INACTIVATION OF SALMONELLA AND SURROGATE BACTERIA ON CASHEWS AND MACADAMIA NUTS EXPOSED TO SATURATED

STEAM AND PROPYLENE OXIDE TREATMENTS

Thomas Philip Saunders

Thesis submitted by the faculty of the Virginia Polytechnic Institute and State University in partial fulfillment of the requirements of the degree of

Master of Science in Life Science

In

Food Science and Technology

Monica A. Ponder, Co-Chair

Robert C. Williams, Co-Chair

Haibo Huang

April 26, 2017

Blacksburg, VA

Keywords: Salmonella, Enterococcus faecium, Pediococcus acidilactici, Staphylococcus carnosus, surrogate, cashews, macadamia nuts, inactivation,

processing, propylene oxide (PPO), saturated steam (SS)

Copyright (2017)

INACTIVATION OF SALMONELLA AND SURROGATE BACTERIA ON CASHEWS AND MACADAMIA NUTS EXPOSED TO SATURATED STEAM AND PROPYLENE

OXIDE TREATMENTS

Thomas Philip Saunders

SCIENTIFIC ABSTRACT

Saturated steam (SS) and propylene oxide (PPO) fumigation are two common methods to

improve microbiological quality and safety of tree nuts. Validation of these processes is needed

to ensure adequate control of bacterial pathogens. Since pathogens cannot be studied in food

processing environments, surrogates with resistance comparable to the pathogens needed to be

identified. The objective was to investigate the suitability of Enterococcus faecium, Pediococcus

acidilactici, and Staphylococcus carnosus as surrogate bacteria for Salmonella spp. on whole

cashews and macadamia nuts, processed with SS or PPO.

Whole cashews and macadamia nuts were co-inoculated with a cocktail of Salmonella

enterica and one of the three potential surrogates. Nuts were dried to original aw, packaged in

poly-woven bags (2.3 kg) and commercially processed using vacuum assisted steam at 80 ͦ C or

PPO fumigation. Salmonella and the potential surrogates were enumerated by serial dilution, and

plated onto TSA with overlay of XLT-4 (Salmonella) or media selective for the potential

surrogates.

Mean log reductions (CFU/g) of Salmonella and each potential surrogate were compared

using a paired T-test. SS results: reduction of Salmonella (6.0 ± 0.14) was significantly larger

than E. faecium (4.3± 0.12), or P. acidilactici (3.7± 0.14) on whole cashews. Salmonella (5.9 ±

0.18) was significantly larger than P. acidilactici (4.4± 0.18) on whole macadamia nuts.

PPO results: reduction of Salmonella (7.3 ± 0.19) was significantly greater than E.

faecium (6.4± 0.31), or P. acidilactici (6.3± 0.33) on whole macadamia nuts. Reduction of

Salmonella was significantly greater than E. faecium and P. acidilactici reduction on cashews.

P. acidilactici may be considered a surrogate for Salmonella reduction on whole

macadamia nuts and whole cashews processed using SS at 80 ͦ C. E. faecium and P. acidilactici

may be considered surrogates for Salmonella reduction on whole macadamia nuts and whole

cashews processed using PPO. Reduction of St. carnosus exceeded that of Salmonella indicating

it is not a suitable surrogate for Salmonella using either processing intervention.

INACTIVATION OF SALMONELLA AND SURROGATE BACTERIA ON CASHEWS AND MACADAMIA NUTS EXPOSED TO SATURATED STEAM AND PROPYLENE

OXIDE TREATMENTS

Thomas Philip Saunders

GENERAL AUDIENCE ABSTRACT

Tree nuts are produced and consumed worldwide, playing a role as a snack or ingredient

in foods in several cultures. Traditionally, tree nuts have been believed to be microbiologically

safe due to their composition and lack of water available to harbor pathogenic bacterial growth.

However, recent years have proven to be problematic for many tree nuts and nut products,

numerous recalls have occurred in the United States for verified contamination of Salmonella.

Since Salmonella can be found in food systems worldwide, and is a leading cause of foodborne

illness due to bacterial contamination of food, steps must be taken to improve the safety of nuts

grown locally and imported products. With several processing options for tree nuts, two that are

commonly practiced to reduce microorganisms include a fumigation of product using propylene

oxide (PPO) and a thermal inactivation treatment through use of saturated steam. The

comparison of Salmonella inactivation and non-pathogenic surrogate bacteria on cashews and

macadamia nuts, being processed in these two manners, was investigated. Possible bacteria

strains that were investigated for surrogacy were Enterococcus faecium NRRL B2354 (ATCC

#8459), Pediococcus acidilactici (ATCC #8042), and Staphylococcus carnosus (ATCC #51365).

Surrogates are bacteria that have similar inactivation characteristics to Salmonella when

processed, that can be purposely introduced before processing to ensure inactivation of

Salmonella and is harmless if consumption occurs. Studies continue to ensure safety of tree nuts

as well as complying with pending and future regulations.

v

ACKNOWLEDGEMENTS

I would like to thank my better half, my girlfriend Mary Gettas, for the support and love these past three years and especially through my graduate education. Long distance relationships are never easy, but I would not have wanted to go through graduate school with anyone else by my side. You have helped make the past two years of research all the better, and I am looking forward to the next chapter in our lives together. Countless late night talks and hearing me discuss aspects of my research, that you may not have always understood, provided me with an irreplaceable outlet, that was always responded with in a loving manor. There is no way I can describe in words how much your support has meant to me, but I hope I can show you even a fraction of all the love that you have shown me.

To my co-advisors:

Dr. Ponder, thank you for all the help and relentless hours you have dedicated to me and our lab group. Without your help, I would not only be without research, but would not have the confidence to produce scientific work, that you were able to facilitate through our lab meetings reviewing various scientific articles and fellow lab mates writing, in the pursuit to help develop my own skills. It has been an honor to work with you and I am a better Food Microbiologist because of you and your passion to teach our lab group whenever the opportunity arises.

Dr. Williams, thank you for all that you have done for me in my graduate career. Without your help, I would have never been accepted into graduate school. You were always available to discuss anything that I needed, scientific or off-topic, and for these things I will be ever grateful.

To my family and friends, thank you all so much for your support and encouragement throughout this process. I’m honored to have such an amazing support system. My parents Doug and Pam Saunders, you made yourselves available at any and all points of the day. Whether it was to discuss my day or just to catch up, I could always count on your phone calls to help me focus back in on the parts of life that matter most to me. My brothers, Cary and Jake Saunders, whether you realized it or not, you both were a huge support system during this process. In the fall semesters, it was great knowing that one or the both of you would be around to watch the Virginia Tech Hokies play football, and that created a balance in my life that was needed after a long week of research.

To some of my closest friends and lab mates, Jian Wu, Kendall Fogler, and Kim Waterman, I can’t thank you enough for your help throughout this entire research. Your flexibility and willingness to help with all portions of my research made the planning and carrying out of each trial much easier. When you have great lab mates around you, the stress levels you experience are exponentially lower. There is no way this thank you will do justice for all you have helped me with, but just know I greatly appreciate every hour spent in lab, every conversation we share, and all the laughs we have had through this journey.

I would like to thank the faculty, staff, and students of the Food Science & Technology Department for your support. I have made so many friendships that have shaped my time at Virginia Tech in some form and I am looking forward to where the future takes us all.

vi

DEDICATION

I would like to dedicate this research to my parents, Richard Douglas Saunders and Pamela Sotos Saunders. Without their influence and encouragement to return to graduate school and pursue my passion of Food Science, I would have never made it this far. They instilled in my brothers and I from an early age to not be afraid to attempt a task and fail, but that it was important to always learn from those experiences, so that we may prove ourselves more successful in the next attempt. I have obtained a drive from my parents to work hard and never give up on a goal, you always finish what you started and complete that task to the best of your ability. I cannot thank them enough for all that they have provided for my brothers and I, and only hope that I can make up a fraction of that to someday.

vii

ATTRIBUTIONS

Monica A. Ponder, PhD, (Food Science & Technology Department at Virginia Tech) is currently an Associate professor and served as the principal investigator of this project. Dr. Ponder provided guidance, funding and helped develop the experimental design for this research. She also performed countless edits and assisted with data analysis. She is a co-author on the manuscripts in Chapters 3 and 4. Dr. Ponder served as a Co-chair on this thesis research committee.

Jian Wu, PhD, (Food Science & Technology Department at Virginia Tech) is currently a project associate for the food safety and food microbiology lab. Dr. Wu aided on all experimental aspects of Chapter 3 and Chapter 4, helping a great deal with preliminary work, which lead to the development of methodology performed. Dr. Wu also designed the PVC pipe apparatus which samples were bound to, preventing location variability though out processing. Dr. Wu is a co-author on Chapter 3 and Chapter 4.

Robert C. Williams, PhD, (Food Science & Technology Department at Virginia Tech) is currently an Associate professor of food microbiology and food safety. Dr. Williams was a co-author on Chapters 3 and 4. Dr. Williams served as a Co-chair on this thesis research committee.

Haibo Huang, PhD, (Food Science & Technology Department at Virginia Tech) is currently an Assistant professor of food processing and engineering. Dr. Huang was a co-author on Chapters 3 and 4, he shared critical thought and advice into statistical analysis of research data. Dr. Huang served as a committee member on this thesis research committee.

Kim M. Waterman, MS, (Food Science & Technology Department at Virginia Tech) is currently the food microbiology lab manager, supporting departmental and graduate research. Ms. Waterman provided technical assistance through out every trial of research. Ms. Waterman played a critical role in maintaining the research time line and that lab processes ran smoothly.

viii

TABLE OF CONTENTS

ABSTRACTS

Scientific

General Audience

ACKNOWLEDGEMENTS v

DEDICATION vi

ATTRIBUTIONS vii

TABLE OF CONTENTS viii

CHAPTER 1: INTRODUCTION AND JUSTIFICATION 1

References 5

CHAPTER 2: LITERATURE REVIEW 6

Microbiological Quality and Safety of Tree Nuts 6

Tree Nuts 6

Macadamia nuts 8

Cashews 11

Saturated Steam Treatment for Nuts 13

Propylene Oxide (PPO) Treatment for Nuts 15

Salmonellosis 18

Salmonella Survival at Low Water Activity 19

Salmonella enterica Strain Consideration 20

Inoculation Preparation Effect on Low Water Activity Foods 21

Wet Inoculation 22

Surrogate Bacteria 23

ix

Enterococcus faecium NRRL B-2354 as a surrogate for Salmonella 23

Pediococcus acidilactici ATCC 8042 as a surrogate for Salmonella 25

Staphylococcus carnosus ATCC #51365 as a surrogate for Salmonella 25

References 27

CHAPTER 3: INACTIVATION OF SALMONELLA ENTERICA AND SURROGATE 33 BACTERIA ON CASHEWS AND MACADAMIA NUTS EXPOSED TO PROPYLENE OXIDE TREATMENTS

Abstract 35

Introduction 37

Materials and Methods 39

Results 44

Discussion 50

Conclusion 54

Acknowledgements 55

References 56

Tables / Figures 58

CHAPTER 4: EVALUATING PEDIOCOCCUS ACIDILACTICI AND ENTEROCOCCUS 60 FAECIUM NRRL B-2354 AS SURROGATES FOR SALMONELLA ENTERICA ON SATURATED STEAM PROCESSED CASHEWS AND MACADAMIA NUTS

Abstract 62

Introduction 63

Materials and Methods 65

Results 70

Discussion 76

Conclusion 80

Acknowledgments 81

References 82

Tables / Figures 83

x

CHAPTER 5: FUTURE RESEARCH 86

1

CHAPTER 1: INTRODUCTION

Salmonella has been responsible for many illnesses and recalls associated with low water

activity (low-aw) foods in the United States (7). Salmonella itself does not grow at water

activities below 0.98. Though food products with aw of 0.7 or lower do not support growth of

pathogenic bacteria such as Salmonella, Salmonella has been documented to survive within low

water activity products for up to 2 years (7). While regulators traditionally concerned themselves

with raw meat products, such as chicken, as a source of Salmonella contamination, it has been

reported that under processed and contaminated food items such as nuts and spices have led to

many recalls and outbreaks (11). Salmonella presence has prompted the recall of a number of

tree nuts including: walnuts, macadamia nuts, pecans, cashews, almonds, pistachios, pine nuts

and hazelnuts (10).

In the United States, tree nuts are routinely processed using propylene oxide (PPO), as

well as saturated steam treatments, for the reduction of bacteria, yeasts and molds. The U.S.

Environmental Protection Agency (EPA) registration for PPO as a pesticide fumigant, for select

foods including nuts, stipulates that (i) exposure temperature must not exceed 51.68 ºC, (ii)

exposure time shall be no more than 4 h, and (iii) PPO residue in the product (prior to release)

shall not exceed 300 ppm. Label instructions additionally dictate that (i) dosage cannot exceed

2.5 kg/m3 and (ii) at the end of treatment there should be no less than four aeration cycles with a

PPO chamber volume of air (EPA registration no. 47870-1, Aberco, Seabrook, Md.)(1).

Commercial processing facilities can use a variety of concentrations of PPO, chamber

temperature, and relative humidity of environment to meet the quality expectations of their

customers as long as these are in compliance with the EPA label designations for tree nuts (6).

2

Raw or processed tree nuts are fumigated at a max rate 2.0 oz ai/ft3. This is completed in the

retort for up to 6 hours and followed by the introduction of four chamber volumes of air (EPA

LABEL). Tree nuts are held at 25°C for 28 days or at 35°C before they are shipped. Considering

that, tree nuts can be shipped if residues of PPO are reported below 300 ppm(6). In addition,

several different packaging configurations may affect the PPO penetration into the sample.

Validation studies must be completed to each of the processing parameters on each system to

deem them successful in improving microbial quality of products.

Saturated steam can be used in processing when attempting to reach a 4-log reduction in

Salmonella (4, 12). Steam is being used more to appeal to consumers concerned with the use of

Ethylene Oxide (ETO) and Propylene Oxide (PPO) treatments in food systems (12). Steam

treatments have been preferred in an attempt to protect organoleptic properties of foods, like tree

nuts, which the consumer demands to meet particular standards. With the capability to use small

amounts of steam with high amounts of heat transferred, it is shown to help prevent major loss

of aromatics and other changes detrimental to the nuts, while thermally inactivating pathogenic

bacteria such as Salmonella (9, 12).

The need to validate many product configurations and processing parameters warrants

the identification of a surrogate to avoid the introduction of Salmonella in the processing

facility. The surrogate should be non-pathogenic, and have similar stability and inactivation

characteristics to the target pathogen under the decontamination process of interest.

Enterococcus faecium NRRL B-2354 has been validated as a surrogate for Salmonella spp. in

several thermal lethality studies (2, 8). While E. faecium NRRL B-2354 has been validated as an

3

appropriate surrogate for Salmonella with regard to thermal processing of almonds, juice and

jerky, there are instances where the inactivation characteristics were not comparable to

Salmonella. In the case of dried pet food and ground and formed beef jerky, Pediococcus

acidilactii ATCC #8042, has been demonstrated to be more heat resistant than Salmonella, and

therefore may serve as a conservative surrogate for Salmonella (3, 5).

Objective 1: Compare the survival of Salmonella enterica with Enterococcus faecium NRRL

B-2354, , Pediococcus acidilactici ATCC #8042, and Staphylococcus carnosus ATCC #51365

on whole cashews and macadamia nuts subjected to propylene oxide fumigation. Bacteria with

comparable or greater survival compared to Salmonella may be considered as a surrogate for

pathogen inactivation on cashews and macadamia nuts processed by a commercial processor

using their proprietary process that is in compliance with US EPA label standards.

H0: There will be no significant differences between average log CFU/g reduction in Salmonella and Enterococcus faecium NRRL B-2354 when using propylene oxide fumigation.

HA: There will be significant differences between average log CFU/g reduction in Salmonella and Enterococcus faecium NRRL B-2354 when using propylene oxide fumigation. H0: There will be no significant differences between average log CFU/g reduction in Salmonella and Pediococcus acidilactici ATCC #8042 when using propylene oxide fumigation. HA: There will be significant differences between average log CFU/g reduction in Salmonella and Pediococcus acidilactici ATCC #8042 when using propylene oxide fumigation. H0: There will be no significant differences between average log CFU/g reduction in Salmonella and Staphylococcus carnosus ATCC #51365 when using propylene oxide fumigation.

4

HA: There will be significant differences between average log CFU/g reduction in Salmonella and Staphylococcus carnosus ATCC #51365 when using propylene oxide fumigation.

Objective 2: Compare the survival of Salmonella enterica with Enterococcus faecium NRRL

B-2354, Pediococcus acidilactici ATCC #8042, and Staphylococcus carnosus ATCC #51365 on

whole cashews and macadamia nuts subjected to saturated steam treatment. Bacteria with

comparable or greater survival compared to Salmonella may be considered as a surrogate for

pathogen inactivation on cashews and macadamia nuts processed by a commercial processor

using their proprietary process that is in compliance with US EPA label standards.

H0: There will be significant differences between average log CFU/g reduction in Salmonella and Enterococcus faecium NRRL B-2354 when using propylene oxide fumigation.

HA: There will be significant differences between average log CFU/g reduction in Salmonella and Enterococcus faecium NRRL B-2354 when using saturated steam treatments.

H0: There will be no significant differences between average log CFU/g reduction in Salmonella and Pediococcus acidilactici ATCC #8042 when using saturated steam treatments. HA: There will be significant differences between average log CFU/g reduction in Salmonella and Pediococcus acidilactici ATCC #8042 when using saturated steam treatments. H0: There will be no significant differences between average log CFU/g reduction in Salmonella and Staphylococcus carnosus ATCC #51365 when using saturated steam treatments. HA: There will be significant differences between average log CFU/g reduction in Salmonella and Staphylococcus carnosus ATCC #51365 when using saturated steam treatments.

5

References:

1. Agency, E. P. 2004. Propylene oxide tolerances for residues. Code of Federal Regulations title 49, chapter 1, section 180.493. 2. Bianchini, A., J. Stratton, S. Weier, T. Hartter, B. Plattner, G. Rokey, G. Hertzel, L. Gompa, B. Martinez, and K. M. Eskridge. 2014. Use of Enterococcus faecium as a Surrogate for Salmonella enterica during Extrusion of a Balanced Carbohydrate-Protein Meal. Journal of Food Protection. 77:75-75. 3. Borowski, A. G., S. C. Ingham, and B. H. Ingham. 2009. Validation of Ground-and-Formed Beef Jerky Processes Using Commercial Lactic Acid Bacteria Starter Cultures as Pathogen Surrogates. Journal of Food Protection. 72:1234-1247. 4. California, A. B. o. April 2007. Considerations for Proprietary Processes Used for Almond Pasteurization and Treatment. 5. Ceylan, E., and D. A. Bautista. 2015. Evaluating Pediococcus acidilactici and Enterococcus faecium NRRL B-2354 as Thermal Surrogate Microorganisms for Salmonella for In-Plant Validation Studies of Low-Moisture Pet Food Products. Journal of Food Protection. 78:934-934. 6. Cosmed. 2017. Label Instructions For Propylene Oxide. 7. Farakos, S. M. S., D. W. Schaffner, and J. F. Frank. 2014. Predicting Survival of Salmonella in Low-Water Activity Foods: An Analysis of Literature Data. Journal of Food Protection. 77:1448-1448. 8. Jeong, S., B. P. Marks, and E. T. Ryser. 2011. Quantifying the Performance of Pediococcus sp. (NRRL B-2354: Enterococcus faecium) as a Nonpathogenic Surrogate for Salmonella Enteritidis PT30 during Moist-Air Convection Heating of Almonds. Journal of Food Protection. 74:603-603. 9. Joosten, H. M., E. Komitopoulou, t. B. H. Kuile, S. Fanning, L. R. Beuchat, F. Bourdichon, R. P. Betts, and H. Beckers. 2013. Low-water activity foods: increased concern as vehicles of foodborne pathogens. Journal of Food Protection. 76:150-72. 10. Palumbo, M., L. R. Beuchat, M. D. Danyluk, and L. J. Harris. 2016. Recalls of tree nuts and peanuts in the U.S., 2001 to present [table and references] U.S. recalls of nuts. 11. Santillana Farakos, S. M., R. Pouillot, N. Anderson, R. Johnson, I. Son, and J. Van Doren. 2016. Modeling the survival kinetics of Salmonella in tree nuts for use in risk assessment. International Journal of Food Microbiology. 227:41-50. 12. Systems, R. P. Date, 2012, Sterilization / Pasteurization of dried products. Available at: http://revtech-process-systems.com/index.php/en/steam-sterilization-pasteurization Accessed, January 2017.

6

CHAPTER 2: LITERATURE REVIEW

Microbiological Quality and Safety of Tree Nuts

Traditionally, tree nuts, a low aw food, would not have been suspect to harbor pathogenic

bacteria. Salmonella has been documented to survive within low water activity products for

periods up to 2 years (30). Microbiological surveying of tree nuts has also risen awareness of

actual microbiological quality of various tree nuts, at variable times throughout their production

(54). Surveying research has been carried out through studies in the United Kingdom assessing

microbiological safety of nuts in market places. While not showing much contamination of

Salmonella during the data collection, there is still evidence that further surveying should be

conducted (54). Research out of Australia, intended to survey retail nuts (read-to-eat foods) for

contamination of Salmonella and E. coli, concluded that while microbiological quality of nuts

and nut products are safe in Australia, there were small samples of macadamia nuts which could

be considered potentially hazardous, due to contamination with Salmonella Aberdeen. Though

there have not been any yet, it is plausible that if there is a high contamination level of nuts in

Australia, an outbreak is highly likely (6).

Tree Nuts

Tree nuts are grown and consumed by many cultures worldwide, they have become a

staple commodity in diets of many Americans. Tree nuts are rich in macronutrients,

micronutrients, and bioactive phytochemicals (1). A tree nut is defined as a hard dried fruit or

seed that has the ability to be separated into parts consisting of a rind or shell and an edible

kernel or “meat” that grows on a tree (64). Tree nuts consist of many varieties including walnuts,

pistachios, macadamia nuts, pecans, almonds, and hazel nuts. These nuts are a major export for

the United States, with the United States being one of the top producers of many of these

7

varieties (44, 79). There is also a demand for tree nuts that are not grown in the United States.

Importing cashews accounts for about 50% of all tree nut imports into the United States, with

India and Vietnam supplying almost 100% of product (79).

Whether tree nuts are grown in the United States or are imported from other supplying

countries, there are various ways in which raw products are harvested, prepared and further

processed. Processes can include forms of thermal treatments; oil roasting, dry roasting,

blanching, steam pasteurization, hot H2O pasteurization, and non-thermal treatments; Propylene

oxide and Ethylene oxide (38). Processing of tree nuts was traditionally intended to extend shelf

life, knowledge of low water activity foods provided the belief that tree nuts could be considered

microbiologically safe, since water activity fell below 0.7 (25). With the main focus on shelf life

extension, there were also a focus on hazards in controlling mycotoxin production. Mycotoxins

are produced by fungi like Aspergillus flavus and have been a source of intoxication in human

and animal consumption of products such as nuts throughout the world (9), pathogenic bacterial

contamination was overlooked for years (6). Bacterial contamination raises concern, with a

number of recalls and outbreaks stemming from contamination of tree nuts and tree nut products

(39), including contamination by Escherichia coli and Salmonella (6). Neither E. coli or

Salmonella can multiply on or in nuts and nut products, due to low water activity, but there have

been studies finding contamination can sustain life for extended periods of time, even up to over

a year (25, 37). Presence of both Salmonella and E. coli provide evidence, typically, of poor

harvest conditions, handling post-harvest, as well as through processing procedures such as

shelling, treating (thermal or non-thermal) and packaging (37).

During routine testing by the United States Food and Drug Administration (FDA),

Salmonella has been isolated from tree nuts including macadamia nuts, cashews, walnuts, brazil

8

nuts, pine nuts, and pistachios (62). In data on U.S. recalls of nuts, specifically tree nuts and

peanuts, a collection of 104 government issued recalls have been recorded from 2001 to January

of 2017, as a result of contaminated nuts and nut ingredients in products (62). Of the 104 product

recalls, only 12 resulted in reported cases of illness and 7 of those listed the source of

contamination to be “raw” nut products (60). The designation of “raw” represent nuts that have

not been treated with a pasteurization step. Pasteurization is an important risk reduction step as

research has shown that if almonds were to be consumed un-pasteurized there is about a 71%

increase contamination rate of Salmonella (24). With the FDA requiring pasteurization of

almonds now, the chance of contaminated ready-to-eat foods drastically drops. “Raw” almonds

sold in the grocery store have been through a pasteurization step, while “raw” almonds from the

local farmer’s market may not, providing a misconception of a safe “raw” product to many(52).

Also, not all tree nuts are required to be pasteurized, though steps to minimize contamination are

recommended and may soon be required. This is an interesting time with many aspects of the

food industry changing due to the Food Safety Modernization Act (FSMA), which has already

played a role changing the process requirements in almond processing (74). With regulations

looming on further nut processing laws, the idea that macadamia nuts and cashews need to be

further investigate is viable. These two very different tree nuts; in composition, fat content, size,

shape, and consumer expectations, have both been involved with various recalls in recent years

(62).

Macadamia Nuts

Macadamia nuts are one of the most sought out nuts falling under the category of one of

the world’s finest dessert nuts (86). The macadamia nut tree or Macadamia integrifolia is an

9

original native of the rainforests of eastern Australia, but was introduced to the Hawaiian Islands

~120 years ago (83). The macadamia nut produced by these trees has a crunchy texture and a

very high oil content, which adds to its smooth, buttery mouth feel and sweet taste (83). In

Hawaii alone, the macadamia nut has a farm value of over $40 million that is only increased

exponentially with the addition of other macadamia nut products that are produced (83).

Macadamia nuts do entail a relatively high production cost due to current harvesting

practices. Since there is a very poor recovery method for nuts that are harvested before the shell

is cracked or incompletely cracked, farmers must wait for the nut to ripen fully and spilt open on

its own (86). Therefore, macadamia nuts are typically harvested off the ground, the nut’s shell

will crack and fall to the ground as a sign that the nut has ripened, ideally, though as eluded to

before, many nuts are harvest before they reach maturity (74). Postharvest, dehydration is the

typical method of preservation in macadamia nuts. Macadamia nuts are dried to a point that there

is lack of water available to help prevent spoilage and microbial growth, all while aiding the

extraction of the kernel or edible nut meat that is desired from the shell (83).

Though it is necessary to allow the shell to break for optimum extraction of the kernel,

this is an area of possible Salmonella contamination in nuts. In a “Report on the prevalence of

Salmonella and E. coli in ready to eat nuts and nut products sold in Australia,” it is stated that in

fact “raw macadamia nuts are frequently contaminated with Salmonella, possibly due to

harvesting techniques”(6). This helps support the reasoning behind a recall that was initiated

after a random selection by a company outsourced by the US Food and Drug Administration

during a routine testing for nuts containing Salmonella. In 2015 recalling products located in 15

different states as well as Washington D.C (2-4). The recall included product labeled as “raw” as

10

well as other macadamia products like “whole roasted” (2-4). It is evident why raw product may

contain microbial contamination, but most would expect a roasted product to be contaminate

free. While the raw products hold more risk of contamination, they were not the only products

affected, providing evidence that critical control points were overlooked in the processing and

packaging steps of the macadamia nuts. Cross contamination between safe and un-safe product is

an assumable hypothesis and a critical control point that is readily investigated. Thus there is a

critical need to re-evaluate macadamia nut processing for a safety purpose.

Currently common practice for macadamia nut production is depicted through the below figure:

Figure 1: Described is a simplified adaptation of macadamia nut processing as published by the National Department of Agriculture in cooperation with ARC-Institute for Tropical and Subtropical Crops (28). This chart can help identify key steps in processing and highlight that contamination of low moisture foods is possible at any step in production, including cultivation/harvesting, processing, and packaging (63).

1) Macadamia nut drops from the tree: matured and collected from the

ground

2) Husk removed from around the nuts

immeadiately following harvest

3) Dried after husking: prevents fungal growth

during storage

4) Dried/hardened "undamaged" nuts put

into storage

5) Shelling: should be dried to a moisture

content of 1.5% to have kernel shrink away from

shell

6) Roasting/fried step typically

7) Package nuts in airtight bottles, tins or plastic

containers

11

Many resources show common practice and list critical areas in processed such as

preventing degradation of nuts through insect infestation as well as list step 1 above, as needing

to be harvested daily to avoid loss of product from rodents and other animals, but there is no

mention of microbiological contamination control points (28). As mentioned by “Improving the

Safety and Quality of nuts,” implementation of a Hazard Analysis Critical Control Point plan

(HACCP plan) can be critical (42). The FDA defines HACCP as “a management system in

which food safety is addressed through analysis and control of biological, chemical, and hazards

from raw material production, procurement and handling, to manufacturing, distribution and

consumption of the finished product (32)” When implementing a HACCP plan, the plan must be

developed for the exact product and exact processing parameters, as does any validation study,

so the implementation of a HACCP plan would be different for macadamia nuts then on would

be for cashews (42).

Cashews

Cashews include another area of agriculture where changing process requirements is in

the eminent future. The cashew tree, Anacardium occidentale, is a fast growing, hardy and

drought tolerant tree that can thrive in areas that are typically less than ideal for many

agricultural products (66). The cashew tree is a tropical tree native to areas of South America,

but was introduced to African and Asian nations by explorers of the 16th century (27, 66). The

trees themselves typically grow anywhere from 8 to 12 meters, and hold host to the cashew tree

fruit that contains the edible nut kernel (66). With the cashew’s high economic, ecological, and

biological potential, it is no surprise that cashew nut production has increased exponentially in

the past few decades (66). The world production of cashew raw nuts reached 4.27 million tons in

12

2011, with the United States importing a great amount of product from countries in Asia and

Africa (27).

Harvesting and processing of cashews can be an arduous undertaking. The cashew, grows

on the end of the “cashew apple” fruit, when matured, the nut falls to the ground, where it is then

collected from the ground surrounding the trees for further processing (7).

Further processing of Cashews:

Figure 2: A simplified overview of steps in cashew harvest and processing to remove from shell as listed in “ Small-scale cashew nut processing” (7). This chart can help identify key steps in processing and highlight that contamination of low moisture foods is possible at any step in production, including cultivation/harvesting, processing, and packaging (63).

1) Mature cashews are collected from the ground

2) Cashews are dried in the sun, may take 1-3

days.

3) Storage of cashews, should be able to keep for

12 months under dry conditions in sealed

polythene bags

4) Nut soaked in water, soften by steaming to

prep for extraction

5) Nut dried to moisture content of 9%

6) Nut cracked mechanically to allow

separation of outer shell and nut meat.

7) Nut meat carefully removed using a nut pick.

13

Cashews are another product that processing can be re-evaluated for safety purposes. The United

States relies on countries like India and Vietnam for much of its cashew import, and with

cashews in high demand it is necessary that the proper processing and evaluations of the imports

are followed (79). The cashew has been recorded with water activity ranging from 0.2 – 0.8

depending on processing parameters and methods, typically measuring around the 0.6 range

when the nut has been dried and processed, with packaging and storage methods factored in as

well (10, 85). Carefully monitoring and processing an imported product can provide a

microbiologically safer product for consumers.

Saturated Steam Treatment for Nuts

Utilization of steam to achieve a required pathogen log reduction in a product, putting it

simply, is what a saturated steam treatment aims to achieve (16, 77). Saturated steam is a form of

thermal pasteurization that consumers tend to approve of when concerned with the use of food

treated with Ethylene Oxide (ETO) as well as Propylene Oxide (PPO) (gas fumigations) (77).

Saturated steam can be used on low moisture foods to help preserve the organoleptic properties.

The ability to use little amounts of steam which transfer high heat, but prevent major losses of

aromatics and changes that are detrimental to desired qualities of the product appeals to the

industry and consumer (45, 77). Research is ongoing, but there is little to no change in

appearance in most products, with no damage to fragile food products (77). Saturated steam,

because it stores a large amount of latent heat, can provide high heat transfer to the outside of a

product. There are steps in processing with saturated steam that may vary between system, but

follow basics of pre-heating used to elevate temperature of the raw product and chamber to

specified degree, the steam treatment which is used to inactivate potential pathogenic bacteria,

and a cooling stage necessary for further processing post-treatment (71). The use of saturated

14

steam can also be conducted in a vacuum chamber and with different pressures, allowing for

lower temperatures to be used to process nuts (42). Steps in steam treatments with a vacuum

consider a pre-vacuum, vacuum, pasteurization, and post vacuum stages that can all be adjusted

in regards to the matrix being processed (71). As described by M.K. Shah et al., in a study

researching the “Efficacy of vacuum steam pasteurization for inactivation of Salmonella PT 30,

Escherichia coli 0157:H7 and Enterococcus faecium on low moisture foods”, steam processing

while under a vacuum allows for processing to occur at temperatures below 100 ºC (71). This is

a notable parameter, allowing for different texture, flavor, and overall mouth feel of end products

by varying processing time and max temperatures reached. Organoleptic properties are important

to the producer to preserve, but some processing is required to obtain certain properties as well,

for example developing flavor in particular nuts by dry or oil roasting them (42).

From a microbiological stand point, the California Almond Board (ABC) discusses using

a saturated steam pasteurization system, using mandatory steam treatments to obtain the

Technical Expert Review Panel’s (TERP) 4-log reduction of Salmonella in steam pasteurization

of almonds as well as a proprietary process that was accepted for pasteurization, obtaining a 5-

log reduction minimum (16). Parameters that have been validated and can be obtained through

contact with ABC or TERP, provided natural almonds treated under these conditions the ability

to be labeled as “pasteurized.” The validation must be followed exactly, with no deviations in

areas such as time, temperature, packaging, and food matrix (16).

15

Propylene Oxide (PPO) treatment for tree nuts:

Propylene oxide (PPO) or C3H6O is a flammable liquid with a boiling point of 34.238 ºC

at normal atmospheric pressures (26). PPO is vaporized to a gas in a fumigation treatment that is

known to reduce populations of bacteria, molds and yeast in food items such as tree nuts, spices,

and cocoa (26). The table below lists commodities with established tolerances in parts per

million treated with PPO per “Reregistration Eligibility Decision (RED) Document for

Propylene Oxide” for Cosmed Group, as archived by the EPA (23, 29).

Table 1. Foods recognized as approved for PPO treatment

Commodity Approved for PPO Treatment Parts Per Million (PPM) Nut, pine 300 Nut, tree (group 14) 300 Nutmeat, processed (excludes peanuts) 300 Pistachio 300 Cacoa bean, dried bean 200 Cacoa bean, coaoa powder 200 Fig 3.0 Grape, raisin 1.0 Plum, prune (dried) 2.0 Herbs and spices, dried (group 19) 300 Garlic, dried 300 Onion, dried 300

Fumigation processes have been extensively discussed in the nut and spice industry. Recently the

Almond Board of California researched and demonstrated that PPO fumigation is an effective

form of pasteurization, for almonds (17). Research has found that PPO can reduce the risk of

Salmonella contaminated almonds from reaching the consumer, achieving a 4-log reduction.

16

Table 2. Pasteurization Operating Parameters to Assure 4-log Reduction of Salmonella

PPO Pasteurization Operating Parameters to Assure 4-log Reduction of Salmonella (17)

Stage Temperature

Initial Product Not less than 86ºF (30ºC)

Inside Chamber: Start & Sterilization 117-125ºF(47-51ºC)

PPO vaporizer temperature at point of PPO injection

140 -160ºF (60-71ºC)

Post ventilation 100-110ºF (38-43ºC) for 2 days or above 59ºF (15ºC) for 5 days

Chamber Vacuum Vacuum (Hg)

Before PPO injection At least 27” Hg

Completion of inert gas injection 5-6 inch Hg

Initial PPO concentration in chamber Not less than 0.5oz PPO/ft3

Duration of Pasteurization 4 hours (industry size)

Aeration Cycles No <4 and not >14

This PPO process validation was taken directly from parameters the California Almond Board

(17) follow when validating a process to produce a 5-log reduction in Salmonella on almonds.

The procedure in table is applicable for bulk-packed almonds that are stacked single or double

stacked pallets, and should be noted that this procedure is in no way applicable for products

packed in retail packaging (17).

17

The U.S. Environmental Protection Agency (EPA) registrater for PPO as a pesticide

fumigant stipulates that (i) exposure temperature must not exceed 51.68 ͦC, (ii) exposure time

shall be no more than 4 h, and (iii) PPO residue in the product (prior to release) shall not exceed

300 ppm. Label instructions additionally dictate that (i) dosage cannot exceed 2.5 kg/m3 and (ii)

at the end of treatment there should be no less than four aeration cycles with a PPO chamber

volume of air (EPA registration no. 47870-1, Aberco, Seabrook, MD.)(26, 29). Requirements by

the EPA have limits on exposure times, and amounts of off gassing allowed by one treatment

center, but processors continue to use a variety of processes, many of which are not validated for

Salmonella (42). PPO does have some drawbacks, the process is lengthy due to off gassing

required by the EPA, which includes day long periods before food products can be further

handled. While PPO is effective in inactivating Salmonella, factors such as chamber size,

product composition, containers which the product is being processed in all play a role in the

efficiency of the process (42).

With strict regulations on how much PPO processors can harness for treatment and

expose to a food system, chamber size must be validated to determine distribution of gas within

the chamber and how much exposure the food system will have. Understanding the exposure to a

product is necessary to verify efficacy of the fumigation treatment and can also be effected by

the food composition being tested (23, 46). How tightly a food sits on top of itself in a container

can change permeability and prevent high enough exposures in the center of a container, which

can lead to under processed product. Not only does the food system need to allow for gas

exposure, the container which foods are processed in have the requirement of being permeable,

just as requirements for medical equipment treated with ethylene oxide (56). Permeable has a

wide definition itself, to PPO standards a carboard box is considered permeable (41). A carboard

18

box would vary in permeability though compared to a bin constructed of wire mesh, further

proving a need to validate every process separately. The promise for treatments such as

saturated steam (8) and PPO (26) in the reduction of Salmonella contamination, and therefore

reduction in cases of Salmonellosis in the United States, do warrant continuing research in

adding additional food items to the list which are already processed in such ways.

Salmonellosis

Salmonellosis is an illness associated with food and waterborne outbreaks worldwide

caused by bacteria known as Salmonella (58). Salmonella is a motile, non-spore forming, Gram

negative, rod-shaped bacterium (21). In the United States, Salmonella is the foodborne bacteria

associated the most foodborne illnesses as well as deaths from foodborne bacteria, with an

estimated 450 each year (69). When infected with Salmonellosis, there are two kinds of illnesses

that could occur. The first of which, non-typhoidal, is gastrointestinal, bringing about nausea,

vomiting, diarrhea and cramps, with additional fever-like symptoms that generally last only 4 to

7 of days and will reside by themselves in healthy patients, and is the most prevalent in the

United States (19, 21).

The second type, typhoidal, is seen more in developing countries and not typically in The

United States, and typically can be associated with travel outside of the country for United States

citizens (19). Typhoidal infections typically arise from undertreated water sources that may

contain the pathogen and items such as raw sewage, transmitting the illness by directly drinking

the source or using the non-potable water for crops and agricultural items, and provides enough

concern to closely monitor where the United States imports products (21). Typhoidal illness is

usually associated with high fever, diarrhea or constipation, aches, pains, drowsiness, as well as

19

sometimes a rash, and tends to be more serious with 10% of untreated cases leading to death

(21). In both instances of Salmonellosis illnesses, patients who are elderly, immunocompromised

or infants are the most likely to develop a severe illness. The CDC estimates that there are about

1.2 million illnesses each year with somewhere around 450 deaths occurring from non-typhoidal

Salmonella in The United States annually, with many cases still going unreported (19).

Salmonellosis caused by non-typhoidal Salmonella traditionally concerned regulators for

contamination of raw meat products, such as chicken. It has been supported over the years that

under processed and contaminated food items such as nuts and spices have led to many recalls

and outbreaks and raise concern if not more for contamination (67). Per the CDC and the

Foodborne Outbreak Online Database (FOOD), there have been 4 outbreaks of Salmonella on

nuts in the United States from 1998 to 2015 (20) warranting further investigation of Salmonella

in low water activity environments.

Salmonella Survival at Low Water Activity

One of the oldest forms of food preservation is the drying or dehydrating of a food

system. By using drying methods for food preservation, the result is lowering the water activity

to that of a point that the environment is no longer suitable for most microbial growth (43, 65).

Drying helps prevent microbial growth due to reduction of available water from growth also

known as water activity (aw ) recorded on a scale from 0-1(34). It specifically refers to the

amount of water that is free or unbound to the food system, that is available for supporting

microbial growth. Food systems with aw of 0.7 or lower typically do not support growth of

bacteria such as Salmonella, however there is still the ability of food to harbor pathogens, even

with this classification of low aw (30). Low aw does not only refer to foods that appear dry or low

in moisture but also, water that would typically be available in some high moisture foods can be

20

bound through the addition of items such as salt, sugar, and other topical items. Foods with low

aw common in diets can include dried nuts and spices, chocolate, dehydrated foods such as meats

and ready-to-eat meal mixes, as well as high moisture items such as jams and jellies.

With the capability to survive on low aw foods, many ready to eat foods, that are popular

in The United States, have potential contamination risk factors. Salmonella has been seen to

survive on tree nuts and tree nut products for extended lengths of time (67), though Salmonella

has only proven to grow at aw as low as 0.93 (11). Bacterial loads and time of contamination also

provide evidence on survival length in low aw systems. Taking into consideration environmental

and storage aspects of food such as temperature, relative humidity, as well as potential for cross

contamination provide insight to the risk of contamination in food (47, 51, 78). Salmonella can

easily be transmitted through cross-contamination at many critical control points during

processing of food. It is likely to occur during inappropriate preparation or unsanitary harvest

and preparation facilities, creating a higher risk of Salmonella contaminated food systems. Cross

contamination is particularly dangerous to a system when it occurs after critical steps are taken to

control for lethality of pathogenic bacteria (35, 58, 63).

Salmonella enterica Strain Consideration

When conducting a process validation, strain consideration of pathogenic bacteria which

you are trying to inactivate is critical. Many factors must be considered when identifying how

Salmonella strains will react with the food system and during the inoculation process. Factors

such as temperature, culture media, serotype, strain, water activity, nutrient matrix as well as

different processing parameters carried out will influence the Salmonella survival (63, 68). For

studies such as these, if low water activity foods are to be researched for process validation, it is

21

ideal to choose strains that have been found in outbreaks in foods of similar composition,

accessible through survey research and government records (62). Salmonella enterica serovars

obtained from low-aw foods are of interest, examples of strains of common interest include;

Montevideo, an isolate from 2010 peppercorn associated outbreak (18) occurred July 1, 2009 -

April 28, 2010, resulting in 272 people from 44 states contracting salmonellosis by Salmonella

Montevideo after consuming salami that had contaminated peppercorn throughout. Salmonella

Tennessee, K4643, a human isolate from 2007 peanut butter associated outbreak that was traced

back to Salmonella contaminated peanut butter. The outbreak consisted of 715 salmonellosis

cases spread over 48 different states (72). Salmonella Ball ARL-SE-085 isolated from black

pepper in 2011(5, 80) along with Salmonella Johannesburg ARL-SE-013 retrieved from dried

ginger in 2010 (31). Salmonella Senftenberg 775W, ATCC 43845, is used in studies which entail

a thermal processing step due to its heat resistant nature, and was obtained from the FDA or

American Type Culture Collection (ATCC, Manassas, VA) (57).

The importance of strain selection is evident with the selected strains all resulting in

larger outbreaks, and being strains that have been found to contaminate low water activity food

in high bacterial loads with the potential of spreading Salmonellosis.

Inoculation Preparation Effect on Low Water Activity Foods

Inoculation methods of the product being tested remains an important step in the testing

of surrogate organisms. Starting inoculum must be sufficient to document 5-log reduction using

plate count methods. With numbers of initial inoculum too low, it would be hard to obtain any

significant data showing sufficient inactivation and regrowth post processing (14, 15).

22

Wet Inoculation:

Wet inoculation methods are used in order to simulate potential Salmonella

contamination that could occur in natural growing habitats (53). The use of wet inoculum

produced by growing Salmonella and potential surrogate organisms in TSB (tryptic soy broth)

and then growing bacterial lawns on TSA (tryptic soy agar) allows for high concentrations of

each microorganism to be introduced into the food system being tested (53). These high counts

allow for proper enumeration techniques to acquire a recoverable number of bacteria to further

validate which surrogate organism express similar characteristics as the Salmonella inactivation

that is also present in the system. With producing high bacterial loads, wet inoculation allows for

easy reproducibility between trials of experiments, almost guaranteeing high counts each time it

is properly performed. Wet inoculation is able to change the surface properties in some food

systems, and includes extra time for drying the test product, but in the case of nuts, it seems to be

an effective way to still study inactivation techniques on Salmonella and surrogate survival (82).

The composition of nuts, macadamia nuts and cashews, provide testable foods which seem to fit

the model for a wet inoculation process. Cashews and macadamia nuts (whole), while not very

porous, do not provide evidence of drastic compositional changes, as would some products when

exposed to a wet inoculation and provide a great enough surface area to retain the inoculum (14,

15). Water activity is recorded post inoculation to ensure that the nuts return to a range with in

their original states, to show that the manipulation of water activity changes were able to reverse

with additional drying time (61).

23

Surrogate Bacteria

Microorganisms used as surrogates must follow a strict criterion. The ideal surrogate is

non-pathogenic, so that it can be introduced into a food system in a processing facility, to

validate the efficiency of a process intended to inactivate a target pathogenic microorganism

(50). Choosing the correct surrogate typically entails that is models inactivation of the target

pathogenic microorganism, consistently exemplifies stable and similar growth patterns, can be

easily prepared in a laboratory setting, and is easily identified and enumerated for pre and post

processing (76). The intention of surrogate use is, if it resists the processing conditions more than

your target pathogenic microorganism, then processors can feel confident if their process

inactivates the bacterial load of the surrogate, the product is safer for consumption by consumers.

Many surrogate organisms have been identified for target pathogen inactivation in a wide variety

of processes. It has been validated to use Bacillus amyoliquefaciens as a surrogate for

Clostridium botulinum in high pressure processing (HPP) of mashed carrots (55), Listeria

innocua has been validated for comparison to Listeria monocytogenes inactivation in flash

pasteurization of frankfurters (75) and the generic E. coli is a valid surrogate for E. coli 0157:H7

in pulsed field treatment of juice (40). The diversity of processes and mechanisms used to

inactivate pathogenic bacteria in the studies do vary, showing the importance of effectiveness of

control points identified. When investigating Salmonella inactivation, thermal mechanisms

remain a focus as a control point when choosing a comparable surrogate, typically.

Enterococcus faecium NRRL B-2354 as a surrogate for Salmonella

Enterococcus faecium NRRL B-2354 has potential as a surrogate microorganism having

been validated to result in comparable inactivation of Salmonella. Enterococcus faecium, a

Gram-positive coccus, non-spore-forming organism, known to survive in temperatures between

24

5-50°C, pH between the range of 4.6-9.9, and has a tolerance of high salt environments (36, 48).

It has been validated by research as surrogate for Salmonella inactivation on almonds with the

use of thermal treatments (76). With this validation study, it was the goal to determine if specific

treatment technology and processing equipment could achieve a 4-log reduction of Salmonella.

The 4-log reduction is the mandated minimum for almonds grown in California as a part of 7

CFR part 981 from the Federal Register in 2007 (76).

The safety of this strain has been evaluated in previous studies based on its genomic and

functional characteristics (49). Enterococcus faecium is used as a surrogate in many different

thermal lethality studies including as a surrogate for Salmonella enterica during the extrusion of

a balanced carbohydrate-protein meal, and has been evaluated in use on almonds processed with

PPO (12, 17, 76). Enterococcus faecium is also already, by the United States Food and Drug

Administration (FDA), Generally Recognized as Safe (GRAS) (33). GRAS status, under 201 and

409 of the Federal Food, Drug, and Cosmetic Act, recognizes the safety in the use of any

substance that is intentionally used as a food additive when used in the conditions which it is

intended (33). To gain GRAS status, a great deal of funding and research must go into all aspects

of an additive, so the positives of finding a surrogate that is already considered GRAS would

prove extremely fortunate to industry producers.

With the California Almond board produced research on using potential surrogate

bacteria such as Enterococcus Faecium, there remains a need to validate separate processes and

food products. These validations must cater to specific needs through every parameter set up, to

test surrogate validity as a whole (66, 83).

25

Pediococcus acidilactici ATCC 8042 as a surrogate for Salmonella

Pediococcus acidilactici ATCC #8042 is a lactobacilli bacteria that has traditionally been

used in meat fermentations (87). Pediococcus acidliactici ATCC #8042 is also recognized as

GRAS (33). Along with E. faecium NRRL B-2354, Pediococcus acidilactici ATCC #8042 has

been used as a surrogate in previous work. Popular with low water activity studies, Pediococcus

acidilactici ATCC #8042 acted as a surrogate in ground-and-formed beef jerky, where overall

reduction indicated >5-log CFU/g reduction, as found sufficient in a process for Salmonella (13)

and whole muscle turkey jerky (84), as have other lactobacilli bacteria.

Pediococcus acidilactici ATCC #8042 is also known for its use as a surrogate for the

dried pet food industry (22).Though it has also been approved as surrogate for Salmonella

inactivation in low water activity foods undergoing thermal processes, validation is still required

of the individual food systems (66, 83). In the study on dried pet food, validity of surrogacy

potential of Pediococcus acidilactici ATCC #8042 and Enterococcus faecium NRRL B-2354

were directly compared for Salmonella (serovars: Anatum, Montevideo, Senftenberg, Tennessee,

Schwarzengrund, Infantis, Mbandaka) (22). In the parameters of the treatment, both non-

pathogenic organisms proved to be valid surrogates, but when validated in the temperatures of

typical processing conditions, Pediococcus acidilactici ATCC #8042 was found a more viable

surrogate to work with in future studies due to inactivation characteristics (22).

Staphylococcus carnosus ATCC #51365 as a surrogate for Salmonella

Staphylococcus carnosus is Gram positive, nonmotile, non-spore forming bacteria. The

bacteria colonies grow as small, raised circular and gray-white appearance in color on mannitol

salt agar (MSA) (70). Staphylococcus aureus will also grow in this medium, but is considered

26

mannitol fermenting, creating MSA to undergo a color change to a yellowish (73), one of the

many testing methods available for identifying Staphylococcus aureus. It is important to be able

to tell Staphylococcus carnosus apart from other cocci bacteria when using them as a indicator

organism. Staphylococcus carnosus ATCC #51365 strains have traditionally been applied as

meat starter cultures in fermented meat products (59). Promoting research in many areas

regarding the safety of the use of this starter culture. With questions arising on safety of the

bacteria to those who may be immunocompromised, and its recent history of antibiotic resistant

strain detection in the food processing industry, it is important to note that Staphylococcus

carnosus has GRAS (generally recognized as Safe) status, making it an ideal subject to test

surrogacy for Salmonella inactivation, from a monetary stand point (33, 59). Thermal resistant

capabilities along with being able to withstand various salt contents, water activities and

environmental conditions provide grounds for research in comparison to Salmonella inactivation

on processed low water activity foods (59, 70).

Staphylococcus carnosus subspecies carnosus ATCC #51365, along with Pediococcus

acidilactici ATCC #8042 have previously been investigated as possible surrogates for

inactivation of Salmonella and E. coli 0157:H7 in ground beef products and frankfurter batter

(81). The thermal tolerance of pathogen cocktails (E.coli, Salmonella) was shown to be

significantly less (P < 0.05) than thermal tolerance of Pediococcus acidilactici ATCC #8042 and

Staphylococcus carnosus ATCC #51365 (81). Approval as surrogates (Staphylococcus carnosus

ATCC #51365, Pediococcus acidilactici ATCC #8042) for Salmonella and E. coli O157:H7

inactivation on frankfurter batter was found, but this is only under limited parameters of thermal

processing (81).

27

References

1. Alasalvar, C., and F. Shahidi. 2008. Tree nuts: composition, phytochemicals, and health effects. CRC Press. 2. Anonymous. 2015. Aurora Products recalls raw macadamia nuts over Salmonella risk. In, Progressive Digital Media Food News U6 - Newspaper Article Progressive Digital Media, London. 3. Anonymous. 2015. Texas Pecan recalls macadamia nuts over potential Salmonella contamination. In, Progressive Digital Media Food News U6 - Newspaper Article Progressive Digital Media, London. 4. Anonymous. 2015. Texas Star recalls Nature's Eats Natural Macadamia Nuts over Salmonella risks. In, Progressive Digital Media Food News U6 - Newspaper Article Progressive Digital Media, London. 5. Anonymous. Date, 2014, Salmonella Ball (presence/25g) in black pepper from Hungary, with raw material from Poland. Accessed, 2015. 6. Authority, N. F. 2013. Report on the prevalence of Salmonella and E. coli in ready to eat nuts and nut products sold in Australia. In. 7. Azam-Ali, S., and E. Judge. 2001. Small-scale cashew nut processing. Coventry (UK): ITDG Schumacher Centre for Technology and Development Bourton on Dunsmore. 8. Ban, G.-H., and D.-H. Kang. 2016. Effectiveness of superheated steam for inactivation of Escherichia coli O157:H7, Salmonella Typhimurium, Salmonella Enteritidis phage type 30, and Listeria monocytogenes on almonds and pistachios. International Journal of Food Microbiology. 220:19-25. 9. Bayman, P., J. L. Baker, and N. E. Mahoney. 2002. Aspergillus on tree nuts: incidence and associations. Mycopathologia. 155:161-169. 10. Beuchat, L. R. 1978. RELATIONSHIP OF WATER ACTIVITY TO MOISTURE CONTENT IN TREE NUTS. Journal of Food Science. 43:754-755. 11. Beuchat, L. R. 2009. Behavior of Salmonella in foods with low water activity. In, IAFP Rapid response symposium, Arlington, VA. 12. Bianchini, A., J. Stratton, S. Weier, T. Hartter, B. Plattner, G. Rokey, G. Hertzel, L. Gompa, B. Martinez, and K. M. Eskridge. 2014. Use of Enterococcus faecium as a Surrogate for Salmonella enterica during Extrusion of a Balanced Carbohydrate-Protein Meal. Journal of Food Protection. 77:75-75. 13. Borowski, A. G., S. C. Ingham, and B. H. Ingham. 2009. Lethality of home-style dehydrator processes against Escherichia coli O157:H7 and Salmonella serovars in the manufacture of ground-and-formed beef jerky and the potential for using a pathogen surrogate in process validation. Journal of Food Protection. 72:2056-2064. 14. Bowman, L. S. 2015. Impacts of inoculation strategy on survival of Salmonella enterica and surrogate Enterococcus faecium at low water activity on dry peppercorn and cumin seeds. In, Food Science and Technology, vol. Master of Science in Life Science. Virginia Tech, Blacksburg, VA. 15. Bowman, L. S., K. M. Waterman, R. C. Williams, and M. A. Ponder. 2015. Inoculation Preparation Affects Survival of Salmonella enterica on Whole Black Peppercorns and Cumin Seeds Stored at Low Water Activity. Journal of Food Protection. 78:1259-1259. 16. California, A. B. o. April 2007. Considerations for Proprietary Processes Used for Almond Pasteurization and Treatment.

28

17. California, A. B. o. October 2008. Guidelines for Validation of Propylene Oxide Pasteurization. 18. CDC. Date, 2010, Multistate outbreak of human Salmonella Montevideo infections (final update). Accessed February 25, 2016. 19. CDC. March 9, 2015. Salmonella 20. CDC, C. f. D. C. a. P.-. Date, August 2016, Foodborne Outbreak Online Database (FOOD Tool). Available at: https://wwwn.cdc.gov/foodborneoutbreaks/ Accessed, March 2017. 21. Center for Food, S., and N. Applied. 1992. The bad bug book: foodborne pathogenic microorganisms and natural toxins handbook. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, Washington, D.C.- eBook. 22. Ceylan, E., and D. A. Bautista. 2015. Evaluating Pediococcus acidilactici and Enterococcus faecium NRRL B-2354 as Thermal Surrogate Microorganisms for Salmonella for In-Plant Validation Studies of Low-Moisture Pet Food Products. Journal of Food Protection. 78:934-934. 23. Cosmed. 2017. Label Instructions For Propylene Oxide. 24. Danyluk, M. D., L. J. Harris, and D. W. Schaffner. 2006. Monte Carlo simulations assessing the risk of salmonellosis from consumption of almonds. Journal of Food Protection®. 69:1594-1599. 25. Danyluk, M. D., T. M. Jones, S. J. Abd, F. Schlitt-Dittrich, M. Jacobs, and L. J. Harris. 2007. Prevalence and amounts of Salmonella found on raw California almonds. Journal of food protection. 70:820-827. 26. 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. Journal of Food Protection. 68:1613-1622. 27. Dendena, B., and S. Corsi. 2014. Cashew, from seed to market: a review. Agronomy for Sustainable Development. 34:753-772. 28. Directorate Communication, N. D. o. A. i. c. w. A.-I. f. T. a. S. C. Date, 2000, Cultivation of macadamias. Available at: http://www.nda.agric.za/docs/Infopaks/macadamia.htm. Accessed, March 2017. 29. EPA. 2004. Propylene oxide tolerances for residues. Code of Federal Regulations title 49, chapter 1, section 180.493. 30. Farakos, S. M. S., D. W. Schaffner, and J. F. Frank. 2014. Predicting Survival of Salmonella in Low-Water Activity Foods: An Analysis of Literature Data. Journal of Food Protection. 77:1448-1448. 31. FDA. 2013. Draft risk profile: Pathogens and filth in spices. 32. FDA. Date, 2015, Hazard Analysis Critical Control Point (HACCP). Available at: https://www.fda.gov/Food/GuidanceRegulation/HACCP/. Accessed, March 2017. 33. FDA. Date, 2017, Generally Recognized as Safe (GRAS). Available at: https://www.fda.gov/Food/IngredientsPackagingLabeling/GRAS/ Accessed, March 2017. 34. Finn, S., O. Condell, P. McClure, A. Amézquita, and S. Fanning. 2013. Mechanisms of survival, responses and sources of Salmonella in low-moisture environments. Frontiers in Microbiology. 4:331. 35. Finn, S., J. C. D. Hinton, P. McClure, A. Amezquita, M. Martins, and S. Fanning. 2013. Phenotypic Characterization of Salmonella Isolated from Food Production Environments Associated with Low-Water Activity Foods. Journal of Food Protection. 76:1488-1488.

29

36. Fisher, K., and C. Phillips. 2009. The ecology, epidemiology and virulence of Enterococcus. Microbiology. 155:1749-1757. 37. Freire, F. d. C. O., and L. Offord. 2002. Bacterial and yeast counts in Brazilian commodities and spices. Brazilian Journal of Microbiology. 33:145-148. 38. GMA. 2010. Industry handbook for safe processing of nuts. GMA Nut Safety Task Force, Grocery Manufacturers Association, Washington, DC. 39. Gurtler, J. B., M. P. Doyle, J. L. Kornacki, and SpringerLink. 2014. The Microbiological Safety of Low Water Activity Foods and Spices. Springer New York, New York, NY. 40. Gurtler, J. B., R. B. Rivera, H. Q. Zhang, and D. J. Geveke. 2010. Selection of surrogate bacteria in place of E. coli O157: H7 and Salmonella Typhimurium for pulsed electric field treatment of orange juice. International journal of food microbiology. 139:1-8. 41. Howe, D. 2017. Cosmed Group. In M. Ponder (ed.). 42. Hurst, W., and L. Harris. 2013. Integrating Hazard Analysis Critical Control Point (HACCP) and Statistical Process Control (SPC) for safer nut processing. Improving the safety and quality of nuts. 15:119-147. 43. ICMSF. 2006. Microorganisms in Foods 6 : Microbial Ecology of Food Commodities. Springer, Boston, MA. 44. Johnson, D. C. 1997. United States is world leader in tree nut production and trade. USDA-ERS Fruit and Tree Nuts Situation and Outlook. FTS-280. 45. Joosten, H. M., E. Komitopoulou, t. B. H. Kuile, S. Fanning, L. R. Beuchat, F. Bourdichon, R. P. Betts, and H. Beckers. 2013. Low-water activity foods: increased concern as vehicles of foodborne pathogens. Journal of Food Protection. 76:150-72. 46. Kashaninejad, M., A. Mortazavi, A. Safekordi, and L. Tabil. 2006. Some physical properties of Pistachio (Pistacia vera L.) nut and its kernel. Journal of Food Engineering. 72:30-38. 47. Keller, S. E., J. M. VanDoren, E. M. Grasso, and L. A. Halik. 2013. Growth and survival of Salmonella in ground black pepper (Piper nigrum). Food Microbiology. 34:182-188. 48. Klein, G. 2003. Taxonomy, ecology and antibiotic resistance of enterococci from food and the gastro-intestinal tract. International journal of food microbiology. 88:123-131. 49. Kopit, L. M., E. B. Kim, R. J. Siezen, L. J. Harris, and M. L. Marco. 2014. Safety of the Surrogate Microorganism Enterococcus faecium NRRL B-2354 for Use in Thermal Process Validation. Applied and Environmental Microbiology. 80:1899-1909. 50. Kornacki, J. L. 2011. Indicator organisms: Chaos, confusion and criteria. In, Food Safety Magazine, Testing. 51. Kotzekidou, P. 1998. Microbial stability and fate of Salmonella Enteritidis in halva, a low-moisture confection. Journal of Food Protection. 61:181-185. 52. Lee, L. E., D. Metz, M. Giovanni, and C. M. Bruhn. 2011. Consumer Knowledge and Handling of Tree Nuts: Food Safety Implications. p. 18. In, vol. 31. Allen Press Publishing Services, Des Moines. 53. Liao, L. 2016. Efficacy of aqueous inoculation in low-water activity matrix on Salmonella enterica serotype Agona thermal resistance and stability during storage. In, vol. Dissertation/Thesis. ProQuest Dissertations Publishing. 54. Little, C. L., W. Jemmott, S. Surman-Lee, L. Hucklesby, and E. de Pinna. 2009. Assessment of the Microbiological Safety of Edible Roasted Nut Kernels on Retail Sale in England, with a Focus on Salmonella. Journal of Food Protection. 72:853-855.

30

55. Margosch, D., M. A. Ehrmann, M. G. Gänzle, and R. F. Vogel. 2004. Comparison of pressure and heat resistance of Clostridium botulinum and other endospores in mashed carrots. Journal of Food Protection®. 67:2530-2537. 56. Mendes, G. C., T. R. Brandao, and C. L. Silva. 2007. Ethylene oxide sterilization of medical devices: a review. American journal of infection control. 35:574-581. 57. Michael, G. T., and C. R. Stumbo. 1970. Ethylene oxide sterilization of Salmonella Senftenberg and Escherichia coli death kinetics and mode of action. Journal of Food Science. 35:631-634. 58. Motarjemi, Y., G. Moy, and E. C. D. Todd. 2014. Encyclopedia of food safety. Academic Press Inc, Boston. 59. Müller, A., R. Reichhardt, G. Fogarassy, R. Bosse, M. Gibis, A. Weiss, J. Weiss, and H. Schmidt. 2016. Safety assessment of selected Staphylococcus carnosus strains with regard to their application as meat starter culture. Food Control. 66:93-99. 60. OHA. Date, 2017, Salmonellosis cases linked to hazelnuts sold at roadside stand. Available at: http://www.oregon.gov/oha/news/Pages/Salmonellosis%20cases%20linked%20to%20hazelnuts%20sold%20at%20roadside%20stand.aspx Accessed, 2017. 61. Palipane, K. B., and R. H. Driscoll. 1993. Moisture sorption characteristics of in-shell macadamia nuts. Journal of Food Engineering. 18:63-76. 62. Palumbo, M., L. R. Beuchat, M. D. Danyluk, and L. J. Harris. 2016. Recalls of tree nuts and peanuts in the U.S., 2001 to present [table and references] U.S. recalls of nuts. 63. Podolak, R., E. Enache, W. Stone, D. G. Black, and P. H. Elliott. 2010. Sources and risk factors for contamination, survival, persistence, and heat resistance of Salmonella in low-moisture foods. J Food Prot. 73:1919-36. 64. Pollack, S., and A. Perez. 2008. Fruit and tree nuts situation and outlook yearbook 2008. Washington, DC: US Department of Agriculture-Economic Research Service. 65. Rojas, F. J., M. Jodral, F. Gosalvez, and R. Pozo. 1991. Mycoflora and toxigenic Aspergillus flavus in Spanish dry-cured ham. International Journal of Food Microbiology. 13:249-255. 66. Salam, M. A., and K. V. Peter. 2010. Cashew : A Monograph. Studium Press (India) Pvt. Ltd., Delhi, IND. 67. Santillana Farakos, S. M., R. Pouillot, N. Anderson, R. Johnson, I. Son, and J. Van Doren. 2016. Modeling the survival kinetics of Salmonella in tree nuts for use in risk assessment. International Journal of Food Microbiology. 227:41-50. 68. Santillana Farakos, S. M., D. W. Schaffner, and J. F. Frank. 2014. Predicting survival of Salmonella in low-water activity foods: An analysis of literature data. Journal of Food Protection. 77:1448-1461. 69. Scallan, E., R. M. Hoekstra, F. J. Angulo, R. V. Tauxe, M.-A. Widdowson, S. L. Roy, J. L. Jones, and P. M. Griffin. 2011. Foodborne illness acquired in the United States—major pathogens. Emerg Infect Dis. 17. 70. Schleifer, K., and U. Fischer. 1982. Description of a new species of the genus Staphylococcus: Staphylococcus carnosus. International Journal of Systematic and Evolutionary Microbiology. 32:153-156. 71. Shah, M. K., G. Asa, J. Sherwood, K. Graber, and T. M. Bergholz. 2017. Efficacy of vacuum steam pasteurization for inactivation of Salmonella PT 30, Escherichia coli O157: H7 and Enterococcus faecium on low moisture foods. International Journal of Food Microbiology.

31

72. Sheth, A. N., M. Hoekstra, N. Patel, G. Ewald, C. Lord, C. Clarke, E. Villamil, K. Niksich, C. Bopp, T.-A. Nguyen, D. Zink, and M. Lynch. 2011. A national outbreak of Salmonella serotype Tennessee infections from contaminated peanut butter: A new food vehicle for Salmonellosis in the United States. Clinical Infectious Diseases. 53:356-362. 73. Simor, A. E., J. Goodfellow, L. Louie, and M. Louie. 2001. Evaluation of a new medium, oxacillin resistance screening agar base, for the detection of methicillin-resistant Staphylococcus aureus from clinical specimens. Journal of clinical microbiology. 39:3422-3422. 74. Singer, C. Date 2016. Hawaii's macadamia nut industry faces salmonella challenge [Type of Medium]. 75. Sommers, C., D. Geveke, and X. Fan. 2008. Inactivation of Listeria innocua on frankfurters that contain potassium lactate and sodium diacetate by flash pasteurization. Journal of food science. 73:M72-M74. 76. Stevenson, K., K. Ito, L. Beuchat, B. P. Marks, and D. Ashton. 2014. Guidelines for using Enterococcus faecium NRRL B-2354 as a surrogate microorganism in almond process validation. In, 2015 Almond Board of California

77. Systems, R. P. Date, 2012, Sterilization / Pasteurization of dried products. Available at: http://revtech-process-systems.com/index.php/en/steam-sterilization-pasteurization Accessed, January 2017. 78. Uesugi, A. R., M. D. Danyluk, and L. J. Harris. 2006. Survival of Salmonella Enteritidis Phage Type 30 on inoculated almonds stored at -20, 4, 23, and 35°C. Journal of Food Protection. 69:1851-1857. 79. USDA, E. Date, 2017, Fruit and Tree Nuts, Trade. Available at: https://www.ers.usda.gov/topics/crops/fruit-tree-nuts/trade/#TreeNuts Accessed March 2017. 80. Van Doren, J. M., D. Kleinmeier, T. S. Hammack, and A. Westerman. 2013. Prevalence, serotype diversity, and antimicrobial resistance of Salmonella in imported shipments of spice offered for entry to the United States, FY2007–FY2009. Food Microbiology. 34:239-251. 81. Vasan, A., R. Geier, S. C. Ingham, and B. H. Ingham. 2014. Thermal Tolerance of O157 and non-O157 Shiga Toxigenic Strains of Escherichia coli, Salmonella, and Potential Pathogen Surrogates, in Frankfurter Batter and Ground Beef of Varying Fat Levels. Journal of Food Protection. 77:1501-1501. 82. Waje, C. K., H.-K. Kim, K.-S. Kim, S. Todoriki, and J.-H. Kwon. 2008. Physicochemical and microbiological qualities of steamed and irradiated ground black pepper (Piper nigrum L.). Journal of Agricultural and Food Chemistry. 56:4592-4596. 83. Wall, M. M., and T. S. Gentry. 2007. Carbohydrate composition and color development during drying and roasting of macadamia nuts ( Macadamia integrifolia). LWT - Food Science and Technology. 40:587-593. 84. Williams, P., W. M. Leong, B. H. Ingham, and S. C. Ingham. 2010. Lethality of small-scale commercial dehydrator and smokehouse/oven drying processes against Escherichia coli O157:H7-, Salmonella spp.-, Listeria monocytogenes-, and Staphylococcus aureus-inoculated turkey jerky and the ability of a lactic acid bacterium to serve as a pathogen surrogate. In. 85. Woodroof, J. G. 1979. Tree nuts: production, processing, products. AVI Pub. Co, Westport, Conn. 86. Y. Sarig, H. B. 2000. Development Of A Method For Separating Macadamia Nut Kernels From Cracked Nuts. Applied Engineering in Agriculture. 16:665-670.

32

87. Zhao, X., J. Zhang, H. Ming, W. Zou, Y. Wang, C. Zhao, S. Mu, H. Zhang, and L. Qiu. 2016. Research on Pork Jerky Obtained Through Fermentation with Pediococcus acidilactici. p. 25. In, Polish Journal of Food and Nutrition Sciences, vol. 66.

33

CHAPTER 3:

Research Paper

Inactivation of Salmonella and surrogate bacteria on cashews and macadamia nuts exposed

to commercial propylene oxide processing conditions

Running title: Salmonella surrogates on nuts processed with PPO

Authors: Thomas Saunders, Jian Wu, Robert C. Williams, Haibo Huang and Monica A. Ponder

Department of Food Science & Technology

Virginia Polytechnic Institute and State University, Blacksburg, Va. 24061

Corresponding author:

Monica Ponder

401D Human and Agricultural Biosciences Building 1 (0924)

1230 Washington St

Virginia Tech

Blacksburg, VA 24061

Phone: 540-231-5031

Fax: 540-231-9293

Email: mponder@vt.edu

34

Keywords: Salmonella, nuts, cashews, propylene oxide, low water activity, macadamia nuts,

surrogates, Enterococcus faecium

35

Abstract:

Propylene oxide (PPO) is a chemical fumigant used to improve the microbial quality of

tree nuts. PPO has been validated to reduce Salmonella on bulk almonds but has not been

evaluated for other tree nuts. The identification of a non-pathogenic microorganism that may be

used to validate the inactivation of Salmonella on cashews and macadamia nuts is needed to

avoid introducing the pathogen to the pasteurization facility, while assuring the different

packaging configurations and processing parameters are effective. The objective of this

research was to compare the reduction of Salmonella spp. and 3 potential surrogate bacteria,

Enterococcus faecium, Pediococcus acidilactici, or Staphylococcus carnosus, on cashews and

macadamia nuts processed using PPO. Whole cashews and macadamia nuts were co-inoculated

with a cocktail of Salmonella spp. and one of three potential surrogates using TSA-grown cells,

and the nuts were dried to original aw around 0.44-0.51. Nuts were packaged in polywoven bags

(2.3 kg) and shipped for PPO treatment at a commercial facility. Samples were returned for

enumeration by plating onto appropriate selective media and TSA overlaid with XLT-4.

Processing was completed three times, with six samples per process. The mean log reductions

(CFU/g) of Salmonella and each potential surrogate, within a sample and amongst all trials, were

compared using a paired T-test. P<0.05 is considered significant. Salmonella and the potential

surrogates were recovered by vigorous shaking, serial diluted, and plated onto TSA with an

overlay of XLT-4 for Salmonella, or media selective for the potential surrogates. Mean

reduction in log CFU/g of Salmonella (7.3 ± 0.19) was significantly greater than that of E.

faecium (6.4± 0.31), or P. acidilactici (6.3± 0.33) on whole macadamia nuts. Reduction of

Salmonella was significantly greater than E. faecium and P. acidilactici reduction on cashews..

Salmonella reduction did not exceed that of St. carnosus and therefore is not a good surrogate. E.

36

faecium and P. acidilactici may be considered as surrogates for Salmonella reduction on whole

macadamia nuts and whole cashews processed using PPO.

37

Introduction

Salmonella has been responsible for many illnesses and recalls associated with low water

activity (low-aw) foods in the United States (10). While water activities below 0.98 do not

support the growth of Salmonella, it has been known to survive for periods up to 2 years (10).

Salmonellosis outbreaks have been traced back to contaminated almonds, cashews, hazelnuts,

pine nuts and pistachios (13). Additionally, Salmonella detection has prompted the recall of

many other types of tree nuts and tree nut containing products including walnuts, macadamia

nuts, pecans, cashews, almonds, pistachios, pine nuts and hazelnuts (22).

Tree nuts are a common snack food and maybe considered a ready to eat. Tree nuts are

not only ready to eat, but are known ingredients in many other food products sold and consumed

in the United States (17, 26). Though there is a high consumption of tree nuts and their products

in the United States, many varieties of nuts must be imported from other countries, including

every continent excluding Antarctica (26). While the United States is a lead exporter in almonds,

walnuts, pistachios and macadamia nuts, they are a lead importer of cashews, pecans, and Brazil

nuts (26). When producing as well as importing a product such as nuts, many considerations

and steps must be taken to ensure that it is safe for human consumption. Importers and

processors of nuts, must comply with the FDA Food Safety Modernization Act (FSMA) final

rule for Preventative Controls for Human Food (12). This is not limited to but includes a focus

on hazard analysis, preventive controls, and over all good manufacturing practices (CGMPs),

not only through the handling and processing of the product in local plants, but also the previous

practices already performed by the exporting countries or local farms supplying the distributor

(12).

38

Considering tree nuts are raw agricultural commodities it is important to validate post-

processing interventions which reduce the risk of Salmonella contamination in products. One

way the producer lessens risk of contamination is through fumigation processing such propylene

oxide. Propylene oxide is used for the reduction of bacteria as well as yeasts and molds. The

U.S. Environmental Protection Agency (EPA) registration for PPO as a pesticide fumigant

stipulates that (i) exposure temperature must not exceed 51.68 ͦC, (ii) exposure time shall be no

more than 4 h, and (iii) PPO residue in the product (prior to release) shall not exceed 300 ppm.

Label instructions additionally dictate that (i) dosage cannot exceed 2.5 kg/m3 and (ii) at the end

of treatment there should be no less than four aeration cycles with a PPO chamber volume of air

(EPA registration no. 47870-1, Aberco, Seabrook, Md.) (9). Commercial processing facilities, in

fact, use a variety of concentrations of PPO, chamber temperature, and relative humidity of

environment to meet the quality expectations of their customers. In addition, different

packaging configurations may affect the penetration of PPO into the sample.

Each process with its set parameters must be validated as a product specific process,

which identifies the need for viable surrogate organisms to avoid the introduction of Salmonella

in the processing facility. The surrogate should be non-pathogenic, and have similar stability

and inactivation characteristics to the target pathogen under the decontamination process of

interest. Enterococcus faecium NRRL B2354 (ATCC #8459) has been validated as a surrogate

for Salmonella spp. in several thermal lethality studies (1, 16). While E. faecium NRRL B2354

(ATCC #8459) has been validated as an appropriate surrogate for Salmonella with regard to

thermal processing of almonds, juice and jerky, there are cases where the inactivation

characteristics were not as comparable to Salmonella, for instance when used in an inactivation

study of peanuts exposed to stagnant heat (5). In the case of dried pet food through extrusion

39

processing and ground, formed beef jerky processed with thermal methods, Pediococcus

acidilactici (ATCC #8042), has been demonstrated to be more heat resistant than Salmonella,

therefore it may serve as a conservative surrogate for Salmonella in these products (2, 7).

Staphylococcus carnosus (ATCC #51365) has been proposed as a surrogate for Salmonella in

heat processed ground meat products and frankfurter batter (27).

The objective of this study was to compare the survival of Salmonella spp. with

Enterococcus faecium NRRL B2354 (ATCC #8459), Pediococcus acidilactici (ATCC #8042),

and Staphylococcus carnosus (ATCC #51365) on whole cashews and macadamia nuts subjected

to commercial propylene oxide fumigation in accordance with the US EPA label standards.

Bacteria with comparable or greater survival compared to Salmonella may be considered as a

surrogate on cashews and macadamia nuts processed by PPO, provided that their process is

validated to achieve comparable temperatures and gas concentrations within the package.

Materials and Methods

Bacterial strains and growth conditions:

Inoculum preparation:

Five Salmonella enterica serovars obtained from low-aw foods (4 strains including

Montevideo, an isolate from 2010 peppercorn associated outbreak, Tennessee, K4643 human

isolate from 2007 peanut butter associated outbreak, Ball ARL-SE-085 isolated from black

pepper in 2011, and Johannesburg ARL-SE-013 from dried ginger in 2010) and Salmonella

Senftenberg 775W, ATCC 43845, frequently used for thermal studies due to increased heat

resistance, were obtained from the FDA or American Type Culture Collection (ATCC,

40

Manassas, VA). Three potential surrogate bacteria Enterococcus faecium (NRRL B-2354,

ATCC strain #8459), Pediococcus acidilactici (ATCC #8042) and Staphylococcus carnosus,

subspecies carnosus (ATCC #51365), were obtained from the ATCC.

Wet inoculation with Salmonella strains and Surrogates grown on TSA:

Inoculation preparation was adapted from the method developed for almonds and is

associated with increased stability of Salmonella on whole peppercorns and cumin seeds

compared to a broth grown method (3, 4). Briefly, each strain was spread onto a 150 x 15 mm

(BD Falcon, Franklin Lakes, NJ) large Tryptic Soy Agar (TSA) (Becton Dickinson, Franklin

Lakes, NJ) plates and incubated at 37 °C for 24 h. Cells were harvested from each of the plates

and placed in separate sterile conical tubes by first applying 0.1% (w/v) peptone-tween (PT)

buffer to each plate and scraping colonies into suspension using a sterile cotton swab. For

Salmonella, the suspensions of 5 strains were combined in comparable volumes corresponding to

similar CFU of each strain as determined previously (6). Equal volumes of the Salmonella

cocktail (12.5 ml) and one of the three potential surrogates (12.5 ml) were mixed in a total

volume of 25 ml. The suspensions were then transferred into 27 oz. sterile Whirl-Pak bags

containing either 400 g of dry whole “Fancy grade” macadamia nuts or whole cashews and a

mixture of shaking, inversion and hand massaged for 1 minute to evenly coat the nuts. After

inoculation, nuts were arranged in a single layer on double layered, sterilized filter paper on top

of drying racks placed inside a biological safety cabinet for drying. Nuts were dried for 24-48 h

to allow the nuts to return to the pre-inoculation aw of around 0.44 for macadamia nuts and aw of

0.51 for cashews at room temperature. Water activity of whole macadamia nuts (5 g) and

41

cashews (4 g) was determined by using an AquaLab 4TE water activity meter (AquaLab,

Pullman, WA).

Heat resistance:

Resistance to heat treatment was examined by heating inoculated cashews and

macadamia nuts (25 g) to 280 ͦ F (138°C) in a lab scale convection oven for 15 minutes, as

referenced by the Almond Board of California.

Packing and Shipping methods:

The dried, inoculated nuts (50g cashew or 50g macadamia) were individually packaged

within muslin drawstring sachets (4 x 6 in.) that were tied to a framework of non-conducting

thin CPVC pipe (12.7 mm diameter pipe, assembly dimensions 38.8 cm x 26.7 cm). This frame

prevented the movement of the sachet throughout a larger bag poly-woven polypropylene bag

(14 x 24 in.), when filled with ~3000g of non-inoculated nuts to achieve a bag thickness of 7cm

depth when placed horizontally. Each bag contained 6 sachets of the specific nut inoculated

with Salmonella and either E. faecium NRRL B2354 (ATCC #8459), P. acidilactici (ATCC

#8042), or Staphylococcus carnosus (ATCC #51365). Each potential surrogate/Salmonella

combination of inoculated nuts were placed within separate poly-woven bags (3 bags per nut

type, 6 bags in total). The bags were sealed by a combination of high strength glue and

industrial staples to prevent any potential cross contamination and the bags were placed in

insulated boxes. The boxes were shipped overnight to a commercial processor for PPO

treatment with instructions on how to prepare the bags to assure uniform and consistent

distribution of the nuts before treatment: to flip each bag gently three times and to lay the bag

flat on the surface for treatment after removal from the shipping packaging. The bags were to be

laid face up (signified by a written label), and the processor included matchstick data loggers

42

with in the PPO run to verify temperatures during processing were reached and remained

consistent through each of the three runs.

PPO Treatments:

Propylene oxide processing was performed by Cosmed Group, Inc. (Cosmed) in

accordance with EPA regulations (9), according to the parameters provided in Table 1 , which

were recorded entirely by processors at Cosmed. These dwell times are much shorter than their

standard process because in a preliminary trial the Salmonella and Enterococcus faecium

NRRL B2354 (ATCC #8459) could not be recovered from these small test bags. A platform

was constructed of stacked pallets and centrally located within a 12-pallet chamber. The six

bags were placed on top of the pallets and temperature data loggers were placed in various

locations around the bags of samples. The average temperature during each of the runs is

reported in figure 1. Dedicated runs were performed for these treatments to minimize the

potential for cross contamination within the facility. Three separate processing runs were

performed on different days. After processing the bags were returned to the insulated shipping

container and shipped back to Virginia Tech for enumeration.

Enumeration:

Salmonella and the corresponding surrogate bacteria were enumerated from nuts at four

time periods: after drying, after heat treatment, after PPO treatment and after storage of non-

treated nuts. The enumeration method was adapted from the method described by Stevenson et

al. for enumerating E. faecium NRRL B2354 (ATCC #8459) from almonds (25). Nuts from

each sachet were mixed with Butterfield’s phosphate buffer (BPB) in a 1:1 ratio in a sterile

sample bag (25 ml to 25 g of treated nuts) and shaken 50 times with vigorous pace in an arc

motion of 1ft, using 27 oz. steril Whirl-pak bags. After a 5 min. rest period the nuts were shaken

43

again 5 times before pouring the liquid into a sterile tube. Tenfold serial dilutions in BPB were

performed. Each dilution was spread plated onto TSA plates in duplicate, incubated at 35 ͦ C for

3h. followed by overlay with 7 ml of XLT4 for recovery of injured Salmonella. Then the plates

were incubated for the remainder of 48h at 35 ͦC. Each dilution was also plated in duplicate onto

the appropriate selective media required to recover the given surrogate strain (MSA for

Staphylococcus carnosus (ATCC #51365) and BEA for P. acidilactici (ATCC #8042) and E.

faecium NRRL B2354 (ATCC #8459)). Plates were incubated at 37°C for 48 h. An aliquot of 1

mL of the supernatant was also transferred to 9 mL of Tryptic soy broth and incubated at 37°C

for 24 h and subsequently streaked onto selective media resulting in a limit of detection of 1.0

log CFU/g.

Experimental Design and Statistical Analysis:

Three separate PPO treatments were performed on three separate days. On each day six

sachets containing a combination of Salmonella and one of the three tested surrogates were

placed within a larger bag. A separate bag was processed for each surrogate/Salmonella pair and

nut type. There was a total of six bags processed each day (3 cashew, 3 macadamia nuts).

Bacterial counts were log transformed prior to statistical analysis. The reduction in log

CFU/g was calculated by subtracting the log CFU/g after PPO processing from the log CFU/g of

unprocessed nuts that were inoculated on the same day as the PPO processed nuts and

enumerated on the day of receipt of the PPO processed nuts to account for any differences. The

unprocessed nuts were stored in a bench top desiccator. The log CFU/g reduction of Salmonella

was compared to the log CFU/g reduction of the corresponding potential surrogate organism

from each sachet, for each nut type using a matched pair-wise t-test. The mean log CFU/g

reduction of Salmonella was compared to the mean log CFU/g reduction of the corresponding

44

potential surrogate organism; each nut type was also compared using a one-way ANOVA using

JMP 13. P < 0.05 were considered significant.

Results

Comparison of reduction of Salmonella (log CFU/g) on cashews and macadamia nuts treated

with PPO to potential surrogate bacteria:

While propylene oxide fumigation significantly reduced the mean populations of

Salmonella and each of the tested potential surrogate bacteria on whole macadamia nuts and

whole cashews (p<0.05), the amount of reduction varied based on type of nut and on bacterial

strain (Table 2). The reductions in Salmonella were compared to E. faecium NRRL B2354

(ATCC #8459) (Fig. 2), P. acidilactici (ATCC #8042) (Fig. 3), St. carnosus (ATCC #51365)

(Fig. 4). It should be noted that only a 1-log reduction in mean Salmonella log CFU/g occurred

during control heat treatment test, demonstrating the majority of reduction is due to PPO

fumigation. E. faecium NRRL B2354 (ATCC #8459) shows significance in inactivation

compared to Salmonella on macadamia nuts and almost identical inactivation rates on cashews,

no samples of E. faecium NRRL B2354 (ATCC #8459) had more reduction log CFU/g than

Salmonella. Pediococcus acidilactici ATCC 8042 had significantly less reduction on cashews

and macadamia nuts compared to Salmonella in all samples. Mean reduction of Staphylococcus

carnosus ATCC 51365 was not statistically different than Salmonella, though there were several

instances where the reduction of St. carnosus exceeded that of Salmonella spp (Fig 4).

No colonies with characteristic black centers indicative of Salmonella appeared on XLT4

plates from non-inoculated nuts. No dark brown/black colonies indicative of E. faecium NRRL

45

B2354 (ATCC #8459) or P. acidilactici (ATCC #8042) appeared on BEA plates from non-

inoculated nuts. No small, white and circular colonies indicative of St. carnosus (ATCC

#51365) growth appeared on MSA.

Table 2. Mean reduction log CFU/g of Salmonella compared to three potential surrogate bacteria co-inoculated onto macadamia nuts or cashews after propylene oxide fumigation. Mean reduction* log CFU/g ± std error of mean (n=18) Macadamia nuts Cashews Notes: Salmonella 7.3 ± 0.19 A 5.4± 0.15 a No samples where

reduction log CFU/g of E. faecium was larger than Salmonella.

Enterococcus faecium NRRL B2354

6.4± 0.31 B 5.2± 0.25 a

Mean difference § 0.93± 0.12 0.60± 0.09 Mean difference Matched pairs T-test P-Value

0.0009 0.0053

Salmonella 7.8± 0.22 A 4.9± 0.22 a Reductions in log

CFU/g Salmonella were significantly larger than Pediococcus in all samples.

Pediococcus acidilactici ATCC 8042

6.3± 0.33 B 4.1± 0.25 b

Mean difference § -1.57± 0.13 -0.83± 0.18 Mean difference Matched pairs T-test P-Value

<0.0001 <0.0036

Salmonella 8.0± 0.19A 4.5± 0.15a While not

statistically different there were several occurrences when the reduction of St. carnosus was larger than Salmonella.

Staphylococcus carnosus ATCC 51365

8.0± 0.36A 5.2± 0.25a

Mean difference § 0.07± 0.12 0.69± 0.22 Mean difference Matched pairs T-test P-Value

0.6191 0.0607

46

*Mean reduction (log CFU/g) reduction is calculated for each co-inoculated surrogate and

Salmonella nut. Each combination was placed in separate bags. Standard error of the means

reflects differences in inactivation from three independent trials (process days). Variability

between Salmonella reported on the table is due to location with processing chamber. Different

letters within surrogate and Salmonella pairs (superscript letters denote significance in results,

when written as A,B or a,b. Upper case lettering is used in Macadamia nut results while lower

case is used in Cashew results) indicates statistical significance at p <0.05 by ANOVA

comparison of Means with Tukeys Adjustment as well as paired comparison as shown in the

following figures. Superscript letters denote significance in results, when written as A,B or a,b.

Upper case lettering is used in Macadamia nut results while lower case is used in Cashew

results.

§ Mean difference is the mean of the difference between the reduction of Salmonella (log

CFU/g) and each co-inoculated surrogate (log CFU/g) packaged within the same sachet and

averaged across each of the three trials. Standard error of the means reflects differences in

inactivation from three independent trials (process days). Output of the matched pairs test of

JMP 13.

47

Figure 2. Comparison in log CFU/g of Salmonella and Entercoccus faecium NRRL B2354 (ATCC #8459) of inoculated nuts processed using PPO on three separate processing days. Each point represents an individual sachet (n=18) processed during the three trials.

48

Figure 3. Comparison in log CFU/g of Salmonella and Pediococcus acidilactici (ATCC #8042) of inoculated nuts processed using PPO on three separate processing days. Each point represents an individual sachet (n=18) processed during the three trials.

49

Figure 4. Comparison in log CFU/g of Salmonella and Staphylococcus carnosus (ATCC #51365) of inoculated nuts processed using PPO on three separate processing days. Each point represents an individual sachet processed during the three trials. (Note: There are only 9 points marked for Macadamia nuts due to inactivation falling below detectable limits.)

.

50

Discussion

The U.S. Environmental Protection Agency (EPA) registration for PPO as a pesticide

fumigant carries many stipulations, and it is necessary to validate that these conditions result in

the inactivation of the target pathogen under the many different packaging configurations and

specific processes for cashews and macadamia nuts that comply with regulation (9). With

Salmonella having been seen to survive on tree nuts extended lengths of time (24), and with

most sources of contamination believed to be through cross-contamination during harvest and

preparation steps leading up to fumigation processes, it is apparent that the use of PPO

fumigation can be justified (11, 21, 23). It is estimated that 64.8 million pounds of products are

annually treated with PPO, with 1.8% total treated crops being nut meat (14). The potential to

have a viable surrogate organism to compare with sufficient inactivation to Salmonella in this

process is a necessity of all processing facilities to deliver a safe product to the consumer.

It should also be noted that potential surrogate bacteria were chosen based on previous

work as surrogates of Salmonella in food systems. E. faecium NRRL B2354 (ATCC #8459) has

been examined as a suitable surrogate in several thermal and non-thermal lethality studies in low

moisture foods. In an extrusion process, results have shown E. faecium NRRL B2354 (ATCC

#8459) requires a temperature of 73.7 ºC to reach a 5-log reduction while Salmonella spp. were

reduced by 5-logs at 60.6 ºC (1). Another lactic acid bacterium, P. acidilactici (ATCC #8042),

has shown similar inactivation compared to Salmonella inactivation in extruded pet food and

ground meat formed jerky products (2, 7). With thermal studies showing promise by using lactic

acid bacteria as surrogates for Salmonella, further inactivation studies proceeded in other

processes.

51

Inoculation preparation was adapted from guidelines used by the Almond Board of

California for processing large batch almonds with PPO (6), however after preliminary trials, we

discovered the need to use a co-inoculation process for proper comparison between inactivation

of Salmonella and potential surrogate bacteria was necessary. Since PPO has been commonly

used to treat commodities like tree nuts for inactivation of potentially pathogenic bacteria and to

increase quality, there is a need to validate this process (8). Populations of Salmonella showed >

5-log CFU/g reduction when initial counts around 8-logs CFU/g compared with counts recorded

5 days post-PPO processing (8). Showing that PPO is a valid process in reducing Salmonella

spp, working on validating surrogate bacteria with that process, included choosing bacteria that

showed to act similar to Salmonella inactivation in other treatment methods. By performing a co-

inoculation, initial log (CFU/g) counts on each replicate of the sample were detected around 8-

logs CFU/g of both Salmonella and potential surrogate. This enable the results to account for

differences in inactivation within a single bag of processed commodity. A difference was noted

through pre-experimentation trials. E. faecium NRRL B2354 (ATCC #8459), P. acidilactici

(ATCC #8042), and Staphylococcus carnosus (ATCC #51365) have all shown this similar

inactivation and provided the research for valid reasons of testing in PPO processing (2, 7, 16,

28, 29).

The need to validate a comparison between Salmonella inactivation and potential

surrogate inactivation for each type of food being processed, is made apparent in this study with

Staphylococcus carnosus (ATCC #51365). Staphylococcus carnosus (ATCC #51365)

inactivation prevents it from being deemed a surrogate on whole cashews and macadamia nuts

processed with PPO, despite inactivation in frankfurter batter compared to Salmonella, also in a

thermal process, where Staphylococcus carnosus (ATCC #51365) is a comparable surrogate

52

(28). This provides that temperature alone is not the only indicator in inactivation of pathogenic

bacteria. Some factors that may contribute to varying results of surrogate’s capability and

Salmonella inactivation include the shape and composition of different food systems as well as

how that food itself reacts in processing conditions (18). Works show that different physical

properties of the pistachio nut (shell) and kernel (nut meat) require different processing in order

to maintain quality and safety, along with improving yield of product (18). This further increases

the importance that validation studies must be developed for a specific product, with specific

processing methods and conditions. While the work with pistachios did not involve PPO, PPO

processing efficiency can be altered due to which commodity is being processed and size of

product, also including factors such as chamber size and containers which the product was

processed (15). With changing variables, the validation of surrogate organisms are only valid

with in the same parameters as they were originally validated in.

These experiments were not designed to demonstrate the effectiveness of PPO for

inactivation of Salmonella in a standard packaging configuration or when using a typical

production load. Most chambers are designed to accommodate ~12 pallets, a truck load, or

multiple truck loads at a time. In relation to “pallets” when referring to tree nuts, nuts can be

processed and contained in boxes, bins, totes, as well as sacks typically weighing in at a total of

~2,000 lbs. In this test situation, the volume of nuts, placement and packaging material was

chosen to assure penetration of steam and PPO, for lab scale results. Cashews are routinely

processed in 40 lb cardboard boxes that may contain a thermal barrier, therefore it cannot be

assumed that the same conditions would produce the same inactivation of Salmonella within the

package.

53

E. faecium NRRL B2354 (ATCC #8459) and P. acidilactici (ATCC #8042) can be

considered surrogate bacteria, this provides useful in-plant surrogates for an industrial scale PPO

nut processing. Thermal and gas distribution need to be well characterized with in the chambers

to assure loads placed in each location receive comparable heat and fumigant. If there is

variability within a chamber, then the inoculated samples should be in the coldest area of the

chamber and/or where concentration of the fumigant PPO is lower. When placing inoculated

nuts in a sample intended for processing, they should be placed within the center of a standard

container. If macadamia nuts are typically processed in a 50lb corrugated cardboard box, the

sachet would need to be placed in the area which would receive the least amount of gas and heat

exposure.

Heat and gas penetration within the packaging the nuts are contained within should be

understood before any validation trials move further. For this study, poly-woven sand bags were

used because of their classification as permeable, as required for fumigant PPO treatments.

These bags work for a lab scale experiment, but are not large enough to be in an industrial scale

process. It is therefore necessary to use large enough containers to efficiently hold product, but

must also meet permeability requirements for PPO fumigant processing (14). Permeability was a

main focus of packaging of medical equipment being sterilized using ethylene oxide (ETO), a

fumigant used in sanitation process and treatment of spices that acts similar to PPO (19, 20).

These factors are not the only that should be considered but do require further investigation to be

in compliance with FSMA Final Rule for Preventative Controls for Human Food (12).

Having a validated surrogate for both cashews and macadamia nuts would serve the

population that wants to continue to use PPO fumigation as a final lethality step in processing, by

verifying that it is a necessary step in consumer safety.

54

Conclusions:

Enterococcus faecium NRRL B2354 (ATCC #8459) and Pediococcus acidilactici

(ATCC #8042) may be considered as surrogates for Salmonella when inoculated onto whole

macadamia nuts and whole cashews processed using propylene oxide fumigation at the

described temperature. Pediococcus acidilactici (ATCC #8042) would be the best choice for

cashews because there was a statistically significant increase in survival compared to

Salmonella. This would provide for more assurance that the process achieves the desired log

reduction of Salmonella. Staphylococcus carnosus (ATCC #51365) should not be used as a

surrogate for Salmonella on either type of nut because the Salmonella was more resistant to the

described PPO process compared to Staphylococcus carnosus (ATCC #51365).

55

Acknowledgements

Funding for this research was provided in part by Cosmed Group, Inc. (Cosmed) and by the

Virginia Agricultural Experiment Station and the Hatch Program of the National Institute of

Food and Agriculture, U.S. Department of Agriculture. MP, RCW and JW designed the study.

Processing conditions were defined in consultation with Cosmed personnel. All PPO processing

was performed by Cosmed staff. Inoculation, enumeration and analysis were performed at

Virginia Tech. Data collection and analysis were performed by TS with assistance from JW and

MP. The manuscript was written by TS, JW, MP, RCW and HH. We would like to thank the US

FDA ORA Arkansas Regional Lab for providing the spice-isolated strains. Special thanks to Kim

Waterman and Kendall Fogler for providing technical assistance.

56

References

1. Bianchini, A., J. Stratton, S. Weier, T. Hartter, B. Plattner, G. Rokey, G. Hertzel, L. Gompa, B. Martinez, and K. M. Eskridge. 2014. Use of Enterococcus faecium as a Surrogate for Salmonella enterica during Extrusion of a Balanced Carbohydrate-Protein Meal. Journal of Food Protection. 77:75-75. 2. Borowski, A. G., S. C. Ingham, and B. H. Ingham. 2009. Validation of ground-and-formed beef jerky processes using commercial lactic acid bacteria starter cultures as pathogen surrogates. Journal of Food Protection. 72:1234-1247. 3. Bowman, L. S. 2015. Impacts of inoculation strategy on survival of Salmonella enterica and surrogate Enterococcus faecium at low water activity on dry peppercorn and cumin seeds. In, Food Science and Technology, vol. Master of Science in Life Science. Virginia Tech, Blacksburg, VA. 4. Bowman, L. S., K. M. Waterman, R. C. Williams, and M. A. Ponder. 2015. Inoculation Preparation Affects Survival of Salmonella enterica on Whole Black Peppercorns and Cumin Seeds Stored at Low Water Activity. Journal of Food Protection. 78:1259-1259. 5. Brar, P. 2014. Determining Validity of Enterococcus Faecium as Surrogate for Salmonella under Stagnant Dry Heating of Peanuts. In, 2014 Annual Meeting Iafp. 6. California, A. B. o. October 2008. Guidelines for Validation of Propylene Oxide Pasteurization. 7. Ceylan, E., and D. A. Bautista. 2015. Evaluating Pediococcus acidilactici and Enterococcus faecium NRRL B-2354 as Thermal Surrogate Microorganisms for Salmonella for In-Plant Validation Studies of Low-Moisture Pet Food Products. Journal of Food Protection. 78:934-934. 8. 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. Journal of Food Protection. 68:1613-1622. 9. EPA. 2004. Propylene oxide tolerances for residues. Code of Federal Regulations title 49, chapter 1, section 180.493. 10. Farakos, S. M. S., D. W. Schaffner, and J. F. Frank. 2014. Predicting Survival of Salmonella in Low-Water Activity Foods: An Analysis of Literature Data. Journal of Food Protection. 77:1448-1448. 11. Finn, S., J. C. D. Hinton, P. McClure, A. Amezquita, M. Martins, and S. Fanning. 2013. Phenotypic Characterization of Salmonella Isolated from Food Production Environments Associated with Low-Water Activity Foods. Journal of Food Protection. 76:1488-1488. 12. GPO. Date, April 6, 2017, FSMA Final Rule for Preventive Controls for Human Food. Available at: https://www.ecfr.gov/cgi-bin/text-idx?SID=01afb009a9de6f6797d02757e59ef157&mc=true&node=pt21.2.117&rgn=div5tsh. Accessed, April 2017. 13. Harris, L. J., M. Palumbo, L. R. Beuchat, and M. D. Danyluk. 2017. Outbreaks of foodborne illness associated with the consumption of tree nuts, peanuts, and sesame seeds [Table and references]. Outbreaks from tree nuts, peanuts, and sesame seeds. 14. Howe, D. 2017. Cosmed Group. In M. Ponder (ed.). 15. Hurst, W., and L. Harris. 2013. Integrating Hazard Analysis Critical Control Point (HACCP) and Statistical Process Control (SPC) for safer nut processing. Improving the safety and quality of nuts. 15:119-147.

57

16. Jeong, S., B. P. Marks, and E. T. Ryser. 2011. Quantifying the Performance of Pediococcus sp. (NRRL B-2354: Enterococcus faecium) as a Nonpathogenic Surrogate for Salmonella Enteritidis PT30 during Moist-Air Convection Heating of Almonds. Journal of Food Protection. 74:603-603. 17. Johnson, D. C. 1997. United States is world leader in tree nut production and trade. USDA-ERS Fruit and Tree Nuts Situation and Outlook. FTS-280. 18. Kashaninejad, M., A. Mortazavi, A. Safekordi, and L. Tabil. 2006. Some physical properties of Pistachio (Pistacia vera L.) nut and its kernel. Journal of Food Engineering. 72:30-38. 19. Labrecque, R., S. Conroy, K. Faucher, T. Swanick, P. Martakos, T. Karwoski, S. Herweck, and T. Carlton. 2006. Packaging and sterilization of medical devices. In Google Patents. 20. Mendes, G. C., T. R. Brandao, and C. L. Silva. 2007. Ethylene oxide sterilization of medical devices: a review. American journal of infection control. 35:574-581. 21. Motarjemi, Y., G. Moy, and E. C. D. Todd. 2014. Encyclopedia of food safety. Academic Press Inc, Boston. 22. Palumbo, M., L. R. Beuchat, M. D. Danyluk, and L. J. Harris. 2016. Recalls of tree nuts and peanuts in the U.S., 2001 to present [table and references] U.S. recalls of nuts. 23. Podolak, R., E. Enache, W. Stone, D. G. Black, and P. H. Elliott. 2010. Sources and risk factors for contamination, survival, persistence, and heat resistance of Salmonella in low-moisture foods. J Food Prot. 73:1919-36. 24. Santillana Farakos, S. M., R. Pouillot, N. Anderson, R. Johnson, I. Son, and J. Van Doren. 2016. Modeling the survival kinetics of Salmonella in tree nuts for use in risk assessment. International Journal of Food Microbiology. 227:41-50. 25. Stevenson, K., K. Ito, L. Beuchat, B. P. Marks, and D. Ashton. 2014. Guidelines for using Enterococcus faecium NRRL B-2354 as a surrogate microorganism in almond process validation. In, 2015 Almond Board of California

26. USDA, E. Date, 2017, Fruit and Tree Nuts, Trade. Available at: https://www.ers.usda.gov/topics/crops/fruit-tree-nuts/trade/#TreeNuts Accessed March 2017. 27. Vasan, A., R. Geier, S. C. Ingham, and B. H. Ingham. 2014. Thermal tolerance of O157 and non-O157 shiga toxigenic strains of Escherichia coli, Salmonella, and potential pathogen surrogates, in frankfurter batter and ground beef of varying fat levels. Journal of Food Protection. 77:1501-1511. 28. Vasan, A., R. Geier, S. C. Ingham, and B. H. Ingham. 2014. Thermal Tolerance of O157 and non-O157 Shiga Toxigenic Strains of Escherichia coli, Salmonella, and Potential Pathogen Surrogates, in Frankfurter Batter and Ground Beef of Varying Fat Levels. Journal of Food Protection. 77:1501-1501. 29. Zhao, X., J. Zhang, H. Ming, W. Zou, Y. Wang, C. Zhao, S. Mu, H. Zhang, and L. Qiu. 2016. Research on Pork Jerky Obtained Through Fermentation with Pediococcus acidilactici. p. 25. In, Polish Journal of Food and Nutrition Sciences, vol. 66.

58

Table 1. Propylene oxide processing parameters

Phase Parameter Set point Minimum Maximum Observed

Start Cycle

Jacket Temperature ( ͦC)

25.56 21.1 51.67 45.56

Product Temperature ( ͦC)

N/A 15.56 N/A 23.89

Phase Parameter Set point Minimum Maximum Observed

Vacuum Pressure (PSIA) 0.5 0.5 1.8 0.5

Nitrogen Pressure (PSIA) 9.5 9.0 11.0 10.0

Vacuum Pressure (PSIA) 0.5 0.25 1.0 0.5

Steam Inject Pressure (PSIA) 1.9 1.5 2.5 2.0

Number of Pulses 1 1 2 1

Vacuum Pressure (PSIA) 1.0 0.75 1.0 1.0

PPO Inject

Pressure (PSIA) 3.0 1.0 6.0 6.0

Weight (LBS) Chamber 3

49 49 99 -

Product Temperature ( ͦC)

N/A 23.89 51.67 43.33

Nitrogen Pressure (PSIA) 11.0 9.0 13.0 11.0

Chamber Temperature ( ͦC)

48.89 37.78 51.67 48.89

PPO Dwell Time (hh:mm:ss) 2:30:00 2:00:00 2:30:00 2:30:00

Chamber Temperature ( ͦC)

48.89 37.78 51.67 51.66

Vacuum Pressure (PSIA) 0.5 0.3 1.0 0.5

Nitrogen Pressure (PSIA) 5.0 4.0 5.5 5.0 Iterations Number of pulses 11 10 11 11

Air Inbleed Pressure (PSIA) Atmospheric 14.0 15.0 15.0

*All measurements were monitored and supplied to the researcher by Cosmed groups

59

Figure 1. Average temperature (°C, ± standard error) of chamber during PPO processing

of cashews and macadamia nuts. Temperatures were recorded every 5 minutes using 6 data

loggers placed on top of the processed bags and underneath the bags.

0

10

20

30

40

50

60

70

1 3 5 7 9 11 13 15 17 1921 23 25 27 29 3133 35 37 39 41 43 4547 49 51 53 55 57 5961 63 65 6769 71 7375 77 79 81 83 8587

Tem

pera

ture

ºC

Time (minutes)

Average Temperature (ºC) per each Trial

Trial 1 Trial 2 Trial 3

60

CHAPTER 4:

Research Paper

Evaluating Pediococcus acidilactici and Enterococcus faecium NRRL B-2354 as Surrogates for Salmonella enterica on Saturated Steam Processed Cashews and Macadamia Nuts

Running title: Salmonella surrogates on nuts processed with saturated steam

Authors: Thomas Saunders, Jian Wu, Robert C. Williams, Haibo Huang and Monica A. Ponder

Department of Food Science & Technology

Virginia Polytechnic Institute and State University, Blacksburg, Va. 24061

Corresponding author:

Monica Ponder

401D Human and Agricultural Biosciences Building 1 (0924)

1230 Washington St

Virginia Tech

Blacksburg, VA 24061

Phone: 540-231-5031

Fax: 540-231-9293

Email: mponder@vt.edu

61

Keywords: Salmonella, nuts, cashews, saturated steam (SS), low water activity, macadamia nuts,

surrogates, Enterococcus faecium

62

Abstract

Enterococcus faecium, Pediococcus acidilactici, and Staphylococcus carnosus were

investigated as surrogates for Salmonella spp. on cashews and macadamia nuts processed using

vacuum assisted steam. Whole cashews and macadamia nuts were co-inoculated with a cocktail

of Salmonella enterica and one of three potential surrogates using tryptic soy agar (TSA) grown

cells, nuts were then dried to original aw around 0.44-0.51. Nuts were packaged in polywoven

bags (2.3 kg) and commercially processed using a vacuum assisted steam at 80 ͦC. Salmonella

and the potential surrogates were recovered from the nuts by vigorous shaking, enumerated by

serial dilution, followed by plating onto TSA with an overlay of XLT-4 for Salmonella, or media

selective for the potential surrogates. Mean log reductions (CFU/g) of Salmonella and each

potential surrogate, within a bag and amongst all trials, were compared using a paired T-test.

Mean reduction (log CFU/g) of Salmonella (6.0 ± 0.14) was significantly larger than that of E.

faecium (4.3± 0.12) (P< 0.0001), or P. acidilactici (3.7± 0.14) (P< 0.0001) on whole cashews.

Mean reduction (log CFU/g) of Salmonella (5.9 ± 0.18) was significantly larger than P.

acidilactici (4.4± 0.18) (P< 0.0001) on whole macadamia nuts. Salmonella and E. faecium

reduction on macadamia nuts were comparable, but instances occurred where Salmonella

reduction was less than E. faecium. Stapylococcus carnosus reduction exceeded that of

Salmonella and is therefore not a comparable surrogate. This data suggests that P. acidilactici

may be considered a surrogate for Salmonella reduction on whole macadamia nuts and whole

cashews processed using saturated steam at 80 ͦC.

63

Introduction

Improvement of microbial quality of tree nuts through process interventions include

forms of thermal treatments (oil roasting, dry roasting, steam), and non-thermal treatments but

also chemical fumigation processes such as Propylene oxide and Ethylene oxide (11).

Salmonella presence has prompted recalls in walnuts, macadamia nuts, pecans, cashews,

almonds, pistachios, pine nuts and hazelnuts (16). Salmonella has been responsible for many

illnesses and recalls associated with low water activity (low-aw) foods in the United States, and

many can be related to under processing and contaminated tree nuts (9, 21). Salmonella cannot

replicate in a system that has a water activity below 0.98, but food products with aw of 0.7 or

lower are commonly contaminated with Salmonella, leaving tree nuts susceptible to harboring

pathogens such as Salmonella (1, 10). In recent studies, Salmonella has been documented to

survive within low water activity products for periods up to 2 years (9, 18). With a wide variety

of (low-aw) foods having been recalled, processing methods which have been traditionally used

to maintain shelf life and quality of nuts, must further be evaluated to acknowledge their

efficacy (22)

Saturated steam in processing of low-aw, is used with the intention to reach a 4-log

reduction in Salmonella (6, 25). Steam has been utilized as a lethality step in many food

processes, and meets the United States Department of Agriculture (USDA) National Organic

Program standards (25, 26). Administration of small amounts of saturated steam that release

large amounts of heat when condensed in a short period of time, has been shown to reduce major

loss of aromatics and those changes detrimental to the organoleptic properties (14, 25).

Maintaining quality in aspect of taste, texture and mouth feel is an important consideration and

packaging and processing parameters are specifically designed for each type of nut to minimize

64

changes to quality (22). With trends showing a rise in consumer’s interest in organic foods as a

whole (29), further validation of saturated steam processes, and the packaging which foods are

processed, will be investigated.

Parameters of saturated steam processing warrant for the identification of a surrogate to

avoid the introduction of Salmonella in the processing facility. A surrogate must be non-

pathogenic, and have similar stability and inactivation characteristics to the target pathogen

under the parameters of each saturated steam process and each type of food being processed

(15). Enterococcus faecium NRRL B-2354 has been demonstrated to have a greater heat

tolerance than many Salmonella spp., on a number of food products(2, 13, 31). E. faecium

NRRL B2354 (ATCC #8459) has been validated as an appropriate surrogate for Salmonella in

regards of thermal processing of almonds in a moist air thermal processing system at various

temperatures and times (13), formed jerky processing with small-business and commercial

thermal processing parameters (3), and thermal processing of dried pet food through extrusion

methods (31). There are instances where the inactivation characteristics were not comparable to

Salmonella, showing the need to validate each process separately and consider multiple

microorganisms as possible surrogates. Pediococcus acidilactici (ATCC #8042) has

demonstrated to possess greater heat resistant than Salmonella in thermal processes, as have

many lactic acid bacteria (3, 8, 30). Dried pet food in thermal processing studies through

extrusion methods and ground and formed beef jerky through a range of thermal processes have

been studied, validating Pediococcus acidilactici (ATCC #8042) as a conservative surrogate for

Salmonella (3, 8). Pediococcus acidilactici (ATCC #8042) has also been used in fermenting

pork jerky products, which provide further reasoning to compare inactivation due to thermal

processing, since available studies provide literature on how it acts in various products and it

65

can have multiple uses in products like dried meats (30). Staphylococcus carnosus (ATCC

#51365), having shown potential surrogacy for Salmonella in frankfurter batter undergoing a

thermal process treatment (27), also provides evidence of potential surrogacy in commodities

being treated in thermal processes. Staphylococcus carnosus (ATCC #51365), while a

comparable surrogate in this study, would need to be further compared in a company’s specific

processing parameters. Staphylococcus carnosus (ATCC #51365) does provide evidence of a

greater heat tolerance compared to Salmonella spp. in recent research (27), but this is dependent

on many intrinsic factors of the food system.

The objective of this study was to compare the survival of Salmonella enterica with

Enterococcus faecium NRRL B2354 (ATCC #8459), Staphylococcus carnosus (ATCC #51365),

and Pediococcus acidilactici (ATCC #8042) on whole cashews and macadamia nuts subjected

to saturated steam processing. Bacteria with comparable or greater survival compared to

Salmonella may be considered as a surrogate on cashews and macadamia nuts processed with

saturated steam, provided their process is validated to achieve comparable temperatures with in

each package

Materials and Methods

Bacterial strains and growth conditions:

Inoculum preparation:

Five Salmonella enterica serovars obtained from low-aw foods (4 strains including

Montevideo, an isolate from 2010 peppercorn associated outbreak, Tennessee, K4643 human

isolate from 2007 peanut butter associated outbreak, Ball ARL-SE-085 isolated from black

pepper in 2011, and Johannesburg ARL-SE-013 from dried ginger in 2010) and Salmonella

66

Senftenberg 775W, ATCC 43845, frequently used for thermal studies due to increased heat

resistance, were obtained from the FDA or American Type Culture Collection (ATCC,

Manassas, VA).Three potential surrogate bacteria Enterococcus faecium (NRRL B-2354, ATCC

strain #8459), Pediococcus acidilactici (ATCC #8042) and Staphylococcus carnosus, subspecies

carnosus (ATCC #51365), were obtained from the ATCC.

Wet inoculation with Salmonella strains and Surrogates grown on TSA:

Inoculation preparation was adapted from the method developed for almonds and is

associated with increased stability of Salmonella on whole peppercorns and cumin seeds

compared to a broth grown method (4, 5). Briefly, each strain was spread onto a 150 x 15 mm

(BD Falcon, Franklin Lakes, NJ) large Tryptic Soy Agar (TSA) (Becton Dickinson, Franklin

Lakes, NJ) plates and incubated at 37 °C for 24 h. Cells were harvested from each of the plates

and placed in separate sterile conical tubes by first applying 0.1% (w/v) peptone-tween (PT)

buffer to each plate and scraping colonies into suspension using a sterile cotton swab. For

Salmonella, the suspensions of 5 strains were combined in comparable volumes corresponding to

similar CFU of each strain as determined previously (6). Equal volumes of the Salmonella

cocktail (12.5 ml) and one of the three potential surrogates (12.5 ml) were mixed in a total

volume of 25 ml. The suspensions were then transferred into 27 oz. sterile Whirl-Pak bags

containing either 400 g of dry whole “Fancy grade” macadamia nuts or whole cashews and a

mixture of shaking, inversion and hand massaged for 1 minute to evenly coat the nuts. After

inoculation, nuts were arranged in a single layer on double layered, sterilized filter paper on top

of drying racks placed inside a biological safety cabinet for drying at room temperature. Nuts

were dried for 24-48 h to allow the nuts to return to the pre-inoculation aw of around 0.44 for

67

macadamia nuts and aw of 0.51 for cashews at room temperature. Water activity of whole

macadamia nuts (5 g) and cashews (4 g) was determined by using an AquaLab 4TE water

activity meter (AquaLab, Pullman, WA).

Packing and Shipping methods:

The dried, inoculated nuts (50 g cashew or 50 g macadamia) were individually packaged

within muslin drawstring sachets (4 x 6 in.) that were tied to a framework of non-conducting

thin CPVC pipe (12.7 mm diameter pipe, assembly dimensions 38.8 cm x 26.7 cm). Framework

prevented the movement of the sachets throughout a larger bag poly-woven polypropylene bag

(14 x 24 in.), when filled with ~3000g of non-inoculated nuts to achieve a bag thickness of 7cm

depth when placed horizontally. Each bag contained 6 sachets of the specific nut inoculated

with Salmonella and either E. faecium NRRL B2354 (ATCC #8459), P. acidilactici (ATCC

#8042), or Staphylococcus carnosus (ATCC #51365). Each potential surrogate/Salmonella

combination of inoculated nuts were placed within separate poly-woven bags (three bags per nut

type, six bags in total). The bags were sealed by a combination of high strength glue and

industrial staples to prevent any potential cross contamination and the bags were placed in

insulated boxes. The boxes were shipped overnight to a commercial processor for saturated

steam treatment with instructions on how to prepare the bags to assure uniform and consistent

distribution of the nuts before treatment: to flip each bag gently three times and to lay the bag

flat on the surface for treatment after removal from the shipping packaging. The bags were to be

laid face up (signified by a written label), and the processor included matchstick data loggers

with in the saturated steam run to verify temperatures during processing were reached and

remained consistent through each of the three runs.

68

Saturated Steam Treatments:

Saturated steam processing was performed by Cosmed ETO, Inc. in accordance with

guide lines set for processing (11) and according to the parameters provided in Table 1. These

process parameters have been modified to take into consideration the lab scale study that was

performed to obtain countable ranges of bacteria in recovery, post lethal treatment. To help

with the study, a platform was constructed of stacked pallets and centrally located within a 12-

pallet chamber. The six bags were placed on top of the pallets and temperature data loggers

were placed in various locations on the load. The average temperature during each of the runs

is reported in figure 1. Dedicated runs were performed for these treatments to minimize the

potential for cross contamination within the facility. Three separate processing runs were

performed on different days. After processing the bags were returned to the insulated shipping

container and shipped back to Virginia Tech for enumeration.

Enumeration:

Salmonella and the corresponding surrogate bacteria were enumerated from nuts at three

time periods: after drying, after saturated steam treatment and after storage of non-treated nuts.

The enumeration method was adapted from the method described by Stevenson et al. for

enumerating E. faecium NRRL B2354 (ATCC #8459) from almonds (24). Nuts from each

sachet were mixed with Butterfield’s phosphate buffer (BPB) in a 1:1 ratio in a sterile sample

bag (25ml to 25g of treated nuts) and shaken 50 times with vigorous pace in an arc motion of

1ft, using 27 oz. steril Whirl-pak bags. After a 5 min. rest period the nuts were shaken again 5

times before pouring the liquid into a sterile tube. Tenfold serial dilutions in BPB were

performed. Each dilution was spread plated onto TSA plates in duplicate, incubated 35 ͦ C for

69

3h. followed by overlay with 7 ml of XLT4 for recovery of injured Salmonella. Then incubated

for the remainder at 35 ͦ C for 24h. Each dilution was also plated in duplicate onto the

appropriate selective media required to recover the given surrogate strain (MSA for

Staphylococcus carnosus (ATCC #51365) and BEA for P. acidilactici (ATCC #8042) and E.

faecium NRRL B2354 (ATCC #8459)). Plates were incubated at 37°C for 24 h. An aliquot of 1

mL of the supernatant was also transferred to 9 mL of Tryptic soy broth and incubated at 37°C

for 24 h and subsequently streaked onto selective media resulting in a limit of detection of 1.0

log CFU/g.

Experimental Design and Statistical Analysis:

Three separate saturated steam treatments were performed on three separate days. On

each day six sachets containing a combination of Salmonella and one of the three tested

surrogates were placed within a larger bag. A separate bag was processed for each

surrogate/Salmonella pair and nut type. There was a total of six bags processed each day (three

cashew, three macadamia nuts).

Bacterial counts were log transformed prior to statistical analysis. The reduction in log

CFU/g was calculated by subtracting the log CFU/g after saturated steam processing from the log

CFU/g of unprocessed nuts that were inoculated on the same day as the saturated steam

processed nuts and enumerated on the day of receipt of the steam processed nuts to account for

any differences. The log CFU/g reduction of Salmonella was compared to the log CFU/g

reduction of the corresponding potential surrogate organism from each sachet, for each nut type

using a matched pairs t-test. The mean log CFU/g reduction of Salmonella was compared to the

mean log CFU/g reduction of the corresponding potential surrogate organism; each nut type was

also compared using a one-way ANOVA using JMP 13.0. P < 0.05 were considered significant.

70

Results

Comparison of reduction of Salmonella (log CFU/g) on cashews and macadamia nuts treated

with saturated steam to potential surrogate bacteria:

While saturated steam processing significantly reduced the mean populations of

Salmonella and each of the tested potential surrogate bacteria on whole macadamia nuts and

whole cashews (p<0.05), the amount of reduction varied based on type of nut and on bacterial

strain (Table 2). The reductions in Salmonella as compared to E. faecium NRRL B2354 (ATCC

#8459) (Fig. 2), P. acidilactici (ATCC #8042) (Fig. 3), St. carnosus (ATCC #51365) (Fig. 4)

provide direct comparison of inactivation between Salmonella spp. and the designated potential

surrogate organism with in each sachet (n=18). While overall the mean reduction of Salmonella

and E. faecium were comparable, the reduction of E. faecium exceeded that of Salmonella on

macadamia nuts in 5/18 sachets (Fig. 2). Salmonella reduction significantly exceeded that of P.

acidilactici (ATCC #8042) in all sachets (Fig. 3). Reduction of St. carnosus (ATCC #51365)

(Fig. 4) on both cashews and macadamia nuts exceeded that of Salmonella in 8/ 12 sachets.

No colonies with characteristic black centers indicative of Salmonella appeared on XLT4

plates from non-inoculated nuts. No dark brown/black colonies indicative of E. faecium NRRL

B2354 (ATCC #8459) or P. acidilactici (ATCC #8042) appeared on BEA plates from non-

inoculated nuts. No small, white and round colonies indicative of Staphylococcus carnosus

(ATCC #51365) growth appeared on MSA.

71

Table 2. Mean reduction log CFU/g of Salmonella compared to three potential surrogate bacteria co-

inoculated onto macadamia nuts or cashews after saturated steam treatments.

Mean reduction* log CFU/g ± std error of mean (n=18) Macadamia nuts Cashews Notes: Salmonella 5.3 ± 0.17 A 6.0± 0.12 a Five points where the

reduction of E. faecium exceeds that of Salmonella on macadamia nuts. No instances on cashews.

Enterococcus faecium NRRL B2354

5.2± 0.17 A 4.3± 0.12 b

Mean difference § -0.102± 0.14 -1.67± 0.13 Mean difference Matched pairs T-test P-Value

0.46 < 0.0001

Salmonella 5.9± 0.18 A 5.9± 0.14 a Reduction of

Salmonella exceeded that of Pediococcus in each instance

Pediococcus acidilactici ATCC 8042

4.4± 0.18 B 3.7± 0.14 b

Mean difference § -1.46± 0.14 -2.20± 0.10 Mean difference Matched pairs T-test P-Value

<0.0001 <0.0001

Salmonella 6.4± 0.19A 5.8± 0.12a Eight points where the

reduction of St. carnosus exceeds that of Salmonella (6 on cashews, 2 on macadamia nuts) (n=12 due to contamination)

Staphylococcus carnosus ATCC 51365

6.5± 0.20A 6.1± 0.14b

Mean difference § -2.3± 0.64 -0.85± 0.52 Mean difference Matched pairs T-test P-Value

0.004 0.13

72

*Mean reduction (log CFU/g) reduction is calculated for each co-inoculated surrogate and

Salmonella nut. Each combination was placed in separate bags. Standard error of the means

reflects differences in inactivation from three independent trials (process days). Variability

between Salmonella reported on the table may be due to location placed within the processing

chamber. Different letters within surrogate and Salmonella pairs (superscript letters denote

significance in results, when written as A,B or a,b. Upper case lettering is used in Macadamia

nut results while lower case is used in Cashew results) indicates statistical significance at p

<0.05 by ANOVA comparison of Means with Tukey’s Adjustment as well as paired comparison

as shown in the following figures.

§ Mean difference is the mean of the difference between the reduction of Salmonella (log

CFU/g) and each co-inoculated surrogate (log CFU/g) packaged within the same sachet and

averaged across each of the three trials. Standard error of the means reflects differences in

inactivation from three independent trials (process days). Output of the matched pairs test of

JMP 13.

73

Figure 2. Comparison in log CFU/g of Salmonella and E. faecium NRRL B2354 (ATCC #8459) of inoculated nuts processed using saturated steam on three separate processing days. Each point represents an individual sachet (n=18) processed during the three trials.

74

Figure 3. Comparison in log CFU/g of Salmonella and P. acidilactici (ATCC #8042) of inoculated nuts processed using saturated steam on three separate processing days. Each point represents an individual sachet (n=18) processed during the three trials.

75

Figure 4. Comparison in log CFU/g of Salmonella and Staphylococcus carnosus (ATCC #51365) of inoculated nuts processed using saturated steam on three separate processing days. Each point represents an individual sachet processed during the three trials (lack of points due to Staphylococcus aureus contamination on trial 1 (n=12)).

76

Discussion

The need to validate various processing parameters regarding process type as well as

food composition and the validation of surrogate bacteria for Salmonella is an ever-growing

need. Utilizing processes such as saturated steam have been successful in reducing Salmonella

by 4-logs (6, 25), through the use of short bursts of steam with high amounts of heat transferred

(25). There is a high appeal of processes such as saturated steam not only for the justified

lethality of the process in low water activity foods, but also due to the lack of organoleptic

changes of the end product (22). Research shows that the potential of having viable surrogate

organisms for saturated steam processing could possibly enable an already high demand market

of tree nut processing to increase consumer safety, with nut consumption and availability

growing in the United States (17).

Considerations for thermal distribution need to be well characterized within the

chambers to assure loads placed in each location receive comparable heat and steam penetration.

If there is variability within a chamber the inoculated packs should be in the coldest area of the

chamber and/or where less steam reaches. Inoculated nuts should be placed within the center of

a standard container, with heat penetration of the packaging of the nuts well understood before

validation trials (11). This will allow placement of inoculated sachets in multiple areas of the

package to assure that differences in product temperature and steam penetration. Equally

important is the type of packaging chosen to hold foods intended for processing. Requirements

for packaging considered valid in chamber processes such as saturated steam, are required to be

permeable (12). Permeability varies between a corregated cardboard box and a wire mesh

container; and it is apparent that food would receiving different distributions of steam when

processed in the same chambers. Interaction between food product and thermal treatment must

77

be observed and therefore validated separately for varying packaging systems. For this research,

poly-woven sand bags were used, for their testable size as well as known permeability. These

bags were not meant to replicate typical packaging or amount of product processed (only

containing 3,000 g of nuts in each), but to show how a saturated steam treatment would interact

with a system co-inoculated with Salmonella spp. and a potential surrogate bacteria. Processing

parameters for the saturated steam chamber were modified to provide enough recovery in each

bacterium to provide a valid comparison on how the treatment affected the system.

With results providing evidence that E. faecium NRRL B2354 (ATCC #8459) and P.

acidilactici (ATCC #8042) are both capable of being considered conservative surrogates for

whole cashews, P. acidilactici (ATCC #8042) conservative for whole macadamia nuts as well,

in this lab scale process, utilizing saturated steam at certain parameters is promising. This

information can help commercial facilities need to validate surrogate bacteria in their own

processes. With steam processing, many factors must be taken into consideration when

validating a surrogate particularly regarding to; the types and brand of equipment being used,

process conditions (steam pressure or vacuum achieved), the amount of product treated per

chamber, chamber size, as well as overall calibrations (11), helping to properly validate a

system.

When validating for food processes and surrogate bacteria, it is important to use proper

methods of enumeration to verify process results. Many media and enumeration techniques have

been developed, though selecting the proper one is almost as important as the food process

itself. For methods in this study, dilution and enumeration techniques were developed based on

one used by the California Almond Board, when processing almonds using Enterococcus

faecium NRRL B2354 (ATCC #8459) as a surrogate (24). Their methods were developed based

78

on the FDA BAM method. Like most methods serial dilutions were plated on selective and non-

selective medias, incubated, and colonies formed are counted to calculate an overall log

reduction. For the best results in recovery of Salmonella spp. and the potential surrogate

organisms, selective media was used.

For Salmonella spp. recovery, traditional overlay methods using XLT-4 was most

efficient, as found by previous thesis research carried out at Virginia Tech (7). Part of this

research observed recovery of Salmonella spp. from steam treated peppercorns and deducted

that traditional overlay of XLT-4 with Tergitol 4 supplement provided the appropriate

environment and nutritional needs to recovery Salmonella spp. that may have been sub-lethally

injured during steam treatments. The method provides the dilution to first be plated onto TSA, a

non-selective media for general aerobic bacteria recovery. The media is incubated for 3 hours,

and then over-laid with 7ml of XLT-4 with Tergitol 4 supplement (7). This method in no way

enriches to provide an enhanced growth in actual number of Salmonella colonies, but provides

an environment best suited for Salmonella spp. recovery over microorganism that may be

present. The allowing of incubation on TSA first provides nutrients for sub-lethally injured

bacteria (23, 28) that may not find the harsh nature of selective XLT-4 suitable by itself (20, 23).

Overlay prevented recovery of potential surrogate organisms as well as other background

microbiota that may still be present on the nuts. Allowing for the recovery of sub-lethally

injured cells provides a more accurate count of Salmonella that remain in the system (19). Sub-

lethally injured Salmonella spp. have the potential to cause problems in a food system and can

result in pending recalls and outbreaks (28).

It should also be noted, that as stated previously, the potential surrogate bacteria were

chosen based on previous use as surrogates of Salmonella in food systems. Each food system

79

must have its own validation; this is apparent with Staphylococcus carnosus (ATCC #51365).

Staphylococcus carnosus (ATCC #51365) was not proven as a valid surrogate in this study on

whole cashews and macadamia nuts, but has been considered comparable for Salmonella in

frankfurter batter in a previous study (27). Some factors that may contribute to varying results

include the shape and composition of different food systems as well as how that food itself reacts

in processing conditions (9). Selection for possible surrogate bacteria was also based on

recoverability as well as bacteria morphology. E. faecium NRRL B2354 (ATCC #8459) and P.

acidilactici (ATCC #8042) are both lactic acid bacteria, which have been proven to have greater

heat resistances than many Salmonella spp. The colony formation also plays a role in surrogate

selection.

The surrogate bacteria should have an easily identifiable colony formation, ideally

varying in appearance enough to not be confused with pathogenic bacteria it is being compared

to. Both E. faecium NRRL B2354 (ATCC #8459) and P. acidilactici (ATCC #8042) form small,

round, colonies when plated on BEA, Enterococcus faecium NRRL B2354 (ATCC #8459)

colonies ferment BEA providing a circle surrounding the white colonies, while P. acidilactici

(ATCC #8042) form white colonies with little to no black appearing on the plate. Staphylococcus

carnosus (ATCC #51365) is selected for by plating on MSA, and appear as small, round,

white/off white colonies, easily compared to a red/pink color of the media. By using these

selective medias for potential surrogate identification, it can be confirmed that all colonies

counted were the intended organism. It is also verified that due to the environment of the media

that Salmonella spp. growth was not supported on these plates, and was itself easily identified as

larger round, white colonies with a black center on XLT-4.

80

Conclusions:

Pediococcus acidilactici (ATCC #8042) may be considered as a surrogate for

Salmonella when saturated steam is used to process whole macadamia nuts and whole cashews

using parameters that achieve comparable temperatures. Enterococcus faecium NRRL B2354

(ATCC #8459) may be considered as a surrogate for Salmonella on cashews when saturated

steam processing parameters achieve comparable temperatures as described in this study. Due

to instances where the reduction of E. faecium NRRL B2354 (ATCC #8459) exceeded that of

Salmonella, its use as a surrogate on stream treated whole macadamia nuts is not recommended.

Staphylococcus carnosus (ATCC #51365) should not be used as a surrogate for Salmonella on

either type of nut because the Salmonella was more frequently resistant to the described

saturated steam process compared to Staphylococcus carnosus (ATCC #51365). Results should

not be extrapolated to other types of nuts as it may result in an unsafe product.

These experiments were not designed to demonstrate the effectiveness of saturated steam

for inactivation of Salmonella in a standard packaging configuration or when using a typical

production load. These chambers are designed to accommodate 8-12 pallets. In this test situation,

the volume of nuts, placement and packaging material was chosen to assure penetration of steam.

Cashews are routinely processed in 40 lb cardboard boxes that may contain a thermal barrier;

therefore it cannot be assumed that the same conditions would produce the same inactivation of

Salmonella within the package.

81

Acknowledgements

Funding for this research was provided in part by Cosmed Eto Inc and by the Virginia

Agricultural Experiment Station and the Hatch Program of the National Institute of Food and

Agriculture, U.S. Department of Agriculture. MP, RCW and JW designed the study.

Processing conditions were defined in consultation with Cosmed personnel. All saturated steam

processing was performed by Cosmed staff. Inoculation, enumeration and analysis were

performed at Virginia Tech. Data collection and analysis were performed by TS with assistance

from JW and MP. The manuscript was written by TS, JW, MP, RCW and HH. We would like to

thank the US FDA ORA Arkansas Regional Lab for providing the spice-isolated strains. Special

thanks to Kim Waterman and Kendall Fogler for providing technical assistance.

82

References

1. Beuchat, L. R. 2009. Behavior of Salmonella in foods with low water activity. In, IAFP Rapid response symposium, Arlington, VA. 2. Bianchini, A., J. Stratton, S. Weier, T. Hartter, B. Plattner, G. Rokey, G. Hertzel, L. Gompa, B. Martinez, and K. M. Eskridge. 2014. Use of Enterococcus faecium as a Surrogate for Salmonella enterica during Extrusion of a Balanced Carbohydrate-Protein Meal. Journal of Food Protection. 77:75-75. 3. Borowski, A. G., S. C. Ingham, and B. H. Ingham. 2009. Validation of Ground-and-Formed Beef Jerky Processes Using Commercial Lactic Acid Bacteria Starter Cultures as Pathogen Surrogates. Journal of Food Protection. 72:1234-1247. 4. Bowman, L. S. 2015. Impacts of inoculation strategy on survival of Salmonella enterica and surrogate Enterococcus faecium at low water activity on dry peppercorn and cumin seeds. In, Food Science and Technology, vol. Master of Science in Life Science. Virginia Tech, Blacksburg, VA. 5. Bowman, L. S., K. M. Waterman, R. C. Williams, and M. A. Ponder. 2015. Inoculation Preparation Affects Survival of Salmonella enterica on Whole Black Peppercorns and Cumin Seeds Stored at Low Water Activity. Journal of Food Protection. 78:1259-1259. 6. California, A. B. o. April 2007. Considerations for Proprietary Processes Used for Almond Pasteurization and Treatment. 7. Caver, C. 2016. Recovery of Salmonella from Steam and Ethylene Oxide-Treated Spices Using Supplemented Agar with Overlay. p. 67. In, College of Food Science and Technology, vol. Master of Science in Food Scinece and Technology. Virginia Polytechnic Institute and State University. 8. Ceylan, E., and D. A. Bautista. 2015. Evaluating Pediococcus acidilactici and Enterococcus faecium NRRL B-2354 as Thermal Surrogate Microorganisms for Salmonella for In-Plant Validation Studies of Low-Moisture Pet Food Products. Journal of Food Protection. 78:934-934. 9. Farakos, S. M. S., D. W. Schaffner, and J. F. Frank. 2014. Predicting Survival of Salmonella in Low-Water Activity Foods: An Analysis of Literature Data. Journal of Food Protection. 77:1448-1448. 10. Farkas, J. 2000. The microbiological safety and quality of food. Aspen Publishers, Gaithersburg, Md. 11. GMA. 2010. Industry handbook for safe processing of nuts. GMA Nut Safety Task Force, Grocery Manufacturers Association, Washington, DC. 12. Howe, D. 2017. Cosmed Group. In M. Ponder (ed.). 13. Jeong, S., B. P. Marks, and E. T. Ryser. 2011. Quantifying the Performance of Pediococcus sp. (NRRL B-2354: Enterococcus faecium) as a Nonpathogenic Surrogate for Salmonella Enteritidis PT30 during Moist-Air Convection Heating of Almonds. Journal of Food Protection. 74:603-603. 14. Joosten, H. M., E. Komitopoulou, t. B. H. Kuile, S. Fanning, L. R. Beuchat, F. Bourdichon, R. P. Betts, and H. Beckers. 2013. Low-water activity foods: increased concern as vehicles of foodborne pathogens. Journal of Food Protection. 76:150-72. 15. Kopit, L. M. 2013. Safety of Enterococcus faecium NRRL B-2354 and Enterococcus faecalis ATCC 29212 for use as surrogate organisms in the thermal processing of dry foods. In, vol. Dissertation/Thesis. ProQuest Dissertations Publishing.

83

16. Palumbo, M., L. R. Beuchat, M. D. Danyluk, and L. J. Harris. 2016. Recalls of tree nuts and peanuts in the U.S., 2001 to present [table and references] U.S. recalls of nuts. 17. Perez, A., and S. Pollack. 2003. Fruit and tree nuts outlook. USDA Electronic report from the economic research service, FTS-302. Accessed in 2017. 19:2003. 18. Podolak, R., E. Enache, W. Stone, D. G. Black, and P. H. Elliott. 2010. Sources and risk factors for contamination, survival, persistence, and heat resistance of Salmonella in low-moisture foods. Journal of Food Protection. 73:1919-1936. 19. Ray, B. 1979. Methods to Detect Stressed Microorganisms 1. Journal of Food Protection. 42:346-355. 20. Rocelle, M., S. Clavero, and L. R. Beuchat. 1995. Suitability of selective plating media for recovering heat-or freeze-stressed Escherichia coli O157: H7 from tryptic soy broth and ground beef. Applied and Environmental Microbiology. 61:3268-3273. 21. Santillana Farakos, S. M., R. Pouillot, N. Anderson, R. Johnson, I. Son, and J. Van Doren. 2016. Modeling the survival kinetics of Salmonella in tree nuts for use in risk assessment. International Journal of Food Microbiology. 227:41-50. 22. Shakerardekani, A., R. Karim, H. M. Ghazali, and N. L. Chin. 2013. Textural, rheological and sensory properties and oxidative stability of nut spreads—a review. International journal of molecular sciences. 14:4223-4241. 23. Speck, M., B. Ray, and R. Read. 1975. Repair and enumeration of injured coliforms by a plating procedure. Applied microbiology. 29:549-550. 24. Stevenson, K., K. Ito, L. Beuchat, B. P. Marks, and D. Ashton. 2014. Guidelines for using Enterococcus faecium NRRL B-2354 as a surrogate microorganism in almond process validation. In, 2015 Almond Board of California

25. Systems, R. P. Date, 2012, Sterilization / Pasteurization of dried products. Available at: http://revtech-process-systems.com/index.php/en/steam-sterilization-pasteurization Accessed, January 2017. 26. USDA. 2016. National Organic Program: USDA Organic Regulations. In Federal Information & News Dispatch, Inc, Lanham. 27. Vasan, A., R. Geier, S. C. Ingham, and B. H. Ingham. 2014. Thermal Tolerance of O157 and non-O157 Shiga Toxigenic Strains of Escherichia coli, Salmonella, and Potential Pathogen Surrogates, in Frankfurter Batter and Ground Beef of Varying Fat Levels. Journal of Food Protection. 77:1501-1501. 28. Wu, V. 2008. A review of microbial injury and recovery methods in food. Food microbiology. 25:735-744. 29. Yadav, R. 2016. Altruistic or egoistic: Which value promotes organic food consumption among young consumers? A study in the context of a developing nation. Journal of Retailing and Consumer Services. 33:92-97. 30. Zhao, X., J. Zhang, H. Ming, W. Zou, Y. Wang, C. Zhao, S. Mu, H. Zhang, and L. Qiu. 2016. Research on Pork Jerky Obtained Through Fermentation with Pediococcus acidilactici. p. 25. In, Polish Journal of Food and Nutrition Sciences, vol. 66. 31. Zhou, T. 2016. Residence time and survival studies for Enterococcus faecium as a surrogate for Salmonella during preconditioning and extrusion processing of dry expanded pet food. In Kansas State University.

84

Table 1. Saturated Steam processing parameters

Phase Parameter Setpoint Minimum Maximum Operator Observed

Start Cycle

Jacket Temperature ( ͦ C)

85.0 76.67 98.89 85.00

Product Temperature ( ͦ C)

N/A 10.0 N/A 22.78

Phase Parameter Setpoint Minimum Maximum Operator Observed

Vacuum Pressure (PSIA) 0.01 0.00 0.31 0.25

Steam Injection Pressure (PSIA) 2.00 2.00 2.50 2.02

Vacuum Pressure (PSIA) 0.50 0.20 0.80 1.97

DEC Pressure (PSIA) 6.00 5.25 6.75 6.00

Product Temperature - - - -

Conditioning Dwell-DEC

Pressure (PSIA) 6.00 5.25 6.75 5.99

Time (hh:mm:ss) Commodity specific

00:01:00 00:02:00 00:01:00

Product Temperature ( ͦ C)

76.67 73.89 82.22 76.11

Avg. Chamber Temperature ( ͦ C)

85.00 76.67 98.89 91.61

Vacuum Pressure (PSIA) 0.01 0.00 0.50 0.15

Air Inbleed Pressure (PSIA) Atmospheric 14.00 15.00 14.68

Total Time Processed

Time (hh:mm:ss) - - - 01:05:00

*All measurements were monitored and supplied to the researcher by Cosmed groups.

85

Figure 1. Average temperature (°C, ± standard error) of chamber during Saturated Steam

processing of cashews and macadamia nuts. Temperatures were recorded every minute using

6 data loggers placed on top of the processed bags and underneath the bags.

0

10

20

30

40

50

60

70

80

90

1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91

Tem

pera

ture

ºC

Time (minutes)

Average Temperature (ºC) per each trial

Trial 1 Trial 2 Trial 3

86

CHAPTER 5: FUTURE RESEARCH

With the need to have validated processes to ensure a 5-log reduction in pathogenic

bacteria on tree nuts and other low water activity foods treated with saturated steam and

propylene oxide, further research with surrogate and indicator organisms is pertinent. Each time

a single parameter in a process (chamber size, commodity, packaging, etc.) deviates from a

previous validated process, the entire process must be re-evaluated. This leaves the research

potential seemingly limitless when it comes to repeating the same style process for each of the

products that producers wish to process.

In regards to the process and methodology that have been carried out in this body of

work, it should be noted that many validation studies include a pre-sanitation process of the

commodity before inoculation of pathogenic and potential surrogate bacteria alike. While this is

common practice in many studies, it should be recognized that we felt inoculum should have to

compete with natural microbiota. Also, noting the importance of choosing surrogate bacteria that

is different from the natural microbiota that already exists. This way the comparison in

inactivation has been performed in an environment most similar to that of a preprocessed product

in the industry. This brings up furthering studies in similar co-inoculation techniques to see how

potential surrogate bacteria will react with the target pathogenic bacteria in the system before,

during and post processing for the lethality treatment. These parameters set up more realistic

scenarios than some studies provide.

As far as inoculation technique, co-inoculation enables to study more exact replicates in a

trial as well as lends data that lends belief to unequal exposure to thermal processing and PPO

fumigate in a single process. Fine-tuning inoculation process as well as implementing dry

87

techniques could lessen the prep time (drying of commodity post-inoculation) for each trial, if

initial log values are high enough to record a 5-log reduction post lethality step.

Repeating experimentation on multiple products, in validated processes, to validate

surrogate bacteria is an avenue to explore, though areas which validate treatable packaging

systems, factoring permeability/penetration and movement of product within, in chamber

processing, can be a revenue for continued research. It has been stated multiple times that

packaging of products treated with both steam and PPO can vary, with a main requirement of

permeability. To be able to develop validation of containers which provide a more stream lined

approach and can be used for multiple types of commodities could be helpful as regulations

require more and more cooperation with validating processes.