EFFECT OF AZOXYSTROBIN ON TURFGRASS PHYLLOSPHERE

MICROBIAL POPULATIONS AND DISEASE ENHANCEMENT

A Thesis

Presented to

The Faculty of Graduate Studies

of

The University of Guelph

by

DANIEL BENEDETTO

In partial fulfilment of requirements

for the degree of

Master of Science

January, 2008

© Daniel Benedetto, 2008 Library and Bibliotheque et 1*1 Archives Canada Archives Canada Published Heritage Direction du Branch Patrimoine de I'edition

395 Wellington Street 395, rue Wellington Ottawa ON K1A0N4 Ottawa ON K1A0N4 Canada Canada

Your file Votre reference ISBN: 978-0-494-41797-3 Our file Notre reference ISBN: 978-0-494-41797-3

NOTICE: AVIS: The author has granted a non­ L'auteur a accorde une licence non exclusive exclusive license allowing Library permettant a la Bibliotheque et Archives and Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par Plntemet, prefer, telecommunication or on the Internet, distribuer et vendre des theses partout dans loan, distribute and sell theses le monde, a des fins commerciales ou autres, worldwide, for commercial or non­ sur support microforme, papier, electronique commercial purposes, in microform, et/ou autres formats. paper, electronic and/or any other formats.

The author retains copyright L'auteur conserve la propriete du droit d'auteur ownership and moral rights in et des droits moraux qui protege cette these. this thesis. Neither the thesis Ni la these ni des extraits substantiels de nor substantial extracts from it celle-ci ne doivent etre imprimes ou autrement may be printed or otherwise reproduits sans son autorisation. reproduced without the author's permission.

In compliance with the Canadian Conformement a la loi canadienne Privacy Act some supporting sur la protection de la vie privee, forms may have been removed quelques formulaires secondaires from this thesis. ont ete enleves de cette these.

While these forms may be included Bien que ces formulaires in the document page count, aient inclus dans la pagination, their removal does not represent il n'y aura aucun contenu manquant. any loss of content from the thesis. Canada ABSTRACT

EFFECT OF AZOXYSTROBIN ON TURFGRASS PHYLLOSPHERE MICROBIAL POPULATIONS AND DISEASE ENHANCEMENT

Daniel Benedetto Advisor: University of Guelph, 2008 Professor T. Hsiang

This study documents dollar spot enhancement months after application of azoxystrobin to stolonifera and . One possible cause of disease enhancement is the perturbation of foliar microbial populations. Dilution plating showed that foliar microbial populations on the grasses treated with azoxystrobin were reduced during spring and summer. This correlated with a five-fold increase of dollar spot incidence. Cultural morphology after 9 days distinguished 11 morphotypes, but longer culturing revealed 26 submorphotypes among yeasts, and filamentous fungi. DNA-RFLP revealed 111 ribotypes and rDNA-sequencing revealed 94 species from 450 culturable isolates.

While identification by morphological characteristics was useful for filamentous fungi, molecular analysis was required for yeasts and bacteria. Thirty-eight microbial species were present only on untreated plots, while 24 species were only associated with azoxystrobin-treated plots. Reduction in microbial populations and absence or gain of some species due to azoxystrobin may help explain increases in dollar spot incidence. ACKNOWLEDGEMENTS I am very grateful with my advisor Dr. Tom Hsiang for giving me the opportunity to obtain a Master's degree. His guidance, patience and expertise were key components that encourage me to keep learning and working through my studies. I particularly appreciated his friendly suggestions, advice and understanding. I also extend my thanks to the members of my advisory committee, Dr. P. Godwin and Dr. M. Habash for their suggestions during this study and to the members of the examination committee Dr. G. Barron and Dr. G

Stephenson.

I would like to thank John McLaughlin, Lynn Tian, Champa Weijekoon and

Xie Weilong for their assistance, useful advice and technical support offered unconditionally throughout this project. Special thanks to Alejandra Cortes for her support and charisma that cheer me up on through difficult times. Thanks to the people of Guelph Turfgrass Institute for allowing me to perform my experiments and prompt support.

I would like to give my love and thanks to my mother for her endless love and encouragement and my father for giving me invaluable help and support. To my future wife Xiaoting (Cathreen) Wei, I thank you for your continuous love, support, encouragement and the sacrifices you have made. I thank you for everything.

I TABLE OF CONTENTS

ACKNOWLEDGEMENTS I

TABLE OF CONTENTS II

LIST OF TABLES VI

LIST OF FIGURES VIII

LIST OF APPENDICES XIII

LIST OF ABBREVIATIONS AND ACRONYMS XIV

CHAPTER 1. LITERATURE REVIEW 1

1.1 FUNGICIDES AND TURFGRASS 1

1.2 STROBILURINS 3

1.2.1 History 4

1.2.2 Structure 5

1.2.3 Mode of action 6

1.2.4 Fungicide resistance 7

1.2.5 Environment persistence and animal safety 7

1.3 DOLLAR SPOT 9

1.3.1 Hosts and symptoms 10

1.3.2 Causal agent 11 1.3.2.1 11 1.3.2.2 Morphology 12 1.3.2.3 Control 13 1.3.3 Dollar spot and fungicides 15

1.4 PHYLLOSPHERE 17

1.4.1 Microbial Biodiversity 18 1.4.1.1 Bacteria phyllosphere communities 23 1.4.1.2 Yeast phyllosphere communities 24

II 1.4.1.3 Fungal phyllosphere communities 25

1.5 ASSESSMENT OF MICROBIAL DIVERSITY 26

1.5.1 Traditional approaches 26

1.5.2 Molecular approach 27

1.6 HYPOTHESES AND OBJECTIVES 32

1.6.1 Hypotheses 32

1.6.2 Objectives 32

CHAPTER 2. EFFECT OF AZOXYSTROBIN ON DOLLAR SPOT DISEASE

DEVELOPMENT 44

2.1 INTRODUCTION 44

2.2 MATERIALS AND METHODS 47 2.2.1 Study site 47 2.2.2 Field Experiments 50 2.2.2.1 Trials 51 2.2.2.2 Weather data 52

2.2.3 Statistic analyses 52

2.3 RESULTS 53

2.3.1 Weather data 53

2.3.2 Plant injury 54 2.3.2.1 Native sand green multi-season trial of 2005-2006 54 2.3.2.2 Poa pratensis multi-season trial of 2005-2006 55 2.3.2.3 Native sand fairway multi-season trial of 2006-2007 56 2.3.2.4 Pathology green multi-season trial of 2006-2007 57

2.3.3 Dollar spot 58 2.3.3.1 Native sand fairway multi-season trial of 2005-2006 58 2.3.3.2 Poa pratensis multi-season trial of 2005-2006 58 2.3.3.3 Native sand green multi-season trial of 2006-2007 58 2.3.3.4 Pathology green multi-season trial of 2006-2007 59 2.3.3.5 Summer plots of 2006 60

2.4 DISCUSSION 60

III CHAPTER 3. EFFECT OF AZOXYSTROBIN ON THE PHYLLOSPHERE MICROBIAL COMMUNITIES OF TURFGRASSES 93

3.1 INTRODUCTION 93

3.2 MATERIALS AND METHODS 96

3.2.1 Study site 96

3.2.2 Sample collection 97

3.2.3 Microbial screening 98 3.2.3.1 Leaf washings 98 3.2.3.2 Media preparation 98 3.2.3.3 Microbial pre-screening 99 3.2.3.4 Microbial screening 100

3.2.4 Leaf weight 101

3.2.5 Colony counting 102

3.2.6 Statistic analyses 104

3.3 RESULTS 105

3.3.1 Green leaves 107

3.3.2 Yellow leaves 110

3.4 DISCUSSION 112

CHAPTER 4. RESTRICTION FRAGMENT LENGTH POLYMORPHISMS AND SEQUENCE ANALYSES OF MICROBIAL ISOLATES FROM TURFGRASS PHYLLOSPHERES 132

4.1 INTRODUCTION 132

4.2 MATERIALS AND METHODS 138

4.2.1 Microbial isolation 138 4.2.1.1 Leaf samples 138 4.2.1.2 Epiphyte isolation 139 4.2.1.3 Endophyte isolation 141

4.2.2 DNA extraction 142 4.2.2.1 Epiphytes 142 4.2.2.2 Endophytes 144

4.2.3 DNA concentration 145

IV 4.2.4 PCR amplifications 146 4.2.4.1 Internally transcribed spacer (ITS) amplification for fungi 146 4.2.4.2 Ribosomal 16S amplification for bacteria 146 4.2.4.3 Restriction Fragment Length Polymorphism (RFLP) 147 4.2.4.4 Restriction enzymes digestions 147 4.2.4.5 Band recording and analysis 149

4.2.5 DNA sequencing and alignment 150

4.3 RESULTS 151

4.3.1 Epiphyte collection 152

4.3.2 Restriction Fragment Length Polymorphism 153 4.3.2.1 Bacterial RFLP 154 4.3.2.2 Yeast RFLP 154 4.3.2.3 Fungal RFLP 155

4.3.3 Sequencing 156 4.3.3.1 Bacteria 156 4.3.3.2 Yeast 158

4.3.3.3 Filamentous fungi 159

4.4 DISCUSSION 161

CHAPTER 5. GENERAL DISCUSSION 191

REFERENCES 201

LIST OF APPENDICES 220

v LIST OF TABLES Table 1.1. Common phyllosphere bacteria of grasses. 34 Table 1.2. Common phyllosphere yeast of grasses. 35 Table 1.3. Common phyllosphere filamentous fungi of grasses. 36 Table 3.1. Colony forming units (CFU) per gram of leaves of microbial morphotypes obtained from the native sand fairway multi-season trial of 2005-2006 at the Guelph Turfgrass Institute, Guelph, Ontario. 118 Table 3.2. Colony forming units (CFU) per gram of leaves of microbial morphotypes obtained from the Poa pratensis multi-season trial of 2005-2006 at the Guelph Turfgrass Institute, Guelph, Ontario. 119 Table 3.3. Colony forming units (CFU) per gram of leaves of microbial morphotypes obtained from the Agrostis stolonifera pathology green multi-season trial of 2006-2007 at the Guelph Turfgrass Institute, Guelph, Ontario. 120 Table 3.4. Comparisons between microbial populations on green leaves of treated and untreated samples of Agrostis stolonifera of the native sand fairway multi-season trial of 2005-2006. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. 121 Table 3.5. Comparisons between microbial populations on green leaves of treated and untreated samples of the Poa pratensis multi-season trial of 2005-2006. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. 122 Table 3.6. Comparisons between microbial populations on green leaves of treated and untreated samples of the Agrostis stolonifera pathology green multi-season trial of 2006-2007. Treatment consisted of a single application of Heritage MAXX® at 12 g azoxystrobin per 100 m2 on 25 November 2006. 123 Table 3.7. Comparisons between microbial populations on yellow leaves of treated and untreated samples of Agrostis stolonifera native sand fairway multi-season trial of 2005-2006. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. 124 Table 3.8. Comparisons between microbial populations on yellow leaves of treated and untreated samples of the Poa pratensis multi-season trial of 2005-2006. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. 125 Table 3.9. Comparisons between microbial populations on yellow leaves of treated and untreated samples of the Agrostis stolonifera pathology

VI green multi-season trial of 2006-2007. Treatment consisted of a single application of Heritage MAXX® at 12 g azoxystrobin per 100 m2 on 25 November 2006. 126 Table 4.1. Numbers of isolates obtained from leaves collected from the Agrostis stolonifera native sand fairway of 2005-2006, Poa pratensis fringe of 2005-2006 and A stolonifera pathology green of 2006-2007. 168 Table 4.2. Number of isolates per morphotypes obtained from leaves collected from the Agrostis stolonifera native sand fairway of 2005-2006, Poa pratensis fringe of 2005-2006 and A. stolonifera pathology green of 2006-2007. RFLP and sequencing refer to the number of isolates used in these analyses and the number in parentheses refers to ribotypes and species. 169 Table 4.3. Bacterial morphotypes and highest matching species with analysis of 16S sequences. 170 Table 4.4. Bacterial species, submorphotype, and number of isolates from azoxystrobin-treated and untreated green or yellow leaves of Agrostis stolonifera as revealed by sequence analysis of DNA amplified with the primer pair F0/PRUN518 172 Table 4.5. List of microbial species found exclusively on each type of leaf. Isolates were obtained from green and yellow leaves collected from the Agrostis stolonifera pathology green either treated with azoxystrobin or untreated. 174 Table 4.6. List of microbial species found exclusively on each type of leaf. Isolates were obtained from green and yellow leaves collected from the Agrostis stolonifera pathology green either treated with azoxystrobin or untreated. 175 Table 4.7. Yeast submorphotypes and highest matching species with analysis of ITS sequences. 177 Table 4.8. Yeast species, submorphotype, and number of isolates from azoxystrobin-treated and untreated green or yellow leaves of Agrostis stolonifera as revealed by sequence analysis of DNA amplified with the primer pair ITS1/ITS4. 178 Table 4.9. Filamentous fungal morphotypes and highest matching species with analysis of ITS sequences. 179 Table 4.10. Filamentous fungal species, submorphotype, and number of isolates from azoxystrobin-treated and untreated green or yellow leaves of Agrostis stolonifera as revealed by sequence analysis of DNA amplified with the primer pair ITS1/ITS4. 181

VII LIST OF FIGURES Figure 1.1. General structure of strobilurins (Adapted from Huang et al., 2007). 37 Figure 1.2. Dollar spot patch caused by homoeocarpa on closely mowed Agrostis stolonifera, next to a Canadian dollar coin. This picture was taken July 2007 at the Guelph Turfgrass Institute, Guelph, Ontario. 38 Figure 1.3. Dollar spot patch on Poa pratensis. The picture was taken July 2006 at the Guelph Turfgrass Institute, Guelph, Ontario. 39 Figure 1.4. Dollar spot hourglass lesion on Poa pratensis. (Courtesy of T. Hsiang). 40 Figure 1.5. Web-like mycelium of Sclerotinia homoeocarpa on Agrostis stolonifera present with morning dew. (Courtesy of T. Hsiang). 41 Figure 1.6. A culture of Sclerotinia homoeocarpa grown at 25C for 10 days on potato dextrose agar plate. (Courtesy of A. Liao). 42 Figure 1.7. Eukaryotic ribosomal DNA (rDNA) gene cassette showing the Small Subunit (SSU), Large Subunit (LSU), Internal Transcribers Spacers (ITS) and Intergenic spacer (IGS). Gray boxes represent highly conserved regions. B) Prokaryotic ribosomal gene showing the 23S, 16S and 5S subunits closely encoded. (Redrawn from Soltis and Soltis, 1998). 43 Figure 2.1. Plot layout for trials at the Guelph Turfgrass Institute, Guelph, Ontario. All trials were set up on Agrostis stolonifera, except for the fringe of the upper green which has Poa pratensis. (a) Native sand fairway multi-season trial of 2005-2006, (b) P. pratensis multi-season trial of 2005-2006, (c) pathology green multi-season trial of 2006-2007, (d) native sand fairway plots multi-season trial of 2006-2007, and (e) summer pathology green plots of 2006 71 Figure 2.2. Plot layout for all trials mentioned in Figure 2.3. Even numbered plots were treated with Heritage®. Odd numbered plots were untreated. 72 Figure 2.3. Temperature and precipitation from 1 October 2005 to 1 December 2006: (a) maximum, minimum and weekly average temperatures recorded at the Waterloo Wellington weather station (Climate ID 6149388), (b) visual estimation snow depth (cm) at the Guelph Turfgrass Institute, Ontario, and (c) total precipitation (mm) recorded at the Waterloo University weather station, Ontario. 73 Figure 2.4. Temperature and precipitation from 1 October 2006 to 1 October 2007: (a) Maximum, minimum and weekly average temperatures recorded at the Waterloo Wellington weather station (Climate ID 6149388), (b) snow depth (cm) measured with a sonic range sensor

VIII and (c) total precipitation (mm) recorded at the Waterloo University weather station, Ontario. 74 Figure 2.5. Native sand fairway trial of 2005-2006 on 28 November 2005. 75 Figure 2.6. Native sand fairway trial of 2005-2006 on 30 March 2006 after snow cover melted on 23 March 2006. 76 Figure 2.7. Plant injury of the Agrostis stolonifera native sand fairway, multi- season trial of 2005-2006. Plant injury was evaluated weekly to monthly from 13 March 2006 until 20 October 2006, and was assessed as percentage area with yellow to straw color grass (produced by biotic and abiotic sources), based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. 77 Figure 2.8. Poa pratensis multi-season trial of 2005-2006 on 28 November 2005. 78 Figure 2.9. Poa pratensis multi-season trial of 2005-2006 on 30 March 2006 after snow coved melted on 23 March 2006. 79 Figure 2.10. Plant injury of the Poa pratensis multi-season trial of 2005-2006. Plant injury was evaluated weekly to monthly from 13 March 2006 until 20 October 2006, and was assessed as percentage area with yellow to straw color grass (produced by biotic and abiotic sources), based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. 80 Figure 2.11. Native sand fairway trial of 2006-2007 on 23 November 2006. 81 Figure 2.12. Two plots of the native sand fairway trial of 2006-2007 after snow melted on 12 March 2007. The treated plot received 12 g/100 m2 azoxystrobin on 25 November 2006, and showed no evidence of snow molds after snow melt, while the untreated plot had some snow mold patches. 82 Figure 2.13. Plant injury of the Agrostis stolonifera native sand fairway multi- season trial of 2006-2007. Plant injury was evaluated weekly to monthly from 23 November 2006 until 8 August 2007 and was assessed as percentage area with yellow to straw color grass (produced by biotic and abiotic sources), based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. 83 Figure 2.14. Pathology green multi-season trial of 2006-2007 on 23 November 2006. 84

IX Figure 2.15. Two plots of the pathology green trial of 2006-2007 after snow melted on 12 March 2007. The treated plot received 12 g/100 m2 azoxystrobin on 25 November 2006, and showed no evidence of snow molds after snow melt, while the untreated plot had some snow mold patches. 85 Figure 2.16. Plant injury of the Agrostis stolonifera pathology green multi-season trial of 2006-2007. Plant injury was evaluated weekly to monthly from 23 November 2006 until 8 August 2007 and was assessed as percentage area with yellow to straw color grass (produced by biotic and abiotic sources), based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. 86 Figure 2.17. Dollar spot incidence of the Agrostis stolonifera native sand fairway multi-season trial of 2005-2006. Dollar spot incidence was evaluated weekly to monthly from 13 March 2006 until 20 October 2006, and was assessed as number of patches per m2, based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. 87 Figure 2.18. Dollar spot incidence of the Poa pratensis multi-season trial of 2005- 2006. Dollar spot incidence was evaluated weekly to monthly from 13 March 2006 until 20 October 2006, and was assessed as number of patches per m2, based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. 88 Figure 2.19. Dollar spot incidence of the Agrostis stolonifera native sand fairway multi-season trial of 2006-2007. Dollar spot incidence was evaluated weekly to monthly from 23 November 2006 until 8 August 2007, and was assessed as number of patches per m2, based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 25 November 2006 at 12 g azoxystrobin per 100 m2. The bars represent standard error. 89 Figure 2.20. Two plots of the pathology green trial of 2006-2007 on 24 July 2007. The treated plot received 12 g azoxystrobin per 100 m2 on 25 November 2006, and showed higher number of dollar spot patches than the untreated plot. 90 Figure 2.21. Dollar spot incidence of the Agrostis stolonifera pathology green multi-season trial of 2006-2007. Dollar spot incidence was evaluated weekly to monthly from 23 November 2006 until 8 August 2007, and was assessed as number of patches per m2, based on four replicate 1 m x 2 m plots per treatment. Heritage® MAXX was applied on 25 November 2006 g azoxystrobin per 100 m2. The bars represent standard error. 91

x Figure 2.22. Agrostis stolonifera summer plots 2006, located on the pathology green at the Guelph Turfgrass Institute, Ontario. Plots were the only area of the green with highly injury caused by biotic and abiotic factors. Because of these confounding factors the data were not used in analyses. 92 Figure 3.1. Yeast colony morphotypes commonly found on potato dextrose agar after 10 days at 10C. 127 Figure 3.2. Bacterial colony morphotypes commonly found on potato dextrose agar after 10 days at 10C. 128 Figure 3.3. Filamentous fungal colony morphotypes commonly found on potato dextrose agar after 10 days at 10C. 129 Figure 3.4. Total microbial populations (bacteria, yeast and filamentous fungi) of treated (full line) and untreated (dashed line) green leaves of the native sand fairway 2005-2006, pathology green 2006-2007 and Poa pratensis fringe 2005-2006. Treatment consisted of a single application of Heritage® 50WDG at 6 g azoxystrobin per 100 m2 applied on 28 November 2005 on the native sand fairway 2005-2006 and P. pratensis fringe 2005-2006 or MAXX® at 12 g azoxystrobin per 100 m2 applied on 25 November 2006 on pathology green. The error bars represent the standard error. 130 Figure 3.5. Total microbial populations (bacteria, yeast and filamentous fungi) of treated (full line) and untreated (dashed line) yellow leaves of the native sand fairway 2005-2006, pathology green 2006-2007 and Poa pratensis fringe 2005-2006. Treatment consisted of a single application of Heritage® 50WDG at 6 g azoxystrobin per 100 m2 applied on 28 November 2005 on the native sand fairway 2005-2006 and P. pratensis fringe 2005-2006 or MAXX® at 12 g azoxystrobin per 100 m2 applied on 25 November 2006 on pathology green. The error bars represent the standard error. 131 Figure 4.1. Filamentous fungal submorphotypes. Isolates were obtained from Agrostis stolonifera leaf surfaces by washing and plating on PDA with antibiotics. Isolates were plated separately on PDA plates and incubated a room temperature for up to 14 days. Top view is shown on left, and bottom view is shown on right, and the notation of color/color refers to top/bottom color. 182 Figure 4.2. Yeast submorphotypes. Isolates were obtained from A. stolonifera leaf surfaces by washing and plating on PDA with antibiotics. Isolates were plated on PDA plates and incubated a room temperature for seven to nine days. 183 Figure 4.3. Bacterial submorphotypes. Isolates were obtained from A. stolonifera leaf surfaces by washing and plating on PDA with benomyl. Isolates

XI were plated on PDA plates and incubated a room temperature for five to seven days. 184 Figure 4.4. Bacterial 16S PCR products digested with restriction enzymes. The product were separately electrophoretically in 2% (w/v) agarose gels, with the weight marker O'GeneRuler™ DNA Ladder, in the first, middle and last lanes. A) digested with /-/pall and B) digested with Haelll. Gels were stained with ethidium bromide solution and observed with UV transilluminator, and the image captured on computer with a frame-grabber card. The images were reversed and enhanced with Paint Shop Pro. 185 Figure 4.5. Yeast ITS PCR products digested with restriction enzymes. The product were separately electrophoretically in 2% (w/v) agarose gels, with the weight marker O'GeneRuler™ DNA Ladder, in the first, middle and last lanes. A) digested with Hae\\\, B) digested with EcoRI, C) digested with Alu\. Gels were stained with ethidium bromide solution and observed with UV transilluminator, and the image captured on computer with a frame-grabber card. The images were reversed and enhanced with Paint Shop Pro. 186 Figure 4.6. Filamentous fungal ITS PCR products digested with restriction enzymes. The product were separately electrophoretically in 2% (w/v) agarose gels, with the weight marker O'GeneRuler™ DNA Ladder, in the first, middle and last lanes. A) non-digested, B) digested with /-/aelll, C) digested with Alu\. Gels were stained with ethidium bromide solution and observed with UV transilluminator, and the image captured on computer with a frame-grabber card. The images were reversed and enhanced with Paint Shop Pro. 187 Figure 4.7. Distance dendrogram of RFLP binary analyses of bacterial 16S PCR products. Banding patterns were obtained using four restriction enzymes separately: Alul, Cfol, Haelll and Hpall, and patterns were recorded in a binary format and analyzed with Windist and Phylip. 188 Figure 4.8. Distance dendrogram of RFLP binary analyses of yeast ITS PCR products. Banding patterns were obtained using four restriction enzymes separately: Alu\, EcoRI, Haelll and Hpall, and patterns were recorded in a binary format and analyzed with Windist and Phylip. 189 Figure 4.9. Distance dendrogram of RFLP binary analyses of filamentous fungal ITS PCR products. Banding patterns were obtained using four restriction enzymes separately: Alu\, CM, EcoRI, Haelll and Hpall, and patterns were recorded in a binary format and analyzed with Windist and Phylip. 190

XII LIST OF APPENDICES Appendix A. Sample SAS statement for comparing plant injury or dollar spot incidence averages between azoxystrobin-treated and untreated plots. 220 Appendix B. Maintenance schedule for the A. stolonifera native soil fairway (B1), the pathology green (B2) and the Poa pratensis fringe (B3). 221 Appendix C. SAS statements used to test homoscedascity and normality of distribution. 224 Appendix D. Normality of distribution based on the Shapiro-Wilk test counts of foliar microbial populations using non-transformed and log- transformed CFU/g of dry leaf. 0 = not normally distributed and 1 =normally distributed. 225 Appendix E. Homogeneity of variance using Bartlett's test of counts of microbial populations using non-transformed and log-transformed CFU/g of dry leaf. Probability values greater than 0.05 then the variance is homogeneous. 228 Appendix F. SAS statements for assessing differences between microbial populations using ANOVA analysis 230 Appendix G. Selection of 16S restriction enzymes based on in silico analyses with GeneRunner of 15 restriction enzymes with 38 common foliar bacteria. 231 Appendix H. Sequences of 16S selected from GenBank which were used for in silico analyses. These 45 species were chosen based on previous reports which depicted them as members of various phyllospheres. 232 Appendix I. Top matches of bacterial 16S sequence of turfgrass bacteria. rDNA amplified with primers F0 and PRUN518. Sequences compared with nr database using megaBLAST. 233 Appendix J. Top matches of yeast ITS sequence of turfgrass yeasts. rDNA amplified with primers ITS1 and ITS4. Sequences compared with nr database using megaBLAST. 238 Appendix K. Top matches of filamentous fungal ITS sequence of turfgrass fungi. rDNA amplified with primers ITS1 and ITS4. Sequences compared with nr database using megaBLAST. 243

XIII LIST OF ABBREVIATIONS AND ACRONYMS ATP: Adenosine Triphosphate BLASTN: Basic Local Alignment Search Tool Nucleotide bp: base pair(s) BSA: Bovine Seric Albumin C: Centigrade DGGE Denaturing Gradient Gel Electrophoresis DNA: Deoxyribonucleic acid EDTA: Ethylene Diamine Tetraacetic Acid EMBL: European Molecular Biology Laboratory FRAC: Fungicide Resistance Action Committee g: gram(s) GTI: Guelph Turfgrass Institute IGS Intergenic Spacer ITS: Internal Transcribed Spacer Pa Pascal 1: liter(s) M: Molar m: meter(s) min: minute(s) NCBI: National Center for Biotechnology Information PCR: Polymerase Chain Reaction PDA: Potato dextrose agar PDB: Protein Data Bank PPFM pink-pigmented, facultative methylotrophic bacterium ppm: part per million Qo cytochrome b ubihydroquinone oxidation center Qol: Qo Inhibitor rDNA: Ribosomal DNA RefSeq: Reference Sequence RFLP: Restriction Fragment Length Polymorphism rRNA: Ribosomal Ribonucleic Acid rpm: revolutions per minute s: second(s) TBE: Tris Borate EDTA TE: Tris EDTA UK: United Kingdom USA: United States of America USGA United States Golf Association UV Ultra violet v/v: volume on volume w/v: weight on volume WDG: Water Dispersible Granules PVP Polyvinyl Pyrrolidone TSA Trypticase Soy Agar

XIV CHAPTER 1. LITERATURE REVIEW 1.1 Fungicides and turfgrass Fungicides are used to protect plants from infectious fungi. If used properly, a fungicide should prevent or stop a disease from establishing or progressing.

The first record of a synthetic fungicide used for plant disease control was the application of Bordeaux mixture to grapevines in France in 1864, to protect grapes from powdery mildew (Sigler et al., 2000). Since then, fungicides have been widely used on plants including turfgrasses. Turfgrasses are a major component of the urban environment such as in home lawns, parks, golf courses, athletic fields, commercial properties and boulevards.

In the US, pesticide used for agriculture and non-agricultural crops in 1993 was estimated at 380 and 120 million kg of active ingredient, respectively (Hodge,

1993). This demonstrates that non-agricultural pesticide usage is very important, reaching almost one third of agricultural pesticide usage. A major portion of non- agricultural pesticide application involves fungicides, especially on golf courses.

For example, Hawaiian golf courses applied on average 8,641 kg of fungicides per year which represented more than 50% of the annual expenditures on turfgrass pesticides in Hawaii (Racke, 2000). In a survey of golf courses in

Alberta, Canada, the study showed that the average amount of pesticide active ingredient per golf courses was 76.9 kg per year, with fungicide composing 64%

(Anonymous, 1998).

Fungicides are widely used to protect golf course turf because of the demand for high quality aesthetically pleasing turf by users and managers.

l Turfgrass fungal diseases, such as snow molds, Pythium blight, anthracnose, and dollar spot, may become difficult to control when conditions are favorable for disease development, which leads to higher or more frequent application of fungicides applications, and which may result in fungicide-resistant populations as has been reported for diseases such as dollar spot (Smiley et al., 2005).

In certain situations when fungicides are applied in turfgrass environments different side effects may occur. Among those side effects, the increase of a particular disease after a fungicide application known as disease enhancement or resurgence is commonly reported. These two terms are often interchangeable, but sometimes are separated according to the target disease of the pesticide. If the fungicide applied increased the incidence of the same disease that was previously controlled or eradicated, then the phenomenon is called resurgence

(Vincelli, 2007). If the increased disease is other than the one the fungicide was originally intended to suppress, then this is referred as disease enhancement

(Vincelli, 2007).

Many examples of disease incidence increased by fungicide have been reported. In Kentucky bluegrass, applications of chlorothalonil lead to more summer patch (caused by Magnaporthe poae) (Vincelli, 2007). Pythium blight is enhanced by benomyl (Warren et al. 1976). Enhancement of dollar spot has been reported by the application of azoxystrobin in creeping bentgrass

(Dernoeden, 2000; Hsiang and Cook, 2006).

2 1.2 Strobilurins The strobilurins are an important class of fungicides used to control a wide spectrum of pathogenic fungi in agriculture and on ornamental plants (Bartlett et al., 2002). They are used to protect important crops, such as cereals, tubers, vegetables, fruits and turfgrasses from common diseases (Bartlett et al., 2002).

Consequently, use of strobilurins has increased tremendously since their introduction, even though they have only been on the market since 1996.

According to McDougall (2001), these fungicides represented over 10% of the world fungicide market in 1999, selling more than US$620 million, with US$415 million from azoxystrobin sales alone (Heaney et al., 2000). This trend looks like it will continue since strobilurins sales predictions for 2006 were estimated to reach 20% of the world fungicide market, becoming the top fungicide group, followed by the triazoles (Sauter et al., 1999).

Although this has been a success story for the strobilurins, many cases of resistance development to strobilurins have been reported around the world. By the second quarter of 2006, the Fungicide Resistance Action Committee (FRAC) reported that 24 plant pathogens present in fruits, field crops, tubers and turfgrasses were resistant to strobilurins pesticides (Anonymous, 2006a). In turfgrasses, FRAC reported that three pathogens were found resistant to strobilurins: Pythium aphanidermatum, Colletotrichum graminicola and

Pyricularia grisea (Anonymous, 2006a).

3 1.2.1 History The discovery and development of the original strobilurin compounds dates from the 1970's. Anke et al. (1977) claimed that the basidiomycetous pine cone , Strobilurus tenacellus, produced novel and powerful antimycotics called strobilurins A and B. After the strobilurins were detected from S. tenacellus, other basidiomycetes species, such as Cyphellopsis and Mycena, were found to also produce strobilurins (Sauter et al., 1999). Surprisingly an ascomycete, Bolinea lutea, was also found to produce this type of chemical, suggesting that strobilurins have an origin before the division of ascomycetes and basidiomycetes (Anke, 1995).

In nature, strobilurin-producing fungi have a survival advantage due to the growth inhibition of possible competitors (Sauter et al., 1999). Laboratory studies of strobilurins have shown them to be specific against fungi and to have a broad spectrum of action inhibiting fungi in the Deuteromycota, ,

Basidiomycota and even Oomycota (Bartlett et al., 2002), the last of which is no longer considered fungal, but are more closely related to other stramenopiles, such as brown algae (Kamoun, 2003).

Since the discovery of strobilurins, several private companies have initiated studies to commercialize strobilurins. Two of these companies, BASF and ICI

(now part of Syngenta), started parallel research programs in the early 1980s for a better understanding of these particular molecules (Sauter et al., 1999). They focused on strobilurins because of the low level of observed non-target effects on mammals, the simplicity of the molecules, the elevated antifungal activity, and the new mode of action, However, field tests with strobilurin A were

4 unsatisfactory because it was unstable (Sauter et al., 1999). This led to a series

of modifications of the original molecule in which an aromatic bridge was used to

stabilize the chemical without losing the antifungal effect (Sauter et al., 1999).

The result was the creation of a large variety of synthetic strobilurins with diverse

properties. From the thousands created, only two chemicals, one from each of the initial companies, became registered in 1996 (Bartlett et al., 2002).

More than 30,000 strobilurin analogues have been synthesized by 20

institutions and companies. Over 500 have been patented, but by 2002, only six were commercialized: azoxystrobin, pyraclostrobin, kresoxim-methyl, trifloxystrobin, metominostrobin and picoxystrobin (Bartlett, et al., 2002). Since

2006, six other strobilurins have been added to the pesticide market: enestrobin, dimoxystrobin, famoxadone, fenamidone, fluoxastrobin and orysastrobin

(Anonymous, 2006b). Since many pathogens have shown resistance to traditional fungicides, new strobilurins analogues are being developed to respond to the demand for effective pest control (Huang, et al., 2007).

1.2.2 Structure The diversity of strobilurins synthesized makes it difficult to recognize that all these thousands of compounds belong to the same family. However, all strobilurins have the same mode of action and share the same toxophore (group conferring biological activity) (Figure 1.1). In almost all strobilurins, the toxophore

(p-methoxyacrylate) links to a benzene ring, which has a long and variable side chain (Gisi et al., 2002). The p-methoxyacrylate is essential because the heteroatoms may have polar and intermolecular interactions with the

5 ubihydroquinone oxidation (Qo) center of the respiratory enzyme (Sauter et al.,

1999). Studies of the co-crystallized structure of the cytochrome bc1 complex with different strobilurins have shown that both the toxophore and benzene ring bind to specific amino acids of enzymes (Gisi et al., 2002).

1.2.3 Mode of action Strobilurins have a toxophore and a variable side chain (Figure 1.1) which modifies the spectrum of action (Bartlett et al., 2002). The toxophore carries a carbonyl oxygen moiety that permits the molecule to bind to the enzyme target

(Gisi et al., 2002). This interaction interrupts the respiratory chain by reversibly binding to the Qo center of the cytochrome b in the complex III (Becker et al.,

1981). Thus, these fungicides are part of a wider group of fungicides called Qo inhibitors or Qols. Qols block the electron transfer between cytochrome b and d producing a deficiency in ATP, which inhibits the fungal growth (Bartlett et al.,

2002). In addition to strobilurins (methoxyacrylates), there are other six types of chemical compounds that have the same mode of action: methoxycarbamates, oximino-acetates, oximino-acetamides, oxazolidine-diones, dihydro-dioxazines and midazolinones (Anonymous, 2006b).

Depending on the particular chemical, strobilurins can have protectant activity, systemic activity or both. Protectant fungicides coat the surface of the plant and protect against pathogen entry. Systemic fungicides can penetrate into the plant and translocate within the plant to provide protection to parts of the plant which were not exposed to the initial fungicide application. Azoxystrobin has systemic activity and moves in the vascular system acropetally from the point

6 of absorption toward the tips of the leaves (Christians, 2007). Pyrasclostrobin on

the other hand, is not systemic and picoxystrobin moves up and down through

the xylem (Bartlett et al., 2001).

1.2.4 Fungicide resistance The development of fungicide resistance is a common place event, and is

expected for most modern pesticides after sufficient usage. This phenomenon is

caused by the artificial selection and eventual domination by naturally occurring

sub-populations of less sensitive strains (Smiley et al., 2005). The risk of

fungicide resistance increases when the fungicide only has one site of action or

when the generation time of the pathogen is short (Forbes, 2001; Ma et al., 2003).

The use of strobilurins against plant pathogens with short generation time places

this group of fungicide in the high-risk category for resistance development for those pathogens. As predicted, only two years after the registration of

azoxystrobin, the first case of resistance was found in the powdery mildew fungus, Blumeria graminis, on wheat in Germany (Heaney et al., 2000). Since then, 24 other pathogens around the world have had confirmed resistance to

Qols or have shown markedly reduced sensitivity (Anonymous, 2006a). For

instance, azoxystrobin has become ineffective in controlling gray leaf spot in perennial ryegrass at several locations in the US (Vincelli and Dixon, 2002), and against turf anthracnose in California (Wong and Midland 2004).

1.2.5 Environment persistence and animal safety In general, strobilurins pose little risk toward humans and animals, but under certain conditions, they may be harmful. These fungicides, with exception

7 of metominostrobin, have low toxicity to mammals (Bartlett et al., 2002). The

Environmental Protection Agency in the US stated that: "Azoxystrobin is of low acute and chronic toxicity to humans, birds, mammal, and bees, but is highly toxic to freshwater fish, freshwater invertebrates, and estuarine/marine fish, and highly toxic to estuarine/marine invertebrates." (Anonymous, 1997). Strobilurins are toxic in aquatic environments probably because of the increased lipophilicity under such conditions (Bartlett et al., 2002). Since Qols might induce different types of negative alterations in aquatic ecosystems, the final fate of strobilurins is an important issue. For this reason, government institutions have required product registration studies on leaching, half-life, byproducts degradation, and other environmental effects.

In general, strobilurins are resistant to hydrolysis, as evidenced by the half- life of azoxystrobin, which is 267 days at 25C (Anonymous, 2000). However strobilurins are sensitive to other sources of degradation such as photolysis or microbial biotransformation (Bartlett et al., 2002). These sources of degradation have diverse effects on each particular strobilurin, but in general, the fast degradation rates make leaching and persistence of strobilurins of little risk. In a study conducted in Nova Scotia, Ontario and British Columbia, leaching scores for kresoxim-methyl indicated that it would be unlikely to leach (Anonymous,

2003a). In persistence studies using azoxystrobin, it was dissipated by 50% in

14-62 days, and was classified as non-persistent to moderately persistent

(Anonymous, 2000).

8 Regarding the effects of strobilurins on non-target species, most studies have focused on multicellular organisms, such as animals. In general, strobilurins are very toxic to aquatic animals, including fish, mollusks and zooplankton

(Anonymous, 1997, Anonymous, 2003a and Anonymous, 2003b). In the same studies, birds, mammals, bees, earthworms and other organisms were negligibly affected, and hence strobilurins are regarded as low risk for terrestrial animals.

For instance, the lethal oral dose at which 50% of test subjects may dies (acute oral LD50) is more than 5 g/kg in rats (Anonymous, 1997), which is considered slightly to non-toxic.

Evaluations on the effect of strobilurins on non-target microorganisms are scarce. Only the strobilurin, 9-methoxystrobilurin L, was reported to have antibacterial activity, but in later years that was found to be erroneous (Sauter et al., 1999). Few direct studies on the effect of strobilurins on yeasts have been published, but it is known that the mitochondrial electron chain reaction of yeasts can be interrupted by these pesticides, since most inhibitory studies have been carried out using yeast mitochondria as a model (Sauter et al., 1999). In 2002,

Buck and Burpee found that numbers of yeasts on turfgrass leaves were significantly reduced when treated with azoxystrobin. Moreover, after several applications, yeasts with less sensitivity toward azoxystrobin were found in treated leaves (Buck and Burpee, 2002).

1.3 Dollar spot Dollar spot is a common turfgrass fungal disease widely distributed around the globe in temperate to semi-tropical zones. It can cause severe damage,

9 mainly in closely mowed turf, producing an uneven and unpleasant golf playing surface. This disease has been reported in North, Central and South America,

Europe, Asia and Australia (Rivera et al., 2004; Smith et al., 1989). Particularly in

North America, dollar spot is consider to be the most prevalent turfgrass disease, especially on golf courses, but it also affects lawns and recreational turf

(Charbonneau and Hsiang, 2003; Walsh et al., 1999). As a result, more money is spent to control dollar spot than any other turfgrass disease in golf courses in

North America (Goodman and Burpee, 1991). In Canada, all cool season turfgrass species are affected by dollar spot in the diverse environments of the

Great Lakes Region, Prairies, Maritimes, and West Coast (Charbonneau and

Hsiang, 2003).

1.3.1 Hosts and symptoms Dollar spot affects many turfgrass species. Around the world it has been reported to infect: Festuca, Agrostis and Poa (Smith et al., 1989). However, other genera also suffer from this disease. For example, in Japan, the warm-season turfgrass japonica, and in Australia, Digitaria didactyla and Agrostis spp., are frequently attacked by dollar spot (Smith et al., 1989). In addition, perennial ryegrass (), bermudagrass (Cynodon dactylon) and bahiagrass

(Paspalum notatum) are also susceptible to dollar spot (Smith et al., 1989).

Including all grass hosts and some rarely infected host plants, the dollar spot fungus has been found to be associated with at least 40 different plant species

(Walsh etal., 1999).

10 Dollar spot symptoms depend on the host affected and management practices. In bentgrasses (Agrostis spp.) maintained as a closely mowed turf, the disease is characterized by sunken, circular, straw-colored spots (Figure 1.2) that are 5 to 7.5 cm in diameter (Couch, 1995). Under conditions favorable to disease, the patches may coalesce to form larger, irregular, tan-colored patches

(Couch, 1995). In turfgrasses which are maintained at higher heights, such as

Kentucky bluegrass (Poa pratensis), perennial ryegrass (Lolium perenne) and fescues (Festuca spp.), straw-colored patches develop (Figure 1.3) that range from 6 to 12 cm and may have an irregular shape (Couch, 1995). Dollar spot in longer leaved grasses forms a characteristic type of lesion (Figure 1.4), distinguished by a dark to reddish margin and a bleached colored and shrunken zone across the leaf blade, and is known as an hourglass lesion (Smiley et al.

2005). In addition, a white floccose web-like mycelium (Figure 1.5) can be observed when dew or high relative humidity is present on warm days

(Charbonneau and Hsiang, 2003; Walsh et al., 1999).

1.3.2 Causal agent

1.3.2.1 Taxonomy The causal agent of dollar spot is currently classified as Sclerotinia homoeocarpa FT. Bennett (Smiley et al., 2005). Originally, Monteith and Dahl

(1932), described the fungus associated with dollar spot symptoms as a species of Rhizoctonia, but shortly afterwards Bennett (1937) reclassified the fungus into the Sclerotinia based on the presence of micro-sclerotia in agar cultures

(Walsh et al., 1999). However, some mycological taxonomists do not consider

n the dollar spot agent to belong to the genus Sclerotinia since it does not produce tuber-like sclerotia as do the other species in this genus, and it lack a fertile teleomorph (Smiley et al., 2005).

Kohn (1993) stated that S. homoeocarpa could not be classified until there was a close examination of the apothecial microanatomy, especially in comparison with the genera Rutstroemia, Lanzia and Mollerodiscus. Carbone and Kohn (1993) contend that S. homoeocarpa is more closely related to substratal stromatal genera, such as Rutstroemia, than to other Sclerotinia species based on DNA sequence analysis. Hoist-Jensen et al. (1997) examined ribosomal DNA sequences of Sclerotiniaceae and concluded that S. homoeocarpa belongs in the same genus as Poculum. Vargas and Powell (1997

& 1999) proposed a reclassification of S. homoeocarpa in the genus Rutstroemia.

The genera Poculum and Rutstroemia are still undergoing revision, and this along with the lack of a teleomorph for North American isolates makes the taxonomic position of S. homoeocarpa uncertain. The causal agent of dollar spot is still commonly referred to as Sclerotinia homoeocarpa among turfgrass managers and turf pathologists.

1.3.2.2 Morphology Isolates of S. homoeocarpa grown on solid media (Figure 1.6) are characterized by a fast-growing, fluffy, white mycelium with occasionally olive, gray, yellow or brown tones as the culture ages (Smiley et al., 2005). In culture, no true sclerotia are formed, but black layers of stromata appear, and black

12 flakes may form small patches or cover the entire mycelial surface (Smiley et al.,

2005; Smith et al., 1989).

Sclerotinia homoeocarpa survives unfavorable periods as a dormant mycelium in infected plants, and as dark stromata in the thatch or on leaf surfaces (Smiley et al., 2005). This pathogen is disseminated through direct contact with neighboring healthy leaves and transport of propagules by equipment, animals, water, or wind (Smiley et al., 2005). When temperatures range from 15C to 30C, the mycelium starts to grow from the thatch reaching healthy leaves (Smiley et al., 2005). Other environmental conditions, such as drought stress and low nitrogen fertility, favor disease development (Smiley et al.,

2005; Smith et al., 1989). The requirement of moderate to higher temperatures makes dollar spot a seasonal disease from late spring to early fall (Smiley et al.,

2005).

Once the S. homoeocarpa mycelium reaches healthy tissue, it invades the plant via stomata or wounds, such as cut leaf tips (Smith et al., 1989). Then the aerial mycelium extends radially from the initial lesion infecting surrounding leaves forming circular patches, but for unknown reasons, the growth of the mycelium stops (Smith et al., 1989), causing the characteristic small spots no larger than a dollar coin.

1.3.2.3 Control Cultural control of dollar spot has focused on preventing or avoiding conditions favorable to disease. Maintaining adequate nitrogen levels and avoiding drought stress are the most common cultural practices for decreasing

13 the severity of dollar spot epidemics (Smiley et al., 2005; Smith et al., 1989).

Reduction of leaf wetness periods by diverse techniques has also been found to reduce dollar spot incidence as well as that of some other turf diseases (Williams et al., 1996).

The previously mentioned cultural control practices are useful to delay the spread of the disease and decrease disease severity, but on intensively managed turfgrass, the aesthetic standards are so high that turf fungicides nearly always required. In the past, dollar spot was controlled with mercury fungicides, but currently, many less persistent fungicides are registered for use (Smiley et al.,

2005; Smith et al., 1989). In Canada, there are seven active ingredients registered for dollar spot control: chlorothalonil, iprodione, boscalid, propiconazole, thiophanate-methyl, and thiram (Anonymous, 2007).

Biological control approaches have been investigated as an alternative for the control of dollar spot. Diverse nutrient amendments, such as fertilizers and composts, have been tested for their effectiveness in controlling dollar spot

(Boulter et al., 2002; Hoyland and Landschoot, 1993; Liu et al., 1997). The other approach consists of applying microbial antagonists, such as Pseudomonas fluorescens (Rodriguez and Pfender, 1997). Although some of these amendments or antagonists have shown some efficacy in controlling dollar spot, the cost, inconsistent results, and faster disease suppression by chemical fungicides have not made biological control a common practice.

14 1.3.3 Dollar spot and fungicides Fungicide application is currently the most commonly used method to control dollar spot. Usually, several applications during a growing season are needed in order to obtain satisfactory control and, as a result, several S. homoeocarpa populations have become resistant to various fungicides. In the US,

S. homoeocarpa has been reported to be resistant to the following fungicides: heavy metals, benzimidazoles, anilazine, dicarboximides and demethylation inhibitors (DMI) (Walsh et al., 1999). Additionally, fungicide cross-resistance also has been reported for S. homoeocarpa. For instance, some DMI-resistant isolates were found to be less sensitive to benzimidazoles and dicarboximides fungicides (Golembiewski et al., 1995).

This development of fungicide resistance had led pesticide companies to develop new fungicides with different sites of action. The most recently introduced major fungicide family, the strobilurins, is not effective in controlling dollar spot, although they show an amazingly broad spectrum of activity against other fungal pathogens (Anonymous, 2000). Product labels for strobilurin fungicides registered in Canada are available at the PMRA website (http://pr- rp.pmra-arla.gc.ca/pub_reg/pls), and neither of the two strobilurins registered on turfgrass in Canada (azoxystrobin and trifloxystrobin) has dollar spot disease on the label. Studies and casual observations have also indicated that dollar spot disease is not controlled by azoxystrobin or pyraclostrobin (Dernoeden et al.,

2000). Several reports in Fungicide and Nematicide Tests have indicated that strobilurins are ineffective for controlling dollar spot and that azoxystrobin increases the incidence and severity of the disease (Dernoeden et al., 2000;

15 Gleason et al., 1996; Grogan and Scott, 1997). In addition previous research in this laboratory has lead to observations of the enhancement of dollar spot on areas were azoxystrobin was applied for snow molds experiments (Cook and

Hsiang, 2004; Hsiang and Cook, 2006). However, turfgrass managers from golf courses have not notice an increase of dollar spot when they have applied commercial fungicides based on azoxystrobin probably because of the intense applications of other fungicides for dollar spot control.

Pesticide-induced outbreaks are not rare and have been reported for other diseases and fungicides, such as the increased severity of Pythium blight after benomyl applications (Warren et al., 1976), or thiophanate-methyl enhancement of crown rust on Lolium perenne (Couch and Joyner, 1975). The severity and occurrence of dollar spot has also been reported to increase after the use of benzimidazoles in bentgrass (Couch and Smith, 1991).

The reasons for these side effects are not known. Two basic explanations have been proposed. First, the fungicide might have a direct effect on the treated plant rendering it more susceptible to particular diseases. Physiological side effects of fungicides on host plants, such as alteration of tissue dry weight and nutrient content support, this hypothesis (Mazur and Hughes, 1976). The other explanation is based on the fungicide suppression or alteration of the non-target antagonistic microorganisms naturally present in or on the host plant that may compete for substrates or inhibit pathogenic fungi, (Melzer and Boland, 1998;

Smiley and Craven, 1979).

16 1.4 Phyllosphere The aerial surfaces of plants represent an enormous area that can be colonized and exploited by many organisms. These surfaces involve leaves, stems, flowers and fruits, but the foliar surface area is the largest component. It has been estimated that the foliar surface of all terrestrial plants represents an area between 2 to 6.4 x 108 km2 depending on the data utilized and the season

(Morris and Kinkel, 2002). If all of this were habitable surface, then it is not surprising that organisms from all kingdoms of life have been found to use this niche and in many different ways. Macroscopic organisms, such as insects and plants, and microscopic organisms such as algae, bacteria, protozoa, lichens and fungi, both single-celled and filamentous, compete for plants surfaces (Atlas and

Bartha, 1987). This niche is called the phylloplane or phyllosphere, and the inhabitants are termed epiphytes and endophytes (Lindow and Leveau, 2002).

Some researchers, such as Carroll et al. (1977), contend that the phyllosphere is part of the living leaf. Other researchers include, buds, flowers, fruits of aerial plants or even the surfaces of organisms such as seaweeds as a part of the phyllosphere (Dickinson, 1982; Morris and Kinkel, 2002).

Microorganisms which establish inside the leaves are usually called endophytes, and are generally fewer in comparison to epiphytes (Lindow and Brandl, 2003).

The study of the phyllosphere formally started with Last (1955), but it was not until 1970 that the phyllosphere became considered a separate ecological field of study (Morris and Kinkel, 2002). Most studies originally were conducted to obtain a better understanding of plant pathogen behavior on important crops

(Lindow and Leveau, 2002). Consequently, the effects and interactions with non-

17 pathogenic microorganisms are not as well documented, especially in non-

cultivated plants. However, a new trend involving broader arrays of plants and

interactions have been developing in phyllosphere research in recent years, in

which food science, biological control and ecophysiology have been emphasized

(Morris and Kinkel, 2002).

Phyllosphere research has contributed to food science safety and

preservation by helping to explain the process of how pathogenic and spoilage

microbes attach and remain on the surfaces of leaves and fruit (Morris and Kinkel,

2002). Studies of microbial phyllosphere communities of inoculated plants with

non-pathogenic bacteria showed a competitive exclusion of the pathogenic

bacteria, which has shed light into biological control of pests (Lindow and Brandl,

2003). In addition, other phyllosphere studies have contributed to a better

understanding of important issues of plant pathology and plant biology, such as

horizontal gene transfer among foliar bacteria and interactions that reduce

herbivory (Morris and Kinkel, 2002).

1.4.1 Microbial Biodiversity Microbial diversity involves an immense range of different kinds of

unicellular organisms including bacteria, archea, protists, yeasts and some multicellular organisms, such as filamentous fungi, arthropods and small plants.

The total number of microbial species is not known, but the number of bacteria species is estimated to be 1 million, even though only about 3,800 species of have been identified (Hawksworth, 1991; Scow et al., 2001).

18 Leaf surfaces are considered an extreme environment. Some characteristics include fluctuating temperatures, high UV radiation and scarcity of nutrients (Yang et al., 2001). However, some plants under normal conditions can sustain over 1 x 108 bacterial colony forming units (CFU/cm2) on leaf surfaces

(Morris and Kinkel, 2002), compared to other surface environments, such as human skin which can sustain 1 x 107 CFU/cm2 (Noble, 1982) or a highly nutritive environments such as bovine manure that sustains 1 x 109 CFU/g.

Biodiversity is measured by indices that reflect the number and relative proportions of species in a community (Morris and Rouse 1986). When studying leaf surface or phyllosphere populations, scientists have found a high diversity of bacteria, filamentous fungi and yeasts. Yang et al. (2001) stated that in only four plant hosts, 85 different phyllosphere bacterial species were identified by culture- dependent methods. Moreover, Stohr and Dighton (2004) reported 107 fungal genera on leaves of American cranberry (Vaccinium macrocarpon). Thus, the phyllosphere is an important ecological niche not only because of the high number of microorganisms, but also because it has a relatively high diversity

(Tables 1.1 to 1.3).

The complex microbial communities of the phyllosphere have different spatial distributions, dynamics, and interactions, with variation at different scale levels. When considering each leaf as a habitat unit (Kinkel et al., 2002), population variability among leaves or leaf growth stage has been reported

(Inacio et al., 2005; Yadav et al., 2005). Phyllosphere microorganisms are usually distributed unevenly on leaf surfaces (Kinkel et al., 2002). They usually group

19 together forming clusters of cells in specific areas of the leaf topography, associating sometimes with specific leaf structures, such as veins or stomata

(Beattie and Lindow, 1995). The reasons for the clustered distribution may depend on the variation of nutrients and water (Beattie and Lindow, 1995).

Populations of leaf groups may also show variable distribution at the canopy level.

For instance, Osono and Mori (2004) described the spatial and temporal patterns of phyllosphere fungi of the giant dogwood (Cornus controversa) canopy. Also

Andrews et al. (1980) found that bacterial populations of apple (Malus domestica) leaves were higher on internal leaves of the canopy.

More recently, several distribution patterns and arrangements on leaves have been found (Morris et al., 2002; Monier and Lindow, 2005; Yadav et al.,

2005). Kinkel et al. (2002) stated that the most particular characteristic of microbial distribution on individual leaves is the high degree of spatial aggregation. This distribution seems to follow a non-random pattern suggesting that intrinsic factors of the leaf surface, such as nutrients or topography are involved (Kinkel et al., 2002).

Competition for nutrients or space between microorganisms that belong to the same functional group is a common event on the phyllosphere. Evidence of such competition between a saprophyte and a pathogenic fungus is the use of

Ulocadium atrum for control of Botrytis aclada in onion (Kohl et al., 1997). Also parasitism of certain fungal species has been found on leaf surfaces, such as

Trichoderma harzianum attacking Botrytis cinerea (Belanger and Avis, 2002).

20 Bacterial populations have also been found to interact on leaf surfaces. On

some leaves, bacterial colonization patterns are similar to biofilm assemblages

found in other environments (Morris et al., 2002). Biofilms are the final stage of a

microbial colonization of a surface and have been described as a highly dense

accumulation of cells arranged in clusters surrounded by interstitial spaces filled with fluid (Habash and Reid, 1999). Biofilms of epiphytic bacteria increase the fitness of their members because these accumulations protect against desiccation, restrict diffusion of toxic chemicals, limit predation, enhance genetic exchange and help retain nutrients (Morris et al., 2002).

The microbial biodiversity of phyllosphere communities normally depends on the microenvironment of the leaf. Several examples have been reported of human activity and direct manipulation affecting phyllosphere microbial biodiversity. Fungicides (Buck and Burpee, 2002), fertilizers and mowing heights

(Behrendt et al., 2004) and pollutants (Dowding, 1986) are only some of the anthropogenic factors that can drastically modify leaf ecosystems. Among the natural factors that alter epiphytic populations are UV, season, wind, temperature, leaf topography and age. All of these factors combined create a highly variable environment in which the microbial populations and plants interact. There is also evidence that leaf surfaces are also being affected directly by the microbial population of the leaf. Water permeability of Prunus and Hedera cuticles increased when bacteria colonize them, which raises the water availability on the leaf surface (Schreiber et al., 2005). Grasses that have symbiotic endophytes in

21 their leaves appear to have an advantage in uptake of nitrogen (Bacon and Hill,

1996).

Turfgrass phyllosphere microorganisms comprise a relatively large number of yeasts, bacteria and filamentous fungi that can be epiphytes or endophytes. In temperate zones, yeasts are important inhabitants of turfgrass foliage in several genera of two major groups: white yeasts and pink or red yeasts (Buck and

Burpee, 2002). The pink/red yeast group contains genera, such as

Sporobolomyces or Rhodotorula, while white yeasts are usually Cryptococcus or

Pseudozyma (Allen et al., 2004b). In contrast to the literature on yeasts in the turfgrass phyllosphere, there is much less information on filamentous fungi and bacteria. There are few studies that have assessed epiphytic filamentous fungi species in turfgrass, but several studies have been done on endophytic fungi, such as Epichloe, which are grass mutualists (Schardl, 1996).

A similar lack of information occurs with turfgrass phyllosphere bacteria.

However, it has been reported that as with phyllospheres of other plants, yellow pigmented colonies can be isolated from grass leaves (Behrendt et al., 1999).

Intensively managed turfgrasses and their phyllosphere may encounter extreme conditions, such as high UV, heat, wind, fertilizer, mowing and intensive pesticide applications among other pressures. Some factors that have been shown to change turfgrass phyllosphere populations include fungicide applications and tissue damage due to disease (Bertelsen et al., 2001;

Gildemacher et al., 2004). Allen et al. (2004b) showed a higher number of total

22 yeast communities on leaves with lesions. This effect is likely related to the leakage of nutrients from the lesions caused by the pathogen (Allen et al., 2004b).

1.4.1.1 Bacteria phyllosphere communities The composition of bacterial species on leaves seems to follow a common pattern in many phyllosphere communities studied, in which few species dominate and others remain at lower frequencies (Hirano and Upper, 2000).

Phyllosphere bacteria are gram-negative or gram-positive, non-spore-forming heterotrophs (Hirano and Upper, 2000) of genera, such as Pseudomonas,

Xanthomonas, Flexibacter, Erwinia, Actinobacter, Staphylococcus, Bacillus,

Corynebacterium, Listeria and Micrococcus (Dickinson, 1982).

Many epiphytic bacteria produce pigments that may increase their UV tolerance, and extracellular polysaccharides that might assist surface attachment and prevent dissection (Beattie and Lindow, 1995). Frequently, yellow foliar bacteria belong in the genera Pantoea, Pseudomonas, Sphingomonas and

Xanthomonas genera (Rudolph et al., 1990), while pink ones are

Methylobacterium (referred as pink-pigmented, facultative methylotrophic bacteria) or Staphylococcus (Hirano and Upper, 2000). In tropical plants, nitrogen-fixing bacteria are found to be part of the phyllosphere community, and

Beijerinckia and Azotobacter are the most common genera (Hirano and Upper,

2000). On the leaf surfaces of grasses, Pseudomonas, Xanthomonas,

Clavibacter and Curtobacterium are the most common genera (Table 1.1)

(Dickinson et al., 1975). Other groups found on grass leaves are considered as

23 transient visitors, such as Microbacterium (Behrendt et al., 2002) and enterococci

(Ottetal.,2001).

1.4.1.2 Yeast phyllosphere communities Yeasts, commonly found on leaf surfaces, are the most abundant type of

fungi in this environment (Dickinson, 1982; Fokkema et al., 1975). Both

ascomycetous and basidiomycetous yeasts are normally part of the phyllosphere,

and Candida, Cryptococcus, Rhodotorula, , Tilletiopsis,

Aureobasidium and Torulopsis are the most regularly encountered genera

(Dickinson, 1982). Most yeasts associated with leaves produce extra-

polysaccharide capsules that presumably help them to adhere to and colonize

leaves and perhaps reduce desiccation (Andrews and Buck, 2002). Yeasts are

thought to provide protection on leaf surfaces against plant pathogens (Fokkema

et al., 1975), and the biocontrol activity of many epiphytic yeasts have been

investigated (Dik and Elad, 1999).

Grass leaves are colonized with the same genera of yeasts commonly found on leaves of other plants (Table 1.2). The species identified usually belong to the genera Rodotorula and Cryptococcus. As with other plants, epiphytic yeasts on grasses are very numerous reaching 2 x 105 CFU/cm2 in spring (Buck and Burpee, 2002). Investigation of yeast communities of turfgrasses has shown that community size and composition are affected by the incidence of foliar diseases, such as dollar spot (Allen et al., 2004) and fungicides (Buck and

Burpee, 2002). In general, application of fungicides reduce yeast populations, but after repeated applications, yeast counts were found to have increase and the

24 yeasts isolated were less sensitive to the fungicide than those obtained from untreated plants (Buck and Burpee, 2002).

1.4.1.3 Fungal phyllosphere communities Filamentous fungi also form part of the phyllosphere (Table 1.3), and the filamentous fungal genera Alternaria and Cladosporium are recorded as inhabitants of a large number of plants studied (Dickinson, 1982).

Representatives of these genera normally are saprophytes, but they can become pathogenic under certain conditions (Dickinson, 1982), particularly when plant tissues are weakened or become senescent. Other filamentous fungal genera, such as Aspergillus, Penicillium, Epicoccum, Myrothecium, Pilobolus and

Stemphylium, have been found as phyllosphere invaders rather than inhabitants, whereas fungal genera, such as Ascochyta, Leptosphaeria, Pleospora and

Phoma, are usually on the leaves, but only grow in highly senescent tissues

(Atlas and Bartha, 1987; Petrini, 1986). In addition, endophytic fungal genera, such as Epichloe, Phyllosticta, Linodochium and Rhabdocline, are listed among phyllosphere fungi (Petrini, 1986).

Reports of fungal species related to grass leaves have generally focused on either mutualism or pathogenesis. A great number of grass endophytes have been identified and investigated for their mutualistic properties, but the main focus of foliar fungi has been on pathogenic epiphytes, while innocuous species are generally ignored. Thus, non-pathogenic filamentous fungi of the phyllosphere of grasses are seldom examined.

25 1.5 Assessment of microbial diversity Contrary to the straightforward approach of animal or plant measurement of

biodiversity which involves counting and identifying individuals, this is not feasible for microbial biodiversity (Scow et al., 2001) because their small size and limited differentiating morphological features does not permit easy counting or identification. Consequently, numerous methods for characterizing microbial communities have been developed. Early studies were based on culturing microorganisms on various media followed by phenotypic identification. Although these methods produced results, it was estimated that no more than 5% of species were being detected using the culture-based approach (Scow et al.,

2001). Molecular biology has provided new methods involving analysis of DNA, with some protocols not requiring culturing. The new methods have also brought along their own problems and biases, but they provide a complement to the traditional methods, giving a better understanding of microbial biodiversity.

1.5.1 Traditional approaches Direct counting of microbes can be done by using a microscope or by plate counts. In the first method, individual cells can be counted based on morphological differences which can be facilitated by the use of dyes and fluorescent compounds (Scow et al., 2001). The second method exploits the fact that individual cells grow into a visible colony making it possible to count viable cells or the number of colony forming units.

Traditional assessment of microbial biodiversity is popular because it does not require expensive equipment or reagents, and because it gives valuable

26 information quickly. However this method underestimates biodiversity because

the counts are limited to cultivatable organisms excluding all that are not suitable

for growth under lab conditions (Scow et al., 2001). This problem is referred as the "great plate count anomaly" (Stanley and Konopka, 1985), and it affects the estimation of diversity and number of microorganisms. Species that cannot grow

under regular laboratory conditions of temperature, pressure and humidity, would

not be counted nor identified. Microbial populations from extreme environments would be highly underestimated when using plating conditions that are common for isolations from other environments.

Obvious factors that can affect the accuracy of the plate counts are pressure and temperature, but there are other factors that are harder to consider.

For instance, the nutrient levels of microenvironment seem to affect the reliability of the plate counting. As Stanley and Konopka (1985) explains, traditional counting from eutrophic (high nutrient) environments represent up to 90% microorganisms found with non traditional methods. In contrast, when counting is perform on samples from oligotrophic (low nutrient) environments, the recovery may be only 1% of that estimated by other enumeration methods. Nevertheless, plate counting is still a useful resource to establish microbial populations of poorly studied environments before using more sophisticated and expensive procedures.

1.5.2 Molecular approach Molecular methods generally focus on nucleic acid sequences. The nucleic acid-based method can be divided into those that involve PCR and those that do

27 not. With PCR methods, a specific gene or region is amplified from a mixed pool

of nucleic acids, and the PCR product can either be sequenced and compared

with previous sequences or separated differentially by electrophoresis creating a

banding pattern (Scow et al., 2001). There are some approaches that target the

cellular lipid composition, but they are limited to specific microorganisms (Scow

et al., 2001). For example, fatty acid methyl ester (FAME) analysis involves using

fatty acids from the sample that are methylated and analyzed by gas

chromatography (Kirk et al., 2004). The obtained profiles indicate the group of

microorganisms based on characteristic fatty acids of each group (Kirk et al.,

2004).

Several molecular methods have been designed using ribosomal DNA

(rDNA) sequences to elucidate phylogenetic relationships and community

composition. Sequences of rDNA are chosen to obtain information on community

diversity because rDNA is found in all forms of life and exhibits a high degree of

conservation (Yonath and Franceschi, 1999).

The basic structure of rDNA (Figure 1.7) is a cassette or unit that can be

repeated tens to thousands of times in a genome. In eukaryotes, each unit

consists of the small subunit (SSU also known as the 18S), the internally transcribed spacer (ITS) 1, the 5.8S gene, the ITS2, the large subunit (LSU also

known as the 28S) and the intergenic spacer region (IGS). Prokaryotes also have

a cassette encoding the rDNA genes, which are repeated in several copies through the genome (Weider, et al., 2005). Also the prokaryotic rDNA is comprised of a small subunit known as the 16S gene (homologous to the

28 eukaryotic 18S gene) closely linked to the large subunit that contains the 5S and

23S rRNA molecules (Weider, et al., 2005).

The differences between rDNA sequences permit comparative studies of phylogenetic relationships over a wide range of taxonomic levels (White et al.,

1991). Indeed, SSU and LSU have been used to study high level phylogenetic relationships (e.g. families and above), while ITS and IGS have been selected for lower levels (e.g. genera and species) (Soltis and Soltis, 1998). Although the ITS is transcribed but not incorporated into ribosomes, this region seems to be involved in some aspects of ribosomal maturation, and is subjected to some levels of evolutionary constraint (Baldwin et al., 1995). This characteristic makes the ITS region suitable for lower taxonomical studies, such as genera or less closely related species within a genus (Soltis and Soltis, 1998). The least conserved region of the rDNA is the IGS, which has been found to be useful in separating closely related species (Soltis and Soltis, 1998).

Many methods have been developed to study polymorphism in the rDNA or other genes. Basically, all methods rely on electrophoresis to obtain different band patterns or fingerprints. The most common methods that have been applied to estimated microbial communities in several environments (Beebee and Rowe,

2004) are restriction fragment length polymorphism (rDNA-RFLP) also known as amplified ribosomal DNA restriction analysis (ARDRA), terminal restriction fragment length polymorphism (T-RFLP), denaturing gradient gel electrophoresis

(DGGE) and thermal gradient gel electrophoresis (TGGE).

29 RFLP is based on the presence of different DNA fragments obtained after

DNA digestion by restriction enzymes. The selected gene is amplified and digested by a series of restriction enzymes that cut in specific sequences generating fragments from the original sequence. Then, the different pieces of

DNA are separated by size using electrophoresis generating a characteristic band pattern. Band patterns are used to compare the same region of DNA of diverse organisms, establishing differences or similarities (Scow et al., 2001).

This method is based on the presence of polymorphisms in the sequences, which are recognized by particular restriction enzymes. This method has been used for rapid comparison of rDNAs (Liu et al., 1997) because it offers a high level of resolution (Tiedje et al., 1999). However, in very complex communities, which are highly diverse and have non-dominant populations, such methods may be inaccurate because it can be difficult to resolve all the bands that are generated (Tiedje et al., 1999). A technique used to avoid this particular problem requires the generation of clone libraries of the PCR products (Liesack and

Stackebrandt, 1992). This method yields a better resolution of the community and the genes can be easily sequenced since they are already cloned, but it can be time-consuming and costly (Tiedje et al., 1999).

T-RFLP is another technique for assessing microbial populations that is quite similar to RFLP. In this method a fluorescent molecule is tagged to a primer, and therefore after digestion of the amplified product with restriction enzymes, only one fragment known as a terminal restriction fragment (T-RF) will be recorded when visualized with the specific fluorescent exiting substance

30 (Blackwood et al., 2003). This method was developed by Liu et al. (1997) to assess the community structure of complex environments avoiding the tedious process of generating libraries associated with the ARDRA technique. However, this method provides a lower resolution than RFLP because only one restriction fragment is considered in each run (Tiedje et al., 1999).

Fischer and Lerman, (1979; 1983) established the denaturing gradient gel electrophoresis (DGGE) technique by using the differential-melting behavior of

DNA molecules from diverse species. Thus, when a pool of PCR products of a specific gene is subjected to electrophoresis separation in a denaturing gradient, the molecules will stop migrating at different rates according to their melting characteristics which varies according with sequence (Muyzer, 1998), The denaturing gradient is created by an increasing concentration of a chemical compound such as urea and formmide or with increasing temperature (Muyzer and Smalla, 1998).

This DGGE method has the advantage that bands can be readily sequenced and used as group-specific hybridizing probes to detect particular microorganisms (Muyzer, 1998). DGGE has been used in many microbial ecology experiments in the last decade. It has helped to assess community complexity and community changes in several environments

The disadvantages of DGGE technique are the initial cost of the specialized equipment necessary to run the gels and the specialized reagents such as clamped primers (Hepburn and Miller, 1998). Other disadvantages include the difficulty in consistent gradient gels, low throughput (Henderson et al., 1997),

31 difficulties with PCR fragments over 400 bp (Bover, 2005), DNA extraction method selective issues, multiple rRNA operons with different 16S rRNA in the same population, and complex primer design to include GC clamps (Ferrari and

Carrera, 2005).

A combination of traditional and molecular techniques for the study of microbial communities can bring a better understanding of the complex microbial world. Especially in very poorly studied environments, a basic knowledge of the microbial diversity can be first addressed with culture dependent methods followed by detailed molecular studies.

1.6 Hypotheses and objectives

1.6.1 Hypotheses 1) The incidence of the turfgrass disease dollar spot is enhanced by single application of azoxystrobin after more than six months of application

2) The enhancement of dollar spot is an indirect effect of azoxystrobin on the composition of the microbial population of leaves as assessed by traditional

(plate count) and molecular techniques.

1.6.2 Objectives 1) To quantify the enhancement of dollar spot disease on turfgrasses caused by azoxystrobin-based commercial fungicides.

2) To refine a method of estimating the number of major epiphytic, culturable, microbial morphotypes from turfgrass leaves using leaf washing and colony counts on artificial media.

32 3) To assess the multi-season effect of a single application of azoxystrobin on the microbial communities of the phyllosphere of Agrostis stolonifera and Poa pratensis.

4) To assess the genetic diversity of the microbial phyllosphere community of A. stolonifera based on restriction fragment length polymorphism of 16S and ITS regions of rDNA.

5) To identify microbial inhabitants of A. stolonifera through sequencing of ribosomal DNA.

33 Table 1.1. Common phyllosphere bacteria of grasses. Species Host plant Type 16S Reference Acinetobacter spp. Lolium perenne Epiphyte yes Dickinson et al., 1975 Bacillus spp. Lolium perenne Epiphyte yes Dickinson etal., 1975 Clavibacter michiganensis Grasses Epiphyte no Behrendtetal., 2002 subsp. insidiosus C. michiganensis subsp. Grasses Epiphyte yes Behrendt etal., 2002 michiganensis C. michiganensis subsp. Grasses Epiphyte yes Behrendt etal., 2002 nebraskensis C. michiganensis subsp. Grasses Epiphyte yes Behrendt etal., 2002 sepedonicus C. michiganensis subsp. Grasses Epiphyte yes Behrendt et al., 2002 tessellarius Curtobacterium faccumfaciens Grasses Epiphyte yes Behrendt etal., 2002 pv betae C. faccumfaciens pv. oortii Grasses Epiphyte yes Behrendt et al., 2002 C. faccumfaciens pv. Grasses Epiphyte yes Behrendt etal., 2002 poinsettiae Curtobacterium albidum Grasses Epiphyte yes Behrendt etal., 2002 Curtobacterium citreum Grasses Epiphyte yes Behrendt etal., 2002 Curtobacterium luteum Grasses Epiphyte yes Behrendt etal., 2002 Curtobacterium pusillum Grasses Epiphyte yes Behrendt etal., 2002 Enterococcus faecalis Grasses Epiphyte yes Ottetal., 2001 Enterococcus mundtii Grasses Epiphyte yes Ott etal., 2001 Enterococcus casseliflavus Grasses Epiphyte yes Ott etal., 2001 Enterococcus faecium Grasses Epiphyte yes Ott etal., 2001 Enterococcus sulfureus Grasses Epiphyte yes Ott etal., 2001 Erwinia herbicola Lolium perenne Epiphyte yes Dickinson et al., 1975 Flexibacter spp. Lolium perenne Epiphyte yes Dickinson et al., 1975 Klebsiella spp. Lolium perenne Epiphyte yes Dickinson et al., 1975 Listeria spp. Lolium perenne Epiphyte yes Dickinson et al., 1975 Micrococcus luteus Lolium perenne Epiphyte yes Dickinson et al., 1975 Pantoea spp Grasses Epiphyte yes Behrendt etal., 1997 Pseudomonas cannabina Grasses Epiphyte yes Behrendt etal., 2003 Pseudomonas cedrina Grasses Epiphyte yes Behrendt etal., 2003 Pseudomonas congelans Grasses Epiphyte yes Behrendt etal., 2003 Pseudomonas fluorescens Lolium perenne Epiphyte yes Dickinson et al., 1975 Pseudomonas poae Grasses Epiphyte yes Behrendt etal., 2003 Pseudomonas rhodesiae Grasses Epiphyte yes Behrendt et al., 2003 Pseudomonas savastanoi pv. Grasses Epiphyte yes Behrendt etal., 2003 glycinea Pseudomonas synxantha Grasses Epiphyte yes Behrendt et al., 2003 Pseudomonas tremae Grasses Epiphyte yes Behrendt etal., 2003 Pseudomonas trivialis Grasses Epiphyte yes Behrendt etal., 2003 Pseudomonas veronii Grasses Epiphyte yes Behrendt et al., 2003 Rathayibacter iranicus Grasses Epiphyte yes Behrendt etal., 2002 Rathayibacter rathayi Grasses Epiphyte yes Behrendt etal., 2002 Rathayibacter toxicus Grasses Epiphyte yes Behrendt etal., 2002 Rathayibacter tritici Grasses Epiphyte yes Behrendt etal., 2002 Staphylococcus saprophytics Lolium perenne Epiphyte yes Dickinson et al., 1975 Stenotrophomonas spp Grasses Epiphyte yes Behrendt etal., 1997 Xanthomonas campestris Lolium perenne Epiphyte yes Dickinson et al., 1975 * Sequences of the 16S ribosc DNA were available on

34 Table 1.2. Common phyllosphere yeast of grasses. Species Host plant Type Reference Aureobasidium pullulans Phleum pratense Epiphyte Bab'eva & Sadykov, 1980 Cryptococcus albidus Phleum pratense Epiphyte Bab'eva & Sadykov, 1980 Cryptococcus flavus Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Cryptococcus laurentii Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Cryptococcus luteolus Phleum pratense Epiphyte Bab'eva & Sadykov, 1980 Cryptococcus macerans Phleum pratense Epiphyte Bab'eva & Sadykov, 1980 Pseudozyma antarctica Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Pseudozyma aphidis Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Pseudozyma parantarctica Agrostis palustris & Festuca Epiphyte Allen et al., 2004 arundinacea Rhodosporidium sp. Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Rhodosporidium capitatum Rhodotorula crocea Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Rhodotorula glutinis Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Rhodotorula mucilaginosa Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Rhodotorula rubra Epiphyte Sakaguchia dacyroidea Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Sporidiobolus pararoseus Agrostis palustris & Festuca Epiphyte Allen etal., 2004 arundinacea Sporobolomyces roseus Phleum pratense Epiphyte Bab'eva & Sadykov, 1980 Tilletiopsis cremea Phleum pratense Epiphyte Bab'eva & Trichosporon sp. Agrostis palustris & Festuca Epiphyte Sadykov, 1980 arundinacea Allen etal., 2004

Trichosporon pullulans Phleum pratense Epiphyte Bab'eva & Sadykov, 1980

35 Table 1.3. Common phyllosphere filamentous fungi of grasses. Species Host plant Type ITS* Reference Epichloe amarillans Agrostis, Sphenopholis, Endophyte yes Siegeletal., 1990 Calamagrostis sp. Epichloe baconii Agrostis, Calamagrostis Endophyte yes White, 1993; Leuchtmann sp. et al., 2000 Epichloe Brachyelytrum erectum Endophyte no Schardl and Leuchtmann, brachyelytri 1999 Epichloe bromicola Bromus sp. Endophyte no Leuchtmann and Schardl, 1998; Leuchtmann et al., 2000 Epichloe clarkii Holcus lanatus Endophyte yes White, 1993; Leuchtmann etal.,2000 Epichloe elymi Elymus sp. Endophyte yes Siegeletal., 1990; Schardl and Leuchtmann, 1999 Epichloe festucae Festuca, Lolium sp. Endophyte yes Siegeletal., 1990; Leuchtmann et al., 2000; Wilkinson etal., 2000 Epichloe glyceriae Glyceria striata Endophyte yes Cheplickand Clay, 1988; Schardl and Leuchtmann, 1999 Epichloe sylvatica Brachypodium sylvaticum Endophyte no Leuchtmann and Schardl, 1998; Leuchtmann et al., 2000; Brem and Leuchtmann, 2001 Epichloe typhina Many Endophyte yes Siegeletal., 1990; Leuchtmann and Schardl, 1998 Microbotryum sp. Agrostis palustris & Epiphyte yes Allen et al., 2004 Festuca arundinacea Neotyphodium Echinopogon ovatus Endophyte no Miles et al., 1998; aotearoae Panaccione et al., 2001 Neotyphodium Echinopogon ovatus Endophyte no Moon et al., 2002 australiense Neotyphodium Lolium arundinaceum Endophyte yes Cheplick and Clay, 1988; coenophialum Clay, 1990; Siegeletal., 1990 Neotyphodium Achnatherum inebrians Endophyte no Miles etal., 1996 inebriansd Neotyphodium lolii Lolium perenne Endophyte yes Siegeletal., 1990; Christensenetal., 1993 Neotyphodium Melica decumbens Endophyte no Moon et al., 2002 melicicola Neotyphodium Lolium pratense Endophyte no Craven etal., 2001 siegelii Neotyphodium sp. Lolium perenne, L Endophyte yes Christensenetal., 1993 arundinaceum Neotyphodium Poa huecu Endophyte no Cabraletal., 1999 tembladerae Neotyphodium Lolium pratense Endophyte yes Leuchtmann et al., 2000 uncinatum * Sequences of the ITS1 ribosomai DNA were: available on SenBan< k

36 Bridge <£=i i=£>Side Chain

Toxophore

Figure 1.1. General structure of strobilurins (Adapted from Huang et al., 2007).

37 * *, =*

# V

. w ,. • -'

1 cm

Figure 1.2. Dollar spot patch caused by Sclerotinia homoeocarpa on closely mowed Agrostis stolonifera, next to a Canadian dollar coin. This picture was taken July 2007 at the Guelph Turfgrass Institute, Guelph, Ontario.

38 \ ••

; ' " '

5

Figure 1.3. Dollar spot patch on Poa pratensis. The picture was taken July 2006 at the Guelph Turfgrass Institute, Guelph, Ontario.

39 40 -*•"".

~ ^ fcj »2*

!... tdcif .•-:.;. :*ss»• « » ,J HHi V

-W^ Figure 1.5. Web-like mycelium of Sclerotinia homoeocarpa on Agrostis stolonifera present with morning dew. (Courtesy of T. Hsiang).

41 / -«*a- *w • jap m t • - : j££i?ft L*"^ **i7 2*-- A. •

//

Figure 1.6. A culture of Sclerotinia homoeocarpa grown at 25C for 10 days on potato dextrose agar plate. (Courtesy of A. Liao).

42 A ITS1 ITS2 IGS

SSU 18S 5.8S LSU 28S SSU 18S V rDNA cassette B

23S 16S 5S V_

rDNA operon

Figure 1.7. Eukaryotic ribosomal DNA (rDNA) gene cassette showing the Small Subunit (SSU), Large Subunit (LSU), Internal Transcribers Spacers (ITS) and Intergenic spacer (IGS). Gray boxes represent highly conserved regions. B) Prokaryotic ribosomal gene showing the 23S, 16S and 5S subunits closely encoded. (Redrawn from Soltis and Soltis, 1998).

43 CHAPTER 2. EFFECT OF AZOXYSTROBIN ON DOLLAR SPOT DISEASE DEVELOPMENT

2.1 Introduction The turfgrass disease dollar spot is caused by the ascomycete Sclerotinia homoeocarpa. It is widespread in cool season grasses worldwide (Smith et al.,

1989). The small, bleached and sunken patches characteristic of this disease pose an aesthetic and payability problem on golf courses since ball speed and trajectory are affected by the altered playing surface. Thus, golf course superintendents spend a great deal of their pest protection budget controlling and preventing this disease (Vargas, 2005), which can lead to other problems such as fungicide resistance development. Sclerotinia homoeocarpa has developed resistance toward demethylation inhibiting (DMI) fungicides (Golembiewski et al.,

1995; Vargas et al, 1992; Hsiang et al. 1997), dicarboximide fungicides

(Detweiler et al. 1983), and benzimidazoles (Warren et al., 1974; Burpee, 1977).

In addition to fungicide resistance issues, interactions with fungicides targeting other diseases may also increase disease spot incidence. Disease increase could be due to resurgence or enhancement, which are different, although they are intimately linked and often referred to as the same effect. The first refers to the chemical effect in which the target disease is successfully controlled, but once the chemical effect fades, the disease surges with more intensity than on untreated areas. Disease enhancement occurs when the target disease is controlled, but the incidence of a non-target disease is higher than in areas where the chemical was not applied (Vincelli, 2002). Disease enhancement or resurgence is a phenomenon which is not well understood.

44 Examples of this can be found in the turfgrass environment such as the increase incidence of summer patch (caused by Magnaporthe poae) in P. pratensis after applications of chlorothalonil (Vincelli, 2007), or the increased severity of Pythium blight after benomyl applications as described by Warren et al. (1976). Dollar spot in particular has been reported to be enhanced by the application of azoxystrobin (Cook and Hsiang, 2002 and 2004; Dernoeden, 2000; Hsiang and

Cook, 2006; Schumann, 2003; Vincelli, 2007). These reports give observational and anecdotal details on disease resurgence or enhancement, but they have not examined the mechanisms behind this effect, nor have they probed the longer term effect of applications of some fungicides.

The mechanisms proposed to produce resurgence and disease enhancement can be segregated in two main factors: direct effects on the plants, and effects on phyllosphere microbial communities (Vincelli, 2007). A direct pesticide effect on the plants includes changes to their physiology, such as hormonal changes, carbohydrate content changes and increased plant stress

(Couch, 1995).

The second common explanation for the occurrence of disease enhancement or resurgence after fungicide application is the suppression of phyllosphere antagonist microorganisms (Smith et al., 1989) Most of the turfgrass pathogens that have been shown to increase in incidence after pesticide application are foliar pathogens attacking and infecting plants through their leaves, such as the following: red thread (Dernoeden et al., 1985), melting out (Dernoeden and Mcintosh, 1991), Rhizoctonia yellow patch (Smiley, 1981)

45 and Helminthosporium leaf spot (Jackson, 1970). The pathogens causing these diseases are part of the phyllosphere and this environment plays a key role in their infection processes. The inhabitants of this microenvironment are affected by conditions such as temperature and nutrient availability (Lindow and Brandl,

2003). Additionally, interactions among the members of the microbial community itself have been shown to affect the survival of certain species (Belanger and

Avis, 2002). The biological control of leaf pathogens relies on the antagonism that certain phyllosphere microorganisms exert on leaf pathogens (Belanger and

Avis, 2002).

Pest resurgence is not exclusively found with plant diseases. Several reports of insect resurgence after insecticide application are common. This phenomenon is better understood than plant disease resurgence. The reappearance of targeted insects occurs mainly because the insecticide reduces populations of antagonists and natural enemies which normally keep the pest population under certain levels (Dutcher, 2007). A well known example is the resurgence of brown plant hopper in rice (Aquino and Heinrichs, 1979). Natural enemies of the brown plant hopper, such as mired bugs, spiders and beetles, are killed by application of broad-spectrum insecticides (Tanaka et al., 1999). Insect resurgence is caused not only by reduced antagonist populations. Other identified causes include behavior, dispersal, development and fecundity of the pest, being affected by pesticide applications (Dutcher, 2007). Therefore, as with plant pathogens, resurgence involves a complex combination of factors that may play different roles depending on the system studied.

46 Dollar spot disease has been repeatedly reported to be enhanced in

bentgrasses treated with commercial fungicides. This includes casual

observations in the golf industry (Demoeden, 2000; Schumann, 2003) as well as

specific experimentally supported results (Hsiang and Cook, 2006; Vincelli, 2002).

Carboxin-derived fungicides such as flutolanil (Vincelli, 2007), strobilurins such

as azoxystrobin (Dernoeden, 2000; Hsiang and Cook, 2006; Schumann, 2003),

pyrrole fungicides such as fludioxonil, chitin synthetase inhibitors such as

polyoxin D (Vincelli, 2007) and organochlorine fungicides such as

pentachloronitrobenzene (Smiley, 1981) are some of the fungicides reportedly

related to dollar spot disease resurgence or enhancement.

In order to investigate the enhancement phenomenon, the effects of azoxystrobin on injury levels and dollar spot incidence were assessed. Long term trials on two turfgrass species, were conducted in two consecutive years treated with different azoxystrobin rates, to obtain results under different conditions.

Furthermore, to observe if azoxystrobin has a short term effect, monthly applications were made to separate plots during the growing season. Because most previous reports were anecdotal, the objective of this study was to obtain numerical data to examine if there is a link between azoxystrobin application and the short term or long term dollar spot enhancement previously observed.

2.2 Materials and Methods

2.2.1 Study site Plots were established at the Guelph Turfgrass Institute (GTI) at the

University of Guelph (Guelph, Ontario, Canada). The first experiments ran from

47 November 2005 to October 2006 (11 months), and the second set of experiments ran from November 2006 to August 2007 (9 months). At the GTI, two areas of creeping bentgrass (Agrostis stolonifera) and one area of Kentucky bluegrass (Poa pratensis) were chosen (Figure 2.1). These areas were maintained using cultural regimes commonly practiced at golf courses in Ontario for putting greens, fairways or fringes.

One set of plots was placed on an area of A. stolonifera referred to as the pathology green. This area had been constructed in 1994 on a 30 cm soil base of

80:20 sand.peat (v/v) mixture on top of a 10 cm drainage layer of gravel following specifications recommended by the United States Golf Association (USGA). The pathology green was fully surrounded by large trees and shrubbery. It was initially seeded and has been frequently overseeded with PennCross®

(Pennington Seed, Madison, Georgia, USA) bentgrass (A. stolonifera). The plot area, however, currently contains approximately 30% annual bluegrass (), which is commonly found as an invader of A. stolonifera swards. The area was irrigated as needed and mowed daily at 5.5 mm height with clippings removed.

On the pathology green in 2006, homogeneous methylene urea complex

Country Club (N-P-K: 18-3-18) (Lebanon Seaboard Corp., Lebanon,

Pennsylvania, USA), was applied monthly as a fertilizer from May to August at 35 kg N/ha, and in September and November, it was applied at 50 kg N/ha.

Additionally, in July 2006, the area was topdressed with 1 cm of calcareous sand

(west half) or non-calcareous sand (east half) and aerated twice with 0.5 inch

48 (1.25 cm) diameter hollow tines on a 2 inch (5.8 cm) center. Similarly, in

September 2006 the area was topdressed twice. Only one application of pesticide occurred during the trial, and this was in mid-July 2006 when the green was sprayed with the insecticide Success SC (Dow Agrosciences, Indianapolis,

Indiana, USA) at 1.0 ml/100 m2. In 2007, the fertilizer Country Club (N-P-K: 18-3-

18) was applied at 50 kg N/ha each month from May until August, and in

September, 2007, Urea Micro-Prill (N-P-K: 46-0-0) (Dyno Nobel, Salt Lake City,

Utah, USA) was applied twice. There were no pesticide treatments in 2007 on the pathology green.

The other A. stolonifera plots were in an area called the native sand fairway.

This is a push-up or natural soil green built in 1994 with local Fox sandy loam, and seeded with PennCross® (A stolonifera), but it currently contains approximately 20% P. annua. It was mowed every other day at 10 mm height and was topdressed using the methods and amounts previously stated for the pathology green. Fertilizers Scotts Pro Turf (N-P-K: 21-3-21) (Scotts Miracle-Gro,

Marysville, Ohio, USA) at 25 kg N/ha or Country Club (N-P-K: 18-3-18) at 25 kg

N/ha were applied in May 2006, followed by monthly fertilizations with Country

Club (N-P-K: 18-3-18) at 35 kg N/ha from June to August 2006. In September and November 2006, the green was fertilized with Country Club at 50 kg N/ha. In

2007, the same fertilizer was used at 50 kg N/ha once a month from May to

August 2007. There were no pesticide treatments in 2006 or 2007 on the native sand fairway plot areas during the course of these experiments.

49 Another set of plots was on a sward of P. pratensis. This area was located on a fringe of a putting green and had been sodded in 1994 on native soil. It was irrigated as needed and cut at 3.8 cm height every other day, and the clippings left in place. The fertilizer Agromart® 25-4-10 (Agromart Group, London, Ontario,

Canada) was applied at 40 kg N/ha in May, August, September and November

2006. No pesticide was applied to this area, except for designated treatments as part of these experiments.

2.2.2 Field Experiments Five set of plots were established in this study at the Guelph Turfgrass

Institute in 2005 and 2006. The plots were established on different areas of the pathology green, native sand fairway and the P. pratensis area surrounding the upper green (Figure 2.1). All plots consisted of areas of 16 m2 (eight plots of 1 m x 2 m) marked with orange spray paint. Plots with odd numbers (1, 3, 5 and 7) were used as the untreated, while even numbered plots (2, 4, 6 and 8) were treated with fungicide (Figure 2.2).

Plots were visually evaluated weekly throughout the course of the trials.

Evaluations consisted of assessing plant injury and incidence of dollar spot.

Symptoms of other diseases, abiotic injury and weed presence were also recorded. Plant injury was assessed as the percentage of plot area with yellow, straw-like or dead leaves caused by all biotic and abiotic agents, such as temperature, drought, insects, pathogens or unknown reasons. Dollar spot incidence was estimated by counting all characteristic dollar spot patches. On the

A. stolonifera plots, the patches counted were less than 4 cm across, round,

50 yellowish and sunken spots (Figure 1.2), while P. pratensis patches were roughly round, about 10 cm in diameter (Figure 1.3) with hourglass symptoms on leaves (Figure 1.4). Once these distinct characteristics had disappeared leaving non-specific straw-like or dead grass, the area was no longer regarded as dollar spot, but was still included in general plant injury assessment as before.

2.2.2.1 Trials There were two types of trials: single season and multi-season. The four multi-season trials were established and treated in the fall (two in 2005 and two in 2006), and evaluations done monthly or weekly until the next fall. On 28

November 2005, the first multi-season trial (2005-2006) was established on the southwest portion of the native sand fairway (A. stolonifera) (Figure 2.1 a). A second trial was established at the same time near the southeast fringe of the upper green in a P. pratensis area (Figure 2.1 b). On 25 November 2006 a third multi-season trial (2006-2007) was established on the southwest portion of the native sand fairway (Figure 2.1 c). The final multi-season trial was established at the west end of the pathology green (Figure 2.1 d).

Short term trials were set up from May to August 2006, and evaluated for up to three months. With the same plot design as the multi-season trials, four trials were established at the beginning of each successive month from June to

September 2006 on the southeast portion of the pathology green (Figure 2.1 e).

The 2005-2006 multi-season plots and the 2006 summer plots were treated with Heritage® 50WDG (Syngenta) applied at the maximum rate used for anthracnose blight of 12 g/100 m2 (Anonymous 2005) giving an active ingredient

51 (azoxystrobin) rate of 6 g/100 m2. The two subsequent multi-season trials of

2006-2007 were treated with Heritage® MAXX® at 126 ml/100 m2 (12 g/100m2 of

active ingredient), which is the amount recommended for gray snow mold

(Anonymous, 2005). The fungicide was applied with a wheel-mounted

compressed air boom sprayer at 140 kPa in water at 10 1/100 m2 using Lurmark

03-F110 nozzles (Hydro EU, Cambridge, Cambridgeshire, United Kingdom]). The

same amount of water was applied to the untreated control plots.

2.2.2.2 Weather data Weather data were obtained for the duration of the trial, specifically snow

cover depth, temperature and precipitation. Snow cover depth of winter 2005-

2006 in Guelph was visually estimated on a daily basis by laboratory colleagues

Champa Wijekoon and Lynn Tian. For winter 2006-2007, snow cover records

were obtained from the Waterloo University weather station web site

(http://www.uwaterloo.ca/), which uses a sonic range sensor to measure the

snow depth. Temperature (ambient air) and precipitation (tipping bucket) data

were also obtained from the same weather station from October 2005 to October

2007.

2.2.3 Statistic analyses Plant injury percentage and the dollar spot incidence data were subjected to

ANOVA using SAS (SAS Institute, Cary, North Caroline, USA) PROC GLM.

When a significant treatment effect was observed in the ANOVA (p=0.05), the

means were compared using Fisher's LSD (Least Significant Difference) at

p=0.05 (Appendix A).

52 2.3 Results

2.3.1 Weather data The winter of 2005-2006 showed both cold and warm extremes (Figure 2.3).

There were two periods of snow cover, in December with up to 15 cm, and in

February with up to 35 cm. However, the snow melted in January with temperatures up to 10C. During the extremely cold temperatures in February, the grass was protected by snow cover. The lack of continuous snow cover prevented high levels of grey snow mold development, and the quick warmup in

January and in March caused snow to melt very quickly which is not favorable for the development of pink snow mold (Hsiang et al. 2007)

The winter of 2006-2007 was unusual in extremes of warmth and cold

(Figure 2.4). Throughout December, temperatures remained warm with many days exceeding 15C. Temperatures remained warm until 9 Jan 2007, when snowfall occurred. Due to very cold temperatures throughout the remainder of

January and February and into the early part of March, snow cover was present on the plots until mid-March. The cold temperatures and snow cover, as well as the greens covers, allowed severe levels of pink snow mold to develop, with a moderate level of gray snow mold. Snow cover for up to two months favors M. nivale, which causes pink snow mold while periods over 3 months favors Typhula snow molds (Hsiang et al., 1999)

In summer the gap between highest and lowest levels was greater in 2007 than 2006. Additionally, in 2007 temperatures remained higher until the end of

September with maximum temperatures exceeding 30C. The warmer weather in

53 2007 was conducive to summer disease development. By the end of September

2006, temperatures dropped to an average of 15C. Precipitation levels were also different between the years. In 2005-2006 precipitation occurred more frequently and at higher levels than in 2006-2007. In 2006-2007 precipitation mostly occurred in the fall of 2006 with some rainfall in summer (Figures 2.3 c and 2.4 c).

2.3.2 Plant injury

2.3.2.1 Native sand green multi-season trial of 2005-2006 The A. stolonifera multi-season plots 2005-2006 on the southern edge of the native sand fairway (Figure 2.1 a) received a simple treatment Heritage® on

28 November 2005. The plot area at time of setup was generally free from disease (see Appendix B for records on seasonal maintenance) as can be seen in Figure 2.5. The plots were evaluated for injury after snow melt in March 2006, and winter damage (yellowing) was evident on these plots (Figure 2.6) with an average 15% area affected on both treated and untreated plots. There were no obvious signs of gray or pink snow mold, and the injury was considered to be mainly of abiotic origin.

By early April, the level of injury dramatically increased (Figure 2.7) due to exposure to cold winds (Figure 2.3 a). Large yellow areas covered nearly 50% of both treated and untreated plots. By mid-April with warmer temperatures (Figure

2.3 a), plants started to recover, and the level of injury decreased to 30% (Figure

2.7). This was seen in both treated and untreated plots with a complete recovery from winter injury by the beginning of June 2006 (Figure 2.7).

54 Throughout summer and the early part of fall, injury levels which included all biotic and abiotic sources, remained under 15% in both untreated and treated plots (Figure 2.7). The minor fluctuations in injury during this period were caused by abiotic stresses, such as drought produced by heat in the summer (Figure 2.3 a) or biotic stress of pathogens, such as S. homoeocarpa causing dollar spot and

Rhizoctonia solani causing brown patch. Although there were no great differences in total plant injury levels between treated and untreated plots during the growing season, on three evaluations dates, the injury levels were significantly greater on plots treated with Heritage®: 22 June 2006, 17 August

2006, 13 October 2006 (Figure 2.7). Injury levels diminished and almost completely disappeared by 22 October 2006.

2.3.2.2 Poa pratensis multi-season trial of 2005-2006 Multi-season plots were established on healthy P. pratensis in November

2005 (Figure 2.8) beside the south-west border of the upper green (Figure 2.1 b). Plant injury percentage was not significantly different across the plots of this trial in November 2005 (see Appendix B for records on seasonal maintenance).

Immediately after snowmelt (Figure 2.9) 30% of all the plot area was yellowish and, in less than 15 days, the area covered with yellow grass reached 100%

(Figure 2.10) probably because of cold desiccating winds (Figure 2.3). The grass remained yellow until the first week of April when it started to grow again, repopulating the plots with healthy green leaves until the green coverage reached -95% on 25 May 2006 (Figure 2.10). In June, the injury levels remained steady with small fluctuations between 5 to 25% of plot area (Figure 2.10). The

55 grass remained green through July and the first half of August 2006. However, by

25 August 2006, plant injury, caused mainly by dollar spot disease, exceeded

30%, but then steadily decreased within two weeks to the previous 5-25% range for the remainder of the growing season (Figure 2.10). In October 2006, the grass did not show any injury, and leaves were dark green across all plots.

2.3.2.3 Native sand fairway multi-season trial of 2006-2007 Plots were established on 25 November 2006 at the native sand fairway and showed no injury (Figure 2.11). These A. stolonifera plots were established on the southwest portion of the native sand fairway (Figure 2.1 c). They were treated with Heritage® 12 g a.i./100 m2 or water and evaluated every two weeks until snowfall and weekly in the growing season.

After snowmelt in March 2007, the treated plots had significantly less plant injury (~10% area affected) than the untreated plots (~25% area affected)

(Figures 2.12 and 2.13). Treatment with the fungicide effectively controlled snow molds. The results of treatments continued to be observable until the end of April when the temperature started to rise, and turfgrass plants of both treatments greened up, reducing injury levels below 15% (Figure 2.13).

In May 2007, the plots had a healthy appearance with no major differences between treatments. This trend continued throughout June until levels of treated plots began to show higher levels of plant injury than the untreated plots. The major reason for the high injury levels observed at the peak of the golf playing period was the development of dollar spot. From mid-June until the last evaluation in September, plots treated with azoxystrobin maintained significantly

56 higher levels of injury than untreated plots. The maximum difference occurred in

August when injury on treated plots was more than twice than the untreated plots

(Figure 2.13).

2.3.2.4 Pathology green multi-season trial of 2006-2007 The pathology green multi-season plots of 2006-2007 (Figure 2.14) were located at the west end of the pathology green (Figure 2.1 d) (see Appendix B for records on seasonal maintenance). These plots showed similar plant injury levels to the native sand fairway plots of 2005-2006 throughout the trial. Plant injury was higher for untreated plots in spring due to pink snow mold disease

(Figure 2.15), and higher for treated plots in summer (Figure 2.16) mainly because of dollar spot disease.

The recovery period for untreated plots after the winter was longer on the pathology green than on the native sand fairway, even though the levels of injury were ~20% higher in the native sand fairway immediately after the snow cover ended (Figure 2.16). The period in which the treated areas showed significantly higher damage from winter injury compare to the untreated plots was longer on the pathology green plots. The differences were noticeable on the pathology green plots in mid-May while they were seen on the native sand fairway plots by the end of June. The difference between treated and untreated plots in the summer was more marked on the native sand fairway plots than on the pathology green plots.

57 2.3.3 Dollar spot

2.3.3.1 Native sand fairway multi-season trial of 2005-2006 The pattern of dollar spot incidence in the native sand fairway multi-season

plots 2005-2006 was completely different than that for general plant injury. Dollar

spot patches began to appear on the plots in 1 June 2006, and immediately a

difference between treated and untreated plots was apparent and significant

(Figure 2.17). From the first appearance until the end of summer, dollar spot

counts on treated plots remained approximately 50% greater than untreated plots.

Only on 13 July 2006 were the counts not significantly different, even though the

mean count for the treated plots was higher than that of the untreated plots.

Treated and untreated plots had the same dollar spot incidence by the end of

August 2006 (Figure 2.17).

2.3.3.2 Poa pratensis multi-season trial of 2005-2006 Dollar spot disease on the P. pratensis plots was first observed on 20 July

2006 at low levels (Figure 2.18). The disease continued to spread reaching 6 spots per square meter, and remained at that level through July until mid August.

On 25 August 2006, the disease level doubled (from 6 to 12 spots/m2) in

untreated plots, and tripled in treated plots (from 6 to 18 spots/m2). On this recording date and 25 August 2006, dollar spot incidence was significantly greater in treated plots than in untreated plots.

2.3.3.3 Native sand green multi-season trial of 2006-2007 Dollar spot disease levels on the native sand fairway plots of 2006-2007 had almost the same pattern in treated and untreated plots (Figure 2.19). Dollar

58 spot disease appeared on 21 June 2007, reaching the highest incidence one

month later. Dollar spot incidence remained steady until the last evaluation when

it decreased. Although disease on untreated and treated plots followed the same

up and down patterns, they were significantly different in levels of dollar spot

incidence. From the beginning of the experiment, treated plots showed more

spots. The ratio varied through the growing season from 2 to 3 times more spots

on the treated plots than untreated plots. These were observed until the levels

plateaued on 19 July 2007 at 80 spots/m2 in the treated plots and 40 spots/m2 in

the untreated plots until the end of observation of 6 August 2007 (Figure 2.19).

2.3.3.4 Pathology green multi-season trial of 2006-2007 Dollar spot disease was first observed on 1 June 2007 on the A. stolonifera

pathology green multi-season trial of 2006-2007. This was earlier than was observed for the A. stolonifera native sand fairway trial of 2006-2007. Ninety days later dollar spot incidence reached a maximum of 80 spots/m2 (Figure 2.20 and 2.21). As on the native sand fairway plots, the disease levels increased until

late August, after which they remained constant. Dollar spot incidence was greater on the treated plots. Dollar spot incidence was observed to increase in treated and untreated plots, but there were higher numbers in the treated plots

(Figure 2.21). Differences were found from the first appearance of dollar spot until the last recording date. The first dollar spot record on 1 June 2007 showed that treated plots had over 5 times more spots than untreated plots, but the difference was reduced by the end of July 2007. However, values of dollar spot

59 incidence continued to be less in untreated plots that in those treated with the fungicide.

2.3.3.5 Summer plots of 2006 Four plots were set up in consecutive months on the growing period of turfgrass in the pathology green (Figure 2.1 e). Plant injury and dollar spot incidence were evaluated weekly. Unfortunately, all plot areas (treated and untreated) suffered extensive plant injury from abiotic stresses that killed over

75% of the plants (Figure 2.22). These areas did not recover and consequently the short term effect of azoxystrobin on dollar spot incidence could not be assessed. No data is presented from these plots and no further mention is made of them.

2.4 Discussion Plant injury percentage and dollar spot incidence were found to be affected by multiple factors such as plant height, (pathology green vs. native sand fairway trial multi-season trials of 2006-2007), environmental conditions (2005-2006 vs.

2006-2007 multi-season trials) and plant species (A stolonifera and P. pratensis).

Plant species was the major determinant of the total plant injury and incidence of dollar spot.

The timing of plant injury (which includes dollar spot injury) was similar in both species differing only in the levels reached. After winter, P. pratensis plots were almost completely yellow (90%) from abiotic sources (Figure 2.10), while A. stolonifera reached 60% (Figure 2.7). Snow mold was scarce and did not make a large contribution to the plant injury recorded. Plots of both species promptly

60 recovered in spring, with less than 10% injury (7-8 of weeks after snow melted).

In summer, injury levels of both turfgrass species were different, because dollar spot began to grow causing more injury on A. stolonifera than P. pratensis, although the numbers of spots were similar at that time. This occurred because the area covered by a single spot on P. pratensis was ~12 times the spot area of closely mowed grasses, such as A. stolonifera (~100 cm2 vs. ~7 cm2 respectively).

As different species, A. stolonifera and P. pratensis have different characteristics that make them naturally more resistant to some diseases than others and tolerant of different environmental and cultural conditions. For instance, Gaeumannomyces graminis causes the disease, take-all patch, of

Agrostis spp., but not of Poa spp. (Smiley et al, 2005). It is not surprising then, that dollar spot disease development was different on P. pratensis plots than on

A. stolonifera plots at the same time of the year. In stands of P. pratensis, dollar spot patches are 6 to 12 cm in diameter (Vargas, 2005), while on bentgrass putting greens, mature patches average 3 cm in diameter (Charbonneau and

Hsiang, 2003). Consequently, the maximum number of dollar spots per square meter will be less in lawns (P. pratensis) than on golf greens (A. stolonifera) by virtue of size patch alone. However, the environmental conditions seemed to have a greater effect on the levels of dollar spot since the maximum value of dollar spot on untreated plots in A. stolonifera was 50 spots/m2 in 2006-2007

(Figure 2.19), while it was 17 spots/m2 in 2005-2006 plots (Figure 2.17).

61 In addition to the size, inherent traits of each plant species affect dollar spot incidence. For A. stolonifera, patch numbers of 126 spots/m2 up to 380 spots/m2 have been reported in different areas of the United States (Tredway and Butler,

2004; Lee et al., 2003). Liu et al. (1995) in a study completed at the Cambridge

Research Station located in Ontario (University of Guelph) reported a maximum spot incidence of over 175 spots/m2 in the uninoculated untreated plots. In a previous study conducted at GTI, the maximum dollar spot incidence was 170 spots/m2 on uninoculated plots (Cook and Hsiang, 2002).

Studies of dollar spot on P. pratensis are scarce, and the disease incidence is not usually recorded as number of spots per area, but rather as a quality rating.

This may be because S. homoeocarpa is not considered an agronomic problem on P. pratensis fairways (Toshikazu 1997). In recent years however, greater incidence of dollar spot on P. pratensis has been reported, such as the first evidence of dollar spot in P. pratensis in Saskatchewan (Smith et al., 2001) and the widespread outbreak in USA (Studzinska et al. in 2006) in recent years. This trend might be mediated by increasing temperature or possibly decreased fertility in recent years.

In addition to the levels of dollar spot incidence, the disease development patterns on untreated plots of the two host species in the 2005-2006 trials were also different. Dollar spots patches were first found on A. stolonifera plots at the end of May and quickly reached a peak on 22 June 2006 with more than 20 spots/m2 (Figure 2.17). The treated plots had at least twice as many spots as untreated plots. Poa pratensis plots had the first dollar spots patches at the end

62 of July with the highest value on 25 August 2006 at 13 spots/m2 (Figure 2.18),

and the treated plots had only 40% more spots than untreated plots. The reason for the delayed start and lesser enhancement effect on P. pratensis was not directly linked to irrigation or meteorological factors, such as temperature or precipitation, since both trials were located in the same area and were irrigated similarly. Nevertheless, local microclimatic conditions in each plot canopy, such as humidity and temperature, could have affected the disease development.

There were two factors that could have made the P. pratensis plots drier.

First, A. stolonifera plots were totally flat, while P. pratensis plots were located on a slight slope that might have allowed water to flow off and hence change the amount of water available. More importantly was the tree shade that covered the

A. stolonifera plots for a large part of the day, while the P. pratensis plots were not shaded. The two factors, plant density and thatch, could affect the optimal conditions of humidity and temperature for S. homoeocarpa and therefore could have delayed disease development in P. pratensis plots.

Nitrogen fertilization could also cause differences between plots in dollar spot incidences, since dollar spot is usually associated with areas deficient in this element (Liu et al., 1995). However, fertilization with 255 kg N/ha applied to the plots in 2006 should produce vigorous growth. Schlossberg and Schmidt (2007) found that A. stolonifera mowed daily at 3.1 mm produced vigorous growth with annual nitrogen rates over 244 kg/ha. This suggests that a nitrogen deficiency was not responsible for the increased number of dollar spots on the native sand fairway.

63 Another difference between P. pratensis and A. stolonifera dollar spot

development was the azoxystrobin treatments. Dollar spot incidence on P.

pratensis plots did not show significant differences of the treated and untreated

plots on 10 of 12 evaluations. This result contrasts with the dollar spot levels

observed over two years in A. stolonifera plots, where there were significant

differences between dollar spot incidence during most of the growing season.

The effect of azoxystrobin on the resurgence or enhancement of dollar spot

incidence has been reported for putting greens containing Agrostis spp. with

some Poa annua (annual bluegrass) (Demoeden 2000; Hsiang and Cook, 2006),

but has not been previously noted on P. pratensis.

Dollar spot disease is not controlled or has been enhanced in A. stolonifera treated with commercial fungicides containing azoxystrobin. This is based on casual observation in the golf industry as well as specific experiments carried out to test the effect (Cook and Hsiang, 2002 and 2003). Application of azoxystrobin did not reduced the number of dollar spot patches in the summer when compared with untreated areas (Demoeden et al., 2000). In other trials, the

number of dollar spot patches were twice as high as the untreated areas (Cook and Hsiang, 2003; Demoeden, 2000; Gleason et al., 1996; Grogan and Scott,

1997; Hsiang and Cook, 2006). In 1996, Schuman et al., reported dollar spot enhancement on A. stolonifera 10 times higher than that two weeks before a single application of azoxystrobin. In this study, all the A. stolonifera plots treated with azoxystrobin had significantly higher dollar spot incidence than the untreated plots six months after a single azoxystrobin application This is the first study in

64 which dollar spot enhancement caused by azoxystrobin was observed after more than one season. Furthermore, although previous studies have reported this effect for putting greens containing Agrostis spp. with some Poa annua (annual bluegrass) (Dernoeden 2000; Hsiang and Cook, 2006), this is the first study to document or even mention the effect of azoxystrobin on the resurgence or enhancement of dollar spot incidence on P. pratensis.

The enhancement of dollar spot caused by azoxystrobin has not been commonly observed by turfgrass managers on golf courses or sport fields. The reason for this may be because other fungicides are commonly used to control dollar spot, and in amenity turf areas that suffer from dollar spot, it would be rare to have no treatments for dollar spot control during the growing season as was done in this study. The cause of the dollar spot enhancement is not well understood, but there are two major hypotheses for enhancement and resurgence in general. First, there may be a direct longer lasting effect of the fungicides on the plants that makes them weaker or stressed facilitating infection.

Secondly, the microbial population of soil or leaves may mediate the enhancement effect, since longer lasting alteration of microbial communities or changes in biodiversity may create different conditions for pathogen growth and infection (Dernoeden, 2000).

In support of the physiological explanation for enhanced disease, azoxystrobin is known to alter plant growth and delay senescence periods of leaves (Grossmann and Retzaff, 1997). These effects have not been reported on turfgrass, but they have been found for other graminaceous plants such as wheat

65 (Wu and Tiedemann, 2001) and barley (Kleven et al., 2003). Alteration of plant hormonal and water-conserving metabolism is considered possible causes of the growth effects of some strobilurins (Vincelli, 2002). These could also alter conditions in which pathogens attack the leaves. The changes produced by azoxystrobin on the plant metabolism could also indirectly modify microbial populations that lead to the disease enhancement, although no study reporting this was found.

Another characteristic of azoxystrobin that may help explain disease enhancement is the way in which the chemical moves inside the plants. Qols normally exhibit translaminar movement (Gold et al., 1996), which is the ability of a chemical to penetrate and cross through the leaf tissue. For fungicides, agronomically significant movement involves the chemical passing through the sprayed surface to the unsprayed lamina at concentrations sufficient to inhibit pathogens on the unsprayed side. Azoxystrobin is also transported and moved in the vascular system acropetally from the point of absorption toward the tips of the leaves (Christians, 2007). The incorporation and movement through the plant of azoxystrobin might alter the metabolism in A. stolonifera allowing the observed disease enhancement.

Although cited as a possible source of disease resurgence and enhancement, some experimental conditions indicate that the plant effect may not be the main cause. First, it seems unlikely that plants treated more than six months previously with azoxystrobin still contain sufficient chemical to cause some effect. A study of azoxystrobin movement suggests that it is localized in

66 intercellular spaces of the leaves and does not enter the cytoplasm of grape plants (Wong and Wilcox, 2001). Thus, the direct effects on plant cellular metabolism may be limited. Since the photodegradation half-life of azoxystrobin in the soil is 11 days (Anonymous, 1997), there is no remaining pool to be drawn from the soil over time. Additionally, biotransformation also increases the rate of azoxystrobin dissipation over time.

Persistence studies of azoxystrobin in the Canadian central provinces indicated a wide range of persistence with a 50% dissipation time (DT50) of 14 to

62 days (Anonymous, 2000). This puts azoxystrobin in a non-persistent to moderately persistent category according to the Goring et al. (1975) classification scheme (Anonymous, 2000). In the same report, application of 5 kg of azoxystrobin per hectare on turfgrass had a DT50 of 135 days on the soil. This implies that the total dissipation (biotransformation) of azoxystrobin can take more than nine months, however the application the maximum annual application

(50 g a.i./100 m2) was four to eight times greater than that used in the current study.

Among other factors that argue against a direct plant effect as the cause of resurgence, is the speed at which turfgrass plants grow. At the time when enhancement is observed, the plant leaves that were treated had already been removed months earlier by continuous growth and mowing. The A. stolonifera shoot growth rate reported by Xu and Huang (2001) ranges from 0.5 to 3 mm/day depending on the temperature. Thus, when mowing at 5.5 mm daily, all treated plant foliage would have been replaced in less than a week (5.5 days assuming a

67 growing rate of 1 mm/day). Consequently, the long lasting effect observed, might be an indirect effect of azoxystrobin such as the alteration of microbial communities rather than a direct effect on plants.

Azoxystrobin is known to have strong inhibitory effects on a variety of organisms such as many of the true fungi (Ascomycota, Deuteromycota

Basidiomycota) and even the stramenopile Oomycota. A single application of this broad spectrum fungicide may have wide reaching and long term effects.

Microbial communities including bacteria and fungi as major groups, are delicate ecosystems that once altered may take months to recover the original biodiversity, richness and distribution, and might never return to their pre- treatment composition. In soils treated with pesticides, for example, the diversity of soil microbial communities was affected even after the last application was performed 135 days before (Fliessbach and Mader, 2004).

The dollar spot disease enhancement effect produced by azoxystrobin on A stolonifera, although clearly observed in both years, differed depending on year and plot location. In the 2005-2006 trial (native sand fairway), the levels of dollar spot were not as high as in the two experiments set up in 2006-2007 (native sand fairway and pathology green). However the ratio of dollar spot incidence between treated vs. untreated plots differed from 2:1 on the native sand fairway plots

(2006-2007), 3:2 native sand fairway plots (2005-2006) to 5:1 on the pathology green plots (2006-2007) (Figures 2.17, 2.19 and 2.20). The most probable causes of these differences are the rate of azoxystrobin applied, and variation in weather and cultural practices. In 2005-2006, the fungicide rate was half (6 g

68 a.i./100 m2) of the rate used in 2006-2007 plots. The doubled azoxystrobin rate might have allowed for the increased disease enhancement observed in 2007, but since the weather conditions were also very different between 2006 and 2007, a definitive conclusion cannot be drawn from the data. Experimental work with different rates within the same experiment are required to resolve this.

Aside from root zone differences, mowing frequency and height differentiated the two A. stolonifera trials of 2006-2007. The pathology green was mowed daily while the native sand fairway was mowed every other day three days a week. Additionally, the height was lower in the pathology green (5.5 mm) vs. 10 mm of the native sand fairway. The mowing rate and height establishes a balance between plant stress and infected plant removal. Intense mowing can lead to weaker overstressed plants with a poor defense response and therefore becoming more disease prone (Salaiz et al., 1995). Nevertheless, the same intense mowing will remove a greater amount of infected tissue possibly reducing disease spread or symptom expansion. The less frequently mowed and presumably less stressed plants (native sand fairway) seemed to be more easily infected, since they showed symptoms 20 days before than the more intensely mowed plots (pathology green). Additionally, within one month, the native sand fairway plots showed the maximum incidence of dollar spot of 80 spots/m2, while the more frequently mowed plots required 1.5 months to reach the peak of infection. Therefore, the mowing effect in this case seemed to delay the onset of macroscopically observable disease symptoms.

69 These results are seemingly contradictory to most dollar spot cultural

treatment guidelines in which maximum height possible is recommended.

However, Ellram et al. (2007) found that dollar spot incidence was lower when

dew was removed daily by mowing rather than on alternate days. This suggests

that in this study the presence of dew might be a cause of the early dollar spot

outbreak in the native sand plots and not the mowing height. Further

investigation of mowing height could shed light on the particular phenomenon of

greater height and more dollar spot symptoms.

In summary, dollar spot disease incidence can be affected by many factors

such as turfgrass species, climatic conditions and cultural practices. Previous

studies and anecdotal reports have found an enhancement or resurgence of dollar spot within weeks or a few months after application. This is the first study to show a longer term effect where, a single azoxystrobin application in fall produced a 2-fold to 5-fold enhancement of dollar spot 4 to 6 months later in summer. In addition, this study also found a slight enhancement on P. pratensis, which has not been previously reported. Although the cause of this disease enhancement phenomenon is not known, the physical and chemical characteristics of azoxystrobin along with turfgrass cultural practices and the finding of the longer term effect from this study, implies that the phenomenon stems from non-target effects on microbial populations of leaves or soil rather than a direct long term effect on the plants. Studies on phyllosphere communities may shed light on this phenomenon and point out the directions for future research.

70 Upper green

Poa pratensis N fringe

a

Native sand fairway flf™ Pathology green

8 m. e

Figure 2.1. Plot layout for trials at the Guelph Turfgrass Institute, Guelph, Ontario. All trials were set up on Agrostis stolonifera, except for the fringe of the upper green which has Poa pratensis. (a) Native sand fairway multi-season trial of 2005-2006, (b) P. pratensis multi-season trial of 2005-2006, (c) pathology green multi-season trial of 2006-2007, (d) native sand fairway plots multi-season trial of 2006-2007, and (e) summer pathology green plots of 2006

71 2 m 1 2 3 4 5 6 7 8

1 m Figure 2.2. Plot layout for all trials mentioned in Figure 2.3. Even numbered plots were treated with Heritage®. Odd numbered plots were untreated.

72 01-Oct-05 30-NOV-05 29-Jan-06 30-Mar-06 29-May-OB 28-Jul-06 26-Sep-06 25-Nov-OB

llriil

u M in„i>,l JUMi.iillli., ill Lit 111 01-Oct-05 30-Nov-05 29-Jan-06 30-Mar-OB 29-May-06 28-Jul-06 26-Sep-06 25-Nov-06 Figure 2.3. Temperature and precipitation from 1 October 2005 to 1 December 2006: (a) maximum, minimum and weekly average temperatures recorded at the Waterloo Wellington weather station (Climate ID 6149388), (b) visual estimation snow depth (cm) at the Guelph Turfgrass Institute, Ontario, and (c) total precipitation (mm) recorded at the Waterloo University weatherstation, Ontario.

73 01-Oct-OB 30-Nov-OB 29-Jan-07 30-Mar-07 29-May-07 2B-Jul-Q7 26-Sep-07

A ilMiiMiilit .. • . c

• i ll ij uLLt. Lit ll i i il i.iiii.ti . i 1 I 01-Oct-OB 30-Nov-OB 29-Jan-07 30-Mar-07 29-May-07 28-Jul-07 26-Sep-07 Figure 2.4. Temperature and precipitation from 1 October 2006 to 1 October 2007: (a) Maximum, minimum and weekly average temperatures recorded at the Waterloo Wellington weather station (Climate ID 6149388), (b) snow depth (cm) measured with a sonic range sensor and (c) total precipitation (mm) recorded at the Waterloo University weatherstation, Ontario.

74 Figure 2.5. Native sand fairway trial of 2005-2006 on 28 November 2005.

75 •Us

V

Figure 2.6. Native sand fairway trial of 2005-2006 on 30 March 2006 after snow cover melted on 23 March 2006.

76 13-Mar 30-Mar 04-May 01-Jun 22-Jun 13-Jul 04-Aug 25-Aug 15-Sep 05-Oct 20-Oct 2006 Figure 2.7. Plant injury of the Agrostis stolonifera native sand fairway, multi-season trial of 2005-2006. Plant injury was evaluated weekly to monthly from 13 March 2006 until 20 October 2006, and was assessed as percentage area with yellow to straw color grass (produced by biotic and abiotic sources), based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. Figure 2.8. Poa pratensis multi-season trial of 2005-2006 on 28 November 2005.

78 3:iSS»=--< **«*JL;

Figure 2.9. Poa pratensis multi-season trial of 2005-2006 on 30 March 2006 after snow coved melted on 23 March 2006.

79 13-Mar 30-Mar 20-Apr 25-May 15-Jun 06-Jul 27-Jul 17-Aug 09-Sep 28-Sep 20-Oct 2006 Figure 2.10. Plant injury of the Poa pratensis multi-season trial of 2005-2006. Plant injury was evaluated weekly to monthly from 13 March 2006 until 20 October 2006, and was assessed as percentage area with yellow to straw color grass (produced by biotic and abiotic sources), based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. W&&&?*.

Figure 2.11. Native sand fairway trial of 2006-2007 on 23 November 2006

81 • • 'Hi*.

Figure 2.12 Two plots of the native sand fairway trial of 2006-2007 after snow melted on 12 March 2007. The treated plot received 12 g/100 m2 azoxystrobin on 25 November 2006, and showed no evidence of snow molds after snow melt while the untreated plot had some snow mold patches.

82 60

Untreated

CO

03 40 ^§

I 20

0 23-Nov 14-Dec 22-Mar 12-Apr 2&-Apr 10-May 24-May 07-Jun 21-Jun 05-Jul 19-Jul 03-Aug 2006 2007 Figure 2.13. Plant injury of the Agrostis stolonifera native sand fairway multi-season trial of 2006-20072006-2007. Plant injury was evaluated weekly to monthly from 23 November 2006 until 8 August 2007 and was assessed as percentage area with yellow to straw color grass (produced by biotic and abiotic sources), based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. hK**i,-i.-%-••'•' Js*L- -'**-•: *-*

Figure 2.14. Pathology green multi-season triar^20oS5oo^r^^ovember

84 Treated Untreated

Figure 2.15. Two plots of the pathology green trial of 2006-2007 after snow melted on 12 March 2007. The treated plot received 12 g/100 m2 azoxystrobin on 25 November 2006, and showed no evidence of snow molds after snow melt, while the untreated plot had some snow mold patches.

85 60"

- -•- - Untreated -Treated

23-Nov 12-Dec 22-Mar 12-Apr 26-Apr 10-May 24-May 07-Jun 21-Jun 05-Jul 19-Jul 26-Jul 2006 2007 Figure 2.16. Plant injury of the Agrostis stolonifera pathology green multi-season trial of 2006-2007. Plant injury was evaluated weekly to monthly from 23 November 2006 until 8 August 2007 and was assessed as percentage area with yellow to straw color grass (produced by biotic and abiotic sources), based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error. 45

25-May 08-Jun 22-Jun 06-Jul 20-Jul 04-Aug 17-Aug 2006 Figure 2.17. Dollar spot incidence of the Agrostis stolonifera native sand fairway multi-season trial of 2005-2006. Dollar spot incidence was evaluated weekly to monthly from 13 March 2006 until 20 October 2006, and was assessed as number of patches per m2, based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 mi . The bars represent standard error.

87 10-Jul 20-Jul 27-Jul 04-Aug 10-Aug 17-Aug 25-Aug 2006 Figure 2.18. Dollar spot incidence of the Poa pratensis multi-season trial of 2005-2006. Dollar spot incidence was evaluated weekly to monthly from 13 March 2006 until 20 October 2006, and was assessed as number of patches per m2, based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 28 November 2005 at 6 g azoxystrobin per 100 m2. The bars represent standard error.

88 24-May 07-Jun 21-Jun 05-Jul 19-Jul 06-Aug 2007 Figure 2.19. Dollar spot incidence of the Agrostis stolonifera native sand fairway multi-season trial of 2006-2007. Dollar spot incidence was evaluated weekly to monthly from 23 November 2006 until 8 August 2007, and was assessed as number of patches per m2, based on four replicate 1 m x 2 m plots per treatment. Heritage® 50WDG was applied on 25 November 2006 at 12 g azoxystrobin per 100 m? The bars represent standard error.

89 Figure 2.20. Two plots of the pathology green trial of 2006-2007 on 24 July 2007. The treated plot received 12 g azoxystrobin per 100 m2 on 25 November 2006, and showed higher number of dollar spot patches than the untreated plot.

90 ••- -Untreated ••—Treated

•5- «*

01-Jun 07-Jun 21-Jun 05-Jul 12-Jul 19-Jul 26-Jul 03-Aug 2007 Figure 2.21. Dollar spot incidence of the Agrostis stolonifera pathology green multi-season trial of 2006-2007. Dollar spot incidence was evaluated weekly to monthly from 23 November 2006 until 8 August 2007, and was assessed as number of patches per m2, based on four replicate 1 m x 2 m plots per treatment. Heritage® MAXX was applied on 25 November 2006 g azoxystrobin per 100 m2. The bars represent standard error.

91 . ——^——————. Figure 2.22. Agrostis stolonifera summer plots 2006, located on the pathology green at the Guelph Turfgrass Institute, Ontario. Plots were the only area of the green with highly injury caused by biotic and abiotic factors. Because of these confounding factors the data were not used in analyses.

92 CHAPTER 3. EFFECT OF AZOXYSTROBIN ON THE PHYLLOSPHERE MICROBIAL COMMUNITIES OF TURFGRASSES

3.1 Introduction The surfaces of terrestrial plants (leaves, stems, flowers and fruits) carry a

complex and dynamic microbial community which is comprised of bacteria,

protozoa, lichens and fungi (Atlas and Bartha, 1987). This niche is referred to as

the phyllosphere and the inhabitants are termed epiphytes and endophytes

(Lindow and Leveau, 2002). Epiphytes are those microorganisms which reside

on plant surfaces, while endophytes are those which live inside the plants

(Lindow and Brandl, 2003). The study of the phyllosphere has contributed to

many fields of knowledge. For instance, studies on phyllosphere microorganisms from vegetables have improved food safety and have provided to a better

understanding of food preservation (Morris and Kinkel, 2002). Studies of

microbial phyllosphere communities of plants inoculated with non-pathogenic and

pathogenic bacteria have shown a competitive exclusion of the pathogenic

bacteria contributing to the field of biological control of pests (Lindow and Brandl,

2003).

Two general types of microorganisms are usually found on leaves. Some are transient inhabitants which come from other niches and establish temporarily on leaves while the other group is persistent and commonly resides on leaves, for long periods (Juniper, 1991). In both cases, microorganisms have special characteristics that allowed them to survive in extreme environments, including resistance to fluctuating temperatures, low humidity, or high UV radiation and

93 scarcity of nutrients (Yang et al., 2001). Many epiphytic bacteria have pigments that protect them from UV radiation, and some produce extracellular polysaccharides that might assist surface attachment and prevent desiccation

(Beattie and Lindow, 1995).

Studies on microbial communities of turfgrass phyllospheres are not common since the bulk of research traditionally focused on food plants and not ornamental plants. Turfgrass phyllosphere microorganisms comprise a relatively large number of epiphytic or endophytic yeasts, bacteria and filamentous fungi.

Two major groups of yeasts (white and pink or red) are commonly associated with turfgrass foliage (Buck and Burpee, 2002). The pink/red yeast group contains genera, such as Sporobolomyces or Rhodotorula, while white yeasts are usually Cryptococcus or Pseudozyma (Allen et al., 2004b).

There are few studies that have assessed epiphytic filamentous fungi species in turfgrass, but several studies have been done on endophytic fungi, such as Epichloe species, which are grass symbionts (Schardl, 1996). Similarly, research on turfgrass phyllosphere bacteria is very limited. In one study, yellow pigmented bacterial colonies were the major morphotype isolated from grass leaves (Behrendt et al., 1999). This yellow pigmented group is composed of

Gram-negative, aerobic, nonspore-forming, rod-shaped bacteria of genera, such as Pantoea, Xanthomonas, Sphingomonas and Pseudomonas, which are commonly found in other plant phyllospheres (Rudolph et al., 1990).

The traditional approach to study phyllosphere microorganisms is to isolate them from leaves, and then count their numbers and identify the species. For

94 epiphytes, many scientists have used leaf washing to dislodge microorganisms into a solution and then grow them on artificial media (Allen et al., 2004b; Osono et al., 2004). As for endophytes, the most common technique also uses artificial media. In this case, plant surfaces are washed and sterilized to remove and kill the epiphytes, and subsequently leaves are either ground or placed directly on artificial media (Santamaria and Bayman, 2005; Yadav et al., 2005). These techniques have advantages and disadvantages. First, they are easy, inexpensive and do not require special training. Thus, easily culturable phyllosphere microorganisms can be obtained, counted and identified. However, only viable microbes are recovered using these techniques, and those that cannot grow in artificial conditions are totally ignored (Stanley and Konopka,

1985).

The phyllosphere is a complex and delicate niche that can be altered by small perturbations, including climatic effects and human management activities.

Fungicides (Buck and Burpee, 2002), fertilizers (Behrendt et al., 2004) and pollutants (Dowding, 1986) are among some human-related factors that can drastically modify foliar ecosystems. Pesticides have been shown to change the microbial composition of many plant species (Dik, 1991). On cereal leaves, fungicides reduced populations of saprophytic yeasts, bacteria and filamentous fungi for several weeks (Fokkema and De Nooij, 1981), sometimes followed by an increase in yeast populations (Fokkema et al., 1987). For the turfgrass phyllosphere, fungicides have been reported to reduce microbial populations

(Bertelsen et al., 2001).

95 Knowledge of the microbial population dynamics of highly managed

turfgrass phyllosphere environments is critical to our understanding of more

complex problems, such as biological control of plant pathogens or the ecological

effects of anthropogenic perturbations. To address this latter issue, seasonal

fluctuations of the phyllosphere epiphyte communities were investigated in two

intensively managed turfgrass species (Agrostis stolonifera and Poa pratensis).

In this study, the effects of a commonly used fungicide, azoxystrobin, on

microbial populations were studied across several seasons, identifying epiphytic

and endophytic bacteria, yeast and filamentous fungi from plants treated with this

fungicide. The bacteria, yeast and filamentous fungi isolated from turfgrass

phyllospheres were subdivided into color morphotypes and enumerated on

diluted plates. The were further examined with molecular techniques as reported

in a subsequent chapter.

3.2 Materials and methods

3.2.1 Study site Turfgrass plots established at the Guelph Turfgrass Institute (GTI) on

November 2005 and 2006 were used as the source of leaves (Figure 2.1 a, b,

and d). These plots are referred to as the native sand fairway trial (2005-2006), P. pratensis fringe trial (2005-2006) and the pathology green trial (2006-2007). Each

trial consisted of eight plots (1 m x 2 m), where four plots (even numbered) were

treated with azoxystrobin, while the other four plots (odd numbered) had an

equivalent amount of water applied (untreated plots). In the 2005-2006 trials, two

turfgrass species (A. stolonifera on the native sand fairway and P. pratensis at

96 the fringe of the upper green) were treated with Heritage® 50WDG at 6 g azoxystrobin per 100 m2 (Figure 2.1 a and b). In the 2006-2007 trial, Heritage®

MAXX® was applied at 12 g azoxystrobin per 100 m2 to the pathology green

(Figure 2.1 d).

3.2.2 Sample collection In the 2005-2006 trials (A stolonifera native sand fairway and Poa pratensis fringe), samples were collected every two months in the spring and fall, and monthly in the summer. In the 2006-2007 trial (A stolonifera pathology green), samples were collected before application of azoxystrobin in November 2006 and every two months from March to July 2007. Plants were cut from 12 different areas of each A. stolonifera plot, and from nine areas of each P. pratensis plot using gloves and scissors surface sterilized with ethanol 70% (v/v) when moving between treated and untreated plots. Obviously diseased leaves or patches in winter (snow mold) or summer (dollar spot) were avoided during sample collection. Harvested leaves from each plot were transferred to a separate sterile

50 ml polypropylene centrifuge tubes and placed in the shade until transported to the lab for storage at 4C for less than three days.

In the lab, samples were maintained at 4C until leaves were segregated into either green or yellow categories based on senescence. Selection was done inside a laminar flow workstation using aseptic techniques. The A. stolonifera green leaves had no noticeable wounds nor any chlorotic areas. In many cases, green leaves of P. pratensis had to be chosen with wounds caused by mower blades, because intact leaves with tips were not available. The "yellow" leaves

97 were yellow or light brown overall or had less than 10% green area across the leaf blade, and were usually not intact. Selected leaves were placed into tubes

(1.5 ml) and stored at 4C for less than two days.

3.2.3 Microbial screening

3.2.3.1 Leaf washings To obtain microorganisms from the surface of leaves, sets of green and yellow leaves were washed using saline solution with agitation and sonication as in other leaf microbial analyses (Yang et al., 2001; Buck and Burpee 2002). First,

1 ml physiological saline solution (0.85% NaCI (w/v)) was added to each 1.5 ml tube and vortexed at maximum speed on a Vortex Genie 2 (Fisher Scientific,

Mississauga, Ontario, Canada) for 30-40 seconds. Tubes were then sonicated with a Crest 175TA sonicator (Crest Ultrasonics, Trenton, New Jersey, USA) at

45 watts for 10 min.

3.2.3.2 Media preparation Unamended Difco® potato dextrose agar (PDA, BD, Franklin Lakes, New

Jersey, USA) was prepared using the manufacturer's instructions: 39 g of PDA powder were mixed with 1 I deonized water and autoclaved for 15 min at 121C.

To obtain bacterial cultures free of most fungi, PDA amended with the fungicide benomyl was prepared by adding benomyl stock solution (200 ug/ml in ethanol

(9.5% (v/v)) made from Tersan 1991 50WP (DuPont, Wilmington, Delaware, USA containing 50% benomyl) to autoclaved molten PDA (50-60C) and mixing thoroughly with a magnetic stir bar to obtain a final 2 ug/ml concentration of benomyl. To obtain fungal cultures free of most bacteria, PDA amended with

98 antibiotics was made by adding streptomycin (100 |jg/ml final concentration) and tetracycline (5 |jg/ml final concentration) directly to autoclaved PDA at ~50C and stirred while dispensing. With a varistaltic dispenser pump (Kate, Manostat, USA), medium was dispensed into 9-cm-diameter plastic Petri plates (15 ml) and glass vials (5 ml) for stock slants.

3.2.3.3 Microbial pre-screening A pre-screening protocol was used on samples collected from August 2006 until July 2007. On samples collected from March 2006 to July 2006, only the full microbial screening was used without pre-screening. Pre-screenings were first done to obtain the general number of bacteria, filamentous fungi and yeast, in treated and control plots, prior to larger scale dilution platings (full microbial screening) for population enumeration. To pre-screen, two to three leaves obtained from each of four replicate plots were placed inside a single 1.5 ml tube for a total of 10 to 12 leaves. This procedure was done separately for treated and untreated green and yellow leaves. Leaf washings were performed as described previously and after sonication, the tube contents were serially diluted.

Pre-screening serial dilutions were used to get a quick preliminary assessment of the total number of bacteria, filamentous fungi and yeasts, and therefore, a wide range of dilutions were used. The original leaf washing suspension was diluted to 1:2, 1:10, 1:50, 1:100, 1:1000, 1:10000, 1:100000 and

1:1000000 using physiological saline as a diluent. When preparing the dilutions, tiny leaf pieces were avoided by sampling from the center of the tube. However, yellow leaves often formed a suspension of very fine pieces that were difficult to

99 avoid. The first four dilutions were plated in triplicate on PDA amended with antibiotics, while the other four dilutions were plated in triplicate on PDA amended with benomyl at 2 ug/ml. For each plating, 50 ul of the vortexed suspensions were placed in a 9-cm-diameter Petri plate on the surface of 15 ml of agar, and spread over the surface with a flamed glass rod.

Pre-screening plates were incubated at 25C, and checked daily until the colonies were visible (two to three days). Then, the total number of colony forming units (CFU) of bacteria, yeast and filamentous fungi were counted using a Quebec dark field 3330 Colony Counter (American Optical Manufacturing

Company, Southbridge, Massachusetts, USA). The protocol for plate counts was to select those dilutions that contained between 20 to 200 colonies per plate for each microbial group. If more than one dilution had colonies in that range, plates at both dilutions were counted and the average CFU/ml calculated.

3.2.3.4 Microbial screening Based on counts obtained from pre-screening, a narrower range of serial dilutions was made for a larger number of samples. These experiments involved colony counts for the major morphotypes obtained on dilution plates. Major morphotypes were defined by color and are more fully described later. The samples were ten A. stolonifera or five P. pratensis leaves from each plot in duplicate. The selection of leaves included yellow and green leaves that were identified as described previously. For the microbial screening, duplicates groups

(ten A. stolonifera or five P. pratensis) of each green and yellow leaves were separated from each plot and transferred separately to 1.5 ml tubes. Thus, from

100 each trial of eight plots (four treated and four untreated), 32 tubes containing 10 leaves would be generated. The samples were stored for up to five days after sample collection before plating. Next, for each sample, four serial dilutions were made based on the number of bacteria, yeasts and filamentous fungi obtained in the pre-screening experiment. The two lowest dilutions were plated in triplicate on PDA amended with antibiotics to obtain counts for the different morphotypes of filamentous fungi and yeasts. The remaining two higher dilutions were plated in triplicate on PDA amended with benomyl to assess bacterial morphotype since bacteria were always more abundant.

Dilution plates were incubated at 10-12C for seven to nine days to allow colonies to grow. Plates were checked daily, looking for anomalies and colony maturity, since some colonies changed in color with maturation. Once colonies had developed their final morphotypes (usually five to six days for bacteria, and seven to eight days for filamentous fungi and yeasts), the number of major morphotypes was recorded as described in microbial prescreening analyses.

3.2.4 Leaf weight After the dilutions had been plated, the tubes containing original leaf washings were transferred to pre-weighed, labeled aluminum dishes individually and placed into an oven (Isotemp 500 series, Fisher Scientific, Mississauga,

Ontario, Canada) at 80C. At 24-hour intervals, leaf weight was checked using an analytical scale (Sartorius TE64, Data Weighing Systems, Chicago Illinois, USA) until the leaves reach a constant weight (usually between 24 and 48 hours). The

101 surface microbial populations were expressed as CFU per gram of dry leaf weight.

3.2.5 Colony counting Based on color and texture, colony morphotype groups were determined for each of the organism groupings: bacteria, yeasts and filamentous fungi. Each morphotype was not necessarily present in all samples, but together they constituted a large fraction of the total population in most collections. Colonies were counted using a black field colony counter, or using white or black paper in the background.

Yeast colonies were separated into four major morphological groups: white, pink, salmon and orange (Figure 3.1). Pink and salmon morpholotypes were easily identified after four days, but orange and white colonies had variations in tone and texture that required more incubation time to develop (up to nine days).

Usually, colony counts for orange and white types were done again after two more days to confirm the results. The total number of yeasts consisted of the sum of the four morphological morphotypes. The white yeast type showed subtle differences among the colonies since some were slimier and shinier. Additionally after long incubation (over two weeks) at 25C, some white colonies changed to different tones of beige to brown.

Three bacterial color morphotypes, yellow, white and orange (Figure 3.2) were identifiable on plates, but it was not possible to classify some colonies because they matured too slowly and hence were grouped with the white morphotype since most colonies started out as white. Differences between white

102 and yellow bacterial colonies were occasionally hard to establish. Therefore, the first colonies were noted on the underside of the plates with different color markers, and two days later colonies were checked for changes in morphotype.

After they were confirmed, colonies were enumerated in their particular morphotypical group. The total number of bacterial colonies was the sum of the four morphotypes counted. The morphotypes had submorphotypes that were revealed when subcultured and allowed to grow at 25C for a longer time. The submorphotypes differed in texture, such as a slimier or drier appearance, but also differed in characters such as paleness or transparency.

Total numbers of filamentous fungal CFU were counted independently of the morphotype since not all colonies could be classified into morphotype.

Recognition of filamentous fungi was based on observation of web-like or branching mycelia on the agar surface. Filamentous fungal colonies were usually flatter, dryer and less shiny than yeast colonies. In some cases, determination of the type of the colony was confirmed by microscopic observation of hyphae or spores.

After three to four days incubation, all filamentous fungal colonies were white or even transparent without visual differences. By day seven to day eight, most filamentous fungal colonies were sufficiently mature, showing different colors or textures or even sporulation (Figure 3.3). However, by that time it was not possible to count colonies on the plates since they became overgrown with mycelia and slimy yeast colonies. Nevertheless, two fungal colony morphotypes matured quickly enough to distinguish them by day four, which allowed colony

103 enumeration of these plus the total filamentous fungal colonies. The two morphotypes were called black and red fungi, based on their main colors. The black fungal morphotype comprised all colonies with plate undersides in shades of dark brown, dark grey or black. The upper surface of these colonies generally had different tones of greens and a variety of textural features. Black colonies were not separated into submorphotypes, because the colors were not easy to discriminate, and occasionally the tones changed over time. The other distinguishable morphotype, red, was more homogenous. All red colonies had the same characteristics of red to burgundy, plate undersides with fluffy, white mycelia on the upper side.

The plates counts for individual morphotypes and species are referred to as population counts. Whereas the total counts for each microbial group (bacteria, yeast or filamentous fungi) are referred as community counts. The total across all groups is referred to as the microbial community.

3.2.6 Statistic analyses To test whether two major assumptions of ANOVA (normality and homoscedascity) were violated, two statistical procedures were used. To assess normality of distribution, the Shapiro-Wilk test was used as implemented in SAS

(SAS Institute, Cary, north Carolina, USA) PROC Univariate Normal at p=0.05.

To assess homoscedascity (homogeneity of variance), the Bartlett's test for equality of variance was used as implemented in SAS PROC GLM with the hovtest=bartlett option selected in Means (examples of these SAS statements can be found in Appendix C.

104 For normality of distribution, the non-transformed data values showed significant (p=0.05) departure from normal distribution (Appendices D). Similarly, non-transformed values did not show homogeneity of variance at p=0.05

(Appendices E). Hence, ANOVA tests were conducted using log transformed values. However, means and graphs are shown as non-transformed values.

The log transformed data were subjected to a one-way ANOVA test using analyzed using SAS PROC GLM. If a significant effect was observed in the

ANOVA, the mean values were compared using Fisher's LSD (Least Significant

Difference) at p=0.05 and the standard error was obtained with PROC MEANS

(Appendix F).

3.3 Results The microbial populations found in the treated and untreated samples differed according to the type of leaf, species of grass and microbial grouping.

Furthermore, the abundance of the two leaf types, green or yellow, differed by season. After snow melt, there were many times more yellow leaves than green leaves on both grass species. During the summer, yellow leaves of A. stolonifera were much less common than green leaves, and in some cases for the native sand fairway trial, insufficient numbers of yellow leaves could be gathered for microbial isolation.

Microbial communities of both turfgrass species (Agrostis stolonifera and

Poa pratensis) always were greater on yellow leaves than green leaves across all sampling dates. This was found in the three organism groupings and morphotype subgroupings for treated and untreated samples on the A.

105 stolonifera native sand fairway (Table 3.1), the P. pratensis fringe (Table 3.2)

and the A. stolonifera pathology green (Table 3.3). Green or yellow leaves had

bacteria as the most populous microbial group independent of the turfgrass

species or treatment. This was followed in most cases by yeast. The number of

filamentous fungi was commonly less than yeasts on both plant species and type

of leaves (Tables 3.1, 3.2 and 3.3).

The two distinguishable filamentous fungal morphotypes, black and red,

contributed only up to 15.5% (black 15% and red less than 0.5%) of total fungi of

green leaves, and up to 9% on yellow leaves (black 8.9% and red less than 0.5%)

(Tables 3.1, 3.2 and 3.3). The remainder of the filamentous fungal colonies

(~90%) was not grouped into morphotypes because they nearly all started as

white colonies which developed different characteristics as they matured, by

which time it was not possible to accurately count individual colonies.

Yeast colonies were grouped into one of four morphotypes (orange, pink,

salmon and white). The major morphotype was white, which accounted for more than 70% of all colonies averaged across plant species and leaf type (Tables 3.1,

3.2 and 3.3). On A. stolonifera green leaves, the salmon morphotype was the

second most abundant comprising 15-18% of the total phyllosphere yeast community (Tables 3.1 and 3.3), while on green P. pratensis leaves; the second

most common morphotype was the orange morphotype (8.4%) (Table 3.2). The pink and orange morphotypes comprised less than 11% of the total phyllosphere yeast community of A. stolonifera green and yellow leaves (Tables 3.1 and 3.3).

On both types of leaves of P. pratensis, pink and salmon morphotypes were the

106 less common amounting to a combined percentage of only 6 to 8 of the total

yeast community (Table 3.2).

The distribution of bacterial morphotypes was variable among plant species,

leaf stage, and treatment. In general, the white and yellow bacterial morphotypes

were the most abundant (Tables 3.1, 3.2 and 3.3). On green leaves, bacterial

morphotypes had similar numbers when pooling data of all sampling periods, but

on yellow leaves, a single morphotype (either orange or white) usually comprised

50% or more of total bacterial community.

The effect of azoxystrobin on the microbial populations varied depending on

the host species, type of leaf and microorganism. In general, on the native sand

fairway azoxystrobin-treated leaves had significantly (p=0.05) fewer filamentous

fungi on green and yellow leaves when averaged across all samplings (Table

3.1). For P. pratensis, the microbial communities seemed only to be affected by

azoxystrobin treatment on green leaves where the total filamentous fungal

community was significantly reduced (p=0.05) by over 30% while yeast and

bacterial populations were increased in some cases by over 100% (Table 3.2).

On the green or yellow leaves of the pathology green azoxystrobin caused a

reduction on the numbers of some filamentous fungi, yeasts and most bacteria

(Table 3.3).

3.3.1 Green leaves The microbial communities of the phyllosphere of A. stolonifera were assessed over two growing seasons and P. pratensis during one season on green leaves (Figure 3.4) and on yellow leaves (Figure 3.5). After snow melted,

107 filamentous fungi counts on A. stolonifera and P. pratensis were low at approximately 0.3 x 106 CFU/g, and then increased in the summer for both years

(Figure 3.4 a, b and c). The filamentous fungal community decreased when the temperature dropped in the fall.

For the yeast community, the highest counts on A. stolonifera were observed after snowmelt in 2006 (Figure 3.4 d) and 2007 (Figure 3.4 f). The populations then declined into the fall (Figures 3.4 d and f). On P. pratensis, the yeast community showed small fluctuations through the seasons (Figure 3.4 e).

Bacterial populations on A. stolonifera showed different patterns for the trial in 2006 on the native sand fairway (Figure 3.4 g) than the trial in 2007 on the pathology green (Figure 3.4 i). On the native sand fairway, the counts started low after snowmelt, and peaked before the start of summer with a decline into the fall (Figure 3.4 g). On the pathology green, the highest bacterial counts were observed right before fungicide treatment in the fall with steady decreases in spring and summer, except for the untreated plot which showed a small increase at the start of summer (Figure 3.4 i). On the P. pratensis fringe, bacterial populations remained at a steady level throughout the sampling period (Figure

3.4 h).

Treatment with azoxystrobin in the fall did lead to some significant changes in microbial populations during the following ten months. Microbial populations on green leaves were affected to different extents depending on the host grass species and year, since each leaf ecosystem has its own particular characteristics. In addition, environmental conditions and cultural practices also

108 influenced the effect of a treatment on the phyllosphere microbial community. In

general, when there were significant differences (p=0.05) between populations

from treated samples or untreated samples, there was a reduction caused by

treatment (Figure 3.4). Averages across all sampling intervals showed that green

leaves from treated plots had fewer filamentous fungi, while bacterial and yeast

counts were varied (Tables 3.1, 3.2 and 3.3).

Although the filamentous fungal community was in some cases reduced by

the treatment, individual morphotypes were not always affected in the same way

(Table 3.4). Azoxystrobin produced an inhibitory effect on the filamentous fungal

community in March on the native sand fairway plots, but the populations

rebounded and were not significantly different from the untreated control summer

(Table 3.4).

In contrast on the native sand fairway plots, the yeast community (total) and populations (morphotypes) were generally not significantly affected by azoxystrobin (Table 3.4), except for the orange morphotype which was significantly decreased by treatment during March and May. (Table 3.4). The bacterial counts on green leaves from treated samples of A. stolonifera on the native sand fairway were found to be higher than from untreated samples only in

March (Table 3.4), and were generally not different throughout the rest of the sampling periods.

Poa pratensis microbial populations (Table 3.5) did not show as large differences between treated and untreated samples, but there were many more instances where morphotype populations were significantly increased (p=0.05)

109 by fungicide treatment than the untreated samples. For example all bacterial morphotypes were higher in samples collected in August from treated plots than untreated plots (Figure 3.4 h, Table 3.5). Yeast populations were also significantly enhanced by fungicide treatment for March and May collections compared to July, August or September collections (Table 3.5).

For the A. stolonifera trial on the pathology green, 2006-2007, there were significant reductions in populations for all microbial groups (Table 3.6) caused by fungicide treatment the previous fall. The reduction was particularly noticeable in at the end of May 2007 when almost all morphotypes, total yeasts and total bacteria were significantly reduced on treated plots compared to untreated plots.

Total fungal and bacterial communities were also significantly reduced on treated plots for the July collection which coincided with the heavy dollar spot infection.

3.3.2 Yellow leaves The seasonal pattern of filamentous fungi and yeasts of yellow A stolonifera leaves were similar to green leaves. The filamentous fungal community had low numbers in spring and increased in the summer for the pathology green and P. pratensis fringe trials (Figure 3.5 b and c). As for the native sand fairway, insufficient numbers of yellow leaves could be gathered for microbial analysis, and hence no data are presented for summer and fall (Figure

3.5 a).

Yeasts, had the highest number in spring declining in summer and fall on A. stolonifera (Figure 3.5 f) and P. pratensis (Figure 3.5 e). Bacteria on yellow leaves of A. stolonifera showed a gentle decline from spring through summer,

110 with peak numbers in fall (Figure 3.5 g and i). Bacteria on yellow leaves of P.

pratensis also showed lower populations in the summer than in the fall (Figure

3.5 h).

Azoxystrobin treatment also caused some significant effects (p=0.05) on

population levels of yellow leaves of A. stolonifera and P. pratensis. On the

native sand fairway, the data from spring 2006 show significantly decreased

counts for treated plots for all three microbial groups (Figure 3.5 a, d and g). On

the pathology green, there were a few instances of significant differences, but the

differences were small (Figure 3.5 c, f and i). On the P. pratensis plots, the May

samples showed reduced counts for bacteria (Figure 3.5 h) and yeasts (Figure

3.5 e) on treated plots, but otherwise the populations were similar.

For morphotypes from yellow leaves of the native sand fairway (Table 3.7),

some fungal and yeast populations were significantly reduced (p=0.05) on treated plots, while bacterial populations were not significantly affected in nearly

all cases. On P. pratensis, there were a few instances where morphotypes were slightly increased on treated plots and some where they were decreased, but the only large difference was for the orange morphotype sampled in May which showed a large reduction in treated plots (Table 3.8) by over 60% (Figure 3.5 e).

Populations from yellow leaves of the pathology green were in most cases reduced by fungicide treatment (Table 3.9). However, pink and salmon yeasts were more numerous in the summer on treated leaves than on untreated leaves

(Table 3.9). The population differences even when significant were not large

(Figure 3.5 c, f and i).

111 3.4 Discussion Treatment with azoxystrobin in late fall had an effect on microbial

populations four months later as snow melted and up to nine months later near

the end of the growing season. After snowmelt, the effect was most apparent on

yellow leaves with most populations from treated samples showing significant

reductions. By the summer, the most noticeable differences were apparent on

green leaves, again with populations reduced in treated samples. At the time of snowmelt, yellow leaves greatly outnumbered green leaves by at least a margin

of 10 to 1, and conversely during the summer, green leaves greatly outnumbered yellow leaves, particularly for A. stolonifera. This implies that overall microbial populations counts in the spring are best reflected by samples from yellow leaves, whereas microbial population counts in the summer are best reflected by samples from green leaves. Under this scenario, then the overall effect of azoxystrobin was a reduction in microbial populations observed right after snowfall and also at monthly or bimonthly sampling periods until the fall. These reduced microbial populations on green leaves during the summer were coincident with the increased dollar spot levels during the summer.

The total filamentous fungi community was either not altered or slightly reduced on treated samples, but the two morphotypes enumerated (black and red) were significantly reduced or increased at diverse sampling dates on treated leaves compared to untreated leaves. This indicates that only certain filamentous fungal species were affected by prior azoxystrobin treatment either directly or indirectly, possibly even by the shift in microbial populations of other groups such as yeasts.

112 The phyllosphere is a complex environment in which several

microorganisms interact with each other and with the plant. This environment

also undergoes many climatic shifts such as high UV radiation during the day

and no UV radiation at night, high temperatures by day and low temperatures by

night, high moisture during rainfall or irrigation and drought stress during dry

periods (Lindow and Leveau, 2002; Sundin 2002). In addition, turfgrass leaves

are affected by cultural practices and intensive management regimes. These

cultural practices include fertilization, pesticide, mowing and wear stress.

Previous studies have shown that fungicide applications can disturb

population levels of yeast and filamentous fungi in turfgrass over short term

periods of less than three months (Buck and Burpee, 2002; Harman et al., 2006).

In other plants, fungal communities have also been reported to be affected by

fungicide treatments (Dik, 1991; Gildemacher et al., 2004). Azoxystrobin in

particular has been shown to exert an inhibitory effect on fungal populations of wheat (Bertelsen et al., 2001) and turfgrass (Buck and Burpee, 2002) leaves. In all cases published, population shifts of the microbial phyllosphere community due to fungicides has been reported in periods less than three months after single or multiple applications.

In the present study, a series of epiphytic groups were significantly altered on areas where the fungicide azoxystrobin was applied once at recommended rates four to six months before. Since the growth rate of A. stolonifera is 0.5-3 mm/day (Xu and Huang, 2001), those leaves collected in May or July would not have been exposed directly to azoxystrobin. Additionally, the remaining

113 azoxystrobin in the plots should already have been degraded by physical,

chemical or biological means as the half life of azoxystrobin is 11 days

(Anonymous 1997). Moreover, the natural leaching and daily mowing with leaf

clippings removed should have reduced azoxystrobin from treated plots even

more. Thus, the inhibitory effect found on many microbial groupings was likely

caused indirectly by either affecting the plant physiology, or by extensively

modifying the phyllosphere community structure so that time was required to

return to an equilibrium found in untreated plots. Another possibility was that the source of phyllosphere microbial colonizers, the soil and thatch was also directly affected. This was not examined in the study because of the complexity of the soil and thatch, microbial populations. Liu et al. (1995) found that applications of

Ringer fertilizers, ammonium nitrate and sulfur coated urea increase the microbial populations on leaves, thatch and soil of A. stolonifera.

The altered microbial population of leaves caused by azoxystrobin found in this investigation may be then related to the ecological succession of microbial species in flux even over a long term. Other studies of longer term (more than 4 months) pesticide effect on phyllosphere microbial communities were not found in the literature. In environments, such as soil, microbial communities were significantly altered even 135 days after herbicide or fungicide application

(Fliessbach and Mader 2004). Additionally, the composition of epiphytic bacterial communities can be influenced by the establishment of antagonistic bacteria at certain times during plant development (Lindow and Brandl, 2003). The long term effect produced by pesticide on soil microflora and that the composition of

114 epiphytic community depends on the types of organisms that establish first

suggesting that the altered microbial community of soil or thatch may still be

different after several months and would eventually establish on the growing

leaves in the spring and summer differently than the leaves from untreated plots.

Alteration of the microbial community on leaves detected in this study may

not be related to a modification on the physiology of plants. Those possibly

altered plant tissues would have been removed from the treated plot by the

continuous mowing. Even the trace amounts of azoxystrobin that may be present

inside the overwintering roots may not produce a drastic physiological response since azoxystrobin is known to only move acropetally (toward leaf tips)

(Christians, 2007), which will be removed by the constant mowing. Nevertheless, the delaying of leaf senescence is a well known effect of azoxystrobin often associated with physiological alteration of plants (Cromey et al., 2004). This effect, however, has also been linked to the reduction of the saprophytic leaf community that reduces stress level of plant and extends the green leaf area retention (Bertelsen et al., 2001). This reduced foliar epiphytes as the indirect cause of delayed leaf senescence, suggesting that microbial alteration of foliar microorganism may have been a cause of this physiological effect produced by azoxystrobin.

The succession and dynamics of microbial community on the phyllosphere in not well understood. Immigration, emigration, growth and death regulate the microbial dynamics on leaves surfaces (Kinkel, 1997). Many fungi associated with turfgrass diseases remain in the thatch and infect healthy leaves when

115 conditions are favorable (Smiley et al., 2005). Perhaps the singular position of

plants in turfgrass environments, in which leaves are intermingled and in close

contact with the thatch increases the chances of interchange of microorganisms

from dead to healthy leaves.

The seasonal fluctuation of microbial community on green leaves from

untreated samples had similar patterns in both trials with A. stolonifera. Basically,

filamentous fungi started low in spring, increased in the summer, and returned to

low levels in fall. This has been found in other phyllosphere environments, such

as leaves of Lytsea polyantha (Rahman et al 2004). Yeasts results, on the other

hand, differed from those reported previously. Leaves of pasture plants were

found to harbor low numbers of some yeast genera (Sporobolomyces spp. and

Rhodotorula spp.) in winter and spring, and high in summer (di Menna, 1959). In

turfgrass, the highest number of yeasts was found in spring, and the numbers

decreased to their lowest levels at the end of summer. This result may be related to the wet period of snow cover when leaves are highly stressed and in intimate

contact with soil and thatch. The wet conditions and the increased amount of

nutrients leached from dying leaves may enhance the growth and colonization of yeasts during and after snow melt.

The phyllosphere community was statistically altered on plants treated with azoxystrobin. A reduction of some microbial groupings was found in spring before the enhancement of dollar spot. This shows that disease enhancement mediated by fungicide may be related to the modification of the microbial phyllosphere community as has been previously suggested (Couch, 1995). The

116 results also show that azoxystrobin has a long term effect over six months in the

turfgrass environment, which affects disease development and microbial

community. These effects have not been previously reported before for any

fungicide for periods greater than four months.

Further evidence of the microbial alteration of turfgrass trigged by

azoxystrobin still must be obtained, and future work could examine whether there

actually are any long-lasting residues of azoxystrobin or its active byproducts in

the soil, thatch or grass plants that somehow manage to persist through a winter

and most of growing season. However, since the enhancement effect was found

six months after a single azoxystrobin treatment, the present study gives strong

support for the role of the microbial community in causing the disease enhancement reported in turfgrass environments after azoxystrobin treatment.

117 Table 3.1. Colony forming units (CFU) per gram of leaves of microbial morphotypes obtained from the Agrostis stolonifera native sand fairway multi- season trial of 2005-2006 at the Guelph Turfgrass Institute, Guelph, Ontario. Means were obtained by pooling all collection data across seven sampling dates from 19 March to 28 September 2006. Leaf refers to the state of the foliage, and the organism categories (filamentous fungi, yeast and bacteria) are subdivided into morphotype groupings. N is the number of plates assessed and SE is the standard error. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. CFU/g (thousands) IJntreate d Treated Leaf Organism Group N Mean SE N Mean SE Diff Green Fungi Total 145 821 66 102 658 87 - Green Fungi Black 119 87 7 72 47 5 - Green Fungi Red 113 3 1 66 4 1 0 Green Yeast Total 150 1,807 145 408 2,225 210 0 Green Yeast Orange 150 100 16 102 67 9 0 Green Yeast Pink 146 183 31 102 185 46 0 Green Yeast Salmon 150 282 40 102 525 67 + Green Yeast White 150 1,235 111 102 1,448 161 0 Green Bacteria Total 140 25,500 1,473 288 31,033 2,281 0 Green Bacteria Orange 137 9,228 1,130 96 12,807 1,093 + Green Bacteria White 136 8,478 774 96 10,844 1,657 0 Green Bacteria Yellow 137 7,794 541 96 7,383 1,069 0

Yellow Fungi Total 54 5,129 555 53 2,945 381 — Yellow Fungi Black 24 394 48 24 208 21 - Yellow Fungi Red 24 70 24 24 45 8 0 Yellow Yeast Total 54 19,761 2,574 212 9,741 1,093 - Yellow Yeast Orange 54 1,670 268 53 290 67 - Yellow Yeast Pink 48 777 206 53 652 187 0 Yellow Yeast Salmon 54 1,595 225 53 1,367 196 0 Yellow Yeast White 54 15,256 1,866 53 7,423 695 — Yellow Bacteria Total 46 672,018 58,920 138 628,682 40,095 0 Yellow Bacteria Orange 46 310,052 45,073 46 284,017 28,301 0 Yellow Bacteria White 46 187,924 23,501 46 195,204 18,275 0 Yellow Bacteria Yellow 46 174,042 26,271 46 149,461 16,984 0 * Diff refers to the difference where the mean for the treated sample was significantly (p=0.05) greater (+) or less (-) than the mean of the untreated sample based on ANOVA of log-transformed values.

118 Table 3.2. Colony forming units (CFU) per gram of leaves of microbial morphotypes obtained from the Poa pratensis multi-season trial of 2005-2006 at the Guelph Turfgrass Institute, Guelph, Ontario. Means were obtained by pooling all collection data across seven sampling dates from 19 March to 9 September 2006. Leaf refers to the state of the foliage, and the organism categories (filamentous fungi, yeast and bacteria) are subdivided into morphotype groupings. N is the number of plates assessed and SE is the standard error. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. CFU/g (thousands) Untreated Treated Leaf Organism Group N Mean SE N Mean SE Diff* Green Fungi Total 102 558 67 90 387 35 — Green Fungi Black 144 20 4 95 18 3 0 Green Fungi Red 143 1 0 95 2 0 + Green Yeast Total 150 844 76 383 1,653 133 0 Green Yeast Orange 126 71 13 77 171 28 0 Green Yeast Pink 147 46 5 102 96 10 + Green Yeast Salmon 150 25 6 102 120 41 0 Green Yeast White 150 676 55 102 1,208 70 + Green Bacteria Total 150 11,915 714 240 22,205 1,387 + Green Bacteria Orange 120 3,396 349 72 4,616 328 + Green Bacteria White 120 4,106 380 72 9,727 946 + Green Bacteria Yellow 144 4,339 466 96 7,747 823 0

Yellow Fungi Total 78 3,428 299 78 3,320 262 0 Yellow Fungi Black 141 213 43 96 154 24 0 Yellow Fungi Red 140 11 3 96 15 4 0 Yellow Yeast Total 150 10,599 1,496 380 8,547 810 0 Yellow Yeast Orange 123 1,749 373 78 1,544 341 0 Yellow Yeast Pink 137 346 71 98 570 127 0 Yellow Yeast Salmon 147 382 94 102 386 83 0 Yellow Yeast White 146 7,852 1,282 102 5,846 540 0 Yellow Bacteria Total 120 664,826 48,332 240 705,214 61,625 0 Yellow Bacteria Orange 66 130,523 25,553 72 101,159 16,613 0 Yellow Bacteria White 68 397,369 32,415 72 483,431 52,773 0 Yellow Bacteria Yellow 92 157,043 17,549 96 149,235 15,232 0 * Diff refers to the difference where the mean for the treated sample was significantly (p=0.05) greater (+) or less (-) than the mean of the untreated sample based on ANOVA of log-transformed values.

119 Table 3.3. Colony forming units (CFU) per gram of leaves of microbial morphotypes obtained from the Agrostis stolonifera pathology green multi- season trial of 2006-2007 at the Guelph Turfgrass Institute, Guelph, Ontario. Means were obtained by pooling all collection data across four sampling dates from 11 November 2006 to 12 July 2006. Leaf refers to the state of the foliage, and the organism categories (filamentous fungi, yeast and bacteria) are subdivided into morphotype groupings. N is the number of plates assessed and SE is the standard error. Treatment consisted of a single application of Heritage MAXX® at 12 g azoxystrobin per 100 m2 on 25 November 2006. CFU/g (thousands) Untreated Treated Leaf Organism Group N Mean SE N Mean SE Diff* Green Fungi Total 117 504 54 68 556 58 0 Green Fungi Black 71 78 10 71 49 11 - Green Fungi Red 72 1 0 72 0 0 — Green Yeast Total 150 2,035 185 277 1,896 214 0 Green Yeast Orange 115 29 7 66 23 3 + Green Yeast Pink 118 143 12 71 108 17 - Green Yeast Salmon 120 380 39 72 207 44 - Green Yeast White 119 1,466 145 68 1,577 146 0 Green Bacteria Total 120 53,290 2,694 198 21,764 1,105 - Green Bacteria Orange 111 18,512 1,658 66 6,968 641 - Green Bacteria White 111 21,771 1,721 66 9,152 713 - Green Bacteria Yellow 111 13,007 1,111 66 5,645 468 - Yellow Fungi Total 111 2,535 231 71 3,036 492 0 Yellow Fungi Black 70 226 54 68 103 21 - Yellow Fungi Red 71 9 5 71 7 2 0 Yellow Yeast Total 150 3,358 355 284 2,785 422 0 Yellow Yeast Orange 113 201 37 71 51 7 0 Yellow Yeast Pink 118 192 20 71 125 16 0 Yellow Yeast Salmon 118 724 152 72 252 38 0 Yellow Yeast White 116 2,237 279 70 2,388 357 0 Yellow Bacteria Total 120 674,544 58,014 199 321,390 32,558 - Yellow Bacteria Orange 111 144,547 16,804 66 47,224 8,442 - Yellow Bacteria White 111 374,805 43,768 66 213,424 26,360 - Yellow Bacteria Yellow 113 156,425 29,833 67 61,434 7,241 - * Diff refers to the difference where the mean for the treated sample was significantly (p=0.05) greater (+) or less (-) than the mean of the untreated sample based on ANOVA of log-transformed values.

120 Table 3.4. Comparisons between microbial populations on green leaves of treated and untreated samples of Agrostis stolonifera of the native sand fairway multi-season trial of 2005-2006. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. 2006* Organism Grouping 19-Mar 31-Mar 19-May 16-Jun 28-Sep Fungi Total - - 0 0 0 Fungi Black - 0 0 Fungi Red 0 0 0 Yeast Total 0 0 0 — 0 Yeast Orange - - - 0 0 Yeast Pink 0 0 + 0 0 Yeast Salmon 0 + 0 0 0 Yeast White 0 — 0 — + Bacteria Total + 0 0 0 Bacteria Orange 0 0 0 0 Bacteria White 0 0 0 0 Bacteria Yellow + - 0 0 The difference of the mean for the treated sample was significantly (p=0.05) greater (+), lesser (-) or not significantly different (0) than the mean of the untreated sample based on ANOVA of log-transformed values. A period (.) indicates when differences were not assessed.

121 Table 3.5. Comparisons between microbial populations on green leaves of treated and untreated samples of the Poa pratensis multi-season trial of 2005- 2006. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. 2006* Organism Grouping 19-Mar 19-May 27-Jul 15-Aug 09-Sep Fungi Total - 0 - 0 0 Fungi Black 0 0 0 0 Fungi Red + 0 + 0 Yeast Total + 0 0 0 0 Yeast Orange + + - 0 0 Yeast Pink 0 + + 0 0 Yeast Salmon + 0 0 0 0 Yeast White 0 + 0 + 0 Bacteria Total 0 + 0 Bacteria Orange 0 + 0 Bacteria White 0 + + Bacteria Yellow 0 + 0 The mean for the treated sample was significantly (p=0.05) greater (+), less (-), or not significantly different (0) than the mean of the untreated sample based on ANOVA of log-transformed values. A period (.) indicates when differences were not assessed.

122 Table 3.6. Comparisons between microbial populations on green leaves of treated and untreated samples of the Agrostis stolonifera pathology green multi- season trial of 2006-2007. Treatment consisted of a single application of Heritage MAXX® at 12 g azoxystrobin per 100 m2 on 25 November 2006. 2007* Organism Grouping 27-Mar 24-May 12-Jul Fungi Total 0 0 - Fungi Black - - 0 Fungi Red 0 — 0 Yeast Total 0 — 0 Yeast Orange - - 0 Yeast Pink 0 - - Yeast Salmon 0 0 0 Yeast White — 0 — Bacteria Total 0 — — Bacteria Orange 0 - - Bacteria White 0 - 0 Bacteria Yellow 0 - - The difference of the mean for the treated sample was significantly (p=0.05) greater (+), less (-) or not significantly different (0) than the mean of the untreated sample based on ANOVA of log-transformed values. A period (.) indicates when differences were not assessed.

123 Table 3.7. Comparisons between microbial populations on yellow leaves of treated and untreated samples of Agrostis stolonifera native sand fairway multi- season trial of 2005-2006. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. 2006* Organism Grouping 19-Mar 31-Mar 19-May Fungi Total - 0 - Fungi Black 0 - Fungi Red 0 0 Yeast Total 0 - 0 Yeast Orange - - - Yeast Pink 0 0 0 Yeast Salmon 0 0 0 Yeast White 0 — 0 Bacteria Total 0 0 Bacteria Orange 0 0 Bacteria White 0 0 Bacteria Yellow 0 + *The difference of the mean for the treated sample was significantly (p=0.05) greater (+), less (-) or not significantly different (0) than the mean of the untreated sample based on ANOVA of log-transformed values. A period (.) indicates when differences were not assessed.

124 Table 3.8. Comparisons between microbial populations on yellow leaves of treated and untreated samples of the Poa pratensis multi-season trial of 2005- 2006. Treatment consisted of a single application of Heritage® (50WDG) at 6 g azoxystrobin per 100 m2 on 28 November 2005. 2006* Organism Grouping 19-Mar 19-May 27-Jul 15-Aug 09-Sep Fungi Total 0 0 0 0 0 Fungi Black - + 0 0 Fungi Red 0 0 0 - Yeast Total 0 - 0 0 0 Yeast Orange 0 - 0 0 0 Yeast Pink 0 0 0 0 0 Yeast Salmon 0 0 + + 0 Yeast White 0 - 0 - 0 Bacteria Total 0 - + Bacteria Orange 0 0 0 Bacteria White 0 0 + Bacteria Yellow 0 0 0 *The difference of the mean for the treated sample was significantly (p=0.05) greater (+), less (-) or not significantly different (0) than the mean of the untreated sample based on ANOVA of log-transformed values. A period (.) indicates when differences were not assessed.

125 Table 3.9. Comparisons between microbial populations on yellow leaves of treated and untreated samples of the Agrostis stolonifera pathology green multi- season trial of 2006-2007. Treatment consisted of a single application of Heritage MAXX® at 12 g azoxystrobin per 100 m2 on 25 November 2006. 2007* Organism Grouping 27-Mar 24-May 12-Jul Fungi Total 0 0 0 Fungi Black - - 0 Fungi Red - 0 0 Yeast Total — — 0 Yeast Orange 0 - 0 Yeast Pink - + + Yeast Salmon - + 0 Yeast White - — - Bacteria Total 0 0 0 Bacteria Orange 0 0 0 Bacteria White + + 0 Bacteria Yellow 0 0 0 The difference of the mean for the treated sample was significantly (p=0.05) greater (+), less (-), or not significantly different (0) than the mean of the untreated sample based on ANOVA of log-transformed values. A period (.) indicates when differences were not assessed.

126 White Salmon

w*fj -$**?£'

Orange Pink

Figure 3.1. Yeast colony morphotypes commonly found on potato dextrose agar after 10 days at 10C.

127 White Yellow

1§H||

HiH

'^\,

Pink Orange

Figure 3.2. Bacterial colony morphotypes commonly found on potato dextrose agar after 10 days at 10C.

128 Black Red

Figure 3.3. Filamentous fungal colony morphotypes commonly found on potato dextrose agar after 10 days at 10C.

129 Native sand fairway Poa pratensis fringe Pathology green '&£, 1.8

w ro / \* g » 1.2 (U T3 |f 0.6 O *1M 0.0 g"10.0 d - "e°n 5.0

a 2.5 •*-••*. Vw *.*- 0.0 ——*su-i v IT 75

Mar 06 Jun 06 Sep 06 Mar 06 Jun 06 Sep 06 Oct 06 Jan 07 Apr 07 Jul 07 Figure 3.4. Total microbial populations (bacteria, yeast and filamentous fungi) of treated (full line) and untreated (dashed line) green leaves of the native sand fairway 2005-2006, pathology green 2006-2007 and Poa pratensis fringe 2005-2006. Treatment consisted of a single application of Heritage® 50WDG at 6 g azoxystrobin per 100 m2 applied on 28 November 2005 on the native sand fairway 2005-2006 and P. pratensis fringe 2005-2006 or MAXX® at 12 g azoxystrobin per 100 m2 applied on 25 November 2006 on pathology green. The error bars represent the standard error. Native sand fairway Poa pratensis fringe Pathology green c J 13 CO o >> c T3 CD E •£? CO u. O

0 12 e < \ f * CO CO CO .co 8 CD * * >^ >- I— * * 3 4 LL. O ^A, 0 900 t. g i CO w 0) CO 600 o ^ ro >. TJ i X O) 300 f u. - • . z O Mar 06 Jun 06 Sep 06 Mar 06 Jun 06 Sep 06 Oct 06 Jan 07 Apr 07 Jul 07 Figure 3.5. Total microbial populations (bacteria, yeast and filamentous fungi) of treated (full line) and untreated (dashed line) yellow leaves of the native sand fairway 2005-2006, pathology green 2006-2007 and Poa pratensis fringe 2005-2006. Treatment consisted of a single application of Heritage® 50WDG at 6 g azoxystrobin per 100 m2 applied on 28 November 2005 on the native sand fairway 2005-2006 and P. pratensis fringe 2005-2006 or MAXX® at 12 g azoxystrobin per 100 m2 applied on 25 November 2006 on pathology green. The error bars represent the standard error. CHAPTER 4. RESTRICTION FRAGMENT LENGTH POLYMORPHISMS AND SEQUENCE ANALYSES OF MICROBIAL ISOLATES FROM TURFGRASS PHYLLOSPHERES

4.1 Introduction When diversity is applied to living organisms, it is called biological diversity or biodiversity, and it refers to "the variety of all forms of life, from genes to species, through to the broad scale of ecosystems" (Faith, 2007). All organisms present in an ecosystem form part of a community that comprises many populations of single species (Odum, 1959). Thus, the estimation of biodiversity can be achieved by counting all organisms and establishing the proportions of populations of the community. This simple approach is not always possible since obtaining and identifying all organisms from an environmental sample is not achievable because time, money, technology and knowledge are limited.

Consequently, methods are used to estimate biodiversity by using indices that reflect the number and relative proportions of species in a community (Morris and

Rouse 1986). In addition, scientists focus their efforts on particular types of organisms within the total community, which can be as diverse as insects of the

Amazon or bacteria from the surface of human teeth. Moreover, when identification of species is not possible, genera rather than species may be enumerated, and for microbes with few differentiating macroscopic characteristics, even broad groupings such as color-based morphotypes may be used.

Microbial diversity encompasses an immense range of different kinds of unicellular organisms, including bacteria, archaea, protists, yeasts and some

132 multicellular organisms, such as filamentous fungi, arthropods and small plants.

The number of microbial species that are still unidentified surpasses by a great

extent those that can be identified (Hawksworth, 1991; Scow et al., 2001).

Morphological identification of microorganisms is not feasible for microbial

biodiversity, since limited differentiating features do not permit easy counting or

identification. Microscopy is used to count and to some extent identify microbes,

but it requires special treatment of samples that can disturb important characteristics of the niche. Small organism size also affects the reliability of sampling, since cells are difficult to separate from environmental sample, such as bacterial cells bound to soil particles (Atlas and Bartha, 1987). Maintaining microorganism viability under lab conditions has also proven to be a major limiting factor in microbial ecology. In 1985 Stanley and Konopka proposed the

"plate count anomaly" referring to the underestimation of many microbial species that cannot grow under regular laboratory conditions. As a consequence of these great challenges, microbial ecologists have developed numerous methods for characterizing microbial communities.

Large advances in microscopic techniques in other fields, such as immunology, fluorometry, computational imaging and molecular genetics, have solved many microbial ecology problems stemming from small size. As for identification and cultivation problems, molecular biology has also provided new methods that do not require culturing and facilitate the identification of organisms that can be cultivated (Tiedje et al., 1999). The new techniques, however, have

133 their own problems and biases, but they provide a complement to the traditional methods, and together give a better understanding of microbial biodiversity.

Molecular methods designed to study microbial diversity have generally focused on nucleic acid sequences, since nucleotide differences offer strong evidence to distinguish organisms that share the same phenotype. The polymerase chain reaction (PCR) is commonly used because it can amplify

(produce many copies of) a specific gene or region from a mixed pool of nucleic acids, and which may be subsequently separated electrophoretically to give diagnostic banding patterns or sequenced for identification (Scow et al., 2001).

Ribosomal DNA (rDNA) is present in all forms of life and exhibits a high degree of conservation (Yonath and Franceschi, 1999). This region has been used to elucidate phylogenetic relationships and community composition of many environmental samples, such as soils (Bomeman and Triplett, 1997), hydrothermal vent systems (Moyer et al., 1994) and plant residues (Matsuyama etal.,2007).

The basic structure of ribosomal DNA (Figure 1.7) is a cassette repeated tens to thousands of times in a genome (Hershkovitz et al., 1999). Eukaryotes have rDNA units formed by the small subunit (SSU or 18S rDNA), the internally transcribed spacer (ITS) 1, the 5.8S gene, the ITS2, the large subunit (LSU or

28S rDNA) and the intergenic spacer region (IGS). In prokaryotes, the rDNA unit is comprised of a SSU known as the 16S rDNA and a LSU that contains the 5S and 23S rRNA (Weider, et al., 2005). The differences between rDNA sequences

134 permit comparative studies of phylogenetic relationships over a wide range of

taxonomic levels (White et al., 1991).

One of the most common methods that has been applied to assess

microbial communities using PCR products of rDNA is restriction fragment length

polymorphism (rDNA-RFLP) (Beebee and Rowe, 2004). The RFLP method is

based on the presence of polymorphisms in the DNA, which is recognized by

particular restriction enzymes that cut specific base sequences. After the

amplified region is digested by a series of restriction enzymes, fragments from

the original sequence are separated by size using electrophoresis to generate a

characteristic banding pattern. Banding patterns are used to compare the same

region of DNA of diverse organisms, establishing differences or similarities (Scow et al., 2001). RFLP has been used for rapid comparison of rDNAs (Liu et al.,

1997) because it offers a high level of resolution (Tiedje et al., 1999).

The rDNA-RFLP method has been used to detect foliar microbial communities of many plants. For instance, phytoplasma genetic variability from leaves of sugar cane (Saccharum spp.) (Wongkaew et al., 1997), pathogenic and non-pathogenic epiphytes of apple (Malus domestica) (Jeng et al., 2001) and bacteria on potato leaves (Solanum tuberosum) (Berg et al., 2005) were studied by the rDNA-RFLP technique.

Foliar niches harbor microbial species that are commonly referred to as epiphytes if they reside on leaf surfaces, or endophytes if reside inside the leaf blade. The microbial component of leaves has high biodiversity of bacteria, filamentous fungi, and yeasts. Yang et al. (2001) found 85 different phyllosphere

135 bacterial species in only four plant hosts. Moreover, Stohr and Dighton (2004) reported 107 fungal genera on leaves of American cranberry (Vaccinium macrocarpon). Thus, the phyllosphere is an important ecological niche not only because of the high number of microorganisms present, but also because it has a relatively high diversity.

Normally bacterial phyllosphere communities are comprised of only a few populations that dominate while others remain at lower frequencies (Hirano and

Upper, 2000). Many epiphytic bacteria produce extracellular polysaccharides that assist surface attachment and prevent desiccation, and pigments that increase their UV tolerance (Beattie and Lindow, 1995). The most common bacterial genera associated with leaves are Actinobacter, Azotobacter, Bacillus,

Beijerinckia, Corynebacterium, Erwinia, Flexibacter, Listeria, Methylobacterium,

Micrococcus, Pantoea, Pseudomonas, Sphingomonas, Staphylococcus and

Xanthomonas (Dickinson, 1982; Hirano and Upper, 2000).

Both ascomycetous and basidiomycetous yeasts are normally part of the phyllosphere, and Aureobasidium, Candida, Cryptococcus, Rhodotorula,

Sporobolomyces, Tilletiopsis, and Torulopsis are the most regularly encountered genera (Dickinson, 1982). Filamentous fungi such as species of Alternaria and

Cladosporium, are commonly found as phyllosphere inhabitants (Dickinson,

1982). Other epiphytic filamentous fungal genera are Ascochyta, Aspergillus,

Epicoccum, Leptosphaeria, Myrothecium, Penicillium, Phoma, Pilobolus,

Pleospora, and Stemphylium (Atlas and Bartha, 1987; Petrini, 1986). In addition,

136 endophytic filamentous fungal genera include Epichloe, Linodochium,

Phyllosticta and Rhabdocline (Petrini, 1986).

Grass leaves harbor similar genera as those of other plants, but additional

bacterial genera have been reported only on grasses. For instance,

Microbacterium, Curtobacterium (Behrendt et al., 2002) and enterococci (Ott et

al., 2001) were isolated from grass leaves. Grass leaves are colonized by the

same genera of yeasts commonly found in or on leaves of other plants, however,

Pseudozyma yeasts are more common on grasses (Allen et al., 2004b). Most

identified phyllosphere filamentous fungi of grasses are pathogenic or endophytic

since most effort has been made to study these types.

In the previous chapter, populations of bacteria, yeast and filamentous fungi were sampled from leaves of Agrostis stolonifera and Poa pratensis. The organisms were further subdivided into morphological groups based on culture characteristics, particularly color. The purpose of this chapter was to further investigate the morphotypes to see whether they were composed of single or multiple species using RFLP and nucleic acid sequencing. Banding patterns from

RFLP of rDNA allowed ribotyping and comparisons to the phenotypic characteristics of the isolates. In addition, rDNA sequences of selected isolates were identified by comparison to publicly available sequence databases to complement the morphological analyses and further resolve differences between similar morphotypes.

137 4.2 Materials and methods

4.2.1 Microbial isolation

4.2.1.1 Leaf samples Leaf samples collected from plots at the Guelph Turfgrass Institute (GTI) were used to obtain isolates of the phyllosphere microbial community. Three groups of plots were used: a native sand fairway trial (2005-2006), P. pratensis fringe trial (2005-2006) and pathology green trial (2006-2007). Each trial consisted of eight plots (1 m x 2 m) where four plots (even numbered) were treated with azoxystrobin (6 g per 100 m2 in 2005-2006 trials and 12 g per 100 m2 in the 2006-2007 trial), while the other four plots (odd numbered) had an equivalent amount of water applied (untreated plots)

Leaves and stems were harvested from 12 locations (Agrostis stolonifera plots) or nine locations (Poa pratensis plots) within each plot using latex gloves and scissors surface sterilized with ethanol 70% (v/v) and placed into sterile 50 ml polypropylene centrifuge tubes. Samples were stored at 4C for less than three days, and then leaves were separated into two groups (green or yellow) based on senescence. First, leaves and stems of each sample tube were mixed by breaking the clusters from of the different subsamples of each plot separately.

Leaves were first separated from stems using flamed forceps and segregated into green or yellow leaves as measure on senescence. Only leaf blades were retained with sheaths cut off at the collar with a flamed scalpel. Green leaves had complete green blades without noticeable wounds or chlorotic areas, while yellow leaves were yellow to brown, had less than 10% green area across the blade and

138 did not have to be intact. Ten to 12 green and yellow leaves of A stolonifera or five green and yellow leaves of P. pratensis from each plot were then placed separately using flamed forceps into sterile 1.5 ml polypropylene tubes (Fisher

Scientific, Mississauga, Ontario, Canada), and stored at 4C until phyllosphere microorganisms were isolated (maximum of 48 hours).

4.2.1.2 Epiphyte isolation Microorganisms were recovered from leaf surfaces by washing and plating.

Leaves from each subplot were separated into subsamples of 10 to 12 leaves of

A. stolonifera or five leaves of P. pratensis leaves in 1.5 ml tubes containing 1 ml sodium chloride solution (0.85% w/v). Tubes were vortexed at maximum speed for 30-40 s followed by sonication for 10 min at 45 watts in a Crest 175TA sonicator (Crest Ultrasonics, Trenton, New Jersey, USA).

After washing, suspensions were serially diluted up to 106 using 0.85%

(w/v) sodium chloride as the diluent. Diluted suspensions (50 pi) were plated on

Difco® potato dextrose agar (PDA) amended with streptomycin at 100 ug/ml plus tetracycline at 5 ug/ml or PDA amended with benomyl at 2 ug/ml. Plates were incubated at 10C until colonies started to form on the agar surface.

Colonies were isolated using a flamed platinum loop or scalpel after 3 to 9 days of incubation depending on the maturity of colonies. Selection of colonies was performed systemically while attempting to maintain an even proportion of the main morphological groups. Additionally, two colonies of the most rarely occurring colonies were also selected and isolated. This procedure was performed for each treatment combination: host (A. stolonifera or P. pratensis)

139 leaf type (green or yellow) and pesticide treatment (azoxystrobin or untreated).

For bacteria and yeasts, selected colonies were streaked onto PDA plates.

Filamentous fungi were subcultured by placing a piece of agar with hyphae on the surface of a PDA plate. Plates were incubated at 25C until morphological characteristics were evident and growth permitted collection for storage.

For the long term storage, approximately 50 mg of bacterial or yeast cells were scraped from the agar surface with a narrow spatula and transferred to 0.6 ml tubes containing 400 ul of glycerol 20% (v/v). Tubes were then gently vortexed to break up clumps of cells and frozen at -20C. Isolates of filamentous fungi were stored in 15 ml vials containing 5 ml PDA. A piece of agar from the border of a subcultured fungal colony was placed inside a vial on the agar surface and incubated at 25C for seven to 10 days with cap loose. When fungal growth covered the agar surface in the vial, the cap was tightened, wrapped with

Parafilm® and stored at 4C.

For DNA extraction, isolates had to be grown for a longer period than was used in dilution plating, and this permit further categorization into submorphotypes beyond what was observed in Chapter 3. Morphotypes of yeast and bacteria were separated into submorphotypes as slimy or not slimy based on the level of wetness. Some white yeast isolates, after incubation periods longer than 10 days, changed their color to light brown or beige and became easily distinguishable from those that remained white. In total, there were seven bacterial submorphotypes: pink, orange, orange slimy, white, white slimy, yellow and yellow slimy. The white yeast morphotype was the only one to be split into

140 submorphotypes: white, white slimy and beige; while the others remained as

originally classified, either pink (always slimy), orange or salmon.

Separation of filamentous fungal isolates was more complex because

filamentous fungi offered more distinguishable morphological characteristics than

bacterial or yeast colonies. Fungal isolates were categorized into 12

submorphotypes by the color and texture present on both sides of PDA plates after incubation at 25C for up to 14 days (Figure 4.1). The 12 morphotypes were as follows (topside/bottom-side): black/black (bl/bl), green/black (gr/bl), green/brown (gr/br), grey (g), orange (o), white/beige (w/b), white/brown (w/br), white fluffy (wf), white/orange (w/o), white/pink (w/p), white/red (w/r) and white/white (w/w). Some isolates had characteristics similar to two or more of the

12 submorphotypes and were categorized as unknown submorphotypes.

4.2.1.3 Endophyte isolation Two surface disinfecting procedures were used in attempts to recover endophytes. Under a dissecting microscope, healthy green leaves from A stolonifera and P. pratensis plants were chosen, selecting those that were complete and without visible wounds from insects, or pathogens or other sources.

Selected leaves were surface-sterilized by soaking in 1.25% (w/v) sodium hypochlorite (diluted household bleach) for 15 min, followed by two rinses with autoclaved deonized water (Moy et al., 2002). In the second protocol, leaves were soaked for 1 min in 70% (v/v) ethanol and then 3 min in 2% hypochlorite

(diluted commercial bleach) followed by 30 s in 70% ethanol (v/v). Leaves were then rinsed twice in sterilized deonized water (Kuklinsky-Sobral et al., 2004).

141 After surface decontamination, four leaves treated as in Moy et al. (2002) or

Kuklinsky-Sobral et al. (2004) were placed on the surface of PDA amended with

streptomycin (100 ug/ml) and tetracycline (5'ug/ml) (Moy et al., 2002). Leaves

from the Kuklinsky-Sobral method were also ground using sterilized, acid- washed sand in autoclaved mortars and pestles. The ground material was transferred to 50 ml flasks with 10 ml of a phosphate-buffered saline medium

(137 mM NaCI, 10 mM phosphate, 2.7 mM KCI, pH 7.4) and incubated at 25C in an oscillator shaker (100 rpm) for 1 hour. Serial dilutions 1:10, 1:20 and 1:50 were then prepared with autoclaved water, and 50 pi were plated on trypticase soy agar (TSA) containing 2 ug/ml benomyl. The plates from both methods were incubated at 25C in darkness for up to 3 weeks.

4.2.2 DNA extraction

4.2.2.1 Epiphytes DNA was extracted from selected isolates obtained from leaf washings.

Bacterial and yeast cells were collected as described above from the surface of

PDA plates, while hyphae of filamentous fungi were obtained using two different methods. In the first method, autoclaved cellophane films were placed on top of the surface of PDA plates, and the fungi allowed to grow over the film. Afterwards, the mycelium was scraped off with a sterile spatula or scalpel and transferred to

1.5 ml tubes. In the second method, fungal inocula from vials were transferred into 125 ml flasks containing 40 ml of 1.5% (w/v) malt broth (BD, Franklin Lakes,

New Jersey, USA). Flasks were shaken (100 rpm) at 25C until the amount of mycelium was enough for DNA extraction (>1 g in usually after five to seven

142 days). The shake cultures were vacuum-filtered through Whatman #1 filter paper

(Brentford, Middlesex, England), and the mycelial residue was then rinsed twice with sterile deionized water. Mycelia were then transferred to 1.5 ml tubes using a flamed spatula. Mycelia and bacterial and yeast cells were preserved at -20C until the DNA was processed.

DNA was extracted from all samples using the method of Edwards et al.

(1991). Approximately 100 mg of microbial cells were transferred to 1.5 ml tubes to which 100 mg of autoclaved, acid-wash sea sand and 150-200 ul of extraction buffer (Tris-HCI 200 mM, EDTA 25 mM, NaCI 250 mM, sodium dodecyl sulfate

0.5% (w/v)) was added. The mixture was vortexed for a few seconds, and cells were mechanically disrupted with pestles tips and a motorized pellet pestle mortar (Kontes, Vineland, New Jersey, USA) for 60 s for bacteria and yeast, and

80 s for filamentous fungi. Next, 200 ul of extraction buffer was added and vortexed for few seconds. The tubes were incubated at room temperature for 2-3 hours, after which they were spun at 12,000 x g in a Spectrofuge 16M centrifuge

(Mandel Scientific Company, Guelph, Ontario, Canada) for 10 min.

The supernatant was transferred to clean 1.5 ml tubes and centrifuged at

12,000 x g for 5 min. The supernatant was then transferred to clean 1.5 ml tubes to which 350-400 ul of -20C isopropanol were added and vortexed immediately.

Tubes were stored at -20C for 30 min to 1 hour and centrifuged at 12,000 x g for

10 min. The supernatant was discarded, and the pellet was retained and then washed with 400 ul of cold (-20C) ethanol 70% (v/v) for 10 min. Next, the ethanol solution was discarded, and the pellets were allowed to dry by blowing air into

143 the tubes. Once the pellet was dried, 100 to 200 ul of Tris-EDTA (TE, 10 mM Tris

HCI, 1 mM EDTA pH 7.5) was added, and the pellet gently disrupted with a

pipette tip. Tubes were stored overnight at 4C to allow the DNA to dissolve into

solution. The tubes were then centrifuged for 5 min at 10,000 x g, and the

supernatant transferred to 0.6 ml tubes which were stored at -20C.

4.2.2.2 Endophytes To obtain DNA from microorganisms living inside the leaves that were not

recovered by the leaf washing procedure, DNA extractions were performed on

leaf tissue. The method chosen was DNA extraction from leaves (Dellaporta et

al., 1983) with modifications by Adams and Demeke (1993). One gram of selected green leaves was submerged in liquid nitrogen and ground immediately

in a chilled mortar and pestle with 0.1 g of sterile acid-washed sea sand.

Approximately 100 mg of leaf tissue and sand were transferred to a 1.5 ml tube with 400 Ml of extraction buffer (100 mM Tris-HCI (pH 8.5), 50 mM EDTA, 1.4%

SDS, 10 mM 2-mercaptoethanol, and 1% (w/v) polyvinyl pyrrolidone (PVP)).

After mixing, the tubes were incubated at 65C for 10 min, and then 250 ul of

5 M potassium acetate was added, followed by 10 s finger vortexing and incubation in crushed ice for 30 min. Tubes were spun at 13,000 x g for 10 min, and the pellet was washed with (-20C) 80% (v/v) ethanol. After the pellets were air-dried, 700 pi of TE buffer (50 mM Tris-HCI (pH 8.0), 10 mM EDTA) was added, followed by 10 s vortexing at maximum speed. The suspension was centrifuged at 13,000 x g for 10 min, and the insoluble fragments discarded. The supernatant was extracted with 700 pi of phenol-chloroform (1:1). The aqueous phase was

144 transferred to a 1.5 ml tube, mixed with 75 ul of 3 M sodium acetate (pH 5.2) and,

500 ul of cold isopropanol to precipitate DNA, which was collected by centrifugation (13,000 x g for 15 min). The pellet was dried in a laminar flow chamber and resuspended in 80 pi of TE buffer (10 mM Tris-HCI, 1 mM EDTA).

The DNA suspensions were stored at -20C until required.

4.2.3 DNA concentration Total DNA extracted from microbial samples and polymerase chain reaction

(PCR) products were visualized and concentration estimated in 1% (w/v) ultra pure agarose gels (Invitrogen, Carlsbad, California, USA) made with Tris-Borate-

EDTA buffer (TBE) (90 mM Tris base, 90 mM boric acid and 2 mM EDTA).

Electrophoresis was done at 100 V and 0.8 A in a Mupid 21 electrophoretical chamber (Helixx Technologies, Toronto, Ontario, Canada). The size marker

O'GeneRuler™ DNA Ladder Mix (5 pi) (Fermentas) was used to assess the DNA size (bp) and estimate the concentration of DNA (ng/ul).

After electrophoretic separation, gels were stained in a 2-4 pg/ml ethidium bromide solution for 5 min. If required, the gels were destained in tap water for 5 to 10 min. An ultraviolet transilluminator Syngene (Synoptics, Cambridge,

Cambridgeshire, UK) permitted the visualization of DNA bands that were recorded with a GBC video camera CCTV (South Hackensack, New Jersey, USA) and a video copy processor P67U (Mitsubishi Electric, Cypress, California, USA).

Images were also saved as black and white jpeg files (resolution 800 x 600) with a frame-grabber card attached to a desktop computer. These images were edited by changing to a negative image, and adjusting the brightness and contrast

145 levels using Jasc Paint Shop Pro version 7.00 (Corel Corporation, Ottawa,

Ontario, Canada).

4.2.4 PCR amplifications

4.2.4.1 Internally transcribed spacer (ITS) amplification for fungi The internally transcribed spacer region was amplified using the conserved fungal primers, ITS1 (5'-TCCGTAGGTGAACCTGCGG) and ITS4 (5-

TCCTCCGCTTATTGATATGC), described by White et al. (1991). These gave a

PCR product between 400 and 700 bp depending on the fungal isolate. The reaction mixture consisted of approximately 1 ng DNA, 1X DNA Tsg buffer (10 mM KCI, 10 mM (NH4)2S04, 20 mM Tris HCI (pH 8.75), 0.1% (v/v) Triton X-100,

0.1 mg/ml bovine serum albumin (BSA) and 2.5 mM MgS04), 0.5 uM of each primer, 0.2 mM of dNTP and 1 U Tsg polymerase per 100 ul of reaction mix

(Biobasic, Markham, Ontario, Canada). The amplification was performed in a

GeneAmp PCR System 2400 (Perkin Elmer, Waltham, Massachusetts, USA) or in a Mastercycler Personal (No 5332 45205) (Eppendorf, Hamburg, Germany) with the following cycling conditions: denaturation at 94C for 2 min, followed by

35 cycles at 94C for 30 s, 55C for 1 min, and 72C for 2 min, and a final extension at 72C for 10 min (Hsiang and Wu, 2000).

4.2.4.2 Ribosomal 16S amplification for bacteria A region of the bacterial 16S gene was amplified by PCR with the forward primer F0 (5'-AGAGTTTGATCCTGGCTCAG-3'), also known as the FD1 of

Weisburg et al. (1991), and the reverse primer, PRUN518r (5-

ATTACCGCGGCTGCTGG) (Muyzer et al., 1993). Primers were chosen based

146 on highly conserved regions of 38 commonly reported phyllosphere bacteria. The primer melting temperatures and binding characteristics were obtained using

GeneRunner 3.04 (Hastings Software, New York, New York, USA), and primer sequences were compared to the GenBank databases using BLASTN (Altschul et al., 1997). The F0/PRUN518r primer pair amplifies a 518 bp fragment in the

16S of Escherichia coli rDNA (Nakatsu et al., 2000). The PCR master mix and the thermocyclers used were the same as in the ITS amplification. The amplification program used was a modification of that by 0vreas et al. (1997), which consisted of a denaturing cycle at 92C for 2 min; 30 cycles of denaturation at 92C for 1 min, annealing at 59C for 30 s, and extension at 72C for 1 min; and a single final extension at 72C for 6 min. Both amplified DNA regions (ITS and

16S) were visualized with the same procedures and equipment used to evaluate

DNA concentration.

4.2.4.3 Restriction Fragment Length Polymorphism (RFLP)

4.2.4.4 Restriction enzymes digestions To choose the enzymes for the RFLP analyses two approaches were used.

For the 16S products, 15 enzymes with restriction sites no longer than six base pairs were chosen for in silico analyses (Appendix G). Forty-five sequences of

22 bacteria genera reported to form part of phyllosphere (Appendix H) were downloaded from GenBank, and the 16S sequences analyzed by with Gene

Runner 3.0 software (Hastings Software) for restriction enzyme cut sites.

Enzymes with more than two restriction sites among several 16S sequences were tested in the laboratory. Up to 12 PCR amplified sequences of

147 morphologically different bacterial isolates amplified with F0/PRUN518r or

F0/R1444 primer sets and restriction digested. From the six enzymes (Alu\, Cfo\,

EcoRI, Haelll (=BsuR\), HindU and HpaU (=Msp\)) tested four (Alu\, CM, Hae\\\

and /-/pall) were chosen based on resolving power of banding patterns observed.

The enzyme EcoRI yielded the same banding pattern in all PCR products tested,

while HindU did not produce cuts and hence were not used.

Restriction enzymes for fungal ITS analyses were chosen based on

previous work and empirical tests. The restriction enzymes, Alu\, HaeU\ and

/-/pall, have been used in other ITS studies on fungal diversity (Gomes et al.,

2002; Krupa, 1999), while Cfo\, EcoRI and HindU were tested with 10 ITS PCR

products in the laboratory. The enzymes selected were Alu\, EcoRI, HaelW, and

Hpall (Fermentas, Burlington, Ontario, Canada) for yeasts. Filamentous fungal

ITS products were additionally digested with C/bl (Promega, Madison, Wisconsin,

USA).

All digestions reactions were done using the manufacturer's buffer at 1 X, and 15 U of restriction enzyme per 100 ul of reaction solution. Between 0.5 and 1

ug of DNA were added to the reaction mix for 15 ul digestion reactions. The digestion reactions were incubated overnight in a water bath at 37C, and the reaction products were either processed immediately, or the reaction was stopped by increasing the temperature to the enzyme denaturing temperature as suggested by the manufacturer.

The buffers used in the enzymatic digestions were as follows: Tango™ used in Alu\ and HpaU (3.3 mM Tris-acetate pH 7.9, 1 mM magnesium acetate,

148 6.6 mM potassium acetate, 0.01 mg/ml BSA), buffer R used with Haelll (1 mM

Tris-HCI pH 8.5, 1 mM MgCI2, 10 mM KCI, 0.01 mg/ml BSA), buffer EcoRI (5 mM

Tris-HCI pH 7.5, 1 mM MgCI2, 10 mM NaCI, 0.002% Triton X-100, 0.01 mg/ml

BSA) and buffer B used for Cfo\ (1 mM Tris-HCI pH 7.5, 1 mM MgCI2, 0.01 mg/ml

BSA). All buffers were supplied with the restriction enzymes.

Digested PCR products were screened using electrophoresis in agarose gels. Electrophoretic parameters, visualizations and image recording were the same as those used for DNA concentration determination, except for the use of

2% (w/v) agarose gels. Fragments of digested PCR products were recorded by size based on two outer lanes loaded with 2 pi (0.5 ug DNA) of GeneRuler™ 50 bp DNA ladder (Fermentas). The minimum band size recorded was 50 bp, and the recorded band intervals were 25 bp or 50 bp depending on the restriction enzyme used.

4.2.4.5 Band recording and analysis Band data were entered in a spreadsheet in a binary format in which 1 indicated the presence of the band, and 0 indicated the absence of that specific band. The data was subsequently saved as a text file for further processing.

Dendrograms were obtained using the distance and similarity analysis tool

WinDist of WinBoot (Yap and Nelson, 1996). The default DICE distance coefficient (dice genetic distance = 1 - 2nxy / (nx + ny) from Nei and Li (1979)) was used for analyses, and the results were obtained in Phylip format

(Felsenstein, 1989). The Phylip file was analyzed using the un-weight pair group method with arithmetic mean (UPGMA) algorithm of Phylip 3.5 (Felsenstein,

149 1989) with a lower-triangular matrix option. The output was visualized as a

phylogram tree with TreeView 1.6.1 (Page, 1996). Additionally, for some

dendrograms, bootstrap analyses were done with WinBoot (Yap and Nelson,

1996) to determine the robustness of the branching patterns.

4.2.5 DNA sequencing and alignment Four hundred and fifty PCR products obtained from isolates of the multi-

season pathology green trial of 2006-2007 were chosen for nucleic acid sequencing. ITS and 16S PCR products, from November 2006 and March 2007

isolates, were selected from a representative variety of morphotypes independently of the type of leaf and treatment to which they belong, to make up

150 samples. Among those 150 samples, 50 were chosen from each microorganism group (filamentous fungi, yeast and bacteria) as representative samples from all morphotypes and conditions (leaf stage and treatment). In May and July 2007, two more batches of 150 isolates were selected from each collection date based on morphotype and leaf type of the isolate. The protocol was to choose at least two isolates from each combination of morphotype, leaf and treatment.

PCR products were sent for nucleic acid sequencing at the College of

Biological Sciences DNA facility of the University of Guelph. Samples were diluted to the specified concentration (10 ng/ul) and sent along with enough forward primer to perform the sequencing. The samples were sequenced using the capillary DNA device 3730 DNA Analyzer (Applied Biosystems, Foster City,

150 California, USA), which gave a text file and a chromatogram for each PCR

product.

The text files were saved individually as FASTA format, deleting the

unknown bases (Ns) from the beginning and end of the sequences until no more

than four consecutive Ns were found. If unknown bases were present in the

middle of the sequence, the chromatogram was examined (Chromas Lite 2.0,

Technelysium Ltd. Tewantin, Queensland, Australia.), and the unknown was

replaced with the base with the highest peak. All the FASTA files obtained were

compared against the GenBank NR database using BLASTN in megaBLAST.

The match with the highest score (e values below 10"50) and with annotated genus and species were selected as the species representative of the leaf isolate

(Appendices I, J and K). If the 50 top matches did not have an annotated species, the best match with a specified genus was selected.

4.3 Results Phyllosphere microflora of turfgrass were isolated and analyzed with molecular techniques to obtain more information about the complexity of the microbial community. Although attempts were made with two different procedures, endophytes were not isolated from foliage on turfgrass samples.

Only on one occasion, a red filamentous fungus was obtained from A. stolonifera leaves collected from the native sand fairway using the endophyte isolation method. This isolate resembled the red morphotype found on leaf wash platings as described in Chapter 3. This isolate was then considered an epiphyte that either survived the surface sterilization or was in a microscopic wound not

151 observed with the stereoscopic microscope. Additionally, DNA extraction of

surface sterilized ground leaves did not yield any PCR amplified fungal ITS or

bacterial 16S products. Thus, the efforts of this study were focused only on

epiphytes.

4.3.1 Epiphyte collection Morphotypes used for dilution plate assessment in the population

experiments (Chapter 3) were found have subtle variations that were only visible

when the colonies were subcultured and allowed to grow for longer periods

(Figures 4.1 to 4.3). Beginning with two, four and three morphotypes found in

Chapter 3, twelve, six and seven submorphotypes were identified for filamentous

fungi, yeast and bacteria, respectively. Even within each submorphotype, other

subtle morphological variations could be observed. These variations were not

further subdivided into more morphotypes, because it was too difficult to

establish consistent characteristics, and this would have increased grouping

errors. The variations of white yeasts were evident when large masses of the

isolated yeasts were grown on PDA plates after seven days (Figure 4.2). Some

of the originally white colonies matured to diverse tones of beige and textures.

These changes, however, were difficult to detect with small colonies. All selected

classified submorphotypes was used for molecular analyses with PCR, restriction

enzymes and DNA sequencing to assess the number of species composing each

submorphotype.

The number of microbial isolates obtained from leaf samples collected in

2006 from the A. stolonifera native sand fairway and the P. pratensis fringe was

152 1,930 (Table 4.1). From the A. stolonifera leaves of the pathology green, 2,232

microbial isolates were obtained to total 4,162 isolates in this study. The

morphological characteristics of each isolate were recorded and stocks were

made. The number of isolates of each morphotype is described in Table 4.2.

Three filamentous fungal morphotypes, black/black, white/beige and green/black,

comprised more than 40% of all isolates. The nine remaining morphotypes

represented less than 9% each. The most commonly isolated yeast morphotype

was white followed by white slimy and pink. White/beige yeasts represented 12%

of the total number of isolates, and orange was the morphotype with least

isolates. The different bacterial morphotypes were more evenly distributed in

number. Yellow had the highest number comprising 21%, and white slimy had the lowest, reaching only 9% of all bacterial isolates.

4.3.2 Restriction Fragment Length Polymorphism Selected microbial isolates were used for the RFLP analyses. The number of selected isolates was based on their representation in each morphotype

(Table 4.2). DNA was extracted successfully from 61 filamentous fungi isolates.

ITS was amplified and digested with one of five restriction enzymes. For yeast,

126 isolates were used for the RFLP analyses, while 39 bacterial isolates were analyzed.

Extraction of DNA (Edwards et al., 1991) usually yielded sufficient quantity and quality of DNA for PCR amplification of the two regions chosen for the RFLP analyses. The 16S fragments amplified with the primer pair, F0/PRUN518r, were slightly larger than 500 bp (Figure 4.4). In yeasts, the range of sizes obtained

153 with primers ITS1 and ITS4 was between 450 and 700 bp, while filamentous fungi had a less variable ITS size, closer to 550 bp (Figures 4.5 and 4.6).

4.3.2.1 Bacterial RFLP RFLP of bacterial PCR products with four enzymes (Alu\, Cfo\, Hae\\\ and

/-/pall) produced banding patterns with one to seven bands depending on the enzyme and PCR product (Figure 4.4). The analyses yielded a dendrogram that did not always cluster by morphotype (Figure 4.7). Bacterial submorphotypes did not necessarily share the same digestion pattern. Consequently, the dendrogram showed many deep branches with similar morphotypes scattered across highly distant branches. The white and white-slimy morphotypes especially had very high variation and were scattered across the dendrogram. Most of the pink and orange bacterial submorphotypes clustered in their own branches.

Representatives of the yellow slimy (ys) morphotype also generally grouped together. However, the separation of morphotypes often did not show high levels of statistical support as seen by the low bootstrap values. The pink group was the submorphotype with the highest similarities, and the representatives were on branches with bootstrap values from 46% to 70%. The orange subgroup had bootstrap values from 41% to 70%. The dendrogram did not show differential clustering of isolates collected from yellow vs. green leaves or treated vs. untreated leaves.

4.3.2.2 Yeast RFLP Restriction digests of PCR amplified DNA of yeasts with four enzymes (Alu\,

EcoRI, Haelll and Hpall) produced clear banding patterns (Figure 4.5). The

154 number of bands obtained with these enzymes ranged from one to three. RFLP

analysis of these banding patterns produced a dendrogram in which

submorphotypes were separated into different clusters (Figure 4.8). All the pink

isolates tested in the RFLP showed exactly the same banding patterns and therefore formed their own branch. The branch with pink isolates was more similar to the salmon yeasts than to the white morphotype, but they still were separated by a long genetic distance and only shared a bootstrap value of 3. The salmon submorphotype was divided into groups which clustered by yellow leaves or green leaves. The white morphotype was interspersed with the white slimy morphotype among 29 deep branches, but many had very low bootstrap support

(Figure 4.8). Most of the white slimy morphotype isolates were found in a cluster with three branches with 67% bootstrap support. Orange and white-beige isolates were divided separately into their own groups with moderate bootstrap support. As with pink isolates, orange, white, white beige and white slimy isolates did not show distinct separation by the type of leaf from which they were isolated.

4.3.2.3 Fungal RFLP Banding patterns obtained with five restriction enzymes (Alu\, Cfo\, EcoRI,

Haelll and /-/pall) of filamentous fungal ITS (Figure 4.6) were used to generate a distance dendrogram (Figure 4.9). The numbers of bands for each enzyme ranged from one to three. The dendrogram showed a strong relationship with morphological features since similar morphotypes generally clustered together.

The most common submorphotype, black/black (bl/bl), was found to cluster together in a group divided into three branches. One of the branches comprised

155 black/black fungal isolated from both types of leaves, while the other two groups only had isolates from either yellow or green leaves. There were some isolates in the black/black cluster that were not classified as black/black, but instead were white/brown (w/br). The black/black submorphotype, as with many other submorphotypes, started out as white, and then after up to two weeks, developed a black appearance. Possibly the three isolates described as white were black with a maturity period longer than the rest of the isolates. Another characteristic found in the bl/bl submorphotype, was that all the isolates obtained from yellow leaves were separated according to the azoxystrobin treatment.

The white/red (w/r), white/beige (w/b) and beige/beige (b/b) submorphotype clustered independently, having in most cases 100% similarity within each cluster

(Figure 4.9). Both of the green submorphotypes described (green/black and green/brown) were grouped in a main cluster, which was divided into many branches with bootstrap values over 50. The same occurred with the white/pink

(w/p) submorphotype, which clustered among highly diverse RFLP banding patterns. Finally the uncategorized isolates did not share banding patterns of the

12 major morphotypes and had independent branches.

4.3.3 Sequencing

4.3.3.1 Bacteria The top matches of bacterial 16S sequences resulted in the tentative identification of 25 genera with 40 different species (Tables 4.3 and 4.4). The most frequent genera were Agrobacterium, Curtobacterium, Pedobacter, and

Sphingomonas representing 14.8, 14.8, 9.0 and 9.0% of total number of

156 sequences analyzed. The bacterial morphotypes comprised many species obtained from sequences comparisons with databases. However, some species share the same submorphotype in most instances. For example, all Pedobacter spp. were described as the pink submorphotype (Table 4.3). For Enterobacter sakazaki, it was white slimy, Dyella yeojuensis was yellow, Subtercola pratensis was yellow and most Sphingomonas spp. Were orange morphotypes. In certain cases, such as Pseudomonas, they were always categorized as white or white slimy because of the difficulty in determining the subtle differences between colonies. The isolates which matched Curtobacterium flaccumfaciens were the most variable with respect to morphotype, since they had representatives in all the submorphotype classifications in this study.

Many of the bacterial top matches (23 species) were found on both yellow and green leaves. However, eight species were only found on green leaves, and

11 species belonged to only yellow leaves (Table 4.5). These species that were associated with a specific type of leaf were found only on a few occasions, and they only accounted for 14.8% of total isolates on green leaves and 20.5% of total isolates on yellow leaves.

Leaves obtained from plots treated with a single application of azoxystrobin had 12 bacterial species that were not found on untreated samples. This represented 14.5% of the total number of sequences analyzed. The untreated leaves had 16 different species that could not be found on treated leaves (Table

4.6), which comprised 27.0% of the total number of sequences analyzed.

157 Green leaves treated with azoxystrobin had five species that were not found on untreated green leaves, but two species were found exclusively of treated green leaves and not on untreated green leaves (Table 4.5). Treated yellow leaves also had different bacterial species than isolates from untreated yellow leaves (Table 4.6). The following species were unique for treated yellow leaves:

Brevundimonas nasdae, Curtobacterium pusillim, Microbacterium imperiale,

Pedobacter koreensis and Stenotrophomonas maltophilia. Species only found on untreated yellow leaves included: Bacillus megaterium, Chryseobacterium proteolyticum, Pedobacter roseus and Zoogloea ramigera. These species only accounted for less than 10% of the total number of sequences obtained from yellow leaves.

4.3.3.2 Yeast The sequences of the yeast ITS had 23 different species matches comprising 10 genera (Tables 4.7 and 4.8). The most common genera were

Cryptococcus , Cystofilobasidium and Rhodotorula comprising 22.7, 13.6 and

38.1% respectively. The 25% remaining sequences contained Caulerpa,

Dioszegia, Leucosporidium, Rhodosporidium, Sakaguchia and Ustilago.

The yeast submorphotypes were in most cases consistent with the species obtained from the sequence comparison with databases, but each submorphotype had more than one species (Tables 4.7 and 4.8). The pink submorphotype included almost all Rhodotorula graminis (13 of 14) and

Rhodosporidium diobovatum (3 of 3). All Dioszegia crocea were always described as the orange submorphotype, and all Cystofilobasidium macerans

158 belonged to the salmon submorphotype, whereas Sakaguchia dacryoidea isolates were found in both the orange and salmon submorphotypes. The most varied was the white morphotype with more than 15 species. The most common white yeasts belonged to the genera, Cryptococcus, Rhodotorula and Ustilago.

The white slimy submorphotype had all species in common with the white submorphotypes, except for Leucosporidium golubevii, which was always in the white slimy group.

The number of yeast species found in green and yellow leaves was almost the same (23 from yellow leaves and 22 from green leaves). In both types of leaves, the genera were essentially the same with some variation in species

(Table 4.5). For example, of the nine species of Cryptococcus, four were only obtained from yellow leaves, while green leaves had sequences that matched five Cryptococcus species and Cryptococcus statzellii (Tables 4.5 and 4.8).

The diversity of species obtained from untreated green leaves was higher than on treated green leaves. Four genera and 10 species were identified in treated samples, whereas untreated samples had sequences matching 10 genera and 14 species (Table 4.8). The same number of species (13) was found on treated and untreated yellow leaves. Dioszegia, Sakaguchia and

Sporobolomyces were unique genera found only on untreated plots (Table 4.6).

4.3.3.3 Filamentous fungi Filamentous fungal epiphytes were the second most variable microbial group with 31 species among 21 genera (Tables 4.9 and 4.10). The most abundant species were Microdochium bolleyi (30.0%), Cladosporium

159 cladosporioides (11.0%), Mucor hiemalis (8.8%) and Epicoccum nigrum (8.7%).

Most isolates of these species were consistently correlated with a morphotype.

Mucor hiemalis isolates were in most cases found as the white fluffy

submorphotype. Cladosporium spp. were always green and E. nigrum were

always white/red. In the case of M. bolleyi, most isolates were described as

black/black submorphotype, but some had a white/beige appearance (Table 4.9).

Both types of leaves had a similar number of filamentous fungal species and genera. The four most common genera which are Cladosporium, Epicoccum,

Microdochium and Mucor were found on the surface of both types of leaves

(Tables 4.10). Some genera were only found on yellow leaves: Alternaria,

Chaenothecopsis, Didymella, Phialemonium, Plectosphaerella and Pyrenochaeta

(Table 4.5). Green leaves only had four exclusive filamentous fungal genera:

Apiosporina, Ascochyta, Epacris and Stachybotrys (Table 4.5).

The total number of species obtained from green and yellow leaves from untreated plots was 12, while only six species were exclusive from treated plots

(Table 4.6). As with the other microbial groups, more diversity was found on green leaves collected from untreated plots than green treated plots. On treated green leaves, 18 species matched the sequences of isolates obtained from green untreated leaf, while only 10 species matched sequences (Tables 4.6 and 4.10).

Untreated yellow leaves had 16 species, while untreated yellow leaves had 14 species (Table 4.10).

160 4.4 Discussion The analyses of phyllosphere microbes with molecular techniques gave a deeper insight into the microbial community, whose numbers and fluctuations were assessed in Chapter 3. These analyses showed the diversity of genera in relation to each other. RFLP analyses showed that the use of morphological characteristics is useful to distinguish groups of culturable fungi, but not for culturable bacteria colonies because of the diversity within each morphotype.

The sequencing of PCR products provided a deeper analysis of the data obtained by RFLP. Species identified by sequences analysis in part confirmed what was found with the RFLP, and directly linked some morphotypes with species.

The RFLP analysis was time consuming since at least four enzymes had to be used to obtain a minimum level of resolution in the dendrograms. For filamentous fungi and to some extent for yeasts, amplification and digestion of

ITS was easy and repeatable, yielding dendrograms with clusters associated with specific diverse morphotypes. This was not the case with the bacterial PCR-

RFLP since the gels were harder to read, and the dendrograms had very low bootstrap values.

Sequencing, however, was successful with nearly all isolates tested in all three microbial groups, and provided identification when the matches were available in GenBank. The sequencing approach produced more relevant data and was less time consuming than the RFLP analyses, but much more expensive ($20 per sequence vs. approximately $2 per RFLP).

161 Most species revealed by the comparison of 16S and ITS sequences with genomic databases have been commonly reported on other leaves. The most commonly found bacterial sequences matched three genera, Sphingomonas,

Curtobacterium and Pedobacter, comprising 40% of all isolates sequenced. The two first genera have been reported to form part of the microbial community of leaves of 11 different families of plants (Enya et al., 2006; Inacio et al., 2002; Kim et al., 1998). Pedobacter and Agrobacterium were more closely related to soil and roots of plants, but since A. stolonifera leaves are very close to the soil, an invasion of these bacteria onto living leaves is a likely scenario.

The enterobacteriaceae, Enterobacter sakazakii and Escherichia coli, were found on the leaf surfaces indicating a possible contamination by polluted water.

However, the occurrence of these bacteria on plants is not uncommon, and leaves of several pasture grasses and other plants have been found to harbor enterobacteriaceae (Behrendt et al., 1997; Yang et al., 2001).

The number of yeast species obtained (23 species) is higher than what has been found on other plant phyllospheres. For instance, only 17 yeast species were identified from leaf surfaces of six deciduous and coniferous species

(Slaavikova and Vadkertiova, 2003). For leaves of five Mediterranean plants, a total 16 species were identified (Inacio et al., 2005). These comparisons, however, are with leaves from very different types of ecosystems, and might not be relevant to this study. In the only previously encountered research on yeast biodiversity of two turfgrass species (A. stolonifera and Festuca arundinacea), only 15 yeast species were observed based on ITS sequence comparisons with

162 data bases (Allen et al., 2004b). The reason for the higher number obtained in

this study (23 vs. 15 species) may be related to the isolation from two types of

leaves (green and yellow) and the multiple isolations during the growing season.

Allen et al. (2004b) also used ITS sequence comparisons, but isolated yeasts

only in summer and fall.

Cryptococcus, Cystofilobasidium and Rhodotorula were the most common genera of yeasts. These genera have been found to be associated with leaf surfaces (Allen et al., 2004b; Buck and Burpee, 2002; Inacio et al., 2005).

Species of Cryptococcus were usually white, Cystofilobasidium were salmon and

Rhodotorula were pink. These genera have been previously found on other leaves including A. stolonifera (Allen et al., 2004b), and were described with the same colors observed in this study. However, the division of the salmon morphotype has never been used before even though their morphological characteristic is easy to distinguish among pink yeasts. The importance of this separation is crucial, since each morphotype was composed of two different genera.

Some ITS sequences obtained from yeasts did not match yeasts. One sequence matched the green algae Caulerpa webbiana, which looked the same as the pink yeasts. Although turfgrass environments can be infested with terrestrial algae, C. webbiana unlikely to survive in this environment, since it is a maritime algae (Silva, 2002) and this might be an incorrect GenBank record. A group of white yeast isolates (11) was found to be the smut fungus Ustilago

163 duriaeana. Smut fungi are microscopic basidiomycetes that have a slimy yeast­ like appearance on artificial media (McKenzie and Vanky, 2001).

The most variable yeast morphotype was white, which was composed of 11 species excluding Ustilago duriaeana since it also has filamentous fungus phase.

This large number of species of white yeasts was found to belong to only two genera, Cryptococcus and Rhodotorula, making the white morphotype highly homogenous at the genus level. White yeasts were also the most abundant morphotype found in the microbial analyses of leaves of A. stolonifera (Chapter

3). Buck and Burpee (2002) found by diluting plating quantification that the white yeast isolates from leaf surfaces of A. stolonifera comprised more 95% of all isolates, and all are presumed to be Cryptococcus spp. In the current study, many isolates of Rhodotorula, especially R. ingeniosa, were white, a result that is in contrast with the assumption that Rhodotorula species are always pink (Allen et al., 2004b; Buck and Burpee, 2002). Interestingly, white yeasts of the genus

Pseudozyma did not match any amplified sequence, even though it was one of the most common yeasts isolated from A. stolonifera leaves by Allen et al.

(2004b) who identified the ITS1 sequence by BLAST searches. ITS from white beige and white slimy colonies also matched the same species found in the white morphotype indicating that this morphotypical grouping was not accurate to species. These results demonstrate that morphological classification of yeasts based on color and texture without further verification is risky because many white-yeasts share the same cultural morphotype.

164 Diversity of filamentous fungi was high with 31 different species from 135

sequences, but over 50% matched only four species, which implies a low level of

biodiversity evenness. This preliminary result shows that the diversity of the

filamentous fungal community is comparatively lower than the reported for fungal

communities from other ecosystems, especially tropical and temperate habitats

(Alhubaishi and Abdel-Kader, 1991; Breeze and Dix, 1981; Heredia 1993).

The most common species was Microdochium bolleyi, a reported plant

pathogen of roots of cereals and grasses including A. stolonifera, but which has

also used as a biological control of take-all patch (Chng et al., 2004; Kirk and

Deacon, 1986). Microdochium bolleyi in most cases had the black/black morphotype, but it was not counted as a black filamentous fungus since at the time of colony enumeration as done in Chapter 3, it was still white or beige.

The black morphotypes enumerated in Chapter 3 corresponded to the green/black and the green/brown submorphotypes of this chapter. Sequences from the green/black isolates usually matched two species of Cladosporium.

These fungi are associated with dead plant tissue and have the characteristic colors of green, brown or black. Cladosporium spores are normally linked to allergic reactions and respiratory diseases (Garret et al., 1998), but

Cladosporium tenuissimum also has mycoparasitic activity towards certain fungal species, such as Uromyces spp. (Assante et al., 2004).

The sequences obtained from the submorphotypes white fluffy in most cases matched Mucor hiemalis which was the only species of the Zygomycota found. This species is associated with soil and plant litter (Lewis, 1994), and its

165 pathogenicity to plants is not well defined. The fourth most common sequence

match was Epicoccum nigrum, which is associated with soil and plant litter

(Franckland, 1998). It has also been tested as biological control of many plant

diseases, such as the fruit disease brown rot (caused by Monilinia spp.) (Larena

et al., 2005) and twig blight (Monilinia laxa) (Pascual et al., 1996). Most other

morphotypes did not derive from single species. For example, of 10 white/white

isolates, eight different species were revealed by sequence analysis.

The large number of filamentous fungi found in this study from A. stolonifera

leaves that are normally near that the soil suggests the soil is the source of filamentous fungi colonizing the continuously growing leaves. However, isolates such as Nectria cinnabarina, which is a pathogen of trees could be incidental on the grass leaf surface.

Endophytic filamentous fungi have not been found in leaves of A. stolonifera (Sun and Brede, 1997). However, on one occasion, a red filamentous fungus which matched the E. nigrum DNA sequence was isolated with the endophyte protocol. Interestingly, E. nigrum has been reported as a common endophytic fungus of Pinus spp. (Yu and LiangDong, 2004).

The use of molecular techniques complemented the community analyses of this study by revealing the possible microbial species that form the complex phyllosphere environment. The diversity revealed is typical of that found with leaves of others plants, but in this study many more species were encountered on foliage which are normally associated with soil. Identification of the microbial organisms by morphological characteristics (color morphotypes) was useful to

166 obtain an overview of the species present on the leaf surfaces of turfgrasses, particularly for filamentous fungi, but molecular analyses were required to resolve species of bacteria and to a lesser extent, yeasts present in the phyllosphere of A stolonifera.

167 Table 4.1. Numbers of isolates obtained from leaves collected from the Agrostis stolonifera native sand fairway of 2005-2006, Poa pratensis fringe of 2005-2006 and A. stolonifera pathology green of 2006-2007.

Collection Plant Trial Fungi Yeast Bacteria Total Mar-2006 A. stolonifera Fairway 54 190 80 324 Mar-2006 P. pratensis Fringe 26 82 108 May-2006 A. stolonifera Fairway 12 30 15 57 May-2006 P. pratensis Fringe 29 46 24 99 Jun-2006 A. stolonifera Fairway 36 34 70 Jul-2006 A. stolonifera Fairway 14 14 Jul-2006 P. pratensis Fringe 38 38 Aug-2006 A. stolonifera Fairway 42 98 86 226 Aug-2006 P. pratensis Fringe 83 108 198 389 Sep-2006 A. stolonifera Fairway 94 135 131 360 Sep-2006 P. pratensis Fringe 57 99 115 271 Nov-2006 A. stolonifera Green 163 190 167 520 Mar-2007 A. stolonifera Green 180 199 182 561 May-2007 A. stolonifera Green 181 192 193 566 Jul-2007 A. stolonifera Green 182 178 199 559 Total 1,139 1,633 1,390 4,162

168 Table 4.2. Number of isolates per morphotypes obtained from leaves collected from the Agrostis stolonifera native sand fairway of 2005-2006, Poa pratensis fringe of 2005-2006 and A. stolonifera pathology green of 2006-2007. RFLP and sequencing refer to the number of isolates used in these analyses and the number in parentheses refers to ribotypes and species.

Chapter 4 Chapter 3 Sub Organism Morphotype morphotype Isolates RFLP Sequence Fungi Total Total 1,139 61 (29) 141 (31) Black Green/black 189 11 (5) 25(6) Black Green/brown 87 3(3) 12(8) Red White/red 49 5(2) 10(3) White Beige/beige 18 4(1) 3(3) White Black/black 209 19(4) 26(4) White Grey 12 0 5(4) White Orange 114 2(1) 9(3) White Others 51 3(3) 6(4) White White/beige 204 3d) 14(3) White White/brown 50 4(2) 5(2) White White fluffy 32 0 13(4) White White/pink 15 4(4) 3(2) White White/white 109 3(3) 10(8) Yeast Total Total 1,633 126(48) 110(23) Orange Orange 54 10(9) 8(4) Pink Pink 341 26(4) 18(5) Salmon Salmon 167 23(7) 20(4) White White 632 34(19) 42(12) White White beige 199 6(2) 9(4) White White slimy 240 27(6) 13(5) Bacteria Total Total 1,390 39 (34) 122(39) Orange Orange 223 8(7) 15(7) Orange Orange slimy 205 4(4) 14(9) Pink Pink 154 6(4) 23(10) Pink Pink slimy 9 4(4) 0 White White 247 4(4) 11 (8) White White slimy 131 3(3) 17(7) Yellow Yellow 298 5(5) 26(12) Yellow Yellow slimy 132 5(4) 16(10)

169 Table 4.3. Bacterial morphotypes and highest matching species with analysis of 16S sequences. Morphotype Species Number Orange Total 15 Agrobacterium tumefaciens 1 Brevundimonas nasdae 1 Chryseobacterium proteolyticum 1 Janthinobacterium sp 2 Microbacterium imperiale 1 Rhodococcus fascians 4 Sphingomonas sp 5 Orange slimy Total 14 Chryseobacterium aurantiacum 1 Chryseobacterium daeguense 1 Curtobacterium flaccumfaciens 2 Curtobacterium herbarum 2 Escherichia coli 2 Microbacterium oleivorans 2 Pedobacter wanjuense 1 Shigella flexneri 1 Sphingomonas sp 2 Pink Total 21 Curtobacterium flaccumfaciens 2 Pedobacter aquatilis 2 Pedobacter aurantiacus 4 Pedobacter koreensis 1 Pedobacter roseus 3 Pedobacter suwonensis 3 Pedobacter wanjuense 3 Shigella dysenteriae 1 Sphingomonas sp 1 Xanthomonas campestris 1 White Total 11 Agrobacterium tumefaciens 2 Chryseobacterium daeguense 1 Escherichia coli 1 Pseudomonas abietaniphila 1 Pseudomonas corrugata 1 Pseudomonas syringae 3 Shigella dysenteriae 1 Sphingomonas sp 1 White slimy Enterobacter sakazakii

170 Table 4.3. cont. Morphotype Species Number White slimy Total 16 Agrobacterium tumefaciens 7 Bacillus megaterium 1 Escherichia coli 1 Pseudomonas putida 1 Shigella flexneri 1 Xanthomonas translucens 1 Yellow Total 25 Agreia bicolorata 4 Chitinophaga ginsengisoli 1 Chryseobacterium aurantiacum 1 Curtobacterium flaccumfaciens 4 Dyella yeojuensis 5 Escherichia coli 1 Flavobacterium sp. 1 Pedobacter roseus 1 Sphingomonas sp. 2 Stenotrophomonas maltophilia 1 Subtercola pratensis 3 Zoogloea ramigera 1 Yellow slimy Total 16 Agreia bicolorata 1 Agrobacterium tumefaciens 1 Arthrobacter ilicis 1 Chryseobacterium aurantiacum 1 Curtobacterium flaccumfaciens 5 Curtobacterium pusillum 2 Dyadobacter hamtensis 1 Flavobactehum sp. 2 Janthinobacterium agaricidamnosum 1 Sphingomonas yabuuchiae 1

171 Table 4.4. Bacterial species, submorphotype, and number of isolates from azoxystrobin-treated and untreated green or yellow leaves of Agrostis stolonifera as revealed by sequence analysis of DNA amplified with the primer pair F0/PRUN518 Sub Green leaves Yellow leaves Species morphotype Untreat Treat Untreat Treat Total Agreia bicolorata Yellow 2 0 2 1 5 Agrobacterium tumefaciens White slimy 2 2 5 2 11 Arthrobacter ilicis Yellow slimy 1 0 0 0 1 Bacillus megaterium White slimy 0 0 1 0 1 bacterium Varied 1 0 2 0 3 Brevundimonas nasdae Orange 0 0 0 1 1 Chitinophaga ginsengisoli Yellow 0 1 0 0 1 Chryseobacterium spp. Varied 2 0 3 1 6 Chryseobacterium Varied 2 1 aurantiacum 0 0 3 Chryseobacterium Varied 2 0 0 0 2 daeguense Chryseobacterium Orange 0 0 1 0 1 proteolyticum Curtobacterium spp. Varied 4 5 1 8 18 Curtobacterium Varied 2 5 1 5 13 flaccumfaciens Curtobacterium herbarum Orange slimy 1 0 0 1 2 Curtobacterium pusillum Yellow slimy 0 0 0 2 2 Dyadobacter hamtensis Yellow slimy 1 0 0 0 1 Dyella yeojuensis Yellow 2 0 1 2 5 Enterobacter sakazakii White slimy 0 1 0 3 4 Escherichia coli Varied 4 0 1 0 5 Flavobacterium sp. Yellow slimy 1 1 1 0 3 Janthinobacterium sp. Varied 2 0 1 0 3 Janthinobacterium Yellow slimy 1 0 0 0 1 agaricidamnosum Microbacterium spp. Varied 0 1 0 2 3 Microbacterium imperiale Orange 0 0 0 1 1 Microbacterium oleivorans Orange slimy 0 1 0 1 2 Pedobacter spp. Pink 2 2 12 2 18 Pedobacter aquatilis Pink 1 0 1 0 2 Pedobacter aurantiacus Pink 0 1 2 1 4 Pedobacter koreensis Pink 0 0 0 1 1 Pedobacter roseus Pink 0 0 4 0 4 Pedobacter suwonensis Pink 0 1 2 0 3 Pedobacter wanjuense Pink 1 0 3 0 4 Pseudomonas spp. Varied 2 2 2 1 7

172 Table 4.4. cont. Sub Green leaves Yellow leaves Species morphotype Untreat Treat Untreat Treat Total Pseudomonas abietaniphila White 1 0 0 0 1 Pseudomonas corrugata White 0 0 0 1 1 Pseudomonas putida White slimy 0 1 0 0 1 Pseudomonas syringae White 1 1 1 0 3 Rhodococcus fascians Orange 1 0 3 0 4 Shigella spp. Varied 0 1 0 3 4 Shigella dysenteriae Varied 0 1 0 1 2 Shigella flexneri Varied 0 0 0 2 2 Sphingomonas spp. Varied 4 4 1 2 11 Sphingomonas yabuuchiae Yellow slimy 0 0 1 0 1 Stenotrophomonas maltophilia Yellow 0 0 0 1 1 Subtercola pratensis Yellow 0 1 0 2 3 Xanthomonas spp. Varied 2 0 0 0 2 Xanthomonas campestris Pink 1 0 0 0 1 Xanthomonas translucens White slimy 1 0 0 0 1 Zoogloea ramigera Yellow 0 0 1 0 1 Total number of sequences 33 21 37 31 122 Total number of species 18 12 18 18

173 Table 4.5. List of microbial species found exclusively on each type of leaf. Isolates were obtained from green and yellow leaves collected from the Agrostis stolonifera pathology green either treated with azoxystrobin or untreated.

Leaf type Organism Green Yellow Bacteria Arthrobacter ilia's Bacillus megaterium Chitinophaga ginsengisoli Brevundimonas nasdae Dyadobacter hamtensis Chryseobacterium aurantiacum Janthinobacterium agaricidamnosum Chryseobacterium proteolyticum Pseudomonas abietaniphila Curtobacterium pusillum Pseudomonas putida Microbacterium imperiale Xanthomonas campestris Pedobacter koreensis Xanthomonas translucens Pedobacter roseus Sphingomonas yabuuchiae Stenotrophomonas maltophilia Zoogloea ramigera Yeast Caulerpa webbiana Cryptococcus dimennae Cryptococcus statzellii Cryptococcus festucosus Rhodotorula glutinis Cryptococcus nyarrowii Sporobolomyces roseus Cryptococcus tephrensis Rhodotorula vanillica Filamentous fungi Apiosporina morbosa Alternaria compacta Ascochyta skagwayensis Chaenothecopsis epithallina Epacris microphylla Cladosporium elatum Fusarium lateritium Didymella rabiei Fusarium proliferatum Fusarium tricinctum Gibberella fujikuroi Gibberella avenacea Phoma medicaginis Phialemonium dimorphosporum Phoma pomorum Phoma cava Stachybotrys parvispora Plectosphaerella cucumerina Pyrenochaeta lycopersici

174 Table 4.6. List of microbial species found exclusively on each type of leaf. Isolates were obtained from green and yellow leaves collected from the Agrostis stolonifera pathology green either treated with azoxystrobin or untreated. Leaf treatment Organism Untreated Treated Bacteria Arthrobacter ilicis Brevundimonas nasdae Bacillus megaterium Chitinophaga ginsengisoli Chryseobacterium daeguense Curtobacterium pusillum Chryseobacterium proteolyticum Microbacterium imperiale Dyadobacter hamtensis Microbacterium oleivorans Escherichia coli Pedobacter koreensis Janthinobacterium agaricidamnosum Pseudomonas corrugate Pedobacter aquatilis Pseudomonas putida Pedobacter roseus Shigella dysentehae Pedobacter wanjuense Shigella flexneri Pseudomonas abietaniphila Stenotrophomonas maltophilia Rhodococcus fascians Subtercola pratensis Sphingomonas yabuuchiae Xanthomonas campestris Xanthomonas translucens Zoogloea ramigera Yeast Caulerpa webbiana Cryptococcus dimennae Cryptococcus laurentii Cryptococcus festucosus Cryptococcus paraflavus Cryptococcus nyarrowii Cryptococcus statzellii Rhodotorula fragaria Cryptococcus tephrensis Rhodotorula glutinis Dioszegia crocea Rhodotorula vanillica Sakaguchia dacryoidea Sporobolomyces roseus Filamentous fungi Alternaria compacta Alternaria longipes Ampelomyces humuli Ascochyta skagwayensis Apiosporina morbosa Chaenothecopsis epithallina Cladosporium elatum Fusarium equiseti Didymella rabiei Plectosphaerella cucumerina Epacris microphylla Pyrenochaeta lycopersici Fusarium proliferatum Gibberella avenacea Gibberella fujikuroi Gibberella spp. Nectria cinnabarina Phialemonium dimorphosporum

175 Table 4.6. cont. Leaf treatment Organism Untreated Treated Filamentous fungi Phoma cava Phoma pomorum Stachybotrys parvispora Thelebolus stercoreus

176 Table 4.7. Yeast submorphotypes and highest matching species with analysis of ITS sequences. Morphotype Species Number Orange Total 8 Cryptococcus statzellii 1 Cystofilobasidium macerans 1 Dioszegia crocea 4 Sakaguchia dacryoidea 2 Pink Total 18 Caulerpa webbiana 1 Cryptococcus dimennae 1 Rhodosporidium diobovatum 3 Rhodotorula glutinis 1 Rhodotorula graminis 12 Salmon Total 20 Cystofilobasidium macerans 15 Rhodotorula graminis 1 Sakaguchia dacryoidea 3 Sporobolomyces roseus 1 White Total 39 Cryptococcus festucosus 1 Cryptococcus flavescens 1 Cryptococcus laurentii 2 Cryptococcus nyarrowii 1 Cryptococcus paraflavus 3 Cryptococcus tephrensis 2 Rhodotorula fragaria 2 Rhodotorula graminis 1 Rhodotorula hordea 2 Rhodotorula ingeniosa 14 Rhodotorula vanillica 1 Ustilago duriaeana 9 White beige Total 9 Cryptococcus laurentii 1 Cryptococcus victoiiae 4 Rhodotorula hordea 2 Ustilago duriaeana 2 White slimy Total 13 Cryptococcus flavescens 4 Leucospohdium golubevii 3 Rhodotorula fragaria 1 Rhodotorula hordea 3 Rhodotorula ingeniosa 2

177 Table 4.8. Yeast species, submorphotype, and number of isolates from azoxystrobin-treated and untreated green or yellow leaves of Agrostis stolonifera as revealed by sequence analysis of DNA amplified with the primer pair ITS1/ITS4.

Sub Green leaves Yellow leaves Species morphotype Untreat Treat Untreat Treat Total Caulerpa webbiana Pink 1 0 0 0 1 Cryptococcus spp. 8 5 8 4 25 Cryptococcus dimennae Pink 0 0 0 1 1 Cryptococcus festucosus White 0 0 0 1 1 Cryptococcus flavescens White slimy 1 2 2 0 5 Cryptococcus laurentii White 3 0 1 0 4 Cryptococcus nyarrowii White 0 0 0 1 1 Cryptococcus paraflavus White 0 1 2 0 3 Cryptococcus statzellii Orange 1 0 0 0 1 Cryptococcus tephrensis White 0 0 2 0 2 Cryptococcus victoriae White 3 2 1 1 7 Cystofilobasidium macerans Salmon 5 3 4 3 15 Dioszegia crocea Orange 3 0 1 0 4 Leucosporidium golubevii White slimy 1 0 0 2 3 Rhodosporidium diobovatum Pink 1 0 0 2 3 Rhodotorula spp. 5 13 15 9 42 Rhodotorula fragaria White slimy 0 1 0 2 3 Rhodotorula glutinis Pink 0 1 0 0 1 Rhodotorula graminis Pink 3 4 4 3 14 Rhodotorula hordea White slimy 0 2 4 2 8 Rhodotorula ingeniosa White 2 5 6 2 15 Rhodotorula vanillica White slimy 0 0 1 0 1 Sakaguchia dacryoidea Salmon 2 0 2 1 5 Sporobolomyces roseus Salmon 1 0 0 0 1 Ustilago duriaeana White 4 2 3 2 11 Total number of sequences 31 23 33 23 110 Total number of species 14 10 13 13

178 Table 4.9. Filamentous fungal morphotypes and highest matching species with analysis of ITS sequences. Morhpotype Species Number black/black Total 24 Cladosporium tenuissimum 2 Epicoccum nigrum 1 Microdochium bolleyi 20 Microdochium nivale 1 white fluffy Total 13 Fusarium tricinctum 1 Microdochium bolleyi 2 Microdochium nivale 1 Mucor hiemalis 9 green/black Total 25 Alternaria compacta 1 Apiosporina morbosa 1 Cladosporium cladospohoides 13 Cladosporium tenuissimum 6 Davidiella tassiana 3 Phoma cava 1 green/brown Total 12 Alternaria longipes 1 Cladosporium cladosporioides 2 Cladosporium tenuissimum 1 Epicoccum nigrum 3 Microdochium bolleyi 1 Phoma medicaginis 2 Stachybotrys parvispora 1 Ascochyta skagwayensis 1 grey Total 5 Alternaria compacta 1 Didymella rabiei 2 Epacris microphylla 1 Phoma pomorum 1 orange Total 9 Microdochium bolleyi 1 Mucor hiemalis 2 Thelebolus stercoreus 6 white/beige Total 14 Microdochium bolleyi 11 Phialemonium dimorphosporum 2 Thelebolus stercoreus 1

179 Table 4.9. cont. Morhpotype Species white/beige Total 14 Microdochium bolleyi 11 Phialemonium dimorphosporum 2 Thelebolus stercoreus 1 white/brown Total 5 Ampelomyces humuli 1 Microdochium bolleyi 4 white/orange Total 3 Fusarium equiseti 1 Fusarium lateritium 2 white/pink Total 3 Fusarium proliferatum 2 Gibberella fujikuroi 1 white/red Total 8 Epicoccum nigrum 6 Gibberella avenacea 1 Pyrenochaeta lycopersici 1 white/white Total 10 Chaenothecopsis epithallina 1 Fusarium equiseti 1 Fusarium lateritium 1 Microdochium bolleyi 1 Microdochium nivale 1 Mucor hiemalis 1 Nectria cinnabarina 3 Plectosphaerella cucumerina 1

180 Table 4.10. Filamentous fungal species, submorphotype, and number of isolates from azoxystrobin-treated and untreated green or yellow leaves of Agrostis stolonifera as revealed by sequence analysis of DNA amplified with the primer pairlTS1/ITS4. Sub Green leaves Yellow leaves Species morphotype Untreat Treat Untreat Treat Total Alternaria spp. Varied 10 2 14 Alternaria compacta green/black 0 0 2 0 2 Alternaria longipes green/brown 0 0 0 11 Ampelomyces humuli white/brown 10 10 2 Apiosporina morbosa green/black 10 0 0 1 Ascochyta skagwayensis white/brown 0 1 0 0 1 Chaenothecopsis epithallina white/red 0 0 0 11 Cladosporium spp. Varied 6 1 10 8 25 Cladosporium cladosporioides green/black 5 0 5 5 15 Cladosporium elatum other 0 0 10 1 Cladosporium tenuissimum green/black 1 1 5 3 10 Davidiella tassiana green/black 0 1 113 Didymella rabiei grey 0 0 10 1 Epacris microphylla grey 10 0 0 1 Epicoccum nigrum Varied 4 4 1 4 13 Fusarium spp. Varied 4 2 0 2 8 Fusarium equiseti Varied 0 1 0 12 Fusarium lateritium Varied 2 1 0 0 3 Fusarium proliferatum white/pink 2 0 0 0 2 Fusarium tricinctum white fluffy 0 0 0 11 Gibberella spp. Varied 10 10 2 Gibberella avenacea white/red 0 0 10 1 Gibberella fujikuroi white/pink 10 0 0 1 Microdochium spp. Varied 11 4 11 18 44 Microdochium bolleyi Varied 10 3 11 17 41 Microdochium nivale Varied 11 0 13 Mucor hiemalis white fluffy 2 4 2 4 12 Nectria cinnabarina white/white 10 2 0 3 Phialemonium dimorphosporum white/white 0 0 2 0 2 Phoma spp. Varied 2 3 2 0 7 Phoma cava green/black 0 0 10 1 Phoma medicaginis green/brown 12 0 0 3 Phoma pomorum grey 10 0 0 1 Plectosphaerella cucumerina white/white 0 0 0 11 Pyrenochaeta lycopersici white/red 0 0 0 11 Stachybotrys parvispora green/brown 10 0 0 1 Thelebolus stercoreus orange 3 0 13 7 Total number of sequences 39 20 38 44 141 Total number of species 18 10 16 14

181 white/white

Figure 4.1. Filamentous fungal submorphotypes. Isolates were obtained from Agrostis stolomfera leaf surfaces by washing and plating on PDA with antibiotics solates were plated separately on PDA plates and incubated a room temperature for up to 14 days. Top view is shown on left, and bottom view is shown on right, and the notation of color/color refers to top/bottom color

182 Figure 4.2. Yeast submorphotypes. Isolates were obtained from A. stolonifera leaf surfaces by washing and plating on PDA with antibiotics. Isolates were plated on PDA plates and incubated a room temperature for seven to nine days

183 Figure 4.3. Bacterial submorphotypes. Isolates were obtained from A. stolonifera leaf surfaces by washing and plating on PDA with benomyl. Isolates were plated on PDA plates and incubated a room temperature for five to seven days.

184 OlCDCDCIIC/lCDCDOOCDDDCDWroroCDCDCDDDCODDCD oooo^ooo-*o-*^oooo->.o^oo OOCOCO<0=*CO• Ol O --J CO N| 00 J Ji \| O O00 8 1000 bp «•> Ml •» 500 bp

250 bp

'«§§» 50 bp

1000 bp B 3 is

500 bp

^t 250 bp

50 bp

Figure 4.4. Bacterial 16S PCR products digested with restriction enzymes. The product were separately electrophoretically in 2% (w/v) agarose gels, with the weight marker O'GeneRuler™ DNA Ladder, in the first, middle and last lanes. A) digested with /-/pall and B) digested with Haelll. Gels were stained with ethidium bromide solution and observed with UV transilluminator, and the image captured on computer with a frame-grabber card. The images were reversed and enhanced with Paint Shop Pro.

185 -<-<--<-<-<-<-<-<--<<-< -u -ft. w U U W U U 0> U ^ ^ U U U » U U N M U N) Nl ->• O 00 CO Ol NJ -»• N> O -v| O -J A 00 Ol -v| -vl N 00 -» ID M N Ol vl > W U S -U O) CO (D U 05 <0 -ft.

A 1000 bp

500 bp Hi mi Hi

250 bp

50 bp

B 1000 bp J- « 500 bp

*« r| 250 bp «i . ';

50 bp

1000 bp B5

** Ml - - _— 500 bp

250 bp y^ "ft #

50 bp

Figure 4.5. Yeast ITS PCR products digested with restriction enzymes. The product were separately electrophoretically in 2% (w/v) agarose gels, with the weight marker O'GeneRuler™ DNA Ladder, in the first, middle and last lanes. A) digested with Hae\\\, B) digested with EcoRI, C) digested with Alu\. Gels were stained with ethidium bromide solution and observed with UV transilluminator, and the image captured on computer with a frame-grabber card. The images were reversed and enhanced with Paint Shop Pro.

186 Tl •n Tl T| Tl T| Tl Tl Tl Tl TiTi-ncoTi-n-n-nTi-n-nco O O w •Nl o03 oCO o-J o-vl o•vl ->l 5 o->l o0> o-g 2 CD 05 O 0> A CO CO -^ Ol co o o A O) CO Ol o> CO CO M N M M O CO -* -» *• 00 O) co Ol Ol O) M O o -ft o>

1500 bp

1000 bp

500 bp *«•>*• «•*•» •§•••• ** •*' 4MH

200 bp

B ;1 1000 bp *MI iMH |p§ ^g^ tfgpgtt jgm^^ >£^^ 500 bp

250 bp

mm 'if 50 bp

1000 bp

500 bp VI •"' 250 bp

50 bp

Figure 4.6. Filamentous fungal ITS PCR products digested with restriction enzymes. The product were separately electrophoretically in 2% (w/v) agarose gels, with the weight marker O'GeneRuler™ DNA Ladder, in the first, middle and last lanes. A) non-digested, B) digested with Hae\\\, C) digested with Alu\. Gels were stained with ethidium bromide solution and observed with UV transilluminator, and the image captured on computer with a frame-grabber card. The images were reversed and enhanced with Paint Shop Pro.

187 ps B0962 OB1032 41 WB1006 24 o B0899 w B0967 16 53 y B0961 21 16 39 yB1012 psB1034] Pink 30! ps B0999 I slimy ps B0938 J p B1035 70 p B1011 57 p B0965 46 • Pink p B0926 64 p B0960 pB1031 ws B0898 60 ws B0946 ys B0904 y B0935 OB0917 53 y B0995 32 OSB1000 53 15 471 os B1009 os B0996 • Orange slimy 821 os B0997 OB1005 0 B0908 23 51 OB0911 • Orange 41 0 B0914 701 OB1007 wB0910 17 ys B0897 161 ys B0892 40 ys B0895 Yellow y B0941 slimy 38 ys B0904 WB0918 0.1 36 ws B0898 Figure 4.7. Distance dendrogram of RFLP binary analyses of bacterial 16S PCR products. Banding patterns were obtained using four restriction enzymes separately: Alul, Cfol, Haelll and Hpall, and patterns were recorded in a binary format and analyzed with Windist and Phylip.

188 Salmon '(23/23)

5T1554 SI White beige

White slimy (20/21)

v White (13/18)

Figure 4.8. Distance dendrogram of RFLP binary analyses of yeast ITS PCR products. Banding patterns were obtained using four restriction enzymes separately: Alu\, EcoRI, Haelll and /-/pall, and patterns were recorded in a binary format and analyzed with Windist and Phylip.

189 GC0697w/w GC0638? 49 871 GC0641w/p] 84 27 eSS? White/Pink 64L GC0713w/p YC0831b/b] GT0746b/bL . „ . GC0660b/bfBeige/Beige GC0718b/bJ YC0760w/br1| 38 YC0792bl/bl A YC0766bl/bl YC0745bl/bl Untreated GC0726w/br> GC0730bl/bl leaves 41 GC0712bl/bl GC0675bl/bl GC0678bl/bl YT0842w/br 35 YT0873bl/bl YT0849bl/bl Yellow Black/Black YT0848bl/bl YT0847bl/bi treated YT0841bl/bl leaves YT0845bl/bl YT0871bl/bl GT0676bl/bl] GC0666bl/bllGreen

GT0664bl/bl[|pawpc GC0665bl/blJleaVeS 10 GC0682bl/bl 71L GC0681? J GC0652w/br GC0708w/r 271 YC0802w/r 89 GC0667w/i4 White/red GT0647w/r 30 10 GC0662w/r 101 YC0796w/w' YT0846w/b] 38i YC0750w/b I white/beige 18 YT0838w/bj YC0777? 54L YT0844g/br-| YT0870g/br Green/brown 82L YT0887g/br YC0803g/bl YC0786g/bl YC0774g/bl GC0661g/bl 81 YC0758g/bl I YC0763g/bl Green/black GC0677g/bl 81 1 GT0714g/bl • YC0894g/bl • YC0888g/bl 30 591 51L •YC0891g/bU iJJSS&Omnge • GC0703w/w • YT0869? 0.1 Figure 4.9. Distance dendrogram of RFLP binary analyses of filamentous fungal ITS PCR products. Banding patterns were obtained using four restriction enzymes separately: Alu\, Cfo\, EcoRI, f/aelll and /-/pall, and patterns were recorded in a binary format and analyzed with Windist and Phylip.

190 CHAPTER 5. GENERAL DISCUSSION An single application of azoxystrobin in the late fall was found to have long lasting residual effects on the microbial community of leaves of A. stolonifera and

P. pratensis and on the turfgrass disease dollar spot in the following growing season six to nine months later. Prior to this study, there were only general casual observations that dollar spot disease could be enhanced by azoxystrobin.

Secondly, the enhancement was found to occur for the first time in Poa pratensis.

Moreover, alteration of the microbial community of young (green) and senescent

(yellow) leaves of two turfgrass species caused by a single azoxystrobin application was documented. The pesticide not only changed the overall epiphytic microbial count, but also altered the species composition of the community reducing biodiversity in most cases, The side effects caused by azoxystrobin occurred after more than six months, which may be the longest lasting case of disease enhancement reported in the turfgrass disease literature.

Finally, this is one of the few reports on seasonal fluctuations of the turfgrass phyllosphere community throughout the growing season with and without fungicide treatment.

To examine the effect of azoxystrobin on dollar spot disease the incidence of dollar spot and the overall level of plant injury were evaluated in trials over two years. Replicated plots of A. stolonifera or P. pratensis received a single application of azoxystrobin in late fall. During the following 10 months, samples of green and yellow leaves were collected from these plots to assess phyllosphere populations of culturable filamentous fungi, yeasts and bacteria by

191 dilution plating techniques. The identification of selected isolates was further investigated using molecular techniques.

Among its many uses, azoxystrobin is registered as a turfgrass fungicide to control diseases, such as brown patch (caused by the basidiomycetous fungus

Rhizoctonia solani), pink snow mold (caused by the ascomycetous fungus

Microdochium nivale) and Pythium blight (caused by the stramenopile genus

Pythium). Although azoxystrobin has a very broad spectrum of activity, one of the most common turfgrass diseases of turfgrass, dollar spot, caused by the ascomycetous fungus Sclerotinia homoeocarpa, is not controlled or even prevented by treatments with azoxystrobin (Mann, 2004). Moreover, in recent years, dollar spot disease has been reported to be enhanced very soon after applications of azoxystrobin (Demoeden, 2000; Schumann, 2003).

The current study revealed that an application of azoxystrobin in the fall could enhance dollar spot disease incidence six to nine months later during the following growing season. This effect was found on A. stolonifera in three separate trials in 2006 and 2007, and on P. pratensis in 2006. The levels of enhancement were higher for A. stolonifera (up to 500%) than for P. pratensis

(up to 75%). This confirms previous observations and casual reports of dollar spot disease enhancement in Agrostis spp, by azoxystrobin (Cook and Hsiang,

2002 and 2004; Demoeden, 2000; Hsiang and Cook, 2006; Schumann, 2003;

Vincelli, 20076), and this is the first report for P. pratensis.

In previous studies of disease enhancement mediated by fungicides, increased disease incidence was observed after one to two months of multiple

192 pesticide applications (Gleason et al., 1996; Grogan and Scott, 1997; Schuman

etal., 1996; Smiley, 1981).

The dollar spot enhancement found in this investigation occurred as soon

as dollar spot appeared annually in June until the end of the growing season (six

to nine months after a single application of azoxystrobin), which makes this study

one of the few to show such a long lasting effect of a fungicide on turfgrass

environments. Based on previous literature documenting the half-life of the

chemical in similar environments, these results also suggest that azoxystrobin

treatment can modify microbial environments long after the pesticide has been

dissipated.

Disease enhancement mediated by pesticides has been commonly reported

in turfgrasses (Couch, 1995), but the causes for this phenomenon are largely

unknown. The two main explanations suggested are physiological alteration of the host plant or perturbation of microbial populations on and in the plant (Couch,

1995; Vincelli, 2007). Direct evidence for either of these two hypothetical causes is lacking. Even though pesticides are known to induce such effects in plant ecosystems, (Andrews and Kenerley, 1979; Bertelsen et al., 2001; Haile et al.,

2000; Slavikova and Vadkertiova, 2003), they have not demonstrated to be correlated with disease enhancement. This investigation sought to relate the alteration of foliar microbial populations with the increase of dollar spot disease triggered by a single azoxystrobin treatment. Since the disease enhancement occurred after azoxystrobin was predicted to be largely degraded or leached away, and because directly treated plant tissues would have been removed by

193 frequent mowing, direct alteration of plant physiology was not considered

plausible and thus was not addressed in this study.

After snowmelt, an application of azoxystrobin the previous fall significantly

reduced most microbial populations on yellow leaves, when yellow leaves greatly

outnumbered green leaves. By the summer when A. stolonifera green leaves

greatly outnumbered yellow leaves, treated green leaves had the most noticeable

reduction of microbial populations, This suggests that overall microbial

populations counts in the spring are best reflected by samples from yellow leaves, whereas microbial population counts in the summer are best reflected by

samples from green leaves. . Therefore, the overall effect of azoxystrobin was an overall reduction in microbial populations on foliage of P. pratensis and A. stolonifera.. These reduced microbial populations on green leaves during the summer were coincident with the increased dollar spot levels during the summer.

The enhancement of dollar spot disease caused by azoxystrobin may be related to the reduction of some groups of microbial phyllosphere inhabitants, as found in this study. This has been suggested as an explanation for cases of turfgrass disease enhancement or resurgence after fungicide treatments (Couch and Joyner 1975; Dernoeden, 2000; Schumann, 2003; Warren et al., 1976). This hypothesis is based on the effects that saprophytes may exert against plant pathogens. Suppression of plant disease mediated by competition of resources or antibiosis and parasitism caused by leaf epiphytes is well documented

(Belanger and Avis, 2002).

194 Based on previous research, yeasts probably inhibit filamentous fungal

pathogens. Studies have shown that naturally occurring yeast microflora on rice

leaves effectively suppressed conidial germination or hyphal growth of the leaf

pathogen Magnaporthe grisea (Kawamata et al., 2004). Yeasts of the genera

Candida, Cryptococcus, Debaryomyces and Rhodotorula have been used as

potential biological controls against the plant pathogens Botrytis cinerea,

Monilinia fructicola, and Penicillium spp. (Chand-Goyal and Spotts 1996; Dik and

Elad 1999; Droby et al. 1989; Janisiewicz and Korsten 2002). Pseudozyma

flocculosa, is an example of a yeast that produces antifungal components which

damage the membrane of several sensitive filamentous fungi (Benyagoub et al.,

1996).

In turfgrass, in vitro studies have shown that many yeasts isolated from

leaves are attached to conidia and hyphae of turfgrass pathogens, including

Rhizoctonia solani and Sclerotinia homoeocarpa (Allen et al., 2004a) which

presumably inhibits pathogen growth. Allen et al. (2004b) found higher number of

pink yeasts on leaves infected with S. homoeocarpa, and suggested that these

yeasts were favored by the plant nutrient leaching of rather than inhibiting the

plant pathogen. Interestingly, A. stolonifera green and yellow leaves from treated

plots harbored more pink yeasts in the summer when dollar spot was enhanced

agreeing with Allen et al. (2004b) who found a higher number of pink yeasts on turfgrass leaves infected with S. homoeocarpa.

The dilution plating experiments showed seasonal fluctuations of microbial communities on the leaves of A. stolonifera and P. pratensis. This differed from

195 the steady augmentation of microbial populations (by microbial growth and

colonization) from spring to fall that has been reported for other phyllosphere

microbial communities (Inacio et al., 2005; Lindow and Brandl, 2003; Osono and

Mori, 2004; Santamaria and Bayman 2005). The most common seasonal pattern

of the filamentous fungal and bacterial communities in untreated plots was low

levels in spring, increase in summer and decrease in fall. The difference is,

perhaps, due to the special characteristics of turfgrass plants. Previous studies of

seasonal fluctuations of phyllosphere microbial populations have been performed on trees (evergreen, deciduous and coniferous), which have leaves that last longer than grass leaves. Therefore, microbial communities have more time to

increase and colonize the leaf surface than microbes from grasses.

Additionally, the particular patterns of microbial phyllosphere organisms on turfgrasses may be related to the constant mowing and the direct contact of leaves with soil and decaying leaves when the grass is stepped upon, being mowed or subjected to other cultural operations. The mowing and removal of leaf tips from putting greens and the constant growth of new leaves and tissues provides new habitats and colonization sites, and repeatedly requires re- colonization, while foliar tissues of trees, broadleaf plants and cereals basically remain intact through a season, with microbial succession on the same tissues throughout the growing season (di Menna, 1959; Glushakova and Chernoy, 2004;

Kinkel, 1997; Thompson et al., 1993; Vardavakis, 1988). The proximity to thatch and soil makes turfgrass leaves easy sites for colonization by soil and thatch microorganisms, rather than just foliar microorganisms from adjacent leaves.

196 Other factors that may affect the seasonal pattern of turfgrass microbial populations on leaves are duration of snow cover, soil type, cultural practices, insect and other pest colonization and nutrient availability.

Investigation of isolates obtained from dilution plating of leaf washings resulted in greater resolution of morphotypes. The original 11 color morphotypes assessed on dilution plates of Chapter 3 could be further subdivided into 26 submorphotypes after isolates were grown for periods (necessary for DNA extraction in Chapter 4), by which time more distinguishing cultural characteristics had developed. The resolution of submorphotypes was especially prominent with filamentous fungi, which went from three morphotypes after 9 days incubation at 9C to 13 submorphotypes after 14 days incubation at 25C for

DNA extraction. Yeast and bacteria morphotypes also changed cultural morphology as they aged in culture, but the overlapping similarities and the variability observed did not allow them to be easily separated into submorphotypes. Nevertheless, the original four bacterial morphotypes were further divided into seven submorphotypes, while the four yeast morphotypes were subdivided to add two new submorphotypes.

Isolates from the 26 submorphotypes of all three microbial groupings were subjected to molecular analysis with rDNA-RFLP and DNA sequencing. DNA was extracted from representatives of all submorphotypes and regions of ITS and 16S rDNA were amplified by PCR. The PCR products were digested with four or five restriction enzymes, and the analyses of banding patterns revealed

107 ribotypes. In distance-based dendrograms, the ribotypes of most filamentous

197 fungi and some yeast isolates clustered by submorphotype, indicating that most phyllosphere fungi from turfgrass leaves can accurately assessed by their morphological characteristics from cultures that are up to two weeks old. In addition, results from the RFLP analyses indicated that some ribotypes were associated with the type of leaf (green or yellow) or to the presence or absence of azoxystrobin treatment.

The results obtained with the RFLP analyses were complemented by rDNA sequencing and comparisons with the GenBank database. The number of microbial species identified was 89, and most of these species were either reported to form part of other phyllospheres or to be associated with soils (Allen et al., 2004b; Buck and Burpee, 2002; Inacio et al., 2002; Kim et al., 1998; Lewis,

1994; Enya et al., 2006; Yang et al., 2001), This suggests that the soil and thatch are the main sources of foliar microbial colonizers of turfgrass leaves. As with the

RFLP ribotypes, DNA sequencing revealed that most species of common filamentous fungi and some yeasts were associated with unique cultural submorphotypes.

In the literature, morphological classification of yeasts is based on color and texture, such as pink yeasts as Rhodotorula spp. and white yeasts as

Cryptococcus spp. (Allen et al., 2004b; Inacio et al., 2002; McCormack et al.,

1994). However, the results from sequencing of such morphotypes showed that many yeasts species share the same cultural morphotype, and classification or designation of taxon without further verification may be risky.

198 Among the 89 microbial species identified from A. stolonifera leaves, some

were present only on leaves collected from untreated plots, while other species

were associated only with leaves from azoxystrobin-treated plots. Fifteen

bacterial, six yeast and 13 filamentous fungal culturable species were present

only on samples from untreated plots. Azoxystrobin-treated plots had leaves with

culturable speceis of 11 bacteria, three yeasts and eight filamentous fungi that

were not found in untreated samples. In a previous investigation, a white

morphotype of yeast and total yeast populations on leaf surfaces of A. stolonifera

were also found to be reduced by applications of azoxystrobin (Buck and Burpee,

2002). Pink yeasts on A. stolonifera leaf surfaces have been in found significantly

larger numbers on leaves infected with S. homoeocarpa (Allen et al., 2004b), but this was not observed in the current study. The alterations caused by azoxystrobin treatment of microbial populations and the loss and gain of particular species indicate that an altered phyllosphere community is perhaps the main reason for the dollar spot disease enhancement caused by azoxystrobin.

In addition to the experimental evidence that dollar spot disease incidence may be increased many months after azoxystrobin application, this study has provided a better understanding of the role of the foliar microbial community in disease enhancement. These results shed light on the complexity of turfgrass microbial population and the non-target and longer term residual effects of azoxystrobin. Alteration of some microbial populations occurred at before and at the same time when dollar spot disease was observed to increase on azoxystrobin-treated plots, therefore further investigation is needed to determine

199 if this effect is the main cause of the disease enhancement rather than possible residual effects of the fungicide on plant physiology. Further studies can focus on the long term analysis of azoxystrobin residues and byproducts in soil, thatch and roots, and their effects on microbial communities. Non-culturable organisms may also be an important part of the phyllosphere and might have also affected dollar spot disease, but these species were not examined in this study. To address this, future studies might use metagenomic approaches to amplify and sequence DNA from complex substrates.

Further work is required to obtain a complete understanding of disease enhancement and its relationship with phyllosphere microbial community, but the findings of this study have contributed to our knowledge of the microorganisms associated with turfgrass environments, and demonstrate the importance of microbial ecology in the epidemiology of turfgrass diseases.

200 REFERENCES Adams RP & Demeke T. 1993. Systematic relationships in Juniperus based on random amplified polymorphic DNAs (RAPDs). Taxon 42:553-571. Alhubaishi AA & Abdel-Kader Ml. 1991. Phyllosphere and phylloplane fungi of qat in Sana'a, Yemen Arab Republic. Journal of Basic Microbiology 31:83-89. Allen TW, Burpee LL & Buck JW. 2004a. In vitro attachment of phylloplane yeasts to Botrytis cinerea, Rhizoctonia solani, and Sclerotinia homoeocarpa. Canadian Journal of Microbiology 50:1041-1048. Allen TW, Quayyum HA, Burpee LL & Buck JW. 2004b. Effect of foliar disease on the epiphytic yeast communities of creeping bentgrass and tall fescue. Canadian Journal of Microbiology 50:853-860. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W & Lipman DJ. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Research 25:3389-3402. Andrews JH & Buck JW. 2002. Adhesion of yeasts to leaf surfaces. In: Phyllosphere Microbiology. Edited by Lindow SE, Hetcht-Poinar E & Elliot VJ. American Phytopathology Society Press. Minneapolis. Minnesota, USA. Andrews JH & Kenerley CM. 1979. The effects of a pesticide program on microbial populations from apple leaf litter. Canadian Journal of Microbiology 25:1331-1344. Andrews JH, Kenerley CM & Nordheim EV. 1980. Positional variation in phylloplane microbial populations within apple tree canopy. Microbial Ecology 6:71-84. Anke T. 1995. The antifungal strobilurins and their possible ecological role. Canadian Journal of Botany 73:S940-S945. Anke T, Oberwinkler F, Steglich W & Schramm G. 1977. The strobilurins: new antibiotics from the basidiomycete Strobilurus tenacellus. Journal of Antibiotics (Tokyo) 30:806-810. Anonymous. 1997. Pesticide fact sheet. Azoxystrobin. Environmental Protection Agency. Washington, DC, USA. Anonymous. 1998. Golf course pesticide use and monitoring fact sheet. Pesticide Management Alberta Environment. Calgary, Alberta, Canada. Anonymous. 2000. Azoxystrobin. Pest management REG2000-15. Regulatory Agency, Health Canada. Ottawa, Ontario, Canada, www.hc-sc.gc.ca/pmra- arla/. Anonymous. 2003a. Kresoxim-methyl technical and Sovran® fungicide. Pest Management Regulatory Agency, Health Canada. Document PRDD2003-05. Ottawa, Ontario, Canada, www.hc-sc.gc.ca/pmra-arla/.

201 Anonymous. 2003b. Pyraclostrobin Headline® EC Cabrio® EG. Pest Management Regulatory Agency, Health Canada. Document REG2003-06. Ottawa, Ontario, Canada, www.hc-sc.gc.ca/pmra-arla/. Anonymous. 2005. Turfgrass management recommendations. Ontario Ministry of Agriculture, Food and Rural Affairs. Publication # 384. Queens Printer. Toronto, Ontario, Canada. Anonymous. 2006a. FRAC List of pathogens with field resistance towards Qol fungicides. http://www.frac.info/frac/meeting/Further_known_cases_of_Qol_resistance_2 006.pdf. Anonymous. 2006b. FRAC code list 1: fungicides sorted by FRAC code. Fungicide Resistance Action Committee. University of Florida. Gainesville, Florida, USA. http://ipm.ifas.ufl.edu/resources/success_stories/T&PGuide/pdfs/Appendices/ Appendix6-FRAC.pdf. Anonymous. 2007. Turfgrass management recommendations. Ontario Ministry of Agriculture, Food and Rural Affairs. Guelph, Ontario Canada. Publication 384S. http://www.omafra.gov.on.ca/english/crops/pub384/p384supplement.htm Aquino G & Heinrichs EA. 1979. Brown planthopper populations on resistant varieties treated with a resurgence causing insecticide. International Rice Research Newsletter 4:12. Assante G, Maffi D, Saracchi M, Farina G, Moricca S & Ragazzi A. 2004. Histological studies on the mycoparasitism of Cladosporium tenuissimum on urediniospores of Uromyces appendiculatus. Mycological Research 108 :170- 182. Atlas RM & Bartha R. 1987. Microbial Ecology. Second edition. The Benjamin/Cummings Publishing Company. Menlo Park, California, USA. Bab'eva IP & Sadikov BF. 1980. Yeast composition and number in plant phyllosphere. Mikologiya i Fitopatologiya 14:473-476. Bacon CW & Hill NS. 1996. Symptomless grass endophytes: products of coevolutionary symbiosis and their role in the ecological adaptation of grasses. In: Endophytic Fungi in Grasses and Woody Plants. Edited by Redlin SC & Carris LM. American Phytopathology Society Press. St. Paul. Minnesota, USA. Baldwin BG, Sanderson MJ, Porter JM & Wojciechowski MF. 1995. The ITS region of nuclear ribosomal DNA: a valuable source of evidence on angiosperm phylogeny. Annals of the Missouri Botanical Garden 82:247-277. Bartlett DW, Clough JM, Godfrey CR, Godwin JR, Hall AA, Heaney SP & Maund SJ. 2001. Understanding the strobilurin fungicides. The Pesticide Outlook August: 143-148.

202 Bartlett DW, Clough JM, Godwin JR, Hall AA, Hamer M & Parr-Dobrzanski B. 2002. The strobilurin fungicides. Pest Management Science 58:649-662. Beattie GA & Lindow SE. 1995. The secret life of foliar bacterial pathogens on leaves. Annual Review of Phytopathology 33:145-172. Becker WF, von Jagow G, Anke T & Steglich W. 1981. Oudemansin, strobilurin- a, strobilurin-b and myxothiazol: new inhibitors of the bc1, segment of the respiratory-chain with an e-B-methoxyacrylate system as common structural element. FEBS Letters 132:329-333. Beebee TJ & Rowe G. 2004. An Introduction to Molecular Ecology. Oxford University Press. New York. Behrendt U, Muller T & Seyfarth W. 1997. The influence of extensification in grassland management on the populations of micro-organisms in the phyllosphere of grasses. Microbiological Research 152:75-85. Behrendt U, Stauber T & Muller T. 2004. Microbial communities in the phyllosphere of grasses on fenland at different intensities of management. Grass and Forage Science 59:169-179. Behrendt U, Urich A & Schumann P. 2003. Fluorescent pseudomonads associated with the phyllosphere of grasses: Pseudomonas trivialis sp. nov., Pseudomonas poae sp. nov. and Pseudomonas congelans sp. nov. International Journal of Systematic and Evolutionary Microbiology 53:1461- 1469. Behrendt U, Ulrich A, Schumann P, Erler W, Burghardt J & Seyfarth W. 1999. A taxonomic study of bacteria isolated from grasses: a proposed new species Pseudomonas graminis sp. nov. International Journal of Systematic Bacteriology 49:297-308. Behrendt U, Urich A, Schumann P, Naumann D & Suzuki K. 2002. Diversity of grass-associated Microbacteriaceae isolated from the phyllosphere and litter layer after mulching the sward; polyphasic characterization of Subtercola pratensis sp. nov., Curtobacterium herbarum sp. nov. and Plantibacter flavus gen. nov., sp. nov. International Journal of Systematic and Evolutionary Microbiology 52:1441-1454. Belanger RR & Avis TJ. 2002. Ecological processes and interaction as occurring in leaf surfaces fungi. In: Phyllosphere Microbiology. Edited by Lindow SE, Hetcht-Poinar E & Elliot VJ. American Phytopathology Society Press. Minneapolis, Minnesota, USA. Belanger RR, Dufour N, Caron J & Benhamou N. 1995. Chronological events associated with the antagonic properties of Trichoderma harzianum against Botrytis cinerea: indirect evidence of sequential role of antibiosis and parasitism. Biocontrol Science and Technology 5:41-53. Bennett FT. 1937. Dollar spot disease of turf and its causal organism Sclerotinia homoeocarpa n. sp. Annals of Applied Biology 24:236-257.

203 Benyagoub M, Bel Rhlid R & Belanger RR. 1996 Purification and characterization of new fatty acids with antibiotic activity produced by Sporothrix flocculosa. Journal of Chemical Ecology 22:405-413. Berg G, Krechel A, Ditz M, Sikora RA, Ulrich A & Hallmann J. 2005. Endophytic and ectophytic potato-associated bacterial communities differ in structure and antagonistic function against plant pathogenic fungi. FEMS Microbiology Ecology 51:215-229. Bertelsen JR, de Neergaard E & Smedegaard-Petersen V. 2001. Fungicidal effects of azoxystrobin and epoxiconazole on phyllosphere fungi, senescence and yield of winter wheat. Plant Pathology 50:190-205. Blackwood CB, Marsh T, Kim S & Paul EA. 2003. Terminal restriction fragment length polymorphism data analysis for quantitative comparison of microbial communities. Applied and Environmental Microbiology 69:926-932. Bomeman J & Triplett EW. 1997. Rapid and direct method for extraction of RNA from soil. Soil Biology and Biochemistry 29:1621-1624. Boulter Jl, Boland GJ & Trevors JT. 2002. Evaluation of composts for suppression of dollar spot (Sclerotinia homoeocarpa) of turfgrass. Plant Disease 86:405-410. Bover C. 2005. DGGE, T-RFLP, LH-PCR, ARISA, Melt Curves. From: http://www.soest.hawaii.edu/oceanography/zij/ocn750lectures/ocn750- 20051103-dgge.pdf. Breeze EM & Dix NJ. 1981. Seasonal analysis of the fungal community on Acer platanoides leaves. Transactions of the British Mycological Society 77:321- 328. Brem D & Leuchtmann A. 2001. Epichloe grass endophytes increase herbivore resistance in the woodland grass. Brachypodium sylvaticum. Oecologia (Berlin.) 126:522-530. Buck JW & Burpee LL. 2002. The effects of fungicides on the phylloplane yeast populations of creeping bentgrass. Canadian Journal of Microbiology 48:522- 529. Burpee LL. 1997. Control of dollar spot of creeping bentgrass caused by an isolate of Sclerotinia homoeocarpa resistant to benzimidazole and demethylation-inhibitor fungicides. Plant Disease 81:1259-1263. Cabral D, Cafaro MJ, Saidman BO, Lugo MA, Reddy PV & White JF. 1999. Evidence supporting the occurrence of a new species of endophyte in some South American grasses. Mycologia 91:315-325. Carbone I & Kohn LM. 1993. Ribosomal DNA sequence divergence within internal transcribed spacer 1 of the Sclerotiniaceae. Mycologia 85:415-427. Carroll GC, Muller EM & Sutton BC. 1977. Preliminary studies on the incidence of needle endophytes in some European conifers. Sydowia 29:87-103.

204 Chand-Goyal T & Spotts RA. 1996. Control of postharvest pear diseases using natural saprophytic yeast colonists and their combination with low dosage of thiabendazole. Postharvest Biological Technology 7:51-64. Charbonneau P & Hsiang T. 2003. Biology and control of dollar spot disease. Ontario Ministry of Agriculture, Food and Rural Affairs. http://www.omafra.gov.on.ca/english/crops/facts/info_turfdollarspot.htm. Accessed February 2006. Cheplick GP & Clay K. 1988. Acquired chemical defenses of grasses: the role of fungal endophytes. Oikos 52:309-318. Chng SF, Cromey MG & Butler RC. 2004. Mycoflora of rhizomes and roots of Elytrigia repens, a potential inoculum source for take-all in wheat. New Zealand Plant Protection 57:10-12. Christensen MJ, Leuchtmann A, Rowan DD & Tapper BA. 1993. Taxonomy of Acremonium endophytes of tall fescue (Festuca arundinacea), meadow fescue (F. pratensis), and perennial ryegrass (Lolium perenne). Mycological Research 97:1083-1092. Christians NE. 2007. Fundamentals of turfgrass management. Third Edition. John Wiley and Sons. Cranford, New Jersey, USA. Clay K. 1990. Insects, endophytic fungi and plants. In: Pests, Pathogens and Plant Communities. Edited by Burdon JJ & Leather SR. Blackwell. Oxford, Oxfordshire, England. Cook S & Hsiang T. 2002. Dollar spot disease control, summer 2002. Guelph Turfgrass Institute 2002 Annual Research Report 16:45-46. Cook S & Hsiang T. 2003. Dollar spot disease control, summer 2003, BASF. Guelph Turfgrass Institute 2003 Annual Research Report 17:57-59. Cook S & Hsiang T. 2004. Chemical trials for dollar spot disease control summer 2004. Guelph Turfgrass Institute 2004 Annual Research Report 18:108-110. Couch HB. 1995. Disease of Turfgrasses. Third edition. Krieger Publisher. Malabar, Florida, USA. Couch HB & Joyner BG. 1975. Interactions among fungicide treatments and nitrogen fertilization in the control of rust on Manhattan rye grass. Proceedings of the American Phytopathological Society 2:122. Couch HB & Smith BD. 1991. Increase in incidence and severity of target turfgrass diseases by certain fungicides. Plant Disease 75:1064-1067. Craven KD, Blankenship JD, Leuchtmann A, Hignight K & Schardl CL. 2001. Hybrid fungal endophytes symbiotic with the grass Lolium pratensis. Sydowia 53:44-73. Cromey MG, Butler RC, Mace MA & Cole ALJ. 2004. Effects of the fungicides azoxystrobin and tebuconazole on Didymella exitialis, leaf senescence and grain yield in wheat. Crop Protection 24:1019-1030.

205 Dernoeden PH. 2000. Dollar spot: getting tougher to manage in creeping bentgrass. Turfax 8:2-6. Dernoeden PH. Krouse JM & Kaminski JE. 2000. Curative dollar spot control with fungicides, 1999. Fungicide and Nematicide Tests 55:491. Dernoeden PH & Mcintosh MS. 1991. Disease enhancement and Kentucky bluegrass quality as influenced by fungicides. Agronomy Journal 83:320-326. Dernoeden PH, Murray JJ & O'Neill NR. 1985. Non-target effects of fungicides on turfgrass growth and enhancement of red thread. In: Proceedings International Turfgrass Research Conference 5th. Edited by Lemaire F. Avignon, Vaucluse, France. Dellaporta SL, Wood J & Hicks JB. 1983. A plant DNA minipreparation: version II. Plant Molecular. Biology 1:19-21. Detweiler AR, Vargas JM & Dannenberger TK. 1983. Resistance of Sclerotinia homoeocarpa to iprodione and benomyl. Plant Disease 67:627-630. di Menna MED. 1959. Yeasts from the leaves of pasture plants. New Zealand Journal of Agricultural Research 2:394-405. Dickinson CH. 1982. The phylloplane and other aerial plant surfaces. In: Experimental Microbial Ecology. Edited by Burns RG & Slater JH. Blackwell Scientific Publishers. Oxford, Oxfordshire, England. Dickinson CH, Austin B & Goodfellow M. 1975. Quantitative and qualitative studies of phylloplane bacteria from Lolium perenne. Journal of General Microbiology 91:157-166. Dik AJ. 1991. Interactions among fungicides, pathogens, yeasts, and nutrients in the phyllosphere. In Microbial Ecology of Leaves. Edited by Andrews JH & Hirano SS. Springer-Verlag. New York, New York, USA. Dik AJ & Elad Y. 1999. Comparison of antagonists of Botrytis cinerea in greenhouse-grown cucumber and tomato under different climatic conditions. European Journal of Plant Pathology. 105:122-137. Dowding P. 1986. Leaf yeasts as indicators of air pollution. In: Microbiology of the Phyllosphere. Edited by Fokkema NJ & Van den Heuvel J. Cambridge University Press. Cambridge, Cambridgeshire, England. Droby S, Chalutz E, Wilson CL & Wisniewski M. 1989. Characterization of the biocontrol activity of Debaryomyces hansenii in the control of Penicillium digitatum on grapefruit. Canadian Journal of Microbiology 35:794-800. Dutcher JD. 2007. A review of resurgence and replacement causing pest outbreaks in IPM. In: General Concepts in Integrated Pest and Disease Management. Edited by:Ciancio A & Mukerji KG. Springer Netherlands. Dortdrech, South Holland, Netherlands.

206 Edwards K, Johnstone C & Thomsom C. 1991. A simple and rapid method for the preparation of plant genomic DNA for PCR analysis. Nucleic Acids Research 19:1349-1349. Ellram A, Horgan B & Hulke B. 2007. Mowing strategies and dew removal to minimize dollar spot on creeping bentgrass. Crop Science 47:2129-2137. Enya J, Shinohara H, Yoshida S, Tsukiboshi T, Negishi H, Suyama K & Tsushima S. 2006. Culturable leaf-associated bacteria on tomato plants and their potential as biological control agents. Microbial Ecology 53:524-536. Faith D. 2007. Biodiversity. In Stanford Encyclopedia of Philosophy. http://plato.stanford.edU/entries/biodiversity/#ConBio Felsenstein, J. 1989. PHYLIP - Phylogeny Inference Package (Version 3.2). Cladistics 5:164-166. Ferrari M & Carrera PS. 2005. Towards Automation for Molecular Diagnosis of Cancer. From: http://www.ifcc.org/ejifcc/vol16no2/160206200513.htm. Fischer SG & Lerman LS. 1979. Length-independent separation of DNA restriction fragments in two-dimensional gel electrophoresis. Cell 16:191-200. Fischer SG & Lerman LS. 1983. DNA fragments differing by single base-pair substitutions are separated in denaturing gradient gels: correspondance with melting theory. Proceedings of the National Academy of Sciences of the United States of America 80:1579-1583. Fliessbach A & Mader P. 2004. Short and long-term effects on soil microorganisms of two potato pesticide spraying sequences with either glufosinate or dinoseb as defoliants. Biology and Fertility of Soils 40:268-276. Fokkema NJ & de Nooij MP. 1981. The effect of fungicides on the microbial balancein the phyllosphere. EPPO bulletin 11:303-310. Fokkema NJ, Van De Laar JAJ, Nelis-Blomberg AL, Schippers B, 1975. The buffering capacity of the natural mycoflora of rye leaves to infection by Cochliobolus sativus, and its susceptibility to benomyl. Netherlands Journal of Plant Pathology 81,176-86. Fokkema NJ, Riphagen I, Poot RJ & de Jong C. 1983. Aphid honeydew, a potential stimulant of Cochliobolus sativus and Septoria nodorum and the competitive role of saprophytic mycoflora. Transactions of the British Mycological Society 81:355-363. Forbes G. 2001. Resistance to fungicides: theory and practice. In: Complementing Resistance to Late Blight (Phytophthora infestans) in the Andes. Edited by Fernandez-Northcote EN. GILB. Cochabamba, Cochabamba, Bolivia. Franckland JC. 1998. Fungal succesion-unravelling the unpredictable. Mycological Research 102:1-15.

207 Garret M, Rayment P, Hooper MA, Abramson M & Hooper BM. 1998. Indoor airborne fungal spores, house dampness and associations with environmental factors and respiratory health in children. Clinical and Experimental Allergy 28 :459-467. Gildemacher PR, Heijne B, Houbraken J, Vromans T, Hoekstra ES & Boekhout T. 2004. Can phyllosphere yeasts explain the effect of scab fungicides on russeting of Elstar apples? European Journal of Plant Pathology 110:929- 937. Gisi U, Sierotzki H, Cook A & McCaffery A. 2002. Mechanisms influencing the evolution of resistance to Qo inhibitor fungicides. Pest Management Science 58:859-867. Gleason ML, Dickson JR & Christians NC. 1996. Evaluation of fungicides for control of dollar spot in creeping bentgrass, 1995. Fungicide and Nematicide Tests 51:342 Glushakova AM & Chernov IY. 2004. Seasonal dynamics in a yeast population on leaves of the common wood sorrel Oxalis acetosella L. Microbiology 73:226-232. Golembiewski RC, Vargas JM, Jones AL & Detweiler AR. 1995. Detection of demethylation inhibitor (DMI) resistance in Sclerotinia homoeocarpa populations. Plant Disease 49:491-493. Gold RE, Ammerman E, Koehl H, Leinhos GME, Lorenz G, Speakman JB, Stark- Urnau M & Sauter H. 1996. The synthetic strobilurin BAS 490 F: Profile of a modern fungicide. In Modern Fungicides and Antifungal Compounds. Edited by Lyr H, Russell PE & Sisler HD. Intercept, Andover, Hampshire, England. Gomes EA, Catarina M, Kasuya M, de Barros EG, Borges AC & Araujo EF. 2002. Polymorphism in the internal transcribed spacer (ITS) of the ribosomal DNA of 26 isolates of ectomycorrhizal fungi. Genetics and Molecular Biology 25:477-483. Goodman DM & Burpee LL. 1991. Biological control of dollar spot disease of creeping bentgrass. Phytopathology 81:1438-1446. Goring CAI, Laskowski DA, Hamaker JW & Meikle RW. 1975. Principles of pesticide degradation in soil. In: Environmental Dynamics of Pesticides. Edited by Haque R & Freed VH. Plenum Press, New York, New York, USA. Grogan JE & Scott DH. 1997. Evaluation of fungicides for control of dollar spot in creeping bentgrass, 1996. Fungicide and Nematicide Tests 52:351-352. Grossmann K & Retzaff G. 1997. Bioregulatory effects of the fungicidal strobilurin kresoxim-methyl in wheat (Triticum aestivum). Pesticide Science 50:11-20. Habash M & Reid G. 1999. Microbial biofilms: their development and significance for medical device-related infections. Journal of Clinical Pharmacology 39:887-898.

208 Haile FJ, Kerns DL, Richarson JM & Higley LG. 2000. Impact of insecticides and surfactant on lettuce physiology and yield. Journal of Economic Entomology 93:788-794. Han D, Rimelspach J, Parisi B & Boehm Mike. 2005. Biological control of dollar spot of golf courses using Pseudomonas aureofaciens TX-1 delivered via the BioJect. http://www.ca.uky.edu/agcollege/plantpathology/PPAExten/Bioject.htm Harman GE, Nelson EB & Ondik KL. 2006. Non-target effects of fungicide applications in microbial populations of putting greens. USGA Turfgrass and Environmental Research 5:1-6. Hawksworth DL. 1991. The fungal dimension of biodiversity: magnitude, significance, and conservation. Mycological Research 95:641-655. Heaney SP, Hall AA, Davies SA & Olaya G. 2000. Resistance to fungicides in the Qol-STAR cross-resistance group: current perspectives. Proceedings of the British Crop Protection Conference: Pests and Disease Franham, Surrey, UK, British Crop Protection Council 755-762. Henderson BG, Wenhama PR, Ashby JP & Blundell G. 1997. Detecting familial defective apolipoprotein B-100: three molecular scanning methods compared. Clinical Chemistry 43:1630-1634. Hepburn M & Miller GA. 1998. A multiple mutation model system as a test development and training tool for denaturing gradient gel electrophoresis. http://www.biocompare.com/technicalarticle/1101/A-Multiple-Mutation-Model- System-As-A-Test-Development-And-Training-Tool-For-Denaturing-Gradient- Gel-Electrophoresis-from-Bio-Rad.html Heredia G. 1993. Mycoflora associated with green leaves and litter of Quercus germana, Quercus sartorri and Liquidambar styraciflua in a Mexican cloud forest. Cryptogamie Mycologie 14:171-183. Hershkovitz MA, Zimmer EA & Hahn WJ. 1999. Ribosomal DNA sequences and angiosperm systematics. In: Molecular Systematics and Plant Evolution. Edited by: Hollingsworth PH, Bateman RM & Gornall RJ. Taylor and Francies, New York, New York, USA. Hirano SS & Upper CD. 2000. Bacteria in the leaf ecosystem with emphasis on Pseudomonas syringae - a pathogen, ice nucleus, and epiphyte. Microbiology and Molecular Biology Reviews 64:624-623. Hodge JE. 1993. Pesticides in Urban Environments: Fate and Significance. Edited by Racke KD & Leslie AR. ACS Symposium Series No. 522. American Chemical Society. Washington, DC. Hoist-Jensen A, Kohn LM & Schumacher T. 1997. Nuclear rDNA phytogeny of the Sclerotiniaceae. Mycologia 89:885-899.

209 Howel CR. 2003. Mechanisms employed by Trichoderma species in the biological control of plant diseases: the history and evolution of current concepts. Plant Disease 97:4-10. Hoyland BF & Landschoot PJ. 1993. Effect of various nitrogen (N) sources on dollar spot suppression. Biological and Cultural Tests for Control of Plant Diseases 8:112. Hsiang T & Cook S. 2006. The effect of winter applications of azoxystrobin on dollar spot. Guelph Turfgrass Institute. 2005 Annual Research Report 20:82- 84. Hsiang T, Yang L & Barton, W. 1997. Baseline sensitivity and cross-resistance to demethylation-inhibiting fungicides by Ontario isolates of Sclerotinia homoeocarpa. European Journal of Plant Pathology 103:409-416. Hsiang T & Wu CR. 2000. Genetic relationships of pathogenic Typhula species assessed by RAPD, ITS-RFLP and ITS sequencing. Mycological Research 104:16-22. Huang W, Zhao PL, Liu CL & Chen Q. 2007. Design, synthesis, and fungicidal activities of new strobilurin derivatives. Journal of Agriculture and Food Chemistry 55:3004-3010. Inacio J, Pereira P, de Carvalho M, Fonseca A, Amaral-Collac MT & Spencer- Martins I. 2002. Estimation and diversity of phylloplane mycobiota on selected plants in a mediterranean-type ecosystem in Portugal. Microbial Ecology 44:344-353. Inacio J, Portugal L, Spencer-Martins I & Fonseca A. 2005. Phylloplane yeasts from Portugal: seven novel anamorphic species in the Tremellales lineage of the Hymenomycetes (Basidiomycota) producing orange-colored colonies. FEMS Yeast Research 5:1167-1183. Jackson NE. 1970. Evaluations of some chemicals for control of stripe smut in Kentucky bluegrass turf. Plant Disease 54:168-170. Janisiewicz WJ & Korsten L. 2002. Biological control of postharvest diseases of fruits. Annual Review of Phytopathology 40:411-441. Jeng RS, Svircev AM, Myers AL, Beliaeva L, Hunter DM & Hubbes M. 2001. The use of 16S and 16S-23S rDNA to easily detect and differentiate common Gram-negative orchard epiphytes. Journal of Microbiological Methods 44:69- 77. Juniper BE. 1991. The leaf from the inside and the outside: a microbe's perspective. In Microbial Ecology of Leaves. Edited by Andrews JH & Hirano SS. Springe-Verlag. New York, New York, USA. Kaumon S. 2003. Molecular genetics of pathogenic oomycetes. Eukaryotic Cell 2:191-199.

210 Kawamata H, Narisawa K & Hashiba T. 2004. Suppression of rice blast by phylloplane fungi isolated from rice plants. Journal of General Plant Pathology 70:131-138. Kim H, Nishiyama M, Kunito T, Senoo K, Kawahara K, Murakami K & Oyaizu H. 1998. High population of Sphingomonas species on plant surface. Journal of Applied Microbiology 85:731-736. Kinkel LL. 1997. Microbial population dynamics on leaves. Annual Review of Phytopathology 35:327-347. Kinkel LL, Newton MR & Leonard KJ. 2002. Resource aggregation in the phyllosphere: implications for microbial dynamics across spatial scales. In: Phyllosphere Microbiology. Edited by Lindow SE, Hetcht-Poinar E & Elliot VJ. American Phytopathology Society Press. Minneapolis, Minnesota, USA. Kirk JL, Beaudette LA, Hart M, Moutoglis P, Klironomos JN, Lee H & Trevors JT. 2004. Methods of studying soil microbial diversity. Journal of Microbiological Methods 58:169-188. Kirk J J & Deacon JW. 1986. Control of the take-all fungus by Microdochium bolleyi, and interactions involving M. bolleyi, Phialophora graminicola and Periconia macrospinosa on cereal roots. Plant and Soil 98:231-237. Kleven S, Dean G, Hacking C & Davidson J. 2003. Evaluation of a new class of fungicides on grain yield in wheat and barley. Australian Society of Agronomy. http://www.regional.org.aU/au/asa/2003/p/7/kleven.htm Kohl J, Belanger RR & Fokkema NJ. 1997. Interaction of four antagonistic fungi with Botrytis aclada in dead onion leaves: a comparative microscopic and ultrastructural study. Phytopathology 87:634-642. Kohn LM. 1993. What do we need to know about discomycetous anamorphs? The fungal holomorph: Mitotic, meiotic and pleomorphic speciation in fungal systematics Edited by Reynolds DR & Taylor JW. CAB International, Wallingford. Krupa P. 1999. Identification by PCR-RFLP of a fungus isolated from mycorrhizal roots of a distinguishable birch growing in areas disturbed by industry. Polish Journal of Environmental Studies 8:161-163. Kuklinsky-Sobral J, Araujo WL, Mendes R, Geraldi IO, Pizzirani-Kleiner AA & Azevedo JL. 2004. Isolation and characterization of soybean-associated bacteria and their potential for plant growth promotion. Environmental Microbiology 6:1244-1251. Larena I, Torres R, De Cal A, Linan M, Melgarejo P, Domenichini P, Bellini A, Mandrin JF, Lichou J, Ochoa de Eribee X & Usall J. 2005. Biological control of postharvest brown rot (Monilinia spp.) of peaches by weld applications of Epicoccum nigrum. Biological Control 32:305-310. Last FT. 1955. Seasonal incidence of Sporobolomyces on cereal leaves. Transactions of the British Mycological Society 38:221-239.

211 Lee L, Fry J & Tisserat N. 2003. Dollar spot in four bentgrass cultivars as affected by acibenzolar-s-methyl and organic fertilizers. Plant Management Network, http://www.plantmanagementnetwork.org/ Leuchtmann A & Schardl CL. 1998. Mating compatibility and phylogenetic relationships among two new species of Epichloe and other congeneric European species. Mycological Research 102:1169-1182. Leuchtmann A, Schmidt D & Bush LP. 2000. Different levels of protective alkaloids in grasses with stroma-forming and seed-transmitted Epichloe I Neotyphodium endophytes. Journal of Chemical Ecology 26:1025-1036. Lewis Rl. 1994. Species diversity and resource relationships of South Georgian fungi. Antarctic Science 6:45-52. Liesack W & Stackebrandt E. 1992. Occurrence of novel groups of the domain bacteria as revealed by analysis of genetic material isolated from an Australian terrestrial environment. Journal of Bacteriology 174:5072-5078. Lindow SE & Brandl MT. 2003. Microbiology of the phyllosphere. Applied and Environmental Microbiology 69:1875-1883. Lindow SE & Leveau JH. 2002. Phyllosphere microbiology. Current Opinion in Biotechnology 13:238-43. Liu LX, Hsiang T, Carey K & Eggens JL. 1995. Microbial populations and suppression of dollar spot disease in creeping bentgrass with inorganic and organic amendments. Plant Disease 79:144-147. Liu W, Marsh TL, Cheng H & Forney LJ. 1997. Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Applied and Environmental Microbiology 45:4516- 4522. Lo CT, Nelson EB & Harman GE. 1997. Improved biocontrol efficacy of Trichoderma harzianum 1295-22 for foliar phases of turf diseases by use of spray applications. Plant Disease 81:1132-1138. Ma Z, Felts D & Michailides TJ. 2003. Resistance to azoxystrobin in Alternaria isolates from pistachio in California. Pesticide Biochemistry and Physiology 77:66-74. Mann R. 2004. Review to identify, collate and assess research on the management and control of the main pests and diseases on European golf courses. The Royal and Ancient Golf Club of St. Andrews. Document No. 2112/1. Fife. Scotland, www.bestcourseforgolf.org/images/pests.pdf. Matsuyama T, Nakajima Y, Matsuya K, Ikenaga M, Asakawa S, Kimura M. 2007. Bacterial community in plant residues in a Japanese paddy field estimated by RFLP and DGGE analyses. Soil Biology and Biochemistry 39:463-472. Mazur AR & Hughes TD. 1976. Chemical composition and quality of Penncross creeping bent grass as affected by ammonium nitrate and several fungicides. Agronomy Journal 68: 721-723.

212 McCormack PJ, Wildman HG & Jeffries P.1994. Production of antibacterial compounds by phylloplane-inhabiting yeasts and yeastlike fungi. Applied Environmental Microbiology 60:927-93. McKeinzie EHC & Vanky K. 2001. Smut fungi of New Zealand: an introduction, and list of recorded species. New Zealand Journal of Botany 39:501-515. McDougall P. 2001. Report 249, July 2001. Crop Protection and Agricultural Biotechnology Consultants. Melzer MJ & Boland GH. 1998. Evaluation of Trichoderma harzianum treatments for dollar spot suppression. Biological Culture Tests control Plant Disease 13:141-141. Monier JM & Lindow SE. 2005. Spatial organization of dual-species bacterial aggregates on leaf surfaces. Applied and Environmental Microbiology 71:5484-5493. Moon CD, Miles CO. Jalfors U & Schardl CL. 2002. The evolutionary origins of three new Neotyphodium endophyte species from grasses indigenous to the Southern Hemisphere. Mycologia 94:694-711. Miles CO, di Menna MED, Jacobs SWL, Garthwaite I, Lane GA, Prestidge RA, Marshall SL, Wilkinson HH, Schardl CL, Ball OJP & Latch GCM. 1998. Endophytic fungi in indigenous Australasian grasses associated with toxicity to livestock. Applied Environmental Entomology 64:601-606. Monteith J & Dahl AS. 1932. Turf diseases and their control. Bulletin US Golf Association. Green Section 12:87-486. Morris CE, Barness MB & McLean RJ. 2002 Biofilms on leaf surface: implications for biology, ecology and management of populations of epiphytic bacteria. In: Phyllosphere Microbiology Edited by Lindow SE, Hetcht-Poinar E & Elliot VJ. American Phytopathology Society Press. Minneapolis, Minnesota, USA. Morris CE & Kinkel LL. 2002. Fifty years of phyllosphere microbiology: significant contributions to research in related field. In: Phyllosphere Microbiology. Edited by Lindow SE, Hetcht-Poinar E & Elliot VJ. American Phytopathology Society Press. Minneapolis, Minnesota, USA. Morris CE & Rouse DI. 1986. Microbiological and sampling considerations for quantification of epiphytic microbial community structure. In: Microbiology of the Phyllosphere. Edited by Fokkema NJ & Van den Heuvel J. Cambridge University Press. Cambridge, Cambridgeshire, England. Moy M, Huaijun ML, Sullivan R, White JF & Belanger FC. 2002. Endophytic fungal p-1,6-Glucanase expression in the infected host grass. Plant Physiology, 130, 1298-1308. Moyer CL, Dobbs FC & Karl DM. 1994. Estimation of diversity and community structure through restriction fragment length polymorphism distribution analysis of bacterial 16S rRNA genes from a microbial mat at an active,

213 hydrothermal vent system, Loihi Seamount, Hawaii. Applied Environmental Microbiology 60:871-879. Muyzer G. 1998. Genetic fingerprinting of microbial communities - present status and future perspectives. In: Microbial Biosystems: New Frontiers. Edited by Bell CR, Brylinsky M & Jhonson-Green P. Atlantic Canada Society for Microbial Ecology. Halifax, Nova Scotia, Canada. Muyzer G, de Waal EC & Uitterlinden AG. 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes encoding for 16S rRNA. Applied and Environmental Microbiology 59:695-700. Muyzer G & Smalla K. 1998. Application of denaturing gradient gel electrophoresis (DGGE) and temperature gradient gel electrophoresis (TGGE) in microbial ecology. Antonie Van Leeuwenhoek International Journal of General and Molecular Microbiology 73:127-141. Nakatsu CH, Torsvik V & 0vreas L. 2000. Soil community analysis using DGGE of 16S rDNA polymerase chain reaction products. Soil Science Society of America 64:1382-1388. Nei M & Li WH. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy of Sciences of the United States of America 76:5269-5273. Noble CW. 1982. Sampling the skin surface. In: Experimental Microbial Ecology. Edited by Burns RG & Slater JH. Blackwell Scientific Publishers. Oxford, Oxfordshire, England. Odum EP. 1959. Fundamentals of Ecology. Third edition. Philadelphia Saunders Co. London, England. Osono T & Mori A. 2004. Distribution of phyllosphere fungi within the canopy of giant dogwood. Mycoscience 45:161-168. Ott EM, MUller T, Muller M, Franz CM, Ulrich A, Gabel M & Seyfarth W. 2001. Population dynamics and antagonistic potential of Enterococci colonizing the phyllosphere of grasses. Journal of Applied Microbiology 91:54-66. 0vreas L, Forney L, Daae FL & Torsvik V. 1997. Distribution of bacterioplankton in meromictic Lake Saslenvannet, as determined by denaturing gradient gel electrophoresis of PCR-amplified gene fragments coding for 16S rRNA. Applied and Environmental Microbiology 63:3367-3373. Panaccione DG, Johnson RD, Wang J, Young CA, Damrongkool P, Scott B & Schardl CL. 2001. Elimination of ergovaline from a grass-Neotyphodium endophyte symbiosis by genetic modification of the endophyte. Proceedings of the National Academy of Sciences of the United States of America 98:12820-12825. Page RDM. 1996. TREEVIEW: An application to display phylogenetic trees on personal computers. Computer Applications in the Biosciences 12: 357-358.

214 Pascual S, Magan N & Melgarejo P. 1996. Improved biological control of peach twig blight by physiological manipulation of Epicoccum nigrum. Brighton Crop Protection Conference: Pests and Diseases 3:411-412. Petrini O. 1986. Taxonomy of endophytic fungi of aerial plant tissues. In: Microbiology of the Phyllosphere. Edited by Fokkema NJ &. van den Hueval J. Cambridge University Press, Cambridge, Cambridgeshire, England. Racke KD. 2000. Pesticides for turfgrass pest management: uses and environmental issues. In: Fate and Management of Turfgrass Chemicals. Edited by Clark JM & Kenna MP. Symposium Series No. 743. American Chemical Society. Washington, DC, USA. Rahman A, Basumatary SK & Ahmed M. 2004. Seasonal variation in microfungi associated with Soalu (Litsea polyantha) leaves. Plant Archives 4:499-501. Rivera MC, Wright ER, Goldring LV, Perez BA & Barreto D. 2004. First report of dollar spot caused by Sclerotinia homoeocarpa on Agrostis stoionifera in Argentina. Plant Disease 88:12. Rodriguez F & Pfender WF. 1997. Antibiosis and antagonism of Sclerotinia homoeocarpa and Drechslera poae by Pseudomonas fluorescens Pf-5 in vitro and in planta. Phytopathology 87:614-621. Rudolph K, Roy MA, Sasser M, Stead DE, Davis DE, Swings M & Gossele F. 1990. Isolation of bacteria. In: Methods in Phytobacteriology, Edited by Klement Z, Rudolph K & Sands DC. Budapest, Hungary. Salaiz TA, Horst GL & Sherman RC. 1995. Mowing height and vertical mowing frequency effects on putting green quality. Crop Science 35:1422-1425. Santamaria J & Bayman P. 2005. Fungal endophytes of coffee leaves (Coffea arabica). Microbial Ecology 50:1-8.New York, USA. Sauter H, Steglich W & Anke T. 1999. Strobilurins: evolution of a new class of active substances. Angewandte Chemie International Edition 38:1328-1349. Schardl CL. 1996. Epichloe species: fungal symbionts of grasses. Annual Reviews of Phytopathology 34:109-130. Schardl CS & Leuchtmann A. 1999. Three new species of Epichloe symbiotic with North American grasses. Mycologia 91:95-107. Schlossberg MJ & Schmidt JP. 2007. Influence of nitrogen rate and form on quality of putting greens cohabited by creeping bentgrass and annual bluegrass. Agronomy Journal 99:99-106. Schreiber L, Krimm U, Knoll D, Sayed M, Auling G & Kroppenstedt RM. 2005. Plant-microbe interactions: identification of epiphytic bacteria and their ability to alter leaf surface permeability. New Phytologist 166:589-594. Schumann GL. 1996. Evaluation of fungicides for control of brown patch and dollar spot on creeping bentgrass, 1995. Fungicide and Nematicide Tests 51:348.

215 Schumann G. 2003. Dollar spot, fungicides, gray leaf spot in NJ. UMAAS. http://www.umassturf.org/mangement_updates/2003_archive/03sep25.htm at September 2005. Scow KM, Schwartz E, Johnson MJ & Macalady JL. 2001. Microbial biodiversity, measurement of. Encyclopedia of Diversity. Volume 4. Edited by Levin SA. Academic Press. San Diego, California, USA. Siegel MR, Latch GCM, Bush LP, Fannin FF, Rowan DD, Tapper BA, Bacon CW & Johnson MC. 1990. Fungal endophyte-infected grasses: Alkaloid accumulation and aphid responses. Journal of Chemical Ecology 16:3301- 3315. Sigler WV, Taylor CP, Throssell CS, Bischoff M & Turco RF. 2000. Environmental fates of fungicides in turfgrass environment: a minireview. In: Fate and Management of Turfgrass Chemicals. Edited by Clark JM & Kenna MP. Symposium Series No. 743. American Chemical Society. Washington, DC, USA. Silva, PC. 2002. An overview of the genus Caulerpa. In: International Caulerpa taxifolia Conference Proceedings. San Diego, California, USA. Slavikova E & Vadkertiova. 2003. Effects of pesticides on yeasts isolated from agricultural soil. Verlag der Zeitschrift fur Naturforschung 58:855-859. Smiley, RW. 1981. Non-target effects of pesticides on turfgrass. Plant Disease 65:17-23. Smiley RW & Craven MM. 1979. Microflora of turfgrass treated with fungicides. Soil Biology and Biochemistry 11:349-353. Smiley RW, Dernoeden PH & Clarke BB. 2005. Compendium of Turfgrass Diseases. Third edition. American Phytopathology Society Press. St. Paul, Minnesota, USA. Smith JD, Gossen BD & Hsiang T. 2001. First report of dollar spot, caused by Sclerotinia homoeocarpa, on Poa pratensis in Saskatchewan, Canada. Plant Disease 85:803 Smith JD, Jackson N & Woolhouse AR. 1989. Fungal Diseases of Amenity Turf Grasses. Third Edition. E. & F.N. Spon. London, England. Soltis DE & Soltis PS. 1998. Choosing an approach and an appropriate gene for phylogenetic analysis. In: Molecular Systematics of Plants II. Edited by Doyle JJ. Kluver Academic Publishers. Boston, Massachusetts, USA. Stanley JT & Konopka AL. 1985. Measurement of in situ activities of non- photosynthetic microorganisms in aquatic and terrestrial habitats. Annual Review of Microbiology 39:321-346. Stier J & Pearson S. 2000. Biocontrol of Dollar Spot on Putting Greens. http://www.hort.wisc.edu/uwturf/pdf/sybspot.pdf.

216 Stohr SN & Dighton J. 2004. Effects of species diversity on establishment and coexistance: a phylloplane fungal community model system. Microbial Ecology 48:431-438. Studzinska A, Mufioz M & Danneberger K. 2006. Dollar Spot rampant on Poa pratensis. http://buckeyeturf.osu.edu. Sun S & Brede AD. 1997. Inoculation of creeping bent with Acremonium endophyte. International Turfgrass Society Research Journal. 8:925-929. Sundin GW. 2002. Ultraviolet radiation on leaves: its influence on microbial communities and their adaptations. In: Phyllosphere Microbiology. Edited by Lindow SE, Hetcht-Poinar E & Elliot VJ. American Phytopathology Society Press. Minneapolis. Minnesota, USA. Tanaka K, Endo S & Kazano H. 1999. Toxicity of insecticides to predators of rice planthoppers: Spiders, the mirid bug and the dryinid wasp. Applied Entomology and Zoology 35:177-187. Thompson IP, Bailey MJ, Fenlon JS, Fermor TR, Lilley AK, Lynch JM, McCormack PJ, McQuilken MP, Purdy KJ, Rainey PB & Whipps JM. 1993. Quantitative and qualitative seasonal changes in the microbial community from the phyllosphere of sugar beet (Beta vulgaris). Plant and Soil 150:172- 191. Tiedje JM, Asuming-Brempong S, Niisslein K, Marsh TL & Flynn SJ. 1999. Opening the black box of soil microbial diversity. Applied Soil Ecology 13:109- 122. Toshikazu T. 1997. Color atlas of turfgrass diseases: disease characteristics and control. John Wiley and Sons. Cranford, New Jersey, USA. Tredway LP & Butler EL. 2004. Preventative control of dollar spot in creeping bentgrass, 2004. http://www.turffiles.ncsu.edu/extension/tredwaytrials/2004/dollar%20spot%20 main.pdf. Vardavakis E. 1988. Seasonal fluctuation of non-pParasitic mycoflora associated with living leaves of Cistus incanus, Arbutus unedo and Quercus coccifera. Mycologia 80:200-210. Vargas JM. 2005. Management of turfgrass diseases. Third Edition. John Wiley and Sons. Cranford, New Jersey, USA Vargas JM, Golembiewski R & Detweiler AR. 1992. Dollar spot resistance to DMI fungicides. Golf Course Management 60:50-54. Vargas JM & Powell JF. 1997. Mycelial compatibility and systematics of Sclerotinia homoeocarpa. Phytopathology 87:S79. Vincelli P. 2002. APS Education Center - Advanced topics - Qol (strobilurin) fungicides: benefits and risks. The Plant Health Instructor. DOI: 10.1094/PHI- I-2002-0809-02. http://www.apsnet.org/education/AdvancedPlantPath/Topics/Strobilurin/

217 Vincelli P. 2007. Chemical control of turfgrass diseases 2008. University of Kentucky Cooperative Extension Service, Lexington, KY. www.ca.uky.edu/agc/pubs/ppa/ppa1/ppa1.pdf Vincelli P & Dixon E. 2002. Resistance to Qol (strobilurin-like) fungicides in isolates of Pyriculaha grisea from perennial ryegrass. Plant Disease 86:235- 240. Walsh B, Ikeda SS & Boland GJ. 1999. Biology and management of dollar spot (Sclerotinia homoeocarpa) an important disease of turfgrass. Hortscience 33:13-21. Warren CG, Sanders PL & Cole H. 1974. Sclerotinia homoeocarpa tolerance to benzimidazole configuration fungicides. Phytopathology 64:1139-1142. Warren CG, Sanders PL & Cole H. 1976. Increased severity of Pythium blight associated with use of benzimidazole fungicides on creeping bent grass. Plant Disease Reporter 60:932-935. Weider LJ, Elser JJ, Crease TJ, Mateos M, Cotner JB & Markow TA. 2005. The functional significance of ribosomal (r)DNA variation: impacts on the evolutionary ecology of organisms. Annual Review Ecological Evolution and Systematics 36:219-42 Weisburg WG, Barns SM, Pelletier DA & Lane DJ. 1991. 16S ribosomal DNA amplification for phylogenetic study. Journal of Bacteriology 173:697-703. White JF. 1993. Endophyte-host associations in grasses. XIX. A systematic study of some sympatric species of Epichloe in England. Mycologia 85:444-455. White TJ, Bruns T, Lee S & Taylor J. 1991. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: PCR Protocols, a Guide to Methods and Applications. Edited by Innis MA, Gelfand DH, Sninsky JJ & White TJ. Academic Press. New York, New York, USA. Williams DW, Powell AJ, Vincelli PC & Dougherty CT. 1996. Dollar spot on bentgrass influenced by displacement of leaf surface moisture, nitrogen and clipping removal. Crop Science 36:1304-1309. Wong FP & Midland S. 2004. Fungicide resistant anthracnose: bad news for greens management. Golf Course Management 72:75-80. Wong FP & Wilcox WF. 2001. Comparative physical modes of action of azoxystrobin, mancozeb, and metalaxyl against Plasmopara viticola (grapevine downy mildew). Plant Disease 85:649-656. Wongkaew P, Hanboonsong Y, Sirithorn P, Choosai C, Boonkrong S, Tinnangwattana T, Kitchareonpanya R & Damak S. 1997. Differentiation of phytoplasmas associated with sugarcane and gramineous weed white leaf disease and sugarcane grassy shoot disease by RFLP and sequencing. Theoretical and Applied Genetics 95:660-663.

218 Wu YX & von Tiedemann A. 2001. Physiological effects of azoxystrobin and epoxiconazole on senescence and the oxidative status of wheat. Pesticide Biochemistry and Physiology 71:1-10. Xu Q & Huang B. 2001. Lowering soil temperatures improves creeping bentgrass growth under heat stress. Crop Science 41:1878-1883. Yadav RKP, Karamanoli K & Vokou D. 2005. Bacterial colonization of the phyllosphere of Mediterranean perennial species as influenced by leaf structural and chemical features. Microbial Ecology 50:185-196. Yang CH, Crowley DE, Borneman J & Keen NT. 2001. Microbial phyllosphere populations are more complex than previously realized. Proceedings of the National Academy of Sciences of the United States of America 98:3889-3894. Yap IV & Nelson RJ. 1996. Winboot: a program for performing bootstrap analysis of binary data to determine the confidence limits of UPGMA-based dendrograms. IRRI Discussion Paper Series No. 14. Manila, Philippines: International Rice Research Institute. Yonath A & Franceschi F. 1999. Ribosomes. In: Encyclopedia of Molecular Biology. Volume 4. Edited by Creighton TE. Wiley-lnterScience Publications. John Wiley & Sons. New York, New York, USA. Yu W & Liang-Dong G. 2004. Morphological and molecular identification of an endophytic fungus Epicoccum nigrum. Mycosystema 23:474-479.

219 LIST OF APPENDICES

Appendix A. Sample SAS statement for comparing plant injury or dollar spot incidence averages between azoxystrobin-treated and untreated plots. date: date of sample (yyyymmdd) loc: plot location (PG: pathology green, NG: native sand fairway) plot: plot replicate (1 to 4) treat: treatment (Water check or Heritage fungicide) injury: plant injury percentage dspot: dollar spot incidence

data temp; infile cards; input date $ loc $ plot treat $ injury dspot; cards; 20061214 PG 1 Water 20 0 20061214 PG 2 Herit 5 0 20061214 PG 3 Water 10 0 20061214 PG 4 Herit 5 0 20061214 PG 5 Water 3 0 20061214 PG 6 Herit 3 0 20061214 PG 7 Water 5 0 20061214 PG 8 Herit 5 0 20061214 NG 1 Water 15 0 20061214 NG 2 Herit 15 0 20061214 NG 3 Water 10 0 20061214 NG 4 Herit 7 0 20061214 NG 5 Water 13 0 20061214 NG 6 Herit 7 0 20061214 NG 7 Water 5 0 20061214 NG 8 Herit 5 0 run; data temp; set; run; proc sort; by loc date / proc glm; by loc date; class treat; model dspot = treat ; means treat/ LSD lines ; run; proc means maxdec=3 data=temp vardef= n mean stderr min max; class loc treat date; var dspot; run;

220 Appendix B. Maintenance schedule for the A stolonifera native soil fairway (B1), the pathology green (B2) and the Poa pratensis fringe (B3).

B1 Native soil fairway Mowing height: 10 mm (0.394 inch) Mowing frequency: Mondays, Wednesday and Fridays Clips: collected at each mowing

2006 19 May 2006 Fertilized with Scotts Pro Turf 21-3-21 at 25 kg N/ha 31 May 2006 Fertilized with Country Club 18-3-18 at 25 kg N/ha 13 Jun 2006 Fertilized with Country Club 18-3-18 at 35 kg N/ha 29 Jun 2006 Fertilized with Country Club 18-3-18 at 35 kg N/ha 05 July 2006 Topdressed with 1 cm of non-calcareous USGA sand 05 July 2006 Aerated using Vk inch hollow tines (1.27 cm) at 2 inch centres (5.08 cm) 17 July 2006 Fertilized with Country Club 18-3-18 at 35 kg N/ha 19 July 2006 Sprayed for Cutworms with Success 480 S at 1.0 ml/100 m2 08 Aug 2006 Fertilized with Country Club 18-3-18 at 35 kg N/ha 30 Aug 2006 Sprayed west section for Dollar Spot with Daconil 2787 at 250 ml/100 m2 (Not done on plots) 05 Sep 2006 Fertilized with Country Club 18-3-18 at 50 kg N/ha 07 Sep 2006 Topdressed using 1 mm non-calcareous USGA sand 21 Sep 2006 Sprayed west section for Dollar Spot with Daconil 2787 at 250 ml/100 m2 (Not done on plots) 27 Sep 2006 Topdressed using 1 mm non-calcareous USGA sand 09 Nov 2006 Fertilized with Country Club 18-3-18 at 50 kg N/ha 2007 09 May 2007 Fertilized with Country Club 18-3-18 at 50 kg N/ha 01 Jun 2007 Fertilized with Country Club 18-3-18 at 50 kg N/ha 05 Jul 2007 Fertilized with Country Club 18-3-18 at 50 kg N/ha 08 Aug 2007 Fertilized with Country Club 18-3-18 at 50 kg N/ha

221 Appendix B. cont.

B2 Pathology green Mowing height: 5.5 mm (0.217 inch) Mowing frequency: Daily from Monday to Friday Clips: collected at each mowing

2006 19 May 2006 Fertilized with Scotts Pro Turf 21-3-21 at 25 kg N/ha 31 May 2006 Fertilized with Country Club 18-3-18 at 25 kg N/ha 13 Jun 2006 Fertilized with Country Club 18-3-18 at 35 kg N/ha 29 Jun 2006 Fertilized with Country Club 18-3-18 at 35 kg N/ha 11 Jul 2006 Topdressed east 14 with 1 cm non-calcareous USGA sand 11 Jul 2006 Aerated East 14 with 14 inch hollow tines at 2 inch centres 13 Jul 2006 Topdressed west 14 with 1 cm calcareous USGA sand 13 Jul 2006 Aerated west 14 with 14 inch hollow tines (1.27 cm) at 2 inch centres (5.08 cm) 17 Jul 2006 Fertilized with Country Club 18-3-18 at 35 kg N/ha 19 Jul 2006 Sprayed for Cutworms with Success 480 s at 1.0 ml/100 m2 (Not done on plots) 08 Aug 2006 Fertilized with Country Club 18-3-18 at 35 kg N/ha 05 Sep 2006 Fertilized with Country Club 18-3-18 at 50 kg N/ha 07 Sep 2006 Topdressed east 14 with 1 mm non-calcareous USGA sand Topdressed west 14 with 1 mm calcareous USGA sand 27 Sep 2006 Topdressed east 14 with 1 mm non-calcareous USGA sand Topdressed west Vz with 1 mm calcareous USGA sand 09 Nov 2006 Fertilized with Country Club 18-3-18 at 50 kg N/ha 2007 09 May 2007 Fertilized with Country Club 18-3-18 at 50 kg N/ha 01 Jun 2007 Fertilized with Country Club 18-3-18 at 50 kg N/ha 05 Jul 2007 Fertilized with Country Club 18-3-18 at 50 kg N/ha 08 Aug 2007 Fertilized with Country Club 18-3-18 at 50 kg N/ha 05 Sep 2007 Urea Micro-Prill 46-0-0 at 12.5 kg N/ha 19 Sep 2007 Urea Micro-Prill 46-0-0 at 12.5 kg N/ha

222 Appendix B. cont.

B3 Poa pratensis fringe

Mowing height: 38.1 mm (1.5 inch) Mowing frequency: Daily from Monday to Friday Clips: not collected 2006 30 May 2006 Agromart 25-4-10 at 40 kg N/ha 08 Aug 2006 Agromart 25-4-10 at 40 kg N/ha 05 Sep 2006 Agromart 25-4-10 at 40 kg N/ha 09 Nov 2006 Agromart 25-4-10 at 40 kg N/ha

223 Appendix C. SAS statements used to test homoscedascity and normality of distribution. date: date of sample (yyyymmdd) trmt: treatment azoxystrobin or untreated plot: plot replicate (1 to 4) sample: replicate within plot (1 or 2) host: turfgrass species (Agrostis stolonifera or Poa pratensis) organism: microbial group (filamentous fungi, yeast or bacteria) morph: morphotype plate: plate replicate (1 to 3) cfu_g: colony forming units per gram of dry leaves

Homoscedascity data temp; infile 'data05-06.prn'; options pagesize=200 linesize=90; input DATE $ trmt $ PLOT SAMPLE HOST $ LEAF $ ORGANISM $ morph $ PLATE $ CFU_G; data temp; set; logcfu=loglO(cfu_g+l) ; run;

PROC SORT; by organism date leaf; PROC GLM; by organism date leaf; class trmt; model logcfu cfu_g= trmt; Means trmt/hovtest=bartlett; run;

Normality of distribution data temp; infile 'data05-06.prn'; options pagesize=200 linesize=90; input DATE $ trmt $ PLOT SAMPLE HOST $ LEAF $ temp $ ORGANISM $ morph $ PLATE $ CFU_G; data temp; set; logcfu=loglO(cfu_g+l); run;

PROC SORT; by date leaf organism trmt; PROC UNIVARIATE PLOT NORMAL; by date leaf organism trmt; VAR CFU_G; VAR logcfu; run;

224 Appendix D. Normality of distribution based on the Shapiro-Wilk test counts of foliar microbial populations using non-transformed and log-transformed CFU/g of dry leaf. 0 = not normally distributed and 1 =normally distributed. Distribution Organism Date Leaf Treatment Log transformed Not transformed Bacteria 2006-03-31 green untreated 0 0 Bacteria 2006-03-31 green azoxystrobin 0 0 Bacteria 2006-03-31 yellow untreated 0 0 Bacteria 2006-03-31 yellow azoxystrobin 0 0 Bacteria 2006-05-06 green untreated 0 0 Bacteria 2006-05-06 green azoxystrobin 0 0 Bacteria 2006-05-06 yellow untreated 0 0 Bacteria 2006-05-06 yellow azoxystrobin 0 0 Bacteria 2006-06-16 green untreated 0 1 Bacteria 2006-06-16 green azoxystrobin 0 0 Bacteria 2006-06-16 yellow untreated 0 0 Bacteria 2006-06-16 yellow azoxystrobin 0 0 Bacteria 2006-07-13 green untreated 0 0 Bacteria 2006-07-13 green azoxystrobin 0 0 Bacteria 2006-07-13 yellow untreated 0 0 Bacteria 2006-07-13 yellow azoxystrobin 0 0 Bacteria 2006-08-15 green untreated 1 0 Bacteria 2006-08-15 green azoxystrobin 1 0 Bacteria 2006-08-15 yellow untreated 0 0 Bacteria 2006-08-15 yellow azoxystrobin 0 0 Bacteria 2006-09-09 green untreated 0 0 Bacteria 2006-09-09 green azoxystrobin 0 0 Bacteria 2006-09-09 yellow untreated 0 0 Bacteria 2006-09-09 yellow azoxystrobin 0 0 Bacteria 2006-09-28 green untreated 0 0 Bacteria 2006-09-28 green azoxystrobin 0 0 Bacteria 2006-09-28 yellow untreated 0 0 Bacteria 2006-09-28 yellow azoxystrobin 0 0 Bacteria 2006-11-23 green untreated 0 0 Bacteria 2006-11-23 green azoxystrobin 0 0 Bacteria 2006-11-23 yellow untreated 0 1 Bacteria 2007-03-27 green untreated 0 0 Bacteria 2007-03-27 green azoxystrobin 0 0 Bacteria 2007-03-27 yellow untreated 0 0 Bacteria 2007-03-27 yellow azoxystrobin 0 0 Bacteria 2007-05-24 green untreated 0 0 Bacteria 2007-05-24 green azoxystrobin 0 0 Bacteria 2007-05-24 yellow untreated 0 0 Bacteria 2007-05-24 yellow azoxystrobin 1 0 Normal

225 OT-OOOOOOOOOT-OO i- O O OOOOT-T-OO o E

c o 3

Ot-O-r-OOOOr-OO OT-I-OOOOOOI-OT-O OOT-OOT-T-OO i

CV

C c c c C g c c C C bi n

bi n bi n bi n bi n bi n CD X! -Q -Q jo. bi n !5 lo JQ In CM TJ TJ P TJ p xi p 73 O TJ p TJ P TJ P TJ o TJ y TJ p TJ P TJ P TJ O TJ TJ TJ TJ p TJ P TJ P TJ 4=

tre a tre a tre a x P trea t trea t trea t oxy s oxy s IIII o 4= 1_ l_ L- £ o 43 d •£= >< P o o o C5 o o o •*-> $ p **^ *->•*-> O m "C N "C c N t_ N c N C N C O i= N C N c N c N c N ~c N ~c N c C C C N N C 3 CO 3 (0 3 (0 3 CO 3 O 4= N C TO 3 CO 3 TO 3 TO 3 CO 3 TO 3 CO 3 O 4= 3 3 3 CO 3 CO 3 CO 3 81 N C TO 3 N C (0 3 TO 3 cc55ccccc oa(DCDOOCDCUCDCDC) = =o(D(DaiaD) LUQ)IDi-LLL.L III! llfilllf D)D)>>>>C3)D)0)C3)D) itfiiffiiiiitiitliifi. (35 CJ5 05 O) T- T- T" CD CD CD CD CD 05 <35 05 CD CO CO CO to co LO m LO en C35 CO CO CO CO N N N |-~ •si" ^t CM CM CM CM CO CO CO O O O O -- T- T- T~ t- T- O O CM CM CM CM CM CM CM CM CM CM i i CO CO CO CO I I I I i i i i CO CO CO CO 00 05 I I I I• i i i i t i i CO CO CO IT) LO m m CD o O p 05 (35 05 TT- — T- CO CO CO CO LO LO LO LO O Q O O m LO LO LO o o o o o o o o o o CD CD CO CO CO CD CO CD o o o o oCD CoD CoO vC-O Tco— i-O- O O o o o C ill i i i • i o o o o I I I I I I I o o o o o o o o I I I I o o o o o o o o o CD CO CD CD CD CD CD CD CD CD CD CO o o o o o o o o o o o o o o o o I"- h- h- o CO CD CD CD O O O O O O O O o o o o CO CD CO CD o o o O O O O O O O O O O O O o o o o CM CM CM CM CM CM CM CM o o o o CM CM CN CM CM CM CM CM o o o CM CM o o o o CM CM CM CM o o o o CM CM CM O CM CM CM CN CM CM CM CM CM CM CM CM CM CM CM CM X E '•& .jo C 'E CD co OJCJ5CJ)a)CJ)CJ)0)C7)CJ)CJ)CJ5CnDJCJ)CJ>a5CJ5C3)CJJC35CJ)a5rj5CJ505C35D)C3) D5C3)CJ)O)C3)D)CJ)D)C3)C3)D)C3)0)C3) CCCCCCCCCCCCCC Q. cccccccccccccccccccccccccccc 3333333333333 3 Q. 3333333333333333333333333333 LL.U-LL.LL.Ll-Li_Ll-lJLLL.LL.LL.LL.LI.LL. < U.U.LUU-Ll.li.LJ.LL.U.lJ.U.LULl.Ll_LLlJ.U.U.U.U.LUU.LL.li.Ll.U.LL.ll. oooooooooooooooooooooooooooooooooooooooooooo

co 3 JO oooooooooooooooooooooooooooooooooooooooooooo (0 5

O

CM CM

0)0)010) CD CO CD CD OCnoCDCOCDCOCO I^-- 1^- 1^- h~ lO 0) CJ) CJ) CO 00 •sr "fr <* CO CO CO CO CM CM CM CM T- in in m o n n N s s s 0)05 0)0)05 T- T- co co co co in o o o o o CD CO CO CO CD CD CD CO CD o o o o o CO CD 1^ K 1^- h- o o o o o o o o o o o o o o o I I I I I T- T- O O O O O I I I I I o o o o c I I I I I I I o o o o o o o o o I I I I CO CO CO CO CO o o o CO CO CO CO CoO coo coo co co co co co oCO CoO CoD CoO o o o o o o o o o CO CO CD CO o o o o o o o o o o o o o o o o o o o o o o —o o CM CM CM CM CM CM CM CM CN o o o o oCMCMCMCMCMCMCMCMCMCMCMCMCMC o o ~ M CMCMCMCMCMCMCMCMCMCMCMCo o o o ~ ~N o o o o o o o o d CM CM CM CM CM CM CM CM X E .!2 'E c (0 CO(/>(/}(/)U)(/>(/>(/)C0(/>(/)(/)(/>(/)(/)(/>(/)U)(/>t/>(/)(/>(/)(0C/)ID0O(DDIDIlllDIDID(DIUIUIDIll(UIDIDID(DIDOIll Q. < Appendix E. Homogeneity of variance using Bartlett's test of counts of microbial populations using non-transformed and log-transformed CFU/g of dry leaf. Probability values greater than 0.05 then the variance is homogeneous. Probability Organism Date Leaf Log CFU/g CFU/g Bacteria 2006-•05-06 green 0.0001 0.0001 Bacteria 2006 •06-16 green 0.0001 0.0001 Bacteria 2006-•03-31 green 0.0006 0.0001 Bacteria 2006-•05-06 yellow 0.4903 0.0001 Bacteria 2006-•03-31 yellow 0.8266 0.0001 Bacteria 2006-09-28 green 0.8973 0.0001 Bacteria 2006-•09-09 yellow 0.9037 0.0001 Bacteria 2006-05-19 green 0.4119 0.0008 Bacteria 2006-•08-15 green 0.1428 0.0026 Bacteria 2006-05-19 yellow 0.0331 0.0034 Bacteria 2006-09-09 green 0.0002 0.0102 Bacteria 2006-08-15 yellow 0.0001 0.0648 Bacteria 2007-03-27 green 0.0001 Bacteria 2007-03-27 yellow 0.0621 Bacteria 2007-05-24 yellow 0.4343 Bacteria 2007-05-24 green 0.6285 Bacteria 2006-07-13 green Bacteria 2006- 11-23 green Bacteria 2006- 11-23 yellow Fungi 2006-•05-06 green 0.0001 0.0002 Fungi 2006-•07-27 green 0.0001 0.0078 Fungi 2007-•03-27 green 0.0001 0.2957 Fungi 2006-•03-19 yellow 0.0042 0.0016 Fungi 2006-•08-15 yellow 0.0082 0.1245 Fungi 2006- 09-28 green 0.0129 0.1474 Fungi 2006-•03-31 green 0.0313 0.5818 Fungi 2007- 05-24 green 0.0334 0.0378 Fungi 2006- 06-16 green 0.0393 0.331 Fungi 2007- 05-24 yellow 0.0693 0.0117 Fungi 2006-•05-19 yellow 0.0906 0.0044 Fungi 2006- 07-27 yellow 0.1976 0.1088 Fungi 2006- 03-31 yellow 0.2008 0.026 Fungi 2006- 09-09 green 0.3471 0.4196 Fungi 2006- 05-06 yellow 0.5323 0.0118 Fungi 2007- 03-27 yellow 0.5975 0.0708 Fungi 2006- 08-15 green 0.6691 0.1511 Fungi 2006- 03-19 green 0.8849 0.0149 Fungi 2006- 05-19 green 0.8936 0.2669 Fungi 2006- 11-23 green 0.0001 Fungi 2006- 11-23 yellow 0.3401 Fungi 2006- 07-13 green

228 <<<<<<<<<<<<<<<<<<<<<<<< (D(D(D(D(Dn(D(D(D(D(D(D(D(1)(D(D0(D(1)(D(Dn)(D(D Q)Q)Q)£DQ)Q)QJQ)Q)Q)Q)£D£DQ)£DQ)Q}Q)CBQ)!B£DQJCDlO 3

I

IO IO IO M IO IO ro ro IO N) ro ro to ro ro ro ro ro ro ro ro ro ro ro O O O O O O o o O O o o o o o o o o o o o o o o O O O O O O o o O O o o o o o o o o o o o o o o O O) G5 0) -J O) OJ at O) O) O) O) --J O) 0> O) 05 -NI ^J O) i i i i I I o o oi io o• io o o o o o o o o - o o en co CD -»J o o o o o -J CD • i I ( I I CO CO o o i i _»00. Ool 0->0• -»O)• CrO o en co en en en co -x -J. (0 IO • i o ro ro co IO IO O) CD en a> en en -^ i i i i i i WW WW CD ~>l co -»• co ->. ro ro -»• -J 00 co ->• co *• -t». co ,<'0(DOOCDOCDOO(DOCDO CO

o oooooooo oooooooooo co^^^cDcncDc_ . - _. _n. ProM^^bbbbbb o OCDOOIOSClMB^OJNA-liOOOOOO (DOIOIC-'OOffl^-'AOCn^OOOOOOO Tl eoa)cocn-».-».a5~jjs.oo*.enco-t».~JOo^-v-i-^ c no"

oooo pppoopopoo o o p p o p o ooocn^cnoo-^^obioooP o o Ko o o o •n O O CD O O O -is|OI0A*OJi>Uti)OOOOO(£) o o -t>. o o o c C005-iO)--JCO-»-->.-vlCO->.-s.rO-»--»-4!>. _i _>. co -* -v ->. iS" Appendix F. SAS statements for assessing differences between microbial populations using ANOVA analysis date: date of sample (yyyymmdd) trmt: treatment azoxystrobin or untreated plot: plot replicate (1 to 4) sample: replicate within plot (1 or 2) host: turfgrass species (Agrostis stolonifera or Poa pratensis) organism: microbial group (filamentous fungi, yeast or bacteria) morph: morphotype plate: plate replicate (1 to 3) cfu_g: colony forming units per gram of dry leaves data temp; infile 'data06-07.prn'; input DATE $ trmt $ PLOT SAMPLE HOST $ LEAF $ ORGANISM $ morph $ PLATE $ CFU_G; data temp; set; logcfu = loglO(CFU_G+1); run;

PROC SORT; by host leaf organism morph; PROC GLM; by host leaf organism morph; class trmt ; model cfu_g logcfu = trmt ; means trmt / lsd lines; run;

PROC SORT; by trmt host leaf organism morph; PR?OC UNIVARIATE noprint; by trmt host leaf organism morph; var cfu_g; output out=b mean=means stderr=standerr n=num; /* create set b with mean and SD for x only */ run; proc print; run;

230 Appendix G. Selection of 16S restriction enzymes based on in silico analyses with GeneRunner of 15 restriction enzymes with 38 common foliar bacteria. Chosen for Enzyme tests Recognition site #Cuts Hpall (Mspl) Yes CCGG/GGCC 6-8

Haelll (BsuRI) Yes GCCC/CCGG 5-8

Alul Yes AGCTVTCGA 4-7

Cfol Yes GCGC/CGCG 4-5

Rsal Yes GTAC/CATG 3-4

Hindll Yes GTYRAC/CARYTR 2

EcoRI Yes GAATTC/CTTAAG 2

Kpnl No GGTACC/CCATGG 1

Pstl No CTGCAG/GACGTC 1

Xbal No TCTAGA/AGATCT 1

BamHI No GGATCC/CCTAGG 0

Eco32l No GATATC/CTATAG 0

Pvul No CGATCG/GCTAGC 0

Pvull No CAGCTG/GTCGAC 0

Xhol No CTCGAG/GAGCTC 0

R = G A (purine)

Y = T C (pyrimidine)

231 Appendix H. Sequences of 16S selected from GenBank which were used for in silico analyses. These 45 species were chosen based on previous reports which depicted them as members of various phyllospheres. Accessio Species Accession Species n number number Methylobacterium Acromobacter sp. AB196248 isbiliense AJ888241 Pantoea Arthrobactersp. AJ893515 agglomerans Z96083 Pseudomonas Azotobacter vinelandii L40329 cannabina AY880218 Pseudomonas Bacillus subtilis AB257199 cedrina AF064461 Pseudomonas Beijerinckia sp. AB119212 congelans AJ492828 Clavibacter Pseudomonas michiganensis X77434 hodesiae AY043360 Pseudomonas Corynebacterium sp. E10982 poae AJ492829 Pseudomonas Corynebacterium sp. AY914072 synxantha D84025 Pseudomonas Curtobacterium albidum AM042692 tremae AJ492826 Pseudomonas Curtobacterium citreum X77436 trivialis AJ492831 Pseudomonas Curtobacterium luteum X77437 veronii AB056120 Curtobacterium pusillum AJ784400 Ralstonia sp. DQ294749 Rathayibacter Curtobacterium sp. AJ967025 iranicus U96184 Rathayibacter Escherichia coli AM 184252 rathayi D45062 Enterococcus Rathayibacter casseliflavus AJ420804 toxicus D84127 Enterococcus faecium AM 157434 Rathayibacter tritici X77438 Enterococcus sulfureus X55133 Sphingomonas sp. Y15514 Staphylococcus Enterococcus faecalis DQ411814 sciuri S83569 Enterococcus mundtii AJ420806 Stenotrophomonas AB196256 Xanthomonas Erwinia amilovorans AJ746202 campestris AJ811695 Flexibacter tractuosus AB078076 Listeria seeligeri X56148 Micrococcus luteus AJ536198

232 Appendix I. Top matches of bacterial 16S sequence of turfgrass bacteria. rDNA amplified with primers FO and PRUN518. Sequences compared with nr database using megaBLAST. Top matches had annotated genera and species, and identities matched over 80%. The plots on the pathology green were not treated or treated with Heritage® 12 g/100 m2.

Collection Leaf Treatment Submorphotype Top Match Score E Value Identities % 23-Nov-2006 Green Untreated Orange Janthinobacterium sp. 848 0.00 439/443 99 23-Nov-2006 Green Untreated Orange slimy Escherichia coli 537 2x10"149 381/436 87 23-NOV-2006 Green Untreated Orange slimy Sphingomonas sp. 811 0.00 412/413 99 23-Nov-2006 Green Untreated Pink Xanthomonas campestris 910 0.00 461/462 99 23-Nov-2006 Green Untreated Pink Pedobacter wanjuense 674 0.00 395/416 94 23-Nov-2006 Green Untreated Pink Bacteroidetes bacterium 741 0.00 432/452 95 23-Nov-2006 Green Untreated Pink Pedobacter aquatilis 98 0.00 443/450 98 23-Nov-2006 Green Untreated White slimy Xanthomonas translucens 912 0.00 460/460 100 23-NOV-2006 Green Untreated Yellow Agreia bicolorata 690 0.00 378/393 96 23-Nov-2006 Green Untreated Yellow Sphingomonas sp. 799 0.00 403/403 100 23-NOV-2006 Green Untreated Yellow Escherichia coli 819 0.00 443/455 97 23-Nov-2006 Green Untreated Yellow Dyella yeojuensis 690 0.00 399/422 94 23-Nov-2006 Green Untreated Yellow Pedobacter roseus 803 0.00 413/421 98 23-Nov-2006 Green Untreated Yellow Sphingomonas sp. 188 10"44 160/192 83 23-Nov-2006 Yellow Untreated Orange Agrobacterium tumefaciens 783 0.00 404/407 99 23-NOV-2006 Yellow Untreated Orange Janthinobacterium sp. 781 0.00 417/433 95 23-Nov-2006 Yellow Untreated Orange Chryseobacterium proteolyticum 622 5x10-175 352/366 96 23-Nov-2006 Yellow Untreated Pink Pedobacter suwonensis 819 0.00 413/413 100 23-Nov-2006 Yellow Untreated Pink Pedobacter roseus 753 0.00 404/414 97 23-Nov-2006 Yellow Untreated Pink Pedobacter roseus 662 0.00 372/389 95 23-Nov-2006 Yellow Untreated Pink Pedobacter roseus 722 0.00 385/393 97 23-Nov-2006 Yellow Untreated Pink Bacteroidetes bacterium 761 0.00 438/456 96 23-NOV-2006 Yellow Untreated White slimy Pseudomonas sp. 539 1(j-1S0 347/361 91 23-Nov-2006 Yellow Untreated White slimy Agrobacterium tumefaciens 773 0.00 404/409 98 free

N> IO ro ro ro IO M N> M IO ro IO ro ro ro ro ro 10 ro ro ro ro ro ro ro ro •si •si -s| •si •s| •si •si -si -si •si •si •si -si -si ^1 CO CO CO CO CO o f f* t f* • *• o. S 2 2 Z Z ^ % k S S k s k 3 k k k k k k k z z (0 £W .9£2 £F 3 CD £ CD 01 01 01 0) 01 03 01 oi 0) 01 0) 01 01 Q] 0 0 0 0 0 *< *< •< >< 7 7 7 7 •^ =; 7 7 7 7 7 7 7 -1 7 < < < < f f M lb IO IO IO IO ho ro 10 ro ro 10 ro ro ro ro ro ro ro fro o o o o o O o O O O o O o o o 0 0 0 0 0 0 0 0 0 0 0 0 o o o o O o O O O o O o o o 0 0 0 0 0 0 0 0 0 0 0 0 0' J ^J sj ^1 sj -si -si •si •si -si -s| --i -s| 0 05 05 "^ "^ - •^ "^ •^ "^ •^ ^i en o> 3

o o Q o O O -< -< -< -< •< -< -< O O 0 0 O 0 O 0 -< •< -< •< -< -3 (D CD CD CD CD CD CD -3 •^ -1 3 -1 CD CD CD CD C-^D CD CD CD C^D CD CD C-^D CD C—\D cp_ cp_ cp_ cp_ cp_ (D (D 0 CD CD CD O O O O= O= O= CD CD CD CD CD CD CD CD 0" 0" 0" 3 Q 3 3 3 3 3 3 5" o" 3 3 3 3 3 S § § S> £ £ S> 3 3 S € S S S

H H H H -i H c c c H H H H c c C -i -H -H H H C c C c c 3 13 -1 3 3 3 —1 "1 -1 3 3 3 3 3 3 CD CD CD CD 3 3 3 3 3 3 i—»- 3 3 3 «-»- .-•• 3 CD CD 0) 0) 0) 0) 0) !-•73• 3 0) 0) 01 0) 3 3 0) 0) 0) 0) 0) —1 3 S- S3- r+ K- CD C-^D CD K- r—- S- CD CD C-^D 1-* r-t- CD C-^D CD C-^D C-^D =•• s- st »-»- ^* ^* CD CD CD CD ite d CD CD CD ite d ite d CD CD CD CD CD ite d CD CD 01 0) CD 0) 0) CD 01 0) 0) 0) 0te d ) Q. a> Q. Q. Q. Q. Q. Q. Q. Q. a. a. a. a. CD CD a. a. CD CD ffi" CD Q. a. Q. O. Yel l Yel l Yel l Yel l Yel l Yel l Yel l Yel l Yel l Yel l Yel l Yel l Yel l Yel l Or a Or a Pin l Pin l Pin l O O O 3- i$ 5! 0) 0) 3 it e sl i 0) 3 ?\ 7\ O 0 3 0 3-* 3 3 3 0 0 0 0 o o CD ^ o o o o o c S to € S !» £ S :> § CO S CD CO CO ca S € g cr w CD CD CD CD CD 0) im y o =r. wlim y =: •a 3- 1 1 o

CD

"0 CO DO CO 1 CO CO ii 03 CO 0 D O N - S S n o o o o 0) cB Tella Creia O %% •°>° & 3" 2 O CD CD Q- 3" 3- O" 3- 3" c c 2 to' 3 ^ CQ c? 3' 3' Co o. S. o llcf^lli^i^ii01 01 n 9- 3- o ST 3 I I' eg § CQ CQ CQ CQ % I"CD 0" o> cu ^ 0- O O O O o o o B: CD CD o >^- o o I §• I m 3 3 §-1

Lien i a 01 CD Cr CP crT cB" CD 3- fi lora o o 01 iS o c c * O X s c a, 0) ? M 3 3 cB=•" 01 3 T3 3 3 0. ' o 0> CD CO 3 •Q CD CD 01 01 01 cF «o o Co 3 c c' 3 3 Co S 2 Co CO o o C3- o C C/1 3 ? a- ca- M cB" 05 3 CO CO CO U3 CO CD c/> CO CO CD CD 01 CO- Co' 5' 5" £2. S CD 3 3 01 T3 •o a •a -a 3 » 0> Qi - CO c c o cB 00 01 is- o CO' 2 3 CQ 01 2 03 01 CD 3 c c O O 3 3 O O 5T oT 3 S' 2 C C o o I o 3_ 3 3 CD' CD' 3- S' 3 3 c o 3* oT Co Co 3 c o o CD' CD' 3 3 Co co Hi OOmOlCOCOOOslslOOOOOOOlOOCDOJOOOIOlClOOOs! 00 00 00 CD -N! o cni^ooco-sicD-sioocncncnioroocoococxioo^i -^ ro ro -si a) Ul -i s| 00 Ol o OKffl^MJiS-iCB01oai-»^01

m p op pp pop opppppppppppp OOOOO 000000000000000000000 O O 000 < 000000000000000000000 o o 000 c CD

CO Ji. Ji. Ji. Ji. Ji. 4i. W js. Js. CO *. Js. -p* CO CO CO on en ro -si ro CO 00 O ->• CO ro en —x 00 CO ^1 00 O) co ^en roro c0o1 00 00 M ^ CO 0 0 -si -J. 4i -si M Ol Ol Ol en •fcv £ 5 5 5 i"s }i 5 4i ji 4*. Ji •li. CO CO CO en co en ro -si en ro Ji it CO 5 s 5 o CJ) —X CO CO •si O O -1 ro 00 03 CD COOOOl^lOO^l-slO00 Oco CD en 00 4i en -si 0 00

COCOCOCOCOCOCOCOCOCOCOCOCQCOCDCDCOCOOCOCO CO CO O CD CO cocDcnoococosjcDoooooococooocDcooocnocooo CO O) O -si CO Appendix I. cont. Collection Leaf Treatment Submorphotype Top Match Score E Value Identities % 24-May-2007 Green Treated Yellow slimy Curtobacterium flaccumfaciens 850 0.00 443/447 99 24-May-2007 Green Untreated Orange slimy Curtobacterium herbarum 882 0.00 448/449 99 24-May-2007 Green Untreated Pink Curtobacterium flaccumfaciens 860 0.00 433/434 99 24-May-2007 Green Untreated White Pseudomonas abietaniphila 773 0.00 446/472 93 24-May-2007 Green Untreated White Pseudomonas syringae 811 0.00 409/409 100 24-May-2007 Green Untreated Yellow Flavobacterium sp. 803 0.00 417/421 99 24-May-2007 Yellow Treated Orange Sphingomonas sp. 765 0.00 409/418 97 24-May-2007 Yellow Treated Orange Brevundimonas nasdae 795 0.00 401/409 98 24-May-2007 Yellow Treated Orange slimy Curtobacterium flaccumfaciens 882 0.00 448/449 99 24-May-2007 Yellow Treated Orange slimy Shigella flexneri 809 0.00 445/465 95 24-May-2007 Yellow Treated Pink Pedobacter aurantiacus 733 0.00 442/454 97 24-May-2007 Yellow Treated Pink Pedobacter koreensis 718 0.00 438/450 96 24-May-2007 Yellow Treated White Agrobacterium tumefaciens 533 2x10"148 322/365 89 24-May-2007 Yellow Treated White Pseudomonas corrugata 884 0.00 455/461 98 24-May-2007 Yellow Treated Yellow Agreia bicolorata 856 0.00 434/435 99 24-May-2007 Yellow Treated Yellow Subtercola pratensis 866 0.00 437/437 100 24-May-2007 Yellow Untreated Orange Rhodococcus fascians 829 0.00 427/434 97 24-May-2007 Yellow Untreated Orange slimy Sphingomonas sp. 797 0.00 404/405 99 24-May-2007 Yellow Untreated Orange slimy Pedobacter wanjuense 684 0.00 430/454 93 24-May-2007 Yellow Untreated Pink Pedobacter wanjuense 694 0.00 387/402 95 24-May-2007 Yellow Untreated Pink Pedobacter aurantiacus 729 0.00 439/452 96 24-May-2007 Yellow Untreated Pink Pedobacter wanjuense 726 0.00 424/436 96 24-May-2007 Yellow Untreated White Agrobacterium tumefaciens 785 0.00 414/420 97 24-May-2007 Yellow Untreated White Pseudomonas syringae 807 0.00 407/407 100 24-May-2007 Yellow Untreated Yellow Chryseobacterium aurantiacum 890 0.00 452/453 99 24-May-2007 Yellow Untreated Yellow Agreia bicolorata 872 0.00 440/440 100 24-May-2007 Yellow Untreated Yellow slimy Flavobacterium sp. 839 0.00 443/450 98 Appendix I. cont. Collection Leaf Treatment Morphotype Top Match Score E Value Identities % 12-Jul-2007 Green Treated Orange slimy Microbacterium oleivorans 852 0.00 433/438 97 12-Jul-2007 Green Treated Pink Pedobacter suwonensis 753 0.00 395/399 98 12-Jul-2007 Green Treated Pink Pedobacter aurantiacus 737 0.00 442/456 96 12-Jul-2007 Green Treated White slimy Agrobacterium tumefaciens 785 0.00 402/405 99 12-Jul-2007 Green Treated White slimy Agrobacterium tumefaciens 787 0.00 407/408 99 12-Jul-2007 Green Treated White slimy Pseudomonas putida 856 0.00 452/455 98 12-Jul-2007 Green Treated Yellow Chitinophaga ginsengisoli 761 0.00 447/473 94 12-Jul-2007 Green Treated Yellow slimy Flavobacterium sp. 809 0.00 416/419 99 12-Jul-2007 Green Untreated Orange Rhodococcus fascians 817 0.00 449/453 99 12-Jul-2007 Green Untreated Orange slimy Chryseobacterium daeguense 805 0.00 447/458 97 12-Jul-2007 Green Untreated Orange slimy Escherichia coli 730 0.00 422/447 94 12-Jul-2007 Green Untreated Pink Pedobacter suwonensis 755 0.00 394/396 99 12-Jul-2007 Green Untreated White Chryseobacterium daeguense 785 0.00 456/474 96 12-Jul-2007 Green Untreated White slimy Agrobacterium tumefaciens 765 0.00 414/422 98 12-Jul-2007 Green Untreated White slimy Enterobacter sakazakii 914 0.00 476/479 99 12-Jul-2007 Green Untreated White slimy Escherichia coli 791 0.00 436/444 98 12-Jul-2007 Green Untreated White slimy Agrobacterium tumefaciens 799 0.00 407/408 99 12-Jul-2007 Green Untreated Yellow slimy Janthinobacterium agaricidamnosum 815 0.00 443/455 97 12-Jul-2007 Green Untreated Yellow slimy Arthrobacter ilicis 862 0.00 442/445 99 12-Jul-2007 Green Untreated Yellow slimy Curtobacterium flaccumfaciens 289 2x10-117 296/359 82 12-Jul-2007 Green Untreated Yellow slimy Dyadobacter hamtensis 732 0.00 427/442 96 12-Jul-2007 Yellow Treated Orange Microbacterium imperiale 755 0.00 440/441 99 12-Jul-2007 Yellow Treated Orange slimy Curtobacterium herbarum 878 0.00 446/447 99 12-Jul-2007 Yellow Treated Orange slimy Curtobacterium flaccumfaciens 817 0.00 447/453 98 12-Jul-2007 Yellow Treated Orange slimy Microbacterium oleivorans 800 0.00 440/443 99 12-Jul-2007 Yellow Treated Pink Shigella dysenteriae 835 0.00 447/455 98 12-Jul-2007 Yellow Treated White Escherichia coli 732 0.00 435/469 92 Appendix I. cont. Collection Leaf Treatment Morphotype Top Match Score E Value Identities % 12-Jul-2007 Yellow Treated White slimy Enterobacter sakazakii 890 0.00 459/464 98 12-Jul-2007 Yellow Treated White slimy Enterobacter sakazakii 890 0.00 460/463 99 12-Jul-2007 Yellow Treated White slimy Enterobacter sakazakii 908 0.00 462/464 99 12-Jul-2007 Yellow Treated Yellow Curtobacterium flaccumfaciens 876 0.00 445/446 99 12-Jul-2007 Yellow Treated Yellow Curtobacterium flaccumfaciens 842 0.00 441/447 98 Appendix J. Top matches of yeast ITS sequence of turfgrass yeasts. rDNA amplified with primers ITS1 and ITS4. Sequences compared with nr database using megaBLAST. Top matches had annotated genera and species, and identities matched over 80%. The plots on the pathology green were not treated or treated with Heritage® 12 g/100 m2.

Collection Leaf Treatment Morphotype Top Match Score E Value Identities % 23-Nov-2006 Green Untreated Orange Dioszegia crocea 761 0.00 421/430 97 23-Nov-2006 Green Untreated Pink Caulerpa webbiana 783 0.00 399/401 99 23-Nov-2006 Green Untreated Salmon Sporobolomyces roseus 971 0.00 492/493 99 23-Nov-2006 Green Untreated Salmon Cystofilobasidium macerans 1053 0.00 531/531 100 23-Nov-2006 Green Untreated White Ustilago duriaeana 886 0.00 456/460 99 23-Nov-2006 Green Untreated White Rhodotorula ingeniosa 904 0.00 482/495 97 23-Nov-2006 Green Untreated White Ustilago duriaeana 874 0.00 444/445 99 23-Nov-2006 Green Untreated White Cryptococcus victoriae 811 0.00 413/415 99 23-Nov-2006 Green Untreated White Cryptococcus sp 787 0.00 417/425 98 23-Nov-2006 Green Untreated White slimy Cryptococcus flavescens 825 0.00 420/422 99 23-Nov-2006 Yellow Untreated Pink Rhodotorula graminis 999 0.00 507/508 99 23-Nov-2006 Yellow Untreated Pink Rhodotorula graminis 1007 0.00 508/508 100 23-Nov-2006 Yellow Untreated Salmon Cystofilobasidium macerans 1003 0.00 415/418 99 23-Nov-2006 Yellow Untreated Salmon Sakaguchia dacryoidea 833 0.00 507/530 95 23-Nov-2006 Yellow Untreated Salmon Cystofilobasidium macerans 997 0.00 513/518 99 23-Nov-2006 Yellow Untreated White Rhodotorula ingeniosa 924 0.00 486/496 97 23-Nov-2006 Yellow Untreated White Cryptococcus tephrensis 973 0.00 577/616 93 23-Nov-2006 Yellow Untreated White Cryptococcus paraflavus 728 0.00 381/383 99 23-Nov-2006 Yellow Untreated White Cryptococcus tephrensis 807 0.00 422/428 98 23-Nov-2006 Yellow Untreated White Rhodotorula ingeniosa 902 0.00 484/496 97 23-Nov-2006 Yellow Untreated White Cryptococcus paraflavus 791 0.00 413/415 99 23-Nov-2006 Yellow Untreated White slimy Rhodotorula ingeniosa 894 0.00 477/490 97 27-Mar-2007 Green Treated Pink Rhodotorula graminis 1013 0.00 513/514 99 OOOfflfflOOJOlrr^O) oimo)0)saioino)0)0)oiso)C) 00)0)0)00)QO)QO)0) 05050)0)C3)C500505CJ50)CT)0)C3)a3

O O t- N- co m CM oo oo CM o o o CO CO CD CM o CM O CN OO 00 r^ co r-» o -f r-co o CM O o O O •sr h- o -- h- in m m •

OOOOOOOCM m OOoOOOOoOoOOOOOOOS 1OOOOOOO V "7 OOT-OOOOT-OT-OOOOOOOJ-J > oodciooo^^oo Xx d d d d xX d O x xcicicicicicidT- UJ CM m CD

T-CMt-CDinOT-CDCM^-G T-oooino)CD05CMcoo)oomcDCDin o C0C0CMCOCD^ra)CDCOCM>< tcotocNno(amo)n(otO)(M^n o (ocnooo)oo(Oooro(D(00)0)0)Nrooo co ooojooQoiroro^^Ni;

co CO CO CO CO C co 2 c CO 5 CO 2 CO co so CD CD CD c o c O O O CD CD CD CD CO CO .CO CO CD 2 CD .CO .CO CO CD CD CO CO C C •a-8 Co co co "" O CO Ber nro CO .£ .2 •S E CD CD O C CD .5 .5 •| 9 co 1«P w CD to £ £ Sec C c £ S CD C s s F F F o CO O CD CD I 3 3 cu S ordi 2 5 5 -2 b, CO CD 5= O CD co co .? 11 CO Cn Co CO CO "_ ^S -^ C-» -_ CO co -C o CO CD •c

* sid sid sid CD co CD CD CO CO 3 3 CO CD •S CO '§ CO CO co co 3 CO o 3 CO CO cn O 3 5 o 5 5 73 s 5 S •o -Q •Q o "O ml 2 -S "° p O ^ jD .o O P co •2 .-i I'+** *•»_ «_J ._- W o P. 13 « o O) .2 p p MS H; >*^ n Ol o -2 5 p p 4C^ I p jo -2 ^5 "6 8J 5 .CO T3 P p. CO T3 P T3 O CO Sw *_i -4^ O * '45 € O O a a O O JC co o 2, 2. 2, 1 CO •c CJ DC ucco ;? ;F o o w >» ;? o: co r> 5 o ;P ;? O o o o o LC ££ DC CO O DC aCD ce or >. E E E E •4— o co w c c c cm c w _: C o a. CD CD 2 CD CD 22CDCD2 ±i ±i ^ _*: jsis E ±i ±: >- J— *-* •*-» ^ -#-« •*-* *—' 4-1 <*-* ^ ^ E E o -C-C C C C 'J £ -C •m"m-C-C:TS-C-C:-C-C-C CD S^Q.CLCLC0^5 ww£<:w£££<:£ C.L£ Q.£. C1O5 CCOD CO

T3 T3 TJ TD T3 T3 •D 73 T3 CD CD CD CD CD CD T3 T3 T3 T3

ccccccc:c:c:§555555c:c:c:cccccc:c:c 000000000OOOOOOO00000000000 000000000 ======00000000000 1-1-.—.—.—._._,>-.—.(D0000001-.—.—>—.—._. —.1— —.—.!_ OOU(500000>->->->>->->-OOCI)00[DO(5000 I— F— I— o O O o o c_> c_> C_D oooooooo s o O O o o o^_,_-ooooooo o o o o CM CN CM CM CM CN CM CN CM CM CM CN CM CM CNCNCNCMCMCMCNCNCNCNCM _. d. i i i i i i i i o CO CO CO CD CO CDCOCDCOCOCOCOCOCOCOCO CO CD CO CDCOCOCOCOC0C0C0 S _2 l*-h-N-l»-l^-l*-l>-h.h-h-h-^--l-^rTl--J--r-l--l--l--l--f o CM CM CM CM CMCMCMCMCNCNCNCNCMCNCMCMCMCNCMCMCMCMCMCMCMCN

239 OfrZ

N3N3MMWN)N)WMMWN)N3MN)N)IV)MWrOMN3MMN)NIO o o_ ?5" 03 03 03 03 01 03 03 03 03 m Q) Q) o •< 5" O O O O O O o o o o o o o OOOOOOOOOOO 3 O O O O O O o o o o o o o ooooooooooo -J -vl -g -vl -si ->J -J ^1 ~J ^J -vj ^1 -J

< < < < < < -< CD CD CD CD CD CD ooooooooooo CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD o o o o o o o o CD CD CD CD CD CD CD CD CD CD CD 3 3 13 3 3 3 3 3 3 3 3 5> 5 =e ^

a c c a c a -\ -j-j-j-i-i-i^czciciczczczczczczc:-) rt.3.3.3.2-3.CDCDCDCDCDCDCDCDCDCD 3333333333 CD „_„„,030)030303030303030w„„„__ 3 Q) (UQ)Q3Q)Q)Q3CDCDCDCDroCDCDCDCDCDo)Q3Q)Q)juQjQ3Q)Q)Q)CCDCDCDCDCDCD5j-cJ-sJ-s»H-a-sJ-'S-s»-sfCD0CDCDCDCDCDCDCDCDsDr i-+^ ^ r-* f*> ft- i-#- i-t-. Q~t_ QA . AQ . QA . .Q_ . .-Qt. ^O . QA _ <*^^ Af* > r-fr-t-«-t-r-t-^^r-t-i-t-r-t-^^-r-<-r-l- <-•- r-* »-»• r-t- i-t- r-t- *-* i-*- r-»-- Q_ CDQCDCDCDCD'"'*''"'"'-'"^'"'"'CDCDCDCDCDCDCDCDCDC" D 0.0.0.0.0-0. 0.0.0.0.0.0.0.0.0.0.

mD)T?<<<<< o o o 1 1 3" 3"O R o 5" o" o 8 0 o 3" ° n Q o u Q. Q. o-' 3; 03 03 5 3 ,-i. IT o- o- o Q) n> 5. ^ 3- Oj 03 3 c o c 03 03 m c 3 c u. 3 3 CO Co ^> c O w (o a 1 c Q- | §.C »O C-Q I03. CO c- 03 03 CO CO •o o ID CD CD §; §; CO 03 S 5.' CQ CD 03 CO c' - a 2 2 CV> 3 03 03 S CQ 03 C CQ c c s o' S 3 S. 3 CO Q3 3 3 3 o

-^CO-vl-gCX3-vJ^->JC»C»COCX)00C33CJ3^IO3^1CDCO^JC»a>C»~4~«JC» o ^03CDCj0-vl-g-».WCDCJ1CDJ:>.CDCn-vJCD0100-I^C35N3N3-NlN3^>.03CJlCJIcn03-vl-».CjON3 o CD

m OpOOpOOOOOOpOOOOOOOOOO^OOOO < bobbooboobbbobbbobbbob ibbbb 0000000000000000000000 soooo 2L c CD a 00 .&. CO *k *. CO CO 4Sw Ji.Ji.^.4S.COCOCOCOCOji.J^JS.4SCOJi. CD 03 CD CD M CJI CD CD O3COCJ1C0004i.CD00C»O3CDN3COCJ1N3 o 3 CJ CJI _>. Ji. -x co co cn go 03 cjicoj^c3)co*.rococoN)a3*. CO 00 Ji. Ji CO j£ ^ ^ ^ CO CO CO 4^. CO ai CD o CO O) CD o~ CJI CJI CO CJI CD o CO 03 CO CO -J CO o en --J O CO NO -J _i CD 03 O o CD C» CD NO 03 N3 ^4

CDCOCDCOCDCDCDCDCDCDCO CDCDCDCDCDCDCDCDCDOCDCDCDCDCD CDCD-vl-vlC33CDC»O3^J0000 COCJIOOCDCJIOOCOCD^JOJ^OOOOCDOO OOJOOT-^OOffi^COCO^COinr-NaSCDOO^-tD^^^S"* O(J)0)O0)0)0)0)0)(3)0)0)0)010)0>000)050)C3)0>t3>0)0)0)0)

CM CO CO CMT - O IT) O O T- T- CD T- CO CM s CM CD in CM o O) in o I - ^r •sf r- CM in •sf O r*- ^r o> i^ in in •* i^- CM "fr 1^- I-- •* j^ 3; 3; ^ co in 5 N r- J 5 5 co 5- ifl ^ n ^ 5 5 5—T CM O CM CM CM (35 CM co 35 ^ "^ co o in "^ So in co 35 in h- CM t^ •* •* 1^ •* CO CO •ST CM CO• * CM CM CO CM O CM CD M- O) O) CO T- "fr •*• T- i- CO NT •sf CM T- •* •*• NT CO if in •* co \r •* CO 'St

00 LO 03 as o o o o o o oooodoooi^rSR oooooooooo T~ O O T- OOOOT-T-OOVTV OOOOOOOOOO o o XXX o o o CM CM CO ddoo'x:xC)C)°°°c>do'oood CM CO

T-OOOT-OOOO^T-nNrfflCMffl^ONOXOnT-T-S^fflS '*mr^'srcoin^t-coo)CDCDCMM-o)OCMcoinr^ococDcj>T-o)T-oo C0CO00C0T-CMCD00r~-l^00C0I^CDCDCDCDO)l^r>-I^h-r-COCDCOr~-

S CO c CO CO CO 3 C CO CO CO co CO CO o CO o co O co CO CO CO CO CO co CO CO -Q .o c .5 o CD «0 CO .o .o .o CO § s •6 CO I i s CO CQ "5 •= :£ c I 5 £ ! o .5 F en CO o o 5 CO Co Q. *!=: I £ T3 C3) CD CD o •c -c CO CO CO 3 CO £ S C CO S-s CO co co .co co •S -5 3 3 -Hi .CO £S CO 3 co co -§'§ .co co 3 o 8 o o g 2 5 5 2 •s •B 6 g e. 9. a o s 5 2 5 -o o 5 c c CO co o S S -2 Pi en 3 o p o T3 T3 T3 T3 T3 T3 3 TJ -fe "C3 5 -8 a o 3 _ T3 T3 -2 o o o o o o o §• o o I •Q co o * -c o o o o p 5 3 oc ft: ft: ft: ft: ft:OOgcocoftlEo ' ft: o £ p;p "o ^ o S CO .CO ac ft: s ° -c: E £ ! CD (D 03 0 0 0 0 0 0 CD CO CO CD CD

o 5 5 O00000000000 5 5 5 5 5 5 5 o o o 0 0 0 o0 o = 00000000000 0 0 0 0 o 0 0 o0 0o o0 > > > > (]) U U L. I— l-l-l-l— U-l— k_ >- >- > > >- > >- >-ooooooooooo r^ i— r^~ o o o o r— r~- r— r— r— r~- •— r^i^i^i^i«wr^i^|s»i^i^|N_|N.|s.|s. o o o o CZ> CT> CI3 CZ3 CZ> CZ> C=>ooooooooooooo o CM CM CM CM CM CM o o o o o o o oooooooooooooo CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CO CO CO CO 3 =3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 -3 ~3 ~3 ~3 ~3 —i ~3 -3 —3 -3 ~3 -3 ~3 ~3 ~3 ~3 I I I I I i i i i •* -!l- •* •* •* CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM

241 Appendix J. cont. Collection Leaf Treatment Morphotype Top Match Score E Value Identities % 12-Jul-2007 Yellow Untreated White Rhodotorula ingeniosa 831 0.00 473/500 94 12-Jul-2007 Yellow Untreated White Rhodotorula ingeniosa 888 0.00 480/496 96 12-Jul-2007 Yellow Untreated White Rhodotorula ingeniosa 274 2x10-70 166/180 92 12-Jul-2007 Yellow Untreated White slimy Cryptococcus flavescens 476 10"131 258/267 96 Appendix K. Top matches of filamentous fungal ITS sequence of turfgrass fungi. rDNA amplified with primers ITS1 and ITS4. Sequences compared with nr database using megaBLAST. Top matches had annotated genera and species, and identities matched over 80%. The plots on the pathology green were not treated or treated with Heritage® 12 g/100 m2.

Collection Leaf Treatment Morphotype Top Match Score E Value Identities % 23-Nov-2006 Green Untreated black/black Microdochium bolleyi 688 0.00 351/353 99 23-Nov-2006 Green Untreated black/black Microdochium bolleyi 884 0.00 448/449 99 23-Nov-2006 Green Untreated black/black Microdochium bolleyi 904 0.00 462/465 99 23-NOV-2006 Green Untreated black/black Microdochium bolleyi 969 0.00 493/495 99 23-Nov-2006 Green Untreated white/brown Ampelomyces humuli 904 0.00 456/456 100 23-Nov-2006 Green Untreated white/brown Ampelomyces humuli 926 0.00 467/467 100 23-Nov-2006 Green Untreated green/black Cladosporium cladosporioides 896 0.00 452/452 100 23-Nov-2006 Green Untreated green/black Cladosporium tenuissimum 918 0.00 463/463 100 23-Nov-2006 Green Untreated grey Phoma pomorum 448 1 Q-123 250/262 95 23-Nov-2006 Green Untreated white/red Epicoccum nigrum 898 0.00 455/456 99 23-Nov-2006 Green Untreated white/brown Ampelomyces humuli 858 0.00 433/433 100 23-Nov-2006 Green Untreated white/pink Fusarium proliferatum 918 0.00 466/467 99 23-Nov-2006 Green Untreated white/pink Gibberella fujikuroi 956 0.00 470/471 99 23-Nov-2006 Green Untreated white/white Nectria cinnabarina 920 0.00 481/484 99 23-Nov-2006 Green Untreated white/white Mucor hiemalis 648 0.00 378/403 93 23-Nov-2006 Yellow Untreated beige/beige Microdochium bolleyi 928 0.00 468/468 100 23-Nov-2006 Yellow Untreated black/black Microdochium bolleyi 894 0.00 451/451 100 23-Nov-2006 Yellow Untreated green/black Cladosporium cladosporioides 839 0.00 423/423 100 23-Nov-2006 Yellow Untreated green/black Cladosporium tenuissimum 886 0.00 447/447 100 23-Nov-2006 Yellow Untreated green/black Cladosporium tenuissimum 884 0.00 446/446 100 23-Nov-2006 Yellow Untreated green/black Cladosporium cladosporioides 936 0.00 472/472 100 23-Nov-2006 Yellow Untreated grey Alternaria compacta 931 0.00 506/508 100 23-Nov-2006 Yellow Untreated white/brown Microdochium bolleyi 987 0.00 498/498 100 23-Nov-2006 Yellow Untreated white/white Microdochium bolleyi 763 0.00 410/417 98 f- o a> O)oicoajooioiNiooiooconoon^ttoojffioioi s oo o O) CDcncnc!)Oc»oooo)0)0)ocna)ooo)0)0)QO)0)ci)D)c s

CD LO T- CD( D Ol N CM OO O) N o o CM en to t- T- CM CM m O CM CM CO o co co in m in o w CO O) 00 O in in m Is- CM CN CD -ST CO -ST 00 v^vs_v5j(\i5555 5 o <* •sr \r t n * 5 o in o •<- t !2 CDCMCoh-cMcocomcncoc!P Jf 5f ? Jf D (O N N O co o in m m CD •* oo a> h~ h- in h- T- T~ -SJ- o h- CO •st- •*- •*-r-- CM . •* 'vT •*!- CO CM CM •^- co co •sr -a- ^ I it w in

CD o o o o o o O O O to CD o o o o o o oofe oooooooo 15 o o o ~ OOTL OOOOOOOO 0.0 0 > X 0.0 0 0.0 0 o o o o o o X oooooooo us •sr CO

CMCoincocDcoa>cococoomi,--cDOi^cDcocMCMr^T-f-oo o coc7)05T-rv.a>o>cMCMcoT-cocoooocoinmocNcoincooo 88 8 90 2 83 5 u 49 6 CDOJcncDoooo'^minoocDoomocoincDCDcocoooi^-r-oo (0

CO CO CO CO £ £ CO 11) CD ci) .•o :t3 .t3 TJ 2 2 o CI o o o o .e c e- c c 5j 5r o o o o o CO o o •*-• o u co o o •c -c CO C co CO CD § e- e- =S E o o rum rum rcoreu bolleyi bolleyi bolleyi bolleyi rcoreu bolleyi bolleyi bolleyi dados a dados dados ipes CD dados « CO co -Q o CD co -Q .O) alis alis nuo nuo

nuo .£ .£ C E E £ £ LUdl £ P £ £ £ £ £ £ £ £ £ CD CO• ° C £ C) Co C co c £ £ £ CO -££ "£o c .3 ori u O/Vl V

lus ste lus ste £ chiu chiu C/7/ U c/?/ u .CD .CO chiu chiu chiu o o £ o •S -£ C) o o O o -C •c u CI o o o n o u. u u. 3 Ti •n v. i. « «^5 Ti 10 CO 10 IU £ £ 5 co O o T» Ti co co o o -o T) Tl o o n O n n o o u o o 0) CI) o O n o y u -2 -2 CD 'C .2 £ ri c n (T ^ ti -5- w s» CI CI (.3 o >». < < 3 3 ^ "ca "co £ CD -5= P o •Hi CD ^ The C/ a Cla C/ a C/ a Mic M c Cla Epi Mic M c Mic Mic Phom Mic § Epi e a. a. 2 § a> a CD CD t o

ac k 2 £ lac k

lac k lac k ^ ac k ac k lac k row n row n B ac k ac k ££ CD CO

XI XI XI XI beig e X! X! XI 3 beig e o X! XI _Q X! ^ £ v CD beig e XI ja XxiI ^ x: cB 35 CD at $ "55 CD a o o .#—» o hit e hit e hite /

co ree n ree n lack / lack / ree n ree n ree n ran g ran g ree n co ree n 1c x: sz o o- XJ xi X! XI s s DJ D) CD O) o o s •$• sJ c~ D) O) O) 5 5 5 X2 2 •OT3T3T3TJT3TO-OT3 CDCDCDCDCDCDCDCDCD vuvawaauvvaiiDovttcDrasnnrararacp.caucucD cococococacococacocococococococoS£££S£SffScococo 0)Q)CD(U I- h- J— I— I— I— l-H-l-33=)33Z)=)3=)l-h-(-

OOOOOOOOOOOOOOOOOOOOOOOOOCDCDCD ======:= = =: = = = :==: ======:= = = CD Q) (D (D(DCDCDCDCD0CDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCD(DCD>-<->- >->->->->->->->->->->->->->->->->->->->->-)->->->-000

h- r~- t^- 1^ 1^- 1^- h-r^ i^. i^- o o o C o o ooooooooooooooo o o o o o O ooooooooooooooo o o o o o o o o ovICNCMCMCMCSICMCMCMINCMCNCMCMCMCNCNCMCNICMCMCMCMCMCN^^ o 0}1 U .2 CO CO CO COc o co co cacococococococo co ca co co co co co co co "5 2 III o CMCMCMCMCMCMCMCSICMCMCNCMCMCMCMCMCMCMCNCMCMCMCMCMCMCMCMCN

244 s^e

to IO IO IO IO IO K> IV) IO IO IO IO IO IO ro ro IO to IO IO IO IO IO ro ro ro ro IO t f> 4* f> 4* 4* -f. -1*. 4* 4*. 4* 4=. 4*. 4* 4* 4*. 4* 4*. *>. 4* t *. •>. •>. •h ^. *. Ji. 3 3 2 3 3 S 01 kCD CkD CkD CkD 0k) CkD CkD CkD CkD CkD CkD kCD 0) Q> kCD kCD 0) mk CD k0) k0) a> k0) k0) kCD s0) kCD f f s< f •< *< >< •< •< »< *< >< << >< •< >< >< •< s< s< f >< •< s< •< •< •< >< ib lb ib lb rb rb rio rb lb lb tio lb lb lb lb IO ib IO IO IO rb to lb o O o o O o O O o o O O o o o O O o o o O o o o O O o o o o o o O o O O o o O O o o o O O o o o o o o o o o o o ^i -si -sl -sl "NJ -J -J -J -J -J -sj -J -s] -si -s| ~s| -si ~s| -sl -sj -J -sl -sl " ^ ^ ^ ^

-< -< -< < CD O o CD CD CD CD O CD CD CD O CD O CD CD CD CD CD CD CD CD CD CD CD CD CD fD —1 -n -1 -i 3 3 -1 —i —i "C^D CD -C^D CD CD CD -^CD -C^D •C^D C—\D CD -C^D C-%D CD CD C—D* fD -C^D -C^D CD CD 0 CD -C^D =O O= o" O= CD CD CD CD CD CD CD CD CD (D CD CD CD CD CD CD fD CD CD CD CD CD CD CD € € S € 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

H H H H C c c c C C C C C C C C C C H H H H H -\ H H H H -1 3 3 3 3 3 3 3 3 3 3 3 3 3 3 -1 CD fD CD CD cB fD CD CD S s 0) s0) s sCD s CD CD i—f & CD CD CD CD CD CD CD CD CD CD CD CD CD CD Si 1—*. Si Si sCD. f&D CD CD CD CD CD 0) 0) CD CD CD CD 0) CD 0) CD CD CD CD C^D CD CD ^ CD CD CD* Q. Q. Q. i-4- r-t- «—»• r-*- i—*- fH- r-t- r-f- «-+- Q. Q. Q. ffi"Q. Q. 2 Q_ Q. CD" CD CD CD CD CD <—»CD- CD »-*CD- (—*CD- CD CD CD a. a. a. Q. Q. Q. Q_ CL a5". a. a. a. Q. a. Q. a. a.

giSS^S^SOOCQCQCQCOCQggg^ggOCQCQCQCQg -1 Cn»«D s —- 3-,-- 3-P" - 3—" r 3—" r T7 I? ^3 ^3 tr Z3 73 T -*- —r -r- -r 3-r"- —i13 !-? -3 73 73 —r

ck/blac k Sck/blac k D ite/whit e ite/pin k ite/beig e ite/beig e ite/re d en/brow n en/brow n en/blac k en/blac k te/whit e te/beig e en/brow n en/brow n en/blac k en/blac k t e fluff y

en/blac k en/blac k t e fluff y t e fluff y te/re d te/re d ng e CD CD CD CD 5: ng e ng e CD CD CD CD 3: CD —\ CD CD 2: ?. z s: 0) —\ 2! CD •< CD Phom David David Epacris Epico Micro Muco, Microi Micro* Epico Epico Phom Muco, Micro Fusat Fusarium proliferatum Epico Micro Cladc Muco Cladc Micro Micro Cladc C/ad o 7??e/ e Tfie/ e Thele lei ium lateritium lei 00 Q. Q. Q. C3- o CJ- o- o a. 00 00 Q. -i Q. Q. o o ^i £D CD 00 -i 'Or/u m dados la tassiana ur n nigrum ur n nigrum c/7/u m bolleyi chium nivale crt/u m nivale ch/u m bolleyi u m nigrum tv m nigrum medicaginis medicaginis la tassiana 'oriu m dados c/7/u m n/Va/ e ch/'u m fao//ey ; cft/u/r ? bolleyi lus stercoreu iemalis O O O /u s stercoreu lus stercoreu microphylla iemalis iemalis •Q O o o o o o -or/uA7 7 dados Q torium dados 3 o 3- o o o o 3- t3 3-

13 Co 00 Co T3 ^3 ^3 O O O O 3 3 3 3 O' O' O' O'

CD CSD; CSD; CSD Cao 00 00 Co 77 9 87 6 82 5 77 3 74 5 88 8 77 7 86 8 52 7 53 7 2 _. _. simcosjcDtofflfflco-sicotojiAooO) UOIlUO-'lOAMsJjj.OfflMOOOOOIsl OU1U(DU1COMUOU09^ISU^K)0)

o o o o o o o o X X x x p 10 " QO opppoooooo oo o o o o o o o o ^ b o o o o o o o 4^ o a> o oooooooooo o o o o

W CO W W Ji * 42. CJ1 4s- 4s. CO 4*. en co co 4*. CO o -l -sl 00 CD CO O CO o -sl co en 4s- _ Js. ro oo go ^ co J*. oo -sl -sl io cOo ^1 -sl en eo £§i§ to co 00 4* en o -^ O *• ji CT> W •li- CO -sl * 5 ^ ^ 4i 4a. 4*. 00 00 00 ^1 O CD CO 05 en oi oi ro en ro ro ro ro en en 00 en o ->• en co co o co CD CD CO CD CO COCOCOOOCOCOCDCOCOCOCOOCOCOOOOOOCO OJ 05 -»--»-01-s]00C00000C00000OC0C00000OC0 00 CD O 00 co en co s Is- 00 OO CM O CO Is- OO 00 05 00 Is- Is- in CT>I - CO T— Is- IO CO Is- in co OO CM o Is- OO O) CT>05 O CO CD o> O) o> oo CO 00 oo 00 05 a> a> 00 OO en 00 oo 00 00 OO OO a>

s _ •* I - CD O O CO T— CM co CO CO T CO o o to 05 OO uo \r 00 CM CM CO XI- CO s CN •* •sr CM 00 CO U> 00 CO CM OO •* •* •^ • CO CM TT I - CO in •sr m o I - CO co •* •a- CM CM CO CM -3-

0) . tn co OO'i-.OOOOO o o o o oOQoo5 o § o a 00~00000 o o o o o o o o o > OO x d d O O O o o x odd X O x X ° XT- X t- O HI in CO 00 CM ^ ^

co CO ?3 3 .<0 O 1 o CO CO • * CD CO o •S CD •a -2 3 E < K a o 5 o 1 c I CD IE CO "5 co 1 .3 co p CD :t; -Q O U , o I! •O .co .co 5 I 11 3 £ E E I- ! E I 3 R -i -C o w i ^ -c 3 o E c O o u CD o o o 1 -^ I II §•?• I o O CD fr CD O PI 8 ~ P § s O "Qo 1 co T•2gd CD CL CD J* CD j«: a) c O) O o o c CJ > T5 CO jo T5 ^1 co o CD J2 CD i5 iS _ o CD J2 CI) & 3= J2 O) I CD c "c "33 52 C "33 "£ "£ c 2 2 CD c? •4-J a> CD CD co "35 CO CD cu 0 cu CD 'JE CD ?° J= _co s en OJ o o 5 5 X! U> D) C3) D) o S 5 J2 C3) O) O) C3) 5

T3-DT3-OT3T3-0-0-OT3 •a -a -a -a _CDCDCDCDCDCDCDCDCDCD__ T3 T3 "O T3 ^ *-« ^-* +-• *- +-» •—> *^ *-• ^-« -w o u •o CD CD CD CD CD CD ID CDC D n)C0C0(OCOCOCOC0COC0C0(DCD -•-» *-• •*-» •*-* «m7s™™-ScUfl>cD(DCDCDCDCD*s-S CD co co co co CO COCOCOCOCOCPij^^^^j^^^^^CpCO CD CD CD (D CD 0) 0 CD (D CD CD "P CD CD filCD £ l0) jD i— L_ i— i_ i— _i- c c c c c c c c c I- H- I- H- ^D33I3Z)333ZJ I- K- I- \- 13 c c c c c c c CD CD CD CD CD CD CD llllllJlllllllllllIII CD CD CD CD CD CD CD >->->>->>->->->->>->->->->->->-oe)oooooo(i)oo

1^ r^- 1^ fs. f-s. r~- h- [N». 1^ h- h- [^ h- Is- Is- Is- |^ s s s s s c o o f~^ O O f~5 o o n O O O (~J O O o O I - [*»*, |N*. [s» f^. [s^ I - I - I - I - o o o O o O o CO o o O O o O co o o o o Q O^* . Q o o o O O O CN (N CN CN CN CN CN CN CM CN CM CN CN CN CN CN CN O O O O i i o o o o o o o o >. i. >* A. >. >, i. >< >. >> i. >. >. i. >, CN CN CN CN CN CO CO CO CO CO CO CO CO CO CO CO 3 3 3 3 3 2 -J -3 —> "o 5 1 2 1 1 * 2 2 1 2 2 1 i } i ^ i % -3 •sr •5f •sf •* N) IO to ro tO M ro ro to ro K> ro IO IO M to ro to to to ro ro to ro to •D "O c_ (D k fck fc k fckI feI i 1 1 k k k k k k 5- k k 3 to to rororororororororo ro ro ro ro ro ro ro ro ro ro ro ro ro ro o o 000000000 00000 0 0 0 0 0 0 000 a o o 000000000 00000 0 0 0 0 0 0 000 -vl ->1 -J-vI-sj-vj-vj^i-vi-vl-vl •vj -vl -vl -si -J -vl -j -j -NJ •vj "v| -vl "vl vj x" 7s o o •< < •<-<-<-<-<•< < O O (t> CD CDCDCDCDCDCDCDCDCD CD CD CD CD CD CD CD CD CD CD CD f> 3 3 0 0 000000000 00000 000000 = CD CD O CD CD § § s s s> SE s s s s s g g S 3 3 c c c c c c c H H H H -I -I H -! -I -I -I C C 3 3 l—»- «-f 3 3 2-ZZcDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCD CD CD CD CD CD CD CD i-»-»-*-«-t-i-<-.-»'r-*-»-i-i-»-<-»-«-t-#-**-»-r-#-«-»-»-»-»-»-CD CD m Q> Q)Q)Q>Q)D)CDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDDJCD CD CD CD CD CD CD CD CD CD a. a. a. a. a. a. a. a a

2" S ai Z 2! 2: a> Z 2". Z 2! CD CD o •"•* ffi ffi »"•* *"*• *••• O »-f «-•" «-*- •-*• 3 T3 1 "IHIIgfEE CT&S — — — CTS g g a- O" OW " 1~l S-> {- 3" L_ QJ — ib_ i__ C nn\. n\ 03 O 5 0 0) O <5 o o ^ v? J o 5-- =•• 5-- S C3 o o _ < 3s 8 ^^^ i CD •D CD

§ Q |5QQ§O O Qo § Q § § Tl W R' 03 03 c c 9? 03 03 O' 05 O' 03 3 Q. Q. o o Q Q 03 S O O o o i §• s 1 a O § §• o S- 3" -S2 Vs -"! 3 3. 01 Cfl Q. 3- 3" T3 T3 O Q. o o O O CD' 2 o 3 o ^33 CD' CD" ? 3. 3= 2 §• §• 3 3 3 03 >? c" c" c~ ' 3 3 C c' CD CD J 01 er C 3 g- c' S" c 03 03 3 C' CD 5" <" 3 3 3 3 3 3 3 3 3. -o ~ O 03 CO o- •§ §• 3 5T «' o a- a- o- CD CD (D CD o CD 3 01 o o- ^ 3 3 oT 3 =-. 03 3 s-l a 2. 9. c & CD c a. « C e o C 0T C/3 •8 01 01 55' 55' ^. CD 0) 03 03 i §• 03 •8 3 o 03 3' o 3' CD = 3 3 o c 3' c c 3 c 2 i s. 03 3 3 O" 3 I 8- CO CD Co

00 CJ1 00 ID(I100O3O)N3WN-»>IOlUO3(OlDUl*4VOlO)rOW 00 o ->• oo^j03cnorooocooo-iocorocooo~jrooowoooro o o -»• C0W-fc.00-i00-»-CDCDCn-vlCT)-vlCJ101CDCD00C0-vlCA3O

00 00 *. to ro to m 0 O 0 0 0 X X 0 X 0 0 0 0 X X X 0 O 0 0 0 0 10-1 1 0 0 0 < 0 O 0 0 0 —A 0 —*• 0 0 —*• 03 0 L0 O 0 0 0 01 O p 0 0 0 ^0 0b 0b 0 0 en -J O C—J^ CO 0 CO -^ 01 0 0 c -si 0 01 -J CO U -». fD Id e •1^ _, ^ .&. ro •^ en C«J •C CO CO ro -t>- CO CO ro -&. en CJl CO ro ro CO *. CO C73 ^1 CO Ji. -fa. -v| ->J ro CJl CO CO 00 ro ro en en •S>. CO ^1 -v| CO 0 CO 3 _k _1. _1 ^. —i. *• a> -t». 0 O CT3 •vl 00 ro CO CO vg ro CO ro .&>. ro ro 0 no <*> CO N3 4^ C/0 •C^ J*. CO —1 -1*. CO CO -P> CO CO CO Ji. ^i ^*1. ->i Ji. C3en1 C33 CO CO 00 03 CroO Ul ^^ CO ro 03 0e0n OenJ 0 CIroO Ji. •v^ l «-tx* CD 1 en en ro en .&. .&. CD ro 00 ro ~ CO 0 00 00 ~^ 05 ~^ 0 00 03 ~* ^. .&. —^ CJl to CO 0 en (A

CO CO CO CO CO CO CO CO 00 00 CO 00 CO CD CO co co to to to CO CO CO 00 Sv ^i 00 ~j <33 to 01 CO -vl 00 .&. ^J Ul •vl CO •vl ^4 $ Oi Ji. 0 CJl •c- 0 en —*