coli
249
O157:H7 and Salmonella were significantly resistant.
250
AC C
EP
TE D
M AN U
SC
232
The resazurin reduction assay (RES) was used to evaluate the viability of biofilm-
- 11 -
ACCEPTED MANUSCRIPT forming cells after a disinfectant treatment (Fig. 2). The range of reduction rate by RES was
252
relatively higher than AWD result and mostly strain dependent. The disinfectant resistance
253
also changed after the biofilm had formed. For example, most Kocuria strains were the most
254
resistant to all disinfectants after the biofilm had formed, while most Bacillus isolates became
255
the most sensitive to BAC among the biofilm-forming isolates. Biofilm-forming L.
256
monocytogenes and E. coli O157:H7 became the most sensitive to NaClO and H2O2,
257
respectively. However, Salmonella and S. aureus became the most resistant bacteria to H2O2,
258
and LA and CA, respectively, after the biofilm had formed.
SC
RI PT
251
To interpret the resistance variations using the five disinfectants and different
260
bacteria strains, a principal components analysis (PCA) of pattern recognition was initially
261
applied to the dataset (Fig. 3). The resistances were based on the AWD assay for each
262
bacterium strain and were partially differentiated using PCA score plots (Fig. 3A, B); no
263
significant differentiations were observed in the RES assay (Fig. 3C, D). We observed a
264
65.84% difference in the first coordinate for AWD, and the data were clearly separated into
265
Gram-positive or Gram-negative groups. L. monocytogenes and Kocuria presented a
266
particularly distinctive skew for relative sensitivity to BAC, CA and LA. The second
267
coordinate explained 17.67% of the difference in the relative sensitivities of S. aureus to
268
H2O2. Unlike AWD, the RES results were not clustered based on bacteria under genus or
269
species level. While the Salmonella and Acinetobacter bacterial strains showed similar
270
susceptibilities using AWD, the bacteria were separated throughout the PCA analysis using
271
the RES data. For the RES data, the first principle component 57.92% of variance distribution
272
for Kocuria showed a relative resistance to BAC, LA and CA, which was the opposite result
273
from the AWD. Bacillus also changed from exhibiting a relatively moderate activity to a
AC C
EP
TE D
M AN U
259
- 12 -
ACCEPTED MANUSCRIPT 274
relative sensitivity to RES.
275
4. Discussion
277
Foodservice facilities are required to ensure that the facilities follow food safety and
RI PT
276
278
hygiene
279
sanitation/disinfection processes because foodservice facilities are the perfect niche for
280
bacteria to thrive due to temperature, humidity and nutrient conditions (Staskel et al., 2007;
281
Stoodley et al., 2002). Bacteria that are “native” to food-contact surfaces have been evaluated
282
in previous studies, such as in the kitchen sink drain (Furuhata, Ishizaki, & Fukuyama, 2010)
283
and cutting board (Abdul-Mutalib et al., 2015), and on household surfaces, including the
284
kitchen counter and sink, bathroom floor, and toilet seat (Gajanan & Singh, 2013). Liu and
285
others (2013) isolated 23 genera out of 117 bacterial strains from fresh-cut produce
286
processing plants (Liu et al., 2013). In this study, we evaluated biofilm formation in bacterial
287
isolates from a foodservice facility in 23 different food-contact and nonfood-contact surfaces
288
using a culturing method. Because the experiment was performed after the regular cleaning
289
and sanitation process, the isolated bacteria should be resistant to sanitation or protected from
290
the sanitation process through biofilm formation.
through
regular
and
appropriate
decontamination
and
EP
TE D
M AN U
SC
practices
Aprons were the most contaminated food-contact surface, and Bacillus, Rhizobium,
292
Kocuria and Pseudoxanthomonas were isolated (Table 1). Aprons are required for workers’
293
personal hygiene, but apron sanitation may have been under-estimated due to indirect food-
294
contact activities. However, aprons can be a great mediator for microorganism cross-
295
contamination or transfer due to frequent direct or indirect food contact. Therefore,
AC C
291
- 13 -
ACCEPTED MANUSCRIPT 296
appropriate sanitation practices for aprons are necessary to minimize the cross-contamination
297
risk. The most diverse bacteria isolation was observed in the floor and included
299
Aeromonas, Kocuria, Bacillus, Pseudomonas and Cloacibacterium (Table 1). The floor does
300
not directly interact with food; however, the floor can collect residue from food ingredients,
301
worker hygienic practices and environmental contaminants to create a diverse microbial
302
community. Liu and others (2013) isolated Bacillus, Enterobacter, Rhnella, Ralstonia and
303
Pseudomonas from nonfood-contact surfaces that are known potential soil, plant, animal and
304
water sources. The sampling method also affected diversity through swabbing various
305
locations with a 5x5 cm surface area.
M AN U
SC
RI PT
298
Bacillus spp. were the most dominant isolate at 33% and were detected on frying
307
pots, dishwashers, fans and other surfaces. Bacillus is ubiquitous and can survive stringent
308
conditions through spore formation (Turnbull, 1996). Therefore, whether Bacillus was
309
transferred from ingredients or food handlers, the bacteria survive under at a high temperature,
310
at low moisture and under sanitizing conditions; thus, it was the most frequently isolated
311
bacteria in the foodservice facilities. Opportunistic pathogens, such as Acinetobacter
312
baumannii, Staphylococcus epidermidis, and Kocuria varians, were isolated; thus, proper
313
handling practice is necessary to minimize growth and cross-contamination (Ryan, 2004).
314
The Enterobacteriaceae family is a hygienic indicator and includes Pantoea, Enterobacter,
315
Raoultella, Escherichia, and Leclercia, which were isolated from 8 surfaces, including cold
316
rooms, countertops, sinks, spice racks, plastic wicker trays, and cutting boards (Table 1). The
317
corresponding surfaces are frequently associated with water and raw ingredients in the pre-
318
treatment area. However, this type of contamination can be controlled during the cooking
AC C
EP
TE D
306
- 14 -
ACCEPTED MANUSCRIPT 319
process. The countertop for the completed menu also contained Enterobacter, and salads with
320
fresh produce are at risk for contamination. Crystal violet (CV) is a dye that binds negatively charged cell surface molecules and
322
exopolysaccharide (EPS) and can efficiently detect bacterial presence and quantify the
323
biomass of biofilm-forming cells (Peeters, Nelis, & Coenye, 2008). In this study, most
324
isolates exhibited a moderate or strong biofilm-forming capacity. The result could be due to
325
the isolation process after the daily cleaning and sanitation practices and the bacteria should
326
have survived the antibiotic treatment. The bacteria were recovered by strongly binding the
327
surfaces or through the protection from food debris or the biofilm EPS. While the tested
328
foodborne pathogens adhere less than the isolates, the shield created by the biofilm
329
community or food debris can increase the chances of survival on food-contact surfaces and
330
eventually cause foodborne illnesses. The CV assay is a high-throughput screening method
331
for biofilm; however, the staining method cannot differentiate viable and non-viable bacteria
332
because it detects the entire biomass. Therefore, the resazurin assay was performed to
333
measure cell viability. The correlation between the CV and resazurin assays was moderate
334
with an r value of 0.57, which validates the reasonable relationship between biomass content
335
and viability (data not shown).
EP
TE D
M AN U
SC
RI PT
321
Depending on the disinfectant, the antimicrobial activity varied for the target bacteria.
337
The surfactant disinfectant BAC primarily kills bacteria through membrane damage where
338
the hydrophilic cationic region destabilizes the bacterial surface through electrostatic
339
interactions, and the hydrophobic region penetrates the cell membrane and causes cell
340
leakage. The cell wall of Gram-positive bacteria is composed of peptidoglycans and teichoic
341
acids that are negatively charged; thus, BAC is more effective against Gram-positive bacteria
AC C
336
- 15 -
ACCEPTED MANUSCRIPT (Fazlara & Ekhtelat, 2012; Fraise, Maillard, & Sattar, 2008). Organic acids, such as lactic
343
acid and citric acid, are active permeabilizers and are effective against Gram–positive, while
344
acetic acid exhibits the opposite result (Lee, Cesario, Owens, Shanbrom, & Thrupp, 2002).
345
Whilst lactic acid can weaken a cell wall by binding the phosphate and carbonyl group
346
protons, it is active against Gram-negative at a low pH. Our results confirm previously
347
accepted antibacterial activities before biofilm forms. However, when bacterial biofilms
348
formed, the susceptibility was transformed from clustered activity to strain-dependent activity.
349
Additionally, the relative susceptibility amongst the bacterial group shifted to relatively
350
resistant or sensitive. When bacterial cells are attached to the abiotic surface, the physiology
351
of the bacteria changes by the stressed condition such as limited nutrients on the surface and
352
the bacteria can cross-protect them from other stresses including disinfectant (Chen & Jiang,
353
2014). For instance, the reduction in bacterial growth rate and changes to its cell membrane
354
composition cause a subsequent decrease in metabolism of antimicrobial agents, stress
355
response modulation, quorum sensing and EPS production, which can limit disinfectant
356
penetration of the target bacteria (Malik & Grohmann, 2012; Otter et al., 2015). In this study,
357
we did not compare the reduced susceptibility using a dose-response or quantify viability
358
using culture method, but we compared the relative responses among bacteria to each
359
disinfectant. Therefore, the absolute difference for each bacterium cannot be compared.
360
However, the bacterial responses to disinfectants under the same concentration were
361
comparable, and we can understand the transformation of relative resistance in each
362
bacterium to determine the pertinent bacterial target for each disinfectant, which can be
363
important information for best hygienic practices.
AC C
EP
TE D
M AN U
SC
RI PT
342
364
- 16 -
ACCEPTED MANUSCRIPT 365
5. Conclusion Consumers’ demand on better quality with safety assured food product has been
367
increased, which can be satisfied using an effective safety program and intervention
368
technology. However, current control steps cannot efficiently kill microorganisms, and the
369
inadequate disinfectant process can limit microbial quality control and promote potential
370
contamination by foodborne pathogens through biofilm formation. This comprehensive study
371
suggests that the relative disinfectant susceptibility of bacteria changes upon forming a
372
biofilm, we must consider re-evaluating the reference bacteria for testing disinfectant
373
activities against sessile bacteria.
M AN U
SC
RI PT
366
374
Acknowledgment
376
This research was funded by a research grant from the Korea Food Research Institute under
377
Grant E0142102-02 and from Gyeongsang National University under Grant 2016-0194.
378
380
References
Abdul-Mutalib, Nordin, N. A., S. A., Osman, M., Ishida, N., Tashiro, K., Sakai, K.,
EP
379
TE D
375
Tashiro, Y., Maeda, T., & Shirai, Y. (2015). Pyrosequencing analysis of microbial
382
community and food-borne bacteria on restaurant cutting boards collected in Seri
383
Kembangan, Malaysia, and their correlation with grades of food premises. International
384
Journal of Food Microbiology, 200, 57–65.
385 386 387
AC C
381
Chen, Z., & Jiang, X. (2014). Microbiological safety of chicken litter or chicken litter-based organic fertilizers: a review. Agriculture, 4, 1-29. Costerton, J. W., Cheng, K. J., Geesey, G. G., Ladd, T. I., Nickel, J. C., Dasgupta, M., &
- 17 -
ACCEPTED MANUSCRIPT 388
Marrie, T. J. (1987). Bacterial biofilms in nature and disease. Annual Review of
389
Microbiology, 41, 435–464.
391
Davies, D. (2003). Understanding biofilm resistance to antibacterial agents. Nature Reviews. Drug Discovery, 2, 114–122.
RI PT
390
Fazlara, A., & Ekhtelat, M. (2012). The Disinfectant Effects of Benzalkonium Chloride on
393
Some Important Foodborne Pathogens. American-Eurasian Journal of Agricultural &
394
Environmental Sciences, 12, 23–29.
395
SC
392
Fraise A. P., Lambert, P. A., & Maillard, J. Y. (2008). Russell, Hugo & Ayliffe’s Principles and Practice of Disinfection, Preservation & Sterilization (4th ed.). England: Wiley-
397
Blackwell.
399 400 401 402
Furuhata, K., Ishizaki, N., & Fukuyama, M. (2010). Characterization of heterotrophic bacteria isolated from the biofilm of a kitchen sink. Biocontrol Science, 15, 21–25. Gajanan, M. V, & Singh, O. V. (2013). Isolation of microbes from common household
TE D
398
M AN U
396
surfaces. Journal of Emerging Investigators, 1, 1–7. Gil, M. I., Selma, M. V., López-Gálvez, F., & Allende, A. (2009). Fresh-cut product sanitation and wash water disinfection: Problems and solutions. International Journal of
404
Food Microbiology, 134, 37–45. Grönholm, L., Wirtanen, G., Ahlgren, K., & Sjöberg, K. N. A. (1999). Screening of
AC C
405
EP
403
406
antimicrobial activities of disinfectants and cleaning agents against foodborne spoilage
407
microbes. Z Lebensm Unters Forsch A, 208, 289–298.
408 409 410
Jessen, B., & Lammert, L. (2003). Biofilm and disinfection in meat processing plants. International Biodeterioration & Biodegradation, 51, 265–269. Kim, O. S., Cho, Y. J., Lee, K. H., Yoon, S. H., Kim, M. C., Na, H. S., Park, S. C., Jeon, Y.
- 18 -
ACCEPTED MANUSCRIPT S., Lee, J. H., Yi, H. N., Won, S. H., & Chun, J. S. (2012). Introducing EzTaxon-e : a
412
prokaryotic 16S rRNA gene sequence database with phylotypes that represent
413
uncultured species. International Journal of Systematic and Evolutionary
414
Microbiology, 62, 716–721.
417 418 419
Activity of Citrate and Acetate. Nutrition, 18, 665–666.
Lindow, S. E., & Brandl, M. T. (2003). Microbiology of the Phyllosphere. Applied and Environmental Microbiology, 69, 1875–1883.
SC
416
Lee, Y. L., Cesario, T., Owens, J., Shanbrom, E., & Thrupp, L. D. (2002). Antibacterial
Liu, N. T., Lefcourt, A. M., Nou, X., Shelton, D. R., Zhang, G., & Lo, Y. M. (2013). Native
M AN U
415
RI PT
411
420
microflora in fresh-cut produce processing plants and their potentials for biofilm
421
formation. Journal of Food Protection, 76, 827–832.
422
Malik, A., & Grohmann, E. (2012). Biofilm Formation by Environmental Bacteria. In M. I. Ansari, K. Schiwon, A. Malik, & E. Grohmann (Eds.), Environmental protection
424
Strategies for Sustainability (pp. 344-345). New York: Springer.
425
TE D
423
Meira, Q. G. D. S., Barbosa, I. D. M., Athayde, A. J. A. A., Siqueira-Júnior, J. P. D., & Souza, E. L. D. (2012). Influence of temperature and surface kind on biofilm formation by
427
Staphylococcus aureus from food-contact surfaces and sensitivity to sanitizers. Food
428
Control, 25, 469–475.
430 431
AC C
429
EP
426
Much, P., Pichler, J., Kasper, S. S., & Allerberger, F. (2009). Foodborne outbreaks, Austria 2007. Wiener Klinische Wochenschrift, 121, 77–85. Naimi, T. S., Wicklund, J. H., Olsen, S. J., Krause, G., Wells, J. G., Bartkus, J. M., Boxrud,
432
D. J., Sullivan, M., Kassenborg, H., Besser, J. M., Mintz, E. D., Osterholm, M. T., &
433
Hedberg, C. W. (2003). Concurrent outbreaks of Shigella sonnei and enterotoxigenic
- 19 -
ACCEPTED MANUSCRIPT 434
Escherichia coli infections associated with parsley: implications for surveillance and
435
control of foodborne illness. Journal of Food Protection, 66, 535–541.
436
Notermans, S., Gallhoff, G., Zwietering, M. H., & Mead, G. C. (1995). The HACCP concept : specification of criteria using quantitative risk assessment. Food Microbiology, 12, 81–
438
90.
440 441
Notermans, S., & Mead, G. C. (1996). Incorporation of elements of quantitative analysis in the HACCP system. International Journal of Food Microbiology, 30, 157–173.
SC
439
RI PT
437
Otter, J. A., Vickery, K., Walker, J. T., Pulcini, E. D., Stoodley, P., Goldenberg, S. D., Salkeld, J. A. G., Chewins, J., Yezli, S., & Edgeworth, J. D. (2015). Surface-attached
443
cells, biofilms and biocide susceptibility: implications for hospital cleaning
444
and disinfection. Journal of Hospital Infection, 89, 16–27.
M AN U
442
Peeters, E., Nelis, H. J., & Coenye, T. (2008). Comparison of multiple methods for
446
quantification of microbial biofilms grown in microtiter plates. Journal of
447
Microbiological Methods, 72, 157–165.
448
TE D
445
Pérez-Rodríguez, F., Valero, A., Carrasco, E., García, R. M., & Zurera, G. (2008). Understanding and modelling bacterial transfer to foods: a review. Trends in Food
450
Science & Technology, 19, 131–144.
452 453 454
Rayner, J., Veeh, R., & Flood, J. (2004). Prevalence of microbial biofilms on selected fresh
AC C
451
EP
449
produce and household surfaces. International Journal of Food Microbiology, 95, 29–39. Ryan, K. J.,& Ray, C. G. (2004). Enterobacteriaceae. In K. J. Ryan (Ed.), Sherris Medical Microbiology (pp. 343-372). New York: McGraw-Hill.
455
Sanchez-Vizuete, P., Orgaz, B., Aymerich, S., Le Coq, D. & Briandet, R. (2015). Pathogens
456
protection against the action of disinfectants in multispecies biofilms. frontiers in
- 20 -
ACCEPTED MANUSCRIPT 457 458 459
Microbiology, 6, 705. Srey, S., Jahid, I. K., & Ha, S. D. (2013). Biofilm formation in food industries: A food safety concern. Food Control, 31, 572–585. Staskel, D. M., Briley, M. E., Field, L. H., & Barth, S. S. (2007). Microbial Evaluation of
461
Foodservice Surfaces in Texas Child-Care Centers. Journal of the American Dietetic
462
Association, 107, 854–859.
Stoodley, P., Sauer, K., Davies, D. G., & Costerton, J. W. (2002). Biofilms as complex
SC
463
RI PT
460
differentiated communities. Annual Review of Microbiology, 56, 187–209.
465
Turnbull, P. C. B. (1996). Bacillus. In S. Baron (Ed.), Medical Microbiology. Texas:
466 467
M AN U
464
Galveston.
Yang, H., Kendall, P. A., Medeiros, L. C., & Sofos, J. N. (2009). Efficacy of Sanitizing Agents against Listeria monocytogenes Biofilms on High-Density Polyethylene Cutting
469
Board Surfaces. Journal of Food Protection, 72, 990–998.
EP
471
AC C
470
TE D
468
- 21 -
ACCEPTED MANUSCRIPT
Table 1. Bacteria identified from different surface area in a cafeteria kitchen. Surface area
Total aerobic count 2 b (log CFU/100cm )
Bacteria
c
RI PT
Surface a No. Cold room
5.90
Brachybacterium, Brevundimonas, Kocuria, Pantoea, Paracoccus, Roseomonas, Sphingobacterium
2
Pretreatment countertops
3.83
Bacillus, Curtobacterium, Gordonia, Kocuria, Pantoea, Pseudomonas, Sphingomonas, Staphylococcus
3
Sink
4.14
Bacillus, Enterobacter, Raoultella
4
Faucet in the sink
5.35
Bacillus, Enhydrobacter, Escherichia, Kocuria
5
Roasting/frying countertop area
3.63
Bacillus, Kocuria, Lysinibacillus
6
Spice rack
4.84
Bacillus, Microbacterium, Pantoea
7
Soup pot
3.59
Acinetobacter, Bacillus
8
Frying pot
2.16
Bacillus
M AN U
SC
1
Countertop for completed menu
3.16
Acidovorax, Acinetobacter, Bacillus, Chryseobacterium, Deinococcus, Enterobacter, Sphingomonas
Rice cooker
2.73
Acinetobacter, Bacillus, Chryseobacterium, Kocuria, Sphingomonas
11
Food waste container
3.21
Bacillus, Staphylococcus
12
Dishwasher
2.53
Bacillus
13
Sink for dishwashing
2.86
Bacillus
14
Stainless steel trays
1.60
Bacillus
15
Plastic wicker tray
4.07
Acinetobacter, Chryseobacterium, Dermacoccus, Exiguobacterium, Kocuria, Pantoea
16
Iron roasting pan
2.00
Acinetobacter, Kocuria
17
Knives
2.00
Acinetobacter, Bacillus
18
Cutting board (Completed menu)
2.45
19
Cutting board (vegetables)
2.97
20
Gloves
21
Apron
22
Fan
23
Floor
EP
TE D
9 10
Arthrobacter, Bacillus, Curtobacterium, Kocuria, Microbacterium, Staphylococcus Acinetobacter, Dermacoccus, Enterobacter, Kocuria, Leclercia, Methylobacterium, Roseomonas
AC C
472
3.97
Acinetobacter, Bacillus, Kocuria
6.85
Bacillus, Kocuria, Pseudoxanthomonas, Rhizobium
4.15
Bacillus Achromobacter, Aeromonas, Bacillus, Cellulosimicrobium, Chryseobacterium, Cloacibacterium, Diaphorobacter, Kocuria, Pseudomonas
4.65
- 22 -
ACCEPTED MANUSCRIPT
a
Please refer to the images in Fig. S1 for the sample number and the surface information.
474
b
Total aerobic count was measured based on the surface swab of 100 cm2 surface area using sterile steel frame and quantified on plate count
475
agar.
476
c
RI PT
473
Bacteria identified on each surface at genus level.
SC
477
M AN U
478 479
AC C
EP
TE D
480
- 23 -
ACCEPTED MANUSCRIPT Figure legends
482
Figure 1. Scatter plot of the crystal violet assay for foodborne bacterial isolates. Each data
483
point represents an average of three replicates. The horizontal lines represent the mean value
484
for each bacterium. The same lowercase letters indicate no significant difference at p<0.05
485
using Tukey’s HSD.
RI PT
481
486
Figure 2. A box plot of the inhibition zone based on the agar well diffusion assay (A-E) and
488
the absorbance of the resazurin reduction assay (F-J) of foodborne bacterial isolates with
489
NaClO (A, F), benzalkonium chloride (B, G), H2O2 (C, H), lactic acid (D, I) and citric acid (E,
490
J). Each data point represents the average of three replicates. The ‘+’ represent the mean
491
value for each bacteria. The same lowercase letters indicate no significant difference at
492
p<0.05 using Tukey’s HSD. The reduction % was calculated as follows. (RFU of treated well
493
with PBS – RFU of treated well with disinfectant) / (RFU of treated well with PBS - RFU of
494
resazurin) x 100
TE D
M AN U
SC
487
495
Figure 3. A principal component analysis score plot (A, C) and loading plot (B, D) based on
497
the agar well diffusion assay (A, B) and resazurin reduction assay (C, D) for foodborne
498
pathogens and bacteria isolates from a cafeteria kitchen.
AC C
499
EP
496
500
Supporting information
501
Figure S1. Images of a cafeteria kitchen. The number indicates the swabbed surface area,
502
which is provided in Table 1.
503 - 24 -
ACCEPTED MANUSCRIPT 504
Table S1. Crystal violet assay for biofilm formation of bacteria from culture collection.
505 506
Table S2. Crystal violet assay for biofilm formation of bacteria isolated in this study.
508
RI PT
507
Table S3. Disinfectant susceptibility of sessile and planktonic bacteria.
AC C
EP
TE D
M AN U
SC
509
- 25 -
SC
RI PT
ACCEPTED MANUSCRIPT
511
M AN U
510
Figure 1.
AC C
EP
TE D
512
- 26 -
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
513
- 27 -
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
Figure 2.
AC C
514
- 28 -
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
515
TE D
518
EP
517
Figure 3.
AC C
516
- 29 -
ACCEPTED MANUSCRIPT Highlights Bacteria isolated from a cafeteria kitchen showed strong biofilm formation capacity. Disinfectant susceptibility changed after biofilm formation in a strain-dependent manner.
RI PT
After biofilm formation, Kocuria increased the resistance to all disinfectants tested. Multivariate analysis revealed a significant increase of relative resistance of
AC C
EP
TE D
M AN U
SC
Kocuria spp. after biofilm.