DETECTION, QUANTIFICATION AND BIOLOGICAL CONTROL OF POST- HARVEST PATHOGENS OF POME FRUIT by Daylin Lindsay Mantyka BSc, The University of Victoria, 2006 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The College of Graduate Studies (Biology) THE UNIVERSITY OF BRITISH COLUMBIA (Okanagan) April 2010 © Daylin Lindsay Mantyka, 2010 ii ABSTRACT Blue mold (Penicillium expansum), grey mold (Botrytis cinerea) and mucor rot (Mucor piriformis) are important post-harvest diseases of pome fruit in British Columbia and throughout the world causing annual losses of 5 – 20%. Identification and quantification using novel DNA macroarray technology may assist in the development of prediction models and disease forecasting. Post-harvest pathogens were monitored and quantified throughout the growing season in four apple orchards in the Okanagan Valley, BC in 2007 and 2008. Their detection was variable due to field and year differences. The average percent detection of P. expansum (27.4%) and M. piriformis (19.2%) was higher than that of B. cinerea (6.2%). There was a positive correlation between total post-harvest pathogen detection in aerial samples just prior to harvest and after harvest in naturally infected fruit (r = 0.74; p = 0.09). Pseudomonas fluorescens (isolates 1-112, 2- 28, 4-6) and Serratia plymuthica (isolate 6-25), isolated from the rhizosphere of legumes, were investigated for their biological control capabilities in semi-commercial storage conditions at 1°C in air and commercial storage conditions in controlled atmosphere and with 1-methylcyclopropene (1-MCP) application. Overall, isolate 6-25 provided control in the greatest number of treatments (51.7%) while isolate 1-112 provided the greatest level of control (75.8%) in treatments where significant control was exhibited. Isolate 4- 6 was tagged with green fluorescent protein to gain insight into bacterial antagonist population and survival dynamics. Alone, its population increased 10 fold after 30 d in storage at 1°C and then decreased to concentrations similar to those at inoculation. In the presence of the pathogen, 4-6-gfp increased then decreased after 30 d in storage at 1°C to undetectable amounts. These data provide greater insight into the prediction, control and iii population dynamics of pathogens and biological control agents as a means of preventing and controlling post-harvest storage diseases in pome fruit. iv TABLE OF CONTENTS ABSTRACT ....................................................................................................................... ii TABLE OF CONTENTS ................................................................................................. iv LIST OF FIGURES .......................................................................................................... vi LIST OF TABLES ......................................................................................................... viii LIST OF ABBREVIATIONS ....................................................................................... xiii ACKNOWLEDGMENTS .............................................................................................. xiv CHAPTER 1: INTRODUCTION ..................................................................................... 1 1.1 Post-harvest pathogens ........................................................................................ 2 1.1.1 Penicillium expansum .................................................................... 3 1.1.2 Botrytis cinerea .............................................................................. 4 1.1.3 Mucor piriformis ............................................................................ 5 1.2 Post-harvest factors that influence post-harvest pathology ................................. 6 1.2.1 Temperature ................................................................................... 7 1.2.2 Controlled atmosphere ................................................................... 7 1.2.3 1-Methylcyclopropene ................................................................... 8 1.3 Methods of detection and identification .............................................................. 8 1.4 Methods of control ............................................................................................ 10 1.4.1 Chemical control .......................................................................... 10 1.4.2 Biological control ......................................................................... 10 1.5 Mechanisms of action ........................................................................................ 13 1.5.1 Antibiosis ..................................................................................... 13 1.5.2 Competition for nutrients ............................................................. 14 1.5.3 Parasitism ..................................................................................... 15 1.5.4 Induced systemic resistance ......................................................... 15 1.6 Green fluorescent protein .................................................................................. 16 1.7 Objectives .......................................................................................................... 18 CHAPTER 2: MATERIALS AND METHODS ........................................................... 20 2.1 Antagonists ........................................................................................................ 20 2.2 Pathogens .......................................................................................................... 20 2.3 Orchard field sites ............................................................................................. 21 2.4 pAG408 ............................................................................................................. 23 2.5 Construction of the gfp-transconjugant ............................................................. 23 2.6 Plasmid stability ................................................................................................ 25 2.7 P. fluorescens 4-6-gfp fitness analysis .............................................................. 25 2.8 Spore, apple tissue collection and fruit washings.............................................. 26 2.9 Validation and field testing of DNA macroarrays............................................. 27 2.9.1 DNA extraction ............................................................................ 27 v 2.9.2 PCR amplification ........................................................................ 28 2.9.3 DNA macroarray construction ..................................................... 29 2.9.4 Hybridization and chemiluminescent detection ........................... 29 2.10 Biological control of post-harvest fungal pathogens ......................................... 33 2.10.1 Inoculation of apples by drenching .............................................. 33 2.10.2 Natural disease incidence ............................................................. 33 2.10.3 Natural disease incidence with bacterial antagonists ................... 34 2.10.4 Biological control of post-harvest pathogens ............................... 35 2.10.5 Bacterial survival on apple ........................................................... 35 2.11 Analysis ............................................................................................................. 36 CHAPTER 3: RESULTS ................................................................................................ 38 3.1 Validation and field testing of DNA macroarrays............................................. 38 3.1.1 Quantification of post-harvest pathogens ..................................... 38 3.1.2 Field application of macroarray technology................................. 43 3.2 Natural disease incidence .................................................................................. 53 3.2.1 Harvest 2007 ................................................................................ 53 3.2.2 Harvest 2008 ................................................................................ 54 3.3 Growing season detection and natural disease incidence .................................. 57 3.4 Antagonist efficacy, harvest 2007 ..................................................................... 58 3.4.1 Natural disease incidence and biocontrol inoculations ................ 58 3.4.2 Post-harvest pathogen and biocontrol inoculations ...................... 58 3.5 Antagonist efficacy, harvest 2008 ..................................................................... 62 3.5.1 Natural disease incidence and biocontrol inoculations ................ 64 3.5.2 Post-harvest pathogen and biocontrol inoculations ...................... 67 3.6 Bacterial survival on apple using GFP .............................................................. 75 3.6.1 Plasmid confirmation ................................................................... 75 3.6.2 Confirmation of gfp transformation ............................................. 75 3.6.3 Plasmid stability ........................................................................... 75 3.6.4 P. fluorescens 4-6-gfp fitness analysis ......................................... 80 3.6.5 Bacterial survival on apple ........................................................... 80 CHAPTER 4: DISCUSSION .......................................................................................... 87 4.1 DNA macroarray for post-harvest apple pathogens .......................................... 87 4.1.1 DNA macroarray quantification ................................................... 87 4.1.2 Field monitoring of apple post-harvest pathogens ....................... 89 4.2 Biological control of post-harvest pathogens .................................................... 93 4.2.1 Bacterial antagonists and CA storage........................................... 95 4.2.2 Bacterial antagonists and 1-MCP ................................................. 98 4.3 Bacterial survival on apple .............................................................................. 100 CHAPTER 5: CONCLUSIONS ................................................................................... 103 REFERENCES ............................................................................................................... 105 APPENDIX ..................................................................................................................... 122 vi LIST OF FIGURES Figure 1. Average daily temperatures and total daily precipitation at Kelowna, BC and Summerland, BC during the growing seasons of 2007 and 2008 ..................................... 22 Figure 2. DNA macroarray template ............................................................................... 31 Figure 3. Quantitative assessment of Penicillium expansum probe PE-H3u illustrating the relationship between greyscale value and DNA concentration ........................................ 39 Figure 4. Quantitative assessment of Botrytis cinerea probe BC-H2d illustrating the relationship between greyscale value and DNA concentration ........................................ 40 Figure 5. Quantitative assessment of Mucor piriformis probe Mpir-ITS-414H1 illustrating the relationship between greyscale value and DNA concentration ................ 41 Figure 6. Concentrations of P. expansum, B. cinerea and M. piriformis from aerial and plant tissue samples collected from Field 3 in 2007 and 2008 using DNA macroarrays . 44 Figure 7. Concentrations of P. expansum, B. cinerea and M. piriformis from aerial and plant tissue samples collected from Field 12 in 2007 and 2008 using DNA macroarrays 45 Figure 8. Concentrations of P. expansum, B. cinerea and M. piriformis from aerial and plant tissue samples collected from the Kiran orchard in 2007 and 2008 using DNA macroarrays ....................................................................................................................... 46 Figure 9. Concentrations of P. expansum, B. cinerea and M. piriformis from aerial and plant tissue samples collected from the Reekie orchard in 2007 and 2008 using DNA macroarrays ....................................................................................................................... 47 Figure 10. Natural disease incidence in Jonagold apples grown in Field 3, Gala apples grown in Field 12 and the Kiran Orchard and Red Delicious apples grown in the Reekie orchard and stored for 4.5 months in CA .......................................................................... 56 Figure 11. Effect of 1-MCP and biological control isolates on natural disease incidence in Gala apples grown in Field 12 (harvest 2007) and stored for 3 months in air at 1°C .. 59 Figure 12. Effect of 1-MCP and biological control isolates on infection severity in Jonagold apples from Field 3 (harvest 2007), inoculated with P. expansum, B. cinerea and M. piriformis and stored for 3 months at 1°C in air ................................................... 61 Figure 13. Effect of 1-MCP and biological control isolates on infection severity in Gala apples from Field 12 (harvest 2007), inoculated with M. piriformis and stored for 3 months at 1°C in air. ......................................................................................................... 63 Figure 14. Average infection severity for all wounded and non-wounded apples from individual biological control treatments conducted in 2008 ............................................. 65 vii Figure 15. Effect of biological control isolates on natural disease incidence in Jonagold apples grown in Field 3 and Gala apples grown in Field 12 (harvest 2008) and stored for 2, 4 or 6 months at 1°C in air ............................................................................................ 66 Figure 16. Effect of biological control isolates on natural disease incidence in Jonagold apples grown in Field 3 (harvest 2008) and stored for 4.5 months in CA ........................ 68 Figure 17. Effect of biological control isolates on natural disease incidence in Gala apples grown in the Kiran orchard (harvest 2008) and stored for 4.5 months in CA ....... 69 Figure 18. Effect of biological control isolates on infection severity in Jonagold apples grown in Field 3 (harvest 2008), inoculated with P. expansum and B. cinerea and stored for 2 months at 1°C in air and inoculated with M. piriformis and stored for 1 month at 1°C in air ........................................................................................................................... 70 Figure 19. Effect of biological control isolates on infection severity in Gala apples grown in Field 12 (harvest 2008), inoculated with P. expansum and B. cinerea and stored for 2 months at 1°C in air and inoculated with M. piriformis and stored for 1 month at 1°C in air ...................................................................................................................................... 71 Figure 20. HindIII restriction digest of gfp-containing plasmid, pAG408 ...................... 76 Figure 21. PCR amplification of gfp in P. fluorescens isolate 4-6 transformed with pAG408 using gfp-specific primers .................................................................................. 77 Figure 22. Fluorescence microscopy of P. fluorescens isolate 4-6 transformed with pAG408 exhibiting green fluorescence ............................................................................ 78 Figure 23. Fluorescence microscopy of an apple wound inoculated with P. fluorescens isolate 4-6-gfp ................................................................................................................... 79 Figure 24. Growth characteristics of P. fluorescens 4-6-wild-type and P. fluorescens 4-6- gfp in half-strength TSB at 28°C for 24 h ......................................................................... 81 Figure 25. Relationship between P. fluorescens isolate 4-6-gfp cell density and relative fluorescent units as determined by dilution plating and direct scanning. ......................... 82 Figure 26. Bacterial survival of 4-6-gfp, P. expansum and P. expansum + 4-6-gfp on Gala apples stored for 55 d at 1°C in air using the direct scanning method ..................... 83 Figure 27. Bacterial survival of 4-6-gfp and 4-6-gfp + P. expansum on Gala apples stored for 55 d at 1°C in air using the dilution plating method ................................................... 86 viii LIST OF TABLES Table 1. Frequency of aerial and plant tissue sampling for Fields 3, 12 and the Kiran and Reekie orchards in the early, middle and late growing seasons of 2007 and 2008. ......... 27 Table 2. DNA macroarray probe sequences..................................................................... 30 Table 3. DNA macroarray detection summary of P. expansum, B. cinerea and M. piriformis from four apple orchards throughout the 2007 and 2008 growing seasons ..... 50 Table 4. Detection of P. expansum, B. cinerea and M. piriformis from fruit washings collected from Jonagold apples from Field 3, Gala apples from Field 12 and the Kiran orchard and Red Delicious apples from the Reekie orchard in 2007 and 2008 using DNA macroarrays ....................................................................................................................... 53 Table 5. Natural disease incidence indicated by infection severity in Jonagold apples grown in Field 3 (harvest 2007), treated or not treated with 1-MCP and stored for 3 or 6 months at 1°C in air .......................................................................................................... 54 Table 6. Natural disease incidence indicated by infection severity in Gala apples grown in Field 12 (harvest 2007), treated or not treated with 1-MCP and stored for 3 or 6 months at 1°C in air .......................................................................................................... 54 Table 7. Natural disease incidence indicated by infection severity in Jonagold apples grown in Field 3, Gala apples grown in Field 12 and the Kiran orchard and Red Delicious apples grown in the Reekie orchard (harvest 2008) and stored for 1, 2, 4 or 6 months at 1°C in air ........................................................................................................................... 55 Table 8. Summary of antagonist efficacy experiments for harvest 2007 and 2008 ......... 73 Table 9. Sole carbon source utilization by P. fluorescens isolates 4-6-wild-type and 4-6- gfp as indicated by Biolog Phenotye MicroArrays™ for sugars and organic acids commonly found in apple juice ........................................................................................ 80 Table 10. Effect of biological control isolate 4-6-gfp on infection severity in Gala apples inoculated with and without P. expansum and stored for 55 days at 1°C in air. .............. 85 Table A1. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. ............................................................................................. 122 Table A2. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. ............................................................................................. 122 ix Table A3. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. ....................................................................................... 122 Table A4. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. ............................................................................................. 123 Table A5. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. ............................................................................................. 123 Table A6. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. ....................................................................................... 123 Table A7. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in all growing season sections for Fields 3, 12 and the Kiran and Reekie orchards. .............................................................................................................. 124 Table A8. Key data extracted from the two-way ANOVA table for IS data collected from Jonagold apples grown in Field 3 and analyzed for NDI after harvest in 2007 ............. 124 Table A9. Key data extracted from the two-way ANOVA table for IS data collected from Gala apples grown in Field 12 analyzed for NDI after harvest in 2007 ......................... 124 Table A10. The effect of incubation period on NDI and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3, Gala apples grown in Field 12 and Kiran and Red Delicious apples grown in Reekie and analyzed for NDI from harvest after 2008. ....................... 125 Table A11. Key data extracted from the one-way ANOVA table constructed from IS data collected from Fields 3, 12, Kiran and Reekie analyzed for NDI for apples stored in CA storage for 4.5 months from harvest 2008 ...................................................................... 125 Table A12. Key data extracted from the one-way ANOVA table constructed from IS data collected from Fields 3, 12, Kiran and Reekie analyzed for NDI for apples stored in CA storage for 4.5 months from harvest 2008 ...................................................................... 125 Table A13. Key data extracted from the two-way ANOVA table for IS data collected from Jonagold apples grown in Field 3 analyzed for NDI in combination with bacterial isolates after harvest in 2007 .......................................................................................... 126 x Table A14. Key data extracted from the two-way ANOVA table for IS data collected from Gala apples grown in Field 12 and analyzed for NDI in combination with bacterial isolates after harvest in 2007 .......................................................................................... 126 Table A15. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for NDI in combination with bacterial isolates from harvest 2007 ............................................................................... 126 Table A16. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are constructed from IS results collected from Field 12 Gala apples and analyzed for NDI in combination with bacterial isolates from harvest 2007 ...................................... 127 Table A17. Key data extracted from the two-way ANOVA table from IS data collected from Jonagold apples grown in Field 3 and inoculated with P. expansum after harvest in 2007 ................................................................................................................................ 127 Table A18. The effect of 1-MCP in six separate biological control categories and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with P. expansum and bacterial isolates after harvest in 2007 ...................... 127 Table A19. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are constructed from IS results collected from Jonagold grown in Field 3 analyzed for disease incidence when inoculated with P. expansum and bacterial isolates from harvest in 2007 ................................................................................................................ 128 Table A20. Key data extracted from the two-way ANOVA table from IS data collected from Jonagold apples grown in Field 3 and inoculated with B. cinerea after harvest in 2007 ................................................................................................................................ 128 Table A21. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2007 .......................... 128 Table A22. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2007. ........................................................................................................................... 129 Table A23. Key data extracted from the two-way ANOVA table constructed from IS data collected from Jonogold apples grown in Field 3 and inoculated with M. piriformis after harvest in 2007 ....................................................................................................... 129 xi Table A24. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data is are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007 ........................................................................................................................................ 129 Table A25. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. ............................................................................................................... 130 Table A26. Key data extracted from the two-way ANOVA table constructed from IS data collected from Gala apples from Field 12 and inoculated with P. expansum after harvest in 2007 ................................................................................................................ 130 Table A27. Key data extracted from the two-way ANOVA table from IS data collected from Gala apples grown in Field 12 and inoculated with M. piriformis after harvest in 2007. ............................................................................................................................... 130 Table A28. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007 ..................... 131 Table A29. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. ........................................................................................................................... 131 Table A30. Key data extracted from the two-way ANOVA table from IS data collected from Jonagold apples grown in Field 3 and analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2008 ............................................... 131 Table A31. Key data extracted from the two-way ANOVA table from IS data collected from Gala apples grown in Field 12 and analyzed for NDI in combination with bacterial isolates after harvest in 2008 .......................................................................................... 132 Table A32. The effect of the incubation period on biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for NDI when inoculated the bacterial isolates after harvest in 2008 ...................................................................... 132 Table A33. The effect of the incubation period on biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for NDI when inoculated the bacterial isolates after harvest in 2008 ............................................................................ 132 xii Table A34. Key data extracted from the two-way ANOVA table from IS data collected from Fields 3 and 12 and the Kiran and Reekie orchards after 4.5 months in CA storage and analyzed for natural disease incidence following inoculation with bacterial isolates after harvest in 2008 ....................................................................................................... 133 Table A35. The effect of the biological control treatments on IS and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3, Gala apples grown in Field 12 and Kiran and Red Delicious apples grown in Reekie after 4.5 months in CA stroage and analyzed for NDI when inoculated the bacterial isolates after harvest in 2008 .......................................... 133 Table A36. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with P. expansum and bacterial isolates after harvest in 2008 ............................................................................................................................ 133 Table A37. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2008 ................................................................................................................................ 134 Table A38. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2008 ............................................................................................................................ 134 Table A39. Growth rate comparisons of P. fluorescens isolates 4-6-wild-type and 4-6-gfp and the resulting key data extracted from the one-way ANOVA table .......................... 134 Table A40. Key data extracted from the two-way ANOVA table constructed from cell densities of P. fluorescens isolates 4-6-wild-type and 4-6-gfp stored for 55 d at 1°C in air ........................................................................................................................................ 135 Table A41. Cell density comparisons determined by dilution plating method of P. fluorescens isolates 4-6-gfp and 4-6-gfp + P. expansum stored for 55 d at 1°C in air and the resulting key data extracted from the one-way ANOVA table ................................. 135 Table A42. Infection severity comparisons of P. fluorescens isolates 4-6-wild-type and 4-6-gfp stored for 55 d at 1°C in air and the resulting key data extracted from the one-way ANOVA table ................................................................................................................. 135 xiii LIST OF ABBREVIATIONS * (one asterisk) – pvalue = 0.05 – 0.01 ** (two asterisks) – pvalue = 0.01 – 0.001 *** (three asterisks) – pvalue = < 0.001 ½ TSA – half-strength tryptic soy agar ½ TSB – half-strength tryptic soy broth 1-MCP – 1-methylcyclopropene 4-6-gfp – transformed P. fluorescens isolate 4-6 with GFP-containing plasmid pAG408 bc – biocontrol BC – British Columbia BCA – biological control agent Bcin – Botrytis cinerea βtub – β tubulin CA – controlled atmosphere CFU – colony forming units CSLM – confocal scanning laser microscopy gen - gentamicin gfp – green fluorescent protein (gene) GFP – green fluorescent protein (protein) IS – infection severity ISR – induced systemic resistance kan - kanamycin log10 – logarithmic to the base 10 LSD – least significant difference MOA – mode of action Mpir – Mucor piriformis NDI – natural disease incidence OD600 – optical density at 600 nm PARC –Pacific Agri-Food Research Centre PBS – phosphate buffered saline PCR – polymerase chain reaction PDA – potato dextrose agar Pex – Penicillium expansum pt – plant tissue rDNA – ribosomal deoxyribonucleic acid RFU – relative fluorescent units TBZ – thiabendazole TSA/B – trypic soy agar/broth xiv ACKNOWLEDGMENTS I would like to firstly thank my primary supervisor, Dr. Louise Nelson for giving me this wonderful opportunity. The applied nature of this program helped me realize where I belong within the scientific community. I would also like to thank my co- supervisor, Dr. Peter Sholberg, for sharing your knowledge and expertise in the area of post-harvest research and the use of the facilities at the Pacific Agri-Food Research Centre (PARC). To the members of my committee, Drs. Dan Durall and Mary Forrest, I appreciate your feedback, patience and encouragement. To my uncle, John Virgl, thank- you for lending me your statistical know-how. I think that I would still be lost without your advice. This project would not have been possible without the generosity of the Advanced Foods and Materials Network STAR/Commercialization program. In addition, I am grateful to the BC Innovation Council for awarding me the BC Innovation Scholars Scholarship. I look forward to collaborating with my business partner, Jenn Schofield, and learning the business of science. I would like to send a mountain of gratitude to Dr. Danielle Hirkala of the Okanagan Tree Fruit Cooperative for access to her bounty of knowledge and for all her help in the lab and in all areas of this project. She was a huge part of my success and I have learned a great deal from her. Thank you for your sense of humour and expertise. To the rest of the Nelson lab- Melissa, Natashia, DeAnna and Chris, your assistance has been greatly appreciated. Thank you to Michael Wies of PARC for teaching me the ins and outs of the confocal microscope and to Dr. Andis Klegeris and his lab group for access to the microplate reader. To fellow graduate students, thank you for providing me xv with tidbits of science and sanity along the way. Finally, I would like to thank my family, especially my parents, Bob and Janet, for their unending support and pushing me to ‘split the apple’. To Barry, my partner in crime, your enquiring mind, patience, support and love has got me through it all. Thanks. 1 CHAPTER 1: INTRODUCTION According to the Food and Agriculture Organization of the United Nations, the world apple production in 2008 was greater than 69 million tonnes (FAOSTAT 2009). Apples (Malus domestica) are ranked fourth (by volume) in the world’s production of fruit crops (BCMAFF 2004). In 2002, the British Columbia apple industry produced 136 000 tonnes or approximately 299 846 000 lbs of apples. Apples are produced in significantly higher volumes than any other tree fruit crop in BC (BCMAFF 2004). One major challenge to apple production is damage incurred by post-harvest diseases; annual losses can vary from 5 – 20% and up to 50% in developing countries (Janisiewicz and Korsten 2002). A variety of fungal pathogens is responsible for post-harvest disease; in BC, three such pathogens include Penicillium expansum (blue mold), Botrytis cinerea (grey mold) and Mucor piriformis (mucor rot) (Sholberg and Haag 1996). A common method of disease prevention is through the use of fungicides. In Canada, two synthetic controls are registered for post-harvest use, Mertect® (thiabendazole) and Scholar® (fludioxonil) (PMRA 2007). However, due to growing health issues, the development of pathogen resistance and the demand for environmentally friendly sustainable practices, pesticides are now a major cause for concern (Errampalli et al. 2006; Michaildes and Spotts 1990; Sholberg and Haag 1996). To reduce post-harvest rot, improved methods for pathogen identification prior to harvest and alternate technologies to chemical control are required to assist in disease prediction and control. Preliminary work has resulted in two novel approaches that address these demands: 1) a rapid DNA-based macroarray system that determines the presence of fungal pathogens responsible for post-harvest rot (Sholberg et al. 2005a) and 2 2) the identification of soil bacteria that can control fungal post-harvest pests (Hirkala et al. 2007; Hynes et al. 2008). This study will assist the project’s long term goal to develop Canadian-based technologies that detect, predict and mitigate post-harvest disease in pome fruit and that are consistent with integrated pest management and sustainable production practices. 1.1 Post-harvest pathogens Fruit infections can occur throughout the growing season, at harvest, during storage, transit and at the retail level (Jones and Aldwinckle 1990). Although global economic losses can be attributed to growing season diseases, post-harvest pathogens, despite modern storage facilities, can cause significant annual losses (Janisiewicz and Korsten 2002). Post-harvest diseases, or latent infections, are asymptomatic throughout the growing season and disease is manifested only after periods of storage. As a result, these pathogens are often difficult to treat and prevent. Post-harvest fungal pathogens can be categorized in many ways, one of which describes the pathogenic point of entry. Rot characterization can include lenticel infections (pores that exist on the surface of the apple that facilitate gas exchange), core and eye rots and wound pathogens (Jijakli and Lepoivre 2004). Common lenticel infections include bitter and bull’s eye rot and are commonly caused by Colletotrichum gleosporioides and Cryptosporiopsis curvispora, respectively (Jijakli and Lepoivre 2004). Core rots develop when an open sinus stems from the calyx end of the apple into the core (Miller 1959), which provides a point of entry for an invading pathogen. Alternaria spp. are the most commonly isolated fungi from core rots, but other fungal species such as Cladosporium, Botrytis, Candida and 3 Fusarium have also been identified (Ellis and Barrat 1983). Dry eye rots, characterized by a shallow, hard rot with a red border are predominantly caused by Botrytis cinerea. Opportunistic wound pathogens invade apples as a result of fruit injury incurred by mishandling or harsh weather conditions. These types of infections are responsible for significant storage losses. Three important fungal pathogens that cause worldwide decay include Penicillium expansum (blue mold), Botrytis cinerea (grey mold) and to lesser extent, Mucor piriformis (mucor rot) (Michailides and Spotts 1990; Sanderson and Spotts 1995; Rosenberger 1990). 1.1.1 Penicillium expansum Penicillium expansum Link, or apple blue mold, is a filamentous Ascomycete that causes one of the most important North American post-harvest storage diseases of pome fruit. Before the introduction of controlled atmosphere (CA) storage and fungicides, it accounted for 90% of post-harvest apple diseases (Rosenberger 1990). P. expansum produces the mycotoxin, patulin, which has been found in apple product derivatives such as apple juice, ciders, puree, vinegar, baby food and whole apples (Abramson et al. 2009; Doores 1983; Piemontese et al. 2005; Watanabe and Shimizu 2005). Its detection and control is, therefore, of utmost importance for food health and safety. Although an opportunistic wound pathogen, P. expansum infection can originate from stem-end invasions, core rots and through lenticels. Infected fruit are characterized by light to dark brown, fleshy, circular lesions that surround wounds. Older lesions may produce a bluish spore mass. P. expansum can also form a dense powdery mass at the centre of the lesion (Rosenberger 1990) and can be characterized by an earthy, musty odour commonly used as a diagnostic tool (Vikram et al. 2004). 4 Spores of P. expansum are ubiquitous and can cause infection within orchards and packinghouses. This fungus can survive in organic debris on the orchard floor and within soil. Conidia also exist in the air and on apple surfaces (Lennox et al. 2003). Within packinghouses, spores have been isolated from fungicide-drench solutions, flume water, dump-tank water, air and storage room walls (Sholberg and Stokes 2006). Conidia can survive from season to season on contaminated bins, picking boxes and storage walls (Rosenberger 1990). Sanitation, harvesting pre-senescent fruit and handling methods are the best strategy for minimizing disease incidence. This control can also be facilitated by fungicides applied as pre-harvest sprays or post-harvest dips (Eckert and Ogawa 1988). For example, pyrimethanil was used as a pre-harvest fungicide and applied twenty days prior to apple harvest. After apples were stored for six months, post-harvest blue mold was significantly reduced when compared to apples that were not treated with fungicides (Sholberg et al. 2005b). Mertect®, with an active ingredient of thiabendazole (TBZ), is an example of a post-harvest fungicide application. It was once highly effective; however, prolonged exposure to such agents has led to TBZ-resistant fungal strains (Sholberg and Haag 1996) that render chemical controls ineffective. 1.1.2 Botrytis cinerea Botrytis cinerea Pers.:Fr is an Ascomycete and, like P. expansum, is an important post-harvest pathogen of pome fruit causing minor rot throughout the growing season and significant rot within packinghouses. It is the most important post-harvest pathogen of pears and is second to blue mold in importance to apples (Rosenberger 1990). B. cinerea infection may originate from wounds, stem punctures, or the stem or calyx portion of the 5 fruit. Upon infection, B. cinerea inflicts a light to dark brown, spongy decayed area around the damaged portion of the fruit. Under high humidity, fluffy white to grey mycelium and spore masses may appear on the decayed area. The optimal growth temperatures for this pathogen range from 20°C to 22°C however, conidia are capable of growth at as low as -2°C (Coley-Smith et al. 1980). B. cinerea spores are ubiquitous, colonizing available organic matter within orchards. Throughout the growing season, this fungus can cause dry eye rot. B. cinerea also can be carried into packinghouses via contaminated storage bins. Once inside, conidia are predominantly spread by air currents and water splash. This pathogen is also known as a nest or cluster rot as secondary infection can occur through fruit to fruit contact; fungi on infected fruit can colonize healthy fruit and spread disease (Rosenberger 1990). 1.1.3 Mucor piriformis Mucor piriformis belongs to the phylum Zygomycota and, until recently, was thought to be of minor importance as a post-harvest pathogen. However, M. piriformis is capable of causing major decay problems in fruit such as strawberries, pome and stone fruit. M. piriformis is typically saprotrophic in orchard soil and infects fruit through the stem or calyx end and puncture wounds (Michailides and Spotts 1990). Upon infection, the area surrounding the lesion becomes soft, watery, light brown and easily separable from the fruit tissue. Often, grey mycelium with dark sporangia appears upon the decayed area. Mucor rot has a distinct sweet smell with a clear, sticky exudate. This fungus sporulates from -1 to 24°C, with optimal growth at 21°C. Fungicides that are currently registered to control other post-harvest pome fruit pathogens are ineffective 6 against M. piriformis and other Zygomycetes (Michailides and Spotts 1990). Although ScholarTM, with active ingredient fludioxonil, has been shown to be effective against M. piriformis (P. Sholberg, personal communication), sanitation methods are often the most effective method for preventing infection (Michailides and Spotts 1990). 1.2 Post-harvest factors that influence post-harvest pathology Because of the ubiquitous nature of fungal spores, their presence throughout the apple growing season can influence post-harvest pathology. P. expansum, B. cinerea and M. piriformis can colonize organic matter on the orchard floor and within soil (Lennox et al. 2003; Michailides and Spotts 1990). P. expansum and B. cinerea conidia can additionally exist in the air and on fruit surfaces (Lennox et al. 2003). Low precipitation, physiological crop condition and the use of pre-harvest fungicides will have a reducing effect on decay levels after harvest (Sholberg and Conway 2004; Sholberg et al. 2003; Sholberg et al. 2005b). However, post-harvest factors can also influence disease incidence. Within packinghouses, spores can survive in fungicide-drench solutions, flume water, dump-tank water and in the air and on walls of storage rooms (Lennox et al. 2003; Sholberg and Stokes 2006). Therefore, sanitation affects disease accumulation. The post-harvest system offers a unique, closed or semi-closed environment where manipulating temperature and atmosphere within the storage room and the use of chemical senescence inhibitors like 1-methylcyclopropene (1-MCP) are often easier to manage in comparison to pre-harvest factors that can contribute to disease. 7 1.2.1 Temperature The first experimental cold storage facility in British Columbia was built in Summerland in 1929 (AAFC 2002). Optimal refrigeration temperatures range from -0.5°C for Braeburn apples in South Africa to as high as 5°C for Belle de Boskoop apples grown in the Netherlands (Kupferman 2003). Lowered temperatures slow the rate of apple respiration, thus retarding ripening (AAFC 2002). Temperature management is also critical to post-harvest disease control. Post-harvest fruit pathogens grow optimally between 20 to 25°C. Some fungi have minimum growth temperatures as low as -2°C and cannot be completely controlled without freezing the fruit. Their growth at these temperatures, however, is significantly reduced which leads to a reduction in post-harvest decay levels (Sholberg and Conway 2004). 1.2.2 Controlled atmosphere “Normal air” refers to atmospheres that consist of 78 – 79% N2, 20 – 21% O2, ~ 0.03% CO2 and trace amounts of other gases. Controlled atmosphere (CA) refers to atmospheres that differ from “normal air” and are under strict control (Yahia 2009). Here, optimal temperatures persist while oxygen and carbon dioxide concentrations are decreased and increased, respectively (Morales et al. 2007). The concept of CA relies on the fact that harvested fruits use oxygen and produce carbon dioxide. If the amount of oxygen is limited, fruits will not ripen or will ripen slowly (Yahia 2009). Apples are often stored under CA as this environment has been shown to be effective in delaying the onset of storage diseases (Smock 1979). Commercial CA storage can range from 1% – 3% O2 and 0.4% – 4.5% CO2 (Kupferman 2003). Additional studies revealed that as high as 13% CO2 prevented close to 100% spore germination of P. expansum after 8 twenty days (Cossentine et al. 2004). However, CO2 levels this high are not commercially acceptable as it may lead to fruit injury resulting in a decrease in fruit quality. 1.2.3 1-Methylcyclopropene 1-Methylcyclopropene (1-MCP) is a synthetic cyclopropene that blocks ethylene receptors and prevents apple ripening (Blankenship and Dole 2003). It has the ability to maintain post-harvest fruit and vegetable quality and its use provides insight into the role of ethylene in fruit senescence (Watkins 2006). Ripening physiology and quality of apples in response to 1-MCP application has been intensively studied (Watkins 2006). Less studied, are the effects of 1-MCP on disease incidence. It is thought that by preventing the ethylene-associated ripening process, apples will be better able to resist pathogens (Watkins 2006). However, prevention of decay in 1-MCP-treated apples has been inconsistent. For example, Golden Delicious apples treated with 1-MCP and P. expansum showed decreased disease incidence (Saftner et al. 2003). Alternatively, disease severity increased in Golden Delicious apples inoculated with P. expansum or Colletotrichum acutatum and treated with 1-MCP (Janisiewicz et al. 2003; Leverentz et al. 2003). Although initially successful, there is the potential of certain decay problems associated with 1-MCP use (P. Sholberg, personal communication). 1.3 Methods of detection and identification Traditional methods for fungal identification were primarily based on morphology or phylogenetic characteristics. However, such techniques have limitations as morphological characterizations rely on fungi to be isolated or cultured, resulting in an 9 underestimation of the microbial community of interest (Bridge and Spooner 2001; Mazzola 2004). The advancement of molecular biology has led to more specific and sensitive fungal DNA-based detection methods that replaced assumptions made in previous studies. For example, Cruickshank and Pitt (1987) used enzyme gel electrophoresis, or a zymogram, to differentiate between Penicillium species. This study hypothesized that like species will display like zymograms. Results mostly confirmed taxonomy based on morphology; however, taxonomic differences were perceived. Detection techniques based on morphology or electrophoresis are time consuming. A rapid method of detection is therefore required. Greater specificity, sensitivity and speed can be attributed to the advancement of the polymerase chain reaction (PCR), a rapid and sensitive primer-mediated enzymatic amplification of target DNA sequences (Saiki et al. 1985). Common PCR-based identification methods include denaturing gradient gel electrophoresis (DGGE), temperature gradient gel electrophoresis (TGGE) or terminal restriction length polymorphisms (T-RFLP) (Mazzola 2004). Although PCR-based identification methods provide a more accurate picture of a microbial community, these methods are often limited by gel resolution or, more importantly, the inability to detect a large number of different pathogens simultaneously (Sholberg et al 2005a). The advent of DNA macroarray technology circumvents this problem. This system utilizes labelled PCR products hybridized to a nylon membrane that contains an assortment of anchored oligonucleotides or probes. Due to a chemiluminescent reaction, pathogen DNA, for example, is captured on X-ray film and is represented as a pattern of pre-determined dots. The presence of a signal indicates the presence of a pathogen that 10 can be identified and quantified. Levesque et al. (1998) were among the first to utilize macroarray technology for plant pathogen identification. This method was further developed for detection of economically important pre-harvest (Sholberg et al. 2005a) and post-harvest diseases of pome fruit (Hirkala et al 2007). These technologies have great potential as they can be used as a high-throughput detection and diagnostic technique that can identify numerous microorganisms across disciplines and environments. 1.4 Methods of control 1.4.1 Chemical control Fungicides are used to prevent post-harvest decay. In Canada, two synthetic post- harvest controls are currently registered, Mertect® and Scholar® with active ingredients thiabendazole (TBZ) and fludioxonil, respectively (PMRA 2007). Although initially effective, fungicides are now a major cause for concern. Pathogens have developed resistance to chemical controls (Errampalli et al. 2006; Sholberg et al. 2005c) which negates their effectiveness. There are also growing health concerns that surround pesticide use (Hancock et al. 2008) and a demand for environmentally friendly sustainable post-harvest practices. An alternative is therefore of high priority. One possibility is the implementation of biological controls. 1.4.2 Biological control Biological control can be defined as “the reduction of the amount of inoculum or disease-producing activity of a pathogen accomplished by or through one or more organisms other than man” (Cook and Baker 1983). As cited in Cook and Baker (1983), 11 biological control dates back to the early 1920s in which Hartley (1921) attempted to control damping off in pine seedlings with fungal antagonists. In 1927, Millard and Taylor tried to control common potato scab by the addition of Streptomyces praecox and grass clippings to autoclaved potted soil. Both studies noted that competition for nutrients may be a putative mechanism of action (Cook and Baker 1983). Sixty years later, the first commercial bacterial biological control agent (BCA), Agrobacterium radiobacter strain K84, was registered with the United States Environmental Protection Agency in 1979. It controlled crown gall caused by Agrobacterium radiobacter pv. tumifaciens. Strain K84 lacked the tumor-inducing (Ti) plasmid present within the pathogenic strain and produced a bacteriocin, which inhibits the growth of certain tumorigenic-strains, and to which K84 was insensitive (Cook and Baker 1983). Although the beginnings of biological control of plant pathogens date back to the early 1920s, practicing biological control within a post-harvest setting has been much less common. Currently, there are three BCAs registered for post-harvest use (Droby et al. 2009). The first and second, developed by JET Harvest Solutions (Longwood, FL), are Bio-Save 10LP and 11LP (Longwood, FL) with active ingredients of Pseudomonas syringae strain ESC-10 and ESC-11, respectively. Bio-Save 10LP targets post-harvest blue and grey mold and mucor rot in pome fruit while Bio-Save 11LP targets Rhizopus soft rot on sweet potatoes. The third, Shemer (Metschnikowia fructicola) is commercially used in Israel for prevention of sweet potato and carrot storage decay. Two yeast-based products, AspireTM and YieldPlus, are no longer available (Droby 2009). Biological control of post-harvest pathogens of pome fruit using bacterial and 12 yeast antagonists has been studied by numerous research groups (Calvo et al. 2007; Etebarian et al. 2005; Janisiewicz 1988; Mikani et al. 2008; Morales et al. 2008; Nunes et al. 2002; Zhou et al. 2001). For example, Candida sake strain CPA-1 significantly decreased P. expansum lesion size by > 80% and reduced disease incidence by 50% (Teixidó et al. 1999). Cryptococcus laurentii and Metschnikowia pulcherrima also provided control against P. expansum (Conway et al. 2007). The epiphytic bacterium, Rahnella aquatilis, significantly inhibited P. expansum and B. cinerea on apples at 15°C and 4°C, respectively (Calvo et al. 2007). A saprophytic strain of Pantoea agglomerans also reduced P. expansum rot on Golden Delicious apples at 1°C in air and at 1°C in a low oxygen atmosphere by 81% and 100%, respectively (Nunes et al. 2002). A study conducted by Etebarian et al. (2005) showed that Pseudomonas fluorescens isolate 1100- 6 significantly reduced the incidence of P. expansum and P. solitium on apples after 11 days at 20°C and 25 days at 5°C. In Canada, there are an increasing number of registrations and products under evaluation that are based upon microbially-active substances (Bailey et al. 2010). However, there are none registered for post-harvest use (Droby et al. 2009). A demand therefore exists for the production and commercialization of a BCA for post-harvest use on pome fruit. A collection of rhizobacteria, isolated from the roots of legumes grown in Saskatchewan soils, were compiled and initially characterized for plant growth promoting traits such as the production of siderophores, ACC deaminase, root elongation and the suppression of legume fungal pathogens (Hynes et al. 2008). Fifteen of these isolates were further tested in vitro for suppression of P. expansum, B. cinerea and M. piriformis (Stokes et al. 2006). Nine isolates provided fungal control and were 13 then tested in situ. Four isolates provided consistent control at both 20ºC and 1ºC; isolates 1-112, 2-28, 4-6 and 6-25 (Hirkala, unpublished data) in combination with 1100-6 were selected for further study. 1.5 Mechanisms of action The interaction between host, antagonist and pathogen is a complex, ecological relationship with high variability depending on the system of focus. The ability of an antagonist to suppress a pathogen may be due to more than one mechanism of action (MOA). A sound mechanistic understanding is essential for formulation development and biocontrol registration (Spadaro and Gullino 2004). Possible mechanisms include antibiosis (Janisiewicz et al. 1991; Kamesnky et al. 2003; Meziane et al. 2006), competition for nutrients (Elad 1996), parasitism (Frankowski et al. 2001; Watanabe et al. 2007) and induction of pathogen resistance in host tissue (Benhamou et al. 2000; Spadaro and Guillino 2004; Terry and Joyce 2004). 1.5.1 Antibiosis One method of antagonistic action may be through the production of antibiotics. For example, pyrrolnitrin produced by strains of Serratia plymuthica, was associated with fungal suppression (Kamesnky et al. 2003; Meziane et al. 2006). To determine whether or not pyrrolnitrin assisted in antifungal activity, Meziane et al. (2006) created an S. plymuthica mutant that lacked the gene responsible for pyrrolnitrin production. The pyrrolnitrin-deficient mutant, IC1270-P1, lost its antifungal activity when tested against pathogens Penicillium digitatum and Penicillium italicum. In addition, purified pyrrolnitrin was effective in suppressing disease symptoms of P. digitatum and P. 14 italicum. Janisiewicz et al. (1991) also isolated pyrrolnitrin from Pseudomonas cepacia for control of B. cinerea and P. expansum on apples and pears. Pyrrolnitrin was found to provide control; however, the level of control varied due to wound type (cut, nail, bruise), pyrrolnitrin concentration and storage temperature (2ºC and 24ºC). A post-harvest dip solution of pyrrolnitrin may therefore be an effective solution against post-harvest pathogens. It should also be noted that the post-harvest fungicide, ScholarTM, has an active ingredient of fludioxonil which belongs to the chemical class of phenylpyrroles. Phenylpyrroles are derived from pyrrolnitrin and are considered a reduced-risk chemical control agent (Errampalli 2004). 1.5.2 Competition for nutrients Competition for nutrients has been widely studied due to nutritional demands of both antagonists and pathogens. B. cinerea, as with most necrotrophs, requires nutrients for germination and the initiation of the pathogenic process. In the absence of nutrients, B. cinerea becomes highly susceptible to degradation and control (Elad 1996; Janisiewicz et al. 2000). It is postulated that microorganisms capable of efficiently utilizing nutrients will make successful biological control agents (Elad 1996). Janisiewicz et al. (2000) proposed a simple way to study competition for nutrients that uses a 24-well tissue culture plate with cylindrical insets with 0.45 µm membranes attached to the bottom. Individual cylinders are placed in each of the 24-wells thereby permitting media nutrient and metabolite interchange while preventing pathogen and antagonist contact due to physical separation. In this system, competition for nutrients can be studied independently from competition for space. Bencheqroun et al. (2007) used this method to determine, in vitro, that the biocontrol antagonist, Aureobasidium pullulans, was 15 competing for nutrients, especially the amino acids glycine, glutamic acid and serine, with the pathogen, Penicillium expansum. 1.5.3 Parasitism Antagonists can directly parasitize pathogens of post-harvest apples. One such method is through the production of lytic or cell wall degrading enzymes such as chitinase and β-1,3-glucanase. Bacterial strains that produce lytic enzymes will most likely have antifungal properties via cell wall hydrolysis. For example, S. plymuthica strain HRO-C48 was isolated from the rhizosphere of oilseed rape and shown to have antifungal properties associated with chitinase production. Frankowski et al. (2001) isolated and characterized two chitinolytic enzymes, one endochitinase (E.C. 3.2.1.14), CHIT60, and one N-acetyl-β-1,4-D-hexosaminidase (E.C. 3.2.1.52), CHIT100. In vitro, CHIT60 and CHIT100 showed direct inhibitory activity on spore germination and germ tube growth of B. cinerea. Alternate parasitic activity may involve direct attachment of antagonist to pathogenic fungal hyphae (Chan and Tian 2005; Watanabe et al. 2007). 1.5.4 Induced systemic resistance Induced systemic resistance (ISR) is an observed phenomenon in which microorganisms activate host plant mechanisms that enhance their defensive capacity against potential pathogenic invasions. Protection can occur both locally and on areas of the plant that did not come into contact with inducing microorganisms (van Loon 2007). For example, Benhamou et al. (2000) determined that S. plymuthica strain R1G64 mediated induced systemic resistance in cucumber to protect against infection caused by Pythium ultimum. Upon pre-treatment with S. plymuthica, P. ultimum disease incidence 16 was reduced; cucumber root cells were structurally and biochemically modified with the formation of phenolic-enriched occluded depositions and structural barriers. These structural barriers were hypothesized to prevent pathogen movement towards the vascular stele. In ISR experiments, it is important to spatially separate inducing microorganisms and challenging pathogens (ie. root and leaf) to ensure that protection is plant- not microorganism-mediated (van Loon 2007). 1.6 Green fluorescent protein Green fluorescent protein (GFP) is made up of 238 amino acids and exhibits green fluorescence at a peak emission of 509 nm by absorbing blue light maximally at 395 nm and minimally at 470 nm (Morin and Hastings 1971). In its native form, GFP was first discovered during the purification of the bioluminescent protein, aequorin, isolated from jellyfish (Aequorea victoria) (Shimomura et al. 1962). In A. victoria, green fluorescence occurs by an intermolecular energy transfer between aequorin and GFP; Ca2+ and aequorin interact and emit blue light, some of which is absorbed by GFP causing a color shift towards green (Morise et al. 1974). However, it wasn’t until 1992 that the influence of GFP was realized. Prasher et al. (1992) paved the way for future GFP application by cloning and sequencing both cDNA and genomic clones from Aequorea victoria. In 1994, Chalfie et al. first expressed GFP within a living organism, highlighting sensory neurons in nematodes. These landmark studies influenced modern science as purified GFP has become one of the most important reporter genes in biology (Bloemberg 2007). GFP is comprised of an eleven-stranded β barrel with a coaxial helix running through the centre. The chromophore is formed from the spontaneous cyclization of the 17 central helix consisting of the tripeptide, Ser65-Try66-Gly67 (Ormo et al. 1996). Introducing random amino acid substitutions into the twenty amino acids that flank the chromophore has led to altered forms of GFP some of which exhibit greater fluorescence. The first GFP modification was derived from a point mutation that altered Serine 65 to Threonine (S65T) (Heim et al.1995). This alteration shifted the excitation and emission maxima to 490 and 510 nm, respectively. Alterations of GFP not only affected intensity, but also increased color variety that is within the blue, cyan and yellow regions of the electromagnetic spectrum (Shaner et al. 2007). Green fluorescent protein can be used as a reporter or tag to mark whole cells, study protein localization and monitor gene interactions and interactions between microorganisms. Its incorporation is non-invasive, it does not require an additional substrate for bioluminescence, nor does visualization necessitate fixation or staining protocols (Bloemberg 2007). The gene product of gfp can be expressed in both prokaryotes and eukaryotes (Chalfie et al. 1994) with no background gene expression (Tombolini et al. 1997). It is highly stable in the presence of denaturants and proteases and persists at high temperatures (65°C) and a wide range of pH values (6-12) (Bloemberg 2007; Chalfie et al. 1994; Ward et al. 1980). Applications of GFP in biological control experiments can help elucidate mechanisms of action (MOA). Watanabe et al. (2007) transformed two fungal species with gfp, Trichoderma asperellum SKT-1, the antagonist, and Gibberella fujikuroi N-68, a pathogenic root fungus responsible for Bakanae disease of rice seedlings. The fluorescent protein was imaged by confocal scanning laser microscopy (CSLM) and mycoparasitism was suggested as the putative MOA. Scanning electron microscopy 18 (SEM) and CSLM showed a disappearance of GFP in G. fujikuroi upon contact with T. asperellum suggesting cell wall degradation. In another example, Bolwerk et al. (2003) transformed red fluorescent protein (RFP) into two putative BCAs, Pseudomonas fluorescens WCS365 and Pseudomonas chlororaphis PCL1391, that have been shown to control Fusarium oxysporum f. sp. radicis-lycopersici, a causal agent of tomato foot and root rot (TFRR). Both Pseudomonas spp. contained RFP, whereas Fusarium oxysporum f. sp. radicis- lycopersici harboured GFP. CSLM revealed that both antagonists colonized the tomato root more quickly than the fungal pathogen and that the bacterial presence hindered root infection. A proposed MOA was that the presence of fungi initiated bacterial colonization of hyphae and subsequent production of fungal secondary metabolites. 1.7 Objectives This thesis will build upon preliminary research and address the following three objectives: 1) validate the use of DNA macroarrays under field conditions; 2) assess bacterial antagonist performance under semi-commercial and commercial conditions and 3) determine how long and in what capacity a selected antagonist will colonize the fruit surface. I hypothesize that: 1) Macroarray technology will accurately identify and quantify post-harvest apple pathogens throughout the growing season and that pathogen prevalence will correlate with disease incidence post-harvest. 2) One or more of the bacterial antagonists will provide significant pathogen control in semi-commercial and commercial storage conditions. 19 3) The use of green fluorescent protein will facilitate visualization of bacterial colonization and survival on apple surfaces. 20 CHAPTER 2: MATERIALS AND METHODS 2.1 Antagonists Four bacterial antagonists were obtained from Dr. L. Nelson and were identified by 16S rRNA sequence and fatty acid methyl ester analysis as Pseudomonas fluorescens (isolates 1-112, 2-28, 4-6) and Serratia plymuthica (isolate 6-25) (Hynes et al. 2008). Isolate 1100-6, Pseudomonas fluorescens (Etebarian et al. 2005), was provided by Dr. Peter Sholberg of the Pacific Agri-Food Research Centre (PARC) in Summerland, BC. The bacteria were grown in half-strength tryptic soy broth (½ TSB) (Becton, Dickinson and Company, Sparks, MD) at 28°C with shaking at 220 rpm. The final culture was centrifuged for 20 min at 3800 g and held at 20°C. Cell pellets were resuspended in sterile phosphate buffered saline solution (PBS) (1.2 g/L Na2HPO4, 0.18 g/L NaH2PO4, 8.5 g/L NaCl). The optical density (OD600) was determined and the bacterial cells were diluted to 108 CFU/ml according to a standard curve relating OD600 to CFU/ml. 2.2 Pathogens Penicillium expansum Link strain 1790, Botrytis cinerea Pers.:Fr strain B27 and Mucor piriformis Fischer strain 536 were provided by Dr. Peter Sholberg and grown on half-strength potato dextrose agar (½ PDA) (HiMedia Laboratories PVT. Ltd., India) for 7 days at 22°C. A spore suspension was created using sterile water and Tween 20 (MP Biomedicals, LLC, Solon, OH) and then enumerated using a Petroff-Hauser counter. 21 2.3 Orchard field sites This study incorporated four conventional (non-organic) orchard field sites from which data were collected over two years. Two research orchards were located at PARC in Summerland, BC. Field 3 (49° 33’ 58.49” N; 119° 38’ 41.93” W) contained 166 Jonagold apple trees planted at high density. Field 12 (49° 33’ 54.81” N; 119° 38’ 56.73” W) contained 120 Gala apple trees planted at high density. The two commercial orchards were located in Kelowna, BC. The Kiran orchard (49° 50’ 41.45” N; 119° 24’ 54.24” W) grew Gala apples planted at high density. The Reekie orchard (49° 50’ 58.78” N; 119° 23’ 32.79” W) grew Red Delicious apples planted at medium density. I-rods were used in this study to collect aerial spore samples. I-rods are clear polystyrene rods that rest within a sampling head of a rotating spore trap. When the motor spins, centrifugal force causes the I-rods coated in silicone grease to extend downwards at a 90º angle and collect airborne particles (Aerobiology Research Laboratories, Nepean, ON). In the 2007 and 2008 growing seasons, Fields 3 and 12 contained one I-rod station per field. The Kiran orchard contained one I-rod station in both 2007 and 2008 and the Reekie orchard contained one I-rod station in 2008 only. Average daily temperatures (°C) and total daily precipitation (mm) from Environment Canada (2008) were plotted and used to compare with pathogen DNA detected throughout the apple growing seasons (Figure 1). Fungicide spray records for Fields 3 and 12 were provided. On May 15, 2007, the fungicide, Nova (myclobutanil) was applied to Field 12. On September 15, 2008, the fungicide Funginex (triforine) was applied to Field 3. 22 0 5 10 15 20 25 30 0 5 10 15 20 25 30 0 25 50 75 100 125 150 T o t a l D a i l y P r e c i p i t a t i o n ( m m ) A v e r a g e D a i l y T e m p e r a t u r e ( ° C ) Days after Blossom 0 5 10 15 20 25 30 0 5 10 15 20 25 30 0 25 50 75 100 125 150 T o t a l D a i l y P r e c i p i t a t i o n ( m m ) A v e r a g e D a i l y T e m p e r a t u r e ( ° C ) Days after Blossom 0 10 20 30 40 50 60 70 0 5 10 15 20 25 30 0 25 50 75 100 125 150 T o t a l D a i l y P r e c i p i t a t i o n ( m m ) A v e r a g e D a i l y T e m p e r a t u r e ( ° C ) Days after Blossom 0 10 20 30 40 50 60 70 0 5 10 15 20 25 30 0 25 50 75 100 125 150 T o t a l D a i l y P r e c i p i a t i o n ( m m ) A v e r a g e D a i l y T e m p e r a t u r e ( ° C ) Days after Blossom A. C. B. D. Figure 1. Average daily temperatures (°C) (solid line) and total daily precipitation (mm) (dashed line) at Kelowna, BC during the growing seasons of 2007 (A) and 2008 (B) and in Summerland, BC during the growing seasons of 2007 (C) and 2008 (D) as indicated by Environment Canada (2008). 23 2.4 pAG408 The gfp-containing plasmid, pAG408 (Suarez et al. 1997), was provided by Dr. Darren Korber, University of Saskatchewan, Saskatoon, Sask. pAG408 was maintained within the donor strain, E. coli S-17 λ pir, on Luria Burtani (LB) agar (10 g/ L tryptone, 5 g/L yeast extract, 10 g/L NaCl) supplemented with 50 µg/ml of kanamycin (kan) and 30 µg/ml of gentamicin (gen). To test for antibiotic efficacy, pAG408 in E. coli S-17 λ pir was streaked onto three replicate plates each of LB agar, LB agar + 50 µg/ml kan, LB agar + 30 µg/ml gen and LB agar + 50 µg/ml of kan and 30 µg/ml of gen and grown at 28°C for 48 h. Furthermore, pAG408 plasmid confirmation was verified by extracting it from E. coli S- 17 λ pir using the Wizard ® Plus Minipreps DNA Purification System (Promega, Madison, WI) as per manufacturer’s instructions, and digesting 1 µg of DNA with 1 U HindIII (Invitrogen, Carlsbad, CA) for 1 h at 37°C. 2.5 Construction of the gfp-transconjugant Bacterial isolate 4-6 was transformed with pAG408 via bacterial conjugation. Donor cells (E. coli S-17 λ pir containing pAG408) and recipient cells (4-6) were grown overnight at 28°C with shaking at 220 rpm. One millilitre of each culture was centrifuged at 10 000 x g for 20 min at 4°C. The supernatants were removed, the pellets washed twice with PBS then resuspended in 50 µl of PBS. Fifty microlitres each of the donors and recipients were combined and vortexed. The combined cultures were spotted on 0.22 µm sterile nylon membrane filters (Whatman, Maidstone, England) placed on LB agar plates supplemented with 50 µg/ml of kan and 30 µg/ml of gen. The plates were 24 incubated for 24 h at 28°C to allow for bacterial conjugation. The bacterial mixtures were then removed from the filters and plated on sodium citrate agar plates (0.2 g/L MgS04, 1.0 g/L NH4H2PO4, 2.0 g/L sodium citrate, 5.0 g/L NaCl, 15 g/L agar) supplemented with 50 µg/ml of kan and 30 µg/ml of gen and then incubated for 48 h at 28°C. The combination of sodium citrate growth medium and antibiotics eliminated E. coli S-17 λ pir pAG408 cells and non-transformed bacterial cells; E. coli cannot use citrate as a carbon source and the wild-type biological control isolates cannot withstand the antibiotics. For a negative control, non-transformed Pseudomonas fluorescens isolate 1-112, Serratia plymuthica isolate 6-25 and E. coli S-17 λ pir containing pAG408 were spread on sodium citrate agar plates supplemented with 50 µg/ml of kan and 30 µg/ml of gen. To confirm successful conjugation, 10 putative transconjugants were selected at random and their genomic DNA was extracted by boiling. One colony was placed in 100 µl of sterile water and the supernatant was used as the DNA template for the PCR that utilized gfp-specific primers. Three microlitres of DNA were used in the 25-µl reaction containing 1x PCR Buffer (TetraLink International, Buffalo, NY), 2.5 mM MgCl2, 0.1 mM dNTP mix (Invitrogen, Carlsbad, CA) , 0.4 µM of each forward (gfp F: 5’- GAGTAAAGGAGAAGAACTTTTCA-3’) and reverse primers (gfp R: 5’- TTATTTGTATAGTTCATCCATG-3’) (Suarez et al. 1997) and 1 U of UltraTherm DNA polymerase (TetraLink International, Buffalo, NY). Amplification was performed on the Techne TC-3000 thermocycler (Barloworld Scientific, London, England) with the following conditions: 96°C for 5 min followed by 35 cycles of 95°C for 1 min, 54.8°C for 1 min and 72°C for 1 min with a final extension cycle at 72°C for 7 min. 25 2.6 Plasmid stability Testing of the stability of exogenous DNA within bacterial isolate 4-6 was conducted according to Bloemberg et al. (1997) except that tryptic soy medium and the antibiotics, kanamycin and gentamicin, were used. Based on replicated stability results and the 2007 biocontrol experimental results, the promising gfp-tranconjgant biological control strain, P. fluorescens isolate 4-6, was selected as a suitable antagonist for bacterial survival and colonization studies. 2.7 P. fluorescens 4-6-gfp fitness analysis Growth rates of 4-6-gfp and its wild-type counterpart (4-6 wild-type) were compared. Three replicates each of 4-6-gfp and 4-6 wild-type were grown in 10 ml of ½ TSB and incubated at 28°C with shaking at 220 rpm for 24 h. Cultures were spun for 20 min at 5000 x g and resuspended in 10 ml of phosphate buffered saline (PBS). Optical density (OD600) was determined and cultures were diluted and standardized with PBS to within 0.100 of each other. One hundred microlitres of standardized culture were added to 10 ml of ½ TSB. OD600 readings were recorded at 0, 2, 4, 6, 8, 10, 12 and 24 h. The three replicates were averaged and represented graphically (time vs OD660). Growth rates (µ) were calculated with the formula µ = (log10 N-log10 No) x 2.303 / t-to) where N was a final cell number, No was an initial cell number, t was the time at N and to was the time at No. Carbon utilization profiles of 4-6-wild-type and 4-6-gfp were also compared by determining carbon profiles using BioLog Phenotype MicroArrays for Microbial Cells 26 (Biolog, Inc., Hayward, CA) (microplate PM1 and PM2) as per the manufacturer’s instructions 2.8 Spore, apple tissue collection and fruit washings Apple leaves, blossoms and I-rods were collected at orchard field sites throughout the apple growing seasons (Table 1). In general, sampling was conducted three times per week at the beginning (early) and end (late) of the growing seasons in 2007 and 2008. Throughout the mid-growing seasons, samples were collected approximately once per week. At harvest, approximately 1250 apples were collected at random from each of Fields 3 and 12 and approximately 300 apples were collected at random from the Kiran and Reekie orchards. Fruit surface washings were conducted on 20 apples each from Fields 3 and 12 in 2007 and 20 apples each from Fields 3, 12, the Kiran and the Reekie orchards in 2008. Twenty apples from each orchard were washed in a 600-ml beaker containing 200 ml of sterile distilled water and 200 µl of Tween 80. Each fruit was shaken for 5 min at 120 rpm and sonicated for 5 min (P. Sholberg, personal communication). After all fruit were washed, the solution was centrifuged at 13 400 g for 10 min. The supernatant was decanted and this process was repeated five times to a final volume of 10 ml that was centrifuged for 15 min at 7430 g. The supernatant was decanted and the remaining cells were transferred to a 2-ml tube and centrifuged at max (16 100 g) for 1 min. The supernatant was removed and 100 µl of water were added for storage at -20°C. 27 Table 1. Frequency of aerial and plant tissue sampling for Fields 3, 12 and the Kiran and Reekie orchards in the early, middle and late growing seasons of 2007 and 2008. Number of samples within sampling period (n) Field Sample type Early Mid Late Total 2007 Growing Season Field 3 aerial 14 6 6 26 plant tissue 2 3 2 7 Field 12 aerial 16 6 5 27 plant tissue 2 3 2 7 Kiran aerial 9 7 6 22 plant tissue 4 7 5 16 Reekie aerial 0 0 0 0 plant tissue 6 8 4 18 2007 Total 53 40 30 123 2008 Growing Season Field 3 aerial 18 11 3 32 plant tissue 4 0 2 6 Field 12 aerial 19 11 3 33 plant tissue 4 0 2 6 Kiran aerial 17 24 11 52 plant tissue 18 24 10 52 Reekie aerial 15 20 11 46 plant tissue 16 24 11 51 2008 Total 111 114 53 278 2.9 Validation and field testing of DNA macroarrays 2.9.1 DNA extraction DNA was extracted from I-rods, plant tissue samples and fruit washings using the Power Soil DNA Isolation Kit (Mo Bio Laboratories, Inc., Carlsbad, CA) according to the manufacturer’s instructions. For DNA isolation, two I-rods were used per aerial 28 extraction and 0.25 g of plant tissue were used for tissue extractions. DNA was eluted in 100 µl of Buffer 5 and stored at -20°C until required for further use. 2.9.2 PCR amplification PCR amplification was used to amplify universally conserved regions of the fungal genome, ribosomal DNA interspacer regions (rDNA ITS) and the β-tubulin gene. Approximately 10 – 20 ng of extracted DNA were used in 25-µl reactions containing 1x PCR Buffer (TetraLink International, Buffalo, NY), 2.5 mM MgCl2, 0.1 mM digoxigenin (Roche Diagnostics GmbH, Mannheim, Germany) (DIG)-dNTP mix , 0.4 µM of each forward and reverse primers and 1 U of UltraTherm DNA polymerase (TetraLink International, Buffalo, NY). Primers for rDNA were ITS1 (5’- TCCGTAGGTGAACCTGCGG-3’) and ITS4 (5’-TCCTCCGCTTATTGATATG-3’) (White et al. 1990) and β-tubulin primers were Bt-LEV-Lo1 (Fwd 5’- GTGAACTCCATCTCGTCCATA-3’) and Bt-LEV-Up4 (5’- CAAGATCCGTGAGGAGT-3’) (de Jong et al. 2001). Amplification was performed on the Techne TC-3000 thermocycler (Barloworld Scientific, London, England) with the following conditions: 96°C for 5 min followed by 35 cycles of 95°C for 1 min, 55°C for 1 min and 72°C for 1 min with a final extension cycle at 72°C for 7 min. DNA concentrations were estimated on 1% agarose gels using a Low Mass ladder (Invitrogen, Gaithersburg, MD). A minimum of 75 ng and a maximum of 100 ng of PCR-amplified DNA were required for hybridization to macroarrays. 29 2.9.3 DNA macroarray construction Twenty-eight oligonucleotide macroarray probes were designed by Dr. Danielle Hirkala and constructed by Operon Technologies, Huntsville, AL (Table 2). The 5’ end of the probe contained an amino C6 linker used to attach the oligonucleotide onto the Immunodyne ABC nylon membranes (PALL Europe Ltd., Portsmouth, England). The probes were resuspended in 0.5 M sodium bicarbonate buffer (pH 8.0) and diluted to a final concentration of 40 µm. In combination with 0.004% bromophenol blue, the resuspended probes were arranged according to a previously designed template (Figure 2). A 384-pin replicater (V&P Scientific, San Diego, CA) and pin guide (V&P Scientific, San Diego, CA) were used to “stamp” the probes onto the nylon membranes (Figure 2). The membranes were blocked for a minimum of 1 hour with hybridization buffer (6x SSC, 0.1% sarcosine, 0.02% SDS) and 1% skim milk at 55°C. 2.9.4 Hybridization and chemiluminescent detection To denature the DIG-labelled PCR products, 75 – 100 ng of DNA were combined with 5 ml of hybridization buffer and placed in boiling water for 10 min. Membranes were added to the denatured probe solution for overnight hybridization at 55°C. Unbound probes were then washed from the membranes with pre-warmed 6x SSC (0.18% NaCl, 0.088% sodium citrate) and 1% SDS (sodium dodecyl sulphate) for 2 x 40 min at 55°C. Hybridized membranes were pooled and washed for 5 min in washing buffer (0.1 M maleic acid, 0.15 M NaCl [pH 7.5], 0.3% Tween 20) and bound with a 1:25 000 dilution of anti-Digoxigenin-AP (alkaline phosphatase) Fab fragments (Roche Diagnostics GmbH, Mannheim, Germany) combined with washing buffer and 1% skim 30 Table 2. DNA macroarray probe sequences. Name Sequence 5’ to 3’ Description rDNA-ITS UN-H1-up [AminoC6]ACCTACGGAAACCTTGTTACGA Universal to all fungi Bot-all-H1b [AminoC6]TTAGCTTGGTATTGAGTCTATGT Universal to all Botrytis spp. BC-H2d [AminoC6]TATGCTCGCCAGAGAATACCAAA Botrytis cinerea BC-H3d [AminoC6]GCTCGCCAGAGAATACCAAAAC Botrytis cinerea BT-H1d [AminoC6]TATGCTCGCCAGAGAAAACCAAA Botrytis tulipae BT-H2d [AminoC6]CTCGCCAGAGAAAACCAAAACT Botrytis tulipae PE-H2c [AminoC6]CCCGAACTCTGCCTGAAGATT Penicillium expansum PE-H3u [AminoC6]CAGACGACAATCTTCAGGCA Penicillium expansum Mpir-ITS-183H1 [AminoC6]TGGTGTCCTTAAAAATTATTATTAT Mucor piriformis Mpir-ITS-414H1 [AminoC6]AACACCCCACATCTTAAAAATC Mucor piriformis β-tubulin UNbt-1 [AminoC6]CAAGAACATGATGGCTGCTTC Universal to all fungi UNbt-2 [AminoC6]CCAAGAACATGATGGCTGC Universal to all fungi UNbt-3 [AminoC6]TGTTCGACCCCAAGAACATG Universal to all fungi Pex-bt-84-H1-gag [AminoC6]TCCGACGAGACTTTCTGTATC Penicillium expansum Pex-bt-84-H1-gcg [AminoC6]TCCGACGCGACTTTCTGTATC Penicillium expansum TBZ sensitive Pex-bt-84-H1-gtg [AminoC6]TCCGACGTGACTTTCTGTATC Penicillium expansum TBZ resistant Pcom-bt-1 [AminoC6]CCGTCAACATGGTCCCCTT Penicillium commune Psol-bt-1 [AminoC6]TCCCTCGTTTGCACTTCTT Penicillium solitum Psol-bt-2 [AminoC6]CCTTTCCGTCCGTCCACCAGCT Penicillium solitum Paur-bt-1 [AminoC6]CACACCTCTGATATCTTGCTAGG Penicillium aureum Paur-bt-2 [AminoC6]CGATGGACAGTAAGTTCTAATGG Penicillium aurantigriseum Bcin-133-H3 [AminoC6]TTACGATATTTGCATGAGAACCT Botrytis cinerea Bstok-144-H4 [AminoC6]GCATGAGAACCCTGAAGCTC Botrytis mali Bot-95-H1-GAG [AminoC6]AACTCTGACGAGACCTTCTG Botrytis cinerea TBZ sensitive Bot-95-H1-GCG [AminoC6]ACTCTGACGCGACCTTCTG Botrytis cinerea TBZ resistant MucorUN-bt-1 [AminoC6]ACATGGTTCCTTTCCCTCGT Universal to all Mucor spp. MucorUN-bt-2 [AminoC6]AAGGCTTTCTTGCATTGGTA Universal to all Mucor spp. MucorUN-bt-3 [AminoC6]GGTGCTGGTAACTCTTGGGC Universal to all Mucor spp. 31 A. Post-Harvest Macrorray Template rDNA-ITS β-tubulin A B C D E F G H 1 UN-H1-up BT-H1d Mpir-ITS-183H1 UNbt-1 Pex-bt-84- H1gag Psol-bt-1 Bcin-133- H3 MucorUN- bt-1 2 Bot-all-H1- b BT-H2d Mpir-ITS- 414H1 UNbt-2 Pex-bt-84- H1gcg Psol-bt-2 Bstok-144- H4 MucorUN- bt-2 3 BC-H2d PE-H2c UNbt-3 Pex-bt-84-H1gtg Paur-bt-1 Bot-95-H1- gag MucorUN- bt-3 4 BC-H3d PE-H3u Pcom-bt-1 Paur-bt-2 Bot-95-H1-gcg B. Post-Harvest Macrorray Template rDNA-ITS β-tubulin A B C D E F G H 1                 2                 3               4           Figure 2. DNA macroarray pathogen template (A) and the corresponding oligonucleotide probe placements that are amine-linked to the nylon membrane (B). Columns A – C detect rDNA ITS PCR products; columns D – H detect β-tubulin PCR products. Each pathogen (A) corresponds with the dotted template (B) as it would appear on a developed macroarray. Each probe is blotted twice for positive confirmation. Grey boxes indicate no probe present. 32 milk powder for 30 min. Membranes were washed for 2 x 15 min then primed for 15 min in buffer 3 (0.1 M Tris-HCl, 0.1 M NaCl [pH 9.5]). For chemiluminescent detection, membranes were incubated for 5 min in buffer 3 plus 1:2000 dilution of chemiluminescent substrate, CDP-Star (Roche Diagnostics GmbH, Mannheim, Germany). Alkaline phosphatase enzymatically dephosphorylates CDP-Star, creating dioxetane phenolate anion, which decomposes and emits light maximally at 466 nm (Roche Diagnostics GmbH, Mannheim, Germany). Membranes were bound in plastic wrap and immediately exposed to x-ray film (CL-X Posure TM Film, Thermo Scientific, Rockford, IL) for 2 h. After fixing and developing for 1 minute each (Kodak, Rochester, NY), the hybridized DIG-amplified PCR products were captured on x-ray film as a series of dots with varying grey intensities. The x-ray film was scanned and saved as a Jpeg picture file using a BenQ 4300u scanner (Taipei, Taiwan) for computer analysis. The hybridized DNA was quantified using ImageJ 1.30v software (National Institutes of Health, Bethesda, MD). Each hybridization signal was assigned a relative greyscale value that ranged from 0 (completely back) to 255 (completely white). Greyscale values were in the form of two averaged measurements. The background was subtracted from each analyzed array using the following formula: Adjusted Greyscale Value = 127.5 x Original Greyscale Value Average Background The adjusted greyscale values were transformed to DNA concentrations when compared to a standard curve produced from known DNA concentrations. 33 2.10 Biological control of post-harvest fungal pathogens 2.10.1 Inoculation of apples by drenching Bacterial cultures were diluted in 5 L of water to a concentration of 108 CFU/ml and fungal cultures were diluted in 5 L of water to concentrations of 104 spores/ml for P. expansum and B. cinerea and 105 spores/ml for M. piriformis. Inoculation concentrations were previously determined by minimum infectious dose and minimum inhibitory concentration experiments (D. Hirkala, personal communication). Apples were surface sterilised with 90% ethanol and wounded in triplicate using a 3-mm diameter nail embedded in a cork. Wounded apples were placed in a plastic net bag and labelled by tagging. Each bag was placed in a 5-L bacterial suspension for 1 min then subsequently placed in a 5-L fungal suspension for 1 min. Following inoculation, apples were air dried on trays for 10 min then placed in mesh bags (10 apples / bag) and in crates (~100 apples / crate) for commercial storage or in covered plastic bins each with three 0.2-µm membrane filters attached to the lid for semi-commercial storage until lesions formed. In this study, commercial storage refers to controlled atmosphere (CA) storage conditions in a commercial packinghouse, whereas semi-commercial storage refers to storage in a research facility at 1°C in air. 2.10.2 Natural disease incidence Following apple harvest in 2007, natural disease incidence (NDI) was determined by wounding 100 Jonagold apples from Field 3 and 100 Gala apples from Field 12. Fifty of the 100 apples were fumigated with 1-Methylcyclopropene (1-MCP) and 50 were not fumigated with 1-MCP. In 2008, NDI was determined by placing 260 apples, each from Fields 3 (Jonagold), 12, Kiran (Gala) and Reekie (Red Delicious) into controlled 34 atmosphere (CA) storage (1.5% O2 and 1.5% CO2) (n = 100 per orchard) in Naramata, BC or in air storage at 1°C (n = 40 per orchard) at PARC in Summerland, BC. Half of the apples were wounded in triplicate and the remaining half were not wounded. 1-MCP treatments were not used in 2008 harvest experiments. Disease incidence (% apples infected) and lesion diameters (mm) were measured after 3 and 6 months for the 2007 harvest, after 4.5 months of sealed CA storage and after 1, 2, 4 and 6 months at 1°C in air storage for apples harvested in 2008. 2.10.3 Natural disease incidence with bacterial antagonists To measure bacterial antagonist efficacy against natural disease incidence for harvest 2007, 120 Jonagold apples from Field 3 and 120 Gala apples from Field 12 were wounded in triplicate and inoculated with one of five bacterial antagonists – isolates 1100-6, 1-112, 2-28, 4-6 or 6-25- and a non-inoculated control (10 apples / antagonist). Sixty of the 120 apples were fumigated with 1-MCP and the remaining 60 apples were not fumigated. Following harvest in 2008, 240 apples from each orchard, Field 3, 12, Kiran and Reekie, were inoculated as in section 2.10.1. One hundred and twenty apples from each orchard were stored in CA (60 wounded and 60 not wounded) and 120 apples were stored at 1°C in air (60 wounded and 60 not wounded). Disease incidence and lesion diameters were measured after 3 months in storage for apples harvested in 2007. For apples harvested in 2008, disease incidence and lesion diameters were measured after 4.5 months of sealed CA storage and after 1, 2, 4 and 6 months at 1°C in air storage. The efficacy of the biological control agents was compared to apples that had been wounded and not inoculated with bacteria. 35 2.10.4 Biological control of post-harvest pathogens For apples harvested in 2007, disease inhibition was measured by inoculating 360 Jonagold apples from Field 3 and 360 Gala apples from Field 12 with one of three fungal pathogens, P. expansum, B. cinerea or M. piriformis and one of five bacterial control isolates, 1100-6, 1-112, 2-28, 4-6, 6-25 and a non-inoculated control. One hundred and eighty of the 360 apples were fumigated with 1-MCP. For apples harvested in 2008, 360 apples each from Fields 3 and 12 and the Kiran and Reekie orchards were inoculated as above. In 2008, 1-MCP was not used, but 180 apples per orchard were wounded and 180 apples per orchard were non-wounded. From each site, there were 10 apples / antagonist + 1 non-inoculated control / pathogen. Disease incidence and lesion diameters were measured after 3 months of storage for apples harvested in 2007 and after 1, 2, 4 and 6 months at 1°C in air storage for apples harvested in 2008. Pathogens were not permitted in commercial CA storage. 2.10.5 Bacterial survival on apple Concentrations of 4-6-gfp were monitored by two experimental techniques- colony counts via dilution plating and direct scanning using the POLARstar Omega microplate reader (BMG Labtech, Germany) as in Etebarian et al. (2005). A standard curve was developed that described the relationship between colony forming units (CFU/ml) and relative fluorescent units (RFU) acquired by direct scanning. The bacterial inoculum was prepared in ½ TSB with 50 µg/ml of kan and 30 µg/ml of gen and grown at 28°C for 24 h. P. expansum spores were harvested from 7-d- old ½ PDA plates. Final inoculation concentrations for 4-6-gfp and P. expansum were 108 CFU/ml and 104 spores/ml, respectively. Apples were prepared as in section 2.9.1 36 and inoculated by pipette (30 µl) with the following treatments: 1) non-inoculated control; 2) 4-6-gfp alone; 3) P. expansum alone and; 4) 4-6-gfp + P. expansum. Isolate 4- 6-gfp was quantified by aseptically removing an apple core (5 mm in diameter x 5 mm deep) from two of the three apple wounds. Both cores were macerated with a sterile rod. One core, to be used for dilution plating, was placed in 1.1 ml of PBS and plated on Pseudomonas F agar (20 g/L peptone, 1.5 g/L K2HPO4, 1.5 g/L anhydrous MgSO4, 10 g/L agar [pH 7.2]) supplemented with 50 µg/ml kan and 30 µg/ml gen. The second core, to be used in direct scanning, was placed in 300 µl of PBS. Two hundred microlitres of the tissue slurry were added to a well in a black Nunc 96-well microplate (Thermo Fisher Scientific, Rochester, NY). Fluorescent readings were obtained using a 405 nm excitation filter and a 510 nm emission filter and a gain of 1200. The relative fluorescent units (RFU) of PBS and non-inoculated control were subtracted from the treatments to correct for background and apple fluorescence, respectively. Bacterial concentrations were determined approximately every 10 days for a total of 55 d. 2.11 Analysis In all biocontrol experiments, disease was measured by disease incidence (percent apple infection) and disease severity (average lesion diameter in mm). From these two values, data were converted to infection severity (IS) (Spotts et al. 1999): Infection Severity (IS) = % Apple Infection * Average Lesion Diameter in mm 100% In order to correct for variance heterogeneity, infection severity values were log10 (n+0.5) transformed. Differences in means were analyzed by either a one-way or two- 37 way analysis of variance (ANOVA). p-values less than 0.05 were considered statistically significant. If the 2-way ANOVA interaction term was significant, the main effects were separated and a one-way ANOVA was used for subsequent analysis. Differences among means were determined by the least significant difference (LSD) test. The Pearson Product Moment Correlation Coefficient (r) was used to measure the linear association of two independent variables. Statistical analyses were conducted using CoStat Statistics Software (CoHort Software, Monterey, CA). 38 CHAPTER 3: RESULTS 3.1 Validation and field testing of DNA macroarrays 3.1.1 Quantification of post-harvest pathogens Macroarray construction for the identification of post-harvest pathogens was completed by Dr. Danielle Hirkala and was based upon the prototype proposed in Sholberg et al. (2005a). The array displayed high specificity, no cross-reactivity and detected as little as 0.1 ng of pure culture fungal DNA extracted from P. expansum and B. cinerea and up to 0.01 ng of pure culture of M. piriformis (Hirkala, unpublished). For the purpose of this study, three macroarray probes were selected for pathogen identification: PE-H3u for P. expansum detection, BC-H2d for B. cinerea detection and Mpir-ITS-414H1 for M. piriformis detection. Each of the three selected probes is comprised of sequences in the rDNA ITS region and is located on the left hand side of the macroarray template (Figure 2). These probes were selected because of their overall higher sensitivity compared to other probes. This decision was based upon all 2007 greyscale values obtained from environmental field data. Standard curves were determined for each of the three probes (Figures 3, 4 and 5). To test the quantitative properties of the amino-linked oligonucleotides, 0 – 100 ng of pure culture fungal DNA was hybridized to the macroarray. Each detector oligonucleotide displayed varying levels of hybridization signal intensity; however, a linear relationship did exist between DNA concentration and hybridization signal strength. The linear range of the probe PE-H3u ((♦) r = -0.974; (●) r = -0.962) was between 0 and 3 ng (Figure 3), whereas the linear range for the probe BC-H2d 39 y = -16.67x + 134.5 y = -41.147x + 123.99 0 20 40 60 80 100 120 140 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Gr ey sca le Va lue DNA Concentration (ng) Figure 3. Quantitative assessment of Penicillium expansum probe PE-H3u illustrating the relationship between greyscale value and DNA concentration (ng). (♦) Represents standard curve used for samples collected in 2007 (r = -0.974); (●) Represents standard curve used for samples collected in 2008 (r = -0.962). Error bars represent standard error of the mean. 40 y = -8.5966x + 126.02 y = -8.1112x + 120.26 0 20 40 60 80 100 120 140 0 1 2 3 4 5 6 7 8 9 10 11 12 Gr ey sca le Va lue DNA Concentration (ng) Figure 4. Quantitative assessment of Botrytis cinerea probe BC-H2d illustrating the relationship between greyscale value and DNA concentration (ng). (♦) Represents standard curve used for samples collected in 2007 (r = -0.960); (●) Represents standard curve used for samples collected in 2008 (r = -0.977). Error bars represent standard error of the mean. 41 y = -148.38x + 129.06 y = -181.28x + 116.61 0 20 40 60 80 100 120 140 0 0.1 0.2 0.3 0.4 0.5 Gr ey sca le Va lue DNA Concentration (ng) Figure 5. Quantitative assessment of Mucor piriformis probe Mpir-ITS-414H1 illustrating the relationship between greyscale value and DNA concentration (ng). (♦) Represents standard curve used for samples collected in 2007 (r = -0.970); (●) Represents standard curve used for samples collected in 2008 (r = -0.990). Error bars represent standard error of the mean. 42 ((♦) r = -0.960; (●) r = -0.977) was between 0 and 12 ng (Figure 4) and the probe Mpir- ITS-414H1 ((♦) r = -0.970; (●) r = -0.990) was between 0 and 0.5 ng (Figure 5). Above these ranges, the curves deviated from linearity as the oligonucleotides reached their saturation level. Throughout the duration of this study, the intensity of the hybridization signals decreased over time which led to the construction of two separate quantification curves for each of the three pathogens (Figures 3, 4 and 5). The two standard curves created for P. expansum DNA quantification (ng) were constructed in August 2008 (♦) and August 2009 (●) (Figure 3). Positive hybridization signals that were present in samples assessed in 2008 were quantified using the formula y = -41.147x + 123.99 (♦). Positive hybridization signals that were present in samples assessed in 2009 were quantified using the formula y = -16.67x + 134.5 (●). The two standard curves created for B. cinerea DNA quantification (ng) were constructed in August 2008 (♦) and October 2009 (●) (Figure 4). Positive hybridization signals that were present in samples assessed in 2008 were quantified using the formula y = -8.1112x + 120.26 (♦). Positive hybridization signals that were present in samples assessed in 2009 were quantified using the formula y = -8.5966x + 126.02 (●). The two standard curves created for M. piriformis DNA quantification (ng) were constructed in October 2008 (♦) and August 2009 (●) (Figure 5). Positive hybridization signals that were present in samples assessed in 2008 were quantified using the formula y = -181.28x + 116.61 (♦). Positive hybridization signals that were present in samples assessed in 2009 were quantified using the formula y = -148.38x + 129.06 (●). 43 3.1.2 Field application of macroarray technology The DNA macroarray was used in two years of field monitoring studies to verify airborne and plant tissue pathogen spore loads in four orchards in the Okanagan Valley, BC: Field 3 in Summerland, BC (Figure 6); Field 12 in Summerland, BC (Figure 7); Kiran orchard in Kelowna, BC (Figure 8); Reekie orchard in Kelowna, BC (Figure 9). All three probes had positive hybridization signals throughout the growing season with high variability between years and sample types. In Field 3 (Figure 6), all three pathogens, P. expansum (A. 2007, D. 2008), B. cinerea (B. 2007, E. 2008) and M. piriformis (C. 2007, F. 2008) were detected most frequently in the early and late growing seasons. In Field 12 (Figure 7), P. expansum was detected at the beginning and middle of the growing season in 2007 (A) and throughout the growing season in 2008 (D). Pathogen levels were much higher in 2008 (D) than in 2007 (A). B. cinerea was detected in the middle of the growing season in 2007 (B) and the beginning and the end of the growing season in 2008 (B). B. cinerea was not detected on plant tissue samples in either 2007 or 2008 (B, E). M. piriformis was detected at the beginning and middle of the growing season in 2007 (C) and at the beginning and the end of the growing season in 2008 (F). In the Kiran orchard (Figure 8), P. expansum (A. 2007, D. 2008), B. cinerea (B. 2007, E. 2008) and M. piriformis (C. 2007, F. 2008) were detected throughout the growing season in both 2007 and 2008. DNA concentrations for P. expansum and B. cinerea were higher in 2008 than in 2007. In the Reekie orchard (Figure 9), P. expansum was detected in the middle of the growing season in 2007 (A) and throughout the growing season in 2008 (D). B. cinerea was not detected in 2007 (B) and was detected throughout the growing season in 2008 (E). M. piriformis was detected in the middle of 44 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 F. E. C. B. A. D. DN A ( ng ) Days After Blossom 2007 2008 Figure 6. Concentrations of P. expansum (A,D), B. cinerea (B,E) and M. piriformis (C,F) from aerial (solid line) and plant tissue samples (dashed line) collected from Field 3 (Summerland, BC) in 2007 (A – C) and 2008 (D – F) using DNA macroarrays and identified by post-harvest pathogen rDNA ITS macroarray probes, PE-H3u (♦), BC-H2d (●) and Mpir-ITS-414H1 (■). 45 0 1 2 3 4 5 6 7 0 25 50 75 100 125 150 0 1 2 3 4 5 6 7 0 25 50 75 100 125 150 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 0 1 2 3 4 5 6 7 0 25 50 75 100 125 150 0 1 2 3 4 5 6 7 0 25 50 75 100 125 150 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 F. E. C. B. A. D. DN A ( ng ) Days After Blossom 2007 2008 Figure 7. Concentrations of P. expansum (A,D), B. cinerea (B,E) and M. piriformis (C,F) from aerial (solid line) and plant tissue samples (dashed line) collected from Field 12 (Summerland, BC) in 2007 (A – C) and 2008 (D – F) using DNA macroarrays and identified by post-harvest pathogen rDNA ITS macroarray probes, PE-H3u (♦), BC-H2d (●) and Mpir-ITS-414H1 (■). 46 0 2 4 6 8 10 0 25 50 75 100 125 150 0 2 4 6 8 10 0 25 50 75 100 125 150 0 2 4 6 8 10 0 25 50 75 100 125 150 175 0 2 4 6 8 10 0 25 50 75 100 125 150 175 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 175 F. E.B. A. D. DN A ( ng ) Days After Blossom 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 C. 2007 2008 Figure 8. Concentrations of P. expansum (A,D), B. cinerea (B,E) and M. piriformis (C,F) from aerial (solid line) and plant tissue samples (dashed line) collected from the Kiran orchard (Kelowna, BC) in 2007 (A – C) and 2008 (D – F) using DNA macroarrays and identified by post-harvest pathogen rDNA ITS macroarray probes, PE-H3u (♦), BC-H2d (●) and Mpir-ITS-414H1 (■). 47 0 2 4 6 8 10 12 0 25 50 75 100 125 150 0 2 4 6 8 10 12 0 25 50 75 100 125 150 0 2 4 6 8 10 12 0 25 50 75 100 125 150 175 0 2 4 6 8 10 12 0 25 50 75 100 125 150 175 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 175 F. E.B. A. D. DN A ( ng ) Days After Blossom 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 C. 2007 2008 Figure 9. Concentrations of P. expansum (A,D), B. cinerea (B,E) and M. piriformis (C,F) from aerial (solid line) and plant tissue samples (dashed line) collected from the Reekie orchard (Kelowna, BC) in 2007 (A – C) and 2008 (D – F) using DNA macroarrays and identified by post-harvest pathogen rDNA ITS macroarray probes, PE-H3u (♦), BC-H2d (●) and Mpir-ITS-414H1 (■). 48 the growing season in 2007 (C) and throughout the growing season in 2008 (F). All pathogen levels were higher in 2008 than in 2007. Due to the observed variability, the data were arranged in an alternate way in order to determine if there were common patterns between pathogens and fields (Table 3). To determine what time in the growing season each of the three main post-harvest pathogens was primarily detected, the season was divided into three sections: 1) early (~1 – 50 days after blossom); 2) mid (~50 – 100 days after blossom) and; 3) late (~100 – 150+ days after blossom) growing season (Table 3). Aerial and plant tissue samples were analyzed separately and in combination by determining percent pathogen detection within either the early, mid or late growing seasons. According to the tabulated data in Table 3, in 2007 and 2008, P. expansum was detected in each of the three growing season segments but differed by field and year. From DNA isolated from aerial samples collected in 2007, there was a significant effect of growing season segment on % P. expansum detected (p = 0.06) (Appendix, Table A1); the % P. expansum detected in the early season was significantly higher than that in the late growing season (Table 3). There was no effect of growing season segment on % P. expansum detected in aerial samples in 2008 (Table 3; Appendix, Table A4). From DNA isolated from plant tissue samples in 2007 there was no significant effect of the growing season segment on the % P. expansum detected (Table 3; Appendix, Table A2), but in 2008, there was a significant effect (p = 0.053) (Appendix, Table A5). The % P. expansum detected in the mid-growing season was significantly higher than that in the late-growing season (Table 3). 49 In 2007, there was no effect of the growing season segment on the % B. cinerea detected in DNA isolated from either aerial or plant tissue samples (Table 3; Appendix, Tables A1 and A2). In 2008, from DNA isolated from plant tissue samples, the % B. cinerea detected in the mid-growing season was significantly higher than that in the early and late-growing seasons (Table 3; Appendix, Table A5). Overall, B. cinerea was detected at much lower levels than either P. expansum or M. piriformis. M. piriformis was detected in each of the three growing season segments but also differed by field and year (Table 3). From DNA isolated from plant tissue samples in 2007, the % M. piriformis detected was significantly higher in the mid-growing season than that in the early or late-growing seasons (Table 3; Appendix, Table A2). From DNA isolated from plant tissue samples in 2008, the % M. piriformis detected was significantly higher in the mid-growing season than that in the late-growing season (p = 0.06) (Table; Appendix, Table A5). In both 2007 and 2008, when aerial and plant tissue DNA samples were analyzed together, there was no effect of growing season segment (Table 3; Appendix, Tables A3 and A6). In 2007 and 2008 combined, the selected probes detected P. expansum (27.4 ± 3.4%) and M. piriformis (19.2 ± 1.4%) most frequently, followed by B. cinerea (6.2 ± 1.4%) (Table 3; Appendix, Table A7). In addition, P. expansum was detected at the highest concentrations, followed by B. cinerea and then M. piriformis. Of the samples collected in 2007, the Kiran orchard had the highest percent detection (81.6%) followed by Field 12 (50%), the Reekie orchard (33.3%) and Field 3 (30.3%). In 2008, the Reekie orchard had the highest percent detection (80.4%) followed by Field 12 (71.8%), Kiran (46.2%) and Field 3 (28.9%) (Table 3). 50 Table 3. DNA macroarray detection (%) summary for P. expansum, B. cinerea and M. piriformis from four apple orchards throughout the 2007 and 2008 growing seasons. Growing season was divided into 3 sampling periods: 1) early-growing season (~ 1-50 days after blossom); 2) mid-growing season (~ 50-100 days after blossom) and; 3) late-growing season (~ 100-150+ days after blossom). Numbers in ( ) indicate standard error of the mean. % pathogen detection within sampling perioda P. expansum B. cinerea M. piriformis % detection per orchard Field Sample type 1 2 3 1 2 3 1 2 3 2007 Growing Season Field 3 aerial 28.6 0.0 0.0 7.1 0.0 0.0 7.1 0.0 16.7 plant tissue 0.0 0.0 100.0 0.0 0.0 0.0 0.0 33.3 0.0 Field ave 25.0b 0.0 25.0 6.3 0.0 0.0 6.3 11.1 12.5 30.3 Field 12 aerial 37.5 33.3 0.0 0.0 33.3 0.0 12.5 0.0 0.0 plant tissue 50.0 33.3 50.0 0.0 0.0 0.0 0.0 66.7 0.0 Field ave 38.9 33.3 14.3 0.0 22.2 0.0 11.1 22.2 0.0 50.0 Kiran aerial 66.7 0.0 16.7 11.1 14.3 0.0 66.7 28.6 50.0 plant tissue 50.0 57.1 0.0 0.0 0.0 20.0 0.0 42.9 20.0 Field ave 61.5 28.6 9.1 7.7 7.1 9.1 46.2 35.7 36.4 81.6 Reekie aerial ND ND ND ND ND ND ND ND ND plant tissue 0.0 25.0 0.0 0.0 0.0 0.0 0.0 50.0 0.0 Field ave 0.0 25.0 0.0 0.0 0.0 0.0 0.0 50.0 0.0 33.3 Ave % detection per sampling period (irod)c 44.2ah (11.5) 11.1ab (11.1) 5.6b (5.6) 6.1a (3.3) 15.9a (9.7) 0.0a (0.0) 28.8a (19.0) 9.5a (9.5) 22.2a (14.7) Ave % detection per sampling period (pt)d 25.0a (14.4) 28.9a (11.8) 37.5a (23.9) 0.0a (0.0) 0.0a (0.0) 5.0a (5.0) 0.0b (0.0) 48.2a (7.0) 5.0b (5.0) 2007 ave % detection per sampling period (irod + pt)e 31.4a (12.9) 21.7a (7.4) 12.1a (5.2) 3.5a (2.0) 7.3a (5.2) 2.3a (2.3) 15.9a (10.3) 29.8a (8.4) 12.2a (8.6) 2007 ave % detection per pathogenf 24.0a (5.6) 4.2b (1.7) 20.6ab (13.8) 51 % pathogen detection within sampling period P. expansum B. cinerea M. piriformis % detection per orchard Field Sample type 1 2 3 1 2 3 1 2 3 2008 Growing Season Field 3 aerial 0.0 18.2 33.3 0.0 9.1 66.7 27.8 0.0 0.0 plant tissue 0.0 ND 0.0 0.0 ND 0.0 0.0 ND 0.0 Field ave 0.0 18.2 20.0 0.0 9.1 40.0 22.7 0.0 0.0 28.9 Field 12 aerial 57.9 54.5 100.0 5.3 9.1 0.0 21.1 0.0 33.3 plant tissue 25.0 ND 0.0 0.0 ND 0.0 0.0 ND 0.0 Field ave 52.2 54.5 60.0 4.3 9.1 0.0 17.4 0.0 20.0 71.8 Kiran aerial 5.9 25.0 18.2 0.0 12.5 18.2 5.9 29.2 18.2 plant tissue 38.9 20.8 10.0 5.6 12.5 0.0 27.8 8.3 0.0 Field ave 22.9 22.9 14.3 2.9 12.5 9.5 17.1 18.8 9.5 46.2 Reekie aerial 73.3 45.0 9.1 40.0 5.0 9.1 40.0 30.0 9.1 plant tissue 50.0 37.5 9.1 6.3 8.3 0.0 37.5 37.5 0.0 Field ave 61.3 40.9 9.1 22.6 6.8 4.5 38.7 34.1 4.5 80.4 Ave % detection per sampling period (irod)c 34.3ah (18.4) 35.7a (8.5) 40.2a (20.6) 11.3a (9.6) 8.9a (1.5) 23.5a (14.9) 23.7a (7.1) 14.8a (8.5) 15.2a (7.1) Ave % detection per sampling period (pt)d 28.5ab (10.8) 29.2a (8.3) 4.8b (2.8) 3.0b (1.7) 10.4a (2.1) 0.0b (0.0) 16.3ab (9.6) 22.2a (14.6) 0.0b (0.0) 2008 ave % detection per sampling period (irod + pt)e 34.1a (14.0) 34.1a (8.4) 25.8a (11.6) 7.4a (5.1) 9.4a (1.2) 13.5a (9.0) 24.0a (5.1) 13.2a (8.2) 8.5a (4.3) 2008 ave % detection per pathogen f 30.8a (10.2) 8.3b (1.3) 17.8ab (3.8) 2007/08 ave % detection per pathogeng 27.4a (3.4) 6.2b (1.4) 19.2a (1.4) 52 a. % pathogen detection: # of times a pathogen was detected within a select sampling period * 100% # of samples taken within a select sampling period b. Field averages were obtained by calculating the average % pathogen detection between both aerial and plant tissue samples c. Ave % detection per sampling period (irod): irod = aerial sample; values were determined by taking the average of the % detection values per sampling period per pathogen within each orchard in aerial samples only d. Ave % detection per sampling period (pt): pt = plant tissue sample; values were determined by taking the average of the % detection values per sampling period per pathogen within each orchard in plant tissue samples only e. Ave % detection per sampling period: values were determined by taking the average of the % detection values per sampling period per pathogen within each orchard in aerial and plant tissue samples combined f. Ave % detection per pathogen: values were determined by taking the average of the % pathogen detection values within all three sampling periods within each orchard with aerial and plant tissue samples combined g. 2007/08 ave % detection per pathogen: values were determined by taking the average of the values obtained in Ave % detection per pathogen in both 2007 and 2008 h. Means followed by different letters within a pathogen and within a row are significantly different (p ≤ 0.06) according to the least significant difference (LSD) test. 53 Fruit washings conducted after harvest in 2007 showed the presence of B. cinerea in Jonagold apples grown in Field 3 and the presence of P. expansum, B. cinera and M. piriformis in Gala apples grown in Field 12. Fruit washings conducted after harvest in 2008 showed the presence of P. expansum in Gala apples grown in the Kiran orchard (Table 4). Table 4. Detection of P. expansum, B. cinerea and M. piriformis from fruit washings collected from Jonagold apples from Field 3, Gala apples from Field 12 and the Kiran orchard and Red Delicious apples from the Reekie orchard in 2007 and 2008 using DNA macroarrays. DNA detected (ng/20 apples) Field P. expansum B. cinerea M. piriformis Harvest 2007 Field 3 NDa 0.48 ND Field 12 0.78 0.66 0.05 Harvest 2008 Field 3 ND ND ND Field 12 ND ND ND Kiran 1.41 ND ND Reekie ND ND ND a. ND = not detected 3.2 Natural disease incidence 3.2.1 Harvest 2007 The effect of incubation period (3 and 6 months) and 1-MCP use (1-MCP+/-) on infection severity (IS) was analyzed using a two-way ANOVA. The interaction terms for apples from both Field 3 and 12 were not significant (Appendix, Tables A8 and A9). In both Field 3 and 12, there was a greater IS at six months than at three months in 1°C storage. However, only in Field 3 was there a significant effect of 1-MCP application; 1- MCP reduced IS by 48.2% in comparison to the non-fumigated control (Tables 5 and 6). 54 Table 5. Natural disease incidence indicated by infection severitya in Jonagold apples grown in Field 3 (harvest 2007), treated or not treated with 1-MCP (1-MCP+/-) and stored for 3 or 6 months at 1°C in air. Field 3 Field 3 Incubation Period 3 months 6 months Overall Mean 1-MCP + 5.51 9.55 7.53 bc 1-MCP - 9.09 22.12 15.61 a Overall Mean 7.30 bb 15.83 a a. Infection severity (IS) = percent apple infection x average lesion diameter (mm)/100 (Spotts et al. 1999). b. Incubation period means with different letters are different at significance levels p ≤0.05 according to least significant difference (LSD) test. c. 1-MCP means with different letters are different at significance levels p ≤0.05 LSD test. Table 6. Natural disease incidence indicated by infection severitya in Gala apples grown in Field 12 (harvest 2007), treated or not treated with 1-MCP (1-MCP+/-) and stored for 3 or 6 months at 1°C in air Field 12 Field 12 3 months 6 months Overall Mean 1-MCP + 7.64 30.28 18.95 ac 1-MCP - 11.78 22.29 16.56 a Overall Mean 9.90 bb 26.28 a a. Infection severity (IS) = percent apple infection x average lesion diameter (mm)/100 (Spotts et al. 1999). b. Incubation period means with different letters are different at significance levels p ≤0.05 according to least significant difference (LSD) test. c. 1-MCP means with different letters are different at significance levels p ≤0.05 LSD test. 3.2.2 Harvest 2008 The effects of incubation period (1, 2, 4 and 6 months) on infection severity in each of the four orchards were analyzed using a one-way ANOVA. Results indicated that in all fields, there was a significant increase in IS from one to six months (Table 7; Appendix, Table A10). A one-way ANOVA also revealed that at two months incubation, Jonagold apples grown in Field 3 had significantly greater infection severities than in all 55 other orchards. At four and six months of incubation, Jonagold apples grown in Field 3 and Gala apples grown in Field 12 had significantly greater infection severities than in the commercial orchard apples, Gala grown in the Kiran orchard and Red Delicious grown in the Reekie orchard (Appendix, Table A11). Table 7. Natural disease incidence indicated by infection severitya in Jonagold apples grown in Field 3, Gala apples grown in Field 12 and the Kiran orchard and Red Delicious apples grown in the Reekie orchard (harvest 2008) and stored for 1, 2, 4 or 6 months at 1°C in air. Apple Variety Field 3 Jonagold Field 12 Gala Kiran Gala Reekie Red Delicious 1 month 0.00 dA 0.00 dA 0.36 bA 0.00 cAc 2 months 1.08 cA 0.54 cB 0.01 bB 0.00 cB 4 months 20.72 bA 19.95 bA 0.08 bB 2.28 bB 6 months 48.12 abA 46.58 aA 2.06 aB 10.88 aB a. Infection severity (IS) = percent apple infection x average lesion diameter (mm)/100 (Spotts et al. 1999). b. Incubation period means with different letters within columns (lowercase) are different at significance levels at p ≤0.05 according to least significant difference (LSD) test. c. Incubation period means with different letters within rows (uppercase) are different at significance levels at p ≤0.05 according to the LSD test. In addition to storing apples at 1°C in air, fruit were also stored in controlled atmosphere (CA) for 4.5 months. Jonagold apples grown in Field 3 had the highest infection severity with significantly greater disease levels compared to Gala apples grown in Fields 12 and Kiran and Red Delicious apples grown in the Reekie orchard (Figure 10; Appendix, Table A12). The high infection severity found in Jonagold apples grown in Field 3 appeared to be due primarily to P. expansum (based on phenotypic observation). 56 a b b b 0 2 4 6 8 10 12 Field 3 Jonagold Field 12 Gala Kiran Gala Reekie Red Delicious Av era ge In fec tio n S ev er ity (IS ) Field Figure 10. Natural disease incidence in Jonagold apples grown in Field 3, Gala apples grown in Field 12 and the Kiran Orchard and Red Delicious apples grown in the Reekie orchard and stored for 4.5 months in controlled atmosphere (CA) in Naramata, BC. IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 57 3.3 Growing season detection and natural disease incidence In this study, there was no correlation between % total pathogen detection (P. expansum, B. cinerea and M. piriformis) detected in the field throughout the entire growing season in both 2007 and 2008 combined (Table 3) and NDI after storage (r = - 0.11; p = 0.8349). With respect to Fields 3 and 12 in 2007 and Fields 3 and 12 and the Kiran and Reekie orchards in 2008, there was no correlation between % pathogens detected in the early-growing season and their corresponding NDIs (r = -0.29; p = 0.5811) and % pathogens detected in the mid-growing season and their corresponding NDIs (r = -0.28; p = 0.5865). However, there was a positive correlation between % pathogens detected in the late-growing season and NDI (r = 0.79; p = 0.0612). When the aerial samples were analyzed separately from the plant tissue samples, in the 2007 and 2008 data combined, there was no correlation between % pathogens detected in the early (r = -0.57; p = 0.2234) and mid-growing seasons and their corresponding NDIs (r = -0.36; p = 0.4838). However, there was a positive correlation between % pathogens detected in late-growing season and NDI (r = 0.74; p = 0.0903). When the plant tissue samples were separated from the aerial samples, in 2007 and 2008 combined, there was no correlation between % pathogens detected in either the mid (r = - 0.22; p = 0.7810) or late-growing seasons (r = -0.21; p = 0.6826) and their corresponding NDIs. However, there was a negative correlation between % pathogens detected in the early-growing season and NDI (r = -0.79; p = 0.0623). 58 3.4 Antagonist efficacy, harvest 2007 3.4.1 Natural disease incidence and biocontrol inoculations The effects of biological control treatment (no biocontrol, 1100-6, 1-112, 2-28, 4- 6 or 6-25) and 1-MCP use (1-MCP+/-) were analyzed using a two-way ANOVA. For Jonagold apples grown in Field 3 and stored at 1°C in air for 3 months, there was no significant effect of the biological controls, 1-MCP use or their interaction (Appendix, Table A13). Results obtained with inoculated Gala apples from Field 12 indicated a significant interaction with both main effects, biological control treatment and 1-MCP use (Appendix, Table A14). Because the interaction term was significant, treatments were separated, followed by a one-way analysis of variance. 1-MCP significantly decreased IS in apples with no biological control treatment and apples that were treated with bacterial isolates 1100-6 and 2-28 (Figure 11; Appendix, Table A15). There were significant reductions in IS means in the absence of 1-MCP in apples treated with bacterial isolates 1-112, 4-6 and 6-25 in comparison to the non-inoculated control (Figure 11). In 1-MCP- treated apples, there was no significant effect of treatments compared to the non-inoculated control (Figure 11; Appendix, Table A16). 3.4.2 Post-harvest pathogen and biocontrol inoculations Infection severity (IS) was determined with apples grown in Fields 3 and 12 and inoculated with one of three pathogens, P. expansum, B. cinerea or M. piriformis and one of five bacterial control isolates and the data were subjected to a two-way ANOVA. IS means for Jonagold apples from Field 3 and inoculated with P. expansum 59 B ab A a B ab A a A bc A d B ab A ab A a A cAab A bc 0 10 20 30 40 50 60 70 MCP+ MCP- Av er ag e I nf ec tio n S ev er ity (IS ) MCP Treatment no biocontrol 1100-6 1-112 2-28 4-6 6-25 Figure 11. Effect of 1-MCP (capital letters) and biological control isolates (lowercase letters) on natural disease incidence in Gala apples grown in Field 12 (harvest 2007) and stored for 3 months in air at 1°C. IS means with different letters within treatments are different at significance levels p ≤0.05 according to LSD test. 1-MCP effect comparisons (capital letters) can be made between 1-MCP treatment groups. Biological control activity comparisons (lowercase letters) can be made within 1-MCP treatment groups. 60 exhibited a significant effect of biological control treatment and of the interaction between the main effects, biological control treatment and 1-MCP use (Appendix, Table A17). The resulting one-way ANOVA showed that the application of 1-MCP significantly decreased IS means in apples inoculated with isolate 1100-6 and increased IS in apples inoculated with isolate 6-25 (Figure 12a; Appendix, Table A18). In comparison to the non-inoculated control, isolates 1100-6, 1-112, 4-6 and 6-25 provided significant control of P. expansum in both the presence and absence of 1-MCP (Figure 12a; Appendix, Table A19). For B. cinerea-treated apples from Field 3, the two-way ANOVA revealed a significant effect of biological control treatment and 1-MCP and of their interaction (Appendix, Table A20). The one-way ANOVAs revealed a significant increase in IS in 1-MCP-treated apples that were not inoculated with a biocontrol and apples inoculated with isolates 2-28 and 4-6 (Figure 12b; Appendix, Table A21). In comparison to a non- inoculated control, isolates 1100-6, 1-112 and 6-25 provided significant control of B. cinerea in the presence of 1-MCP. There was no significant reduction in IS in apples in the absence of 1-MCP (Figure 12b; Appendix, Table A22). For M. piriformis-treated apples grown in Field 3, there was a significant effect of the biological control treatment and of the interaction between the main effects, biological control treatment and 1-MCP use (Appendix, Table A23). The one-way ANOVAs revealed a significant increase in IS in non-1-MCP treated apples inoculated with isolate 4-6 (Figure 12c; Appendix, Table A24). There was no effect of inoculation with any of the biological control isolates in the presence of 1-MCP; however, there was a significant increase in IS in non-1-MCP-treated apples inoculated with isolate 4-6 when 61 A a A a B c A b A d A c A ab A ab A d A c A b B c 0 10 20 30 40 50 60 70 MCP+ MCP- Av era ge In fec tio n S ev eri ty (IS ) MCP Treatment A a B a A b A a A b A a A a B a A a B a A b A a 0 10 20 30 40 50 60 70 MCP+ MCP- Av era ge In fec tio n S ev eri ty (IS ) MCP Treatment A a A bc A a A bA a A bc A a A c B a A a A a A c 0 10 20 30 40 50 60 70 MCP+ MCP- Av era ge In fec tio n S ev eri ty (IS ) MCP Treatment A. P. expansum C. M. piriformis B. B. cinerea no biocontrol 1100-6 1-112 2-28 4-6 6-25 Figure 12. Effect of 1-MCP (capital letters) and biological control isolates (lowercase letters) on infection severity in Jonagold apples from Field 3 (harvest 2007), inoculated with P. expansum (A), B. cinerea (B) and M. piriformis (C) and stored for 3 months at 1°C in air. IS means with different letters within treatments are different at significance levels p ≤0.05 according to LSD test. 1-MCP effect comparisons (capital letters) can be made between 1-MCP treatment groups. Biological control activity comparisons (lowercase letters) can be made within 1-MCP treatment groups. 62 compared to 1-MCP-treated apples (Figure 12c; Appendix, Table A25). For Gala apples from Field 12 treated with P. expansum and each of five biological control isolates, there was a significant effect of 1-MCP use and no significant interaction between the main effects, biological control treatment and 1-MCP use (Appendix, Table A26). 1-MCP significantly increased the mean IS (48.16 ± 0.50) compared to that obtained in the absence of 1-MCP (45.69 ± 0.68). There was no overall significant effect of the biological control isolates (Appendix, Table A26). There were no data for Gala apples from Field 12 inoculated with B. cinerea and the five biological control isolates as there was complete apple decay after 3 months of storage at 1°C. For Gala apples from Field 12 treated with M. piriformis and each of the five biological controls, there was a significant interaction between the main effects, biological control treatment and 1-MCP use (Appendix, Table A27). The one-way ANOVAs revealed a significant decrease in IS in 1-MCP-treated apples inoculated with isolate 1-112 compared to non-1-MCP-treated apples and a significant increase in IS in non-1-MCP-treated apples inoculated with isolate 2-28 when compared to 1-MCP-treated apples (Figure 13; Appendix, Table A28). In the presence of 1-MCP, isolate 2-28 exhibited significant biological control activity compared to the non-inoculated control. There was no significant reduction in IS in non-1-MCP treated apples when compared to a non-inoculated control (Figure 13; Appendix, Table A29). 3.5 Antagonist efficacy, harvest 2008 For all 2008 biological control experiments, apples were wounded and non- wounded. Non-wounded apples had a very low frequency of infection with low 63 A ab A ab A a B bB c A a A ab A abA bc A b 0 10 20 30 40 50 60 70 MCP+ MCP- Av era ge In fec tio n S ev eri ty (IS ) MCP Treatment no biocontrol 1-112 2-28 4-6 6-25 Figure 13. Effect of 1-MCP (capital letters) and biological control isolates (lowercase letters) on infection severity in Gala apples from Field 12 (harvest 2007), inoculated with M. piriformis and stored for 3 months at 1°C in air. IS means with different letters within treatments are different at significant levels p ≤0.05 according to LSD test. 1-MCP effect comparisons (capital letters) can be made between 1-MCP treatment groups. Biological control activity comparisons (lowercase letters) can be made within 1-MCP treatment groups. 64 variation. As a result, the variable was eliminated from 2008 biological control experimental analysis as it explained little about the corresponding dependent and independent variables (Figure 14). 3.5.1 Natural disease incidence and biocontrol inoculations The effects of biological control treatment (no biocontrol, 1100-6, 1-112, 2-28, 4- 6 or 6-25) and incubation period (1, 2, 4 and 6 months) were analyzed using a two-way ANOVA. For Jonagold and Gala apples grown in Fields 3 and 12, respectively, then stored at 1°C in air, there was a significant effect of incubation period and of the interaction between the main effects, biological control treatment and incubation period (Appendix, Tables A30 and A31). For Jonagold apples grown in Field 3, there was no significant reduction in IS in apples treated with the five biological controls after 2 and 4 months in storage (Figure 15a and 15b). However, after six months in storage, isolates 2- 28 and 6-25 significantly reduced IS compared to the non- inoculated control (Figure 15c; Appendix, Table A32). For Gala apples grown in Field 12, isolates 1100-6, 1-112, 4-6 and 6-25 significantly reduced IS compared to the non-inoculated control after two months in storage (Figure 15d; Appendix, Table A33). There was no significant reduction in IS after four and six months storage compared to the non-inoculated control (Figure 15e and 15f). Jonagold apples grown in Field 3, Gala apples grown in Fields 12 and the Kiran orchard and Red Delicious apples grown in the Reekie orchard were also stored for 4.5 months in CA storage. A two-way ANOVA with replication indicated there was a significant effect of biological control treatment and location and of the interaction between biological control treatment and location (Appendix, Table A34). The one-way 65 0 5 10 15 20 25 NDI (1°C air) NDI (CA) NDI + bc (1°C air) NDI + bc (CA) Pathogen + bc (1°C air) Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Experiment wounded non-wounded Figure 14. Average infection severity for all wounded and non-wounded apples from individual biological control treatments conducted in 2008. Experiments were as follows: 1) Natural disease incidence (NDI) at 1°C in air; 2) Natural disease incidence (NDI) in controlled atmosphere (CA); 3) Natural disease incidence (NDI) in combination with biocontrols (bc) at 1°C in air; 4) Natural disease incidence (NDI) in combination with biocontrols (bc) in CA; 5) Pathogen in combination with biocontrols at 1°C in air. Error bars represent standard error of the mean. 66 a b b ab b b 0 2 4 6 8 10 no biocontrol 1100-6 1-112 2-28 4-6 6-25 b a b b b b 0 2 4 6 8 10 no biocontrol 1100-6 1-112 2-28 4-6 6-25 a ab ab bc a c 0 10 20 30 40 50 60 70 80 no biocontrol 1100-6 1-112 2-28 4-6 6-25 a a a a a a 0 10 20 30 40 50 60 70 80 no biocontrol 1100-6 1-112 2-28 4-6 6-25 bc c c ab bc a 0 10 20 30 40 50 60 70 80 no biocontrol 1100-6 1-112 2-28 4-6 6-25 b b b ab ab b 0 10 20 30 40 50 60 70 80 no biocontrol 1100-6 1-112 2-28 4-6 6-25 E.B. A. D. C. F. Av er ag e I nfe cti on Se ve rit y ( IS) Biocontrol Treatment Field 3, 2008 Field 12, 2008 Figure 15. Effect of biological control isolates on natural disease incidence in Jonagold apples grown in Field 3 (A – C) and Gala apples grown in Field 12 (D – F) (harvest 2008) and stored for 2 (A&D), 4 (B&E) or 6 (C&F) months at 1°C in air. IS means with different letters within figures are different at significance levels p ≤0.05 according to LSD test. 67 ANOVAs revealed a significant decrease in IS in Jonagold apples grown in Field 3 and inoculated with biological control isolates 1100-6, 1-112, 2-28, 4-6 and 6-25 (Figure 16; Appendix, Table A35). There was no significant reduction in IS in Gala apples grown in Field 12 and inoculated with one of five biological controls (data not shown; Appendix, Table A35). In Gala apples grown in the Kiran orchard, there was little overall disease. However, isolates 1100-6, 2-28, 4-6 and 6-25 significantly reduced IS compared to the non-inoculated control (Figure 17; Appendix, Table A35). For Red Delicious apples grown in the Reekie orchard, there was no significant effect on IS when inoculated with the biological control isolates (data not shown; Appendix, Table A35). 3.5.2 Post-harvest pathogen and biocontrol inoculations Infection severity (IS) was determined with apples grown in Fields 3 and 12 and inoculated with one of three pathogens, P. expansum, B. cinerea or M. piriformis and one of five bacterial control isolates. One way ANOVAs were computed to determine if there were significant effects of the biological control agents on infection severities within each field. For Jonagold apples grown in Field 3 and inoculated with P. expansum there was a significant reduction in IS in apples treated with isolates 4-6 and 6-25 compared to the non-inoculated control (Figure 18a; Appendix, Table A36). For Gala apples grown in Field 12 and inoculated with P. expansum, there was a significant reduction in IS in apples treated with isolates 1100-6, 2-28 and 6-25 compared to the non-inoculated control (Figure 19, Appendix, Table A36). In Jonagold apples grown in Field 3 and Gala 68 a c bc b bc bc 0 2 4 6 8 10 12 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev er ity (IS ) Biocontrol Treatment Figure 16. Effect of biological control isolates on natural disease incidence in Jonagold apples grown in Field 3 (harvest 2008) and stored for 4.5 months in CA. IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 69 a b ab b b b 0 0.1 0.2 0.3 0.4 0.5 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev er ity (IS ) Biocontrol Treatment Figure 17. Effect of biological control isolates on natural disease incidence in Gala apples grown in the Kiran orchard (harvest 2008) and stored for 4.5 months in CA. IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 70 a bc bc b bc c 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment b bc a bc c c 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment a b b b b b 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment A. P. expansum C. M. piriformis B. B. cinerea no biocontrol 1100-6 1-112 2-28 4-6 6-25 Figure 18. Effect of biological control isolates on infection severity in Jonagold apples grown in Field 3 (harvest 2008), inoculated with P. expansum (A) and B. cinerea (B) and stored for 2 months at 1°C in air and inoculated with M. piriformis (C) and stored for 1 month at 1°C in air . IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 71 a b b b b b 0 2 4 6 8 10 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment a c abc bc ab c 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment a bc bc b bc c 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment A. P. expansum C. M. piriformis B. B. cinerea no biocontrol 1100-6 1-112 2-28 4-6 6-25 Figure 19. Effect of biological control isolates on infection severity in Gala apples grown in Field 12 (harvest 2008), inoculated with P. expansum (A) and B. cinerea (B) and stored for 2 months at 1°C in air and inoculated with M. piriformis (C) and stored for 1 month at 1°C in air. IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 72 apples grown in Field 12 and inoculated with either B. cinerea (Figures 18b and 19b; Appendix, Table A37) or M. piriformis (Figures 18c and 19c; Appendix, Table A38) there was a significant reduction in IS in apples treated with all isolates, 1100-6, 1-112, 2-28, 4-6 and 6-25. Table 8 summarizes all biological control treatments in 2007 and 2008 and indicates the percent decrease (or increase) of all BCAs. Overall, results obtained from two years of experimental trials indicated that the bacterial isolate that provided significant control in the greatest number of treatments was Serratia plymuthica isolate 6-25 (51.7%) followed by P. fluorescens isolates 4-6 (41.4%), 1100-6 (37.9%), 1-112 (34.5%) and 2-28 (31%). The bacterial isolate that provided the greatest reduction in infection severity, in cases where significant control was exhibited, was isolate 1-112 followed by isolates 1100-6, 4-6, 6-25 and 2-28. Their respective average reduction in IS was 75.8%, 61%, 59%, 57.6% and 56.4%. In 2007, isolates 1100-6, 1-112 and 6-25 provided significant control of both P. expansum and B. cinerea. Isolate 4-6 provided significant control of P. expansum, only. Isolate 2-28 was the only bacterial strain to provide significant control of M. piriformis. In 2008, isolates 1100-6, 2-28, 4-6 and 6-25 provided significant control of all pathogens, P. expansum, B. cinerea and M. piriformis. Isolate 1-112 provided significant control of both P. expansum and B. cinerea (Table 8). All bacterial antagonists provided greater than 90% control in at least two experiments (Table 8). 73 Table 8. Summary of antagonist efficacy experiments for harvest 2007 and 2008. Percent disease decrease or (increase)a Experimentb Storage Typec Field Apple Variety Storage Time (months) 1-MCP+/- 1100-6 1-112 2-28 4-6 6-25 Harvest 2007 NDI + bc 1C air 3 Jonagold 3 MCP + NS NS NS NS NS 1C air 3 Jonagold 3 MCP - NS NS NS NS NS NDI + bc 1C air 12 Gala 3 MCP + NS NS NS NS NS 1C air 12 Gala 3 MCP - NS 82.1 NS 49.2 39.9 Pex + bc 1C air 3 Jonagold 3 MCP + 78.7 94.3 NS 97.3 53.2 1C air 3 Jonagold 3 MCP - 58.6 94.1 NS 96.9 96.2 1C air 12 Gala 3 MCP+&- NS NS NS 7.8 5.7 Bcin + bc 1C air 3 Jonagold 3 MCP + 94.7 98.1 NS NS 96.1 1C air 3 Jonagold 3 MCP - NS NS NS NS NS Mpir + bc 1C air 3 Jonagold 3 MCP + NS NS NS NS NS 1C air 3 Jonagold 3 MCP - NS NS NS (98.1) NS 1C air 12 Gala 3 MCP + NS NS 91.6 NS NS 1C air 12 Gala 3 MCP - NS NS NS NS NS Harvest 2008 NDI + bc 1C air 3 Jonagold 2 N/A (69.6) NS NS NS NS 1C air 3 Jonagold 4 N/A NS NS NS NS (52.9) 1C air 3 Jonagold 6 N/A NS NS 47.4 NS 67.5 1C air 12 Gala 2 N/A 74.7 88.3 NS 70.4 80.8 1C air 12 Gala 4 N/A NS NS NS NS NS 1C air 12 Gala 6 N/A NS NS NS NS NS 74 Percent disease decrease or (increase)a Experimentb Storage Typec Field Apple Variety Storage Time (months) 1-MCP+/- 1100-6 1-112 2-28 4-6 6-25 Harvest 2008 NDI + bc CA 3 Jonagold 4.5 N/A 94.9 83.8 43.0 63.9 45.5 CA 12 Gala 4.5 N/A NS NS NS NS NS CA Kiran Gala 4.5 N/A 100 NS 100 100 81.1 CA Reekie Red Delicious 4.5 N/A NS NS NS NS NS Pex + bc 1C air 3 Jonagold 2 N/A NS (26.0) NS 32.8 33.9 1C air 12 Gala 2 N/A 26.7 NS 20.0 NS 26.7 Bcin + bc 1C air 3 Jonagold 2 N/A 57.4 38.1 56.8 33.1 37.1 1C air 12 Gala 2 N/A 26.9 28.6 18.3 22.3 33.9 Mpir + bc 1C air 3 Jonagold 1 N/A 65.4 62.8 39.0 45.0 74.6 1C air 12 Gala 1 N/A 93.0 87.6 91.8 90.0 85.0 a. Percentages indicate a decrease in IS; percentages in brackets indicate an increase in IS; NS = not significant b. NDI = natural disease incidence; bc = biocontrol; Pex = P. expansum; Bcin = B. cinerea; Mpir = M. piriformis c. CA = controlled atmosphere 75 3.6 Bacterial survival on apple using GFP 3.6.1 Plasmid confirmation E. coli S-17 λ pir containing the gfp-plasmid, pAG408, grew on LB agar without antibiotics and LB agar supplemented with either 50 µg/ml kan or 30 µg/ml gen or both antibiotics combined. Alternatively, the bacterial isolates 1100-6, 1-112, 2-28, 4-6 and 6- 25 only grew on agar plates that were not supplemented with antibiotics. When pAG408 was digested with HindIII, the plasmid was cut into two fragments which measured approximately 2.1 and 3.6 kb (Figure 20). These fragments approximated the total plasmid size of 5.7 kb (Suarez et al. 1997). The undigested plasmid migrated to approximately 10 kb due to its supercoiled conformation 3.6.2 Confirmation of gfp transformation After transformation of genetic material from the host (E. coli S-17 λ pir) into the donor (isolate 4-6), transconjugants were amplified by PCR using gfp-specific primers. Successful gfp amplification was observed when a band was present around 714 bp (Figure 21). The absence of banding suggested an unsuccessful transformation. Confirmation of gfp transformation was additionally confirmed by fluorescence microscopy in vitro (Figure 22) and in vivo (Figure 23). 3.6.3 Plasmid stability After four successive days of sub-culturing and growth in the absence of antibiotic selection, 80.3 ± 6.8% of the transformed P. fluorescens isolate 4-6 (4-6-gfp) retained kanamycin and gentamicin resistance. These results, in combination with prior biological control experimental results, indicated that 4-6-gfp was an appropriate tool for 76 4.0 1 2 43 5 2.1 kb 3.6 kb3.0 2.0 kb Figure 20. HindIII restriction digest of gfp-containing plasmid, pAG408. Lane 1: 1 kb ladder (New England Biolabs, Ipswich, MA); Lanes 2 and 4: undigested pAG408; Lanes 3 and 5: digested pAG408. 77 Figure 21. PCR amplification of gfp in P. fluorescens isolate 4-6 transformed with pAG408 using gfp-specific primers. Sample were loaded on a 1% TBE agarose gel and stained with Sybr Safe (Invitrogen, Carlsbad, CA). Lane 1) Low DNA Mass Ladder (Invitrogen, Carlsbad, CA); 2) negative control (ddH20); 3) extracted plasmid, pAG408; 4 and 5) gfp transconjugant of bacterial isolate 4-6. 78 Figure 22. Fluorescence microscopy of P. fluorescens isolate 4-6 transformed with pAG408 exhibiting green fluorescence in ½ TSB supplemented with 50 µg/ml kan and 30µg/ml gen. 79 Figure 23. Fluorescence microscopy of an apple wound inoculated with P. fluorescens isolate 4-6-gfp. AP = apple peel; AW = apple wound; PF = P. fluorescens isolate 4-6-gfp; R = receptacle or fruit flesh. 80 studying bacterial survival on apples in the absence of antibiotic selection. 3.6.4 P. fluorescens 4-6-gfp fitness analysis There was no significant difference between the growth rates (µ) (Figure 24; Appendix, Table A39) of P. fluorescens isolate 4-6-wild-type (0.71 ± 0.04/h) and 4-6-gfp (0.75 ± 0.06/h) and sole source carbon utilization patterns (Table 9). Table 9. Sole carbon source utilization by P. fluorescens isolates 4-6-wild-type and 4-6- gfp as indicated by Biolog Phenotye MicroArrays™ (Biolog, Inc., Hayward, CA) for sugars and organic acids commonly found in apple juice (Eisele and Drake 2005). Apple Carbon Sources Bacterial Isolate 4-6-wild-type 4-6-gfp Fructose + + Sucrose + + Glucose + + Sorbitol + + L-Malic Acid + + Quinic Acid + + Citric Acid + + Fumaric Acid + + 3.6.5 Bacterial survival on apple For standard bacterial suspensions of 4-6-gfp, there was a strong linear relationship between CFU/ml and relative fluorescent units (RFU) (r = 0.999) (Figure 25). The fluorescent values obtained by direct scanning of inoculated apple samples (Figure 26) were too low to extrapolate to cell densities from the formula y = 1.8041x + 6.5652 (Figure 25). However, trends could be discerned based on RFUs (Figure 26). 81 0.01 0.1 1 10 0 4 8 12 16 20 24 log 10 OD 60 0 Time (h) 4-6-gfp 4-6-wild-type Figure 24. Growth characteristics of P. fluorescens 4-6-wild-type and P. fluorescens 4-6- gfp in half-strength TSB at 28°C for 24 h. Error bars represent standard error of the mean (n = 3). 82 y = 1.8041x + 6.5652 0 20 40 60 80 100 120 140 0 10 20 30 40 50 60 70 Re lat ive Flu or esc en t U nit s (R FU ) (x 10 3 ) CFU/ml (x 108) Figure 25. Relationship between P. fluorescens isolate 4-6-gfp cell density (CFU/ml) and relative fluorescent units (RFU) as determined by dilution plating and direct scanning (r = 0.999). Values are an average of three replicates. 83 -1000 0 1000 2000 3000 4000 5000 6000 0 10 20 30 40 50 60 Re lat ive Flu or esc en t U nit s (R FU ) Days in Storage (d) 4-6-gfp P. expansum 4-6-gfp + P. expansum Figure 26. Bacterial survival of 4-6-gfp (♦), P. expansum (●) and P. expansum + 4-6-gfp (■) on Gala apples stored for 55 d at 1°C in air using the direct scanning method. Error bars represent standard error of the mean (n = 4). 84 Isolate 4-6-gfp alone increased during 20 d in storage then decreased after 30 d in storage, followed by an increase at 55 d in storage. P. expansum alone followed the same trend; however, the overall RFUs were higher. The RFUs obtained for 4-6-gfp in the presence of P. expansum initially decreased in the first 10 d of storage and increased until 45 d in storage, followed by a decrease at 55 d in storage. The selective growth medium, PF agar supplemented with 50 µg/ml kan and 30 µg/ml gen, was effective in isolating 4-6-gfp and no other resident bacteria on the apple surface were detected. Over 55 d of dilutions and platings, there was no bacterial growth on the plates with extracted material from the untreated and P. expansum only-inoculated apples. A two-way ANOVA indicated a significant interaction between bacterial strain, 4-6-wild-type and 4-6-gfp, and time (Appendix, Table A40). One-way ANOVAs revealed no significant differences in cell densities between apples inoculated with 4-6- gfp + P. expansum and apples inoculated solely with 4-6-gfp up to 20 d in 4°C storage (Figure 27; Appendix, Table A41). From 30 to 55 d of storage, populations of 4-6-gfp in the presence of P. expansum differed significantly from populations of 4-6-gfp alone. When challenged with the pathogen, bacterial populations remained constant until lesions developed after 30 d, after which populations rapidly declined. When 4-6-gfp was unchallenged with the pathogen, its population increased from 20 – 30 d in storage then declined to slightly greater than its population at time 0. From 0 to 40 d in storage, 4-6- gfp significantly reduced IS compared to the control (P. expansum alone) (Table 10; Appendix, Table A42). There was no significant reduction in IS of apples treated with 4- 85 6-gfp in the presence of P. expansum after 40 d in storage (Table 10; Appendix, Table A42). Table 10. Effect of biological control isolate 4-6-gfp on infection severity in Gala apples inoculated with and without P. expansum and stored for 55 days at 1°C in air. IS means with different letters within columns are different at significant levels p ≤0.05 according to the least significant difference (LSD) test. Treatments Days in storage 10 20 30 40 45 55 Untreated 0.0 0.0 0.0 0.0 0.0 0.0 4-6-gfp alone 0.0 0.0 0.0 0.0 0.0 0.0 P. expansum alone 0.0 0.0 2.7 a 7.8 a 11.5 a 19.6 a 4-6-gfp + P. expansum 0.0 0.0 0.0 b 3.0 b 8.4 a 16.9 a 86 a a a a a a a a a b b b0 2 4 6 8 0 10 20 30 40 50 60 log 10 CF U/ ml Days in Storage (d) 4-6-gfp 4-6-gfp + P. expansum Figure 27. Bacterial survival of 4-6-gfp (♦) and 4-6-gfp + P. expansum (●) on Gala apples stored for 55 d at 1°C in air using the dilution plating method CFU/ml means with different letters within sampling times are different at significance levels p≤0.05 according to LSD test (n = 4). 87 CHAPTER 4: DISCUSSION 4.1 DNA macroarray for post-harvest apple pathogens 4.1.1 DNA macroarray quantification In the current study, it was hypothesized that macroarray technology would accurately identify and quantify post-harvest apple pathogens throughout the growing season and that the frequency of pathogen detection would correlate with disease incidence post-harvest. This study reports that post-harvest pathogen quantification can be accomplished using DNA macroarray technology and that apple post-harvest pathogens can be detected throughout the apple growing season. Although many studies have used macroarray technology to detect and identify various plant pathogens and biological control microorganisms (Fessehaie et al. 2003; Izzo and Mazzola 2009; Le Floch et al. 2007; Levesque et al. 1998; Lievens et al. 2003; Robideau et al. 2008; Sholberg et al. 2005a; Tambong et al. 2006), few have quantified macroarray hybridization signal intensity (Lievens et al. 2005; Lievens et al. 2007; Sholberg et al. 2006) as we have done in this study. Pathogen DNA was quantified by hybridizing 0 – 100 ng of pure culture DNA extracts of P. expansum, B. cinerea and M. piriformis to the macroarray. The detector probes or oligonucleotide sequences had varying levels of hybridization signal intensity. As found by Sholberg et al. (2006) who correlated greyscale value and cell density, a linear relationship was found between greyscale value and DNA concentration. However, in this study, the hybridization signals became saturated outside of the linear range at >3 ng for P. expansum, >12 ng for B. cinerea and >0.5 ng for M. piriformis (data not shown). Lievens et al. (2005), who quantified DNA concentrations of 88 phytopathogenic fungi, Verticillium albo-atrum, Verticillium dahlia, Fusarium oxysporum, Fusarium solani, Pythium ultimum and Rhizoctonia solani, also reported a linear relationship between signal intensity and DNA concentration. Linearity was established when 0.5 fmol of detector probe was spotted on a membrane and template concentrations ranged from 0.25 pg to 2.5 ng which was in the same range or somewhat lower than the linear range in the present study. Below and above 0.5 fmol, the curves deviated from linearity. Lievens et al. (2007) also reported that hybridization signals between 0.25 to 25 pg (0.00025 to 0.025 ng), that detect Trichoderma spp. and Trichoderma hamatum, a disease suppressive organism, resulted in a logarithmic relationship between signal intensity and template DNA concentration. This was lower than the linear range reported in the present study. Above 25 pg, the standard curves deviated from linearity. Although a range of detector probe concentrations were not tested within this study, Lievens et al. (2005) noted that altering the amount of oligonucleotide spotted on the membrane can affect the range of DNA concentrations detected by a specific macroarray probe. Once a greater knowledge base exists about the level of post-harvest pathogenic DNA found throughout a crop’s growing season and its correlation with disease manifestation, the DNA oligonucleotide detector probe concentrations can be altered to detect ranges typically found in field conditions. This will allow for more accurate DNA quantification and disease management. For enhanced and more precise quantification, it would be advantageous to simultaneously employ hybridization signal analysis with real-time PCR. The ability to accurately quantify a pathogen is useful in 89 plant disease management as information can be used to assess potential risk and disease development (Lievens and Thomma 2005). 4.1.2 Field monitoring of apple post-harvest pathogens The current study was the first of its kind to monitor post-harvest pathogen presence via molecular technologies prior to commercial storage and disease development. Of the samples collected in each of the three growing season sampling periods, there was a tendency to detect total populations of all three post-harvest fungi more frequently in the early and mid-growing seasons of 2007 (56.6% - early; 60% - mid; 33.3% - late) and 2008 (67.6% - early; 63.2% - mid; 34% - late). This tendency to detect lower pathogen levels from air and plant tissue samples in the late growing season corresponds with the lower pathogen levels detected on whole apples collected at harvest (Table 4). In addition, the increase in early season fungal detection corresponds with results obtained by Teixidó et al. (1999) who observed total bacterial, yeast and filamentous fungal populations on Golden Delicious apples. However, they found an even greater increase in microorganisms at the apple bud stage followed by a decrease in populations at blossom. It was suggested that as buds form blossoms, there is a loss of external structure due to blossom drop, which accounts for the reported decrease in microorganisms after the apple budding stage. In the current investigation, samples were not collected at the apple bud stage. The rise in total fungal propagules may therefore correspond with precipitation events in the mid-growing season of 2007 and the early and mid-growing season of 2008 (Figure 1). Of the three major post-harvest pathogens, P. expansum, on average, was detected most frequently (27.4. ± 3.4%) and was found throughout the growing season. In 2008 in 90 Fields 3 and 12 there was an increased detection frequency at the end of the growing season (Table 3). The abundance of P. expansum found within this study contradicts results of Amiri and Bompeix (2005) and Teixidó et al. (1999) who reported low levels of Penicillium inoculum within the orchard atmosphere. However, the aerial sampling period of Amiri and Bompeix (2005) was between July and September which excludes the early growing season. In contrast to P. expansum, B. cinerea was detected the least number of times (6.2 ± 1.4%) and found throughout the growing season of 2007 and 2008. The most important infective units of B. cinerea are the conidia produced in late winter and early spring from over-wintering mycelia and/or sclerotia on host tissue or soil surfaces (Elad et al. 2007); grey mold is rarely seen in the field (Rosenberger 1990). The low levels of B. cinerea as indicated by the macroarray correspond with these results. M. piriformis was detected the second most frequently (19.2 ± 1.4%) and found most frequently in the mid-growing season of 2007 and the early to mid-growing season of 2008 in DNA isolated from plant tissue samples. These findings support Guo et al. (1999) who found an increase in M. piriformis populations within orchard soil from December to March and April, followed by a decrease afterwards. It was suggested that the increase in M. piriformis corresponds with apples left on the orchard floor and cool weather. Alternatively, these results contradict those found by Dobson and Spotts (1988) who recovered no M. piriformis from orchard air but found an abundance of B. cinerea and P. expansum. Sporangiospores of M. piriformis are soilborne (Michailides and Spotts 1986); greater than 75% are found in the top 2 cm of the soil profile and are dispersed primarily by rain splash and insects (Dobson and Spotts 1988). The detection 91 of an abundance of M. piriformis in this study may correspond to precipitation bursts found in both 2007 and 2008 (Figure 9). Sporangiospores may be aerosolized by the falling rain and then trapped by the aerial spore sampler. Earlier studies, as mentioned above, that detected fungal abundance throughout the orchard growing season utilized culturing methods for fungal detection and identification (Amiri and Bompeix 2005; Dobson and Spotts 1988; Teixidó et al. 1999). In the case of P. expansum and M. piriformis, the macroarray detected the presence of pathogens not commonly found within the orchard. Culturing methodologies for detection and relative quantification may underestimate fungal abundance, thus showing the power of macroarray technology. Data obtained using the macroarray may be useful in developing a disease forecasting model such as that presented by Spotts et al. (2009). For future experiments, it would be beneficial to correlate pathogen findings throughout the growing season with species-specific identification of natural disease incidence found within the packinghouse. In this study, a positive correlation was only evident between natural disease incidence and percent pathogens detected in the late-growing season for aerial and plant tissue samples analyzed together (r = 0.79) and for aerial samples (r = 0.74) when analysed separately from plant tissue samples. This finding corresponds with Lennox et al. (2003) who observed a significant correlation between the density of B. cinerea conidia on the pear fruit surface and grey mold disease incidence in cold storage. In addition, Walter et al. (1997) found a strong correlation between B. cinerea berry fruit contamination and latent boysenberry infection in New Zealand. Given the cost and time involved in collecting and analyzing DNA for macroarray identification and 92 quantification, the majority of samples could be collected, for example, near the end of the growing season. It should be noted, however, that Spotts et al. (2009) found that in creating an at-harvest prediction model for grey mold risk in pears, surface DNA was the least important predictor in the model as opposed to orchard rating and fungicide applications. Also of note, within this study, there was a negative correlation between percent pathogens detected in the early-growing season and natural disease incidence when plant tissue samples were analysed separately from aerial samples (r = -0.79). This data suggests that perhaps plant tissue samples collected in the early-growing season are not optimal predictors for natural disease incidence, thereby supporting the correlations mentioned above. More work is required in order to determine the relatedness between growing season pathogen levels and post-harvest disease prevalence. The variability in the results of this study suggests that a multitude of factors contribute to disease incidence found post-harvest. Coley-Smith et al. (1980) proposed that disease outbursts may be a result of complex interactions between the production and dispersal of various inocula, infection pathogenesis, pathogen survival along with temperature, rainfall, humidity and crop protection, nutrition and phenology. Variability may also be explained by the application of pre-harvest fungicides and their effect on resident microflora. According to Sholberg and Boulé (2008), the application of Nova®, a fungicide used to control powdery mildew, reduced apple leaf and fruit microflora in two years of study. In the current study, Nova® was applied to Field 12 on May 15, 2007 which may help to explain the reduced pathogen loads when compared to the 2008 results (Figure 6). It should not be forgotten that the majority of post-harvest diseases are 93 a result of pathogens entering the fruit via wounds, all of which should be considered in accurate disease forecasting models. 4.2 Biological control of post-harvest pathogens Five putative bacterial control agents were applied to apples harvested from commercial and experimental orchards in the Okanagan Valley, BC. It was hypothesized that one or more of the bacterial antagonists would provide significant pathogen control in semi-commercial and commercial storage conditions. This study showed that the four P. fluorescens isolates and one S. plymuthica isolate exhibited some control and antagonistic efficacy against post-harvest apple pathogens (Table 8). As indicated, Serratia plymuthica isolate 6-25 provided significant control in the greatest number of treatments (57.6%) (Table 8). These results suggest that this bacterium may provide control against a variety of post-harvest diseases and storage conditions and its use may be applied to other crops. Using S. plymuthica as a BCA, however, is a challenging endeavour as concerns surround its biosafety, a major consideration when selecting an antagonist (Droby et al. 2009). Serratia plymuthica belongs to the family Enterobacteriaceae. Members of this family, such as Escherichia coli strain O157:H7, have been known to cause human infections (Health Protection Agency 2007). Although opportunistic pathogenic strains of S. plymuthica have been isolated (Carrero et al. 1995; Clark and Janda 1985; Domingo et al. 1994) infection is rare compared to other members of the genus, Serratia, or of the family, Enterobacteriaceae. Despite concerns surrounding its use as a biological control agent, S. plymuthica isolate 6-25 provided significant control in this study, against apple post-harvest pathogens, P. expansum, B. cinerea and M. piriformis. S. plymuthica biological control 94 efficacy has also been demonstrated by other research groups, although not on post- harvest apple pathogens. S. plymuthica strain HRO-C48 has been studied to control Verticillium dahliae, the causal agent of Verticillum wilt in oil-seed rape (Müller and Berg 2008). HRO-C48 also suppressed growth of V. dahliae and Phytophthora cactorum in strawberries (Kurze et al. 2001) and a commercial product based on strain HRO-C48 has been developed, RhizoStar® (e-nema GmbH, Raisdorf, Germany) (Müller and Berg 2008). S. plymuthica strain 5-6 also showed significant control of dry rot of potato caused by Fusarium sambucinum (Gould et al. 2008). The post-harvest biological control provided by S. plymuthica isolate 6-25 of this study should not be disregarded as the potential exists, as evident by other studies, for commercial application. This study also showed that pseudomonads, specifically P. fluorescens, are also capable of providing control against post-harvest pathogens. According to Stockwell and Stack (2007), Pseudomonas spp., have been studied for decades as model organisms for biological control of plant diseases. In this study the greatest reduction in IS, in cases where significant control was exhibited, was with P. fluorescens isolate 1-112 followed by 1100-6 (Table 8). Other pseudomonads are also being studied for their efficacy against post-harvest pathogens. Mikani et al. (2008) reported significant control of Botrytis mali, formerly thought of as B. cinerea and recently revived by O’Gorman et al. (2008), using 10 strains of P. fluorescens isolated from leaf surfaces and apple fruit. Similar to the level of control found in this study, the results of Mikani et al. (2008), reported approximately 34.9% to 95.9% B. mali inhibition on Golden Delicious apples after 25 d at 5°C. In addition, Zhou et al. (2001) studied four strains of Pseudomonas syringae that were isolated from the phyllosphere of apple trees and controlled blue 95 mold, following inoculation by pipette, by 64 – 70% at 4°C after 28 d. The bacterial isolates also controlled grey mold, but to a lesser extent. Similar to methods used in this study, Zhou et al. (2001) also inoculated apples by dipping and found that a P. syringae isolate controlled blue mold by 93% after 133 d at 1°C in air, with the overall incidence of blue mold at 30%. In many cases, (Figure 12a) the bacterial isolates used in the present study exhibited similar and greater levels of control than those reported by Mikani et al. (2008) and Zhou et al. (2001). The effectiveness of these isolates may be due to the unique environment in which they were isolated. Many BCAs are found on the fruit surface or directly within a wound. The microbial agents used in this study, however, were isolated from the rhizosphere of legumes grown in cold Saskatchewan soils and may be more adapted to cold storage conditions. Droby et al. (2009) outlined the importance of antagonist selection and the use of a variety of screening procedures in finding effective BCAs. An assortment of selection techniques would increase biocontrol species diversity and therefore the potential for a successful product. These experiments illustrate the potential of S. plymuthica and P. fluorescens isolates for biological control application and commercialization. 4.2.1 Bacterial antagonists and CA storage One type of commercial storage application used in this study was controlled atmosphere (CA). This research confirmed the hypothesis that one or more antagonists would provide control in commercial conditions. In all experiments, the lowest disease levels were found in apples stored in CA storage for 4.5 months following harvest in 2008. Many suggestions have been made regarding the selection criteria for potential BCAs (Wilson and Wisniewski 1989). One such criterion is the ability of the antagonist 96 to be compatible with current post-harvest commercial processing procedures such as CA (Janisiewicz and Korsten 2002; Wilson and Wisniewski 1989). This current study indicated that the selected antagonists were compatible with CA storage practices and may actually extend apple storage time due to enhanced pathogen suppression when compared to storage at 1°C. For example, the highest disease incidence recorded in CA was in Jonagold apples which reached an average IS of 10.39 ± 1.6 after 4.5 months in storage (Figure 10). The biological antagonists provided significant control against naturally occurring diseases in all orchards but Field 12 and the Reekie orchard (data not shown). In comparison, for Jonagold and Gala apples grown in Fields 3 and 12 then stored at 1°C in air for 4 months, non-inoculated controls had IS levels of 15.65 ± 5.0 (Figure 16b) and 19.90 ± 4.8 (Figure 16e), respectively. These values were almost double the infection severities found in CA. In both Jonagold and Gala apples harvested from Fields 3 and 12, respectively, the bacterial antagonists did not provide a significant reduction in IS after 4 months in storage (Figure 16b and e). However, isolate 6-25 reduced IS in Jonagold apples grown in Field 3 after 6 months in storage (Figure 16c). The decrease in apple disease supports the findings of Smock (1979) as CA effectively delayed the onset of storage decay. In the CA-stored apples, natural disease incidence was very low even after 4.5 months of storage, except in the case of Jonagold apples from Field 3 which appeared to be due primarily to P. expansum (based on phenotypic observation). The gas composition of a packinghouse can directly or indirectly suppress fungal decay. Direct inhibition affects mycelium development and spore germination of resident pathogens. Alternatively, storage conditions also affect senescence and ripening of hosts, which, in 97 turn, affects their ability to resist infections, thereby indirectly controlling fungal decay (Barkai-Golan 2001). The implementation of CA is a balance between O2 and CO2 concentrations. General recommendations include 1 – 2 % O2 and 0.5 – 2% CO2. Concentrations outside of these ranges may lead to fruit injury which can affect fruit flavour and appearance (Kupferman 2003). However, post-harvest pathogens are only strikingly suppressed after O2 concentrations have been lowered to less than 1% and CO2 concentrations increased to greater than 10% (El-Goorani and Sommer 1981). The CA storage conditions in this study were approximately 1.5% O2 and 1.5% CO2, a range in which, if the conditions were right, may lead to the fungal outbreak that was observed in the Jonagold apples. In the current study, the antagonistic effects of the bacterial isolates were difficult to observe because there was little natural disease present; if little disease was present, then there was little disease to control. However, the outbreak in Jonagold apples from Field 3 demonstrated the antagonistic potential of the bacteria because significant disease was present. While all of the antagonists demonstrated significant control, isolate 1100-6 and 1-112 were particularly effective, providing 94.9% and 83.8% reduction in disease symptoms in Jonagold apples from Field 3 (Figure 17). Although not directly inoculated with a pathogen, these BCAs performed better than the commercially registered BCA, Bio-Save 10 LPTM, in studies conducted by Janisiewicz and Jeffers (1997) and Zhou et al. (2001). Using Bio-SaveTM during 133 and 123 days in CA storage, Zhou et al. (2001) reported a 57% and a 73% reduction of blue mold in Empire and Delicious apples, respectively. It would be beneficial to test the efficacy of the bacterial antagonists against artificially inoculated pathogens under CA conditions as in Conway et al. (2007), 98 Janisiewicz et al. (2008) and Tian et al. (2002). However, during the present study, permission for pathogen inoculation within a commercial storage facility was not given. 4.2.2 Bacterial antagonists and 1-MCP Another type of commercial storage application tested in this study was the use of 1-MCP. This research confirmed the hypothesis that one or more antagonists would provide control in commercial conditions, although to a lesser extent. 1-MCP was applied to Jonagold apples grown in Field 3 and Gala apples grown in Field 12 in the biological control experiments using apples from the 2007 harvest. It was postulated that the application of 1-MCP would result in a decrease in disease incidence and severity of non-inoculated wounded apples or apples inoculated with putative biological control agents and/or post-harvest pathogens. However, when the overall effect of 1-MCP was examined, a reduction in infection severity occurred in only 14% of the experimental treatments. For example, 1-MCP application reduced IS caused by natural disease incidence in Jonagold apples from Field 3 (Table 5) by 51.8% and in Gala apples from Field 12 (Figure 11) by 67.1%. Variable levels of control confirm findings of Saftner et al. (2003) who studied the effects of pre-storage heat, CA storage and pre- and post- application of MCP on inhibition of fungal decay and maintenance of Golden Delicious fruit quality. Pre-storage treatment of MCP reduced lesion size (mm) of fruit inoculated with B. cinerea and P. expansum by 34% and 17.9%, respectively. Post-storage application of MCP had no effect on reducing lesion diameter. Variable responses to 1- MCP application are not uncommon. Watkins (2008) indicated that the apple is a fruit with many cultivars that differ in their ripening rates, harvest criteria, post-harvest handling practices and storage periods in air and CA. In addition, time between harvest 99 and 1-MCP treatment and repeated application have variable effects in disease resistance responses. In the current study, when a decrease in infection severity was observed, it appeared to be primarily in experiments where there was no artificial inoculation of P. expansum, B. cinerea or M. piriformis; the only pathogens present were those that occurred naturally. Contradictory to the intended outcome of 1-MCP application, some treatments in the present study resulted in an increase in infection severity of apples treated with 1- MCP, especially those apples inoculated with B. cinerea (Figure 12b). An increase in decay after 1-MCP application was also observed by Janisiewicz et al. (2003) and Leverentz et al. (2003) on Golden Delicious apple fruit and by Bedford et al. (2002) who reported an increase in B. cinerea decay when apples were fumigated with 1-MCP. Commercial application of 1-MCP as SmartFresh® is thought to delay apple ripening by preventing ethylene production via preferential binding to the ethylene receptor site and extending the action of natural defence mechanisms due to delayed ripening (Blankenship and Dole 2003). However, the reverse was observed in some cases as 1- MCP appeared to hinder plant defence mechanisms. According to Marcos et al. (2005), endogenous ethylene may be an important component in plant resistance and defence gene regulation; the application of 1-MCP may therefore compromise a plant’s defence response system (Jiang et al. 2001). Despite the controversy surrounding chemical control agents, fungicides still remain the preferred choice for the prevention of post-harvest diseases. Although fungicide-resistant pathogens have increased within the population, chemical control provides a consistent and effective reduction in disease. Within the agricultural 100 community, there is a relatively high level of scepticism surrounding the use of biological control products (Droby et al. 2009). Commercially acceptable decay levels are generally below 5% (Janisiewicz and Korsten 2002) and because of the inconsistency and variability of biofungicides, the packinghouse industry will often choose the most economically viable option that consistently reduces rot. Disease control was exhibited by the bacterial antagonists in this study, but efficacy was highly variable, as observed by other researchers. The tritrophic interaction between host, pathogen and antagonist is a highly complex one that demands a more sound and fundamental understanding. In addition to exploring the mechanism(s) of action for fungal suppression, combination and alternative treatments for enhanced efficacy should be explored. Such alternate biocontrol treatments, outlined by Sharma et al. (2009) include, for example, the addition of low doses of fungicides, salt additives, or nutrients and plant products, as well as the use of mixed cultures or manipulation of the physical environment. Rapid antagonist colonization of fruit wounds is critical to controlling decay; therefore, enhancements leading to improved wound colonization should be further explored (Janisiewicz and Korsten 2002). 4.3 Bacterial survival on apple Acquiring basic knowledge about potential antagonists will provide a greater understanding into the mechanisms of actions, which will, in turn, offer insight into means for achieving more consistent control. Disease prevention relies on a quantitative relationship between antagonist and pathogen (Janisiewicz 1988) and knowledge of antagonist cell number during time in storage will assist in enhanced disease suppression (Etebarian and Sholberg 2005). In the present study, Pseudomonas fluorescens isolate 4- 101 6 was tagged with GFP to determine the survival characteristics of this antagonist alone and when challenged with P. expansum. It was hypothesized that green fluorescent protein would facilitate visualization of bacterial colonization and survival on apple surfaces. In this study, gfp did facilitate bacterial visualization, but indirectly as pAG408 provided antibiotic resistance of isolate 4-6 which permitted selective plating that excluded other organisms during survival analysis. Bacterial concentrations were determined using direct scanning and colony counts. There was a strong linear relationship between fluorescent intensity and cell density (Figure 26). These results corresponded with those found by Etebarian and Sholberg (2005) who studied population levels of P. fluorescens isolate 1100-6 labelled with GFP using direct scanning, cell counts and microscopy. This and the current study illustrated that measuring fluorescent intensity as a means of estimating cell concentration was an effective method of population assessment. The initial inoculation concentration of 4-6-gfp (108 CFU/ml), however, combined with the method of recovery, resulted in fluorescent readings that were not high enough to quantify using the linear equation y = 1.8041x + 6.5652 (Figure 26). Using a Fast Prep machine for tissue maceration as in Etebarian and Sholberg (2005) or in Etebarian et al. (2005) may overcome issues incurred by extraction. Despite low values, the raw microplate data were compared to plate counting data. Relative fluorescent units (RFUs) (Figure 27) and cell densities (CFU/ml) (Figure 28) of isolate 4-6-gfp followed similar patterns; when cell counts increased, so did RFUs. However, there was an inverse relationship between RFUs and CFU/ml for 4-6-gfp inoculated in the presence of P. expansum; RFUs increased while cell counts decreased. 102 Two possible explanations are that P. expansum auto-fluoresces or that the decayed apple tissue auto-fluoresces. The latter explanation corresponds with the formation of lesions at 30 d (Table 10). Increased antagonist populations after 30 d of storage were also found by Janisiewicz and Marchi (1992) who reported a 10 – 100 fold increase in populations of Pseudomonas syringae inoculated on pears stored at 1°C. Etebarian and Sholberg (2005) also found an increase in P. fluorescens isolate 1100-6 GFP on apples after 35 days in 5°C storage. However, unlike the survival characteristics of 4-6-gfp + P. expansum in this study, 1100-6 GFP + P. expansum maintained as high or higher cell numbers than 1100-6 GFP alone. Low cell counts after lesion development may be attributed to difficulties in extracting apple cores from completely decayed tissue. Even so, after 30 d at 1°C in air storage, 4-6-gfp appeared to lose some of its antagonistic capabilities (Table 10). Information such as this should be considered when implementing control strategies for optimal post-harvest disease management. 103 CHAPTER 5: CONCLUSIONS This thesis built upon preliminary research and addressed the following three objectives: 1) to validate the performance of DNA macroarrays under field conditions; 2) to assess bacterial antagonist performance under semi-commercial and commercial conditions and; 3) to determine how long and in what capacity a selected antagonist will colonize the fruit surface. The first hypothesis addressed the first objective and stated that macroarray technology would accurately identify and quantify post-harvest apple pathogens throughout the growing season and that disease prevalence would correlate with disease incidence post-harvest. Results confirmed that in two years of field trials, the DNA macroarray is a fast and easy technique capable of detecting apple post-harvest pathogens throughout the apple growing season (Table 3). Commercial application of this diagnostic and quantitative technique would provide accurate information to assist growers and field personnel in crop management. A positive correlation existed between pathogen detection at the end of the growing season and natural disease incidence found post-harvest. More work is required to obtain information about the importance of monitoring pre-harvest pathogen loads and their contribution to post-harvest disease. The second hypothesis addressed the second objective and stated that one or more of the bacterial antagonists would provide significant pathogen control in semi- commercial and commercial storage conditions. Results indicated that despite variability between years, variety and biological control efficacy, the potential BCAs demonstrated control at 1°C in air and particularly in CA storage (Table 8). Biocontrol efficacy in combination with 1-MCP was less effective and the application of 1-MCP greatly 104 increased disease levels with B. cinerea artificial inoculation (Figure 12b). Combining some or all of the bacterial control agents used in this study in conjunction with other organic amendments such as salts (ie. sodium bicarbonate) or manipulating the physical environment (ie. heat treatment) may provide consistency in control which the industry demands. The third hypothesis addressed the third objective and stated that the use of green fluorescent protein would facilitate visualization of bacterial colonization and survival on apple surfaces. Results indicated that green fluorescent protein was an effective technique to monitor population levels as the gfp-marked strain behaved similarly to the wild-type strain (Figure 25; Table 9). The fluorescence levels obtained by direct scanning, however, were not sufficient to distinguish 4-6-gfp from background auto- fluorescence. The presence of the gfp-containing plasmid within P. fluorescens isolate 4- 6 therefore provided an indirect method of monitoring population levels as pAG408 contained antibiotic resistant genes that allowed for strain-specific isolation from an inoculated apple wound. Bacterial populations were found to initially increase then decrease within an apple wound during 55 d in storage at 1°C (Figure 28). In the presence of P. expansum, 4-6-gfp populations declined, which coincided with P. expansum lesion development. The information obtained within this study will assist the fruit industry in detecting, quantifying and biologically controlling post-harvest pathogens of pome fruit. 105 REFERENCES AAFC 2002. Canadian Apple Online [online]. Available from http://ats.agr.ca/applecanada/home-e.htm [accessed 6 March 2008]. Abramson, D., Lombaert, G., Clear, R.M., Sholberg, P., Trelka, R., and Rosin, E. 2009. Production of patulin and citrinin by Penicillium expansum from British Columbia (Canada) apples. Mycotox. Res. 25(2): 85-88. Amiri, A., and Bompeix, G. 2005. Diversity and population dynamics of Penicillium spp. on apples in pre- and postharvest environments: consequences for decay development. Plant Pathol. 54(1): 74-81. Bailey, K.L., Boyetchko, S.M., and Längle, T. 2010. Social and economic drivers shaping the future of biological control: a Canadian perspective on the factors affecting the development and use of microbial biopesticides. Biol. Control 52(3): 221-229. Barkai-Golan, R. 2001. Postharvest diseases of fruits and vegetables: development and control. Elsevier Science B.V. Amsterdam, The Netherlands. Bedford, K.E., Stokes, S.C., Sholberg, P.L., and Lau, O.L. 2002. Effects of 1-MCP on postharvest decay of apples caused by Penicillium expansum and Botrytis cinerea. Can. J. Plant Pathol. 24(3): 381. Bencheqroun, S.K., Bajji, M., Massart, S., Labhilili, M., El Jaafari, S., and Jijakli, M.H. 2007. In vitro and in situ study of postharvest apple blue mold biocontrol by Aureobasidium pullulans: Evidence for the involvement of competition for nutrients. Postharvest Biol. Technol. 46(2): 128-135. 106 Benhamou, N., Gagné, S., Le Quéré, D., and Dehbi, L. 2000. Bacterial-mediated induced resistance in cucumber: beneficial effect of the endophytic bacterium Serratia plymuthica on the protection against infection by Pythium ultimum. Biochem. Cell Biol. 90(1): 45-56. Blankenship, S.M., and Dole, J.M. 2003. 1-methylcyclopropene: a review. Postharvest Biol. Technol. 28(1): 1-25. Bloemberg, G.V. 2007. Microscopic analysis of plant–bacterium interactions using auto fluorescent proteins. Eur. J. Plant Pathol. 119(3): 301–309. Bloemberg, G.V., O’Toole, G.A., Lugtenberg, B.J.J., and Kolter, R. 1997. Green fluorescent protein as a marker for Pseudomonas spp. Appl. Environ. Microbiol. 63(11): 4543-4551. Bolwerk, A., Lagopodi, A.L., Wijfjes, A.H.M., Lamers, G.E.M., Chin-A-Woeng, T.F.C., Lugtenberg, B.J.J., and Bloemberg, G.V. 2003. Interactions in the tomato rhizosphere of two Pseudomonas biocontrol strains with the phytopathogenic fungus Fusarium oxysporum f. sp. radicis-lycopersici. Mol. Plant Microbe Interact. 16(11): 983-993. Bridge, P., and Spooner, B. 2001. Soil fungi: diversity and detection. Plant Soil 232(1-2): 147-154. British Columbia Ministry of Agriculture, Food and Fisheries. 2004. Industry Overview: Apples and Pears in British Columbia [online]. Available from http://www.agf.gov.bc.ca/treefrt/pubs/pubs_toc.htm [accessed 24 February 2008]. Calvo, J., Calvente, V., de Orellano, M.E., Benuzzi, D., and de Tosetti, M.I.S. 2007. Biological control of postharvest spoilage caused by Penicillium expansum and 107 Botrytis cinerea in apple by using the bacterium Rahnella aquatilis. Int. J. Food Microbiol. 113(3): 251-257. Carrero, P., Garrote, J.A., Pacheco, A., Garcia, A.I., Gil, R. and, Carbajosa, S.G. 1995. Report of six cases of human infection by Serratia plymuthica. J. Clin. Microbiol. 33(2): 275-275. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W.W., and Prasher, D.C. 1994. Green fluorescent protein as a marker for gene expression. Sci. 263(5148): 802-805. Chan, Z., and Tian, S. 2005. Interaction of antagonistic yeasts against postharvest pathogens of apple fruit and possible mode of action. Postharvest Biol. Technol. 36(2): 215-223. Clark, R.B., and Janda, J.M. 1985. Isolation of Serratia plymuthica from a human burn site. J. Clin. Microbiol. 21(4): 656-657. Coley-Smith, J.R., Verhoeff, K., and Jarvis, W.R. 1980. The Biology of Botrytis. Academic Press, New York, NY. Conway, W.S., Janisiewicz, W.J., Leverentz, B., Saftner, R.A., and Camp, M.J. 2007. Control of blue mold of apple by combining controlled atmosphere, an antagonist mixture, and sodium bicarbonate. Postharvest Biol. Technol. 45(3): 326-332. Cook, R.J., and Baker, K.F. 1983. The nature and practice of biological control of plant pathogens. The American Phytopathological Society, St. Paul, MN. Cossentine, J.E., Sholberg, P.L., Jensen, L.B.J., Bedford, K.E., and Shephard, T.C. 2004. Fumigation of empty fruit bins with carbon dioxide to control diapausing coldling moth larvae and Penicillium expansum Link. ex Thom spores. HortSci. 39(2): 429- 432. 108 Cruickshank, R.H., and Pitt, J.I. 1987. Identification of species in Penicillium subgenus Penicillium by enzyme electrophoresis. Mycol. 79(4): 614-620. de Jong, S.N., Lévesque, C.A., Verkley, G.J.M., Abeln, E.C.A., Rahe, J.E., and Braun, P.G. 2001. Phylogenetic relationships among Neofabraea species causing tree cankers and bull's-eye rot of apple based on DNA sequencing of ITS nuclear rDNA, mitochondrial rDNA, and the β-tubulin gene. Mycol. Res. 105(6): 658-669. Dobson, R.L., and Spotts, R.A. 1988. Distribution of sporangiospores of Mucor piriformis in pear orchard soils. Plant Dis. 72(8): 702-705. Domingo, D., Limia, A., Alarcon, T., Sanz, J.C., Del Rey, M.C. and, Lopez-Brea, M. 1994. Nosocomial septicemia caused by Serratia plymuthica. J. Clin. Microbiol. 32(2): 575-577. Doores, S. 1983. The microbiology of apples and apple products. Crit. Rev. Food Sci. Nutr. 19: 133-149. Droby, S., Wisniewski, M., Macarisin, D., and Wilson, C. 2009. Twenty years of postharvest biocontrol research: Is it time for a new paradigm? Postharvest Biol. Technol. 52(2): 137-145. Eckert, J.W., and Ogawa, J.M. 1988. The chemical control of postharvest diseases: Deciduous fruits, berries, vegetables and root/tuber crops. Ann. Rev. Phytopathol. 26: 433-469. Eisele, T.A., and Drake, S.R. 2005. The partial compositional characteristics of apple juice from 175 apple varieties. J. Food Compost. Anal. 18(2-3): 213-221. Elad, Y. 1996. Mechanisms involved in the biological control of Botrytis cinerea incited diseases. Eur. J. Plant Pathol. 102(8): 719-732. 109 Elad, Y., Williamson, B., Tudzynski, P., and Delen, N. 2007. Botrytis: biology, pathology and control. Springer, Dordrecht, The Netherlands. El-Goorani, M.A., and Sommer, M.F. 1981. Effects of modified atmospheres of postharvest pathogens of fruits and vegetables. Hort. Rev. 3: 412-461. Ellis, M.A., and Barrat, G.J. 1983. Colonization of Delicious apple fruits by Alternaria spp. and effect of fungicide sprays on moldy-core. Plant Dis. 67(2): 150-152. Errampalli, D. 2004. Effect of fludioxonil on germination and growth of Penicillium expansum and decay in apple cvs. Empire and Gala. Crop Prot. 23(9): 811-817. Errampalli, D., Brubacher, N., and DeEll, J. 2006. Sensitivity of Penicillium expansum to diphenylamine and thiabendazole and postharvest control of blue mold with fludioxonil in ‘McIntosh’ apples. Postharvest Biol. Technol. 39(1): 101-107. Etebarian, H.R., Sholberg, P.L., Eastwell, K.C., and Sayler, R.J. 2005. Biological control of apple blue mold with Pseudomonas fluorescens. Can. J. Microbiol. 51(7): 591- 598. Etebarian, R.H., and Sholberg, P.L. 2005. Monitoring population levels of Pseudomonas fluorescens labelled with green fluorescent protein, using fluorescence microscopy and direct fluorescence scanning techniques. Can. J. Plant Pathol. 28(1): 125-130. FAOSTAT. 2009. FAO statistics division [online]. Available from http://faostat.fao.org [accessed 13 January 2010]. Fessehaie, A. De Boer, S.H., and Lévesque, A.C. 2003. An oligonucleotide array for the identification and differentiation of bacteria pathogenic on potato. Phytopathol. 93(3): 262-269. 110 Frankowski, J., Lorito, M., Scala, F., Schmid, R. Berg, G., and Bahl, H. 2001. Purification and properties of two chitinolytic enzymes of Serratia plymuthica HRO-C48. Arch. Microbiol. 176(6): 421-426. Gould, M., Nelson, L.M., Waterer, D., and Hynes, R.K. 2008. Biocontrol of Fusarium sambucinum, dry rot of potato, by Serratia plymuthica 5-6. Biocontrol Sci. Technol. 18(9): 1005-1016. Guo, L., Michailides, T.J., and Morgan, D.P. 1999. Survival of Mucor piriformis in soil of apple orchards in California. Plant Dis. 83(2): 189-193. Hancock, D.B., Martin, E.R., Mayhew, G.M., Stajich, J.M., Jewett, R., Stacy, M.A., Scott, B.L., Vance, J.M., and Scott, W.K. 2008. Pesticide exposure and risk of Parkinson's disease: A family-based case-control study. BMC Neurol. 8(6). Health Protection Agency. 2007. Identification of Enterobacteriaceae [online]. Available from www.hpa-standardmethods.org.uk/documents/bsopid/pdf/bsopid16.pdf [accessed 14 April 2008]. Heim, R., Cubitt, A.B., and Tsien, R.Y. 1995. Improved green fluorescence. Nature 373(6516): 663-664. Hirkala, D., Nelson, L., Stokes, S., and Sholberg, P. 2007. Postharvest biological control of blue and grey mold on apples. Can. J. Plant Pathol. 29(1): 84-85. Hynes, R.K., Leung, G.C., Hirkala, D.L., and Nelson, L.M. 2008. Isolation, selection and characterization of beneficial rhizobacteria from pea, lentil and chickpea grown in western Canada. Can. J. Microbiol. 54(4): 248-258. 111 Izzo, A.B., and Mazzola, M. 2009. Hybridization of an ITS-based macroarray with ITS community probes for characterization of complex communities of fungi and fungal-like protists. Mycol. Res. 113: 802-812. Janisiewicz, W., Yourman, L., Roitman, J., and Mahoney, N. 1991. Postharvest control of blue mold and grey mold of apples and pears by dip treatment with pyrrolnitrin, a metabolite of Pseudomonas cepacia. Plant Dis. 75(5): 490-494. Janisiewicz, W.J. 1988. Biocontrol of postharvest diseases of apples with antagonist mixtures. Phytopathol. 78(2): 194-198. Janisiewicz, W.J., and Korsten, L. 2002. Biological control of postharvest diseases of fruit. Annu. Rev. Phytopathol. 40: 411-441. Janisiewicz, W.J., and Jeffers, S.N. 1997. Efficacy of commercial formulation of two biofungicides for control of blue mold and gray mold of apples in cold storage. Crop Prot. 16(7): 629-633. Janisiewicz, W.J., and Marchi, A. 1992. Control of storage rots on various pear cultivars with a saprophytic strain of Pseudomonas syringae. Plant Dis. 76(6): 555-560. Janisiewicz, W.J., Leverentz, B., Conway, W.S., Saftner, R.A., Reed, A.N., and Camp, M.J. 2003. Control of bitter rot and blue mold of apples by integrating heat and antagonist treatments on 1-MCP treated fruit stored under controlled atmosphere conditions. Postharvest Biol. Technol. 29(2): 129-143. Janisiewicz, W.J., Saftner, R.A., Conway, W.S., and Yoder, K.S. 2008. Control of blue mold decay of apple during commercial controlled atmosphere storage with yeast antagonists and sodium bicarbonate. Postharvest Biol. Technol. 49(3): 374-378. 112 Janisiewicz, W.J., Tworkoski, T.J., and Sharer, C. 2000. Characterizing the mechanism of biological control of postharvest diseases on fruits with a simple method to study competition for nutrients. Phytopathol. 90(11): 1196-1200. Jiang, Y., Joyce, D.C., and Terry, L.A. 2001. 1-Methylcyclopropene treatment affects strawberry fruit decay. Postharvest Biol. Technol. 23(3): 227-232. Jijakli, M.H., and Lepoivre, P. 2004. Disease management of fruits and vegetables. In Fruit and vegetable diseases. Edited by K.G. Mukerji. Kluwer Academic Publishers, Dordrecht, The Netherlands. pp. 59-94. Jones, A.L., and Aldwinckle, H.S. 1990. Compendium of apple and pear diseases. The American Phytopathological Society, St. Paul, Minnesota. Kamensky, M., Ovadis, M., Chet, I., and Chernin, L. 2003. Soil-borne strain IC14 of Serratia plymuthica with multiple mechanisms of antifungal activity provides biocontrol of Botrytis cinerea and Sclerotinia sclerotiorum diseases. Soil Biol. Biochem. 35(2): 323-331. Kupferman, E. 2003. Controlled atmosphere storage of apples and pears. Postharvest Information Network, Tree Fruit Research and Extension Center, Washington State University, Wenatchee, WA. Kurze, S., Bahl, H., Dahl, R., and Berg, G. 2001. Biological control of fungal strawberry diseases by Serratia plymuthica HRO-C48. Plant Dis. 85(5): 529-534. Le Floch, G., Tambong, J., Vallance, J., Tirilly, Y., Lévesque, A., and Rey, P. 2007. Rhizosphere persistence of three Pythium oligandrum strains in tomato soilless culture assessed by DNA macroarray and real-time PCR. FEMS Microbiol. Ecol. 61(2): 317-326. 113 Lennox, C.L., Spotts, R.A., and Cervantes, L.A. 2003. Populations of Botrytis cinerea and Penicillium spp. on pear fruit, and in orchards and packinghouses, and their relationship to postharvest decay. Plant Dis. 87(6): 639-644. Leverentz, B., Conway, W.S., Janisiewicz, W.J., Saftner, R.A., and Camp, M.J. 2003. Effect of combining MCP treatment, heat treatment, and biocontrol on the reduction of postharvest decay of 'Golden Delicious' apples. Postharvest Biol. Technol. 27(3): 221-233. Levesque, C.A., Harlton, C.E., and de Cock, A.W.A.M. 1998. Identification of some oomycetes by reverse dot blot hybridization. Phytopathol. 88(3): 213-222. Lievens, B., and Thomma, P.H.J. 2005. Recent developments in pathogen detection arrays: implications for fungal plant pathogens and use in practice. Phytopathol. 95(12): 1374-1380. Lievens, B., Brouwer, M., Vanachter, A.C.R.C, Lévesque, A., Cammue, B.P.A., and Thomma, B.P.H.J. 2003. Design and development of a DNA array for rapid detection and identification of multiple tomato vascular wilt pathogens. FEMS Microbiol. Lett. 223(1): 119-122. Lievens, B., Brouwer, M., Vanachter, A.C.R.C, Lévesque, A.C., Cammue, B.P.A, and Thomma, B.P.H.J. 2005. Quantitative assessment of phytopathogenic fungi in various substrates using a DNA macroarray. Environ. Microbiol. 7(11): 1698-1710. Lievens, B., Claes, L., Krause, M.S., Vanachter, A.C.R.C., Cammue, B.P.A., and Thomma, B.P.H.J. 2007. Assessing populations of a disease suppressive microorganism and a plant pathogen using DNA arrays. Plant Sci. 172(3): 505-514. 114 Marcos, J.F., González-Candelas, L., and Zacarías, L. 2005. Involvement of ethylene biosynthesis and perception in the susceptibility of citrus fruits to Penicillium digitatum infection and the accumulation of defence-related mRNAs. J. Exp. Bot. 56(418): 2183-2193. Mazzola, M. 2004. Assessment and management of soil microbial community structure for disease suppression. Annu. Rev. Phytopathol. 42: 35–59. Meziane, H., Gavriel, S., Ismailov, Z., Chet, I., Chernin, L., and Höfte, M. 2006. Control of green and blue mould on orange fruit by Serratia plymuthica strains IC14 and IC1270 and putative modes of action. Postharvest Biol. Technol. 39(2): 125-133. Michailides, T.J., and Spotts, R.A. 1986. Factors effecting dispersal of Mucor piriformis in pear orchards and into the packinghouse. Plant Dis. 70(11): 1060-1063. Michailides, T.J., and Spotts, R.A. 1990. Postharvest disease of pome and stone fruits caused by Mucor piriformis in the Pacific Northwest and California. Plant Dis. 74(8): 537-543. Mikani, A., Etebarian, H.R., Sholberg, P.L., O’Gorman, D.T., Stokes, S., and Alizadeh, A. 2008. Biological control of apple grey mold caused by Botrytis mali with Pseudomonas fluorescens strains. Postharvest Biol. Technol. 48(1): 107-112. Miller, P. M. 1959. Open calyx tubes as a factor contributing to carpel discoloration and decay of apples. Phytopathol. 49(8): 520-523. Morales, H., Sanchis, V., Rovira, A., Ramos, A., and Marin, S. 2007. Patulin accumulation in apples during postharvest: Effect of controlled atmosphere storage and fungicide treatments. Food Cont. 18(11): 1443-1448. 115 Morales, H., Sanchis, V., Usall, J., Ramos, A.J., and Marín, S. 2008. Effect of biocontrol agents Candida sake and Pantoea agglomerans on Penicillium expansum growth and patulin accumulation in apples. Int. J. Food Microbiol. 122(1-2): 61-67. Morin, J.G., and Hastings, J.W. 1971. Energy transfer in a bioluminescent system. J. Cell. Physiol. 77(3): 313. Morise, H., Shimomura, O., Johnson, F., and Winant, J. 1974. Intermolecular energy transfer in the bioluminescent system of Aequorea. Biochem. 13(12): 2656–62. Müller, H., and Berg, G. 2008. Impact of formulation procedures on the effect of the biocontrol agent Serratia plymuthica HRO-C48 on Verticillium wilt in oilseed rape. Biocont. 53(6): 905-916. Nunes, C., Usall, J., Teixido, N., Fons, E., and Vinas, I. 2002. Post-harvest biological control by Pantoea agglomerans (CPA-2) on Golden Delicious apples. J. Appl. Microbiol. 92(2): 247-255. O’Gorman, D.T., Sholberg, P.L., Stokes, S.C., and Ginns, J. 2008. DNA sequence analysis of herbarium specimens facilitates the revival of Botrytis mali, a postharvest pathogen of apple. Mycol. 100(2): 227-235. Ormo, M., Cubitt, A.B., Kallio, K., Gross, L.A., Tsien, R.Y., and Remington, S.J. 1996. Crystal structure of the Aequorea victoria green fluorescent protein. Sci. 273(5280): 1392-1395. Pest Management Regulatory Agency. 2007. Available from http://www.pmra- arla.gc.ca/english/pubs/pubs-e.html [accessed 9 March 2008]. 116 Piemontese, L., Solfrizzo, M., and A. Visconti. 2005. Occurrence of patulin in conventional and organic fruit products in Italy and subsequent exposure assessment. Food Add. Contamin. 22: 437-442. Prasher, D., Eckenrode ,V., Ward, W., Prendergast, F., and Cormier, M. 1992. Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111(2): 229–33. Robideau, G.P., Caruso, F.L., Oudemans, P.V., McManus, P.S., Renaud, M.A., Auclair, M.E., Bilodeau, G.J., Yee, D., Désaulniers, N.L., DeVerna, J.W., and Lévesque, C.A. 2008. Detection of cranberry fruit rot fungi using DNA array hybridization. Can. J. Plant Pathol. 30(2): 226-240. Rosenberger, D.A. 1990. Blue mold. In Compendium of apple and pear diseases. Edited by A.L. Jones, and H.S. Aldwinkle. The American Phytopathological Society, St. Paul, Minn. pp. 54–55. Saftner, R.A, Abbott, J.A., and Conway, W.S. 2003. Effects of 1-methylcyclopropene and heat treatments on ripening and postharvest decay in Golden Delicious apples. J. Amer. Soc. Hort. Sci. 128(1): 120-127. Saiki, R.K., Scharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A., and Arnheim, N. 1985. Enzymatic amplification of β-globlin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Sci. 230(4732): 1350-1354. Sanderson, P.G., and Spotts, R.A. 1995. Postharvest decay of winter pear and apple fruit caused by species of Penicillium. Phytopathol. 85(1): 103-110. Shaner, N.C., Patterson, G.H., and Davidson, M.W. 2007. Advances in fluorescent protein technology. J. Cell. Sci. 120(24): 4247-4260. 117 Sharma, R.R., Singh, D., and Singh, R. 2009. Biological control of postharvest diseases of fruits and vegetables by microbial antagonists: a review. Biol. Control 50(3): 205-221. Shimomura, O., Johnson, F., and Saiga, Y. 1962. Extraction, purification and properties of aequorin, a bioluminescent protein from the luminous hydromedusan, Aequorea. J Cell Comp Physiol 59(3): 223–39. Sholberg, P., and Boulé, J. 2008. Interaction of powdery mildew treatments on apple microflora. In Plant-microbe interactions. Edited by E. Ait Barka and C. Clément. Research Signpost, Kerala, India. pp. 393-406. Sholberg, P., and Stokes, S. 2006. Apple packinghouse sanitation in British Columbia in 2005. Can. Plant Dis. Survey 86: 123-124. Sholberg, P., O’Gorman, D., Bedford, K., and Lévesque, C.A. 2005a. Development of a DNA macroarray for detection and monitoring of economically important apple diseases. Plant Dis. 89(11): 1143-1150. Sholberg, P.L., and Conway, W.S. 2004. Postharvest Pathology. In The Commercial Storage of Fruits, Vegetables, and Florist and Nursery Stocks. Edited by K.C. Gross, C.Y. Wang and M. Saltveit. Available from http://www.ba.ars.usda.gov/hb66/ [accessed 14 March 2010]. Sholberg, P.L., O’Gorman, D., and Bedford, K.E. 2006. Monitoring Erwinia amylovora in pome fruit orchards using reverse-dot-blot hybridization. Acta Hortic. 704: 91- 97. Sholberg, P.L., and Haag, P. D. 1996. Incidence of postharvest pathogens of stored apples in British Columbia. Can. J. Plant Pathol. 18(1): 81-85. 118 Sholberg, P.L., Bedford, K.E., and Stokes, S. 2003. Effect of preharvest application of cyprodinil on postharvest decay of apples caused by Botrytis cinerea. Plant Dis. 87(9): 1067-1071. Sholberg, P.L., Bedford, K.E., and Stokes, S. 2005b. Sensitivity of Penicillium spp. and Botrytis cinerea to pyrimethanil and its control of blue and gray mold of stored apples. Crop Prot. 24(2): 127-134. Sholberg, P.L., Harlton, C., Haag, P., Levesque, C.A., O’Gorman, D., and Siefert, K. 2005c. Benzimidazole and diphenylamine sensitivity and identity of Penicillium spp. that cause postharvest blue mold of apples using beta-tubulin gene sequences. Postharvest Biol. Technol. 36(1): 41-49. Smock, R. 1979. Controlled atmosphere storage of fruits. Hortic. Rev. 1: 301-333. Spadaro, D., and Gullino, M.L. 2004. State of the art and future prospects of the biological control of postharvest fruit diseases. Int. J. Food Microbiol. 91(2): 185- 194. Spotts, R.A., Cervantes, L.A., and Mielke, E.A. 1999. Variability in postharvest decay among apple cultivars. Plant Dis. 83(11): 1051-1054. Spotts, R.A., Serdani, M., Wallis, K.M., Walter, M., Harris-Virgin, T., Spotts, K., Sugar, D., Xiao, C.L., and Qu, A. 2009. At-harvest prediction of grey mould risk in pear fruit in long-term cold storage. Crop Prot. 28(5): 414-420. Stockwell, V.O., and Stack, J.P. 2007. Using Pseudomonas spp. for integrated biological control. Phytopathol. 97(2): 244-249. 119 Stokes, S.C., Sholberg, P.L., Nelson, L.M., and Sergeeva, E. 2006. The screening of bacterial isolates for biological control of blue and gray molds of apples. Can. J. Plant Pathol. 28(1): 163. Suarez, A., Güttler, A., Strätz, M., Staendner, L.H., Timmis, K.N., and Guzmán, C.A. 1997. Green fluorescent protein-based reporter systems for genetic analysis of bacteria including monocopy applications. Gene 196(1-2): 69-74. Tambong, J.T., de Cock, A.W.W.M., Tinker, A., and Lévesque, C.A. 2006. Oligonucleotide array for identification and detection of Pythium species. Appl. Environ. Microbiol. 72(4): 2691-2706. Teixidó, N., Usall, J., and Vinas, I. 1999. Efficacy of preharvest and postharvest Candida sake biocontrol treatments to prevent blue mould on apples during cold storage. Int. J. Food Microbiol. 509(3): 203-210. Teixidó, N., Usall, J., Magan, N., and Vinas, I. 1999. Microbial population dynamics on Golden Delicious apples from bud to harvest and effect of fungicide applications. Ann. Appl. Biol. 134(1): 109-116. Terry, L.A., and Joyce, D.C. 2004. Elicitors of induced disease resistance in postharvest horticultural crops: a brief review. Postharvest Biol. Technol. 32(1): 1-13. Tian, S., Fan, Q., Xu, Y., and Liu, H. 2002. Biocontrol efficacy of antagonist yeasts to gray mold and blue mold on apples and pears in controlled atmospheres. Plant Dis. 86(8): 848-853. Tombolini, R., Unge, A., Davey, M.E., de Bruijn, F.J., and Jansson, J.K. 1997. Flow cytometric and microscopic analysis of GFP-tagged Pseudomonas fluorescens bacteria. FEMS Microbiol. Ecol. 22(1): 17-28. 120 van Loon, L.C. 2007. Plant responses to plant growth-promoting rhizobacteria. Eur. J. Plant. Pathol. 119(3): 243-254. Vikram, A., Prithiviraj, B., Hamzehzarghani, H., and Kushalappa, A.C. 2004. Volatile metabolite profiling to discriminate diseases of McIntosh apple inoculated with fungal pathogens. J. Sci. Food Agric. 84(11): 1333-1340. Walter, M., Boyd-Wilson, K.S.H., Stanley, J., Harris-Virgin, P., Morgan, C., and Pyke, N.B. 1997. Epidemiology of Botrytis cinerea in boysenberry (Rubus spp.). Proc. 50th N.Z. Plant Protection Conf. 50: 93-100. Ward, W.W., Cody, C.W., Hart, R.C., and Cormier, M.J. 1980. Spectrophotometric identity of the energy transfer chromophores in Renilla and Aequorea green- fluorescent proteins. Photochem. Photobiol. Rev. 31: 611-615. Watanabe, M., and H. Shimizu. 2005. Detection of patulin in apple juices marketed in the Tohoku district, Japan. J. Food Prot. 68: 610-612. Watanabe, S., Kumakura, K., Izawa, N., Nagayama, K., Mitachi, T., Kanamori, M., Teraoka, T., and Arie, T. 2007. Mode of action of Trichoderma asperellum SKT-1, a biocontrol agent against Gibberella fujikuroi. J. Pestic. Sci. 32(3): 222-228. Watkins, C.B. 2006. The use of 1-methylcyclopropene (1-MCP) on fruits and vegetables. Biotechnol. Adv. 24(4): 389-409. Watkins, C.B. 2008. Overview of 1-methylcyclopropene trials and uses for edible horticultural crops. HortSci. 43(1): 86-94. White, T. J., Burns, T.D., Lee, S.B., and Taylor, J.W. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics,. In PCR protocols: a 121 guide to methods and applications. Edited by M. A. Innis, D. H. Helfand, J. J. Sninsky, and T. J. White. Academic Press, Inc., San Diego, CA. pp. 315-322. Wilson, C.L., and Wisniewski, M.E. 1989. Biological control of postharvest diseases of fruits and vegetables: an emerging technology. Annu. Rev. Phytopathol. 27:425- 411. Yahia, E.M. 2009. Introduction. In Modified and controlled atmospheres for the storage, transportation, and packaging of horticultural commodities. Edited by E.M. Yahia. CRC Press, Boca Raton, FL. pp. 1-16. Zhou, T., Chu, C.L., Liu, W.T., and Schaneider, E., 2001. Postharvest control of blue mold and grey mold on apples using isolates of Pseudomonas syringae. Can. J. Plant Pathol. 23(3): 246-252. 122 APPENDIX Appendix A Table A1. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 2628.336 4.585 0.0619 B. cinerea 2 384.596 1.857 0.2357 M. piriformis 2 573.896 0.429 0.6697 Table A2. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 327.860 0.134 0.8766 B. cinerea 2 66.667 1.0 0.4053 M. piriformis 2 5625.402 28.266 0.0001*** Table A3. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 741.125 1.119 0.3681 B. cinerea 2 55.512 0.567 0.5863 M. piriformis 2 683.272 1.018 0.3994 123 Table A4. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 75.335 0.034 0.9668 B. cinerea 2 488.562 0.579 0.5800 M. piriformis 2 203.247 0.438 0.6583 Table A5. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 1702.334 4.590 0.0533 B. cinerea 2 148.328 12.683 0.0047** M. piriformis 2 1336.256 4.316 0.0601 Table A6. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 182.052 0.171 0.8461 B. cinerea 2 76.432 0.262 0.7753 M. piriformis 2 503.152 1.681 0.2398 124 Table A7. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in all growing season sections for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Detected pathogens (P. expansum, B. cinerea or M. piriformis) Category DF Type III SS F-value p-value 2007 growing season 2 896.063 4.116 0.0538 2008 growing season 2 1022.252 3.222 0.0880 2007/08 growing season 2 454.769 19.248 0.0194* Table A8. Key data extracted from the two-way ANOVA table for infection severity data collected from Jonagold apples grown in Field 3 and analyzed for natural disease incidence after harvest in 2007. DF Type III SS F-value p-value Main Effects Incubation Period 1 3639.591 13.081 0.0004*** 1-MCP+/- 1 3260.433 11.718 0.0008*** Interaction Incubation period x MCP+/- 1 1008.275 3.624 0.0584 Table A9. Key data extracted from the two-way ANOVA table for infection severity data collected from Gala apples grown in Field 12 analyzed for natural disease incidence after harvest in 2007. DF Type III SS F-value p-value Main Effects Incubation Period 1 13.767 36.073 0.0000*** 1-MCP+/- 1 0.175 0.458 0.4995 Interaction Incubation period x MCP+/- 1 0.286 0.750 0.3876 125 Table A10. The effect of incubation period (1, 2, 4 and 6 months) on natural disease incidence and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3, Gala apples grown in Field 12 and Kiran and Red Delicious apples grown in Reekie and analyzed for natural disease incidence from harvest after 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value Field 3 3 61.3201 74.6611 0.0000*** Field 12 3 79.6754 276.1476 0.0000*** Kiran 3 0.8366 3.2372 0.0268* Reekie 3 11.1988 17.9952 0.0000*** Table A11. Key data extracted from the one-way ANOVA table constructed from infection severity data collected from Fields 3, 12, Kiran and Reekie analyzed for natural disease incidence for apples stored in CA storage for 4.5 months from harvest 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Category DF Type III SS F-value p-value 1 month storage 3 0.9484 1.5636 0.1195 2 months storage 3 10.0842 7.1456 0.0001*** 4 months storage 3 8863.9197 27.1892 0.0000*** 6 months storage 3 40909.7401 48.0806 0.0000*** Table A12. Key data extracted from the one-way ANOVA table constructed from infection severity data collected from Fields 3, 12, Kiran and Reekie analyzed for natural disease incidence for apples stored in CA storage for 4.5 months from harvest 2008. Category DF Type III SS F-value p-value Field 3 29.2208 55.0013 0.0000*** 126 Table A13. Key data extracted from the two-way ANOVA table for infection severity data collected from Jonagold apples grown in Field 3 analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2007.. Table A14. Key data extracted from the two-way ANOVA table for infection severity data collected from Gala apples grown in Field 12 and analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2007.. Table A15. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for natural disease incidence in combination with bacterial isolates from harvest 2007. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 2310.748 18.214 0.0005*** 1100-6 1 1674.926 11.450 0.0033** 1-112 1 7.358 0.243 0.6279 2-28 1 1542.633 26.057 0.0001*** 4-6 1 0.020 1.49e-4 0.9904 6-25 1 130.302 0.966 0.3387 DF Type III SS F-value p-value Main Effects Biological Control 1 0.900 1.192 0.3182 1-MCP+/- 1 0.342 2.262 0.1355 Interaction Biological Control x MCP+/- 5 0.332 0.439 0.8202 DF Type III SS F-value p-value Main Effects Biological Control 1 3039.868 5.773 0.0001*** 1-MCP+/- 1 3435.342 32.620 0.0000*** Interaction Biological Control x MCP+/- 5 2779.181 0.439 0.0002*** 127 Table A16. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are constructed from IS results collected from Field 12 Gala apples and analyzed for natural disease incidence in combination with bacterial isolates from harvest 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 624.112 1.329 0.2659 1-MCP- 5 5194.937 8.902 0.0000*** Table A17. Key data extracted from the two-way ANOVA table from infection severity data collected from Jonagold apples grown in Field 3 and inoculated with P. expansum after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 5 32.862 61.071 0.0000*** 1-MCP+/- 1 0.015 0.413 0.7077 Interaction Biological Control x MCP+/- 5 5.626 1.125 0.0000*** Table A18. The effect of 1-MCP in six separate biological control categories and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with P. expansum and bacterial isolates after harvest in 2007. Individual one- way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 0.0420 1.775 0.1994 1100-6 1 1.4724 7.422 0.0139* 1-112 1 0.0425 0.215 0.6482 2-28 1 0.0872 2.057 0.1687 4-6 1 0.0329 0.302 0.5893 6-25 1 3.9641 52.958 0.0000*** 128 Table A19. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are constructed from infection severity results collected from Jonagold grown in Field 3 analyzed for disease incidence when inoculated with P. expansum and bacterial isolates from harvest in 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 16.7121 31.798 0.0000*** 1-MCP- 5 21.7763 39.548 0.0000*** Table A20. Key data extracted from the two-way ANOVA table from infection severity data collected from Jonagold apples grown in Field 3 and inoculated with B. cinerea after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 5 21.222 20.760 0.0000*** 1-MCP+/- 1 15.998 78.249 0.0000*** Interaction Biological Control x MCP+/- 5 13.841 13.534 0.0000*** Table A21. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 20077. Individual one- way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 9.706 45.191 0.0000*** 1100-6 1 0.3931 1.793 0.1973 1-112 1 0.2607 1.060 0.3169 2-28 1 13.3139 80.502 0.0000*** 4-6 1 5.9336 29.726 0.0000*** 6-25 1 0.2310 1.271 0.2743 129 Table A22. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 33.7372 53.215 0.0000 1-MCP- 5 1.3253 0.940 0.4630 Table A23. Key data extracted from the two-way ANOVA table constructed from infection severity data collected from Jonogold apples grown in Field 3 and inoculated with M. piriformis after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 5 14.346 10.660 0.0000*** 1-MCP+/- 1 0.784 2.913 0.0908 Interaction Biological Control x MCP+/- 5 7.164 5.323 0.0002*** Table A24. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data is are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 0.0020 0.0181 0.8943 1100-6 1 0.6318 1.3963 0.2527 1-112 1 0.0194 0.0729 0.7902 2-28 1 0.6870 4.3055 0.0526 4-6 1 6.5402 13.1253 0.0019** 6-25 1 0.0677 0.5298 0.4761 130 Table A25. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 1.1507 0.9683 0.4455 1-MCP- 5 20.3591 13.5439 0.0000 Table A26. Key data extracted from the two-way ANOVA table constructed from infection severity data collected from Gala apples from Field 12 and inoculated with P. expansum after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 5 192.271 1.940 0.0936 1-MCP+/- 1 182.887 9.226 0.0030** Interaction Biological Control x MCP+/- 5 164.951 1.664 0.1494 Table A27. Key data extracted from the two-way ANOVA table from infection severity data collected from Gala apples grown in Field 12 and inoculated with M. piriformis after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 4 3.015 1.729 0.1505 1-MCP+/- 1 6.578 1.508e-5 0.9969 Interaction Biological Control x MCP+/- 4 8.469 4.856 0.0014*** 131 Table A28. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. Individual one- way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 0.0242 0.0487 0.8278 1-112 1 1.9035 4.7987 0.0419* 2-28 1 6.0306 14.5425 0.0013** 4-6 1 0.4423 1.1814 0.2914 6-25 1 0.0683 0.1370 0.7156 Table A29. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 7.0326 3.9878 0.0075** 1-MCP- 5 4.4516 2.5814 0.0498* Table A30. Key data extracted from the two-way ANOVA table from infection severity data collected from Jonagold apples grown in Field 3 and analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2008. DF Type III SS F-value p-value Main Effects Biological Control 5 1.2214 0.8840 0.4926 Incubation Period 3 105.9743 127.8366 0.0000*** Interaction Biological Control x Incubation Period 15 14.3647 3.4656 0.0000*** 132 Table A31. Key data extracted from the two-way ANOVA table from infection severity data collected from Gala apples grown in Field 12 and analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2008. DF Type III SS F-value p-value Main Effects Biological Control 5 0.5763 0.8804 0.4950 Incubation Period 3 149.6138 380.9182 0.0000*** Interaction Biological Control x Incubation Period 15 3.6138 1.8400 0.0309* Table A32. The effect of the incubation period on biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for natural disease incidence when inoculated the bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value 2 months 5 5.5753 4.3605 0.0021** 4 months 5 5067.4972 3.9609 0.0039** 6 months 5 15646.1017 5.2559 0.0005*** Table A33. The effect of the incubation period on biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for natural disease incidence when inoculated the bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value 2 months 5 2.1850 2.8870 0.0221* 4 months 5 1876.4999 2.5715 0.0370* 6 months 5 743.3109 0.7762 0.5782 133 Table A34. Key data extracted from the two-way ANOVA table from infection severity data collected from Jonagold (Field 3), Gala (Fields 12 and Kiran) and Red Delicious (Reekie) apples after 4.5 months in CA storage and analyzed for natural disease incidence following inoculation with bacterial isolates after harvest in 2008. DF Type III SS F-value p-value Main Effects Biological Control 5 3.3247 5.4749 0.0001*** Field 3 12.3636 33.9323 0.0000*** Interaction Biological Control x Field 15 5.8261 3.1980 0.0000*** Table A35. The effect of the biological control treatments on infection severity and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3, Gala apples grown in Field 12 and Kiran and Red Delicious apples grown in Reekie after 4.5 months in CA stroage and analyzed for natural disease incidence when inoculated the bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value Field 3 5 7.7543 4.8907 0.0009*** Field 12 5 0.1933 1.1827 0.3298 Kiran 5 0.1136 2.2402 0.0633 Reekie 5 1.0897 1.7312 0.1432 Table A36. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with P. expansum and bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value Field 3, Jonagold 5 3041.4892 6.8050 0.0001*** Field 12, Gala 5 401.2654 3.1239 0.0151* 134 Table A37. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value Field 3, Jonagold 5 5981.4974 4.9960 0.0008*** Field 12, Gala 5 1962.7067 6.4391 0.0001*** Table A38. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value Field 3, Jonagold 5 2115.5372 4.7271 0.0012** Field 12, Gala 5 3.0721 4.1326 0.0030** Table A39. Growth rate (µ) comparisons of P. fluorescens isolates 4-6-wild-type and 4- 6-gfp and the resulting key data extracted from the one-way ANOVA table. Main Effect Bacterial Isolate 4-6-wt vs 4-6-gfp Category DF Type III SS F-value p-value Growth Rate (µ) 1 0.0022 0.6492 0.4656 135 Table A40. Key data extracted from the two-way ANOVA table constructed from cell densities (CFU/ml) of P. fluorescens isolates 4-6-wild-type and 4-6-gfp stored for 55 d at 1°C in air. DF Type III SS F-value p-value Main Effects Bacterial strain 1 50.5367 670.6401 0.0000*** Time 5 72.3075 191.9023 0.0000*** Interaction Bacterial strain x Time 5 72.7250 171.9775 0.0000*** Table A41. Cell density comparisons determined by dilution plating method of P. fluorescens isolates 4-6-gfp and 4-6-gfp + P. expansum stored for 55 d at 1°C in air and the resulting key data extracted from the one-way ANOVA table. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Bacterial Isolate 4-6-gfp vs 4-6-gfp + P. expansum Category DF Type III SS F-value p-value 0 d storage 1 0.1061 2.3329 0.1610 10 d storage 1 8.4137e-4 0.0157 0.9024 20 d storage 1 0.5121 4.0261 0.0758 30 d storage 1 2.3762 36.5429 0.0005*** 45 d storage 1 16.6408 120.1450 0.0000*** 55 d storage 1 4.418e13 8.8646 0.0206* Table A42. Infection severity comparisons of P. fluorescens isolates 4-6-wild-type and 4-6-gfp stored for 55 d at 1°C in air and the resulting key data extracted from the one-way ANOVA table. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Bacterial Isolate 4-6-gfp vs 4-6-gfp + P. expansum Category DF Type III SS F-value p-value 30 d storage 1 21.6680 7.6629 0.0040** 40 d storage 1 46.8838 14.1739 0.0093** 45 d storage 1 18.9113 1.3841 0.2840 55 d storage 1 13.9568 0.5615 0.4820 DETECTION, QUANTIFICATION AND BIOLOGICAL CONTROL OF POST- HARVEST PATHOGENS OF POME FRUIT by Daylin Lindsay Mantyka BSc, The University of Victoria, 2006 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The College of Graduate Studies (Biology) THE UNIVERSITY OF BRITISH COLUMBIA (Okanagan) April 2010 © Daylin Lindsay Mantyka, 2010 ii ABSTRACT Blue mold (Penicillium expansum), grey mold (Botrytis cinerea) and mucor rot (Mucor piriformis) are important post-harvest diseases of pome fruit in British Columbia and throughout the world causing annual losses of 5 – 20%. Identification and quantification using novel DNA macroarray technology may assist in the development of prediction models and disease forecasting. Post-harvest pathogens were monitored and quantified throughout the growing season in four apple orchards in the Okanagan Valley, BC in 2007 and 2008. Their detection was variable due to field and year differences. The average percent detection of P. expansum (27.4%) and M. piriformis (19.2%) was higher than that of B. cinerea (6.2%). There was a positive correlation between total post-harvest pathogen detection in aerial samples just prior to harvest and after harvest in naturally infected fruit (r = 0.74; p = 0.09). Pseudomonas fluorescens (isolates 1-112, 2- 28, 4-6) and Serratia plymuthica (isolate 6-25), isolated from the rhizosphere of legumes, were investigated for their biological control capabilities in semi-commercial storage conditions at 1°C in air and commercial storage conditions in controlled atmosphere and with 1-methylcyclopropene (1-MCP) application. Overall, isolate 6-25 provided control in the greatest number of treatments (51.7%) while isolate 1-112 provided the greatest level of control (75.8%) in treatments where significant control was exhibited. Isolate 4- 6 was tagged with green fluorescent protein to gain insight into bacterial antagonist population and survival dynamics. Alone, its population increased 10 fold after 30 d in storage at 1°C and then decreased to concentrations similar to those at inoculation. In the presence of the pathogen, 4-6-gfp increased then decreased after 30 d in storage at 1°C to undetectable amounts. These data provide greater insight into the prediction, control and iii population dynamics of pathogens and biological control agents as a means of preventing and controlling post-harvest storage diseases in pome fruit. iv TABLE OF CONTENTS ABSTRACT ....................................................................................................................... ii TABLE OF CONTENTS ................................................................................................. iv LIST OF FIGURES .......................................................................................................... vi LIST OF TABLES ......................................................................................................... viii LIST OF ABBREVIATIONS ....................................................................................... xiii ACKNOWLEDGMENTS .............................................................................................. xiv CHAPTER 1: INTRODUCTION ..................................................................................... 1 1.1 Post-harvest pathogens ........................................................................................ 2 1.1.1 Penicillium expansum .................................................................... 3 1.1.2 Botrytis cinerea .............................................................................. 4 1.1.3 Mucor piriformis ............................................................................ 5 1.2 Post-harvest factors that influence post-harvest pathology ................................. 6 1.2.1 Temperature ................................................................................... 7 1.2.2 Controlled atmosphere ................................................................... 7 1.2.3 1-Methylcyclopropene ................................................................... 8 1.3 Methods of detection and identification .............................................................. 8 1.4 Methods of control ............................................................................................ 10 1.4.1 Chemical control .......................................................................... 10 1.4.2 Biological control ......................................................................... 10 1.5 Mechanisms of action ........................................................................................ 13 1.5.1 Antibiosis ..................................................................................... 13 1.5.2 Competition for nutrients ............................................................. 14 1.5.3 Parasitism ..................................................................................... 15 1.5.4 Induced systemic resistance ......................................................... 15 1.6 Green fluorescent protein .................................................................................. 16 1.7 Objectives .......................................................................................................... 18 CHAPTER 2: MATERIALS AND METHODS ........................................................... 20 2.1 Antagonists ........................................................................................................ 20 2.2 Pathogens .......................................................................................................... 20 2.3 Orchard field sites ............................................................................................. 21 2.4 pAG408 ............................................................................................................. 23 2.5 Construction of the gfp-transconjugant ............................................................. 23 2.6 Plasmid stability ................................................................................................ 25 2.7 P. fluorescens 4-6-gfp fitness analysis .............................................................. 25 2.8 Spore, apple tissue collection and fruit washings.............................................. 26 2.9 Validation and field testing of DNA macroarrays............................................. 27 2.9.1 DNA extraction ............................................................................ 27 v 2.9.2 PCR amplification ........................................................................ 28 2.9.3 DNA macroarray construction ..................................................... 29 2.9.4 Hybridization and chemiluminescent detection ........................... 29 2.10 Biological control of post-harvest fungal pathogens ......................................... 33 2.10.1 Inoculation of apples by drenching .............................................. 33 2.10.2 Natural disease incidence ............................................................. 33 2.10.3 Natural disease incidence with bacterial antagonists ................... 34 2.10.4 Biological control of post-harvest pathogens ............................... 35 2.10.5 Bacterial survival on apple ........................................................... 35 2.11 Analysis ............................................................................................................. 36 CHAPTER 3: RESULTS ................................................................................................ 38 3.1 Validation and field testing of DNA macroarrays............................................. 38 3.1.1 Quantification of post-harvest pathogens ..................................... 38 3.1.2 Field application of macroarray technology................................. 43 3.2 Natural disease incidence .................................................................................. 53 3.2.1 Harvest 2007 ................................................................................ 53 3.2.2 Harvest 2008 ................................................................................ 54 3.3 Growing season detection and natural disease incidence .................................. 57 3.4 Antagonist efficacy, harvest 2007 ..................................................................... 58 3.4.1 Natural disease incidence and biocontrol inoculations ................ 58 3.4.2 Post-harvest pathogen and biocontrol inoculations ...................... 58 3.5 Antagonist efficacy, harvest 2008 ..................................................................... 62 3.5.1 Natural disease incidence and biocontrol inoculations ................ 64 3.5.2 Post-harvest pathogen and biocontrol inoculations ...................... 67 3.6 Bacterial survival on apple using GFP .............................................................. 75 3.6.1 Plasmid confirmation ................................................................... 75 3.6.2 Confirmation of gfp transformation ............................................. 75 3.6.3 Plasmid stability ........................................................................... 75 3.6.4 P. fluorescens 4-6-gfp fitness analysis ......................................... 80 3.6.5 Bacterial survival on apple ........................................................... 80 CHAPTER 4: DISCUSSION .......................................................................................... 87 4.1 DNA macroarray for post-harvest apple pathogens .......................................... 87 4.1.1 DNA macroarray quantification ................................................... 87 4.1.2 Field monitoring of apple post-harvest pathogens ....................... 89 4.2 Biological control of post-harvest pathogens .................................................... 93 4.2.1 Bacterial antagonists and CA storage........................................... 95 4.2.2 Bacterial antagonists and 1-MCP ................................................. 98 4.3 Bacterial survival on apple .............................................................................. 100 CHAPTER 5: CONCLUSIONS ................................................................................... 103 REFERENCES ............................................................................................................... 105 APPENDIX ..................................................................................................................... 122 vi LIST OF FIGURES Figure 1. Average daily temperatures and total daily precipitation at Kelowna, BC and Summerland, BC during the growing seasons of 2007 and 2008 ..................................... 22 Figure 2. DNA macroarray template ............................................................................... 31 Figure 3. Quantitative assessment of Penicillium expansum probe PE-H3u illustrating the relationship between greyscale value and DNA concentration ........................................ 39 Figure 4. Quantitative assessment of Botrytis cinerea probe BC-H2d illustrating the relationship between greyscale value and DNA concentration ........................................ 40 Figure 5. Quantitative assessment of Mucor piriformis probe Mpir-ITS-414H1 illustrating the relationship between greyscale value and DNA concentration ................ 41 Figure 6. Concentrations of P. expansum, B. cinerea and M. piriformis from aerial and plant tissue samples collected from Field 3 in 2007 and 2008 using DNA macroarrays . 44 Figure 7. Concentrations of P. expansum, B. cinerea and M. piriformis from aerial and plant tissue samples collected from Field 12 in 2007 and 2008 using DNA macroarrays 45 Figure 8. Concentrations of P. expansum, B. cinerea and M. piriformis from aerial and plant tissue samples collected from the Kiran orchard in 2007 and 2008 using DNA macroarrays ....................................................................................................................... 46 Figure 9. Concentrations of P. expansum, B. cinerea and M. piriformis from aerial and plant tissue samples collected from the Reekie orchard in 2007 and 2008 using DNA macroarrays ....................................................................................................................... 47 Figure 10. Natural disease incidence in Jonagold apples grown in Field 3, Gala apples grown in Field 12 and the Kiran Orchard and Red Delicious apples grown in the Reekie orchard and stored for 4.5 months in CA .......................................................................... 56 Figure 11. Effect of 1-MCP and biological control isolates on natural disease incidence in Gala apples grown in Field 12 (harvest 2007) and stored for 3 months in air at 1°C .. 59 Figure 12. Effect of 1-MCP and biological control isolates on infection severity in Jonagold apples from Field 3 (harvest 2007), inoculated with P. expansum, B. cinerea and M. piriformis and stored for 3 months at 1°C in air ................................................... 61 Figure 13. Effect of 1-MCP and biological control isolates on infection severity in Gala apples from Field 12 (harvest 2007), inoculated with M. piriformis and stored for 3 months at 1°C in air. ......................................................................................................... 63 Figure 14. Average infection severity for all wounded and non-wounded apples from individual biological control treatments conducted in 2008 ............................................. 65 vii Figure 15. Effect of biological control isolates on natural disease incidence in Jonagold apples grown in Field 3 and Gala apples grown in Field 12 (harvest 2008) and stored for 2, 4 or 6 months at 1°C in air ............................................................................................ 66 Figure 16. Effect of biological control isolates on natural disease incidence in Jonagold apples grown in Field 3 (harvest 2008) and stored for 4.5 months in CA ........................ 68 Figure 17. Effect of biological control isolates on natural disease incidence in Gala apples grown in the Kiran orchard (harvest 2008) and stored for 4.5 months in CA ....... 69 Figure 18. Effect of biological control isolates on infection severity in Jonagold apples grown in Field 3 (harvest 2008), inoculated with P. expansum and B. cinerea and stored for 2 months at 1°C in air and inoculated with M. piriformis and stored for 1 month at 1°C in air ........................................................................................................................... 70 Figure 19. Effect of biological control isolates on infection severity in Gala apples grown in Field 12 (harvest 2008), inoculated with P. expansum and B. cinerea and stored for 2 months at 1°C in air and inoculated with M. piriformis and stored for 1 month at 1°C in air ...................................................................................................................................... 71 Figure 20. HindIII restriction digest of gfp-containing plasmid, pAG408 ...................... 76 Figure 21. PCR amplification of gfp in P. fluorescens isolate 4-6 transformed with pAG408 using gfp-specific primers .................................................................................. 77 Figure 22. Fluorescence microscopy of P. fluorescens isolate 4-6 transformed with pAG408 exhibiting green fluorescence ............................................................................ 78 Figure 23. Fluorescence microscopy of an apple wound inoculated with P. fluorescens isolate 4-6-gfp ................................................................................................................... 79 Figure 24. Growth characteristics of P. fluorescens 4-6-wild-type and P. fluorescens 4-6- gfp in half-strength TSB at 28°C for 24 h ......................................................................... 81 Figure 25. Relationship between P. fluorescens isolate 4-6-gfp cell density and relative fluorescent units as determined by dilution plating and direct scanning. ......................... 82 Figure 26. Bacterial survival of 4-6-gfp, P. expansum and P. expansum + 4-6-gfp on Gala apples stored for 55 d at 1°C in air using the direct scanning method ..................... 83 Figure 27. Bacterial survival of 4-6-gfp and 4-6-gfp + P. expansum on Gala apples stored for 55 d at 1°C in air using the dilution plating method ................................................... 86 viii LIST OF TABLES Table 1. Frequency of aerial and plant tissue sampling for Fields 3, 12 and the Kiran and Reekie orchards in the early, middle and late growing seasons of 2007 and 2008. ......... 27 Table 2. DNA macroarray probe sequences..................................................................... 30 Table 3. DNA macroarray detection summary of P. expansum, B. cinerea and M. piriformis from four apple orchards throughout the 2007 and 2008 growing seasons ..... 50 Table 4. Detection of P. expansum, B. cinerea and M. piriformis from fruit washings collected from Jonagold apples from Field 3, Gala apples from Field 12 and the Kiran orchard and Red Delicious apples from the Reekie orchard in 2007 and 2008 using DNA macroarrays ....................................................................................................................... 53 Table 5. Natural disease incidence indicated by infection severity in Jonagold apples grown in Field 3 (harvest 2007), treated or not treated with 1-MCP and stored for 3 or 6 months at 1°C in air .......................................................................................................... 54 Table 6. Natural disease incidence indicated by infection severity in Gala apples grown in Field 12 (harvest 2007), treated or not treated with 1-MCP and stored for 3 or 6 months at 1°C in air .......................................................................................................... 54 Table 7. Natural disease incidence indicated by infection severity in Jonagold apples grown in Field 3, Gala apples grown in Field 12 and the Kiran orchard and Red Delicious apples grown in the Reekie orchard (harvest 2008) and stored for 1, 2, 4 or 6 months at 1°C in air ........................................................................................................................... 55 Table 8. Summary of antagonist efficacy experiments for harvest 2007 and 2008 ......... 73 Table 9. Sole carbon source utilization by P. fluorescens isolates 4-6-wild-type and 4-6- gfp as indicated by Biolog Phenotye MicroArrays™ for sugars and organic acids commonly found in apple juice ........................................................................................ 80 Table 10. Effect of biological control isolate 4-6-gfp on infection severity in Gala apples inoculated with and without P. expansum and stored for 55 days at 1°C in air. .............. 85 Table A1. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. ............................................................................................. 122 Table A2. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. ............................................................................................. 122 ix Table A3. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. ....................................................................................... 122 Table A4. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. ............................................................................................. 123 Table A5. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. ............................................................................................. 123 Table A6. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. ....................................................................................... 123 Table A7. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in all growing season sections for Fields 3, 12 and the Kiran and Reekie orchards. .............................................................................................................. 124 Table A8. Key data extracted from the two-way ANOVA table for IS data collected from Jonagold apples grown in Field 3 and analyzed for NDI after harvest in 2007 ............. 124 Table A9. Key data extracted from the two-way ANOVA table for IS data collected from Gala apples grown in Field 12 analyzed for NDI after harvest in 2007 ......................... 124 Table A10. The effect of incubation period on NDI and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3, Gala apples grown in Field 12 and Kiran and Red Delicious apples grown in Reekie and analyzed for NDI from harvest after 2008. ....................... 125 Table A11. Key data extracted from the one-way ANOVA table constructed from IS data collected from Fields 3, 12, Kiran and Reekie analyzed for NDI for apples stored in CA storage for 4.5 months from harvest 2008 ...................................................................... 125 Table A12. Key data extracted from the one-way ANOVA table constructed from IS data collected from Fields 3, 12, Kiran and Reekie analyzed for NDI for apples stored in CA storage for 4.5 months from harvest 2008 ...................................................................... 125 Table A13. Key data extracted from the two-way ANOVA table for IS data collected from Jonagold apples grown in Field 3 analyzed for NDI in combination with bacterial isolates after harvest in 2007 .......................................................................................... 126 x Table A14. Key data extracted from the two-way ANOVA table for IS data collected from Gala apples grown in Field 12 and analyzed for NDI in combination with bacterial isolates after harvest in 2007 .......................................................................................... 126 Table A15. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for NDI in combination with bacterial isolates from harvest 2007 ............................................................................... 126 Table A16. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are constructed from IS results collected from Field 12 Gala apples and analyzed for NDI in combination with bacterial isolates from harvest 2007 ...................................... 127 Table A17. Key data extracted from the two-way ANOVA table from IS data collected from Jonagold apples grown in Field 3 and inoculated with P. expansum after harvest in 2007 ................................................................................................................................ 127 Table A18. The effect of 1-MCP in six separate biological control categories and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with P. expansum and bacterial isolates after harvest in 2007 ...................... 127 Table A19. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are constructed from IS results collected from Jonagold grown in Field 3 analyzed for disease incidence when inoculated with P. expansum and bacterial isolates from harvest in 2007 ................................................................................................................ 128 Table A20. Key data extracted from the two-way ANOVA table from IS data collected from Jonagold apples grown in Field 3 and inoculated with B. cinerea after harvest in 2007 ................................................................................................................................ 128 Table A21. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2007 .......................... 128 Table A22. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2007. ........................................................................................................................... 129 Table A23. Key data extracted from the two-way ANOVA table constructed from IS data collected from Jonogold apples grown in Field 3 and inoculated with M. piriformis after harvest in 2007 ....................................................................................................... 129 xi Table A24. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data is are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007 ........................................................................................................................................ 129 Table A25. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. ............................................................................................................... 130 Table A26. Key data extracted from the two-way ANOVA table constructed from IS data collected from Gala apples from Field 12 and inoculated with P. expansum after harvest in 2007 ................................................................................................................ 130 Table A27. Key data extracted from the two-way ANOVA table from IS data collected from Gala apples grown in Field 12 and inoculated with M. piriformis after harvest in 2007. ............................................................................................................................... 130 Table A28. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007 ..................... 131 Table A29. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. ........................................................................................................................... 131 Table A30. Key data extracted from the two-way ANOVA table from IS data collected from Jonagold apples grown in Field 3 and analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2008 ............................................... 131 Table A31. Key data extracted from the two-way ANOVA table from IS data collected from Gala apples grown in Field 12 and analyzed for NDI in combination with bacterial isolates after harvest in 2008 .......................................................................................... 132 Table A32. The effect of the incubation period on biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for NDI when inoculated the bacterial isolates after harvest in 2008 ...................................................................... 132 Table A33. The effect of the incubation period on biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for NDI when inoculated the bacterial isolates after harvest in 2008 ............................................................................ 132 xii Table A34. Key data extracted from the two-way ANOVA table from IS data collected from Fields 3 and 12 and the Kiran and Reekie orchards after 4.5 months in CA storage and analyzed for natural disease incidence following inoculation with bacterial isolates after harvest in 2008 ....................................................................................................... 133 Table A35. The effect of the biological control treatments on IS and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3, Gala apples grown in Field 12 and Kiran and Red Delicious apples grown in Reekie after 4.5 months in CA stroage and analyzed for NDI when inoculated the bacterial isolates after harvest in 2008 .......................................... 133 Table A36. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with P. expansum and bacterial isolates after harvest in 2008 ............................................................................................................................ 133 Table A37. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2008 ................................................................................................................................ 134 Table A38. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2008 ............................................................................................................................ 134 Table A39. Growth rate comparisons of P. fluorescens isolates 4-6-wild-type and 4-6-gfp and the resulting key data extracted from the one-way ANOVA table .......................... 134 Table A40. Key data extracted from the two-way ANOVA table constructed from cell densities of P. fluorescens isolates 4-6-wild-type and 4-6-gfp stored for 55 d at 1°C in air ........................................................................................................................................ 135 Table A41. Cell density comparisons determined by dilution plating method of P. fluorescens isolates 4-6-gfp and 4-6-gfp + P. expansum stored for 55 d at 1°C in air and the resulting key data extracted from the one-way ANOVA table ................................. 135 Table A42. Infection severity comparisons of P. fluorescens isolates 4-6-wild-type and 4-6-gfp stored for 55 d at 1°C in air and the resulting key data extracted from the one-way ANOVA table ................................................................................................................. 135 xiii LIST OF ABBREVIATIONS * (one asterisk) – pvalue = 0.05 – 0.01 ** (two asterisks) – pvalue = 0.01 – 0.001 *** (three asterisks) – pvalue = < 0.001 ½ TSA – half-strength tryptic soy agar ½ TSB – half-strength tryptic soy broth 1-MCP – 1-methylcyclopropene 4-6-gfp – transformed P. fluorescens isolate 4-6 with GFP-containing plasmid pAG408 bc – biocontrol BC – British Columbia BCA – biological control agent Bcin – Botrytis cinerea βtub – β tubulin CA – controlled atmosphere CFU – colony forming units CSLM – confocal scanning laser microscopy gen - gentamicin gfp – green fluorescent protein (gene) GFP – green fluorescent protein (protein) IS – infection severity ISR – induced systemic resistance kan - kanamycin log10 – logarithmic to the base 10 LSD – least significant difference MOA – mode of action Mpir – Mucor piriformis NDI – natural disease incidence OD600 – optical density at 600 nm PARC –Pacific Agri-Food Research Centre PBS – phosphate buffered saline PCR – polymerase chain reaction PDA – potato dextrose agar Pex – Penicillium expansum pt – plant tissue rDNA – ribosomal deoxyribonucleic acid RFU – relative fluorescent units TBZ – thiabendazole TSA/B – trypic soy agar/broth xiv ACKNOWLEDGMENTS I would like to firstly thank my primary supervisor, Dr. Louise Nelson for giving me this wonderful opportunity. The applied nature of this program helped me realize where I belong within the scientific community. I would also like to thank my co- supervisor, Dr. Peter Sholberg, for sharing your knowledge and expertise in the area of post-harvest research and the use of the facilities at the Pacific Agri-Food Research Centre (PARC). To the members of my committee, Drs. Dan Durall and Mary Forrest, I appreciate your feedback, patience and encouragement. To my uncle, John Virgl, thank- you for lending me your statistical know-how. I think that I would still be lost without your advice. This project would not have been possible without the generosity of the Advanced Foods and Materials Network STAR/Commercialization program. In addition, I am grateful to the BC Innovation Council for awarding me the BC Innovation Scholars Scholarship. I look forward to collaborating with my business partner, Jenn Schofield, and learning the business of science. I would like to send a mountain of gratitude to Dr. Danielle Hirkala of the Okanagan Tree Fruit Cooperative for access to her bounty of knowledge and for all her help in the lab and in all areas of this project. She was a huge part of my success and I have learned a great deal from her. Thank you for your sense of humour and expertise. To the rest of the Nelson lab- Melissa, Natashia, DeAnna and Chris, your assistance has been greatly appreciated. Thank you to Michael Wies of PARC for teaching me the ins and outs of the confocal microscope and to Dr. Andis Klegeris and his lab group for access to the microplate reader. To fellow graduate students, thank you for providing me xv with tidbits of science and sanity along the way. Finally, I would like to thank my family, especially my parents, Bob and Janet, for their unending support and pushing me to ‘split the apple’. To Barry, my partner in crime, your enquiring mind, patience, support and love has got me through it all. Thanks. 1 CHAPTER 1: INTRODUCTION According to the Food and Agriculture Organization of the United Nations, the world apple production in 2008 was greater than 69 million tonnes (FAOSTAT 2009). Apples (Malus domestica) are ranked fourth (by volume) in the world’s production of fruit crops (BCMAFF 2004). In 2002, the British Columbia apple industry produced 136 000 tonnes or approximately 299 846 000 lbs of apples. Apples are produced in significantly higher volumes than any other tree fruit crop in BC (BCMAFF 2004). One major challenge to apple production is damage incurred by post-harvest diseases; annual losses can vary from 5 – 20% and up to 50% in developing countries (Janisiewicz and Korsten 2002). A variety of fungal pathogens is responsible for post-harvest disease; in BC, three such pathogens include Penicillium expansum (blue mold), Botrytis cinerea (grey mold) and Mucor piriformis (mucor rot) (Sholberg and Haag 1996). A common method of disease prevention is through the use of fungicides. In Canada, two synthetic controls are registered for post-harvest use, Mertect® (thiabendazole) and Scholar® (fludioxonil) (PMRA 2007). However, due to growing health issues, the development of pathogen resistance and the demand for environmentally friendly sustainable practices, pesticides are now a major cause for concern (Errampalli et al. 2006; Michaildes and Spotts 1990; Sholberg and Haag 1996). To reduce post-harvest rot, improved methods for pathogen identification prior to harvest and alternate technologies to chemical control are required to assist in disease prediction and control. Preliminary work has resulted in two novel approaches that address these demands: 1) a rapid DNA-based macroarray system that determines the presence of fungal pathogens responsible for post-harvest rot (Sholberg et al. 2005a) and 2 2) the identification of soil bacteria that can control fungal post-harvest pests (Hirkala et al. 2007; Hynes et al. 2008). This study will assist the project’s long term goal to develop Canadian-based technologies that detect, predict and mitigate post-harvest disease in pome fruit and that are consistent with integrated pest management and sustainable production practices. 1.1 Post-harvest pathogens Fruit infections can occur throughout the growing season, at harvest, during storage, transit and at the retail level (Jones and Aldwinckle 1990). Although global economic losses can be attributed to growing season diseases, post-harvest pathogens, despite modern storage facilities, can cause significant annual losses (Janisiewicz and Korsten 2002). Post-harvest diseases, or latent infections, are asymptomatic throughout the growing season and disease is manifested only after periods of storage. As a result, these pathogens are often difficult to treat and prevent. Post-harvest fungal pathogens can be categorized in many ways, one of which describes the pathogenic point of entry. Rot characterization can include lenticel infections (pores that exist on the surface of the apple that facilitate gas exchange), core and eye rots and wound pathogens (Jijakli and Lepoivre 2004). Common lenticel infections include bitter and bull’s eye rot and are commonly caused by Colletotrichum gleosporioides and Cryptosporiopsis curvispora, respectively (Jijakli and Lepoivre 2004). Core rots develop when an open sinus stems from the calyx end of the apple into the core (Miller 1959), which provides a point of entry for an invading pathogen. Alternaria spp. are the most commonly isolated fungi from core rots, but other fungal species such as Cladosporium, Botrytis, Candida and 3 Fusarium have also been identified (Ellis and Barrat 1983). Dry eye rots, characterized by a shallow, hard rot with a red border are predominantly caused by Botrytis cinerea. Opportunistic wound pathogens invade apples as a result of fruit injury incurred by mishandling or harsh weather conditions. These types of infections are responsible for significant storage losses. Three important fungal pathogens that cause worldwide decay include Penicillium expansum (blue mold), Botrytis cinerea (grey mold) and to lesser extent, Mucor piriformis (mucor rot) (Michailides and Spotts 1990; Sanderson and Spotts 1995; Rosenberger 1990). 1.1.1 Penicillium expansum Penicillium expansum Link, or apple blue mold, is a filamentous Ascomycete that causes one of the most important North American post-harvest storage diseases of pome fruit. Before the introduction of controlled atmosphere (CA) storage and fungicides, it accounted for 90% of post-harvest apple diseases (Rosenberger 1990). P. expansum produces the mycotoxin, patulin, which has been found in apple product derivatives such as apple juice, ciders, puree, vinegar, baby food and whole apples (Abramson et al. 2009; Doores 1983; Piemontese et al. 2005; Watanabe and Shimizu 2005). Its detection and control is, therefore, of utmost importance for food health and safety. Although an opportunistic wound pathogen, P. expansum infection can originate from stem-end invasions, core rots and through lenticels. Infected fruit are characterized by light to dark brown, fleshy, circular lesions that surround wounds. Older lesions may produce a bluish spore mass. P. expansum can also form a dense powdery mass at the centre of the lesion (Rosenberger 1990) and can be characterized by an earthy, musty odour commonly used as a diagnostic tool (Vikram et al. 2004). 4 Spores of P. expansum are ubiquitous and can cause infection within orchards and packinghouses. This fungus can survive in organic debris on the orchard floor and within soil. Conidia also exist in the air and on apple surfaces (Lennox et al. 2003). Within packinghouses, spores have been isolated from fungicide-drench solutions, flume water, dump-tank water, air and storage room walls (Sholberg and Stokes 2006). Conidia can survive from season to season on contaminated bins, picking boxes and storage walls (Rosenberger 1990). Sanitation, harvesting pre-senescent fruit and handling methods are the best strategy for minimizing disease incidence. This control can also be facilitated by fungicides applied as pre-harvest sprays or post-harvest dips (Eckert and Ogawa 1988). For example, pyrimethanil was used as a pre-harvest fungicide and applied twenty days prior to apple harvest. After apples were stored for six months, post-harvest blue mold was significantly reduced when compared to apples that were not treated with fungicides (Sholberg et al. 2005b). Mertect®, with an active ingredient of thiabendazole (TBZ), is an example of a post-harvest fungicide application. It was once highly effective; however, prolonged exposure to such agents has led to TBZ-resistant fungal strains (Sholberg and Haag 1996) that render chemical controls ineffective. 1.1.2 Botrytis cinerea Botrytis cinerea Pers.:Fr is an Ascomycete and, like P. expansum, is an important post-harvest pathogen of pome fruit causing minor rot throughout the growing season and significant rot within packinghouses. It is the most important post-harvest pathogen of pears and is second to blue mold in importance to apples (Rosenberger 1990). B. cinerea infection may originate from wounds, stem punctures, or the stem or calyx portion of the 5 fruit. Upon infection, B. cinerea inflicts a light to dark brown, spongy decayed area around the damaged portion of the fruit. Under high humidity, fluffy white to grey mycelium and spore masses may appear on the decayed area. The optimal growth temperatures for this pathogen range from 20°C to 22°C however, conidia are capable of growth at as low as -2°C (Coley-Smith et al. 1980). B. cinerea spores are ubiquitous, colonizing available organic matter within orchards. Throughout the growing season, this fungus can cause dry eye rot. B. cinerea also can be carried into packinghouses via contaminated storage bins. Once inside, conidia are predominantly spread by air currents and water splash. This pathogen is also known as a nest or cluster rot as secondary infection can occur through fruit to fruit contact; fungi on infected fruit can colonize healthy fruit and spread disease (Rosenberger 1990). 1.1.3 Mucor piriformis Mucor piriformis belongs to the phylum Zygomycota and, until recently, was thought to be of minor importance as a post-harvest pathogen. However, M. piriformis is capable of causing major decay problems in fruit such as strawberries, pome and stone fruit. M. piriformis is typically saprotrophic in orchard soil and infects fruit through the stem or calyx end and puncture wounds (Michailides and Spotts 1990). Upon infection, the area surrounding the lesion becomes soft, watery, light brown and easily separable from the fruit tissue. Often, grey mycelium with dark sporangia appears upon the decayed area. Mucor rot has a distinct sweet smell with a clear, sticky exudate. This fungus sporulates from -1 to 24°C, with optimal growth at 21°C. Fungicides that are currently registered to control other post-harvest pome fruit pathogens are ineffective 6 against M. piriformis and other Zygomycetes (Michailides and Spotts 1990). Although ScholarTM, with active ingredient fludioxonil, has been shown to be effective against M. piriformis (P. Sholberg, personal communication), sanitation methods are often the most effective method for preventing infection (Michailides and Spotts 1990). 1.2 Post-harvest factors that influence post-harvest pathology Because of the ubiquitous nature of fungal spores, their presence throughout the apple growing season can influence post-harvest pathology. P. expansum, B. cinerea and M. piriformis can colonize organic matter on the orchard floor and within soil (Lennox et al. 2003; Michailides and Spotts 1990). P. expansum and B. cinerea conidia can additionally exist in the air and on fruit surfaces (Lennox et al. 2003). Low precipitation, physiological crop condition and the use of pre-harvest fungicides will have a reducing effect on decay levels after harvest (Sholberg and Conway 2004; Sholberg et al. 2003; Sholberg et al. 2005b). However, post-harvest factors can also influence disease incidence. Within packinghouses, spores can survive in fungicide-drench solutions, flume water, dump-tank water and in the air and on walls of storage rooms (Lennox et al. 2003; Sholberg and Stokes 2006). Therefore, sanitation affects disease accumulation. The post-harvest system offers a unique, closed or semi-closed environment where manipulating temperature and atmosphere within the storage room and the use of chemical senescence inhibitors like 1-methylcyclopropene (1-MCP) are often easier to manage in comparison to pre-harvest factors that can contribute to disease. 7 1.2.1 Temperature The first experimental cold storage facility in British Columbia was built in Summerland in 1929 (AAFC 2002). Optimal refrigeration temperatures range from -0.5°C for Braeburn apples in South Africa to as high as 5°C for Belle de Boskoop apples grown in the Netherlands (Kupferman 2003). Lowered temperatures slow the rate of apple respiration, thus retarding ripening (AAFC 2002). Temperature management is also critical to post-harvest disease control. Post-harvest fruit pathogens grow optimally between 20 to 25°C. Some fungi have minimum growth temperatures as low as -2°C and cannot be completely controlled without freezing the fruit. Their growth at these temperatures, however, is significantly reduced which leads to a reduction in post-harvest decay levels (Sholberg and Conway 2004). 1.2.2 Controlled atmosphere “Normal air” refers to atmospheres that consist of 78 – 79% N2, 20 – 21% O2, ~ 0.03% CO2 and trace amounts of other gases. Controlled atmosphere (CA) refers to atmospheres that differ from “normal air” and are under strict control (Yahia 2009). Here, optimal temperatures persist while oxygen and carbon dioxide concentrations are decreased and increased, respectively (Morales et al. 2007). The concept of CA relies on the fact that harvested fruits use oxygen and produce carbon dioxide. If the amount of oxygen is limited, fruits will not ripen or will ripen slowly (Yahia 2009). Apples are often stored under CA as this environment has been shown to be effective in delaying the onset of storage diseases (Smock 1979). Commercial CA storage can range from 1% – 3% O2 and 0.4% – 4.5% CO2 (Kupferman 2003). Additional studies revealed that as high as 13% CO2 prevented close to 100% spore germination of P. expansum after 8 twenty days (Cossentine et al. 2004). However, CO2 levels this high are not commercially acceptable as it may lead to fruit injury resulting in a decrease in fruit quality. 1.2.3 1-Methylcyclopropene 1-Methylcyclopropene (1-MCP) is a synthetic cyclopropene that blocks ethylene receptors and prevents apple ripening (Blankenship and Dole 2003). It has the ability to maintain post-harvest fruit and vegetable quality and its use provides insight into the role of ethylene in fruit senescence (Watkins 2006). Ripening physiology and quality of apples in response to 1-MCP application has been intensively studied (Watkins 2006). Less studied, are the effects of 1-MCP on disease incidence. It is thought that by preventing the ethylene-associated ripening process, apples will be better able to resist pathogens (Watkins 2006). However, prevention of decay in 1-MCP-treated apples has been inconsistent. For example, Golden Delicious apples treated with 1-MCP and P. expansum showed decreased disease incidence (Saftner et al. 2003). Alternatively, disease severity increased in Golden Delicious apples inoculated with P. expansum or Colletotrichum acutatum and treated with 1-MCP (Janisiewicz et al. 2003; Leverentz et al. 2003). Although initially successful, there is the potential of certain decay problems associated with 1-MCP use (P. Sholberg, personal communication). 1.3 Methods of detection and identification Traditional methods for fungal identification were primarily based on morphology or phylogenetic characteristics. However, such techniques have limitations as morphological characterizations rely on fungi to be isolated or cultured, resulting in an 9 underestimation of the microbial community of interest (Bridge and Spooner 2001; Mazzola 2004). The advancement of molecular biology has led to more specific and sensitive fungal DNA-based detection methods that replaced assumptions made in previous studies. For example, Cruickshank and Pitt (1987) used enzyme gel electrophoresis, or a zymogram, to differentiate between Penicillium species. This study hypothesized that like species will display like zymograms. Results mostly confirmed taxonomy based on morphology; however, taxonomic differences were perceived. Detection techniques based on morphology or electrophoresis are time consuming. A rapid method of detection is therefore required. Greater specificity, sensitivity and speed can be attributed to the advancement of the polymerase chain reaction (PCR), a rapid and sensitive primer-mediated enzymatic amplification of target DNA sequences (Saiki et al. 1985). Common PCR-based identification methods include denaturing gradient gel electrophoresis (DGGE), temperature gradient gel electrophoresis (TGGE) or terminal restriction length polymorphisms (T-RFLP) (Mazzola 2004). Although PCR-based identification methods provide a more accurate picture of a microbial community, these methods are often limited by gel resolution or, more importantly, the inability to detect a large number of different pathogens simultaneously (Sholberg et al 2005a). The advent of DNA macroarray technology circumvents this problem. This system utilizes labelled PCR products hybridized to a nylon membrane that contains an assortment of anchored oligonucleotides or probes. Due to a chemiluminescent reaction, pathogen DNA, for example, is captured on X-ray film and is represented as a pattern of pre-determined dots. The presence of a signal indicates the presence of a pathogen that 10 can be identified and quantified. Levesque et al. (1998) were among the first to utilize macroarray technology for plant pathogen identification. This method was further developed for detection of economically important pre-harvest (Sholberg et al. 2005a) and post-harvest diseases of pome fruit (Hirkala et al 2007). These technologies have great potential as they can be used as a high-throughput detection and diagnostic technique that can identify numerous microorganisms across disciplines and environments. 1.4 Methods of control 1.4.1 Chemical control Fungicides are used to prevent post-harvest decay. In Canada, two synthetic post- harvest controls are currently registered, Mertect® and Scholar® with active ingredients thiabendazole (TBZ) and fludioxonil, respectively (PMRA 2007). Although initially effective, fungicides are now a major cause for concern. Pathogens have developed resistance to chemical controls (Errampalli et al. 2006; Sholberg et al. 2005c) which negates their effectiveness. There are also growing health concerns that surround pesticide use (Hancock et al. 2008) and a demand for environmentally friendly sustainable post-harvest practices. An alternative is therefore of high priority. One possibility is the implementation of biological controls. 1.4.2 Biological control Biological control can be defined as “the reduction of the amount of inoculum or disease-producing activity of a pathogen accomplished by or through one or more organisms other than man” (Cook and Baker 1983). As cited in Cook and Baker (1983), 11 biological control dates back to the early 1920s in which Hartley (1921) attempted to control damping off in pine seedlings with fungal antagonists. In 1927, Millard and Taylor tried to control common potato scab by the addition of Streptomyces praecox and grass clippings to autoclaved potted soil. Both studies noted that competition for nutrients may be a putative mechanism of action (Cook and Baker 1983). Sixty years later, the first commercial bacterial biological control agent (BCA), Agrobacterium radiobacter strain K84, was registered with the United States Environmental Protection Agency in 1979. It controlled crown gall caused by Agrobacterium radiobacter pv. tumifaciens. Strain K84 lacked the tumor-inducing (Ti) plasmid present within the pathogenic strain and produced a bacteriocin, which inhibits the growth of certain tumorigenic-strains, and to which K84 was insensitive (Cook and Baker 1983). Although the beginnings of biological control of plant pathogens date back to the early 1920s, practicing biological control within a post-harvest setting has been much less common. Currently, there are three BCAs registered for post-harvest use (Droby et al. 2009). The first and second, developed by JET Harvest Solutions (Longwood, FL), are Bio-Save 10LP and 11LP (Longwood, FL) with active ingredients of Pseudomonas syringae strain ESC-10 and ESC-11, respectively. Bio-Save 10LP targets post-harvest blue and grey mold and mucor rot in pome fruit while Bio-Save 11LP targets Rhizopus soft rot on sweet potatoes. The third, Shemer (Metschnikowia fructicola) is commercially used in Israel for prevention of sweet potato and carrot storage decay. Two yeast-based products, AspireTM and YieldPlus, are no longer available (Droby 2009). Biological control of post-harvest pathogens of pome fruit using bacterial and 12 yeast antagonists has been studied by numerous research groups (Calvo et al. 2007; Etebarian et al. 2005; Janisiewicz 1988; Mikani et al. 2008; Morales et al. 2008; Nunes et al. 2002; Zhou et al. 2001). For example, Candida sake strain CPA-1 significantly decreased P. expansum lesion size by > 80% and reduced disease incidence by 50% (Teixidó et al. 1999). Cryptococcus laurentii and Metschnikowia pulcherrima also provided control against P. expansum (Conway et al. 2007). The epiphytic bacterium, Rahnella aquatilis, significantly inhibited P. expansum and B. cinerea on apples at 15°C and 4°C, respectively (Calvo et al. 2007). A saprophytic strain of Pantoea agglomerans also reduced P. expansum rot on Golden Delicious apples at 1°C in air and at 1°C in a low oxygen atmosphere by 81% and 100%, respectively (Nunes et al. 2002). A study conducted by Etebarian et al. (2005) showed that Pseudomonas fluorescens isolate 1100- 6 significantly reduced the incidence of P. expansum and P. solitium on apples after 11 days at 20°C and 25 days at 5°C. In Canada, there are an increasing number of registrations and products under evaluation that are based upon microbially-active substances (Bailey et al. 2010). However, there are none registered for post-harvest use (Droby et al. 2009). A demand therefore exists for the production and commercialization of a BCA for post-harvest use on pome fruit. A collection of rhizobacteria, isolated from the roots of legumes grown in Saskatchewan soils, were compiled and initially characterized for plant growth promoting traits such as the production of siderophores, ACC deaminase, root elongation and the suppression of legume fungal pathogens (Hynes et al. 2008). Fifteen of these isolates were further tested in vitro for suppression of P. expansum, B. cinerea and M. piriformis (Stokes et al. 2006). Nine isolates provided fungal control and were 13 then tested in situ. Four isolates provided consistent control at both 20ºC and 1ºC; isolates 1-112, 2-28, 4-6 and 6-25 (Hirkala, unpublished data) in combination with 1100-6 were selected for further study. 1.5 Mechanisms of action The interaction between host, antagonist and pathogen is a complex, ecological relationship with high variability depending on the system of focus. The ability of an antagonist to suppress a pathogen may be due to more than one mechanism of action (MOA). A sound mechanistic understanding is essential for formulation development and biocontrol registration (Spadaro and Gullino 2004). Possible mechanisms include antibiosis (Janisiewicz et al. 1991; Kamesnky et al. 2003; Meziane et al. 2006), competition for nutrients (Elad 1996), parasitism (Frankowski et al. 2001; Watanabe et al. 2007) and induction of pathogen resistance in host tissue (Benhamou et al. 2000; Spadaro and Guillino 2004; Terry and Joyce 2004). 1.5.1 Antibiosis One method of antagonistic action may be through the production of antibiotics. For example, pyrrolnitrin produced by strains of Serratia plymuthica, was associated with fungal suppression (Kamesnky et al. 2003; Meziane et al. 2006). To determine whether or not pyrrolnitrin assisted in antifungal activity, Meziane et al. (2006) created an S. plymuthica mutant that lacked the gene responsible for pyrrolnitrin production. The pyrrolnitrin-deficient mutant, IC1270-P1, lost its antifungal activity when tested against pathogens Penicillium digitatum and Penicillium italicum. In addition, purified pyrrolnitrin was effective in suppressing disease symptoms of P. digitatum and P. 14 italicum. Janisiewicz et al. (1991) also isolated pyrrolnitrin from Pseudomonas cepacia for control of B. cinerea and P. expansum on apples and pears. Pyrrolnitrin was found to provide control; however, the level of control varied due to wound type (cut, nail, bruise), pyrrolnitrin concentration and storage temperature (2ºC and 24ºC). A post-harvest dip solution of pyrrolnitrin may therefore be an effective solution against post-harvest pathogens. It should also be noted that the post-harvest fungicide, ScholarTM, has an active ingredient of fludioxonil which belongs to the chemical class of phenylpyrroles. Phenylpyrroles are derived from pyrrolnitrin and are considered a reduced-risk chemical control agent (Errampalli 2004). 1.5.2 Competition for nutrients Competition for nutrients has been widely studied due to nutritional demands of both antagonists and pathogens. B. cinerea, as with most necrotrophs, requires nutrients for germination and the initiation of the pathogenic process. In the absence of nutrients, B. cinerea becomes highly susceptible to degradation and control (Elad 1996; Janisiewicz et al. 2000). It is postulated that microorganisms capable of efficiently utilizing nutrients will make successful biological control agents (Elad 1996). Janisiewicz et al. (2000) proposed a simple way to study competition for nutrients that uses a 24-well tissue culture plate with cylindrical insets with 0.45 µm membranes attached to the bottom. Individual cylinders are placed in each of the 24-wells thereby permitting media nutrient and metabolite interchange while preventing pathogen and antagonist contact due to physical separation. In this system, competition for nutrients can be studied independently from competition for space. Bencheqroun et al. (2007) used this method to determine, in vitro, that the biocontrol antagonist, Aureobasidium pullulans, was 15 competing for nutrients, especially the amino acids glycine, glutamic acid and serine, with the pathogen, Penicillium expansum. 1.5.3 Parasitism Antagonists can directly parasitize pathogens of post-harvest apples. One such method is through the production of lytic or cell wall degrading enzymes such as chitinase and β-1,3-glucanase. Bacterial strains that produce lytic enzymes will most likely have antifungal properties via cell wall hydrolysis. For example, S. plymuthica strain HRO-C48 was isolated from the rhizosphere of oilseed rape and shown to have antifungal properties associated with chitinase production. Frankowski et al. (2001) isolated and characterized two chitinolytic enzymes, one endochitinase (E.C. 3.2.1.14), CHIT60, and one N-acetyl-β-1,4-D-hexosaminidase (E.C. 3.2.1.52), CHIT100. In vitro, CHIT60 and CHIT100 showed direct inhibitory activity on spore germination and germ tube growth of B. cinerea. Alternate parasitic activity may involve direct attachment of antagonist to pathogenic fungal hyphae (Chan and Tian 2005; Watanabe et al. 2007). 1.5.4 Induced systemic resistance Induced systemic resistance (ISR) is an observed phenomenon in which microorganisms activate host plant mechanisms that enhance their defensive capacity against potential pathogenic invasions. Protection can occur both locally and on areas of the plant that did not come into contact with inducing microorganisms (van Loon 2007). For example, Benhamou et al. (2000) determined that S. plymuthica strain R1G64 mediated induced systemic resistance in cucumber to protect against infection caused by Pythium ultimum. Upon pre-treatment with S. plymuthica, P. ultimum disease incidence 16 was reduced; cucumber root cells were structurally and biochemically modified with the formation of phenolic-enriched occluded depositions and structural barriers. These structural barriers were hypothesized to prevent pathogen movement towards the vascular stele. In ISR experiments, it is important to spatially separate inducing microorganisms and challenging pathogens (ie. root and leaf) to ensure that protection is plant- not microorganism-mediated (van Loon 2007). 1.6 Green fluorescent protein Green fluorescent protein (GFP) is made up of 238 amino acids and exhibits green fluorescence at a peak emission of 509 nm by absorbing blue light maximally at 395 nm and minimally at 470 nm (Morin and Hastings 1971). In its native form, GFP was first discovered during the purification of the bioluminescent protein, aequorin, isolated from jellyfish (Aequorea victoria) (Shimomura et al. 1962). In A. victoria, green fluorescence occurs by an intermolecular energy transfer between aequorin and GFP; Ca2+ and aequorin interact and emit blue light, some of which is absorbed by GFP causing a color shift towards green (Morise et al. 1974). However, it wasn’t until 1992 that the influence of GFP was realized. Prasher et al. (1992) paved the way for future GFP application by cloning and sequencing both cDNA and genomic clones from Aequorea victoria. In 1994, Chalfie et al. first expressed GFP within a living organism, highlighting sensory neurons in nematodes. These landmark studies influenced modern science as purified GFP has become one of the most important reporter genes in biology (Bloemberg 2007). GFP is comprised of an eleven-stranded β barrel with a coaxial helix running through the centre. The chromophore is formed from the spontaneous cyclization of the 17 central helix consisting of the tripeptide, Ser65-Try66-Gly67 (Ormo et al. 1996). Introducing random amino acid substitutions into the twenty amino acids that flank the chromophore has led to altered forms of GFP some of which exhibit greater fluorescence. The first GFP modification was derived from a point mutation that altered Serine 65 to Threonine (S65T) (Heim et al.1995). This alteration shifted the excitation and emission maxima to 490 and 510 nm, respectively. Alterations of GFP not only affected intensity, but also increased color variety that is within the blue, cyan and yellow regions of the electromagnetic spectrum (Shaner et al. 2007). Green fluorescent protein can be used as a reporter or tag to mark whole cells, study protein localization and monitor gene interactions and interactions between microorganisms. Its incorporation is non-invasive, it does not require an additional substrate for bioluminescence, nor does visualization necessitate fixation or staining protocols (Bloemberg 2007). The gene product of gfp can be expressed in both prokaryotes and eukaryotes (Chalfie et al. 1994) with no background gene expression (Tombolini et al. 1997). It is highly stable in the presence of denaturants and proteases and persists at high temperatures (65°C) and a wide range of pH values (6-12) (Bloemberg 2007; Chalfie et al. 1994; Ward et al. 1980). Applications of GFP in biological control experiments can help elucidate mechanisms of action (MOA). Watanabe et al. (2007) transformed two fungal species with gfp, Trichoderma asperellum SKT-1, the antagonist, and Gibberella fujikuroi N-68, a pathogenic root fungus responsible for Bakanae disease of rice seedlings. The fluorescent protein was imaged by confocal scanning laser microscopy (CSLM) and mycoparasitism was suggested as the putative MOA. Scanning electron microscopy 18 (SEM) and CSLM showed a disappearance of GFP in G. fujikuroi upon contact with T. asperellum suggesting cell wall degradation. In another example, Bolwerk et al. (2003) transformed red fluorescent protein (RFP) into two putative BCAs, Pseudomonas fluorescens WCS365 and Pseudomonas chlororaphis PCL1391, that have been shown to control Fusarium oxysporum f. sp. radicis-lycopersici, a causal agent of tomato foot and root rot (TFRR). Both Pseudomonas spp. contained RFP, whereas Fusarium oxysporum f. sp. radicis- lycopersici harboured GFP. CSLM revealed that both antagonists colonized the tomato root more quickly than the fungal pathogen and that the bacterial presence hindered root infection. A proposed MOA was that the presence of fungi initiated bacterial colonization of hyphae and subsequent production of fungal secondary metabolites. 1.7 Objectives This thesis will build upon preliminary research and address the following three objectives: 1) validate the use of DNA macroarrays under field conditions; 2) assess bacterial antagonist performance under semi-commercial and commercial conditions and 3) determine how long and in what capacity a selected antagonist will colonize the fruit surface. I hypothesize that: 1) Macroarray technology will accurately identify and quantify post-harvest apple pathogens throughout the growing season and that pathogen prevalence will correlate with disease incidence post-harvest. 2) One or more of the bacterial antagonists will provide significant pathogen control in semi-commercial and commercial storage conditions. 19 3) The use of green fluorescent protein will facilitate visualization of bacterial colonization and survival on apple surfaces. 20 CHAPTER 2: MATERIALS AND METHODS 2.1 Antagonists Four bacterial antagonists were obtained from Dr. L. Nelson and were identified by 16S rRNA sequence and fatty acid methyl ester analysis as Pseudomonas fluorescens (isolates 1-112, 2-28, 4-6) and Serratia plymuthica (isolate 6-25) (Hynes et al. 2008). Isolate 1100-6, Pseudomonas fluorescens (Etebarian et al. 2005), was provided by Dr. Peter Sholberg of the Pacific Agri-Food Research Centre (PARC) in Summerland, BC. The bacteria were grown in half-strength tryptic soy broth (½ TSB) (Becton, Dickinson and Company, Sparks, MD) at 28°C with shaking at 220 rpm. The final culture was centrifuged for 20 min at 3800 g and held at 20°C. Cell pellets were resuspended in sterile phosphate buffered saline solution (PBS) (1.2 g/L Na2HPO4, 0.18 g/L NaH2PO4, 8.5 g/L NaCl). The optical density (OD600) was determined and the bacterial cells were diluted to 108 CFU/ml according to a standard curve relating OD600 to CFU/ml. 2.2 Pathogens Penicillium expansum Link strain 1790, Botrytis cinerea Pers.:Fr strain B27 and Mucor piriformis Fischer strain 536 were provided by Dr. Peter Sholberg and grown on half-strength potato dextrose agar (½ PDA) (HiMedia Laboratories PVT. Ltd., India) for 7 days at 22°C. A spore suspension was created using sterile water and Tween 20 (MP Biomedicals, LLC, Solon, OH) and then enumerated using a Petroff-Hauser counter. 21 2.3 Orchard field sites This study incorporated four conventional (non-organic) orchard field sites from which data were collected over two years. Two research orchards were located at PARC in Summerland, BC. Field 3 (49° 33’ 58.49” N; 119° 38’ 41.93” W) contained 166 Jonagold apple trees planted at high density. Field 12 (49° 33’ 54.81” N; 119° 38’ 56.73” W) contained 120 Gala apple trees planted at high density. The two commercial orchards were located in Kelowna, BC. The Kiran orchard (49° 50’ 41.45” N; 119° 24’ 54.24” W) grew Gala apples planted at high density. The Reekie orchard (49° 50’ 58.78” N; 119° 23’ 32.79” W) grew Red Delicious apples planted at medium density. I-rods were used in this study to collect aerial spore samples. I-rods are clear polystyrene rods that rest within a sampling head of a rotating spore trap. When the motor spins, centrifugal force causes the I-rods coated in silicone grease to extend downwards at a 90º angle and collect airborne particles (Aerobiology Research Laboratories, Nepean, ON). In the 2007 and 2008 growing seasons, Fields 3 and 12 contained one I-rod station per field. The Kiran orchard contained one I-rod station in both 2007 and 2008 and the Reekie orchard contained one I-rod station in 2008 only. Average daily temperatures (°C) and total daily precipitation (mm) from Environment Canada (2008) were plotted and used to compare with pathogen DNA detected throughout the apple growing seasons (Figure 1). Fungicide spray records for Fields 3 and 12 were provided. On May 15, 2007, the fungicide, Nova (myclobutanil) was applied to Field 12. On September 15, 2008, the fungicide Funginex (triforine) was applied to Field 3. 22 0 5 10 15 20 25 30 0 5 10 15 20 25 30 0 25 50 75 100 125 150 T o t a l D a i l y P r e c i p i t a t i o n ( m m ) A v e r a g e D a i l y T e m p e r a t u r e ( ° C ) Days after Blossom 0 5 10 15 20 25 30 0 5 10 15 20 25 30 0 25 50 75 100 125 150 T o t a l D a i l y P r e c i p i t a t i o n ( m m ) A v e r a g e D a i l y T e m p e r a t u r e ( ° C ) Days after Blossom 0 10 20 30 40 50 60 70 0 5 10 15 20 25 30 0 25 50 75 100 125 150 T o t a l D a i l y P r e c i p i t a t i o n ( m m ) A v e r a g e D a i l y T e m p e r a t u r e ( ° C ) Days after Blossom 0 10 20 30 40 50 60 70 0 5 10 15 20 25 30 0 25 50 75 100 125 150 T o t a l D a i l y P r e c i p i a t i o n ( m m ) A v e r a g e D a i l y T e m p e r a t u r e ( ° C ) Days after Blossom A. C. B. D. Figure 1. Average daily temperatures (°C) (solid line) and total daily precipitation (mm) (dashed line) at Kelowna, BC during the growing seasons of 2007 (A) and 2008 (B) and in Summerland, BC during the growing seasons of 2007 (C) and 2008 (D) as indicated by Environment Canada (2008). 23 2.4 pAG408 The gfp-containing plasmid, pAG408 (Suarez et al. 1997), was provided by Dr. Darren Korber, University of Saskatchewan, Saskatoon, Sask. pAG408 was maintained within the donor strain, E. coli S-17 λ pir, on Luria Burtani (LB) agar (10 g/ L tryptone, 5 g/L yeast extract, 10 g/L NaCl) supplemented with 50 µg/ml of kanamycin (kan) and 30 µg/ml of gentamicin (gen). To test for antibiotic efficacy, pAG408 in E. coli S-17 λ pir was streaked onto three replicate plates each of LB agar, LB agar + 50 µg/ml kan, LB agar + 30 µg/ml gen and LB agar + 50 µg/ml of kan and 30 µg/ml of gen and grown at 28°C for 48 h. Furthermore, pAG408 plasmid confirmation was verified by extracting it from E. coli S- 17 λ pir using the Wizard ® Plus Minipreps DNA Purification System (Promega, Madison, WI) as per manufacturer’s instructions, and digesting 1 µg of DNA with 1 U HindIII (Invitrogen, Carlsbad, CA) for 1 h at 37°C. 2.5 Construction of the gfp-transconjugant Bacterial isolate 4-6 was transformed with pAG408 via bacterial conjugation. Donor cells (E. coli S-17 λ pir containing pAG408) and recipient cells (4-6) were grown overnight at 28°C with shaking at 220 rpm. One millilitre of each culture was centrifuged at 10 000 x g for 20 min at 4°C. The supernatants were removed, the pellets washed twice with PBS then resuspended in 50 µl of PBS. Fifty microlitres each of the donors and recipients were combined and vortexed. The combined cultures were spotted on 0.22 µm sterile nylon membrane filters (Whatman, Maidstone, England) placed on LB agar plates supplemented with 50 µg/ml of kan and 30 µg/ml of gen. The plates were 24 incubated for 24 h at 28°C to allow for bacterial conjugation. The bacterial mixtures were then removed from the filters and plated on sodium citrate agar plates (0.2 g/L MgS04, 1.0 g/L NH4H2PO4, 2.0 g/L sodium citrate, 5.0 g/L NaCl, 15 g/L agar) supplemented with 50 µg/ml of kan and 30 µg/ml of gen and then incubated for 48 h at 28°C. The combination of sodium citrate growth medium and antibiotics eliminated E. coli S-17 λ pir pAG408 cells and non-transformed bacterial cells; E. coli cannot use citrate as a carbon source and the wild-type biological control isolates cannot withstand the antibiotics. For a negative control, non-transformed Pseudomonas fluorescens isolate 1-112, Serratia plymuthica isolate 6-25 and E. coli S-17 λ pir containing pAG408 were spread on sodium citrate agar plates supplemented with 50 µg/ml of kan and 30 µg/ml of gen. To confirm successful conjugation, 10 putative transconjugants were selected at random and their genomic DNA was extracted by boiling. One colony was placed in 100 µl of sterile water and the supernatant was used as the DNA template for the PCR that utilized gfp-specific primers. Three microlitres of DNA were used in the 25-µl reaction containing 1x PCR Buffer (TetraLink International, Buffalo, NY), 2.5 mM MgCl2, 0.1 mM dNTP mix (Invitrogen, Carlsbad, CA) , 0.4 µM of each forward (gfp F: 5’- GAGTAAAGGAGAAGAACTTTTCA-3’) and reverse primers (gfp R: 5’- TTATTTGTATAGTTCATCCATG-3’) (Suarez et al. 1997) and 1 U of UltraTherm DNA polymerase (TetraLink International, Buffalo, NY). Amplification was performed on the Techne TC-3000 thermocycler (Barloworld Scientific, London, England) with the following conditions: 96°C for 5 min followed by 35 cycles of 95°C for 1 min, 54.8°C for 1 min and 72°C for 1 min with a final extension cycle at 72°C for 7 min. 25 2.6 Plasmid stability Testing of the stability of exogenous DNA within bacterial isolate 4-6 was conducted according to Bloemberg et al. (1997) except that tryptic soy medium and the antibiotics, kanamycin and gentamicin, were used. Based on replicated stability results and the 2007 biocontrol experimental results, the promising gfp-tranconjgant biological control strain, P. fluorescens isolate 4-6, was selected as a suitable antagonist for bacterial survival and colonization studies. 2.7 P. fluorescens 4-6-gfp fitness analysis Growth rates of 4-6-gfp and its wild-type counterpart (4-6 wild-type) were compared. Three replicates each of 4-6-gfp and 4-6 wild-type were grown in 10 ml of ½ TSB and incubated at 28°C with shaking at 220 rpm for 24 h. Cultures were spun for 20 min at 5000 x g and resuspended in 10 ml of phosphate buffered saline (PBS). Optical density (OD600) was determined and cultures were diluted and standardized with PBS to within 0.100 of each other. One hundred microlitres of standardized culture were added to 10 ml of ½ TSB. OD600 readings were recorded at 0, 2, 4, 6, 8, 10, 12 and 24 h. The three replicates were averaged and represented graphically (time vs OD660). Growth rates (µ) were calculated with the formula µ = (log10 N-log10 No) x 2.303 / t-to) where N was a final cell number, No was an initial cell number, t was the time at N and to was the time at No. Carbon utilization profiles of 4-6-wild-type and 4-6-gfp were also compared by determining carbon profiles using BioLog Phenotype MicroArrays for Microbial Cells 26 (Biolog, Inc., Hayward, CA) (microplate PM1 and PM2) as per the manufacturer’s instructions 2.8 Spore, apple tissue collection and fruit washings Apple leaves, blossoms and I-rods were collected at orchard field sites throughout the apple growing seasons (Table 1). In general, sampling was conducted three times per week at the beginning (early) and end (late) of the growing seasons in 2007 and 2008. Throughout the mid-growing seasons, samples were collected approximately once per week. At harvest, approximately 1250 apples were collected at random from each of Fields 3 and 12 and approximately 300 apples were collected at random from the Kiran and Reekie orchards. Fruit surface washings were conducted on 20 apples each from Fields 3 and 12 in 2007 and 20 apples each from Fields 3, 12, the Kiran and the Reekie orchards in 2008. Twenty apples from each orchard were washed in a 600-ml beaker containing 200 ml of sterile distilled water and 200 µl of Tween 80. Each fruit was shaken for 5 min at 120 rpm and sonicated for 5 min (P. Sholberg, personal communication). After all fruit were washed, the solution was centrifuged at 13 400 g for 10 min. The supernatant was decanted and this process was repeated five times to a final volume of 10 ml that was centrifuged for 15 min at 7430 g. The supernatant was decanted and the remaining cells were transferred to a 2-ml tube and centrifuged at max (16 100 g) for 1 min. The supernatant was removed and 100 µl of water were added for storage at -20°C. 27 Table 1. Frequency of aerial and plant tissue sampling for Fields 3, 12 and the Kiran and Reekie orchards in the early, middle and late growing seasons of 2007 and 2008. Number of samples within sampling period (n) Field Sample type Early Mid Late Total 2007 Growing Season Field 3 aerial 14 6 6 26 plant tissue 2 3 2 7 Field 12 aerial 16 6 5 27 plant tissue 2 3 2 7 Kiran aerial 9 7 6 22 plant tissue 4 7 5 16 Reekie aerial 0 0 0 0 plant tissue 6 8 4 18 2007 Total 53 40 30 123 2008 Growing Season Field 3 aerial 18 11 3 32 plant tissue 4 0 2 6 Field 12 aerial 19 11 3 33 plant tissue 4 0 2 6 Kiran aerial 17 24 11 52 plant tissue 18 24 10 52 Reekie aerial 15 20 11 46 plant tissue 16 24 11 51 2008 Total 111 114 53 278 2.9 Validation and field testing of DNA macroarrays 2.9.1 DNA extraction DNA was extracted from I-rods, plant tissue samples and fruit washings using the Power Soil DNA Isolation Kit (Mo Bio Laboratories, Inc., Carlsbad, CA) according to the manufacturer’s instructions. For DNA isolation, two I-rods were used per aerial 28 extraction and 0.25 g of plant tissue were used for tissue extractions. DNA was eluted in 100 µl of Buffer 5 and stored at -20°C until required for further use. 2.9.2 PCR amplification PCR amplification was used to amplify universally conserved regions of the fungal genome, ribosomal DNA interspacer regions (rDNA ITS) and the β-tubulin gene. Approximately 10 – 20 ng of extracted DNA were used in 25-µl reactions containing 1x PCR Buffer (TetraLink International, Buffalo, NY), 2.5 mM MgCl2, 0.1 mM digoxigenin (Roche Diagnostics GmbH, Mannheim, Germany) (DIG)-dNTP mix , 0.4 µM of each forward and reverse primers and 1 U of UltraTherm DNA polymerase (TetraLink International, Buffalo, NY). Primers for rDNA were ITS1 (5’- TCCGTAGGTGAACCTGCGG-3’) and ITS4 (5’-TCCTCCGCTTATTGATATG-3’) (White et al. 1990) and β-tubulin primers were Bt-LEV-Lo1 (Fwd 5’- GTGAACTCCATCTCGTCCATA-3’) and Bt-LEV-Up4 (5’- CAAGATCCGTGAGGAGT-3’) (de Jong et al. 2001). Amplification was performed on the Techne TC-3000 thermocycler (Barloworld Scientific, London, England) with the following conditions: 96°C for 5 min followed by 35 cycles of 95°C for 1 min, 55°C for 1 min and 72°C for 1 min with a final extension cycle at 72°C for 7 min. DNA concentrations were estimated on 1% agarose gels using a Low Mass ladder (Invitrogen, Gaithersburg, MD). A minimum of 75 ng and a maximum of 100 ng of PCR-amplified DNA were required for hybridization to macroarrays. 29 2.9.3 DNA macroarray construction Twenty-eight oligonucleotide macroarray probes were designed by Dr. Danielle Hirkala and constructed by Operon Technologies, Huntsville, AL (Table 2). The 5’ end of the probe contained an amino C6 linker used to attach the oligonucleotide onto the Immunodyne ABC nylon membranes (PALL Europe Ltd., Portsmouth, England). The probes were resuspended in 0.5 M sodium bicarbonate buffer (pH 8.0) and diluted to a final concentration of 40 µm. In combination with 0.004% bromophenol blue, the resuspended probes were arranged according to a previously designed template (Figure 2). A 384-pin replicater (V&P Scientific, San Diego, CA) and pin guide (V&P Scientific, San Diego, CA) were used to “stamp” the probes onto the nylon membranes (Figure 2). The membranes were blocked for a minimum of 1 hour with hybridization buffer (6x SSC, 0.1% sarcosine, 0.02% SDS) and 1% skim milk at 55°C. 2.9.4 Hybridization and chemiluminescent detection To denature the DIG-labelled PCR products, 75 – 100 ng of DNA were combined with 5 ml of hybridization buffer and placed in boiling water for 10 min. Membranes were added to the denatured probe solution for overnight hybridization at 55°C. Unbound probes were then washed from the membranes with pre-warmed 6x SSC (0.18% NaCl, 0.088% sodium citrate) and 1% SDS (sodium dodecyl sulphate) for 2 x 40 min at 55°C. Hybridized membranes were pooled and washed for 5 min in washing buffer (0.1 M maleic acid, 0.15 M NaCl [pH 7.5], 0.3% Tween 20) and bound with a 1:25 000 dilution of anti-Digoxigenin-AP (alkaline phosphatase) Fab fragments (Roche Diagnostics GmbH, Mannheim, Germany) combined with washing buffer and 1% skim 30 Table 2. DNA macroarray probe sequences. Name Sequence 5’ to 3’ Description rDNA-ITS UN-H1-up [AminoC6]ACCTACGGAAACCTTGTTACGA Universal to all fungi Bot-all-H1b [AminoC6]TTAGCTTGGTATTGAGTCTATGT Universal to all Botrytis spp. BC-H2d [AminoC6]TATGCTCGCCAGAGAATACCAAA Botrytis cinerea BC-H3d [AminoC6]GCTCGCCAGAGAATACCAAAAC Botrytis cinerea BT-H1d [AminoC6]TATGCTCGCCAGAGAAAACCAAA Botrytis tulipae BT-H2d [AminoC6]CTCGCCAGAGAAAACCAAAACT Botrytis tulipae PE-H2c [AminoC6]CCCGAACTCTGCCTGAAGATT Penicillium expansum PE-H3u [AminoC6]CAGACGACAATCTTCAGGCA Penicillium expansum Mpir-ITS-183H1 [AminoC6]TGGTGTCCTTAAAAATTATTATTAT Mucor piriformis Mpir-ITS-414H1 [AminoC6]AACACCCCACATCTTAAAAATC Mucor piriformis β-tubulin UNbt-1 [AminoC6]CAAGAACATGATGGCTGCTTC Universal to all fungi UNbt-2 [AminoC6]CCAAGAACATGATGGCTGC Universal to all fungi UNbt-3 [AminoC6]TGTTCGACCCCAAGAACATG Universal to all fungi Pex-bt-84-H1-gag [AminoC6]TCCGACGAGACTTTCTGTATC Penicillium expansum Pex-bt-84-H1-gcg [AminoC6]TCCGACGCGACTTTCTGTATC Penicillium expansum TBZ sensitive Pex-bt-84-H1-gtg [AminoC6]TCCGACGTGACTTTCTGTATC Penicillium expansum TBZ resistant Pcom-bt-1 [AminoC6]CCGTCAACATGGTCCCCTT Penicillium commune Psol-bt-1 [AminoC6]TCCCTCGTTTGCACTTCTT Penicillium solitum Psol-bt-2 [AminoC6]CCTTTCCGTCCGTCCACCAGCT Penicillium solitum Paur-bt-1 [AminoC6]CACACCTCTGATATCTTGCTAGG Penicillium aureum Paur-bt-2 [AminoC6]CGATGGACAGTAAGTTCTAATGG Penicillium aurantigriseum Bcin-133-H3 [AminoC6]TTACGATATTTGCATGAGAACCT Botrytis cinerea Bstok-144-H4 [AminoC6]GCATGAGAACCCTGAAGCTC Botrytis mali Bot-95-H1-GAG [AminoC6]AACTCTGACGAGACCTTCTG Botrytis cinerea TBZ sensitive Bot-95-H1-GCG [AminoC6]ACTCTGACGCGACCTTCTG Botrytis cinerea TBZ resistant MucorUN-bt-1 [AminoC6]ACATGGTTCCTTTCCCTCGT Universal to all Mucor spp. MucorUN-bt-2 [AminoC6]AAGGCTTTCTTGCATTGGTA Universal to all Mucor spp. MucorUN-bt-3 [AminoC6]GGTGCTGGTAACTCTTGGGC Universal to all Mucor spp. 31 A. Post-Harvest Macrorray Template rDNA-ITS β-tubulin A B C D E F G H 1 UN-H1-up BT-H1d Mpir-ITS-183H1 UNbt-1 Pex-bt-84- H1gag Psol-bt-1 Bcin-133- H3 MucorUN- bt-1 2 Bot-all-H1- b BT-H2d Mpir-ITS- 414H1 UNbt-2 Pex-bt-84- H1gcg Psol-bt-2 Bstok-144- H4 MucorUN- bt-2 3 BC-H2d PE-H2c UNbt-3 Pex-bt-84-H1gtg Paur-bt-1 Bot-95-H1- gag MucorUN- bt-3 4 BC-H3d PE-H3u Pcom-bt-1 Paur-bt-2 Bot-95-H1-gcg B. Post-Harvest Macrorray Template rDNA-ITS β-tubulin A B C D E F G H 1                 2                 3               4           Figure 2. DNA macroarray pathogen template (A) and the corresponding oligonucleotide probe placements that are amine-linked to the nylon membrane (B). Columns A – C detect rDNA ITS PCR products; columns D – H detect β-tubulin PCR products. Each pathogen (A) corresponds with the dotted template (B) as it would appear on a developed macroarray. Each probe is blotted twice for positive confirmation. Grey boxes indicate no probe present. 32 milk powder for 30 min. Membranes were washed for 2 x 15 min then primed for 15 min in buffer 3 (0.1 M Tris-HCl, 0.1 M NaCl [pH 9.5]). For chemiluminescent detection, membranes were incubated for 5 min in buffer 3 plus 1:2000 dilution of chemiluminescent substrate, CDP-Star (Roche Diagnostics GmbH, Mannheim, Germany). Alkaline phosphatase enzymatically dephosphorylates CDP-Star, creating dioxetane phenolate anion, which decomposes and emits light maximally at 466 nm (Roche Diagnostics GmbH, Mannheim, Germany). Membranes were bound in plastic wrap and immediately exposed to x-ray film (CL-X Posure TM Film, Thermo Scientific, Rockford, IL) for 2 h. After fixing and developing for 1 minute each (Kodak, Rochester, NY), the hybridized DIG-amplified PCR products were captured on x-ray film as a series of dots with varying grey intensities. The x-ray film was scanned and saved as a Jpeg picture file using a BenQ 4300u scanner (Taipei, Taiwan) for computer analysis. The hybridized DNA was quantified using ImageJ 1.30v software (National Institutes of Health, Bethesda, MD). Each hybridization signal was assigned a relative greyscale value that ranged from 0 (completely back) to 255 (completely white). Greyscale values were in the form of two averaged measurements. The background was subtracted from each analyzed array using the following formula: Adjusted Greyscale Value = 127.5 x Original Greyscale Value Average Background The adjusted greyscale values were transformed to DNA concentrations when compared to a standard curve produced from known DNA concentrations. 33 2.10 Biological control of post-harvest fungal pathogens 2.10.1 Inoculation of apples by drenching Bacterial cultures were diluted in 5 L of water to a concentration of 108 CFU/ml and fungal cultures were diluted in 5 L of water to concentrations of 104 spores/ml for P. expansum and B. cinerea and 105 spores/ml for M. piriformis. Inoculation concentrations were previously determined by minimum infectious dose and minimum inhibitory concentration experiments (D. Hirkala, personal communication). Apples were surface sterilised with 90% ethanol and wounded in triplicate using a 3-mm diameter nail embedded in a cork. Wounded apples were placed in a plastic net bag and labelled by tagging. Each bag was placed in a 5-L bacterial suspension for 1 min then subsequently placed in a 5-L fungal suspension for 1 min. Following inoculation, apples were air dried on trays for 10 min then placed in mesh bags (10 apples / bag) and in crates (~100 apples / crate) for commercial storage or in covered plastic bins each with three 0.2-µm membrane filters attached to the lid for semi-commercial storage until lesions formed. In this study, commercial storage refers to controlled atmosphere (CA) storage conditions in a commercial packinghouse, whereas semi-commercial storage refers to storage in a research facility at 1°C in air. 2.10.2 Natural disease incidence Following apple harvest in 2007, natural disease incidence (NDI) was determined by wounding 100 Jonagold apples from Field 3 and 100 Gala apples from Field 12. Fifty of the 100 apples were fumigated with 1-Methylcyclopropene (1-MCP) and 50 were not fumigated with 1-MCP. In 2008, NDI was determined by placing 260 apples, each from Fields 3 (Jonagold), 12, Kiran (Gala) and Reekie (Red Delicious) into controlled 34 atmosphere (CA) storage (1.5% O2 and 1.5% CO2) (n = 100 per orchard) in Naramata, BC or in air storage at 1°C (n = 40 per orchard) at PARC in Summerland, BC. Half of the apples were wounded in triplicate and the remaining half were not wounded. 1-MCP treatments were not used in 2008 harvest experiments. Disease incidence (% apples infected) and lesion diameters (mm) were measured after 3 and 6 months for the 2007 harvest, after 4.5 months of sealed CA storage and after 1, 2, 4 and 6 months at 1°C in air storage for apples harvested in 2008. 2.10.3 Natural disease incidence with bacterial antagonists To measure bacterial antagonist efficacy against natural disease incidence for harvest 2007, 120 Jonagold apples from Field 3 and 120 Gala apples from Field 12 were wounded in triplicate and inoculated with one of five bacterial antagonists – isolates 1100-6, 1-112, 2-28, 4-6 or 6-25- and a non-inoculated control (10 apples / antagonist). Sixty of the 120 apples were fumigated with 1-MCP and the remaining 60 apples were not fumigated. Following harvest in 2008, 240 apples from each orchard, Field 3, 12, Kiran and Reekie, were inoculated as in section 2.10.1. One hundred and twenty apples from each orchard were stored in CA (60 wounded and 60 not wounded) and 120 apples were stored at 1°C in air (60 wounded and 60 not wounded). Disease incidence and lesion diameters were measured after 3 months in storage for apples harvested in 2007. For apples harvested in 2008, disease incidence and lesion diameters were measured after 4.5 months of sealed CA storage and after 1, 2, 4 and 6 months at 1°C in air storage. The efficacy of the biological control agents was compared to apples that had been wounded and not inoculated with bacteria. 35 2.10.4 Biological control of post-harvest pathogens For apples harvested in 2007, disease inhibition was measured by inoculating 360 Jonagold apples from Field 3 and 360 Gala apples from Field 12 with one of three fungal pathogens, P. expansum, B. cinerea or M. piriformis and one of five bacterial control isolates, 1100-6, 1-112, 2-28, 4-6, 6-25 and a non-inoculated control. One hundred and eighty of the 360 apples were fumigated with 1-MCP. For apples harvested in 2008, 360 apples each from Fields 3 and 12 and the Kiran and Reekie orchards were inoculated as above. In 2008, 1-MCP was not used, but 180 apples per orchard were wounded and 180 apples per orchard were non-wounded. From each site, there were 10 apples / antagonist + 1 non-inoculated control / pathogen. Disease incidence and lesion diameters were measured after 3 months of storage for apples harvested in 2007 and after 1, 2, 4 and 6 months at 1°C in air storage for apples harvested in 2008. Pathogens were not permitted in commercial CA storage. 2.10.5 Bacterial survival on apple Concentrations of 4-6-gfp were monitored by two experimental techniques- colony counts via dilution plating and direct scanning using the POLARstar Omega microplate reader (BMG Labtech, Germany) as in Etebarian et al. (2005). A standard curve was developed that described the relationship between colony forming units (CFU/ml) and relative fluorescent units (RFU) acquired by direct scanning. The bacterial inoculum was prepared in ½ TSB with 50 µg/ml of kan and 30 µg/ml of gen and grown at 28°C for 24 h. P. expansum spores were harvested from 7-d- old ½ PDA plates. Final inoculation concentrations for 4-6-gfp and P. expansum were 108 CFU/ml and 104 spores/ml, respectively. Apples were prepared as in section 2.9.1 36 and inoculated by pipette (30 µl) with the following treatments: 1) non-inoculated control; 2) 4-6-gfp alone; 3) P. expansum alone and; 4) 4-6-gfp + P. expansum. Isolate 4- 6-gfp was quantified by aseptically removing an apple core (5 mm in diameter x 5 mm deep) from two of the three apple wounds. Both cores were macerated with a sterile rod. One core, to be used for dilution plating, was placed in 1.1 ml of PBS and plated on Pseudomonas F agar (20 g/L peptone, 1.5 g/L K2HPO4, 1.5 g/L anhydrous MgSO4, 10 g/L agar [pH 7.2]) supplemented with 50 µg/ml kan and 30 µg/ml gen. The second core, to be used in direct scanning, was placed in 300 µl of PBS. Two hundred microlitres of the tissue slurry were added to a well in a black Nunc 96-well microplate (Thermo Fisher Scientific, Rochester, NY). Fluorescent readings were obtained using a 405 nm excitation filter and a 510 nm emission filter and a gain of 1200. The relative fluorescent units (RFU) of PBS and non-inoculated control were subtracted from the treatments to correct for background and apple fluorescence, respectively. Bacterial concentrations were determined approximately every 10 days for a total of 55 d. 2.11 Analysis In all biocontrol experiments, disease was measured by disease incidence (percent apple infection) and disease severity (average lesion diameter in mm). From these two values, data were converted to infection severity (IS) (Spotts et al. 1999): Infection Severity (IS) = % Apple Infection * Average Lesion Diameter in mm 100% In order to correct for variance heterogeneity, infection severity values were log10 (n+0.5) transformed. Differences in means were analyzed by either a one-way or two- 37 way analysis of variance (ANOVA). p-values less than 0.05 were considered statistically significant. If the 2-way ANOVA interaction term was significant, the main effects were separated and a one-way ANOVA was used for subsequent analysis. Differences among means were determined by the least significant difference (LSD) test. The Pearson Product Moment Correlation Coefficient (r) was used to measure the linear association of two independent variables. Statistical analyses were conducted using CoStat Statistics Software (CoHort Software, Monterey, CA). 38 CHAPTER 3: RESULTS 3.1 Validation and field testing of DNA macroarrays 3.1.1 Quantification of post-harvest pathogens Macroarray construction for the identification of post-harvest pathogens was completed by Dr. Danielle Hirkala and was based upon the prototype proposed in Sholberg et al. (2005a). The array displayed high specificity, no cross-reactivity and detected as little as 0.1 ng of pure culture fungal DNA extracted from P. expansum and B. cinerea and up to 0.01 ng of pure culture of M. piriformis (Hirkala, unpublished). For the purpose of this study, three macroarray probes were selected for pathogen identification: PE-H3u for P. expansum detection, BC-H2d for B. cinerea detection and Mpir-ITS-414H1 for M. piriformis detection. Each of the three selected probes is comprised of sequences in the rDNA ITS region and is located on the left hand side of the macroarray template (Figure 2). These probes were selected because of their overall higher sensitivity compared to other probes. This decision was based upon all 2007 greyscale values obtained from environmental field data. Standard curves were determined for each of the three probes (Figures 3, 4 and 5). To test the quantitative properties of the amino-linked oligonucleotides, 0 – 100 ng of pure culture fungal DNA was hybridized to the macroarray. Each detector oligonucleotide displayed varying levels of hybridization signal intensity; however, a linear relationship did exist between DNA concentration and hybridization signal strength. The linear range of the probe PE-H3u ((♦) r = -0.974; (●) r = -0.962) was between 0 and 3 ng (Figure 3), whereas the linear range for the probe BC-H2d 39 y = -16.67x + 134.5 y = -41.147x + 123.99 0 20 40 60 80 100 120 140 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Gr ey sca le Va lue DNA Concentration (ng) Figure 3. Quantitative assessment of Penicillium expansum probe PE-H3u illustrating the relationship between greyscale value and DNA concentration (ng). (♦) Represents standard curve used for samples collected in 2007 (r = -0.974); (●) Represents standard curve used for samples collected in 2008 (r = -0.962). Error bars represent standard error of the mean. 40 y = -8.5966x + 126.02 y = -8.1112x + 120.26 0 20 40 60 80 100 120 140 0 1 2 3 4 5 6 7 8 9 10 11 12 Gr ey sca le Va lue DNA Concentration (ng) Figure 4. Quantitative assessment of Botrytis cinerea probe BC-H2d illustrating the relationship between greyscale value and DNA concentration (ng). (♦) Represents standard curve used for samples collected in 2007 (r = -0.960); (●) Represents standard curve used for samples collected in 2008 (r = -0.977). Error bars represent standard error of the mean. 41 y = -148.38x + 129.06 y = -181.28x + 116.61 0 20 40 60 80 100 120 140 0 0.1 0.2 0.3 0.4 0.5 Gr ey sca le Va lue DNA Concentration (ng) Figure 5. Quantitative assessment of Mucor piriformis probe Mpir-ITS-414H1 illustrating the relationship between greyscale value and DNA concentration (ng). (♦) Represents standard curve used for samples collected in 2007 (r = -0.970); (●) Represents standard curve used for samples collected in 2008 (r = -0.990). Error bars represent standard error of the mean. 42 ((♦) r = -0.960; (●) r = -0.977) was between 0 and 12 ng (Figure 4) and the probe Mpir- ITS-414H1 ((♦) r = -0.970; (●) r = -0.990) was between 0 and 0.5 ng (Figure 5). Above these ranges, the curves deviated from linearity as the oligonucleotides reached their saturation level. Throughout the duration of this study, the intensity of the hybridization signals decreased over time which led to the construction of two separate quantification curves for each of the three pathogens (Figures 3, 4 and 5). The two standard curves created for P. expansum DNA quantification (ng) were constructed in August 2008 (♦) and August 2009 (●) (Figure 3). Positive hybridization signals that were present in samples assessed in 2008 were quantified using the formula y = -41.147x + 123.99 (♦). Positive hybridization signals that were present in samples assessed in 2009 were quantified using the formula y = -16.67x + 134.5 (●). The two standard curves created for B. cinerea DNA quantification (ng) were constructed in August 2008 (♦) and October 2009 (●) (Figure 4). Positive hybridization signals that were present in samples assessed in 2008 were quantified using the formula y = -8.1112x + 120.26 (♦). Positive hybridization signals that were present in samples assessed in 2009 were quantified using the formula y = -8.5966x + 126.02 (●). The two standard curves created for M. piriformis DNA quantification (ng) were constructed in October 2008 (♦) and August 2009 (●) (Figure 5). Positive hybridization signals that were present in samples assessed in 2008 were quantified using the formula y = -181.28x + 116.61 (♦). Positive hybridization signals that were present in samples assessed in 2009 were quantified using the formula y = -148.38x + 129.06 (●). 43 3.1.2 Field application of macroarray technology The DNA macroarray was used in two years of field monitoring studies to verify airborne and plant tissue pathogen spore loads in four orchards in the Okanagan Valley, BC: Field 3 in Summerland, BC (Figure 6); Field 12 in Summerland, BC (Figure 7); Kiran orchard in Kelowna, BC (Figure 8); Reekie orchard in Kelowna, BC (Figure 9). All three probes had positive hybridization signals throughout the growing season with high variability between years and sample types. In Field 3 (Figure 6), all three pathogens, P. expansum (A. 2007, D. 2008), B. cinerea (B. 2007, E. 2008) and M. piriformis (C. 2007, F. 2008) were detected most frequently in the early and late growing seasons. In Field 12 (Figure 7), P. expansum was detected at the beginning and middle of the growing season in 2007 (A) and throughout the growing season in 2008 (D). Pathogen levels were much higher in 2008 (D) than in 2007 (A). B. cinerea was detected in the middle of the growing season in 2007 (B) and the beginning and the end of the growing season in 2008 (B). B. cinerea was not detected on plant tissue samples in either 2007 or 2008 (B, E). M. piriformis was detected at the beginning and middle of the growing season in 2007 (C) and at the beginning and the end of the growing season in 2008 (F). In the Kiran orchard (Figure 8), P. expansum (A. 2007, D. 2008), B. cinerea (B. 2007, E. 2008) and M. piriformis (C. 2007, F. 2008) were detected throughout the growing season in both 2007 and 2008. DNA concentrations for P. expansum and B. cinerea were higher in 2008 than in 2007. In the Reekie orchard (Figure 9), P. expansum was detected in the middle of the growing season in 2007 (A) and throughout the growing season in 2008 (D). B. cinerea was not detected in 2007 (B) and was detected throughout the growing season in 2008 (E). M. piriformis was detected in the middle of 44 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 0 1 2 3 4 0 25 50 75 100 125 150 F. E. C. B. A. D. DN A ( ng ) Days After Blossom 2007 2008 Figure 6. Concentrations of P. expansum (A,D), B. cinerea (B,E) and M. piriformis (C,F) from aerial (solid line) and plant tissue samples (dashed line) collected from Field 3 (Summerland, BC) in 2007 (A – C) and 2008 (D – F) using DNA macroarrays and identified by post-harvest pathogen rDNA ITS macroarray probes, PE-H3u (♦), BC-H2d (●) and Mpir-ITS-414H1 (■). 45 0 1 2 3 4 5 6 7 0 25 50 75 100 125 150 0 1 2 3 4 5 6 7 0 25 50 75 100 125 150 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 0 1 2 3 4 5 6 7 0 25 50 75 100 125 150 0 1 2 3 4 5 6 7 0 25 50 75 100 125 150 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 F. E. C. B. A. D. DN A ( ng ) Days After Blossom 2007 2008 Figure 7. Concentrations of P. expansum (A,D), B. cinerea (B,E) and M. piriformis (C,F) from aerial (solid line) and plant tissue samples (dashed line) collected from Field 12 (Summerland, BC) in 2007 (A – C) and 2008 (D – F) using DNA macroarrays and identified by post-harvest pathogen rDNA ITS macroarray probes, PE-H3u (♦), BC-H2d (●) and Mpir-ITS-414H1 (■). 46 0 2 4 6 8 10 0 25 50 75 100 125 150 0 2 4 6 8 10 0 25 50 75 100 125 150 0 2 4 6 8 10 0 25 50 75 100 125 150 175 0 2 4 6 8 10 0 25 50 75 100 125 150 175 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 175 F. E.B. A. D. DN A ( ng ) Days After Blossom 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 C. 2007 2008 Figure 8. Concentrations of P. expansum (A,D), B. cinerea (B,E) and M. piriformis (C,F) from aerial (solid line) and plant tissue samples (dashed line) collected from the Kiran orchard (Kelowna, BC) in 2007 (A – C) and 2008 (D – F) using DNA macroarrays and identified by post-harvest pathogen rDNA ITS macroarray probes, PE-H3u (♦), BC-H2d (●) and Mpir-ITS-414H1 (■). 47 0 2 4 6 8 10 12 0 25 50 75 100 125 150 0 2 4 6 8 10 12 0 25 50 75 100 125 150 0 2 4 6 8 10 12 0 25 50 75 100 125 150 175 0 2 4 6 8 10 12 0 25 50 75 100 125 150 175 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 175 F. E.B. A. D. DN A ( ng ) Days After Blossom 0.0 0.2 0.4 0.6 0.8 1.0 0 25 50 75 100 125 150 C. 2007 2008 Figure 9. Concentrations of P. expansum (A,D), B. cinerea (B,E) and M. piriformis (C,F) from aerial (solid line) and plant tissue samples (dashed line) collected from the Reekie orchard (Kelowna, BC) in 2007 (A – C) and 2008 (D – F) using DNA macroarrays and identified by post-harvest pathogen rDNA ITS macroarray probes, PE-H3u (♦), BC-H2d (●) and Mpir-ITS-414H1 (■). 48 the growing season in 2007 (C) and throughout the growing season in 2008 (F). All pathogen levels were higher in 2008 than in 2007. Due to the observed variability, the data were arranged in an alternate way in order to determine if there were common patterns between pathogens and fields (Table 3). To determine what time in the growing season each of the three main post-harvest pathogens was primarily detected, the season was divided into three sections: 1) early (~1 – 50 days after blossom); 2) mid (~50 – 100 days after blossom) and; 3) late (~100 – 150+ days after blossom) growing season (Table 3). Aerial and plant tissue samples were analyzed separately and in combination by determining percent pathogen detection within either the early, mid or late growing seasons. According to the tabulated data in Table 3, in 2007 and 2008, P. expansum was detected in each of the three growing season segments but differed by field and year. From DNA isolated from aerial samples collected in 2007, there was a significant effect of growing season segment on % P. expansum detected (p = 0.06) (Appendix, Table A1); the % P. expansum detected in the early season was significantly higher than that in the late growing season (Table 3). There was no effect of growing season segment on % P. expansum detected in aerial samples in 2008 (Table 3; Appendix, Table A4). From DNA isolated from plant tissue samples in 2007 there was no significant effect of the growing season segment on the % P. expansum detected (Table 3; Appendix, Table A2), but in 2008, there was a significant effect (p = 0.053) (Appendix, Table A5). The % P. expansum detected in the mid-growing season was significantly higher than that in the late-growing season (Table 3). 49 In 2007, there was no effect of the growing season segment on the % B. cinerea detected in DNA isolated from either aerial or plant tissue samples (Table 3; Appendix, Tables A1 and A2). In 2008, from DNA isolated from plant tissue samples, the % B. cinerea detected in the mid-growing season was significantly higher than that in the early and late-growing seasons (Table 3; Appendix, Table A5). Overall, B. cinerea was detected at much lower levels than either P. expansum or M. piriformis. M. piriformis was detected in each of the three growing season segments but also differed by field and year (Table 3). From DNA isolated from plant tissue samples in 2007, the % M. piriformis detected was significantly higher in the mid-growing season than that in the early or late-growing seasons (Table 3; Appendix, Table A2). From DNA isolated from plant tissue samples in 2008, the % M. piriformis detected was significantly higher in the mid-growing season than that in the late-growing season (p = 0.06) (Table; Appendix, Table A5). In both 2007 and 2008, when aerial and plant tissue DNA samples were analyzed together, there was no effect of growing season segment (Table 3; Appendix, Tables A3 and A6). In 2007 and 2008 combined, the selected probes detected P. expansum (27.4 ± 3.4%) and M. piriformis (19.2 ± 1.4%) most frequently, followed by B. cinerea (6.2 ± 1.4%) (Table 3; Appendix, Table A7). In addition, P. expansum was detected at the highest concentrations, followed by B. cinerea and then M. piriformis. Of the samples collected in 2007, the Kiran orchard had the highest percent detection (81.6%) followed by Field 12 (50%), the Reekie orchard (33.3%) and Field 3 (30.3%). In 2008, the Reekie orchard had the highest percent detection (80.4%) followed by Field 12 (71.8%), Kiran (46.2%) and Field 3 (28.9%) (Table 3). 50 Table 3. DNA macroarray detection (%) summary for P. expansum, B. cinerea and M. piriformis from four apple orchards throughout the 2007 and 2008 growing seasons. Growing season was divided into 3 sampling periods: 1) early-growing season (~ 1-50 days after blossom); 2) mid-growing season (~ 50-100 days after blossom) and; 3) late-growing season (~ 100-150+ days after blossom). Numbers in ( ) indicate standard error of the mean. % pathogen detection within sampling perioda P. expansum B. cinerea M. piriformis % detection per orchard Field Sample type 1 2 3 1 2 3 1 2 3 2007 Growing Season Field 3 aerial 28.6 0.0 0.0 7.1 0.0 0.0 7.1 0.0 16.7 plant tissue 0.0 0.0 100.0 0.0 0.0 0.0 0.0 33.3 0.0 Field ave 25.0b 0.0 25.0 6.3 0.0 0.0 6.3 11.1 12.5 30.3 Field 12 aerial 37.5 33.3 0.0 0.0 33.3 0.0 12.5 0.0 0.0 plant tissue 50.0 33.3 50.0 0.0 0.0 0.0 0.0 66.7 0.0 Field ave 38.9 33.3 14.3 0.0 22.2 0.0 11.1 22.2 0.0 50.0 Kiran aerial 66.7 0.0 16.7 11.1 14.3 0.0 66.7 28.6 50.0 plant tissue 50.0 57.1 0.0 0.0 0.0 20.0 0.0 42.9 20.0 Field ave 61.5 28.6 9.1 7.7 7.1 9.1 46.2 35.7 36.4 81.6 Reekie aerial ND ND ND ND ND ND ND ND ND plant tissue 0.0 25.0 0.0 0.0 0.0 0.0 0.0 50.0 0.0 Field ave 0.0 25.0 0.0 0.0 0.0 0.0 0.0 50.0 0.0 33.3 Ave % detection per sampling period (irod)c 44.2ah (11.5) 11.1ab (11.1) 5.6b (5.6) 6.1a (3.3) 15.9a (9.7) 0.0a (0.0) 28.8a (19.0) 9.5a (9.5) 22.2a (14.7) Ave % detection per sampling period (pt)d 25.0a (14.4) 28.9a (11.8) 37.5a (23.9) 0.0a (0.0) 0.0a (0.0) 5.0a (5.0) 0.0b (0.0) 48.2a (7.0) 5.0b (5.0) 2007 ave % detection per sampling period (irod + pt)e 31.4a (12.9) 21.7a (7.4) 12.1a (5.2) 3.5a (2.0) 7.3a (5.2) 2.3a (2.3) 15.9a (10.3) 29.8a (8.4) 12.2a (8.6) 2007 ave % detection per pathogenf 24.0a (5.6) 4.2b (1.7) 20.6ab (13.8) 51 % pathogen detection within sampling period P. expansum B. cinerea M. piriformis % detection per orchard Field Sample type 1 2 3 1 2 3 1 2 3 2008 Growing Season Field 3 aerial 0.0 18.2 33.3 0.0 9.1 66.7 27.8 0.0 0.0 plant tissue 0.0 ND 0.0 0.0 ND 0.0 0.0 ND 0.0 Field ave 0.0 18.2 20.0 0.0 9.1 40.0 22.7 0.0 0.0 28.9 Field 12 aerial 57.9 54.5 100.0 5.3 9.1 0.0 21.1 0.0 33.3 plant tissue 25.0 ND 0.0 0.0 ND 0.0 0.0 ND 0.0 Field ave 52.2 54.5 60.0 4.3 9.1 0.0 17.4 0.0 20.0 71.8 Kiran aerial 5.9 25.0 18.2 0.0 12.5 18.2 5.9 29.2 18.2 plant tissue 38.9 20.8 10.0 5.6 12.5 0.0 27.8 8.3 0.0 Field ave 22.9 22.9 14.3 2.9 12.5 9.5 17.1 18.8 9.5 46.2 Reekie aerial 73.3 45.0 9.1 40.0 5.0 9.1 40.0 30.0 9.1 plant tissue 50.0 37.5 9.1 6.3 8.3 0.0 37.5 37.5 0.0 Field ave 61.3 40.9 9.1 22.6 6.8 4.5 38.7 34.1 4.5 80.4 Ave % detection per sampling period (irod)c 34.3ah (18.4) 35.7a (8.5) 40.2a (20.6) 11.3a (9.6) 8.9a (1.5) 23.5a (14.9) 23.7a (7.1) 14.8a (8.5) 15.2a (7.1) Ave % detection per sampling period (pt)d 28.5ab (10.8) 29.2a (8.3) 4.8b (2.8) 3.0b (1.7) 10.4a (2.1) 0.0b (0.0) 16.3ab (9.6) 22.2a (14.6) 0.0b (0.0) 2008 ave % detection per sampling period (irod + pt)e 34.1a (14.0) 34.1a (8.4) 25.8a (11.6) 7.4a (5.1) 9.4a (1.2) 13.5a (9.0) 24.0a (5.1) 13.2a (8.2) 8.5a (4.3) 2008 ave % detection per pathogen f 30.8a (10.2) 8.3b (1.3) 17.8ab (3.8) 2007/08 ave % detection per pathogeng 27.4a (3.4) 6.2b (1.4) 19.2a (1.4) 52 a. % pathogen detection: # of times a pathogen was detected within a select sampling period * 100% # of samples taken within a select sampling period b. Field averages were obtained by calculating the average % pathogen detection between both aerial and plant tissue samples c. Ave % detection per sampling period (irod): irod = aerial sample; values were determined by taking the average of the % detection values per sampling period per pathogen within each orchard in aerial samples only d. Ave % detection per sampling period (pt): pt = plant tissue sample; values were determined by taking the average of the % detection values per sampling period per pathogen within each orchard in plant tissue samples only e. Ave % detection per sampling period: values were determined by taking the average of the % detection values per sampling period per pathogen within each orchard in aerial and plant tissue samples combined f. Ave % detection per pathogen: values were determined by taking the average of the % pathogen detection values within all three sampling periods within each orchard with aerial and plant tissue samples combined g. 2007/08 ave % detection per pathogen: values were determined by taking the average of the values obtained in Ave % detection per pathogen in both 2007 and 2008 h. Means followed by different letters within a pathogen and within a row are significantly different (p ≤ 0.06) according to the least significant difference (LSD) test. 53 Fruit washings conducted after harvest in 2007 showed the presence of B. cinerea in Jonagold apples grown in Field 3 and the presence of P. expansum, B. cinera and M. piriformis in Gala apples grown in Field 12. Fruit washings conducted after harvest in 2008 showed the presence of P. expansum in Gala apples grown in the Kiran orchard (Table 4). Table 4. Detection of P. expansum, B. cinerea and M. piriformis from fruit washings collected from Jonagold apples from Field 3, Gala apples from Field 12 and the Kiran orchard and Red Delicious apples from the Reekie orchard in 2007 and 2008 using DNA macroarrays. DNA detected (ng/20 apples) Field P. expansum B. cinerea M. piriformis Harvest 2007 Field 3 NDa 0.48 ND Field 12 0.78 0.66 0.05 Harvest 2008 Field 3 ND ND ND Field 12 ND ND ND Kiran 1.41 ND ND Reekie ND ND ND a. ND = not detected 3.2 Natural disease incidence 3.2.1 Harvest 2007 The effect of incubation period (3 and 6 months) and 1-MCP use (1-MCP+/-) on infection severity (IS) was analyzed using a two-way ANOVA. The interaction terms for apples from both Field 3 and 12 were not significant (Appendix, Tables A8 and A9). In both Field 3 and 12, there was a greater IS at six months than at three months in 1°C storage. However, only in Field 3 was there a significant effect of 1-MCP application; 1- MCP reduced IS by 48.2% in comparison to the non-fumigated control (Tables 5 and 6). 54 Table 5. Natural disease incidence indicated by infection severitya in Jonagold apples grown in Field 3 (harvest 2007), treated or not treated with 1-MCP (1-MCP+/-) and stored for 3 or 6 months at 1°C in air. Field 3 Field 3 Incubation Period 3 months 6 months Overall Mean 1-MCP + 5.51 9.55 7.53 bc 1-MCP - 9.09 22.12 15.61 a Overall Mean 7.30 bb 15.83 a a. Infection severity (IS) = percent apple infection x average lesion diameter (mm)/100 (Spotts et al. 1999). b. Incubation period means with different letters are different at significance levels p ≤0.05 according to least significant difference (LSD) test. c. 1-MCP means with different letters are different at significance levels p ≤0.05 LSD test. Table 6. Natural disease incidence indicated by infection severitya in Gala apples grown in Field 12 (harvest 2007), treated or not treated with 1-MCP (1-MCP+/-) and stored for 3 or 6 months at 1°C in air Field 12 Field 12 3 months 6 months Overall Mean 1-MCP + 7.64 30.28 18.95 ac 1-MCP - 11.78 22.29 16.56 a Overall Mean 9.90 bb 26.28 a a. Infection severity (IS) = percent apple infection x average lesion diameter (mm)/100 (Spotts et al. 1999). b. Incubation period means with different letters are different at significance levels p ≤0.05 according to least significant difference (LSD) test. c. 1-MCP means with different letters are different at significance levels p ≤0.05 LSD test. 3.2.2 Harvest 2008 The effects of incubation period (1, 2, 4 and 6 months) on infection severity in each of the four orchards were analyzed using a one-way ANOVA. Results indicated that in all fields, there was a significant increase in IS from one to six months (Table 7; Appendix, Table A10). A one-way ANOVA also revealed that at two months incubation, Jonagold apples grown in Field 3 had significantly greater infection severities than in all 55 other orchards. At four and six months of incubation, Jonagold apples grown in Field 3 and Gala apples grown in Field 12 had significantly greater infection severities than in the commercial orchard apples, Gala grown in the Kiran orchard and Red Delicious grown in the Reekie orchard (Appendix, Table A11). Table 7. Natural disease incidence indicated by infection severitya in Jonagold apples grown in Field 3, Gala apples grown in Field 12 and the Kiran orchard and Red Delicious apples grown in the Reekie orchard (harvest 2008) and stored for 1, 2, 4 or 6 months at 1°C in air. Apple Variety Field 3 Jonagold Field 12 Gala Kiran Gala Reekie Red Delicious 1 month 0.00 dA 0.00 dA 0.36 bA 0.00 cAc 2 months 1.08 cA 0.54 cB 0.01 bB 0.00 cB 4 months 20.72 bA 19.95 bA 0.08 bB 2.28 bB 6 months 48.12 abA 46.58 aA 2.06 aB 10.88 aB a. Infection severity (IS) = percent apple infection x average lesion diameter (mm)/100 (Spotts et al. 1999). b. Incubation period means with different letters within columns (lowercase) are different at significance levels at p ≤0.05 according to least significant difference (LSD) test. c. Incubation period means with different letters within rows (uppercase) are different at significance levels at p ≤0.05 according to the LSD test. In addition to storing apples at 1°C in air, fruit were also stored in controlled atmosphere (CA) for 4.5 months. Jonagold apples grown in Field 3 had the highest infection severity with significantly greater disease levels compared to Gala apples grown in Fields 12 and Kiran and Red Delicious apples grown in the Reekie orchard (Figure 10; Appendix, Table A12). The high infection severity found in Jonagold apples grown in Field 3 appeared to be due primarily to P. expansum (based on phenotypic observation). 56 a b b b 0 2 4 6 8 10 12 Field 3 Jonagold Field 12 Gala Kiran Gala Reekie Red Delicious Av era ge In fec tio n S ev er ity (IS ) Field Figure 10. Natural disease incidence in Jonagold apples grown in Field 3, Gala apples grown in Field 12 and the Kiran Orchard and Red Delicious apples grown in the Reekie orchard and stored for 4.5 months in controlled atmosphere (CA) in Naramata, BC. IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 57 3.3 Growing season detection and natural disease incidence In this study, there was no correlation between % total pathogen detection (P. expansum, B. cinerea and M. piriformis) detected in the field throughout the entire growing season in both 2007 and 2008 combined (Table 3) and NDI after storage (r = - 0.11; p = 0.8349). With respect to Fields 3 and 12 in 2007 and Fields 3 and 12 and the Kiran and Reekie orchards in 2008, there was no correlation between % pathogens detected in the early-growing season and their corresponding NDIs (r = -0.29; p = 0.5811) and % pathogens detected in the mid-growing season and their corresponding NDIs (r = -0.28; p = 0.5865). However, there was a positive correlation between % pathogens detected in the late-growing season and NDI (r = 0.79; p = 0.0612). When the aerial samples were analyzed separately from the plant tissue samples, in the 2007 and 2008 data combined, there was no correlation between % pathogens detected in the early (r = -0.57; p = 0.2234) and mid-growing seasons and their corresponding NDIs (r = -0.36; p = 0.4838). However, there was a positive correlation between % pathogens detected in late-growing season and NDI (r = 0.74; p = 0.0903). When the plant tissue samples were separated from the aerial samples, in 2007 and 2008 combined, there was no correlation between % pathogens detected in either the mid (r = - 0.22; p = 0.7810) or late-growing seasons (r = -0.21; p = 0.6826) and their corresponding NDIs. However, there was a negative correlation between % pathogens detected in the early-growing season and NDI (r = -0.79; p = 0.0623). 58 3.4 Antagonist efficacy, harvest 2007 3.4.1 Natural disease incidence and biocontrol inoculations The effects of biological control treatment (no biocontrol, 1100-6, 1-112, 2-28, 4- 6 or 6-25) and 1-MCP use (1-MCP+/-) were analyzed using a two-way ANOVA. For Jonagold apples grown in Field 3 and stored at 1°C in air for 3 months, there was no significant effect of the biological controls, 1-MCP use or their interaction (Appendix, Table A13). Results obtained with inoculated Gala apples from Field 12 indicated a significant interaction with both main effects, biological control treatment and 1-MCP use (Appendix, Table A14). Because the interaction term was significant, treatments were separated, followed by a one-way analysis of variance. 1-MCP significantly decreased IS in apples with no biological control treatment and apples that were treated with bacterial isolates 1100-6 and 2-28 (Figure 11; Appendix, Table A15). There were significant reductions in IS means in the absence of 1-MCP in apples treated with bacterial isolates 1-112, 4-6 and 6-25 in comparison to the non-inoculated control (Figure 11). In 1-MCP- treated apples, there was no significant effect of treatments compared to the non-inoculated control (Figure 11; Appendix, Table A16). 3.4.2 Post-harvest pathogen and biocontrol inoculations Infection severity (IS) was determined with apples grown in Fields 3 and 12 and inoculated with one of three pathogens, P. expansum, B. cinerea or M. piriformis and one of five bacterial control isolates and the data were subjected to a two-way ANOVA. IS means for Jonagold apples from Field 3 and inoculated with P. expansum 59 B ab A a B ab A a A bc A d B ab A ab A a A cAab A bc 0 10 20 30 40 50 60 70 MCP+ MCP- Av er ag e I nf ec tio n S ev er ity (IS ) MCP Treatment no biocontrol 1100-6 1-112 2-28 4-6 6-25 Figure 11. Effect of 1-MCP (capital letters) and biological control isolates (lowercase letters) on natural disease incidence in Gala apples grown in Field 12 (harvest 2007) and stored for 3 months in air at 1°C. IS means with different letters within treatments are different at significance levels p ≤0.05 according to LSD test. 1-MCP effect comparisons (capital letters) can be made between 1-MCP treatment groups. Biological control activity comparisons (lowercase letters) can be made within 1-MCP treatment groups. 60 exhibited a significant effect of biological control treatment and of the interaction between the main effects, biological control treatment and 1-MCP use (Appendix, Table A17). The resulting one-way ANOVA showed that the application of 1-MCP significantly decreased IS means in apples inoculated with isolate 1100-6 and increased IS in apples inoculated with isolate 6-25 (Figure 12a; Appendix, Table A18). In comparison to the non-inoculated control, isolates 1100-6, 1-112, 4-6 and 6-25 provided significant control of P. expansum in both the presence and absence of 1-MCP (Figure 12a; Appendix, Table A19). For B. cinerea-treated apples from Field 3, the two-way ANOVA revealed a significant effect of biological control treatment and 1-MCP and of their interaction (Appendix, Table A20). The one-way ANOVAs revealed a significant increase in IS in 1-MCP-treated apples that were not inoculated with a biocontrol and apples inoculated with isolates 2-28 and 4-6 (Figure 12b; Appendix, Table A21). In comparison to a non- inoculated control, isolates 1100-6, 1-112 and 6-25 provided significant control of B. cinerea in the presence of 1-MCP. There was no significant reduction in IS in apples in the absence of 1-MCP (Figure 12b; Appendix, Table A22). For M. piriformis-treated apples grown in Field 3, there was a significant effect of the biological control treatment and of the interaction between the main effects, biological control treatment and 1-MCP use (Appendix, Table A23). The one-way ANOVAs revealed a significant increase in IS in non-1-MCP treated apples inoculated with isolate 4-6 (Figure 12c; Appendix, Table A24). There was no effect of inoculation with any of the biological control isolates in the presence of 1-MCP; however, there was a significant increase in IS in non-1-MCP-treated apples inoculated with isolate 4-6 when 61 A a A a B c A b A d A c A ab A ab A d A c A b B c 0 10 20 30 40 50 60 70 MCP+ MCP- Av era ge In fec tio n S ev eri ty (IS ) MCP Treatment A a B a A b A a A b A a A a B a A a B a A b A a 0 10 20 30 40 50 60 70 MCP+ MCP- Av era ge In fec tio n S ev eri ty (IS ) MCP Treatment A a A bc A a A bA a A bc A a A c B a A a A a A c 0 10 20 30 40 50 60 70 MCP+ MCP- Av era ge In fec tio n S ev eri ty (IS ) MCP Treatment A. P. expansum C. M. piriformis B. B. cinerea no biocontrol 1100-6 1-112 2-28 4-6 6-25 Figure 12. Effect of 1-MCP (capital letters) and biological control isolates (lowercase letters) on infection severity in Jonagold apples from Field 3 (harvest 2007), inoculated with P. expansum (A), B. cinerea (B) and M. piriformis (C) and stored for 3 months at 1°C in air. IS means with different letters within treatments are different at significance levels p ≤0.05 according to LSD test. 1-MCP effect comparisons (capital letters) can be made between 1-MCP treatment groups. Biological control activity comparisons (lowercase letters) can be made within 1-MCP treatment groups. 62 compared to 1-MCP-treated apples (Figure 12c; Appendix, Table A25). For Gala apples from Field 12 treated with P. expansum and each of five biological control isolates, there was a significant effect of 1-MCP use and no significant interaction between the main effects, biological control treatment and 1-MCP use (Appendix, Table A26). 1-MCP significantly increased the mean IS (48.16 ± 0.50) compared to that obtained in the absence of 1-MCP (45.69 ± 0.68). There was no overall significant effect of the biological control isolates (Appendix, Table A26). There were no data for Gala apples from Field 12 inoculated with B. cinerea and the five biological control isolates as there was complete apple decay after 3 months of storage at 1°C. For Gala apples from Field 12 treated with M. piriformis and each of the five biological controls, there was a significant interaction between the main effects, biological control treatment and 1-MCP use (Appendix, Table A27). The one-way ANOVAs revealed a significant decrease in IS in 1-MCP-treated apples inoculated with isolate 1-112 compared to non-1-MCP-treated apples and a significant increase in IS in non-1-MCP-treated apples inoculated with isolate 2-28 when compared to 1-MCP-treated apples (Figure 13; Appendix, Table A28). In the presence of 1-MCP, isolate 2-28 exhibited significant biological control activity compared to the non-inoculated control. There was no significant reduction in IS in non-1-MCP treated apples when compared to a non-inoculated control (Figure 13; Appendix, Table A29). 3.5 Antagonist efficacy, harvest 2008 For all 2008 biological control experiments, apples were wounded and non- wounded. Non-wounded apples had a very low frequency of infection with low 63 A ab A ab A a B bB c A a A ab A abA bc A b 0 10 20 30 40 50 60 70 MCP+ MCP- Av era ge In fec tio n S ev eri ty (IS ) MCP Treatment no biocontrol 1-112 2-28 4-6 6-25 Figure 13. Effect of 1-MCP (capital letters) and biological control isolates (lowercase letters) on infection severity in Gala apples from Field 12 (harvest 2007), inoculated with M. piriformis and stored for 3 months at 1°C in air. IS means with different letters within treatments are different at significant levels p ≤0.05 according to LSD test. 1-MCP effect comparisons (capital letters) can be made between 1-MCP treatment groups. Biological control activity comparisons (lowercase letters) can be made within 1-MCP treatment groups. 64 variation. As a result, the variable was eliminated from 2008 biological control experimental analysis as it explained little about the corresponding dependent and independent variables (Figure 14). 3.5.1 Natural disease incidence and biocontrol inoculations The effects of biological control treatment (no biocontrol, 1100-6, 1-112, 2-28, 4- 6 or 6-25) and incubation period (1, 2, 4 and 6 months) were analyzed using a two-way ANOVA. For Jonagold and Gala apples grown in Fields 3 and 12, respectively, then stored at 1°C in air, there was a significant effect of incubation period and of the interaction between the main effects, biological control treatment and incubation period (Appendix, Tables A30 and A31). For Jonagold apples grown in Field 3, there was no significant reduction in IS in apples treated with the five biological controls after 2 and 4 months in storage (Figure 15a and 15b). However, after six months in storage, isolates 2- 28 and 6-25 significantly reduced IS compared to the non- inoculated control (Figure 15c; Appendix, Table A32). For Gala apples grown in Field 12, isolates 1100-6, 1-112, 4-6 and 6-25 significantly reduced IS compared to the non-inoculated control after two months in storage (Figure 15d; Appendix, Table A33). There was no significant reduction in IS after four and six months storage compared to the non-inoculated control (Figure 15e and 15f). Jonagold apples grown in Field 3, Gala apples grown in Fields 12 and the Kiran orchard and Red Delicious apples grown in the Reekie orchard were also stored for 4.5 months in CA storage. A two-way ANOVA with replication indicated there was a significant effect of biological control treatment and location and of the interaction between biological control treatment and location (Appendix, Table A34). The one-way 65 0 5 10 15 20 25 NDI (1°C air) NDI (CA) NDI + bc (1°C air) NDI + bc (CA) Pathogen + bc (1°C air) Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Experiment wounded non-wounded Figure 14. Average infection severity for all wounded and non-wounded apples from individual biological control treatments conducted in 2008. Experiments were as follows: 1) Natural disease incidence (NDI) at 1°C in air; 2) Natural disease incidence (NDI) in controlled atmosphere (CA); 3) Natural disease incidence (NDI) in combination with biocontrols (bc) at 1°C in air; 4) Natural disease incidence (NDI) in combination with biocontrols (bc) in CA; 5) Pathogen in combination with biocontrols at 1°C in air. Error bars represent standard error of the mean. 66 a b b ab b b 0 2 4 6 8 10 no biocontrol 1100-6 1-112 2-28 4-6 6-25 b a b b b b 0 2 4 6 8 10 no biocontrol 1100-6 1-112 2-28 4-6 6-25 a ab ab bc a c 0 10 20 30 40 50 60 70 80 no biocontrol 1100-6 1-112 2-28 4-6 6-25 a a a a a a 0 10 20 30 40 50 60 70 80 no biocontrol 1100-6 1-112 2-28 4-6 6-25 bc c c ab bc a 0 10 20 30 40 50 60 70 80 no biocontrol 1100-6 1-112 2-28 4-6 6-25 b b b ab ab b 0 10 20 30 40 50 60 70 80 no biocontrol 1100-6 1-112 2-28 4-6 6-25 E.B. A. D. C. F. Av er ag e I nfe cti on Se ve rit y ( IS) Biocontrol Treatment Field 3, 2008 Field 12, 2008 Figure 15. Effect of biological control isolates on natural disease incidence in Jonagold apples grown in Field 3 (A – C) and Gala apples grown in Field 12 (D – F) (harvest 2008) and stored for 2 (A&D), 4 (B&E) or 6 (C&F) months at 1°C in air. IS means with different letters within figures are different at significance levels p ≤0.05 according to LSD test. 67 ANOVAs revealed a significant decrease in IS in Jonagold apples grown in Field 3 and inoculated with biological control isolates 1100-6, 1-112, 2-28, 4-6 and 6-25 (Figure 16; Appendix, Table A35). There was no significant reduction in IS in Gala apples grown in Field 12 and inoculated with one of five biological controls (data not shown; Appendix, Table A35). In Gala apples grown in the Kiran orchard, there was little overall disease. However, isolates 1100-6, 2-28, 4-6 and 6-25 significantly reduced IS compared to the non-inoculated control (Figure 17; Appendix, Table A35). For Red Delicious apples grown in the Reekie orchard, there was no significant effect on IS when inoculated with the biological control isolates (data not shown; Appendix, Table A35). 3.5.2 Post-harvest pathogen and biocontrol inoculations Infection severity (IS) was determined with apples grown in Fields 3 and 12 and inoculated with one of three pathogens, P. expansum, B. cinerea or M. piriformis and one of five bacterial control isolates. One way ANOVAs were computed to determine if there were significant effects of the biological control agents on infection severities within each field. For Jonagold apples grown in Field 3 and inoculated with P. expansum there was a significant reduction in IS in apples treated with isolates 4-6 and 6-25 compared to the non-inoculated control (Figure 18a; Appendix, Table A36). For Gala apples grown in Field 12 and inoculated with P. expansum, there was a significant reduction in IS in apples treated with isolates 1100-6, 2-28 and 6-25 compared to the non-inoculated control (Figure 19, Appendix, Table A36). In Jonagold apples grown in Field 3 and Gala 68 a c bc b bc bc 0 2 4 6 8 10 12 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev er ity (IS ) Biocontrol Treatment Figure 16. Effect of biological control isolates on natural disease incidence in Jonagold apples grown in Field 3 (harvest 2008) and stored for 4.5 months in CA. IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 69 a b ab b b b 0 0.1 0.2 0.3 0.4 0.5 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev er ity (IS ) Biocontrol Treatment Figure 17. Effect of biological control isolates on natural disease incidence in Gala apples grown in the Kiran orchard (harvest 2008) and stored for 4.5 months in CA. IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 70 a bc bc b bc c 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment b bc a bc c c 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment a b b b b b 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment A. P. expansum C. M. piriformis B. B. cinerea no biocontrol 1100-6 1-112 2-28 4-6 6-25 Figure 18. Effect of biological control isolates on infection severity in Jonagold apples grown in Field 3 (harvest 2008), inoculated with P. expansum (A) and B. cinerea (B) and stored for 2 months at 1°C in air and inoculated with M. piriformis (C) and stored for 1 month at 1°C in air . IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 71 a b b b b b 0 2 4 6 8 10 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment a c abc bc ab c 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment a bc bc b bc c 0 10 20 30 40 50 60 no biocontrol 1100-6 1-112 2-28 4-6 6-25 Av era ge In fec tio n S ev eri ty (IS ) Biocontrol Treatment A. P. expansum C. M. piriformis B. B. cinerea no biocontrol 1100-6 1-112 2-28 4-6 6-25 Figure 19. Effect of biological control isolates on infection severity in Gala apples grown in Field 12 (harvest 2008), inoculated with P. expansum (A) and B. cinerea (B) and stored for 2 months at 1°C in air and inoculated with M. piriformis (C) and stored for 1 month at 1°C in air. IS means with different letters are different at significance levels p ≤0.05 according to LSD test. 72 apples grown in Field 12 and inoculated with either B. cinerea (Figures 18b and 19b; Appendix, Table A37) or M. piriformis (Figures 18c and 19c; Appendix, Table A38) there was a significant reduction in IS in apples treated with all isolates, 1100-6, 1-112, 2-28, 4-6 and 6-25. Table 8 summarizes all biological control treatments in 2007 and 2008 and indicates the percent decrease (or increase) of all BCAs. Overall, results obtained from two years of experimental trials indicated that the bacterial isolate that provided significant control in the greatest number of treatments was Serratia plymuthica isolate 6-25 (51.7%) followed by P. fluorescens isolates 4-6 (41.4%), 1100-6 (37.9%), 1-112 (34.5%) and 2-28 (31%). The bacterial isolate that provided the greatest reduction in infection severity, in cases where significant control was exhibited, was isolate 1-112 followed by isolates 1100-6, 4-6, 6-25 and 2-28. Their respective average reduction in IS was 75.8%, 61%, 59%, 57.6% and 56.4%. In 2007, isolates 1100-6, 1-112 and 6-25 provided significant control of both P. expansum and B. cinerea. Isolate 4-6 provided significant control of P. expansum, only. Isolate 2-28 was the only bacterial strain to provide significant control of M. piriformis. In 2008, isolates 1100-6, 2-28, 4-6 and 6-25 provided significant control of all pathogens, P. expansum, B. cinerea and M. piriformis. Isolate 1-112 provided significant control of both P. expansum and B. cinerea (Table 8). All bacterial antagonists provided greater than 90% control in at least two experiments (Table 8). 73 Table 8. Summary of antagonist efficacy experiments for harvest 2007 and 2008. Percent disease decrease or (increase)a Experimentb Storage Typec Field Apple Variety Storage Time (months) 1-MCP+/- 1100-6 1-112 2-28 4-6 6-25 Harvest 2007 NDI + bc 1C air 3 Jonagold 3 MCP + NS NS NS NS NS 1C air 3 Jonagold 3 MCP - NS NS NS NS NS NDI + bc 1C air 12 Gala 3 MCP + NS NS NS NS NS 1C air 12 Gala 3 MCP - NS 82.1 NS 49.2 39.9 Pex + bc 1C air 3 Jonagold 3 MCP + 78.7 94.3 NS 97.3 53.2 1C air 3 Jonagold 3 MCP - 58.6 94.1 NS 96.9 96.2 1C air 12 Gala 3 MCP+&- NS NS NS 7.8 5.7 Bcin + bc 1C air 3 Jonagold 3 MCP + 94.7 98.1 NS NS 96.1 1C air 3 Jonagold 3 MCP - NS NS NS NS NS Mpir + bc 1C air 3 Jonagold 3 MCP + NS NS NS NS NS 1C air 3 Jonagold 3 MCP - NS NS NS (98.1) NS 1C air 12 Gala 3 MCP + NS NS 91.6 NS NS 1C air 12 Gala 3 MCP - NS NS NS NS NS Harvest 2008 NDI + bc 1C air 3 Jonagold 2 N/A (69.6) NS NS NS NS 1C air 3 Jonagold 4 N/A NS NS NS NS (52.9) 1C air 3 Jonagold 6 N/A NS NS 47.4 NS 67.5 1C air 12 Gala 2 N/A 74.7 88.3 NS 70.4 80.8 1C air 12 Gala 4 N/A NS NS NS NS NS 1C air 12 Gala 6 N/A NS NS NS NS NS 74 Percent disease decrease or (increase)a Experimentb Storage Typec Field Apple Variety Storage Time (months) 1-MCP+/- 1100-6 1-112 2-28 4-6 6-25 Harvest 2008 NDI + bc CA 3 Jonagold 4.5 N/A 94.9 83.8 43.0 63.9 45.5 CA 12 Gala 4.5 N/A NS NS NS NS NS CA Kiran Gala 4.5 N/A 100 NS 100 100 81.1 CA Reekie Red Delicious 4.5 N/A NS NS NS NS NS Pex + bc 1C air 3 Jonagold 2 N/A NS (26.0) NS 32.8 33.9 1C air 12 Gala 2 N/A 26.7 NS 20.0 NS 26.7 Bcin + bc 1C air 3 Jonagold 2 N/A 57.4 38.1 56.8 33.1 37.1 1C air 12 Gala 2 N/A 26.9 28.6 18.3 22.3 33.9 Mpir + bc 1C air 3 Jonagold 1 N/A 65.4 62.8 39.0 45.0 74.6 1C air 12 Gala 1 N/A 93.0 87.6 91.8 90.0 85.0 a. Percentages indicate a decrease in IS; percentages in brackets indicate an increase in IS; NS = not significant b. NDI = natural disease incidence; bc = biocontrol; Pex = P. expansum; Bcin = B. cinerea; Mpir = M. piriformis c. CA = controlled atmosphere 75 3.6 Bacterial survival on apple using GFP 3.6.1 Plasmid confirmation E. coli S-17 λ pir containing the gfp-plasmid, pAG408, grew on LB agar without antibiotics and LB agar supplemented with either 50 µg/ml kan or 30 µg/ml gen or both antibiotics combined. Alternatively, the bacterial isolates 1100-6, 1-112, 2-28, 4-6 and 6- 25 only grew on agar plates that were not supplemented with antibiotics. When pAG408 was digested with HindIII, the plasmid was cut into two fragments which measured approximately 2.1 and 3.6 kb (Figure 20). These fragments approximated the total plasmid size of 5.7 kb (Suarez et al. 1997). The undigested plasmid migrated to approximately 10 kb due to its supercoiled conformation 3.6.2 Confirmation of gfp transformation After transformation of genetic material from the host (E. coli S-17 λ pir) into the donor (isolate 4-6), transconjugants were amplified by PCR using gfp-specific primers. Successful gfp amplification was observed when a band was present around 714 bp (Figure 21). The absence of banding suggested an unsuccessful transformation. Confirmation of gfp transformation was additionally confirmed by fluorescence microscopy in vitro (Figure 22) and in vivo (Figure 23). 3.6.3 Plasmid stability After four successive days of sub-culturing and growth in the absence of antibiotic selection, 80.3 ± 6.8% of the transformed P. fluorescens isolate 4-6 (4-6-gfp) retained kanamycin and gentamicin resistance. These results, in combination with prior biological control experimental results, indicated that 4-6-gfp was an appropriate tool for 76 4.0 1 2 43 5 2.1 kb 3.6 kb3.0 2.0 kb Figure 20. HindIII restriction digest of gfp-containing plasmid, pAG408. Lane 1: 1 kb ladder (New England Biolabs, Ipswich, MA); Lanes 2 and 4: undigested pAG408; Lanes 3 and 5: digested pAG408. 77 Figure 21. PCR amplification of gfp in P. fluorescens isolate 4-6 transformed with pAG408 using gfp-specific primers. Sample were loaded on a 1% TBE agarose gel and stained with Sybr Safe (Invitrogen, Carlsbad, CA). Lane 1) Low DNA Mass Ladder (Invitrogen, Carlsbad, CA); 2) negative control (ddH20); 3) extracted plasmid, pAG408; 4 and 5) gfp transconjugant of bacterial isolate 4-6. 78 Figure 22. Fluorescence microscopy of P. fluorescens isolate 4-6 transformed with pAG408 exhibiting green fluorescence in ½ TSB supplemented with 50 µg/ml kan and 30µg/ml gen. 79 Figure 23. Fluorescence microscopy of an apple wound inoculated with P. fluorescens isolate 4-6-gfp. AP = apple peel; AW = apple wound; PF = P. fluorescens isolate 4-6-gfp; R = receptacle or fruit flesh. 80 studying bacterial survival on apples in the absence of antibiotic selection. 3.6.4 P. fluorescens 4-6-gfp fitness analysis There was no significant difference between the growth rates (µ) (Figure 24; Appendix, Table A39) of P. fluorescens isolate 4-6-wild-type (0.71 ± 0.04/h) and 4-6-gfp (0.75 ± 0.06/h) and sole source carbon utilization patterns (Table 9). Table 9. Sole carbon source utilization by P. fluorescens isolates 4-6-wild-type and 4-6- gfp as indicated by Biolog Phenotye MicroArrays™ (Biolog, Inc., Hayward, CA) for sugars and organic acids commonly found in apple juice (Eisele and Drake 2005). Apple Carbon Sources Bacterial Isolate 4-6-wild-type 4-6-gfp Fructose + + Sucrose + + Glucose + + Sorbitol + + L-Malic Acid + + Quinic Acid + + Citric Acid + + Fumaric Acid + + 3.6.5 Bacterial survival on apple For standard bacterial suspensions of 4-6-gfp, there was a strong linear relationship between CFU/ml and relative fluorescent units (RFU) (r = 0.999) (Figure 25). The fluorescent values obtained by direct scanning of inoculated apple samples (Figure 26) were too low to extrapolate to cell densities from the formula y = 1.8041x + 6.5652 (Figure 25). However, trends could be discerned based on RFUs (Figure 26). 81 0.01 0.1 1 10 0 4 8 12 16 20 24 log 10 OD 60 0 Time (h) 4-6-gfp 4-6-wild-type Figure 24. Growth characteristics of P. fluorescens 4-6-wild-type and P. fluorescens 4-6- gfp in half-strength TSB at 28°C for 24 h. Error bars represent standard error of the mean (n = 3). 82 y = 1.8041x + 6.5652 0 20 40 60 80 100 120 140 0 10 20 30 40 50 60 70 Re lat ive Flu or esc en t U nit s (R FU ) (x 10 3 ) CFU/ml (x 108) Figure 25. Relationship between P. fluorescens isolate 4-6-gfp cell density (CFU/ml) and relative fluorescent units (RFU) as determined by dilution plating and direct scanning (r = 0.999). Values are an average of three replicates. 83 -1000 0 1000 2000 3000 4000 5000 6000 0 10 20 30 40 50 60 Re lat ive Flu or esc en t U nit s (R FU ) Days in Storage (d) 4-6-gfp P. expansum 4-6-gfp + P. expansum Figure 26. Bacterial survival of 4-6-gfp (♦), P. expansum (●) and P. expansum + 4-6-gfp (■) on Gala apples stored for 55 d at 1°C in air using the direct scanning method. Error bars represent standard error of the mean (n = 4). 84 Isolate 4-6-gfp alone increased during 20 d in storage then decreased after 30 d in storage, followed by an increase at 55 d in storage. P. expansum alone followed the same trend; however, the overall RFUs were higher. The RFUs obtained for 4-6-gfp in the presence of P. expansum initially decreased in the first 10 d of storage and increased until 45 d in storage, followed by a decrease at 55 d in storage. The selective growth medium, PF agar supplemented with 50 µg/ml kan and 30 µg/ml gen, was effective in isolating 4-6-gfp and no other resident bacteria on the apple surface were detected. Over 55 d of dilutions and platings, there was no bacterial growth on the plates with extracted material from the untreated and P. expansum only-inoculated apples. A two-way ANOVA indicated a significant interaction between bacterial strain, 4-6-wild-type and 4-6-gfp, and time (Appendix, Table A40). One-way ANOVAs revealed no significant differences in cell densities between apples inoculated with 4-6- gfp + P. expansum and apples inoculated solely with 4-6-gfp up to 20 d in 4°C storage (Figure 27; Appendix, Table A41). From 30 to 55 d of storage, populations of 4-6-gfp in the presence of P. expansum differed significantly from populations of 4-6-gfp alone. When challenged with the pathogen, bacterial populations remained constant until lesions developed after 30 d, after which populations rapidly declined. When 4-6-gfp was unchallenged with the pathogen, its population increased from 20 – 30 d in storage then declined to slightly greater than its population at time 0. From 0 to 40 d in storage, 4-6- gfp significantly reduced IS compared to the control (P. expansum alone) (Table 10; Appendix, Table A42). There was no significant reduction in IS of apples treated with 4- 85 6-gfp in the presence of P. expansum after 40 d in storage (Table 10; Appendix, Table A42). Table 10. Effect of biological control isolate 4-6-gfp on infection severity in Gala apples inoculated with and without P. expansum and stored for 55 days at 1°C in air. IS means with different letters within columns are different at significant levels p ≤0.05 according to the least significant difference (LSD) test. Treatments Days in storage 10 20 30 40 45 55 Untreated 0.0 0.0 0.0 0.0 0.0 0.0 4-6-gfp alone 0.0 0.0 0.0 0.0 0.0 0.0 P. expansum alone 0.0 0.0 2.7 a 7.8 a 11.5 a 19.6 a 4-6-gfp + P. expansum 0.0 0.0 0.0 b 3.0 b 8.4 a 16.9 a 86 a a a a a a a a a b b b0 2 4 6 8 0 10 20 30 40 50 60 log 10 CF U/ ml Days in Storage (d) 4-6-gfp 4-6-gfp + P. expansum Figure 27. Bacterial survival of 4-6-gfp (♦) and 4-6-gfp + P. expansum (●) on Gala apples stored for 55 d at 1°C in air using the dilution plating method CFU/ml means with different letters within sampling times are different at significance levels p≤0.05 according to LSD test (n = 4). 87 CHAPTER 4: DISCUSSION 4.1 DNA macroarray for post-harvest apple pathogens 4.1.1 DNA macroarray quantification In the current study, it was hypothesized that macroarray technology would accurately identify and quantify post-harvest apple pathogens throughout the growing season and that the frequency of pathogen detection would correlate with disease incidence post-harvest. This study reports that post-harvest pathogen quantification can be accomplished using DNA macroarray technology and that apple post-harvest pathogens can be detected throughout the apple growing season. Although many studies have used macroarray technology to detect and identify various plant pathogens and biological control microorganisms (Fessehaie et al. 2003; Izzo and Mazzola 2009; Le Floch et al. 2007; Levesque et al. 1998; Lievens et al. 2003; Robideau et al. 2008; Sholberg et al. 2005a; Tambong et al. 2006), few have quantified macroarray hybridization signal intensity (Lievens et al. 2005; Lievens et al. 2007; Sholberg et al. 2006) as we have done in this study. Pathogen DNA was quantified by hybridizing 0 – 100 ng of pure culture DNA extracts of P. expansum, B. cinerea and M. piriformis to the macroarray. The detector probes or oligonucleotide sequences had varying levels of hybridization signal intensity. As found by Sholberg et al. (2006) who correlated greyscale value and cell density, a linear relationship was found between greyscale value and DNA concentration. However, in this study, the hybridization signals became saturated outside of the linear range at >3 ng for P. expansum, >12 ng for B. cinerea and >0.5 ng for M. piriformis (data not shown). Lievens et al. (2005), who quantified DNA concentrations of 88 phytopathogenic fungi, Verticillium albo-atrum, Verticillium dahlia, Fusarium oxysporum, Fusarium solani, Pythium ultimum and Rhizoctonia solani, also reported a linear relationship between signal intensity and DNA concentration. Linearity was established when 0.5 fmol of detector probe was spotted on a membrane and template concentrations ranged from 0.25 pg to 2.5 ng which was in the same range or somewhat lower than the linear range in the present study. Below and above 0.5 fmol, the curves deviated from linearity. Lievens et al. (2007) also reported that hybridization signals between 0.25 to 25 pg (0.00025 to 0.025 ng), that detect Trichoderma spp. and Trichoderma hamatum, a disease suppressive organism, resulted in a logarithmic relationship between signal intensity and template DNA concentration. This was lower than the linear range reported in the present study. Above 25 pg, the standard curves deviated from linearity. Although a range of detector probe concentrations were not tested within this study, Lievens et al. (2005) noted that altering the amount of oligonucleotide spotted on the membrane can affect the range of DNA concentrations detected by a specific macroarray probe. Once a greater knowledge base exists about the level of post-harvest pathogenic DNA found throughout a crop’s growing season and its correlation with disease manifestation, the DNA oligonucleotide detector probe concentrations can be altered to detect ranges typically found in field conditions. This will allow for more accurate DNA quantification and disease management. For enhanced and more precise quantification, it would be advantageous to simultaneously employ hybridization signal analysis with real-time PCR. The ability to accurately quantify a pathogen is useful in 89 plant disease management as information can be used to assess potential risk and disease development (Lievens and Thomma 2005). 4.1.2 Field monitoring of apple post-harvest pathogens The current study was the first of its kind to monitor post-harvest pathogen presence via molecular technologies prior to commercial storage and disease development. Of the samples collected in each of the three growing season sampling periods, there was a tendency to detect total populations of all three post-harvest fungi more frequently in the early and mid-growing seasons of 2007 (56.6% - early; 60% - mid; 33.3% - late) and 2008 (67.6% - early; 63.2% - mid; 34% - late). This tendency to detect lower pathogen levels from air and plant tissue samples in the late growing season corresponds with the lower pathogen levels detected on whole apples collected at harvest (Table 4). In addition, the increase in early season fungal detection corresponds with results obtained by Teixidó et al. (1999) who observed total bacterial, yeast and filamentous fungal populations on Golden Delicious apples. However, they found an even greater increase in microorganisms at the apple bud stage followed by a decrease in populations at blossom. It was suggested that as buds form blossoms, there is a loss of external structure due to blossom drop, which accounts for the reported decrease in microorganisms after the apple budding stage. In the current investigation, samples were not collected at the apple bud stage. The rise in total fungal propagules may therefore correspond with precipitation events in the mid-growing season of 2007 and the early and mid-growing season of 2008 (Figure 1). Of the three major post-harvest pathogens, P. expansum, on average, was detected most frequently (27.4. ± 3.4%) and was found throughout the growing season. In 2008 in 90 Fields 3 and 12 there was an increased detection frequency at the end of the growing season (Table 3). The abundance of P. expansum found within this study contradicts results of Amiri and Bompeix (2005) and Teixidó et al. (1999) who reported low levels of Penicillium inoculum within the orchard atmosphere. However, the aerial sampling period of Amiri and Bompeix (2005) was between July and September which excludes the early growing season. In contrast to P. expansum, B. cinerea was detected the least number of times (6.2 ± 1.4%) and found throughout the growing season of 2007 and 2008. The most important infective units of B. cinerea are the conidia produced in late winter and early spring from over-wintering mycelia and/or sclerotia on host tissue or soil surfaces (Elad et al. 2007); grey mold is rarely seen in the field (Rosenberger 1990). The low levels of B. cinerea as indicated by the macroarray correspond with these results. M. piriformis was detected the second most frequently (19.2 ± 1.4%) and found most frequently in the mid-growing season of 2007 and the early to mid-growing season of 2008 in DNA isolated from plant tissue samples. These findings support Guo et al. (1999) who found an increase in M. piriformis populations within orchard soil from December to March and April, followed by a decrease afterwards. It was suggested that the increase in M. piriformis corresponds with apples left on the orchard floor and cool weather. Alternatively, these results contradict those found by Dobson and Spotts (1988) who recovered no M. piriformis from orchard air but found an abundance of B. cinerea and P. expansum. Sporangiospores of M. piriformis are soilborne (Michailides and Spotts 1986); greater than 75% are found in the top 2 cm of the soil profile and are dispersed primarily by rain splash and insects (Dobson and Spotts 1988). The detection 91 of an abundance of M. piriformis in this study may correspond to precipitation bursts found in both 2007 and 2008 (Figure 9). Sporangiospores may be aerosolized by the falling rain and then trapped by the aerial spore sampler. Earlier studies, as mentioned above, that detected fungal abundance throughout the orchard growing season utilized culturing methods for fungal detection and identification (Amiri and Bompeix 2005; Dobson and Spotts 1988; Teixidó et al. 1999). In the case of P. expansum and M. piriformis, the macroarray detected the presence of pathogens not commonly found within the orchard. Culturing methodologies for detection and relative quantification may underestimate fungal abundance, thus showing the power of macroarray technology. Data obtained using the macroarray may be useful in developing a disease forecasting model such as that presented by Spotts et al. (2009). For future experiments, it would be beneficial to correlate pathogen findings throughout the growing season with species-specific identification of natural disease incidence found within the packinghouse. In this study, a positive correlation was only evident between natural disease incidence and percent pathogens detected in the late-growing season for aerial and plant tissue samples analyzed together (r = 0.79) and for aerial samples (r = 0.74) when analysed separately from plant tissue samples. This finding corresponds with Lennox et al. (2003) who observed a significant correlation between the density of B. cinerea conidia on the pear fruit surface and grey mold disease incidence in cold storage. In addition, Walter et al. (1997) found a strong correlation between B. cinerea berry fruit contamination and latent boysenberry infection in New Zealand. Given the cost and time involved in collecting and analyzing DNA for macroarray identification and 92 quantification, the majority of samples could be collected, for example, near the end of the growing season. It should be noted, however, that Spotts et al. (2009) found that in creating an at-harvest prediction model for grey mold risk in pears, surface DNA was the least important predictor in the model as opposed to orchard rating and fungicide applications. Also of note, within this study, there was a negative correlation between percent pathogens detected in the early-growing season and natural disease incidence when plant tissue samples were analysed separately from aerial samples (r = -0.79). This data suggests that perhaps plant tissue samples collected in the early-growing season are not optimal predictors for natural disease incidence, thereby supporting the correlations mentioned above. More work is required in order to determine the relatedness between growing season pathogen levels and post-harvest disease prevalence. The variability in the results of this study suggests that a multitude of factors contribute to disease incidence found post-harvest. Coley-Smith et al. (1980) proposed that disease outbursts may be a result of complex interactions between the production and dispersal of various inocula, infection pathogenesis, pathogen survival along with temperature, rainfall, humidity and crop protection, nutrition and phenology. Variability may also be explained by the application of pre-harvest fungicides and their effect on resident microflora. According to Sholberg and Boulé (2008), the application of Nova®, a fungicide used to control powdery mildew, reduced apple leaf and fruit microflora in two years of study. In the current study, Nova® was applied to Field 12 on May 15, 2007 which may help to explain the reduced pathogen loads when compared to the 2008 results (Figure 6). It should not be forgotten that the majority of post-harvest diseases are 93 a result of pathogens entering the fruit via wounds, all of which should be considered in accurate disease forecasting models. 4.2 Biological control of post-harvest pathogens Five putative bacterial control agents were applied to apples harvested from commercial and experimental orchards in the Okanagan Valley, BC. It was hypothesized that one or more of the bacterial antagonists would provide significant pathogen control in semi-commercial and commercial storage conditions. This study showed that the four P. fluorescens isolates and one S. plymuthica isolate exhibited some control and antagonistic efficacy against post-harvest apple pathogens (Table 8). As indicated, Serratia plymuthica isolate 6-25 provided significant control in the greatest number of treatments (57.6%) (Table 8). These results suggest that this bacterium may provide control against a variety of post-harvest diseases and storage conditions and its use may be applied to other crops. Using S. plymuthica as a BCA, however, is a challenging endeavour as concerns surround its biosafety, a major consideration when selecting an antagonist (Droby et al. 2009). Serratia plymuthica belongs to the family Enterobacteriaceae. Members of this family, such as Escherichia coli strain O157:H7, have been known to cause human infections (Health Protection Agency 2007). Although opportunistic pathogenic strains of S. plymuthica have been isolated (Carrero et al. 1995; Clark and Janda 1985; Domingo et al. 1994) infection is rare compared to other members of the genus, Serratia, or of the family, Enterobacteriaceae. Despite concerns surrounding its use as a biological control agent, S. plymuthica isolate 6-25 provided significant control in this study, against apple post-harvest pathogens, P. expansum, B. cinerea and M. piriformis. S. plymuthica biological control 94 efficacy has also been demonstrated by other research groups, although not on post- harvest apple pathogens. S. plymuthica strain HRO-C48 has been studied to control Verticillium dahliae, the causal agent of Verticillum wilt in oil-seed rape (Müller and Berg 2008). HRO-C48 also suppressed growth of V. dahliae and Phytophthora cactorum in strawberries (Kurze et al. 2001) and a commercial product based on strain HRO-C48 has been developed, RhizoStar® (e-nema GmbH, Raisdorf, Germany) (Müller and Berg 2008). S. plymuthica strain 5-6 also showed significant control of dry rot of potato caused by Fusarium sambucinum (Gould et al. 2008). The post-harvest biological control provided by S. plymuthica isolate 6-25 of this study should not be disregarded as the potential exists, as evident by other studies, for commercial application. This study also showed that pseudomonads, specifically P. fluorescens, are also capable of providing control against post-harvest pathogens. According to Stockwell and Stack (2007), Pseudomonas spp., have been studied for decades as model organisms for biological control of plant diseases. In this study the greatest reduction in IS, in cases where significant control was exhibited, was with P. fluorescens isolate 1-112 followed by 1100-6 (Table 8). Other pseudomonads are also being studied for their efficacy against post-harvest pathogens. Mikani et al. (2008) reported significant control of Botrytis mali, formerly thought of as B. cinerea and recently revived by O’Gorman et al. (2008), using 10 strains of P. fluorescens isolated from leaf surfaces and apple fruit. Similar to the level of control found in this study, the results of Mikani et al. (2008), reported approximately 34.9% to 95.9% B. mali inhibition on Golden Delicious apples after 25 d at 5°C. In addition, Zhou et al. (2001) studied four strains of Pseudomonas syringae that were isolated from the phyllosphere of apple trees and controlled blue 95 mold, following inoculation by pipette, by 64 – 70% at 4°C after 28 d. The bacterial isolates also controlled grey mold, but to a lesser extent. Similar to methods used in this study, Zhou et al. (2001) also inoculated apples by dipping and found that a P. syringae isolate controlled blue mold by 93% after 133 d at 1°C in air, with the overall incidence of blue mold at 30%. In many cases, (Figure 12a) the bacterial isolates used in the present study exhibited similar and greater levels of control than those reported by Mikani et al. (2008) and Zhou et al. (2001). The effectiveness of these isolates may be due to the unique environment in which they were isolated. Many BCAs are found on the fruit surface or directly within a wound. The microbial agents used in this study, however, were isolated from the rhizosphere of legumes grown in cold Saskatchewan soils and may be more adapted to cold storage conditions. Droby et al. (2009) outlined the importance of antagonist selection and the use of a variety of screening procedures in finding effective BCAs. An assortment of selection techniques would increase biocontrol species diversity and therefore the potential for a successful product. These experiments illustrate the potential of S. plymuthica and P. fluorescens isolates for biological control application and commercialization. 4.2.1 Bacterial antagonists and CA storage One type of commercial storage application used in this study was controlled atmosphere (CA). This research confirmed the hypothesis that one or more antagonists would provide control in commercial conditions. In all experiments, the lowest disease levels were found in apples stored in CA storage for 4.5 months following harvest in 2008. Many suggestions have been made regarding the selection criteria for potential BCAs (Wilson and Wisniewski 1989). One such criterion is the ability of the antagonist 96 to be compatible with current post-harvest commercial processing procedures such as CA (Janisiewicz and Korsten 2002; Wilson and Wisniewski 1989). This current study indicated that the selected antagonists were compatible with CA storage practices and may actually extend apple storage time due to enhanced pathogen suppression when compared to storage at 1°C. For example, the highest disease incidence recorded in CA was in Jonagold apples which reached an average IS of 10.39 ± 1.6 after 4.5 months in storage (Figure 10). The biological antagonists provided significant control against naturally occurring diseases in all orchards but Field 12 and the Reekie orchard (data not shown). In comparison, for Jonagold and Gala apples grown in Fields 3 and 12 then stored at 1°C in air for 4 months, non-inoculated controls had IS levels of 15.65 ± 5.0 (Figure 16b) and 19.90 ± 4.8 (Figure 16e), respectively. These values were almost double the infection severities found in CA. In both Jonagold and Gala apples harvested from Fields 3 and 12, respectively, the bacterial antagonists did not provide a significant reduction in IS after 4 months in storage (Figure 16b and e). However, isolate 6-25 reduced IS in Jonagold apples grown in Field 3 after 6 months in storage (Figure 16c). The decrease in apple disease supports the findings of Smock (1979) as CA effectively delayed the onset of storage decay. In the CA-stored apples, natural disease incidence was very low even after 4.5 months of storage, except in the case of Jonagold apples from Field 3 which appeared to be due primarily to P. expansum (based on phenotypic observation). The gas composition of a packinghouse can directly or indirectly suppress fungal decay. Direct inhibition affects mycelium development and spore germination of resident pathogens. Alternatively, storage conditions also affect senescence and ripening of hosts, which, in 97 turn, affects their ability to resist infections, thereby indirectly controlling fungal decay (Barkai-Golan 2001). The implementation of CA is a balance between O2 and CO2 concentrations. General recommendations include 1 – 2 % O2 and 0.5 – 2% CO2. Concentrations outside of these ranges may lead to fruit injury which can affect fruit flavour and appearance (Kupferman 2003). However, post-harvest pathogens are only strikingly suppressed after O2 concentrations have been lowered to less than 1% and CO2 concentrations increased to greater than 10% (El-Goorani and Sommer 1981). The CA storage conditions in this study were approximately 1.5% O2 and 1.5% CO2, a range in which, if the conditions were right, may lead to the fungal outbreak that was observed in the Jonagold apples. In the current study, the antagonistic effects of the bacterial isolates were difficult to observe because there was little natural disease present; if little disease was present, then there was little disease to control. However, the outbreak in Jonagold apples from Field 3 demonstrated the antagonistic potential of the bacteria because significant disease was present. While all of the antagonists demonstrated significant control, isolate 1100-6 and 1-112 were particularly effective, providing 94.9% and 83.8% reduction in disease symptoms in Jonagold apples from Field 3 (Figure 17). Although not directly inoculated with a pathogen, these BCAs performed better than the commercially registered BCA, Bio-Save 10 LPTM, in studies conducted by Janisiewicz and Jeffers (1997) and Zhou et al. (2001). Using Bio-SaveTM during 133 and 123 days in CA storage, Zhou et al. (2001) reported a 57% and a 73% reduction of blue mold in Empire and Delicious apples, respectively. It would be beneficial to test the efficacy of the bacterial antagonists against artificially inoculated pathogens under CA conditions as in Conway et al. (2007), 98 Janisiewicz et al. (2008) and Tian et al. (2002). However, during the present study, permission for pathogen inoculation within a commercial storage facility was not given. 4.2.2 Bacterial antagonists and 1-MCP Another type of commercial storage application tested in this study was the use of 1-MCP. This research confirmed the hypothesis that one or more antagonists would provide control in commercial conditions, although to a lesser extent. 1-MCP was applied to Jonagold apples grown in Field 3 and Gala apples grown in Field 12 in the biological control experiments using apples from the 2007 harvest. It was postulated that the application of 1-MCP would result in a decrease in disease incidence and severity of non-inoculated wounded apples or apples inoculated with putative biological control agents and/or post-harvest pathogens. However, when the overall effect of 1-MCP was examined, a reduction in infection severity occurred in only 14% of the experimental treatments. For example, 1-MCP application reduced IS caused by natural disease incidence in Jonagold apples from Field 3 (Table 5) by 51.8% and in Gala apples from Field 12 (Figure 11) by 67.1%. Variable levels of control confirm findings of Saftner et al. (2003) who studied the effects of pre-storage heat, CA storage and pre- and post- application of MCP on inhibition of fungal decay and maintenance of Golden Delicious fruit quality. Pre-storage treatment of MCP reduced lesion size (mm) of fruit inoculated with B. cinerea and P. expansum by 34% and 17.9%, respectively. Post-storage application of MCP had no effect on reducing lesion diameter. Variable responses to 1- MCP application are not uncommon. Watkins (2008) indicated that the apple is a fruit with many cultivars that differ in their ripening rates, harvest criteria, post-harvest handling practices and storage periods in air and CA. In addition, time between harvest 99 and 1-MCP treatment and repeated application have variable effects in disease resistance responses. In the current study, when a decrease in infection severity was observed, it appeared to be primarily in experiments where there was no artificial inoculation of P. expansum, B. cinerea or M. piriformis; the only pathogens present were those that occurred naturally. Contradictory to the intended outcome of 1-MCP application, some treatments in the present study resulted in an increase in infection severity of apples treated with 1- MCP, especially those apples inoculated with B. cinerea (Figure 12b). An increase in decay after 1-MCP application was also observed by Janisiewicz et al. (2003) and Leverentz et al. (2003) on Golden Delicious apple fruit and by Bedford et al. (2002) who reported an increase in B. cinerea decay when apples were fumigated with 1-MCP. Commercial application of 1-MCP as SmartFresh® is thought to delay apple ripening by preventing ethylene production via preferential binding to the ethylene receptor site and extending the action of natural defence mechanisms due to delayed ripening (Blankenship and Dole 2003). However, the reverse was observed in some cases as 1- MCP appeared to hinder plant defence mechanisms. According to Marcos et al. (2005), endogenous ethylene may be an important component in plant resistance and defence gene regulation; the application of 1-MCP may therefore compromise a plant’s defence response system (Jiang et al. 2001). Despite the controversy surrounding chemical control agents, fungicides still remain the preferred choice for the prevention of post-harvest diseases. Although fungicide-resistant pathogens have increased within the population, chemical control provides a consistent and effective reduction in disease. Within the agricultural 100 community, there is a relatively high level of scepticism surrounding the use of biological control products (Droby et al. 2009). Commercially acceptable decay levels are generally below 5% (Janisiewicz and Korsten 2002) and because of the inconsistency and variability of biofungicides, the packinghouse industry will often choose the most economically viable option that consistently reduces rot. Disease control was exhibited by the bacterial antagonists in this study, but efficacy was highly variable, as observed by other researchers. The tritrophic interaction between host, pathogen and antagonist is a highly complex one that demands a more sound and fundamental understanding. In addition to exploring the mechanism(s) of action for fungal suppression, combination and alternative treatments for enhanced efficacy should be explored. Such alternate biocontrol treatments, outlined by Sharma et al. (2009) include, for example, the addition of low doses of fungicides, salt additives, or nutrients and plant products, as well as the use of mixed cultures or manipulation of the physical environment. Rapid antagonist colonization of fruit wounds is critical to controlling decay; therefore, enhancements leading to improved wound colonization should be further explored (Janisiewicz and Korsten 2002). 4.3 Bacterial survival on apple Acquiring basic knowledge about potential antagonists will provide a greater understanding into the mechanisms of actions, which will, in turn, offer insight into means for achieving more consistent control. Disease prevention relies on a quantitative relationship between antagonist and pathogen (Janisiewicz 1988) and knowledge of antagonist cell number during time in storage will assist in enhanced disease suppression (Etebarian and Sholberg 2005). In the present study, Pseudomonas fluorescens isolate 4- 101 6 was tagged with GFP to determine the survival characteristics of this antagonist alone and when challenged with P. expansum. It was hypothesized that green fluorescent protein would facilitate visualization of bacterial colonization and survival on apple surfaces. In this study, gfp did facilitate bacterial visualization, but indirectly as pAG408 provided antibiotic resistance of isolate 4-6 which permitted selective plating that excluded other organisms during survival analysis. Bacterial concentrations were determined using direct scanning and colony counts. There was a strong linear relationship between fluorescent intensity and cell density (Figure 26). These results corresponded with those found by Etebarian and Sholberg (2005) who studied population levels of P. fluorescens isolate 1100-6 labelled with GFP using direct scanning, cell counts and microscopy. This and the current study illustrated that measuring fluorescent intensity as a means of estimating cell concentration was an effective method of population assessment. The initial inoculation concentration of 4-6-gfp (108 CFU/ml), however, combined with the method of recovery, resulted in fluorescent readings that were not high enough to quantify using the linear equation y = 1.8041x + 6.5652 (Figure 26). Using a Fast Prep machine for tissue maceration as in Etebarian and Sholberg (2005) or in Etebarian et al. (2005) may overcome issues incurred by extraction. Despite low values, the raw microplate data were compared to plate counting data. Relative fluorescent units (RFUs) (Figure 27) and cell densities (CFU/ml) (Figure 28) of isolate 4-6-gfp followed similar patterns; when cell counts increased, so did RFUs. However, there was an inverse relationship between RFUs and CFU/ml for 4-6-gfp inoculated in the presence of P. expansum; RFUs increased while cell counts decreased. 102 Two possible explanations are that P. expansum auto-fluoresces or that the decayed apple tissue auto-fluoresces. The latter explanation corresponds with the formation of lesions at 30 d (Table 10). Increased antagonist populations after 30 d of storage were also found by Janisiewicz and Marchi (1992) who reported a 10 – 100 fold increase in populations of Pseudomonas syringae inoculated on pears stored at 1°C. Etebarian and Sholberg (2005) also found an increase in P. fluorescens isolate 1100-6 GFP on apples after 35 days in 5°C storage. However, unlike the survival characteristics of 4-6-gfp + P. expansum in this study, 1100-6 GFP + P. expansum maintained as high or higher cell numbers than 1100-6 GFP alone. Low cell counts after lesion development may be attributed to difficulties in extracting apple cores from completely decayed tissue. Even so, after 30 d at 1°C in air storage, 4-6-gfp appeared to lose some of its antagonistic capabilities (Table 10). Information such as this should be considered when implementing control strategies for optimal post-harvest disease management. 103 CHAPTER 5: CONCLUSIONS This thesis built upon preliminary research and addressed the following three objectives: 1) to validate the performance of DNA macroarrays under field conditions; 2) to assess bacterial antagonist performance under semi-commercial and commercial conditions and; 3) to determine how long and in what capacity a selected antagonist will colonize the fruit surface. The first hypothesis addressed the first objective and stated that macroarray technology would accurately identify and quantify post-harvest apple pathogens throughout the growing season and that disease prevalence would correlate with disease incidence post-harvest. Results confirmed that in two years of field trials, the DNA macroarray is a fast and easy technique capable of detecting apple post-harvest pathogens throughout the apple growing season (Table 3). Commercial application of this diagnostic and quantitative technique would provide accurate information to assist growers and field personnel in crop management. A positive correlation existed between pathogen detection at the end of the growing season and natural disease incidence found post-harvest. More work is required to obtain information about the importance of monitoring pre-harvest pathogen loads and their contribution to post-harvest disease. The second hypothesis addressed the second objective and stated that one or more of the bacterial antagonists would provide significant pathogen control in semi- commercial and commercial storage conditions. Results indicated that despite variability between years, variety and biological control efficacy, the potential BCAs demonstrated control at 1°C in air and particularly in CA storage (Table 8). Biocontrol efficacy in combination with 1-MCP was less effective and the application of 1-MCP greatly 104 increased disease levels with B. cinerea artificial inoculation (Figure 12b). Combining some or all of the bacterial control agents used in this study in conjunction with other organic amendments such as salts (ie. sodium bicarbonate) or manipulating the physical environment (ie. heat treatment) may provide consistency in control which the industry demands. The third hypothesis addressed the third objective and stated that the use of green fluorescent protein would facilitate visualization of bacterial colonization and survival on apple surfaces. Results indicated that green fluorescent protein was an effective technique to monitor population levels as the gfp-marked strain behaved similarly to the wild-type strain (Figure 25; Table 9). The fluorescence levels obtained by direct scanning, however, were not sufficient to distinguish 4-6-gfp from background auto- fluorescence. The presence of the gfp-containing plasmid within P. fluorescens isolate 4- 6 therefore provided an indirect method of monitoring population levels as pAG408 contained antibiotic resistant genes that allowed for strain-specific isolation from an inoculated apple wound. Bacterial populations were found to initially increase then decrease within an apple wound during 55 d in storage at 1°C (Figure 28). In the presence of P. expansum, 4-6-gfp populations declined, which coincided with P. expansum lesion development. The information obtained within this study will assist the fruit industry in detecting, quantifying and biologically controlling post-harvest pathogens of pome fruit. 105 REFERENCES AAFC 2002. Canadian Apple Online [online]. Available from http://ats.agr.ca/applecanada/home-e.htm [accessed 6 March 2008]. Abramson, D., Lombaert, G., Clear, R.M., Sholberg, P., Trelka, R., and Rosin, E. 2009. Production of patulin and citrinin by Penicillium expansum from British Columbia (Canada) apples. Mycotox. Res. 25(2): 85-88. Amiri, A., and Bompeix, G. 2005. Diversity and population dynamics of Penicillium spp. on apples in pre- and postharvest environments: consequences for decay development. Plant Pathol. 54(1): 74-81. Bailey, K.L., Boyetchko, S.M., and Längle, T. 2010. Social and economic drivers shaping the future of biological control: a Canadian perspective on the factors affecting the development and use of microbial biopesticides. Biol. Control 52(3): 221-229. Barkai-Golan, R. 2001. Postharvest diseases of fruits and vegetables: development and control. Elsevier Science B.V. Amsterdam, The Netherlands. Bedford, K.E., Stokes, S.C., Sholberg, P.L., and Lau, O.L. 2002. Effects of 1-MCP on postharvest decay of apples caused by Penicillium expansum and Botrytis cinerea. Can. J. Plant Pathol. 24(3): 381. Bencheqroun, S.K., Bajji, M., Massart, S., Labhilili, M., El Jaafari, S., and Jijakli, M.H. 2007. In vitro and in situ study of postharvest apple blue mold biocontrol by Aureobasidium pullulans: Evidence for the involvement of competition for nutrients. Postharvest Biol. Technol. 46(2): 128-135. 106 Benhamou, N., Gagné, S., Le Quéré, D., and Dehbi, L. 2000. Bacterial-mediated induced resistance in cucumber: beneficial effect of the endophytic bacterium Serratia plymuthica on the protection against infection by Pythium ultimum. Biochem. Cell Biol. 90(1): 45-56. Blankenship, S.M., and Dole, J.M. 2003. 1-methylcyclopropene: a review. Postharvest Biol. Technol. 28(1): 1-25. Bloemberg, G.V. 2007. Microscopic analysis of plant–bacterium interactions using auto fluorescent proteins. Eur. J. Plant Pathol. 119(3): 301–309. Bloemberg, G.V., O’Toole, G.A., Lugtenberg, B.J.J., and Kolter, R. 1997. Green fluorescent protein as a marker for Pseudomonas spp. Appl. Environ. Microbiol. 63(11): 4543-4551. Bolwerk, A., Lagopodi, A.L., Wijfjes, A.H.M., Lamers, G.E.M., Chin-A-Woeng, T.F.C., Lugtenberg, B.J.J., and Bloemberg, G.V. 2003. Interactions in the tomato rhizosphere of two Pseudomonas biocontrol strains with the phytopathogenic fungus Fusarium oxysporum f. sp. radicis-lycopersici. Mol. Plant Microbe Interact. 16(11): 983-993. Bridge, P., and Spooner, B. 2001. Soil fungi: diversity and detection. Plant Soil 232(1-2): 147-154. British Columbia Ministry of Agriculture, Food and Fisheries. 2004. Industry Overview: Apples and Pears in British Columbia [online]. Available from http://www.agf.gov.bc.ca/treefrt/pubs/pubs_toc.htm [accessed 24 February 2008]. Calvo, J., Calvente, V., de Orellano, M.E., Benuzzi, D., and de Tosetti, M.I.S. 2007. Biological control of postharvest spoilage caused by Penicillium expansum and 107 Botrytis cinerea in apple by using the bacterium Rahnella aquatilis. Int. J. Food Microbiol. 113(3): 251-257. Carrero, P., Garrote, J.A., Pacheco, A., Garcia, A.I., Gil, R. and, Carbajosa, S.G. 1995. Report of six cases of human infection by Serratia plymuthica. J. Clin. Microbiol. 33(2): 275-275. Chalfie, M., Tu, Y., Euskirchen, G., Ward, W.W., and Prasher, D.C. 1994. Green fluorescent protein as a marker for gene expression. Sci. 263(5148): 802-805. Chan, Z., and Tian, S. 2005. Interaction of antagonistic yeasts against postharvest pathogens of apple fruit and possible mode of action. Postharvest Biol. Technol. 36(2): 215-223. Clark, R.B., and Janda, J.M. 1985. Isolation of Serratia plymuthica from a human burn site. J. Clin. Microbiol. 21(4): 656-657. Coley-Smith, J.R., Verhoeff, K., and Jarvis, W.R. 1980. The Biology of Botrytis. Academic Press, New York, NY. Conway, W.S., Janisiewicz, W.J., Leverentz, B., Saftner, R.A., and Camp, M.J. 2007. Control of blue mold of apple by combining controlled atmosphere, an antagonist mixture, and sodium bicarbonate. Postharvest Biol. Technol. 45(3): 326-332. Cook, R.J., and Baker, K.F. 1983. The nature and practice of biological control of plant pathogens. The American Phytopathological Society, St. Paul, MN. Cossentine, J.E., Sholberg, P.L., Jensen, L.B.J., Bedford, K.E., and Shephard, T.C. 2004. Fumigation of empty fruit bins with carbon dioxide to control diapausing coldling moth larvae and Penicillium expansum Link. ex Thom spores. HortSci. 39(2): 429- 432. 108 Cruickshank, R.H., and Pitt, J.I. 1987. Identification of species in Penicillium subgenus Penicillium by enzyme electrophoresis. Mycol. 79(4): 614-620. de Jong, S.N., Lévesque, C.A., Verkley, G.J.M., Abeln, E.C.A., Rahe, J.E., and Braun, P.G. 2001. Phylogenetic relationships among Neofabraea species causing tree cankers and bull's-eye rot of apple based on DNA sequencing of ITS nuclear rDNA, mitochondrial rDNA, and the β-tubulin gene. Mycol. Res. 105(6): 658-669. Dobson, R.L., and Spotts, R.A. 1988. Distribution of sporangiospores of Mucor piriformis in pear orchard soils. Plant Dis. 72(8): 702-705. Domingo, D., Limia, A., Alarcon, T., Sanz, J.C., Del Rey, M.C. and, Lopez-Brea, M. 1994. Nosocomial septicemia caused by Serratia plymuthica. J. Clin. Microbiol. 32(2): 575-577. Doores, S. 1983. The microbiology of apples and apple products. Crit. Rev. Food Sci. Nutr. 19: 133-149. Droby, S., Wisniewski, M., Macarisin, D., and Wilson, C. 2009. Twenty years of postharvest biocontrol research: Is it time for a new paradigm? Postharvest Biol. Technol. 52(2): 137-145. Eckert, J.W., and Ogawa, J.M. 1988. The chemical control of postharvest diseases: Deciduous fruits, berries, vegetables and root/tuber crops. Ann. Rev. Phytopathol. 26: 433-469. Eisele, T.A., and Drake, S.R. 2005. The partial compositional characteristics of apple juice from 175 apple varieties. J. Food Compost. Anal. 18(2-3): 213-221. Elad, Y. 1996. Mechanisms involved in the biological control of Botrytis cinerea incited diseases. Eur. J. Plant Pathol. 102(8): 719-732. 109 Elad, Y., Williamson, B., Tudzynski, P., and Delen, N. 2007. Botrytis: biology, pathology and control. Springer, Dordrecht, The Netherlands. El-Goorani, M.A., and Sommer, M.F. 1981. Effects of modified atmospheres of postharvest pathogens of fruits and vegetables. Hort. Rev. 3: 412-461. Ellis, M.A., and Barrat, G.J. 1983. Colonization of Delicious apple fruits by Alternaria spp. and effect of fungicide sprays on moldy-core. Plant Dis. 67(2): 150-152. Errampalli, D. 2004. Effect of fludioxonil on germination and growth of Penicillium expansum and decay in apple cvs. Empire and Gala. Crop Prot. 23(9): 811-817. Errampalli, D., Brubacher, N., and DeEll, J. 2006. Sensitivity of Penicillium expansum to diphenylamine and thiabendazole and postharvest control of blue mold with fludioxonil in ‘McIntosh’ apples. Postharvest Biol. Technol. 39(1): 101-107. Etebarian, H.R., Sholberg, P.L., Eastwell, K.C., and Sayler, R.J. 2005. Biological control of apple blue mold with Pseudomonas fluorescens. Can. J. Microbiol. 51(7): 591- 598. Etebarian, R.H., and Sholberg, P.L. 2005. Monitoring population levels of Pseudomonas fluorescens labelled with green fluorescent protein, using fluorescence microscopy and direct fluorescence scanning techniques. Can. J. Plant Pathol. 28(1): 125-130. FAOSTAT. 2009. FAO statistics division [online]. Available from http://faostat.fao.org [accessed 13 January 2010]. Fessehaie, A. De Boer, S.H., and Lévesque, A.C. 2003. An oligonucleotide array for the identification and differentiation of bacteria pathogenic on potato. Phytopathol. 93(3): 262-269. 110 Frankowski, J., Lorito, M., Scala, F., Schmid, R. Berg, G., and Bahl, H. 2001. Purification and properties of two chitinolytic enzymes of Serratia plymuthica HRO-C48. Arch. Microbiol. 176(6): 421-426. Gould, M., Nelson, L.M., Waterer, D., and Hynes, R.K. 2008. Biocontrol of Fusarium sambucinum, dry rot of potato, by Serratia plymuthica 5-6. Biocontrol Sci. Technol. 18(9): 1005-1016. Guo, L., Michailides, T.J., and Morgan, D.P. 1999. Survival of Mucor piriformis in soil of apple orchards in California. Plant Dis. 83(2): 189-193. Hancock, D.B., Martin, E.R., Mayhew, G.M., Stajich, J.M., Jewett, R., Stacy, M.A., Scott, B.L., Vance, J.M., and Scott, W.K. 2008. Pesticide exposure and risk of Parkinson's disease: A family-based case-control study. BMC Neurol. 8(6). Health Protection Agency. 2007. Identification of Enterobacteriaceae [online]. Available from www.hpa-standardmethods.org.uk/documents/bsopid/pdf/bsopid16.pdf [accessed 14 April 2008]. Heim, R., Cubitt, A.B., and Tsien, R.Y. 1995. Improved green fluorescence. Nature 373(6516): 663-664. Hirkala, D., Nelson, L., Stokes, S., and Sholberg, P. 2007. Postharvest biological control of blue and grey mold on apples. Can. J. Plant Pathol. 29(1): 84-85. Hynes, R.K., Leung, G.C., Hirkala, D.L., and Nelson, L.M. 2008. Isolation, selection and characterization of beneficial rhizobacteria from pea, lentil and chickpea grown in western Canada. Can. J. Microbiol. 54(4): 248-258. 111 Izzo, A.B., and Mazzola, M. 2009. Hybridization of an ITS-based macroarray with ITS community probes for characterization of complex communities of fungi and fungal-like protists. Mycol. Res. 113: 802-812. Janisiewicz, W., Yourman, L., Roitman, J., and Mahoney, N. 1991. Postharvest control of blue mold and grey mold of apples and pears by dip treatment with pyrrolnitrin, a metabolite of Pseudomonas cepacia. Plant Dis. 75(5): 490-494. Janisiewicz, W.J. 1988. Biocontrol of postharvest diseases of apples with antagonist mixtures. Phytopathol. 78(2): 194-198. Janisiewicz, W.J., and Korsten, L. 2002. Biological control of postharvest diseases of fruit. Annu. Rev. Phytopathol. 40: 411-441. Janisiewicz, W.J., and Jeffers, S.N. 1997. Efficacy of commercial formulation of two biofungicides for control of blue mold and gray mold of apples in cold storage. Crop Prot. 16(7): 629-633. Janisiewicz, W.J., and Marchi, A. 1992. Control of storage rots on various pear cultivars with a saprophytic strain of Pseudomonas syringae. Plant Dis. 76(6): 555-560. Janisiewicz, W.J., Leverentz, B., Conway, W.S., Saftner, R.A., Reed, A.N., and Camp, M.J. 2003. Control of bitter rot and blue mold of apples by integrating heat and antagonist treatments on 1-MCP treated fruit stored under controlled atmosphere conditions. Postharvest Biol. Technol. 29(2): 129-143. Janisiewicz, W.J., Saftner, R.A., Conway, W.S., and Yoder, K.S. 2008. Control of blue mold decay of apple during commercial controlled atmosphere storage with yeast antagonists and sodium bicarbonate. Postharvest Biol. Technol. 49(3): 374-378. 112 Janisiewicz, W.J., Tworkoski, T.J., and Sharer, C. 2000. Characterizing the mechanism of biological control of postharvest diseases on fruits with a simple method to study competition for nutrients. Phytopathol. 90(11): 1196-1200. Jiang, Y., Joyce, D.C., and Terry, L.A. 2001. 1-Methylcyclopropene treatment affects strawberry fruit decay. Postharvest Biol. Technol. 23(3): 227-232. Jijakli, M.H., and Lepoivre, P. 2004. Disease management of fruits and vegetables. In Fruit and vegetable diseases. Edited by K.G. Mukerji. Kluwer Academic Publishers, Dordrecht, The Netherlands. pp. 59-94. Jones, A.L., and Aldwinckle, H.S. 1990. Compendium of apple and pear diseases. The American Phytopathological Society, St. Paul, Minnesota. Kamensky, M., Ovadis, M., Chet, I., and Chernin, L. 2003. Soil-borne strain IC14 of Serratia plymuthica with multiple mechanisms of antifungal activity provides biocontrol of Botrytis cinerea and Sclerotinia sclerotiorum diseases. Soil Biol. Biochem. 35(2): 323-331. Kupferman, E. 2003. Controlled atmosphere storage of apples and pears. Postharvest Information Network, Tree Fruit Research and Extension Center, Washington State University, Wenatchee, WA. Kurze, S., Bahl, H., Dahl, R., and Berg, G. 2001. Biological control of fungal strawberry diseases by Serratia plymuthica HRO-C48. Plant Dis. 85(5): 529-534. Le Floch, G., Tambong, J., Vallance, J., Tirilly, Y., Lévesque, A., and Rey, P. 2007. Rhizosphere persistence of three Pythium oligandrum strains in tomato soilless culture assessed by DNA macroarray and real-time PCR. FEMS Microbiol. Ecol. 61(2): 317-326. 113 Lennox, C.L., Spotts, R.A., and Cervantes, L.A. 2003. Populations of Botrytis cinerea and Penicillium spp. on pear fruit, and in orchards and packinghouses, and their relationship to postharvest decay. Plant Dis. 87(6): 639-644. Leverentz, B., Conway, W.S., Janisiewicz, W.J., Saftner, R.A., and Camp, M.J. 2003. Effect of combining MCP treatment, heat treatment, and biocontrol on the reduction of postharvest decay of 'Golden Delicious' apples. Postharvest Biol. Technol. 27(3): 221-233. Levesque, C.A., Harlton, C.E., and de Cock, A.W.A.M. 1998. Identification of some oomycetes by reverse dot blot hybridization. Phytopathol. 88(3): 213-222. Lievens, B., and Thomma, P.H.J. 2005. Recent developments in pathogen detection arrays: implications for fungal plant pathogens and use in practice. Phytopathol. 95(12): 1374-1380. Lievens, B., Brouwer, M., Vanachter, A.C.R.C, Lévesque, A., Cammue, B.P.A., and Thomma, B.P.H.J. 2003. Design and development of a DNA array for rapid detection and identification of multiple tomato vascular wilt pathogens. FEMS Microbiol. Lett. 223(1): 119-122. Lievens, B., Brouwer, M., Vanachter, A.C.R.C, Lévesque, A.C., Cammue, B.P.A, and Thomma, B.P.H.J. 2005. Quantitative assessment of phytopathogenic fungi in various substrates using a DNA macroarray. Environ. Microbiol. 7(11): 1698-1710. Lievens, B., Claes, L., Krause, M.S., Vanachter, A.C.R.C., Cammue, B.P.A., and Thomma, B.P.H.J. 2007. Assessing populations of a disease suppressive microorganism and a plant pathogen using DNA arrays. Plant Sci. 172(3): 505-514. 114 Marcos, J.F., González-Candelas, L., and Zacarías, L. 2005. Involvement of ethylene biosynthesis and perception in the susceptibility of citrus fruits to Penicillium digitatum infection and the accumulation of defence-related mRNAs. J. Exp. Bot. 56(418): 2183-2193. Mazzola, M. 2004. Assessment and management of soil microbial community structure for disease suppression. Annu. Rev. Phytopathol. 42: 35–59. Meziane, H., Gavriel, S., Ismailov, Z., Chet, I., Chernin, L., and Höfte, M. 2006. Control of green and blue mould on orange fruit by Serratia plymuthica strains IC14 and IC1270 and putative modes of action. Postharvest Biol. Technol. 39(2): 125-133. Michailides, T.J., and Spotts, R.A. 1986. Factors effecting dispersal of Mucor piriformis in pear orchards and into the packinghouse. Plant Dis. 70(11): 1060-1063. Michailides, T.J., and Spotts, R.A. 1990. Postharvest disease of pome and stone fruits caused by Mucor piriformis in the Pacific Northwest and California. Plant Dis. 74(8): 537-543. Mikani, A., Etebarian, H.R., Sholberg, P.L., O’Gorman, D.T., Stokes, S., and Alizadeh, A. 2008. Biological control of apple grey mold caused by Botrytis mali with Pseudomonas fluorescens strains. Postharvest Biol. Technol. 48(1): 107-112. Miller, P. M. 1959. Open calyx tubes as a factor contributing to carpel discoloration and decay of apples. Phytopathol. 49(8): 520-523. Morales, H., Sanchis, V., Rovira, A., Ramos, A., and Marin, S. 2007. Patulin accumulation in apples during postharvest: Effect of controlled atmosphere storage and fungicide treatments. Food Cont. 18(11): 1443-1448. 115 Morales, H., Sanchis, V., Usall, J., Ramos, A.J., and Marín, S. 2008. Effect of biocontrol agents Candida sake and Pantoea agglomerans on Penicillium expansum growth and patulin accumulation in apples. Int. J. Food Microbiol. 122(1-2): 61-67. Morin, J.G., and Hastings, J.W. 1971. Energy transfer in a bioluminescent system. J. Cell. Physiol. 77(3): 313. Morise, H., Shimomura, O., Johnson, F., and Winant, J. 1974. Intermolecular energy transfer in the bioluminescent system of Aequorea. Biochem. 13(12): 2656–62. Müller, H., and Berg, G. 2008. Impact of formulation procedures on the effect of the biocontrol agent Serratia plymuthica HRO-C48 on Verticillium wilt in oilseed rape. Biocont. 53(6): 905-916. Nunes, C., Usall, J., Teixido, N., Fons, E., and Vinas, I. 2002. Post-harvest biological control by Pantoea agglomerans (CPA-2) on Golden Delicious apples. J. Appl. Microbiol. 92(2): 247-255. O’Gorman, D.T., Sholberg, P.L., Stokes, S.C., and Ginns, J. 2008. DNA sequence analysis of herbarium specimens facilitates the revival of Botrytis mali, a postharvest pathogen of apple. Mycol. 100(2): 227-235. Ormo, M., Cubitt, A.B., Kallio, K., Gross, L.A., Tsien, R.Y., and Remington, S.J. 1996. Crystal structure of the Aequorea victoria green fluorescent protein. Sci. 273(5280): 1392-1395. Pest Management Regulatory Agency. 2007. Available from http://www.pmra- arla.gc.ca/english/pubs/pubs-e.html [accessed 9 March 2008]. 116 Piemontese, L., Solfrizzo, M., and A. Visconti. 2005. Occurrence of patulin in conventional and organic fruit products in Italy and subsequent exposure assessment. Food Add. Contamin. 22: 437-442. Prasher, D., Eckenrode ,V., Ward, W., Prendergast, F., and Cormier, M. 1992. Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111(2): 229–33. Robideau, G.P., Caruso, F.L., Oudemans, P.V., McManus, P.S., Renaud, M.A., Auclair, M.E., Bilodeau, G.J., Yee, D., Désaulniers, N.L., DeVerna, J.W., and Lévesque, C.A. 2008. Detection of cranberry fruit rot fungi using DNA array hybridization. Can. J. Plant Pathol. 30(2): 226-240. Rosenberger, D.A. 1990. Blue mold. In Compendium of apple and pear diseases. Edited by A.L. Jones, and H.S. Aldwinkle. The American Phytopathological Society, St. Paul, Minn. pp. 54–55. Saftner, R.A, Abbott, J.A., and Conway, W.S. 2003. Effects of 1-methylcyclopropene and heat treatments on ripening and postharvest decay in Golden Delicious apples. J. Amer. Soc. Hort. Sci. 128(1): 120-127. Saiki, R.K., Scharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A., and Arnheim, N. 1985. Enzymatic amplification of β-globlin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Sci. 230(4732): 1350-1354. Sanderson, P.G., and Spotts, R.A. 1995. Postharvest decay of winter pear and apple fruit caused by species of Penicillium. Phytopathol. 85(1): 103-110. Shaner, N.C., Patterson, G.H., and Davidson, M.W. 2007. Advances in fluorescent protein technology. J. Cell. Sci. 120(24): 4247-4260. 117 Sharma, R.R., Singh, D., and Singh, R. 2009. Biological control of postharvest diseases of fruits and vegetables by microbial antagonists: a review. Biol. Control 50(3): 205-221. Shimomura, O., Johnson, F., and Saiga, Y. 1962. Extraction, purification and properties of aequorin, a bioluminescent protein from the luminous hydromedusan, Aequorea. J Cell Comp Physiol 59(3): 223–39. Sholberg, P., and Boulé, J. 2008. Interaction of powdery mildew treatments on apple microflora. In Plant-microbe interactions. Edited by E. Ait Barka and C. Clément. Research Signpost, Kerala, India. pp. 393-406. Sholberg, P., and Stokes, S. 2006. Apple packinghouse sanitation in British Columbia in 2005. Can. Plant Dis. Survey 86: 123-124. Sholberg, P., O’Gorman, D., Bedford, K., and Lévesque, C.A. 2005a. Development of a DNA macroarray for detection and monitoring of economically important apple diseases. Plant Dis. 89(11): 1143-1150. Sholberg, P.L., and Conway, W.S. 2004. Postharvest Pathology. In The Commercial Storage of Fruits, Vegetables, and Florist and Nursery Stocks. Edited by K.C. Gross, C.Y. Wang and M. Saltveit. Available from http://www.ba.ars.usda.gov/hb66/ [accessed 14 March 2010]. Sholberg, P.L., O’Gorman, D., and Bedford, K.E. 2006. Monitoring Erwinia amylovora in pome fruit orchards using reverse-dot-blot hybridization. Acta Hortic. 704: 91- 97. Sholberg, P.L., and Haag, P. D. 1996. Incidence of postharvest pathogens of stored apples in British Columbia. Can. J. Plant Pathol. 18(1): 81-85. 118 Sholberg, P.L., Bedford, K.E., and Stokes, S. 2003. Effect of preharvest application of cyprodinil on postharvest decay of apples caused by Botrytis cinerea. Plant Dis. 87(9): 1067-1071. Sholberg, P.L., Bedford, K.E., and Stokes, S. 2005b. Sensitivity of Penicillium spp. and Botrytis cinerea to pyrimethanil and its control of blue and gray mold of stored apples. Crop Prot. 24(2): 127-134. Sholberg, P.L., Harlton, C., Haag, P., Levesque, C.A., O’Gorman, D., and Siefert, K. 2005c. Benzimidazole and diphenylamine sensitivity and identity of Penicillium spp. that cause postharvest blue mold of apples using beta-tubulin gene sequences. Postharvest Biol. Technol. 36(1): 41-49. Smock, R. 1979. Controlled atmosphere storage of fruits. Hortic. Rev. 1: 301-333. Spadaro, D., and Gullino, M.L. 2004. State of the art and future prospects of the biological control of postharvest fruit diseases. Int. J. Food Microbiol. 91(2): 185- 194. Spotts, R.A., Cervantes, L.A., and Mielke, E.A. 1999. Variability in postharvest decay among apple cultivars. Plant Dis. 83(11): 1051-1054. Spotts, R.A., Serdani, M., Wallis, K.M., Walter, M., Harris-Virgin, T., Spotts, K., Sugar, D., Xiao, C.L., and Qu, A. 2009. At-harvest prediction of grey mould risk in pear fruit in long-term cold storage. Crop Prot. 28(5): 414-420. Stockwell, V.O., and Stack, J.P. 2007. Using Pseudomonas spp. for integrated biological control. Phytopathol. 97(2): 244-249. 119 Stokes, S.C., Sholberg, P.L., Nelson, L.M., and Sergeeva, E. 2006. The screening of bacterial isolates for biological control of blue and gray molds of apples. Can. J. Plant Pathol. 28(1): 163. Suarez, A., Güttler, A., Strätz, M., Staendner, L.H., Timmis, K.N., and Guzmán, C.A. 1997. Green fluorescent protein-based reporter systems for genetic analysis of bacteria including monocopy applications. Gene 196(1-2): 69-74. Tambong, J.T., de Cock, A.W.W.M., Tinker, A., and Lévesque, C.A. 2006. Oligonucleotide array for identification and detection of Pythium species. Appl. Environ. Microbiol. 72(4): 2691-2706. Teixidó, N., Usall, J., and Vinas, I. 1999. Efficacy of preharvest and postharvest Candida sake biocontrol treatments to prevent blue mould on apples during cold storage. Int. J. Food Microbiol. 509(3): 203-210. Teixidó, N., Usall, J., Magan, N., and Vinas, I. 1999. Microbial population dynamics on Golden Delicious apples from bud to harvest and effect of fungicide applications. Ann. Appl. Biol. 134(1): 109-116. Terry, L.A., and Joyce, D.C. 2004. Elicitors of induced disease resistance in postharvest horticultural crops: a brief review. Postharvest Biol. Technol. 32(1): 1-13. Tian, S., Fan, Q., Xu, Y., and Liu, H. 2002. Biocontrol efficacy of antagonist yeasts to gray mold and blue mold on apples and pears in controlled atmospheres. Plant Dis. 86(8): 848-853. Tombolini, R., Unge, A., Davey, M.E., de Bruijn, F.J., and Jansson, J.K. 1997. Flow cytometric and microscopic analysis of GFP-tagged Pseudomonas fluorescens bacteria. FEMS Microbiol. Ecol. 22(1): 17-28. 120 van Loon, L.C. 2007. Plant responses to plant growth-promoting rhizobacteria. Eur. J. Plant. Pathol. 119(3): 243-254. Vikram, A., Prithiviraj, B., Hamzehzarghani, H., and Kushalappa, A.C. 2004. Volatile metabolite profiling to discriminate diseases of McIntosh apple inoculated with fungal pathogens. J. Sci. Food Agric. 84(11): 1333-1340. Walter, M., Boyd-Wilson, K.S.H., Stanley, J., Harris-Virgin, P., Morgan, C., and Pyke, N.B. 1997. Epidemiology of Botrytis cinerea in boysenberry (Rubus spp.). Proc. 50th N.Z. Plant Protection Conf. 50: 93-100. Ward, W.W., Cody, C.W., Hart, R.C., and Cormier, M.J. 1980. Spectrophotometric identity of the energy transfer chromophores in Renilla and Aequorea green- fluorescent proteins. Photochem. Photobiol. Rev. 31: 611-615. Watanabe, M., and H. Shimizu. 2005. Detection of patulin in apple juices marketed in the Tohoku district, Japan. J. Food Prot. 68: 610-612. Watanabe, S., Kumakura, K., Izawa, N., Nagayama, K., Mitachi, T., Kanamori, M., Teraoka, T., and Arie, T. 2007. Mode of action of Trichoderma asperellum SKT-1, a biocontrol agent against Gibberella fujikuroi. J. Pestic. Sci. 32(3): 222-228. Watkins, C.B. 2006. The use of 1-methylcyclopropene (1-MCP) on fruits and vegetables. Biotechnol. Adv. 24(4): 389-409. Watkins, C.B. 2008. Overview of 1-methylcyclopropene trials and uses for edible horticultural crops. HortSci. 43(1): 86-94. White, T. J., Burns, T.D., Lee, S.B., and Taylor, J.W. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics,. In PCR protocols: a 121 guide to methods and applications. Edited by M. A. Innis, D. H. Helfand, J. J. Sninsky, and T. J. White. Academic Press, Inc., San Diego, CA. pp. 315-322. Wilson, C.L., and Wisniewski, M.E. 1989. Biological control of postharvest diseases of fruits and vegetables: an emerging technology. Annu. Rev. Phytopathol. 27:425- 411. Yahia, E.M. 2009. Introduction. In Modified and controlled atmospheres for the storage, transportation, and packaging of horticultural commodities. Edited by E.M. Yahia. CRC Press, Boca Raton, FL. pp. 1-16. Zhou, T., Chu, C.L., Liu, W.T., and Schaneider, E., 2001. Postharvest control of blue mold and grey mold on apples using isolates of Pseudomonas syringae. Can. J. Plant Pathol. 23(3): 246-252. 122 APPENDIX Appendix A Table A1. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 2628.336 4.585 0.0619 B. cinerea 2 384.596 1.857 0.2357 M. piriformis 2 573.896 0.429 0.6697 Table A2. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 327.860 0.134 0.8766 B. cinerea 2 66.667 1.0 0.4053 M. piriformis 2 5625.402 28.266 0.0001*** Table A3. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2007 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 741.125 1.119 0.3681 B. cinerea 2 55.512 0.567 0.5863 M. piriformis 2 683.272 1.018 0.3994 123 Table A4. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 75.335 0.034 0.9668 B. cinerea 2 488.562 0.579 0.5800 M. piriformis 2 203.247 0.438 0.6583 Table A5. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 1702.334 4.590 0.0533 B. cinerea 2 148.328 12.683 0.0047** M. piriformis 2 1336.256 4.316 0.0601 Table A6. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in the early, mid or late growing season in 2008 for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Growing season section (early, mid, late) Category DF Type III SS F-value p-value P. expansum 2 182.052 0.171 0.8461 B. cinerea 2 76.432 0.262 0.7753 M. piriformis 2 503.152 1.681 0.2398 124 Table A7. Key data extracted from the one-way ANOVA tables for the % P. expansum, B. cinerea or M. piriformis detected from I-rod and plant tissue samples as indicated by DNA macroarrays in all growing season sections for Fields 3, 12 and the Kiran and Reekie orchards. Main Effect Detected pathogens (P. expansum, B. cinerea or M. piriformis) Category DF Type III SS F-value p-value 2007 growing season 2 896.063 4.116 0.0538 2008 growing season 2 1022.252 3.222 0.0880 2007/08 growing season 2 454.769 19.248 0.0194* Table A8. Key data extracted from the two-way ANOVA table for infection severity data collected from Jonagold apples grown in Field 3 and analyzed for natural disease incidence after harvest in 2007. DF Type III SS F-value p-value Main Effects Incubation Period 1 3639.591 13.081 0.0004*** 1-MCP+/- 1 3260.433 11.718 0.0008*** Interaction Incubation period x MCP+/- 1 1008.275 3.624 0.0584 Table A9. Key data extracted from the two-way ANOVA table for infection severity data collected from Gala apples grown in Field 12 analyzed for natural disease incidence after harvest in 2007. DF Type III SS F-value p-value Main Effects Incubation Period 1 13.767 36.073 0.0000*** 1-MCP+/- 1 0.175 0.458 0.4995 Interaction Incubation period x MCP+/- 1 0.286 0.750 0.3876 125 Table A10. The effect of incubation period (1, 2, 4 and 6 months) on natural disease incidence and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3, Gala apples grown in Field 12 and Kiran and Red Delicious apples grown in Reekie and analyzed for natural disease incidence from harvest after 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value Field 3 3 61.3201 74.6611 0.0000*** Field 12 3 79.6754 276.1476 0.0000*** Kiran 3 0.8366 3.2372 0.0268* Reekie 3 11.1988 17.9952 0.0000*** Table A11. Key data extracted from the one-way ANOVA table constructed from infection severity data collected from Fields 3, 12, Kiran and Reekie analyzed for natural disease incidence for apples stored in CA storage for 4.5 months from harvest 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Category DF Type III SS F-value p-value 1 month storage 3 0.9484 1.5636 0.1195 2 months storage 3 10.0842 7.1456 0.0001*** 4 months storage 3 8863.9197 27.1892 0.0000*** 6 months storage 3 40909.7401 48.0806 0.0000*** Table A12. Key data extracted from the one-way ANOVA table constructed from infection severity data collected from Fields 3, 12, Kiran and Reekie analyzed for natural disease incidence for apples stored in CA storage for 4.5 months from harvest 2008. Category DF Type III SS F-value p-value Field 3 29.2208 55.0013 0.0000*** 126 Table A13. Key data extracted from the two-way ANOVA table for infection severity data collected from Jonagold apples grown in Field 3 analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2007.. Table A14. Key data extracted from the two-way ANOVA table for infection severity data collected from Gala apples grown in Field 12 and analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2007.. Table A15. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for natural disease incidence in combination with bacterial isolates from harvest 2007. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 2310.748 18.214 0.0005*** 1100-6 1 1674.926 11.450 0.0033** 1-112 1 7.358 0.243 0.6279 2-28 1 1542.633 26.057 0.0001*** 4-6 1 0.020 1.49e-4 0.9904 6-25 1 130.302 0.966 0.3387 DF Type III SS F-value p-value Main Effects Biological Control 1 0.900 1.192 0.3182 1-MCP+/- 1 0.342 2.262 0.1355 Interaction Biological Control x MCP+/- 5 0.332 0.439 0.8202 DF Type III SS F-value p-value Main Effects Biological Control 1 3039.868 5.773 0.0001*** 1-MCP+/- 1 3435.342 32.620 0.0000*** Interaction Biological Control x MCP+/- 5 2779.181 0.439 0.0002*** 127 Table A16. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are constructed from IS results collected from Field 12 Gala apples and analyzed for natural disease incidence in combination with bacterial isolates from harvest 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 624.112 1.329 0.2659 1-MCP- 5 5194.937 8.902 0.0000*** Table A17. Key data extracted from the two-way ANOVA table from infection severity data collected from Jonagold apples grown in Field 3 and inoculated with P. expansum after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 5 32.862 61.071 0.0000*** 1-MCP+/- 1 0.015 0.413 0.7077 Interaction Biological Control x MCP+/- 5 5.626 1.125 0.0000*** Table A18. The effect of 1-MCP in six separate biological control categories and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with P. expansum and bacterial isolates after harvest in 2007. Individual one- way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 0.0420 1.775 0.1994 1100-6 1 1.4724 7.422 0.0139* 1-112 1 0.0425 0.215 0.6482 2-28 1 0.0872 2.057 0.1687 4-6 1 0.0329 0.302 0.5893 6-25 1 3.9641 52.958 0.0000*** 128 Table A19. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are constructed from infection severity results collected from Jonagold grown in Field 3 analyzed for disease incidence when inoculated with P. expansum and bacterial isolates from harvest in 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 16.7121 31.798 0.0000*** 1-MCP- 5 21.7763 39.548 0.0000*** Table A20. Key data extracted from the two-way ANOVA table from infection severity data collected from Jonagold apples grown in Field 3 and inoculated with B. cinerea after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 5 21.222 20.760 0.0000*** 1-MCP+/- 1 15.998 78.249 0.0000*** Interaction Biological Control x MCP+/- 5 13.841 13.534 0.0000*** Table A21. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 20077. Individual one- way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 9.706 45.191 0.0000*** 1100-6 1 0.3931 1.793 0.1973 1-112 1 0.2607 1.060 0.3169 2-28 1 13.3139 80.502 0.0000*** 4-6 1 5.9336 29.726 0.0000*** 6-25 1 0.2310 1.271 0.2743 129 Table A22. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 33.7372 53.215 0.0000 1-MCP- 5 1.3253 0.940 0.4630 Table A23. Key data extracted from the two-way ANOVA table constructed from infection severity data collected from Jonogold apples grown in Field 3 and inoculated with M. piriformis after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 5 14.346 10.660 0.0000*** 1-MCP+/- 1 0.784 2.913 0.0908 Interaction Biological Control x MCP+/- 5 7.164 5.323 0.0002*** Table A24. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data is are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 0.0020 0.0181 0.8943 1100-6 1 0.6318 1.3963 0.2527 1-112 1 0.0194 0.0729 0.7902 2-28 1 0.6870 4.3055 0.0526 4-6 1 6.5402 13.1253 0.0019** 6-25 1 0.0677 0.5298 0.4761 130 Table A25. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 1.1507 0.9683 0.4455 1-MCP- 5 20.3591 13.5439 0.0000 Table A26. Key data extracted from the two-way ANOVA table constructed from infection severity data collected from Gala apples from Field 12 and inoculated with P. expansum after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 5 192.271 1.940 0.0936 1-MCP+/- 1 182.887 9.226 0.0030** Interaction Biological Control x MCP+/- 5 164.951 1.664 0.1494 Table A27. Key data extracted from the two-way ANOVA table from infection severity data collected from Gala apples grown in Field 12 and inoculated with M. piriformis after harvest in 2007. DF Type III SS F-value p-value Main Effects Biological Control 4 3.015 1.729 0.1505 1-MCP+/- 1 6.578 1.508e-5 0.9969 Interaction Biological Control x MCP+/- 4 8.469 4.856 0.0014*** 131 Table A28. The effect of 1-MCP in six separate biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. Individual one- way ANOVAs were determined for each category and combined into a single table. Main Effect 1-MCP+/- Category DF Type III SS F-value p-value No biocontrol 1 0.0242 0.0487 0.8278 1-112 1 1.9035 4.7987 0.0419* 2-28 1 6.0306 14.5425 0.0013** 4-6 1 0.4423 1.1814 0.2914 6-25 1 0.0683 0.1370 0.7156 Table A29. The effect of the biological control treatments in two categories, 1-MCP+ and 1-MCP- and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2007. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value 1-MCP+ 5 7.0326 3.9878 0.0075** 1-MCP- 5 4.4516 2.5814 0.0498* Table A30. Key data extracted from the two-way ANOVA table from infection severity data collected from Jonagold apples grown in Field 3 and analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2008. DF Type III SS F-value p-value Main Effects Biological Control 5 1.2214 0.8840 0.4926 Incubation Period 3 105.9743 127.8366 0.0000*** Interaction Biological Control x Incubation Period 15 14.3647 3.4656 0.0000*** 132 Table A31. Key data extracted from the two-way ANOVA table from infection severity data collected from Gala apples grown in Field 12 and analyzed for natural disease incidence in combination with bacterial isolates after harvest in 2008. DF Type III SS F-value p-value Main Effects Biological Control 5 0.5763 0.8804 0.4950 Incubation Period 3 149.6138 380.9182 0.0000*** Interaction Biological Control x Incubation Period 15 3.6138 1.8400 0.0309* Table A32. The effect of the incubation period on biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3 and analyzed for natural disease incidence when inoculated the bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value 2 months 5 5.5753 4.3605 0.0021** 4 months 5 5067.4972 3.9609 0.0039** 6 months 5 15646.1017 5.2559 0.0005*** Table A33. The effect of the incubation period on biological control treatments and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Gala apples grown in Field 12 and analyzed for natural disease incidence when inoculated the bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value 2 months 5 2.1850 2.8870 0.0221* 4 months 5 1876.4999 2.5715 0.0370* 6 months 5 743.3109 0.7762 0.5782 133 Table A34. Key data extracted from the two-way ANOVA table from infection severity data collected from Jonagold (Field 3), Gala (Fields 12 and Kiran) and Red Delicious (Reekie) apples after 4.5 months in CA storage and analyzed for natural disease incidence following inoculation with bacterial isolates after harvest in 2008. DF Type III SS F-value p-value Main Effects Biological Control 5 3.3247 5.4749 0.0001*** Field 3 12.3636 33.9323 0.0000*** Interaction Biological Control x Field 15 5.8261 3.1980 0.0000*** Table A35. The effect of the biological control treatments on infection severity and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from Jonagold apples grown in Field 3, Gala apples grown in Field 12 and Kiran and Red Delicious apples grown in Reekie after 4.5 months in CA stroage and analyzed for natural disease incidence when inoculated the bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Treatment Category DF Type III SS F-value p-value Field 3 5 7.7543 4.8907 0.0009*** Field 12 5 0.1933 1.1827 0.3298 Kiran 5 0.1136 2.2402 0.0633 Reekie 5 1.0897 1.7312 0.1432 Table A36. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with P. expansum and bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value Field 3, Jonagold 5 3041.4892 6.8050 0.0001*** Field 12, Gala 5 401.2654 3.1239 0.0151* 134 Table A37. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with B. cinerea and bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value Field 3, Jonagold 5 5981.4974 4.9960 0.0008*** Field 12, Gala 5 1962.7067 6.4391 0.0001*** Table A38. The effect of the biological control treatments in two categories, Field 3 Jonagold apples and Field 12 Gala apples and the resulting key data extracted from the one-way ANOVA results. Data are from IS results collected from apples analyzed for disease incidence when inoculated with M. piriformis and bacterial isolates after harvest in 2008. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Biological Control Category DF Type III SS F-value p-value Field 3, Jonagold 5 2115.5372 4.7271 0.0012** Field 12, Gala 5 3.0721 4.1326 0.0030** Table A39. Growth rate (µ) comparisons of P. fluorescens isolates 4-6-wild-type and 4- 6-gfp and the resulting key data extracted from the one-way ANOVA table. Main Effect Bacterial Isolate 4-6-wt vs 4-6-gfp Category DF Type III SS F-value p-value Growth Rate (µ) 1 0.0022 0.6492 0.4656 135 Table A40. Key data extracted from the two-way ANOVA table constructed from cell densities (CFU/ml) of P. fluorescens isolates 4-6-wild-type and 4-6-gfp stored for 55 d at 1°C in air. DF Type III SS F-value p-value Main Effects Bacterial strain 1 50.5367 670.6401 0.0000*** Time 5 72.3075 191.9023 0.0000*** Interaction Bacterial strain x Time 5 72.7250 171.9775 0.0000*** Table A41. Cell density comparisons determined by dilution plating method of P. fluorescens isolates 4-6-gfp and 4-6-gfp + P. expansum stored for 55 d at 1°C in air and the resulting key data extracted from the one-way ANOVA table. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Bacterial Isolate 4-6-gfp vs 4-6-gfp + P. expansum Category DF Type III SS F-value p-value 0 d storage 1 0.1061 2.3329 0.1610 10 d storage 1 8.4137e-4 0.0157 0.9024 20 d storage 1 0.5121 4.0261 0.0758 30 d storage 1 2.3762 36.5429 0.0005*** 45 d storage 1 16.6408 120.1450 0.0000*** 55 d storage 1 4.418e13 8.8646 0.0206* Table A42. Infection severity comparisons of P. fluorescens isolates 4-6-wild-type and 4-6-gfp stored for 55 d at 1°C in air and the resulting key data extracted from the one-way ANOVA table. Individual one-way ANOVAs were determined for each category and combined into a single table. Main Effect Bacterial Isolate 4-6-gfp vs 4-6-gfp + P. expansum Category DF Type III SS F-value p-value 30 d storage 1 21.6680 7.6629 0.0040** 40 d storage 1 46.8838 14.1739 0.0093** 45 d storage 1 18.9113 1.3841 0.2840 55 d storage 1 13.9568 0.5615 0.4820