Molecular Analysis of Host Resistance and Pathogenicity Of

MOLECULAR ANALYSIS OF HOST RESISTANCE AND PATHOGENICITY OF

RICE BLAST IN EAST AFRICA

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Emmanuel M. Mgonja, MS

Graduate Program in Plant Pathology

The Ohio State University

2016

Dissertation Committee:

Professor Guo-Liang Wang, Advisor

Professor Thomas K. Mitchell, Co-advisor

Professor Sally A. Miller

Professor Esther Van der Knaap

Copyrighted by

Emmanuel M. Mgonja

2016

ABSTRACT

Africa is endowed with rich rice genetic resources that can be utilized to boost rice production and productivity on the continent. Detailed understanding of rice genetic resources and diversity of germplasm collections is significantly important for the management and sustainable use of genetic stocks and rice breeding programs. One of the major constraints to rice production in Africa is rice blast disease, caused by the fungal pathogen Magnaporthe oryzae. Deployment of host resistance combined with the use of good cultural practices remains the most promising approach for blast disease management. In this dissertation, findings from genotyping-by-sequencing (GBS) based diversity analyses, association mapping of blast resistance loci and screening for blast resistant cultivars are presented. Genetic diversity analysis of 190 African rice cultivars using 184K single nucleotide polymorphisms (SNPs) generated by GBS, revealed the existence of three major groups in the African rice population. The grouping was independent of ecologies, and varieties in different subspecies were mixed in all the three clusters. The possible explanation to such trend could be due to misclassification of these cultivars or possible admixtures that could have happened in handling seeds or may be a result of natural hybridization between O. sativa and O. glaberrima. Misclassifications of some rice cultivars were discovered using phylogenetic and PCR analyses. Association mapping using the SNP data from 190 African cultivars and 162 Rice Diversity panel 1

(RDP1) cultivars and the inoculation data of these rice cultivars with a total of nine African ii

M. oryzae isolates identified 56 loci that are significantly associated with blast resistance.

Moreover, we established the linkage of two major loci on chromosome 1 and 12 to two resistance (R) genes (Pish and Pita), respectively, by using PCR analysis. A total of 30 highly blast resistant rice cultivars to the nine isolates were obtained. Further screening of these cultivars using seven West African M. oryzae isolates showed that 50 percent of them conferred a high level of resistance. These findings demonstrate that effectiveness of GBS on population studies and the power of association mapping for quick identification of rice blast R genes and quantitative trait loci (QTLs). The identified loci are effective against many African M. oryzae isolates and could be valuable in blast resistance breeding programs in Africa.

iii

Dedicated to my family

ACKNOWLEDGMENTS

First and foremost, I thank the Almighty God for his unfathomable grace and faithfulness. His glory is supreme and His Word and promises are ultimate.

Special words of appreciation to my advisor, Dr. Guo-Liang Wang for his outstanding support and guidance throughout my research work. He provided me with all possible opportunities and encouragement in reaching my research targets and I am deeply humbled and so grateful. I am also so thankful to my co-advisor, Dr. Thomas Mitchel and my local advisor Dr. Robert Mabagala (Sokoine University of Agriculture, Tanzania) for their technical support and useful advice during the whole course of my study. I would like to express my heartfelt gratitude to my dissertation committee; Dr. Sally Miller, Dr. Esther

Van der Knaap, Dr. Thomas Mitchel and Dr. Guo-Liang Wang for the guidance and support, without which it would have been impossible to complete this work.

I wish to thank all the graduate students in Dr. Wang’s Lab (Pengfei Bai, Pavinee

Suttiviriya and Da-Young Lee) for their social and moral support. Special thanks to Maria

Bellizzi (Lab technician) and Chan Ho Park (Post Doc) for their help and technical support.

To all other members in Dr. Wang’s lab, I thank them all. To my colleague Elias

Balimponya, his efforts and contribution to this work are highly appreciated.

I would like to express my gratitude to the iAGRI/USAID project for funding my studies here in OSU and in Tanzania (fieldwork). Without their financial support this work would have been impossible to accomplish. This project was partly supported by grants from Innovative Agricultural Research Initiative of the USAID, the SCPRID program of

BBSRC and BM Gates Foundation, USDA-NIFA Hatch Project, Borlaug-LEAP fellowship and EAAPP project. The RDP1 seeds were obtained from the Genetic Stocks

Oryza (GSOR) collection laboratory, the USDA Dale Bumpers National Rice Research

Center. Africa rice cultivars were provided by AfricaRice, Benin and ARI-Katrin,

Tanzania.

Last, but definitely not least, I would like to say “Thank you so much” to my family and friends. To my wife Irene and to my son Ethan for their love, support and huge sacrifice in each single day “I love and thank you so much”. To my mother Annah and my father

Mohamed “You are so special and wonderful to me”. To all my brothers and sisters, I am so thankful for all their social and moral support.

VITA

August 20, 1980 ...... Born-Mwanga, Kilimanjaro, Tanzania

2003-2006 ...... B.S. Agronomy, Sokoine University of

Agriculture, Morogoro, Tanzania

2007...... Research Scientist, ARI-Katrin, Ifakara,

Tanzania

2008-2009 ...... JICA fellow, Tsukuba International Center,

Tsukuba, Ibaraki, Japan

2009-2011 ...... MS, International Agricultural

Development, Tokyo University of

Agriculture, Tokyo, Japan

2012 to present ...... PhD, Plant Pathology, The Ohio State

University, Columbus OH, USA

FIELDS OF STUDY

Major Field: Plant Pathology

vii

TABLE OF CONTENTS

ABSTRACT ...... ii ACKNOWLEDGMENTS ...... v VITA ...... vii FIELDS OF STUDY...... vii LIST OF ABBREVIATIONS AND ACRONYMS ...... xi LIST OF TABLES ...... xv LIST OF FIGURES ...... xvi

CHAPTER 1: Introduction Pathogenesis of Magnaporthe oryzae and host resistance to the pathogen ...... 1 The status of rice production in Africa...... 2 Blast disease, a major rice production constraint in Africa...... 3 Pathogenesis of M. oryzae ...... 3 Rice resistance mechanism and resistance gene mapping ...... 5 References ...... 8

CHAPTER 2: Genome-wide Association Mapping of Rice Resistance Genes against Magnaporthe oryzae Isolates from Four African Countries Introduction ...... 11 Materials and methods ...... 14 Plant and fungal materials ...... 14 RAPD and PCR analysis ...... 15 Evaluation of blast resistance phenotypes ...... 16 Genome-wide association analysis ...... 17 Results ...... 18 Molecular characterization of the eight M. oryzae isolates ...... 18 viii

Resistance of the 162 RDP1 cultivars to the eight M. oryzae isolates ...... 18 Identification of rice QTLs associated with resistance to eight M. oryzae isolates .. 19 RABR_2 is linked to Pish R gene ...... 21 Discussion ...... 22 References ...... 26

CHAPTER 3: GBS-Based Diversity Analysis of African Rice Cultivars and Association Mapping of Rice Blast Resistance Genes. Introduction ...... 52 Materials and methods ...... 57 Plant and fungal materials ...... 57 DNA extraction and GBS ...... 59 GBS based diversity ...... 59 PCR analyses ...... 60 Evaluation of blast resistance phenotypes ...... 61 Association analysis ...... 62 Results ...... 62 GBS-based diversity of African rice cultivars ...... 62 Blast resistance phenotypes ...... 63 Identification of rice QTLs associated with resistance to six isolates ...... 64 RABR_23 is linked to Pita R gene ...... 65 Discussion ...... 66 References ...... 70

CHAPTER 4: Rice Blast Resistance Spectrum of Thirty African Rice Cultivars against Seven Magnaporthe oryzae Isolates from Benin Introduction ...... 106 Materials and methods ...... 108

Plant and fungal materials ...... 108 Blast resistance screening ...... 109 Results ...... 110 Blast resistance phenotypes ...... 110 Discussion ...... 111 References ...... 113

CHAPTER 5: Project Summary and Conclusion ...... 131

BIBLIOGRAPHY ...... 133

LIST OF ABBREVIATIONS AND ACRONYMS

Abbreviation/Acronym Term

AR Africa Rice lines

ARS Agricultural Research Service

BBSRC Biotechnology and Biological Sciences Research Council

BF Burkina Faso

BM Bill & Melinda

CTAB Cetyltrimethyl ammonium bromide

DNA Deoxyribonucleic Acid dNTPs Deoxynucleotide triphosphates

EAAPP Eastern Africa Agricultural Productivity Project

EDTA Ethylene diamine tetra-acetic acid

EN External source rice lines

FAO Food and Agriculture Organization of the United Nations

GAPIT Genome Association and Prediction Integrated Tool

GBS Genotype by Sequencing

GSOR Genetic Stocks – Oryza

GWAS Genome-wide Association Studies

iAGRI Innovative Agricultural Research Initiative

IRRI Rice Research Institute

KE Kenya

LABR Locus Associated with Blast Resistance

LD Linkage Disequilibrium

LEAP Leadership Enhancement in Agriculture Program

L-NER Lowland NERICA lines

LOD Logarithm of Odds

LRR Leucine rich repeat

MAF Minor Allele Frequency

NBS Nucleotide binding-site

NB-ARC Nucleotide-Binding adaptor shared by APAF-1, R proteins

and CED-4

NERICA New rice for Africa

NIFA National Institute of Food and Agriculture

OSU The Ohio State University

PCA Principal Component Analysis

PCR Polymerase Chain Reaction

Pi genes Pyricularia genes

xii

QQ Quantile-Quantile

QTL Quantitative Trait Locus

QTLs Quantitative Trait Loci

RABR Region Associated with Blast Resistance

RAPD Random Amplified Polymorphic DNA

RDP1 Rice Diversity panel 1

R genes Resistance genes

SCPRID Sustainable Crop Production Research for International

Development

SNPs single nucleotide polymorphisms

SUA Sokoine University of Agriculture

TASSEL Trait Analysis by aSSociation, Evolution and Linkage

TEJ Temperate japonica lines of rice

JIRCAS Japan International Research Center for Agricultural Sciences

TZ Tanzania

TZLR Tanzania landrace rice lines

UG Uganda

UP-NER Upland NERICA

USAID United States Agency for International Development

xiii

USDA United States Department of Agriculture

UV Ultraviolet

xiv

LIST OF TABLES

Table 1. Random Amplified Polymorphic DNA (RAPD) primer sequences ...... 31

Table 2. Summary of Rice Diversity Panel-1 disease phenotypes ...... 32

Table 3. Resistant and susceptible Rice Diversity Panel-1 (RDP1)...... 33

Table 4. The regions associated with rice resistance (RABRs)...... 34

Table 5. Gene annotations for the candidate genes...... 36

Table 6. List of primers used for Pish gene amplification...... 38

Table 7. Origin and ecologies of 190 African rice cultivars...... 78

Table 8. List of specific primer pairs ...... 84

Table 9. List of 42 rice cultivars used for PCR confirmation...... 85

Table 10. List of rice resistant and susceptible cultivars...... 87

Table 11. Summary of 190 African rice cultivar disease phenotypes ...... 88

Table 12. African rice cultivars resistant to six African M. oryza isolates ...... 89

Table 13. The regions associated with rice resistance (RABRs)...... 90

Table 14. Gene annotations for the candidate genes...... 91

Table 15. List of primers used for Pita gene amplification ...... 93

Table 16. Blast disease scores of 60 rice cultivars...... 122

Table 17. Rice cultivars highly resistant to seven Magnaporthe oryzae isolates ...... 124

LIST OF FIGURES

Figure 1. Genetic relationship among the eight isolates ...... 39

Figure 2. An example of disease score distribution of the RDP1 cultivarsr...... 40

Figure 3. Rice blast disease score distribution of the RDP1 cultivars ...... 41

Figure 4. Rice blast disease score distribution of the RDP1 cultivars ...... 42

Figure 5. Rice blast disease score distribution of the RDP1 cultivars ...... 43

Figure 6. Cluster analysis of the thirteen M. oryzae isolates ...... 44

Figure 7. Principal component analysis of the eight M. oryzae isolates ...... 45

Figure 8. A combined Manhattan plot summarizing GWAS results ...... 46

Figure 9. Genome-wide association analysis for two Kenyan M. oryzae isolates ...... 47

Figure 10. Genome-wide association analysis for two Burkina-Faso M. oryzae ...... 48

Figure 11. Genome-wide association analysis for two Tanzanian M. oryzae isolates ...... 49

Figure 12. Genome-wide association analysis results of the RDP1 cultivars...... 50

Figure 13. Gel images of the amplified Pish gene fragments...... 51

Figure 14. Phylogenetic tree showing the relationship among rice accessions ...... 94

Figure 15. Gel images of the amplified African rice subspecies specific markers ...... 95

Figure 16. Disease score distribution of the African rice cultivars...... 96

Figure 17. Rice blast disease score distribution of the African rice cultivars...... 97

Figure 18. Rice blast disease score distribution of the African cultivars ...... 98 xvi

Figure 19. Cluster analysis of the six M. oryzae isolates based on disease scores...... 99

Figure 20. A combined Manhattan plot summarizing GWAS results ...... 100

Figure 21. Genome-wide association analysis for two Ugandan M. oryzae isolates...... 101

Figure 22. Genome-wide association analysis for two African M. oryzae isolates ...... 102

Figure 23. Genome-wide association analysis Tanzanian M. oryzae isolate TZ-01 ...... 103

Figure 24. Genome-wide association analysis for Tanzanian M. oryzae ...... 104

Figure 25. Gel images of the amplified Pita gene fragments ...... 105

Figure 26. Rice blast disease score distribution of the 30 African rice cultivars ...... 125

Figure 27. Rice blast disease score distribution of the 30 African rice cultivars ...... 126

Figure 28. Rice blast disease score distribution of the 30 African rice cultivars ...... 127

Figure 29. Rice blast disease score distribution of the 30 African rice cultivars ...... 128

Figure 30. Cluster analysis of the seven M. oryzae isolates based on disease scores. .... 129

Figure 31. Cluster analysis of the seven M. oryzae isolates based on disease scores ..... 130

xvii

CHAPTER 1

INTRODUCTION

Pathogenesis of Magnaporthe oryzae and host resistance to the pathogen

Rice is a staple food that feeds more than half of the world’s population (Khush 2005; Way and Heong 1994) and its grains are the most superior source of energy amongst cereals.

Rice grains are considered as both human food and cash crop. Rice stalks are used as animal feed and thatching, and the husks are used as litter for poultry as well as mulch. Rice belongs to the genus Oryza which consists of twenty-six species distributed throughout tropical and subtropical areas, but there are only two cultivated species, Oryza glaberrima

Steud and Oryza sativa Linn. Oryza glaberrima is an upland crop and is confined to Africa while O. sativa is grown worldwide (USDA 2012). Rice farming is practiced in a wide range of ecological conditions, although most occurs in warm/cool humid subtropics, warm humid tropics, and in warm sub-humid tropics. The International Rice Research Institute

(IRRI) (1993) categorized rice land ecosystems into four types: irrigated rice, rainfed lowland rice, upland rice, and flood-prone rice. Apart from the upland system, the others are characterized as wet rice cultivation. Rice can be grown as a single crop or in mixed systems with other crops like wheat, maize and pulses

(Swaminathan 1984; Timsina and Connor 2001). The time period required to reach maturity differs among varieties. While most traditional varieties require about 150 days of growth to reach grain maturity, improved high yielding and early maturing varieties only require about 90 days before harvesting (Moldenhauer and Gibbons 2003). This chapter presents the background information on rice blast disease in Africa, pathogenicity of the blast fungus and mapping of blast resistance loci/genes.

The status of rice production in Africa

Rice has gained popularity all over Africa due to its uniqueness as both food and cash crop. In 2012, Africa cultivated about 6.2 % of the total global rice area and contributed

3.8% of total global rice production (FAOSTAT data 2012). In Africa, agriculture is not a business but a way of life. Rice is currently becoming the major crop for food security and income generation for African farmers (AfricaRice 2012). In 2012, the rice crop generated

$7 billion, ranking fifth in Africa on economic value among all agricultural products including dairy. The current increasing demand for rice on the continent is mainly due to a general dietary shift from conventional foods as a result of improved income among middle class and high rate of urbanization (Khush and Jena 2009). Major rice producing areas in

Africa are found in the Eastern and Western parts of the continent. In Eastern Africa,

Tanzania is leading in terms of rice production (1.8 million metric tons) and consumption

(60kg/person/year) after Madagascar. Uganda and Kenya have lower production and consumption rates (FAOSTAT data 2012)

Blast disease, a major rice production constraint in Africa

Despite the increasing importance and demand for rice, its production is facing many biotic and abiotic constraints including drought, salinity, pests and diseases. These constraints are even further aggravated by current climatic change and crop intensification.

Rice crop diseases pose a major threat to rice productivity in Africa. Of particular concern is rice blast disease caused by a fungal pathogen Magnaporthe oryzae (Syn: Pyricularia oryzae) (Talbot 2003; Valent and Chumley 1991). Rice blast disease can cause between

10% to 30% yield loss depending on variety susceptibility (Talbot 2003) and up to 100% during an epidemic (Khush and Jena 2009). The disease is devastating in many parts of

Africa where rice is grown (Séré et al. 2013).

Pathogenesis of M. oryzae

Rice blast fungus has a very wide host range infecting more than 50 plant species, most of which belongs to grass family (gramineae) including rice, wheat, finger millet and barley (Talbot 2003). Different races of the fungus can infect different parts of rice plants and in some cases a single race can infect all aerial parts of the plant at the same time. But the disease is most conspicuous when the pathogen attacks the panicles, leaf collar, neck, nodes and leaf blades. However, leaf blast is the most common type and is generally characterized by elliptical or spindle shaped lesions. The lesions or spots first appear as minute brown specks, and eventually grows to become spindle shaped. The center is greyish with a brown margin. The lesions may widen and eventually coalesce, killing the entire leaf. The fungus overwinters by means of mycelia and spores inside the infected

plant remains or in infected seeds. The pathogen can also infect and overwinter in the alternative hosts, i.e., most poaceae/gramineae grass family, making the disease cycle even more complex. Rice blast disease development is favored by a number of factors such as high relative humidity (above 80%), low temperature (150C-260C), cloudy weather, high number of wet or rainy days, long dew durations, slow wind movement, availability of alternative hosts and excessive doses of nitrogenous fertilizers (Rousk et al. 2009).

When a spore lands on the surface of a rice leaf, the infection process is initiated. The spore attaches itself on the waxy cuticle layer on the plant surface by means of adhesive substances secreted at the tip of the spore (Hamer et al. 1988). For spore germination to occur, a water film is required. The germination process is rapid, taking only few hours for the formation of a germ tube that grows by apical extension and a hooking process occurs followed by appressorium formation (Bourett and Howard 1990). Formation of an appressorium (penetration structure) is vital in the infection process (pathogenicity). The appressorium requires a hydrophobic surface to build turgor pressure and push itself into the host cell (Dean 1997). After successful penetration, the penetration hyphae differentiate into highly branched infectious hyphae that colonize the entire host cell starting as a bio- trophic then suddenly switching to necrotrophic phase, killing the entire host cell. The mechanism underlying this switching from bio-trophic to necrotrophic is not well understood. Upon infection and colonization, the fungus can produce either three-celled asexual spores (conidia) in clusters on long conidiophores or four celled sexual spores

(ascospores) in perithecia. These spores can be dispersed by wind or water and land on other parts of the same plant or different plants causing new infections (Rossman et al.

1990). The disease cycle is rather short (5-7 days) with most of the damage from the secondary infection.

Rice resistance mechanism and resistance gene mapping

Single gene resistance to rice blast disease follows a classical hypothesis of the gene- for-gene interaction. This type of resistance is also referred to as monogenic or complete resistance. In complete resistance, the host plant single dominant R gene confers resistance against a specific pathogen race carrying a corresponding dominant avirulence (AVR) gene.

In the process of infection, the pathogen produces and releases specific avirulence proteins

(effectors) encoded by AVR gene to facilitate its pathogenicity. These effector proteins are recognized by plant surveillance system possessing the corresponding resistance R gene and trigger resistance against the pathogen (Jia et al. 2000; Talbot 2003). Since these effectors are often race or strain‐specific, a plant host may be resistant to one race of a pathogen but susceptible to another.

To date, nearly 100 blast resistance R genes has been identified and 22 of them are already cloned and characterized (Liu et al. 2014; Ramkumar et al. 2014; Wang et al.

2014c). Although, these single R genes can provide complete resistance to the targeted races of pathogen, this type of resistance is easily broken down under high pathogen population pressure and due to the fact that M. oryzae is highly variable and undergoes genetic modifications to overcome the host surveillance system. To overcome this problem, researchers have come up with a way of stacking many R genes in a single line, a method referred to as gene pyramiding. In this case the resulting resistance becomes broader and more effective to a number of pathogen races (Pinta et al. 2013).

In recent years, the attention has been shifting to consider another form of disease resistance, partial resistance. Partial resistance also referred to as incomplete resistance is controlled by many genes with small effects. This type of resistance is inherited quantitatively, involving many genes located on different chromosome loci. These loci are often referred as quantitative trait loci (QTLs). Unlike complete resistance, incomplete resistance is robust and hard to break.

Several studies have been carried out to map blast resistance-associated QTLs and showing their usefulness in proving a resistance spectrum against races of M. oryzae. To date, about 347 QTLs have already been identified and mapped in the rice genome. The availability of high quality rice genome sequence has contributed immensely to such rapid advancement in QTL mapping (Project 2005). The analysis of these QTLs have shown that

165 QTLs have major effects while the remaining have minor effects but significantly contribute to broad and robust resistance to rice blast (Ballini et al. 2008). These findings point to the usefulness of combining major R genes and QTLs to produce elite cultivars with broad and robust resistance against rice blast disease. More studies are still needed to identify new loci associated with rice blast resistance in diverse rice germplasm collections.

This dissertation demonstrates the usefulness of association mapping for the identification of resistance R genes and QTLs associated with blast resistance and contribute in understanding African rice diversity and pathogenicity of M. oryza. This study is addressing the following research questions and hypotheses.

Research questions:

1. Do the East African M. oryzae isolates differ in virulence level?

2. How diverse are the East African M. oryzae isolates?

3. Do the African rice cultivars contain new resistance genes to M. oryzae?

Hypotheses:

H : East African M. oryzae isolates are highly diverse with different virulence levels. 1

H : The African rice varieties contain new resistance genes against M. oryzae infection. 2

References

AfricaRice. 2012. Africa Rice Centre (AfricaRice) Annual Report 2011: A new rice research for development strategy for Africa. Cotonou, Benin.

Ballini, E., Morel, J. B., Droc, G., Price, A., Courtois, B., Notteghem, J. L., and Tharreau, D. 2008. A genome-wide meta-analysis of rice blast resistance genes and quantitative trait loci provides new insights into partial and complete resistance. Mol Plant Microbe Interact 21:859-868.

Bourett, T. M., and Howard, R. J. 1990. In vitro development of penetration structures in the rice blast fungus Magnaporthe grisea. Canadian Journal of Botany 68:329-342.

Dean, R. A. 1997. Signal pathways and appressorium morphogenesis. Annual Review of Phytopathology 35:211-234.

FAOSTAT data. 2012. FAOSTAT.http://faostat.fao.org/site/339/default.aspx. Retrieved June 12, 2016

Hamer, J. E., Howard, R. J., Chumley, F. G., and Valent, B. 1988. A mechanism for surface attachment in spores of a plant pathogenic fungus. Science 239:288-290.

Jia, Y., McAdams, S. A., Bryan, G. T., Hershey, H. P., and Valent, B. 2000. Direct interaction of resistance gene and avirulence gene products confers rice blast resistance. The EMBO journal 19:4004-4014.

Khush, G. S. 2005. What it will take to feed 5.0 billion rice consumers in 2030. Plant Molecular Biology 59:1-6.

Khush, G. S., and Jena, K. 2009. Current status and future prospects for research on blast resistance in rice (Oryza sativa L.). Pages 1-10 in: Advances in genetics, genomics and control of rice blast disease. Springer, Dordrecht, Netherlands.

Maclean, J. L., and Dawe, D. C. 2002. Rice almanac: Source book for the most important economic activity on earth. CAB International, Wallingford, UK, in association with the International Rice Research Institute, Los Banos, the Philippines.

Mew, T. W., Leung, H., Savary, S., Vera Cruz, C. M., and Leach, J. E. 2004. Looking ahead in rice disease research and management. Critical Reviews in Plant Sciences 23:103-127.

Moldenhauer, K. A., and Gibbons, J. H. 2003. Rice morphology and development. Rice: Origin, History, Technology, and Production. Hoboken, NJ. John Wiley and Sons.

Pinta, W., Toojinda, T., Thummabenjapone, P., and Sanitchon, J. 2013. Pyramiding of blast and bacterial leaf blight resistance genes into rice cultivar RD6 using marker assisted selection. African Journal of Biotechnology 12.

Project, I. R. G. S. 2005. The map-based sequence of the rice genome. Nature 436:793- 800.

Ramkumar, G., Madhav, M. S., Rama Devi, S. J. S., Manimaran, P., Mohan, K. M., Balachandran, S. M., Neeraja, C. N., Sundaram, R. M., Viraktamath, B. C., and Prasad, M. S. 2014. Nucleotide diversity of Pita, a major blast resistance gene and identification of its minimal promoter. Gene 546:250-256.

Rossman, A. Y., Howard, R. J., and Valent, B. 1990. Pyricularia grisea, the Correct Name for the Rice Blast Disease Fungus. Mycologia 82:509-512.

Rousk, J., Brookes, P. C., and Bååth, E. 2009. Contrasting soil pH effects on fungal and bacterial growth suggest functional redundancy in carbon mineralization. Applied and Environmental Microbiology 75:1589-1596.

Savary, S., Horgan, F., Willocquet, L., and Heong, K. 2012. A review of principles for sustainable pest management in rice. Crop protection 32:54-63.

Séré, Y., Fargette, D., Abo, M. E., Wydra, K., Bimerew, M., Onasanya, A., and Akator, S. K. 2013. Managing the Major Diseases of Rice in Africa. Realizing Africa's Rice Promise:213-228

Swaminathan, M. S. 1984. Rice in 2000 AD. In: Abrol and Sulochana Gadgil (eds.), Rice in a variable climate. APC Publications Pvt. Ltd., New Delhi-110005, India.

Talbot, N. J. 2003. On the trail of a cereal killer: exploring the biology of Magnaporthe grisea. Annual Reviews in Microbiology 57:177-202.

Timsina, J., and Connor, D. 2001. Productivity and management of rice–wheat cropping systems: issues and challenges. Field Crops Research 69:93-132.

USDA, National Genetic Resources Program. 2012. Germplasm Resources Information Network – (GRIN) [Online Database]. National Germplasm Resources Laboratory, Beltsville, Maryland. www.ars-grin.gov/cgi-bin/npgs/html/index.pl (accessed 12 May 2016).

Valent, B., and Chumley, F. G. 1991. Molecular genetic analysis of the rice blast fungus, Magnaporthe grisea. Annu Rev Phytopathol 29:443-467.

Wang, X., Lee, S., Wang, J., Ma, J., Bianco, T., and Jia, Y. 2014. Current advances on genetic resistance to rice blast disease. Pages 195-217 in: Rice Germplasm, Genetics and Improvement. W. Yan and J. Bao, eds. InTech, Rijeka, Croatia.

Way, M., and Heong, K. 1994. The role of biodiversity in the dynamics and management of insect pests of tropical irrigated rice-A review. Bulletin of Entomological Research 84:567-588.

CHAPTER 2

Genome-wide Association Mapping of Rice Resistance Genes against Magnaporthe

oryzae Isolates from Four African Countries

Introduction

The popularity of rice as a food crop is rapidly increasing throughout Africa. Rice consumption in urban areas of Africa has significantly increased, and per capita consumption has doubled since 1970 (Muthayya et al. 2014). The demand for rice is becoming higher because of a general dietary shift away from traditional foods such as maize, cassava, yams, millet, and sorghum (Nayar 2014). The current per capital consumption of rice in Africa is estimated to be 23.3 kg per annum (FAO 2012). Despite this increase in rice consumption, only 40% of the rice consumed in Africa is produced in

Africa (IRRI 2013).

As in other rice-growing areas of the world, rice blast disease caused by the fungus

Magnaporthe oryzae (Bourett and Howard 1990; Ribot et al. 2008), is a major constraint for rice production in Africa. The disease is spreading in both lowland and upland ecosystems (Séré et al. 2013). Deployment of resistant varieties combined with the use of good cultural practices remains the most promising approach to blast disease management

(Maclean and Dawe 2002; Mew et al. 2004; Savary et al. 2012) .

Single resistance (R) gene-mediated resistance is qualitative and provides complete resistance against specific races of M. oryzae exhibiting a gene-for-gene interaction with pathogens.

In contrast to complete resistance, partial resistance is quantitatively inherited and controlled by multiple quantitative trait loci (QTLs); in addition, partial resistance is considered to be more durable than R gene-mediated resistance (Mundt 2014). Because of its clear resistance phenotype and dominant expression, however, complete resistance has long been favored over partial resistance by plant breeders. Over sixty major R genes and

17 minor R genes against M. oryzae have been mapped in the rice genome, and 22 of these have already been cloned (Liu et al. 2014; Wang et al. 2014b). The effectiveness of these

R genes against the M. oryzae populations in Africa has not been extensively tested.

Although mutational and conventional breeding techniques including pedigree selection, anther culture, wide hybridization between O. glaberrima and O. sativa, and backcrossing have been widely used in Africa (Habarurema et al. 2012; Luzi-Kihupi et al. 2009; Singh et al. 2001), the efficiency of breeding for disease resistance has been low on the continent because of inefficient field screening, high pathogen population pressure and lack of information on the prevailing M. oryzae population structure and virulence (Tharreau et al.

2009; Zeigler et al. 1994).

Traditional gene mapping methods using F2 populations and recombinant inbred lines

(RILs) have been used to localize target genes in many crop plants. Although it will continue to be an important tool for gene mapping, this approach is time-consuming and provides only low mapping resolution. These limitations, however, can be reduced with

the use of the genome-wide association study (GWAS) approach (Abdurakhmonov and

Abdukarimov 2008). Unlike traditional gene-tagging using bi-parental crosses, GWAS uses natural populations or germplasm collections and linkage disequilibrium (LD)-based association to rapidly map target genes in a large collection of diverse genotypes.

Application of GWAS for the dissection of complex traits in maize, rice, foxtail millet, and sorghum has been reported in recent years (Huang et al. 2010; Jia et al. 2013; Kump et al.

2011; Li et al. 2013; Morris et al. 2013). For example, GWAS was used to identify multiple genes that potentially control complex traits such as ﬂowering time, leaf size, leaf angle, and disease resistance (Buckler et al. 2009; Poland et al. 2011; Tian et al. 2011).

In rice, researchers have used GWAS with high-density single nucleotide polymorphism (SNP) chips and low-coverage sequencing methods to identify genes and

QTLs associated with traits related to abiotic stress, grain quality, and agronomic performance. In a recent GWAS, 366 diverse indica accessions with 0.8 million SNPs were inoculated with 16 M. oryzae isolates. A total of 30 loci associated with blast disease resistance were identified, and four of these loci were linked to previously mapped R genes while 26 were novel candidate genes that may contribute to defense responses (Wang et al.

2014a).

Our laboratory recently used the rice diversity panel 1 (RDP1) containing 0.7 million

SNPs and five M. oryzae isolates to identify 97 loci associated with blast resistance

(LABRs). Fifteen of these LABRs were linked to known R gene loci, and 82 were new.

Some of the LABRs are strongly linked to R and defense genes encoding nucleotide binding site-leucine rich repeats (NBS-LRR), defense-related proteins, transcription factors, and

receptor-like protein kinases (Kang et al. 2015). In this study, 162 RDP1 cultivars (70 indica and 92 temperate japonica) (Eizenga et al. 2013) were used. These cultivars were inoculated with eight isolates from four African countries. Association mapping showed that 31 regions associated with blast resistance (RABRs) in the rice genome were involved in the resistance against the eight isolates. PCR analysis revealed that a major RABR on chromosome 1 is associated with resistance to four M. oryzae isolates. This study demonstrates the effectiveness of GWAS for the rapid identification of R/QTLs genes in rice and provides linked SNP markers for breeding for resistance against rice blast in

Africa.

Materials and methods

Plant and fungal materials

The 162 O. sativa accessions (70 indica and 92 temperate japonica) used in this study were part of the RDP1 (Eizenga et al. 2013) and were obtained from the Genetic Stocks–

Oryza (GSOR) Collection, USDA ARS Dale Bumpers National Rice Research Center.

These cultivars were collected from 82 countries. Seeds were increased in a greenhouse at

The Ohio State University in the summer of 2013. The eight M. oryzae isolates used in this study were part of M. oryzae collection from a field survey conducted in the summer of

2013 in four African countries: Tanzania, Uganda, Kenya and Burkina Faso. The selection of the eight isolates was based on pathogenicity results of four rice cultivars (IR64, Toride,

Nipponbare, and CO-39) with different levels of resistance inoculated with 60 M. oryzae isolates. The eight M. oryzae isolates selected for this study were: TZ-01 and TZ-12 from

Tanzania; UG-05 and UG-11 from Uganda; KE-14 and KE-37 from Kenya; and BF-05 and

BF-27 from Burkina Faso.

RAPD and PCR analysis

Random amplified polymorphic DNA (RAPD) markers were used to assess the genetic variability of the eight most virulent isolates ((Govarthanan et al. 2011; Williams et al. 1990). Fungal DNA was extracted from the mycelia using the cetyltrimethyl ammonium bromide (CTAB) method (Bashir et al. 2014). DNA was quantified with a

Nanodrop spectrophotometer. PCR amplifications were performed in a 25-μl reaction mixture consisting of 50 ng/μl genomic DNA, 1X reaction buffer (Promega), 0.25 mM dNTPs, 0.2 μM random primer, 2.5 μM MgCl2, and 1 unit of Taq polymerase. The amplification included one denaturing cycle of 4 min at 94°C; followed by 45 cycles of 1 min at 94°C, 1 min at 40°C, and 1 min at 72°C; and a final extension step of 2 min at 72°C.

The amplified products were resolved by electrophoresis on a 1.4% agarose gel using TAE buffer (45 mM Tris-acetate, 1 mM EDTA, pH 8.0) at 100 volts for 2 h. A 1-kb ladder was included as a molecular size marker. Gels were stained with an ethidium bromide solution

(0.5 μg/ml), and band patterns were visualized with UV light. Definite, bright polymorphic bands of various molecular weights that were generated by the RAPD markers were scored

“1” for presence and “0” for absence. Euclidean Distance, Complete Linkage in Minitab software was used to perform a cluster analysis and to create a dendrogram of the eight M. oryzae isolates based on RAPD marker scores.

PCR analysis of RABR_2 was done using ten resistant and 15 susceptible rice cultivars from RDP1 based on the inoculation results. Six specific primers (Table 6) designed using the sequences of the cloned R gene Pish as a reference were used. The PCR reaction started with a denaturation step for 3 min at 95°C followed by 33 cycles of 30 seconds min at

95°C, 30 seconds min at 58°C for annealing of primers, 1 min at 72°C for extension and stopped after a final extension step of 72°C for 10 min. The amplified products were resolved by electrophoresis on a 1.4% agarose gel using TAE buffer (45 mM Tris-acetate,

1 mM EDTA, pH 8.0) at 100 volts for 2 h. A 1-kb ladder was included as a molecular size marker. Gels were stained with an ethidium bromide solution (0.5 μg/ml), and band patterns were visualized with UV light.

Evaluation of blast resistance phenotypes

Rice seedlings and fungal cultures were grown and prepared for spray inoculation as described by Park et al. (2012). About 15 seedlings of each of the 162 rice cultivars were grown in a growth chamber maintained with 12 h light/12 h, a temperature of 26°C with light and 21°C with darkness, and 80% humidity (Park et al. 2012). Seedlings with three or four fully expanded leaves (18-20 days old) were sprayed with fungal spore suspensions

(5x105 conidia/ml in 0.1% Tween-20) until droplets formed on the leaves. Each sprayed seedling was kept in a plastic bag (≥90% humidity) in the dark for 24 h and was then returned to the growth chamber (Wang et al., 2012). Six days after inoculation, the seedlings were scored for disease using a 0-9 scale, in which 0 indicates no blast symptoms and 9 indicates severe blast symptoms (Zhu et al. 2012). The inoculation experiment was

performed twice under the same conditions. If the results from two experiments were different, a third experiment was performed and data from the two similar experiments were used. Data analysis was carried out using the R statistical package. The cluster analysis and principal components analysis (PCA) were conducted based on the interactions between rice genotypes and M. oryzae isolates.

Genome-wide association analysis

GWAS was performed using the phenotypic data (blast disease scores) of the 162 rice accessions of the RDP1 and genotypic data (publicly available 44 K-SNP dataset) previously described by Zhao et al. (2011). A mixed linear model (MLM) (Bradbury et al.

2007a) was implemented with Tassel 5.0 software (http://www.maizegenetics.net/tassel/).

The computational power of the MLM was further increased by using a compressed MLM, which jointly uses kinship (K) matrix and population structure (Q) matrices The K matrix is the variance-covariance matrix between the individuals, and the Q matrix is obtained through both STRUCTURE and PCA. This model can be summarized using Henderson’s matrix notation as Y = Xβ + Zu + e, where Y is the vector of observed phenotypes; β is an unknown vector containing fixed effects, including the genetic marker, population structure (Q), and the intercept; u is an unknown vector of random additive genetic effects from multiple background QTLs for individuals/lines; X and Z are the known design matrices; and e is the unobserved vector of random residuals (Henderson 1975; Lipka et al. 2012). Based on PERL (Christiansen et al. 2012) and its SVG module (scalable vector graphics), and using the TASSEL results as input files, we developed a Perl script to

combine multiple Manhattan plot figures into an integrated one (combined Manhattan plot). Manhattan and QQ plots were produced using the R package.

Results

Molecular characterization of the eight M. oryzae isolates

IR64, Toride, Nipponbare, and CO-39 were inoculated with 60 M. oryzae isolates that were collected from Tanzania, Kenya, Uganda, and Burkina Faso for virulence tests. Based on the inoculation results, the eight most virulent isolates were selected for inoculations.

To determine the genetic diversity of the eight M. oryzae isolates, ten RAPD markers were selected based on reproducibility of amplification. The RAPD marker sequences are shown in Table 1. A total of 48 bright, polymorphic bands were scored. The size of the amplified fragments ranged from 300 to 3,000 bp. An example of the PCR amplification is shown in

Fig. 1A. Phylogenetic analysis showed that the eight isolates were genetically diverse although those from the same country clustered together (Fig. 1B).

Resistance of the 162 RDP1 cultivars to the eight M. oryzae isolates

The distributions of rice blast disease scores obtained by inoculating 162 cultivars with eight isolates M. oryzae are shown in Fig. 2, 3, 4 and 5. The disease scores were generally skewed toward susceptibility (disease score ≥ 8.0) or resistance (disease score ≤ 2.0). The percentages of cultivars that were resistant or susceptible to each isolate are shown in Table

Among the 162 tested cultivars, 11 were highly resistant (disease score ≤ 2.0) and 15 were highly susceptible (disease score > 8.0) to all eight isolates (Table 3). Cluster analysis and

PCA based on disease severity indicated that the eight isolates were diverse but clustered based on the country of origin (Fig. 6 and 7). For example, the two isolates from Tanzania

(TZ-01 and TZ-12) were clustered together (≈75% similarity) and also related to those from Uganda (UG-05 and UG-11), and the two isolates from Kenya (KE-14 and KE-37) were also related to each other (≈84% similarity). Interestingly, the two isolates from

Burkina Faso (BF05 and BF27) were quite different from the other isolates. These results are consistent with those obtained from the RAPD-based PCR analysis (Fig. 1B). The five isolates used by Kang et al (2015) were also included in our pathotype analysis to determine their relationship with our eight African isolates. Cluster analysis results of the thirteen isolates based on the pathotype data (disease phenotypes) of 162 RPD1 cultivars showed low pathotype similarity level between the eight African isolates used in this study and those used by Kang et al (2015) (Fig. 6). Four among the five isolates used by Kang et al

(2015); ROI-1 (from South Korea), RB22 (from China), 75-1-127 (from Colombia) and

P06-6 (from Philippines) were grouped in one separate cluster while isolate 0-249 (from

India) related (68% pathotype similarity) to the two Kenyan isolates (KE-14 and KE-37) used in our study (Fig. 6).

Identification of rice QTLs associated with resistance to eight M. oryzae isolates

By using the 44 K-SNP dataset (36,900 high quality SNPs) (Zhao et al. 2011) and blast inoculation data, 31 non-redundant RABRs (LOD >3.5) significantly associated with

resistance against the eight M. oryzae isolates were identified (Table 4). RABR_1,

RABR_19, RABR_28, RABR_29 and RABR_30 are located in the regions with the five previously mapped R genes Pi27(t), Pi50, Piy(t), Pi34 and Pi43, respectively. RABR_2, and RABR_11 are located in regions with the previously cloned R genes Pish and Pi21, respectively. The remaining 24 RABRs are novel candidate genes (Fig. 8). Detailed information about chromosome locations and candidate genes in the regions for all 31

RABRs is summarized in Table 4. At least three RABRs were associated with each of seven isolates but no RABR was identified for isolate BF-27. The RABRs identified for the isolates from the same country were, however, not the same. For example, no RABR was common between the two Kenyan isolates (KE-14 and KE-37) (Fig. 9) or between the two

Burkina Faso isolates (BF-05 and BF-27) (Fig. 10). A total of 13 RABRs were identified for two Tanzanian isolates (TZ-01 and TZ-12), but only three in common (Fig. 8 and Fig.

11). The two Uganda isolates (UG-05 and UG-11) with a total of 17 RABRs, shared only five RABRs (Fig. 8 and Fig. 12). Some similarity was evident in the RABRs for Tanzanian and Ugandan isolates. TZ-01 shared four RABRs with UG-05 and three with UG-11. UG-

05 and UG-11 each shared six RABRs with TZ-12. Of note, BF-05 (from West Africa) shared eight RABRs with the four isolates from Tanzania and Uganda (East Africa). The two Kenyan (East Africa) isolates did not share any RABRs with the other isolates.

The 31 RABRs are distributed across ten of the 12 rice chromosomes. The highest number of RABRs per chromosome was six (observed on chromosomes 2 and 4) while none were detected on chromosomes 3 or 7. Chromosomes 6 and 11 had four RABRs each, chromosomes 1 and 5 had three RABRs each, chromosomes 8 and 10 had two RABRs

each, and chromosomes 9 and 12 had one RABR each. Thirteen RABRs were associated with resistance to at least two isolates. Two RABRs were associated with resistance to three isolates, and three RABRs were associated for resistance to four isolates. RABR_2, located on chromosome 1, was strongly associated with resistance to five isolates. The SNP signals for the RABR_2 locus were the highest in response to isolate UG-05 (P=9.52E-16,

LOD=13.5) (Fig. 12A).

The reference Nipponbare genomic sequence (MSU V7.0) was then searched using gene ontology (GO) terms to identify candidate R genes associated with the 31 RABRs.

Annotation analysis of the RABRs led to the identification of 36 R or R-related candidate genes (Table 4). The 36 candidate genes can be categorized into eight main functional groups: NBS-LRR genes, Nucleotide-Binding adaptor shared by APAF-1, R proteins, and

CED-4 (NB-ARC) domain genes, protein phosphorylation related genes, transcription factor genes, ubiquitination-related genes, DNA-binding genes, receptor-like protein kinase genes, and oxidase/reductase family genes (Table 4).

RABR_2 is linked to Pish R gene

The strong association of RABR_2 with resistance to five of the eight isolates prompted a detailed analysis of the region. Based on the GWAS results, there were 18

SNPs with a LOD value above 3.5 corresponding to RABR_2. Sequence analysis showed that the cloned R gene Pish is closely linked to the RABR_2 region. A genome analysis by

Takahashi et al (2010) revealed that Pish, along with other three highly conserved NBS-

LRR paralogous genes, belong to the NBS-LRR class of R genes (Takahashi et al. 2010).

Pish gene was first discovered by Imbe and Matsumoto in 1985 (Imbe and Matsumoto

1985). Using the sequences of the cloned R gene Pish as a reference, six primer pairs (Table

6) were designed to amplify the sequence of the candidate gene in ten blast-resistant and

15 blast-susceptible cultivars that were identified in our pathogenicity assay. PCR amplifications revealed that a fragment was generated with all primer pairs for all of the ten resistant cultivars but for none of the 15 susceptible cultivars (Fig. 13A). As a control, amplification was observed in all 25 cultivars when the primer pair N10F+N10R was used

(Fig. 13A). The relative position of the amplified fragments of the six primer pairs in the

Pish coding region is shown in Fig. 13B.

To confirm the association of Pish gene to a major resistance locus on chromosome 1

(RABR_2), Nipponbare that contains the Pish gene was inoculated with three of the eight isolates that showed strong association at the Pish locus (TZ-01, UG-05 and UG-11).

Nipponbare conferred resistance against all the three isolates, confirming that Pish is associated to RABR_2 and contributes to the resistance phenotype in some cultivars observed in our study. These results suggest that RABR_2 might be a new allele of the Pish gene.

Discussion

Rice blast is becoming an important disease in many rice-growing areas in Africa because of the lack of highly resistant rice cultivars, increased fertilizer application, and recent changes in climatic conditions. Germplasm that is resistant to the prevailing African

M. oryzae populations is therefore urgently needed. In this study, two subpopulations

(Indica and temperate Japonica) in RDP1 were used to map the R genes and QTLs for resistance to eight diverse isolates from four African countries. Thirty-one RABRs involved in resistance to these isolates were identified. Among them, 42% are associated with resistance to more than one isolate, and RABR_2 is associated with resistance to five isolates. These findings suggest that some RABRs may contain multiple R genes or QTL or a single locus in the region may confer resistance to multiple isolates. Seven of the

RABRs are co-localized with the blast resistance loci previously identified by Wang et al

(2014) and Kang et al (2015). Five RABRs (RABR_1, RABR_19, RABR_28, RABR_29 and

RABR_30) closely relate to loci associated with blast resistance (LABRs) previously reported by Kang et al. (2015). RABR-1 and RABR-28 were also reported by Wang et al

(2014). Twenty four RABRs are new. These results suggest that many of the rice loci associated with resistance against African populations may be different from the loci with resistance against non-African populations of M. oryzae. To determinate the relationship between M. oryzae isolates and RABRs, the study used the pathotype data (disease phenotypes) of 162 RPD1 cultivars inoculated with our eight African isolates and five isolates used by Kang et al (2015) to perform cluster analysis. The results showed low pathotype similarity level between the eight African isolates used in this study and those used by Kang et al (2015). Also, there was no clear relationship between the isolates similarity and the RABRs identified in this study and those (LABRs) identified by Kang et al (2015). However, it would be useful in the future to compare the eight African isolates used in this study and the five isolates used by Kang et al. (2015) at the DNA level in order to determinate the potential relevance between M. oryzae isolates and RABRs. Further fine

mapping will help to detect the precise physical locations of these 31 RABRs in the rice genome and to determine their relationship with other known R and QTL genes effective against non-African populations of M. oryzae.

Durable resistance to M. oryzae is a complex trait and involves in both complete and partial resistance genes. In a well-known example, the durably resistant upland rice cultivar

Moroberekan contains two R genes (Pi5 and Pi7) and ten QTLs (Wang et al. 1994). Recent

GWASs have also shown that resistance to M. oryzae isolates in China, South America,

India, Philippines, and Korea is controlled by multiple loci with R genes and minor QTL genes (Liu et al. 2013). In this study, 162 RPD1 cultivars were inoculated with eight M. oryzae isolates from four African countries. About 1/3 of the RABRs had LOD scores >

5.0. The effects of these RABRs on resistance to M. oryzae were as strong as those of R genes. The remaining 2/3 of the RABRs had low LOD scores and were considered to be

QTLs. The results confirm that resistance to M. oryzae populations in Africa is complex and is controlled by both major and minor genes. Although the functions of the candidate

QTLs associated with blast resistance require confirmation, detection of loci associated with partial resistance against African populations of M. oryzae provides starting materials for detailed molecular analysis. In addition, some of the SNP markers in those RABRs with high LOD scores (>5.0) can be used in marker-aided selection of blast-resistant cultivars in Africa.

Among the RABRs identified in this study, RABR_2 on chromosome 1 has the strongest association with blast resistance. The search of the reference Nipponbare genomic sequence revealed that RABR_2 is co-localized with the R gene Pish, which

encodes an NBS-LRR disease-resistance protein (Kang et al. 2015; Takahashi et al. 2010).

The PCR amplification of the RABR_2-related candidate gene in resistant and susceptible cultivars confirmed the association of the locus with resistance. RABR_2 is involved in resistance to five isolates from East Africa. Two of these isolates are from Tanzania (TZ-

01 and TZ-12), two are from Uganda (UG-05 and UG-11), and one is from Kenya (KE-

37). However, RABR_2 is not associated with resistance to either of the two isolates from

Burkina Faso (West Africa). These results suggest that RABR_2 is a good candidate R gene for control of rice blast in East Africa but not in West Africa.

Eleven cultivars that are highly resistant (disease score ≤ 2.0) to all eight of the tested isolates of M. oryzae from Africa were found. Among them, six are indica type and five are temperate japonica type. These varieties may contain multiple R genes and/or QTLs that are effective against African M. oryzae populations. Crossing these resistant rice cultivars with highly susceptible but popular African cultivars will facilitate the breeding of resistance against rice blast in Africa.

References

Abdurakhmonov, I. Y., and Abdukarimov, A. 2008. Application of association mapping to understanding the genetic diversity of plant germplasm resources. International Journal of Plant Genomics 2008:574927.

Bashir, Uzma, Sobia, M., and Naureen, A. 2014. First report of alternaria metachromatica from Pakistan causing leaf spot of tomato. Pakistan Journal of Agricultural Science 51:305-308.

Bourett, T. M., and Howard, R. J. 1990. In vitro development of penetration structures in the rice blast fungus Magnaporthe grisea. Canadian Journal of Botany 68:329-342.

Bradbury, P. J., Zhiwu, Z., Dallas, E. K., Terry, M. C., Yogesh, R., and Edward, S. B. 2007. TASSEL: software for association mapping of complex traits in diverse samples. Bioinformatics 23:2633-2635.

Buckler, E. S., Holland, J. B., Bradbury, P. J., Acharya, C. B., Brown, P. J., Browne, C., Ersoz, E., Flint-Garcia, S., Garcia, A., and Glaubitz, J. C. 2009. The genetic architecture of maize flowering time. Science 325:714-718.

Christiansen, T., Foy B.D., and L., W. 2012. Programming perl: Unmatched power for text processing and scripting. O'Reilly Media, Inc. Sebastopol, California, USA.

Eizenga, G. C., Ali, M., Bryant, R. J., Yeater, K. M., McClung, A. M., and McCouch, S. R. 2013. Registration of the rice diversity panel 1 for genomewide association studies. Journal of Plant Registrations 8:109-116.

FAO. 2012. The State of Food Insecurity in the World 2012: Economic growth is necessary but not sufficient to accelerate reduction of hunger and malnutrition. FAO, Rome. doi 10.

Govarthanan, M., Guruchandar, A., Arunapriya, S., Selvankumar, T., and Selvam, K. 2011. Genetic variability among Coleus sp. studied by RAPD banding pattern analysis. International Journal of Biotechnology and Molecular Biology Research 2:202- 208.

Habarurema, I., Asea, G., Lamo, J., Gibson, P., Edema, R., Séré, Y., and Onasanya, R. O. 2012. Genetic analysis of resistance to rice bacterial blight in Uganda. African Crop Science Journal 20:105 –112.

Henderson, C. R. 1975. Best linear unbiased estimation and prediction under a selection model. Biometrics 31:423-447.

Huang, X., Wei, X., Sang, T., Zhao, Q., Feng, Q., Zhao, Y., Li, C., Zhu, C., Lu, T., Zhang, Z., Li, M., Fan, D., Guo, Y., Wang, A., Wang, L., Deng, L., Li, W., Weng, Q., Liu, K., Huang, T., Zhou, T., Jing, Y., Li, W., Lin, Z., Buckler, E. S., Qian, Q., Zhang, Q., Li, J., and Han, B. 2010. Genome-wide association studies of 14 agronomic traits in rice landraces. Nature Genetics 42:961-967.

Imbe, and Matsumoto. 1985. Inheritance of resistance of rice varieties to the blast fungus strains virulent to the variety" Reiho". Japanese Journal of breeding 35:332-339.

IRRI. 2013. International Rice Research Institute. World Rice Statistics. Los Baños, the Philippines.

Jia, G., Huang, X., Zhi, H., Zhao, Y., Zhao, Q., Li, W., Chai, Y., Yang, L., Liu, K., and Lu, H. 2013. A haplotype map of genomic variations and genome-wide association studies of agronomic traits in foxtail millet (Setaria italica). Nature Genetics 45:957-961.

Kang, H., Yue, W., Shasha, P., Yanli, Z., Yinghui, X., Dan, W., Shaohong, Q., Zhiqiang, L., Shuangyong, Y., Zhilong, W., Wende, L., Yuese, N., Pavel, K., Hei, L., Jason, M., Susan, R. M., and Wang, G. L. 2015. Dissection of the genetic architecture of rice resistance to the blast fungus Magnaporthe oryzae. Molecular Plant Pathology: In press.

Kump, K. L., Bradbury, P. J., Wisser, R. J., Buckler, E. S., Belcher, A. R., Oropeza-Rosas, M. A., Zwonitzer, J. C., Kresovich, S., McMullen, M. D., and Ware, D. 2011. Genome-wide association study of quantitative resistance to southern leaf blight in the maize nested association mapping population. Nature Genetics 43:163-168.

Li, H., Peng, Z., Yang, X., Wang, W., Fu, J., Wang, J., Han, Y., Chai, Y., Guo, T., and Yang, N. 2013. Genome-wide association study dissects the genetic architecture of oil biosynthesis in maize kernels. Nature Genetics 45:43-50.

Lipka, A. E., Tian, F., Wang, Q., Peiffer, J., Li, M., Bradbury, P. J., Gore, M. A., Buckler, E. S., and Zhang, Z. 2012. GAPIT: genome association and prediction integrated tool. Bioinformatics 28:2397-2399.

Liu, W., Liu, J., Triplett, L., Leach, J. E., and Wang, G. L. 2014. Novel insights into rice innate immunity against bacterial and fungal pathogens. Annual Review of Phytopathology 52:213-241.

Liu, Y., Liu, B., Zhu, X., Yang, J., Bordeos, A., Wang, G., Leach, J. E., and Leung, H. 2013. Fine-mapping and molecular marker development for Pi56(t), a NBS-LRR gene conferring broad-spectrum resistance to Magnaporthe oryzae in rice. Theoretical and Applied Genetics 126:985-998.

Luzi-Kihupi, A., Zakayo, J., Tusekelege, H., Mkuya, M., Kibanda, N., Khatib, K., and Maerere, A. 2009. Mutation breeding for rice improvement in Tanzania. In Q. Y. Shu (Ed.), Induced Plant Mutations in the Genomics Era (pp. 385-387). Rome. Food and Agriculture Organization of the United Nations.

Mew, T. W., Leung, H., Savary, S., Vera Cruz, C. M., and Leach, J. E. 2004. Looking ahead in rice disease research and management. Critical Reviews in Plant Sciences 23:103-127.

Morris, G. P., Ramu, P., Deshpande, S. P., Hash, C. T., Shah, T., Upadhyaya, H. D., Riera- Lizarazu, O., Brown, P. J., Acharya, C. B., and Mitchell, S. E. 2013. Population genomic and genome-wide association studies of agroclimatic traits in sorghum. Proceedings of the National Academy of Sciences 110:453-458.

Mundt, C. C. 2014. Durable resistance: a key to sustainable management of pathogens and pests. Infection, Genetics and Evolution 27:446-455.

Muthayya, S., Sugimoto, J. D., Montgomery, S., and Maberly, G. F. 2014. An overview of global rice production, supply, trade, and consumption. Annals of the New York Academy of Sciences 1324:7-14.

Nayar, N. M. 2014. Rice in Africa. Pages 1-7 in: Encyclopaedia of the History of Science, Technology, and Medicine in Non-Western Cultures. H. Selin, ed. Springer, Dordrecht, Netherlands.

Park, C. H., Songbiao, C., Gautam, S., Bo, Z., Chang, H. K., Pattavipha, S., and al, A. J. A. e. 2012. The Magnaporthe oryzae effector AvrPiz-t targets the RING E3 Ubiquitin Ligase APIP6 to suppress pathogen-associated molecular pattern– triggered immunity in rice. The Plant Cell 24:4748-4762.

Poland, J. A., Bradbury, P. J., Buckler, E. S., and Nelson, R. J. 2011. Genome-wide nested association mapping of quantitative resistance to northern leaf blight in maize. Proceedings of the National Academy of Sciences 108:6893-6898.

Ribot, C., Hirsch, J., Balzergue, S., Tharreau, D., Nottéghem, J.-L., Lebrun, M.-H., and Morel, J.-B. 2008. Susceptibility of rice to the blast fungus, Magnaporthe grisea. Journal of Plant Physiology 165:114-124.

Savary, S., Horgan, F., Willocquet, L., and Heong, K. 2012. A review of principles for sustainable pest management in rice. Crop Protection 32:54-63.

Séré, Y., Fargette, D., Abo, M. E., Wydra, K., Bimerew, M., Onasanya, A., and Akator, S. K. 2013. Managing the Major Diseases of Rice in Africa. Realizing Africa's Rice Promise:213-228

Singh, S., Sidhu, J. S., Huang, N., Vikal, Y., Li, Z., Brar, D. S., Dhaliwal, H. S., and G.S, K. 2001. Pyramiding three bacterial blight resistance genes (xa5, xa13 and Xa21) using marker-assisted selection into indica rice cultivar PR106. Theoretical and Applied Genetics 102:1011–1015.

Takahashi, A., Hayashi, N., Miyao, A., and Hirochika, H. 2010. Unique features of the rice blast resistance Pish locus revealed by large scale retrotransposon-tagging. BMC Plant Biology 10:175.

Tharreau, D., Fudal, I., Andriantsimialona, D., Utami, D., Fournier, E., Lebrun, M.-H., and Nottéghem, J.-L. 2009. World population structure and migration of the rice blast fungus, Magnaporthe oryzae. Pages 209-215 in: Advances in Genetics, Genomics and Control of Rice Blast Disease. Springer, Dordrecht, Netherlands.

Tian, F., Bradbury, P. J., Brown, P. J., Hung, H., Sun, Q., Flint-Garcia, S., Rocheford, T. R., McMullen, M. D., Holland, J. B., and Buckler, E. S. 2011. Genome-wide association study of leaf architecture in the maize nested association mapping population. Nature Genetics 43:159-162.

Wang, C., Yang, Y., Yuan, X., Xu, Q., Feng, Y., Yu, H., Wang, Y., and Wei, X. 2014a. Genome-wide association study of blast resistance in indica rice. BMC Plant Biology 14:1-11.

Wang, G. L., Mackill, D. J., Bonman, J. M., McCouch, S. R., Champoux, M. C., and Nelson, R. J. 1994. RFLP mapping of genes conferring complete and partial resistance to blast in a durably resistant rice cultivar. Genetics 136:1421-1434.

Wang, X., Lee, S., Wang, J., Ma, J., Bianco, T., and Jia, Y. 2014b. Current advances on genetic resistance to rice blast disease. Pages 195-217 in: Rice Germplasm, Genetics and Improvement. W. Yan and J. Bao, eds. InTech, Rijeka, Croatia.

Williams, J. G. K., Kubelik, A. R., Livak, K. J., Rafalski, J. A., and Tingey, S. V. 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18:6531–6535.

Zeigler, R. S., Leong, S. A., and Teng, P. S. 1994. Rice blast disease. CAB International, Wallingford, UK, in association with the International Rice Research Institute, Los Baños, the Philippines.

Zhao, K., Tung, C. W., Eizenga, G. C., Wright, M. H., Ali, M. L., Price, A. H., Norton, G. J., Islam, M. R., Reynolds, A., Mezey, J., McClung, A. M., Bustamante, C. D., and McCouch, S. R. 2011. Genome-wide association mapping reveals a rich genetic architecture of complex traits in Oryza sativa. Nature Communications:1467.

Zhu, X. Y., Chen, S., Yang, J. Y., Zhou, S. C., Zeng, L. X., Han, J. L., and al., e. 2012. The identification of Pi50(t), a new member of the rice blast resistance Pi2/Pi9 multigene family. Theoretical and Applied Genetics 124:1295-1304.

PRIMER KIT NAME NAME SEQUENCE

OPG-08 TCACGTCCAC

KIT G OPG-20 TCTCCCTCAG OPJ-08 CATACCGTGG

KIT J OPJ-11 ACTCCTGCGA KIT N OPN-03 GGTACTCCCC

OPP-01 GTAGCACTCC OPP-03 CTGATACGCC KIT P OPP-09 GTGGTCCGCA KIT W OPW-03 GTCCGGAGTG

KIT AB OPAB-03 TGGCGCACAC

Table 1. Random Amplified Polymorphic DNA (RAPD) primer sequences used for M. oryzae variability study

Magnaporthe oryzae isolate TZ01 TZ12 UG05 UG11 KE14 KE37 BF05 BF27 % Resistance 17.3 34.0 30.2 29.6 35.8 53.1 17.4 13.6 % Susceptibility 22.2 14.2 21.6 31.5 16.7 8.6 22.4 39.5

Table 2. Summary of Rice Diversity Panel-1 disease phenotypes for eight African Magnaporthe oryzae isolates. Resistance=disease score < 2.0 and Susceptibility=disease score > 8.0

Resistant rice cultivars across eight Susceptible rice cultivars across eight Magnaporthe oryzae isolates Magnaporthe oryzae isolates GSOR ID NSFTV.ID Variety Name GSOR ID NSFTV.ID Variety Name 301015 NSFTV17 Binulawan 301009 NSFTV9 Baber 301030 NSFTV32 Chondongji 301070 NSFTV77 JC149 301065 NSFTV72 IR8 301107 NSFTV115 NPE 835 301075 NSFTV83 Kamenoo 301145 NSFTV154 Ta Hung Ku 301154 NSFTV163 Taducan 301146 NSFTV155 Ta Mao Tsao 301222 NSFTV232 Shangyu 394 301168 NSFTV177 68-2 301281 NSFTV291 Topolea 70/76 301210 NSFTV219 Nucleoryza 301416 NSFTV616 RT0034 301211 NSFTV220 Azerbaidjanica 301028 NSFTV30 Chiem Chanh 301279 NSFTV289 Lusitano 301386 NSFTV62 Gyehwa 3 301291 NSFTV301 Melanotrix 301274 NSFTV284 IR-44595 301292 NSFTV302 WIR 3039 301324 NSFTV334 Lomello 301237 NSFTV247 Desvauxii 301271 NSFTV281 Patna 301246 NSFTV256 Romanica

Table 3. Resistant (disease score ≤ 2.0) and susceptible (disease score ≥ 8.0) Rice Diversity Panel-1 (RDP1) rice genotypes to eight African Magnaporthe oryzae isolates.

Locus CHR LOD Value Known R gene Candidate genes RABR_1 1 3.860 Pi27(t) LOC_Os01g09850.1 RABR_2 1 13.612 Pi37, Pish LOC_Os01g57270.1 RABR_3 1 10.779 LOC_Os01g57350.1 RABR_4 2 4.424 LOC_Os02g05610.1 RABR_5 2 6.219 LOC_Os02g10900.1 RABR_6 2 6.459 LOC_Os02g13960.1 RABR_7 2 6.476 LOC_Os02g17400.1 RABR_8 2 8.648 LOC_Os02g41904.1 RABR_9 4 4.497 LOC_Os04g57090.1 RABR_10 4 5.688 LOC_Os04g54680.1 RABR_11 4 4.151 Pi21 LOC_Os04g12970.1 RABR_12 4 5.197 LOC_Os04g49460 .1 RABR_13 4 4.099 LOC_Os04g51830 .1 RABR_14 4 4.518 LOC_Os04g57390 .1 RABR_15 5 4.770 LOC_Os05g04450.1 RABR_16 5 3.972 LOC_Os05g24550.1 RABR_17 5 6.208 LOC_Os05g30220.1 RABR_18 6 4.388 LOC_Os06g14540.1 RABR_19 6 6.209 Pi50 LOC_Os06g15680.1 RABR_20 6 5.132 LOC_Os06g43620.1

Continued

Table 4. The regions associated with rice resistance (RABRs) to eight Magnaporthe oryzae isolates. Gene annotations for all the candidate genes corresponding to 31 RABRs are shown in Table 5.

Table 4 continued

RABR_21 6 5.451 LOC_Os06g43510.1 RABR_22 8 6.182 LOC_Os08g06510.1 RABR_23 8 7.492 LOC_Os08g44620.1 RABR_24 9 5.706 LOC_Os09g16980.1 RABR_25 10 4.372 LOC_Os10g06760.1 RABR_26 10 4.714 LOC_Os10g41900.1 RABR_27 11 5.076 LOC_Os11g08950.1 RABR_28 11 4.786 Pi-y(t) LOC_Os11g11810.1 LOC_Os11g11790.1 LOC_Os11g11950.1 LOC_Os11g11920.1 LOC_Os11g11940.1 RABR_29 11 4.493 Pi-34 LOC_Os11g40430.1 RABR_30 11 6.131 Pi-43 LOC_Os11g43500.1 LOC_Os11g43480.1 RABR_31 12 4.210 LOC_Os12g39070.1

Region/ Locus Candidate genes Annotation

RABR_1 LOC_Os01g09850.1 ZOS1-04 - C2H2 zinc finger protein, expressed NBS-LRR disease resistance RPP13-like RABR_2 LOC_Os01g57270.1 protein 1, putative, expressed

RABR_3 LOC_Os01g57350.1 Diacylglycerol kinase, putative, expressed

RABR_4 LOC_Os02g05610.1 ZOS2-03 - C2H2 zinc finger protein, expressed

RABR_5 LOC_Os02g10900.1 LRR-protein, putative, expressed

RABR_6 LOC_Os02g13960.1 Spotted leaf 11, putative, expressed

RABR_7 LOC_Os02g17400.1 LRR-protein, putative, expressed DEF7 - Defensin and Defensin-like DEFL RABR_8 LOC_Os02g41904.1 family, expressed Jasmonate O-methyltransferase, putative, RABR_9 LOC_Os04g57090.1 expressed ulp1 protease family protein, putative, RABR_10 LOC_Os04g54680.1 expressed UDP-glucoronosyl/UDP-glucosyl transferase, RABR_11 LOC_Os04g12970.1 putative, expressed Protein kinase domain containing protein, RABR_12 LOC_Os04g49460 .1 expressed

RABR_13 LOC_Os04g51830 .1 OsHKT1;4 - Na+ transporter, expressed

RABR_14 LOC_Os04g57390 .1 Acyl-protein thioesterase, putative, expressed

RABR_15 LOC_Os05g04450.1 Peroxidase precursor, putative, expressed Papain family cysteine protease domain RABR_16 LOC_Os05g24550.1 containing protein, expressed

Continued

Table 5. Gene annotations for the candidate genes corresponding to the thirty-one regions associated with rice resistance (RABRs) to eight Magnaporthe oryzae isolates.

Table 5 continued

NBS-LRR disease resistance RPP13-like RABR_17 LOC_Os05g30220.1 Protein 1, putative, expressed

RABR_18 LOC_Os06g14540.1 Endoglucanase, putative, expressed

RABR_19 LOC_Os06g15680.1 Cytochrome P450 71A6, putative, expressed

RABR_20 LOC_Os06g43620.1 Haemolysin-III, putative, expressed

RABR_21 LOC_Os06g43510.1 Cytochrome P450 71D6, putative, expressed Zinc finger, C3HC4 type domain containing RABR_22 LOC_Os08g06510.1 Protein, expressed

RABR_23 LOC_Os08g44620.1 Zinc-binding protein, putative, expressed

RABR_24 LOC_Os09g16980.1 OsWAK86 - OsWAK pseudogene, expressed LRR-Receptor-like protein kinase precursor, RABR_25 LOC_Os10g06760.1 putative, expressed LRR-disease resistance mla1 protein, putative, RABR_26 LOC_Os10g41900.1 expressed Protein kinase family protein, putative, RABR_27 LOC_Os11g08950.1 expressed NBS-LRR disease resistance protein, putative, RABR_28 LOC_Os11g11810.1 expressed NBS-LRR type disease resistance protein, LOC_Os11g11790.1 putative, expressed NBS-LRR disease resistance protein RPM1, LOC_Os11g11950.1 putative, expressed NB-ARC domain disease resistance protein, LOC_Os11g11920.1 putative, expressed NB-ARC domain containing protein MLA10, LOC_Os11g11940.1 putative, expressed Cell wall-associated receptor kinase-like 2 RABR_29 LOC_Os11g40430.1 precursor, putative, expressed NBS-LRR type disease resistance protein, RABR_30 LOC_Os11g43500.1 putative, expressed LOC_Os11g43480.1 LRR family protein, expressed

RABR_31 LOC_Os12g39070.1 TATA-binding protein, putative, expressed

OLIGO'S NAME SEQUENCE N1F TGAGGGTCCCAGTGTACCAA N1R TGCGTGAATCAACTAGCCCC N3R CTTATCCCCCTGGTTGCGTG 5' 6R AGGCAAGCTGCTACTGTTGT 5' 4R TCATGTCCCTCAGTTGTCGC 3' 7F ACAACAGTAGCAGCTTGCCT 3' 7R CAGGGAACAGAGTGGACAGC N10F GCTTGCGATCTGCTACCTCT N10R CGGACAGACCACAGGACTAC

Table 6. List of primers used for Pish gene amplification. The primers were designed using Pish gene sequence as a reference.

Figure 1. Genetic relationship among the eight Magnaporthe oryzae isolates. A. An electrophoresis image of Random Amplified Polymorphic DNA (RAPD) amplification with primer P-01 separated in a 1.4% agarose gel. The DNA amplifications and gel electrophoresis were conducted three times for each primer and identical results were obtained. Only the clearly amplified polymorphic bright bands were analyzed. Numbers at the top correspond to isolates: 1=TZ-01, 2=TZ-12, 3=UG-05, 4=UG-11, 5=KE14, 6=KE- 37, 7=BF-05 and 8=BF-27. Lane M=1kb molecular marker. (TZ = Tanzania; UG = Uganda; KE = Kenya; BF = Burkina Faso) B. A dendrogram of the eight M. oryzae isolates created using Minitab’s cluster analysis function computed with Euclidean Distance Complete Linkage. The scale indicates the genetic similarity in percentage.

Figure 2. An example of disease score distribution (0-9) of the Rice Diversity Panel-1 (RDP1) cultivars inoculated with Tanzanian Magnaporthe oryzae isolate TZ-01 (A) and TZ-12 (B). The experiment was conducted two times and the disease scores of the two biological replicates were similar.

Figure 3. Rice blast disease score distribution (0-9) of the Rice Diversity Panel-1 (RDP1) cultivars inoculated with Kenyan Magnaporthe oryzae isolates KE-14 (A) and KE-37 (B). The experiment was conducted two times and the disease scores of the two biological replicates were similar.

Figure 4. Rice blast disease score distribution (0-9) of the Rice Diversity Panel-1 (RDP1) cultivars inoculated with Ugandan Magnaporthe oryzae isolates UG-05 (A) and UG-11 (B). The inoculation experiment performed twice and the disease scores of the two biological replicates were similar.

Figure 5. Rice blast disease score distribution (0-9) of the Rice Diversity Panel-1 (RDP1) cultivars inoculated with Burkina Faso Magnaporthe oryzae isolates BF-05 (A) and BF-27 (B). The inoculation experiment was performed twice and the disease scores of the two biological replicates were similar.

Relationship among blast isolates

56.68

) 71.12

(

85.56

100.00 R01-1 RB22 75-1-127 P06-6 O-249 KE-14 KE-37 TZ-01 TZ-12 UG-05 UG-11 BF-05 BF-27 Isolates

Figure 6. Cluster analysis of the thirteen Magnaporthe oryzae isolates based on disease scores of the 162 Rice Diversity Panel-1 (RDP1) cultivars. The first (left to right) five isolates were previously used by Kang et al (2015) while the last eight were used in this study. The scale indicates the genetic similarity in percentage. The M. oryzae isolates from Tanzania (TZ-01 and TZ-12) and those from Uganda (UG-05 and UG-11) are grouped in the same cluster but different from those from Kenyan (KE-37 and KE-14) and Burkina Faso (BF-05 and BF-27). The two Kenyan isolates are grouped together as well as the two isolates from Burkina Faso forming a total of three clusters based on genetic relatedness.

Figure 7. Principal component analysis of the eight Magnaporthe oryzae isolates based on disease scores of the 162 Rice Diversity Panel-1 (RDP1) cultivars. The M. oryzae isolates from Tanzania (TZ-01 and TZ-12) and those from Uganda (UG05 and UG11) are grouped together but separated from the Kenyan (KE37 and KE14) and Burkina Faso (BF05 and BF27) isolates.

Figure 8. A combined Manhattan plot summarizing Genome Wide Association Study (GWAS) results for the eight Magnaporthe oryzae isolates. The X-axis is the genomic position of each SNP, and the Y-axis is the negative logarithm of the P-value obtained from the GWAS model (specifically from the F-test for testing H0: No association between the single nucleotide polymorphism (SNP) and trait). SNPs with strong associations for the trait have higher Y-coordinate value. The red arrows indicate that identified regions associated with blast resistance (RABRs) are co-localized with previously mapped or cloned R gene regions while blue triangles indicates shared RABRs among two isolates from Tanzania (TZ-01 and TZ-12) and blue diamonds indicate shared RABRs among two M. oryzae isolates from Uganda (UG-05 and UG-11).

Figure 9. Genome-wide association analysis for two Kenyan Magnaporthe oryzae isolates. (A) Manhattan plots that summarizes Genome Wide Association Study (GWAS) results. The X-axis is the genomic position of each single nucleotide polymorphism (SNP), and the Y-axis is the negative logarithm of the P-value obtained from the GWAS model (specifically from the F-test for testing H0: No association between the SNP and trait). SNPs with strong associations for the trait (blast disease resistance) have higher Y- coordinate value. (B) Quantile-Quantile plots that shows fitness of the GWAS analysis model for the two isolates. The X-axis corresponds to the expected values of negative logarithm of P and the Y-axis corresponds to the observed values of negative logarithm of P obtained from the GWAS analysis.

Figure 10. Genome-wide association analysis for two Burkina-Faso Magnaporthe oryzae isolates. (A) Manhattan plots that summarizes Genome Wide Association Study (GWAS) results. The X-axis is the genomic position of each single nucleotide polymorphism (SNP), and the Y-axis is the negative logarithm of the P-value obtained from the GWAS model (specifically from the F-test for testing H0: No association between the SNP and trait). SNPs with strong associations for the trait (blast disease resistance) have higher Y- coordinate value. (B) Quantile-Quantile plots that shows fitness of the GWAS analysis model for the two isolates. The X-axis corresponds to the expected values of negative logarithm of P and the Y-axis corresponds to the observed values of negative logarithm of P obtained from the GWAS analysis.

Figure 11. Genome-wide association analysis for two Tanzanian Magnaporthe oryzae isolates. (A) Manhattan plots that summarizes Genome Wide Association Study (GWAS) results. The X-axis is the genomic position of each single nucleotide polymorphism, and the Y-axis is the negative logarithm of the P-value obtained from the GWAS model (specifically from the F-test for testing H0: No association between the SNP and trait). SNPs with strong associations for the trait (blast disease resistance) have higher Y- coordinate value. (B) Quantile-Quantile plots that shows fitness of the GWAS analysis model for the two isolates. The X-axis corresponds to the expected values of negative logarithm of P and the Y-axis corresponds to the observed values of negative logarithm of P obtained from the GWAS analysis.

B UG-05 UG-11

Figure 12. Genome-wide association analysis results of the Rice Diversity Panel-1 (RDP1) cultivars to two Ugandan Magnaporthe oryzae isolates. (A) Manhattan plots that summarizes Genome Association Study (GWAS) results for two Ugandan isolates (UG- 05 and UG-11). The X-axis is the genomic position of each single nucleotide polymorphism (SNP), and the Y-axis is the negative logarithm of the P-value obtained from the GWAS model (specifically from the F-test for testing H0: No association between the SNP and trait). SNPs with strong associations for the trait (blast disease resistance) have higher Y-coordinate value. (B) Quantile-Quantile plots that shows fitness of the GWAS analysis model for the two isolates. The X axis corresponds to the expected values of negative logarithm of P and the Y-axis corresponds to the observed values of negative logarithm of P obtained from the GWAS analysis.

R S A B 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 + - N1F + N1R

N1F + N3R

N1F + 5’ 6R

N1F + 5’ 4R

33’’ 7F ++ 53’’ 7R7R

N10F + N10R

Figure 13. Gel images of the amplified Pish gene fragments and the regions of designed primers used for amplification. (A) Twenty-five temperate japonica (TEJ) rice cultivars were used in the PCR analysis. The primer sequence can be found in Table 6. Lanes 01- 10 contain amplicons from varieties resistant to blast and lanes 11-25 are from varieties susceptible to blast; lane (+) is a Nipponbare DNA as a positive control and lane (-) is water as negative control. (B) Schematic diagram showing the location of each amplicon (primer) in the Pish gene coding region. The molecular weight for the five primers are; N1F + N1R=938 bp, N1F+N3R=952 bp, N1F+5’ 6R=1,336 bp, N1F+5’ 4R=1,661 bp, 3’ 7F+5’ 7R= 2,343 bp and N10F+N10R=895 bp.

CHAPTER 3

GBS-Based Diversity Analysis of African Rice Cultivars and Association Mapping

of Rice Blast Resistance Genes

Introduction

Rice is one of the important cereal crops worldwide feeding more than 50% of the global human population (Khush 2005). In Africa, rice consumption has dramatically increased mostly in urban areas. The current estimated per capital consumption of rice in

Africa is 23.3kg per annum, double that of early 1970’s (FAO 2012; Muthayya et al. 2014).

Despite this increase in rice consumption, rice production sector in Africa is facing many biotic and abiotic constraints such as drought, low soil fertility, soil salinity, pests and diseases. Molecular breeding and biotechnology plays a vital role in improving important agronomic traits including resistance to pests and disease (Stewart and Ow 2008).

Therefore, for sustainable rice breeding and production, detailed understanding and proper management of rice genetic resources and diversity is significantly important (Guimaraes

2002). Africa has a tremendous wealth of rice genetic resources that can be utilized to expand the genetic base of rice varieties that are widely grown by farmers across the continent and make them well adapted to abiotic and biotic stresses.

The history of rice cultivation in Africa goes back to the first century AD when rice was introduced for the first time in East Africa by Indian traders (Harlan and Stemler 1976).

Later on in the middle of the second century AD, the crop was introduced in West Africa by the Portuguese (Portères 1962). While O. glaberrima is only mainly cultivated in West

Africa, O. sativa is cultivated all across the continent where rice is grown. Generally, O. glaberrima has yield and genetic diversity compared to O. sativa. However, O. glaberrima is more adapted to drastic conditions than its cousin O. sativa and can tolerate or resist soil salinity, iron toxicity, pests and diseases (Linares 2002). For these reasons, O. glaberrima can be utilized as a valuable resource for O. sativa improvement through interspecific hybridization (Orjuela et al. 2014).

Conservation and utilization of African rice genetic resources are under the custody of the Africa Rice Center (AfricaRice), Benin. Over several years, AfricaRice has been collecting rice germplasm from all across Africa to ensure that these valuable genetic materials are conserved for sustainable utilization in rice improvement programs across the continent and beyond. About 2,500 O. glaberrima and 17,000 O. sativa accessions from all across Africa have been conserved in AfricaRice, Benin (Sanni et al. 2013).

Several studies were conducted on molecular diversity analysis of African rice cultivars over the past years. By using 93 simple sequence repeat (SSR) markers, Semon et al. (2004) analyzed the population structure of 198 O. glaberrima accessions and identified five genetically diverse groups. The accessions in the first group are floating type originated from the upper delta of river Niger. The second group is non-floating (lowland) cultivated in Nigeria believed to have originated from areas along river Niger. The third

group is upland type originating from Liberia. These first three groups are linked to the three ecotypes previously reported by Portères (1970). The fourth and fifth groups share ancestry with other two subspecies of O. sativa (Indica and Japonica, respectively)

(Portères 1970; Sanni et al. 2013; Semon et al. 2005). These results point out that some

O. glaberrima accessions possess some degree of admixtures (Sanni et al. 2013). Two separate studies conducted by Barry et al. (2007) and Dramé et al. (2011) using SSR markers on both O. glaberrima and O. sativa accessions from West Africa reported similar findings. In these studies three major ecotypes were identified; floating type grown in deep water, erect type grown in upland areas and an intermediate group between O. sativa and

O. glaberrima, suggesting the occurrence of O. glaberrima × O. sativa recombination.

These findings suggest that, despite the reproductive barrier, hybrid forms may occur in nature producing intermediate genotypes between O. sativa and O. glaberrima (Barry et al. 2007; Dramé et al. 2013; Semon et al. 2005).

Another extensive analysis of the African rice genetic diversity was done by Orjuela et al. (2014) using global genotyping. In their study, 279 O. glaberrima accessions were genotyped and a set of 235 SNPs was used to structure the individual accessions in two major populations groups. The two populations are not linked to any previous phenotyped trait. Misclassification in O. glaberrima was also identified. Interestingly, Orjuela et al.

(2014) identified a new form of O. sativa from the set of African varieties (Orjuela et al.

2014).

These findings and many others show that the African rice cultivars (both O. glaberrima and O. sativa) have a rich diversity that can be further explored. Current

advances in molecular technologies have simplified the process of analyzing species diversity both in terms of time and cost. For example, next generation sequencing (NGS) technologies are currently used for whole genome sequencing projects where a high number of sample genomes are sequenced to explore large number of single nucleotide polymorphisms (SNPs). These SNPs can be used in studying population diversity, constructing haplotype maps and performing genome-wide association studies (GWAS)

(Metzker 2010). Recent development of simple and highly multiplexed, genotyping-by- sequencing (GBS) approach, which is based on NGS, has been used as a molecular tool in conducting various studies in population diversity, germplasm characterization, breeding, and trait mapping in complex and diverse organisms (Elshire et al. 2011; Orjuela et al.

2014). These new molecular tools can be used to explore Africa rice diversity and identify novel genes that can be used in breeding programs to produce varieties that can resist against major rice production constraints such as rice diseases in Africa. Rice blast disease, caused by the fungus Magnaporthe oryzae (Bourett and Howard 1990; Ribot et al. 2008), is one of the major setbacks for rice production in Africa. Use of resistant varieties seems to be the most promising approach for blast disease management (Savary et al. 2012) and

Africa has rich rice genetic resources that can be utilized in identifying new resistance (R) genes to M. oryzae. Monogenic resistance that provides complete resistance against specific races of M. oryzae exhibiting a gene-for-gene interaction has been proven to be easily broken making breeding for resistant varieties a repeating exercise. Contrary to complete resistance, partial resistance controlled by multiple small effect genes or quantitative trait loci (QTLs) is more durable even under high pathogen population

pressure (Mundt 2014). Thus, in order to significantly improve rice blast resistance in elite cultivars, it is necessary to utilize rice germplasm that contains both major and minor genes combining both vertical and horizontal resistance.

GWAS uses natural populations or germplasm collections and linkage disequilibrium

(LD)-based association to map target genes quickly in a large collection of diverse genotypes. It addresses the shortcomings of traditional gene mapping using bi-parental crosses and efficiently maps with high resolution all the loci that are associated with the trait across the whole genome (Abdurakhmonov and Abdukarimov 2008). GWAS can be used to dissect complex traits in crops (Huang et al. 2010; Jia et al. 2013) and identify multiple genes that are associated with those traits.

In our previous study, we used 162 cultivars of the rice diversity panel 1 (RDP1) containing 36,900 SNPs and eight M. oryzae isolates to identify 31 genomic regions associated with blast resistance (RABRs) in the rice genome. Seven of these RABRs were linked to known R gene loci, and 24 were new. Some of the loci associated with blast resistance (LABRs) are strongly associated with R and defense genes encoding NBS-LRR, defense-related proteins, transcription factors, and receptor-like protein kinases. Here I present the findings of a study on GBS-based diversity analysis of African rice cultivars and association mapping of rice blast resistance genes.

In this study, 190 African rice cultivars from AfricaRice and the Tanzania Ministry of

Agriculture germplasm center were used. These cultivars were genotyped by sequencing and generated 184K SNPs that were used for diversity analysis and association mapping of rice blast resistance genes. Six rice M. oryzae isolates from Africa were used to inoculate

the 190 African rice cultivars. In the diversity study three major population groups were identified. This study also detected the misclassification of some cultivars. Association mapping showed that 25 RABRs in the rice genome were associated with blast disease resistance against the six isolates. PCR analysis revealed that a major RABR on chromosome 12 is highly linked with resistance to four M. oryzae isolates. This study demonstrates the usefulness of GBS in diversity analysis and effectiveness of GWAS for quick identification of R/QTLs genes in rice and provides SNP markers which are highly linked for immediate use in breeding for resistance against rice blast in Africa.

Materials and methods

Plant and fungal materials

The 190 O. glaberrima, O. sativa and New Rice for Africa (NERICA, hybrids of O. glaberrima and O. sativa) accessions used in this study were obtained from AfricaRice and the Tanzania Ministry of Agriculture (Katrin Agricultural Research Institute) germplasm.

These cultivars were collected from all African countries where rice is grown and represent the geographical distribution and ecologies of the cultivated rice species in Africa (Table

7). Seed increase for these cultivars was done in a greenhouse at The Ohio State University in the summer of 2013. De-husked seeds were placed in small 15 ml conical tubes and sterilized with 70% ethanol for 1 minute and then 40% Clorox for 30 minutes. The seeds were then washed with sterilized water and transferred into Petri dish containing ½ MS medium for germination. The ½ MS medium was prepared by adding 2.2g Murashige &

Skoog salt, 30g sucrose, and 2.2g phytagel in 1L distilled water. The pH was adjusted to 57

5.6 pH using 1M HCl and KOH solutions. Five days after germination, the plantlets were transplanted into sterilized soil in small pots and kept in a growth chamber. Nutrient solution including iron was added weekly until the plants were ready for inoculation. Single spore isolation of M. oryzae from infected samples was done using water agar and oatmeal agar media. Leaf samples were cut into small pieces and placed on wet filter papers in small sterile container to induce sporulation. The container was incubated under florescent light at room temperature for 24 hours. Fungal spores of each sample were observed under a light microscope, picked using a sterile loop and placed in a Petri dish containing water agar medium. The plates were incubated for 2 days at room temperature (25°C) in the dark.

The growing hyphal tips from each single spore mycelial colony were cut and transferred to oatmeal agar media separately and placed in an incubator (25°C and white florescent light) for 7 days to induce sporulation. Each culture plate colonized with M. oryzae fungi was considered an isolate and assigned a name code. These fungal isolates were then propagated in new sterile plates with oatmeal medium. Sterile Whatman filter papers were placed in the plates and left for 5 days to be colonized by fungal mycelia. The papers were then desiccated and transferred into sterilized vial tubes for long term storage of fungal isolates at -20°C. The selected six M. oryzae isolates include: TZ-01 and TZ-12 from

Tanzania, UG-05 and UG-11 from Uganda, KE-37 from Kenya, and BF-07 from Burkina

Faso.

DNA extraction and GBS

For GBS, at least 50μl of 50-100 ng/ul total genomic DNA was extracted from leaf tissues of a single plant per each accession for all the 190 cultivars using Qiagen DNeasy

DNA extraction kit as per manufacturer’s protocol. The DNA concentration for each sample was quantified using a NanoDrop ND-8000 spectrophotometer. Thirty microlitter of DNA for each of the 190 cultivars were sent to Cornell University Institute of

Biotechnology in two 96-wells plates where GBS was done using a 96-plex ApeKI GBS protocol. For SNPs calling, individual libraries (i.e. cultivars) were identified by their unique names. The reads were then aligned to the O. glaberrima and O. sativa reference genomes using Burrows-Wheeler transform (BWA) (Li and Durbin 2010). The resulting

BAM alignment files were sorted in order to facilitate the variant calling step. The

"samtools pileup’’, "bcftools view" and "vcfutils.pl varFilter" commands (using a max depth of 1000) were used in a pipeline to call the SNPs (SAMTOOLS) (Li et al. 2009).

SNPs calling using both O. glaberrima and O. sativa reference genomes generated 184K

SNPs.

GBS based diversity

A total of 184K SNP markers generated from GBS were used to assess the genetic diversity of the 190 African rice cultivars. Genetic distances between the cultivars was calculated using TASSEL (Bradbury et al. 2007b), and a phylogenetic tree was constructed using ggtree R package (G Yu et al. 2016).

PCR analyses

For confirmation of GBS-based diversity results, PCR analysis was conducted using rice subspecies-specific primers (Table 8). Thirty-eight randomly chosen varieties from the

190 African rice accessions were used. Three O. glaberrima (Tog 5603, Tog 6126 and Tog

84117), two O. sativa (NPB (japonica) and CO-39 (indica)) were included as controls

(Table 9). Amplifications were performed in a 25-μl reaction mixture consisting of 50 ng/μl genomic DNA, 1X reaction buffer (Promega), 0.25 mM dNTPs, 0.2 μM random primer,

2.5 μM MgCl2, and 1 unit of Taq polymerase. The amplification included one denaturing cycle of 4 min at 94°C; followed by 45 cycles of 1 min at 94°C, 1 min at 40°C, and 1 min at 72°C; and a final extension step of 2 min at 72°C. The amplified products were resolved by electrophoresis on a 1.4% agarose gel using TAE buffer (45 mM Tris-acetate, 1 mM

EDTA, pH 8.0) at 100 volts for 2 h. A 1-kb ladder was included as a molecular size marker.

Gels were stained with an ethidium bromide solution (0.5 μg/ml), and band patterns were visualized with UV light. To investigate the association of the Pita gene to observed resistance phenotypes, PCR analysis of RABR_23 was done using Pita specific primers.

DNA from 10 cultivars with resistance phenotype and 10 cultivars with susceptible phenotype were used for the analysis (Table 10). The PCR conditions were set to; 3 minutes at 95°C for denaturation followed by 33 cycles for 30 seconds at 95°C, 1 min at 58°C for primers annealing, 1 min for extension step at 72°C followed by final extension step for 5 min at 72°C. The amplified products and 1-kb ladder were loaded on 1.4% agarose gel and electrophoresis run in TAE buffer (45 mM Tris-acetate, 1 mM EDTA, pH 8.0) at 120 volts

for 1 hour. Ethidium bromide solution was used for gel staining and band patterns were visualized under UV light.

Evaluation of blast resistance phenotypes

Rice seedlings were germinated and fungal cultures were grown and prepared for spray inoculation as described by Park et al. (2012). About fifteen rice seedlings per cultivar per pot (experimental unit) for all the 190 cultivars were grown in a growth chamber conditioned for 26°C temperature for 12 hours of light (day) and 21°C for 12 hours darkness (night), 80% to 90% humidity as described by Park et al. (2012). Seedlings with three to four fully expanded leaves (18-21 days after germination) were sprayed with fungal spore suspensions with spore concentration of 5x105 conidia/ml in 0.1% Tween-20 until run-off. Spore concentration was quantified using a hemocytometer under microscope observation. The inoculated seedlings were kept in a plastic container (≥90% humidity) for

24 hours and then transferred to the growth chamber (Park et al. 2012). Disease scoring was done 6 days post inoculation using a 0-9 blast scoring scale (IRRI, 1996, where 0 indicates no blast symptoms (highly resistant) and 9 indicates severe blast symptoms

(highly susceptible) (Zhu et al. 2012). The inoculation experiment was performed twice under the same conditions. If the results from two experiments were different, a third experiment was performed and data from the two similar experiments were used. R statistical package was used for data analysis. The cluster analysis and principal components analysis (PCA) were conducted based on the interactions between rice genotypes and M. oryzae isolates.

Association analysis

Phenotypic data (blast disease scores of 190 African rice accessions) and the 184K-

SNP dataset generated by GBS were utilized to perform GWAS as previously described by Zhao et al. (2011). Mixed linear model (MLM) (Bradbury et al. 2007b) was employed using Tassel 5.0 software (http://www.maizegenetics.net/tassel/) to run the analysis. A compressed Mixed Linear Model (MLM) was conducted, taking into account the kinship

(K) matrix and population structure (Q) matrices. (Henderson 1975; Lipka et al. 2012).

TASSEL results were utilized as input files to combine multiple Manhattan plots into an integrated one (combined Manhattan plot) using homemade Perl scripts based on PERL

(Christiansen et al. 2012) and its SVG module (scalable vector graphics).

Results

GBS-based diversity of African rice cultivars

Based on the results, the 190 African cultivars were grouped into three population clusters, two major ones and a minor cluster that consists of only few cultivars (Fig. 14).

Surprisingly, the groupings were independent of ecologies and not specific to subspecies categories. The first group (Grp I) was the most diverse and mostly comprised of cultivars originated from Tanzania (TZLRs) and few from the AfricaRice germplasm center. The majority of cultivars in this group are classified as O. sativa although some few varieties that are classified as O. glaberrima were also found in this group. Moreover, the majority of NERICA varieties (hybrids of O. glaberrima and O. sativa) are also clustered together within this group. Group II and III have fewer number of cultivars compared to group I and

mostly comprised of varieties obtained from the AfricaRice germplasm (ARs) center,,with a few from Tanzania.

To confirm the results of genetic diversity using GBS data and detect possible misclassification of African rice cultivars at the subspecies level, we performed PCR analysis using primers designed from genomic regions specific to O. glaberrima. The PCR analysis revealed that some of the cultivars classified as O. glaberrima were actually O. sativa and vice versa. Two cultivars, WAR42-82-2-3-1 and Sahel-217 classified as O. sativa were amplified by O. glaberrima specific primer set (M14) (Fig. 15, lane 8 and 23 respectively) but not amplified with O. sativa primers. Similarly, one cultivar, RAM-100 classified as O. glaberrima was amplified by O. sativaspecific primers (M3) (Fig. 15, lane

14) but not amplified with O. glaberrima primers. The interspecific hybrids (NERICAs) were amplified with both O. glaberrima and O. sativa specific primers (Fig. 15, lane 17,

18 and 20). The control (known) varieties for both O. sativa and O. glaberrima gave expected amplifications (Fig. 15, lane 38, 39, 40, 41 and 42). This experiment was conducted three times with consistent results.

Blast resistance phenotypes

The blast disease score distributions of the 190 African rice cultivars inoculated with the six M. oryzae isolates are shown in Fig. 16, 17 and 18. The disease scores for all the isolates were evenly distributed (range 0-9) with a mean score range of 3-5. The percentages of cultivars that were resistant and susceptible to each isolate and mean disease scores are shown in Table 11. Based on the inoculation phenotypes of the 190 African

cultivars, 15 were highly resistant (disease score ≤ 2.0) to all six isolates (Table 12). Cluster analysis based on disease phenotypes of the inoculated rice cultivars indicates that the six isolates were diverse from each other and independent of their origin (Fig. 19). Only the two isolates from Uganda (UG-05 and UG-11) were clustered together (≈93% similarity) and somehow related to TZ-01 from Tanzania. Interestingly, one isolate from Tanzania

(TZ-12) was clustered together with Burkina Faso isolate (BF-07).

Identification of rice QTLs associated with resistance to six isolates

Using genotypic (SNPs data set derived from GBS) and phenotypic (blast disease scores) data sets, we conducted GWAS and identified 25 non-redundant RABRs (LOD ≥

4.0) that were significantly associated with resistance phenotypes in the African rice cultivars against the six M. oryzae isolates. RABR_3, RABR_20, RABR_22, and RABR_23 are located in the regions with 4 previously mapped R genes, Pish, Pi-y(t), Pi-6 and Pita, respectively.

Two of these four RABRs (RABR_3 and RABR_20) and other two (RABR_7 and

RABR_19) were also identified in our previous study using 162 accessions from the Rice

Diversity Panel 1 (RDP1). The other 19 RABRs are novel candidate genes (Fig. 20).

RABR-7 and RABR-8 were associated with all the isolates (from East Africa) except isolate

BF-07 (from Burkina Faso). Thirteen RABRs were at least shared with two isolates.

However, there was no association between shared RABRs and an isolate’s country of origin.

The 25 RABRs were distributed across 12 rice chromosomes (Fig 21, 22 and 23). The highest number of RABRs per chromosome was four, which were observed on chromosomes 1, 2 and 12. Three RABRs (RABR_22, RABR_23 and RABR_24) were clustered together on chromosome 12. These RABRs were strongly associated with resistance to four isolates (UG-05, UG-11, TZ-12 and BF-07). RABR_23 had the highest

SNP signals in response to isolate TZ-12 (LOD ˃20) (Fig. 24). We further identified candidate R genes associated with the 25 RABRs by exploring the reference Nipponbare genomic sequence (MSU V7.0) using gene ontology (GO) terms. Twenty five R or R- related candidate genes were identified using annotation analysis of the 25 RABRs (Table

13). The 25 candidate genes included NBS-LRR genes, receptor-like protein kinases, protein phosphorylation related genes, transcription factors, ubiquitination-related genes and DNA-binding genes. The identified RABRs may have different biological functions associated with both qualitative and quantitative resistance against rice blast pathogens.

RABR_23 is linked to Pita R gene

The strong association of RABR_23 with resistance to four isolates incited us to perform a detailed analysis of this genomic region on chromosome 12. Sequence analysis showed that RABR_23 region is closely linked to the Pita gene. Pita belongs to the NBS-

LRR domain-containing class of R gene (Bryan et al. 2000) and provides broad spectrum resistance against different isolates of M. oryzae in India and other rice growing countries in Asia (Ramkumar et al. 2014). Using Pita as a reference sequence, three primer pairs were designed (Table 15) to amplify the sequence fragment of the candidate gene in 10 resistant and 10 susceptible cultivars that were selected from our pathogenicity assay. Pita5 65

primer set was able to amplify a fragment related to Pita R gene in eight out of ten blast- resistant rice cultivars and two out of ten blast-susceptible rice cultivars (Fig. 25). To confirm the association of Pita gene to a major resistance locus on chromosome 12

(RABR_23), Tadukan rice variety that harbors Pita gene was inoculated with three isolates that showed strong association at the Pita locus (TZ-12, UG-05 and UG-11). Tadukan showed resistance against all the three isolates, confirming that Pita is associated to

RABR_23 and contributes significantly to the resistance phenotype in some cultivars observed in an inoculation assay of the 190 African rice cultivars. These results suggest that RABR_23 might be a new allele of Pita R gene and may be utilized in rice blast resistance breeding programs.

Discussion

The importance of preserving genetic materials for sustainable rice crop productivity cannot be underestimated. With increased rice crop intensification and rapid evolution of strains of pathogens, monogenic resistance can easily be broken down over time and we often need to go back to our genetic pool to identify QTLs/R genes that can be utilized for resistance breeding programs. It is therefore necessary to understand in detail the diversity of our genetic materials for proper management of genetic resources (Guimaraes 2002).

Africa has a huge wealth of rice genetic resources that can be exploited for genetic improvement of popular rice varieties (Sanni et al. 2013) widely grown all across the continent against biotic and abiotic challenges.

A number of studies conducted on molecular diversity analysis of African rice cultivars over the past years have revealed that African rice is diverse but often

misclassified (Orjuela et al. 2014; Semon et al. 2005). By using GBS, a diversity analysis of 190 African rice cultivars including O. sativa, O. glaberrima and interspecific hybrid

NERICAs was conducted. The analysis classified the cultivars into three major groups.

However, the groupings were independent of ecologies and varieties in different subspecies were mixed in all the three clusters. This could be due to misclassification of these cultivars or possible admixtures that could have happened in handling seeds, or may be a result of natural hybridization between O. sativa and O. glaberrima (Barry et al. 2007; Dramé et al. 2013). The PCR results using O. glaberrima- and O. sativa- specific primers addressed our concern of possible misclassification of African rice cultivars. Previous studies reported the occurrence of natural hybrids between O. sativa and O. glaberrima (Barry et al. 2007; Semon et al. 2005). Later on, following a series of phenotypic and genotypic screening, these cultivars were proven to be true O. sativa and not interspecific hybrids

(Garris et al. 2005). However, they were neither classified as O. sativa spp indica nor O. sativa spp japonica and might have been introduced a long time ago in Africa from Asia

(Orjuela et al. 2014). Our PCR results show that, some O. glaberrima cultivars were amplified by O. sativa-specific primers (M3) and not by O. glaberrima-specific primers

(M14) (Fig. 15). Likewise some O. sativa cultivars were amplified by O. glaberrima- specific primers (M14) and not by O. sativa-specific primers (M3). These results show the possibility that most of the African rice cultivars are misclassified in terms of their naming and ecologies. This raises a flag for a need to conduct studies with the aim of properly classifying the African germplasm that would boost its value as genetic stock and enable sustainable and easy exploitation of their genetic pool.

A big challenge facing African rice productivity is pests and diseases. Rice blast is one of the important diseases in many rice-growing areas of Africa. Lack of highly resistant rice cultivars, climate change and rice intensification makes the disease even more devastating. Identification of new resistance genes in African rice germplasm is highly needed. This study used a diverse African rice population with 190 cultivars of different origins and ecologies to map R genes and QTLs that confers resistance to rice blast disease.

Using six M. oryzae isolates from Tanzania, Uganda, Kenya and Burkina Faso, 25 RABRs associated to blast disease resistance were identified. About 50% of them were associated with resistance to two or more isolates. These findings point to two possibilities, either one

RABR may contain multiple R genes/QTL or a single locus may confer resistance to multiple isolates.

Six of the 25 RABRs are co-localized with blast resistance loci previously identified in the paper by Kang et al (2015) and also in previous association study using a different rice population. Three of the RABRs (RABR_20, RABR_22, and RABR_23) closely relate to loci associated with blast resistance (LABRs) previously reported by Kang et al. (2015).

RABR-20 was also reported by Wang et al (2014). Nineteen RABRs are new and only unique to this study. The results suggest that African rice populations may contain many blast resistance loci that are different from other non-African populations providing an avenue for new R gene exploration and utilization in breeding programs.

Many association mapping studies have shown that resistance to M. oryzae isolates is controlled by multiple loci with R genes and minor QTL genes (Liu et al. 2013). In this study, about 2/3 of the RABRs had LOD scores > 5.0, which implies that their contribution

to resistance phenotypes are as strong as those of major R genes. The remaining 1/3 of the

RABRs had LOD scores of < 4 and can be considered as QTLs. This illustrates the complexity of both African rice germplasm and M. oryzae populations and that the resistance phenotypes observed are controlled by both R genes and minor QTLs. While the functional analysis of QTLs associated with rice blast partial resistance is absolutely crucial, their identification sets a foundation for detailed molecular analysis and subsequent use as markers in breeding programs in Africa.

RABR_23 on chromosome 12 showed very strong association with blast resistance.

This locus was found to be linked with the Pita gene. PCR amplifications of the RABR_23- related candidate gene in ten resistant and ten susceptible cultivars confirmed the association of the locus with resistance phenotype. Moreover, two susceptible cultivars were amplified with Pita specific primers and two resistant cultivars were not amplified.

This shows that Pita is not the only locus contributing to resistance, hence other R genes/QTLs may play a role. RABR_23 is associated with resistance to three isolates from

East Africa (UG-05, UG-11 and TZ-12) and one isolate from West Africa (BF-07) suggesting its usefulness as a candidate R gene for managing rice blast in East and West

Africa. Based on the pathogenicity results, 15 cultivars were highly resistant (disease score

≤ 2.0) to all six M. oryzae isolates. These varieties may contain multiple R genes and/or

QTLs that are effective against African M. oryzae populations and useful in facilitating the breeding of resistance against rice blast in Africa

References

AfricaRice. 2012. Africa Rice Centre (AfricaRice) Annual Report 2011: A new rice research for development strategy for Africa. Cotonou, Benin.

Barry, M. B., Pham, J. L., Noyer, J. L., Billot, C., Courtois, B., and Ahmadi, N. 2007. Genetic diversity of the two cultivated rice species (O. sativa & O. glaberrima) in Maritime Guinea. Evidence for interspecific recombination. Euphytica 154:127- 137.

Bashir, Uzma, Sobia, M., and Naureen, A. 2014. First report of alternaria metachromatica from Pakistan causing leaf spot of tomato. Pakistan Journal of Agricultural Science 51:305-308.

Bourett, T. M., and Howard, R. J. 1990. In vitro development of penetration structures in the rice blast fungus Magnaporthe grisea. Canadian Journal of Botany 68:329-342.

Bradbury, P. J., Zhang, Z., Kroon, D. E., Casstevens, T. M., Ramdoss, Y., and Buckler, E. S. 2007a. TASSEL: software for association mapping of complex traits in diverse samples. Bioinformatics 23:2633-2635.

Bradbury, P. J., Zhiwu, Z., Dallas, E. K., Terry, M. C., Yogesh, R., and Edward, S. B. 2007b. TASSEL: software for association mapping of complex traits in diverse samples. Bioinformatics 23:2633-2635.

Bryan, G. T., Wu, K.-S., Farrall, L., Jia, Y., Hershey, H. P., McAdams, S. A., Faulk, K. N., Donaldson, G. K., Tarchini, R., and Valent, B. 2000. A single amino acid difference distinguishes resistant and susceptible alleles of the rice blast resistance gene Pita. The Plant Cell 12:2033-2045.

Christiansen, T., Foy B.D., and L., W. 2012. Programming perl: Unmatched power for text processing and scripting. O'Reilly Media, Inc. Sebastopol, California, USA.

Dean, R. A. 1997. Signal pathways and appressorium morphogenesis. Annual Review of Phytopathology 35:211-234.

Dramé, K. N., Sanchez, I., Gregorio, G., and Ndjiondjop, M. N. 2013. Suitability of a selected set of simple sequence repeats (SSR) markers for multiplexing and rapid molecular characterization of African rice (Oryza glaberrima Steud.). African Journal of Biotechnology 10:6675-6685.

Elshire, R. J., Glaubitz, J. C., Sun, Q., Poland, J. A., Kawamoto, K., Buckler, E. S., and Mitchell, S. E. 2011. A Robust, Simple Genotyping-by-Sequencing (GBS) Approach for High Diversity Species. PLoS ONE 6:e19379.

FAO. 2012. The State of Food Insecurity in the World 2012: Economic growth is necessary but not sufficient to accelerate reduction of hunger and malnutrition. FAO, Rome. doi 10.

FAOSTAT data. 2012. FAOSTAT.http://faostat.fao.org/site/339/default.aspx. Retrieved June 12, 2016

Yu G., Smith D., Zhu H., Guan Y., and Lam T. 2016. ggtree: an R package for visualization and annotation of phylogenetic tree with different types of meta-data. Methods in Ecology and Evolution: In press.

Garris, A. J., Tai, T. H., Coburn, J., Kresovich, S., and McCouch, S. 2005. Genetic structure and diversity in Oryza sativa L. Genetics 169:1631-1638.

Guimaraes, E. P. 2002. Genetic diversity of rice production in Brazil. In: Nguyen, V.N. (ed.) Genetic Diversity in Rice Production, Case Studies from Brazil, India and

Nigeria. Food and Agriculture Organization of the United Nations (FAO), Rome, Italy:11-35.

Hamer, J. E., Howard, R. J., Chumley, F. G., and Valent, B. 1988. A mechanism for surface attachment in spores of a plant pathogenic fungus. Science 239:288-290.

Harlan, J. R., and Stemler, A. 1976. The races of sorghum in Africa. In: Harlan, J.R., De Wet, J.M. and Stemler, A.B. (eds) Origin of African Plant Domestication. Mouton, The Hague, Netherlands.

Henderson, C. R. 1975. Best linear unbiased estimation and prediction under a selection model. Biometrics 31:423-447.

Imbe, and Matsumoto. 1985. Inheritance of resistance of rice varieties to the blast fungus strains virulent to the variety" Reiho". Japanese Journal of breeding 35:332-339.

IRRI. 2013. International Rice Research Institute. World Rice Statistics: . Los Baños, the Philippines: IRRI. June 29, 2013.

Khush, G. S. 2005. What it will take to feed 5.0 billion rice consumers in 2030. Plant Molecular Biology 59:1-6.

Li, H., and Durbin, R. 2010. Fast and accurate long-read alignment with Burrows–Wheeler transform. Bioinformatics 26:589-595.

Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, G., and Durbin, R. 2009. The sequence alignment/map format and SAMtools. Bioinformatics 25:2078-2079.

Linares, O. F. 2002. African rice (Oryza glaberrima): history and future potential. Proceedings of the National Academy of Sciences 99:16360-16365.

Liu, W., Liu, J., Triplett, L., Leach, J. E., and Wang, G. L. 2014. Novel insights into rice innate immunity against bacterial and fungal pathogens. Annu Rev Phytopathol 52:213-241.

Metzker, M. L. 2010. Sequencing technologies - the next generation. Nature Reviews. Genetics 11:31-46.

Mew, T. W., Leung, H., Savary, S., Vera Cruz, C. M., and Leach, J. E. 2004. Looking ahead in rice disease research and management. Critical Reviews in Plant Sciences 23:103-127.

Moldenhauer, K. A., and Gibbons, J. H. 2003. Rice morphology and development. Rice: Origin, History, Technology, and Production. Hoboken, NJ. John Wiley and Sons.

Mundt, C. C. 2014. Durable resistance: a key to sustainable management of pathogens and pests. Infection, Genetics and Evolution 27:446-455.

Muthayya, S., Sugimoto, J. D., Montgomery, S., and Maberly, G. F. 2014. An overview of global rice production, supply, trade, and consumption. Annals of the New York Academy of Sciences 1324:7-14.

Nayar, N. M. 2014. Rice in Africa. Pages 1-7 in: Encyclopaedia of the History of Science, Technology, and Medicine in Non-Western Cultures. H. Selin, ed. Springer, Dordrecht, Netherlands.

Orjuela, J., Sabot, F., Chéron, S., Vigouroux, Y., Adam, H., Chrestin, H., Sanni, K., Lorieux, M., and Ghesquière, A. 2014. An extensive analysis of the African rice genetic diversity through a global genotyping. Theoretical and Applied Genetics 127:2211-2223.

Portères, R. 1962. Berceaux agricoles primaires sur le continent africain. The Journal of African History 3:195-210.

Portères, R. 1970. Primary cradles of agriculture in the African continent. Papers in African Prehistory:43-58.

Project, I. R. G. S. 2005. The map-based sequence of the rice genome. Nature 436:793- 800.

Rossman, A. Y., Howard, R. J., and Valent, B. 1990. Pyricularia grisea, the Correct Name for the Rice Blast Disease Fungus. Mycologia 82:509-512.

Sanni, K. A., Tia, D. D., Ojo, D. K., Ogunbayo, A. S., Sikirou, M., and Hamilton, N. R. S. 2013. 7 Diversity of Rice and Related Wild Species in Africa. Realizing Africa's Rice Promise:87.

Savary, S., Horgan, F., Willocquet, L., and Heong, K. 2012. A review of principles for sustainable pest management in rice. Crop protection 32:54-63.

Semon, M., Nielsen, R., Jones, M. P., and McCouch, S. R. 2005. The population structure of African cultivated rice Oryza glaberrima (Steud.) evidence for elevated levels of linkage disequilibrium caused by admixture with O. sativa and ecological adaptation. Genetics 169:1639-1647.

Séré, Y., Fargette, D., Abo, M. E., Wydra, K., Bimerew, M., Onasanya, A., and Akator, S. K. 2013. Managing the Major Diseases of Rice in Africa. Realizing Africa's Rice Promise:213-228

Stewart, C. N., and Ow, D. W. 2008. The Future of Plant Biotechnology. Plant Biotechnology and Genetics: Principles, Techniques, and Applications:357-369.

Swaminathan, M. S. 1984. Rice in 2000 AD. In: Abrol and Sulochana Gadgil (eds.), Rice in a variable climate. APC Publications Pvt. Ltd., New Delhi-110005, India.

Takahashi, A., Hayashi, N., Miyao, A., and Hirochika, H. 2010. Unique features of the rice blast resistance Pish locus revealed by large scale retrotransposon-tagging. BMC plant biology 10:175.

Talbot, N. J. 2003. On the trail of a cereal killer: exploring the biology of Magnaporthe grisea. Annual Reviews in Microbiology 57:177-202.

Timsina, J., and Connor, D. 2001. Productivity and management of rice–wheat cropping systems: issues and challenges. Field Crops Research 69:93-132.

USDA, A., National Genetic Resources Program. 2012. Germplasm Resources Information Network – (GRIN) [Online Database]. National Germplasm Resources Laboratory, Beltsville, Maryland. www.ars-grin.gov/cgi-bin/npgs/html/index.pl (accessed 12 May 2016).

Valent, B., and Chumley, F. G. 1991. Molecular genetic analysis of the rice blast fungus, Magnaporthe grisea. Annual Revision of Phytopathology 29:443-467.

Wang, C., Yang, Y., Yuan, X., Xu, Q., Feng, Y., Yu, H., Wang, Y., and Wei, X. 2014a. Genome-wide association study of blast resistance in indica rice. BMC Plant Biology. 14:1-11.

Wang, X., Lee, S., Wang, J., Ma, J., Bianco, T., and Jia, Y. 2014b. Current advances on genetic resistance to rice blast disease. Wengui Yan (Ed) 1501:70.

Wang, X., Lee, S., Wang, J., Ma, J., Bianco, T., and Jia, Y. 2014c. Current advances on genetic resistance to rice blast disease. Pages 195-217 in: Rice Germplasm, Genetics and Improvement. W. Yan and J. Bao, eds. InTech, Rijeka, Croatia.

Way, M., and Heong, K. 1994. The role of biodiversity in the dynamics and management of insect pests of tropical irrigated rice-A review. Bulletin of Entomological Research 84:567-588.

Zeigler, R. S., Leong, S. A., and Teng, P. S. 1994. Rice blast disease. CAB International, Wallingford, UK, in association with the International Rice Research Institute, Los Baños, the Philippines

Zhu, X. Y., Chen, S., Yang, J. Y., Zhou, S. C., Zeng, L. X., and Han, J. L. 2012. The identification of Pi50(t), a new member of the rice blast resistance Pi2/Pi9 multigene family. Theoretical and Applied Genetics 124:1295-1304.

S/N OSU Code Cultivar Name Ecology Sub-species Source 1 AR-1 Bouake 189 Lowland rainfed Sativa AfricaRice-Benin 2 AR-10 ITA 257 Upland Sativa AfricaRice-Benin 3 AR-100 Balanta Mangrove Sativa AfricaRice-Benin 4 AR-101 VENUTOHE Upland Glaberrima AfricaRice-Benin 5 AR-102 EKOUDZI Upland Glaberrima AfricaRice-Benin 6 AR_103 DANYI-MOLI Upland Glaberrima AfricaRice-Benin 7 AR-105 EWENTOHI Upland Glaberrima AfricaRice-Benin 8 AR-106 EWINTO YIBO Upland Glaberrima AfricaRice-Benin 9 AR-107 EWEMOLI Upland Glaberrima AfricaRice-Benin 10 AR-109 EWINTOHI Upland Glaberrima AfricaRice-Benin 11 AR-11 Kogoni (91-1) Irrigated Sativa AfricaRice-Benin 12 AR-110 WONKIFONG Mangrove Sativa AfricaRice-Benin 13 AR-12 Moroberekan Upland Sativa AfricaRice-Benin 14 AR-13 NERICA 4 Upland Interspecific AfricaRice-Benin 15 AR-14 NERICA-L 19 Lowland rainfed Interspecific AfricaRice-Benin 16 AR-15 Sahel 108 Irrigated Sativa AfricaRice-Benin 17 AR-16 Sahel 159 Irrigated Sativa AfricaRice-Benin 18 AR-17 Sahel 177 Irrigated Sativa AfricaRice-Benin 19 AR-18 Sahel 208 Irrigated Sativa AfricaRice-Benin 20 AR-19 Sahel 209 Irrigated Sativa AfricaRice-Benin 21 AR-2 CG 14 Upland Glaberrima AfricaRice-Benin 22 AR-20 Sahel 217 Irrigated Sativa AfricaRice-Benin 23 AR-22 Sahel 305 Irrigated Sativa AfricaRice-Benin 24 AR-23 Sahel 328 Irrigated Sativa AfricaRice-Benin 25 AR-24 Sahel 329 Irrigated Sativa AfricaRice-Benin 26 AR-27 WAB128-B-B-13-HB Upland Sativa AfricaRice-Benin 27 AR-28 WAB176-B-8-HB Upland Sativa AfricaRice-Benin 28 AR-3 FKR 16 (4456) Lowland rainfed Sativa AfricaRice-Benin 29 AR-30 WAB217-B-B-2-HB Upland Sativa AfricaRice-Benin 30 AR-32 WAB272-B-B-1-H1 Upland Sativa AfricaRice-Benin

Continued

Table 7. Origin and ecologies of 190 African rice cultivars used in genome wide association study (GWAS) and genotype-by-sequencing (GBS)-based diversity study.

Table 7 continued

31 AR-34 WAB306-B-B-6-L2-L1-LB Upland Sativa AfricaRice-Benin 32 AR-35 WAB307-B-B-B1-L3-L1-LB Upland Sativa AfricaRice-Benin 33 AR-36 WAB337-B-B7-H4 Upland Glaberrima AfricaRice-Benin 34 AR-37 WAB365-B-2-H1-HB Upland Sativa AfricaRice-Benin 35 AR-38 WAB384-B-11-H2-H1-HB Upland Sativa AfricaRice-Benin 36 AR-39 WAB492-81-2 Upland Sativa AfricaRice-Benin 37 AR-4 FKR 19 (Tox 728-1) Lowland rainfed Sativa AfricaRice-Benin 38 AR-40 WAB497-25-1 Upland Sativa AfricaRice-Benin 39 AR-41 WAB506-125-3 Upland Sativa AfricaRice-Benin 40 AR-42 WAB519-55-3 Upland Sativa AfricaRice-Benin 41 AR-43 WAB570-35-53 Upland Sativa AfricaRice-Benin 42 AR-44 WAB583-6-1 Upland Sativa AfricaRice-Benin 43 AR-46 WAB706-2-K1-K1 Upland Sativa AfricaRice-Benin 44 AR-47 WAB96-1-1 Upland Sativa AfricaRice-Benin 45 AR-48 WAR102-1-2-1 Mangrove Sativa AfricaRice-Benin 46 AR-49 WAR42-82-2-3-1 Mangrove Sativa AfricaRice-Benin 47 AR-5 Gambiaka kokum Mali Lowland rainfed Sativa AfricaRice-Benin 48 AR-50 WAR77-3-2-2 Mangrove Sativa AfricaRice-Benin 49 AR-52 WAS161-B-9-3 Irrigated Sativa AfricaRice-Benin 50 AR-53 WAT100-TGR-2-4 Lowland rainfed Sativa AfricaRice-Benin 51 AR-54 WAT107-TGR-2-2 Lowland rainfed Sativa AfricaRice-Benin 52 AR-55 WAT307-WAS-B-24-8-4-4-2 Lowland rainfed Sativa AfricaRice-Benin 53 AR-56 WAT311-WAS-B-B-23-7-1 Lowland rainfed Sativa AfricaRice-Benin 54 AR-57 WAT317-WAS-B-B-55-4-3 Lowland rainfed Sativa AfricaRice-Benin 55 AR-6 FKR 54 (WABIR 12979) Lowland rainfed Sativa AfricaRice-Benin 56 AR-60 WAT46-TGR-1-2 Lowland rainfed Sativa AfricaRice-Benin 57 AR-61 WAT50-TGR-4-1 Lowland rainfed Sativa AfricaRice-Benin 58 AR-63 WIT 10 Lowland rainfed Sativa AfricaRice-Benin 59 AR-64 WITA 11 Lowland rainfed Sativa AfricaRice-Benin 60 AR-65 WITA 12 Lowland rainfed Sativa AfricaRice-Benin 61 AR-66 WITA 2 Lowland rainfed Sativa AfricaRice-Benin 62 AR-67 WITA 3 Lowland rainfed Sativa AfricaRice-Benin

Continued

Table 7 continued

63 AR-68 WITA 4 Lowland rainfed Sativa AfricaRice-Benin 64 AR-7 IRAT 104 Upland Sativa AfricaRice-Benin 65 AR-71 WITA 7 Lowland rainfed Sativa AfricaRice-Benin 66 AR-72 WITA 8 Lowland rainfed Sativa AfricaRice-Benin 67 AR-73 WITA 9 Lowland rainfed Sativa AfricaRice-Benin 68 AR-74 NERICA 1 Upland Interspecific AfricaRice-Benin 69 AR-75 NERICA 2 Upland Interspecific AfricaRice-Benin 70 AR-77 NERICA 5 Upland Interspecific AfricaRice-Benin 71 AR-78 NERICA 6 Upland Interspecific AfricaRice-Benin 72 AR-79 NERICA 7 Upland Interspecific AfricaRice-Benin 73 AR-8 ITA 123 Upland Sativa AfricaRice-Benin 74 AR-80 NERICA 8 Upland Interspecific AfricaRice-Benin 75 AR-81 NERICA 9 Upland Interspecific AfricaRice-Benin 76 AR-82 NERICA 10 Upland Interspecific AfricaRice-Benin 77 AR-83 NERICA 11 Upland Interspecific AfricaRice-Benin 78 AR-84 NERICA 12 Upland Interspecific AfricaRice-Benin 79 AR-85 NERICA 13 Upland Interspecific AfricaRice-Benin 80 AR-86 NERICA 14 Upland Interspecific AfricaRice-Benin 81 AR-88 NERICA 16 Upland Interspecific AfricaRice-Benin 82 AR-89 NERICA 17 Upland Interspecific AfricaRice-Benin 83 AR-9 ITA 150 Upland Sativa AfricaRice-Benin 84 AR-90 NERICA 18 Upland Interspecific AfricaRice-Benin 85 AR-91 IDSA 6 Upland Sativa AfricaRice-Benin 86 AR-92 IDSA 85 Upland Sativa AfricaRice-Benin 87 AR-95 RAM 130 Deep-water Glaberrima AfricaRice-Benin 88 AR-96 RAM 1 Deep-water Glaberrima AfricaRice-Benin 89 AR-97 RAM 100 Deep-water Glaberrima AfricaRice-Benin 90 AR-98 RAM 154 Deep-water Glaberrima AfricaRice-Benin 91 AR-99 CK 4 Irrigated Sativa AfricaRice-Benin 92 CO-39 Suscept. Check Irrigated Sativa Katrin-Tanzania 93 EN-10 IR 77713 Irrigated Sativa Katrin-Tanzania

Continued

Table 7 continued

94 EN-11 IR 79511 Irrigated Sativa Katrin-Tanzania INTSINDAGIRA 95 EN-12 BIGEGA Irrigated Sativa Katrin-Tanzania 96 EN-13 INTSINZI Irrigated Sativa Katrin-Tanzania 97 EN-14 RUMBUKA Lowland rainfed Sativa Katrin-Tanzania 98 EN-15 IR 05N221 Irrigated Sativa Katrin-Tanzania 99 EN-16 IR 03A262 Irrigated Sativa Katrin-Tanzania 100 EN-18 IR 07A167 Irrigated Sativa Katrin-Tanzania 101 EN-2 BKN/SUPA Lowland rainfed Sativa Katrin-Tanzania 102 EN-20 IR 06A107 Irrigated Sativa Katrin-Tanzania 103 EN-22 IR O3A550 Irrigated Sativa Katrin-Tanzania 104 EN-24 WITA 9 Lowland Sativa Katrin-Tanzania 105 EN-25 K5 Lowland Sativa Katrin-Tanzania EDIGET(WAB189-B-B- 106 EN-3 B-HB) Irrigated Sativa Katrin-Tanzania 107 EN-4 ROJOMENA 271/10 Irrigated Sativa Katrin-Tanzania 108 EN-5 IR 80482 Irrigated Sativa Katrin-Tanzania 109 EN-6 IR 77080 Irrigated Sativa Katrin-Tanzania 110 EN-7 HUA 565 Irrigated Sativa Katrin-Tanzania 111 EN-9 FRX 92-14 Irrigated Sativa Katrin-Tanzania 112 L-NER-13 LOW LAND NERICA 13 Lowland rainfed Interspecific Katrin-Tanzania 113 L-NER-32 LOW LAND NERICA 32 Lowland rainfed Interspecific Katrin-Tanzania 114 L-NER-33 LOW LAND NERICA 33 Lowland rainfed Interspecific Katrin-Tanzania 115 L-NER-35 LOW LAND NERICA 35 Lowland rainfed Interspecific Katrin-Tanzania 116 L-NER-46 LOW LAND NERICA 46 Lowland rainfed Interspecific Katrin-Tanzania 117 L-NER-48 LOW LAND NERICA 48 Lowland rainfed Interspecific Katrin-Tanzania 118 L-NER-50 LOW LAND NERICA 50 Lowland rainfed Interspecific Katrin-Tanzania 119 L-NER-52 LOW LAND NERICA 52 Lowland rainfed Interspecific Katrin-Tanzania 120 L-NER-56 LOW LAND NERICA 55 Lowland rainfed Interspecific Katrin-Tanzania 121 L-NER-59 LOW LAND NERICA 59 Lowland rainfed Interspecific Katrin-Tanzania 122 L-NER-6 LOW LAND NERICA 6 Lowland rainfed Interspecific Katrin-Tanzania 123 L-NER-60 LOW LAND NERICA 60 Lowland rainfed Interspecific Katrin-Tanzania 124 L-NER-7 LOW LAND NERICA 7 Lowland rainfed Interspecific Katrin-Tanzania 125 L-NER-8 LOW LAND NERICA 8 Lowland rainfed Interspecific Katrin-Tanzania 126 TORIDE Resist Check Irrigated Sativa Katrin-Tanzania 127 TZLR-1 JARIBU Lowland rainfed Sativa Katrin-Tanzania

Continued 81

Table 7 continued

128 TZLR-10 LINGWELINGWELI Lowland rainfed Sativa Katrin-Tanzania 129 TZLR-11 SINDANO KUBWA Lowland rainfed Sativa Katrin-Tanzania 130 TZLR-13 KIA LA NGAWA Upland Sativa Katrin-Tanzania 131 TZLR-15 KIHOGO RED MOR Upland Sativa Katrin-Tanzania 132 TZLR-16 CHAMBENA Upland Sativa Katrin-Tanzania 133 TZLR-17 MBAWA YA NJIWA Upland Sativa Katrin-Tanzania 134 TZLR-18 TONDOGOSO Upland Sativa Katrin-Tanzania 135 TZLR-19 FAYA MAFUTA Upland Sativa Katrin-Tanzania 136 TZLR-2 ZAMBIA Lowland Sativa Katrin-Tanzania 137 TZLR-21 MWANZA Lowland rainfed Sativa Katrin-Tanzania 138 TZLR-22 MPAKA WA BIBI Lowland rainfed Sativa Katrin-Tanzania 139 TZLR-23 RANGI MBILI Lowland rainfed Sativa Katrin-Tanzania 140 TZLR-24 LIFUMBA Lowland rainfed Sativa Katrin-Tanzania 141 TZLR-25 GOMBE Lowland rainfed Sativa Katrin-Tanzania 142 TZLR-27 RINGA MSONGA Lowland rainfed Sativa Katrin-Tanzania 143 TZLR-3 AFAA MWANZA 1/159 Lowland rainfed Sativa Katrin-Tanzania 144 TZLR-30 MSONGA Lowland rainfed Sativa Katrin-Tanzania 145 TZLR-31 PISHORI (BROWN) Irrigated Sativa Katrin-Tanzania 146 TZLR-32 MLEKE ALONGOLE Upland Sativa Katrin-Tanzania MBAWAMBILI 147 TZLR-33 MWEKUNDU Upland Sativa Katrin-Tanzania 148 TZLR-34 NONDO Upland Sativa Katrin-Tanzania 149 TZLR-35 RANGI MBILI NYEKUNDU Upland Sativa Katrin-Tanzania 150 TZLR-36 SUKARI Lowland rainfed Sativa Katrin-Tanzania 151 TZLR-37 GAMTI Lowland rainfed Sativa Katrin-Tanzania 152 TZLR-38 USINIGUSE Lowland rainfed Sativa Katrin-Tanzania 153 TZLR-40 KALING'ANAULA Lowland rainfed Sativa Katrin-Tanzania 154 TZLR-41 SI MZITO Lowland rainfed Sativa Katrin-Tanzania 155 TZLR-43 KALUNDI Lowland rainfed Sativa Katrin-Tanzania 156 TZLR-44 SUPA KIJIVU Lowland rainfed Sativa Katrin-Tanzania 157 TZLR-45 SUPA Lowland rainfed Sativa Katrin-Tanzania 158 TZLR-47 AFAA KIKANGAGA Lowland rainfed Sativa Katrin-Tanzania 159 TZLR-50 LOYA Lowland rainfed Sativa Katrin-Tanzania 160 TZLR-51 SHINGO YA MWALI Lowland rainfed Sativa Katrin-Tanzania

Continued

Table 7 continued

161 TZLR-53 KALIVUMBULA Lowland rainfed Sativa Katrin-Tanzania 162 TZLR-54 LUNYUKI Lowland rainfed Sativa Katrin-Tanzania 163 TZLR-55 KATUMAHI Lowland rainfed Sativa Katrin-Tanzania 164 TZLR-57 TUNDURU Upland Sativa Katrin-Tanzania 165 TZLR-58 JAMBO TWENDE Lowland rainfed Sativa Katrin-Tanzania 166 TZLR-6 MZUNGU Lowland rainfed Sativa Katrin-Tanzania 167 TZLR-60 LIMOTA Lowland rainfed Sativa Katrin-Tanzania MBAWAMBILI 168 TZLR-62 RANGIMBILI Upland Sativa Katrin-Tanzania 169 TZLR-63 KISEGESE Lowland rainfed Sativa Katrin-Tanzania 170 TZLR-64 MOSHI Lowland rainfed Sativa Katrin-Tanzania 171 TZLR-65 UMANHO Lowland rainfed Sativa Katrin-Tanzania 172 TZLR-66 KIHOGO RED Upland Sativa Katrin-Tanzania 173 TZLR-67 MASANTULA Lowland rainfed Sativa Katrin-Tanzania 174 TZLR-68 MWARABU Lowland rainfed Sativa Katrin-Tanzania 175 TZLR-7 SOTEA Lowland rainfed Sativa Katrin-Tanzania 176 TZLR-70 MWASUNGO Lowland rainfed Sativa Katrin-Tanzania 177 TZLR-71 MBAWAMBILI Upland Sativa Katrin-Tanzania UROO 1 KIKUSYA 178 TZLR-73 (IMPROVED) Irrigated Sativa Katrin-Tanzania 179 TZLR-74 TXD 306 (IMPROVED) Irrigated Sativa Katrin-Tanzania 180 TZLR-75 TXD 85 (IMPROVED) Irrigated Sativa Katrin-Tanzania 181 TZLR-76 TXD 88 (IMPROVED) Irrigated Sativa Katrin-Tanzania 182 TZLR-77 SUPA BC (IMPROVED) Irrigated Sativa Katrin-Tanzania 183 TZLR-78 FAYA DUME-1 Upland Sativa Katrin-Tanzania 184 TZLR-79 FAYA DUME-2 Upland Sativa Katrin-Tanzania 185 TZLR-8 SIFARA Lowland rainfed Sativa Katrin-Tanzania 186 TZLR-80 FAYA DUME-3 Upland Sativa Katrin-Tanzania 187 TZLR-81 FAYA DUME-4 Upland Sativa Katrin-Tanzania 188 TZLR-82 FAYA DUME-5 Upland Sativa Katrin-Tanzania 189 TZLR-83 MWANGAZA Irrigated Sativa Katrin-Tanzania 190 TZLR-9 SUPA SURUNGAI Lowland rainfed Sativa Katrin-Tanzania

OLIGO'S NAME SEQUENCE S11004AF (M3) TCTCTGGCCTTCTACTCATGG S11004AR (M3) TTGTGTTTCTACTTGGACTCTTTTT SINEI-r2F (M14) GGAGGACGTGCAGATCGTTC SINEI-r2R (M14) TTGCCCGGATACTTCTCCTC UBQF CGCAAGTTCGCTTCTTGACC UBQR GTGGAGCTACCTGAGTCACA

Table 8. List of specific primer pairs used for rice subspecies identification in PCR amplification

S/N/Lane OSU Code Cultivar Name Ecology Sub-species 1 AR_43 WAB570-35-53 Upland Sativa 2 AR_19 Sahel 209 Irrigated Sativa

3 AR_14 NERICA-L 19 Lowland rainfed Interspecific 4 AR_71 WITA 7 Lowland rainfed Sativa 5 TZLR_54 LUNYUKI Lowland rainfed Sativa 6 AR_11 Kogoni (91-1) Irrigated Sativa EDIGET(WAB189-B-B-B- 7 EN_3 HB) Irrigated Sativa 8 AR_49 WAR42-82-2-3-1 Mangrove Sativa 9 EN_5 IR 80482 Irrigated Sativa 10 AR_37 WAB365-B-2-H1-HB Upland Sativa 11 EN_4 ROJOMENA 271/10 Irrigated Sativa 12 EN_24 WITA 9 Lowland Sativa 13 AR_53 WAT100-TGR-2-4 Lowland rainfed Sativa 14 AR_97 RAM 100 Deep-water Glaberrima 15 TZLR_3 AFAA MWANZA 1/159 Lowland rainfed Sativa 16 TZLR_74 TXD 306 (IMPROVED) Irrigated Sativa

17 L_NER_48 LOW LAND NERICA 48 Lowland rainfed Interspecific 18 L_NER_13 LOW LAND NERICA 13 Lowland rainfed Interspecific 19 EN_2 BKN/SUPA Lowland rainfed Sativa 20 L_NER_56 LOW LAND NERICA 55 Lowland rainfed Interspecific 21 EN_18 IR 07A167 Irrigated Sativa 22 AR_23 Sahel 328 Irrigated Sativa 23 AR_20 Sahel 217 Irrigated Sativa 24 EN_11 IR 79511 Irrigated Sativa 25 AR_90 NERICA 18 Upland Interspecific 26 AR_34 WAB306-B-B-6-L2-L1-LB Upland Sativa 27 TZLR_32 MLEKE ALONGOLE Upland Sativa MBAWAMBILI 28 TZLR_62 RANGIMBILI Upland Sativa 29 AR_22-1 Sahel 305 Irrigated Sativa 30 TZLR_81 FAYA DUME-4 Upland Sativa

Continued

Table 9. List of 42 rice cultivars used for PCR confirmation of Genotype-by-Sequencing (GBS) -based diversity of 190 African rice cultivars.

Table 9 continued

31 TZLR_10 LINGWELINGWELI Lowland rainfed Sativa 32 TZLR_27 RINGA MSONGA Lowland rainfed Sativa 33 TZLR_64 MOSHI Lowland rainfed Sativa 34 TZLR_30 MSONGA Lowland rainfed Sativa 35 TZLR_66 KIHOGO RED Upland Sativa 36 TZLR_41 SI MZITO Lowland rainfed Sativa 37 TZLR_44 SUPA KIJIVU Lowland rainfed Sativa 38 Tog 5603 Tog 5603 Lowland Glaberrima 39 Tog 6126 Tog 6126 Lowland Glaberrima 40 Tog 84117 Tog 84117 Lowland Glaberrima 41 Nipponbare Nipponbare Irrigated Sativa 42 CO39 CO39 Irrigated Sativa

10 Resistant 10 Susceptible varieties

S/N/Lanes cultivars S/N/Lanes and controls

1 AR-47 11 TZLR-65

2 EN-19 12 TZLR-75

3 L-NER-33 13 TZLR-32

4 AR-34 14 TZLR-26

5 AR-36 15 AR-61

6 EN-20 16 TZLR-55

7 L-NER-50 17 EN-9

8 L-NER-7 18 EN-2

9 EN-25 19 TZLR-36

10 AR-28 20 TZLR-19

21 NPB (-)

22 Katy (+)

- Water (-)

Table 10. List of rice cultivars resistant (disease score < 2.0) or susceptible (disease score > 7.0) to Magnaporthe oryzae used in PCR amplification of Pita gene fragments. 1-22 and (–) represents the 23 lanes of the gel image in Fig. 19.

Magnaporthe oryzae isolate TZ-01 TZ-12 UG-05 UG-11 KE-37 BF-07 Mean Score 4.7 4.4 4.9 4.5 4.4 3.4 % Resistance 31 30 25 32.9 24 34 % Susceptibility 16 22 25 26 11.8 4.3

Table 11. Summary of 190 African rice cultivar disease phenotypes inoculated with six African Magnaporthe oryzae isolates. Resistance=disease score ≤ 2.0 and Susceptibility=disease score ≥ 8.0

OSU ID Variety Name

AR-106 EWINTO YIBO

AR-23 Sahel 328

AR-28 WAB176-B-8-HB

AR-34 WAB306-B-B-6-L2-L1-LB

AR-36 WAB337-B-B7-H4

AR-47 WAB96-1-1

AR-67 WITA 3

EN-10 IR 77713

EN-15 IR 05N221

EN-19 IR 07A166

L-NER-33 LOW LAND NERICA 33

L-NER-50 LOW LAND NERICA 50

L-NER-56 LOW LAND NERICA 55

L-NER-7 LOW LAND NERICA 7

TZLR-74 TXD 306 (IMPROVED)

Table 12. African rice cultivars resistant to six African Magnaporthe oryzae isolates (TZ- 01, TZ-12, UG-05, UG-11, KE-37 and BF-07).

Locus CHR Best P value Known R gene Candidate genes RABR_1 1 3.07E-05 LOC_Os01g23970.1 RABR_2 1 2.47E-09 LOC_Os01g66680.1 RABR_3 1 1.17E-06 Pish LOC_Os01g57270.1 RABR_4 1 2.77E-11 LOC_Os01g66560.1 RABR_5 2 2.75E-05 LOC_Os02g06320.1 RABR_6 2 1.41E-05 LOC_Os02g29040.1 RABR_7 2 1.45E-13 LOC_Os02g57180.1 RABR_8 2 1.01E-12 LOC_Os02g57160.1 RABR_9 3 7.92E-06 LOC_Os03g25370.1 RABR_10 3 5.63E-08 LOC_Os03g43880.1 RABR_11 4 4.72E-06 LOC_Os04g25820.1 RABR_12 4 3.73E-05 LOC_Os04g50680.1 RABR_13 5 3.79E-05 LOC_Os05g09550.1 RABR_14 6 5.14E-05 LOC_Os06g12220.1 RABR_15 6 1.42E-05 LOC_Os06g02980.1 RABR_16 6 1.68E-05 LOC_Os06g19660.1 RABR_17 7 1.10E-05 LOC_Os07g05740.1 RABR_18 8 3.07E-05 LOC_Os08g14210.1 RABR_19 11 1.30E-06 LOC_Os11g05850.1 RABR_20 11 3.85E-06 Pi-y(t) LOC_Os11g12300.1 RABR_21 11 1.75E-05 LOC_Os11g33120.1 RABR_22 12 1.70E-17 Pi-6 LOC_Os12g15470.1 RABR_23 12 2.75E-24 Pita LOC_Os12g22110.1 RABR_24 12 1.08E-18 LOC_Os12g17840.1 RABR_25 12 7.28E-09 LOC_Os12g23180.1

Table 13. The regions associated with rice resistance (RABRs) to six Magnaporthe oryzae isolates. Gene annotations for all the candidate genes corresponding to twenty-five RABRs are shown in Table 13.

Region/ Candidate genes Annotation Locus Cysteine-rich receptor protein kinase, putative, RABR_1 LOC_Os01g23970.1 expressed S-domain receptor-like protein kinase, putative, RABR_2 LOC_Os01g66680.1 expressed NBS-LRR disease resistance RPP13-like protein RABR_3 LOC_Os01g57270.1 1, putative, expressed Signal recognition particle 72 kDa protein, RABR_4 LOC_Os01g66560.1 putative, expressed RNA recognition motif 2 domain containing RABR_5 LOC_Os02g06320.1 protein, expressed Ankyrin repeat domain containing protein, RABR_6 LOC_Os02g29040.1 putative, expressed NADH dehydrogenase 1 alpha subcomplex RABR_7 LOC_Os02g57180.1 subunit 9,mitochondrial precursor, putative, expressed ELMO/CED-12 family protein, putative, RABR_8 LOC_Os02g57160.1 expressed RABR_9 LOC_Os03g25370.1 Peroxidase precursor, putative, expressed RABR_10 LOC_Os03g43880.1 PLA IIIA/PLP7, putative, expressed RABR_11 LOC_Os04g25820.1 Glycosyltransferase, putative, expressed MYB family transcription factor, putative, RABR_12 LOC_Os04g50680.1 expressed Der1-like family domain containing protein, RABR_13 LOC_Os05g09550.1 expressed RABR_14 LOC_Os06g12220.1 HVA22, putative, expressed ATP synthase F1, epsilon subunit family protein, RABR_15 LOC_Os06g02980.1 expressed WD domain, G-beta repeat domain containing RABR_16 LOC_Os06g19660.1 protein, expressed

Continued

Table 14. Gene annotations for the candidate genes corresponding to the twenty-five regions associated with rice resistance (RABRs) to eight Magnaporthe oryzae isolates.

Table 14 continued

Receptor-like protein kinase 2 precursor, RABR_17 LOC_Os07g05740.1 putative, expressed Glycosyl hydrolases family 16, putative, RABR_18 LOC_Os08g14210.1 expressed RABR_19 LOC_Os11g05850.1 Cyclin-T1-3, putative, expressed NBS-LRR disease resistance protein, putative, RABR_20 LOC_Os11g12300.1 expressed RABR_21 LOC_Os11g33120.1 Respiratory burst oxidase, putative, expressed OsSCP65 - Putative Serine Carboxypeptidase RABR_22 LOC_Os12g15470.1 homologue, expressed NBS-LRR resistance protein MLA13, putative, RABR_23 LOC_Os12g17340.1 expressed

RABR_24 LOC_Os12g17840.1 Ubiquitin family protein, expressed

3-beta hydroxysteroid dehydrogenase/isomerase RABR_25 LOC_Os12g23180.1 family protein, putative, expressed

OLIGO'S NAME SEQUENCE Pita5F CAGCGAACTCCTTCGCATACGCA Pita 5R CGAAAGGTGTATGCACTATAGTATCC Pita577F ATGAACACCACAGCCTAAACC Pita577R CAGACCCGAAACAACACTAGG

Pita 801F CAAGCCAAATCTGAATCTTACCAT

Pita 801R TATGGAAATGTTGCCCCAATCTG

Table 15. List of primers used for Pita gene amplification

Grp I Grp II

Grp III

Figure 14. Phylogenetic tree showing the relationship among the 190 African rice accessions. The different colors depict groupings based on genetic diversity among groups. Genetic distances between the cultivars were calculated using TASSEL, and phylogenetic tree was constructed using ggtree (R package for visualization and annotation of phylogenetic tree).

Figure 15. Gel images of the amplified African rice subspecies-specific markers. Thirty- seven randomly chosen varieties from 190 African rice accessions were used in the PCR analysis to confirm the Genotyping-by sequencing (GBS- diversity analysis results. Three Oryza glaberrima, two Oryza sativa, Nipponbare (japonica) and CO39 (indica), were included as controls. Water (lane -) was used as an additional negative control. M14 and M3 are O. glaberrima and O. sativa specific primers. The primers sequence are in Table 8.

Figure 16. Disease score distribution (0-9) of the 190 African rice cultivars inoculated with Magnaporthe oryzae isolate TZ-01 (A) and TZ-12 (B). The inoculation experiment was conducted twice and produced similar results.

Figure 17. Rice blast disease score distribution (0-9) of the 190 African rice cultivars inoculated with Ugandan Maganaporthe isolates isolates UG-05 (A) and UG-11 (B). The inoculation experiment was performed twice with consistent results.

Figure 18. Rice blast disease score distribution (0-9) of the 190 African cultivars inoculated with Kenyan Magnaporthe oryzae isolate KE-37 (A) and Burkina Faso M. oryzae isolate BF-07 (B). The inoculation experiment was performed twice with consistent results.

Figure 19. Cluster analysis of the six Magnaporthe oryzae isolates based on disease scores of the 190 African rice cultivars. The scale indicates the genetic similarity in percentage. The six isolates are clustered in three major groups. The M. oryzae isolates from Uganda (UG-05 and UG-11) are grouped together in one group while those from Tanzania (TZ-01 and TZ-12) are grouped separately. The Kenyan isolate (KE37) is in its own group whereas Burkina Faso (BF05) and Tanzania isolate (TZ-12) are clustered together.

Figure 20. A combined Manhattan plot summarizing genome wide association study (GWAS) results for the six Magnaporthe oryzae isolates. The X-axis is the genomic position of each single nucleotide polymorphism (SNP), and the Y-axis is the negative logarithm of the P-value obtained from the GWAS. SNPs with strong associations for the trait have higher Y-coordinate value. The red arrows indicate that identified regions associated with blast resistance (RABRs) are co-localized with previously mapped or cloned R gene regions.

100

Figure 21. Genome-wide association analysis for two Ugandan Magnaporthe oryzae isolates (UG-05 and UG-11) (A). Manhattan plots that summarizes genome wide association study (GWAS) results. The X-axis is the genomic position of each single nucleotide polymorphism (SNP), and the Y-axis is the negative logarithm of the P-value obtained from the GWAS model. (B) Quantile-Quantile plots that shows fitness of the GWAS analysis model for the two isolates. The X-axis corresponds to the expected values of negative logarithm of P and the Y-axis corresponds to the observed negative logarithm of P values.

101

Figure 22. Genome-wide association analysis for Kenyan (KE-37) and Burkina Faso (BF- 07) Magnaporthe oryzae isolates. (A) Manhattan plots that summarizes genome wide association study (GWAS) results. The X-axis is the genomic position of each single nucleotide polymorphism (SNP), and the Y-axis is the negative logarithm of the P-value obtained from the GWAS model. (B) Quantile-Quantile plots that shows fitness of the GWAS analysis model for the two isolates. The X-axis corresponds to the expected values of negative logarithm of P and the Y-axis corresponds to the observed negative logarithm of P values.

102

Figure 23. Genome-wide association analysis Tanzanian Magnaporthe oryzae isolate TZ- 01. (A) Quantile-Quantile plot that shows fitness of the genome wide association study (GWAS) analysis model for the isolate. The X-axis corresponds to the expected values of negative logarithm of P and the Y-axis corresponds to the observed values of negative logarithm of P. (B) Manhattan plots that summarizes genome wide association study (GWAS) results. The X-axis is the genomic position of each single nucleotide polymorphism (SNP), and the Y-axis is the negative logarithm of the P-value obtained from the GWAS model.

103

Figure 24. Genome-wide association analysis Tanzanian Magnaporthe oryzae isolate TZ- 12. (A) Quantile-Quantile plot that shows fitness of the genome wide association study (GWAS) analysis model for the isolate. The X-axis corresponds to the expected values of negative logarithm of P and the Y-axis corresponds to the observed values of negative logarithm of P. (B) Manhattan plots that summarizes genome wide association study (GWAS) results. The X-axis is the genomic position of each single nucleotide polymorphism (SNP), and the Y-axis is the negative logarithm of the P-value obtained from the GWAS model (specifically from the F-test for testing H0: No association between the SNP and trait). SNPs with strong associations for the trait (blast disease resistance) have higher Y-coordinate value.

104

Figure 25. Gel images of the amplified Pita gene fragments. Twenty African rice cultivars were used in the PCR analysis. The primer sequence and variety name can be found in Table 15. Lane 01-10 contain amplicons from varieties resistant to blast and lanes 11-20 are from varieties susceptible to blast. Nipponbare (lane 21) and Katy (lane 22) are negative and positive controls, respectively. Water (lane -) is an extra negative control. The resistance and susceptibility were based on inoculation results of these varieties to six Magnaporthe oryzae isolates (TZ-01, TZ-12, UG-05, UG-11, KE-37 and BF-07).

105

CHAPTER 4

Rice Blast Resistance Spectrum of Thirty African Rice Cultivars against Seven

Magnaporthe oryzae Isolates from Benin

Introduction

Rice (Oryza sativa L. and O. glaberrima S.) is the staple food that feeds nearly half of the world population (Way and Heong 1994). It is the most important grain in terms of human nutrition, providing more than 20 percent of calories consumed (Smith 1998). In 2012, rice was ranked second after maize worldwide in terms of production (FAOSTAT data 2012).

Asia contributes over 90 percent of the global rice production. The United States of

America (USA), the only North American exporter of rice, currently accounts for approximately 1.5% of the total amount of rice produced globally (Childs and Nathan

2012). The consumption per capital varies by country, from 210 kg/year in Myanmar to

4kg/year in France (Abdullah et al. 2006). In Africa, rice consumption has been increasing at an annual rate of 4.52% (between 1961 and 2005). In 2008, Africa cultivated about 5.6

% of the total rice area and was responsible for 3.4% of total production in the world (FAO

2012; Nayar 2014). However, global rice production has not been able to keep up with consumption patterns, and the self-sufficiency ratio has significantly dropped. Based on current statistics, rice is number one among food crops in terms of economic value in the

106

developing countries. Based on production size and market value, rice is of high importance both ecologically and economically in many Asian and African countries (IRRI

2013)

Despite the technological advancement in crop productivity, about 800 million people, especially in developing countries of Africa, South America and Asia live on one meal per day. In addition, global population increase outpaces the increase in food crop productivity. To be able to meet the global demand for rice, we need to produce 40% more rice by 2030 (Khush 2005). To arrive to such projections, improvement of important agronomic traits including resistance to insects and pathogens should be taken into consideration (Stewart and Ow 2008). Recent advancement in molecular breeding and biotechnology can be utilized to generate new rice cultivars resistant to insects and pathogens.

Rice is probably the most diverse crop with a wide gene pool. Africa has rich and diverse rice genetic stock that can be utilized in rice crop improvement programs (Sanni et al. 2013). In our two recent studies on association mapping of rice blast resistance genes, we identified a total of 56 loci that are significantly associated with blast resistance to nine

(used in Chapter 2 and 3 experiments) Magnaporthe oryzae isolates from four African countries (Tanzania, Kenya, Uganda and Burkina Faso). In addition, among 162 Rice

Diversity Panel 1 (RDP1) and 190 diverse African cultivars used in our studies, 30 were highly resistant (disease score ≤ 2) to the nine African M. oryzae isolates. This chapter present new findings on the screening of the previously identified 30 blast resistant rice cultivars using seven West African M. oryzae isolates. The intention of this study was to

107

test the resistance spectrum of the 30 cultivars using more geographic genetic diverse

African M. oryzae isolates. This work contributes to the understanding and characterization of our resistant rice germplasm and provides foundation for identification and utilization of new rice blast resistance genes in rice improvement programs.

Materials and methods

Plant and fungal materials

We used 30 rice cultivars resistant to nine African M. oryzae isolates (form RDP1 and

AfricaRice germplasm). Another set of 28 rice blast differential lines (LTH) containing R genes for blast resistance were also included in this study (Kobayashi et al. 2007). Rice accessions 75-1-127 (containing the Pi-9 gene) and CO-39 were included as resistant and susceptible controls, respectively, hence a total of 60 rice accessions were used for inoculations (Table 15). Seed increase for these cultivars were done in the rice fields at the

University of Arkansas, Fayetteville, AR, USA and AfricaRice Center, Benin. The seven

M. oryzae isolates used in this study were selected based on diversity and pathotype information. Each of the isolates contained the Avr genes different from each other and hence suitable for rice blast resistance screening (Dr. Silue, AfricaRice Center, personal communication). The selected seven M. oryzae isolates were: BN-0050, BN-0013, BN-

0040, BN-0082, BN-0119 and BN-0094, all from Benin.

108

Blast resistance screening

Fungal cultures and rice seedlings were grown and prepared for spray inoculation assay as described previously. The seeds were sown in small plastic pots containing sterilized soil and inserted in rectangular wood tray. The containers were arranged in 12 columns, each column with six mini pots. Seeds for each rice cultivar were sown in a column of six pots hence 12 different cultivars per wood tray. Five trays made a single set of 60 rice cultivars, which were then kept in green house for germination and growth. The fungal isolates were grown and propagated on rice bran-agar medium under 25°C and white florescent light incubation for 10 days to induce sporulation.

Seedlings (21 days after germination) were sprayed with fungal spore suspensions with spore concentration of 3x104 conidia/ml in 0.1% gelatin until run-off. Spore concentration was quantified using a hemocytometer. The inoculated seedlings were kept in a wooden cage covered by thick wet cloth. An electric humidifier was placed inside the cage to maintain high (≥90%) humidity. The plants were left inside the cage for 24 hours and then transferred to a wooden bench and sprayed with water mist 4 hours a day. Disease scoring was done six days after inoculation using a 0-5 scoring system developed by IRRI and JIRCAS (where 0 indicates no blast symptoms (highly resistant) and 5 indicates severe blast symptoms (highly susceptible) (Hayashi and Yoshida 2009). The inoculation experiment was performed twice under the same conditions. If the results from two experiments were different, a third experiment was performed and data from the two similar experiments were used in analysis. Minitab 17 statistical package was used for data

109

analysis. The cluster analysis was performed based on the interactions between rice genotypes and M. oryzae isolates.

Results

Blast resistance phenotypes

The blast disease scores for all the 60 rice accessions are presented in Table 16

(Differentials, RDP1 and AfricaRice lines). The disease score distributions of 30 cultivars

(RDP1 and African rice accessions) tested for their resistance spectrum are shown in Fig

26, 27, 28 and 29 for each of the seven M. oryzae isolates. The disease scores for all the isolates were generally skewed towards the resistance side (disease score ≤ 2). Based on the blast disease severity, 16 cultivars were highly resistant (disease score ≤ 2.0) to all seven M. oryzae isolates tested (Table 17).

Seven M. oryzae isolates were clustered into three major groups based on disease phenotypes of the 30 resistant cultivars (Fig. 30). The first group comprised of BN-0050,

BN-0094 and BN-0119. BN-0050 and BN-0094 had 93% similarity while BN-0119 related to 0050 and BN-0094 at 87% similarity level. BN-0013, BN-0066 and BN-0040 were clustered together in the second group. BN-0013 and BN-0066 related more to each other

(87%) than to BN-0040 (82%). BN-0082 distantly related to the rest of the isolates (60%) and was on its own category. The same trend was observed for the inoculation phenotypes of 28 blast differential lines (Fig. 31).

110

Discussion

The 3thirty rice cultivars used in this study had shown resistance to nine African rice

M. oryzae isolates. In our two previous association mapping studies, 56 RABRs were identified in these cultivars. This prompted us to screen these cultivars using more M. oryzae isolates from Africa to test the resistance spectrum of the thirty cultivars using more

Africa M. oryzae isolates. Since six out of nine isolates that were used in association mapping originated from East Africa, it was more practical to use West African M. oryzae isolates for further resistance screening. In the gene‐for‐gene resistance, the plant surveillance system recognizes the pathogen via pathogen specific effectors (avirulence proteins) coded by avirulence genes in the pathogen (Jia et al. 2000). Since these effectors are often isolate‐specific, the selection of M. oryzae isolates in this study was based on their Avr genes information in order to capture a wide range of isolate diversity and pathogenicity. Based on the inoculation results, the seven M. oryzae isolates were clustered into different groups, indicating their genetic diversity. Cluster analysis results suggests that the seven isolates may contain different Avr genes that resulted in their difference in virulence to the thirty rice accession. The same trend observed for the 28 rice blast differential lines as expected suggesting that these isolates may belong into different races.

The inoculation results show that different rice accessions respond differently to the seven M. oryzae isolates. Some accessions were susceptible to one isolate and resistant to another suggesting that they may contain different R genes or QTLs. This may be reason for the observed different levels of resistance to different M. oryzae isolates among rice accessions. Moreover, about 50 percent of the thirty rice accessions/cultivars tested were

111

susceptible to at least one of the seven isolates. All the thirty rice accessions in this study were resistant to nine M. oryzae isolates used in our two previous association mapping studies. Together, these results suggest that the seven isolates used in this study differ from the nine isolates used in our previous two association mapping studies and may contain more Avr genes. Since the remaining 50 percent of the cultivars were resistant to all the seven isolates, it shows that the kind of resistance in these cultivars is more robust and they may contain new R genes or QTLs that may be used to improve other cultivars that have good agronomic qualities but lack resistance to rice blast disease.

In addition to 30 rice cultivars, 28 rice blast differential lines developed by IRRI-Japan

Collaborative Research Project were also included. This variety set composes of monogenic lines for 24 R genes (Pia, Pib, Pii, Pik, Pik-h, Pik-m, Pik-p, Pik-s, Pish, Pit,

Pita, Pita-2, Piz, Piz-5 (Pi2), Piz-t, Pi1, Pi3, Pi5(t), Pi7, Pi9, Pi11(t), Pi12(t), Pi19(t), and

Pi20) (Kobayashi et al. 2007). Twenty-four of the differential lines were susceptible to at least one of the seven M. oryzae isolates. Only four lines were moderately resistant with a disease score range of 1 to 2. Two major R genes (Pita-2 and Piz-5) were involved for resistance of these four lines. These findings offer useful information for the use of appropriate R genes for rice breeding programs in West African countries.

112

References

Abdullah, A.B., S. Ito, K. Adhana. 2005. Estimate of rice consumption in Asian countries and the world towards 2050. Tottori University, Japan. P 28-42. Online: http://worldfood.apionet.or.jp/alias pdf.

AfricaRice. 2012. Africa Rice Centre (AfricaRice) Annual Report 2011: A new rice research for development strategy for Africa. Cotonou, Benin.

Bashir, Uzma, Sobia, M., and Naureen, A. 2014. First report of alternaria metachromatica from Pakistan causing leaf spot of tomato. Pakistan Journal of Agricultural Science 51:305-308.

Bourett, T. M., and Howard, R. J. 1990. In vitro development of penetration structures in the rice blast fungus Magnaporthe grisea. Canadian Journal of Botany 68:329-342.

Bryan, G. T., Wu, K.-S., Farrall, L., Jia, Y., Hershey, H. P., McAdams, S. A., Faulk, K. N., Donaldson, G. K., Tarchini, R., and Valent, B. 2000. A single amino acid difference 113

distinguishes resistant and susceptible alleles of the rice blast resistance gene Pita. The Plant Cell 12:2033-2045.

Childs, and Nathan. 2012. Rice Outlook. A Report from Economic Research Service (ERS).USDA. Washington, D. C. Available at: www.ers.usda.org.

Christiansen, T., Foy B.D., and L., W. 2012. Programming perl: Unmatched power for text processing and scripting. O'Reilly Media Inc. Sebastopol, California, USA.

Dean, R. A. 1997. Signal pathways and appressorium morphogenesis. Annual Review of Phytopathology 35:211-234.

FAO. 2012. The State of Food Insecurity in the World 2012: Economic growth is necessary but not sufficient to accelerate reduction of hunger and malnutrition. FAO, Rome. doi 10.

FAOSTAT data. 2012. FAOSTAT.http://faostat.fao.org/site/339/default.aspx. Retrieved June 12, 2016

Garris, A. J., Tai, T. H., Coburn, J., Kresovich, S., and McCouch, S. 2005. Genetic structure and diversity in Oryza sativa L. Genetics 169:1631-1638.

114

Guimaraes, E. P. 2002. Genetic diversity of rice production in Brazil. In: Nguyen, V.N. (ed.) Genetic Diversity in Rice Production, Case Studies from Brazil, India and Nigeria. Food and Agriculture Organization of the United Nations (FAO), Rome, Italy:11-35.

Hamer, J. E., Howard, R. J., Chumley, F. G., and Valent, B. 1988. A mechanism for surface attachment in spores of a plant pathogenic fungus. Science 239:288-290.

Harlan, J. R., and Stemler, A. 1976. The races of sorghum in Africa. In: Harlan, J.R., De Wet, J.M. and Stemler, A.B. (eds) Origin of African Plant Domestication. Mouton, The Hague, Netherlands.

Henderson, C. R. 1975. Best linear unbiased estimation and prediction under a selection model. Biometrics 31:423-447.

Imbe, and Matsumoto. 1985. Inheritance of resistance of rice varieties to the blast fungus strains virulent to the variety" Reiho". Japanese Journal of breeding 35:332-339.

IRRI. 2013. International Rice Research Institute. World Rice Statistics: . Los Baños, the Philippines: IRRI. June 29, 2013.

115

Khush, G. S. 2005. What it will take to feed 5.0 billion rice consumers in 2030. Plant molecular biology 59:1-6.

Kobayashi, N., Telebanco-Yanoria, M. J., Tsunematsu, H., Kato, H., Imbe, T., and Fukuta, Y. 2007. Development of new sets of international standard differential varieties for blast resistance in rice (Oryza sativa L.). Japan Agricultural Research Quarterly: JARQ 41:31-37.

Li, H., and Durbin, R. 2010. Fast and accurate long-read alignment with Burrows–Wheeler transform. Bioinformatics 26:589-595.

Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, G., and Durbin, R. 2009. The sequence alignment/map format and SAMtools. Bioinformatics 25:2078-2079.

Linares, O. F. 2002. African rice (Oryza glaberrima): history and future potential. Proceedings of the National Academy of Sciences 99:16360-16365.

Liu, W., Liu, J., Triplett, L., Leach, J. E., and Wang, G. L. 2014. Novel insights into rice innate immunity against bacterial and fungal pathogens. Annual Revision of Phytopathology 52:213-241.

116

Metzker, M. L. 2010. Sequencing technologies - the next generation. Nature eviews. Genetics 11:31-46.

Mew, T. W., Leung, H., Savary, S., Vera Cruz, C. M., and Leach, J. E. 2004. Looking ahead in rice disease research and management. Critical Reviews in Plant Sciences 23:103-127.

Moldenhauer, K. A., and Gibbons, J. H. 2003. Rice morphology and development. Rice: Origin, History, Technology, and Production. Hoboken, NJ. John Wiley and Sons.

Mundt, C. C. 2014. Durable resistance: a key to sustainable management of pathogens and pests. Infection, Genetics and Evolution 27:446-455.

Muthayya, S., Sugimoto, J. D., Montgomery, S., and Maberly, G. F. 2014. An overview of global rice production, supply, trade, and consumption. Annals of the New York Academy of Sciences 1324:7-14.

Nayar, N. M. 2014. Rice in Africa. Pages 1-7 in: Encyclopaedia of the History of Science, Technology, and Medicine in Non-Western Cultures. H. Selin, ed. Springer, Dordrecht, Netherlands.

Orjuela, J., Sabot, F., Chéron, S., Vigouroux, Y., Adam, H., Chrestin, H., Sanni, K., Lorieux, M., and Ghesquière, A. 2014. An extensive analysis of the African rice

117

genetic diversity through a global genotyping. Theoretical and Applied Genetics 127:2211-2223.

Portères, R. 1962. Berceaux agricoles primaires sur le continent africain. The Journal of African History 3:195-210.

Portères, R. 1970. Primary cradles of agriculture in the African continent. Papers in African Prehistory:43-58.

Project, I. R. G. S. 2005. The map-based sequence of the rice genome. Nature 436:793- 800.

Rossman, A. Y., Howard, R. J., and Valent, B. 1990. Pyricularia grisea, the Correct Name for the Rice Blast Disease Fungus. Mycologia 82:509-512.

Sanni, K. A., Tia, D. D., Ojo, D. K., Ogunbayo, A. S., Sikirou, M., and Hamilton, N. R. S. 2013. 7 Diversity of Rice and Related Wild Species in Africa. Realizing Africa's Rice Promise:87.

118

Savary, S., Horgan, F., Willocquet, L., and Heong, K. 2012. A review of principles for sustainable pest management in rice. Crop Protection 32:54-63.

Séré, Y., Fargette, D., Abo, M. E., Wydra, K., Bimerew, M., Onasanya, A., and Akator, S. K. 2013. Managing the Major Diseases of Rice in Africa. Realizing Africa's Rice Promise:213-228

Smith, B. D. 1998. The Emergence of Agriculture. Scientific American Library, A Division of HPHLP, New York: ISBN 0-7167-6030-7164.

Stewart, C. N., and Ow, D. W. 2008. The Future of Plant Biotechnology. Plant Biotechnology and Genetics: Principles, Techniques, and Applications:357-369.

Swaminathan, M. S. 1984. Rice in 2000 AD. In: Abrol and Sulochana Gadgil (eds.), Rice in a variable climate. APC Publications Pvt. Ltd., New Delhi-110005, India.

Takahashi, A., Hayashi, N., Miyao, A., and Hirochika, H. 2010. Unique features of the rice blast resistance Pish locus revealed by large scale retrotransposon-tagging. BMC Plant Biology 10:175.

Talbot, N. J. 2003. On the trail of a cereal killer: exploring the biology of Magnaporthe grisea. Annual Reviews in Microbiology 57:177-202.

119

Timsina, J., and Connor, D. 2001. Productivity and management of rice–wheat cropping systems: issues and challenges. Field Crops Research 69:93-132.

Valent, B., and Chumley, F. G. 1991. Molecular genetic analysis of the rice blast fungus, Magnaporthe grisea. Annual Review of Phytopathology 29:443-467.

Wang, C., Yang, Y., Yuan, X., Xu, Q., Feng, Y., Yu, H., Wang, Y., and Wei, X. 2014a. Genome-wide association study of blast resistance in indica rice. BMC Plant Biology 14:1-11.

Wang, X., Lee, S., Wang, J., Ma, J., Bianco, T., and Jia, Y. 2014c. Current advances on genetic resistance to rice blast disease. Wengui Yan (Ed) 1501:70.

Way, M., and Heong, K. 1994. The role of biodiversity in the dynamics and management of insect pests of tropical irrigated rice--A review. Bulletin of Entomological Research 84:567-588.

Zeigler, R. S., Leong, S. A., and Teng, P. S. 1994. Rice blast disease. CAB International, Wallingford, UK, in association with the International Rice Research Institute, Los Banos, the Philippines.

120

121

UA-id/ Disease Scores of Seven Magnaporthe oryzae Isolates S/N OSU-id Variety Name Code BN0050 BN0013 BN0040 BN0066 BN0082 BN0119 BN0094 1 1-Nov IRBLA-A 4 4 3 3 0 5 5 2 2-Nov IRBLA-C 4 5 4 4 3 5 5 3 3-Nov IRBLI-F5 3 4 3 3 3 5 5 4 4-Nov IRBLKS-F5 4 5 4 4 3 5 5 5 5-Nov IRBLKS-S 2 4 2 3 0 5 4 6 6-Nov IRBLK-KA 4 3 2 3 2 5 5 7 7-Nov IRBLKP-K60 5 4 2 4 0 5 5 8 8-Nov IRBLKH-K3 4 2 2 1 3 4 5 9 9-Nov IRBLZ-FU 2 4 1 2 2 2 3 10 10-Nov IRBLZ5-CA 2 1 0 0 0 2 3 11 11-Nov IRBLZT-T 3 1 0 1 2 4 4 IRBLTA CT 12 13-Nov 2 2 3 2 4 3 3 3 13 14-Nov IRBLB-B 3 5 4 5 3 5 4 14 15-Nov IRBLT-K59 4 5 3 4 3 4 4 15 16-Nov IRBLSH-S 2 2 1 2 1 4 2 16 17-Nov IRBLSH-B 2 4 1 3 2 4 3 17 18-Nov IRBL 1-CL 5 4 2 3 1 5 4 18 19-Nov IRBL 3-CP 4 4 4 2 4 2 5 4 19 20-Nov IRBL 5-M 5 4 3 4 2 5 4 20 21-Nov IRBL 7-M 4 4 2 4 3 5 5 21 22-Nov IRBL 9-W 1 2 0 1 0 2 2 22 23-Nov IRBL 12-M 3 2 0 0 0 3 2 23 24-Nov IRBL 19-A 5 4 2 4 3 5 5 24 25-Nov IRBLKM TS 5 2 1 3 1 5 5 IRBL 20-IR 25 26-Nov 24 3 0 4 3 3 5 4 26 27-Nov IRBLTA 2-PI 0 0 1 2 2 1 1 IRBLTA 2- 27 28-Nov RE 0 0 0 1 2 1 0 28 29-Nov IRBLTA CP 1 4 4 2 4 3 3 4 29 30-Nov IRBL 11-ZH 2 3 2 4 2 5 4

Continued

Table 16. Blast disease scores of 60 rice cultivars inoculated with seven Magnaporthe oryza isolates from Benin.

122

Table 16 continued

30 Nov-31 IRBLZ 5-CA ® 2 1 0 0 0 2 0 31 Nov-32 IRTP 16211 (LTH) 5 3 3 3 3 5 5 32 Nov-53 CO39 5 4 3 4 3 4 5 33 L-NER-7 Lowland NERICA 7 2 2 1 3 0 3 1 34 L-NER-50 Lowland NERICA 50 0 0 0 0 2 0 0 35 75-1-127 75-1-127 0 0 0 0 0 0 0 36 AR36 WAB337-B-B7-H4 0 3 1 4 3 1 0 37 L-NER-33 Lowland NERICA 33 0 0 0 0 0 0 0 WAB306-B-B-6-L2-L1- 38 AR34 LB 1 1 1 2 0 0 0 39 AR28 WAB176-B-8-HB 1 1 1 1 0 0 0 40 Toride Toride 2 0 0 0 0 4 0 41 EN-19 IR 07A166 0 1 1 1 1 0 0 42 EN-10 IR 77713 0 2 1 2 1 2 0 43 TZLR-74 TXD 306 0 1 1 3 1 3 1 44 AR-67 WITA 3 0 0 0 0 1 0 0 45 AR-47 WAB96-1-1 0 1 1 0 1 1 0 46 AR-106 Ewinto Yibo 0 0 0 1 1 0 0 47 EN-15 IR 05N221 0 3 0 3 0 1 0 48 L-NER-56 Lowland NERICA 56 0 1 0 0 1 0 0 49 EN-25 K5 1 3 3 2 0 5 3 50 NSFTV17 Binulawan 0 0 1 2 0 1 0 51 NSFTV32 Chondongji 3 3 2 3 1 4 2 52 NSFTV72 IR8 1 0 3 3 0 4 2 53 NSFTV83 Kamenoo 3 2 0 2 0 3 2 54 NSFTV163 Taducan 2 1 1 2 1 0 1 55 NSFTV232 Shangyu 394 1 0 0 1 0 3 2 56 NSFTV291 Topolea 70/76 3 1 0 2 0 4 2 57 NSFTV616 RT0034 0 0 0 1 0 0 0 58 NSFTV30 Chiem Chanh 2 2 0 2 0 0 0 59 NSFTV62 Gyehwa 3 0 0 0 0 0 0 0 60 NSFTV284 0 0 0 1 0 0 0

123

Resistance level based on disease scores 0 0-1 1-2 75-1-127 NSFTV284 Lowland NERICA 50 L-NER-33 RT0034 WAB306-B-B-6-L2-L1-LB Gyehwa 3 L-NER-56 IR 77713 WITA 3 Binulawan WAB96-1-1 Taducan Ewinto Yibo Chiem Chanh IR 07A166 IRBLZ5-CA WAB176-B-8-HB IRBLTA 2-PI IRBLZ 5-CA ® IRBLTA 2-RE

Table 17. Rice cultivars highly resistant (disease score ≤ 2) to seven Magnaporthe oryzae isolates from Benin.

124

Figure 26. Rice blast disease score distribution (0-5) of the thirty African rice cultivars inoculated with Benin isolates BN-0050 (A) and BN-0013 (B). The inoculation experiment was conducted two times with consistent results.

125

Figure 27. Rice blast disease score distribution (0-5) of the thirty African rice cultivars inoculated with Benin isolates BN-0040 (A) and BN-0066 (B). The inoculation experiment was conducted two times with consistent results.

126

Figure 28. Rice blast disease score distribution (0-5) of the thirty African rice cultivars inoculated with Benin isolates BN-0082 (A) and BN-0119 (B). The inoculation experiment was conducted two times with consistent results.

127

Figure 29. Rice blast disease score distribution (0-5) of the thirty African rice cultivars inoculated with Benin isolates BN-0094. The inoculation experiment was conducted two times with consistent results.

128

Relationship among blast isolates based on disease phenotypes

59.67

)

(

73.11

S 86.56

100.00 BN0050 BN0094 BN0119 BN0013 BN0066 BN0040 BN0082 Blast isolates

Figure 30. Cluster analysis of the seven M. oryzae isolates based on disease scores of the thirty resistant rice cultivars. The scale indicates their similarity in percentage. The seven isolates are clustered in three major groups.

129

Relationship among blast isolates based on disease phenotypes

61.86

)

(

74.57

S 87.29

100.00 BN0050 BN0119 BN0094 BN0013 BN0040 BN0066 BN0082 Blast isolates

Figure 31. Cluster analysis of the seven M. oryzae isolates based on disease scores of the twenty-eight rice blast differential lines. The scale indicates their similarity in percentage. The seven isolates are clustered in three major groups.

130

CHAPTER 5

PROJECT SUMMARY AND CONCLUSION

The significance of genetic resistance in controlling rice blast disease cannot be over emphasized. This chapter summarizes the results demonstrated in previous chapters on association mapping of two rice germplasm populations and GBS-based diversity of the African rice germplasm. Genome-wide association study (GWAS) with high-density single nucleotide polymorphism (SNP) chips was used to identify R genes and quantitative trait loci (QTLs) associated with blast resistance in two rice populations. The 162 diverse accessions selected from the rice diversity panel 1 (RDP1) are genotyped with 36,900 SNPs and the 190 diverse African rice accessions are genotyped with 184K SNPs. The mapping for these two populations was done separately using eight Magnaporthe oryzae isolates for RDP1 accessions and six M. oryzae isolates for African accessions. A total of 56 loci significantly associated with blast resistance were identified. Thirteen of these loci are linked to previously mapped R genes and QTLs while 43 are novel candidate genes/QTLs. PCR analyses revealed close linkage of two major loci on chromosome 1 (RABR_2) and 12 (RABR_23) to Pish and Pita R genes, respectively. The newly identified loci contain genes encoding defense-related proteins, transcription factors, and receptor-like protein kinases. In addition, 30 highly resistant (disease score ≤ 2) rice cultivars were identified. These cultivars were screened using seven West African M. oryzae isolates to test their resistance spectrum. About 50 percent of these cultivars showed resistance to all the seven isolates, thus they can be used to improve elite cultivars with good agronomic qualities but lack resistance to M. oryzae.

131

We also studied the diversity of African rice accessions using 184K SNPs and identified three highly diverse major population groups. We detected the misclassification of some rice cultivars. The analysis shows that the African rice cultivars (both O. glaberrima and O. sativa) are highly diverse and can be further explored. The findings in this research project also demonstrates the usefulness of GBS in diversity analysis and effectiveness of GWAS for quick identification of R/QTLs genes in rice. The SNP markers that are tightly linked to major R genes can be immediately used in breeding for resistance against M. oryzae in Africa.

132

BIBLIOGRAPHY

AfricaRice. 2012. Africa Rice Centre (AfricaRice) Annual Report 2011: A new rice research for development strategy for Africa. Cotonou, Benin.

Bashir, Uzma, Sobia, M., and Naureen, A. 2014. First report of alternaria metachromatica from Pakistan causing leaf spot of tomato. Pakistan Journal of Agricultural Science 51:305-308.

Bourett, T. M., and Howard, R. J. 1990. In vitro development of penetration structures in the rice blast fungus Magnaporthe grisea. Canadian Journal of Botany 68:329-342.

133

Childs, and Nathan. 2012. Rice Outlook. A Report from Economic Research Service (ERS).USDA. Washington, D. C. Available at: www.ers.usda.org.

Christiansen, T., Foy B.D., and L., W. 2012. Programming perl: Unmatched power for text processing and scripting. O'Reilly Media Inc. Sebastopol, California, USA.

Dean, R. A. 1997. Signal pathways and appressorium morphogenesis. Annual Review of Phytopathology 35:211-234.

FAO. 2012. The State of Food Insecurity in the World 2012: Economic growth is necessary but not sufficient to accelerate reduction of hunger and malnutrition. FAO, Rome. doi 10.

FAOSTAT data. 2012. FAOSTAT.http://faostat.fao.org/site/339/default.aspx. Retrieved June 12, 2016

Garris, A. J., Tai, T. H., Coburn, J., Kresovich, S., and McCouch, S. 2005. Genetic structure and diversity in Oryza sativa L. Genetics 169:1631-1638.

134

Hamer, J. E., Howard, R. J., Chumley, F. G., and Valent, B. 1988. A mechanism for surface attachment in spores of a plant pathogenic fungus. Science 239:288-290.

Harlan, J. R., and Stemler, A. 1976. The races of sorghum in Africa. In: Harlan, J.R., De Wet, J.M. and Stemler, A.B. (eds) Origin of African Plant Domestication. Mouton, The Hague, Netherlands.

Henderson, C. R. 1975. Best linear unbiased estimation and prediction under a selection model. Biometrics 31:423-447.

Imbe, and Matsumoto. 1985. Inheritance of resistance of rice varieties to the blast fungus strains virulent to the variety" Reiho". Japanese Journal of breeding 35:332-339.

IRRI. 2013. International Rice Research Institute. World Rice Statistics: . Los Baños, the Philippines: IRRI. June 29, 2013.

135

Khush, G. S. 2005. What it will take to feed 5.0 billion rice consumers in 2030. Plant molecular biology 59:1-6.

Li, H., and Durbin, R. 2010. Fast and accurate long-read alignment with Burrows–Wheeler transform. Bioinformatics 26:589-595.

Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, G., and Durbin, R. 2009. The sequence alignment/map format and SAMtools. Bioinformatics 25:2078-2079.

Linares, O. F. 2002. African rice (Oryza glaberrima): history and future potential. Proceedings of the National Academy of Sciences 99:16360-16365.

Liu, W., Liu, J., Triplett, L., Leach, J. E., and Wang, G. L. 2014. Novel insights into rice innate immunity against bacterial and fungal pathogens. Annual Revision of Phytopathology 52:213-241.

136

Metzker, M. L. 2010. Sequencing technologies - the next generation. Nature eviews. Genetics 11:31-46.

Mew, T. W., Leung, H., Savary, S., Vera Cruz, C. M., and Leach, J. E. 2004. Looking ahead in rice disease research and management. Critical Reviews in Plant Sciences 23:103-127.

Moldenhauer, K. A., and Gibbons, J. H. 2003. Rice morphology and development. Rice: Origin, History, Technology, and Production. Hoboken, NJ. John Wiley and Sons.

Mundt, C. C. 2014. Durable resistance: a key to sustainable management of pathogens and pests. Infection, Genetics and Evolution 27:446-455.

Muthayya, S., Sugimoto, J. D., Montgomery, S., and Maberly, G. F. 2014. An overview of global rice production, supply, trade, and consumption. Annals of the New York Academy of Sciences 1324:7-14.

Nayar, N. M. 2014. Rice in Africa. Pages 1-7 in: Encyclopaedia of the History of Science, Technology, and Medicine in Non-Western Cultures. H. Selin, ed. Springer, Dordrecht, Netherlands.

137

Portères, R. 1962. Berceaux agricoles primaires sur le continent africain. The Journal of African History 3:195-210.

Portères, R. 1970. Primary cradles of agriculture in the African continent. Papers in African Prehistory:43-58.

Project, I. R. G. S. 2005. The map-based sequence of the rice genome. Nature 436:793- 800.

Rossman, A. Y., Howard, R. J., and Valent, B. 1990. Pyricularia grisea, the Correct Name for the Rice Blast Disease Fungus. Mycologia 82:509-512.

Sanni, K. A., Tia, D. D., Ojo, D. K., Ogunbayo, A. S., Sikirou, M., and Hamilton, N. R. S. 2013. 7 Diversity of Rice and Related Wild Species in Africa. Realizing Africa's Rice Promise:87.

138

Savary, S., Horgan, F., Willocquet, L., and Heong, K. 2012. A review of principles for sustainable pest management in rice. Crop Protection 32:54-63.

Séré, Y., Fargette, D., Abo, M. E., Wydra, K., Bimerew, M., Onasanya, A., and Akator, S. K. 2013. Managing the Major Diseases of Rice in Africa. Realizing Africa's Rice Promise:213-228

Smith, B. D. 1998. The Emergence of Agriculture. Scientific American Library, A Division of HPHLP, New York: ISBN 0-7167-6030-7164.

Stewart, C. N., and Ow, D. W. 2008. The Future of Plant Biotechnology. Plant Biotechnology and Genetics: Principles, Techniques, and Applications:357-369.

Swaminathan, M. S. 1984. Rice in 2000 AD. In: Abrol and Sulochana Gadgil (eds.), Rice in a variable climate. APC Publications Pvt. Ltd., New Delhi-110005, India.

Takahashi, A., Hayashi, N., Miyao, A., and Hirochika, H. 2010. Unique features of the rice blast resistance Pish locus revealed by large scale retrotransposon-tagging. BMC Plant Biology 10:175.

Talbot, N. J. 2003. On the trail of a cereal killer: exploring the biology of Magnaporthe grisea. Annual Reviews in Microbiology 57:177-202.

139

Timsina, J., and Connor, D. 2001. Productivity and management of rice–wheat cropping systems: issues and challenges. Field Crops Research 69:93-132.

Valent, B., and Chumley, F. G. 1991. Molecular genetic analysis of the rice blast fungus, Magnaporthe grisea. Annual Review of Phytopathology 29:443-467.

Wang, C., Yang, Y., Yuan, X., Xu, Q., Feng, Y., Yu, H., Wang, Y., and Wei, X. 2014a. Genome-wide association study of blast resistance in indica rice. BMC Plant Biology 14:1-11.

Wang, X., Lee, S., Wang, J., Ma, J., Bianco, T., and Jia, Y. 2014c. Current advances on genetic resistance to rice blast disease. Wengui Yan (Ed) 1501:70.

Way, M., and Heong, K. 1994. The role of biodiversity in the dynamics and management of insect pests of tropical irrigated rice--A review. Bulletin of Entomological Research 84:567-588.

Zeigler, R. S., Leong, S. A., and Teng, P. S. 1994. Rice blast disease. CAB International, Wallingford, UK, in association with the International Rice Research Institute, Los Banos, the Philippines.

140

141