Accepted Manuscript

Cleaning and Sanitation of Salmonella-Contaminated Peanut Butter Processing

Equipment

Elizabeth M. Grasso, Stephen F. Grove, Lindsay A. Halik, Fletcher Arritt, Susanne E.

Keller

PII: S0740-0020(14)00056-2
DOI: 10.1016/j.fm.2014.03.003
Reference: YFMIC 2122

To appear in: Food Microbiology

Received Date: 10 July 2013

Revised Date: 22 January 2014
Accepted Date: 5 March 2014

Please cite this article as: Grasso, E.M., Grove, S.F., Halik, L.A., Arritt, F., Keller, S.E., Cleaning and
Sanitation of Salmonella-Contaminated Peanut Butter Processing Equipment, Food Microbiology (2014),
doi: 10.1016/j.fm.2014.03.003.

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

1 Cleaning and Sanitation of Salmonella-Contaminated Peanut Butter Processing Equipment

3 Elizabeth M. Grassoa1*, Stephen F. Groveb, Lindsay A. Halikc1, Fletcher Arrittd, Susanne E.

4 Kellerc

PT
5

RI
a
6 Office of the Commissioner, U.S. Food and Drug Administration, 6502 S. Archer Road,

7 Bedford Park, IL 60501-1957 USA; egrasso@iit.edu,

SC
8
b
9 Institute for Food Safety and Health, Illinois Institute of Technology, 6502 S. Archer Road,

10 Bedford Park, IL 60501-1957 USA; sgrove@iit.edu

U
AN
11
c
12 Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, 6502 S.
M

13 Archer Road, Bedford Park, IL 60501-1957 USA; lhalik@iit.edu, Susanne.Keller@fda.hhs.gov

D

14
TE

d
15 Department of Food, Bioprocessing & Nutrition Sciences, North Carolina State University, 129

16 Schaub Hall, Raleigh, NC 27695-7624 USA; fmarritt@ncsu.edu

EP

17

18 1Current Address: Institute for Food Safety and Health, Illinois Institute of Technology, 6502 S.
C

19 Archer Road, Bedford Park, IL 60501-1957 USA; egrasso@iit.edu

AC

20

21 * Corresponding author: Institute for Food Safety and Health, 6502 S. Archer Road, Bedford

22 Park, IL 60501-1957 USA; E-mail: egrasso@iit.edu; Tel: (708) 563-8824; Fax: (708) 728-4177

23
 ACCEPTED MANUSCRIPT

24 Abstract

25 Microbial contamination of peanut butter by Salmonella poses a significant health risk as

26 Salmonella may remain viable throughout the product shelf life. Effective cleaning and

27 sanitation of processing lines are essential for preventing cross-contamination. The objective of

PT
28 this study was to evaluate the efficacy of a cleaning and sanitation procedure involving hot oil

RI
29 and 60% isopropanol, ± quaternary ammonium compounds, to decontaminate pilot-scale

30 processing equipment harboring Salmonella. Peanut butter inoculated with a cocktail of four

SC
31 Salmonella serovars (~7 log CFU/g) was used to contaminate the equipment (~75 L). The

32 system was then emptied of peanut butter and treated with hot oil (90oC) for 2 h followed by

33
U
sanitizer for 1 h. Microbial analysis of food-contact surfaces (7 locations), peanut butter, and oil
AN
34 were conducted. Oil contained ~3.2 log CFU/mL on both trypticase soy agar with yeast extract

35 (TSAYE) and xylose lysine deoxycholate (XLD), indicating hot oil alone was not sufficient to
M

36 inactivate Salmonella. Environmental sampling found 0.25-1.12 log CFU/cm2 remaining on

D

37 processing equipment. After the isopropanol sanitation (± quaternary ammonium compounds),

TE

38 no Salmonella was detected in environmental samples on XLD (< 0.16 log CFU/cm2). These

39 data suggest that a two-step hot oil clean and isopropanol sanitization treatment may eliminate
EP

40 pathogenic Salmonella from contaminated equipment.

41 Highlights
C

42 • Hot oil alone is not sufficient to remove Salmonella on processing equipment.

AC

43 • Hot oil may be a viable first step in a two part cleaning-sanitization process.

44 • Hot oil followed by 60% isopropanol removes Salmonella on processing equipment.

45 • Hot oil followed by isopropanol + QACs removes Salmonella on processing equipment.

46 Keywords: peanut butter, Salmonella, sanitation

2
 ACCEPTED MANUSCRIPT

47 1. Introduction

48 Low-moisture foods, including nut butters, were once thought to be relatively safe with respect

49 to foodborne illness risk since low water activities do not permit the growth of foodborne

50 microorganisms. Research has shown that while pathogenic foodborne microorganisms are

PT
51 unable to grow in these foods, they are able to persist in the products at both refrigerated and

RI
52 ambient conditions for a long period of time (Burnett et al., 2000; Grasso et al., 2010; Keller et

53 al., 2012; Park et al., 2008). Furthermore, low-moisture foods such as peanuts, almonds,

SC
54 hazelnuts, sesame seeds, pine nuts, walnuts, and pistachios have all been implicated in outbreaks

55 or recalled due to contamination with bacterial pathogens (Centers for Disease Control and

56
U
Prevention, 2011; U. S. Food and Drug Administration, 2004; U. S. Food and Drug
AN
57 Administration, 2009a; U. S. Food and Drug Administration, 2010). Nuts may become

58 contaminated with pathogens, such as Salmonella, at any point from growth to processing
M

59 (Schaffner et al., 2013). Since the mid 1990’s there have been five recorded Salmonella
D

60 outbreaks associated with commercial peanut butter, peanut-based products, and peanut flavored
TE

61 savory snacks (Centers for Disease Control and Prevention, 2007; Centers for Disease Control

62 and Prevention, 2009; Centers for Disease Control and Prevention, 2012; Killalea et al., 1996;
EP

63 Scheil et al., 2008). More than 1,150 people became ill consuming Salmonella-contaminated

64 peanut butter products during two particularly high-profile U.S. outbreaks in 2007 and 2009
C

65 (Centers for Disease Control and Prevention, 2011), which highlighted the vulnerability of such
AC

66 products to contamination by pathogenic microorganisms. The 2012 outbreak, involving a

67 variety of nut butter products processed at the same facility, was likely the result of poor

68 equipment sanitation (U. S. Food and Drug Administration, 2012).

69

3
 ACCEPTED MANUSCRIPT

70 Typical grinding and packaging of roasted peanuts occurs at 48oC (Woodroof, 1983) yet

71 inactivation of microorganisms in peanut butter and peanut butter products is minimal at this

72 temperature (Mattick et al., 2001). Keller et al. (2012), found a 6 log CFU/g reduction in

73 Salmonella over 50 min at 85oC. Ma et al. (2009) and Shachar and Yaron (2006) found similar

PT
74 results. These findings indicate thermal processing of peanut butter is not a practical

RI
75 intervention step. Consequently, to reduce the risk of foodborne illness in such products,

76 prevention of contamination rather than processing is essential.

SC
77

78 Effective cleaning and sanitation of nut butter lines are essential for preventing both initial and

79
U
cross-contamination with microbial hazards such as Salmonella (U. S. Food and Drug
AN
80 Administration, 2009b; Grocery Manufacturers Association, 2009a; Grocery Manufacturers

81 Association, 2009b; Grocery Manufacturers Association, 2010; International Life Sciences

M

82 Institute, 2011; Podolak et al., 2010; Scott et al., 2009). In a May 2007 survey of Grocery
D

83 Manufacturers Association (GMA) members, 53% of respondents (n = 17 respondents) had

TE

84 manufacturing periods for low-moisture areas of processing that extended 7 days or longer prior

85 to shutting down for sanitation (Scott et al., 2009). Several companies reported production
EP

86 periods as long as 28-35 days prior to shutting down for sanitation (Scott et al., 2009).

87 Environmental investigations of consecutive Salmonella outbreaks originating from a single

C

88 bakery found that inadequately cleaned piping nozzles may have been responsible for a continual
AC

89 cross-contamination problem (Evans et al., 1996). Traditional wet-cleaning methods are

90 generally not used in such a setting, as the introduction of moisture may increase the risk of

91 equipment contamination (Beuchat et al., 2013; Codex Alimentarius, 1979). Traditional dry-

92 cleaning methods are not suitable for all dry cleaning situations, such as nut butters (International

4
 ACCEPTED MANUSCRIPT

93 Life Sciences Institute, 2011). Currently there are no validated sanitation programs for use in

94 high fat emulsified processing plants. As this type of product was considered to be of minimal

95 risk, good manufacturing practices have not been implemented and collectively used throughout

96 the nut butter manufacturing industry. Non-aqueous based chemical sanitation methods are

PT
97 needed to ensure the killing/removal of pathogenic microorganisms from nut butter processing

RI
98 equipment. However, the efficacy of such treatments needs further investigation. Potential

99 non-aqueous chemical sanitation products, such as quaternary ammonium compounds or

SC
100 isopropyl alcohol containing quaternary ammonium compounds, have shown promising effects

101 against Salmonella in low-moisture environments (Du et al., 2007; Du et al., 2010; Kamineni et

102
U
al., 2011). A range of products are commercially available for disinfection of low-moisture food
AN
103 production areas). Isopropanol has the added benefit of evaporating quickly and leaving no

104 residue on food production surfaces. The objective of this research was to evaluate the efficacy
M

105 of a model cleaning and sanitation method involving 60% isopropanol with and without the
D

106 addition of quaternary ammonium compounds on Salmonella removal from pilot-scale peanut
TE

107 butter processing equipment.

108
EP

109 2. Methods and methods

110 2.1 Organisms and growth conditions

C

111 Salmonella Tennessee K4643 and Salmonella Anatum 5802 were obtained as a gift from Dr. L.
AC

112 Beuchat, University of Georgia (Athens, GA). S. Tennessee K4643 was originally isolated from

113 the 2006 United States peanut butter outbreak (Centers for Disease Control and Prevention,

114 2007). S. Anatum 5802 was isolated from a raw pecan sample. Salmonella Typhimurium 09-

115 0001-E5 and Salmonella Typhimurium 09-0001-A1 cultures were obtained as a gift from the

5
 ACCEPTED MANUSCRIPT

116 Minnesota Department of Agriculture (St. Paul, MN) and originally isolated from peanut butter

117 involved in a 2008 multi-state outbreak in the United States. Stock cultures were maintained

118 frozen at -20oC.

119

PT
120 Overnight cultures of the four Salmonella serovars were transferred from TSAYE to test tubes

RI
121 containing 10 mL trypticase soy broth (TSB; Becton, Dickinson, and Co.). After 24 h incubation

122 (37ºC), 0.1 mL was spread on the surface of TSAYE plates and incubated for 24 h at 37oC to

SC
123 obtain lawn cultures. After incubation, cells were harvested from plates by adding 1 mL TSB

124 and mixed using a sterile L-shaped plate spreader (Fisher Scientific, Pittsburgh, PA). The

125
U
suspension was removed from the plate using a 1-mL pipette and collected in a 50 mL conical
AN
126 centrifuge tube (Becton Dickinson, and Co., Franklin Lake, NJ). Approximately 0.5 mL cell

127 suspension was recovered from each plate. Ten plates of each serovar were harvested separately
M

128 for each experiment. Approximately 20 mL of cell suspension was recovered from 40 plates.
D

129 Five mL of each serovar cell suspension was pipetted together to form the Salmonella cocktail
TE

130 used in this study. The concentration of harvested cells was 11.6 ± 0.6 log CFU/mL. This cell

131 suspension was then mixed 1:1 with peanut oil and approximately 1 mL of Tween 80 (Fisher
EP

132 Chemical, Whippany, NJ).

133
C

134 The cell:oil:Tween 80 mixture was thoroughly mixed with a vortex mixer. The mixture, ~41 mL
AC

135 total, was equally divided between two 400 g lots of commercial creamy peanut butter in

136 separate Whirl-Pak bags (24 oz, Nasco, Fisher Scientific, Pittsburgh, PA) and mixed by

137 alternating hand mixing and stomaching (StomacherTM 400, Seward Ltd, West Sussex, UK) at 30

6
 ACCEPTED MANUSCRIPT

138 s intervals at 250 rpm three times each. These two contaminated lots were subsequently used to

139 contaminate peanut butter used in pilot plant processing experiments.

140

141 2.2 Peanut butter processing equipment

PT
142 A peanut butter pumping machine consisting of two steam-jacketed vessels and two pumps, with

RI
143 stainless steel piping (inner diameter 1.5”) to circulate molten peanut butter (donated by North

144 Carolina State University and funded through a FDA Centers of Excellence grant awarded to

SC
145 Western Center for Food Safety, University of California at Davis) was installed into the IFSH

146 Bio-Containment Pilot Plant (BCPP) to allow safe experimentation with large quantities of

147
U
pathogen (Figure 1). Biohazard suits with breathing air were used for trials involving large
AN
148 quantities of pathogenic microorganisms or procedures that presented a risk of aerosolization.

149 The steam-jacketed vessels (capacity ~75L) included thermocouple units to provide continuous
M

150 in-tank temperature profiles of the heating process. The positive displacement pump from the
D

151 peanut butter tank allows re-circulation of product, or disposal from system and had a flow rate
TE

152 of ~17.4 L/min.

153
EP

154 2.3 Processing

155 Approximately 75 L of soybean oil, purchased from a wholesale supplier (Gordon Food
C

156 Services, Chicago, IL), was added to the steam-jacketed tank 1 (T1) and heated to ~93°C (Figure
AC

157 1).

158

159 Oil heated in T1 was pumped through the piping, via the centrifugal pump until the exiting oil

160 was visibly clear (~5 L), to flush out any liquid or debris that may have been present in the

7
 ACCEPTED MANUSCRIPT

161 piping prior to the start of each trial. Rinse oil was disposed of by pumping to a collection tank.

162 Once clear, the remaining oil was re-circulated through the piping back to T1.

163

164 Approximately 63.5 kg of creamy peanut butter, purchased from a manufacturer in bulk

PT
165 containers (15.9 kg each), was autoclaved at 121°C for 30 min (total run time 1:08 h) to soften

RI
166 the peanut butter for ease of handling prior to each trial. Immediately following autoclaving, the

167 temperature of the peanut butter was ~55-60oC, as measured by a digital thermometer probe

SC
168 inserted near the middle of the container. Peanut butter treated in this manner was softened, but

169 otherwise unchanged in character. After autoclaving, peanut butter in each container was

170
U
sampled for initial bacterial counts, and then mixed with a metal 5-gal spiral mixer (Model
AN
171 #SM5HD, Allway Tools, Bronx, NY) attached to an electric hand drill (DW130V, Dewalt,

172 Baltimore, MD) for approximately 1 min. Four containers of softened creamy peanut butter
M

173 were emptied into the steam-jacketed tank 2 (T2), and heated to ~65oC. Agitation was provided
D

174 inside T2 by the spiral mixer. While heating in T2, the peanut butter was pumped at a flow rate
TE

175 of 17.4 ± 0.3 L/min from T2 through the piping to flush out any remaining oil. Peanut butter was

176 discarded until peanut butter of a regular consistency was observed exiting the equipment (~5 L
EP

177 removed). Peanut butter was then re-circulated through the piping and T2 until a holding

178 temperature of 65oC was reached, at which time approximately 30 g of peanut butter was
C

179 removed in triplicate from the exit piping (Figure 1, #1), for determination of both microbial
AC

180 populations and water activity.

181

182 2.3.1 Inoculation

8
 ACCEPTED MANUSCRIPT

183 Approximately 60 kg heated peanut butter was inoculated with two 400-g lots of the Salmonella-

184 inoculated peanut butter (described earlier) and mixed as described earlier and by re-circulating

185 the mixture through the piping for 15 min, to ensure a fully contaminated processing line. Final

186 Salmonella populations in the product were determined in triplicate samples removed at the end

PT
187 of the 15 min mixing period. Remaining contaminated peanut butter was then pumped into a

RI
188 waste container.

189

SC
190 After the peanut butter was pumped from the system, a rinse using approximately 20 L of

191 soybean oil at ambient temperature (~22°C) was added directly to T2 to remove any large peanut

192
U
butter clumps still remaining in the system. This initial oil rinse was re-circulated through the
AN
193 piping and T2 for 5 min. After 5 min the oil/peanut butter mixture was sampled in triplicate for

194 microbial testing, and the remaining mixture was discarded.

M

195
D

196 2.3.2 Hot oil clean

TE

197 Soybean oil heated to 93oC in T1 was pumped to T2 to begin a “hot oil” cleaning procedure.

198 The hot oil was recirculated through the piping and T2 for 5 min to ensure thorough mixing.
EP

199 After the initial 5 min recirculation, a ‘0 time’ sample of the oil was taken. Subsequent samples

200 of oil were taken at regular intervals up to 2 h (0, 30, 60, 90, and 120 min). The oil was then
C

201 discarded, and environmental samples (swabs) taken as described section 2.4.
AC

202

203 2.3.3 Chemical sanitation

9
 ACCEPTED MANUSCRIPT

204 Two types of sanitation protocols, both containing isopropanol, were utilized. Equipment was

205 cooled for 24 h prior to any chemical sanitation to ensure that isopropanol was not used near its

206 flash point (40oC). All experimental trials were repeated three times for each sanitizer.

207

PT
208 2.3.3.1 Isopropanol sanitation

RI
209 Isopropanol (Fisher Scientific, Pittsburgh, PA) was diluted to a 60% (v/v) concentration using

210 deionized water. Approximately 75 L of 60% isopropanol was poured into T2. Pumping

SC
211 commenced, and the alcohol, which initially had a cloudy appearance due to residual oil and

212 peanut butter in the system, was diverted to a waste container until it appeared clear (~7.5-10 L).

213
U
The remaining isopropanol (~65 L) was re-circulated for 60 min at room temperature, and then
AN
214 diverted to waste. Environmental samples were taken as described in section 2.4 (Figure 1b).

215
M

216 2.3.3.2 Isopropanol with added quaternary ammonium compounds sanitation

D

217 Redi san RTU hard surface sanitizer (Ecolab, St. Paul, MN) containing 59% isopropanol and
TE

218 0.02% quaternary ammonium compounds was used. Redi san is commercially-available and

219 EPA-registered for food-contact surfaces, when used as the manufacturer instructs.
EP

220 Approximately 65 L of the commercial sanitizer was circulated through the piping and T2.

221 Sanitizer visibly contaminated with peanut butter was diverted to a waste container until it
C

222 appeared clear (~5 L). The remaining isopropanol based sanitizer was again re-circulated for 60
AC

223 min at room temperature, and then diverted to waste. Environmental samples were taken as

224 described in section 2.4.

225

226 2.4 Environmental sampling

10
 ACCEPTED MANUSCRIPT

227 Environmental samples were taken at four time points during the experiment; prior to the start of

228 the experiment, after hot oil treatment, after cooling, immediately prior to sanitizer treatment,

229 and immediately following any sanitizer treatment. Sampling sites were as follows: ; 2. Upper 1.

230 Exit piping of tank 2 (T2); 2. Elbow of the upper recirculation pipe of T2; 3. Left elbow of lower

PT
231 pipe, left of positive displacement pump; 4. Right side of lower pipe, immediately before

RI
232 positive displacement pump; 5. Lower pipe, immediately after positive displacement pump; 6.

233 Bottom exit of T2; and 7. Bottom exit of tank 1 (T1) (Table 1, Figure 1b). These sites were

SC
234 chosen as the most likely to contain detectable levels of Salmonella after the cleaning and

235 sanitation process. All sampling sites with the exception of the bottom exit of tank 1 (T1) had

236
U
direct contact with the contaminated peanut butter. Sample times and location were the same for
AN
237 each trial.

238 Environmental samples were taken by swabbing an area of ~6.25 cm2 (2.5 cm x 2.5 cm) at the
M

239 predetermined locations. The limit of detection was 0.16 log CFU/cm2. A commercial
D

240 environmental swab consisting of a five-inch, rayon-tipped swab and containing a neutralizing
TE

241 buffer (3MTM, St. Paul, MN) was used. Environmental swabs were stored on ice (less than 2 h)

242 until microbial analysis.

EP

243

244 2.5 Microbial analysis

C

245 To determine Salmonella populations in peanut butter and oil samples, 10 g samples were taken
AC

246 and serially diluted 1:10 in 0.1% peptone water (Becton, Dickinson, and Co.) and plated on

247 TSAYE and xylose deoxycholate agar (XLD, Becton, Dickinson, and Co.). One to two drops of

248 Tween 80 was added to the first of each dilution series with the peanut butter or oil.

249 Environmental samples were serially diluted in 0.1% peptone water and plated onto TSAYE and

11
 ACCEPTED MANUSCRIPT

250 XLD. Plates were incubated at 37oC for 24 h. For each trial, a few presumptive positive

251 colonies were picked and confirmed as Salmonella species using Gram negative cards in the

252 Vitek® 2 Compact (bioMérieux, Marcy l’Etoile, France).

253

PT
254 2.6 Statistical analysis

RI
255 One-way analysis of variance (ANOVA) and paired t tests between sampling points for the

256 environmental swabs and also between product contamination from inoculation through hot oil

SC
257 treatment were calculated with SAS 9.2 (SAS Institute, Inc, Cary, NC) at a significance level of

258 95% (α = 0.05).

259
U
AN
260 3. Results and Discussion

261 Although traditional water-based cleaning and sanitizing methods are very effective at reducing
M

262 the presence of microorganisms in the processing environment, they are not recommended for
D

263 use in the low-moisture food processing environment (Du et al., 2007). During the manufacture
TE

264 of peanut butter products and cleaning/sanitation of peanut butter processing equipment, the

265 environment should be kept as free from moisture as possible. The use of water during cleaning
EP

266 can allow the water activity to reach levels that promote the growth of microorganisms (Beuchat

267 et al., 2013; Codex Alimentarius, 1979). Dry sanitation procedures are used in processing
C

268 environments where water must not be used, but current methods may not be appropriate for all
AC

269 dry cleaning situations (Beuchat et al., 2013; International Life Sciences Institute, 2011).

270 Currently, nut butter producers frequently push through fresh product or internal pipeline

271 cleaning devices to remove unwanted product or clean their piping, and do not have validated

272 chemical methods to sanitize their processing equipment. The lack of effective sanitation

12
 ACCEPTED MANUSCRIPT

273 procedures could lead to microbial cross-contamination (Evans et al., 1996; Grocery

274 Manufacturers Association, 2009b; Kusumaningrum et al., 2003), as seen in the 2012 Salmonella

275 outbreak in Sunland, Inc. products (Centers for Disease Control and Prevention, 2012; U. S.

276 Food and Drug Administration, 2012).

PT
277

RI
278 Cleaning and sanitation trials were conducted on pilot-scale stainless steel nut butter pumping

279 equipment designed to mimic optimal processing conditions currently used in the nut butter

SC
280 industry (Codex Alimentarius, 2008).

281

282
U
Three independent cleaning/sanitizing trials were conducted for each sanitation treatment.
AN
283 Approximately 63.5 kg of commercial creamy peanut butter was heated in the steam jacketed

284 tank 2 (T2). Once the temperature of the peanut butter reached ~60oC, 800 g of inoculated
M

285 peanut butter was added to T2. Surface grown cultures (sessile) were used as studies have
D

286 shown their thermal resistance follows a linear trend (Keller et al., 2012). Water activity samples
TE

287 were collected immediately before inoculation and after 15 min mixing. The water activities of

288 the peanut butter before and after inoculation were 0.200 ± 0.045 and 0.203 ± 0.052,
EP

289 respectively. It was concluded that addition of the inoculum did not have a significant effect (n =

290 6; p > 0.05) on the water activity of the peanut butter.

C

291
AC

292 A temperature range of 60-65oC was chosen as the inoculation and processing temperature for

293 the peanut butter. In this temperature range the peanut butter remained in a molten state that

294 flowed freely through the piping. In addition, previous research has shown that Salmonella

295 serovars are thermally stable at temperatures under ~80oC in peanut butter (Keller et al., 2012;

13
 ACCEPTED MANUSCRIPT

296 Ma et al., 2009; Shachar and Yaron, 2006). Holding the peanut butter between 60-65oC allowed

297 the inoculated peanut butter to fully contaminate the system and ensured any microbial reduction

298 was not due to thermal death during the contamination process. The final microbial count of the

299 peanut butter used to contaminate the equipment was 7.3-7.4 log CFU/g (Table 2).

PT
300

RI
301 Although homogeneous contamination of the processing equipment was desired, it is noted that

302 in many foods, especially low-moisture foods, distribution of pathogens may not be

SC
303 homogeneous. Environmental samples were taken at areas of the equipment that came into

304 contact with the inoculated peanut butter product and on an area known not to come into contact

305
U
with the contaminated peanut butter (Table 1). The environmental swabs were taken from
AN
306 different locations in the peanut butter processing line to determine the background

307 contamination remaining on the equipment, and to determine the efficacy of the
M

308 cleaning/sanitation of the hot oil and isopropanol-based treatments (Figures 2 and 3).
D

309
TE

310 In addition to environmental swabs, samples were taken of inoculated peanut butter, cold oil

311 flush, and periodically of the hot oil circulated throughout the system (Table 2). Triplicate
EP

312 samples were obtained for each product at each sampling point. Following the removal of

313 inoculated peanut butter from the system, the initial cold oil flush provided ~1-log CFU/mL
C

314 reduction of Salmonella than initially present in the peanut butter present in the system. After
AC

315 the initial oil flush was drained, the hot oil treatment provided a second immediate reduction

316 from the initial microbial load. Both reductions with oil could be attributed to a simple dilution

317 effect. During the 2 h hot oil treatment, the temperature of the oil ranged from 86oC to 90oC

318 (measured at t = 0 and 120 min, respectively). Heat was applied to the system through a steam-

14
 ACCEPTED MANUSCRIPT

319 jacketed insulated main tank. Circulatory piping was not heated. Therefore, the oil in portions

320 of the equipment that were not heated and/or insulated may have been at a lower temperature

321 than oil in the heated tank. This is not atypical of what would be found in the industry.

322 Although there was variability in oil temperature for the duration of the hot oil treatment,

PT
323 microbial levels enumerated from all oil samples collected during the two hour hot oil circulation

RI
324 were not statistically different (p > 0.05; Table 2), indicating no significant thermal inactivation

325 occurred during hot oil treatment. This was not unexpected based on thermal kinetics found in

SC
326 the literature related to thermal destruction of Salmonella in this temperature range in low-

327 moisture environments. Salmonella inoculated into peanut butter is resistant to heat, with the

328
U
greatest inactivation occurring within the first 10-20 min of heat treatment (Ma et al., 2009;
AN
329 Shachar and Yaron, 2006). The same authors observed a maximum decrease of 3.2 log CFU/g at

330 90oC held for 50 min, for a cocktail of Salmonella Agona, S. Enteritidis, and S. Typhimurium.
M

331
D

332 The lack of a reduction in Salmonella counts in the hot oil throughout the hot oil treatment is also
TE

333 reflected in the environmental swab results after hot oil treatment (Figures 2 and 3). Total

334 microbial populations are clearly higher after hot oil treatment indicating the failure of hot oil to
EP

335 reduce populations to pre-contamination levels (data not shown). This is again obvious in

336 presumptive Salmonella populations (Figure 2).

C

337
AC

338 Since the results of all three trials of hot oil treatments failed to remove Salmonella from the

339 peanut butter processing equipment or provide a reasonable reduction in Salmonella populations,

340 using an oil treatment as a sanitation step would unlikely provide appropriate levels of

15
 ACCEPTED MANUSCRIPT

341 decontamination. However, no visible peanut butter residual was observed following the hot oil

342 treatment.

343

344 It is critical that such sanitizing is undertaken in a manner that avoids the creation of a further

PT
345 safety risk. Consequently, all processing equipment was allowed to return to ambient

RI
346 temperatures prior to the isopropanol based sanitation procedure. Equipment was allowed to

347 cool for 24 h prior to the use of isopropanol to ensure treatment below the flash point of

SC
348 isopropanol (40oC).

349

350
U
Following circulation of 60% isopropanol at ambient temperature for 1 hour, low levels of
AN
351 microbial contamination were detected on the processing equipment after plating on TSAYE. A

352 number of colonies on TSAYE were tested using the Vitek microbial identification system, but
M

353 none were found to be Salmonella. In addition, after isopropanol treatment there were no
D

354 presumptive Salmonella detected in environmental swabs when plated onto XLD (detection level
TE

355 0.16 log CFU/cm2) (Figure 2). Similar results were observed for sanitizing treatment using

356 isopropanol with added quaternary ammonium compounds as were found for the 60%
EP

357 isopropanol treatment (Figure 3). Less than 1.15 CFU/cm2 were enumerated on TSAYE from

358 equipment surfaces following the sanitation step, but again no Salmonella were detected after
C

359 treatment (Figure 3).

AC

360

361 Du et al. (2007), also investigated the use of isopropanol-based quaternary ammonium sanitizer

362 on the treatment of almond dust inoculated stainless steel surfaces, and found a reduction in

363 aerobic plate counts of ~3 log CFU/g. Further studies on the use of isopropanol-based

16
 ACCEPTED MANUSCRIPT

364 quaternary ammonium sanitizers found that they were more effective than aqueous quaternary

365 ammonium sanitizers (Du et al., 2010). The same authors also showed similar reductions in

366 Salmonella populations to results to those found here. In the previous study both isopropanol

367 and isopropanol-based quaternary ammonium sanitizer were able to reduce Salmonella counts

PT
368 from an initial contamination level of 5.2 log CFU/g to below 1.3 log CFU/g in almond huller-

RI
369 sheller dust. In our studies Salmonella reductions were found to be as great as 4-6 log CFU/cm2

370 depending on the initial contamination (Figures 2 and 3). The use of 60% isopropanol or 59%

SC
371 isopropanol with 0.02% added quaternary ammonium compounds aided in the sanitation of the

372 pilot-scale nut butter pumping equipment.

373
U
AN
374 4. Conclusions

375 Sanitation guidelines are lacking for facilities where the use of water-based cleaning and
M

376 sanitation procedures are not desirable. In this study, we have tested the use of hot oil and
D

377 isopropanol sanitizers to decontaminate a simulated peanut butter processing system.

TE

378 Unfortunately, hot oil did not result in any appreciable reduction in Salmonella levels other than

379 what would be expected through a dilution effect. However, hot oil was effective in
EP

380 cleaning/removing peanut butter from the processing equipment. A cleaning step is critical prior

381 to the application of a chemical sanitizer, particularly in the presence of high-fat foods such as
C

382 peanut butter, which may confer stability to the bacterial cells. Following removal of peanut
AC

383 butter, both 60% isopropanol and an isopropanol sanitizer containing quaternary ammonium

384 compounds, examined independently, were highly effective with respect to reduction of

385 Salmonella in the system. Both isopropanol-based sanitizers were able to effectively reduce

386 Salmonella contamination by as much as 5-log CFU/g, to below detection limits.

17
 ACCEPTED MANUSCRIPT

387

388 Typical industry recommendations with respect to good manufacturing practices call for

389 effective cleaning and sanitation at regularly scheduled intervals (Codex Alimentarius 1979,

390 Grocery Manufactures Association 2009a). A two-step, hot oil clean followed by isopropanol-

PT
391 based sanitizer as outlined here may provide such a procedure for an environment frequently

RI
392 deficient in documented effective cleaning and sanitation regiments. In addition, implementation

393 of such procedures could allow for effective lot separation, thus reducing the risk of recalls

SC
394 should contamination occur and can provide an effective means for reducing or eliminating

395 Salmonella contamination of nut butter processing equipment.

396
U
AN
397 5. Acknowledgements

398 This work was supported by the U.S. Food and Drug Administration and by appointments to the
M

399 Research Participation Program at the Center for Food Safety and Applied Nutrition
D

400 administered by the Oak Ridge Institute for Science and Education (ORISE) through
TE

401 an interagency agreement between the U.S. Department of Energy and the U.S. Food and Drug

402 Administration. This project was also supported by the U.S. Food and Drug Administration,
EP

403 Office of the Commissioner, Commissioner’s Fellowship Program (CFP). This project was also

404 supported by the International Life Sciences Institute (ILSI) North America.
C

405
AC

406 6. References

407 Beuchat, L.R., Komitopoulou, E., Beckers, H., Betts, R.P., Bourdichon, F., Fanning, S. Joosten,

408 H.M., Ter Kuile, B.H., 2013. Low-water activity foods: Increased concern as vehicles of

409 foodborne pathogens. Journal of Food Protection 76, 150-172.

18
 ACCEPTED MANUSCRIPT

410 Burnett, S.L., Gehm, E.R., Weissinger, W.R., Beuchat, L.R., 2000. Survival of Salmonella in

411 peanut butter and peanut butter spread. Journal of Applied Microbiology 89, 72–77.

412 Centers for Disease Control and Prevention, 2007. Multistate outbreak of Salmonella serotype

413 Tennessee infections associated with peanut butter—United States, 2006–2007.

PT
414 Morbidity and Mortality Weekly Report. 56(21):521–524.

RI
415 Centers for Disease Control and Prevention, 2009. Multistate outbreak of Salmonella infections

416 associated with peanut butter and peanut butter containing products—United States,

SC
417 2008–2009. Morbidity and Mortality Weekly Report 58, 85-90.

418 Centers for Disease Control and Prevention, 2011. Salmonella outbreaks.

419
U
http://www.cdc.gov/salmonella/outbreaks.html. Accessed: September 2011.
AN
420 Centers for Disease Control and Prevention, 2012. Multistate outbreak of Salmonella Bredeney

421 infections linked to peanut butter manufactured by Sunland, Inc. (Final Update).
M

422 http://www.cdc.gov/salmonella/bredeney-09-12/index.html. Accessed: November 2012.

D

423 Codex Alimentarius, 1979. Recommended international code of hygienic practice for groundnuts
TE

424 (peanuts). CAC/RCP 22-1979.

425 http://www.codexalimentarius.net/input/download/standards/267/CXP_022e.pdf .
EP

426 Accessed: November 2012.

427 Codex Alimentarius, 2008. Guidelines for the validation of food safety control measures.
C

428 CAC/GL 69-2008.

AC

429 http://www.codexalimentarius.net/input/download/standards/11022/cxg_069e.pdf.

430 Accessed: December 2012.

431 Du, W.-X., Danyluk, M.D., Harris, L.J., 2007. Evaluation of cleaning treatments for almond-

432 contact surfaces in hulling and shelling facilities. Food Protection Trends 27, 678-683.

19
 ACCEPTED MANUSCRIPT

433 Du, W.-X., Danyluk, M.D., Harris, L.J., 2010. Efficacy of aqueous and alcohol-based

434 quaternary ammonium sanitizers for reducing Salmonella in dusts generated in almond

435 hulling and shelling facilities. Journal of Food Science 75, M7-13.

436 Evans, M.R., Tromans, J.P., Dexter, E.L.S., Ribeiro, C.D., 1996. Consecutive Salmonella

PT
437 outbreaks traced to the same bakery. Epidemiology and Infection 116, 161-167.

RI
438 Grasso, E.M., Somerville, J.A., Balasubramaniam, V.M., Lee, K., 2010. Minimal effects of high-

439 pressure treatment on Salmonella enterica serovar Typhimurium inoculated into peanut

SC
440 butter and peanut products. Journal of Food Science 75, E522-526.

441 Grocery Manufacturers Association, 2009a. Control of Salmonella in low-moisture foods.

442
U
http://www.gmaonline.org/downloads/technical-guidance-and-
AN
443 tools/SalmonellaControlGuidance.pdf. Accessed: November 2010.

444 Grocery Manufacturers Association, 2009b. Annex to control of Salmonella in low-moisture

M

445 foods. http://www.gmaonline.org/downloads/technical-guidance-and-

D

446 tools/Salmonellaguidanceannex.pdf. Accessed: November 2010.

TE

447 Grocery Manufacturers Association, 2010. Industry handbook for safe processing of nuts.

448 http://www.gmaonline.org/downloads/technical-guidance-and-
EP

449 tools/Executive_Summary_NutIndustryHandbook_22Feb10.pdf. Accessed: November

450 2011.
C

451 International Life Science Institute Europe, 2011. Persistence and survival of pathogens in dry
AC

452 foods and dry food processing environments.

453 http://www.ilsi.org/Europe/Documents/Persistence%20and%20survival%20report.pdf.

454 Accessed: January 2013.

20
 ACCEPTED MANUSCRIPT

455 Kamineni, P., Halik, L., Grasso, E., Keller, S., Grove, S., Jackson, L., Chirtel, S., 2011.

456 Sanitation guidelines for almond and nut butter processing. Final Report, Almond Board

457 of California.

458 Keller, S., Grasso, E., Halik, L., Fleischman, G., Chirtel, S., Grove, S., 2012. Effect of growth on

PT
459 the thermal resistance and survival of Salmonella Tennessee and Oranienburg in peanut

RI
460 butter, measured by a new thin-layer thermal death time device. Journal of Food

461 Protection 75, 1125-1130.

SC
462 Killalea, D., Ward, L.R., Roberts, D., de Louvois, J., Sufi, F., Stuart, J.M., Wall, P.G., Susman,

463 M., Schwieger, M., Sanderson, P.J., Fisher, I.S.T., Mead,P.S., Gill, O.M., Bartlett,

464
U
C.L.R., Rowe, B., 1996. International epidemiological and microbiological study of
AN
465 outbreak of Salmonella Agona infection from a ready to eat savoury snack—I: England

466 and Wales and the United States. British Medical Journal 313, 1105-1107.
M

467 Kusumaningrum, H.D., Riboldi, G., Hazeleger, W.C., Beumer, R.R., 2003. Survival of
D

468 foodborne pathogens on stainless steel surfaces and cross-contamination to foods.

TE

469 International Journal of Food Microbiology 85, 227-236.

470 Ma, L., Zhang, G., Gerner-Smidt, P., Mantripragada, V., Ezeoke, I., Doyle, M.P., 2009. Thermal
EP

471 inactivation of Salmonella in peanut butter. Journal of Food Protection 72, 1596-1601.

472 Mattick, K.M., Jørgensen, F., Wang, P., Pound, J., Vandeven, M.H., Ward, L.R., Legan, J.D.,
C

473 Lappin-Scott, H.M., Humphrey, T.J. 2001, Effect of challenge temperature and solute
AC

474 type on heat tolerance of Salmonella serovars at low water activity. Applied and

475 Environmental Microbiology 67, 4128-4136.

476 Park, E.J., Oh, S.W., Kang, D.H., 2008. Fate of Salmonella Tennessee in peanut butter at 4 and

477 22°C. Journal of Food Science 73, M82-86.

21
 ACCEPTED MANUSCRIPT

478 Podolak, R., Enache, E., Stone, W., Black, D.G., Elliot, P.H., 2010. Sources and risk factors for

479 contamination, survival, persistence, and heat resistance of Salmonella in low-moisture

480 foods. Journal of Food Protection 73, 1919-1936.

481 Schaffner, D.W., Buchanan, R.L., Calhoun, S., Danyluk, M.D., Harris, L.J., Djordjevic, D.,

PT
482 Whiting, R.C., Kottapalli, B., Wiedmann, M., 2013. Issues to consider when setting

RI
483 intervention targets with limited data for low-moisture food commodities: A peanut case

484 study. Journal of Food Protection 76, 360-369.

SC
485 Scheil, W., Cameron, S., Dalton, C., Murray, C., Wilson, D., 2008. A South Australian

486 Salmonella Mbandaka outbreak investigation using a database to select controls.

487
U
Australian and New Zealand Journal of Public Health 22, 536-539.
AN
488 Scott, V.N., Chen, Y., Freier, T.A., Keuhm, J., Moorman, M., Meyer, J., Morille-Hinds, T., Post,

489 L., Smoot, L., Hood, S., Shebuski, J., Banks, S., 2009. Control of Salmonella in low-
M

490 moisture foods I: minimizing entry of Salmonella into a processing facility. Food
D

491 Protection Trends 29, 342-353.

TE

492 Shachar, D.,Yaron, S., 2006. Heat tolerance of Salmonella enterica serovars Agona, Enteritidis,

493 and Typhimurium in peanut butter. Journal of Food Protection 69, 2687-2691.
EP

494 U. S. Food and Drug Administration, 2004. 2004 Recalls, market withdrawals, & safety alerts.

495 http://www.fda.gov/Safety/Recalls/ArchiveRecalls/2004/default.htm. Accessed October

C

496 2013.
AC

497 U. S. Food and Drug Administration, 2009a. 2009 Recalls, market withdrawals, & safety alerts.

498 http://www.fda.gov/Safety/Recalls/ArchiveRecalls/2009/default.htm. Accessed October

499 2013.

22
 ACCEPTED MANUSCRIPT

500 U. S. Food and Drug Administration, 2009b. Guidance for industry: Measures to address the risk

501 for contamination by Salmonella species in food containing a peanut-derived product as

502 an ingredient.

503 http://www.fda.gov/Food/GuidanceComplianceRegulatoryInformation/GuidanceDocuments/Pro

PT
504 duceandPlanProducts/ucm115386.htm. Accessed: May 2011.

RI
505 U. S. Food and Drug Administration, 2010. 2010 Recalls, market withdrawals, & safety alerts.

506 http://www.fda.gov/Safety/Recalls/ArchiveRecalls/2010/default.htm. Accessed October

SC
507 2013.

508 U.S. Food and Drug Administration, 2012. Inspection report: FEI number 1000117188.

509
U
http://www.fda.gov/downloads/AboutFDA/CentersOffices/OfficeofGlobalRegulatoryOpe
AN
510 rationsandPolicy/ORA/ORAElectronicReadingRoom/UCM324793.pdf. Accessed April

511 2013.
M

512 Woodroof, J.G. 1983, Peanuts: production, processing, products, 3rd edition. AVI Publishing
D

513 Company, Inc., Westport CT.

TE

514

515
C EP
AC

23
 ACCEPTED MANUSCRIPT

a
2
1

PT
4

7 6 5

RI
3

U SC
AN
516

517 b
M

518

519
D

520
TE

521

522
EP

523

524
C

525
AC

526 Figure 1. a. A schematic diagram of the peanut butter pumping machine. bLocations of

527 environmental samples taken from the machine. 1. Exit piping; 2. Upper elbow; 3. Left elbow of

528 lower pipe, left of positive displacement pump; 4. Right side of lower pipe, immediately before

24
 ACCEPTED MANUSCRIPT

529 positive displacement pump; 5. Lower pipe, immediately after positive displacement pump; 6.

530 Bottom exit of T2; 7. Bottom exit of T1.

531

532

PT
533

RI
534

535

SC
536

537

538
U
AN
539

540
M
D
TE
C EP
AC

25
 ACCEPTED MANUSCRIPT

PT
RI
SC
541

U
542 Figure 2. Salmonella populations remaining on the peanut butter equipment at seven different
AN
543 locations as tested with environmental swabs plated onto xylose lysine deoxycholate (XLD) agar

544 with a limit of detection of 0.16 log CFU/cm2. Samples were taken before the start of a new
M

545 trial, after the hot oil treatment, after system cool down, and after the isopropanol treatment.

546
D

547
TE
C EP
AC

26
 ACCEPTED MANUSCRIPT

PT
RI
SC
548

U
549 Figure 3. Salmonella populations remaining on the peanut butter equipment at seven different
AN
550 locations as tested with environmental swabs plated onto xylose lysine deoxycholate (XLD) agar

551 with a limit of detection of 0.16 log CFU/cm2. Samples were taken before the start of a new
M

552 trial, after the hot oil treatment, after system cool down, and after the isopropanol with added

553 quaternary ammonium compounds treatment.

D

554
TE

555
C EP
AC

27
 ACCEPTED MANUSCRIPT

556 Table 1. Environmental sampling sites and rationale for sampling site inclusion.

Environment sampling Direct contact with

Sample number Rationale
site peanut butter/sanitizer
This would be where
Exit piping of tank 2
1 product was filled into yes
(T2)
packaging

PT
Complete fill of this
Elbow of the upper angle of pipe may be
2 yes
recirculation pipe of T2 difficult and have a

RI
decreased contact-time
Higher chance or
contamination due to
Left elbow of lower

SC
gravitational force
3 pipe, left of positive yes
leading towards
displacement pump
positive displacement
pump

U
Higher chance or
Right side of lower contamination due to
AN
pipe, immediately gravitational force
4 yes
before positive leading towards
displacement pump positive displacement
M

pump
Small voids may occur
Lower pipe,
due to slippage of
immediately after
5 positive displacement yes
D

positive displacement
pump leading to lower
pump
contact-time
TE

All product/sanitizer
leaving the heated
vessel exits through
EP

6 Bottom exit of T2 this port, higher yes

chance or
contamination and
longer contact-time
C

Contamination should
Bottom exit of tank 1 not occur unless there
AC

7 no
(T1) were pump failure
issues
557

558

28
 ACCEPTED MANUSCRIPT

559 Table 2. Microbial counts in the product sample at various time points throughout the peanut

560 butter inoculation. Samples were plated onto tryptic soy agar with yeast extract (TSAYE) and

561 xylose lysine deoxycholate (XLD) agar.

Microbial survival (log CFU/g) of producta

PT
Product sample TSAYE XLD
Inoculated peanut butter 7.4 ± 0.4 A 7.3 ± 0.4 a
Initial oil flush 6.3 ± 0.5 B 6.5 ± 1.3 b

RI
Hot oil (t = 0 min) 4.5 ± 1.0 C 4.5 ± 1.0 c
Hot oil (t = 30 min) 4.0 ± 0.3 C 3.7 ± 0.5 c
Hot oil (t = 60 min) 3.6 ± 0.4 C 3.6 ± 0.9 c

SC
Hot oil (t = 90 min) 3.6 ± 0.8 C 3.4 ± 1.4 c
Hot oil (t = 120 min) 3.2 ± 1.2 C 3.3 ± 1.6 c
a
562 Different capital letters indicate significant differences (p < 0.05) between sampling points
563 measured on TSAYE media; different lowercase letters indicate significant differences (p < 0.05)

U
564 between sampling points measured on XLD media.
565
AN
M
D
TE
C EP
AC

29