Role of Plant Growth-Promoting Rhizobacteria in Integrated

ROLE OF PLANT GROWTH-PROMOTING RHIZOBACTERIA IN INTEGRATED

DISEASE MANAGEMENT AND PRODUCTIVITY OF TOMATO

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of the Ohio State University

Cristian Nava Diaz, M.S.

* * *

The Ohio State University

2006

Dissertation Committee:

Ph.D. Sally A. Miller, Adviser ______

Ph.D. Michael A. Ellis

Ph.D. Matthew D. Kleinhenz

Ph.D. Douglas J. Doohan

ABSTRACT

The ability of three rifampicin-resistant strains of Bacillus spp. (B. subtilis MI600,

B. subtilis GBO3 and B. amyloliquefaciens IN937) to improve fresh market

tomato growth and productivity and reduce the intensity of diseases caused by

Xanthomonas euvesicatoria, Alternaria solani, Septoria lycopersici and

Pseudomonas cichorii was evaluated under greenhouse, field, and controlled environmental conditions. Rifampicin-resistant mutants of each strain were generated and used to monitor colonization of tomato roots. There were no differences in colony morphology, endospore production or ERIC fingerprint pattern between wild-type and rifampicin resistant mutants of MBI600 and IN937.

The rifampicin-resistant mutant of GBO3 had the same colony morphology and

ERIC-fingerprint pattern as the wild-type, but produced significantly fewer endospores. Under greenhouse conditions, no growth increase or bacterial leaf spot (X. euvesicatoria) suppression was observed in tomato plants inoculated with any of the Bacillus spp. evaluated in this study. Under field conditions,

MBI600 and GB03+IN937 were integrated in an intensive tomato management program that included mulch, drip irrigation and a forecasted fungicide spray program. Bacillus spp. population densities were 104 - 106 CFU g-1 during the

seedling stage and dropped to less than 103 CFU g-1 during the flowering and

fruiting stages. Significant increases in plant height were observed in all Bacillus- inoculated tomato plants, however, foliar diseases incited by A. solani and S. lycopersici were not reduced. GBO3+IN937-inoculated plants were more susceptible to bacterial stem rot caused by Pseudomonas cichorii than those inoculated with MBI600 or non-inoculated control plants. In controlled environmental studies, Bacillus spp. population densities ranged from 101-105

CFU g-1 tomato root. Neither nitrogen concentration in the nutrient solution nor activation of the systemic acquired resistance defense mechanism had a significant effect on population density of MBI600, GBO3 or IN937. In a separate experiment under controlled environmental conditions, Bacillus spp. strains

MBI600, and GBO3+IN937 did not increase nitrogen concentration in tomato leaves. Lesions induced by Pseudomonas cichorii on tomato stems tended to be

longer in GBO3+IN937 inoculated plants than those inoculated with MBI600 or a

water-treated control.

iii

DEDICATION

To my parents

Maria Cristina Diaz Lozano

Jose Guadalupe Nava Bernal

reri

ACKNOWLEDGMENTS

The author would like to express his deepest gratitude to all members of his

advisory committee, Dr. Sally A. Miller, Dr. Michael A. Ellis, Dr. Douglas J.

Doohan, and Dr. Matthew D. Kleinhenz for their various inputs during course and

laboratory work as well as guidance in this research.

The author would like to thank Ms. Melanie L. Lewis Ivey and Dr. Annette L.

Wszelaki for their important inputs and guidance during his stay at The Ohio

State University Department of Plant Pathology.

The author would like to acknowledge the economical support of the National

Council for Science and Technology (CONACyT) Mexico and the Ohio State

University

The author would like to acknowledge Mr. Bert Bishop, Dr. Larry Madden, and

Dr. Paul Pierce for patient support in statistical analysis.

The author would like to acknowledge the support of Dr. Fulya Baysal Tustas.

The author would like to thank the help and support of Emilia Gabriela Briceno

Montero.

The author acknowledges the support of Fatthy and Samia Abdelalim.

The author acknowledges the support of friends: Katia A. Figueroa R., Mizuho

Nita, Angel Rebollar Alviter, Jhony Mera, Carilyn Perry and Santiago X. Mideros.

VITAE

1971………….…Born, Mexico city

1994………….…Research Assistant, Phytopathogenic Fungi

1995………….... BSc Agricultural Parasitology, Autonomous Chapingo University

1997………….... M.S. Phytopathology, Postgraduate College

1997…………… Technician, Plant Parasitic Nematodes

2001………….…Research Assistant, Integrated Disease Management

2002 – present…Student Research Assistant, The Ohio State University

PUBLICATIONS

Nava-Diaz, C., Abdelalim, F., Kleinhenz, M.D., Doohan, D.J., Lewis Ivey M.L., and Miller, S.A. 2005. Effect of mulch, irrigation, fungicide program and Bacillus spp. on fresh market tomato. Phytopathology 95: s73 Lewis-Ivey, M.L., Nava-Diaz, C., and Miller, S.A. 2004. Identification and management of Colletotrichum acutatum on immature bell peppers. Plant Disease 88: 1198-1204 Nava-Diaz, C., Kleinhenz, M.D., Doohan, D.J., Lewis-Ivey, M.L. and Miller, S.A. 2004. Bacillus spp. with potential as biological control agents. Phytopathology 84(6): s74 Teliz, O.D. and Nava, D.C. 2001. Integrated pest management: bases and philosophy. 15-22 (8p.). Teliz, O. D. 2001. Integrated Pest Management. Simposio. Annual Congress. (ISBN 968-5284-07-5). The Postgraduate College, Mexican Entomological Society and Mexican Phytopathological Society. Queretaro, México. July 2001: 129p

FIELDS OF STUDY

Plant Pathology

TABLE OF CONTENTS

ABSTRACT ...... ii DEDICATION ...... iv ACKNOWLEDGMENTS ...... v VITAE ...... vi LIST OF TABLES ...... ix LIST OF FIGURES ...... xx LIST OF ABBREVIATIONS ...... xxv INTRODUCTION ...... 1 CHAPTER 1 ...... 28 INTRODUCTION ...... 28

MATERIALS AND METHODS ...... 34

RESULTS ...... 39

DISCUSSION ...... 49

CONCLUSIONS...... 53

LIST OF REFERENCES...... 54

CHAPTER 2 ...... 60 INTRODUCTION ...... 60

MATERIALS AND METHODS ...... 65

RESULTS ...... 71

DISCUSSION ...... 91

CONCLUSIONS...... 97

LIST OF REFERENCES...... 98

CHAPTER 3 ...... 105 INTRODUCTION ...... 105

vii

MATERIALS AND METHODS ...... 110

RESULTS ...... 120

DISCUSSION ...... 142

CONCLUSIONS...... 147

LIST OF REFERENCES...... 148

CHAPTER 4 ...... 153 INTRODUCTION ...... 153

MATERIALS AND METHODS ...... 158

RESULTS ...... 167

DISCUSSION ...... 181

CONCLUSIONS...... 185

LIST OF REFERENCES...... 186

CHAPTER 5 ...... 192 INTRODUCTION ...... 192

MATERIALS AND METHODS ...... 197

RESULTS ...... 206

DISCUSSION ...... 214

CONCLUSIONS...... 217

LIST OF REFERENCES...... 218

CONCLUSIONS...... 220

LIST OF REFERENCES...... 223

viii

LIST OF TABLES

Table 1.1. Analysis of variance of endospore production by Bacillus subtilis

(MBI600 and GBO3) and B. amyloliquefaciens (IN937) on endospore-forming medium after 14 days incubation. Means followed by the same letter are not significantly different (LSD 0.05)...... 42

Table 1.2. Mathematical models that best describe Bacillus subtilis (MBI600 and

GBO3) and B. amyloliquefaciens (IN937) endospore formation over time...... 46

Table 2.1. Composition of nutrient solution used to fertilize tomato seedlings. .. 67

Table 2.2. Bacterial population composition in 49 day-old tomato roots inoculated with Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937. Colonies were isolated from roots on rifampicin- amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on three colonies per sample (Experiment I)...... 74

Table 2.3. Analysis of variance of Bacillus subtilis strain MBI600, B. subtilis strain

GBO3 and B. amyloliquefaciens strain IN937 population densities on 49 day-old tomato roots. Colonies were isolated from roots on rifampicin-amended medium.

Identification was based on colony morphology and rep-PCR with ERIC primers carried out on three colonies per sample (Experiment I). Barlett’s p-value 1.0.

Normality assumed...... 74

Table 2.4. Bacterial population composition on 56 day-old tomato roots

inoculated with Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B.

amyloliquefaciens strain IN937. Colonies were isolated from roots on rifampicin-

amended medium. Identification was based on colony morphology and rep-PCR

with ERIC primers carried out on three colonies per sample (Experiment II). .... 75

Table 2.5. Analysis of variance of Bacillus subtilis strain MBI600, B. subtilis strain

GBO3 and B. amyloliquefaciens strain IN937 population densities on 56 day-old

tomato roots. Colonies were isolated from roots on rifampicin-amended medium.

Identification was based on colony morphology and rep-PCR with ERIC primers

carried out on three colonies per sample (Experiment II). Barlett’s p-value 1.0.

Normality assumed...... 75

Table 2.6 Growth promotion and induced resistance against Xanthomonas

euvesicatoria in three varieties of tomato inoculated with rifampicin-resistant

mutants of Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B.

amyloliquefaciens strain IN937 under greenhouse conditions (Experiment I).

Plant height was evaluated over time. The area under the plant height curve was

used for statistical analysis. Root length, fresh weight, dry weight and bacterial

leaf spot were evaluated at the end of the experiment. Tomato varieties

‘Mountain Fresh’ and ‘Mountain Spring’ are abbreviated as ‘Mfresh’ and

‘Mspring’, respectively. Inoculation and non inoculation with X. euvesicatoria are

abbreviated as ‘xc’ and ‘xo’, respectively. Levene’s p-values were 0.493, 0.359,

0.063, 0.032, and 0.000 for plant height, root length, fresh weight, dry weight and

bacterial leaf spot density, respectively. Normality was assumed...... 80

Table 2.7 Growth promotion and induced resistance against Xanthomonas

euvesicatoria on four varieties of tomato inoculated with rifampicin-resistant

mutants of Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B.

amyloliquefaciens strain IN937 under greenhouse conditions (Experiment II).

Plant height was evaluated over time. The area under the plant height curve was

used for statistical analysis. Root length, fresh weight, dry weight and bacterial

leaf spot were evaluated at the end of the experiment. Tomato varieties

‘Mountain Fresh’ and ‘Mountain Spring’ are abbreviated as ‘Mfresh’ and

‘Mspring’, respectively. Inoculation and non inoculation with X. euvesicatoria are

abbreviated as ‘xc’ and ‘xo’, respectively. Levene’s p-values were 1.000, 0.504,

0.169, 0.045, and 0.009 for plant height, root length, fresh weight, dry weight and

bacterial leaf spot density, respectively. Normality was assumed...... 85

Table 3.1. Bacterial population composition on tomato roots inoculated with

Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937. Colonies were isolated from roots on rifampicin-amended medium.

Identification was based on colony morphology and rep-PCR with ERIC primers carried out on three colonies per sample. Seedling (29 day old), vegetative (49

day old), flowering (102 day old) and fruiting (132 day old) stages (Experiment I).

...... 123

Table 3.2. Bacterial population composition in tomato roots inoculated with

Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937. Colonies were isolated from roots on rifampicin-amended medium.

Identification was based on colony morphology and rep-PCR with ERIC primers carried out on three colonies per sample. Seedling (36 day old), vegetative (60 day old), flowering (114 day old) and fruiting (143 day old) stages (Experiment II).

...... 124

Table 3.3. Growth promotion of ‘Mountain Spring’ tomato inoculated with rifampicin-resistant mutants of Bacillus subtilis strain MBI600, B. subtilis strain

GBO3 and B. amyloliquefaciens strain IN937 under field conditions (Experiment

I). Plant height, number of leaves, leaf length, leaf width, nitrate concentration, foliar fresh and dry weights and leaf area were evaluated three times, on 81, 95, and 116 day-old plants. The area under the curve was used for statistical analysis. Averages are shown. Irrigation, fungicide, and PGPR treatment are abbreviated as ‘irrig’, ‘fungi’, and ‘treat’, respectively. Levene’s p-values were

0.950, 0.893, 0.996, 0.998, 0.983, 0.999, 1.000, and 0.999 for plant height, number of leaves, leaf length, leaf width, nitrate concentration, foliar fresh and dry weights and leaf area, respectively. Normality was assumed...... 131

xii

Table 3.4. Growth promotion of ‘Mountain Spring’ tomato inoculated with

rifampicin-resistant mutants of Bacillus subtilis strain MBI600, B. subtilis strain

GBO3 and B. amyloliquefaciens strain IN937 under field conditions (Experiment

II). Plant height, number of leaves, leaf length, leaf width, nitrate concentration,

foliar fresh and dry weights and leaf area were evaluated three times on 101,

118, and 141 day-old plants. The area under the curve was used for statistical

analysis. Averages are shown. Irrigation, fungicide, and PGPR treatment are

abbreviated as ‘irrig’, ‘fungi’, and ‘treat’, respectively. Levene’s p-values were

0.926, 0.991, 0.994, 0.610, 0.999, 0.603, 0.573, and 0.879 for plant height,

number of leaves, leaf length, leaf width, nitrate concentration, foliar fresh and

dry weights and leaf area, respectively. Normality was assumed...... 133

Table 4.1. Composition of nutrient solutions used to fertilize tomato seedlings.

...... 161

Table 4.2. Characteristics of Bacillus spp. inoculum suspensions drenched on 14 and 21 day-old tomato seedlings, variety ‘Mountain Spring’ (Experiment I / II).163

Table 4.3. Bacterial population composition in 49 day-old tomato roots inoculated

with rifampicin-resistant Bacillus subtilis strains MBI600, B. subtilis strain GBO3

and B. amyloliquefaciens strain IN937. Colonies were isolated from roots on

rifampicin-amended medium. Identification was based on colony morphology and

xiii

rep-PCR with ERIC primers carried out on one colony per sample (Experiments I

and II)...... 168

Table 4.4. Analysis of variance of rifampicin-resistant Bacillus subtilis strains

MBI600 and GBO3 and B. amyloliquefaciens strain IN937 population densities

on 49 day-old tomato roots. Colonies were isolated from roots on rifampicin-

amended medium. Identification was based on colony morphology and rep-PCR

with ERIC primers carried out on one colony per sample. Levene’s p-value were

0.814 and 0.953 for Experiment I and II, respectively. Normality was assumed.

...... 169

Table 4.5. Analysis of variance of the effect of nitrogen concentration on

rifampicin-resistant Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 population densities on 49 day-old tomato roots.

Colonies were isolated from roots of on rifampicin-amended medium.

Identification was based on colony morphology and rep-PCR with ERIC primers carried out on one colony per sample. Levene’s p-value were 0.814 and 0.953 for

Experiment I and II, respectively. Normality was assumed...... 169

Table 4.6. Analysis of variance of the effect of acibenzolar-S-methyl on

rifampicin-resistant Bacillus subtilis strain MBI600 and GBO3 and B. amyloliquefaciens strain IN937 population densities on 49 day-old tomato roots.

Colonies were isolated from roots of on rifampicin-amended medium.

xiv

Identification was based on colony morphology and rep-PCR with ERIC primers

carried out on one colony per sample. Levene’s p-value were 0.814 and 0.953 for

Experiment I and II, respectively. Normality was assumed...... 170

Table 4.7. Analysis of variance of the effect of activation of systemic aquired resistance by acibenzolar-S-methyl before and after inoculation with rifampicin-

resistant Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens

strain IN937 on bacterial population densities on 49 day-old tomato roots.

Colonies were isolated from roots of on rifampicin-amended medium.

Identification was based on colony morphology and rep-PCR with ERIC primers

carried out on one colony per sample. Levene’s p-values were 0.814 and 0.953

for Experiments I and II, respectively. Normality was assumed...... 171

Table 4.8. Analysis of variance of the effect of rifampicin-resistant Bacillus subtilis

strain MBI600 and GBO3 and B. amyloliquefaciens strain IN937 on nitrogen

absorption by tomato plants. Area under the curve of nitrogen concentration in

the effluent was used for statistical analysis. Averages are shown. Levene’s p-

value were 0.220 and 0.665 for Experiments I and II, respectively. Normality was

assumed...... 172

Table 4.9. Analysis of variance of the effect of acibenzolar-S-methyl on nitrogen absorption by tomato plants. Area under the curve of nitrogen concentration in the effluent was used for statistical analysis. Averages are shown. Levene’s p-

values were 0.220 and 0.665 for Experiments I and II, respectively. Normality

was assumed...... 173

Table 4.10. Analysis of variance of the effect of nitrogen concentration supplied

on nitrogen absorption by tomato plants. Area under the curve of nitrogen

concentration in the effluent was used for statistical analysis. Averages are

shown. Levene’s p-values were 0.220 and 0.665 for Experiments I and II,

respectively. Normality was assumed...... 173

Table 4.11. Analysis of variance of the effect of rifampicin-resistant Bacillus

subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 on

biomass of 49 day-old tomato plants. Area under the curve of plant height, stem

diameter and number of leaves was used for statistical analysis. Averages are

shown. Levene’s p-values for plant height, stem diameter, number of leaves, total fresh weight, total dry weight and root length were 0.981, 0.864, 0.972, 0.414,

0.284, 0.901 (Experiment I) and 1.000, 0.961, 0.966, 0.988, 0.991, 0.982

(Experiment II), respectively. Normality was assumed...... 176

Table 4.12. Analysis of variance of the effect of acibenzolar-S-methyl on biomass

of 49 day-old tomato plants. Area under the curve of plant height, stem diameter

and number of leaves was used for statistical analysis. Averages are shown.

Levene’s p-values for plant height, stem diameter, number of leaves, total fresh weight, total dry weight and root length were 0.981, 0.864, 0.972, 0.414, 0.284,

xvi

0.901 (Experiment I) and 1.000, 0.961, 0.966, 0.988, 0.991, 0.982 (Experiment

II), respectively. Normality was assumed...... 177

Table 4.13. Analysis of variance of the effect of nitrogen concentration supplied on biomass of 49 day-old tomato plants. Area under the curve of plant height, stem diameter and number of leaves was used for statistical analysis. Averages are shown. Levene’s p-values for plant height, stem diameter, number of leaves, total fresh weight, total dry weight and root length were 0.981, 0.864, 0.972,

0.414, 0.284, 0.901 (Experiment I) and 1.000, 0.961, 0.966, 0.988, 0.991, 0.982

(Experiment II), respectively. Normality was assumed...... 178

Table 4.14. Analysis of variance of the effect of the interaction of nitrogen concentration and acibenzolar-S-methyl on biomass of 49 day-old tomato plants.

Averages are shown. Levene’s p-values for total fresh weight and total dry weight were 0.414, 0.284 (Experiment I) and 0.988, 0.991 (Experiment II), respectively.

Normality was assumed...... 179

Table 4.15. Analysis of variance of the effect of the interaction of acibenzolar-S- methyl and plant growth-promoting rhizobacteria on total fresh weight of 49 day- old tomato seedlings. Averages are shown. Levene’s p-values for total fresh weight were 0.414 (Experiment I) and 0.988 (Experiment II). Normality was assumed...... 180

Table 5.1. Composition of nutrient solution used to fertilize tomato seedlings. 200

xvii

Table 5.2. Bacterial population composition in 60 day-old tomato roots inoculated with rifampicin-resistant Bacillus subtilis strains MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937. Colonies were isolated from roots on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on one colony per sample (Experiment I and II)...... 207

Table 5.3. Analysis of variance of rifampicin-resistant Bacillus subtilis strains

MBI600 and GBO3 and B. amyloliquefaciens strain IN937 population densities

(log CFU g-1 root +1) on 60 day-old tomato roots. Colonies were isolated from roots of on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on one colony per sample. Levene’s p-values were 0.635 and 0.280 for Experiment I and II, respectively. Normality was assumed...... 207

Levene’s p-values for plant height, root length, nitrogen in sap and lesion size were 0.667, 0.388, 0.775, 0.023 (Experiment I) and 0.648, 0.471, 0.838, 0.045

(Experiment II), respectively. Normality was assumed...... 209

xviii

Table 5.5. Analysis of variance of the effect of irrigation rate on tomato biomass and bacterial stem rot severity. Area under the curve of plant height and nitrogen concentration was used for statistical analysis. Averages are shown. Levene’s p- values for plant height, root length, nitrogen in sap and lesion size were 0.667,

0.388, 0.775, 0.023 (Experiment I) and 0.648, 0.471, 0.838, 0.045 (Experiment

II), respectively. Normality was assumed...... 210

xix

LIST OF FIGURES

Figure 1.1. Colony morphology of Bacillus spp. on nutrient agar medium after 24 hours incubation at 28oC. A. B. subtilis MBI600 wild-type; B. B. subtilis MBI600

rifampicin-resistant mutant; C. B. subtilis GBO3 wild-type; D. B. subtilis GBO3

rifampicin-resistant mutant; E. B. amyloliquefaciens IN937 wild-type; F. B.

amyloliquefaciens IN937 rifampicin-resistant mutant...... 40

Figure 1.2. (*) Vegetative cells, (**) endospores within cell, and (***) free

endospores of Bacillus subtilis MBI600 rif on endospore-forming medium

incubated for 8 days at room temperature (945 x)...... 41

Figure 1.3. Production of endospores by wild-type and rifampicin-resistant strains

of Bacillus subtilis (MBI600 and GBO3) and B. amyloliquefaciens (IN937) on

endospore-forming medium incubated at room temperature under white light for

14 days. Experiment I: (+) Endospores; (∆) Rate of production of endospores;

Experiment II: (x) Endospores; (◊) Rate of production of endospores...... 44

Figure 1.4. Enterobacterial repetitive intergenic consensus - polymerase chain

reaction (ERIC-PCR) products obtained from genomic DNA from Bacillus subtilis

strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937. Arrows

indicate polymorphisms...... 48

Figure 2.1. Chromelosporium sp. A. Conidiophore; B. Detail of conidiophores

and conidia; C. Antagonistic effect of IN937rif on a Chromelosporium spp.

colony...... 73

Figure 3.1 Categories for tomato fruit assessment: A. red marketable; B. green

marketable; C. physiological disorders; D. anthracnose; E. bacterial diseases; F.

other diseases; G. insect or bird damage and H. other diseases on green fruit.

...... 117

Figure 3.2. Identification of Bacillus subtilis strains MBI600 and GBO3 and B.

amyloliquefaciens strain IN937 isolated from 29-day old tomato roots during

seedling stage, using enterobacterial repetitive intergenic consensus -

polymerase chain reaction (ERIC-PCR) fingerprints. Wild-type and rifampicin-

resistant strain fingerprints were used as a reference. The banding patterns of

three colonies isolated from treatments (control, MBI600 and GBO3+IN937) were

compared with the reference patterns for identification (Experiment I)...... 125

Figure 3.3 Tomato root colonization by Bacillus spp. 29 (seedling stage), 49

(seedling stage), 102 (flowering stage) and 132 (fruiting stage) days after seeding, Experiment I. Control plants were colonized by MBI600 (flowering and fruiting stages), GBO3 (vegetative, flowering and fruiting stages), and IN937

(seedling, vegetative, flowering and fruiting stages). Means followed by the same

letter are not significantly different (p-value < 0.05)...... 126

xxi

Figure 3.4. Tomato root colonization by Bacillus spp. 36 (seedling stage), 60

(seedling stage), 114 (flowering stage) and 143 (fruiting stage) days after seeding, Experiment II. Control plants were colonized by MBI600 (vegetative, flowering and fruiting stages), GBO3 (fruiting stage), and IN937 (seedling and fruiting stages). Means followed by the same letter are not significantly different.

...... 127

Figure 3.5. Effect of irrigation on Bacillus spp. population density on 102-day-old

‘Mountain Spring’ tomato roots. Means followed by the same letter are not

significantly different (p-value < 0.05). Experiment I...... 128

Figure 3.6. Effect of soil drenches with rifampicin-resistant mutants of plant

growth-promoting rhizobacteria (PGPR) (Bacillus subtilis MBI600 and GBO3, and

B. amyloliquefaciens IN937) on height of ‘Mountain Spring’ tomato seedlings 49

(Experiment I) and 60 days (Experiment II) after planting. Means followed by the

same letter are not significantly different (p-value < 0.05)...... 135

Figure 3.7. Effect of soil drenches with rifampicin-resistant strains of plant

growth-promoting rhizobacteria (PGPR) (Bacillus subtilis MBI600 and GBO3, and

B. amyloliquefaciens IN937) on height of ‘Mountain Spring’ tomato plants under

field conditions. Plant height was evaluated on 81, 95 and 116 (Experiment I) and

101, 118, and 141 day-old plants (Experiment II). Means followed by the same

letter are not significantly different (p-value < 0.05)...... 136

xxii

Figure 3.8. Effect of plastic and rye residue mulches on height of ‘Mountain

Spring’ tomato plants under field conditions. Plant height was evaluated on 81,

95 and 116 (Experiment I) and 101, 118, and 141 day-old plants (Experiment II).

Means followed by the same letter are not significantly different (p-value < 0.05).

...... 136

Figure 3.9. Diseases and pathogens observed on ‘Mountain Spring’ tomato

plants. A. early blight; B. Alternaria solani conidia; C. septoria leaf spot; D. pycnidia and conidia of Septoria lycopersici; E. bacterial stem rot; and F. pith desintegration induced by Pseudomonas cichorii...... 138

Figure 3.10. Effect of plastic and plant residue mulches on severity of early blight and septoria leaf spot on ‘Mountain Spring’ tomato plants under field conditions.

Severity of foliar diseases was evaluated every 7 days. Area under the disease curve was used for statistical analysis. Means followed by the same letter are not significantly different (p-value < 0.05)...... 139

Figure 3.11. Effect of plant growth-promoting rhizobacteria (PGPR) (Bacillus

subtilis MBI600 and GBO3, and B. amyloliquefaciens IN937) on bacterial stem

rot severity of ‘Mountain Spring’ tomato plants under field conditions. Severity of

bacterial stem rot was evaluated every 7 days. Area under the disease curve was

used for statistical analysis. Means followed by the same letter are not

significantly different (p-value < 0.05)...... 139

xxiii

Figure 3.12. Effect of plastic and plant residue mulches on severity of bacterial stem rot on ‘Mountain Spring’ tomato plants under field conditions. Severity of bacterial stem rot was evaluated every 7 days. Area under the disease curve was used for statistical analysis. Means followed by the same letter are not significantly different (p-value < 0.05)...... 140

Figure 3.13. Effect of plant growth-promoting rhizobacteria (PGPR) (Bacillus

subtilis MBI600 and GBO3, and B. amyloliquefaciens IN937) on marketable yield

of ‘Mountain Spring’ tomato plants. Means followed by the same letter are not

significantly different (p-value < 0.05)...... 141

Figure 3.14. Effect of plastic and plant residue mulches on marketable yield of

‘Mountain Spring’ tomato plants under field conditions. Means followed by the

same letter are not significantly different (p-value < 0.05)...... 141

Figure 4.1. Irrigation system. A. General view. B. The single line serving each pot

was situated so that solution dripped freely into the vermiculite. C. A piece of transparent plastic covered the surface of the growing medium to reduce evapotranspiration...... 160

Figure 5.1. Bacterial stem rot (Pseudomonas cichorii). A. Brown lesions on

tomato variety ‘Mountain Spring’; B. Lesion on tomato variety ‘Big Beef’...... 212

xxiv

LIST OF ABBREVIATIONS

g = gram. Unit of weight. h = hour. Unit of time. ha = hectare; acre = 0.404686 hectares. k =prefix kilo (103). Kilogram. l = unit of volume. Liter.

M = Mol. Molecular mass of a particular molecule expressed in grams per liter.

Metric ton = 1000 kg x g = gram-force. 1 gram of mass subject to standard gravity = 980.66 cm s-2 =

9.806 mN

μ = prefix micro (10-6). m = prefix milli (10-3). Milliliter, milligram. rpm = revolutions per minute.

Ton = unit of weight equal to 1000 kg. u = unit. Amount of enzyme that incorporates 10 nanomoles of dNTPs into acid insoluble form within 30 minutes at 74 oC.

xxv

INTRODUCTION

Tomato (Lycopersicon esculentum Mill.) is a very important vegetable crop worldwide. In 2004, 120 million metric tons of tomatoes (fresh market and

processing) were produced worldwide (FAO, 2004). In The United States, 52,892

ha of fresh market tomatoes were planted in 2004 producing 35,903 kg/ha and a

value of $ 1,342,478,000. Ohio is third among the states in tomato production for fresh market, with 2,832 planted hectares, yielding 20,713 kg/ha and a value of

$49,549,000. Ohio is also third among the states in tomatoes for processing, with

2,670 planted hectares, 70,672 kg/ha and a value of $ 13,902,000 (USDA, 2005).

‘Mountain Fresh’, ‘Mountain Spring’, ‘Florida 47’, and ‘BHN543’ are high value

varieties commonly used for fresh market tomato production in Ohio. ‘BHN543’

matures at mid-season, producing large or extra large, globe-shaped fruit. The plant has determinate growth and is reported to be resistant to Fusarium wilt

races 1 and 2, Meloidogyne incognita, M. javanica, M. arenaria, and Verticillium

dahliae race 1. ‘Florida 47’ is a mid-season variety that produces large fruit that are flattened at the poles. The variety has determinate growth and shows resistance to Fusarium wilt races 1 and 2, Stemphylium sp., and Verticillium dahliae race 1. ‘Mountain Spring’ is a mid-season variety that produces extra

large, crack resistant, oblate fruit. The variety has determinate growth and resistance to Fusarium wilt races 1 and 2, Stemphylium, and Verticillium dahliae

race 1. ‘Mountain Fresh’ is a mid-season variety that produces large fruit, from

oblate to globose in shape. The plant has determinate growth (Anonymous,

2000).

A few elements have been determined to be essential for plant growth. According

to their relative concentration in plant tissue, essential nutrients are classified as

macronutrients (nitrogen, potassium, calcium, magnesium, phosphorus, sulfur

and silicon) and micronutrients (chlorine, iron, boron, manganese, sodium, zinc, copper, nickel and molybdenum). Absence or inadequate supply of essential nutrients prevents a plant from completing its life cycle or causes nutritional disorders manifested by characteristic symtoms (Taiz and Zeiger, 2002).

Nitrogen-deficient plants grow slowly. Leaves are small, thick, and light green.

Stems are thick and hard. Flower buds generally drop off, fruits are small, and

yield is reduced (Scholberg et al. 2000; Wilcox, 1993). Phosphorous-deficient plants grow slowly, and leaves become dark, with purple interveinal tissue on the underside. Stems become slender and hard (Wilcox, 1993). Potassium-deficient

tomato plants have small stems and shortened internodes. Young leaves are

dark green, crinkled and curled. Old leaves are chlorotic and bronzed. Fruits are slender and surfaces are blotchy. They ripen unevenly and drop off the plant after ripening (Wilcox, 1993). Calcium deficiency results in reduced growth of tomato. Root tips die and seedling leaves show upward cupping and have necrotic margins. Limited calcium availability during the fruiting stage results in the disorder blossom end rot, since translocation from vegetative portions and root uptake cannot compensate for the deficiency in the fruit during development

(Wilcox, 1993).

Total nitrogen uptake by tomato averages 197 kg/ha (Wilcox, 1993). The variety

‘Mountain Spring’ requires a nitrogen rate of only 67 kg/ha to produce 48 ton/ha

(Taber, 1998). Nitrogen accumulation is highest during the vegetative stage (185

mg/plant/day). Average nitrogen concentration in plant tissue is relatively high

during seedling (4.38% dry weight) and vegetative stages (4.30% dry weight) and

decreases at flowering (3.99% dry weight) and fruiting (3.34% dry weight) stages.

Nitrogen concentrations in petiole sap of the tomato variety ‘Mountain Spring’ at

seedling, flowering, and fruiting stages was reported to be 180, 203, and 90 ppm

respectively (Taber, 1998). Average total phosphorous uptake by tomatoes is 26

kg/ha or about 1.052 g/plant, of which 75% is in the fruit and the rest in leaves

and stems. Phosphorus uptake is particularly rapid during the fruiting stage

(Wilcox, 1993). Average total potassium uptake is 296 kg/ha or about 12 g/plant. 3

Nearly 70% of potassium is in the fruit. Potassium uptake is rapid during vegetative and fruiting stages (Wilcox, 1993).

During vegetative growth, the root system, stems and leaves are developed (first

42 days) then, flowers appear and develop. Fruits begin to develop about 55 days after emergence. During flowering and fruiting, a rapid increase in dry weight of stems and leaves occurs up to 70 days after emergence, with little increase after 77 days. Accumulation of dry weight in the fruits is almost linear from day 70 to 105. Fruit ripening begins at 84 days and progresses to ripeness at 112 days (Wilcox, 1993).

Water management influences uptake and utilization of nutrients, development of plants, disease development, and quality and quantity of tomato production.

Irrigation is particularly important in tomato in late summer when there is a reduction in rainfall. Soil moisture deficiencies during vegetative, flowering or fruiting stages resulted in yield reductions of about 25, 52 and 43%, respectively

(Rutledge et al. 1999). Irrigation delivery system (sprinklers vs. furrow vs. buried drip) and irrigation frequency have a significant impact not only on quality and quantity of tomato production, but also on the development of diseases (Strange et al. 2000). Drip irrigation is one of the most common delivery systems in which

water under pressure flows through a pipe. This system utilizes less water than

over-the-top irrigation. Drip irrigation places water directly where it is needed and reduces diseases since it avoids wetting the foliage. One disadvantage of the system is that it requires clean water since soil and mineral deposits result in line

blockage and non-uniform water application (Rutledge et al. 1999).

The use of black plastic is widely accepted for growing tomatos. Plastic mulch

controls weeds and certain diseases, conserves moisture, and increases quality

and quantity of marketable fruit. Plastic mulch is generally installed over a 100

cm wide x 12 cm high bed. The irrigation system is set up ahead of laying the

plastic. Transplant holes are punched through the plastic. The main

disadvantage of plastic mulch is the expense associated with installation,

removal, and disposal of the black plastic (Rutledge et al. 1999).

Nearly 200 biotic diseases have been reported on tomato throughout the USA

(Jones et al. 1997). Worldwide, one of the most important foliar bacterial

diseases is bacterial spot caused by Xanthomonas euvesicatoria (formerly X.

campestris pv. vesicatoria, X. axonopodis pv. vesicatoria, race T1, phenotypic

group A), X. vesicatoria (formerly X. vesicatoria, race T2, phenotypic group B), X.

perforans (formerly X. axonopodis pv. vesicatoria, race T3, phenotypic group C),

and X. gardneri (formerly X. vesicatoria, race T4, phenotypic group D) (Jones et

al. 2004a; Jones et al. 2004c).

Xanthomonas euvesicatoria and X. vesicatoria are widely distributed throughout

the world (Bouzar et al. 1994; Bouzar et al. 2004). These species of

Xanthomonas can reduce tomato production under environmental conditions that occur in Ohio (Abbasi et al. 2002; Sahin, 1997), Alabama and the southeastern

United States (Cambell et al. 1998). In Ohio, three species have been reported as causal agents of bacterial leaf spot: X. euvesicatoria, X. vesicatoria, and X.

perforans (Sahin, 1997). X. perforans has been reported in several tomato production areas in the United States and in Mexico. X. gardneri was isolated in

Costa Rica, and reported from Michigan and Brazil (Bouzar et al. 2004).

Xanthomonas euvesicatoria, X. vesicatoria and X. perforans are gram negative,

aerobic, oxidase negative, catalase positive, rod-shaped bacteria with a single

polar flagellum. The yellow color of Xanthomonas spp. colonies is caused by

brominated arylpolyene esters. Colonies are mucoid on yeast extract-dextrose-

CaCO3 medium (Schaad et al. 2001). These species can be distinguished easily

by physiological tests, by distinctive SDS PAGE profile, and DNA similarities. X.

euvesicatoria is weakly amylolytic and pectolytic, able to use cis-aconitic acid

and has a distinctive SDS PAGE profile. X. vesicatoria is strongly amylolytic and pectolytic, utilizes acetic acid, but does not utilize cis-aconitate. X. perforans is

strongly amylolytic and pectolytic, but does not utilize acetic acid and has a

distinctive SDS PAGE profile (Jones et al. 2004c).

Xanthomonas euvesicatoria and X. vesicatoria induce brown, roughly circular

spots on stems, leaves, and petioles of tomato and pepper plants. The lesions

are generally smaller than 3 mm in diameter, surrounded by a prominent halo, and become water-soaked during periods with high moisture. When environmental conditions are optimal for disease development, the lesions

coalesce. On affected fruits symptoms begin as minute slightly raised blisters

that become brown and scab like (Jones et al. 1997; Basim et al. 2004).

The pathogens of bacterial spot survive on infected crop residue at least 6

months (Bashan et al. 1982; Jones et al. 1997), and or tomato volunteers and weeds (Jones et al. 1986). The pathogen persists for 16 days in sandy loam soil, and at least 3 months in tomato seed (Bashan et al. 1982). Development of bacterial spot disease is favored by temperatures of 24-30 oC and abundant

moisture. Dissemination is mainly by wind-driven rain droplets and clipping of

transplants. The bacteria penetrate plant tissue through stomata and wounds

(Jones et al. 1997).

Partial control of bacterial leaf spot is achieved by crop rotation, disease free

transplants, destruction of infected leaves, and eradication of weeds or volunteer

plants. Pesticides with efficacy against bacterial leaf spot include bactericides

such as Cuprofix 40 DF (2.24 kg/ha sprayed every 7-10 days), and the plant

activator Actigard 50WG (35 g/ha sprayed every 7-10 days) (Lewis Ivey et al.

2005; Briceno and Miller, 2004; Jones et al. 2004b; Obradovic et al. 2005;

Vavrina et al. 2004).

Stem rot of tomato is a bacterial disease that is characterized by breakdown of

pith, vascular discoloration, external black lesions, eventual wilting and death of

the plant. Bacteria that cause stem rot are Erwinia carotovora (Speights et al.

1967), E. carotovora ssp. carotovora (Dhanvantari and Dirks, 1987),

Pseudomonas viridiflava (Lukezic et al. 1983), P. corrugata (Scarlett et al. 1978)

and P. cichorii (Wilkie and Dye, 1974).

Pseudomonas cichorii is a gram negative, rod-shaped bacterium with one to

several polar flagella. The bacterium is aerobic, levan negative, oxidase positive,

non-pectolytic and arginine dihydrolase negative, inducing a positive hypersensitivity reaction on tobacco. Colonies produce diffusable fluorescent pigments that are visible on iron-deficient medium such as King’s medium B, and

Pseudomonas agar F (Braun-Kiewnick and Sands, 2001).

Pseudomonas cichorii induces elongated dark lesions on the surface of tomato stems. Lesions may extend along petioles and leaves, causing dark green water- soaked blotches without halos. Internally, vascular tissues show dark brown discoloration. Infected stem pith is brown and watery then disintegrates leaving a hollow stem. Dark brown spots appear on fruits. Eventually the plant wilts and dies (Wilkie and Dye, 1974).

Pseudomonas cichorii infection is favored by extended leaf wetness and temperatures around 20 oC. Symptoms appear within 48 hours after inoculation under these conditions (Wilkie and Dye, 1974). Pseudomonas cichorii severity may be reduced by applying copper hydroxide and Actigard (Ustun, 2004).

Early blight (Alternaria solani, A. alternata) has been reported from most tomato production areas. Alternaria solani is present in England, India, Australia, and the

United States. It is particularly damaging in the central states (Jones, 1997). This

fungus produces septate and branched mycelium. Conidiophores are simple and

dark. Conidia are obclavate to elliptical, 150 - 300 x 15 – 19 μm, dark, with both

transversal and longitudinal septa, and a very long beak (almost the size of the

body), developing through pores in the outer wall at the apex of conidiophore

(Barnett and Hunter, 1998; Evans and Howards, 1994; Menzies and Jarvis,

1994a). The pathogen affects leaves, stems, flowers, and fruits. On leaves, A.

solani induces small circular spots with concentric rings, becoming irregular in shape and generally surrounded by a yellow area. Black elongated, slightly sunken lesions occur on stems and pedicels. Black, leathery, sunken lesions appear on the fruit, generally around the calix, wounds or cracks. Heavy fruit load, age, and inadequate nutrition increase tomato susceptibility to early blight

(Jones, 1997; Menzies and Jarvis, 1994a; Pitblado and Howard, 1994).

Alternaria solani survives in plant residue, tomato seed or volunteer Solanaceous

weeds. Spores are spread by wind, rain, and infested plant debris. Conidia

germinate within 2 hours in water at 6 - 34 oC and penetrate directly through the

cuticle. Lesions usually appear on the older leaves in 2 - 3 days and spread to

the upper leaves. Extended leaf wetness and high temperature (24 – 29 oC) favor

the disease, which can completely defoliate tomato plants and expose fruit to

sunscald and other diseases, and reduce yield (Jones, 1997; Pitblado and

Howard, 1994).

Management of early blight may be achieved by crop rotation, use of disease-

free transplants, minimizing plant injury, balancing nutrition, destruction of

infected leaves, eradication of weeds or volunteer plants, irrigation, and use of

resistant varieties (e.g. ‘HY9478’, ‘Malinta’ and ‘Medalist’) and fungicides (Jones,

1997; Menzies and Jarvis, 1994a; Pitblado and Howard, 1994).

Septoria leaf spot (Septoria lycopersici) is another common disease of tomato in

Ohio that can significantly reduce yield if not properly managed. It is widely distributed throughout the world, having been reported from Europe, Asia, Africa,

Australia and North and South America (Stevenson, 1997). Septoria lycopersici produces septate and immerse mycelium. The pycnidia are globose, about 66

μm in width with a thin wall of textura angularis and a smaller-celled inner layer.

The ostiole is circular and central. The conidiogenous cells are holoblastic, hyaline, ampuliform, doliform or short cylindrical that produce hyaline, multiseptate, filiform 67 x 3.2 μm conidia (Menzies and Jarvis, 1994b; Sutton,

1980).

Septoria leaf spot is a common foliar disease late in the growing season. Circular

brown spots (2-5 mm) generally with a narrow halo appear on the lower leaves.

The center of old lesions turns light brown with dark margins and black pycnidia

appear within the lesion. Small circular, water-soaked spots occur in stems and

petioles. Under favorable conditions, the disease spreads from lower leaves to stems and upper leaves, inducing defoliation and causing exposure of fruit to sunscald and other pathogens, with potentially heavy yield losses (Menzies and

Jarvis, 1994b; Pitblado, 1994; Stevenson, 1997). Septoria lycopersici survives on tomato seed, wood stakes and plant debris. Spores are spread by water, workers, equipments, insects, wind and infested plant debris. Conidia penetrate tomato plants through stomates. Symptoms appear within 6 days and pycnidia are produced about 14 days after inoculation. Temperatures between 20 - 25 oC

and extended leaf wetness favor the disease. At 100% relative humidity, 10 days

are required to complete the life cycle (Menzies and Jarvis, 1994b; Pitblado,

1994; Stevenson, 1997). As for early blight, management of septoria leaf spot is

achieved by crop rotation, use of disease-free transplants, elimination of weeds,

crop debris removal, minimizing plant injury, balancing nutrition, timing of

irrigation, and application of fungicides (Stevenson, 1997; Pitblado, 1994).

Most growers rely on fungicides/bactericides to manage foliar diseases and

achieve acceptable yield of tomato. However, new resistant strains of

Xanthomonas species causing bacterial spot (Cambell et al. 1997; Jones et al.

1991; Marco and Stall, 1983) and fungal and oomycetes pathogens (Day et al.

2004; Grech, 1990; Yun et al. 1999) appear, reducing the effectiveness of these products. Furthermore, alternative measures of disease control that require fewer or low-risk pesticides are needed to prevent human health problems and damage to the environment (Bell et al. 2000; Cambell et al. 1997). Under these circumstances, the use of pathogen-free tomato seed, as well as cultural and biological control tactics are needed for integrated management of diseases caused by both bacteria and fungi. The effectiveness of hot-water treatment of tomato seed to reduce bacterial spot has been demonstrated (Miller et al. 2004).

Optimizing the fertilizer program may be a practical approach to help manage bacterial diseases (Harkness and Marlatt, 1970). Nitrogen has been studied in relation to disease intensity in several crop-disease combinations (Huber and

Watson, 1974). High rates of nitrogen induce a reduction (Chase and Poole,

1987; Harkness and Marlatt, 1970; Chase and Jones, 1986) or increase

(Bachelder et al. 1956; Nayudu and Walker, 1961; Thomas, 1965; Haygood et al.

1982; Jones et al. 1985) in the susceptibility of the host to the pathogen.

Promising results have been reported using biorational products such as

bacteriophages (Snowden, 2004; Jackson, 2004; Jones et al. 2004b), Bacillus

amyloliquefaciens (Briceno and Miller, 2004), B. subtilis (Briceno and Miller,

2004), Pseudomonas syringae (Briceno and Miller, 2004), P. fluorescens

(Briceno and Miller, 2004), Bacillus subtilis strains QRD131, 132, 137 and 141

(Highland, 2004) against bacterial spot and foliar diseases of tomato.

The use of plant growth-promoting rhizobacteria (PGPR) is a potentially attractive

approach to disease management and improved crop productivity in sustainable

agriculture, since PGPR have been reported to increase yield and protect crops from disease simultaneously (Ramamoorthy et al. 2002; Raupach, 1998). PGPR are strains of bacteria that live in the rhizosphere, stimulate plant growth, and improve stand under stress conditions (van Loon et al. 1998). Several classes of

PGPR have been reported to enhance plant growth and suppress pathogens: biofertilizers, such as Rhizobium, Sinorhizobium, Mesorhizobium,

Bradyrhizobium, Azorhizobium and Allorhizobium fix nitrogen into a form that can

be used by the plant (Bloemberg and Lugtenberg, 2001). Phytostimulator PGPR

enhance plant growth, usually by the production of phytohormones such as

auxins (Asghar et al. 2002). Biocontrol agents protect plants (niche exclusion,

antibiotic synthesis, competition for nutrients, antagonism) and enhance defense

mechanisms by inducing systemic resistance (Bloemberg and Lugtenberg,

2001). Induced systemic resistance (ISR) is a phenomenon characterized by: 1)

absence of toxic effects of the PGPR on the pathogen, 2) suppression of ISR by

gene expression inhibitors such as actinomycin, 3) requirement of an interval of

time between application of inducer and protection, 4) absence of a dose- response correlation, 5) non-specific protection that is local and/or systemic, and

6) dependence on genotype of the plant (van Loon et al. 1998).

Rhizobacteria that are known to induce ISR include: Pseudomonas putida 89B-

27, P. fluorescens 89B-27, Serratia marcescens 90-166 (Liu et al. 1995;

Raupach et al. 1996), B. amyloliquefaciens strain IN937a, B. subtilis strain

IN937b, B. pumilis strain SE34, Kluyvera cryocrescens strain IN114 (Benhamou et al. 1998; Zehnder et al. 2000), and Bacillus sp. isolates B and J (Braun-

Kiewnick et al. 1998). Induction of systemic resistance and enhacement of plant

growth by Bacillus subtilis and B. amyloliquefaciens have been reported on

cabbage (Wulff et al. 2002), onion (Reddy and Rahe, 1989a, 1989b), sugar beet

(Braun-Kiewnick et al. 1998; Collins et al. 2003), tomato varieties ‘Solar Set’

(Reddy et al. 2000; Martinez-Ochoa, 2000), and ‘Mountain Pride’ (Zehnder et al.

2000), tobacco, and cucumber (Reddy et al. 2000).

Several studies have been performed to elucidate the mechanism of biocontrol

by B. subtilis and B. amyloliquefaciens. Braun-Kiewnick et al. (1998) suggested

that ISR is a mode of action for Bacillus subtilis. The authors found that application of B. subtilis strains J and B or benzothiadiazole to sugar beet induced systemic resistance and reduction of cercospora leaf spot. Protein patterns, enzyme activity and elevated concentration of peroxidase in crude extracts were observed in Bacillus-inoculated plants. Recently, two studies presented additional evidence in favor of ISR hypothesis. Collins et al. (2003) reported that hyperparasitism, competition, and antibiosis are not important mechanisms of cercospora leaf spot control in sugar beet influenced by Bacillus subtilis, since its population density was not correlated with Cercospora beticola disease severity. Wulff et al. (2002) found that in vitro inhibitory effects of Bacillus on Xanthomonas campestris pv. campestris were not correlated to its ability to reduce black rot incidence in cabbage.

Inconsistency of disease control is one common characteristic of biological

control agents (Weller, 1988). B. subtilis strain MBI600 is recognized as

biological control agent (EPA, 2004). However, when MBI600 was inoculated

onto sugar beet, no significant antagonism against Pythium ultimun was observed (Schmidt et al. 2004). Bacillus subtilis strain GBO3, a successful

biological control agent in numerous crops (Reddy et al. 2000; Martinez-Ochoa,

2000) was reported to be ineffective against bacterial leaf spot in tomato variety

‘Florida 47’ (Vavrina et al. 2004).

The main objective of this study was to evaluate biological, cultural and chemical

approaches to tomato management. Antibiotic-resistant mutants of commercially

available B. subtilis (strains MBI600 and GBO3) and B. amyloliquefaciens (strain

IN937) were selected and characterized based on physiological, morphological,

and molecular characteristics. Under greenhouse conditions, the efficiency of

these PGPR in colonizing tomato seedling roots and their effects on tomato

seedling growth promotion and induced resistance to bacterial leaf spot were

evaluated. In the field, the single and combined effects of mulch, irrigation,

fungicide application program and Bacillus spp. were evaluated in orded to

optimize resources to control foliar diseases induced by Alternaria solani,

Septoria lycopersici and bacterial stem rot (Pseudomonas cichorii) and increase

marketable yield. The integrated effect of systemic acquired resistance (SAR)

and nitrogen concentration on Bacillus spp. population density have not been

explored previously. In order to evaluate their compatibility, the effects of the

SAR inductor Actigard and nitrogen on Bacillus spp. population density were

evaluated. Finally, the effect of two rates of irrigation and Bacillus spp. on the

tomato variety ‘Mountain Spring’ were evaluated in terms of nitrogen absorption

and bacterial stem rot suppression, mechanisms that may explain the increase in susceptibility to bacterial stem rot observed on GBO3+IN937 inoculated tomato plants in the field.

LIST OF REFERENCES

Abbasi, P.A., Al-Dahmani, J., Sahin, F., Hoitink, H.A.J., and Miller, S.A. 2002. Effect of compost amendments on disease severity and yield of tomato in conventional and organic production systems. Plant Disease 86 (2): 156- 161.

Anonymous. Siegers 2000. Hybrid round red tomato. http://www.siegers.com/shop/kind.asp?kind_id=BBB.

Asghar, H.N., Zahir, Z.A., Arshad, M. and Khaliq, A. 2002. Relationship between in vitro production of auxins by rhizobacteria and their growth-promoting activities in Brassica juncea L. Biol Fertil Soils 35: 231-237.

Bachelder, S., Daines, R.H. and Bartley, C.E. 1956. Effect of nutrition on the incidence of bacterial leaf spot of peach. (Abstr.) Phytopathology 46: 240.

Barnett, H.L. and Hunter, B.B. 1998. Illustrated genera of imperfect fungi. APS press. Fourth edition. St. Paul, Minnesota USA: 218 p.

Bashan, Y., Diab, S. and Okon, Y. 1982. Survival of Xanthomonas campestris pv. vesicatoria in tomato seeds and roots in symptomless and dry leaves in non-host plants and in the soil. Plant and Soil 68: 161-170.

Basim, J.B., Jones, G., Minsavage, V., and Dickstein, E.R. 2004. Bacterial spot of tomato and pepper caused by Xanthomonas axonopodis pv. vesicatoria in the Western Meditterranean region of Turkey. Plant Disease 88 (1): 85.

Bell, C.E., Masiunas, J.B., and Stall, W.M. 2000. My view. Weed Science 48: 1.

Benhamou, N., Kloepper, J.W., and Tuzun, S. 1998. Induction of resistance against Fusarium wilt of tomato by combination of chitosan with an endophytic bacterial strain: ultrastructure and cytochemistry of the host response. Planta 204: 153-168.

Bloemberg, G.V., and Lugtenberg, B.J.J. 2001. Molecular basis of plant growth promotion and biocontrol by rhizobacteria. Current Opinion in Plant Biology 4: 343-350.

Bouzar, H., Jones, J.B., Stall, R.E. Hodge, N.C., Minsavage, G.V., Benedict, A.A. and Alvarez, A.M. 1994. Physiological, chemical, serological, and pathogenic analyses of a wordwide collection of Xanthomonas campestris pv. vesicatoria strains. Phytopathology 84: 663-671.

Bouzar, L.H., Minsavage, G.V., Stall, R.E., and Schaad, N.W. 2004. Bacterial spot- worldwide distribution, importance and review. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 21 pp.

Braun-Kiewnick A., Kiewnick, S., and Jacobsen, B.J. 1998. Induction of systemic resistance by antagonistic Bacillus sp. and the chemical inducer benzothiadiazole controls Cercospora leaf spot of sugar beet. Phytopathology 88: s10.

Braun-Kiewnick, A. and Sands, D.C. 2001. Gram-negative bacteria. Pseudomonas. In Schaad, N.W., Jones, J.B., and Chun, W. 2001. Laboratory guide for identification of plant pathogenic bacteria. Third edition. APS press. USA: 82 – 120 pp.

Briceno, M.E.G. and Miller, S.A. 2004. Evaluation of biorational options for bacterial spot management during transplant production. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 21.

Cambell, H.L., Wilson, M. and Byrne J.M. 1997. Novel chemicals control bacterial spot and may reduce copper contamination problems. Highlights of agricultural research 44 (2): http://www.ag.auburn.edu/aaes/communications/highlights/summer97/che micals.htm.

Cambell, L. Moss, B., Byrne, J., Dianese, A., and Wilson, M. 1998. The good, the bad and the ugly: biocontrol of bacterial spot of tomato. Highlights of agricultural research 45 (2): http://www.ag.auburn.edu/aaes/communications/highlights/summer98/bact erial.html.

Chase, A.R. and Jones, J.B. 1986. Effects of host nutrition, leaf age, and preinoculation light levels on severity of leaf spot of dwarf schefflera caused by Pseudomonas cichorii. Plant Disease 70: 561-563.

Chase, A.R. and Poole, R.T. 1987. Effects of fertilizer rate on severity of Xanthomonas leaf spot of schefflera and dwarf schefflera. Plant Disease 71: 527-529.

Collins, D.P., Jacobsen B.J., and Maxwell, B. 2003. Spatial and temporal population dynamics of a phyllosphere colonizing Bacillus subtilis biological agent of sugar beet cercospora leaf spot. Biological control 26: 224-232.

Day, J.P., Wattier, R.A.M., Shaw, D.S. and Shattock, R.C. 2004. Phenotypic and genotypic diversity in Phytophthora infestans on potato in Great Britain, 1995-98. Plant Pathology 53: 303-315.

Dhanvantari, B.N. and Dirks, V.A. 1987. Bacterial stem rot of greenhouse tomato: Etiology, spatial distribution, and the effect of high humidity. Phytopathology 77: 1457-1463.

EPA. 2004. Bacillus subtilis MBI600 129082. http://www.epa.gov/pesticides/biopesticides/ingredients/factsheets/factshe et_129082.htm.

Evans, I.R. and Howard, R.J. 1994. Early blight. In Howard, R.J., Garland, J.A. and Seaman, W.L. 1994. Diseases and pest of vegetable crops in Canada. The Canadian Phytopathological Society. Entomological Society of Canada. Canada: 232 pp.

FAO. 2004. FAOSTAT agriculture data. http://apps.fao.org/page/collections?subset=agriculture.

Grech, N.M. 1990. Fungicide resistance in plant pathogenic fungi: the problems and ways to cope. Calif-Grow 14(4): 36-37.

Harkness, R.W., and Marlatt, R.B. 1970. Effect of nitrogen, phosphorus and potassium on growth and Xanthomonas disease of Philodendron oxycardium. J. Amer. Soc. Hort. Sci. 95 (1): 37-41.

Haygood, R.A., Strider, D.L., and Nelson, P.V. 1982. Influence of nitrogen and potassium on growth and bacterial leaf blight of Philodendron selloum. Plant Disease 66: 728-730.

Highland, H.B. 2004. Yield and disease control characteristics after the use of serenade biofungicide in fresh market tomatoes. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 46 pp.

Huber, D.M. and Watson, R.D. 1974. Nitrogen form and plant disease. Annual Review of Phytopathology 12: 139-165.

Jackson, L. 2004. Bacteriophage: A viable bacteria control solution. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 53 pp.

Jones J.B., Momol, M.T., Obradovic, A., Balogh, B., and Olson S.M. 2004b. Bacterial spot management on tomatoes. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 55 pp.

Jones J.B., Stall, R.E. and Zitter, T.A. 1997. Compendium of Tomato Diseases. APS press. Minnesota. USA: 73 p.

Jones, J. B., Woltz, S. S., Jones, J. P., and Portier, K. L. 1991. Population dynamics of Xanthomonas campestris pv. vesicatoria on tomato leaflets treated with copper bactericides. Phytopathology 81:714-719.

Jones, J.B., Chase, A.R., Harbaugh, B.K. and Raju, B.J. 1985. Effect of leaf wetness, fertilizer rate, leaf age, and light intensity before inoculation on bacterial leaf spot of chrysanthemum. Plant Disease 69: 782-784.

Jones, J.B., Lacy G.H., Bouzar, H., Minsavage, G.V., Stall, R.E., and Schaad, N.W. 2004a. Bacterial spot-worldwide distribution, importance and review. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 21 pp. 22

Jones, J.B., Lacy, G.H., Bouzar, H., Stall, R.E., and Schaad, N.W. 2004c. Reclassification of the Xanthomonads associated with bacterial spot disease of tomato and pepper. System. Appl. Microbiol. 27: 755-762.

Jones, J.B., Pohronezny, K.L. Stall, R.E. and Jones, J.P. 1986. Survival of Xanthomonas campestris pv. vesicatoria in Florida on tomato crop residue, weeds, and volunteer tomato plants. Phytopathology 76: 430 434.

Jones, J.P. 1997. Early blight. In Jones J.B., Stall, R.E. and Zitter, T.A. 1997. Compendium of Tomato Diseases. APS press. Minnesota. USA: 13 pp.

Lewis Ivey, M.L., Mera, J.R., and Miller, S.A. 2005. Evaluation of fungicides and bactericides for the control of foliar and fruit diseases of processing tomatoes, 2004. Vegetable disease management research reports 2004. Plant Pathology series No. 126. The Ohio State University. 4 p.

Liu, L., Kloepper, J.W., and Tuzun, S. 1995. Induction of systemic resistance in cucumber against Fusarium wilt by plant growth-promoting rhizobacteria. Phytopathology 85: 695-698.

Lukezik, F.L., Levine, R.G. and MacNab, A.A. 1983. Pseudomonas viridiflava associated with stem necrosis of mature tomato plants (Abst) Phytopathology 73: 370.

Marco, G. M., and Stall, R. E. 1983. Control of bacterial spot of pepper initiated by strains of Xanthomonas campestris pv. vesicatoria that differ sensitivity to copper. Plant Disease. 67:779-781.

Martinez-Ochoa, N. 2000. Biological control of the root-knot nematode with rhizobacteria and organic amendments. Ph.D. dissertation. Auburn University. Alabama: 120 p.

Menzies, J.G. and Jarvis, W.R. 1994a. Early blight. In Howard, R.J., Garland, J.A. and Seaman, W.L. 1994. Diseases and pest of vegetable crops in Canada. The Canadian Phytopathological Society. Entomological Society of Canada. Canada: 346 pp.

Menzies, J.G. and Jarvis, W.R. 1994b. Septoria blight. In Howard, R.J., Garland, J.A. and Seaman, W.L. 1994. Diseases and pest of vegetable crops in 23

Canada. The Canadian Phytopathological Society. Entomological Society of Canada. Canada: 350 pp.

Miller, S.A., Lewis Ivey, M.L. and Hirzel, E. 2004. Bacterial disease management in tomatoes: the role of clean seed. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 70 pp.

Nayudu, M.V. and Walker, J.C. 1961. Tomato leaf composition in relation to bacterial leaf spot. Phytopathology 51: 368-372.

Obradovic A., Jones, J.B., Momol, M.T., Olson, S.M., Jackson, L.E., Balogh, B., Guven, K., and Iriarte, F.B. 2005. Integration of biological control agents and systemic acquired resistance inducers against bacterial spot on tomato. Plant Disease 89: 712-716.

Pitblado, R.E. 1994. Septoria leaf spot. In Howard, R.J., Garland, J.A. and Seaman, W.L. 1994. Diseases and pest of vegetable crops in Canada. The Canadian Phytopathological Society. Entomological Society of Canada. Canada: 275 pp.

Pitblado, R.E. and Howard, R.J. 1994. Early blight. In Howard, R.J., Garland, J.A. and Seaman, W.L. 1994. Diseases and pest of vegetable crops in Canada. The Canadian Phytopathological Society. Entomological Society of Canada. Canada: 274 pp.

Ramamoorthy, V., Raguchander, T., and Samiyappan, R. 2002. Enhancing resistance of tomato and hot pepper to Pythium diseases by seed treatment with fluorescent pseudomonads. European Journal of Plant Pathology 108: 429-441.

Raupach, G.S. 1998. Induced systemic resistance as the principal component of integrated pest management of Cucumis sativus L. by PGPR. Ph.D. dissertation. Auburn University. Alabama. USA: 120 p.

Raupach, G.S., Lui, L., Murphy, J.F., Tuzun, S. and Koeppler, J.W. 1996. Induced systemic resistance in cucumber against cucumber mosaic cucumovirus using plant growth-promoting rhizobacteria (PGPR). Plant Disease 80: 891-894.

Reddy, M.S. and Rahe, J.E. 1989a. Growth effects associated with seed bacterization not correlated with populations of Bacillus subtilis inoculant in onion seedling rhizospheres. Soil Biology and Biochemistry 21(3): 373- 378.

Reddy, M.S., and Rahe, J.E. 1989b. Bacillus subtilis B-2 and selected onion rhizobacteria in onion seedling rhizospheres: Effects on seedling growth and indigenous rhizosphere microflora. Soil Biology and Biochemistry 21(3): 379-383.

Reddy, M.S., Ryu, C. M., Zhang, S., Yan, Z. and Kloepper, J. W. 2000. Aqueous formulations of plant growth-promoting rhizobacteria for control of foliar pathogens. http://www.ag.auburn.edu/~mreddy/.

Rutledge, A.D., Wills, J.B., and Bost, S. 1999. Commercial tomato production. Agricultural Extension Service. The Universiy of Tennessee. PB 737: 31 p.

Sahin, F. 1997. Detection, identification and characterization of strains of Xanthomonas campestris pv. vesicatoria by traditional and molecular methods, and resistance in Capsicum species to Xanthomonas campestris pv. vesicatoria pepper race 6. Ph.D. Dissertation. The Ohio State University. 182 p.

Scarlett, C.M., Fletcher, J.T, Roberts, P. and Lelliott, R.A. 1978. Tomato pith necrosis caused by Pseudomonas corrugata n.sp. Ann. Appl. Biol. 88:105- 114.

Schaad, N.W., Jones, J.B., and Lacy, G.H. 2001. Gram negative bacteria. In Schaad, N.W., Jones, J.B., and Chun, W. 2001. Laboratory guide for identification of plant pathogenic bacteria. Third edition. APS press. USA: 175-200 pp.

Schmidt, C.S., Agostini, F., Leifert, C., Killham, K., and Mullins, C.E. 2004. Influence of soil temperature and matric potential on sugar beet seedling colonization and suppression of Pythium damping-off by the antagonistic bacteria Pseudomonas fluorecens and Bacillus subtillis. Phytopathology 94: 351-363.

Scholberg, J., McNeal, B.L. Boote, K.J., Jones, J.W., Locascio, S.J., and Olson, S.M. 2000. Nitrogen stress effects on growth and nitrogen accumulation by field-grown tomato. Agronomy Journal 92: 159 – 167.

Snowden, C. 2004. Agriphage: safe and effective bacterial control. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 47 pp.

Speights, D.E., Hallivell, R.S., Horne, C.W. and Hughes, A.B. 1967. A bacterial stem rot of greenhouse tomato plants. Phytopathology 57: 902-904.

Stevenson, W.R. 1997. Septoria leaf spot. In Jones J.B., Stall, R.E. and Zitter, T.A. 1997. Compendium of Tomato Diseases. APS press. Minnesota. USA: 22 pp.

Strange, M.L., Schrader, W.L. and Hartz, T.K. 2000. Fresh market tomato production in California. Publication 8017. University of California. Division of Agriculture and Natural Resources: 8 http://anrcatalog.ucdavis.edu/pdf/8017.pdf.

Sutton, B.C. The Coelomycetes. 1980. Commonwealth Mycological Institute. Kew, Surrey, England. 696 p.

Taber, H.G. 1998. Petiole sap nitrate sufficiency values for fresh market tomato production. http://www.hort.iastate.edu/facultystaff/meetfaculty/taber/nitrate/index.php.

Taiz, L., and Zeiger, E. 2002. Plant Physiology. Sinauer Associates. Third edition. Sunderland Massachusetts, USA: 68-86.

Thomas, C.A. 1965. Effect of photoperiod and nitrogen on reaction of sesame to Pseudomonas sesami and Xanthomonas sesami. Plant Disease Reporter 49 (2): 119-120.

USDA. 2005. Vegetables 2004 summary. United States Department of Agriculture and National Agriculture Statistics Service. 1-2 (05) http://usda.mannlib.cornell.edu/reports/nassr/fruit/pvg-bban/vgan0105.pdf.

Ustun, N., Demir, G., and Saygili, H. 2004. Possibilities for control of tomato pith necrosis by using copper compounds and plant activators. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 105 pp.

Van Loon, L.C., Bakker, P.A.H.M., and Pieterse, C.M.J. 1998. Systemic resistance induced by rhizosphere bacteria. Annual Review of Phytopathology 36: 453-483.

Vavrina, C.S., Roberts, P.D., Kokalis-Burelle, N., and Ontermaa, E.O. 2004. Greenhouse screening of commercial products marketed as systemic resistance and plant growth promotion inducers. HortScience 39 (2): 433- 437.

Weller, D.M. 1988, Biological control of soilborne plant pathogens in the rhizosphere with bacteria. Annual Review of Phytopathology 26 379-407.

Wilcox, G.E. 1993. Tomato. In Bennet, W.F. 1993. Nutrient deficiencies and toxicities in crop plants. APS press. St. Paul Minnesota. USA: 137-142.

Wilkie, J.P. and Dye, D.W. 1974. Pseudomonas cichorii causing tomato and celery diseases in New Zealand. N.Z. Journal of Aricultural Research 17: 123 – 130.

Wulff E.G., Mguni, C.M., Mansfeld-Giese, K., Fels, J., Lubeck, M., and Hockenhull, J. 2002. Biochemical and molecular characterization of Bacillus amyloliquefaciens, B. subtilis and B. pumilus isolates with distinct antagonistic potential against Xanthomonas campestris pv. campestris. Plant Pathology 51: 574-587.

Yun, L.T., Mizubuti, E. and Fry, W.E. 1999. Genetics of metalaxyl resistance in Phytophthora infestans. Fungal Genetics and Biology 26: 118-130.

Zehnder, G.W., Yao, C., Murphy, J.F., Sikora, E.R. and Kloepper, J.W. 2000. Induction of resistance in tomato against cucumber mosaic cucumovirus by plant growth-promoting rhizobacteria. Biocontrol 45: 127-137.

CHAPTER 1

Morphological, Physiological, and Molecular Characterization of Rifampicin- Resistant Mutants of Bacillus spp. Plant Growth-Promoting Rhizobacteria

INTRODUCTION

Bacillus is a gram positive, rod-shaped, endospore-forming bacterium with one

chromosome (Kunst et al. 1997; Nicholson et al. 2000) that has the ability to switch from binary fission to asymmetric division (Levin and Losick, 2000).

Several species of this genus are important human and animal pathogens, others are commonly used in the detergent and food industry (Sonenshein, 2000) and some have been reported as agents of biocontrol of plant diseases (Martinez-

Ochoa, 2000; Zehnder et al. 2000).

Bacillus species are commonly found associated with plants, and some have

been reported to enhance growth and suppress pathogens (Kunst et al. 1997;

Nicholson et al. 2000; Reddy and Rahe, 1989a; Reddy and Rahe, 1989b;

Masashiro, 1997; Zhinong, 2000; Benhamou et al. 1998; Martinez-Ochoa, 2000;

Zehnder et al. 2000). Bacillus subtilis strains MBI600 and GBO3, and Bacillus

amyloliquefaciens strain IN937 have been reported as biological control agents in 28

numerous crop plants (Martinez-Ochoa, 2000; Zehnder et al. 2000; Reddy et al.

2000; Raupach, 1998; Ryu et al. 2000; EPA, 2004a; EPA, 2004b). B. subtilis strain MBI600 has been licenced for sale as a pesticide by Microbio Ltd. (Bolder,

CO, USA) since 1994. It was labeled as a treatment for seeds of cotton, beans, barley, wheat, corn, peas, peanuts, and soybeans (EPA, 2004a). B. subtilis strain

GBO3 was licenced for sale as a pesticide to Gustafson Inc. (Plano, TX, USA) since 1992, labeled on flowers and ornamental seeds and as a seed treatment for cotton, vegetables, peanuts, and soybeans (EPA, 2004b).

One of the reasons Bacillus species have been sought as biocontrol agents is

their resistance to environmental stresses and subsequent ability to be

formulated and stored prior to use (Schisler et al. 2004). Endospores produced

by Bacillus spp. can tolerate a variety of chemical and physical factors such as

heat, pressure, gravity force, vacuum, ultraviolet light, gamma radiation and

extreme desiccation (Driks and Setlow, 2000; Nicholson et al. 2000; Sonenshein,

2000). The ability of the bacterium to withstand adverse conditions is related in

part to the presence of DNA binding proteins such as small acid-soluble spore

proteins (SASP). In vitro studies have shown that DNA binding proteins and DNA

form a complex that changes the properties of the DNA, which becomes

extremely resistant to enzymatic, chemical and physical factors (Driks and

Setlow, 2000).

Endospores are formed by asymmetrical division that yields a small cell

(forespore) and a large mother cell that at the beginning engulfs the forespore, and finally lyses, releasing the endospore (Burkholder and Grossman, 2000).

Two main factors are involved in sporulation induction: limitation of nutrients including carbon, nitrogen or phosphorus, and quorum sensing (Sonenshein,

2000). At least two specific germinants are known for Bacillus subtilis: L-alanine and the mixture of L-asparganine, fructuose, glucose and potassium ions

(Sonenshein, 2000).

Bacillus spp. are usually identified using tests that include gram reaction, motility, position of the spore within the mother cell, growth at 45 oC, growth at pH 5.7, growth in NaCl, utilization of citrate, anaerobic growth, acid formation from arabinose, mannitol, and xylose, and starch hydrolysis. However, strains give variable results when tested physiologically (Chun and Vidaver, 2001). The enterobacterial repetitive intergenic consensus PCR (ERIC-PCR) assay is a molecular technique based on rep-PCR that results in unique fingerprints that can be used to identify rhizobial genera, species, or strains that could not be

distinguished by other methods (Bruijn, 1992; Bruijn et al. 1996; Louws et al.

1999). In this technique, ERIC sequences are used as primer binding sites for

PCR, enabling amplification of DNA fragments of different sizes, that can be

separated by electrophoresis, forming a DNA fingerprint pattern specific for

individual bacterial strains (Versalovic et al. 1994). There is evidence that ERIC

primers anneal to anonymous binding sites resulting in bands that do not

represent sequences that lie between genuine ERIC elements (Gillings and

Holley, 1997). However, in general fingerprint patterns and reproducibility observed in several species have made this technique one of the most useful

tools for bacterial identification (Louws et al. 1995; Bruijn et al. 1996; Lupski et al.

1996; Louws et al. 1999; Herman and Heyndrickx, 2000; Lima et al. 2002).

The use of MBI600, GBO3 and IN937 as biocontrol agents on tomato requires understanding the rhizosphere community of these PGPR under the conditions in

which they are expected to perform. Rifampicin-resistant mutants of MBI600,

GBO3 and IN937 may facilitate monitoring of these strains on tomato roots.

Rifampicin is a semisynthetic derivative of rifamycin that is produced by

Streptomyces. It is active against gram positive and some gram negative bacteria

by binding to the β-subunit of RNA polymerase and blocking mRNA synthesis

and protein production (Todar, 2002). Spontaneous rifampicin-resistant mutants

have been observed in Bacillus subtilis (Butterworth and McCartney, 1991). The stability of these mutations and the ability of recovery of Bacillus spp. from plant tissue have made these mutants an important research tool. For instance, Collins et al. (2003) was able to monitor in time and space the biological control agent

Bacillus subtilis in the phyllosphere of sugar beet by using a rifampicin-resistant mutant. Martinez Ochoa (2000) also verified colonization of two biocontrol agents

(B. subtilis and B. amyloliquefaciens) on tomato roots using rifampicin-resistant mutants. Morphological and biological control characteristics of B. subtilis remained unchanged after rifampicin selection in a study by Collins et al. (2003), but in Burkolderia cepacia the ability to supress stem rot on poinsettia was variable after rifampicin selection (Benson, 1998). The potential risk of introduction of rifampicin-resistant strains and engineered bacteria was studied by Natsch et al. (1998) who showed that introduction of Pseudomonas fluorescens strains CHA0-rif and CHA0-rif/pME3424 did not affect the abundance of dominant genotypic groups of culturable bacteria and did not change the proportion of sensitive and resistant bateria in situ.

The main goal of the study reported in this chapter was to select and characterize rifampicin-resistant mutants of Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937. Rifampicin selection

would facilitate monitoring of these biocontrol agents from the tomato

rhizosphere. Characterization of rifampicin-resistant mutants based on colony morphology and ERIC fingerprint patterns would make possible unambiguous identification of re-isolated Bacillus strains. It was hypothesized that rifampicin selection would not modify colony morphology, endospore production or fingerprint pattern, that endospore proportion increases over time as nutrients in medium are depleted under our conditions, and that polymorphisms in ERIC fingerprint patterns are sufficient to distinguish between Bacillus subtilis and B.

amyloliquefaciens strains and other bacteria isolated from tomato rhizospheres.

MATERIALS AND METHODS

Isolates

Three previously reported plant growth-promoting rhizobacteria were used

throughout the experiments: B. subtilis strain MBI600 (Microbio Ltd., Bolder, CO,

USA), B. subtilis strain GBO3 (Gustafson Inc., Plano, TX, USA), and B.

amyloliquefaciens strain IN937 (Auburn University, AL, USA).

Rifampicin-Resistant Mutant Selection

Selection of spontaneous rifampicin-resistant strains was performed using the

gradient plate technique described by Eisenstadt et al. (1994). B. subtilis

(MBI600, GBO3) and B. amyloliquefaciens (IN937) were streaked from the low (0

ppm) to the high (100 ppm) concentration of rifampicin (Calbiochem, La Jolla,

CA, USA) and incubated for 24 - 48 hours. Streaked colonies that grew in the

high rifampicin concentration zone were selected.

Long-Term Storage

Wild-type and rifampicin-resistant strains of Bacillus spp. were plated on nutrient

agar (8 g nutrient broth Difco BD, Sparks, MD, USA; 15 g agar Difco BD, Sparks,

MD, USA; 1000 ml distilled water) and rifampicin-amended (60 mg/l) nutrient

agar medium, respectively, and incubated at 28 oC for 48 hours. A loopful of

bacterial cells was transferred into a 2 ml cryogenic vial (Nalgene, Rochester,

NY, USA) containing 500 μl glucose-free nutrient broth yeast extract medium

(Schaad et al. 2001) mixed (1:1 v:v) with 30% glycerol. Isolates were stored at –

80 oC.

Colony Morphology

Wild-type and rifampicin-resistant strains of Bacillus spp. were plated on nutrient

agar. Plates were incubated at room temperature, under white light (Sylvania

20W, warm light) for 48 hours. Morphological characteristics (form, elevation, edge and consistency) of the colonies were recorded for rifampicin-resistant mutants and wild-type cultures.

Endospore Production

Wild-type and rifampicin-resistant strains of Bacillus spp. were plated on NA

medium and incubated at 28 oC for 24 hours in dark. Four single colonies of each

strain were transferred to individual plates that contained endospore-forming medium [Modified nutrient-broth yeast extract agar (Schaad et al. 2001): 8 g nutrient broth; 2 g yeast extract; 0.5 g KH2PO4; 2 g K2HPO4; 2.5 g glucose; 0.03

g MnSO4 H2O; 18 g Agar; 1000 ml distilled water; 1 ml of 1 M MgSO4 was added

after cooling at 50 oC). The cultures were randomly arranged on a laboratory

bench and incubated at ambient temperature under white light (Sylvania 20W,

warm light) for 14 days. A sample from each culture was picked up with a

bacteriological loop every two days, suspended in 20 μl water placed on a glass

slide and fixed by passing the slide above a flame until the water evaporated.

Spores and bacteria were stained using malachite green (80%) and safranin

(0.5%) according to the procedure described by Schaeffer and Fulton (1933).

Bacteria were photographed using a digital camera (Magnafire, Optronics, Goleta

California, USA) mounted on an inverted microscope (DMIRB, Leica, USA) with the following settings: condenser s23, ph 5.0, filter DLF, magnification 1.5x, objective 63x, ocular 10x.

The proportion of free endospores was quantified from the digital images. Using

Photodraw (Microsoft), the top-left visible field of each photograph was expanded

until the resulting image included one hundred bacteria / free endospores. The

proportion of free endospores in this 100 cell sample was determined.

DNA Fingerprinting

Genomic DNA was extracted from wild-type and rifampicin-resistant colonies of

each Bacillus strains. Strains were plated on NA and incubated at 28 oC for 48 hours. A single colony was transferred to a 1.5 ml Eppendorf tube containing 250

μl sterile distilled water. Bacteria were pelleted by centrifugation (16000 x g for 10 minutes). Pellets were suspended in 186 μl lysis buffer (20 mM TRIS; 2 mM

EDTA; 1.2 % Triton x-100; 20 mg/l lysozyme) and incubated at 37 oC for 40 minutes. Genomic DNA was extracted using the Qiagen DNeasy tissue kit

(Valencia, CA. USA) according to the manufacturer’s instructions.

rep-PCR was carried out as described in Louws et al. (1996) with 0.8 mM of each of the primers [ERIC 1R (5’-ATGTAAGCTCCTGGGGATTCAC-3’); ERIC 2 (5’-

AAGTAAGTGACTGGGGTGAGCG-3’)], 1.25 mM of each of four dNTP’s, 1.6 u

Taq polymerase (Promega, Madison, WI, USA), and variable concentrations of template. After amplification, 7 μl reaction mixture plus 3 μl loading dye [5 mg bromophenol blue; 5 ml 5 x TBE buffer (0.45 M Tris-Borate; 0.01 M EDTA; pH

8.3); 2 g sucrose] were loaded into 1.5 % agarose gels in 0.5 x TBE and amplified DNA fragments were separated by horizontal gel electrophoresis

(Midicell e350, E-C apparatus corporation, St. Petersburg, Florida, USA) at 50 V for 240 minutes at 10 oC. Amplification products (stained in 2 μg ml-1 ethidium

bromide for 15 minutes) were analyzed under a UV light transilluminator. Images

were photographed using an EDAS290 (Kodak, Rochester, N.Y., USA) system ,

gel size 13 x 17 cm with 1 - 3.5 seconds of exposure time.

Statistical Analysis

The area under the endospore proportion curve was calculated using the

trapezoid method that consists of breaking up the curve into a series of

rectangles, calculating the area of each rectangle and adding the areas together.

The area under the curve was used for comparisons in a totally randomized

design with four replications (proc glm; class bacillus rep; model area = rep bacillus; means bacillus / lsd lines; run; SAS 8.0). Mathematical modeling of the proportion of endospores over time was carried out using Haldane corrected data

(add 0.5 to the numerator and 1 to the denominator). Data were adjusted to

Exponential, Monomolecular, Logistic, and Gompertz models. The model that best described the data was selected based on r2, sum of squares of error, and

distribution of residuals versus time (SAS 8.0). Rate of production of endospores

was the slope in each model. Weighted mean rates were calculated (Richards,

1959) to facilitate comparisons among rates of endospore production of Bacillus

strains (Weighted rate ω = r / (2η+2) η=0 monomolecular; η=1 gompertz; η=2 logistic). The experiment as a whole was repeated once.

RESULTS

Rifampicin-Resistant Mutant Selection and Colony Morphology of Bacillus spp.

Spontaneous mutants of B. subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 were isolated from NA amended with rifampicin.

Morphological characteristics of rifampicin-resistant mutants and wild-type colonies on NA medium were similar (Figure 1.1). Bacillus subtilis strain MBI600 produced circular, white, elevated colonies with smooth margins and mucilaginous consistency. Bacillus subtilis strain GBO3 produced circular, white, flat colonies with irregular margins and dry consistency. Bacillus amyloliquefaciens strain IN937 produced irregular, white colonies with smooth margins and mucilaginous consistency.

Figure 1.1. Colony morphology of Bacillus spp. on nutrient agar medium after 24 hours incubation at 28oC. A. B. subtilis MBI600 wild-type; B. B. subtilis MBI600 rifampicin-resistant mutant; C. B. subtilis GBO3 wild-type; D. B. subtilis GBO3 rifampicin-resistant mutant; E. B. amyloliquefaciens IN937 wild-type; F. B. amyloliquefaciens IN937 rifampicin-resistant mutant.

Endospore Production

Vegetative cells, endospores within mother cells and free endospores were

observed after 2 days of incubation on endospore-forming medium (Figure 1.2).

Figure 1.2. (*) Vegetative cells, (**) endospores within cell, and (***) free endospores of Bacillus subtilis MBI600 rif on endospore-forming medium incubated for 8 days at room temperature (945 x).

Significant differences were observed in endospore production, as determined by

the area under the endospore production curve, between MBI600 and IN937

when compared with GB03 (P <0.0001) in both experiments (Table 1.1). No significant differences were observed in endospore production between wild-type

and rifampicin-resistant MBI600 and IN937 strains (Table 1.1). However, the

rifampicin-resistant mutant of GBO3 produced significantly fewer endospores than the wild-type (P <0.0001) in both experiments (Table 1.1).

Experiment I Experiment II Strain Endospore Area under curve Endospore Area under curve proportion after proportion after 14 days 14 days MBI600wild 0.99 9.67 AB 0.96 8.90 A MBI600rif 1.00 10.11 A 0.94 8.73 A

GBO3wild 0.97 6.73 C 0.94 6.21 B GBO3rif 0.86 5.23 D 0.70 3.49 C

IN937wild 0.92 9.69 AB 0.96 9.35 A IN937rif 0.96 8.78 B 0.96 8.18 A Levene’s p-values for homogeneity of variances were 0.202 and 0.451 for experiment I and II, respectively. Normality was assumed in both experiments. Analysis of variance p-value was <0.0001 in both experiments.

Table 1.1. Analysis of variance of endospore production by Bacillus subtilis (MBI600 and GBO3) and B. amyloliquefaciens (IN937) on endospore-forming medium after 14 days incubation. Means followed by the same letter are not significantly different (LSD 0.05).

Production of endospores increased over time until day 14 (Figure 1.3).

Endospore production rates of B. subtilis strain MBI600 and B. amyloliquefaciens

increased over time reaching a maximum in about 4 days. Weighted rates of

endospore production were higher for MBI600 and IN937 than for B. subtilis

GBO3 (Table 1.2).

MBI600 wild MBI600 rif

1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2 Spore (proportion) Spore Spore (proportion)

0 0 2 4 6 8 101214 2468101214 -0.2 -0.2 Time (days) Time (days)

GBO3 wild GBO3 rif

1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2 Spore (proportion) Spore Spore (proportion) Spore

0 0 2 4 6 8 101214 2 4 6 8 10 12 14 -0.2 -0.2 Time (days) Time (days)

IN937 wild IN937 rif

1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2 Spore (proportion) Spore (proportion) Spore

0 0 2 4 6 8 10 12 14 2 4 6 8 10 12 14 -0.2 -0.2 Time (days) Time (days)

Figure 1.3. Production of endospores by wild-type and rifampicin-resistant strains of Bacillus subtilis (MBI600 and GBO3) and B. amyloliquefaciens (IN937) on endospore-forming medium incubated at room temperature under white light for 14 days. Experiment I: (+) Endospores; (∆) Rate of production of endospores; Experiment II: (x) Endospores; (◊) Rate of production of endospores.

The mathematical model that described 75% of the endospore production curves of B. subtilis strain MBI600 was the monomolecular model (Table 1.2). The

Gompertz model described 75% of the endospore production curves of B. subtilis strain GBO3 (Table 1.2). Fifty percent of the endospore production curves of B. amyloliquefaciens strain IN937 were described by the Logistic model (Table 1.2).

Weighted mean endospore production rates ranged from 0.09 - 0.18, 0.04 - 0.10, and 0.05 - 0.13 for strains MBI600, GBO3 and IN937, respectively. The rifampicin-resistant mutant and wild-type of Bacillus subtilis strain MBI600 and B. amyloliquefaciens strain IN937 did not show significant differences in the production of endospores (Table 1.1) and required 6 - 12 days incubation on endospore-forming medium to reach 90% endospores (Table 1.2). The rifampicin-resistant mutant of Bacillus subtilis strain GBO3 produced significantly fewer endospores than the wild-type (Table 1.1) and required from 12 to 21 days to produce 90% endospores (Table 1.2). An average of 11.29 days incubation on endospore-forming medium was required among Bacillus strains to reach a vegetative-cell:endospore ratio of 1:9.

Days to Weighted Experiment Strain Model Equation r2 reach 0.9 mean endospores Rate I MBI600 wild Gompertz -ln (-ln (y)) = 0.38 t - 0.89 0.61 8.26 0.09 MBI600 rif Monomolecular ln (1 / (1-y)) = 0.36 t - 0.18 0.82 6.89 0.18 GBO3 wild Gompertz -ln (-ln (y)) = 0.39 t - 2.30 0.97 11.66 0.09 GBO3 rif Logistic ln (y / (1-y)) = 0.64 t - 5.63 0.87 12.23 0.10 IN937 wild Logistic ln (y / (1-y)) = 0.33 t - 1.00 0.59 9.68 0.05 IN937 rif Gompertz -ln (-ln (y)) = 0.38 t - 1.50 0.78 9.86 0.09 II MBI600 wild Monomolecular ln (1 / (1-y)) = 0.26 t - 0.18 0.93 9.54 0.13 MBI600 rif Monomolecular ln (1 / (1-y)) = 0.20 t - 0.03 0.93 11.66 0.10 GBO3 wild Gompertz -ln (-ln (y)) = 0.31 t - 1.98 0.97 13.54 0.07 GBO3 rif Gompertz -ln (-ln (y)) = 0.19 t - 1.8 0.96 21.31 0.04 IN937 wild Monomolecular ln (1 / (1-y)) = 0.26 t - 0.018 0.84 8.92 0.13 IN937 rif Logistic ln (y / (1-y)) = 0.47 t - 3.45 0.59 12.01 0.07

Table 1.2. Mathematical models that best describe Bacillus subtilis (MBI600 and GBO3) and B. amyloliquefaciens (IN937) endospore formation over time.

DNA Fingerprinting of Bacillus spp. Strains

DNA sequences from B. subtilis strains MBI600 and GBO3 and B.

amyloliquefaciens strain IN937 amplified by rep-PCR with ERIC primers ranged

from 100 to 6000 bp. Polymorphic sequences differentiated Bacillus spp. from

one another. Unique sequences of approximately 696, 750 and 2750 bp were

amplified for B. subtilis while characteristic sequences of approximately 1529,

1650, 2000, 3000 and 5368 bp were amplified in B. amyloliquefaciens (Figure

1.4).

Bacillus subtilis strains MBI600 and GBO3 were differentiated from one another based on amplified sequences of approximately 776, 1000, 1553, and 3666 bp present only in GBO3. No differences in banding patterns were detected between wild-type and rifampicin-resistant mutants of each strain (Figure 1.4).

Figure 1.4. Enterobacterial repetitive intergenic consensus - polymerase chain reaction (ERIC-PCR) products obtained from genomic DNA from Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937. Arrows indicate polymorphisms.

DISCUSSION

Rifampicin-resistant mutants of B. subtilis and B. amyloliquefaciens were readily obtained from wild-type strains. Mutation in these strains is likely to have occurred in the RNA polymerase gene (rpoB), preventing rifampicin from complexing with the β-subunit of the RNA polymerase. Butterworth and

McCartney (1991) reported spontaneous rifampicin-resistant mutants in B. subtilis var. niger that tolerated 100 mg/l rifampicin. Rifampicin-resistance is a useful genetic marker because it provides a readily selectable phenotype. These mutants are stable over time and can be easily recovered from plant tissue using rifampicin-amended medium. For instance, Collins et al. (2003) used rifampicin- resistant mutants to monitor colonization of B. subtilis in the phyllosphere of sugar beet. Martinez-Ochoa (2000) selected rifampicin-resistant mutants to monitor colonization of B. subtilis and B. amyloliquefaciens in the rhizosphere of tomato.

In this study, colonies of B. subtilis and B. amyloliquefaciens were characterized by shape, color, elevation, type of margin and consistency. These morphological characteristics remained unchanged after rifampicin selection. Similar

observations have been reported by others (Collins et al. 2003). Colony

morphology of B. subtilis and B. amyloliquefaciens is the most important criterion

for selection of bacterial colonies from the isolation plate when numerous other

bacteria are present.

Endospores directly impact the ability of Bacillus cereus to survive and colonize

the rhizosphere (Young et al. 1995). This ability is an important factor in PGPR efficacy as inoculants in promoting biological control and plant growth enhancement (Young and Burns, 1993). The ability to withstand adverse conditions could contribute to the establishment, persistence, and colonization of

MBI600, GBO3 and IN937 in agricultural systems as well as facilitate the formulation of rational products.

In this study, vegetative cells, endospores within mother cells, and free

endospores were observed in endospore-forming medium after 2 days of

incubation for all strains tested. Endospore production is triggered by depletion of

nutrients and quorum sensing (Nicholson et al. 2000; Sonenshein, 2000). No

significant differences in endospore production were observed between the wild- type and rifampicin-resistant mutants of MBI600 and IN937. Both strains produced significantly more endospores than did GBO3, which showed

differences in endospore production between the wild-type and rifampicin- resistant mutant strains.

The monomolecular model efficiently described endospore production in B. subtilis MBI600. The model implied that the original population of vegetative cells was the source of endospores, without further bacterial reproduction on the medium. The Gompertz model efficiently described endospore production of B. subtilis strain GBO3. This model implies that endospores were formed in the original vegetative cell population, and in new vegetative cells that were produced on the medium, with a maximum rate of production at the beginning of the curve. The logistic model efficiently described endospore production of B. amyloliquefaciens strain IN937. This model implies that endospores resulted from the original cell population, and in new vegetative cells that were produced on the medium, with a maximum rate of production at the middle of the curve.

Based on these models, the average time to reach 90% endospores under the described conditions was estimated in 11.29 days.

Polymorphisms in ERIC-PCR fingerprint banding patterns of Bacillus subtilis

(MBI600 and GBO3) and B. amyloliquefaciens (IN937) were sufficient for differentiation of both species and strains in the same species. Thus, visual

inspection was sufficient to document whether or not an isolate belonged to a

particular species or strain. Faint bands observed in the fingerprint patterns were avoided during the identification process since such bands are not always

amplified to the same extent or not equally visible (Louws et al. 1994). In

previous studies rep-PCR using ERIC primers has demonstrated sufficient

resolution to identify rhizobial genera, species, or strains that could not be

distinguished by any other method (Bruijn, 1992; Bruijn et al. 1996; Louws et al.

1999). rep-PCR using ERIC primers is a specific, sensitive, and reliable method

(Louws et al. 1999) that can be easily incorporated to monitor PGPR bacteria in biocontrol experiments. Other workers (Collins et al. 2003; Martinez-Ochoa,

2000) have used morphological and physiological characteristics to identify

PGPR isolated from plant material. However, colony morphology is not a

definitive characteristic for identification, and the variability in physiological tests

(Chun and Vidaver, 2001) obligated us to look for a new approach.

CONCLUSIONS

• Rifampicin-resistant mutants were readily derived from wild-type Bacillus

subtilis strains MBI600 and GBO3, and B. amyloliquefaciens strain IN937.

• Selection for rifampicin resistance did not alter colony morphology, endospore

proportion, or ERIC fingerprint pattern of B. subtilis strain MBI600 or B.

amyloliquefaciens strain IN937.

• Wild-type and rifampicin-resistant mutants of B. subtilis strain GBO3 had the

same colony morphology and ERIC-fingerprint pattern, but differed in the

proportion of endospores produced in vitro.

• For all strains, the proportion of endospores in endospore-forming medium

increased over time. B. subtilis strain MBI600 and B. amyloliquefaciens strain

IN937 produced more spores that B. subtilis strain GBO3. It was estimated

that on average 12 days of incubation on endospore-forming medium were

required to reach 90% endospores.

• Polymorphisms in ERIC-PCR fingerprint patterns were sufficient to distinguish

not only between species, but also among strains of these biocontrol agents.

LIST OF REFERENCES

Benhamou, N., Kloepper, J.W., and Tuzun, S. 1998. Induction of resistance against Fusarium wilt of tomato by combination of chitosan with an endophytic bacterial strain: ultra structure and cytochemistry of the host response. Planta 204: 153-168.

Benson, M. 1998. Integrated biocontrol strategies for disease control on greenhouse ornamentals using poinsettia as a model system. American Floral Endowmen. http://www.endowment.org/pr/p4998pr1.htm.

Bruijn, F.J. 1992. Use of repetitive (Repetitive Extragenic Palindromic and Enterobacterial Repetitive Intergenic Consensus) sequences and the polymerase chain reaction to fingerprint the genomes of Rhizobium meliloti isolates and other soil bacteria. Applied and Environmental Microbiology 58 (7): 2180-2187.

Bruijn, F.J., Rademaker, J., Schneider, M., Rossbach, U. and Louws, F.J. 1996. Biology of Plant-Microbe Interactions. Proceedings of the 8th international Congress of Molecular Plant-Microbe Interactions. APS press: 497-502 pp.

Butterworth, J. and McCartney, H.A. 1991. The dispersal of bacteria from leaf surfaces by water splash. Journal of Applied Bacteriology 71: 484-496.

Burkholder, W.F. and Grossman, A.D. 2000. Regulation of the initiation of endospore formation in Bacillus subtilis. In Brun, Y.V. and Shimkets, L.J. 2000. Prokaryotic Development. American Society for Microbiology. Washington, DC. USA: 151-166 pp.

Chun, W., and Vidaver, A.K. 2001. Gram positive bacteria. In Schaad, N.W., Jones, J.B. and Chun, W. 2001. Laboratory guide for identification of plant pathogenic bacteria. Third edition. APS press. USA: 250-260 pp.

Collins, D.P., Jacobsen, B.J., and Maxwell, B. 2003. Spatial and temporal population dynamics of a phyllosphere colonizing Bacillus subtilis biological agent of sugar beet cercospora leaf spot. Biological Control 26: 224-232. 54

Driks, A. and Setlow, P. 2000. Morphogenesis and properties of the bacterial spore. In Brun, Y.V. and Shimkets, L.J. 2000. Prokaryotic Development. American Society for Microbiology. Washington, DC. USA: 191-218.

Eisenstadt, E., Carlton, B.C. and Brown, B.J. 1994. Gene mutation. In Gerhardt, P., Murray, R.G.E., Wood, W.A. and Krieg, N.R. 1994. Methods for general and molecular bacteriology. American Society for Microbiology. USA: 297-316.

EPA. 2004a. Bacillus subtilis MBI600 129082. http://www.epa.gov/pesticides/biopesticides/ingredients/factsheets/factshe et_129082.htm.

EPA. 2004b. Bacillus subtilis GBO3 (129068). http://www.epa.gov/pesticides/biopesticides/ingredients/factsheets/factshe et_129068.htm.

Gillings, M. and Holley, M. 1997. Repetitive element PCR fingerprinting (rep- PCR) using enterobaterial repetitive intergenic consensus (ERIC) primers is not necessarily directed at ERIC elements. Letters in Applied Mircobiology 25: 17-21.

Herman L., Heyndrickx, M. 2000. The presence of intragenically located REP-like elements in Bacillus sporothermodurans is sufficient for REP-PCR typing. Res. Microbiol. 151: 255-261.

J., Lazarevic, V., Lee, S.-M., Levine, A., Liu, H., Masuda, S., Maüel, C., Médigue, C., Medina, N., Mellado, R.P., Mizuno, M., Moestl, D., Nakai, S., Noback, M., Noone, D., O’Reilly, M., Ogawa, K., Ogiwara, A., Oudega, B., Park, S.-H., Parro, V., Pohl, T.M. Portetelle, D., Porwollik, S., Prescott, A.M, Presecan, E., Pujic, P., Purnelle, B., Rapoport, G., Rey, M., Reynolds, S., Rieger, M., Rivolta, C., Rocha, E., Roche, B., Rose, M., Sadaie, Y., Sato, T., Scanlan, E., Schleich, S., Schroeter, R., Scoffone, F., Sekiguchi, J., Sekowska, A., Seror, S.J., Serror, P., Shin, B.-S., Soldo, B., Sorokin, A., Tacconi, E., Takagi, T., Takahashi, H., Takemaru, K., Takeuchi, M., Tamakoshi, A., Tanaka, T., Terpstra, P., Tognoni, A., Tosato,V., Uchiyama, S., Vandenbol, M.,. Vannier, F., Vassarotti, A., Viari, A., Wambutt, R., Wedler, E., Wedler, H., Weitzenegger, T., Winters, T., Wipat, A., Yamamoto, H., Yamane, K., Yasumoto, K., Yata, K., Yoshida, K., Yoshikawa, H.-F., Zumstein, E., Yoshikawa, H and Danchin, A. 1997. The complete genome sequence of the gram-positive bacterium Bacillus subtilis. Nature 390: 249-256.

Levin, P.A., and Losick, R. 2000. Asymmetric division and cell fate during sporulation in Bacillus subtilis. In Brun, Y.V. and Shimkets, L.J. 2000. Prokaryotic Development. American Society for Microbiology. Washington, DC. USA: 167-189 pp.

Lima, A.S.G., Guidelli, A.M., Abreu, I.L., and Lemos, M.V.F. 2002. Identification of new isolates of Bacillus thuringiensis using rep-PCR products and δ−endotoxin electron microscopy. Genetics and Molecular Biology 25 (2): 225-229.

Louws, F.J., Fulbright, D.W., Stephens, C.T., and de Bruijn, F. 1994. Specific genomic fingerprints of phytopathogenic Xanthomonas and Pseudomonas pathovars and strains generated with repetitive sequences and PCR. Applied and Environmental Microbiology 60 (7): 2286-2295.

Louws, F.J., Fulbright, D.W., Stephens, C.T., and de Bruijn, F. 1995. Differentiation of genomic structure by rep-PCR fingerprinting to rapidly classify Xanthomonas campestris pv. vesicatoria. Phytopathology 85: 528- 536.

Louws, F.J., Rademaker, J.L.W., and de Bruijn, F.J. 1996. REP-PCR: A rapid technique for bacterial identification and assessment of genetic diversity. Workshop APS annual meeting 16 p. 56

Louws, F.J., Rademaker, J.L.W., and de Bruijn, F.J. 1999. The three Ds of PCR- based genomic analysis of phytobacteria: diversity, detection, and disease diagnostic. Annual Review of Phytopathology 37: 81-125.

Lupski, J.R., Versalovic, J., and Koeuth, T. 1996. Oligonucleotide primers for fingerprint bacterial strains using repetitive DNA sequence PCR (rep- PCR). http://www.bcm.edu/bcmt/techs/tech-92-20.html.

Martinez-Ochoa, N. 2000. Biological control of the root-knot nematode with rhizobacteria and organic amendments. Ph.D. dissertation. Auburn University. Alabama: 120 p.

Masahiro, S. 1997. Plant growth promoting rhizobacteria (PGPR) for interior spruce (Picea engelmannii x P. glauca) seedlings. Ph.D. dissertation. The University of British Columbia. Canada: 324 p.

Natsch, A., Keel, C., Hebecker, N., Laasik, E. and Defago, G. 1998. Impact of Pseudomonas fluorescens strain CHA0 and a derivative with improved biocontrol activity on the culturable resident bacterial community on cucumber roots. FEMS Microbiology Ecology 27 : 365 – 380.

Nicholson, W.L., Munakata, N., Horneck, G., Melosh, H.J., and Setlow, P. 2000. Resistance of Bacillus endospores to extreme terrestrial and extraterrestrial environments. Microbiology and Molecular Biology Reviews 64 (3): 548-572.

Raupach, G.S. 1998. Induced systemic resistance as the principal component of integrated pest management of Cucumis sativus L. by PGPR. Ph.D. dissertation. Auburn University. Alabama. USA: 120 p.

Reddy, M.S., Ryu, C. M., Zhang, S., Yan, Z. and Kloepper, J. W. 2000. Aqueous formulations of plant growth-promoting rhizobacteria for control of foliar pathogens. http://www.ag.auburn.edu/~mreddy/.

Richards, F.J. 1959. A flexible growth function for empirical use. Journal of Experimental Botany 10 (29): 290 – 300.

Ryu, C.M. Reddy, M.S., Zhang S., Murphy, J.F. and Kloepper, J.W. 2000. Plant growth promotion of tomato by biological preparation (LS213) and evaluation for protection against cucumber mosaic virus. http://www.ag.auburn.edu/~mreddy/.

Schaad, N.W., Jones, J.B. and Chun, W. 2001. Laboratory guide for identification of plant pathogenic bacteria. Third edition. APS press. USA: 4 pp.

Schaeffer, A.B., and Fulton, M.D. 1933. A simplified method of staining endospores. Science 77 (1990): 194.

Schisler, D.A., Slininger, P.J., Behle, R.W. and Jackson, M.A. 2004. Formulation of Bacillus spp. for biological control of plant diseases. Phytopathology 94: 1264 – 1271.

Sonenshein, A.L. 2000. Endospore-Forming Bacteria: An Overview. In Brun, Y.V. and Shimkets, L.J. 2000. Prokaryotic Development. American Society for Microbiology. Washington, DC. USA: 133-150 pp.

Todar, K. 2002. Antimicrobial agents used in treatment of infectious disease. University of Wisconsin-Madison. Department of Bacteriology. http://textbookofbacteriology.net/antimicrobial.html.

Versalovic, J., Schineider, M., de Bruijn, F.J., and Lupski, J.R. 1994. Genomic fingerprinting of bacteria using repetitive sequence-based polymerase chain reaction. Methods in Molecular and Cellular Biology 5: 25-40.

Young, C.S. and Burns, R.G. 1993. Detection, survival, and activity of bacteria added to soil. In Bollag J.M. and Stotzky, G. 1993. Soil Biochemistry. Vol. 8. Dekker. New York, USA: 1-64 pp.

Young, C.S., Lethbridge G., Shaw, L.J., and Burns, R.G. 1995. Survival of inoculated Bacillus cereus spores and vegetative cells in non-planted rhizosphere soil. Soil Biology and Biochemistry 27 (8): 1017-1026.

Zhinong, Y. 2000. Induced systemic resistance and plant growth promotion in tomato elicited by rhizobacteria. Ph.D. dissertation. Auburn University. Alabama: 131 p.

CHAPTER 2

Growth Promotion, Induction of Resistance to Bacterial Leaf Spot and Root Colonization of Tomato by Selected Plant Growth-Promoting Rhizobacteria

INTRODUCTION

Xanthomonas euvesicatoria, X. vesicatoria, X. perforans, and X. gardneri, the causal agents of bacterial leaf spot (BLS) of processing and fresh market tomatoes, occur worldwide (Bouzar et al. 2004; Cambell et al. 1998; Jones et al.

2004a; Jones et al. 2004b, Jones et al. 2004c). The disease can account for up to 70% of marketable yield losses in some regions (Basim et al. 2004; Jones et al. 1997). The use of bactericides such as copper or streptomycin alone does not provide adequate control because of low efficacy and the ability of the bacterial populations to acquire resistance (Cambell et al. 1997; Jones et al. 1991; Marco and Stall, 1983). Thus, alternative control methods are needed. Some potential alternatives, either alone or in combination that are desirable include biological control, activation of natural plant defense mechanisms, cultural practices that promote stronger and healthier plants and the use of bacterial spot-resistant varieties.

To be highly effective, genetic resistance against all of the species causing

bacterial leaf spot is necessary. Resistance against bacterial leaf spot in tomato

is controlled by several genes in different loci (Scott et al. 2003). Transference of

resistance genes from pepper to tomato resulted in resistance only against X.

vesicatoria (Tai et al. 1999). Scott et al. (2003) found that selection for resistance

against X. vesicatoria derived from crosses to tomato PI114490 would result in

resistance to both X. vesicatoria and X. euvesicatoria, but selection for X.

perforans or X. vesicatoria was less likely to result in resistance to the other

species.

In the absence of effective resistant varieties, control of bacterial leaf spot is

achieved by crop rotation, use of disease free transplants, destruction of infected

leaves, eradication of weeds or volunteer plants, and the use of bactericides, bacteriophages, and plant growth-promoting rhizobacteria (Highland, 2004;

Briceno and Miller, 2004; Jackson, 2004; Jones et al. 2004b; Lewis Ivey et al.

2005; Obradovic et al. 2005; Snowden, 2004; Vavrina et al. 2004). Adjusting the fertilizer program, particularly the nitrogen component, also affects the intensity of bacterial diseases (Harkness and Marlatt, 1970). Nitrogen has been studied in relation to disease intensity in several crop-disease combinations (Huber and

Watson, 1974). High rates of nitrogen were shown to reduce susceptibility in

Brassaia actinophylla and Schefflera arboricola to X. campestris pv. hederae

(Chase and Poole, 1987), in Philodendron oxycardium to Xanthomonas sp.

(Harkness and Marlatt, 1970), and in Schefflera arboricola to Pseudomonas

cichorii (Chase and Jones, 1986). On the other hand, high rates of nitrogen

increased the susceptibility of peach to Xanthomonas pruni (Bachelder et al.

1956), tomato to X. vesicatoria (Nayudu and Walker, 1961), sesame to P. sesami

and X. sesami (Thomas, 1965), philodendron to Erwinia chrysanthemi (Haygood

et al. 1982), and chrysanthemum to P. cichorii (Jones et al. 1985). Several species of Bacillus have been reported to fix nitrogen from the atmosphere

(Chanway and Holl 1991; Jacobs et al. 1985; Li et al. 1992; Shawky, 1983). For

instance, Grau and Wilson (1962) found that Bacillus polymyxa fixed up to 100

μg nitrogen / ml after only 200 hours of culture on nitrogen-free medium. The role

of Bacillus sp. and tuberculate ectomycorrhizae was studied on Douglas fir by Li

et al. (1992), who found a close relationship between this bacterium and the

tuberculate mycorrhizae and an indication of their contribution to the nitrogen

dynamics of Douglas fir.

The use of plant growth-promoting rhizobacteria (PGPR) is a potentially attractive

approach to managing diseases since PGPR have been shown to increase yield

and protect crops simultaneously (Ramamoorthy et al. 2002; Raupach, 1998). It

is implicit that PGPR establish, colonize, and reach population densities sufficient

to exert beneficial effects (Bloemberg and Lugtenberg 2001; Benizri et al. 2001).

The population density that triggers disease suppression or induced systemic

resistance has been established for several PGPR species including

Pseudomonas putida WCS358 (5 log CFU g-1 root; van Loon et al. 1998), P.

fluorescens WCS374 and WCS417 (5 log CFU g-1 root; van Loon et al. 1998),

and B. megaterium (6 log CFU g-1; Liu and Sinclair, 1992). Once systemic

resistance is induced, further protection is independent of the remaining

population of the PGPR (van Loon et al. 1998). However, inconsistency of

disease control is one common characteristic of biological control agents

including PGPR (Weller, 1988). Factors that contribute to variable induction of

systemic resistance by PGPR may include high concentration of pathogen

inoculum, favorable conditions for disease development, decreased competitive

activity of PGPR (Obradovic, et al. 2005), and the fact that induced systemic

resistance depends on genotype of the plant (van Loon et al. 1998).

The protective mechanisms in plants may be also elicited upon contact with pathogens in a phenomenon that is termed systemic aquired resistance (SAR).

For instance, Colletotrichum lagenarium induced systemic aquired resistance

against fungal and bacterial diseases in cucumber (Sticher et al. 1997). In most

cases inoculation with the organism serving as the inductor of SAR leads to localized necrosis. The time needed to induce systemic aquired resistance and the level of protection depends on the plant and the type of organism used for the primary inoculation (Sticher et al. 1997). Acibenzolar-S-methyl (Actigard 50 WG

Syngenta Crop Protection, Greensboro, North Carolina, USA) is a synthetic inducer of systemic aquired resistance in plants (Oostendrop et al. 2001) that has been tested against bacterial leaf spot with promising results (Briceno and Miller,

2004; Jones et al. 2004b; Louws et al. 2001; Obradovic et al. 2005).

In this study, plant growth-promoting rhizobacteria were characterized in terms of

their effects on tomato seedling growth promotion, induced resistance to bacterial

leaf spot, and efficiency in colonizing tomato seedling roots. For this study, four

fresh market tomato varieties, three PGPR strains (Bacillus spp.) and two concentrations of nitrogen were combined in a factorial experiment. The following specific hypothesis were tested: 1) Bacillus spp. colonization of tomato roots depends on genotype, 2) inoculation of tomato transplants with PGPR improves plant growth and disease suppression and 3) PGPR may increase nitrogen absortion, improving plant growth and decreasing susceptibility of tomato to bacterial leaf spot.

MATERIALS AND METHODS

Inoculation with Plant Growth-Promoting Rhizobacteria

Inoculum of rifampicin-resistant strains of MBI600, GBO3 and IN937 was prepared by suspending 15 day-old cultures grown on endospore-forming medium (Schaad et al. 2001, see Chapter 1) in sterile distilled water. The optical density at 600 nanometers was adjusted to 0.5 - 0.8 absorbance (7.2-7.4 log

CFU ml-1, confirmed by plating 10-fold serial dilutions onto nutrient agar medium).

Percent of endospores in the inoculum suspension was 99, 93 and 27% for

MBI600, IN937 and GBO3 respectively. Tomato varieties ‘Mountain Spring’,

‘Florida 47’, ‘Mountain Fresh’, and ‘BHN543’ (Siegers Seed company, Rochester,

NY, USA) were germinated in plastic Petri dishes containing moist filter paper

(Whatman 1 Maidstone, England). Plates were incubated at 28 oC for 7 - 8 days.

Seedling roots were dipped in the inoculum suspension or autoclaved distilled water for 15 seconds. Tomato seedlings were immediately transplanted into 253 cm3 pots containing fine vermiculite (Therm-o-rock East Inc, New Eagle,

Pennsylvania, USA). Treatments were arranged in a factorial design (treatment,

nitrogen, variety, pathogen) with five replications. Three-hundred (Experiment I)

and four-hundred tomato seedlings were randomly arranged on the bench in the

greenhouse. Seedlings were maintained at 27 ± 5 oC for 49 days (Experiment I) or 56 days (Experiment II) until evaluation. Acibenzolar-S-methyl (Actigard 50WG

Syngenta Crop Protection, Greensboro, NC, USA) was used as a positive control of induction of systemic resistance at the recommended rate of 56 mg l-1 (52 g ha-1) applied to 30-day-old plants using a hand-held sprayer.

Nutrient Solution Preparation and Application

Macro- and micro-nutrient solutions were prepared separately to prevent precipitation. Two macronutrient solutions were prepared (Table 2.1), each containing a different concentration of nitrogen (25 or 150 ppm). Twenty ml of nutrient solution was applied to each tomato plant daily.

Element 2N 12N ppm ppm Macronutrients N 26.88 149.56 P 17.44 17.44 K 101.72 101.68 Ca 163.90 147.25 Mg 36.80 36.80 S 28.53 28.55 Micronutrients Fe 4.64 4.64 Co 0.01 0.01 Cu 0.05 0.05 Mn 1.10 1.10 Mo 0.05 0.05 Zn 0.04 0.04 Na 1.91 1.91

Table 2.1. Composition of nutrient solution used to fertilize tomato seedlings.

Inoculation of Tomato Seedlings with Xanthomonas euvesicatoria

X. euvesicatoria (formerly X. campestris pv. vesicatoria) strain 110C T1P3 was plated on yeast extract dextrose carbonate (YDC, Schaad et al. 2001) medium

and incubated at 28 oC for 72 hours. Cultures were flooded with approximately

10 ml sterile distilled water and bacteria were dislodged with a glass rod. The

-1 optical density (OD600) was adjusted to 0.17 (7.7 -7.8 log CFU ml , confirmed by

dilution plating). Tomato plants (35-37 day-old) were misted (30 seconds every

12 minutes) for 24 hours prior to inoculation. Inoculum was applied until runoff to

the upper and lower surfaces of the leaves using a hand-held sprayer. In the 67

repeated experiment, plants were inoculated twice (35 and 49 day-old plants)

with X. euvesicatoria as described above, because of the low initial density of

bacterial leaf spot symptoms on tomato leaves.

Determination of PGPR Colonization of Tomato Roots

Whole roots of two randomly selected plants per treatment (49 and 56 day-old

Experiments I and II, respectively) were excised, shaken to remove all but tightly

adhered vermiculite particles and pooled into one sample. Samples were shaken

in 20 ml potassium phosphate washing buffer (KPB buffer: 10 mM K2HPO4, 10

mM KH2PO4 pH 7.4) for 10 minutes on a rotatory shaker (New Brunswick

Scientific Co, New Brunswick, NJ, USA) at approximately 120 rpm. Twelve ml

washing buffer was recovered and centrifuged for 10 minutes at 16000 x g to sediment the bacteria. The bacterial pellet was re-suspended in 1 ml KPB buffer, and 10-fold serial dilutions were made. One hundred μl of dilutions 10-3, 10-4, and

10-5 were spread onto nutrient agar amended with 60 mg l-1 rifampicin (two plates

per dilution). Plates were incubated for 7 days at room temperature (24 oC) under

8 hours white light (Sylvania 20W, warm light) and the number of colonies that were morphologically similar to MBI600, GBO3 and IN937 in each treatment was

recorded. Three colonies representative of the inoculum type from each sample

were purified on nutrient agar, and the isolate identity was confirmed using rep-

PCR with ERIC primers. Colony counts were adjusted based on the proportion of

PCR fingerprints corresponding to the inoculated Bacillus strain in the sample.

Colony forming units per gram of root (fresh weight) were calculated for each dilution, the average was calculated, a unit was added (to avoid obtaining logarithm of zero) and the data were Log10 transformed before statistical

analysis. Data were analyzed using a general lineal model in a factorial

experimental design with three factors (treatment, nitrogen, varienty: proc glm; class rep var treat fert path; model lcfug1 = rep var|treat|fert / ss3; lsmeans

var|treat|fert / stderr pdiff; means treat / lsd lines; run; SAS 8.0). Inoculation with

X. euvesicatoria was not included as a factor in the model to increase the

number of replications in the analysis of variance of Bacillus spp. population

density. Population densities reported are the average of all sampled roots.

Data Collection and Statistical Analysis

Plant height was measured at 7-day intervals starting 14-16 days after

inoculation with PGPR. Increase of biomass was evaluated at the end of the

experiment by measuring shoot and root fresh and dry weight and root length.

The area under the plant height curve, total fresh weight, total dry weight and

root length were determined. Colony forming units per gram of root were

calculated at the end of the experiment. Density of disease (number of spots per

plant) was recorded at 49 (Experiment I) and 46, 47, 53, and 56 (Experiment II) day-old plants. Disease density and the area under disease progress curve

(AUDPC) (Experiment II only) were used for analysis of variance and comparisons.

Data were analyzed using a general lineal model in a factorial experimental design with four factors (treatment, nitrogen, variety, pathogen) and five replications (proc glm; class rep var treat fert path; model area rl tfw tdw = rep var|treat|fert|path var*treat*fert / ss3; lsmeans var|treat|fert|path var*treat*fert / stderr pdiff; run; SAS 8.0). Least square means were used to compare treatments and interactions. The experiment as a whole was repeated once.

RESULTS

Due to poor germination, emergence, and low vigor of tomato seedlings, ‘Florida

47’ variety was not included in the analysis of variance in Experiment I.

Root Colonization by PGPR

In Experiment I, MBI600, GBO3, and IN937 were recovered from 70, 34 and 50%

of the root samples inoculated with the same strain (Table 2.2). The population

density of MBI600 was higher than that of IN937 and GBO3 (Table 2.3).

Population density averaged across all plants (including those not colonized)

ranged from 1.1 to 2.6 log CFU g-1 root. The average populations in colonized roots were 3.6, 3.4, and 3.5 log CFU g-1 root for MBI600, GBO3 and IN937,

respectively. Cross contamination was detected in up to 16% of the root

samples. The most common strains detected as contaminants were MBI600

(8%) and GBO3 (4 – 8%) (Table 2.2). IN937 was detected as a contaminant in

8% of the root samples in water-treated plants (Table 2.2) with an average

population density of 0.4 log CFU g-1 root (average population in colonized roots

was 5.1 log CFU g-1 root) (Table 2.3). No Bacillus spp. or other bacterial colonies

were detected in plants that were treated with Actigard (Table 2.2). No significant

differences in population densities of MBI600, GBO3 or IN937 were detected among tomato varieties ‘Mountain Spring’, ‘Mountain Fresh’ and ‘BHN543’ (Table

2.3). No other interactions among treatments affected root colonization of tomato by Bacillus spp.

In Experiment II, MBI600, GBO3, and IN937 were recovered from 20, 6, and 63% of the root samples inoculated with the same strain (Table 2.4). In this experiment the population density of IN937 was higher than that of MBI600 or

GBO3 (Table 2.5). Population density averaged across all plants (including those not colonized) ranged from 0.2 to 2.7 log CFU g-1 root. The average population in colonized roots were 3.3, 4.3, 4.4 log CFU g-1 root for MBI600, GBO3 and IN937, respectively. Cross contamination was detected in up to 12 % of the root samples (Table 2.4). The most common strains detected as contaminants were

MBI600 (3 - 6%) and IN937 (6 %) (Table 2.4). The three strains were detected in water-treated plants at a population density of 0.6 log CFU g-1 (average population in colonized roots were 2.7, 3.9, 3.4 log CFU g-1 root for MBI600,

GBO3 and IN937, respectively) (Table 2.5). No Bacillus spp. colonies were detected in plants that were treated with Actigard (Table 2.4). No significant differences in population densities of MBI600, GBO3 and IN937 were detected among tomato varieties ‘Florida 47’, ‘Mountain Spring’, ‘Mountain Fresh’ and

‘BHN543’ (Table 2.5). No other interactions among treatments affected root

colonization of tomato by Bacillus spp.

Chromelosporium sp., an inhabitant of soil that is usually found in greenhouse on sterilized vermiculite, was isolated from 92% of the samples that were processed for Bacillus isolation during Experiment II. The production of antibiotics with antagonistic activity by MBI600, GBO3 and IN937 was observed in nutrient agar amended with rifampicin as mycelial growth inhibition zones around the PGPR colonies (Figure 2.1).

A B C

Figure 2.1. Chromelosporium sp. A. Conidiophore; B. Detail of conidiophores and conidia; C. Antagonistic effect of IN937rif on a Chromelosporium spp. colony.

Percent of plants colonized by Inoculum / Treatment MBI600 GBO3 IN937 Other None Bacillus subtilis MBI600 70.83 4.17 0.00 0.00 25.00 Bacillus subtilis GBO3 0.00 34.78 0.00 30.43 34.78 Bacillus amyloliquefaciens IN937 8.33 8.33 50.00 0.00 33.33 Actigard 0.00 0.00 0.00 0.00 100.00 Water 0.00 0.00 8.33 41.67 50.00

Population density Log (CFU g-1 root + 1) Tomato variety Plant growth-promoting rhizobacteria p-value 0.2441 f-value 1.43 p-value <0.0001 f-value 14.04 BHN543 1.47 A Average Colonized roots Mountain Fresh 1.09 A MBI600 2.63 A 3.69 B Mountain Spring 0.97 A GBO3 1.14 BC 3.43 B IN937 1.73 B 3.52 B Water 0.42 C 5.17 A Actigard 0.00 C 0.00

Table 2.3. Analysis of variance of Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937 population densities on 49 day-old tomato roots. Colonies were isolated from roots on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on three colonies per sample (Experiment I). Barlett’s p-value 1.0. Normality assumed.

Percent of plants colonized by Inoculum MBI600 GBO3 IN937 Other None Bacillus subtilis MBI600 20.00 0.00 6.67 13.33 60.00 Bacillus subtilis GBO3 0.00 6.25 0.00 0.00 93.75 Bacillus amyloliquefaciens IN937 3.33 0.00 63.33 0.00 33.33 Actigard 0.00 0.00 0.00 6.25 93.75 Water 6.25 6.25 6.25 0.00 81.25

Table 2.4. Bacterial population composition on 56 day-old tomato roots inoculated with Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937. Colonies were isolated from roots on rifampicin- amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on three colonies per sample (Experiment II).

Population density Log (CFU g-1 root + 1) Tomato variety Plant growth-promoting rhizobacteria p-value 0.2731 f-value 1.31 p-value <0.0001 f-value 22.79 Colonized Average roots Average Colonized roots BHN543 1.10 A 3.83 A MBI600 0.58 B 3.37 A Florida 47 0.67 A 4.55 A GBO3 0.27 B 4.32 A Mountain Fresh 0.99 A 3.91 A IN937 2.77 A 4.41 A Mountain Spring 0.62 A 4.15 A Water 0.62 B 3.37 A Actigard 0.00 B 0.00

Table 2.5. Analysis of variance of Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937 population densities on 56 day-old tomato roots. Colonies were isolated from roots on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on three colonies per sample (Experiment II). Barlett’s p-value 1.0. Normality assumed.

Tomato Growth Promotion by PGPR

No significant increase in plant height was induced by Bacillus subtilis strain

MBI600, B. subtilis strain GBO3 or B. amyloliquefaciens strain IN937 when inoculated onto tomato varieties ‘Florida 47’, ‘BHN543’, ‘Mountain Fresh’ and

‘Mountain Spring’ compared with water-treated plants (Tables 2.6 and 2.7). No significant decrease in plant height was observed in Actigard-treated seedlings

(Tables 2.6 and 2.7). However, tomato seedlings inoculated with B. amyloliquefaciens strain IN937 and fertilized with the low rate of nitrogen (25 ppm) were significantly taller than water-treated control plants at the same rate of nitrogen in Experiment II but not Experiment I. The opposite occured at the high rate (150 ppm) of nitrogen, plants treated with IN937 were shorter than non- inoculated control plants (Table 2.7). Other factors that affected plant height included tomato variety and nitrogen concentration. ‘BHN543’ seedlings were tallest, followed by seedlings of ‘Mountain Spring’ and ‘Mountain Fresh’ (Tables

2.6 and 2.7). Nitrogen concentration of the nutrient solution had a significant effect on plant height in the second but not in the first experiment. One-hundred fifty ppm nitrogen significantly increased plant height of tomato seedlings compared to 25 ppm nitrogen (Table 2.7). No other interactions among treatments affected plant height of tomato.

No significant increase in the root length was induced by Bacillus subtilis strain

MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937 inoculated

onto tomato varieties ‘Florida 47’, ‘BHN543’, ‘Mountain Fresh’ and ‘Mountain

Spring’ compared with water-treated plants (Tables 2.6 and 2.7). Foliar application of Actigard on tomato seedlings did not affect root length in either experiment (Tables 2.6 and 2.7). The roots of ‘BHN543’ were longer than those of ‘Mountain Fresh’ but not of ‘Mountain Spring’ in the first but not in the second experiment (Table 2.6). Nitrogen rate significantly influenced root length. Tomato plants fertilized with 150 ppm nitrogen had shorter roots than those fertilized with

25 ppm nitrogen (Tables 2.6 and 2.7). There were no other interactions among treatments affecting root length.

No significant increase in fresh and dry weights was induced by Bacillus subtilis

strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937

inoculated onto tomato varieties ‘Florida 47’, ‘BHN543’, ‘Mountain Fresh’ and

‘Mountain Spring’ compared with the control (Tables 2.6 and 2.7). Fresh weight

of tomato was reduced in Actigard-treated plants when compared with water-

treated plants in Experiment I (Tables 2.6). Higher fresh and dry weights were

observed for ‘BHN543’ seedlings than for seedlings of ‘Mountain Fresh’,

‘Mountain Spring’ or ‘Florida 47’ (Tables 2.6 and 2.7). Fresh and dry weight of

seedlings were affected by nitrogen rate. The higher rate of nitrogen (150 ppm) induced an increase in both of these variables when compared with the plants fertilized with 25 ppm nitrogen (Tables 2.6 and 2.7). No other interactions among treatments affected fresh and dry weight of tomato.

Resistance to Bacterial Leaf Spot Induced by PGPR

Tomato plants developed typical lesions on leaves within 12 days after inoculation. Plants that were inoculated with X. euvesicatoria developed 43-46 lesions per plant in these experiments, while 7-9 lesions/plant were observed on non-inoculated plants (Tables 2.6 and 2.7). No significant increase in disease suppression against bacterial leaf spot density was induced by Bacillus subtilis strain MBI600, B. subtilis strain GBO3 or B. amyloliquefaciens strain IN937 inoculated onto tomato varieties ‘Florida 47’, ‘BHN543’, ‘Mountain Fresh’ and

‘Mountain Spring’ compared with the control (Tables 2.6 and 2.7). However, the bacterial leaf spot density was significanlty reduced in seedlings that were treated with Actigard in both experiments (Tables 2.6 and 2.7). The bacterial leaf spot density on ‘BHN543’ was significantly higher than on ‘Mountain Fresh’ or

‘Mountain Spring’ in the first but not in the second experiment (Tables 2.6 and

2.7). Nitrogen rate significantly affected bacterial leaf spot density: 36-37 lesions/plant developed on seedlings fertilized with 150 ppm nitrogen, while 15-

16 lesions/plant were observed for plants fertilized with 25 ppm nitrogen (Tables

2.6 and 2.7). There were interactions among treatments that affected bacterial leaf spot density. First, a high level of disease density was observed at the high concentration of nitrogen in both inoculated and non inoculated plants (Tables

2.6 and 2.7). Secondly, no significant reduction of bacterial leaf spot density was observed in Bacillus-treated plants compared with the control at any level of inoculum. However, the bacterial leaf spot density in seedlings inoculated with X. euvesicatoria and treated with Actigard was significantly lower than in the

Xanthomonas-inoculated control (Tables 2.6 and 2.7). Third, no significant reduction of bacterial leaf spot density was observed in Bacillus-treated plants compared with the control at any concentration of nitrogen. However, the bacterial leaf spot density in seedlings treated with Actigard was significantly lower compared with the control at the high rate of nitrogen (Tables 2.6 and 2.7).

Finally, the bacterial leaf spot density on ‘BHN543’, ‘Mountain Fresh’ and

‘Mountain Spring’ fertilized with 150 ppm nitrogen was significantly higher than the bacterial spot density on the same varieties fertilized with 25 ppm nitrogen

(Table 2.6).

Plant height Root length Fresh weight Dry weight Bacterial leaf spot (cm day) (cm) (g) (g) (spots plant-1) Variety Variety Variety Variety Variety p-value <0.0001 p-value 0.0192 p-value <0.0001 p-value <0.0001 p-value 0.0421 F 26.71 F 4.02 F 18.74 F 27.98 f 3.21 BHN543 312.82 A BHN543 14.66 A BHN543 10.27 A BHN543 1.22 A BHN543 31.36 A Mspring 292.37 B Mspring 13.72 AB Mfresh 7.97 B Mfresh 0.89 B Mspring 27.14 AB Mfresh 249.58 C Mfresh 12.71 B Mspring 7.79 B Mspring 0.80 B Mfresh 22.06 B PGPR PGPR PGPR PGPR PGPR p-value 0.0592 p-value 0.3480 p-value 0.0114 p-value 0.2354 p-value 0.0001 F 2.30 F 1.12 F 3.32 F 1.40 F 6.15 Water 305.41 A Water 14.20 A Water 9.42 A Water 1.06 A Water 31.77 A IN937 276.16 A IN937 14.52 A IN937 9.11 A IN937 0.98 A MBI600 31.16 A MBI600 282.88 A MBI600 13.04 A MBI600 8.76 A MBI600 0.97 A GBO3r 30.84 A GBO3 276.16 A GBO3 13.56 A GBO3 8.61 AB GBO3 0.98 A IN937 28.30 A Actigard 275.08 A Actigard 13.15 A Actigard 7.49 B Actigard 0.88 A Actigard 12.19 B

80 Nitrogen Nitrogen Nitrogen Nitrogen Nitrogen p-value 0.1527 p-value <0.0001 p-value <0.0001 p-value <0.0001 p-value <0.0001 F 2.06 F 55.76 F 158.23 F 115.44 F 51.96 12N 290.11 A 12N 11.61 B 12N 10.96 A 12N 1.23 A 12N 37.58 A 2N 279.74 B 2N 15.78 A 2N 6.40 B 2N 0.71 B 2N 16.12 B

Continued

Table 2.6 Growth promotion and induced resistance against Xanthomonas euvesicatoria in three varieties of tomato inoculated with rifampicin- resistant mutants of Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937 under greenhouse conditions (Experiment I). Plant height was evaluated over time. The area under the plant height curve was used for statistical analysis. Root length, fresh weight, dry weight and bacterial leaf spot were evaluated at the end of the experiment. Tomato varieties ‘Mountain Fresh’ and ‘Mountain Spring’ are abbreviated as ‘Mfresh’ and ‘Mspring’, respectively. Inoculation and non inoculation with X. euvesicatoria are abbreviated as ‘xc’ and ‘xo’, respectively. Levene’s p-values were 0.493, 0.359, 0.063, 0.032, and 0.000 for plant height, root length, fresh weight, dry weight and bacterial leaf spot density, respectively. Normality was assumed.

Table 2.6 continued

Plant height Root length Fresh weight Dry weight Bacterial leaf spot (cm day) (cm) (g) (g) (spots plant-1) Pathogen Pathogen Pathogen Pathogen Pathogen p-value 0.6663 p-value 0.8316 p-value 0.1874 p-value 0.2641 p-value<0.0001 F 0.19 F 0.05 F 1.75 F 1.25 F 169.42 xc 283.37 A xc 13.75 A xc 8.44 A xc 0.94 A xc 46.24 A xo 286.48 A xo 13.69 A xo 8.92 A xo 1.00 A xo 7.46 B Variety*Nitrogen Variety*Nitrogen Variety*Nitrogen Variety*Nitrogen Variety*Nitrogen p-value 0.2252 p-value 0.4890 p-value 0.0231 p-value <0.0001 p-value 0.0303 F 1.50 F 0.72 F 3.83 F 9.67 F 3.55 BHN543 12N 326.62 A BHN543 12N 12.23 A BHN543 12N 13.18 A BHN543 12N 1.63 A BHN543 12N 47.69 A Mfresh 12N 248.31 A Mfresh 12N 11.08 A Mfresh 12N 10.22 B Mfresh 12N 1.11 B Mspring 12N 35.94 B Mspring 12N 295.39 A Mspring 12N 11.52 A Mspring 12N 9.47 B Mspring 12N 0.95 BC Mfresh 12N 29.12 B BHN543 2N 299.03 A BHN543 2N 17.08 A BHN543 2N 7.36 C BHN543 2N 0.82 C Mspring 2N 18.34 C

81 Mspring 2N 289.35 A Mspring 2N 15.91 A Mspring 2N 6.12 CD Mfresh 2N 0.68 CD BHN543 2N 15.03 C Mfresh 2N 250.85 A Mfresh 2N 14.34 A Mfresh 2N 5.72 D Mspring 2N 0.65 D Mfresh 2N 15.00 C PGPR*Nitrogen PGPR*Nitrogen PGPR*Nitrogen PGPR*Nitrogen PGPR*Nitrogen p-value 0.3117 p-value 0.9167 p-value 0.2259 p-value 0.7218 p-value 0.0345 F 1.20 F 0.24 F 1.43 F 0.52 F 2.64 Actigard 12N 270.0 A Actigard 12N 11.39 A Actigard 12N 8.95 A Actigard 12N 1.08 A Actigard 12N 13.65 B Actigard 2N 280.0 A Actigard 2N 14.91 A Actigard 2N 6.02 A Actigard 2N 0.67 A Actigard 2N 10.73 B GBO3 12N 274.7 A GBO3 12N 11.11 A GBO3 12N 10.81 A GBO3 12N 1.23 A GBO3 12N 43.97 A GBO3 2N 277.5 A GBO3 2N 16.01 A GBO3 2N 6.41 A GBO3 2N 0.73 A GBO3 2N 17.70 B IN937 12N 295.1 A IN937 12N 12.61 A IN937 12N 11.72 A IN937 12N 1.23 A IN937 12N 39.83 A IN937 2N 275.0 A IN937 2N 16.42 A IN937 2N 6.50 A IN937 2N 0.73 A IN937 2N 16.76 B MBI600 12N 300.0 A MBI600 12N 11.11 A MBI600 12N 11.33 A MBI600 12N 1.23 A MBI600 12N 46.83 A MBI600 2N 265.7 A MBI600 2N 14.97 A MBI600 2N 6.18 A MBI600 2N 0.70 A MBI600 2N 15.50 B Water 12N 310.5 A Water 12N 11.83 A Water 12N 11.97 A Water 12N 1.37 A Water 12N 43.63 A Water 2N 300.2 A Water 2N 16.58 A Water 2N 6.88 A Water 2N 0.74 A Water 2N 19.92 B

Continued

Table 2.6 continued

Plant height Root length Fresh weight Dry weight Bacterial leaf spot (cm day) (cm) (g) (g) (spots plant-1) PGPR*Pathogen PGPR*Pathogen PGPR*Pathogen PGPR*Pathogen PGPR*Pathogen p-value 0.7374 p-value 0.5474 p-value 0.3440 p-value 0.9317 p-value 0.0015 F 0.50 F 0.77 F 1.13 F 0.21 F 4.56 Actigard xc 279.2 A Actigard xc 13.06 A Actigard xc 7.87 A Actigard xc 0.87 A Actigard xc 19.39 B Actigard xo 270.8 A Actigard xo 13.24 A Actigard xo 7.10 A Actigard xo 0.88 A Actigard xo 4.99 B GBO3 xc 273.2 A GBO3 xc 14.24 A GBO3 xc 8.31 A GBO3 xc 0.98 A GBO3 xc 53.96 A GBO3 xo 279.0 A GBO3 xo 12.88 A GBO3 xo 8.91 A GBO3 xo 0.99 A GBO3 xo 7.71 B IN937 xc 274.5 A IN937 xc 14.62 A IN937 xc 8.91 A IN937 xc 0.92 A IN937 xc 47.10 A IN937 xo 295.6 A IN937 xo 14.41 A IN937 xo 9.31 A IN937 xo 1.03 A IN937 xo 9.50 B MBI600 xc 283.9 A MBI600 xc 13.41 A MBI600 xc 7.93 A MBI600 xc 0.91 A MBI600 xc 55.56 A MBI600 xo 281.8 A MBI600 xo 12.66 A MBI600 xo 9.58 A MBI600 xo 1.02 A MBI600 xo 6.76 B

82 Water xc 305.7 A Water xc 13.43 A Water xc 9.17 A Water xc 1.04 A Water xc 55.21 A Water xo 305.0 A Water xo 14.98 A Water xo 9.68 A Water xo 1.07 A Water xo 8.34 B Nitrogen*pathogen Nitrogen*pathogen Nitrogen*pathogen Nitrogen*pathogen Nitrogen*pathogen p-value 0.8644 p-value 0.4219 p-value 0.7136 p-value 0.7801 p-value <0.0001 F 0.03 F 0.65 F 0.14 F 0.08 F 25.95 12N xc 287.9 A 12N xc 11.90 A 12N xc 10.78 A 12N xc 1.19 A 12N xc 64.56 A 12N xo 292.2 A 12N xo 11.33 A 12N xo 11.13 A 12N xo 1.26 A 12N xo 10.60 C 2N xc 278.8 A 2N xc 15.61 A 2N xc 6.09 A 2N xc 0.69 A 2N xc 27.92 B 2N xo 280.6 A 2N xo 15.94 A 2N xo 6.70 A 2N xo 0.73 A 2N xo 4.32 C

Continued

Table 2.6 continued

Plant height Root length Fresh weight (cm day) (cm) (g) Variety*PGPR Variety*PGPR Variety*PGPR p-value 0.8792 p-value 0.4890 p-value 0.9040 F 0.47 F 0.72 F 0.43 Mean Std Dev Mean Std Dev Mean Std Dev BHN543 Actigard 304.39 A 109.51 BHN543 Actigard 13.51 A 4.04 BHN543 Actigard 8.96 A 4.56 BHN543 GBO3 311.41 A 41.32 BHN543 GBO3 13.86 A 4.59 BHN543 GBO3 10.66 A 4.74 BHN543 IN937 320.67 A 68.44 BHN543 IN937 17.14 A 6.53 BHN543 IN937 10.89 A 4.57 BHN543 MBI600 301.09 A 58.99 BHN543 MBI600 13.78 A 4.50 BHN543 MBI600 9.64 A 3.91 BHN543 Water 326.55 A 39.28 BHN543 Water 15.01 A 7.00 BHN543 Water 11.18 A 4.54

Mfresh Actigard 249.02 A 59.76 Mfresh Actigard 12.20 A 2.80 Mfresh Actigard 6.55 A 2.99 Mfresh GBO3 230.05 A 43.00 Mfresh GBO3 12.14 A 5.03 Mfresh GBO3 7.93 A 3.36

83 Mfresh IN937 247.14 A 37.33 Mfresh IN937 13.08 A 3.83 Mfresh IN937 8.57 A 4.24 Mfresh MBI600 248.16 A 46.25 Mfresh MBI600 12.16 A 3.56 Mfresh MBI600 8.23 A 3.42 Mfresh Water 273.55 A 28.65 Mfresh Water 13.97 A 5.57 Mfresh Water 8.57 A 4.29

Mspring Actigard 271.81 A 71.52 Mspring Actigard 13.74 A 5.36 Mspring Actigard 6.95 A 2.91 Mspring GBO3 287.03 A 89.27 Mspring GBO3 14.69 A 6.59 Mspring GBO3 7.24 A 2.67 Mspring IN937 287.50 A 49.48 Mspring IN937 13.34 A 4.42 Mspring IN937 7.87 A 3.36 Mspring MBI600 299.40 A 64.33 Mspring MBI600 13.18 A 3.47 Mspring MBI600 8.40 A 3.54 Mspring Water 316.13 A 60.10 Mspring Water 13.63 A 4.58 Mspring Water 8.52 A 3.32 Variety*PGPR*Nitrogen*Pathogen p-value 0.0964 F 1.71

Continued

Table 2.6 continued

Dry weight Bacterial leaf spot (g) (spots plant day) Variety*PGPR Variety*PGPR p-value 0.8855 p-value 0.4280 F 0.46 F 1.01 Mean Std Dev Mean Std Dev BHN543 Actigard 1.19A 0.47 BHN543 Actigard 13.28A 21.83 BHN543 GBO3 1.25A 0.69 BHN543 GBO3 37.19A 40.49 BHN543 IN937 1.17A 0.61 BHN543 IN937 29.05A 39.60 BHN543 MBI600 1.15A 0.54 BHN543 MBI600 38.40A 50.82 BHN543 Water 1.35A 0.64 BHN543 Water 38.88A 37.28 Mfresh Actigard 0.73A 0.34 Mfresh Actigard 11.95A 25.63 Mfresh GBO3 0.85A 0.40 Mfresh GBO3 20.56A 27.11 Mfresh IN937 0.96A 0.50 Mfresh IN937 32.05A 39.11

84 Mfresh MBI600 0.94A 0.46 Mfresh MBI600 20.10A 32.12 Mfresh Water 0.97A 0.52 Mfresh Water 25.65A 33.49 Mspring Actigard 0.71A 0.41 Mspring Actigard 11.35A 13.65 Mspring GBO3 0.84A 0.26 Mspring GBO3 34.77A 48.87 Mspring IN937 0.79A 0.30 Mspring IN937 23.80A 26.46 Mspring MBI600 0.81A 0.41 Mspring MBI600 35.00A 42.38 Mspring Water 0.84A 0.42 Mspring Water 30.80A 33.06

Plant height Root length Fresh weight Dry weight Bacterial leaf spot (cm day) (cm) (g) (g) (spots plant day) Variety Variety Variety Variety Variety p-value <0.0001 p-value 0.1558 p-value <0.0001 p-value <0.0001 p-value 0.0926 F 24.00 F 1.75 F 8.23 F 11.12 F 2.16 BHN543 374.44 A BHN543 13.61 A BHN543 15.13 A BHN543 1.61 A BHN543 20.56 B Florid47 310.86 C Florid47 12.45 A Florida47 11.58 B Florida47 1.11 B Florida47 29.40 A Mfresh 296.10 C Mfresh 13.12 A Mfresh 12.64 B Mfresh 1.28 B Mfresh 22.38 AB Mspring 339.14 B Mspring 13.86 A Mspring 12.32 B Mspring 1.16 B Mspring 32.85 A PGPR PGPR PGPR PGPR PGPR p-value 0.5430 p-value 0.0136 p-value 0.1446 p-value 0.0095 p-value<0.0001 F 0.77 F 3.20 F 1.72 F 3.41 F 6.73 Actigard 339.8 A Actigard 11.97 C Actigard 12.38 A Actigard 1.12 B Actigard 6.63 B GBO3 332.2 A GBO3 14.10 AB GBO3 13.35 A GBO3 1.42 A GBO3 31.43 A IN937 322.0 A IN937 12.65 BC IN937 11.81 A IN937 1.17 B IN937 31.81 A MBI600 331.3 A MBI600 14.15 A MBI600 13.42 A MBI600 1.43 A MBI600 35.05 A

85 Water 325.2 A Water 13.42 ABC Water 13.63 A Water 1.31 AB Water 26.55 A Nitrogen Nitrogen Nitrogen Nitrogen Nitrogen p-value <0.0001 p-value <0.0001 p-value <0.0001 p-value <0.0001 p-value <0.0001 F 62.25 F 29.52 F 210.50 F 149.87 F 27.63 12N 357.87 A 12N 11.98 B 12N 16.81 A 12N 1.71 A 12N 36.67 A 2N 302.39 B 2N 14.54 A 2N 9.02 B 2N 0.87 B 2N 15.92 B

Continued

Table 2.7 Growth promotion and induced resistance against Xanthomonas euvesicatoria on four varieties of tomato inoculated with rifampicin- resistant mutants of Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937 under greenhouse conditions (Experiment II). Plant height was evaluated over time. The area under the plant height curve was used for statistical analysis. Root length, fresh weight, dry weight and bacterial leaf spot were evaluated at the end of the experiment. Tomato varieties ‘Mountain Fresh’ and ‘Mountain Spring’ are abbreviated as ‘Mfresh’ and ‘Mspring’, respectively. Inoculation and non inoculation with X. euvesicatoria are abbreviated as ‘xc’ and ‘xo’, respectively. Levene’s p-values were 1.000, 0.504, 0.169, 0.045, and 0.009 for plant height, root length, fresh weight, dry weight and bacterial leaf spot density, respectively. Normality was assumed.

Table 2.7 continued

Plant height Root length Fresh weight Dry weight Bacterial leaf spot (cm day) (cm) (g) (g) (spots plant day) Pathogen Pathogen Pathogen Pathogen Pathogen p-value 0.2153 p-value 0.0125 p-value 0.2295 p-value 0.3499 p-value<0.0001 F 1.54 F 6.31 F 1.45 F 0.88 F 75.01 xc 325.77 A xc 12.67 B xc 12.59 A xc 1.26 A xc 43.39 A xo 334.50 A xo 13.85 A xo 13.24 A xo 1.32 A xo 9.20 B Variety*Nitrogen Variety*Nitrogen Variety*Nitrogen Variety*Nitrogen Variety*Nitrogen p-value 0.8269 p-value 0.4134 p-value 0.2962 p-value 0.0124 p-value 0.2240 F 0.30 F 0.96 F 1.24 F 3.68 F 1.47 BHN543 12N 406 A BHN543 12N 11.8 A BHN543 12N 19.8 A BHN543 12N 2.22 A BHN543 12N 24.7 A BHN543 2N 342 A BHN543 2N 15.3 A BHN543 2N 10.3 A BHN543 2N 1.01 C BHN543 2N 16.3 A Florida47 12N 334 A Florida47 12N 11.5 A Florida47 12N 15.0 A Florida47 12N 1.43 B Florida47 12N 38.4 A Florida47 2N 287 A Florida47 2N 13.3 A Florida47 2N 8.08 A Florida47 2N 0.78 C Florida47 2N 20.3 A 86 Mfresh 12N 325 A Mfresh 12N 12.2 A Mfresh 12N 16.4 A Mfresh 12N 1.69 B Mfresh 12N 36.2 A Mfresh 2N 266 A Mfresh 2N 13.9 A Mfresh 2N 8.83 A Mfresh 2N 0.87 C Mfresh 2N 8.56 A Mspring 12N 365 A Mspring 12N 12.2 A Mspring 12N 15.8 A Mspring 12N 1.49 B Mspring 12N 47.3 A Mspring 2N 313 A Mspring 2N 15.5 A Mspring 2N 8.82 A Mspring 2N 0.83 C Mspring 2N 18.4 A

Continued

Table 2.7 continued

Plant height Root length Fresh weight Dry weight Bacterial leaf spot (cm day) (cm) (g) (g) (spots plant day) PGPR*Nitrogen PGPR*Nitrogen PGPR*Nitrogen PGPR*Nitrogen PGPR*Nitrogen p-value 0.0364 p-value 0.0957 p-value 0.1128 p-value 0.3451 p-value 0.1666 F 2.60 F 1.99 F 1.89 F 1.12 F 1.63 Actigard 12N 371.93 A Actigard 12N 11.86 A Actigard 12N 16.12 A Actigard 12N 1.48 A Actigard 12N 10.86 A Actigard 2N 307.66 BC Actigard 2N 12.08 A Actigard 2N 8.64 A Actigard 2N 0.76 A Actigard 2N 2.41 A GBO3 12N 361.41 A GBO3 12N 13.05 A GBO3 12N 17.55 A GBO3 12N 1.87 A GBO3 12N 43.25 A GBO3 2N 303.13 BC GBO3 2N 15.14 A GBO3 2N 9.15 A GBO3 2N 0.97 A GBO3 2N 19.61 A IN937 12N 329.05 B IN937 12N 10.84 A IN937 12N 14.36 A IN937 12N 1.48 A IN937 12N 50.55 A IN937 2N 315.03 B IN937 2N 14.46 A IN937 2N 9.25 A IN937 2N 0.86 A IN937 2N 13.08 A MBI600 12N 360.46 A MBI600 12N 12.19 A MBI600 12N 18.11 A MBI600 12N 1.94 A MBI600 12N 45.77 A MBI600 2N 302.15 BC MBI600 2N 16.11 A MBI600 2N 8.72 A MBI600 2N 0.91 A MBI600 2N 24.33 A

87 Water 12N 366.53 A Water 12N 11.94 A Water 12N 17.90 A Water 12N 1.76 A Water 12N 32.94 A Water 2N 283.99 C Water 2N 14.89 A Water 2N 9.35 A Water 2N 0.87 A Water 2N 20.16 A PGPR*Pathogen PGPR*Pathogen PGPR*Pathogen PGPR*Pathogen PGPR*Pathogen p-value 0.1795 p-value 0.9586 p-value 0.2641 p-value 0.2367 p-value 0.0166 F 1.58 F 0.16 F 1.32 F 1.39 F 3.07 Actigard xc 340.8 A Actigard xc 11.34 A Actigard xc 12.14 A Actigard xc 1.11 A Actigard xc 11.55 B Actigard xo 338.7 A Actigard xo 12.60 A Actigard xo 12.63 A Actigard xo 1.13 A Actigard xo 1.72 B GBO3 xc 334.1 A GBO3 xc 13.27 A GBO3 xc 13.45 A GBO3 xc 1.43 A GBO3 xc 49.69 A GBO3 xo 330.4 A GBO3 xo 14.93 A GBO3 xo 13.25 A GBO3 xo 1.41 A GBO3 xo 13.16 B IN937 xc 326.9 A IN937 xc 12.17 A IN937 xc 12.32 A IN937 xc 1.25 A IN937 xc 47.55 A IN937 xo 317.0 A IN937 xo 13.13 A IN937 xo 11.29 A IN937 xo 1.09 A IN937 xo 16.08 B MBI600 xc 319.5 A MBI600 xc 13.85 A MBI600 xc 12.19 A MBI600 xc 1.27 A MBI600 xc 60.54 A MBI600 xo 343.0 A MBI600 xo 14.44 A MBI600 xo 14.64 A MBI600 xo 1.59 A MBI600 xo 9.56 B Water xc 307.4 A Water xc 12.69 A Water xc 12.86 A Water xc 1.23 A Water xc 47.62 A Water xo 343.1 A Water xo 14.15 A Water xo 14.39 A Water xo 1.39 A Water xo 5.47 B

Continued

Table 2.7 continued

Plant height Root length Fresh weight Dry weight Bacterial leaf spot (cm day) (cm) (g) (g) (spots plant day) Nitrogen*Pathogen Nitrogen*Pathogen Nitrogen*Pathogen Nitrogen*Pathogen Nitrogen*Pathogen p-value 0.8622 p-value 0.0633 p-value 0.5338 p-value 0.9108 p-value 0.0276 F 0.03 F 3.47 F 0.39 F 0.01 F 4.90 12N xc 352.9 A 12N xc 11.82 A 12N xc 16.65 A 12N xc 1.68 A 12N xc 58.14 A 12N xo 362.8 A 12N xo 12.13 A 12N xo 16.96 A 12N xo 1.73 A 12N xo 15.21 C 2N xc 298.6 A 2N xc 13.51 A 2N xc 8.53 A 2N xc 0.84 A 2N xc 28.64 B 2N xo 306.1 A 2N xo 15.57 A 2N xo 9.51 A 2N xo 0.91 A 2N xo 3.19 D

Continued

Table 2.7 continued

Plant height Root length Fresh weight (cm day) (cm) (g) Variety*PGPR Variety*PGPR Variety*PGPR p-value 0.6783 p-value 0.2186 p-value 0.0822 F 0.77 F 1.30 F 1.63 Mean Std Dev Mean Std Dev Mean Std Dev BHN543 Actigard 358.36 A 95.58 BHN543 Actigard 11.78 A 4.98 BHN543 Actigard 12.59 A 6.39 BHN543 GBO3 376.84 A 62.65 BHN543 GBO3 15.04 A 5.71 BHN543 GBO3 15.21 A 7.68 BHN543 IN937 371.43 A 63.74 BHN543 IN937 12.80 A 4.62 BHN543 IN937 15.30 A 7.40 BHN543 MBI600 382.14 A 100.42 BHN543 MBI600 14.76 A 5.44 BHN543 MBI600 16.01 A 8.12 BHN543 Water 383.44 A 83.46 BHN543 Water 13.65 A 4.36 BHN543 Water 16.54 A 7.28 Florida47 Actigard 342.93 A 100.50 Florida47 Actigard 11.52 A 4.46 Florida47 Actigard 13.52 A 7.14 Florida47 GBO3 303.90 A 81.56 Florida47 GBO3 12.50 A 3.45 Florida47 GBO3 10.54 A 6.23 Florida47 IN937 288.89 A 81.34 Florida47 IN937 11.35 A 3.27 Florida47 IN937 9.36 A 4.29

89 Florida47 MBI600 316.94 A 74.20 Florida47 MBI600 15.07 A 4.98 Florida47 MBI600 13.14 A 6.58 Florida47 Water 301.64 A 111.63 Florida47 Water 11.80 A 4.57 Florida47 Water 11.33 A 6.57 Mfresh Actigard 309.27 A 92.08 Mfresh Actigard 12.97 A 3.60 Mfresh Actigard 12.86 A 6.42 Mfresh GBO3 308.58 A 91.82 Mfresh GBO3 14.08 A 5.62 Mfresh GBO3 14.56 A 7.38 Mfresh IN937 296.76 A 95.69 Mfresh IN937 13.60 A 2.79 Mfresh IN937 11.87 A 7.22 Mfresh MBI600 289.96 A 69.32 Mfresh MBI600 12.92 A 6.76 Mfresh MBI600 11.72 A 5.80 Mfresh Water 275.90 A 98.93 Mfresh Water 12.01 A 4.62 Mfresh Water 12.19 A 7.97 Mspring Actigard 348.63 A 94.57 Mspring Actigard 11.61 A 5.63 Mspring Actigard 10.56 A 5.60 Mspring GBO3 339.76 A 77.72 Mspring GBO3 14.78 A 3.79 Mspring GBO3 13.09 A 6.20 Mspring IN937 331.08 A 69.18 Mspring IN937 12.86 A 5.33 Mspring IN937 10.69 A 4.95 Mspring MBI600 336.17 A 76.46 Mspring MBI600 13.85 A 4.82 Mspring MBI600 12.81 A 6.82 Mspring Water 340.07 A 92.28 Mspring Water 16.20 A 7.12 Mspring Water 14.46 A 8.26 Variety*PGPR*Nitrogen*Pathogen p-value 0.0006 F 2.96

Continued

16.08 56.16 3.62 38.76 51.25 12.10 30.40 32.42 35.63 59.95 76.33 13.88 Std Dev Std 36.30 47.34 44.25 44.46 71.13 65.31 20.20 59.63 A A

A A A A A A A A A A A A A A A A A A 33.01 7.90 Mean 26.44 2.17 20.47 29.55 24.07 8.95 48.92 7.52 23.95 28.72 38.50 36.83 25.17 21.95 47.90 33.72 19.98 40.20 (spots plant day) Bacterial leaf spot Bacterial IN937 Water Actigard GBO3 MBI600

IN937 Water IN937 Water IN937 Water GBO3 GBO3 GBO3 Actigard MBI600 Actigard MBI600 Actigard MBI600 BHN543 BHN543 BHN543 BHN543 BHN543 Florida47 Florida47 Florida47 Florida47 Florida47 Mfresh Mfresh Mfresh Mfresh Mfresh Mspring Mspring Mspring Mspring Mspring

Variety*PGPR p-value 0.7676 F 0.68 0.72 0.82 0.66 1.12 0.96 0.79 0.91 1.15 1.19 0.76 0.55 0.75 0.71 0.81 0.95 0.61 Std Dev Std 0.50 0.52 0.70 0.87 CDEF EFG EFG FG FG FG G BCDEF ABCDE ABCD DEFG BCDEF

FG ABC CDEFG A FG G FG AB (g) 1.20 1.24 1.31 1.62 1.16 1.10 1.18 Mean 0.88 1.34 1.27 1.35 1.85 1.02 1.66 1.69 0.98 1.28 0.88 1.03 1.75 Dry weight weight Dry IN937 Water GBO3 MBI600 Actigard

Water Water IN937 IN937 IN937 Water Water GBO3 MBI600 GBO3 GBO3 Actigard MBI600 Water Water Actigard MBI600 Actigard BHN543 BHN543 BHN543 BHN543 BHN543 Florida47 Florida47 Florida47 Florida47 Florida47 Mfresh Mfresh Mfresh Mfresh Mfresh Mspring Mspring Mspring Mspring Mspring Variety*PGPR p-value 0.0408 F 1.84

90 Table 2.7 continued DISCUSSION

Our data show that Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B.

amyloliquefaciens strain IN937 colonized roots of tomato varieties ‘BHN543’,

‘Florida 47’, ‘Mountain Fresh’ and ‘Mountain Spring’ resulting in similar population

densities. However, the percentage of colonized roots and the population density

of these Bacillus strains was low when compared with those reported by

McSpadden Gardener (2004). Even though the concentration of active colony forming units in the applied inoculum should have been sufficient to trigger beneficial effects on tomato, the established population density of these mutants

on roots may not be sufficient to trigger induced systemic resistance or enhance

plant growth if the mode of action of the PGPR is similar to that reported

elsewhere (van Loon et al. 1998; Liu and Sinclair, 1992). The percent recovery

and the population densities of MBI600 and IN937 tended to be higher than

those of GBO3 in both experiments. Perhaps the higher proportion of

endospores in the inoculum suspension of MBI600 and IN937 compared with

those in GBO3 inoculum facilitated colonization and establishment of MBI600

and IN937 populations in tomato roots. Young et al. (1995) found that the final

populations of Bacillus cereus were similar to the initial levels when inoculation 91

was done using spores, whereas populations declined rapidly in the first 5 days

when vegetative cells were used as source of inoculum. Possibly the colonization

of tomato roots by Bacillus spp. was aided by the presence of spores that endure

environmental conditions and germinate when conditions are optimal for bacterial

development (Driks and Setlow, 2000; Nicholson et al. 2000; Sonenshein, 2000).

The analysis of the bacterial population composition in roots of tomato showed that contamination of control plants by MBI600, GBO3 and IN937 reached up to

8% at population densities up to 100.62 CFU g-1. Interestingly, previous reseach

on Bacillus spp. (Braun-Kiewnick et al. 1998; Collins et al. 2003; Reddy et al.

2000; Martinez-Ochoa, 2000; Reddy and Rahe, 1989a, 1989b; Wulff et al. 2002;

Zehnder et al. 2000) did not report contamination of control plants or cross contamination among Bacillus spp. treatments. Perhaps contamination was not observed or the identification of Bacillus spp. was based on morphological characteristics of the colony and physiological tests that made identification up to strain level impossible. In our experiments, several factors may contribute to contamination of control plants. First, the factorial experimental design required the randomization of the treatments on the work surface; therefore, PGPR- inoculated plants were placed near or adjacent to non-inoculated plants in the same greenhouse. Second, the fertigation was performed by adding 20 ml of

nutrient solution per day per plant by hand. Even though plants were irrigated individually, water splash could carry bacterial cells from one treatment to other.

Finally, the air movement created by the cooling system in the greenhouse may have caused the movement of soil and bacterial cells from inoculated to noninoculated plants. Individualized irrigation and reduction of air movement may serve to avoid cross contamination.

Growth promotion was quantified as an increase in plant height, root length, and/or fresh and dry weights. Our data suggested that no significant growth promotion was induced by Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937 inoculated onto tomato varieties ‘Florida

47’, ‘BHN543’, ‘Mountain Fresh’ and ‘Mountain Spring’ compared with the control. However, tomato seedlings inoculated with B. amyloliquefaciens strain

IN937 and fertilized with low rate of nitrogen (25 ppm) were significantly taller compared with the water-treated control at the same rate of nitrogen. Several mechanisms have been reported to enhance plant growth such as fixation of nitrogen into a form that can be used by the plant (Bloemberg and Lugtenberg,

2001). Bacillus spp. has the ability to fix nitrogen from the atmosphere (Chanway and Holl 1991; Jacobs et al. 1985; Li et al. 1992; Shawky, 1983; Taiz and Zeiger,

2002). Perhaps B. amyloliquefaciens strain IN937 increased plant height by fixing

nitrogen or improved stand under low nitrogen (stress conditions, van Loon et al.

1998). Tomato growth parameters were influenced by genotype, nitrogen rate

and Actigard treatment. One hundred-fifty parts per million nitrogen increased

plant height by 18%, fresh weight by 78% and dry weight by 82% compared to fertigation with 25 ppm nitrogen. The availability of nitrogen in the immediate

microenvironment also regulated root length. In these studies, the high rate of

nitrogen induced a relatively small root system. This is in agreement with the

results of Bloom et al. (1993) who showed that the high concentration of nitrogen

in the root zone may have resulted in high nitrogen uptake per unit of root, and

relatively low translocation of carbohydrate to the root. Actigard reduced by 25%

the average fresh weight of tomato seedlings. Actigard is a synthetic inducer of

systemic aquired resistance in plants (Oostendrop et al. 2001), which is

expressed against a broad sprectrum of organisms and involves production of

lignin, papillae, pathogenesis related proteins (β-1,3-glucanases, chitinases,

osmotin, etc.), furanocoumarin, phytoalexins, and glycine rich proteins (Sticher et

al. 1997). Deployment of these defense mechanisms is thought to have a

physiological cost measured in terms of vegetative and reproductive growth

(Vallad and Goodman, 2004). For instance, reduction in growth and yield has

been reported in tomato variety ‘Mountain Fresh’ as a result of Actigard treatment

(Ferguson et al. 2002).

Due to randomization of treatments in the experiment, tomato plants that were inoculated with Xanthomonas euvesicatoria were placed near or adjacent to noninoculated plants in the same greenhouse. Dispersion of X. euvesicatoria

from inoculated plants to non-inoculated plants may have occurred due to mist,

air movement, and contact between plants. It was observed that Xanthomonas-

inoculated plants had significantly more spots per plant that noninoculated plants

in both experiments. No significant reduction of bacterial leaf spot density was

observed in tomato varieties ‘Florida 47’, ‘BHN543’, ‘Mountain Fresh’ and

‘Mountain Spring’ inoculated with Bacillus subtilis strain MBI600, B. subtilis strain

GBO3 and B. amyloliquefaciens strain IN937. Unlike successful treatment with

Pseudomonas fluorescens A506 (foliar application), P. syringae TLP2 (foliar

application), P. syringae Cit 7 (foliar application), Burkholderia gladioli IN-26 (root

application), Bacillus pumilus SE-34 (root application) and B. cereus M-22 (root

application) against bacterial speck of tomato (Ji et al. 1996; Wilson et al. 1996;

Wilson et al. 2002), MBI600, GBO3 and IN937 treatment did not reduce bacterial

leaf spot in the greenhouse. Obradovic et al. (2005) reported similar observations

and concluded that the high concentration of inoculum and favorable conditions

for disease development contributed to decreased effectiveness of Bacillus

pumilus B122 and Pseudomonas fluorescens B130 against bacterial spot. In our

experiments, acibenzolar-S-methyl (Actigard) induced systemic aquired

resistance in the tomato varieties tested. Bacterial leaf spot density was reduced

by 61-81% in Actigard treated plants compared with non-treated plants. Similar

observations have been reported elsewhere (Briceno and Miller, 2004; Jones et

al. 2004b). Our data showed that 150 ppm nitrogen affected bacterial leaf spot

density in tomato seedlings of varieties ‘Florida 47’, ‘BHN543’, ‘Mountain Fresh’

and ‘Mountain Spring’. Fertilization with this rate of nitrogen resulted in

production of young tissue (shown by significantly increased biomass) that

developed 230% more bacterial leaf spot lesions than seedlings fertilized with 25

ppm nitrogen. Nayudu and Walker (1961) observed that a decrese in bacterial leaf spot was correlated with reduced growth and nitrogen concentration of tomato leaves.

CONCLUSIONS

• Bacillus sp. was an aggressive colonizer of tomato roots. Dipping geminating

roots in 7 log CFU ml-1 for 15 seconds yielded colonization levels that ranged

from 3.3 to 5.0 log CFU g-1 49 days after inoculation. However, not all

inoculated plants were colonized, resulting in low average colonization rate.

MBI600 and IN937 were more effective than GBO3 in colonizing tomato

roots.

• No plant growth increase or disease suppression was observed in tomato

plants of varieties ‘Florida 47’, ‘BHN543’, ‘Mountain Fresh’ and ‘Mountain

Spring’ inoculated with Bacillus subtilis strain MBI600, B. subtilis strain GBO3

and B. amyloliquefaciens strain IN937.

• Plants fertilized with 150 ppm nitrogen had higher biomass and increased

susceptibility to bacterial leaf spot than those fertilized with 25 ppm nitrogen.

• Systemic resistance induced by Actigard (56mg/l) reduced bacterial leaf spot

in tomato seedlings by 61-81% compared to the untreated control.

LIST OF REFERENCES

Bachelder, S., Daines, R.H. and Bartley, C.E. 1956. Effect of nutrition on the incidence of bacterial leaf spot of peach. (Abstr.) Phytopathology 46: 240.

Benizri, E., Baudoin, E., and Guckert, A. 2001. Root colonization by inoculated plant growth promoting rhizobacteria. Biocontrol Sci. Technol. 11:557-574.

Bloemberg, G. V., and Lugtenberg, B. J. J. 2001. Molecular basis of plant growth promotion and biocontrol by rhizobacteria. Current Opinion in Plant Biology 4: 343-350.

Bloom, A.J., Jackson L.E., and Smart, D.R. 1993. Root growth as a function of ammonium and nitrate in the root zone. Plant Cell and Environment 16: 199 – 206.

Chanway, C.P. Holl, F.B. 1991. Biomass increase and associative nitrogen- fixation of mycorrhizal pinus-contorta seedlings inoculated with a plant- growth promoting Bacillus strain. Canadian Journal of Botany-Revue Canadienne de Botanique 69(3): 507-511.

Chase, A.R. and Poole, R.T. 1987. Effects of fertilizer rate on severity of Xanthomonas leaf spot of schefflera and dwarf schefflera. Plant Disease 71: 527-529.

Ferguson, A., Jones, R.T., Snyder J. and Nesmith W. 2002. Effects of Actigard on tomato yields. Fruit and vegetable crops research report. University of Kentucky. Agricultural Experimental Station. Technical Report 470: 53-56.

Grau, F.H. and Wilson, P.W. 1962. Physiology of nitrogen fixation by Bacillus polymyxa. Journal of Bacteriology 83: 490-496. 99

Harkness, R.W., and Marlatt, R.B. 1970. Effect of nitrogen, phosphorus and potassium on growth and Xanthomonas disease of Philodendron oxycardium. J. Amer. Soc. Hort. Sci. 95 (1): 37-41.

Haygood, R.A., Strider, D.L., and Nelson, P.V. 1982. Influence of nitrogen and potassium on growth and bacterial leaf blight of Philodendron selloum. Plant Disease 66: 728-730.

Huber, D.M. and Watson, R.D. 1974. Nitrogen form and plant disease. Annual Review of Phytopathology 12: 139-165.

Jacobs, M.J., Funke, B.R., Merrill, S.D. 1985. Enumeration and characterization of nitrogen-fixing bacteria in a stripmine site. Plant and Soil 84(1): 3-10.

Ji, P., Kloepper, J.W., Wilson, M., and Campbell, H.L. 1996. Rhizobacterial- induced systemic resistance in tomato againts bacterial speck. (Abstract) Phytopathology 86: s50.

Jones J.B., Stall, R.E. and Zitter, T.A. 1997. Compendium of Tomato Diseases. APS press. Minnesota. USA: 73 p.

100

Jones, J.B., Woltz, S. S., Jones, J. P., and Portier, K. L. 1991. Population dynamics of Xanthomonas campestris pv. vesicatoria on tomato leaflets treated with copper bactericides. Phytopathology 81:714-719.

Li, C.Y., Massicote, H.B., and Moore, L.V.H. 1992. Nitrogen-fixing Bacillus sp. associated with Douglas-fir tuberculate ectomycorrhizae. Plant and Soil 140(1): 35-40.

Liu, Z.L. and Sinclair, J.B. 1992. Population dynamics of Bacillus megaterium strain B153-2-2 in the rhizosphere of soybean. Phytopathology 82: 1297- 1301.

Louws, F. J., Wilson, M., Campbell, H. L., Cuppels, D. A., Jones, J. B., Shoemaker, P. B., Sahin, F., and Miller, S. A. 2001. Field control of bacterial spot and bacterial speck of tomato using a plant activator. Plant Disease 85:481-488.

Marco, G. M., and Stall, R. E. 1983. Control of bacterial spot of pepper initiated by strains of Xanthomonas campestris pv. vesicatoria that differ sensitivity to copper. Plant Disease. 67:779-781.

101

Martinez-Ochoa, N. 2000. Biological control of the root-knot nematode with rhizobacteria and organic amendments. Ph.D. dissertation. Auburn University. Alabama: 120 p.

McSpadden Gardener, B. 2004. Ecology of Bacillus and Paenibacillus spp. in agricultural systems. Phytopathology 94: 1252-1258.

Nayudu, M.V. and Walker, J.C. 1961. Tomato leaf composition in relation to bacterial leaf spot. Phytopathology 51: 368-372.

Oostendrop, M., Kunz, W., Dietrich, B., and Staub, T. 2001. Induced disease resistance in plants by chemicals. Eur. J. Plant Pathol. 107: 19-28.

Raupach, G.S. 1998. Induced systemic resistance as the principal component of integrated pest management of Cucumis sativus L. by PGPR. Ph.D. dissertation. Auburn University. Alabama. USA: 120 p.

Reddy, M.S., and Rahe, J.E. 1989b. Bacillus subtilis B-2 and selected onion rhizobacteria in onion seedling rhizospheres: Effects on seedling growth

102

and indigenous rhizosphere microflora. Soil Biology and Biochemistry 21(3): 379-383.

Reddy, M.S., Ryu, C. M., Zhang, S., Yan, Z. and Kloepper, J. W. 2000. Aqueous formulations of plant growth-promoting rhizobacteria for control of foliar pathogens. http://www.ag.auburn.edu/~mreddy/.

Schaad, N.W., Jones, J.B. and Chun, W. 2001. Laboratory guide for identification of plant pathogenic bacteria. Third edition. APS press. USA: 4 pp.

Scott, J,W., Francis, D.M., Miller, S.A., Somodi, G.C. and Jones J.B. 2003 Tomato bacterial spot resistance derived from PI114490; inheritance of resistance to race T2 and relationship across three pathogen races. J. Amer. Soc. Hort. Sci. 128 (5): 698-703.

Shawky, B.T. 1983. Growth and nitrogen fixation by cutures of Bacillus sp. S 77g, Pseudomonas sp. S 44b, and Klebsiella sp. S145m, isolated from Egyptian soild. Zentralbl Mikrobiol 138(1): 9-15.

Sonenshein, A.L. 2000. Endospore-Forming Bacteria: An Overview. In Brun, Y.V. and Shimkets, L.J. 2000. Prokaryotic Development. American Society for Microbiology. Washington, DC. USA: 133-150 pp.

Sticher L., Mauch-Mani B., and Métraux J.P. 1997. Systemic aquired resistance. Annual Review of Phytopathology 35:235-270.

Tai, T.H., Dahlbeck, D., Clark, E.T., Gajiwala, P., Pasion, R., Whalen, M.C., Stall, R.E. and Staskawicz, B.J. 1999. Expression of the Bs2 pepper gene confers resistance to bacterial leaf spot disease in tomato. PNAS 96 (24): 14153 – 14158.

Taiz, L., and Zeiger E. 2002. Plant Physiology. Sinauer Associates. Third edition. Sunderland Massachusetts, USA: 68-86.

103

Thomas, C.A. 1965. Effect of photoperiod and nitrogen on reaction of sesame to Pseudomonas sesami and Xanthomonas sesami. Plant Disease Reporter 49 (2): 119-120.

Vallad, G.E. and Gooman R.M. 2004. Systemic aquired resistance and induced systemic resistance in conventional agriculture. Crop Sci. 44: 1920-1934.

Van Loon, L.C., Bakker, P.A.H.M., and Pieterse, C.M.J. 1998. Systemic resistance induced by rhizosphere bacteria. Annual Review of Phytopathology 36: 453-483.

Weller, D.M. 1988, Biological control of soilborne plant pathogens in the rhizosphere with bacteria. Annual Review of Phytopathology 26 379-407.

Wilson, M., Campbell, H.L., Ji, P., Jones, J.B., and Cuppels, D.A. 2002. Biological control of bacterial speck of tomato under field conditions at several locations in North America. Phytopathology 92: 1284-1292.

Wilson, M., Campbell, H.L., Jones, J.B., Suslow, T.V., and Cuppels, D.A. 1996. Biological control of bacterial speck of tomato. (abstract) Phytopathology 86: s49.

CHAPTER 3

Integrated Effects of Mulch, Irrigation, Fungicides and Bacillus spp. on Fresh Market Tomato

INTRODUCTION

Fresh market tomato is an important crop in Ohio, with 2,832 planted hectares, an average yield per hectare of 46,448 kg and a value of $ 89,244,000 (USDA,

2003). ‘Mountain Spring’ is one of the most commonly planted varieties and like most commercial fresh market tomato varieties is intensively managed.

The intensive management of tomato includes fertilization, water management, use of black plastic mulch, and use of fungicides (Hockmuth, 1992; Masiunas et al. 1997). Nitrogen, phosphorous and potassium are the most common macronutrients included in a tomato fertilization program. Nitrogen is usually supplied (197 kg/ha) during early stages of development (Wilcox, 1993). Certain varieties of tomato, such as ‘Mountain Spring’ require only 67 kg/ha to produce

48 ton/ha (Taber, 1998). Tomato total uptake of phosphorous is 26 kg/ha, which is particularly rapid during the fruiting stage (Wilcox, 1993). Potassium uptake is about 12 g/plant during the vegetative and fruiting stages (Wilcox, 1993). Water 105

management not only affects uptake and utilization of nutrients, but also the

development of certain diseases, herbicide activation and quality and quantity of

tomato production. Drip irrigation is one of the most common delivery systems, in

which water under pressure flows through a pipe. Drip irrigation places water

directly where it is needed. Water deficiency during vegetative, flowering or

fruiting stages results in yield reductions of 25, 52 and 43 % respectively

(Rutledge et al. 1999). The use of black plastic mulch is widely accepted since

plastic mulch controls weeds and reduces certain diseases, conserves moisture

and increases quality and quantity of marketable fruit. Plastic mulch is generally installed over a 100 cm wide x 12 cm high bed. The main disadvantage of the use of plastic is the expense associated with installation, removal and disposal of the black plastic (Rutledge et al. 1999). Mulching fresh market tomato with residue from the previous winter annual crop may be an alternative to plastic mulch to control weeds and reduce production costs and impact on the environment. The fact that the severity of foliar diseases in fresh market tomato is lower in plants grown under plant residue mulch compared with those grown on plastic mulch is also attractive (Mills et al. 2002). Nearly 200 biotic diseases have been reported on tomato throughout the USA (Jones et al. 1997). The most important foliar diseases in Ohio are early blight (Alternaria solani) and septoria leaf spot (Septoria lycopersici). Control of foliar diseases of tomato is achieved

106

by crop rotation, balanced nutrition, destruction of infected leaves, eradication of

weed or volunteer plants (Jones, 1997; Menzies and Jarvis, 1994a; Menzies and

Jarvis, 1994b; Pitblado, 1994; Pitblado and Howard, 1994; Stevenson, 1997) and the use of fungicides. Amistar 80WG, Bravo Weather Stik, Dithane M45, Tanos

50DF, Kocide 2000, Manzate75DF, and Cabrio 20 WG are among the fungicides applied to manage foliar and fruit diseases, particularly early blight (Alternaria

solani), septoria leaf spot (Septoria lycopersici), and anthracnose (Colletotrichum

gloeosporioides, C coccodes) on tomato (Lewis Ivey et al. 2005; Lewis Ivey et al.

2003). However, tomato pathogens may acquire resistance to fungicides (Day et al. 2004; Grech, 1990; Yun et al. 1999). For instance, Phytophthora infestans

isolates resistant to metalaxyl are widespread in tomato commercial sites in

England and USA (Day et al. 2004; Yun et al. 1999).

The use of plant growth-promoting rhizobacteria (PGPR) represents a potentially

attractive alternative disease management approach since PGPR have been

reported to increase yield and protect crops simultaneously (Ramamoorthy et al.

2002; Raupach, 1998). PGPR have been reported to stimulate plant growth and

improve stand under stress conditions (van Loon et al. 1998). Three strains of

PGPR, Bacillus subtilis MBI600 (Microbio, Bolder, CO) and GBO3 (Gustafson

Plano, TX), and Bacillus amyloliquefaciens IN937 (Auburn University, AL), have

107

been reported to act as biological control agents against various plant pathogens

in numerous field and vegetable crops (Martinez-Ochoa, 2000; Zehnder et al.

2000; Reddy et al. 2000; Raupach, 1998; Ryu et al. 2000; EPA, 2004a; EPA,

2004b). The combined effect of B. subtilis strain GBO3 and B. amyloliquefaciens

strain IN937 has been studied by Martinez-Ochoa (2000), who found a reduction

in galling of tomato plants inoculated with a mix of the two compared with control

plants, and Reddy et al. (2000), who found a reduction of severity of bacterial leaf

spot, angular leaf spot, and blue mold in PGPR-inoculated plants of tomato,

cucumber and tobacco, respectively, compared with control plants. The use of

the mixture of GBO3+IN937 was first patented (United States Patent 6524998)

and then commercialized by Gustafson. Plant disease control using PGPR has been variable across locations and crops, several factors may influence the ability of PGPR to affect plant growth parameters and disease suppression. It is necessary that PGPR establish, colonize, and reach population densities sufficient to exert beneficial effects (Bloemberg and Lugtenberg 2001; Benizri et

al. 2001). Soil moisture may influence the ability of PGPR to establish in

rhizosphere. For instance, the survival and activity of Pseudomonas fluorescens

in soil was significantly greater at -30 and -750 kPa compared with -1500 kPa

(Meikle et al. 1995). Competition may also be a factor. The survival of Bacillus

cereus B11 was negatively affected by the presence of Flavobacterium P25 and

108

Arthrobacter A109 in wheat rhizosphere and non-planted soil (Young et al. 1995).

Nutrients, inorganic compounds and plant-derived factors can modulate antibiotic

production in Bacillus cereus UW85 (Milner et al. 1995). Antibiosis and

antagonism in Pseudomonas flourescens strain Pf-5 were modulated by the

composition of the medium used to perform the test in vitro (Rodriguez and

Pfender, 1997). These observations may help explain the variable effectiveness

of PGPR in numerous studies.

The goal of this study was to evaluate the effect of soil moisture and mulch type

on PGPR colonization of tomato rhizospheres and subsequent effect on plant height, foliar diseases and yield of tomato. We hypothesized that colonization of tomato roots with one or more PGPR would reduce disease and enhance crop yield and quality. It was also hypothesized that moisture availability would

influence PGPR colonization. Finally, we also hypothesized that the use of rye mulch would decrease severity of foliar diseases, increase yield, conserve

moisture, and influence PGPR colonization.

109

MATERIALS AND METHODS

Experimental Design

This study was conducted in 2002 (Experiment I) and 2003 (Experiment II) on

Wooster silt loam soil at the Ohio Agricultural Research and Development

Center, Wooster. The experimental plots were prepared according to standard

commercial practices. Plots consisted of 12 subplots spaced 3 m apart. Each

subplot contained 5 rows (1 m wide, 18 m long) spaced 1.5 m apart. Three row

treatments were located in the center of each subplot. Treatments consisted of a

row of 11 plants spaced 60 cm apart within the row. Treatments were arranged in

a split-split-split plot design (mulch, irrigation, fungicide, PGPR) with three

replications in the first experiment and in a factorial design (mulch, pesticide,

PGPR) with six replications in the second experiment. Mulches used were

opaque black plastic and plant residue. Rye used as plant residue mulch was

planted by the middle of October at a density of 100 kg/ha. Irrigation was

provided at 50% and 100% replacement of estimated crop evapotranspiration.

The irrigation tape (TSX 510-30-340, T-systems International, San Diego, CA) was installed on the bed under the mulch. The irrigation demand was estimated

110

by monitoring soil moisture using tensiometers installed at depths of 17 and 38 cm.

Production of PGPR-Inoculated Tomato Seedlings

Rifampicin-resistant strains were isolated as described previously (Chapter 1).

Inoculum was prepared by suspending 13 day-old cultures grown on endospore- forming medium in sterile distilled water. The concentration was adjusted to 7.7-

-1 8.3 log CFU ml (A600 = 0.6-0.8). Endospores constituted 95, 98, 77% and 98,

94, 26% of the MBI600, IN937, GBO3 inoculum suspensions for the first and second experiment, respectively. Seeds of ‘Mountain Spring’ tomatoes (Siegers

Seed Co. Rochester, NY) were germinated in 288 cell plastic flats containing soil- less potting mix (Conrad Fafard Inc, Agawam, MN). Fifteen-day-old seedlings were inoculated in the flats by drenching each plant with 1 ml PGPR suspension

(MBI600, or GBO3+IN937). Control plants were drenched with 1 ml sterilized distilled water. Seedlings were maintained in a greenhouse at 27 oC for 25 days.

Seedlings (40 days old) were hardened off for 7 days prior planting by reducing temperature and water, increasing ventilation and providing full sunlight to the seedlings.

111

Assessment of Tomato Root Colonization by PGPR

Root colonization by Bacillus spp. was evaluated in 29 and 49 day-old seedlings

in the greenhouse prior to planting and in 102 and 132 day-old plants during the

flowering and fruiting stages, respectively in the field in Experiment I. In

Experiment II, root colonization by Bacillus spp. was evaluated in 36 and 60 day- old seedlings in the greenhouse prior to planting and 114 and 143 day-old plants during the flowering and fruiting stages in the field.

In Experiment I, 24 plants were selected at random from MBI600-, GBO3+IN937-

and water- inoculated plants from 288 cell plastic flats at both evaluation dates.

The roots of four seedlings were pooled into one sample and processed as

described below. In Experiment II, thirty-six plants were selected at random from

MBI600-, GBO3+IN937- and water- inoculated plants from 288 cell plastic flats at both evaluation dates. Individual roots were processed as described below.

Under field conditions, the weeds and debris around four plants from each

replicate plot were removed. Eight soil cores (2.5 cm diameter 7-15 inches depth)

were taken from these four plants at each sampling date (576 soil core samples

from 288 plants). The soil cores were placed in a plastic bag and stored at 4 oC

112

up to 72 hours before processing. The soil cores were mixed until uniform in

appearance. The tomato roots were collected by hand from the soil.

Whole roots (seedling and vegetative stages) or root fragments (flowering and

fruiting stages) were excised and shaken to remove all but tightly adhered soil.

Roots were weighed and shaken in 20 ml potassium phosphate washing buffer

(KPB: 10mM K2HPO4, 10mM KH2PO4 pH 7.4) for 10 minutes on a rotary shaker

at approximately 120 rpm. A 12 ml aliquot was removed from each sample and

centrifuged for 10 minutes at 16000 x g. The bacterial pellet was re-suspended in

1ml KPB buffer, and 10-fold serial dilutions were made. One-hundred μl of

dilutions 10-3, 10-4, and 10-5 were spread onto nutrient agar amended with 60 mg

l-1 rifampicin (two plates per dilution). Plates were incubated for 7 days at room

temperature (24 oC) under 8 h white light (Sylvania 20W, warm light) and the

number of colonies that were morphologically similar to MBI600, GBO3, and

IN937 in each treatment was recorded. Three colonies representative of the

inoculated strain from each sample were purified on nutrient agar, and the isolate

identity was confirmed using rep-PCR with ERIC primers. A single colony was

transferred to a 1.5 ml Eppendorf tube containing 250 μl sterile distilled water.

Bacteria were pelleted by centrifugation (16000 x g for 10 minutes). Pellets were suspended in 186 μl lysis buffer (20mM TRIS; 2mM EDTA; 1.2 % Triton x-100;

113

20mg/l lysozyme) and incubated at 37 oC for 40 minutes. Genomic DNA was

extracted using the Qiagen DNeasy tissue kit (Valencia, CA. USA) according to the manufacturer’s instructions. rep-PCR was carried out essentially as described by Louws et al. (1996) with 0.8 mM of each of the primers [ERIC 1R

(5’-ATGTAAGCTCCTGGGGATTCAC-3’); ERIC 2 (5’-

AAGTAAGTGACTGGGGTGAGCG-3’)], 1.25mM of each of four dNTP’s, 1.6 u

Taq polymerase, and variable concentration of template. After amplification, 7μl

reaction mixture plus 3 μl loading dye [5 mg bromophenol blue; 5 ml 5x TBE

(0.45 M Tris-Borate; 0.01 M EDTA; pH 8.3); 2 g sucrose] were loaded into 1.5 %

agarose gels in 0.5 x TBE and amplified DNA fragments were separated by

horizontal gel electrophoresis (Midicell e350, E-C apparatus corporation, St.

Petersburg, FL, USA) at 50 V for 240 minutes at 10 oC. Amplification products

(stained in 2 μg ml-1 ethidium bromide for 15 minutes) were analyzed under UV

light transilluminator. Images were photographed using an EDAS290 (Kodak,

Rochester, N.Y., USA) system, gel size 13 x 17 cm and 1 - 3.5 seconds of

exposure time. Colony counts were adjusted based on the proportion of PCR

fingerprints corresponding to the inoculated Bacillus strain in the sample. Colony

forming units per gram of root were calculated and one unit was added (to avoid

obtaining a logarithm of zero) and the data were Log10 transformed before

114

statistical analysis. Population densities reported are the average of all sampled roots.

Plant Growth Evaluation

Plant height, number of leaves, leaf length and width, leaf area, nitrate concentration in sap and fresh and dry weight were evaluated on four plants at the edges of each replicate plot on 81, 95 and 116 (Experiment I) and 101, 118, and 141 day-old plants (Experiment II). The third leaf of each plant was cut and used to quantify nitrate concentration in the sap in situ. Nitrate concentration in fresh sap was measured with a Cardy nitrate meter (Spectrum Technologies,

East Plainfield, IL, USA), which was recalibrated every 30 samples according to the manufacturer’s instructions. The samples were placed in a plastic bag, refrigerated, and transferred to the laboratory for the other measurements. The area under the curve of these variables was used to compare treatments.

Disease Management and Evaluations

Timing of fungicide applications was based on TOMCAST (http://www.ag.ohio- state.edu/~vegnet/tomcats/tomfrm.htm) daily severity values (DSV) supplied by

SkyBit, Inc. (Boalsburg, PA, http://www.skybit.com). Fungicide was applied using a CO2-pressurized backpack sprayer when 15 DSV accumulated. Applications of

115

chlorothalonil (Bravo Ultrex, 1.65 kg/ha, Syngenta Crop Protection, Greensboro,

NC) or mancozeb (Manex ii, 4.2 L/ha, Griffin LLC, Valdosta GA) were alternated

with azoxystrobin (Quadris, 0.45 L/ha, Syngenta Crop Protection). The severity

and incidence of disease on seven plants at the center of each replicate plot was

evaluated weekly using a modified Horsfall-Barratt scale (Scott et al. 1997). The

Area Under the Disease Progress Curve (AUDPC) was calculated.

Yield Determination

Seven plants located at the center in each replicate plot were harvested at least

three times before the end of the growing season. Mature green tomatos that had pink coloration on the blossom end of the fruit were harvested by hand into bulk bins. The fruit were classified in eight categories: red marketable, green marketable, physiological disorders, anthracnose, bacterial diseases, other diseases, insect or bird damage, and green with other diseases (Figure 3.1).

Total marketable fruit was the sum of green and red marketable fruit.

116

Figure 3.1 Categories for tomato fruit assessment: A. red marketable; B. green marketable; C. physiological disorders; D. anthracnose; E. bacterial diseases; F. other diseases; G. insect or bird damage and H. other diseases on green fruit.

117

Weather Data

Experiment I: Average maximum temperatures for 10-30 Jun, Jul, Aug, and 1-5

Sep were 31.7, 31.0, 30.0 and 29.8 oC; average minimum temperatures were

17.3, 16.8, 15.7 and 13.8 oC; and total rainfall was 18.5, 21.8, 50.0 and 4.3 mm.,

respectively. Experiment II: Average maximum temperatures 23-31 May, Jun,

Jul, Aug, and 1-9 Sep were 20.1, 24.8, 28.1, 28.2 and 23.6 oC; average minimum

temperatures were 6.5, 12.6, 14.8, 15.1 and 10.9 oC; and rainfall was 27.6, 67.3,

137.6, 88.9 and 45.7 mm., respectively.

Data Analysis

The general linear model was used to perform Analysis of Variance. Least squares means were used to compare interactions (proc glm; class rep mulch irrig fungi treat; model markett = rep mulch irrig mulch*irrig mulch(rep irrig) fungi mulch*fungi irrig*fungi mulch*irrig*fungi fungi(rep irrig mulch) treat treat*mulch treat*irrig treat*mulch*irrig treat*fungi treat*mulch*fungi treat*irrig*fungi treat*mulch*irrig*fungi / ss3; test h = mulch irrig mulch*irrig e = mulch(rep irrig); test h = fungi fungi*mulch fungi*irrig fungi*mulch*irrig e = fungi(rep irrig mulch); means mulch|irrig / lsd lines e = mulch(rep irrig); means fungi mulch*fungi irrig*fungi mulch*irrig*fungi / lsd lines; means treat treat*mulch treat*irrig treat*mulch*irrig treat*fungi treat*mulch*fungi treat*irrig*fungi

118

treat*mulch*irrig*fungi / lsd lines; lsmeans mulch*irrig mulch*fungi treat*irrig*fungi treat*mulch*irrig*fungi / pdiff; run; in the second experiment the following line was added: rep_ = irrig + rep;).

119

RESULTS

Root colonization by Bacillus spp.

In Experiment I, MBI600 was recovered from 100, 100, 79 and 58 % of the root samples inoculated with this strain during the seedling, vegetative, flowering and fruiting stages, respectively (Table 3.1). Strain identity was confirmed by ERIC-

PCR. Cross contamination by GBO3 was detected in up to 4% of the root samples during the fruiting stage but not at any other stage (Table 3.1).

GBO3+IN937 were recovered from 100, 100, 8 and 4 % of the root samples inoculated with these strains during the seedling, vegetative, flowering and fruiting stages, respectively (Table 3.1). Cross contamination by MBI600 was detected in up to 4% of the root samples during the flowering and fruiting stages

(Table 3.1). All of the Bacillus strains were recovered from water-inoculated plants (control) at some time during the experiment. MBI600 was detected in 8% of the root samples in water-treated plants only during the flowering and fruiting stages, respectively (Table 3.1). GBO3 was detected in 18, 6, and 12% of the root samples in water-treated plants during the vegetative, flowering and fruiting stages, respectively (Table 3.1). IN937 was detected in 33, 63, 39 and 4% of the

120

root samples in water-treated plants during the seedling, vegetative, flowering and fruiting stages, respectively (Table 3.1).

In Experiment II, MBI600, was recovered from 100, 100, 50 and 58 % of the root samples inoculated with this strain during the seedling, vegetative, flowering and fruiting stages, respectively (Table 3.2). Strain identity was confirmed by ERIC-

PCR. No cross contamination by GBO3 or IN937 was observed (Table 3.2).

Either GBO3 or IN937 was recovered from 83, 83, 29 and 25 % of the root samples inoculated with these strains during the seedling, vegetative, flowering and fruiting stages, respectively (Table 3.2). Cross contamination by MBI600 was detected only in 4% of the root samples evaluated during the flowering stage

(Table 3.2). All of the Bacillus strains were recovered from water-inoculated plants (control) at some time during the experiment. MBI600 was detected in 25,

4 and 18 % of the root samples in water-treated plants during the vegetative, flowering and fruiting stages, respectively (Table 3.2). GBO3 was detected only in 2% of the root samples in water-treated plants during the fruiting stage (Table

3.2). IN937 was detected in 8 and 4% of the root samples in water-treated plants during the seedling and fruiting stages, respectively (Table 3.2).

121

PGPR population densities of 104-106 CFU g-1 were detected in tomato roots during the seedling stage, then decreased over time to less than 103 CFU g-1

(Figs. 3.3, 3.4). In Experiment I, B. subtilis MBI600 population density was significantly higher than that of GBO3+IN937 on 49, 102, and 132 day-old plants.

Control plants were contaminated at the seedling stage with IN937 and GBO3 at densities of 103 - 104 CFU g-1. In Experiment II, the GBO3+IN937 and MBI600

treatments had similar population densities, except on 36 day-old seedlings in

which the MBI600 population was significantly higher than that of GBO3+IN937.

Contamination of non-inoculated seedlings by IN937 and MBI600 was negligible

remaining less than 101 CFU g-1. The mean population density of MBI600 and

GBO3+IN937 on tomato roots across all treatments and samples tended to be

higher in plots irrigated at 100% than those irrigated at the 50% rate during the

flowering stage in the first but not second year (Figure 3.5). Application of

fungicides or type of mulch did not affect Bacillus spp. population density in either

experiment.

122

Inoculum Stage Percent of plants colonized by MBI600 GBO3 IN937 Other None Seedling 100.00 0.00 0.00 0.00 0.00 Vegetative 100.00 0.00 0.00 0.00 0.00 Bacillus subtilis MBI600 Flowering 79.16 0.00 0.00 4.16 16.66

Fruiting 58.33 4.16 0.00 8.33 29.16 Seedling 0.00 33.33 66.66 0.00 0.00 Vegetative 0.00 37.50 62.50 0.00 0.00 B. subtilis GBO3 + B. amyloliquefaciens IN937 Flowering 4.16 0.00 8.33 58.33 29.16 Fruiting 4.16 2.08 2.08 54.16 37.50 Seedling 0.00 0.00 33.33 66.66 0.00 Vegetative 0.00 18.18 63.63 18.18 0.00 Water

123 Flowering 8.33 6.25 39.58 25.00 20.83 Fruiting 8.33 12.50 4.16 54.16 20.83

Table 3.1. Bacterial population composition on tomato roots inoculated with Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937. Colonies were isolated from roots on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on three colonies per sample. Seedling (29 day old), vegetative (49 day old), flowering (102 day old) and fruiting (132 day old) stages (Experiment I).

B. ng (114 ng (114 day old) 0.00 0.00 0.00 0.00 strain GBO3 and and GBO3 strain None 45.83 58.33 41.66 66.66 41.66 62.50 41.66 25.00 0.00 0.00 8.33 0.00 Other 29.16 16.66 25.00 33.33 33.33 16.66 41.66 16.66 B. subtilis 4.16 8.33 0.00 0.00 0.00 0.00 0.00 0.00 25.00 75.00 25.00 79.16 IN937 00 16 00 4. 0. 0. 0.00 2.08 8.33 0.00 0.00 4.16 0.00 0.00 0.00 GBO3 strain MBI600, strain MBI600, Percent of plants colonized by Percent of plants colonized 0.00 0.00 0.00 0.00 4.16 4.16 25.00 18.75 58.33 50.00 100.00 100.00 MBI600 mended medium. Identification was based on colony morphology morphology colony on was based Identification medium. mended Bacillus subtilis

Stage Vegetative Flowering Fruiting Seedling Vegetative Flowering Fruiting Seedling Vegetative Flowering Fruiting Seedling r sample. Seedling (36 day old), vegetative (60 day old), floweri old), day (60 old), vegetative day (36 Seedling sample. r

IN937

Inoculum Inoculum B. amyloliquefaciens B. amyloliquefaciens MBI600 GBO3 + GBO3 strain IN937. Colonies were isolated from roots on rifampicin-a on roots from isolated were Colonies IN937. strain Bacillus subtilis B. subtilis Water

124 amyloliquefaciens amyloliquefaciens and rep-PCR with ERIC primers carried out on three colonies pe colonies three out on carried primers with ERIC rep-PCR and and fruiting (143 day old) stages (Experiment II). (Experiment stages day old) (143 and fruiting with inoculated roots tomato in composition population 3.2. Bacterial Table

Figure 3.2. Identification of Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 isolated from 29-day old tomato roots during seedling stage, using enterobacterial repetitive intergenic consensus - polymerase chain reaction (ERIC-PCR) fingerprints. Wild-type and rifampicin- resistant strain fingerprints were used as a reference. The banding patterns of three colonies isolated from treatments (control, MBI600 and GBO3+IN937) were compared with the reference patterns for identification (Experiment I).

125

6.17 A 6 MBI600rif 5.59 A GBO3rif+IN937rif 5 5.03 A Water

4.31 AB 4 4.13 A 3.64 A 4.23 B

3 2.62 B 2.68 A Log CFU/g root Log CFU/g

1 1.03 B

0.4 C 0 0.13 B 29 49 102 132 Plant age (day-old)

Figure 3.3 Tomato root colonization by Bacillus spp. 29 (seedling stage), 49 (seedling stage), 102 (flowering stage) and 132 (fruiting stage) days after seeding, Experiment I. Control plants were colonized by MBI600 (flowering and fruiting stages), GBO3 (vegetative, flowering and fruiting stages), and IN937 (seedling, vegetative, flowering and fruiting stages). Means followed by the same letter are not significantly different (p-value < 0.05).

126

6 6.09 A 6.07 A MBI600rif GBO3rif+IN937rif 5 4.82 B 4.87 A Water

Log CFU/g root CFU/g Log 2.56 A

2 1.89 A 1.38 A 1.20 B 1.24 A 1 1.09 A

0.35 C 0 0.12 B 36 60 114 143 Plant age (day-old)

Figure 3.4. Tomato root colonization by Bacillus spp. 36 (seedling stage), 60 (seedling stage), 114 (flowering stage) and 143 (fruiting stage) days after seeding, Experiment II. Control plants were colonized by MBI600 (vegetative, flowering and fruiting stages), GBO3 (fruiting stage), and IN937 (seedling and fruiting stages). Means followed by the same letter are not significantly different.

127

5 4.5 3.89 A 4.37 A 4 3.5 3 MBI600rif 2.5 GBO3rif+IN937rif 2

Log CFU/groot 1.5 1 0.8 BC 0.5 0 C 0 50 100 Irrigation (replacement of estimated crop evapotranspiration)

Figure 3.5. Effect of irrigation on Bacillus spp. population density on 102-day-old ‘Mountain Spring’ tomato roots. Means followed by the same letter are not significantly different (p-value < 0.05). Experiment I.

Effect of Treatments on Plant Growth

Bacillus sp. IN937+GBO3 significantly increased the height of tomato seedlings

both years compared to the control (Figure 3.6). Under field conditions, both

IN937+GBO3 and MBI600-inoculated plants were taller than non-inoculated plants in both years (Figure 3.7, Tables 3.3 and 3.4). In Experiment I, plants grown using plastic mulch were taller than those grown using rye residue mulch; the opposite was observed in the second year (Figure 3.8). No effect on plant height was induced by irrigation in either year (Tables 3.3 and 3.4).

128

The number of leaves per plant was not significantly affected by Bacillus sp.

treatment, irrigation rate, nor fungicides compared to the control. However, plants

grown under plastic mulch had significantly more leaves than plants grown under

plant residue mulch in Experiment I (Table 3.3). In Experiment II, the number of

leaves was higher in plants grown on plant residue mulch than those grown

under plastic. The number of leaves per plant was significantly reduced in plants

that were inoculated with GBO3+IN937 compared with water-inoculated plants,

but MBI600-inoculated plants did not affect leaf number (Table 3.4)

Leaf length and width were not significanlty affected by Bacillus sp. inoculation,

fungicide nor irrigation rate compared to the control. A significant increase in

these variables was observed in plants grown under plastic mulch, compared

with those grown under plant residue mulch (Experiment I, Table 3.3). In

Experiment II, leaf length and width were significantly reduced in plants that were

inoculated with GBO3+IN937 compared with water-inoculated plants, but but

MBI600-inoculated plants did not affect leaf length nor width. No significant effect

on length and width was observed on plants grown under different types of

mulches or fungicide applications.

129

In Experiment I, nitrate concentration in sap was not significanlty affected by

Bacillus sp. treatment, irrigation rate, nor fungicides compared to the control.

(Table 3.3). In Experiment II, nitrate concentration in sap was significantly increased in plants that were inoculated with GBO3+IN937 compared with water- and MBI600- inoculated plants (Table 3.4). In both experiments, plants grown under plastic mulch had significantly more nitrate than the ones grown under plant residue mulch (Tables 3.3, 3.4). No effect of fungicides on nitrate concentration in sap was observed.

Foliar fresh and dry weights were not significantly affected by Bacillus sp.

inoculation, irrigation rate, nor fungicides compared to the control in either

experiment. However, plants grown under plastic mulch had significantly higher

fresh and dry weight than those grown under plant residue mulch in both

experiments (Tables 3.3, 3.4).

130

Source DF Levels Plant height Number of leaves Leaf length Leaf width Mulch 1 p 0.0006 F 41.83 p 0.0005 F 46.27 p 0.0009 F 36.61 p 0.0014 F 31.58 Plastic 35.33 A 10.09 A 23.91 A 22.77 A Rye residue 25.03 B 7.13 B 15.66 B 12.99 B Irrig 1 p 0.2921 F 1.33 p 0.1874 F 2.21 p 0.2147 F 1.92 p 0.2736 F 1.45 50% 31.12 A 8.94 A 20.71 A 18.86 A 100% 29.25 A 8.28 A 18.86 A 16.91 A Mulch*irrig 1 p 0.5224 F 0.46 p 0.6416 F 0.24 p 0.7477 F 0.11 p 0.7635 F 0.10 fungi 1 p 0.2394 F 1.62 p 0.3549 F 0.96 p 0.6757 F 0.19 p 0.5950 F 0.31 Yes 30.65 A 8.72 A 19.97 A 18.05 A No 29.72 A 8.50 A 19.61 A 17.71 A Mulch*fungi 1 p 0.6369 F 0.24 p 0.4730 F 0.57 p 0.6163 F 0.27 p 0.7057 F 0.15 irrig*fungi 1 p 0.8975 F 0.02 p 0.4259 F 0.70 p 0.1701 F 2.27 p 0.2272 F 1.56 Mulch*irrig*fungi 1 p 0.4456 F 0.64 p 0.3775 F 0.87 p 0.9411 F 0.01 p 0.6840 F 0.18 treat 2 p 0.0286 F 3.98 p 0.4717 F 0.77 p 0.1489 F 2.02 p 0.1000 F 2.48 MBI600 30.35 AB 8.77 A 20.47 A 18.73 A GBO3+IN937 31.06 A 8.59 A 19.21 A 17.24 A

131 Control 29.14 B 8.47 A 19.69 A 17.68 A mulch*treat 2 p 0.9348 F 0.07 p 0.5954 F 0.53 p 0.2584 F 1.41 p 0.3686 F 1.03 irrig*treat 2 p 0.0161 F 4.71 p 0.0808 F 2.72 p 0.1993 F 1.70 p 0.3541 F 1.07 mulch*irrig*treat 2 p 0.4442 F 0.83 p 0.2062 F 1.66 p 0.1760 F 1.83 p 0.2336 F 1.52 fungi*treat 2 p 0.0036 F 6.72 p 0.2322 F 1.53 p 0.1161 F 2.30 p 0.1643 F 1.91 mulch*fungi*treat 2 p 0.7267 F 0.32 p 0.6197 F 0.49 p 0.6225 F 0.48 p 0.6646 F 0.41 irrig*fungi*treat 2 p 0.0174 F 4.61 p 0.1579 F 1.96 p 0.7130 F 0.34 p 0.9191 F 0.08 mulc*irri*fung*treat 2 p 0.9424 F 0.06 p 0.3625 F 1.05 p 0.3657 F 1.04 p 0.5732 F 0.57

Continued

Table 3.3. Growth promotion of ‘Mountain Spring’ tomato inoculated with rifampicin-resistant mutants of Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937 under field conditions (Experiment I). Plant height, number of leaves, leaf length, leaf width, nitrate concentration, foliar fresh and dry weights and leaf area were evaluated three times, on 81, 95, and 116 day-old plants. The area under the curve was used for statistical analysis. Averages are shown. Irrigation, fungicide, and PGPR treatment are abbreviated as ‘irrig’, ‘fungi’, and ‘treat’, respectively. Levene’s p-values were 0.950, 0.893, 0.996, 0.998, 0.983, 0.999, 1.000, and 0.999 for plant height, number of leaves, leaf length, leaf width, nitrate concentration, foliar fresh and dry weights and leaf area, respectively. Normality was assumed.

Table 3.3 continued

Source DF Levels Nitrate concentration Foliar fresh weight Foliar dry weight Leaf area Mulch 1 p<0.0001 F 117.85 p 0.0005 F 47.04 p 0.0002 F 61.58 p 0.0005 F 47.41 Plastic 4129.2 A 16.11 A 2.11 A 133.51 A Rye residue 1668.0 B 5.49 B 0.73 B 56.68 B Irrig 1 p 0.1082 F 3.56 p 0.2596 F 1.55 p 0.1490 F 2.74 p 0.1672 F 2.47 50% 3142.4 A 11.71 A 1.57 A 103.20 A 100% 2654.9 A 9.89 A 1.28 A 86.99 A mulch*irrig 1 p 0.2264 F 1.82 p 0.9316 F 0.01 p 0.9324 F 0.01 p 0.9319 F 0.01 fungi 1 p 0.0524 F 5.18 p 0.2223 F 1.75 p 0.1074 F 3.29 p 0.4117 F 0.75 Yes 3091.9 A 11.28 A 1.50 A 98.65 A No 2705.4 A 10.32 A 1.35 A 91.55 A mulch*fungi 1 p 0.8838 F 0.02 p 0.9973 F 0.00 p 0.7285 F 0.13 p 0.8678 F 0.03 irrig*fungi 1 p 0.6197 F 0.27 p 0.2095 F 1.86 p 0.0686 F 4.42 p 0.6007 F 0.30

132 mulch*irrig*fungi 1 p 0.4883 F 0.53 p 0.9443 F 0.01 p 0.8732 F 0.03 p 0.6440 F 0.23 treat 2 p 0.8277 F 0.19 p 0.0755 F 2.80 p 0.0373 F 3.65 p 0.1008 F 2.47 MBI600 2912.0 A 11.82 A 1.58 A 105.47 A GBO3+IN937 2854.0 A 9.73 A 1.27 B 89.45 A Control 2929.9 A 10.85 A 1.41 AB 90.37 A mulch*treat 2 p 0.6518 F 0.43 p 0.0773 F 2.78 p 0.0678 F 2.93 p 0.5280 F 0.65 irrig*treat 2 p 0.9376 F 0.06 p 0.2034 F 1.67 p 0.2728 F 1.35 p 0.0272 F 4.04 mulch*irrig*treat 2 p 0.1642 F 1.91 p 0.8968 F 0.11 p 0.7277 F 0.32 p 0.9781 F 0.02 fungi*treat 2 p 0.0986 F 2.49 p 0.0987 F 2.49 p 0.0431 F 3.47 p 0.1081 F 2.39 mulch*fungi*treat 2 p 0.3628 F 1.05 p 0.9277 F 0.08 p 0.9786 F 0.02 p 0.8811 F 0.13 irrig*fungi*treat 2 p 0.7156 F 0.34 p 0.1182 F 2.28 p 0.1291 F 2.18 p 0.1645 F 1.91 mulc*irri*fung*treat 2 p 0.6042 F 0.51 p 0.3433 F 1.11 p 0.4719 F 0.77 p 0.3373 F 1.12

Source DF Levels Plant height Number of leaves Leaf length Leaf width Mulch 1 p 0.0119 F 14.19 p 0.0239 F 10.27 p 0.7045 F 0.16 p 0.3689 F 0.97 Plastic 43.21 B 9.69 B 20.31 A 19.45 A Rye residue 47.24 A 10.37 A 20.87 A 19.01 A fungi 1 p 0.8350 F 0.05 p 0.7028 F 0.15 p 0.8275 F 0.05 p 0.8007 F 0.07 Yes 45.06 A 10.05 A 20.69 A 19.64 A No 45.39 A 10.01 A 20.50 A 18.82 A mulch*fungi 1 p 0.8360 F 0.05 p 0.8632 F 0.03 p 0.2062 F 1.83 p 0.9291 F 0.01 treat 2 p<0.0001 F 40.21 p<0.0001 F 31.76 p 0.0001 F 11.69 p 0.0687 F 2.87 MBI600 50.12 A 10.76 A 21.90 A 19.99 AB GBO3+IN937 39.86 C 9.02 B 19.02 B 17.22 B Control 45.68 B 10.31 A 20.86 A 20.49 A mulch*treat 2 p 0.0653 F 2.92 p 0.0462 F 3.32 p 0.0350 F 3.65 p 0.0916 F 2.54 fungi*treat 2 p 0.5040 F 0.70 p 0.3232 F 1.16 p 0.9216 F 0.08 p 0.1902 F 1.73 mulch*fungi*treat 2 p 0.5663 F 0.58 p 0.5565 F 0.59 p 0.9743 F 0.03 p 0.1444 F 2.03

133 Continued

Table 3.4. Growth promotion of ‘Mountain Spring’ tomato inoculated with rifampicin-resistant mutants of Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937 under field conditions (Experiment II). Plant height, number of leaves, leaf length, leaf width, nitrate concentration, foliar fresh and dry weights and leaf area were evaluated three times on 101, 118, and 141 day-old plants. The area under the curve was used for statistical analysis. Averages are shown. Irrigation, fungicide, and PGPR treatment are abbreviated as ‘irrig’, ‘fungi’, and ‘treat’, respectively. Levene’s p-values were 0.926, 0.991, 0.994, 0.610, 0.999, 0.603, 0.573, and 0.879 for plant height, number of leaves, leaf length, leaf width, nitrate concentration, foliar fresh and dry weights and leaf area, respectively. Normality was assumed.

F 1.15 F 2.85 A A F 0.10 A A F 10.42 A B A F 0.05 F 0.34 F 1.26 77.23 96.87 88.13 85.97 67.28 93.71 100.17 p 0.3266 p 0.1522 p 0.7636 p 0.0002 p 0.8358 p 0.7114 p 0.2948 Leaf area Leaf area F 0.44 F 26.36 A B F 0.16 A A F 4.84 A B AB F 0.26 F 6.23 F 0.17 1.89 1.56 1.78 1.67 1.94 1.46 1.78 p 0.6490 p 0.0037 p 0.6994 p 0.0131 p 0.6179 p 0.0044 p 0.8478 Foliar dry weight weight Foliar dry F 2.99 A A F 2.35 A A F 1.22 A A A F 0.34 F 2.50 F 0.28 F 0.49 9.15 9.74 7.34 r fresh weight r fresh 6174 15.78 15.19 15.14 14.91 p 0. p 0.7544 p 0.7544 p 0.1441 p 0.1560 p 0.3049 p 0.1446 p 0.1446 p 0.7142 Folia F 1.15 F 6.99 A B F 1.11 A A F 3.33 B A B F 1.06 F 1.33 F 2.45 concentration concentration 3444.8 2223.9 2954.8 2713.9 2742.5 3096.2 2664.5 p 0.3280 p 0.0458 p 0.3160 p 0.0460 p 0.3554 p 0.2752 p 0.0995 Nitrate No Yes Plastic Control MBI600 Rye residue GBO3+IN937 Levels 2 DF 1 2 2 1 1 2 Mulch fungi mulch*fungi treat mulch*treat fungi*treat mulch*fungi*treat Source

134 Table 3.4 continued

Figure 3.6. Effect of soil drenches with rifampicin-resistant mutants of plant growth-promoting rhizobacteria (PGPR) (Bacillus subtilis MBI600 and GBO3, and B. amyloliquefaciens IN937) on height of ‘Mountain Spring’ tomato seedlings 49 (Experiment I) and 60 days (Experiment II) after planting. Means followed by the same letter are not significantly different (p-value < 0.05).

135

Figure 3.7. Effect of soil drenches with rifampicin-resistant strains of plant growth-promoting rhizobacteria (PGPR) (Bacillus subtilis MBI600 and GBO3, and B. amyloliquefaciens IN937) on height of ‘Mountain Spring’ tomato plants under field conditions. Plant height was evaluated on 81, 95 and 116 (Experiment I) and 101, 118, and 141 day-old plants (Experiment II). Means followed by the same letter are not significantly different (p-value < 0.05).

Figure 3.8. Effect of plastic and rye residue mulches on height of ‘Mountain Spring’ tomato plants under field conditions. Plant height was evaluated on 81, 95 and 116 (Experiment I) and 101, 118, and 141 day-old plants (Experiment II). Means followed by the same letter are not significantly different (p-value < 0.05).

136

Effect of Treatments on Diseases

Early blight (Alternaria solani) and septoria leaf spot (Septoria lycopersici Speg.)

were the primary foliar diseases first observed at 117 and 103 days after seeding

for the first and second years respectively (Figure 3.9). Treatment with Bacillus

spp. did not reduce early blight or septoria leaf spot in either year. The severity of

foliar diseases in plants grown on plant residue mulch was significantly lower

than for plants grown on plastic mulch in both experiments (Figure 3.10). No

effect on severity of foliar diseases was induced by drip irrigation level in either

experiment.

In Experiment II, an outbreak of stem rot (Pseudomonas cichorii) was observed

(Figure 3.9). Tomato plants inoculated with GBO3+IN937 had higher bacterial

stem rot severity than the control or those inoculated with MBI600 (Figure 3.11).

Tomato plants grown on plastic residue mulch were more affected by stem rot than plants grown on plant residue mulch (Figure 3.12). Neither fungicide application nor irrigation level affected stem rot severity.

137

Figure 3.9. Diseases and pathogens observed on ‘Mountain Spring’ tomato plants. A. early blight; B. Alternaria solani conidia; C. septoria leaf spot; D. pycnidia and conidia of Septoria lycopersici; E. bacterial stem rot; and F. pith desintegration induced by Pseudomonas cichorii.

138

Figure 3.10. Effect of plastic and plant residue mulches on severity of early blight and septoria leaf spot on ‘Mountain Spring’ tomato plants under field conditions. Severity of foliar diseases was evaluated every 7 days. Area under the disease curve was used for statistical analysis. Means followed by the same letter are not significantly different (p-value < 0.05).

Figure 3.11. Effect of plant growth-promoting rhizobacteria (PGPR) (Bacillus subtilis MBI600 and GBO3, and B. amyloliquefaciens IN937) on bacterial stem rot severity of ‘Mountain Spring’ tomato plants under field conditions. Severity of bacterial stem rot was evaluated every 7 days. Area under the disease curve was used for statistical analysis. Means followed by the same letter are not significantly different (p-value < 0.05).

139

Effect of Treatments on Marketable Yield

No significant increase in yield was induced by Bacillus spp. in either experiment.

In Experiment II, the yield of plants inoculated with GBO3+IN937 was significantly lower than that of control plants and plants inoculated with MBI600

(Figure 3.13). In Experiment I, marketable yields were 64 and 32 ton/ha for plants

grown under plastic and plant residue mulches, respectively. In Experiment II,

marketable yields were 10 and 23 ton/ha for plants grown under plastic and plant

residue mulches, respectively (Figure 3.14). Marketable yield of tomato was not

affected by irrigation level or fungicide in either experiment.

140

Figure 3.13. Effect of plant growth-promoting rhizobacteria (PGPR) (Bacillus subtilis MBI600 and GBO3, and B. amyloliquefaciens IN937) on marketable yield of ‘Mountain Spring’ tomato plants. Means followed by the same letter are not significantly different (p-value < 0.05).

Figure 3.14. Effect of plastic and plant residue mulches on marketable yield of ‘Mountain Spring’ tomato plants under field conditions. Means followed by the same letter are not significantly different (p-value < 0.05).

141

DISCUSSION

Fresh market tomato is an intensively managed, high value crop requiring

significant economic input. Most growers rely on fungicides to manage diseases.

However, alternative disease management strategies that offer acceptable yield

and quality while requiring less pesticide are needed. Integration of cultural and

biological methods may offer much-needed opportunities to overcome current

and developing challenges in vegetable pest management. These strategies

must be integrated with other crucial crop management inputs such as water

management.

Promising results have been published on increasing yield and suppressing plant

pathogens by application of Bacillus subtilis strains MBI600 and GBO3 and

Bacillus amyloliquefaciens strain IN937 (Martinez-Ochoa, 2000; Zehnder et al.

2000; Reddy et al. 2000; Raupach, 1998; Ryu et al. 2000; EPA, 2004a; EPA,

2004b). However, in those studies the PGPR were used as single component treatments. In this study, irrigation, mulch, and PGPR were integrated to explore their single and combined effect on diseases, plant growth and yield of fresh market tomato.

142

Colonization of tomato roots by rifampicin-resistant mutants of MBI600, GBO3 and IN937 was verified using rep-PCR over a period of 107 days. In both experiments, the percent of plants colonized by PGPR decreased from 100% to

58% and from 100% to 4% for MBI600 and GBO3+IN937, respectively.

Population densities were high (104 - 106 CFU g-1) during the seedling stage but

dropped to low levels (less than 103 CFU g-1) during the flowering and fruiting stages, when foliar diseases appeared. Perhaps the intensive care of the seedlings and protected environment during the first two stages of development

(greenhouse conditions) facilitated PGPR colonization of tomato roots by providing relatively constant temperature and moisture in the soil, factors that were not controlled in the field. Contamination of water-inoculated plants (control) by PGPR was high in Experiment I (up to 63 % of the sampled plants with 104

CFU g-1 root) but it was reduced in Experiment II (up to 18 % of the sampled

plants with 101 CFU g-1 root). Several facts may explain this observation. First,

MBI600-, GBO3+IN937- and water- inoculated seedlings were maintained in the

same greenhouse in Experiment I and in different greenhouses in Experiment II.

Second, even though treatments were separated on the work surface in

Experiment I, water splash during irrigation may have carried bacterial cells from one treatment to other. Finally, the air movement created by the cooling system in the greenhouse may have caused the movement of soil and bacterial cells

143

from inoculated to noninoculated plants. Our results show that soil moisture did not significantly affect rhizosphere populations of MBI600 and GBO3+IN937 on tomato rhizosphere in Experiment I, which was performed in a relatively dry year.

However, both population densities tended to be lower at the low rate of irrigation than the high rate of irrigation. Similarly, Schmidt et al. (2004) reported that populations of B. subtilis MBI600 were slightly reduced under low matric potentials. In Experiment II, which was carried out in a relatively rainy year, no differences in Bacillus spp. population densities were observed. Frequent and heavy rains accumulated vast amounts of water over the experimental plot and soil moisture did not fluctuate between 50 and 100% irrigation rate. Our data show that Bacillus spp. population densities on tomato rhizosphere were not affected by type of mulch.

MBI600 and GBO3+IN937 increased the height of tomato plants in this study.

However neither of these two treatments provided satisfactory control of early blight or septoria leaf spot. Although the average population density of these strains on roots during the seedling stage should have been sufficient to trigger induced systemic resistance or enhance plant growth if the mode of action of these PGPR is similar to those reported elsewhere (van Loon et al. 1998; Lui and

Sinclair, 1992), systemic induced resistance was not observed against early

144

blight and septoria leaf spot. The percent of colonization and population density

of MBI600 and GBO3+IN937 dropped to low level when foliar diseases

appeared. Perhaps induction of systemic resistance and growth enhancement in tomato variety ‘Mountain Spring’ are dependent on PGPR population density. At harvest, no increase in marketable yield was observed for plants treated with

PGPR, when compared with water-treated plants. On the contrary, plants that were inoculated with GBO3+IN937 were more susceptible to P. cichorii and a significant reduction in yield was observed. In both experiments, GBO3+IN937- inoculation resulted in production of young tissue (shown by significantly increased plant height and nitrate concentration in sap) that was more susceptible to P. cichorii. Similar increase in biomass of tomato fertigated with a high rate of nitrogen and increase of the susceptibility of four varieties of tomato to X. euvesicatoria was observed in two experiments performed previously

(Chapter 2).

Black plastic and rye residue mulches affected plant height, foliar and bacterial

disease susceptibility, and yield of fresh market tomato in both years. In the first

year, plants grown on plastic mulch were taller, had higher severity of foliar

diseases and higher marketable yield than those grown on plant residue mulch.

In the second year, plants grown in plastic mulch were smaller, had higher

145

severity of foliar and bacterial diseases and less marketable yield than those

grown on rye residue mulch. The lower severity of foliar and bacterial diseases in

plants grown on rye mulch may have been a result of reduced pathogen

inoculum dispersal as observed in other studies. Ntahimpera et al. (1998) found

that the presence of a cover crop reduced dispersal of Colletotrichum acutatum

conidia compared with bare soil. Ntahimpera et al. (1999) found that straw mulch

had low splash response (distribution and total mass of splashed droplets

produced) while plastic cover had the highest response. In our study, even

though the amount of rainfall was greater in the second year than in the first year,

the severity of foliar diseases was lower in plants grown on rye mulch,

suggesting that rye mulch reduced foliar disease severity more effectively than

plastic mulch. Similar observations have been recorded elsewhere. For instance

Mills et al. (2002) found that early blight (Alternaria solani) and septoria leaf spot

(Septoria lycopersici) intensity was lower in tomato grown on beds with hairy vetch mulch than in uncovered or plastic mulch covered beds.

146

CONCLUSIONS

• Bacillus subtilis MBI600 colonized tomato roots more actively and

persisted longer in tomato rhizospheres than GBO3 and IN937. High

population density and significant increases in plant height suggest that

these strains were metabolically active in fresh market tomato.

• Metabolically active colonies of Bacillus spp. in tomato rhizosphere did not

induce systemic resistance against foliar diseases or increase the number

of leaves, foliar fresh or dry weights, or marketable yield.

• High levels of nitrate in sap were detected in plants grown on plastic

mulch compared with those grown on rye residue mulch.

• Plants that were inoculated with GBO3+IN937 had significantly more

nitrate in sap than those inoculated with MBI600 or water-treated control in

the second experiment.

• Plants grown on plastic mulch had more biomass than those grown on rye

residue mulch.

• Plants grown on rye mulch had lower severity of foliar diseases and stem

rot than those grown on plastic mulch.

147

LIST OF REFERENCES

Benizri, E., Baudoin, E., and Guckert, A. 2001. Root colonization by inoculated plant growth promoting rhizobacteria. Biocontrol Sci. Technol. 11:557-574.

Bloemberg, G. V., and Lugtenberg, B. J. J. 2001. Molecular basis of plant growth promotion and biocontrol by rhizobacteria. Current Opinion in Plant Biology 4: 343-350.

Day, J.P., Wattier, R.A.M., Shaw, D.S. and Shattock, R.C. 2004. Phenotypic and genotypic diversity in Phytophthora infestans on potato in Great Britain, 1995-98. Plant Pathology 53: 303-315

EPA. 2004a. Bacillus subtilis MBI600 129082. http://www.epa.gov/pesticides/biopesticides/ingredients/factsheets/factshe et_129082.htm

EPA. 2004b. Bacillus subtilis GBO3 (129068). http://www.epa.gov/pesticides/biopesticides/ingredients/factsheets/factshe et_129068.htm

Grech, N.M. 1990. Fungicide resistance in plant pathogenic fungi: the problems and ways to cope. Calif-Grow 14(4): 36-37

Hockmuth, G.J. 1992. Fertilizer management for drip-irrigated vegetables in Florida. Hort Technol 2: 27 – 32.

Jones J.B., Stall, R.E. and Zitter, T.A. 1997. Compendium of Tomato Diseases. APS press. Minnesota. USA: 73 p.

Jones, J.P. 1997. Early blight. In Jones J.B., Stall, R.E. and Zitter, T.A. 1997. Compendium of Tomato Diseases. APS press. Minnesota. USA: 13 pp.

Lewis Ivey, M.L., Mera, J.R., and Miller, S.A. 2003. Evaluation of fungicides for the control of foliar and fruit diseases of processing tomatoes 2003. Vegetable disease management research reports 2003. Plant Pathology series No. 120. The Ohio State University. 1 p.

148

Liu, Z.L. and Sinclair, J.B. 1992. Population dynamics of Bacillus megaterium strain B153-2-2 in the rhizosphere of soybean. Phytopathology 82: 1297- 1301.

Louws, F.J., Rademaker, J.L.W., and de Bruijn, F.J. 1996. REP-PCR: A rapid technique for bacterial identification and assessment of genetic diversity. Workshop APS annual meeting 16 p.

Martinez-Ochoa, N. 2000. Biological control of the root-knot nematode with rhizobacteria and organic amendments. Ph.D. dissertation. Auburn University. Alabama: 120 p.

Masiunas, J., McGiffen, M., Wilen, C., Bell, C., Lanini, T., Derr, J., and Kolasani, G. 1997. Integrated weed management in horticultural crops. In Weed management in Horticultural Crops. American Society for Horticultural Science and Weed Science Society of America. Joint Workshop 6-7 February 1997. M.E.McGiffen: 1-16

Meikle, A., Amin-Hanjani, S., Glover, L.A., Killham, K., and Prosser, J.I. 1995. Matric potential and the survival and activity of a Pseudomonas fluorescens inoculum in soil. Soil Biology and Biochemistry 27(7): 881-892

Menzies, J.G. and Jarvis, W.R. 1994b. Septoria blight. In Howard, R.J., Garland, J.A. and Seaman, W.L. 1994. Diseases and pest of vegetable crops in Canada. The Canadian Phytopathological Society. Entomological Society of Canada. Canada: 350 pp.

Mills, D.J., Coffman, C.B., Teasdale, J.R., Everts, K.L., Abdul-Baki, A.A. Lydon, J., and Anderson, J.D. 2002. Foliar disease in fresh-market tomato grown

149

in different bed strategies and fungicide spray programs. Plant Disease 86: 955 – 959.

Milner, J.L., Raffel, S.J., Lethbridge, B.J., and Handelsman, J. 1995. Culture and conditions that influence accumulation of zwittermycin A by Bacillus cereus. Appl Microbiol Biotechnol 43: 685-691

Ntahimpera, N., Ellis, M.A., Wilson, L.L. and Madden, L.V. 1998. Effects of a cover crop on splash dispersal of Colletotrichum acutatum conidia. Phytopathology 88: 536 – 543.

Ntahimpera, N., Hacker, J.K., Wilson, L.L., Hall, F.R., and Madden, L.V. 1999. Characterization of splash droplets from different surfaces with a phase doppler particle analyzer. Agricultural and Forest Meteorology 97: 9 - 19.

Raupach, G.S. 1998. Induced systemic resistance as the principal component of integrated pest management of Cucumis sativus L. by PGPR. Ph.D. dissertation. Auburn University. Alabama. USA: 120 p.

Reddy, M.S., Ryu, C. M., Zhang, S., Yan, Z. and Kloepper, J. W. 2000. Aqueous formulations of plant growth-promoting rhizobacteria for control of foliar pathogens. http://www.ag.auburn.edu/~mreddy/.

150

Rodriguez, F., and Pfender, W.F. 1997. Antibiosis and antagonism of Sclerotinia homeocarpa and Drechslera poae by Pseudomonas fluorescens PF-5 in vitro and in planta. Phytopathology 87: 614-621

Rutledge, A.D., Wills, J.B., and Bost, S. 1999. Commercial tomato production. Agricultural Extension Service. The Universiy of Tennessee. PB 737: 31 p.

Stevenson, W.R. 1997. Septoria leaf spot. In Jones J.B., Stall, R.E. and Zitter, T.A. 1997. Compendium of Tomato Diseases. APS press. Minnesota. USA: 22 pp.

Taber, H.G. 1998. Petiole sap nitrate sufficiency values for fresh market tomato production. http://www.hort.iastate.edu/facultystaff/meetfaculty/taber/nitrate/index.php.

USDA. 2003. United States Department of Agriculture and National Agriculture Statistics Service. 1-2 (03) http://usda.mannlib.cornell.edu/reports/nassr/fruit/pvg-bban/vgan0103.pdf

Van Loon, L.C., Bakker, P.A.H.M., and Pieterse, C.M.J. 1998. Systemic resistance induced by rhizosphere bacteria. Annual Review of Phytopathology 36: 453-483.

Wilcox, G.E. 1993. Tomato. In Bennet, W.F. 1993. Nutrient deficiencies and toxicities in crop plants. APS press. St. Paul Minnesota. USA: 137-142.

Yun, L.T., Mizubuti, E. and Fry, W.E. 1999. Genetics of metalaxyl resistance in Phytophthora infestans. Fungal Genetics and Biology 26: 118-130

152

CHAPTER 4

Effect of a Plant Activator and Nitrogen on Tomato Root Colonization and Growth promotion by Bacillus spp. under Controlled Environmental Conditions

INTRODUCTION

Sustainable management of plant diseases requires the integration of cultural, genetic, biological, and chemical control tactics. Where no resistant varieties are available, cutural practices and biological control are important methods in plant disease management, since they reduce dependence on chemical approaches and are environmental friendly.

Among cultural practices, adjusting the fertilizer program particularly the nitrogen concentration has long been known to affect bacterial disease incidence and/or severity (Harkness and Marlatt, 1970). Nitrogen has been studied in relation to disease intensity in several crop-disease combinations (Bachelder et al. 1956;

Chase and Jones, 1986; Chase and Poole, 1987; Harkness and Marlatt, 1970;

Haygood et al. 1982; Huber and Watson, 1974; Jones et al. 1985; Nayudu and

Walker, 1961; Thomas, 1965). High rates of nitrogen induce a reduction (Chase and Poole, 1987; Harkness and Marlatt, 1970; Chase and Jones, 1986) or 153

increase (Bachelder et al. 1956; Nayudu and Walker, 1961; Thomas, 1965;

Haygood et al. 1982; Jones et al. 1985) in disease susceptibility, depending on the host and the pathogen.

In tomato, absortion of nitrogen is highest during the vegetative stage (185

mg/plant/day). Nitrogen concentration in plant tissue is relatively high during

seedling (4.38% dry weight) and vegetative stages (4.30% dry weight) and

decreases at flowering (3.99% dry weight) and fruiting (3.34% dry weight) stages.

Nitrogen concentrations in petiole sap of the tomato variety ‘Mountain Spring’ at

seedling, flowering, and fruiting stages was reported to be 180, 203, and 90 ppm

respectively (Taber, 1998).

PGPR are strains of bacteria that live in the rhizosphere, stimulate plant growth,

and improve stand under stress conditions (van Loon et al. 1998). PGPR that are

known to induce ISR include: Pseudomonas putida 89B-27, P. fluorescens 89B-

27, Serratia marcescens 90-166 (Liu et al. 1995; Raupach et al. 1996), B.

amyloliquefaciens strain IN937a, B. subtilis strain IN937b, B. pumilis strain SE34,

Kluyvera cryocrescens strain IN114 (Benhamou et al. 1998; Zehnder et al. 2000),

and Bacillus sp. isolates B and J (Braun-Kiewnick et al. 1998). Induction of

systemic resistance and enhacement of plant growth by Bacillus subtilis and B.

154

amyloliquefaciens have been reported on cabbage (Wulff et al. 2002), onion

(Reddy and Rahe, 1989a, 1989b), sugar beet (Braun-Kiewnick et al. 1998;

Collins et al. 2003), tomato variety ‘Solar Set’ (Reddy et al. 2000; Martinez-

Ochoa, 2000), tomato variety ‘Mountain Pride’ (Zehnder et al. 2000), tobacco,

and cucumber (Reddy et al. 2000). Bacillus spp. have the ability to fix nitrogen from the atmosphere (Chanway and Holl 1991; Jacobs et al. 1985; Li et al. 1992;

Shawky, 1983; Taiz and Zeiger, 2002). For instance, Grau and Wilson (1962) found that Bacillus polymyxa fixed up to 100 μg nitrogen / ml after only 200 hours of culture on nitrogen-free medium. In this process, the molecular nitrogen (N2), present in vast quantities in the atmosphere (78% by volume), is converted into ammonia (NH3) and then ammonium (NH4) or nitrate (NO3) by breaking the triple

covalent bond between two nitrogen atoms. Nitrate in the soil is acquired by the

+ - root via H NO3 symporter, translocated to the shoot and incorporated into

organic nitrogen compounds by reduction of nitrate to nitrite (nitrate reductase),

from nitrite to ammonium (nitrite reductase), from ammonium to glutamine and

glutamate (glutamine and glutamate synthetase) and transferred to organic

compounds (Taiz and Zeiger, 2002). It is not known whether nitrogen

concentration in the plant rhizosphere influences root colonization by PGPR.

155

Systemic Aquired Resistance (SAR) is characterized by the activation of plant

defense mechanisms that results in quantitative protection against a broad

spectrum of microorganisms (Sticher et al. 1997). It has been documented that

defense mechanisms can be elicited by inorganic (Dubos and Plomion, 2001;

Rakwal et al. 2003; Sivaguru et al. 2003) and organic compounds (Fellbrich et al.

2002; Chen et al. 2001). Acibenzolar-S-methyl (Actigard, Novartis, Greensboro,

North Carolina, USA) is a chemical that elicites SAR in plants (Ostendrop et al.

2001) and effectively reduces the intensity of several bacterial diseases (Anith et al. 2004; Briceno and Miller, 2004; Jones et al. 2004; Louws et al. 2001; Maxon-

Stein et al. 2002). This class of compounds have been called ‘plant activators’ due to the fact that deployment of these defense mechanisms involves activation of genes that encode for production of lignin, papillae, pathogenesis related proteins (β-1,3-glucanases, chitinases, osmotin, etc.), furanocoumarin, phytoalexins, and glycine rich proteins involved in plant defense against pathogens (Sticher et al. 1997).

In order to exhibit measurable effects PGPR must establish, colonize, and reach

population densities sufficient to ‘activate’ plant responses (Bloemberg and

Lugtenberg 2001; Benizri et al. 2001). Colonization of bacterized roots and

spatial localization of biocontrol agents has been monitored using

156

bioluminescence marker genes (Liu, et al. 1995; Meikle et al. 1995) and

rifampicin-resistant mutants (Martinez-Ochoa, 2000; Young et al. 1995). rep-PCR

with ERIC primers has been used recently to identify PGPR isolated from tomato

roots up to the strain level (Chapter 2). In that study, cross contamination by

different strains of Bacillus spp. applied to tomato roots reached 18 % in non-

inoculated roots. Interestingly, these PGPR were not detected as contaminants

on roots of Actigard-treated plants that were not PGPR-inoculated (Chapter 2). It

is possible that defense mechanisms activated by Actigard on tomato affected

Bacillus spp. population on roots.

The objectives of this study were to determine the combined effect of nitrogen

concentration and acibenzolar-S-methyl on tomato root colonization by three

PGPR, and to determine if Bacillus spp. improve nitrogen absorption as a possible mechanism of plant growth enhancement. It was hypothetized that the defense mechanism induced in tomato plants by acibenzolar-S-methyl will negatively affect Bacillus spp. population density on tomato roots, and that

Bacillus spp. fix and/or facilitate nitrogen absorption by tomato roots, enhancing plant growth.

157

MATERIALS AND METHODS

Experimental design

Treatments were arranged in a factorial design (acibenzolar-S-methyl, nitrogen,

PGPR) with three replications. One-hundred twenty-six tomato plants were randomly arranged on the bench in a controlled environment chamber. Nitrogen was provided at two concentrations (25 and 150 ppm). Actigard 50 WG sprayed at two different times (14 and 21 day-old plants) activated the mechanism of defense in tomato plants before and after inoculation with PGPR. Nontreated plants served as a control. Three strains of Bacillus spp. (MBI600, GBO3, and

IN937) were inoculated onto 14- and 21-day-old plants, or left non-inoculated

(water-treated control). The experiment was repeated once.

Germination and Growth of Tomato Plants

‘Mountain Spring’ (Siegers Seed Company, Rochester, NY, USA) seeds were germinated in plastic Petri dishes containing moist filter paper (Whatman 1,

Maidstone, England). Plates were incubated at 28 oC for 8 days and seedlings were planted in 10 cm square pots containing a square piece of paper (Kleenex,

Kimberly-Clark, Roswell, GA, USA) on the bottom, fine vermiculite (Therm-o-rock

East Inc, New Eagle, Pennsylvania, USA) (one seedling per pot) and a 9 x 9 cm

158

piece of square transparent plastic (polyethylene ULINE Waukegan, IL, USA) in the top to reduce evapotranspiration. The pots were placed in a growth chamber

(24/22 oC, day/night 40% HR, 14 h light).

Nutrient Solution, Irrigation System, and Soil Moisture

Begining at transplanting into vermiculite, plants were auto-irrigated to excess once a day with a complete nutrient solution containing one of two concentrations of nitrogen (Table 4.1). Nutrient solution was dispensed at a rate of 71.2 ml/day/pot to the top of the pot through a siphon delivery system as described in

Kleinhenz and Palta (2002) with the following modifications: siphons to each pot were established through 1.6 mm diameter, 3 m long tubing (R-3603, id 1.6 mm, od 3.2 mm, wall 0.8 mm, Tygon, Saint Gobain, Akron, OH, USA) and a variable hydrostatic head (Figure 4.1). The single line serving each pot was situated so that solution dripped freely into the vermiculite. Work surfaces were horizontal and exactly at the same vertical distance from the floor. Solutions were held in opaque reservoirs in the growth chamber. Solution delivery rate declined with reductions in the hydraulic head. Reservoirs containing nutrient solutions were refilled daily at 11:00-12:00 hrs. The effluent from each pot, always present, was collected for nitrogen analysis conducted using an Accumet 950 Ion meter

(Fisher Scientific, Pittsburg, PA, USA).

159

Figure 4.1. Irrigation system. A. General view. B. The single line serving each pot was situated so that solution dripped freely into the vermiculite. C. A piece of transparent plastic covered the surface of the growing medium to reduce evapotranspiration.

160

Element Concentration (ppm) Macronutrients N 26.88 149.56 P 17.44 17.44 K 101.72 101.68 Ca 163.90 147.25 Mg 36.80 36.80 S 28.53 28.55 Micronutrients Fe 4.64 4.64 Co 0.01 0.01 Cu 0.05 0.05 Mn 1.10 1.10 Mo 0.05 0.05 Zn 0.04 0.04 Na 1.91 1.91

Table 4.1. Composition of nutrient solutions used to fertilize tomato seedlings.

Soil samples were collected twice a day (before and after irrigation) from three pots at each concentration of nitrogen supplied to determine the amount of water per gram of soil based on gravimetry. The moisture in the soil was relatively constant and ranged from 3.86 to 4.07 and from 4.07 to 4.32 ml/g soil for the first and second experiment, respectively.

161

Bacillus spp. Culture and Plant Inoculation

Rifampicin-resistant strains of B. subtilis MBI600 (Microbio Bolder, CO), B.

subtilis GBO3 (Gustafson Plano TX), and B. amyloliquefaciens IN937 (Auburn

University AL) (Chapter 1) stored at -80oC were plated on nutrient agar amended with rifampicin (60mg/l) and incubated at 28 oC for 48 hours. They were streaked

onto endospore-forming medium and incubated at ambient temperature under

white lights (Sylvania 20W, warm light). Inoculum were prepared by suspending

12 day-old cultures in sterile distilled water. Optical density was adjusted to 0.6-

0.8 absorbance (600 nm). The number of colony forming units per milliliter for

each suspension was determined by plating 10-fold serial dilutions onto nutrient

agar (NA) medium. Endospore content (proportion) of each inoculum suspension

was quantified using the staining procedure described by Schaeffer and Fulton

(1933). One milliliter of bacterial inoculum suspension was drenched on 14- and

21-day old plants, according to the treatment (Table 4.2). Autoclaved distilled

water was used in controls.

162

Tomato Bacillus strain Absorbance Concentration Endospores (day-old) (600 nm) (log CFU ml-1) (Percent) 14 MBI600 0.816 / 0.865 7.9 / 8.5 100 / 100 GBO3 0.760 / 0.769 7.3 / 8.6 82 / 47 IN937 0.600 / 0.663 6.8 / 7.8 95 / 98

21 MBI600 0.795 / 0.870 7.7 / 7.8 98 / 98 GBO3 0.720 / 0.752 7.9 / 8.3 78 / 93 IN937 0.666 / 0.621 7.0 / 8.3 98 / 98

Table 4.2. Characteristics of Bacillus spp. inoculum suspensions drenched on 14 and 21 day-old tomato seedlings, variety ‘Mountain Spring’ (Experiment I / II).

Actigard Treatment

The systemic acquired resistance inducer acibenzolar-S-methyl (Actigard 50WG,

Syngenta Crop Protection, Greensboro, NC, USA) was applied until run off

(Experiment I) and glistening (Experiment II) at the recommended rate of 56 mg/l

(52 g / ha) onto 14 and 21 day-old plants using a hand-held sprayer. Control plants were sprayed with sterile distilled water.

Isolation and Identification of PGPR from Tomato Roots

Whole roots of 126 plants were excised, shaken to remove all but tightly adhered vermiculite and shaken (120 rpm) in potassium phosphate buffer (KPB 10 mM

K2HPO4, 10 mM KH2PO4 pH 7.4) for 10 minutes in a rotatory shaker (New

Brunswick Scientific Co, New Brunswick, NJ, USA) at room temperature. Twelve

163

ml washing buffer were recovered and centrifuged for 10 minutes at 16000 x g to

pellet the bacteria. The bacterial pellet was re-suspended in 1 ml KPB, and 10-

fold serial dilutions were made. One hundred μl of dilutions 10-3, 10-4, and 10-5 were spread onto nutrient agar amended with 60 mg/l rifampicin. Plates were incubated for 7 days at room temperature and the number of colonies that were morphologically similar to MBI600, GBO3 and IN937 in each treatment was recorded. One representative colony from each root sample was purified on nutrient agar. Purified colonies were transferred to 1.5 ml microcentrifuge tubes and washed once in 250 μl sterile distilled water. Pellets were suspended in 186

μl lysis buffer (20 mM TRIS; 2 mM EDTA; 1.2% Triton x-100; 20mg/l lysozyme) and incubated at 37 oC for 40 minutes. Genomic DNA was extracted using the

Qiagen DNAeasy tissue kit (Valencia, CA, USA) according to the manufacturer’s

instructions. A repetitive extragenic palindromic PCR (rep-PCR) assay was

carried out as described by Louws et al. (1996) with the following modifications:

0.8 mM each of the primers [ERIC1R (5’-ATGTAAGCTCCTGGGGATTCAC-3’);

ERIC2 (5’-AAGTAAGTGACTGGGGTGAGCG-3’)], 1.25 mM dNTP, 1.6 ul Taq

polymerase, and 1 μl genomic DNA (20-50 ng/μl). The polymerase chain reaction

was performed in a PTC-100 thermocycler (MJ Research, Waltham, MA, USA)

according to the program described by Louws et al. (1996). After amplification, 7

μl PCR product was mixed with 3 μl loading dye [5 mg bromophenol blue; 5 ml

164

5x TBE (0.45 M Tris-Borate; 0.01 M EDTA; pH 8.3); 2 g sucrose] were loaded into 1.5 % agarose gels in 0.5 x TBE and amplicons were separated by horizontal gel electrophoresis (Midicell e350, E-C apparatus corporation, St.

Petersburg, Florida, USA) at 50 V for 240 minutes at 10 oC. Amplification

products (stained in 2 μg ml-1 ethidium bromide for 15 minutes) were analyzed

under a UV light transilluminator. Images were photographed using an EDAS20

(Kodak, Rochester, N.Y., USA) system, gel size 13 x 17 cm with 1 - 3.5 seconds of exposure time. Population counts were adjusted based on ERIC-PCR identification. Colony counts for which the fingerprint of the representative isolate did not match the expected PGPR pattern were replaced with zero.

Contamination by any of the PGPR detected in control plants was counted.

Colony forming units per gram fresh weight of root were determined, a unit was added and data were transformed to the base 10 logarithm before statistical analysis.

Data Collection and Statistical Analysis

Plant height, stem diameter, number of leaves and nitrogen concentration in the

effluent were recorded at 7-day intervals starting on 14 day-old plants. The area

under the curve values for these variables were calculated using the trapezoid

method that consists of breaking up the curve into a series of rectangles,

165

calculating the area of each rectangle and adding the areas. Fresh weight, dry

weight and root length were recorded at the end of the experiments (49 day-old plants). The area under the curve of plant height, stem diameter, number of leaves and nitrogen concentration in the effluent were used for statistical analysis. Root length and fresh and dry weights were evaluated at the end of the

experiment. The General Linear Model was used to perform analysis of variance

(SAS 8.0) and least squares means were used to compare treatments and

interactions (proc glm; class nitro actigard bacillus rep; model auch aucd aucl

aucn tfw rl tdw = rep nitro|actigard|bacillus / ss3; lsmeans nitro|actigard|bacillus /

stderr pdiff; run;)

166

RESULTS

Tomato root colonization by Bacillus spp.

MBI600, GBO3, and IN937 were re-isolated from 94, 66 and 97% (Experiment I)

and 88, 33 and 85% (Experiment II) of tomato roots tested in this study (Table

4.3). Bacillus spp. population densities ranged from 101-105 CFU g-1 root (Table

4.4). There was no significant effect of seedling age (14 or 21 day-old) on

MBI600, GBO3 or IN937 final population densities. GBO3 population density

(101-103 CFU g-1 root) was significantly lower than that of MBI600 and IN937

(104-105 CFU g-1 root) in both experiments (Table 4.4). Up to 11% of the water-

treated seedlings were contaminated with MBI600 (Table 4.3) at a low population

density (100.5 CFU g-1 root, Table 4.4). Cross contamination reached a maximum

level of 5.4% of the roots tested (Table 4.3). Population density of MBI600,

GBO3 and IN937 was not significantly affected by nitrogen concentration (Table

4.5) or Actigard (Table 4.6) applied on 14 or 21 day-old plants and there were not

significant interactions among these factors. Table 4.7 shows the analysis of

variance of the effect of activation of systemic aquired resistance by acibenzolar-

S-methyl before or after PGPR inoculation on B. subtilis strain MBI600 and

GBO3 and B. amyloliquefaciens strains IN937 population densities. Activation of

167

systemic aquired resistance 7 days before, at the time of PGPR inoculation, or 7 days after PGPR inoculation did not affect final population densities of MBI600,

GBO3 or IN937 compared with the non-Actigard-treated control.

Treatment rep-PCR identification (percent in each category) MBI600 GBO3 IN937 Other No Colonies MBI600 94.44 2.77 0.00 2.77 0.00 Experiment I GBO3 2.77 66.66 2.77 0.00 27.77 IN937 0.00 0.00 97.22 2.77 0.00 Water 5.55 0.00 0.00 0.00 94.44

MBI600 88.57 0.00 0.00 0.00 11.42 Experiment II GBO3 0.00 33.33 2.77 0.00 63.88 IN937 0.00 0.00 85.29 0.00 14.70 Water 11.11 0.00 0.00 0.00 88.88

Table 4.3. Bacterial population composition in 49 day-old tomato roots inoculated with rifampicin-resistant Bacillus subtilis strains MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937. Colonies were isolated from roots on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on one colony per sample (Experiments I and II).

168

Treatment Applied on Population density seedlings Log (CFU g-1 root +1) Experiment I Experiment II p<0.0001 F 39.48 p<0.0001 F 17.19 MBI600 14 day-old 5.63 A 4.11 A MBI600 21 day-old 5.49 A 5.29 A GBO3 14 day-old 2.55 B 1.07 BC GBO3 21 day-old 3.02 B 2.00 B IN937 14 day-old 5.94 A 5.23 A IN937 21 day-old 5.78 A 4.56 A Water 0.18 C 0.50 C

Table 4.4. Analysis of variance of rifampicin-resistant Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 population densities on 49 day-old tomato roots. Colonies were isolated from roots on rifampicin- amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on one colony per sample. Levene’s p-value were 0.814 and 0.953 for Experiment I and II, respectively. Normality was assumed.

Nitrogen Population density (ppm) Log (CFU g-1 root +1) Experiment I Experiment II p 0.5238 F 0.41 p 0.9651 F 0.00 25 4.00 A 3.25 A 150 4.17 A 3.26 A

Table 4.5. Analysis of variance of the effect of nitrogen concentration on rifampicin-resistant Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 population densities on 49 day-old tomato roots. Colonies were isolated from roots of on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on one colony per sample. Levene’s p-value were 0.814 and 0.953 for Experiment I and II, respectively. Normality was assumed.

169

Treatment Applied on Population density seedlings Log (CFU g-1 root +1) Experiment I Experiment II p 0.2872 F 1.27 P 0.7110 F 0.34 Actigard 14 day-old 4.26 A 3.23 A Actigard 21 day-old 4.21 A 3.08 A Control 3.79 A 3.45 A

Table 4.6. Analysis of variance of the effect of acibenzolar-S-methyl on rifampicin-resistant Bacillus subtilis strain MBI600 and GBO3 and B. amyloliquefaciens strain IN937 population densities on 49 day-old tomato roots. Colonies were isolated from roots of on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on one colony per sample. Levene’s p-value were 0.814 and 0.953 for Experiment I and II, respectively. Normality was assumed.

170

Treatment Applied to PGPR Inoculated Population density seedlings on Log (CFU g-1 root +1) seedlings Experiment I Experiment II p 0.4048 F 1.06 p 0.3931 F 1.07 Actigard 14 day old MBI600 14 day-old 6.01 A 4.26 A Actigard 14 day old MBI600 21 day-old 5.98 A 5.29 A Actigard 21 day-old MBI600 14 day-old 5.12 A 3.44 A Actigard 21 day-old MBI600 21 day-old 5.72 A 5.50 A Control MBI600 14 day-old 5.75 A 4.62 A Control MBI600 21 day-old 4.77 A 5.10 A

Actigard 14 day old GBO3 14 day-old 2.74 A 0.59 A Actigard 14 day old GBO3 21 day-old 3.48 A 3.09 A Actigard 21 day-old GBO3 14 day-old 3.75 A 1.30 A Actigard 21 day-old GBO3 21 day-old 2.80 A 2.10 A Control GBO3 14 day-old 1.15 A 1.33 A Control GBO3 21 day-old 2.77 A 0.82 A

Actigard 14 day old IN937 14 day-old 5.82 A 4.67 A Actigard 14 day old IN937 21 day-old 5.22 A 4.76 A Actigard 21 day-old IN937 14 day-old 6.00 A 6.01 A Actigard 21 day-old IN937 21 day-old 6.06 A 3.19 A Control IN937 14 day-old 5.98 A 5.02 A Control IN937 21 day-old 6.07 A 5.74 A

Actigard 14 day old Control 0.55 A 0.00 A Actigard 21 day-old Control 0.00 A 0.00 A Control Control 0.00 A 1.52 A

Table 4.7. Analysis of variance of the effect of activation of systemic aquired resistance by acibenzolar-S-methyl before and after inoculation with rifampicin- resistant Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 on bacterial population densities on 49 day-old tomato roots. Colonies were isolated from roots of on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on one colony per sample. Levene’s p-values were 0.814 and 0.953 for Experiments I and II, respectively. Normality was assumed.

171

Nitrogen Absorption by PGPR- and Actigard-Treated Plants

Neither MBI600, GBO3, IN937 nor Actigard affected the concentration of nitrogen

in the liquid phase that was collected after irrigation (Tables 4.8 and 4.9).

Significant differences in the amount of nitrogen collected in the effluent were

detected at the two concentrations of nitrogen supplied. Nitrogen concentrations in the effluent were 1.3-11.6 and 18.8-120.0 ppm when 25 and 150 ppm nitrogen were supplied, respectively (Table 4.10). There were no significant interactions among PGPR and Actigard on the concentration of nitrogen in the effluent.

Treatment Applied on Mean nitrogen concentration in effluent seedlings (ppm) Experiment I Experiment II p 0.2236 F 1.40 p 0.2219 F 1.41 MBI600 14 day-old 9.19 A 68.17 A MBI600 21 day-old 8.17 A 66.26 A GBO3 14 day-old 8.90 A 68.52 A GBO3 21 day-old 10.23 A 64.26 A IN937 14 day-old 14.05 A 68.44 A IN937 21 day-old 9.43 A 64.71 A Water 10.59 A 63.54 A

Table 4.8. Analysis of variance of the effect of rifampicin-resistant Bacillus subtilis strain MBI600 and GBO3 and B. amyloliquefaciens strain IN937 on nitrogen absorption by tomato plants. Area under the curve of nitrogen concentration in the effluent was used for statistical analysis. Averages are shown. Levene’s p- value were 0.220 and 0.665 for Experiments I and II, respectively. Normality was assumed.

172

Treatment Applied on Mean nitrogen concentration in effluent seedlings (ppm) Experiment I Experiment II p 0.2548 F 1.39 P 0.4221 F 0.87 Actigard 14 day-old 9.97 A 65.34 A Actigard 21 day-old 9.12 A 65.75 A Control 11.14 A 67.72 A

Nitrogen supplied Mean nitrogen concentration in effluent (ppm) (ppm) Experiment I Experiment II p <0.0001 F 370.80 P<0.0001 F 2946.30 25 1.30 B 11.60 B 150 18.86 A 120.94 A

Table 4.10. Analysis of variance of the effect of nitrogen concentration supplied on nitrogen absorption by tomato plants. Area under the curve of nitrogen concentration in the effluent was used for statistical analysis. Averages are shown. Levene’s p-values were 0.220 and 0.665 for Experiments I and II, respectively. Normality was assumed.

173

Effect of PGPR, Actigard and Nitrogen Concentration on Tomato Biomass

Inoculation of tomato seedlings with MBI600, GBO3, and IN937 did not affect plant height, stem diameter, number of leaves, root length or total fresh and dry weight in this study (Table 4.11). Actigard significantly reduced plant height, stem diameter, number of leaves, total fresh and dry weight and root length compared with the untreated control in Experiment I but not in Experiment II (Table 4.12 ).

Tomato seedlings fertigated with 150 ppm nitrogen had greater plant height, stem diameter, fresh weight and dry weight and less root length than those fertigated with 25 ppm nitrogen (Table 4.13). There was a significant interaction between acibenzolar-S-methyl, age of the seedling during application, and nitrogen rate related to fresh and dry weights in Experiment I (Table 4.14). At the low rate of nitrogen, Actigard applied to 14 or 21 day-old seedlings did not reduce fresh or dry weight compared with untreated control. At the high rate of nitrogen,

Actigard applied to 14 or 21 day-old seedlings significantly reduced fresh or dry weight compared with untreated control. This reduction was greater when

Actigard was applied to 14 than for 21 day-old seedlings. A significant interaction between acibenzolar-S-methyl and Bacillus spp. also was detected in Experiment

I (Table 4.15). When Actigard was applied to 14 or 21 day-old plants, no increase in fresh or dry weight was observed on Bacillus-inoculated seedlings compared with the non PGPR-inoculated control. However, non-Actigard treated plants that

174

were inoculated with IN937 at 14 days and GBO3 at 21 days had a significant higher fresh weight than the non PGPR-inoculated, non-Actigard-treated control.

175

Treatment Applied on Biomass of 49 day old tomato plants seedlings Plant Stem Number Total Total dry Root height diameter of leaves fresh weight length (cm) (mm) weight (g) (cm) (g) MBI600 14 day-old 9.02 A 2.08 A 2.66 A 7.39 A 0.45 A 11.73 A MBI600 21 day-old 8.09 A 1.93 A 2.47 A 6.20 A 0.38 A 9.77 A Experiment I GBO3 14 day-old 9.34 A 2.06 A 2.89 A 7.54 A 0.46 A 10.44 A GBO3 21 day-old 9.03 A 2.13 A 2.62 A 8.74 A 0.54 A 11.66 A IN937 14 day-old 9.31 A 2.11 A 2.63 A 8.10 A 0.55 A 11.12 A IN937 21 day-old 10.28 A 2.16 A 2.70 A 7.31 A 0.44 A 11.05 A Water 9.34 A 2.16 A 2.81 A 7.24 A 0.45 A 10.05 A p-value 0.1865 0.4472 0.4656 0.6618 0.7201 0.8032 F-value 1.51 0.98 0.95 0.69 0.61 0.50

176 MBI600 14 day-old 5.46 A 1.80 A 1.74 A 5.81 A 0.46 A 6.49 A MBI600 21 day-old 6.66 A 1.83 A 1.77 A 5.56 A 0.45 A 8.10 A GBO3 14 day-old 6.86 A 1.96 A 1.94 A 6.23 A 0.49 A 6.65 A GBO3 21 day-old 6.14 A 1.77 A 1.66 A 6.45 A 0.47 A 7.48 A Experiment II IN937 14 day-old 5.88 A 1.79 A 1.58 A 6.53 A 0.51 A 7.85 A IN937 21 day-old 7.15 A 1.90 A 1.69 A 6.19 A 0.49 A 7.73 A Water 7.34 A 1.94 A 1.90 A 6.14 A 0.50 A 6.91 A p-value 0.7713 0.9586 0.8764 0.9990 0.9991 0.7841 F-value 0.55 0.25 0.40 0.06 0.06 0.53

Table 4.11. Analysis of variance of the effect of rifampicin-resistant Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 on biomass of 49 day-old tomato plants. Area under the curve of plant height, stem diameter and number of leaves was used for statistical analysis. Averages are shown. Levene’s p-values for plant height, stem diameter, number of leaves, total fresh weight, total dry weight and root length were 0.981, 0.864, 0.972, 0.414, 0.284, 0.901 (Experiment I) and 1.000, 0.961, 0.966, 0.988, 0.991, 0.982 (Experiment II), respectively. Normality was assumed.

Treatment Applied on Biomass of 49 day old tomato plants seedlings Plant Stem Number Total Total dry Root height diameter of leaves fresh weight length (cm) (mm) weight (g) (cm) (g) Experiment I Actigard 14 day-old 8.45 B 1.97 B 2.38 C 5.36 C 0.29 C 9.80 B Actigard 21 day-old 8.85 B 2.02 B 2.69 B 7.42 B 0.45 B 10.10 B Control 10.30 A 2.28 A 2.98 A 9.72 A 0.66 A 12.59 A p-value 0.0018 0.0002 <0.0001 <0.0001 <0.0001 0.0118

177 F-value 6.84 9.77 11.63 12.32 14.06 4.69

Experiment II Actigard 14 day-old 6.44 A 1.85 A 1.72 A 5.10 A 0.38 A 6.71 A Actigard 21 day-old 6.51 A 1.80 A 1.72 A 6.35 A 0.51 A 7.78 A Control 6.55 A 1.91 A 1.82 A 6.95 A 0.55 A 7.45 A p-value 0.8728 0.8682 0.6436 0.3423 0.1321 0.3896 F-value 0.14 0.14 0.44 1.09 2.08 0.95

Table 4.12. Analysis of variance of the effect of acibenzolar-S-methyl on biomass of 49 day-old tomato plants. Area under the curve of plant height, stem diameter and number of leaves was used for statistical analysis. Averages are shown. Levene’s p-values for plant height, stem diameter, number of leaves, total fresh weight, total dry weight and root length were 0.981, 0.864, 0.972, 0.414, 0.284, 0.901 (Experiment I) and 1.000, 0.961, 0.966, 0.988, 0.991, 0.982 (Experiment II), respectively. Normality was assumed.

Suplemented Biomass of 49 day old tomato plants nitrogen (ppm) Plant Stem Number Total Total dry Root height diameter of leaves fresh weight length (cm) (mm) weight (g) (cm) (g) Experiment I 150 9.44 A 2.28 A 2.64 A 9.80 A 0.64 A 8.80 B 25 8.96 A 1.90 B 2.73 A 5.20 B 0.30 B 12.86 A p-value 0.3286 <0.0001 0.4564 <0.0001 <0.0001 <0.0001

178 F-value 0.97 30.56 0.56 41.17 33.79 24.80

Experiment II 150 7.54 A 2.05 A 1.88 A 8.11 A 0.60 A 7.31 A 25 5.46 B 1.66 B 1.63 A 4.15 B 0.36 B 7.32 A p-value 0.0137 0.0084 0.1216 0.0003 0.0015 0.9849 F-value 6.38 7.32 2.45 14.40 10.86 0.00

Nitrogen Treatment Applied on seedlings Biomass of 49 day old tomato plants supplied (ppm) Total fresh weight (g) Total dry weight (g) Experiment I 25 Actigard 14 day-old 4.30 C 0.22 C 25 Actigard 21 day-old 5.31 C 0.29 C 25 Control 5.99 C 0.38 C 150 Actigard 14 day-old 6.42 C 0.36 C 150 Actigard 21 day-old 9.54 B 0.61 B 150 Control 13.45 A 0.94 A p-value 0.0119 0.0118 F-value 4.68 4.69

179 25 Actigard 14 day-old 3.51 A 0.30 A 25 Actigard 21 day-old 4.99 A 0.43 A Experiment II 25 Control 3.96 A 0.36 A 150 Actigard 14 day-old 6.69 A 0.46 A 150 Actigard 21 day-old 7.71 A 0.59 A 150 Control 9.92 A 0.36 A p-value 0.3915 0.3204 F-value 0.95 1.15

Table 4.14. Analysis of variance of the effect of the interaction of nitrogen concentration and acibenzolar-S-methyl on biomass of 49 day-old tomato plants. Averages are shown. Levene’s p-values for total fresh weight and total dry weight were 0.414, 0.284 (Experiment I) and 0.988, 0.991 (Experiment II), respectively. Normality was assumed.

’s p-values for total fresh

A A A A A A A A A A A A A A A A A A A A A l and plant growth-promoting 5.27 7.31 5.58 5.54 6.09 6.60 4.50 7.60 5.78 5.12 6.42 6.69 6.66 6.78 4.78 3.77 7.71 4.69 6.60 7.14 8.11 Experiment II p 0.9987 F 0.19 p 0.9987 F 0.19 I BCD

BCD CD BCD BCD CD CD CD BCD BCD BCD BCD BCD BCD BCD D BCD BC A AB A

8.13 6.69 6.90 5.48 5.89 5.38 7.74 7.25 5.36 3.78 4.23 5.86 8.27 6.24 8.39 7.55 8.49 7.63 Experiment 14.60 10.21 13.45 p 0.0378 F 1.97 p 0.0378 F 1.97 Fresh weight of 49 day old tomato plants weight Fresh ings. Averages are shown. Levene raction of acibenzolar-S-methy

on day-old day-old day-old day-old day-old seedlings Inoculated Inoculated 14 day-old day-old day-old 14 21 21 day-old 21 day-old 21 day-old 21 day-old 14 21 14 day-old 14 day-old 21 day-old 14 day-old day-old 14 14 day-old 14 day-old 14

21 day-old day-old 21 21 day-old day-old 21

14 day-old 14 day-old 21 day-old 21 day-old 21 day-old 14 day-old 14 day-old 14 day-old day-old 14 IN937 IN937 IN937 IN937 IN937 IN937

GBO3 GBO3 GBO3 GBO3 GBO3 GBO3 Control Control Control MBI600 MBI600 MBI600 MBI600 MBI600 MBI600 PGPR

on e of the effect inte Sprayed seedlings I) and 0.988 (Experiment II). Normality was assumed. 21 day-old 21 day-old 14 day old 14 day old 21 day-old 21 day-old 21 day-old day-old day-old 21 21 14 day old day-old 21 14 day old 14 day old 14 day old day-old 21

21 day-old 21 day-old 14 day old

Treatment Treatment Actigard Actigard Actigard Actigard Actigard Actigard Actigard Actigard Actigard Actigard Actigard Actigard Actigard Actigard Control Control Control Control Control Control Control

180 weight were 0.414 (Experiment rhizobacteria on total fresh weight of 49 day-old tomato seedl Table 4.15. Analysis of varianc DISCUSSION

In this study, colonization of tomato roots by rifampicin-resistant mutants of

MBI600, GBO3 and IN937 was verified using rep-PCR. Our data show that

Bacillus subtilis strain MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens

strain IN937 colonized roots of tomato variety ‘Mountain Spring’ resulting in

colonization percentages that ranged from 88-94%, 33-66% and 85-97% for

MBI600, GBO3 and IN937, respectively. Final population densities were not

affected by the seedling age at the time of inoculation (14 or 21 day old

seedling). Population densities of MBI600 (104-105 CFU g-1 root) or IN937 (104-

105 CFU g-1 root) were significantly higher that the population density of GBO3

(101-103 CFU g-1 root). The higher proportion of endospores in the inoculum

suspension of MBI600 and IN937 compared with those in GBO3 inoculum may

have facilitated colonization and establishment of MBI600 and IN937 populations

in tomato roots, in agreement with Young et al. (1995), who found that the final

populations of Bacillus cereus on wheat rhizosphere and non-planted soil were similar to the initial levels when inoculation was done using spores, whereas populations declined rapidly in the first 5 days when vegetative cells were used as source of inoculum. 181

Our data also show that the percentage of colonized roots and the population

density of these Bacillus strains was at similar levels to those reported by

McSpadden Gardner (2004) and should have been sufficient to enhance plant

growth if the mode of action of the PGPR is similar to that reported elsewhere

(van Loon et al. 1998; Liu and Sinclair, 1992).

Neither the concentration of nitrogen in the solution used to fertilize tomato

seedlings nor Actigard application (before and after Bacillus spp. inoculation)

affected the percentage colonization or population density of MBI600, GBO3 or

IN937 in tomato roots.

No significant differences in nitrogen absorption by tomato roots were detected in

Bacillus-treated plants. Our data suggest that the increase in biomass observed in Bacillus-treated plants is not the result of the increase in nitrogen absortion by

+ - tomato roots in a H NO3 symporter via measured as a reduction in the

concentration of nitrogen in the effluent at each concentration of nitrogen

supplied. Future studies should consider nitrogen measurements in plant tissue.

In contrast with previous studies that show that MBI600, GBO3, and IN937

increase plant height of tomato variety ‘Mountain Spring’ (Chapter 3), no

182

significant increase in biomass was observed in Bacillus-treated plants. However, a significant interaction between acibenzolar-S-methyl and Bacillus spp. was detected. When Actigard was sprayed on 14 or 21 day-old plants, no increase in fresh and dry weight was observed on Bacillus-inoculated seedlings compared with the non PGPR-inoculated control. When Actigard was not applied, tomato plants that were inoculated with IN937 and GBO3 had a significant increase in fresh weight compared with the non PGPR-inoculated control. These results agreed with previous observations (Reddy et al. 2000; Reddy and Rahe, 1989a,

1989b; Chapter 3) and suggest that the deployment of defense mechanism induced by Actigard may have a physiological cost measured as reduction of tomato biomass that cannot be compensate by PGPR.

When Actigard was applied until run off, a significant decrease in plant biomass was observed (Experiment I). In Experiment II, Actigard was applied until glistening and no significant decrease in plant biomass was observed. However, the evaluated biomass variables tended to decrease compared with control.

Ferguson et al. (2002) observed that Actigard application on ‘Mountain Fresh’ variety not only reduced the total yield but also the size of the fruit, when compared with control. Vavrina et al. (2004) reported that Actigard significantly

183

reduced root and shoot weight, shoot length, and stem diameter in tomato variety

‘Florida 47’.

184

CONCLUSIONS

• Inoculation by drench of PGPR on 14 or 21 day-old seedlings resulted in

similar Bacillus spp root population densities after 49 days.

• Neither nitrogen concentration in nutrient solution nor activation of defense

mechanims induced by Actigard before of after inoculation with Bacillus

spp. had a significant effect on population density of MBI600, GBO3 or

IN937 on tomato roots.

• Significant increase in fresh and weight was observed in Bacillus-

inoculated plants in one of the experiments.

• Bacillus spp. did not increase absorption of nitrogen by tomato roots.

• Run off aplication of 56 mg/l Actigard significantly reduced biomass of

‘Mountain Spring’ tomato compared to control plants.

185

LIST OF REFERENCES

Anith, K. N., Momol, M. T., Kloepper, J. W., Marois, J. J., Olson, S. M., and Jones, J. B. 2004. Efficacy of plant growth–promoting rhizobacteria, acibenzolar-S-methyl, and soil amendment for integrated management of bacterial wilt on tomato. Plant Disease. 88: 669-673.

Bachelder, S., Daines, R.H. and Bartley, C.E. 1956. Effect of nutrition on the incidence of bacterial leaf spot of peach. (Abstr.) Phytopathology 46: 240.

Benizri, E., Baudoin, E., and Guckert, A. 2001. Root colonization by inoculated plant growth promoting rhizobacteria. Biocontrol Sci. Technol. 11:557-574

Bloemberg, G. V., and Lugtenberg, B. J. J. 2001. Molecular basis of plant growth promotion and biocontrol by rhizobacteria. Current Opinion in Plant Biology 4: 343-350.

186

Chase, A.R. and Poole, R.T. 1987. Effects of fertilizer rate on severity of Xanthomonas leaf spot of schefflera and dwarf schefflera. Plant Disease 71: 527-529.

Chen, H-J., Hou, W.C. Kuc, J., and Lin, Y.H. 2001. Salicylic acid mediates alternative signal transduction pathways for pathogenesis-related acidic b- 1,3-glucanase (protein N) induction in tobacco cell suspension culture. J. Plant Physiol. 159: 331–337.

Dubos, C., and Plomion C. 2001. Drough differentially affects expression of a PR-10 proteon, in needles of maritime pine (Pinus pinaster Ait) seedlings. Journal of Experimental Botany. 52 (358): 1143 – 1144.

Fellbrich, G., Romanski, A., Varet, A., Blume, B., Brunner, F., Engelhardt, S., Felix, G.,Kemmerling, B., Krzymowska, M., and Nürnberger, T. 2002. NPP1, a Phytophthora-associated trigger of plant defense in parsley and Arabidopsis. The Plant Journal 32: 375–390.

Grau, F.H. and Wilson, P.W. 1962. Physiology of nitrogen fixation by Bacillus polymyxa. Journal of Bacteriology 83: 490-496.

Harkness, R.W., and Marlatt, R.B. 1970. Effect of nitrogen, phosphorus and potassium on growth and Xanthomonas disease of Philodendron oxycardium. J. Amer. Soc. Hort. Sci. 95 (1): 37-41.

187

Haygood, R.A., Strider, D.L., and Nelson, P.V. 1982. Influence of nitrogen and potassium on growth and bacterial leaf blight of Philodendron selloum. Plant Disease 66: 728-730.

Huber, D.M. and Watson, R.D. 1974. Nitrogen form and plant disease. Annual Review of Phytopathology 12: 139-165.

Jacobs, M.J., Funke, B.R., Merrill, S.D. 1985. Enumeration and characterization of nitrogen-fixing bacteria in a stripmine site. Plant and Soil 84(1): 3-10.

Jones J.B., Momol, M.T., Obradovic, A., Balogh, B., and Olson S.M. 2004. Bacterial spot management on tomatoes. In Momol M.T. and Jones J.B. 2004. 1st international Symposium on Tomato Diseases and 19th Annual Tomato Disease Workshop. University of Florida. Orlando, Florida, USA: 55

Kleinhenz M.D. and Palta J.P. 2002. Root zone calcium modulates the response of potato plants to heat stress. Physiologia Plantarum 115: 111-118.

Li, C.Y., Massicote, H.B., and Moore, L.V.H. 1992. Nitrogen-fixing Bacillus sp. associated with Douglas-fir tuberculate ectomycorrhizae. Plant and Soil 140(1): 35-40.

Liu, L., Kloepper, J.W., and Tuzun, S. 1995. Induction of systemic resistance in cucumber against Fusarium wilt by plant growth-promoting rhizobacteria. Phytopathology 85: 695-698.

Liu, Z.L. and Sinclair, J.B. 1992. Population dynamics of Bacillus megaterium strain B153-2-2 in the rhizosphere of soybean. Phytopathology 82: 1297- 1301.

Louws, F.J., Rademaker, J.L.W., and de Bruijn, F.J. 1996. REP-PCR: A rapid technique for bacterial identification and assessment of genetic diversity. Workshop APS annual meeting 16p.

188

Martinez-Ochoa, N. 2000. Biological control of the root-knot nematode with rhizobacteria and organic amendments. Ph.D. dissertation. Auburn University. Alabama: 120 p.

Maxon-Stein, K., He, S.Y., Hammerschmidt, R., and Jones, A.L. 2002. Effect of treating apple trees with acibenzolar-s-methyl on fire blight and expression of pathogenesis-related protein genes. Plant disease 86 (7): 785 – 790.

McSpadden Gardener, B. 2004. Ecology of Bacillus and Paenibacillus spp. in agricultural systems. Phytopathology 94: 1252-1258.

Meikle, A., Amin-Hanjani, S., Glover, L.N., Killham, K., and Prosser, J. 1995. Matric potential and the survival and activity of a Pseudomonas fluorescens in soil. Soil Biology and Biochemistry 27(7): 881-892.

Nayudu, M.V. and Walker, J.C. 1961. Tomato leaf composition in relation to bacterial leaf spot. Phytopathology 51: 368-372.

Oostendrop, M., Kunz, W., Dietrich, B., and Staub, T. 2001. Induced disease resistance in plants by chemicals. Eur. J. Plant Pathol.107: 19-28.

Rakwal, R, Agrawal, G.K., Kubo, A., Yonekura M., Tamogami, S., Saji, H., Iwahashi, H. 2003. Defense/stress responses elicited in rice seedlings exposed to the gaseous air pollutant sulfur dioxide. Environmental and Experimental Botany 49: 223-235.

Reddy, M.S., Ryu, C. M., Zhang, S., Yan, Z. and Kloepper, J. W. 2000. Aqueous formulations of plant growth-promoting rhizobacteria for control of foliar pathogens. http://www.ag.auburn.edu/~mreddy/.

Schaeffer, A.B., and Fulton, M.D. 1933. A simplified method of staining endospores. Science 77 (1990): 194

Shawky, B.T. 1983. Growth and nitrogen fixation by cutures of Bacillus sp. S 77g, Pseudomonas sp. S 44b, and Klebsiella sp. S145m, isolated from Egyptian soild. Zentralbl Mikrobiol 138(1): 9-15.

Sivaguru, M., Ezaki, B., He, Z-H., Tong, H., Osawa, H., Baluška, F., Volkmann, D., and Matsumoto, H. 2003. Aluminum-Induced Gene Expression and Protein Localization of a Cell Wall-Associated Receptor Kinase in Arabidopsis. Plant physiology. 132: 2256-2266.

Sticher L., Mauch-Mani B., and Métraux J.P. 1997. Systemic aquired resistance. Annual Review of Phytopathology 35:235-270.

Taber, H.G. 1998. Petiole sap nitrate sufficiency values for fresh market tomato production. http://www.hort.iastate.edu/facultystaff/meetfaculty/taber/nitrate/index.php.

Taiz, L., and Zeiger E. 2002. Plant Physiology. Sinauer Associates. Third edition. Sunderland Massachusetts, USA: 68-86.

Thomas, C.A. 1965. Effect of photoperiod and nitrogen on reaction of sesame to Pseudomonas sesami and Xanthomonas sesami. Plant Disease Reporter 49 (2): 119-120.

Van Loon, L.C., Bakker, P.A.H.M., and Pieterse, C.M.J. 1998. Systemic resistance induced by rhizosphere bacteria. Annual Review of Phytopathology 36: 453-483.

190

191

CHAPTER 5

Effect of Soil Moisture and Bacillus spp. on Bacterial Stem Rot

INTRODUCTION

Stem rot of tomato is a bacterial disease that is characterized by breakdown of

pith, vascular discoloration, external black lesions, eventual wilting and death of

the plant. Several bacteria have been attributed as causal agents of stem rot,

such as Erwinia carotovora (Speights et al. 1967), E. carotovora ssp. carotovora

(Dhanvantari and Dirks, 1987), Pseudomonas viridiflava (Lukezic et al. 1983), P.

corrugata (Scarlett et al. 1978) and P. cichorii (Wilkie and Dye, 1974).

Pseudomonas cichorii is a gram negative, aerobic, levan negative, oxidase positive, non-pectolytic, arginine dihydrolase negative bacterium that induces a

positive hypersensitivity reaction on tobacco. Colonies produce diffusable

fluorescent pigments that are visible on iron-deficient medium such as KB and

PF (Braun-Kiewnick and Sands, 2001). Pseudomonas cichorii induces elongated

dark lesions on the surface of tomato stems. Lesions may extend along petioles and leaves, causing dark green water-soaked blotches without haloes. Internally, 192

vascular tissues show dark brown discoloration. Infected stem pith is brown and

watery then disintegrates leaving a hollow stem. Dark brown spots appear on

fruits. Eventually the plant wilts and dies (Wilkie and Dye, 1974). The infection is

favored by extended leaf wetness and temperatures around 20 oC. Symptoms

appear after 48 hours under these conditions (Wilkie and Dye, 1974).

During a rainy year, an outbreak of stem rot (Pseudomonas cichorii) was

observed in an experimental plot of fresh market tomato variety ‘Mountain Spring’

in Wooster, Ohio. In this experiment tomato plants inoculated with a combination

of Bacillus spp. PGPR (GBO3+IN937) had higher bacterial stem rot severity than the control or those inoculated with Bacillus subtilis strain MBI600. Tomato plants grown on plastic residue mulch were more affected by stem rot than plants grown on plant residue mulch. Interestingly, tomato grown under plastic mulch had

significantly more nitrate in sap that those grown under plant residue mulch. Also,

GBO3+IN937-inoculated tomato had higher nitrate in sap compared to control

plants (Chapter 3).

The use of plant growth-promoting rhizobacteria (PGPR) represents a potentially

attractive approach to bacterial stem rot management in tomato, since currently

193

available bactericides are inefficient in managing this disease (Ustum, 2004) and alternative methods are needed.

Optimal conditions for plant growth-promoting rhizobacterial root colonization have been reported. For instance, the optimal temperature for root colonization by Pseudomonas fluorescens and P. putida is below 20 oC. Matric potentials between –0.3 and –0.7 bars were optimal for bacterial cell growth (Weller, 1988).

In field experiments, it was observed that during the flowering stage of tomato in a dry year, Bacillus spp. population densities of MBI600 and GBO3+IN937 tended to decrease at the low rate of irrigation (50% replacement of estimated crop evapotranspiration, Chapter 3).

Several mechanisms have been reported for PGPR to enhance plant growth such as fixation of nitrogen into a form that can be used by the plant (Bloemberg and Lugtenberg, 2001). Molecular nitrogen (N2) is present in vast quantities in the atmosphere (78% by volume). However, this reservoir is not directly available to living organisms. Nitrogen-fixing bacteria, including some Bacillus spp., convert molecular nitrogen (N2) into ammonia (NH3) and then ammonium (NH4) or nitrate (NO3) by breaking the triple covalent bond between two nitrogen atoms.

Nearly 171 million tons of nitrogen are fixed every year through this process

194

+ - worldwide. Nitrate in the soil is acquired by the root via H NO3 symporter, translocated to the shoot and incorporated into organic nitrogen compounds by reduction of nitrate to nitrite (nitrate reductase), from nitrite to ammonium (nitrite reductase), from ammonium to gutamine and glutamate (glutamine and glutamate synthetase) and transferred to organic compounds (Taiz and Zeiger,

2002).

Mycorrhizal fungi, nitrogen-fixing bacteria and plant growth-promoting rhizobacteria participate with roots in the acquisition of nutrients (Taiz and Zeiger,

2002). The role of Bacillus sp. and tuberculate ectomycorrhizae was studied in

Douglas fir by Li et al. (1992) who found an increase in nitrogenase activity in the external layer of the ectomycorrhizae, from where Bacillus spp. were isolated.

The authors conclude that the fungus and associated mycorrhizosphere microbes contribute to maintain a niche where nitrogen fixation by Bacillus spp. takes place.

In order to further investigate our field observations, the etiology of bacterial stem rot was verified. Two rates of irrigation were investigated to determine if they influenced Bacillus spp. population density on tomato roots and bacterial stem rot severity. It was also investigated if PGPR increased nitrogen concentration in

195

tomato leaves and if this increase of nitrate in tomato tissue would increase the

suceptibility of tomato to bacterial stem rot. It was hypothetized that a low rate of

irrigation would reduce Bacillus spp. population density in tomato roots, that tomato seedlings inoculated with GBO3+IN937 would have an increase in nitrate in the leaves compared with the control, and that this increase in nitrate would increase suceptibility of tomato to bacterial stem rot.

196

MATERIALS AND METHODS

Inoculation of Tomato Seedlings with PGPR

Inoculum of rifampicin-resistant mutants (Chapter 1) of Bacillus subtilis strain

MBI600, B. subtilis strain GBO3 and B. amyloliquefaciens strain IN937 was prepared by suspending 12 day-old cultures grown on endospore-forming medium in sterile distilled water. Optical density was adjusted to 0.6-0.8 absorbance (600 nm). The number of colony forming units per milliliter for each suspension (6.8-7.9 log CFU ml-1) was determined by plating 10-fold serial dilutions onto nutrient agar (NA) medium. Endospore content (proportion) of each inoculum suspension was quantified using the staining procedure described by

Schaeffer and Fulton (1933). Endospores constituted 94, 98, 90 and 100, 94,

79% of the inoculum suspension for experiments I and II of MBI600, IN937,

GBO3, respectively. Seeds of ‘Mountain Spring’ tomatoes (Siegers Seed Co.

Rochester, NY) were planted in 10 cm square pots containing a square piece of paper (Kleenex, Kimberly-Clark, Roswell, GA, USA) on the bottom, fine vermiculite (Therm-o-rock East Inc, New Eagle, Pennsylvania, USA) (one seedling per pot) and placed in a growth chamber (24/22 oC, day/night 40% HR,

14 h light). One milliliter of bacterial inoculum suspension was drenched on 14-

197

and 12-day old plants, according to the treatment. Autoclaved distilled water was

used in controls.

Bacterial Stem Rot Isolation from Diseased Tissue

Tomato stems from a 2003 field experiment (Chapter 3) that showed typical

vascular discoloration and pith necrosis were sterilized with 70% alcohol and split

open. Margins of discolored tissue were excised, suspended in KPB buffer

(10mM K2HPO4, 10mM KH2PO4 pH 7.4) and shaken for 20 minutes at 172 rpm.

The diffusate was diluted and plated onto King’s medium B. Twelve different colonies were isolated and purified.

Identification of the Causal Agent of Bacterial Stem Rot

Identification was based on fatty acid composition (Dickstein et al. 2001 ) and the

following additional tests were done according to the protocols described in

Schaad et al. (2001): hypersensitivity test, gram test, oxidative/fermentative

metabolism, production of fluorescent pigments, and oxidase.

Inoculation of Bacterial Stem Rot on Tomato Seedlings

Pseudomonas cichorii was plated on Pseudomonas agar F medium (Difco BD,

Sparks, MD, USA) and incubated at 28 oC for 24 hours. Inoculums of P. cichorii

198

were prepared by suspending 24 h cultures in sterile distilled water. Optical

density at 600 nanometers was adjusted to 0.2 absorbance (8.3 - 8.9 log CFU ml-

1, confirmed by 10-fold dilutions onto PF medium). Fifty-four day-old tomato

stems were punctured by a needle previously dipped into the bacterial

suspension. The point of inoculation was located at approximately half of the total

plant height. After inoculation, 100% relative humidity was provided by two

humidifiers (Larchmont WF HP 226, Walton, Moonachie, NJ) running 24 hours a

day up to the end of the experiment.

Nutrient solution

Macro and micro-nutrient solutions were prepared separately to prevent

precipitation (Chapter 2). The macronutrient solutions (25 and 150 ppm nitrogen)

were mixed (1:1). The resulting solution (175ppm) was diluted (1:1) with water

and used to fertilize tomato seedlings (Table 5.1).

199

Concentration Element (ppm) Macronutrients N 88.22 P 17.44 K 101.70 Ca 155.57 Mg 36.80 S 28.54 Micronutrients Fe 4.64 Co 0.01 Cu 0.05 Mn 1.10 Mo 0.05 Zn 0.04 Na 1.91

Table 5.1. Composition of nutrient solution used to fertilize tomato seedlings.

Irrigation Rate and Soil Moisture

Begining at seeding into vermiculite, plants were autoirrigated once a day with a complete nutrient solution (Table 5.1). Irrigation was provided at two levels: 50%

(32.5 ml/plant/day of 88 ppm nitrogen solution) and 100% (32.5 ml/plant/day of

88 ppm nitrogen solution plus 32.5 ml/plant/day water) estimated evapotranspiration (data not shown). Soil moisture was determined using gravimetric methods from three plants per level of irrigation. Nutrient solution was dispensed to the top of the pot through a siphon delivery system as described in 200

Kleinhenz and Palta (2002) with the following modifications: siphons to each pot

were established through 1.6 mm diameter 3 m long tubing (R-3603, id 1.6 mm,

od 3.2, wall 0.8 mm, Tygon, Saint Gobain, Akron, OH, USA) and variable

hydrostatic head. The single line serving each pot was situated so that solution

dripped freely into the vermiculite. Work surfaces were horizontal and exactly at

the same vertical distance from the floor. Solution was held in opaque reservoirs

in the growth chamber. Solution delivery rate declined with reductions in the

hydraulic head. Reservoirs containing nutrient solutions were refilled daily at

11:00-12:00 hrs.

Nitrogen Concentration in Tomato Leaves

The basal leaf from each replicate was taken every 12 days beginning with 24

day-old plants. Basal leaves were collected between 11:00 and 13:00 hr. The

sample was weighed and placed in a plastic bag. Samples were stored (4 oC) up to 72 hours prior to nitrate analysis. One milliliter water (adjusted with nitrate ionic strength adjustor: 26 g ammonium sulfate; 74 ml water; add 1 for every 50 ml water) was added and the sample was ground. Nitrogen concentration was measured from the suspension using a specific ion electrode (Accumet 950 Ion meter, Fisher Scientific, Pittsburg, PA, USA).

201

Tomato Root Colonization by PGPR

Root colonization by Bacillus spp. was evaluated in 60 day-old seedlings. Whole

roots from all the plants in the experiment were excised and shaken to remove all

but tightly adhered vermiculite particles. Samples were shaken in 20 ml

potassium phosphate washing buffer (KPB 10 mM K2HPO4, 10 mM KH2PO4 pH

7.4) for 10 minutes on a rotary shaker (New Brunswick Scientific Co, New

Brunswick, NJ, USA) at approximately 120 rpm. Twelve ml washing buffer was recovered and centrifuged for 10 minutes at 16000 x g. The bacterial pellet was re-suspended in 1 ml KPB buffer, and 10-fold serial dilutions were made. One hundred μl of dilutions 10-3, 10-4, and 10-5 were spread onto nutrient agar

amended with 60 mg l-1 rifampicin (two plates per dilution). Plates were incubated

for 7 days at room temperature (24 oC) under 8 hours white light (Sylvania 20W,

warm light) and the number of colonies that were morphologically similar to

MBI600, GBO3 and IN937 in each treatment was recorded. One representative

colony from each sample was purified on nutrient agar. Purified colonies were

transferred to a 1.5 ml microcentrifuge tube and washed once in 250 μl sterile

distilled water. Pellets were suspended in 186 μl lysis buffer (20 mM TRIS; 2 mM

EDTA; 1.2% Triton x-100; 20mg/l lysozyme) and incubated at 37 oC for 40

minutes. Genomic DNA was extracted using the Qiagen DNAeasy tissue kit

(Valencia, CA. USA) according to the manufacturer’s instructions. A repetitive

202

extragenic palindromic PCR (rep-PCR) assay was carried out as described by

Louws et al. (1996) with the following modifications: 0.8 mM each of the primers

[ERIC1R (5’-ATGTAAGCTCCTGGGGATTCAC-3’); ERIC2 (5’-

AAGTAAGTGACTGGGGTGAGCG-3’)], 1.25 mM dNTP, 1.6 ul Taq polymerase,

and 1 μl genomic DNA (20-50 ng/μl). The polymerase chain reaction was

performed in a PTC-100 thermocycler (MJ Research, Waltham, MA, USA)

according to the program described by Louws et al. (1996). After amplification, 7

μl PCR product was mixed with 3 μl loading dye [5 mg bromophenol blue; 5 ml

Petersburg, Florida, USA) at 50 V for 240 minutes at 10 oC. Amplification

products (stained in 2 μg ml-1 ethidium bromide for 15 minutes) were analyzed

under a UV light transilluminator. Images were photographed using an EDAS290

Contamination by any of the PGPR detected in control-plants was counted.

Colony forming units per gram fresh weight of root were determined, a unit was

203

added and data were transformed to the base 10 logarithm before statistical

analysis.

Experimental Design

Treatments were arranged in a factorial design (irrigation, pathogen, PGPR) with

five replications. Sixty tomato plants were randomly located on the work

surfaces. Irrigation was provided at two levels: 50% and 100% replacement of

estimated evapotraspiration. The pathogen (Pseudomonas cichorii) was

inoculated on 54-day old tomato plants by puncturing the stems with a needle

previously dipped into the bacterial suspension. Autoclaved distilled water was

used on control plants. Three strains of Bacillus spp. were grouped in two

treatments (MBI600 and GBO3+IN937). Water was used in control plants. The experiment was repeated once.

Data Collection and Statistical Analysis

Plant height, nitrogen concentration in leaves, and soil moisture were evaluated

at 12-day intervals. The area under the curve of these variables was calculated using the trapezoid method that consists of breaking up the curve into a series of rectangles, calculating the area of each rectangle and adding the areas together.

Root length and length of the lesion induced by Pseudomonas cichorii was

204

evaluated at the end of the experiment (60 day-old plants). The tomato stems were split open. Vascular discoloration was measured using a ruler. The area under the curve of plant height, nitrogen concentration in leaves, and soil moisture and the length of the lesion were used for statistical analysis. The

General Linear Model was used in factorial experimental design with three factors to analyze the variance (proc glm; class moisture pgpr pathogen; model aucph aucn lesion rl logCFUg1 = moisture|pgpr|pathogen / ss3; lsmeans moisture|pgpr|pathogen / stderr pdiff; run; SAS 8.0). Least square means were used to compare treatments and interactions.

205

RESULTS

Tomato Root Colonization by Bacillus spp.

MBI600 and GBO3+IN937 were re-isolated from 90 and 90% (Experiment I) and

60 and 85% (Experiment II) of tomato roots, respectively (Table 5.2). Bacillus

spp. population densities ranged from 105.0-105.2 CFU g-1 root in Experiment I and

from 102.9-104.4 CFU g-1 root in Experiment II (Table 5.3). There were no

significant differences in the population densities of MBI600 compared with those

of GBO3+IN937 in Experiment I, but the population density of GBO3+IN937 was

significantly higher than that MBI600 in Experiment II (Table 5.3). Five percent of

the water-treated seedlings were contaminated with MBI600 at a population

density of 100.2 CFU g-1 root in Experiment I (Tables 5.2 and 5.3). Five percent of

the water-treated seedlings were contamined with IN937 at a density of 100.2

CFU g-1 in Experiment II (Tables 5.2 and 5.3). No cross contamination among

PGPR-inoculated plants was detected in either experiment (Table 5.2). The

average soil moisture was 2.00 and 2.01 ml/g soil at 50% and 100% replacement

estimated evapotranspiration for Experiment I. Average moisture in soil was 3.21

and 3.80 ml/g soil at 50% and 100% replacement estimated evapotranspiration

for Experiment II. No differences in the average soil moisture were detected

between 50% or 100% replacement of estimated crop evapotraspiration rates.

206

Population density of MBI600 and GBO3+IN937 in tomato roots was not

significantly affected by irrigation rate.

Treatment Percent of plants colonized by MBI600 GBO3 IN937 Other None MBI600 90.00 0.00 0.00 0.00 10.00 Experiment I GBO3+IN937 0.00 0.00 90.00 10.00 0.00 Control 5.00 0 .00 0.00 10.00 85.00

MBI600 60.00 0.00 0.00 0.00 40.00 Experiment II GBO3+IN937 0.00 0.00 85.00 0.00 15.00 Control 0.00 0 .00 5.00 0.00 95.00

Treatment Population density Log (CFU g-1 root +1) Experiment I Experiment II MBI600 5.08 A 2.93 B GBO3+IN937 5.23 A 4.48 A Control 0.26 B 0.21 C

Table 5.3. Analysis of variance of rifampicin-resistant Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 population densities (log CFU g-1 root +1) on 60 day-old tomato roots. Colonies were isolated from roots of on rifampicin-amended medium. Identification was based on colony morphology and rep-PCR with ERIC primers carried out on one colony per sample. Levene’s p-values were 0.635 and 0.280 for Experiment I and II, respectively. Normality was assumed.

207

Effect of Irrigation Rate and PGPR on Tomato Plant Height and Root Length

Inoculation of tomato seedlings with MBI600, GBO3, and IN937 did not affect

plant height or root length in either experiment (Table 5.4). The area under the

curve of plant height was used for statistical analyisis. Plant height and root

length averages are presented. The average plant height of plants treated with

MBI600-, GBO3+IN937- and water- ranged from 5-8, 5-8, 5-9 cm, respectively.

No significant differences in plant height were observed in Bacillus-treated plants

compared with control. Plants irrigated with 50 and 100% estimated crop

evapotranspiration were 4–8 and 5-9 cm tall, respectively. No significant

differences in plant height were observed between 50 and 100% estimated crop

evapotranspiration irrigation rate in either experiment (Table 5.5). Plants irrigated

with 50 and 100% estimated crop evapotranspiration had 10–11 and 12-13 cm

roots, respectively. No significant differences in root size were observed between

50 and 100% estimated crop evapotranspiration irrigation rate in Experiment I but

roots of plants irrigated at 100% rate were significantly bigger than the roots of plants irrigated at the 50% rate in Experiment II (Table 5.5).

208

Treatment Plant height Root length Mean nitrogen Lesion length concentration in leaves MBI600 8.51 A 11.88 A 138.21 A 3.16 A Experiment I GBO3+IN937 8.98 A 12.23 A 151.30 A 3.52 A Control 9.26 A 11.74 A 159.97 A 3.07 A p value 0.5466 0.8852 0.8658 0.8610 F value 0.61 0.12 0.14 0.15

MBI600 5.07 A 10.76 A 108.99 B 2.35 A Experiment II GBO3+IN937 5.00 A 13.20 A 113.85 B 2.66 A Control 5.71 A 10.69 A 181.27 A 2.29 A p value 0.0939 0.1311 0.0124 0.6785 F value 2.49 2.12 4.82 0.39

Table 5.4. Analysis of variance of the effect of rifampicin-resistant Bacillus subtilis strains MBI600 and GBO3 and B. amyloliquefaciens strain IN937 on tomato biomass and bacterial stem rot severity. Area under the curve of plant height and nitrogen concentration was used for statistical analysis. Averages are shown. Levene’s p-values for plant height, root length, nitrogen in sap and lesion size were 0.667, 0.388, 0.775, 0.023 (Experiment I) and 0.648, 0.471, 0.838, 0.045 (Experiment II), respectively. Normality was assumed.

209

Irrigation Plant height Root length Nitrogen Lesion length rate concentration in leaves 50% 8.72 A 11.10 A 168.42 A 3.52 A Experiment I 100% 9.12 A 12.79 A 131.23 A 2.98 A P value 0.4634 0.0505 0.1387 0.4450 F value 0.55 4.02 2.27 0.59

50% 4.98 A 10.03 B 136.88 A 2.54 A Experiment II 100% 5.53 A 13.07 A 132.53 A 2.32 A P value 0.0601 0.0101 0.7329 0.5549 F value 3.71 7.18 0.12 0.35

Nitrogen Concentration in Tomato Leaves

The average nitrogen concentration in plants that were inoculated with MBI600 or

GBO3+IN937 or the water control were 108-138, 113-151, 159-181 ppm, respectively. No significant differences in nitrogen concentration were observed between Bacillus-inoculated plants and control plants in Experiment I but significantly lower concentrations of leaf nitrogen were detected in Bacillus- inoculated plants compared with control non-PGPR-inoculated plants in

Experiment II (Table 5.4). Nitrogen in leaves ranged from 136-168 and 131-132 ppm for 50 and 100% replacement of estimated evapotranspiration, respectively.

210

No differences of nitrogen concentration in leaves was observed in plants

irrigated with the 50 or 100% rate.

Bacterial Stem Rot Identification

Naturally infected tomato plants in the field showed dark brown lesions on stems,

pith necrosis, and eventual plant wilting and death of the plant (Chapter 3).

Isolations were made on King’s medium B from 12 tomato stems and petioles

collected at random from the experimental plot yielding several types of bacteria,

that were predominatly flat, white colonies, producing green fluorescent pigments

diffusing into the medium. When inoculated onto ‘Big Beef’ and ‘Mountain Spring’ tomato varieties, the bacteria produced dicoloration of vascular tissue and brown lesions on tomato stems in about 4 days. The bacteria were gram negative, oxidase positive, oxidized glucose rapidly, did not ferment glucose, and was positive for the hypersensitivity reaction in tobaco. Fatty acids composition

matched for Pseudomonas cichorii (similarity value 0.897).

211

A B

Figure 5.1. Bacterial stem rot (Pseudomonas cichorii). A. Brown lesions on tomato variety ‘Mountain Spring’; B. Lesion on tomato variety ‘Big Beef’.

Effect of PGPR and Irrigation Rate on Lesion Length

Discoloration of vascular tissue spread in both directions, but was more evident

in the upper part of the stems. Plants that were treated with MBI600,

GBO3+IN937 and water had mean lesion lengths of 3.16, 3.52, and 3.07 cm in

Experiment I and 2.35, 2.66, and 2.29 cm in Experiment II, respectively. No

significant differences in lesion length were observed in Bacillus treated plants when compared with the control (Table 5.4). Plants irrigated with 50 and 100%

estimated crop evapotranspiration had lesion lenghts of 2.54-3.52 and 2.32-2.98

cm, respectively. No significant differences in lesion length were observed in 212

tomato plants irrigated with 50 or 100% estimated crop evapotranspiration irrigation in either experiment.

213

DISCUSSION

In this study, colonization of tomato roots by rifampicin-resistant strains of

Bacillus subtilis MBI600, B. subtilis GBO3 and B. amyloliquefaciens IN937 was

verified using rep-PCR and ERIC primers. It was hypothetized that a low rate of

irrigation would result in a low soil moisture and consequently in a reduction of

Bacillus spp. population density in tomato roots. However, the soil moisture at the

low rate of irrigation was not significantly different than the moisture at the high

rate of irrigation (data not shown). In our experiments, the population density of

MBI600 and GBO3+IN937 in tomato roots was not significantly affected by either

50% or 100% irrigation rate.

The population densities of MBI600 and GBO3+IN937 (6.8–7.9 log CFU ml-1)

applied onto 12 day-old seed and recovered after 60 days (2.9-5.2 log CFU g-1 root) should have been sufficient to trigger plant growth enhancement and beneficial effects in tomato if the mode of action of these plant growth-promoting rhizobacteria is similar to those reported elsewhere (van Loon et al. 1998; Liu and Sinclair, 1992). However, the height of plants that were inoculated with

MBI600 or GBO3+IN937 was not significantly different from that of water-treated

214

controls. Neither irrigation rate nor inoculation with P.cichorii affected tomato plant height.

One of the mechanisms that has been reported to enhance plant growth induced

by PGPR is fixation of nitrogen into a form that can be used by the plant

(Bloemberg and Lugtenberg, 2001). In our experiments, it was considered that

fixation of nitrogen and/or facilitation of nitrogen compound acquirement could

result in an increase of nitrogen in plant tissue and consequently in an increase

in growth. In order to evaluate the result of these alternatives, nitrogen

concentration in leaves was monitored every 12 days. Our results showed that

inoculation of tomato seedlings with Bacillus did not result in an increase of

nitrogen concentration in leaves compared to control. On the contrary, lower

concentrations of nitrogen in leaves were detected in Bacillus-inoculated plants in

one of the experiments and no effect on plant height was observed.

It was established that P. cichorii was the causal agent of bacterial stem rot of tomato in Wooster, Ohio in summer 2003. The pathogen induced typical symptoms on tomato varieties ‘Big Beef’ and ‘Mountain Spring’ within 4 days after inoculation. During field experiments, GBO3+IN937 inoculated plants had higher severity of bacterial stem rot when compared to control plants (Chapter 3).

215

Data of this experiment show no significant differences in lesion length on stems of tomato inoculated with Bacillus- compared with non PGPR-inoculated control.

However, lesions in GBO3+IN937 inoculated plants tended to be longer than in

MBI600- or water-control inoculated plants in both experiments. This result is in agreement with field observations (Chapter 3).

216

CONCLUSIONS

• Bacillus spp. population density was not affected by irrigation rate. The

two irrigation rates resulted in similar soil moisture.

• MBI600, GBO3+IN937 did not increase nitrogen concentration in tomato

leaves.

• Tomato seedlings inoculated with GBO3+IN937 tended to have longer

lesions than seedling inoculated with MBI600 or water control in both

experiments, although the difference was not statistically significant.

217

LIST OF REFERENCES

Bloemberg, G. V., and Lugtenberg, B. J. J. 2001. Molecular basis of plant growth promotion and biocontrol by rhizobacteria. Current Opinion in Plant Biology 4: 343-350.

Dhanvantari, B.N. and Dirks, V.A. 1987. Bacterial stem rot of greenhouse tomato: Etiology, spatial distribution, and the effect of high humidity. Phytopathology 77: 1457 – 1463.

Kleinhenz M.D. and Palta J.P. 2002. Root zone calcium modulates the response of potato plants to heat stress. Physiologia Plantarum 115: 111-118.

Li, C.Y., Massicote, H.B., and Moore, L.V.H. 1992. Nitrogen-fixing Bacillus sp. associated with Douglas-fir tuberculate ectomycorrhizae. Plant and Soil 140(1): 35-40.

Liu, Z.L. and Sinclair, J.B. 1992. Population dynamics of Bacillus megaterium strain B153-2-2 in the rhizosphere of soybean. Phytopathology 82: 1297- 1301.

Louws, F.J., Rademaker, J.L.W., and de Bruijn, F.J. 1996. REP-PCR: A rapid technique for bacterial identification and assessment of genetic diversity. Workshop APS annual meeting 16p.

Lukezik, F.L., Levine, R.G. and MacNab, A.A. 1983. Pseudomonas viridiflava associated with stem necrosis of mature tomato plants (Abst) Phytopathology 73: 370. 218

Scarlett, C.M., Fletcher, J.T, Roberts, P. and Lelliott, R.A. 1978. Tomato pith necrosis caused by Pseudomonas corrugata n.sp. Ann. Appl. Biol. 88:105- 114.

Schaad, N.W., Jones, J.B., and Chun, W. 2001. Laboratory guide for identification of plant pathogenic bacteria. Third edition. APS press. USA: 373.

Schaeffer, A.B., and Fulton, M.D. 1933. A simplified method of staining endospores. Science 77 (1990): 194.

Speights, D.E., Hallivell, R.S., Horne, C.W. and Hughes, A.B. 1967. A bacterial stem rot of greenhouse tomato plants. Phytopathology 57: 902-904.

Taiz, L., and Zeiger E. 2002. Plant Physiology. Sinauer Associates. Third edition. Sunderland Massachusetts, USA: 68-86.

Van Loon, L.C., Bakker, P.A.H.M., and Pieterse, C.M.J. 1998. Systemic resistance induced by rhizosphere bacteria. Annual Review of Phytopathology 36: 453-483.

Weller, D.M. 1988, Biological control of soilborne plant pathogens in the rhizosphere with bacteria. Annual Review of Phytopathology 26 379-407.

Wilkie, J.P. and Dye, D.W. 1974. Pseudomonas cichorii causing tomato and celery diseases in New Zealand. N.Z. Journal of Aricultural Research 17: 123 – 130.

219

CONCLUSIONS

Rifampicin-resistant strains of three biological control agents (B. subtilis MI600,

B. subtilis GBO3 and B. amyloliquefaciens IN937) were characterized based on molecular, physiological and morphological characteristics. Selection for rifampicin resistance did not alter colony morphology, endospore proportion, or

ERIC fingerprint pattern of B. subtilis strain MBI600 or B. amyloliquefaciens strain

IN937. The rifampicin-resistant mutant of B. subtilis strain GBO3 had the same colony morphology and ERIC-fingerprint pattern as the wild-type, but produced significantly fewer endospores than MBI600, IN937 or GBO3 wild-type. Colony morphology and fingerprint pattern were used to identify colonies that were reisolated from tomato roots, using rifampicin-amended medium. Of three strains evaluated, GBO3 had the lowest levels of root colonization and population density, which may have been related to its significantly lower endospore production.

220

Although metabolically active colonies of Bacillus spp. were reisolated from dip- inoculated tomato roots, no significant increase in plant height was observed in

MBI600-, GBO3- or IN937- inoculated plants compared with control plants under greenhouse conditions. However, when seedlings were drench inoculated, plants treated with MBI600 and GBO3+IN937 were significantly taller than the nontreated control plants. Plant growth enhancement by PGPR was a variable phenomenon that may be dependent on inoculation method, population density and environmental conditions.

Plastic mulch significantly affected the uptake of nitrogen by tomato plants in field experiments. Plants that were grown under plastic mulch had higher concentrations of nitrate in sap than those grown under rye residue mulch. This increase in nitrate likely resulted in an increase of susceptibility of tomato to fungal and bacterial diseases. Further, it is possible that inoculum dispersal may have been reduced in rye mulch compared to dispersal in plastic mulch.

Similarly, GBO3+IN937- inoculated tomato plants had significantly higher levels of nitrate in their sap than MBI600- or control plants and were more susceptible to bacterial stem rot. Further research showed that GBO3+IN937 inoculated plants did not increase nitrogen concentration in leaf tissue but tended to develop larger bacterial stem rot lesions than MBI600- or water- treated plants.

221

Activation of SAR in four varieties of tomato was achieved by application of

Actigard. SAR significantly reduced bacterial leaf spot on tomato plants fertilized

with 150 and 25 ppm nitrogen. Under controlled environmental conditions

application of Actigard (56 ml/l) until run off significantly reduced biomass of

tomato plants and this reduction was not compensated by 150 ppm nitrogen

supplied as fertilizer. Actigard application appeared to inhibit PGPR colonization

when population densities of Bacillus spp. on tomato roots were low. However,

Actigard did not affect final population densities of MBI600, GBO3 or IN937 on tomato roots when Bacillus spp. inoculum contained log 7 CFU/ml.

The use of MBI600, GBO3 or IN937 in an intensive tomato management

program should consider that while plant growth may increase, susceptibility to

diseases, particularly bacterial stem rot may also increase. Integration of a

fertilization program that includes a low rate of nitrogen, application of

acibenzolar-S-methyl, and use of plant residue mulch and a forecasted fungicide

spray program should be considered to optimize management of foliar and

bacterial diseases that limit tomato production in Ohio.

222

LIST OF REFERENCES

Anonymous. Siegers 2000. Hybrid round red tomato. http://www.siegers.com/shop/kind.asp?kind_id=BBB.

Bachelder, S., Daines, R.H. and Bartley, C.E. 1956. Effect of nutrition on the incidence of bacterial leaf spot of peach. (Abstr.) Phytopathology 46: 240.

Barnett, H.L. and Hunter, B.B. 1998. Illustrated genera of imperfect fungi. APS press. Fourth edition. St. Paul, Minnesota USA: 218 p.

Bell, C.E., Masiunas, J.B., and Stall, W.M. 2000. My view. Weed Science 48: 1.

Benhamou, N., Kloepper, J.W., and Tuzun, S. 1998. Induction of resistance against Fusarium wilt of tomato by combination of chitosan with an

223

endophytic bacterial strain: ultrastructure and cytochemistry of the host response. Planta 204: 153-168.

Benizri, E., Baudoin, E., and Guckert, A. 2001. Root colonization by inoculated plant growth promoting rhizobacteria. Biocontrol Sci. Technol. 11:557-574.

Benson, M. 1998. Integrated biocontrol strategies for disease control on greenhouse ornamentals using poinsettia as a model system. American Floral Endowmen. http://www.endowment.org/pr/p4998pr1.htm.

Bloemberg, G. V., and Lugtenberg, B. J. J. 2001. Molecular basis of plant growth promotion and biocontrol by rhizobacteria. Current Opinion in Plant Biology 4: 343-350.

Bloom, A.J., Jackson L.E., and Smart, D.R. 1993. Root growth as a function of ammonium and nitrate in the root zone. Plant Cell and Environment 16: 199 – 206.

224

Butterworth, J. and McCartney, H.A. 1991. The dispersal of bacteria from leaf surfaces by water splash. Journal of Applied Bacteriology 71: 484-496.

Chanway, C.P. Holl, F.B. 1991. Biomass increase and associative nitrogen- fixation of mycorrhizal pinus-contorta seedlings inoculated with a plant-

225

growth promoting Bacillus strain. Canadian Journal of Botany-Revue Canadienne de Botanique 69(3): 507-511.

Chase, A.R. and Poole, R.T. 1987. Effects of fertilizer rate on severity of Xanthomonas leaf spot of schefflera and dwarf schefflera. Plant Disease 71: 527-529.

Day, J.P., Wattier, R.A.M., Shaw, D.S. and Shattock, R.C. 2004. Phenotypic and genotypic diversity in Phytophthora infestans on potato in Great Britain, 1995-98. Plant Pathology 53: 303-315.

Dhanvantari, B.N. and Dirks, V.A. 1987. Bacterial stem rot of greenhouse tomato: Etiology, spatial distribution, and the effect of high humidity. Phytopathology 77: 1457-1463.

Dickstein, E.R., Jones J.B. and Stead, D.E. 2001. Automated tecniques. In Schaad, N.W., Jones, J.B., and Chun, W. 2001. Laboratory guide for identification of plant pathogenic bacteria. Third edition. APS press. USA: 343-358. Driks, A. and Setlow, P. 2000. Morphogenesis and properties of the bacterial spore. In Brun, Y.V. and Shimkets, L.J. 2000. Prokaryotic Development. American Society for Microbiology. Washington, DC. USA: 191-218. 226

EPA. 2004a. Bacillus subtilis MBI600 129082. http://www.epa.gov/pesticides/biopesticides/ingredients/factsheets/factshe et_129082.htm

EPA. 2004b. Bacillus subtilis GBO3 (129068). http://www.epa.gov/pesticides/biopesticides/ingredients/factsheets/factshe et_129068.htm.

FAO. 2004. FAOSTAT agriculture data. http://apps.fao.org/page/collections?subset=agriculture.

227