Influence of water activity on processing resistance of Salmonella serovars and implications on sanitization of pistachios by heat and ozone

THESIS

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of The Ohio State University

By

Marilia Peña-Meléndez

Graduate Program in Food Science and Technology

The Ohio State University

2011

Master's Examination Committee:

Ahmed E. Yousef, Advisor

Luis Rodriguez-Saona

Sheryl Barringer

Copyrighted by

Marilia Peña-Meléndez

2011

ABSTRACT

Recent salmonellosis outbreaks due to consumption of low moisture foods such as nuts have been well documented. Low water activity (aw) plays an important role in these foods because it limits the growth of microorganisms, especially foodborne pathogens.

Additionally, the cross-protective effect that happens in microorganisms when they are exposed to more than one stress condition (e.g. low aw and heat) is well documented and has become a concern to the food industry.

The objectives of this study were to: (i) investigate the effect of low aw, physiological state, and type of humectants on the heat resistance of three Salmonella enterica serovars (Saintpaul 02-109, Tennessee 2053H, and Elmsbuettel 1236H); (ii) evaluate the use of a heat-ozone combination treatment to reduce populations of

Salmonella Enteritidis on inoculated pistachios, (iii) determine the effect of inoculation method and storage time on the effectiveness of the heat-ozone combination treatments; and (iv) assess the potential of using a heat resistant pistachio isolate, Enterococcus faecium OSY 31284, as a Salmonella surrogate when determining the effectiveness of heat-ozone combination treatments.

For the first part of this study, Salmonella serovars that were adapted or non-adapted to low aw were heat-treated at 55°C for up to 45 min; samples were removed at 5 min intervals. Lowering aw was achieved by modifying Tryptic Soy Broth (TSB) composition to contain 20% glycerol (aw 0.94), 4% NaCl (aw 0.97), or 35% sucrose (aw 0.95). Results showed that type of humectant and physiological state (adapted vs. non-adapted) significantly affected Salmonella D55°C-value (P<0.05). Cells adapted in 20% glycerol ii showed lower D55°C-values, compared to the non-adapted cells. Presence of NaCl and sucrose in the broth had a protective effect under both the adapted and non-adapted conditions for the three serovars tested as compared to glycerol.

For the second part of the study, pistachios were inoculated with either Salmonella

Enteritidis or E. faecium OSY 31284. In addition to variations in bacterial species, other variables tested were inoculation methods (vacuum and no vacuum) and storage times

(24hrs and 72hrs); all possible combinations were performed. Treatments were comprised of heat (5% brine at 70°C for 10 minutes), ozonation (160 g/m3, 12.5 psig, 30 min holding time), heat followed by ozonation or ozonation followed by heat. Populations of the inoculated bacteria were determined before and after treatments. Results showed significant reduction of E. faecium on pistachios when treated with heat-ozone combinations (P<0.05), but the reduction was not significantly greater than that achieved by heat alone. Various treatments inactivated Salmonella Enteritidis to a greater extent than that observed with E. faecium. For both bacteria, the application of gaseous ozone alone showed little reduction. Additionally, inoculation method alone and storage time alone did not have a significant effect on the log reductions of both bacteria. However, the interaction of 72hrs storage and inoculation using vacuum influenced the efficacy of the treatments the most. The data also suggest that E. faecium is not a suitable surrogate for Salmonella when evaluating pistachio decontamination processes, because the survivors of the two bacteria in response to combination treatments differ by more than 2 log/g.

iii

DEDICATION To my parents, Orlando and Awilda.

iv

ACKNOWLEDGMENTS

I would like to thank my advisor Dr. Ahmed Yousef and Dr. Jennifer Perry for their support, continuous guidance and encouragement. I wouldn’t have got to this point if it weren’t for them. To my committee members Dr. Luis Rodriguez-Saona and Dr. Sheryl

Barringer for their suggestions and guidance.

To the rest of the lab members: Amrish Chawla, Goksel Tirpanci, Mustafa Yesil, Yuan

Yan, En Huang, David Kasler, and Seth Costello, I thank them for their support and suggestions throughout this time.

My great appreciation goes to Edward for his patience and encouragement to continue, to my family Orlando, Awilda, Andria and Orlando, Jr. for their unconditional support, and finally, to my friends Leyre, Melvili, Lumarie, and Lizanel for being there when I needed them the most.

v

VITA

November 24, 1986 ...... Born, Mayaguez, Puerto Rico.

2009...... B.S. Industrial Microbiology, University of

Puerto Rico, Mayaguez

2009 to present ...... Graduate Research Associate, The Ohio

State University

FIELDS OF STUDY

Major Field: Food Science and Technology

vi

TABLE OF CONTENTS

Abstract…...... ii

Dedication ...... iv

Acknowledgments...... v

Vita……...... vi

List of Tables ...... ix

List of Figures ...... x

Chapter 1. Literature Review ...... 1

1.1 Low Moisture Foods and associated outbreaks ...... 1

1.2 Tree nuts ...... 3

1.3 Salmonella ...... 5

1.4 Validation studies ...... 10

1.5 Decontamination Treatments ...... 12

1.6 Ozone and its use in food ...... 19

vii

Chapter 2. Effect of low Water Activity, physiological state, and type of humectant on the

heat resistance of Salmonella serovars Saintpaul 02-109, Tennessee 2053H and

Elmsbuettel 1236H ...... 28

Chapter 3. Decontamination of pistachios using heat and gaseous ozone combination ... 49

References ...... 66

viii

LIST OF TABLES

Table 1. 1. PPO pasteurization operating parameters. Obtained from Almond Pasteurization Using Propylene Oxide (PPO) Standard Operating Procedure (SOP), Almond Board of California...... 17

Table 2.1. D55°C- value of Salmonella serovars Saintpaul 02-109, Tennessee 2053H, and Elmsbuettel 1236H after heat treatment at 55°C in low aw broths...... 39

Table 2.2. Average of D55°C- value for Salmonella serovars Saintpaul 02-109, Tennessee 2053H, and Elmsbuettel 1236H after heat treatment at 55°C in low aw broths...... 39

Table 3.1 Summary of results from pistachios inoculated with E. faecium and treated with heat-ozone combination...... 57

Table 3.2 Summary of results from pistachios inoculated with S. Enteritidis and treated with heat-ozone combination...... 58

ix

LIST OF FIGURES

Figure 1.1. Hulled, unshelled pistachios...... 4

Figure 1.2. Comparison of the reduction of Geobacillus stearothermophilus spores and Salmonella Enteritidis PT 30 in almonds during and after commercial propylene oxide (PPO) treatment. Obtained from (Danyluk, Uesugi and Harris 2005)...... 19

Figure 1. 3. Corona discharge representation. Adapted from Kim, Yousef, and Dave 1999...... 22

Figure 2.1. Representation of procedure used for transfers of cultures (a total of three, three days, four 24hr cycles including transfer of stock culture) and preparation of samples before heat treatment...... 33

Figure 2.2. Growth of Salmonella enterica serovars Saintpaul 02-109, Tennessee 2053H, and Elmsbuettel 1236H in the low aw broths containing 20% glycerol (A), 4% NaCl (B), and 35% sucrose (C) during three transfers (4 growth cycles of 24hrs each)...... 38

Figure 2.3. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Saintpaul 02- 109 adapted and non-adapted in 20% glycerol...... 40

Figure 2.4. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Saintpaul 02- 109 adapted and non-adapted in 4% NaCl ...... 41

Figure 2.5. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Saintpaul 02- 109 adapted and non-adapted in 35% sucrose ...... 41

Figure 2.6. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Tennessee 2053H adapted and non-adapted in 20% glycerol ...... 42

Figure 2.7. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Tennessee 2053H adapted and non-adapted in 4% NaCl...... 42

Figure 2.8. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Tennessee 2053H adapted and non-adapted in 35% sucrose...... 43 x

Figure 2.9. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Elmsbuettel 1236H adapted and non-adapted in 20% glycerol ...... 43

Figure 2.10. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Elmsbuettel 1236H adapted and non-adapted in 4% NaCl...... 44

Figure 2.11. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Elmsbuettel 1236H adapted and non-adapted in 35% sucrose...... 44

Figure 3.1. Ozone concentration (g/m3) and pressure (psig) as it changes by time. Total treatment time is 30 min, in addition to 30 min exhaust; total time is 60 min...... 54

Figure 3.2. Log reduction of Salmonella Enteritidis inoculated pistachios with and without use of vacuum and storage of 24hrs or 72hrs after combination treatments (heat/ozone)...... 59

Figure 3.3. Log reduction of Enterococcus faecium OSY 31284 inoculated pistachios and without use of vacuum and storage of 24hrs or 72hrs after combination treatments (heat/ozone)...... 60

xi

CHAPTER 1. LITERATURE REVIEW

1.1 LOW MOISTURE FOODS AND ASSOCIATED OUTBREAKS

Low moisture foods can be defined as foods with a water activity (aw) below 0.85

(FDA 2010). Some of these foods include nuts (peanuts, almonds, pistachios), cereals, chocolate, pet food, and powdered milk, among others. Most bacteria do not grow at aw below 0.85 and in general, bacteria require higher aw values than fungi (Jay, Loessner and

Golden 2005). Foods with low aw were once considered to be safe since it was thought that the environment would not provide the necessary conditions for bacterial growth, especially growth of foodborne pathogens. However, recent cases of food recalls and outbreaks have changed how the food industry views and handles these types of products.

In 2006, a multistate outbreak of Salmonella Tennessee associated with the consumption of two brands of peanut butter (Peter Pan and Great Value) occurred (CDC

2007). Both brands were produced in the same plant and the outbreak strain was isolated from opened and unopened jars as well as from environmental samples taken at the plant

(CDC 2007). Cases were reported from 47 states and a total of 628 individuals were infected with the outbreak strain (CDC 2007). In 2008, another multistate outbreak associated with peanut butter and peanut butter containing products was reported.

Salmonella Typhimurium was isolated and identified as the outbreak strain and the source was traced to a single brand: King Nut, produced by Peanut Corporation of

America (PCA) (CDC 2009a). The company distributes their product only in bulk (not sold directly to consumers). A total of 529 people from 43 states and one person from 1

Canada were reported to have been infected with the outbreak strain (CDC 2009a).

Peanut products; including peanut butter, peanut paste and peanut meal, were also produced by this same company and sold to other food companies for use in diverse products (CDC 2009a). As a result, the recall affected many companies. Pistachios and pistachio products have also been associated with a Salmonella outbreak in 2009. The source was linked to one company in California and different Salmonella strains, including Montevideo, Newport, and Senftenberg, were isolated (CDC 2009b).

In 2008, another outbreak was investigated, but this time it was associated with cereal (unsweetened puffed rice and wheat cereals) (CDC 2008). Routine testing of the food detected the presence of Salmonella and following this discovery the recall was issued (CDC 2008). Salmonella Agona was identified as the outbreak strain and was isolated from the people affected, a total of 28 (CDC 2008).

There are many other cases of recalls and outbreaks related to low aw foods, including chocolate, toasted oats cereal, and milk powder, among others (GMA 2009).

These products have become a major concern for the food industry. Although in many cases the source of contamination is still unknown, there are several factors that may have played a role, including inappropriate sanitation practices, inadequate maintenance of equipment and facilities, inadequate handling of ingredients, and untrained personnel, among others. It is also important to keep track and obtain ingredients from reputable suppliers.

2

1.2 TREE NUTS

Composition

Tree nuts are widely consumed for their nutritious attributes and convenience.

Nutritional content varies with type of nut, geographical location where grown, and climate, among other factors. They are eaten alone, roasted, flavored, and added to different dishes. Storage of nuts has become important to the food industry because of safety and quality issues.

Tree nuts and edible seeds are rich in nutrients and are composed mainly of lipids and proteins (Alasalvar and Shahidi 2009; Venkatachalam and Sathe 2006). The moisture content of nuts is usually less than 10% and is influenced by the climate at the time of growth, storage conditions and time of harvest (Alasalvar and Shahidi 2009). The lipid content generally consists of monounsaturated fatty acids (MUFAs) and polyunsaturated fatty acids (PUFAs) (Venkatachalam and Sathe 2006). Typically, the lipid content ranges from 44% (e.g. almond, pistachio) up to more than 60% (e.g. hazelnut, walnut), depending on the nut, and growing conditions (Alasalvar and Shahidi

2009). Other important components include: amino acids, carbohydrates (primarily fibers and simple sugars), vitamins, phenolics, and antioxidants (Alasalvar and Shahidi

2009).

3

Pistachios

General characteristics

Pistachio trees (Pistacia vera L.) were native to Asia before being introduced to

Europe (Woodroof 1967). The nut grows in a ―grape-like cluster‖ and has an outer skin or hull that covers it (Alasalvar and Shahidi 2009). During ripening, the shell, which is light wood in color, splits open and the kernel (seed) can be seen. The kernel has a bright green color which can be eaten raw, roasted, salted or incorporated in different products.

Currently, the biggest grower and producer of pistachios in the world is located in San

Joaquin Valley, CA (Paramount Farms 2011).

Figure 1.1. Hulled, unshelled pistachios.

4

Harvesting and Processing

The pistachio growing season is from April to September and automated harvesting starts in late August, lasting all the way through September (Paramount Farms 2008).

During harvesting, a practice employed by some companies is the use of equipment that prevents pistachios from touching the ground because this minimizes microbial contamination. However, nuts may be collected from the ground surface. Pistachios need to be processed within 24 hours after they are harvested to eliminate the moisture trapped by the hull (Alasalvar and Shahidi 2009). Common processing steps include: coarse cleaning to remove leaves and twigs; hulling to remove outer layer or hull of the nut; pre- drying to remove excess water; and drying to reduce moisture content before storage

(Paramount Farms 2008). Subsequently, pistachios are stored in ventilated silos for up to two years before going into packaging. Pistachios are graded, hand sorted and then packaged or transported for further processing. Nuts that are transported are brined

(product is soaked in saturated salt brine), strained, roasted (i.e. placed in rotary roasters at 360-400°F), and finally, packaged (Paramount Farms 2008).

1.3 SALMONELLA

Characteristics and general information

Salmonella spp. are small, Gram-negative, non-sporeforming rods that are commonly distributed in nature (Jay, Loessner and Golden 2005; FDA 2009). Members of this genus are facultative anaerobes, mostly motile with peritrichous flagella, catalase positive, and oxidase negative (Bell and Kyriakides 2002). Although the main reservoir is 5 the intestinal tract of animals (in particular poultry) and humans, salmonellae can also be found in soil, water, insects, and anything that comes in contact with a contaminated source (Jay, Loessner and Golden 2005; Bad Bug Book 2009). The optimum temperature for growth is 37°C, but members of this genus can survive and grow between 46.2°C and

5.3°C (Bell and Kyriakides 2002; Jay, Loessner and Golden 2005). Growth at a pH of

4.05 ± 0.05 has been reported under ideal conditions (high aw, absence of inhibitors, high nutrient content) (Chung and Goepfert 1970) and minimum aw for growth has been determined at 0.94; but cells can remain dormant and viable below this value (Jay,

Loessner and Golden 2005; Mattick et al. 2000).

Illness

Salmonella spp. are considered to be one of the most important foodborne pathogens

(Bell and Kyriakides 2002) and are the causative agent of salmonellosis. Salmonellosis is a type of gastroenteritis that presents typical symptoms including nausea, vomiting and fever and can become chronic (FDA 2009). The disease is caused when the bacteria grows and multiplies in the host, colonizes and penetrates small intestine tissues, and subsequently, causes inflammation (possibly due to the production of an enterotoxin)

(FDA 2009; Bell and Kyriakides 2002).

Although salmonellosis is the most common illness caused by Salmonella spp., typhoid fever, a more severe illness, can also result from exposure to members of this genus. Typhoid fever is caused by the serovars Typhi and Paratyphi and symptoms include fever, malaise, and cough. Typhoid fever can progress to severe complications 6 including hemorrhages myocarditis, hepatitis, pneumonia, thrombocytopenia and hemolytic uremic syndrome (WHO 2003).

There is a diverse array of foods associated with Salmonella spp. ranging from raw animal products to dry foods and produce. Some of the most common implicated foods are: raw/undercooked meats, poultry, milk and other dairy products, eggs, and most recently, nuts (FDA 2009; Bell and Kyriakides 2002; Jay, Loessner and Golden 2005).

Taxonomy

Serotyping

Salmonella spp. belong to the family of Enterobacteriaceae which also includes

Escherichia coli. Initially, each serotype was designated as a different species and as a result the genus consisted of several species (Brenner et al. 2000). Today, the genus

Salmonella is comprised of two species: S. enterica and S. bongori, with most of the serotypes belonging to S. enterica (Brenner et al. 2000). To differentiate between serotypes several methods can be used. One of these methods consists of establishing differences among the antigens present, also known as the antigenic formula: somatic antigen (O); flagellar antigen (H), which can be present in one of two phases; and, in some cases the capsular antigen (V) (Bell and Kyriakides 2002).

7

Phage typing

A second method used to distinguish strains is called phage typing. This method consists of establishing the sensitivity of the microorganisms to the lytic action of bacteriophages (Bell and Kyriakides 2002). The type is designated by assigning a definite type (DT) and provisional type (PT) number, which indicate the status of the designation

(Anderson et al. 1977).

Within S. enterica, there are six subspecies: enterica I, salamae II, arizonae IIIa, diarizonae IIIb, boutenae IV, and indica VI (Bell and Kyriakides 2002; Brenner et al.

2000). Genus, species and subspecies names should be in italics and the serotype name should be capitalized, but not in italics. The numbers I, II, IIIa, IIIb, IV and VI represent the DNA hybridization subgroup; S. bongori belongs to the subgroup V (Bell and

Kyriakides 2002). In other words, the complete name should be for example, Salmonella enterica subspecies enterica I serovar Enteritidis.

Resistance to unfavorable conditions

Salmonella spp. have been reported to have the ability to survive harsh environmental conditions including low aw, low and high pH and heat. They are not eliminated during drying, refrigeration or freezing (GMA 2010). Water activity of foods can be lowered by the addition of salts, sugars and other water binding compounds.

Salmonella spp. require a aw of 0.94 or above for growth, but there have been many reports suggesting that cells can survive for extensive periods of time at lower aw (Bell

8 and Kyriakides 2002; Burnett et al. 2000; Kieboom et al. 2006). Mattick et al. (2000b) conducted a study with Salmonella enterica serovar Enteritidis PT4 and Salmonella enterica serovar Typhimurium DT104 which showed that these serovars survived at low aw for long periods of time (TSB supplemented with 8% NaCl over a 5 month period).

They also suggest, consistent with other studies, that the survival of the microorganisms is dependent on the type of solute added to the media or food, the temperature at which they are growing (more survival at 21°C than at 37°C) and the expression of a stationary- phase sigma factor known as RpoS (Mattick et al. 2000b). RpoS is important for survival when stressful conditions like osmotic stress, high temperatures and presence of some chemicals occur (White 2007).

As mentioned previously, the lowest pH at which Salmonella spp. has been shown to grow is 4.05 (Chung and Goepfert 1970). Studies have found that adapting Salmonella spp. to acidic conditions favors their survival and persistence in fermented foods, such as cheeses (Leyer and Johnson 1992). Conversely, it has been shown that exposure and adaptation of cells to acidic conditions (sub-optimal pH) may aid the effectiveness of a heat treatment (Bell and Kyriakides 2002). Differences among serotypes show notable variability in pH resistance and sensitivity (Bell and Kyriakides 2002).

Salmonella spp., as well as other foodborne pathogens, have been shown to be tolerant or resistant to diverse environments. For this reason, in food processing there are many factors to take into consideration prior to establishing processing parameters.

9

1.4 VALIDATION STUDIES

When evaluating a new technology or treatment for use in food it is important to perform validation studies in order to determine process efficacy and ensure finished product safety. Validation studies consist of: using a surrogate microorganism that is of equal or greater resistance than the target microorganism and that it is safe to use in a food processing plant; and of evaluating the operation and equipment itself (GMA 2010).

The resistance to the treatment under study of the surrogate selected should be correlated to the resistance of the microorganism of concern. It is also important to determine the target reduction level of the organism of concern for a particular product and the type of process under evaluation (GMA 2009). Enterococcus faecium NRRL B-2354 is an example of a Salmonella surrogate that has been widely studied for use in almond and thermal processes (ABC 2007c; Harris 2009).

The use of nonpathogenic surrogates, when selected properly, provide a good substitute for the pathogenic microorganism that help in the design of new and effective decontamination technologies as well as minimizing the introduction of potentially hazardous waste to active processing facilities. In addition, it is important to keep in mind that a surrogate that works for a specific treatment and a particular food may not be as effective for other treatments and foods.

Enterococcus faecium

E. faecium, previously designated Pediococcus sp., are Gram positive cocci, homofermentative and thermoduric lactic acid bacteria (Jay, Loessner and Golden 2005; 10

Jeong, Marks and Ryser 2011). They have been found to be conservative surrogates for specific strains of Salmonella in many studies evaluating thermal treatments in almonds

(Jeong et al. 2011).

The Almond Board of California (ABC) (2007c) in conjunction with different universities and laboratories have attempted to identify surrogates for Salmonella

Enteritidis Phage Type 30. They identified E. faecium NRRL B-2354 as a surrogate for use in almond thermal processes and established the ―Guidelines for process validation using E. faecium NRRL B-2354‖. In this document, the ABC presents a protocol where they describe how to carry out an experiment using a surrogate; from preparation of the inoculum to interpretation of results (ABC 2007c).

Several studies have been conducted to evaluate the use of nonpathogenic microorganisms, such as E. faecium, as surrogates for Salmonella and other foodborne pathogens. Jeong and collegues (2011) evaluated E. faecium as a nonpathogenic surrogate for Salmonella Enteritidis PT30 and its thermal inactivation on the surface of almonds undergoing moist-air, forced convection heating. Their results show that E. faecium can be a useful, although conservative, surrogate for S. Enteritidis PT30 during moist-air heating of almonds in the targeted reduction range of 4 to 5 log, with average log reductions resulting lower for E. faecium when compared with S. Enteritidis PT30

(Jeong, Marks and Ryser 2011).

11

1.5 DECONTAMINATION TREATMENTS

In recent years, various technologies have been developed to minimize and/or eliminate harmful bacteria from food, in particular low moisture foods. There is ongoing research evaluating the use of different treatments in nuts, mostly almonds, and some of these have already been approved as pasteurization treatments. Some treatments that have been approved and are currently used by processors for nuts include: dry roasting, oil roasting, blanching, steam pasteurization, hot water pasteurization, propylene oxide, ethylene oxide, or combinations of treatments (GMA 2010). In order for a process to be considered effective, a minimum of 4 log reduction should be achieved for almonds

(Federal Register 2007) and 5 log reduction for peanuts (FDA 2011b) and pistachios

(FDA 2011a).

Oil and Dry Roasting

Oil and dry roasting are common thermal processes used in the industry to reduce or eliminate microorganisms in raw almonds and to give the product an appealing roasted flavor. Dry roasting can be accomplished by using a continuous conveyor roaster (single or multiple staged) or a rotary roaster (ABC 2007b). Different time and temperature combinations can be used as long as a minimum of 4 log reduction is achieved. Typical dry roasting temperatures range from 265°F to 310°F with time variations depending on the temperature (ABC 2007b). The ABC recommends 121°C/250°F for 100 min or

149°C/300°F for 9 minutes for a 4 log reduction (Harris 2009).

12

Oil roasting is considered to be a faster process when compared to dry roasting and it consists of a pre-heated oil tank containing a continuous conveyor that transports the almonds through the tank; almonds will be submerged in the oil (ABC 2007d). The temperatures most commonly used in industry are 280 to 350°F (138 to 149°C) for 3 to

15 minutes, with a minimum exposure time of 2 minutes (ABC 2007d). Temperature and time parameters may vary depending on initial temperature of the oil, moisture level of the almonds, equipment (e.g. volume of tank/oil), and roasting degree (ABC 2007d).

A study conducted by Du et al. (2010) evaluated the effects of hot oil on almonds inoculated with Salmonella Enteritidis PT 30 and Salmonella Senftenberg 775W. Their results show a reduction of 2.9, 3.0, and 3.6 log CFU/g for S. Enteritidis after the first 30 seconds of oil exposure at temperatures of 116, 121, and 127°C, respectively (Du et al.

2010). S. Senftenberg showed similar results at 127°C and an approximate 5 log reduction was achieved in less than 1.5 minutes at 127°C for both serotypes (Du et al.

2010). Therefore the authors conclude that at the temperatures established for commercial oil roasting of almonds (280 to 350°F [138 to 149°C]) for 3 to 15 minutes, over 5 log reduction of Salmonella can be achieved (Du et al. 2010).

Even though oil roasting and dry roasting are effective decontamination treatments, they may present processing challenges when trying to maintain a constant processing temperature and an even distribution of heat.

13

Infrared (IR) heating

Infrared (IR) heating has been recently incorporated into different food processing operations, including baked goods and meats, because of its efficiency and fast heating capabilities. During IR heating, the heat generated will be absorbed by the food when exposed to radiation within the wavelength of 0.78-1000µm (Jun and Irudayaraj 2009).

The thermal energy transferred by IR heating is in the form of electromagnetic (EM) waves and in the spectrum it is located between visible light and microwaves (Jun and

Irudayaraj 2009). Since foods are composed of diverse components, like carbohydrates, amino acids and lipids, these components will absorb at different wavelengths within the

IR spectrum (Jun and Irudayaraj 2009).

A study conducted by Brandl and colleagues (2008), evaluated the use of IR heating for dry pasteurization of inoculated raw almond kernels and its effect in reducing

Salmonella Enteritidis populations. Their results show that exposing the almond kernels to IR heat for 30, 35 and 45 seconds raised their surface temperatures to 90, 102, and

113°C, and subsequently, when placed at room temperature to cool, it resulted in a 0.63,

1.03 and 1.51 log reduction of S. Enteritidis, respectively (Brandl et al. 2008).

Additionally, they state that the most effective treatment consisted of IR heating, followed by holding at a warm temperature (initially a maximum of 109°C and then decreased to 80°C) for 60 minutes (Brandl et al. 2008). This process yielded a 7.5 log reduction of S. Enteritidis. Other holding temperatures of 104 and 100°C were also evaluated and these resulted in 5.3 and 4.2 log reductions, respectively (Brandl et al.

14

2008). The authors conclude that this is an effective pasteurization treatment for raw almonds because it is effective against S. Enteritidis and it did not visibly affect the quality of the nuts.

IR heating is a relatively new technology that is currently being investigated for different food processing applications. Although it has several advantages like fast heating rate and uniform drying temperature, it may present other problems such as low penetration power and if exposure is prolonged, it may negatively affect the quality of the nut

(Krishnamurthy et al. 2008).

Blanching and steaming

Blanching of almonds is a thermal process that is mainly used in the industry to remove the skins of the nuts (ABC 2007a). Studies done to evaluate its effectiveness against bacterial contamination (e.g. Salmonella), have determined that a treatment of 2 minutes at 190°C is sufficient to produce a 5 log or greater reduction of Salmonella (ABC

2007a; ABC 2007e).

The process of blanching consists of soaking the almond kernels in hot water or steam-injected water for a determined time (e.g. 2 min) and subsequently, passing nuts through rubber rollers which will remove the skins (ABC 2007a). Then, the kernels may be subjected to shaking and/or sprayed with water in order to remove unwanted material, including remaining skins (ABC 2007a). Finally, the kernels are dried, cooled and sorted

(ABC 2007a).

15

Steam is another thermal treatment used by the industry to decontaminate nuts. This is a surface treatment that is mainly used for organic almonds (ABC 2009). One advantage of using steam as opposed to blanching is that steam can move into the different crevices present in nuts (Lee et al. 2006) and as a result it will have the ability to reach bacterial cells that might be embedded in those areas. A study conducted by Lee and colleagues (2006) evaluated the effectiveness of steam in the reduction of S.

Enteritidis on the surface of two cultivars of raw almonds (Nonpareil and Mission). The treatment consisted of subjecting the inoculated almonds to steam at 93°C for 5, 15, 25,

35, 45, 55 and 65 seconds and the recovery of surviving cells by plating (Lee et al. 2006).

Results of this study show that steam produced a 5.7 and 5.8 log reduction of S.

Enteritidis after 65s of treatment for the Nonpareil cultivar and a 4.0 and 4.1 log reduction for the Mission cultivar (Lee et al. 2006). The authors conclude that steam is an effective treatment to inactivate Salmonella on almonds and that cultivar differences should be taken into consideration when using steam as a decontamination treatment (Lee et al. 2006).

Both blanching and steam have been shown to be effective treatments for the control of Salmonella in almonds, but these treatments may be energy intensive and require additional processing in order to remove the additional added moisture before storage.

Propylene oxide (PPO)

16

Propylene oxide is registered as a fumigant in the US that is commonly used in the

nut industry for the control of insects and microorganisms (ABC 2008). In 2004, the FDA

approved the use of PPO for pasteurization of raw almond kernels in bulk and currently it

is used for treatment of in-shell almonds to reduce Salmonella populations; treatment

should achieve at least a 4 log reduction (ABC 2008). The tolerance residue established

by the Environmental Protection Agency (EPA) when PPO is used as a postharvest

fumigant of tree nuts is 300ppm (EPA 2010). The ABC has established guidelines (2008)

as well as Standard Operating Procedures (SOP) (ABC) for the use of PPO as a

pasteurization method. The parameters established by the ABC can be seen in Table 1.1.

Table 1. 1. PPO pasteurization operating parameters. Obtained from Almond Pasteurization Using Propylene Oxide (PPO) Standard Operating Procedure (SOP), Almond Board of California. Parameters Operational Level Initial product temperature Not less than 86°F (30°C) Chamber temperature 120-125°F (49-51°C) Chamber vacuum before PPO injection At least 27inHg PPO vaporizer temperature 140-145°F (60-63°C) PPO concentration Not less than 0.5oz PPO/ft3 Chamber vacuum after injection of inert 5-6 inHg gas Duration of pasteurization 4 hours Aeration cycles Not <4 and not >14 Post Ventilation 100-110°F (38-43°C) for 2 days or above 59°F (15°C) for 5 days

Danyluk and colleagues (2005) investigated the efficacy of using a commercial PPO

treatment in reducing S. Enteritidis Phage Type 30 on inoculated almonds and the use of

17

Bacillus stearothermophilus as a surrogate for future PPO validation studies. Their results show that PPO was effective in reducing populations of S. Enteritidis PT 30 on almonds and that B. stearothermophilus is not a suitable surrogate for S. Enteritidis because of the low correlation of results between the two (Figure 1.2) (Danyluk, Uesugi and Harris

2005). A 5 log or greater reduction was achieved when initial counts were performed and

5 days after the treatment; only slight reductions were recorded after storage (1.2 to 4.4 log CFU/g) (Danyluk, Uesugi and Harris 2005). In this experiment, the researchers used the minimum concentration of PPO that is used in industry (0.5 kg/m3), the maximum time (4 hours) exposure as recommended by the EPA and the almonds were treated in bulk since this facilitates diffusion of the gas.

Three disadvantages regarding the use of PPO are worth mentioning. Although residue levels have been established in the U.S., some countries may not have regulations for PPO, limiting the distribution of nuts (such as almonds) that are produced here and undergo this type of decontamination treatment. Additionally, PPO cannot be used on certified organic nuts and it has been reported to be a probable human carcinogen based on animal studies (ABC 2009; U.S. Dept. of Health and Human Services 2011).

18

Figure 1.2. Comparison of the reduction of Geobacillus stearothermophilus spores and Salmonella Enteritidis PT 30 in almonds during and after commercial propylene oxide (PPO) treatment. Obtained from (Danyluk, Uesugi and Harris 2005).

1.6 OZONE AND ITS USE IN FOOD

General characteristics

19

Ozone is a triatomic oxygen molecule (O3) that is colorless in the gaseous state, but dark blue in the liquid state (Horvath, Bilittzky, and Huttner 1985). In the O3 molecule, the oxygen atoms are arranged in an isosceles triangle with an obtuse angle (Durrant and

Durrant 1970). The molecular structure of ozone contributes to its oxidizing capabilities.

It has a reduction potential of 2.07, making it one of the strongest oxidizers (Horvath,

Bilittzky, and Huttner 1985). The melting and boiling points are approximately 80 K and

161 K, respectively (Horvath, Bilittzky, and Huttner 1985). Stability of ozone depends on several factors like temperature, pH and presence of contaminants (Kim, Yousef, and

Khadre. 2003). Increasing pH and the temperature decreases ozone stability (Kim 1998).

Ozone has a half life of 12 hours in the gaseous state at room temperature but only 20 to

30 minutes in water, depending on water purity and pH (Kim, Yousef, and Khadre.

2003). Ozone has been described to have a very strong and pungent odor that can be detected by humans at a minimum concentration of 0.02 ppm (Horvath, Bilittzky, and

Huttner 1985).

In nature, ozone is mostly formed in the stratosphere when short ultraviolet (UV) radiation (< 240nm) passes through molecular oxygen (Kim, Yousef, and Khadre. 2003).

This ozone layer serves as a protective layer to Earth (Horvath, Bilittzky, and Huttner

1985) and some of the ozone produced here is transferred to the troposphere (Kim,

Yousef, and Khadre. 2003). In the troposphere, ozone can be produced in different ways by human activity when solar radiation reacts with nitrogen oxides and hydrocarbon pollutants (Horvath, Bilittzky, and Huttner 1985). When generated in excess, ozone can

20 become an air pollutant. Nitrogen oxides and hydrocarbons derive mainly from the exhaust gases of motor vehicles.

Ozone production and decomposition

Ozone is found commonly in nature, but can also be produced synthetically in an industry or lab setting. Since ozone is very unstable and can decompose easily, it is commonly produced onsite and as needed; it cannot be collected or stored for extended periods of time. Formation of ozone can occur by: application of electric discharge to oxygen (i.e. corona discharge), exposure of oxygen to ultraviolet light (i.e. photochemical method), chemical reactions in which oxygen is involved, and by electrolysis of water

(Durrant and Durrant 1970; Horvath, Bilittzky, and Huttner; Kim, Yousef, and Dave

1999). The most common method used to generate high concentrations of ozone is the corona discharge method (Figure 1.3). Here, a high voltage alternating current is applied transversely through a discharge gap in the presence of oxygen (mainly when generating high concentrations) or air (Kim, Yousef, and Dave 1999). Electrons in the oxygen molecules will get excited and eventually this causes separation of the molecules into atoms. Sequentially, the oxygen atoms will combine with oxygen molecules and form the ozone molecule (O3).

21

Electrode

Voltage O3 O2 Discharge gap

Electrode

Figure 1. 3. Corona discharge representation. Adapted from Kim, Yousef, and Dave 1999.

Because of its high reactivity, ozone readily reacts with any organic compounds it encounters, contributing to the rapid decomposition of the molecule. Decomposition of ozone produces several oxidative radicals, including hydroperoxyl radical (·HO2), superoxide anion radical (·O2-) and the hydroxyl radical (·OH) (Kim, Yousef, and Dave

1999). The production of these radicals contributes to the reactivity of ozone, in particular the hydroxyl radical, and the subsequent chain reactions perpetuated by these radicals may be responsible for ozone’s toxicity and antimicrobial activity (Grimes et al.

1983).

22

Ozone measurement

The rapid decomposition of ozone and its high reactivity poses difficulty in determining its concentration. A number of methods have been employed to determine ozone concentration in the gaseous and aqueous states. These include chemical (using redox reactions), electrochemical, thermal, and optical methods (Horvath, Bilittzky, and

Huttner 1985).

A common chemical method used for ozone determination in the aqueous phase is the indigo method. This method consists of using indigo trisulfonate, which is blue in color, and spectrophotometric techniques to determine absorbance (Bader and Hoigné

1981). The indigo reagent has only one carbon-carbon double bond which is disrupted upon contact with ozone (Bader and Hoigné 1981; Kim, Yousef, and Dave 1999). As a result, the reagent is decolorized and this can be detected by spectrophotometry.

Ozone in the gas phase can absorb infrared, visible and ultraviolet radiation, with distinctly strong absorption in the ultraviolet (UV) region (maximum absorption at 260 nm) (Horvath, Bilittzky, and Huttner 1985). For this reason, UV light is one of the most common methods to accurately measure gaseous ozone. UV monitors for ozone quantification consist of a source of UV radiation, e.g. mercury (Hg) lamps, which produce radiation at about 254 nm (Dunlea et al. 2006). The absorbance measured at this wavelength determines ozone concentration.

23

Antimicrobial activity

The effect of ozone in microorganisms is well documented and broadly distributed.

Studies have found the use of ozone to be effective against Gram positive and Gram negative bacteria (Thanomsub et al. 2002), fungi (Barbosa-Martínez et al. 2002), bacterial spores (Novak and Yuan 2004), and viruses (Shin and Sobsey 2003). Microbial sensitivity to ozone varies depending on the physiological state of the cells being treated; stationary phase cells are more resistant than exponential phase cells and bacterial spores are more resistant than vegetative cells (Kim, Yousef, and Khadre. 2003). Additionally, the environment where the microorganisms are present also influences ozone sensitivity, i.e. cells in a dry environment have been showed to be very resistant to gaseous ozone

(Kim, Yousef, and Khadre. 2003).

Several mechanisms of action have been investigated in order to determine the effect of ozone on bacterial cells. The cell membrane is the primary target: ozone reacts with the unsaturated lipids present within the lipopolysaccharide layer of Gram negative bacteria (Kim, Yousef, and Khadre. 2003). When ozone damages the membrane, it causes leakage of cellular contents and subsequent cell lysis (Thanomsub et al. 2002).

Once ozone enters the cell, it can cause damage to its genetic material by targeting the

DNA, causing separation of the strands (Ishizaki et al. 1987).

24

Applications of ozone in food

The effectiveness of ozone in targeting a broad range of microorganisms has been reported in various foods including fresh produce and animal products (Perry and Yousef

2011). It has been shown to be useful in extending the shelf life and maintaining the quality of fresh fruits and vegetables (Liangji 1999). Initially, ozone was mainly used to disinfect wastewater and to purify drinking water in European countries (Horvath,

Bilittzky, and Huttner 1985). At a slow pace it was incorporated in the treatment of water in North America as an alternative to chlorine (Horvath, Bilittzky, and Huttner 1985).

Currently, a significant number of studies are being conducted to evaluate different applications of ozone in the food industry. The U.S. Food and Drug Administration

(FDA) has approved the use of ozone in food and on food contact surfaces (CFR 2011).

Dry foods such as nuts and grains are usually not considered to be prone to bacterial contamination due to their low aw. These conditions (low aw) can also provide disadvantages during food processing because any microorganism present will be harder to inactivate. Presence of aflatoxins, bacterial spores and insects are usually the main concern for dry foods. Recently, many cases of foodborne outbreaks have been reported, raising concern about these types of products. As mentioned previously, various technologies have been and are currently under study in order to develop a safer food supply. Ozone is a good candidate for decontamination of dry foods and it has been investigated by several researchers.

25

Oztekin and colleagues (2006) investigated the effect of gaseous ozone on the microflora of dried figs. They found that an ozone concentration of as low as 5 ppm and an exposure time of at least 3 hours was effective in reducing aerobic microorganism counts, coliforms, yeast and mold (Oztekin et al. 2006). Another study investigated the effect of different ozone concentrations and exposure times on the inactivation of

Escherichia coli and Bacillus cereus as well as changes on the physico-chemical characteristics in pistachios (Akbas and Ozdemir 2006). They tested three different gaseous ozone concentrations (0.1, 0.5 and 1.0 ppm) and several times (up to 360 min) and reported that a concentration of 1.0 ppm and an exposure time of 360 min was effective in reducing E. coli and B. cereus populations in kernels and shelled pistachios

(3.5 log and 3.0 log reductions, respectively) (Akbas and Ozdemir 2006). Additionally, a concentration lower than 1.0 ppm was found effective in reducing both bacteria in ground pistachios; about 1.5 log reduction (Akbas and Ozdemir 2006). Physico-chemical properties (pH, FFA, peroxide values, color and fatty acid compositions) were not significantly affected, but for ground pistachios treated at 1.0 ppm, the peroxide values were significantly higher (Akbas and Ozdemir 2006). The authors indicate that surface area is an important factor when treating pistachios with gaseous ozone.

Akbas and Ozdemir (2008) inoculated red pepper flakes with Escherichia coli,

Bacillus cereus cells, and B. cereus spores and treated with gaseous ozone. A treatment of 1.0 ppm ozone concentration for 360 min reduced B. cereus and E. coli populations by

1.5 and 2.0 log, respectively (Akbas and Ozdemir 2008). Reduction of B. cereus spores

26

(1.5 log) occurred at ozone concentrations of 7 ppm or higher for 360 min, but flavor and appearance were slightly affected at these concentrations (Akbas and Ozdemir 2008).

Based on the findings of these studies and many other published articles, gaseous ozone may be a good treatment for inclusion in dried food processing practices.

Safety

Since ozone is a very strong oxidizer, it is important to choose the correct material and equipment when handling it. Inhaling high or low concentrations of ozone for extended or repeated periods of time can cause diverse and serious symptoms in humans.

The primary target is the respiratory system and symptoms include dizziness, throat pain, headache, and coughing (Horvath, Bilittzky, and Huttner 1985). If exposed for too long, a person may have lung tissue damage and eventually develop severe pulmonary edema

(Scheel et al. 1959). The Occupational Safety and Health Administration (OSHA) of the

United States recommend an exposure limit of 0.1ppm for extended periods of time (US

Dept. of Labor 2007).

27

CHAPTER 2. EFFECT OF LOW WATER ACTIVITY, PHYSIOLOGICAL STATE, AND TYPE OF HUMECTANT ON THE HEAT RESISTANCE OF SALMONELLA SEROVARS SAINTPAUL 02-109, TENNESSEE 2053H AND ELMSBUETTEL 1236H

Abstract

The purpose of this study is to investigate the effect of low water activity (aw), type of humectant and physiological state on the heat resistance of three Salmonella enterica serovars (Saintpaul 02-109, Tennessee 2053H, and Elmsbuettel 1236H).

Serovars were grown (adapted) or just placed (non-adapted) in the low aw broths prior to the heat treatment at 55°C for a total time of 45 min; samples were removed at 5 min intervals. Water activity of Tryptic Soy Broth (TSB) was lowered by adjusting its composition to contain 20% glycerol (aw 0.94), 4% NaCl (aw 0.97), or 35% sucrose (aw

0.95). Type of humectant and cell adaptation significantly affected the D55°C-value. Cells adapted in 20% glycerol showed the lowest D55°C-value (0.86 min for Saintpaul 02-109 and Tennessee 2053H and 0.98 min for Elmsbuettel 1236H). Non-adapted cells placed in the 20% glycerol broth prior to the heat treatment showed an increase in the D55°C-value when compared to the adapted ones (D55°C-value of 3.0, 3.8, and 3.9 min for Saintpaul

02-109, Tennessee 2053H and Elmsbuettel 1236H, respectively). NaCl and sucrose showed a more protective effect under both the adapted and non-adapted conditions for the three serovars tested as compared to glycerol. More research needs to be done in order to fully understand how the cells behave in the presence of different humectants and their subsequent response to heat.

28

Introduction

Foods may contain different added components that bind water and lower the water activity (aw) of the product. Lowering the aw has been used as a preservation method for preventing growth of foodborne microorganisms thus protecting the food against pathogens and spoilage microbiota (Jay, Loessner and Golden 2005). Lately, this approach has been questioned as disease outbreaks linked to low aw foods are on the rise.

Disease outbreaks linked to the consumption of peanut butter, pistachios and cereals have occurred in the past few years (CDC 2007; CDC 2009b; CDC 2008).

Low moisture foods can be defined as foods with a water activity (aw) of or below

0.85 (FDA 2010). Most microorganisms do not grow at < 0.85 aw and in general, bacteria require higher aw values than fungi for growth (Jay, Loessner and Golden 2005). Foods with low aw were generally assumed to be safe since the prevailing environment would not support growth of microorganisms, especially foodborne pathogens. This assumption is now questionable since strains of some pathogens survive well under dry conditions.

Examples of low moisture foods that harbor dry resistant strains include nuts (peanuts, almonds, pistachios), cereals, and chocolate, among others.

Salmonella enterica is a small Gram-negative rod and facultative anaerobe (Jay,

Loessner and Golden 2005; FDA 2009). Although the main reservoir of salmonellae is the intestinal tract of animals (particularly poultry) and humans, this bacteria can also be found in soil, water, insects, and anything that comes in contact with a contaminated

29 source (Jay, Loessner and Golden 2005; FDA 2009). The minimum aw for growth of

Salmonella spp. has been determined at 0.94, but cells can remain dormant at aw below this value (Jay, Loessner and Golden 2005; Mattick et al. 2000b).

In addition to low aw tolerance, some Salmonella serovars were found to be resistant to heat. The resistance or sensitivity to heat of particular Salmonella serovars may depend on the conditions to which the microorganism has been exposed to before being subjected to a heat treatment as well as the food matrix or medium in which it is present at the time of the treatment (Bell and Kyriakides 2002). Heat resistance is also dependent on the serovar; some serovars (e.g Salmonella Senftenberg 775W) are more resistant to heat than others (Bayne and Garibaldi 1969). It has been suggested that a cross protection effect occurs when some Salmonella strains are successively exposed to various stress conditions (Yousef and Juneja 2003). Goepfert and colleagues (1970) evaluated the effect of lowering aw by different humectants on heat resistance of eight Salmonella strains.

According to these researchers, sucrose provided more protection to cells than the other solutes tested (fructose, glycerol, and sorbitol) and the heat resistance is not only dependant on the aw, but also on the type of solute used and pathogen strain. Another research team (Mattick et al., 2000b) investigated whether habituation of Salmonella spp. at low aw prior to a heat treatment increased heat tolerance. Their findings show that habituation at 0.95 aw increased heat tolerance at a treatment temperature of 54°C with the tested solutes. The solutes used to lower the aw included glycerol, NaCl and a mixture of glucose and fructose. It was also noted that the extent of the heat tolerance varies with the type of solute. In addition to habituation at low aw, subjecting Salmonella spp. to 30 sublethal temperatures confers heat tolerance. A study by Bunning et al. (1990) demonstrated that heat-shocking Salmonella enterica serovar Typhimurium at various temperatures (42°C, 48°C and 52°C, vs. 35°C for the control) increases pathogen’s heat tolerance.

This study investigates the effect of low aw, type of humectant (NaCl, sucrose or glycerol) and physiological state (adapted vs. non-adapted) on the heat tolerance of three

Salmonella serovars.

Materials and Methods

Preparation of low aw broths

Tryptic Soy Broth (TSB; Bacto™, Difco Laboratories, Sparks, MD) was used to grow and heat-treat Salmonella serovars. The medium was prepared following manufacturer’s instructions. Three different humectants were then added at different concentrations to lower the aw of the media (TSB, aw of 0.99); the modified media contained 35% sucrose (aw 0.95), 4% NaCl (aw 0.97) and 20% glycerol (aw 0.94). Level of humectants was chosen after preliminary experimentation and represents the highest concentration allowing growth of bacterial cultures.

Culture preparation

Salmonella enterica serovars tested in this study were Saintpaul 02-109, Tennessee

2053H and Elmsbuettel 1236H. These are outbreak strains provided by the Food and

31

Drug Administration (FDA) and were originally isolated from cantaloupe rind, thyme, and peanut butter, respectively. Stock cultures of all three serovars were grown in 9 ml of

TSB and incubated at 37°C for 24 hours prior to use in further experiments (stock culture).

Adaptation to reduced aw

To determine the effect of adaptation on heat tolerance, 10 µl of stock cultures of each serovar were placed in 9 ml of the different low aw broths and incubated at 37°C for

24hrs. After incubation, subsequent transfers (up to three, a total of three days, four 24hr cycles including transfer of stock culture) to the respective fresh low aw broths were performed, followed by incubation. Heat treatment was carried out on the third day. Cells subjected to these transfers will be referred to as ―adapted.‖ Growth of serovars in the reduced aw media is displayed in Fig. 2.2.

Sterile, thin-walled 0.2 ml PCR tubes (Midwest Scientific, St. Louis, MO) were labeled as: 0, 5, 10, 15, 20, 25, 30, 35, 40 and 45 min. A total volume of 100µl of broth was added to each tube. The broth added depended on the experiment being performed.

For example, low aw broths were used when treating adapted and non-adapted cultures, but TSB was used for controls. Following broth addition, 20µl of the corresponding cultures were added to their respective tubes.

32

Figure 2.1. Representation of procedure used for transfers of cultures (a total of three, three days, four 24hr cycles including transfer of stock culture) and preparation of samples before heat treatment.

Effect of humectants

Overnight cultures (20 µl) of each serovar (grown in TSB and incubated for 24hrs at

37°C) were added to 100 µl of low aw broths prior to heat treatment; no adaptation was performed. Labeling of PCR tubes was the same as shown previously.

Heat treatment

A water bath (Precision 2864 Circulating Water Bath, Thermo Fisher Scientific Inc.,

Marietta, OH) was set to a temperature of 55°C, tubes were submerged in water, and then removed from the heat at 5 min intervals over the course of 45 min. Once out of the heat, tubes were immediately placed in an ice-water bath until analysis. To maintain aseptic technique, tubes were dipped in 70% ethanol prior to determining populations.

Determination of survivors was done by dilution and plating of cell suspensions on

33

Tryptic Soy Agar (TSA; Bacto™, Difco Laboratories, Sparks, MD). Plates were incubated for 24hrs at 37°C.

D value calculation

D value was calculated as (-1/slope) and can be expressed by the following formula:

where N0 is the initial population count at time 0, N is the population count after heating at a steady temperature (T) for a specific period of time (t), and DT is the decimal reduction time at the specific T (Yousef et al. 2011).

Statistical analysis

Statistical analysis was performed using PROC GLM from SAS v. 9.2 (SAS

Institute, Inc. 2009). The following model was used with the variables serovar, physiological state, and humectants classified as class variables:

D value= µ + Serovar + Physiological state + Humectant + (Serovar*Physiological state) + (Serovar*Humectant) + (Humectant*Physiological state) +

(Serovar*Humectant*Physiological state) + error. Significant differences were determined by using difference of least square means comparisons. Probability value of

<0.05 was considered significant.

34

Results

The three humectants used in this study (glycerol, NaCl, and sucrose) added in different percentages lowered the water activity of the media to different extents. Heat resistance was influenced by the type of humectants and the physiological state of bacterial cells at the time of the treatment (TSB, adapted, and non-adapted). Growth of serovars in the reduced aw media is displayed in Fig. 2.2. The D values at 55°C of the three serovars (Table 2.1) show significant differences depending on the humectant used, physiological state of cells, and their response to the constant heat over a period of time

(0-45 min) (P < 0.05). When the serovars were adapted in 20% glycerol, a decrease in the

D55°C-value can be observed when compared to the controls (TSB), but this difference was not significant (Tables 2.1 and 2.2). However, when the strains were placed in the

20% glycerol prior to the heat treatment (non-adapted), an increase in the heat resistance is evident for all serovars.

Serovars showed increased heat resistance in both the adapted and non-adapted states when placed in the NaCl and sucrose broths as compared to the controls (TSB). Serovars adapted in sucrose displayed a greater heat resistance than the non-adapted ones. NaCl showed clear differences among the serovars; Saintpaul 02-109 was more resistant when adapted, Tennessee 2053H was more resistant when non-adapted, and Elmsbuettel 1236H presented a similar resistance under the two physiological states (Table 2.1). However, serovar alone did not significantly affect the D55°C-value (P>0.05). The effect of

35 humectants and physiological state on the D55°C-value value was significant (P<0.05)

(Table 2.2). Thermal death curves for all treatments are displayed in Figures 2.3-2.11.

36

A

BB

37

C

Figure 2.2. Growth of Salmonella enterica serovars Saintpaul 02-109, Tennessee 2053H, and Elmsbuettel 1236H in the low aw broths containing 20% glycerol (A), 4% NaCl (B), and 35% sucrose (C) during three transfers (4 growth cycles of 24hrs each).

B

38

Table 2.1. D55°C- value of Salmonella serovars Saintpaul 02-109, Tennessee 2053H, and Elmsbuettel 1236H after heat treatment at 55°C in low aw broths. P value of < 0.05 was considered significant.

a D55°C- value 20% 4% 35% glycerolb NaClb sucroseb Non- Non- Non- Serovar TSB Adapted adapted TSB Adapted adapted TSB Adapted adapted Saintpaul 1.0 ± 0.86 ± 3.0 ± 1.1± 5.5 ± 2.6 ± 1.0 ± 8.0 ± 3.9 ± 02-109 0.04 0.07 0.39 0.10 1.15A 0.21 0.23 4.26 1.14 Tennessee 1.1 ± 0.86 ± 3.9 ± 1.2 ± 2.3 ± 4.0 ± 1.6 ± 6.5 ± 4.0 ± 2053H 0.13 0.22 1.92 0.15 0.45 0.76 0.31 2.37 2.1 Elmsbuettel 1.4 ± 0.98 ± 3.8 ± 1.4 ± 4.1 ± 4.0 ± 1.2 ± 7.6 ± 5.4 ± 1236H 0.22 0.57 0.90 0.25 0.46 0.06 0.15 2.50 0.98

aHeat treatment at 55°C for 0-45 min. b Average D55°C ± standard deviation.

Table 3.2. Average of D55°C- value for Salmonella serovars Saintpaul 02-109, Tennessee 2053H, and Elmsbuettel 1236H after heat treatment at 55°C in low aw broths. P value of < 0.05 was considered significant.

Physiological 20% 4% 35% state Glycerol NaCl Sucrose

TSB 1.2±0.21A 1.2±0.15A 1.3±0.31A A BD C Adapted 0.9±0.07 4.0±1.60 7.4±0.78 Non-Adapted 3.6±0.49 DF 3.5±0.81EF 4.4±0.84DF a Average D55°C ± standard deviation. Superscripts represent significant differences among D55°C values (P<0.05).

39

Figure 2.3. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Saintpaul 02-109 adapted and non-adapted in 20% glycerol. Error bars represent standard deviations (n=9).

40

Figure 2.4. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Saintpaul 02-109 adapted and non-adapted in 4% NaCl. Error bars represent standard deviations (n=9).

Figure 2.5. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Saintpaul 02-109 adapted and non-adapted in 35% sucrose. Error bars represent standard deviations (n=9).

41

Figure 2.6. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Tennessee 2053H adapted and non-adapted in 20% glycerol. Error bars represent standard deviations (n=9).

Figure 2.7. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Tennessee 2053H adapted and non-adapted in 4% NaCl. Error bars represent standard deviations (n=9).

42

Figure 2.8. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Tennessee 2053H adapted and non-adapted in 35% sucrose. Error bars represent standard deviations (n=9).

Figure 2.9. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Elmsbuettel 1236H adapted and non-adapted in 20% glycerol. Error bars represent standard deviations (n=9). 43

Figure 2.10. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Elmsbuettel 1236H adapted and non-adapted in 4% NaCl. Error bars represent standard deviations (n=9).

Figure 2.11. Log CFU/ml after heat treatment at 55°C of Salmonella serovar Elmsbuettel 1236H adapted and non-adapted in 35% sucrose. Error bars represent standard deviations (n=9).

44

Discussion

The cross protective effect that occurs when a microorganism is sequentially exposed to various stress conditions has been studied extensively (Yousef and Juneja 2003). The results of this study suggest that type of humectant (solutes) and physiological state influence the heat resistance of the Salmonella serovars tested. A study by Corry (1973) reported an increase in the heat resistance at 65°C of three Salmonella serovars

(Senftenberg 775W, Typhimurium 7M 4987 and Typhimurium 39H) as the concentrations of sugar and polyols increased. The researcher found that there is no linear relationship between aw and heat resistance (Corry 1973), which is consistent with our results where the lowest D55°C-value was observed when the serovars were adapted in

20% glycerol (aw of 0.94, lowest aw tested). The type of humectants used to lower the aw of a food or media can significantly affect the heat resistance of a microorganism, e.g., an ionic humectant may reduce the heat resistance when added in low concentrations as opposed to nonionic humectants which may have different effects depending on their molecular weight (Tapia, Alzamora and Chirife 2007). In the current study, the ionic humectant added (NaCl) provided a protective effect at two physiological states (adapted and non-adapted) (Figs. 2.4, 2.7 and 2.10). The nonionic humectants (sucrose and glycerol) differed in the protection they provided to the serovars (Figs. 2.3, 2.5, 2.6, 2.8,

2.9, 2.11). Glycerol decreased the heat resistance of all three serovars studied when they were adapted in the low aw broth, but the non-adapted cultures showed an increase in the resistance. Cells adapted in the broth with 20% glycerol, seemed to become more sensitive to the heat treatment, thus resulting in the lowest D55°C-value observed. Sucrose

45 showed a more protective effect than glycerol and NaCl for the three serovars under both the adapted and non-adapted conditions, with the one exception being the non-adapted

Tennessee 2053H in NaCl and sucrose where these cultures exhibited the same D55°C- value (4.0). Overall, adaptation in sucrose offered the most heat protection of any condition tested (Table 2.2).

It has been suggested that a microorganism may have a higher heat resistance under low aw conditions because there is a reduced amount of water present and as a result the water molecule vibrations will be comparatively less as the cells are heated (Tapia,

Alzamora and Chirife 2007). When there is a high water content in the medium, water molecule vibrations increase during heating and this may cause protein denaturation; i.e. disulfide and hydrogen bonds break (Tapia, Alzamora and Chirife 2007).

During adaptation of the cells in the low aw broth, they may have been taking up the solutes present and differences in the type of solutes may have resulted in a variability in the D55°C-value after the heat challenge. Microbial cells posses an internal osmotic pressure higher than that of the environment, causing a turgor pressure or a pressure that is ―exerted outward on the cell wall‖ (Gutiérrez, Abee and Booth 1995). Microorganisms lose turgor when they are placed in a low aw environment since the water in the cytoplasm of the cell migrates to the outside and as a consequence cells are unable to replicate (Tapia, Alzamora and Chirife 2007). In order to reestablish the turgor, microorganisms will accumulate the solutes in the cytoplasm and this may provide an adaptive response to the osmotic stress they were subjected to (Gutiérrez, Abee and

46

Booth 1995). A particular class of solutes, called compatible solutes, does not affect the metabolic and reproductive functions of the cell even at high concentrations (Gutiérrez,

Abee and Booth 1995). These include glycerol and sucrose. Glycerol is known to easily permeate the cell and cause little or no plasmolysis. Sucrose and NaCl, on the other hand, do not permeate as easily and can cause a more severe plasmolysis; osmoregulatory responses are different depending on how permeable the solutes are (Gould 1989; Tapia,

Alzamora and Chirife 2007).

Although not investigated in this study, researchers have documented the expression of different genes when microbial cells are placed under stress conditions. This stress response possibly increases cell ability to survive the stress. An example is the gene rpoS which encodes for the stationary phase sigma factor known as RpoS (Mattick et al.

2000a). RpoS regulates the expression of important genes that play a role in the survival of the microorganisms when they are exposed to stress conditions like low aw and high temperature (Tapia, Alzamora and Chirife 2007).

Despite the fact that low aw plays an important role in limiting the growth of microorganisms in food, some microorganisms are able to survive these conditions and depending on the solute present, they may be able to resist subsequent treatments, e.g. heat. As seen in this study, it is evident that type of solute greatly influences the heat resistance, but it is unclear as to what exactly is happening in the cell during exposure to the different solutes. More research needs to be done in order to understand how the cell

47 is behaving in the low aw broth and how it is responding to the heat challenge on the molecular level.

48

CHAPTER 3. DECONTAMINATION OF PISTACHIOS USING HEAT AND GASEOUS OZONE COMBINATION

Abstract

The purpose of this study is to evaluate the effectiveness of heat/ozone combination treatments in reducing Salmonella enterica serovar Enteritidis poulation on pistachios. A heat resistant non-pathogenic isolate, Enterococcus faecium OSY 31284, was tested as a potential Salmonella surrogate for pistachio decontamination processes.

Additionally, influence of inoculation method (i.e., with or without vacuum) and storage time (24 or 72hrs) on the efficacy of the combination treatments was investigated.

Sanitization treatments were comprised of heat (5% brine at 70°C for 10 minutes), ozonation (160 g/m3, 12.5 psig, 30 min holding time), heat followed by ozonation or ozonation followed by heat. Populations of the inoculated bacteria were determined before and after treatments using appropriate plating techniques.

Results showed significant reduction of E. faecium on pistachios when treated with heat-ozone combination treatment (P<0.05), but these reductions were not significantly different than those achieved by heat alone. Various treatments inactivated Salmonella

Enteritidis to a greater extent than that observed with E. faecium. For both bacteria, the application of gaseous ozone alone showed little reduction probably due to the entrapment of microorganisms between the shell and nut meat. Additionally, inoculation method alone and storage time alone did not have a significant effect on the log reductions of either bacteria. However, the interaction of 72hrs storage and inoculation using vacuum significantly influenced the efficacy of the treatments. The data also 49 suggest that E. faecium is not a suitable surrogate for Salmonella when evaluating pistachios decontamination processes, because the survivors of the two bacteria in response to combination treatments differ by more than 2 log.

Introduction

Recent cases of foodborne outbreaks associated with low moisture foods, such as nuts (peanuts, almonds, pistachios), have raised concerns about the safety of these types of foods. Different technologies have been evaluated in order to determine feasibility for utilization against Salmonella on nuts. Gaseous treatment with polypropylene oxide

(PPO) is currently used in the industry to pasteurize almonds (ABC; GMA 2010) and it has been reported that a 5 log reduction or higher of Salmonella Enteritidis Phage Type

30 can be achieved on almonds (Danyluk, Uesugi, and Harris 2005). Although PPO has been approved as an effective decontamination treatment, it presents three disadvantages:

(1) some countries may not have regulations for PPO, limiting the distribution of nuts

(such as almonds) that are produced in the US and undergo this type of treatment; (2)

PPO cannot be used on certified organic nuts (ABC 2009); and (3) it has been reported to be a probable human carcinogen based on animal studies (U.S. Dept. of Health and

Human Services 2011). Alternatively, ozone is an antimicrobial gas that may be suitable for integration into nut processing practices. Ozone (O3) is a triatomic oxygen molecule that occurs naturally in the atmosphere, but can also be generated onsite (Horvath,

Bilittzky, and Huttner 1985). It is a very strong oxidizer that has been approved by the

FDA to be used on food and food contact surfaces (CFR 2011). Its effectiveness in

50 targeting a broad range of bacteria has been reported in various foods including produce and animal products (Perry and Yousef 2011).

When evaluating a new technology to decontaminate foods, it is important to carry out validation studies using a surrogate of equivalent or greater resistance to that of the target microorganism (GMA 2010). Enterococcus faecium is a Gram positive bacterium that has been used as a surrogate for Salmonella during thermal treatment of almonds

(Jeong, Marks and Ryser 2011). The current study evaluates: (i) the effect of the combination treatments (heat/ozone) on reducing Salmonella Enteritidis populations on inoculated pistachios; (ii) the effect of inoculation method and storage time on the effectiveness of the combination treatments; and (iii) the use of a heat resistant pistachio isolate (E. faecium OSY 31284) when determining the effectiveness of non-thermal treatments on inoculated pistachios.

Materials and Methods

Culture preparation

Cultures of a heat resistant pistachio isolate, Enterococcus faecium OSY 31284, and

Salmonella enterica serovar Enteritidis were prepared by inoculating 555 μl of stock culture into 500ml of suitable broth media. These bacteria were grown in de Man, Rogosa and Sharpe (MRS) broth (Criterion, Hardy Diagnostics, Santa Maria, CA) and Tryptic

Soy Broth (TSB) (Bacto™, Difco Laboratories, Sparks, MD), respectively. Inoculated

MRS broth was incubated at 30°C and inoculated TSB at 37°C, both for 24 hours.

51

Enterococcus faecium OSY 31284 is a pistachio isolate, whereas Salmonella Enteritidis is an egg isolate obtained from the Ohio Department of Agriculture, Reynoldsburg, OH.

Inoculation of pistachios

Unroasted, hulled and unsalted pistachios (shell on) were obtained from Paramount

Farms (San Joaquin Valley, CA) and from Horizon Growers Cooperative (Tulare, CA).

The aw of the pistachios ranged from 0.7 to 0.3 depending on storage time and conditions.

The variables for inoculation consisted of: two inoculation methods (vacuum and no vacuum), two storage times (24hrs and 72hrs) for inoculated samples, and two bacterial isolates (Salmonella Enteritidis and E. faecium OSY 31284). All possible combinations were performed.

Inoculation using vacuum consisted of submerging 400g of pistachios in 500ml of overnight culture (either E. faecium OSY 31284 or S. Enteritidis) in a 2,800ml flask. A cotton stopper was placed before positioning flask inside a custom treatment vessel for inoculation. Once in vessel, 5 inHg vacuum was applied and held for 15 minutes. After release of vacuum, excess liquid was removed and pistachios were placed in sterile metal baskets. Pistachios were dried for 24 or 72 hours at room temperature, depending on the combination needed. Similarly, pistachios (400g) inoculated without vacuum were also placed in a 2,800ml flask, submerged in the overnight culture (500ml), and shaken by hand for 15 min. Excess liquid was removed and pistachios were placed in sterile metal baskets to dry at room temperature.

52

Decontamination treatments

Treatments consisted of heat (5% brine at 70°C for 10 minutes), ozonation (10 in Hg vacuum, 160 g/m3 ozone, 12.5 psig, 30 minutes holding time), heat followed by ozonation (heat + ozone) or ozonation followed by heat (ozone + heat). During the brining step, pistachios were completely submerged in the solution and constant agitation was achieved by use of magnetic stir bars. Brine was comprised of table salt and distilled water. For combination treatments, steps were applied sequentially with as little delay as possible. Gaseous ozone was produced using pure oxygen and an ozone generator that uses corona discharge (Ozonia Triogen Ltd., Zurich, Switzerland). The gaseous ozone was then pumped into a custom treatment vessel and held for 30 min. After the holding period, an additional 30 min was used to exhaust and decompose excess ozone to oxygen by the use of a thermal destruct unit (Ozonia, Elmwood Park, NJ). Ozone concentration was monitored throughout the experiment with the use of an ultraviolet ozone monitor

(Mini-Hicon, IN USA, Inc., Norwood, MA). Figure 3.1 illustrates how ozone concentration (g/m3) and pressure (psig) changed throughout the process. Populations of the inoculated bacteria were determined before and after treatments using appropriate plating techniques.

53

Figure 3.1. Ozone concentration (g/m3) and pressure (psig) as it changes by time. Total treatment time is 30 min, in addition to 30 min exhaust; total time is 60 min.

54

Sample analysis

Treated pistachios (25g or 50g samples) were placed in a blender (Waring

Commercial Blender 51BL32, Torrington, CT) and cold 0.1% peptone water was added.

Samples were blended for 25 seconds and aliquot of the slurry was removed in order to carry out serial dilutions. Appropriate dilutions were consequently plated onto MRS agar

(E. faecium OSY 31284) or TSA (Salmonella Enteritidis) and plates were incubated for

48 hours at 30°C for E. faecium OSY 31284 and 37°C for Salmonella Enteritidis.

Incubated plates were examined and populations of survivors were determined.

Statistical analysis

Statistical analysis was performed using SAS v. 9.2 (SAS Institute, Inc. 2009). A mixed model was used to compare log reductions between treatments. The following model was used, with treatment, inoculation, and storage classified as class variables:

Log reduction= µ + Treatment + Inoculation + Storage + (Treatment*inoculation) +

(Inoculation*storage) + (Treatment*storage) + (Treatment*inoculation*storage) + error.

Significant differences were determined by using difference of least square means comparisons. Probability value of <0.05 was considered significant.

Results

Effect of inoculation method and storage of inoculated pistachios

55

Results show that the inoculation method used and the storage time after inoculation influence the effectiveness of the heat-ozone combination treatments. Pistachios inoculated with either E. faecium OSY 31284 or Salmonella Enteritidis without the use of vacuum and stored for 24hrs showed a higher log reduction for both bacteria when compared with the pistachios inoculated using vacuum and stored for 72hrs (Figs.3.2 and

3.3). The log reductions for Salmonella Enteritidis inoculated pistachios (no vacuum,

24hrs storage) were 6.1, 0.3. 6.6, and 6.6 for heat alone, ozone alone, heat followed by ozone and ozone followed by heat, respectively. The log reductions of the combination treatments (heat-ozone), compared to that of the heat alone, were not significantly different (P>0.05). Pistachios inoculated using vacuum and a 72hrs storage displayed much lower log reductions (4.8 for heat alone, 0.1 for ozone alone, 5.3 for heat followed by ozone and 5.0 for ozone followed by heat) and this interaction affected the efficacy of the treatments the most (P<0.05).

Pistachios inoculated with E. faecium OSY 31284 without the use of vacuum and stored for 24hrs also showed a much higher log reduction (3.8 for heat alone, 0.6 for ozone alone, 4.0 for heat followed by ozone, and 4.6 for ozone followed by heat) than those inoculated using vacuum and stored for 72hrs (2.0 for heat alone, 0.5 for ozone alone, 1.9 for heat followed by ozone, and 1.9 for ozone followed by heat) (Figs. 3.2 and

3.3). In this case also, the interaction of vacuum inoculation and 72hr storage affected the efficacy of the treatments the most (P<0.05).

56

Additional treatments were conducted using pistachios inoculated with Salmonella

Enteritidis only. Pistachios inoculated using vacuum and stored for 24hrs showed the following log reductions: 7.0 log for heat alone, heat followed by ozone and ozone followed by heat, and 0.7 log for ozone alone (Fig. 3.2). On the other hand, pistachios inoculated without the use of vacuum and stored for 72hrs showed log reductions of 5.4 for heat alone, 0.1 for ozone alone, 5.7 for heat followed by ozone, and 6.5 for ozone followed by heat (Fig. 3.2). In these treatments, log reductions caused by heat-ozone combination and that caused by heat alone were not significantly different (P>0.05).

All samples (for either E. faecium OSY 31284 or S. Enteritidis) treated only with ozone showed no significant log reductions (P>0.05). A summary of results can be seen in Tables 3.1 and 3.2.

Table 4.1 Summary of results from pistachios inoculated with E. faecium and treated with heat-ozone combination.

Treatment, Log reduction Inoculation Storage Bacteria method timec Heatd Ozonee Heat + ozonef Ozone + heatg E. faecium OSY 31284 No vacuuma 24hrs 3.8 ± 0.12A 0.6 ± 0.26C 4.0 ± 0.57A 4.6 ± 0.35D Vacuumb 72hrs 2.0 ± 0.19 B 0.5 ± 0.16C 1.9 ± 0.42B 1.9 ± 0.33B aInoculation by submerging pistachios in broth and shaking for 15min. b Inoculation by submerging pistachios in broth and applying vacuum (5 inHg) for 15min. cAt room temperature. d70°C, for 10min, 5% brine. e160 g/m3, 12.5 psig, 30 min holding time. fHeat followed immediately by ozone. gOzone followed immediately by heat. h Average Log reductions ± standard deviation. Superscripts represent significant differences among Log reductions (P<0.05).

57

Table 5.2 Summary of results from pistachios inoculated with S. Enteritidis and treated with heat-ozone combination.

Treatment, Log reduction Inoculation Storage Bacteria method timec Heatd Ozonee Heat + ozonef Ozone + heatg S. Enteritidis No vacuuma 24hrs 6.1 ± 0.3ABC 0.3 ± 0.12G 6.6 ± 0.18CEF 6.6 ± 0.18CEF No vacuuma 72hrs 5.4 ± 0.32AD 0.1 ± 0.25G 5.7 ± 0.98ACD 6.5 ± 0.69BC Vacuumb 24hrs 7.0 ± 0.10BF 0.7 ± 0.23H 7.0 ± 0.10BE 7.0 ± 0.10BE Vacuumb 72hrs 4.8 ± 0.74D 0.1 ± 0.17G 5.3 ± 0.82AD 5.0 ± 0.96D aInoculation by submerging pistachios in broth and shaking for 15min. b Inoculation by submerging pistachios in broth and applying vacuum (5 inHg) for 15min. cAt room temperature. d70°C, for 10min, 5% brine. e160 g/m3, 12.5 psig, 30 min holding time. fHeat followed immediately by ozone. gOzone followed immediately by heat. h Average Log reductions ± standard deviation. Superscripts represent significant differences among Log reductions (P<0.05).

58

Evaluation of the surrogate

E. faecium was proposed as a Salmonella surrogate for pistachio decontamination processes. When inoculated pistachios were treated with heat or heat-ozone combinations, Salmonella Enteritidis and E. faecium OSY 31284 responded to the treatment quite differently (Fig 3.2 and 3.3). Greater than 2 log difference was observed between the reductions experienced by the two bacteria. Under both conditions evaluated

(no vacuum inoculation, 24hr storage vs. vacuum inoculation, 72hr storage) the same difference can be observed.

Figure 3.2. Log reduction of Salmonella Enteritidis inoculated pistachios

with and without use of vacuum and storage of 24hrs or 72hrs after combination treatments (heat/ozone). Error bars represent standard deviations (n = 48).

59

Figure 3.3. Log reduction of Enterococcus faecium OSY 31284 inoculated pistachios and without use of vacuum and storage of 24hrs or 72hrs after combination treatments (heat/ozone). Error bars represent standard deviations (n = 24).

60

Physical changes of pistachios

No visible defects were observed after treatments (Fig. 3.4). Difference in appearance is attributed to altered moisture content.

c a b

c d

Figure 3. 4. Uninoculated and untreated pistachios (a), inoculated pistachios before treatment (b), after heat-ozone combination treatment (c), and after ozone alone (d).

61

Discussion

As previously mentioned, Salmonella spp. require aw of 0.94 for growth, but studies have shown that some serovars have the ability to remain dormant in dry environments

(aw below 0.94) for extended periods of time (Bell and Kyriakides 2002; Burnett et al.

2000; Kieboom et al. 2006). This is also evident in recent outbreaks involving peanuts, pistachios and other dry products (CDC 2007; CDC 2009a; CDC 2008).

Pistachio processing involves storage of up to two years, commonly in ventilated silos (Paramount Farms 2008), posing for potential bacterial contamination. Additionally, cross contamination can occur during growing and harvest and at any step during processing, transportation and packaging. The two inoculation methods employed in this study, i.e., use or no use of vacuum and varied storage time (24hr or 72hr), evaluated the possible contamination scenarios. By inoculating with the use of vacuum, we ensured an even distribution of Salmonella cells inside and outside of the nut; all the way to the end where the shell is attached to the kernel (data not shown). This may have aided in cell attachment to the nuts and it simulated a possible scenario than can occur during the pistachio processing chain. Inoculation without vacuum did not show an even distribution of cells, but it may simulate pistachio cross contamination during processing and/or transportation. It was hypothesized that differences in inoculation methods would significantly influence the effectiveness of the combination treatments. Although inoculation method was important, the results suggest that the storage time after inoculation had the most influence on the effectiveness of the combination treatments

62

(Table 3.1). The initial population (about 8 log/g) remained similar regardless of the inoculation method used.

Significant differences in the log reductions of the combination treatments can be observed for Salmonella Enteritidis and E. faecium OSY 31284. Pistachios stored for

24hrs showed a higher log reduction, especially for the heat-ozone combination treatments, when compared to the ones stored for 72hrs, but interaction of 72hr storage and vaccum inoculation affected the efficacy of the treatments the most (P<0.05), for both bacterial isolates. A 72hr storage may promote attachment of cells to the nut and internalization and as a result bacteria become hardier and more difficult to inactivate.

Internalization of cells makes it difficult for the ozone to reach them. Additionally, storage time alone and inoculation method alone did not have a significant effect on the log reductions of both bacteria (P>0.05).

Akbas and Ozdemir (2006) inoculated kernels, shelled and ground pistachios with

Escherichia coli and Bacillus cereus and treated the product with ozone. The researchers reported a 3.5 log reduction of E. coli and a 3 log reduction of B. cereus in kernels and shelled pistachios and a 2 log reduction of both bacteria in ground pistachios after treatment with ozone at 1 ppm for 360 min (Akbas and Ozdemir 2006). It is important to note that the ozone concentration they used (1 ppm), storage time after inoculation (1hr), and processing of samples (use of stomacher) are different from what was used in the current study; therefore there is variability in the results. The storage time of 1hr is not

63 enough to completely dry the pistachios and as a result the nut may have had some moisture left, possibly causing cell sensitivity to the treatments.

When comparing the log reductions of Salmonella Enteritidis and E. faecium OSY

31284 after heat and heat-ozone combination treatments, more than a 2 log difference can be observed (Figs. 3.2 and 3.3). A potential surrogate microorganism should be of equal or slightly greater resistance than the target microorganism (GMA 2010). In this particular case, E. faecium OSY 31284 showed a much higher resistance to the treatments than Salmonella Enteritidis, suggesting that it is not a suitable surrogate when evaluating the decontamination process tested in this study. E. faecium OSY 31284 has been found to be a useful surrogate for evaluation of thermal processes in almond studies

(Jeong, Marks, and Ryser 2011). Selecting a surrogate with such a greater resistance than the target microorganism may lead to over processing of the food product and as a result it will decrease its quality.

Physical appearance of the pistachios after treatment did not show visible defects; no color change was observed in the kernel. Shell appearance was altered by combination treatments, but this is expected to be the result of moisture loss, and would not be considered an adverse effect. Although not evaluated in this study, Akbas and Ozdemir

(2006) reported that no significant changes in free fatty acids occurred after ozone treatment of kernels, shelled and ground pistachios when exposed to 0.1 and 1ppm of ozone for 360 min. They also reported that peroxide values were not significantly

64 changed for kernels and shelled pistachios, but were significantly higher for ground pistachios (Akbas and Ozdemir 2006).

More research needs to be done in order to fully establish the effectiveness of ozone to inactivate Salmonella on nuts and the extent at which the quality of the nut is affected.

65

REFERENCES

Akbas, M. Y., and M. Ozdemir. 2008. Effect of gaseous ozone on microbial inactivation and sensory of flaked red peppers. Int. J. Food Sci. Tech. 43:1657-1662.

Akbas, M. Y., and M. Ozdemir. 2006. Effectiveness of ozone for inactivation of Escherichia coli and Bacillus cereus in pistachios. Int. J. Food Sci. Tech. 41:513-519.

Alasalvar, C., and F. Shahidi (ed.). 2009. Tree nuts Composition, Phytochemicals, and Heatth effects. CRC Press, Boca Rato:.

Almond Board of California (ABC). 2009. Almond Pasteurization. [Online.] http://www.almondboard.com/FoodProfessionals/Documents/Pasteurization_Sheet%205. 22.09.pdf. Accessed 4 July 2011.

Almond Board of California (ABC). Almond Pasteurization Using Propylene Oxide (PPO) Standard Operating Procedure (SOP). [Online.] http://www.almondboard.com/Handlers/Documents/Pasteurization-Using-PPO-SOP.pdf. Accessed 27 June 2011.

Almond Board of California (ABC). 2007a. Guidelines for Validation of Blanching Processes. [Online.] http://www.almondboard.com/Handlers/Documents/Blanching- Validation-Guidelines.pdf. Accessed 3 July 2011.

Almond Board of California (ABC). 2007b. Guidelines for Validation of Dry Roasting Processes. [Online.] http://www.almondboard.com/Handlers/Documents/Dry-Roast- Validation-Guidelines.pdf. Accessed 26 June 2011.

Almond Board of California (ABC). 2007c. Guidelines for Process Validation Using Enterococcus faecium NRRL B-2354. [Online.] http://www.almondboard.com/Handlers/Documents/Enterococcus-Validation- Guidelines.pdf. Accessed November 2009.

Almond Board of California (ABC). 2007d. Guidelines for Validation of Oil Roasting Processes. [Online.] http://www.almondboard.com/Handlers/Documents/Oil-Roast- Validation-Guidelines.pdf. Accessed 26 June 2011.

Almond Board of California (ABC). 2008. Guidelines for Validation of Propylene Oxide Treatment for In-Shell Almonds. [Online.] http://www.almondboard.com/Handlers/Documents/Inshell-PPO-Pasteurization- Guidelines.pdf. Accessed 26 June 2011.

66

Almond Board of California (ABC). 2007e. Pasteurization Treatments. [Online.] http://www.almondboard.com/Handlers/Documents/Pasteurization-Treatments.pdf. Accessed 3 July 2011.

Anderson E. S., Ward L. R., DeSax M. J., and DeSa J. D. H. 1977. Bacteriophage- typing designations of Salmonella typhimurium. J. Hyg. 78:297-300.

Bader, H., and J. Hoigné. 1981. Determination of ozone in water by the indigo method. Water Research. 15: 449-456.

Barbosa-Martinez, C., L. Ponce de León-Garcia, J. Sepúlveda-Sánchez, and D. Nieto-Angel. Effects of ozone, iodine and chlorine on spore germination of fungi isolated from mango fruits. Revista Mexicana de Fitopatología. 20:60-65.

Bell, C. and A. Kyriakides. 2002. Salmonella: A practical approach to the organism and its control in foods. Blackwell Science, UK.

Brandl, M.T., Z. Pan, S. Huynh, Y. Zhu, and T.H. Mchugh. 2008. Reduction of Salmonella Enteritidis population sizes on almond kernels with infrared. J. Food Prot. 71:897-902.

Brenner, F.W., R.G. Villar, F.J. Angulo, R. Tauxe, and B. Swaminathan. 2000. Salmonella nomenclature, guest commentary. J. Clinical Microbiol. 38:2465-2467.

Bunning, V.K., R.G. Crawford, J.T. Tierney, and J.T. Peeler, 1990. Thermotolerance of Listeria monocytogenes and Salmonella typhimurium after sublethal heat shock. J. Appl. and Environ. Microbiol. 56:3216-3219.

Burnett, S. L., E. R. Gehm, W. R. Weissinger, and L. R. Beuchat. 2000. A note: Survival of Salmonella in peanut butter and peanut butter spread. J. Appl. Microbiol. 89:472-477.

Centers for Disease Control and Prevention (CDC). 2008. Investigation of Outbreak of Infections Caused by Salmonella Agona. Department of Health and Human Services. [Online.] http://www.cdc.gov/salmonella/agona/. Accessed 15 June 2011.

Centers for Disease Control and Prevention (CDC). 2009a. Multistate Outbreak of Salmonella Infections Associated with Peanut Butter and Peanut Butter--Containing Products --- United States, 2008—2009. Department of Health and Human Services. [Online.] http://www.cdc.gov/mmwr/preview/mmwrhtml/mm58e0129a1.htm. Accessed 15 June 2011.

67

Centers for Disease Control and Prevention (CDC) (). 2009b. Salmonella in Pistachio Nuts, 2009. Department of Health and Human Services. [Online.] http://www.cdc.gov/salmonella/pistachios/update.html. Accessed 15 June 2011.

Centers for Disease Control and Prevention(CDC) . 2007. Multistate Outbreak of Salmonella Serovar Tennessee Infections Associated with Peanut Butter --- United States, 2006—2007. Morbidity and Mortality Weekly Report. [Online.] http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5621a1.htm. Accessed 15 June 2011.

Code of Federal Regulations (CFR). 2011. Direct food substances affirmed as generally recognize as safe: Title 21, 184. [Online.] http://ecfr.gpoaccess.gov/cgi/t/text/text- idx?c=ecfr&sid=786bafc6f6343634fbf79fcdca7061e1&rgn=div5&view=text&node=21:3 .0.1.1.14&idno=21#21:3.0.1.1.14.2.1.142. Accessed 16 May 2011.

Chung, K.C., and J.M. Goepfert. 1970. Growth of Salmonella at low pH. J. Food Sci. 35:326-328.

Danyluk, M.D., A.R. Uesugi, and L.J. Harris. 2005. Survival of Salmonella Enteritidis PT 30 on Inoculated Almonds after Commercial Fumigation with Propylene Oxide. J. Food Sci. 68:1613-1622.

Du, W.X., S.J. ABD, K.L. McCarthy, and L.J. Harris. 2010. Reduction of Salmonella on inoculated almonds exposed to hot oil. J. Food Prot.: 1238-1246.

Dunlea, E.J., et. al. 2006. Technical note: Evaluation of standard UV absorption ozone monitors in a polluted urban environment. Atmos. Chem. Phys. 6:3163-3180.

Durrant, P.J., and B. Durrant. 1970. Chapter 22: Sub-group VI N: Oxygen, sulphur, selenium, and tellurium, p. 796-880. In: Introduction to inorganic chemistry. John Wiley & Sons, Inc., New York.

Environmental Protection Agengy (EPA). 2010. Code of Federal Regulations: Title 49, Chapter 1, Section 180.491. [Online.] http://frwebgate1.access.gpo.gov/cgi- bin/PDFgate.cgi?WAISdocID=ZplLdl/7/2/0&WAISaction=retrieve. Accessed 30 June 2010.

Food and Drug Administration (FDA) . 2009. Bad Bug Book: Foodborne Pathogenic Microorganisms and Natural Toxins Handbook Salmonella spp. Department of Health and Human Services. [Online.] http://www.fda.gov/food/foodsafety/foodborneillness/foodborneillnessfoodbornepathoge nsnaturaltoxins/badbugbook/ucm069966.htm. Accessed 16 June 2011.

68

Food and Drug Administration (FDA) . 2011a. Guidance for Industry: Measures to address the risk for contamination by Salmonella species in food containing a peanut- derived product as an ingredient. [Online.] http://www.fda.gov/Food/GuidanceComplianceRegulatoryInformation/GuidanceDocume nts/ProduceandPlanProducts/ucm115386.htm. Accessed 26 June 2011.

Food and Drug Administration (FDA) . 2011b. Guidance for Industry: Measures to address the risk for contamination by Salmonella species in food containing a pistachio- derived product As an ingredient; Draft Guidance. [Online.] http://www.fda.gov/Food/GuidanceComplianceRegulatoryInformation/GuidanceDocume nts/ProduceandPlanProducts/ucm169160.htm. Accessed 26 June 2011.

Food and Drug Administration (FDA) . 2010. Water Activity (aw) in Foods. Department of Health and Human Services. [Online.] http://www.fda.gov/ICECI/Inspections/InspectionGuides/InspectionTechnicalGuides/uc m072916.htm. Accessed 15 June 2011.

Federal Register. 2007. Federal Register/Vol. 72, No. 61/Friday, March 30, 2007/Rules and Regulations. [Online.] http://www.almondboard.com/Growers/Documents/Outgoing- Quality-Control-Rule.pdf. Accessed 3 July 2011. p 15021-15036.

Grocery Manufacturers Association (GMA). 2009. Control of Salmonella in Low- Moisture Foods. Grocery Manufacturers Association, Washington, D.C. [Online.] http://graphics8.nytimes.com/packages/pdf/business/20090515_moss_ingredients/Salmon ellaControlGuidance.pdf. Accessed 25 June 2011.

Goepfert, J. M., I. K. Iskander, and C. H. Amudson. 1970. Relation of the heat resistance of salmonellae to the water activity of the environment. Appl. Microbiol. 19:429-433.

Gould, G. W. (ed). 1989. Chapter 5: Drying, raised osmotic pressure and low water activity. In: Mechanisms of action of food preservation procedures, p. 97-117. Elsevier Applied Science, London.

Grimes, H.D., K.K. Perkins, and W.F. Boss. 1983. Ozone degrades into hydroxyl radical under physiological conditions. Plant Physiol. 72:1016-1020.

Grocery Manufacturers Association (GMA). 2010. Industry handbook for safe processing of nuts. The Association of Food, Beverage, and Consumer Products Companies. [Online.] http://www.gmaonline.org/downloads/technical-guidance-and- tools/Industry_Handbook_for_Safe_Processing_of_Nuts_1st_Edition_22Feb10.pdf. Accessed 16 April 2011. 69

Gutierrez, C., T. Abee, and I. R. Booth. 1995. Physiology of the osmotic stress response in Microorganisms. Inter. J. Food Microbiol. 28:233-244.

Harris, L. 2009. Process Validation for Low Water Activity Foods. Presentation at International Association for Food Protection. [Online.] http://www.foodprotection.org/files/rr_presentations/RR_07.pdf. Accessed June 26, 2011.

Horvath, M., L.Bilittzky, and J.Huttner. 1985. Ozone. Elsevier, Budapest.

Ishizaki, K., K. Sawadaishi, K. Miura, and N. Shinriki. 1987. Effect of ozone on plasmid DNA of Escherichia coli in situ. Wat. Res. 21: 823-827.

Jay, J.M., M.J. Loessner, and D. A. Golden. 2005. Modern Food Microbiology. Springer, New York.

Jeong, S., B.P. Marks, and E.T. Ryser. 2011. Quantifying the performance of Pediococcus sp. (NRRL B-2324: Enterococcus faecium) as a nonpathogenic surrogate for Salmonella Enteritidis PT30 during moist-air convection heating of almonds. J. Food Prot. 74:603-609.

Jun, S. and J. Irudayaraj (ed). 2009. Chapter 5: Infrared radiation for food processing. In: Food processing operations modeling: design and analysis, p 113-123. CRC Press, Boca Raton.

Kieboom, J., H. D. Kusumaningrum, M. H. Tempelaars, W. C. Hazeleger, T. Abee, and R. R. Beumer. 2006. Survival, elongation, and elevated tolerance of Salmonella enterica serovar Enteritidis at reduced water activity. J. Food Prot. 69:2681–2686.

Kim, J.G. 1998. Ozone as an antimicrobial agent in minimally processed foods. PhD thesis. The Ohio State University, Columbus.

Kim, J.G., A.E. Yousef, and M.A. Khadre. 2003. Ozone and its current and future application in the food industry. Advances Food Nutr. Res. 45:167-218.

Kim, J.G., A.E. Yousef, and S. Dave. 1999. Application of ozone for enhancing the microbiological safety and quality of foods: A Review. J. Food Prot. 62:1071-1087.

Lee, S.Y., S.W. Oh, H.J. Chung, J.I. Reyes de Corcuera, J.R. Powers, and D.H. Kang. 2006. Reduction of Salmonela enterica serovar Enteritidis on the surface of raw shelled almonds by exposure to steam. J. Food Prot. 69:591-595.

70

Leyer, G.J. and E.A. Johnson. 1992. Acid adaptation promotes survival of Salmonella spp. in cheese. J. Appl. and Environ. Microbiol. 58:2075-2080.

Liangji, X. 1999. Use of ozone to improve the safety of fresh fruits and vegetables. Food Technology. 53:58–61.

Mattick, K.L., F. Jorgensen, J. D. Legan, H. M. Lappin-Scott, and T. J. Humphrey. 2000a. Habituation of Salmonella spp. at reduced water activity and its effect on heat tolerance. J. Appl. and Environ. Microbiol. 66:4921-4925.

Mattick, K. L., F. Jorgensen, J. D. Legan, M. B. Cole, J. Porter, H. M. Lappin-Scott, and T.J. Humphrey. 2000b. Survival and filamentation of Salmonella enterica Serovar Enteritidis PT4 and Salmonella enterica Serovar Typhimurium DT104 at low water activity. J. Appl. and Environ. Microbiol. 66:1274-1279.

Ng, H., H. G. Bayne, and J. A. Garibaldi. 1969. Heat Resistance of Salmonella: the Uniqueness of Salmonella senftenberg 775W. J. Appl. Microbiol. 17:78-82.

Novak, J.S. and J.T.C. Yuan. 2004. Increased inactivation of ozone-treated Clostridium perfringens vegetative cells and spores on fabricated beef surfaces using mild heat. J. Food Prot. 67:342-346.

Oztekin, S., B. Zorlugenc, F. K. Zorlugenc. 2006. Effect of ozone treatment on microflora of dried figs. J. Food Eng. 8: 8-12.

Paramount Farms. 2008. Quality from the ground up. DVD. Los Angeles, CA.

Paramount Farms. 2011. [Online.] http://www.paramountfarms.com/#home. Accessed 5 August 2011.

Perry, J.J., and A.E. Yousef. 2011. Decontamination of raw foods using ozone-based sanitization techniques. Annual Rev. Food Sci. Tech. 2:281-298.

Scheel, L.D., O.J. Dobrogorski, J.T. Mountain, J.L. Svirbely, and Stokinger, H.E. 1959. Physiological, biochemical, immunologic, and pathologic changes following ozone exposure. J. Appl. Physiol. 14:67-80.

Shin, G.A. and M.D. Sobsey. 2003. Reduction of Norwalk virus, poliovirus 1, and bacteriophage MS2 by ozone disinfection of water. Appl. and Environ. Microbiol. 69:3975-3978.

Tapia, M. S., S. M. Alzamora, and J. Chirife. 2007. Chapter 10: Effects of Water Activity (aw) on Microbial Stability: As a Hurdle in Food Preservation, p. 239-255. In: G. 71

V Barbosa-Cánovas, A. J. Fontana, Jr., S. J.Schmidt, T. P. Labuza (ed.), Water Activity in Foods: Fundamentals and Applications. John Wiley & Sons, Iowa.

Thanomsub, B., V. Anupunpisit, S. Chanphetch, T. Watcharachaipong, R. Poonkhum, and C. Srisukonth. 2002. Effects of ozone treatment on cell growth and ultrastructural changes in bacteria. J. Gen. Appl. Mictobiol. 48:193-199.

US Department of Labor, Occupational Safety and Health Administration. 2007. Chemical sampling information: ozone. [Online.] http://www.osha.gov/dts/chemicalsampling/data/CH_259300.html. Accessed 9 July 2011.

Venkatachalam, M. and S. K. Sathe. 2006. Chemical composition of selected edible nut seeds. J. Agric. Food Chem. 54:4705-4714.

White, D. 2007. Chapter 2: Growth and Cell Division, p. 53-82. In: The Physiology and Biochemistry of Prokaryots. Oxford University Press, New York.

Woodroof, J. G. 1967. Tree nuts: Production, processing and products. The Avi Publishing Company, Inc., Conneticut.

World Health Organization (WHO). 2003. Background Document: The diagnosis, treatment and prevention of typhoid fever. Communicable Disease Surveillance and Response Vaccines and Biologicals. [Online.] http://www.who.int.proxy.lib.ohio- state.edu/rpc/TFGuideWHO.pdf. Accessed 25 August 2011.

Yousef, A. E., J. J. Perry, and J. Waite. 2011. Chapter 8: Thermal resistance of microorganisms in food. In: Food Microbiology Laboratory Manual. The Ohio State University, Columbus.

Yousef, A.E., and V.K. Juneja. 2003. Microbial Stress Adaptation and Food Safety. CRC Press, Boca Raton.

72