Host Preferences and Pathogenomics of Pseudomonas fuscovaginae in Australia

Kankanthri Nirodha Sandeepani Weeraratne

B.Sc. Agriculture (Hons.)

A thesis submitted to Charles Sturt University, Australia

for the degree of Doctor of Philosophy

December 2017

School of Agricultural and Wine Sciences

Faculty of Science

Charles Sturt University

Table of Contents

List of Figures ...... VII List of Tables ...... X Certificate of Authorship ...... XII Acknowledgements ...... XIII Abstract ...... 1 Chapter 1. Introduction and Literature Review ...... 6 1.1. Important bacterial diseases of rice ...... 7 1.2. Sheath rot diseases of rice ...... 9 1.2.1. Symptoms and Diagnosis ...... 9 1.2.2. Diversity of causal organisms ...... 10 1.2.3. Economic impact and global distribution of the disease ...... 11 1.2.4. Control ...... 13 1.3. Burkholderia glumae ...... 16 1.3.1. Characteristics ...... 16 1.3.2. Symptoms and diagnosis ...... 17 1.3.3. Pathogenicity and virulence ...... 17 1.3.4. Host range ...... 18 1.3.5. Control ...... 19 1.4. Acidovorax avenae ssp. avenae ...... 20 1.4.1. Characteristics ...... 20 1.4.2. Symptoms and diagnosis ...... 20 1.4.3. Pathogenicity and virulence ...... 21 1.4.4. Host range ...... 22 1.4.5. Control ...... 23 1.5. Pantoea ananatis ...... 23 1.5.1. Characteristics ...... 24 1.5.2. Symptoms and diagnosis ...... 24 1.5.3. Pathogenicity and virulence ...... 25 1.5.4. Host range ...... 26 1.5.5. Control ...... 27

II

1.6. Pseudomonas fuscovaginae ...... 27 1.6.1. Classification and nomenclature ...... 28 1.6.2. Morphological and physiological characteristics ...... 28 1.6.3. Disease symptoms and diagnostics ...... 29 1.6.4. Epidemiology ...... 30 1.6.5. Geographical distribution ...... 31 1.6.6. Genetic diversity ...... 31 1.6.7. Pathogenicity and virulence ...... 32 1.6.8. Host range ...... 36 1.6.9. Control ...... 37 1.6.10. P. fuscovaginae in Australia ...... 39 Chapter 2. Potential hosts of P. fuscovaginae strains isolated from rice in Australia and their host-pathogen interactions ...... 44 2.1. Introduction ...... 44 2.1.1 Host range of P. fuscovaginae ...... 44 2.1.2 The range of potential hosts for P. fuscovaginae in Australia ...... 45 2.1.3 Occurrence of P. fuscovaginae in rice cropping regions in Australia...... 47 2.1.4 Sources of host plant resistance to P. fuscovaginae ...... 49 2.1.5 Techniques for screening of host plant resistance...... 52 2.2 Materials and Methods ...... 54 2.2.1 Strains of P. fuscovaginae ...... 54 2.2.2 Selection of cultivars ...... 57 2.2.3 Plant growth conditions ...... 62 2.2.4 Preparation of inoculum and inoculation techniques ...... 62 2.2.5 Data collection ...... 63 2.2.6 Data analysis ...... 63 2.2.7 Survey of sheath brown rot disease ...... 64 2.2.8 Preparation of plant samples collected during the survey ...... 65 2.2.9 Use of Loop mediated isothermal amplification (LAMP) technique...... 66 2.2.10 Isolation and screening of putative P. fuscovaginae from diseased plant samples ...... 67 2.3 Results ...... 67 2.3.1 Screening of existing collection of putative P. fuscovaginae isolates ...... 67 2.3.2 Response of rice cultivars to inoculation with P. fuscovaginae...... 73 2.3.3 Response of wheat cultivars to inoculation with P. fuscovaginae ...... 77

III

2.3.4 Response of durum-wheat cultivars to inoculation with P. fuscovaginae 84 2.3.5 Response of triticale cultivars to inoculation with P. fuscovaginae ...... 85 2.3.6 Response of barley cultivars to inoculation with P. fuscovaginae...... 87 2.3.7 Outcomes of the disease survey ...... 88 2.3.8 Results of LAMP assay for detecting P. fuscovaginae under field conditions ...... 90 2.3.9 Results of screening of bacteria isolated from diseased plant samples from the survey ...... 90 2.3.10 Response of rice and wheat to inoculation with P. fuscovaginae strains isolated from wheat ...... 91 2.4 Discussion ...... 92 2.5 Conclusions ...... 101 Chapter 3. Comparative genomics of P. fuscovaginae from a perspective of host- specificity ...... 103 3.1. Introduction ...... 103 3.2. Materials and methods ...... 107 3.2.1. P. fuscovaginae genomes ...... 107 3.2.2. Whole genome comparisons ...... 109 3.3. Results ...... 110 3.3.1. Rice NRPS I ...... 114 3.3.2. Rice NRPS II ...... 115 3.3.3. Rice NRPS III ...... 116 3.3.4. Rice NRPS IV ...... 117 3.3.5. Rice NRPS V ...... 118 3.3.6. Rice PS I ...... 118 3.3.7. Rice PS II ...... 119 3.3.8. Rice LC-FAC I ...... 120 3.3.9. Rice LC-FAC II ...... 121 3.3.10. Rice LC-FAC III ...... 122 3.3.11. Rice LC-FAC IV ...... 123 3.3.12. Rice LC-FAC V ...... 124 3.3.13. Rice LC-FAC VI ...... 125 3.4. Discussion ...... 126 3.5. Conclusion ...... 132

IV

Chapter 4. Identification and evaluation of a SypA homologue from P. fuscovaginae ...... 134 4.1. Introduction ...... 134 4.2. Materials and Methods ...... 137 4.2.1. Bacterial strains, plasmids ...... 137 4.2.2. Recombinant DNA techniques and sequence analysis ...... 139 4.2.3. Mutagenesis of the SypA gene homologue in P. fuscovaginae ...... 140 4.2.4. Locating sypA mutation in whole genome sequences of P. fuscovaginae strains ...... 144 4.2.5. Screening of P. fuscovaginae SypA mutants for virulence deficiency ...... 145 4.2.6. HPLC analysis and isolation of syringopeptin ...... 146 4.2.7. Statistical analysis ...... 147 4.3. Results ...... 148 4.3.1. Sequence analysis indicates that the mutation occurred in sypA homologue ...... 148 4.3.2. HPLC analysis reveals lack of production of phytotoxins in mutants ...... 152 4.3.3. SypA mutant P. fuscovaginae strains are less virulent than their respective wild types ...... 154 4.4. Discussion ...... 158 4.5. Conclusions ...... 165 Chapter 5. Identification and Characterization of Adhesins in P. fuscovaginae .. 166 5.1. Introduction ...... 166 5.2. Materials and Methods ...... 172 5.2.1. Bacterial strains, plasmids and culture media ...... 172 5.2.2. Recombinant DNA techniques and sequence analysis ...... 174 5.2.3. Mutagenesis of the Ad1 gene in P. fuscovaginae ...... 177 5.2.4. Preparation of P. fuscovaginae competent cells ...... 179 5.2.5. Sequence analysis and Genome mining ...... 183 5.3. Results ...... 184 5.3.1. Identification of adhesion like functions in the genome of P. fuscovaginae ...... 184 5.3.2. Mutagenesis in Ad1 gene ...... 188 5.3.3. P. fuscovaginae competent cells ...... 188 5.3.4. Sequence analysis...... 189

V

5.3.5. Genome mining ...... 191 5.4. Discussion ...... 194 5.5. Conclusions ...... 201 Chapter 6. General Discussion ...... 203 6.2. Directions for future studies ...... 220 7. References ...... 221 8. Appendix ...... 265

VI

List of Figures

Figure 1.1: Global distribution of Pseudomonas fuscovaginae (CABI, 2017b) ...... 31 Figure 2.1: Major rice growing regions in Australia (Kealey, Clampett et al. 2000) ..... 46 Figure 2.2: Sampling locations of rice-wheat cropping rotations in the Murrumbidgee Irrigation area (MIA), Coleambally Irrigation Area (CIA) and Murray Valley (MV) regions ...... 65 Figure 2.3: Virulence of Pseudomonas fuscovaginae strains on wheat cultivars inoculated with 107 cfu/mL of P. fuscovaginae by pinprick method ...... 84 Figure 2.4: Growth response of rice seedlings (cv. Amaroo) inoculated with 107 cfu/mL of Pseudomonas fuscovaginae by seed-soaking method. Data represent seedling height 10 days after inoculation...... 91 Figure 2.5: Growth response of wheat seedlings (cv. Rosella) inoculated with 107 cfu/mL of P. fuscovaginae by seed-soaking method. Data represent seedling height 10 days after incolation...... 92 Figure 3.1: Comparison of five rice-infecting Pseudomonas fuscovaginae genomes (inwards: CB98818, DAR77795, DAR77800, ICMP5940, SE-1), aligned to the reference genome of the rice-infecting strain UPB0736...... 112 Figure 3.2: Comparison of four wheat-infecting Pseudomonas fuscovaginae genomes (inwards: UPB0588, UPB0526, UPB1013, UPB0589) aligned to the reference genome of the rice-infecting strain UPB0736...... 113 Figure 4.2: pKNOCK- Km plasmid (Addgene plasmid # 46262) ...... 142 Figure 4.3: Insertional mutagenesis using pKNOCK plasmids (modified from Alexeyev (1999)) ...... 142 Figure 4.4: PCR amplification of F2R and R2F with PfSy-F2/pKNOCK-NewR and PfSy- R2/pKNOCK-NewF primers combinations (modified from Alexeyev (1999)) ...... 143 Figure 4.5: BLAST results for Pseudomonas fuscovaginae DAR77795 hypothetical protein (2026 bp) ...... 149 Figure 4.6: Conserved domain hits that match with the mutated region (1023 bp) of the sypA homologue in Pseudomonas fuscovaginae DAR77795 ...... 150 Figure 4.7: Location of sypA homologue in Pseudomonas fuscovaginae DAR77795 and DAR77800 genome assembly, as indicated by the blue coloured arrows ...... 150

VII

Figure 4.8: Sequence alignment of selected 1631 bp regions of sypA homologues from Pseudomonas fuscovaginae whole genome sequences corresponding to DAR77795 sypA homologue 358 – 1989 bp...... 151 Figure 4.9: Molecular phylogenetic analysis of the selected 1631 bp regions of sypA homologues from Pseudomonas fuscovaginae whole genome sequences corresponding to DAR77795 sypA homologue 358-1989 bp, inferred using the maximum-likelihood method. The tree is drawn to scale with branch lengths measured in number of substitutions per site...... 151 Figure 4.10: UPLC-MS chromatogram for bacterial exudates of Pseudomonas fuscovaginae strains DAR77795, DAR77800, and their respective sypA homologue mutants, in comparison to reference strain UPB0736 prepared in KB medium. The chromatographic peaks seen at retention time from 2.00 to 2.50 for the reference P. fuscovaginae UPB0736 wild type strain, have masses corresponding to fuscopeptins...... 153 Figure 4.11: Growth of Pseudomonas fuscovaginae cultures in nutrient broth, shaking at

180 rpm, at 28 °C as indicated by change in OD600 over time ...... 154 Figure 4.12: Growth response of rice (cv. Amaroo) to seed soaking with 107 cfu/mL of Pseudomonas fuscovaginae. Length of shoot and root for each treatment is the mean of four replicates, each containing 15 seedlings. Bars with the same letter (upper case and lower case) are not significantly different at the 5% level...... 156 Figure 4.13: Growth response of wheat (cv. Rosella) to seed soaking with 107 cfu/mL of Pseudomonas fuscovaginae. Length of shoot and root for each treatment is the mean of four replicates, each containing 15 seedlings. Bars with the same letter (upper case and lower case) are not significantly different at the 5% level...... 156 Figure 4.14: Percent disease index of rice (cv. Nipponbare) plantlets inoculated with 107 cfu/mL of Pseudomonas fuscovaginae. PDI for each treatment was calculated with 24 seedlings equally distributed in to two blocks. Bars with the same letter are not significantly different at the 5% level...... 157 Figure 5.1: pEX 18-Gm plasmid (Suksomtip & Tungpradabkul, 2005)...... 178 Figure 5.2: Graphical representation of pMP4655 (egfp) and pMP4662 (rfp) ...... 182 Table 5.4: BLASTN results for Ad1 nucleotide sequence from Pseudomonas fuscovaginae DAR77795 in NCBI ...... 185 Figure 5.3: Putative conserved domains in Ad1 gene of Pseudomonas fuscovaginae strain DAR77795...... 187 VIII

Figure 5.4: egfp-expressing colonies of Pseudomonas fuscovaginae strain DAR77795 WT...... 188 Figure 5.5: egfp-expressing bacteria cells of Pseudomonas fuscovaginae strain DAR77795 WT ...... 189 Figure 5.6: Putative conserved domain hits from NCBI BLASTX of 179Ad1 insert. . 190 Figure 5.7: Molecular phylogenetic analysis of the selected regions extracted from Pseudomonas fuscovaginae whole genome sequences homologous to Ad1 of DAR77795, inferred using the maximum-likelihood method. The tree is drawn to scale with branch lengths measured in number of substitutions per site...... 193

IX

List of Tables

Table 2.1: Bacteria strains used in this study ...... 56 Table 2.2: Details of the selected cultivars of rice and the winter cereals ...... 58 Table 2.3: Identification of putative P. fuscovaginae isolates by BACTID ...... 69 Table 2.4: Disease response (percent sheath lesion length) according to the interactions between rice cultivars and strains of Pseudomonas fuscovaginae ...... 74 Table 2.5: Growth response of wheat cultivars inoculated with Pseudomonas fuscovaginae 107 cfu/mL by seed-soaking method, showing the interactions between wheat cultivars and strains of P. fuscovaginae ...... 78 Table 2.6: Growth response of two durum-wheat cultivars inoculated with 107 cfu/mL of Pseudomonas fuscovaginae by seed-soaking method, showing the interactions between durum-wheat cultivars and P. fuscovaginae strains ...... 85 Table 2.7: Growth response of two triticale cultivars inoculated with 107 cfu/mL of Pseudomonas fuscovaginae by seed-soaking method, showing the interactions between triticale cultivars and P. fuscovaginae strains ...... 86 Table 2.8: Growth response of two barley cultivars inoculated with 107 cfu/mL of Pseudomonas fuscovaginae by seed-soaking method, showing the interactions between barley cultivars and P. fuscovaginae strains ...... 87 Table 2.9: Details of sampling locations of the survey for sheath brown rot during the 2013/14 rice and wheat cropping seasons ...... 89 Table 3.1: Background information of the genomes of Pseudomonas fuscovaginae used in this study ...... 109 Table 3.2: Features of the genomes of Pseudomonas fuscovaginae used in this study ...... 110 Table 4.1: Bacterial strains used in this study ...... 138 Table 4.2: Plasmids used and generated in this study ...... 138 Table 4.3: Primers used in this study ...... 139 Table 4.4: PCR conditions for PsfSyp ...... 141 Table 4.5: PCR conditions for the amplification of F2R and R2F...... 144 Table 5.1: Bacterial strains used in this study ...... 173 Table 5.2: Plasmids used and generated in this study ...... 173

X

Table 5.3: Primers used in this study ...... 176 Table 5.4: BLASTN results for Ad1 nucleotide sequence from Pseudomonas fuscovaginae DAR77795 in NCBI ...... 185 Table 5.5: BLASTN results for Ad1 nucleotide sequence from Pseudomonas fuscovaginae DAR77795, in draft genomes of P. fuscovaginae in NCBI/GenBank.... 191 Table 5.6: BLASTN results for Ad1 of Pseudomonas fuscovaginae DAR77795 in P. fuscovaginae genomes ...... 192

XI

Certificate of Authorship

I hereby declare that this submission is my own work and to the best of my knowledge and belief, understand that it contains no material previously published or written by another person, nor material which to a substantial extent has been accepted for the award of any other degree or diploma at Charles Sturt University or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by colleagues with whom I have worked at Charles Sturt University or elsewhere during my candidature is fully acknowledged. I agree that this thesis be accessible for the purpose of study and research in accordance with normal conditions established by the Executive Director, Library Services, Charles Sturt University or nominee, for the care, loan and reproduction of thesis, subject to confidentiality provisions as approved by the University.

Name: Kankanthri Nirodha Sandeepani Weeraratne

Signature:

Date: 20.12.2017

XII Acknowledgements

First, I would like to thank my principal supervisor Prof. Chris Steel for accepting me as a PhD student, and Prof. Gavin Ash for directing me to Prof. Steel, being the former

Postgraduate coordinator of Faculty of Science. I am very much indebted to Prof. Steel and the co-supervisors, Prof. Ash, Dr. Sandra Savocchia and Dr. Ben Stodart for their invaluable advice and kind support in every possible manner through the journey towards formulating this thesis. Special thanks to Dr. Stodart for guiding me through genomics and bioinformatics, a new area of research for me. Furthermore, I am grateful to Dr.

Savocchia, who supported me to maintain my PhD candidature through difficult times, as the Sub Dean Graduate studies for the Faculty of Science.

Then, I would like to express my sincere gratitude to Dr. Vittorio Venturi of the

Bacteriology Research Group at the International Centre for Genetic Engineering and

Biotechnology (ICGEB), Trieste (Italy), for accepting me as a graduate student, and Prof.

Ash for directing me to ICGEB. I am grateful to Dr. Venturi, Dr. Giulia Devescovi and

Dr. Iris Bertani for guiding me through molecular microbiology and bioinformatics, leading to the success of studies described in Chapter 4 and 5 of this thesis.

Many thanks to Andrew Watson, New South Wales Department of Primary Industries

(NSW DPI) in Yanco for his kind support in various ways. I am obliged to Dr. Ben

Ovenden and Peter Snell at NSW DPI, Yanco, for providing directions and rice seeds and

Dr. Andrew Milgate of NSW DPI, Wagga Wagga, for providing directions and wheat, durum, triticale and barely seeds, leading to the success of studies mentioned in Chapter

2 of this thesis.

I am indebted to Dr. Monica Hofte and Dr. Khuong Hua of Ghent University, Belgium and Marc Ongena of University of Liège, Belgium, for conducting the analyses

XIII mentioned in Chapter 4. In addition, I am thankful to Prof. Claude Bragard of Universite

Catholique de Leuven, Belgium for kindly sending the Pseudomonas fuscovaginae isolates from the UPB collection.

I would like to express my sincere gratitude to rice grower families, the Houghtons of

Leeton, the Bullers of Murrami, the Andreazzas and the Menzies of Griffith, the Mannus, the Pattersons, the Evans, the Sheppards and the Sutherlands of Coleambally, the

Scoullars, the Hands and the Allits in Deniliquin, the Bradfords of Mayrung, and the

Strongs of Moone, for their generous support in letting me inspect their rice and wheat crops and to take samples as required.

My special thanks to the Wagga Wagga Plant and Invertebrate Pathology Group, especially Dr. Bree Wilson, Dr. Dante Adorada and Dr. John Harper for their valuable insights and support in numerous ways. I am obliged to Encar Adorada and Gurli Nielson for their assistance during difficult times. In addition, I am thankful to Dr. Padraig Strappe and Dr. Sergio Moroni for their help in need. Furthermore, I am obliged to Dr. David

Luckett of NSW DPI, Wagga Wagga, for his guidance in stastical analysis.

Without the support of the technical and administration staff of the Faculty of Science, none of these studies could have been conducted. Particularly, I am thankful to Kamala

Anggamuthu, Natalie Allison, David Thompson, Charmaine Carlisle, Trent Smith,

Ashleigh Van Oosterum, Melanie Snell, Annette Doherty and Michael Gentle. In addition, I am obliged to Jacquie Blomfield and Katie Murrell-Orgill of CSU student support services. Furthermore, I am grateful to all the staff at School of Agricultural and

Wine Sciences, Sutherland labs, National Life Sciences Hub, Lab Stores, National Wine and Grape and Industry Centre and the CSU Research office for offering support when it is much needed.

XIV Without the camaraderie of fellow postgraduate students, Sujeewa Rathnayake, Aruni

Buddhika, Kylie Crampton, Thushara Abayawickrama, Ronnie Dotaona, Razia Shaik,

Mostafa Kamal, Nadika Bandara, Navideh Sadoughi, Amir Sohail, Shamsul Haque,

Jesmin Aktar, Joe Moore, Sana Hanif, Aurelie Quade, Rowan Alden, Sebastian De Mattia and Diego Zhu, the tedious life of a PhD student would have been more difficult. In addition, I am grateful to fellow postgraduate students, Hitendra Patel, Gordana Uzelac,

Daniel Passos, Xiangfa Meng and Bruna Cutinho at ICGEB for making my stay there memorable.

It has been a wonderful journey of learning and perseverance, but I would not have survived it without the unconditional love and the infinite support of my husband Uvindu

Warnakulasooriya, my mother Madhavi Wijesinghe, my father Gamini Weeraratne, my aunt Padamaltha Wijesinghe, my brother Udara and two sisters Yasoda and Prabodha.

Moreover, I am blessed to have the love of my baby Bevin who grew up together with this thesis.

I am grateful to the family of Sri Lankans in Wagga Wagga, for their generous support in numerous ways to go through this journey.

Last, but not the least, my sincere gratitude goes to the Faculty of Science, Charles Sturt

University for providing me the Compact Scholarship, funding the research and the stipend for three and a half years, and the Graham Centre for Agricultural Innovation and the Australasian Plant Pathology Society for providing financial support to attend conferences.

XV Abstract

The bacterial plant pathogen Pseudomonas fuscovaginae causes sheath brown rot disease on a broad range of hosts. Since the first detection from Japan in 1976, it has been reported from various agro-ecological regions around the world. In 2005, P. fuscovaginae infecting rice was first reported from Australia in the major rice-growing region of southern New South Wales (NSW). The pathogenicity and aggressiveness of five strains namely, DAR77318, DAR77320, DAR77795, DAR77797 and DAR77800, were identified and characterised. Although there are several reports on host preference and pathogenomics of strains of P. fuscovaginae, no such research has been conducted for the identified Australian strains. Nor is it known whether strains isolated from rice pose a threat to cereal crops including wheat.

In this thesis, the potential host range of the five strains of P. fuscovaginae from Australia were investigated utilising pathogenicity and virulence assays. Comparisons were made with four strains of P. fuscovaginae isolated from rice, originating from Japan

(ICMP5940), Burundi (ICMP9995) and Madagascar (ICMP9996, ICMP9998), and strain

ICMP10137 isolated from wheat in Brazil. Pathogenicity and virulence assays were conducted on ten cultivars of wheat, by both seed-soaking and pinprick inoculation methods. In addition, two cultivars each of durum wheat, triticale and barley were inoculated by the seed-soaking method. All five of the Australian strains of P. fuscovaginae demonstrated pathogenicity and virulence at heading on wheat cultivars,

Chara, Ventura, Mace, Spitfire, EGA Gregory, Lincoln, Sunvale and Wyalkatchem, as indicated by percent lesion length on flag leaf sheaths. These and two other wheat cultivars namely, EGA Wedgetail and Rosella were susceptible to infection by P. fuscovaginae at seed germination, as indicated by reduced seedling height. Similar effects 1 were observed for the triticale (cv. Hawkeye, cv. Tahara), durum-wheat (cv. EGA

Bellaroi, cv. Tjilkuri) and barley (cv. Schooner, cv. Commander) at seed germination.

This is the first report on the pathogenicity of rice-infecting strains of P. fuscovaginae from Australia on wheat.

Furthermore, ten cultivars of rice that are commonly grown in southern NSW were inoculated with the aforementioned strains of P. fuscovaginae by the pinprick method.

Based on the percent lesion length on flag leaf sheaths at heading, all the rice cultivars investigated, Amaroo, Doongara, Illabong, Koshihikari, Kyeema, Langi, Opus, Quest,

Reiziq, Sherpa, were susceptible to P. fuscovaginae. Strain ICMP10137, isolated from wheat, expressed pathogenicity and virulence on all the rice cultivars, demonstrating the cross-species pathogenicity of P. fuscovaginae. By providing host resistance information on a range of cultivars, this host preference study provides insights for variety selection for cultivation and breeding of resistant cultivars.

Four additional strains of P. fuscovaginae isolated from wheat, UPB0526, UPB0588 and

UPB0589 originating from Mexico and UPB1013 from Nepal, and UPB0407 isolated from Leersia hexandra (Swamp rice grass) in Burundi, were able to cause significant (α

= 0.05) pathogenicity on rice (cv. Amaroo), at the seed germination stage. Strain

UPB0407 was also virulent on both rice (cv. Amaroo) and wheat (cv. Rosella) at seed germination. The results demonstrate the potential threat of strains infecting rice to infect wheat and other cereals crops. Furthermore, it indicates the possibility of the pathogen to be present in rice-wheat cropping systems.

In order to determine the presence of P. fuscovaginae in rice-wheat cropping systems in southern NSW, a survey of both rice and wheat crops was conducted, covering the

Murrumbidgee and Coleambally irrigation areas, and the Murray valley. Disease symptoms associated with sheath brown rot were observed in three out of 16 rice crops

2 and one out of 17 wheat crops. However, P. fuscovaginae could not be identified from the bacteria isolated from them.

Notably, there were significant (α = 0.05) strain to cultivar interactions observed in expressing disease symptoms in rice, at heading. However, the strain to cultivar interactions were not significant (α = 0.05) in wheat at heading. Similar observations were made with wheat, durum wheat, triticale, and barley, at seed germination. These results suggested that different host resistance mechanisms are engaged at different stages of plant growth in wheat. To explain these observations of host-pathogen interactions, further investigations were carried out on the molecular interactions between the hosts and strains of P. fuscovaginae.

Few molecular studies have been conducted on the pathogenicity and virulence factors of

P. fuscovaginae and only a small number of pathogenomic studies supported by whole genome analyses have been conducted on rice-infecting strains. However, none of these included strains from Australia. In this thesis, a comparative genome analysis was conducted of P. fuscovaginae, with a perspective of host-preference. Assembled draft genomes of rice-infecting DAR77795, DAR77800, ICMP5940, SE-1 from the

Philippines, CB98818 from China and UPB0736 from Madagascar were compared with the newly assembled genomes of wheat-infecting UPB0526, UPB0588 and UPB0589 from Mexico, UPB1013 from Nepal and UPB0497 isolated from swamp rice grass in

Burundi. The results show a highly diverse core-genome of P. fuscovaginae. Some distinct differences between the strains could be observed. Strains of P. fuscovaginae isolated from rice had 119 unique genes encoding hypothetical proteins, non-ribosomal peptide synthetases, peptide synthetases, long chain fatty acid CoA ligases, and large exo- proteins involved in haem utilisation and adhesion functions. Sixty-five genes appeared to be more common to wheat-infecting P. fuscovaginae strains. The genome of UPB0407

3 appeared to have acquired a number of phage-related genes and hypothetical proteins that were not present in other strains.

Further analysis revealed that one of the genes coding for hypothetical proteins/non- ribosomal protein synthesis to be a homologue of phytotoxic syringopeptin synthetase A

(sypA) of P. syringae pv. syringae. Therefore, a mutagenesis protocol was undertaken utilising the pKNOCK system to knock-out this gene in strains DAR77795 and

DAR77800. The subsequent effect of this knock-out on pathogenicity and virulence on both rice and wheat seeds and rice seedlings was analysed. The inability to synthesize this syringopeptin homologue was found to affect the virulence of DAR77795 and

DAR77800 mutants, as indicated by the root lengths and shoot lengths of inoculated seeds that were comparable to those of the control treatment of sterile distilled water, and inability of the mutants to cause necrotic lesions on the inoculated seedlings in comparison to the respective wildtypes. The syringopeptin homologue appears to be a non-host-specific toxin affecting both rice and wheat, contributing to the pathogenicity of both strains of P. fuscovaginae. Genome mining for the region targeted by the mutagenesis within the syringopeptin gene homologue showed that it is conserved in all

10 of the P. fuscovaginae genomes (DAR77795, DAR77800, SE-1, UPB0736, UPB0526,

UPB0588 and UPB0589, UPB1013, ICMP5940, CB98818) assessed. Therefore, this phytotoxin could be an important factor responsible for non-host specific pathogenicity and virulence reported for P. fuscovaginae.

Another gene, which was present in only seven of the examined genomes and showed high variability among the strains of P. fuscovaginae, was also targeted for mutagenesis to determine its role in pathogenicity and virulence. Several techniques such as the deletion-insertion by pEX system, and gene knockout by pKNOCK system failed to generate stable mutants in both DAR77795 and DAR77800. Therefore, these large exo- 4 proteins involved in haemaglutinin and adhesion functions could not be studied further.

However, during the experimental process, the ability to produce competent cells from P. fuscovaginae was demonstrated. Thus, direct transformation with plasmids could be possible for future studies of the pathogen.

5 Chapter 1. Introduction and Literature Review

Rice (Oryza sativa L.) is the second most widely grown cereal crop in the world (Statista,

2017) and the staple food for more than half the world’s population, which is congregated in developing countries in Asia and the Pacific (Ricepedia, 2017). Ensuring that there is sufficient rice available and affordable for everyone is of paramount importance to guarantee food security in Asia and the Pacific, where 90% of the global rice produce is consumed (IRRI, 2017b). Rice is becoming an important part of the staple diet in Africa and Latin America as well, where the population is rapidly growing (Ricepedia, 2017), further increasing the demand for rice as food. It is predicted that current food production will have to be increased by 50% by 2050 to cater for the world population which is estimated to reach approximately 10 billion by 2050 (FAO, IFAD, UNICEF, WFP, &

WHO, 2017). At present, rice is cultivated on 161.1 million ha producing 483.9 million t of milled rice (Statista, 2017).

As a way of increasing production, the possibility of expanding the area under rice cultivation is considered as remote, because of limited land and water resources

(Papademetriou, 2005). In the developing countries in Asia, land is becoming less available due to increased demand from expanding urban and industrial sectors, whereas, in Sub-Saharan Africa and Latin America, developing new lands to suit paddy cultivation is expensive (Van Nguyen & Ferrero, 2006). Furthermore, around the world, available water for agriculture has become depleted, while rice cultivation is water intensive

(Bouman, Humphreys, Tuong, & Barker, 2007). Therefore, increasing the productivity of rice cultivation is the only option for increasing total rice production. However, since the 1990s, the average rate of increase in the global yield of rice, has been just over one percent per year (FAO, 2017). Major factors that contribute to decline of the growth rice

6 production include abiotic and biotic stresses, declining productivity in intensive rice production systems, limited returns as the potential for increasing yield by breeding high yielding cultivars has reached the maximum, increasing production costs and increasing public attention for the protection of environmental resources (Van Nguyen & Ferrero,

2006).

Improvements in rice production have been obtained via breeding efforts aimed at producing cultivars that are high yielding, short season and stress-tolerant (Brennan &

Malabayabas, 2011). However, biological threats such as weeds, pests and diseases remain an obstacle to the efforts to increase productivity (Sundström et al., 2014). It is estimated that, in rice, pests and diseases cause yield losses up to 25%, annually (Van

Nguyen & Ferrero, 2006). Bacterial diseases are a key constraint for improving rice productivity, particularly in Asia (Saha, Garg, Biswas, & Rai, 2015).

1.1. Important bacterial diseases of rice

There are about 10 major rice diseases known to be caused by bacteria, which are classified into three genera, Erwinia, Pseudomonas, and Xanthomonas. Species of the genus Erwinia isolated from rice are commonly considered to be saprophytic, although they may also be opportunistic pathogens (Cottyn, Certz, & Mew, 1994a). Erwinia herbicola Löhnis 1911 Dye 1964 (syn. Pantoea agglomerans Ewing and Fife 1972

Gavini et al. 1989 (Tindall, 2014)), Erwinia uredevora Serrano 1928 (syn. Pantoea ananatis (Mergaert, Verdonck, & Kersters, 1993)) and Erwinia chrysanthemi

Burkholder et al. 1953 (syn. Dickeya dadantii (Samson et al., 2005)) have been reported as pathogens of rice. E. uredovora causes grain discolouration (Cortesi, Bartoli, Pizzatti,

Bertocchi, & Schaad, 2008; Yan et al., 2010), stem necrosis (Cother, Reinke, McKenzie,

Lanoiselet, & Noble, 2004), new bacterial leaf blight (Kini et al., 2016; Mondal, Mani,

7

Singh, Kim, & Mudgett, 2011) and E. herbicola is the causal organism of palea browning

(Azegami, 2013) and leaf blight (Lee, Hong, & Kim, 2010), while E. chrysanthemi causes severe foot rot (Goto, 1979) and sheath browning (Cottyn et al., 1994a).

The genus Xanthomonas contains many species that are considered as important plant pathogens, and have been reported from a wide variety of plant hosts. Bacterial blight of rice, caused by Xanthomonas oryzae pv. oryzae Ishiyama 1922 (Swings et al., 1990) is seed borne and considered the most serious and widespread bacterial disease in rice production (Zhang & Wang, 2013). X. oryzae pv. oryzicola Fang et al. 1957 (Swings et al., 1990) causes bacterial leaf streak of rice and is considered less important than X. oryzae pv. oryzae (CABI, 2017c).

The genus Pseudomonas includes many devastating plant pathogens of a wide range of economically important plant hosts cultivated around the world. The characteristics of the members of this genus are complex, thus it has undergone considerable taxonomic reclassification (Ozen & Ussery, 2012). These heterogeneous species are grouped based on rRNA homology, and the rRNA-homology group I consists of fluorescent

Pseudomonas spp., because of their ability to produce pyoverdines (Aremu & Babalola,

2015). One of these species is Pseudomonas syringae pv. syringae Van Hall 1904

(Skerman, McGowan, & Sneath, 1980) that causes glume blotch (Cother, 1974;

Whitworth et al., 2013), and another one; Pseudomonas fuscovaginae Tanii, Miyajima and Akita 1976, causes bacterial sheath brown rot of rice (Tanii, Miyajima, & Akita,

1976).

In addition, Pseudomonas avenae ssp. avenae Manns 1909 (syn. Acidovorax avenae ssp. avenae Willems et al. 1992 (Schaad et al., 2008)), which is non-fluorescent, causes bacterial brown stripe/ bacterial stripe of rice (Li et al., 2011; Somda, Veena, &

Mortensen, 2001; Song, Kim, Hwang, & Schaad, 2004; Xie, Sun, & Mew, 1998).

8

Pseudomonas glumae Kurita and Tabei 1967 (syn. Burkholderia glumae (Urakami et al.,

1994)), which is also non-fluorescent, causes bacterial grain rot of rice (Goto &

Watanabe, 1975; Tsushima, 1996; Uematsu, Yoshimura, Nishiyama, Ibaraki, & Fujii,

1976), grain discolouration (Zeigler & Alvarez, 1989), bacterial wilt (Jeong et al., 2003;

Kim, Kang, Kim, Choi, & Hwang, 2010) and bacterial panicle blight (Nandakumar et al.,

2009; Riera-Ruiz et al., 2014) and is considered as one of the most important bacterial disease of rice (Ham, Melanson, & Rush, 2011; Zhou-qi, Bo, Guan-lin, Bin, & Shi-wen,

2016).

1.2. Sheath rot diseases of rice

Sheath rot of rice is a disease complex that is caused by a number of fungal and bacterial pathogens (Cottyn et al., 1996a). Until recently, it was considered as geographically limited and a minor disease although, there is an increased interest in rice sheath rot as an emerging and widespread disease (Bigirimana, Hua, Nyamangyoku, & Höfte, 2015).

1.2.1. Symptoms and Diagnosis

Pathogens affecting the flag leaf sheath cause sheath rot at panicle initiation (booting), panicle emergence (heading), grain filling (milk) and grain maturation (dough and mature) stages of the rice plant. Typical sheath rot symptoms include brown discolouration and irregular spots or lesions on the flag leaf sheath that encloses the young panicles (Rostami, Ghasemi, Rahimian, & Khosravi, 2010), poor panicle emergence

(Duveiller, Notteghem, Rott, Snacken, & Maraite, 1990), necrosis of infected leaf sheath

(Saberi, Safaie, & Rahimian, 2013), discolouration of grains (Cottyn et al., 1996a), and in the worst cases, result in unfilled or partially-filled grains and panicle sterility (Cortesi et al., 2008), which may lead to yield losses of 80-100% (Bigirimana et al., 2015). Florets

9

turn dark reddish brown to black in colour. The intensity of grain discolouration ranges from sporadic discolouration to discolouration of the whole glume (Phat, Duong, & Du,

2005). Due to the similarity of symptoms caused by the pathogens, reliable diagnosis based only on symptomology is difficult (Cottyn et al., 1996a; Duveiller, Miyajima,

Snacken, Autrique, & Maraite, 1988).

1.2.2. Diversity of causal organisms

Among the fungal pathogens associated with sheath rot of rice, Sarocladium oryzae

Sawada 1922 Gams and Hawksworth 1976, is considered the most virulent (Sakthivel,

2001). In addition, Helminthosporium spp., Cercospora spp. Gerlachia spp.,

Trichoconiella spp. Cuvularia spp. Alternaria padwickii Ganguly 1947 Ellis 1971 (syn.

Trichoconiella padwickii Jain 1975 (Dugan & Peever, 2002)), Bipolaris oryzae de Haan

1900 Shoemaker 1959 (syn. Cochliobolus miyabeanus (Manamgoda et al., 2014)),

Fusarium spp., Tilletia barclayana Sacc. and Syd. 1899 (TullÍs & Johnson, 1952), Phoma sorghina Sacc. and Syd. 1899 Boerema, Dorenb. and Kesteren 1973 (Boerema, 2004),

Cephalosporium oryzae, and Ustilaginoidea virens Cooke 1828 Takahashi 1896 (syn.

Villosiclava virens Nakata 1934 (Tanaka, Ashizawa, Sonoda, & Tanaka, 2008)) have been often isolated from rice plants showing symptoms of sheath rot (Cother, Stodart, &

Ash, 2009a; Cottyn, 2003; Du, Loan, Cuong, Nghiep, & Thach, 2001; Naeimi, Okhovvat,

Hedjaroude, & Khosravi, 2003; Sharma, Sthapit, Pradhanang, & Joshi, 1996).

Over 5,600 different bacteria have been isolated from rice plants and seeds with sheath rot and grain discolouration symptoms (Cottyn et al., 2001; Cottyn et al., 1996b; Cottyn,

2003). Pseudomonas spp. are the most common group of bacteria associated with sheath diseases, and the following species have been isolated: Pseudomonas stutzeri Lehmann and Neumann 1896 Sijderius 1946 (Lalucat, Bennasar, Bosch, García-Valdés, &

10

Palleroni, 2006) and Pseudomonas fulva Iizuka and Komagata 1963 (Uchino, Shida,

Uchimura, & Komagata, 2001) (Cortesi et al., 2008), P. syringae pv. syringae,

Pseudomonas aeruginosa Schroeter 1872 Migula 1900 (Skerman et al., 1980), P. fuscovaginae, Pseudomonas aureofaciens Kluyver (1956), Pseudomonas corrugata

Scarlett, Fletcher, Roberts, and Lelliott (1978), Pseudomonas fluorescens Flügge 1886

Migula 1895 (Bergey, Krieg, & Holt, 1984), Pseudomonas putida Trevisan 1889 (Anzai,

Kim, Park, Wakabayashi, & Oyaizu, 2000) and Pseudomonas marginalis Brown 1918

Stevens 1925 (Anzai et al., 2000); (Adorada, Stodart, Tpoi, Costa, & Ash, 2013b; Cother et al., 2010; Cottyn et al., 2001; Cottyn et al., 1996b; Saberi et al., 2013; Zeigler &

Alvarez, 1990). Opportunistic bacteria such as Erwinia spp. contribute to discoloured grains (Yan et al., 2010), affect germination and cause seedling diseases, in contaminated seeds (Chinara, Mishra, & Dhal, 2007).

However, the extent to which each individual bacteria species contribute to total disease development of sheath rot is not known. Adorada (2013) noted that the bacteria to be considered are P. fuscovaginae, Acidovorax avenae ssp. avenae, B. glumae, P. ananatis,

Erwinia sp., and Xanthomonas sp.. Bigirimana et al. (2015) considered S. oryzae,

Fusarium sp., and P. fuscovaginae to be the most important pathogens involved in rice sheath rot disease complex.

1.2.3. Economic impact and global distribution of the disease

Bacterial sheath diseases have been reported to cause significant damage to rice crops

(Cother et al., 2009a), with more than 30% discoloured grains, panicle sterility up to 50%

(Cortesi et al., 2008) and yield losses as high as 100% for severe infections (Cahyaniati

& Mortensen, 1995; Naeimi et al., 2003; Rott, Notteghem, & Frossard, 1989). Since colour is an important quality parameter of rice, discoloured grains having an unattractive

11

yellow appearance or black spots on milled rice can downgrade the market quality, as well as significantly reduce the rice milling recovery, and affect cooking and eating quality (Cortesi et al., 2008; Du et al., 2001; Phat et al., 2005).

Bacterial sheath rot of rice has been observed to thrive in climates with low temperatures and high humidity levels (Detry, Chapeaux, & Tilquin, 1991a; Sharma et al., 1996), although Jaunet, Notteghem, and Rapilly (1996) concluded that there is no direct relationship between low temperature and disease development. Furthermore, bacterial sheath rot has been reported from regions with high temperatures (Ham & Groth, 2011;

Nandakumar et al., 2009; Razak et al., 2009b) and high humidity (Mew, Leung, Savary,

Vera Cruz, & Leach, 2004) especially in the rainy season where disease severity has been mild to severe (Cottyn et al., 1996a).

Presumably, the changes in rice farming systems since the green revolution in Asia around the 1960s, such as; the use of high-yielding, photoperiod-insensitive, semi-dwarf cultivars; increased use of fertilizers, particularly nitrogen fertilizers; increased plant density, have led to favourable conditions for diseases (Bigirimana et al., 2015; Saha et al., 2015; Van Nguyen & Ferrero, 2006; Zhou-qi et al., 2016) such as sheath rot. It is hypothesised that the photoperiod-insensitive cultivars are likely to continue flowering under conditions of high humidity and high temperature that are conducive to the disease

(Mew et al., 2004). Furthermore, increased global migration and trade has provided many opportunities for exchange of planting material, thus, enabling the spread of disease during the last two decades (Adorada et al., 2013b; Falk & Wallace, 2011; Lanoiselet,

Cother, & Ash, 2001; Meyerson & Reaser, 2002).

Symptoms of bacterial sheath rot disease have been reported from rice crops in Asia

(Adorada et al., 2013b; Cahyaniati & Mortensen, 1995; Chiba, Chiba, Shimada, Ota, &

Kuwata, 1977; Cother et al., 2010; Cottyn et al., 1996a; Cottyn, George, & Vera Cruz,

12

2002; Cottyn et al., 1996b; Kim, Choi, & Kim, 2015; Razak et al., 2009b; Rostami et al.,

2010; Sharma et al., 1996; Ura et al., 2006; Xie, 2003), South America (Malavolta, De

Almeida, & Malavolta, 1997b; Zeigler & Alvarez, 1986; Zeigler & Alvarez, 1987), Africa

(Duveiller et al., 1988; Duveiller et al., 1990; Macapuguay & Mnzaya, 1988; Rott et al.,

1989; Somda et al., 2001), North America (Ham & Groth, 2011; Nandakumar, Rush,

Shahjahan, O'Reilly, & Groth, 2005; Yuan, 2004), Europe (Cortesi et al., 2008) and

Australia (Cother et al., 2004; Cother, Stodart, Noble, Reinke, & van de Ven, 2009b).

1.2.4. Control

Maintaining crop sanitation by removal of alternative host weeds (Kaur & Singh, 2015;

Kumar, Gupta, Maheshwari, & Atwal, 2014), use of healthy seed paddy (Adorada,

Stodart, Pangga, & Ash, 2014; Baker & Smith, 1966; Cottyn, 2003), and removal of infected plants and infected stubble after harvest is vital for disease control (Cother et al.,

2009a; Lanoiselet et al., 2005). Maintaining optimum plant spacing, providing the nutrient requirement of the crop at right time (Chau et al., 2003), especially application of potassium at tillering stage (Sakthivel, 2001) and silicon (Zhang, Dai, & Zhang, 2006) reduce diseases.

For sheath rot caused by fungi, foliar application of calcium sulphate and zinc sulphate and seed treatment and foliar spraying at booting stage with the fungicides carbendazim or mancozeb, benomyl and copper oxychloride has been found to be effective (Sakthivel,

2001). However, in recent years, development of fungicide resistance and concerns of environmental pollution have revived interest in devising novel plant-based chemicals that are environmentally sustainable (Ganesan, Vadivel, & Jayaraman, 2015; Yoon, Cha,

& Kim, 2013). Botanicals such as the extract of Chromolaena odorata L. (Khoa, Thuy,

Thuy, Collinge, & Jorgensen, 2011), Cocculus hirsutus L. (Shivalingaiah, Umesha, &

13

Sateesh, 2013), Piper sarmentosum Roxb. (Chanprapai & Chavasiri, 2017) have been found to be effective against some of the fungal organisms associated with rice sheath diseases.

Several studies have been conducted on the biological control of sheath rot (Cottyn,

Debode, Regalado, Mew, & Swings, 2009; Gnanamanickam, 2009). P. fluorescence was found to effectively control S. oryzae (Reddy, Choudary, & Reddy, 2007; Sakthivel &

Gnanamanickam, 1987). Delftia tsuruhatensis has shown potential as a biocontrol agent against P. syringae pv. syringae (Ghazemi, Rostami, & Yazdaninia, 2012). Application of bacteriophages both alone and with non-pathogenic P. ananatis, has shown potential to suppress bacterial palea browning caused by pathogenic P. ananatis (Azegami, 2013).

An isolate of rhizobacterium, Bacillus amyloliquefaciens, has shown potential as a biocontrol agent of brown sheath rot disease of rice caused by P. fuscovaginae (Kakar et al., 2014). Rice-associated bacterial isolates of B. amyloliquefaciens have demonstrated effectiveness in controlling bacterial panicle blight caused by B. glumae in field conditions (Shrestha, Karki, Groth, Jungkhun, & Ham, 2016).

Screening germplasm for resistance and breeding of resistant rice cultivars is seen as the best option for controlling sheath rot disease (Bigirimana et al., 2015; Christopher,

Cordeiro, Waters, & Henry, 2004; Collard, Vera Cruz, McNally, Virk, & Mackill, 2008;

Ham et al., 2011; Sakthivel, 2001). Various techniques for screening rice cultivars have been developed and carried out targeting some individual species involved in this disease complex; S. oryzae (Kalpana et al., 2006; Mahadevaiah, Shailaja, Uday, & Kumar, 2015;

Mvuyekure, Sibiya, Derera, Nzungize, & Nkima, 2017), P. fuscovaginae (Adorada et al.,

2013c; Detry, Duveiller, & Maraite, 1991b; Tilquin & Detry, 1993), B. glumae (Chien &

Chang, 1987; Goto & Watanabe, 1975; Mizobuchi et al., 2013; Mogi & Tsushima, 1985;

14

Nandakumar & Rush, 2008), Acidovorax avenae ssp. avenae (Heydari-Nezhad,

Babaeizad, & Rahimian, 2015).

However, the inability to predict exact involvement of each individual pathogen in the disease complex has made it difficult to take effective control measures (Bigirimana et al., 2015). The complex selective influence of diverse pathogens in populations is a challenge in breeding rice cultivars resistant to multiple pathogens (Mundt, 2014). The unpredictable nature of the factors acting in the sheath rot diseases might explain the variability in losses attributed to an individual pathogen, e.g. the yield losses attributed to

S. oryzae can be as variable as in the range of 20 - 85% (Bigirimana et al., 2015; Sakthivel,

2001). Currently, there is a lack of comprehensive studies on the link between the presence and quantity of disease inoculum and yield loss, with regard to sheath rot diseases (Bigirimana et al., 2015; Mew & Gonzales, 2002). Particularly, there is limited information available for resistance breeding regarding sheath rot diseases of bacterial origin. Furthermore, due to the interference of environmental factors such as temperature and humidity, it is difficult to screen a diverse range of cultivars of which the heading occurs at different times in the field by conventional inoculation techniques (Mizobuchi et al., 2013). Thus, molecular marker (such as quantitative trait loci associated with resistance) -assisted breeding has become the future of resistant breeding of rice

(Arunakumari et al., 2016; Ashkani et al., 2015; Das, Patra, & Baek, 2017; Fiyaz et al.,

2016). However, unlike other bacterial diseases of rice such as bacterial blight, no marker genes for complete resistance to bacterial sheath rot diseases have been discovered and expression of partial resistance is subjected to environmental conditions (Mizobuchi,

Fukuoka, Tsushima, Yano, & Sato, 2016). Sundin, Castiblanco, Yuan, Zeng, and Yang

(2016) noted that further research has to be carried out on methods besides host resistance to manage systemic bacterial pathogens; hence, a comprehensive understanding of

15

bacterial pathosystems is required to identify optimal targets in the pathogens and to optimise seasonal timings for the deployment of control measures.

Given the extraordinary genetic variability among the strains of each individual pathogen

(Ayyadurai, Kirubakaran, Srisha, & Sakthivel, 2005; Mann et al., 2013; Quibod et al.,

2015b; Seo et al., 2015; Taghipour, Rahimian, & Babaeazad, 2015), studying the interactions of the strains of these pathogens with certain rice cultivars and also with each other in the plant disease microbiome, will be advantageous for devising novel sustainable disease management techniques. Therefore, this literature review aims to analyse the research that has been carried out to-date on the four most important bacterial pathogens involved in sheath rot disease P. fuscovaginae, B. glumae, P. ananatis and

Acidovorax avenae ssp. avenae.

1.3. Burkholderia glumae

Burkholderia glumae is considered among one of the major bacterial pathogens of rice in many areas around the world (Zhou-qi et al., 2016). It causes bacterial panicle blight, which is an increasingly important disease problem in global rice production and seedling and grain rot of rice (Ham et al., 2011). Burkholderia spp. has been consistently isolated from bacterial leaf-sheath browning, grain discolouration and grain rot of rice (Cortesi et al., 2008; Cottyn et al., 2009; Cottyn et al., 2001; Cottyn et al., 1996b; Zeigler & Alvarez,

1990).

1.3.1. Characteristics

B. glumae is a Gram-negative, non-fluorescent, capsulated, motile, lophotrichous flagella, pectinolytic bacterium producing a yellow-green, water-soluble pigment on various media. Colonies appear yellow on King’s B (KB) medium (King, Ward et al. 1954) due

16

to this pigment. The organism is rod-shaped, with 1-3 polar flagella. Basic biochemical characteristics of B. glumae include a negative reaction to arginine dihydrolase test, oxidase test and nitrate reduction test and a positive reaction to lecithinase production and inositol test (Yuan, 2004; Zhu et al., 2008).

1.3.2. Symptoms and diagnosis

Bacterial panicle blight is characterised by affected panicles having blighted florets, which initially display white or light grey colouration on the basal third with a dark-brown margin, which eventually become brown coloured. The florets then turn dark with growth of fungi or bacteria on the surface. In addition, seedling blighting and sheath rot is observed. Extensive occurrence of upright panicles as a result of aborted seeds, empty grains due to failure of grain filling and seed development is a typical phenomenon observed in a severely infested field (Nandakumar et al., 2009).

1.3.3. Pathogenicity and virulence

B. glumae produces two major virulence factors: the phytotoxin toxoflavin and a secreted lipase and type III effectors. Toxoflavin and lipase are known to be major virulence factors of this pathogen, and their production is dependent on the TofI/TofR quorum- sensing system, which is mediated by N-acyl homoserine lactone (AHL)-dependent, cell- density-dependent quorum-sensing regulation system. A LysR-type transcriptional activator ToxR was found to regulate toxoflavin biosynthesis and transport in B. glumae

(Kim et al., 2004). Flagellar biogenesis and a type III secretion system are also required for full virulence of B. glumae, and are regulated by quorum sensing (Chen, Barphagha,

Karki, & Ham, 2012; Degrassi, Devescovi, Kim, Hwang, & Venturi, 2008; Devescovi et al., 2007; Ham et al., 2011; Kim et al., 2007; Liu et al., 2008).

17

Optimum temperature for disease development is reported as 30 - 35 ℃ with a range of

11 - 40 °C and a thermal death point at 70 °C (Tsushima, Mogi, & Saito, 1986). Severe epidemics causing up to 40% yield losses have been reported when there has been prolonged high night temperature during the growing season (Ham et al., 2011; Tsushima,

1996). With the recent changes in global climate, increased occurrence of disease has been reported from North, Central, and South America as well as in Asia (Kim et al.,

2010; Lim et al., 2009; Luo, Xie, Li, & Lihui, 2007; Nandakumar et al., 2009; Sayler,

Cartwright, & Yang, 2006). The pathogen is considered to be seed-borne, but also survives in the soil (Nandakumar, Bollich, Shahjahan, Groth, & Rush, 2008). B. glumae behaves as an endophyte in seedlings grown from infected seeds moving acropetally to the panicle. Flowers become infected when the threshold bacterial population is reached and environment conditions are favourable (Tsushima, Mogi, Naito, & Saito, 1991;

Tsushima, Naito, & Koitabashi, 1995). Spatial distribution patterns of the disease in infested rice fields suggest dissemination of disease from severely diseased panicles to neighbouring healthy plants (Tsushima & Naito, 1991). However, the disease cycle is not fully understood and long distance dissemination and the role of insect vectors is not known (Ham et al., 2011).

1.3.4. Host range

Burkholderia spp. isolated from rice plants with bacterial leaf-sheath browning and grain rot were pathogenic and induced symptoms of seedling rot, grain rot, and leaf-sheath browning in rice plants, as well as in some orchidaceous plants (cymbidium, dendrobium, and oncidium leaves), gladiolus leaves, and onion bulbs (Hiroyuki et al., 2006) and also cause bacterial wilt on many other field crops (Jeong et al., 2003; Kim et al., 2010). In

18

addition, B. glumae was reported to cause chronic granulomatous disease (CGD) in humans (Weinberg et al., 2007).

1.3.5. Control

Despite the economic importance of this widespread disease, effective control measurements are not widely available. A better understanding of the molecular mechanisms underlying virulence and of the rice defence mechanisms is required for development of better disease control methods (Ham et al., 2011). Potential control measures include seed sterilizsation and planting of partially resistant lines (Nandakumar,

Bollich, Groth, & Rush, 2007). Breeding of a completely resistant variety has not yet been realised (Zhou-qi et al., 2016). A few biocontrol agents have also been tested (Cho et al.,

2007). Cultural practices such as early planting to avoid high temperatures during the panicle growth stages and use of less nitrogen fertilizers reduce disease severity (Ham &

Groth, 2011). There are no recommended pesticides available, although, copper compounds are weakly effective. Seed treatment and foliar spray with oxolinic acid has been found effective for controlling the disease (Hikichi, Okuno, & Furusawa, 1995).

Mapping resistance genes of the rice plant may provide information for marker-assisted selective breeding (Mizobuchi et al., 2013; Sayler et al., 2006). The potential of disrupting quorum sensing as a control tactic has been demonstrated. An N-acyl-homoserine lactonase (aiiA) gene from Bacillus thuringiensis was introduced into bacterial endophyte inhibited production of quorum-sensing signals and reduced the disease incidence of rice seedling rot caused by B. glumae in situ. It shows the possibility of using a transformed bacterial epiphyte with the aiiA gene as an environment-friendly biological control agent against B. glumae, that inhabits the same ecological niche as B. glumae (Cho et al., 2007).

In addition, the induction of a rice defence system by pre-treatment of various chemical

19

materials, which leads to enhanced disease resistance, is being studied (Ham & Groth,

2011).

1.4. Acidovorax avenae ssp. avenae

Acidovorax avenae ssp. avenae has been repetitively isolated from discoloured grain and sterile panicles, and sheath rot symptoms of rice (Cortesi et al., 2008; Cother et al., 2010;

Cottyn et al., 2009; Cottyn et al., 2001; Cottyn et al., 1996b; Cottyn, 2003; Saberi et al.,

2013; Zeigler & Alvarez, 1990). It causes bacterial brown stripe disease in rice and has been reported in many countries in Asia, Africa, the Americas, and Europe (Claflin,

Ramundo, Leach, & Erinle, 1989; Cortesi et al., 2008; Saberi et al., 2013; Shakya &

Chung, 1983; Somda et al., 2001; Xie et al., 1998; Zeigler & Alvarez, 1989).

1.4.1. Characteristics

Colonies of A. avenae ssp. avenae are small (2-3 mm) whitish-grey or cream-tan coloured, raised, fringed, entire, non-fluorescent and translucent on nutrient agar. The bacterium is gram-negative, oxidase positive, nitrate positive, arginine dihydrolase negative and starch-negative. It gives positive reactions for tobacco hypersensitive reaction and produces acid from arabinose, fructose, galactose, glucose, glycerol, and sorbitol (Claflin et al., 1989; Mudingotto, Veena, & Mortensen, 2001; Shakya & Chung,

1983; Somda et al., 2001).

1.4.2. Symptoms and diagnosis

The symptoms start as water-soaked, dark green-brown, longitudinal stripes on the bottom of stems soon after emergence and frequently extend into the sheaths, spreading along the leaf midrib (CABI, 2017a; Goto & Ohata, 1956; Somda et al., 2001) and

20

throughout the seedling at the one-leaf stage (Li et al., 2011). Later the stripes turn dark brown (Somda et al., 2001). The coleoptile and leaf blade may also show symptoms

(Shakya & Chung, 1983; Shakya, Vinther, & Mathur, 1985). Mildly infected seedlings recover (Shakya, Chung, & Vinther, 1986; Shakya et al., 1985). Severely affected seedlings are stunted and eventually become dead (Plantwise, 2017a). On mature plants, brown stripes on leaf blades and sheaths can be observed and in the severest infections, brown strips extends over the whole leaf blade (Shakya et al., 1985; Webster & Gunnell,

1992).

1.4.3. Pathogenicity and virulence

A. avenae ssp. avenae is a seed-borne pathogen (Shakya & Chung, 1983). Rice seeds contaminated with this pathogen are important sources of the primary inoculum and a means of dissemination of the pathogen to new areas (Shakya et al., 1986; Shakya et al.,

1985). The bacterium is transmitted from seeds to seedlings and vice versa (Shakya et al.,

1986). It has been isolated from discoloured seeds as well as from asymptomatic seeds of rice (Xie et al., 1998). Latent infections in plants transmitting disease to second generation seeds have been observed (Shakya et al., 1986). Paddy nurseries in both upland and wetland ecosystems get affected. High humidity favours disease development (Cottyn,

Certz, & Mew, 1994b).

In studies conducted on the mechanism of virulence factors of A. avenae spp. avenae, hair-like appendages called Type IV pili encoded by the pilP gene were found to play a key role in plant pathogenicity, twitching motility and biofilm formation (Liu et al., 2012).

A. avenae spp. avenae has a wide host range among monocotyledonous plants; however, individual strains of this pathogen infect only one or a few host species. Incompatible strains have been found to induce an immune response of rapid cell death which is

21

characterised as programmed cell death (Che et al., 1999). Flagellin, a key component of flagella, induces plant immune responses such as rapid generation of H2O2, accompanying hypersensitive cell death and the expression of defence genes such as PAL, Cht-1, PBZ1, and LOX (Tanaka et al., 2003). The resistance response in rice cells is induced by the flagellin existing in the incompatible strain of P. avenae but not in the flagellin of the compatible strain (Che et al., 2000). Specific induction of immune responses by A. avenae flagellins is due to a structural difference between the virulent and avirulent flagellins. Induction specificity of the immune responses to flagellin is regulated by transcriptional sugar chain modifications. Four amino acids residues are glycosylated in the flagellin from the virulent strain which may prevent epitope recognition in rice (Hirai et al., 2011). The flagellin perception system in rice is a conserved flagellin perception system utilizing the FLS2 receptor which, when upregulated, hardly affects resistance against a compatible strain (Takai, Isogai, Takayama, & Che, 2008).

1.4.4. Host range

In addition to rice, A. avenae ssp. avenae causes disease in a wide range of economically important plants such as, bacterial stalk rot of corn, bacterial leaf blight of oats, red stripe of sugarcane (Song et al., 2004; Zia-ul-Hussnain et al., 2011) and bacterial stripe disease of barley, maize, oats, pearl millet, corn, millet, sorghum, foxtail millet, sugarcane and finger millet (Claflin et al., 1989; Furuya, Ito, & Tsuchiya, 2009; Mudingotto et al., 2001;

Shakya et al., 1985). However, individual strains can infect only one or a few host species

(Che et al., 1999). In addition, it has been found to infect watermelon, anthurium, orchids

(Schaad et al., 2008), and creeping bentgrass (Furuya et al., 2009).

22

1.4.5. Control

As rice seeds are a very important infection source for the transmission of A. avenae spp. avenae, care must be taken in the introduction of seeds. Therefore, use of healthy seeds is important to prevent seedling death in nurseries (Somda et al., 2001; Xie et al., 1998).

No specific chemical control is available for the disease, however, kasugamycin is effective against P. avenae and controls damage in nursery boxes (Cottyn et al., 1994b).

An effective method for controlling bacterial brown stripe of rice is to maintain adequate levels of temperature and humidity in growing rice seedlings (Kadota and Ohuchi, 1990).

Seed treatment by applying dry heat treatment at 65 °C for 6 days effectively eliminates the pathogen (Zeigler & Alvarez, 1989). No resistant breeding lines are available for rice.

1.5. Pantoea ananatis

Pantoea ananatis has been found in association with sheath rot and grain discolouration symptoms on rice from Asia (Cother et al., 2010; Cottyn, 2003; Yan et al., 2010), Europe

(Cortesi et al., 2008) and Australia (Cother et al., 2004). It is regarded as an emerging pathogen causing diseases on a wide range of economically important agricultural crops

(Carr, Bonasera, Zaid, Lorbeer, & Beer, 2010; Lana et al., 2012; Mondal et al., 2011) and forest tree species (De Maayer et al., 2010) around the world. Unlike other plant pathogens, it is capable of infecting humans and occurs in diverse ecological niches

(Coutinho & Venter, 2009; Weller-Stuart, De Maayer, & Coutinho, 2017). It has antimicrobial properties, and the potential as a biological control agent against some insect pests (Coutinho & Venter, 2009). The CrtI gene producing carotene desaturase from P. ananatis is used to increase the β-carotene content in Golden Rice (Al-Babili,

Hoa, & Schaub, 2006).

23

1.5.1. Characteristics

P. ananatis belongs to the family Enterobacteriaceae in the group Gammaproteobacteria. They are gram-negative, facultatively anaerobic, rod shaped bacteria that have three to six peritrichous flagella. Colonies are yellow, raised with smooth margins on nutrient agar. On yeast dextrose calcium carbonate agar, colonies are light yellow-to-orange, glistening and convex. P. ananatis is positive for β- galactosidase, catalase and negative for arginine dihydrolase, lysine decarboxylase, ornithine decarboxylase, tryptophan deaminase, phenylalanine deaminase, oxidase, urease, nitrate reductase, methyl red test and H2S production. P. ananatis can utilise citrate and tartrate, d-glucose, d-mannitol, d-melibiose, l-arabinose, sucrose, meso- inositol, glycerol, d-sorbitol, amygdalin, d-mannose, d-cellobiose, l-inositol, d-arabinose, cellulose, glycerol and d-arabitol, but not glycogen, N-acetyl-d-galactosamine, malonic acid, l-fucose, or xylitol. P. ananatis produces a yellow pigment and a positive tobacco hypersensitive reaction (Alippi & López, 2010; Carr et al., 2010; Coutinho & Venter,

2009; Yan et al., 2010).

1.5.2. Symptoms and diagnosis

Necrotic lesions occur on the rachis and stem, extending into the flag leaf sheath and stop at the second node. Therefore, the lesions can be seen on the flag leaf sheath near the collar. At early flowering of rice, light, rusty, water-soaked lesions appear on the lemma or palea and later turn brown. Affected stems are weaker than the healthy stems. Severely discoloured, immature and lighter grains are observed at harvest. Infection with P. ananatis significantly affects the percentage of whole grains remaining after milling and the yellowness index, degrading the grain quality to a greater extent (Azegami, 2013;

Cortesi et al., 2008; Cother et al., 2004; Yan et al., 2010).

24

1.5.3. Pathogenicity and virulence

Little is known about the major virulence factors of P. ananatis (Coutinho & Venter,

2009). The genome of a virulent strain of P. ananatis isolated from Eucalyptus has been completely sequenced and initial comparative genomics revealed the absence of the Type

II, Type III and Type IV secretion systems that are usually associated with pathogenicity of other plant pathogens. However, this particular strain of P.ananatis possess a Type VI secretion system (T6SS), which plays a major role in virulence in both animal and plant pathogens (De Maayer et al., 2011) and genes encoding proteins which show similarity to polyketide synthases (PKS)/non-ribosomal peptide synthases (NRPS). These proteins are involved in the synthesis of a wide range of secondary metabolites including antibiotics, siderophores as well as phytotoxins (De Maayer et al., 2012). Siderophore production is a common trait of the species P. ananatis and it gives the bacteria competitive advantages to colonise rice plant tissues (Loaces, Ferrando, & Fernández

Scavino, 2011). P. ananatis produces two AHLs, N-hexanoyl-l-homoserine lactone (C6-

HSL) and N-(3-oxohexanoyl)-l-homoserine lactone (3-oxo-C6-HSL) which regulate expression of some virulence factors through AHL-mediated quorum sensing, controlling biosynthesis of exopolymeric substances, biofilm formation, inhibits cell aggregation and ability of infection (Morohoshi et al., 2007). Disrupted cell aggregation or exopolysaccharide production does not necessarily affect the pathogenicity of P. ananatis

(Morohoshi, Ogata, & Ikeda, 2011; Morohoshi, Oseki, & Ikeda, 2011). Dissemination of

P. ananatis is encouraged by high temperature and relative humidity during the growing season (Azad, Holmes, & Cooksey, 2000). However, Cother et al. (2004) observed that hot, dry and windy conditions aggravates the disease severity in rice.

25

1.5.4. Host range

P. ananatis has a very broad host range including cereals, fruits, vegetables, condiments, forage grass, horticultural crops and forest tree species (Coutinho & Venter, 2009). It has been reported to cause leaf blotch disease characterised by symptoms of necrotic streaks and dark blotches on leaves of maize, sudan grass, sorghum, corn and oat in Argentina,

Brazil and Mexico and USA (Alippi & López, 2010; Azad et al., 2000; Cota et al., 2010).

Also, it can infect stored yellow onion bulbs and cause symptoms of odourless, light brown to brown scales (Carr et al., 2010). P. ananatis is also the causal organism of

Eucalyptus blight and dieback in South Africa (De Maayer et al., 2010). The host range of P. ananatis may be predicted by causing significant blight symptoms on Welsh onion upon stab inoculation and inducing a hypersensitive response (HR)-like reaction on tobacco by infiltration, and also the detection of iaaM, iaaH and etz genes by PCR (Kido,

Hasegawa, Matsumoto, Kobayashi, & Takikawa, 2010). Strains of Group Ι were characterised by positive reactions on both Welsh onion and tobacco and the absence of iaaM, iaaH, or etz genes, as well as palea browning on rice, but not causing internal fruit rot on melon. Isolates from foxtail millet, hydrangea, pineapple and rice belong to group Ι (Kido et al., 2010). Group II is characterised by possessing iaaM, iaaH and etz genes and causing internal fruit rot on melon but not causing blight on Welsh onion, no hypersensitivity reaction on tobacco and no palea browning on rice.

Isolates from melon belong to group II (Kido et al., 2010). Isolates from Group III do not possess iaaM, iaaH, and etz genes, do not cause blight on Welsh onion or no HR-like reaction on tobacco and are not pathogenic on rice or melon. Strains of Group III are from bamboo grass, Chinese silver grass, citrus, dogwood, melon, mugwort, silk tree, sweet corn, tea and Welsh onion (Kido et al., 2010).

26

1.5.5. Control

New reports of P. ananatis on previously unknown hosts emphasises the re-emergence of this bacterium as a potentially economically important plant pathogen and the need for disease control measurements. In eucalyptus, resistant/tolerant cultivars/clones are deployed but no resistant lines are available for rice (Coutinho & Venter, 2009).

Although there are no reports on the use of chemicals to control the disease on rice, the fungicide Mancozeb has been proven to be effective in controlling white spot disease in maize caused by P. ananatis, when applied during the early stages of the disease development in the fields, by completely inhibiting the growth of the bacterium (Bomfeti et al., 2007; Coutinho & Venter, 2009). However, given the omnipresent nature of this pathogen, alternative control measures that do not rely on chemical sprays are needed. A bacteriophage lytic to pathogenic P. ananatis has the potential to impede the growth of the pathogen on rice plants at the flowering stage (Azegami, 2013). Avoidance and eradication of the initial inoculum are probably the most appropriate management strategies. Genomic information will elucidate information pertaining to pathogenicity and host specificity and will facilitate the development of novel approaches to pathogen control in the future (Coutinho & Venter, 2009).

1.6. Pseudomonas fuscovaginae

Amidst the other established and well known rice diseases, sheath brown rot caused by

P. fuscovaginae (Tanii et al., 1976), is considered as an emerging, serious disease (Quibod et al., 2015a). It is becoming widespread over many different agro-ecological regions in

Asia (Cahyaniati & Mortensen, 1995; Chiba et al., 1977; Cother et al., 2010; Cottyn et al., 2002; Cottyn et al., 2001; Cottyn, 2003; Kim et al., 2015; Razak et al., 2009b; Rostami et al., 2010; Sharma et al., 1996; Xie, 2003), Africa (Detry et al., 1991a; Duveiller et al.,

27

1988; Duveiller et al., 1990; Onasanya et al., 2010; Rott, Honegger, Notteghem, &

Ranomenjanahary, 1991; Rott et al., 1989), South, Central and North America

(Malavolta, de Almeida, & Malavolta Jr, 1997a; Taunous, Kempf, & Duarte, 1997;

Zeigler & Alvarez, 1986; Zeigler & Alvarez, 1987; Zeigler & Alvarez, 1990; Zeigler,

Aricapa, & Hoyos, 1987), the Caribbean (Rivero-González et al., 2017) and Europe

(Arsenijevic, 1991). Sheath brown rot was first detected and rated as the most important bacterial disease of rice in Hokkaido, Japan in 1976 (Tanii et al., 1976). Disease symptoms resembling that of sheath brown rot were observed from rice fields in Australia for the first time in 2005 and P. fuscovaginae was identified as the causal organism

(Cother et al., 2009b).

1.6.1. Classification and nomenclature

Initially, Pseudomonas marginalis (Pseudomonas fluorescens biovar II) was reported to be the causal agent of sheath brown rot of rice in Japan in 1976. However, following detailed studies, the bacterium was re-identified as a new species and was given a new name of P. fuscovaginae (Miyajima, Tanii, & Tadahiko, 1983b). Genome analyses based on the draft genome sequence of P. fuscovaginae suggests that it is most closely related to the beneficial P. fluorescens than to pathogenic P. syringae and P. aeruginosa (Patel,

2012).

1.6.2. Morphological and physiological characteristics

P. fuscovaginae is described as a strictly aerobic, Gram negative, non-spore-forming and rod-shaped bacterium with a single, polar flagellum. Colonies appear as light brown to yellowish-brown on solid nutrient agar. P. fuscovaginae produces a green fluorescent pigment that is clearly observed on King’s B media and Pseudomonas agar (in the

28

presence of cetrimide) and grows well in Luria-Bertani media. The optimal growth temperature under laboratory conditions is 28 °C (Miyajima, Tanii, & Akita, 1983a). P. fuscovaginae oxidizes glucose in oxidation-fermentation medium and gives a positive reaction to oxidase and arginine dihydrolase (Rott et al., 1991). It differs from other oxidase and arginine dihydrolase positive non-pathogenic fluorescent Pseudomonas spp., by not utilizing 2-ketogluconate or inositol (Duveiller et al., 1988). In addition, it does produce acid from trehalose, nitrate reductase or β-glucosidase. It gives a hypersensitive reaction in tobacco (Zeigler & Alvarez, 1990).

1.6.3. Disease symptoms and diagnostics

Infection by P. fuscovaginae results in lesions on the leaf sheath, grain discolouration and panicle sterility in rice. Typical symptoms of the disease start to appear from seedling stage but are most prominently expressed at the panicle emergence stage (Batoko,

Bouharmont, Kinet, & Maraite, 1997b). Initially, lower leaf sheaths of infected seedlings show yellow-brown discolouration that later turn grey-brown to dark brown (Duveiller et al., 1988). Infected seedlings often die (Adorada et al., 2014; Razak et al., 2009a). On adult plants, the infected flag leaf sheaths show oblong to irregular dark green and water- soaked lesions on the adaxial side of the leaf sheath, which later become grey-brown or brown surrounded by an effuse dark brown margin (Xie, 2003). With severe infection, the entire leaf sheath turns necrotic, greyish brown or dark brown, wither and dry (Cother et al., 2009b; Xie, 2003). The disease affects upper internode elongation, reducing panicle emergence and results in various levels of sterility (Batoko, Bouharmont, Kinet, &

Maraite, 1997a). Grains of infected panicles are discoloured, deformed, poorly filled, empty, sterile, or may be symptomless except for small brown spots (Adorada et al., 2014;

Cother et al., 2009b; Razak et al., 2009a; Xie, 2003). The degree of disease susceptibility

29

could be determined after heading, by monitoring panicle emergence rate of diseased plants (Batoko et al., 1997b). No bacterial ooze is seen from any part of the infected plant

(Xie, 2003).

1.6.4. Epidemiology

P. fuscovaginae is often found associated with rice seeds and seeds are considered as the primary source of inoculum for disease transmission (Cottyn, 2003). It has been detected from infected seeds as well as asymptomatic, seemingly healthy seeds (Adorada et al.,

2014; Taunous et al., 1997; Zeigler & Alvarez, 1990; Zeigler et al., 1987).

Intercontinental movement of the pathogen can be attributed to trading of rice seeds

(Duveiller et al., 1988). Seeds of hosts as well as of non-hosts can be a source of inoculum for plant pathogenic bacteria in general (Darrasse et al., 2010). In temperate regions, P. fuscovaginae can survive in rice straw when stored indoors (Miyajima, 1980). In the tropics, other host plants or seeds harbour the bacterium (Cottyn, 2003). Bacterial population proliferating on or overwintering on symptomless leaf blades and sheaths can give rise to a secondary infection or disease cycle which is more infectious at the booting stage (Miyajima, 1980).

High elevation, low temperature and high humidity are reported to contribute to the high incidence of disease in irrigated temperate regions and rain-fed upland rice ecosystems

(Duveiller et al., 1990; Sharma et al., 1996). Cold temperature stress probably predisposes plants to severe infections (Detry et al., 1991a). However, severe disease incidence and yield losses from 50% to 100% have been reported from many countries regardless of the agro-climate being temperate or tropical (Cahyaniati & Mortensen, 1995; Razak et al.,

2009a; Rott et al., 1989). In addition, low temperatures act negatively on the pathogenicity process of P. fuscovaginae; thus, the occurrence of P. fuscovaginae in rice cultivation in

30

low temperature areas cannot be explained as a direct effect of temperatures on pathogenicity (Jaunet et al., 1996).

1.6.5. Geographical distribution

According to the Crop Protection Compendium, P. fuscovaginae is found in 35 countries

(CABI, 2017b), occupying a diverse range of agro-ecological regions (temperate, sub- tropical, tropical) distributed over all the continents of the world, except for Antarctica.

Figure 1.1: Global distribution of Pseudomonas fuscovaginae (CABI, 2017b)

1.6.6. Genetic diversity

Genomes of Pseudomonas species in general are known to display plasticity, with intrinsic genetic variation and continuous distribution of genetic diversity (Silby,

Winstanley, Godfrey, Levy, & Jackson, 2011). During an earlier analysis of genetic diversity of fluorescent pseudomonads isolated from diseased rice showing symptoms of sheath rot, by restricted fragment length polymorphism (RFLP) of 16S rRNA gene amplicons, the P. fuscovaginae strains isolated from Japan, Madagascar, Nepal,

31

Indonesia, and the Philippines were distinctly delineated from the other fluorescent

Pseudomonas spp. of 16S rRNA homology group I (Jaunet, Laguerre, Lemanceau,

Frutos, & Notteghem, 1995). The authors also suggested that intra-species polymorphism exists within this group of P. fuscovaginae strains akin to the P. fluorescens cluster that consists of closely related subgroup biovars I, II, III, and V. In fact, Cottyn et al. (1996b) questioned if P. fuscovaginae can be defined as a species given the genetic variability amongst the strains of this species. Similarly, Xie et al., 2012Cother et al. (2009b) reported single nucleotide polymorphisms (SNP) within the conserved region of 16S rRNA nucleotide sequence, during comparison of 15 P. fuscovaginae isolates, indicating the possibility of these isolates to form a cluster of closely related subgroups.

Recent genetic analysis studies conducted on P. fuscovaginae and P. fuscovaginae-like strains show that they have very flexible genomes, i.e. the genome of each strain comprises an open pan genome of a very distinct lineage accompanied by a repertoire of accessory genes (Patel et al., 2012; Quibod et al., 2015b; ). This means that P. fuscovaginae and P. fuscovaginae-like strains might possess diverse pathogenicity and virulence characteristics, with a potential to adopt to various hosts and environment conditions by genetic mutations and subsequent selection.

1.6.7. Pathogenicity and virulence

P. fuscovaginae produces phytotoxic lipodepsinonapeptide syringotoxin and two hydrophobic lipodepsipeptides, Fuscopeptin-A (FP-A) and Fuscopeptin-B (FP-B), concurrently which generate the symptoms observed on diseased plants (Ballio et al.,

1996; Batoko et al., 1997b; Batoko et al., 1998; Batoko, Flamand, Boutry, Kinet, &

Maraite, 1997d). These phytotoxins are considered as an integral component of the plant pathogen-interaction. The toxins induce rapid and high electrolyte leakage from calli on

32

rice cells (Batoko et al., 1997d). Syringotoxin is a structural analogue of syringomycin, produced by P. syringae pv. syringae (Batoko et al., 1998). Syringotoxin and syringomycin E at low concentrations stimulate H+-ATPase activity of native right-side out vesicles on rice shoot plasma membranes, and the reverse at higher concentrations and also with inside-out membrane vesicles. There is a synergistic action by the two types of toxins on inhibition of ATPase activity of plasma membranes. In contrast, FP-A and

FP-B induce inhibition of the H+-ATPase regardless the orientation of the vesicles

(Batoko et al., 1998; Batoko et al., 1997d). Fuscopeptins produced by P. fuscovaginae are structurally similar to syringopeptins produced by strains of P. syringae pv. syringae, although the optimum conditions for syringotoxin production by P. fuscovaginae are different from those reported for P. syringae pv. syringae (Ballio et al., 1996; Bare et al.,

1999; Coraiola, Paletti, Flore, Fogliano, & Serra, 2008; Flamand, Ewbank, Goret, &

Maraite, 1997).

Rice plants are sensitive to these toxins at all growth stages. The phytotoxins are non- host-specific, although, the severity of the toxin damage is related to the degree of cultivar susceptibility to the pathogen (Batoko et al., 1997b). The toxins induce a drastic inhibition of seedling elongation and affect the elongation of the peduncle and of the first internode, resulting in partial or total inhibition of panicle exertion (Batoko et al., 1997a; Batoko,

Bouharmont, & Maraite, 1994). The activity of toxins on germinating seeds and seedlings is considered as a reliable tool for screening genotype susceptibility/resistance to P. fuscovaginae at early growth stages (Batoko et al., 1994). It was reported that the purified toxins of P. fuscovaginae enhanced germination and have no apparent effect on the number of roots of the seedlings (Batoko et al., 1994), although the crude extract of bacterial exudates from P. fuscovaginae reduced the germination and lowered the number of roots of the seedlings as reported by Adorada et al. (2013c).

33

P. fuscovaginae produces Fe(III)-chelating siderophores called pyoverdins that are yellow-green in colour and fluoresce under UV light, in large quantities under iron- limited growth conditions. Pyoverdins are important in determining the pathogenicity of fluorescent Pseudomonas spp. P. fuscovaginae produces an atypical pyoverdin, which possess two beta-hydroxyaspartic acids as iron ligands which enables the bacterium to chelate more Fe(III) (Bultreys, Gheysen, Wathelet, Maraite, & de Hoffmann, 2003).

Recently, several studies have been published on the genes that are involved in synthesis and secretion of pathogenicity and virulence associated proteins and the regulation of those processes. P. fuscovaginae strain UPB0736 has been found to possess a unique complex regulatory network composed of two conserved N-acyl homoserine lactone

(AHL) quorum sensing (QS) systems designated as PfsI/R and PfvI/R regulated by two different repressors; RsaL and RsaM. RsaL is located between PfvI and PfvR and negatively regulates the PfvI/R system. RsaM located between PfsI and PfsR, negatively regulates both of the QS systems (PfvI/R and PfsI/R). PfsI is the synthase gene of PfsI/R system, which produces N-acyl homoserine lactones (N-AHLs) which are recognised by cognate PfsR. PfvI produces 3-oxo-N-AHLs that are recognised by the cognate PfvR that also recognises the AHLs produced by PfsI albeit at high concentrations. Both systems are required for virulence and pathogenicity of P. fuscovaginae on rice (Mattiuzzo et al.,

2011). Furthermore, Uzelac, Patel, Devescovi, Licastro, and Venturi (2017) found that the PfsI/R system regulates 98 genes while PfvI/R system regulates 26 genes but, only two genes are regulated by both systems. However, RsaM repressor regulates over 400 genes of which 206 are negatively regulated and 260 are positively regulated. Because

RsaM regulates both PfsI/R and PfvI/R systems, it regulates more than half of the genes controlled by the PfsI/R system and 65 % controlled by the PfvI/R system. Thus, it can be concluded that the RsaM is a universal regulator, which mediates gene expression through

34

the two QS systems as well as independently, while the two QS systems each regulate an exclusive set of genes. Quorum sensing regulatory systems in bacteria, modulate gene expression as a function of their cell density and regulate pathogenesis/virulence factors like toxins, motility, enzymes, biofilm related proteins, and induction of systemic resistance in the plant (Camilli & Bassler, 2006; Chin-A-Woeng, van den Broek,

Lugtenberg, & Bloemberg, 2005; Fuqua, Parsek, & Greenberg, 2001; Gelencsér et al.,

2012; Venturi, 2006). In P. fuscovaginae strains DAR77795 and DAR77780 isolated from Australia, the RsaL regulon appears to have degraded (Stodart, 2014). However, considering that the Australian P. fuscovaginae strains are highly virulent (Adorada,

2013), it can be hypothesised that the regulatory functions of the more common RsaL repressor does not have an influence on the pathogenicity and virulence of the pathogen as much as that of RsaM repressor. It should be noted that RsaM is a novel repressor in

QS systems that appears to be unique to P. fuscovaginae (Mattiuzzo et al., 2011). In addition, random transposon mutagenesis of P. fuscovaginae strain UPB0736 generated nine Tn5 mutants with both known and novel virulence-related loci affected (Patel et al.,

2014). These genes encode functions for biosynthesis of phytotoxic syringopeptin, type

IV pili, biosynthesis of arginine, sulphur metabolism and type VI secretion systems.

Mutation of these genes affected the pathogenicity and virulence of P. fuscovaginae strain

UPB0736.

Genome analysis of the draft genome sequence reports of P. fuscovaginae strains

UPB0736 and CB98818 provide more information about the pathogenicity and virulence associated genes that are important for bacterial survival and fitness functions such as adherence and movement, quorum sensing and protein secretion (Patel et al., 2012; Xie et al., 2012). Functional annotation of P. fuscovaginae strain CB98818 revealed the presence of pathogenicity-related genes like type III, IV and VI secretion systems, Hcp-

35

and VgR-like protein, Hrp protein, flagellin, multidrug resistance efflux pumps, multidrug transporter MdtB and ABC transporter subunits (Xie et al., 2012).

Another study analysing the draft genomes of P. fuscovaginae and P. fuscovaginae-like strains have shown that they have very flexible genomes with a diverse range of metabolic capabilities (Quibod et al., 2015b). The high structural and functional diversity indicates the adaptations of P. fuscovaginae to occupy multiple environments and frequent gene exchange through insertions and deletions. These data are supported by the analysis of secretome during infection on rice, which indicate diverse functional adaptations indicating the ability of P. fuscovaginae to occupy multiple environmental niches

(Quibod et al., 2015b).

Apart from the above-mentioned, studies on the pathogenicity and virulence functions of

P. fuscovaginae are limited, despite its global presence and emergence as a serious plant pathogen of a number of economically important crops (Bigirimana et al., 2015).

Therefore, there is a need to conduct further studies on the host-pathogen interactions of

P. fuscovaginae, considering not only the factors affecting pathogenicity and virulence, but also host range.

1.6.8. Host range

When first reported, P. fuscovaginae was identified as a pathogen of eight species of

Graminae other than rice (Miyajima et al., 1983a). To date, P. fuscovaginae has been reported from economically important crops including wheat (Triticum aestivum), maize

(Zea mays), oat (Avena sativa), triticale, barley (Hordeum vulgare), rye (Secale cereale) and sorghum (Sorghum bicolour) (Arsenijevic, 1991; Duveiller & Maraite, 1990;

Duveiller, Snacken, & Maraite, 1989; Malavolta et al., 1997a). P. fuscovaginae has also been isolated from both symptomatic and asymptomatic Graminae weeds such as

36

bentgrasses (Agrostis spp.), mountain bromegrass (Bromus marginatus), perennial ryegrass (Lolium perenne), smooth-stalked meadowgrass (Poa pratensis), and swamp rice grass (Leerisa hexandra) (Mulet, Lalucat, & Garcia-Valdes, 2010; Vera Cruz &

Opulencia, 2013; Xie, 2003). Although P. fuscovaginae was known only to affect monocots, P. fuscovaginae isolated from rice produced disease symptoms when inoculated on quinoa (Chenopodium quinoa Willd.) belonging to the family

Amaranthaceae (Mattiuzzo et al., 2011; Patel et al., 2014) and therefore indicating the possibility of cross pathogenicity. However, it is not known whether the P. fuscovaginae strains isolated from rice can cause disease in wheat and other cereal crops and vice versa.

1.6.9. Control

Crop sanitation involving weed control, immediate eradication of infected plants, and immediate eradication of crop stubble after harvest is an effective cultural method of disease control (IRRI, 2017a), given that P. fuscovaginae has been shown to spread through contaminated plant material (Adorada et al., 2014) and alternative plant hosts

(Miyajima, 1980). Breaking the disease cycle by off-season cultivation of a non-host plants such as a legume or a vegetable may be effective (Plantwise, 2017b). Use of healthy seeds (IRRI, 2017a) and adjusting the sowing date, use of 20 - 30 days old seedlings for transplanting or using early maturity cultivars to avoid cold stress towards the panicle emergence/panicle filling stage, have been recommended (IRRI, 2017a; Macapuguay &

Mnzaya, 1988; Plantwise, 2017b).

Dry heat treatment at 65 °C for 6 days eliminates P. fuscovaginae from seeds (Zeigler &

Alvarez, 1989), as does hot water treatment at 65 °C (IRRI, 2017a). Seed treatment with streptomycin alone or combined with oxytetracyclin (15% + 1.5%) effectively controls bacterial sheath brown rot (Cottyn et al., 1994a). Kasugamycin is also effective against

37

P. fuscovaginae but does not eliminate it completely from the seed (Cottyn et al., 1994b).

Adorada et al. (2014) reported that soaking seeds in sodium hypochlorite increased the germination of seeds contaminated by P. fuscovaginae. Plant extracts of Piper sarmentosum have antibacterial properties against P. fuscovaginae (Rahman, Sijam, &

Omar, 2014).

The rhizospheric bacterium, B. amyloliquefaciens strain Bk7, has shown potential biocontrol activity against P. fuscovaginae by suppressing the growth of the pathogen by

93% efficacy in vitro and 76.6% efficacy in reducing disease incidence of sheath brown rot on rice plants in glasshouse (Kakar et al., 2014). However, further studies need to be conducted in order to evaluate the efficacy of this potential bio-control agent under field conditions. Further studies are required to find potential bio control agents for P. fuscovaginae as this has been scarcely explored as a control method.

Breeding of disease resistant cultivars is generally considered as the best option for managing a plant pathogen. Breeding for resistant rice cultivars for P. fuscovaginae started soon after the disease was first diagnosed. These studies focused on developing tools of resistance screening (Batoko, Bouharmont, & Maraite, 1997c; Detry et al.,

1991b), such as mapping quantitative trait loci in rice germplasm for resistance against bacterial sheath brown rot (Tilquin & Detry, 1993), standardising inoculation methods and optimum inoculums concentration for different plant growth stages (Adorada et al.,

2013c; Detry et al., 1991b). Another aspect was breeding resistant cultivars that are suitable for different environmental conditions (Detry et al., 1991b). It was also demonstrated that the host-strain interaction as depicted by disease resistance level could differ depending on the maturity level of the plant (Adorada et al., 2013c). Therefore, host resistance of a certain crop variety should be tested at both the seed and mature stages of plant.

38

1.6.10. P. fuscovaginae in Australia

In Australia, the major rice production regions are in southern New South Wales

(Riverina) and northern Victoria, and less significantly in some parts of Queensland and

Northern Territory (Bajwa & Chauhan, 2017). Rice fields in southern New South Wales are irrigated by the Murrumbidgee irrigation scheme, and therefore an intensive water management regime is followed (Thompson, 2002). Rice production in this region is primarily for export and the industry is input-intensive and profit-oriented (RGA, 2017).

The scale of production in Australia is enormous compared to subsistence-level rice farming that is seen in most of rice producing countries that are in Asia and Africa. This level of scale demands intensive water, fertiliser and disease management practices in a cost-effective manner. Therefore, the decisions taken by rice growers in each cultivation season regarding the choice of rice variety and cultivation practices require accurate information about pest and disease tolerance as well as water and fertiliser use efficiency

(RGA, 2017). Due to advanced crop sanitation management, rice in Australia has been free from most of the major rice diseases that occur in the neighbouring rice producing countries (Lanoiselet et al., 2001). Diseases that are considered as a threat to Australian rice industry include rice blast (Magnaporthe grisea), bakanae (Gibberella fujikoroi), bacterial grain rot/panicle blight (Burkholderia gluamae) and kernal smut (Tilletia barclayana) (Adorada, 2013; Whitworth et al., 2013). The rice pest exclusion zone under

New South Wales legislation prevents the importation of rice plants and cereals such as millet or brown rice or rice products into this zone (Adorada et al., 2013b; Lanoiselet et al., 2001). Despite these strict quarantine measures, pathogens causing devastating diseases of rice such as P. ananatis have been identified in the Riverina (Cother et al.,

2004).

39

In 2005, symptoms of brown sheath rot caused by P. fuscovaginae were first observed on rice plants (cv. Amaroo) at the NSW DPI field station in Leeton in southern NSW. The causal organism was confirmed as P. fuscovaginae by polyphasic identification by Cother et al. (2009b) and five strains which are hereafter mentioned as “Australian strains” were identified. These Australian strains, which are deposited in the NSW DPI Plant Pathology

Herbarium (Orange, NSW) collection, are identified as, DAR77138, DAR77320,

DAR77800, DAR77795, DAR77797. The authors noted that some of the SNPs in the 16S rRNA gene are consistent within the five strains isolated from Australia and the type P. fuscovaginae strain ICMP5940 (GenBank; AB0213181.1), which could indicate their lineage. Later studies have also demonstrated that these strains are closely related to the

Japanese strains (Ash et al., 2014).

Currently, this pathogen is considered as a minor threat to the rice industry of Australia

(Whitworth et al., 2013). Surveys conducted in 2010 in the rice growing Riverina region of Australia, failed to find P. fuscovaginae (Adorada, 2013). However, the adverse effect of climate change, erratic temperature and humidity conditions that can occur during the

Australian summer in which rice is grown, may contribute to conducive conditions required for sudden and devastating outbreaks of P. fuscovaginae in Australia. Therefore, routine inspections assisted by rapid and accurate diagnostic techniques in rice growing regions are essential to identify a possible epidemic at the earlier stages and employ adequate control measures. A better understanding of the epidemiology, particularly of pathogenomics of the Australian strains of P. fuscovaginae, will contribute to broaden the knowledge of this pathogen in order to devise better disease prevention and management strategies, in both Australia and around the world.

Considering the currently employed crop management strategies and the scale of production in Australia, the search for sources of host plant resistance can be considered

40

as the most effective and economical tactic of disease management. Although previous studies have shown that breeding of disease resistant cultivars as the best option for managing P. fuscovaginae, the level of resistance to the pathogen within the rice cultivars grown in Australia are not yet known. Therefore, a study including the Australian strains against the popular rice cultivars that are currently grown in Australia is required.

The details regarding the interactions of the pathogen in Australian cropping systems are also limited. Although P. fuscovaginae is known as a devastating pathogen on wheat and other winter cereal crops, it is not known whether the Australian strains that are isolated from rice, are capable of causing disease in other cereal crops such as wheat. Up to now,

P. fuscovaginae has not been reported from the wheat crops that are grown in rotation in rice-wheat cropping systems in southern NSW. However, recent studies of genetic analysis conducted on two of the Australian strains of P. fuscovaginae along with several other P. fuscovaginae and P. fuscovaginae-like strains show that they have a highly diverse range of metabolic functions (Quibod et al., 2015a) indicating the ability of P. fuscovaginae to adapt to occupy various environments. Cother et al. (2009b) noted that there are some SNPs unique to each of the Australian isolates, although all five of them shared SNPs with type P. fuscovaginae strain ICMP5940, which demonstrates that the

Australian strains might have originated from the Japanese strains, but have adapted to

Australian cropping systems by rapid genetic mutations upon introduction to Australia.

The authors also noted that the Australian strains possess genetic differences that could be detectable by their metabolic profiles. Therefore, the response of winter cereals, particularly of wheat, to the infection with Australian P. fuscovaginae strains needs to be investigated, taking into consideration that the cross-pathogenicity of any of the known

P. fuscovaginae strains has not been established yet.

41

Furthermore, the genetic characteristics responsible for the host preferences of P. fuscovaginae remain largely unknown, mostly due to that lack of genome information on

P. fuscovaginae strains isolated from a diverse range of host plants. However, some distinct mechanisms involved in the colonisation of rice plants have been identified, indicating a set of unique genetic characteristics within rice-infecting P. fuscovaginae

(Quibod et al., 2015b). Also, the genetic background of the pathogenicity and virulence factors of Australian strains remain largely unexplored, though several studies have demonstrated the role of phytotoxins, siderophores, quorum-sensing systems and, secretion systems (type III, IV and, VI) in pathogenicity and virulence of other P. fuscovaginae strains (Patel et al., 2012; Patel et al., 2014; Xie et al., 2012). Some of the

Australian strains are more pathogenic and aggressive compared to the Japanese strains

(Adorada, Stodart, & Ash, 2013a). Furthermore within the Australian strains there are differences in virulence (Adorada et al., 2013a). This indicates that the Australian strains have rapidly evolved to have distinctly different metabolic capabilities. Functional annotation of the pathogenicity and virulence related genes of Australian P. fuscovaginae strains will assist to elaborate their unique traits. Therefore, a whole-genome comparison and studies on pathogenicity and virulence factors may provide more information on their conserved and unique mechanisms of infection, which will in turn facilitate better understanding of their host-pathogen interactions.

Thus, this research project reflecting on the heightened interest on P. fuscovaginae as an emerging pathogen of economic importance, aims to achieve the following objectives:

- To establish the pathogenicity and virulence of existing strains of P. fuscovaginae

originating from Australia on a range of cereal crop cultivars (selected to account

for the genetic variation among those cultivars), that are grown in rice-wheat

cropping systems in southern NSW, Australia;

42

- To survey for the prevalence of P. fuscovaginae within rice-wheat cropping

systems of the three major rice growing regions; Murrumbidgee Irrigation Area

(MIA), Coleambally Irrigation Area (CIA), and Murray Valley (MV);

- To utilise current genomic data to identify genetic elements associated with host-

specificity and host-specific virulence within P. fuscovaginae;

- To investigate the role of pathogenicity and virulence related genes present in

strains of P. fuscovaginae originating from Australia.

This thesis presents the background, methodology and, results of the experiments conducted to achieve the above-mentioned objectives, and discusses the outcomes and shortcomings, in order to arrive at conclusions and to suggest the future directions for studies to be conducted on P. fuscovaginae.

43

Chapter 2. Potential hosts of P. fuscovaginae strains isolated from rice in Australia and their host-pathogen interactions

2.1. Introduction

2.1.1 Host range of P. fuscovaginae

P. fuscovaginae is known as a pathogen of a broad range of plant families. It has been mostly reported from cereal crops of family Poaceae, particularly from Oryza sativa L.

(rice) and Triticum aestivum L. (wheat) causing sheath brown rot, grain discolouration, and seedling death and panicle sterility. These plant species are therefore considered the major hosts of P. fuscovaginae, while other cereals including Zea mays L. (maize),

Sorghum bicolor (L.) Moench (sorghum), Hordeum vulgare L. (barley), Secale cereale

L. (rye), Triticale x Triticosecale Wittm. Ex A. Camus and Avena sativa L. (oats) are considered as minor hosts.

Soon after P. fuscovaginae was first identified in 1976 from rice fields in Japan, Miyajima

(1980) reported that the pathogen was detected in Agrostis clavata var. nukabo and Poa pratensis growing in and around rice fields in Japan. Weed plants of family Poaceae such as Agrostis L. (bent-grass), Bromus marginatus Nees ex Steud. (Mountain brome-grass),

Lolium perenne L. (perennial ryegrass), Poa pratensis L. (smooth-stalked meadow- grass), Phleum pratense L. (timothy-grass), and Phalaris arundinacea L. (reed canary- grass) (Miyajima et al., 1983a) which are often found associated with cereal crops have been identified as minor hosts. In Burundi, it has been isolated from Leersia hexandra

Swart (Swamp rice-grass) (Bultreys et al., 2003).

The ability of P. fuscovaginae strains that are pathogenic to rice to over-winter in and remain undetected in other cereal crops and grassy weeds is plausible, although the cross- species pathogenicity of P. fuscovagiane has not been thoroughly investigated. However,

44

artificial inoculation tests indicate a range of potential hosts from several other monocotyledons belonging to families Commelinaceae, Cyperaceae and Pontederiaceae as well as several dicotyledonous species that are found in tropical American rice- growing areas (Zeigler & Alvarez, 1986). P. fuscovaginae was found to affect these plants at flowering, seedling and vegetative growth stages, causing disease on the inflorescence, leaves, seeds and the whole plant. In addition, it has been demonstrated that dicotyledonous crop species Chenopodium quinoa Willd (quinoa) of the family

Amaranthaceae can be used as a model to study the pathogenicity and virulence of P. fuscovaginae (Mattiuzzo et al., 2011; Patel et al., 2014).

2.1.2 The range of potential hosts for P. fuscovaginae in Australia

In Australia, sheath brown rot has been reported only from rice. Symptoms similar to those of the sheath brown rot disease were first observed in the rice fields in Leeton located in the Riverina region of NSW in 2005 and five ‘Australian’ strains of P. fuscovaginae were isolated (Cother et al., 2009b). In the Riverina region, rice is cultivated as a summer cereal crop from October to May. Californian cultivars were identified to be suitable to the region, according to the trials conducted in 1922 - 24 at the beginning of rice cultivation in the Riverina (Kealey & Clampett, 2000). The temperate climate of high summer temperatures and low humidity and the heavy soils in the Riverina are uniquely suited to growing medium-grain rice, which are commonly known as temperate cultivars.

Temperate rice cultivars of Asian Japonica origin are grown in the three major growing areas (Figure 2.1), MIA, CIA, and MV, over approximately 140,000 hectares of land

(Bajwa & Chauhan, 2017).

45

Figure 2.1: Major rice growing regions in Australia (Kealey, Clampett et al. 2000)

Rice cultivation relies entirely on irrigation and the crops are grown in 5-25 cm of standing water, depending on the growth stage of the plant (RGA, 2017; Thompson,

2002). The water is managed to insulate the plant from cold overnight temperatures during January (Farrell et al., 2001; RGA, 2017; Williams & Angus, 1994). Harvest occurs from March to May. It should be noted that exposure to cold night temperatures is assumed to predispose the rice crop to infection by P. fuscovaginae (Detry et al., 1991a;

Jaunet et al., 1996; Sharma et al., 1996), especially during the panicle initiation, panicle exertion and panicle filling stages of the plant. Therefore, sheath brown rot disease symptoms could be expected to be observed at any time from January to April. Once rice is harvested, a winter cereal; mostly wheat or pasture is grown (Smith & Humphreys,

2001). As far as the author is aware, none of these winter cereal crops or pastures that are grown in rotation with rice have ever been reported to express symptoms similar to that of sheath brown rot, nor have they been screened for the existence of P. fuscovaginae. In this study, cultivars of winter cereals such as wheat, durum wheat, barley, triticale that 46

are commonly grown in rotation with rice in the Riverina region were screened for their potential as alternative hosts for strains of P. fuscovaginae from Australia.

Occurrence of weeds is a significant problem in rice cultivation, although considered as pests themselves rather than potential hosts for diseases (Bajwa & Chauhan, 2017; Kealey

& Clampett, 2000; McIntyre, Finlayson, Ladiges, & Mitchell, 1991). Several grass species that are known to be alternative hosts for P. fuscovaginae such as; Agrostis spp.,

Poa pratensis (Miyajima, 1980), Lolium perenne, Bromus marginatus, Phleum pratense,

Phalaris arundinacea (Miyajima et al., 1983a), are found in southern NSW (Biosecurity

Queensland, 2016a; Biosecurity Queensland, 2016b; HerbiGuide, 2017; NSW

WeedWise, 2017; PlantNET, 2017b), and Leersia hexandra (Bultreys et al., 2003) grows in and around swamps and creeks in north eastern NSW (PlantNET, 2017a). In addition to these, plant species belonging to Alismataceae (Ash, Chung, McKenzie, & Cother,

2008), Commelinaceae, Cyperaceae and Pontederiaceae (Whitworth et al., 2013) that could be potential hosts of P. fuscovaginae (Zeigler & Alvarez, 1986) are found frequently associated with rice crops (Kealey & Clampett, 2000). These species or any other weeds that are commonly found in rice fields have not been accounted for in previous disease surveys. Nor have they been tested by artificial inoculation to determine if they could be potential hosts for P. fuscovaginae.

2.1.3 Occurrence of P. fuscovaginae in rice cropping regions in Australia

Surveys of disease conducted in 2010 and 2012 showed low prevalence of diseases in the rice fields of the Riverina (Adorada, 2013). This may be due to changes in cultivars or management practices, but does not necessarily indicate that the organism is no longer present in Australia. The survey of disease in 2010 was conducted within rice crops in and around the NSW DPI field research station in Leeton, and in 2012, it was conducted

47

examining 20 rice crops in MIA (Watson, 2013). Additional disease surveys were planned to be conducted in the CIA in 2013 and the MV in 2014, examining 20 crops each, just after sowing and near harvest, with focus on identifying issues with crop establishment due to common oomycetes infecting rice at germination such as Pythium spp. and Achlya spp. (NSW DPI, 2012). Therefore, in this study, an extensive survey looking for the symptoms of sheath brown rot disease in rice fields, covering the three major growing regions was conducted.

The previous surveys were conducted based on the visual symptoms of sheath brown rot, although P. fuscovaginae is a seed-borne pathogen of endophytic nature and has been reported to exist asymptomatically in all parts of infected rice plants (Adorada et al.,

2014). It has also been demonstrated that P. fuscovaginae could overwinter in infected rice plants kept indoors (Miyajima, 1980). In addition, seedlings with disease symptoms are reported to have arisen from asymptomatic seeds (Adorada et al., 2014). Thus, any new survey for the pathogen should be facilitated by a rapid and accurate diagnostic technique with the potential of application under field conditions. Loop-mediated isothermal amplification (LAMP) assay is a real-time quantitative polymerase chain reaction technique of high throughput and high accuracy (Eriko, Shoichi, & Naoto, 2008;

Li et al., 2009). It employs a set of four to six primers having specific base sequences that anneal to a specific gene region. Amplification does not rely on temperature cycling, but occurs at 60 - 65 °C, making use of a novel thermostable polymerase (Notomi et al.,

2000). Unlike PCR, LAMP is relatively tolerant to contaminants (Kaneko, Kawana,

Fukushima, & Suzutani, 2007). The template DNA could be pure DNA, heat-killed cells or live cells (Lang et al., 2014). As LAMP does not require temperature cycling, it can be performed by staff with minimal training and resources. The technique can be used in paddy fields, seed stores and in quarantine inspection stations (Hamburger et al., 2013).

48

Therefore, it has shown a strong potential to be used in the diagnosis of pathogens such as P. fuscovaginae. Ash et al. (2014) described a LAMP protocol for accurate detection of P. fuscovaginae. Their method has shown potential to be applied under field conditions, due to the set of PCR primers that can be used to amplify the target DNA fragments with high affinity under ambient temperature, and due to the portable nature and the graphical user-interface of the thermal cycler equipment.

With the continuing changes in the cropping environment due to germplasm exchanges, the presence of alternative hosts in the field and the climate change, an outbreak of disease caused by Australian strains of P. fuscovaginae can be anticipated. Thus, the presence of this pathogen in rice fields can be considered as a great risk to the Australia’s rice industry. Therefore, a management system needs to be developed to prevent this pathogen from dissemination and exerting its destructive effects. Considering the scale of production of rice in Australia, where the average size of a rice farm is around 400 hectares, the most feasible method of managing this disease is the use of host plant resistance.

2.1.4 Sources of host plant resistance to P. fuscovaginae

Tilquin and Detry (1993) demonstrated that host-plant resistance could be used to successfully control the sheath brown rot disease caused by P. fuscovaginae. The authors found that the tall cultivars grown in central Africa such as “Ambalalava” and IR50 that contain the gene eui, which is responsible for elongated uppermost internode, to be well adapted against the infection by P. fuscovaginae. In Madagascar, local cultivars that are taller than the introduced semi-dwarf cultivars were found to be the most tolerant. The introduced cultivars were susceptible to the disease though they are cold-tolerant

(Duveiller et al., 1990). In Tanzania, rice cultivar KM67 was reported to show some level

49

of resistance (Macapuguay & Mnzaya, 1988). Razak et al. (2009b) reported that the cultivars MR240, MR243, MR244, MR245, MR246, MR248 and MR249 grown in

Malaysia could be classified as moderately resistant to the disease.

In Australia, Cother et al. (2009b) reported that there appears to be a genetic component to susceptibility of rice cultivars to P. fuscovaginae, as the symptoms observed in the breeding plots in Yanco, NSW were more evident on the breeding lines with the parent

M201 in the pedigree. Since then, M201 is being gradually discontinued from the breeding programs (Ovenden, 2016), although resistance screening of rice cultivars against P. fuscovaginae specifically, is not being conducted regularly as the pathogen is considered a minor threat (Whitworth et al., 2013).

Adorada (2013) recommended that the screening of available germplasm should be performed in order to identify possible sources of resistance to sheath brown rot disease that are readily accessible in Australia, because of the limited access to currently available sources of resistance to P. fuscovaginae. Therefore, identification of sources of resistance and development of resistant cultivars to P. fuscovaginae in Australia is of greatest importance.

Eighty per cent of rice produced in Australia is of medium grain Japonica cultivars

(DAWR, 2015). A unique advantage for Australia with medium grain rice is that it is produced by only a few countries, as it is not suited to the tropical climate of a majority of rice producing counties (RGA, 2017). Australia also produces long grain Indica type cultivars such as ‘fragrant’ rice and short grain cultivars such as Koshihikari. Until recently, Amaroo and Millin were the most popular medium grain cultivars and Langi was the largest selling long grain (DAWR, 2015). Studies have shown that the cultivar

Amaroo, from which the Australian strains were isolated, is susceptible to the disease

(Adorada et al., 2013c). From 2012 onwards, Amaroo has not been recommended for

50

commercial cultivation and has been replaced by newly developed, high yielding and short season cultivars (Ovenden, 2016). However, according to the author’s knowledge, these new cultivars have not been screened for resistance against P. fuscovaginae.

Therefore, it is essential that the currently grown, popular, high-yielding rice cultivars such as Reiziq, Sherpa, Opus or Illabong (Troldahl, Snell, Dunn, Ovenden, & Pallas,

2016), should be assessed for resistance against P. fuscovaginae.

Furthermore, recently, there has been efforts to extend rice cultivation to regions of

Australia where the climate is tropical (Lyon, 2013; Sherrington, 2015). Trial rice crops have been conducted in the Ord River irrigation area of Western Australia (Sivapalan,

2014) and in rotation with pasture and pulses in the Northern Territory (ABC, 2014) and sugarcane crops in the Burdekin (Rogers, 2015) and Mackay regions (Sparkes, 2015;

Webster, 2016) in north Queensland. However, these regions are still struggling to establish rice cultivation on a large scale (Cockfield, Mushtaq, & White, 2012; Sivapalan,

2017). With this change of climate, rice-breeding goals should consider including the resistance to potentially devastating diseases such as sheath brown rot, in addition to developing high-yielding rice cultivars with less water consumption (Clarke, 2016).

Furthermore, it should be considered that wheat and other winter cereals are grown in rotation with rice in the Riverina region and the wheat cultivation in the Riverina region is an important component of the Southern cropping region, which is one the major wheat cropping regions in Australia. Although the cross-species pathogenicity of strains of P. fuscovaginae isolated from Australian has not been investigated before, there is a great potential of these to pose a significant threat to both industries. Therefore, it is essential that a screening of a potential range of hosts of these strains needs to be conducted including several cultivars of each crop.

51

2.1.5 Techniques for screening of host plant resistance

A suitable inoculation method that is practical, reliable and reproducible is essential to screen the crop cultivars for sources of resistance against diseases (Chaudhary, 1996;

Mahadevaiah et al., 2015) such as sheath brown rot. In addition, such an inoculation method must be able to discriminate between resistant and susceptible genotypes

(Challagulla, Bhattarai, & Midmore, 2015; Rosas et al., 2016). A number of inoculation methods have been utilized previously in studies for evaluating disease resistance against

P. fuscovaginae. Batoko, Kinet, Maraite, and Bouharmont (1991) observed that the susceptibility of rice cultivars to P. fuscovaginae correlates to the sensitivity to a toxin produced by the pathogen, and repeatedly demonstrated that the injection of the purified toxin to the plant can be successfully used for resistance screening (Batoko et al., 1997a;

Batoko et al., 1997b; Batoko et al., 1997c). Furthermore, soaking the seeds in the toxin prior to sowing, induced a drastic inhibition of rice seedling elongation which in turn correlates to varietal susceptibility to the disease in the field and hence, this test too was suggested as a reliable tool for screening genotype susceptibility to sheath brown rot disease caused by P. fuscovaginae (Batoko et al., 1994). However, Adorada (2013) remarked that the complexities of this assay makes it unappealing to researchers and a simpler method using the crude bacterial suspension could be evaluated as an alternative to screen genotype susceptibility to the disease.

Spraying bacterial suspension on leaf sheaths of rice plants have been used as an inoculation method for P. fuscovaginae by Zeigler and Alvarez (1987) and Cother et al.

(2009b). Jaunet et al. (1996) poured the bacterial suspension on the adaxial side of the flag leaf sheath as an inoculation method. Introducing P. fuscovaginae bacteria to the plant by injection (Cottyn et al., 2002; Duveiller et al., 1988) or by puncturing the stem or the leaf sheath with a needle contaminated with bacteria (Detry et al., 1991b) have been

52

employed to inoculate rice seedlings. Valencia-Botin et al. (2007) evaluated the aggressiveness of P. fuscovaginae on initial growth of wheat seedlings by inoculation on seedlings by spraying on, and puncture of seedlings, as well as inoculating seeds by vacuum-infiltration.

Recent studies by Adorada (2013) emphasises the use of crude bacterial cell suspensions introduced to the plant by pricking with a pin tip (pinprick method) and the resulting disease measured based on the extent of symptoms can be used for screening host plant resistance against P. fuscovaginae. Adorada et al. (2013c) demonstrated that using the pinprick method at the late booting or early panicle exertion stage is the most suitable for identifying sources of resistance to P. fuscovaginae in mature plants, while the seed- soaking method at the seed to germination stage is useful for detection of early disease resistance. For the pinprick method, a single point assessment of disease severity at 14 days after inoculation (DAI) could be used instead of the area under disease progress curve (AUDPC) values to classify genotypes. For the seed-soaking method, Adorada et al. (2013c) established an index of reduction in seedling height 10 days after seed soaking, for the classification of the genotypes reaction to the disease. They also suggested that the disease severity at 14 DAI could be used for assessing the response of the different cultivars to inoculation by P. fuscovaginae on both seedlings and mature plants and the mean disease severity values at 14 DAI can be used to classify the resistant phenotypes as Resistant (R), Moderately resistant /Intermediate (I) or Susceptible (S). Furthermore, they recommended that a 107 cfu/mL inoculum concentration as best for discriminating between resistant and susceptible genotypes as all the growth stages of rice were susceptible to the pathogen at this concentration. The authors also proposed that the host- strain interactions (as depicted by host resistance levels) could differ depending on the growth stage of the plant, as there could be different mechanisms of resistance to P.

53

fuscovaginae in rice at each growth stage. Therefore, host resistance of a certain cultivar should be tested at both the early and mature stages of plant growth as they found no correlation between the resistance classifications of cultivars between the two inoculation methods. Rice cultivar Amaroo was used by Adorada et al. (2013c) as a positive control which indicated a successful inoculation by developing the symptoms of sheath brown rot.

2.2 Materials and Methods

2.2.1 Strains of P. fuscovaginae

Initially, twenty-nine bacterial isolates that were collected from rice plants showing symptoms of sheath brown rot were obtained from the culture collection at Charles Sturt

University (CSU), Wagga Wagga, NSW and sub-cultured. In addition, eleven strains of

P. fuscovaginae identified in previous studies were sub-cultured from the culture collection held at CSU. Then, the sub-cultures were screened by their biochemical characteristics using the BACTID bacteriological identification system (Black, Holt, &

Sweetmore, 1996a) and LOPAT tests for fluorescent Pseudomonas spp. (Lelliott, Billing,

& Hayward, 1966) and their pathogenicity on rice plants by pin-prick inoculation of known susceptible rice cv. Amaroo.

The biochemical characteristics tested were, Gram stain, fluorescence on KB medium under UV light (F), presence of cytochrome C oxidase enzyme (O), presence of nitrate reductase enzyme (NIT), presence of dehydrogenase enzymes for oxidation (Triphenyl tetrazolium chloride (TTC) test), oxidative/fermentative metabolism of glucose, presence of arginine dihydrolase enzyme (A), presence of levan sucrase enzyme (L), pectinolytic activity (P) and hypersensitivity on tobacco plants (THS).

54

From the eleven strains known to be P. fuscovaginae, five were isolated in Australia;

DAR77138, DAR77320, DAR77795, DAR77797, DAR77800 and five international strains of P. fuscovaginae; ICMP5940, ICMP9995, ICMP9996, ICMP9998, ICMP10137, were used for the inoculation on rice and winter cereal crop cultivars. In addition to this, five strains P. fuscovaginae, UPB0526, UPB0588, UPB0589, UPB1013 and UPB0407, that were isolated from plants other than rice, internationally, were imported from the

Université Catholique de Louvain, Belgium under the permit to import quarantine material numbered IP14013852 subject to the Quarantine Act 1908 Section 13 (2AA) and stored at CSU. The background information of all the above-mentioned strains of bacteria are given in Table 2.1.

55

Table 2.1: Bacteria strains used in this study

Isolate Host Origin

MB286 ICMP5939 NCPPB3598 LMG2192 Rice Japan MB281 ICMP5940 LMG2158 Rice Japan MB287 ICMP9995 CFBP2805 LMG5742 Rice Burundi MB282 ICMP9996 Rice Madagascar MB289 ICMP9998 UPB0736 LMG12426 Rice Madagascar MB291 ICMP10137 NCPPB3757 Wheat Brazil MB178 DAR77318 Rice Australia MB179 DAR77795 ICMP17628 Rice Australia MB181 DAR77797 Rice Australia MB189 DAR77320 Rice Australia MB194 DAR77800 ICMP17627 Rice Australia MB247 Rice Cambodia MB253 Rice Cambodia MB248 Rice Cambodia MB249 Rice Cambodia MB232 Rice Cambodia MB250 Rice Cambodia MB234 Rice Cambodia MB235 Rice Cambodia MB236 Rice Cambodia MB251 Rice Cambodia MB239 Rice Cambodia MB240 Rice Cambodia MB246 Rice Cambodia MB245 Rice Cambodia DLA99 Rice Australia DLA100 Rice Australia DLA101 Rice Australia DLA102 Rice Australia

56

Isolate Host Origin

DLA103 Rice Australia DLA104 Rice Australia DLA105 Rice Australia DLA106 Rice Australia DB4S1 Rice Australia DB4S2 Rice Australia DB4S3 Rice Australia DB4S4 Rice Australia DB4S5 Rice Australia DB4S6 Rice Australia DB4S7 Rice Australia UPB0588 Wheat Mexico UPB0526 Wheat Mexico UPB1013 Wheat Nepal UPB0407 Swamp rice grass Burundi UPB0589 Wheat Mexico

Source: (Adorada, 2013; Cother et al., 2009b)

2.2.2 Selection of cultivars

Based on the aspects of popularity among the growers and interests of the breeders, ten cultivars each from rice and wheat and two cultivars each from durum wheat, triticale and barley were selected across a range of genetic characteristics, to counter any bias towards disease susceptibility / resistance, upon consulting the rice and winter cereal breeders of

NSW DPI.

57

Table 2.2: Details of the selected cultivars of rice and the winter cereals

Crop Cultivar Growth and yield characteristics Disease resistance* Rice Amaroo Medium grain, semi-dwarf, high yielding, no longer in commercial production, Japonica type Susceptible to P. fuscovaginae Rice Doongara Long grain, semi-dwarf, high amylose content, ‘hard’ quality of cooking, susceptible to cold temperatures during the reproductive period, resistant to lodging Rice Reiziq Medium grain, semi-dwarf, Japonica type, high yield potential, strong establishment vigour, resistant to lodging, moderately susceptible to cold temperatures during the reproductive period, the most widely grown variety in Australia Rice Koshihikari Short grain, tall, ‘Japanese’ quality of cooking, low yielding, Australian selection of a very old traditional Japonica type, a very pubescent variety, susceptible to lodging

Rice Langi Long grain, semi-dwarf, ‘soft jade’ quality of cooking, low amylose content, moderate establishment vigour and cold stress tolerance, moderately resistant to lodging, only grown in the MIA and CIA Rice Quest Medium grain, semi-dwarf, short season, Japonica type, no longer in commercial production Susceptible to P. fuscovaginae Rice Opus Short grain, semi-dwarf, ‘Japanese’ quality of cooking, high yielding, Japonica type Australian cultivar, moderate establishment vigour, resistant to lodging, moderately resistant to cold temperatures during the reproductive period, a pubescent variety, only grown in the Murray Valley Rice Kyeema Long grain, tall, ‘fragrant’ quality of cooking, low yielding, Indica type

58

Crop Cultivar Growth and yield characteristics Disease resistance* Rice Illabong Medium grain, semi-dwarf, ‘arborio’ quality of cooking, high yield potential, moderate establishment vigour, lower germination percentage, moderate cold stress tolerance, moderately resistant to lodging Rice Sherpa Medium grain, semi-dwarf, high cold stress tolerance, moderate establishment vigour, high yield potential, resistant to lodging Wheat EGA Gregory Early to mid season, medium to slow maturity, Australian premium white (APW) quality of Stem rust MR, Stripe grain rust MR, Leaf rust MR, Yellow leaf spot MS-S Wheat EGA A mid to long season winter wheat, April sowing, mid-season maturity once cold requirement Stem rust MR-MS, Wedgetail is met, APW quality of grain Stripe rust MR-MS, Leaf rust MS, Yellow leaf spot MS-S Wheat Spitfire Early to mid season maturity, medium plant height, Australian Hard (AH) quality of grain Stem rust MR, Stripe rust MR, Leaf rust MS, Yellow leaf spot MS-S Wheat Chara Short coleoptile, slow early vigour, AH quality of grain, ideal for southern NSW, high yields Stem rust MR-MS, under irrigation Stripe rust MS-S, Leaf rust MS, Yellow leaf spot MS-S Wheat Lincoln Mid season, semi dwarf, AH quality of grain Stem rust MR, Stripe rust R-MR, Leaf rust MR, Yellow leaf spot MR-MS

59

Crop Cultivar Growth and yield characteristics Disease resistance* Wheat Mace Medium height, AH quality of grain Stripe rust S, Stem rust MR, Stripe rust S-VS, Leaf rust MR, Yellow leaf spot MR-MS Wheat Wyalkatchem Short stature, large grains, APW quality of grain Stem rust MS, Stripe rust S, Leaf rust MS, Yellow leaf spot MR Wheat Ventura Main season spring wheat, semi-dwarf, ideal for NSW, AH quality of grain Stem rust R-MR, Leaf rust R, Stripe rust MS, Crown rot MS-S, Yellow leaf spot S, Septoria tritici blotch MS Wheat Sunvale Main season, Australian Prime Hard (APH) of grain Stem rust R, Leaf rust MR-MS, Stripe rust MR, Crown rot MS-S, Common root rot MS Wheat Rosella Widely adapted winter wheat, good seedling vigour, mid-season maturity once cold Stripe rust MR-MS, Leaf requirement is met, Australian Standard White (ASW) quality of grain and stem rust MR-MS, Crown rot MS-S Durum EGA Bellaroi Australian Premium Durum (APDR) quality of grain, grain yield is typically better in Stem, leaf and stripe rust Northern NSW, best for irrigated production in southern NSW. MR, Yellow leaf spot MR-MS, Common root rot MR, Crown rot VS

60

Crop Cultivar Growth and yield characteristics Disease resistance* Durum Tjilkuri Mid-season flowering, moderate height improved grain yield, APDR quality of grain Stem, stripe and leaf rust R, Yellow leaf spot MS, Black point S Barley Schooner Malting quality of grain, lower-yielding Powdery mildew S, Leaf rust VS-S, Scald S Barley Commander Malting quality of grain, high yield potential than Schooner Scald S Triticale Hawkeye Mid season, excellent yield stability and adaptability over a broad range of environments, Stripe rust MR - MS good early vigour, high yield potential, excellent physical quality of grain Triticale Tahara Main season, wide adaptability, high yielding Stripe rust MS

*VS - Very susceptible, S - susceptible, MS - Moderately susceptible, MR - moderately resistant, R - Resistant

Source: (Matthews, McCaffery, & Jenkins, 2013; Troldahl et al., 2016)

61

2.2.3 Plant growth conditions

Rice, wheat, barley and triticale seeds were treated with 5% NaOCl for 5 minutes with continuous agitation to disinfect the surface of the seed and then washed thoroughly with sterile distilled water (SDW). Subsequently, the seeds were soaked overnight in SDW to induce germination. For rice and wheat planted in pots, seeds were allowed to germinate and grow and under sterile conditions for two weeks at 28 °C. Three seedlings were planted in plastic pots (18 cm dia.) containing steam-pasteurised sandy clay loam (1:1) soil. Rice plants were maintained in the glasshouse at 30 - 35 °C day and 20 - 25 °C night temperatures, under flooded soil conditions. Wheat plants were maintained in the glasshouse at 25 °C day and 15 °C night temperatures, under well-drained soil conditions.

Nitrogen, phosphorous and potassium nutrients were supplied by adding urea, super phosphate and muriate of potash, at the recommended rates. The experiment had three replicates.

2.2.4 Preparation of inoculum and inoculation techniques

Bacterial cultures were grown in nutrient broth (NB) at 28 °C, shaking at 120 rpm, until

9 they reached 10 cfu/ml (OD600 = 1), and then they were diluted 1:10, twice, in SDW, to a final concentration of 107 cfu/ml. Seeds were inoculated by the seed soaking method described previously (Adorada et al., 2013b). Prior to inoculation, the surfaces of the seeds were disinfected and placed on sterile filter paper disks in petri plates, and then inoculated with 10 mL of 107 cfu/ml bacterial suspension and sealed with parafilm.

Inoculated seeds were incubated at 28 °C for 10 days. Each treatment consisted of four replicates and treatment with SDW was used as a control.

62

Rice and wheat plants were inoculated at the panicle exertion/heading (Zadoks Code DC:

50 - 60) stage on the flag leaf sheath by using the pinprick method and were immediately covered for 24 hours to increase humidity. Afterwards, the rice plants were kept in the glasshouse at 30 - 35 °C / 20 - 25 ℃ day/night temperatures, and wheat plants were kept in the glasshouse at 25 °C day and 15 °C night temperatures. A treatment with SDW was used as a control.

2.2.5 Data collection

For the seed-soaking assay, the shoot length (from the base of the plant to the tip of the longest leaf) and root length (from the base of the plant to the tip of the longest root) and the number of roots of each seedling were measured after ten days from inoculation. In the pinprick assay, the disease symptoms were measured fourteen days after inoculation, based on length of the brown coloured lesions on the inoculated sheath with respect to the total length of the sheath.

2.2.6 Data analysis

For the seed soaking assays, measurements of the three parameters: seedling height

(length in mm from the base of the seedling to the tip of the longest leaf), length of the roots (length in mm from the base of the seedling to the tip of the longest root), and number of roots were taken. An analysis of variance (ANOVA) was performed for both the seedling height and the length of the roots. In the pinprick assays, the length of the lesion at the point of inoculation (POI) on the flag leaf sheath, width of the lesion at the

POI, the length of the flag leaf sheath, width of the flag leaf sheath, were measured in mm. The number of discoloured seeds in each head and the number of total seeds in each head were also counted. An ANOVA was performed for the percent disease incidence

63

(the length of lesion at POI on sheath expressed as a percentage of the total length of the sheath). Data were transformed by either square-root transformation or arc-sine transformation as appropriate, in order to reduce variation and to obtain a normal distribution before analysis of variance. Means of the treatments of ten P. fuscovaginae strains and the control for each variety were compared and separated by Tukey’s test at

5% confidence interval using SAS software, version 6.12 (SAS Institute, USA). Also, means for each cultivar were compared and separated by Tukey’s test at 5% confidence interval) for each crop.

2.2.7 Survey of sheath brown rot disease

A survey for disease symptoms of sheath brown rot was conducted during mid-October to mid-December 2013 in wheat fields and in late-February to late-March 2014 in the rice fields, when the particular fields were at heading stage, in the three major rice-growing regions (MIA, CIA and MV). Sampling points were selected at five farms, which had practiced rice-wheat in rotation during the last five seasons, from each growing region

(Figure 2.2). Farms were selected randomly to represent a range of cultivars and to cover the entire growing area as much as possible. The crops were observed for the symptoms resembling sheath brown rot and samples were taken. At each sampling point where a sample with symptoms was taken, another sample without symptoms was taken. The whole plant was removed from the soil and then put into clean room manufactured zip- lock bags, and kept on ice until taken to the lab. In the lab, the samples were kept in a cool room at 4 °C, until processed.

64

Figure 2.2: Sampling locations of rice-wheat cropping rotations in the Murrumbidgee Irrigation area (MIA), Coleambally Irrigation Area (CIA) and Murray Valley (MV) regions

2.2.8 Preparation of plant samples collected during the survey

The plant samples were processed within 24 hours. From each sample, three panicles/heads with flag leaves and the leaf sheaths were taken and aseptically cut in to 3

- 4 cm long pieces using scissors disinfected with 70% ethanol. They were put in 50 mL of sterile pre-prepared saline wash solution containing 0.9% (W/V) NaCl and 0.2% (V/V)

Tween 80 (Renouf, Claisse, & Lonvaud-Funel, 2005), in 250 mL clean room manufactured plastic containers, and stored temporarily on ice in a cool box. Total epiphytic microbial population was released into the solution by shaking the containers at 120 rpm for 3 hours on a rotary shaker at 25 °C (Setati, Jacobson, Andong, & Bauer,

2012). The wash solutions containing the epiphytic microbiome extracted from the plant samples were transferred to sterile 50 mL conical bottomed Falcon tube and centrifuged at 5000x g for 10 minutes. The pellet was resuspended in 1 mL of sterile saline wash solution and stored in 2 mL Eppendorf® microcentrifuge tubes at -20 °C. 65

2.2.9 Use of Loop mediated isothermal amplification (LAMP) technique

Isothermal master mix containing the Pf8 primer set for LAMP protocol described by Ash et al. (2014) was prepared and stored in a sterile and darkened polypropylene tube, to be used in the field and in the laboratory. In the field, plant tissue samples with disease symptoms resembling sheath brown rot, such as discoloured rice seeds and leaf sheath with brown lesions were shredded with a sterile pair of scissors and then shaken by hand in a sterile polypropylene tube with either; (a). sterile wash solution or (b). SDW. Then 1

µL of the suspension was taken as the template, in to an Optigene® LAMP tube containing

11 µL of the LAMP reaction mixture which is described below.

Constituent Volume Pf8 F3 primer (10 µM) 0.384 µL Pf8 B3 primer (10 µM) 0.384 µL Pf8 FIP primer (1 µM) 0.384 µL Pf8 BIP primer(1 µM) 0.384 µL Pf8 LoopF (10 µM) 0.192 µL Pf8 LoopB (10 µM) 0.192 µL Isothermal Master Mix (Optigene®) 7.2 µL Sterile distilled water 1.8 µL

Source : Ash et al. (2014)

The final LAMP reactions of 12 µL volume were amplified as a batch of eight; the numbers of wells in each Genie PCR tube strip (GeneWorks Pty Ltd), and was run in a

Genie II® (Optigene Ltd) instrument. The temperature profile was 65 °C for 30 minutes and 85 °C for 5 mins. A negative control containing a LAMP reaction with no template

66

and a positive control containing a bacterial suspension of P. fuscovaginae strain

DAR77795 (107 cfu/mL) were included in each run.

2.2.10 Isolation and screening of putative P. fuscovaginae from diseased plant

samples

Parts of plant samples showing symptoms similar to that of sheath brown rot disease, such as discoloured rice seeds and leaf sheath with brown lesions, were first dipped in 70% ethanol for five minutes and washed thoroughly with SDW. Then, they were crushed with sterile forceps and scalpels and placed on petri plates containing NA media, and incubated at 28 °C for several days, until the bacterial colonies appeared. Then bacterial colonies were further isolated several times based on their growth characteristics to obtain pure isolates. The bacterial isolates were screened for florescence using KB medium. Bacterial isolates identified as florescent Pseudomonas spp. were screened further using the biochemical tests in the BACTID bacteriological identification system and LOPAT tests.

2.3 Results

2.3.1 Screening of existing collection of putative P. fuscovaginae isolates

Out of the fourty isolates screened, fourteen isolates, namely; MB286, MB281, MB287,

MB253, MB248, MB232, MB250, MB234, MB235, MB236, MB251, MB246 and

MB245 were clustered by the BACTID system as ‘Florescent Pseudomonads’, of which the biochemical characteristics are; Gram negative, bacilli, florescent, non-nitrate reducing, oxidase positive, TTC positive, oxidation positive and fermentation negative.

Six isolates, namely, MB282, MB289, MB291, MB178, MB240 and DLA103 were grouped as ‘Agrobacterium and non-fluorescent Pseudomonads’. Five isolates, namely,

MB247, MB239, DLA99, DLA104 and DLA106 came together as ‘Yellow, mostly non-

67

pathogenic bacteria’. Nine isolates, namely, DLA102, DLA105, DB4S1, DB4S2, DB4S3,

DB4S4, DB4S5, DB4S6 and DB4S7, were collected as ‘Yellow Erwinia’. Out of the forty isolates, six isolates, namely, MB179, MB181, MB189, MB194, MB249 and DLA101 did not belong to any group. The details of the results are presented in Table 2.3.

68

Table 2.3: Identification of putative P. fuscovaginae isolates by BACTID

Isolate Shape Gram F NIT TTC Oxi. Ferm. L O P A THS BACTID Pathogenicity Fluorescent MB286 bacilli - + - + + - - + - + + + Pseudomonads Fluorescent MB281 bacilli - + - + + - - + - + - + Pseudomonads Fluorescent MB287 bacilli - + - + + - - + + + + + Pseudomonads Agrobacterium & non- MB282 bacilli - - - + + - - + + + + fluorescent + Pseudomonads Agrobacterium & non- MB289 bacilli - - - + + - - + + + - fluorescent + Pseudomonads Agrobacterium & non- MB291 bacilli - - + + + - - + + + + fluorescent + Pseudomonads Agrobacterium & non- MB178 bacilli - - - + + - - + + + + fluorescent - Pseudomonads MB179 bacilli - - - + + + - + + + + No match + 69

Isolate Shape Gram F NIT TTC Oxi. Ferm. L O P A THS BACTID Pathogenicity

MB181 bacilli - - - + + + - + + + + No match +

MB189 bacilli - - + + + + - + + + + No match + MB194 bacilli - - - + + + - + + + + No match + Yellow, mostly non- MB247 bacilli - - - + + - - - - + - + pathogenic bacteria Fluorescent MB253 bacilli - + - + + - - + + + - - Pseudomonads Fluorescent MB248 bacilli - + - + + - - + - + - - Pseudomonads MB249 bacilli - + - + + + - + - + - No match + Fluorescent MB232 bacilli - + - + + - - + + + - + Pseudomonads Fluorescent MB250 bacilli - + - + + - - + + + - + Pseudomonads Fluorescent MB234 bacilli - + - + + - - + + + - - Pseudomonads Fluorescent MB235 bacilli - + - + + - - + - + - + Pseudomonads

70

Isolate Shape Gram F NIT TTC Oxi. Ferm. L O P A THS BACTID Pathogenicity Fluorescent MB236 bacilli - + + + + - - + - + - - Pseudomonads Fluorescent MB251 bacilli - + - + + - - + + + - + Pseudomonads Yellow, mostly non- MB239 cocci - - - + + - - - - + - - pathogenic bacteria Agrobacterium & non- MB240 bacilli - - + + + - - + - - - fluorescent - Pseudomonads Fluorescent MB246 bacilli - + - + + - - + - + - + Pseudomonads Fluorescent MB245 bacilli - + - + + - - - - + - + Pseudomonads Yellow, mostly non- DLA99 bacilli - - - + + - + + - - - - pathogenic bacteria Fluorescent DLA100 bacilli - + - + + - + - + + - + Pseudomonads DLA101 cocci - - - + + + + - + - + No match + DLA102 cocci - - - + + + + - - - - Yellow Erwinia -

71

Isolate Shape Gram F NIT TTC Oxi. Ferm. L O P A THS BACTID Pathogenicity Agrobacterium & non- DLA103 bacilli ------+ - - + - fluorescent - Pseudomonads Yellow, mostly non- DLA104 bacilli - - - + + ------pathogenic bacteria DLA105 bacilli - - - + + + + - - - + Yellow Erwinia + Yellow, mostly non- DLA106 bacilli - - - + - - - - - + - - pathogenic bacteria DB4S1 bacilli - - - + + + + - - - + Yellow Erwinia + DB4S2 bacilli - - - + + + + - - - - Yellow Erwinia - DB4S3 cocci + - - + + + + - - - + Yellow Erwinia - DB4S4 bacilli - - - + + + + - - - + Yellow Erwinia - DB4S5 bacilli - - - + + + + - - - + Yellow Erwinia - DB4S6 bacilli - - - + + + + - - - - Yellow Erwinia - DB4S7 bacilli - - - + + + + - - - - Yellow Erwinia +

Gram: Gram stain, F: fluorescence on KB medium under UV light, O: presence of cytochrome C oxidase enzyme, NIT: presence of nitrate reductase enzyme, TTC: Triphenyl tetrazolium chloride test for presence of dehydrogenase enzymes for oxidation, Oxi./Ferm.: oxidative/fermentative metabolism of glucose, A: presence of arginine dihydrolase enzyme, L: presence of levan sucrase enzyme, P: pectinolytic activity, THS: hypersensitivity on tobacco plants 72

2.3.2 Response of rice cultivars to inoculation with P. fuscovaginae

All ten of the rice cultivars were susceptible to infection with strains of P. fuscovaginae by the pinprick method on leaf sheaths at heading. In general, the virulence of the tested strains ranged from 7.86% to 16.66%, as expressed by the severity of the disease symptoms on the tested rice cultivars. However, there were differences in disease responses shown by the rice cultivars to each strain, as depicted by percent sheath lesion

(length of leaf sheath lesion at the point of inoculation in mm as a proportion of the total area of the flag leaf sheath in mm). There were significant (α = 0.05, Pr > F = 0.001, MSD

= 43.88) levels of interactions between the rice cultivars and strains of P. fuscovaginae.

The cultivar x strain interaction cv. Koshihikari x DAR77800 demonstrated the highest levels of disease response. A classification of the nature of these interactions at 5% significance level are presented in Table 2.4. Interestingly, ICMP 10137, which is a strain isolated from wheat in Brazil, hence known as a wheat-infecting P. fuscovaginae strain, also showed pathogenicity and virulence on rice.

73

Table 2.4: Disease response (percent sheath lesion length) according to the interactions between rice cultivars and strains of Pseudomonas fuscovaginae

Rice P. fuscovaginae strain Mean lesion Tukey Rice P. fuscovaginae strain Mean lesion Tukey variety accession number length grouping* variety accession number length grouping* Koshihikari DAR77800 76.96 A Kyeema ICMP9996 16.18 BC Koshihikari ICMP10137 48.62 AB Reiziq DAR77318 15.90 BC Doongara ICMP5940 42.51 ABC Doongara ICMP9998 15.18 BC Amaroo ICMP9998 42.42 ABC Amaroo ICMP9996 14.82 BC Koshihikari DAR77318 36.73 ABC Koshihikari ICMP9998 14.61 BC Koshihikari DAR77320 30.92 BC Illabong ICMP9998 14.06 BC Amaroo DAR77800 29.86 BC Quest DAR77795 13.82 BC Kyeema DAR77318 29.73 BC Kyeema ICMP9998 13.46 BC Quest DAR77318 28.65 BC Langi ICMP9996 13.26 BC Koshihikari DAR77795 27.29 BC Koshihikari ICMP9996 12.44 BC Koshihikari DAR77797 26.91 BC Illabong ICMP9995 10.94 BC Koshihikari ICMP9995 25.03 BC Koshihikari ICMP5940 10.78 BC Langi DAR77320 24.98 BC Kyeema DAR77320 10.46 BC Kyeema DAR77795 22.54 BC Amaroo ICMP9995 10.30 BC Amaroo DAR77318 22.37 BC Opus ICMP9995 10.14 BC Illabong DAR77795 21.99 BC Quest ICMP9996 9.97 BC Langi ICMP10137 21.14 BC Langi DAR77795 9.61 BC Doongara ICMP10137 20.73 BC Kyeema DAR77800 9.38 BC Amaroo DAR77797 20.50 BC Sherpa ICMP9998 9.26 BC Opus DAR77318 19.47 BC Langi ICMP5940 16.72 BC 74

Rice P. fuscovaginae strain Mean lesion Tukey Rice P. fuscovaginae strain Mean lesion Tukey variety accession number length grouping* variety accession number length grouping* Opus DAR77800 8.86 BC Doongara DAR77795 6.24 BC Reiziq ICMP9998 8.76 BC Kyeema ICMP10137 6.16 BC Opus DAR77320 8.64 BC Sherpa DAR77795 5.86 BC Quest DAR77800 8.63 BC Doongara DAR77320 5.61 BC Quest ICMP9998 8.30 BC Opus ICMP9996 5.41 BC Langi DAR77800 8.28 BC Sherpa ICMP10137 5.33 BC Reiziq ICMP9996 8.06 BC Amaroo DAR77795 5.17 BC Sherpa ICMP9995 8.00 BC Illabong DAR77797 5.08 BC Kyeema ICMP9995 7.87 BC Illabong DAR77320 5.07 BC Quest ICMP5940 7.86 BC Amaroo DAR77320 5.05 BC Langi ICMP9995 7.58 BC Illabong DAR77318 4.99 BC Illabong ICMP5940 7.54 BC Sherpa DAR77320 4.97 BC Langi ICMP9998 7.49 BC Illabong ICMP10137 4.94 BC Illabong DAR77800 7.24 BC Amaroo ICMP5940 4.66 C Opus DAR77795 7.12 BC Reiziq DAR77795 4.54 C Kyeema DAR77797 7.00 BC Illabong ICMP9996 4.51 C Doongara ICMP9995 6.87 BC Amaroo ICMP10137 4.39 C Quest ICMP10137 6.78 BC Reiziq DAR77800 3.97 C Doongara DAR77800 6.76 BC Quest ICMP9995 3.95 C Opus ICMP9998 6.76 BC Doongara DAR77797 3.90 C Sherpa DAR77800 6.66 BC Opus ICMP5940 3.74 C Reiziq DAR77797 6.49 BC Sherpa ICMP5940 3.65 C

75

Rice P. fuscovaginae strain Mean lesion Tukey Rice P. fuscovaginae strain Mean lesion Tukey variety accession number length grouping* variety accession number length grouping* Sherpa ICMP9996 3.63 C Illabong SDW 0.00 C Reiziq ICMP5940 3.25 C Koshihikari SDW 0.00 C Langi DAR77318 3.24 C Kyeema SDW 0.00 C Reiziq ICMP9995 3.18 C Langi SDW 0.00 C Doongara ICMP9996 3.15 C Opus SDW 0.00 C Sherpa DAR77318 2.98 C Quest SDW 0.00 C Sherpa DAR77797 2.68 C Reiziq SDW 0.00 C Reiziq DAR77320 2.65 C Sherpa SDW 0.00 C Reiziq ICMP10137 2.59 C Opus DAR77797 2.55 C Langi DAR77797 2.54 C Doongara DAR77318 2.33 C Quest DAR77320 2.09 C Kyeema ICMP5940 1.99 C Opus ICMP10137 1.94 C Quest DAR77797 1.11 C Amaroo SDW 0.00 C Doongara SDW 0.00 C

*Significant differences at α = 0.05, Pr > F = 0.001, MSD = 43.88 SDW = Sterile distilled water

76

2.3.3 Response of wheat cultivars to inoculation with P. fuscovaginae

The results of the inoculation of wheat seeds with P. fuscovaginae 107 cfu/mL by the seed-soaking method indicated that all the cultivars examined are highly susceptible to infection by all ten strains of P. fuscovaginae. However, it should be noted that there were significant interactions between the P. fuscovaginae strains and the wheat cultivars (α =

0.05, Pr > F = 0.001, MSD = 2.10). The growth response of inoculated wheat seeds

(average seedling height) according to the interactions between the P. fuscovaginae strains and the wheat cultivars are presented in Table 2.5. The least significant effect on seedling height was demonstrated by ICMP5940 on cv. Mace, comparable to that of SDW treatment. The most significant reduction of seedling height was shown by DAR77797, on both cv. Rosella and cv. Lincoln.

77

Table 2.5: Growth response of wheat cultivars inoculated with Pseudomonas fuscovaginae 107 cfu/mL by seed-soaking method, showing the interactions between wheat cultivars and strains of P. fuscovaginae

Wheat variety P. fuscovaginae strain accession Mean seedling height* Tukey grouping** number Chara SDW 11.54 A Sunvale SDW 10.69 BA EGA Wedgetail SDW 9.90 BA Lincoln SDW 9.85 BA Wyalkatchem SDW 9.41 B EGA Gregory SDW 8.97 B Mace SDW 8.97 B Ventura SDW 8.84 B Mace ICMP5940 8.67 B Rosella SDW 8.60 B Spitfire SDW 6.32 C Spitfire DAR77800 5.98 C Ventura ICMP5940 5.96 D C Chara ICMP5940 5.80 D C E Mace DAR77800 5.43 D F C E Ventura DAR77800 5.42 D F C E EGA Gregory DAR77800 5.29 D F C E Wyalkatchem DAR77800 5.03 G D F C E Wyalkatchem ICMP5940 4.98 G D F C E Ventura DAR77320 4.80 G D F C E H

78

Wheat variety P. fuscovaginae strain accession Mean seedling height* Tukey grouping** number Sunvale DAR77800 4.49 G D F C I E H Mace DAR77320 4.34 G D F J C I E H EGA Wedgetail ICMP5940 4.27 G D K F J C I E H Chara DAR77800 4.23 G L D K F J C I E H EGA Gregory ICMP5940 4.19 G L D K F J I E H Sunvale ICMP5940 3.98 G L D K F J M I E H Lincoln DAR77800 3.87 N G L D K F J M I E H EGA Wedgetail DAR77800 3.87 N G L D K F J M I E H Ventura DAR77318 3.81 N G L O K F J M I E H Wyalkatchem ICMP9995 3.80 N G L O K F J M I E H Wyalkatchem DAR77318 3.76 N G L O K F J M I E H P Spitfire ICMP5940 3.74 N G L O K F J M I E H P Wyalkatchem ICMP10137 3.64 N G L O K F J M I Q H P Lincoln ICMP5940 3.61 N G L O K F J M I Q H P Mace ICMP9996 3.57 N G L O K F J M I Q H P Lincoln DAR77797 3.11 N G L O K R J M I Q H P Rosella ICMP5940 2.99 N G L O K R J M I Q S H P EGA Wedgetail DAR77318 2.93 N G L O K R J M I T Q S H P Wyalkatchem DAR77320 2.74 N U L O K R J M I T Q S H P Mace DAR77318 2.71 N U L O K R J M I T Q S H P EGA Gregory DAR77318 2.62 N U L O K R J M I T Q S P EGA Gregory DAR77320 2.61 N U L O K R J M I T Q S P

79

Wheat variety P. fuscovaginae strain accession Mean seedling height* Tukey grouping** number Chara DAR77318 2.55 N U L O K R J M I T Q S P Ventura ICMP9995 2.43 N U L O K R J M I T Q S P Spitfire DAR77320 2.40 N U L O K R J M I T Q S P EGA Gregory ICMP9995 2.36 N U L O K R J M T Q S P Spitfire DAR77318 2.35 N U L O K R J M T Q S P Ventura ICMP9996 2.31 N U L O K R J M T Q S P EGA Wedgetail DAR77320 2.18 N U L O K R M T Q S P Rosella DAR77800 2.13 N U L O R M T Q S P Lincoln ICMP9995 2.07 N U O R M T Q S P Ventura DAR77795 2.06 N U O R M T Q S P Sunvale DAR77320 2.00 N U O R M T Q S P Lincoln DAR77320 1.97 N U O R M T Q S P Ventura ICMP10137 1.94 N U O R M T Q S P Sunvale ICMP9995 1.94 N U O R M T Q S P EGA Wedgetail ICMP9995 1.92 N U O R M T Q S P Wyalkatchem ICMP9996 1.91 N U O R M T Q S P Chara ICMP10137 1.83 N U O R T Q S P Mace DAR77795 1.82 N U O R T Q S P Rosella DAR77320 1.81 N U O R T Q S P EGA Gregory ICMP9996 1.81 N U O R T Q S P Chara ICMP9995 1.78 N U O R T Q S P Chara DAR77795 1.76 U O R T Q S P

80

Wheat variety P. fuscovaginae strain accession Mean seedling height* Tukey grouping** number Spitfire ICMP9995 1.74 U O R T Q S P Rosella DAR77795 1.74 U O R T Q S P Chara ICMP9996 1.71 U O R T Q S P EGA Wedgetail ICMP9996 1.69 U R T Q S P EGA Wedgetail ICMP10137 1.58 U R T Q S EGA Gregory ICMP10137 1.55 U R T Q S Spitfire ICMP9996 1.45 U R T S Sunvale DAR77318 1.43 U R T S Lincoln DAR77318 1.42 U R T S Rosella ICMP9995 1.40 U R T S Wyalkatchem DAR77795 1.38 U R T S Spitfire ICMP10137 1.37 U R T S Mace ICMP 10137 1.36 U R T S Ventura DAR77797 1.34 U R T S Spitfire DAR77795 1.26 U R T S Rosella DAR77318 1.26 U R T S Lincoln ICMP10137 1.22 U R T S Lincoln DAR77795 1.21 U R T S Rosella ICMP10137 1.16 U R T S Sunvale DAR77795 1.13 U R T S Mace DAR77797 1.12 U R T S Wyalkatchem DAR77797 1.10 U R T S

81

Wheat variety P. fuscovaginae strain accession Mean seedling height* Tukey grouping** number Chara DAR77797 1.07 U R T S Lincoln ICMP9996 1.06 U R T S EGA Wedgetail DAR77795 1.03 U R T S Mace ICMP9995 1.03 U R T S Sunvale ICMP10137 1.02 U R T S EGA Wedgetail DAR77797 0.99 U T S Sunvale DAR77797 0.96 U T S Spitfire DAR77797 0.91 U T S Rosella ICMP 9996 0.91 U T S EGA Gregory DAR77795 0.89 U T S EGA Gregory DAR77797 0.86 U T Sunvale ICMP9996 0.84 U T Lincoln DAR77797 0.81 U Rosella DAR77797 0.81 U

*Average seedling height in mm, square – root transformed. **Significant differences at α = 0.05, Pr > F = 0.001, MSD = 2.10 SDW = Sterile distilled water

82

There were no significant interactions (α = 0.05, Pr > F = 0.8962) observed between the wheat cultivars and P. fuscovaginae strains in the pinprick assay, which was in contrast to the seed-soaking assay. When the leaf sheaf of eight wheat cultivars (Chara, Ventura,

Mace, Spitfire, EGA Gregory, Lincoln, Sunvale and Wyalkatchem) was inoculated using the pinprick method, all were equally susceptible to infection with P. fuscovaginae. The other two wheat cultivars, namely EGA Wedgetail and Rosella failed to produce heads to be inoculated. There were no significant differences (α = 0.05, Pr > F = 0.6576, MSD =

6.79) amongst the cultivars in their level of susceptibility as reflected by the percent

sheath lesion (area of leaf sheath lesion at the point of inoculation in mm as a proportion of the total area of the flag leaf sheath in mm).

However, there were significant differences (α = 0.05, Pr > F = 0.005, MSD = 8.46) amongst the P. fuscovaginae strains with respect to their virulence on wheat plants as reflected by percent sheath lesion. Highest virulence was shown by DAR77797, while

ICMP5940, ICMP9995 and ICMP9998 showed no significant difference from the negative control treatment of SDW. The other six P. fuscovaginae strains DAR77795,

DAR77320, DAR77800, ICMP9996, DAR77318 and ICMP10137, (the latter of which is the positive control), showed intermediate levels of virulence (Figure 2.3).

83

Figure 2.3: Virulence of Pseudomonas fuscovaginae strains on wheat cultivars inoculated with 107 cfu/mL of P. fuscovaginae by pinprick method. Bars with the same letter (upper case and lower case) are not significantly different at the 5% level.

2.3.4 Response of durum-wheat cultivars to inoculation with P. fuscovaginae

Both cultivars of durum-wheat showed significant reduction in growth (α = 0.05, Pr > F

= 0.001) with all the strains of P. fuscovaginae, during the seed soaking assay. There were significant (α = 0.05, Pr > F = 0.001, MSD = 1.59) interactions between the two durum- wheat cultivars and the P. fuscovaginae strains. ICMP5940 on EGA Bellaroi was the least virulent combination. The most virulent strain x variety interactions were, DAR77797 on both cultivars, DAR77320 x Tjilkuri, ICMP9996 x EGA Bellaroi and ICMP10137 x EGA

Bellaroi (Table 2.6).

84

Table 2.6: Growth response of two durum-wheat cultivars inoculated with 107 cfu/mL of Pseudomonas fuscovaginae by seed-soaking method, showing the interactions between durum-wheat cultivars and P. fuscovaginae strains

P. fuscovaginae Durum-wheat Mean seedling strain accession Tukey Grouping** variety height* number EGA Bellaroi SDW 10.62 A Tjilkuri SDW 7.10 B EGA Bellaroi ICMP5940 4.56 C EGA Bellaroi DAR77800 3.78 CD Tjilkuri ICMP9995 2.67 DE Tjilkuri ICMP5940 2.63 DE Tjilkuri DAR77318 2.32 DEF EGA Bellaroi ICMP9995 2.17 EF Tjilkuri ICMP9996 2.06 EF EGA Bellaroi DAR77320 1.96 EF Tjilkuri DAR77800 1.85 EF EGA Bellaroi DAR77318 1.78 EF Tjilkuri ICMP10137 1.48 EF EGA Bellaroi DAR77795 1.25 EF Tjilkuri DAR77795 1.22 EF EGA Bellaroi ICMP9996 1.01 F EGA Bellaroi ICMP10137 0.99 F Tjilkuri DAR77320 0.83 F EGA Bellaroi DAR77797 0.79 F Tjilkuri DAR77797 0.77 F

*Average seedling height in mm, square – root transformed. **Significant differences at α = 0.05, Pr > F = 0.001, MSD = 1.59. SDW = Sterile distilled water

2.3.5 Response of triticale cultivars to inoculation with P. fuscovaginae

There were significant differences in the response of the two Triticale cultivars to treatment with each P. fuscovaginae strain, in the seed-soaking assay. Although both cultivars were susceptible to infection by all the P. fucovaginae strains, there were significant (α = 0.05, Pr > F = 0.001, MSD = 1.60) interactions between the Triticale

85

cultivars and P. fuscovaginae strains. The effect of ICMP5940 on cv. Hawkeye was the least as shown by the average seedling height. The effect of DAR77797 and DAR77318 on both cv. Tahara and cv. Hawkeye were the highest along with the effect of DAR7795,

ICMP9995 and ICMP10137 on cv. Tahara (Table 2.7).

Table 2.7: Growth response of two triticale cultivars inoculated with 107 cfu/mL of Pseudomonas fuscovaginae by seed-soaking method, showing the interactions between triticale cultivars and P. fuscovaginae strains

P. fuscovaginae Tukey Triticale variety strain accession Mean seedling height* Grouping** number Hawkeye SDW 11.53 A Tahara SDW 6.84 B Hawkeye ICMP5940 5.62 B Tahara ICMP5940 3.45 C Hawkeye DAR77800 2.73 CD Hawkeye ICMP9995 1.74 DE Hawkeye DAR77795 1.61 DE Hawkeye DAR77320 1.51 DE Tahara DAR77320 1.48 DE Hawkeye ICMP9996 1.42 DE Tahara DAR77800 1.35 DE Tahara ICMP9996 1.18 DE Hawkeye ICMP10137 1.15 DE Hawkeye DAR77797 1.03 E Tahara ICMP9995 1.02 E Tahara ICMP10137 1.00 E Tahara DAR77795 0.95 E Tahara DAR77318 0.86 E Hawkeye DAR77318 0.83 E Tahara DAR77797 0.82 E *Average seedling height in mm, square –root transformed. **Significant differences at α = 0.05, Pr > F = 0.001, MSD = 1.60. SDW = Sterile distilled water

86

2.3.6 Response of barley cultivars to inoculation with P. fuscovaginae

There were significant (α = 0.05, Pr > F = 0.001, MSD = 1.39) interactions between the two barley cultivars, Schooner and Commander with the P. fuscovaginae strains, when inoculated at a concentration of 107 cfu/mL of bacteria by seed-soaking method, as demonstrated by their average seedling height (Table 2.8).

Table 2.8: Growth response of two barley cultivars inoculated with 107 cfu/mL of Pseudomonas fuscovaginae by seed-soaking method, showing the interactions between barley cultivars and P. fuscovaginae strains

Barley variety Mean Tukey Grouping** P. fuscovaginae strain seedling accession number height* Commander SDW 11.92 A Schooner SDW 5.99 B Schooner ICMP5940 3.56 C Schooner DAR77800 3.13 C D Commander DAR77800 1.87 D E Schooner DAR77320 1.39 E Commander ICMP5940 1.17 E Schooner ICMP9995 0.96 E Schooner DAR77795 0.76 E Schooner ICMP9996 0.75 E Commander DAR77318 0.74 E Schooner ICMP10137 0.70 E Commander ICMP9995 0.70 E Schooner DAR77318 0.70 E Commander ICMP10137 0.70 E Commander ICMP9996 0.70 E Commander DAR77795 0.70 E Commander DAR77797 0.70 E Commander DAR77320 0.70 E Schooner DAR77797 0.70 E *Average seedling height in mm, square –root transformed. **Significant differences at α = 0.05, Pr > F = 0.001, MSD = 1.39. SDW = Sterile distilled water

87

2.3.7 Outcomes of the disease survey

In the MIA, out of the four sampling locations with rice, samples were taken from six different rice crops, considering variations in the cultivar and management practice. From the rice crops in the CIA and MV, samples were taken from five locations in each. Out of the five locations of the wheat crops in the MIA, samples were taken from fourteen different wheat crops, when factors such as the variety, management practice, and the presence of disease symptoms similar to that of sheath brown rot, were taken in to consideration. In the CIA, only three wheat crops were sampled, and from MV, no samples could be taken due to difficulties in logistics. Each sample had five random replicates taken in a zig-zag pattern within the crop. One each showed symptoms similar to that of sheath brown rot, out of the six rice crops in MIA, and the five rice crops in the

CIA. One out of the fourteen wheat crops in the MIA displayed symptoms similar to that of sheath brown rot. The details of the crops surveyed are presented in Table 2.9.

88

Table 2.9: Details of sampling locations of the survey for sheath brown rot during the 2013/14 rice and wheat cropping seasons

Farm Location Crop Variety Disease Farm Location Crop Variety Disease Symptoms* Symptoms* 731 Leeton Wheat Merinda Yes 731 Leeton Rice Kyeema Yes 731 Leeton Wheat Merinda No 1034 Murrami Rice Langi No 1034 Murrami Wheat Phantom No 1034 Murrami Rice Langi No 1034 Murrami Wheat #3278 No 1074 Griffith Rice Doongara No 1034 Murrami Wheat #3278 No 1074 Griffith Rice Doongara No 1034 Murrami Wheat Crusader No 90 Coleambally Rice Sherpa Yes 1034 Murrami Wheat EGA Gregory No 181 Coleambally Rice Reiziq No 1034 Murrami Wheat #0079 No 212 Coleambally Rice Kyeema No 1034 Murrami Wheat Dart No 78 Coleambally Rice Reiziq No 1034 Murrami Wheat Spitfire No 79 Coleambally Rice Reiziq No 1034 Murrami Wheat Cobra No 1996 Griffith Rice Reiziq Yes 1074 Griffith Wheat QAL2000 No William park Deniliquin Rice Reiziq No 90 Coleambally Wheat Lincoln No Eddington Deniliquin Rice Reiziq No 1996 Griffith Wheat QAL2000 No Blighty Mayrung Rice Reiziq No 4160 Griffith Wheat QAL2000 No Strong Moone Rice Reiziq No 79 Coleambally Wheat Spitfire No Holmwood Deniliquin Rice Reiziq No 64 Coleambally Wheat Lincoln No *Disease symptoms of sheath brown rot. 89

2.3.8 Results of LAMP assay for detecting P. fuscovaginae under field conditions

Use of LAMP for P. fuscovaginae did not yield reliable results under the field conditions.

The samples prepared in saline wash solution appeared to interrupt the PCR reaction as observed by the lack of real-time quantification of PCR products displayed by the

GenieII®. When SDW was used to extract epiphytic bacteria from the plant samples by manually shaking, there still was no result.

2.3.9 Results of screening of bacteria isolated from diseased plant samples from

the survey

Bacteria isolated from the symptomatic rice and wheat plant samples collected in the survey were characterised to be fluorescent and non-fluorescent Pseudomonas spp. and

Erwinia spp. The fluorescent Pseudomonas bacteria were isolated on Pseudomonas selective media and showed florescence on KB medium, but did not comply with one or more of the other biochemical characteristics of P. fuscovaginae, in contrast to the fluorescent Pseudomonas bacteria that were re-isolated from rice and wheat plants inoculated with P. fuscovaginae and kept in the glasshouse. The results of profiling the microbiome associated with the disease symptoms (from wheat plant samples) revealed the diversity of bacteria in the phyllosphere (Appendix). In addition, fungi isolated from symptomatic wheat plant samples were identified as Alternaria spp., Fusarium spp. and

Cladosporium spp.

90

2.3.10 Response of rice and wheat to inoculation with P. fuscovaginae strains

isolated from wheat

Four P. fuscovaginae strains that were originally isolated from wheat, thus known as wheat-infecting strains (UPB0526, UPB0588, UPB0589, UPB1013) and one P. fuscovaginae strain (UPB0407), isolated from swamp rice grass, demonstrated significant

(α = 0.05, Pr > F = 0.001, MSD = 0.78) pathogenicity and varying levels of virulence on inoculated rice seeds using the seed-soaking method. The resulting seedling height of germinating rice seeds of cv. Amaroo is depicted in the Figure 2.4. As shown by the reduction in seedling height, UPB0407 and UPB0588 expressed virulence equal to each other but greater than the other strains. All the five strains had highly significant (α =

0.05, Pr > F = 0.001, MSD = 1.00) virulence, similar to each other when inoculated on the wheat seeds of cv. Rosella (Figure 2.5).

Figure 2.4: Growth response of rice seedlings (cv. Amaroo) inoculated with 107 cfu/mL of Pseudomonas fuscovaginae by seed-soaking method. Data represent seedling height 10 days after inoculation. Bars with the same letter (upper case and lower case) are not significantly different at the 5% level.

91

Figure 2.5: Growth response of wheat seedlings (cv. Rosella) inoculated with 107 cfu/mL of P. fuscovaginae by seed-soaking method. Data represent seedling height 10 days after incolation. Bars with the same letter (upper case and lower case) are not significantly different at the 5% level.

2.4 Discussion

P. fuscovaginae is generally known as a pathogen that has been reported to cause disease on a broad-range of host plants, causing sheath brown rot in rice and a number of other cereal crops. However, the pathogenicity of a particular P. fuscovaginae strain isolated from a particular crop, across a range of cereal crops has been demonstrated rarely.

Duveiller and Maraite (1990) demonstrated that P. fuscovaginae isolated from wheat in the central highlands of Mexico could cause symptoms of sheath brown rot when inoculated on rice. Malavolta et al. (1997a) reported that the P. fuscovaginae strains isolated from rice in Sao Paulo, Brazil, produced disease symptoms when inoculated on wheat. Quibod et al. (2015b) used the term “rice-infecting” to describe eight P. fuscovaginae strains that were isolated from rice thus, assumed to possess a repertoire of genes that may give them specific advantage in colonising and infecting rice plants. That study found that these rice-infecting P. fuscovaginae strains have the genetic capability

92

to occupy a range of cropping environments, implying the ability to occupy a range of host plants. The above-mentioned rice-infecting strains included DAR77795 and

DAR77800 from Australia. However, in Australia, P. fuscovaginae has been detected exclusively from rice and the ability of these strains to cause disease on other cereals that are grown in rotation with rice has not been investigated.

Therefore, in this study, plant inoculations were conducted to identify the potential of the five Australian strains (DAR77318, DAR77320, DAR77795, DAR77797, DAR77800), to cause sheath brown rot disease on wheat, durum-wheat, triticale and barley, as well as to screen a number of cultivars of rice and wheat for sources of resistance against P. fuscovaginae. In comparison, four strains isolated from rice in other countries

(ICMP5940, ICMP9995, ICMP9996, ICMP9998) and one strain isolated from wheat

(ICMP10137) were used. Furthermore, five additional strains isolated from wheat

(UPB0526, UPB0588, UPB0589, UPB1013) and one strain isolated from swamp rice grass (UPB0407) were used to establish the cross-pathogenicity of P. fuscovaginae. In general, all the P. fuscovaginae strains tested in this study were pathogenic to both rice and wheat as well as on triticale, durum-wheat and barley. This is the first report on the pathogenicity of Australian strains of P. fuscovaginae, on cereal crops other than rice.

The ability to occupy a wide range of hosts and the non-host specificity of the pathogenicity P. fuscovaginae could be attributed to the non-host specific factors of pathogenicity and virulence such as the phytotoxins produced by this pathogen. It has been demonstrated that the syringotoxins similar to syringomycin of P. syringae and the fuscopeptins similar to syringopeptins of P. syringae (Coraiola et al., 2008) produced by

P. fuscovaginae, are mainly responsible for the symptoms of sheath brown rot. These non-host specific phytotoxins cause stunted seedlings due to hindrance of seedling elongation (Batoko et al., 1994), poor panicle emergence due to impeded culm elongation

93

at heading (Batoko et al., 1997b) and necrotic lesions on leaf sheaths, plant stems and leaves due to increased permeability of cells causing leakage of cellular contents (Batoko et al., 1997d).

In the current study, the role of phytotoxins as a pathogenicity and virulence factor of P. fuscovaginae appears to have a greater effect at early stages of plant growth i.e. seedling growth, in comparison to causing sheath brown rot on mature plants, because all the strains showed high levels of virulence on all the cultivars during the seed-soaking assay, although some strains were not virulent on some cultivars of rice and wheat during the pinprick assay. The discrepancies in the degree of cultivar susceptibility to the pathogen at different stages of plant growth has been observed in earlier studies conducted on rice

(Adorada et al., 2013c; Malavolta et al., 1997a). However, it should be noted that this is the first report showing inconsistencies in the degree of susceptibility of wheat cultivars to P. fuscovaginae at seed germination and heading stages of plant growth.

Adorada et al. (2013c) notes that there could be different mechanisms of host plant resistance at the different stages of plant growth. In addition to that, the above-mentioned observation could be attributed to the growth conditions provided for P. fuscovaginae under the two assays. Seed-soaking assay may have provided favourable conditions for bacteria to proliferate by increased moisture, and more opportunity of quorum sensing to induce production of phytotoxins, compared to pinprick assay. The results indicate that the seed-soaking method employed in this study to identify disease resistance at early stages may not be appropriate for wheat, because the effect of phytotoxins appears to be entirely lethal to wheat seeds. However, the use of the seed-soaking method to identify disease resistance at early stages in wheat could be improved by optimising the inoculum concentrations. It should be noted that the inoculation techniques and inoculum concentrations used in this study for seed-soaking and pinprick assays conducted on

94

wheat and other cereals, are essentially the recommendations arising from the studies conducted for screening of rice germplasm. When conducting resistance screening of germplasm, method of inoculation and inoculum density can greatly influence the conclusion (Cottyn, 2003). Hence, alternative inoculation techniques and inoculum concentrations may have to be explored, developed and optimised, in order to identify sources of resistance in wheat, barley, triticale and other winter cereals.

Furthermore, physical barriers of the seed rather than the effect of host immune system are likely to be involved in the plant-pathogen interactions with respect to the mechanisms of host plant resistance at seed germination. Differences in the components and the characteristics of seed could be the reason for the wheat seeds showing greater susceptibility when inoculated with P. fuscovaginae by seed-soaking method, compared to that of rice seeds. For example, rice seed contain more silica than wheat seed (Sahebi et al., 2015), making the rice seed less permeable to bacteria and bacterial toxins in comparison to wheat seeds. In addition, the differences in plasma membrane intrinsic proteins such as ‘aquaporins’ between the wheat and rice seed cells (Palmer et al., 2015) may have contributed to the discriminatory penetration of bacteria and toxins into the seeds.

In this study, significant interactions were observed between the crop cultivars and P. fuscovaginae strains, which could be partially attributed to the host resistance mechanisms of a particular cultivar. It has been reported that the severity of the damage by phytotoxins in P. fuscovaginae correlates with the degree of susceptibility of the cultivars to the pathogen, which indicates that the phytotoxins are an integral component of the plant-pathogen interaction (Batoko et al., 1991). However, it should be noted that in this study, inoculation with live bacteria was used as the technique to screen the host resistance rather than the toxin. Therefore, the intrinsic differences in virulence factors in

95

different P. fuscovaginae strains that are supplementary to the phytotoxins, could contribute to the specific strain-cultivar interactions. The variations in virulence characteristics of some of the P. fuscovaginae strains used in this study has been also observed previously by Adorada et al. (2013a), who also used the inoculation with live bacteria.

In the present study, all the P. fuscovaginae strains including the Australian strains were found to be virulent on all the ten rice cultivars as well as eight wheat cultivars, at heading stage. Adorada (2013) observed that all the P. fuscovaginae strains from Australia are aggressively pathogenic on rice cv. Amaroo. Although the phytotoxins produced by P. fuscovaginae are not host-specific, the differences in virulence levels in relation to the origin of a particular strain as observed in this study, indicates a certain amount of host- specificity and specific strain-host interactions for P. fuscovaginae strains. This could be due to factors other than the phytotoxins that were acquired by each strain giving them a certain advantage in particular crop environment over the resistance mechanisms of their preferred host plant. Recent studies conducted on host specificity of similar bacterial pathogens points to other pathogenicity and virulence factors that could be related to host specificity, such as the Type III secretory systems, which produce the infection- and motility-related proteins (Sarkar, Gordon, Martin, & Guttman, 2006). Although, there have been recent studies conducted on the virulence factors other than the phytotoxins of

P. fuscovaginae (Patel et al., 2014), a detailed study of these factors including Australian

P. fuscovaginae strains are yet to be conducted.

It should be noted that there are a number of genes in the gene cassette coding for syringotoxin-syringopeptin phytotoxins in P. syringae pv. syringae (Scholz-Schroeder,

Soule, & Gross, 2003; Wang, Lu, Yang, Sze, & Gross, 2006) to which the phytotoxins in

P. fuscovaginae have shown structural and functional similarity (Ballio et al., 1996). This

96

complex consists of many proteins that functions as a cascade and governs the expression of these phytotoxins (Bender, Alarcón-Chaidez, & Gross, 1999). However, the nature of the phytotoxic protein complex in P. fuscovaginae strains in the context of their relative virulence has not been investigated.

Cother et al. (2009b) reported that single nucleotide polymorphisms (SNPs) appear to exist in the conserved regions within the group of Australian P. fuscovaginae strains. This suggests that there is potential for other SNPs to occur in the genetic codes that would translate to proteins responsible for pathogenicity and virulence, within a very short period upon introduction to a particular cropping environment. Previous studies have shown that P. fuscovaginae isolated from the sheath lesions and discoloured seeds from the same plant have different virulence characteristics (Malavolta et al., 1997a).

Furthermore, it should be noted that the bacteria acquire random fragments of DNA from their environment for their survival under certain conditions such as the plasmid DNA for resistance against antimicrobials (von Wintersdorff et al., 2016). Thus, a further investigation in to the genetic basis of the factors determining host specificity, such as the proteins involved in host-pathogen interactions and the host-specific virulence and pathogenicity, should be conducted including a number of P. fuscovaginae strains that were isolated from different hosts. It is evident that the Australian strains of P. fuscovaginae have unique characteristics in relation to pathogenicity and virulence, which would be very interesting to be studied in the context of host-pathogen interactions.

This study shows that there is a potential threat from rice-infecting Australian P. fuscovaginae strains to the other cereals such as wheat, barley and triticale. It should be of concern especially under the cropping system that is practiced in the Riverina region, where rice is a summer crop and other cereals are grown in the winter, in rotation.

Although not common, this crop rotation is practiced mostly to utilise residual soil

97

moisture in irrigated lands in southern NSW. However, in the present study, the survey for the sheath brown rot disease conducted in the three major growing regions did not result in P. fuscovaginae being isolated from the plants showing symptoms similar to that of disease, from both the rice and wheat crops surveyed. The results of the disease survey resonates with the findings of disease surveys conducted in 2010 and 2012 in the MIA

(Adorada, 2013), and is likely to reflect the changes in crop management practices. The rice cv. Amaroo from which the sheath brown rot disease was first observed in 2005, and the Australian strains of P. fuscovaginae were isolated, has been removed from commercial production since 2012. In addition to that, the current practice of burning rice stubble may contribute to the suppression of disease by preventing build-up of a P. fuscovaginae population. However, it should be noted that burning of stubble might not be complete; leaving the root system of infected rice with overwintering bacteria with the possibility of emerging on a winter cereal crop, especially in low temperatures. Previous observations suggests that the low winter temperatures with high humidity are favourable to P. fuscovaginae, although some studies dispute this (Jaunet et al., 1996). In general, the rice crops in Australia remain free of recurring bacterial diseases owing to the hot summer temperatures (Lanoiselet et al., 2001). The variation in expression of virulence characteristics of Australian P. fuscovaginae strains under diverse climate conditions has not been investigated yet. Overall, it can be hypothesised that the above-mentioned factors have hindered the P. fuscovaginae population from reaching the quorum required to express the pathogenicity and virulence factors, thus leading to the absence of sheath rot disease symptoms in the field. Although, inoculum densities of 106 to 107 cfu/mL have been generally used in pathogenicity testing of bacteria, as was done in the present study, it is not a likely population density for a single bacterial strain to occur on plants in nature

(Cottyn, 2003). Considering that the present study demonstrates that all the popular rice

98

cultivars such as Reiziq, Langi, Kyeema, Doongara, Sherpa and Opus, which are currently in large-scale commercial production, are susceptible to Australian P. fuscovaginae strains, it is essential that the rice crops are screened routinely preferably aided with accurate novel diagnostic tests for P. fuscovaginae.

This study also highlights the devastating effect the contaminated planting material would have on crop establishment and crop yield, provided the favourable conditions for the pathogen to proliferate. Cross contamination of successive crops via irrigation has a great likelihood to occur, as the rice cultivation is entirely irrigated and a water table is maintained throughout the cropping season until the harvesting.

Furthermore, the symptoms of sheath rot in rice crops observed during the survey could be arising from other pathogens which are known to exist in the rice fields in southern

NSW, such as Waitea circinata (sheath rot), Rhizoctonia oryzae-sativae (aggregate sheath rot), Pseudomonas syrinage pv. syringae (glume blotch), Pantoea ananatis (sheath and glume rot), that are often found in association with the incidents of sheath brown rot. In this study, the presence of aforementioned pathogens was not investigated. However, it would be of interest to study the profile of the microbial population associated with sheath rot diseases and their interactions.

This study demonstrated the potential of the infected seeds to cause a devastating effect on seedling establishment. Although the seed lots are often thoroughly checked for seed discoloration as a sign of disease, discoloured seeds of rice due to infection with sheath brown rot has been observed to disappear the brown colour and have bleached appearance after sometime (Adorada et al., 2013c). It should be noted that, the wheat seeds infected with P. fuscovaginae might not show any discolouration as with infected rice seeds.

Implications of invisible contamination of seed lots on crop establishment (Adorada et al., 2014) and on biosecurity (Adorada et al., 2013b) have been discussed before. Thus,

99

seed testing with advanced diagnostic techniques might be appropriate for screening for

P. fuscovaginae in both rice and wheat seed lots. As evident from the results of biochemical characterisation of the existing collection of P. fuscovaginae isolates, traditional diagnostics tests such as the BACTID and LOPAT tests have a limited capability in accurately identifying the causal organism of sheath brown rot disease. For example, in this study, four out of six isolates, which had no matching group in BACTID system, were Australian P. fuscovaginae isolates (DAR77795, DAR77797, DAR77320, and DAR77800) that were highly virulent. In addition, out of the isolates that were grouped as ‘Yellow, mostly non-pathogenic bacteria’, MB247 showed pathogenicity on rice. Similarly, Adorada (2013) reported that the type strain of P. fuscovaginae

ICMP5940 and other known P. fuscovaginae strains ICMP5939, ICMP9995 and

ICMP9998 was grouped into BACTID group 5, in which the characteristics are; Gram negative, bacilli, non-florescent, non-nitrate reducing, non- fermenting and oxidase activity positive.

Furthermore, in the present study, nine isolates out of the 40 isolates screened belong to

‘Yellow Erwinia’, which are a group of bacteria that are commonly found on discoloured grains. Because there are several pathogens causing sheath rot diseases showing very similar disease symptoms, and the existing diagnostic tests are not accurate to distinguish them, it is obvious that the accurate and rapid novel diagnostic tests such as LAMP should be routinely employed in both the crop fields and biosecurity screenings. In this study, the LAMP protocol with pf8 primer set developed for the detection of P. fuscovaginae

(Ash et al., 2014) was used in crop fields for the first time. Plant parts suspected to have sheath brown rot were screened in-situ with crude extract of epiphytic microbiome taken as the template, using a portable thermal cycler and isothermal master mix with primers that are anneal to targets in ambient temperature. However, the test failed to yield reliable

100

positive results, either due to the obstruction from the high salt content in the wash solution which was used to extract the existing epiphytic microbial population from the plant surface in to the solution, and/or the lack of template P. fuscovaginae DNA in the situation where just shaking by hand in SDW was used to extract epiphytic microorganisms. Therefore, it should be noted that there is opportunity to improve and optimise the sample preparation methods used in this study in order to detect P. fuscovaginae in field conditions, using the LAMP protocol. For example, use of a 0.1 M potassium phosphate buffer at pH 8.0 with sonication as described by Bodenhausen,

Horton, and Bergelson (2013) could be employed to reduce the salt content in the plant wash solution while increasing the extraction of bacterial cells in to the plant wash.

2.5 Conclusions

It is evident from the results of this study, that the Australian strains of P. fuscovaginae isolated from rice are capable of causing sheath brown rot disease to a range of cultivars of rice, wheat, durum-wheat, triticale and barley, thus highlighting a significant biosecurity risk. All 10 of the rice cultivars tested in this study, namely Amaroo,

Doongara, Illabong, Koshihikari, Kyeema, Langi, Opus, Quest, Reiziq, Sherpa were susceptible to infection with P. fuscovaginae, by pinprick method on leaf sheaths at heading. Likewise, all the Australian strains of P. fuscovaginae demonstrated pathogenicity and virulence at heading stage on all the wheat cultivars namely, Chara,

Ventura, Mace, Spitfire, EGA Gregory, Lincoln, Sunvale and Wyalkatchem. The aforementioned wheat cultivars and two other wheat cultivars namely, EGA Wedgetail and Rosella were also susceptible to infection with P. fuscovaginae at the seed germination stage. Similar effects were observed for triticale, durum-wheat and barley cultivars namely, Hawkeye, Tahara, EGA Bellaroi, Tjilkuri, Schooner and Commander,

101

at seed germination stage. In addition, P. fuscovaginae isolated from wheat and swamp rice grass were able to cause significant pathogenicity on rice cv. Amaroo, at seed germination stage. Notably, strain ICMP10137 isolated from wheat, expressed pathogenicity and virulence on all the rice cultivars, demonstrating the cross-species pathogenicity of P. fuscovaginae.

Although the results establish the cross-species pathogenicity of P. fuscovaginae, significant interactions between the strains examined and the crop cultivars were present.

These differences in the degree of cultivar susceptibility as shown in the study could be used as a basis for cultivar selection by estimating the tendency of a certain crop to have sheath brown rot, and as a guidance to identify molecular resistance markers for marker- assisted breeding of resistant cultivars.

Furthermore, the significant interactions shown by the P. fuscovaginae strains and the crop cultivars in this study, has a potential to be used as a precursor to study the molecular host-pathogen interactions. There could be a complex of pathogenicity and virulence factors involved in the host-specificity of P. fuscovaginae. Therefore, to identify which of them are responsible for the host-pathogen interactions seen here, it would be worthwhile to conduct a comparative analysis of the pathogenicity and virulence factors of P. fuscovaginae strains isolated from different hosts. In order to recognise such, a comparative analysis of the genomes of some of these strains would be advantageous as a starting point.

102

Chapter 3. Comparative genomics of P. fuscovaginae from a perspective of host-specificity

3.1. Introduction

P. fuscovaginae is mainly known as a pathogen of rice, but it has also been reported as a pathogen on wheat, triticale, sorghum and maize (Arsenijevic, 1991; Duveiller & Maraite,

1990; Duveiller et al., 1989). In addition, it has been reported from a number of wild grass species (Duveiller et al., 1988; Miyajima, 1980). Despite its importance as an emerging, economically important and globally distributed pathogen of a broad range of host plants, few studies have been conducted to compare the genetic characteristics of P. fuscovaginae strains in order to understand the factors involved in its pathogenicity, virulence and host- pathogen interactions. However, of the comparative genomics conducted on P. fuscovaginae to date, most have been conducted from a taxonomic perspective.

Phylogenetic analysis of 16S rRNA, gyrB, rpoB and rpoD gene sequences of rice- infecting P. fuscovaginae resulted in strains from Australia, Japan and China clustering into a single group (Adorada, 2013). A similar phylogenetic analysis of these genes found that P. fuscovaginae is closely related to the plant pathogenic P. fluorescens lineage (Patel et al., 2012). Generally, species in the P. fluorescens group show high genetic diversity and intra-species heterogeneity. For example, Silby et al. (2009) identified only 61% of genes were shared among three P. fluorescens strains, and comparative genomic analysis showed that a high number of genes were unique to each species (Silby et al., 2009).

Publication of draft genome sequences of P. fuscovaginae (Patel et al., 2012; Xie et al.,

2012) has provided novel opportunities to conduct comparative genomic analysis on a basis of functionality, while revealing more information about the genomic composition, structure and diversity of P. fuscovaginae. 103

Recently published comparative analysis of draft genomes and genome assemblies of rice-infecting strains of P. fuscovaginae and strains classed as P. fuscovaginae-like (i.e. plant pathogenic bacteria isolated from rice and closely related to P. fuscovaginae, but not characterised as such), shows that both P. fuscovaginae and P. fuscovaginae-like rice- infecting bacterial strains have plastic genomes with a high proportion of strain-specific genes and unique metabolic capabilities (Patel et al., 2014; Quibod et al., 2015b).

Although these analyses sufficiently reveal that the genomes of P. fuscovaginae are highly diverse, the available information may represent only a fraction of the overall variability, due to their limitation to rice-infecting strains. More diversity can be expected if the genomes of P. fuscovaginae isolated from different host plants are included in the comparisons. If a descriptive genome analysis of P. fuscovaginae is conducted based on host-specificity, it will not only expand the currently known pan-genome of P. fuscovaginae by adding to a diverse range of pathogenicity and virulence related factors, but also will reveal novel factors involved in host-pathogen interactions.

With respect to plant pathogenic bacteria, there are several key factors yet unknown about host-specificity (Sarkar et al., 2006). In general, bacterial pathogens use a variety of mechanisms to suppress or avoid host defences. However, a single strain of a bacterial pathogen may use several pathways to adopt to various hosts while, there could be multiple ways the strains of a particular pathogen to adopt to the same plant host (Sarkar et al., 2006). Previous studies conducted on bacterial plant pathogens related to P. fuscovaginae provide information on several factors that appear to have a role in host- specificity.

Comparative genome analysis of rice-pathogenic Burkholderia spp., B. glumae (causing grain rot), B. gladioli (causing sheath rot) and B. plantarii (causing seedling blight) found

104

that they have a unique set of genes that are involved in phytotoxin (toxoflavin, tropolone and rhizotoxin) production, of which the presence or absence in the genome of a particular strain appeared to be an adaptation to a particular host environment (Seo et al., 2015).

Although these phytotoxins are non-host specific (Ura et al., 2006), the regulation of their production in the above-mentioned Burkholderia spp. by the bacterial quorum sensing

(QS) (Chen et al., 2012), could be a factor determining the host preference in the above- mentioned Burkholderia spp. as they also possess distinctive QS-mediated AHL synthase-receptor circuits (Seo et al., 2015). Therefore, the genes involved in phytotoxin synthesis and secretion of P. fuscovaginae as well as QS should be considered in recognising the factors determining its host preferences.

Evidence for horizontal gene acquisition during change of hosts have been observed in relation to host-specific virulence in P. syringae (Sarkar et al., 2006). Genes acquired through horizontal gene transfer such as plasmids, provided a competitive advantage of survival of a non-pathogenic strain of the tomato pathogen, Clavibacter michiganensis subsp. michiganensis, despite its lack of a number of pathogenicity and virulence related factors (Załuga et al., 2014). Therefore, components of horizontal gene transfer, such as plasmids or virus/phage genes, could be another aspect to consider with regard to host specificity in P. fuscovaginae. In addition, elements of horizontal gene transfer are controlled by the clustered regularly interspaced short palindromic sequence repeats

(CRISPR)-mediated bacterial immune system, thus determining the fitness of a particular bacterial species in a given host environment (Rath, Amlinger, Rath, & Lundgren, 2015).

Seo et al. (2015) described that the presence of CRISPR systems vary among rice- infecting Burkholderia spp. particularly, B. plantarii was found to possess unique

CRISPR/CRISPR-associated (Cas) protein systems, while the other species B. glumae and B. gladioli did not. Given the potential of mobile genetic elements to alter the genetic

105

fitness of a bacterium and the CRISPR-Cas proteins to regulate the acquisition of such genes, these could be features of interest to explore in P. fuscovaginae, with regard to host specificity.

Furthermore, comparative genomic analysis of the phytopathogen Ralstonia solanacearum, which has an unusually broad host range, revealed that loss of genes, non- synonymous polymorphisms and horizontal gene transfer among type III secreted effectors are associated with the host preferences (Ailloud et al., 2015). Similarly, type

III secreted effectors appeared to be the most important factor for host specific virulence of Pseudomonas syringe, which demonstrates a wide range of adaptations to a number of hosts (Sarkar et al., 2006). Moreover, P. syringae exhibits a wide diversity in both the type III secreted effectors repertoire and pathways affecting the synthesis of secreted phytotoxins, suggesting a cumulative effect of both types of virulence factors in determining the host preferences (Loper et al., 2012). In many Gram negative bacteria, type III specialised protein secretory system is used by the pathogens to inject type III secreted effectors directly into host cells by suppressing the host defence response and prompt pathogen growth and transmission, thus type III secretion system and type II secreted effectors are considered as essential constituents of pathogenicity and virulence factors, although these demonstrate a wide diversity between the strains. Therefore, type

III secreted effectors could be another feature of determining the host preferences of P. fuscovaginae.

Although understanding the genomics of host preferences is an essential element of controlling a plant pathogen, host-related pathogenicity and virulence factors are yet to be fully investigated or characterised for P. fuscovaginae. With respect to factors influencing the host preferences of P. fuscovaginae, recent studies suggest that the pan-

106

genome of rice-infecting strains of P. fuscovaginae includes a set of common virulence factors that are useful for colonising the rice sheath (Quibod et al., 2015b). And the core- secretome of P. fuscovaginae-like species showed strong signals with positive selection that associate with pathogen lifestyles (Quibod et al., 2015b). Understanding the synthesis and secretion of these molecules will be beneficial to determine the host preferences of

P. fuscovaginae. The information generated from the above-mentioned comparative genomic analyses with regard to pathogenicity and virulence factors can be used as a starting point. However, it should be noted that these studies have been conducted only on P. fuscovaginae isolated from and known to infect rice. Therefore, available genomic information on P. fuscovaginae needs to be expanded by investigating the genomes of P. fuscovaginae isolated from several other hosts. However, according to the author’s knowledge, such information is not publicly available yet. The present study attempts to recognize potential genomic adaptations for host-specificity of P. fuscovaginae, by comparison of the draft genomes from strains of P. fuscovaginae strains isolated from rice, wheat, and a wild grass species.

3.2. Materials and methods

3.2.1. P. fuscovaginae genomes

Previously sequenced and assembled genomes of rice-infecting strains of P. fuscovaginae

(DAR77795, DAR77800, ICMP5940, SE-1, CB98818 and UPB0736) were compared with the newly sequenced and assembled genomes of wheat-infecting strains of P. fuscovaginae (UPB0526, UPB0588, UPB0589, UPB1013) and UPB0407, a P. fuscovaginae strain isolated from swamp rice grass (Leersia hexandra). The background information of these draft genomes are summarised in Table 3.1. The genomes of

107

DAR77795, DAR77800, ICMP5940, SE-1, CB98818 and UPB0736 were accessed at

NCBI, uploaded to the Rapid Annotations using Subsystems Technology (RAST) server

(Aziz et al., 2008), and re-annotated. The P. fuscovaginae strains UPB0526, UPB0588,

UPB0589, UPB1013 and UPB0407 were imported from the Université Catholique de

Louvain, Belgium under the permit to import quarantine material numbered IP14013852 subject to the Quarantine Act 1908 Section 13 (2AA) and stored at CSU, Wagga Wagga.

They were subcultured, basic identification tests performed and pathogenicity assays were conducted (data presented in Chapter 1). De novo assembled (Velvet assembler 1.2.10) draft genomes of these strains were obtained from CSU, following sequencing by the Australian

Genome Research Facility. They were annotated using RAST server. All the 11 genomes were accessed for comparison through the RAST server and The SEED Viewer (Overbeek et al., 2014). Comparisons of protein-coding sequences were conducted by a sequence- based genome comparison tool, with the RAST-annotated draft-genome of UPB0736 as the reference genome (Patel et al., 2012).

108

Table 3.1: Background information of the genomes of Pseudomonas fuscovaginae used in this study

Strain Host Location isolated Year NCBI isolated accession number UPB0736 Rice Antsirabe, 1986 AIEU01 Madagascar CB98818 Rice Zhejiang, China 1998/99 ALAQ01 SE-1 Rice Siniloan, Philippines 2000 AQOH01 DAR77795 Rice Leeton, Australia 2005 BATD01 ICMP5940 Rice Hokkaido, Japan 1976 BATG01 DAR77800 Rice Leeton, Australia 2005 JSYZ01 UPB0588 Wheat Codagem, Mexico 1987 - UPB0526 Wheat Codagem, Mexico 1987 - UPB1013 Wheat Nigale, Nepal 1995 - UPB0407 Swamp rice Gisha, Burundi 1988 - grass UPB0589 Wheat Codagem, Mexico 1987 -

3.2.2. Whole genome comparisons

Genes were considered to be unique to rice-infecting P. fuscovaginae, if they were present in at least three out of six strains isolated from rice including UPB0736, and absent in three out of four strains from wheat. Genes were selected for further analysis if they appeared to be entirely absent in rice-infecting strains, except in UPB0736. Genes considered unique to wheat-infecting P. fuscovaginae were selected if they appeared to be common for wheat-infecting strains rather than rice-infecting. Some of these genes may be found in UPB0736 and only one of the rice-infecting P. fuscovaginae strains.

Selected genes were characterised for their functions using The SEED viewer psi-BLAST tool (Altschul et al., 1997) and SmartBLAST algorithms using the NCBI blastp suite.

From the BLAST results, only those showing the highest homology and covering 100% of the query were selected based on E-value. In addition, characterisation of the features

109

based on homology to nucleotide sequences was considered by RAST/The SEED viewer comparison of the region of the focus gene with genes from annotated genomes that are available on the RAST server.

3.3. Results

General features of the RAST-annotated P. fuscovaginae draft genomes are summarised in Table 3.2. The total number of protein-coding sequences in the reference genome

UPB0736 were 5849, while the numbers varied between 5801-9000 in other P. fuscovaginae genomes. Compared to that, RNA coding sequences varied less among the genomes from 53 of UPB0736 to 74 of UPB0407. The number of the subsystems that these protein-coding sequences were categorised into ranged from 505 to 536, while in

UPB0736 it was 528. There was also a great difference in the number of contigs (162-

459).

Table 3.2: Features of the genomes of Pseudomonas fuscovaginae used in this study

Strain Size GC Number Number of Number Number (bp) content of contigs subsystems of of RNA- (%) with protein- coding Protein coding sequences Encoding sequences Genes (PEGs) UPB0736 6350233 61.5 273 528 5849 53 CB98818 6541075 61.4 251 530 6685 69 SE-1 6525804 63.1 407 470 6071 46 DAR77795 6298008 61.3 162 523 6998 58 ICMP5940 6378970 61.2 459 525 7603 56 DAR77800 6188479 61.2 175 505 9000 65 UPB0588 6574114 61.0 288 529 5984 73 UPB0526 6785712 60.8 278 534 6278 70 UPB1013 6439171 60.9 311 520 5802 71 UPB0407 6695693 61.1 230 533 6118 74 UPB0589 6753763 60.9 258 536 6246 67

110

Graphical representations of functional annotations highlighting the differences among the P. fuscovaginae genomes within the rice-infecting and wheat-infecting groups are presented in Figure 3.1 and 3.2, respectively. These show high variability of functionality within each group, indicating distinctive metabolic features for each P. fuscovaginae strain. The diversity within the wheat-infecting P. fuscovaginae was greater than that was among rice-infecting P. fuscovaginae. UPB1013 was found to be uniquely different from the rest.

111

Figure 3.1: Comparison of five rice-infecting Pseudomonas fuscovaginae genomes (inwards: CB98818, DAR77795, DAR77800, ICMP5940, SE-1), aligned to the reference genome of the rice-infecting strain UPB0736.

112

Figure 3.2: Comparison of four wheat-infecting Pseudomonas fuscovaginae genomes (inwards: UPB0588, UPB0526, UPB1013, UPB0589) aligned to the reference genome of the rice-infecting strain UPB0736.

113

From the 5849 protein-coding sequences in the reference genome, 63 genes were found to be unique to UPB0736. Of those, 71% were hypothetical proteins, the rest were replication-related and membrane-related proteins, and 26 genes were found only in

UPB0407 and UPB0736. Of these, 55% were hypothetical proteins and the rest were phage-related proteins. For rice-infecting P. fuscovaginae, 119 genes appear to be unique, of which 28.5% were hypothetical proteins. From the rest, 26 were long-chain-fatty-acid-

-CoA ligase (LC-FAC) and 11 were non-ribosomal peptide synthetase modules and related proteins (NRPS). Another eleven were peptide synthetases (PS) and eight were large exoproteins involved in heme utilisation or adhesion of the ShlA/HecA/FhaA family

(adhesins). Compared to rice-infecting P. fuscovaginae, 65 genes appear to be more common to wheat-infecting strains of P. fuscovaginae (UPB0526, UPB0588, UPB0589 and UPB1013). Out of these genes, 81.5% were hypothetical proteins.

Functions of some of the selected genes that were entirely absent in wheat-infecting P. fuscovaginae genomes and present in all the rice-infecting genomes are presented below.

3.3.1. Rice NRPS I

This protein-coding sequence (RAST Feature ID: fig|50340.26.peg.63) consisting of 138 amino acids was found to have multiple putative conserved domains belonging to adenylate forming domains class I superfamily, with similarity to the adenylation domain of A-NRPS-Sfm-like including Saframycin A gene cluster from Streptomyces lavendulae. Others include, amino acid adenylation domain, AMP-binding enzyme,

PRK12316 peptide synthase, NRPS component F (EntF) for secondary metabolites biosynthesis, transport and catabolism. BLASTp revealed >90% similarity to aa- adenylation domains/NRPS of P. syringae species, which is a component of 114

Syringopeptin synthetase B (SypB syn.) of P. syringae pv. syringae (97%) and P. syringae pv. solidagae (96%) and Syringopeptin synthetase A (SypA syn.) of P. syringae pv. solidagae (95%) and P. syringae pv. apatata (94%) genes. Chromosomal region comparison showed that this gene groups with Ralstonia solanacearum cell processes, protection responses and cell killing, Nostoc punctiforme NRPS modules and related proteins, Thioesterase domains of type I polyketide synthases or NRPS, Myxococcus xanthus linear gramicidin synthetase subunit D [Includes: ATP-dependent tryptophan adenylase (TrpA) (Tryptophan activase), ATP-dependent D-leucine adenylase (D-LeuA)

(D-Leucine activase), Leucine racemase [ATP-hydrolyzing], ATP-dependent tryptophan adenylase (TrpA) (Tryptophan activase), ATP-dependent glycine adenylase (GlyA)

(Glycine activase) and linear gramicidin--PCP reductase]. In addition, this gene is located in association with a NRPS/Transcriptional regulator, LuxR family protein, Choline- sulfatase, putative lipoprotein polyketide synthase of M. xanthus, Malonyl CoA-acyl carrier protein transacylase of N. punctiforme and Polyketide synthetase of

Agrobacterium tumefaciens.

3.3.2. Rice NRPS II

This protein sequence comprises 146 amino acids and noted as RAST Feature ID: fig|50340.26.peg.5487. It was found to have multiple putative conserved domains belonging to the Adenylate forming domain Class I superfamily, namely, the adenylating domain of A –NRPS-Sfm-like; amino acid adenylation domains, AMP-binding enzyme,

PRK12316 peptide synthase and EntF. BLASTp showed >90% homology to NRPS and aa-adenylation domain proteins of P. syringae species, more specifically, syringopeptin synthetase A, B and C of P. syringae pv. syringae (92, 97, 94% homology, respectively).

115

A comparison of the chromosomal regions puts this gene in the same group with R. solanacearum cell processes, protection responses and cell killing; M. xanthus NRPS/

Linear gramicidin synthetase subunit D [Includes: TrpA, D-LeuA, Leucine racemase,

TrpA, GlyA, Linear gramicidin--PCP reductase; NRPS/peptide synthetase of

Pseudoalteromonas tunicata, Microcystis aeruginosa: Pseudoalteromonas tunicate; N. punctiforme Thioesterase domains of type I polyketide synthases or NRPS Hahella chejuensis NRPS terminal component, peptide synthetase of A. tumefaciens;

Pseudomonas aeruginosa PAO1 NRPS modules and related proteins, Taurine transport system permease protein TauC, Taurine transport ATP-binding protein TauB, Taurine- binding periplasmic protein TauA, Alpha-ketoglutarate-dependent taurine dioxygenase;

Actinosynnema mirum: NRPS modules and related proteins, Thioesterase domains of type

I polyketide synthases or NRPS. This gene is also located in association with M. xanthus

Choline-sulfatase, polyketide synthase; M. aeruginosa polyketide synthase; N. punctiforme Malonyl CoA-acyl carrier protein transacylase; A. tumefaciens Polyketide synthetase; P. aeruginosa PAO1: SyrP-like protein Alpha-ketoglutarate-dependent taurine dioxygenase; A. mirum polyketide synthase type I genes.

3.3.3. Rice NRPS III

This gene with RAST Feature ID: fig|50340.26.peg.613, is 171 amino acids long and has multiple putative conserved domains, specifically, A-NRPS-Srf-like; the adenylation domain of NRPS, including Bacillus subtilis termination module Surfactin (SrfA-C).

Others are amino acid adenylation domain, AMP-binding enzyme, EntF, peptide synthase

PRK12467. It showed homology (76-91%) to the NRPS and amino acid adenylation domain of Pseudomonas species, SypB syn., SypB syn. and SypC syn. of a number of P.

116

syringae (78-91%) sub species, and CipE of P. cichori (77%). The best match was

NRPS/SypB syn. of P. syringae pv. syringae (91%). Chromosomal region comparison places this gene in the same group with the Ralstonia solanacearum gene set for functions of cell processes, protection responses, cell killing, peptide synthetase of

Chromobacterium violaceum, Photorhabdus asymbiotica subsp. asymbiotica and

Serratia marcescens, NRPS/polyketide synthase and N-acetylmuramoyl-L-alanine amidase AmiB precursor of Myxococcus xanthus. It was found to be located closer to the gene cassettes of LemA protein, death incurring protein, Doc toxin, RTX toxins determinant A and related Ca2+-binding proteins of Photorhabdus asymbiotica subsp. asymbiotica and outer membrane component of tripartite multidrug resistance system,

Secretion protein HlyD, LC-FAC (part of Biotin biosynthesis, fatty acid metabolism,

Phenylalkanoic acid degradation clusters) of P. fuscovaginae.

3.3.4. Rice NRPS IV

The 174 amino acid-long protein of RAST Feature ID: fig|50340.26.peg.4891 has several conserved domains, A-NRPS-Sfm-like, A-NRPS TubE like: the adenylation domain (A domain) of a family of NRPSs synthesising toxins and antitumor agents; for example,

TubE for tubulysine, CrpA for cryptophycin, TdiA for terrequinone A, KtzG for kutzneride, and Vlm1/Vlm2 for valinomycin, AMP-binding enzyme, peptide synthase

PRK12316, EntF. It showed BLASTp homology (76-91%) to NRPS and amino acid adenylation domain of a number of Pseudomonas spp. (up to 91%) and Syp A, B, C syn. of several P. syringae subsp. (86-91%). The best match was to SypC syn. of P. syringae pv. syringae (90%) and SypB syn. of P. syringae pv. aceris (91%). Comparison of the chromosomal region of the focus gene found it in the same group with NRPS / peptide

117

synthetase repeats of Pseudoalteromonas tunicata, peptide synthetase repeats of P. florescence and P. asymbiotica subsp. asymbiotica and M. xanthus, linear gramicidin synthetase subunit D (TrpA, D-LeuA, leucine racemase TrpA, GlyA), linear gramicidin-

-PCP reductase and cell processes, protection responses and, cell killing of R. solanacearum.

3.3.5. Rice NRPS V

This protein consists of 114 amino acids and named in RAST as Feature ID: fig|50340.26.peg.3074, has only one putative conserved domain; peptide synthase

PRK12467. BLASTp homology resulted in 77%-83% to amino acid adenylation domain proteins and NRPS of a number of Pseudomonas species. The best match was 83% of

SypB syn. of P. syringae pv. syringae. Comparison of chromosomal region of this gene provided no grouping with other organisms. In P. fuscovaginae UPB0736, this protein is involved with NRPS related proteins/peptide synthetase. Predicted functions for it are; permease of ABC transporter, ATP-binding component of ABC transporter, transcriptional regulator for fatty acid degradation FadQ, TetR family, multi-copper oxidases, cytochrome oxidase biogenesis protein Sco1/SenC/PrrC, copper metallochaperone, bacterial analog of Cox17 protein.

3.3.6. Rice PS I

The protein of RAST Feature ID: fig|50340.26.peg.1886 contains 204 amino acids, and multiple putative conserved domains found were; Phosphopantetheine attachment site,

A-NRPS, MP-binding enzyme C-terminal domain, peptide synthase, L-aminoadipate-

118

semialdehyde dehydrogenase, acyl-CoA synthetase (AMP-forming)/AMP-acid ligase II

(involved in lipid transport and metabolism and secondary metabolites biosynthesis, transport and catabolism). BLASTp results show homology to peptide synthetase (73%), amino acid adenylation protein (66%) of P. syringae SypC syn. of P. syringae pv. atrofaciens (71%). Best matches include NRPS/ MassB of P. synxantha (72%) and P. syrinage pv. syringae (73%). Comparison of chromosomal region of this gene gives only grouping with peptide synthetase of P. florescence. In P. fuscovaginae UPB036, this gene is associated with secreted oxygenase, transcriptional regulator, LysR family, and multidrug resistance efflux pumps (RND efflux system, outer membrane lipoprotein

CmeC).

3.3.7. Rice PS II

This 207 amino-acid protein has RAST Feature ID: fig|50340.26.peg.2310, and contains multiple conserved domains; phosphopantetheine attachment site, condensation domain,

PKS phosphopantetheine attachment site, PRK12467 peptide synthase, L-aminoadipate- semialdehyde dehydrogenase. BLASTp results show homology to NRPS of a number of

Pseudomonas species (76-80 %), amphisin synthetase of Pseudomonas sp. (76-81%), gramidicin S synthase 2 of P. protegens (79%) WLIP synthetase B of P. florescence

(78%), SypC syn. of P. syringae pv. solidagae (70%), arthrofactin synthatase B of

Pseudomonas sp. (77%), TaaB of P. constantinii (77%), xantholysin synthetase B of P. putida (76%). Best match was to NRPS of P. chloraraphis (82%). Comparison of the chromosomal region did not find functional similarities to other organisms.

119

3.3.8. Rice LC-FAC I

This protein of 82 amino acids is denoted as RAST Feature ID: fig|50340.26.peg.5306.

It was shown to have conserved domains A NRPS, amino acid adenylation domain, EntF,

AMP-binding enzyme. BLASTp shows homology to NRPS, amino acid adenylation domain proteins, SypA, B, C syn. of P. syringae. The best match was to NRPS of P. syringae pv. atrofaciens (89%). This LC-FAC is a component of biotin biosynthesis/biotin synthesis cluster, fatty acid metabolism cluster, n-phenylalkanoic acid degradation subsystems of P. fuscovaginae. Chromosomal comparison put this feature in the same group with NRPS/peptide synthetase of N. punctiforme, pepetide synthetase of

P. fluorescens and Cyanothece sp., NRPS/polyketide synthase, linear gramicidin synthetase subunit C [includes: ATP-dependent valine adenylase (ValA) (Valine activase); ATP-dependent D-valine adenylase (D-ValA) (D-valine activase); valine racemase [ATP-hydrolyzing], TrpA, D-LeuA, leucine racemase [ATP-hydrolyzing];

ATP-dependent tryptophan/phenylalanine/tyrosine adenylase (Trp/Phe/TyrA)

(tryptophan/phenylalanine/tyrosine activase, major facilitator superfamily MFS 1, thioesterase domains of type I polyketide synthases or NRPS. Also, belonging to the same group with Microcystis aeruginosa non-NRPS modules and related proteins, Anabaena variabilis peptide synthetase and macrolide-efflux protein, Streptosporangium roseum

NRPS/aa-adenylation proteins, Serratia marcescens peptide synthetases. The position of this gene on the chromosome is close to SyrP-like, heterocyst specific ABC-transporter, membrane fusion protein DevB homolog, ABC transporter permease protein, serine/threonine kinase and ABC transporter-ATP binding proteins in A. variabilis and lysine biosynthesis DAP pathway genes (N-acyl-D-amino-acid deacylase, diaminopimelate decarboxylase), MbtH-like protein, ABC superfamily/membrane transport protein, arginine/ornithine transport ATP-binding protein and SryP-like proteins

120

in S. roseum. In N. punctiforme, it is a part of the exopolysaccharide biosynthesis cluster

(non-functioning in this organism), cell division ribosomal stress proteins, DNA replication, transcription factors/ transcription repair clusters, and is associated with transcription-repair coupling factors, hemolysins and related proteins containing CBS domains, aldo/keto reductase family, undecaprenyl-phosphate galactosephosphotransferase, membrane protein involved in the export of O-antigen, teichoic acid lipoteichoic acids, glycosyltransferase and exopolysaccharide production protein ExoQ. In M. aeruginosa, it is associated with ABC transporter related and mobile element proteins. In Cyanothece sp, this gene is located close to cyanobacterial polyprenyltransferase (UbiA homolog), mobile element protein and transposase genes.

3.3.9. Rice LC-FAC II

This protein was denoted as RAST Feature ID: fig|50340.26.peg.4255 and had 110 amino acids, conserved domains of A-NRPS, AMP-binding enzyme C-terminal domain, peptide synthase, L-aminoadipate-semialdehyde dehydrogenase, acyl-CoA synthetase

(AMP-forming)/AMP-acid ligase II. BLASTp yielded homology (<91%) to NRPS, peptide synthases, AA adenylation domain proteins of many Pseudomonas species and

Syp A, B, C syn. (75-81%) of several P. syringae subspecies. In addition, there was 75% homology to syringomycin syn. of P. syringae pv. syringae (75%), syringomycin syn. E of P. syringae pv. apatata (75%). In addition, NupA P. florescence (77%) and CipC P. cichori (76%) were significantly homologous. However, the best match was to NRPS of

P. fuscovaginae (91%) and LC-FAC ligase of P. batumici (88%). This LC-FAC is a part of biotin synthesis cluster, fatty acid metabolism cluster, n-phenylalkanoic acid degradation subsystems of P. fuscovaginae. A comparison of the chromosomal region,

121

grouped this feature with only P. fluorescens peptide synthetases and R. solanacearum

NRPS.

3.3.10. Rice LC-FAC III

This protein-coding sequence of RAST Feature ID: fig|50340.26.peg.2976 of 157 amino acids has multiple conserved domains; A-NRPS, AMP-binding enzyme, peptide synthase, L-aminoadipate-semialdehyde dehydrogenase, Acyl-CoA synthetase (AMP- forming)/AMP-acid ligase II, AMP-binding enzyme. In BLASTp, it was homologous to an AA adenylation enzyme/thioester reductase family protein of Pseudomonas spp.

(81%), NRPS/AA adenylation of Pseudomonas sp. (<82%), ViscC of P. chlororaphis

(79%), arthrofactin syn. B of Pseudomonas sp. (80%), TaaB of P. constantinii (79%),

TaaA of P. constantinii (75%), Syringomycin syn. of Pseudomonas sp. (73%) and NupA of P. florescence (72%). However, the best match was with the NRPS of P. fuscovaginae

(97%). This LC-FAC is a part of the biotin biosynthesis/biotin synthesis cluster, fatty acid metabolism cluster and n-phenylalkanoic acid degradation subsystems of P. fuscovaginae. A comparison of the chromosomal region, grouped this feature only with peptide synthetases of P. florescence and Chromobacterium violaceum. In C. violaceum it was found to be assigned closer to putative diaminobutyrate-pyruvate transaminase and amino acid ABC transporter proteins, in the contig. In P. florescence, it was associated with nicotinate-nucleotide--dimethylbenzimidazole phosphoribosyltransferase, and adenosylcobinamide-phosphate guanylyltransferase that are components of cobalamin synthesis and co-enzyme B12 biosynthesis cluster.

122

3.3.11. Rice LC-FAC IV

This protein-coding sequence of 170 amino acid, denoted as RAST Feature ID: fig|50340.26.peg.4368 contains multiple conserved domains; A-NRPS, AMP-binding enzyme C-terminal domain, EntF; enterobactin synthase subunit F, amino acid adenylation domain, EntF; non-ribosomal peptide synthetase component F, and AMP- binding enzyme. BLASTp shows homology to NRPS of Pseudomonas sp. (<72%), AA adenylation enzymes of Pseudomonas sp. (<71%), tolaasin of P. tolaasii (72%), TaaC

(75%), TaaB (74%) and TaaD of P. constantinii (75%). In addition, it shows high homology to peptide syn. of P. syringae pv. syringae (70%) and SypA syn. (71%), SypB syn. (72%), SypC syn. of several subspecies of P. syringae, syringomycin syn. of P. syringae pv. syringae (68%) and NupA of P. florescence (68%). However, the best match was NRPS of P. fuscovaginae (86%). This LC-FAC is a part of the biotin synthesis cluster, fatty acid metabolism cluster, n-phenylalkanoic acid degradation subsystems of

P. fuscovaginae. Comparison of the chromosomal region grouped this gene with

NRPS/peptide synthetases of P. florescence, S. marcescens, and P. asymbiotica subsp. asymbiotica and siderophore biosynthesis non-ribosomal peptide synthetase modules of

S. roseum. In P. fuscovagiane, this gene was found to be located close to macrolide- specific efflux protein MacA and macrolide export ATP-binding/permease protein MacB that belong to a multidrug resistance efflux pump subsystem. In S. roseum, it is assigned next to the ABC-type Fe3+-siderophore transport system ATPase component and permease 2 component, which are part of the flavohaemoglobin subsystem. In P. asymbiotica subsp. asymbiotica, this gene is associated with autoinducer 2 (AI-2) kinase

LsrK gene of AI2 transport and processing (lsr ACDBFGE operon) subsystem.

123

3.3.12. Rice LC-FAC V

The gene of RAST Feature ID: fig|50340.26.peg.2988 has 175 amino acids has multiple conserved domains ANRPS, AMP-binding enzyme C-terminal domain, phosphopantetheine attachment site, peptide synthase, L-aminoadipate-semialdehyde dehydrogenase, CaiC; Acyl-CoA synthetase (AMP-forming)/AMP-acid ligase I, AMP- binding enzyme. BLASTp homologues include; NRPS of Pseudomonas sp. (<77%), AA adenylation enzyme / thioester reductase family of Pseudomonas spp. (73%), amphisin syn. of Pseudomonas sp. (73%), TaaC of P. constantinii (74%), TaaE of P. constantinii

(72%), arthrofactin syn. B of Pseudomonas sp. (74%), arthrofactin syn. A of

Pseudomonas sp. (73%), SypA syn of P. syringae (72%), SypC of P. syringae pv. lapsa

(70%), CipB of P. cichori (70%). The best match was NRPS of P. fuscovaginae (94%).

This LC-FAC is a part of the biotin biosynthesis/ biotin synthesis cluster, fatty acid metabolism cluster, n-phenylalkanoic acid degradation subsystems of P. fuscovaginae.

Comparison of the chromosomal region grouped this gene with NRPS/peptide synthetases of P. fuscovaginae and P. florescence. In P. fuscovaginae, this gene was found to be located adjacent to the RND efflux system, outer membrane lipoprotein

CmeC, which are part of the multidrug resistance efflux pump subsystem and transcriptional regulators of LuxR family, and LysR family along with oxygenase secreted protein. In P. florescence, it was closely associated with polymyxin resistance protein ArnT, undecaprenyl phosphate-alpha-L-Ara4N transferase, melittin resistance protein PqaB; which are parts of lipid A Ara4n pathway (polymixin resistance) subsystem and RND efflux system outer membrane lipoprotein CmeC, which are components of the multidrug resistance efflux pump subsystem. In addition, ABC transporter ATP-binding protein, transcriptional regulator of LuxR family, YbbM seven transmembrane helix

124

protein, iron-sulfur binding protein YPO1417 and transmembrane protein were found adjacent to this gene.

3.3.13. Rice LC-FAC VI

This 248 amino acid long gene of RAST Feature ID: fig|50340.26.peg.3398 has two conserved domains; condensation domain and peptide synthase. Significant BLASTp homologies were NRPS of Pseudomonas sp. (<88%), NRPS/polyketide synthase of R. solanacearum (63%), amino acid adenylation enzyme/ thioester reductase family of P. fuscovaginae (75%), AA adenylation of several P. syringae sub species (<88%), CipF of

P. cichori (68%), SypC syn. of P. syringae (<88%), Syringomycin syn E of P. syringae pv. syringae (75%) and NunD of P. florescence (71%). The best match was to amino acid adenylation of P. syringae pv. syringae (86%). This LC-FAC is a part of biotin biosynthesis cluster, fatty acid metabolism cluster, n-phenylalkanoic acid degradation subsystems of P. fuscovaginae. Comparison of the chromosomal region placed this gene with the same group as NRPS of R. solanacearum, NRPS/pyoverdine of P. fuscovaginae, peptide synthetases of P. florescence, Bradyrhizobium japonicum, S. marcescens,

NRPS/polyketide synthase of M. xanthus and NRPS/ pyoverdine synthetase D/ polyketide synthase type I of Opitutus terrae. In P. fuscovaginae, it was found to be accompanying macrolide-specific efflux protein MacA, macrolide export ATP- binding/permease protein MacB and outer membrane component of tripartite multidrug resistance system proteins which are components of the multidrug resistance efflux pump subsystem.

125

3.4. Discussion

The study described in Chapter 2 demonstrates that the strains of P. fuscovaginae isolated from rice are capable of infecting wheat and vice versa, thus establishing the cross- pathogenicity of P. fuscovaginae. Although that study highlights the potential of wheat and other cereal crops to become hosts for strains of P. fuscovaginae isolated from rice, in Australia, the symptoms of the sheath brown rot disease caused by P. fuscovaginae have been observed only from rice crops in cereal cropping systems in the Riverina region of southern NSW (Adorada, 2013; Cother et al., 2009b). However, the environmental conditions in these cropping systems are favourable for P. fuscovaginae, with cold night temperatures that is assumed to predispose the crops to infection (Macapuguay &

Mnzaya, 1988) with relatively high humidity in the temperate to sub-tropical agroclimate of the Riverina region (Williams, Hook, & Hamblin, 2002). In addition, the study described in Chapter 2 shows that a number of crop cultivars of rice, wheat, barley and triticale which are regularly grown in southern NSW to be susceptible to P. fuscovaginae.

Moreover, the crop cultivars that could be potential hosts for P. fuscovaginae are usually cultivated in rotation. Therefore, the present study was conducted to investigate the pathogen factors that could be determining the host preferences of P. fuscovaginae.

Recent studies conducted by genome analysis show that P. fuscovaginae has a very plastic genome with diverse metabolic capabilities (Quibod et al., 2015b), enabling the pathogen to adapt to various host environments (Patel et al., 2014). However, these studies have been conducted by comparison of genomes of “rice-infecting P. fuscovaginae” strains only. Therefore, in the present study, to determine if a genetic basis for host specificity of P. fuscovaginae exists, a comparison of eleven genomes of P. fuscovaginae strains isolated from rice, wheat and swamp rice grass (Leersia hexandra) was conducted.

126

The results showed that these are highly diverse genomes with a large proportion of strain-specific variability. Some distinct differences related to host preference could be observed although the pathogenicity assays which were described in Chapter 2 revealed that all the strains used in this study are capable of infecting both rice and wheat seeds.

There was great variability amongst the eleven genomes used in this study, with respect to the basic features such as the size, number of contigs, number of protein coding sequences, RNA coding sequences and the number of subsystems. These differences could be attributed to either the intrinsic variances among the P. fuscovaginae strains and the issues arising from sequencing errors and unresolved problems of assembly. Since half of these genomes have been sequenced and assembled by several research groups, at various times and using different techniques, such discrepancies can be expected. It should also be noted that these strains have been isolated from a number of locations with a variety of environmental conditions over a period of three decades (1976-2005), thus sampling bias may exist. In addition, in previous studies, some of these isolates have shown variation in virulence on rice (Adorada et al., 2013a; Cother et al., 2009b).

Similarly, recently conducted genome comparison studies of P. fuscovaginae which included six of the genomes used in this study along with two newly drafted genomes, observed inconsistencies in basic features of these genomes (Quibod et al., 2015b). The authors also reported that the two newly drafted P. fuscovaginae-like strains, which were isolated from the same location around the same time, from the same host, had considerable variations in genome size and number of protein coding sequences.

Furthermore, with the possibilities of contamination and low-quality reads excluded, the authors suggested that the genome structures of these strains may be more complex than previously thought (Quibod et al., 2015b).

127

In contrast to the high variability among all the genomes, the newly sequenced and drafted genomes in this study appeared to be more comparable to each other in size, number of contigs, number of protein and RNA coding sequences and the subsystems. These draft genomes seem to be of high quality with an average size of 6.65Mbs (SE=76bp) consisting of 273 contigs on average (SE=1). They have 6086 (SE=3) protein-coding sequences falling into 530 (SE=6) subsystems and 71 (SE=0) RNA-coding sequences on average, and less fragmented to contigs. As the genome from UPB0736 was the best characterised, it was selected as the reference for comparison.

Whole genome alignments based on the annotated gene content obtained by RAST showed a high level of polymorphism among the strains of P. fuscovaginae examined.

The differences observed in the comparisons of genes with the reference genome, are very diverse even within the strains in the same host group. These show high variability of functionality within each group, indicating distinctive metabolic features for each strain. The high diversity in the rice-infecting group has possibly arisen from the variations in genome size and the number of coding sequences. However, similar variability in functions was observed even within the wheat-infecting group that has uniform basic features among the genomes, thus, this great diversity could be attributed to the intrinsic differences among genomes. Previous studies have also demonstrated variations in pathogenicity and virulence among the rice-infecting strains of P. fuscovaginae, by both biological and genetic evidence (Adorada, 2013). However, these intrinsic differences could conceal the host-specific variability among the strains thus, being a constraint to find host-specificity related functions that are unique to genomes in a particular group.

128

Comparison of functional annotations of individual genes reveal the metabolic versatility of P. fuscovaginae. Quibod et al. (2015b) noted that a large majority of the gene clusters were strain-specific, suggesting that P. fuscovaginae genomes have a highly accessorised pan-genome consisting of a reduced core genome and a large number of dispensable and strain-specific genes. However, in this study, genes that were selected for further analysis with respect to host-specificity were well conserved among the strains in the particular group.

Although it was difficult to recognise a large number of genes responsible for host- specificity, several hypothetical proteins, non-ribosomal peptide synthetases (NRPS), peptide synthetases, long chain fatty acid CoA ligases (LC-FAC), and large exo-proteins involved in haem utilisation and adhesion functions were found to be unique to rice- infecting strains of P. fuscovaginae. Out of 119 of those genes, 5 NRPS, 2 peptide synthatases and 6 LC-FACs, that were entirely absent in wheat-infecting strains of P. fuscovaginae, were analysed using SEED viewer PSI-blast tool and Smart-blast on the

NCBI blastp suite.

All the NRPS were found to be components of amino acid adenylation domains/NRPS domains of various syringopeptin synthetases (A, B, C,) of a number of P. syringae sub- species. The role of syringopeptins as phytotoxins are well documented (Serra et al.,

1999). P. fuscovaginae produces fuscopeptin which is a phytotoxic homologue to syringopeptin (Ballio et al., 1996). However, while both of these phytotoxins play a major role in determining virulence of P. fuscovaginae, neither are known to be host-specific

(Batoko et al., 1997b; Iacobellis, Lavermicocca, Grgurina, Simmaco, & Ballio, 1992).

Therefore, it can be hypothesised that these NRPS, being short amino acid sequences, are part of a vast and complex secretory module. Studies show that there are a large number

129

of synthetase modules involved in the non-ribosomal peptide synthesis of syringopeptin in Pseudomonas syringae pv. syringae (Scholz-Schroeder et al., 2003). The presence or absence of one or more of these phytotoxin synthetase modules may have a role in determining the host-specificity, similar to that observed in P. syringae pv. syringae

(Rezaei & Taghavi, 2014). However, feature analysis with SEED viewer comparison placed different portions of these genes with a number of pathogenicity and virulence related genes other than syringopeptins. These differences in the results can be explained by the differences in algorithms.

Of the genes that are distinctive to wheat-infecting P. fuscovaginae strains, 81.5% were hypothetical proteins. Although, 65 genes were categorised as exclusive to wheat- infecting P. fuscovaginae strains, it should be noted that these include the genes present in two out of six rice-infecting P. fuscovaginae strains (incl. UPB0736). Given that the reference genome is of a rice-infecting strain, which may not be the most suitable to identify the genes that are unique to wheat-infecting genomes.

A number of hypothetical proteins and transcriptional regulators were found to be exclusive to P. fuscovaginae strains DAR77795 and DAR77800 that were isolated from rice in Australia. In addition, UPB1013 isolated from Nepal was significantly different from other strains in the wheat-infecting group that were isolated from Mexico, suggesting a geographical variation.

Twenty-five phage-related proteins and hypothetical proteins were found only in

UPB0407 (host: Leersia hexandra, commonly known as swamp rice grass) and UPB0736

(host: rice). This deviation from other P. fuscovaginae may be a geographical variation considering that both these strains are isolated from locality in close proximity, i.e.

UPB0736 from Madagascar and UPB0407 from Burundi, with respect to the distance to

130

the origins of other strains. In previous studies, prophage-related genes have been found in large proportions in P. fuscovaginae genomes, indicating that they may have played an important evolutionary role by facilitating lateral gene transfer and may have contributed to the internal rearrangement of the genomes thus contributing to the genetic diversity and functional adaptations (Quibod et al., 2015b).

Furthermore, the genome of UPB0407 appeared to have acquired a number of phage- related genes and hypothetical proteins that were not present in other strains. Having a large number of phage-related proteins may have given UPB0407 a distinct advantage in occupying a niche of its host Leersia spp. It should be noted that, UPB0407 strain showed the highest virulence on rice seeds in pathogenicity assays, compared to strains isolated from wheat. Quibod et al. (2015b) suggested the plasticity of P. fuscovaginae genome equipped with mechanisms that allow frequent gene exchange, allows it to be an opportunistic pathogen capable of occupying a broad host range, capable of using multiple resources and colonising in different environments.

The lack of a complete reference genome is a limitation in a study such as this. For example, the total number of coding sequences of the P. fuscovaginae strains analysed in this study was greater than the number of reference coding sequences of UPB0736, except for UPB1013. This could be another reason for the low number of host-specific genes. In addition, there are a large number of genes uncharacterised. For example, the percentage of hypothetical proteins in newly sequenced genomes (UPB0526, UPB0588, UPB0589,

UPB1013, and UPB0407) is around 40%.

Although the number of genes that appear to be related to host specificity is small, most of them appear to be involved in a diverse array of pathogenicity and virulence functions.

The selected genes showed homology to functions that are considered as useful in

131

adapting to different host environments such as components of antibiotic resistance. The elements beneficial for horizontal gene transfer and components of secretory modules that are advantageous for epiphytic fitness such as phytotoxins, immunosuppressants, enzyme inhibitors, antibiotics, bacteriocins and sideophores were found. However, it should be noted that the specific function of these genes in P. fuscovaginae have not been analysed yet. Also, more than one fourth (28%) of protein encoding genes that are unique to rice-infecting strains and 81.5% of protein encoding genes that appear to be unique to wheat-infecting strains of P. fuscovaginae, remain uncharacterised. Thus, functional validation of these proteins may elucidate their role in host specificity.

3.5. Conclusion

In this study, a comparative genome analysis was conducted with a perspective of host- specificity. Draft genomes of P. fuscovaginae strains isolated from rice; DAR77795,

DAR77800, ICMP5940, SE-1, CB98818 and UPB0736 were compared with the newly- assembled draft genomes of P. fuscovaginae strains isolated from wheat; UPB0526,

UPB0588 and UPB0589, UPB1013 and UPB0497 isolated from swamp rice grass

(Leersia hexandra); a weed. The results show a highly diverse genome of P. fuscovaginae. Several hypothetical proteins, non-ribosomal peptide synthetases, peptide synthetases, long chain fatty acid CoA ligases, and large exo-proteins involved in haem utilization and adhesion functions were found to be unique to strains isolated from rice.

A large number of genes that could be associated with host specificity in both the rice- infecting and wheat-infecting groups were hypothetical proteins that are yet to be characterised. The pathogenicity assays revealed that all the strains used in this study are capable of infecting both rice and wheat. Although a limited number of P. fuscovaginae

132

strains from a small range of hosts were studied, the highly diverse genomes of P. fuscovaginae strains demonstrate their capability to adapt to different environment and hosts, highlighting the importance of P. fuscovaginae as a potential pathogen of a wide range of crops.

The availability of a completely characterised genome of P. fuscovaginae and the lack of assembled genomes of P. fuscovaginae strains isolated from different hosts, were the major limitations in this study. Therefore, genome sequencing of more strains of P. fuscovaginae followed by a transcriptional analysis will be beneficial to identify genes that are related to host specificity. Particularly, the role of various non-ribosomal peptide synthases needs to be studied further, as the previous studies conducted on similar pathogens have demonstrated that these genes, which are involved in secretion, play a role in host-specificity in addition to being major determinants of pathogenicity and virulence.

133

Chapter 4. Identification and evaluation of a sypA homologue from P. fuscovaginae

4.1. Introduction

When invading a plant, phytopathogens rely on the host plant for nutrients for their growth and survival. During this process, pathogen cells attach to the surface of a plant, overcome plant cellular barriers to enter the host plant cells and then colonise and proliferate inside the host plant by suppressing the plant immune system, while utilizing nutrients and water from the plant (Cao, Baldini, & Rahme, 2001). Infection of the host plant is a complex process that relies on the coordinated expression of a number of virulence factors. During infection, plant pathogens secrete a number of proteins to override or to evade the plant cellular defences and the plant immune system

(Abramovitch, Anderson, & Martin, 2006). Some proteins alter the plant host environment to facilitate the infection process (Büttner & He, 2009). In addition, secretory systems are in place to deliver these molecules to interact with apoplastic and cytoplasmic components of plant cell defence mechanisms (Wang et al., 2017). Key pathogenicity and virulence factors identified in Gram negative phytopathogenic bacteria are; cell wall degrading enzymes, phytotoxins, extracellular polysaccharides, phytohormones and type III effector proteins that deliver these in to host cells (Collmer,

1998). In addition to that, signal transduction (Venturi & Fuqua, 2013), iron chelation

(Loaces et al., 2011) and quorum sensing (von Bodman, Bauer, & Coplin, 2003) are also important virulence factors. Understanding the genomics behind the synthesis and secretion of these molecules is essential for managing plant diseases and designing novel techniques for disease management.

134

Infection by P. fuscovaginae results in leaf sheath lesions, sheath rot, panicle sterility and grain discolouration of rice (Tanii et al., 1976). Typical symptoms of the disease may start to appear from the seedling stage, however, most are clearly expressed at the panicle emergence stage (Chiba et al., 1977). Initially, lower leaf sheaths of infected seedlings show yellow-brown discolouration which later turn grey-brown to dark brown and often die (Cottyn et al., 1994b; Razak et al., 2009b). On adult plants, the infected flag leaf sheaths show oblong to irregular dark green and water-soaked lesions, which later become grey-brown or brown surrounded by an effuse dark brown margin (Xie, 2003). With severe infection, the entire leaf sheath turns necrotic, greyish brown or dark brown, wither and dry (Rostami et al., 2010). The disease affects the elongation of upper internodes, reduces panicle emergence, retards growth and results in various levels of sterility

(Batoko et al., 1997a). Grains of infected panicles are discoloured, deformed, poorly filled, empty and sterile or may be symptomless except for small brown spots (Cother et al., 2009b; Cottyn, 2003; Razak et al., 2009b).

These symptoms of disease are attributed to phytotoxins produced by P. fuscovaginae

(Batoko et al., 1997b). The bacterium has been shown to produce phytotoxic lipodepsinonapeptide syringotoxin and two hydrophobic lipodepsipeptides; fuscopeptin-

A (FP-A) and fuscopeptin-B (FP-B) simultaneously (Batoko et al., 1997b). These toxins collectively induce rapid and high electrolyte leakage from calli on rice cells, leading to the symptoms observed on diseased plants (Batoko et al., 1997d).

Syringotoxin is a structural analogue of syringomycin, produced by P. syringae pv. syringae (Batoko et al., 1998). Syringotoxin at low concentrations stimulate H+-ATPase activity of native right-side out vesicles on plasma membranes of cells of rice shoot, and the reverse at higher concentrations and also with inside-out membrane vesicles while,

FP-A and FP-B induce inhibition of the H+-ATPase regardless of the orientation of the

135

vesicles. However, these toxins act synergistically to inhibit ATPase activity of the plasma membranes (Batoko et al., 1998; Batoko et al., 1997d).

Fuscopeptins produced by P. fuscovaginae are structurally similar to syringopeptins produced by strains of P. syringae pv. syringae, even though the optimum conditions for syringotoxin production by P. fuscovaginae are different from those reported for P. syringae pv. syringae (Ballio et al., 1996; Bare et al., 1999; Coraiola et al., 2008; Flamand et al., 1997). Rice plants are sensitive to these toxins at all stages of growth (Batoko et al., 1997b). Although, these phytotoxins are non-host specific, the severity of the toxin damage is related to the degree of cultivar susceptibility to the pathogen (Batoko et al.,

1997b). The toxins induce a drastic inhibition of elongation of seedling and affect the elongation of the peduncle and the first internode, resulting in partial or total inhibition of panicle exertion (Batoko et al., 1997a; Batoko et al., 1994). However, the toxins enhance seed germination and appear to have no effect on the number of roots produced by seedlings (Batoko et al., 1994). Therefore, the activity of toxins on germinating seeds is a reliable tool for screening genotype susceptibility/resistance to P. fuscovaginae at early growth stages (Batoko et al., 1994).

Limited biochemical studies have demonstrated the production of these phytotoxins and their role in pathogenicity and virulence, although genetic and molecular studies on phytotoxins produced by P. fuscovaginae are limited (Patel et al., 2014). Whole genome sequencing information of DAR77795, which is one of the P. fuscovaginae strains isolated from Australia (Stodart, Pattermore, & Ash, 2013), revealed the presence of a protein homologous to syringopeptin synthetase A (sypA) of P. syringae pv. syringae, initially identified by Scholz-Schroeder, Hutchison, Grgurina, and Gross (2001). The syringopeptins of P. syringae pv. syringae are known to be key determinants of pathogenicity, virulence (Scholz-Schroeder et al., 2003) and host-specificity (Rezaei &

136

Taghavi, 2014) of the pathogen. The present study was conducted to investigate this hypothetical protein and determine its prospective role in the pathogenicity, virulence and host-specificity of P. fuscovaginae. In order to achieve this objective, a site-specific mutation of the gene corresponding to this sypA homologue protein and the biochemical and in planta functional analysis of the resulting mutant P. fuscovaginae strains were conducted.

4.2. Materials and Methods

4.2.1. Bacterial strains, plasmids

The bacterial strains and plasmids used and generated in this study are listed in Table 4.1 and Table 4.2, respectively. Escherichia coli strains DH5α, C118 and DH5α (pRK2013) were cultured at 37 °C in Luria-Bertani (LB) broth shaking at 180 rpm or on LB agar

(LBA) (Sambrook, Fritsch, & Maniatis, 1989). P. fuscovaginae strains were cultured in either KB or LB broth with shaking at 180 rpm and on LBA, at 28 °C. LB broth and LBA supplemented with appropriate antibiotics were used for selection of transformed E. coli strains and transconjugated P. fuscovaginae. Antibiotics (Sigma-Aldrich Co. LLC) were added to media as required at the following final concentrations per millilitre: 100 µg of ampicillin, 50 µg of kanamycin, 150 µg of nitrofurantoin, 25 µg of nalidixic acid. In addition, X-gal (Sigma-Aldrich Co. LLC) was used at a final concentration of 40 µg/mL in LBA medium to select transformed E. coli.

137

Table 4.1: Bacterial strains used in this study

Strains Relevant characteristics* Reference / Source Escherichia coli DH5α Cloning strain, Nalr (Hanahan, 1985) ICGEB** laboratory stock E. coli DH5α (pRK2013) Helper strain for tri- (Figurski & Helinski, parental conjugation, Kmr 1979) ICGEB laboratory stock E. coli C118 Cloning strain ICGEB laboratory stock Pseudomonas Wild-type strain isolated (Cother et al., 2009b) fuscovaginae DAR77795 from diseased rice in WT Australia, Nfr, Ampr P. fuscovaginae Wild-type strain isolated (Cother et al., 2009b) DAR77800 WT from diseased rice in Australia, Nfr, Ampr P. fuscovaginae pKNOCK mutant of This study DAR77795 SypΔ DAR77795 Nfr, Ampr, Kmr P. fuscovaginae pKNOCK mutant of This study DAR77800 SypΔ DAR77800 Nfr, Ampr, Kmr *Nalr, Kmr, Nfr, Ampr indicates nalidixic acid, kanamycin, nitrofurantoin, and ampicillin respectively. **International Centre for Genetic Engineering and Biotechnology, Trieste, Italy.

Table 4.2: Plasmids used and generated in this study

Plasmids Relevant characteristics* Reference / Source pGEM-T Easy Cloning vector, Ampr Promega Corporation pKNCOK-Km Conjugative suicide vector, Kmr (Alexeyev, 1999) pGEM+PsfSyp pGEM-T easy vector containing This study internal fragment from Pseudomonas fuscovaginae DAR77795 WT Syp homologue, Ampr pKNOCK+PsfSyp pKNoCK vector containing This study internal fragment from P. fuscovaginae DAR77795 WT Syp homologue, Kmr *Kmr, Ampr indicates kanamycin and ampicillin respectively.

138

4.2.2. Recombinant DNA techniques and sequence analysis

Genomic DNA from P. fuscovaginae strains DAR77795 and DAR77800 were extracted by Sarkosyl/Pronase lysis (Better, Lewis, Corbin, Ditta, & Helinski, 1983). For plasmid

DNA extraction, EuroGold Xchange plasmid miniprep and midiprep kits (EuroClone

S.p.A.) were used. Routine procedures for DNA manipulation such as agarose gel electrophoresis, ligations with T4 ligase, digestion with restriction endonucleases and transformation of E. coli were performed as described previously (Sambrook et al., 1989).

Primers used were designed using the Web Primer tool of the Saccharomyces Genome

Database (Stanford University, Stanford, CA94305) and manufactured by Integrated

DNA Technologies Inc. Table 4.3 presents the list of primers used in this study. PCR amplifications were performed with the GoTaq amplification (Promega) kits. Automated sequencing of some of the PCR amplified fragments were cloned into pGEM-T Easy

(Promega) and sequenced using SP6/T7 primers by Macrogen sequencing services

(Europe). Automated sequencing of some of the purified PCR fragments was also performed by the Australian Genome Research Facility (Brisbane).

Table 4.3: Primers used in this study

Primers Sequence 5’-3’ Source SP6 ATTTAGGTGACACTATAG ICGEB* laboratory stock T7 TAATACGACTCACTATAGGG ICGEB laboratory stock pKNOCK-New F CTTAACCGCTGACATGGAA ICGEB laboratory stock pKNOCK-New F TTTATTCGGACACGCGTCCT ICGEB laboratory stock PfSy-F CAATGGCAGATCGCCCAG This study PfSy-R AACCCAGGTCACCGGTCTT This study PfSy-F2 TTTTTCCAGGTGCACATACG This study PfSy-R2 ACAACACCTGGCCTACCTGAT This study *International Centre for Genetic Engineering and Biotechnology, Trieste, Italy. 139

4.2.3. Mutagenesis of the sypA gene homologue in P. fuscovaginae

In order to create a site-specific inactivation of the target gene, the gene knockout method by insertion of a suicide vector plasmid namely, pKNOCK–Km, as described by

Alexeyev (1999) was employed.

Firstly, the internal fragment PsfSyp, which is approximately 1/3 (660 bp) in size of the target gene (~2 kb) was amplified from the genomic DNA of P. fuscovaginae, DAR77795

WT, using the primers PfSy-F/PfSy-R (Table 4.3). The purified PCR fragment was ligated into pGEM-T Easy vector plasmid (Figure 4.1) and transformed in E.coli DH5α.

Figure 4.1: pGEM-T Easy vector plasmid (Promega Corporation)

Transformed E.coli DH5α bacteria cells were then spread plated on LBA containing 40

µg/mL of X-gal and Amp100. Successful transformants were selected based on ampicillin resistance and the disruption of the lacZ promoter of pGEM-T easy vector by the insertion of PsfSyp in successfully ligated plasmids. The white colonies grown on the selective medium contained the successfully ligated vector construct pGEM+PsfSyp, amongst the blue colonies, which contained non-ligated pGEM plasmids. Selected white colonies

140

were then screened for the presence of cloned internal fragment PsfSyp by colony PCR using PfSy-F/R primers under the conditions mentioned in Table 4.4.

Table 4.4: PCR conditions for PsfSyp

PCR reaction (30µL) Temp PCR program GoTaq® Green 5x PCR buffer 6µL 94 °C 5 min. 1 cycle

25mM MgCl2 1.8µL 95 °C 30 s 5mM dNTP mix 0.6µL 50 °C 30 s 30 cycles GoTaq® polymerase 5U/µL 0.15µL 72 °C 30 s 100mM primer PfSy-F 0.15µL 72 °C 7 min. 1 cycle 100mM primer PfSy-R 0.15µL 4 °C ∞ Water 21.15µL

In addition, the presence of pGEM+PsfSyp was confirmed by restriction digestions with enzymes SmaI and StuI. There are corresponding restriction sites to these enzymes within the inserted PsfSyp fragment but not in pGEM. Then, the PsfSyp internal fragment was excised from pGEM by digesting with EcoRI and ligated to pKNOCK-Km plasmid that was linearised with EcoRI and de-phosphorylated with shrimp alkaline phosphatase. The pKNOCK-Km plasmid has a kanamycin resistant marker, and a polylinker region with nine unique restriction sites (Figure 4.2).

141

Figure 4.2: pKNOCK- Km plasmid (Addgene plasmid # 46262)

Figure 4.3: Insertional mutagenesis using pKNOCK plasmids (modified from Alexeyev (1999)) (Abbreviations: ori, R6K plasmid γ -origin of replication; mob, RP4 plasmid oriT region; Res, Kanamycin antibiotic-resistance gene)

142

The ligation was transformed into E. coli C118. Successful transformations were selected based on kanamycin resistance, colony PCR using PfSy-F/R primers and restriction digestion of purified pKNOCK + PsfSyp with EcoRI. Tri-parental conjugations between transformed E. coli C118 and P. fuscovaginae wild type (WT) strains were carried with the helper strain E. coli DH5α (pRK2013). Fresh colonies of recipient P. fuscovaginae, donor E. coli C118 + pKNOCK + PsfSyp and E. coli DH5α (pRK2013) were mixed in abundance and cultured on LBA and incubated at 28 °C. Putative P. fuscovaginae sypA gene mutants were selected based on the resistance to kanamycin and nitrofurantoin.

To confirm site-specific mutation, colony PCR on putative mutants was performed with a combination of primers; PfSy-F2/pKNOCK-NewR and PfSy-R2/pKNOCK-NewF, to amplify two specific targets; F2R and R2F (approximately 1 kb each) respectively, from the sypA gene homologue in P. fuscovaginae, which includes and is located external to the inserted PsfSyp internal fragment (Figure 4.4). The PCR conditions are detailed in

Table 4.5.

PfSyF2 pKNOCK-NewF

geneX’ Res mob ori ‘geneX

pKNOCK-NewR PfSyR2

Figure 4.4: PCR amplification of F2R and R2F with PfSy-F2/pKNOCK-NewR and PfSy- R2/pKNOCK-NewF primers combinations (modified from Alexeyev (1999))

143

Table 4.5: PCR conditions for the amplification of F2R and R2F

PCR reaction (30µl) Temp PCR program GoTaq® Green 5x PCR buffer 6 µl 94 °C 5 min. 1 cycle 25mM MgCl2 1.8 µl 95 °C 30 s 5mM dNTP mix 0.6 µl 51 °C 30 s 30 cycles GoTaq® polymerase 5U/µl 0.15 µl 72 °C 30 s 100mM Primer F 0.15 µl 72 °C 7 min. 1 cycle 100mM Primer R 0.15 µl 4 °C ∞ Water 21.15 µl

The position of mutation was confirmed to be in the target region of the sypA gene homologue by sequencing of F2R and R2F amplified from both P. fuscovaginae strains

DAR77795 SypΔ and DAR77800 SypΔ.

4.2.4. Locating sypA mutation in whole genome sequences of P. fuscovaginae

strains

Verification of F2R and R2F sequences were conducted by local alignments and sequence editing using Serial Cloner version 1.3-11, CLC Genomics work bench version 5.5.1.

F2R and R2F sequences were compared and modified to construct a region approximately

1kb in size, where the mutation had occurred. The mutated region from each P. fuscovaginae strain was subjected to homology searches using BLASTX 2.2.31

(Altschul, Gish, Miller, Myers, & Lipman, 1990; Zhang, Schwartz, Wagner, & Miller,

2000) and BLASTN 2.2.32 (Altschul et al., 1997), via NCBI (Ma, McAuley et al. 2009) and GenBank databases (Benson et al., 2012), including the whole genome shot-gun assembly sequences of P. fuscovaginae DAR77795 and DAR77800 (Stodart et al., 2013).

144

4.2.5. Screening of P. fuscovaginae sypA mutants for virulence deficiency

Before preparing bacterial suspensions for further analysis, P. fuscovaginae strains

DAR77795WT, DAR77800WT and their respective sypA mutants were checked for their growth rates in nutrient rich liquid media, LB and NB (Amyl Media Pty Ltd.). The virulence of two sypA mutant P. fuscovaginae strains were compared to their respective wild types by inoculating rice and wheat seeds using the seed soaking method described previously (Adorada et al., 2013c). Bacterial cultures were grown in NB until their optical density (OD) at 600 nm was equal to 1, and diluted to a concentration of a 107 cfu/ml in sterile distilled water (SDW). Rice (cv. Amaroo) and wheat (cv. Rosella) seeds were surface sterilised with 5% NaOCl, rinsed five times with SDW and soaked for 12 hours in SDW. Seeds were then placed on sterile filter paper disks in Petri plates, inoculated with 10 mL of 107 cfu/mL bacterial suspension and sealed with Parafilm®. A treatment with SDW was used as a control. Inoculated seeds were incubated at 28 °C for 10 days.

Virulence was evaluated by measuring the shoot length (from the base of the plant to the tip of the longest leaf) and root length (from the base of the plant to the tip of the longest root) in millimetres and counting the number of roots of each seedling. Each treatment consisted of 60 seedlings, which were equally distributed among four replicates to have

15 seedlings in each. The entire experiment was repeated twice. Bacterial populations for each treatment were determined 10 days post-inoculation by standard serial dilution plating on nutrient agar (Bacto Laboratories) and Pseudomonas agar (Amyl media), with and without kanamycin50.

Virulence assays were also conducted by inoculation of rice plantlets (cv. Nipponbare) as described previously (Mattiuzzo et al., 2011). Each treatment consisted of 24 plantlets, which were equally distributed in two trays (22 x 15 x 6 cm) containing 700g of non- sterile potting soil (Structural; Snebbout, Kaprijke, Belgium) each. Bacteria were cultured

145

in LB broth and OD600 was adjusted to 1 by diluting with LB broth, before bacterial suspensions were centrifuged at 13000 rpm for 2 min. After centrifugation, the pellets were 10 times diluted with sterile saline (0.85% NaCl), with reference to the volume of

LB medium used to culture the bacteria. Bacterial suspensions were then injected into the stem of 4 weeks old rice plantlets until a bacterial droplet was observed at the stem apex.

Plants of the control treatment were injected with sterile saline. After the inoculation, the plants were placed in a saturated humid chamber (28 °C, relative humidity (RH) = 100%) for one day before they were moved to a growth chamber (28 °C, RH = 60%, 16 h photoperiod). Disease severity was evaluated 10 days after inoculation based on a rating scale (Mattiuzzo et al., 2011), with the following brief amendments, in which:

Score 0: No symptoms, only the sign of the injection puncture

1: Necrosis around the puncture extending up to 1 cm

2: Necrosis around the puncture and chlorosis from 1 to 3 cm on the new leaf

3: Necrosis around the puncture and chlorosis extending up to 5 cm on the new leaf

4: Necrosis around the puncture and chlorosis for the two-thirds of the new leaf

5: Necrosis around the puncture and chlorosis of the new leaf

6: Necrosis around the puncture and chlorosis throughout two or more leaves

The entire experiment was repeated once.

4.2.6. HPLC analysis and isolation of syringopeptin

Analyses of lipopeptides produced by P. fuscovaginae wild type and mutant strains were carried out on a reversed-phase ultra-high performance liquid chromatography-mass spectrometry (UPLC-MS) system by Marc Ongena at Microbial Processes and

Interactions (MiPI) Research Unit, Gembloux Agro-Bio Tech Faculty, University of

Liège, Belgium. For UPLC-MS, a 10 µL volume of each sample prepared from bacterial

146

cultures grown in both KB and LB media was injected. In KB, bacterial cultures were grown in 5 mL liquid medium with shaking at 150 rpm and incubated at 28 °C for 24 h.

After 24 h, the OD620 was measured and the cultures were centrifuged for 2 min at 13400 rpm. The supernatant was filter-sterilised using a 0.22 µm filter and 200 µL was taken into vials to be injected to UPLC-MS. For LB medium, cultures were grown in 6-well plates without shaking for 24 h and were treated as mentioned as above.

Analyses were carried out on a UPLC-MS (Waters, Acquity class H) coupled with a single quadrupole detector (Waters, Acquity) on an Acquity UPLC BEH C18 (Φ

2.1 × 50 mm, 1.7μm, Waters, Zellik, Belgium). Both instruments were controlled with the

MassLynx software (Waters). Compounds were eluted with a linear gradient (from 15% to 95% within 3.5 min) of acetnitrile acidified with 0.1% formic acid in water acidified with 0.1% formic acid at a constant flow rate 0.6 mL/min at 40 °C. Compounds were monitored in the positive ion mode and electrospray in-source settings in the SQD were as follows: source temp: 130 °C; desolvation temp: 400 °C; nitrogen flow: 1000 liters/h; cone voltage: 60V. The wild type P. fuscovaginae strain UPB0736 (Patel et al., 2014) was used as a reference strain. Each analysis was repeated three times.

4.2.7. Statistical analysis

In seed soaking assays, measurements of the three parameters (shoot length, root length, and number of roots) were taken from a total of 60 seedlings per treatment. Count data

(i.e. number of roots) were square root transformed to obtain a normal distribution of data before analysis of variance. One-way analysis of variance test (ANOVA) was performed for each parameter. Means of the five treatments (P. fuscovaginae DAR77795WT, P. fuscovaginae DAR77800WT, P. fuscovaginae DAR77795SypΔ, P. fuscovaginae

DAR77800SypΔ, SDW control) were compared and separated by Tukey’s test at 5%

147

confidence interval using SAS software, version 6.12 (SAS Institute, USA). The error bars were calculated by determining the standard errors of the means for each treatment.

The results from each time the experiment was conducted were analysed independently.

Disease score data from virulence assays by inoculation of rice plantlets were averaged and then converted to percent disease index (PDI), taking the score of six as the highest disease incidence (100%). The PDIs of treatments were analysed using chi-square test at

5% confidence level. Data from each time-wise replication were analysed separately.

4.3. Results

4.3.1. Sequence analysis indicates that the mutation occurred in sypA

homologue

A partial gene encoding for a hypothetical protein (2026bp) from P. fuscovaginae

DAR77795 genome was identified and 80% of its nucleotide sequence showed 81% homology to a gene encoding for syringopeptin synthetase (sypA) of Pseudomonas syringae pv. syringae (Pss) (GenBank: AF286216.2) (Figure 4.5). Furthermore, 60% of the nucleotide sequence showed 77% homology to a gene in the sessilin biosynthesis gene cluster of Pseudomonas sp. CMR12a (GenBank: JQ309920.1).

148

Figure 4.5: BLAST results for Pseudomonas fuscovaginae DAR77795 hypothetical protein (2026 bp)

PsfSyp insert (661 bp) used in the site-specific inactivation by pKNOCK suicide vector was selected from the middle third (845-1505 bp) of the hypothetical protein homologous to sypA, which was 2026 bp long. Sequence analysis revealed that F2R and R2F sequences amplified from the putative mutants were approximately 1 kb in length and contained the PsfSyp insert flanked by parts of the pKNOCK genome and parts of the P. fuscovagiane sypA homologue gene that are located external to PsfSyp insert. The ~1 kb region where the insertional mutation occurred is located within the 555 – 1576 bp region of sypA homologue that corresponds to the 5220-6599 bp region of Pss sypA. This particular region in sypA encodes for peptide synthetases. The sypA homologue is located in P. fuscovaginae DAR77795 whole genome assembly contigs 317 and 35. PsfSyp insert and mutated region correspond to 267-1376 bp and 6-1373 bp region of DAR77795WGS contig 317, respectively. In P. fuscovaginae strain DAR877800, the mutated region is

149

located in the 733-1640 bp region in contig 538 of the whole genome assembly of

DAR77800 (Figure 4.7). Mutated regions of both P. fuscovaginae strains were identical.

The mutation was found to have occurred in a region, which encodes for amino acid adenylation domains of non-ribosomal peptide synthetases (Figure 4.6).

Figure 4.6: Conserved domain hits that match with the mutated region (1023 bp) of the sypA homologue in Pseudomonas fuscovaginae DAR77795 In this concise display, RF indicates the open reading frames; +1, +2, +3, -1, -2, -3. The cyan-coloured regions in each RF of the query represent compositionally biased regions that were detected by the low-complexity filter, which alleviates the inaccurate annotation of the query sequence. Under each RF, the superfamily to which a specific hit belongs is shown in pastel- pink color denoting less confidence association of the query region with the conserved domain model, thus indicating low confidence level in the inferred function of the query protein. The small triangles in light pink colour, under the query indicate the conserved features/ sites such as binding or catalytic sites.

Figure 4.7: Location of sypA homologue in Pseudomonas fuscovaginae DAR77795 and DAR77800 genome assembly, as indicated by the blue coloured arrows

150

The sypA homologue in DAR77795 is found to be associated with other pathogenicity related genes such as syr-P like protein and ABC-type siderophore export system. In Pss, this gene is located in a syp-syr gene cassette. Therefore, all available whole genome sequences of P. fuscovaginae were BLASTed with Pss B301D (Wang et al., 2006) sypA,

B, C gene sets. In at least five genomes (DAR77795, DAR77800, UPB0407, UPB0526,

UPB0588, UPB0589, UPB1013) of P. fuscovaginae, the sypA, sypB, sypC genes are found to be located in the same or adjacent contigs, indicating the presence of a syp gene cassette in P. fuscovaginae.

Figure 4.8: Sequence alignment of selected 1631 bp regions of sypA homologues from Pseudomonas fuscovaginae whole genome sequences corresponding to DAR77795 sypA homologue 358 – 1989 bp.

Figure 4.9: Molecular phylogenetic analysis of the selected 1631 bp regions of sypA homologues from Pseudomonas fuscovaginae whole genome sequences corresponding to DAR77795 sypA homologue 358-1989 bp, inferred using the maximum-likelihood method. The tree is drawn to scale with branch lengths measured in number of substitutions per site.

151

4.3.2. HPLC analysis reveals lack of production of phytotoxins in mutants

As shown in Figure 4.10, chromatographic peaks with masses corresponding to fuscopeptins were observed for the reference P. fuscovaginae UPB0736 wild type strain at retention time from 2.00 - 2.50. The same compounds were detected for the two P. fuscovaginae strains DAR77795WT and DAR77800WT grown in KB medium.

Comparison of relative amounts of fuscopeptins produced by the five isolates in KB (y axis in same scale) clearly shows suppression of synthesis in the mutant P. fuscovaginae

DAR77795SypΔ and DAR77800SypΔ. For DAR77795WT and DAR77795SypΔ, the two main components (ions in the range m/z 1550) accumulated in much larger quantites in LB compared to KB (data not shown).

152

Figure 4.10: UPLC-MS chromatogram for bacterial exudates of Pseudomonas fuscovaginae strains DAR77795, DAR77800, and their respective sypA homologue mutants, in comparison to reference strain UPB0736 prepared in KB medium. The chromatographic peaks seen at retention time from 2.00 to 2.50 for the reference P. fuscovaginae UPB0736 wild type strain, have masses corresponding to fuscopeptins.

153

4.3.3. sypA mutant P. fuscovaginae strains are less virulent than their

respective wild types

P. fuscovaginae strains DAR77795WT, DAR77800WT and their respective sypA

mutants were found to have similar growth rates as indicated by OD600 values of the broth cultures (Figure 4.11).

Figure 4.11: Growth of Pseudomonas fuscovaginae cultures in nutrient broth, shaking at 180 rpm, at 28 °C as indicated by change in OD600 over time

Seed inoculation

Shoot lengths and root lengths of rice and wheat seedlings treated with sypA mutants

DAR77795SypΔ and DAR77800SypΔ were significantly (P>0.05) greater compared to those treated with their respective wild types (Figures 4.12 and 4.13). There was no significant (P>0.05) difference on root or shoot growth of seedlings of rice or wheat inoculated with wild type strains DAR77800WT and DAR77795WT. Similarly, the effects of mutation on seedling growth showed no significant difference (P>0.05) between the two mutant strains, DAR77795SypΔ and DAR77800SypΔ for both rice and wheat. When compared to the control SDW treatment, the shoot and root lengths for both 154

rice and wheat were significantly (P>0.05) reduced when treated with the mutants as compared to their respective wild type strains. Although the numbers of roots of seedlings treated with both mutants were comparable to that of the control on rice seeds, on wheat seeds, both the mutants caused significantly (P>0.05) lower numbers of roots compared to the control. In general, there was a clear contrast between the virulence effects of the wild types and the mutants, on both rice and wheat seeds.

155

Figure 4.12: Growth response of rice (cv. Amaroo) to seed soaking with 107 cfu/mL of Pseudomonas fuscovaginae. Length of shoot and root for each treatment is the mean of four replicates, each containing 15 seedlings. Bars with the same letter (upper case and lower case) are not significantly different at the 5% level.

Figure 4.13: Growth response of wheat (cv. Rosella) to seed soaking with 107 cfu/mL of Pseudomonas fuscovaginae. Length of shoot and root for each treatment is the mean of four replicates, each containing 15 seedlings. Bars with the same letter (upper case and lower case) are not significantly different at the 5% level.

156

Inoculation of seedlings

Seedlings inoculated with sypA mutants of P. fuscovaginae DAR77800WT and

DAR77795WT expressed significantly (P>0.05) lower PDIs in comparison to those inoculated with the wild type, and significantly (P>0.05) higher PDIs compared to the control treatment (Figure 4.14). There was no significant difference (P>0.05) in PDI among the seedlings treated with the two mutant P. fuscovaginae strains,

DAR77795SypΔ and DAR77800SypΔ.

Figure 4.14: Percent disease index of rice (cv. Nipponbare) plantlets inoculated with 107 cfu/mL of Pseudomonas fuscovaginae. PDI for each treatment was calculated with 24 seedlings equally distributed in to two blocks. Bars with the same letter are not significantly different at the 5% level.

157

4.4. Discussion

A hypothetical protein from P. fuscovaginae DAR77795 genome revealed significant homology to syringopeptin synthetase (sypA) of Pseudomonas syringae pv. syringae

(Pss) and the sessilin biosynthesis gene cluster of Pseudomonas sp. CMR12a. In Pss strain B301D, sypA gene was 16140 bp long and is a part of a gene cluster of 73,800 bp, which includes syringopeptin synthetase B (sypB) and syringopeptin synthetase C (sypC) genes (Scholz-Schroeder et al., 2003). This gene cluster exists in a 132 kb operon along with a cluster of syringomycin synthetase genes (Wang et al., 2006). These non- ribosomal peptide synthetases encoded by syr and syp genes, are responsible for the biosynthesis of the lipodepsipeptide toxins, syringopeptin and syringomycin, in Pss

(Bender et al., 1999). These syr and syp gene clusters are located adjacent to one another on the chromosome (Scholz-Schroeder et al., 2003). Known as the syr-syp box, conserved promoter sequences of the syr and syp genes co-regulate the production of syringomycin and syringopeptin (Wang et al., 2006). Both syringomycin and syringopeptin are major virulence determinants of the plant pathogen Pss, due to their necrosis-inducing properties (Iacobellis et al., 1992).

P. fuscovaginae produces lipodepsipepetides fuscopeptins A and B which are structurally and functionally similar to syringopeptins produced by Pss (Ballio et al., 1996). In addition, syringotoxin secreted by P. fuscovaginae is a structural analogue of syringomycin secreted by Pss (Batoko et al., 1998). A common functional trait of these bacterial toxins is their permeabilising activity on biological membranes (Coraiola et al.,

2008). The interaction of these hydrophobic molecules with the lipid bilayer of cell membranes creates cation-selective channels, alters membrane potential and cause the intracellular fluids to leak out of the cell (Coraiola et al., 2008), thus facilitating the point of entry and providing nutrients to invading pathogens. Therefore, these phytotoxins can

158

be determined as an essential component of the infection process, which are thus vital to the pathogenicity, and virulence of P. fuscovaginae.

Disease symptoms of sheath brown rot, particularly cell necrosis in rice leaf sheath

(Flamand et al., 1997) and poor panicle emergence due to inhibition of culm elongation during heading (Batoko et al., 1997b), are attributed to the bioactivity of bacterial toxins produced by P. fuscovaginae. Although these phytotoxins are non-host specific, the severity of the toxin damage was observed to be related to the degree of cultivar susceptibility to the pathogen (Batoko et al., 1997b). P. fuscovaginae bacterial toxins also have surfactant and anti-fungal properties (Flamand et al., 1997) which might provide the pathogen a competitive advantage over the other microorganisms present on the plant surfaces. Therefore, these toxins are considered as an integral component of host- pathogen interactions.

Many of the earlier studies on the phytotoxins of P. fuscovaginae focused on biochemical characterisation. Recently, the role of a sypC homolog gene in biosynthesis of P. fuscovaginae phytotoxins has been reported (Patel et al., 2014). In this study, a disruption in biosynthesis of the sypA gene homologue in P. fuscovaginae demonstrates inability to produce phytotoxic lipopeptide fuscopeptins followed by a significant reduction in its virulence and disease symptoms on rice and wheat seedlings, and hence provides more genetic information on P. fuscovaginae phytotoxins.

In addition, 60% of the nucleotide sequence of the hypothetical protein of P. fuscovaginae

DAR77795 showed 77% homology to the sessilin biosynthesis gene cluster of the bio- control strain Pseudomonas sp. CMR12a (GenBank:JQ309920.1). These are non- ribosomal peptide synthetase genes that are responsible for the biosynthesis of sessilin, which is a cyclic lipopeptide having bio-surfactant and antimicrobial properties (D'aes,

2012). Mutant Pseudomonas sp. CMR12a, incapable of sessilin production, showed

159

reduced biofilm formation but enhanced swarming motility, indicating that it facilitates a sessile lifestyle. If this non-ribosomal peptide synthetase gene in P. fuscovaginae is also involved in sessilin biosynthesis, it can be hypothesised that sessilin might be involved in the pathogenicity and virulence of P. fuscovaginae by increasing its competitive advantage over other microorganisms on the plant surfaces. In addition to that, virulence of P. fuscovaginae may be increased by inducing expression of virulence related genes that are governed by quorum sensing (Mattiuzzo et al., 2011) which in turn may be facilitated by swarming. Sessilin mutants were completely incapable of controlling plant pathogenic fungi, which indicates that sessilin plays an important role in the biocontrol ability of Pseudomonas sp. CMR12a (D'aes et al., 2014).

Further validation of the genetic screening by complementation of the knockout mutants was not attempted in this study. In a previous study, knock-out mutation on a similar gene target of sypC in P. fuscovaginae failed to generate complementation due to the target having an unusually large open reading frame (ORF) (Patel et al., 2014). Likewise, the sypA gene of Pss B301D (to which the gene mutated in this study is homologous) has a very large ORF. This sypA gene is 16140 bp in length and is a part of a gene cluster of

73,800 bp (Scholz-Schroeder et al., 2003), which is present in a 132 kb operon along with a cluster of syringomycin synthetase genes (Wang et al., 2006).

Seed inoculation with purified bacterial toxins (Batoko et al., 1994; Batoko et al., 1997c) and crude bacterial suspension (Adorada et al., 2013c) have both been validated as a reliable tool for early detection of resistance of rice cultivars to P. fuscovaginae. As for the optimum bacterial suspension concentration for inoculations, 108-1011 cfu/mL concentrations are considered too high and 104-105 cfu/mL concentrations are considered too low. An optimum concentration of 107 cfu/mL has been established for inoculating rice seeds (Adorada et al., 2013c) and for recording of disease severity by inoculation of

160

plantlets of rice using either the pinprick method (Patel et al., 2014) or injection of bacterial suspension (Mattiuzzo et al., 2011).

Reduction in the length of seedlings raised from the seeds soaked with purified bacterial toxins of P. fuscovaginae has been reported previously (Batoko et al., 1994). Instead of purified toxins, crude bacterial suspensions were used in the present study. Seedling heights of the seeds treated with mutants of phytotoxin sypA homologue were greater than those treated with their respective wild types, validating the previous findings. Reduction of seedling height, root length and number of roots in rice seedlings from rice seeds inoculated with P. fuscovaginae bacterial suspension at 107 cfu/mL was also reported

(Adorada et al., 2013c).

Although seed soaking in purified bacterial toxins had no effect on root number and root growth (Batoko et al., 1994), in this study, both DAR77795SypΔ and DAR77795SypΔ, which were incapable of producing a phytotoxin homologue, showed significantly

(P>0.05) higher root growth in inoculated rice and wheat seeds, compared to seeds treated with their respective wild type strains. Root numbers of rice seedlings infected with mutants were comparable to those from the control treatment of sterile distilled water.

This indicates the potential effect of the toxin on root growth of infected seeds, which could hinder the establishment of seedlings in field. Poor seedling establishment had been reported previously as a result of P. fuscovaginae infection on seeds (Adorada et al.,

2014). This can be attributed to the poor development of root systems in seedlings infected with P. fuscovaginae.

Although the rice cv. Amaroo used in this study is known to be moderately susceptible to

P. fuscovaginae at both seedling and mature plant stages (Adorada et al., 2013c), little is known about the resistance status of the wheat cv. Rosella to P. fuscovaginae. However,

P. fuscovaginae is known to be a pathogen of wheat (Duveiller & Maraite, 1990) and

161

previous experiments have demonstrated that cv. Rosella is susceptible to infection by P. fuscovaginae at the seedling stage. The suitable concentration (cfu/mL) of bacterial cell suspension to be used for resistance/susceptibility screening of wheat germplasm is not known. Therefore, it is not clear if the bacterial suspension of 107 cfu/mL concentration was too high for wheat seed inoculation or the wheat seedlings of cv. Rosella were more susceptible to the phytotoxins produced by P. fuscovaginae, in comparison to rice seedlings of cv. Amaroo. In previous experiments described in Chapter 2, conducted with ten wheat cultivars and ten P. fuscovaginae strains, wheat seedlings arising from seeds inoculated with bacterial cells at a concentration of 107 cfu/mL showed susceptibility to

P. fuscovaginae with significantly (P>0.05) different cultivar x strain interactions.

Therefore, it cannot be concluded if the susceptibility of wheat cv. Rosella to phytotoxins produced by P. fuscovaginae is a trait that could be attributed to cultivar susceptibility or a trait that is common to wheat in general, such as the high permeability of the wheat seed coat to bacterial phytotoxins compared to rice seed coat. Reactions of the seed coat to invading pathogens are very specific to cultivar type (Radchuk & Borisjuk, 2014). The activity of purified bacterial toxins has been studied on rice cells and were shown to change the membrane potential and exert a detergent-like activity on the lipid bi-layer of cell membranes (Batoko et al., 1998) causing the cells to leak electrolytes irrespective to the orientation of transport vesicles (Batoko et al., 1997d). This provides insight into the role of bacterial toxins in facilitating pathogenicity, but no such studies have been conducted specifically on wheat cells. Therefore, further studies should be conducted to gain a deeper understanding of the molecular plant-pathogen interactions of P. fuscovaginae with wheat seeds and seedlings.

Different levels of resistance in rice cultivars are displayed to P. fuscovaginae and are dependent on the growth stage of the plant (Adorada et al., 2013c), although rice is

162

susceptible to bacterial toxins at all growth stages (Batoko et al., 1997b). Inoculation with a needle dipped in a bacterial cell suspension (pinprick method) is considered as the most appropriate method of inoculation to obtain an accurate estimate of the amount of disease produced (Bua, Adipala, & Opio, 1998). This method has been used in previous studies where P. fuscovaginae was inoculated into rice at different growth stages, such as 1- month old plantlets (Patel et al., 2014), seedling, panicle initiation (early booting) and panicle exertion (heading) stages (Adorada et al., 2013c). Similarly, the method was also used to successfully evaluate the disease severity of P. fuscovaginae on 1-month old rice plantlets (Mattiuzzo et al., 2011) and at early booting stage (Detry et al., 1991b).

Appearance of brown coloured, water-soaked and necrotic lesions at the point of infection is a characteristic symptom of the sheath brown rot disease caused by P. fuscovaginae.

These symptoms are attributed to the necrotic activity of phytotoxins produced by the bacteria and are assumed to be common for a range of host plants and cultivars. Therefore, the size of the lesions is an appropriate measurement of virulence to discriminate P. fuscovaginae sypA mutants from their wild types. In this study, a scoring method

(Mattiuzzo et al., 2011) was used to evaluate the disease incidence based on the appearance of necrotic lesions. In the results of this experiment, the sizes of necrotic lesions were expressed as a percentage disease index (PDI), based on the scoring system.

The PDIs from sypA mutants were significantly (P>0.05) lower than their respective wild types. In addition, there was no significant (P>0.05) difference in disease severity between the two P. fuscovaginae strains, for both the two sypA mutants and their respective wild types. These observations regarding the virulence of sypA mutants and their respective wild types agree with the observations from the inoculation of seeds

(shoot length, root length, number of roots).

163

In general, DAR77800WT and DAR77795WT were observed to have similar virulence to each other as measured by the growth of infected seedlings (shoot length and root length) and disease severity on infected rice and wheat plantlets (PDI). However, the observations from earlier virulence assays conducted on mature rice plants reported that

DAR77800WT is less virulent than DAR77795WT (Cother et al., 2009b). The results of the present study agree with the observation that P. fuscovaginae toxins, such as the sypA homologue studied here, are known to be effective at all growth stages of the host plant

(Batoko et al., 1997b). However, it should be noted that the results of inoculation at seedling or plantlet stage might not accurately represent the impact of disease at panicle initiation (booting) stage and panicle exertion (heading) stage, which are the more economically significant as reflected in the harvest. Degrees of cultivar susceptibility of rice that were determined based on disease incidence (measured by necrotic lesion like symptoms) is reported not to correlate with those that were determined based on the inhibition of panicle exertion (Detry et al., 1991b). Furthermore, it has been reported that there are different levels of host-pathogen interactions of P. fuscovaginae at different growth stages of the host plant, that are expressed by varying disease severity (Adorada et al., 2013b). Therefore, it is important to investigate the pathogen factors other than the phytotoxins, which could determine the different host-pathogen interactions, causing variations in disease severity.

This study shows that the sypA homologue in P. fuscovaginae is involved in virulence related functions and the bacterial toxin, which sypA homologue gene mutants are incapable of producing, is an integral component of host-pathogen interactions, although, it is unlikely to be involved in host-specificity. These observations agree with previous reports that purified bacterial toxins of P. fuscovaginae are non-host specific and disease severity correlates with cultivar susceptibility (Batoko et al., 1997b).

164

4.5. Conclusions

Although it has been recognised as a prevalent and serious plant pathogen of a broad range of economically important hosts, limited studies have been conducted on the pathogenicity and virulence mechanisms of P. fuscovaginae. This study investigated the role of a hypothetical protein identified from the whole genome sequences of P. fuscovaginae strains DAR77795 and DAR77800, with homology to syringopeptin synthetase A (sypA) of P. syringae pv. syringae, which is a non-ribosomal peptide synthetase. This hypothetical protein is likely to be involved in the production of phytotoxic fuscopeptins produced by P. fuscovaginae, which are structurally and functionally similar to syringopeptins of P. syringae pv. syringae. Virulence assays were conducted on mutant P. fuscovaginae strains, in which a region encoding for amino acid adenylation function of this hypothetical protein is obstructed. Inoculation of seeds and plantlets of rice and wheat with P. fuscovaginae wild types and their respective sypA mutants showed that the mutation significantly reduced the appearance of disease symptoms, which in turn is known as an effect of phytotoxins. It was evident that virulence of the mutants is reduced when compared to that of their respective wild types.

This study confirms that this hypothetical protein homologous to sypA is a major pathogenicity and virulence determinant of P. fuscovaginae, subjected to further validation by complementation of the knockout mutants, which was not attempted in this study. Apparently, this sypA homologue is not involved in determining the host- specificity. Therefore, other hypothetical proteins found in P. fuscovaginae genomes such as those analogous to large exo-proteins with adhesion functions, would be worth characterising.

165

Chapter 5. Identification and Characterization of Adhesins in P. fuscovaginae

5.1. Introduction

For bacterial pathogens, attachment (adherence), penetration (entry), movement, colonisation, invasion and suppression of host immunity are the most common steps in pathogenesis (Bastas & Kannan, 2015). Attachment to the plant surface is the first and one of the most important processes in pathogenicity, as it is the first point of contact between the pathogen and the plant (Romantschuk, 1992). Most plant pathogenic bacteria enter the plant through natural openings such as stomata and hydathodes, which are natural openings on the surface and the tips and margins of leaves, respectively (Huang,

1986; Zeng, Melotto, & He, 2010). Plants have a system to detect invading pathogens, which triggers the plant immune responses when a pathogen cell is in the vicinity (Kline,

Fälker, Dahlberg, Normark, & Henriques-Normark, 2009). Plant surface defence mechanisms either deter the attachment of pathogen to plant surface or entrap and destroy the pathogen (Jones & Dangl, 2006). In order to commence the pathogenesis process, plant pathogens must override these and attach successfully. Therefore, bacterial attachment is a crucial step in pathogenicity. Bacterial attachment to the host surface is mediated by specific bacterial adhesins which dock and bind with host surface receptors

(Klemm & Schembri, 2000).

Adhesins are a group of specialised surface-associated proteins that collectively contribute to and facilitate this attachment and colonisation by the pathogen (Kline et al.,

2009). They are polypeptides that can be divided into two types; fimbrial and afimbrial

(non-fimbrial) adhesins (Klemm & Schembri, 2000). Afimbrial adhesins are short hair- like structures made up of single proteins which enable closer contact with the host plant

166

surface (Gottig, Garavaglia, Garofalo, Orellano, & Ottado, 2009). Fimbrial adhesins are multiprotein structures that produce long hair-like appendages called pili, protruding from the bacterial cell surface (Soto & Hultgren, 1999). Fimbriae are an important feature for adherence in Gram-negative bacteria (Klemm, 1985). Feil, Feil, and Lindow (2007) proposed that fimbrial adhesins are involved in cell-to-cell aggregation while afimbrial adhesins contribute to initial attachment to host surfaces. Compared to pathogens of humans and animals, the role of these proteins in virulence in plant pathogens has been not been studied in detail (Rojas, Ham, Deng, Doyle, & Collmer, 2002).

Many members of the plant root-associated bacteria family Rhizobiaceae mediate direct attachment to plant root hair cells though a Ca2+ ion-dependent adhesins called rhicadhesin (Smit, Logman, Boerrigter, Kijne, & Lugtenberg, 1989). Pili- like structures are involved in attachment of Bradyrhizobium japonicum to soybean roots (Vesper &

Bauer, 1986). Monitoring bacterial attachment to leaf surface and entry to leaves at early stages by enhanced green florescent protein (egfp) tagged X. oryzae pv. oryzae cells showed that there is a set of adhesin-like proteins associated with the virulence of the rice pathogen, and have preferential involvement of each one at different stages of the disease process (Das, Rangaraj, & Sonti, 2009). These various adhesins are secreted by different secretory pathways that are highly conserved (Gerlach & Hensel, 2007).

The role of fimbrial adhesin Type IV pili has been established as a virulence factor for a number of plant pathogenic bacteria. Type IV pili of X. campestris pv. hyacinthi have been suggested to play a role in attachment to stomata of hyacinth leaves (van Doorn,

Boonekamp, & Oudega, 1994). In Ralstonia solanacearum, mutation in pilA gene encoding for Type IV pilin has been shown to inhibit the capability of polar attachment to tobacco cells in suspension culture and tomato roots (Kang, Liu, Genin, Schell, &

Denny, 2002). Mutation in the pilB gene encoding for nucleotide binding protein involved

167

in providing energy for pilin translocation and assembly in X. fastidiosa, affects its colonisation and upstream migration in xylem vessels (Meng et al., 2005). In addition, a mutation in pilQ gene encoding for the type IV pilus secretin (Drake & Koomey, 1995) reduces virulence of R. solanacearum in tomato plants (Liu, Kang, Genin, Schell, &

Denny, 2001) and affects colonisation and upstream migration of X. fastidiosa in xylem vessels (Meng et al., 2005). However, in X. oryzae pv. oryzae, pilQ have no effect on attachment to leaf surface and entry, or the ability to cause disease after surface inoculation, although, it is always important for virulence after wound inoculation (Das et al., 2009).

Type IV pilus is one of the best-characterised adhesins in Pseudomonas spp. as well. Pili like structures have been shown to be involved in stomatal attachment of P. syringae pv. phaseolicola to bean leaves (Romantschuk & Bamford, 1986). The type IV pilus is important for the adhesion, the epiphytic fitness and the survival of the bean pathogen P. syringae pv. syringae (Suoniemi, Björklöf, Haahtela, & Romantschuk, 1995). Although

Type IV pilus is reported to have no role in bacteria attachment of the tomato pathogen

P. syringae pv. tomato strain DC3000, it appeared to be important for epiphytic fitness and survival of the pathogen (Roine, Raineri, Romantschuk, Wilson, & Nunn, 1998). In addition, it is involved in the surface motility and virulence of P. syringae pv. tabaci on tobacco plants (Nguyen et al., 2012; Taguchi & Ichinose, 2011).

Type V protein secretion system of gram negative bacteria is a simple protein secretion system associated with translocation of larger proteins or protein domains; mostly adhesion and haemolysins, to the outer membrane. It comprises two distinct pathways, the autotransporter pathway and the two-partner secretion pathway (TPS) (van Ulsen, ur

Rahman, Jong, Daleke-Schermerhorn, & Luirink, 2014). TPS consists of a transported protein TpsA and the specific transporter TpsB. TpsA have TPS domain essential for

168

protein secretion and directs the secretion of entire protein through the channel-forming, porin-like TpsB that localizes in the outer membrane (Jacob-Dubuisson, Fernandez, &

Coutte, 2004). Numerous studies on Type V adhesins from animal and human pathogens show that they are required for optimal virulence. In X. oryzae pv. oryzae, mutants of

Yersinia autotransporter-like protein H (YapH) showed lack of attachment to leaves or entry through wounds. Although this gene appeared to be not essential for causing disease by surface inoculation, it was crucial for virulence by wound inoculation (Das et al.,

2009).

TPS type filamentous haemagglutinin (Fha) of Bordetella pertussis consists of a fimbrial adhesin FhaB which is the TpsA, secreted through the TPS partner FhaC, which is the

TpsB (Jacob‐Dubuisson, Locht, & Antoine, 2001). In Erwinia chrysanthemi (Syn.

Dickeya dadantii), a mutation in the hecA gene which is homologous to FhaB of B. pertussis, causes deficiency in surface attachment and aggregate formation on tobacco leaves thus reducing virulence following surface inoculation (Rojas et al., 2002). In

Xanthomonas axonopodis pv. citri, FhaB-like protein XacFhaB is expressed in planta during plant-pathogen interaction and is demonstrated to be involved in virulence in both epiphytic and wound inoculations. Mutation of this gene affected the attachment to leaf surfaces and biofilm formation. Mutant cells showed faster movement in swarming attributed to the lack of cell to cell aggregations because no hyperflagellation was observed (Gottig et al., 2009). In addition, the mutants excreted more exopolysaccharides facilitating motility. XacFhaB has a transporter protein XacFhaC close to it in the genome of X. axonopodis pv. citri. However, mutation of XacFhaC showed only intermediate virulence phenotype indicating that there is another transporter protein involved for

XacFhaB (Gottig et al., 2009). Mutations in Fha-like paralogs XadA and XadB in X. oryzae pv. oryzae showed deficiency on leaf attachment or entry. XadA function appeared

169

to be the most important factor to cause disease by surface inoculation, compared to

XadB. However, both XadA and XadB appeared to have no effect on virulence following wound inoculation (Das et al., 2009). In X. fastidiosa, mutations in the fimbrial adhesins fimA, fimF, and afimbrial adhesins xadA and hxfB shows reduced virulence on grape plants. However, random insertional mutation of a Fha homologue in X. fastidiosa produced hypervirulent strains with more severe symptoms and earlier grapevine death

(Feil et al., 2007).

The aforementioned studies sufficiently demonstrate that adhesins/filamentous hemagglutinins mediate the contact between bacterial cells; therefore, a mutation impairs the movement of the pathogen inside the plant and reduces the virulence. However, very limited information is available for adhesins/filamentous hemagglutinins of P. fuscovaginae.

P. fuscovaginae is capable of epiphytic and endophytic colonisation on both asymptomatic and symptomatic rice seeds and endophytic colonisation in roots, stems and leaves (Adorada et al., 2014). The ability to colonise a range of plant cell surfaces both epiphytically and endophytically indicates strong and diverse adherence properties.

Spray inoculation of P. fuscovaginae has been shown to cause the same amount of disease on rice plants, thus, it is as effective as wound inoculation (Adorada et al., 2013c).

Therefore, the mechanisms used by P. fuscovaginae to attach to leaf surface and the involvement of them in host-pathogen interactions; particularly, in pathogenicity and virulence of the pathogen, is an area of interest to study.

Patel et al. (2014) reported the inactivation of a locus of a gene in P. fuscovaginae strain

UPB0736, which showed significant homology to Type IV pili biogenesis gene pilZ, resulted in reduced virulence on rice. Another study conducted on gene expression of P. fuscovaginae-like pathogen during infection on rice reveals the expression of Type IV

170

pili biosynthesis gene for membrane adhesion during the first three hours after infiltration

(Quibod et al., 2015b). As demonstrated by the aforementioned studies, Type IV pili have a number of functions including adhesion, motility, secretion and virulence in Gram- negative bacteria. In addition, a gene encoding for flagellin with motility functions was largely expressed throughout the entire time of infection of P. fuscovaginae-like pathogen on rice, indicating its role in bacterial movement inside the host tissue (Quibod et al.,

2015b). It is evident that these adhesin-like proteins are involved in pathogenicity and virulence of P. fuscovaginae.

Analysis of draft genomes of P. fuscovaginae strains DAR77795 and DAR77800 reveals the presence of genes encoding several large hypothetical proteins, which have sequence homology to some of the well-characterised adhesins and hemagglutinins. The role of these type of adhesins and hemagglutinins in the host-pathogen interactions of P. fuscovaginae has not been studied yet.

With other Gram negative bacteria that are pathogenic to humans and animals, adhesins have been identified as a factor governing host specificity (Kline et al., 2009; MacKenzie,

Palmer, Köster, & White, 2017; Stones & Krachler, 2016; Yue et al., 2015). It has been suggested that the knowledge of bacterial adhesins could be utilized to devise novel methods to control diseases (Pan, Yang, & Zhang, 2014; Stones & Krachler, 2015). As demonstrated in the previous studies in this thesis, P. fuscovaginae strains display significant host cultivar-pathogen strain interactions. However, studies on the role of adhesins in host specificity in plant pathogens have hardly been explored. It has been demonstrated that the adhesins in Xanthomonads are likely to be involved in adaptation to their host plants (Mhedbi-Hajri et al., 2011). Therefore, the aim of the study described in this chapter was to investigate the role of one of the aforementioned adhesin-like proteins in the host-pathogen interactions of P. fuscovaginae.

171

5.2. Materials and Methods

To study the role of Adhesin 1 in virulence and pathogenicity of P. fuscovaginae, two mutagenic techniques were attempted to create a mutant with site-specific inactivation of the target gene.

- by deletion with replacement vector plasmid pEX18 (Suksomtip &

Tungpradabkul, 2005)

- by insertion of suicide vector pKNOCK (Alexeyev, 1999)

5.2.1. Bacterial strains, plasmids and culture media

The bacterial strains used and generated in this study are listed in Table 5.1. The plasmids used and generated in this study are listed in Table 5.2. Escherichia coli strains DH5α,

C118 and were cultured at 37 °C in Luria-Bertani (LB) broth shaking at 180 rpm or on

LB agar (LBA) (Sambrook et al., 1989). P. fuscovaginae strains were cultured in KB and/or LB broth shaking at 180rpm and on LBA, at 28 °C. LB broth and LBA supplemented with appropriate antibiotics were used for selection of transformed E. coli strains and transconjugated P. fuscovaginae. Antibiotics (Sigma-Aldrich Co. LLC) were added to media as required at the following final concentrations per millilitre: 100 µg of ampicillin, 20 µg of gentamycin, 50 µg of kanamycin, 150 µg of nitrofurantoin, 25 µg of nalidixic acid. In addition, a final concentration of 40 µg/mL X-gal® and/or 250 µg/mL

S-gal® (Sigma-Aldrich Co. LLC) supplied with Fe3+ ions at a concentration of 500 µg/mL was used in LBA medium to select transformed E. coli.

172

Table 5.1: Bacterial strains used in this study

Strains Relevant characteristics* Reference / Source Escherichia coli DH5α Cloning strain, Nalr (Hanahan, 1985) ICGEB** laboratory stock E. coli C118 Cloning strain ICGEB laboratory stock Pseudomonas Wild-type strain isolated (Cother et al., 2009b) fuscovaginae DAR77795 from diseased rice in WT Australia, Nfr, Ampr P. fuscovaginae Wild-type strain isolated (Cother et al., 2009b) DAR77800 WT from diseased rice in Australia, Nfr, Ampr

*Nalr, Nfr, Ampr indicates nalidixic acid, nitrofurantoin, and ampicillin respectively. **International Centre for Genetic Engineering and Biotechnology, Trieste, Italy.

Table 5.2: Plasmids used and generated in this study

Plasmids Relevant characteristics* Reference / Source pGEM-T Easy Cloning vector, Ampr Promega pGEM+Ad1I pGEM-T Easy vector This study containing internal fragment from Pseudomonas fuscovaginae DAR77795 WT Ad1 gene, Ampr pEX18-Gm Gene replacement vector ICGEB** laboratory stock multiple cloning sites, Gmr, OriT+; conjugation mediated plasmid transfer SacB+: Sucrose tolerance pKNOCK-Km Conjugative suicide (Alexeyev, 1999) vector, Kmr pKNOCK+Ad1I pKNoCK vector This study containing internal fragment from P. fuscovaginae DAR77795 WT Ad1 gene, Kmr

*Kmr, Ampr indicates kanamycin and ampicillin respectively. **International Centre for Genetic Engineering and Biotechnology, Trieste, Italy.

173

5.2.2. Recombinant DNA techniques and sequence analysis

Deletion mutants

Genomic DNA from P. fuscovaginae strains were extracted by Sarkosyl/Pronase lysis

(Better et al., 1983). For plasmid DNA extraction, EuroClone® EuroGold® Xchange® plasmid Miniprep and Midiprep kits were used. Routine procedures for DNA manipulation such as agarose gel electrophoresis, ligations with T4 ligase, digestion with restriction endonucleases and transformation of E. coli were performed as described previously (Sambrook et al., 1989). Primers used were designed using the Web Primer tool of Saccharomyces Genome Database (Stanford University, Stanford, CA94305) and manufactured by Integrated DNA Technologies Inc. The list of primers used in this study are in Table 5.3. PCR amplifications were performed with GoTaq amplification

(Promega) kits. Amplified PCR fragments were cloned on pGEM-T Easy (Promega) and were sequenced using SP6/T7 primers by Macrogen sequencing services (Europe).

Knockout mutants

P. fuscovaginae strains were grown in NB medium (Amyl Media) overnight at 28 °C and bacterial genomic DNA were extracted with the Sigma™ GenElute™ Bacterial Genomic

DNA extraction kit. Plasmid DNA of pGEM were extracted using Wizard® Plus SV

Miniprep DNA purification system and pKNOCK Plasmid DNA was extracted using

Bioline® ISOLATE II plasmid mini kit (protocol for low copy number plasmids).

Routine procedures for DNA manipulation such as agarose gel electrophoresis, ligations with T4 ligase, digestion with restriction endonucleases and transformation of E. coli were performed as described previously (Sambrook et al., 1989). Primers used were designed using the Web Primer tool of Saccharomyces Genome Database (Stanford

University Stanford, CA94305) and manufactured by GeneWorks Pvt. Ltd. (Australia).

174

The list of primers used in this study are in Table 5.3. PCR amplifications were performed with MyTaq (Bioline™) kits. PCR fragments were purified using Promega™ Wizard®

SV gel and PCR clean-up system. Cloning was done with pGEM-T Easy kit (Promega™).

Automated sequencing of PCR fragments was performed by Australian Genome

Research Facility (Brisbane node).

175

Table 5.3: Primers used in this study

Primers Sequence 5’-3’ Restriction Source sites SP6 ATTTAGGTGACACTATAG ICGEB* laboratory stock T7 TAATACGACTCACTATAGGG ICGEB laboratory stock Ad1Up-F AGGATCCTGAACAAGCATCTGT BamHI This study ACCG Ad1Up-R GTCTAGAACGGTTATCAATGTCG XbaI This Study CCAT Ad1Down-F GTCTAGATCTCAGAGGCCAGTG XbaI This study CCAA Ad1Down-R GAAGCTTTGACAAAATGTGGGC HindIII This Study CAAA pKNOCK- CTTAACCGCTGACATGGAA ICGEB New F laboratory stock pKNOCK- TTTATTCGGACACGCGTCCT ICGEB New F laboratory stock M13 F GTAAAACGACGGCCAG Topo cloning kit M13 R CAGGAAACAGCTATGACCATG Topo cloning kit 179Ad1InsF ACAGTCGTTTGTCGATCACCG This study 179Ad1InsR TCGTAGCGATTGGGTGAGC This Study 179Ad1ExtF ACTGCTGAGCAATCGCAAT This study 179Ad1ExtR AACGGTTCCAGTTGTCCTTGA This Study

*International Centre for Genetic Engineering and Biotechnology, Trieste, Italy.

176

5.2.3. Mutagenesis of the Ad1 gene in P. fuscovaginae

Deletion mutants

From Ad1 (7467 bp), a 1002 bp region (2038 bp to 3038 bp) which is 100% homologous in both DAR77795 and DAR77800, was selected as the insert for pKNOCK mutagenic construct. It was checked to be clear of restriction sites that were used in the study. The flanking regions of Ad1Ins were amplified by a two-step PCR using two pairs of oligonucleotides. The upstream fragment Ad1Up was amplified by Ad1UpF and

Ad1UpR yielding a fragment of 1094 bp. Ad1Up fragment has two flanking restriction sites namely BamHI and XbaI, at 3’- and 5’- ends of the fragment, respectively. The downstream fragment Ad1Down of 1176 bp was amplified by the pair of primers

Ad1DownF and Ad1DownR. Ad1Down has two flanking restriction sites, XbaI and

HindIII at the 3’- and 5’- ends of the fragment, respectively. Amplified Ad1Up and

Ad1Down were separately cloned on pGEM and verified by sequencing. They were excised by digestion with the respective restriction enzymes. Ad1Up was first cloned into pEX plasmid digested by BamHI and XbaI restriction enzymes and the resulting construct

Ad1Up+pEX was transformed in E. coli Dh5α. Successful transformations were selected based on gentamycin resistance and colony PCR for Ad1Up.

Plasmid DNA of pEX+ Ad1Up was extracted and purified. Then Ad1Down was cloned into pEX+Ad1Up plasmid construct linearised by digesting with XbaI and HindIII restriction enzymes. The complete pEX construct with Ad1Up and Ad1Down was then transformed into E. coli DH5α. Successful transformation were selected on gentamycin resistance and colony PCR for both Ad1Up and Ad1Down.

Successful transformants were then selected for the first homologous recombination event

(Sucrose sensitive, Gmr) on Gm30 media with and without 5% sucrose. Selected colonies

177

were then screened for second homologous recombination event (Sucr Gms) on 5% sucrose with and without Gm20.

Figure 5.1: pEX 18-Gm plasmid (Suksomtip & Tungpradabkul, 2005). The location of genes and their transcriptional orientation of the pEX18GM gene replacement vector with multiple cloning cites from pUC18; aacC1, represents the gentamicin (Gm) resistance genes; ori, origin of replication; oriT, the origin of transfer; sacB, the levansucrase-encoding gene; lacZ, the reporter gene encoding beta- galactosidase.

Knockout mutants

From Ad1 (7467 bp), a 1002 bp region (2038 bp to 3038 bp) which is 100% homologous in both DAR77795 and DAR77800 was selected as the insert for pKNOCK mutagenic construct. This insert Ad1I was amplified from genomic DNA of DAR77795 by PCR with primers 179Ad1InsF and 179Ad1InsR. Purified Ad1I was ligated to pGEM and transformed in competent cells of E. coli DH5α. Successful transformations were selected based on the disruption of lacZ promoter, ampicillin resistance and colony PCR with 178

179Ad1InsF/R primers. First, pGEM plasmid containing Ad1I was digested with ECoRI, to excise Ad1I fragment. Ad1I was separated by gel electrophoresis and subsequently purified. Because the quality of DNA obtained by the above method was suboptimal,

PCR amplification of a region in Ad1I + pGEM which contains Ad1I was amplified using

M13 F/R primers (Invitrogen). The purified M13 PCR fragment was digested with ECoRI and Ad1I was separated by gel electrophoresis. Purified Ad1I fragment was then ligated to pKNOCK plasmid linearized by digestion with EcoRI and dephosphorylated with

TSAP (Promega). The ligation was transformed into E. coli C118. Successful transformations were selected based on kanamycin resistance, colony PCR with

179Ad1InsF/R primers and restriction digestion of extracted and purified pKNOCK+Ad1I plasmid DNA with EcoRI. The Ad1I fragments resulting from PCR amplifications and restriction digestion were purified and sequenced for further confirmation of ligation.

Tri-parental conjugations between transformed E. coli C118 and P. fuscovaginae wild type (WT) strains were carried with the helper strain E. coli DH5α (pRK2013). Fresh colonies of recipient P. fuscovaginae, donor E. coli C118+ pKNOCK+Ad1I and E. coli

DH5α (pRK2013) were mixed in abundance and cultured on LBA and incubated at 28

°C. Putative P. fuscovaginae Ad1 gene mutants were selected based on the resistance to kanamycin and nitrofurantoin and colony PCR with pKNCOK F/ Ad1Ext R and Ad1Ext

F/ pKNCOK R primers combinations. The resulting PCR amplifications were sequenced for confirmation of mutation.

5.2.4. Preparation of P. fuscovaginae competent cells

As there are no previous reports of preparing competent cells from P. fuscovaginae strains, a protocol was adapted from rapid benchtop and microcentrifuge based methods

179

used to prepare chemically-competent E. coli and P. aeruginosa cells (Chuanchuen,

Narasaki, & Schweizer, 2002) by using CaCl2 /RbCl salts.

P. fuscovaginae strains DAR77795 and DAR77800 were grown in LB broth (NaCl: 5 g,

Tryptone: 10 g, yeast extract: 5 g per litre) at 25 °C overnight until the cultures reached

OD = 1. All the following steps were performed in pre-cooled tubes sitting on ice and centrifugations were done at 4 °C. Ten mL of broth culture was centrifuged in a 15 mL centrifuge tube at 4500 rpm for 10 minutes. The cell pellet was resuspended in 4 mL of ice-cold sterile CaCl2.2H2O, and kept on ice overnight at 4 °C. The cell suspension was centrifuged at 4500 rpm for 10 minutes and the cell pellet was then re-suspended in ice- cold sterile transformation salts and glycerol solution (CaCl2.2H2O 75 mM, RbCl 10 mM,

MOPS 10 mM, glycerol 15% w/v, pH=6.8). The cell suspension was then aliquoted (200

µL) into 1.5 mL microcentrifuge tubes, and immediately stored at -80 °C. The prepared competent cells from the two P. fuscovaginae strains were transformed with both replicating (pGEM-T, egfp) and non-replicating (pKNOCK) plasmids.

The general procedure for transformation of P. fuscovaginae competent cells was as follows. Half of the ligation mix (7.5 µL; approximately 30 – 75 ng of DNA) was mixed with 100 µL of freshly thawed competent cells and incubated for 30 minutes on ice. It was then subjected to heat shock by dipping in a water bath at 42 °C, for 90 seconds, and then quickly moved back to ice. After adding 1 mL of LB broth, it was then incubated at

25 °C for 1 hour with agitation (180 rpm). The transformation mix was then centrifuged at 5000 rpm for 5 minutes to pellet the cells. The supernatant was discarded. To wash away the excess salts and free plasmids, the cell pellet was resuspended in 1 mL of LB broth and was centrifuged again and the supernatant discarded. The cell pellet was again resuspended in 1 mL of LB broth and 50 µL was spread on appropriate LB agar plate for screening and to count cells for transformation efficiency. The rest of the cell suspension

180

was centrifuged and supernatant discarded. The cell pellet was resuspended in 50 µL of

LB and spread plated onto the appropriate LB agar plate for screening (10 times concentrated).

Transformation with pGEM-T easy plasmid

In order to measure the viability and transformability of the prepared competent cells, they were transformed with self- replicating pGEM-T plasmid (~3 kb) which was ligated with the control DNA insert provided in the pGEM-T Easy kit by Promega. Transformed colonies were screened on LB agar containing ampicillin(aq) 100 µg/mL, S-gal(aq) 250

µg/mL and FeCl3(aq) (500 µg/mL).

Transformation with pKNOCK + Ad1 plasmid construct

P. fuscovaginae competent cells were tested for direct transformation with non-self- replicating suicide vector plasmid pKNOCK constructed with Ad1I. Transformation was screened on LBA plates containing kanamycin (aq) 100 µg/mL.

Transformation with fluorescent protein expressing plasmids

To test the transformation efficiency of competent cells of P. fuscovaginae, they were modified to express enhanced green florescent protein (egfp) and fluorescent DsRED protein (rfp) by transforming the prepared P. fuscovaginae competent cells with pMP4655 (green) and pMP4662 (red) fluorescent protein expressing plasmids (illustrated in Figure 5.2), which contain a tetracycline resistance marker, in addition to the fluorescent protein.

181

Figure 5.2: Graphical representation of pMP4655 (egfp) and pMP4662 (rfp) The location of genes and their transcriptional orientation of the rhizosphere-stable reporter plasmids pMP4655 and pMP4662 expressing autofluorescent protein-encoding genes egfp and rfp, respectively, under the control of the lacZ promoter gene; Tc, tetracycline resistance gene; Plac, lacZ promoter gene; egfp, enhanced green fluorescent protein; rfp, DsRED protein. Abbreviations: B = BamHI, Bg = BglII, Sc = SacI, Xb = XbaI (Bloemberg, Wijfjes, Lamers, Stuurman, & Lugtenberg, 2000).

For each, 75 ng plasmid DNA was transformed in 50 µL freshly thawed competent cells.

Successful transformations were selected on LB agar with a final concentration of tetracycline at 10 µg/mL (Tc10). For confirmation, single colonies were observed for florescence under UV at Vilber-Lourmat Vision-Capt Imaging system and viewed with a

Nikon Eclipse TE200 microscope fitted with Intensilight C-HGFI, using florescent viewing under FITC filter for green florescence and TRITC filter for red florescence

(Figure 5.4). Bacterial smear slides prepared from fluorescent colonies were viewed under a Nikon Eclipse Ti-E A1 confocal laser-scanning fluorescence microscope, to confirm fluorescence of individual transformed bacteria cells (Figure 5.5).

182

5.2.5. Sequence analysis and Genome mining

Verification of Ad1Ext F/ pKNOCKR (F2R) and pKNOCKF/ Ad1Ext R (R2F) sequences were conducted by local alignments and sequence editing using Serial Cloner version 1.3-

11 and CLC Genomics workbench version 5.5.1. F2R and R2F sequences were compared and modified to construct a region approximately 1 kb, where the mutation has occurred.

The mutated region from each P. fuscovaginae strain was subjected to homology searches using BLASTX 2.2.31 (Altschul et al., 1990; Zhang et al., 2000) and BLASTN 2.2.32

(Altschul et al., 1997), accessing NCBI/GenBank databases including whole genome shot-gun assembly sequences of P. fuscovaginae DAR77795 and DAR77800 (Stodart et al., 2013).

The nucleotide sequence of Ad1 protein (7467 bp) of DAR77795 was used for genome mining in available P. fucovaginae genomes. Homology searches were conducted using

BLASTN and BLASTX tools in NCBI/GenBank databases, CLC Genomics workbench version 5.5.1 and RAST server, on the draft genome sequences of P. fuscovaginae mentioned in Chapter 3. Briefly, the assembled genomes of strains DAR77795,

DAR77800, ICMP5940, SE-1, CB98818 and UPB0736 available from NCBI/Genbank databases were downloaded from NCBI. The strains UPB0407, UPB0526, UPB0588,

UPB0589 and UPB1013, which were imported from the Université Catholique de

Louvain, Belgium, and were sequenced and assembled by Dr. Ben Stodart at Charles

Sturt University, were uploaded to RAST server and re-annotated. All the 11 genomes were searched with nucleotide sequence-based genome comparison tool BLASTN in

CLC Genomics workbench version 5.5.1, using the Ad1 nucleotide sequence of

DAR77795 (7467 bp) as the query.

Results obtained by the above-mentioned process were assessed and the nucleotide sequences corresponding to Ad1 of DAR77795 were extracted from each genome and

183

were edited to have a single continuous nucleotide sequence file from each genome, denoted as the Ad1 from that particular P. fuscovaginae genome. These Ad1 homologues were then aligned to determine the phylogenetic relationship and maximum likelihood phylogeny tree was generated using CLC Genomics workbench 5.5.1. In addition, all 11 genomes which were uploaded to RAST server were re-annotated and were searched with amino acid sequence-based protein database search using compositional score matrix adjustment (Altshul et al., 2005) known as gapped BLAST and Position-Specific Iterated

BLAST (Psi-BLAST) tools of RAST server, using the Ad1 of DAR77795 as the query.

5.3. Results

5.3.1. Identification of adhesion like functions in the genome of P.

fuscovaginae

In silico analysis of Australian strains DAR77795 and DAR77800 revealed the presence of several adhesin–like genes named Ad1, Ad2, Ad3, Ad4 which showed homology to genes known to have adhesion and hemagglutinin functions. Adhesin1 (Ad1) is present in the WGS of DAR77795 contig: DAR77795_125, (Sequence

ID: dbj|BATD01000125.1|, Range 1: 15866 to 23200). And in WGS of DAR77800 contig: DAR77800_302 (Sequence ID: dbj|BATE01000302.1|) Range 1: 13300 to 15224,

Range 2: 17027 to 18249, Range 3: 15222 to 16218, Range 4: 16218 to 17026, Range 5:

18249 to 18778. Also in DAR77800 contig: DAR77800_412, (Sequence

ID: dbj|BATE01000412.1|) Range 1: 1096 to 2221 and Range 2: 2221 to 2470).

Query with Ad1 (7467 bp) BLASTN in NCBI/GenBank showed homology to adhesins and haemaglutinins of a number of pathogenic Pseudomonas species and other pathogenic bacteria (Table 5.4).

184

Table 5.4: BLASTN results for Ad1 nucleotide sequence from Pseudomonas fuscovaginae DAR77795 in NCBI

Bacteria species Accession Homologue Query Sequence number cover homology % % Pseudomonas WP_040066257.1 adhesin 99 74 batumici P. nitroreducens WP_024763514.1 adhesin 99 54 P. fluorescens WP_046068325.1 adhesin 99 54 P. poae WP_041931575.1 adhesin 90 50 P. parafulva WP_039581480.1 adhesin 64 54 P. libenensis WP_057012657.1 adhesin 99 52 P. brassicacearum WP_039013747.1 adhesin 84 54 P. cannabina WP_046062683.1 adhesin 72 55 P. syringae WP_032606191.1 adhesin 90 50 P. chlororaphis WP_047736554.1 adhesin 84 53 P. trivialis WP_057007264.1 adhesin 90 50 P. putida WP_043861311.1 adhesin 87 52 P. monteilii WP_028699555.1 adhesin 88 52 P. fluorescens gb|AEV64460.1, haemaglutinin 90 50 WP_003238096.1 P. poae WP_015372903.1 haemaglutinin 90 50 Ralstonia WP_043898196.1 haemaglutinin 84 50 solanacearum Enterobacter WP_041162275.1 filamentous 54 50 asburiae hemagglutinin Erwinia amylovora WP_004165239.1 adhesion 84 38 WP_006481050.1 filamentous haemaglutinin Erwinia pyrifoliae WP_012669128.1 adhesin 84 38 Klebsiella gb|ERN61300.1 adhesin 88 38 pneumoniae K. variicola WP_044643675.1 adhesin 88 38 Enterobacter WP_015704285.1 adhesin 88 38 aerogenes K. variicola emb|CEL87303.1 filamentous 88 38 hemagglutinin family outer membrane protein K. pneumoniae emb|CDI25543.1 filamentous 88 38 subsp. pneumoniae hemagglutinin family outer membrane protein 185

In BLASTx search, the 9 to 1715-letter portion of amino acid sequence of this protein shows 99% homology to a partial adhesion protein of P. fuscovaginae

(WP_036988017.1) in NCBI/GenBank protein database. This corresponds to the 2344 -

7464 bp region that is 68% of the nucleotide sequence of Ad1 of DAR77795, which was the query.

In addition, 1582-aa long hypothetical protein of P. fuscovaginae (WP_051356164.1) corresponded to the 1-1535 and 1705-6324 regions of the query while 1260-aa long hypothetical protein of P. fuscovaginae (WP_029530826.1) corresponded to the 1-1260 and 3685-7464 regions of the query. Both displayed 99% homology.

During the search for putative conserved domains with the BLASTX tool in the

NCBI/GenBank, Ad1 of DAR77795 was found to contain multiple conserved domains; two large filamentous hemagglutinins (FhaB), exo-proteins associated with haeme- utilisation, adhesion, intracellular trafficking, secretion, and vesicular transport and a filamentous hemagglutinin repeat, which occurs in number of proteins implicated in cell aggregation (Figure 5.3).

186

Figure 5.3: Putative conserved domains in Ad1 gene of Pseudomonas fuscovaginae strain DAR77795.

In this concise display, RF indicates the open reading frames; +1, +3, and -1. Under each RF, specific hits are shown in bright colours that denote high confidence association between the query region and the conserved domain model thus, indicating high confidence level in the inferred function of the query protein. Under each specific hit, the superfamily to which a specific hit belongs is shown in pastel color, indicating less confidence association of the query region with the conserved domain model. The multi domains shown in grey bars are computationally detected domain models, which could contain multiple single domains. Cyan coloured regions in each RF of the query represent compositionally biased regions that were detected by the low- complexity filter, which alleviates the inaccurate annotation of the query sequence. 187

5.3.2. Mutagenesis in Ad1 gene

Deletion and Knockout mutants

Out of the 450 colonies screened for Sucr Gms putative Ad1 gene mutants of P. fuscovaginae, none of them were found to be successfully transformed. There were no successfully transformed Kmr Ad1 gene knockout mutants of P. fuscovaginae.

5.3.3. P. fuscovaginae competent cells

Competent P. fuscovaginae cells were made, however, with self-replicating plasmid pGem-T Easy, the transformed colonies could not be differentiated from non-transformed colonies. Despite this, P. fuscovaginae competent cells were successfully transformed with non-self-replicating pKNOCK plasmid. This method was also useful for calculating the transformation efficiency of the P. fuscovaginae competent cells. The transformation efficiency of P. fuscovaginae competent cells with non-replicating pKNOCK plasmid was 1 x 104 cfu/µg of DNA. Transformations of P. fuscovaginae competent cells with both the pMP4655 (egfp) and pMP4662 (rfp) fluorescent protein expressing plasmids were successful, yielding both green and red florescent protein expressing P. fuscovaginae bacteria from both the strains (Figure 5.4 and Figure 5.5).

Figure 5.4: egfp-expressing colonies of Pseudomonas fuscovaginae strain DAR77795 WT

188

Figure 5.5: egfp-expressing bacteria cells of Pseudomonas fuscovaginae strain DAR77795 WT

5.3.4. Sequence analysis

The 179Ad1 insert (1002 bp) used to make the pKNOCK mutagenic construct for Ad1 gene, was found to code for intracellular trafficking, secretion, and vesicular transport.

According to BLASTX results (Figure 5.6), a portion of 179Ad1 insert shows homology to multi-domain PRK15319 that is an AIDA autotransporter-like protein ShdA. In addition, a part of 179Ad1 insert shows homology to HrpB4, which is a type III secretion protein always found in type III secretion operons, in a limited number of species including Burkholderia, Xanthmonas and Ralstonia.

189

Figure 5.6: Putative conserved domain hits from NCBI BLASTX of 179Ad1 insert.

In this concise display, RF indicates the open reading frames; +1, +2, +3, -1, -2, -3. The superfamily to which a specific hit belongs is shown in pastel color, indicating less confidence association of the query region with the conserved domain model. The multi domains shown in grey bars are computationally detected domain models, which could contain multiple single domains. Cyan coloured regions in each RF of the query represent compositionally biased regions that were detected by the low-complexity filter, which alleviates the inaccurate annotation of the query sequence.

190

5.3.5. Genome mining

From BLASTN search on P. fuscovaginae draft genomes accessed at NCBI/GenBank database, with the query of Ad1 (100% homologous in both DAR77795 and DAR77800) nucleotide sequence of 7467 bp, the results presented in Table 5.5 were obtained.

Table 5.5: BLASTN results for Ad1 nucleotide sequence from Pseudomonas fuscovaginae DAR77795, in draft genomes of P. fuscovaginae in NCBI/GenBank

P. fuscovaginae Strain taxid Adhesion1 ICMP 5940 1367451 97% query cover, 99% sequence homology SE -1 1316934 Nil Tanii, Miyajima et al.1976 50340 Nil UPB0736 / E124 1150862 Nil CB98818 1176490 99% query cover, 99% sequence homology

The results obtained by BLASTN of Ad1 nucleotide sequence from DAR77795 using

CLC genomics workbench 5.5.1, on newly sequenced, assembled and annotated P. fuscovaginae genomes along with the re-annotated P. fuscovaginae genomes, are summarized in Table 5.6. When the nucleotide sequences, which are homologous to Ad1 of DAR77795 and are presented in Table 5.6, were extracted and edited as individual nucleotide sequences and were aligned, distinctive phylogenetic relationships were revealed (Figure 5.7).

191

Table 5.6: BLASTN results for Ad1 of Pseudomonas fuscovaginae DAR77795 in P. fuscovaginae genomes

Pseudomonas Location in assembled Sequence Hit Hit Hit fuscovaginae genome homology start end frame/ WGS % strand DAR77795 scaffold24size109228 100 53139 60473 minus DAR77800 scaffold23size96127 100 51134 58374 minus UPB0407 No significant similarities. UPB0526 UPB526_contig_4 100 50601 58067 plus UPB0588 UPB588_contig_1 100 24538 32004 plus UPB0589 UPB589_contig_32 100 50601 58067 plus UPB0736 No significant similarities UPB1013 No significant similarities CB98818 sid1014967accnALAQ01 100 22468 27615 minus 0000088 100 2747 5062 plus sid958772accnALAQ010 00197 ICMP5940 Sid2005150accnBATG0 99 20344 27607 plus 1000101 SE-1 No significant similarities

192

Figure 5.7: Molecular phylogenetic analysis of the selected regions extracted from Pseudomonas fuscovaginae whole genome sequences homologous to Ad1 of DAR77795, inferred using the maximum-likelihood method. The tree is drawn to scale with branch lengths measured in number of substitutions per site.

193

In the RAST annotated DAR77795 whole genome, fig|50340.32.peg.3053 (1561 aa long from 193 to 1514 aa) on the plus strand, showed 99% homology to 9 -1335 aa region of the query that was the partial-adhesin protein (1715 aa) of P. fuscovaginae

WP_036988017.1, which corresponds to the 2344 - 7464 bp region of Ad1 of DAR77795.

In addition, a 4640 bp region (54228 - 58852 bp) on minus strand of scaffold24/size109228 of DAR77795 genome that was annotated in RAST was found to have 99% identity to of the 1685 - 6324 bp region of the plus strand of the Ad1 of

DAR77795. Furthermore, the -3 frame of this scaffold24/size109228 of the RAST annotated DAR77795 genome (4686 bp) was found to belong to protein subsystem Two

Partner Secretion pathway (TPS) in to which the above-mentioned CDS feature fig|50340.32.peg.3053 also belongs.

5.4. Discussion

Understanding the factors that regulate host specificity is crucial in controlling emerging and spreadable diseases caused by bacterial plant pathogens (Jacques et al.,

2016) such as sheath brown rot of rice caused by P. fuscovaginae (Bigirimana et al.,

2015). Biomolecules produced by pathogens are essential determinants of host specificity through their involvement in attachment to host surfaces and colonisation in host tissues

(Che et al., 2000). In pathogenic bacteria, the group of proteins collectively known as adhesins facilitate the attachment of bacteria to host surfaces and colonisation, both extra- and intra-cellular. Thus, adhesins have been associated with host specificity (Klemm &

Schembri, 2000; Mhedbi-Hajri et al., 2011).

194

In this study, a large hypothetical protein that shows sequence homology to known large exo-proteins that function as adhesins and hemagglutinins, was functionally analysed.

This protein, Adhesin 1 (Ad1) was found initially in the analysis of whole genomes of P. fuscovaginae strains DAR77795 and DAR77800, with 100% homology between the two strains. In the nucleotide-based homology search in NCBI/GenBank databases, it was found to be absent in the P. fuscovaginae draft genomes of SE-1 and UPB0736, although it was present in ICMP5940 and CB98818 with 99% homology. P. fuscovaginae strains

ICMP5940 and SE-1 are known to have low to moderate virulence (Adorada, 2013), while UPB0736 and CB98818 are known as moderate to highly virulent pathogens on rice (Patel et al., 2014; Xie et al., 2012). It has been reported that DAR77795 and

DAR77800 show different levels of virulence on rice (Adorada et al., 2013a). Since these

P. fuscovaginae strains display variable host-pathogen interactions, it was of importance to investigate the role of this particular protein in the host-pathogen interactions of P. fuscovaginae.

The initial BLASTN search using Ad1 of DAR77795 yielded different results from the whole genome sequences that are supposed to be from the same strain, indicating incompleteness and inaccuracies of whole genomes of P. fuscovaginae available in

NCBI/GenBank databases. For example, the whole genome sequences of ICMP5940 which is the type strain of P. fuscovaginae (Tanii et al., 1976), were recorded twice in

NCBI/GenBank databases as ICMP5940 (taxid:1367451) and Tanii, Miyajima et al.1976

(taxid:50340). Therefore, as mentioned in Chapter 3, all available P. fuscovaginae genomes were uploaded to RAST server and re-annotated. By this process, the accuracy of the results from genome mining for Ad1 and the comparisons of individual Ad1 homologues could be improved, whilst reducing the data misinterpretations as much as possible.

195

Genome mining in the 11 whole genome sequences from different P. fuscovaginae strains, namely DAR77795, DAR77800, UPB0407, UPB0526, UPB0588, UPB0589,

UPB0736, UPB1013, CB98818, ICMP5940 and SE-1, revealed that UPB0407,

UPB0736, UPB1013 and SE-1 do not possess the nucleotide sequence of Ad1, while the rest of the strains contain nucleotide sequences that are 99 - 100% homologous to Ad1.

When these Ad1 homologues were extracted, modified and aligned to show maximum likelihood phylogeny, the single nucleotide differences among these were revealed, indicating the possibility of displaying variable host-pathogen interactions due to the allelic variances in this particular gene.

Allelic variations, particularly, nonsynonymous single-nucleotide polymorphisms

(nsSNPs), in surface-exposed molecules such as adhesins that promote host colonisation, have been reported to be involved in host specificity of zoonotic pathogens (Yue et al., 2015). Allelic variations in multiple adhesins have been found to contribute to bacterial adaptation to certain preferential hosts without losing the capacity to maintain a broad host range (De Masi et al., 2017). Lu et al. (2008) reported that the differentiation of the core genome of Xanthomonads with respect to host- and tissue- specificity, involved subtle changes in a small number of genes or in non-coding sequences, among regulatory targets or secretory substrates. However, the role of single- nucleotide polymorphisms in adhesins in determining the host preferences of plant pathogens has been barely studied. Therefore, it was interesting to study the functions of Ad1 in P. fuscovaginae, which appears to be absent in some strains and show allelic variations among the others.

Ad1 is highly homologous to adhesins and filamentous hemagglutinins of many plant pathogenic Pseudomonas spp., according to the BLASTN results. The BLASTX results indicates that Ad1 is likey to contain conserved domains that are similar to filamentous

196

hemagglutinin adhesin B (FhaB). FhaB is a precursor to filamentous hemagglutinin adhesin (FHA), which conforms to FHA, after removal of an N-terminal signal peptide and a large C-terminal prodomain, during secretion. FHA is assumed as the biologically active form, which is a large, filamentous protein that functions as an essential element for attachment to host surfaces. In the zoonotic pathogen Bordetella bronchiseptica, mutant strains that produced an altered FhaB molecule were eliminated from the lower respiratory tract faster than the wild-type suggesting a defect in resistance, thus indicating that the pre-protein FhaB plays a critical role in the persistence of B. bronchiseptica

(Melvin, Scheller, Noël, & Cotter, 2015). In the citrus canker pathogen Xanthomonas axenopodis pv. citri, FhaB was found to be essential for tissue colonisation, surface attachment and biofilm formation thus, determining virulence (Gottig et al., 2009).

Several methods for gene mutagenesis have been employed to study adhesion function in plant pathogenic bacteria. Examples include site-specific gene disruption by integration of recombinant plasmid through homologous recombination (Das et al., 2009), random gene disruption by transposon-induced mutagenesis (Pradhan, Ranjan, & Chatterjee,

2012) and site-specific gene deletion by marker exchange mutagenesis through double recombination (Gottig et al., 2009). Considering the large size of the target Ad1 gene

(7467 bp) and the requirement for site - specific mutation, gene deletion by marker exchange through double recombination mutagenic technique (Suksomtip &

Tungpradabkul, 2005) and gene disruption by integration of recombinant plasmid through homologous recombination (Alexeyev, 1999) were applied. However, these methods could not generate a mutation in the nucleotide sequence of Ad1 protein. This could be because the chosen region for mutation was essential for pathogen survival. Alternatively, the mutation was created in a region of which the function is not essential for the protein to be synthesised and secreted. The second explanation is more reasonable given that

197

there is more than one mode of transportation for a particular protein. A number of protein secretion mechanisms have been identified in Gram-negative pathogens (Newman &

Stathopoulos, 2004). If the mutation occured in a domain associated with transport, it may have not been effective in disrupting the functions of Ad1, as much as it would be, if the mutation occurred on a domain involved in synthesis.

The Famfig protein subsystem search in RAST server revealed that a part of the Ad1 encodes a two-partner secretion pathway (TPS). In Gram-negative bacteria, the TPS exports large, mostly virulence-related proteins designated as TpsA across the outer membrane through specific transporter proteins designated as TpsB (Jacob-Dubuisson,

Guerin, Baelen, & Clantin, 2013). TpsB transporters belong to the Omp85 superfamily, which is a Type V secretion system involved in protein translocation across cell membranes or integration of proteins to cell membranes (Jacob-Dubuisson et al., 2004; van Ulsen et al., 2014). In the model two-partner secretion system FHA /FhaC of

Bordetella pertussis, the exoprotein FHA contains a large C-terminal prodomain that is removed during translocation (Fan, Fiedler, Jacob-Dubuisson, & Müller, 2012). The

FhaB prodomain is required for the extracellularly located mature C-terminal domain

(MCD) of FHA to achieve its proper conformation, thus FhaB prodomain is essential for the function of FHA in vivo, through its role in maturation of FHA (Noël, Mazar, Melvin,

Sexton, & Cotter, 2012).

Furthermore, the BLASTX analysis of the Ad1 gene shows that the mutated region corresponds to conserved domains for AIDA autotransporter-like protein ShdA. Secretion similarities between autotransporter and two-partner secretion systems have been observed previously (Newman & Stathopoulos, 2004). The authors also reported similarities between autotransporter secretion mechanism with that of intimin and invasins, which are essential for pathogen infection and colonisation of host cells.

198

The functional annotation of Ad1 with NCBI Protein database and the RAST server leads to the conclusion that the target region for mutation was in fact a portion of a large gene cassette, involved in transportation. The mutagenic techniques used for this protein were not successful probably because these are large proteins, thus, a deletion or disruption of a small portion of the larger gene cassette may not necessarily halt the synthesis and the secretion of the protein. The presence of Ad1 in the context of the gene cassette or operon should be assessed to identify its role in adhesion functions. This step might help to select a better target for mutagenesis of Ad1. Instead of selecting a random target for mutation from the large Ad1 gene, conducting further in silico analysis such as test PCR before selecting the target region for mutation, preferably in a synthesis region, would be more effective to study the functions of the Ad1, in future.

Furthermore, this study has the potential to be expanded by utilizing the information generated on the putative adhesins discovered in this study for in silico analyses such as translating nucleotide sequences and analysing the ORFs and homology of resulting protein sequences. The nucleotide sequence information of each Ad1 homologue extracted from each P. fuscovaginae strain and modified in this study could be used to predict the molecular structures of proteins and their morphological and functional variations, if there are any. To test the potential role played by adhesins in the adaptation of Xanthomonads to their host plants, candidate genes that likely play distinct roles in host adaptation were tested in silico by inspecting molecular signatures of selection on candidate genes extracted from sequenced genomes of strains belonging to different pathovars (Mhedbi-Hajri et al., 2011). The authors found strong evidence of adaptive divergence acting on most candidate genes, highlighting a correspondence between pathovar clustering and repertoires of adhesins.

199

Despite the drawbacks, this study reports the successful making of competent cells from

P. fuscovaginae, for the first time, as far as the author is aware. The method of preparing chemically competent cells used in this study seems not to have affected the viability of

P. fuscovaginae cells, because the growth of cultures in LB Tc10 Km50, kept at 28 °C and agitated at 120 rpm, for over 12 hours, was greater than OD600 = 0.5. However, no screening of transformed colonies against non - transformed colonies could be achieved by transforming with pGem -T Easy. The self-replicating pGEM-T Easy plasmid has an ampicillin resistant marker and a lacZ reporter gene which codes for a peptide that couples with a peptide produced by a mutant lacZΔM15 gene in bacteria chromosome, which then produces β-galactosidase that gives a distinct colour (in this case, black) by cleaving substrate S-gal (Sigma) at the presence of Fe3+ ions. P. fuscovaginae is ampicillin resistant and apparently, lacks a mutant lacZΔM15 gene in bacterial chromosome, which may have prevented any differentiation between transformed and non- transformed colonies.

The competent cells of P. fuscovaginae to express green fluorescent protein have many downstream applications. For example, the attachment of P. fuscovaginae cells expressing green fluorecent protein, to plant surfaces, can be visualised with confocal microscopy. It is known that P. fuscovaginae behaves as both an endophyte and epiphyte.

With the use of confocal microscopy, the movement of P. fuscovaginae within plant tissues and interactions with plant cells and other microorganisms in the disease microbiome can be studied. Particularly, it will be useful in studying the mode of pathogenicity. The formation of biofilms is known to be related to pathogenicity and virulence however the biofilm lifecycle of P. fuscovaginae is hitherto unknown.

However, with P. fuscovaginae cells expressing green fluorecent protein, microscopic viewing of planktonic cells and biofilm formation could validate the biofilm cycle of P. fuscovaginae. Samples of bacteria expressing green fluorescent protein taken from

200

biofilms and transferred to glass slides have been visualised by florescence microscopy and confocal laser scanning microscopy (Gottig et al., 2009). In addition, quantification of pathogen loading in plant cells of different plant cell types and various plant species could be conducted by florescence spectrophotometry.

Considering that many secreted proteins are virulence factors (Newman & Stathopoulos,

2004), a better understanding of these secretion mechanisms in P. fuscovginae will support developing better control measures for sheath brown rot. New gene editing techniques such as CRISPR- Cas 9 (Clustered Regularly Interspaced Short Palindromic

Repeats and CRISPR-associated system) and TALEN (Transcription activator-like effector nucleases) (Nemudryi, Valetdinova, Medvedev, & Zakian, 2014; Rath et al.,

2015) have been employed successfully to create target-specific mutagenesis on large exo-proteins like Ad1, in order to study their functions (Nakashima & Miyazaki, 2014;

Shapiro et al., 2017; Sundin et al., 2016).

5.5. Conclusions

A gene encoding a large hypothetical exo-protein, which showed nucleotide sequence homology to a precursor of filamentous hemagglutinin adhesin (FhaB) was identified from the draft genomes of P. fuscovginae strains DAR77795 and DAR77800. Designated as Ad1, this particular gene showed allelic differences among the draft genomes of the 11

P. fuscovaginae stains, DAR77795, DAR77800, UPB0407, UPB0526, UPB0588,

UPB0589, UPB0736, UPB1013, CB98818, ICMP5940 and SE-1 which have previously demonstrated variable host- pathogen interactions. The adhesin FHaB was found to be an essential component of two partner secretory pathway (TPS) that is a branch of type V secretory systems, which is well- characterised in Gram-negative bacteria and known as an essential determinant of pathogenicity and virulence of many pathogens. The

201

attempted mutagenesis of the gene by site-specific mutagenic techniques, using the suicide vector PKNOCK-Km for gene disruption and gene deleting vector pEX18-Gm was not successful, most likely due to the large size of the Ad1 gene. However, during the process, competent cells of P. fuscovaginae was produced for the first time, thus empowering future studies to observe attachment of P. fuscovaginae to plant surfaces, movement within plant tissues, interaction with plant cells and other microorganisms, by transforming the bacterium with florescent protein expressing plasmids. Given the importance of the adhesins in plant pathogenic bacteria and the increasing importance of manipulating adhesins as a method of controlling diseases, it is suggested that the latest gene editing techniques such as CRISPR-Cas 9 and TALEN could be used for mutagenesis and to study the functions of this adhesin in the host- plant interactions of P. fuscovaginae.

202

Chapter 6. General Discussion

In recent times, sheath brown rot has revived the interest of plant pathologists worldwide as an emerging and important disease of rice (Bigirimana et al., 2015). There have been several reports of sheath brown rot disease, particularly from countries in Asia where rice is the staple food and an important commodity (Cother et al., 2010; Cottyn et al., 2002;

Kim et al., 2015; Razak et al., 2009b; Rostami et al., 2010; Xie, 2003). Because of this, studies on the bacterium P. fuscovaginae, the causal organism of sheath brown rot, has gained momentum. There is a heightened interest in exploring the genetic background of

P. fuscovaginae in order to develop new tactics for diagnostics and control of the disease

(Ash et al., 2014; Cother et al., 2009b; Mattiuzzo et al., 2011; Onasanya et al., 2010; Patel et al., 2012; Patel et al., 2014; Quibod et al., 2015b; Uzelac et al., 2017; Xie et al., 2012).

The major purpose of the studies presented in this thesis was to investigate strains of P. fuscovaginae from Australia (Cother et al., 2009b), which are relatively recently isolated compared to the other strains of P. fuscovaginae studied so far.

During the studies conducted by Cother et al. (2009b), on identification and characterisation of the five Australian P. fuscovaginae strains namely, DAR77318,

DAR77320, DAR77795, DAR77797, DAR77800, they have shown unique features in the conserved regions in their genome sequences, compared to the genome sequences of

P. fuscovaginae that were available at the time. The studies conducted on the lineage of the Australian strains of P. fuscovaginae indicate that these are closely related to the strain isolated in 1976 from Japan, ICMP5940, yet have exclusive characteristics of virulence on rice compared to ICMP5940 (Adorada, 2013). Thus, it can be assumed that these have rapidly evolved in the unique agro-climate in southeastern Australia, after introduction.

The main objective of this thesis was to study the host-pathogen interactions of P.

203

fuscovaginae, particularly focusing on the aspects of host preferences of the pathogen and genetic basis of the genetic factors that regulate the host-pathogen interactions.

P. fuscovaginae has been isolated from a broad range of host plants, across various agro- ecological regions around the world. In addition to being recognised as the most important bacterial pathogen contributing to sheath rot disease complex of rice, it is known as a pathogen causing significant damage on cereal crops such as maize (Duveiller et al.,

1989), wheat (Duveiller & Maraite, 1990) and sorghum (Duveiller et al., 1989). It has also been demonstrated to be a potential threat for other cereal crops such as triticale

(Duveiller & Maraite, 1990), barley (Plantwise, 2017b), oat and rye (Malavolta et al.,

1997a). Duveiller and Maraite (1990) showed that a strain of P. fuscovaginae isolated from wheat could cause symptoms of sheath brown rot on rice. Malavolta et al. (1997a) reported that a strain isolated from rice could cause disease symptoms on wheat.

However, prior to the current study there were no studies on the host range or the cross- species pathogenicity for the strains isolated from Australia.

In the second chapter of this thesis, the pathogenicity and virulence of a range of P. fuscovaginae strains were determined. To establish whether the Australian strains, which were originally isolated from rice, could pose a threat to wheat and other cereal crops, pathogenicity assays were conducted on cultivars of rice, wheat, triticale and barley.

Comparisons were drawn with the pathogenicity and virulence profiles of strains originating from Japan (ICMP 5940), Burundi (ICMP 9995) and Madagascar (ICMP

9996), which were isolated from rice, and strain ICMP10137 isolated from wheat in

Brazil. The results indicated that all five of the strains isolated from Australia can adversely affect seedling elongation of wheat, triticale and barley, and cause disease symptoms on wheat at the heading stage. This is the first report on the pathogenicity of

Australian P. fuscovaginae strains on cereal crops other than rice.

204

In this study, plant inoculations were conducted to screen a number of cultivars of rice and wheat, in order to find sources of resistance against P. fuscovaginae. However, no resistant rice cultivar could be identified out of the 10 cultivars, namely, Amaroo,

Doongara, Illabong, Koshihikari, Kyeema, Langi, Opus, Quest, Reiziq, Sherpa. Likewise, all 10 of the wheat cultivars, namely, Chara, Ventura, Mace, Spitfire, EGA Gregory,

Lincoln, Sunvale and Wyalkatchem, were susceptible to the pathogen. In addition, these wheat cultivars and two other wheat cultivars, namely, EGA Wedgetail and Rosella, were susceptible to the pathogen, at the seed germination stage. Similarly, triticale, durum- wheat and barley cultivars, namely, Hawkeye, Tahara, EGA Bellaroi, Tjilkuri, Schooner and Commander, were susceptible at seed germination stage. This study highlights the ability of Australian strains to cause disease on other cereals that are grown in rotation with rice, and the need for developing resistant cultivars. The cereal crop cultivars for this study were selected based on the popularity among growers, and the interests of breeders.

Cultivars that are commonly grown in rice-wheat cropping systems of southern NSW were also included.

The need for on-going screening of rice cultivars grown in Australia has been highlighted in the previous studies as well, considering the highly virulent nature of the Australian P. fucovaginae strains (Adorada et al., 2013a), and the lack of resistant rice cultivars in

Australia (Adorada, 2013). The present study shows specific strain-cultivar interactions between P. fuscovaginae and rice, indicating the presence of resistance genes in cultivars, albeit partially. Hence, the information generated in this study could be used as a preliminary guidance for identifying genes to be used in multi-locus sequence typing aided selection of cultivars and molecular marker assisted breeding of resistant cultivars.

Future studies that will be conducted to find the sources of resistance of rice, wheat,

205

durum-wheat, triticale and barley, against P. fuscovaginae, such as screening of germplasm collections should consider using multiple strains of the pathogen.

In Australia, P. fuscovaginae has been detected exclusively from rice and the symptoms of sheath brown rot disease have not been reported from Australia after 2005 (Adorada,

2013). However, the results of the study in Chapter 2 indicate the potential of the pathogen to infect wheat and other cereal crops. Quibod et al. (2015b) noted that, although P. fuscovaginae strains have been isolated from rice possess a repertoire of genes that may give them specific advantage in colonizing and infecting rice plants, they have the genetic capability to occupy a range of host environments. Thus, it is likely that P. fuscovaginae may remain in the Riverina region, where rice as a summer crop and wheat as a winter crop are grown in rotation (RGA, 2017).

Therefore, as mentioned in Chapter 2 of this thesis, a survey of sheath brown rot disease was conducted in three major rice-growing areas, Murrumbidgee Irrigation Area (MIA),

Coleambally Irrigation Area (CIA) and Murray Valley (MV) (Bajwa & Chauhan, 2017).

Of the 16 rice crops surveyed, three contained plants displaying symptoms typically associated with sheath brown rot, while for wheat, only one of seventeen contained symptomatic plants. However, P. fuscovaginae was not detected from plants showing these symptoms. The analysis of plants collected during similar surveys of the MIA conducted in 2010 and 2012 (Adorada, 2013) also failed to isolate P. fuscovaginae. The author attributed those results to the changes in crop management practices. It is assumed that the rice cv. Amaroo from which the symptoms of sheath brown rot was first observed and the parent line cv. M201, have an inherent susceptibility to P. fuscovaginae (Cother et al., 2009b). Since 2012, cv. Amaroo is no longer cultivated commercially and cv. M201 is being gradually removed from breeding lines (Ovenden, 2016). According to

Lanoiselet et al. (2001), rice crops in Australia remain free of recurring bacterial diseases

206

owing to the hot summer temperatures. In addition, the practice of burning rice stubble

(Lanoiselet et al., 2001) may contribute to the suppression of disease development by preventing P. fuscovaginae from amassing a quorum density that is perhaps required to express pathogenicity and virulence factors, resulting in disease. Still, there is a risk of partial burning of stubble, leaving the root system of infected rice with overwintering bacteria, with the possibility of emerging on the winter cereal crop, especially with the low temperatures that are considered to be favourable to P. fuscovaginae (IRRI, 2017a).

However, some studies dispute the correlation between low temperatures and proliferation of P. fuscovaginae. Jaunet et al. (1996) observed that the temperatures of 13

°C and 18 °C hindered the proliferation of P. fuscovaginae in planta. In addition, Alden

(2016) observed that the optimal temperatures for disease development in germinating rice ranged from 23 °C to 28 °C. The author concluded that the effect of temperature on sheath brown rot was highly dependent on the rice plant, the pathogen and their interaction. However, the response of P. fuscovaginae to variation in weather by expression of virulence characteristics is yet to be investigated. Overall, it can be hypothesised that all these factors have contributed to suppress the P. fuscovaginae population from reaching the quorum that is required to express pathogenicity and virulence factors, thus leading to the absence of sheath brown rot disease symptoms in the fields. Although inoculum densities of 106 to 107 cfu/mL have been generally used in pathogenicity testing of bacteria, as was done in this study, it is unlikely such a population density for a single bacterial strain would occur on plants, naturally (Cottyn, 2003).

In addition, the symptoms of sheath brown rot in rice crops observed during the survey could be due to other pathogens such as Waitea circinata (sheath rot), Rhizoctonia oryzae- sativae (Aggregate sheath rot), P. syrinage pv. syringae (Glume blotch), P. ananatis

(sheath and glume rot), that express similar symptoms and are known to be present in the

207

rice fields in southern NSW (Whitworth et al., 2013). In the present study, the bacteria isolated from the symptomatic rice and wheat plants, included fluorescent and non- fluorescent Pseudomonas spp., Erwinia spp. and fungi such as Alternaria spp., Fusarium spp. and Cladosporium spp.

Furthermore, Bigirimana et al. (2015) suggests that sheath rot is a disease complex, which is a collective expression of indistinguishable symptoms of sheath rot diseases caused by the fungi; Sarocladium oryzae, Fusarium fujikuroi, and the fluorescent Pseudomonads, particularly, P. fuscovaginae. Hence, there is a likelihood for an incident of sheath rot to escalate to economic thresholds due to the cumulative infection by several bacterial and fungal pathogens. Supplementary to this study, the epiphytic bacterial population associated with the disease symptoms on wheat were profiled and it showed the diversity of bacteria of the phyllosphere (Appendix). Profiling the plant disease microbiome is a novel area of interest in phytopathology, particularly because of its importance in planning sustainable strategies for disease control (Busby et al., 2017; Trivedi, Trivedi,

Grinyer, Anderson, & Singh, 2016). Therefore, it would be advantageous to include the other fungal and bacterial pathogens that have a potential to contribute to sheath rot, in disease risk assessments.

The current study demonstrates that the rice cultivars Reiziq, Langi, Kyeema, Doongara,

Sherpa and Opus (all of which are currently in large-scale commercial production (RGA,

2017)) are susceptible to a range of strains of P. fuscovaginae. The cultivars of wheat, durum wheat, triticale and barley that were tested in the study are susceptible to the pathogen, as well. In addition, cross-contamination of successive cereal crops has a great likelihood to occur. Thus, routine monitoring for the symptoms of sheath rot in both rice and wheat crops in southern New South Wales is highly recommended.

208

Contaminated plant material contributes to the persistence of the disease and cause a devastating effect on crop establishment and yield, given the favourable conditions.

Although seed lots are often screened for diseases by the symptoms of seed discoloration, discoloured rice seeds may have a bleached appearance after sometime (Adorada, 2013) and wheat seeds infected with P. fuscovaginae might not show any discoloration as with infected rice seeds. Consequences of undetectable contamination of seed lots could be detrimental for crop establishment (Adorada et al., 2014). Therefore, it is essential that cereal crops are regularly monitored and the seeds are routinely screened preferably with an accurate and rapid diagnostic technique.

Chapter 2 of this thesis discusses the application of BACTID and LOPAT tests, which are conventional diagnostics tools, that employ biochemical characterization of bacterial pathogens (Black, Holt, & Sweetmore, 1996b). Screening of the existing collection of P. fuscovaginae isolates demonstrated that these tests have limited accuracy in identifying the pathogen in a timely manner. In addition, there are several pathogens causing sheath rot diseases with comparable disease symptoms that could not be diagnosed accurately and efficiently with the existing diagnostics. In this study, the accurate and rapid novel diagnostic tool LAMP (Loop-mediated isothermal amplification) for the detection of P. fuscovaginae (Ash et al., 2014) was used in rice and wheat crop fields in NSW, for the first time. The assay did not yield positive results, which could be either due to obstruction from the high salt content in the wash solution which was used to extract the existing epiphytic microbial population from the plant surface, or lack of adequate template P. fuscovaginae DNA in situ because just shaking by hand in SDW was used to extract the epiphytic microbes. However, if the extraction of the microbial population and isolating the DNA of epiphytic of microbiome could be improved, this LAMP protocol has great

209

potential to be the used in field conditions to accurately detect P. fuscovaginae within half an hour.

During the aforementioned plant inoculation studies, significant interactions were observed between the crop cultivars and P. fuscovaginae strains. In addition, discrepancies in the degree of susceptibility of wheat cultivars to P. fuscovaginae between the seed germination and heading stages of plant growth were observed, for the first time.

These observations could be attributed equally to host resistance mechanisms of a particular cultivar and the intrinsic differences in virulence factors of different P. fuscovaginae strains. Phytotoxins produced by P. fuscovaginae are major determinants of virulence and considered as integral components of the plant-pathogen interactions.

Thus, the severity of the damage by phytotoxins is considered to correlate with the degree of susceptibility of the cultivars to the pathogen (Batoko et al., 1991). Adorada et al.

(2013a) noted that there were variations in virulence characteristics among the P. fuscovaginae strains that were used in this study. Malavolta et al. (1997a) reported that the P. fuscovaginae strains isolated from different parts of the same rice plants differed in pathogenicity. Furthermore, Adorada et al. (2013c) suggested that there could be different mechanisms of host plant resistance at the different stages of plant growth. All of these factors may have contributed to the specific strain-cultivar interactions, as well as the degree of cultivar susceptibility at different growth stages of a particular crop cultivar.

The methods of inoculation and inoculum concentrations used to screen cultivars of wheat and other cereals in this study, were based on the recommendations for screening rice cultivars. As the method of inoculation and inoculum density can greatly influence the results of resistance screening of germplasms (Cottyn, 2003), inoculation techniques and

210

inoculum concentrations have to be developed and optimized for identifying sources of resistance in wheat, barley, triticale and other winter cereals.

In order to determine the cross-pathogenicity of P. fuscovaginae further, five strains isolated from wheat (UPB0526, UPB0588, UPB0589, UPB1013) and one strain isolated from Swamp rice grass (UPB0407) were inoculated to seeds of both rice (cv. Amaroo) and wheat (cv. Rosella). All these P. fuscovaginae strains were pathogenic to both the crops, indicating the ability of the pathogen to occupy a range of host plants of the family

Poaceae. The non-host specificity of P. fuscovaginae could be attributed to the non-host specific factors of pathogenicity and virulence such as the phytotoxins produced by this pathogen (Batoko et al., 1997b). The phytotoxic lipodepsipeptides, i.e. syringotoxin and the fuscopeptins, produced by P. fuscovaginae (Flamand et al., 1997), are responsible for the symptoms of sheath brown rot such as stunted seedlings (Batoko et al., 1994), impaired panicle emergence (Batoko et al., 1997b) and necrotic lesions on leaf sheaths, plant stems and leaves.

The differences in virulence levels among strains, dissimilar strain-cultivar interactions and the variations in pathogenicity on different parts of the plant and at different growth stages of the plant, indicate that there is a certain amount of specificity associated with each P. fuscovaginae strain. This could be attributed to virulence factors other than the phytotoxins that were acquired by each strain, giving them a certain advantage in a particular environment to overcome resistance mechanisms of their preferred host.

However, some phytotoxins have been reported to influence the host range of a particular plant pathogen, determining the epiphytic fitness of the pathogen (Arrebola et al., 2009) by acting as surfactants (Bender et al., 1999; Hutchison & Gross, 1997), effectors (Gavin

& Satchell, 2015; Tsuge et al., 2013) and anti-microbials (Cirvilleri, Bonaccorsi, Scuderi,

& Scortichini, 2005).

211

The host preferences of pathogenic bacteria have been reported to be dependent on mobile genetic elements (Jacques et al., 2016; Sheibani-Tezerji et al., 2015). Mobile genetic elements (MGE) are sequences of DNA found in bacteria that appear to have originated from other organisms (Frost, Leplae, Summers, & Toussaint, 2005). Bacteria acquire traits that are required for their survival in a particular environment such as resistance against antimicrobials (von Wintersdorff et al., 2016) by horizontal transfer of MGE such as plasmids, bacteriophages and transposons (von Wintersdorff et al., 2016). Host specificity of bacterial pathogens similar to P. fuscovaginae is reported to be associated with pathogenicity and virulence factors such as the Type III secretory systems, which produce the infection-and motility-related proteins (Sarkar et al., 2006). However, a study of factors associated with the host – pathogen interactions of the strains of P. fuscovaginae have yet to be conducted.

In Chapter 3 of this thesis, a comparative genomics study of P. fuscovaginae strains

DAR77795 (Australia), DAR77800 (Australia), ICMP5940 (Japan), SE-1 (Philippines),

CB98818 (China), UPB0736 (Madagascar) that are isolated from rice and UPB0526

(Mexico), UPB0588 (Mexico) and UPB0589 (Mexico), UPB1013 (Nepal) that are isolated from wheat and UPB0407 (Burundi) isolated from the weed; Swamp rice grass

(Leersia hexandra L.), is discussed. This is the first study that has been conducted using the newly assembled draft genomes of UPB0526, UPB0588, UPB0589, UPB1013 and

UPB0407.

The genomic analyses showed a highly diverse core-genome of P. fuscovaginae, similar to that observed by Quibod et al. (2015b). Some distinct differences between the three groups of rice-infecting, wheat-infecting and weed-infecting strains could be identified.

However, the pathogenicity assays of these P. fuscovaginae strains conducted on rice

(cv. Amaroo) and wheat (cv. Rosella), which are detailed in the Chapter 3, revealed that

212

all the strains used in this study are capable of infecting both rice and wheat, at least at seed germination stage. Several hypothetical proteins, non-ribosomal peptide synthetases, peptide synthetases, long chain fatty acid CoA ligases, and large exo-proteins involved in haem-utilization and adhesion functions were found to be unique to rice-infecting P. fuscovaginae. Weed-infecting strain UPB0407 appeared to have acquired a number of phage-related genes and hypothetical proteins that were not present in other strains. The wheat-infecting P. fuscovaginae group possess a large number of hypothetical proteins, as well. The results agree with the observations of Quibod et al. (2015b) who noted that the genomes of P. fuscovaginae possess diverse metabolic capacities enabling it to adapt to various host environments.

It should be noted that there was a great variability amongst the eleven genomes used in this study, with respect to the basic features such as the size, number of contigs, number of protein coding sequences, RNA coding sequences and the number of subsystems.

These genomes have been sequenced and assembled by several research groups at various times using different techniques. In addition, these strains have been isolated from a number of locations around the world where the environmental conditions vary, over a period of three decades (1976-2005). A genome comparison study of P. fuscovaginae, which included six of the genomes used in this study along with two newly drafted genomes (Quibod et al., 2015b) reported similar discrepancies. Some of these isolates were previously found to vary in virulence on rice (Adorada et al., 2013a; Cother et al.,

2009b). Quibod et al. (2015b) also reported that the two newly drafted P. fuscovaginae- like strains, which were isolated from the same location around the same time, and from the same host, had considerable variations in genome size and number of protein coding sequences. Thus, the high variability among these eleven P. fuscovaginae genomes could be attributed to the intrinsic variances among the strains or the issues arising from

213

sequencing errors. The newly sequenced and assembled genomes used in in this study appeared to be more comparable to each other in size, number of contigs, number of protein and RNA coding sequences, the subsystems and of high quality with less fragmentation to contigs.

As the genome from UPB0736 has been characterised more extensively (Patel et al.,

2012) compared to the others, it was selected as the reference isolate. There was a high level of polymorphism among the strains of P. fuscovaginae, with respect to the reference genome, and within the strains in the same host group. Quibod et al. (2015b) suggested that the genome structures of P. fuscovaginae strains might be more complex than previously thought even without the possibilities of contamination and low-quality reads.

High variability of functionality within each group indicates distinctive metabolic features for each strain although that may conceal the host-specific variability. Quibod et al. (2015b) too, noted that a large majority of the gene clusters were strain-specific.

Out of the 210 genes that were found to be related to host-specificity, most appear to be involved in a diverse range of pathogenicity and virulence functions that are beneficial in adapting to different host environments. These genes were homologous to the elements of horizontal gene transfer and components of secretory modules that are advantageous for epiphytic fitness such as phytotoxins, immunosuppressants, enzyme inhibitors, antibiotics, bacteriocins and sideophores. Particularly, a large number of phage-related genes appeared to be exclusive to UPB0407 that may provide a distinct advantage in occupying the environmental niche of its host, Swamp rice grass. In the same way, a large proportion of prophage-related genes have been found in P. fuscovaginae genomes, which may have facilitated horizontal gene transfer contributing to functional adaptations

214

enabling the pathogen to occupy a broad range of hosts and colonise in various environments. (Quibod et al., 2015b).

Among the genes identified to be involved in host specificity, non-ribosomal peptide synthetases (NRPS) were found to be the highest in number. All these NRPS were found to be components of amino acid adenylation domains/NRPS domains of various syringopeptin synthetases (A, B, C,) of P. syringae sub-species. Syringopeptins are well known phytotoxins (Serra et al., 1999) and P. fuscovaginae produces fuscopeptins that are structural and functional homologues to syringopeptin (Ballio et al., 1996). While both of these phytotoxins play a major role in determining virulence of P. fuscovaginae, neither are known to be host-specific (Batoko et al., 1997b; Iacobellis et al., 1992).

Therefore, it can be hypothesised that these NRPS, being short amino acid sequences, are part of a vast and complex secretory module.

There are a number of genes encoding synthetase modules in the gene cassette which codes for syringomycin and syringopeptin phytotoxins of P. syringae pv. syringae

(Scholz-Schroeder et al., 2003; Wang et al., 2006). This complex consists of many proteins that function as a cascade to govern the expression of these phytotoxins (Bender et al., 1999). The presence or absence of one or more of these synthetase modules might determine the host-specificity, similar to what is observed in P. syringae pv. syringae

(Rezaei & Taghavi, 2014). However, the specific function of these genes in P. fuscovaginae has yet to be validated, along with a large number of genes which appear unique to wheat-infecting strains of P. fuscovaginae that are yet to be characterised.

Therefore, a functional validation of the hypothetical proteins that were identified in this study might elucidate their contribution to in host specificity, pathogenicity and virulence of P. fuscovaginae.

215

In this study, the reference genome was of UPB0736, which is a rice-infecting strain. If there was a well-characterised and complete reference genome of P. fuscovaginae, it could have been advantageous. However, as the first study comparing the genomes of P. fuscovaginae strains isolated from different hosts, the present study provides an initiation to compilation of a complete reference genome for P. fuscovaginae.

In Chapter 4 of this thesis, the results for the functional analysis of a gene encoding a hypothetical NRPS identified within the genomes of P. fuscovaginae was undertaken.

The identified gene was homologous to the syringopeptin synthetase A (sypA) of P. syringae pv. syringae. The amino acid adenylating domain of the gene was targeted in P. fuscovaginae strains DAR77795 and DAR77800, using the pKNOCK suicide vector system (Alexeyev, 1999) that produces a site-specific insertional mutagenesis obstructing transcription of the target gene. Mutant strains of P. fuscovaginae could not produce metabolites that are produced by their respective wildtypes, affecting the virulence of these mutated strains. The non-functional target gene resulted in unaffected shoot lengths of inoculated seeds, and the absence of necrotic lesions on the inoculated seedlings, when compared to their respective wildtypes. It appears to be a non-host-specific toxin, which affects both rice and wheat at seed germination stage, and as a significant determinant of virulence in both DAR77795 and DAR77800. However, the pathogenicity and virulence of the mutants in comparison to wildtypes would need to be verified in mature wheat plants as, in this current research, only mature rice plants were examined. This would assist in fully establishing the role of this protein in determining host-specificity of P. fuscovaginae. Additionally, the complementation of the mutation, thus reinstating the expression of the target gene, which in turn will produce the sypA homologue phytotoxin and hence restoring the virulence, will further confirm the function of this gene.

216

The nucleotide sequence of the sypA homologue gene was found to be conserved in all

11 genomes examined i.e. DAR77795, DAR77800, SE-1, UPB0736, UPB0526,

UPB0588, UPB589, UPB1013, ICMP5940, and CB98818. Therefore, it is highly likely that this gene is involved in virulence-related functions in P. fuscovaginae, encoding a bacterial phytotoxin that is similar to syringopeptin A, which the mutants were incapable of producing. Hence, it can be considered as an integral component of host-pathogen interactions, eventhough it is unlikely to be involved in host-specificity. This concurs with Batoko et al. (1997b) who demonstrated that purified bacterial toxins of P. fuscovaginae were non-host specific. The authors also reported that these toxins are effective at all growth stages of the rice and the susceptibility of rice cultivars to the pathogen, equivalents to the severity of disease symptoms induced by the purified bacterial toxins.

Detry et al. (1991b) reported that the degrees of cultivar susceptibility of rice that were determined based on disease incidence (measured by necrotic lesion like symptoms on leaves and leaf sheaths) at seedling and booting stages, did not correlate with the degrees of cultivar susceptibility of rice that were determined based on the inhibition of panicle emergence in the field. This could be partly attributed to virulence factors other than the phytotoxins that could govern host-pathogen interactions, particularly the epiphytic fitness of the pathogen on various host plants, on different parts of the host, and at each stage of plant growth, that could cause variations in severity of disease. Therefore, the role of the hypothetical proteins such as those analogous to large exo-proteins with adhesion functions, in host-pathogen interactions of P. fuscovaginae, needs to be studied.

In plant pathogenic bacteria, adhesins are a group of proteins that facilitate the attachment and colonisation of bacteria to a plant surface and hence, have been associated with host specificity and virulence (Klemm & Schembri, 2000; Mhedbi-Hajri et al., 2011). In

217

chapter 5 of this thesis, the functional analysis of an adhesin-like gene that was assumed to be involved in host-specificity of P. fuscovaginae, is described. Whole genome analysis of strains DAR77795 and DAR77800 revealed the presence of four large hypothetical proteins, which showed sequence homology to known exo-proteins that functions as adhesins and hemagglutinins. One of these genes named Ad1 was present in only seven of the 11 genomes of P. fucovaginae examined (DAR77795, DAR77800,

UPB0526, UPB0588, UPB0589, CB98818, and ICMP5940). Within these seven genomes, sequence variability was identified consisting of 23 single nucleotide polymorphisms (SNPs) in the 7241 bp long sequence of Ad1. These allelic variances indicated the possibility of displaying distinctive host-pathogen interactions. Previous studies have demonstrated that allelic variations due to SNPs in adhesins could cause significant differences in protein structure, thus determining the host specificity (De Masi et al., 2017; Yue et al., 2015).

Functional annotations of the Ad1 gene revealed that it is highly homologous to fhaB that encodes the prodomain FhaB of the exo-protein filamentous haemaglutinin A (FHA), which is a component of the FHA/FhaC two-partner secretory pathway (TPS), a Type V like secretory system in Gram-negative bacteria (Noël et al., 2012). The FhaB prodomain has been found to be essential for the virulence of Bordetella pertussis through its requirement in FHA maturation, which is necessary for the function of FHA in vivo

(Melvin et al., 2015). As far as the author is aware, the present study is the first report of functional analysis of adhesins of plant pathogenic bacteria in silico.

Thus, Ad1 was functionally analysed by site-specific mutation of the gene using gene deletion by marker exchange through double recombination mutagenic technique described by Suksomtip and Tungpradabkul (2005) and gene disruption by integration of recombinant plasmid through homologous recombination described by (Alexeyev, 1999).

218

However, both of these techniques failed to yield successful mutants out of this gene probably because of the deletion or disruption of a small portion of these large genes might not affect the synthesis and the secretion of the protein. Analysis of the targeted site for mutation in Ad1 gene showed that it is homologous to conserved domains with transporter fucntions. This leads to the conclusion that the target region for mutation could be a portion of a large gene cassette, which is involved in transportation.

However, during the experimental process, the ability to produce competent cells from P. fuscovaginae was demonstrated, thus, facilitating direct transformation of P. fuscovaginae with plasmids in future studies. Particularly, it was demonstrated that the P. fuscovaginae cells could be modified to express green fluorescent protein enabling a number of downstream applications such as the study of the attachment to surfaces as both an endophyte and epiphyte, interactions with the microbiome and quantification of pathogen loading. This is the first report of preparing P. fuscovaginae cells competent.

Considering the importance of adhesins in host-pathogen interactions, particularly to host-specificity and the role of exploiting adhesins in controlling disease, it is suggested that the latest techniques of gene editing such as CRISPR- Cas 9 (Clustered Regularly

Interspaced Short Palindromic Repeats and CRISPR-associated system), and TALEN

(Transcription activator-like effector nucleases) (Nemudryi et al., 2014; Rath et al., 2015), could be used to attempt mutation of the Ad1 target, along with other adhesins, to study their function within P. fuscovaginae.

219

6.2. Directions for future studies

The studies presented in this thesis demonstrate the cross-species pathogenicity of strains of P. fuscovaginae, which is generally known as the causal organism of sheath brown rot of rice, hence, indicate its potential to survive in cereal cropping systems and cause significant yield loss, given the favourable envrionmental conditions. For the first time, unique strain-cultivar interactions between the pathogen and cereal crops is reported and the presence of a large number of genes that could be governing to host- selectivitity of

P. fuscovaginae is revealed. Of these, the functionality of a phytotoxic metabolite and a puatative adhesin was discovered. In addition, several new techniques that would be useful for studying the biology, epidemiology, pathogeniciy and virulence of the pathogen, were applied in the studies discussed in this thesis.

Thus, the information generated in this thesis would be highly beneficial for conducting the following studies, in the future;

- Functional analysis of the genes that were found to be unique to each host group,

through a transcriptomic study, i.e. in planta analysis of gene expression, thus,

validating their involvement in the host- specificity of P. fuscovaginae.

- Profiling the disease microbiome that is associated with sheath rot of rice in

Australia by qualitative analysis and quantification of the microbial community,

thus enumerating the contribution of each of the major pathogens involved in this

disease complex.

- Analyses of secretome of P. fuscovaginae both in vivo and in planta, under

different temperature and humidity conditions, to observe the expression of

pathogenicity and virulence genes, in order to understand the absence of disease

symptoms in the rice-wheat cropping systems in southern NSW.

220

7. References ABC. (2014, 23 May 2014, 5:25pm). Rice crops replace cattle as Northern Territory pastoralist diversifies. Retrieved from http://www.abc.net.au/news/2014-05- 23/rice-seedlings-key-to-future-of-norther-australia-cattle-farms/5474354 Abramovitch, R. B., Anderson, J. C., & Martin, G. B. (2006). Bacterial elicitation and evasion of plant innate immunity. Nature reviews. Molecular cell biology, 7(8), 601-611. doi:10.1038/nrm1984 Adorada, D. L. (2013). Pathogenicity, diversity, biology and sources of resistance to Pseudomonas fuscovaginae in rice. (Doctor of Philosophy), Charles Sturt University. Retrieved from http://researchoutput.csu.edu.au/R/-?func=dbin-jump- full&object_id=49875&local_base=GEN01-CSU01

Adorada, D. L., Stodart, B. J., & Ash, G. J. (2013a). Pseudomonas fuscovaginae in Australian rice: identification, pathogenicity and aggressiveness. Australasian Plant Pathology. Adorada, D. L., Stodart, B. J., Pangga, I. B., & Ash, G. J. (2014). Implications of bacterial contaminated seed lots and endophytic colonization by Pseudomonas fuscovaginae on rice establishment. Plant Pathology, 64, 43-50. doi:10.1111/ppa.12243 Adorada, D. L., Stodart, B. J., Tpoi, R. P., Costa, S. S., & Ash, G. J. (2013b). Bacteria associated with sheath browning and grain discoloration of rice in East Timor and implications for Australia’s biosecurity. Australasian Plant Disease Notes 1-5. doi:10.1007/s13314-013-0092-z Adorada, D. L., Stodart, B. J., Vera Cruz, C., Gregorio, G., Pangga, I., & Ash, G. J. (2013c). Standardizing resistance screening to Pseudomonas fuscovaginae and evaluation of rice germplasm at seedling and adult plant growth stages. Euphytica, 192(1), 1-16. doi:10.1007/s10681-012-0804-z Ailloud, F., Lowe, T., Cellier, G., Roche, D., Allen, C., & Prior, P. (2015). Comparative genomic analysis of Ralstonia solanacearum reveals candidate genes for host specificity. BMC Genomics, 16(1), 270. Retrieved from http://www.biomedcentral.com/1471-2164/16/270

221

Al-Babili, S., Hoa, T. T. C., & Schaub, P. (2006). Exploring the potential of the bacterial carotene desaturase CrtI to increase the β-carotene content in Golden Rice. Journal of Experimental Botany, 57(4), 1007-1014. doi:10.1093/jxb/erj086 Alden, R. (2016). The effect of temperature on rice infected with Pseudomonas fuscovaginae. Honours Thesis. Faculty of Science/ School of Agricultural and Wine Sciences. Charles Sturt University. Alexeyev, M. F. (1999). The pKNOCK series of broad-host-range mobilizable suicide vectors for gene knockout and targeted DNA insertion into the chromosome of gram-negative bacteria. Biotechniques, 26(5), 824-826, 828. Alippi, A. M., & López, A. C. (2010). First report of leaf spot disease of maize caused by Pantoea ananatis in Argentina. Plant Disease, 94(4), 487-487. doi:10.1094/pdis- 94-4-0487a Altschul, S. F., Gish, W., Miller, W., Myers, E. W., & Lipman, D. J. (1990). Basic local alignment search tool. Journal of Molecular Biology, 215(3), 403-410. doi:10.1016/s0022-2836(05)80360-2 Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W., & Lipman, D. J. (1997). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Research, 25, 3389-3402. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC146917/pdf/253389.pdf Anzai, Y., Kim, H., Park, J. Y., Wakabayashi, H., & Oyaizu, H. (2000). Phylogenetic affiliation of the Pseudomonads based on 16S rRNA sequence. International Journal of Systematic and Evolutionary Microbiology, 50(4), 1563-1589. Retrieved from http://ijs.sgmjournals.org/cgi/content/abstract/50/4/1563 Aremu, B. R., & Babalola, O. O. (2015). Classification and taxonomy of vegetable macergens. Frontiers in Microbiology, 6(1361). doi:10.3389/fmicb.2015.01361 Arrebola, E., Cazorla, F. M., Codina, J. C., Gutierrez-Barranquero, J. A., Perez-Garcia, A., & de Vicente, A. (2009). Contribution of mangotoxin to the virulence and epiphytic fitness of Pseudomonas syringae pv. syringae. International Microbiology 12(2), 87-95. Arsenijevic, M. (1991). Bacterial sheath brown rot of rice, wheat, maize and sorghum plants. [Contemporary Agriculture]. Savremena Poljoprivreda 39(1), 66-71. Arunakumari, K., Durgarani, C. V., Satturu, V., Sarikonda, K. R., Chittoor, P. D. R., Vutukuri, B., Laha, G. S., Nelli, A. P. K., Gattu, S., Jamal, M., Prasadbabu, A.,

222

Hajira, S., & Sundaram, R. M. (2016). Marker-assisted pyramiding of genes conferring resistance against bacterial blight and blast diseases into Indian rice variety MTU1010. Rice Science, 23(6), 306-316. doi:https://doi.org/10.1016/j.rsci.2016.04.005 Ash, G. J., Chung, Y. R., McKenzie, C., & Cother, E. J. (2008). A phylogenetic and pathogenic comparison of potential biocontrol agents for weeds in the family Alismataceae from Australia and Korea. Australasian Plant Pathology, 37(4), 402- 405. doi:10.1071/ap08023 Ash, G. J., Lang, J. M., Triplett, L. R., Stodart, B. J., Verdier, V., Cruz, C. V., Rott, P., & Leach, J. E. (2014). Development of a genomics-based LAMP (loop-mediated isothermal amplification) assay for detection of Pseudomonas fuscovaginae from rice. Plant Disease, 98(7), 909-915. doi:10.1094/PDIS-09-13-0957-RE Ashkani, S., Rafii, M. Y., Shabanimofrad, M., Miah, G., Sahebi, M., Azizi, P., Tanweer, F. A., Akhtar, M. S., & Nasehi, A. (2015). Molecular breeding strategy and challenges towards improvement of blast disease resistance in rice crop. Frontiers in Plant Science, 6(886). doi:10.3389/fpls.2015.00886 Ayyadurai, N., Kirubakaran, S. I., Srisha, S., & Sakthivel, N. (2005). Biological and molecular variability of Sarocladium oryzae, the sheath rot pathogen of rice (Oryza sativa L.). Curr Microbiol, 50(6), 319-323. doi:10.1007/s00284-005-4509-6 Azad, H. R., Holmes, G. J., & Cooksey, D. A. (2000). A new leaf blotch disease of Sudangrass caused by Pantoea ananas and Pantoea stewartii. Plant Disease, 84(9), 973-979. doi:10.1094/pdis.2000.84.9.973 Azegami, K. (2013). Suppressive effect of bacteriophage on bacterial palea browning of rice caused by Pantoea ananatis. Journal of General Plant Pathology, 79(2), 145- 154. doi:10.1007/s10327-013-0435-9 Aziz, R. K., Bartels, D., Best, A. A., DeJongh, M., Disz, T., Edwards, R. A., Formsma, K., Gerdes, S., Glass, E. M., Kubal, M., Meyer, F., Olsen, G. J., Olson, R., Osterman, A. L., Overbeek, R. A., McNeil, L. K., Paarmann, D., Paczian, T., Parrello, B., Pusch, G. D., Reich, C., Stevens, R., Vassieva, O., Vonstein, V., Wilke, A., & Zagnitko, O. (2008). The RAST Server: rapid annotations using subsystems technology. BMC Genomics, 9, 75. doi:10.1186/1471-2164-9-75

223

Bajwa, A. A., & Chauhan, B. S. (2017). Rice production in Australia. In Chauhan, B. S., Jabran, K., & Mahajan, G. (Eds.), Rice Production Worldwide (pp. 169-184): Springer International Publishing. Baker, K. F., & Smith, S. H. (1966). Dynamics of seed transmission of plant pathogens. Annual Review of Phytopathology, 4(1), 311-332. doi:10.1146/annurev.py.04.090166.001523 Ballio, A., Bossa, F., Camoni, L., di Giorgio, D., Flamand, M. C., Maraite, H., Nitti, G., Pucci, P., & Scaloni, A. (1996). Structure of fuscopeptins, phytotoxic metabolites of Pseudomonas fuscovaginae. FEBS Letters, 381(3), 213-216. Retrieved from http://www.sciencedirect.com/science/article/B6T36-3VXHMSH- 17/2/e99542f68d300b881326d8aaf04ab6f1 Bare, S., Coiro, V. M., Scaloni, A., di Nola, A., Paci, M., Segre, A. L., & Ballio, A. (1999). Conformations in solution of the fuscopeptins - phytotoxic metabolites of Pseudomonas fuscovaginae. European Journal of Biochemistry, 266(2), 484-492. Retrieved from http://onlinelibrary.wiley.com/store/10.1046/j.1432- 1327.1999.00883.x/asset/j.1432- 1327.1999.00883.x.pdf?v=1&t=i7y5o1zi&s=fbc6201901378441d26d38ed6ffb9a6 d0411be8e Bastas, K. K., & Kannan, V. R. (2015). Pathogenesis of plant pathogenic bacteria Sustainable approaches to controlling plant pathogenic bacteria (pp. 17-48): CRC Press. Batoko, H., Bouharmont, J., Kinet, J. M., & Maraite, H. (1997a). Inhibition of rice (Oryza sativa L.) internode elongation by Pseudomonas fuscovaginae toxins. In Rudolph, K., Burr, T. J., Mansfield, J. W., Stead, D., Vivian, A., & Kietzell, J. (Eds.), Pseudomonas syringae pathovars and related pathogens (Vol. 9, pp. 227-229): Springer Netherlands. Batoko, H., Bouharmont, J., Kinet, J. M., & Maraite, H. (1997b). Involvement of toxins produced by Pseudomonas fuscovaginae in aetiology of rice bacterial sheath brown rot. Journal of Phytopathology, 145(11-12), 525-531. Retrieved from ://000071797600011 Batoko, H., Bouharmont, J., & Maraite, H. (1994). Inhibition of rice (Oryza sativa L.) seedling elongation by a Pseudomonas fuscovaginae toxin. Euphytica, 76(1-2), 139-143. doi:10.1007/bf00024032

224

Batoko, H., Bouharmont, J., & Maraite, H. (1997c). Biological effects of toxins produced by Pseudomonas fuscovaginae on rice and in vitro selection for resistance. In Poisson, C. & Rakotoarisoa, J. (Eds.), Conference Rice for Highlands (pp. 113- 118). 75116 Paris: Centre Cooperation Int Rech Agronomique Developpement. Batoko, H., d'Exaerde, A. D., Kinet, J. M., Bouharmont, J., Gage, R. A., Maraite, H., & Boutry, M. (1998). Modulation of plant plasma membrane H+-ATPase by phytotoxic lipodepsipeptides produced by the plant pathogen Pseudomonas fuscovaginae. Biochimica et Biophysica Acta-Biomembranes 1372(2), 216-226. Retrieved from ://000075343800007 Batoko, H., Flamand, M.-C., Boutry, M., Kinet, J.-M., & Maraite, H. (1997d). Biological effects of Pseudomonas fuscovaginae toxins on rice cells. In Rudolph, K., Burr, T. J., Mansfield, J. W., Stead, D., Vivian, A., & Kietzell, J. (Eds.), Pseudomonas syringae pathovars and related pathogens (Vol. 9, pp. 215-220): Springer Netherlands. Batoko, H., Kinet, J. M., Maraite, H., & Bouharmont, J. (1991). Susceptibility to brown sheath rot of rice correlated to sensitivity to bacterial toxin. Parasitica, 47, 151-163. Retrieved from http://blogfile.paran.com/BLOG_862327/200808/1223387728_Susceptibility%20 to%20brown%20sheath%20rot%20of%20rice%20correlated%20to%20sensitivity %20to%20bacterial%20toxin%20(Rice%20Genetics%20Newsletter%20Volume %2011).pdf Bender, C. L., Alarcón-Chaidez, F., & Gross, D. C. (1999). Pseudomonas syringae phytotoxins: mode of action, regulation, and biosynthesis by peptide and polyketide synthetases. Microbiology and Molecular Biology Reviews, 63(2), 266-292. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC98966/pdf/mr000266.pdf Benson, D. A., Karsch-Mizrachi, I., Clark, K., Lipman, D. J., Ostell, J., & Sayers, E. W. (2012). GenBank. Nucleic Acids Research 40 (Database issue), D48-53. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/22144687 Bergey, D. H., Krieg, N. R., & Holt, J. G. (1984). Bergey's manual of systematic bacteriology. Baltimore, MD: Williams & Wilkins.

225

Better, M., Lewis, B., Corbin, D., Ditta, G., & Helinski, D. R. (1983). Structural relationships among Rhizobium meliloti symbiotic promoters. Cell, 35(2 Pt 1), 479- 485. Bigirimana, V. d. P., Hua, K. H. G., Nyamangyoku, O. I., & Höfte, M. (2015). Rice sheath rot: an emerging ubiquitous destructive disease complex. Frontiers in Plant Science 6. doi:10.3389/fpls.2015.01066 Biosecurity Queensland. (2016a). Fact sheet (Lolium perenne L.) Environmental Weeds of Australia Identic Pty Ltd. Biosecurity Queensland. (2016b). Fact sheet (Phleum pratense L.) Environmental Weeds of Australia Identic Pty Ltd. Black, R., Holt, J., & Sweetmore, A. (1996a). BACTID : bacteriological identification for resource-poor plant pathology laboratories : incorporating BACTID kit and expert system. Chatham: Natural Resources Institute. Black, R., Holt, J., & Sweetmore, A. (1996b). BACTID bacteriological identification for resource-poor plant pathology laboratories: incorporating BACTID kit and expert system (Version 5.0 ed.). Chatham, UK: Practical Action Publishing. Bloemberg, G. V., Wijfjes, A. H., Lamers, G. E., Stuurman, N., & Lugtenberg, B. J. (2000). Simultaneous imaging of Pseudomonas fluorescens WCS365 populations expressing three different autofluorescent proteins in the rhizosphere: new perspectives for studying microbial communities. Molecular plant-microbe interactions 13(11), 1170-1176. doi:10.1094/mpmi.2000.13.11.1170 Bodenhausen, N., Horton, M. W., & Bergelson, J. (2013). Bacterial communities associated with the leaves and the roots of Arabidopsis thaliana. PLoS ONE, 8(2), e56329. doi:10.1371/journal.pone.0056329 Boerema, G. H. (2004). Phoma identification manual: differentiation of specific and infra-specific taxa in culture: CABI. Bomfeti, C. A., Meirelles, W. F., Souza-Paccola, E. A., Casela, C. R., Ferreira, A. d. S., Marriel, I. E., & Paccola-Meirelles, L. D. (2007). Avaliação de produtos químicos comerciais, in vitro e in vivo, no controle da doença foliar, mancha branca do milho, causada por Pantoea ananatis. Summa Phytopathologica, 33, 63-67. Retrieved from http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0100- 54052007000100009&nrm=iso

226

Bouman, B. A. M., Humphreys, E., Tuong, T. P., & Barker, R. (2007). Rice and water. Advances in agronomy, 92, 187-237. Brennan, J. P., & Malabayabas, A. (2011). International Rice Research Institute's contribution to rice varietal yield improvement in South-East Asia: Australian Centre for International Agricultural Research, Canberra. Bua, B., Adipala, E., & Opio, F. (1998). Screening cowpea germplasm for resistance to bacterial blight in Uganda. International Journal of Pest Management, 44(3), 185- 189. doi:10.1080/096708798228293 Bultreys, A., Gheysen, I., Wathelet, B., Maraite, H., & de Hoffmann, E. (2003). High- performance liquid chromatography analyses of pyoverdin siderophores differentiate among phytopathogenic fluorescent Pseudomonas species. Applied and Environmental Microbiology, 69(2), 1143-1153. doi:10.1128/aem.69.2.1143- 1153.2003 Busby, P. E., Soman, C., Wagner, M. R., Friesen, M. L., Kremer, J., Bennett, A., Morsy, M., Eisen, J. A., Leach, J. E., & Dangl, J. L. (2017). Research priorities for harnessing plant microbiomes in sustainable agriculture. PLoS Biology, 15(3), e2001793. doi:10.1371/journal.pbio.2001793 Büttner, D., & He, S. Y. (2009). Type III protein secretion in plant pathogenic bacteria. Plant Physiology, 150(4), 1656-1664. doi:10.1104/pp.109.139089 CABI. (2017a). Acidovorax avenae subsp. avenae (bacterial leaf blight). Invasive Species Compendium: CABI. CABI. (2017b, 11 October 2017). Pseudomonas fuscovaginae (sheath brown rot). Invasive Species Compendium. Retrieved from https://www.cabi.org/isc/datasheet/44957 CABI. (2017c). Xanthomonas oryzae pv. oryzicola (bacterial leaf streak of rice). Cahyaniati, C. N., & Mortensen, C. N. (1995). Bacterial sheath brown rot of rice (Pseudomonas fuscovaginae) grown in Indonesia. Retrieved from Copenhagen, Denmark: Camilli, A., & Bassler, B. L. (2006). Bacterial small-molecule signaling pathways. Science, 311(5764), 1113-1116. doi:10.1126/science.1121357 Cao, H., Baldini, R. L., & Rahme, L. G. (2001). Common mechanisms for pathogens of plants and animals. Annual Review of Phytopathology, 39(1), 259. Retrieved from

227

http://ezproxy.csu.edu.au/login?url=http://search.ebscohost.com/login.aspx?direct =true&db=a9h&AN=5367177&site=ehost-live Carr, E. A., Bonasera, J. M., Zaid, A. M., Lorbeer, J. W., & Beer, S. V. (2010). First report of bulb disease of onion caused by Pantoea ananatis in New York. Plant Disease, 94(7), 916-916. doi:10.1094/pdis-94-7-0916b Challagulla, V., Bhattarai, S., & Midmore, D. J. (2015). In-vitro Vs in-vivo inoculation: Screening for resistance of Australian rice genotypes against blast fungus. Rice Science, 22(3), 132-137. doi:https://doi.org/10.1016/j.rsci.2015.05.017 Chanprapai, P., & Chavasiri, W. (2017). Antimicrobial activity from Piper sarmentosum Roxb. against rice pathogenic bacteria and fungi. Journal of Integrative Agriculture, 16(11), 2513-2524. doi:https://doi.org/10.1016/S2095- 3119(17)61693-9 Chau, L. M., Cat, H. D., Ben, P. T., Phuong, L. T., Cheng, J., & Heong, K. L. (2003). Impacts of nutrition management on insect pests and diseases of rice. Omonrice, 11(93-102), 93-102. Retrieved from http://www.clrri.org/lib/omonrice/11-11.pdf Chaudhary, R. N. (1996). Techniques for screening rice genotypes for resistance to the blast fungus and it's crude extract toxin. Che, F.-S., Iwano, M., Tanaka, N., Takayama, S., Minami, E., Shibuya, N., Kadota, I., & Isogai, A. (1999). Biochemical and morphological features of rice cell death induced by Pseudomonas avenae. Plant and Cell Physiology, 40(10), 1036-1045. Retrieved from http://pcp.oxfordjournals.org/content/40/10/1036.abstract Che, F.-S., Nakajima, Y., Tanaka, N., Iwano, M., Yoshida, T., Takayama, S., Kadota, I., & Isogai, A. (2000). Flagellin from an incompatible strain of Pseudomonas avenae induces a resistance response in cultured rice cells. Journal of Biological Chemistry, 275(41), 32347-32356. doi:10.1074/jbc.M004796200 Chen, R., Barphagha, I. K., Karki, H. S., & Ham, J. H. (2012). Dissection of quorum- sensing genes in Burkholderia glumae reveals non-canonical regulation and the new regulatory gene tofM for toxoflavin production. PLoS ONE, 7(12), e52150. doi:10.1371/journal.pone.0052150 Chiba, S., Chiba, J., Shimada, K., Ota, K., & Kuwata, H. (1977). The outbreak of bacterial sheath brown rot of rice plant in Aomori prefecture and its effect on yield and quality of rice [Pseudomonas fuscovaginae]. Retrieved from

228

Chien, C. C., & Chang, Y. C. (1987). The susceptibility of rice plants at different growth stages and 21 commercial rice varieties to Pseudomonas glumae. Journal of Agriculture Research China, 36, 302 - 310. Chin-A-Woeng, T. F. C., van den Broek, D., Lugtenberg, B. J. J., & Bloemberg, G. V. (2005). The Pseudomonas chlororaphis PCL1391 sigma regulator psrA represses the production of the antifungal metabolite phenazine-1-carboxamide. Molecular Plant-Microbe Interactions, 18(3), 244-253. doi:10.1094/mpmi-18-0244 Chinara, N., Mishra, B. D., & Dhal, N. K. (2007). Effect of bacterial seed infection on rice seed germination. Journal of Mycopathological Research, 45(2), 269-271. Retrieved from http://www.cabdirect.org/abstracts/20083204277.html;jsessionid=9855F592406A F4A18625074F01F2F475# Cho, H.-S., Park, S.-Y., Ryu, C.-M., Kim, J. F., Kim, J.-G., & Park, S.-H. (2007). Interference of quorum sensing and virulence of the rice pathogen Burkholderia glumae by an engineered endophytic bacterium. FEMS Microbiology Ecology, 60(1), 14-23. doi:10.1111/j.1574-6941.2007.00280.x Christopher, M., Cordeiro, G., Waters, D., & Henry, R. (2004). Marker-assisted selection in rice improvement (04/011). Retrieved from Kingston ACT: Chuanchuen, R., Narasaki, C. T., & Schweizer, H. P. (2002). Benchtop and microcentrifuge preparation of Pseudomonas aeruginosa competent cells. Biotechniques, 33(4), 760-763. Cirvilleri, G., Bonaccorsi, A., Scuderi, G., & Scortichini, M. (2005). Potential biological control activity and genetic diversity of Pseudomonas syringae pv. syringae strains. Journal of Phytopathology, 153(11-12), 654-666. Retrieved from ://000233148700004 Claflin, L. E., Ramundo, B. A., Leach, J. E., & Erinle, I. D. (1989). Pseudomonas avenae, causal agent of bacterial leaf stripe of pearl millet. Plant Disease, 73, 1010-1014. doi:10.1094/PD-73-1010 Clarke, M. (2016). Rice program five year RD&E plan 2016/17 to 2021/22: Rural Industries Research and Development Corporation. Cockfield, G., Mushtaq, S., & White, N. (2012). Relocation of intensive agriculture to northern Australia: the case of the rice industry. Retrieved from http://era.daf.qld.gov.au/id/eprint/3603/1/Output2RiceCaseStudy.pdf

229

Collard, B., Vera Cruz, C. M., McNally, K. L., Virk, P. S., & Mackill, D. J. (2008). Rice molecular breeding laboratories in the genomics era: current status and future considerations. International Journal of Plant Genomics, 2008, 25. doi:10.1155/2008/524847 Collmer, A. (1998). Determinants of pathogenicity and avirulence in plant pathogenic bacteria. Current Opinion in Plant Biology, 1(4), 329-335. Retrieved from http://www.sciencedirect.com/science/article/B6VS4-49WMMNN- 20/2/90c147c1b93817341438f54517203820 Coraiola, M., Paletti, R., Flore, A., Fogliano, V., & Serra, M. D. (2008). Fuscopeptins, antimicrobial lipodepsipeptides from Pseudomonas fuscovaginae, are channel forming peptides active on biological and model membranes. Journal of Peptide Science, 14(4), 496-502. doi:10.1002/psc.970 Cortesi, P., Bartoli, F., Pizzatti, C., Bertocchi, D., & Schaad, N. W. (2008). Panicle sterility and grain discolouration: new and emerging bacterial diseases of rice in Italy. In Fatmi, M. C., A.; Iacobellis, N.S.; Mansfield, J.; Murillo, J.; Schaad, N.W.; Ullrich, M. (Ed.), Pseudomonas syringae pathovars and related pathogens - identification, epidemiology and genomics (pp. 391-411): Springer Netherlands. Cota, L. V., Costa, R. V., Silva, D. D., Parreira, D. F., Lana, U. G. P., & Casela, C. R. (2010). First report of pathogenicity of Pantoea ananatis in sorghum (Sorghum bicolor) in Brazil. Australasian Plant Disease Notes, 5(1), 120-122. doi:10.1007/bf03215904 Cother, E. J. (1974). Bacterial glume blotch of rice. Plant disease reporter, 63(2), 131- 135. Retrieved from http://aims.fao.org/serials/c_8f53e167 Cother, E. J., Noble, D. H., Van de Ven, R. J., Lanoiselet, V., Ash, G. J., Vuthy, N., Visarto, P., & Stodart, B. S. (2010). Bacterial pathogens of rice in the Kingdom of Cambodia and description of a new pathogen causing a serious sheath rot disease. Plant Pathology, 59(5), 944-953. doi:10.1111/j.1365-3059.2010.02310.x Cother, E. J., Reinke, R., McKenzie, C., Lanoiselet, V. M., & Noble, D. H. (2004). An unusual stem necrosis of rice caused by Pantoea ananas and the first record of this pathogen on rice in Australia. Australasian Plant Pathology, 33, 495-503. doi:10.1071/AP04053

230

Cother, E. J., Stodart, B., & Ash, G. (2009a). Improving understanding and management of rice pathogens in Cambodia project - final report (FR2009-46). Retrieved from Canberra ACT 2601 Australia: Cother, E. J., Stodart, B. S., Noble, D. H., Reinke, R., & van de Ven, R. J. (2009b). Polyphasic identification of Pseudomonas fuscovaginae causing sheath and glume lesions on rice in Australia. Australasian Plant Pathology, 38(3), 247-261. doi:10.1071/ap08103 Cottyn, B., Cerez, M. T., van Outryve, M. F., Barroga, J., Swings, J., & Mew, T. W. (1996a). Bacterial diseases of rice I. Pathogenic bacteria associated with sheath rot complex and grain discoloration of rice in the Philippines. Plant Disease, 80(4), 429-437. Cottyn, B., Certz, M. T., & Mew, T. W. (1994a). Bacteria. In Mew, T. W. & Misra, J. K. (Eds.), A Manual of rice seed health testing (pp. 29-46). Manila, Philippines: International Rice Research Institute. Cottyn, B., Certz, M. T., & Mew, T. W. (1994b). Bacterial pathogens In Mew, T. W. & Misra, J. K. (Eds.), A manual of rice seed health testing (pp. 91-96). Manila, Philippines: International Rice Research Institute. Cottyn, B., Debode, J., Regalado, E., Mew, T. W., & Swings, J. (2009). Phenotypic and genetic diversity of rice seed-associated bacteria and their role in pathogenicity and biological control. Journal of Applied Microbiology, 107(3), 885-897. doi:10.1111/j.1365-2672.2009.04268.x Cottyn, B., George, T., & Vera Cruz, C. M. (2002). Characterization of rice sheath rot from Siniloan, Philippines. International Rice Research Newsletter, 27(1), 39-40. Retrieved from http://www.irri.org/publications/irrn/pdfs/vol27no1/irrn27- 1pestsci.pdf Cottyn, B., Regalado, E., Lanoot, B., de Cleene, M., Mew, T. W., & Swings, J. (2001). Bacterial populations associated with rice seed in the tropical environment. Phytopathology, 91(3), 282-292. doi:10.1094/phyto.2001.91.3.282 Cottyn, B., van Outryve, M. F., Cerez, M. T., de Cleene, M., Swings, J., & Mew, T. W. (1996b). Bacterial diseases of rice II. Characterization of pathogenic bacteria associated with sheath rot complex and grain discoloration of rice in the Philippines. Plant Disease, 80(4), 429-437.

231

Cottyn, B. J. G. (2003). Bacteria associated with rice seed from Philippine farmers' fields. (PhD), Universiteit Gent (Belgium), Belgium. Retrieved from http://proquest.umi.com/pqdweb?did=764701491&Fmt=7&clientId=20870&RQT =309&VName=PQD Coutinho, T. A., & Venter, S. N. (2009). Pantoea ananatis: an unconventional plant pathogen. Molecular Plant Pathology, 10(3), 325-335. doi:10.1111/j.1364- 3703.2009.00542.x D'aes, J. (2012). Biological activity and regulation of cyclic lipopeptides and phenazines produced by biocontrol strain Pseudomonas CMR12a. Ghent University, Belgium. Retrieved from http://hdl.handle.net/1854/LU-2087965 D'aes, J., Kieu, N. P., Léclère, V., Tokarski, C., Olorunleke, F. E., De Maeyer, K., Jacques, P., Höfte, M., & Ongena, M. (2014). To settle or to move? The interplay between two classes of cyclic lipopeptides in the biocontrol strain Pseudomonas CMR12a. Environmental Microbiology, 16(7), 2282-2300. doi:10.1111/1462- 2920.12462 Darrasse, A., Darsonval, A., Boureau, T., Brisset, M.-N., Durand, K., & Jacques, M.-A. (2010). Transmission of plant-pathogenic bacteria by nonhost seeds without induction of an associated defense reaction at emergence. Applied and Environmental Microbiology, 76(20), 6787-6796. doi:10.1128/aem.01098-10 Das, A., Rangaraj, N., & Sonti, R. V. (2009). Multiple adhesin-like functions of Xanthomonas oryzae pv. oryzae are involved in promoting leaf attachment, entry, and virulence on rice. Molecular plant-microbe interactions 22(1), 73-85. doi:10.1094/mpmi-22-1-0073 Das, G., Patra, J. K., & Baek, K.-H. (2017). Insight into MAS: a molecular tool for development of stress resistant and quality of rice through gene stacking. Frontiers in Plant Science, 8, 985. doi:10.3389/fpls.2017.00985 DAWR. (2015, 26 Oct 2015). Rice. Agriculture, farming and food: Crops. Retrieved from http://www.agriculture.gov.au/ag-farm-food/crops/rice De Maayer, P., Chan, W.-Y., Blom, J., Venter, S., Duffy, B., Smits, T., & Coutinho, T. (2012). The large universal Pantoea plasmid LPP-1 plays a major role in biological and ecological diversification. BMC Genomics, 13(1), 625. Retrieved from http://www.biomedcentral.com/content/pdf/1471-2164-13-625.pdf

232

De Maayer, P., Chan, W. Y., Venter, S. N., Toth, I. K., Birch, P. R. J., Joubert, F., & Coutinho, T. A. (2010). Genome sequence of Pantoea ananatis LMG 20103, the causative agent of Eucalyptus blight and dieback. Journal of Bacteriology, 192, 2936 - 2937. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2876505/pdf/0060-10.pdf De Maayer, P., Venter, S., Kamber, T., Duffy, B., Coutinho, T., & Smits, T. (2011). Comparative genomics of the type VI secretion systems of Pantoea and Erwinia species reveals the presence of putative effector islands that may be translocated by the VgrG and Hcp proteins. BMC Genomics, 12(1), 576. Retrieved from http://www.biomedcentral.com/content/pdf/1471-2164-12-576.pdf De Masi, L., Yue, M., Hu, C., Rakov, A. V., Rankin, S. C., & Schifferli, D. M. (2017). Cooperation of adhesin alleles in Salmonella - host tropism. mSphere, 2(2). doi:10.1128/mSphere.00066-17 Degrassi, G., Devescovi, G., Kim, J., Hwang, I., & Venturi, V. (2008). Identification, characterization and regulation of two secreted polygalacturonases of the emerging rice pathogen Burkholderia glumae. FEMS Microbiology Ecology, 65(2), 251-262. doi:10.1111/j.1574-6941.2008.00516.x Detry, J. F., Chapeaux, J. P., & Tilquin, J. P. (1991a). Estimation of rice bacterial sheath brown rot (BSR) and rice blast (Bl) severity in five Burundi highland swamps. International Rice Research Newsletter, 16(6), 20-21. Detry, J. F., Duveiller, E., & Maraite, H. (1991b). Development of a method for evaluation of resistance of rice cultivars to Pseudomonas fuscovaginae. Parasitica, 47(4), 151-163. Devescovi, G., Bigirimana, J., Degrassi, G., Cabrio, L., LiPuma, J. J., Kim, J., Hwang, I., & Venturi, V. (2007). Involvement of a quorum-sensing-regulated lipase secreted by a clinical isolate of Burkholderia glumae in severe disease symptoms in rice. Applied and Environmental Microbiology, 73(15), 4950-4958. doi:10.1128/aem.00105-07 Drake, S. L., & Koomey, M. (1995). The product of the pilQ gene is essential for the biogenesis of type IV pili in Neisseria gonorrhoeae. Molecular microbiology, 18(5), 975-986.

233

Du, P. V., Loan, L. C., Cuong, N. D., Nghiep, H. V., & Thach, N. D. (2001). Survey on seed borne fungi and its effects on grain quality of common rice cultivars in the Mekong Delta. Omonrice, 9, 107 -113. Dugan, F. M., & Peever, T. L. (2002). Morphological and cultural differentiation of described species of Alternaria from Poaceae. Mycotaxon, 83, 229-264. Duveiller, E., & Maraite, H. (1990). Bacterial sheath rot of wheat caused by Pseudomonas fuscovaginae in the highlands of Mexico. Plant Disease, 74(11), 932-935. Duveiller, E., Miyajima, K., Snacken, F., Autrique, A., & Maraite, H. (1988). Characterization of Pseudomonas fuscovaginae and differentiation from other fluorescent Pseudomonads occurring on rice in Burundi. Journal of Phytopathology, 122(2), 97-107. doi:10.1111/j.1439-0434.1988.tb00995.x Duveiller, E., Notteghem, J. L., Rott, P., Snacken, F., & Maraite, H. (1990). Bacterial sheath brown rot of rice caused by Pseudomonas fuscovaginae in Malagasy. Tropical Pest Management, 36(2), 151-153. doi:10.1080/09670879009371461 Duveiller, E., Snacken, F., & Maraite, H. (1989). First detection of Pseudomonas fuscovaginae on maize and sorghum in Burundi. Plant Disease, 73(5), 514 - 517. Eriko, H., Shoichi, T., & Naoto, A. (2008). Japan Patent No. 39839562. Falk, I., & Wallace, R. (2011). Managing plant biosecurity across borders. In Falk, I., Wallace, R., & Ndoen, M. L. (Eds.), Managing biosecurity across borders (pp. 3- 19): Springer Netherlands. Fan, E., Fiedler, S., Jacob-Dubuisson, F., & Müller, M. (2012). Two-partner secretion of Gram-negative bacteria: a single β-barrel protein enables transport across the outer membrane. Journal of Biological Chemistry, 287(4), 2591-2599. doi:10.1074/jbc.M111.293068 FAO. (2017). The future of food and agriculture: trends and challenges. Retrieved from http://www.fao.org/3/a-i6881e.pdf FAO, IFAD, UNICEF, WFP, & WHO. (2017). The state of food security and nutrition in the world 2017: building resilience for peace and food security. Retrieved from http://www.fao.org/state-of-food-security-nutrition/en/ Farrell, T., Fox, K. M., Williams, R. L., Fukai, S., Reinke, R. F., & Lewin, L. G. (2001). Temperature constraints to rice production in Australia and Laos: a shared problem. Paper presented at the Increased Lowland Rice Production in the Mekong Region.

234

Feil, H., Feil, W. S., & Lindow, S. E. (2007). Contribution of fimbrial and afimbrial adhesins of Xylella fastidiosa to attachment to surfaces and virulence to grape. Phytopathology, 97(3), 318-324. doi:10.1094/PHYTO-97-3-0318 Figurski, D. H., & Helinski, D. R. (1979). Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proceedings of the National Academy of Sciences of the United States of America, 76(4), 1648- 1652. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC383447/ Fiyaz, R. A., Yadav, A. K., Krishnan, S. G., Ellur, R. K., Bashyal, B. M., Grover, N., Bhowmick, P. K., Nagarajan, M., Vinod, K. K., Singh, N. K., Prabhu, K. V., & Singh, A. K. (2016). Mapping quantitative trait loci responsible for resistance to Bakanae disease in rice. Rice, 9(1), 45. doi:10.1186/s12284-016-0117-2 Flamand, M.-C., Ewbank, E., Goret, B., & Maraite, H. (1997). Production of phytotoxic lipodepsipeptides by Pseudomonas fuscovaginae. In Rudolph, K., Burr, T. J., Mansfield, J. W., Stead, D., Vivian, A., & Kietzell, J. (Eds.), Pseudomonas syringae pathovars and related pathogens (Vol. 9, pp. 221-226): Springer Netherlands. Frost, L. S., Leplae, R., Summers, A. O., & Toussaint, A. (2005). Mobile genetic elements: the agents of open source evolution. Nature reviews Microbiology, 3(9), 722-732. doi:10.1038/nrmicro1235 Fuqua, C., Parsek, M. R., & Greenberg, E. P. (2001). Regulation of gene expression by cell-to-cell communication: acyl-homoserine lactone quorum sensing. Annual Review of Genetics, 35(1), 439-468. doi:doi:10.1146/annurev.genet.35.102401.090913 Furuya, N., Ito, T., & Tsuchiya, K. (2009). Occurrence of bacterial brown stripe of creeping bentgrass on golf course green in Kyushu. Journal of the Faculty of Agriculture, Kyushu University, 54(1), 13 -17. Ganesan, S., Vadivel, K., & Jayaraman, J. (2015). Sustainable crop disease management using natural products: CABI. Gavin, H. E., & Satchell, K. J. F. (2015). MARTX toxins as effector delivery platforms. Pathogens and Disease, 73(9), ftv092. doi:10.1093/femspd/ftv092 Gelencsér, Z., Galbáts, B., Gonzalez, J. F., Choudhary, K. S., Hudaiberdiev, S., Venturi, V., & Pongor, S. (2012). Chromosomal arrangement of AHL-driven quorum sensing circuits in Pseudomonas. ISRN Microbiology, 2012, 6. doi:10.5402/2012/484176

235

Gerlach, R. G., & Hensel, M. (2007). Protein secretion systems and adhesins: the molecular armory of Gram-negative pathogens. International Journal of Medical Microbiology, 297(6), 401-415. doi:10.1016/j.ijmm.2007.03.017 Ghazemi, A., Rostami, M., & Yazdaninia, T. (2012). In vitro antagonistic and biodegradation activity of a newly isoalted Delftia tsuruhatensis from rice plant in Iran. International Journal of Agriculture and Crop Sciences, 4(2), 40-44. Gnanamanickam, S. S. (2009). Biological control of sheath-rot and other fungal diseases Biological control of rice diseases (Vol. 8, pp. 91-93): Springer Netherlands. Goto, K., & Ohata, K. (1956). New bacterial diseases of rice (brown stripe and grain rot). Annals of Phytophathological Society of Japan, 21, 46 - 47. Goto, M. (1979). Bacterial foot rot of rice caused by a strain of Erwinia chrysanthemi. Goto, T., & Watanabe, B. (1975). Varietal resistance to bacterial grain rot of rice, caused by Pseudomonas glumae Proceedings of the Association for Plant Protection, Kyushu, 21, 141 - 143. Gottig, N., Garavaglia, B. S., Garofalo, C. G., Orellano, E. G., & Ottado, J. (2009). A filamentous hemagglutinin-like protein of Xanthomonas axonopodis pv. citri, the phytopathogen responsible for citrus canker, is involved in bacterial virulence. PLoS ONE, 4(2), e4358. doi:10.1371/journal.pone.0004358 Ham, J. H., & Groth, D. E. (2011). Bacterial panicle blight an emerging rice disease. Louisiana Agriculture. Ham, J. H., Melanson, R. A., & Rush, M. C. (2011). Burkholderia glumae: next major pathogen of rice? Molecular Plant Pathology, 12(4), 329-339. doi:10.1111/j.1364- 3703.2010.00676.x Hamburger, J., Abbasi, I., Kariuki, C., Wanjala, A., Mzungu, E., Mungai, P., Muchiri, E., & King, C. H. (2013). Evaluation of loop-mediated isothermal amplification suitable for molecular monitoring of schistosome-infected snails in field laboratories. The American journal of tropical medicine and hygiene, 88(2). doi:10.4269/ajtmh.2012.12-0208 Hanahan, D. (1985). Techniques for transformation of E. coli. In Glover, D. M. (Ed.), DNA cloning : a practical approach (Vol. 1, pp. 109-135). Oxford, United Kingdom: IRL Press. HerbiGuide. (2017). Kentucky blue grass. Retrieved from http://www.herbiguide.com.au/Descriptions/hg_Kentucky_Blue_Grass.htm

236

Heydari-Nezhad, A. M., Babaeizad, V., & Rahimian, H. (2015). Role of resistance- related genes in the interaction of rice with the causal agent of bacterial brown stripe. Iranian Journal of Plant Pathology, 51(4), Pe523-Pe539. Hikichi, Y., Okuno, T., & Furusawa, I. (1995). Mode of action of oxolinic acid against bacterial seedling rot of rice caused by Pseudomonas glumae. III. Infection with P. glumae into plumules. Annals of Phytopathology Society of Japan, 61, 134 -136. Hirai, H., Takai, R., Iwano, M., Nakai, M., Kondo, M., Takayama, S., Isogai, A., & Che, F.-S. (2011). Glycosylation regulates specific induction of rice immune responses by Acidovorax avenae flagellin. Journal of Biological Chemistry, 286(29), 25519- 25530. doi:10.1074/jbc.M111.254029 Hiroyuki, U., Naruto, F., Kazuhiro, I., Masatomo, H., Kenichi, T., & Nobuaki, M. (2006). Burkholderia gladioli associated with symptoms of bacterial grain rot and leaf- sheath browning of rice plants. Journal of General Plant Pathology, 72(2), 98. Retrieved from http://proquest.umi.com/pqdweb?did=1013193001&Fmt=7&clientId=20870&RQ T=309&VName=PQD Huang, J. (1986). Ultrastructure of bacterial penetration in plants. Annual Review of Phytopathology, 24(1), 141-157. doi:doi:10.1146/annurev.py.24.090186.001041 Hutchison, M. L., & Gross, D. C. (1997). Lipopeptide phytotoxins produced by Pseudomonas syringae pv. syringae: comparison of the biosurfactant and ion channel-forming activities of syringopeptin and syringomycin. Molecular Plant- Microbe Interactions, 10(3), 347-354. doi:10.1094/MPMI.1997.10.3.347 Iacobellis, N. S., Lavermicocca, P., Grgurina, I., Simmaco, M., & Ballio, A. (1992). Phytotoxic properties of Pseudomonas syringae pv. syringae toxins. Physiological and Molecular Plant Pathology 40(2), 107-116. doi:http://dx.doi.org/10.1016/0885-5765(92)90038-W IRRI. (2017a). Bacterial sheath brown rot. Rice Knowledge Bank. Retrieved from http://www.knowledgebank.irri.org/decision-tools/rice-doctor/rice-doctor-fact- sheets/item/bacterial-sheath-brown-rot?category_id=345 IRRI. (2017b). Increasing food security. Retrieved from http://irri.org/our- impact/increase-food-security Jacob-Dubuisson, F., Fernandez, R., & Coutte, L. (2004). Protein secretion through autotransporter and two-partner pathways. Biochimica et Biophysica Acta (BBA) -

237

Molecular Cell Research, 1694(1), 235-257. doi:https://doi.org/10.1016/j.bbamcr.2004.03.008 Jacob-Dubuisson, F., Guerin, J., Baelen, S., & Clantin, B. (2013). Two-partner secretion: as simple as it sounds? Research in microbiology 164(6), 583-595. doi:10.1016/j.resmic.2013.03.009 Jacob‐Dubuisson, F., Locht, C., & Antoine, R. (2001). Two‐partner secretion in Gram‐ negative bacteria: a thrifty, specific pathway for large virulence proteins. Molecular microbiology, 40(2), 306-313. Jacques, M. A., Arlat, M., Boulanger, A., Boureau, T., Carrere, S., Cesbron, S., Chen, N. W., Cociancich, S., Darrasse, A., Denance, N., Fischer-Le Saux, M., Gagnevin, L., Koebnik, R., Lauber, E., Noel, L. D., Pieretti, I., Portier, P., Pruvost, O., Rieux, A., Robene, I., Royer, M., Szurek, B., Verdier, V., & Verniere, C. (2016). Using ecology, physiology, and genomics to understand host specificity in Xanthomonas: French network on Xanthomonads (FNX). Annual review of phytopathology doi:10.1146/annurev-phyto-080615-100147 Jaunet, T., Laguerre, G., Lemanceau, P., Frutos, R., & Notteghem, J. L. (1995). Diversity of Pseudomonas fuscovaginae and other fluorescent pseudomonads isolated from diseased rice. Phytopathology, 85(12), 1534-1541. doi:10.1094/Phyto-85-1534 Jaunet, T., Notteghem, J. L., & Rapilly, F. (1996). Pathogenicity process of Pseudomonas fuscovaginae, the causal agent of sheath brown rot of rice. Journal of Phytopathology, 144(9-10), 425-430. doi:10.1111/j.1439-0434.1996.tb00318.x Jeong, Y., Kim, J., Kim, S., Kang, Y., Nagamatsu, T., & Hwang, I. (2003). Toxoflavin produced by Burkholderia glumae causing rice grain rot is responsible for inducing bacterial wilt in many field crops. Plant Disease, 87, 890 - 895. Jones, J. D. G., & Dangl, J. L. (2006). The plant immune system. Nature, 444(7117), 323- 329. Retrieved from http://dx.doi.org/10.1038/nature05286 Kakar, K. U., Duan, Y. P., Nawaz, Z., Sun, G., Almoneafy, A. A., Hassan, M. A., Elshakh, A., Li, B., & Xie, G.-L. (2014). A novel rhizobacterium Bk7 for biological control of brown sheath rot of rice caused by Pseudomonas fuscovaginae and its mode of action. European Journal of Plant Pathology, 138(4), 819-834. doi:10.1007/s10658-013-0356-7 Kalpana, K., Maruthasalam, S., Rajesh, T., Poovannan, K., Kumar, K. K., Kokiladevi, E., Raja, J. A. J., Sudhakar, D., Velazhahan, R., Samiyappan, R., & Balasubramanian,

238

P. (2006). Engineering sheath blight resistance in elite indica rice cultivars using genes encoding defense proteins. Plant Science, 170(2), 203-215. doi:https://doi.org/10.1016/j.plantsci.2005.08.002 Kaneko, H., Kawana, T., Fukushima, E., & Suzutani, T. (2007). Tolerance of loop- mediated isothermal amplification to a culture medium and biological substances. Journal of Biochemical and Biophysical Methods, 70(3), 499-501. doi:10.1016/j.jbbm.2006.08.008 Kang, Y., Liu, H., Genin, S., Schell, M. A., & Denny, T. P. (2002). Ralstonia solanacearum requires type 4 pili to adhere to multiple surfaces and for natural transformation and virulence. Molecular Microbiology, 46(2), 427-437. doi:10.1046/j.1365-2958.2002.03187.x Kaur, S., & Singh, S. (2015). Impact of crop establishment methods and weed control on weeds, insect-pest and disease infestation in rice in north-western indo-gangetic plains. International Journal of Agriculture Sciences, 7(4), 487-491. Kealey, L. M., & Clampett, W. S. (2000). Production of quality rice in south eastern Australia : overview. Kingston, A.C.T.: Rural Industries Research Development Corporation. Khoa, N. D., Thuy, P. T., Thuy, T. T., Collinge, D. B., & Jorgensen, H. J. (2011). Disease- reducing effect of Chromolaena odorata extract on sheath blight and other rice diseases. Phytopathology, 101(2), 231-240. doi:10.1094/phyto-04-10-0113 Kido, K., Hasegawa, M., Matsumoto, H., Kobayashi, M., & Takikawa, Y. (2010). Pantoea ananatis strains are differentiated into three groups based on reactions of tobacco and welsh onion and on genetic characteristics. Journal of General Plant Pathology, 76(3), 208-218. doi:10.1007/s10327-010-0230-9 Kim, J.-W., Kang, Y.-S., Kim, J.-G., Choi, O.-H., & Hwang, I.-G. (2010). Occurrence of Burkholderia glumae on rice and field crops in Korea. The Plant Pathology Journal, 26(3), 271 - 272. Kim, J., Choi, O., & Kim, W. I. (2015). First report of sheath brown rot of rice caused by Pseudomonas fuscovaginae in Korea. Plant Disease 99(7), 1033-1033. doi:10.1094/PDIS-11-14-1219-PDN Kim, J., Kang, Y., Choi, O., Jeong, Y., Jeong, J.-E., Lim, J. Y., Kim, M., Moon, J. S., Suga, H., & Hwang, I. (2007). Regulation of polar flagellum genes is mediated by

239

quorum sensing and FlhDC in Burkholderia glumae. Molecular Microbiology, 64(1), 165-179. doi:10.1111/j.1365-2958.2007.05646.x Kim, J., Kim, J.-G., Kang, Y., Jang, J. Y., Jog, G. J., Lim, J. Y., Kim, S., Suga, H., Nagamatsu, T., & Hwang, I. (2004). Quorum sensing and the LysR-type transcriptional activator ToxR regulate toxoflavin biosynthesis and transport in Burkholderia glumae. Molecular Microbiology, 54(4), 921-934. doi:10.1111/j.1365-2958.2004.04338.x Kini, K., Agnimonhan, R., Afolabi, O., Milan, B., Soglonou, B., Gbogbo, V., Koebnik, R., & Silué, D. (2016). First report of a new bacterial leaf blight of rice caused by Pantoea ananatis and Pantoea stewartii in Benin. Plant Disease, 101(1), 242-242. doi:10.1094/PDIS-06-16-0940-PDN Klemm, P. (1985). Fimbrial adhesions of Escherichia coli. Reviews of Infectious Diseases 7(3), 321-340. Klemm, P., & Schembri, M. A. (2000). Bacterial adhesins: function and structure. International Journal of Medical Microbiology, 290(1), 27-35. doi:http://dx.doi.org/10.1016/S1438-4221(00)80102-2 Kline, K. A., Fälker, S., Dahlberg, S., Normark, S., & Henriques-Normark, B. (2009). Bacterial adhesins in host-microbe interactions. Cell Host & Microbe, 5(6), 580- 592. doi:https://doi.org/10.1016/j.chom.2009.05.011 Kluyver, A. J. (1956). Pseudomonas aureofaciens nov. spec. and its pigments. Journal of Bacteriology, 72(3), 406-411. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC357921/ Kumar, R., Gupta, A., Maheshwari, V. K., & Atwal, S. S. (2014, 23 June 2014). Major diseases and their management in basmati rice. Retrieved from http://www.krishisewa.com/articles/disease-management/419-diseases-basmati- rice.html Lalucat, J., Bennasar, A., Bosch, R., García-Valdés, E., & Palleroni, N. J. (2006). Biology of Pseudomonas stutzeri. Microbiology and Molecular Biology Reviews, 70(2), 510-547. doi:10.1128/MMBR.00047-05 Lana, U. G. D. P., Gomes, E. A., Silva, D. D., Costa, R. V., Cota, L. V., Parreira, D. F., Souza, I. R. P., & Guimarães, C. T. (2012). Detection and molecular diversity of Pantoea ananatis associated with white spot disease in maize, sorghum and

240

crabgrass in Brazil. Journal of Phytopathology, 160(9), 441-448. doi:10.1111/j.1439-0434.2012.01924.x Lang, J. M., Langlois, P., Nguyen, M. H. R., Triplett, L. R., Purdie, L., Holton, T. A., Djikeng, A., Vera Cruz, C. M., Verdier, V., & Leach, J. E. (2014). Sensitive detection of Xanthomonas oryzae pathovars oryzae and oryzicola by loop-mediated isothermal amplification. Applied and Environmental Microbiology, 80(15), 4519- 4530. doi:10.1128/AEM.00274-14 Lanoiselet, V., Cother, E. J., & Ash, G. J. (2001). Risk assessment of exotic plant diseases to the Australian rice industry with emphasis on rice blast. Lanoiselet, V. M., Cother, E. J., Ash, G. J., Hind-Lanoiselet, T. L., Murray, G. M., & Harper, J. D. I. (2005). Prevalence and survival, with emphasis on stubble burning, of Rhizoctonia spp., causal agents of sheath diseases of rice in Australia. Australasian Plant Pathology, 34(2), 135-142. Lee, H. B., Hong, J. P., & Kim, S. B. (2010). First report of leaf blight caused by Pantoea agglomerans on rice in Korea. Plant Disease, 94(11), 1372-1372. doi:10.1094/PDIS-05-10-0374 Lelliott, R. A., Billing, E., & Hayward, A. C. (1966). A determinative scheme for the fluorescent plant pathogenic Pseudomonads. Journal of Applied Bacteriology, 29(3), 470-489. doi:10.1111/j.1365-2672.1966.tb03499.x Li, B., Liu, B., Yu, R., Tao, Z., Wang, Y., Xie, G., Li, H., & Sun, G. (2011). Bacterial brown stripe of rice in soil-less culture system caused by Acidovorax avenae subsp. avenae in China. Journal of General Plant Pathology, 77(1), 64-67. doi:10.1007/s10327-010-0280-z Li, X., Nie, J., Ward, L., Madani, M., Hsiang, T., Zhao, Y., & De Boer, S. H. (2009). Comparative genomics-guided loop-mediated isothermal amplification for characterization of Pseudomonas syringae pv. phaseolicola. Journal of Applied Microbiology, 107(3), 717-726. doi:10.1111/j.1365-2672.2009.04262.x Lim, J. Y., Lee, T.-H., Nahm, B. H., Choi, Y. D., Kim, M., & Hwang, I. (2009). Complete genome sequence of Burkholderia glumae BGR1. Journal of bacteriology, 191(11), 3758-3759. doi:10.1128/jb.00349-09 Liu, H., Coulthurst, S. J., Pritchard, L., Hedley, P. E., Ravensdale, M., Burr, T., Takle, G., Brurberg, M.-B., Birch, P. R. J., Salmond, G. P. C., & Toth, I. K. (2008).

241

Quorum sensing coordinates brute force and stealth modes of infection in the plant pathogen Pectobacterium atrosepticum. PLoS Pathology, 4, e1000093. Liu, H., Kang, Y., Genin, S., Schell, M. A., & Denny, T. P. (2001). Twitching motility of Ralstonia solanacearum requires a type IV pilus system. Microbiology, 147(12), 3215-3229. doi:10.1099/00221287-147-12-3215 Liu, H., Tian, W.-X., Ibrahim, M., Li, B., Zhang, G.-Q., Zhu, B., & Xie, G.-L. (2012). Characterization of pilP, a gene required for twitching motility, pathogenicity, and biofilm formation of Acidovorax avenae subsp. avenae RS-1. European Journal of Plant Pathology, 134(3), 551-560. doi:10.1007/s10658-012-0038-x Loaces, I., Ferrando, L., & Fernández Scavino, A. (2011). Dynamics, diversity and function of endophytic siderophore-producing bacteria in rice. Microbial Ecology, 61(3), 606-618. doi:10.1007/s00248-010-9780-9 Loper, J. E., Hassan, K. A., Mavrodi, D. V., Davis, E. W., Lim, C. K., Shaffer, B. T., Elbourne, L. D. H., Stockwell, V. O., Hartney, S. L., Breakwell, K., Henkels, M. D., Tetu, S. G., Rangel, L. I., Kidarsa, T. A., Wilson, N. L., de Mortel, J. E. V., Song, C. X., Blumhagen, R., Radune, D., Hostetler, J. B., Brinkac, L. M., Durkin, A. S., Kluepfel, D. A., Wechter, W. P., Anderson, A. J., Kim, Y. C., Pierson, L. S., Pierson, E. A., Lindow, S. E., Kobayashi, D. Y., Raaijmakers, J. M., Weller, D. M., Thomashow, L. S., Allen, A. E., & Paulsen, I. T. (2012). Comparative genomics of plant-associated Pseudomonas spp.: insights into diversity and inheritance of traits involved in multitrophic interactions. PLoS Genetics, 8(7). doi:10.1371/journal.pgen.1002784 Lu, H., Patil, P., Van Sluys, M. A., White, F. F., Ryan, R. P., Dow, J. M., Rabinowicz, P., Salzberg, S. L., Leach, J. E., Sonti, R., Brendel, V., & Bogdanove, A. J. (2008). Acquisition and evolution of plant pathogenesis-associated gene clusters and candidate determinants of tissue-specificity in Xanthomonas. PLoS One, 3(11), e3828. doi:10.1371/journal.pone.0003828 Luo, J., Xie, G., Li, B., & Lihui, X. (2007). First report of Burkholderia glumae isolated from symptomless rice seeds in China. Plant Disease, 91, 1363 - 1363. Lyon, N. (2013, 2 Jul 2013, 4 a.m.). Dryland rice trial proves potential. Retrieved from http://www.queenslandcountrylife.com.au/story/3588880/dryland-rice-trial- proves-potential/#!

242

Macapuguay, F., & Mnzaya, M. (1988). Progress on irrigated rice agronomy at Usangu village, M'beya zone, Tanzania. International Rice Research Newsletter, 37, 40-44. MacKenzie, K. D., Palmer, M. B., Köster, W. L., & White, A. P. (2017). Examining the link between biofilm formation and the ability of pathogenic Salmonella strains to colonize multiple host species. Frontiers in Veterinary Science, 4, 138. doi:10.3389/fvets.2017.00138 Mahadevaiah, C., Shailaja, H., Uday, G., & Kumar, M. K. P. (2015). Standardization of disease screening protocol for sheath rot disease in rice. International Journal of Agricultural Science and Research (IJASR), 5(5), 129-137. Malavolta, V. M. A., de Almeida, I. M. G., & Malavolta Jr, V. A. (1997a). Characterization of Pseudomonas fuscovaginae on rice in Sao Paulo State, Brazil, and reaction of rice cultivars to the bacterium. Summa Phytopathologica 23(1), 29 - 35. Malavolta, V. M. A., De Almeida, I. M. G., & Malavolta, J. V. A. (1997b). Characterization of Pseudomonas fuscovaginae on rice in Sao Paulo State, Brazil, and reaction of rice cultivars to the bacterium. Summa Phytopathologica, 23(1), 29 - 35. Manamgoda, D. S., Rossman, A. Y., Castlebury, L. A., Crous, P. W., Madrid, H., Chukeatirote, E., & Hyde, K. D. (2014). The genus Bipolaris. Studies in Mycology, 79(Supplement C), 221-288. doi:https://doi.org/10.1016/j.simyco.2014.10.002 Mann, R. A., Smits, T. H. M., Bühlmann, A., Blom, J., Goesmann, A., Frey, J. E., Plummer, K. M., Beer, S. V., Luck, J., Duffy, B., & Rodoni, B. (2013). Comparative genomics of 12 strains of Erwinia amylovora identifies a pan-genome with a large conserved core. PLoS ONE, 8(2), e55644. doi:10.1371/journal.pone.0055644 Matthews, P., McCaffery, D., & Jenkins, L. (2013). Winter crop variety sowing guide 2013. Orange, NSW, Australia: NSW Department of Primary Industries. Mattiuzzo, M., Bertani, I., Ferluga, S., Cabrio, L., Bigirimana, J., Guarnaccia, C., Pongor, S., Maraite, H., & Venturi, V. (2011). The plant pathogen Pseudomonas fuscovaginae contains two conserved quorum sensing systems involved in virulence and negatively regulated by RsaL and the novel regulator RsaM. Environmental Microbiology 13(1), 145-162. doi:10.1111/j.1462- 2920.2010.02316.x

243

McIntyre, S., Finlayson, C. M., Ladiges, P. Y., & Mitchell, D. S. (1991). Weed community composition and rice husbandry practices in New South Wales, Australia. Agriculture, Ecosystems & Environment, 35(1), 27-45. doi:https://doi.org/10.1016/0167-8809(91)90074-8 Melvin, J. A., Scheller, E. V., Noël, C. R., & Cotter, P. A. (2015). New insight into filamentous hemagglutinin secretion reveals a role for full-length FhaB in Bordetella Virulence. mBio, 6(4). doi:10.1128/mBio.01189-15 Meng, Y., Li, Y., Galvani, C. D., Hao, G., Turner, J. N., Burr, T. J., & Hoch, H. C. (2005). Upstream migration of Xylella fastidiosa via pilus-driven twitching motility. Journal of Bacteriology, 187(16), 5560-5567. doi:10.1128/JB.187.16.5560- 5567.2005 Mergaert, J., Verdonck, L., & Kersters, K. (1993). Transfer of Erwinia ananas (synonym, Erwinia uredovora) and Erwinia stewartii to the Genus Pantoea emend. as Pantoea ananas (Serrano 1928) comb. nov. and Pantoea stewartii (Smith 1898) comb. nov., respectively, and description of Pantoea stewartii subsp. indologenes subsp. nov. International Journal of Systematic and Evolutionary Microbiology, 43(1), 162- 173. doi:doi:10.1099/00207713-43-1-162 Mew, T. W., & Gonzales, P. (2002). A handbook of rice seedborne fungi: International Rice Research Institute. Mew, T. W., Leung, H., Savary, S., Vera Cruz, C. M., & Leach, J. E. (2004). Looking ahead in rice disease research and management. Critical reviews in plant sciences, 23(2), 103-127. Meyerson, L. A., & Reaser, J. K. (2002). Biosecurity: moving towards a comprehensive approach. BioScience, 52(7), 593-600. doi:10.1641/0006- 3568(2002)052[0593:bmtaca]2.0.co;2 Mhedbi-Hajri, N., Darrasse, A., Pigne, S., Durand, K., Fouteau, S., Barbe, V., Manceau, C., Lemaire, C., & Jacques, M. A. (2011). Sensing and adhesion are adaptive functions in the plant pathogenic Xanthomonads. BMC Evolutionary Biology 11, 67. doi:10.1186/1471-2148-11-67 Miyajima, K. (1980). Ecological study of Pseudomonas fuscovaginae, the causal bacterium of sheath brown rot of rice, by the phage method. Bulletin of Hokkaido Prefectural Agricultural Experiment Stations, 43, 42 -51.

244

Miyajima, K., Tanii, A., & Akita, T. (1983a). Pseudomonas fuscovaginae sp. nov., nom. rev. International Journal of Systematic Bacteriology 33(3), 656-657. doi:10.1099/00207713-33-3-656 Miyajima, K., Tanii, A., & Tadahiko, A. (1983b). Pseudomonas fuscovaginae sp. nov., nom. rev. International Journal of Systematic Bacteriology, 33(3), 656-657. doi:10.1099/00207713-33-3-656 Mizobuchi, R., Fukuoka, S., Tsushima, S., Yano, M., & Sato, H. (2016). QTLs for resistance to major rice diseases exacerbated by global warming: brown spot, bacterial seedling rot, and bacterial grain rot. Rice, 9, 23. doi:10.1186/s12284-016- 0095-4 Mizobuchi, R., Sato, H., Fukuoka, S., Tanabata, T., Tsushima, S., Imbe, T., & Yano, M. (2013). Mapping a quantitative trait locus for resistance to bacterial grain rot in rice. Rice, 6(1), 1-10. doi:10.1186/1939-8433-6-13 Mogi, S., & Tsushima, S. (1985). Varietal resistance to bacterial grain rot in rice, caused by Pseudomonas glumae. Kyushu Agricultural Research, 47, 103. Mondal, K. K., Mani, C., Singh, J., Kim, J. G., & Mudgett, M. B. (2011). A new leaf blight of rice caused by Pantoea ananatis in India. Plant Disease, 95(12), 1582- 1582. doi:10.1094/pdis-06-11-0533 Morohoshi, T., Nakamura, Y., Yamazaki, G., Ishida, A., Kato, N., & Ikeda, T. (2007). The plant pathogen Pantoea ananatis produces N-acyl homoserine lactone and causes center rot disease of onion by quorum sensing. Journal of bacteriology, 189(22), 8333-8338. doi:10.1128/jb.01054-07 Morohoshi, T., Ogata, Y., & Ikeda, T. (2011). Cell aggregation is negatively regulated by N-acyl homoserine lactone-mediated quorum sensing in Pantoea ananatis SK-1. Journal of Bioscience and Bioengineering, 112(6), 566-569. doi:http://dx.doi.org/10.1016/j.jbiosc.2011.08.004 Morohoshi, T., Oseki, K., & Ikeda, T. (2011). Exopolysaccharide production is influenced by sugars, N-acyl homoserine lactone, and transcriptional regulators RcsA and RcsB, but does not affect pathogenicity in the plant pathogen Pantoea ananatis. Bioscience, Biotechnology, and Biochemistry, 75(5), 997-999. Retrieved from http://www.tandfonline.com/doi/pdf/10.1271/bbb.100888

245

Mudingotto, P. J., Veena, M. S., & Mortensen, C. N. (2001). First report of bacterial blight caused by Acidovorax avenae subsp. avenae associated with finger millet seeds from Uganda. New Disease Reports, 4, 15. Mulet, M., Lalucat, J., & Garcia-Valdes, E. (2010). DNA sequence-based analysis of the Pseudomonas species. Environmental Microbiology, 12(6), 1513-1530. doi:10.1111/j.1462-2920.2010.02181.x Mundt, C. C. (2014). Durable resistance: a key to sustainable management of pathogens and pests. Infection, Genetics and Evolution, 27(Supplement C), 446-455. doi:https://doi.org/10.1016/j.meegid.2014.01.011 Mvuyekure, S. M., Sibiya, J., Derera, J., Nzungize, J., & Nkima, G. (2017). Assessment of genetic diversity of rice based on SNP markers for selection of parents for sheath rot (Sarocladium oryzae) resistance breeding. South African Journal of Plant and Soil, 0-0. doi:10.1080/02571862.2017.1333636 Naeimi, S., Okhovvat, S. M., Hedjaroude, G. A., & Khosravi, V. (2003). Sheath rot of rice in Iran. Communications in Agricultural and Applied Biological Sciences, 61(4), 681 - 684. Nakashima, N., & Miyazaki, K. (2014). Bacterial cellular engineering by genome editing and gene silencing. International Journal of Molecular Sciences, 15(2), 2773-2793. doi:10.3390/ijms15022773 Nandakumar, R., Bollich, P., Groth, D., & Rush, M. C. (2007). Confirmation of the partial resistance of Jupiter rice to bacterial panicle blight caused by Burkholderia glumae through reduced disease and yield loss in inoculated field tests. Phytopathology, 97, S82 - S83. Nandakumar, R., Bollich, P. A., Shahjahan, A. K. M., Groth, D. E., & Rush, M. C. (2008). Evidence for the soilborne nature of the rice sheath rot and panicle blight pathogen, Burkholderia gladioli 1. Canadian Journal of Plant Pathology, 30(1), 148-154. doi:10.1080/07060660809507505 Nandakumar, R., & Rush, M. C. (2008). Analysis of gene expression in Jupiter rice showing partial resistance to rice panicle blight caused by Burkholderia glumae. Phytopathology, 98, S112. Nandakumar, R., Rush, M. C., Shahjahan, A. K. M., O'Reilly, K., & Groth, D. (2005). Bacterial panicle blight of rice in the southern United States caused by Burkholderia glumae and B. gladioli. Phytopathology, 95, S73.

246

Nandakumar, R., Shahjahan, A. K. M., Yuan, X. L., Dickstein, E. R., Groth, D. E., Clark, C. A., Cartwright, R. D., & Rush, M. C. (2009). Burkholderia glumae and B. gladioli cause bacterial panicle blight in rice in the southern United States. Plant Disease, 93(9), 896-905. doi:10.1094/pdis-93-9-0896 Nemudryi, A. A., Valetdinova, K. R., Medvedev, S. P., & Zakian, S. M. (2014). TALEN and CRISPR/Cas genome editing systems: tools of discovery. Acta Naturae, 6(3), 19-40. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4207558/ Newman, C. L., & Stathopoulos, C. (2004). Autotransporter and two-partner secretion: delivery of large-size virulence factors by Gram-negative bacterial pathogens. Critical Reviews in Microbiology, 30(4), 275-286. doi:10.1080/10408410490499872 Nguyen, L. C., Taguchi, F., Tran, Q. M., Naito, K., Yamamoto, M., Ohnishi-Kameyama, M., Ono, H., Yoshida, M., Chiku, K., Ishii, T., Inagaki, Y., Toyoda, K., Shiraishi, T., & Ichinose, Y. (2012). Type IV pilin is glycosylated in Pseudomonas syringae pv. tabaci 6605 and is required for surface motility and virulence. Molecular Plant Pathology 13(7), 764-774. doi:10.1111/j.1364-3703.2012.00789.x Noël, C. R., Mazar, J., Melvin, J. A., Sexton, J. A., & Cotter, P. A. (2012). The prodomain of the Bordetella two-partner secretion pathway protein FhaB remains intracellular yet affects the conformation of the mature C-terminal domain. Molecular Microbiology, 86(4), 988-1006. doi:10.1111/mmi.12036 Notomi, T., Okayama, H., Masubuchi, H., Yonekawa, T., Watanabe, K., Amino, N., & Hase, T. (2000). Loop-mediated isothermal amplification of DNA. Nucleic Acids Research, 28(12). doi:10.1093/nar/28.12.e63 NSW DPI. (2012). Rice pre-season meeting summary Aug-Sept 2012 [PowerPoint Presentation]: Rice Growers Association Australia. NSW WeedWise. (2017). Brown-top bent (Agrostis capillaris): Department of Primary Industries, New South Wales. Onasanya, A., Basso, A., Somado, E., Gasore, E. R., Nwilene, F. E., Ingelbrecht, I., Lamo, J., Wydra, K., Ekperigin, M. M., Langa, M., Oyelakin, O., Sere, Y., Winter, S., & Onasanya, R. O. (2010). Development of a combined molecular diagnostic and DNA fingerprinting technique for rice bacteria pathogens in Africa. Biotechnology, 9(2), 89-105. Retrieved from

247

http://ezproxy.csu.edu.au/login?url=http://search.ebscohost.com/login.aspx?direct =true&db=a9h&AN=56576101&site=ehost-live Ovenden, B. (2016, 2/15/2016). [Information on breeding background of rice varieties]. Overbeek, R., Olson, R., Pusch, G. D., Olsen, G. J., Davis, J. J., Disz, T., Edwards, R. A., Gerdes, S., Parrello, B., Shukla, M., Vonstein, V., Wattam, A. R., Xia, F., & Stevens, R. (2014). The SEED and the rapid annotation of microbial genomes using subsystems technology (RAST). Nucleic Acids Research, 42(Database issue), D206-214. doi:10.1093/nar/gkt1226 Ozen, A. I., & Ussery, D. W. (2012). Defining the Pseudomonas genus: where do we draw the line with Azotobacter? Microbial Ecology, 63(2), 239-248. doi:10.1007/s00248-011-9914-8 Palmer, R., Cornuault, V., Marcus, S. E., Knox, J. P., Shewry, P. R., & Tosi, P. (2015). Comparative in situ analyses of cell wall matrix polysaccharide dynamics in developing rice and wheat grain. Planta, 241(3), 669-685. doi:10.1007/s00425- 014-2201-4 Pan, X., Yang, Y., & Zhang, J.-R. (2014). Molecular basis of host specificity in human pathogenic bacteria. Emerging Microbes & Infections, 3(3), e23. doi:10.1038/emi.2014.23 Papademetriou, M. K. (2005). Rice production in the Asia-Pacific region: issues and perspectives. Retrieved from http://www.fao.org/docrep/003/x6905e/x6905e04.htm Patel, H. K., Da Silva, D. P., Devescovi, G., Maraite, H., Paszkiewicz, K., Studholme, D. J., & Venturi, V. (2012). Draft genome sequence of Pseudomonas fuscovaginae, a broad-host-range pathogen of plants. Journal of Bacteriology, 194(10), 2765-2766. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3347198/pdf/zjb2765.pdf Patel, H. K., Matiuzzo, M., Bertani, I., Bigirimana, V. d. P., Ash, G. J., Höfte, M., & Venturi, V. (2014). Identification of virulence associated loci in the emerging broad host range plant pathogen Pseudomonas fuscovaginae. BMC Microbiology, 14(1), 1-13. doi:10.1186/s12866-014-0274-7 Patel, H. K. D. S., Daniel Passos ; Devescovi, Giulia ; Maraite, Henri ; Paszkiewicz, Konrad ; Studholme, David J ; Venturi, Vittorio. (2012). Draft genome sequence of

248

Pseudomonas fuscovaginae, a broad-host-range pathogen of plants. Journal of bacteriology, 194(10), 2765-2766. Phat, C. T., Duong, N. T., & Du, L. T. (2005). Influence of grain discoloration to seed quality. Omonrice, 13, 139 -144. PlantNET. (2017a). Fact sheet ( Leersia hexandra Sw.) New South Wales Flora Online. Retrieved from http://plantnet.rbgsyd.nsw.gov.au/cgi- bin/NSWfl.pl?page=nswfl&lvl=sp&name=Leersia~hexandra PlantNET. (2017b). Fact sheet (Phalaris arundinacea L.) New South Wales Flora Online. Retrieved from http://plantnet.rbgsyd.nsw.gov.au/cgi- bin/NSWfl.pl?page=nswfl&lvl=sp&name=Phalaris~arundinacea Plantwise. (2017a). Plantwise technical factsheet - bacterial leaf blight (Acidovorax avenae subsp. avenae). Plantwise Knowledge Bank: CABI. Plantwise. (2017b). Sheath brown rot (Pseudomonas fuscovaginae) Plantwise Technical Factsheet: Plantwise Knowledge Bank. Pradhan, B. B., Ranjan, M., & Chatterjee, S. (2012). XadM, a novel adhesin of Xanthomonas oryzae pv. oryzae, exhibits similarity to Rhs family proteins and is required for optimum attachment, biofilm formation, and virulence. Molecular Plant-Microbe Interactions 25(9), 1157-1170. doi:10.1094/mpmi-02-12-0049-r Quibod, I. L., Grande, G., Oreiro, E. G., Borja, F. N., Dossa, G. S., Mauleon, R., Cruz, C. V., & Oliva, R. (2015a). Rice-Infecting Pseudomonas Genomes Are Highly Accessorized and Harbor Multiple Putative Virulence Mechanisms to Cause Sheath Brown Rot. PLoS ONE, 10(9), e0139256. doi:10.1371/journal.pone.0139256 Quibod, I. L., Grande, G., Oreiro, E. G., Borja, F. N., Dossa, G. S., Mauleon, R., Cruz, C. V., & Oliva, R. (2015b). Rice-infecting Pseudomonas genomes are highly accessorized and harbor multiple putative virulence mechanisms to cause sheath brown rot. PLoS ONE, 10(9), e0139256. doi:10.1371/journal.pone.0139256 Radchuk, V., & Borisjuk, L. (2014). Physical, metabolic and developmental functions of the seed coat. Frontiers in Plant Science, 5, 510. doi:10.3389/fpls.2014.00510 Rahman, S. F. S. A., Sijam, K., & Omar, D. (2014). Chemical composition of Piper sarmentosum extracts and antibacterial activity against the plant pathogenic bacteria Pseudomonas fuscovaginae and Xanthomonas oryzae pv. oryzae. Journal

249

of Plant Diseases and Protection, 121(6), 237-242. Retrieved from http://www.jstor.org/stable/24618863 Rath, D., Amlinger, L., Rath, A., & Lundgren, M. (2015). The CRISPR-Cas immune system: biology, mechanisms and applications. Biochimie, 117(Supplement C), 119-128. doi:https://doi.org/10.1016/j.biochi.2015.03.025 Razak, A., Zainudin, N., Sidiqe, S., Ismail, N., Mohamad, N., & Salleh, B. (2009a). Sheath Brown Rot Disease of Rice Caused by Pseudomonas Fuscovaginae in the Peninsular Malaysia Journal of Plant Protection Research (Vol. 49, pp. 244). Razak, A., Zainudin, N., Sidiqe, S., Ismail, N., Mohamad, N., & Salleh, B. (2009b). Sheath brown rot disease of rice caused by Pseudomonas fuscovaginae in the peninsular Malaysia. Journal of Plant Protection Research, 49(3), 244. doi:10.2478/v10045-009-0037-x Reddy, K. R. N., Choudary, K. A., & Reddy, M. S. (2007). Antifungal metabolites of Pseudomonas fluorescens isolated from rhizosphere of rice crop. Journal of Mycology and Plant Pathology, 37(2). Renouf, V., Claisse, O., & Lonvaud-Funel, A. (2005). Understanding the microbial ecosystem on the grape berry surface through numeration and identification of yeast and bacteria. Australian Journal of Grape and Wine Research, 11(3), 316-327. doi:10.1111/j.1755-0238.2005.tb00031.x Rezaei, R., & Taghavi, S. M. (2014). Host specificity, pathogenicity and the presence of virulence genes in Iranian strains of Pseudomonas syringae pv. syringae from different hosts. Archives of Phytopathology and Plant Protection, 47(19), 2377- 2391. doi:10.1080/03235408.2013.877192 RGA. (2017). About the rice industry. Retrieved from http://www.rga.org.au/the-rice- industry.aspx Ricepedia. (2017). The global staple. Retrieved from http://ricepedia.org/rice-as- food/the-global-staple-rice-consumers Riera-Ruiz, C., Vargas, J., Cedeño, C., Quirola, P., Escobar, M., Cevallos-Cevallos, J. M., Ratti, M., & Peralta, E. L. (2014). First report of Burkholderia glumae causing bacterial panicle blight on rice in Ecuador. Plant Disease, 98(7), 988-988. doi:10.1094/PDIS-10-13-1024-PDN Rivero-González, D., Corzo, M., Plasencia, O., Cruz, A., Martínez, B., & Martínez, Y. (2017). Characterization and diagnosis of Pseudomonas fuscovaginae Miyajima,

250

Tanii and Akita, causal agent of the brown sheath rot in rice. Biotecnologia Aplicada, 34(2), 2101-2108. Retrieved from http://ezproxy.csu.edu.au/login?url=http://search.ebscohost.com/login.aspx?direct =true&db=a9h&AN=125103547&site=ehost-live Rogers, E. (2015, 19 May 2015 at 2:39 pm). Queensland rice growers join national association with hopes to build export market. ABC Rural. Retrieved from http://www.abc.net.au/news/rural/2015-05-19/queensland-rice-growth/6480682 Roine, E., Raineri, D. M., Romantschuk, M., Wilson, M., & Nunn, D. N. (1998). Characterization of type IV pilus genes in Pseudomonas syringae pv. tomato DC3000. Molecular Plant-Microbe Interactions, 11(11), 1048-1056. doi:10.1094/mpmi.1998.11.11.1048 Rojas, C. M., Ham, J. H., Deng, W.-L., Doyle, J. J., & Collmer, A. (2002). HecA, a member of a class of adhesins produced by diverse pathogenic bacteria, contributes to the attachment, aggregation, epidermal cell killing, and virulence phenotypes of Erwinia chrysanthemi EC16 on Nicotiana clevelandii seedlings. Proceedings of the National Academy of Sciences, 99(20), 13142-13147. doi:10.1073/pnas.202358699 Romantschuk, M. (1992). Attachment of plant pathogenic bacteria to plant surfaces. Annual Review of Phytopathology, 30(1), 225-243. doi:doi:10.1146/annurev.py.30.090192.001301 Romantschuk, M., & Bamford, D. H. (1986). The causal agent of halo blight in bean, Pseudomonas syringae pv. phaseolicola, attaches to stomata via its pili. Microbial Pathogenesis, 1(2), 139-148. Rosas, J. E., Martínez, S., Bonnecarrère, V., Pérez de Vida, F., Blanco, P., Malosetti Zunin, M., Jannink, J. L., & Gutiérrez, L. (2016). Comparison of phenotyping methods for resistance to stem rot and aggregated sheath spot in rice. Crop Science, 56(4), 1619-1627. Retrieved from http://edepot.wur.nl/386266 Rostami, M., Ghasemi, A. G., Rahimian, H., & Khosravi, V. (2010). Bacterial sheath rot of rice in Mazandaran province Iranian Journal of Plant Pathology, 45(3), 61 - 63. Rott, P., Honegger, J., Notteghem, J. L., & Ranomenjanahary, S. (1991). Identification of Pseudomonas fuscovaginae with biochemical, serological, and pathogenicity tests. Plant Disease, 75(8), 843-846. Rott, P., Notteghem, J. L., & Frossard, P. (1989). Identification and characterization of Psuedomonas fuscovaginae, the causal agent of bacterial sheath brown rot of rice,

251

from Madagascar and other countries. Plant Disease, 73(2), 133-137. doi:10.1094/pd-73-0133 Saberi, E., Safaie, N., & Rahimian, H. (2013). Characterization of bacterial strains associated with sheath rot complex and grain discoloration of rice in north of Iran (Mazandaran province). Journal of Bacteriology Research, 5(5), 51-61. doi:10.5897/JBR2013.0115 Saha, S., Garg, R., Biswas, A., & Rai, A. B. (2015). Bacterial diseases of rice: an overview. Journal of Pure and Applied Microbiology, 9(1), 12. Retrieved from https://www.researchgate.net/publication/282209547_Bacterial_diseases_of_rice_ An_overview Sahebi, M., Hanafi, M. M., Akmar, A. S. N., Rafii, M. Y., Azizi, P., Tengoua, F. F., Azwa, J. N. M., & Shabanimofrad, M. (2015). Importance of silicon and mechanisms of biosilica formation in plants. BioMed Research International, 2015, 396010. doi:10.1155/2015/396010 Sakthivel, N. (2001). Sheath rot disease of rice: current status and control strategies Major fungal diseases of rice: recent advances (pp. 271-283). Dordrecht: Springer Netherlands. Sakthivel, N., & Gnanamanickam, S. S. (1987). Evaluation of Pseudomonas fluorescens for suppression of sheath rot disease and for enhancement of grain yields in rice (Oryza sativa L.). Applied and Environmental Microbiology, 53(9), 2056 - 2059. Retrieved from http://aem.asm.org/content/53/9/2056.full.pdf Sambrook, J., Fritsch, E. F., & Maniatis, T. (1989). Molecular Cloning (Vol. 2): Cold Spring Harbor Laboratory Press New York. Samson, R., Legendre, J. B., Christen, R., Fischer-Le Saux, M., Achouak, W., & Gardan, L. (2005). Transfer of Pectobacterium chrysanthemi (Burkholder et al. 1953) Brenner et al. 1973 and Brenneria paradisiaca to the genus Dickeya gen. nov. as Dickeya chrysanthemi comb. nov. and Dickeya paradisiaca comb. nov. and delineation of four novel species, Dickeya dadantii sp. nov., Dickeya dianthicola sp. nov., Dickeya dieffenbachiae sp. nov. and Dickeya zeae sp. nov. International Journal of Systematic and Evolutionary Microbiology, 55(Pt 4), 1415-1427. doi:10.1099/ijs.0.02791-0

252

Sarkar, S. F., Gordon, J. S., Martin, G. B., & Guttman, D. S. (2006). Comparative genomics of host-specific virulence in Pseudomonas syringae. Genetics, 174(2), 1041-1056. doi:10.1534/genetics.106.060996 Sayler, R. J., Cartwright, R. D., & Yang, Y. (2006). Genetic characterization and real- time PCR detection of Burkholderia glumae, a newly emerging bacterial pathogen of rice in the United States. Plant Disease, 90, 603 - 610. Scarlett, C. M., Fletcher, J. T., Roberts, P., & Lelliott, R. A. (1978). Tomato pith necrosis caused by Pseudomonas corrugata n. sp. Annals of Applied Biology, 88(1), 105- 114. doi:10.1111/j.1744-7348.1978.tb00684.x Schaad, N. W., Postnikova, E., Sechler, A., Claflin, L. E., Vidaver, A. K., Jones, J. B., Agarkova, I., Ignatov, A., Dickstein, E., & Ramundo, B. A. (2008). Reclassification of subspecies of Acidovorax avenae as A. avenae (Manns 1905) emend., A. cattleyae (Pavarino, 1911) comb. nov., A. citrulli Schaad et al., 1978) comb. nov., and proposal of A. oryzae sp. nov. Systematic and Applied Microbiology, 31(6–8), 434-446. doi:http://dx.doi.org/10.1016/j.syapm.2008.09.003 Scholz-Schroeder, B. K., Hutchison, M. L., Grgurina, I., & Gross, D. C. (2001). The contribution of syringopeptin and syringomycin to virulence of Pseudomonas syringae pv. syringae strain B301D on the basis of sypA and syrB1 biosynthesis mutant analysis. Molecular Plant-Microbe Interactions, 14(3), 336-348. doi:10.1094/MPMI.2001.14.3.336 Scholz-Schroeder, B. K., Soule, J. D., & Gross, D. C. (2003). The sypA, sypB, and sypC synthetase genes encode twenty-two modules involved in the nonribosomal peptide synthesis of syringopeptin by Pseudomonas syringae pv. syringae B301D. Molecular Plant-Microbe Interactions 16(4), 271-280. doi:10.1094/mpmi.2003.16.4.271 Seo, Y.-S., Lim, J. Y., Park, J., Kim, S., Lee, H.-H., Cheong, H., Kim, S.-M., Moon, J. S., & Hwang, I. (2015). Comparative genome analysis of rice-pathogenic Burkholderia provides insight into capacity to adapt to different environments and hosts. BMC Genomics, 16(1), 1-11. doi:10.1186/s12864-015-1558-5 Serra, M. D., Fagiuoli, G., Nordera, P., Bernhart, I., Volpe, C. D., Di Giorgio, D., Ballio, A., & Menestrina, G. (1999). The interaction of lipodepsipeptide toxins from Pseudomonas syringae pv. syringae with biological and model membranes: a

253

comparison of syringotoxin, syringomycin, and two syringopeptins. Molecular Plant-Microbe Interactions, 12(5), 391-400. doi:10.1094/MPMI.1999.12.5.391 Setati, M. E., Jacobson, D., Andong, U.-C., & Bauer, F. (2012). The vineyard yeast microbiome, a mixed model microbial map. PLoS ONE, 7(12), e52609. doi:10.1371/journal.pone.0052609 Shakya, D. D., & Chung, H. S. (1983). Detection of Pseudomonas avenae in rice seed. Seed Science and Technology, 11(3), 583-587. Shakya, D. D., Chung, H. S., & Vinther, F. (1986). Transmission of Pseudomonas avenae, the cause of bacterial stripe of rice. Journal of Phytopathology, 116(1), 92- 96. doi:10.1111/j.1439-0434.1986.tb00899.x Shakya, D. D., Vinther, F., & Mathur, S. B. (1985). World wide distribution of a bacterial stripe pathogen of rice identified as Pseudomonas avenae. Journal of Phytopathology, 114(3), 256-259. doi:10.1111/j.1439-0434.1985.tb00850.x Shapiro, R. S., Chavez, A., Porter, C. B. M., Hamblin, M., Kaas, C. S., DiCarlo, J. E., Zeng, G., Xu, X., Revtovich, A. V., Kirienko, N. V., Wang, Y., Church, G. M., & Collins, J. J. (2017). A CRISPR–Cas9-based gene drive platform for genetic interaction analysis in Candida albicans. Nature Microbiology. doi:10.1038/s41564-017-0043-0 Sharma, S., Sthapit, B. R., Pradhanang, P. M., & Joshi, K. D. (1996). Bacterial sheath brown rot of rice caused by Pseudomonas fuscovaginae in Nepal. Paper presented at the Rice cultivation in highland areas, Proceedings of the CIRAD conference held at Antananarivo, Madagascar, 29 March-5 April 1996., Antananarivo, Madagascar. Sheibani-Tezerji, R., Naveed, M., Jehl, M.-A., Sessitsch, A., Rattei, T., & Mitter, B. (2015). The genomes of closely related Pantoea ananatis maize seed endophytes having different effects on the host plant differ in secretion system genes and mobile genetic elements. Frontiers in Microbiology, 6(440). doi:10.3389/fmicb.2015.00440 Sherrington, M. (2015, 3 Dec 2015, 1 p.m.). SunRice's NQ operation succeeding. Retrieved from http://www.northqueenslandregister.com.au/story/3532472/nq- rice-boom-continues/ Shivalingaiah, Umesha, S., & Sateesh, M. K. (2013). Cocculus hirsutus extract inhibits the Xanthomonas oryzae pv. oryzae, the bacterial leaf blight pathogen in rice.

254

Archives of Phytopathology and Plant Protection, 46(15), 1885-1894. doi:10.1080/03235408.2013.781354 Shrestha, B. K., Karki, H. S., Groth, D. E., Jungkhun, N., & Ham, J. H. (2016). Biological control activities of rice-associated Bacillus sp. strains against sheath blight and bacterial panicle blight of rice. PLoS ONE, 11(1), e0146764. doi:10.1371/journal.pone.0146764 Silby, M. W., Cerdeno-Tarraga, A. M., Vernikos, G. S., Giddens, S. R., Jackson, R. W., Preston, G. M., Zhang, X. X., Moon, C. D., Gehrig, S. M., Godfrey, S. A., Knight, C. G., Malone, J. G., Robinson, Z., Spiers, A. J., Harris, S., Challis, G. L., Yaxley, A. M., Harris, D., Seeger, K., Murphy, L., Rutter, S., Squares, R., Quail, M. A., Saunders, E., Mavromatis, K., Brettin, T. S., Bentley, S. D., Hothersall, J., Stephens, E., Thomas, C. M., Parkhill, J., Levy, S. B., Rainey, P. B., & Thomson, N. R. (2009). Genomic and genetic analyses of diversity and plant interactions of Pseudomonas fluorescens. Genome Biology 10(5), R51. doi:10.1186/gb-2009-10- 5-r51 Silby, M. W., Winstanley, C., Godfrey, S. A. C., Levy, S. B., & Jackson, R. W. (2011). Pseudomonas genomes: diverse and adaptable. Fems Microbiology Reviews, 35(4), 652-680. doi:10.1111/j.1574-6976.2011.00269.x Sivapalan, S. (2014). Research in the Ord has found ways to improve rice yields. Australian Grain, 5. Sivapalan, S. (2017, Wednesday, 8 March 2017 - 2:14pm). Rice in the Ord river irrigation area. Retrieved from https://www.agric.wa.gov.au/grains-research- development/rice-ord-river-irrigation-area Skerman, V. B. D., McGowan, V., & Sneath, P. H. A. (1980). Approved lists of bacterial names. International Journal of Systematic and Evolutionary Microbiology, 30(1), 225-420. doi:doi:10.1099/00207713-30-1-225 Smit, G., Logman, T. J., Boerrigter, M. E., Kijne, J. W., & Lugtenberg, B. J. (1989). Purification and partial characterization of the Rhizobium leguminosarum biovar viciae Ca2+-dependent adhesin, which mediates the first step in attachment of cells of the family Rhizobiaceae to plant root hair tips. Journal of Bacteriology, 171(7), 4054-4062. doi:10.1128/jb.171.7.4054-4062.1989

255

Smith, D. J., & Humphreys, E. (2001). The benefits of winter crops after rice harvest part 2: models to predict what will happen in your situation. Farmers' Newsletter Large Area(157), 36-38. Somda, I., Veena, M. S., & Mortensen, C. N. (2001). First report on the occurrence of bacterial stripe organism Acidovorax avenae subsp. avenae in rice seeds from Burkina Faso. Plant Disease, 85(7), 804-804. doi:10.1094/pdis.2001.85.7.804a Song, W. Y., Kim, H. M., Hwang, C. Y., & Schaad, N. W. (2004). Detection of Acidovorax avenae ssp. avenae in rice seeds using BIO-PCR. Journal of Phytopathology, 152(11-12), 667-676. doi:10.1111/j.1439-0434.2004.00914.x Soto, G. E., & Hultgren, S. J. (1999). Bacterial adhesins: common themes and variations in architecture and assembly. Journal of Bacteriology, 181(4), 1059-1071. Retrieved from http://jb.asm.org/content/181/4/1059.short Sparkes, D. (2015, 13 November 2015 at 12:32 pm). Rice breeder focuses on new, high- value varieties for Mackay and north Queensland. QLD Country hour. Rural Retrieved from http://www.abc.net.au/news/rural/2015-11-13/rice-breeder-works- on-new-varieties-for-north-queensland/6938280 Statista. (2017). Rice - Statistics & Facts. Retrieved from https://www.statista.com/topics/1443/rice/ Stodart, B. (2014, 2/17/2014). [QS in Aust isolates]. Stodart, B. J., Pattermore, J. P., & Ash, G. J. (2013). The Pseudomonas fuscovaginae DAR 77795 whole genome shotgun (WGS) project (Accession: BATD00000000). Genome sequences for strains of plant pathogenic bacteria from rice in Australia. Genome assembly data. Agricultural and Wine Sciences. Charles Sturt University. Stones, D. H., & Krachler, A.-M. (2015). Fatal attraction: how bacterial adhesins affect host signaling and what we can learn from them. International Journal of Molecular Sciences, 16(2), 2626-2640. doi:10.3390/ijms16022626 Stones, D. H., & Krachler, A. M. (2016). Against the tide: the role of bacterial adhesion in host colonization. Biochemical Society Transactions, 44(6), 1571-1580. doi:10.1042/BST20160186 Suksomtip, M., & Tungpradabkul, S. (2005). An easy method for generating deletion mutants in Agrobacterium tumefaciens using a simple replacement vector. ScienceAsia, 31, 349-357. doi:10.2306/scienceasia1513-1874.2005.31.349

256

Sundin, G. W., Castiblanco, L. F., Yuan, X., Zeng, Q., & Yang, C.-H. (2016). Bacterial disease management: challenges, experience, innovation and future prospects. Molecular Plant Pathology, 17(9), 1506-1518. doi:10.1111/mpp.12436 Sundström, J. F., Albihn, A., Boqvist, S., Ljungvall, K., Marstorp, H., Martiin, C., Nyberg, K., Vågsholm, I., Yuen, J., & Magnusson, U. (2014). Future threats to agricultural food production posed by environmental degradation, climate change, and animal and plant diseases – a risk analysis in three economic and climate settings. Food Security, 6(2), 201-215. doi:10.1007/s12571-014-0331-y Suoniemi, A., Björklöf, K., Haahtela, K., & Romantschuk, M. (1995). Pili of Pseudomonas syringae pathovar syringae enhance initiation of bacterial epiphytic colonization of bean. Microbiology, 141(2), 497-503. doi:doi:10.1099/13500872- 141-2-497 Swings, J., van den Mooter, M., Vauterin, L., Hoste, B., Gillis, M., Mew, T. W., & Kersters, K. (1990). Reclassification of the causal agents of bacterial blight (Xanthomonas campestris pv. oryzae) and bacterial leaf streak (Xanthomonas campestris pv. oryzicola) of rice as pathovars of Xanthomonas oryzae (ex Ishiyama 1922) sp. nov., nom. rev. International Journal of Systematic and Evolutionary Microbiology, 40(3), 309-311. doi:doi:10.1099/00207713-40-3-309 Taghipour, M., Rahimian, H., & Babaeazad, V. (2015). Genetic diversity of strains of Acidovorax oryzae causal agent of brown stripe of rice in Mazandaran province. Iranian Journal of Plant Pathology, 51(1), Pe12-Pe25, En11. Taguchi, F., & Ichinose, Y. (2011). Role of type IV pili in virulence of Pseudomonas syringae pv. tabaci 6605: correlation of motility, multidrug resistance, and HR- inducing activity on a nonhost plant. Molecular Plant-Microbe Interactions, 24(9), 1001-1011. doi:10.1094/mpmi-02-11-0026 Takai, R., Isogai, A., Takayama, S., & Che, F.-S. (2008). Analysis of flagellin perception mediated by flg22 receptor OsFLS2 in rice. Molecular Plant-Microbe Interactions, 21(12), 1635-1642. doi:10.1094/mpmi-21-12-1635 Tanaka, E., Ashizawa, T., Sonoda, R., & Tanaka, C. (2008). Villosiclava virens gen. nov., comb. nov., teleomorph of Ustilaginoidea virens, the causal agent of rice false smut. Mycotaxon, 106, 491-501. Tanaka, N., Che, F.-S., Watanabe, N., Fujiwara, S., Takayama, S., & Isogai, A. (2003).

Flagellin from an incompatible strain of Acidovorax avenae mediates H2O2

257

generation accompanying hypersensitive cell death and expression of PAL, Cht-1, and PBZ1, but not of LOX in rice. Molecular Plant-Microbe Interactions, 16(5), 422-428. doi:10.1094/mpmi.2003.16.5.422 Tanii, A., Miyajima, K., & Akita, T. (1976). The sheath brown rot disease of rice plant and its causal bacterium, Pseudomonas fuscovaginae A. Tanii, K. Miyajima & T. Akita sp. nov. Annals of the Phytopathological Society of Japan, 42(5), 540-548. Taunous, I. D., Kempf, D., & Duarte, V. (1997). Virulence of isolates of fluorescent pseudomonads on irrigated rice cultivars. Fitopatologia Brasileira, 22(4), 525 -528. Thompson, J. A. (2002). Water management of rice in southern New South Wales, Australia. In Bouman, B. A. M., Hengsdijk, H., Hardy, B., Bindraban, P. S., Tuong, T. P., & Ladha, J. K. (Eds.), Water-wise rice production. Proceedings of the International Workshop on Water-wise Rice Production, 8-11 April 2002, Los Baños, Philippines (pp. 356). Los Baños (Philippines): International Rice Research Institute. Tilquin, J. P., & Detry, J. F. (1993). Efficiency of natural selection for bacterial sheath rot (BSR) in bulked families. International Rice Research Notes, 18(1), 23-24. Tindall, B. J. (2014). The combination Enterobacter agglomerans is to be cited as Enterobacter agglomerans (Beijerinck 1888) Ewing and Fife 1972 and the combination Pantoea agglomerans is to be cited as Pantoea agglomerans (Beijerinck 1888) Gavini et al. 1989. Opinion 90. Judicial Commission of the International Committee on Systematics of Prokaryotes. International Journal of sSystematic and Evolutionary Microbiology, 64(Pt 10), 3582-3583. doi:10.1099/ijs.0.069161-0 Trivedi, P., Trivedi, C., Grinyer, J., Anderson, I. C., & Singh, B. K. (2016). Harnessing host-vector microbiome for sustainable plant disease management of phloem- limited bacteria. Frontiers in Plant Science, 7(1423). doi:10.3389/fpls.2016.01423 Troldahl, D., Snell, P., Dunn, B., Ovenden, B., & Pallas, L. (2016). Rice variety guide 2016 – 17 (6 ed., Vol. Primefact 1112): NSW Department of Primary Industries. Tsuge, T., Harimoto, Y., Akimitsu, K., Ohtani, K., Kodama, M., Akagi, Y., Egusa, M., Yamamoto, M., & Otani, H. (2013). Host-selective toxins produced by the plant pathogenic fungus Alternaria alternata. FEMS Microbiology Reviews 37(1), 44-66. doi:10.1111/j.1574-6976.2012.00350.x

258

Tsushima, S. (1996). Epidemiology of bacterial grain rot of rice caused by Pseudomonas glumae. Japan Agricultural Research Quarterly, 30, 85 - 89. Tsushima, S., Mogi, S., Naito, H., & Saito, H. (1991). Populations of Pseudomonas glumae on rice plants. Annals of the Phytopathological Society of Japan, 57(2), 145 -152. doi:10.3186/jjphytopath.57.145 Tsushima, S., Mogi, S., & Saito, H. (1986). Effect of temperature on the growth of Pseudomonas glumae and the development of rice bacterial grain rot. Proceedings of the Association for Plant Protection of Kyushu, 32, 14 - 16. Tsushima, S., & Naito, H. (1991). Spatial distribution and dissemination of bacterial grain rot of rice caused by Pseudomonas glumae. Annals of the Phytopathological Society of Japan, 57(2), 180 - 187. doi:10.3186/jjphytopath.57.180 Tsushima, S., Naito, H., & Koitabashi, M. (1995). Change in panicle susceptibility associated with flowering rate of spikelets in bacterial grain rot of rice caused by Pseudomonas glumae. Annals of the Phytopathological Society of Japan, 61, 109 - 113. TullÍs, E. C., & Johnson, A. G. (1952). Synonymy of Tilletia hórrida and Neovossia barclayana. Mycologia, 44(6), 773-788. Uchino, M., Shida, O., Uchimura, T., & Komagata, K. (2001). Recharacterization of Pseudomonas fulva Iizuka and Komagata 1963, and proposals of Pseudomonas parafulva sp. nov. and Pseudomonas cremoricolorata sp. nov. The Journal of General and Applied Microbiology 47(5), 247-261. Uematsu, T., Yoshimura, D., Nishiyama, K., Ibaraki, T., & Fujii, H. (1976). Occurrence of bacterial seedling rot in nursery flat, caused by grain rot bacterium Pseudomonas glumae. Annals of the Phytopathological Society of Japan, 42, 310 - 312. Ura, H., Furuya, N., Iiyama, K., Hidaka, M., Tsuchiya, K., & Matsuyama, N. (2006). Burkholderia gladioli associated with symptoms of bacterial grain rot and leaf- sheath browning of rice plants. Journal of General Plant Pathology, 72(2), 98-103. doi:10.1007/s10327-005-0256-6 Urakami, T., Ito-Yoshida, C., Araki, H., Kijima, T., Suzuki, K.-I., & Komagata, K. (1994). Transfer of Pseudomonas plantarii and Pseudomonas glumae to Burkholderia as Burkholderia spp. and description of Burkholderia vandii sp. nov. International Journal of Systematic and Evolutionary Microbiology, 44(2), 235- 245. doi:doi:10.1099/00207713-44-2-235

259

Uzelac, G., Patel, H. K., Devescovi, G., Licastro, D., & Venturi, V. (2017). Quorum sensing and RsaM regulons of the rice pathogen Pseudomonas fuscovaginae. Microbiology, 163(5), 765-777. Retrieved from http://www.microbiologyresearch.org/docserver/fulltext/micro/163/5/765_micro0 00454.pdf?expires=1536465981&id=id&accname=guest&checksum=978C7B6B 4F3FB2E39637DE8A086EF9A6 Valencia-Botin, A. J., Mendoza-Onofre, L. E., Silva-Rojas, H. V., Cordova-Tellez, L., Espinosa-Victoria, D., Valadez-Moctezuma, E., & Villasenor-Mir, H. E. (2007). Aggressiveness estimates and inoculation methods of pathogenic bacteria on seeds and seedlings of wheat 'seri M82'. Revista Fitotecnia Mexicana, 30(3), 255-259. van Doorn, J., Boonekamp, P. M., & Oudega, B. (1994). Partial characterization of fimbriae of Xanthomonas campestris pv. hyacinthi. Molecular Plant-Microbe Interactions 7(3), 334-344. Van Nguyen, N., & Ferrero, A. (2006). Meeting the challenges of global rice production. Paddy and Water Environment, 4(1), 1-9. van Ulsen, P., ur Rahman, S., Jong, W. S. P., Daleke-Schermerhorn, M. H., & Luirink, J. (2014). Type V secretion: from biogenesis to biotechnology. Biochimica et Biophysica Acta, 1843(8), 1592-1611. doi:https://doi.org/10.1016/j.bbamcr.2013.11.006 Venturi, V. (2006). Regulation of quorum sensing in Pseudomonas. FEMS Microbiology Reviews, 30(2), 274-291. doi:10.1111/j.1574-6976.2005.00012.x Venturi, V., & Fuqua, C. (2013). Chemical signaling between plants and plant-pathogenic bacteria. Annual Review of Phytopathology, 51(1), 17-37. doi:10.1146/annurev- phyto-082712-102239 Vera Cruz, C., & Opulencia, R. (2013). Bacterial sheath brown rot. Retrieved from http://www.knowledgebank.irri.org/decision-tools/rice-doctor/rice-doctor-fact- sheets/item/bacterial-sheath-brown-rot?category_id=345 Vesper, S. J., & Bauer, W. D. (1986). Role of pili (fimbriae) in attachment of Bradyrhizobium japonicum to soybean roots. Applied and Environmental Microbiology, 52(1), 134-141. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC203409/

260

von Bodman, S. B., Bauer, W. D., & Coplin, D. L. (2003). Quorum sensing in plant- pathogenic bacteria. Annual Review of Phytopathology, 41, 455-482. doi:10.1146/annurev.phyto.41.052002.095652 von Wintersdorff, C. J. H., Penders, J., van Niekerk, J. M., Mills, N. D., Majumder, S., van Alphen, L. B., Savelkoul, P. H. M., & Wolffs, P. F. G. (2016). Dissemination of antimicrobial resistance in microbial ecosystems through horizontal gene transfer. Frontiers in Microbiology, 7, 173. doi:10.3389/fmicb.2016.00173 Wang, N., Lu, S.-E., Yang, Q., Sze, S.-H., & Gross, D. C. (2006). Identification of the syr-syp box in the promoter regions of genes dedicated to syringomycin and syringopeptin production by Pseudomonas syringae pv. syringae B301D. Journal of Bacteriology, 188(1), 160-168. doi:10.1128/JB.188.1.160-168.2006 Wang, S., Boevink, P. C., Welsh, L., Zhang, R., Whisson, S. C., & Birch, P. R. J. (2017). Delivery of cytoplasmic and apoplastic effectors from Phytophthora infestans haustoria by distinct secretion pathways. The New Phytologist, 216(1), 205-215. doi:10.1111/nph.14696 Watson, A. (2013). [Disease surveys on rice]. Webster, L. (2016, 13 April 2016 at 1:42 pm). More rice varieties starting to show promise for the tropical north. QLD Country Hour. Retrieved from http://www.abc.net.au/news/rural/2016-04-13/mackay-rice-trials-showing-more- promise/7322374 Webster, R. K., & Gunnell, P. S. (1992). Compendium of rice diseases. St. Paul, MN.: APS Press. Weinberg, J. B., Alexander, B. D., Majure, J. M., Williams, L. W., Kim, J. Y., Vandamme, P., & LiPuma, J. J. (2007). Burkholderia glumae infection in an infant with chronic granulomatous disease. Journal of Clinical Microbiology, 45(2), 662- 665. doi:10.1128/jcm.02058-06 Weller-Stuart, T., De Maayer, P., & Coutinho, T. (2017). Pantoea ananatis: genomic insights into a versatile pathogen. Molecular Plant Pathology, 18(9), 1191-1198. doi:10.1111/mpp.12517 Whitworth, R., Lattimore, M.-A., Beecher, G., Dunn, B., Warden, L., O'Keeffe, K., Fowler, J., Muirhead, T., & Troldahl, D. (2013). Rice field guide - to pests , diseases and weeds in southern New South Wales: Department of Trade and Investment, Regional Infrastructure and Services, NSW, Australia.

261

Williams, J., Hook, R. A., & Hamblin, A. (2002). Agro-Ecological Regions of Australia: methodologies for their derivation and key issues in resource management. Canberra: CSIRO Land and Water. Williams, R. L., & Angus, J. F. (1994). Deep floodwater protects high-nitrogen rice crops from low-temperature damage. Australian Journal of Experimental Agriculture, 34(7), 927-932. doi:https://doi.org/10.1071/EA9940927 Xie, G. (2003). First report of sheath brown rot of rice in China and characterization of the causal organism by phenotypic tests and biolog. International Rice Research Notes, 28(1), 50 - 52. Retrieved from http://ejournals.ph/index.php?journal=IRRN&page=article&op=view&path%5B %5D=4170&path%5B%5D=4366

Xie, G., Cui, Z., Tao, Z., Qiu, H., Liu, H., Ibrahim, M., Zhu, B., Jin, G., Sun, G., Almoneafy, A., & Li, B. (2012). Genome sequence of the rice pathogen Pseudomonas fuscovaginae CB98818. Journal of Bacteriology, 194(19), 5479- 5480. doi:10.1128/jb.01273-12 Xie, G., Sun, X. L., & Mew, T. W. (1998). Characterization of Acidovorax avenae subsp. avenae from rice seeds. Chinese Journal of Rice Science, 12(3), 165 - 171. Yan, H., Yu, S. H., Xie, G. L., Fang, W., Su, T., & Li, B. (2010). Grain discoloration of rice caused by Pantoea ananatis (synonym Erwinia uredovora) in China. Plant Disease, 94(4), 482-482. doi:10.1094/pdis-94-4-0482b Yoon, M.-Y., Cha, B., & Kim, J.-C. (2013). Recent trends in studies on botanical fungicides in agriculture. The Plant Pathology Journal, 29(1), 1-9. doi:10.5423/PPJ.RW.05.2012.0072 Yuan, X. (2004). Identification of bacterial pathogens causing panicle blight of rice in Louisiana. (Master of Science), Louisiana State University, Baton Rouge, LA. Yue, M., Han, X., de Masi, L., Zhu, C., Ma, X., Zhang, J., Wu, R., Schmieder, R., Kaushik, R. S., Fraser, G. P., Zhao, S., McDermott, P. F., Weill, F.-X., Mainil, J. G., Arze, C., Fricke, W. F., Edwards, R. A., Brisson, D., Zhang, N. R., Rankin, S. C., & Schifferli, D. M. (2015). Allelic variation contributes to bacterial host specificity. Nature Communications, 6, 8754. doi:10.1038/ncomms9754 Załuga, J., Stragier, P., Baeyen, S., Haegeman, A., van Vaerenbergh, J., Maes, M., & de Vos, P. (2014). Comparative genome analysis of pathogenic and non-pathogenic

262

Clavibacter strains reveals adaptations to their lifestyle. BMC Genomics, 15(1), 1- 14. doi:10.1186/1471-2164-15-392 Zeigler, R. S., & Alvarez, E. (1986). Bacterial sheath brown rot (BSBR) in Latin America. International Rice Research Newsletter, 11(5), 15 -16. Zeigler, R. S., & Alvarez, E. (1987). Bacterial sheath brown rot of rice caused by Pseudomonas fuscovaginae in Latin America. Plant Disease, 71(7), 592-597. Zeigler, R. S., & Alvarez, E. (1989). Grain discoloration of rice caused by Pseudomonas glumae in Latin America. Plant Disease, 73(4), 368. doi:10.1094/PD-73-0368B Zeigler, R. S., & Alvarez, E. (1990). Characteristics of Pseudomonas spp. causing grain discoloration and sheath rot of rice and associated Pseudomonad epiphytes. Plant Disease, 74(11), 917 - 922. doi:10.1094/PD-74-0917 Zeigler, R. S., Aricapa, G., & Hoyos, E. (1987). Distribution of fluorescent Pseudomonas spp. causing grain and sheath discoloration of rice in Latin America. Plant Disease, 71, 894 - 900. doi:10.1094/PD-71-0896. Zeng, W., Melotto, M., & He, S. Y. (2010). Plant stomata: a checkpoint of host immunity and pathogen virulence. Current Opinion in Biotechnology, 21(5), 599-603. doi:http://dx.doi.org/10.1016/j.copbio.2010.05.006 Zhang, G. L., Dai, Q. G., & Zhang, H. C. (2006). Silicon application enhances resistance to sheath blight (Rhizoctonia solani) in rice. Journal of Plant Physiology and Molecular Biology, 32(5), 600-606. Zhang, H., & Wang, S. (2013). Rice versus Xanthomonas oryzae pv. oryzae: a unique pathosystem. Current Opinion in Plant Biology, 16(2), 188-195. doi:https://doi.org/10.1016/j.pbi.2013.02.008 Zhang, Z., Schwartz, S., Wagner, L., & Miller, W. (2000). A greedy algorithm for aligning DNA sequences. Journal of Computational Biology, 7(1-2), 203-214. doi:10.1089/10665270050081478 Zhou-qi, C., Bo, Z., Guan-lin, X., Bin, L., & Shi-wen, H. (2016). Research status and prospect of Burkholderia glumae, the pathogen causing bacterial panicle blight. Rice Science, 23(3), 111-118. doi:https://doi.org/10.1016/j.rsci.2016.01.007 Zhu, B., Lou, M.-m., Huai, Y., Xie, G., Luo, J.-y., & Xu, L.-h. (2008). Isolation and identification of Burkholderia glumae from symptomless rice seeds. Rice Science, 15(2), 145-149. doi:http://dx.doi.org/10.1016/S1672-6308(08)60033-5

263

Zia-ul-Hussnain, S., Haque, M. I., Mughal, S. M., Shah, K. N., Irfan, A., Afghan, S., Shahazad, A., Batool, A., Khanum, P., Hussain, K., Nawaz, K., Hassan, M. N., & Hafeez, F. Y. (2011). Isolation and biochemical characterizations of the bacteria (Acidovorax avenae subsp. avenae) associated with red stripe disease of sugarcane. African Journal of Biotechnology, 10(37), 7191 -7197.

264

8. Appendix Profiling a Disease Microbiome

Methods and Materials

I. Sampling

Wheat plants (cv. Merinda) were sampled from farm 731 in Leeton at early maturity.

Seven to ten sampling points were chosen from a block of approximately 450 x 400m.

Two areas in the field that were separated by a large earthen bund were designated as 1a and 1b. In the area designated as 1a, the crop was cultivated on partially burnt rice stubble. Symptoms similar to sheath brown rot i.e. brown coloured lesions on glume, lesions on leaf sheath and stem were observed. It was estimated that 80% of heads displaying brown lesions were associated with prominent stem and sheath lesions. The plants with symptoms were well spread throughout the block, with an incidence of ~5%.

In the area designated as 1b, the crop was grown on fallow, with a recent history of rice.

There, the symptoms were rare with 1-2 spikelets with brown lesion appearing occasionally, throughout the block. Glume lesions are not always found with stem lesions. Affected plants were less than 1% of the total crop. At each sampling point where a sample with symptoms was taken, another sample without symptoms was also taken. The whole plant was detached from soil aseptically and then put into clean room manufactured zip-lock bags, and kept on ice until taken to the lab. In the lab, the samples were maintained at 4 °C until processed.

II. Sample preparation

From each sample, the heads with the flag leaf and the leaf sheath was aseptically cut into

3 - 4 cm long pieces using scissors cleaned with 70% ethanol and put into 50 mL of sterile saline wash solution comprising 0.9% (W/V) NaCl and 0.25 (V/V) Tween 80, in 250 mL

265

clean room manufactured plastic containers. The containers were kept at 4 °C until processed further.

III. Extraction of total microbial population

Total epiphytic microbial population was released into the solution by agitating the containers at 120 rpm for 3 hours on a rotary shaker at 25 °C. Then, the solution was transferred to sterile 50 mL conical centrifuge tubes and centrifuged at 5000g for 10 minutes. Then, the supernatant was decanted, the pellet was resuspended in 1 mL of sterile saline wash solution in sterile 2 mL microcentrifuge tubes and immediately stored at -20

°C.

IV. DNA Extraction

Total bacterial DNA was extracted using a FastDNA™ SPIN Kit (MP Biomedicals LLC) with bead beating by FastPrep® instrument. Samples were thawed to room temperature and vortexed before transferring 200 µL into 2 mL screw-capped lysis tubes that contained a mixture of ceramic and silica particles, and the protocol for FastDNA™ SPIN

Kit was followed. The DNA was eluted in 100 µL of ultra-pure water supplied by the manufacturer. For further analysis, 50µL of extracted DNA was taken representing a single sampling point.

V. DNA quality and quantification

The quantity and the quality of eluted DNA was checked using NanoDrop™ spectrophotometer (absorbance at 260 nm, ratio absorbance at 260/230 nm and 260/280 nm). In addition, DNA samples were diluted to 10 ng of template with DNase free sterile water and three µL of 10 ng template DNA was electrophoresed on a 1% agarose gel.

Furthermore, a PCR amplification of 16SrRNA gene was conducted, followed by electrophoresis of PCR products on 1% agarose gel.

266

VI. Profiling bacterial diversity

DNA extracts from three samples [1a with disease symptoms (1a23D), 1a without disease symptoms (1a23), 1b without disease symptoms (1b24)] were prepared according to the

Australian Genome Research Facility (AGRF) guidelines and sent to AGRF for diversity profiling by 16SrRNA gene sequencing.

Results and Discussion

I. Diversity of bacteria

Identification of species was not complete, and only three species were identified namely,

Erwinia dispersa, Enterobacter hormaechei, Ewingella americana. Therefore, the diversity indices were calculated based on genus level information. Based on relative proportions of genera, Shannon’s diversity index and Evenness index of these samples are as below:

Sample Shannon’s diversity Evenness Index (EH) index (H) 1a23 0.546 0.213 1a23D 1.307 0.422 1b24 0.662 0.238

The diversity and evenness indices of the bacterial population associated with disease symptoms (1a23D) are higher than that without disease symptoms (1a23 and 1b24).

However, the general assumption is that the microbiome of a healthy plant has more diversity and evenness than that of a plant showing disease symptoms. It should be noted that this study considered only the epiphytic bacterial population; hence, the total diversity and evenness indices of the microbiome could have been entirely different.

267

II. Composition at taxonomic level : Family

The epiphytic bacterial microbiome from symptomatic plants was found to contain families Paenibacillaceae, Streptococcaceae, Sphingomonadaceae, Aeromonadaceae and an unspecified family belonging to the Order Pseudomonadales, which were not present in the epiphytic bacterial microbiome from the asymptomatic plants, that was taken from the same sampling point with plants located 15-30 cm apart.

Family 1a23D 1a23 1b24 Unassigned + + + Paenibacillaceae + - - Streptococcaceae + - - Clostridiaceae - + + Sphingomonadaceae + - + Oxalobacteraceae - - + Gammaproteobacteria;Other;Other + + + Aeromonadaceae + - + Enterobacteriaceae + + + Pseudomonadales;Other + - - Moraxellaceae + + - Pseudomonadaceae + + +

Although Paenibacillaceae, Streptococcaceae, Sphingomonadaceae and

Aeromonadaceae are not considered as plant pathogens, the presence of Order

Pseudomonadales is interesting. However, sample 1a23 was dominated by

Enterobacteriaceae (84%), 1a23D with Aeromonadaceae (53%) and

Enterobacteriacease (27%) and 1b24 by Aeromonadaceae (84%), which not prominently associated with the wheat phyllosphere, but are present in submerged soil. Both

Aeromonadaceae and Enterobacteriaceae proliferate under anaerobic conditions, which can be explained by the delay in sample processing facilitating bacteria proliferation in the wash solution under hermetic conditions.

268