University of Florida Thesis Or Dissertation Formatting

INNOVATIVE TOOLS FOR STUDYING PLANT-RELATED OOMYCETE POPULATIONS OF POTENTIAL AND CURRENT THREAT IN AGRICULTURE IN ECUADOR

MARIA FERNANDA RATTI TORRES

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2018

To my Obi-Wan Roberto, Yoda one for me. You have been nothing but supportive during these years, you make me proud of being your wife, but above all, you make me immensely happy. This is for you and for a marriage that has been put to rest for so long and it is ready to resume.

ACKNOWLEDGMENTS

I wish to thank my parents for all their effort and the patience, they are my pillar without whom I could not have pursued my Ph.D. studies. Also to my siblings Andrea and Pablo, my brother in law Carlos and my nephews, not only for their support, but for joking around all the time and cheering me up during this journey.

Erica M. Goss deserves a special section only for her, but the formatting will not allow it. I cannot imagine having spent these years under anybody else’s guidance. She always challenged me to be better, to be calmed during stressful situations and to trust in myself. Her advices will be forever in my mind.

Doing this research would have been impossible without helping hands around the world: Thanks to Esther Lilia P. for all her support, to Juan C., Carlos A., Jerry L. and Daynet S, for assisting with logistic and lab space in Ecuador. Also to Ricardo O.,

Frank M., Leonardo S., Janna B, Roberto Faedda, Natalia Peres, Shad Ali, and Edzard v.S., for providing isolates/DNA, advice and assistance that were required for this research. Also, to my Committee members: Randy P., Karen G. and Jiri H. that were very helpful with advices, suggestions, and corrections, they were a wonderful team and

I learnt a lot from them.

I want to thank my lab members: Marina A., who was always kind and whose help was extremely valuable; Jonelle J., Parker N., and Kelcey H., the smartest undergraduates in our Department. To my lab mates throughout my stay: Jianan W.,

Jackson P., Kola F., Fernanda I., Jeannie K., Brett L. and Ashish A. who were always willing to help, talk, or just listen. Thanks to all of them for making our lab such a nice place to work.

There are many people who were there pushing me giving words of courage when I needed it the most. I guess they call themselves friends, but they might as well be angels. Ellen D., Shankar K., Mpoki S., were my first lab and office mates, I wish the best for them. Also, to Ariena v.B. for giving me the opportunity to come here, she was the first person opening doors for me and I will be eternally grateful with her. A special thank for my chocolate club Ying and James, who will always be in my heart and in my mind. Thanks to my bestie Ana G., whose words of support always kept me up when I thought I could not have the strength to continue. This section would not be complete without my Ecuadorian fellas Lisbeth E., Daniel M., and Andres O. for making

Gainesville an even more interesting place to live. But especially to my “Carachitas”

Jorge and Miguel (honorary Ecuadorian), I do not know what I would have done without them, hope we can get together soon.

I want to thank to my natural science teacher at Elementary school, Miss Betty V. and all the women around the world who are involved in science, they are an everyday inspiration.

Finally, I cannot thank enough to my ultimate inspiration Roberto J., he is my Jedi master, partner, friend and my love.

The Force will be with you, always.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 8

LIST OF FIGURES ...... 10

LIST OF ABBREVIATIONS ...... 12

ABSTRACT ...... 13

CHAPTER

1 OOMYCETE PATHOGENS IN TROPICAL AGRO-SYSTEMS ...... 15

Emerging Pathogens in Crops and Agro-Ecological Interfaces ...... 15 Overview of Oomycetes ...... 17 Oomycete Communities in Tropical Soil: Cacao Farms ...... 19 Phytophthora nicotianae, Another Versatile Pathogen of Tropical and Subtropical Crops ...... 20 Hybrid Phytophthora Species ...... 22 Objectives ...... 25

2 EVALUATION OF TWO BARCODING MARKERS FOR HIGH THROUGHPUT SEQUENCING OF OOMYCETES AS APPLIED TO CACAO AGRO- ECOSYSTEMS IN ECUADOR ...... 26

Introduction ...... 26 Materials and Methods...... 30 Sampling Sites and Strategies ...... 30 Library Preparation and Sequencing ...... 30 Database Assemblage and Taxonomy Assignment of Sequence Reads ...... 33 Statistical Analyses ...... 34 Results ...... 36 Quality of Reads and General Statistics of Runs ...... 36 Mock Communities ...... 36 Species Composition of Soil Samples ...... 37 Taxa in Relationship with Land Use ...... 40 Discussion ...... 41

3 PINEAPPLE HEART ROT ISOLATES FROM ECUADOR REAVEAL A NEW GENOTYPE OF PHYTOPHTHORA NICOTIANAE ...... 65

Introduction ...... 65 Materials and Methods...... 68

Sampling and Morphological Identification of Isolates ...... 68 DNA Extraction and PCR Conditions for Microsatellites and Mitochondrial Markers ...... 69 SSR Genotyping and Analysis ...... 71 Multi-Locus Sequencing Analysis ...... 72 Results ...... 73 Identity of Pineapple Isolates and Morphological Description...... 73 SSR Diversity and Host-Population Genetic Distance ...... 74 Population Structure Across Host-Populations ...... 75 Multi-Locus Sequencing ...... 76 Discussion ...... 76

4 EVALUATION OF HIGH RESOLUTION MELTING ANALYSIS AS A METHOD TO DIFFERENTIATE PHYTOPHTHORA HYBRIDS FROM THEIR PARENTAL SPECIES ...... 88

Introduction ...... 88 Materials and Methods...... 92 Isolates Used in this Study ...... 92 Polymorphic Sites Mining and Primer Design ...... 93 PCR and High Resolution Melting ...... 94 Results ...... 95 Primer Design and Evaluation ...... 95 P. infestans and P. andina Melting Curves ...... 95 P. nicotianae, P. cactorum and P. x pelgrandis Melting Curves ...... 96 Discussion ...... 97

5 GENERAL DISCUSSION AND CONCLUDING REMARKS ...... 120

APPENDIX

A SUPPLEMENTARY TABLES OF CHAPTER 2 ...... 122

B SUPPLEMENTARY FIGURES OF CHAPTER 2 ...... 126

C SUPPLEMENTARY TABLES OF CHAPTER 3 ...... 127

D SUPLEMENTAL FIGURES OF CHAPTER 3 ...... 144

E SUPLEMENTAL TABLES OF CHAPTER 4 ...... 147

LIST OF REFERENCES ...... 149

BIOGRAPHICAL SKETCH ...... 163

LIST OF TABLES

Table page

2-1 Sampling sites location and characteristics ...... 47

2-2 Names and sequences of primer names used in this study ...... 48

2-3 Number of reads assigned to known taxa and number of taxa per sample using cox2 marker based on usearch61 OTU-picking ...... 49

2-4 Number of reads assigned to known taxa and number of taxa per sample using ITS marker based on usearch61 OTU-picking ...... 50

2-5 Blast hits with at least 90% identity using cox2 marker ...... 51

2-6 Blast hits with at least 90% identity using ITS marker ...... 53

2-7 Taxa detected by both markers cox2 and ITS with ≥90% ID ...... 54

2-8 Simpson’s D diversity index calculated for each sample and land use type considered in this study...... 55

2-9 ANOVA results for effects of land use and farm on Simpson’s D index of diversity ...... 55

2-10 PERMANOVA results per sample from cox2 unrarefied OTU table ...... 56

2-11 PERMANOVA results per farm from cox2 unrarefied OTU table...... 56

2-12 PERMANOVA results per farm from ITS unrarefied OTU table...... 57

3-1 Isolates collected in this study, their origin and mating type...... 80

3-2 Allele sizes (bp) for SSR loci and the resulting multi-loci genotype (MLG) designation for each isolate...... 81

3-3 Genotypic diversity of populations with sample sizes of greater than 10 isolates ...... 82

3-4 Pairwise genetic differentiation based on FST values across the eight populations defined in Table 3-3 ...... 83

4-1 Species, name, host and collection site of isolates used in this study ...... 101

4-2 Primer information and details of amplicons...... 102

4-3 Pre and post-melting temperature for each primer set ...... 103

A-1 OTU table generated by usearch61 method for cox2 marker ...... 122

A-2 OTU table generated by usearch61 method for ITS marker ...... 124

C-1 Original allele calls before correction to allow comparison with data published by Biasi et al. 2016 ...... 127

C-2 General information of P. nicotianae isolates used for analysis...... 128

C-3 P. nicotianae allelic dataset used for analysis...... 134

E-1 Primer sets designed for HRM analysis of P. andina hybrid and its parental species P. infestans ...... 147

E-2 Primer sets designed for HRM analysis of P. xpelgrandis hybrid and its parental species P. cactorum and P. nicotianae ...... 148

LIST OF FIGURES

Figure page

2-1 Relative abundance of Oomycete clades by farm and land use obtained with usearch61 OTU picking against cox2 Oomycete database...... 58

2-2 Relative abundance of Oomycete genera for each sample obtained by merging outputs from usearch61 and BLAST search with at least 90% identity using cox2 database ...... 59

2-3 Unrarefied total number of reads of Oomycete taxa by farm and land use with cox2 marker; y axis is in logarithmic scale...... 60

2-4 Relative abundance of Oomycete clades for each sample type and farm obtained with usearch61 OTU picking against ITS Oomycete database...... 61

2-5 Relative abundance of Oomycete genera for each sample obtained by merging outputs from unrarefied usearch61 and BLAST with at least 90% identity using ITS...... 62

2-6 Unrarefied total number of reads of Oomycete taxa by farm and land with unrarefied OTU data obtained by usearch91 with ITS marker...... 63

2-7 NMDS of Bray-Curtis dissimilarity between communities of Crop and AE samples...... 64

3-1 Minimum spanning network using Bruvo’s distance for P. nicotianae isolates by host and country...... 84

3-2 Principal coordinates analysis of multi-locus genotypes using Lynch distance. . 85

3-3 Phytophthora nicotianae population structure by host and country...... 86

3-4 Maximum likelihood phylogeny inferred from concatenated DNA sequences from cox2+spacer and trnG_rns mitochondrial markers...... 87

4-1 Derivative melt curves plots, generated for primer sets to differentiate P. infestans and its hybrid P. andina...... 104

4-2 Aligned melt curves for primer set PaM2 ...... 105

4-3 Aligned melt curves for primer set PaM18 ...... 106

4-4 Aligned melt curves for primer set PaM19 ...... 107

4-5 Aligned melt curves for primer set PaM27 ...... 108

4-6 Aligned melt curves for primer set PaM33 ...... 109

4-7 Aligned melt curves for primer set PaM57 ...... 110

4-8 Difference plots, generated for primer sets to differentiate P. infestans and its hybrid P. andina...... 111

4-9 Derivative Melt Curves plots, generated for primer sets to differentiate the hybrid P. xpelgrandis from its parental species P. nicotianae and P. cactorum...... 112

4-10 Aligned melt curves for primer set Ph9 ...... 113

4-11 Aligned melt curves for primer set Ph11 ...... 114

4-12 Aligned melt curves for primer set Ph14 ...... 115

4-13 Aligned melt curves for primer set Ph25 ...... 116

4-14 Aligned melt curves for primer set Ph29 ...... 117

4-15 Aligned melt curves for primer set Ph29 ...... 118

4-16 Difference plots, generated for primer sets to differentiate P. nicotianae, P. cactorum and its hybrid P. xpelgrandis...... 119

B-1 Geographic location of sampling sites in the Coastal Region of Ecuador...... 126

D-1 Maximum likelihood rooted tree using cox2+spacer sequences ...... 144

D-2 Delta K plot for Structure result shown in Fig. 3-3A...... 145

D-3 Delta K plot for Structure result shown in Fig. 3-3B...... 146

LIST OF ABBREVIATIONS

AE Agro-ecological interface

AFLP Amplified fragment length polymorphism

ANOVA Analysis of variance

Btub Beta (β) tubulin

ENL Enolase gene

HRM High Resolution Melting

Hsp90 Heat shock protein 90

ML Maximum Likelihood

MLS Multi-locus Sequencing

MSN Minimum spanning network

OTU Operational Taxonomic Unit

PCoA Principal components analysis

PHR Pineapple Heart Rot

RAPD Random Amplified Polymorphic DNA

RFLP Restriction fragment length polymorphism

SCAR Sequence Characterized Amplified Region

SSR Single Sequence Repeat

TEF1 Translation elongation factor EF1 alpha

TigA Triosephosphate isomerase/glyceraldehyde-3-phosphate dehydrogenase

TRP1 Tryptophan biosynthesis protein

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

INNOVATIVE TOOLS FOR STUDYING PLANT-RELATED OOMYCETE POPULATIONS OF POTENTIAL AND CURRENT THREAT IN AGRICULTURE IN ECUADOR

Maria Fernanda Ratti Torres

May 2018

Chair: Erica M. Goss Major: Plant Pathology

Interactions of emerging pathogens with their hosts are a function of their evolutionary history and ecological context. Three different approaches are described to study oomycete communities associated with economically important crops in Ecuador.

With high-throughput multiplexed sequencing of two universal markers, oomycete species composition was studied in soil samples from cacao farms and contrasted with uncultivated zones. Species of Phytophthora and Pythium and most known pathogens were present in both types of soil, suggesting that zones that border cacao plantations affect the ecology of oomycetes in crop areas. Second, intra-specific genetic variation of

Phytophthora nicotianae populations was analyzed with nuclear and mitochondrial markers. These loci revealed a single clonal lineage from pineapple heart rot in Ecuador that is distinct but closely related to isolates collected from vegetables and ornamentals.

Finally, we designed high resolution melting experiments to discern parental and hybrid

Phytophthora species that represent a risk for agriculture because of their ability to emerge as new pathogens. These tools have the ability to predict the evolutionary potential of oomycete pathogen populations to overcome control methods (1), which

becomes important for understanding their risks and developing management strategies.

CHAPTER 1 OOMYCETE PATHOGENS IN TROPICAL AGRO-SYSTEMS

Emerging Pathogens in Crops and Agro-Ecological Interfaces

The emergence of aggressive plant pathogens can occur rapidly and has increased in frequency due to the expansion of global commerce and other human activities (2). Plants and their pathogens evolve over vast evolutionary time; however, domesticated crops hasten evolution by increasing selection pressure on pathogens that in turn specialize on these crops. As humans have developed new crops, pathogens of wild ancestors of these crops have acquired new strategies to affect them.

Importantly, the genetic uniformity of most crops has enabled rapid host specialization and pathogen spread, leading to faster evolution of pathogens in agro-systems compared to natural systems (3).

Although agriculture is evolutionarily recent, agricultural crops show complex ecological relationships. Likewise, plant pathogens may attack a broad range of hosts or be very specialized. Some pathogens and plant hosts are thought to have evolved together since the beginnings of agriculture ~10000 years ago, while other crop pathogens have recently emerged via host shifts and hybridization. Horizontal genetic transfer and interspecific hybridization are “radical genetic events” that can dramatically change host-pathogen interactions over very short periods of time (4).

Crop-pathogen interactions are influenced by the community context of the interaction. Jones 2009 described the outcomes of relationships among plants and virus pathogens as depending on their interactions both inside and outside agro-systems, with cultivated and uncultivated hosts (5). Disease can be diminished by natural selection for host resistance in co-evolved interactions, by high diversity plant

communities that include non-hosts, and by genetic diversity within hosts. Pathogen movement across cultivated and wild interfaces influences both epidemiological and co- evolutionary processes (6). These dynamics are essential for understanding the evolution of both broad and narrow host range pathogens.

The area of contact between crops and semi-wild vegetation has been termed the agro-ecological (AE) interface (7, 8). AE interfaces can vary in size, complexity and seasonality as a result of agricultural practices. There can be very dramatic or practically indistinguishable transitions between crop and surrounding vegetation, with different consequences on pathogen evolution. In the absence of cultivated hosts, pathogens may survive on crop volunteers or develop other strategies that ensure their survival (8). This may involve colonization of alternative hosts or survival in soil

(including on decaying plant material). For all of the above reasons, AE interfaces may facilitate rapid evolutionary change. Beyond pathogens, they harbor insects, birds, carabid beetles and spiders that can act as pollinators or natural enemies of pests (8-

10). AE interfaces also harbor fungal plant pathogens. Puccinia graminis f. sp. tritici, causal agent of wheat stem rust, complete its sexual cycle and produces primary inoculum on wild barberry. Barberry eradication during the first third of the 20th century significantly reduced the pathogenic diversity of P. graminis f. sp. tritici populations (7).

Fusarium wilt disease in Australian commercial species of cotton (Gossypium hirsutum

L.) arose from infections of wild Gossypium species that were interspersed among cotton fields (8). Despite potentially dramatic impacts on pathogen evolution and epidemiology, the microbial population dynamics of these zones of interaction between crop and wild species have been understudied. To advance understanding of emerging

pathogens, there is a need to investigate the contribution of AE interfaces to pathogen diversity and evolution across many systems.

Overview of Oomycetes

Oomycetes are a group of fungal-like eukaryotes that are closely related to diatoms, brown algae and other Stramenopiles (11). Oomycetes share ecology

(osmotrophic life style) and morphology (hyphal growth) with Fungi, but they produce biflagellated spores, typical of heterokonts, and molecular phylogenies place them in the supergroup Heterokonta (12). Unlike fungi, the cell wall of oomycetes are composed of

β-1,3-glucan polymers and cellulose, and hyphae are coenocytic and devoid of cross walls (13). Oomycetes store energy in form of mycolaminarin, which is also found in kelps and diatoms, and some genera are auxotrophic for sterol and thiamine. Genome size and number of chromosomes vary among oomycete genera. Genomes are around

58% GC, have few introns in genes, and are rich in repetitive sequences. Oomycetes are also known for their extreme phenotypic variation, even within strictly asexual lineages. The genetic basis of this variation is not fully understood, but may be generated by mitotic recombination, gene conversion, transposable elements or even expendable chromosomes (14). During most of their life cycle they are diploid and reproduction can be asexual by means of sporangia or sexual by male and female gametangia, respectively, the antheridium and oogonium. While sporangia allow rapid reproduction and dispersal (13), sexual reproduction generates genotypic diversity and thick-walled sexual spores serve as overseasoning survival structures.

Oomycetes include obligate biotrophs, hemibiotrophs and necrotrophs. Many are saprophytic in terrestrial or aquatic ecosystems and contribute to nutrient cycling, while those in the orders Peronosporales, Albuginales, Pythiales, Saprolegniales,

Leptomitales, and Rhipidiales are primarily pathogenic (15). Oomycete plant pathogens parasitize crops important for food and economic security (12). The most well-known oomycete may be Phytophthora infestans, the causal agent of late blight in potato, which caused widespread, devastating crop loss in Europe in the 19th century and led to the Irish Potato Famine.

Oomycete species are found in a wide range of environments, rapidly spread across wide geographic areas, and can be very successful in occupying new ecological niches (16). Adequate identification of pathogens across varied landscapes is the first step to implement control strategies for avoiding economic impacts they cause on crops. Universal genetic markers for eukaryotes are responsible for major advances in taxonomic and phylogenetic analysis of oomycetes and have renewed biodiversity studies of this group (17). New DNA-dependent technologies and the availability of sequence databases have increased the interest in developing tools to identify emerging oomycete pathogens. DNA markers have advantages in not requiring culturing of organisms and have the potential to produce standard methods that can be replicated within and among environments. Platforms supporting taxon identification and molecular systematics are now available for many different organisms using barcode- like approaches (18, 19). For many eukaryotes, cytochrome oxidase genes are the primary barcode sequence. Cytochrome c oxidase subunit I (COI or cox1) has been used as an alternative to the spacers between ribosomal subunits (ITS) for oomycete identification. This mitochondrial gene was proposed as the primary marker for oomycete identification due to its resolution of species of Pythium and Phytophthora

(16). Later, cox2 was suggested for DNA barcoding of oomycetes, due to its utility with downy mildew pathogens in addition to Pythium and Phytophthora (20).

Oomycete Communities in Tropical Soil: Cacao Farms

Metabarcoding studies of environmental samples is gaining space in biodiversity studies. This approach is already commonly used to describe microbial inhabitants in soil, which were often viewed as a “black box” in ecological studies. Massive sequencing of DNA barcodes provides a new tool to begin to unravel the interactions between microorganisms, how microorganisms affect ecosystem processes, and the effects of environmental factors on soil communities (21, 22). These approaches have been applied to soilborne plant pathogens and are providing robust data on microbial community diversity and pathogen biology in agroecosystems, including temporal and spatial variation in soil communities. Most metabarcoding studies use characterization of Operational Taxonomic Units (OTUs) based on specific genomic markers, and assume each OTU can be identified through a specific DNA sequence (23).

In agricultural soil, microbial communities are strongly influenced by plants because they provide exudates and litter as sources of organic matter; however, other edaphic factors also impact microbial diversity. In plantations of cacao, Theobroma cacao, management systems and cover cropping contribute to variation in the composition of bacterial, archaeal and fungal communities (22). However, oomycete pathogens in cacao plantations have not been fully characterized. The majority of oomycete species that have been described and characterized occupy terrestrial niches associated with agriculture (24), and cacao soils are not an exception. Phytophthora palmivora and P. megakarya are cacao pathogens that cause severe yield losses every year. Cacao is the only known host for P. megakarya, which is only present in West and

Central Africa (25, 26). In contrast, P. palmivora has a broad host range of hosts including tropical fruit crops, monocots and ornamentals and is globally distributed (27).

Presumably, that flexibility allows the establishment of P. palmivora outside of cultivated areas. Two other Phytophthora species that have been recovered from cacao are P. capsici and P. tropicalis. Both are known pathogens in Phytophthora clade 2 and, despite being morphologically similar, are phylogenetically diverged and differ in host range (28).

Ecuador is one of the eight global crop domestication centers described by

Vavilov (5); therefore, it is likely a center of pathogen evolution as well. This makes it an important location to study pathogens across wild and agro-ecosystems, as the diverse native flora would be expected to facilitate rapid evolution. Thus, the establishment of baseline knowledge of major genera of oomycetes in Ecuador could advance studies of plant-pathogen interactions, phylogeography and epidemiology in diverse plant communities.

Phytophthora nicotianae, Another Versatile Pathogen of Tropical and Subtropical Crops

Phytophthora nicotianae Breda de Haan (syn. P. parasitica) infects numerous hosts and survives for long periods of time in the soil, causing polycyclic outbreaks that ultimately result in plant mortality and reduction of crop density (29-31). P. nicotianae has been isolated from five continents and in numerous natural, agricultural and urban ecosystems. It is common in irrigation systems and is widely disseminated by commercial trade of ornamentals (32). Despite its wide host range, the population biology of P. nicotianae has been studied mostly in tobacco and citrus production.

One of the hosts that is affected by P. nicotianae in tropical agro-systems is pineapple (Ananas comosus). The history of pineapple begins before Spanish colonization of the Americas. European explorers brought the fruit back to Europe, but by that time, the fruit was domesticated and well known to the indigenous people of

South America and Caribbean (33). Nevertheless, the Spanish, Portuguese and Dutch were responsible for its worldwide distribution. There are an assortment of pineapple varieties currently available; Ecuadorian farmers prefer the MD-2 cultivar due to its sweetness, size consistency and preference for consumption. The MD-2 hybrid was developed by Del Monte Fresh Produce Hawaii Inc. from the cross of two PRI

(Pineapple Research Institute) hybrids, 58–1184 and 59–443. MD-2 is characterized by yellow-green leaves with reddish tips and fruits that range from 1.3-2.5 Kg with an intense yellow pulp. MD-2 is susceptible to P. cinnamomi and P. nicotianae (34).

Classically, studying populations of plant pathogens has required the development of genetic markers to infer species identity, classify races, study resistance to chemical controls, designate mating types or assign individuals to structured populations. Phytophthora species have been studied with an array of markers, including phenotypic markers like mating type and specific virulence, and molecular markers such as allozymes, RFLPs, mitochondrial haplotypes, AFLPs, SSR and SNPs. Among these, simple sequence repeats (SSR) and single nucleotide polymorphisms (SNP) in housekeeping genes remain the most widely used markers for

Phytophthora population genetics (4, 35).

Microsatellites, or simple-sequence repeats (SSR), are a subcategory of tandem repeats located in prokaryotes and eukaryotes. They are widely distributed in genomes

and highly polymorphic, hence, they are commonly used PCR-based genetic markers for population biology studies (4, 36). SSRs are more common in noncoding than coding regions, which make them desirable markers for fingerprinting, genetic mapping and genetic structure analyses that assume neutral evolution (37). Polymorphic loci usually emerge from changes in the number of repeat units (motifs) by additions or deletions caused by polymerase slippage or recombination errors (36). Replication slippages arise when DNA temporarily dissociates and then misalign during renaturation. Mutation rates increase with the number of repeats as slippages are more likely to occur (4). Development and optimization of an SSR panel can be costly and time consuming, although the advancement of sequencing technologies and bioinformatics applications has considerably improved this process. SSR genotyping is considered a low throughput method; however, PCR multiplexing has allowed many markers to be genotyped in a single reaction (38). Microsatellites have been developed for the study of P. nicotianae populations, revealing significant population structure that suggests possible host specialization (39). The pathogen is widely distributed and appears to have different population structures across different geographic regions and hosts. To better understand the drivers of P. nicotianae population biology and evolution, focused studies by host and geographic location are required.

Hybrid Phytophthora Species

Hybridization, by sexual or parasexual recombination between genetically distant individuals, combines the genetic information of two diverged genomes (40). At its most simple, a hybrid is the result of breeding between different species (41). Speciation requires the establishment of reproductive isolation (42); thus, organisms from different

geographical areas are more likely to hybridize because of weak or absent genetic barriers (40).

Hybridization among plant pathogenic fungi has been hypothesized since the early 1930s when Flor noted atypical morphological phenotypes in Tilletia species (43).

Interspecific hybrids have been reported under natural and laboratory conditions for the past 20-30 years; but the strongest evidence of this phenomenon became available in the molecular era. In the mid-1990s, hybrid species were detected in Ascomycota and

Basidiomycota fungi as well as Phytophthora and Pythium genera (44). These hybrids are individuals with intermediate phenotypes that have been formed by sexual mating or parasexuality (4). Sympatric species that reproduce sexually generally have strong reproductive barriers compared to allopatric species, thereby preventing hybridization and maintaining genetic isolation (4, 40). However, human transportation and other forms of migration have made encounters between previously allopatric species possible. As a result, hybrids are formed and may emerge as new species when there is reproductive isolation from parental species (40). These hybrids may have different host ranges and pathogenicity from the parental species. Fungi and oomycetes appear to hybridize readily, possibly due to their genome plasticity (41, 45, 46). In many cases, resulting hybrids are polyploids and aneuploids (4). Cytogenetic studies have determined abnormal number of chromosomes; in contrast, molecular methods rely on finding multiple alleles compared to the parents (4). In allopolyploid organisms, new features may emerge and disadvantageous alleles can be buffered by extra gene copies (47). In contrast, homoploid hybrid speciation, which does not include change in chromosome number, is usually accomplished by geographic isolation, niche

divergence and rearrangement of chromosomes (42, 48). Host range change via hybridization has been explored through laboratory crosses between Phytophthora mirabilis and P. infestans, which produced progeny that were not pathogenic on

Mirabilis jalapa (host of P. mirabilis) or potato (host of P. infestans) but some progeny could infect tomato (as P. infestans) (45). These results suggested that these species are sexually compatible and diverged via host specialization (44). Hybrids represent a specific pathogenic risk due to their capacity to colonize new hosts, and hybridization is a mechanism of pathogen emergence (46).

The following hybrid Phytophthora species have been formally reported: P. alni

(from P. cambivora and a taxon related to P. fragarie) (49-51), P. andina (from P. infestans and an unknown parent related to P. mirabilis and P. ipomoeae) (52, 53), P. xpelgrandis (P. cactorum X P. nicotianae) (47, 54), P. xserendipita (P. cactorum X P. hedraiandra) (47), P. xstagnum (P. chlamydospora and a species related to P. mississippiae) (55), possibly P. meadii (56), among other unnamed hybrids in clade 6 and 8 (57, 58). Also, hybrids produced in laboratory conditions have been described: P. infestans X P.mirabilis, P. nicotianae X P. capsici, P. sojae X P. vignae and P. capsici X

P. tropicalis. Methods for their production included culture pairing, zoospore fusion and nuclear transplantation (55).

The alder pathogen P. alni is one of the most studied Phytophthora hybrids. The hybrid was identified by variable numbers of chromosomes among isolates, which doubled the number of chromosomes compared to its putative parental species (4). P. xserendipita was described in 2007 as a hybrid between P. cactorum and P. hedraiandra by analysis of single-strand conformation polymorphism (SCCP) of the ITS

region, showing an intermediate profile between both parental species (47). In Virginia, a hybrid found in irrigation water in ornamental nurseries that was confirmed by cloning and analysis of ITS and beta tubulin DNA sequences. The authors named it

Phytophthora xstagnum and its parental species are believed to be P. taxon

PgChlamydo (now P. chlamydospora (59)) and a species genetically close to P. mississippiae (55). P. andina and P. xpelgrandis are described in detail in Chapter 4.

In Phytophthora, the lack of morphological characteristics makes interspecific hybridization difficult to detect (44). Cytology studies have been used to demonstrate polyploidy or an unstable chromosome structure as result of hybridization (56).

Polyploidy may remain stable or be a temporary phase before returning to a diploid state. Although not completely understood, it is believed that polyploidy is related to hybridization (58). Flow cytometry has been used to detect DNA content variation and heterokaryosis in P. infestans (60) and P. andina (61).

Interspecific hybrids of Phytophthora are difficult to definitively identify by means of phenotypic characters. However, modern molecular techniques have produced a dramatic increase in the number of putative hybrids (62). While sequencing continues to be used to initially characterize interspecific hybrids, there is a need for routine differentiation of hybrids from parental species, to understand the ecological interactions of hybrid species relative to their parental species.

Objectives

The objectives of this research were to determine species richness, composition, and structure of major groups of oomycetes associated with commercially important crops in Ecuador with innovative methods that take advantage of genetic markers available for oomycete research.

CHAPTER 2 EVALUATION OF TWO BARCODING MARKERS FOR HIGH THROUGHPUT SEQUENCING OF OOMYCETES AS APPLIED TO CACAO AGRO-ECOSYSTEMS IN ECUADOR

Introduction

Oomycetes are ubiquitous both in natural ecosystems and crops, but diversity in forests and other natural ecosystems has mainly been examined in temperate regions

(2). The best-known genus of oomycetes is Phytophthora, whose species cause severe economic losses in crop plants and natural ecosystems. This genus includes more than

100 species that together affect a broad range of dicotyledonous and a few monocotyledonous plants (14, 15). The infamous P. infestans changed the course of human history when it attacked potato crops in Ireland and provoked famines and massive migrations (63, 64). Other notable species are P. sojae causing root rot in soybean; P. cinnamomi, which is responsible of several dieback and root rot diseases, and P. megakarya and P. palmivora, which cause black pod on cacao (14). Although they cause less lethal diseases than Phytophthora, Pythium species are important pathogens responsible for early damping-off and reducing vigor of mature plants. With

~120 species, they are considered common soil inhabitants occupying a wide spectrum of ecological niches, with life styles ranging from saprophytic to pathogenic (65, 66). In addition to Phytophthora and Pythium species, other oomycetes are able to survive in soil and many have been described as plant pathogens. However, their abundance and distribution are not well known.

Systematic sampling of natural ecosystems for Phytophthora species has revealed extensive, previously unknown species diversity (67). In agricultural ecosystems multiple Phytophthora species can be found; thus, it is expected that

natural ecosystems contain higher diversity. Results from different surveys of temperate forests and streams show similar numbers of species among studies but few overlapping taxa. For example, species composition in Western Australia (mostly from

Phytophthora clade 1 and 2) is different from temperate forests of South Africa (P. cinnamomi), South America (clade 6 and 9) and Asia (clade 2 and clade 6) (67-70).

Subtropical genera Halophytophthora and Salisapilia species are common in saline habitats (24). Studies of Pythium species composition has focused more on agricultural fields, irrigation systems, greenhouses and nurseries than natural ecosystems. Tropical and subtropical regions have not been extensively explored, especially beyond agriculture frontiers. In Ecuador, oomycete species have been reported as disease causing agents of important crops; nonetheless, few studies have occurred in wild plant habitats. Among these, a hybrid Phytophthora species, P. andina, was first isolated from wild solanum adjacent to potato crops in the Andean region (71). Tropical ecosystems, besides being important sources of plant pathogens, have received little attention as putative centers of oomycete diversity.

Cacao (Theobroma cacao) is a tropical perennial crop native to South America

(72). Although the center of domestication is Central America, no wild plants have been found there; thus, it has been hypothesized that the origin is in the region of Orinoco and the Amazon basin (73). Cacao farms are characterized by low plants and a canopy of shade trees. The leaf litter is usually retained, which provides organic matter that serves as a resource for soil inhabiting organisms (74). Researchers have examined biodiversity of animals and plants in agroforests (75), as well as soil microbial diversity

(22). However, the impact of the forest in and around T. cacao plants on plant pathogenic taxa is not clear.

Agricultural soil is mostly inhabited by communities whose niches are defined by their relationship with cropped plants. These systems are dominated by organisms that depend on crop plants for survival. Meanwhile, agro-ecological (AE) interfaces are the remnants of natural areas that interact directly or indirectly with plant crops. Their influence has been recognized in terms of arthropod pests and predators, as well as pathogenicity and resistance genes. These areas connect crop and wild plants and are important for the establishment of new pathogens (6, 7). The movement of fauna from patches of non-agricultural soil to crops is well known (76), which suggests that crop edges do not prevent the interplay between these two environments. Pathogens may move between these environments in water and wind. AEs contain alternative hosts

(weeds, volunteers or wild species very similar to cropped plants) for crop pathogens and, thus, may maintain genetic diversity in crop pathogen populations (8). The presence of pathogens in these uncultivated areas usually goes unnoticed because weeds may not show symptoms of infection or are not scouted for disease (77).

New culture-independent methods for assessing taxonomic diversity of microbes have the potential to provide large amounts of data for comparisons within and among communities. Genetic markers have been adapted for high throughput sequencing studies of oomycete communities. Most efforts to develop markers for culture- independent identification have used the ITS regions of the ribosomal DNA, mitochondrial cytochrome oxidase (cox) genes and spacers, and nuclear Ras-related

Ypt1 introns (78). For many eukaryotes, the mitochondrial cytochrome oxidase genes

are the primary barcode sequence. The use of cytochrome c oxidase subunit I (cox1) was proposed as the ultimate toolbox for Stramenopiles identification due to its resolution in Pythium and Phytophthora (16). There is also a history of phylogeny and diversity studies using cox2, because of the higher interspecific and intraspecific variation in this gene (20, 24, 79, 80). Later, cox2 was suggested for DNA barcoding instead of cox1 as universal Oomycete barcode, due to its ability to also cover downy mildew pathogens (20). However, its use has been limited to culture collections, and the specificity of cox2 primers for use on environmental samples has not been considered.

The most extensively used marker is ITS. Cooke et al., 2000 produced a dataset of ITS sequences for oomycetes, which became the reference for posterior studies (62).

However, some distinct species have very similar ITS sequences, which has hindered identification. For this reason, other markers have been developed. Robideau et al.

(2011) indicated that cox1 showed better resolution than ITS and the nuclear large subunit (LSU), but acknowledged the versatility of ITS if used in isolates rather than environmental samples (16). For high-throughput sequencing, Riit et al. 2016 modified standard ITS1 primers to make them more oomycete-specific. The resulting ITS1oo and

ITS4ngs were tested with twenty soil samples and infected plant tissue, recovering more than 400 oomycete OTUs from forest nurseries (81).

The main objective of this study was to determine oomycete species relative abundance and composition based on cox2 and ITS massive amplicon sequencing in agriculture soil as compared to AE interfaces of cacao farms in Ecuador. We had three sub-objectives: (1) test the utility of cox2 as a taxonomic barcode for high-throughput sequencing for oomycetes and compare the results to sequencing with primers ITS1oo-

ITS4ngs; (2) describe oomycete communities in cacao farms and their associated AEs; and (3) test for differences in oomycete communities in soil from cacao versus neighboring AEs. We hypothesized that the type of land use would determine the composition and relative abundance of oomycetes present in soil.

Materials and Methods

Sampling Sites and Strategies

Eight farms were sampled, covering three precincts of the Coastal Region of

Ecuador. From each farm, nine samples were taken from cultivated (C) soil and nine from neighboring soil (N). From cultivated soil, each sample was composed of three subsamples circling a cacao tree and taken at 1 m distance from the trunk, where oomycete communities interact with root systems. From neighboring soil, in zones adjacent to cacao plots, each sample was composed of 3 subsamples encircling cultivated trees, shrubs or weeds. Here, oomycetes may be associating with a more diverse flora. Table 1 summarizes the characteristics of sampling sites and locations of the regions considered in this study, which belong to two provinces of Ecuador: Guayas and Los Rios. Fig. B-1 describes the geographic locations of sampling sites in Ecuador.

Soil was collected with small shovels that were disinfected with 10% bleach after each use. Samples were taken at ~5 cm of depth and placed in sealable bags, labeled and kept cool until further processing.

Library Preparation and Sequencing

Soil samples were homogenized and consolidated into 3 samples per zone per site before DNA extraction with MO-BIO PowerSoil DNA isolation kit (MO BIO

Laboratories, Carlsbad, CA). DNA was measured with a spectrophotometer and aliquoted at ~15 ng/µL.

Libraries were constructed by a three-step PCR modified from Lundberg et al.,

2013 (82), wherein: 1) the locus of interest was enriched, 2) frameshift primers were added to add complexity to the sequence run, and 3) unique Golay barcodes added to each sample (83). Single-indexing, wherein a single barcode is added, was used during the third step. In this approach, costs are reduced, because barcoded primers are not amplicon-specific. For the first step, primer sets for ITS regions were ITS1oo and

ITS4ngs (81) and for cox2, Cox2-F and Cox2-RC4 (20). For the second step, primer sets contained sequencing and frameshift sections in addition to the amplicon-based region, and for the third step, the forward primer was the same for all samples, but the reverse contained the unique barcode to identify the sample during bioinformatics analysis. Primer names and sequences are detailed in Table 2.

PCR conditions used for cox2 amplicons were as follows: PCR1 was a 12.5 µL reaction containing 1 X Buffer, 2 mM MgCl2, 0.2 mM dNTP mix, 0.2 µM each primer,

0.05 U Phusion High Fidelity Polymerase (New England Biolabs, Ipswich, MA) and 5-15 ng DNA. Cycling was as follows: 95C for 1 min, 10 cycles of 95C 30 s, 51C 30 s,

72C 30 s, followed by an extension step of 72C for 5min. PCR2 was also in a reaction of 12.5 µL with 1X Buffer, 2 mM MgCl2, 0.2 mM dNTP mix, 0.2 µM of primer mixes (one mix included 6 forward and the other had 6 reverse primers), 0.05 U Phusion polymerase, and 2.5 µL of PCR1 product as template. Cycling program was 95C for 3 min, 20 cycles of 95C 30 s, 70C 45 s, 72C 45 s, then 5 min of extension at 72C.

PCR3 was done in 50 µL reactions including 1X Buffer, 2 mM dNTP mix, 0.1 µM each primer (barcoded primer was added individually), 0.1 U Phusion polymerase and 6 µL of

PCR2 product. Cycling included denaturation for 3 min at 95C followed by 10 cycles of

95C for 30 s, 70C for 45 s and 72C for 45 s, then an extension step of 5 min at 72C.

PCR conditions used for ITS amplicons were similar to cox2 but instead of using

Phusion polymerase and reagents, we used AccuStart ToughmixII (Quanta Biosciences

Inc., Gaithersburg, MD). Reactions for ITS used 20 cycles in PCR1, with annealing temperature of 55C. PCR2 and 3 had the same setup as cox2 but with 10 cycles each.

After the third PCR, amplicons were visualized in 1% agarose gels. Cleaning of products was done with AMPure XP reagent (Beckman Coulter, Brea, CA, USA) following manufacturer instructions for Agencourt SPRIPlate 96 Super Magnet Plate using 30 µL of PCR3 product. Once the products were clean, we quantified each sample individually using Quant-IT Pico-Green dsDNA reagent measured in a Synergy plate reader (BioTek®, Winooski, VT, USA) adapted from a BioTek application note by

Brescia and Banks, 2010 (84). Pooling of samples before submission for sequencing was done by adjusting each sampling to a concentration of ~12 ng/µL and adding 10 µL of each sample to a 1.5 µL microcentrifuge tube. Libraries were submitted to Illumina

MiSeq 2x300, for which sequences of ~600 bp (cox2) and ~750 bp (ITS) were obtained.

To test for sequencing bias and accuracy of genetic markers, a mock community comprising three known Phytophthora species was included per marker. The DNA of each Phytophthora isolate was amplified separately using the same procedures as described above, but using the same barcode; then, the amplicons were visualized in agarose gel, quantified and added in equal proportions. For the cox2 marker, the mock community was composed of P. andina (isolate 2309, Ecuador), P. nicotinae (isolate

Pn1, USA) and P. megakarya (isolate Pm1, Nigeria). The ITS mock community was

composed of P. andina (isolate 2470, Ecuador), P. nicotianae (isolate G194, Ecuador) and P. cactorum (isolate 11-1, USA).

Database Assemblage and Taxonomy Assignment of Sequence Reads

Two databases were assembled for use with cox2 and ITS markers. These databases included sequences and identifiers, and additional files for assigning taxonomy, in which the identifiers match with the name of corresponding taxa. cox2 database identifiers were GenInfo Identifiers (GIs), which were used for the function

Fetch Taxonomy Representation of Metagenomics Analysis section in Galaxy. The curated data from PhytophthoraDB did not have GIs and were coded starting at

990000001. ITS database identifiers were GenBank accession numbers and records from PhytophthoraID used the same system. Sequences longer than 2000 bp were removed due to their slowing the OTU picking process.

The cox2 and ITS databases were assembled from available sequences from

GenBank (http://www.ncbi.nlm.nih.gov), PhytophthoraDB

(http://www.PhytophthoraDB.org) and PhytophthoraID.org (85). Noncurated and curated sequences from GenBank and PhytophthoraDB, respectively, were used to assemble a cox2 database, due to the limited availability of Oomycete taxa in other repositories outside of GenBank. A curated database (from PhytophthoraID.org) and a noncurated database were assembled for the ITS database. Sequences were organized in fasta files that were merged and parsed with text editing tools in BioLinux 8 (Ubuntu

16.04.03) to remove descriptors and keep only one identifier/sequence per record. In addition, we reformatted databases to make them compatible with BLASTn tool in the

Galaxy framework (86).

Sequencing reads were quality checked using FastQC (Andrews S., 2010). Our sequences are larger than the optimum length for Illumina Miseq 2x300; therefore, we analyzed the read that had better quality and yielded more sequences. For cox2 sequences we used R2, whereas for ITS we used R1. All further analyses were done in qiime/1.9.1 (87), which combines bioinformatics tools for different steps of metagenomics studies.

We used split_libraries_fastq.py command to apply quality filters based on Phred quality scores (88, 89) with at least a value of 14. Files were duplicated to add qiime labels with command add_qiime_labels.py to produce one file with all samples merged with unique identifiers. When R2 was used, sequences were reverse complimented with adjust_seq_orientation.py before conducting OTU picking with pick_open_reference_otus.py. The usearch61 method was used to match the sequences against our custom Oomycete sequence databases (90).

Sequences that did not match the reference databases during open reference

OTU picking were denominated failures.fasta and searched against custom BLAST databases in Galaxy using BLASTn search with a minimum of 90% identity and 50% coverage of query sequence.

Statistical Analyses

We rarefied OTU tables with known taxonomy (single_rarefaction.py) to the lowest number of reads per sample to construct bar charts for species composition

(summarize_taxa_through_plots.py). Unrarefied data were used in combination with

BLAST results to generate bar charts representing total number of reads and relative abundance of taxa per farm and sample.

Simpson’s D diversity index (91) was calculated based on species richness and proportion of taxa for each sample using the unrarefied OTU table generated with cox2 marker. Variation in Simpson’s index was analyzed with two-way analysis of variance

(ANOVA) using factors land use, farm, and their interaction.

To examine variation by land use, unrarefied OTU tables were edited to perform

PERMANOVA (Permutational ANOVA) (92) with Bray-Curtis dissimilarity (93) and 1000 permutations using the adonis function in the R package vegan (94). For each markers, we analyzed the effect of land use on: 1) OTU relative abundance using Bray-Curtis dissimilarity and 2) OTU composition (presence/absence) using the Jaccard index.

There had to be at least 2 reads per OTU to be included in the analysis: ITS used 9 taxa, whereas cox2 used 8 taxa. ITS samples considered only 5 farms due to low read counts. The cox2 analysis used all 8 farms. To test the effect of geographic region on

OTU relative abundance and composition, we combined the three samples by land use

(crop or AE) for each farm and tested factors land use, region, and their interaction. For cox2, OTUs were also analyzed using the three replicate samples per farm and land use, and examined the main effects of land use, farm, and interaction between land use and farm. Non-metric multidimensional scaling (NMDS) was used to visualize differences in read numbers of taxa between crop samples and AEs based on Bray-

Curtis dissimilarity using R packages vegan and MASS (95). Because NMDS uses rank-orders instead of relative abundance of OTUs, it is a suitable way to represent non-linear data, in this case, samples with high variation in read numbers.

Results

Quality of Reads and General Statistics of Runs

The cox2 raw fastq files obtained from the MiSeq instrument contained a total of

~26,700,000 reads. According to FastQC, quality scores of the reads were >28 until cycle 150, where they started to drop. After filtering for quality, 5,290,954 reads were retained, and from those, ~3.9 million reads were OTU-picked resulting in 88,291 reads with known taxa. The raw Mock sample (not considered in the total count described above) contained ~920,000 reads.

The ITS raw fastq files contained a total of ~3,400,000 reads with quality scores under 28 after 180-200 cycles. Thus, the quality was better than cox2 reads. However, an issue with barcode sequencing displaced ~3,000,000 reads into unassigned samples. These unassigned reads were searched with BLASTn, which verified that they matched our database, but were discarded because efforts to rescue the barcodes were not successful. After quality filtering, ~388,000 assigned reads were retained and 2073 known OTUs were assigned after open reference OTU picking. The mock sample contained 320 reads and the blank sample contained 83 reads.

Mock Communities

When we OTU-picked the cox2 mock community with usearch61 method,

202,713 of 923,572 reads were assigned to known OTUs belonging to the community we assembled: P. andina (and closely related P. infestans, P. mirabilis, P. phaseoli and

P. ipomoeae), P. nicotianae and P. megakarya. However, 135 reads (0.01% of assigned reads) matched to P. meadii (30), P. citrophthora (78), P. botryosa (7), P. iranica (4), P. tropicalis (1) and Pythium splendens (15). This discordance with the expected mock community indicates the possibility of obtaining false hits at low rates

and principally at species level. Reads that were discarded during this process were blasted against our database, from there, 134,176 reads with ≥97% ID matched the expected records, except for 26 that hit P. meadii (1), P. colocasiae (1), Phytopythium vexans (12), Pythium splendens (10), Pythium catenulatum (1) and Pythium arrhenomanes (1).

The ITS mock community raw fastq file consisted of 389 reads. After OTU picking with the usearch61 method, 242 were assigned to known OTUs, 223 to P. infestans (and 2 reads of closely related P. mirabilis) and 14 P. nicotianae; no reads were assigned to P. cactorum. After BLAST search with the reads that failed to be assigned with usearch, 47 hits were found with at least 97% ID, from which 6 were reported as P. nicotianae and 39 were P. infestans. In addition, 1 was identified as

Phytophthora sp. and 1 as Pythium graminicola.

Species Composition of Soil Samples

Species richness is expressed in putative species assigned by usearch61. With this method, only 10 species were assigned with cox2 and 7 with ITS. Total numbers of

OTUs and species assigned for each sample for cox2 and ITS are summarized in

Tables 2-3 and 2-4, respectively. OTU tables for cox2 and ITS markers are shown in appendices A-1 and A-2, respectively.

Total numbers of reads (assigned to species) for cox2 showed considerable variation among samples. The numbers of reads in cacao soil (samples designated C) ranged from 24 to 4930, while in AEs (N samples) ranged from 11 to 53407. The sample with the highest number of reads was AE sample 6N2, the second highest was

AE sample 3N1 at 5462 OTUs. There was no apparent trend in which type of sample

(AE or cacao crop) produced more reads. Rarefied data were used to examine relative

abundance of OTUs by farm (Figure 2-1) with taxa represented by clade as defined by

Blair et al. 2008 (96) and Lévesque and Cook 2004 (97).

The cox2 marker identified 48 known OTUs that were assigned to 10 species in the genera Pythium, Phytophthora or Phytopythium. Pythium clades included P. splendens (clade I), P. catenulatum, P. torulosum (clade B1a), P. arrhenomanes, P. aristoporum (clade B1e), and P. myriotylum (clade B1c). Also found was Phytopythium vexans (98, 99), previously Pythium vexans in Pythium clade K. Only three

Phytophthora clades were detected; clade 1 included P. infestans, P. andina, P. iranica and P. nicotianae; clade 2 P. capsici, P. botryose and P. tropicalis; and clade 4 P. quercetorum, P. megakarya, P. palmivora. Other genera were not assigned to species

(Fig. 1). Clade I was the most abundant Pythium taxon, whilst clade 4 was the most abundant Phytophthora. For samples 1C, 3N, 6N and 8N, the majority of reads belonged to one abundant OTU. There were two OTUs that were assigned to P. megakarya, but we believe it is more likely that these belong to the closely related P. palmivora.

From 3,974,359 OTUs, 88,291 (2.22%) corresponded to known oomycete species/clades. Sequence reads that were not assigned as known taxa during OTU- picking with usearch61 were used for BLAST search against our databases with lower identity threshold. The purpose was to place oomycete data that would not be recognized as known species, but may come from oomycete orders with poor database representation or taxa that have not been described. The OTU picking (closed reference, usearch61) discarded ~1,600,000 reads that were used for BLAST (these reads included also the de novo OTUs), for which the % identity required for

identification was lowered to 90, to include new or misidentified oomycetes. Unrarefied data from OTU-picking were merged with the BLAST hits. The bar charts of relative abundance show the diversity of mostly low-frequency genera that were not initially assigned to species (Figures 2-2 and 2-3). The BLAST hits (at 90% identity) are shown in Table 2-5.

For the ITS data, there were fewer reads, therefore, variation in numbers of reads were less than for the cox2 data. The range of read numbers assigned to taxa were 1-292 for crop samples, and 1-245 for AE samples. Almost half of the samples

(21) resulted in 0 OTUs assigned to a species (1C2, 1N3, 2C3, 2N1, 3C1, 3N2, 3N3,

4N1, 4N2, 4N3, 5C2, 5N2, 5N3, 6C1, 6N1, 6N2, 7C1, 7N1, 7N3, 8C1 and 8N2). Due to the barcoding error, assigned reads for each sample using the ITS marker ranged from

0 to 53,407 in 7N2. AE samples from four farms (2N, 4N, 5N and 7N) were discarded due to low or no reads. Rarefied data were used for construction of a relative abundance bar chart (Figure 2-4) for each farm, with taxa described as clades.

The ITS marker identified 51 OTUs representing 5 genera (Achlya, Dictyuchus,

Pythium, Phytophthora and Phytopythium) and 21 different species that were reduced to 16 taxa after grouping by clade. Achlya species included A. bisexualis and A. prolifera. Pythium clades included P. deliense (clade A), P. catenulatum, P. torulosum

(B1a), P. myriotylum (B1c), P. inflatum (B1d), P. aristosporum, P. arrhenomanes (B1e),

P. acanthophoron (B1d), P. splendens (I), and P. acanthophoron (J). There were other

Pythium sp. for which clade was not specified. The Phytopythium species identified was

P. helicoides. Phytophthora clades included: P. infestans, P. nicotianae (clade 1) and P. palmivora (clade 4). Phytophthora clade 1 was the most abundant Phytophthora taxon

using the ITS marker, Pythium clade J was the most abundant Pythium taxon, but a large portion of reads were assigned to Achlya spp. Two Pythium species were found in nearly every sample sequenced for ITS: P. arrhenomanes and P. splendens.

Phytophthora botryosa was also in the majority of samples and in high frequency in 7N2

(Fig. 2).

Similar to cox2 data, from 374,651 OTUs picked, 2073 (0.55%) matched our database. The number of reads discarded after OTU picking (closed reference, usearch61) were ~380,000. Additional genera were found by BLAST using OTUs not assigned to taxa in the database and discarded reads (Figures 2-5 and 2-6, Table 2-6).

The taxa that could be detected with both markers includes the most important

Phytophthora species for cacao production (P. palmivora, P. capsici, P. citrophthora) and many Pythium and Phytopythium that may be surviving in soil (Table 2-7). In addition, a few genera from orders Saprolegniales, one from order Lagenidiales and one from Albuginales were detected by both markers.

Taxa in Relationship with Land Use

In general, most taxa that were present in cacao soil were also present in AE soil. Phytophthora palmivora, causal agent of black pod in cacao, was found both in crop and AE soil, with potentially higher frequency in AEs due specific samples.

Simpson’s diversity index showed lower values in AE relative to Crop samples

(Table 2-8). Simpson’s D is the probability of two randomly selected individuals in the sample belonging to the same taxa. The higher the value, the less diverse the community. ANOVA on Simpson’s D values showed significantly higher diversity in AEs

(P = 0.047; Table 2-9).

We also investigated the effect of land use on OTU relative abundance and composition using PERMANOVA analysis performed with unrarefied reads and presence/absence OTU data. For the cox2 marker, we used OTU data per sample and per farm, whereas missing samples for the ITS only allowed combined samples by farm.

We found a significant effect of land use (crop versus AE) and significant variation among farms using the cox2 dataset (Table 2-10), taking advantage of the three samples per farm/land use. There was not a significant effect of land use when samples per farm were combined (Table 2-11), which suggests that there may not be enough power to reject the null hypothesis when farm samples are combined. We also found no significant effect of region for the combined samples. For the presence/absence data, there was no significant effect of land use for either data set. Because the 6N2 sample was an outlier in read number and relative abundance of taxa, we tested whether removing this sample changed the PERMANOVA results. We saw no changes in statistical significance when 6N2 was removed. For the ITS data, we obtained no significant effect of land use on OTU relative abundance. However, land use significantly impacted the presence/absence of OTUs (Table 2-12). The subtle effect of land use on OTU relative abundance using the cox2 marker can be seen in a NMDS plot using Bray-Curtis dissimilarity (Figure 2-7).

Discussion

Agro-ecological interfaces (AEs), the semi-natural areas that surround and interact with plant crops, are often unvalued in studies of pathogen diversity, yet support diverse plant communities that serve as alternative hosts, reservoirs of inoculum and impact the evolution of agricultural pathogens (7, 77). Understanding pathogen diversity in AEs is the first step in studying interactions between crop pathogens and plant hosts

in AEs and may ultimately contribute to a better understanding of pathogen population biology for management strategies. When they are not affecting crops, communities of oomycete plant pathogens either survive as resting structures or affect weeds and other wild plants. Although oomycetes are important plant pathogens, they are understudied in tropical regions and are generally underrepresented in soil microbiome studies. In the present study, the composition and distribution of oomycetes was examined soil from cacao and associated natural systems in Ecuador.

Cacao pathogens were not limited to the cultivated zones. According to our combined data from usearch61 OTU-picking and BLAST identification, cacao pathogens or their close relatives were present in cacao soil. Phytophthora palmivora,

P. megakarya, P. citrophthora, P. arecae, P. capsici, P. nicotianae and P. megasperma have been isolated from cacao (100). P. arecae is a synonym of P. palmivora, so we only report P. palmivora (101). Phytophthora palmivora was the most abundant OTU found overall, followed by P. nicotianae and closely by P. megasperma. P. palmivora was abundant in both type of samples but mostly in AEs areas, which is consistent with our hypothesis that inoculum of this wide host range species would be associated with plants in addition to cacao. P. nicotianae was prevalent, especially in AEs from farms 5 and 8, which differed considerably in their flora. P. nicotianae has a wide host range and it is not surprising that it was present in semi-natural areas, yet the impact of P. nicotianae in cacao plantations in Ecuador has not been examined in previous publications. P. megasperma was mainly found in farms 5 and 6 (South East Guayas) and did not show an obvious preference for land use type. P. capsici was found among the assigned OTUs in crop soil, whereas we had only one read of P. citrophthora.

Several species share similar cox2 sequences; thus, results by species should be interpreted with caution. We found OTUs assigned to clade 4 species P. palmivora, P. megakarya and P. quercetorum but their presence should be confirmed with isolations, especially for P. megakarya, which to date has been only reported in West Africa.

Pythium species are less lethal to adult plants than Phytophthora, but there are exceptions. In general, Pythium species cause damping-off of seedlings of many plant genera. Pythium splendens was the most common taxon assigned from this genus followed by Pythium arrhenomanes and Pythium irregulare. P. splendens was found in every sample that was analyzed and seems to be more abundant in crop than AE samples. A similar pattern was noted for P. arrhenomanes, but the opposite pattern was found for P. irregulare.

Phytopythium vexans is a known inhabitant of cacao soil (102), although its role in this ecosystem is not clear. It may cause underground stem canker, but the effects of this pathogen are either mild or undiagnosed. Phytopythium vexans was found in every sample analyzed with cox2, confirming its ubiquitous nature in cacao-related soil. It did not seem to have preference for crop or AE soil.

Other genera were detected, but their read numbers were low compared to

Phytophthora and Pythium. This may be reflect actual abundances in soil or bias in the markers. The databases we assembled included species from other oomycete orders; thus, these methods have the ability to detect “neglected” oomycetes.

For many OTUs identified either marker, taxonomy could not be assigned. Both cox2 and ITS amplify for most eukaryotes, and many nonspecific products were obtained but discarded during the database search. There was also inaccurate

assignment of closely related species with both markers in the mock communities.

Although we identified OTUs that had greater than 97% identity to known taxa, we decreased the resolution of the taxonomic units to clades and subclades to avoid inflating species richness. We chose to use cox2 due to its better resolution of species and ITS due to the greater availability of reference sequences, which is also important for accurate identification. These markers are expected to work well with small amounts of initial material (103) as occurs in soil samples. We evaluated the atp9-nad9 marker

(104), which has successfully identified a wide variety of Phytophthora species from cultured and environmental samples, and might be useful for environmental samples and high-throughput scale due to its short-length (~340bp). Unfortunately no amplicons were produced by this marker (data not shown), possibly due to the low amounts of

Phytophthora DNA that were present in the samples. In summary, the present results indicate that additional work is needed to identify markers that are both sensitive and specific for oomycete detection by high-throughput sequencing of environmental samples.

One of our long-term goals is to understand the factors that shape oomycete composition in agricultural and ecological soil, especially as relates to tropical crops. We found a significant effect of land use in our PERMANOVA analysis of OTU relative abundance based on cox2 and presence/absence with ITS. To make informed management decisions, it is necessary to understand the pathosystems within and beyond crops. The environment impacts diversity and distribution of plant pathogens at different scales; therefore, ecological factors like source of irrigation water and other inputs, as well as surrounding fields need to be considered as sources of inoculum

(105). Oomycete pathogens are major problems in nurseries, and a study of the distribution of Phytophthora in nurseries successfully identified major sources of inoculum of soil borne pathogens (106). However, teasing apart effects in complex environments is challenging. In an extensive study of the community of Oomycetes on soybean seedlings, 16% of the variability was explained by environmental and edaphic factors (105). We found that important plant pathogens were present and relatively abundant in zones adjacent to cacao. The presence of pathogens outside crops indicates the potential for pathogen movement between crop and AE. Specifically, AE could serve as reservoirs of oomycete communities and affect the composition of the community of oomycetes in the crop.

Among the challenges faced in this study were the paucity of high quality reads that could be assigned to OTUs and described taxa. OTUs determined with 90-97% identity do not accurately identify species they hit, but may represent new or poorly identified taxa. Unfortunately, we could not determine the extent of cryptic species present in our samples due to the limited resolution of our markers. Cryptic speciation is common in plant pathogens, but their presence in the less studied AE interfaces requires more research (7). Another consideration is the difficulty with which DNA is extracted from soil. Improvements are needed to ensure that oomycete inoculum is not masked by other, high abundant taxa that are present in soil. The need for meticulous

DNA extraction protocols has been pointed out for fungi (107).

Despite the progress in soil microbiology, fundamental research on associations of crop pathogens with weeds and voluntary vegetation is scarce. This study shows that soils of semi-natural areas are reservoirs for oomycete taxa, including cacao pathogens.

Future research on oomycete pathogens in cacao agro-systems should combine more extensive sampling that allows for increased statistical power and confirmation of the most prevalent species by culturing. Future collections could also combine foliar and rhizosphere samples of plants in AE interfaces. As the tools and technology for culture- dependent and independent identification of microbes progress, and curated databases expand, the analysis of large datasets will be faster and more accurate. However, these advances will need to be accompanied by more specific markers. In addition, analysis of mRNA could be included, to start deciphering the functions of many of the taxa found in this study. Furthermore, pathogens found by these methods could be analyzed with disease severity data to increase knowledge on epidemiology of important oomycete pathogens in tropical crops and better understand disease risks.

Table 2-1. Sampling sites location and characteristics Sample Name Locations Regions/Province Characteristics of AE zones 1 La Nueva Unión 1 - Central Guayas Weeds, herbaceus, flowering plants, cereals 2 San Rafael 1 - Central Guayas Weeds, herbaceus, flowering trees 3 Milagro 3 1 - Central Guayas Weeds, berries, ornamentals, living fence, fruit trees 4 La Sulla 1 2 - South-East Guayas Weeds, shrubs, flowering trees 5 La Sulla 2 2 - South-East Guayas Weeds, woody trees, shrubs, fruit trees Weeds, woody trees, shrubs, fruit trees, ornamentals, 6 La Sulla 3 2 - South-East Guayas palms, semi-natural forest, riparian zone 7 Buena Fe 1 3 - South Los Rios Weeds, shrubs, fruit trees, herbaceus 8 Buena Fe 2 3 - South Los Rios Weeds, African palm, maize

Table 2-2. Names and sequences of primer names used in this study, details on the barcodes and frameshift regions can be found in Caporaso et al. 2012 (59) and Lundberg et al. 2013 (58). Primer Sequence Name ITS1oo GGAAGGATCATTACCACA ITS4ngs TCCTCCGCTTATTGATATGC Cox2-F GGCAAATGGGTTTTCAAGATCC Cox2-RC4 TGATTWAYNCCACAAATTTCRCTACATTG 2xxxF1* ACACTCTTTCCCTACACGACGCTCTTCCGATCT GATCT XXXXXXXXXXXXXXXXXXX** 2xxxF2 ACACTCTTTCCCTACACGACGCTCTTCCGATCT GATC XXXXXXXXXXXXXXXXXXX 2xxxF3 ACACTCTTTCCCTACACGACGCTCTTCCGATCT GAT XXXXXXXXXXXXXXXXXXX 2xxxF4 ACACTCTTTCCCTACACGACGCTCTTCCGATCT GA XXXXXXXXXXXXXXXXXXX 2xxxF5 ACACTCTTTCCCTACACGACGCTCTTCCGATCT G XXXXXXXXXXXXXXXXXXX 2xxxF6 ACACTCTTTCCCTACACGACGCTCTTCCGATCT XXXXXXXXXXXXXXXXXXX 2xxxR1 GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT GATCT XXXXXXXXXXXXXXXXXXX 2xxxR2 GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT GATC XXXXXXXXXXXXXXXXXXX 2xxxR3 GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT GAT XXXXXXXXXXXXXXXXXXX 2xxxR4 GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT GA XXXXXXXXXXXXXXXXXXX 2xxxR5 GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT G XXXXXXXXXXXXXXXXXXX 2xxxR6 GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT XXXXXXXXXXXXXXXXXXX 3cF AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCT CAAGCAGAAGACGGCATACGAGATNNNNNNNNNNNNTGACTGGAGTTCAGACGTGTGCTCTTCCGAT 3cR#*** -CT**** * xxx refers to the locus to be amplified (cox2 or ITS) ** XXXXXXXXXXXXXXXXXXX refers to the amplicon annealing region, which varies between loci. *** # refers to the number of reverse primer according to the barcode used for that particular sample **** NNNNNNNNNNNN refers to the unique barcode per sample

Table 2-3. Number of reads assigned to known taxa and number of taxa per sample using cox2 marker based on usearch61 OTU-picking Total Total Crop number Assigned AE number Assigned Farm Sample of reads species Sample of reads species Nueva Union 1C1 506 6 1N1 57 5 Nueva Union 1C2 445 4 1N2 11 5 Nueva Union 1C3 24 3 1N3 58 6 San Rafael 2C1 3320 5 2N1 91 6 San Rafael 2C2 942 5 2N2 16 3 San Rafael 2C3 254 6 2N3 13 5 Milagro 3 3C1 909 6 3N1 5462 8 Milagro 3 3C2 343 7 3N2 123 4 Milagro 3 3C3 1259 3 3N3 25 4 La Sulla 1 4C1 57 5 4N1 63 3 La Sulla 1 4C2 84 4 4N2 65 5 La Sulla 1 4C3 240 6 4N3 148 5 La Sulla 2 5C1 1342 7 5N1 264 8 La Sulla 2 5C2 408 7 5N2 511 8 La Sulla 2 5C3 309 7 5N3 148 6 La Sulla 3 6C1 321 3 6N1 349 8 La Sulla 3 6C2 373 7 6N2 53407 9 La Sulla 3 6C3 166 6 6N3 278 7 Buena Fe 1 7C1 4930 8 7N1 489 8 Buena Fe 1 7C2 1790 10 7N2 816 9 Buena Fe 1 7C3 1149 7 7N3 484 8 Buena Fe 2 8C1 299 6 8N1 297 5 Buena Fe 2 8C2 1300 5 8N2 324 8 Buena Fe 2 8C3 257 8 8N3 3765 6

Table 2-4. Number of reads assigned to known taxa and number of taxa per sample using ITS marker based on usearch61 OTU-picking Total Total Crop number Assigned AE number Assigned Farm Sample of reads species Sample of reads species Nueva Union 1C1 39 3 1N1 49 7 Nueva Union 1C2 0 0 1N2 1 1 Nueva Union 1C3 1 1 1N3 0 0 San Rafael 2C1 292 7 2N1 0 0 San Rafael 2C2 1 1 2N2 5 2 San Rafael 2C3 0 0 2N3 1 1 Milagro 3 3C1 0 0 3N1 91 3 Milagro 3 3C2 31 1 3N2 0 0 Milagro 3 3C3 29 4 3N3 0 0 La Sulla 1 4C1 104 4 4N1 0 0 La Sulla 1 4C2 6 3 4N2 0 0 La Sulla 1 4C3 84 5 4N3 0 0 La Sulla 2 5C1 38 3 5N1 31 2 La Sulla 2 5C2 0 0 5N2 0 0 La Sulla 2 5C3 146 3 5N3 0 0 La Sulla 3 6C1 0 0 6N1 0 0 La Sulla 3 6C2 162 2 6N2 0 0 La Sulla 3 6C3 5 2 6N3 78 4 Buena Fe 1 7C1 0 0 7N1 0 0 Buena Fe 1 7C2 271 2 7N2 1 1 Buena Fe 1 7C3 2 1 7N3 0 0 Buena Fe 2 8C1 0 0 8N1 245 6 Buena Fe 2 8C2 66 2 8N2 0 0 Buena Fe 2 8C3 73 2 8N3 221 2

Table 2-5. Blast hits with at least 90% identity using cox2 marker Order Putative species Albuginales Albugo candida, Albugo resedae Lagenidiales Lagenidium giganteum, Lagenidium sp, Lagenidium thermophilum Leptomitales Apodachlya pyrifer, Leptomitus lacteus Peronosporales Bremia lactucae, Hyaloperonospora parasitica Sclerosporales Pachymetra chaunorhiza Unknown Peronophythora litchii Peronosporales Peronospora aestivalis, Peronospora aestivalis, Peronospora phacae, Peronospora sanguisorbae, Peronospora sordida, Peronospora trifolii-repentis, Peronospora variabilis, Phytophthora andina, Phytophthora asparagi, Phytophthora austrocedri, Phytophthora bisheria, Phytophthora boehmeriae, Phytophthora botryosa, Phytophthora brassicae, Phytophthora cambivora, Phytophthora captiosa, Phytophthora citrophthora, Phytophthora clandestina, Phytophthora colocasiae, Phytophthora cryptogea, Phytophthora drechsleria, Phytophthora erytrhoseptica, Phytophthora fallax, Phytophthora foliorum, Phytophthora gallica, Phytophthora glovera, Phytophthora hedraiandra, Phytophthora heveae, Phytophthora idaei, Phytophthora ilicis, Phytophthora infestans, Phytophthora insolita, Phytophthora inundata, Phytophthora ipomoeae, Phytophthora iranica, Phytophthora katsurae, Phytophthora kernoviae, Phytophthora lagoariana, Phytophthora macrochlamydospora, Phytophthora meadii, Phytophthora medicaginis, Phytophthora megakarya, Phytophthora megasperma, Phytophthora mengei, Phytophthora mexicana , Phytophthora multivora , Phytophthora nemorosa, Phytophthora nemorosa , Phytophthora nicotianae, Phytophthora obscura, Phytophthora palmivora, Phytophthora parvispora, Phytophthora phaseoli, Phytophthora pisi, Phytophthora plurivora , Phytophthora polonica, Phytophthora primulae, Phytophthora pseudosyringae , Phytophthora quercetorum, Phytophthora ramorum, Phytophthora richardiae, Phytophthora sansomeana, Phytophthora siskiyouensis , Phytophthora sojae, Phytophthora sp, Phytophthora sp canalensis, Phytophthora sp cuyabensis, Phytophthora sp kelmania, Phytophthora sp lagoariana, Phytophthora sp napoensis, Phytophthora sp novaeguine, Phytophthora syringae, Phytophthora syringae, Phytophthora tentaculata, Phytophthora tropicalis, Phytophthora tropicalis , Phytophthora vignae, Phytopythium cucurbitacearum, Phytopythium helicoides, Phytopythium sp, Plasmopara angustiterminalis, Plasmopara densa, Plasmopara sphagneticolae, Pseudoperonospora urticae, Sclerospora graminicola, Viennotia oplismeni

Table 2-5. Continued Order Putative species Pythiales Pythiogeton ramosum, Pythium acanthicum, Pythium acanthophoron, Pythium aphanidermatum, Pythium apinafurcum, Pythium aristosporum, Pythium arrhenomanes, Pythium catenulatum, Pythium chamaihyphon, Pythium conidiophorum, Pythium cylindrosporum, Pythium deliense, Pythium graminicola, Pythium helicoides, Pythium heterothallicum, Pythium inflatum, Pythium insidiosum, Pythium intermedium, Pythium irregulare, Pythium iwagamai, Pythium litorale, Pythium mamillatum, Pythium middletonii, Pythium monospermum, Pythium myriotylum, Pythium nodosum, Pythium nunn, Pythium oligandrum, Pythium paddicum, Pythium periplocum, Pythium porphyrae, Pythium pulchrum, Pythium pyrilobum, Pythium rhizo-oryzae, Pythium rostratum, Pythium schmitthenneri, Pythium selbyi, Pythium senticosum, Pythium sp, Pythium spinosum, Pythium splendens, Pythium sulcatum, Pythium sylvaticum, Pythium torulosum, Pythium ultimum, Pythium ultimum var sporangiiferum, Pythium undulatum, Pythium volutum Saprolegniales Achlya ambisexualis, Aphanomyces euteiches, Aphanomyces laevis, Aphanomyces sp, Dictyuchus sterilis, Plectospira myriandra, Pythiopsis cymosa, Saprolegnia ferax, Saprolegnia parasitica, Saprolegnia sp, Thraustotheca clavata

Table 2-6. Blast hits with at least 90% identity using ITS marker Order Putative species Albuginales Albugo laibachii Lagenidiales Lagenidium sp Peronosporales Peronospora grisea, Peronospora sp, Peronospora verbenae, Phytophthora cactorum, Phytophthora capsici, Phytophthora cinnamomi, Phytophthora colocasiae, Phytophthora frigida, Phytophthora infestans, Phytophthora palmivora, Phytophthora parasitica, Phytophthora sp, Phytophthora tropicalis, Phytopythium cucurbitacearum, Phytopythium helicoides, Phytopythium sp, Phytopythium vexans Pythiales Pythiogeton ramosum, Pythiogeton sp, Pythium acanthicum, Pythium acanthophoron, Pythium aphanidermatum, Pythium aristosporum, Pythium arrhenomanes, Pythium camurandrum, Pythium catenulatum, Pythium contiguanum, Pythium cucurbitacearum, Pythium deliense, Pythium graminicola, Pythium grandisporangium, Pythium helicoides, Pythium insidiosum, Pythium irregulare, Pythium kashmirense, Pythium marsipium, Pythium myriotylum, Pythium nodosum, Pythium oopapillum, Pythium pyrilobum, Pythium rhizo oryzae, Pythium sp, Pythium splendens, Pythium sulcatum, Pythium torulosum Saprolegniales Achlya sp, Achlya sparrowii, Aphanomyces astaci, Aphanomyces cochlioides, Brevilegnia longicaulis, Brevilegnia minutandra, Dictyuchus monosporus, Dictyuchus sp, Geolegnia helicoides, Pythiopsis irregularis, Pythiopsis terrestris, Saprolegnia monoica, Saprolegnia sp

Table 2-7. Taxa detected by both markers cox2 and ITS with ≥90% ID Order Putative species Lagenidiales Lagenidium sp Peronosporales Phytophthora colocasiae Phytophthora infestans Phytophthora palmivora Phytophthora sp Phytophthora tropicalis Phytopythium cucurbitacearum Phytopythium helicoides Phytopythium sp

Pythiales Pythium acanthicum Pythium acanthophoron Pythium aphanidermatum Pythium aristosporum Pythium arrhenomanes Pythium catenulatum Pythium deliense Pythium graminicola Pythium helicoides Pythium insidiosum Pythium irregulare Pythium myriotylum Pythium nodosum Pythium pyrilobum Pythium sp Pythium splendens Pythium sulcatum Pythium torulosum

Saprolegniales Saprolegnia sp

Table 2-8. Simpson’s D diversity index calculated for each sample and land use type considered in this study. Crop AE Samples SimpsonsD Samples SimpsonsD 1C1 0.65 1N1 0.36 1C2 0.93 1N2 0.36 1C3 0.43 1N3 0.39 2C1 0.88 2N1 0.41 2C2 0.90 2N2 0.49 2C3 0.49 2N3 0.35 3C1 0.93 3N1 0.83 3C2 0.59 3N2 0.58 3C3 0.98 3N3 0.56 4C1 0.33 4N1 0.63 4C2 0.55 4N2 0.42 4C3 0.74 4N3 0.41 5C1 0.67 5N1 0.52 5C2 0.47 5N2 0.54 5C3 0.46 5N3 0.30 6C1 0.50 6N1 0.50 6C2 0.53 6N2 0.68 6C3 0.30 6N3 0.41 7C1 0.45 7N1 0.58 7C2 0.67 7N2 0.55 7C3 0.49 7N3 0.42 8C1 0.46 8N1 0.37 8C2 0.96 8N2 0.51 8C3 0.43 8N3 1.00 Average 0.61 0.51 Standard deviation 0.21 0.16

Table 2-9. ANOVA results for effects of land use and farm on Simpson’s D index of diversity Df SumsOfSqs MeanSqs F Value Pr(>F) LandUse 1 0.14 0.14 4.26 0.047 Farm 7 0.32 0.05 1.38 0.249 LandUse:Farm 7 0.24 0.03 1.02 0.439 Residuals 32 1.06 0.03

Table 2-10. PERMANOVA results per sample from cox2 unrarefied OTU table Dataset: OTU reads per sample Df SumsOfSqs MeanSqs F.Model R2 Pr(>F) LandUse 1 0.94 0.94 4.22 0.07 0.002 Farm 7 3.41 0.49 2.18 0.26 0.001 LandUse:Farm 7 1.61 0.23 1.03 0.12 0.413 Residuals 32 7.14 0.22 0.54 Total 47 13.10 1 Dataset: OTU presence/absence per sample Df SumsOfSqs MeanSqs F.Model R2 Pr(>F) LandUse 1 0.02 0.02 0.33 0.01 0.773 Farm 7 0.58 0.08 1.56 0.23 0.074 LandUse:Farm 7 0.26 0.04 0.71 0.10 0.788 Residuals 32 1.71 0.05 0.66 Total 47 2.58 1

Table 2-11. PERMANOVA results per farm from cox2 unrarefied OTU table. Dataset: OTU reads per farm Df SumsOfSqs MeanSqs F.Model R2 Pr(>F) LandUse 1 0.49 0.49 2.01 0.12 0.079 Region 2 0.86 0.43 1.77 0.21 0.088 LandUse:Region 2 0.36 0.18 0.75 0.09 0.672 Residuals 10 2.42 0.24 0.59 Total 15 4.12 1 Dataset: OTU presence/absence per farm Df SumsOfSqs MeanSqs F.Model R2 Pr(>F) LandUse 1 -0.002 -0.002 -0.136 -0.011 0.879 Region 2 0.03 0.01 1.08 0.17 0.418 LandUse:Region 2 0.00 0.00 0.16 0.03 0.9 Residuals 10 0.13 0.01 0.81 Total 15 0.16 1

Table 2-12. PERMANOVA results per farm from ITS unrarefied OTU table. Dataset: OTU reads per farm Df SumsOfSqs MeanSqs F.Model R2 Pr(>F) LandUse 1 0.25 0.25 1.19 0.11 0.367 Region 2 0.86 0.43 2.05 0.37 0.09 LandUse:Region 2 0.36 0.18 0.85 0.15 0.604 Residuals 4 0.84 0.21 0.36 Total 9 2.31 1 Dataset: OTU presence/absence per farm Df SumsOfSqs MeanSqs F.Model R2 Pr(>F) LandUse 1 0.66 0.66 3.14 0.30 0.017 Region 2 0.31 0.15 0.73 0.14 0.752 LandUse:Region 2 0.39 0.20 0.93 0.18 0.551 Residuals 4 0.84 0.21 0.38 Total 9 2.21 1

Figure 2-1. Relative abundance of Oomycete clades by farm and land use obtained with usearch61 OTU picking against cox2 Oomycete database. Land use type corresponded to crop (C) or AEs (N). Taxa is grouped by clade, and percentages are indicated on the right of the legend.

Figure 2-2. Relative abundance of Oomycete genera for each sample obtained by merging outputs from usearch61 and BLAST search with at least 90% identity using cox2 database

Total Number of Reads - cox2

100000

10000

1000

100

1 1C 1N 2C 2N 3C 3N 4C 4N 5C 5N 6C 6N 7C 7N 8C 8N

Figure 2-3. Unrarefied total number of reads of Oomycete taxa by farm and land use with cox2 marker; y axis is in logarithmic scale.

Figure 2-4. Relative abundance of Oomycete clades for each sample type and farm obtained with usearch61 OTU picking against ITS Oomycete database. Samples are coded by farm with the numbers 1-8 according to Table 2-1 and by land use indicating if they corresponded to crop (C) or AEs (N). Taxa is grouped by clades. Note that some of the samples are not shown (2N, 4N, 5N and 7N) due to absence of reads assigned by this method.

Figure 2-5. Relative abundance of Oomycete genera for each sample obtained by merging outputs from unrarefied usearch61 and BLAST with at least 90% identity using ITS.

Total Number of Reads - ITS 5000

4500

4000

3500

3000

2500

2000

1500

1000

500

0 1C 1N 2C 2N 3C 3N 4C 4N 5C 5N 6C 6N 7C 7N 8C 8N

Figure 2-6. Unrarefied total number of reads of Oomycete taxa by farm and land with unrarefied OTU data obtained by usearch91 with ITS marker.

Figure 2-7. NMDS of Bray-Curtis dissimilarity between communities of Crop and AE samples. Centroids of each group are indicated in the squares and each point (sample) is connected to the centroids with blue lines. Centroids of each group are connected to the sample scores with vectors. Oomycete clades (represented in red) are located according to their scores based on Bray- Curtis dissimilarities that are weighted and rotated during NMDS analysis.

CHAPTER 3 PINEAPPLE HEART ROT ISOLATES FROM ECUADOR REAVEAL A NEW GENOTYPE OF PHYTOPHTHORA NICOTIANAE

Introduction

Pineapple (Ananas comosus), a perennial monocot in the Bromeliaceae family, is the most commercialized bromeliad and has become the third most important tropical fruit crop after banana and citrus (108). In Ecuador, pineapple crops are located in 6 provinces, with expanded production areas for export in Santo Domingo de los

Tsáchilas. National annual production is estimated to be ~126000 tons [FAOSTAT,

2014], 45% of which are exported [MAGAP bulletin, 2014]. Ecuadorian pineapple production has improved since the adoption of the MD-2 hybrid, which has better quality than the formerly most commercialized cultivar ‘Smooth Cayenne’. MD-2 was developed by Del Monte Fresh Produce Hawaii Inc., and it is characterized by yellow- green leaves with reddish tips and fruits with an intense yellow pulp. MD-2 is preferred for consumption due to its sweetness and size consistency.

Pineapple heart rot (PHR) is a disease that drastically diminishes the yield of harvested fruit per year. PHR is caused by Phytophthora nicotianae, P. palmivora and

P. cinnamomi. These species cause the same symptoms, but differ in occurrence based on climate and altitude; consequently, they are rarely found together (31). Phytophthora nicotianae is the most frequent causal agent of PHR, especially in tropical latitudes or during warm seasons (109-112). MD-2 is more susceptible to P. nicotianae than

‘Smooth Cayenne’ (34), especially if soil is alkaline and poorly drained (31). P. nicotianae causes polycyclic outbreaks that ultimately result in plant mortality and reduced crop density (29-31). Pineapple can be affected by P. nicotianae as soon as it is planted; young leaves stop expanding normally and become yellow. Eventually,

water-soaked lesions appear at the base of the apical meristem, a soft rot develops, and leaves are easily detached. Ultimately, severely affected plants decline and die.

Although younger plants are more vulnerable, P. nicotianae also affects mature plants.

When disease occurs in plants bearing fruit, the peduncle and fruit can also be affected

(31, 109, 113).

Depending on the edaphology and humidity, PHR can cause up to 100% plant mortality. Due to its abundant production of dormant propagules, P. nicotianae is known to persist in soil and infected tissues for long periods of time (27). Chlamydospores, oospores and mycelia can be transported by running water or farming tools and equipment (113). This pathogen presents a challenge for producers (112).

P. nicotianae affects hosts in 225 genera, from herbaceous plants to vegetables and even forest trees in natural ecosystems (29, 32). The P. nicotianae life cycle involves the production of multiple spore types. Swimming zoospores allow dissemination through water, while chlamydospores, mycelia and oospores can spread in soil attached to field attire and tools (113). Zoospores are believed to be the propagules that start the disease cycle; once they attach to host tissues, they encyst and germinate. Appressoria aid the infection of roots and leaves, which allow hyphae to invade plant cells using haustoria-like structures. Finally, sporangia are produced on the plant surface to restart the cycle, and chlamydospores are formed directly from hyphae and serve as survival structures (114, 115).

Like other heterothallic Phytophthora, this pathogen is able to reproduce sexually with opposite mating types (A1 and A2). Sexual reproduction purges deleterious mutations and enables reassortment of chromosomes leading to new multilocus

genotypes (4, 39). However, P. nicotianae may not complete regular sexual cycles on its major hosts and causes significant damage even when not reproducing sexually (29,

114).

Several molecular strategies have been developed to understand the genetic variability, population genetic structure and dispersal of P. nicotianae, including multilocus sequence analysis and simple sequence repeats (SSR) (2, 30, 116). Intraspecific variation occurs among populations of P. nicotianae without evident correlation to the geographic location of recovery. Moreover, nucleotide sequence variation in nuclear and mitochondrial markers revealed an association between crops and genetic groups, suggesting the possibility of host specialization (30, 39, 117). However, isolates from

Citrus were genetically similar, regardless of the country of origin. P. nicotianae on tobacco also exhibited clonality within populations (118, 119), but presents genetic differentiation among regions due to the nature of tobacco seed propagation, which limits pathogen spread across production regions. Conversely, isolates from ornamental potted plants presented more genetic variation within populations and less differentiation among populations of host genera and regions (39). Biasi et al. hypothesized that host specialization may develop in intensive farming systems. In high- density ornamental nurseries, the presence of opposite mating types may increase the possibility of sexual reproduction and the rise of new genotypes that have higher fitness on particular hosts (32, 39). In contrast, monoculture systems of genetically uniform crops, such as pineapple, can lead to specialization and divergence from other host populations due to intensified host selection pressure (4).

The first objective of this study was to determine the level of intra-specific genetic variation in P. nicotianae causing PHR in MD-2 produced in Santo Domingo de los

Tsáchilas and Los Ríos provinces, Ecuador. Because MD-2 pineapples in Ecuador are grown in continuous monoculture, ranging from 100-600 ha, we hypothesized that this population of P. nicotianae could be clonal, as in Citrus and Nicotiana. Second, host genetic diversity was evaluated by comparing the genotypes and haplotypes reported from other hosts with those found on pineapple in Ecuador. Specifically, genetic relationships between P. nicotianae from pineapple in Ecuador and other populations found in different hosts worldwide were determined. No previous population genetic studies had incorporated P. nicotianae isolated from pineapple, thus, this study advances understanding of its evolution and population structure.

Materials and Methods

Sampling and Morphological Identification of Isolates

During the rainy season (February-April) of 2016, samples were taken from four pineapple farms. In Santo Domingo de los Tsáchilas, farms M and G were ~2 km apart, and farm F was ~16 km from M and G. Farm A in Los Ríos province was located ~48 km from farm F. We hypothesized that a single population of the pathogen was found on all farms, because the overall sampled area was small, was primarily dedicated to the production of pineapple, and shared similar soil and weather characteristics.

In each location, we surveyed MD-2 for PHR. Disease distribution was patchy, and was prevalent on sloped hills and in close proximity to water sources. Yellowed and water-soaked leaves were pulled from affected plants, and 3 to 5 2x2 mm leaf pieces were cut from the border between healthy and necrotic tissues, surface disinfected for 2 minutes in 20% bleach, ethanol and sterile water and dried with autoclaved paper

towels. The surface-disinfected leaf pieces were plated on 20% V8-PARPH agar medium and incubated at 25 °C for 72 h. Colonies were checked for bacterial/fungal contaminants and re-isolated onto new plates. Isolates were tentatively identified as P. nicotianae based on caenocytic hyphae, abundant swellings from which hyphae grow radially and observation of terminal or intercalary chlamydospores and lemon-shaped papillated sporangia. Single-spore isolates of resulting cultures were obtained by placing plugs of 48 h old cultures in sterile water and incubating for 24 h at 25 °C in presence of light. After sporulation occurred, spores were captured by pipette suction,

10-fold diluted in sterile water and plated on 20% V8-PARPH. Colonies formed from a single spore were transferred to new V8-plates and were used for DNA extraction. In parallel, mating type was determined according to Erwin and Ribeiro 1996. Known A1 and A2 mating types from Florida nurseries were used in pairings with isolates of unknown mating type. Pineapple isolates were also paired with themselves to check for self-fertility and as controls.

DNA Extraction and PCR Conditions for Microsatellites and Mitochondrial Markers

DNA was isolated from fresh hyphae with the Qiagen DNeasy kit (QIAGEN Inc.,

Valencia, CA, USA) according to manufacturer recommendations for fungal DNA extraction. The cytochrome oxidase II gene was amplified using 1X GoTaq colorless master mix, 0.2 uM each primer and ~10 ng of DNA template. Amplicons were purified using 0.5 µL of ExoSAP-IT PCR Product Cleanup Reagent (Affymetrix, Cleveland, OH,

USA) per 20 µL of reaction. The mix was incubated overnight at 37°C and inactivation of the enzyme was performed at 80 °C for 15 min. The products were sequenced to confirm species identification.

SSR markers for P. nicotianae developed by Biasi et al.2015 were used to genotype the pineapple isolates. Forward primers were labeled with fluorescent tags at the 5’ end to allow duplexing during the genotyping process. The primers for loci P1509,

P643, P1129, P5 and P2040 were tagged with 6-FAM, while HEX dye was used with primers for P15, P788, P2039, and P17. Amplification conditions were as follows: 15 ul

PCR reactions consisted of 1X GoTaq colorless master mix, 4 mM MgCl2, 0.3 uM of primers and ~10 ng of DNA template. PCR runs were as follows: 3 minutes of denaturation at 94°C, then 35 cycles of 30 seconds at 94°C, 30 seconds of 59 and 45 seconds at 72°C, followed by a final extension of 40 minutes at 72°C. In addition to the

30 PHR isolates, 10 isolates with published genotypes (39) were used to calibrate the scoring of alleles. Amplicons were diluted and genotyped by capillary electrophoresis in the bioanalyzer Applied Biosystems AB3730 using LIZ600 as the fragment size standard.

Based on the microsatellite genotyping results, 5 individuals were selected for sequencing of additional mitochondrial markers. We used primers from Mammella et al.

2013 to amplify mitochondrial loci cox2+spacer, atp1-nad5, trnG-rns and rns-cox.

Primer pairs for loci atp1-nad5 and rns-cox did not amplify well for the isolates selected, resulting in faint or no bands. We proceeded with cox2+spacer and trnG-rns. For sequencing, 50 µL reactions consisting of 1X GoTaq, 0.5 µM each primer and ~15ng of

DNA were used to amplify the mitochondrial markers. The products were purified using

ExoSap-IT as described above. Sanger sequencing was done with forward and reverse primers to obtain both strands for each marker.

SSR Genotyping and Analysis

Allele calls were performed with GeneMarker V2.7.0 (Softgenetics, State

College, PA) (120), with Liz600 standard for size calibration. Stutter peaks were reduced by using 5’ PIG tails (GTTT) and, when present, they were manually removed based on comparisons with other peaks (stutter peaks were at least 50% smaller).

Calibration of scores was performed using the 10 isolates previously genotyped by comparisons of the size called in this study with the score reported by Biasi et al. Then, allele calls from pineapple isolates were modified accordingly (Table C-1).

Combining our data with the Biasi et al. (2016) data, global populations were a priori determined based both on the host genera and country. We assigned genotyped isolates to 12 populations: Ananas/Ecuador, Citrus/Italy, Citrus/Vietnam,

Convolvulus/Italy, Correa/Italy, Dodonaea/Italy, Lavandula/Italy, Lycopersicon/Italy,

Myrtus/Italy, Nicotiana/USA, Nicotiana/Australia, Ruta/Italy, totaling 236 isolates (Table

C-2 and C-3). Other isolates were not included in analyses due to low number (n<5) of individuals per host/country. Data were formatted for the GENALEX 6.501 add-in to calculate allele frequencies and Fst values. The GENALEX file was imported to the

Poppr R package (121, 122) for analysis of genetic diversity and construction of minimum spanning networks (MSN) with Bruvo’s distance (123, 124). Bruvo’s distance was designed for polyploids, and while we did not have any apparent polyploids in our isolates, the previously reported data included polyploid isolates (39). Principal coordinate analysis (PCoA) was performed with Lynch’s distance using Polysat package in R (125).

Finally, population structure was analyzed using Bayesian clustering implemented in the software STRUCTURE 2.3.4 (126). For this process, data was

clone-corrected to censor multiple identical genotypes that can obscure the analysis by causing non-random associations (127). Only unique host and MLG combinations were kept, i.e., a MLG could be present multiple times if it was isolated from different hosts.

After clone correction and removal of 16 triploid and tetraploid isolates, the dataset contained 74 diploid individuals. The simulations were done with 50,000 Markov chain

Monte Carlo (MCMC) steps of burnin followed by 100,000 steps for sampling. We used

10 independent iterations for 1-10 clusters (K). The model considered admixture and did not take into account the source population of the individuals. The online tool Structure

Harvester (128), which compares and analyzes STRUCTURE results, was used to find the most likely number of clusters based on the Evanno et al. method (129). A subset of hosts was analyzed separately using similar parameters.

Multi-Locus Sequencing Analysis

All sequence processing, including quality trimming, assembly, alignment and phylogeny were performed using Geneious 10.1.3 software (Biomatters, Auckland, New

Zealand), which integrates tools to edit and analyze sequence data (130). We started with multiple-alignment of the cox2 region of all pineapple isolates. Based on this preliminary analysis, we confirmed that our isolates were P. nicotianae and contained no polymorphisms in this mitochondrial gene. We then examined the phylogenetic relationship of pineapple isolates to 96 isolates from other hosts (117) using the concatenated alignment of the two markers cox2+spacer and trnG_rns. All sequences were aligned using MUSCLE (131, 132) with a maximum of 8 iterations. To obtain a suitable nucleotide substitution model for a maximum likelihood (ML) phylogenetic tree, we computed likelihood scores with the software jModeltest 2.1.10 (133, 134). Based on

Akaike information criterion (AIC) scores, the most appropriate model was Tamura and

Nei with invariant sites. A phylogenetic tree was constructed by maximum likelihood

(ML) using the PhyML (135) plugin incorporated in the software Geneious 10.1.3.

Gamma shape parameter was set to 1.00 and the proportion of invariant sites was estimated. The Subtree Pruning and Regrafting (SPR) method (136) was used for topology search and a total of 1000 bootstrap samples were assessed for statistical support. The tree was rooted based on a ML tree (Figure D-1) built with cox2+spacer sequences that used an isolate of P. infestans as outgroup.

Results

Identity of Pineapple Isolates and Morphological Description

Isolates recovered from PHR lesions were initially assigned to genus based on morphology. Thirty single-spore isolates were recovered and successfully maintained for further experiments. When mating type assays were performed, rounded amphigynous oospores of ~25 µM were observed. Species identification was confirmed by sequencing the cytochrome oxidase II (cox2) gene for each single-spore isolate.

Sequences were compared to existing records in GenBank, using the BLAST tool.

Isolates had 100% identity and 0.0 e-value with published P. nicotianae sequences.

All pineapple isolates produced oospores in pairings with the A1 mating type reference isolate. However, some also produced very few oospores in the presence of the A2 mating type reference isolate. We considered our isolates to be the A2 mating type, because oospore production in pairings with A2 isolates was repeated and only a small number of oospores were found. No oospores were reported when only one isolate was plated or when different isolates from pineapple were plated together. Table

3-1 summarizes the origin, host, farm code and mating type of isolates used in this study.

SSR Diversity and Host-Population Genetic Distance

SSR genotyping of P. nicotianae from PHR revealed a single multi-locus genotype (MLG) in 29 isolates, defined as MLG100, plus a singleton MLG101. The

MLGs are differentiated by a 2 bp size difference in locus P1509. Table 3-2 lists the allele calls for the nine SSR loci genotyped in Ananas isolates. When compared to genotyped isolates from other hosts, Ananas isolates had no private alleles but unique genotypes for loci P2039 (114/120) and P17 (132/144). For the other seven loci, the observed genotypes had been observed in isolates from other hosts. Locus P1509 genotype 122/122 was present in Citrus from Vietnam primarily, but also in one isolate from USA and another from Italy. P15 96/96 was only present in one other isolate from the host genus Rosmarinus. P788 135/137 was shared with isolates from genera

Convolvulus, Lavandula and Polygala. P643 158/164 was found in two Lavandula isolates with two different MLG. P1129 154/160 was a very common genotype in isolates from Chamaleucium, Lavandula, Polygala and Hebe. P1129 190/190 was the most shared genotype; it was found in several individuals from Citrus, Lycopersicum,

Capsicum, Myrtus, Nicotiana and Correa and belonged to several MLG. Finally, locus

2040 155/155 was reported in Chamaleucium, Convolvulus, Hebe, Lavandula and

Rosmarinus.

Genotypic diversity was lower for P. nicotianae from Ananas compared to samples of similar sizes from other host-country populations (Table 3-3). P. nicotianae from Ananas showed high pairwise Fst values to other population samples (Table 3-4), possibly due in part to the high level of clonality within the sample.

A minimum spanning network (Fig. 3-1) was constructed based on a matrix of pairwise Bruvo’s distances to visualize genetic distance among genotypes. Several

clusters of MLGs could be identified within the network. MLGs from Citrus in both Italy and Vietnam clustered as well as Nicotiana isolates from the United States. A Nicotiana-

Australia cluster was clearly separated from the closest MLG from Lycopersicon. Some host populations were distributed across the network, including two MLG clusters from

Myrtus and multiple MLGs from Lavandula. The Ananas population was distant from the closest MLG, Lavandula (MLG3).

We complemented the MSN with a principal coordinate analysis (PCoA) using

Lynch’s distance (Fig. 3-2). PCoA is a multidimensional scaling method that uses a two- dimensional ordination space to map objects based on their ranked distances; where proximity corresponds to similarity, but is not the original distance among objects (137,

138). Consistent with the network, Ananas MLGs showed separation from MLGs representing populations from other hosts, forming a small cluster at the bottom left of the plot. The other distinct cluster contained MLGs from Citrus in Vietnam and Italy.

Population Structure Across Host-Populations

Motivated by the MSN and PCoA, we examined whether the observed clusters could be considered subpopulations using model-based Bayesian clustering of P. nicotianae genotypes. Each isolate was assigned with high probability to one of 3 genetic clusters (Fig. 3-3A) (ΔK =12.481877; Figure D-2). The first group (in light blue) comprised aCll isolates from Citrus from Vietnam and all but one of the Citrus isolates from Italy. The second group (in violet) consisted of all Nicotiana isolates from the

United States and Australia in addition to one Lavandula isolate and the only Dodonaea isolate. The third group (in yellow) included isolates from ornamentals, and Ananas.

Because we were concerned that the largely clonal reproduction in populations in the first and second (light blue and violet) groups could affect the inference of population

structure, we re-analyzed only isolates assigned to the third (yellow) group (40 isolates).

This analysis provided support for K=2 genetic clusters (Fig. 3-3B, ΔK =43.195994,

Figure D-3). The orange group contained all but one of the Myrtus isolates, as well as two Lavandula isolates and one Convolvulus isolate. The second group (blue) contained Ananas MLGs clustered with isolates from ornamentals, including most

Convolvulus and Lavandula, one Myrtus isolate and all isolates from Ruta, Correa and

Lycopersicon.

Multi-Locus Sequencing

Sequences of P. nicotianae isolated from pinneapple were identical across 889 bp for the cox2+spacer. The trimmed trnG-rns sequence was 579 bp and contained small indels among isolates: isolate A512 contained a T insertion in position 401 of the alignment, whereas isolates F20 and F2 had a triplet of Ts inserted in positions 402-

404. This region is A-T rich and these polymorphisms may be due to replication slippage. The maximum likelihood tree using both markers conserved most of the topology of the phylogenetic tree based on four mitochondrial and two nuclear markers in Mammella et al. (2013), although some of the branches in our 2-locus tree had low bootstrap support (Fig.3-4). Mammella et al. designated phylogenetic clades within P. nicotianae and we use their nomenclature. Nicotiana and Citrus isolates formed clades that are nearly exclusive to these host genera (N2 and N3). Ananas isolates formed a distinct branch that was most closely related to groups N3 and N4. The Ananas haplotype may represent a new mtDNA group N7.

Discussion

Phytophthora nicotianae is a multifaceted pathogen whose population structure remains poorly understood across the entirety of its host and geographic range (32).

Therefore, studies of hierarchical samples of isolates from additional hosts are needed to better define the genetic variation in this pathogen (39). We genotyped P. nicotianae isolates collected within and among farms across the pineapple production regions of

Ecuador and found a very homogeneous population of P. nicotianae. Out of 30 isolates from two provinces, 29 isolates shared the same 9-locus SSR genotype, whereas a single variant in one locus produced a second genotype found only in Santo Domingo de los Tsáchilas province. Likewise, we found only one mating type across isolates, which is consistent with the SSR data and suggests that the sampled population of P. nicotianae is clonal. The isolates from Ananas are genetically distinct from isolates from other hosts. Our data suggest a new genetic group of P. nicotianae associated with pineapple heart rot in Ecuador.

Based on isolates that have been genotyped to date, the global population structure of P. nicotianae consists of three main genetic groups: one represented by isolates from Citrus, another on Nicotiana and a third group composed of isolates from all other hosts, including Ananas, While the PCoA indicated genetic distance between

Ananas and other isolate genotypes, Structure analysis showed Ananas isolates in the same group with ornamentals. Indeed, Ananas isolates shared some alleles with other populations, especially with isolates from lavender. But pineapple and lavender have very different movement pathways and are not cultivated in the same regions; while the first has its origins in South America, the latter is grown on the Mediterranean Sea.

Their histories of importations and commercial trade indicate that both hosts were present in Europe during colonization of the Americas (108) when there could have been shifts among multiple hosts by P. nicotianae. Alternatively, Ananas isolates could

share alleles with isolates from other hosts due to homoplasy (convergent evolution).

The lack of data from other South American crops hinders our ability to separate the effect of geography from that of host. This data gap may explain the genetic distance between isolates from Ananas and other hosts in many of our analyses.

The global population of P. nicotianae may be panmictic for the majority of hosts due to their means of cultivation, propagation and distribution in the plant trade (32), and isolates from ornamentals are a genetically diverse group. Conversely, isolates from Citrus and Nicotiana form distinct genetic groups, and we have found a novel cluster associated with Ananas in Ecuador. Some of these subpopulations exhibit high clonality and reduction of heterozygosity. Clonal reproduction provides advantages, including maintenance of successful combinations of co-adapted alleles for a given environment and a lower cost of reproduction compared to sexual reproduction (1, 4).

Pineapple monoculture is a stable environment that can favor the persistence of highly fit genotypes and reduce the effect of random genetic drift. The clonal population on pineapple also may be the result of a founder effect. There were a limited number of importations of pineapple seed from Costa Rica to Ecuador and, since the late 1990s, producers in Ecuador have used vegetative propagation from their own nurseries. It is possible that only one mating type of the pathogen was imported with seeds. Testing this hypothesis requires sampling from other MD-2 pineapple production regions as well as other potential hosts in Ecuador to determine if this genetic group is specific to pineapple.

Through global trade, P. nicotianae has established new niches and colonized a diversity of host genera. Based on this and previous studies, P. nicotianae population

structure may be different between nurseries and more intensively monocultured crops.

It is challenging to assess population biology in a pathogen with such a wide host range, on crops with varying management strategies where host pathogen populations face different evolutionary forces. Assessments of phenotypic variation in P. nicotianae populations, including fungicide resistance and virulence, have been mostly limited to studies on tobacco (118, 139). Management of this economically important pathogen will require better understanding of its evolutionary trajectories on specific hosts and in different cropping systems, as well as its history and potential for migration among global regions and hosts.

Table 3-1. Isolates collected in this study, their origin and mating type. Geographical Mating Isolate Designations Host Farm Origin Type A53, A58, A512, A61, A. comosus - Los Ríos, A A2 A611, A67, A69 cultivar MD-2 Ecuador

F11, F12, F13, F20, A. comosus - Santo F A2 F2 cultivar MD-2 Domingo de los Tsáchilas, Ecuador

G181, G191, G192, A. comosus - Santo G A2 G193, G194 cultivar MD-2 Domingo de los Tsáchilas, Ecuador

M132, M161, M162, A. comosus - Santo M A2 M211, M21, M231, cultivar MD-2 Domingo de M241, M281, M283, los Tsáchilas, M301, M302, M33, Ecuador M71

Table 3-2. Allele sizes (bp) for SSR loci and the resulting multi-loci genotype (MLG) designation for each isolate. Repeat motifs, according to Biasi et al. 2015, are indicated under each locus name Locus Isolates P15091 P15 P788 P643 P20392 P1129 P17 P5 P2040 MLG GT TGTC GA GT CGA GTA AAC TGTC AGT A53, A58, A512, A61, A611, A67, A69, F12, F13, F20, F2, G181, G191, G192, G193, G194, M132, M161, M162, M211, M21, M231, M241, M281, M283, M301, M302, M33, M71 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 100 F11 120/120 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 101 Correa1*, Correa2* 160/160 66/78 129/135 158/162 99/120 139/154 105/144 190/190 158/158 94 RCGS22* 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158 15 M5rlc* 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 71 CM1f* 116/116 66/93 129/137 164/168 120/120 151/154 129/132 194/194 158/164 6 C301* 120/120 75/93 129/129 152/164 99/99 139/154 129/159 190/190 152/158 45 Serravalle1*, FerraraR3* 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 52 Dodoneacol5* 132/132 93/111 127/137 166/174 111/111 151/154 126/126 190/238 158/158 75 Lav5* 124/126 78/78 135/137 158/166 99/99 154/160 105/105 186/202 155/167 62 1. Only locus P1509 exhibited SSR allelic variation (underlined) among Ananas from Ecuador. 2. Allele calls in italics represent new genotypes in comparison with other previously studied isolates. * Isolates previously genotyped were used as standards to adjust allele calls and allow comparisons across studies.

Table 3-3. Genotypic diversity of populations with sample sizes of greater than 10 isolates. Populations were defined by host genera and country of origin. Shanon-Wiener Simpson’s Genotypic Host genera Location N MLG Index Index Evenness Ananas Ecuador 30 2 0.146 0.064 0.437 Correa Italy 11 2 0.305 0.165 0.556 Ruta Italy 10 3 0.639 0.340 0.576 Citrus Vietnam 35 8 1.524 0.684 0.603 Citrus Italy 49 16 1.624 0.589 0.352 Myrtus Italy 37 13 1.966 0.790 0.614 Nicotiana USA 13 10 2.205 0.876 0.873 Lavandula Italy 23 12 2.292 0.881 0.832 Total 236 85 3.595 0.942 0.459

Table 3-4. Pairwise genetic differentiation based on FST values across the eight populations defined in Table 3-3 Ananas/ Citrus/ Citrus/ Correa/ Lavandula/ Myrtus/ Nicotiana/ Ruta/ Populations Ecuador Italy Vietnam Italy Italy Italy USA Italy Ananas/Ecuador - Citrus/Italy 0.258 Citrus/ Vietnam 0.404 0.112 Correa/Italy 0.476 0.160 0.045 Lavandula/Italy 0.280 0.067 0.099 0.104 Myrtus/Italy 0.366 0.113 0.240 0.278 0.128 Nicotiana/USA 0.294 0.086 0.164 0.206 0.106 0.141 Ruta/Italy 0.345 0.100 0.145 0.172 0.102 0.210 0.073 -

Figure 3-1. Minimum spanning network using Bruvo’s distance for P. nicotianae isolates by host and country. Genetic distance is represented in the network edges as indicated in the scale bar: the smaller the genetic distance, the thicker and darker the edge. Ananas MLGs are located within a branch of MLGs from ornamental hosts, closest to isolates from Lavandula and Ruta. The network clearly separates Nicotiana and Citrus isolates from other hosts. The original graphic was modified to show edges that were hidden under nodes.

Figure 3-2. Principal coordinates analysis of multi-locus genotypes using Lynch distance. Host genera and country are indicated by color. Each symbol represents a MLG. Ananas MLGs are shown as squares, while all other hosts are represented by triangles.

B Figure 3-3. Phytophthora nicotianae population structure by host and country. A) The 74 P. nicotianae MLGs were assigned with high probability to one of three distinct clusters (K = 3). One cluster (light blue) contained most isolates from Citrus (regardless of country of origin), a second group (violet) was composed of isolates from Nicotiana (also without considering origin); while the third cluster (yellow) contained isolates from the remaining host genera, including Ananas and ornamentals. B) Analysis of the subset of P. nicotianae isolates corresponding to the yellow cluster in A) showed support for K=2 genetic clusters. Ananas MLGs were assigned to the blue cluster along with isolates from several ornamental genera; while the orange group was predominantly Myrtus in addition to one Convolvulus and two Lavandula isolates.

Figure 3-4. Maximum likelihood phylogeny inferred from concatenated DNA sequences from cox2+spacer and trnG_rns mitochondrial markers. Numbers on nodes represent statistical support based on 1000 bootstrap samples, with values >50% shown. Sequences from Ananas isolates (N7) were genetically distinct from isolates from other hosts. Clade nomenclature follows Mammella et al. 2013. Branch lengths are in substitutions per site as indicated by the scale bar.

CHAPTER 4 EVALUATION OF HIGH RESOLUTION MELTING ANALYSIS AS A METHOD TO DIFFERENTIATE PHYTOPHTHORA HYBRIDS FROM THEIR PARENTAL SPECIES

Introduction

Establishment of pathogens in new environments depends on their ability to adapt to local conditions, which may also require infection of new hosts. One of the mechanisms proposed for rapid evolution of introduced organisms is interspecific hybridization. Hybridization rapidly introduces genetic variation to a population that has likely experienced a genetic bottleneck (140) Following hybridization, hybrids may continue to evolve by introgression (backcrossing to one of the parents), mitotic chromosomal rearrangements, chromosome loss and epigenetic alterations (47, 141).

Interspecific hybrid plant pathogens are not common but have been reported repeatedly in fungi and oomycetes and are of potentially high impact. Hybridization may be under- reported in these microbes because molecular markers are generally required to confirm hybrid status (142).

Species in the genus Phytophthora, the “plant destroyer”, have been remarkably successful in agricultural environments. They cause severe losses of hundreds of hosts genera (41, 143). Evolutionary adaptability and effective dispersal have enabled

Phytophthora species to colonize new niches and become invasive pathogens. Human- aided movement of Phytophthora has also facilitated hybridizations of previously allopatric species (40). Interspecific hybridization that has occurred between several

Phytophthora species (46, 47, 49, 52, 53, 55), has resulted in host shifts and aggressive new taxa that cause new disease outbreaks. The most well-explored case is that of P. xalni, which is hypothesized to have emerged through several hybridization events,

possibly in nurseries, before establishing as an aggressive alder pathogen in natural ecosystems in Europe (50).

The formation of interspecific hybrids depends on many factors including frequent niche overlap, weak reproductive barriers between species, and compatibility between genomes (40). Phytophthora infestans is known for its genome plasticity (144) and variation in ploidy and genome size (61, 145). In the Andean region of Ecuador, P. infestans was thought to have a simple population genetic structure, until blight symptoms were found on wild Solanum species. Atypical P. infestans isolates were isolated from wild hosts, as well as S. quitoense (lulo) and S. betaceum (tree tomato), which are commercialized in the Andes and cultivated elsewhere (52, 71). The atypical isolates were described as a new species, P. andina, based on AFLP, RFLP and allozyme analyses (52). P. andina was subsequently identified as an interspecific hybrid of P. infestans and an unknown closely related species (53, 146). The second parental species may reside cryptically in wild Solanum species that occur in the Andean region and nearby Amazon basin of Ecuador, Colombia and Peru (44, 53, 147). Both P. infestans and P. andina are able to infect S. muricatum, which could have served as a bridge between the parental species. S. betaceum is attacked by P. andina genotype

EC-3, recently redescribed as P. betacei (148), but may also be affected by P. infestans and intermediate genotypes in Colombia (149). A recent report from Peru indicated the presence of mating type A2 of P. andina (or possibly P. betacei), whereas only mating type A1 isolates were previously found on this host. The presence of both mating types could lead to sexual reproduction among apparently asexual lineages. Presently,

isolates causing late blight on Solanum hosts in the Andes must be genotyped to definitively assign them to species (150).

The interspecific hybrid Phytophthora xpelgrandis was reported for the first time in Dutch hydroponic greenhouses, where it was recovered from Spathiphyllum and

Primula ornamentals with root and crown rot (151). Although it exhibited distinct morphology, its hybrid status was not detected by morphology. RAPD and isozyme analysis were required to determine its relationship with P. nicotianae, followed by ITS sequencing and AFLP to recognize that P. cactorum was also a parent of this interspecific hybrid (44, 47). P. cactorum and P. nicotianae are closely related (clade I) species (62, 96). Mitochondrial DNA from Dutch P. xpelgrandis isolates was of P. nicotianae origin and the mechanism for hybridization was not clear at the time they were reported (151).The mitochondrial coxI in German isolates were of P. nicotianae origin, and RAPD analysis of nuclear DNA indicated a greater resemblance to P. cactorum than P. nicotianae, although some bands were identical to the first , indicating the difficulty of identification of P. xpelgrandis with traditional methods (152). P. xpelgrandis produces abortive oospores, suggesting that its sexual reproductive system is nonfunctional (47). P. xpelgrandis represents a more widespread risk to agriculture than P. andina, as it is widely distributed (the Netherlands, Taiwan, Peru, Germany,

USA, Hungary and Italy) (151-156) and has a wide host range including ornamentals of commercial importance as well as conifers.

Hybrids have been detected based on unusual or intermediate morphology, genetic fingerprinting, as well as cytology studies that show polyploidy or an unstable chromosome structure as result of hybridization (56). Cloning and sequencing of RAPD

fragments for the generation of SCAR-based PCR primers was used to discriminate P. alni subspecies alni (Paa), uniformis (Pau) and multiformis (Pam) (157). This distinguished the variants, but was not able to determine that Paa was a hybrid between

Pau and Pam (158). The origin of Paa was determined by sequencing nuclear single- copy genes (158) and quantification of allele copy number of single-copy genes (46).

Currently, sequencing of cloned amplicons of nuclear loci, for which hybrid individuals often show distant, phylogenetically distinct alleles (4, 44, 53), is the most common detection strategy to detect hybrids (53). Rapid detection of Phytophthora hybrid species would facilitate study of their emergence and distribution, and accelerate management decisions.

Quantitative polymerase chain reaction (qPCR) has improved the diagnosis of plant pathogens, because it is a reliable and sensitive method to detect and quantify pathogen DNA (159). To rapidly identify hybrids, genotyping sites with single nucleotide polymorphisms (SNPs) is proposed that would differentiate parental species. High resolution melting (HRM) is a qPCR-dependent technique that measures the rate dissociation of double-stranded to single-stranded DNA as temperature increases (160,

161). This technique allows the identification of different SNP genotypes by taking advantage of their distinct melting temperatures (Tm) (162). During the HRM, incremental increases in temperature cause double-stranded DNA to disassociate, releasing a dye bound to the double-stranded DNA and decreasing the fluorescence level. The rate of dissociation depends on the thermodynamics of the DNA sequence being analyzed. Polymorphisms within and among samples cause small but detectable differences in Tm (162-164), thus forming unique fluorescence profiles. HRM is a

closed-tube assay that produces rapid results faster compared to other mutation scanning methods that rely on strand separation, but expose the PCR products to the environment and are laborious and time-consuming (165). HRM has been used in a variety of plant pathology-related studies where accurate differentiation of one or a few polymorphisms is required. These methods have been used to distinguish seven

Fusarium oxysporum formae speciales (164), discriminate closely related Monilinia spp. associated with brown rot of pome fruits (166), and identify P. infestans lineages in the

United States (162), among others. HRM also helped discriminate the beneficial strain of Pseudomonas protegens Pf4 from normal strains present in the rhizosphere (160).

We hypothesized that HRM analysis has the potential to discriminate hybrid

Phytophthora from their parent species using SNPs that differentiate parental species and are heterozygous in the hybrid. Thus, our objectives were to: 1) identify SNPs that distinguish P. andina and P. xpelgrandis from parental species; and 2) design and test

HRM assays based on these SNPs. Here we show proof of concept for using HRM for rapid discrimination of specific Phytophtora hybrids from parental species, with the ultimate goal of establishing these assays as diagnostic tools.

Materials and Methods

Isolates Used in this Study

Isolates or DNA of P. infestans, P. andina, P. nicotianae, P. cactorum and P. xpelgrandis were kindly provided by collaborators. Host and country of origin are described in Table 4-1. Genomic DNA extraction used the FastDNA isolation kit (MP

Biomedicals, USA) for P. infestans and P. andina, and DNeasy Plant mini kit (Qiagen,

Valencia, USA) for P. nicotianae, P. cactorum and P. xpelgrandis. We adjusted extracted genomic DNA to a concentration of 20 ng/µL.

Polymorphic Sites Mining and Primer Design

To identify polymorphic sites distinguishing P. infestans from P. andina we used heterozygous sites identified from Illumina short reads of P. andina mapped against P. infestans isolate T-30 contigs. A pileup file summarizing polymorphisms in 50 core genes of P. andina isolate EC3425 were kindly provided by Sophien Kamoun. These data were explored using Artemis 16.0.0 (167). Sections of 80-100 bp containing heterozygous sites in P. andina were selected to design primers suitable for amplifying these short regions. Primer3 v. 0.4.0 (168, 169) generated several sets of primers whose amplicons were tested in silico using uMelt (170). uMelt produces a virtual melting profile curve for each sample, with values of fluorescence vs. temperature at each 0.5C intervals, and the fluorescence derivative (-dF/dT) vs. temperature.

Derivative values were transferred to a spreadsheet where they were plotted. The temperature where the peak reached a maximum was recorded and compared between

P. infestans and P. andina. If this difference was ≥0.5C, we considered the amplicon a candidate for HRM.

To identify heterozygous SNPs in P. xpelgrandis, available sequences of nuclear genes from P. nicotianae and P. cactorum were aligned and compared visually in

MEGA version 6.0 (171). Regions of 80-100 bp containing one or more SNPs between

P. nicotianae and P. cactorum were identified. These regions were then tested in uMelt, used to generate primers, and virtual amplicons were evaluated as above.

Resulting primers were tested across hybrid and parental isolates by visualizing

PCR products in 2% agarose gels after end-point PCR before qPCR and HRM.

PCR and High Resolution Melting

PCR and High Resolution Melting was performed in the Applied Biosystems

7500Fast system. MeltDoctor reagents or master mix were used in 20 µL reactions.

MeltDoctor HRM reagent kit was added in the following manner: 1 X Buffer, 3 mM

MgCl2, 0.2 mM dNTP mix, 0.4 µM each primer, 1 X HRM dye, 0.1 U of AmpliTaq Gold

DNA polymerase and 10 ng DNA. PCR cycling was done as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 15 s, annealing temperature for 30 s and 60°C for 45 s. The HRM protocol followed the manufacturer instructions. Fluorescence was recorded every 0.1°C. Each isolate was repeated 3 times for each primer set.

Melting curves were visualized and analyzed with Applied Biosystems HRM software v3.0.1 (Life Technologies). Melting profiles were generated by the software, which involved calculating pre- and post-melting regions, derivative melting curve, aligned and difference melting curves, and plotting of these curves. The HRM software automatically sets variant calls according to the differences in curves. If variant calls were consistent with species tested, we considered the primer set to be successful in hybrid/parental discrimination.

Initially, one or two isolates of each parental and hybrid species were run for each primer set. When they provided sufficient differences in melting profiles, these primer sets were re-tested with more isolates. Information of gene targets, annealing temperature and amplicon sizes of candidate primer sets are detailed in Table 4-2.

To compare melting profiles among samples amplified with the same primer set, we used change in fluorescence at each temperature step. Graphic observation of aligned melting curves between pre- and post-melting temperature shows the divergence of curves according to species. Then, we used difference data, where one

curve was designated as the “standard” and was set to 0 for each temperature and other reactions using the same primer set were compared to this standard. When amplicons contained more than one SNP, melting profiles were complex. Most primer sets used default determination of pre- and post-melting temperature, but if complex curves were generated, we considered only one peak, which was either the first peak or the peak present for all isolates.

Results

Primer Design and Evaluation

For P. infestans and P. andina, polymorphic sites were found at several loci that were annotated mostly as putative proteins, but also as putative known proteins (Table

4-2). We tested 24 primer sets based on amplicon simulations in uMelt. End-point PCR produced 14 primer sets that amplified in both P. infestans and P. andina (Table E-1).

Six primer sets successfully detected P. infestans and P. andina as two different variants by HRM.

A total of 7 nuclear genes were examined for polymorphic sites between P. cactorum and P. nicotianae: Btub, TRP1, Ypt1, TEF1, TigA, Hsp90 and ENL. Based on these loci, 22 primer sets were designed (Table E-2) and 10 successfully amplified both parental species and the hybrid. Five sets successfully distinguished P. xpelgrandis as a variant distinct from one or both parental species based on melting profiles.

P. infestans and P. andina Melting Curves

Primer sets PaM2, PaM18, PaM19, PaM27, PaM33 and PaM57 generated variant calls consistent with species P. infestans and P. andina. Figures 4-3 to 4-7 shows aligned melting curves graphics for every primer set, where panel A indicates the curves by isolate and panel B curves are colored by variant type. All graphics were

generated by the HRM software. Primers PaM2 and PaM18 were tested with two isolates per species, whereas the rest were run with three isolates per species.

Divergence of curves is more obvious for primer sets PaM2, PaM18, PaM19 and

PaM277, and less appreciable for sets PaM33 (Figure 4-6) and PaM57 (Figure 4-7).

After checking derivative melt curves, pre- and post-melting temperatures were modified to exclude regions where peaks were either small or not present in all isolates/species. Figure 4-1 shows the derivative melting graphic for the six primer sets and values are provided in Table 4-3.

Melting curve discrimination and variation between species is visualized more clearly with difference graphics (Figure 4-8). We observed variation among replicates and isolates within a species, but variants could be differentiated. For primer sets

PaM33 and PaM57, the differential curves are narrow, and the variation between species is more distinct inside a smaller temperature range compared to other primer sets.

P. nicotianae, P. cactorum and P. x pelgrandis Melting Curves

We proceeded similarly with primer sets to distinguish P. xpelgrandis from its parental species. Settings for pre and post-melting temperatures are described in Table

4-3 and derivative graphics are shown in Figure 4-9. Primer set Ph29 produced complex melting profiles and thus we examined two different pre- and post-melting settings.

Ph9 was able to distinguish P. nicoianae but not P. cactorum from P. xpelgrandis. Ph11 and Ph25 distinguished only P. cactorum from P. xpelgrandis, producing two different variants. Ph29 was able to differentiate all three species as variants, although the curve separation was only strong for P. cactorum versus P. xpelgrandis. Ph14 successfully differentiated all three species. Aligned melting curves

per primer set are shown in Figures 4-10 to 4-15. Note that we used only one isolate of

P. xpelgrandis for assays, thus we were not able to assess variation among isolates of this hybrid.

Difference plots (Figure 4-16) show variation among replicates of isolates and also within species, when curves vary in dissociation, the difference plot marks drastically this behavior. This is especially noticeable for primer Ph29 when used with three species (Panel F) where despite the variation of P. nicotianae isolates (shades of blue), there is still clear differentiation from P. xpelgrandis curves (gray). Ph9 showed less separation compared to the rest of primer sets in the aligned curves graphic (Figure

4-10), and in the difference plot its curve is more narrow (Figure 4-16, panel A), which makes it more specific for a short range of temperature where the species differ. Still, species distinction is accomplishable.

Discussion

Interspecific hybrids of Phytophthora species are being reported with increasing frequency and include emerging pathogens. HRM was shown to differentiate the hybrid pathogens P. andina and P. xpelgrandis from their parental species. For researchers interested in identifying these hybrids and their parental species from diseased samples, this technique may produce results faster than traditional genotyping or sequencing.

Survival in a changing environment is a driving force in evolution, and hybridization is a remarkable way to obtain significant genetic changes in relatively short time. Interspecific hybridization impacts genome evolution enormously and can result in increased pathogenicity of filamentous microbes (141). Interspecific hybridization is limited by reproductive barriers and genome divergence (41, 47, 141, 142, 156, 172,

173). The movement of live plants and their pathogens into new environments by commercial trade increases the possibility of hybridization among closely related species that have not encountered each other for thousands or millions of years.

Multiple hybrids have been found in nurseries where plant genera such as

Rhododendron can act as melting pots for multiple Phytophthora pathogens (41).

Hybrids may then shift or spread to new hosts or escape from the nursery to wild hosts, causing economic and environmental problems. Both P. xpelgrandis and P. andina were initially investigated because they were causing disease on new hosts, relative to parental species. When hybrids compete with parental strains for the same host, the result may be an increase in virulence. For example, the hybrid Phytophthora xalni causes alder decline, and parental lineages, P. xmultiformis and P. uniformis, are less aggressive than P. xalni. Hybrids outcompete parental species on their primary host or may shift from the parental host when they are less fit or competitive on that host than the parental species (141). Detection of these host range shifts and host jumps helps understand the magnitude and importance of these new problems as they arise and spread.

The HRM method could help reduce the risk posed by P. andina, P. xpelgrandis and other hybrids by providing a rapid method to distinguish the hybrids from their parents. Multiple primer sets can be used in parallel to increase the accuracy of the assays, specifically because the targeted heterozygous SNPs may be present in some loci but not in others due to loss of heterozygosity (174). For P. andina, we developed six primer sets across six conserved genes, which together should provide robust detection of hybrid status. For P. xpelgrandis, differentiation from both parental species

can be obtained with only Ph24 or Ph29, but these primers could be combined with the others that distinguish between P. xpelgrandis and one or the other parent. We were initially limited to multilocus sequencing data, but now that a draft genome of P. cactorum is available (175), it could be used with P. nicotianae genome sequences to identify additional loci for HRM analysis.

While HRM discrimination of hybrid is rapid, there are considerations of cost and convenience. “New generation” saturation dyes increase the resolution and accuracy of

HRM assays, but are more expensive than non-saturating dyes. Because oligo-specific fluorescent probes are not required, the same dye can be used across loci, which lowers the start-up cost of HRM analysis. The method also requires DNA content uniformity. If there are large differences in DNA concentration among isolates, the behavior of melting curves could be affected and cause false variants. This can be overcome by standardizing extraction protocols among isolates being tested and using aliquots with similar DNA concentrations. Finally, access to HRM software is essential for rapid analysis and costs depend on the platform used to generate data. However, open-source programs are available for processing HRM data (176).

In conclusion, HRM could be used for rapid discrimination of hybrids from parental species. Globally, many hybrids have been identified, some more worrisome than others. HRM could be deployed locally for discrimination of hybrid species of particular concern. Multilocus assays are expected to be relatively stable through time due to the asexual nature of P. andina, P. xpelgrandis, and other hybrid Phytophthora of concern. As long as routine genome sequencing of Phytophthora isolates for

identification remains cost prohibitive, HRM has the potential to improve understandings of the distribution and host range of Phytophthora hybrids.

100

Table 4-1. Species, name, host and collection site of isolates used in this study Species Isolate name Host genus Origin Provided by P. infestans J107 Solanum Mexico S. Fernandez-Pavia Mich7043 Solanum Mexico S. Fernandez-Pavia Mich7039 Solanum Mexico S. Fernandez-Pavia Mich7061 Solanum Mexico S. Fernandez-Pavia P. andina 2470 Anarrichomenum Ecuador R. Oliva/G. Forbes 2479 Solanum Ecuador R. Oliva/G. Forbes 2309 Anarrichomenum Ecuador R. Oliva/G. Forbes P. nicotianae F11 Ananas Ecuador M. Ratti M231 Ananas Ecuador M. Ratti G194 Ananas Ecuador M. Ratti Corr2 Correa Italy F. Martin/L. Schena C301 Myrtus Italy F. Martin/L. Schena FerraraR1 Citrus Italy F. Martin/L. Schena P. cactorum 11-1 Fragaria USA N. Peres 12-12 Fragaria USA N. Peres

P. xpelgrandis 956/07 Lavandula Italy R. Faedda

101

Table 4-2. Primer information and details of amplicons. Annealing Primer Polymorphic Temperature Amplicon Name Location sites (°C) size PITG_18049 PaM2 Chromosome segregation protein 2 50 87 PITG_03974 PaM18 Conserved hypothetical protein 1 50 63 PITG_04010 APS kinase/ATP sulfurlyase/ pyrophosphatase fusion PaM19 protein 1 51 51 PITG_08406 PaM27 Conserved hypothetical protein 1 52 56 PITG_09705 PaM33 SUMO-conjugating enzyme (SCE), putative 1 52 63 PITG_20799 PaM57 Conserved hypothetical protein 1 51 57 Ph9 Beta (β) tubulin 3 51 66 Ph11 Beta (β) tubulin 4 54 67 Ph14 Tryptophan biosynthesis protein (TRP1) 2 53 59 Triosephosphate isomerase/glyceraldehyde-3- Ph25 phosphate dehydrogenase (TigA) 5 51 70 Triosephosphate isomerase/glyceraldehyde-3- Ph29 phosphate dehydrogenase (TigA) 12 52 90

102

Table 4-3. Pre and post-melting temperature for each primer set. Prefixes before isolate names indicate species, Pi corresponds to Phytophthora infestans, Pa to P. andina, Pn to P. nicotianae, Pp to P. xpelgrandis and Pc to P. cactorum Primer Pre-melting Post-melting Set stop start Isolates successfully discriminated PaM2 79.6 84.2 PiJ107, PiMich7061, Pa2470, Pa2479 PaM18 80.1 84.8 PiJ107, PiMich7061, Pa2470, Pa2479 PaM19 77.1 83.3 PiJ107, PiMich7043, PiMich7061, Pa2309 Pa2470, Pa2479 PaM27 76.3 83.1 PiJ107, PiMich7043, PiMich7061, Pa2309 Pa2470, Pa2479 PaM33 77.1 83.1 PiJ107, PiMich7043, PiMich7061, Pa2309 Pa2470, Pa2479 PaM57 73.5 83.7 PiJ107, PiMich7043, PiMich7061, Pa2309 Pa2470, Pa2479 Ph9 78.7 87.1 PnFerr1, PnG194, PnC301, Pp956/07 Ph11 77.9 83.6 Pc11-1, Pc12-12, Pp956/07 Ph14 78 84.4 PnC301, PnCorr2, PnM231,Pc11-1, Pc12-12, Pp956/08 Ph25 81.7 88.2 Pc11-1, Pc12-12, Pp956/07 Ph29 76.4 83.6 Pc11-1, Pc12-12, Pp956/07 Ph29 76.4 88.3 PnC301, PnCorr2, PnM231,Pc11-1, Pc12-12, Pp956/08

103

Figure 4-1. Derivative melt curves plots, generated for primer sets to differentiate P. infestans and its hybrid P. andina. Primer sets: A) PaM2, B) PaM18, C) PaM19, D) PaM27, E) PaM33 and F) PaM57. If two peaks were present, only one was considered for setting pre- and post-melting temperatures.

104

Figure 4-2. Aligned melt curves for primer set PaM2, A) shows the curves color coded by sample and B) by variant call

105

Figure 4-3. Aligned melt curves for primer set PaM18, A) shows the curves color coded by sample and B) by variant call

106

Figure 4-4. Aligned melt curves for primer set PaM19, A) shows the curves color coded by sample and B) by variant call

107

Figure 4-5. Aligned melt curves for primer set PaM27, A) shows the curves color coded by sample and B) by variant call

108

Figure 4-6. Aligned melt curves for primer set PaM33, A) shows the curves color coded by sample and B) by variant call

109

Figure 4-7. Aligned melt curves for primer set PaM57, A) shows the curves color coded by sample and B) by variant call

110

Figure 4-8. Difference plots, generated for primer sets to differentiate P. infestans and its hybrid P. andina. A) PaM2, B) PaM18, C) PaM19, D) PaM27, E) PaM33 and F) PaM57. One of the replicates of an isolate was set as 0 and the others were adjusted according to this reference. Note the separation between that P. infestans and P. andina curves separate clearly.

111

Figure 4-9. Derivative Melt Curves plots, generated for primer sets to differentiate the hybrid P. xpelgrandis from its parental species P. nicotianae and P. cactorum. A) Ph9 discriminates P. nicotianae from P. xpelgrandis, B) Ph11 differentiates P. cactorum and P. xpelgrandis, C) Ph14 differentiates all three species, D) Ph25 differentiates P. cactorum and P. xpelgrandis, E) and F) Ph29 successfully separates P. cactorum from P. xpelgrandis, if one peak is considered. To additionally differentiate P. xpelgrandis from P. nicotianae two peaks (and different pre- and post-melting temperature) were included between pre and post-melting temperature.

112

Figure 4-10. Aligned melt curves for primer set Ph9, A) shows the curves color coded by sample and B) by variant call

113

Figure 4-11. Aligned melt curves for primer set Ph11, A) shows the curves color coded by sample and B) by variant call

114

Figure 4-12. Aligned melt curves for primer set Ph14, A) shows the curves color coded by sample and B) by variant call

115

Figure 4-13. Aligned melt curves for primer set Ph25, A) shows the curves color coded by sample and B) by variant call

116

Figure 4-14. Aligned melt curves for primer set Ph29, A) shows the curves color coded by sample and B) by variant call

117

Figure 4-15. Aligned melt curves for primer set Ph29, A) shows the curves color coded by sample and B) by variant call

118

Figure 4-16. Difference plots, generated for primer sets to differentiate P. nicotianae, P. cactorum and its hybrid P. xpelgrandis. A) Ph9, B) Ph11, C) Ph14, D) Ph25, E) Ph29 with P. cactorum and P. xpelgrandis and F) Ph29 with all species. One of the replicates of an isolate was set as 0 and the others were adjusted according to this reference. Note that P. infestans and P. andina curves separate clearly from their counterparts

119

CHAPTER 5 GENERAL DISCUSSION AND CONCLUDING REMARKS

Emerging pathogens cause extensive economic and ecological losses in forests and crop ecosystems. The introduction, establishment and evolution of pathogens has been facilitated by the expansion of agriculture and global commerce, which provide increased opportunities for the long-distance movement of pathogens between wild and crop hosts. In preceding chapters were described three different approaches to study oomycete communities associated with economically important crops in Ecuador: a high-throughput method to assess species richness in cacao soils, analysis of intraspecific genetic variation of Phytophthora nicotianae populations on pineapple, and a high resolution melting assay to discriminate Phytophthora hybrids P. andina and P. pelgrandis.

In Chapter 2, oomycete pathogens of cacao were shown to reside in both cacao soil and in soil in adjacent plant communities. In addition, the utility of cox2 and ITS as barcoding markers was demonstrated and the advantages and disadvantages of each marker were highlighted.

Chapter 3 considered the population structure of Phytophthora nicotianae, which is an economically costly pathogen with a wide host range. We focused on a population causing pineapple heart rot in Ecuador and revealed a genetic lineage that is closely related to, but distinct from, populations on ornamental hosts in other parts of the world.

The need was emphasized for more population studies in tropical crops to characterize how host and geography structure the genetic variation in this pathogen.

Chapter 4 focused on the design and application of an innovative technique for detecting subtle differences between parental and hybrid species of Phytophthora that

120

represent a risk for agriculture in Ecuador and elsewhere. HRM could advance knowledge of hybrid Phytophthora by providing a tool to rapidly differentiate hybrid from parental isolates. This method could be adapted for use in diagnostic clinics with additional markers obtained by high throughput sequencing methods.

Together, this research provides new data and tools to enhance research on oomycetes in Ecuador. These studies provide key initial data for future studies of oomycete diversity in tropical agroecosystems and for studying the population biology of

Phytophthora pathogens. Increased knowledge of oomycetes in tropical ecosystems could lead to better prediction of their evolutionary potential, which is important for understanding and reducing their risks, and developing management strategies.

121

APPENDIX A SUPPLEMENTARY TABLES OF CHAPTER 2

Table A-1. OTU table generated by usearch61 method for cox2 marker 1C1 1C2 1C3 1N1 1N2 1N3 2C1 2C2 2C3 2N1 2N2 2N3 3C1 3C2 3C3 3N1 taxonomy 401 430 11 15 5 4 3107 894 147 36 10 2 875 259 1248 4950 Pythium_cladeI 3 0 1 3 0 30 0 1 2 43 0 2 12 5 0 5 Phytophthora_clade4 1 0 1 0 3 1 0 1 1 0 0 1 2 1 0 4 Phytophthora_clade1 48 4 0 10 0 1 25 1 1 4 1 1 9 19 2 2 Phytophthora_clade2 0 0 0 0 0 2 26 0 3 1 0 0 0 10 0 8 Pythium_cladeB1a 53 11 12 29 3 21 168 47 100 4 5 7 11 51 11 499 Pythium_cladeB1e 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Pythium_cladeB1c 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 3 Phytopythium

Table A-1. Continued 3N2 3N3 4C1 4C2 4C3 4N1 4N2 4N3 5C1 5C2 5C3 5N1 5N2 5N3 6C1 6C2 taxonomy 91 5 12 23 205 48 21 16 142 210 195 30 18 62 174 259 Pythium_cladeI 2 1 7 1 2 0 0 37 15 3 11 10 25 11 0 7 Phytophthora_clade4 0 1 1 0 0 0 4 0 2 9 3 6 5 1 2 3 Phytophthora_clade1 15 0 29 0 6 0 3 3 97 3 29 33 102 41 0 30 Phytophthora_clade2 0 0 0 2 0 1 1 6 0 0 0 0 0 0 0 0 Pythium_cladeB1a 15 18 8 58 26 14 36 86 1086 183 72 186 359 3 145 75 Pythium_cladeB1e 0 0 0 0 2 0 0 0 0 0 0 0 3 30 0 0 Pythium_cladeB1c 0 0 1 0 0 0 0 0 2 0 0 0 0 0 0 0 Phytopythium

122

Table A-1. Continued 6C3 6N1 6N2 6N3 7C1 7C2 7C3 7N1 7N2 7N3 8C1 8C2 8C3 8N1 8N2 8N3 taxonomy 46 222 401 93 1449 298 445 119 191 128 70 1273 158 146 210 3759 Pythium_cladeI 14 3 10090 12 102 3 10 5 1 11 7 0 10 0 6 1 Phytophthora_clade4 1 6 4 3 4 4 3 4 19 5 3 3 3 95 2 3 Phytophthora_clade1 70 6 42786 14 2965 2 21 8 3 38 32 3 41 2 5 2 Phytophthora_clade2 4 0 4 6 0 47 3 0 28 20 0 0 3 4 3 0 Pythium_cladeB1a 31 113 204 150 415 1438 668 353 575 283 187 21 42 50 98 2 Pythium_cladeB1e 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Pythium_cladeB1c 0 1 2 1 0 1 0 0 3 0 0 0 1 0 1 1 Phytopythium

123

Table A-2. OTU table generated by usearch61 method for ITS marker 1C1 1C3 1N1 1N2 2C1 2C2 2N2 2N3 3C2 3C3 3N1 3N3 4C1 4C2 taxonomy 23 0 29 0 0 0 4 1 0 0 0 0 0 0 Achlya 0 0 0 0 0 0 0 0 0 0 0 0 3 0 Dictyuchus 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Phytophthora_clade1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Phytophthora_clade2 0 0 0 0 0 0 0 0 0 1 0 0 0 0 Phytophthora_clade4 0 0 11 1 4 0 1 0 0 17 4 0 98 5 Phytopythium 0 0 3 0 8 0 0 0 0 0 4 0 0 0 Pythium 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Pythium_cladeA 0 0 5 0 192 0 0 0 0 0 0 0 0 1 Pythium_cladeB1a 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Pythium_cladeB1c 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Pythium_cladeB1d 0 0 0 0 30 0 0 0 0 0 0 0 0 0 Pythium_cladeB1e 16 1 0 0 58 1 0 0 31 9 83 0 3 0 Pythium_cladeI 0 0 0 0 0 0 0 0 0 2 0 0 0 0 Pythium_cladeJ

124

Table A-2. Continued. 4C3 5C1 5C3 5N1 6C2 6C3 6N3 7C2 7C3 7N2 8C2 8C3 8N1 8N3 taxonomy 0 0 0 0 0 0 22 0 0 0 0 0 2 0 Achlya 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Dictyuchus 2 0 1 0 0 0 0 0 0 0 0 0 42 0 Phytophthora_clade1 0 2 0 0 0 0 0 0 0 0 0 0 0 0 Phytophthora_clade2 0 6 0 0 0 1 0 0 0 0 0 0 0 0 Phytophthora_clade4 1 30 90 19 94 4 56 268 2 1 0 7 152 Phytopythium 0 0 0 0 0 0 0 0 0 0 0 0 14 0 Pythium 0 0 0 12 0 0 0 0 0 0 0 0 19 0 Pythium_cladeA 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Pythium_cladeB1a 7 0 0 0 0 0 0 0 0 0 0 0 0 0 Pythium_cladeB1c 2 0 0 0 0 0 0 0 0 0 0 0 0 0 Pythium_cladeB1d 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Pythium_cladeB1e 72 0 55 0 68 0 0 3 0 0 64 66 16 173 Pythium_cladeI 0 0 0 0 0 0 0 0 0 0 2 0 0 0 Pythium_cladeJ

125

APPENDIX B SUPPLEMENTARY FIGURES OF CHAPTER 2

Figure B-1. Geographic location of sampling sites in the Coastal Region of Ecuador. Colored circles represent the zones of sampling: Yellow is Zone 3 –South Los Rios, Blue is Zone 1- Central Guayas and Red is Zone 2 - South-East Guayas. Flags indicate the locations of the farms sampled per region.

126

APPENDIX C SUPPLEMENTARY TABLES OF CHAPTER 3

Table C-1. Original allele calls before correction to allow comparison with data published by Biasi et al. 2016 Isolates Markers: P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 Corr2 In this study 198/198 64/77 129/135 157/161 99/120 128/143 105/143 190/190 177/177 Biasi et al. 160/160 66/78 129/135 158/162 99/120 139/154 105/144 190/190 158/158 ACGS22 In this study unclear 77/93 129/135 157/165 112/121 128/152 105/143 190/194 152/158 Biasi et al. 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158 M5rlc In this study 164/182 77/108 135/139 161/178 100/112 140/140 106/129 194/232 183/183 Biasi et al. 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 CM1f In this study 154/154 64/93 129/137 163/167 121/121 140/153 129/132 194/194 177/183 Biasi et al. 116/116 66/93 129/137 164/168 120/120 151/154 129/132 194/194 158/164 C301 In this study 158/158 74/93 129/129 151/164 100/100 128/143 129/158 190/190 170/177 Biasi et al. 120/120 75/93 129/129 152/164 99/99 139/154 129/159 190/190 152/158 Corr1 In this study unclear 64/77 129/135 157/161 99/120 128/143 105/143 190/190 177/177 Biasi et al. 160/160 66/78 129/135 158/162 99/120 139/154 105/144 190/190 158/158 Serr1 In this study 162/162 64/74 127/129 147/162 100/115 128/159 129/143 190/190 158/164 Biasi et al. 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Dodon5 In this study 170/170 93/111 127/137 166/175 112/112 140/143 126/126 190/240 158/158 Biasi et al. 132/132 93/111 127/137 166/174 111/111 151/154 126/126 190/238 158/158 Ferr R3 In this study 162/162 64/74 127/129 147/162 100/115 128/159 129/143 190/190 158/164 Biasi et al. 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Lav5 In this study 162/164 77/77 135/137 157/166 100/100 143/149 106/106 188/204 174/186 Biasi et al. 126/126 78/78 135/137 158/166 99/99 154/160 105/105 186/202 155/167 A512 In this study 160/160 96/96 135/137 158/164 115/121 143/149 132/143 190/190 155/155 Modified 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 F11 In this study 158/158 96/96 135/137 158/164 115/121 143/149 132/143 190/190 155/155 Modified 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155

127

Table C-2. General information of P. nicotianae isolates used for analysis Isolate Host genus Origin A512 Ananas Ecuador A53 Ananas Ecuador A58 Ananas Ecuador A611 Ananas Ecuador A61 Ananas Ecuador A67 Ananas Ecuador A69 Ananas Ecuador F11 Ananas Ecuador F12 Ananas Ecuador F13 Ananas Ecuador F20 Ananas Ecuador F2 Ananas Ecuador G181 Ananas Ecuador G191 Ananas Ecuador G192 Ananas Ecuador G193 Ananas Ecuador G194 Ananas Ecuador M132 Ananas Ecuador M161 Ananas Ecuador M162 Ananas Ecuador M211 Ananas Ecuador M21 Ananas Ecuador M231 Ananas Ecuador M241 Ananas Ecuador M281 Ananas Ecuador M283 Ananas Ecuador M301 Ananas Ecuador M302 Ananas Ecuador M33 Ananas Ecuador M71 Ananas Ecuador M1f1h Myrtus Italy 22Sa Lavandula Italy 23c Lavandula Italy 23ra Lavandula Italy 24c Lavandula Italy 22sb Lavandula Italy Cm1a Convolvulus Italy Cm1f Convolvulus Italy Cm2d Convolvulus Italy Cm5a Convolvulus Italy

128

Table C-2. Continued Isolate Host genus Origin Lavb1a Lavandula Italy Lavb2b Lavandula Italy M1f1d Myrtus Italy M1r2f Myrtus Italy M1r2c Myrtus Italy RCGS21 Ruta Italy RCGS22 Ruta Italy RCGS23 Ruta Italy RGRS14 Ruta Italy RGRS16 Ruta Italy RGRS21 Ruta Italy RGRS23 Ruta Italy RGRS24 Ruta Italy RGRS15 Ruta Italy 178 Lycopersicon Italy RGRS22 Ruta Italy 203 Lycopersicon Italy F16 Lycopersicon Italy Lavanda3 Lavandula Italy Lavanda1 Lavandula Italy Lavanda4 Lavandula Italy Pf1i Citrus Vietnam Pf1l Citrus Vietnam 1A2 Citrus Italy 1B3 Citrus Italy 1C3 Citrus Italy 1B41 Citrus Italy 4B4 Citrus Italy 4C1 Citrus Italy 2C Citrus Italy Lavb2a Lavandula Italy Mirto p5 Myrtus Italy Cm2a Convolvulus Italy Cm2c Convolvulus Italy Cm4a Convolvulus Italy Cm4b Convolvulus Italy F10 Lycopersicon Italy Pomodoro Lycopersicon Italy P1 Lycopersicon Italy P1577 Citrus Italy C301 Myrtus Italy

129

Table C-2. Continued Isolate Host genus Origin scrp465 Lycopersicon Italy CH230 Citrus Italy Pf1a Citrus Vietnam Pf1b Citrus Vietnam Pf1c Citrus Vietnam Pf1d Citrus Vietnam Pf1e Citrus Vietnam Pf1f Citrus Vietnam Pf1h Citrus Vietnam Pf1m Citrus Vietnam Pf1o Citrus Vietnam Pf1p Citrus Vietnam Pf2a Citrus Vietnam Pf2b Citrus Vietnam Pf2c Citrus Vietnam Pf2d Citrus Vietnam Pf2e Citrus Vietnam Pf2f Citrus Vietnam Pf2g Citrus Vietnam CH229 Citrus Vietnam Pf1g Citrus Vietnam E2at Citrus Italy CH281 Citrus Vietnam Ph3 Citrus Italy RADSIJ Citrus Italy Ph87 Citrus Italy CH280 Citrus Vietnam 1B2 Citrus Italy 1C1 Citrus Italy 2A Citrus Italy 2A4 Citrus Italy 2B3 Citrus Italy 3B Citrus Italy 4A1 Citrus Italy 4A4 Citrus Italy 4B1 Citrus Italy FerraraR1 Citrus Italy FerraraR10 Citrus Italy FerraraR11 Citrus Italy Ferrara R3 Citrus Italy Ferrara R5 Citrus Italy

130

Table C-2. Continued Isolate Host genus Origin Ferrara R6 Citrus Italy Phtast2 Citrus Italy Serravalle1 Citrus Italy Serravalle2 Citrus Italy Serravalle4 Citrus Italy CH237 Citrus Vietnam CH236 Citrus Vietnam CH231 Citrus Vietnam CH233 Citrus Vietnam Cedro 10a Citrus Italy Cedro 1c Citrus Italy Cedro 3b Citrus Italy Cedro 6c Citrus Italy Cedro 7d Citrus Italy Cedro 8a Citrus Italy Cedro 8a1 Citrus Italy Cedro 8a2 Citrus Italy CH228 Citrus Vietnam CH235 Citrus Vietnam CH232 Citrus Vietnam M1r1c Myrtus Italy Ph5 Citrus Italy Ph9 Citrus Italy Serravalle3 Citrus Italy Lav5 Lavandula Italy Lavca Lavandula Italy Lavcb Lavandula Italy Lav2ra Lavandula Italy Lav6 Lavandula Italy 2B2 Citrus Italy 3A Citrus Italy Lav3 Lavandula Italy Lav3c Lavandula Italy Lav4 Lavandula Italy Lav1 Lavandula Italy M1f1a Myrtus Italy M1f1b Myrtus Italy M1f1c Myrtus Italy M1f1e Myrtus Italy M1f1g Myrtus Italy M1f2a Myrtus Italy

131

Table C-2 Continued Isolate Host genus Origin M1f2b Myrtus Italy M1f2c Myrtus Italy M1f2f Myrtus Italy M1r2a Myrtus Italy M13r1a Myrtus Italy M13r1b Myrtus Italy M1f1f Myrtus Italy M1r1a Myrtus Italy M1r1e Myrtus Italy M1r2d Myrtus Italy M1r1d Myrtus Italy M1r2b Myrtus Italy M5r1b Myrtus Italy M5r1c Myrtus Italy M5r1d Myrtus Italy M5r2b Myrtus Italy M5r4a Myrtus Italy M5r4b Myrtus Italy M5r4c Myrtus Italy M5r4e Myrtus Italy M5r4f Myrtus Italy M6r1 Myrtus Italy M1r2e Myrtus Italy Lav2 Lavandula Italy Dodoneacol2 Dodonaea Italy Dodoneacol5 Dodonaea Italy Dodoneacol1 Dodonaea Italy Dodonearad1 Dodonaea Italy Dodonearad4 Dodonaea Italy P0582 Nicotiana USA P1333 Nicotiana USA P1334 Nicotiana USA P1335 Nicotiana USA 310 Nicotiana Australia P1752 Nicotiana Australia P1751 Nicotiana Australia P1753 Nicotiana Australia P1494 Nicotiana Australia PH121 Nicotiana USA PH122 Nicotiana USA PH124 Nicotiana USA

132

Table C-2. Continued Isolate Host genus Origin PH125 Nicotiana USA P0583 Nicotiana USA P1350 Nicotiana USA PN26 Nicotiana USA PH123 Nicotiana USA Correa1 Correa Italy Correa11 Correa Italy Correa12 Correa Italy Correa3 Correa Italy Correa4 Correa Italy Correa5 Correa Italy Correa8 Correa Italy Correa10 Correa Italy Correa9 Correa Italy Correa2 Correa Italy Correa6 Correa Italy CH270 Citrus Vietnam CH271 Citrus Vietnam CH272 Citrus Vietnam CH275 Citrus Vietnam P1495 Nicotiana Australia Cm5c Convolvulus Italy 4A2 Citrus Italy 1B42 Citrus Italy 2B4 Citrus Italy F11B Lycopersicon Italy Nic8vasi Lavandula Italy TL8VP Lavandula Italy Pf1n Citrus Vietnam 3A1 Citrus Italy M1r1b Myrtus Italy PN23 Nicotiana USA

133

Table C-3. P. nicotianae allelic dataset used for analysis. Data from non-Ananas hosts were obtained from Biasi et al. 2016 Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 A512 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 A53 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 A58 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 A611 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 A61 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 A67 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 A69 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 F11 100 120/120 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 F12 101 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 F13 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 F20 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 F2 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 G181 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 G191 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 G192 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 G193 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 G194 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M132 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M161 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M162 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M211 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M21 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M231 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M241 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M281 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155

134

Table C-3. Continued Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 M283 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M301 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M302 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M33 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M71 100 122/122 96/96 135/137 158/164 114/120 154/160 132/144 190/190 155/155 M1f1h 2 0/0 66/78 129/137 164/168 99/99 151/154 105/129 186/194 158/164 22Sa 3 116/116 78/96 129/129 164/166 111/120 154/154 105/144 194/194 155/158 23c 3 116/116 78/96 129/129 164/166 111/120 154/154 105/144 194/194 155/158 23ra 3 116/116 78/96 129/129 164/166 111/120 154/154 105/144 194/194 155/158 24c 3 116/116 78/96 129/129 164/166 111/120 154/154 105/144 194/194 155/158 22sb 4 116/116 78/96 129/129 164/166 99/120 154/154 105/144 194/194 155/158 Cm1a 6 116/116 66/93 129/137 164/168 120/120 151/154 129/132 194/194 158/164 Cm1f 6 116/116 66/93 129/137 164/168 120/120 151/154 129/132 194/194 158/164 Cm2d 6 116/116 66/93 129/137 164/168 120/120 151/154 129/132 194/194 158/164 Cm5a 7 116/116 78/90 135/137 168/174 99/120 139/154 126/129 194/202 158/161 Lavb1a 10 116/116 78/78 129/137 155/164 99/120 154/154 105/105 190/194 158/167 Lavb2b 11 116/116 78/99 129/135 158/164 99/99 154/154 105/144 190/230 149/155 M1f1d 12 116/116 66/78 129/137 164/168 99/120 151/154 105/129 186/194 158/164 M1r2f 12 116/116 66/78 129/137 164/168 99/120 151/154 105/129 186/194 158/164 M1r2c 13 116/116 66/78 129/137 164/168 99/120 151/154 105/129 194/194 158/164 RCGS21 15 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158 RCGS22 15 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158 RCGS23 15 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158 RGRS14 15 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158 RGRS16 15 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158

135

Table C-3. Continued Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 RGRS21 15 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158 RGRS23 15 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158 RGRS24 15 116/116 78/93 129/135 158/166 111/120 139/163 105/144 190/194 152/158 RGRS15 16 116/116 78/93 129/135 158/166 120/120 139/163 105/144 190/194 152/158 178 17 116/118 75/78 127/135 164/164 99/120 139/154 105/129 190/214 152/152 RGRS22 17 116/118 75/78 127/135 164/164 99/120 139/154 105/129 190/214 152/152 203 18 116/118 75/78 129/135 164/164 99/120 139/154 105/129 190/214 152/152 F16 19 116/118 78/93 129/135 166/168 111/120 139/163 105/144 190/194 152/158 Lavanda3 20 116/118 75/78 127/135 164/164 99/111 139/151 105/129 194/230 164/164 Lavanda1 20 116/118 75/78 127/135 164/164 99/111 139/151 105/129 194/230 164/164 Lavanda4 21 116/118 75/78 127/135 164/164 99/111 151/151 105/129 194/230 164/164 Pf1i 23 116/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1l 23 116/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 1A2 24 116/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 155/164 1B3 24 116/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 155/164 1C3 24 116/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 155/164 1B41 25 116/124 66/75 127/129 148/162 99/114 139/139 126/129 190/190 155/164 4B4 27 116/124 84/84 127/135 148/162 99/114 139/139 129/144 190/190 158/164 4C1 28 116/124 66/66 129/129 148/162 99/114 139/139 129/144 190/190 155/164 2C 30 116/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Lavb2a 34 116/130 78/99 129/135 158/164 99/99 154/154 105/144 190/230 149/155 Mirto p5 35 116/144 78/108 127/135 166/178 99/111 154/154 105/129 190/194 158/158 Cm2a 36 118/118 78/96 129/135 166/168 99/111 139/163 102/126 202/202 155/155 Cm2c 37 118/118 78/96 129/135 166/168 99/111 139/163 102/126 194/202 155/155 Cm4a 37 118/118 78/96 129/135 166/168 99/111 139/163 102/126 194/202 155/155

136

Table C-3. Continued Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 Cm4b 37 118/118 78/96 129/135 166/168 99/111 139/163 102/126 194/202 155/155 F10 39 118/120 75/84 127/143 158/168 99/120 139/154 105/159 190/210 152/158 Pomodoro 41 118/120 75/78 129/135 164/164 99/99 139/139 105/159 190/190 152/158 P1 42 118/122 75/84 129/135 164/164 99/120 139/154 105/159 190/210 152/158 P1577 43 118/130 66/75 129/129 164/164 99/111 139/151 105/129 190/202 161/164 C301 45 120/120 75/93 129/129 152/164 99/99 139/154 129/159 190/190 152/158 scrp465 46 120/120 75/78 129/129 152/168 99/111 139/139 105/129 190/194 158/158 CH230 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1a 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1b 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1c 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1d 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1e 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1f 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1h 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1m 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1o 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1p 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf2a 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf2b 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf2c 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf2d 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf2e 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf2f 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf2g 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164

137

Table C-3. Continued Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 CH229 48 122/122 66/75 127/129 148/172 111/111 139/139 129/144 190/190 158/164 Pf1g 50 122/122 66/75 129/129 148/172 111/111 139/139 129/144 190/190 158/164 E2at 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 CH281 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Ph3 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 RADSIJ 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Ph87 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 CH280 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 1B2 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 1C1 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 2A 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 2A4 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 2B3 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 3B 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 4A1 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 4A4 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 4B1 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 FerraraR1 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 FerraraR10 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 FerraraR11 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Ferrara R3 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Ferrara R5 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Ferrara R6 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Phtast2 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Serravalle1 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164

138

Table C-3. Continued. Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 Serravalle2 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Serravalle4 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 CH237 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 CH236 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 CH231 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 CH233 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Cedro 10a 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Cedro 1c 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Cedro 3b 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Cedro 6c 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Cedro 7d 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Cedro 8a 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Cedro 8a1 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Cedro 8a2 52 124/124 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 CH228 54 124/124 66/75 127/129 148/164 99/114 139/139 129/144 190/190 158/161 CH235 55 124/124 66/75 127/129 148/162 99/114 139/139 129/141 190/190 158/164 CH232 55 124/124 66/75 127/129 148/162 99/114 139/139 129/141 190/190 158/164 M1r1c 57 124/124 66/78 129/137 164/168 99/99 151/154 105/129 186/194 158/164 Ph5 59 124/124 66/75 127/129 164/164 99/114 139/139 129/144 190/190 158/164 Ph9 60 124/124 66/75 129/129 148/162 99/114 139/139 129/144 190/190 158/164 Serravalle3 61 124/124 66/75 129/143 148/162 99/114 139/139 129/144 190/190 158/164 Lav5 62 124/126 78/78 135/137 158/166 99/99 154/160 105/105 186/202 155/167 Lavca 62 124/126 78/78 135/137 158/166 99/99 154/160 105/105 186/202 155/167 Lavcb 62 124/126 78/78 135/137 158/166 99/99 154/160 105/105 186/202 155/167 Lav2ra 62 124/126 78/78 135/137 158/166 99/99 154/160 105/105 186/202 155/167

139

Table C-3. Continued. Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 Lav6 63 124/126 78/78 135/137 158/166 99/99 154/160 105/105 202/202 155/167 2B2 64 126/126 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 3A 64 126/126 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 Lav3 65 126/126 75/96 129/135 166/168 99/99 154/160 102/105 202/230 155/155 Lav3c 65 126/126 75/96 129/135 166/168 99/99 154/160 102/105 202/230 155/155 Lav4 65 126/126 75/96 129/135 166/168 99/99 154/160 102/105 202/230 155/155 Lav1 65 126/126 75/96 129/135 166/168 99/99 154/160 102/105 202/230 155/155 M1f1a 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M1f1b 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M1f1c 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M1f1e 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M1f1g 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M1f2a 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M1f2b 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M1f2c 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M1f2f 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M1r2a 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M13r1a 66 126/144 78/108 135/139 162/178 99/111 139/151 105/129 194/230 164/164 M13r1b 67 126/144 66/78 129/137 164/168 99/120 151/154 105/129 186/194 158/167 M1f1f 68 126/144 78/108 129/129 162/178 99/111 139/151 105/129 194/230 164/164 M1r1a 69 126/144 66/78 129/137 164/168 99/120 151/154 105/129 186/194 158/164 M1r1e 69 126/144 66/78 129/137 164/168 99/120 151/154 105/129 186/194 158/164 M1r2d 69 126/144 66/78 129/137 164/168 99/120 151/154 105/129 186/194 158/164 M1r1d 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M1r2b 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164

140

Table C-3. Continued Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 M5r1b 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M5r1c 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M5r1d 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M5r2b 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M5r4a 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M5r4b 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M5r4c 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M5r4e 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M5r4f 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M6r1 71 126/144 78/108 135/139 162/178 99/111 151/151 105/129 194/230 164/164 M1r2e 72 126/144 78/108 135/139 162/178 99/99 139/151 105/129 194/230 164/164 Lav2 74 132/132 93/111 127/137 166/174 111/111 151/154 126/126 190/234 158/158 Dodoneacol2 75 132/132 93/111 127/137 166/174 111/111 151/154 126/126 190/238 158/158 Dodoneacol5 75 132/132 93/111 127/137 166/174 111/111 151/154 126/126 190/238 158/158 Dodoneacol1 75 132/132 93/111 127/137 166/174 111/111 151/154 126/126 190/238 158/158 Dodonearad1 75 132/132 93/111 127/137 166/174 111/111 151/154 126/126 190/238 158/158 Dodonearad4 75 132/132 93/111 127/137 166/174 111/111 151/154 126/126 190/238 158/158 P0582 76 136/142 75/75 127/139 160/170 99/111 151/151 126/165 190/190 155/158 P1333 77 136/144 75/75 127/139 162/170 99/111 151/151 126/165 190/190 155/158 P1334 77 136/144 75/75 127/139 162/170 99/111 151/151 126/165 190/190 155/158 Correa8 94 160/160 66/78 129/129 158/162 99/120 139/154 105/144 190/190 158/158 Correa10 94 160/160 66/78 129/129 158/162 99/120 139/154 105/144 190/190 158/158 Correa9 94 160/160 66/78 129/129 158/162 99/120 139/154 105/144 190/190 158/158 Correa2 94 160/160 66/78 129/129 158/162 99/120 139/154 105/144 190/190 158/158 Correa6 95 160/160 66/78 129/137 158/162 99/120 154/154 105/144 190/190 158/158

141

Table C-3. Continued Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 CH270 97 174/176 66/75 135/139 148/162 99/114 139/139 108/144 190/190 158/164 CH271 97 174/176 66/75 135/139 148/162 99/114 139/139 108/144 190/190 158/164 CH272 97 174/176 66/75 135/139 148/162 99/114 139/139 108/144 190/190 158/164 CH275 97 174/176 66/75 135/139 148/162 99/114 139/139 108/144 190/190 158/164 133/139 P1495 1 0/0 75/81 131/135 155/178 99/120 126/144 190/194 158/158 /154 129/135 Cm5c 8 116/116 75/90 168/174 99/120 139/154 126/129 194/202 158/161 /137 99/111 4A2 26 116/124 66/75 127/129 148/162 139/139 129/144 190/190 155/164 /114 66/75 1B42 31 116/124 127/129 148/162 99/114 139/139 129/144 190/190 158/164 /84 116/124 2B4 32 66/75 127/129 148/162 99/114 139/139 129/144 190/190 158/164 /126 75/84 F11B 40 118/120 127/129 148/162 99/120 139/154 105/159 190/210 152/158 /96 118/168 Nic8vasi 44 78/96 135/139 164/166 111/120 151/151 105/129 194/210 164/164 /170 118/168 TL8VP 44 78/96 135/139 164/166 111/120 151/151 105/129 194/210 164/164 /170 66/75 Pf1n 51 122/122 127/129 148/172 111/111 139/139 129/144 190/190 158/164 /84 66/75 3A1 53 124/124 127/129 148/162 99/114 139/139 129/144 190/190 158/164 /84

142

Table C-3. Continued Markers Isolate MLG P1509 P15 P788 P643 P2039 P1129 P17 P5 P2040 139/151 M1r1b 70 126/144 66/78 129/137 164/168 99/120 105/129 186/194 158/164 /154 144/148 PN23 88 87/90 131/135 172/172 99/111 151/151 102/147 190/190 149/158 /150

143

APPENDIX D SUPLEMENTAL FIGURES OF CHAPTER 3

Figure D-1. Maximum likelihood rooted tree using cox2+spacer sequences

144

Figure D-2. Delta K plot for Structure result shown in Fig. 3-3A. Graphic generated by Structure Harvester (102). Selection of the number of distinct clusters was based on the evaluation of the ΔK statistic (103)

145

Figure D-3. Delta K plot for Structure result shown in Fig. 3-3B. Graphic generated by Structure Harvester (102). Selection of the number of distinct clusters was based on the evaluation of the ΔK statistic (103)

146

APPENDIX E SUPLEMENTAL TABLES OF CHAPTER 4

Table E-1. Primer sets designed for HRM analysis of P. andina hybrid and its parental species P. infestans Primer Name Location Sequence (Fw) Sequence (Rv) PaM1* PITG_18049 GGGGTTGTTGAGGGACTCTA TTTGTTGCTTGAGTGTTCGAT PaM2* PITG_18049 GGGGTTGTTGAGGGACTCTAT CGAAGCTCTTCAGTTTGTTGC PaM6* PITG_08426 CGATGCTCAAGGTGAAACAA TCCTCCTCCAGGGGTATACAG PaM7 PITG_08426 CGATGCTCAAGGTGAAACAA TAGCATACCCTCCTCCTCCA PaM9* PITG_08301 CTAACGGGACTGGTGCAAG TGACGGGTCTTCACGTACAA PaM16* PITG_03633 TGACTACGCCCTGGCATC TTAGCAATCAGCACGCAGTT PaM17 PITG_03974 CTTGGCAACAAACTCACCAC GCTTGTGCAAGTTCTTGAAGG PaM18* PITG_03974 GTTACCGTTCAACCCTGTCG TGATTGCCCTTCAGTGTCAG PaM19* PITG_04010 GACATGTCGCGTTCGTTCT GTAGTCGGTGCGACCAAAC PaM21* PITG_05143 CGACCATAGGCCTCACGACT CATCGAGTACTTGCCGTTTG PaM22* PITG_05200 GTCGGGGGTCATTACACG CCCGATACGATCTCCATTTG PaM27* PITG_08406 GGCCCGTGTAGTGAAGCTG GTTGACGACACACCAAGTCG PaM32 PITG_08426 ATCACTTCGGCATTGGATG GTCCGGTGCACAGCAGTT PaM33* PITG_09705 ATATGGAGGACCCAGCACAG TGTAGAGCTCCGGGTCGTT PaM34* PITG_09752 GGCCAAAGCTGAGAGTATGC TCCCTCCTAAGCGTCTTGTT PaM36 PITG_09819 ATGGCAATGGAAATCACCTC TTCTCCAGGACAGGGTTCAC PaM39* PITG_11250 CTCTTCCTAGTGGCCGTGCT CGACCAGGAAGATCATGACA PaM42 PITG_11366 CTTACGCGCCTTGTTCCTT GATCCACGCGTTGGTACAG PaM44 PITG_12669 GCGCTTGCGTCTAGAGAAAAT TGCAACTCTTGCTCGTGTGT PaM45 PITG_12669 CGTTGCGAAAGGATAAGGAA GCTCTTTGTTATCGCGCTGT PaM54 PITG_18238 CGTCAAGATAGCGACCAAGG ACCGGGTTTAAATGCCCTAC PaM57* PITG_20799 CTGACGAAGGCACAGGATCT GCAGCTTCGTTCAAGGATGT PaM60 PITG_18049 GCCGAAACGAAAAGTTGAAG TCCAGCGCCTTCTTATTCAC PaM62 PITG_18243 CTGCGTCACATGCTTCTTGT AAGGTCGAGCCTAACGACAG *Primer sets that successfully amplified for all the isolates of both species considered in this study.

147

Table E-2. Primer sets designed for HRM analysis of P. xpelgrandis hybrid and its parental species P. cactorum and P. nicotianae Primer Name Gene Sequence (Fw) Sequence (Rv) Ph1 ypt1 GCTGCAYGAGATCGATAGGT YARARCASTCAGCTCTTTTCC Ph4 Btub TGCTTCCGTACGCTGAA CAGGCACGTGGTGATA Ph5 Btub GTAAGCTGGCCGTGAACCT GGCGAAACCAATCATAAAGAA Ph6 Btub CACGGCCGCTATTTAACTG CCTTYGTGCTCATACGTCC Ph8 Btub ATTTGCTTCCGTACGCTGAA CGTGGTGATACCARACATGG Ph9* Btub ATTTGCTTCCGTACGCTGA GGCRCACACSAGGTGGTT Ph10* Btub CGTGGCTCGCAGCAGTA ATCATGTTCTTGGCRTCGAA Ph11* Btub TGTCGGTGCACCAGCTT GTACAGGGCCTCGTTATCCA Ph12 TRP1 GCTCGAGCGTCTYAACG TTGCRATAWYAGYAGCAYACAA Ph13 TRP1 TYGCAATTAWRACAGAAGGA CTCGCRCGCTTRAACT Ph14* TRP1 GCTCTGGCTGCCGAGTT AGCTCCGTGGCRATGTC Ph16 TRP1 CTRGAYGACATGATGGMAGCT AGTCCTTGCGCAGGATGG Ph18* Ypt1 TCCGCACGATCACTAGCA GTCCGTCACGTCGTACACC Ph19 Ypt1 TGGCTGCAYGAGATCGATAG YCASTCAGCTCTTTTCCKTSG Ph20 Tef1 CCCGTTACGAGGAGATCAAG TGTAGCCRACCTTCTTCAGG Ph22* tigA CGCYGCCACCGTCAA GTTGCCCTGCTTCCAGAC Ph23* tigA GGGTGCTGAGTACACRCTCG GCCACGATCTCGTTGGTCT Ph25* tigA GTTGCTATCAACGGCTTCG TGATGAGCGGGTTCTTGG Ph27* tigA ACATCTTYGTGAACGGYAAG TACTGCACCTGYTCCTCKCC Ph29 tigA AGCGAGAACGAGATGAAGG GAAGATCGACGAGTGCGAGT Ph32 ENL GTGATCAAGGGCCGCTAC ATGTTCGGGGCGAAACC Ph33 ENL CGTTCGACCAGGAYGACT ACGATCTGMACGTCCTTGC *Primer sets that successfully amplified for all the isolates of the three species considered in this study.

148

LIST OF REFERENCES

1. McDonald BA & Linde C (2002) Pathogen Population Genetics, Evolutionary Potential and Durable Resistance. Annual Review of Phytopathology 40(1):349- 379.

2. Cooke D. SL, Cacciola S.O. (2007) Tools to Detect, Identify and Monitor Phytophthora Species in Natural Ecosystems. Journal of Plant Pathology Vol 89((1)):13-28.

3. Stukenbrock EH & McDonald BA (2008) The Origins of Plant Pathogens in Agro- Ecosystems. Annual Review of Phytopathology 46(1):75-100.

4. Milgroom MG (2015) Population Biology of Plant Pathogens: Genetics, Ecology, and Evolution (The American Phytopathology Society, St. Paul, Minnesota 55121, U.S.A.) p 399.

5. Jones RAC (2009) Plant virus emergence and evolution: Origins, new encounter scenarios, factors driving emergence, effects of changing world conditions, and prospects for control. Virus Research 141(2):113-130.

6. Alexander HM, Mauck KE, Whitfield AE, Garrett KA, & Malmstrom CM (2013) Plant-virus interactions and the agro-ecological interface. European Journal of Plant Pathology 138(3):529-547.

7. Burdon JJ & Thrall PH (2008) Pathogen evolution across the agro-ecological interface: implications for disease management. Evolutionary Applications 1(1):57-65.

8. Papaïx J, Burdon JJ, Zhan J, & Thrall PH (2015) Crop pathogen emergence and evolution in agro-ecological landscapes. Evolutionary Applications 8(4):385-402.

9. Bianchi FJJA, Booij CJH, & Tscharntke T (2006) Sustainable pest regulation in agricultural landscapes: a review on landscape composition, biodiversity and natural pest control. Proceedings of the Royal Society of London B: Biological Sciences 273(1595):1715-1727.

10. Duelli P & Obrist MK (2003) Regional biodiversity in an agricultural landscape: the contribution of seminatural habitat islands. Basic and Applied Ecology 4(2):129-138.

11. Matari N & Blair J (2014) A multilocus timescale for oomycete evolution estimated under three distinct molecular clock models. BMC Evolutionary Biology 14(1):101.

12. Thines M & Kamoun S (2010) Oomycete-plant coevolution: recent advances and future prospects. Current Opinion in Plant Biology 13:1-7(1879-0356 (Electronic)).

149

13. Hardham AR (2007) Cell biology of plant–oomycete interactions. Cellular Microbiology 9(1):31-39.

14. Kamoun S (2003) Molecular Genetics of Pathogenic Oomycetes. Eukaryotic Cell 2(2):191-199.

15. Beakes G, Glockling S, & Sekimoto S (2012) The evolutionary phylogeny of the oomycete “fungi”. Protoplasma 249(1):3-19.

16. Robideau GP, et al. (2011) DNA barcoding of oomycetes with cytochrome c oxidase subunit I and internal transcribed spacer. Molecular Ecology Resources 11(6):1002-1011.

17. Giresse X, Ahmed S, Richard-Cervera S, & Delmotte F (2010) Development of New Oomycete Taxon-specific Mitochondrial Cytochrome b Region Primers for Use in Phylogenetic and Phylogeographic Studies. Journal of Phytopathology 158(4):321-327.

18. Ratnasingham S & Hebert PDN (2007) bold: The Barcode of Life Data System (http://www.barcodinglife.org). Molecular Ecology Notes 7(3):355-364.

19. Kang S Fau - Mansfield MA, et al. (2010) The promise and pitfalls of sequence- based identification of plant-pathogenic fungi and oomycetes. (0031-949X (Print)).

20. Choi Y-J, et al. (2015) Towards a universal barcode of oomycetes – a comparison of the cox1 and cox2 loci. Molecular Ecology Resources:n/a-n/a.

21. Orgiazzi A, Dunbar MB, Panagos P, de Groot GA, & Lemanceau P (2015) Soil biodiversity and DNA barcodes: opportunities and challenges. Soil Biology and Biochemistry 80:244-250.

22. Buyer JS, Baligar VC, He Z, & Arévalo-Gardini E (2017) Soil microbial communities under cacao agroforestry and cover crop systems in Peru. Applied Soil Ecology 120:273-280.

23. Bates ST, et al. (2010) Examining the global distribution of dominant archaeal populations in soil. The Isme Journal 5:908.

24. Nigrelli L & Thines M (2013) Tropical oomycetes in the German Bight – Climate warming or overlooked diversity? Fungal Ecology 6(2):152-160.

25. Zentmyer GA (1987) Taxonomic relationships and distribution of species of Phytophthora causing black pod of cocoa. in Tenth International Cocoa Research Conference (Cocoa Producers' Alliance London UK, Santo Domingo, Dominican Republic), pp 391-395.

150

26. Ploetz R (2016) The Impact of Diseases on Cacao Production: A Global Overview. Cacao Diseases: A History of Old Enemies and New Encounters, eds Bailey BA & Meinhardt LW (Springer International Publishing, Cham), pp 33-59.

27. Erwin Donald C. ROK (1996) Phytophthora diseases worldwide (The American Phytopathological Society., St. Paul Minnesota 55121-2097, USA) p 562.

28. Surujdeo-Maharaj S, Sreenivasan TN, Motilal LA, & Umaharan P (2016) Black Pod and Other Phytophthora Induced Diseases of Cacao: History, Biology, and Control. Cacao Diseases: A History of Old Enemies and New Encounters, eds Bailey BA & Meinhardt LW (Springer International Publishing, Cham), pp 213- 266.

29. Ludowici V., Zhang W., Blackman L., & A. H (2013) Phytophthora nicotianae. Phytophthora A Global Perspective, CABI Plant Protection Series, ed K. L (CAB International, UK), p 256.

30. Mammella MA, Cacciola SO, Martin F, & Schena L (2011) Genetic characterization of Phytophthora nicotianae by the analysis of polymorphic regions of the mitochondrial DNA. Fungal Biology 115(4–5):432-442.

31. Rohrbach KG & Johnson MW (2003) Pests, diseases and weeds. The pineapple: botany, production and uses, eds Bartholomew DP, Paull RE, & Rohrbach KG (CABI Publishing, UK), pp 203-251.

32. Panabieres F, et al. (2016) Phytophthora nicotianae diseases worldwide: new knowledge of a long-recognised pathogen.

33. Rohrbach KG, Leal F, & D'Eeckenbrugge GC (2003) History, distribution and world production. The pineapple: botany, production and uses, eds Bartholomew DP, Paull RE, & Rohrbach KG (CABI Publishing).

34. Chan Y.K., G. CdE, & G.M. S (2003) Breeding and Variety Improvement. The pineapple: botany, production, and uses, eds Bartholomew D.P., Paull R.E., & K.G. R (CABI Publishing, UK), pp 33-55.

35. Fry WE, et al. (2008) Population Genetics and Population Diversity of Phytophthora infestans. Oomycete Genetics and Genomics, (John Wiley & Sons, Inc.), pp 139-164.

36. Vieira MLC, Santini L, Diniz AL, & Munhoz CdF (2016) Microsatellite markers: what they mean and why they are so useful. Genetics and Molecular Biology 39(3):312-328.

37. Ellegren H (2004) Microsatellites: simple sequences with complex evolution. Nature Reviews Genetics 5:435.

151

38. Guichoux E, et al. (2011) Current trends in microsatellite genotyping. Molecular Ecology Resources 11(4):591-611.

39. Biasi A, et al. (2016) Genetic Analysis of Phytophthora nicotianae Populations from Different Hosts Using Microsatellite Markers. Phytopathology 106(9):1006- 1014.

40. Brasier C (2000) Plant pathology: The rise of the hybrid fungi. Nature 405(6783):134-135.

41. Beckerman J, Goodwin S, & Gibson K (2014) The threat of hybrid Phytophthoras (Fort Collins (CO)), (Service UF).

42. Schumer M, Rosenthal GG, & Andolfatto P (2014) Hpw common is homoploid hybrid speciation? Evolution 68(6):1553-1560.

43. Flor HH (1932) Heterothallism and hybridization in Tilletia tritici and T. levis. Journal of Agricultural Research 44(1):49-58.

44. Érsek T & Man in 't Veld WA (2013) Phytophthora species hybrids: a novel threat to crops and natural ecosystems. Phytophthora: a global perspective 2013 pp. 37-47, CABI Plant Protection Series, No.2, ed Lamour K).

45. Goodwin SB & Fry WE (1994) Genetic Analyses of Interspecific Hybrids between Phytophthora infestans and Phytophthora mirabilis. Experimental Mycology 18(1):20-32.

46. Husson C, et al. (2015) Evidence for homoploid speciation in Phytophthora alni supports taxonomic reclassification in this species complex. Fungal Genetics and Biology 77:12-21.

47. Man In 't Veld WA, Rosendahl Kc Fau - Hong C, & Hong C (2012) Phytophthora xserendipita sp. nov. and P. xpelgrandis, two destructive pathogens generated by natural hybridization. (0027-5514 (Print)).

48. Gross BL (2012) Genetic and phenotypic divergence of homoploid hybrid species from parental species. Heredity 108(3):157-158.

49. Brasier CM, Cooke DEL, & Duncan JM (1999) Origin of a new Phytophthora pathogen through interspecific hybridization. Proceedings of the National Academy of Sciences 96(10):5878-5883.

50. Brasier CM, et al. (2004) Phytophthora alni sp. nov. and its variants: designation of emerging heteroploid hybrid pathogens spreading on Alnus trees. Mycological Research 108(10):1172-1184.

152

51. Érsek T & Nagy Z (2008) Species hybrids in the genus Phytophthora with emphasis on the alder pathogen Phytophthora alni: a review. European Journal of Plant Pathology 122(1):31-39.

52. Oliva RF, et al. (2010) Phytophthora andina sp. nov., a newly identified heterothallic pathogen of solanaceous hosts in the Andean highlands. Plant Pathology 59(4):613-625.

53. Goss EM, et al. (2011) The Plant Pathogen Phytophthora andina Emerged via Hybridization of an Unknown Phytophthora Species and the Irish Potato Famine Pathogen, P. infestans. PLoS ONE 6(9):e24543.

54. Man in 't Veld WA (2001) First Report of Natural Hybrids of Phytophthora nicotianae and P. cactorum on Loquat in Taiwan. Plant Disease 85(1):98-98.

55. Yang X, Richardson PA, & Hong C (2014) Phytophthora × stagnum nothosp. nov., a New Hybrid from Irrigation Reservoirs at Ornamental Plant Nurseries in Virginia. PLoS ONE 9(7):e103450.

56. Sansome E, Brasier CM, & Hamm PB (1991) Phytophthora meadii may be a species hybrid. Mycological Research 95(3):273-277.

57. Nagel JH, et al. (2013) Characterization of Phytophthora hybrids from ITS clade 6 associated with riparian ecosystems in South Africa and Australia. Fungal Biology 117(5):329-347.

58. Bertier L, Leus L, D’hondt L, de Cock AWAM, & Höfte M (2013) Host Adaptation and Speciation through Hybridization and Polyploidy in Phytophthora. PLoS ONE 8(12):e85385.

59. Hansen EM, Reeser P, Sutton W, & Brasier CM (2015) Redesignation of Phytophthora taxon Pgchlamydo as Phytophthora chlamydospora sp. nov. 2015 10:14.

60. Catal M, et al. (2010) Heterokaryotic nuclear conditions and a heterogeneous nuclear population are observed by flow cytometry in Phytophthora infestans. Cytometry Part A 77A(8):769-775.

61. Wang J, Presser JW, & Goss EM (2016) Nuclear DNA content of the hybrid plant pathogen Phytophthora andina determined by flow cytometry. Mycologia 108(5):899-904.

62. Cooke D, Drenth A, Duncan JM, Wagels G, & Brasier CM (2000) A Molecular Phylogeny of Phytophthora and Related Oomycetes. Fungal Genetics and Biology 30(1):17-32.

63. Kamoun S (2001) Nonhost resistance to Phytophthora: novel prospects for a classical problem. Current Opinion in Plant Biology 4(4):295-300.

153

64. Lara E & Belbahri L (2011) SSU rRNA reveals major trends in oomycete evolution. Fungal Diversity 49(1):93-100.

65. Martin F (2008) Pythium Genetics. Oomycete Genetics and Genomics, (John Wiley & Sons, Inc.), pp 213-239.

66. Weiland JE (2011) Influence of Isolation Method on Recovery of Pythium Species from Forest Nursery Soils in Oregon and Washington. Plant Disease 95(5):547-553.

67. Hansen EM, Reeser PW, & Sutton W (2012) Phytophthora Beyond Agriculture. Annual Review of Phytopathology 50(1):359-378.

68. Hansen EM (2015) Phytophthora Species Emerging as Pathogens of Forest Trees. Current Forestry Reports 1(1):16-24.

69. Burgess TI, et al. (2009) Re-evaluation of Phytophthora Species Isolated During 30 Years of Vegetation Health Surveys in Western Australia Using Molecular Techniques. Plant Disease 93(3):215-223.

70. Huai Wx, et al. (2013) Identification of Phytophthora species baited and isolated from forest soil and streams in northwestern Yunnan province, China. Forest Pathology 43(2):87-103.

71. Ordoñez ME, et al. (2000) A Novel Population of Phytophthora, Similar to P. infestans, Attacks Wild Solanum Species in Ecuador. Phytopathology 90(2):197- 202.

72. Zhang D & Motilal L (2016) Origin, Dispersal, and Current Global Distribution of Cacao Genetic Diversity. Cacao Diseases: A History of Old Enemies and New Encounters, eds Bailey BA & Meinhardt LW (Springer International Publishing, Cham), pp 3-31.

73. Motamayor JC, et al. (2002) Cacao domestication I: the origin of the cacao cultivated by the Mayas. Heredity 89(5):380-386.

74. Rice RA & Greenberg R (2000) Cacao Cultivation and the Conservation of Biological Diversity. AMBIO: A Journal of the Human Environment 29(3):167-173.

75. De Beenhouwer M, Aerts R, & Honnay O (2013) A global meta-analysis of the biodiversity and ecosystem service benefits of coffee and cacao agroforestry. Agriculture, Ecosystems & Environment 175:1-7.

76. Macfadyen S & Muller W (2013) Edges in Agricultural Landscapes: Species Interactions and Movement of Natural Enemies. PLOS ONE 8(3):e59659.

77. Wisler GC & Norris RF (2005) Interactions between Weeds and Cultivated Plants as Related to Management of Plant Pathogens. Weed Science 53(6):914-917.

154

78. Martin FN, Abad ZG, Balci Y, & Ivors K (2012) Identification and Detection of Phytophthora: Reviewing Our Progress, Identifying Our Needs. Plant Disease 96(8):1080-1103.

79. Hudspeth DSS, Stenger D, & Hudspeth MES (2003) A cox2 phylogenetic hypothesis for the downy mildews and white rusts. Fungal Diversity 13:47-57.

80. Cook K, Hudspeth S, & Hudspeth M (2001) A cox2 phylogeny of representative marine Peronosporomycetes (Oomycetes). (Nova Hedwigia), pp 231-243.

81. Riit T, et al. (2016) Oomycete-specific ITS primers for identification and metabarcoding. MycoKeys 14:17-30.

82. Lundberg DS, Yourstone S, Mieczkowski P, Jones CD, & Dangl JL (2013) Practical innovations for high-throughput amplicon sequencing. Nat Meth 10(10):999-1002.

83. Caporaso JG, et al. (2012) Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. ISME J 6.

84. Brescia P & Banks P (2010) Micro-Volume PicoGreen® DNA Quantification using BioTek’s Take3™ Multi-Volume Plate. (BioTek Instruments, Inc., Winooski, VT).

85. Grünwald NJ, et al. (2010) Phytophthora-ID.org: A Sequence-Based Phytophthora Identification Tool. Plant Disease 95(3):337-342.

86. Afgan E, et al. (2016) The Galaxy platform for accessible, reproducible and collaborative biomedical analyses: 2016 update. Nucleic Acids Research 44(W1):W3-W10.

87. Caporaso JG, et al. (2010) QIIME allows analysis of high-throughput community sequencing data. Nat Meth 7(5):335-336.

88. Ewing B, Hillier L, Wendl MC, & Green P (1998) Base-Calling of Automated Sequencer Traces UsingPhred. I. Accuracy Assessment. Genome Research 8(3):175-185.

89. Ewing B & Green P (1998) Base-Calling of Automated Sequencer Traces Using Phred. II. Error Probabilities. Genome Research 8(3):186-194.

90. Edgar RC (2010) Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26(19):2460-2461.

91. Simpson EH (1949) Measurement of Diversity. Nature 163:688.

92. Anderson MJ (2001) A new method for non-parametric multivariate analysis of variance. Austral Ecology 26(1):32-46.

155

93. Bray JR & Curtis JT (1957) An Ordination of the Upland Forest Communities of Southern Wisconsin. Ecological Monographs 27(4):325-349.

94. Oksanen JF, et al. (2017) vegan: Community Ecology Package. R package version 2.4-3. https://CRAN.R-project.org/package=vegan.

95. Venables WN & Ripley BD (2002) Modern Applied Statistics with S (Springer, New York, USA) Fourth Ed.

96. Blair JE, Coffey Md Fau - Park S-Y, Park Sy Fau - Geiser DM, Geiser Dm Fau - Kang S, & Kang S (2008) A multi-locus phylogeny for Phytophthora utilizing markers derived from complete genome sequences. (1096-0937 (Electronic)).

97. LéVesque CA & De Cock AWAM (2004) Molecular phylogeny and taxonomy of the genus Pythium. Mycological Research 108(12):1363-1383.

98. Bala K, Robideau GP, & Lévesque CA (2010) Phytopythium Abad, de Cock, Bala, Robideau, Lodhi & Lévesque, gen. nov. and Phytopythium sindhum Lodhi, Shahzad & Lévesque, sp. nov. Persoonia - Molecular Phylogeny and Evolution of Fungi 24(1):136-137.

99. de Cock A, et al. (2015) Phytopythium: molecular phylogeny and systematics. Persoonia : Molecular Phylogeny and Evolution of Fungi 34:25-39.

100. Drenth A & Guest DI (2004) Diversity and management of Phytophthora in Southeast Asia (Australian Centre for International Agricultural Research (ACIAR), Canberra) p 238 pp.

101. Martin FN & Tooley PW (2003) Phylogenetic relationships of Phytophthora ramorum, P. nemorosa, and P. pseudosyringae, three species recovered from areas in California with sudden oak death. Mycological research 107.

102. Newhook FJ & Jackson GVH (1977) Phytophthora palmivora in cocoa plantation soils in the Solomon Islands. Transactions of the British Mycological Society 69(1):31-38.

103. Telle S & Thines M (2008) Amplification of cox2 (~620 bp) from 2 mg of Up to 129 Years Old Herbarium Specimens, Comparing 19 Extraction Methods and 15 Polymerases. PLOS ONE 3(10):e3584.

104. Bilodeau GJ, Martin FN, Coffey MD, & Blomquist CL (2014) Development of a Multiplex Assay for Genus- and Species-Specific Detection of Phytophthora Based on Differences in Mitochondrial Gene Order. Phytopathology 104(7):733- 748.

105. Rojas JA, et al. (2016) Oomycete Species Associated with Soybean Seedlings in North America—Part II: Diversity and Ecology in Relation to Environmental and Edaphic Factors. Phytopathology 107(3):293-304.

156

106. Parke JL, Knaus BJ, Fieland VJ, Lewis C, & Grünwald NJ (2014) Phytophthora Community Structure Analyses in Oregon Nurseries Inform Systems Approaches to Disease Management. Phytopathology 104(10):1052-1062.

107. Bilodeau GJ, Koike ST, Uribe P, & Martin FN (2011) Development of an Assay for Rapid Detection and Quantification of Verticillium dahliae in Soil. Phytopathology 102(3):331-343.

108. Bartholomew DP, Paull RE, & Rohrbach KG (2003) The pineapple: botany, production and uses (CABI publishing) pp x-301 pp.

109. Shen HF, Lin BR, Zhan JX, & Pu XM (2012) First Report of Pineapple Heart Rot Caused by Phytophthora nicotianae in Hainan Province, China. Plant Disease 97(4):560-560.

110. Rohrbach KG & Schenck S (1985) Control of Pineapple Heart Rot, Caused by Phytophthora parasitica and P. cinnamomi, with Metalaxyl, Fosetyl Al, and Phosphorous Acid. The American Phytopathological Society.

111. Espinosa-Rodríguez CJ, et al. (2015) Etiología de la pudrición del cogollo de la piña (Ananas comosus. L. Merril) cultivar MD2 en Isla, Veracruz, México. Revista mexicana de fitopatología 33:104-115.

112. Rodríguez Y, et al. (2002) Bioassay for in vitro differentiation of pineapple cultivar resistance levels to heart rot disease. In Vitro Cellular & Developmental Biology - Plant 38(6):613-616.

113. Green J. & S. N (2015) Heart and Root Rots of Pineapple. (College of Tropical Agriculture and Human Resources (CTAHR), Honolulu, Hawai‘i).

114. Meng Y, Zhang Q, Ding W, & Shan W (2014) Phytophthora parasitica: a model oomycete plant pathogen. Mycology 5(2):43-51.

115. Wang YAN, et al. (2011) Infection of Arabidopsis thaliana by Phytophthora parasitica and identification of variation in host specificity. Molecular Plant Pathology 12(2):187-201.

116. McDonald B & Linde C (2002) The population genetics of plant pathogens and breeding strategies for durable resistance. Euphytica 124(2):163-180.

117. Mammella MA, et al. (2013) Analyses of the Population Structure in a Global Collection of Phytophthora nicotianae Isolates Inferred from Mitochondrial and Nuclear DNA Sequences. Phytopathology 103(6):610-622.

118. Parkunan V, Johnson CS, Bowman BC, & Hong CX (2010) Population Structure, Mating Type, and Mefenoxam Sensitivity of Phytophthora nicotianae in Virginia Tobacco Fields. Plant Disease 94(11):1361-1365.

157

119. Zhang XG, Sun WX, Guo L, Yu JF, & Chang CJ (2003) Genetic and Pathogenic Variation Among Tobacco Black Shank Strains of Phytophthora parasitica var. nicotianae from the Main Tobacco Growing in China. Journal of Phytopathology 151(5):259-266.

120. Hulce D, Li X, Snyder-Leiby T, & Johathan Liu CS (2011) GeneMarker® Genotyping Software: Tools to Increase the Statistical Power of DNA Fragment Analysis. Journal of Biomolecular Techniques : JBT 22(Suppl):S35-S36.

121. Kamvar ZN, Brooks JC, & Grünwald NJ (2015) Novel R tools for analysis of genome-wide population genetic data with emphasis on clonality. Frontiers in Genetics 6(208).

122. Kamvar ZN, Tabima JF, & NJ G (2014) Poppr: an R package for genetic analysis of populations with clonal, partially clonal, and/or sexual reproduction. PeerJ 2 : e281.

123. Bruvo R, Michiels NK, D’Souza TG, & Schulenburg H (2004) A simple method for the calculation of microsatellite genotype distances irrespective of ploidy level. Molecular Ecology 13(7):2101-2106.

124. Cassens I, Mardulyn P, Milinkovitch MC, & Steel M (2005) Evaluating Intraspecific “Network” Construction Methods Using Simulated Sequence Data: Do Existing Algorithms Outperform the Global Maximum Parsimony Approach? Systematic Biology 54(3):363-372.

125. Clark LV & Jasieniuk M (2011) polysat: an R package for polyploid microsatellite analysis. Molecular Ecology Resources 11(3):562-566.

126. Pritchard JK, Stephens M, & Donnelly P (2000) Inference of Population Structure Using Multilocus Genotype Data. Genetics 155(2):945.

127. Milgroom MG (1996) Recombination and the Multilocus Structure of Fungal Populations. Annual Review of Phytopathology 34(1):457-477.

128. Earl DA & vonHoldt BM (2012) STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources 4(2):359-361.

129. Evanno G, Regnaut S, & Goudet J (2005) Detecting the number of clusters of individuals using the software structure: a simulation study. Molecular Ecology 14(8):2611-2620.

130. Kearse M, et al. (2012) Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28(12):1647-1649.

158

131. Edgar RC (2004) MUSCLE: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinformatics 5(1):113.

132. Edgar RC (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32(5):1792-1797.

133. Darriba D, Taboada GL, Doallo R, & Posada D (2012) jModelTest 2: more models, new heuristics and parallel computing. Nat Meth 9(8):772-772.

134. Guindon S & Gascuel O (A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. (1063-5157 (Print)).

135. Guindon S, et al. (New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. (1076-836X (Electronic)).

136. Swofford DL, Olsen GJ, Waddell PJ, & Hillis DM (1996) Phylogenetic inference. Molecular Systematics, eds Hilis DM, Moritz C, & Mable BK (Sinauer Associates Inc., Sunderland, Massachusetts USA), 2nd Ed, pp 407–514.

137. Shepard RN (1966) Metric structures in ordinal data. Journal of Mathematical Psychology 3(2):287-315.

138. Ramette A (2007) Multivariate analyses in microbial ecology. Fems Microbiology Ecology 62(2):142-160.

139. Colas V, et al. (1998) Diversity of Virulence in Phytophthora parasitica on Tobacco, as Reflected by Nuclear RFLPs. Phytopathology 88(3):205-212.

140. Aguayo J, et al. (2016) Genetic Diversity and Origins of the Homoploid-Type Hybrid Phytophthora ×alni. Applied and Environmental Microbiology 82(24):7142- 7153.

141. Depotter JRL, Seidl MF, Wood TA, & Thomma BPHJ (2016) Interspecific hybridization impacts host range and pathogenicity of filamentous microbes. Current Opinion in Microbiology 32:7-13.

142. Burgess TI (2015) Molecular Characterization of Natural Hybrids Formed between Five Related Indigenous Clade 6 Phytophthora Species. PLOS ONE 10(8):e0134225.

143. Erwin DC & Ribeiro OK (1996) Phytophthora diseases worldwide (The American Phytopathological Society., St. Paul Minnesota 55121-2097, USA) p 562.

144. Raffaele S, et al. (2010) Genome Evolution Following Host Jumps in the Irish Potato Famine Pathogen Lineage. Science 330(6010):1540-1543.

159

145. Li Y, et al. (2016) Changing Ploidy as a Strategy: The Irish Potato Famine Pathogen Shifts Ploidy in Relation to Its Sexuality. Molecular Plant-Microbe Interactions 30(1):45-52.

146. Blair JE, Coffey MD, & Martin FN (2012) Species Tree Estimation for the Late Blight Pathogen, Phytophthora infestans, and Close Relatives. PLoS ONE 7(5):e37003.

147. Adler N. CG, Forbes G., Flier W. (2002) Phytophthora infestans sensu lato in South America Population substructuring through host-specificity. Global Initiative on Late Blight Conference, ed Lizarraga C (CIP. Global Initiative on Late Blight (GILB)), pp pp. 13-17.

148. Mideros MF, et al. (2018) Phytophthora betacei, a new species within Phytophthora clade 1c causing late blight on Solanum betaceum in Colombia. Persoonia : Molecular Phylogeny and Evolution of Fungi 41(03):39-55.

149. Forbes GA, et al. (2013) Phytophthora infestans and Phytophthora andina on Solanaceous hosts in South America. Phytophthora: a global perspective, CABI Plant Protection Series, ed Lamour K (CABI), Vol 2, pp 48-58.

150. Forbes GA, Gamboa S, Lindqvist‐Kreuze H, Oliva RF, & Perez W (2016) Identification of an A2 population of Phythophthora andina attacking tree tomato in Peru indicates a risk of sexual reproduction in this pathosystem. Plant Pathology 65(7):1109-1117.

151. Man in 't Veld WA, et al. (1998) Natural Hybrids of Phytophthora nicotianae and Phytophthora cactorum Demonstrated by Isozyme Analysis and Random Amplified Polymorphic DNA. Phytopathology 88(9):922-929.

152. Nirenberg HI, Gerlach Wf Fau - Grafenhan T, & Grafenhan T (2009) Phytophthora x pelgrandis, a new natural hybrid pathogenic to Pelargonium grandiflorum hort. Mycologia 101(2):220-231.

153. Hurtado-Gonzales OP, Aragon-Caballero LM, Flores-Torres JG, Veld WMit, & Lamour KH (2009) Molecular comparison of natural hybrids of Phytophthora nicotianae and P. cactorum infecting loquat trees in Peru and Taiwan. Mycologia 101(4):496-502.

154. Szigethy A, et al. (2012) First Report of Phytophthora × pelgrandis Causing Root Rot and Lower Stem Necrosis of Common Box, Lavender and Port-Orford-Cedar in Hungary. Plant Disease 97(1):152-152.

155. Leonberger AJ, Speers C, Ruhl G, Creswell T, & Beckerman JL (2012) A Survey of Phytophthora spp. in Midwest Nurseries, Greenhouses, and Landscapes. Plant Disease 97(5):635-640.

160

156. Faedda R, et al. (2013) Phytophthora × pelgrandis Causes Root and Collar Rot of Lavandula stoechas in Italy. Plant Disease 97(8):1091-1096.

157. Ioos R, Husson C, Andrieux A, & Frey P (2005) SCAR–based PCR primers to detect the hybrid pathogen Phytophthora alni and its subspecies causing alder disease in Europe. European Journal of Plant Pathology 112(4):323-335.

158. Ioos R, Andrieux A, Marçais B, & Frey P (2006) Genetic characterization of the natural hybrid species Phytophthora alni as inferred from nuclear and mitochondrial DNA analyses. Fungal Genetics and Biology 43(7):511-529.

159. Mirmajlessi SM, Loit E, Mänd M, & Mansouripour SM (2015) Real-time PCR applied to study on plant pathogens: potential applications in diagnosis - a review. Plant Protection Science 51(4):177-190.

160. Martini M, Moruzzi S, Ermacora P, Loi N, & Firrao G (2015) Quantitative real-time PCR and high-resolution melting (HRM) analysis for strain-specific monitoring of fluorescent pseudomonads used as biocontrol agents against soil-borne pathogens of food crops. Trends in Food Science & Technology 46(2, Part B):277-285.

161. Taylor Claire F (2009) Mutation scanning using high-resolution melting. Biochemical Society Transactions 37(2):433.

162. Hansen ZR, et al. (2015) SNP-Based Differentiation of Phytophthora infestans Clonal Lineages Using Locked Nucleic Acid Probes and High-Resolution Melt Analysis. Plant Disease 100(7):1297-1306.

163. Reed GH & Wittwer CT (2004) Sensitivity and Specificity of Single-Nucleotide Polymorphism Scanning by High-Resolution Melting Analysis. Clinical Chemistry 50(10):1748.

164. Ganopoulos I, Madesis P, Zambounis A, & Tsaftaris A (2012) High-resolution melting analysis allowed fast and accurate closed-tube genotyping of Fusarium oxysporum formae speciales complex. FEMS Microbiology Letters 334(1):16-21.

165. Gundry CN, et al. (2003) Amplicon Melting Analysis with Labeled Primers: A Closed-Tube Method for Differentiating Homozygotes and Heterozygotes. Clinical Chemistry 49(3):396.

166. Papavasileiou A, Madesis PB, & Karaoglanidis GS (2016) Identification and Differentiation of Monilinia Species Causing Brown Rot of Pome and Stone Fruit using High-Resolution Melting (HRM) Analysis. Phytopathology 106(9):1055- 1064.

167. Rutherford K, et al. (2000) Artemis: sequence visualization and annotation. Bioinformatics (Oxford, England) 16(10):944-945.

161

168. Untergasser A, et al. (2012) Primer3—new capabilities and interfaces. Nucleic Acids Research 40(15):e115-e115.

169. Koressaar T & Remm M (2007) Enhancements and modifications of primer design program Primer3. Bioinformatics 23(10):1289-1291.

170. Dwight Z, Palais R, & Wittwer CT (2011) uMELT: prediction of high-resolution melting curves and dynamic melting profiles of PCR products in a rich web application. Bioinformatics 27(7):1019-1020.

171. Tamura K, Stecher G, Peterson D, Filipski A, & Kumar S (2013) MEGA6: Molecular Evolutionary Genetics Analysis Version 6.0. Molecular Biology and Evolution 30(12):2725-2729.

172. Stukenbrock EH (2016) Hybridization speeds up the emergence and evolution of a new pathogen species. Nature Genetics 48:113.

173. Safaiefarahani B, Mostowfizadeh-Ghalamfarsa R, Hardy GESJ, & Burgess TI (2016) Species from within the Phytophthora cryptogea complex and related species, P. erythroseptica and P. sansomeana, readily hybridize. Fungal Biology 120(8):975-987.

174. Lamour KH, et al. (2012) Genome sequencing and mapping reveal loss of heterozygosity as a mechanism for rapid adaptation in the vegetable pathogen Phytophthora capsici. Molecular plant-microbe interactions : MPMI 25(10):1350- 1360.

175. Grenville-Briggs LJ, et al. (2017) Draft genome of the oomycete pathogen Phytophthora cactorum strain LV007 isolated from European beech (Fagus sylvatica). Genomics Data 12:155-156.

176. Cousins MM, Swan D, Magaret CA, Hoover DR, & Eshleman SH (2012) Analysis of HIV Using a High Resolution Melting (HRM) Diversity Assay: Automation of HRM Data Analysis Enhances the Utility of the Assay for Analysis of HIV Incidence. PLoS ONE 7(12):e51359.

162

BIOGRAPHICAL SKETCH

Maria Fernanda Ratti Torres was born in Esmeraldas, Ecuador. She completed her Bachelor of Science degree in biology at Escuela Superior Politecnica del Litoral

(ESPOL) in 2010. Afterwards, she started working in the Phytopathology and

Microbiology Department at Centro de Investigaciones Biotecnologicas del Ecuador

(CIBE-ESPOL), where she gained experience on bacterial pathogens and soil microbial ecology. She was awarded with a full scholarship of the government of Ecuador to pursue her Doctoral degree. Maria joined the Plant Pathology Department at University of Florida in 2014 under the supervision of Doctor Erica Goss and has studied population genetics and soil communities of Oomycetes in tropical and subtropical environments. Maria will become a Doctor of Philosophy by May of 2018.

163