UTILIZATION OF ENTEROCOCCUS FAECIUM NRRL B - 2354 AS A SALMONELLA SURROGATE FOR VALIDATING THERMAL TREATMENT OF LOW - MOISTURE FOODS By Nurul Hawa Ahmad A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Food Science - Doctor of Philosophy 2020 ABST R ACT UTILIZATION OF ENTEROCOCCUS FAECIUM NRRL B - 2354 AS A SALMONELLA SURROGATE FOR VALIDATING THERMAL TREATMENT OF LOW - MOISTURE FOODS By Nurul Hawa Ahmad L ow moisture foods (L MF ) have been associated with numerous Salmonella outbreaks and recalls , as a result of either an insufficient pathogen lethality step or post - processing contamination . The use of a non - pathogenic surrogate microorganism, Enterococcus faecium NRRL B2354, has been proposed to be used as a Salmonella surrogate for validating thermal treatment of LMF since the use of Salmonella is prohibited in the food processing facilities. The overall goal of this study was to validate the efficacy of E. faecium as a Salmonella surrogate in validating the thermal processing of LMF products. This research goal was achieved via three specific objectives : i ) to assess the influence of talc powder as a dry - inocul um carrier on thermal resistance of E. faecium i n almond meal at 0.45 a w ; ii) to compare the thermal resistance of E. faecium and Salmonella in almond meal, date paste, wheat flour, peanut butter, ground black pepper , and non - fat dried milk powder via standardized methodology in an interlaboratory study ; iii) to determine the effect of lactose and protein content on thermal resistance of E. faecium and Salmonella in dairy powders. Log - linear and Bigelow model s were used to estimate D - and z - values in this study. The findings s howed that talc powder is not recommended as a dry inoculation carrier , because the presence of talc influenced the t hermal resis tance of E. faecium in almond mea l . T he interlaboratory comparison study demonstrated that E. faecium was more thermal ly resistan t ( P < 0.05) than Salmonella in the 5 out of 6 low - moisture products , and these trends were highly impacted by product composition. S tandardized methodolog y yield e d reproducible D - and z - values between laboratories. The magnitude of difference of thermal resistance between microorganisms was most substantial in nonfat dry milk powder . Lastly, the presence of lactose, but not protein content , most likely contributed to the magnitude difference of thermal resistance between E. faecium and Salmonella in dairy powders. In summary, E. faecium NRRL B - 2354 can be considered as a relevant biological validation tool for thermal processing of LMF. The positive impact of this study is to support the use of E. faecium NRRL B - 2354 for validation of LMF thermal pasteurization and to ultimately r educe Salmonella outbreaks and recalls linked to LMF. iv To Muhammad Amsyar Jafri, thank you for the sacrifices you have made for me; To my dear parents and family in Malaysia, thank you for supporting me throughout this journey. v ACKNOWLEDGEMENT Praise be for Allah. This success would not be possible without a continuous support from important folks as follows: My dearest husband, Muhammad Amsya r , who shows me a different perspective of life and inspire me to be the best version of myself. My major advisor , Dr. Elliot Ryser , who a lways guide s and encourage s me to be an excellent researcher and teacher. My supervisory committee member, Dr. Bradley Marks , a cheerful and brilliant engineer, who has trusted me to lead the surrogate project. Despite all challenges, this experience has t aught me how to interact with people and overcome difficult situation and decision making. Dr. Kirk Dolan , an enthusiastic professor, who has introduced me to MATLAB coding. Dr. James Pes t ka , who is a supportive and easy - going committee member. My research colleagues ( Ian, Phil, Carly, Quincy , Pichamon, Be atriz, Kaitlyn, Francisco , Michael James , and Nicole Hall ) and numerous undergraduates , who always assist me and be there for me despite their hectic schedules. Last but not the least , I would like to express my gratitude to Ministry of Education Malaysia and Universiti Putra Malaysia for sponsoring my doctoral study. This work is supporte d by the National Institute of Food and Agriculture, U.S. Department of Agriculture, award number 2015 - 68003 - 23415. vi TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ........................ x LIST OF FIGURES ................................ ................................ ................................ .................... xv KEY TO ABBREVIATIONS AND SYMBOLS ................................ ................................ .... xvii 1. INTRODUCTION ................................ ................................ ................................ ..................... 1 1.1 Background of research ................................ ................................ ................................ .. 1 1.2 Problem statement ................................ ................................ ................................ ........... 2 2. LITERATURE REVIEW ................................ ................................ ................................ ........ 4 2.1 Low - moisture foods (LMF) ................................ ................................ ............................ 4 2.2 Salmonella ................................ ................................ ................................ ....................... 6 2.2.1 Survival mechanism of Salmonella in the desiccated state ................................ ........ 7 2.3 Salmonella in specific commodity of LMF ................................ ................................ .... 9 2.3.1 Nuts and tree - nuts ................................ ................................ ................................ ....... 9 2.3.2 Dried fruits ................................ ................................ ................................ ................ 11 2.3.3 Spices and herbs ................................ ................................ ................................ ........ 12 2.3.4 Cereal and grains ................................ ................................ ................................ ....... 14 2.3.5 Dairy powders ................................ ................................ ................................ ........... 15 2.4 Factors influencing thermal resistance of Salmonella ................................ .................. 17 2.4.1 Treatment temperature ................................ ................................ .............................. 17 2.4.2 Water activity ................................ ................................ ................................ ............ 18 2.4.3 Recovery media and enumeratio n ..18 2.4.4 Inoculation method ................................ ................................ ................................ ... 19 2.4.5 Food composition and structure ................................ ................................ ................ 22 2.5 Enterococcus faecium NRRL B - 2354 as a potential Salmonella surrogate ................. 27 2.5.1 Desirable characteristics of a surrogate microorganism ................................ ........... 27 2.5.2 Characteristics of Enterococci ................................ ................................ .................. 28 2.5.3 Safety evaluation of E. faecium ................................ ................................ ................ 29 2.5.4 E. faecium NRRL B - 2354 in low - moisture food studies ................................ .......... 30 2.6 Microbial inactivation model ................................ ................................ ........................ 31 2.7 Literature summary and knowledge gaps ................................ ................................ ..... 33 3. EFFECT OF TALC AS A D RY - INOCULATION CARRIER ON THERMAL RESISTANCE OF ENTEROCOCCUS FAECIUM NRRL B - 2354 IN ALMOND MEAL .. 34 3.1 Introduction ................................ ................................ ................................ ................... 34 3.2 Materials and methods ................................ ................................ ................................ .. 36 3.2.1 Experimental design ................................ ................................ ................................ .. 36 3.2.2 Talc ................................ ................................ ................................ ........................... 37 3.2.3 Almo nds . 37 3.2.4 Inoculum preparation ................................ ................................ ................................ 38 3.2.5 Wet inoculation ................................ ................................ ................................ ......... 39 vii 3.2.6 Dry inoculation ................................ ................................ ................................ ......... 39 3.2.7 Isothermal treatment ................................ ................................ ................................ . 40 3.3 Results and discussion ................................ ................................ ................................ . 41 3.3.1 Particle size distribution of almond meals ................................ ................................ 41 3.3.2 Microbial background populations on whole almonds and heat - treated talc. .......... 42 3.3.3 E. faecium populations on almonds, almond meal, and talc before heat treatment .. 42 3.3.4 Thermal resistance of E. faecium ................................ ................................ .............. 43 3.4 Conclusion ................................ ................................ ................................ .................... 45 4. INTERLABORATORY EVALUATION OF ENTEROCOCCUS FAECIUM NRRL B - 2354 AS A SALMONELLA SURROGATE FOR VALIDATING THERMAL TREATMENT OF MULTIPLE LOW - MOISTURE FOODS ................................ ............... 46 4.1 Introduction ................................ ................................ ................................ ................... 46 4.2 Materials and methods ................................ ................................ ................................ .. 49 4.2.1 Overall study design ................................ ................................ ................................ . 49 4.2.2 Inoculum preparation ................................ ................................ ................................ 50 4.2.3 Proximate analyses ................................ ................................ ................................ .... 51 4.2.4 Background microbial counts ................................ ................................ ................... 52 4.2.5 Product inoculation, equilibration, and dissemination ................................ .............. 52 4.2.6 Come - up time ................................ ................................ ................................ ............ 57 4.2.7 Isothermal treatment and survivor enumerat ion ................................ ....................... 57 4.2.8 Water activity measurement at 80°C ................................ ................................ ........ 58 4.2.9 Data analysis ................................ ................................ ................................ ............. 59 4.3 Results ................................ ................................ ................................ ........................... 60 4.3.1 Proximate analyses ................................ ................................ ................................ .... 60 4.3.2 Background microbial counts ................................ ................................ ................... 61 4.3.3 Thermal come - up time and initial bacterial population ................................ ............ 62 4.3.4 Inactivation kinetics of Salmonella and E. faecium ................................ .................. 62 4.3.5 Thermal resistance of Salmonella and E. faecium between laboratories .................. 66 4.3.6 Influence of product composition on thermal resistance ................................ .......... 68 4.3.7 a w level at high temperature ................................ ................................ ...................... 69 4.4 Discussion ................................ ................................ ................................ ..................... 70 4.5 Conclusion ................................ ................................ ................................ .................... 75 5. EFFECT OF LACTOSE AND PROTEIN ON THERMAL RESISTANCE OF ENTEROCOCCUS FAECIUM NRRL B - 2354 AND SALMONELLA IN DAIRY POWDERS 76 5.1 Introduction ................................ ................................ ................................ ................... 76 5.2 Materials and method ................................ ................................ ................................ .... 81 5.2.1 Dairy powders ................................ ................................ ................................ ........... 81 5.2.2 Overall experimental design ................................ ................................ ..................... 81 5.2.3 Background microbial counts ................................ ................................ ................... 83 5.2.4 Inoculum preparation ................................ ................................ ................................ 83 5.2.5 Inoculation of dairy powders ................................ ................................ .................... 84 5.2.6 Come - up time and isothermal treatment ................................ ................................ ... 85 5.2.7 Survivor enumeration and estimation of D and z T values ................................ ........ 86 viii 5.2.8 Measurement of a w at 80 °C ................................ ................................ ....................... 87 5.2.9 Glass transition measurement using DSC ................................ ................................ . 87 5.3 Results ................................ ................................ ................................ ........................... 88 5.3.1 Proximate analysis and background microbial counts ................................ .............. 88 5.3.2 Initial populations and come - up time ................................ ................................ ....... 88 5.3.3 Isothermal inactivation curves ................................ ................................ .................. 89 5.3.4 Effect of lactose and protein content ................................ ................................ ........ 89 5.3.5 Effect of sugars o n resuscitation of survivor in SMP and LSMP ............................. 92 5.3.6 A w of dairy powders at 80 °C . ................................ ................................ ................... 94 5.3.7 Glass transition temperature (Tg) of dairy powders ................................ ................. 95 5.4 Discussion ................................ ................................ ................................ ..................... 96 5.5 Conclusion ................................ ................................ ................................ .................. 100 6. OVERALL CONCLUSION AND RECOMMENDATIONS ................................ ....... 101 6.1 Overall conclusions ................................ ................................ ................................ ..... 101 6.2 Recommendations for future work ................................ ................................ ............. 102 APPENDICES ................................ ................................ ................................ ........................... 105 APPENDIX A. Previous validation studies using Enterococcus faecium NRRL B - 2354 ..... 106 APPENDIX B. Isothermal inactivation data for almond meal and talc (Chapter 3) .............. 109 APPENDIX C. Isothermal i nactivation data for nonfat dry milk powder at 0.25 a w (chapter 4) ................................ ................................ ................................ ................................ ................. 112 APPENDIX D. Isothermal inactivation data for peanut butt er at 0.25 a w (chapter 4) ............ 118 APPENDIX E. Isothermal inactivation data for almond meal at 0.45 a w (chapter 4) ............ 124 APPENDIX F. Isothermal inactivation data for wheat flour at 0.45 a w (chapter 4) ............... 130 APPENDIX G. Isot hermal inactivation data for ground black pepper at 0.45 a w (chapter 4) 137 APPENDIX H. Isothermal inactivation data for date paste at 0.65 a w (chapter 4) ................. 143 APPENDIX I. Isothermal inactivation data for skim milk powder (chapter 5) ...................... 149 APPENDIX J. Isothermal inactivation data for lactose - free skim milk powder (chapter 5) .. 152 APPENDIX K. Isothermal inactivation data for lactose powder (chapter 5) ......................... 155 APPENDIX L. Isothermal inactivation data for milk protein isolate 90% powder (chapter 5) ................................ ................................ ................................ ................................ ................. 158 APPENDIX M. Homogeneity data (Chapter 3) ................................ ................................ ..... 161 APPENDIX N. Homogeneity data (chapter 4) ................................ ................................ ....... 164 APPENDIX O. Homogeneity data (chapter 5) ................................ ................................ ....... 166 APPENDIX P. Come - up time for almond meal and talc (Chapter 3) ................................ .... 167 APPENDIX Q. Come - u p time (chapter 4) ................................ ................................ ............. 168 APPENDIX R. Come - up time (chapter 5) ................................ ................................ .............. 170 APPENDIX S. a w changes during heating at 80 °C - example (chapter 4) ............................. 171 APPENDIX T. MATLAB codes for log - linear model (chapter 3) ................................ ......... 172 APPENDIX U. MATLAB codes for aggregated D - values of log - linear model (chapter 4 and 5) ................................ ................................ ................................ ................................ ............. 174 APPENDIX V. MATLAB codes for global estimates of log - linear model and Bigelow model (chapter 4) ................................ ................................ ................................ ............................... 176 APPENDIX W. MATLAB codes for global estimates of log - linear model and Bigelo w model with optimum Tref and selected Tref (chapter 5) ................................ ................................ ... 183 ix APPENDIX X. MATLAB codes for effect of sugar types (chapter 5) ................................ .. 187 APPENDIX Y. Scaled sensitivity coe fficient (SSC) and residual plots (chapter 4) .............. 190 APPENDIX Z. Scaled sensitivity coefficient (SSC) and residual plots (chapter 5) ............... 191 APPENDIX AA. Optimal experimental design plot (chapter 4) ................................ ............ 192 APPENDIX AB. Optimal experimental design (chapter 5) ................................ ................... 193 APPENDIX AC. Inoculation flow chart (Chapter 3) ................................ ............................. 194 APPENDIX AD. Photographs of isothermal inactivation experiment ................................ ... 195 APPENDIX AE. Standardized methodol ogy (chapter 5) ................................ ....................... 196 BIBLIOGRAPHY ................................ ................................ ................................ ..................... 217 x L IST O F TABLES Table 2.1 Water activity of low - moisture food ingredients/products at specified temperature ...... 5 Table 3.1 Thermal resistance (D values at 80ºC) of E. faecium in wet inoculated (WI), dry inoculated (DI), wet talc inoculated (WI) almond meal, and inoculated talc powder (TP) at 0.45 a w . ................................ ................................ ................................ ................................ .................. 44 Table 4.1 Moisture, native a w , and chemical composition of tested low - moisture products ........ 61 Table 4.2 D - values for Salmonella and E. faecium across all laboratories an d products ............. 67 Table 4.3 z T for Salmonella and E. faecium across all laboratories (individually and combined) and products ................................ ................................ ................................ ................................ .. 68 Table 4.4 Water activity ( a w ) measurements at 23°C (before and after heating) and at 80°C (n=2) ................................ ................................ ................................ ................................ ....................... 69 Table 5.1 Proximate composition of dairy powders ................................ ................................ ..... 82 Table 5.2 D - and z T - values (T ref = 90°C) for Salmonella and E. faecium in skim milk powder, lactose - free skim milk powder, lactose powder and milk protein isolate 90% at 0.25 a w ............ 93 Table 5.3 D - values for both microorganisms in skim milk powder (SMP) at 90 °C and lactose - free ski m milk powder (LSMP) at 70 °C. ................................ ................................ ...................... 94 Table 5.4 Water activity (a w ) at 23°C (before and after heating) and at 80°C (n=3) .................... 95 Table 5.5 Glass transition and melting temperatures for dairy powders ................................ ...... 95 Table A.1 List of previous validation studies using Enterococcus faecium NRRL B - 2354 in low - moisture foods. ................................ ................................ ................................ ............................ 106 Table B. 1 Enterococcus faecium NRRL B - 2354 inactivation in wet - inoculated almond meal (0.45 a w ) at 80°C. ................................ ................................ ................................ ........................ 109 Table B.2 Enterococcus faecium NRRL B - 2354 inactivation in dry - inoculated almond meal (0.45 a w ) at 80°C. ................................ ................................ ................................ ........................ 110 Table B.3 Raw survivor data for Enterococcus faecium NRRL B - 2354 in wet - talc - inoculated almond meal (0.45 a w ) at 80°C. ................................ ................................ ................................ .. 110 xi Table B.4 Raw survivor data for Enterococcus faecium NRRL B - 2354 in talc powder (0.45 a w ). ................................ ................................ ................................ ................................ ..................... 111 Table C.1 Raw survival data for Salmonella in nonfat dry milk powder (0.25 a w ) at 85°C generated by MSU and WSU. ................................ ................................ ................................ ..... 112 Table C.2 Raw survival data for Salmonella in nonfat dry milk powder (0.25 a w ) at 90°C generated by MSU and WSU. ................................ ................................ ................................ ..... 113 Table C.3 Raw survival data for Salmonella in nonfat dry milk powder (0.25 a w ) at 95°C generated by MSU and WSU. ................................ ................................ ................................ ..... 114 Table C.4 Raw survival data for Enterococcus faecium NRRL B - 2354 in nonfat dry milk powder (0.25 a w ) at 85°C ................................ ................................ ................................ ......................... 115 Table C.5 Raw survival data for Enterococcus faecium NRRL B - 2354 in nonfat dry milk powder (0.25 a w ) at 90°C ................................ ................................ ................................ ......................... 116 Table C.6 Raw survival data for Enterococcus faecium NRRL B - 2354 in nonfat dry milk powder (0.25 a w ) at 95°C ................................ ................................ ................................ ......................... 117 Table D.1 Raw survival data for Salmonella in peanut butter (0.25 a w ) at 85°C generated by UNL and IFSH. ................................ ................................ ................................ ................................ .... 118 Table D.2 Raw survival data for Salmonella in peanut butter (0.25 a w ) at 90°C generated by UNL and IFSH. ................................ ................................ ................................ ................................ .... 119 Table D.3 Raw survival data for Salmonella in peanut butter (0.25 a w ) at 95°C generated by UNL and IFSH. ................................ ................................ ................................ ................................ .... 120 Table D.4 Raw survival data for Enterococcus faecium NRRL - B2354 in peanut butter (0.25 a w ) at 90°C generated by UNL and IFSH. ................................ ................................ ........................ 121 Table D.5 Raw survival data for Enterococcus faecium NRRL - B2354 in peanut butter (0.25 a w ) at 95°C generated by UNL and IFSH. ................................ ................................ ........................ 122 Table D.6 Raw survival data for Enterococcus faecium NRRL - B2354 in peanut butter (0.25 a w ) at 100°C generated by UNL and IFSH. ................................ ................................ ...................... 123 Table E.1 Raw survival data for Salmonella in almond meal (0.45 a w ) at 80°C generated by MSU and WSU. ................................ ................................ ................................ ................................ .... 124 Table E.2 Raw survival data for Salmonella in almond meal (0.45 a w ) at 85°C generated by MSU and WSU. ................................ ................................ ................................ ................................ .... 125 xii Table E.3 Raw survival data for Salmonella in almond meal (0.45 a w ) at 90°C generated by MSU and WSU. ................................ ................................ ................................ ................................ .... 126 Table E.4 Raw survival data for Enterococcus faecium NRRL B - 2354 in almond meal (0.45 a w ) at 80°C generated by MSU and WSU. ................................ ................................ ....................... 127 Table E.5 Raw survival data for Enterococcus faecium NRRL B - 2354 in almond meal (0.45 a w ) at 85°C generated by ................................ ................................ ................................ ................... 128 Table E.6 Raw survival data for Enterococcus faecium NRRL B - 2354 in almond meal (0.45 a w ) at 90°C generated by ................................ ................................ ................................ ................... 129 Table F.1 Raw survival data for Salmonella in wheat flour (0.45 a w ) at 70°C generated by WSU and IFSH. ................................ ................................ ................................ ................................ .... 130 Table F.2 Raw survival data for Salmonella in wheat flour (0.45 a w ) at 75°C generated by WSU and IFSH. ................................ ................................ ................................ ................................ .... 13 2 Table F.3 Raw survival data for Salmonella in wheat flour (0.45 a w ) at 80°C generated by WSU and IFSH. ................................ ................................ ................................ ................................ .... 133 Table F.4 Raw survival data for E. faecium NRRL B - 2354 in wheat flour (0.45 a w ) at 75°C generated by WSU and IFSH. ................................ ................................ ................................ ..... 134 Table F.5 Raw survival data for E. faecium NRRL B - 2354 in wheat flour (0.45 a w ) at 80°C generated by WSU and IFSH. ................................ ................................ ................................ ..... 135 Table F.6 Raw survival data for E. faecium NRRL B - 2354 in wheat flour (0.45 a w ) at 85°C generated by WSU and IFSH. ................................ ................................ ................................ ..... 136 Table G.1 Raw survival data for Salmonella in ground black pepper (0.45 a w ) at 65°C generated by UNL and IFSH. ................................ ................................ ................................ ...................... 137 Table G.2 Raw survival data for Salmonella in ground black pepper (0.45 a w ) at 70°C generated by UNL and IFSH. ................................ ................................ ................................ ...................... 138 Table G.3 Raw survival data for Salmonella in ground black pepper (0.45 a w ) at 75°C generated by UNL and IFSH. ................................ ................................ ................................ ...................... 139 Table G.4 Raw survival data for E. faecium NRRL B - 2354 in ground black pepper (0.45 a w ) at 70°C generated by UNL and IFSH. ................................ ................................ ............................ 140 Table G.5 Raw survival data for E. faecium NRRL B - 2354 in ground black pepper (0.45 a w ) at 75°C generated by UNL and IFSH. ................................ ................................ ............................ 141 xiii Table G.6 Raw survival data for E. faecium NRRL B - 2354 in ground black pepper (0.45 a w ) at 80°C generated by UNL and IFSH. ................................ ................................ ............................ 142 Table H.1 Raw survival data for Salmonella in date paste (0.65 a w ) at 65°C generated by MSU and UGA. ................................ ................................ ................................ ................................ .... 143 Table H.2 Raw survival data for Salmonella in date paste (0.65 a w ) at 70°C generated by MSU and UGA. ................................ ................................ ................................ ................................ .... 144 Table H.3 Raw survival data for Salmonella in date paste (0.65 a w ) at 75°C generated by MSU and UGA. ................................ ................................ ................................ ................................ .... 145 Table H.4 Raw survival data for E. faecium NRRL B - 2354 in date paste (0.65 a w ) at 70°C generated by MSU and UGA. ................................ ................................ ................................ ..... 146 Table H.5 Raw survival data for E. faecium NRRL B - 2354 in date paste (0.65 a w ) at 75°C generated by MSU and UGA. ................................ ................................ ................................ ..... 147 Table H.6 Raw survival data for E. faecium NRRL B - 2354 in date paste (0.65 a w ) at 80°C generated by MSU and UGA. ................................ ................................ ................................ ..... 148 Table I.1 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in skim milk powder (0.25 a w ) at 85°C. ................................ ................................ ................................ ........................ 149 Table I.2 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in skim milk powder (0.25 a w ) at 90°C. ................................ ................................ ................................ ........................ 150 Table I.3 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in skim milk powder (0.25 a w ) at 95°C. ................................ ................................ ................................ ........................ 151 Table J.1 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose - free skim milk powder (0.25 a w ) at 65°C. ................................ ................................ ................................ ... 152 Table J.2 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose - free skim milk powder (0.25 a w ) at 70°C. ................................ ................................ ................................ ... 153 Table J.3 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose - free skim milk powder (0.25 a w ) at 75°C. ................................ ................................ ................................ ... 154 Table K.1 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose powder (0.25 a w ) at 85°C. ................................ ................................ ................................ ........................ 155 xiv Table K.2 Raw surviv al data for Salmonella and E. faecium NRRL B - 2354 in lactose powder (0.25 a w ) at 90°C. ................................ ................................ ................................ ........................ 156 Table K.3 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose powder (0.25 a w ) at 95°C. ................................ ................................ ................................ ........................ 157 Table L.1 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in milk protein isolate 90% powder (0.25 a w ) at 80°C. ................................ ................................ ................................ ... 158 Table L.2 Raw survival data for Salmonella and E. faecium NRR L B - 2354 in milk protein isolate 90% powder (0.25 a w ) at 85°C. ................................ ................................ ................................ ... 159 Table L.3 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in milk protein isolate 90% powder (0.25 a w ) at 90°C. ................................ ................................ ................................ ... 160 Table M.1 Enterococcus faecium NRR L B - 2354 population on whole almond and almond meal during wet - inoculation. ................................ ................................ ................................ ............... 161 Table M.2 Enterococcus faecium NRRL B - 2354 populatio n on whole almond and almond meal during dry - inoculation. ................................ ................................ ................................ ............... 162 Table M.3 Enterococcus faecium NRRL B - 2354 population on whole almond and almond meal during wet - talc - inoculation. ................................ ................................ ................................ ........ 163 Table M.4 Enterococcus faecium NRRL B - 2354 population on talc powder alone. ................. 163 Table N.1 Salmonella and E. faecium population in almond meal, wheat flour, ground black pepper, nonfat dry milk powder, peanut butter, and date paste. ................................ ................. 164 Table O.1 Homogeneity of Salmonella and E. faecium population in dairy powders at 0.25 a w . ................................ ................................ ................................ ................................ ..................... 166 Table P. 1 Come - up time (CUT) for talc powder and almond meal at 0.45 a w . ......................... 167 Table Q.1 Come - up time (CUT) for peanut butter, almond meal, wheat flour, ground black pepper, nonfat dry milk powder, and date pas te. ................................ ................................ ........ 168 Table R.1 Come - up time for dairy powders (0.25 a w ) at three level of temperatures. ............... 170 xv LIST OF FIGURES Figure 3.1 Overall experimental design for wet inoculation (WI), dry inoculation (DI), wet talc inoculatio n (WT), and talc powder inoculation (TP). ................................ ................................ ... 37 Figure 3.2 Populations of E. faecium on almonds, almond meal, and talc using the wet inoculation (WI), dry inoculation (DI), wet inoculation talc (WT), and talc powder (TP) methods before thermal treatment. Each point represents the mean of triplicate samples. ......................... 43 Figure 3.3 Inactivation curve (80ºC) for E. faecium NRRL B - 2354 in talc and almond meal inoculated using the wet inoculation (WI), dry inoculation (DI), and we t talc inoculation (WT) methods. Each point represents triplicate subsamples at each time point. ................................ ... 44 Figure 4.1 Experimental design of multi - laboratory thermal inactivation of E. faecium and Salmonella in six low - moisture products at designated water activities (aw). Each organism was tested at three temperatures; where four temperatures are listed, the top three were for E. faecium , and the bottom three were for Salmonella , based on preliminary works. Arrow indicates laboratory that was responsible for product inoculation. ................................ .............................. 50 Figure 4.2 Interlaboratory isothermal inactivation curves for Salmonella cocktail in: (A) nonfat dried milk powder - 0.25 a w , (B) peanut butter - 0.25 a w , (C) almond meal - 0.45 a w , (D) wheat flour - 0.45 a w , (E) ground black pepper - 0.45a w , (F) date paste - 0.65a w , and the corresponding log - linear model. Data shown are from three independent batches from each laboratory. .......... 63 Figure 4.3 Interlaboratory isothermal inactivation curves for E. faecium cocktail in: (A) nonfat dried milk powder - 0.25 a w , (B) peanut butter - 0.25 a w , (C) almond meal - 0.45 a w , (D) wheat flour - 0.45 a w , (E) ground black pepper - 0.45a w , (F) date paste - 0.65a w , and the corresponding log - linear model. Data shown are from three independent batches from each laboratory. .......... 64 Figure 4.4 D 80 °C (both laboratories) of Salmonella and E. faecium for each product. (* = E. faecium thermal resistance was higher ( P < 0.05) than that of Salmonella . ** = E. faecium thermal resistance was lower ( P < 0.05) than that of Salmonella ) . ................................ .............. 65 Figure 5.1 Isothermal inactivatio n curves for the Salmonella cocktail and Enterococcus faecium NRRL B - 2354) skim milk powder, lactose - free in skim milk powder, lactose powder, and milk protein isolate 90% at 0.25 a w. ................................ ................................ ................................ ...... 91 Figure S.1 Changes of a w during heating from room temperature to 80°C for nonfat dry milk powder indicated by the solid lines ( n=2). Temperature is indicated by dotted line. ................. 171 Figure Y.1 Example of scaled sensitivity coefficient and residual scatter plots for chapter 4 ... 190 xvi Figure Z.1 Example of scaled sensitivity co efficient and residual plots for dairy powders ....... 191 Figure AA.1 Example of optimal experimental design plo t for chapter 4. ................................ . 192 Figure AB.1 Example of optimal experimental design plot for chapter 5. ................................ . 193 Figure AC.1 Flowchart of experimental methodology for chapter 3 ................................ .......... 194 Figure AD.1 Materials and equipment for isothermal inactivation experiment A) conditioning chamber, B) 4TE Aqualab meter to (TDT) cells, and D) TDT cells packed with sample, prior to water bath treatment. .................. 195 xvii KEY TO ABBREVIATIONS AND SYMBOLS LMF Low - moisture foods RMSE Root mean square d error a w Water activity t Time 0 N Number of microorganism (CFU/g) at a specific time point N 0 Initial number of microorganism (CFU/g) at t0 D(T) Time (min) needed to reduce a bacterial population by a 1 log at a specific t emperature D ref Time (min) needed to reduce a bacterial population by a 1 log at T ref T ref Reference temperature ( °C ) z T Temperature ( °C ) required for a log change of D - value 1 1. INTRODUCTION 1.1 Background of research Salmonella is the most frequent pathogen that has been associated with worldwide low - moisture food outbreaks , with 9 major outbreaks (200 0 to 201 7 ) (Harris et al. 2019) and 82 recalls (2001 to 2019) (Yada & Harris, 2019) linked to nuts and tree nuts , 7 outbreaks and 4 recalls linked to grains and cereals ( 1998 to 201 9 ) (Harris & Yada, 2019) , and 2 major outbreaks linked to infant formula (2018 to 2019) (European Food Safety Authority, 2018; Jones, Pardos de la Gandara, et al., 2019) . Although various food safety interventions have been a dopted to mitigate Salmonella contamination , foodborne outbreaks and product recall persist . In 2011, FDA introduced the Food Modernization Act (FSMA) , which requires that all preventive measures taken to inactivate Salmonella in food products be documented as part of the ri sk assessment (U.S. Food and Drug Administration, 2018a) . Microbial challenge studies can be used to provide scientific evidence for the effectiveness of different Salmonella reduction strategies (National Advisory Committee on Microbiologica l Criteria for Foods, 2010) . Hence, Enterococcus faecium NRRL B - 2354 has been proposed as a non - pathogenic surrogate for Salmonella in validating thermal processes of LMF , because the introduction of a pathog en to a food processing facili ty is not acceptable . A surrogate microorganism is considered a good substitute for Salmonella substitute in validation studies if it exhibits similar growth and maintains a high level of population, does not stimulate food spoilage, does not pose pathogenic attributes according to scientific evaluation, and possesses similar susceptibility when exposed to specific microbial inactivation treatments (Busta et al., 2003) . With the use of E. faecium NRRL B - 2354 as a potential Salmonella surrogate, most researche r s are of the opinion that th e surrogate microorganism should exhibit similar or higher thermal resistance than Salmonella as a conservative approach. In this regard, the Almond Board 2 of California established a validation protocol for thermal processing of almonds using E. faecium to evaluate the recommend ed 4 log reduction (Almond Board of Cali fornia, 2014) . This recommendation has become a basic guideline for other LMF, since there are no specific federal regulat ion s prescribing specific surrogates or protocols . Given that the thermal resistance of Salmonella is influenced by various intrins ic and extrinsic factors, there is a critical need to evaluate the feasibility of E. faecium NRRL B - 2354 as a Salmonella surrogate in other LMF . 1.2 Problem statement Low - moisture foods (LMF) c overs a wide range of food compositions (for example: high - fat, high - protein , or high - sugar) . Currently, validation studies using various treatments and product characteristics have been conducted but are limited to almond s (Jeong et al. , 2011) , peanut (Poirier et al. , 2014) , formulated peanut paste (Enache et al., 2015) , pet food (Bianchini et al., 2014; Ceylan & Bautista, 2015) , hamburger dough (Channaiah et al., 2016) , wheat flour (Villa - Rojas et al. , 2017 ; Liu et al. , 2018) , whole black peppercorn (Wei et al., 2018) , macadamia and cashew (Saunders et al. , 2018) , cocoa powder (Tsai et al., 2019) , cumin seeds (Chen et al., 2019) , corn flour (Ozturk et al., 2019) , and paprika powder (Ozturk et al., 2020) . While previous studies focused solely on one type of food composition, Rachon et al. ( 2016 ) and Shah et al. ( 2017) have performed validation studies using several LMF to represent multiple food compositio ns w i thin a single study . Due to variations in methodology, which influenced the thermal resistance of Salmonella , it is challenging to draw a consensus as to what exten t E. faecium NRRL B - 2354 is a n acceptable surrogate microorganism for Salmonella . There fore, t he overall goal of this study was to determine the acceptability of E. faecium NRRL B - 2354 as a Salmonella surrogate for LMF . Th e specific aims of this research were as follows: 3 i ) Determine the influence of talc powder as a dry - inocul um carrier on thermal resistance of E. faecium i n almond meal at 0.45 a w ; ii ) Q uantify the effect of temperature and product composition on thermal resistance of E. faecium and Salmonella for almond meal, date paste, wheat flour, peanut butter, ground black pepper , and non - fat dried milk powder ; iii ) Determine the effect of milk protein and lactose composition on thermal resistance of E. faecium and Salmonella in dairy powders. The research hypotheses were th at: i ) Talc powders influences thermal resistance of E. faecium on almond meal ; ii ) E. faecium is more thermally resistant than Salmonella across all low - moisture products , and the results are reproducible across laborat ories. Pr oduct composition affect s the thermal resistance of both microorganisms. iii ) Changes in protein and lactose content in dairy powders affect thermal resistance of both Salmonella and E. faecium 4 2. LITERATURE REVIEW 2.1 Low - moisture food s (LMF) Water activity (a w ) is the ratio of the vapor pressure of the food to the vapor pressure of pure water at the same temperature, assuming an equilibrium state of the food with the surrounding atmosphere. Based on physical chemis try, the definition of w ater activity can be further elucidated as equal to the ratio of the fugacity of the water to its fugacity under reference conditions (f 0 ). The fugacity (f) is the real gas equivalent of an ideal gas's partial pressure and may be thought of as the escaping tendency of a substance where f 0 is escaping tendency of pure material (Sippola & Taskinen, 2018) . The range of a w of a product is between 0 (total absence ) to 1 (pure water) (Labuza, 1975) . The growth of mos t pathogenic bacteria, including Salmonella , is hindered below 0.9 5 a w (Labuza, 1975) . Although bacteria cannot grow at low a w , chemical reactions such as lipid oxidation, non - enzymatic browning, enzyme hydrolysis, sucrose hydrolysis, and ascorbic acid degradation may still occur during storage (Adams et al. 2016) . Thus, a w is used as a quantitative measurement for estimat ing the shelf - life of food s based on microbial physical changes. Foods with a water activity (a w ) of 0.85 and below are d efined as low - moisture foods (LMF) (Codex Alimentarius, 2015; U.S. Food and Drug Administration, 2014) which cover s nuts and nut products, edible seeds, cereal s and grains , spices and herbs, confection e ry powders , milk - based powders, chocolate , dried fruits an d pet food s . Dried meats and fish products can be considered as LMF; however, these products are excluded in th e context of this literature review because the a w range of these products can reach above 0.85 and is highly variable, depending on the product formulation and preparation method (Food and Agriculture Organization of the United Nations and World Health Organization, 2014) . Moreover, dried meat product s are a subset of 5 RTE meat products , in which the s afety guideline of RTE meat products against Salmonella hazard has been established by USDA (U.S. Department of Agriculture, 2017) . I t is important to highlight that the term low - moisture foods and low water activity foods have been used in terchangeably in the literature . H owever, the term low - moisture foods is adopted throughout th is dissertation for consistency . The a w values for selected LMF products are shown in Table 2.1. Table 2. 1 Water activity of low - moisture food ingredients/ product s at specified temperature adapt ed from (Schmidt and Fontana Jr. , 2007) Product Water activity, a w Temp erature (°C) Jam (orange, strawberry, peaches) 0.83 25 Honey 0.55 25 Raisins 0.53 22 Black pepper (ground) 0.72 25 Cinnamon (ground) 0.59 25 Almond (sliced) 0.48 20 Walnut 0.43 25 Peanut (salted) 0.26 22 Peanut butter (creamy) 0.20 - 0.26 22 Wheat grain (whole) 0.68 - 0.70 25 All - purpose flour 0.45 20 Nonfat dry milk powder 0.14 30 6 2.2 Salmonella Salmonella are rod - shaped, non - spore forming, and Gram - negative members of the Family Enterobacteriaceae. They are also facultative anaerobes, with peritrichous flagella (Ryan et al. 2017) . Salmonella can grow at temperature s ranging from 5 to 47 ° C (optimum 37 ° C), pH values between 4 and 9 (optimum ~pH 6.5 to 7.5) (Graziani et al., 2017) and water activity values of 0.93 or higher (Adams et al. 2016) . Major reservoirs for Salmonella are domestic animals such as poultry, cattle, and pigs and cold - blooded animals such as reptiles and rodents (Heredia & García, 2018) . Salmonellosis, the illness caused by Salmonella e, can be life - threatening, particularly in immunocompromised groups such as the elderly and children (<5 years old), due to excessive loss of water. However, most of the symptoms are relatively mild, which include acute onset of fever, dizziness, nausea, and abdominal cramping. Upon ingestion of Salmonella (as low as one cell), the symptoms of illness may appear 12 to 36 hours later and persist for up to 7 days (World Health Organization 2018) . Globally, it was estimated that 80.3 million cases of foodborne related gastroenteritis are caused by nontyphoidal Salmonella each year (Majowicz et al., 2010) . According to the most recent Kauffmann - White - Le Minor scheme, there are at least 2659 Salmonella serovars, with 1586 serovars belonging to subsp. enterica (I) (Issenhuth - Jeanjean et al., 2014) . Salmonella serovars are distinguished based on the composition of the surface antigens; oligopolysaccharides (O) and flagellar proteins (H) (Ryan et al., 2017) . Salmonella serovars within subsp. enterica are mostly responsible for human - related illnesses, and one of the most c ommon nontyphoidal Salmonella serovars can be transmitted from animals to humans is Salmonella enterica subsp. enterica Enteritidis or abbreviated as Salmonella Enteritidis (World Hea lt h Organization 2018) . 7 In the U.S., nontyphoidal Salmonella is the second dominant pathogen causing foodborne illnesses (1 million out of 9.4 million cases) and is the leading pathog en that cause hospitalization each year (Scallan et al., 2011) . Although chicken, eggs and seeded vegetables were identified as common food vehicles, LMF such as Turkish pine nuts and chia seed powder, were recognized as novel sources of salmonellosis (Dewey - Mattia, Manikonda, Hall, Wise, & Crowe, 2018) . For LMF, a study by Young et al. (2015 ) found that Salmonella was linked to nearly 45% of LMF outbreaks reported in the North America and Europe between 1953 to 2013 . 2.2.1 Survival mechanism of Salm onella in the desiccated state Salmonella is capable of surviv ing in peanut butter (0.24 a w ) and chocolate cookie fillings (0.38 a w ) (Beuchat & Mann, 2015) as well as in different compositions of peanut butter and peanut butter spread for up to 6 month s (Burnett et al., 2000) . Salmonella can also survive up to 12 months in peanut butter fondant candy during storage at room temperature (Nummer et al., 2012) . Despite the high - sugar content , which was previously thought to cause osmotic shock to Salmonella , the high - fat content in peanut butter may provide a protective barrier for its survival. Salmonella was also detected in dried fruits ( e.g., cranberries, date paste and raisins) for up to 8 months i ndicat ing that Salmonella can adapt to acidic environment s such as in cranberries (pH 2.52), in addition to low - a w (B euchat & Mann , 2014) . Salmonella survival was also described in milk powder ( Lian et al. , 2015) . In other instances, Salmonella was shown to survive on products contai ning natural antimicrobial compounds such as ground black pepper (Keller et al., 2013) . Despite low - a w , high tolerance of Salmonella in LMF products is concernin g because LMF products have extended period of shelf - life and there are no obvious signs of contamination for consumer to notice ( Santillana & Frank , 2014) . 8 The ability of Salmonella to withstand dry environment s could be driven by synthesis and uptake of osmoprotectant molecules which help to equilibrate cytoplasmic osmotic pressure. A model proposed by Finn et al. ( 2013) suggested proline, glycine - betaine , and ectoine (osmoprotectants solutes) concentrations increased when pro U and osmU genes were up - regulated . In addition, proP gene expression, particularly for betaine uptake, was also expressed and this gene was crucial for long - term persistence of Salmonella in dry environments (Cairney et al., 1985) . Disaccharides such as trehalose, which are stimulate d by fatty acid catabolism and sigma factors, wer e also produced (Finn et al., 2013) . In response to high os motic pressure in the cytoplasm , OmpF and OmpC, two porins located at the outer cell membrane, begin to passive ly diffus e osm oprotectants (Balaji, Connor, Lucas, Anderson, & Csonka, 2005) . Additionally, the cell membrane was observed to adjust its trans to cis unsaturated fatty acid ratio to accommodate low - moisture stress (Cronan, 2012). As desiccation progresses, Salmonella is transformed into viable but nonculturable (VBNC) state (Gruzdev et al. , 2012) with ribosomal RNA molecules degraded as a source of nutrient s (Deng et al. , 2012) . Other responses such as production of filament s , induced by inhibition of cell division, as well as protective glycolyx layers during desiccation have been correlated with extended survival of desiccated Salmonella (Mattick et al., 2001) . Lastly , Salmonella has the ability to modify its genetic machinery to accommodate desiccated conditions, which imply that Salmonella might not be detectable in LMF products because they typically pose an extended shelf - life (Burgess et al., 2016) . 9 2.3 Salmonella in specific commodity of LMF LMF covers a broad range of composition s that can be used as raw ingredients or consumed as ready - to - eat (RTE) product s . Salmonella can contaminate LMF at any point of processing via cross - contamination with the environment during harvest, poorly maintained facilities and equipment, and inadequate hy giene practices . Considering increas ed trend of Salmonella outbreaks /recalls linked to LMF, Food and Agriculture Organization of the United Nations (FAO) and World Health Organization (WHO) established the ranking of LMF for microbiological risk assessment in 2014 (Food and Agriculture Organization of the United Nations and World Health Organization, 2014) . Based on FAO/WHO preliminary report, s everal different LMF food categories have been adapted to represent high - fat and high - protein ( nuts and tree nuts ), high - carbohydrate ( cereal and grains ), high - protein and high sugar (dairy powders), high - sugar ( dried fruits ), and antimicrobial - rich (black pepper) products are discussed below. 2.3.1 Nuts and tree - nuts U.S. is the leading producer of tree nuts across the globe (International Nut and Dried Fruit, 2018) . T he U. S . export value of tree nuts increased from $7.9 billion to $8.5 billion in 2016/2017 (U.S. Department of Agriculture, 2019) , suggesting that this increase is a positive indicator of market growth. A national survey reported that 38% of U. S . adults in the period of 2009 to 2010 and 32% of U. S . youths (2 - 19 years old) in the period of 2009 to 2012, consumed nuts or nut butters (as an individual - item food) on a daily basis (Nielsen et al., 2016; Nielsen et al. , 2014) . Tree nuts (e.g., almond, pecan, hazelnut, walnut, pistachio, and macadamia) and ground nuts (peanut) are considered as a nutritious source of protein and unsaturated fat. 10 Major Salmonella outbreaks in the nut and tree - nut industry have involved peanut butter and almonds . Salmonella Enteritidis outbreaks linked to raw al monds in 2004 (Centers for Disease Control and Prevention, 2004) involved 12 states in the U.S. and Canada and resulted the recalls of 13 million pounds of raw almonds. In 2006, peanut butter (Peter Pan and Great Value brands) contaminated with Salmonella Tennessee sickened 425 persons, with 71 out of 351 patients hospitalized (Centers for Disease Control and Prevention, 2007) . Later in 2009, Salmonella Typhimurium tainted King Nut creamy peanut butter and products contai ning peanut butter of the same brand were recalled (Centers for Disease Control and Prevention, 2009) . This outbreak resulted in 714 cases of illness including 9 deaths. The source of Salmonella contamination for both peanut butter outbreaks was traced to processing facilities in Georgia (Centers for Disease Control and Preve ntion, 2009) . A series of peanut butter and nut butter outbreaks linked to Salmonella from 2001 to 2019 were comprehensively itemized by Yada & Harri s ( 2018) , which impl ies recurring contamination events associated with Salmonella in nut - products. Salmonella can contaminate peanut butter and almonds at any point of manufacturing stages . For pe anut butter production , shelled peanuts are surface - cleaned to remove dirt and foreign materials, graded , and stored after harvest (Calhoun et al., 2018) . Shelled p eanut s are then roasted ( 160 ° C / 40 to 60 min ) to establish desirable flavors , cooled , dry - blanched at (135 ° C / 25 min) , and de - shelled via brushing. Peanut kernels are then ground and packed a t 48 ° C (Woodroof, 1983) . After de - shelling, cross - contamination with water droplet s , dust, personnel, processing equipment, and food packaging materia ls can introduce Salmonella into peanut butter (Grasso - Kelley, 2016) . It is also possible that the addition of other ingredients , such as sweete ner and emulsifier , after peanut roasting can increase the entry of Salmonella in peanut butter , if those ingredients are untreated. Although m uch efforts have been emphasized on implementing proper hygiene 11 practices in the peanut processing facilities, an intervention during post - processing of peanuts should be included to reduce recurring outbreaks. For raw a lmonds , soil can be one of the sources of Salmonella , because the almonds are typically harvested using a mechanical shaker that allows the almonds to fall on the ground . Raw a lmonds are then de - hulled , sorted, pack ed, and stored. A lmonds can be consumed and inc orporated into food products in numerous forms, including chopped, sliced, meal, and flour . For instance, a lmond meal can be contaminated with Salmonella during the grinding of almonds or during almond meal packaging . Salmonella can be introduced in to the finished baked goods such as macaroons, cookies, and granola bars , because baked goods are minimally heat - treated. While most Salmonella inactivation strategies have been focused on whole almonds, there is still limited studies on various form of almonds such as almond meal. 2.3.2 Dried fruits In 2014, n early 60 ,000 Kirkland Signature dried fruit snack s were voluntarily recalled due to possible contamination of Salmonella . Although no confirmed illness es were reporte d, this reca - brand snack for Costco (Anonymous, 2014) . This is concerning because d ried fruits such as dates, prunes and dried apricots are consumed as ready - to - eat snacks for their die tary fibers , natural sugars , as well as antioxidant compounds, including p olyphenols, carotenoids, and tocopherols (Hernández - Alonso, Camacho - Barcia, Bulló, & Salas - Salvadó, 2017) . Dates , particularly, offer versatil e food applications given that there are at least 10 variet ies of date cultivars . In addition, date fruits undergo several ripening stages, w hich distinct physiochemical profile develop s at each ripening stage (Ghnimi et al., 2017) . Dried d ates contain 12 at least 66 to 88 % of carbohydrate s which mostly fructose, glucose, and sucrose (Ahmed & Al - jasass, 2014) . Dates can be consumed as snacks, juices, and jam, or incorporated into granola bars, cereals, and baked - good as added - value nutrients. The U. S . import value for dates alone reached nearly $1 billion in 2017/2018 (U.S. Department of Agriculture, 2019) , with the majority of dates imported from Iran and Saudi Arabia (Internationa l Nut and Dried Fruit, 2018) . After harves t , dates are hand - sorted to remove damaged dates, pitted, and fumigated with methyl bromide gas for pest control. Heat treatment can also be used only if the dates are converted into other types of date products, such as , d ate paste . Date paste are typically used as fillings in bakery products. For date paste, p itted dates that are commonly with lesser quality are soaked (95 ° C / ~ 15 s) or steamed (69 kPa/3 min) before grinding (Ashraf & Hamidi - Esfahani, 2011). Cross - contamination with so il, dirt, workers, and equipment can increase the chances of Salmonella entry into date paste. Salmonella exposure to sub - lethal injury conditions ( minimal heat treatment and osmotic stress due to high - sugar environment ) , may yield to greater resistance to the subsequent heat treatment, such as baking (Podolak et al,. 2010) . Given that Salmonella can survive up to 8 months in date paste (Beuchat & Mann, 2014) on top of absence of post - grinding pasteurization, date paste can be the source of Salmonella in cookies, fruit cakes, and granola bars. 2.3.3 Spices and herbs Spices and herbs serve as a good source of antimicrobial compounds and are typically used as natural food preservatives (Adams et al., 2016) . Spices such as black pepper, and mixed herb powder s have desirable aroma and flavor, hence e xtensively used as meat and poultry seasonings or added raw as condiment . A recent survey by Isbill et al. (2018) revealed that 54% of U.S. adults in the Midwest region (n total = 703) were using at least one type of spice on a daily basis, with black 13 pepper one of the most commonly used spices among 10 spices surveyed. Spices are imported from India, with the U.S. im port value of spices representing 17% of the global import value (Nguyen et al. 2019) . Despite antimicrobial - rich compounds in spices, Salmonella has been detected in variety of spices including black pepper, paprika, ground oregano, ginger, garlic powder, mustard seeds as documented in the risk profile of pa thogens and filth in spices by U.S. Food and Drug Administration (2017c) . Black pepper is one of the spices that has been involved in a nationwide Salmonella outbreak. The prese nce of Salmonella Montevideo in the salami seasoned with contaminated black and red peppercorns has affected 272 persons from 44 state s . The peppercorn was imported from a single supplier and used in many products. As a result, 1.2 million pounds of peppercorns were recalled due to this outbreak (Centers for Disease Control and Prevention, 2010) . In the black pepper production, immature pepper corn (gree n in color) are briefly blanched in hot water (80 ° C / 2 min) to clean the berries and to stimul at e the browning enzymes (Pruthi, 1992) . T he peppercorns are then dried on bamboo mat, cement floor or polyethylene fabric . Appearance of wrinkle d and black color are typically observed at the end of drying step (Balasubramanian et al., 2016) . Salmonella that may be introduced during harvesting may survive during minimal heat treatment of unripe peppercorn , hence enhance Salmonella persistence during drying stage. In addition, cross - contamination with dirt, dust, and soil during drying and with the equipment during gr indin g (for coarsely and finely ground) can be the sources of Salmonella in ready - to - eat foods. Thus, black pepper manufacturers need to perform p ost - processing lethality step to ensure the safety of raw black peppers. 14 2.3.4 Cereal and grains Wheat is a commonly grown crop in the U.S . and composed of 70 - 75% of starch, 11 - 15% protein, ~3% lipids, and a considerable amount of minerals and vitamins (Gallaher & Anderson, 2019) . Wheat flour is a n essential raw ingredient for b aked goods such as cereals, breads, pastries, biscuits, cak es, crackers, and cookies, as well as noodle products. In addition to wheat, other types of grains such as oat, corn, rice, and barley are also typically ground to make flour. In recent years, Salmonella outbreaks/recalls linked to flours and flour - based products ha ve posed a great concern to the public . In 2018, was implicated with Salmonella Mbandaka outbreak which caused 100 cases and 30 hospitalizations from 33 states (Centers for Disease Control and Prevention, 2018b) . In 2019, Duncan Hines Classic White Cake mixes were recalled by Conagra Brands , and Pillsbury unbleached all - purpose flour with expiration date of April 2020, was voluntarily recalled by Hometown Food Company due to potential Salmonella contamination (U.S. Food and Drug Administration, 2019a, 2019b) . Production of wheat flour involves wheat storing, cleaning, conditioning, milling, and sifting (Morris & Rose, 1996) . During storage, wheat grains are sometimes f umigat ed to kill insects and rodents but not bacteria. W heat grains are cleaned and purified by multiple sieving to discard any impurities . Next , t he grains are tempered by soak ing in sanitized water for at most 24 h t o facilitat e the removal of the germ and bran from the inner endosperm during milling . Adding moisture to the grains during tempering can promote Salmonella growth , and the subsequent milling process can most likely dis seminate Salmonella to the entire lot of flour. G iven that t here is no known pasteurization step for either wheat kernel or wheat flour , prior to distribution , such processing setting clearly pose Salmonella risk to consumer s , especially when no heat treatment is needed for making the end - product , such as raw cookie dough . Therefore, there is an urgent 15 need for post - processing pasteurization of whole kernels or wheat flour to reduce the occurrence of Salmonella outbreak. 2.3.5 D airy powders Milk is a nutrient - packed liquid that is spray - dried into powders for a longer shelf - life. Dairy powders are low - moisture ingredients that are used in food formulations or instantly reconstituted as beverage . D airy powders are commercially manufactured to c ompose different levels of protein, fat, and sugars to serve desired functional properties , including flowability, solubility, and dispersibility of the food products (Sharma et al., 2012) . It is also important to highlight that the alteration of milk components to manufacture dairy powder variants is commonly tak ing place after milk pasteurization, in which c ross - contamination with equipment, personnel, and under - maintained facilities can introduce Salmonella into the dairy powders. For instance, Salmonella Agona contaminated powdered infant formula sickened 39 infants (< 1 year old) in 2018 from France, Greece and Spain (European Food Safety Authori ty, 2018) . A broad product recall was issued by the French manufacturing company, including products other than infant formula, involving 67 countries. The isolated Salmonella strain was identical to the strain found in 2004 - 2005 infant formula outbreak (European Food Safety Authority, 2018) . Given that there is no pathogen lethality step after spray - drying, consumption of milk powders contaminated during post - processing may lead to s almonellosis , especially in immunocompromised groups such as children and the elderly. Overall, LMF makes up a pivotal commodity in the U.S. because of their nutritional values and their extens ive applications in the food formulation s . P revious Salmonella outbreak s/recalls linked to LMF s were most likely derived from Salmonella entry via cross - contamination with the 16 environment, equipment, and personnel during LMF processing . T he a ddition of moisture into LMFs during processing undoubtedly can promote Salmonella growth . Although Salmonella contamination level in LMFs was reasonably low, adaptive capacity of Salmonella in dry environment allow them to remain in the dormant state (Finn et al., 2013) . Salmonella exposure towards minimal heat treatment may also enhance Salmonella resistance to a subsequent kill step , if any . With this understanding, a practical food safety intervention is to validat e Salmonella lethality steps i n current LMF processing to ensure the safety of LMF ingredients. In this dissertation, specific LMF product s , including peanut butter and almond meal (high - fat, high - pro tein) , wheat flour (high - carbohydrate), date paste (high - sugar) , non - fat dried milk powder (high protein, high sugar), and ground black pepper (antimicrobial - rich ) , were selected to cover a myriad composition of LMF materials that have been implicated with Salmonella outbreaks/recalls. Assessing multiple LMF compositions in a single validation study allow ed a better understanding of how food composition influence s the thermal resistance of Salmonella in LMF . To achieve the most effective validation strategy , other f actors that can impact thermal resistance of Salmonella are briefly discussed because these factors can also impact the thermal resistance of a surrogate microorganism that is purposely used as a bi ological validation tool in the thermal processing of LMF . 17 2. 4 Factors influencing thermal resistance of Salmonella T hermal resistance of Salmonella is influence d b y both intrinsic and extrinsic factors that are strain - , product - and process - dependent . It is important that these factors are identified before conducting validation studies for thermal processing of low - moisture foods. P revious studies demonstrated that several factors , such as temperature, water activity , inoculation protocol, recovery media, and food composition , are among important parameters affecting thermal resistance of Salmonella , which are further elaborated in the sections below. 2. 4 .1 Treatment temperature Treatment temperature is an extrinsic parameter that can impact the inactivation trend of Salmonella . Salmonella inactivation curves can be linear or nonlinear concave upwards depending on the treatment temperature ( Villa - R ojas et al. , 2013) . In a study by Rachon et al. ( 2016) , a linear relationship was observed during thermal inactivation of Salmonella in seasonings at 80°C , while nonlinear concave upwards curves were reported for confectionery powder at 100°C . Based on our visual assessment, inactivation curves for confectionery powder at 100°C indicate a linear trend if survivors at the initial treatment time are not plotte d. In another study, an upwardly concave curve with asymptomatic tailing was observed for Salmonella inactivatio n in peanut butter at 70 to 90°C (Shachar & Yaron, 2006) . The tailing effect indicated that increasing treatment time may not effective ly eliminate Salmonella in the LMF processing for pean ut butter. Most importantly, extrapolatio n of linear Salmonella inactivation data outside experimental parameters poses a major risk , since thermal processing of LMF must be validate under the actual processing conditions used 18 2. 4 .2 Water activity Water activity is one of the intrinsic factors that can affec t Salmonella thermal resistance, where p re - exposure of Salmonella to dry environment can increase its thermal resistance. For instance, Salmonella was found to be more thermally resistant in whe at flour equilibrated at 0.3 a w as compared to 0.6 a w , regardless of the time required for equilibration (Smith & Marks, 2015) . Increased Salmonella resistance has been observed in a wide range of LMF , including almond s (Jeong et al. 2011) , almond flour (Villa - Rojas et al. 2013), peanut paste (Enache et al., 2014) , pet food (Bianchini et al., 2014) , and hamburger dough (Channaiah et al., 2016) . One possible reason could be related to microscopic water distribution (Syamaladevi et al. , 2016) . It is believed that the vibration of water molecules i n high a w environment s disrupts the cell membrane of Salmonella during heating (Pena - Melendez et al. , 2014) ; th erefore , the lower availability of water molecules in low a w foods might lead to less cell membr ane disruption, and therefore increased thermal resistance. 2.4. 3 Recovery media and enumeration Quantification of Salmonella survivors can be affected by recovery media. Nonselective media, such as modified tryptic soy agar supplemented with yeast extract (TSAYE) allow the growth of injured Salmonella survivors, whereas, selective media, such as xylose lysine desoxycholate agar (XLD), are more restrictive for survivor growth (Busta et al., 2003) . Use of nonselective and selective media for recovery of hea lthy and heat - treated Salmonella in LMF has been previously reported (Channaiah et al. 2016; J in et al. , 2018) . One possible implication of using selective media is the underestimation of survivors after a given inactivation treatment. However, an enrichment step for Salmonella recovery may lead to an overestimation of Salmonella 19 survivors. Given that recovery media and enumeration methods could influence Salmonella resistance, higher accuracy of survivor counts need to be considered before conducting LMF thermal processing validation studies. 2.4. 4 Inoculation metho d Ideally, in vitro Salmonella preparation should emulate the physiological state of Salmonella at the point of LMF contamination . The d ifferent physiological state s of Salmonella ( planktonic, colon y or biofil m ) can be ac hie ved via culture preparation ; b roth - grown Salmonella ( planktonic state ) , agar - grown Salmonella ( sessile state ), and living/ inert surface - grown Salmonella ( biofilm state ) (Aviles et al., 2013; Keller et al., 2012) . Typically, Tryptic Soy A gar (TSA) and Tryptic Soy B roth (TSB) are used to gr ow Salmonella . Inclusion of Salmonella in an aqueous culture is typically used for inoculation (Bowman et al. , 2015; Channaiah et al., 2016; Keller et al., 2013; Liu et al., 2018; Rachon et al., 2016; Tsai et al., 2019; Wei et al., 2018) . In contrast , dry inoc ulation involves the use of wet - inocu lated dry carriers , such as chalk (Beuchat & Mann, 2011) , sand (Blessington et al. , 2013) , silica beads (Hildebrandt et al., 2017) , and talc powder (Enache et al., 2015) to exclude post - inoculation drying and to limit physic al alteration of the LMF matrix . D ry - carrier material s shoul d be able to maintain high population levels, be easily removed from the food matrix without any residue, and not influence the thermal resistance of Salmonella in LMF. So far, there is no known inert material that can satisfy all the characteristics of an ideal carrier. For Salmonella heat - inactivation studies, a high initial population is desirable to ensure that the survivor counts are above the limit of detection, which typically is achieved using the wet inoculum method. 20 Salmonella stability in LMF during equilibration will vary based on the culture prepa ration method s (Aviles et al. , 2013; Bowman et al. , 2015; Wiertzema et al., 2019) . Salmonella in dried milk powder was more stable in a biofilm state (1.6 log reduction) as op posed to the planktonic state (5.4 log reduction) after 30 days of storage at an a w of 0.30 (Aviles et. al , 2013) . Likewise, Bowman et al. (2015) showed that Salmonella populations on peppercorn were most stable in the biofilm state followed by TSA and TSB - grown Salmonella after 28 days of storage at an a w of 0.30. In other instances , the greatest de crease s in Salmonella and E. faecium in wheat flour, milk powder and soy powder , were observed using broth - grown as opposed to as compared to agar - grown cultures ( Wiertzema et al., 2019) . I noculation protocols are product dependent and variations in executing the inoculation step can certainly influence thermal resistance of Salmonella in LMF. The suspension medium used for harvesting agar - grown Salmonella can impact the the rmal resistance of Salmonella . For example, t he inactivation of Salmonella in peanut butter required a longer heating time pronounced tailing of the inactivation curve observed when agar - grown cells were harvested into peptone water supplemented with an oil suspension as compared to a suspension without oil ( Li et al. 2014) . Previous studies have also used oil in the culture suspension to retain low a w condition s . ( He et al. 2011; Keller et al., 2012 ) . In other instance s , the use of agar - grown Salmonella harvested in or as a pellet yielded reproducible Salmonella thermotolerance data in wheat flour across two laboratories, as compared to using other culture prepara tions including broth - grown Salmonella ( Hildebrandt et al. 2016) . Despite using the same culture preparation method , Limcharoenchat et al. (2018) adding the inoculum during pre - fab rication of almond products (whole, meal, and butter), resulted in greater Salmonella resistance than those during post - fabricat ion , however the reverse d was seen for wheat products (kernel and flour). 21 Dry transfer of Salmonella to LMF h as been previously performed using chalk to inoculate peca ns (Beuchat & Mann, 2011) , and sand to inoculate almonds and walnuts (Blessington et al., 2013) , and silica beads to inoculate spices (Hildebrandt et al., 2017) , and t alc powder to inoculate peanut paste (Enache et al., 2015) . T he initial Salmonella load in LMF using dry a nd wet inoculation is carrier - and product - dependent . T he initial population in dry - inoculated pecan s , which used chalk as dry carrier, was reported to be ~6.7 CFU/g, comparable to that seen for wet - inoculated pecans (Beuchat & Ma nn, 2011) . S and - inoculated nut kernels yielded lower initial population s of Salmonella (~5 log CFU/g) as compared to wet inoculation (~8 log CFU/g) (Blessington et al., 2013) . Salmonella population s in ground cloves and oregano, but not ground black pepper or ground ginger, were greater using dry silica beads, rather than an aqueo us inoculum (Hildebrandt et al., 2017) . For spices, t he addition of moisture could stimulate inherent antimicrobial bioactive compounds during inoculation and sample recovery , thus reducing the initial population of Salmonella , with this observ ation being spice dependent. It is important to note that sand and silica beads that were used in the previous studies were removed from the matrix by sieving . Neither talc (magnesium s ilicate ) no r chalk (calcium carbonate) powder were not removed from the matrix after dry inoculation (Beuchat & Mann, 2011; Enache et al., 2015) and were utilized to achieve both a high level of in itial inoculum and homogenous bacterial distribution (Hoffmans & Fung, 1992) . While t h e dry carrier residue in the LMF during he at treatment will mo st likely impact the thermal resistance of Salmonella , but the influence of the dry carrier has not yet been evaluated . In dry - inoculated pecans, Salmonella was found to exhibit a lower D - value than those for wet - inoculated pecans (Beuchat & Mann, 2011) . The authors explained that lower Salmonella heat resistance in dry - inoculated pecan s was due to the location of Salmonella on the 22 nut surface , being more exposed to the heat as compared to those for wet - inoculated pecan. H owever, the lower heat resistance of Salmonella was more likely due to the presence of chalk coating the surface of the pecans. In another study, E. faecium was ~3 times more heat resistant than Salmonella in dry - inoculated peanut paste (50% fat, 0.6 a w ). While both wet and dry inoculation represent potential transmission route s for Salmonella in LMF , none of the above studies evaluated the influence of dry carrier s on thermal resistance of Salmonella , which could lead to over predictions of thermal resistance for both Salmonella and its potential surrogate. Therefore, there is a need to further understan d the most practical inoculation protocol prior to conducting va lidation thermal processe s for LMF using E. faecium NRRL B - 2354. 2. 4 . 5 Food composition and structure LMF that were commonly associated with Salmonella outbreaks/recalls comp rise of a wide range of products that can be grouped based on their major macronutrient component s as discussed in section 2.3. In the literature, t he effect of food compositions on thermal resistance of Salmonella in LMF ha s been investigated but a fundamental understanding of how ositions influence the thermal resistance of Salmonella is still lacking. One of the earlie st studies evaluated the influence of different humectants (sucrose, glucose - fructose, and NaCl) employed in low - a w broth on thermal inactivation of Salmonella ( 55 - 74 °C ) (Mattick et al., 2001) . When compared with other humectant s , t he term protective effect was used to describ es humectant s that promot ed Salmonella heat resistance under specif ic conditio ns . A t 0.90 relative vapor pressure (rvp ) (expression of a w in a diluted system ) f or example, Salmonella in s ucrose broth afforded more thermal protection to Salmonella compared glucose - fructose , as well as another that contained NaCl . The p rotective effect of humectants ( either 23 sucrose, glycerol, or NaCl) o n Salmonella heat resistance w as als o assessed by Pena - Melendez et al. (2014) , who specifically examined whether osmotic shock or bacterial cell habituation in low - a w Tryptic soy broth (TSB) significantly affected Salmonella heat resistance at 55 °C . Similar ly, Mattick et al., ( 2001) found that su crose was more than NaCl in a low - a w system . However, there was no further explanation as to why sucrose, as compared to other humectants, stimulat ed increased Salmonella resistance during heat ing . U sing peanut butter as a food model, previous studies have assessed the effect of fat and carbohydrate levels on thermal resistance of Salmonella . P eanut butter is considered as high - fat LMF , but commercially available p eanut butter products tend to contain a significant amount of carbohydrate (as sweetener) in the formulation, especially in regular and reduced - fat peanut butter. He et al. (2011) demonstrated that Salmonella in regular peanut butt er ( 33% fat , 0.4 a w ) exhibited higher heat resistance than Salmonella in reduced - fat peanut butter ( 6.3% fat , 0.7 a w ) , despite a similar carbohydrate content (~40%) . Salmonella also exhibited similar thermal resistance in samples containing 33% or 50% fat. In another study, a decreas ing Salmonella thermal resistance trend in four different compositions of peanut butter ( 70 - 90 °C , ~0.4 a w ) as follows : reduced - fat (33% fat , 42% carb ohydrate ) > regular (50% fat, 22% carbohydrate) , reduced suga r (53% fat, 19% carbohydrate) > omega - 3 peanut butter (49% fat, 24% carbohydrate) ( Li et al. , 2014) . In another study, Enache et al. ( 2014) tested 16 peanut paste formulation s composed of varying fat content (47% to 56%) and a w levels (0.3 to 0.6 a w ) . Overall, Salmonella was mos t thermally resistant in formulations containing 47% regardless of a w level as compared to peanut paste with higher fat content ( at 85 - 90 °C ) . This observation might be due to the lower carbohydrate level in organic peanut butter (22% carbohydrate) , given that both organic and regular peanut butter were similar in a w (0.4 a w ) as compared to regular and reduced - fat peanut butter which differed in a w . 24 P revious studies have shown that for different fat and carbohydrate level s within a sing le LMF ingredient (like peanut butter) , Salmonella t hermotolerance was observed to shift when high - fa t and high - carbohydrate component s were represented by two different LMF ingredient s . For instance, Salmonella exhibited higher thermal resistant in peanut butter ( 50 % fat, 20% protein, 20% carbohydrate ) than all - purpose wheat flour ( 70% carbohydrate, 10 % protein ) at 80 °C , 0.45 a w (Syamaladevi et al., 2016) . Limcharoenchat et al. ( 2019) evaluated the effect of structure (whole, particulate, and paste) on thermal resistance of Salmonella using several LMF ingredients; almond products, wheat products , and date products . The authors observed greater Salmonella heat resistance in almond meal and butter, than in whole almonds; in date paste than in date pieces; and in wheat kernel than in wheat flour . The authors also found that Salmonella in almond products (high - fat) showed greater resistance than in wheat (high - carbohydrate) and date products (high - sugar). In a multi - ingredient LMF matrix, the tr end in Salmonella thermal resistance can be complicated . For example, L i et al. ( 2014) mentioned th at contamination source s affect ed the heat resistance of Salmonella represented in the binary LMF matrix ( nonfat dry milk powder , peanut butter ) and quaternary LMF matrix (cookies: sugar, butter, flour, and egg). For the binary LMF matrix treated at 90 °C , Salmonella exhibited decreased heat resistance when contaminat ion occurred in nonfat dried milk powder as compared to those from contamination occurred in peanut butter . Further, when cookies prepared with contaminated flour , Salmonella was least heat resistance at w hen baked at 177°C as opposed to when present in the other ingredients (egg, butter , and sugar), which in case Salmonella thermotolerance was statistically similar. In term of explaining the effect of food compositions, no correlation could be made between the binary (food model) and quaternary models , because the ingredients used to compose those food models were 25 completely different. Later, Jin et al. ( 2018) assessed t wo food models, differing in protein (34% protein, 9% fat) and fat content (17% protein, 27% fat), which included wheat flour, soy powder, and soybean oil . At 79.5 °C (0.63 a w ) Salmonella was similar ly heat resistan t in high - protein and high - fat samples with greater resistance than in high - fat matrix and lower resistance than high - fat matrix, seen above and below 79.5 °C . Several studies (Li et al., 2014; Nummer et al., 2012; Syamaladevi et al., 2016) hypothesized that water mole cule s distribution within the food matrix play s an important role in Salmonella inactivation during heat ing . In the presence of water, possible inactivation mechanism of Salmonella is most likely caused by the v ibration of water molecules that lead to cell membrane disruption during thermal treatment (Tapia et al. 2007) . Li et al. ( 2014) described the no nhomogeneous water , fat, and protein distribution in a nonfat dry milk: peanut butter matrix using an attenuated total reflection Fourier IR spectrophotometer (ATR - FT - IR ) imaging. W ater molecules partitioned in the high - lipid region , due to water and lipid phase separation , thereby r educ ing the number of water molecule s available for Salmonella inactivation . Conversely, dissemination of water molecules within the spatial locations of milk proteins , resulted in more water molecules available in the food system , hence p romoting Salmonella lethality in the high - protein matrix . Another study proposed that w ater activity ( a w ) changes in LMF ingredients at elevated temperature could be a key factor in affecting Salmonella thermal resistance in different food compositions ( Syamaladevi et al. , 2016) . Given that a w is temperature - dependent and previous studies only measured a w of LMF at ambient temperature, a thermal cell equipped with relative humidity sensor was used to measure the a w o f peanut butter (high - fat) at 80 ° C . R educed a w of peanut butter (from 0.45 to 0.04 a w ) at 80°C was observed and the author speculated that this 26 change was the reason why Salmonella was more heat resistant in the high - lipid matrix. Syamaladevi et al. ( 2016) also further explained that that fewer water molecules in the LMF matrix caused the bacterial cytoplasm components and cell membrane to become more rigid , resulting in less susceptib ility to heat injury . In the same study, increased a w of all - purpose flour (from 0.45 to 0.80 a w ) at 80°C was also observed, which could explain the l ower Salmonella heat resistance in a high - carbohydrate matrix as compared to those in a high - fat matrix . Using a similar method , Tadapaneni et al. (2017) observed an increased in a w shift in wheat flour, almond flour, and nonfat dry milk powder at elevated temperature for all moisture content s tested. The authors inferred that molecular linkag es between water molecules and hydrophilic groups of the foods experienced temperature - induced breakage , with the magnitude of molecular breakage s depend ing on the numbers of hydrophilic groups available in the foods. Altogether, the effect of food compositions on thermal resistance of Salmonella in LMF has been corroborated wit h the accessibility of water molecules in the a given LMF matrix. Previous studies surmise d that a h igh - fat matrix protected Salmonella , whereas a high - carbohydrate or high - protein matrix offered less protection for Salmonella from heat inactivation. This concept cannot be generalized because the reverse Salmonella lethality trend was observed when carbohydrate and lipid components were within a single LMF ingredient , such as peanut butter. Therefore, t here are several limitations that can be highligh ted from the LMF literature described in section 2.4.5 . First, the knowledge advancement regarding the effect of food compositions on Salmonella thermotolerance has been mostly demonstrated using high - fat matrix as a food model , denoting the importance to explore the impact of other macro - components of the LMF food matrix that have been associated with Salmonella outbreaks/recalls, including high - 27 sugar, high - protein and antimicrobial - rich matri ces . Second , the measurement of a w at elevated temperature needs to be documented since a w is influenced by temperature, however no known devices can be used to accurately measure the changes of a w during thermal treatment. Therefore, the use of a w changes at elevated temperature could not exp lain the magnitude of Salmonella heat resistance in different LMF composition s . To sum up, there are numerous factors affecting Salmonella inactivation in LMF because Salmonella thermal resistance is both product - and process - dependent. U nderstanding the implication of these factors on Salmonella inactivation is crucial for both the food industry and regulators to ensure effective validati on step for LMF thermal processing using a surrogate micro or ganism . H owever , it was difficult to achi eve a strong conclusion due to difference in methodology between previous studies . 2. 5 Enterococcus faecium NRRL B - 2354 as a potential Salmonella surrogate 2.5.1 Desirable characteristics of a surrogate microorganism The Food Safety Modernization Act (FSMA) requires scientifically proven data to demonstrate effective control of microbial hazards for a given processing kill step (U.S. Food and Drug Administration, 2018a) . Micr obial challenge stud ies are one of the validation approaches that food manufacturers can use to ensure the safety of LMF (National Advisory Committee on Microbiological Criteria for Food s, 2010) . Surrogate microorganisms are not naturally present in the food but are introduced in to food products to evaluate the effectiveness of the processing kill steps (Busta et al., 2003) . Although it might seem ideal to utilize the actual pathogens isolated from the outbreaks for validating thermal processing of low - moisture foods, the use of pathogens in food processing faci lities is clearly 28 unacceptable. Given that exposure to foodborne pathogens without comprehensive precautions can be life - threatening, such challenge studies with pathogens must be conducted in biosafety level 2 facilities by trained researchers or profess ionals. (Hu & Gurtler, 2017) . Despite taking precautionary measures, a public health laboratory worker in Saskatchewan, Canada, was diagnose d with enterocolitis, due to Salmonella Typhimurium 14028 infection in 2011 (Alexander et al., 2015) . In the period of 2011 to 2017, the number of Salmonella Typhimurium acquired infections from teaching laboratories in the U.S. were 174 cases, including one death (Centers fo r Disease Control and Prevention, 2012, 2014a, 2017) . Therefore, there is a heightened need to use nonpathogenic surrogate bacteria as an alternative to the target pathogen. A surrogate microorganism is a nonpathogenic strain responding to a given treatment in a manner identical to the pathogenic strain . Busta et al. ( 2003) further elaborated that a suitable surrogate micr oorganism, predominantly for fresh and fresh - cut produce applications, should 1) show similar growth and inactivation trends to that of the target pathogen when exposed to processing conditions, 2) be easy to prepare, 3) remain genetically unchanged, 4) ma intain its population until used, 5) not cause spoilage of the food, 6) be easily distinguished from background microflora, 7) possess equivalent characteristics to the target pathogen in terms of attachment to the food and susceptibility to injury, 8) be easily recovered. Any surrogate strain must have an extensive record of being non - pathogenic, which implies that a thorough search of the literature is needed to select appropriate surrogate microorganisms for validation studies. 2. 5 . 2 Characteristics of Enterococci Enterococci are gram - positive lactic acid bacteria (LAB), that natural ly inhabit the gastrointestinal tract of human s and livestock, such as pigs, cattle, and sheep (Franz, Huch, 29 Abriouel, Holzapfel, & Gálvez, 2011) . Enterococci are omnipresent in the environment and considered resilient microorganisms due to their resistance to a broad range of pH values and temperature s (Gira, 2002) . They are able to grow at 10 ° C to 45 ° C, at pH 9.6, in 6.5% sodium chloride, and survive 30 min of heating at 60 ° C (Schleifer and Killpper - Balz , 1984) . It is not surprising that Enterococcus spp. are present in many adverse environmental niches and used a s indicator microorganisms in the dairy industry (Franz et al. , 1999) . Enterococcus faecium NRRL B - 2354 , also designated as Enterococcus faecium ATCC 8459, ha s an equivocal history. Several papers mention that th is strain was isolated by G.J. Hucker from a dairy processing environment in the 1920s , with the earliest name documented as NRRL B - 2354 ( in the USDA NRRL culture collection ) and later as Micrococcus freudenreichii ATCC 8459 (in the American Type Culture Collection) (Bergan et al. , 1970 ; Hu and Gurtler , 2017) . T h is strain was later assigned to Pediococcus sp p. (Ma et al., 2007) . The genus Enterococc us was not introduced until 1984 , which separated S treptococcus faecalis and Streptococcus faecium from other Lancefield Group D Streptococc i based on 16S rRNA identification (Schleifer and Killpper - Balz , 1984) . However, genus transition to Enterococcus faecium , from Pediococcus spp. and Streptococcus spp. , was not fully understoo d . 2. 5 . 3 Safety evaluation of E . faecium To date, there are 17 species within the Enterococcus group, compris ing commensal and pathogenic strains . Pathogenic isolates are commonly associated w ith nosocomial infections and multi - drug resistan t traits (Santa gati et al., 2012) . Genetic and biochemical investigations on E. faecium NRRL B - 2354 demosntrated that this particular strain lack s the virulence factor s and other genotypic attributes that are present in clinical strains of E. faecium (Kop it et al., 2014) . 30 Furthermore, food isolates of E. faecium and E. faecalis have not been cor relat ed with human illnesses based on genomic analysis of 25 Enterococcus strains (J. Bonacina et al., 2017) . Such strains from food have also been deemed to be desirabl e for the fermentation of dairy, meat, and vegetable products (Kornacki, 2012) . For example , raw milk a rtisan cheeses in Southern Europe w ere produced using E. faecium strains as non - starter l actic acid bacteria to improve organoleptic properties of the cheese during ripening (Gira, 2002) . M ore recent use s for bacteriocins from e nterococci along with the European Food Safety Authority ( EFSA ) are described by Hanchi et al. ( 2018) . Given that E. faecium NRRL B - 2354 was originally isolated from the dairy environment , E. faecium NRRL B - 2354 has undergone a careful and extensive background check prior its utilization in food. Based on the characteristics of E. faecium NRRL B - 2354 mentioned, this surrogate microorganism has the potential to be used in validat e Salmonella inactivation in LMF . 2 . 5 . 4 E . faecium NRRL B - 2354 in low - moisture food studies E. faecium was initially used by the Almond Board of California as a biological validation tool for dry and moist roasting of almond s because E. faecium was more thermal ly resistan t than Salmonella Enteritidis PT30 (Almond Board of California, 2014) . Due to this attribute, E. faecium has been inve s tigated as a potential surrogate for Salmonella strains in thermal processing of other LMF (Table A.1) . H igher thermal resistance of E. faecium as compared to Salmonella was observed in most inactivation studie s (Bianchini et al., 2014; Ceylan & Bautista, 2015; Channaiah et al., 2016; Enache et al., 2015; Jeong et al., 20 11; Liu, Rojas, et al., 2018; Rachon et al., 2016; Saunders et al., 2018; Shah et al., 2017; Villa - Rojas et al., 2017; Wei et al., 2018) . Salmonella co c ktail s were more common ly used rather than single Salmonella strain s , especially when 31 multiple single food ingredients or multi - ingredient food matrix were tested. With the aim to determine relative thermal resistance between Salmonella and E. faecium in multiple LMFs, use of a Salmonella cocktail is a better approach to r educe the experimental work, given that the heat sensitivity of a ny particular Salmonella strain is highly dependent on product composition (Enache et al., 2014; Limcharoenchat & Marks, 2018; Shah et al., 2017) . Since t he inoculati on method s , recovery media , inactivation treatments , and result ing presentation s of the data differred , it is hard to draw a n overall conclusion whether E. faecium is a suitable surrogate for Salmonella based on these studies. 2. 6 Microbial inactivation model Microbial models are practical tools that enable researchers to predict microbial growth, survival, or death, and - case - experiments for a ny food processing treatmeny (Dolan & Mishra, 2013) . Microbial inactivation models are mathematical descriptions used to quantify microbial reduction with time, when pathogens are subjected to a given kill step (Marks, 2008) . At a constant temperature, plotting pathogen counts for a series of time intervals can describe the inactivation trend for a given thermal processing condition, in which the inactivation curve can be either linear or nonlinea r (Dolan & Mishra, 2013) . A c lassical log - linear model describes heat inactivation following a linear pattern. The primary log - linear model is expressed in Eq. 2. 1 (Peleg & Cole, 1998) , where D (T) is the time required to reduce the microbial population by a 1 - log at a given temperature T (mi n ); N t is the number of survivors at time t (log CFU/g); and N 0 is the initial microbial population (log CFU/g). D (T) - the estimated parameter of the log - linear model , is used to compare the heat resistance of microorganisms, and is commonly denoted as D 80 °C (for instance) (Smelt & Brul, 2014) . The 32 secondary log - linear model, called Bigelow model (Eq. 2. 2) (Bigelow, 1920), is used to describe the relationship of log D(T) with temperature, where z T is the temperature change needed to achieve 1 - log change in D(T). L o g N / N 0 = t / D(T) (Eq. 2.1 ) Log D(T) = log D ref + (T ref T ) / z T (Eq. 2 .2 ) The Weibull model can be used to describe inactivation curves that exhibit downward or upward concave patterns (Eq. 2.3) (Peleg & Cole, 1998) . The Weibull model can also (scale parameter expressed in time ) and p (shape parameter) of the inactivation curve, where p > 1 indicate s a concave downward trend , whereas p < 1 indicate s a concave upward trend of the inactivation curve. Weibull models consider biological variation attributes of microbial inactivation (Van Boekel, 200 2) . L o g N t / N 0 = ( t / p (Eq. 2. 3) T o evaluate the appropriateness of model fitting, several statistical analyses such as root mean square d error (RMSE) and Akaike Information Criteria (AIC) can be used as shown in Eq. 2. 4 and Eq. 2. 5 , respectively: RMSE = (Eq. 2. 4 ) AIC c = (Eq. 2. 5) RMSE is a model error term , which can be used to describe the appropriateness of the model fit. In Eq. 2.4, n is the number of observations and y is the population of survivors. A lower RMSE value indicates lower variability of both the experimental data and residuals of the model (Santillana Farakos et al., 2013) . 33 AIC c (Eq. 2.5) evaluates the probability of one model appropriately describ ing a set of inactivation data over other model, taking into account goodness of the model fit (represented by the sum of squared errors, RMSE) and the number of parameters ( ) (Tenario - Bernal et al., 2013) . While microbial inactivation model s can be used to predict the resistance of Salmonella in low - moisture foods during thermal processing, interpretation of the results is limited to parameters used to develop the model. Therefore, an y interpretation of results obtained from a given model need s to be carefully used to avoid inappropriate extrapolation. 2. 7 Literature s ummary and k nowledge gaps Fundamentally, an i ncreasing number of Salmonella outbreaks linked to LMFs in recent years may be at tributed to the adaptive capacity of Salmonella in dry environment s . The b urden of illness and profit loss as previously reported illustrates that this problem needs immediate intervention to ensure that f ood manufacturer s deliver safe food. Validating the effectiveness of the Salmonella lethality step is crucial, however, Salmonella cannot be used in food production facilities . The s afety evaluation s of E. faecium NRRL B - 2354 indicated a lack of virulence and antibiotic genes , hence it is safe for food validation procedure s . A ny suitable surrogate should demonstrate similar or higher thermal resistance than Salmonella for a given condition. P revious studies , however, did not address to wh at E. faecium NRRL B - 2354 was valid for use . Variations in methods make the comparison of previous findings difficult . Hence, t here is a critical need to standardize the method used to validate thermal processes for LMF for thermal when using E. faecium a s a surrogate for Salmonella . 34 3. EFFECT OF TALC AS A DRY - INOCULATION CARRIER ON THERMAL RESISTANCE OF ENTEROCOCCUS FAECIUM NRRL B - 2354 IN ALMOND MEAL 3.1 Introduction Microbiological safety of nuts and low - moisture foods (water activity [a w ] < 0.85) (Codex Alimentarius, 2015) remains an ongoing concern, with 73 of 95 nut - product recalls from 2001 to 2017 linked to Salmonella contamination (Yada & Harris, 2018) . Almond meal is a coarsely ground product (with skin) that i s commonly used in baked goods such as cookies and granola. Continuing outbreaks and recalls of almonds (Centers for Disease Control and Prevention, 2004, 2014b, 2018b; U.S. Food and Drug Administration, 2018b, 2018c, 2018e) , in addition to the Food Safety and Modernization Act preventive control rules (U.S. Food and Drug Administration, 2018a) , reinforce the need for pathogen control strategies for low - moisture foods and for the validation of methods for such processes. Entry of Salmonella into a low - moisture food as a wet or dry contaminant would be expected to affect its survival rate and thermal resistance (Li et al. 2014) . Most Salmonella inactivation studies involving low - moisture foods have used wet - inoculation methods that can easily achieve levels of 7 to 8 log CFU/g (Hildebrandt e t al. 2016; Jin et al. 2018; Keller et al. 2012; Li et al. 2014; Limcharoenchat et al. 2018; Syamaladevi et al. 2016) . Such high initial populations are crucial for validating processes aimed at decreasing the target organism by at least 5 log. Because adding a wet inoculum to a low - moisture product can cause stickiness and caking, dry inocula are often preferred for certain types of low - moisture foods, such as powders (Hoffmans & Fung, 1992) . Although dry inocula tion (DI) avoids the introduction of water into the test material, high inoculation levels are more difficult to achieve ( Hildebrandt et al. 2017; Zhang et 35 al. 2017) , and the impact of the dry inoculum carrier on thermal resistance of the test organism is not well understood. Various dry carriers have been assessed, including sa nd to inoculate both nuts (Blessington et al. 2013) and sucrose (Beuchat, Mann, Kelly, & Ortega, 2017) , talc powder to inoculate peanut paste (Enache et al., 2015) , and silica beads to inocu late spices (Hildebrandt et al. 2017) . The ideal carrier should retain high numbers of the test organism in its natural physiological state and transfer the population to the food product with easy and complete remov al of the carrier thereafter. However, no identified carrier meets these criteria. Based on several reports, Salmonella populations of 7 to 10 log CFU/g were achieved in silica beads (Hildebrandt et al. 2017) , talc p owder (Enache et al., 2015) , and sand (Blessington et al. 2013) , with 2 to 3 log subsequently transferred to nuts, spices, and peanut butter. While sand and the silica beads were l ar gely removable from dried spices and nuts after inoculation, the same would not be true for peanut butter with such carriers, including ta lc, remaining in the product after inoculation. Use of chalk (calcium carbonate) as a carrier was previously reported to be superior to wet inoculation (WI) in terms of homogeneity of the inoculum and physical properties of the material (Hoffmans & Fung, 1992) . Enache et al. ( 2015 ) showed that talc (hydrous magnesium silicate) did not affect the survival of Salmonella Tennessee or Enterococcus faecium in a model peanut butter paste (0.6 a w , 50% fat) during 30 days of storage. However, the influence of talc on bacterial thermal resistance was not assessed. In addition, mineral powders are non - hygroscopic materials and have hydrophilic - hydrophobic interactions with water (Galet et al. 2010; Rotenberg et al. 2011) , w hich complicates a w testing, leading to further uncertainties in determining bacterial thermal resistance. 36 E. faecium NRRL - B2354 is a nonpathogenic bacterium previously shown to be a suitable surrogate for Salmonella in thermal processing studies (Kopit et al., 2014) . E. faecium was initially used by the Almond Board of California as a biological validation tool for dry and moist roasting of almonds (Almond Board of California, 2014) . E. faecium has since been used to validate thermal and nonthermal protocols for inactivation of Salmonella in other low - moisture prod ucts, including pet food (Bianchini et al., 2014) , wheat flour (Liu, Ozturk, et al., 2018) , hamburger bun dough (Channaiah et al., 2016) , and cereals (Lucore et al. 2017) . Therefore, the objective of this study was to quantify the influence of residual talc on thermal resistance of E. faecium in almond meal at 0.45 a w . 3.2 Materials and methods 3.2.1 Experimental design In this study, WI refers to the direct addition of the inoculum to whole almonds before fabrication into meal, whereas in DI, talc powder is used as the inoculum carrier to mimic possible dry entry of Salmonella into almond meal. These two inoculation methods served as experimental controls. To assess the influence of talc powder as a dry inoculum carrier, wet - inoculated whole almonds were mixed with talc powder before fabrication into almond meal, refe rred to as wet talc (WT). Finally, inoculated talc powder alone (TP; without contact with almonds) was included as another control. After inoculation, the almond and talc powder samples were equilibrated to 0.45 a w and then heated at 80 ° C as shown in Fi g 3. 1. 37 Figure 3. 1 Overall experimental design for wet inoculation (WI), dry inoculation (DI), wet talc inoculation (WT), and talc powder inoculation (TP). 3.2.2 Talc A 300 - g batch of talc powder (hydrous magnesium No. 14807 - 96 - 6, Sigma, St. Louis, MO) was placed on a sterile tray and heated in an oven at 100 ° C for 2 h. Thereafter, five 1 - g samples were randomly selected from the tray and plated on tryptic soy agar (Difco, BD, Fr anklin Lakes, NJ) supplemented with 0.6% (w/v) yeast extract (Difco, BD) (TSAYE) to quantify background microorganisms (Enache et al., 2015) . Heat - treated talc powder was stored in a tightly closed ster ile stainless - steel jar at room temperature before use. 3.2.3 Almonds Raw almonds (no shell, nonpareil, 20/22, propylene oxide pasteurized, 49.5% crude fat; Nuts.com, Cranford, NJ) were vacuum sealed in plastic bags and stored at 48 ° C. Before use, popula tions of coliforms and yeast or mold were determined in triplicate 25 - g samples by diluting in 225 mL of 0.1% buffered peptone water (BPW), stomaching for 3 min (model 1381/471, NEU - TEC Group Inc., Farmingdale, NY), serially diluting, and plating on Petrif ilm E. coli or coliform 38 and yeast and mold count plates (3M Company, St. Paul, MN), respectively. Particle size distribution of fabricated almond meals. A 100 - g uninoculated almond sample was ground using a food processor (model FP21, Hamilton Beach Brands , Inc., Glen Allen, VA) at a speed setting 1 for 45 s, stopping every 9 s to stir the almonds with a sterile spatula, to obtain a meal. Approximately 0.1 g of this meal was added to 60 mL of isopropanol and mixed with an overhead stirrer at 1,200 rpm to en sure uniformity. While stirring, a 3 - mL aliquot was transferred to the chamber of a Microtrac Laser light scattering system (S3500 Tri - laser System, Microtrac Inc., York, PA) to obtain particle size measurements. Particle size distribution was determined f or two representative batches of almond meal. 3.2. 4 Inoculum preparation E. faecium NRRL B - 2354 was received from the Institute for Food Safety and Health (Bedford, IL) and maintained at 80 ° C in tryptic soy broth (Difco, BD) supplemented with 0.6% (w/v) yeast extract (Difco, BD) (TSBYE) containing 20% (v/v) glycerol. Using the method of Keller et al. (2012) , one vial of frozen stock (1.0 mL) was submerged in a 37 ° C water bath for 30 s, after which 0.1 mL was transferred to 9 mL of TSBYE and incubated for 24 h at 37 ° C. A loopful of this overnight culture was then streaked to plates of TSAYE to obtain working stock cultures (transferred monthly) after 24 h of incubation at 37 ° C. For each experiment, a single colony from the working stock was transferred to TSB YE and incubated for 24 h at 37 ° C, after which 1 mL was spread on a 150 - mm - diameter plate of TSAYE to obtain confluent growth after 24 h of incubation at 3 7° C . The bacterial lawn was then harvested in 2.25 mL of BPW using a spreader. The E. faecium populat ion in the suspension was determined by directly plating appropriately diluted samples on Trypticase soy agar supplemented with esculin (0.025%) (97% esculin hydrate; 39 Sigma) and ammonium ferric citrate (0.05%; Sigma) (ETSA) - a nonselective or differential medium that produces black colonies of E. faecium after 48 h of incubation at 37 ° C. 3.2.5 Wet inoculation Whole almonds (100 g) were inoculated with 1 mL of the E. faecium suspension (about 10 log CFU/mL), massaged by hand in a Whirl - Pak bag (Nasco, Modesto, CA) for 10 min, and transferred to a steel tray. After drying in a biosafety cabinet for 1 h around 238C, three 1 - g samples were stomached for 3 min, serially diluted i n 0.1% BPW, and plated on ETSA to determine the initial population of E. faecium . The WI whole almonds were then transferred to a computer - controlled conditioning chamber (Hildebrandt et al., 2016) at 45% relative humidity. After reaching an a w of 0.45 ( ± 0. 25) within 2 to 3 days, the almonds were ground into a meal as previously described and then reconditioned in the chamber to 0.45 aw before isothermal treatment. To confirm homogeneity of the inoculum, three 1 - g almond meal samples were collected and ass essed for numbers of E. faecium using ETSA as previously described. The inoculum was considered homogeneous if the standard deviation for three subsamples was less than 0.3 log CFU/g (Lang et al., 2017) . 3.2.6 Dry inoculation One mL of inoculum was added to 100 g of heat - treated talc powder, which was then massaged by hand in a Whirl - Pak bag f or 10 min, aseptically transferred to a sterile tray, and dried in a biosafety cabinet for 24 h. Thereafter, an equal weight of whole almonds was added to dried inoculated talc powder in a new Whirl - Pak bag and massaged by hand for 10 min to ensure that ea ch almond was evenly coated. Excess talc was removed from the DI almonds by sieving (U.S. 40 Standard 20) for 2 min with no clumping or caking observed, after which the talc - coated almonds were weighed. DI almonds were conditioned to 0.45 aw and ground into a lmond meal in the same manner as for WI almonds. E. faecium populations for talc, DI whole almonds, and DI almond meal were determined by plating on ETSA as previously described. WT and TP. A 100 - g batch of whole almonds was inoculated using WI. Thereafter , 100 g of uninoculated talc powder was added into the dried WI almonds, sieved, and weighed. After reaching the targeted a w of 0.45, these WT almonds were ground into meal as previously described. For TP, 100 g of talc powder was inoculated with 1.0 mL of the culture and held for 2 days at 45% relative humidity, which was the same equilibration period for almond meal. Inoculum homogeneity testing of WT whole almonds, WT almond meal, and TP alone was performed as previously described. WI, DI, WT almond meal , and TP were independently prepared in triplicate batches and then subjected to isothermal heating at 80 ° C . 3.2.7 Isothermal treatment Isothermal heating targeting a 3 - log reduction was conducted by loading 0.5 - to 0.8 - g samples of the respective materials into aluminum test cells (sample thickness of about 1 mm (9)) that were then submerged in a water bath (GP400, Neslab, Newington, NH) at 80 ± 0.2 ° C, with the temperature verified using a T - type thermocouple probe connected to a handhel d thermometer (Omega RDXL4SD, Norwalk, CT). Come - up time was defined as the time for the samples to reach 0.5 ° C below the treatment temperature, as determined by inserting a temperature probe into the center of the test cell cavity containing the sample. T he average of six such measurements plus two standard deviations was used as time zero (t0) for subsequent analyses to ensure that an isothermal condition had been achieved before sampling. Starting at t0, three test cells were 41 removed at each of five or m ore equally spaced time intervals and immediately placed on ice (about 30 s). Three 1 - g samples of untreated almond meal were used as positive controls for each isothermal trial. All isothermal trials were performed in triplicate. Survivor enumeration and estimation of D - value. Treated samples were aseptically transferred from the test cells into Whirl - Pak bags, serially diluted in 0.1% BPW, stomached for 3 min (NEU - TEC Group), further diluted as needed, and then plated in duplicate on ETSA. After 48 h of i ncubation at 37 ° C, all black colonies were counted as E. faecium . Based on the pooled data from triplicate trials, D - values obtained from each inoculation method were determined by one - step regression of log (N/N 0 )=t/D using nlinfit, a nonlinear regression tool in MATLAB (version R2017b, Math Works Inc., Natick, MA). Root mean square d error and the 95% confidence interval were used to determine the goodness of fit of the model. The slopes of the regression were then compared using analysis of covariance and 3.3 Results and discussion 3.3.1 Particle size distribution of almond meals The two batches of fabricated almond meal consistently exhibited three major particle size distribution peaks reported in micromete rs. Volume percentages for each peak were 40.2% (897.8 - moisture food, wet analysis of the particle s ize measurement was necessary to avoid clumping. However, this analysis did not decrease the reliability or accuracy of the light scattering measurement, because the Microtrac Trilaser System uses three laser sources at 780 nm. 42 3.3.2 Microbial background populations on whole almonds and heat - treated talc. Whole almonds (three 25 - g samples) yielded coliform and yeast or mold counts of 4.6 ± 0.5 and 3.3 ± 0.3 log CFU/g, respectively. Heat - treated talc powder had a mesophilic aerobic bacteria population of 2.7 ± 0.2 log CFU/g. However, given the high inoculation levels used for E. faecium , the native background microflora did not interfere with survivor enu meration after thermal treatment. 3.3.3 E. faecium populations on almonds, almond meal, and talc before heat treatment E. faecium populations on WI almonds and WI almond meal ranged from 7.9 to 8.4 and 7.4 to 7.8 log CFU/g, respectively (Fig 3. 2). For DI , initial E. faecium populations ranged from 8.0 to 8.5, 4.8 to 5.8, and 4.9 to 5.4 log CFU/g for talc, almonds, and almond meal, respectively. DI almonds (100 - g batch) retained 1.4 6 0.2 g of inoculated talc, or about 1% of that used for inoculation, with these inoculated almonds yielding E. faecium of around 6 log CFU/g. DI almonds contained E. faecium at 4.8 to 5.8 log CFU/g, with this lower population expected because of some retention of talc on the sieve. Similar populations of 4.2 to 5.1 log CFU/g we re previously reported using sand as the inoculum carrier . WT almonds (7.6 to 8.5 log CFU/g) and WT almond meal (7.7 to 8.0 log CFU/g) yielded similar E. faecium populations when compared with WI almonds and WI almond meal, respectively. These E. faecium populations were similar to TP alone (7.5 to 9.2 log CFU/g) and DI talc. WT and DI almonds contained 1.6 6 1.2 and 1.4 6 0.2 g of talc post - inoculation, respectively. However, after using WT inoculation to mix uninoculated talc with the almonds, the sieved talc yielded E. faecium populations of 5.1 to 7.2 log CFU/g, suggesting preferential adherence to the talc. High variability in the E. faecium population of sieved talc may have been because of inadequate mixing with the spatula. Before isothermal treatme nt, E. faecium 43 was homogeneously distributed in each batch of WI, DI, and WT almond meal with a standard deviation of 0.04 to 0.09 log CFU/g, whereas higher variability was seen for TP (0.10 to 0.29 log CFU/g). Nonetheless, all batches were within the esta blished standard deviation of 0.30 log CFU/g and therefore considered homogeneous. Figure 3. 2 Populations of E. faecium on almonds, almond meal, and talc using the wet inoculation (WI), dry inoculation (DI), wet inoculation talc (WT), and talc powder (TP) methods before thermal treatment. Each point represents the mean of triplicate samples. 3.3.4 Thermal resistance of E. faecium Overall, this study aimed to determine the efficacy of talc as a carrier for DI of almonds and the impact of carrier residue on thermal resistance of E. faecium . Almond meal and talc yielded thermal come - up times of 2.0 and 3.1 min, respectively. Total heating times at 80 ± 0.2 ° C were 120 min for almond meal and 63 min for talc, respectively. All E. faecium inactivation curves were log linear without tailing (Fig 3. 3). Over the course of isothermal heating, E. faecium populations 44 for E. faecium in TP being the lowest (Table 3.1) . Overall, the log - linear model was a good fit for the isothermal data, with the relatively low root mean square error values, comparable to what is typically expected for predictive modeling (Farakos et al., 2014) . The presence of about 1 g of talc residue in both DI and WT almond meal after inoculation increased the thermal resistance of E. faecium . Despite having similar amounts of talc, E. faecium thermal resistance differed between Figure 3. 3 Inactivation curve (80ºC) for E. faecium NRRL B - 2354 in talc and almond meal inoculated using the wet inoculation (WI), dry inoculation (DI), and wet talc inoculation (WT) methods. Each point represents triplicate subsamples at each time point. Table 3. 1 Thermal res istance (D values at 80ºC) of E. faecium in wet inoculated (WI), dry inoculated (DI), wet talc inoculated (WI) almond meal, and inoculated talc powder (TP) at 0.45 a w . Sample 45 D - values with differing superscript letters were significantly different ( P < 0.05), based on analysis of covariance (ANCOVA). a Root mean square error DI and WT almond meal. This may be because of the stage at which the talc was added, or it may reflect a synergistic interaction between talc and almond meal; otherwise, the D - v alues for E. faecium in DI and WT almond meal should have been between that of talc and that of WI almond meal (20 to 40 min). The surface characteristics of talc differ greatly from those of hygroscopic foods, such that talc can be hydrophilic when intera cting with water molecules under low humidity but hydrophobic when interacting with droplets of water under saturated conditions (Rotenberg et al. 2011) . Talc may not have been able to absorb additional moisture in the conditioning chamber to reach a state corresponding to 0.45 a w for almond meal, which may have affected the measured thermal resistance of E. faecium . Given the high fat content of almond meal, the combination of talc and almond meal may have increased surface hydrophobicity, leading to increased thermal resistance of E. faecium . Absence of a food matrix, particularly fat, which would have partially shielded the organism from heat stress (Shachar & Yaron, 2006) , might also be responsible for talc alone having the lowes t D - value. 3.4 Conclusion In summary, although the use of dry carriers for inoculation of low - moisture foods may better mimic how these products become contaminated, the presence of a dry carrier such as talc may alter thermal resistance of the target organism and lead to either over - or underestimation of lethality. Therefore, such carriers should not be used indiscriminately in bacterial inactivation or process validation studies without first knowing their impact on thermal resistance of the t arget organism . 46 4. INTERLABORATORY EVALUATION OF ENTEROCOCCUS FAECIUM NRRL B - 2354 AS A SALMONELLA S URROGATE FOR V ALIDATING T HERMAL T REATMENT OF MULTIPLE L OW - M OISTURE F OODS 4.1 Introduction During the last 20 years, Salmonella outbreaks/recalls have been linked to a wide range of low - moisture foods (LMFs) and ingredients, including nuts and nut butter (Centers for Disease Control and Prevention, 2007, 2009, 2014b, 2016a) , cereal (Centers for Disease Control and Prevention, 1998, 2008b) , dried fruit (Anonymous, 2014) , infant formula (European Food Safety Authority, 2018) , herbs/spices - seasoning powder (Centers for Disease Control and Prevention, 2010; U.S. Food and Drug Administration, 2017b, 2017a) , pet food (Centers for Disease Control and Prevention, 2008a; U.S. Food and Drug Administration, 2018f) , and supplement powders (C enters for Disease Control and Prevention, 2016b, 2018a) . Despite current food safety standards and the preventive controls now mandated by the Food Safety Modernization Act (U.S. Food and Drug Ad ministration, 2018a) , these Salmonella outbreaks/recalls have expanded to include an ever - widening number of LMFs (Centers for Disease Control and Prevention, 2018b; U.S. Food and Drug Administration, 2018e, 2018c, 2018b, 2018d) . Validating Salmonella inactivation strategies for LMFs remains challenging due to the complex interactions between the pathogen, process, and product (National Advisory Committee on Microbiological Criteria for Foods, 2010) . In addition, because these pathogen reduction processes cannot be validated using Salmonella in food production facilities, an ap propriate, nonpathogenic surrogate is the most common validation tool (Hu & Gurtler, 2017) . Enterococcus faecium NRRL B - 2354, previously cl assified as a Pediococcus sp. (Jeong et al., 2011) , is a nonpath ogenic bacterium. Based on its higher thermal resistance, E. faecium NRRL B - 2354 was identified as a suitable surrogate for thermal inactivation of Salmonella Enteritidis 47 PT30 in almonds by the Almond Board of California (Almond Board of California, 2014) . Further work by Kopit et al. (Kopit et al., 2014) demonstr ated that E. faecium NRRL B - 2354 is unlikely to harbor genes for virulence or antibiotic resistance. Given these findings, E. faecium NRRL B - 2354 has been used as a surrogate for Salmonella in various LMF validation studies involving beef jerky (Borowski & Ingham, 2009) , peanuts (Poirier et al., 2014; Sanders & C alhoun, 2014) , pet food (Bianchini et al., 2014; Ceylan & Bautista, 2015) , wheat flour (Liu et al., 2018; Villa - Rojas et al., 2017) , hamburger bun dough (Channaiah et al., 2016) , cereals products (Lucore, et al., 2017) , black peppercorn (Wei et al., 2018) , cashew and macadamia nuts (Saunders et al., 2018) , sunflower kernels, flaxseeds, and quinoa (Shah et al ., 2017) . Nevertheless, there is a need to confirm the validity of E. faecium as a Salmonella surrogate across a wide variety of low - moisture products, using standardized protocols to eliminate differences related to culture preparation (Hildebrandt et al., 2016; Ma et al., 2 009) , recovery media (C. W. Clark & John Ordal, 1969; Gurtler, 2009) , and inoculation protocol (Enache et al., 2015; Limcharoenchat et al., 2018) , all of which have been reported to influence thermal resistance of Salmonella . Salmonella thermotolerance in LMFs varies with serovar (Santillana, 2014) , growth conditions (Hildebrandt et al., 2016) , processing relative humidity (Jeong et al., 2011) , water activity (a w ) (Syamaladevi et al., 2016) , treatment temperature (Farakos et al ., 2014; Lang et al., 2017) , and physical/chemical characteristics of the product (Enache et al., 2014; Jin et al., 2018; Li et al., 2014; L imcharoenchat et al., 2018) . Several previous studies suggest that high - fat foods enhance Salmonella thermal resistance. For example, Salmonella was more thermally resistant in peanut butter than in wheat flour (Syamaladevi et al., 2016) . Using different peanut butter formulations, Salmonella was least heat resistant (at 72°C) in reduced - fat peanut butter (6 vs. 33% fat) (He et al., 2011) . In a similar study, fat levels of 47 to 56% in formulated peanut paste (a 48 smaller range than the prior study) did not significantly influence Salmonella thermotolerance (Enache et al., 2014) . More recently, Jin et al. ( 2018) assessed thermal resistance of Salmonella Agona in a high - fat, high - protein model food. In that study, the D 75 °C at 0.50 and 0.63 a w was higher for the high - fat than for the high - protein matrix; however, the D 75 °C at 0.73 to 0.90 a w showed the opposite result, with higher values in the high - protein matrix. In addition to product formulation, product structure can also affect Salmonella thermal resistance, with higher resistance reported in almond meal than in whole almonds at multiple water activities and treatment temperatures (Limcharoenchat et al. , 2019) . Overall, food composition must be considered when justifying the use of a Salmonella surrogate for validation of LMF thermal processes. To further understand the effect of food co mposition on thermal resistance of Salmonella . Syamaladevi et al. ( 2016) proposed that temperature - induced changes in a w could explain some observed differences in thermal resistance, using wheat flour (high - carbohydrate) and peanut butter (high - fat) as model materials. Therefore, reporting a w values of other LMF products at elevated test temperatures also should help advance the understanding of the influence of a w on thermotolerance of Salmonella . Method variations, in addition to critical parameters affecting Salmonella thermal resistance, make any comparison of findings from previous studies difficult (Van Asselt & Zwietering, 2006) however, reproducibility of thermal resistance results across laboratories is possible when methods are carefully standardized and controlled (Hildebrandt et al., 2020) . Consequently, a critical need exists to standardiz e the methodology used in studies that validate the use of E. faecium as a Salmonella surrogate in LMF. T his study aimed to: (i) quantify the effect of temperature and product composition on relative thermal resistance of E. faecium and 49 Salmonella in peanut butter, nonfat dry milk powder, almond meal, ground black pepper, wheat flour, and date paste, and (ii) determine the reproducibility of results across laboratories. 4.2 M aterials and methods 4.2.1 Overall study design Six low - moisture food pr oducts (nonfat dry milk powder, peanut butter, almond meal, wheat flour, ground black pepper, and date paste) were chosen based on their association with previous Salmonella outbreaks/recalls, varying composition (fat - , protein - , carbohydrate - , sugar - , ant imicrobial - rich), and structure (powder, particulate, paste) (Fig 4. 1). Each product was assigned to two independent laboratories as follows: nonfat dry milk powder and almond meal - Michigan State University (MSU) and Washington State University (WSU); pe anut butter and ground black pepper - Institute for Food Safety and Health (IFSH) and University of Nebraska - Lincoln (UNL); wheat flour - IFSH and WSU; and date paste - MSU and University of Georgia (UGA). Thermal resistance trials for both E. faecium and Salmonella were conducted at three different temperatures, with one of the two institutions responsible for product inoculation, equilibration at the designated a w , and then shipping half of the product to the collaborating institution for testing using identical protocols. All inoculated products were thermally treated within 21 days after inoculation. Survivors were enumerated on nonselective differential media (d escribed below), with the counts entered into identical data - recording templates in a single cloud - based data folder. MSU analyzed all thermal inactivation data using log - linear and Bigelow models, via a one - step regression (detailed below). 50 Figure 4. 1 Experimental design of multi - laboratory thermal inactivation of E. faecium and Salmonella in six low - moisture products at designated water activities (aw). Each organism was tested at three temperatures; where four te mperatures are listed, the top three were for E. faecium , and the bottom three were for Salmonella , based on preliminary works. Arrow indicates laboratory that was responsible for product inoculation. 4.2.2 Inoculum preparation A cocktail of five differe nt Salmonella serovars was selected based on their previous association with the low - moisture products being assessed . Salmonella Agona 447967 (2008 rice puffed cereal outbreak), Salmonella Montevideo 488275 (2009 - 2010 black pepper outbreak), and Salmonella Mbandaka 698538 (2013 sesame tahini outbreak) were obtained from the U.S. Food and Drug Administration (FDA, ORA Arkansas Regional Lab, Jefferson, AR); Salmonella Tennessee K4643 (2006 pe anut butter outbreak) was obtained from Dr. Larry Beuchat at the University of Georgia (Athens, GA); Salmonella Reading Moff 180418 (associated with cumin) was obtained from the FDA culture collection (FDA CFSAN, Bedford Park, IL). Enterococcus faecium NR RL B - 2354, which served as the surrogate (hereafter referred to as E. faecium ), was obtained from the U.S. Department of Agriculture Agriculture Research Service (USDA - ARS, 51 Peoria, IL). All strains were grown on slants of Trypticase soy agar ( Difco, BD, Franklin Lakes, NJ) containing 0.6% yeast extract (Difco, BD) [TSAYE] at IFSH and then sent by overnight mail to the four other participating laboratories. Upon receipt, fr ozen stocks were prepared by transferring the culture to Trypticase Soy Broth (Difc o, BD, Franklin Lakes, NJ) supplemented with 0.6% yeast extract (Difco, BD), [TSBYE] for 24 h (37°C), mixing with 20% (v/v) glycerol, and then freezing multiple 2 - ml aliquots at - 80°C for long - term storage. To create working stock cultures , each vial of th e frozen (1 ml) stock was thawed in a water bath for 30 s (37°C), subjected to two sequential 24 h (37°C) transfers in 10 mL TSBYE and then streaked onto plates of TSAYE (working stocks). After incubation, the plates were sealed with parafilm and stored ( 4°C), with the culture transferred monthly as needed. The inoculum was prepared using the agar - lawn method of Keller et al. (Keller et al., 2013) with slight modifications. A single colony from the working stock was transferred into 10 mL of TSBYE and incubated for 24 h (37°C), after which 100 µL was spread onto a TSAYE plate (100 x 15 mm) for confluent growth. After 24 h (37°C), the bacterial lawn was harvested in 1 ml of 0.1% buffered peptone w ater (Difco, BD), and transferred into a sterile test tube. Equal volumes of the five Salmonella strains were then combined to obtain a five - strain cocktail containing ~11 log CFU/mL. E. faecium was similarly grown and harvested to obtain a single strain s uspension containing ~10 log CFU/mL. Both cell suspensions were used within 2 h of preparation. 4.2.3 Proximate analyses Moisture, crude protein, and crude fat were analyzed by an independent third party (Dairy One Forage Lab, Ithaca, NY). Total soluble s olids for date paste was determined using an A bbe 3 - L refractometer (Bausch & Lomb Optical Co., Rochester, NY). The native a w of each product was 52 measured using an AquaLab series 4TE meter (Meter Group, Pullman, WA). All analyses included duplicate samples with duplicate measurements. 4.2.4 Background microbial counts Coliform and yeast/mold background counts for all samples were obtained using 3M Petrifilm (3M Company, St. Paul, MN) for coliform (incubated overnight at 37 °C) and yeast/mold (incubated for 3 to 5 days at 25 °C) , respectively. Presence of any Salmonella was evaluated by plating a 1:10 dilution of the sample in 0.1% buffered peptone water (BPW) ( Difco, BD ) on Tryptic Soy Agar ( Difco ) supplemented with 0.6% (w/v) yeast extract ( Difco ), which was modified by adding ammonium ferric citrate (0.05%) (Sigma Aldrich, St. Louis, MO) and sodium thiosulfate (0.03%) (Sigma Aldrich) (MTSA), to obtain a non - selective, differential medium. The appearance of black colonies on MTSA signified presump tive Salmonella e. 4.2.5 Product inoculation, equilibration, and dissemination Detailed inoculation procedures for each product type are described in the following sections. Overall, t o achieve an initial population of ~8 log CFU/g , a ratio of 1 mL of ino culum per 100 g was used for all products, except for peanut butter (3 mL per 200 g) and black pepper (2 mL per 100 g). After inoculating three batches per product with either the Salmonella cocktail or E. faecium , each batch was equilibrated in a custom - designed, controlled - humidity conditioning chamber, with similar chambers at each laboratory (Hildebrandt et al., 2016) . Each product was equilibrated to either 0.25, 0.45, or 0.65 a w , according to which was value was closest to the native a w of that product. To ensure homogeneity of the inoculum, ten 1 - g samples per batch were diluted 1:10 in 0.1% BPW, stomached for 3 min, appropriately diluted, and surface plated on either MTSA 53 for Salmonella or Tryptic Soy Agar (Difco) supplemented with yeast extract (0.06%) (Difco), ammonium ferric citrate (0.05%) (Sigma - Aldrich), and esculin hydrate (0.025% ) (Acros Organics, Morris, NJ) (ETSA) for E. faecium , in duplicate , with both organisms enumerated after incubation for 24 h (37°C). The inoculum was considered homogeneous if the standard deviation in bacterial counts for plates with 25 to 250 colonies wa s < 0.3 log CFU/g . After equilibrating to the designated a w w , the inoculated samples were packed in multi - layer, vapor - w level) and sent to the paired laboratory via an overnight shipping service (following appropriate biosafety protocols for shipping/receiving). Once received, the samples were held at the appropriate equilibrium RH in a controlled chamber and used within 2 1 days of inoculation. 4.2.5.1 Peanut butter A commercial brand of creamy peanut butter with less than 2% of hydrogenated vegetable oils and salt (batch no. 52735315000244H, Peter Pan, ConAgra Food Inc., IL) was inoculated at IFSH. Briefly, the i noculum was mixed with peanut oil (Signature Kitchens, Better Living Brands, LLC, Pleasanton, CA) at a 1:1 ratio (v/v), followed by a few drops (~0.25 ml) of Tween 80 (Fisher Chemical, Whippany, NJ) to form an emulsion and then vortexed . After adding 3 ml of this inoculum to 200 g of peanut butter in a Whirl - Pak bag (Nasco, Fort Atkinson, WI), the inoculum was evenly dispersed by repeated hand - massaging and stomaching for at least 6 min. The inoculated peanut butter was placed in a small container that was attache d to a continuous mixing apparatus to avoid peanut butter separation during equilibration in a humidity - controlled chamber to 0.25 a w (3 to 4 days). After confirming homogeneity of the inoculum, half of the peanut butter 54 was packed in FoilPAK pouches (KSP - 150 - 1MB, ProAmpac, Cincinnati, OH) and shipped to UNL. 4.2.5.2 Almond meal Raw almonds ( Prunus dulcis ) (no shell; nonpareil; 20/22 size; propylene oxide pasteurized) were obtained from a retail supplier (Nuts.com, batch no. p38572195 , Cranford, NJ), vacuum - packaged (500 g), and stored at ~2.5 °C until use. Inoculation and fabrication of almond meal were pe rformed at MSU as described by Limcharoenchat et al. (Limcharoench at et al., 2018) , with slight modification. Briefly, 400 g of almonds were mixed with 4 mL of inoculum in a Whirl - Pak bag (Nasco, Modesto, CA). After hand - massaging the bag for 1 min, the inoculated almonds were transferred to a sterile tray, dried in a biosafety cabinet for 1 h, and then transferred and held in a conditioning chamber until the a w of the almonds reached 0.45 (2 to 3 days). The inoculated almonds (~100 g) were then ground for 45 s using a food processor (model FP21, Hamilton Beach Brand, Glen Allen, VA) at a speed of 1, with the resulting almond meal re - conditioned to 0.45 a w (1 to 2 days). After homogeneity of the inoculum was confirmed, the almond meal was packed in a 250 mL - PET bottle, which was capped, sealed with parafilm, and packed in double Whirl - Pak bags before shipping to WSU. 4.2.5.3 Date paste Jumbo Medjool dates ( Phoenix dactyl ifera ) (5.1 to 5.7 cm in length;) were purchased from Sun Garden Growers (batch no. 281557, Bard, CA), vacuum - packed (200 g), and stored at ~2.5 °C until used. Inoculation of dates and fabrication into date paste was performed at MSU according to Limcharoen chat et al. (2018), with slight modification. Whole dates (600 g) were pitted and 55 cut in half, after which ~10 g of date halves were cut into 6 pieces and spot - inoculated with 100 µL of inoculum. After drying for 1 h and conditioning to 0.65 a w over 3 to 4 days in the conditioning chamber, the date pieces were subjected to three passages through a meat grinder (Model no KV25GOXOB, Kitchen Aid, Benton Harbor, MI) with 1 cm hole grinding plates, at a speed setting of 2. Date paste was then formed into a smal l ball (~ 1 g each) using a spatula and re - equilibrated to 0.65 a w (3 to 4 days) in the conditioning chamber. After confirming homogeneity of the inoculum, the date paste was packed in PET bottles and shipped to UGA. 4.2.5.4 Nonfat dry milk powder Lo w heat pasteurized nonfat dry milk powder was acquired from the Michigan Milk Producers Association (lot no. 2201604900010 26, Ovid, MI). Milk powder was vacuum - packed (500 g) and stored at room temperature (~23°C) to avoid caking. A batch (300 g) of nonfat dry milk powder was inoculated at WSU by mixing 30 - g of nonfat dry milk powder in a Whirl - Pak bag with 3 mL of inoculum and hand - massaging until no visible inoculum droplet was observed. The inoculated nonfat dry milk powder was then transferred to a steri le tray for equilibration to 0.25 a w (2 to 3 days in a conditioning chamber), and g round using an aluminum cylinder (18 mm width x 80 mm height) until no large particle were observed. This pulverized milk powder was used as a seed inoculum to inoculate the remaining nonfat dry milk powder (270 g) , by mixing 10 g of seed inoculum with 90 g of uninoculated milk powder and stomaching for 3 min (260 rpm) at a time, before re - equilibration of the whole batch to 0.25 a w (3 to 5 days). The sample was packed in PET bottles and shipped to MSU. 56 4.2.5. 5 Wheat flour Unbleached all - purpose wheat flour (batch no. 3701961701, Gold Medal, General Mills Inc., MN;) was purchased from a retailer in Lansing, MI, vacuum - packaged in 500 - g quantities, and shipped to WSU. Wheat flour was inoculated at WSU by first mixing 50 g of flour with 6 mL of inoculum in a Whirl - Pak bag. After repeating 3 min of stomaching followed by 1 min of hand - massaging three times, 5 g of this seed inoculum was added to 55 g of uninoculated flour, ha nd - massaged in a new bag for 1 min and subsequently stomached for 3 min, with this process repeated twice. All inoculated flour was transferred to a Whirl - Pak bag, mixed using a sterile spatula for 3 min and then moved to the conditioning chamber for equil ibration to 0.45 a w . After confirming homogeneity of the inoculum, the samples were packed in PET bottles and shipped to IFSH. 4.2.5.6 Ground black pepper Steamed - pasteurized whole black peppercorn ( Piper nigrum ) was obtained from McCormick and Company, Inc. (Sparks, MD, USA) and stored in double - sealed polypropylene bags at - 20 °C until inoculation at UNL. Whole black peppercorn (1 kg) was pre - equilibrated to 0.45 a w over 2 days in a conditioning chamber, placed in a Whirl - Pak bag, and sprayed with 5 x 4 mL of inoculum, with alternate hand - massaging for 2 min. Following re - equilibration to 0.45 a w over 2 days , black peppercorn was refrigerated for 2 to 3 h before grinding. Whole black peppercorn (300 g) was ground using a (powder grinder , Waring, Lancaster, PA) for 30 s. After 10 min to allow the particulates to settle, the ground black pepper was sieved for 3 min (U.S Standard No. 20 sieve), followed by 10 min of settling, after which the sieved product was weighe d . Any black peppercorn particulates remaining on the sieve were returned to the grinder with another 300 g of black peppercorn for subsequent processing. Finely ground black pepper 57 was re - conditioned to 0.45 a w (1 day in the conditioning chamber ). After confirming homogeneity of the inoculum, the finely ground black pepper was transferred to an aluminized plastic pouch ( PAKVF4C, IMPAK Corp., Los Angeles, CA), which was heat - sealed (IPKHS - 606T, IMPAK Corp., Los Angeles, CA) and then shipped to IFSH. 4.2.6 Come - up time Prior to thermal come - up time measurement, the temperature of the water bath (GP - 400, Neslab, Newington, NH) was verified (± 0.2°C). Come - up time (the time needed to reach 0.5°C below the treatment temperature) was measured by immersing a T - type thermocouple probe connected to a handheld thermometer (Omega RDXL4SD, Norwalk, CT) in the water bath, with a second probe also attached to the center of an aluminum test cell (0.5 to 0.8 g, 4 mm thick; Washington State University, Pullman, WA) (Chung, Birla, & Tang, 2008) containing an equilibra ted, uninoculated sample (~1 g). To mimic experimental conditions , 18 similar test cells were also immersed in the water bath when establishing the come - up time. The average of six measurements (+ 2 standard deviations) was used to determine time zero (t 0 ) for each isothermal treatment. 4.2.7 Isothermal treatment and survivor enumeration Aluminum test cells were loaded with 1 g of the inoculated sample and tightly sealed. For each isothermal experiment, the aim was to achieve a target reduction of 3 - 5 logs across 6 to 8 equally spaced time intervals, with three test cells pulled at each time interval, starting at t 0 , and then immediately placed on ice (for ~30 s) to stop thermal inactivation. After aseptically opening the test cell, the contents were se rially diluted in 0.1% BPW and stomached for 3 min (NEU - TEC 58 Group Inc, Farmingdale, NY), followed by spreading 100 µl of the appropriate dilutions on duplicate plates of MTSA or ETSA for enumeration of Salmonella and E. faecium , respectively. Survivors (black colonies) were counted after 24 (MTSA) and 24 to 48 h (ETSA) of incubation at 37°C, with acceptable colony counts for subsequent analysis ranging from 25 to 250, giving a limit of quantification (LOQ) of 25 CFU. 4.2.8 Wat er activity measurement at 80°C Uninoculated samples of each product (100 g) were equilibrated to their target a w , which was verified using an AquaLab a w meter (Meter Group) as previously described. After equilibration, 1 - g samples were packed in a a w meta l cup with lid, wrapped in parafilm, and transferred to a Whirl - Pak bag to avoid any a w changes before analysis. A custom - designed, high - temperature a w meter (Meter Group, Pullman, WA, USA) (Syamaladev i et al., 2016) was calibrated using three a w standards of 0.25, 0.50, and 0.76. The sample was then allowed to equilibrate at 23°C for 15 min and placed in the air - sealed sample chamber for an additional 5 min to measure a w at room temperature (23°C). To mimic the high temperatures during isothermal treatment , the air - sealed chamber containing the sample then was placed in an oven at 80°C (± 0.2°C) . After reaching and remaining at 79.5°C for 10 min, the a w was recorded, and the sample chamber was removed from the oven and cooled to room temperature (23°C) before removing the sample cup, capping and wrapping it in parafilm, and placing in a Whirl - Pak bag fo r post - heating a w measurement using the Aqualab a w meter . All measurements were performed in duplicate. 59 4.2.9 Data analysis For each isothermal treatment, survivor data were collected from t hree biologically independent trials conducted in triplicate and analyzed by MSU to obtain D - and z - values. The log - linear (primary) and Bigelow (secondary) models were used to estimate D ( T ) (min) and z T ( °C), respectively, using a nonlinear tool ( nlinfit ) from MATLAB software (version R2017a Math Works Inc, Natick, MA) via one - step regression according to Eq. 4. 1. log (Eq. 4.1) where N is the population of survivors in (CFU/g); N 0 is the initial bacterial population (CFU/g); T is the specific treatment temperature (°C); D ( T ) is the time needed to achieve a 1 log reduction in the bacterial population (min); D ref is the D - value (min) at the reference temperature; T ref is the reference temperature (°C); and z T (°C) is the temperature change needed for a 1 log change in D ( T ). Model performance (goodness - of - fit) was quantified by root mean squared error (RMSE). Weibull model (Jin e t al., 2018) parameter that was not significantly different than a value of 1, indicating a log - linear trend. T ref was set as 80°C for all products and both organisms. Although not nec essarily statistically ideal, given that the various product - organism combinations were run at different temperature ranges (due to differences in resistances), establishing a common reference temperature was necessary to enable direct comparisons across t he various products and both organisms. The acceptability of this choice was evaluated in terms of the model RMSE. The significance tests for differences in thermal resistance between laboratories , food composition, and isothermal treatment temperature wer e t - 60 unbalanced sample sizes and variances (Welch, 1938) . Changes in a w due to temperature evaluated t - w measurements). 4 .3 Results 4.3.1 Proximate analyses The test products were selected primarily to establish a domain of food compositions (Table 4. 1) that reasonably covers LMFs that have been associated with previous Salmonella outbreaks or recalls. In this study, almo nd meal and peanut butter represented nut products, typically rich in fats and proteins (Venkatachalam & Sathe, 2006) ; date paste represented dried fruits that are rich in natural simple sugars and dietary fibers (Ahmed & Al - jasass, 2014) ; wheat flour represented grain - based products, high in carbohydrates (Anonymous, 2017) ; nonfat dry milk powder represented a wi de range of spray - dried milk and dairy - based powders that are high in protein and lactose (Sharma et al., 2012) ; ground black pepper represented s pices, which contain antimicrobial compounds (Meghwal et al. , 2012) . Hence, evaluating multiple, diverse LMF compositions in a single study yielded a broader, systemmatic evaluation of the efficacy of E. faecium NRRL B - 2354 as a biological validation tool in thermal treatment of LMFs. 61 Table 4. 1 Moisture, native a w , and chemical composition of tested low - moisture products Standard deviation is reported in parentheses (n=2) . a Native a w represents a w at retail condition, measured once the package was received . b Total soluble solids of date was 64.9 ± 5.2°Brix. 4.3.2 Background microbial counts Coliform counts (log CFU/g) and its corresponding standard deviation were reported as follows: <0.04 for black pepper, 2.0 (0.57) for dates, 2.4 (0.12) for nonfat dry milk powder , 4.6 (0.51) for almond, 4.8 (0.18) for wheat flour, and 2.0 for peanut butter (1 out of triplicates). Y easts and molds (log CFU/g) and its corresponding standard deviation were 3.3 (0.33) for peanut butter, 4.3 (0.38) for dates, 3.4 for wheat flour (1 out of triplicates), and <0.04 for both nonfat dry milk and black pepper. As expected, coliform and yeast/mold counts were higher in unprocessed products (almonds, dates, and wheat flour) as compared to processed products (peanut butter, nonfat dry milk powder, and black pepper). Prior steam pasteurization and the presence of va rious antimicrobial compounds in black pepper can help explain the apparent absence of coliforms, yeasts, and molds. Salmonella could not be quantified in any of the uninoculated samples (<0.04 log CFU/g). Overall, the low populations of background microbi al counts did not interfere with survivor enumeration, due to the high inoculum level. 62 4.3.3 Thermal come - up time and initial bacterial population The thermal come - up times measured for all produc ts were in the range of 107 to 363 s. The post - conditionin g populations of E. faecium and Salmonella (log CFU/g) ranged as follows: 8.0 to 8.4 and 8.1 to 8.4, respectively, for peanut butter; 7.7 to 8.1 and 7.9 to 8.1, respectively, for nonfat dry milk powder; 7.8 to 8.1 and 8.0 to 8.5, respectively, for almond m eal; 8.1 to 8.2 and 8.2 to 8.7, respectively, for wheat flour; 7.2 to 7.5 and 6.7 to 6.9, respectively, for ground black pepper; and 8.2 to 8.4 and 8.0 to 8.4, respectively, for date paste. The standard deviations within all batches across all products wer e below 0.3 log CFU/g (ranging between 0.05 and 0.24 log CFU/g), indicating that all laboratories achieved homogenous inoculation. Among all products, the lowest initial populations of E. faecium and Salmonella were seen in black pepper, possibly due to th e activation of water soluble antimicrobial compounds after inoculation (Wei et al., 2018) . Nevertheless, the inoculation level of black pepper for both microorganisms was still considered adequate to achieve a minimum 3 log reduction (CFU/g) during isothermal treatment. 4.3.4 Inactivation kinetics of Salmonella and E. faecium Salmonella (Fig 4. 2) and E. faecium (Fig 4. 3) survivor curves decreased with isothermal treatment time. Because t he majority of the individual isothermal survivor curves (138 of 22 9 ) exhibited a Weibull shape parameter that were not statistically different ( P > 0.05) than 1, the log - linear model was employed to analyze all the curves to enable simple, direct comparison of thermal resistance between Salmonella and E. faecium in this study. Using t he log - linear model, the global RMSE ranged from 0.23 to 0.78 log CFU/g across all of the products, with only 4 of 24 RMSEs greater than 0.6 log CFU/g, which is a reasonable model error, given that the standard deviation among replicate observations (i.e., the underlying experimental variability) across the 63 Figure 4. 2 Interlaboratory isothermal inactivation curves for Salmonella cocktail in: (A) nonfat dried milk powder - 0.25 a w , (B) peanut butter - 0.25 a w , (C) almond meal - 0.45 a w , (D) wheat flour - 0.45 a w , (E) ground black pepper - 0.45 a w , (F) date paste - 0.65a w , and the corresponding log - linear model. Data shown are from three independent batches from each laboratory. -6 -5 -4 -3 -2 -1 0 100 200 Time (mi n) Survivors (log N/N o ) A -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Time (min) Survivors (log N/N o ) B -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Survivors (log N/N o ) Time (min) C -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Survivors (log N/N o ) Time (min) D -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Time Survivors (log N/N o ) E -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Survivors (log N/N o ) Time F 64 Figure 4. 3 Interlaboratory isothermal inactivation curves for E. faecium cocktail in: (A) nonfat dried milk powder - 0.25 a w , (B) peanut butter - 0.25 a w , (C) almond meal - 0.45 a w , (D) wheat flour - 0.45 a w , (E) ground black pepper - 0.45a w , (F) date paste - 0.65a w , and the corresponding log - linear model. Data shown are from three independent batches from each laboratory. -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Survivors (log N/N o ) Time (min) A -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Time (min) Survivors (log N/N o ) B -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Time (min) Survivors (log N/N o ) C Survivors (log N/N o ) -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Time (min) D C -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Time (min) Survivors (log N/N o ) E Survivors (log N/N o ) -6 -5 -4 -3 -2 -1 0 50 100 150 200 250 Time (min) F 65 entire study was 0. 17 log CFU/g. The combined laboratory D 80 ° C comparisons (Fig 4 .4 ) demonstrated that E. faecium exhibited greater thermal resistance ( P < 0.05) than Salmonella in nonfat dry milk powder (~3× greater ), peanut butter (~2× greater), almond meal (~1.5× greater), black pepper (~2× greater) and dates (3× greater). However, E. faecium was slightly less resistant ( P < 0.05) (~10% lower D 80 ° C ) than Salmonella in wheat flour (Table 2), but with distinct results at the two laboratories that conducted these tests. D 80 ° C - WSU for Salmonella (10.4 ± 0.3 min) in wheat flour was significantly higher ( P < 0.05) than that observed for E . faecium (9.1 ± 0.1 min), but the D 80 ° C - IFSH for Salmonella (9.7 ± 0.3 min) and E. faecium (9.7 ± 0.1 min) were statistically equivalent ( P > 0.05). Figure 4. 4 D 80 °C (both laboratories) of Salmonella and E. faecium for each product . (* = E. faecium thermal resistance was higher ( P < 0.05) than that of Salmonella . ** = E. faecium thermal resistance was lower ( P < 0.05) than that of Salmonella ) . 28.9 100.2 57.6 133.5 26.1 34.2 10.1 9.4 1.5 3.2 0.5 1.5 0 20 40 60 80 100 120 140 Nonfat dry milk powder (0.25 a w ) Peanut butter (0.25 a w ) Almond meal (0.45 a w ) Wheat flour (0.45 a w ) Ground black pepper (0.45 a w ) Date paste (0.65 a w ) ° 66 4.3.5 Thermal resistance of Salmonella and E. faecium between laboratories Cross - laboratory D - values and z T - values were deemed reproducible when the corresponding parameters were statistically similar ( P > 0.05) for a given product at a given temperature across two laboratories. T he cross - laboratory D - values for Salmonella and E. faecium in peanut butter (IFSH - UNL) were reproducible at all temperatures (Table 4. 2). However, there were differences ( P < 0.05 ) across laboratories in the determined thermal resistance of b oth microorganisms in date paste (MSU - UGA) a nd nonfat dry milk powder (WSU - MSU) at all temperatures. The remaining product/thermal treatment combinations also yielded some differences ( P < 0.05 ) across laboratories in D - values as follows: D 80°C - Salmonella and D - E. faecium at all temperatures for almond meal (MSU - WSU); D 7 0°C - Salmonella , D 8 0°C - Salmonella and D 85°C - E. faecium for wheat flour (WSU - IFSH); and D 7 0°C - E. faecium for ground black pepper (UNL - IFSH). T hese cross - laboratory differences in D - values ranged from 6 to 50%, with an average difference of 23%. For all products, the cross - laboratory z T - Salmonella values were reproducible ( P > 0.05) (Table 4. 3). However, the cross - laboratory z T - E. faecium values were only reproducible ( P > 0.05) for nonfat dry milk powder and almond meal. There were cross - laboratory differences ( P < 0.05) in the z T - E. faecium for peanut butter, ground black pepper, wheat flour, and date paste; however, those differences were relatively small, ranging from 0.9 to 1.4 °C (nominal difference of 7 to 12%). Despite these cross - laboratory variations in z T - values, the combined - laboratory z T - Salmonella and z T - E. faecium values were significantly differe nt ( P < 0.05) from each other for all products except nonfat dry milk powder. 67 Table 4. 2 D - value s for Salmonella and E. faecium across all laboratories and products Standard error is in parentheses. D - value s between laboratories at a given temperature within the same micro organism and product. Product Temperature ( °C) Salmonella E. faecium D - value (min) D - value (min) Peanut butter (0.25 a w ) IFSH UNL IFSH UNL 85 27.3 (0.7) 27.7 (0.7) - - 90 11.7 (0.3) 11.4 (0.4) 22.1 (0.3) 20.3 (0.3) 95 5.9 (0.1) 6.6 (0.1) 8.9 (0.2) 8.6 (0.1) 100 - - 3.4 (0.1) 3.4 (0.1) Nonfat milk powder (0.25 a w ) MSU WSU MSU WSU 85 16.1 (0.4)* 13.5 (0.2)* 59.3 (3.1)* 46.2 (2.4)* 90 8.7 (0.4)* 6.7 (0.1)* 26.6 (1.5)* 21.1 (1.0)* 95 4.2 (0.2)* 3.6 (0.1)* 14.1 (0.7)* 11.8 (0.7)* Almond meal (0.45 a w ) MSU WSU MSU WSU 80 25.8 (0.3)* 27.3 (0.3)* 33.1 (0.5)* 36.1 (0.7)* 85 13.0 (0.2) 13.3 (0.1) 16.5 (0.3)* 18.9 (0.5)* 90 7.2 (0.1) 7.4 (0.1) 8.3 (0.2)* 9.1 (0.2)* Wheat flour (0.45 a w ) WSU IFSH WSU IFSH 70 45.5 (0.9)* 41.8 (1.4)* - - 75 19.0 (0.7) 20.3 (0.7) 25.3 (0.4) 25.2 (0.5) 80 11.1 (0.2)* 9.7 (0.3)* 9.9 (0.1) 9.9 (0.3) 85 - - 2.9 (0.1)* 3.7 (0.1)* Ground black pepper (0.45 a w ) UNL IFSH UNL IFSH 65 20.0 (0.7) 21.5 (0.7) - - 70 8.6 (0.3) 9.8 (0.4) 25.6 (0.8)* 33.7 (1.6)* 75 3.3 (0.1) 3.4 (0.1) 8.6 (0.3) 9.2 (0.4) 80 - - 3.3 (0.1) 3.3 (0.1) Date paste (0.65 a w ) MSU UGA MSU UGA 65 14.6 (0.5)* 11.1 (0.4)* - - 70 5.1 (0.2)* 4.1 (0.1)* 8.8 (0.2)* 5.3 (0.2)* 75 1.7 (0.1)* 1.4 (0.1)* 4.4 (0.1)* 2.7 (0.1)* 80 - - 1.8 (0.1)* 1.3 (0.1)* 68 Table 4. 3 z T for Salmonella and E. faecium across all laboratories (individ ually and combined) and products RMSE: root mean squared - error. Standard error is indicated in parentheses. Individual laboratory z T followed by Combined laboratory z T followed by dagger symbol ( microorganisms within the product. 4.3.6 Influence of product composition on thermal resistance In general, both microorganisms exhibited higher thermal resistance in products with lower a w (Fig. .4 4). To illustrate, the D 80 ° C for Salmonella in nonfat dry milk (0.25 a w ) was ~25 times greater than that for date paste (0.65 a w ). Moreover, the D 80 ° C values for Salmonella and E. faecium were higher in peanut butter (0.25 a w ) than in almond meal (0.45 a w ), at the same temperatures, despite similar fat and protein content (Table 4. 1). For products having the same a w , both microorganisms differed in thermal resistance across the products. For instance, the D 80 ° C for Salmonella was higher ( P < 0.05) in nonfat dry milk powder than in peanut butter at 0.25 a w . Product Salmonella E. faecium Individual laboratory z T (°C) Combined laboratory z T (°C) c Combined laboratory RMSE (log CFU/g) Individual laboratory z T (°C) b Combined laboratory z T (°C) c Combined laboratory RMSE (log CFU/g) Nonfat dry milk powder (0.25 a w ) MSU: 17.9 (0.7) 17.4 (0.3) 0.50 MSU: 16.5 (0.9) 16.6 (0.7) 0.74 WSU: 17.3 (0.4) WSU: 16.7 (0.9) Peanut butter (0.25 a w ) IFSH: 14.9 (0.3) 15.2 (0.3) 0.52 IFSH: 12.2 (0.1)* 12.6 (0.1) 0.35 UNL: 15.6 (0.5) UNL: 13.1 (0.2)* Almond meal (0.45 a w ) MSU: 18.0 (0.3) 17.7 (0.2) 0.28 MSU: 16.6 (0.3) 16.7 (0.3) 0.39 WSU: 17.6 (0.2) WSU: 16.7 (0.4) Wheat flour (0.45 a w ) WSU:16.0 (0.4) 15.9 (0.3) 0.54 WSU: 10.6 (0.1)* 11.3 (0.1) 0.38 IFSH: 15.6 (0.5) IFSH: 12.0 (0.2)* Black pepper (0.45 a w ) UNL: 13.2 (0.4) 12.9 (0.3) 0.48 UNL: 11.2 (0.3)* 10.7 (0.2) 0.63 IFSH: 12.5 (0.4) IFSH: 10.1 (0.3)* Date paste (0.65 a w ) MSU: 10.6 (0.2) 10.7 (0.2) 0.55 MSU: 15.1 (0.5)* 14.7 (0.5) 0.68 UGA: 11.0 (0.2) UGA: 16.5 (0.7)* 69 Similarly, for those products at 0.45 a w , the D 80 ° C values for Salmonella and E. faecium were largest in almond meal, followed by wheat flour and ground black pepper. 4.3.7 a w level at high temperature When heated to 80 °C, the a w of wheat flour, ground black pepper and almond meal yielded a significant increase ( P < 0.1) from 23°C (Table 4 .4 ). Table 4. 4 Water activity ( a w ) measurements at 23°C (before and after heating) and at 80°C (n=2) a Standard deviation is indicated in parentheses and means followed by asterisk (*) were significantly different - heating HT sensor 23 ° C and HT sensor 80 ° C) within the same product. Date paste was indicated with double asterisk (**) because the variance between two trials was 0. b a w c Increase of aw as measured by the HT sensor from 23°C (preheating) to 80°C w as indicated by (+) symbol, and decrease of aw was indicated by ( - ) symbol Product a w measurement a Pre - heating Aqualab b Pre - heating HT sensor 23°C HT sensor 80°C a w difference c Post - heating HT sensor 23°C Post - heating Aqualab b Nonfat dry milk powder (0.25 a w ) 0.26 0.26(0.01) 0.28(0.00) +0.02 0.21(0.00) 0.22 Peanut butter (0.25 a w ) 0.23 0.24(0.01) 0.26(0.01) +0.02 0.31(0.00) 0.20 Wheat flour (0.45 a w ) 0.43 0.44(0.01)* 0.66(0.01)* +0.22 0.50(0.00) 0.44 Ground black pepper (0.45 a w ) 0.45 0.43(0.01)* 0.62(0.01)* +0.19 0.53(0.00) 0.47 Almond meal (0.45 a w ) 0.45 0.45(0.01)* 0.51(0.00)* +0.06 0.48(0.02) 0.41 Date paste (0.65 a w ) 0.64 0.64(0.00)** 0.61(0.00)** - 0.03 0.66(0.00) 0.58 70 4.4 Discussion This interlaboratory study aimed to evaluate the efficacy of E. faecium as a surrogate for Salmonella in the thermal processing of multiple LMFs, with the goal of generating reproducible results across laboratories. Prior to beginning any work, the study d esign and experimental protocols (including culture maintenance, inoculum preparation, and isothermal treatment protocols) were thoroughly reviewed and standardized for all five institutions. Subsequently, each laboratory established a standard inoculation protocol for the assigned product, given that each LMF possessed specific challenges concerning inoculation and equilibration. A series of criteria were included in the study design to reduce uncertainty in microbiological measurements (Weissfeld, 2010) . For example, the lower plate count of 25 CFU was selected to improve plate count accuracy (Jongenburger et al., 2010) , which in turn improved model parameter estimation (Garce & Marks, 2014) . An upper limit of 0.3 log CFU/g was established for homogeneity of the inoculum within the sample batch (n=10), given that cell counts varying by no more than ± 0.5 log CFU/g are generally deemed as being not significantly different (European Union, 2005) . The inoculums were prepared using the broth to lawn - harvest method, because the thermal resistance of Salmonella in wheat flour was previously reported to be mor e reproducible between two laboratories using this procedure (Hildebrandt et al., 2016) . Black peppercorn was inoculated with 2 mL instead of 1 mL of inoculum to achieve a population of ~ 8 log CFU/g, which helped to account for the microbial reduction after grinding and during the come - up time, so that the targeted 3 - 5 log reduction could be achieved. Wheat flour and nonf at dry milk powder were inoculated with a seeded sample to prevent product clumping. Owing to the attention given to standardizing the methodology, our cross - laboratory study was specifically designed to provide 71 more reliable conclusions compared to other single - laboratory studies (Ceylan & Bautista, 2015; Channaiah et al., 2016; Rachon et al., 2016; Shah et al., 2017) . Based on the combined - laboratory D 80 ° C values, E. faecium exhibited greater thermal resistance than Salmonella in peanut butter (0.25 a w ), nonfat dry milk powder (0.25 a w ), almond meal (0.45 a w ), ground black pepper (0.45 a w ) and date paste (0.65 a w ), but slightly lower heat resistance than Sal monella in wheat flour (0.45 a w ) (Fig. 4). T hermal resistance of both microorganisms for all products were compared at a T ref of 80°C, given that four of the six products were tested at 80°C, with this temperature also having been used in previous inactiv ation studies with LMFs (Limcharoenchat et al., 2018; Liu et al. , 2018; Syamaladevi et al., 2016) . Based on our cross - laboratory validation study, E. faecium was more thermally resistant than Salmonella in all products except wheat flour, despite variations in the D - and z T - values between laboratories. For instance, in nonfat dry milk powder, the D - E. f aecium and D - Salmonella from MSU were significantly higher than those from WSU at all temperatures, and, in date paste, the D - E. faecium and D - Salmonella from UGA were consistently lower than those from MSU (Table 2). Regardless of these case - specific D - v alue variations, both laboratories demonstrated that E. faecium was more thermally resistant than Salmonella in nonfat dry milk powder and date paste. Although the cross - laboratory z T - E.faecium values also were statistically different in peanut butter, wheat flour, and ground black pepper, the relative differences were quite low (average ~9.5%) . Nevertheless, the difference in z T - values between laboratories did not change the fact that E. faeciu m was more resistant than Salmonella in all products except wheat flour (where the difference between the two organisms was only ~10%). Ideally, standardized methodology should yield reproducible results (I. M. Hildebrandt et al., 2020) . We have observed several variabilities in the D - E. faecium and D - Salmonella values between 72 laboratories in this study. Nevertheless, given that our study were executed using a strictly - controlled proto col, variations in D - values did not entail that the results were irreproducible between laboratories, instead these varibilities emerged as a results of adhering to standardized methodology, as evidenced by relatively lower standard errors than those repor ted in the previous Salmonella thermal inactivation studies. In fact, it should be more concerning if D - values posing high standard error that were obtained from two independent studies appeared to be statistically similar with each other. In such scenario , the sources of variability could be challenging to be identified because both studies most likely used different experimental protocols. Most importantly, high variability of D - values could mask the effect of independent parameter studied as demonstrated by van Asselt and Zwietering (Van Asselt & Zwietering, 2006) . Our results showed that the food matrix affected the thermal resistance of Salmonella and E. faecium in LMFs (H. Li et al., 2014) and that thermal resistance of E. faecium and Salmonella generally increases at lower a w , as previously reported (Archer , Jervis, Bird, & Gaze, 1998; Liu, Tang, et al., 2018; Syamaladevi, Kiran, et al., 2016; Villa - rojas et al., 2013) . At the same a w , h igher bacterial resistance in peanut butter, as compared to nonfat dry milk powder, was presumably due to thermal protec tion afforded by lipids during heating (Juneja, Eblen, & Marks, 2000; Kataoka et al., 2013; Syamal adevi, Kiran, et al., 2016; Syamaladevi, Tang, Villa - Rojas, et al., 2016) . This study showed that a w of nonfat dry milk powder behaved similarly to peanut butter at 80°C. Although composition of almond meal was similar to peanut butter, almond meal showed 13% increase of a w at 80°C. Similar a w increased for ground black pepper (51%) was thought to be due to similar carbohydrate and protein content of ground black pepper to that wheat flour (Meghwal et al., 2012) . While it is important to measu re the sample a w at treatment temperature , 73 the changes of a w at elevated temperature may not be the dominant factor controlling thermal resistance of Salmonella in the LMFs tested. A desirable surrogate should exhibit higher thermal resistance than Salmon ella (Hu & Gurtler, 2017) . Based solely on this characteristic, previous studies corroborate E. faecium as a viable Salmonella surrogate for thermal processing of LMFs (Ceylan & Bautista, 2015; Channaiah et al., 2016; Enache et al., 2015; Liu, Rojas, et al., 2018; Rachon et al., 2016; Shah et al., 2017) . - case - consider the limitations of this premise, given that longer treatment times or increased treatment temperatures can negatively affect product quality. For ins tance, Wright et al. , 2018) observed changes in the texture and viscosity of commercial peanut butter and formulated tree nut butters in glass jars when treated at > 90 ° C. Therefore, the goal of the surrogate being more thermally resistant than Salmonella is not an unbounded criterion and must be balanced against practical con siderations in conducting commercial process validations. Currently, the only standard practices for utilizing a surrogate in thermal process validation presume that a process that achieves the lethality performance standard with the surrogate therefore is sufficient for the Salmonella (Almond Board of California, 2014) . Although there could be defendable rationale for utilizing other surrogate - to - patho gen statistical relationships in interpreting surrogate lethality results, there are no such published standards methods. Previous studies demonstrated that E. faecium exhibited greater thermal resistance than Salmonella in peanut butter (Enache et al., 2014) , almonds (Jeong et al., 2011) , and black pepper (Wei et al., 2018) , which is consistent with the outcomes of this study. Rachon et al. ( 2016) reported that E. faecium was also more thermally resistant than Salmonella , Escherichia coli , and Listeria monocytogenes in four low - moisture matrices . The exceptional thermotolerance of E. 74 faecium in low - environmental stresses, including elevated temperature, acidic and alkaline conditions, detergents, heavy metals, high osmolari ty, and desiccation (Ramsey et al., 2014) . Like other enterococci, E. faecium has a complex cell wall structure (Gao et al. , 2018) and is also capable o f expressing extensive cell wall peptidoglycan cross - linkages (Oliver, 2010; Pfeffer et al., 2006) as well as generalized stress response proteins, including as GroEL and DnaK chaperones (Ramsey et al., 2014) after weeks of exposure when in a viable but not culturable (VBNC) state. Enterococci are also capable of changing their cell membrane unsaturated - to - satura ted fatty acids ratio, leading to increased membrane fluidity and enhanced protection during heat stress (Fisher & Philips, 2009) , as also observed in Salmonella (Porta et al., 2010) . Given the requirement to validate thermal processes for LMFs, food processors can still gain valuab le insight into the thermal resistance of both Salmonella and E. faecium in LMFs from the previously published literature. However, the lack of standardized methodologies in these studies introduces a degree of uncertainty. Our cross - laboratory D 90 ° C - Salmonella values for peanut butter (0.25 a w ) were 27.3±0.7 min for IFSH and 27.7±0.7 min for UNL. However, a larger nominal difference was observed in D 90 ° C - Salmonella values for peanut butter from two prior independent studies, which were 2.25±0.2 min at 0.40 a w (Li et al., 2014) and 13.4±0.9 min at 0.45 a w (Ma et al., 2009) . In fact, it would be expected that D 90 ° C - Salmonella at 0.40 a w would be higher than D 90 ° C - Salmonella at 0.45 a w , but other experimental factors obviously contributed to the opposite cross - study inference. Additionally, basing process validations on incompletely documented experimental conditions, as found in much of the publi shed literature, remains problematic. For this reason, standardized methodologies that employ a cross - validation approach can enhance the validity and robustness of thermal inactivation data. 75 LMFs covers a broad range of food compositions. Food matrices in this study were selected to cover most common LMFs that have been associated with previous Salmonella outbreaks and recalls. Therefore, for other products that are close within the range of c omposition of the products in this study, limited additional data might be sufficient to confirm the acceptability of E. faecium as a surrogate and therefore to use it in commercial process validations. For example, cashew butter would be of similar compos ition as the peanut butter in this study, so that the relative D - and z - values for E. faecium and Salmonella should be close to those observed for peanut butter. In contrast, for other products significantly different from those in this study, such as drie d green tea powder, a thorough set of experimental trials would be needed to identify the critical factors affecting thermal resistance of both E. faecium and Salmonella , and thereby confirming the acceptability of E. faecium as a Salmonella surrogate for thermal process validations . 4. 5 C onclusion E. faecium was generally more thermally resistant than Salmonella in the low - moisture foods tested in this study ; however, the magnitude of differences in heat resistance between these two microorganisms was s trongly influenced by the food matrix. Our interlaboratory, multi - product data support E. faecium NRRL B - 2354 as a fairly robust surrogate for Salmonella within the product composition domains tested . Food processors should be able to utilize E. faecium NRRL B - 2354 as a biological validation tool for the products tested in this study; however, care should be taken when interpreting and applying the isothermal results for LMF commercial process validat ions, where other factors (e.g., dynamic product moisture content, process humidity, etc.) may affect the relative thermal resistance of these two organisms . 76 5. EFFECT OF LACTOSE AND PROTEIN ON THERMAL RESISTANCE OF ENTEROCOCCUS FAECIUM NRRL B - 2354 AND SA LMONELLA IN DAIRY POWDERS 5.1 Introduction Dairy powders are very common low - moisture ingredients for powdered infant formula, supplement beverages, confectionery powders , baked - goods, instant soup s and gravy , ice cream, cheese , and yogurt formulations . Da iry powder variants , such as skim milk powder, lactose - free skim milk powder, lactose powder , and milk protein isolate , are incorporated as value - added ingredient s in food formulation s to achieve desired functional properties or to provide milk nutrients to s pecific cluster s of individuals , such as infant s and those who are lactose intolerant . The different milk fractionation technologies allow the use of milk components as functional ingredient s in other food s (Thomas et al., 2004) , and maximize the use of milk constituents during dairy processing (Burke et al., 2018) . Furthermore, t he m anipul ation of liquid milk components involves enzymatic reaction s , or micro , ultra - or nano - filtration, after milk pasteurization but before spray - drying (Sharma et al., 2012) . G iven that there is no pathogen lethality step after spray - drying, dairy powders contaminated during the post - processing stage can be one of the major sources of foodborne illness, particularly for immunocompromised individuals. I n the period from 199 3 to 2019 , at least seven major Salmonella outbreaks have been associated with dairy powders worldwide (Centers for Disease Control and Prevention, 1993; Threlfall et al., 1998 ; Park et al., 2004 ; Brouard et al., 2007 ; Rodríguez - Urrego et al., 2010 ; European Food Safety Authority, 2018; Jones et al., 2019) . For instance, i n 2018, a global Salmonella Agona ou tbreak traced to infant formula has sickened 39 infants in France , and one infant in Greec e and Spain , respectively. The impact of this outbreak also was devastating economically since the product had to be recalled from 63 countries (European Food Safety 77 Authority, 2018) . In the following year , powdered infant formula manufactured in France , containing Salmonella Poona, sickened 31 infants with 13 hospitalizations (Jones et al., 2019) . The same strains of Salmonella Agona and Salmonella Poona from these outbreaks were isolated from the processing facilities, with these same str ains also responsible for outbreaks dating back to 2005 (Brouard et al., 2007) and 2010 - 2011 (Rodríguez - Urrego et al., 2010) outbreak s, respectively . Therefore, there is a n eed to validate post - processing thermal inactivation strategies for dairy powders due to recurring Salmonella contamination events. The use of a surrogate microorganism for LMF process v alid ation process is seen as a feasible food safety intervention , because the use of a pathogenic bacteria is clearly prohibited in the food processing facilities . Ideally, the surrogate microorganism should pose similar or higher thermal resistance than that of Salmonella at any given testing parameters. Additionally, the surrogate should also pos sess a significant record of non - pathogenicity, be able to maintain a high populat ion in the product after inoculation, not induce spoilage to the food product, be easy to prepare, be readily (Busta et al., 2003) . Enterococcus faecium NRRL B - 2354 (ATCC 8459) is a nonpathogenic, non - spore forming lactic acid bacteri um . E. faecium can withstand various adverse conditions such as 30 min of and grow at both extreme pH (9.6) and elevated salt concentrations (6.5% NaCl) (Gira, 2002) . Originally of dairy origin (Hu & Gurtler, 2017) , E. faec i um NRRL B - 2354 found its application in food as a starter culture in dairy, meat, and vegetables fermentation s (Gira, 2002; Kornacki, 2012) . Genomic relatedness of E. faecium NRRL B2354 with clinical E. faecium strains is superficial, due to the indiscernible virulence and antibiotic resistance genes expressed by E. faecium NRRL B - 2354 (Kopit et al., 2014) . Based on these sa fety evaluations and its resistance to adverse conditions, E. faecium has been utilized as a potential Salmonella surrogate in low - 78 moisture validation studies, which have overall demonstrated greater thermal resistance of E. faecium as compared to Salmonel la (Bianchini et al., 2014; Ceylan & Bautista, 2015; Channaiah et al., 2016; Enache et al., 2014; Jeong et al., 2011; Liu et al., 2018; Rachon et al., 2016) . I ntrinsic and extrinsic factors affecting Salmonella thermal resistance are important for representative validation output. These factors include Salmonella serotype (Diez - G onzalez et al., 2017; Pena - Melendez et al., 2014; Shachar & Yaron, 2006) , product water activity (Bianchini et al., 2014; Channaiah et al., 2016; Enache et al., 2015; Smith & Marks, 2015; Syamaladevi et al., 2016) , treatment temperature (Podolak et al. , 2017; Rachon et al., 2016; Villa - rojas et al., 2013) , inoculum preparation (Enache et al., 2015; Hildebrandt et al., 2016) , and recovery media (Busta et al., 2003) , as well as the structure and che mical composition of the food products (Limcharoenchat et al., 2019 ; Enache et al., 2014; Jin et al., 2018; Li et al., 2014; Rachon et al., 2016) . In term s of food composition, p revious low - moisture studies have focused on the effect of fat, carbohydrate, and protein in lipid - rich LMF matri ce s (Enache et al., 2014; He et al., 2011; Jin et al., 2018; Li et al., 2014; Li et al., 2014) . Others have investigated the influence of sugar solutes using low - a w broth system s (Mattick et al., 2001; Pena - Melendez et al., 2014) . Because food composition can affect thermal resistance of Salmonella , there is a need to evaluate the effect of sugar and protein on thermal resistance of Salmonella and its surrogate microorganism in dairy powder variants , given that the proportion s of sugar and protein in milk are altered to produce low - fat dairy powder variants . The r elative thermal resistance of E. faecium to Salmonella , can be referred to as the kill ratio. Within this context, the kill ratio can be defined as the relative D - value of E. faecium as compared to Salmonella . For instance, the kill ratio of E. faecium to Salmonella in nonfat dry milk powder at 80 °C was ~3 (section 4.3.6) . Previous results also suggest that the kill ratio s for E. 79 faecium and Salmonella were greatly influenced by food composition. For high - suga r LMF such as nonfat dry milk powder (~56% lactose) and date paste (~65% simple sugars in the form of glucose, fructose, and sucrose), a kill ratio of ~3 was observed , whereas a high - fat and high - protein LMF such as almond meal exhibited a kill ratio of ~1 .5 . T he magnitude of D 80 °C for both microorganisms in nonfat dry milk powder w as greater than that for date paste, despite the similar kill ratio , where components of nonfat dry milk powder were negatively impacted due to browning and caking. Thus far, there is no establish ed kill ratio to justify the acceptability of E. faecium as a Salmonella surrogate in validating thermal processing of LMF. Because food composit ion influenced the kill ratio of E. faecium and Salmonella in LMF, the feasibility of E. faecium as a Salmonella surrogate in thermal processing of dairy powders as affected by lactose and protein compositions, needs to be assessed. T hermal resistance of Salmonella is impacted by the types of sugar found in LMF (Mattick et al., 2001; Pena - Melendez et al., 2014) , however the extent of thermal protect ion by different sugar solutes was not addressed. In comparing thermal resistance between E. faecium and Salmonella in dairy powder, sugar ut ilization may differ during resuscitation of survivors on recovery media with the survivor counts affect ing the estimat ed D - value and ultimately the thermal resistance of both microorganisms. Because Enterococcus spp. are capable of metabolizing 13 differ ent sugars, including lactose, D - glucose and galactose (Manero & Blanch, 1999; Ramsey et al., 2014) , greater thermal resistance in SMP might be due to the inherent characteristics of E. faecium in fermented dairy products (Bonacina et al., 2017; Gira, 2002) . In this study, the relative thermal resistance of E. faecium and Salmonella in dairy powders as a function of sugar type w as investigated to justify the use of E. faecium as a Salmonella surrogate in validating thermal treatment of dairy powders . 80 In a previous study, Syamaladevi et al. ( 2016) speculated that a w changes at elevated tempera ture from the rearrangement of water molecules distribution within food matrices might partly explain the enhanced thermal resistance of Salmonella in LMF . However, Based on the findings in chapter 4, no appreciable change in a w was observed in nonfat dry milk powders during heating. Since the quality of nonfat dry milk powder was negatively impacted at elevated temperature, physical changes in dairy powders during heat ing might better explain the magnitude of Salmonella thermal r esistance in dairy powder variants. One feasible approach to determine the physical changes is to measu re the glass transition temperature (T g ) of dairy powders during heating (Thomas et al. ,2004) . T g is the temperature at which dairy powders converted from the amorphous to crystalline state, with this change dictated by sugar composition (Schuck et al., 2005; Thomas et al. , 2004) . The overall aim of this study was to assess the impact of both lactose and protein on thermal resistance of Salmonella and E. faecium , particularly in skim milk powder (SMP), lactose - free skim milk powder (LSMP), lactose powder (LP), and milk protein isolate 90% (MPI ) via the following t hree objectives : ( 1) compare the thermal resistance of Salmonella and E. faecium in selected dairy powd ers at 0.25 a w at three treatment temperatures , (2) assess the effect of sugars (lactose vs. glucose - galactose) on bacterial survivor resuscitation on non - selective differential media using SMP and LSMP as LMF models , and (3) determine the a w changes and g lass transition temperature (Tg) of dairy powders at elevated temperatures . 81 5 .2 M aterials and method 5.2.1 Dairy powders Bovine spray - dried s kim milk powder (SMP) (batch no. 10MR01), lactose - free skim milk powder (LFSMP) ( batch no. 04MU01 ) , and lactose powders (LP) (batch no. 11MO09) were obtained from Prolactal (Hartberg, Austria). Milk protein isolate (MPI) (l ot no. 11900 ) was obtained from Idaho Milk Products Inc., ( Jerome, ID, US A). Proximate compositions for all products were su pplied by the manufacture r s . Su gar compositions of SMP and LSMP w as determined using hi gh p erformance l iquid c hromatography (HPLC) via a third - party laboratory service (Medallion Labs, MN) . Water activity was measured using an AquaLab series 4TE meter (Meter Group, Pullman, WA) in duplicates (Table 5.1) . All samples were kept sealed in their original packages and stored at room temperature. 5.2. 2 Overall experimental design Dairy powders were inoculated either with a Salmonella cocktail or E. faecium (1 mL inoculum per 100 g powder) , pre - equilibrated at 0.25 a w , ground to determine bacterial population homogeneity , and re - equilibrated at 0.25 a w prior to being subjected to isothermal treatment in a water bath ( GP - 400, Neslab, Newington, NH ) . For the first objective, t he samples were subjected to three heat treatments (based on preliminary findings), followed by recovery and enumeration of survivor s on nonselective - differential media after 24 h of incubation at 37 °C . 82 Table 5. 1 Proximate composition of dairy powders Products Moisture (%) Water activity c Fat (%) Lactose (%) Glucose (%) Galactose (%) Protein (%) Ash (%) pH Skim milk powder (SMP) 3.3 0.156 (0.001) 0.1 48 (0.7) a - - 37.7 7.9 6.6 Lactose - free skim milk powder (LSMP) 4.0 0.131 (0.004) 0.1 <0.01 b 26 (0.3) b 24 (0.9) b 38.0 8.3 6.4 Lactose powder (LP) 0.1 0.211 (0.001) - > 99 - - 0.1 0.1 6.4 Milk protein isolate 90% (MPI) 5.4 0.171 (0.005) 1.2 1.2 - - 91.6 6.7 6.7 Proximate analys e s for all products were provided by manufacturers , except as indicated with superscript letters. a,b Sugar profile s for SMP and LSFMP were obtained from a third - party laboratory. c Water activity w as measured using an AquaLab 4TE meter. The mean (standard deviation) values were resulted from duplicate measurements (n=2) . In the second objective, SMP and LSMP were heat - tr eated at a single temperature and then split into two separate samples for analysis. T he sugar composition of SMP and LSMP was then equilibrated in half of the samples by adding glucose - galactose to SMP (SMPX) , and lactose to LSMP (LSMPX) (Table 5.1) , af ter which survivors were enumerated on nonselective differential media at 37 °C for 24 h. For the third objective, the a w of non - inoculated dairy powders (SMP, LSMP, LP, and MPI) pre - equilibrated at 0.25 a w was measure d at 80 °C using a custom - designed, high - temperature a w meter (Meter Group, Pullman, WA, USA) (Syamaladevi et al., 2016) with the glass transition ( T g ) also determined by DSC. (detailed below). 83 5.2.3 Background microbial counts Aerobic plate counts and p resence of Salmonella were evaluated by plating a 1:10 dilut ion of the sample in 0.1% buffered peptone water ( BPW ) ( Difco, BD ) on Tryptic Soy Agar ( Difco ) supplemented with 0.6% (w/v) yeast extract ( Difco ), referred as TSAYE, and MTSA (TSAYE supplemented with 0.05% ammonium ferric citrate ( ACROS Organics, Morris, NJ) and 0.03% sodium thiosulfate ( Fischer Scientific, Waltham, MA), respectively. Any black colon ies on MTSA were counted as p resum ptive Salmonella e . Y east/mold background counts for all samples were obtained using 3M Petri fil m (3M Company, St. Paul, MN) . TSAYE and MTSA were incubated for 24 h ( 37 °C ) , whereas yeast/mold Petrifilm were incubated for 3 to 5 days ( 25 °C ). 5.2.4 Inoculum preparation Salmonella Agona 447967 (2008 rice puffed cereal outbreak), Salmonella Montevideo 488275 (2009 - 2010 black pepper outbreak), and Salmonella Mbandaka 698538 (2013 sesame tahini outbreak) were obtained from the FDA (ORA Arkansas Regional Lab, Jefferson, AR); Salmonella Tennessee K4643 (2006 peanut butter outbreak) was obtained from Dr. Larry Beuchat at the University of Georgia (Athens, GA); Salmonella Reading Moff 180418 (associated with cumin) was obtained from the FDA culture c ollection (FDA CFSAN, Bedford Park, IL). Enterococcus faecium NRRL B - 2354 was received from the Institute for Food Safety and Health (IFSH, Bedford, IL) and maintained at - 80°C in TSBYE containing 20% (v/v) glycerol. Using the method of Keller et al. ( 2012) , one vial of frozen stock (1.0 ml) was submerged in a 37°C water bath for 30 s, after which 0.1 ml was transferred to 9 ml of TSBYE and incubated for 24 h ( 3 7°C ) . A loopful of this overnight culture then was streaked to plates of TSAYE to obtain working stock cultures (transferred monthly) after 24 h of incubation ( 37°C ) . For each independent replicate , a 84 single colony from the working stock was transferred to TSBYE and incubated for 24 h ( 37°C ) , after which 1 ml was spread on a 150 mm diameter plate of TSAYE to obtain confluent growth after 24 h ( 37°C ) . The bacterial lawn was then harvested in 2.25 ml of BPW using a spreader. An equal volume of the 5 individual Salmonella strain s were combined as cocktail , and the inoculum was used within 2 h of preparation . 5.2.5 Inoculation of dairy powders A 100 - g s ample of each powder was mixed with 1 mL of either the Salmonella cocktail or E. faecium in a 710 mL VWR bag (VWR International, Canada). The bag was shaken and hand - massaged for 3 min, followed by mixing in a circular - motion using a spatula. The sample wa s pre - equilibrated at 0.25 a w for 24 to 48 h in a conditioning chamber (Hildebrandt et al., 2016) by spreading the powder into a thin layer i n a filter paper boat (~7 cm x 11 cm). At this point, the powder was dry , however , some clump ing could be observed . While in the conditioning chamber, the sample was transferred into a 3.8 L Whirl - Pak bag (Nasco ) , laid flat to spread the powder into a thin layer, and pulverized using a porcelain pestle (7.3 cm in length, HIC Harold Import Co., Lakewood, NJ) fo r 5 min . Thereafter the powder was collected in the bottom of the bag and hand - massag ed. The pulverizing and mixing steps were repeated at least three times until the powde r was free of visible clumps. The powder was then mixed with a spatula, transferr ed to a filter boat, and equilibrated at 0.25 a w (± 0.025) for another 1 to 2 days for SMP, LSMP and MPI and 5 to 10 days for LP. Homogeneity of the inoculum in the powder was confirmed by diluting three 1 - g subsamples in 0.1% BPW followed by plat ing on either MTSA or ETSA (TSAYE supplemented with 0.05% ammonium ferric citrate (Acros Organics, Morris, 85 NJ) and 0.025% esculin hydrate (Acros Organ ics, Morris, NJ) . The inoculum was considered homogenous if the standard deviation of the initial population was < 0.3 log CFU/g. 5.2.6 Come - up time and i sothermal treatment Prior to isothermal treatment, the come - up time needed to reach 0.5 °C below the treatment temperature was measured . The temperature of the water bath (GP - 400, Neslab, Newington, NH) was verified (± 0.2 °C ). A T - type thermocouple probe aluminum test cell (0.5 to 0.8 g, 4 mm thick; Washington State University, Pullman, WA) (Ch ung et al., 2008) was packed with uninoculated, equilibrated powder (~ 0.7 g), and connected to a handheld thermometer (Omega RDXL4SD, Norwalk, CT). A second temperature probe was connected to the handheld thermometer and immersed in the water bath. To mimic the experimental load, 18 similar test cells were immersed in the water bath for come - up time measurement . Six independent measurements were taken, with the mean plus two standard deviations was used as t0. The Inoculated s amples were packed in aluminum test cells , tightly sealed , and subjected to isothermal treatment at three different t emperatures (± 0.2 °C ), targeting 3 to 5 log reductions. Starting at t0, three subsamples were collected at each of 7 equally spaced time intervals per trial , and immediately immersed in an ice water bath for ~ 30 s to stop the thermal inactivation. The content s of each test cell was aseptically transferred to a Whirl - Pak bag (Nasco), diluted in 0.1% BPW, and pummeled for 3 min (NEU - TEC Group Inc. Famingdale, NY ). Isothermal trials were performed in triplicate using three biologically independent batches. For the second objective , SMP and LSMP were treated at 90 °C and 70 °C , respectively. Triplicate samples were collected at five equally spaced time intervals for each of three trials as 86 in objective 1. Thereafter the samples were dilut ed (1:10 dilution) in 0.1% buffered peptone water (BPW), and pummeled using a stomacher (NEUTEC Group Inc., Farmingdale, NY) for 3 min, after which 5 mL of the diluted SMP or LSMP was transferred to a second 118 mL Whirl - Pak bag (Nasco, Fort Atkinson, WI) containing 0.1 g each of glucose and galactose powder (Sigma - Aldrich, St. Louis, MO) ( SMPX) or 0.2 g of lactose powder for LSMP (LSMPX) with SMP and LSMP serving as controls. The amou nt of added sugar was calculated based on the sugar composition of SMP and LSMP in Table 5.1. 5.2.7 S urvivor enumeration and estimation of D and z T value s Survivors were recovered by spread plating 0.1 mL of a 1:10 diluted in BPW on MTSA and ETSA , for Salmonella and E. faecium counts, respectively , with all black colonies counted after 24 h of incubation at 37 °C . A cceptable colony counts were in the range of 25 to 250 colonies, yielding a limit of quantification of 25 00 CFU /g . To estimate the D and z T values, the l og - linear and Bigelow model s were fitted to log - transformed survivor counts (log CFU/g) using a nlinfit function in MATLAB (version 2017a, MathWorks Inc., Natick, MA), via one step global regression (Equation 4.1) . Temperature reference, T ref , of 90 °C was used in the Bigelow model for comparing D - values of Salmon ella and E. faecium across powders and microorganisms. Root mean squared error (RMSE) of the model parameters w as also estimated . When t he Weibull model (Jin et al., 2018) was also fitted to the isothermal data , the shape parameter was significantly different from a value of 1, signifying a log - linear trend. as performed to determine the effect of temperature and organism on D - values , whereas the effect 87 of organism based on D 90 °C and z T w as - test due to unbalanced sampling size s and unequal variances (Welch, 193 8) . 5.2.8 Measurement of a w at 80 °C The measurement of a w at 80 °C was performed according to the method described in section 4.2.8. Briefly, n on - in oculated dairy powders at 0.25 a w ( ± 0.025 ) w ere placed in a metal a w cup with the lid on, sealed with parafilm, packed in a double layer of Whirl - Pak bags, and placed in a cooler to avoid a w changes. Prior to measurement, the sealed samples were placed on the bench for equilibration at room temperature ( 23 °C ) for 15 min . After calibrating the high temp erature a w sensor using 0.25 and 0.50 a w standard solution s , the sample was loaded in to the HT sensor chamber until it reached a constant room temperature ( 23°C ) . Thereafter, the HT sensor was placed in an oven at 80°C (± 0.2°C) . A fter reaching and remaining at 79.5°C for 10 min, the a w was recorded, and the sample chamber was removed from the oven and cooled to room temperature (23°C) . Thereafter, the sample cup was removed from the chamber, capp ed, wrapp ed in parafilm, and plac ed in a Whirl - Pak bag for post - heating a w measurement using the Aqualab a w m eter (Decagon) . All measurements were performed in tri plicate. The changes in a w between 23 °C and 80 °C were evaluated using a t - test 5.2.9 Glass transition measurement using DSC Non - inoculated dairy powders at 0.25 a w (± 0.025) w ere placed in a a w cup with the lid on, sealed with parafilm, and double bagged in Whirl - Pak bags to avoid a w changes. F or DSC measurement, a 5 mg (± 1 mg) sample was loaded in a hermetic aluminum pan (#900793.901, TA Instruments, New Castle, DE , USA ) , covered with a aluminum hermetic lid (#900794.901, TA 88 Instruments), and sealed twice using a hermetic pan press (Tzero press, TA Instruments) . Th e pan was carefully placed in side the sampl ing chamber of the DSC (Q1000 series, TA Instruments) using tweezer s . A reference pan (an empty sealed hermetic aluminum pan ) was placed in the neighboring disk as a control . The set temperature for DSC before analysis was 4 2.5 °C . The powders were first cooled to 20 °C , at a scanning rate of 10 °C /min and then heated from 20 °C to +1 8 0 °C at a rate of 10 °C /min. Tg range was determined from the DSC curve of the heating cycle and analyzed using TA universal analysis 2000 software (TA Instruments ). All DSC measurements were performed in duplicate . 5.3 R esults 5.3.1 Proximate analysis and background microbial counts The composition and a w of all dairy powders are shown in Table 5. 1. Salmonella was not detected (<0.4 log CFU/g) in any of the uninoculated powders . For APC , only LP yielded 2 log CFU/g for both replicates , with the remaining powders having counts below the limit of detection (<0.4 log CFU/g) . Excep t for one replicate of SMP (3 log CFU/g), y east and molds were not detected . 5.3.2 Initial populations and come - up time Initial population s in the inoculated samples range d from 7.7 to 8.0 and 8 .4 to 8.5 for Salmonella and E. faecium , respectively, in SMP; 7.7 to 8.0 and 8.4 to 8.6 for Salmonella and E. faecium , respectively, in LSMP; 8.3 to 8.6 and 7.6 to 8.2 for Salmonella and E. faecium , respectively, in MPI; and 7. 6 to 7.8 and 8.0 to 8.4 for Salmonella and E. faecium , respectively, in 89 LP . Come - up time s were in the range of 221 to 316 s for SMP, 81 to 121 s for LSMP, 93 to 167 s for MPI, and 164 to 171 s for LP . 5.3.3 Isothermal inactivation curves Salmonella and E. faecium population s decreased in all powders during thermal treatment ( Fig 5. 1 ) . Linearity test ing showed that 51 of 6 9 inactivation curves for both Salmonella and E. faecium were linear. Except for D 95 °C for SMP, D - E . f aecium w as greater ( P < 0.05) than D - Salmonella for SMP, LP, and MPI at all treatment temperatures . However , the inverse trend was observed for LSMP at all temperatures (Table 5.2). The comparison of D 9 0 °C between microorganism for SMP, LP, and MPI demonstrated that E. faecium was more thermally resistant ( P < 0.05) than Salmonella ; however , the D 9 0 °C for both microorganisms in LSMP w as statistically similar ( P > 0.05). The z T values were significantly different between microorganisms ( P < 0.05) only for SMP and LP, in which z T - Salmonella was higher than z T - E. faecium for both powders. F ive out of 8 RMSE values were less than 0.6 log CFU/g , indicating the appropriateness of the model for fitting the isothermal inactivation data . 5.3.4 Effect of lactose and protein content At the reference temperature of 90 °C , the estimated D 90 °C values for both microorganisms w ere greater in SMP ( P < 0.05) than in LSMP, despite similar protein content (~38%) (Table 5. 2 ) . D 90 °C values for both microorganisms were also greater in SMP (48.9% lactose) than in LP (> 99% lactose), which most likely due to the the presence of protein in SMP (~38% protein). In powder having the highest protein content (91.6%), the D 90 °C of both microorganisms was substantially lower in MPI than in SMP, but substantially higher than LSMP. 90 For the first objective, the relative thermal resistance of E. faecium to Salmonella can be expressed as the ratio of D 90 °C - E. faecium to D 90 °C - Salmonella . For SMP, D 90 °C - E. faecium was almost 2 × greater than that for Salmonella (18.4 min and 10.6 min, respectively). A similar kill ratio was observed for LP, in which D 90 °C - E. fa ecium was greater than D 90 °C - Salmonella (7.3 min and 2.9 min, respectively), even though LP had almost twice the lactose content (> 99%) compared to SMP (48.9%). For MPI, the D 90 °C - E. faecium to D 90 °C - Salmonella kill ratio was ~1.6. N o difference ( P > 0.05) was seen between D 90 °C - E. faecium and D 90 °C - Salmonella in LSMP. Based on these findings , sugar composition impacted thermal resistance of both microorganisms in SMP and LSMP, w ith lactose most likely protecting both microorganisms, particularly E. faecium in SMP . While protein content alone might not enhance thermal resistance , the kill ratio of E. faecium and Salmonella was influenced by the presence of lactose in dairy powders. 91 Figure 5. 1 Isothermal inactivation curves for the Salmonella cocktail and Enterococcus faecium NRRL B - 2354 ) in skim milk powder , lactose - free skim milk powder, lactose powder, and milk protein isolate 90% at 0.25 a w. 92 5 .3.5 Effect of sugars on resuscitation of survivor in SMP and LSMP The D - values for both microorganisms in SMP, SMPX, LSMP, and LSMPX are reported in Table 5.3. E. faecium exhibited similar thermal resistance (P > 0.05) in SMP and SMPX, whereas Salmonella was more therm ally resistant in SMP (P < 0.05) (~14% greater) as compared to SMPX, at 90 °C . Both E. faecium and Salmonella exhibited similar thermal resistance in LSMP and LSMPX, at 70 °C . For the second objective, the relative thermal resistance s of E. faecium and Salmonella (kill ratio) in SMP and LSMP were determined at 90 °C and 70 °C , respectively . Based on the results, the addition of sugars during survivor resuscitation did not impact t he kill ratio s for E. faecium and Salmonella in SMP and LSMP. T he kill rati o s in LSMP and LSMPX were almost equal (~1), indicating that addition of lactose did not influence resuscitation of survivors, and hence thermal resistance. Similarly, the kill ratio s for E. faecium and Salmonella in SMP and SMPX nearly equivalent (~2). Th ese results show that thermal resistance of Salmonella and E. faecium was not influenced by sugar composition during recovery . 93 Table 5. 2 D - and z T - values (T ref = 90°C) for Salmonella and E. faecium in skim milk powder, lactose - free skim milk powder, lactose powder , and milk protein isolate 90% at 0.25 a w Dairy powder Temperature (°C) Salmonella E. faecium D - value (min) a D ref (min) b z T (°C) c RMSE (log CFU/g) D - value (min) a D ref (min) b z T (°C) c RMSE (log CFU/g) SMP 85 21.7 (0.7)* 10.6 (0.2)* * 17.6 (0.7) 0.50 41.3 (2.9)* 18.4 (0.7)* * 12.9 (0.7) 0.57 90 10.8 (0.4)* 19.0 (1.2)* 95 5.9 (0.2) 7.4 (0.3) LSMP 65 38.2 (1.3)* 0.40 (0.0) 12.9 (0.3) 0.45 31.9 (0.9)* 0.50 (0.0) 13.7 (0.3) 0.50 70 15.6 (0.4)* 14.5 (0.4)* 75 6.4 (0.1)* 6.0 (0.1)* LP 85 5.0 (0. 4) * 2.9 (0. 1 )* * 22.2 ( 2.7 ) 0. 66 12.4 ( 0.7 )* 7. 0 (0. 3 )* * 1 7.5 (1. 5 ) 0. 64 90 2.8 (0.2)* 8.3 (0.5)* 95 1.8 (0. 2 ) * 3.2 (0.3) * MPI 80 43.3 (2.1)* 8.3 (0.2)* * 13.7 (0.4) 0.54 56.3 (2.7)* 13.2 (0.6)* * 15.3 (0.8) 0.62 85 20.2 (0.8)* 30.9 (1.6)* 90 8.2 (0.2)* 12.5 (0.7)* Standard error s are in the parentheses; RMSE is root mean squared error. a D - values followed by (*) are significantly temperature, within the same product . b D ref followed by (* * ) . c z T followed by dagger symbol ( microorganisms within the same product . 94 Table 5. 3 D - values for both microorganisms in skim milk powder (SMP) at 90 °C and lactose - free skim milk powder (LSMP) at 70 °C . Microorganism Powders D - value (min) RMSE (log CFU/g) 95% CI Interval (min) 90 °C Salmonella SMP 8.4 (0.3)* 0.32 [7.8 , 9.0] SMP X 7.4 (0.3) * 0.28 [6.9, 7.8] E. faecium SMP 16.1 (1.3) 0.81 [13.4, 18.9] SMP X 14.4 (1.0) 0.74 [12.5, 16.4] 70 °C Salmonella LSMP 14.6 (0.4) 0.37 [ 13.7, 15.5 ] LSMP X 14.6 (0.3) 0.32 [14.0 , 15.2] E. faecium LSMP 12.7 (0.4) 0.60 [11.8, 13.6] LSMP X 13.1 (0.4) 0.59 [12.3, 14.0] X indicates added sugar ; gl ucose and galactose for SMP and lactose for LSMP . RMSE= Root mean square d error Asterisk (*) indicates significant different between D - value within the same powder and microorganism. 5.3. 6 A w of dairy powders at 80 °C . Water activity values for of SMP and LSMP at room temperature (23 ± 1 °C ) and 80 °C were statistically similar ( P > 0.05) (Table 5.4) . A w values for LP at 80 °C decreased ( - 0.11) , whereas those for MPI increased (+0.09). The a w values for all products at room temperature before and a fter heat treatment w ere similar ( P > 0.05), except for LP, which was highly variable during measurement as compared to the other dairy powders. After heating at 80 °C , only SMP and LSMP exhibited physical changes in the form of caking and browning. 95 Table 5. 4 Water activity (a w ) at 23°C (before and after heating) and at 80°C (n= 3 ) Dairy powder a w measurement Visual assessment on observed change after 80°C Pre - heating AquaLab b 23°C pre - heating HT senso r a 80°C HT sensor a a w differences c 23°C Post - heating HT sensor Post - Heating AquaLab SMP 0.25 0.22 (0.02) 0.25 (0.01) + 0.03 0.21 (0.00) 0.22 Yes LSMP 0.23 0.20 (0.01) 0.21 (0.00) + 0.01 0.19 (0.00) 0.24 Yes LP 0.27 0.30 (0.01)* 0.19 (0.02)* - 0.11 0.25 (0.02) 0.48 No MPI 0.25 0.27 (0.02)* 0.35 (0.01)* + 0.08 0.25 (0.01) 0.26 No a Standard deviation is in parentheses , means followed by asterisk (*) are column s across the same row (pre - heating HT sensor 23 ° C and HT sensor 80 ° C) within the same product (n=3). b Powder a w values ( ± 0.025 a w ) w ere determined in duplicate and standard deviation s . c An i ncrease in a w from pre - heating HT sensor 23 ° C to 80 ° C is indicated by (+) symbol and a decreased in a w is indicated by ( - ) symbol 5.3. 7 G lass transition temperature (T g ) of dairy powders G lass transition (T g ) (Table 5.4) w as only observed for SMP and MPI, not for LSMP or LP. The melting peak s for LSMP and LP indicate that physical changes occurred across the - 20 to 180 ° C heating cycl e (Table 5.4) . Table 5. 5 Glass transition and melting temperatures for dairy powders Glass transition, Tg ( ° C) Melting peak ( ° C) Powders Tg i Tg Tg e Tp i Tp Tp e SMP 43.1 (0.3) 58.3 (0.3) 74.3 (0.1) NA NA NA LSMP NA NA NA 152.1 (3.7) 162.6 (6.8) 174.9 (3.0) MPI 109.8 (0.5) 138.7 (7.1) 160.9 (13.3) NA NA NA LP NA NA NA 130.0 (0.4) 150.1 (0.1) 176.9 (0.007) Standard deviation is in parenthesis . Tg i = Tg initial and Tg e = Tg end; Tp i = peak initial and Tp e = peak end 96 5. 4 D iscussion The composition of dairy powders is globally standardized (World Heal th Organization & Food and Agriculture Organization of the United Nations, 2011) . L actose is typically used as an alternative to sucrose in baked products, because the relative sweetness of lactose to sucrose is 0.2 to 0.4 (Davis, 1995) . Milk proteins (composed of 80% casein and 20% whey) are used as foaming, emulsifying, gelling, and water - adsorbing agent s in foods such as baked goods, instant soups, gravy, beverages, salad dressings and ice cream ( Thomas et al., 2004) . Skim milk powder , comprised of milk protein and lactose, is commonly used as bulking agent since end - product s weetness is only marginally affected . Lactose - free skim milk powder can provide essential nutrients to thoese who are lactose intolerant (Storhaug et a., 2017) . Since d airy powders are important ingredients in many food formulations and prone to post - processing contamination, any Salmonella reduction step for dairy powders need s to be properly validated . In this study, SMP , LSMP, MPI , and LP were selecte d to represent varying lactose and milk protein compositions in commercially available dairy powders . SMP and LSMP were used to compare the effect of l actose and hydrolyzed lactose (equal amount of glucose and galactose) on thermal resistance of both microorganisms. LP and MPI were selected to investigate the sole effect of lactose and milk protein, respectively, on thermal resistance of both microorgani sms. Because the physicochemical properties of dairy powders are dictated by the major macronutrient component (Schuck et al., 2005) , LP was compared wit h SMP , which has a high p roportion of lactose ( 48.9 %) as opposed to protein and other minerals. MPI (9 1.6 % protein, 1 .2 % lactose) and SMP (37.7% protein, 48.9% lactose) were chosen to represent the highest and lowest milk protein content , respectively. Based on their compositional differences, the dairy powders selected provide new insight into the thermal resistance of Salmonella and E. faecium as a surrogate. 97 The water activi ty of lactose powder was highly variable and required a longer time (about 5 to 10 days) to reach 0.25 a w during equilibration , as compared to other dairy powders. Commercially available lactose is usually obtained as a crystalline alpha - monohydrate, appearing as a fine white powder with the consistency powdered sugar, although a small amount of amorphous lactose can form on the surface of crystalline lactose during spray - drying (Clark et al., 2016) . Because amorphous lactose absorb s nearly 100 times more moisture than crystalline lactose, the small amount of liquid introduced into lactose powder during inoculation might alter the shape of th e product isotherm during storage (Bronlund & Paterson, 2004) . L actose powder is also sensitive to changes in relative humidity. These observations may partly explain why the a w of lactose was highly variable during equilibration , Overall, t he findings suggest that the presence of lactose in SMP mi ght be partly responsible for the differences in the kill ratio of E. faecium to Salmonella , between SMP and LSMP . This observation is supported by the fact that different kill ratio s were seen in SMP and LP, due to t he presence of lactose . Particularly for MPI, where the protein content is the highest, the observed kill ratio of ~1.6 could be due to the small percentage of lactose present (1.2%) in MPI. In addition, when lactose was hydrolyzed into glucose and galactose (represented by L S MP) , E. faecium was no more thermally resistan t tha n Salmonella . Furthermore , E. faecium did not exhibit increased thermal resistance in LSMPX (LSMP with added lactose). Because thermal resistance of both microorganisms in SMP and LSMP did not differ upon addition of complementary sugars during resuscitation of survivors, thermal resistance of both microorganisms was primarily impacted by the product microenvironment of da iry powders , rather than the means of recovery. 98 T he magnitude of the D 9 0 °C value was substantially greater in SMP than in LSMP for the same target organism (Table 5.2). Because SMP and LSMP differ in sugar composition, the se findings imply that lactose ( a disaccharides), but not glucose or galactose (monosaccharides), is thermally protective for both microorganisms . A possible explanation can be linked to lactose crystall ization of SMP during heat treatment (Morgan et al., 2005) . L actose crystallization releases water molecules, and this transition state of lactose occurs when the SMP temperature exceeds its Tg . The rate of crystallization is controlled by relative humidity , the difference in product temperature (T) and the glass transition temperature (Tg) (Jouppila & Roos, 1994) . In dairy powders, increased T - Tg accelerate s lactose crystallization, which also promote s other physical changes including M aillard browning , stickiness, and caking (Jouppila et al., 1997) . The Tg of SMP was 58.3 °C, indicat ing lactose crystallization in SMP during heat ing . LSMP c an also undergone crystallizat ion due to the presence of monosaccharides, with the Tg for glucose and galactose previously reported to be similar at 3 0 °C and 31 °C (Roos, 1993) . The Tg of LSMP was not determined in this study, however Shuck et al. (2005) reported that the Tg of LSMP was lower than that for SMP, in the range of - 15 to - 5 °C at a relative humidity of 22% to 33%. Since lactose crystallization is a rapid phenomenon (Clark et al., 2016) , Tg measurements using DSC will be affected by the heating rate during heating cycle (Thomas et al., 2004) . Tg of LSMP might also not have been captured since heating cycle began at - 20°C. Therefore, the T - Tg may have been lower for SMP than for LSMP, with the observed physical changes in LSMP , particularly caking and browning, occurr ing faster at the treatment temperature. Lactose is a reducing disaccharide , whereas glucose and gala ctose are reducing monosaccharides (Davis, 1995) . Maillard reaction results in nonenzymatic browning, which involves the reaction of lysine (amino acid) with a reducing sugar at high temperature (Morgan et 99 al., 2005) . In dairy products, t he Maillard reaction occurs more rapidly in the presence of monosaccharides as compared to lactose (Naranjo et al., 2013) . Water is one product derived from the early stage of the Maillard reaction B rien et al., 19 89) . Lactose has low solubility in water since the hydroxyl groups of lactose form hydrogen bonds between glucose and galactose (the building monosaccharides of lactose) . In contrast monosaccharides which are more soluble possess a higher number of hydr oxyl groups that can serve as binding sites for water molecules, forming intermolecular hydrogen bond s (BeMiller, 2019) . Therefore, the higher number of water molecules generated from the Maillard reaction may have bound to the monosaccharides in LSMP, whereas water molecules in SMP likely bound to milk protein rather than lactose. Since the vibration of water molecules provide one means to inactivate microorganism s at high temperature (Pena - Melendez et al., 2014) , the vibrational differences that oc cur when water molecules are abound to milk protein in SMP as compared to monosaccharides in LSMP , could also partly explain the observed differences in microbial inactivation, B oth SMP and LSMP were visually brown after heat ing , however, the browning rate s for SMP and LSMP were not measured. B rowning of SMP and LSMP correlated with caking, with these two phenomena not seen in LP and MPI. Caking results from thermal plasticization, where water acts as a plasticizer, forming liquid bridg es to the hydroxyl gr oups of sugars at high temperature (Roos, 2009) . Lactose crystallization releases water molecules, which stimulate the formation of liquid bridges, leading to particle adhesion of SMP and LSMP. Despite t he molecular interaction of water molecules for SMP and LSMP, these molecular events did not significantly affect the a w changes seen for SMP and LSMP at 80 °C , although a slight increase in a w was observed for SMP and LSMP (+0.02). 100 5.5 Conclusion This study assessed the influence of lactose and protein composition on thermal resistance of E. faecium and Salmonella , with the aim to determine the acceptability of E. faecium as Salmonella surrogate in dairy powders . In summary, E . faecium was more t hermally resistant than Salmonella in SMP, LP, and MPI, but similar ly resistan t to Salmonella in LSMP. Although dairy powder composition greatly impact ed the thermal resistance of both microorganisms, the kill ratio relative for E. faecium and Salmonella was consistently higher in the presence of lactose. Th ese findings , however, must not be extrapolated to other dairy powders such as whole milk powder tha t have distinct compositional differences. In practical terms, Salmonella contamination levels in commercially produced dairy powders are extremely low since pasteurization of fluid milk effective ly eliminate all foodborne pathogens including Salmonella . S pray - drying is typically used for commercial production of dairy powders. Although this study did not address the efficacy of spray - drying on Salmonella inactivation, spray dryers are a potential source for Salmonella contamination based on previous outbreaks. Thus, understanding the influence of lactose, milk protein, or both components on the thermal resistance of Salmonella and E. faecium will help the dairy industry evaluate and validate post - processing lethality strategies for dairy powders. 101 6. OVERALL CONCLUSION AND RECOMMENDATIONS 6.1 Overall conclusions In summary, Enterococcus faecium NRRL B - 2354 consistently exhibited greater thermal resistance than Salmonella in 5 out of 6 products (chapter 4), and 3 out of 4 dairy powders (chapter 5). Product composition influence d the thermal resistance of both microorganisms across all LMF tested. Comparison studies of the thermal resistance of Salmonella and E. faecium across laboratories play a major role in robustly supporting the effect of product composition as previously stated. Since the product inoculation method is an important component of standardized methodology, inoculation method s that influence the thermal resistance of microorganism in the product, as demonstrated by dry inoculation using talc powder (chapter 3) , need to be avoid ed . Th ese finding offer greater credence for utilizing E. faecium NRRL B - 2354 as a Salmonella surrogate for validating thermal processing of LMF . Given the pressing need for the food industr y to validate thermal processes for enhanced safety, this study wa s able to provide scientific justification for Enterococcus faecium NRRL B - 2354 as an acceptable Salmonella surrogate for validating thermal processes for LMF , however product composition need s to be considered when validating thermal pasteurization of LMF . Validating thermal treatment s for LMF is part of risk management (Food and Agriculture Organization of the United Nations and World Health Organization, 2014) . Currently, there is no specified minimum log reduction required for such validation s . A minimum 4 - log reduction in validating thermal process es for almonds was adopted by Almond Board of California, when using E. faecium as a surrogate for Salmonella Enteritidis PT 30 (Almond Board of California, 2014) since E. faecium exhibited slightly higher thermal resistance than Salmonella Enteritidis PT 30. 102 Because E. faecium was consistently more thermally resistant than Salmonella in LMF, as demonstrated in thi s work , a higher kill ratio may lead to undesirable outcomes from over - processing , including deterioration of the final product and higher processing cost s , if a minimum 4 - log reduction is required for validating thermal processes for LMF using E. faecium . In order to establish and validat e minimum log reduction s for Salmonella in different LMF , the proper kill ratio of E. faecium to Salmonella must be chosen based on produ ct composition to obtain a safe and consumer acceptable product . T his dissertation demonstrated that p roduct composition can affect thermal resistance of Salmonella and E. faecium ; therefore , food manufacturers need to consider product composition when using E. faecium to validate thermal processing of LMF. R egulatory bodies may consider establishing a lower minimum log reduction when using E. faecium in validating thermal processes for produ ct s such as nonfat dry milk powder that yield exceptional ly high kill ratio s of E. faecium to Salmonella ( e.g., kill ratio of 3) . Typically, thermal inactivation process es target a minimum 5 - log reduction (Sm elt & Brul, 2014) , so t o illustrate, a 3 - log reduction for E. faecium in nonfat dry milk powder would be equivalent to a 9 - log reduction for Salmonella , which lead s to severe over - processing . Since contamination levels for Salmonella in nonfat dry milk powder are typically extremely low , a thermal treatment yielding a 9 - log reduction for Salmonella in nonfat dry milk powder would be inappropriate . 6.2 Recommendations for future work The main research contribution of this dissertation is that E. faecium NRRL B - 2354 has again been shown to be a suitable surrogate for Salmonella , given that its thermal resistance was generally similar or higher than Salmonella in the low - moisture food products tested. This key 103 finding suggest s that food manufacturers can consider E. faecium as a biological validation tool in the thermal processin g of low - moisture foods to provide scientifically based justification as require d by FSMA. With such a research milestone, there are several recommendations for improving the certainty of E. faecium utilization in ensuring the safety of LMF . One of the cop ing mechanisms for Salmonella to survive desiccat ed condition is through biosynthesis of compatible solutes, such as trehalose (disaccharides) and osmoprotectant molecules (glycine, betaine, proline), that increase osmotic solutes in the cytoplasm (Finn et al., 2013) . Because relative thermal resistance of E. faecium to Salmonella is partly sugar - dependent, at least in skim milk powder , it is crucial to understand how sugar types influence production of trehalose and osmoprotectants in both microorganisms, and subsequently their resistances to heat. Therefore, it is proposed to determine the impact of sugar type s ; glucose, galactose, gl ucose - galactose (monosaccharides), and lactose (disaccharides) , via quantification of osmotic solutes, on cell viability when exposed to subsequent heat stress. If cell viability increases with increasing accumulation of osmotic solutes, it can be inferred that the degree of osmotic shock induced by the different type of sugars, influence the degree of bacterial cross - tolerance towards subsequent heat treatment. In this study, comparison of thermal resistance with E. faecium , was made with a Salmonella cock tail, as opposed to a single strain of Salmonella . Because we observed a higher kill ratio of E. faecium to Salmonella in SMP, but not LSMP, it is important to confirm that the differences in kill ratio were not due to differences in thermal resistance of the individual Salmonella st rain s . Therefore, a future study can investigate the impact of sugar composition on strain selectivity of Salmonella survivors in dairy powders. If different Salmonella strain s are identified in SMP and LSMP, it can be inferred that the differences in kill ratio of E. faecium to 104 Salmonella in SMP and LSMP can be derived from a comparison of thermal resistance between E. faecium and the different strain of Salmonella . U se of a stand ardized protocol is a strategic approach to determine the reproducibility of results between two or more validation studies. To extend the impact of the current findings, the LMF validation studies can be scaled up using a pilot - study, so that the critical parameters selected for the thermal process can better represent the critical parameters in actual LMF processing. Furthermore, foodborne pathogens that have been associated with low - moisture foods are not limited to only Salmonella , but also E. coli O157 :H7, Listeria monocytogenes , Cronobacter sakazakii , Staphylococcus aureus , and Bacillus cereus . It is recommended to test whether E. faecium is also a suitable surrogate for these target pathogens. 105 \ APPENDI CES 106 APPENDIX A. Previous validation studies using Enterococcus faecium NRRL B - 2354 Table A. 1 List of previous v alidation studies using Enterococcus faecium NRRL B - 2354 in low - moisture foods. Product Target pathogen(s) Surrogate Inactivation step Pertinent findings Reference Macadamia , cashew nuts S . Senftenberg 775 W S . Montevideo 1449 S . Tennessee K4643 S . Tennessee Ball ARL - SE - 085 S . Tennessee ARL - SE - 013 Enterococcus faecium NRRL B - 2354 Propylene oxide (PPO) fumigation Log (CFU/g) reduction : Salmonella - Macadamia: 7.3 log Cashew: 5.4 log E. faecium Macadamia: 6.4 log Cashew: 5.1 log Saunders et al. ( 2018) Black peppercorn S . Agona 447967 S . Reading S . Tennessee K4643 S . Mbandaka 488275 S . Montevideo 698538 Enterococcus faecium NRRL B - 2354 Radiofrequency Log (CFU/g) reduction after 2.5 min: Salmonella - 5.3 log E. faecium - 5.3 log Wei et al. ( 2018) Wheat flou r S. Enteritidis PT 30 Enterococcus faecium NRRL B - 2354 Oil bath D 75 - 85°C values: Salmonella - 0.30 a w : 5.9 - 24.5 min 0.45 a w : 2.9 - 17.7 min 0.60 a w : 1.1 - 12.0 min E. faecium - 0.30 a w : 15.9 - 65.8 min 0.45 a w : 4.1 - 29.4 min 0.60 a w : 2.7 - 25.5 min Liu et al. ( 2018) Whole flaxseed milled flaxseed, sunflower kernel, quinoa whole black peppercorn S. Enteritidis PT 30 E. coli O157:H7 Enterococcus faecium NRRL B - 2354 Vacuum pasteurization Log (CFU/g) reduction after 1 min at 75°C: Salmonella - Whole flaxseed: 5.5 log Sunflower kernel: 4.0 log Quinoa: 4.3 log Milled flaxseed: 3.0 log Black peppercorn: 6.0 log E. faecium - Whole flaxseed: 5.2 log Sunflower kernel: 3.0 log Quinoa: 2.4 log Milled flaxseed and black peppercorn: n/ a Shah et al. ( 2017) 107 Table A .1 ( c ) Product Target pathogen(s) Surrogate Inactivation step Pertinent findings Reference Wheat flour S. Enteritidis PT 30 Enterococcus faecium NRRL B - 2354 Radiofrequency Log (CFU/g) reduction Salmonella - 0.45, 0.65 a w : ~7 log 0.25 a w : 5 log E. faecium - 0.45, 0.65 a w : n/a 0.25 a w : 3 log Villa - Rojas et al. ( 2017) Confectionery formulation, savory seasoning powder, chicken meat powder and pet food S. Enteritidis PT 30 S. Typhimurium S. Anatum S. Montevideo S. Senftenburg 775W S. Tennessee Enterococcus faecium NRRL B - 2354 Oil bath D 80°C for 5 - log reduction (Weibull model): Salmonella - Confectionery : 40.6 min Seasoning :9.5 min Chicken meat powder 67.2 min Pet food : 4.2 min E. faecium - Confectionery : 25 min Seasoning : 35.3 min Chicken meat powder 126.2 min Pet food : 15.7 min Rachon et al. ( 2016) Hamburger bun dough c S. Typhimurium S. Newport S. Senftenburg 775W Enterococcus faecium NRRL B - 2354 Oven and water bath Time needed to reach > 5 - log reduction after baking (oven) Salmonella 6 min (below LOD after 9 min by enrichment) E. faecium 7.8 min (below LOD after 11.5 min by enrichment ) D 55 - 61 °C of dough (water bath) Salmonella - 3.1 - 28.6 min E. faecium - 14.7 - 133.3 min Channa iah et al. ( 2016) Commercial pet food b S. Anatum S. Montevideo S. Senftenburg 775W S. Tennessee S. Scwarzengrund S. Infantis S. Mbandaka Enterococcus faecium NRRL B - 2354 Water bath D - values at different moisture (%) levels: Salmonella 9.1% : 1.1 - 6.5 min 16.9% : 0.4 - 5.5 min 27% : 0.5 - 17.3 min E.faecium : 9.1 % : 1.7 - 11.7 min 16.9% : 1.1 - 13.8 min 27% : 1.8 - 24 min Ceylan & Bautista ( 2015) 108 Table A. 1 ( c ) a Carbohydrate - protein meal made of almost equal percentage of chicken meal and r ice flour b Major ingredients: corn and soybean meal c In gredients: bread flour, granulated sugar, shortening, yeast, salt, and water . Product Target pathogen(s) Surrogate Inactivation step Pertinent findings Reference Formulated peanut paste Five strains of S. Tennessee: 5010 H S 13952 (782) S 13972 (783) S 13999 (784) FSL R8 - 5221 Enterococcus faecium NRRL B - 2354 Water bath D 85 °C values Salmonella 0.9 - 1.1 log E. faecium 2.5 - 3.4 log Enache et al. ( 2015) Formulated pet food a S. Oranienburg S. Typhimurium S. Branderup S. Heidelberg/Sheldon S. Enteritidis Enterococcus faecium NRRL B - 2354 Extrusion Log reduction (log CFU/g): Salmonella - 4.2 - 6.5 log; ( 55 - 62 °C) E. faecium - 1.5 - 5.8 log (55 - 75 °C ) Bianchini et al. ( 2014) Whole almond S. Enteritidis PT30 Enterococcus faecium NRRL B - 2354 (formerly known as Pediococcus acidilactici ) Moist - air heating Log (CFU/g) reduction ranges for 5 - 90% humidity level at each temperature Salmonella 121°C: 2.8 5.1 log 149°C: 2.8 6.2 log 177°C : 3.5 - 6.3 log 204°C : 1.6 6.6 log P.acidilactici : 121°C: 2.2 5.0 log 149°C: 3.0 4.4 log 177°C : 2.7 5.3 log 204°C : 1.0 4.5 log Jeong et al. ( 2011) 109 APPENDIX B . Isothermal inactivation data for almond meal and talc (Chapter 3) Table B. 1 Enterococcus faecium NRRL B - 2354 inactivation in wet - inoculated almond meal (0.45 a w ) at 80 °C . Time (min) Log CFU/g Rep 1 R ep 2 R ep 3 0 0 0 30 30 30 60 60 60 90 90 90 120 120 120 7.50 7.53 7.52 7.04 6.92 6.69 5.89 5.89 5.72 5.35 5.33 5.31 4.97 4.97 4.98 7.37 7.66 7.4 0 6.51 6.48 6.29 5.77 5.72 5.68 5.56 5.58 5.54 5.18 4.61 4.55 7.59 7.58 7.63 6.64 6.65 6.2 0 5.88 5.7 0 5.48 5.08 5 .00 4.97 4.47 4.69 4.54 110 Table B. 2 Enterococcus faecium NRRL B - 2354 inactivation in dry - inoculated almond meal (0.45 a w ) at 80 °C . Table B. 3 Raw survivor data for Enterococcus faecium NRRL B - 2354 in wet - talc - inoculated almond meal (0.45 a w ) at 80 °C . Time (min) Log CFU/g Rep 1 Rep 2 Rep 3 0 8.02 7.80 7.69 0 7.86 7.75 7.45 0 7.96 7.80 7.48 30 7.03 7.00 6.61 30 7.00 6.90 6.72 30 6.75 6.89 6.69 60 6.73 6.30 6.15 60 6.53 6.17 6.26 60 6.64 6.23 6.34 90 5.97 5.89 5.72 90 6.23 5.85 5.79 90 6.21 5.85 5.66 120 5.83 5.08 4.99 120 5.50 5.04 5.13 120 5.51 4.99 5.33 Time (min) Log CFU/g Rep 1 Rep 2 Rep 3 0 5.27 n/a 5.22 0 4.89 4.95 5.25 0 4.88 5.03 5.13 30 - 4.40 4.43 30 - 4.48 4.55 30 - 4.48 4.59 60 3.79 3.83 3.95 60 3.73 3.73 3.93 60 3.71 3.71 4.06 90 3.46 3.46 3.80 90 3.39 3.39 3.61 90 3.45 3.49 3.65 120 3.26 3.59 3.43 120 3.14 3.34 3.01 120 3.11 3.31 3.37 111 Table B. 4 Raw survivor data for Enterococcus faecium NRRL B - 2354 in talc powder (0. 45 a w ). at 80 °C . Time (min) Log CFU/g Rep 1 Rep 2 Rep 3 0 8.22 7.99 - 0 8.06 - 8.09 0 8.03 - - 9 7.73 7.24 6.57 9 6.81 - - 9 7.81 - 7.6 0 18 6.96 - 7.21 18 6.99 6.41 - 18 7.22 7.22 7.54 27 6.59 - 5.06 27 6.12 - - 27 6.03 6.84 7.3 0 36 5.64 6.37 6.24 36 6.45 6.82 - 36 6.82 n/a 5.62 45 6.11 n/a 4.63 45 5.99 n/a 4.25 45 5.23 5.17 - 54 6.06 6.24 - 54 5.36 5.78 4.73 54 5.92 - - 63 - 5.99 5 .00 63 - 6.23 6.11 63 - - - 112 APPENDIX C . Isothermal i nactivation data for nonfat dry milk powder at 0.25 a w (chapter 4) Table C. 1 Raw survival data for Salmonella in nonfat dry milk powder (0.25 a w ) at 85 °C generated by MSU and WSU. Salmonella - 85 °C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.87 0 8.02 0 7.37 0 7.51 0 7.7 0 0 7.07 0 7.83 0 8.07 0 7.38 0 7.65 0 7.56 0 7.18 0 7.78 0 - 0 - 0 7.7 0 0 7.79 0 6.93 10 7.1 0 10 7.97 10 6.91 10 6.75 10 7.08 10 5.95 10 6.93 10 7.48 10 6.71 10 6.69 10 7.18 10 6.14 10 7.02 10 7.65 10 6.76 10 - 10 7.08 10 6.00 20 6.72 20 7.17 20 - 20 6.05 20 6.62 20 - 20 6.6 0 20 7.09 20 6.13 20 6.13 20 6.57 20 - 20 6.64 20 7.31 20 6.14 20 n/a 20 6.48 20 5.34 30 6.07 30 6.8 0 30 5.58 30 5.55 30 5.61 30 n/a 30 5.93 30 6.91 30 - 30 5.5 0 30 5.84 30 4.75 30 5.7 0 30 6.58 30 - 30 5.24 30 5.45 30 4.67 40 5.25 40 6.17 40 5.14 40 4.79 40 4.89 40 - 40 4.96 40 5.85 40 5.84 40 4.95 40 4.53 40 4.52 40 5.31 40 6.07 40 - 40 4.84 40 4.93 40 - 50 - 50 5.43 50 3.56 50 3.42 50 4.04 50 3.39 50 4.71 50 4.85 50 3.6 0 50 3.38 50 3.97 50 3.31 50 4.38 50 5.68 50 3.67 50 3.42 50 3.92 50 3.63 60 3.69 60 3.98 60 2.91 60 4.42 60 4.08 60 3.61 60 3.66 60 8.02 60 - 113 Table C. 2 Raw survival data for Salmonella in nonfat dry milk powder (0.25 a w ) at 90 °C generated by MSU and WSU. Salmonella - 90 °C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.01 0 7.12 0 7.29 0 7.47 0 7.83 0 7.26 0 6.94 0 7.04 0 7.27 0 7.47 0 7.83 0 7.21 0 7.04 0 6.88 0 7.23 0 7.3 0 0 7.74 0 7.13 5 7.09 5 6.55 5 6.7 0 5 6.57 5 7.1 5 6.52 5 7.24 5 6.71 5 6.96 5 6.34 5 7.07 5 6.52 5 7.55 5 6.40 5 - 5 6.26 5 7.08 5 6.82 10 6.39 10 6.47 10 6.18 10 5.85 10 6.53 10 6.05 10 6.58 10 6.18 10 7.72 10 6.13 10 6.52 10 5.74 10 6.54 10 6.20 10 6.77 10 5.94 10 6.47 10 5.9 0 15 5.74 15 5.54 15 5.54 15 5.18 15 5.44 15 - 15 5.93 15 5.86 15 5.43 15 5.08 15 5.59 15 - 15 5.78 15 5.93 15 5.52 15 - 15 5.84 15 5.84 20 4.87 20 5.02 20 - 20 - 20 4.53 20 4.48 20 - 20 5.06 20 - 20 3.6 20 4.75 20 4.6 0 20 - 20 5.12 20 4.56 20 3.64 20 4.7 0 20 - 25 - 25 3.73 25 3.91 25 3.21 25 3.77 25 3.83 25 3.73 25 4.22 25 3.74 25 - 25 3.6 0 25 3.83 25 3.67 25 - 25 - 25 - 25 4.01 25 3.88 30 - 30 3.18 30 - 30 - 30 3.59 30 2.68 30 2.55 30 - 30 - 114 Table C. 3 Raw survival data for Salmonella in nonfat dry milk powder (0.25 a w ) at 95 °C generated by MSU and WSU. Salmonella - 95 °C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.13 0 6.67 0 6.41 0 7.14 0 7.45 0 6.85 0 7.21 0 - 0 6.75 0 6.94 0 7.52 0 6.77 0 7.32 0 6.64 0 6.41 0 7.03 0 7.61 0 6.76 2.5 8.01 2.5 6.39 2.5 6.12 2.5 6.45 2.5 - 2.5 6.39 2.5 7.56 2.5 5.99 2.5 5.91 2.5 6.74 2.5 6.86 2.5 6.16 2.5 7.37 2.5 5.73 2.5 5.94 2.5 6.59 2.5 7.14 2.5 6.11 5 6.08 5 5.98 5 5.05 5 5.36 5 6.52 5 5.77 5 6.16 5 6.93 5 5.24 5 5.01 5 6.72 5 5.57 5 6.29 5 5.51 5 4.93 5 5.53 5 6.47 5 - 7.5 4.84 7.5 5.46 7.5 4.46 7.5 5.11 7.5 - 7.5 5.37 7.5 5.38 7.5 6.47 7.5 - 7.5 - 7.5 5.6 7.5 5.66 7.5 4.92 7.5 5.59 7.5 4.59 7.5 - 7.5 5.35 7.5 5.35 10 4.64 10 4.82 10 - 10 3.81 10 4.22 10 4.19 10 - 10 - 10 3.8 0 10 4.25 10 4.24 10 3.79 10 - 10 6.49 10 3.51 10 4.23 10 4.22 10 - 12.5 - 12.5 - 12.5 2.89 12.5 2.99 12.5 3.94 12.5 3.49 12.5 - 12.5 - 12.5 4.21 12.5 3.14 12.5 4.15 12.5 3.12 12.5 4.08 12.5 5.68 12.5 3.18 12.5 3.42 12.5 3.79 12.5 - 15 2.99 15 - 15 - 15 - 15 - 15 2.67 15 4.53 15 4.01 15 3.05 15 - 15 2.69 15 - 115 Table C. 4 Raw survival data for Enterococcus faecium NRRL B - 2354 in nonfat dry milk powder (0.25 a w ) at 85 °C generated by MSU and WSU. E. faecium - 85 °C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.96 0 7.95 0 7.38 0 7.83 0 7.69 0 7.44 0 8.13 0 7.78 0 7.4 0 0 7.81 0 7.82 0 7.38 0 - 0 7.8 0 0 7.4 0 0 - 0 7.86 0 7.43 30 7.83 30 7.58 30 7.4 0 40 7.56 40 7.61 40 7.34 30 7.72 30 7.7 0 30 7.26 40 7.77 40 7.59 40 - 30 7.65 30 7.53 30 7.27 40 7.52 40 7.65 40 7.34 60 7.66 60 7.35 60 7.15 80 7.25 80 7.12 80 6.97 60 7.39 60 n/a 60 7.1 0 80 7.26 80 7.38 80 6.98 60 7.64 60 n/a 60 7.21 80 7.17 80 n/a 80 6.99 90 7.26 90 6.84 90 6.82 120 5.41 120 6.47 120 5.97 90 7.17 90 6.94 90 6.68 120 - 120 n/a 120 - 90 7.22 90 6.89 90 6.51 120 5.45 120 n/a 120 5.81 120 5.84 120 6.08 120 6.04 160 3.65 160 4.45 160 3.16 120 6.06 120 5.55 120 5.76 160 3.59 160 4.38 160 3.48 120 5.87 120 5.68 120 5.78 160 3.75 160 4.57 160 3.32 150 5.56 150 4.46 150 n/a 200 3.1 0 200 - 200 - 150 4.51 150 - 150 3.74 200 3.15 200 - 200 - 150 3.76 150 4.59 150 - 200 3.12 200 - 200 - 180 4.86 180 - 180 3.35 180 - 180 - 180 - 180 - 180 - 180 - 116 Table C. 5 Raw survival data for Enterococcus faecium NRRL B - 2354 in nonfat dry milk powder (0.25 a w ) at 90 °C generated by MSU and WSU. E. faecium - 90 °C MSU WSU Time (min) Rep 1 ( L og CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.9 0 0 7.79 0 7.32 0 7.77 0 7.95 0 7.24 0 7.87 0 7.71 0 7.28 0 7.85 0 7.81 0 7.22 0 - 0 7.67 0 - 0 7.74 0 - 0 7.3 0 15 7.92 15 7.48 15 7.26 20 7.68 16 7.65 16 6.97 15 7.86 15 7.59 15 7.12 20 7.59 16 7.57 16 7.01 15 7.99 15 - 15 - 20 7.45 16 7.56 16 6.89 30 7.58 30 7.29 30 7.16 40 6.85 32 7.35 32 6.66 30 7.62 30 7.18 30 7.03 40 7.05 32 7.20 32 6.57 30 7.69 30 7.21 30 6.85 40 6.98 32 - 32 6.52 45 7.09 45 6.7 0 45 6.66 60 4.78 48 6.84 48 4.61 45 7.02 45 6.82 45 6.49 60 4.51 48 6.87 48 - 45 7.00 45 6.81 45 6.19 60 n/a 48 - 48 5.12 60 6.19 60 5.84 60 5.83 80 2.56 64 5.54 64 3.57 60 6.16 60 6.01 60 5.6 0 80 2.44 64 5.51 64 3.36 60 5.77 60 5.7 60 5.63 80 2.59 64 - 64 3.59 75 2.95 75 4.86 75 3.58 100 - 80 4.11 80 3.47 75 3.33 75 3.66 75 4.09 100 - 80 4.31 80 3.33 75 4.29 75 4.47 75 3.66 100 - 80 3.69 80 7.24 90 - 90 2.56 90 3.67 90 - 90 - 90 3.52 90 - 90 - 90 - 117 Table C. 6 Raw survival data for Enterococcus faecium NRRL B - 2354 in nonfat dry milk powder (0.25 a w ) at 95 °C generated by MSU and WSU. E. faecium - 95 °C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.93 0 7.67 0 7.09 0 7.77 0 7.8 0 0 7.3 0 0 7.92 0 7.68 0 7.03 0 7.63 0 7.7 0 0 7.45 0 7.8 0 0 7.62 0 7.06 0 7.86 0 7.8 0 0 7.38 6 7.58 6 7.43 6 7.04 10 7.43 10 7.35 10 - 6 7.7 0 6 7.45 6 7.05 10 7.41 10 7.43 10 7.28 6 7.87 6 - 6 6.88 10 - 10 7.63 10 7.28 12 7.56 12 7.41 12 7.01 20 7.10 20 7.26 20 - 12 7.2 0 12 7.3 0 12 6.82 20 6.98 20 7.16 20 5.97 12 7.38 12 7.12 12 6.82 20 7.06 20 7.08 20 6.08 18 - 18 7.11 18 6.56 30 5.14 30 5.87 30 5.21 18 6.99 18 - 18 6.71 30 5.49 30 5.96 30 5.08 18 7.1 0 18 - 18 6.79 30 5.54 30 5.94 30 5.16 24 6.14 24 6.4 0 24 6.13 40 2.47 40 3.86 40 4.07 24 6.54 24 6.41 24 5.94 40 2.59 40 3.12 40 4.33 24 6.44 24 6.66 24 6.00 40 2.43 40 3.72 40 4.32 30 5.74 30 5.4 0 30 - 50 n/a 50 - 50 - 30 5.95 30 5.57 30 5.32 50 n/a 50 - 50 - 30 5.33 30 5.75 30 5.08 50 n/a 50 - 50 - 36 3.92 36 4.54 36 3.49 36 - 36 3.83 36 4.35 36 3.47 36 - 36 3.70 118 APPENDIX D . Isothermal inactivation data for peanut butter at 0.25 a w (chapter 4) Table D. 1 Raw survival data for Salmonella in peanut butter (0.25 a w ) at 85 °C generated by UNL and IFSH. Salmonella - 85 °C UNL IFSH Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU/ g) Time (min) Rep 4 (Log CFU /g) Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU /g) 0 6.7 0 7.25 0 6.89 0 7.15 0 7.66 0 7.87 0 8.17 0 7.70 0 7.03 0 7.15 0 7.1 0 0 6.96 0 7.76 0 7.55 0 7.90 0 7.62 0 7.04 0 7.36 0 7.25 0 7.02 0 7.82 0 7.74 0 7.76 0 7.69 25 5.09 25 5.39 25 4.9 25 5.24 25 6.01 25 5.92 25 6.06 25 5.84 25 5.17 25 5.07 25 5.07 25 5.19 25 5.81 25 5.86 25 6.18 25 6.07 25 5.2 0 25 5.53 25 5.29 25 5.61 25 5.95 25 6.00 25 6.02 25 6.44 50 4.23 50 4.32 50 4.21 50 4.6 50 4.94 50 4.91 50 5.08 50 5.49 50 4.42 50 4.87 50 4.18 50 4.7 50 5.02 50 5.12 50 5.11 50 5.72 50 4.49 50 4.76 50 4.55 50 4.69 50 5.17 50 4.90 50 5.40 50 5.71 75 4.22 75 3.9 75 4.01 75 3.91 75 4.47 75 4.68 75 4.87 75 4.79 75 3.98 75 4.28 75 3.77 75 3.97 75 4.66 75 4.66 75 4.76 75 4.79 75 3.87 75 4.15 75 3.98 75 4.23 75 4.62 75 4.37 75 4.89 75 4.45 100 3.65 100 3.77 100 3.49 100 3.59 100 4.04 100 4.26 100 4.38 100 4.45 100 3.66 100 3.59 100 3.93 100 3.45 100 3.94 100 3.86 100 4.13 100 4.25 100 3.63 100 3.80 100 n/a 100 3.47 100 4.14 100 4.22 100 4.40 100 4.32 125 2.54 125 3.23 125 2.9 125 2.98 125 3.53 125 3.39 125 3.71 125 4.09 125 2.97 125 3.48 125 3.51 125 3.27 125 3.04 125 3.96 125 3.77 125 4.24 125 2.98 125 3.37 125 3.36 125 3.21 125 3.74 125 3.72 125 4.01 125 4.29 119 Table D. 2 Raw survival data for Salmonella in peanut butter (0.25 a w ) at 90 °C generated by UNL and IFSH. Salmonella - 90 °C UNL IFSH Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU/ g) Time (min) Rep 4 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU /g) 0 6.64 0 6.73 0 6.25 0 6.75 0 7.41 0 7.64 0 7.60 0 6.29 0 6.53 0 6.81 0 6.27 0 7.47 0 7.46 0 7.22 0 6.76 0 6.8 0 0 6.73 0 6.72 0 7.22 0 7.48 0 7.44 8 - 8 5.39 8 5.12 8 5.7 0 8 6.30 8 6.23 8 6.70 8 5.56 8 5.34 8 5.21 8 5.64 8 6.03 8 6.20 8 6.46 8 - 8 5.6 0 8 5.15 8 5.69 8 6.11 8 6.47 8 6.78 16 4.25 16 4.7 0 16 4.22 16 4.46 16 5.46 16 5.82 16 5.96 16 - 16 4.65 16 4.51 16 5.38 16 5.50 16 5.55 16 6.02 16 4.68 16 4.82 16 4.58 16 4.8 0 16 5.08 16 5.81 16 5.96 24 4.1 0 24 4.21 24 3.48 24 4.3 0 24 4.72 24 5.13 24 5.01 24 4.16 24 4.2 0 24 3.98 24 4.46 24 4.69 24 5.01 24 4.82 24 4.04 24 4.37 24 3.72 24 4.86 24 4.56 24 5.20 24 5.20 32 4.08 32 3.32 32 3.05 32 4.36 32 4.22 32 4.92 32 4.99 32 3.92 32 3.53 32 3.39 32 4.31 32 4.37 32 4.93 32 4.68 32 4.04 32 3.61 32 4.44 32 4.32 32 4.58 32 4.95 32 5.32 40 3.5 0 40 3.78 40 2.81 40 3.96 40 4.37 40 4.78 40 4.41 40 3.74 40 3.94 40 2.89 40 3.98 40 3.87 40 4.63 40 5.00 40 3.86 40 3.98 40 - 40 3.83 40 4.17 40 4.61 40 4.45 120 Table D. 3 Raw survival data for Salmonella in peanut butter (0.25 a w ) at 95 °C generated by UNL and IFSH. Salmonella - 95 °C UNL IFSH Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU/g) Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU /g) 0 6.12 0 6.2 0 5.89 0 6.97 0 6.58 0 7.23 0 6.96 0 5.99 0 6.1 0 5.74 0 7.30 0 6.99 0 7.18 0 6.93 0 5.77 0 6.05 0 5.76 0 6.70 0 7.06 0 7.01 0 6.81 4 - 4 5.38 4 4.93 4 5.69 4 6.01 4 6.32 4 6.15 4 5.41 4 5.23 4 4.91 4 5.94 4 5.92 4 6.16 4 6.42 4 5.48 4 5.16 4 5.04 4 6.01 4 5.84 4 6.01 4 6.15 8 4.54 8 4.59 8 4.38 8 5.01 8 5.24 8 5.44 8 5.25 8 4.36 8 4.15 8 4.63 8 4.98 8 5.33 8 5.33 8 5.65 8 4.70 8 4.77 8 4.24 8 5.05 8 5.25 8 5.66 8 5.62 12 4.00 12 4 .00 12 3.99 12 4.01 12 4.64 12 4.58 12 5.25 12 4.25 12 3.86 12 3.71 12 4.63 12 4.99 12 4.57 12 5.20 12 3.96 12 3.91 12 3.71 12 3.86 12 4.83 12 4.67 12 5.29 16 3.81 16 3.37 16 3.3 16 3.98 16 4.66 16 3.79 16 4.64 16 3.65 16 3.95 16 3.17 16 3.95 16 4.48 16 3.71 16 4.62 16 3.95 16 3.96 16 3.17 16 3.92 16 4.79 16 3.83 16 4.62 20 2.93 20 3.41 20 3.03 20 3.58 20 4.01 20 3.75 20 4.26 20 3.05 20 3.65 20 2.89 20 3.65 20 4.15 20 3.60 20 4.28 20 3.08 20 3.04 20 3.02 20 3.55 20 4.21 20 3.73 20 4.25 121 Table D. 4 Raw survival data for Enterococcus faecium NRRL - B2354 in peanut butter (0.25 a w ) at 90 °C generated by UNL and IFSH. E. faecium - 90 °C UNL IFSH Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU/ g) Time (min) Rep 4 (Log CFU /g) Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU /g) 0 8.07 0 8.23 0 8.13 0 8.19 0 8.14 0 8.35 0 8.18 0 8.30 0 8.12 0 8.36 0 8.29 0 8.14 0 8.16 0 8.27 0 8.23 0 8.17 0 7.97 0 8.27 0 8.17 0 8.06 0 8.03 0 8.47 0 8.21 0 8.07 20 7.41 20 7.7 0 20 7.53 20 7.72 20 7.52 20 7.78 20 7.98 20 7.64 20 7.53 20 7.72 20 7.65 20 7.68 20 7.34 20 7.73 20 7.32 20 7.53 20 7.71 20 7.76 20 7.39 20 7.65 20 7.33 20 8.06 20 7.49 20 7.74 40 5.82 40 6.32 40 5.87 40 6.13 40 6.30 40 6.38 40 5.70 40 5.97 40 5.87 40 6.33 40 5.83 40 6.10 40 6.08 40 6.53 40 5.72 40 6.04 40 6.14 40 6.28 40 5.76 40 6.00 40 6.30 40 6.19 40 5.78 40 6.19 60 4.74 60 5.07 60 5.13 60 4.85 60 4.86 60 5.51 60 5.45 60 5.10 60 4.82 60 5.24 60 5.16 60 5.08 60 4.79 60 5.44 60 5.33 60 5.11 60 4.6 0 60 5.09 60 4.97 60 5.13 60 - 60 5.23 60 5.44 60 4.92 80 3.73 80 4.35 80 4.73 80 4.21 80 4.42 80 4.79 80 4.75 80 4.37 80 4.31 80 4.24 80 4.56 80 4.10 80 4.39 80 4.66 80 4.65 80 4.42 80 3.9 0 80 4.57 80 4.55 80 4.27 80 4.32 80 4.77 80 4.65 80 4.59 100 3.05 100 3.11 100 4.20 100 3.34 100 3.41 100 3.94 100 4.25 100 4.57 100 3.23 100 3.44 100 2.54 100 3.50 100 2.98 100 3.98 100 4.22 100 3.84 100 2.90 100 3.47 100 3.56 100 - 100 3.65 100 3.96 100 4.35 100 4.15 122 Table D. 5 Raw survival data for Enterococcus faecium NRRL - B2354 in peanut butter (0.25 a w ) at 95 °C generated by UNL and IFSH. E. faecium - 95 °C MSU WSU Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 4 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 4 (Log CFU/g) 0 8.03 0 8.22 0 8.07 0 7.65 0 8.17 0 7.87 0 8.19 0 7.87 0 8.03 0 7.86 0 8.23 0 8.02 0 8.01 0 8.2 0 7.93 0 7.84 0 8.16 0 8.07 8 7.51 8 7.27 8 7.19 10 7.30 8 7.48 8 7.61 8 7.49 8 7.44 8 7.06 10 7.20 8 7.41 8 7.46 8 7.64 8 7.19 8 7.04 10 7.20 8 7.59 8 7.62 16 6.15 16 5.68 16 6.04 20 5.30 16 5.90 16 6.13 16 6.37 16 5.74 16 5.65 20 5.63 16 6.06 16 5.85 16 6.73 16 5.74 16 5.64 20 5.46 16 5.93 16 5.81 24 5.05 24 5.12 24 4.79 30 4.27 24 5.46 24 4.99 24 5.04 24 5.12 24 4.69 30 4.17 24 4.98 24 4.85 24 4.98 24 5.1 0 24 4.98 30 4.10 24 5.56 24 4.67 32 4.55 32 4.82 32 3.86 40 3.21 32 4.78 32 4.32 32 4.25 32 4.61 32 4.4 0 40 3.09 32 4.67 32 4.21 32 4.20 32 4.62 32 4.26 40 2.95 32 n/a 32 4.29 40 3.64 40 3.9 0 40 3.14 40 4.40 40 3.88 40 3.57 40 3.82 40 3.35 40 4.05 40 3.59 40 3.57 40 3.84 40 3.57 40 4.66 40 3.64 123 Table D. 6 Raw survival data for Enterococcus faecium NRRL - B2354 in peanut butter (0.25 a w ) at 100 °C generated by UNL and IFSH. E. faecium - 100 °C UNL IFSH Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU/ g) Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU /g) 0 7.59 0 7.75 0 7.99 0 7.63 0 8.38 0 7.95 0 7.96 0 7.87 0 7.68 0 7.79 0 7.82 0 8.20 0 8.10 0 8.03 0 7.90 0 7.58 0 7.77 0 7.75 0 8.16 0 7.96 0 7.96 4 6.39 4 5.85 4 6.27 4 7.02 3 7.77 3 6.98 3 7.78 4 6.74 4 6.09 4 6.72 4 6.95 3 7.69 3 7.59 3 7.55 4 6.76 4 - 4 - 4 7.16 3 7.77 3 7.43 3 7.56 8 5.05 8 4.77 8 4.82 8 5.29 6 7.09 6 5.96 6 6.81 8 5.61 8 5.06 8 4.69 8 5.39 6 7.15 6 5.78 6 6.37 8 4.94 8 4.76 8 4.8 8 5.49 6 6.93 6 5.81 6 6.41 12 4.06 12 4.07 12 3.86 12 4.42 9 5.70 9 4.57 9 5.24 12 - 12 3.97 12 4.00 12 4.47 9 5.54 9 4.96 9 5.18 12 4.44 12 3.98 12 3.96 12 4.14 9 5.51 9 5.10 9 5.43 16 - 16 3.31 16 2.84 16 2.81 12 4.88 12 3.62 12 4.07 16 - 16 3.52 16 2.74 16 2.49 12 4.75 12 3.96 12 4.18 16 - 16 3.16 16 2.88 16 3.18 12 4.66 12 4.34 12 4.39 20 2.60 20 - 20 - 15 3.72 15 3.47 15 3.70 20 2.78 20 - 20 - 15 3.41 15 3.50 15 4.19 20 - 20 2.71 20 - 15 3.71 15 4.09 15 3.11 124 APPENDIX E . Isothermal inactivation data for almond meal at 0.45 a w (chapter 4) Table E. 1 Raw survival data for Salmonella in almond meal (0.45 a w ) at 80 °C generated by MSU and WSU. Salmonella - 80°C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 8.19 0 8.32 0 7.85 0 7.72 0 7.85 0 7.85 0 8.25 0 8.42 0 7.75 0 7.49 0 8.15 0 7.87 0 8.40 0 8.32 0 7.91 0 7.69 0 8.11 0 7.83 20 7.59 20 7.44 20 6.99 20 7.01 20 7.14 20 7.05 20 7.59 20 7.59 20 6.9 0 20 7.05 20 - 20 7.07 20 - 20 7.51 20 7.03 20 7.11 20 7.23 20 6.69 40 6.72 40 6.49 40 7.17 40 5.94 40 6.42 40 6.41 40 6.7 0 40 6.72 40 6.10 40 6.39 40 6.3 40 6.28 40 6.79 40 6.56 40 n/a 40 6.23 40 6.25 40 6.27 60 5.81 60 6.05 60 5.33 60 5.34 60 5.42 60 6.14 60 5.76 60 5.72 60 5.18 60 5.52 60 5.46 60 6.07 60 5.80 60 5.71 60 5.15 60 n/a 60 5.27 60 5.81 80 5.05 80 4.99 80 4.54 80 5.23 80 4.93 80 4.79 80 5.01 80 - 80 4.46 80 4.96 80 5.06 80 4.71 80 4.93 80 5.04 80 n/a 80 4.88 80 - 80 4.77 100 4.09 100 4.21 100 3.98 100 4.22 100 4.36 100 4.14 100 4.48 100 4.85 100 3.67 100 4.17 100 4.42 100 4.09 100 4.23 100 4.32 100 3.99 100 4.43 100 3.87 100 4.07 120 3.90 120 3.44 120 - 120 3.85 120 3.83 120 - 120 3.74 120 - 120 4.10 140 - 140 - 140 - 140 - 140 3.54 140 - 140 - 140 3.37 140 - 125 Table E. 2 Raw survival data for Salmonella in almond meal (0.45 a w ) at 85 °C generated by MSU and WSU. Salmonella - 85°C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 8.19 0 7.94 0 7.59 0 7.71 0 7.90 0 7.81 0 8.19 0 8.05 0 7.83 0 7.67 0 7.88 0 7.63 0 8.07 0 7.97 0 7.57 0 7.57 0 7.77 0 7.64 10 6.48 10 7.18 10 7.00 10 7.15 10 7.20 10 7.16 10 7.18 10 7.18 10 6.92 10 7.11 10 7.24 10 7.20 10 7.24 10 - 10 7.00 10 6.96 10 7.17 10 - 20 6.11 20 6.43 20 6.08 20 5.95 20 6.63 20 6.28 20 6.21 20 6.35 20 6.07 20 6.12 20 6.26 20 6.35 20 6.68 20 6.34 20 5.96 20 6.09 20 6.04 20 6.40 30 5.42 30 5.45 30 5.11 30 5.23 30 5.45 30 5.44 30 5.47 30 5.72 30 5.20 30 5.28 30 5.19 30 5.29 30 5.61 30 5.78 30 5.17 30 n/a 30 5.29 30 5.17 40 5.02 40 4.79 40 4.56 40 4.78 40 4.64 40 4.61 40 4.8 0 40 5.09 40 4.37 40 4.86 40 4.88 40 4.57 40 4.92 40 4.94 40 4.38 40 4.90 40 - 40 4.70 50 4.42 50 4.36 50 3.66 50 4.04 50 4.06 50 3.91 50 4.48 50 4.5 50 4.02 50 4.45 50 3.84 50 3.90 50 3.89 50 4.46 50 4.07 50 - 50 4.09 50 3.85 60 3.79 60 3.62 60 - 60 3.51 60 3.73 60 3.49 60 3.71 60 3.64 60 - 60 3.11 60 3.37 60 3.01 60 3.66 60 3.57 60 - 60 2.92 60 3.24 60 2.80 70 - 70 - 70 - 70 - 70 - 70 - 70 - 70 - 70 - 126 Table E. 3 Raw survival data for Salmonella in almond meal (0.45 a w ) at 90 °C generated by MSU and WSU. Salmonella - 90°C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 8.18 0 8.11 0 7.36 0 7.07 0 7.49 0 7.69 0 8.13 0 8.07 0 7.50 0 7.26 0 7.64 0 7.36 0 8.02 0 8.01 0 7.64 0 7.07 0 7.34 0 7.40 5 - 5 7.13 5 6.80 5 6.72 5 6.88 5 6.93 5 7.26 5 7.15 5 6.74 5 6.77 5 6.98 5 7.05 5 7.47 5 7.28 5 5.98 5 6.69 5 6.75 5 6.80 10 6.58 10 6.60 10 6.78 10 5.57 10 6.07 10 5.77 10 6.65 10 6.68 10 6.05 10 5.5 0 10 6.06 10 - 10 6.31 10 6.69 10 5.93 10 5.95 10 - 10 5.94 15 5.84 15 6.11 15 5.21 15 4.95 15 4.96 15 5.27 15 5.74 15 5.90 15 5.40 15 5.17 15 5.07 15 5.19 15 5.80 15 - 15 5.32 15 4.83 15 5.1 0 15 5.35 20 4.89 20 4.83 20 4.53 20 4.90 20 4.39 20 4.66 20 4.92 20 - 20 4.62 20 4.57 20 4.67 20 4.48 20 5.11 20 5.19 20 - 20 4.65 20 4.78 20 4.77 25 4.21 25 4.50 25 3.54 25 4.08 25 4.10 25 4.10 25 4.45 25 4.22 25 3.60 25 3.66 25 4.08 25 3.95 25 4.56 25 4.18 25 3.52 25 4.23 25 - 25 3.99 30 3.96 30 3.92 30 - 30 3.4 0 30 3.22 30 3.39 30 3.68 30 3.99 30 5.14 30 3.09 30 3.40 30 3.69 30 3.45 30 3.54 30 - 30 3.45 30 3.28 30 3.60 35 3.54 35 4.37 35 - 35 3.28 35 3.07 35 3.56 35 3.54 35 3.08 35 - 127 Table E. 4 Raw survival data for Enterococcus faecium NRRL B - 2354 in almond meal (0.45 a w ) at 80 °C generated by MSU and WSU. E. faecium - 80°C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 - 0 8.45 0 7.98 0 7.49 0 7.49 0 8.03 0 7.75 0 8.09 0 8.07 0 7.56 0 7.05 0 7.99 0 7.49 0 7.97 0 8.00 0 - 0 7.04 0 7.86 20 6.83 20 7.25 20 7.14 20 6.45 20 6.95 20 7.15 20 7.09 20 7.05 20 7.17 20 6.45 20 6.91 20 7.22 20 6.82 20 6.98 20 7.23 20 - 20 6.85 20 6.93 40 6.06 40 6.37 40 6.51 40 6.07 40 6.6 0 40 6.32 40 6.07 40 6.03 40 6.31 40 5.96 40 5.96 40 6.60 40 5.96 40 6.07 40 6.45 40 5.91 40 6.03 40 - 60 5.78 60 5.55 60 5.80 60 - 60 5.83 60 6.02 60 5.69 60 5.69 60 5.78 60 5.84 60 5.29 60 5.95 60 5.64 60 5.64 60 5.78 60 5.49 60 5.08 60 5.97 80 5.15 80 5.21 80 5.35 80 5.33 80 5.45 80 5.42 80 5.26 80 5.14 80 5.56 80 5.18 80 5.25 80 5.53 80 5.27 80 5.13 80 5.73 80 5.11 80 4.87 80 5.26 100 4.76 100 4.85 100 4.87 100 4.81 100 4.85 100 4.94 100 4.87 100 4.91 100 4.99 100 5.06 100 4.8 0 100 4.84 100 4.65 100 4.89 100 4.95 100 4.71 100 4.67 100 4.76 120 4.24 120 4.80 120 4.51 120 4.12 120 4.77 120 4.76 120 4.12 120 4.24 120 4.71 120 4.10 120 4.11 120 4.81 120 3.77 120 4.38 120 4.57 120 4.12 120 4.25 120 - 140 3.81 140 3.91 140 4.42 140 3.61 140 - 140 4.25 140 3.41 140 3.83 140 4.18 128 Table E. 5 Raw survival data for Enterococcus faecium NRRL B - 2354 in almond meal (0.45 a w ) at 85 °C generated by MSU and WSU. E.faecium - 85°C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.40 0 8.02 0 8.04 0 7.62 0 6.69 0 7.68 0 7.53 0 8.10 0 7.83 0 7.81 0 6.75 0 7.51 0 7.56 0 8.10 0 7.95 0 7.55 0 6.56 0 8.00 10 6.86 10 - 10 7.02 10 7.29 10 6.33 10 6.99 10 6.86 10 6.49 10 6.98 10 7.01 10 6.47 10 6.94 10 6.82 10 6.78 10 7.02 10 7.13 10 - 10 7.03 20 5.97 20 5.98 20 6.29 20 6.83 20 - 20 6.41 20 6.01 20 5.8 0 20 6.19 20 6.33 20 5.67 20 5.77 20 6.10 20 6.00 20 6.32 20 6.42 20 5.75 20 - 30 5.66 30 5.57 30 5.75 30 5.50 30 5.15 30 5.94 30 5.5 0 30 5.55 30 5.69 30 5.35 30 5.15 30 5.89 30 5.67 30 5.51 30 5.96 30 5.19 30 - 30 6.03 40 5.22 40 5.13 40 5.37 40 5.87 40 5.03 40 5.10 40 5.26 40 5.14 40 5.38 40 5.9 40 4.95 40 5.29 40 5.46 40 5.09 40 5.31 40 5.29 40 4.76 40 - 50 4.84 50 4.66 50 4.96 50 5.24 50 4.53 50 4.81 50 4.97 50 4.58 50 4.86 50 5.18 50 4.60 50 4.71 50 4.96 50 4.83 50 4.77 50 5.31 50 4.73 50 4.69 60 4.69 60 4.16 60 4.63 60 4.48 60 3.84 60 4.01 60 4.21 60 4.35 60 4.50 60 4.52 60 3.97 60 4.38 60 4.23 60 3.91 60 4.60 60 4.45 60 6.69 60 4.42 70 3.87 70 3.52 70 4.21 70 3.99 70 3.42 70 4.01 70 4.11 70 - 70 4.03 129 Table E. 6 Raw survival data for Enterococcus faecium NRRL B - 2354 in almond meal (0.45 a w ) at 90 °C generated by MSU and WSU. E. faecium - 90°C MSU WSU Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 - 0 7.94 0 7.62 0 7.50 0 6.66 0 7.93 0 7.36 0 7.78 0 7.77 0 7.52 0 6.60 0 7.68 0 7.55 0 7.78 0 7.88 0 - 0 6.39 0 - 5 6.60 5 6.70 5 6.72 5 6.76 5 6.13 5 7.31 5 6.62 5 6.69 5 6.73 5 6.74 5 6.10 5 7.19 5 6.17 5 6.44 5 6.02 5 7.00 5 6.39 5 7.14 10 5.91 10 5.72 10 6.67 10 5.64 10 - 10 6.00 10 5.96 10 5.85 10 6.04 10 5.67 10 5.47 10 5.65 10 - 10 5.83 10 5.92 10 - 10 5.74 10 5.85 15 5.68 15 5.73 15 5.63 15 5.76 15 5.19 15 6.04 15 5.47 15 5.68 15 5.65 15 6.04 15 5.09 15 5.92 15 5.15 15 5.62 15 5.55 15 5.34 15 4.92 15 5.78 20 5.34 20 4.94 20 5.25 20 5.27 20 4.89 20 5.53 20 5.29 20 4.98 20 5.19 20 5.27 20 4.89 20 4.88 20 5.16 20 5.17 20 5.22 20 4.97 20 4.95 20 4.86 25 3.81 25 4.87 25 4.86 25 5.16 25 4.42 25 4.70 25 3.92 25 4.70 25 4.83 25 4.99 25 4.06 25 4.69 25 4.83 25 4.83 25 4.87 25 - 25 4.00 25 4.81 30 4.27 30 4.10 30 4.35 30 4.49 30 3.78 30 4.36 30 4.45 30 4.17 30 4.59 30 4.49 30 3.42 30 4.45 30 3.57 30 4.23 30 4.49 30 - 30 3.53 30 4.09 35 3.35 35 3.86 35 4.02 35 3.18 35 3.76 35 3.97 35 3.71 35 3.69 35 3.87 130 APPENDIX F . Isothermal inactivation data for wheat flour at 0.45 a w (chapter 4) Table F. 1 Raw survival data for Salmonella in wheat flour (0.45 a w ) at 70 °C generated by WSU and IFSH. Salmonella - 70°C WSU IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 8.17 0 7.64 0 8.19 0 8.52 0 8.29 0 8.19 0 8.12 0 7.79 0 8.21 0 8.01 0 8.28 0 8.22 0 8.14 0 7.75 0 8.32 0 8.01 0 8.2 0 8.15 10 7.65 35 6.50 35 6.87 30 6.67 30 6.88 30 7.00 10 7.87 35 6.39 35 6.82 30 6.78 30 6.97 30 7.02 10 7.65 35 6.33 35 7.32 30 8.24 30 8.24 30 8.24 20 7.66 70 5.46 70 6.2 60 n/a 60 n/a 60 - 20 7.67 70 5.58 70 5.95 60 6.63 60 6.15 60 6.24 20 7.34 70 5.38 70 6.44 60 6.21 60 6.15 60 6.37 40 7.36 105 4.80 105 6.43 90 4.62 90 5.7 90 5.49 40 7.23 105 4.94 105 5.77 90 5.93 90 5.93 90 5.93 40 7.16 105 4.63 105 6.07 90 5.86 90 5.63 90 5.66 60 6.58 140 4.23 140 4.64 120 5.24 120 5.33 120 3.96 60 6.33 140 3.85 140 5.38 120 5.66 120 4.53 120 4.46 60 6.88 140 4.07 140 4.85 120 5.86 120 5.33 120 4.56 80 6.32 175 3.45 175 4.42 180 4.00 180 4.5 180 4.04 80 6.08 175 3.90 175 4.46 180 4.38 180 5.07 180 4.55 80 6.30 175 3.89 175 4.58 180 4.09 180 5.25 180 4.31 100 6.43 210 - 210 3.76 100 6.28 210 3.48 210 3.87 131 Table F Salmonella - 70 °C W SU IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 100 5.86 210 - 210 3.75 130 5.27 130 5.72 130 5.78 160 5.52 160 5.25 160 4.58 190 4.67 190 4.46 190 4.58 132 Table F. 2 Raw survival data for Salmonella in wheat flour (0.45 a w ) at 75 °C generated by WSU and IFSH. Salmonella - 75°C WSU IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.86 0 7.43 0 8.19 0 n/a 0 7.76 0 - 0 7.63 0 7.47 0 8.23 0 7.96 0 7.78 0 7.77 0 7.85 0 7.48 0 7.94 0 8.10 0 7.69 0 7.63 5 7.16 5 6.22 5 8.11 21 5.75 21 5.74 21 6.58 5 7.89 5 6.24 5 6.85 21 6.16 21 6.18 21 6.20 5 8.01 5 6.22 5 6.86 21 6.12 21 5.89 21 6.00 15 6.14 15 5.18 15 6.92 40 5.80 40 5.02 40 4.57 15 6.89 15 5.80 15 6.63 40 5.60 40 4.82 40 6.07 15 7.01 15 5.16 15 6.63 40 5.75 40 4.96 40 6.15 30 5.64 30 4.85 30 6.49 60 5.77 60 3.75 60 4.55 30 6.34 30 4.86 30 5.8 0 60 5.19 60 4.00 60 4.46 30 6.43 30 4.76 30 6.01 60 5.25 60 4.29 60 4.62 45 5.67 45 4.48 45 5.87 82 - 82 3.86 82 4.03 45 5.64 45 3.61 45 5.1 0 82 4.26 82 3.86 82 3.69 45 5.69 45 4.10 45 5.62 82 4.20 82 3.88 82 3.78 60 5.39 60 4.20 60 5.19 100 - 100 - 100 3.80 60 5.16 60 4.34 60 4.45 100 3.98 100 3.87 100 - 60 5.26 60 3.76 60 4.81 100 - 100 - 100 - 75 3.71 75 3.73 75 4.91 75 4.18 75 3.72 75 4.28 75 4.78 75 - 75 3.91 90 4.14 90 - 90 3.99 90 - 90 4.22 90 - 133 Table F. 3 Raw survival data for Salmonella in wheat flour (0.45 a w ) at 80 °C generated by WSU and IFSH. Salmonella - 80°C WSU IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.62 0 7.06 0 7.35 0 8.00 0 7.39 0 7.66 0 7.65 0 7.17 0 7.24 0 7.69 0 7.20 0 7.82 0 7.44 0 7.19 0 7.59 0 7.71 0 7.22 0 7.67 2 6.82 2 6.83 2 6.99 9 6.32 21 5.61 21 6.88 2 7.07 2 6.92 2 7.57 9 6.76 21 5.57 21 6.96 2 - 2 6.95 2 7.29 9 6.92 21 6.00 21 6.00 10 6.25 10 6.09 10 6.97 18 6.47 40 4.64 40 5.15 10 6.20 10 6.00 10 7.03 18 5.77 40 4.79 40 5.40 10 6.37 10 5.82 10 6.48 18 6.01 40 4.72 40 5.34 18 5.60 18 5.44 18 6.03 27 - 60 3.95 60 4.34 18 5.43 18 5.07 18 5.57 27 3.95 60 4.22 60 4.22 18 6.06 18 5.05 18 5.84 27 3.61 60 4.17 60 5.26 26 5.98 26 4.65 26 5.17 36 - 82 3.99 82 4.61 26 5.87 26 4.34 26 5.37 36 3.78 82 - 82 4.34 26 4.78 26 4.42 26 5.32 36 4.05 82 4.42 82 4.10 34 5.13 34 4.04 34 4.55 45 n/a 100 - 100 - 34 4.82 34 3.99 34 4.23 45 3.81 100 - 100 3.63 34 4.88 34 3.96 34 4.02 45 - 100 3.58 100 4.04 42 - 39 3.77 39 3.41 42 4.36 39 3.63 39 3.56 42 3.55 39 3.72 39 3.61 50 - 50 - 50 - 134 Table F. 4 Raw survival data for E. faecium NRRL B - 2354 in wheat flour (0.45 a w ) at 75°C generated by WSU and IFSH. E. faecium - 75°C WSU IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.81 0 8.34 0 8.12 0 8.01 0 8.23 0 8.01 0 7.97 0 8.11 0 8.01 0 8.01 0 8.31 0 8.16 0 8.08 0 8.04 0 8.06 0 8.26 0 8.37 0 8.40 20 7.36 20 7.53 20 7.22 20 7.40 20 7.21 20 7.17 20 7.31 20 7.20 20 7.84 20 7.32 20 7.25 20 6.95 20 7.24 20 7.27 20 7.37 20 8.24 20 7.27 20 7.18 40 6.56 40 6.54 40 6.79 40 - 40 6.24 40 6.06 40 6.66 40 6.58 40 6.91 40 6.70 40 6.44 40 6.30 40 6.58 40 6.59 40 6.86 40 6.71 40 6.34 40 6.48 60 6.01 60 5.92 60 5.74 60 5.93 60 5.99 60 5.83 60 5.91 60 5.95 60 6.05 60 5.93 60 5.99 60 5.84 60 5.99 60 6.27 60 5.87 60 5.82 60 6.17 60 6.46 80 4.57 80 5.21 80 4.32 80 4.84 80 5.04 80 - 80 4.75 80 5.23 80 4.34 80 4.92 80 5.03 80 5.43 80 4.51 80 5.15 80 4.44 80 5.10 80 5.05 80 4.69 100 3.72 95 4.52 100 3.57 100 4.30 100 4.69 100 4.15 100 3.53 95 4.75 100 3.91 100 4.03 100 4.72 100 4.53 100 3.75 95 4.91 100 3.65 100 3.98 100 4.06 100 3.96 120 - 120 - 120 - 135 Table F. 5 Raw survival data for E. faecium NRRL B - 2354 in wheat flour (0.45 a w ) at 80 °C generated by WSU and IFSH. E. faecium - 80°C WSU IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.93 0 8.05 0 7.97 0 7.71 0 8.21 0 7.73 0 8.01 0 8.19 0 7.84 0 8.00 0 8.15 0 7.79 0 7.86 0 8.15 0 7.88 0 7.98 0 8.24 0 7.71 7 7.01 7 7.26 7 7.27 8 7.04 8 6.99 8 6.79 7 6.97 7 7.24 7 7.28 8 7.28 8 7.14 8 6.54 7 6.94 7 7.32 7 7.31 8 6.92 8 7.21 8 6.89 14 6.18 14 6.59 14 6.62 16 6.27 16 6.59 16 6.00 14 6.45 14 6.40 14 6.95 16 5.85 16 4.16 16 6.43 14 6.30 14 6.59 14 6.67 16 6.57 16 6.31 16 5.77 21 5.74 21 6.19 21 5.93 24 5.99 24 5.53 24 5.09 21 5.74 21 6.19 21 5.71 24 6.01 24 5.66 24 4.96 21 5.70 21 6.22 21 5.60 24 5.42 24 5.69 24 5.04 28 5.18 28 5.59 28 5.00 32 4.91 32 5.11 32 4.31 28 5.11 28 5.76 28 4.85 32 5.09 32 4.97 32 3.97 28 5.32 28 5.69 28 4.88 32 5.00 32 5.23 32 - 35 4.45 35 5.14 35 3.79 40 4.47 40 3.90 40 3.57 35 4.17 35 4.82 35 4.08 40 4.6 0 40 4.02 40 4.44 35 4.44 35 5.02 35 3.79 40 4.74 40 3.72 40 3.57 42 3.39 42 3.99 42 3.42 42 3.53 42 4.38 42 - 42 3.47 42 4.01 42 - 136 Table F. 6 Raw survival data for E. faecium NRRL B - 2354 in wheat flour (0.45 a w ) at 85 °C generated by WSU and IFSH. E. faecium - 85°C WSU IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.45 0 7.56 0 7.47 0 7.84 0 7.83 0 7.02 0 7.56 0 7.90 0 7.85 0 7.82 0 8.33 0 7.60 0 7.55 0 7.69 0 7.99 0 7.91 0 8.07 0 7.54 2 7.11 2 7.04 2 6.86 3 - 3 7.37 3 6.13 2 6.81 2 7.22 2 6.81 3 8.08 3 7.30 3 6.25 2 6.86 2 7.18 2 6.76 3 7.02 3 7.27 3 6.11 4 5.97 4 6.44 4 6.13 6 5.98 6 6.72 6 5.33 4 6.15 4 6.49 4 5.92 6 5.78 6 6.87 6 5.20 4 5.95 4 6.54 4 5.95 6 6.31 6 6.85 6 5.29 6 5.39 6 5.93 6 5.48 9 5.28 9 5.28 9 4.40 6 5.46 6 5.86 6 5.04 9 5.33 9 5.33 9 4.34 6 5.79 6 5.71 6 5.40 9 5.41 9 5.61 9 4.26 9 4.39 9 5.18 9 4.73 12 5.01 12 4.75 12 3.75 9 4.35 9 5.00 9 4.53 12 5.32 12 3.90 12 - 9 4.50 9 5.06 9 4.61 12 5.02 12 4.83 12 3.95 12 3.62 12 5.04 12 3.74 15 4.15 15 4.58 15 3.59 12 3.71 12 4.81 12 3.63 15 3.92 15 4.63 15 3.59 12 3.74 12 4.59 12 4.09 15 4.91 15 4.29 15 3.63 15 - 15 3.48 15 3.47 15 - 15 3.53 15 3.54 15 - 15 4.46 15 3.77 137 APPENDIX G . Isothermal inactivation data for ground black pepper at 0.45 a w (chapter 4) Table G. 1 Raw survival data for Salmonella in ground black pepper (0.45 a w ) at 65 °C generated by UNL and IFSH. Salmonella - 65°C UNL IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 6.45 0 6.31 0 8.19 0 6.07 0 6.50 0 6.60 0 6.38 0 6.51 0 8.21 0 6.27 0 6.53 0 6.67 0 6.51 0 - 0 8.32 0 6.16 0 6.44 0 6.57 10 5.64 10 5.37 15 6.87 12 5.97 12 5.31 15 6.18 10 5.31 10 5.47 15 6.82 12 5.85 12 5.56 15 6.52 10 5.43 10 5.33 15 7.32 12 5.71 12 5.18 15 6.28 20 4.87 20 4.97 30 6.2 0 24 - 24 4.86 30 5.42 20 5.00 20 5.05 30 5.95 24 - 24 5.05 30 5.38 20 4.90 20 5.05 30 6.44 24 - 24 4.63 30 5.26 30 4.33 30 4.54 45 6.43 36 4.14 36 4.36 45 4.74 30 4.19 30 4.42 45 5.77 36 - 36 4.55 45 4.73 30 4.27 30 4.37 45 6.07 36 3.61 36 4.14 45 4.81 40 4.03 40 3.66 60 4.64 48 3.45 48 3.97 60 4.39 40 3.84 40 4.00 60 5.38 48 3.47 48 4.18 60 4.32 40 4.11 40 3.83 60 4.85 48 3.95 48 4.09 60 4.27 50 3.03 50 3.58 75 4.42 60 2.81 60 3.62 75 3.37 50 3.65 50 - 75 4.46 60 3.04 60 3.24 75 3.36 50 3.51 50 3.35 75 4.58 60 3.22 60 3.78 75 3.68 90 2.48 90 2.45 90 - 138 Table G. 2 Raw survival data for Salmonella in ground black pepper (0.45 a w ) at 70 °C generated by UNL and IFSH. Salmonella - 70°C UNL IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 5.93 0 5.82 0 5.77 0 5.97 0 6.27 0 6.51 0 5.94 0 6.07 0 5.86 0 5.88 0 5.99 0 6.38 0 5.92 0 5.96 0 5.97 0 5.93 0 6.03 0 6.64 6 4.80 6 4.70 6 4.69 5 5.07 5 5.03 8 5.67 6 4.82 6 4.65 6 4.81 5 4.63 5 5.08 8 6.00 6 4.80 6 4.71 6 4.56 5 4.70 5 5.16 8 5.84 12 3.96 12 3.87 12 4.27 10 3.87 10 4.8 16 5.04 12 3.91 12 3.79 12 - 10 3.77 10 4.72 16 5.03 12 3.97 12 4.06 12 4.24 10 3.88 10 4.76 16 4.64 18 2.86 18 3.35 18 3.63 15 3.85 15 4.04 24 4.46 18 3.12 18 3.42 18 4.10 15 3.55 15 3.81 24 4.45 18 3.44 18 3.47 18 3.69 15 3.58 15 4.04 24 4.29 24 - 24 2.75 24 3.24 20 2.51 20 3.98 32 3.76 24 3.00 24 2.75 24 3.39 20 2.88 20 4.05 32 3.88 24 2.66 24 2.90 24 - 20 3.36 20 4.44 32 4.15 30 - 30 3.18 30 3.32 25 2.77 25 3.33 40 3.33 30 2.50 30 2.65 30 3.25 25 2.77 25 3.28 40 3.34 30 - 30 2.81 30 3.09 25 3.71 25 3.46 40 3.07 36 2.98 36 - 36 2.90 139 Table G. 3 Raw survival data for Salmonella in ground black pepper (0.45 a w ) at 75 °C generated by UNL and IFSH. Salmonella - 75°C UNL IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 5.58 0 5.21 0 5.69 0 5.27 0 5.24 0 5.77 0 5.71 0 5.77 0 5.73 0 5.63 0 5.57 0 5.90 0 5.73 0 5.28 0 5.52 0 5.26 0 5.22 0 5.90 2. 8 4.56 2. 8 4.92 2.5 4.24 2 4.29 2 4.36 2 4.91 2. 8 4.24 2. 8 4.67 2.5 4.15 2 4.56 2 4.61 2 5.27 2. 8 4.73 2. 8 4.47 2.5 4.70 2 4.68 2 4.67 2 5.08 5.5 3.82 5.5 3.43 5 3.51 4 - 4 3.98 4 4.78 5.5 3.29 5.5 3.54 5 3.68 4 3.68 4 4.08 4 4.64 5.5 3.17 5.5 3.90 5 3.67 4 3.95 4 3.81 4 5.05 8. 3 2.56 8. 3 2.69 7.5 3.50 6 3.21 6 3.69 6 4.10 8. 3 2.72 8. 3 2.98 7.5 3.80 6 3.17 6 3.17 6 3.98 8. 3 2.66 8. 3 3.13 7.5 3.70 6 3.59 6 3.08 6 4.24 11 - 11 - 10 2.96 8 2.63 8 3.58 8 3.36 11 - 11 2.65 10 2.95 8 2.87 8 3.19 8 3.65 11 2.66 11 2.64 10 3.00 8 3.14 8 2.90 8 3.53 13. 8 - 13. 8 - 12.5 2.49 10 - 10 - 10 2.95 13. 8 - 13. 8 - 12.5 2.90 10 - 10 - 10 2.45 13. 8 - 13. 8 - 12.5 - 10 - 10 - 10 3.02 15 - 15 2.40 15 - 140 Table G. 4 Raw survival data for E. faecium NRRL B - 2354 in ground black pepper (0.45 a w ) at 70 °C generated by UNL and IFSH. E. faecium - 70°C UNL IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.26 0 7.17 0 7.56 0 7.72 0 7.37 0 n/a 0 7.29 0 7.17 0 7.58 0 7.52 0 7.29 0 7.34 0 - 0 7.31 0 7.59 0 7.33 0 7.43 0 7.38 16 6.72 30 6.43 30 6.08 16 6.88 20 7.13 24 7.08 16 6.73 30 6.35 30 5.90 16 6.64 20 7.17 24 7.06 16 6.75 30 6.21 30 5.97 16 6.63 20 6.95 24 6.87 32 5.95 60 5.29 60 4.60 32 5.66 40 6.70 48 6.23 32 6.11 60 5.37 60 4.47 32 5.53 40 6.62 48 6.28 32 6.09 60 5.28 60 4.09 32 - 40 6.50 48 6.43 48 4.99 90 4.39 90 3.45 48 5.04 60 6.16 72 5.83 48 5.01 90 - 90 3.23 48 5.46 60 5.85 72 5.95 48 5.18 90 4.47 90 3.29 48 5.15 60 5.72 72 5.84 64 4.68 120 3.61 120 2.61 64 4.58 80 5.48 96 5.07 64 4.61 120 3.58 120 2.41 64 4.60 80 5.46 96 5.20 64 - 120 3.99 120 - 64 4.21 80 5.69 96 4.36 80 - 150 - 80 - 100 4.37 120 3.40 80 4.06 150 - 80 3.56 100 4.34 120 3.85 80 3.86 150 - 80 3.68 100 4.30 120 3.66 141 Table G. 5 Raw survival data for E. faecium NRRL B - 2354 in ground black pepper (0.45 a w ) at 75 °C generated by UNL and IFSH. E. faecium - 75°C UNL IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 - 0 7.22 0 7.30 0 8.17 0 7.56 0 7.58 0 7.20 0 7.37 0 7.37 0 - 0 - 0 7.62 0 - 0 7.32 0 7.54 0 7.18 0 7.64 0 7.56 7 6.31 7 6.72 7 6.83 7 6.94 7 7.12 10 6.25 7 - 7 6.65 7 6.64 7 6.26 7 7.1 0 10 6.25 7 6.03 7 6.45 7 6.53 7 7.02 7 6.86 10 6.36 14 5.23 14 5.89 14 5.48 14 5.84 14 6.3 0 20 5.22 14 5.32 14 6.43 14 5.25 14 5.54 14 6.38 20 5.45 14 5.03 14 - 14 5.17 14 5.31 14 6.51 20 5.12 21 4.44 21 5.54 21 4.27 21 4.72 21 5.83 30 4.3 0 21 4.86 21 4.95 21 4.42 21 4.82 21 6.18 30 4.37 21 4.20 21 - 21 4.48 21 4.48 21 6.14 30 4.49 28 3.64 28 5.25 28 3.06 28 3.69 28 5.42 40 3.64 28 3.45 28 5.22 28 3.22 28 4.08 28 5.51 40 - 28 3.73 28 5.44 28 3.07 28 3.94 28 5.41 40 3.83 35 3.04 35 4.35 35 - 35 3.75 35 4.67 50 - 35 3.25 35 4.34 35 2.61 35 3.29 35 4.84 50 - 35 3.40 35 4.31 35 2.64 35 3.20 35 4.88 50 - 142 Table G. 6 Raw survival data for E. faecium NRRL B - 2354 in ground black pepper (0.45 a w ) at 80 °C generated by UNL and IFSH. E. faecium - 80°C UNL IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 6.99 0 7.07 0 7.10 0 6.81 0 7.03 0 7.86 0 6.94 0 7.13 0 6.97 0 7.57 0 7.46 0 7.44 0 6.90 0 7.13 0 6.89 0 7.05 0 7.30 0 7.29 3 5.71 3 6.31 3 5.71 2 6.29 2 6.61 4 5.86 3 5.72 3 6.36 3 6.10 2 6.59 2 6.33 4 5.85 3 5.57 3 6.44 3 5.71 2 6.10 2 6.77 4 6.21 6 4.18 6 - 6 4.73 4 5.64 4 5.59 8 4.82 6 4.51 6 - 6 4.81 4 5.31 4 5.85 8 4.76 6 4.72 6 - 6 4.91 4 5.74 4 5.34 8 4.98 9 3.02 9 5.1 0 9 3.42 6 5.10 6 4.88 12 3.74 9 3.24 9 4.84 9 3.71 6 4.92 6 5.02 12 - 9 4.29 9 4.23 9 4.00 6 5.86 6 4.88 12 3.92 12 2.59 12 4.06 12 2.81 8 3.99 8 4.59 16 3.86 12 2.44 12 4.38 12 2.65 8 4.35 8 4.66 16 2.82 12 2.56 12 4.14 12 2.83 8 3.93 8 4.07 16 3.16 15 - 15 3.98 15 - 10 3.24 10 4.10 20 2.98 15 - 15 3.67 15 2.53 10 3.31 10 4.04 20 2.64 15 - 15 4.07 15 - 10 3.73 10 3.76 20 - 143 APPENDIX H . Isothermal inactivation data for date paste at 0.65 a w (chapter 4) Table H. 1 Raw survival data for Salmonella in date paste (0. 6 5 a w ) at 65 °C generated by MSU and UGA . Salmonella - 65°C MSU UGA Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU/ g) Time (min) Rep 4 (Log CFU /g) Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU /g) 0 7.27 0 7.30 0 7.56 0 7.87 0 7.28 0 7.77 0 7.75 0 7.73 0 7.63 0 7.18 0 7.54 0 7.89 0 7.16 0 7.77 0 7.73 0 8.17 0 7.08 0 7.19 0 - 0 8.02 0 - 0 7.69 0 7.64 0 8.02 8 6.26 8 6.70 8 6.94 8 7.16 7 6.06 8 7.12 8 6.94 8 7.15 8 6.71 8 7.34 8 6.71 8 7.80 7 5.87 8 7.09 8 6.93 8 6.94 8 6.63 8 6.64 8 6.79 8 7.05 7 6.12 8 6.82 8 6.97 8 6.90 16 6.55 16 6.11 16 6.24 16 n/a 14 5.86 16 6.69 16 6.69 16 7.33 16 6.10 16 6.09 16 6.07 16 n/a 14 6.26 16 6.68 16 5.93 16 7.23 16 5.89 16 6.63 16 6.53 16 6.78 14 5.98 16 6.71 16 6.07 16 7.04 24 5.92 24 6.66 24 6.48 24 6.97 21 5.45 24 6.38 24 6.10 24 6.30 24 5.49 24 5.74 24 6.2 24 6.78 21 6.12 24 5.5 24 5.92 24 6.59 24 5.17 24 5.88 24 6.28 24 6.85 21 5.23 24 6.34 24 5.98 24 7.14 32 5.85 32 5.67 32 5.23 32 5.98 28 3.96 32 4.89 32 4.14 32 5.19 32 5.10 32 5.70 32 4.87 32 n/a 28 6.01 32 4.76 32 4.92 32 5.64 32 4.88 32 5.03 32 5.62 32 n/a 28 4.01 32 5.07 32 5.00 32 5.05 40 - 40 4.94 40 3.67 40 5.61 35 - 40 4.11 40 2.65 40 - 40 4.02 40 - 40 3.73 40 6.27 35 - 40 4.13 40 - 40 3.38 40 3.74 40 4.90 40 3.57 40 6.17 35 - 40 4.13 40 2.64 40 3.71 48 - 48 2.79 48 5.04 48 - 48 - 48 - 48 4.96 48 - 48 - 48 - 48 3.80 48 n/a 144 Table H. 2 Raw survival data for Salmonella in date paste (0. 6 5 a w ) at 70 °C generated by MSU and UGA. Salmonella - 70 °C MSU UGA Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU/ g) Time (min) Rep 4 (Log CFU /g) Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU /g) 0 7.27 0 7.62 0 7.1 0 0 7.51 0 7.28 0 7.77 0 7.75 0 7.73 0 7.63 0 7.51 0 7.23 0 7.44 0 7.16 0 7.77 0 7.73 0 8.17 0 7.08 0 7.39 0 7.41 0 7.62 0 - 0 7.69 0 7.64 0 8.02 2.5 6.26 2.5 7.12 2.5 6.62 2.5 7.24 2 6.06 8 7.12 8 6.94 8 7.15 2.5 6.71 2.5 6.62 2.5 6.68 2.5 7.11 2 5.87 8 7.09 8 6.93 8 6.94 2.5 6.63 2.5 6.8 2.5 6.67 2.5 7.07 2 6.12 8 6.82 8 6.97 8 6.90 5 6.55 5 6.1 0 5 6.63 5 6.75 4 5.86 16 6.69 16 6.69 16 7.33 5 6.10 5 6.34 5 5.84 5 6.7 4 6.26 16 6.68 16 5.93 16 7.23 5 5.89 5 6.51 5 n/a 5 6.75 4 5.98 16 6.71 16 6.07 16 7.04 7.5 5.92 7.5 6.1 0 7.5 5.78 7.5 6.77 6 5.45 24 6.38 24 6.10 24 6.30 7.5 5.49 7.5 6.34 7.5 6.65 7.5 6.49 6 6.12 24 5.5 24 5.92 24 6.59 7.5 5.17 7.5 6.48 7.5 5.99 7.5 6.55 6 5.23 24 6.34 24 5.98 24 7.14 10 5.85 10 4.86 10 5.15 10 6 .00 8 3.96 32 4.89 32 4.14 32 5.19 10 5.10 10 5.35 10 5.67 10 6.06 8 6.01 32 4.76 32 4.92 32 5.64 10 4.88 10 5.52 10 - 10 - 8 4.01 32 5.07 32 5.00 32 5.05 12.5 - 12.5 4.08 12.5 4.69 12.5 5.07 10 - 40 4.11 40 2.65 40 - 12.5 4.02 12.5 4.37 12.5 5.19 12.5 5.8 0 10 - 40 4.13 40 - 40 3.38 12.5 3.74 12.5 5.63 12.5 5.48 12.5 5.31 10 - 40 4.13 40 2.64 40 3.71 15 - 15 3.22 15 4.67 15 4.89 12 48 - 15 - 15 2.8 0 15 4.63 15 - 12 48 - 15 - 15 4.11 15 4.92 15 - 12 48 - 17.5 3.27 14 17.5 - 14 145 Table H. 3 Raw survival data for Salmonella in date paste (0. 6 5 a w ) at 75 °C generated by MSU and UGA. 17.5 3.24 14 Salmonella - 75°C MSU UGA Time (min) Rep 1 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU/ g) Time (min) Rep 4 (Log CFU /g) Time (min) Rep 2 (Log CFU /g) Time (min) Rep 3 (Log CFU /g) Time (min) Rep 4 (Log CFU /g) 0 6.02 0 6.01 0 6.78 0 5.92 0 7.37 0 7.39 0 - 0 6.35 0 5.91 0 6.06 0 6.48 0 7.1 0 0 7.35 0 7.76 0 5.37 0 5.56 0 6.84 0 6.70 0 7.29 0 7.51 0 6.71 0.7 5.55 0.7 5.40 0.7 5.53 1 6.17 1 7.31 1 6.45 1 6.33 0.7 5.59 0.7 5.14 0.7 6.12 1 5.96 1 6.12 1 6.88 1 7.33 0.7 - 0.7 5.68 0.7 5.45 1 5.9 1 5.96 1 6.88 1 5.49 1.3 4.94 1.3 5.73 1.3 6.17 2 5.18 2 5.68 2 4.93 2 5.24 1.3 5.41 1.3 4.84 1.3 5.1 0 2 5.09 2 6.33 2 5.63 2 5.62 1.3 5.04 1.3 5.60 1.3 4.83 2 4.82 2 5.13 2 5.89 2 4.22 2 4.95 2 3.87 2 5.15 3 5.01 3 4.75 3 5.35 3 4.98 2 4.88 2 4.73 2 5.6 0 3 5.77 3 5.36 3 4.80 3 4.67 2 4.75 2 4.34 2 5.27 3 - 3 4.78 3 6.21 3 4.05 2.7 4.40 2.7 - 2.7 4.06 4 3.89 4 3.88 4 4.81 4 4.46 2.7 - 2.7 4.32 2.7 5.56 4 4.83 4 4.81 4 4.52 4 4.14 2.7 4.30 2.7 - 2.7 4.12 4 3.80 4 4.85 4 4.14 4 3.11 3.3 3.20 3.3 4.79 3.3 3.94 5 - 5 3.05 5 n/a 5 4.18 3.3 3.58 3.3 4.58 3.3 4.62 5 4.02 5 4.50 5 4.10 5 3.68 3.3 4.16 3.3 3.64 3.3 3.94 5 - 5 - 5 4.91 5 2.68 4 3.48 4 - 4 3.52 6 2.98 6 - 6 - 6 3.57 4 3.64 4 2.77 4 4.51 6 3.01 6 - 6 - 6 - 4 2.96 4 3.45 4 - 6 3.28 6 - 6 - 6 - 146 Table H. 4 Raw survival data for E. faecium NRRL B - 2354 in date paste (0. 6 5 a w ) at 70 °C generated by MSU and UGA. E. faecium - 7 0 °C MSU UGA Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.3 0 0 7.36 0 7.85 0 7.87 0 7.9 0 0 8.07 0 7.49 0 7.48 0 7.82 0 7.99 0 7.7 0 0 8.13 0 7.53 0 7.53 0 8.08 0 7.96 0 7.77 0 8.10 4 6.89 4 6.85 4 - 5 6.32 5 5.4 0 5 7.32 4 6.87 4 6.93 4 - 5 6.23 5 5.55 5 6.78 4 6.94 4 7.03 4 - 5 6.13 5 5.62 5 7.04 8 6.62 8 6.74 8 7.30 10 6.13 10 4.78 10 6.42 8 6.52 8 6.61 8 7.35 10 5.79 10 4.97 10 6.42 8 6.43 8 6.49 8 7.00 10 5.92 10 4.99 10 6.54 12 5.95 12 5.79 12 6.57 15 5.48 15 3.62 15 5.94 12 5.95 12 6.02 12 6.84 15 5.74 15 3.57 15 5.68 12 6.00 12 6.04 12 6.70 15 4.79 15 3.72 15 5.62 16 5.85 16 5.38 16 6.27 20 4.36 20 1.74 20 4.61 16 5.66 16 5.60 16 6.10 20 4.23 20 2.6 20 5.10 16 5.9 0 16 5.34 16 n/a 20 4.74 20 2.57 20 5.19 20 5.6 0 20 5.21 20 5.81 25 2.26 25 - 25 4.20 20 5.92 20 5.18 20 - 25 3.72 25 - 25 4.29 20 5.82 20 5.41 20 6.01 25 - 25 - 25 - 24 5.07 24 4.93 24 4.84 30 3.34 24 4.69 24 4.94 24 5.08 30 2.45 24 4.81 24 4.7 24 5.00 30 - 28 4.42 28 3.45 28 4.87 28 4.38 28 3.48 28 4.42 28 4.75 28 3.20 28 4.90 147 Table H. 5 Raw survival data for E. faecium NRRL B - 2354 in date paste (0. 6 5 a w ) at 75 °C generated by MSU and UGA. E. faecium - 75°C UNL IFSH Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.19 0 6.68 0 7.11 0 7.62 0 7.07 0 8.03 0 7.31 0 6.66 0 7.02 0 7.37 0 7.06 0 7.92 0 7.19 0 7.14 0 7.19 0 7.59 0 6.94 0 7.96 2 6.67 2 6.50 2 6.46 2 6.15 2 6.17 2 7.14 2 6.86 2 6.70 2 6.59 2 6.04 2 6.06 2 7.17 2 6.73 2 6.19 2 6.78 2 6.17 2 6.12 2 6.98 4 6.33 4 5.87 4 6.66 4 5.81 4 5.00 4 6.65 4 6.65 4 5.77 4 6.41 4 5.75 4 5.21 4 6.57 4 6.24 4 5.72 4 5.91 4 5.63 4 5.31 4 6.66 6 6.12 6 5.47 6 5.67 6 5.13 6 4.18 6 6.11 6 6.38 6 5.45 6 5.82 6 4.71 6 4.18 6 6.30 6 6.35 6 - 6 5.84 6 5.12 6 3.95 6 6.37 8 5.95 8 4.92 8 5.44 8 4.08 8 3.38 8 5.85 8 5.55 8 5.12 8 5.49 8 4.5 0 8 3.15 8 5.77 8 5.42 8 4.82 8 5.42 8 4.11 8 3.43 8 5.47 10 5.11 10 4.58 10 5.15 10 n/a 10 2.53 10 - 10 5.26 10 4.10 10 4.30 10 3.82 10 2.29 10 4.58 10 5.50 10 4.13 10 5.50 10 - 10 2.41 10 4.92 12 4.04 12 3.61 12 4.43 12 - 12 - 12 3.68 12 4.41 12 3.35 12 4.78 12 - 12 - 12 4.04 12 4.53 12 - 12 4.77 12 - 12 - 12 3.88 14 2.58 14 3.38 14 - 14 2.68 14 - 14 2.54 148 Table H. 6 Raw survival data for E. faecium NRRL B - 2354 in date paste (0. 6 5 a w ) at 80 °C generated by MSU and UGA. E. faecium - 80°C MSU UGA Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 6.52 0 5.6 0 0 5.90 0 7.37 0 7.32 0 7.63 0 5.95 0 5.75 0 6.19 0 6.47 0 7.33 0 7.68 0 6.02 0 5.52 0 6.42 0 6.26 0 7.41 0 7.26 1 5.82 0. 8 5.79 0. 8 6.5 0 0. 8 5.83 0. 8 6.45 0. 8 6.89 1 6.15 0. 8 4.99 0. 8 6.07 0. 8 6.47 0. 8 6.64 0. 8 7.16 1 5.66 0. 8 4.89 0. 8 5.78 0. 8 6.01 0. 8 6.68 0. 8 7.24 2 5.56 1.5 4.4 0 1.5 5.61 1.5 5.37 1.5 5.13 1.5 6.65 2 5.47 1.5 4.52 1.5 5.9 0 1.5 4.99 1.5 5.8 0 1.5 6.08 2 4.89 1.5 4.49 1.5 6.18 1.5 5.63 1.5 5.29 1.5 6.19 3 4.75 2. 3 3.89 2. 3 5.15 2. 3 2. 3 4.43 2. 3 5.69 3 4.42 2. 3 3.83 2. 3 5.42 2. 3 4.71 2. 3 n/a 2. 3 5.99 3 4.46 2. 3 3.94 2. 3 5.03 2. 3 4.73 2. 3 4.53 2. 3 6.24 4 3.58 3 - 3 4.32 3 4.09 3 4.1 3 5.53 4 3.66 3 - 3 3.86 3 4.4 3 4.16 3 5.56 4 4.54 3 3.57 3 4.66 3 3.44 3 4.16 3 6.13 3. 8 - 3. 8 4.41 3. 8 4.58 3. 8 2.71 3. 8 - 3. 8 - 3. 8 - 3. 8 - 3. 8 - 3. 8 4.96 3. 8 2.46 3. 8 4.99 3. 8 - 3. 8 - 3. 8 - 149 APPENDIX I . Isothermal inactivation data for skim milk powder (chapter 5) Table I. 1 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in skim milk powder (0.25 a w ) at 85°C. 85°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.55 0 7.15 0 8.17 0 8.22 0 8.69 0 8.72 0 - 0 7.59 0 8.00 0 8.18 0 8.38 0 8.48 0 - 0 7.87 0 7.84 0 8.3 0 0 8.24 0 9.36 12 6.6 12 7.00 12 7.50 10 8.57 15 8.11 15 8.06 12 6.47 12 6.77 12 7.47 10 8.2 15 8.04 15 8.22 12 6.49 12 8.15 12 7.47 10 8.32 15 8.23 15 8.19 24 6.08 24 6.22 24 7.23 20 8.07 30 7.42 30 8.10 24 6.11 24 6.25 24 6.97 20 7.6 30 7.10 30 7.73 24 6.1 0 24 - 24 7.16 20 7.83 30 7.04 30 7.88 36 5.22 36 5.37 36 6.82 30 7.76 45 3.82 45 8.03 36 5.47 36 5.66 36 6.99 30 7.76 45 - 45 7.88 36 5.37 36 - 36 6.83 30 7.79 45 3.73 45 7.95 48 4.12 48 5.24 48 4.93 40 7.2 60 - 60 7.57 48 4.2 0 48 5.87 48 5.31 40 7.13 60 - 60 7.73 60 - 48 5.84 48 5.98 40 6.8 60 - 60 8.15 60 - 60 5.45 60 4.95 50 6.49 75 - 75 7.28 60 - 60 5.23 60 4.27 50 5.84 75 - 75 7.13 72 - 60 5.31 60 5.55 50 6.13 75 - 75 7.40 72 - 72 3.97 72 4.08 72 - 72 - 72 5.19 72 - 72 4.04 150 Table I. 2 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in skim milk powder (0.25 a w ) at 90 °C. 90°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 6.9 0 7.45 0 7.14 0 8.31 0 - 0 8.06 0 6.7 0 7.58 0 7.19 0 - 0 8.4 0 0 8.22 0 6.78 0 7.25 0 8.09 0 8.05 0 - 0 8.09 6 6.85 6 7.36 6 7.06 12 8.08 12 7.47 12 7.72 6 6.13 6 6.91 6 6.91 12 8.35 12 7.8 12 7.83 6 6.18 6 7.21 6 7.33 12 8.37 12 7.73 12 8.05 12 5.47 12 6.64 12 - 24 - 24 6.2 0 24 7.65 12 5.93 12 6.06 12 6.38 24 - 24 - 24 7.74 12 - 12 6.27 12 6.64 24 - 24 6.49 24 7.58 18 5.56 18 5.89 18 6.15 36 7.3 1 36 - 36 - 18 5.56 18 5.58 18 5.56 36 7.18 36 4.47 36 6.81 18 6.24 18 5.74 18 6.10 36 7.28 36 - 36 6.83 24 3.98 24 6.31 24 5.16 48 6.39 48 - 48 5.91 24 3.95 24 - 24 5.48 48 6.4 48 - 48 5.7 24 3.72 24 - 24 5.78 48 5.94 48 - 48 5.55 30 - 30 5.56 30 4.61 60 4.98 60 - 60 3.71 30 - 30 3.83 30 4.17 60 4.01 60 - 60 3.86 30 - 30 4.77 30 4.74 60 4.02 60 - 60 3.68 151 Table I. 3 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in skim milk powder (0.25 a w ) at 95 °C. 85°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 6.49 0 7.39 0 6.88 0 8.38 0 7.74 0 8.07 0 6.52 0 6.69 0 6.69 0 8.24 0 8.14 0 8.36 0 - 0 7.14 0 - 0 8.53 0 8.28 0 8.26 3 5.78 3 7.09 3 6.18 6 7.62 6 7.03 6 7.58 3 5.91 3 7.57 3 6.23 6 7.99 6 7.14 6 7.63 3 6.23 3 7.26 3 6.17 6 7.96 6 - 6 7.44 6 5.28 6 5.68 6 6.63 12 7.42 12 5.42 12 7.54 6 5.17 6 7.17 6 5.87 12 7.28 12 5.5 0 12 7.16 6 4.82 6 5.92 6 5.78 12 7.63 12 5.9 0 12 7.77 9 3.95 9 5.29 9 5.49 18 6.77 18 - 18 5.73 9 4.21 9 5.19 9 5.10 18 6.73 18 - 18 6.39 9 4.09 9 5.08 9 5.36 18 6.23 18 - 18 6.56 12 - 12 4.12 12 - 24 5.78 24 - 24 5.59 12 - 12 4.75 12 4.70 24 4.76 24 - 24 4.74 12 - 12 - 12 4.61 24 - 24 - 24 4.83 15 - 15 4.42 15 4.28 30 3.49 30 - 30 3.47 15 - 15 - 15 4.75 30 - 30 - 30 3.17 15 - 15 4.74 15 - 30 3.62 30 - 30 3.35 18 3.06 18 3.29 18 - 18 4.96 18 - 18 - 152 APPENDIX J . Isothermal inactivation data for lactose - free skim milk powder (chapter 5) Table J. 1 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose - free skim milk powder (0.25 a w ) at 65 °C. 65°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.98 0 7.16 0 8.39 0 8.26 0 8.7 0 0 8.88 0 7.17 0 7.80 0 6.87 0 8.68 0 8.58 0 8.83 0 7.79 0 6.33 0 7.35 0 - 0 8.56 0 8.51 15 7.28 20 6.78 20 7.33 20 - 20 8.30 20 8.87 15 7.41 20 6.78 20 6.85 20 - 20 - 20 8.88 15 - 20 7.01 20 7.23 20 - 20 - 20 - 30 6.54 40 - 40 6.10 40 - 40 8.16 40 8.10 30 6.13 40 6.03 40 4.99 40 8.10 40 8.30 40 - 30 6.66 40 6.06 40 6.77 40 8.27 40 8.26 40 - 45 5.40 60 5.66 60 5.39 60 7.29 60 7.23 60 - 45 7.17 60 5.60 60 5.34 60 6.93 60 7.05 60 6.93 45 6.67 60 5.58 60 5.67 60 6.89 60 7.15 60 7.52 60 6.29 80 5.24 80 5.01 80 5.78 80 6.00 80 5.72 60 5.82 80 5.86 80 5.59 80 5.78 80 6.27 80 5.66 60 5.64 80 5.38 80 5.01 80 5.47 80 5.87 80 - 75 5.20 100 4.93 100 4.63 100 4.97 100 5.17 100 5.36 75 5.65 100 4.87 100 4.55 100 5.11 100 5.38 100 4.92 75 5.85 100 4.66 100 4.93 100 5.20 100 5.50 100 6.03 90 5.26 120 4.07 120 4.00 120 4.58 120 4.94 120 4.83 90 5.30 120 5.02 120 3.81 120 4.69 120 4.56 120 4.45 90 5.26 120 4.96 120 4.20 120 4.89 120 4.65 120 - 153 Table J. 2 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose - free skim milk powder (0.25 a w ) at 70 °C. 70°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.60 0 7.71 0 7.72 0 8.88 0 8.74 0 8.68 0 7.46 0 7.69 0 - 0 8.18 0 8.32 0 8.38 0 - 0 7.35 0 7.46 0 8.55 0 7.96 0 8.17 8 7.53 10 6.64 10 6.61 12 - 10 - 10 - 8 7.61 10 5.65 10 6.87 12 - 10 8.37 10 8.24 8 7.06 10 6.67 10 - 12 8.13 10 8.39 10 8.53 16 5.93 20 5.85 20 6.09 24 7.78 20 7.68 20 7.94 16 6.56 20 5.82 20 5.76 24 7.76 20 7.78 20 8.12 16 6.50 20 6.35 20 6.18 24 7.81 20 7.79 20 7.86 24 5.66 30 5.89 30 5.56 36 5.56 30 6.43 30 5.65 24 5.71 30 5.78 30 5.6 0 36 - 30 6.25 30 5.74 24 5.8 0 30 5.73 30 5.13 36 5.76 30 6.38 30 6.17 32 6.08 40 5.49 40 5.11 48 4.81 40 5.66 40 5.17 32 5.09 40 4.96 40 5.00 48 5.28 40 5.59 40 5.01 32 - 40 - 40 4.82 48 5.00 40 5.66 40 4.79 40 5.04 50 3.74 50 4.42 60 4.29 50 5.16 50 4.79 40 5.04 50 4.6 50 4.17 60 4.43 50 4.96 50 4.22 40 5.09 50 3.94 50 4.01 60 3.89 50 5.05 50 4.63 48 5.09 60 3.63 60 3.71 72 3.93 60 4.13 60 4.07 48 4.79 60 4.07 60 4.96 72 3.96 60 4.16 60 4.03 48 - 60 3.94 60 3.34 72 4.01 60 4.49 60 3.68 154 Table J. 3 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose - free skim milk powder (0.25 a w ) at 75 °C. 75°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.57 0 8.06 0 8.03 0 8.33 0 8.50 0 8.04 0 7.40 0 7.78 0 7.81 0 8.62 0 - 0 8.64 0 7.58 0 7.87 0 8.18 0 8.3 0 - 0 8.70 5 6.5 0 5 6.55 5 6.37 5 8.11 5 - 5 8.40 5 6.93 5 6.75 5 7.02 5 8.07 5 - 5 8.09 5 5.93 5 6.50 5 6.74 5 8.25 5 8.02 5 7.81 10 5.75 10 6.48 10 5.84 10 7.73 10 7.60 10 8.01 10 5.56 10 6.54 10 5.64 10 7.17 10 7.80 10 7.79 10 5.70 10 6.13 10 5.73 10 - 10 7.74 10 8.01 15 - 15 5.60 15 4.93 15 5.31 15 5.66 15 - 15 4.83 15 5.68 15 4.95 15 5.07 15 5.67 15 - 15 5.42 15 5.67 15 4.72 15 4.93 15 5.80 15 5.66 20 3.92 20 4.85 20 4.68 20 5.2 20 5.73 20 4.55 20 - 20 4.76 20 4.55 20 5.13 20 5.52 20 4.67 20 4.18 20 5.50 20 4.69 20 5.48 20 4.96 20 4.85 25 3.52 25 5.07 25 3.65 25 3.65 25 - 25 4.60 25 3.53 25 4.93 25 4.11 25 3.84 25 4.12 25 3.72 25 - 25 4.48 25 4.52 25 3.99 25 3.65 25 4.08 30 2.54 30 - 30 2.86 30 - 30 3.52 30 3.59 30 2.84 30 4.22 30 2.80 30 3.58 30 3.62 30 2.98 30 3.14 30 4.22 30 3.01 30 3.74 30 3.44 30 3.75 155 APPENDIX K . Isothermal inactivation data for lactose powder (chapter 5) Table K. 1 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose powder (0.25 a w ) at 85 °C. 8 5°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.20 0 6.92 0 6.50 0 7.96 0 8.63 0 7.8 0 0 6.94 0 6.29 0 - 0 8.02 0 8.42 0 7.57 0 6.94 0 6.27 0 6.81 0 7.85 0 8.24 0 7.7 2 7.03 2 6.14 4 7.00 8 8.26 6 8.41 7 7.58 2 6.93 2 5.58 4 - 8 7.68 6 6.93 7 7.18 2 - 2 5.8 4 5.75 8 7.66 6 6.99 7 - 4 5.95 4 5.84 8 6.67 16 - 12 7.29 14 - 4 - 4 6.97 8 4.86 16 - 12 - 14 6.77 4 - 4 5.86 8 5.22 16 - 12 - 14 6.26 6 6.10 6 5.96 10 - 24 6.92 18 6.5 0 21 6.76 6 5.82 6 6.02 10 5.54 24 - 18 5.91 21 6.66 6 5.41 6 4.61 10 - 24 - 18 - 21 6.85 8 3.86 8 4.74 12 5.32 32 6.02 24 - 28 - 8 - 8 4.1 12 - 32 - 24 - 28 - 8 5.10 8 3.99 12 4.02 32 - 24 - 28 - 10 - 10 4.91 14 - 40 5.7 30 - 35 3.43 10 4.11 10 - 14 - 40 30 - 35 5.64 10 - 10 - 14 - 40 4.93 30 - 35 3.98 156 Table K. 2 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose powder (0.25 a w ) at 90 °C. 90°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 6.70 0 6.24 0 6.53 0 7.79 0 8.2 0 0 7.58 0 7.08 0 6.92 0 6.65 0 8.12 0 8.29 0 7.5 0 0 6.76 0 6.66 0 6.7 0 0 - 0 - 0 7.87 1 6.43 1 - 1 6.38 4 7.82 3 8.25 4 7.08 1 6.44 1 6.77 1 6.09 4 7.7 3 - 4 7.03 1 - 1 5.97 1 5.52 4 7.25 3 8.28 4 7.3 0 2 5.86 2 6.19 2 5.98 8 - 6 7.06 8 - 2 6.52 2 5.86 2 5.5 8 - 6 7.37 8 6.98 2 6.67 2 - 2 4.73 8 6.85 6 - 8 7.04 3 5.92 3 5.89 3 4.84 12 6.72 9 6.94 12 6.83 3 6.12 3 5.28 3 4.63 12 4.86 9 - 12 6.62 3 6.02 3 6.37 3 5.57 12 - 9 6.71 12 6.76 4 - 4 6.09 4 5.67 16 - 12 - 16 5.4 0 4 - 4 5.49 4 - 16 - 12 5.82 16 5.62 4 5.45 4 4.96 4 5.29 16 - 12 - 16 - 5 4.95 5 3.58 5 - 20 - 15 6.25 20 5.54 5 4.86 5 5.93 5 4.78 20 - 15 20 5.81 5 5.41 5 5.23 5 2.81 6 - 6 - 6 - 6 - 6 - 6 - 6 - 6 - 6 - 7 - 7 - 7 - 7 - 7 - 7 - 7 - 7 - 7 - 157 Table K. 3 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in lactose powder (0.25 a w ) at 95 °C. 95°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 6.49 0 5.06 0 6.1 0 0 7.09 0 - 0 7.95 0 6.17 0 5.81 0 4.72 0 7.22 0 6.66 0 7.57 0 6.27 0 6.84 0 6.54 0 8.35 0 8.29 0 - 0.5 6.51 0.5 6.57 0.5 5.72 2 7.13 1.5 8.16 1.5 7.22 0.5 6.13 0.5 6.48 0.5 6.76 2 6.79 1.5 7.55 1.5 7.66 0.5 - 0.5 - 0.5 5.74 2 7.51 1.5 7.4 0 1.5 7.87 1 6.05 1 6.48 1 - 4 6.88 3 - 3 7.05 1 5.94 1 4.57 1 5.19 4 6.71 3 4.64 3 - 1 5.85 1 4.78 1 5.77 4 6.68 3 - 3 - 1.5 6.07 1.5 4.59 1.5 - 6 6.88 4.5 5.23 4.5 6.87 1.5 5.83 1.5 5.83 1.5 5.52 6 - 4.5 4.7 0 4.5 - 1.5 4.45 1.5 - 1.5 4.58 6 - 4.5 - 4.5 - 2 5.6 0 2 6.39 2 - 8 5.53 6 - 6 - 2 - 2 4.07 2 4.4 0 8 5.3 0 6 - 6 - 2 - 2 - 2 3.96 8 4.71 6 - 6 - 2.5 5.19 2.5 5.37 2.5 5.12 10 2.44 7.5 - 7.5 5.96 2.5 4.09 2.5 - 2.5 - 10 4.95 7.5 - 7.5 - 2.5 4.74 2.5 5.58 2.5 3.37 10 - 7.5 - 7.5 - 3 - 3 - 3 - 12 - 9 - 9 - 3 - 3 - 3 - 12 - 9 - 9 - 3 - 3 - 3 - 12 - 9 - 9 - 158 APPENDIX L . Isothermal inactivation data for milk protein isolate 90% powder (chapter 5) Table L. 1 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in milk protein isolate 90% powder (0.25 a w ) at 80 °C. 80°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.28 0 7.98 0 7.78 0 8.25 0 7.91 0 7.25 0 7.73 0 7.63 0 7.89 0 7.16 0 8.14 0 7.39 0 8.13 0 8.42 0 7.87 0 7.62 0 8.13 0 7.46 24 7.28 24 7.9 0 24 7.12 25 7.73 25 8.31 25 7.29 24 7.12 24 8.22 24 8.37 25 7.14 25 7.72 25 7.72 24 7.6 0 24 8.04 24 7.77 25 - 25 7.89 25 7.36 48 6.01 48 7.07 48 7.1 0 50 6.40 50 7.63 50 6.86 48 6.83 48 7.41 48 7.24 50 7.27 50 6.78 50 6.74 48 6.75 48 7.15 48 6.53 50 6.67 50 7.21 50 7.02 72 5.1 0 72 - 72 6.16 75 7.37 75 6.86 75 5.68 72 5.57 72 6.39 72 n/a 75 - 75 6.34 75 6.22 72 5.41 72 6.40 72 6.79 75 6.96 75 6.92 75 n/a 96 5.51 96 5.39 96 - 100 5.52 100 6.60 100 6.36 96 - 96 6.90 96 5.72 100 5.27 100 6.76 100 6.53 96 6 .00 96 7.28 96 4.88 100 6.54 100 7.08 100 6.24 120 - 120 6.01 120 3.67 125 6.13 125 6.01 125 3.76 120 - 120 - 120 3.65 125 5.11 125 5.64 125 6.14 120 4.8 0 120 5.55 120 4.08 125 3.89 125 - 125 4.45 144 - 144 5.18 144 n/a 150 4.81 150 5.21 150 4.59 144 - 144 3.84 144 n/a 150 4.59 150 4.67 150 5.28 144 - 144 5.61 144 4.63 150 3.71 150 5.74 150 n/a 159 Table L. 2 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in milk protein isolate 90% powder (0.25 a w ) at 85 °C. 85°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 8.70 0 7.14 0 7.49 0 7.48 0 8.10 0 7.68 0 7.90 0 7.51 0 7.72 0 7.4 0 7.45 0 7.64 0 7.53 0 7.06 0 8.75 0 7.98 0 8.22 0 7.66 12 7.65 12 - 12 6.74 15 - 15 - 15 7.67 12 7.55 12 7.23 12 6.58 15 7.01 15 7.96 15 7.06 12 8.07 12 6.56 12 7.05 15 7.05 15 7.05 15 7.13 24 7.52 24 6.8 24 6.60 30 6.03 30 7.86 30 7.25 24 6.22 24 6.75 24 6.73 30 6.38 30 6.91 30 6.48 24 6.80 24 5.81 24 7.78 30 6.07 30 n/a 30 6.60 36 6.98 36 6.18 36 7.33 45 - 45 5.84 45 6.94 36 7.38 36 6.25 36 6.53 45 7.11 45 5.67 45 6.84 36 6.78 36 5.42 36 7.19 45 5.72 45 6.07 45 6.58 48 6.10 48 4.99 48 5.19 60 6.87 60 6.39 60 4.87 48 4.80 48 5.03 48 6.15 60 5.66 60 6.27 60 6.20 48 5.05 48 4.78 48 5.62 60 6.53 60 5.46 60 5.48 60 5.00 60 4.75 60 4.93 75 4.72 75 4.81 75 5.29 60 4.92 60 3.94 60 5.25 75 - 75 3.59 75 5.51 60 5.80 60 - 60 3.76 75 5.13 75 4.85 75 6.24 72 5.34 72 - 72 4.36 90 3.05 90 3.08 90 5.46 72 - 72 3.54 72 3.44 90 2.62 90 3.58 90 - 72 - 72 n/a 72 3.59 90 3.85 90 3.79 90 4.90 160 Table L. 3 Raw survival data for Salmonella and E. faecium NRRL B - 2354 in milk protein isolate 90% powder (0.25 a w ) at 90 °C. 90°C Salmonella E. faecium Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) Time (min) Rep 1 (Log CFU/g) Time (min) Rep 2 (Log CFU/g) Time (min) Rep 3 (Log CFU/g) 0 7.93 0 7.81 0 7.56 0 7.87 0 7.53 0 7.46 0 7.85 0 8.22 0 8.57 0 6.30 0 7.54 0 7.71 0 7.26 0 8.04 0 8.12 0 8.20 0 7.76 0 7.94 6 7.40 6 7.41 6 7.22 6 7.10 6 7.71 6 6.65 6 7.06 6 7.12 6 7.30 6 7.16 6 6.18 6 7.49 6 7.02 6 7.06 6 6.67 6 7.06 6 7.31 6 6.64 12 7.51 12 6.63 12 6.48 12 6.26 12 6.67 12 7.14 12 6.68 12 6.70 12 6.79 12 6.64 12 7.11 12 6.31 12 6.39 12 6.17 12 6.73 12 7.97 12 7.04 12 6.10 18 6.39 18 4.90 18 5.70 18 6.46 18 5.00 18 6.64 18 6.30 18 5.57 18 6.50 18 6.59 18 4.64 18 6.98 18 6.29 18 n/a 18 6.32 18 4.71 18 n/a 18 6.63 24 5.21 24 4.93 24 6.18 24 5.07 24 4.87 24 6.72 24 5.11 24 6.59 24 4.90 24 4.84 24 5.61 24 n/a 24 5.77 24 5.56 24 n/a 24 n/a 24 6.31 24 7.13 30 4.17 30 3.92 30 4.27 30 4.65 30 5.33 30 5.97 30 4.08 30 3.98 30 n/a 30 n/a 30 5.83 30 4.97 30 3.93 30 n/a 30 n/a 30 5.52 30 4.98 30 n/a 36 2.91 36 n/a 36 4.29 36 4.64 36 n/a 36 5.21 36 3.76 36 n/a 36 3.87 36 4.90 36 5.08 36 5.73 36 4.19 36 n/a 36 n/a 36 3.81 36 3.78 36 3.52 161 APPENDIX M . Homogeneity data (Chapter 3) Table M. 1 Enterococcus faecium NRRL B - 2354 population on whole almond and almond meal during wet - inoculation. Rep Log CFU/g Whole almond Almond meal 1 7.85 7.38 1 8.08 7.39 1 7.9 7.45 Mean 7.94 7.41 S tdev 0.12 0.04 2 7.95 7.63 2 8.23 7.69 2 8.35 7.54 Mean 8.18 7.62 Stdev 0.21 0.08 3 7.98 7.65 3 8.15 7.83 3 8.01 7.78 Mean 8.05 7.75 Stdev 0.09 0.09 162 Table M. 2 Enterococcus faecium NRRL B - 2354 population on whole almond and almond meal during dry - inoculation. Rep Log CFU/g Talc powder Whole almond Almond meal 1 8.51 4.84 4.94 1 8.40 5.57 4.92 1 8.53 5.41 4.98 Mean 8.48 5.27 4.95 Stdev 0.07 0.38 0.03 2 8.3 5.49 5.33 2 7.95 5.37 5.23 2 8.24 5.51 5.2 Mean 8.16 5.46 5.25 Stdev 0.19 0.08 0.07 3 8.07 5.66 5.38 3 8.27 5.74 5.41 3 8.3 5.83 5.38 Mean 8.21 5.74 5.39 Stdev 0.13 0.09 0.02 163 Table M. 3 Enterococcus faecium NRRL B - 2354 population on whole almond and almond meal during wet - talc - inoculation. Table M. 4 Enterococcus faecium NRRL B - 2354 population on talc powder alone. Rep Log CFU/g Whole almond Sieved talc powder Almond meal 1 8.51 5.09 7.87 1 8.14 5.1 0 7.79 1 7.57 5.19 7.96 Mean 8.07 5.13 7.87 Stdev 0.47 0.06 0.09 2 8.37 5.87 7.84 2 7.47 7.23 7.79 2 8.11 5.4 0 7.7 Mean 7.98 6.17 7.78 Stdev 0.46 0.95 0.07 3 7.81 5.64 7.84 3 8.15 5.46 7.79 3 7.95 5.72 7.7 Mean 7.97 5.61 7.78 Stdev 0.17 0.13 0.07 Rep Log CFU/g Talc powder 1 8.15 1 7.96 1 8.13 Mean 8.08 Stdev 0.10 2 7.77 2 8.11 2 n/a Mean 7.94 Stdev 0.24 3 7.99 3 7.43 3 7.61 Mean 7.68 Stdev 0.29 164 APPENDIX N . Homogeneity data (chapter 4) Table N. 1 Salmonella and E. faecium population in almond meal, wheat flour, ground black pepper, nonfat dry milk powder, peanut butter, and date paste. Almond meal (0.45 a w ) Wheat flour (0.45 a w ) Sample Salmonella (log CFU/g) E. faecium (log CFU/g) Salmonella (log CFU/g) E. faecium (log CFU/g) Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 1 8.60 8.12 8.02 7.53 8.00 8.04 8.23 8.44 8.71 8.02 8.07 8.28 2 8.61 8.18 7.91 7.69 7.93 8.09 8.51 8.48 8.56 7.96 8.09 8.15 3 8.25 8.27 7.88 8.43 8.22 7.94 8.26 8.52 8.87 8.00 8.15 8.07 4 8.47 8.18 7.91 7.72 8.15 8.01 8.06 8.47 8.80 8.06 8.07 8.02 5 8.72 8.24 8.00 7.94 8.33 8.00 8.17 8.67 8.65 8.10 8.07 8.09 6 8.51 8.26 8.12 7.69 8.00 8.04 8.18 8.61 8.67 8.33 8.07 8.31 7 8.51 8.49 8.07 7.87 7.85 7.92 7.98 8.61 8.73 7.93 8.04 8.28 8 8.51 8.56 7.97 7.80 8.34 8.02 8.16 8.69 8.64 7.96 8.01 8.20 9 8.15 8.54 7.99 7.65 7.97 8.06 8.37 8.78 8.55 7.98 7.96 8.25 10 8.58 8.54 8.27 7.77 8.03 8.02 8.18 8.68 8.61 7.93 8.02 8.14 Mean 8.49 8.34 8.01 7.81 8.08 8.01 8.21 8.59 8.68 8.03 8.06 8.18 Stdev 0.17 0.17 0.12 0.24 0.17 0.05 0.15 0.11 0.10 0.12 0.05 0.10 Ground black pepper (0.45 a w ) Nonfat dry milk powder (0.25 a w ) Sample Salmonella (log CFU/g) E. faecium (log CFU/g) Salmonella (log CFU/g) E. faecium (log CFU/g) Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 1 6.94 6.60 6.86 7.64 7.35 7.41 7.84 8.19 7.88 7.98 7.99 7.53 2 6.87 6.73 6.67 7.36 7.12 7.26 7.93 8.18 7.98 7.99 8.10 7.69 3 6.93 6.53 6.76 7.41 7.08 7.35 7.93 8.14 7.95 7.90 8.04 7.71 4 6.96 6.74 6.53 7.55 6.89 7.46 7.96 8.14 7.90 7.94 8.14 7.67 5 6.93 6.84 6.51 7.45 7.13 7.34 7.97 7.96 7.96 7.95 8.08 7.71 6 6.92 6.68 6.62 7.48 6.86 7.13 7.94 8.31 7.94 7.98 8.08 7.58 7 6.80 6.58 6.83 7.38 7.34 7.28 7.85 8.08 7.89 7.96 8.07 7.67 8 6.80 6.58 6.83 7.41 7.21 7.53 7.97 8.00 7.92 7.99 8.08 7.71 9 6.97 6.75 6.59 7.41 7.21 7.53 7.88 8.04 7.95 7.91 8.12 7.82 10 6.97 6.75 6.59 7.45 7.37 7.10 7.95 8.19 7.99 7.96 8.05 7.74 Mean 6.91 6.68 6.68 7.46 7.16 7.34 7.92 8.12 7.94 7.96 8.07 7.68 Stdev 0.06 0.10 0.13 0.08 0.18 0.15 0.05 0.11 0.04 0.03 0.04 0.08 165 Table N .1 Peanut butter (0.25 a w ) Sample Salmonella (log CFU/g) E. faecium (log CFU/g) Rep 1 Rep 2 Rep 3 Rep 4 Rep 1 Rep 2 Rep 3 Rep 4 1 8.18 8.05 8.35 8.29 8.11 8.30 8.22 8.34 2 8.13 8.07 8.24 8.51 8.06 8.32 8.40 8.27 3 8.03 8.07 8.30 8.19 8.08 8.52 8.26 8.32 4 8.11 8.11 8.29 8.29 8.10 8.46 8.47 8.17 5 8.12 8.17 8.45 8.14 7.86 8.20 8.28 8.42 6 8.10 8.13 8.45 8.17 7.92 8.31 8.16 8.14 7 8.18 8.21 8.44 8.22 8.00 8.25 8.22 8.34 8 8.06 8.19 8.33 8.27 8.23 8.39 8.37 8.32 9 8.14 8.15 8.51 8.23 8.01 8.54 8.33 8.29 10 8.02 8.10 8.46 8.08 8.02 8.39 8.38 8.35 Mean 8.11 8.13 8.38 8.24 8.04 8.37 8.31 8.30 Stdev 0.05 0.05 0.09 0.11 0.10 0.11 0.10 0.08 Date paste (0.65 a w ) Sample Salmonella (log CFU/g) E. faecium (log CFU/g) Rep 1 Rep 2 Rep 3 Rep 4 Rep 1 Rep 2 Rep 3 1 8.16 7.99 7.92 8.21 n/a n/a 8.19 2 8.15 8.04 8.12 8.34 8.42 8.23 8.26 3 8.02 7.87 n/a 8.49 8.29 n/a 8.30 4 8.37 7.98 8.08 8.54 n/a 8.27 8.23 5 8.16 7.98 8.14 8.28 8.45 8.32 8.26 6 8.25 7.95 7.98 n/a 8.42 8.24 8.18 7 8.07 7.96 7.83 8.57 n/a 8.37 8.23 8 8.02 8.05 7.90 8.25 8.43 8.21 8.30 9 8.17 8.25 7.87 8.43 8.36 8.20 8.18 10 8.28 7.93 n/a 8.31 8.29 8.09 8.30 Mean 8.16 8.00 7.98 8.38 8.38 8.24 8.24 Stdev 0.11 0.10 0.12 0.13 0.07 0.08 0.05 166 APPENDIX O . Homogeneity data (chapter 5) Table O. 1 Homogeneity of Salmonella and E. faecium population in dairy powders at 0.25 a w . Skim milk powder Lactose - free skim milk powder Sample Salmonella (log CFU/g) E. faecium (log CFU/g) Salmonella (log CFU/g) E. faecium (log CFU/g) Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 1 8.08 7.58 8.08 8.80 8.28 8.24 7.84 7.89 8.10 8.58 8.74 8.35 2 7.75 7.93 7.55 8.29 8.52 8.51 7.87 7.52 7.82 8.54 8.37 8.32 3 8.04 7.75 7.43 8.46 8.64 8.51 7.85 7.69 8.08 8.13 8.61 8.61 Mean 7.96 7.75 7.69 8.52 8.48 8.42 7.85 7.70 8.00 8.42 8.57 8.43 Stdev 0.18 0.18 0.34 0.26 0.18 0.16 0.02 0.18 0.16 0.25 0.19 0.16 Lactose powder Milk protein isolate 90% Sample Salmonella (log CFU/g) E. faecium (log CFU/g) Salmonella (log CFU/g) E. faecium (log CFU/g) Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3 1 7.76 7.79 7.77 8.30 8.54 7.95 8.07 8.53 8.22 8.00 8.03 7.96 2 7.70 7.90 7.73 8.48 8.27 8.10 8.05 8.83 8.36 7.32 8.41 7.83 3 7.47 7.80 7.57 8.30 8.50 8.08 8.30 8.40 - 7.54 8.06 7.61 Mean 7.64 7.83 7.69 8.36 8.44 8.05 8.45 8.58 8.29 7.62 8.17 7.80 Stdev 0.16 0.06 0.11 0.10 0.15 0.08 0.14 0.22 0.10 0.35 0.21 0.18 167 APPENDIX P . Come - up time for almond meal and talc (Chapter 3) This APPENDIX shows six measurements of come - up time for non - inoculated almond meal and talc powder at 0.45 a w to achieve 0.5 °C below set temperature of 80 °C . Table P. 1 Come - up time (CUT) for talc powder and almond meal at 0.45 a w . Rep Time (s) Talc powder Almond meal 1 72 110 2 100 102 3 114 102 4 128 100 5 172 100 6 134 120 Mean 120 112 Stdev 33.8 5 Mean + 2 Stdev 187.6 122 168 APPENDIX Q . Come - up time (chapter 4) Table Q. 1 Come - up time (CUT) for peanut butter, almond meal, wheat flour, ground black pepper, nonfat dry milk powder, and date paste. Time (s) Peanut butter (0.25 a w ) Nonfat dry milk powder (0.25 a w ) Rep IFSH UNL MSU WSU Temperature (°C) Temperature (°C) Temperature (°C) Temperature (°C) 85 90 95 100 85 90 95 100 85 90 95 85 90 95 1 258 248 257 265 164 173 178 - 94 186 202 - - - 2 274 226 286 250 156 188 172 - 96 182 210 - - - 3 276 244 265 308 182 181 173 - 96 188 174 - - - 4 260 250 271 272 164 180 182 - 86 150 234 - - - 5 299 229 275 308 173 189 192 - 74 174 160 - - - 6 241 249 275 256 188 189 176 - 76 190 248 - - - Mean 268 241 272 277 171 183 179 - 87 178 205 - - - Stdev 20 11 10 26 12.1 6.5 7.4 - 10 15 33.8 - - - Mean + 2 stdev 308 262 291 328 195 196 194 - 107 208 272 - - - Time (s) Rep Ground black pepper (0.45 a w ) Almond meal (0.45 a w ) IFSH UNL MSU WSU* Temperature (°C) Temperature (°C) Temperature (°C) Temperature (°C) 65 70 75 80 65 70 75 80 80 85 90 80 85 90 1 228 202 198 - 126 133 118 139 118 118 144 - - - 2 218 210 210 - 94 105 113 127 110 118 128 - - - 3 211 216 209 - 73 115 106 128 110 116 118 - - - 4 218 207 235 - 112 95 98 152 108 112 118 - - - 5 211 210 223 - 113 94 111 143 108 108 112 - - - 6 208 220 209 - 89 168 121 127 120 116 124 - - - Mean 216 211 214 - 101 118 111 136 112 115 124 - - - Stdev 7.3 6.4 13 - 19.3 28.3 8.3 10.4 5.3 3.9 11.2 - - - Mean + 2 stdev 230 224 240 - 140 175 128 157 123 123 146 120 120 120 169 Wheat flour (0. 4 5 a w ) Rep IFSH ** WSU * Temperature (°C) Temperature (°C) 70 75 80 85 70 75 80 85 1 96 108 112 - - - - - 2 163 121 108 - - - - - 3 135 156 - - - - - - 4 97 345 154 - - - - - 5 - 103 106 174 - - - - 6 - 120 120 120 - - - - Mean - - - - - - - - Stdev - - - - - - - - Mean + 2 stdev 180 180 180 180 Rep Date paste (0. 6 5 a w ) MSU UNL Temperature (°C) Temperature (°C) 65 70 75 80 65 70 75 80 1 290 206 168 196 18 5 133 135 13 8 2 180 170 178 202 17 4 9 6 100 92 3 110 174 180 186 264 8 4 97.9 9 7 4 284 152 164 186 267 10 9 11 9 10 6 5 186 180 164 188 10 5 122 156 144 6 346 180 198 182 102 91 9 5 103 7 - - - - 132 92 8 6 100 8 - - - - 122 11 6 10 7 113 9 - - - - 9 8 13 9 125 14 2 10 - - - - 9 5 9 7 9 4 92 11 - - - - 11 8 81 8 7 90 12 - - - - 11 4 1 20 10 9 110 Mean 210 176 171 192 14 8 10 7 109 11 1 Stdev 76.4 19.6 7.7 7.1 61.9 18.4 20.3 18.9 Mean+ 2 stdev 363 216 186 206 2 7 2 143 150 149 170 APPENDIX R . Come - up time (chapter 5) Table R. 1 Come - up time for dairy powders (0.25 a w ) at three level of temperatures. *Outlier was determined using 1.5 interquartil e analysis using Excel 2016 Time (s) Rep Skim milk powder Lactose - free skim milk powder 8 5 ° C 9 0 ° C 95 ° C 65 ° C 70 ° C 75 ° C 1 214 216 268 66 74 82 2 180 260 366* 56 52 96 3 156 120 240 58 52 114 4 140 134 300 66 80 68 5 150 152 424* 56 46 62 6 138 160 204 78 70 52 Mean 163 174 252 63 62 77 Stdev 29 54 32 9 14 22 Mean + 2 Stdev 221 281 316 81 90 121 Rep Lactose powder Milk protein isolate 90% 8 5 ° C 9 0 ° C 95 ° C 8 0 ° C 85 ° C 90 ° C 1 92 100 110 78 72 54 2 94 110 60 82 72 66 3 134 82 116 80 76 68 4 114 268* 134 62 58 70 5 154 158 124 78 82 122 6 68 132 66 84 70 158 Mean 109 116 102 77 72 89 Stdev 31 29 31 8 8 39 Mean + 2 Stdev 171 174 164 93 88 167 171 APPENDIX S . a w changes during heating at 80 ° C - example (chapter 4) Figure S. 1 Changes of a w during heating from room temperature to 80°C for nonfat dry milk powder indicated by the solid lines (n=2). Temperature is indicated by dotted line. 172 APPENDIX T. MATLAB codes for log - linear model (chapter 3) Housekeeping and data read clear; % Clear the workspace. close all ; % Close all figures. format compact clc % Select data in its corresponding range: data=xlsread( 'Data_Dry_Inoculation' , 'Log CFU for Matlab' ); % Col 6:time (min) % Col 5:Log N/No % Wet Inoculation: x1=data(1:45,6); y1=data(1:45,5); % Dry inoculation: x2=data(46:86,6); y2=data(46:86,5); % Wet - talc inoculation: x3=data(87:131,6); y3=data(87:131,5); % Talc: x4=data(132:178,6); y4=data(132:178,5); nlinfit beta0(1)=40; %Initial parameter guess for almond meal at 80C D - value in min % beta0(4)=20; % Initial guess for talc - close to wheat flour % f1 is a function for beta % J is sensitivity matrix % RMSE unit is the same as y and should be 10% less than ypred % sigma is standard err or of vector [beta1, resids1, J1, COVB1, mse1]=nlinfit(x1,y1,@f1,beta0); [beta2, resids2, J2, COVB2, mse2]=nlinfit(x2,y2,@f1,beta0); [beta3, resids3, J3, COVB3, mse3]=nlinfit(x3,y3,@f1,beta0); [beta4, resids4, J4, COVB4, mse4]=nlinfit(x4,y4,@f1,beta0); bet a1 beta2 beta3 beta4 rmse1=sqrt(mse1) rmse2=sqrt(mse2) rmse3=sqrt(mse3) rmse4=sqrt(mse4) [~, sigma1]=corrcov(COVB1) [~, sigma2]=corrcov(COVB2) [~, sigma3]=corrcov(COVB3) [~, sigma4]=corrcov(COVB4) 173 Confidence interval for parameter ci1=nlparci(beta1, resids1, 'jacobian' ,J1) ci2=nlparci(beta2, resids2, 'jacobian' ,J2) ci3=nlparci(beta3, resids3, 'jacobian' ,J3) ci4=nlparci(beta4, resids4, 'jacobian' ,J4) Confidence and prediction intervals for the dependent variable %nonlinear regression confidence intervals -- 'on' means simultaneous %bounds; 'off' is for nonsimultaneous bounds; must use 'curve' for %regression line, 'observation' for prediction interval [ypred1, delta1] = nlpredci(@f1,x1,beta1,resids1,J1,0.05, 'on' , 'curve' ); %confidence band for regression line [ypred1, deltaob1] =nlpredci(@f1,x1,beta1,resids1,J1,0.05, 'on' , 'observation' ); %prediction band for individual points [ypred2, delta2] = nlpredci(@f1,x2,beta2,resids2,J2,0.05, 'on' , 'curve' ); [ypred2, deltaob2]= nlpredci(@f1,x2,be ta2,resids2,J2,0.05, 'on' , 'observation' ); [ypred3, delta3] = nlpredci(@f1,x3,beta3,resids3,J3,0.05, 'on' , 'curve' ); [ypred3, deltaob3] =nlpredci(@f1,x3,beta3,resids3,J3,0.05, 'on' , 'observation' ); [ypred4, delta4] = nlpredci(@f1,x4,beta4,resids4,J4,0.05, 'on' , 'curve' ); [ypred4, deltaob4] =nlpredci(@f1,x4,beta4,resids4,J4,0.05, 'on' , 'observation' ) function [ y ] = f1(beta,t) %Function to calculate D - value at a single temperature using log - linear % model y= - t./beta(1); end 174 APPENDIX U. MATLAB codes for aggregated D - values of log - linear model (chapter 4 and 5) These codes were used to estimate aggregated D - values of isothermal inactivation data for chapter 4. Similar codes were used for chapter 5 with slight modifications in the data selection section. %Housekeeping clear; % Clear the workspace. close all ; % Close all figures. format compact clc % Lab 1 - IFSH, 2 - MSU, 3 - UNL, 4 - UGA, 5 - WSU % Product 1 - Peanut butter, 2 - Wheat flour, 3 - Black pepper, 4 - Almond meal, 5 - Date paste, 6 - Nonfat dried milk powder % Organism 1 - Salmonella cockta il, 2 - E.faecium % Batch code Replicates % Exp code Data set for specific temperature for specific replicate/batch % Time Time points % Temp Target treatment temperature % aw Target water activity % l pop Survivor in log CFU/g % Log N/N0 Log reduction (average T0, and deducted all lpop with obtained value) % Column Order = 1Lab|2Product|3Organism|4Batch|5Exp_code|6Time|7Temp|8aw|9lpop|10Log N/N0 results=[]; % empty matrix to o rganize results obtained ypredmatrix=[]; % need the ypred for isothermal curve plotting data=xlsread( 'Surrogate_Complete Raw Data All Products.xlsx' , 'Peanut butter Matlab' , 'A:J' ); % Purpose: To run 4 data sets, consist of IFSH - SAL, IFSH - EF, UNL - SAL, UNL - EF, at the same time labs=unique(data(:,1)); % Col 1 is lab IFSH/UNL [1,3] orgs=unique(data(:,3)); % Col 3 is organism [1,2] for i=1:length(labs); j=labs(i); for m=1:length(orgs); k=orgs(m); sec=find(data(:,3)==k & data(:,1)==j); temps=unique(data(sec,7)); % (sec,7) This line will find temp at specified data section for n=1:length(temps); g=temps(n); sec=find(data(:,3)==k & data(:,1)==j & data(:,7)==g); x=data(sec ,6); yobs=data(sec,10); % Log N/No 175 beta0=25; % aggregated % fnameINV appears only in the nlinfit and nlpredci statements. Everywhere % else, use fnameFOR [beta,resids,J,COVB,mse] = nlinfit(x,yobs,@f1,beta0); beta; rmse=sqrt(mse); [R,sigma] =corrcov(COVB); R; % R is the correlation matrix for the parameters sigma; % sigma is the standard error vector relerr=sigma./beta'; % relative error (coefficient of variance) for each parameter ss=resids'*resids; n=length( x); p=length(beta); condX=cond(J); %needs to be < 10^6. J is the sensitivity matrix = X detXTX=det(J'*J); %needs to be far from zero % Organizing Dref(SE), zT(SE), RMSE, Relative errors in a matrix: results=[results; j k g beta(1) sigma(1) rmse relerr(1)] function [ y ] = f1(beta,t) %Function to calculate D - value at a single temperature using log - linear % model y= - t./beta(1); end 176 APPENDIX V. MATLAB codes for global estimates of log - linear model and Bigelow model (chapter 4 ) %Housekeeping clear; % Clear the workspace. close all ; % Close all figures. format compact clc % Lab 1 - IFSH, 2 - MSU, 3 - UNL, 4 - UGA, 5 - WSU % Product 1 - Peanut butt er, 2 - Wheat flour, 3 - Black pepper, 4 - Almond meal, 5 - Date paste, 6 - Nonfat dried milk powder % Organism 1 - Salmonella cocktail, 2 - E.faecium % Batch code Replicates % Exp code Data set for specific temperature for specific replicate/batch % Temp Target treatment temperature % aw Target water activity % lpop Survivor in log CFU/g % Log N/N0 Log reduction (average T0, and deducted all lpop with obtained value) % Column Order = 1Lab|2Product|3Organism|4Batch|5Exp_code|6Time|7Temp|8aw|9lpop|10Log N/N0 results=[]; % empty matrix to organize results obtained ypredmatrix=[]; % need the ypred for isothermal curve plotting data=xlsread( 'Surrogate_Complete Raw Data All Products.xlsx' , 'Almond meal Matlab' , 'A:J' ); % Purpose: To run 4 data sets, consist of MSU - SAL, MSU - EF, WSU - SAL, WSU - EF, at the same time labs=unique(data(:,1)); % Col 1 is lab IFSH/UNL [2,5] orgs=unique(data(:,3)); % Col 3 is organism [1,2] for i=1:length(labs); j=labs(i); for m=1:length(orgs); k=orgs(m); sec=find(data(:,3)==k & data(:,1)==j); x=[data(sec,6), data(sec,7)]; % Col 6 time, Col 7 temp [x=time_temp matrix for nlinfit] yobs=data(sec,10); % Log N/No %beta(1) = Dref %beta(2) = z beta0(1)=25; %initial guess Dref for almond meal - Sal at 80C beta0(2)=16; %initial guess z global Tref ; Tref = 8 5 ; 177 nlinfit % ****IMPORTANT**** Usually, fnameFOR will be different from fnameINV % fnameINV=@forderexp_project; %Only in simple models can fnameFOR be the same as fnameINV % fnameINV appears only in the nlinfit and nlpredci statements. Everywhere % else, use fnameFOR [beta,resids,J,COVB,mse] = nlinfit(x,yobs,@forderexp_project,beta0); beta; rmse=sqrt(mse); [R,sigma]=corrcov(COVB); R; % R is the correlation matrix for the parameters sigma; % sigma is the standard error vector relerr=sigma./beta'; % relative error (coefficient of variance) for each parameter ss=resids'*resids; n=length(x); p=length(beta); condX=cond(J); %needs to be < 10^6. J is the sensitivity matrix = X detXTX=det(J'*J); %needs to be far from zero % Organizing Dref(SE), zT(SE), RMSE, Re lative errors in a matrix: results=[results; j k beta(1) sigma(1) Tref beta(2) sigma(2) rmse relerr(1) relerr(2)] Confidence interval and Predicted Band % CONFIDENCE INTERVAL FOR PARAMETERS ci=nlparci(beta, resids, J) % CONFIDENCE AND PREDICTION INTERVALS FOR DEPENDENT VARIABLES, Y % nonlinear regression confidence intervals -- 'on' means simultaneous bounds; 'off' is for nonsimultaneous bounds; % must use 'curve' for regression line, 'observation' for prediction interval [ypred, delta] = nlpredci(@forderexp_project,x,beta,resids,J,0.05, 'on' , 'curve' ); %confidence band for regression line [ypred, deltaob] =nlpredci(@forderexp_project,x,beta,resids,J,0.05, 'on' , 'observation' ); %prediction band for individual points yspan=range(ypred); % total span of ypred relrmse=rmse/yspan; % ratio of rmse vs. yspan % Organizing ypred in a matrix: ypredmatrix=[ypredmatrix; data(sec,1),data(sec,3),x,ypred] % 1 - lab, 3 - orgs x - timetemp end end %% RESIDUAL SCATTER PLOT figure hold on plot(x(:,1), resids, 'square' , 'Markerfacecolor' , 'k' ); YLine = [0 0]; 178 XLine = [0 max(x(:,1))]; plot (XLine, YLine, 'R' ); %plot a straight red line at zero ylabel( 'Observed y - Predicted y' , 'fontsize' ,16, 'fontweight' , 'bold' ) xlabel( 'Time (min)' , 'fontsize' ,16, 'fontweight' , 'bold' ) title ( 'Residual scatter plot for almond meal' , 'fontsize' ,16, 'fontweight' , 'bold' ) grid on mean=mean(resids) %% number of runs = number of times moving from one residual to the next crosses zero rescross=resids(2:n).*resid s(1:n - 1); %multiply each pair of residuals res_sign=sign(rescross); %get the sign of each multiplied pair count=0; for i=1:n - 1 if res_sign(i)<0 %if product of pair is < 0, that's a run count=count+1; end end fprintf( 'number of runs = %5.2f \ n' ,count); minrun=(n+1)/2; %count should be >=minrun fprintf( 'Minimum required number of runs = %5.2f \ n' ,minrun); %% residuals histogram -- same as dfittool, but no curve fit here [n1, xout] = hist(resids,10); %10 is the number of bins figure hold on set(gca, 'fontsize' ,14, 'fontweight' , 'bold' ); bar(xout, n1) % plots the histogram xlabel( 'Y_{observed} - Y_{predicted}' , 'fontsize' ,16, 'fontweight' , 'bold' ) ylabel( 'Frequency' , 'fontsize' ,16, 'fontweight' , 'bold' ) title ( 'Residual histogram for almond meal' , 'fontsize' ,16, 'fontweight' , 'bold' ) %% SCALED SENSITIVITY COEFFICIENT PLOT fnameFOR=@forderdiff_project xs=linspace(max(x(:,1)),max(x(:,1)),1000; Xp= SSC_project(beta,xs,fnameFOR) title ( ' SSC using beta estimates for almond meal' ) function y = forderexp_project(beta,t) %Make t a 2 column matrix, has time and temperature %first - order reaction equation, explicit form %beta are the parameters, and t are the independent variables values global Tref tt=t(:,1); %all times read in for all temperatures TT=t(:,2); %all temperatures y= - tt./beta(1).*(10.^((TT - Tref)./beta(2))); % y= - tt./(beta(1)*10.^(((TT - Tref)/beta(2)))); End 179 function y = forderdiff_project( beta,t ) %first - order model, differential form % called by inv_soln tspan=t; global Tref; T=@(t) 80+(90 - 80)/tspan(length(tspan))*t; [t,y]=ode45(@ss,tspan,0); %Take the derivative of your function with respect to each independent %variable function dy=ss(t,y) %function that computes the dydt dy(1)= - 1/beta(1)*10^((T(t) - Tref)/beta(2)); %Use global Tref later on end end function Xp = SSC_project( beta,x,func ) %computes scaled sensitivity coefficients % uses forward - difference approximation % beta are the parameters % x is the independent variable % func is the model d=0.001; ypred=func(beta,x); figure for i = 1:length(beta) %scaled sens coeff for forward problem betain = beta; %reset beta betain(i) = beta(i)*(1+d); yhat{i} = func(betain,x); %function with only one perturbed parameter Xp{i} = (yhat{i} - ypred)/d; %scaled sens coeff for ith parameter ysensf=Xp{i}; hold on h2(i) = plot(x,ysensf, ' -- b' , 'LineWidth' ,2); end ysensf1=Xp{1}; ysensf2=Xp{2}; %extract data from cell array into vec tors hold on YLine = [0 0]; XLine = [0 max(x)]; set(gca, 'fontsize' ,14, 'fontweight' , 'bold' ); h2(1) = plot(x,ysensf1, ' -- k' , 'LineWidth' ,2); h2(2) = plot(x,ysensf2, ' - k' , 'LineWidth' ,2); h2(3) = plot(x,ypred, ':k' , 'LineWidth' ,3); %X label is temperature, bec ause this is for simulated experiment, where 180 %temperature is increasing at a constant rate. xlabel( 'Time (min)' , 'fontsize' ,16, 'fontweight' , 'bold' ) %Y label units are ___________ ylabel( 'SSC or Log(CFU/g)' , 'fontsize' ,16, 'fontweight' , 'bold' ) plot (XLine, YLine, 'k' ); %plot a straight black line at zero legend(h2, 'X''_{Dref}' , 'X''_{z}' , 'Y_{pred}' , 'location' , 'best' ) grid on end %% %%%%% OPTIMAL EXP ERIMENTA L DESIGN %%%%%%%%%%%%% % this is a forward problem only % reproduce Beck and Arnold, p. 442, Fig. 8.10 % exponential decay eta = beta(1)*exp( - beta(2)*t) Eq. 8.4.3 %Found in last 3rd to last uploaded section of Beck Book on D2L %For our purposes, optimal exp design must be for simulated experiment. %This is the only way to get SSC's, which the opt exp is dependent upon. %% set up times close all clear clc t=0:1:140; %time m=length(t); %% parameter values beta(1)=20; % Because it's normalized beta(2)=16; % Because it's normalized %% Call and plot the function Y=optexp_func(beta,t); set (gca, 'fontsize' ,14, 'fontweight' , 'bold' ); plot(t,Y) xlabel( 'Time (min)' , 'fontsize' ,16, 'fontweight' , 'bold' ) ylabel( 'Reduction (Log CFU/g)' , 'fontsize' ,16, 'fontweight' , 'bold' ) title( 'Predicted data' , 'fontsize' ,16, 'fontweight' , 'bold' ) %% Compute C11, C12,...delta for opt exptl design fname=@optexp_func; %Xp are the scaled sensitivity coefficients %path(path,'C: \ Users \ dolank \ Dropbox (MSU) \ My Documents \ class \ BE 835 \ matlabgenericodes \ GenericSSC'); 181 Xp=SSC_V4(beta,t,fname); %compute scaled sensitivity coeffi cients as a cell array p=length(beta); %compute entire C matrix for i=1:p for j=1:p intgrnd=Xp{i}.*Xp{j}; %integrand for Eq. 8.3.5, Beck and Arnold, p. 434 C{i,j}=(1./t).*cumtrapz(t,intgrnd); % trapezoidal rule integral of Eq. 8.3.5 , Beck and Arnold, p. 434 clear intgrnd end end %to compute delta, must set up the C matrix for each time %extract C into a 3 - D matrix we call "CC" for i=1:p for j=1:p CC(i,j,:)=C{i,j}; %gives a 3D matrix that is m (depth) in time end end CC(:,:,1)=0; %beginning values at time = 0 are zero CC(1,1,1)=1; %except for C11, for the initial value delta(1)=0; %determinant at time zero = 0. for k=2:m delta(k)=det(CC(:,:,k)); end % delta=sqrt(delta);%converts units of delta to same as units for C %convert 3D to 2D for plotting %C matrix is symmetrical, so need only one half i=1; d=1; % "Cp" is "C for plotting" %for first row, i =1 for j=1:p Cp(d,:)=CC(i,j,:); d=d+1; end %f or second row, i=2 i=2; for j=2:p Cp(d,:)=CC(i,j,:); d=d+1; 182 end %% plot the results figure hold on linewidth = 1; % fig=figure; % left_color = [.5 .5 0]; % right_color = [0 .5 .5]; % set(fig,'defaultAxesColorOrder',[left_color; right_color]); yyaxis left f=10; %Size of each curve doesnt matter, as long as you can find the maximum set(gca, 'fontsize' ,16, 'fontweight' , 'bold' ); h(1)=plot(t,Cp(1,:), ':k' , 'linewidth' ,2); %C11 h(2)=plot(t, - Cp( 2,:), ' -- b' , 'linewidth' ,linewidth); %C12 h(3)=plot(t,10*Cp(3,:), ' - og' , 'linewidth' ,linewidth); %C22 beta2max = max(Cp(3,:)); h(4)=plot(t,delta, ' - ^m' , 'linewidth' ,linewidth); plot([0 max(t)],[0 0], ' - k' , 'linewidth' ,linewidth) axis([0 140 - 2 10]) ylabel( 'Sensitivi ty' ) %Temp line for simulated exp: simtemp = 65+(75 - 65)*t/max(t); tempbeta2max = simtemp(55); yyaxis right orange = [.91 .41 .17]; h(5) = plot(t,simtemp, ' - ' , 'Color' ,orange, 'linewidth' ,linewidth) ylabel( 'Temperature (C)' ) xlabel( 'Time (min)' , 'fontsize' ,16, 'fontweight' , 'bold' ) title( 'Lactose free SMP powder' , 'fontsize' ,12, 'fontweight' , 'bold' ) % Edit legend to fit the plot, coefficients, etc. legend(h, 'Dref' , 'sensitivity' , 'Z_T' , 'Delta' , 'Simulated temperature' , 'Location' , 'Best' ) grid on 183 APPENDIX W. MATLAB codes for global estimates of log - linear model and Bigelow model with optimum Tref and selected Tref (chapter 5) %nlinfit using file name = inv_soln.m %Housekeeping clear; % Clear the workspace. close all ; % Close all figures. format c ompact clc %READ IN DATA results=[]; % empty matrix to organize results obtained ypredmatrix=[]; % need the ypred for isothermal curve plotting % data=xlsread ('Dairy powders Data.xlsx','Matlab Lactose','A:H'); % data=xlsread ('Dairy powders Data.xlsx','M atlab Skim Milk Powder','A:H'); % data=xlsread ('Dairy powders Data.xlsx','Matlab Lactose - free SMP','A:H'); data=xlsread ( 'Dairy powders Data.xlsx' , 'Matlab MPI' , 'A:H' ); % COLUMN: % Product % Organism % Batch code % Exp code % Time % lpop % Temp % Log N/No orgs=unique(data(:,2)); % Organism [1,2] m=1; % This is used to find Tref for each organism % for m=1:length(orgs); k=orgs(m); sec=find(data(:,2)==k); % batch=unique(data(sec,3)); % batch % for n=1:length(batch); % g=batch(n); % sec=find(data(:,2)==k & data(:,3)==g); x=[data(sec,5), data(sec,7)]; % Col 5 time, Col 7 temp [x=time_temp matrix for nlinfit] yobs=data(sec,8); % Log N/No %INITIAL PARAMETER GUESS %beta(1) = Dref %beta(2) = z beta0(1)=25; %initial guess Dref beta0(2)=16; %initial guess z global Tref %Tref=90 184 % ****IMPORTANT**** Usually, fnameFOR will be different from fnameINV % fnameINV=@forderexp_project; %Only in simple models can fnameFOR be the same as fnameINV % fnameINV appears only in the nlinfit and nlpredci statements. Everywhere % else, use fnameFOR %This portion sets up the nlinfit to test Trefs from 65 to 95 Tlow=65; Tup=95;Rout=[]; Trange=linspace(Tlow,Tup,1000); for j=1:1:length(Trange); Tref =Trange(j); % This portion is standard code to estimate model parameters at the various Trefs [beta,resids,J,COVB,mse] = nlinfit(x,yobs,@forderexp_project,beta0); beta; rmse=sqrt(mse); [R,sigma]=corrcov(COVB); R; % R is the correlation matrix for the parameter s % This portion compiles the correlation values for an output and estimates Tref - optimal Rout(j)=R(1,2); end %RESIDUAL SCATTER PLOT figure; scatter(Trange,Rout) Topt=Trange(find(abs(Rout)==min(abs(Rout)))) xlabel( 'Temperature' ) ylabel( 'covariance' ) title( 'Minimum covariance plot for Tref' , 'FontSize' ,16) % This portion uses Tref optimal to estimate the Dref and Z at the optimal temperature value Tref =Topt; [beta,resids,J,COVB,mse] = nlinfit(x,yobs,@forderexp_project,beta0); beta; rmse=s qrt(mse); [R,sigma]=corrcov(COVB); R; sigma; % sigma is the standard error vector relerr=sigma./beta'; % relative error (coefficient of variance) for each parameter ss=resids'*resids; n=length(x); p=length(beta); condX=cond(J); %needs to be < 10^6. J is the sensitivity matrix = X detXTX=det(J'*J); %needs to be far from zero % Organizing Dref(SE), zT(SE), RMSE, Relative errors in a matrix: % results=[results; k beta(1) sigma(1) Tref beta(2) sigma(2) rmse relerr( 1) relerr(2)] % all batches so omit variable g % results=[results; k g beta(1) sigma(1) Tref beta(2) sigma(2) rmse relerr(1) relerr(2)] 185 % CONFIDENCE INTERVAL FOR PARAMETERS % ci=nlparci(beta, resids, J) % CONFIDENCE AND PREDICTION INTERVALS FOR DEPENDENT VARIABLES, Y % nonlinear regression confidence intervals -- 'on' means simultaneous bounds; 'off' is for nonsimultaneous bounds; % must use 'curve' for regression line, 'observation' for prediction interval [ypred, delta] = nlpredci(@forderexp_project,x,be ta,resids,J,0.05, 'on' , 'curve' ); %confidence band for regression line [ypred, deltaob] =nlpredci(@forderexp_project,x,beta,resids,J,0.05, 'on' , 'observation' ); %prediction band for individual points yspan=range(ypred); % total span of ypred relrmse=rmse/yspan; % ratio of rmse vs. yspan % Organizing ypred in a matrix: ypredmatrix=[ypredmatrix; data(sec,2),x,ypred] % 2 - orgs % end % end % omit one end for all batches figure hold on plot(x(:,1), resids, 'square' , 'Markerfacecolor' , 'b' ); %%%Why s ome have red outline? YLine = [0 0]; XLine = [0 max(x(:,1))]; plot (XLine, YLine, 'R' ); %plot a straight red line at zero ylabel( 'Observed y - Predicted y' , 'fontsize' ,16, 'fontweight' , 'bold' ) xlabel( 'Time (min)' , 'fontsize' ,16, 'fontweight' , 'bold' ) grid on title( 'residual scatter plot for Milk protein isolate' ) mean=mean(resids) %% number of runs = number of times moving from one residual to the next crosses zero rescross=resids(2:n).*resids(1:n - 1); %multiply each pair of residuals res_sign=sign(rescross); %ge t the sign of each multiplied pair count=0; for i=1:n - 1 if res_sign(i)<0 %if product of pair is < 0, that's a run count=count+1; end end fprintf( 'number of runs = %5.2f \ n' ,count); minrun=(n+1)/2; %count should be >=minrun fprintf( 'Minimum required number of runs = %5.2f \ n' ,minrun); %% residuals histogram -- same as dfittool, but no curve fit here [n1, xout] = hist(resids,10); %10 is the number of bins figure hold on set(gca, 'fontsize' ,14, 'fontweight' , 'bold' ); bar(xout, n1) % plo ts the histogram 186 xlabel( 'Y_{observed} - Y_{predicted}' , 'fontsize' ,16, 'fontweight' , 'bold' ) ylabel( 'Frequency' , 'fontsize' ,16, 'fontweight' , 'bold' ) SSC Tref = Topt fnameFOR = @forderdiff_project; xs=linspace(min(x(:,1)),max(x(:,1)),1000); Xp=SSC_project(beta ,xs,fnameFOR); title( 'SSC using beta estimates for Milk protein isolate' ) 187 APPENDIX X. MATLAB codes for effect of sugar types (chapter 5) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%% % D - VALUES FOR EFFECT OF SUGAR TYPES ON RESUSCITATION OF SURVIVORS ON % NONSELECTIVE DIFFERENTIAL MEDIA IN SMP AND LSMP AT 90C AND 70C, RESPECTIVELY %%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%%%%%%%%%%%%% clear; % Clear the workspace. close all ; % Close all figures. format compact clc %Select data in its corresponding range: data=xlsread( 'Dairy powders Data' , 'Matlab sugar effect' ); % Col 5:time (min) % Col 8:Log N/No %% Skim milk powder (SMP) % SMP NORMAL SAL: x1=data(2:31,5); y1=data(2:31,8); % SMP ADDED SUGAR SAL: x2=data(32:61,5); y2=data(32:61,8); % SMP NORMAL EF : x3=data(62:90,5); y3=data(62:90,8); % SMP ADDED SUGAR EF: x4=data(91:119,5); y4=data(91:119,8); %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Lactose - free skim milk powder (LSMP) % % lSMP NORMAL SAL: % x1=data(120:153,5); % y1=data(120:153,8); % % % lSMP ADDED SUGAR SAL: % x2=data(154:193,5); % y2=data(154:193, 8); % % % lSMP NORMAL EF : % x3=data(194:233,5); % y3=data(194:233,8); % % % lSMP ADDED SUGAR EF: % x4=data(234:278,5); % y4=data(234:278,8); beta0(1)=15; %Initial parameter guess for both products % f1 is a function for beta % J is sensitivity matrix % RMSE unit is the same as y and should be 10% less than ypred 188 % sigma is standard error of vector [beta1, resids1, J1, COVB1, mse1]=nlinfit(x1,y1,@f1,beta0); [beta2, resids2, J2, COVB2, mse2]=nlinfit(x2,y2,@f1,beta0); [beta3, resids3, J3, COVB3, mse3]=nlinfit(x3,y3,@f1,beta0); [beta4, resids4, J4, COVB4, mse4]=nlinfit(x4,y4,@f1,beta0); beta1 beta2 beta3 beta4 rmse1=sqrt(mse1) rmse2=sqrt(mse2) rmse3=sqrt(mse3) rmse4=sqrt(mse4) [~, sigma1]=corrcov(COVB1) [~, sigma2]=corrcov(COVB2) [~, sigma 3]=corrcov(COVB3) [~, sigma4]=corrcov(COVB4) %Confidence interval for parameter ci1=nlparci(beta1, resids1, 'jacobian' ,J1) ci2=nlparci(beta2, resids2, 'jacobian' ,J2) ci3=nlparci(beta3, resids3, 'jacobian' ,J3) ci4=nlparci(beta4, resids4, 'jacobian' ,J4) %Confidence and prediction intervals for the dependent variable %nonlinear regression confidence intervals -- 'on' means simultaneous %bounds; 'off' is for non - simultaneous bounds; must use 'curve' for %regression line, 'observation' for prediction interval [ypred1, delta1] = nlpredci(@f1,x1,beta1,resids1,J1,0.05, 'on' , 'curve' ); %confidence band for regression line [ypred1, deltaob1] =nlpredci(@f1,x1,beta1,resids1,J1,0.05, 'on' , 'observation' ); %prediction band for individual points [ypred2, delta2] = nlpredci(@f1,x2,beta2,resids2,J2,0.05, 'on' , 'curve' ); [ypred2, deltaob2] =nlpredci(@f1,x2,beta2,resids2,J2,0.05, 'on' , 'observation' ); [ypred3, delta3] = nlpredci(@f1,x3,beta3,resids3,J3,0.05, 'on' , 'curve' ); [ypred3, deltaob3] =nlpredci(@f1,x3,beta3,resids3 ,J3,0.05, 'on' , 'observation' ); [ypred4, delta4] = nlpredci(@f1,x4,beta4,resids4,J4,0.05, 'on' , 'curve' ); [ypred4, deltaob4] =nlpredci(@f1,x4,beta4,resids4,J4,0.05, 'on' , 'observation' ) ANCOVA data=xlsread( 'Dairy powders Data' , 'Matlab sugar effect' ); x=data(:,5); % time in min y=data(:,8); % Normalized data log N/No 189 g1=data(:,1); % Product 1 - SMP NORMAL, 2 - SMP ADDED SUGAR 3 - LSMP NORMAL 4 - LSMP ADDED SUGAR g2=data(:,2); % 1 - Sal 2 - Ef % COMPARE SMP SAL normal vs SMP SAL added sugar sec=find(g1~=3 & g1~=4 & g2~=2); % % COMPARE SMP EF normal vs SMP EF added sugar % sec=find(g1~=3 & g1~=4 & g2~=1); % % % COMPARE LSMP SAL normal vs LSMP SAL added sugar % sec=find(g1~=1 & g1~=2 & g2~=2); % % % COMPARE SMP EF normal vs SMP EF added su gar % sec=find(g1~=1 & g1~=2 & g2~=1); x=data(sec,5); y=data(sec,8); g1=g1(sec); % product g2=g2(sec); % organism [h,atab,ctab,stats]=aoctool(x,y,g1); stats 190 APPENDIX Y. Scaled s ensitivity coefficient (SSC) and residual plots (chapter 4) Figure Y. 1 Example of s caled sensitivity coefficient and residual scatter plots for chapter 4 191 APPENDIX Z. Scaled sensitivity coefficient (SSC) and residual plots (chapter 5 ) Figure Z. 1 Example of scaled sensitivity coefficient and residual plots for dairy powders 192 APPENDIX AA. Optimal e xperimental design plot (chapter 4) Figure AA. 1 Example of optimal e xperimental design plot for chapter 4. 193 APPENDIX AB. Optimal experimental design (chapter 5) Figure AB. 1 Example of optimal experimental design plot for chapter 5. 194 APPENDIX A C . I noculation flow chart (Chapter 3) Figure AC. 1 Flowchart of experimental methodology for chapter 3 195 APPENDIX A D . Photographs of isothermal inactivation experiment Figure AD. 1 Materials and equipment for isothermal inactivation experiment A ) conditioning chamber, B C ) thermal death test (TDT) cells, and D ) TDT cells packed with sample, prior to water bath treatment. A ) B ) C D 196 APPENDIX A E . Standardized methodology (chapter 5) This APPENDIX includes step - by - step standardized protocols for culture maintenance, inoculum harvesting and propagation, product inoculation, and isothermal treatment . MOFFETT CENTER TITLE : CULTURE MAINTENANCE AUTHOR : IFSH (Grasso - Kelley) DOCUMENT NO. : USDA CAP - MC - QA - 213 - SOP001 - V03 EFFECTIVE DATE : 09/02/2016 Purpose: To obtain and maintain microbial cell cultures for subsequent experimental use. Materials and Equipment : Microorganisms - Salmonella Agona 447967 Source: FDA, ORA Arkansas Regional Lab (Jefferson, AR) Associated recall: Puffed rice cereal, Minnesota Salmonella Enteritidis PT 30 ATCC BAA - 1045 Raw almonds Salmonella Tennessee K4643 Source: Dr. L. Beuchat, University of Georgia (Athens, GA) Assoicated recall: 2006 peanut butter outbreak Salmonella Montevideo 488275 Source: FDA, ORA Arkansas Regional Lab (Jefferson, AR) Associated recall: 2009 - 2010 black pepper outbreak Salmonella Mbandaka 698538 Source: FDA, ORA Arkansas Regional Lab (Jefferson, AR) Associated recall: Sesame tahini from Turkey Enterococcus faecium NRRL B - 2354 Source: USDA, ARS (Peoria, IL) Media sourced from BD Difco, Franklin Lakes, NJ Tryptic Soy Agar (Soybean - Casein Digest Agar) (Fisher Catalogue: DF0369 - 07 - 8) Tryptic Soy Broth (Soybean - Casein Digest Medium) (Fisher Catalogue: DF0370 - 07 - 5) Yeast extract (Fisher Catalogue:DF1810 - 07 ) 197 Prepared media media may be prepared up to 1 month prior to the use. Store prepared media at refrigerated conditions (4°C). If media is not made fresh, remove media from refrigerator (~2 - 24h prior to use) to allow media to warm to room temperature (~22°C). Tryptic Soy A gar with 0.6% yeast extract plates (TSAYE) Tryptic Soy Broth with 0.6%, 10 mL (TSBYE) Supplies and Equipment 80% sterile glycerol Vortex 1 mL sterile cryopreservation vials 37°C incubator 1 mL pipettor 37°C water bath 1 mL sterile pip ette tips 4°C incubator/refrigerator 10 mL culture tubes with lids - 80°C freezer Biosafety cabinet 100 x 15 petri dishes Paper towels Parafilm Pertinent personal protective equipment 70% ethanol (o r other disinfectant) Biohazard bag(s) Permanent marker (and labelling tape, if desired) Procedure Upon receipt of microorganisms, each laboratory should; - Confirm receipt of 6 cultures (listed above) - Resuscitate cultures - Prepare multiple individual frozen stock cultures of each microorganism for future use. - Prepare stock cultures See APPENDIX A) Working culture active stock plate stored at 4°C, used for research studies ( use for 1 month ) Permanent culture stock plate used to prepare the subsequent months working and permanent culture ( transfer monthly, up to 5 transfers ) 1. Resuscitation of stock culture from frozen vials or original receipt of microorganism a. Remove individual vials to be resuscitated from freezer. b. Thaw in a 37°C water bath with gentle agitation for ~2 min, or until ice crystals have melted. c. Remove the vial from the water bath and decontaminate it by dipping or spraying with 70% ethanol. d. In biosafety laminar flow cabinet, unscrew vial and transfer entire contents to 10 mL room temperature TSBYE using a 1 mL pipettor. Ensure proper labelling of microorganism identification, init ials, and date. e. Vortex TSBYE + culture. f. Loosen cap, incubate TSB + culture at 37°C for 24 ± 2 h. g. Examine TSBYE + culture overnight culture for turbidity. If turbid, continue to step h. If not turbid, return to step a. h. Vortex. L sterile loop(s) streak 1 loopful of overnight culture to surface of room 198 temperature TSAYE, for isolation. (Do a three - point isolation streak, using a new loop for each streak.) j. Incubate inverted plate(s) at 37°C for 24 ± 2 h. k. Remove plates from incubator and confirm the presence of isolated colonies. If not present, go to step a. l. Wrap plate with parafilm for storage. m. Store inverted, wrapped plates at 4°C for up to 1 month. This permanent stock plate may be used to create additional fro zen stock, permanent stock, and working stock (as needed). 2. Creation of frozen stock plate. b. Transfer harvested colony to 10 mL TSBYE. c. Vortex to mix thoroughly . d. Loosen cap, incubate TSBYE + culture at 37°C for 24 ± 2 h. e. Pipet 0.5 mL overnight culture and 0.5 mL 80% sterile glycerol to an appropriately labeled cryopreservation vial, seal vial. f. Vortex. g. Freeze the glycerol stock at - 80°C. h. Repeat steps e. through g. to create additional stock cultures. 3. Creation of working and permanent stock plates permanent stock plate. b. Aseptically streak harvested colony to surface of room temperature TSAYE, for isolation. (Do a three - point isolation streak, using a new loop for each streak.) c. Label TSAYE plate with the following, at a minimum, microorgan ism, current date, initials, working or permanent plate, # of transfers and/or date retrieved from frozen stock. d. Incubate inverted plate(s) at 37°C for 24 ± 2 h. e. Remove plates from incubator and confirm the presence of isolated colonies. If not p resent, go to step a. f. Wrap plate with parafilm for storage. g. Store inverted, wrapped plates at 4°C for up to 1 month. 3. Frequency of transfer basis. b. All act every 6 months. Additional transfers may be completed if: a. Contamination of stock plate(s) occurs, or is thought to have occurred. due to issue s related to media drying out, cultures not maintained at 4°C, etc. c. A new research aim is started. Fresh cultures are recommended upon starting any new research aim. 199 Notes: Always follow good laboratory practices; decontaminate surfaces before and after performing microbial transfer techniques, use aseptic techniques, perform microbial manipulation in a biosafety laminar flow cabinet. Prepare frozen stock cultures as needed to replenish stock. Prepare permanent and working stock cultures monthly. Al ways use working stock cultures to prepare cultures for individual experiments. 200 MOFFETT CENTER TITLE : INOCULUM PROPOGATION AND HARVESTING AUTHOR : IFSH (Grasso - Kelley) DOCUMENT NO. : USDA CAP - MC - QA - 213 - SOP002 - V03 EFFECTIVE DATE : 09/02/2016 Purpose: To prepare inoculums for food matrix inoculations. Materials and Equipment : Microorganisms as stated in Culture maintenance protocol . Media sourced from BD Difco, Franklin Lakes, NJ Tryptic Soy Agar (Soybean - Casein Digest Agar) (Fisher Catalogue: DF0369 - 07 - 8) Tryptic Soy Broth (Soybean - Casein Digest Medium) (Fisher Catalogue: DF0370 - 07 - 5) Yeast extract (Fisher Catalogue:DF0127 - 07 - 1 ) Buffer ed peptone water (Fisher Catalogue: DF1807 - 17 - 4) Ammonium Iron Citrate Sodium Thiosulfate Esculin Hydrate Prepared media media may be prepared up to 1 month prior to the use. Store prepared media at refrigerated conditions (4°C). If media is not made fresh, remove media from refrigerator (~2 - 24h prior to use) to allow media to warm to room temperature (~22°C). Tryptic Soy Agar with 0.6% yeast extract plates (TSAYE) Sodium Thiosulfate Modified Tryptic Soy Agar with 0.6% yeast extract (mTSAYE) Esculin Modified Tryptic Soy Agar with 0.6% yeast extract (eTSAYE) 0.1% buffered peptone water, 9 mL (BPW) Supplies and Equipment - Vortex 1 mL pipettor 37°C incubator 1 mL sterile pipette tips 4°C incubator/refrigerator Biosafety cabinet 70% ethanol (or other disinfectant) L - shaped spreaders Permanent marker Sterile tu bes (10 - 50 mL conical tubes) Paper towels Parafilm Pertinent personal protective equipment Procedure 1. Creation of lawns for inoculation complete, individually, for all 5 Salmonella serovars or E. faecium, as needed. loop, aseptically harvest one isolated colony from working stock plate (USDA CAP - MC - QA - 213 - SOP001 - V03). b. Transfer harvested colony to 10 mL TSBYE. 201 c. Vortex to mix thoroughly. d. Loosen cap, incubate TSB + culture at 37°C for 24 ± 2 h. e . Vortex. d f . Using an L - shaped spreader, spread culture across the surface of the plate. Prepare the appropriate amount of plates to provide the adequate volume of inoculum needed. Note: 1 plate provides 0.5 - 1mL of inoculum. g . Allow culture to be completely absorbed onto the surface of the agar before inversion. h . Incubate inverted plate(s) at 37°C for 24 ± 2 h. i . Wrap working stock culture with parafilm and store at 4°C. j . C omplete steps a. g. for remaining microbial serovars, individually. 2. Harvesting plate grown cells (lawns) a. Pipette 1 mL BPW to the surface of the 100 mm diameter lawn plate (adjusting volumes if larger sized plates are used). b. Using an L - shape d spreader gently agitate the cells into solution. c. Tilt plate and gather cell suspension along the edge of the plate. d. Using a 1 mL pipettor, draw up as much liquid as possible. e. Transfer harvested culture to a sterile tube. f. Repeat steps a. e. for the remaining plates of the same culture. g. Pool harvested culture from the same serovar into 1 tube. h. Repeat steps a. g. for each serovar, separately. i. Vortex each coni cal tube, individually. Note: Target initial concentration ~ 11 log CFU/mL. 3. To prepare a Salmonella cocktail a. Using individually harvested cultures from step 2, aseptically pipette 1 mL of each serovar into a separate sterile tube. b. Vortex Salmonella cocktail. c. Use harvested cocktail within 2 h. d. Enumerate Salmonella cocktail inoculum a. Serially dilute an aliquot of the Salmonella cocktail inoculum using BPW. b. Spread plate 0.1 mL of appropriate dilutions onto mTSAYE (at least 3 dilutions). c. Incubate inverted plates at 37°C for 24 ± 2 h. d. Count mTSAYE plates. e. Calculate starting inoculum level. 4. Working with E. faecium a. Vortex harvested culture. b. Use harvested cocktail within 2 h. c. Enumera te E. faecium inoculum a. Serially dilute an aliquot using BPW. b. Spread plate 0.1 mL of appropriate dilutions onto eTSAYE (at least 3 dilutions). c. Incubate inverted plates at 37°C for 24 ± 2 h. 202 d. Count eTSAYE plates. e. Calculate starting inocul um level. Notes: Always follow good laboratory practices; decontaminate surfaces before and after performing microbial transfer techniques, use aseptic techniques, perform microbial manipulation in a biosafety laminar flow cabinet. Always use working sto ck cultures to prepare cultures for individual experiments. Use equal volumes of all 5 Salmonella serovars if working with the Salmonella cocktail. Use E. faecium separately, if desired work is with the surrogate. 203 MOFFETT CENTER TITLE : LIQUID INOCULATION OF PEANUT BUTTER AUTHOR : IFSH (Grasso - Kelley/Pickens) DOCUMENT NO. : USDA CAP - MC - QA - 213 - SOP004 - V03 EFFECTIVE DATE : 09/02/2016 Purpose: To create an inoculated, homogenous low - water activity high - fat em ulsion food product for further inactivation or storage studies. Note: Follow USDA CAP - MC - QA - 213 - SOP002 - V03 for culture growth and harvesting before proceeding. Media sourced from BD Difco, Franklin Lakes, NJ Tryptic Soy Agar (Soybean - Casein Digest Agar) (Fisher Catalogue: DF0369 - 07 - 8) Tryptic Soy Broth (Soybean - Casein Digest Medium) (Fisher Catalogue: DF0370 - 07 - 5) Yeast extract (Fisher Catalogue:DF0127 - 07 - 1 ) Buffe red peptone water (Fisher Catalogue: DF1807 - 17 - 4) Ammonium Iron Citrate Sodium Thiosulfate Esculin Hydrate Prepared media media may be prepared up to 1 month prior to the use. Store prepared media at refrigerated conditions (4°C). If media is not made fresh, remove media from refrigerator (~2 - 24h prior to use) to allow media to warm to room temperature (~22°C). Sodium Thiosu lfate Modified Tryptic Soy Agar with 0.6% yeast extract (mTSAYE) Esculin Modified Tryptic Soy Agar with 0.6% yeast extract (eTSAYE) 0.1% buffered peptone water, 9 mL (BPW) Supplies and Equipment Vortex Round plastic tubs 1 mL pipettor Biosafety cabinet 1 mL sterile pipette tips Peanut butter Plate harvested inoculum 24 oz Whirl - Pak style bag or plastic bucket Weigh boat 70% ethanol or other disinfectant Balanc e Permanent marker Timer Paper towels Sterile tongue depressor Personal protective equipment 2 or 4 oz Whirl - Pak style bags Peanut oil 4TE water activity meter L - shaped spreaders Water activity standards Tween 80 Water activity cups a n lids Stomacher Controlled environment chamber with mixing apparatus 204 Procedure: ALL CULTURE WORK should be completed in a biosafety cabinet. 1. Preparation of peanut butter a. Obtain creamy peanut butter from a commercial source. b. Perform background microflora analysis. c. Aseptically, sample at least 3 1 - g samples into 2 or 4 oz Whirl - Pak bags. d. Serial dilute each sample using BPW. e. Individually spread - plate 0.1 mL of appropriate dilutions onto TSAYE, in duplicate (at least 3 dilutions). f. Individually plate onto Petrifilm. g. Incubate inverted plates or non - inverted Petrifilm at 37°C for 24 ± 2 h. h. Count TSAYE and Petrifilm i. Calculate background aerobic mi croflora load and associated other results. j. Store materials, closed, at room conditions prior to step 2. 2. Preparation of inoculum for peanut butter a. Obtain Salmonella cocktail or Enterococcus inoculum from USDA CAP - MC - QA - 213 - SOP002 - V03. b. Mi x peanut oil 1:1 with inoculum. c. Add Tween 80 to form emulsion (volume is dependent on volume of inoculum; usually a few drops). d. Vortex to form a suspension. 3. Inoculation of particulate based matrix : For each 200 g product inoculated use 3 mL inoculum (USDA CAP - MC - QA - 213 - SOP002 - V02). a. Weigh 200 g peanut butter into large Whirl - Pak style bag. b. Vortex culture obtained from following USDA CAP - MC - QA - 213 - SOP002 - V03. c. Pipette 3 mL target inoculum throughout the entire sample (rather than one spot). d. Loosely fold over and secure the top of the Whirl - Pak bag. e. Hand - massage inoculum into ingredients until there are minimal spots of visible liquid throughout bag (~1 min). butter back towards bottom of bag with tongue depressor (or other similar instrument). g. Stomach Whirl - Pak bag for 30 s at 230 rpm. h. Scrape peanut butter bag to bottom of bag, as in step 3.f. i. Repeat steps 3.e - h. three times, each. j. Transfer inoculated peanut butter to small round plastic container. k. Place inside environmental chamber with continuously mixing apparatus (set to 35% RH, or other RH , as appropriate). l. Let inoculated sample sit for at least 1 h to allow for moisture equilibration before use before continuing on to step 4. 4. Product measurements a. Prepare water activity meter for analysis by turning on meter as indicated in th e 205 instruction manual. b. Run appropriate water activity standards; 0.250, 0.500, 0.760, and 0.984. c. Aseptically transfer at least 3 samples of inoculated, equilibrated product to separate water activity sample cups in a glove box. d. Cover sample cups with lids to minimize water activity changes. e. Measure water activity at 25°C. f. Record results, if water activity analysis of samples is not within +/ - 0.025 a w, do not proceed. Allow material to equilibrate for 1 d longer and rest art at step 3. 5. Homogeneity testing a. Working in glove box, aseptically sample ten 1 g samples randomly from the equilibrated inoculated food material. Place samples into 2 oz or 4 oz Whirl - Pak bags. b. Serial dilute each sample using BPW. c. Indi vidually spread - plate 0.1 mL of appropriate dilutions onto mTSAYE or eTSAYE (as appropriate), in duplicate (at least 3 dilutions). d. Incubate inverted plates at 37°C for 24 ± 2 h. e. Count mTSAYE/eTSAYE plates (use 25 - 250 as countable range, recording all results for dilutions plated). f. Calculate Salmonella /surrogate. If standard deviation is > 0.3 log CFU/g, the sample is not homogeneous. Do not use sample for further testing. Restart inoculation with fresh materials. 6. Storage a. Continue to store inoculated material at 35% RH in a controlled environmental chamber (or other RH, as appropriate). b. Conduct storage and/or inactivation study within 4 - 21 d of inoculation. Ensure starting microbial population is enumerated on testin g date. 206 TITLE: ALMOND MEAL INOCULATION AUTHOR: MSU ( N urul Hawa Ahmad) Inoculation of whole almond 1. Obtain almond from the walk - in cooler and put in the Whirl - Pack bag. 2. Harvest lawn plates as previously described (100 g product inoculated use 1 mL inoculum). 3. Pipette inoculum to the entire almond in the Whirl - Pack. 4. Hand - massage the bag for even inoculum distribution (until there is no visible liquid spot observed). 5. Transfer the inoculated almond on a filter paper on top of sterile tray and spread them for uniform layer. 6. Leave the hood blower running and let them dry at least for 1 h. Do not close the hood door. Fabrication of almond meal 1. Obtain a food processor and feed ~ 100 g inoculated almond. 2. Grind the almond for 45 s at the speed of 1, stopping every 9 s to push almond chunk at the side of the food processor wall. 3. Transfer almond meal to a filter paper boat(s) and scrape the leftover using spatula 4. Spread the almond meal for uniform layer. The depth of the layer should be less than 2.5 cm deep. 5. Carefully place the almond meal in the equilibration chamber to reach targeted aw. Homogeneity testing (obtained from standardized protocol) 1. Aseptically sample ten 1 g samples randomly from the equilibrated inoculated food material. P lace samples into 2 oz or 4 oz Whirl - Pak bags. 2. Serial dilute each sample using BPW. 3. Individually spread - plate 0.1 mL of appropriate dilutions onto TSAYE, in duplicate (at least 3 dilutions). 4. Incubate inverted plates at 37°C for 24 ± 2 h. 5. Count TSAYE plates (use 25 - 250 as countable range). a. Calculate Salmonella /surrogate. If standard deviation is > 0.3 log CFU/g, the sample is not homogeneous. Do not use sample for further testing. Restart inoculation with fresh materials. b. Frequency: 2 day after inoculation 207 TITLE: DATE PASTE INOCULATION AUTHOR: MSU ( N urul Hawa Ahmad) Inoculation of date halves 1. Obtain dates from the walk - in cooler. 2. Pit the date and cut each date into 6 pieces. 3. Using 100 g product per 1 mL inoculum rate, pipette approp riate volume of inoculum to each date pieces,being careful not to let any run over the side (For instance, if you have 20 date pieces you may want to pipette 50 uL on each piece) 4. Leave the hood blower running and let the halves dry for 30 minutes. Do not c lose the hood door. Fabrication of date paste 1. Assemble the Kitchen - Aid grinder machine and turn it on to speed number 2. 2. Feed date halves through the top of the machine and collect the paste that comes out the side. 3. Feed the resulting paste back through the machine 2 more times to achieve a very uniform paste. 4. Scrape the date paste leftover inside the grinder using a spatula 5. Be sure to bleach the parts of the grinder after use if the date halves were inoculated. Equilibration of date paste 1. Using spatula, make a small ball of date paste (almost the same size that can fit in the test cell) and arrange it on the filter paper (this filter paper is used to overlay the rack in equilibration chamber) 2. Carefully place the inoculated date paste balls in th e equilibration chamber that has been set to 0.65 aw 3. After 2 days of equilibration, perform homogeneity test as described below. Homogeneity testing (obtained from standardized protocol) 1. Aseptically sample ten 1 g samples randomly from the equilibrated in oculated food material. Place samples into 2 oz or 4 oz Whirl - Pak bags. 2. Serial dilute each sample using BPW. 3. Individually spread - plate 0.1 mL of appropriate dilutions onto sal - TSAYE, in duplicate (at least 3 dilutions). 4. Incubate inverted plates at 37°C fo r 24 ± 2 h. 5. Count TSAYE plates (use 25 - 250 as countable range). a. Calculate Salmonella /surrogate. If standard deviation is > 0.3 log CFU/g, the sample is not homogeneous. Do not use sample for further testing. Restart inoculation with fresh materials. b. Freq uency: 2 day after inoculation 208 TITLE: GROUND BLACK PEPPER INOCULATION AUTHOR: UNL (Sabrina Vasquez) S tandard Operating Procedures (SOP) Inoculation, equilibration and grinding of whole black pepper 1. Introduction To study the behavior of Salmonella spp. on whole and ground black pepper, proper inoculation, equilibration and grinding procedures must be followed. Five strains of Salmonella are used to prepare a cocktail to inoculate whole black peppercorns. The inocu lated sample is equilibrated to a desired water activity using custom - made relative humidity - controlled chambers. Once the inoculated samples have reached the target water activity, a Waring Powder Grinder is used to grind the samples. To minimize potenti al exposure to personnel and environmental contamination, the procedures described in this SOP must be followed. During inoculation, equilibration and grinding procedures, personnel must wear latex/nitrile shoulder gloves, dust mask, and a lab coat of app ropriate size. 2. Inoculation procedures 2.1 Inoculation of material inside biosafety cabinet 1. When necessary equilibrate non - inoculated whole black peppercorns to a desired water activity using a relative humidity controlled chamber. 2. Each batch of equilibrated whole black peppercorns ( 1 kg) will be added to a Sterile bag. 3. The bag will be placed inside the biosafety cabinet. The sample should be uniformly distributed along the bottom of the bag. 4. 20 mL of Salmonella cocktail will be added to the bag using a spray device. Figure 1. Addition of the Salmonella cocktail should be done in increments of 4 mL. Once each 4mL have been sprayed onto the whole black pepper, the sterile bag should be sealed and hand - massaged for 2 minutes. Continue with this pr ocedure until all 20 mL of Salmonella cocktail has been added to the whole black pepper samples. The initial population at this step is 7 - 8 Log CFU/g. 5. Using a sterile metal scoop pour the inoculated whole black peppercorns on aluminum trays. Figure 2. Pla ce each tray onto the aluminum stand. 6. Carefully remove the aluminum stand, while holding the trays, from the biosafety cabinet and transfer it to a humidity - controlled chamber, previously placed next to the biosafety cabinet. Figure 3. 7. Place the lid on t he chamber and secure it with the aluminum locks. 8. Return the humidity - controlled chamber to its rank and connect the corresponding cables. Figure 4. 9. Set the relative humidity - controlled chamber to the appropriate % RH associated with the desired target water activity. Keep inoculated samples under this condition for 36 - 48 hours. 3. Grinding whole black pepper procedure 1. Disconnect the airflow tubes, sensor cable and fan from the relative humidity chamber containing the equilibrated inoculated whol e black pepper. 2. Carefully take the chamber and place it beside the biosafety cabinet. 3. Open the chamber and remove the stand with trays containing the inoculated sample. 209 4. Place the stand inside the biosafety cabinet. 5. Place the chamber lid back and return the chamber to the rank if it disturbs movements around the biosafety cabinet. 6. Using a metal scoop pour all the inoculated samples into a sterile bag. 7. Refrigerate the samples in bag for 2 - 3 hours in a refrigerator designated for biohazardous material. Als o introduce into the refrigerator the Waring cups placed inside sterile bag. 8. Before removing the sample and Waring cups from the refrigerator, place the following items inside the biosafety cabinet: a. Waring powder grinder b. U.S 20 mesh sieve c. Sieve holder d. Press n Seal e. Masking tape 9. Remove the Waring cups and bag containing the inoculated sample and place them inside the biosafety cabinet. 10. Using a metal scoop, add black peppercorns to the Waring cup until the 2 - cup level ( 300 grams). 11. Cover the Waring cup with a Press n Seal wrap. 12. Place the cup into the Waring powder grinder and place the plastic lid to lock the equipment. Figure 5. 13. 14. Once the sample has been grou nded, wait for 10 minutes before opening the Waring cup. This is important to prevent aerosols particles to flow excessively inside the biosafety cabinet. 15. Carefully, remove the plastic lock container and the Press n Seal plastic from the Waring cup. Disca rd the Press n Seal in a biosafety bag. 16. Carefully, pour the grounded product on a U.S 20 mesh sieve with its holder. 17. Place Press n Seal on top of sieve and secure it with masking tape. Figure 5. 18. Using your hands, shake the sieve for at least 3 min. 19. Wai t for 10 min before removing the Press n Seal from the sieve. 20. Remove the masking tape and Press n Seal and discard them in a biosafety bag. 21. Pour the finely ground black pepper into a sterile bag. Record its weight. 22. Collect the ground black pepper that did not passed through the sieve and add it to a sterile bag. Record its weight. 23. Repeat the grinding procedure with another 300 grams. 24. Once 900 grams of whole black pepper have been grounded (3 runs), add all of the pepper particles retained by the sieve i nto a Waring cup and grind them for 30 seconds, following the procedure described for the whole peppercorns. 25. Calculate the amount of finely ground black pepper that should be mixed with all the second finely ground samples obtained to maintain the same pr oportion of particles. Use the following equation: A= total mass of finely ground black pepper collected after 3 runs. B= total mass of finely ground black pepper obtained from grinding the retained black pepper. 210 C= total mass of retained black pepper on sieve. 26. Mix both corresponding amounts of finely black pepper on a sterile bag. 27. Using a sterile scoop, add the finely ground black pepper back to the aluminum trays. 28. Place the relative humidity chamber next to the biosafety cabinet. 29. Place the trays in the stand and carefully place the stand back in the relative humidity chamber. 30. Close the chamber and return it to the rack. 31. Connect all the airflow tubes, sensor and fan. 32. Set the relative humidity - controlled chamber to the appropriate % RH associated with the desired target water activity. Keep inoculated samples under this condition for 36 - 48 hours. 4. Equipment and biosafety cabinet sanitizing after whole black pepper inoculation and/or grinding procedures 4. 1 Equ ipment sanitizing a) U.S Mesh Sieve and Sieve Holder a. Inside the biosafety cabinet, spray 70% ethanol to the U.S Mesh Sieve and holder. Place them inside a sterile bag. b. Autoclave for at least 15 min at 121 ° C. b) Waring powder grinder, cups and cover a. Disconnect the Waring Powder Grinder from the electricity source. b. Soak paper towels with 70% ethanol and wipe the Waring powder grinder and its cable. c. Let it dry inside the biosafety cabinet. d. Soak the used Waring cup and plastic cover into 70% ethanol. Let stand for 15 min and then let it dry inside the biosafety cabinet. c) Metal Scoop and Aluminum stands a. Wipe the metal scoop with paper towels soaked with 70% ethanol. b. Place them inside a sterile bag. c. Autoclave for at least 15 min at 121 °C. 4.2 Biosafety cabinet sanitizing a) Spray 70% ethanol inside the whole biosafety cabinet. b) Place paper towels onto a Swiffer sweeper and soak it into 70% ethanol. c) Clean the walls and front glass of the biosafety cabinet using the Swiffer sweeper. d) Discard all used paper towels into a biohazardous bag. e) Autoclave for at least 15 min at 121 °C. 4.3 Relative Humidity Controlled Chambers a) Follow the SOP for decontamination of Relative Humidity Controlled Chambers. 211 1 2 3 4 5 Figure 1. Spray inoculation of whole black pepper Figure 2. Pour inoculated whole black pepper into aluminum trays Figure 3. Humidity - controlled chamber beside biosafety cabinet Figure 4. Humidity - controlled chamber with inoculated product Figure 5. Waring cup and sieve with Press n Seal 212 TITLE: WHEAT FLOUR INOCULATION AUTHOR: WSU (Yuqiao Jin) Example: Inoculate 6 mL to 600 g wheat flour 1. Transfer 50 g wheat flour into a whirl - pack bag. 2. Inoculate 6 mL harvested bacteria into 50 g wheat flour. 3. Place the whir - pack bag into stomacher blender (stomacher 400 circulator) for 3 min with 230 strokes/min. 4. Take the whirl - pack bag out, hand massage the inoculated wheat flour for 1 min. 5. Place the whir - pack bag back to the stomacher blen der for 3 min with 230 strokes/min. 6. Take the whirl - pack bag out, hand massage the inoculated wheat flour for 1 min. 7. Place the whir - pack bag back to the stomacher blender for 3 min with 230 strokes/min. 8. Take the whirl - pack bag out, hand massag e the inoculated wheat flour for 1 min. a new whirl - pack bag. Transfer 55 g of un - inoculated wheat flour into the bag, too. 10. Hand massage the mixture for 1 min. 11. Place the whir - pack bag into stomacher blender for 3 min with 230 strokes/min. 12. Take the whirl - pack bag out, hand massage the inoculated wheat flour for 1 min. 13. Place the whir - pack bag back to the stomacher blender for 3 min with 230 strokes/ min. 14. Take the whirl - pack bag out, hand massage the inoculated wheat flour for 1 min. 15. Repeat the same procedure from 9 to 14 for the rest of wheat flour. 16. Put all the inoculated wheat flour (600 g) into a large sterile bag, mix with a sterile disposal spatula by hand randomly for 3 min. 213 TITLE: NONFAT DRY MILK POWDER INOCULATION AUTHOR: WSU (Shuxiang Liu) 1. Grinding non - fat milk powder procedure and inoculation 2. Pack 100 g non - fat milk powder into a sterile whirl bag; 3. Direct pipette 1 ml inoculum into 100 g non - fat milk powder and hand mix well; 4. Put inoculated and wet non - fat milk power into chamber for equilibration 24 h; 5. Use a sterile scoop pour all the dry, inoculated samples into a sterile bag; 6. Double bag the s terile bag and remove the air in between as much as possible; 7. Seal the bags separately and appropriately without too much air inside; 8. Use a small tool (whatever is harder than the non - fat milk powder) and break the particles gently. Watch out for the break age of sterile bags and change them if necessary (repeat e - g); 9. When there are no visible large - particles in inoculated non - fat milk powder, transfer it into a larger sterile bag for further equilibration; 10. Set the relative humidity - controlled chamber to the appropriate % RH associated with the desired target water activity. Keep inoculated samples under this condition for 36 hours. Note: equilibrated samples can be sealed in a closed container for storage up to a month (from the inoculation date). Make sur e the equilibrated and inoculated samples will not be exposed to other RH% for more than 1 hour (in case the water activity alters). 214 MOFFETT CENTER TITLE : ISOTHERMAL INACTIVATION STUDIES USING TEST CELLS AUTHOR : IFSH (Grasso - Kelley/Hildebrandt) DOCUMENT NO. : USDA CAP - MC - QA - 213 - SOP005 - V02 EFFECTIVE DATE : 09/02/2016 Purpose : To conduct thermal death time trials to determine microbial inactivation kinetics at iso - thermal, iso - water activity conditions. Media sourced from BD Difco, Franklin Lakes, NJ Tryptic Soy Agar (Soybean - Casein Digest Agar) (Fisher Catalogue: DF0369 - 07 - 8) Tryptic Soy Broth (Soybean - Casein Digest Medium) (Fisher Catalogue: DF0370 - 07 - 5) Yeast extract (Fisher Catalogue:DF0127 - 07 - 1 ) Buffered peptone water (Fisher Catalogue: DF1807 - 17 - 4) Ammonium Iron Citrate Sodium Thiosulfate Esculin Hydrate Prepared media media may be prepared up to 1 month prior to the use. Store prepared media at refrigerated conditions (4°C). If media is not made fresh, remove media from refrigerator (~2 - 24 h prior to use) to allow media to warm to room temperature (~22°C). Sodium Thiosulfate Modified Tryptic Soy Agar with 0.6% yea st extract (mTSAYE) Esculin Modified Tryptic Soy Agar with 0.6% yeast extract (eTSAYE) 0.1% buffered peptone water, 9 mL (BPW) Supplies and Equipment Inoculated and equilibrated matrix Water activity cups and lids Test cells, set (aluminum, stainless steel, etc) Sterile spatula/scoop Test cells, with thermocouple Balance Water activity meter (4TEV) Pipettor Water activity standards (as appropriate) Sterile pipettor tips Glove box Sterile bags (2/ 4 oz, Whirl - Pak style) Biosafety cabinet Timer Circulating oil or water bath Ice bath 37°C incubator L - shaped spreaders 70% ethanol or other disinfectant Permanent marker Paper towels Personal protective equipment Thermometer wi th compatible thermocouple 215 Procedure: PACKING test cells should be completed in an equilibrated glove box. ALL CULTURE WORK should be completed in a biosafety cabinet. PREWORK : Determine come - up time to ensure isothermal conditions during inactivation trials. Use a K - type thermocouple (handh eld thermometer) inserted into the center of the test cell containing equilibrated, non - inoculated sample - 0.2°C). treatment, by calibrated thermocouple/thermometer) and measure water/oil bath temperature. submerge test cell into oil bath and simultaneously start timer. s for the center of the sample to reach 0.5°C below of set temperature in a fully loaded water/oil bath. Monitor heating medium temperature. has cooled to room tempe rature prior to repeating. Use new sample for each measurement.) 1. Packing test cells (Use of a static eliminator may aid in reduction of static effects on inoculated materials.) a. Allow glove box to reach/maintain appropriate RH conditions. b. Fully pack test cells with inoculated and equilibrated matrix to each test cell. (Pack at least 18 test cells per trial.) c. Transfer at least 3 samples of material to sample cups. Cover lids to minimize water activity changes. 2. Water activity measurements a. Prepare water activity meter for analysis by turning on meter as instructed in the instruction manual. b. Run appropriate water activity standards; 0.250, 0.500, 0.760, and 0.984. c. Measure water activity at 25°C. 3. Isothermal treatment If the water bath does not fit all 18 test cells, use randomized pull in the same water bath OR try to put 9 test cell s. a. Set oil or water bath to test temperature. (Allow oil bath to heat/cool to set temperature.) b. Use calibrated thermometer to measure heating medium temperature. Temperature should be within 0.2°C. c. Submerge test cells into water/oil bath and simu ltaneously start timer. d. Heat samples for appropriate time intervals (taking into account CUT - see PREWORK Come - up time) e. At all treatment times, immediately submerge treated samples in an ice bath to cool fully. 216 Notes: d be conducted; more may be tested, as appropriate/required. (Actual intervals to be used will be dependent on matrix, microorganism, and temperature. A minimum of 3 - log reduction should be seen, ideally a 5 - log reduction will be achieved. Ideally time intervals will let us see ~ 1 - log reduction per time point, but uniform time intervals are not required.) 4. Enumeration a. Asepticall y transfer the content of the treated test cell into sterile Whirl - Pak style bag. Note mass of sample transferred using a balance. b. Serial dilute each sample using BPW (As appropriate, dilution scheme may change depending on treatment conditions. Prewo rk may be required to estimate expected inactivation). c. Individually spread - plate 0.1 mL of appropriate dilutions onto appropriate modified TSAYE, in duplicate (plating at least 3 separate tubes from dilution scheme). d. Complete for all treated sample s. e. Incubate inverted plates at 37°C for 24 ± 2 h. f. Count plates (use 25 - 250 as countable range, but record all results). Notes: Always follow good laboratory practices; decontaminate surfaces before and after performing microbial transfer techniques, use aseptic techniques, perform microbial manipulation in a biosafety laminar flow cabinet. 217 BIBLIOGRAPHY 218 BIBLIOGRAPHY Adams, MR, Moss, MO, McClure, P. (2016). Food microbiology (4th ed.). Royal Society of Chemistry. Ahmed, J., & Al - jasass, F. M. (2014). Date fruit composition and nutrition. In A. A. K. Muhammad Siddiq, Salah M . Aleid (Ed.), Dates:postharvest science, processing technology and health benefits (1st ed., pp. 261 283). West Sussex, UK: John Wiley & Sons, Ltd. Alexander, D. C., Fitzgerald, S. F., Depaulo, R., Kitzul, R., Daku, D., Levett, P. N., & Cameron, A. D. S. (2015). Laboratory - a cquired i nfection with Salmonella enterica Serovar Typhimurium e xposed by W hole - Genome Sequencing. Journal of Clinical Microbiology , 54 , 190 - 193. https://doi.org/10.1128/JCM.02720 - 15 Almond Board of California. (2014). Guidelines for using Enterococcus faecium NRRL B - 2354 as a surrogate microorganism in almond process validation . Modesto, CA. Retrieved from https://www.almonds.com/sites/default/files/guidelines_for_using_enterococcus_faecium_n rrl_b - 2354_as_a_surrogate_microorganism_in_ almond_process_validation.pdf Anonymous. (2014). Kirkland Signature sliced fruit recalled from Costco for potential Salmonella . Retrieved November 10, 2018, from https://www.foodsafetynews.com/2014/03/kirkland - signature - sliced - fruit - recalled - from - costco - for - potential - Salmonella / Anonymous. (2017). Composition of commercial flour. In S. Finnie & W. A. Atwell (Eds.), Wheat flour (2nd ed., pp. 31 48). St. Paul, MN: W oodhead Publishing. https://doi.org/10.1016/B978 - 1 - 891127 - 90 - 8.50003 - 6 Archer, J., Jervis, E. T., Bird, J. O. N., & Gaze, J. O. Y. E. (1998). Heat resistance of Salmonella weltevreden in low - moisture environments. Journal of Food Protection , 61 (8), 969 97 3. Aviles, B., Klotz, C., Eifert, J., Williams, R., & Ponder, M. (2013). Biofilms promote survival and virulence of Salmonella enterica Tennessee during prolonged dry storage and after passage through an in vitro digestion system. International Journal of Food Microbiology , 162 (3), 252 259. https://doi.org/10.1016/j.ijfoodmicro.2013.01.026 Balaji, B., Connor, K. O., Lucas, J. R., Anderson, J. M., & Csonka, L. N. (2005). Timing of induction of osmotically controlled genes in Salmonella enterica serovar Typhimurium , 219 determined with quantitative real - time reverse transcription - PCR. Applied and Environmental Microbiology , 71 (12), 8273 8283. https://doi.org/10.1128/AEM.71.12.8273 Balasubramanian, S., Roselin, P., Singh, K. K., Zachariah, J., & Saxena, S. N. (2016). Postharvest Processing and Benefits of Black Pepper, Coriander, Cinnamon, Fenugre ek, and Turmeric Spices. Critical Reviews in Food Science and Nutrition , 56 (10), 1585 1607. https://doi.org/10.1080/10408398.2012.759901 BeMiller, J. N. (2019). Monosaccharides. In Carbohydrate Chemistry for Food Scientists (pp. 1 23). AACC International Press. https://doi.org/10.1016/B978 - 0 - 12 - 812069 - 9.00001 - 7 Bergan, T, Bovre, K, Hovig, B. (1970). Present status of the species Micrococcus freudenreichii guillebeau 1891. International Journal of Systematic Bacteriology , 20 , 249 254. Retrieved from https: //www.microbiologyresearch.org/docserver/fulltext/ijsem/20/3/ijs - 20 - 3 - 249.pdf?expires=1565639530&id=id&accname=sgid026736&checksum=43334F28FA4B8 660B99DB73F4E8644F5 Beuchat, LR., Mann, D. . (2014). Survival of Salmonella on dried fruits and in aqueous drie d fruit homogenates as affected by temperature. Journal of Food Protection , 77 , 1102 1109. Beuchat, L. R., & Mann, D. A. (2011). Inactivation of Salmonella on pecan nutmeats by hot air treatment and oil roasting. Journal of Food Protection , 74 (9), 1441 14 50. https://doi.org/10.4315/0362 - 028X.JFP - 11 - 080 Beuchat, L. R., & Mann, D. A. (2015). Survival of Salmonella in cookie and cracker sandwiches containing inoculated, low water activity fillings. Journal of Food Protection , 78 (10), 1828 1834. https://doi.org/10.4315/0362 - 028X.JFP - 15 - 142 Beuchat, L. R., Mann, D. A., Kelly, C. A., & Ortega, Y. R. (2017). Retention of viability of Salmonella in sucrose as affected by type of inoculum, water activity, and storage temperature. Journal of Food Protection , 80 (9), 1408 1414. https://doi.org/10.4315/0362 - 028X.JFP - 16 - 537 Bianchini, A., Stratton, J., Weier, S., Hartter, T., (2014). Use of Enterococcus faecium as a surrogate for Salmonella enterica during extrusion of a balanced carbohydrate - protein meal. Journal of Food Protection , 77 (1), 75 82. https://doi.org/10.4315/0362 - 028X.JFP - 13 - 220 Blessington, T., Theofel, C. G., & Harris, L. J. (2013). Short communication A dry - inoculation method for nut kernels. Food Microbiology , 33 (2), 292 297. 220 Bonacina, J., Suarez, N., Hormigo, R., Fadda, S., Lechner, M., & Saavedra, L. (2017). A geno mic view of food - related and probiotic Enterococcus strains. DNA Research , 24 , 11 24. Bonacina, Julieta, Suárez, N., Hormigo, R., Fadda, S., Lechner, M., & Saavedra, L. (2017). A genomic view of food - related and probiotic Enterococcus strains. DNA Resear International Journal for Rapid Publication of Reports on Genes and Genomes , 24 (1), 11 24. https://doi.org/10.1093/dnares/dsw043 Borowski, A. G., & Ingham, S. C. (2009). Lethality of home - style dehydrator processes against Escherichia coli 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 (10), 2056 2064. Bowman, L. S., Waterman, K. M., Williams, R. C., & Ponder, M. A. (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 (7). https://doi .org/10.4315/0362 - 028X.JFP - 14 - 483 Bronlund, J., & Paterson, T. (2004). Moisture sorption isotherms for crystalline, amorphous and predominantly crystalline lactose powders. International Dairy Journal , 14 , 247 254. H. (2007). Two consecutive large outbreaks of Salmonella enterica serotype Agona infections in infants linked to the consumption of powdered infant formula. Pedi atric Infectious Disease Journal , 26 (2), 148 152. https://doi.org/10.1097/01.inf.0000253219.06258.23 (2016). The response of foodborne pathogens to osmotic and des iccation stresses in the food chain. International Journal of Food Microbiology , 221 , 37 53. https://doi.org/10.1016/j.ijfoodmicro.2015.12.014 Burke, N., Zacharski, K., Southern, M., Hogan, P., Ryan, M., & Adley, C. (2018). The dairy industry: process, mo nitoring, standards, and quality. In A. Díaz & R. García - Gimeno (Eds.), Descriptive food science (pp. 4 17). IntechOpen. https://doi.org/http://dx.doi.org/10.5772/57353 Burnett, SL., Gehm, ER, Weissinger, WR., Beuchat, L. (2000). Survival of Salmonella in peanut butter and peanut butter spread. Journal of Applied Microbiology , 89 , 472 477. 221 Busta, F. F., Suslow, T. V., Parish, M. E., Beuchat, L. R., Farber, J. N., Garrett, E. H., & Harris, L. J. (2003). The Use of indicators and surrogate microorganisms f or the evaluation of pathogens in fresh and fresh - cut produce. Comprehensive Reviews in Food Science and Food Safety , 2 (s1), 179 185. https://doi.org/10.1111/j.1541 - 4337.2003.tb00035.x Cairney, J., Booth, IR, Higgins, C. (1985). Salmonella typhimurium pro P gene encodes a transport the osmoprotectant betaine. Journal of Bacteriology , 1218 1223. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC219318/pdf/jbacter00217 - 0252.pdf Calhoun, S., Post, L., Warren, B., Thompson, S., & Bontempo, A. R. (201 8). Prevalence and concentration of Salmonella on raw, shelled peanuts in the United States. Journal of Food Protection , 81 , 1755 1760. https://doi.org/10.4315/0362 - 028X.JFP - 18 - 114 Centers for Disease Control and Prevention. (1993). Salmonella Serotype Te nnessee in Powdered Milk Products and Infant Formula -- Canada and United States, 1993 . Retrieved from https://www.cdc.gov/mmwr/preview/mmwrhtml/00021081.htm Centers for Disease Control and Prevention. (1998). Multistate outbreak of Salmonella serotype Ag ona infections linked to toasted oats cereal -- United States, April - May, 1998. MMWR. , 47 (22), 462 464. Retrieved from https://www.cdc.gov/mmwr/preview/mmwrhtml/00053368.htm Centers for Disease Control and Prevention. (2004). Outbreak of Salmonella serotype Enteritidis ifections associated with raw almonds --- United States and Canada, 2003 -- 2004. MMWR , 53 (22), 484 487. Retrieved from https://www.cdc.gov/mmwr/preview/mmwrhtml/ mm5322a8.htm Centers for Disease Control and Prevention. (2007). Multistate outbreak of Salmonella serotype Tennessee infections associated with peanut butter --- United States, 2006 -- 2007. MMWR. , 56 (21), 521 524. Retrieved from https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5621a1.htm Centers for Disease Control and Prevention. (2008a). Multistate outbreak of human Salmonella infections caused by contaminated dry dog food --- United States, 2006 -- 2007. MMWR. , 57 (19), 521 524. Retri eved from https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5719a4.htm Centers for Disease Control and Prevention. (2008b). Multistate outbreak of Salmonella Agona infections linked to rice and wheat puff cereal (final update). Retrieved November 10, 2018, 222 from https://www.cdc.gov/ Salmonella /2008/rice - wheat - puff - cereal - 5 - 13 - 2008.html Centers for Disease Control and Prevention. (2009). Multistate outbreak of Salmonella infections associated with peanut butter and peanut butter -- containing products --- United Stat es, 2008 -- 2009. MMWR. , 58 (4), 85 90. Retrieved from https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5804a4.htm Centers for Disease Control and Prevention. (2010). Salmonella Montevideo infections associated with salami products made with contaminated imported black and red pepper --- United States, July 2009 -- April 2010. MMWR , 59 (50), 1647 1650. Retrieved from https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5950a3.htm Centers for Disease Control and Prevention. (2012). Human Salmonella Typhimurium infections asso ciated with exposure to clinical and teaching microbiology laboratories (final update). Retrieved August 8, 2019, from https://www.cdc.gov/ Salmonella /2011/lab - exposure - 1 - 17 - 2012.html?CDC_AA_refVal=https%3A%2F%2Fwww.cdc.gov%2F Salmonella %2Ftyphim urium - labora tory%2Findex.html Centers for Disease Control and Prevention. (2014a). Human Salmonella Typhimurium Infections Linked to Exposure to Clinical and Teaching Microbiology Laboratories (Final Update) | Typhimurium Infections Linked to Microbiology Laboratorie s | June 2014 | Salmonella | CDC. Retrieved August 8, 2019, from https://www.cdc.gov/ Salmonella /typhimurium - labs - 06 - 14/index.html Centers for Disease Control and Prevention. (2014b). Multistate outbreak of Salmonella Braenderup infections linked to nut bu tter manufactured by nSpired Natural Foods, Inc. (final update). https://doi.org/https://www.cdc.gov/ Salmonella /braenderup - 08 - 14/index.html Centers for Disease Control and Prevention. (2016a). Multistate outbreak of Salmonella Montevideo and Salmonella Se nftenberg infections linked to Wonderful pistachios. Retrieved November 10, 2018, from https://www.cdc.gov/ Salmonella /montevideo - 03 - 16/index.html Centers for Disease Control and Prevention. (2016b). Multistate outbreak of Salmonella Virchow infections lin ked to Garden of Life raw meal organic shake & meal products. Retrieved November 10, 2018, from https://www.cdc.gov/ Salmonella /virchow - 02 - 16/index.html Centers for Disease Control and Prevention. (2017). Human Salmonella Typhimurium infections linked to exposure to clinical and teaching microbiology laboratories. Retrieved August 8, 2019, from https://www.cdc.gov/ Salmonella /typhimurium - 07 - 17/index.html 223 C enters for Disease Control and Prevention. (2018a). Multistate outbreak of Salmonella I 4,[5],12:b: - infections linked to kratom products. Retrieved November 10, 2018, from https://www.cdc.gov/ Salmonella /kratom - 02 - 18/index.html Centers for Disease Control and Prevention. (2018b). Multistate outbreak of Salmonella https://doi.org/https://www.cdc.gov/ Salmonella /mbandaka - 06 - 18/index.html Ceylan, E., & Bautista, D. A. (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 (5), 934 939. https://doi.org/10.4315/0362 - 028X.JFP - 14 - 271 Milliken, G. (2016). Validation of baking to control Salmonella serovars in hamburger bun manufacturin g, and evaluation of Enterococcus faecium ATCC 8459 and Saccharomyces cerevisiae as nonpathogenic surrogate indicators. Journal of Food Protection , 79 (4), 544 552. https://doi.org/10.4315/0362 - 028X.JFP - 15 - 241 Chen, L., Wei, X., Irmak, S., Chaves, B. D., & Subbiah, J. (2019). Inactivation of Salmonella enterica and Enterococcus faecium NRRL B - 2354 in cumin seeds by radiofrequency heating. Food Control , 103 , 59 69. https://doi.org/10.1016/j.foodcont.2019.04.004 Chung, H. - J., Birla, S. L., & Tang, J. (2008). Performance evaluation of aluminum test cell designed for determining the heat resistance of bacterial spores in foods. LWT , 41 , 1351 1359. https://doi.org/10.1016/j.lwt.2007.08.024 Clark, C. W., & John Ordal, Z. (1969). Thermal injury and recovery of Sa lmonella typhimurium and its effect on enumeration procedures. APPLIED MICROBIOLOGY , 332 336. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC377980/pdf/applmicro00009 - 0052.pdf Clark, Z., Paterson, A. H. J., Joe, R., & Mcleod, J. S. (2016). Am orphous lactose crystallisation kinetics. International Dairy Journal . https://doi.org/10.1016/j.idairyj.2015.12.012 Codex Alimentarius. (2015). Code of hygienic low - moisture foods CAC/RCP 75 - 2015 . Retrieved from http://www.fao.org/fao - who - codexalimentarius/sh - proxy/en/?lnk=1&url=https%253A%252F%252Fworkspace.fao.org%252Fsites%252Fcode x%252FStandards%252FCAC%2BRCP%2B75 - 2015%252FCXC_075e.pdf 224 Davis, E. A. (1995). Functionality of sugars: Physicochemical interactions in foods. American Journal of Clinical Nutrition , 62 (1 SUPPL.), 170 177. https://doi.org/10.1093/ajcn/62.1.170S Deng, X., Li, Z., & Zhang, W. (2012). Transcriptome sequencing of Salmonella enterica serovar Enteritidis under desiccation and starvation str ess in peanut oil. Food Microbiology , 30 (1), 311 315. https://doi.org/10.1016/j.fm.2011.11.001 Dewey - Mattia, D., Manikonda, K., Hall, A. J., Wise, M. E., & Crowe, S. J. (2018). Surveillance for foodborne disease outbreaks - United States, 2009 - 2015. MMWR Su rveill. Summ. , 67 (10), 1 11. Retrieved from https://www.cdc.gov/mmwr/volumes/67/ss/pdfs/ss6710a1 - H.pdf Diez - gonzalez, F., Chick, M., Lourenco, A., Maserati, A., & Ryan, C. (2017). Thermal Death Kinetics of Salmonella enterica in a Toasted Oat Cereal Model . Frontiers in Microbiology . Dolan, K. D., & Mishra, D. K. (2013). Parameter Estimation in Food Science. Annu. Rev. Food Sci. Technol , 4 , 401 422. https://doi.org/10.1146/annurev - food - 022811 - 101247 Enache, E., Kataoka, A. I., Black, D. G., Napier, C. D., Podolak, R., & Hayman, M. M. (2015). Development of a dry inoculation method for thermal challenge studies in low - moisture foods by using talc as a carrier for Salmonella and a surrogate ( Enterococcus faecium ). Journal of Food Protection , 78 (6), 1106 111 2. https://doi.org/10.4315/0362 - 028X.JFP - 14 - 396 (2014). The heat resistance of Salmonella Tennessee in peanut paste formulations at four different levels of fat and wa ter activity. In Annual Meeting International Association of Food Protection . https://doi.org/10.13140/2.1.3729.2806 European Food Safety Authority. (2018). Multi - country outbreak of Salmonella Agona infections linked to infant formula . https://doi.org/10.2903/sp.efsa.2018.EN - 1365 European Union. (2005). Commission regulation (EC) No 2073/2005 of 15 November 2005 on microbiological criteria for foodstuffs. Official J. L338, 1 26. Retrieved from https://www.fsai.ie/uploadedFiles/Consol_Reg2073_2005.pdf Farakos, M. S., & Frank, J. F. (2014). Challenges in the Control of Foodborne Pathogens in Low - Water Activity Foods and Spices. In The microbiology safety of low water activtiy foods (pp. 15 30). https://doi.org/10.1007/978 - 1 - 4939 - 2062 - 4 225 Farakos, S M Santillana, Frank, J. F., & Schaffner, D. W. (2013). International Journal of Food Microbiology Modeling the in fl uence of temperature , water activity and water mobility on the persistence of Salmonella in low - moisture foods . International Journal of Food Microbiology , 166 (2), 280 293. https://doi.org/10.1016/j.ijfoodmicro.2013.07.007 Farakos, Sofia M Santillana, Schaffner, D. W., & Frank, J. F. (2014). Predicting survival of Salmonella in l ow Journal of Food Protection , 77 (9), 1448 1461. https://doi.org/10.4315/0362 - 028X.JFP - 14 - 013 Finn, S., Condell, O., Mcclure, P., Amézquita, A., & Fanning, S. (2013). Mechanisms of survival , respon ses , and sources of Salmonella in low - moisture environments. Frontiers in Microbiology , 4 (November), 1 15. https://doi.org/10.3389/fmicb.2013.00331 Fisher, K.,Philips, C. (2009). The ecology, epidemiology and virulence of Enterococcus . Microbiol. , 155 , 1 749 1757. Food and Agriculture Organization of the United Nations and World Health Organization. (2014). Ranking of low moisture foods in support of microbiological risk management . Retrieved from http://ucfoodsafety.ucdavis.edu/files/209893.pdf Franz, C. M. A. P., Holzapfel, W. H., & Stiles, M. E. (1999). Enterococci at the crossroads of food safety? International Journal of Food Microbiology . https://doi.org/10.1016/S0168 - 1605(99)00007 - 0 Franz, C. M. A. P., Huch, M., Abriouel, H., Holzapfel, W., & Gál vez, A. (2011). Enterococci as probiotics and their implications in food safety. International Journal of Food Microbiology . https://doi.org/10.1016/j.ijfoodmicro.2011.08.014 Galet, L., Goalard, C., & Dodds, J. A. (2010). The importance of surface energy in the dispersion behaviour of talc particles in aqueous media. Powder Technology , 190 (1 2), 242 246. https://doi.org/10.1016/j.powtec.2008.04.086 Gallaher, D. D., & A. Anderson, J. (2019). Wheat. In Whole Grains and their Bioactives (pp. 19 43). Wiley. h ttps://doi.org/10.1002/9781119129486.ch2 Gao, W., Howden, B. P., Stinear, T. P., Mecsas, J., & Buchrieser, C. (2018). Evolution of virulence in Enterococcus faecium , a hospital - adapted opportunistic pathogen. Current Opinion in Microbiology , 41 , 76 82. https://doi.org/10.1016/j.mib.2017.11.030 226 Garce, F., & Marks, B. P. (2014). Use of simulation tools to illustrate the effect of data management practices for low and negative plate counts on the estimated parameters of microbial reduction models. Journal of Food Protection , 77 (8), 1372 1379. https ://doi.org/10.4315/0362 - 028X.JFP - 13 - 462 Ghnimi, S., Umer, S., Karim, A., & Kamal - Eldin, A. (2017). Date fruit ( Phoenix dactylifera L.): an underutilized food seeking industrial valorization. NFS Journal , 6 , 1 10. https://doi.org/10.1016/j.nfs.2016.12.001 Gira, G. (2002). Enterococci from foods. FEMS Microbiology Reviews , 26 (April), 163 171. Grasso - Kelley, E. M. (2016). Foodborne pathogens in ready - to - eat peanut butter - containing products. https://doi.org/10.1016/B978 - 0 - 12 - 801916 - 0.00006 - 6 Graziani, C., Losasso, C., Luzzi, I., Ricci, A., Scavia, G., & Pasquali, P. (2017). Chapter 5 Salmonella . Foodborne Diseases (Third Edit). Elsevier Inc. https://doi.org/10.1016/B978 - 0 - 12 - 385007 - 2.00005 - X Gruzdev, N., Pinto, R., & Sela (Saldinger), S. (2012). Persiste nce of Salmonella enterica during dehydration and subsequent cold storage. Food Microbiology , 32 (2), 415 422. https://doi.org/10.1016/j.fm.2012.08.003 Gurtler, J. B. (2009). Evaluation of plating media for recovering Salmonella from thermally treated egg albumen 1. J. Appl. Poult. Res , 18 (April), 297 309. https://doi.org/10.3382/japr.2008 - 00109 Hanchi, H., Mottawea, W., Sebei, K., & Hammami, R. (2018). The genus Enterococcus : Between probiotic potential and safety concerns - an update. Frontiers in Microbio logy , 9 (AUG), 1 16. https://doi.org/10.3389/fmicb.2018.01791 Harris, L. J., & Yada, S. (2019). Flour and cereal grain products: foodborne illness outbreaks and product recalls [Tables and references] . https://doi.org/10.1093/cid/cir831 Harris, L. J., Yad a, S., Beuchat, L. R., & Danyluk, M. D. (2019). Outbreaks of foodborne illness associated with the consumption of tree nuts, peanuts, and sesame seeds . Retrieved from https://ucfoodsafety.ucdavis.edu/low - moisture - foods/nuts - and - nut - pastes. He, Y., Guo, D. , Yang, J., Tortorello, M. Lou, & Zhang, W. (2011). Survival and heat resistance of Salmonella enterica and Escherichia coli O157:H7 in peanut butter. Applied and 227 E nvironmental M icrobiology , 77 (23), 8434 8438. https://doi.org/10.1128/AEM.06270 - 11 Heredia, N., & García, S. (2018). Animals as sources of food - borne pathogens: A review. Animal Nutrition , 4 , 250 255. https://doi.org/10.1016/j.aninu.2018.04.006 Hernández - Alonso, P., Camacho - Barcia, L., Bulló, M., & Salas - Salvadó, J. (2017). Nuts and Dried Fruits: An Update of Their Beneficial Effects on Type 2 Diabetes. Nutrients , 9 (7), 673. https://doi.org/10.3390/nu9070673 Hildebrandt, I. A. N. M., Hu, C., Gra sso - kelley, E. M., Ye, P., Anderson, N. M., & Keller, S. E. (2017). Dry transfer inoculation of low - moisture spices containing antimicrobial compounds. Journal of Food Protection , 80 (2), 338 344. https://doi.org/10.4315/0362 - 028X.JFP - 16 - 279 Hildebrandt, I . M., Marks, B., Anderson, N. M., & Grasso - Kelley, E. M. (2020). Reproducibility of Salmonella thermal resistance measurements via multi - laboratory isothermal inactivation experiments. Journal of Food Protection . In press. Hildebrandt, I. M., Marks, B. P. , Ryser, E. T., Villa - Rojas, R., Tang, J., Garces - Vega, F. J., & Buchholz, S. E. (2016). Effects of inoculation procedures on variability and repeatability of Salmonella thermal resistance in wheat flour. Journal of Food Protection , 79 (11), 1833 1839. http s://doi.org/10.4315/0362 - 028X.JFP - 16 - 057 Hoffmans, C. M., & Fung, D. Y. C. (1992). Effective method for dry inoculation of bacterial cultures. Journal of Rapid Methods & Automation in Microbiology , 1 (4), 287 294. https://doi.org/10.1111/j.1745 - 4581.1992.t b00275.x Hu, M., & Gurtler, J. B. (2017). Selection of surrogate bacteria for use in food safety challenge studies: a review. Journal of Food Protection , 80 (9), 1506 1536. https://doi.org/10.4315/0362 - 028X.JFP - 16 - 536 International Nut and Dried Fruit. (2 018). Nuts and dried fruits statistical yearbook 2017/2018 . Retrieved from https://www.nutfruit.org/files/tech/1524481168_INC_Statistical_Yearbook_2017 - 2018.pdf Isbill, J., Kandiah, J., & Khubchandani, J. (2018). Use of ethnic spices by adults in the Unit ed States: An exploratory study. Health Promoting Perspectives , 8 , 33 40. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5797306/pdf/hpp - 8 - 33.pdf 228 Issenhuth - Jeanjean, S., Roggentin, P., Mikoleit, M., Guibourdenche, M., De Pinna, E., Nair, S., - X. (2014). Supplement 2008 - 2010 (no. 48) to the White - Kauffmann - Le Minor scheme. Research in Microbiologoy , 165 , 526 530. https://doi.org/10.1016/j. resmic.2014.07.004 Jeong, S., Marks, B. P., & Ryser, E. T. (2011). Quantifying the performance of Pediococcus sp . ( NRRL B - Enterococcus faecium ) as a nonpathogenic surrogate for Salmonella Enteritidis PT30 during moist - air convection heating of a lmonds. Journal of Food Protection , 74 (4), 603 609. https://doi.org/10.4315/0362 - 028X.JFP - 10 - 416 Jin, Yuqiao, Pickens, S. R., Hildebrandt, I. M., Burbick, S. J., Grasso - Kelley, E. M., Keller, S. E., & Anderson, N. M. (2018). Thermal inactivation of Salmon ella Agona in low - water activity foods: predictive models for the combined effect of temperature, water activity, and food component. Journal of Food Protection , 81 , 1411 1417. https://doi.org/10.4315/0362 - 028X.JFP - 18 - 041 Jones, G., de la Gandara, M. P., Herrera - Leon, L., Herrera - Leon, S., Martinez, C. V., Hureaux - - Da Silva, N. (2019). Outbreak of Salmonella enterica serotype Poona in infants linked to persistent Salmonella contamination in an infant formula manufacturing facility, France , August 2018 to February 2019. Euro Surveill. , 24 (13), 1900161. https://doi.org/10.2807/1560 - 7917.ES.2019.24.13.1900161 Jongenburger, I., Reij, M. W., Boer, E. P. J., Gorris, L. G. M., & Zwietering, M. H. (2010). Factors influencing the accuracy of the p lating method used to enumerate low numbers of viable micro - organisms in food. International Journal of Food Microbiology , 143 , 32 40. https://doi.org/10.1016/j.ijfoodmicro.2010.07.025 Jouppila, K., Kansikas, J., & Roos, Y. H. (1997). Glass Transition, Wa ter Plasticization, and Lactose Crystallization in Skim Milk Powder. Journal of Dairy Science , 80 , 3125 3160. https://doi.org/10.3168/jds.S0022 - 0302(97)76286 - 6 Jouppila, K., & Roos, Y. H. (1994). Water Sorption and Time - Dependent Phenomena of Milk Powders. Journal of Dairy Science , 77 , 1798 1808. https://doi.org/10.3168/jds.S0022 - 0302(94)77121 - 6 Juneja, V. K., Eblen, B. S., & Marks, H. M. (2000). Thermal inactiv ation of Salmonella serotypes in red meat as affected by fat content. Quantitative Microbiology , 2 , 189 225. Retrieved from https://link.springer.com/content/pdf/10.1023%2FA%3A1013995111581.pdf Kataoka, A. I., Enache, E., Black, D. G., Elliott, P. H., Nap ier, C. D., Podolak, R., & Hayman, 229 M. M. (2013). Survival of Salmonella Tennessee , Salmonella Typhimurium DT104 , and Enterococcus faecium in peanut paste formulations at two different levels of water activity and fat. Journal of Food Protection , 77 (8), 1 252 1259. https://doi.org/10.4315/0362 - 028X.JFP - 13 - 553 Keller, S. E., Grasso, E. M., Halik, L. A., Fleischman, G. J., Chirtel, S. J., & Grove, S. F. (2012). Effect of Growth on the Thermal Resistance and Survival of Salmonella Tennessee and Oranienburg in Peanut Butter , Measured by a New Thin - Layer Thermal De ath Time Device. Journal of Food Protection , 75 (6), 1125 1130. https://doi.org/10.4315/0362 - 028X.JFP - 11 - 477 Keller, S. E., Vandoren, J. M., Grasso, E. M., & Halik, L. A. (2013). Growth and survival of Salmonella in ground black pepper ( Piper nigrum ). Fo od Microbiology , 34 (1), 182 188. https://doi.org/10.1016/j.fm.2012.12.002 Kopit, L. M., Kim, B., Siezen, R. J., Harris, L. J., & Marco, L. (2014). Safety of the surrogate microorganism Enterococcus faecium NRRL B - 2354 for use in thermal process validation . Applied and Environmental Microbiology , 80 (6), 1899 1909. https://doi.org/10.1128/AEM.03859 - 13 Kornacki, J. L. (2012). Enterococcus faecium NRRL B - 2354: Tempest in a Teapot or Serious Foodborne Pathogen? Food Safety Magazine . Retrieved from https://www. foodsafetymagazine.com/magazine - archive1/april - may - 2012/enterococcus - faecium - nrrl - b - 2354 - tempest - in - a - teapot - or - serious - foodborne - pathogen/ Labuza, T. P. (1975). Sorption phenomena in foods: theoretical and practical aspects. In R. Chokyun (Ed.), Theory, determination and control of physical properties of food materials (pp. 197 219). Boston: D. Reidel Publishing Company. https://doi.org/10.1007/978 - 94 - 010 - 1731 - 2 Lang, E., Chemlal, L., Molin, P., Guyot, S., Alvarez - martin, P., Perrier - s, P. relevance of the substrate water activity. Food Research International , 99 , 577 585. Li, C., Huang, L., & Chen, J. (2014). Comparative study of thermal inactivation kinetics of Salmonella spp. in peanut butter and peanut butter spread. Food Control , 45 , 143 149. https://doi.org/10.1016/j.foodcont.2014.04.028 (2014). Effect of the local microenvironment on survival and thermal inactivation of Salmonella in low - and intermediate - moisture multi - ingredient foods. Journal of Food 230 Protection , 77 (1), 67 74. https://doi.org/10.4315/0362 - 028X.JFP - 13 - 277 Lian, F., Zhao, W., Yang, R., T ang, Y., & Katiyo, W. (2015). Survival of Salmonella enteric in skim milk powder with different water activity and water mobility. Food Control , 47 , 1 6. https://doi.org/10.1016/j.foodcont.2014.06.036 Limcharoenchat, P., M.K. James, K.D. Dolan, E.T. Ryser , and B. P. M. 2019. (2019). Effects of product structure, temperature, and water activity on the thermal resistance of Salmonella Enteritidis PT30 in low - moisture foods. Journal of Food Protection , In review. Limcharoenchat, P., Buchholz, S. E., James, M. K., Hall, N. O., Ryser, E. T., & Marks, B. P. (2018). Inoculation protocols influence the thermal resistance of Salmonella Enteritidis PT 30 in fabricated almond, wheat, and date products. Journal of Food Prot ection , 81 , 606 613. https://doi.org/10.4315/0362 - 028X.JFP - 17 - 297 Limcharoenchat, P., & Marks, B. P. (2018). Effects of Product Structure, Temperature, Water Activity, and Storage on the Thermal Resistance of Salmonella Enteritidis PT 30 in Low - Moisture F oods . ProQuest Dissertations and Theses . Michigan State University. Retrieved from http://sfx.scholarsportal.info/guelph/docview/2031162129?accountid=11233 Liu, S., Ozturk, S., Xu, J., Kong, F., Gray, P., Zhu, M. - validatio n of radio frequency pasteurization of wheat flour by inoculated pack studies. Journal of Food Engineering , 217 , 68 74. https://doi.org/10.1016/j.jfoodeng.2017.08.013 Liu, S., Rojas, R. V., Gray, P., Zhu, M. - J., & Tang, J. (2018). Enterococcus faecium as a Salmonella surrogate in the thermal processing of wheat flour: Influence of water activity at high temperatures. Food Microbiology , 74 , 92 99. https://doi.org/10.1016/J.FM.2018.03.001 Liu, S., Tang, J., Tadapaneni, R. K., Yang, R., & Zhu, M. - J. (2018). Exponentially increased thermal resistance of Salmonella spp. and Enterococcus faecium at reduced water activity. Applied and Environmental Microbiology , 84 (8), 1 12. Lucore, L.A, Anthony, J .G, Shirin, J. A. (2017). A thermal process lethality model for low water activity food. Food Protection Trends , 37 (1), 43 55. Ma, L. I., Zhang, G., Gerner - Smidt, P., Mantripragada, V., Ezeoke, I., & Doyle, M. P. (2009). Thermal inactivation of Salmonella in peanut butter. Journal of Food Protection , 72 (8), 1596 1601 . 231 Ma, L., Kornacki, J., Zhang, G., Lin, C. - H., & Doyle, M. (2007). Development of Thermal Surrogate Microorganisms in Ground Beef for In - Plant Critical Control Point Validation Studies. Journal of Food Protection , 70 , 952 957. Retrieved from https://jfoo dprotection.org/doi/pdfplus/10.4315/0362 - 028X - 70.4.952 M. (2010). The Global Burden of Nontyphoidal Salmonella Gastroenteritis . Clinical Infectious Diseases , 50 (6), 882 889. https://doi.org/10.1086/650733 Manero, A., & Blanch, A. R. (1999). Identification of Enterococcus spp. with a biochemical key. Applied and Environmental Microbiology , 65 (10), 4425 4430. Marks, B. P. (2008). Microbial modeling in food process models. Comprehensive Reviews in Food Science and Food Safety , 7 , 137 143. Mattick, K. L., Jørgensen, F., Wang, P., Pound, J., Ward, L. R., Legan, J. D., & Humphrey, T. J. (2001). Effect of Challenge Temperature and Solute Type on Heat Tolerance of Salmonella Serovars at Low Water Activity Effect of Challenge Temperature and Solute Type on Heat Tolerance of Salmonella Serovars at Low Water Activity. Applied and Environmental Microbiology , 67 (9), 4128 4136. https://doi.org/10.1128/AEM.67.9.4128 Meghwal, M., Goswami, T. K., & Goswami. (2012). Chemical composition , nutritional , Open Access Scientific Reports , 1 (2), 1 5. https://doi.org/10. 4172/scientificreports.1 Morgan, F., Nouzille, C. A., Baechler, R., Vuataz, G., & Raemy, A. (2005). Lactose crystallisation and early Maillard reaction in skim milk powder and whey protein allisation and early Maillard reaction in skim milk powder and whey protein. Lait , 85 , 315 323. https://doi.org/10.1051/lait Morris, C. F., & Rose, S. P. (1996). Wheat. In R. J. Henry & P. Kettlewell (Eds.), Cereal grain quality (pp. 4 54). Chapman & Hall. Naranjo, G. B., Gonzales, A. S. P., Leiva, G. E., & Malec, L. S. (2013). The kinetics of Maillard reaction in lactose - hydro lysed milk powder and related systems containing carbohydrate mixtures. Food Chemistry , 141 (4), 3790 3795. https://doi.org/10.1016/j.foodchem.2013.06.093 National Advisory Committee on Microbiological Criteria for Foods. (2010). Parameters for 232 determining inoculated pack/challenge study protocols. Journal of Food Protection , 73 (1), 140 202. Nguyen, L, Duong, LT, Mentreddy, RS. (2019). The U.S. import demand for spices and herbs by differentiated sources. Journal of Applied Research on Medicinal and Aromat ic Plants , 12 , 13 20. Nielsen, S. J., Div, M., Herrick, K. A., Akinbami, L. J., & Ogden, C. L. (2016). Nut consumption among U.S. youth, 2009 - 2012 . Hyattsville, MD. Retrieved from http://www.cdc.gov/nchs/data/databriefs/db238_table.pdf#3. Nielsen, S. J. , Kit, B. K., & Ogden, C. L. (2014). Nut consumption among U.S. adults, 2009 - 2010 . Hyattsville, MD. Retrieved from https://www.cdc.gov/nchs/data/databriefs/db176.pdf Nummer, B. A., Shrestha, S., & Smith, J. V. (2012). Survival of Salmonell a in a high sugar , low water - activity , peanut butter flavored candy fondant. Food Control , 27 (1), 184 187. https://doi.org/10.1016/j.foodcont.2011.11.037 Maillard browning reaction in foods, 28 (3). https://doi.org/10.1080/10408398909527499 Oliver, J. D. (2010). Recent fndings on the viable but nonculturable state in pathogenic bacteria. FEMS Microbiology Reviews , 34 , 415 425. https://doi.org/10.1111/j.1574 - 6976.2009.00200.x Ozturk, S., Kong, F., & Singh, R. K. (2020). Evaluation of Enterococcus faecium NRRL B - 2354 as a potential surrogate of Salmonella in packaged paprika, white pepper and cumin powder during radio frequency heating. Food Control , 108 . htt ps://doi.org/10.1016/j.foodcont.2019.106833 Ozturk, S., Liu, S., Xu, J., Tang, J., Chen, J., Singh, R. K., & Kong, F. (2019). Inactivation of Salmonella Enteritidis and Enterococcus faecium NRRL B - 2354 in corn flour by radio frequency heating with subsequ ent freezing. LWT - Food Science and Technology , 111 , 782 789. https://doi.org/10.1016/j.lwt.2019.04.090 Park, J. - K., Seok, W. - S., Choi, B. J., Kim, H. M., Yoon, S. - Salmonella enterica London infections associated with c onsumption of infant formula. Yonsei Medical Journal , 45 , 43 48. 233 Peleg, M., & Cole, M. (1998). Reinterpretation of Microbial Survival Curves. Critical Reviews in Food Science , 38 , 353 380. Pena - Melendez, M., Perry, J. J., & Yousef, A. E. (2014). Changes in thermal resistance of three Salmonella serovars in response to osmotic shock and adaptation at water activities reduced by different humectants. Journal of Food Protection , 77 (6), 914 918. https://doi.org/10.4315/0362 - 028X.JFP - 13 - 201 Pfeffer, J. M., Strating, H., Weadge, J. T., & Clarke, A. J. (2006). Peptidoglycan O acetylation and autolysin profile of Enterococcus faecalis in the viable but nonculturable state. J ournal of B acteriology , 188 (3), 902 908. https://doi.org/10.1128/JB.188.3.902 - 908.2006 Podolak, R., Black, D. G., & Wiley, J. (2017). Control of Salmonella and Other Bacterial Pathogens in Low - Moisture Foods . Retrieved from http://www.wiley.com/go/permissions. Podolak, R., En ache, E., Stone, W., Black, D. G., & Elliott, P. H. (2010). Sources and risk factors for contamination , survival , persistence , and heat resistance of Salmonella in low - moisture foods. Journal of Food Processing and Preservation , 73 (10), 1919 1936. https ://doi.org/10.4315/0362 - 028X - 73.10.1919 Poirier, D., Sanders, T. H., & Davis, J. P. (2014). Salmonella surrogate reduction using industrial peanut dry roasting parameters. Peanut Science , 41 , 72 84. Retrieved from http://www.peanutscience.com/doi/pdf/10.3 146/PS13 - 21.1 Porta, A., Török, Z., Horvath, I., Franceschelli, S., Vígh, L., & Maresca, B. (2010). Genetic modification of the Salmonella membrane physical state alters the pattern of heat shock response. JOURNAL OF BACTERIOLOGY , 192 (7), 1988. https://do i.org/10.1128/JB.00988 - 09 Rachon, G., Peñaloza, W., & Gibbs, P. A. (2016). Inactivation of Salmonella , Listeria monocytogenes and Enterococcus faecium NRRL B - 2354 in a selection of low moisture foods. International Journal of Food Microbiology , 231 , 16 2 5. https://doi.org/10.1016/j.ijfoodmicro.2016.04.022 Ramsey, M., Hartke, A., & Huycke, M. (2014). The physiology and metabolism of Enterococci [Internet]. In Y. Gilmore, M.S., Clewell, D.B., Ike (Ed.), Enterococci: from commensals to leading causes of dru g resistant infection (Vol. 1964, pp. 1 43). Boston, MA. Rodríguez - Urrego, J., Herrera - León, S., Echeita - Sarriondia, A., Soler, P., Simon, F., & Mateo, S. (2010). Nationwide outbreak of Salmonella serotype Kedougou associated with infant 234 formula, Spain, 2 008. Eurosurveillance , 15 (22), 1 5. https://doi.org/10.2807/ese.15.22.19582 - en Roos, Y. (1993). Melting and glass transitions of low molecular weight carbohydrates . Carbohydrate Research (Vol. 238). Roos, Y. H. (2009). Solid and Liquid States of Lactose. In P. F. McSweeney, P.L.H., Fox (Ed.), Advanced Dairy Chemistry (Vol. 3, pp. 17 34). Springer. https://doi.org/10.1007/978 - 0 - 387 - 84865 - 5 Rotenberg, B., Patel, A. J., & Chandler, D. (2011). Molecular explanation for why talc surfaces can be both hydrophilic and hydrophobic. J. Am. Chem. Soc , 133 , 20521 20527. https://doi.org/10.1021/ja208687a the clinically and veterinary significant pathogen Salmonella . BioMed Research International , 2017 , 1 6. https://doi.org/10.1155/2017/3782182 Sanders, T. H., & Calh oun, R. S. (2014). Effect of oil and dry roasting of peanuts at various temperatures and times on survival of Salmonella and Enterococcus faecium . Peanut Science , 41 , 65 71. Santagati, M., Campanile, F., Stefani, S., & Martinez, J. L. (2012). Genomic div ersification of enterococci in hosts: the role of the mobilome. Front. Microbiol , 3 , 1 9. https://doi.org/10.3389/fmicb.2012.00095 Santillana, S. M. (2014). Relative survival of four serotypes of Salmonella enterica in low - water activity whey protein powd er held at 36 and 70C at various water activity level. Journal of Food Protection , 7 , 1198 1200. Saunders, T., Wu, J., Williams, R. C., Huang, H., & Ponder, M. A. (2018). Inactivation of Salmonella and surrogate bacteria on cashews and macadamia nuts expo sed to commercial propylene oxide processing conditions. Journal of Food Protection , 81 , 417 423. https://doi.org/10.4315/0362 - 028X.JFP - 17 - 252 Scallan, E., Hoekstra, R.M., Angulo, F.J., Tauxe, R. V., Widdowson, M., Roy, S. L., Jones, J. L., Griffin, P. M. (2011). Foodborne illness acquired in the United States - major pathogens. Emerging Infectious Disease , 17 , 7 15. https://doi.org/10.3201/eid1701.P11101 235 Schleifer, Karl H, Killpper - Balz, R. (1984). Transfer of Streptococcus faecalis and Streptococcus faec ium to the genus Enterococcus nom. rev. as Enterococcus faecalis comb. nov. and Enterococcus faecium comb. nov. International Journal of Systematic Biology , 34 , 31 34 . Schmidt, SJ, Fontana Jr., A. (2007). Water Activity Values of Select Food Ingredients and Products. In Gustavo Barbosa - Canovas et al. (Ed.), Water Activity in Foods: Fundamentals and Applications (pp. 407 420). Oxford, UK: Blackwell Publishing Ltd. Retriev ed from https://onlinelibrary.wiley.com/doi/pdf/10.1002/9780470376454.app5 Schuck, P., Blanchard, E., Dolivet, A., Méjean, S., Onillon, E., & Jeantet, R. (2005). Water activity and glass transition in dairy ingredients. In Lait (Vol. 85, pp. 295 304). htt ps://doi.org/10.1051/lait:2005020 Shachar, D., & Yaron, S. (2006). Heat tolerance of Salmonella enterica serovars Agona , Enteritidis, and Typhimurium in peanut butter. Journal of Food Protection , 69 (11), 2687 2691. Shah, M. K., Asa, G., Sherwood, J., Gr aber, K., & Bergholz, T. M. (2017). Efficacy of vacuum steam pasteurization for inactivation of Salmonella PT 30 , Escherichia coli Enterococcus faecium on low moisture foods. International Journal of Food Microbiology , 244 , 111 118. https:// doi.org/10.1016/j.ijfoodmicro.2017.01.003 Sharma, A., Jana, A. H., & Chavan, R. S. (2012). Functionality of milk powders and milk - based powders for end use applications a review. Comprehensive Reviews in Food Science and Food Safety , 11 , 521 528. https: //doi.org/10.1111/j.1541 - 4337.2012.00199.x Sippola, H., & Taskinen, P. (2018). Activity of Supercooled Water on the Ice Curve and Other Thermodynamic Properties of Liquid Water up to the Boiling Point at Standard Pressure. Journal of Chemical and Engineering , 63 , 2986 2998. https://doi.org/10.1021/ acs.jced.8b00251 Smelt, J. P. P. M., & Brul, & S. (2014). Thermal Inactivation of Microorganisms. Critical Reviews in Food Science and Nutrition , 54 (10), 1371 1385. https://doi.org/10.1080/10408398.2011.637645 Smith, D. F., & Marks, B. P. (2015). Effect of rapid product desiccation or hydration on thermal resistance of Salmonella enterica serovar Enteritidis PT 30 in wheat flour. Journal of Food Protection , 78 (2), 281 286. https://doi.org/10.4315/0362 - 028X.JFP - 14 - 403 Storhaug, C. L., Fosse, S. K., & Fadnes, L. T. (2017). Country, regional, and global estimates for 236 lactose malabsorption in adults: a systematic review and meta - analysis. The Lancet Gastroenterology and Hepatology , 2 (10), 738 746. https://doi.org/10.1016/S 2468 - 1253(17)30154 - 1 Syamaladevi, R. M., Kiran, R., Xu, J., Villa - Water activity change at elevated temperatures and thermal resistance of Salmonella in all purpose wheat flour and peanut butter. Food R esearch International , 81 , 163 170. https://doi.org/10.1016/j.foodres.2016.01.008 Syamaladevi, R. M., Tang, J., Villa - Rojas, R., Sablani, S., Carter, B., & Campbell, G. (2016). Influence of water activity on thermal resistance of microorganisms in low - moi sture foods: a review. Comprehensive Reviews in Food Science and Food Safety , 15 (2), 353 370. https://doi.org/10.1111/1541 - 4337.12190 Syamaladevi, R. M., Tang, J., & Zhong, Q. (2016). Water Diffusion from a Bacterial Cell in Low - Moisture Foods. Journal of Food Science , 81 (9), 2129 2134. https://doi.org/10.1111/1750 - 3841.13412 Tadapaneni, R. K., Yang, R., Carter, B., & Tang, J. (2017). A new method to determine the water activity and the net isosteric heats of sorption for low moisture foods at elevated te mperatures. Food Research International , 102 , 203 212. https://doi.org/10.1016/j.foodres.2017.09.070 Tenario - Bernal, M., Marks, B. P., & Ryser, E. T. (2013). Evaluating the Predictive Ability of a Path - Dependent Thermal Inactivation Model for Salmonella S ubjected to Prior Sublethal Heating in Ground Turkey , Beef , and Pork. Journal of Food Protection , 76 (2), 220 226. https://doi.org/10.4315/0362 - 028X.JFP - 12 - 279 Thomas, Marie E C, Scher, J, Desobry - effec t on physical and functional. Critical Reviews in Food Science and Nutrition , 8398 , 297 322. https://doi.org/10.1080/10408690490464041 P. (1998). Molecular fin gerprinting defines a strain of Salmonella enterica serotype Anatum responsible for an international outbreak associated with formula - dried milk. Epidemiology and Infection , 121 (2), 289 293. https://doi.org/10.1017/S0950268898001149 Tsai, H. - C., Ballom, K . F., Xia, S., Tang, J., Marks, B. P., & Zhu, M. - J. (2019). Evaluation of Enterococcus faecium NRRL B - 2354 as a surrogate for Salmonella during cocoa powder thermal processing. Food Microbiology , 82 , 135 141. https://doi.org/10.1016/j.fm.2019.01.005 237 U.S. D epartment of Agriculture. (2017). Salmonella compliance guidelines for small and very small meat and poultry establishments that produce ready - to - eat (RTE) products and revised APPENDIX A . Retrieved from https://www.fsis.usda.gov/wps/wcm/connect/bf3f01a1 - a0b7 - 4902 - a2df - a87c73d1b633/ Salmonella - Compliance - Guideline - SVSP - RTE - APPENDIX - A.pdf?MOD=AJPERES U.S. Department of Agriculture. (2019). U.S. exports: tree nuts . Retrieved from https://data.ers.usda.gov/re ports.aspx?programArea=fruit&groupName=Tree nuts&reportPath=/FruitAndVeg/MarketSegmentSummary_exp&ID=17886#P9dae959842e 54aefa8699ca533e2059e_2_26iT0R0x0 U.S. Food and Drug Administration. (2014). Water Activity (aw) in Foods. Retrieved April 29, 2019, from https://www.fda.gov/inspections - compliance - enforcement - and - criminal - investigations/inspection - technical - guides/water - activity - aw - foods U.S. Food and Drug Administration. (2017a). Recalls, market withdrawals, safety alerts - Conagra brands recal Salmonella in spice packet. Retrieved November 10, 2018, from https://www.fda.gov/Safety/Recalls/ucm550184.htm U.S. Food and Drug Administration. (2017b). Recalls, market withdrawals, safety alerts - Spice ly Organics recalls organic tarragon because of possible health risk. Retrieved November 10, 2018, from https://www.fda.gov/Safety/Recalls/ucm580600.htm U.S. Food a nd Drug Administration. (2018a). Food Safety Modernization Act (FSMA) - FSMA final rule for preventive controls for human food . Silver Spring. https://doi.org/https://www.fda.gov/food/guidanceregulation/fsma/ucm334115.htm U.S. Food and Drug Administration . (2018b). Recalls, market withdrawals, safety alerts - AMPI recalls limited amount of dry whey powder because of possible health risk. Retrieved October 17, 2018, from https://www.fda.gov/Safety/Recalls/ucm614759.htm U.S. Food and Drug Administration. (2 018c). Recalls, market withdrawals, safety alerts - Bazzini LLC recalls certain pistachio products because of possible health risk. Retrieved October 17, 2018, from https://www.fda.gov/Safety/Recalls/ucm622867.htm U.S. Food and Drug Administration. (2018d ). Recalls, market withdrawals, safety alerts - 238 International Harvest, Inc. recalls organic Go Smile! raw coconut because of possible health risk. Retrieved November 10, 2018, from https://www.fda.gov/Safety/Recalls/ucm601529.htm U.S. Food and Drug Admini stration. (2018e). Recalls, market withdrawals, safety alerts - Pepperidge Farm® announces voluntary recall of four varieties of Goldfish® crackers. Retrieved October 17, 2018, from https://www.fda.gov/Safety/Recalls/ucm614548.htm U.S. Food and Drug Admin istration. (2018f). Recalls, market withdrawals, safety alerts - Raws For Paws recalls turkey pet food because of possible Salmonella health risk. Retrieved from https://www.fda.gov/Safety/Recalls/ucm596043.htm U.S. Food and Drug Administration. (2019a). FDA Investigated Recalled Duncan Hines Cake Mixes Potentially Linked to Salmonella Agbeni Illnesses | FDA. Retrieved November 4, 2019, from https://www.fda.gov/food/outbreaks - foodborne - illness/fda - investigated - recalled - duncan - hines - cake - mixes - potentially - l inked - Salmonella - agbeni - illnesses U.S. Food and Drug Administration. (2019b). Hometown Food Company Recalls Two Production LOT Codes of Pillsbury® Unbleached All - Purpose 5lb Flour Due to Possible Health Risk | FDA. Retrieved November 4, 2019, from https:/ /www.fda.gov/safety/recalls - market - withdrawals - safety - alerts/hometown - food - company - recalls - two - production - lot - codes - pillsburyr - unbleached - all - purpose - 5lb - flour Van Asselt, E. D., & Zwietering, M. H. (2006). A systematic approach to determine global therma l inactivation parameters for various food pathogens. International Journal of Food Microbiology , 107 (1), 73 82. https://doi.org/10.1016/j.ijfoodmicro.2005.08.014 Van Boekel, M. A. J. S. (2002). On the use of the Weibull model to describe thermal inactivation of microbial vegetative cells. International Journal of Food Microbiology , 74 (1 2), 139 159. https://doi.org/10.1016/S0168 - 1605(01)00742 - 5 Venkatachalam, M., & Sathe, S. K. (2006). Chemical composition of selected edible nut seeds. Agricultural and Food Chemistry , 54 , 4705 4714. https://doi.org/10.1021/jf0606959 Villa - inactivation of Salmonella Enteritidis PT 30 in almond kernels as influenced by water activity. Journal of Food Protection , 76 (1), 26 32. https://doi.org/10.4315/0362 - 028X.JFP - 11 - 509 Villa - Rojas, R., Zhu, M., Marks, B. P., & Tang, J. (2017). Radiofrequency inactivation of 239 Salmonella Enteritidis PT 30 and Enterococcus faecium in wheat flour at different water activities. Biosystems Engineering , 156 , 7 16. https://doi.o rg/10.1016/j.biosystemseng.2017.01.001 Wei, X., Lau, S. K., Stratton, J., Irmak, S., Bianchini, A., & Subbiah, J. (2018). Radio - frequency processing for inactivation of Salmonella enterica and Enterococcus faecium NRRL B - 2354 in black peppercorn. Journal of Food Protection , 81 (10), 1685 1695. https://doi.org/10.4315/0362 - 028X.JFP - 18 - 080 Weissfeld, A. S. (2010). Estimation of uncertainty of measurement in microbiology. Clinical Microbiology Newsletter , 32 (22), 171 175. https://doi.org/10.1016/j.clinmicnews .2010.10.004 Welch, B. L. (1938). The significant of the difference between two means when population variances are unequal. Biometrika , 29 , 350 362. Retrieved from https://about.jstor.org/terms Wiertzema, J. R., Borchardt, C., Beckstrom, A. K., Dev, K., J. (2019). Evaluation of m ethods for i noculating d ry p owder f oods with Salmonella enterica, Enterococcus faecium , or Cronobacter sakazakii . Journal of Food Protection , 82 , 1082 1088. https://doi.org/10.4315/0362 - 028X.JFP - 18 - 284 World Health Organization, & Food and Agriculture Organization of the United Nations. (2011). Codex Alimentarus: Milk and Milk Products . Retrieved from http://www.fao.org/3/i2085e/i2085e00.pdf World Heatlh Organization. (2018). Salmonella (non - typhoidal). Retrieved May 4, 2019, from https://www.who.int/news - room/fact - sheets/detail/ Salmonella - (non - typhoidal) Wright, D. G., Minarsich, J., & Daeschel, M. A. (2018). Thermal inactivation of Salmonel la spp. in commercial tree nut and peanut butters in finished packaging. Journal of Food Safety , (December 2016), 1 6. https://doi.org/10.1111/jfs.12371 Yada, S., & Harris, L. J. (2018). Recalls of tree nuts and peanuts in the U.S., 2001 to present (versi on 2) [Table and references]. Retrieved October 17, 2018, from http://ucfoodsafety.ucdavis.edu/Nuts_and_Nut_Pastes. Yada, S., & Harris, L. J. (2019). Recalls of tree nuts and peanuts in the U.S., 2001 to present (version 2) [Table and references] . Retriev ed from https://ucfoodsafety.ucdavis.edu/low - moisture - foods/nuts - and - nut - pastes. 240 Young, I. A. N., Waddell, L., Cahill, S., Kojima, M., & Clarke, R. (2015). Application of a Rapid Knowledge Synthesis and Transfer Approach To Assess the Microbial Safety of L ow - Journal of Food Protection , 78 (12), 2264 2278. https://doi.org/10.4315/0362 - 028X.JFP - 15 - 146 Development and v alidation of a c ultural m ethod for the d etection and i solation of Salmonella in c loves, 80 (3), 376 382. https://doi.org/10.4315/0362 - 028X.JFP - 16 - 376